From 3dacc900c00bad7275ee8c096dbcaab699d83e36 Mon Sep 17 00:00:00 2001 From: Kriish Sharma Date: Mon, 13 Oct 2025 01:43:19 +0000 Subject: [PATCH 0001/1381] hdlc_ppp: fix potential null pointer in ppp_cp_event logging MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit drivers/net/wan/hdlc_ppp.c: In function ‘ppp_cp_event’: drivers/net/wan/hdlc_ppp.c:353:17: warning: ‘%s’ directive argument is null [-Wformat-overflow=] 353 | netdev_info(dev, "%s down\n", proto_name(pid)); | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ drivers/net/wan/hdlc_ppp.c:342:17: warning: ‘%s’ directive argument is null [-Wformat-overflow=] 342 | netdev_info(dev, "%s up\n", proto_name(pid)); | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Update proto_name() to return "LCP" by default instead of NULL. This change silences the compiler without changing existing behavior and removes the need for the local 'pname' variable in ppp_cp_event. Suggested-by: Krzysztof Hałasa Signed-off-by: Kriish Sharma Acked-by: Krzysztof Hałasa Link: https://patch.msgid.link/20251013014319.1608706-1-kriish.sharma2006@gmail.com Signed-off-by: Paolo Abeni --- drivers/net/wan/hdlc_ppp.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/drivers/net/wan/hdlc_ppp.c b/drivers/net/wan/hdlc_ppp.c index 7496a2e9a282..159295c4bd6d 100644 --- a/drivers/net/wan/hdlc_ppp.c +++ b/drivers/net/wan/hdlc_ppp.c @@ -126,14 +126,12 @@ static inline struct proto *get_proto(struct net_device *dev, u16 pid) static inline const char *proto_name(u16 pid) { switch (pid) { - case PID_LCP: - return "LCP"; case PID_IPCP: return "IPCP"; case PID_IPV6CP: return "IPV6CP"; default: - return NULL; + return "LCP"; } } From 53615ad26e9789bfcdf3a4dccbcecb15294ea024 Mon Sep 17 00:00:00 2001 From: Byungchul Park Date: Mon, 13 Oct 2025 13:41:33 +0900 Subject: [PATCH 0002/1381] netmem: replace __netmem_clear_lsb() with netmem_to_nmdesc() Now that we have struct netmem_desc, it'd better access the pp fields via struct netmem_desc rather than struct net_iov. Introduce netmem_to_nmdesc() for safely converting netmem_ref to netmem_desc regardless of the type underneath e.i. netmem_desc, net_iov. While at it, remove __netmem_clear_lsb() and make netmem_to_nmdesc() used instead. Suggested-by: Pavel Begunkov Signed-off-by: Byungchul Park Reviewed-by: Mina Almasry Link: https://patch.msgid.link/20251013044133.69472-1-byungchul@sk.com Signed-off-by: Paolo Abeni --- include/net/netmem.h | 68 +++++++++++++++++++++--------------------- net/core/netmem_priv.h | 16 +++++----- 2 files changed, 42 insertions(+), 42 deletions(-) diff --git a/include/net/netmem.h b/include/net/netmem.h index f7dacc9e75fd..651e2c62d1dd 100644 --- a/include/net/netmem.h +++ b/include/net/netmem.h @@ -247,6 +247,23 @@ static inline unsigned long netmem_pfn_trace(netmem_ref netmem) return page_to_pfn(netmem_to_page(netmem)); } +/* XXX: How to extract netmem_desc from page must be changed, once + * netmem_desc no longer overlays on page and will be allocated through + * slab. + */ +#define __pp_page_to_nmdesc(p) (_Generic((p), \ + const struct page * : (const struct netmem_desc *)(p), \ + struct page * : (struct netmem_desc *)(p))) + +/* CAUTION: Check if the page is a pp page before calling this helper or + * know it's a pp page. + */ +#define pp_page_to_nmdesc(p) \ +({ \ + DEBUG_NET_WARN_ON_ONCE(!page_pool_page_is_pp(p)); \ + __pp_page_to_nmdesc(p); \ +}) + /** * __netmem_to_nmdesc - unsafely get pointer to the &netmem_desc backing * @netmem @@ -265,43 +282,26 @@ static inline struct netmem_desc *__netmem_to_nmdesc(netmem_ref netmem) return (__force struct netmem_desc *)netmem; } -/* __netmem_clear_lsb - convert netmem_ref to struct net_iov * for access to - * common fields. - * @netmem: netmem reference to extract as net_iov. +/* netmem_to_nmdesc - convert netmem_ref to struct netmem_desc * for + * access to common fields. + * @netmem: netmem reference to get netmem_desc. * - * All the sub types of netmem_ref (page, net_iov) have the same pp, pp_magic, - * dma_addr, and pp_ref_count fields at the same offsets. Thus, we can access - * these fields without a type check to make sure that the underlying mem is - * net_iov or page. + * All the sub types of netmem_ref (netmem_desc, net_iov) have the same + * pp, pp_magic, dma_addr, and pp_ref_count fields via netmem_desc. * - * The resulting value of this function can only be used to access the fields - * that are NET_IOV_ASSERT_OFFSET'd. Accessing any other fields will result in - * undefined behavior. - * - * Return: the netmem_ref cast to net_iov* regardless of its underlying type. + * Return: the pointer to struct netmem_desc * regardless of its + * underlying type. */ -static inline struct net_iov *__netmem_clear_lsb(netmem_ref netmem) +static inline struct netmem_desc *netmem_to_nmdesc(netmem_ref netmem) { - return (struct net_iov *)((__force unsigned long)netmem & ~NET_IOV); + void *p = (void *)((__force unsigned long)netmem & ~NET_IOV); + + if (netmem_is_net_iov(netmem)) + return &((struct net_iov *)p)->desc; + + return __pp_page_to_nmdesc((struct page *)p); } -/* XXX: How to extract netmem_desc from page must be changed, once - * netmem_desc no longer overlays on page and will be allocated through - * slab. - */ -#define __pp_page_to_nmdesc(p) (_Generic((p), \ - const struct page * : (const struct netmem_desc *)(p), \ - struct page * : (struct netmem_desc *)(p))) - -/* CAUTION: Check if the page is a pp page before calling this helper or - * know it's a pp page. - */ -#define pp_page_to_nmdesc(p) \ -({ \ - DEBUG_NET_WARN_ON_ONCE(!page_pool_page_is_pp(p)); \ - __pp_page_to_nmdesc(p); \ -}) - /** * __netmem_get_pp - unsafely get pointer to the &page_pool backing @netmem * @netmem: netmem reference to get the pointer from @@ -320,12 +320,12 @@ static inline struct page_pool *__netmem_get_pp(netmem_ref netmem) static inline struct page_pool *netmem_get_pp(netmem_ref netmem) { - return __netmem_clear_lsb(netmem)->pp; + return netmem_to_nmdesc(netmem)->pp; } static inline atomic_long_t *netmem_get_pp_ref_count_ref(netmem_ref netmem) { - return &__netmem_clear_lsb(netmem)->pp_ref_count; + return &netmem_to_nmdesc(netmem)->pp_ref_count; } static inline bool netmem_is_pref_nid(netmem_ref netmem, int pref_nid) @@ -390,7 +390,7 @@ static inline bool netmem_is_pfmemalloc(netmem_ref netmem) static inline unsigned long netmem_get_dma_addr(netmem_ref netmem) { - return __netmem_clear_lsb(netmem)->dma_addr; + return netmem_to_nmdesc(netmem)->dma_addr; } void get_netmem(netmem_ref netmem); diff --git a/net/core/netmem_priv.h b/net/core/netmem_priv.h index cd95394399b4..23175cb2bd86 100644 --- a/net/core/netmem_priv.h +++ b/net/core/netmem_priv.h @@ -5,19 +5,19 @@ static inline unsigned long netmem_get_pp_magic(netmem_ref netmem) { - return __netmem_clear_lsb(netmem)->pp_magic & ~PP_DMA_INDEX_MASK; + return netmem_to_nmdesc(netmem)->pp_magic & ~PP_DMA_INDEX_MASK; } static inline void netmem_or_pp_magic(netmem_ref netmem, unsigned long pp_magic) { - __netmem_clear_lsb(netmem)->pp_magic |= pp_magic; + netmem_to_nmdesc(netmem)->pp_magic |= pp_magic; } static inline void netmem_clear_pp_magic(netmem_ref netmem) { - WARN_ON_ONCE(__netmem_clear_lsb(netmem)->pp_magic & PP_DMA_INDEX_MASK); + WARN_ON_ONCE(netmem_to_nmdesc(netmem)->pp_magic & PP_DMA_INDEX_MASK); - __netmem_clear_lsb(netmem)->pp_magic = 0; + netmem_to_nmdesc(netmem)->pp_magic = 0; } static inline bool netmem_is_pp(netmem_ref netmem) @@ -27,13 +27,13 @@ static inline bool netmem_is_pp(netmem_ref netmem) static inline void netmem_set_pp(netmem_ref netmem, struct page_pool *pool) { - __netmem_clear_lsb(netmem)->pp = pool; + netmem_to_nmdesc(netmem)->pp = pool; } static inline void netmem_set_dma_addr(netmem_ref netmem, unsigned long dma_addr) { - __netmem_clear_lsb(netmem)->dma_addr = dma_addr; + netmem_to_nmdesc(netmem)->dma_addr = dma_addr; } static inline unsigned long netmem_get_dma_index(netmem_ref netmem) @@ -43,7 +43,7 @@ static inline unsigned long netmem_get_dma_index(netmem_ref netmem) if (WARN_ON_ONCE(netmem_is_net_iov(netmem))) return 0; - magic = __netmem_clear_lsb(netmem)->pp_magic; + magic = netmem_to_nmdesc(netmem)->pp_magic; return (magic & PP_DMA_INDEX_MASK) >> PP_DMA_INDEX_SHIFT; } @@ -57,6 +57,6 @@ static inline void netmem_set_dma_index(netmem_ref netmem, return; magic = netmem_get_pp_magic(netmem) | (id << PP_DMA_INDEX_SHIFT); - __netmem_clear_lsb(netmem)->pp_magic = magic; + netmem_to_nmdesc(netmem)->pp_magic = magic; } #endif From 16a2206354d169bfd13552ad577e07ce66e439ab Mon Sep 17 00:00:00 2001 From: Jan Vaclav Date: Thu, 9 Oct 2025 23:09:08 +0200 Subject: [PATCH 0003/1381] net/hsr: add protocol version to fill_info output Currently, it is possible to configure IFLA_HSR_VERSION, but there is no way to check in userspace what the currently configured HSR protocol version is. Add it to the output of hsr_fill_info(), when the interface is using the HSR protocol. Let's not expose it when using the PRP protocol, since it only has one version and it's not possible to set it from userspace. This info could then be used by e.g. ip(8), like so: $ ip -d link show hsr0 12: hsr0: mtu ... ... hsr slave1 veth0 slave2 veth1 ... proto 0 version 1 Reviewed-by: Fernando Fernandez Mancera Signed-off-by: Jan Vaclav Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20251009210903.1055187-6-jvaclav@redhat.com Signed-off-by: Paolo Abeni --- net/hsr/hsr_netlink.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/hsr/hsr_netlink.c b/net/hsr/hsr_netlink.c index b120470246cc..4461adf69623 100644 --- a/net/hsr/hsr_netlink.c +++ b/net/hsr/hsr_netlink.c @@ -166,6 +166,8 @@ static int hsr_fill_info(struct sk_buff *skb, const struct net_device *dev) goto nla_put_failure; if (hsr->prot_version == PRP_V1) proto = HSR_PROTOCOL_PRP; + else if (nla_put_u8(skb, IFLA_HSR_VERSION, hsr->prot_version)) + goto nla_put_failure; if (nla_put_u8(skb, IFLA_HSR_PROTOCOL, proto)) goto nla_put_failure; From 0513a3f97b96aaf41e91e02bc3b5d08c9dce5bfa Mon Sep 17 00:00:00 2001 From: Alok Tiwari Date: Mon, 13 Oct 2025 03:01:16 -0700 Subject: [PATCH 0004/1381] net: bridge: correct debug message function name in br_fill_ifinfo The debug message in br_fill_ifinfo() incorrectly refers to br_fill_info instead of the actual function name. Update it for clarity in debugging output. Signed-off-by: Alok Tiwari Reviewed-by: Simon Horman Acked-by: Nikolay Aleksandrov Link: https://patch.msgid.link/20251013100121.755899-1-alok.a.tiwari@oracle.com Signed-off-by: Jakub Kicinski --- net/bridge/br_netlink.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/bridge/br_netlink.c b/net/bridge/br_netlink.c index 4e2d53b27221..0264730938f4 100644 --- a/net/bridge/br_netlink.c +++ b/net/bridge/br_netlink.c @@ -467,7 +467,7 @@ static int br_fill_ifinfo(struct sk_buff *skb, else br = netdev_priv(dev); - br_debug(br, "br_fill_info event %d port %s master %s\n", + br_debug(br, "br_fill_ifinfo event %d port %s master %s\n", event, dev->name, br->dev->name); nlh = nlmsg_put(skb, pid, seq, event, sizeof(*hdr), flags); From e0aa115271394d68992e4a0369479e3632038c2a Mon Sep 17 00:00:00 2001 From: Alok Tiwari Date: Mon, 13 Oct 2025 09:05:02 -0700 Subject: [PATCH 0005/1381] eth: fbnic: fix various typos in comments and strings Fix several minor typos and grammatical errors in comments and log (in fbnic firmware, PCI, and time modules) Changes include: - "cordeump" -> "coredump" - "of" -> "off" in RPC config comment - "healty" -> "healthy" in firmware heartbeat comment - "Firmware crashed detected!" -> "Firmware crash detected!" - "The could be caused" -> "This could be caused" - "lockng" -> "locking" in fbnic_time.c Signed-off-by: Alok Tiwari Reviewed-by: Simon Horman Link: https://patch.msgid.link/20251013160507.768820-1-alok.a.tiwari@oracle.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/meta/fbnic/fbnic_fw.c | 6 +++--- drivers/net/ethernet/meta/fbnic/fbnic_pci.c | 6 +++--- drivers/net/ethernet/meta/fbnic/fbnic_time.c | 2 +- 3 files changed, 7 insertions(+), 7 deletions(-) diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_fw.c b/drivers/net/ethernet/meta/fbnic/fbnic_fw.c index c87cb9ed09e7..1166fa17438d 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_fw.c +++ b/drivers/net/ethernet/meta/fbnic/fbnic_fw.c @@ -878,11 +878,11 @@ fbnic_fw_parse_coredump_info_resp(void *opaque, struct fbnic_tlv_msg **results) * @fbd: FBNIC device structure * @cmpl_data: Completion struct to store coredump * @offset: Offset into coredump requested - * @length: Length of section of cordeump to fetch + * @length: Length of section of coredump to fetch * * Return: zero on success, negative errno on failure * - * Asks the firmware to provide a section of the cordeump back in a message. + * Asks the firmware to provide a section of the coredump back in a message. * The response will have an offset and size matching the values provided. */ int fbnic_fw_xmit_coredump_read_msg(struct fbnic_dev *fbd, @@ -1868,7 +1868,7 @@ int fbnic_fw_xmit_rpc_macda_sync(struct fbnic_dev *fbd) if (err) goto free_message; - /* Send message of to FW notifying it of current RPC config */ + /* Send message off to FW notifying it of current RPC config */ err = fbnic_mbx_map_tlv_msg(fbd, msg); if (err) goto free_message; diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_pci.c b/drivers/net/ethernet/meta/fbnic/fbnic_pci.c index a7a6b4db8016..4620f1847f2e 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_pci.c +++ b/drivers/net/ethernet/meta/fbnic/fbnic_pci.c @@ -185,7 +185,7 @@ static void fbnic_health_check(struct fbnic_dev *fbd) { struct fbnic_fw_mbx *tx_mbx = &fbd->mbx[FBNIC_IPC_MBX_TX_IDX]; - /* As long as the heart is beating the FW is healty */ + /* As long as the heart is beating the FW is healthy */ if (fbd->fw_heartbeat_enabled) return; @@ -196,7 +196,7 @@ static void fbnic_health_check(struct fbnic_dev *fbd) if (tx_mbx->head != tx_mbx->tail) return; - fbnic_devlink_fw_report(fbd, "Firmware crashed detected!"); + fbnic_devlink_fw_report(fbd, "Firmware crash detected!"); fbnic_devlink_otp_check(fbd, "error detected after firmware recovery"); if (fbnic_fw_config_after_crash(fbd)) @@ -378,7 +378,7 @@ static int fbnic_probe(struct pci_dev *pdev, const struct pci_device_id *ent) * @pdev: PCI device information struct * * Called by the PCI subsystem to alert the driver that it should release - * a PCI device. The could be caused by a Hot-Plug event, or because the + * a PCI device. This could be caused by a Hot-Plug event, or because the * driver is going to be removed from memory. **/ static void fbnic_remove(struct pci_dev *pdev) diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_time.c b/drivers/net/ethernet/meta/fbnic/fbnic_time.c index 39d99677b71e..db7748189f45 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_time.c +++ b/drivers/net/ethernet/meta/fbnic/fbnic_time.c @@ -253,7 +253,7 @@ static void fbnic_ptp_reset(struct fbnic_dev *fbd) void fbnic_time_init(struct fbnic_net *fbn) { - /* This is not really a statistic, but the lockng primitive fits + /* This is not really a statistic, but the locking primitive fits * our usecase perfectly, we need an atomic 8 bytes READ_ONCE() / * WRITE_ONCE() behavior. */ From bdec4271e8081fed339759c45f1db81ea7f8f8ed Mon Sep 17 00:00:00 2001 From: Jonas Gorski Date: Mon, 13 Oct 2025 17:28:34 +0200 Subject: [PATCH 0006/1381] net: dsa: b53: implement port isolation support Implement port isolation support via the Protected Ports register. Protected ports can only communicate with unprotected ports, but not with each other, matching the expected behaviour of isolated ports. Tested on BCM963268BU. Signed-off-by: Jonas Gorski Reviewed-by: Florian Fainelli Reviewed-by: Vladimir Oltean Link: https://patch.msgid.link/20251013152834.100169-1-jonas.gorski@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/b53/b53_common.c | 25 ++++++++++++++++++++++++- drivers/net/dsa/b53/b53_regs.h | 4 ++++ 2 files changed, 28 insertions(+), 1 deletion(-) diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c index 2f846381d5a7..ad4990da9f7c 100644 --- a/drivers/net/dsa/b53/b53_common.c +++ b/drivers/net/dsa/b53/b53_common.c @@ -632,6 +632,25 @@ static void b53_port_set_learning(struct b53_device *dev, int port, b53_write16(dev, B53_CTRL_PAGE, B53_DIS_LEARNING, reg); } +static void b53_port_set_isolated(struct b53_device *dev, int port, + bool isolated) +{ + u8 offset; + u16 reg; + + if (is5325(dev)) + offset = B53_PROTECTED_PORT_SEL_25; + else + offset = B53_PROTECTED_PORT_SEL; + + b53_read16(dev, B53_CTRL_PAGE, offset, ®); + if (isolated) + reg |= BIT(port); + else + reg &= ~BIT(port); + b53_write16(dev, B53_CTRL_PAGE, offset, reg); +} + static void b53_eee_enable_set(struct dsa_switch *ds, int port, bool enable) { struct b53_device *dev = ds->priv; @@ -652,6 +671,7 @@ int b53_setup_port(struct dsa_switch *ds, int port) b53_port_set_ucast_flood(dev, port, true); b53_port_set_mcast_flood(dev, port, true); b53_port_set_learning(dev, port, false); + b53_port_set_isolated(dev, port, false); /* Force all traffic to go to the CPU port to prevent the ASIC from * trying to forward to bridged ports on matching FDB entries, then @@ -2318,7 +2338,7 @@ int b53_br_flags_pre(struct dsa_switch *ds, int port, struct netlink_ext_ack *extack) { struct b53_device *dev = ds->priv; - unsigned long mask = (BR_FLOOD | BR_MCAST_FLOOD); + unsigned long mask = (BR_FLOOD | BR_MCAST_FLOOD | BR_ISOLATED); if (!is5325(dev)) mask |= BR_LEARNING; @@ -2343,6 +2363,9 @@ int b53_br_flags(struct dsa_switch *ds, int port, if (flags.mask & BR_LEARNING) b53_port_set_learning(ds->priv, port, !!(flags.val & BR_LEARNING)); + if (flags.mask & BR_ISOLATED) + b53_port_set_isolated(ds->priv, port, + !!(flags.val & BR_ISOLATED)); return 0; } diff --git a/drivers/net/dsa/b53/b53_regs.h b/drivers/net/dsa/b53/b53_regs.h index 309fe0e46dad..c16b3e3e8227 100644 --- a/drivers/net/dsa/b53/b53_regs.h +++ b/drivers/net/dsa/b53/b53_regs.h @@ -120,6 +120,10 @@ #define B53_SWITCH_CTRL 0x22 #define B53_MII_DUMB_FWDG_EN BIT(6) +/* Protected Port Selection (16 bit) */ +#define B53_PROTECTED_PORT_SEL 0x24 +#define B53_PROTECTED_PORT_SEL_25 0x26 + /* (16 bit) */ #define B53_UC_FLOOD_MASK 0x32 #define B53_MC_FLOOD_MASK 0x34 From 10c4b4f60f5d0dbd29fa819be76e888501c7b729 Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Mon, 13 Oct 2025 22:50:27 +0200 Subject: [PATCH 0007/1381] net: mdio: use macro module_driver to avoid boilerplate code Use macro module_driver to avoid boilerplate code. Signed-off-by: Heiner Kallweit Link: https://patch.msgid.link/e5c37417-4984-4b57-8154-264deef61e0d@gmail.com Signed-off-by: Jakub Kicinski --- include/linux/mdio.h | 13 ++----------- 1 file changed, 2 insertions(+), 11 deletions(-) diff --git a/include/linux/mdio.h b/include/linux/mdio.h index c640ba44dd6e..42d6d47e445b 100644 --- a/include/linux/mdio.h +++ b/include/linux/mdio.h @@ -689,16 +689,7 @@ struct phy_device *mdiobus_get_phy(struct mii_bus *bus, int addr); * init/exit. Each module may only use this macro once, and calling it * replaces module_init() and module_exit(). */ -#define mdio_module_driver(_mdio_driver) \ -static int __init mdio_module_init(void) \ -{ \ - return mdio_driver_register(&_mdio_driver); \ -} \ -module_init(mdio_module_init); \ -static void __exit mdio_module_exit(void) \ -{ \ - mdio_driver_unregister(&_mdio_driver); \ -} \ -module_exit(mdio_module_exit) +#define mdio_module_driver(_mdio_driver) \ + module_driver(_mdio_driver, mdio_driver_register, mdio_driver_unregister) #endif /* __LINUX_MDIO_H__ */ From c3527eeb65cfe6fb93f9e06bc0429616a3a23592 Mon Sep 17 00:00:00 2001 From: Denis Benato Date: Mon, 13 Oct 2025 20:36:32 +0200 Subject: [PATCH 0008/1381] eth: fealnx: fix typo in comments There are a few typos in comments: - replace "avilable" with "available" - replace "mutlicast" with "multicast" Signed-off-by: Denis Benato Reviewed-by: Simon Horman Link: https://patch.msgid.link/20251013183632.1226627-1-benato.denis96@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/fealnx.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/fealnx.c b/drivers/net/ethernet/fealnx.c index 6ac8547ef9b8..3c9961806f75 100644 --- a/drivers/net/ethernet/fealnx.c +++ b/drivers/net/ethernet/fealnx.c @@ -196,7 +196,7 @@ enum intr_status_bits { ERI = 0x00000080, /* receive early int */ CNTOVF = 0x00000040, /* counter overflow */ RBU = 0x00000020, /* receive buffer unavailable */ - TBU = 0x00000010, /* transmit buffer unavilable */ + TBU = 0x00000010, /* transmit buffer unavailable */ TI = 0x00000008, /* transmit interrupt */ RI = 0x00000004, /* receive interrupt */ RxErr = 0x00000002, /* receive error */ @@ -215,7 +215,7 @@ enum rx_mode_bits { CR_W_RXMODEMASK = 0x000000e0, CR_W_PROM = 0x00000080, /* promiscuous mode */ CR_W_AB = 0x00000040, /* accept broadcast */ - CR_W_AM = 0x00000020, /* accept mutlicast */ + CR_W_AM = 0x00000020, /* accept multicast */ CR_W_ARP = 0x00000008, /* receive runt pkt */ CR_W_ALP = 0x00000004, /* receive long pkt */ CR_W_SEP = 0x00000002, /* receive error pkt */ From 331f8a8bea22aecf99437f3561453a85f40026de Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Mon, 13 Oct 2025 16:29:41 +0200 Subject: [PATCH 0009/1381] net: airoha: Add missing stats to ethtool_eth_mac_stats Add the following stats to ethtool ethtool_eth_mac_stats stats: - FramesTransmittedOK - OctetsTransmittedOK - FramesReceivedOK - OctetsReceivedOK Signed-off-by: Lorenzo Bianconi Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20251013-airoha-ethtool-improvements-v1-1-fdd1c6fc9be1@kernel.org Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/airoha/airoha_eth.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/net/ethernet/airoha/airoha_eth.c b/drivers/net/ethernet/airoha/airoha_eth.c index 833dd911980b..2fe1f39558b8 100644 --- a/drivers/net/ethernet/airoha/airoha_eth.c +++ b/drivers/net/ethernet/airoha/airoha_eth.c @@ -2022,8 +2022,12 @@ static void airoha_ethtool_get_mac_stats(struct net_device *dev, airoha_update_hw_stats(port); do { start = u64_stats_fetch_begin(&port->stats.syncp); + stats->FramesTransmittedOK = port->stats.tx_ok_pkts; + stats->OctetsTransmittedOK = port->stats.tx_ok_bytes; stats->MulticastFramesXmittedOK = port->stats.tx_multicast; stats->BroadcastFramesXmittedOK = port->stats.tx_broadcast; + stats->FramesReceivedOK = port->stats.rx_ok_pkts; + stats->OctetsReceivedOK = port->stats.rx_ok_bytes; stats->BroadcastFramesReceivedOK = port->stats.rx_broadcast; } while (u64_stats_fetch_retry(&port->stats.syncp, start)); } From fc4fed9054ef5b5269d4395dd9db36fe98fce9e3 Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Mon, 13 Oct 2025 16:29:42 +0200 Subject: [PATCH 0010/1381] net: airoha: Add get_link ethtool callback Set get_link ethtool callback to ethtool_op_get_link routine. Signed-off-by: Lorenzo Bianconi Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20251013-airoha-ethtool-improvements-v1-2-fdd1c6fc9be1@kernel.org Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/airoha/airoha_eth.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/airoha/airoha_eth.c b/drivers/net/ethernet/airoha/airoha_eth.c index 2fe1f39558b8..6effdda64380 100644 --- a/drivers/net/ethernet/airoha/airoha_eth.c +++ b/drivers/net/ethernet/airoha/airoha_eth.c @@ -2770,6 +2770,7 @@ static const struct ethtool_ops airoha_ethtool_ops = { .get_drvinfo = airoha_ethtool_get_drvinfo, .get_eth_mac_stats = airoha_ethtool_get_mac_stats, .get_rmon_stats = airoha_ethtool_get_rmon_stats, + .get_link = ethtool_op_get_link, }; static int airoha_metadata_dst_alloc(struct airoha_gdm_port *port) From 6378e25ee1ca2ed687eee78eff7bd588d52a4e14 Mon Sep 17 00:00:00 2001 From: Frank Li Date: Fri, 10 Oct 2025 14:34:17 -0400 Subject: [PATCH 0011/1381] dt-bindings: net: dsa: nxp,sja1105: Add optional clock Add optional clock for OSC_IN and fix the below CHECK_DTBS warnings: arch/arm/boot/dts/nxp/imx/imx6qp-prtwd3.dtb: switch@0 (nxp,sja1105q): Unevaluated properties are not allowed ('clocks' was unexpected) Signed-off-by: Frank Li Acked-by: Conor Dooley Reviewed-by: Vladimir Oltean Link: https://patch.msgid.link/20251010183418.2179063-1-Frank.Li@nxp.com Signed-off-by: Jakub Kicinski --- Documentation/devicetree/bindings/net/dsa/nxp,sja1105.yaml | 3 +++ 1 file changed, 3 insertions(+) diff --git a/Documentation/devicetree/bindings/net/dsa/nxp,sja1105.yaml b/Documentation/devicetree/bindings/net/dsa/nxp,sja1105.yaml index e9dd914b0734..607b7fe8d28e 100644 --- a/Documentation/devicetree/bindings/net/dsa/nxp,sja1105.yaml +++ b/Documentation/devicetree/bindings/net/dsa/nxp,sja1105.yaml @@ -41,6 +41,9 @@ properties: therefore discouraged. maxItems: 1 + clocks: + maxItems: 1 + spi-cpha: true spi-cpol: true From 1c51450f1afff1e7419797720df3fbd9ccbf610c Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 13 Oct 2025 14:59:26 +0000 Subject: [PATCH 0012/1381] tcp: better handle TCP_TX_DELAY on established flows Some applications uses TCP_TX_DELAY socket option after TCP flow is established. Some metrics need to be updated, otherwise TCP might take time to adapt to the new (emulated) RTT. This patch adjusts tp->srtt_us, tp->rtt_min, icsk_rto and sk->sk_pacing_rate. This is best effort, and for instance icsk_rto is reset without taking backoff into account. Signed-off-by: Eric Dumazet Link: https://patch.msgid.link/20251013145926.833198-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- include/net/tcp.h | 2 ++ net/ipv4/tcp.c | 31 +++++++++++++++++++++++++++---- net/ipv4/tcp_input.c | 4 ++-- 3 files changed, 31 insertions(+), 6 deletions(-) diff --git a/include/net/tcp.h b/include/net/tcp.h index 5ca230ed526a..1e547138f4fb 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -461,6 +461,8 @@ enum skb_drop_reason tcp_child_process(struct sock *parent, struct sock *child, void tcp_enter_loss(struct sock *sk); void tcp_cwnd_reduction(struct sock *sk, int newly_acked_sacked, int newly_lost, int flag); void tcp_clear_retrans(struct tcp_sock *tp); +void tcp_update_pacing_rate(struct sock *sk); +void tcp_set_rto(struct sock *sk); void tcp_update_metrics(struct sock *sk); void tcp_init_metrics(struct sock *sk); void tcp_metrics_init(void); diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 8a18aeca7ab0..4d720aa09a4c 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -3583,9 +3583,12 @@ static int tcp_repair_options_est(struct sock *sk, sockptr_t optbuf, DEFINE_STATIC_KEY_FALSE(tcp_tx_delay_enabled); EXPORT_IPV6_MOD(tcp_tx_delay_enabled); -static void tcp_enable_tx_delay(void) +static void tcp_enable_tx_delay(struct sock *sk, int val) { - if (!static_branch_unlikely(&tcp_tx_delay_enabled)) { + struct tcp_sock *tp = tcp_sk(sk); + s32 delta = (val - tp->tcp_tx_delay) << 3; + + if (val && !static_branch_unlikely(&tcp_tx_delay_enabled)) { static int __tcp_tx_delay_enabled = 0; if (cmpxchg(&__tcp_tx_delay_enabled, 0, 1) == 0) { @@ -3593,6 +3596,22 @@ static void tcp_enable_tx_delay(void) pr_info("TCP_TX_DELAY enabled\n"); } } + /* If we change tcp_tx_delay on a live flow, adjust tp->srtt_us, + * tp->rtt_min, icsk_rto and sk->sk_pacing_rate. + * This is best effort. + */ + if (delta && sk->sk_state == TCP_ESTABLISHED) { + s64 srtt = (s64)tp->srtt_us + delta; + + tp->srtt_us = clamp_t(s64, srtt, 1, ~0U); + + /* Note: does not deal with non zero icsk_backoff */ + tcp_set_rto(sk); + + minmax_reset(&tp->rtt_min, tcp_jiffies32, ~0U); + + tcp_update_pacing_rate(sk); + } } /* When set indicates to always queue non-full frames. Later the user clears @@ -4119,8 +4138,12 @@ int do_tcp_setsockopt(struct sock *sk, int level, int optname, tp->recvmsg_inq = val; break; case TCP_TX_DELAY: - if (val) - tcp_enable_tx_delay(); + /* tp->srtt_us is u32, and is shifted by 3 */ + if (val < 0 || val >= (1U << (31 - 3))) { + err = -EINVAL; + break; + } + tcp_enable_tx_delay(sk, val); WRITE_ONCE(tp->tcp_tx_delay, val); break; default: diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 31ea5af49f2d..8fc97f4d8a6b 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -1095,7 +1095,7 @@ static void tcp_rtt_estimator(struct sock *sk, long mrtt_us) tp->srtt_us = max(1U, srtt); } -static void tcp_update_pacing_rate(struct sock *sk) +void tcp_update_pacing_rate(struct sock *sk) { const struct tcp_sock *tp = tcp_sk(sk); u64 rate; @@ -1132,7 +1132,7 @@ static void tcp_update_pacing_rate(struct sock *sk) /* Calculate rto without backoff. This is the second half of Van Jacobson's * routine referred to above. */ -static void tcp_set_rto(struct sock *sk) +void tcp_set_rto(struct sock *sk) { const struct tcp_sock *tp = tcp_sk(sk); /* Old crap is replaced with new one. 8) From 6ddb811a579f87b8506344020002d396f814f7c8 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 13 Oct 2025 15:22:31 +0000 Subject: [PATCH 0013/1381] net: add SK_WMEM_ALLOC_BIAS constant sk->sk_wmem_alloc is initialized to 1, and sk_wmem_alloc_get() takes care of this initial value. Add SK_WMEM_ALLOC_BIAS define to not spread this magic value. Signed-off-by: Eric Dumazet Reviewed-by: Neal Cardwell Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20251013152234.842065-2-edumazet@google.com Signed-off-by: Jakub Kicinski --- include/net/sock.h | 3 ++- net/atm/common.c | 2 +- net/core/sock.c | 5 ++--- 3 files changed, 5 insertions(+), 5 deletions(-) diff --git a/include/net/sock.h b/include/net/sock.h index 60bcb13f045c..2794bc5c5654 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -2303,6 +2303,7 @@ static inline int skb_copy_to_page_nocache(struct sock *sk, struct iov_iter *fro return 0; } +#define SK_WMEM_ALLOC_BIAS 1 /** * sk_wmem_alloc_get - returns write allocations * @sk: socket @@ -2311,7 +2312,7 @@ static inline int skb_copy_to_page_nocache(struct sock *sk, struct iov_iter *fro */ static inline int sk_wmem_alloc_get(const struct sock *sk) { - return refcount_read(&sk->sk_wmem_alloc) - 1; + return refcount_read(&sk->sk_wmem_alloc) - SK_WMEM_ALLOC_BIAS; } /** diff --git a/net/atm/common.c b/net/atm/common.c index 881c7f259dbd..cecc71a8bee1 100644 --- a/net/atm/common.c +++ b/net/atm/common.c @@ -157,7 +157,7 @@ int vcc_create(struct net *net, struct socket *sock, int protocol, int family, i memset(&vcc->local, 0, sizeof(struct sockaddr_atmsvc)); memset(&vcc->remote, 0, sizeof(struct sockaddr_atmsvc)); vcc->qos.txtp.max_sdu = 1 << 16; /* for meta VCs */ - refcount_set(&sk->sk_wmem_alloc, 1); + refcount_set(&sk->sk_wmem_alloc, SK_WMEM_ALLOC_BIAS); atomic_set(&sk->sk_rmem_alloc, 0); vcc->push = NULL; vcc->pop = NULL; diff --git a/net/core/sock.c b/net/core/sock.c index dc03d4b5909a..542cfa16ee12 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -2313,7 +2313,7 @@ struct sock *sk_alloc(struct net *net, int family, gfp_t priority, } sock_net_set(sk, net); - refcount_set(&sk->sk_wmem_alloc, 1); + refcount_set(&sk->sk_wmem_alloc, SK_WMEM_ALLOC_BIAS); mem_cgroup_sk_alloc(sk); cgroup_sk_alloc(&sk->sk_cgrp_data); @@ -2494,8 +2494,7 @@ struct sock *sk_clone_lock(const struct sock *sk, const gfp_t priority) atomic_set(&newsk->sk_rmem_alloc, 0); - /* sk_wmem_alloc set to one (see sk_free() and sock_wfree()) */ - refcount_set(&newsk->sk_wmem_alloc, 1); + refcount_set(&newsk->sk_wmem_alloc, SK_WMEM_ALLOC_BIAS); atomic_set(&newsk->sk_omem_alloc, 0); sk_init_common(newsk); From d365c9bca35cdeb534aac279c81d1fc9730bb100 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 13 Oct 2025 15:22:32 +0000 Subject: [PATCH 0014/1381] net: control skb->ooo_okay from skb_set_owner_w() 15 years after Tom Herbert added skb->ooo_okay, only TCP transport benefits from it. We can support other transports directly from skb_set_owner_w(). If no other TX packet for this socket is in a host queue (qdisc, NIC queue) there is no risk of self-inflicted reordering, we can set skb->ooo_okay. This allows netdev_pick_tx() to choose a TX queue based on XPS settings, instead of reusing the queue chosen at the time the first packet was sent for connected sockets. Tested: 500 concurrent UDP_RR connected UDP flows, host with 32 TX queues, 512 cpus, XPS setup. super_netperf 500 -t UDP_RR -H -l 1000 -- -r 100,100 -Nn & This patch saves between 10% and 20% of cycles, depending on how process scheduler migrates threads among cpus. Using following bpftrace script, we can see the effect on Qdisc/NIC tx queues being better used (less cache line misses). bpftrace -e ' k:__dev_queue_xmit { @start[cpu] = nsecs; } kr:__dev_queue_xmit { if (@start[cpu]) { $delay = nsecs - @start[cpu]; delete(@start[cpu]); @__dev_queue_xmit_ns = hist($delay); } } END { clear(@start); }' Before: @__dev_queue_xmit_ns: [128, 256) 6 | | [256, 512) 116283 | | [512, 1K) 1888205 |@@@@@@@@@@@ | [1K, 2K) 8106167 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ | [2K, 4K) 8699293 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@| [4K, 8K) 2600676 |@@@@@@@@@@@@@@@ | [8K, 16K) 721688 |@@@@ | [16K, 32K) 122995 | | [32K, 64K) 10639 | | [64K, 128K) 119 | | [128K, 256K) 1 | | After: @__dev_queue_xmit_ns: [128, 256) 3 | | [256, 512) 651112 |@@ | [512, 1K) 8109938 |@@@@@@@@@@@@@@@@@@@@@@@@@@ | [1K, 2K) 16081031 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@| [2K, 4K) 2411692 |@@@@@@@ | [4K, 8K) 98994 | | [8K, 16K) 1536 | | [16K, 32K) 587 | | [32K, 64K) 2 | | Signed-off-by: Eric Dumazet Reviewed-by: Neal Cardwell Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20251013152234.842065-3-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/core/sock.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) diff --git a/net/core/sock.c b/net/core/sock.c index 542cfa16ee12..08ae20069b6d 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -2694,6 +2694,8 @@ void __sock_wfree(struct sk_buff *skb) void skb_set_owner_w(struct sk_buff *skb, struct sock *sk) { + int old_wmem; + skb_orphan(skb); #ifdef CONFIG_INET if (unlikely(!sk_fullsock(sk))) @@ -2707,7 +2709,15 @@ void skb_set_owner_w(struct sk_buff *skb, struct sock *sk) * is enough to guarantee sk_free() won't free this sock until * all in-flight packets are completed */ - refcount_add(skb->truesize, &sk->sk_wmem_alloc); + __refcount_add(skb->truesize, &sk->sk_wmem_alloc, &old_wmem); + + /* (old_wmem == SK_WMEM_ALLOC_BIAS) if no other TX packet for this socket + * is in a host queue (qdisc, NIC queue). + * Set skb->ooo_okay so that netdev_pick_tx() can choose a TX queue + * based on XPS for better performance. + * Otherwise clear ooo_okay to not risk Out Of Order delivery. + */ + skb->ooo_okay = (old_wmem == SK_WMEM_ALLOC_BIAS); } EXPORT_SYMBOL(skb_set_owner_w); From 2ddef3462b3a5d62e5485e22ce128a5c02276438 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 13 Oct 2025 15:22:33 +0000 Subject: [PATCH 0015/1381] net: add /proc/sys/net/core/txq_reselection_ms control Add a new sysctl to control how often a queue reselection can happen even if a flow has a persistent queue of skbs in a Qdisc or NIC queue. A value of zero means the feature is disabled. Default is 1000 (1 second). This sysctl is used in the following patch. Signed-off-by: Eric Dumazet Reviewed-by: Neal Cardwell Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20251013152234.842065-4-edumazet@google.com Signed-off-by: Jakub Kicinski --- Documentation/admin-guide/sysctl/net.rst | 17 +++++++++++++++++ include/net/netns/core.h | 1 + net/core/net_namespace.c | 1 + net/core/sysctl_net_core.c | 7 +++++++ 4 files changed, 26 insertions(+) diff --git a/Documentation/admin-guide/sysctl/net.rst b/Documentation/admin-guide/sysctl/net.rst index 2ef50828aff1..40749b3cd356 100644 --- a/Documentation/admin-guide/sysctl/net.rst +++ b/Documentation/admin-guide/sysctl/net.rst @@ -406,6 +406,23 @@ to SOCK_TXREHASH_DEFAULT (i. e. not overridden by setsockopt). If set to 1 (default), hash rethink is performed on listening socket. If set to 0, hash rethink is not performed. +txq_reselection_ms +------------------ + +Controls how often (in ms) a busy connected flow can select another tx queue. + +A resection is desirable when/if user thread has migrated and XPS +would select a different queue. Same can occur without XPS +if the flow hash has changed. + +But switching txq can introduce reorders, especially if the +old queue is under high pressure. Modern TCP stacks deal +well with reorders if they happen not too often. + +To disable this feature, set the value to 0. + +Default : 1000 + gro_normal_batch ---------------- diff --git a/include/net/netns/core.h b/include/net/netns/core.h index 9b36f0ff0c20..cb9c3e4cd738 100644 --- a/include/net/netns/core.h +++ b/include/net/netns/core.h @@ -13,6 +13,7 @@ struct netns_core { struct ctl_table_header *sysctl_hdr; int sysctl_somaxconn; + int sysctl_txq_reselection; int sysctl_optmem_max; u8 sysctl_txrehash; u8 sysctl_tstamp_allow_data; diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c index b0e0f22d7b21..adcfef55a66f 100644 --- a/net/core/net_namespace.c +++ b/net/core/net_namespace.c @@ -395,6 +395,7 @@ static __net_init void preinit_net_sysctl(struct net *net) net->core.sysctl_optmem_max = 128 * 1024; net->core.sysctl_txrehash = SOCK_TXREHASH_ENABLED; net->core.sysctl_tstamp_allow_data = 1; + net->core.sysctl_txq_reselection = msecs_to_jiffies(1000); } /* init code that must occur even if setup_net() is not called. */ diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c index 8cf04b57ade1..f79137826d7f 100644 --- a/net/core/sysctl_net_core.c +++ b/net/core/sysctl_net_core.c @@ -667,6 +667,13 @@ static struct ctl_table netns_core_table[] = { .extra2 = SYSCTL_ONE, .proc_handler = proc_dou8vec_minmax, }, + { + .procname = "txq_reselection_ms", + .data = &init_net.core.sysctl_txq_reselection, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_ms_jiffies, + }, { .procname = "tstamp_allow_data", .data = &init_net.core.sysctl_tstamp_allow_data, From 4a7708443dec13b074bc43855f494358fedbd3c0 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 13 Oct 2025 15:22:34 +0000 Subject: [PATCH 0016/1381] net: allow busy connected flows to switch tx queues This is a followup of commit 726e9e8b94b9 ("tcp: refine skb->ooo_okay setting") and of prior commit in this series ("net: control skb->ooo_okay from skb_set_owner_w()") skb->ooo_okay might never be set for bulk flows that always have at least one skb in a qdisc queue of NIC queue, especially if TX completion is delayed because of a stressed cpu. The so-called "strange attractors" has caused many performance issues (see for instance 9b462d02d6dd ("tcp: TCP Small Queues and strange attractors")), we need to do better. We have tried very hard to avoid reorders because TCP was not dealing with them nicely a decade ago. Use the new net.core.txq_reselection_ms sysctl to let flows follow XPS and select a more efficient queue. After this patch, we no longer have to make sure threads are pinned to cpus, they now can be migrated without adding too much spinlock/qdisc/TX completion pressure anymore. TX completion part was problematic, because it added false sharing on various socket fields, but also added false sharing and spinlock contention in mm layers. Calling skb_orphan() from ndo_start_xmit() is not an option unfortunately. Note for later: 1) move sk->sk_tx_queue_mapping closer to sk_tx_queue_mapping_jiffies for better cache locality. 2) Study if 9b462d02d6dd ("tcp: TCP Small Queues and strange attractors") could be revised. Tested: Used a host with 32 TX queues, shared by groups of 8 cores. XPS setup : echo ff >/sys/class/net/eth1/queue/tx-0/xps_cpus echo ff00 >/sys/class/net/eth1/queue/tx-1/xps_cpus echo ff0000 >/sys/class/net/eth1/queue/tx-2/xps_cpus echo ff000000 >/sys/class/net/eth1/queue/tx-3/xps_cpus echo ff,00000000 >/sys/class/net/eth1/queue/tx-4/xps_cpus echo ff00,00000000 >/sys/class/net/eth1/queue/tx-5/xps_cpus echo ff0000,00000000 >/sys/class/net/eth1/queue/tx-6/xps_cpus echo ff000000,00000000 >/sys/class/net/eth1/queue/tx-7/xps_cpus ... Launched a tcp_stream with 15 threads and 1000 flows, initially affined to core 0-15 taskset -c 0-15 tcp_stream -T15 -F1000 -l1000 -c -H target_host Checked that only queues 0 and 1 are used as instructed by XPS : tc -s qdisc show dev eth1|grep backlog|grep -v "backlog 0b 0p" backlog 123489410b 1890p backlog 69809026b 1064p backlog 52401054b 805p Then force each thread to run on cpu 1,9,17,25,33,41,49,57,65,73,81,89,97,105,113,121 C=1;PID=`pidof tcp_stream`;for P in `ls /proc/$PID/task`; do taskset -pc $C $P; C=$(($C + 8));done Set txq_reselection_ms to 1000 echo 1000 > /proc/sys/net/core/txq_reselection_ms Check that the flows have migrated nicely: tc -s qdisc show dev eth1|grep backlog|grep -v "backlog 0b 0p" backlog 130508314b 1916p backlog 8584380b 126p backlog 8584380b 126p backlog 8379990b 123p backlog 8584380b 126p backlog 8487484b 125p backlog 8584380b 126p backlog 8448120b 124p backlog 8584380b 126p backlog 8720640b 128p backlog 8856900b 130p backlog 8584380b 126p backlog 8652510b 127p backlog 8448120b 124p backlog 8516250b 125p backlog 7834950b 115p Signed-off-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20251013152234.842065-5-edumazet@google.com Signed-off-by: Jakub Kicinski --- include/net/sock.h | 26 ++++++++++++-------------- net/core/dev.c | 29 +++++++++++++++++++++++++++-- 2 files changed, 39 insertions(+), 16 deletions(-) diff --git a/include/net/sock.h b/include/net/sock.h index 2794bc5c5654..f0d00928db9e 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -313,6 +313,7 @@ struct sk_filter; * @sk_bind_phc: SO_TIMESTAMPING bind PHC index of PTP virtual clock * for timestamping * @sk_tskey: counter to disambiguate concurrent tstamp requests + * @sk_tx_queue_mapping_jiffies: time in jiffies of last @sk_tx_queue_mapping refresh. * @sk_zckey: counter to order MSG_ZEROCOPY notifications * @sk_socket: Identd and reporting IO signals * @sk_user_data: RPC layer private data. Write-protected by @sk_callback_lock. @@ -485,6 +486,7 @@ struct sock { unsigned long sk_pacing_rate; /* bytes per second */ atomic_t sk_zckey; atomic_t sk_tskey; + unsigned long sk_tx_queue_mapping_jiffies; __cacheline_group_end(sock_write_tx); __cacheline_group_begin(sock_read_tx); @@ -1992,7 +1994,15 @@ static inline void sk_tx_queue_set(struct sock *sk, int tx_queue) /* Paired with READ_ONCE() in sk_tx_queue_get() and * other WRITE_ONCE() because socket lock might be not held. */ - WRITE_ONCE(sk->sk_tx_queue_mapping, tx_queue); + if (READ_ONCE(sk->sk_tx_queue_mapping) != tx_queue) { + WRITE_ONCE(sk->sk_tx_queue_mapping, tx_queue); + WRITE_ONCE(sk->sk_tx_queue_mapping_jiffies, jiffies); + return; + } + + /* Refresh sk_tx_queue_mapping_jiffies if too old. */ + if (time_is_before_jiffies(READ_ONCE(sk->sk_tx_queue_mapping_jiffies) + HZ)) + WRITE_ONCE(sk->sk_tx_queue_mapping_jiffies, jiffies); } #define NO_QUEUE_MAPPING USHRT_MAX @@ -2005,19 +2015,7 @@ static inline void sk_tx_queue_clear(struct sock *sk) WRITE_ONCE(sk->sk_tx_queue_mapping, NO_QUEUE_MAPPING); } -static inline int sk_tx_queue_get(const struct sock *sk) -{ - if (sk) { - /* Paired with WRITE_ONCE() in sk_tx_queue_clear() - * and sk_tx_queue_set(). - */ - int val = READ_ONCE(sk->sk_tx_queue_mapping); - - if (val != NO_QUEUE_MAPPING) - return val; - } - return -1; -} +int sk_tx_queue_get(const struct sock *sk); static inline void __sk_rx_queue_set(struct sock *sk, const struct sk_buff *skb, diff --git a/net/core/dev.c b/net/core/dev.c index a64cef2c537e..33e6101dbc45 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4591,6 +4591,32 @@ u16 dev_pick_tx_zero(struct net_device *dev, struct sk_buff *skb, } EXPORT_SYMBOL(dev_pick_tx_zero); +int sk_tx_queue_get(const struct sock *sk) +{ + int resel, val; + + if (!sk) + return -1; + /* Paired with WRITE_ONCE() in sk_tx_queue_clear() + * and sk_tx_queue_set(). + */ + val = READ_ONCE(sk->sk_tx_queue_mapping); + + if (val == NO_QUEUE_MAPPING) + return -1; + + if (!sk_fullsock(sk)) + return val; + + resel = READ_ONCE(sock_net(sk)->core.sysctl_txq_reselection); + if (resel && time_is_before_jiffies( + READ_ONCE(sk->sk_tx_queue_mapping_jiffies) + resel)) + return -1; + + return val; +} +EXPORT_SYMBOL(sk_tx_queue_get); + u16 netdev_pick_tx(struct net_device *dev, struct sk_buff *skb, struct net_device *sb_dev) { @@ -4606,8 +4632,7 @@ u16 netdev_pick_tx(struct net_device *dev, struct sk_buff *skb, if (new_index < 0) new_index = skb_tx_hash(dev, sb_dev, skb); - if (queue_index != new_index && sk && - sk_fullsock(sk) && + if (sk && sk_fullsock(sk) && rcu_access_pointer(sk->sk_dst_cache)) sk_tx_queue_set(sk, new_index); From a8e846b8d93de748485653dd8a6a8efd8f5d7613 Mon Sep 17 00:00:00 2001 From: Abhishek Rawal Date: Tue, 14 Oct 2025 11:22:33 +0530 Subject: [PATCH 0017/1381] r8152: Advertise software timestamp information. Driver calls skb_tx_timestamp(skb) in rtl8152_start_xmit(), but does not advertise the capability in ethtool. Advertise software timestamp capabilities on struct ethtool_ops. Signed-off-by: Abhishek Rawal Reviewed-by: Jamie Bainbridge Reviewed-by: Vadim Fedorenko Link: https://patch.msgid.link/20251014055234.46527-1-rawal.abhishek92@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/usb/r8152.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/usb/r8152.c b/drivers/net/usb/r8152.c index 44cba7acfe7d..f896e9f28c3b 100644 --- a/drivers/net/usb/r8152.c +++ b/drivers/net/usb/r8152.c @@ -9311,6 +9311,7 @@ static const struct ethtool_ops ops = { .set_ringparam = rtl8152_set_ringparam, .get_pauseparam = rtl8152_get_pauseparam, .set_pauseparam = rtl8152_set_pauseparam, + .get_ts_info = ethtool_op_get_ts_info, }; static int rtl8152_ioctl(struct net_device *netdev, struct ifreq *rq, int cmd) From 378e6523ebb1e80b3955b7675cfe40b07028d085 Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Tue, 14 Oct 2025 08:02:47 +0200 Subject: [PATCH 0018/1381] net: bcmgenet: remove unused platform code This effectively reverts b0ba512e25d7 ("net: bcmgenet: enable driver to work without a device tree"). There has never been an in-tree user of struct bcmgenet_platform_data, all devices use OF or ACPI. Signed-off-by: Heiner Kallweit Reviewed-by: Simon Horman Reviewed-by: Florian Fainelli Link: https://patch.msgid.link/108b4e64-55d4-4b4e-9a11-3c810c319d66@gmail.com Signed-off-by: Jakub Kicinski --- MAINTAINERS | 1 - .../net/ethernet/broadcom/genet/bcmgenet.c | 20 ++--- drivers/net/ethernet/broadcom/genet/bcmmii.c | 75 +------------------ include/linux/platform_data/bcmgenet.h | 19 ----- 4 files changed, 7 insertions(+), 108 deletions(-) delete mode 100644 include/linux/platform_data/bcmgenet.h diff --git a/MAINTAINERS b/MAINTAINERS index 3a27901781c2..4c4b519171f3 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -5123,7 +5123,6 @@ F: Documentation/devicetree/bindings/net/brcm,unimac-mdio.yaml F: drivers/net/ethernet/broadcom/genet/ F: drivers/net/ethernet/broadcom/unimac.h F: drivers/net/mdio/mdio-bcm-unimac.c -F: include/linux/platform_data/bcmgenet.h F: include/linux/platform_data/mdio-bcm-unimac.h BROADCOM IPROC ARM ARCHITECTURE diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.c b/drivers/net/ethernet/broadcom/genet/bcmgenet.c index 98971ae4f87d..d99ef92feb82 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c +++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c @@ -35,7 +35,6 @@ #include #include #include -#include #include @@ -3926,7 +3925,6 @@ MODULE_DEVICE_TABLE(of, bcmgenet_match); static int bcmgenet_probe(struct platform_device *pdev) { - struct bcmgenet_platform_data *pd = pdev->dev.platform_data; const struct bcmgenet_plat_data *pdata; struct bcmgenet_priv *priv; struct net_device *dev; @@ -4010,9 +4008,6 @@ static int bcmgenet_probe(struct platform_device *pdev) priv->version = pdata->version; priv->dma_max_burst_length = pdata->dma_max_burst_length; priv->flags = pdata->flags; - } else { - priv->version = pd->genet_version; - priv->dma_max_burst_length = DMA_MAX_BURST_LENGTH; } priv->clk = devm_clk_get_optional(&priv->pdev->dev, "enet"); @@ -4062,16 +4057,13 @@ static int bcmgenet_probe(struct platform_device *pdev) if (device_get_phy_mode(&pdev->dev) == PHY_INTERFACE_MODE_INTERNAL) bcmgenet_power_up(priv, GENET_POWER_PASSIVE); - if (pd && !IS_ERR_OR_NULL(pd->mac_address)) - eth_hw_addr_set(dev, pd->mac_address); - else - if (device_get_ethdev_address(&pdev->dev, dev)) - if (has_acpi_companion(&pdev->dev)) { - u8 addr[ETH_ALEN]; + if (device_get_ethdev_address(&pdev->dev, dev)) + if (has_acpi_companion(&pdev->dev)) { + u8 addr[ETH_ALEN]; - bcmgenet_get_hw_addr(priv, addr); - eth_hw_addr_set(dev, addr); - } + bcmgenet_get_hw_addr(priv, addr); + eth_hw_addr_set(dev, addr); + } if (!is_valid_ether_addr(dev->dev_addr)) { dev_warn(&pdev->dev, "using random Ethernet MAC\n"); diff --git a/drivers/net/ethernet/broadcom/genet/bcmmii.c b/drivers/net/ethernet/broadcom/genet/bcmmii.c index 573e8b279e52..38f854b94a79 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmmii.c +++ b/drivers/net/ethernet/broadcom/genet/bcmmii.c @@ -20,7 +20,6 @@ #include #include #include -#include #include #include "bcmgenet.h" @@ -436,23 +435,6 @@ static struct device_node *bcmgenet_mii_of_find_mdio(struct bcmgenet_priv *priv) return priv->mdio_dn; } -static void bcmgenet_mii_pdata_init(struct bcmgenet_priv *priv, - struct unimac_mdio_pdata *ppd) -{ - struct device *kdev = &priv->pdev->dev; - struct bcmgenet_platform_data *pd = kdev->platform_data; - - if (pd->phy_interface != PHY_INTERFACE_MODE_MOCA && pd->mdio_enabled) { - /* - * Internal or external PHY with MDIO access - */ - if (pd->phy_address >= 0 && pd->phy_address < PHY_MAX_ADDR) - ppd->phy_mask = 1 << pd->phy_address; - else - ppd->phy_mask = 0; - } -} - static int bcmgenet_mii_wait(void *wait_func_data) { struct bcmgenet_priv *priv = wait_func_data; @@ -467,7 +449,6 @@ static int bcmgenet_mii_wait(void *wait_func_data) static int bcmgenet_mii_register(struct bcmgenet_priv *priv) { struct platform_device *pdev = priv->pdev; - struct bcmgenet_platform_data *pdata = pdev->dev.platform_data; struct device_node *dn = pdev->dev.of_node; struct unimac_mdio_pdata ppd; struct platform_device *ppdev; @@ -511,8 +492,6 @@ static int bcmgenet_mii_register(struct bcmgenet_priv *priv) ppdev->dev.parent = &pdev->dev; if (dn) ppdev->dev.of_node = bcmgenet_mii_of_find_mdio(priv); - else if (pdata) - bcmgenet_mii_pdata_init(priv, &ppd); else ppd.phy_mask = ~0; @@ -594,58 +573,6 @@ static int bcmgenet_mii_of_init(struct bcmgenet_priv *priv) return 0; } -static int bcmgenet_mii_pd_init(struct bcmgenet_priv *priv) -{ - struct device *kdev = &priv->pdev->dev; - struct bcmgenet_platform_data *pd = kdev->platform_data; - char phy_name[MII_BUS_ID_SIZE + 3]; - char mdio_bus_id[MII_BUS_ID_SIZE]; - struct phy_device *phydev; - - snprintf(mdio_bus_id, MII_BUS_ID_SIZE, "%s-%d", - UNIMAC_MDIO_DRV_NAME, priv->pdev->id); - - if (pd->phy_interface != PHY_INTERFACE_MODE_MOCA && pd->mdio_enabled) { - snprintf(phy_name, MII_BUS_ID_SIZE, PHY_ID_FMT, - mdio_bus_id, pd->phy_address); - - /* - * Internal or external PHY with MDIO access - */ - phydev = phy_attach(priv->dev, phy_name, pd->phy_interface); - if (IS_ERR(phydev)) { - dev_err(kdev, "failed to register PHY device\n"); - return PTR_ERR(phydev); - } - } else { - /* - * MoCA port or no MDIO access. - * Use fixed PHY to represent the link layer. - */ - struct fixed_phy_status fphy_status = { - .link = 1, - .speed = pd->phy_speed, - .duplex = pd->phy_duplex, - .pause = 0, - .asym_pause = 0, - }; - - phydev = fixed_phy_register(&fphy_status, NULL); - if (IS_ERR(phydev)) { - dev_err(kdev, "failed to register fixed PHY device\n"); - return PTR_ERR(phydev); - } - - /* Make sure we initialize MoCA PHYs with a link down */ - phydev->link = 0; - - } - - priv->phy_interface = pd->phy_interface; - - return 0; -} - static int bcmgenet_mii_bus_init(struct bcmgenet_priv *priv) { struct device *kdev = &priv->pdev->dev; @@ -656,7 +583,7 @@ static int bcmgenet_mii_bus_init(struct bcmgenet_priv *priv) else if (has_acpi_companion(kdev)) return bcmgenet_phy_interface_init(priv); else - return bcmgenet_mii_pd_init(priv); + return -EINVAL; } int bcmgenet_mii_init(struct net_device *dev) diff --git a/include/linux/platform_data/bcmgenet.h b/include/linux/platform_data/bcmgenet.h deleted file mode 100644 index d8f8738629d2..000000000000 --- a/include/linux/platform_data/bcmgenet.h +++ /dev/null @@ -1,19 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0 */ -#ifndef __LINUX_PLATFORM_DATA_BCMGENET_H__ -#define __LINUX_PLATFORM_DATA_BCMGENET_H__ - -#include -#include -#include - -struct bcmgenet_platform_data { - bool mdio_enabled; - phy_interface_t phy_interface; - int phy_address; - int phy_speed; - int phy_duplex; - u8 mac_address[ETH_ALEN]; - int genet_version; -}; - -#endif From 4077d7fb27be990a8ddcff9b49f7e1788a960f3a Mon Sep 17 00:00:00 2001 From: "Dr. David Alan Gilbert" Date: Sat, 11 Oct 2025 01:10:38 +0100 Subject: [PATCH 0019/1381] wifi: wcn36xx: Remove unused wcn36xx_smd_update_scan_params wcn36xx_smd_update_scan_params() last use was removed in 2020 by commit 5973a2947430 ("wcn36xx: Fix software-driven scan") Remove it. This leaves the wcn36xx_hal_update_scan_params_req_ex and wcn36xx_hal_update_scan_params_resp structs unused. Remove them, together with the unused wcn36xx_hal_update_scan_params_req. Signed-off-by: Dr. David Alan Gilbert Acked-by: Loic Poulain Link: https://patch.msgid.link/20251011001038.352393-1-linux@treblig.org Signed-off-by: Jeff Johnson --- drivers/net/wireless/ath/wcn36xx/hal.h | 74 -------------------------- drivers/net/wireless/ath/wcn36xx/smd.c | 60 --------------------- drivers/net/wireless/ath/wcn36xx/smd.h | 1 - 3 files changed, 135 deletions(-) diff --git a/drivers/net/wireless/ath/wcn36xx/hal.h b/drivers/net/wireless/ath/wcn36xx/hal.h index d3a9d00e65e1..ef9ea4ff891b 100644 --- a/drivers/net/wireless/ath/wcn36xx/hal.h +++ b/drivers/net/wireless/ath/wcn36xx/hal.h @@ -4484,80 +4484,6 @@ struct set_rssi_filter_resp { u32 status; }; -/* Update scan params - sent from host to PNO to be used during PNO - * scanningx */ -struct wcn36xx_hal_update_scan_params_req { - - struct wcn36xx_hal_msg_header header; - - /* Host setting for 11d */ - u8 dot11d_enabled; - - /* Lets PNO know that host has determined the regulatory domain */ - u8 dot11d_resolved; - - /* Channels on which PNO is allowed to scan */ - u8 channel_count; - u8 channels[WCN36XX_HAL_PNO_MAX_NETW_CHANNELS]; - - /* Minimum channel time */ - u16 active_min_ch_time; - - /* Maximum channel time */ - u16 active_max_ch_time; - - /* Minimum channel time */ - u16 passive_min_ch_time; - - /* Maximum channel time */ - u16 passive_max_ch_time; - - /* Cb State */ - enum phy_chan_bond_state state; -} __packed; - -/* Update scan params - sent from host to PNO to be used during PNO - * scanningx */ -struct wcn36xx_hal_update_scan_params_req_ex { - - struct wcn36xx_hal_msg_header header; - - /* Host setting for 11d */ - u8 dot11d_enabled; - - /* Lets PNO know that host has determined the regulatory domain */ - u8 dot11d_resolved; - - /* Channels on which PNO is allowed to scan */ - u8 channel_count; - u8 channels[WCN36XX_HAL_PNO_MAX_NETW_CHANNELS_EX]; - - /* Minimum channel time */ - u16 active_min_ch_time; - - /* Maximum channel time */ - u16 active_max_ch_time; - - /* Minimum channel time */ - u16 passive_min_ch_time; - - /* Maximum channel time */ - u16 passive_max_ch_time; - - /* Cb State */ - enum phy_chan_bond_state state; -} __packed; - -/* Update scan params - sent from host to PNO to be used during PNO - * scanningx */ -struct wcn36xx_hal_update_scan_params_resp { - - struct wcn36xx_hal_msg_header header; - - /* status of the request */ - u32 status; -} __packed; - struct wcn36xx_hal_set_tx_per_tracking_req_msg { struct wcn36xx_hal_msg_header header; diff --git a/drivers/net/wireless/ath/wcn36xx/smd.c b/drivers/net/wireless/ath/wcn36xx/smd.c index 2cf86fc3f8fe..136acc414714 100644 --- a/drivers/net/wireless/ath/wcn36xx/smd.c +++ b/drivers/net/wireless/ath/wcn36xx/smd.c @@ -1127,66 +1127,6 @@ int wcn36xx_smd_process_ptt_msg(struct wcn36xx *wcn, return ret; } -static int wcn36xx_smd_update_scan_params_rsp(void *buf, size_t len) -{ - struct wcn36xx_hal_update_scan_params_resp *rsp; - - rsp = buf; - - /* Remove the PNO version bit */ - rsp->status &= (~(WCN36XX_FW_MSG_PNO_VERSION_MASK)); - - if (WCN36XX_FW_MSG_RESULT_SUCCESS != rsp->status) { - wcn36xx_warn("error response from update scan\n"); - return rsp->status; - } - - return 0; -} - -int wcn36xx_smd_update_scan_params(struct wcn36xx *wcn, - u8 *channels, size_t channel_count) -{ - struct wcn36xx_hal_update_scan_params_req_ex msg_body; - int ret; - - mutex_lock(&wcn->hal_mutex); - INIT_HAL_MSG(msg_body, WCN36XX_HAL_UPDATE_SCAN_PARAM_REQ); - - msg_body.dot11d_enabled = false; - msg_body.dot11d_resolved = true; - - msg_body.channel_count = channel_count; - memcpy(msg_body.channels, channels, channel_count); - msg_body.active_min_ch_time = 60; - msg_body.active_max_ch_time = 120; - msg_body.passive_min_ch_time = 60; - msg_body.passive_max_ch_time = 110; - msg_body.state = PHY_SINGLE_CHANNEL_CENTERED; - - PREPARE_HAL_BUF(wcn->hal_buf, msg_body); - - wcn36xx_dbg(WCN36XX_DBG_HAL, - "hal update scan params channel_count %d\n", - msg_body.channel_count); - - ret = wcn36xx_smd_send_and_wait(wcn, msg_body.header.len); - if (ret) { - wcn36xx_err("Sending hal_update_scan_params failed\n"); - goto out; - } - ret = wcn36xx_smd_update_scan_params_rsp(wcn->hal_buf, - wcn->hal_rsp_len); - if (ret) { - wcn36xx_err("hal_update_scan_params response failed err=%d\n", - ret); - goto out; - } -out: - mutex_unlock(&wcn->hal_mutex); - return ret; -} - static int wcn36xx_smd_add_sta_self_rsp(struct wcn36xx *wcn, struct ieee80211_vif *vif, void *buf, diff --git a/drivers/net/wireless/ath/wcn36xx/smd.h b/drivers/net/wireless/ath/wcn36xx/smd.h index 2c1ed9e570bf..4e39df5589b3 100644 --- a/drivers/net/wireless/ath/wcn36xx/smd.h +++ b/drivers/net/wireless/ath/wcn36xx/smd.h @@ -66,7 +66,6 @@ int wcn36xx_smd_finish_scan(struct wcn36xx *wcn, enum wcn36xx_hal_sys_mode mode, int wcn36xx_smd_init_scan(struct wcn36xx *wcn, enum wcn36xx_hal_sys_mode mode, struct ieee80211_vif *vif); -int wcn36xx_smd_update_scan_params(struct wcn36xx *wcn, u8 *channels, size_t channel_count); int wcn36xx_smd_start_hw_scan(struct wcn36xx *wcn, struct ieee80211_vif *vif, struct cfg80211_scan_request *req); int wcn36xx_smd_stop_hw_scan(struct wcn36xx *wcn); From f35a07a4842a88801d9182b1a76d178bfa616978 Mon Sep 17 00:00:00 2001 From: Kang Yang Date: Tue, 14 Oct 2025 19:07:57 +0800 Subject: [PATCH 0020/1381] wifi: ath10k: move recovery check logic into a new work MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Currently, ath10k has a recovery check logic. It will wait for the last recovery to finish by wait_for_completion_timeout(); But in SDIO scenarios, the recovery function may be invoked from interrupt context, where long blocking waits are undesirable and can lead to system instability. Additionally, Linux’s ordered workqueue processes one task at a time. If a previous recovery is still queued or executing, new triggers are ignored. This prevents accurate tracking of consecutive failures and delays transition to the WEDGED state. To address this, move the recovery check logic into a different workqueue. Tested-on: QCA6174 hw3.2 PCI WLAN.RM.4.4.1-00288-QCARMSWPZ-1 Tested-on: QCA6174 hw3.2 SDIO WLAN.RMH.4.4.1-00189 Fixes: c256a94d1b1b ("wifi: ath10k: shutdown driver when hardware is unreliable") Signed-off-by: Kang Yang Reviewed-by: Baochen Qiang Link: https://patch.msgid.link/20251014110757.155-1-kang.yang@oss.qualcomm.com Signed-off-by: Jeff Johnson --- drivers/net/wireless/ath/ath10k/core.c | 20 +++++++++----------- drivers/net/wireless/ath/ath10k/core.h | 2 +- drivers/net/wireless/ath/ath10k/mac.c | 2 +- 3 files changed, 11 insertions(+), 13 deletions(-) diff --git a/drivers/net/wireless/ath/ath10k/core.c b/drivers/net/wireless/ath/ath10k/core.c index 6f78f1752cd6..9ae3595fb698 100644 --- a/drivers/net/wireless/ath/ath10k/core.c +++ b/drivers/net/wireless/ath/ath10k/core.c @@ -3,7 +3,6 @@ * Copyright (c) 2005-2011 Atheros Communications Inc. * Copyright (c) 2011-2017 Qualcomm Atheros, Inc. * Copyright (c) 2018-2019, The Linux Foundation. All rights reserved. - * Copyright (c) 2021-2024 Qualcomm Innovation Center, Inc. All rights reserved. * Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. */ @@ -2493,8 +2492,9 @@ static int ath10k_init_hw_params(struct ath10k *ar) return 0; } -static bool ath10k_core_needs_recovery(struct ath10k *ar) +static void ath10k_core_recovery_check_work(struct work_struct *work) { + struct ath10k *ar = container_of(work, struct ath10k, recovery_check_work); long time_left; /* Sometimes the recovery will fail and then the next all recovery fail, @@ -2504,7 +2504,7 @@ static bool ath10k_core_needs_recovery(struct ath10k *ar) ath10k_err(ar, "consecutive fail %d times, will shutdown driver!", atomic_read(&ar->fail_cont_count)); ar->state = ATH10K_STATE_WEDGED; - return false; + return; } ath10k_dbg(ar, ATH10K_DBG_BOOT, "total recovery count: %d", ++ar->recovery_count); @@ -2518,27 +2518,24 @@ static bool ath10k_core_needs_recovery(struct ath10k *ar) ATH10K_RECOVERY_TIMEOUT_HZ); if (time_left) { ath10k_warn(ar, "previous recovery succeeded, skip this!\n"); - return false; + return; } /* Record the continuous recovery fail count when recovery failed. */ atomic_inc(&ar->fail_cont_count); /* Avoid having multiple recoveries at the same time. */ - return false; + return; } atomic_inc(&ar->pending_recovery); - - return true; + queue_work(ar->workqueue, &ar->restart_work); } void ath10k_core_start_recovery(struct ath10k *ar) { - if (!ath10k_core_needs_recovery(ar)) - return; - - queue_work(ar->workqueue, &ar->restart_work); + /* Use workqueue_aux to avoid blocking recovery tracking */ + queue_work(ar->workqueue_aux, &ar->recovery_check_work); } EXPORT_SYMBOL(ath10k_core_start_recovery); @@ -3734,6 +3731,7 @@ struct ath10k *ath10k_core_create(size_t priv_size, struct device *dev, INIT_WORK(&ar->register_work, ath10k_core_register_work); INIT_WORK(&ar->restart_work, ath10k_core_restart); + INIT_WORK(&ar->recovery_check_work, ath10k_core_recovery_check_work); INIT_WORK(&ar->set_coverage_class_work, ath10k_core_set_coverage_class_work); diff --git a/drivers/net/wireless/ath/ath10k/core.h b/drivers/net/wireless/ath/ath10k/core.h index 8c72ed386edb..859176fcb5a2 100644 --- a/drivers/net/wireless/ath/ath10k/core.h +++ b/drivers/net/wireless/ath/ath10k/core.h @@ -3,7 +3,6 @@ * Copyright (c) 2005-2011 Atheros Communications Inc. * Copyright (c) 2011-2017 Qualcomm Atheros, Inc. * Copyright (c) 2018-2019, The Linux Foundation. All rights reserved. - * Copyright (c) 2022 Qualcomm Innovation Center, Inc. All rights reserved. * Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. */ @@ -1208,6 +1207,7 @@ struct ath10k { struct work_struct register_work; struct work_struct restart_work; + struct work_struct recovery_check_work; struct work_struct bundle_tx_work; struct work_struct tx_complete_work; diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c index 154ac7a70982..da6f7957a0ae 100644 --- a/drivers/net/wireless/ath/ath10k/mac.c +++ b/drivers/net/wireless/ath/ath10k/mac.c @@ -3,7 +3,6 @@ * Copyright (c) 2005-2011 Atheros Communications Inc. * Copyright (c) 2011-2017 Qualcomm Atheros, Inc. * Copyright (c) 2018-2019, The Linux Foundation. All rights reserved. - * Copyright (c) 2021-2024 Qualcomm Innovation Center, Inc. All rights reserved. * Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. */ @@ -5428,6 +5427,7 @@ static void ath10k_stop(struct ieee80211_hw *hw, bool suspend) cancel_work_sync(&ar->set_coverage_class_work); cancel_delayed_work_sync(&ar->scan.timeout); cancel_work_sync(&ar->restart_work); + cancel_work_sync(&ar->recovery_check_work); } static int ath10k_config_ps(struct ath10k *ar) From 960fc268a9fc269190014773c81507e695bec3d4 Mon Sep 17 00:00:00 2001 From: Jeff Johnson Date: Fri, 10 Oct 2025 09:46:27 -0700 Subject: [PATCH 0021/1381] wifi: ath11k: Remove struct wmi_bcn_send_from_host_cmd struct wmi_bcn_send_from_host_cmd is unused, so remove it. Compile tested only. Link: https://patch.msgid.link/20251010-ath11k-nuke-wmi_bcn_send_from_host_cmd-v1-1-bfb5118d9018@oss.qualcomm.com Signed-off-by: Jeff Johnson --- drivers/net/wireless/ath/ath11k/wmi.h | 16 +--------------- 1 file changed, 1 insertion(+), 15 deletions(-) diff --git a/drivers/net/wireless/ath/ath11k/wmi.h b/drivers/net/wireless/ath/ath11k/wmi.h index 9fcffaa2f383..3c4885a12855 100644 --- a/drivers/net/wireless/ath/ath11k/wmi.h +++ b/drivers/net/wireless/ath/ath11k/wmi.h @@ -1,7 +1,7 @@ /* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved. - * Copyright (c) 2021-2025 Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. */ #ifndef ATH11K_WMI_H @@ -3463,20 +3463,6 @@ struct scan_cancel_param { u32 pdev_id; }; -struct wmi_bcn_send_from_host_cmd { - u32 tlv_header; - u32 vdev_id; - u32 data_len; - union { - u32 frag_ptr; - u32 frag_ptr_lo; - }; - u32 frame_ctrl; - u32 dtim_flag; - u32 bcn_antenna; - u32 frag_ptr_hi; -}; - #define WMI_CHAN_INFO_MODE GENMASK(5, 0) #define WMI_CHAN_INFO_HT40_PLUS BIT(6) #define WMI_CHAN_INFO_PASSIVE BIT(7) From d34a368be24d029544cc97feb87729a9f7984a78 Mon Sep 17 00:00:00 2001 From: Jeff Johnson Date: Fri, 10 Oct 2025 09:47:58 -0700 Subject: [PATCH 0022/1381] wifi: ath12k: Remove struct wmi_bcn_send_from_host_cmd struct wmi_bcn_send_from_host_cmd is unused, so remove it. Compile tested only. Link: https://patch.msgid.link/20251010-ath12k-nuke-wmi_bcn_send_from_host_cmd-v1-1-6f1172b77848@oss.qualcomm.com Signed-off-by: Jeff Johnson --- drivers/net/wireless/ath/ath12k/wmi.h | 16 +--------------- 1 file changed, 1 insertion(+), 15 deletions(-) diff --git a/drivers/net/wireless/ath/ath12k/wmi.h b/drivers/net/wireless/ath/ath12k/wmi.h index a8c3190e8ad9..d9fd6a6b708d 100644 --- a/drivers/net/wireless/ath/ath12k/wmi.h +++ b/drivers/net/wireless/ath/ath12k/wmi.h @@ -1,7 +1,7 @@ /* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* * Copyright (c) 2018-2021 The Linux Foundation. All rights reserved. - * Copyright (c) 2021-2025 Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. */ #ifndef ATH12K_WMI_H @@ -3609,20 +3609,6 @@ struct ath12k_wmi_scan_cancel_arg { u32 pdev_id; }; -struct wmi_bcn_send_from_host_cmd { - __le32 tlv_header; - __le32 vdev_id; - __le32 data_len; - union { - __le32 frag_ptr; - __le32 frag_ptr_lo; - }; - __le32 frame_ctrl; - __le32 dtim_flag; - __le32 bcn_antenna; - __le32 frag_ptr_hi; -}; - #define WMI_CHAN_INFO_MODE GENMASK(5, 0) #define WMI_CHAN_INFO_HT40_PLUS BIT(6) #define WMI_CHAN_INFO_PASSIVE BIT(7) From 596b911644cc19ecba0dbc9c92849fb59390e29a Mon Sep 17 00:00:00 2001 From: Baochen Qiang Date: Tue, 14 Oct 2025 10:30:20 +0800 Subject: [PATCH 0023/1381] wifi: ath11k: restore register window after global reset Hardware target implements an address space larger than that PCI BAR can map. In order to be able to access the whole target address space, the BAR space is split into 4 segments, of which the last 3, called windows, can be dynamically mapped to the desired area. This is achieved by updating window register with appropriate window value. Currently each time when accessing a register that beyond ATH11K_PCI_WINDOW_START, host calculates the window value and caches it after window update, this way next time when accessing a register falling in the same window, host knows that the window is already good hence no additional update needed. However this mechanism breaks after global reset is triggered in ath11k_pci_soc_global_reset(), because with global reset hardware resets window register hence the window is not properly mapped any more. Current host does nothing about this, as a result a subsequent register access may not work as expected if it falls in a window same as before. Although there is no obvious issue seen now, better to fix it to avoid future problem. The fix is done by restoring the window register after global reset. Tested-on: WCN6855 hw2.0 PCI WLAN.HSP.1.1-03125-QCAHSPSWPL_V1_V2_SILICONZ_LITE-3.6510.30 Fixes: d5c65159f289 ("ath11k: driver for Qualcomm IEEE 802.11ax devices") Signed-off-by: Baochen Qiang Reviewed-by: Vasanthakumar Thiagarajan Link: https://patch.msgid.link/20251014-ath11k-reset-window-cache-v1-1-b85271b111dd@oss.qualcomm.com Signed-off-by: Jeff Johnson --- drivers/net/wireless/ath/ath11k/pci.c | 20 +++++++++++++++++++- 1 file changed, 19 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/ath/ath11k/pci.c b/drivers/net/wireless/ath/ath11k/pci.c index d8655badd96d..7114eca8810d 100644 --- a/drivers/net/wireless/ath/ath11k/pci.c +++ b/drivers/net/wireless/ath/ath11k/pci.c @@ -1,7 +1,7 @@ // SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (c) 2019-2020 The Linux Foundation. All rights reserved. - * Copyright (c) 2021-2025 Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. */ #include @@ -177,6 +177,19 @@ static inline void ath11k_pci_select_static_window(struct ath11k_pci *ab_pci) ab_pci->ab->mem + ATH11K_PCI_WINDOW_REG_ADDRESS); } +static void ath11k_pci_restore_window(struct ath11k_base *ab) +{ + struct ath11k_pci *ab_pci = ath11k_pci_priv(ab); + + spin_lock_bh(&ab_pci->window_lock); + + iowrite32(ATH11K_PCI_WINDOW_ENABLE_BIT | ab_pci->register_window, + ab->mem + ATH11K_PCI_WINDOW_REG_ADDRESS); + ioread32(ab->mem + ATH11K_PCI_WINDOW_REG_ADDRESS); + + spin_unlock_bh(&ab_pci->window_lock); +} + static void ath11k_pci_soc_global_reset(struct ath11k_base *ab) { u32 val, delay; @@ -201,6 +214,11 @@ static void ath11k_pci_soc_global_reset(struct ath11k_base *ab) val = ath11k_pcic_read32(ab, PCIE_SOC_GLOBAL_RESET); if (val == 0xffffffff) ath11k_warn(ab, "link down error during global reset\n"); + + /* Restore window register as its content is cleared during + * hardware global reset, such that it aligns with host cache. + */ + ath11k_pci_restore_window(ab); } static void ath11k_pci_clear_dbg_registers(struct ath11k_base *ab) From 36f9edbb9d0fc36c865c74f3c1ad8e1261ad3981 Mon Sep 17 00:00:00 2001 From: Sarika Sharma Date: Tue, 30 Sep 2025 14:45:50 +0530 Subject: [PATCH 0024/1381] wifi: ath12k: Fix MSDU buffer types handling in RX error path Currently, packets received on the REO exception ring from unassociated peers are of MSDU buffer type, while the driver expects link descriptor type packets. These packets are not parsed further due to a return check on packet type in ath12k_hal_desc_reo_parse_err(), but the associated skb is not freed. This may lead to kernel crashes and buffer leaks. Hence to fix, update the RX error handler to explicitly drop MSDU buffer type packets received on the REO exception ring. This prevents further processing of invalid packets and ensures stability in the RX error handling path. Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.4.1-00199-QCAHKSWPL_SILICONZ-1 Fixes: d889913205cf ("wifi: ath12k: driver for Qualcomm Wi-Fi 7 devices") Signed-off-by: Sarika Sharma Reviewed-by: Baochen Qiang Reviewed-by: Vasanthakumar Thiagarajan Link: https://patch.msgid.link/20250930091551.3305312-2-sarika.sharma@oss.qualcomm.com Signed-off-by: Jeff Johnson --- drivers/net/wireless/ath/ath12k/dp_rx.c | 70 ++++++++++++++++++++++-- drivers/net/wireless/ath/ath12k/hal_rx.c | 10 +--- 2 files changed, 66 insertions(+), 14 deletions(-) diff --git a/drivers/net/wireless/ath/ath12k/dp_rx.c b/drivers/net/wireless/ath/ath12k/dp_rx.c index 5e5c14a70316..99d29eda26cf 100644 --- a/drivers/net/wireless/ath/ath12k/dp_rx.c +++ b/drivers/net/wireless/ath/ath12k/dp_rx.c @@ -1,7 +1,7 @@ // SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (c) 2018-2021 The Linux Foundation. All rights reserved. - * Copyright (c) 2021-2025 Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. */ #include @@ -3781,6 +3781,48 @@ ath12k_dp_process_rx_err_buf(struct ath12k *ar, struct hal_reo_dest_ring *desc, return 0; } +static int ath12k_dp_h_msdu_buffer_type(struct ath12k_base *ab, + struct list_head *list, + struct hal_reo_dest_ring *desc) +{ + struct ath12k_rx_desc_info *desc_info; + struct ath12k_skb_rxcb *rxcb; + struct sk_buff *msdu; + u64 desc_va; + + desc_va = (u64)le32_to_cpu(desc->buf_va_hi) << 32 | + le32_to_cpu(desc->buf_va_lo); + desc_info = (struct ath12k_rx_desc_info *)(uintptr_t)desc_va; + if (!desc_info) { + u32 cookie; + + cookie = le32_get_bits(desc->buf_addr_info.info1, + BUFFER_ADDR_INFO1_SW_COOKIE); + desc_info = ath12k_dp_get_rx_desc(ab, cookie); + if (!desc_info) { + ath12k_warn(ab, "Invalid cookie in manual descriptor retrieval: 0x%x\n", + cookie); + return -EINVAL; + } + } + + if (desc_info->magic != ATH12K_DP_RX_DESC_MAGIC) { + ath12k_warn(ab, "rx exception, magic check failed with value: %u\n", + desc_info->magic); + return -EINVAL; + } + + msdu = desc_info->skb; + desc_info->skb = NULL; + list_add_tail(&desc_info->list, list); + rxcb = ATH12K_SKB_RXCB(msdu); + dma_unmap_single(ab->dev, rxcb->paddr, msdu->len + skb_tailroom(msdu), + DMA_FROM_DEVICE); + dev_kfree_skb_any(msdu); + + return 0; +} + int ath12k_dp_rx_process_err(struct ath12k_base *ab, struct napi_struct *napi, int budget) { @@ -3825,6 +3867,26 @@ int ath12k_dp_rx_process_err(struct ath12k_base *ab, struct napi_struct *napi, drop = false; ab->device_stats.err_ring_pkts++; + hw_link_id = le32_get_bits(reo_desc->info0, + HAL_REO_DEST_RING_INFO0_SRC_LINK_ID); + device_id = hw_links[hw_link_id].device_id; + partner_ab = ath12k_ag_to_ab(ag, device_id); + + /* Below case is added to handle data packet from un-associated clients. + * As it is expected that AST lookup will fail for + * un-associated station's data packets. + */ + if (le32_get_bits(reo_desc->info0, HAL_REO_DEST_RING_INFO0_BUFFER_TYPE) == + HAL_REO_DEST_RING_BUFFER_TYPE_MSDU) { + if (!ath12k_dp_h_msdu_buffer_type(partner_ab, + &rx_desc_used_list[device_id], + reo_desc)) { + num_buffs_reaped[device_id]++; + tot_n_bufs_reaped++; + } + goto next_desc; + } + ret = ath12k_hal_desc_reo_parse_err(ab, reo_desc, &paddr, &desc_bank); if (ret) { @@ -3833,11 +3895,6 @@ int ath12k_dp_rx_process_err(struct ath12k_base *ab, struct napi_struct *napi, continue; } - hw_link_id = le32_get_bits(reo_desc->info0, - HAL_REO_DEST_RING_INFO0_SRC_LINK_ID); - device_id = hw_links[hw_link_id].device_id; - partner_ab = ath12k_ag_to_ab(ag, device_id); - pdev_id = ath12k_hw_mac_id_to_pdev_id(partner_ab->hw_params, hw_links[hw_link_id].pdev_idx); ar = partner_ab->pdevs[pdev_id].ar; @@ -3886,6 +3943,7 @@ int ath12k_dp_rx_process_err(struct ath12k_base *ab, struct napi_struct *napi, } } +next_desc: if (tot_n_bufs_reaped >= quota) { tot_n_bufs_reaped = quota; goto exit; diff --git a/drivers/net/wireless/ath/ath12k/hal_rx.c b/drivers/net/wireless/ath/ath12k/hal_rx.c index 669096278fdd..c4443ca05cd6 100644 --- a/drivers/net/wireless/ath/ath12k/hal_rx.c +++ b/drivers/net/wireless/ath/ath12k/hal_rx.c @@ -1,7 +1,7 @@ // SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (c) 2018-2021 The Linux Foundation. All rights reserved. - * Copyright (c) 2021-2025 Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. */ #include "debug.h" @@ -323,7 +323,7 @@ int ath12k_hal_desc_reo_parse_err(struct ath12k_base *ab, { enum hal_reo_dest_ring_push_reason push_reason; enum hal_reo_dest_ring_error_code err_code; - u32 cookie, val; + u32 cookie; push_reason = le32_get_bits(desc->info0, HAL_REO_DEST_RING_INFO0_PUSH_REASON); @@ -338,12 +338,6 @@ int ath12k_hal_desc_reo_parse_err(struct ath12k_base *ab, return -EINVAL; } - val = le32_get_bits(desc->info0, HAL_REO_DEST_RING_INFO0_BUFFER_TYPE); - if (val != HAL_REO_DEST_RING_BUFFER_TYPE_LINK_DESC) { - ath12k_warn(ab, "expected buffer type link_desc"); - return -EINVAL; - } - ath12k_hal_rx_reo_ent_paddr_get(ab, &desc->buf_addr_info, paddr, &cookie); *desc_bank = u32_get_bits(cookie, DP_LINK_DESC_BANK_MASK); From 43ba986e7ac7d9420e26e9a9b03c73054bc2149c Mon Sep 17 00:00:00 2001 From: Sarika Sharma Date: Tue, 30 Sep 2025 14:45:51 +0530 Subject: [PATCH 0025/1381] wifi: ath12k: track dropped MSDU buffer type packets in REO exception ring Add a counter "reo_excep_msdu_buf_type" in ath12k_debugfs_dump_device_dp_stats() to account for packets dropped due to unexpected MSDU buffer types in the RX error path. These packets are discarded to prevent incorrect parsing and potential kernel crashes. This helps in debugging and monitoring RX error handling behavior. Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.4.1-00199-QCAHKSWPL_SILICONZ-1 Signed-off-by: Sarika Sharma Reviewed-by: Baochen Qiang Reviewed-by: Vasanthakumar Thiagarajan Link: https://patch.msgid.link/20250930091551.3305312-3-sarika.sharma@oss.qualcomm.com Signed-off-by: Jeff Johnson --- drivers/net/wireless/ath/ath12k/core.h | 1 + drivers/net/wireless/ath/ath12k/debugfs.c | 5 ++++- drivers/net/wireless/ath/ath12k/dp_rx.c | 2 ++ 3 files changed, 7 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/ath/ath12k/core.h b/drivers/net/wireless/ath/ath12k/core.h index 3d1956966a48..48d95ea7b3db 100644 --- a/drivers/net/wireless/ath/ath12k/core.h +++ b/drivers/net/wireless/ath/ath12k/core.h @@ -963,6 +963,7 @@ struct ath12k_device_dp_stats { u32 tx_wbm_rel_source[HAL_WBM_REL_SRC_MODULE_MAX]; u32 tx_enqueued[DP_TCL_NUM_RING_MAX]; u32 tx_completed[DP_TCL_NUM_RING_MAX]; + u32 reo_excep_msdu_buf_type; }; struct ath12k_reg_freq { diff --git a/drivers/net/wireless/ath/ath12k/debugfs.c b/drivers/net/wireless/ath/ath12k/debugfs.c index 16601a8c3644..15219429d4ed 100644 --- a/drivers/net/wireless/ath/ath12k/debugfs.c +++ b/drivers/net/wireless/ath/ath12k/debugfs.c @@ -1,7 +1,7 @@ // SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (c) 2018-2021 The Linux Foundation. All rights reserved. - * Copyright (c) 2021-2025 Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. */ #include "core.h" @@ -1178,6 +1178,9 @@ static ssize_t ath12k_debugfs_dump_device_dp_stats(struct file *file, len += scnprintf(buf + len, size - len, "\n"); } + len += scnprintf(buf + len, size - len, "\nREO excep MSDU buf type:%u\n", + device_stats->reo_excep_msdu_buf_type); + len += scnprintf(buf + len, size - len, "\nRx WBM REL SRC Errors:\n"); for (i = 0; i < HAL_WBM_REL_SRC_MODULE_MAX; i++) { diff --git a/drivers/net/wireless/ath/ath12k/dp_rx.c b/drivers/net/wireless/ath/ath12k/dp_rx.c index 99d29eda26cf..6c9f0839c83a 100644 --- a/drivers/net/wireless/ath/ath12k/dp_rx.c +++ b/drivers/net/wireless/ath/ath12k/dp_rx.c @@ -3790,6 +3790,8 @@ static int ath12k_dp_h_msdu_buffer_type(struct ath12k_base *ab, struct sk_buff *msdu; u64 desc_va; + ab->device_stats.reo_excep_msdu_buf_type++; + desc_va = (u64)le32_to_cpu(desc->buf_va_hi) << 32 | le32_to_cpu(desc->buf_va_lo); desc_info = (struct ath12k_rx_desc_info *)(uintptr_t)desc_va; From 6917e268c4338ceb5916c8695423597ed8c8b38e Mon Sep 17 00:00:00 2001 From: Aditya Kumar Singh Date: Wed, 24 Sep 2025 19:13:36 +0530 Subject: [PATCH 0026/1381] wifi: ath12k: Defer vdev bring-up until CSA finalize to avoid stale beacon Mac80211 schedules CSA finalize work twice during a channel switch: first during the reserved switch phase and again during the finalize phase. The beacon content is updated only during the second schedule, which occurs after the reserved switch completes. However, the ath12k driver attempts to bring up the VDEV during the channel switch callback (ath12k_mac_update_vif_chan()), which leads to premature installation of stale beacon templates before the updated content is available. This premature VDEV bring-up causes outdated beacon information to be broadcast, which can result in updated channel parameters during the transition. In MBSSID scenarios, this behavior is particularly problematic because the transmitting interface's beacon must be updated before non-transmitting interfaces are brought up. Failing to do so can lead to beacon mismatches across interfaces. Introduce a is_csa_in_progress flag to defer VDEV_UP until CSA finalize is complete. Set this flag during the channel switch callback when CSA is active. In bss_info_changed(), check this flag and issue VDEV_UP only after the beacon template has been updated. Ensure that in MBSSID cases, the transmitting interface is brought up first, followed by all non-transmitting interfaces. This ordering makes sure correct beacon propagation and avoids stale beacon installation during CSA transitions. Additionally, move the call to ath12k_mac_update_peer_puncturing_width() before VDEV bring-up during CSA handling. This ensures that the puncturing bitmap and bandwidth settings are applied before the VDEV is brought up. Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.3.1-00173-QCAHKSWPL_SILICONZ-1 Fixes: 8c6faa56bfb2 ("wifi: ath12k: add MBSSID beacon support") Signed-off-by: Aditya Kumar Singh Signed-off-by: Maharaja Kennadyrajan Reviewed-by: Vasanthakumar Thiagarajan Link: https://patch.msgid.link/20250924134336.888-1-maharaja.kennadyrajan@oss.qualcomm.com Signed-off-by: Jeff Johnson --- drivers/net/wireless/ath/ath12k/core.h | 1 + drivers/net/wireless/ath/ath12k/mac.c | 90 ++++++++++++++++++++++---- 2 files changed, 79 insertions(+), 12 deletions(-) diff --git a/drivers/net/wireless/ath/ath12k/core.h b/drivers/net/wireless/ath/ath12k/core.h index 48d95ea7b3db..02a32b9f3ac2 100644 --- a/drivers/net/wireless/ath/ath12k/core.h +++ b/drivers/net/wireless/ath/ath12k/core.h @@ -355,6 +355,7 @@ struct ath12k_link_vif { struct wmi_vdev_install_key_arg group_key; bool pairwise_key_done; u16 num_stations; + bool is_csa_in_progress; }; struct ath12k_vif { diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c index 1d7b60aa5cb0..84473dbf22e1 100644 --- a/drivers/net/wireless/ath/ath12k/mac.c +++ b/drivers/net/wireless/ath/ath12k/mac.c @@ -4221,6 +4221,30 @@ static bool ath12k_mac_supports_tpc(struct ath12k *ar, struct ath12k_vif *ahvif, chandef->chan->band == NL80211_BAND_6GHZ; } +static void ath12k_wmi_vdev_params_up(struct ath12k *ar, + struct ath12k_link_vif *arvif, + struct ath12k_link_vif *tx_arvif, + struct ieee80211_bss_conf *info, u16 aid) +{ + struct ath12k_wmi_vdev_up_params params = { + .vdev_id = arvif->vdev_id, + .aid = aid, + .bssid = arvif->bssid + }; + int ret; + + if (tx_arvif) { + params.tx_bssid = tx_arvif->bssid; + params.nontx_profile_idx = info->bssid_index; + params.nontx_profile_cnt = 1 << info->bssid_indicator; + } + + ret = ath12k_wmi_vdev_up(arvif->ar, ¶ms); + if (ret) + ath12k_warn(ar->ab, "failed to bring vdev up %d: %d\n", + arvif->vdev_id, ret); +} + static void ath12k_mac_bss_info_changed(struct ath12k *ar, struct ath12k_link_vif *arvif, struct ieee80211_bss_conf *info, @@ -4228,6 +4252,7 @@ static void ath12k_mac_bss_info_changed(struct ath12k *ar, { struct ath12k_vif *ahvif = arvif->ahvif; struct ieee80211_vif *vif = ath12k_ahvif_to_vif(ahvif); + struct ath12k_link_vif *tx_arvif; struct cfg80211_chan_def def; u32 param_id, param_value; enum nl80211_band band; @@ -4236,9 +4261,9 @@ static void ath12k_mac_bss_info_changed(struct ath12k *ar, u32 preamble; u16 hw_value; u16 bitrate; - int ret; u8 rateidx; u32 rate; + int ret; lockdep_assert_wiphy(ath12k_ar_to_hw(ar)->wiphy); @@ -4271,12 +4296,41 @@ static void ath12k_mac_bss_info_changed(struct ath12k *ar, "Set burst beacon mode for VDEV: %d\n", arvif->vdev_id); + /* In MBSSID case, need to install transmitting VIF's template first */ + ret = ath12k_mac_setup_bcn_tmpl(arvif); if (ret) ath12k_warn(ar->ab, "failed to update bcn template: %d\n", ret); + + if (!arvif->is_csa_in_progress) + goto skip_vdev_up; + + tx_arvif = ath12k_mac_get_tx_arvif(arvif, info); + if (tx_arvif && arvif != tx_arvif && tx_arvif->is_csa_in_progress) + /* skip non tx vif's */ + goto skip_vdev_up; + + ath12k_wmi_vdev_params_up(ar, arvif, tx_arvif, info, ahvif->aid); + + arvif->is_csa_in_progress = false; + + if (tx_arvif && arvif == tx_arvif) { + struct ath12k_link_vif *arvif_itr; + + list_for_each_entry(arvif_itr, &ar->arvifs, list) { + if (!arvif_itr->is_csa_in_progress) + continue; + + ath12k_wmi_vdev_params_up(ar, arvif, tx_arvif, + info, ahvif->aid); + arvif_itr->is_csa_in_progress = false; + } + } } +skip_vdev_up: + if (changed & (BSS_CHANGED_BEACON_INFO | BSS_CHANGED_BEACON)) { arvif->dtim_period = info->dtim_period; @@ -10862,9 +10916,9 @@ ath12k_mac_update_vif_chan(struct ath12k *ar, int n_vifs) { struct ath12k_wmi_vdev_up_params params = {}; - struct ath12k_link_vif *arvif; struct ieee80211_bss_conf *link_conf; struct ath12k_base *ab = ar->ab; + struct ath12k_link_vif *arvif; struct ieee80211_vif *vif; struct ath12k_vif *ahvif; u8 link_id; @@ -10925,6 +10979,28 @@ ath12k_mac_update_vif_chan(struct ath12k *ar, continue; } + ret = ath12k_mac_update_peer_puncturing_width(arvif->ar, arvif, + vifs[i].new_ctx->def); + if (ret) { + ath12k_warn(ar->ab, + "failed to update puncturing bitmap %02x and width %d: %d\n", + vifs[i].new_ctx->def.punctured, + vifs[i].new_ctx->def.width, ret); + continue; + } + + /* Defer VDEV bring-up during CSA to avoid installing stale + * beacon templates. The beacon content is updated only + * after CSA finalize, so we mark CSA in progress and skip + * VDEV_UP for now. It will be handled later in + * bss_info_changed(). + */ + if (link_conf->csa_active && + arvif->ahvif->vdev_type == WMI_VDEV_TYPE_AP) { + arvif->is_csa_in_progress = true; + continue; + } + ret = ath12k_mac_setup_bcn_tmpl(arvif); if (ret) ath12k_warn(ab, "failed to update bcn tmpl during csa: %d\n", @@ -10945,16 +11021,6 @@ ath12k_mac_update_vif_chan(struct ath12k *ar, arvif->vdev_id, ret); continue; } - - ret = ath12k_mac_update_peer_puncturing_width(arvif->ar, arvif, - vifs[i].new_ctx->def); - if (ret) { - ath12k_warn(ar->ab, - "failed to update puncturing bitmap %02x and width %d: %d\n", - vifs[i].new_ctx->def.punctured, - vifs[i].new_ctx->def.width, ret); - continue; - } } /* Restart the internal monitor vdev on new channel */ From b94f523cc5a19108ff4687a4bce9e5d484f0f9c5 Mon Sep 17 00:00:00 2001 From: Thiraviyam Mariyappan Date: Tue, 7 Oct 2025 19:03:32 +0530 Subject: [PATCH 0027/1381] wifi: ath12k: Fix NSS value update in ext_rx_stats Currently, in ext_rx_stats, the NSS value is taken directly from the firmware, which results in incorrect mapping: 4x4, 3x3, 2x2, 1x1 SS are incorrectly updated as 3x3, 2x2, 1x1, 0x0 SS respectively. Fix the issue by incrementing the NSS value by 1 while updating the PPDU info to ensure accurate spatial stream statistics. Remove the redundant +1 increment in the radiotap header when monitor mode is enabled to prevent double counting. Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.5-01651-QCAHKSWPL_SILICONZ-1 Signed-off-by: Thiraviyam Mariyappan Reviewed-by: Vasanthakumar Thiagarajan Link: https://patch.msgid.link/20251007133332.1092178-1-thiraviyam.mariyappan@oss.qualcomm.com Signed-off-by: Jeff Johnson --- drivers/net/wireless/ath/ath12k/dp_mon.c | 19 ++++++++++++------- 1 file changed, 12 insertions(+), 7 deletions(-) diff --git a/drivers/net/wireless/ath/ath12k/dp_mon.c b/drivers/net/wireless/ath/ath12k/dp_mon.c index 009c49502148..39d1967584db 100644 --- a/drivers/net/wireless/ath/ath12k/dp_mon.c +++ b/drivers/net/wireless/ath/ath12k/dp_mon.c @@ -1,7 +1,7 @@ // SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (c) 2019-2021 The Linux Foundation. All rights reserved. - * Copyright (c) 2021-2025 Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. */ #include "dp_mon.h" @@ -105,7 +105,7 @@ static void ath12k_dp_mon_parse_vht_sig_a(const struct hal_rx_vht_sig_a_info *vh if (ppdu_info->is_stbc && nsts > 0) nsts = ((nsts + 1) >> 1) - 1; - ppdu_info->nss = u32_get_bits(nsts, VHT_SIG_SU_NSS_MASK); + ppdu_info->nss = u32_get_bits(nsts, VHT_SIG_SU_NSS_MASK) + 1; ppdu_info->bw = u32_get_bits(info0, HAL_RX_VHT_SIG_A_INFO_INFO0_BW); ppdu_info->beamformed = u32_get_bits(info1, HAL_RX_VHT_SIG_A_INFO_INFO1_BEAMFORMED); @@ -129,7 +129,7 @@ static void ath12k_dp_mon_parse_ht_sig(const struct hal_rx_ht_sig_info *ht_sig, ppdu_info->is_stbc = u32_get_bits(info1, HAL_RX_HT_SIG_INFO_INFO1_STBC); ppdu_info->ldpc = u32_get_bits(info1, HAL_RX_HT_SIG_INFO_INFO1_FEC_CODING); ppdu_info->gi = u32_get_bits(info1, HAL_RX_HT_SIG_INFO_INFO1_GI); - ppdu_info->nss = (ppdu_info->mcs >> 3); + ppdu_info->nss = (ppdu_info->mcs >> 3) + 1; } static void ath12k_dp_mon_parse_l_sig_b(const struct hal_rx_lsig_b_info *lsigb, @@ -233,7 +233,9 @@ ath12k_dp_mon_parse_he_sig_b2_ofdma(const struct hal_rx_he_sig_b2_ofdma_info *of value = value << HE_STA_ID_SHIFT; ppdu_info->he_data4 |= value; - ppdu_info->nss = u32_get_bits(info0, HAL_RX_HE_SIG_B2_OFDMA_INFO_INFO0_STA_NSTS); + ppdu_info->nss = + u32_get_bits(info0, + HAL_RX_HE_SIG_B2_OFDMA_INFO_INFO0_STA_NSTS) + 1; ppdu_info->beamformed = u32_get_bits(info0, HAL_RX_HE_SIG_B2_OFDMA_INFO_INFO0_STA_TXBF); } @@ -261,7 +263,9 @@ ath12k_dp_mon_parse_he_sig_b2_mu(const struct hal_rx_he_sig_b2_mu_info *he_sig_b value = value << HE_STA_ID_SHIFT; ppdu_info->he_data4 |= value; - ppdu_info->nss = u32_get_bits(info0, HAL_RX_HE_SIG_B2_MU_INFO_INFO0_STA_NSTS); + ppdu_info->nss = + u32_get_bits(info0, + HAL_RX_HE_SIG_B2_MU_INFO_INFO0_STA_NSTS) + 1; } static void @@ -553,7 +557,8 @@ static void ath12k_dp_mon_parse_he_sig_su(const struct hal_rx_he_sig_a_su_info * ppdu_info->is_stbc = u32_get_bits(info1, HAL_RX_HE_SIG_A_SU_INFO_INFO1_STBC); ppdu_info->beamformed = u32_get_bits(info1, HAL_RX_HE_SIG_A_SU_INFO_INFO1_TXBF); dcm = u32_get_bits(info0, HAL_RX_HE_SIG_A_SU_INFO_INFO0_DCM); - ppdu_info->nss = u32_get_bits(info0, HAL_RX_HE_SIG_A_SU_INFO_INFO0_NSTS); + ppdu_info->nss = u32_get_bits(info0, + HAL_RX_HE_SIG_A_SU_INFO_INFO0_NSTS) + 1; ppdu_info->dcm = dcm; } @@ -2179,7 +2184,7 @@ static void ath12k_dp_mon_update_radiotap(struct ath12k *ar, spin_unlock_bh(&ar->data_lock); rxs->flag |= RX_FLAG_MACTIME_START; - rxs->nss = ppduinfo->nss + 1; + rxs->nss = ppduinfo->nss; if (test_bit(WMI_TLV_SERVICE_HW_DB2DBM_CONVERSION_SUPPORT, ar->ab->wmi_ab.svc_map)) rxs->signal = ppduinfo->rssi_comb; From 8c21b32c2cc82224c7fc1a9f67318f3b1199744b Mon Sep 17 00:00:00 2001 From: Baochen Qiang Date: Thu, 9 Oct 2025 14:16:55 -0700 Subject: [PATCH 0028/1381] wifi: ath12k: fix VHT MCS assignment While associating, firmware needs the peer's receive capability to calculate its own VHT transmit MCS. Currently, the host sends this information via mcs->rx_mcs_set field, but firmware actually reads it from mcs->tx_mcs_set field. This mismatch is incorrect. This issue has not caused failures so far because most peers advertise identical TX and RX capabilities. Fix this by assigning the value to tx_mcs_set as expected. Additionally, the rate control mask is intended to limit our transmit MCS, so it should also apply to the peer's receive capability. Update the logic accordingly. Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1 Fixes: d889913205cf ("wifi: ath12k: driver for Qualcomm Wi-Fi 7 devices") Signed-off-by: Baochen Qiang Signed-off-by: Pradeep Kumar Chitrapu Reviewed-by: Vasanthakumar Thiagarajan Reviewed-by: Baochen Qiang Link: https://patch.msgid.link/20251009211656.2386085-2-quic_pradeepc@quicinc.com Signed-off-by: Jeff Johnson --- drivers/net/wireless/ath/ath12k/mac.c | 7 +++---- drivers/net/wireless/ath/ath12k/wmi.c | 13 ++++++++----- drivers/net/wireless/ath/ath12k/wmi.h | 2 ++ 3 files changed, 13 insertions(+), 9 deletions(-) diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c index 84473dbf22e1..303707606076 100644 --- a/drivers/net/wireless/ath/ath12k/mac.c +++ b/drivers/net/wireless/ath/ath12k/mac.c @@ -2249,7 +2249,6 @@ static void ath12k_peer_assoc_h_vht(struct ath12k *ar, struct cfg80211_chan_def def; enum nl80211_band band; u16 *vht_mcs_mask; - u16 tx_mcs_map; u8 ampdu_factor; u8 max_nss, vht_mcs; int i, vht_nss, nss_idx; @@ -2340,10 +2339,10 @@ static void ath12k_peer_assoc_h_vht(struct ath12k *ar, arg->peer_nss = min(link_sta->rx_nss, max_nss); arg->rx_max_rate = __le16_to_cpu(vht_cap->vht_mcs.rx_highest); arg->rx_mcs_set = __le16_to_cpu(vht_cap->vht_mcs.rx_mcs_map); - arg->tx_max_rate = __le16_to_cpu(vht_cap->vht_mcs.tx_highest); + arg->rx_mcs_set = ath12k_peer_assoc_h_vht_limit(arg->rx_mcs_set, vht_mcs_mask); - tx_mcs_map = __le16_to_cpu(vht_cap->vht_mcs.tx_mcs_map); - arg->tx_mcs_set = ath12k_peer_assoc_h_vht_limit(tx_mcs_map, vht_mcs_mask); + arg->tx_max_rate = __le16_to_cpu(vht_cap->vht_mcs.tx_highest); + arg->tx_mcs_set = __le16_to_cpu(vht_cap->vht_mcs.tx_mcs_map); /* In QCN9274 platform, VHT MCS rate 10 and 11 is enabled by default. * VHT MCS rate 10 and 11 is not supported in 11ac standard. diff --git a/drivers/net/wireless/ath/ath12k/wmi.c b/drivers/net/wireless/ath/ath12k/wmi.c index ff6b3d4ea820..e76275bd6916 100644 --- a/drivers/net/wireless/ath/ath12k/wmi.c +++ b/drivers/net/wireless/ath/ath12k/wmi.c @@ -1,7 +1,7 @@ // SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (c) 2018-2021 The Linux Foundation. All rights reserved. - * Copyright (c) 2021-2025 Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. */ #include #include @@ -2367,10 +2367,13 @@ int ath12k_wmi_send_peer_assoc_cmd(struct ath12k *ar, cmd->peer_bw_rxnss_override |= cpu_to_le32(arg->peer_bw_rxnss_override); if (arg->vht_capable) { - mcs->rx_max_rate = cpu_to_le32(arg->rx_max_rate); - mcs->rx_mcs_set = cpu_to_le32(arg->rx_mcs_set); - mcs->tx_max_rate = cpu_to_le32(arg->tx_max_rate); - mcs->tx_mcs_set = cpu_to_le32(arg->tx_mcs_set); + /* Firmware interprets mcs->tx_mcs_set field as peer's + * RX capability + */ + mcs->rx_max_rate = cpu_to_le32(arg->tx_max_rate); + mcs->rx_mcs_set = cpu_to_le32(arg->tx_mcs_set); + mcs->tx_max_rate = cpu_to_le32(arg->rx_max_rate); + mcs->tx_mcs_set = cpu_to_le32(arg->rx_mcs_set); } /* HE Rates */ diff --git a/drivers/net/wireless/ath/ath12k/wmi.h b/drivers/net/wireless/ath/ath12k/wmi.h index d9fd6a6b708d..64bd968989c8 100644 --- a/drivers/net/wireless/ath/ath12k/wmi.h +++ b/drivers/net/wireless/ath/ath12k/wmi.h @@ -4204,8 +4204,10 @@ struct wmi_unit_test_cmd { struct ath12k_wmi_vht_rate_set_params { __le32 tlv_header; __le32 rx_max_rate; + /* MCS at which the peer can transmit */ __le32 rx_mcs_set; __le32 tx_max_rate; + /* MCS at which the peer can receive */ __le32 tx_mcs_set; __le32 tx_max_mcs_nss; } __packed; From 9c5f229b1312a31aff762b2111f6751e4e3722fe Mon Sep 17 00:00:00 2001 From: Pradeep Kumar Chitrapu Date: Thu, 9 Oct 2025 14:16:56 -0700 Subject: [PATCH 0029/1381] wifi: ath12k: fix TX and RX MCS rate configurations in HE mode Currently, the TX and RX MCS rate configurations per peer are reversed when sent to the firmware. As a result, RX MCS rates are configured for TX, and vice versa. This commit rectifies the configuration to match what the firmware expects. Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1 Fixes: d889913205cf ("wifi: ath12k: driver for Qualcomm Wi-Fi 7 devices") Signed-off-by: Pradeep Kumar Chitrapu Reviewed-by: Vasanthakumar Thiagarajan Reviewed-by: Baochen Qiang Link: https://patch.msgid.link/20251009211656.2386085-3-quic_pradeepc@quicinc.com Signed-off-by: Jeff Johnson --- drivers/net/wireless/ath/ath12k/mac.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c index 303707606076..d7bc19cea2a6 100644 --- a/drivers/net/wireless/ath/ath12k/mac.c +++ b/drivers/net/wireless/ath/ath12k/mac.c @@ -2624,9 +2624,10 @@ static void ath12k_peer_assoc_h_he(struct ath12k *ar, switch (link_sta->bandwidth) { case IEEE80211_STA_RX_BW_160: v = le16_to_cpu(he_cap->he_mcs_nss_supp.rx_mcs_160); + v = ath12k_peer_assoc_h_he_limit(v, he_mcs_mask); arg->peer_he_rx_mcs_set[WMI_HECAP_TXRX_MCS_NSS_IDX_160] = v; - v = ath12k_peer_assoc_h_he_limit(v, he_mcs_mask); + v = le16_to_cpu(he_cap->he_mcs_nss_supp.tx_mcs_160); arg->peer_he_tx_mcs_set[WMI_HECAP_TXRX_MCS_NSS_IDX_160] = v; arg->peer_he_mcs_count++; @@ -2636,10 +2637,10 @@ static void ath12k_peer_assoc_h_he(struct ath12k *ar, default: v = le16_to_cpu(he_cap->he_mcs_nss_supp.rx_mcs_80); + v = ath12k_peer_assoc_h_he_limit(v, he_mcs_mask); arg->peer_he_rx_mcs_set[WMI_HECAP_TXRX_MCS_NSS_IDX_80] = v; v = le16_to_cpu(he_cap->he_mcs_nss_supp.tx_mcs_80); - v = ath12k_peer_assoc_h_he_limit(v, he_mcs_mask); arg->peer_he_tx_mcs_set[WMI_HECAP_TXRX_MCS_NSS_IDX_80] = v; arg->peer_he_mcs_count++; From e5b670e5439bda09ea7e3dd3dd32edb2f367c0d3 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 14 Oct 2025 14:06:05 +0000 Subject: [PATCH 0030/1381] net: remove obsolete WARN_ON(refcount_read(&sk->sk_refcnt) == 1) sk->sk_refcnt has been converted to refcount_t in 2017. __sock_put(sk) being refcount_dec(&sk->sk_refcnt), it will complain loudly if the current refcnt is 1 (or less) in a non racy way. We can remove four WARN_ON() in favor of the generic refcount_dec() check. Signed-off-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Reviewed-by: Xuanqiang Luo Link: https://patch.msgid.link/20251014140605.2982703-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- include/net/sock.h | 12 ++++-------- net/netlink/af_netlink.c | 4 +--- net/tipc/socket.c | 4 +--- 3 files changed, 6 insertions(+), 14 deletions(-) diff --git a/include/net/sock.h b/include/net/sock.h index f0d00928db9e..30ac2eb4ef9b 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -830,11 +830,9 @@ static inline bool sk_del_node_init(struct sock *sk) { bool rc = __sk_del_node_init(sk); - if (rc) { - /* paranoid for a while -acme */ - WARN_ON(refcount_read(&sk->sk_refcnt) == 1); + if (rc) __sock_put(sk); - } + return rc; } #define sk_del_node_init_rcu(sk) sk_del_node_init(sk) @@ -852,11 +850,9 @@ static inline bool sk_nulls_del_node_init_rcu(struct sock *sk) { bool rc = __sk_nulls_del_node_init_rcu(sk); - if (rc) { - /* paranoid for a while -acme */ - WARN_ON(refcount_read(&sk->sk_refcnt) == 1); + if (rc) __sock_put(sk); - } + return rc; } diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index 2b46c0cd752a..687a84c48882 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -596,10 +596,8 @@ static void netlink_remove(struct sock *sk) table = &nl_table[sk->sk_protocol]; if (!rhashtable_remove_fast(&table->hash, &nlk_sk(sk)->node, - netlink_rhashtable_params)) { - WARN_ON(refcount_read(&sk->sk_refcnt) == 1); + netlink_rhashtable_params)) __sock_put(sk); - } netlink_table_grab(); if (nlk_sk(sk)->subscriptions) { diff --git a/net/tipc/socket.c b/net/tipc/socket.c index 1574a83384f8..bc614a1f019c 100644 --- a/net/tipc/socket.c +++ b/net/tipc/socket.c @@ -3031,10 +3031,8 @@ static void tipc_sk_remove(struct tipc_sock *tsk) struct sock *sk = &tsk->sk; struct tipc_net *tn = net_generic(sock_net(sk), tipc_net_id); - if (!rhashtable_remove_fast(&tn->sk_rht, &tsk->node, tsk_rht_params)) { - WARN_ON(refcount_read(&sk->sk_refcnt) == 1); + if (!rhashtable_remove_fast(&tn->sk_rht, &tsk->node, tsk_rht_params)) __sock_put(sk); - } } static const struct rhashtable_params tsk_rht_params = { From e1f5bb196f0b0eee197e06d361f8ac5f091c2963 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Mon, 13 Oct 2025 10:23:06 -0700 Subject: [PATCH 0031/1381] net: bcmasp: Add support for PHY-based Wake-on-LAN If available, interrogate the PHY to find out whether we can use it for Wake-on-LAN. This can be a more power efficient way of implementing that feature, especially when the MAC is powered off in low power states. Signed-off-by: Florian Fainelli Link: https://patch.msgid.link/20251013172306.2250223-1-florian.fainelli@broadcom.com Signed-off-by: Jakub Kicinski --- .../ethernet/broadcom/asp2/bcmasp_ethtool.c | 34 +++++++++++++++++-- 1 file changed, 32 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/broadcom/asp2/bcmasp_ethtool.c b/drivers/net/ethernet/broadcom/asp2/bcmasp_ethtool.c index 63f1a8c3a7fb..dd80ccfca19d 100644 --- a/drivers/net/ethernet/broadcom/asp2/bcmasp_ethtool.c +++ b/drivers/net/ethernet/broadcom/asp2/bcmasp_ethtool.c @@ -163,11 +163,30 @@ static void bcmasp_set_msglevel(struct net_device *dev, u32 level) static void bcmasp_get_wol(struct net_device *dev, struct ethtool_wolinfo *wol) { struct bcmasp_intf *intf = netdev_priv(dev); + struct bcmasp_priv *priv = intf->parent; + struct device *kdev = &priv->pdev->dev; + u32 phy_wolopts = 0; + + if (dev->phydev) { + phy_ethtool_get_wol(dev->phydev, wol); + phy_wolopts = wol->wolopts; + } + + /* MAC is not wake-up capable, return what the PHY does */ + if (!device_can_wakeup(kdev)) + return; + + /* Overlay MAC capabilities with that of the PHY queried before */ + wol->supported |= BCMASP_SUPPORTED_WAKE; + wol->wolopts |= intf->wolopts; + + /* Return the PHY configured magic password */ + if (phy_wolopts & WAKE_MAGICSECURE) + return; - wol->supported = BCMASP_SUPPORTED_WAKE; - wol->wolopts = intf->wolopts; memset(wol->sopass, 0, sizeof(wol->sopass)); + /* Otherwise the MAC one */ if (wol->wolopts & WAKE_MAGICSECURE) memcpy(wol->sopass, intf->sopass, sizeof(intf->sopass)); } @@ -177,10 +196,21 @@ static int bcmasp_set_wol(struct net_device *dev, struct ethtool_wolinfo *wol) struct bcmasp_intf *intf = netdev_priv(dev); struct bcmasp_priv *priv = intf->parent; struct device *kdev = &priv->pdev->dev; + int ret = 0; + + /* Try Wake-on-LAN from the PHY first */ + if (dev->phydev) { + ret = phy_ethtool_set_wol(dev->phydev, wol); + if (ret != -EOPNOTSUPP && wol->wolopts) + return ret; + } if (!device_can_wakeup(kdev)) return -EOPNOTSUPP; + if (wol->wolopts & ~BCMASP_SUPPORTED_WAKE) + return -EINVAL; + /* Interface Specific */ intf->wolopts = wol->wolopts; if (intf->wolopts & WAKE_MAGICSECURE) From f197902cd21ae833850679b216bb62c0d056bbb3 Mon Sep 17 00:00:00 2001 From: "Kory Maincent (Dent Project)" Date: Mon, 13 Oct 2025 16:05:31 +0200 Subject: [PATCH 0032/1381] net: pse-pd: pd692x0: Replace __free macro with explicit kfree calls Replace __free(kfree) with explicit kfree() calls to follow the net subsystem policy of avoiding automatic cleanup macros as described in the documentation. Signed-off-by: Kory Maincent Reviewed-by: Simon Horman Link: https://patch.msgid.link/20251013-feature_pd692x0_reboot_keep_conf-v2-1-68ab082a93dd@bootlin.com Signed-off-by: Jakub Kicinski --- drivers/net/pse-pd/pd692x0.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/drivers/net/pse-pd/pd692x0.c b/drivers/net/pse-pd/pd692x0.c index f4e91ba64a66..055e925c853e 100644 --- a/drivers/net/pse-pd/pd692x0.c +++ b/drivers/net/pse-pd/pd692x0.c @@ -1200,9 +1200,9 @@ static void pd692x0_managers_free_pw_budget(struct pd692x0_priv *priv) static int pd692x0_setup_pi_matrix(struct pse_controller_dev *pcdev) { - struct pd692x0_manager *manager __free(kfree) = NULL; struct pd692x0_priv *priv = to_pd692x0_priv(pcdev); struct pd692x0_matrix port_matrix[PD692X0_MAX_PIS]; + struct pd692x0_manager *manager; int ret, nmanagers; /* Should we flash the port matrix */ @@ -1216,7 +1216,7 @@ static int pd692x0_setup_pi_matrix(struct pse_controller_dev *pcdev) ret = pd692x0_of_get_managers(priv, manager); if (ret < 0) - return ret; + goto err_free_manager; nmanagers = ret; ret = pd692x0_register_managers_regulator(priv, manager, nmanagers); @@ -1236,12 +1236,15 @@ static int pd692x0_setup_pi_matrix(struct pse_controller_dev *pcdev) goto err_managers_req_pw; pd692x0_of_put_managers(priv, manager, nmanagers); + kfree(manager); return 0; err_managers_req_pw: pd692x0_managers_free_pw_budget(priv); err_of_managers: pd692x0_of_put_managers(priv, manager, nmanagers); +err_free_manager: + kfree(manager); return ret; } From 6fa1f8b64a47edd7d8420d8fd1008507aee2853e Mon Sep 17 00:00:00 2001 From: "Kory Maincent (Dent Project)" Date: Mon, 13 Oct 2025 16:05:32 +0200 Subject: [PATCH 0033/1381] net: pse-pd: pd692x0: Separate configuration parsing from hardware setup Cache the port matrix configuration in driver private data to enable PSE controller reconfiguration. This refactoring separates device tree parsing from hardware configuration application, allowing settings to be reapplied without reparsing the device tree. This refactoring is a prerequisite for preserving PSE configuration across reboots to prevent power disruption to connected devices. Signed-off-by: Kory Maincent Reviewed-by: Simon Horman Link: https://patch.msgid.link/20251013-feature_pd692x0_reboot_keep_conf-v2-2-68ab082a93dd@bootlin.com Signed-off-by: Jakub Kicinski --- drivers/net/pse-pd/pd692x0.c | 115 +++++++++++++++++++++++------------ 1 file changed, 76 insertions(+), 39 deletions(-) diff --git a/drivers/net/pse-pd/pd692x0.c b/drivers/net/pse-pd/pd692x0.c index 055e925c853e..782b1abf94cb 100644 --- a/drivers/net/pse-pd/pd692x0.c +++ b/drivers/net/pse-pd/pd692x0.c @@ -85,6 +85,11 @@ enum { PD692X0_MSG_CNT }; +struct pd692x0_matrix { + u8 hw_port_a; + u8 hw_port_b; +}; + struct pd692x0_priv { struct i2c_client *client; struct pse_controller_dev pcdev; @@ -101,6 +106,8 @@ struct pd692x0_priv { enum ethtool_c33_pse_admin_state admin_state[PD692X0_MAX_PIS]; struct regulator_dev *manager_reg[PD692X0_MAX_MANAGERS]; int manager_pw_budget[PD692X0_MAX_MANAGERS]; + int nmanagers; + struct pd692x0_matrix *port_matrix; }; /* Template list of communication messages. The non-null bytes defined here @@ -809,11 +816,6 @@ struct pd692x0_manager { int nports; }; -struct pd692x0_matrix { - u8 hw_port_a; - u8 hw_port_b; -}; - static int pd692x0_of_get_ports_manager(struct pd692x0_priv *priv, struct pd692x0_manager *manager, @@ -903,7 +905,8 @@ pd692x0_of_get_managers(struct pd692x0_priv *priv, } of_node_put(managers_node); - return nmanagers; + priv->nmanagers = nmanagers; + return 0; out: for (i = 0; i < nmanagers; i++) { @@ -963,8 +966,7 @@ pd692x0_register_manager_regulator(struct device *dev, char *reg_name, static int pd692x0_register_managers_regulator(struct pd692x0_priv *priv, - const struct pd692x0_manager *manager, - int nmanagers) + const struct pd692x0_manager *manager) { struct device *dev = &priv->client->dev; size_t reg_name_len; @@ -975,7 +977,7 @@ pd692x0_register_managers_regulator(struct pd692x0_priv *priv, */ reg_name_len = strlen(dev_name(dev)) + 23; - for (i = 0; i < nmanagers; i++) { + for (i = 0; i < priv->nmanagers; i++) { static const char * const regulators[] = { "vaux5", "vaux3p3" }; struct regulator_dev *rdev; char *reg_name; @@ -1008,10 +1010,14 @@ pd692x0_register_managers_regulator(struct pd692x0_priv *priv, } static int -pd692x0_conf_manager_power_budget(struct pd692x0_priv *priv, int id, int pw) +pd692x0_conf_manager_power_budget(struct pd692x0_priv *priv, int id) { struct pd692x0_msg msg, buf; - int ret, pw_mW = pw / 1000; + int ret, pw_mW; + + pw_mW = priv->manager_pw_budget[id] / 1000; + if (!pw_mW) + return 0; msg = pd692x0_msg_template_list[PD692X0_MSG_GET_POWER_BANK]; msg.data[0] = id; @@ -1032,11 +1038,11 @@ pd692x0_conf_manager_power_budget(struct pd692x0_priv *priv, int id, int pw) } static int -pd692x0_configure_managers(struct pd692x0_priv *priv, int nmanagers) +pd692x0_req_managers_pw_budget(struct pd692x0_priv *priv) { int i, ret; - for (i = 0; i < nmanagers; i++) { + for (i = 0; i < priv->nmanagers; i++) { struct regulator *supply = priv->manager_reg[i]->supply; int pw_budget; @@ -1053,7 +1059,18 @@ pd692x0_configure_managers(struct pd692x0_priv *priv, int nmanagers) return ret; priv->manager_pw_budget[i] = pw_budget; - ret = pd692x0_conf_manager_power_budget(priv, i, pw_budget); + } + + return 0; +} + +static int +pd692x0_configure_managers(struct pd692x0_priv *priv) +{ + int i, ret; + + for (i = 0; i < priv->nmanagers; i++) { + ret = pd692x0_conf_manager_power_budget(priv, i); if (ret < 0) return ret; } @@ -1101,10 +1118,9 @@ pd692x0_set_port_matrix(const struct pse_pi_pairset *pairset, static int pd692x0_set_ports_matrix(struct pd692x0_priv *priv, - const struct pd692x0_manager *manager, - int nmanagers, - struct pd692x0_matrix port_matrix[PD692X0_MAX_PIS]) + const struct pd692x0_manager *manager) { + struct pd692x0_matrix *port_matrix = priv->port_matrix; struct pse_controller_dev *pcdev = &priv->pcdev; int i, ret; @@ -1117,7 +1133,7 @@ pd692x0_set_ports_matrix(struct pd692x0_priv *priv, /* Update with values for every PSE PIs */ for (i = 0; i < pcdev->nr_lines; i++) { ret = pd692x0_set_port_matrix(&pcdev->pi[i].pairset[0], - manager, nmanagers, + manager, priv->nmanagers, &port_matrix[i]); if (ret) { dev_err(&priv->client->dev, @@ -1126,7 +1142,7 @@ pd692x0_set_ports_matrix(struct pd692x0_priv *priv, } ret = pd692x0_set_port_matrix(&pcdev->pi[i].pairset[1], - manager, nmanagers, + manager, priv->nmanagers, &port_matrix[i]); if (ret) { dev_err(&priv->client->dev, @@ -1139,9 +1155,9 @@ pd692x0_set_ports_matrix(struct pd692x0_priv *priv, } static int -pd692x0_write_ports_matrix(struct pd692x0_priv *priv, - const struct pd692x0_matrix port_matrix[PD692X0_MAX_PIS]) +pd692x0_write_ports_matrix(struct pd692x0_priv *priv) { + struct pd692x0_matrix *port_matrix = priv->port_matrix; struct pd692x0_msg msg, buf; int ret, i; @@ -1166,13 +1182,32 @@ pd692x0_write_ports_matrix(struct pd692x0_priv *priv, return 0; } +static int pd692x0_hw_conf_init(struct pd692x0_priv *priv) +{ + int ret; + + /* Is PD692x0 ready to be configured? */ + if (priv->fw_state != PD692X0_FW_OK && + priv->fw_state != PD692X0_FW_COMPLETE) + return 0; + + ret = pd692x0_configure_managers(priv); + if (ret) + return ret; + + ret = pd692x0_write_ports_matrix(priv); + if (ret) + return ret; + + return 0; +} + static void pd692x0_of_put_managers(struct pd692x0_priv *priv, - struct pd692x0_manager *manager, - int nmanagers) + struct pd692x0_manager *manager) { int i, j; - for (i = 0; i < nmanagers; i++) { + for (i = 0; i < priv->nmanagers; i++) { for (j = 0; j < manager[i].nports; j++) of_node_put(manager[i].port_node[j]); of_node_put(manager[i].node); @@ -1201,48 +1236,50 @@ static void pd692x0_managers_free_pw_budget(struct pd692x0_priv *priv) static int pd692x0_setup_pi_matrix(struct pse_controller_dev *pcdev) { struct pd692x0_priv *priv = to_pd692x0_priv(pcdev); - struct pd692x0_matrix port_matrix[PD692X0_MAX_PIS]; + struct pd692x0_matrix *port_matrix; struct pd692x0_manager *manager; - int ret, nmanagers; - - /* Should we flash the port matrix */ - if (priv->fw_state != PD692X0_FW_OK && - priv->fw_state != PD692X0_FW_COMPLETE) - return 0; + int ret; manager = kcalloc(PD692X0_MAX_MANAGERS, sizeof(*manager), GFP_KERNEL); if (!manager) return -ENOMEM; + port_matrix = devm_kcalloc(&priv->client->dev, PD692X0_MAX_PIS, + sizeof(*port_matrix), GFP_KERNEL); + if (!port_matrix) { + ret = -ENOMEM; + goto err_free_manager; + } + priv->port_matrix = port_matrix; + ret = pd692x0_of_get_managers(priv, manager); if (ret < 0) goto err_free_manager; - nmanagers = ret; - ret = pd692x0_register_managers_regulator(priv, manager, nmanagers); + ret = pd692x0_register_managers_regulator(priv, manager); if (ret) goto err_of_managers; - ret = pd692x0_configure_managers(priv, nmanagers); + ret = pd692x0_req_managers_pw_budget(priv); if (ret) goto err_of_managers; - ret = pd692x0_set_ports_matrix(priv, manager, nmanagers, port_matrix); + ret = pd692x0_set_ports_matrix(priv, manager); if (ret) goto err_managers_req_pw; - ret = pd692x0_write_ports_matrix(priv, port_matrix); + ret = pd692x0_hw_conf_init(priv); if (ret) goto err_managers_req_pw; - pd692x0_of_put_managers(priv, manager, nmanagers); + pd692x0_of_put_managers(priv, manager); kfree(manager); return 0; err_managers_req_pw: pd692x0_managers_free_pw_budget(priv); err_of_managers: - pd692x0_of_put_managers(priv, manager, nmanagers); + pd692x0_of_put_managers(priv, manager); err_free_manager: kfree(manager); return ret; @@ -1647,7 +1684,7 @@ static enum fw_upload_err pd692x0_fw_poll_complete(struct fw_upload *fwl) return FW_UPLOAD_ERR_FW_INVALID; } - ret = pd692x0_setup_pi_matrix(&priv->pcdev); + ret = pd692x0_hw_conf_init(priv); if (ret < 0) { dev_err(&client->dev, "Error configuring ports matrix (%pe)\n", ERR_PTR(ret)); From 8f3d044b34fe99b894046edb84605456195cabc0 Mon Sep 17 00:00:00 2001 From: "Kory Maincent (Dent Project)" Date: Mon, 13 Oct 2025 16:05:33 +0200 Subject: [PATCH 0034/1381] net: pse-pd: pd692x0: Preserve PSE configuration across reboots Detect when PSE hardware is already configured (user byte == 42) and skip hardware initialization to prevent power interruption to connected devices during system reboots. Previously, the driver would always reconfigure the PSE hardware on probe, causing a port matrix reflash that resulted in temporary power loss to all connected devices. This change maintains power continuity by preserving existing configuration when the PSE has been previously initialized. Signed-off-by: Kory Maincent Reviewed-by: Simon Horman Link: https://patch.msgid.link/20251013-feature_pd692x0_reboot_keep_conf-v2-3-68ab082a93dd@bootlin.com Signed-off-by: Jakub Kicinski --- drivers/net/pse-pd/pd692x0.c | 35 ++++++++++++++++++++++++++++++++--- 1 file changed, 32 insertions(+), 3 deletions(-) diff --git a/drivers/net/pse-pd/pd692x0.c b/drivers/net/pse-pd/pd692x0.c index 782b1abf94cb..134435e90073 100644 --- a/drivers/net/pse-pd/pd692x0.c +++ b/drivers/net/pse-pd/pd692x0.c @@ -30,6 +30,8 @@ #define PD692X0_FW_MIN_VER 5 #define PD692X0_FW_PATCH_VER 5 +#define PD692X0_USER_BYTE 42 + enum pd692x0_fw_state { PD692X0_FW_UNKNOWN, PD692X0_FW_OK, @@ -80,6 +82,7 @@ enum { PD692X0_MSG_GET_PORT_PARAM, PD692X0_MSG_GET_POWER_BANK, PD692X0_MSG_SET_POWER_BANK, + PD692X0_MSG_SET_USER_BYTE, /* add new message above here */ PD692X0_MSG_CNT @@ -103,6 +106,7 @@ struct pd692x0_priv { bool last_cmd_key; unsigned long last_cmd_key_time; + bool cfg_saved; enum ethtool_c33_pse_admin_state admin_state[PD692X0_MAX_PIS]; struct regulator_dev *manager_reg[PD692X0_MAX_MANAGERS]; int manager_pw_budget[PD692X0_MAX_MANAGERS]; @@ -193,6 +197,12 @@ static const struct pd692x0_msg pd692x0_msg_template_list[PD692X0_MSG_CNT] = { .key = PD692X0_KEY_CMD, .sub = {0x07, 0x0b, 0x57}, }, + [PD692X0_MSG_SET_USER_BYTE] = { + .key = PD692X0_KEY_PRG, + .sub = {0x41, PD692X0_USER_BYTE}, + .data = {0x4e, 0x4e, 0x4e, 0x4e, + 0x4e, 0x4e, 0x4e, 0x4e}, + }, }; static u8 pd692x0_build_msg(struct pd692x0_msg *msg, u8 echo) @@ -1233,6 +1243,15 @@ static void pd692x0_managers_free_pw_budget(struct pd692x0_priv *priv) } } +static int +pd692x0_save_user_byte(struct pd692x0_priv *priv) +{ + struct pd692x0_msg msg, buf; + + msg = pd692x0_msg_template_list[PD692X0_MSG_SET_USER_BYTE]; + return pd692x0_sendrecv_msg(priv, &msg, &buf); +} + static int pd692x0_setup_pi_matrix(struct pse_controller_dev *pcdev) { struct pd692x0_priv *priv = to_pd692x0_priv(pcdev); @@ -1268,9 +1287,16 @@ static int pd692x0_setup_pi_matrix(struct pse_controller_dev *pcdev) if (ret) goto err_managers_req_pw; - ret = pd692x0_hw_conf_init(priv); - if (ret) - goto err_managers_req_pw; + /* Do not init the conf if it is already saved */ + if (!priv->cfg_saved) { + ret = pd692x0_hw_conf_init(priv); + if (ret) + goto err_managers_req_pw; + + ret = pd692x0_save_user_byte(priv); + if (ret) + goto err_managers_req_pw; + } pd692x0_of_put_managers(priv, manager); kfree(manager); @@ -1793,6 +1819,9 @@ static int pd692x0_i2c_probe(struct i2c_client *client) } } + if (buf.data[2] == PD692X0_USER_BYTE) + priv->cfg_saved = true; + priv->np = dev->of_node; priv->pcdev.nr_lines = PD692X0_MAX_PIS; priv->pcdev.owner = THIS_MODULE; From 9fbafbfa5b992187d6e4bc85dd0479eb660b3cc1 Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Mon, 13 Oct 2025 15:58:49 +0200 Subject: [PATCH 0035/1381] dt-bindings: net: airoha: npu: Add AN7583 support Introduce AN7583 NPU support to Airoha EN7581 NPU device-tree bindings. Acked-by: Rob Herring (Arm) Signed-off-by: Lorenzo Bianconi Link: https://patch.msgid.link/20251013-airoha-npu-7583-v3-1-00f748b5a0c7@kernel.org Signed-off-by: Jakub Kicinski --- Documentation/devicetree/bindings/net/airoha,en7581-npu.yaml | 1 + 1 file changed, 1 insertion(+) diff --git a/Documentation/devicetree/bindings/net/airoha,en7581-npu.yaml b/Documentation/devicetree/bindings/net/airoha,en7581-npu.yaml index c7644e6586d3..59c57f58116b 100644 --- a/Documentation/devicetree/bindings/net/airoha,en7581-npu.yaml +++ b/Documentation/devicetree/bindings/net/airoha,en7581-npu.yaml @@ -18,6 +18,7 @@ properties: compatible: enum: - airoha,en7581-npu + - airoha,an7583-npu reg: maxItems: 1 From 0850ae496d534847ec2c26744521c1bce04ec59d Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Mon, 13 Oct 2025 15:58:50 +0200 Subject: [PATCH 0036/1381] net: airoha: npu: Add airoha_npu_soc_data struct Introduce airoha_npu_soc_data structure in order to generalize per-SoC NPU firmware info. Introduce airoha_npu_load_firmware utility routine. This is a preliminary patch in order to introduce AN7583 NPU support. Signed-off-by: Lorenzo Bianconi Reviewed-by: Simon Horman Link: https://patch.msgid.link/20251013-airoha-npu-7583-v3-2-00f748b5a0c7@kernel.org Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/airoha/airoha_npu.c | 79 ++++++++++++++++-------- 1 file changed, 52 insertions(+), 27 deletions(-) diff --git a/drivers/net/ethernet/airoha/airoha_npu.c b/drivers/net/ethernet/airoha/airoha_npu.c index 8c883f2b2d36..41944cc5f6b0 100644 --- a/drivers/net/ethernet/airoha/airoha_npu.c +++ b/drivers/net/ethernet/airoha/airoha_npu.c @@ -103,6 +103,16 @@ enum { QDMA_WAN_PON_XDSL, }; +struct airoha_npu_fw { + const char *name; + int max_size; +}; + +struct airoha_npu_soc_data { + struct airoha_npu_fw fw_rv32; + struct airoha_npu_fw fw_data; +}; + #define MBOX_MSG_FUNC_ID GENMASK(14, 11) #define MBOX_MSG_STATIC_BUF BIT(5) #define MBOX_MSG_STATUS GENMASK(4, 2) @@ -182,51 +192,55 @@ static int airoha_npu_send_msg(struct airoha_npu *npu, int func_id, return ret; } -static int airoha_npu_run_firmware(struct device *dev, void __iomem *base, - struct resource *res) +static int airoha_npu_load_firmware(struct device *dev, void __iomem *addr, + const struct airoha_npu_fw *fw_info) { const struct firmware *fw; - void __iomem *addr; int ret; - ret = request_firmware(&fw, NPU_EN7581_FIRMWARE_RV32, dev); + ret = request_firmware(&fw, fw_info->name, dev); if (ret) return ret == -ENOENT ? -EPROBE_DEFER : ret; - if (fw->size > NPU_EN7581_FIRMWARE_RV32_MAX_SIZE) { + if (fw->size > fw_info->max_size) { dev_err(dev, "%s: fw size too overlimit (%zu)\n", - NPU_EN7581_FIRMWARE_RV32, fw->size); + fw_info->name, fw->size); ret = -E2BIG; goto out; } - addr = devm_ioremap_resource(dev, res); - if (IS_ERR(addr)) { - ret = PTR_ERR(addr); - goto out; - } - memcpy_toio(addr, fw->data, fw->size); - release_firmware(fw); - - ret = request_firmware(&fw, NPU_EN7581_FIRMWARE_DATA, dev); - if (ret) - return ret == -ENOENT ? -EPROBE_DEFER : ret; - - if (fw->size > NPU_EN7581_FIRMWARE_DATA_MAX_SIZE) { - dev_err(dev, "%s: fw size too overlimit (%zu)\n", - NPU_EN7581_FIRMWARE_DATA, fw->size); - ret = -E2BIG; - goto out; - } - - memcpy_toio(base + REG_NPU_LOCAL_SRAM, fw->data, fw->size); out: release_firmware(fw); return ret; } +static int airoha_npu_run_firmware(struct device *dev, void __iomem *base, + struct resource *res) +{ + const struct airoha_npu_soc_data *soc; + void __iomem *addr; + int ret; + + soc = of_device_get_match_data(dev); + if (!soc) + return -EINVAL; + + addr = devm_ioremap_resource(dev, res); + if (IS_ERR(addr)) + return PTR_ERR(addr); + + /* Load rv32 npu firmware */ + ret = airoha_npu_load_firmware(dev, addr, &soc->fw_rv32); + if (ret) + return ret; + + /* Load data npu firmware */ + return airoha_npu_load_firmware(dev, base + REG_NPU_LOCAL_SRAM, + &soc->fw_data); +} + static irqreturn_t airoha_npu_mbox_handler(int irq, void *npu_instance) { struct airoha_npu *npu = npu_instance; @@ -597,8 +611,19 @@ void airoha_npu_put(struct airoha_npu *npu) } EXPORT_SYMBOL_GPL(airoha_npu_put); +static const struct airoha_npu_soc_data en7581_npu_soc_data = { + .fw_rv32 = { + .name = NPU_EN7581_FIRMWARE_RV32, + .max_size = NPU_EN7581_FIRMWARE_RV32_MAX_SIZE, + }, + .fw_data = { + .name = NPU_EN7581_FIRMWARE_DATA, + .max_size = NPU_EN7581_FIRMWARE_DATA_MAX_SIZE, + }, +}; + static const struct of_device_id of_airoha_npu_match[] = { - { .compatible = "airoha,en7581-npu" }, + { .compatible = "airoha,en7581-npu", .data = &en7581_npu_soc_data }, { /* sentinel */ } }; MODULE_DEVICE_TABLE(of, of_airoha_npu_match); From 4478596f71d92060c9093bdf1d2d940881f41bcc Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Mon, 13 Oct 2025 15:58:51 +0200 Subject: [PATCH 0037/1381] net: airoha: npu: Add 7583 SoC support Introduce support for Airoha 7583 SoC NPU selecting proper firmware images. Signed-off-by: Lorenzo Bianconi Reviewed-by: Simon Horman Link: https://patch.msgid.link/20251013-airoha-npu-7583-v3-3-00f748b5a0c7@kernel.org Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/airoha/airoha_npu.c | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) diff --git a/drivers/net/ethernet/airoha/airoha_npu.c b/drivers/net/ethernet/airoha/airoha_npu.c index 41944cc5f6b0..68b7f9684dc7 100644 --- a/drivers/net/ethernet/airoha/airoha_npu.c +++ b/drivers/net/ethernet/airoha/airoha_npu.c @@ -16,6 +16,8 @@ #define NPU_EN7581_FIRMWARE_DATA "airoha/en7581_npu_data.bin" #define NPU_EN7581_FIRMWARE_RV32 "airoha/en7581_npu_rv32.bin" +#define NPU_AN7583_FIRMWARE_DATA "airoha/an7583_npu_data.bin" +#define NPU_AN7583_FIRMWARE_RV32 "airoha/an7583_npu_rv32.bin" #define NPU_EN7581_FIRMWARE_RV32_MAX_SIZE 0x200000 #define NPU_EN7581_FIRMWARE_DATA_MAX_SIZE 0x10000 #define NPU_DUMP_SIZE 512 @@ -622,8 +624,20 @@ static const struct airoha_npu_soc_data en7581_npu_soc_data = { }, }; +static const struct airoha_npu_soc_data an7583_npu_soc_data = { + .fw_rv32 = { + .name = NPU_AN7583_FIRMWARE_RV32, + .max_size = NPU_EN7581_FIRMWARE_RV32_MAX_SIZE, + }, + .fw_data = { + .name = NPU_AN7583_FIRMWARE_DATA, + .max_size = NPU_EN7581_FIRMWARE_DATA_MAX_SIZE, + }, +}; + static const struct of_device_id of_airoha_npu_match[] = { { .compatible = "airoha,en7581-npu", .data = &en7581_npu_soc_data }, + { .compatible = "airoha,an7583-npu", .data = &an7583_npu_soc_data }, { /* sentinel */ } }; MODULE_DEVICE_TABLE(of, of_airoha_npu_match); @@ -762,6 +776,8 @@ module_platform_driver(airoha_npu_driver); MODULE_FIRMWARE(NPU_EN7581_FIRMWARE_DATA); MODULE_FIRMWARE(NPU_EN7581_FIRMWARE_RV32); +MODULE_FIRMWARE(NPU_AN7583_FIRMWARE_DATA); +MODULE_FIRMWARE(NPU_AN7583_FIRMWARE_RV32); MODULE_LICENSE("GPL"); MODULE_AUTHOR("Lorenzo Bianconi "); MODULE_DESCRIPTION("Airoha Network Processor Unit driver"); From 4bd451f4c2851eee7b6e17bb6fd6c9caaadbdc18 Mon Sep 17 00:00:00 2001 From: Dimitri Daskalakis Date: Mon, 13 Oct 2025 14:14:48 -0700 Subject: [PATCH 0038/1381] net: fbnic: Fix page chunking logic when PAGE_SIZE > 4K The HW always works on a 4K page size. When the OS supports larger pages, we fragment them across multiple BDQ descriptors. We were not properly incrementing the descriptor, which resulted in us specifying the last chunks id/addr and then 15 zero descriptors. This would cause packet loss and driver crashes. This is not a fix since the Kconfig prevents use outside of x86. Signed-off-by: Dimitri Daskalakis Signed-off-by: Jakub Kicinski Reviewed-by: Simon Horman Link: https://patch.msgid.link/20251013211449.1377054-2-dimitri.daskalakis1@gmail.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/meta/fbnic/fbnic_txrx.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_txrx.c b/drivers/net/ethernet/meta/fbnic/fbnic_txrx.c index b1e8ce89870f..57e18a68f5d2 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_txrx.c +++ b/drivers/net/ethernet/meta/fbnic/fbnic_txrx.c @@ -887,6 +887,7 @@ static void fbnic_bd_prep(struct fbnic_ring *bdq, u16 id, netmem_ref netmem) *bdq_desc = cpu_to_le64(bd); bd += FIELD_PREP(FBNIC_BD_DESC_ADDR_MASK, 1) | FIELD_PREP(FBNIC_BD_DESC_ID_MASK, 1); + bdq_desc++; } while (--i); } From 75b350839b9e6a0ee73b5a4835a8c61e68851ae8 Mon Sep 17 00:00:00 2001 From: Dimitri Daskalakis Date: Mon, 13 Oct 2025 14:14:49 -0700 Subject: [PATCH 0039/1381] net: fbnic: Allow builds for all 64 bit architectures This enables aarch64 testing, but there's no reason we cannot support other architectures. Signed-off-by: Dimitri Daskalakis Signed-off-by: Jakub Kicinski Reviewed-by: Simon Horman Link: https://patch.msgid.link/20251013211449.1377054-3-dimitri.daskalakis1@gmail.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/meta/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/meta/Kconfig b/drivers/net/ethernet/meta/Kconfig index 3ba527514f1e..dff51f23d295 100644 --- a/drivers/net/ethernet/meta/Kconfig +++ b/drivers/net/ethernet/meta/Kconfig @@ -19,7 +19,7 @@ if NET_VENDOR_META config FBNIC tristate "Meta Platforms Host Network Interface" - depends on X86_64 || COMPILE_TEST + depends on 64BIT || COMPILE_TEST depends on !S390 depends on MAX_SKB_FRAGS < 22 depends on PCI_MSI From 00922eeaca3c5c2001781bcad40e0bd54d0fdbb6 Mon Sep 17 00:00:00 2001 From: "Rob Herring (Arm)" Date: Mon, 13 Oct 2025 16:30:49 -0500 Subject: [PATCH 0040/1381] dt-bindings: net: Convert amd,xgbe-seattle-v1a to DT schema Convert amd,xgbe-seattle-v1a binding to DT schema format. It's a straight-forward conversion. Signed-off-by: Rob Herring (Arm) Link: https://patch.msgid.link/20251013213049.686797-2-robh@kernel.org Signed-off-by: Paolo Abeni --- .../bindings/net/amd,xgbe-seattle-v1a.yaml | 147 ++++++++++++++++++ .../devicetree/bindings/net/amd-xgbe.txt | 76 --------- 2 files changed, 147 insertions(+), 76 deletions(-) create mode 100644 Documentation/devicetree/bindings/net/amd,xgbe-seattle-v1a.yaml delete mode 100644 Documentation/devicetree/bindings/net/amd-xgbe.txt diff --git a/Documentation/devicetree/bindings/net/amd,xgbe-seattle-v1a.yaml b/Documentation/devicetree/bindings/net/amd,xgbe-seattle-v1a.yaml new file mode 100644 index 000000000000..006add8b6410 --- /dev/null +++ b/Documentation/devicetree/bindings/net/amd,xgbe-seattle-v1a.yaml @@ -0,0 +1,147 @@ +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/net/amd,xgbe-seattle-v1a.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: AMD XGBE Seattle v1a + +maintainers: + - Shyam Sundar S K + +allOf: + - $ref: /schemas/net/ethernet-controller.yaml# + +properties: + compatible: + const: amd,xgbe-seattle-v1a + + reg: + items: + - description: MAC registers + - description: PCS registers + - description: SerDes Rx/Tx registers + - description: SerDes integration registers (1/2) + - description: SerDes integration registers (2/2) + + interrupts: + description: Device interrupts. The first entry is the general device + interrupt. If amd,per-channel-interrupt is specified, each DMA channel + interrupt must be specified. The last entry is the PCS auto-negotiation + interrupt. + minItems: 2 + maxItems: 6 + + clocks: + items: + - description: DMA clock for the device + - description: PTP clock for the device + + clock-names: + items: + - const: dma_clk + - const: ptp_clk + + iommus: + maxItems: 1 + + phy-mode: true + + dma-coherent: true + + amd,per-channel-interrupt: + description: Indicates that Rx and Tx complete will generate a unique + interrupt for each DMA channel. + type: boolean + + amd,speed-set: + description: > + Speed capabilities of the device. + 0 = 1GbE and 10GbE + 1 = 2.5GbE and 10GbE + $ref: /schemas/types.yaml#/definitions/uint32 + enum: [0, 1] + + amd,serdes-blwc: + description: Baseline wandering correction enablement for each speed. + $ref: /schemas/types.yaml#/definitions/uint32-array + minItems: 3 + maxItems: 3 + items: + enum: [0, 1] + + amd,serdes-cdr-rate: + description: CDR rate speed selection for each speed. + $ref: /schemas/types.yaml#/definitions/uint32-array + items: + - description: CDR rate for 1GbE + - description: CDR rate for 2.5GbE + - description: CDR rate for 10GbE + + amd,serdes-pq-skew: + description: PQ data sampling skew for each speed. + $ref: /schemas/types.yaml#/definitions/uint32-array + items: + - description: PQ skew for 1GbE + - description: PQ skew for 2.5GbE + - description: PQ skew for 10GbE + + amd,serdes-tx-amp: + description: TX amplitude boost for each speed. + $ref: /schemas/types.yaml#/definitions/uint32-array + items: + - description: TX amplitude for 1GbE + - description: TX amplitude for 2.5GbE + - description: TX amplitude for 10GbE + + amd,serdes-dfe-tap-config: + description: DFE taps available to run for each speed. + $ref: /schemas/types.yaml#/definitions/uint32-array + items: + - description: DFE taps available for 1GbE + - description: DFE taps available for 2.5GbE + - description: DFE taps available for 10GbE + + amd,serdes-dfe-tap-enable: + description: DFE taps to enable for each speed. + $ref: /schemas/types.yaml#/definitions/uint32-array + items: + - description: DFE taps to enable for 1GbE + - description: DFE taps to enable for 2.5GbE + - description: DFE taps to enable for 10GbE + +required: + - compatible + - reg + - interrupts + - clocks + - clock-names + - phy-mode + +unevaluatedProperties: false + +examples: + - | + ethernet@e0700000 { + compatible = "amd,xgbe-seattle-v1a"; + reg = <0xe0700000 0x80000>, + <0xe0780000 0x80000>, + <0xe1240800 0x00400>, + <0xe1250000 0x00060>, + <0xe1250080 0x00004>; + interrupts = <0 325 4>, + <0 326 1>, <0 327 1>, <0 328 1>, <0 329 1>, + <0 323 4>; + amd,per-channel-interrupt; + clocks = <&xgbe_dma_clk>, <&xgbe_ptp_clk>; + clock-names = "dma_clk", "ptp_clk"; + phy-mode = "xgmii"; + mac-address = [ 02 a1 a2 a3 a4 a5 ]; + amd,speed-set = <0>; + amd,serdes-blwc = <1>, <1>, <0>; + amd,serdes-cdr-rate = <2>, <2>, <7>; + amd,serdes-pq-skew = <10>, <10>, <30>; + amd,serdes-tx-amp = <15>, <15>, <10>; + amd,serdes-dfe-tap-config = <3>, <3>, <1>; + amd,serdes-dfe-tap-enable = <0>, <0>, <127>; + }; diff --git a/Documentation/devicetree/bindings/net/amd-xgbe.txt b/Documentation/devicetree/bindings/net/amd-xgbe.txt deleted file mode 100644 index 9c27dfcd1133..000000000000 --- a/Documentation/devicetree/bindings/net/amd-xgbe.txt +++ /dev/null @@ -1,76 +0,0 @@ -* AMD 10GbE driver (amd-xgbe) - -Required properties: -- compatible: Should be "amd,xgbe-seattle-v1a" -- reg: Address and length of the register sets for the device - - MAC registers - - PCS registers - - SerDes Rx/Tx registers - - SerDes integration registers (1/2) - - SerDes integration registers (2/2) -- interrupts: Should contain the amd-xgbe interrupt(s). The first interrupt - listed is required and is the general device interrupt. If the optional - amd,per-channel-interrupt property is specified, then one additional - interrupt for each DMA channel supported by the device should be specified. - The last interrupt listed should be the PCS auto-negotiation interrupt. -- clocks: - - DMA clock for the amd-xgbe device (used for calculating the - correct Rx interrupt watchdog timer value on a DMA channel - for coalescing) - - PTP clock for the amd-xgbe device -- clock-names: Should be the names of the clocks - - "dma_clk" for the DMA clock - - "ptp_clk" for the PTP clock -- phy-mode: See ethernet.txt file in the same directory - -Optional properties: -- dma-coherent: Present if dma operations are coherent -- amd,per-channel-interrupt: Indicates that Rx and Tx complete will generate - a unique interrupt for each DMA channel - this requires an additional - interrupt be configured for each DMA channel -- amd,speed-set: Speed capabilities of the device - 0 - 1GbE and 10GbE (default) - 1 - 2.5GbE and 10GbE - -The MAC address will be determined using the optional properties defined in -ethernet.txt. - -The following optional properties are represented by an array with each -value corresponding to a particular speed. The first array value represents -the setting for the 1GbE speed, the second value for the 2.5GbE speed and -the third value for the 10GbE speed. All three values are required if the -property is used. -- amd,serdes-blwc: Baseline wandering correction enablement - 0 - Off - 1 - On -- amd,serdes-cdr-rate: CDR rate speed selection -- amd,serdes-pq-skew: PQ (data sampling) skew -- amd,serdes-tx-amp: TX amplitude boost -- amd,serdes-dfe-tap-config: DFE taps available to run -- amd,serdes-dfe-tap-enable: DFE taps to enable - -Example: - xgbe@e0700000 { - compatible = "amd,xgbe-seattle-v1a"; - reg = <0 0xe0700000 0 0x80000>, - <0 0xe0780000 0 0x80000>, - <0 0xe1240800 0 0x00400>, - <0 0xe1250000 0 0x00060>, - <0 0xe1250080 0 0x00004>; - interrupt-parent = <&gic>; - interrupts = <0 325 4>, - <0 326 1>, <0 327 1>, <0 328 1>, <0 329 1>, - <0 323 4>; - amd,per-channel-interrupt; - clocks = <&xgbe_dma_clk>, <&xgbe_ptp_clk>; - clock-names = "dma_clk", "ptp_clk"; - phy-mode = "xgmii"; - mac-address = [ 02 a1 a2 a3 a4 a5 ]; - amd,speed-set = <0>; - amd,serdes-blwc = <1>, <1>, <0>; - amd,serdes-cdr-rate = <2>, <2>, <7>; - amd,serdes-pq-skew = <10>, <10>, <30>; - amd,serdes-tx-amp = <15>, <15>, <10>; - amd,serdes-dfe-tap-config = <3>, <3>, <1>; - amd,serdes-dfe-tap-enable = <0>, <0>, <127>; - }; From af3fce9f1bb41cdca32b24e7f19a902b7fe2906a Mon Sep 17 00:00:00 2001 From: Jiawen Wu Date: Tue, 14 Oct 2025 14:17:24 +0800 Subject: [PATCH 0041/1381] net: txgbe: expend SW-FW mailbox buffer size to identify QSFP module Recent firmware updates introduce additional fields in the mailbox message to provide more information for identifying 40G and 100G QSFP modules. To accommodate these new fields, expand the mailbox buffer size by 4 bytes. Without this change, drivers built against the updated firmware cannot properly identify modules due to mismatched mailbox message lengths. The old firmware version that used the smaller mailbox buffer has never been publicly released, so there are no backward-compatibility concerns. Signed-off-by: Jiawen Wu Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20251014061726.36660-2-jiawenwu@trustnetic.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/wangxun/txgbe/txgbe_type.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h b/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h index 41915d7dd372..9d53c7413f7b 100644 --- a/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h +++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h @@ -352,7 +352,9 @@ struct txgbe_sfp_id { u8 vendor_oui0; /* A0H 0x25 */ u8 vendor_oui1; /* A0H 0x26 */ u8 vendor_oui2; /* A0H 0x27 */ - u8 reserved[3]; + u8 transceiver_type; /* A0H 0x83 */ + u8 sff_opt1; /* A0H 0xC0 */ + u8 reserved[5]; }; struct txgbe_hic_i2c_read { From 1f863ce5c71276710a7689c88bf4003fa5173998 Mon Sep 17 00:00:00 2001 From: Jiawen Wu Date: Tue, 14 Oct 2025 14:17:25 +0800 Subject: [PATCH 0042/1381] net: txgbe: optimize the flow to setup PHY for AML devices To adapt to new firmware for AML devices, the driver should send the "SET_LINK_CMD" to the firmware only once when switching PHY interface mode, and no longer needs to re-trigger PHY configuration based on the RX signal interrupt (TXGBE_GPIOBIT_3). In previous firmware versions, the PHY was configured only after receiving "SET_LINK_CMD", and might remain incomplete if the RX signal was lost. To handle this case, the driver used TXGBE_GPIOBIT_3 interrupt to resend the command. This workaround is no longer necessary with the new firmware. And the unknown link speed is permitted in the mailbox buffer. Signed-off-by: Jiawen Wu Link: https://patch.msgid.link/20251014061726.36660-3-jiawenwu@trustnetic.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/wangxun/libwx/wx_type.h | 2 - .../net/ethernet/wangxun/txgbe/txgbe_aml.c | 50 ++++++------------- .../net/ethernet/wangxun/txgbe/txgbe_type.h | 1 + 3 files changed, 15 insertions(+), 38 deletions(-) diff --git a/drivers/net/ethernet/wangxun/libwx/wx_type.h b/drivers/net/ethernet/wangxun/libwx/wx_type.h index d89b9b8a0a2c..4880268b620e 100644 --- a/drivers/net/ethernet/wangxun/libwx/wx_type.h +++ b/drivers/net/ethernet/wangxun/libwx/wx_type.h @@ -1271,8 +1271,6 @@ struct wx { /* PHY stuff */ bool notify_down; - int adv_speed; - int adv_duplex; unsigned int link; int speed; int duplex; diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_aml.c b/drivers/net/ethernet/wangxun/txgbe/txgbe_aml.c index dc87ccad9652..1da92431c324 100644 --- a/drivers/net/ethernet/wangxun/txgbe/txgbe_aml.c +++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_aml.c @@ -19,8 +19,8 @@ void txgbe_gpio_init_aml(struct wx *wx) { u32 status; - wr32(wx, WX_GPIO_INTTYPE_LEVEL, TXGBE_GPIOBIT_2 | TXGBE_GPIOBIT_3); - wr32(wx, WX_GPIO_INTEN, TXGBE_GPIOBIT_2 | TXGBE_GPIOBIT_3); + wr32(wx, WX_GPIO_INTTYPE_LEVEL, TXGBE_GPIOBIT_2); + wr32(wx, WX_GPIO_INTEN, TXGBE_GPIOBIT_2); status = rd32(wx, WX_GPIO_INTSTATUS); for (int i = 0; i < 6; i++) { @@ -42,11 +42,6 @@ irqreturn_t txgbe_gpio_irq_handler_aml(int irq, void *data) wr32(wx, WX_GPIO_EOI, TXGBE_GPIOBIT_2); wx_service_event_schedule(wx); } - if (status & TXGBE_GPIOBIT_3) { - set_bit(WX_FLAG_NEED_LINK_CONFIG, wx->flags); - wx_service_event_schedule(wx); - wr32(wx, WX_GPIO_EOI, TXGBE_GPIOBIT_3); - } wr32(wx, WX_GPIO_INTMASK, 0); return IRQ_HANDLED; @@ -96,6 +91,9 @@ static int txgbe_set_phy_link_hostif(struct wx *wx, int speed, int autoneg, int case SPEED_10000: buffer.speed = TXGBE_LINK_SPEED_10GB_FULL; break; + default: + buffer.speed = TXGBE_LINK_SPEED_UNKNOWN; + break; } buffer.fec_mode = TXGBE_PHY_FEC_AUTO; @@ -106,19 +104,18 @@ static int txgbe_set_phy_link_hostif(struct wx *wx, int speed, int autoneg, int WX_HI_COMMAND_TIMEOUT, true); } -static void txgbe_get_link_capabilities(struct wx *wx) +static void txgbe_get_link_capabilities(struct wx *wx, int *speed, int *duplex) { struct txgbe *txgbe = wx->priv; if (test_bit(PHY_INTERFACE_MODE_25GBASER, txgbe->sfp_interfaces)) - wx->adv_speed = SPEED_25000; + *speed = SPEED_25000; else if (test_bit(PHY_INTERFACE_MODE_10GBASER, txgbe->sfp_interfaces)) - wx->adv_speed = SPEED_10000; + *speed = SPEED_10000; else - wx->adv_speed = SPEED_UNKNOWN; + *speed = SPEED_UNKNOWN; - wx->adv_duplex = wx->adv_speed == SPEED_UNKNOWN ? - DUPLEX_HALF : DUPLEX_FULL; + *duplex = *speed == SPEED_UNKNOWN ? DUPLEX_HALF : DUPLEX_FULL; } static void txgbe_get_phy_link(struct wx *wx, int *speed) @@ -138,23 +135,11 @@ static void txgbe_get_phy_link(struct wx *wx, int *speed) int txgbe_set_phy_link(struct wx *wx) { - int speed, err; - u32 gpio; + int speed, duplex, err; - /* Check RX signal */ - gpio = rd32(wx, WX_GPIO_EXT); - if (gpio & TXGBE_GPIOBIT_3) - return -ENODEV; + txgbe_get_link_capabilities(wx, &speed, &duplex); - txgbe_get_link_capabilities(wx); - if (wx->adv_speed == SPEED_UNKNOWN) - return -ENODEV; - - txgbe_get_phy_link(wx, &speed); - if (speed == wx->adv_speed) - return 0; - - err = txgbe_set_phy_link_hostif(wx, wx->adv_speed, 0, wx->adv_duplex); + err = txgbe_set_phy_link_hostif(wx, speed, 0, duplex); if (err) { wx_err(wx, "Failed to setup link\n"); return err; @@ -230,14 +215,7 @@ int txgbe_identify_sfp(struct wx *wx) return -ENODEV; } - err = txgbe_sfp_to_linkmodes(wx, id); - if (err) - return err; - - if (gpio & TXGBE_GPIOBIT_3) - set_bit(WX_FLAG_NEED_LINK_CONFIG, wx->flags); - - return 0; + return txgbe_sfp_to_linkmodes(wx, id); } void txgbe_setup_link(struct wx *wx) diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h b/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h index 9d53c7413f7b..b9a4ba48f5b9 100644 --- a/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h +++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h @@ -314,6 +314,7 @@ void txgbe_up(struct wx *wx); int txgbe_setup_tc(struct net_device *dev, u8 tc); void txgbe_do_reset(struct net_device *netdev); +#define TXGBE_LINK_SPEED_UNKNOWN 0 #define TXGBE_LINK_SPEED_10GB_FULL 4 #define TXGBE_LINK_SPEED_25GB_FULL 0x10 From a058de9262f4d0e58b8ed1d4a05758bbb92cc10e Mon Sep 17 00:00:00 2001 From: Jiawen Wu Date: Tue, 14 Oct 2025 14:17:26 +0800 Subject: [PATCH 0043/1381] net: txgbe: rename txgbe_get_phy_link() The function txgbe_get_phy_link() is more appropriately named txgbe_get_mac_link(), since it reads the link status from the MAC register. Signed-off-by: Jiawen Wu Link: https://patch.msgid.link/20251014061726.36660-4-jiawenwu@trustnetic.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/wangxun/txgbe/txgbe_aml.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_aml.c b/drivers/net/ethernet/wangxun/txgbe/txgbe_aml.c index 1da92431c324..35eebdb07761 100644 --- a/drivers/net/ethernet/wangxun/txgbe/txgbe_aml.c +++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_aml.c @@ -118,7 +118,7 @@ static void txgbe_get_link_capabilities(struct wx *wx, int *speed, int *duplex) *duplex = *speed == SPEED_UNKNOWN ? DUPLEX_HALF : DUPLEX_FULL; } -static void txgbe_get_phy_link(struct wx *wx, int *speed) +static void txgbe_get_mac_link(struct wx *wx, int *speed) { u32 status; @@ -234,7 +234,7 @@ static void txgbe_get_link_state(struct phylink_config *config, struct wx *wx = phylink_to_wx(config); int speed; - txgbe_get_phy_link(wx, &speed); + txgbe_get_mac_link(wx, &speed); state->link = speed != SPEED_UNKNOWN; state->speed = speed; state->duplex = state->link ? DUPLEX_FULL : DUPLEX_UNKNOWN; From 4a997d49d92ad9dda603f60881faa6c800d435e9 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Tue, 14 Oct 2025 23:54:54 +0000 Subject: [PATCH 0044/1381] tcp: Save lock_sock() for memcg in inet_csk_accept(). If memcg is enabled, accept() acquires lock_sock() twice for each new TCP/MPTCP socket in inet_csk_accept() and __inet_accept(). Let's move memcg operations from inet_csk_accept() to __inet_accept(). Note that SCTP somehow allocates a new socket by sk_alloc() in sk->sk_prot->accept() and clones fields manually, instead of using sk_clone_lock(). mem_cgroup_sk_alloc() is called for SCTP before __inet_accept(), so I added the protocol check in __inet_accept(), but this can be removed once SCTP uses sk_clone_lock(). Signed-off-by: Kuniyuki Iwashima Signed-off-by: Martin KaFai Lau Reviewed-by: Shakeel Butt Reviewed-by: Eric Dumazet Acked-by: Roman Gushchin Link: https://patch.msgid.link/20251014235604.3057003-2-kuniyu@google.com --- net/ipv4/af_inet.c | 22 ++++++++++++++++++++++ net/ipv4/inet_connection_sock.c | 25 ------------------------- 2 files changed, 22 insertions(+), 25 deletions(-) diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index 3109c5ec38f3..e8771faa5bbf 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -755,6 +755,28 @@ EXPORT_SYMBOL(inet_stream_connect); void __inet_accept(struct socket *sock, struct socket *newsock, struct sock *newsk) { + /* TODO: use sk_clone_lock() in SCTP and remove protocol checks */ + if (mem_cgroup_sockets_enabled && + (!IS_ENABLED(CONFIG_IP_SCTP) || sk_is_tcp(newsk))) { + gfp_t gfp = GFP_KERNEL | __GFP_NOFAIL; + + mem_cgroup_sk_alloc(newsk); + + if (mem_cgroup_from_sk(newsk)) { + int amt; + + /* The socket has not been accepted yet, no need + * to look at newsk->sk_wmem_queued. + */ + amt = sk_mem_pages(newsk->sk_forward_alloc + + atomic_read(&newsk->sk_rmem_alloc)); + if (amt) + mem_cgroup_sk_charge(newsk, amt, gfp); + } + + kmem_cache_charge(newsk, gfp); + } + sock_rps_record_flow(newsk); WARN_ON(!((1 << newsk->sk_state) & (TCPF_ESTABLISHED | TCPF_SYN_RECV | diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index cdd1e12aac8c..3b83b66b2284 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -712,31 +712,6 @@ struct sock *inet_csk_accept(struct sock *sk, struct proto_accept_arg *arg) release_sock(sk); - if (mem_cgroup_sockets_enabled) { - gfp_t gfp = GFP_KERNEL | __GFP_NOFAIL; - int amt = 0; - - /* atomically get the memory usage, set and charge the - * newsk->sk_memcg. - */ - lock_sock(newsk); - - mem_cgroup_sk_alloc(newsk); - if (mem_cgroup_from_sk(newsk)) { - /* The socket has not been accepted yet, no need - * to look at newsk->sk_wmem_queued. - */ - amt = sk_mem_pages(newsk->sk_forward_alloc + - atomic_read(&newsk->sk_rmem_alloc)); - } - - if (amt) - mem_cgroup_sk_charge(newsk, amt, gfp); - kmem_cache_charge(newsk, gfp); - - release_sock(newsk); - } - if (req) reqsk_put(req); From 7c268eaeec6388b7bee36aef3fb5e62c9222ad3b Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Tue, 14 Oct 2025 23:54:55 +0000 Subject: [PATCH 0045/1381] net: Allow opt-out from global protocol memory accounting. Some protocols (e.g., TCP, UDP) implement memory accounting for socket buffers and charge memory to per-protocol global counters pointed to by sk->sk_proto->memory_allocated. Sometimes, system processes do not want that limitation. For a similar purpose, there is SO_RESERVE_MEM for sockets under memcg. Also, by opting out of the per-protocol accounting, sockets under memcg can avoid paying costs for two orthogonal memory accounting mechanisms. A microbenchmark result is in the subsequent bpf patch. Let's allow opt-out from the per-protocol memory accounting if sk->sk_bypass_prot_mem is true. sk->sk_bypass_prot_mem and sk->sk_prot are placed in the same cache line, and sk_has_account() always fetches sk->sk_prot before accessing sk->sk_bypass_prot_mem, so there is no extra cache miss for this patch. The following patches will set sk->sk_bypass_prot_mem to true, and then, the per-protocol memory accounting will be skipped. Note that this does NOT disable memcg, but rather the per-protocol one. Another option not to use the hole in struct sock_common is create sk_prot variants like tcp_prot_bypass, but this would complicate SOCKMAP logic, tcp_bpf_prots etc. Signed-off-by: Kuniyuki Iwashima Signed-off-by: Martin KaFai Lau Reviewed-by: Shakeel Butt Reviewed-by: Eric Dumazet Acked-by: Roman Gushchin Link: https://patch.msgid.link/20251014235604.3057003-3-kuniyu@google.com --- include/net/proto_memory.h | 3 +++ include/net/sock.h | 3 +++ include/net/tcp.h | 3 +++ net/core/sock.c | 32 +++++++++++++++++++++++++------- net/ipv4/tcp.c | 3 ++- net/ipv4/tcp_output.c | 7 ++++++- net/mptcp/protocol.c | 7 ++++--- net/tls/tls_device.c | 3 ++- 8 files changed, 48 insertions(+), 13 deletions(-) diff --git a/include/net/proto_memory.h b/include/net/proto_memory.h index 8e91a8fa31b5..ad6d703ce6fe 100644 --- a/include/net/proto_memory.h +++ b/include/net/proto_memory.h @@ -35,6 +35,9 @@ static inline bool sk_under_memory_pressure(const struct sock *sk) mem_cgroup_sk_under_memory_pressure(sk)) return true; + if (sk->sk_bypass_prot_mem) + return false; + return !!READ_ONCE(*sk->sk_prot->memory_pressure); } diff --git a/include/net/sock.h b/include/net/sock.h index 30ac2eb4ef9b..415e7381aa50 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -118,6 +118,7 @@ typedef __u64 __bitwise __addrpair; * @skc_reuseport: %SO_REUSEPORT setting * @skc_ipv6only: socket is IPV6 only * @skc_net_refcnt: socket is using net ref counting + * @skc_bypass_prot_mem: bypass the per-protocol memory accounting for skb * @skc_bound_dev_if: bound device index if != 0 * @skc_bind_node: bind hash linkage for various protocol lookup tables * @skc_portaddr_node: second hash linkage for UDP/UDP-Lite protocol @@ -174,6 +175,7 @@ struct sock_common { unsigned char skc_reuseport:1; unsigned char skc_ipv6only:1; unsigned char skc_net_refcnt:1; + unsigned char skc_bypass_prot_mem:1; int skc_bound_dev_if; union { struct hlist_node skc_bind_node; @@ -381,6 +383,7 @@ struct sock { #define sk_reuseport __sk_common.skc_reuseport #define sk_ipv6only __sk_common.skc_ipv6only #define sk_net_refcnt __sk_common.skc_net_refcnt +#define sk_bypass_prot_mem __sk_common.skc_bypass_prot_mem #define sk_bound_dev_if __sk_common.skc_bound_dev_if #define sk_bind_node __sk_common.skc_bind_node #define sk_prot __sk_common.skc_prot diff --git a/include/net/tcp.h b/include/net/tcp.h index 1e547138f4fb..439e327fdbfa 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -303,6 +303,9 @@ static inline bool tcp_under_memory_pressure(const struct sock *sk) mem_cgroup_sk_under_memory_pressure(sk)) return true; + if (sk->sk_bypass_prot_mem) + return false; + return READ_ONCE(tcp_memory_pressure); } /* diff --git a/net/core/sock.c b/net/core/sock.c index 08ae20069b6d..5bf208579c02 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -1046,9 +1046,13 @@ static int sock_reserve_memory(struct sock *sk, int bytes) if (!charged) return -ENOMEM; + if (sk->sk_bypass_prot_mem) + goto success; + /* pre-charge to forward_alloc */ sk_memory_allocated_add(sk, pages); allocated = sk_memory_allocated(sk); + /* If the system goes into memory pressure with this * precharge, give up and return error. */ @@ -1057,6 +1061,8 @@ static int sock_reserve_memory(struct sock *sk, int bytes) mem_cgroup_sk_uncharge(sk, pages); return -ENOMEM; } + +success: sk_forward_alloc_add(sk, pages << PAGE_SHIFT); WRITE_ONCE(sk->sk_reserved_mem, @@ -3145,8 +3151,11 @@ bool sk_page_frag_refill(struct sock *sk, struct page_frag *pfrag) if (likely(skb_page_frag_refill(32U, pfrag, sk->sk_allocation))) return true; - sk_enter_memory_pressure(sk); + if (!sk->sk_bypass_prot_mem) + sk_enter_memory_pressure(sk); + sk_stream_moderate_sndbuf(sk); + return false; } EXPORT_SYMBOL(sk_page_frag_refill); @@ -3263,10 +3272,12 @@ int __sk_mem_raise_allocated(struct sock *sk, int size, int amt, int kind) { bool memcg_enabled = false, charged = false; struct proto *prot = sk->sk_prot; - long allocated; + long allocated = 0; - sk_memory_allocated_add(sk, amt); - allocated = sk_memory_allocated(sk); + if (!sk->sk_bypass_prot_mem) { + sk_memory_allocated_add(sk, amt); + allocated = sk_memory_allocated(sk); + } if (mem_cgroup_sk_enabled(sk)) { memcg_enabled = true; @@ -3275,6 +3286,9 @@ int __sk_mem_raise_allocated(struct sock *sk, int size, int amt, int kind) goto suppress_allocation; } + if (!allocated) + return 1; + /* Under limit. */ if (allocated <= sk_prot_mem_limits(sk, 0)) { sk_leave_memory_pressure(sk); @@ -3353,7 +3367,8 @@ int __sk_mem_raise_allocated(struct sock *sk, int size, int amt, int kind) trace_sock_exceed_buf_limit(sk, prot, allocated, kind); - sk_memory_allocated_sub(sk, amt); + if (allocated) + sk_memory_allocated_sub(sk, amt); if (charged) mem_cgroup_sk_uncharge(sk, amt); @@ -3392,11 +3407,14 @@ EXPORT_SYMBOL(__sk_mem_schedule); */ void __sk_mem_reduce_allocated(struct sock *sk, int amount) { - sk_memory_allocated_sub(sk, amount); - if (mem_cgroup_sk_enabled(sk)) mem_cgroup_sk_uncharge(sk, amount); + if (sk->sk_bypass_prot_mem) + return; + + sk_memory_allocated_sub(sk, amount); + if (sk_under_global_memory_pressure(sk) && (sk_memory_allocated(sk) < sk_prot_mem_limits(sk, 0))) sk_leave_memory_pressure(sk); diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 4d720aa09a4c..54def27326f1 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -928,7 +928,8 @@ struct sk_buff *tcp_stream_alloc_skb(struct sock *sk, gfp_t gfp, } __kfree_skb(skb); } else { - sk->sk_prot->enter_memory_pressure(sk); + if (!sk->sk_bypass_prot_mem) + tcp_enter_memory_pressure(sk); sk_stream_moderate_sndbuf(sk); } return NULL; diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index b94efb3050d2..7f5df7a71f62 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -3743,12 +3743,17 @@ void sk_forced_mem_schedule(struct sock *sk, int size) delta = size - sk->sk_forward_alloc; if (delta <= 0) return; + amt = sk_mem_pages(delta); sk_forward_alloc_add(sk, amt << PAGE_SHIFT); - sk_memory_allocated_add(sk, amt); if (mem_cgroup_sk_enabled(sk)) mem_cgroup_sk_charge(sk, amt, gfp_memcg_charge() | __GFP_NOFAIL); + + if (sk->sk_bypass_prot_mem) + return; + + sk_memory_allocated_add(sk, amt); } /* Send a FIN. The caller locks the socket for us. diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index 0292162a14ee..94a5f6dcc577 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -1065,11 +1065,12 @@ static void mptcp_enter_memory_pressure(struct sock *sk) mptcp_for_each_subflow(msk, subflow) { struct sock *ssk = mptcp_subflow_tcp_sock(subflow); - if (first) + if (first && !ssk->sk_bypass_prot_mem) { tcp_enter_memory_pressure(ssk); - sk_stream_moderate_sndbuf(ssk); + first = false; + } - first = false; + sk_stream_moderate_sndbuf(ssk); } __mptcp_sync_sndbuf(sk); } diff --git a/net/tls/tls_device.c b/net/tls/tls_device.c index a64ae15b1a60..caa2b5d24622 100644 --- a/net/tls/tls_device.c +++ b/net/tls/tls_device.c @@ -373,7 +373,8 @@ static int tls_do_allocation(struct sock *sk, if (!offload_ctx->open_record) { if (unlikely(!skb_page_frag_refill(prepend_size, pfrag, sk->sk_allocation))) { - READ_ONCE(sk->sk_prot)->enter_memory_pressure(sk); + if (!sk->sk_bypass_prot_mem) + READ_ONCE(sk->sk_prot)->enter_memory_pressure(sk); sk_stream_moderate_sndbuf(sk); return -ENOMEM; } From b46ab63181ff973ddce44ebc9ac24b269d42f481 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Tue, 14 Oct 2025 23:54:56 +0000 Subject: [PATCH 0046/1381] net: Introduce net.core.bypass_prot_mem sysctl. If a socket has sk->sk_bypass_prot_mem flagged, the socket opts out of the global protocol memory accounting. Let's control the flag by a new sysctl knob. The flag is written once during socket(2) and is inherited to child sockets. Tested with a script that creates local socket pairs and send()s a bunch of data without recv()ing. Setup: # mkdir /sys/fs/cgroup/test # echo $$ >> /sys/fs/cgroup/test/cgroup.procs # sysctl -q net.ipv4.tcp_mem="1000 1000 1000" # ulimit -n 524288 Without net.core.bypass_prot_mem, charged to tcp_mem & memcg # python3 pressure.py & # cat /sys/fs/cgroup/test/memory.stat | grep sock sock 22642688 <-------------------------------------- charged to memcg # cat /proc/net/sockstat| grep TCP TCP: inuse 2006 orphan 0 tw 0 alloc 2008 mem 5376 <-- charged to tcp_mem # ss -tn | head -n 5 State Recv-Q Send-Q Local Address:Port Peer Address:Port ESTAB 2000 0 127.0.0.1:34479 127.0.0.1:53188 ESTAB 2000 0 127.0.0.1:34479 127.0.0.1:49972 ESTAB 2000 0 127.0.0.1:34479 127.0.0.1:53868 ESTAB 2000 0 127.0.0.1:34479 127.0.0.1:53554 # nstat | grep Pressure || echo no pressure TcpExtTCPMemoryPressures 1 0.0 With net.core.bypass_prot_mem=1, charged to memcg only: # sysctl -q net.core.bypass_prot_mem=1 # python3 pressure.py & # cat /sys/fs/cgroup/test/memory.stat | grep sock sock 2757468160 <------------------------------------ charged to memcg # cat /proc/net/sockstat | grep TCP TCP: inuse 2006 orphan 0 tw 0 alloc 2008 mem 0 <- NOT charged to tcp_mem # ss -tn | head -n 5 State Recv-Q Send-Q Local Address:Port Peer Address:Port ESTAB 111000 0 127.0.0.1:36019 127.0.0.1:49026 ESTAB 110000 0 127.0.0.1:36019 127.0.0.1:45630 ESTAB 110000 0 127.0.0.1:36019 127.0.0.1:44870 ESTAB 111000 0 127.0.0.1:36019 127.0.0.1:45274 # nstat | grep Pressure || echo no pressure no pressure Signed-off-by: Kuniyuki Iwashima Signed-off-by: Martin KaFai Lau Reviewed-by: Shakeel Butt Reviewed-by: Eric Dumazet Acked-by: Roman Gushchin Link: https://patch.msgid.link/20251014235604.3057003-4-kuniyu@google.com --- Documentation/admin-guide/sysctl/net.rst | 8 ++++++++ include/net/netns/core.h | 1 + net/core/sock.c | 5 +++++ net/core/sysctl_net_core.c | 9 +++++++++ 4 files changed, 23 insertions(+) diff --git a/Documentation/admin-guide/sysctl/net.rst b/Documentation/admin-guide/sysctl/net.rst index 40749b3cd356..991773dcb9cf 100644 --- a/Documentation/admin-guide/sysctl/net.rst +++ b/Documentation/admin-guide/sysctl/net.rst @@ -212,6 +212,14 @@ mem_pcpu_rsv Per-cpu reserved forward alloc cache size in page units. Default 1MB per CPU. +bypass_prot_mem +--------------- + +Skip charging socket buffers to the global per-protocol memory +accounting controlled by net.ipv4.tcp_mem, net.ipv4.udp_mem, etc. + +Default: 0 (off) + rmem_default ------------ diff --git a/include/net/netns/core.h b/include/net/netns/core.h index cb9c3e4cd738..9ef3d70e5e9c 100644 --- a/include/net/netns/core.h +++ b/include/net/netns/core.h @@ -17,6 +17,7 @@ struct netns_core { int sysctl_optmem_max; u8 sysctl_txrehash; u8 sysctl_tstamp_allow_data; + u8 sysctl_bypass_prot_mem; #ifdef CONFIG_PROC_FS struct prot_inuse __percpu *prot_inuse; diff --git a/net/core/sock.c b/net/core/sock.c index 5bf208579c02..b78533fb9268 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -2306,8 +2306,13 @@ struct sock *sk_alloc(struct net *net, int family, gfp_t priority, * why we need sk_prot_creator -acme */ sk->sk_prot = sk->sk_prot_creator = prot; + + if (READ_ONCE(net->core.sysctl_bypass_prot_mem)) + sk->sk_bypass_prot_mem = 1; + sk->sk_kern_sock = kern; sock_lock_init(sk); + sk->sk_net_refcnt = kern ? 0 : 1; if (likely(sk->sk_net_refcnt)) { get_net_track(net, &sk->ns_tracker, priority); diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c index f79137826d7f..8d4decb2606f 100644 --- a/net/core/sysctl_net_core.c +++ b/net/core/sysctl_net_core.c @@ -683,6 +683,15 @@ static struct ctl_table netns_core_table[] = { .extra1 = SYSCTL_ZERO, .extra2 = SYSCTL_ONE }, + { + .procname = "bypass_prot_mem", + .data = &init_net.core.sysctl_bypass_prot_mem, + .maxlen = sizeof(u8), + .mode = 0644, + .proc_handler = proc_dou8vec_minmax, + .extra1 = SYSCTL_ZERO, + .extra2 = SYSCTL_ONE + }, /* sysctl_core_net_init() will set the values after this * to readonly in network namespaces */ From 5ed17896a09e2cc3f07b7d8f6a8b1d6be54550f7 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Tue, 14 Oct 2025 23:54:57 +0000 Subject: [PATCH 0047/1381] bpf: Support bpf_setsockopt() for BPF_CGROUP_INET_SOCK_CREATE. We will support flagging sk->sk_bypass_prot_mem via bpf_setsockopt() at the BPF_CGROUP_INET_SOCK_CREATE hook. BPF_CGROUP_INET_SOCK_CREATE is invoked by __cgroup_bpf_run_filter_sk() that passes a pointer to struct sock to the bpf prog as void *ctx. But there are no bpf_func_proto for bpf_setsockopt() that receives the ctx as a pointer to struct sock. Also, bpf_getsockopt() will be necessary for a cgroup with multiple bpf progs running. Let's add new bpf_setsockopt() and bpf_getsockopt() variants for BPF_CGROUP_INET_SOCK_CREATE. Note that inet_create() is not under lock_sock() and has the same semantics with bpf_lsm_unlocked_sockopt_hooks. Signed-off-by: Kuniyuki Iwashima Signed-off-by: Martin KaFai Lau Acked-by: Roman Gushchin Link: https://patch.msgid.link/20251014235604.3057003-5-kuniyu@google.com --- net/core/filter.c | 48 +++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 48 insertions(+) diff --git a/net/core/filter.c b/net/core/filter.c index 76628df1fc82..ed3f0e536059 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -5733,6 +5733,40 @@ static const struct bpf_func_proto bpf_sock_addr_getsockopt_proto = { .arg5_type = ARG_CONST_SIZE, }; +BPF_CALL_5(bpf_sock_create_setsockopt, struct sock *, sk, int, level, + int, optname, char *, optval, int, optlen) +{ + return __bpf_setsockopt(sk, level, optname, optval, optlen); +} + +static const struct bpf_func_proto bpf_sock_create_setsockopt_proto = { + .func = bpf_sock_create_setsockopt, + .gpl_only = false, + .ret_type = RET_INTEGER, + .arg1_type = ARG_PTR_TO_CTX, + .arg2_type = ARG_ANYTHING, + .arg3_type = ARG_ANYTHING, + .arg4_type = ARG_PTR_TO_MEM | MEM_RDONLY, + .arg5_type = ARG_CONST_SIZE, +}; + +BPF_CALL_5(bpf_sock_create_getsockopt, struct sock *, sk, int, level, + int, optname, char *, optval, int, optlen) +{ + return __bpf_getsockopt(sk, level, optname, optval, optlen); +} + +static const struct bpf_func_proto bpf_sock_create_getsockopt_proto = { + .func = bpf_sock_create_getsockopt, + .gpl_only = false, + .ret_type = RET_INTEGER, + .arg1_type = ARG_PTR_TO_CTX, + .arg2_type = ARG_ANYTHING, + .arg3_type = ARG_ANYTHING, + .arg4_type = ARG_PTR_TO_UNINIT_MEM, + .arg5_type = ARG_CONST_SIZE, +}; + BPF_CALL_5(bpf_sock_ops_setsockopt, struct bpf_sock_ops_kern *, bpf_sock, int, level, int, optname, char *, optval, int, optlen) { @@ -8062,6 +8096,20 @@ sock_filter_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog) return &bpf_sk_storage_get_cg_sock_proto; case BPF_FUNC_ktime_get_coarse_ns: return &bpf_ktime_get_coarse_ns_proto; + case BPF_FUNC_setsockopt: + switch (prog->expected_attach_type) { + case BPF_CGROUP_INET_SOCK_CREATE: + return &bpf_sock_create_setsockopt_proto; + default: + return NULL; + } + case BPF_FUNC_getsockopt: + switch (prog->expected_attach_type) { + case BPF_CGROUP_INET_SOCK_CREATE: + return &bpf_sock_create_getsockopt_proto; + default: + return NULL; + } default: return bpf_base_func_proto(func_id, prog); } From 38163af068810b388f6723a681dfd8c7b3680d38 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Tue, 14 Oct 2025 23:54:58 +0000 Subject: [PATCH 0048/1381] bpf: Introduce SK_BPF_BYPASS_PROT_MEM. If a socket has sk->sk_bypass_prot_mem flagged, the socket opts out of the global protocol memory accounting. This is easily controlled by net.core.bypass_prot_mem sysctl, but it lacks flexibility. Let's support flagging (and clearing) sk->sk_bypass_prot_mem via bpf_setsockopt() at the BPF_CGROUP_INET_SOCK_CREATE hook. int val = 1; bpf_setsockopt(ctx, SOL_SOCKET, SK_BPF_BYPASS_PROT_MEM, &val, sizeof(val)); As with net.core.bypass_prot_mem, this is inherited to child sockets, and BPF always takes precedence over sysctl at socket(2) and accept(2). SK_BPF_BYPASS_PROT_MEM is only supported at BPF_CGROUP_INET_SOCK_CREATE and not supported on other hooks for some reasons: 1. UDP charges memory under sk->sk_receive_queue.lock instead of lock_sock() 2. Modifying the flag after skb is charged to sk requires such adjustment during bpf_setsockopt() and complicates the logic unnecessarily We can support other hooks later if a real use case justifies that. Most changes are inline and hard to trace, but a microbenchmark on __sk_mem_raise_allocated() during neper/tcp_stream showed that more samples completed faster with sk->sk_bypass_prot_mem == 1. This will be more visible under tcp_mem pressure (but it's not a fair comparison). # bpftrace -e 'kprobe:__sk_mem_raise_allocated { @start[tid] = nsecs; } kretprobe:__sk_mem_raise_allocated /@start[tid]/ { @end[tid] = nsecs - @start[tid]; @times = hist(@end[tid]); delete(@start[tid]); }' # tcp_stream -6 -F 1000 -N -T 256 Without bpf prog: [128, 256) 3846 | | [256, 512) 1505326 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@| [512, 1K) 1371006 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ | [1K, 2K) 198207 |@@@@@@ | [2K, 4K) 31199 |@ | With bpf prog in the next patch: (must be attached before tcp_stream) # bpftool prog load sk_bypass_prot_mem.bpf.o /sys/fs/bpf/test type cgroup/sock_create # bpftool cgroup attach /sys/fs/cgroup/test cgroup_inet_sock_create pinned /sys/fs/bpf/test [128, 256) 6413 | | [256, 512) 1868425 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@| [512, 1K) 1101697 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ | [1K, 2K) 117031 |@@@@ | [2K, 4K) 11773 | | Signed-off-by: Kuniyuki Iwashima Signed-off-by: Martin KaFai Lau Acked-by: Roman Gushchin Link: https://patch.msgid.link/20251014235604.3057003-6-kuniyu@google.com --- include/uapi/linux/bpf.h | 2 ++ net/core/filter.c | 37 ++++++++++++++++++++++++++++++++++ tools/include/uapi/linux/bpf.h | 1 + 3 files changed, 40 insertions(+) diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index 6829936d33f5..6eb75ad900b1 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -7200,6 +7200,8 @@ enum { TCP_BPF_SYN_MAC = 1007, /* Copy the MAC, IP[46], and TCP header */ TCP_BPF_SOCK_OPS_CB_FLAGS = 1008, /* Get or Set TCP sock ops flags */ SK_BPF_CB_FLAGS = 1009, /* Get or set sock ops flags in socket */ + SK_BPF_BYPASS_PROT_MEM = 1010, /* Get or Set sk->sk_bypass_prot_mem */ + }; enum { diff --git a/net/core/filter.c b/net/core/filter.c index ed3f0e536059..16105f52927d 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -5733,9 +5733,37 @@ static const struct bpf_func_proto bpf_sock_addr_getsockopt_proto = { .arg5_type = ARG_CONST_SIZE, }; +static int sk_bpf_set_get_bypass_prot_mem(struct sock *sk, + char *optval, int optlen, + bool getopt) +{ + int val; + + if (optlen != sizeof(int)) + return -EINVAL; + + if (!sk_has_account(sk)) + return -EOPNOTSUPP; + + if (getopt) { + *(int *)optval = sk->sk_bypass_prot_mem; + return 0; + } + + val = *(int *)optval; + if (val < 0 || val > 1) + return -EINVAL; + + sk->sk_bypass_prot_mem = val; + return 0; +} + BPF_CALL_5(bpf_sock_create_setsockopt, struct sock *, sk, int, level, int, optname, char *, optval, int, optlen) { + if (level == SOL_SOCKET && optname == SK_BPF_BYPASS_PROT_MEM) + return sk_bpf_set_get_bypass_prot_mem(sk, optval, optlen, false); + return __bpf_setsockopt(sk, level, optname, optval, optlen); } @@ -5753,6 +5781,15 @@ static const struct bpf_func_proto bpf_sock_create_setsockopt_proto = { BPF_CALL_5(bpf_sock_create_getsockopt, struct sock *, sk, int, level, int, optname, char *, optval, int, optlen) { + if (level == SOL_SOCKET && optname == SK_BPF_BYPASS_PROT_MEM) { + int err = sk_bpf_set_get_bypass_prot_mem(sk, optval, optlen, true); + + if (err) + memset(optval, 0, optlen); + + return err; + } + return __bpf_getsockopt(sk, level, optname, optval, optlen); } diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h index 6829936d33f5..9b17d937edf7 100644 --- a/tools/include/uapi/linux/bpf.h +++ b/tools/include/uapi/linux/bpf.h @@ -7200,6 +7200,7 @@ enum { TCP_BPF_SYN_MAC = 1007, /* Copy the MAC, IP[46], and TCP header */ TCP_BPF_SOCK_OPS_CB_FLAGS = 1008, /* Get or Set TCP sock ops flags */ SK_BPF_CB_FLAGS = 1009, /* Get or set sock ops flags in socket */ + SK_BPF_BYPASS_PROT_MEM = 1010, /* Get or Set sk->sk_bypass_prot_mem */ }; enum { From 5f941dd87b0a82dd690821c6e0f427db87a4453b Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Tue, 14 Oct 2025 23:54:59 +0000 Subject: [PATCH 0049/1381] selftests/bpf: Add test for sk->sk_bypass_prot_mem. The test does the following for IPv4/IPv6 x TCP/UDP sockets with/without sk->sk_bypass_prot_mem, which can be turned on by net.core.bypass_prot_mem or bpf_setsockopt(SK_BPF_BYPASS_PROT_MEM). 1. Create socket pairs 2. Send NR_PAGES (32) of data (TCP consumes around 35 pages, and UDP consuems 66 pages due to skb overhead) 3. Read memory_allocated from sk->sk_prot->memory_allocated and sk->sk_prot->memory_per_cpu_fw_alloc 4. Check if unread data is charged to memory_allocated If sk->sk_bypass_prot_mem is set, memory_allocated should not be changed, but we allow a small error (up to 10 pages) in case other processes on the host use some amounts of TCP/UDP memory. The amount of allocated pages are buffered to per-cpu variable {tcp,udp}_memory_per_cpu_fw_alloc up to +/- net.core.mem_pcpu_rsv before reported to {tcp,udp}_memory_allocated. At 3., memory_allocated is calculated from the 2 variables at fentry of socket create function. We drain the receive queue only for UDP before close() because UDP recv queue is destroyed after RCU grace period. When I printed memory_allocated, UDP bypass cases sometimes saw the no-bypass case's leftover, but it's still in the small error range (<10 pages). bpf_trace_printk: memory_allocated: 0 <-- TCP no-bypass bpf_trace_printk: memory_allocated: 35 bpf_trace_printk: memory_allocated: 0 <-- TCP w/ sysctl bpf_trace_printk: memory_allocated: 0 bpf_trace_printk: memory_allocated: 0 <-- TCP w/ bpf bpf_trace_printk: memory_allocated: 0 bpf_trace_printk: memory_allocated: 0 <-- UDP no-bypass bpf_trace_printk: memory_allocated: 66 bpf_trace_printk: memory_allocated: 2 <-- UDP w/ sysctl (2 pages leftover) bpf_trace_printk: memory_allocated: 2 bpf_trace_printk: memory_allocated: 2 <-- UDP w/ bpf (2 pages leftover) bpf_trace_printk: memory_allocated: 2 We prefer finishing tests faster than oversleeping for call_rcu() + sk_destruct(). The test completes within 2s on QEMU (64 CPUs) w/ KVM. # time ./test_progs -t sk_bypass #371/1 sk_bypass_prot_mem/TCP :OK #371/2 sk_bypass_prot_mem/UDP :OK #371/3 sk_bypass_prot_mem/TCPv6:OK #371/4 sk_bypass_prot_mem/UDPv6:OK #371 sk_bypass_prot_mem:OK Summary: 1/4 PASSED, 0 SKIPPED, 0 FAILED real 0m1.481s user 0m0.181s sys 0m0.441s Signed-off-by: Kuniyuki Iwashima Signed-off-by: Martin KaFai Lau Acked-by: Roman Gushchin Link: https://patch.msgid.link/20251014235604.3057003-7-kuniyu@google.com --- .../bpf/prog_tests/sk_bypass_prot_mem.c | 292 ++++++++++++++++++ .../selftests/bpf/progs/sk_bypass_prot_mem.c | 104 +++++++ 2 files changed, 396 insertions(+) create mode 100644 tools/testing/selftests/bpf/prog_tests/sk_bypass_prot_mem.c create mode 100644 tools/testing/selftests/bpf/progs/sk_bypass_prot_mem.c diff --git a/tools/testing/selftests/bpf/prog_tests/sk_bypass_prot_mem.c b/tools/testing/selftests/bpf/prog_tests/sk_bypass_prot_mem.c new file mode 100644 index 000000000000..e4940583924b --- /dev/null +++ b/tools/testing/selftests/bpf/prog_tests/sk_bypass_prot_mem.c @@ -0,0 +1,292 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright 2025 Google LLC */ + +#include +#include "sk_bypass_prot_mem.skel.h" +#include "network_helpers.h" + +#define NR_PAGES 32 +#define NR_SOCKETS 2 +#define BUF_TOTAL (NR_PAGES * 4096 / NR_SOCKETS) +#define BUF_SINGLE 1024 +#define NR_SEND (BUF_TOTAL / BUF_SINGLE) + +struct test_case { + char name[8]; + int family; + int type; + int (*create_sockets)(struct test_case *test_case, int sk[], int len); + long (*get_memory_allocated)(struct test_case *test_case, struct sk_bypass_prot_mem *skel); +}; + +static int tcp_create_sockets(struct test_case *test_case, int sk[], int len) +{ + int server, i, err = 0; + + server = start_server(test_case->family, test_case->type, NULL, 0, 0); + if (!ASSERT_GE(server, 0, "start_server_str")) + return server; + + /* Keep for-loop so we can change NR_SOCKETS easily. */ + for (i = 0; i < len; i += 2) { + sk[i] = connect_to_fd(server, 0); + if (sk[i] < 0) { + ASSERT_GE(sk[i], 0, "connect_to_fd"); + err = sk[i]; + break; + } + + sk[i + 1] = accept(server, NULL, NULL); + if (sk[i + 1] < 0) { + ASSERT_GE(sk[i + 1], 0, "accept"); + err = sk[i + 1]; + break; + } + } + + close(server); + + return err; +} + +static int udp_create_sockets(struct test_case *test_case, int sk[], int len) +{ + int i, j, err, rcvbuf = BUF_TOTAL; + + /* Keep for-loop so we can change NR_SOCKETS easily. */ + for (i = 0; i < len; i += 2) { + sk[i] = start_server(test_case->family, test_case->type, NULL, 0, 0); + if (sk[i] < 0) { + ASSERT_GE(sk[i], 0, "start_server"); + return sk[i]; + } + + sk[i + 1] = connect_to_fd(sk[i], 0); + if (sk[i + 1] < 0) { + ASSERT_GE(sk[i + 1], 0, "connect_to_fd"); + return sk[i + 1]; + } + + err = connect_fd_to_fd(sk[i], sk[i + 1], 0); + if (err) { + ASSERT_EQ(err, 0, "connect_fd_to_fd"); + return err; + } + + for (j = 0; j < 2; j++) { + err = setsockopt(sk[i + j], SOL_SOCKET, SO_RCVBUF, &rcvbuf, sizeof(int)); + if (err) { + ASSERT_EQ(err, 0, "setsockopt(SO_RCVBUF)"); + return err; + } + } + } + + return 0; +} + +static long get_memory_allocated(struct test_case *test_case, + bool *activated, long *memory_allocated) +{ + int sk; + + *activated = true; + + /* AF_INET and AF_INET6 share the same memory_allocated. + * tcp_init_sock() is called by AF_INET and AF_INET6, + * but udp_lib_init_sock() is inline. + */ + sk = socket(AF_INET, test_case->type, 0); + if (!ASSERT_GE(sk, 0, "get_memory_allocated")) + return -1; + + close(sk); + + return *memory_allocated; +} + +static long tcp_get_memory_allocated(struct test_case *test_case, struct sk_bypass_prot_mem *skel) +{ + return get_memory_allocated(test_case, + &skel->bss->tcp_activated, + &skel->bss->tcp_memory_allocated); +} + +static long udp_get_memory_allocated(struct test_case *test_case, struct sk_bypass_prot_mem *skel) +{ + return get_memory_allocated(test_case, + &skel->bss->udp_activated, + &skel->bss->udp_memory_allocated); +} + +static int check_bypass(struct test_case *test_case, + struct sk_bypass_prot_mem *skel, bool bypass) +{ + char buf[BUF_SINGLE] = {}; + long memory_allocated[2]; + int sk[NR_SOCKETS]; + int err, i, j; + + for (i = 0; i < ARRAY_SIZE(sk); i++) + sk[i] = -1; + + err = test_case->create_sockets(test_case, sk, ARRAY_SIZE(sk)); + if (err) + goto close; + + memory_allocated[0] = test_case->get_memory_allocated(test_case, skel); + + /* allocate pages >= NR_PAGES */ + for (i = 0; i < ARRAY_SIZE(sk); i++) { + for (j = 0; j < NR_SEND; j++) { + int bytes = send(sk[i], buf, sizeof(buf), 0); + + /* Avoid too noisy logs when something failed. */ + if (bytes != sizeof(buf)) { + ASSERT_EQ(bytes, sizeof(buf), "send"); + if (bytes < 0) { + err = bytes; + goto drain; + } + } + } + } + + memory_allocated[1] = test_case->get_memory_allocated(test_case, skel); + + if (bypass) + ASSERT_LE(memory_allocated[1], memory_allocated[0] + 10, "bypass"); + else + ASSERT_GT(memory_allocated[1], memory_allocated[0] + NR_PAGES, "no bypass"); + +drain: + if (test_case->type == SOCK_DGRAM) { + /* UDP starts purging sk->sk_receive_queue after one RCU + * grace period, then udp_memory_allocated goes down, + * so drain the queue before close(). + */ + for (i = 0; i < ARRAY_SIZE(sk); i++) { + for (j = 0; j < NR_SEND; j++) { + int bytes = recv(sk[i], buf, 1, MSG_DONTWAIT | MSG_TRUNC); + + if (bytes == sizeof(buf)) + continue; + if (bytes != -1 || errno != EAGAIN) + PRINT_FAIL("bytes: %d, errno: %s\n", bytes, strerror(errno)); + break; + } + } + } + +close: + for (i = 0; i < ARRAY_SIZE(sk); i++) { + if (sk[i] < 0) + break; + + close(sk[i]); + } + + return err; +} + +static void run_test(struct test_case *test_case) +{ + struct sk_bypass_prot_mem *skel; + struct nstoken *nstoken; + int cgroup, err; + + skel = sk_bypass_prot_mem__open_and_load(); + if (!ASSERT_OK_PTR(skel, "open_and_load")) + return; + + skel->bss->nr_cpus = libbpf_num_possible_cpus(); + + err = sk_bypass_prot_mem__attach(skel); + if (!ASSERT_OK(err, "attach")) + goto destroy_skel; + + cgroup = test__join_cgroup("/sk_bypass_prot_mem"); + if (!ASSERT_GE(cgroup, 0, "join_cgroup")) + goto destroy_skel; + + err = make_netns("sk_bypass_prot_mem"); + if (!ASSERT_EQ(err, 0, "make_netns")) + goto close_cgroup; + + nstoken = open_netns("sk_bypass_prot_mem"); + if (!ASSERT_OK_PTR(nstoken, "open_netns")) + goto remove_netns; + + err = check_bypass(test_case, skel, false); + if (!ASSERT_EQ(err, 0, "test_bypass(false)")) + goto close_netns; + + err = write_sysctl("/proc/sys/net/core/bypass_prot_mem", "1"); + if (!ASSERT_EQ(err, 0, "write_sysctl(1)")) + goto close_netns; + + err = check_bypass(test_case, skel, true); + if (!ASSERT_EQ(err, 0, "test_bypass(true by sysctl)")) + goto close_netns; + + err = write_sysctl("/proc/sys/net/core/bypass_prot_mem", "0"); + if (!ASSERT_EQ(err, 0, "write_sysctl(0)")) + goto close_netns; + + skel->links.sock_create = bpf_program__attach_cgroup(skel->progs.sock_create, cgroup); + if (!ASSERT_OK_PTR(skel->links.sock_create, "attach_cgroup(sock_create)")) + goto close_netns; + + err = check_bypass(test_case, skel, true); + ASSERT_EQ(err, 0, "test_bypass(true by bpf)"); + +close_netns: + close_netns(nstoken); +remove_netns: + remove_netns("sk_bypass_prot_mem"); +close_cgroup: + close(cgroup); +destroy_skel: + sk_bypass_prot_mem__destroy(skel); +} + +static struct test_case test_cases[] = { + { + .name = "TCP ", + .family = AF_INET, + .type = SOCK_STREAM, + .create_sockets = tcp_create_sockets, + .get_memory_allocated = tcp_get_memory_allocated, + }, + { + .name = "UDP ", + .family = AF_INET, + .type = SOCK_DGRAM, + .create_sockets = udp_create_sockets, + .get_memory_allocated = udp_get_memory_allocated, + }, + { + .name = "TCPv6", + .family = AF_INET6, + .type = SOCK_STREAM, + .create_sockets = tcp_create_sockets, + .get_memory_allocated = tcp_get_memory_allocated, + }, + { + .name = "UDPv6", + .family = AF_INET6, + .type = SOCK_DGRAM, + .create_sockets = udp_create_sockets, + .get_memory_allocated = udp_get_memory_allocated, + }, +}; + +void serial_test_sk_bypass_prot_mem(void) +{ + int i; + + for (i = 0; i < ARRAY_SIZE(test_cases); i++) { + if (test__start_subtest(test_cases[i].name)) + run_test(&test_cases[i]); + } +} diff --git a/tools/testing/selftests/bpf/progs/sk_bypass_prot_mem.c b/tools/testing/selftests/bpf/progs/sk_bypass_prot_mem.c new file mode 100644 index 000000000000..09a00d11ffcc --- /dev/null +++ b/tools/testing/selftests/bpf/progs/sk_bypass_prot_mem.c @@ -0,0 +1,104 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright 2025 Google LLC */ + +#include "bpf_tracing_net.h" +#include +#include +#include + +extern int tcp_memory_per_cpu_fw_alloc __ksym; +extern int udp_memory_per_cpu_fw_alloc __ksym; + +int nr_cpus; +bool tcp_activated, udp_activated; +long tcp_memory_allocated, udp_memory_allocated; + +struct sk_prot { + long *memory_allocated; + int *memory_per_cpu_fw_alloc; +}; + +static int drain_memory_per_cpu_fw_alloc(__u32 i, struct sk_prot *sk_prot_ctx) +{ + int *memory_per_cpu_fw_alloc; + + memory_per_cpu_fw_alloc = bpf_per_cpu_ptr(sk_prot_ctx->memory_per_cpu_fw_alloc, i); + if (memory_per_cpu_fw_alloc) + *sk_prot_ctx->memory_allocated += *memory_per_cpu_fw_alloc; + + return 0; +} + +static long get_memory_allocated(struct sock *_sk, int *memory_per_cpu_fw_alloc) +{ + struct sock *sk = bpf_core_cast(_sk, struct sock); + struct sk_prot sk_prot_ctx; + long memory_allocated; + + /* net_aligned_data.{tcp,udp}_memory_allocated was not available. */ + memory_allocated = sk->__sk_common.skc_prot->memory_allocated->counter; + + sk_prot_ctx.memory_allocated = &memory_allocated; + sk_prot_ctx.memory_per_cpu_fw_alloc = memory_per_cpu_fw_alloc; + + bpf_loop(nr_cpus, drain_memory_per_cpu_fw_alloc, &sk_prot_ctx, 0); + + return memory_allocated; +} + +static void fentry_init_sock(struct sock *sk, bool *activated, + long *memory_allocated, int *memory_per_cpu_fw_alloc) +{ + if (!*activated) + return; + + *memory_allocated = get_memory_allocated(sk, memory_per_cpu_fw_alloc); + *activated = false; +} + +SEC("fentry/tcp_init_sock") +int BPF_PROG(fentry_tcp_init_sock, struct sock *sk) +{ + fentry_init_sock(sk, &tcp_activated, + &tcp_memory_allocated, &tcp_memory_per_cpu_fw_alloc); + return 0; +} + +SEC("fentry/udp_init_sock") +int BPF_PROG(fentry_udp_init_sock, struct sock *sk) +{ + fentry_init_sock(sk, &udp_activated, + &udp_memory_allocated, &udp_memory_per_cpu_fw_alloc); + return 0; +} + +SEC("cgroup/sock_create") +int sock_create(struct bpf_sock *ctx) +{ + int err, val = 1; + + err = bpf_setsockopt(ctx, SOL_SOCKET, SK_BPF_BYPASS_PROT_MEM, + &val, sizeof(val)); + if (err) + goto err; + + val = 0; + + err = bpf_getsockopt(ctx, SOL_SOCKET, SK_BPF_BYPASS_PROT_MEM, + &val, sizeof(val)); + if (err) + goto err; + + if (val != 1) { + err = -EINVAL; + goto err; + } + + return 1; + +err: + bpf_set_retval(err); + return 0; +} + +char LICENSE[] SEC("license") = "GPL"; From bd853a59a87ecf9fc060dd54b10a3a94e9786182 Mon Sep 17 00:00:00 2001 From: Alok Tiwari Date: Tue, 14 Oct 2025 19:57:43 -0700 Subject: [PATCH 0050/1381] net: amd-xgbe: use EOPNOTSUPP instead of ENOTSUPP in xgbe_phy_mii_read_c45 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The MDIO read callback xgbe_phy_mii_read_c45() can propagate its return value up through phylink_mii_ioctl() to user space via netdev ioctls such as SIOCGMIIREG. Returning ENOTSUPP results in user space seeing "Unknown error", since ENOTSUPP is not a standard errno value. Replace ENOTSUPP with EOPNOTSUPP to align with the MDIO core’s usage and ensure user space receives a proper "Operation not supported" error instead of an unknown code. Signed-off-by: Alok Tiwari Reviewed-by: Simon Horman Acked-by: Shyam Sundar S K Link: https://patch.msgid.link/20251015025751.1532149-1-alok.a.tiwari@oracle.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/amd/xgbe/xgbe-phy-v2.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-phy-v2.c b/drivers/net/ethernet/amd/xgbe/xgbe-phy-v2.c index a56efc1bee33..35a381a83647 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-phy-v2.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-phy-v2.c @@ -668,7 +668,7 @@ static int xgbe_phy_mii_read_c45(struct mii_bus *mii, int addr, int devad, else if (phy_data->conn_type & XGBE_CONN_TYPE_MDIO) ret = xgbe_phy_mdio_mii_read_c45(pdata, addr, devad, reg); else - ret = -ENOTSUPP; + ret = -EOPNOTSUPP; xgbe_phy_put_comm_ownership(pdata); From 38c31c2620de4e570539a2a461eb62d0e7e692f7 Mon Sep 17 00:00:00 2001 From: Hangbin Liu Date: Wed, 15 Oct 2025 08:36:49 +0000 Subject: [PATCH 0051/1381] netdevsim: add ipsec hw_features Currently, netdevsim only sets dev->features, which makes the ESP features fixed. For example: # ethtool -k eni0np1 | grep esp tx-esp-segmentation: on [fixed] esp-hw-offload: on [fixed] esp-tx-csum-hw-offload: on [fixed] This patch adds the ESP features to hw_features, allowing them to be changed manually. For example: # ethtool -k eni0np1 | grep esp tx-esp-segmentation: on esp-hw-offload: on esp-tx-csum-hw-offload: on Suggested-by: Sabrina Dubroca Signed-off-by: Hangbin Liu Reviewed-by: Sabrina Dubroca Link: https://patch.msgid.link/20251015083649.54744-1-liuhangbin@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/netdevsim/ipsec.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/netdevsim/ipsec.c b/drivers/net/netdevsim/ipsec.c index 47cdee5577d4..36a1be4923d6 100644 --- a/drivers/net/netdevsim/ipsec.c +++ b/drivers/net/netdevsim/ipsec.c @@ -277,6 +277,7 @@ void nsim_ipsec_init(struct netdevsim *ns) NETIF_F_GSO_ESP) ns->netdev->features |= NSIM_ESP_FEATURES; + ns->netdev->hw_features |= NSIM_ESP_FEATURES; ns->netdev->hw_enc_features |= NSIM_ESP_FEATURES; ns->ipsec.pfile = debugfs_create_file("ipsec", 0400, From e1048520750dd9369ec97554ab308ff1ff932ec6 Mon Sep 17 00:00:00 2001 From: Harshit Mogalapalli Date: Wed, 15 Oct 2025 02:01:17 -0700 Subject: [PATCH 0052/1381] Octeontx2-af: Fix pci_alloc_irq_vectors() return value check In cgx_probe() when pci_alloc_irq_vectors() fails the error value will be negative and that check is sufficient. err = pci_alloc_irq_vectors(pdev, nvec, nvec, PCI_IRQ_MSIX); if (err < 0 || err != nvec) { ... } When pci_alloc_irq_vectors() fail to allocate nvec number of vectors, -ENOSPC is returned, so it would be safe to remove the check that compares err with nvec. Suggested-by: Paolo Abeni Signed-off-by: Harshit Mogalapalli Reviewed-by: Simon Horman Link: https://patch.msgid.link/20251015090117.1557870-1-harshit.m.mogalapalli@oracle.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/marvell/octeontx2/af/cgx.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/marvell/octeontx2/af/cgx.c b/drivers/net/ethernet/marvell/octeontx2/af/cgx.c index ec0e11c77cbf..42044cd810b1 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/cgx.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/cgx.c @@ -1994,7 +1994,7 @@ static int cgx_probe(struct pci_dev *pdev, const struct pci_device_id *id) nvec = pci_msix_vec_count(cgx->pdev); err = pci_alloc_irq_vectors(pdev, nvec, nvec, PCI_IRQ_MSIX); - if (err < 0 || err != nvec) { + if (err < 0) { dev_err(dev, "Request for %d msix vectors failed, err %d\n", nvec, err); goto err_release_regions; From f18c231fb12a9662bb9ebd7613e8f71fc497baf1 Mon Sep 17 00:00:00 2001 From: Jan Vaclav Date: Wed, 15 Oct 2025 12:10:02 +0200 Subject: [PATCH 0053/1381] net/hsr: add interlink to fill_info output Currently, it is possible to configure the interlink port, but no way to read it back from userspace. Add it to the output of hsr_fill_info(), so it can be read from userspace, for example: $ ip -d link show hsr0 12: hsr0: mtu ... ... hsr slave1 veth0 slave2 veth1 interlink veth2 ... Signed-off-by: Jan Vaclav Reviewed-by: Fernando Fernandez Mancera Link: https://patch.msgid.link/20251015101001.25670-2-jvaclav@redhat.com Signed-off-by: Jakub Kicinski --- net/hsr/hsr_netlink.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/net/hsr/hsr_netlink.c b/net/hsr/hsr_netlink.c index 4461adf69623..851187130755 100644 --- a/net/hsr/hsr_netlink.c +++ b/net/hsr/hsr_netlink.c @@ -160,6 +160,12 @@ static int hsr_fill_info(struct sk_buff *skb, const struct net_device *dev) goto nla_put_failure; } + port = hsr_port_get_hsr(hsr, HSR_PT_INTERLINK); + if (port) { + if (nla_put_u32(skb, IFLA_HSR_INTERLINK, port->dev->ifindex)) + goto nla_put_failure; + } + if (nla_put(skb, IFLA_HSR_SUPERVISION_ADDR, ETH_ALEN, hsr->sup_multicast_addr) || nla_put_u16(skb, IFLA_HSR_SEQ_NR, hsr->sequence_nr)) From 12a7b7bc14273e19a57a49e0ab92e7eb635de8a4 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Wed, 15 Oct 2025 17:10:41 +0100 Subject: [PATCH 0054/1381] net: stmmac: dwc-qos-eth: move MDIO bus locking into stmmac_mdio Rather than dwc-qos-eth manipulating the MDIO bus lock directly, add helpers to the stmmac MDIO layer and use them in dwc-qos-eth. This improves my commit 87f43e6f06a2 ("net: stmmac: dwc-qos: calibrate tegra with mdio bus idle"). Signed-off-by: Russell King (Oracle) Reviewed-by: Maxime Chevallier Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/E1v945J-0000000AmeJ-1GOb@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../ethernet/stmicro/stmmac/dwmac-dwc-qos-eth.c | 4 ++-- drivers/net/ethernet/stmicro/stmmac/stmmac.h | 2 ++ drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c | 14 ++++++++++++++ 3 files changed, 18 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-dwc-qos-eth.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-dwc-qos-eth.c index e8539cad4602..f1c2e35badf7 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-dwc-qos-eth.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-dwc-qos-eth.c @@ -162,7 +162,7 @@ static void tegra_eqos_fix_speed(void *bsp_priv, int speed, unsigned int mode) priv = netdev_priv(dev_get_drvdata(eqos->dev)); /* Calibration should be done with the MDIO bus idle */ - mutex_lock(&priv->mii->mdio_lock); + stmmac_mdio_lock(priv); /* calibrate */ value = readl(eqos->regs + SDMEMCOMPPADCTRL); @@ -198,7 +198,7 @@ static void tegra_eqos_fix_speed(void *bsp_priv, int speed, unsigned int mode) value &= ~SDMEMCOMPPADCTRL_PAD_E_INPUT_OR_E_PWRD; writel(value, eqos->regs + SDMEMCOMPPADCTRL); - mutex_unlock(&priv->mii->mdio_lock); + stmmac_mdio_unlock(priv); } else { value = readl(eqos->regs + AUTO_CAL_CONFIG); value &= ~AUTO_CAL_CONFIG_ENABLE; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac.h b/drivers/net/ethernet/stmicro/stmmac/stmmac.h index 7ca5477be390..ec8bddc1c37f 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac.h +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac.h @@ -388,6 +388,8 @@ static inline bool stmmac_wol_enabled_phy(struct stmmac_priv *priv) int stmmac_mdio_unregister(struct net_device *ndev); int stmmac_mdio_register(struct net_device *ndev); int stmmac_mdio_reset(struct mii_bus *mii); +void stmmac_mdio_lock(struct stmmac_priv *priv); +void stmmac_mdio_unlock(struct stmmac_priv *priv); int stmmac_pcs_setup(struct net_device *ndev); void stmmac_pcs_clean(struct net_device *ndev); void stmmac_set_ethtool_ops(struct net_device *netdev); diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c index f408737f6fc7..d62b2870899d 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c @@ -734,3 +734,17 @@ int stmmac_mdio_unregister(struct net_device *ndev) return 0; } + +void stmmac_mdio_lock(struct stmmac_priv *priv) +{ + if (priv->mii) + mutex_lock(&priv->mii->mdio_lock); +} +EXPORT_SYMBOL_GPL(stmmac_mdio_lock); + +void stmmac_mdio_unlock(struct stmmac_priv *priv) +{ + if (priv->mii) + mutex_unlock(&priv->mii->mdio_lock); +} +EXPORT_SYMBOL_GPL(stmmac_mdio_unlock); From 0bc832a54d274c15c54f2a55995a485e490309b9 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Wed, 15 Oct 2025 17:10:46 +0100 Subject: [PATCH 0055/1381] net: stmmac: place .mac_finish() method more appropriately Place the .mac_finish() initialiser and implementation after the .mac_config() initialiser and method which reflects the order that they appear in struct phylink_mac_ops, and the order in which they are called. This keeps logically similar code together. Signed-off-by: Russell King (Oracle) Reviewed-by: Maxime Chevallier Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/E1v945O-0000000AmeP-1k0t@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../net/ethernet/stmicro/stmmac/stmmac_main.c | 26 +++++++++---------- 1 file changed, 13 insertions(+), 13 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index 650d75b73e0b..3728afa701c6 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -859,6 +859,18 @@ static void stmmac_mac_config(struct phylink_config *config, unsigned int mode, /* Nothing to do, xpcs_config() handles everything */ } +static int stmmac_mac_finish(struct phylink_config *config, unsigned int mode, + phy_interface_t interface) +{ + struct net_device *ndev = to_net_dev(config->dev); + struct stmmac_priv *priv = netdev_priv(ndev); + + if (priv->plat->mac_finish) + priv->plat->mac_finish(ndev, priv->plat->bsp_priv, mode, interface); + + return 0; +} + static void stmmac_mac_link_down(struct phylink_config *config, unsigned int mode, phy_interface_t interface) { @@ -1053,27 +1065,15 @@ static int stmmac_mac_enable_tx_lpi(struct phylink_config *config, u32 timer, return 0; } -static int stmmac_mac_finish(struct phylink_config *config, unsigned int mode, - phy_interface_t interface) -{ - struct net_device *ndev = to_net_dev(config->dev); - struct stmmac_priv *priv = netdev_priv(ndev); - - if (priv->plat->mac_finish) - priv->plat->mac_finish(ndev, priv->plat->bsp_priv, mode, interface); - - return 0; -} - static const struct phylink_mac_ops stmmac_phylink_mac_ops = { .mac_get_caps = stmmac_mac_get_caps, .mac_select_pcs = stmmac_mac_select_pcs, .mac_config = stmmac_mac_config, + .mac_finish = stmmac_mac_finish, .mac_link_down = stmmac_mac_link_down, .mac_link_up = stmmac_mac_link_up, .mac_disable_tx_lpi = stmmac_mac_disable_tx_lpi, .mac_enable_tx_lpi = stmmac_mac_enable_tx_lpi, - .mac_finish = stmmac_mac_finish, }; /** From e82c64be9b45a1fec556c8fbd35f30ccaf852b6a Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Wed, 15 Oct 2025 17:10:51 +0100 Subject: [PATCH 0056/1381] net: stmmac: avoid PHY speed change when configuring MTU There is no need to do the speed-down, speed-up dance when changing the MTU as there is little power saving that can be gained from such a brief interval between these, and the autonegotiation they cause takes much longer. Move the calls to phylink_speed_up() and phylink_speed_down() into stmmac_open() and stmmac_release() respectively, reducing the work done in the __-variants of these functions. Signed-off-by: Russell King (Oracle) Reviewed-by: Maxime Chevallier Reviewed-by: Gatien Chevallier Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/E1v945T-0000000AmeV-2BvU@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../net/ethernet/stmicro/stmmac/stmmac_main.c | 19 ++++++++++--------- 1 file changed, 10 insertions(+), 9 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index 3728afa701c6..500cfd19e6b5 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -3963,8 +3963,6 @@ static int __stmmac_open(struct net_device *dev, stmmac_init_coalesce(priv); phylink_start(priv->phylink); - /* We may have called phylink_speed_down before */ - phylink_speed_up(priv->phylink); ret = stmmac_request_irq(dev); if (ret) @@ -4015,6 +4013,9 @@ static int stmmac_open(struct net_device *dev) kfree(dma_conf); + /* We may have called phylink_speed_down before */ + phylink_speed_up(priv->phylink); + return ret; err_disconnect_phy: @@ -4032,13 +4033,6 @@ static void __stmmac_release(struct net_device *dev) struct stmmac_priv *priv = netdev_priv(dev); u32 chan; - /* If the PHY or MAC has WoL enabled, then the PHY will not be - * suspended when phylink_stop() is called below. Set the PHY - * to its slowest speed to save power. - */ - if (device_may_wakeup(priv->device)) - phylink_speed_down(priv->phylink, false); - /* Stop and disconnect the PHY */ phylink_stop(priv->phylink); @@ -4078,6 +4072,13 @@ static int stmmac_release(struct net_device *dev) { struct stmmac_priv *priv = netdev_priv(dev); + /* If the PHY or MAC has WoL enabled, then the PHY will not be + * suspended when phylink_stop() is called below. Set the PHY + * to its slowest speed to save power. + */ + if (device_may_wakeup(priv->device)) + phylink_speed_down(priv->phylink, false); + __stmmac_release(dev); phylink_disconnect_phy(priv->phylink); From 07d91ec99a8a37d9b9f6d8ec3f16b97622a74893 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Wed, 15 Oct 2025 17:10:56 +0100 Subject: [PATCH 0057/1381] net: stmmac: rearrange tc_init() To make future changes easier, rearrange the use of dma_cap->l3l4fnum vs priv->flow_entries_max. Always initialise priv->flow_entries_max from dma_cap->l3l4fnum, then use priv->flow_entries_max to determine whether we allocate priv->flow_entries and set it up. This change is safe because tc_init() is only called once from stmmac_dvr_probe(). Signed-off-by: Russell King (Oracle) Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/E1v945Y-0000000Ameb-2gDI@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/stmmac_tc.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_tc.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_tc.c index 97e89a604abd..ef65cf511f3e 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_tc.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_tc.c @@ -262,10 +262,10 @@ static int tc_init(struct stmmac_priv *priv) unsigned int count; int ret, i; - if (dma_cap->l3l4fnum) { - priv->flow_entries_max = dma_cap->l3l4fnum; + priv->flow_entries_max = dma_cap->l3l4fnum; + if (priv->flow_entries_max) { priv->flow_entries = devm_kcalloc(priv->device, - dma_cap->l3l4fnum, + priv->flow_entries_max, sizeof(*priv->flow_entries), GFP_KERNEL); if (!priv->flow_entries) From 4a4094ba7ad2563a96f0482db53d042b695892b0 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Wed, 15 Oct 2025 17:11:01 +0100 Subject: [PATCH 0058/1381] net: stmmac: rename stmmac_phy_setup() to include phylink stmmac_phy_setup() does not set up any PHY, but does setup phylink. Rename this function to stmmac_phylink_setup() to reflect more what it is doing. Signed-off-by: Russell King (Oracle) Reviewed-by: Maxime Chevallier Reviewed-by: Gatien Chevallier Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/E1v945d-0000000Ameh-3Bs7@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index 500cfd19e6b5..c9fa965c8566 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -1185,7 +1185,7 @@ static int stmmac_init_phy(struct net_device *dev) return 0; } -static int stmmac_phy_setup(struct stmmac_priv *priv) +static int stmmac_phylink_setup(struct stmmac_priv *priv) { struct stmmac_mdio_bus_data *mdio_bus_data; struct phylink_config *config; @@ -7642,7 +7642,7 @@ int stmmac_dvr_probe(struct device *device, if (ret) goto error_pcs_setup; - ret = stmmac_phy_setup(priv); + ret = stmmac_phylink_setup(priv); if (ret) { netdev_err(ndev, "failed to setup phy (%d)\n", ret); goto error_phy_setup; From 888bd0eca93c8c4ef4510d6b21bc43679e256167 Mon Sep 17 00:00:00 2001 From: Shangjuan Wei Date: Wed, 15 Oct 2025 19:40:41 +0800 Subject: [PATCH 0059/1381] dt-bindings: ethernet: eswin: Document for EIC7700 SoC Add ESWIN EIC7700 Ethernet controller, supporting clock configuration, delay adjustment and speed adaptive functions. Signed-off-by: Zhi Li Signed-off-by: Shangjuan Wei Reviewed-by: Krzysztof Kozlowski Link: https://patch.msgid.link/20251015114041.1166-1-weishangjuan@eswincomputing.com Signed-off-by: Jakub Kicinski --- .../bindings/net/eswin,eic7700-eth.yaml | 127 ++++++++++++++++++ 1 file changed, 127 insertions(+) create mode 100644 Documentation/devicetree/bindings/net/eswin,eic7700-eth.yaml diff --git a/Documentation/devicetree/bindings/net/eswin,eic7700-eth.yaml b/Documentation/devicetree/bindings/net/eswin,eic7700-eth.yaml new file mode 100644 index 000000000000..9ddbfe219ae2 --- /dev/null +++ b/Documentation/devicetree/bindings/net/eswin,eic7700-eth.yaml @@ -0,0 +1,127 @@ +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/net/eswin,eic7700-eth.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Eswin EIC7700 SOC Eth Controller + +maintainers: + - Shuang Liang + - Zhi Li + - Shangjuan Wei + +description: + Platform glue layer implementation for STMMAC Ethernet driver. + +select: + properties: + compatible: + contains: + enum: + - eswin,eic7700-qos-eth + required: + - compatible + +allOf: + - $ref: snps,dwmac.yaml# + +properties: + compatible: + items: + - const: eswin,eic7700-qos-eth + - const: snps,dwmac-5.20 + + reg: + maxItems: 1 + + interrupts: + maxItems: 1 + + interrupt-names: + const: macirq + + clocks: + items: + - description: AXI clock + - description: Configuration clock + - description: GMAC main clock + - description: Tx clock + + clock-names: + items: + - const: axi + - const: cfg + - const: stmmaceth + - const: tx + + resets: + maxItems: 1 + + reset-names: + items: + - const: stmmaceth + + rx-internal-delay-ps: + enum: [0, 200, 600, 1200, 1600, 1800, 2000, 2200, 2400] + + tx-internal-delay-ps: + enum: [0, 200, 600, 1200, 1600, 1800, 2000, 2200, 2400] + + eswin,hsp-sp-csr: + $ref: /schemas/types.yaml#/definitions/phandle-array + items: + - description: Phandle to HSP(High-Speed Peripheral) device + - description: Offset of phy control register for internal + or external clock selection + - description: Offset of AXI clock controller Low-Power request + register + - description: Offset of register controlling TX/RX clock delay + description: | + High-Speed Peripheral device needed to configure clock selection, + clock low-power mode and clock delay. + +required: + - compatible + - reg + - clocks + - clock-names + - interrupts + - interrupt-names + - phy-mode + - resets + - reset-names + - rx-internal-delay-ps + - tx-internal-delay-ps + - eswin,hsp-sp-csr + +unevaluatedProperties: false + +examples: + - | + ethernet@50400000 { + compatible = "eswin,eic7700-qos-eth", "snps,dwmac-5.20"; + reg = <0x50400000 0x10000>; + clocks = <&d0_clock 186>, <&d0_clock 171>, <&d0_clock 40>, + <&d0_clock 193>; + clock-names = "axi", "cfg", "stmmaceth", "tx"; + interrupt-parent = <&plic>; + interrupts = <61>; + interrupt-names = "macirq"; + phy-mode = "rgmii-id"; + phy-handle = <&phy0>; + resets = <&reset 95>; + reset-names = "stmmaceth"; + rx-internal-delay-ps = <200>; + tx-internal-delay-ps = <200>; + eswin,hsp-sp-csr = <&hsp_sp_csr 0x100 0x108 0x118>; + snps,axi-config = <&stmmac_axi_setup>; + snps,aal; + snps,fixed-burst; + snps,tso; + stmmac_axi_setup: stmmac-axi-config { + snps,blen = <0 0 0 0 16 8 4>; + snps,rd_osr_lmt = <2>; + snps,wr_osr_lmt = <2>; + }; + }; From ea77dbbdbc4e0f95ad12d4a9179e1ca785f2cd2c Mon Sep 17 00:00:00 2001 From: Shangjuan Wei Date: Wed, 15 Oct 2025 19:41:01 +0800 Subject: [PATCH 0060/1381] net: stmmac: add Eswin EIC7700 glue driver Add Ethernet controller support for Eswin's eic7700 SoC. The driver implements hardware initialization, clock configuration, delay adjustment functions based on DWC Ethernet controller, and supports device tree configuration and platform driver integration. Signed-off-by: Zhi Li Signed-off-by: Shangjuan Wei Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20251015114101.1218-1-weishangjuan@eswincomputing.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/Kconfig | 9 + drivers/net/ethernet/stmicro/stmmac/Makefile | 1 + .../ethernet/stmicro/stmmac/dwmac-eic7700.c | 235 ++++++++++++++++++ 3 files changed, 245 insertions(+) create mode 100644 drivers/net/ethernet/stmicro/stmmac/dwmac-eic7700.c diff --git a/drivers/net/ethernet/stmicro/stmmac/Kconfig b/drivers/net/ethernet/stmicro/stmmac/Kconfig index 9507131875b2..716daa51df7e 100644 --- a/drivers/net/ethernet/stmicro/stmmac/Kconfig +++ b/drivers/net/ethernet/stmicro/stmmac/Kconfig @@ -67,6 +67,15 @@ config DWMAC_ANARION This selects the Anarion SoC glue layer support for the stmmac driver. +config DWMAC_EIC7700 + tristate "Support for Eswin eic7700 ethernet driver" + depends on OF && HAS_DMA && ARCH_ESWIN || COMPILE_TEST + help + This driver supports the Eswin EIC7700 Ethernet controller, + which integrates Synopsys DesignWare QoS features. It enables + high-speed networking with DMA acceleration and is optimized + for embedded systems. + config DWMAC_INGENIC tristate "Ingenic MAC support" default MACH_INGENIC diff --git a/drivers/net/ethernet/stmicro/stmmac/Makefile b/drivers/net/ethernet/stmicro/stmmac/Makefile index 51e068e26ce4..ec56bcf2db62 100644 --- a/drivers/net/ethernet/stmicro/stmmac/Makefile +++ b/drivers/net/ethernet/stmicro/stmmac/Makefile @@ -14,6 +14,7 @@ stmmac-$(CONFIG_STMMAC_SELFTESTS) += stmmac_selftests.o # Ordering matters. Generic driver must be last. obj-$(CONFIG_STMMAC_PLATFORM) += stmmac-platform.o obj-$(CONFIG_DWMAC_ANARION) += dwmac-anarion.o +obj-$(CONFIG_DWMAC_EIC7700) += dwmac-eic7700.o obj-$(CONFIG_DWMAC_INGENIC) += dwmac-ingenic.o obj-$(CONFIG_DWMAC_IPQ806X) += dwmac-ipq806x.o obj-$(CONFIG_DWMAC_LPC18XX) += dwmac-lpc18xx.o diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-eic7700.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-eic7700.c new file mode 100644 index 000000000000..1dcf2037001e --- /dev/null +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-eic7700.c @@ -0,0 +1,235 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Eswin DWC Ethernet linux driver + * + * Copyright 2025, Beijing ESWIN Computing Technology Co., Ltd. + * + * Authors: + * Zhi Li + * Shuang Liang + * Shangjuan Wei + */ + +#include +#include +#include +#include +#include +#include + +#include "stmmac_platform.h" + +/* eth_phy_ctrl_offset eth0:0x100 */ +#define EIC7700_ETH_TX_CLK_SEL BIT(16) +#define EIC7700_ETH_PHY_INTF_SELI BIT(0) + +/* eth_axi_lp_ctrl_offset eth0:0x108 */ +#define EIC7700_ETH_CSYSREQ_VAL BIT(0) + +/* + * TX/RX Clock Delay Bit Masks: + * - TX Delay: bits [14:8] — TX_CLK delay (unit: 0.1ns per bit) + * - RX Delay: bits [30:24] — RX_CLK delay (unit: 0.1ns per bit) + */ +#define EIC7700_ETH_TX_ADJ_DELAY GENMASK(14, 8) +#define EIC7700_ETH_RX_ADJ_DELAY GENMASK(30, 24) + +#define EIC7700_MAX_DELAY_UNIT 0x7F + +static const char * const eic7700_clk_names[] = { + "tx", "axi", "cfg", +}; + +struct eic7700_qos_priv { + struct plat_stmmacenet_data *plat_dat; +}; + +static int eic7700_clks_config(void *priv, bool enabled) +{ + struct eic7700_qos_priv *dwc = (struct eic7700_qos_priv *)priv; + struct plat_stmmacenet_data *plat = dwc->plat_dat; + int ret = 0; + + if (enabled) + ret = clk_bulk_prepare_enable(plat->num_clks, plat->clks); + else + clk_bulk_disable_unprepare(plat->num_clks, plat->clks); + + return ret; +} + +static int eic7700_dwmac_init(struct platform_device *pdev, void *priv) +{ + struct eic7700_qos_priv *dwc = priv; + + return eic7700_clks_config(dwc, true); +} + +static void eic7700_dwmac_exit(struct platform_device *pdev, void *priv) +{ + struct eic7700_qos_priv *dwc = priv; + + eic7700_clks_config(dwc, false); +} + +static int eic7700_dwmac_suspend(struct device *dev, void *priv) +{ + return pm_runtime_force_suspend(dev); +} + +static int eic7700_dwmac_resume(struct device *dev, void *priv) +{ + int ret; + + ret = pm_runtime_force_resume(dev); + if (ret) + dev_err(dev, "%s failed: %d\n", __func__, ret); + + return ret; +} + +static int eic7700_dwmac_probe(struct platform_device *pdev) +{ + struct plat_stmmacenet_data *plat_dat; + struct stmmac_resources stmmac_res; + struct eic7700_qos_priv *dwc_priv; + struct regmap *eic7700_hsp_regmap; + u32 eth_axi_lp_ctrl_offset; + u32 eth_phy_ctrl_offset; + u32 eth_phy_ctrl_regset; + u32 eth_rxd_dly_offset; + u32 eth_dly_param = 0; + u32 delay_ps; + int i, ret; + + ret = stmmac_get_platform_resources(pdev, &stmmac_res); + if (ret) + return dev_err_probe(&pdev->dev, ret, + "failed to get resources\n"); + + plat_dat = devm_stmmac_probe_config_dt(pdev, stmmac_res.mac); + if (IS_ERR(plat_dat)) + return dev_err_probe(&pdev->dev, PTR_ERR(plat_dat), + "dt configuration failed\n"); + + dwc_priv = devm_kzalloc(&pdev->dev, sizeof(*dwc_priv), GFP_KERNEL); + if (!dwc_priv) + return -ENOMEM; + + /* Read rx-internal-delay-ps and update rx_clk delay */ + if (!of_property_read_u32(pdev->dev.of_node, + "rx-internal-delay-ps", &delay_ps)) { + u32 val = min(delay_ps / 100, EIC7700_MAX_DELAY_UNIT); + + eth_dly_param &= ~EIC7700_ETH_RX_ADJ_DELAY; + eth_dly_param |= FIELD_PREP(EIC7700_ETH_RX_ADJ_DELAY, val); + } else { + return dev_err_probe(&pdev->dev, -EINVAL, + "missing required property rx-internal-delay-ps\n"); + } + + /* Read tx-internal-delay-ps and update tx_clk delay */ + if (!of_property_read_u32(pdev->dev.of_node, + "tx-internal-delay-ps", &delay_ps)) { + u32 val = min(delay_ps / 100, EIC7700_MAX_DELAY_UNIT); + + eth_dly_param &= ~EIC7700_ETH_TX_ADJ_DELAY; + eth_dly_param |= FIELD_PREP(EIC7700_ETH_TX_ADJ_DELAY, val); + } else { + return dev_err_probe(&pdev->dev, -EINVAL, + "missing required property tx-internal-delay-ps\n"); + } + + eic7700_hsp_regmap = syscon_regmap_lookup_by_phandle(pdev->dev.of_node, + "eswin,hsp-sp-csr"); + if (IS_ERR(eic7700_hsp_regmap)) + return dev_err_probe(&pdev->dev, + PTR_ERR(eic7700_hsp_regmap), + "Failed to get hsp-sp-csr regmap\n"); + + ret = of_property_read_u32_index(pdev->dev.of_node, + "eswin,hsp-sp-csr", + 1, ð_phy_ctrl_offset); + if (ret) + return dev_err_probe(&pdev->dev, ret, + "can't get eth_phy_ctrl_offset\n"); + + regmap_read(eic7700_hsp_regmap, eth_phy_ctrl_offset, + ð_phy_ctrl_regset); + eth_phy_ctrl_regset |= + (EIC7700_ETH_TX_CLK_SEL | EIC7700_ETH_PHY_INTF_SELI); + regmap_write(eic7700_hsp_regmap, eth_phy_ctrl_offset, + eth_phy_ctrl_regset); + + ret = of_property_read_u32_index(pdev->dev.of_node, + "eswin,hsp-sp-csr", + 2, ð_axi_lp_ctrl_offset); + if (ret) + return dev_err_probe(&pdev->dev, ret, + "can't get eth_axi_lp_ctrl_offset\n"); + + regmap_write(eic7700_hsp_regmap, eth_axi_lp_ctrl_offset, + EIC7700_ETH_CSYSREQ_VAL); + + ret = of_property_read_u32_index(pdev->dev.of_node, + "eswin,hsp-sp-csr", + 3, ð_rxd_dly_offset); + if (ret) + return dev_err_probe(&pdev->dev, ret, + "can't get eth_rxd_dly_offset\n"); + + regmap_write(eic7700_hsp_regmap, eth_rxd_dly_offset, + eth_dly_param); + + plat_dat->num_clks = ARRAY_SIZE(eic7700_clk_names); + plat_dat->clks = devm_kcalloc(&pdev->dev, + plat_dat->num_clks, + sizeof(*plat_dat->clks), + GFP_KERNEL); + if (!plat_dat->clks) + return -ENOMEM; + + for (i = 0; i < ARRAY_SIZE(eic7700_clk_names); i++) + plat_dat->clks[i].id = eic7700_clk_names[i]; + + ret = devm_clk_bulk_get_optional(&pdev->dev, + plat_dat->num_clks, + plat_dat->clks); + if (ret) + return dev_err_probe(&pdev->dev, ret, + "Failed to get clocks\n"); + + plat_dat->clk_tx_i = stmmac_pltfr_find_clk(plat_dat, "tx"); + plat_dat->set_clk_tx_rate = stmmac_set_clk_tx_rate; + plat_dat->clks_config = eic7700_clks_config; + plat_dat->bsp_priv = dwc_priv; + dwc_priv->plat_dat = plat_dat; + plat_dat->init = eic7700_dwmac_init; + plat_dat->exit = eic7700_dwmac_exit; + plat_dat->suspend = eic7700_dwmac_suspend; + plat_dat->resume = eic7700_dwmac_resume; + + return devm_stmmac_pltfr_probe(pdev, plat_dat, &stmmac_res); +} + +static const struct of_device_id eic7700_dwmac_match[] = { + { .compatible = "eswin,eic7700-qos-eth" }, + { } +}; +MODULE_DEVICE_TABLE(of, eic7700_dwmac_match); + +static struct platform_driver eic7700_dwmac_driver = { + .probe = eic7700_dwmac_probe, + .driver = { + .name = "eic7700-eth-dwmac", + .pm = &stmmac_pltfr_pm_ops, + .of_match_table = eic7700_dwmac_match, + }, +}; +module_platform_driver(eic7700_dwmac_driver); + +MODULE_AUTHOR("Zhi Li "); +MODULE_AUTHOR("Shuang Liang "); +MODULE_AUTHOR("Shangjuan Wei "); +MODULE_DESCRIPTION("Eswin eic7700 qos ethernet driver"); +MODULE_LICENSE("GPL"); From 01b6aca22bb9f8fbbebbf8bdbb80aadf11318e3d Mon Sep 17 00:00:00 2001 From: Sagar Cheluvegowda Date: Wed, 15 Oct 2025 18:26:12 +0200 Subject: [PATCH 0061/1381] dt-bindings: net: qcom: ethernet: Add interconnect properties Add documentation for the interconnect and interconnect-names properties required when voting for AHB and AXI buses. Suggested-by: Andrew Halaney Signed-off-by: Sagar Cheluvegowda Reviewed-by: Andrew Lunn Reviewed-by: Krzysztof Kozlowski Signed-off-by: Konrad Dybcio Link: https://patch.msgid.link/20251015-topic-qc_stmmac_icc_bindings-v5-1-da39126cff28@oss.qualcomm.com Signed-off-by: Jakub Kicinski --- Documentation/devicetree/bindings/net/qcom,ethqos.yaml | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/Documentation/devicetree/bindings/net/qcom,ethqos.yaml b/Documentation/devicetree/bindings/net/qcom,ethqos.yaml index e7ee0d9efed8..423959cb928d 100644 --- a/Documentation/devicetree/bindings/net/qcom,ethqos.yaml +++ b/Documentation/devicetree/bindings/net/qcom,ethqos.yaml @@ -73,6 +73,14 @@ properties: dma-coherent: true + interconnects: + maxItems: 2 + + interconnect-names: + items: + - const: cpu-mac + - const: mac-mem + phys: true phy-names: From 56cef47c28dc7d40af8959ceb5a3f49e907e2a89 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 14 Oct 2025 17:19:02 +0000 Subject: [PATCH 0062/1381] selftests/net: packetdrill: unflake tcp_user_timeout_user-timeout-probe.pkt This test fails the first time I am running it after a fresh virtme-ng boot. tcp_user_timeout_user-timeout-probe.pkt:33: runtime error in write call: Expected result -1 but got 24 with errno 2 (No such file or directory) Tweaks the timings a bit, to reduce flakiness. Signed-off-by: Eric Dumazet Cc: Soham Chakradeo Cc: Willem de Bruijn Reviewed-by: Kuniyuki Iwashima Tested-by: Jamal Hadi Salim Acked-by: Jamal Hadi Salim Link: https://patch.msgid.link/20251014171907.3554413-2-edumazet@google.com Signed-off-by: Jakub Kicinski --- .../net/packetdrill/tcp_user_timeout_user-timeout-probe.pkt | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/tools/testing/selftests/net/packetdrill/tcp_user_timeout_user-timeout-probe.pkt b/tools/testing/selftests/net/packetdrill/tcp_user_timeout_user-timeout-probe.pkt index 183051ba0cae..6882b8240a8a 100644 --- a/tools/testing/selftests/net/packetdrill/tcp_user_timeout_user-timeout-probe.pkt +++ b/tools/testing/selftests/net/packetdrill/tcp_user_timeout_user-timeout-probe.pkt @@ -23,14 +23,16 @@ // install a qdisc dropping all packets +0 `tc qdisc delete dev tun0 root 2>/dev/null ; tc qdisc add dev tun0 root pfifo limit 0` + +0 write(4, ..., 24) = 24 // When qdisc is congested we retry every 500ms // (TCP_RESOURCE_PROBE_INTERVAL) and therefore // we retry 6 times before hitting 3s timeout. // First verify that the connection is alive: -+3.250 write(4, ..., 24) = 24 ++3 write(4, ..., 24) = 24 + // Now verify that shortly after that the socket is dead: - +.100 write(4, ..., 24) = -1 ETIMEDOUT (Connection timed out) ++1 write(4, ..., 24) = -1 ETIMEDOUT (Connection timed out) +0 %{ assert tcpi_probes == 6, tcpi_probes; \ assert tcpi_backoff == 0, tcpi_backoff }% From 5b2b7dec05f3ab80d7bb7fa09acd43222fab6aa2 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 14 Oct 2025 17:19:03 +0000 Subject: [PATCH 0063/1381] net: add add indirect call wrapper in skb_release_head_state() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit While stress testing UDP senders on a host with expensive indirect calls, I found cpus processing TX completions where showing a very high cost (20%) in sock_wfree() due to CONFIG_MITIGATION_RETPOLINE=y. Take care of TCP and UDP TX destructors and use INDIRECT_CALL_3() macro. Signed-off-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Reviewed-by: Toke Høiland-Jørgensen Tested-by: Jamal Hadi Salim Acked-by: Jamal Hadi Salim Link: https://patch.msgid.link/20251014171907.3554413-3-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/core/skbuff.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 6be01454f262..8eb3c5820724 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -1136,7 +1136,16 @@ void skb_release_head_state(struct sk_buff *skb) skb_dst_drop(skb); if (skb->destructor) { DEBUG_NET_WARN_ON_ONCE(in_hardirq()); - skb->destructor(skb); +#ifdef CONFIG_INET + INDIRECT_CALL_3(skb->destructor, + tcp_wfree, __sock_wfree, sock_wfree, + skb); +#else + INDIRECT_CALL_1(skb->destructor, + sock_wfree, + skb); + +#endif } #if IS_ENABLED(CONFIG_NF_CONNTRACK) nf_conntrack_put(skb_nfct(skb)); From fe946a751d9b52b7c45ca34899723b314b79b249 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 14 Oct 2025 17:19:04 +0000 Subject: [PATCH 0064/1381] net/sched: act_mirred: add loop detection MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Commit 0f022d32c3ec ("net/sched: Fix mirred deadlock on device recursion") added code in the fast path, even when act_mirred is not used. Prepare its revert by implementing loop detection in act_mirred. Adds an array of device pointers in struct netdev_xmit. tcf_mirred_is_act_redirect() can detect if the array already contains the target device. Signed-off-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Reviewed-by: Toke Høiland-Jørgensen Tested-by: Jamal Hadi Salim Acked-by: Jamal Hadi Salim Link: https://patch.msgid.link/20251014171907.3554413-4-edumazet@google.com Signed-off-by: Jakub Kicinski --- include/linux/netdevice_xmit.h | 9 ++++- net/sched/act_mirred.c | 62 +++++++++++++--------------------- 2 files changed, 31 insertions(+), 40 deletions(-) diff --git a/include/linux/netdevice_xmit.h b/include/linux/netdevice_xmit.h index 813a19122ebb..cc232508e695 100644 --- a/include/linux/netdevice_xmit.h +++ b/include/linux/netdevice_xmit.h @@ -2,6 +2,12 @@ #ifndef _LINUX_NETDEVICE_XMIT_H #define _LINUX_NETDEVICE_XMIT_H +#if IS_ENABLED(CONFIG_NET_ACT_MIRRED) +#define MIRRED_NEST_LIMIT 4 +#endif + +struct net_device; + struct netdev_xmit { u16 recursion; u8 more; @@ -9,7 +15,8 @@ struct netdev_xmit { u8 skip_txqueue; #endif #if IS_ENABLED(CONFIG_NET_ACT_MIRRED) - u8 sched_mirred_nest; + u8 sched_mirred_nest; + struct net_device *sched_mirred_dev[MIRRED_NEST_LIMIT]; #endif #if IS_ENABLED(CONFIG_NF_DUP_NETDEV) u8 nf_dup_skb_recursion; diff --git a/net/sched/act_mirred.c b/net/sched/act_mirred.c index 5f01f567c934..f27b583def78 100644 --- a/net/sched/act_mirred.c +++ b/net/sched/act_mirred.c @@ -29,31 +29,6 @@ static LIST_HEAD(mirred_list); static DEFINE_SPINLOCK(mirred_list_lock); -#define MIRRED_NEST_LIMIT 4 - -#ifndef CONFIG_PREEMPT_RT -static u8 tcf_mirred_nest_level_inc_return(void) -{ - return __this_cpu_inc_return(softnet_data.xmit.sched_mirred_nest); -} - -static void tcf_mirred_nest_level_dec(void) -{ - __this_cpu_dec(softnet_data.xmit.sched_mirred_nest); -} - -#else -static u8 tcf_mirred_nest_level_inc_return(void) -{ - return current->net_xmit.sched_mirred_nest++; -} - -static void tcf_mirred_nest_level_dec(void) -{ - current->net_xmit.sched_mirred_nest--; -} -#endif - static bool tcf_mirred_is_act_redirect(int action) { return action == TCA_EGRESS_REDIR || action == TCA_INGRESS_REDIR; @@ -439,44 +414,53 @@ TC_INDIRECT_SCOPE int tcf_mirred_act(struct sk_buff *skb, { struct tcf_mirred *m = to_mirred(a); int retval = READ_ONCE(m->tcf_action); - unsigned int nest_level; + struct netdev_xmit *xmit; bool m_mac_header_xmit; struct net_device *dev; - int m_eaction; + int i, m_eaction; u32 blockid; - nest_level = tcf_mirred_nest_level_inc_return(); - if (unlikely(nest_level > MIRRED_NEST_LIMIT)) { +#ifdef CONFIG_PREEMPT_RT + xmit = ¤t->net_xmit; +#else + xmit = this_cpu_ptr(&softnet_data.xmit); +#endif + if (unlikely(xmit->sched_mirred_nest >= MIRRED_NEST_LIMIT)) { net_warn_ratelimited("Packet exceeded mirred recursion limit on dev %s\n", netdev_name(skb->dev)); - retval = TC_ACT_SHOT; - goto dec_nest_level; + return TC_ACT_SHOT; } tcf_lastuse_update(&m->tcf_tm); tcf_action_update_bstats(&m->common, skb); blockid = READ_ONCE(m->tcfm_blockid); - if (blockid) { - retval = tcf_blockcast(skb, m, blockid, res, retval); - goto dec_nest_level; - } + if (blockid) + return tcf_blockcast(skb, m, blockid, res, retval); dev = rcu_dereference_bh(m->tcfm_dev); if (unlikely(!dev)) { pr_notice_once("tc mirred: target device is gone\n"); tcf_action_inc_overlimit_qstats(&m->common); - goto dec_nest_level; + return retval; } + for (i = 0; i < xmit->sched_mirred_nest; i++) { + if (xmit->sched_mirred_dev[i] != dev) + continue; + pr_notice_once("tc mirred: loop on device %s\n", + netdev_name(dev)); + tcf_action_inc_overlimit_qstats(&m->common); + return retval; + } + + xmit->sched_mirred_dev[xmit->sched_mirred_nest++] = dev; m_mac_header_xmit = READ_ONCE(m->tcfm_mac_header_xmit); m_eaction = READ_ONCE(m->tcfm_eaction); retval = tcf_mirred_to_dev(skb, m, dev, m_mac_header_xmit, m_eaction, retval); - -dec_nest_level: - tcf_mirred_nest_level_dec(); + xmit->sched_mirred_nest--; return retval; } From 178ca30889a13b555dddab7689fd2cc58c8e5dac Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 14 Oct 2025 17:19:05 +0000 Subject: [PATCH 0065/1381] Revert "net/sched: Fix mirred deadlock on device recursion" MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This reverts commits 0f022d32c3eca477fbf79a205243a6123ed0fe11 and 44180feaccf266d9b0b28cc4ceaac019817deb5c. Prior patch in this series implemented loop detection in act_mirred, we can remove q->owner to save some cycles in the fast path. Signed-off-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Reviewed-by: Toke Høiland-Jørgensen Reviewed-by: Victor Nogueira Tested-by: Jamal Hadi Salim Acked-by: Jamal Hadi Salim Link: https://patch.msgid.link/20251014171907.3554413-5-edumazet@google.com Signed-off-by: Jakub Kicinski --- include/net/sch_generic.h | 1 - net/core/dev.c | 6 ------ net/sched/sch_generic.c | 2 -- 3 files changed, 9 deletions(-) diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index 738cd5b13c62..32e9961570b4 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -117,7 +117,6 @@ struct Qdisc { struct qdisc_skb_head q; struct gnet_stats_basic_sync bstats; struct gnet_stats_queue qstats; - int owner; unsigned long state; unsigned long state2; /* must be written under qdisc spinlock */ struct Qdisc *next_sched; diff --git a/net/core/dev.c b/net/core/dev.c index 35010faf0b78..1d8e7a76d83b 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4167,10 +4167,6 @@ static inline int __dev_xmit_skb(struct sk_buff *skb, struct Qdisc *q, return rc; } - if (unlikely(READ_ONCE(q->owner) == smp_processor_id())) { - kfree_skb_reason(skb, SKB_DROP_REASON_TC_RECLASSIFY_LOOP); - return NET_XMIT_DROP; - } /* * Heuristic to force contended enqueues to serialize on a * separate lock before trying to get qdisc main lock. @@ -4210,9 +4206,7 @@ static inline int __dev_xmit_skb(struct sk_buff *skb, struct Qdisc *q, qdisc_run_end(q); rc = NET_XMIT_SUCCESS; } else { - WRITE_ONCE(q->owner, smp_processor_id()); rc = dev_qdisc_enqueue(skb, q, &to_free, txq); - WRITE_ONCE(q->owner, -1); if (qdisc_run_begin(q)) { if (unlikely(contended)) { spin_unlock(&q->busylock); diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index 1e008a228ebd..dfa8e8e667d2 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -679,7 +679,6 @@ struct Qdisc noop_qdisc = { .qlen = 0, .lock = __SPIN_LOCK_UNLOCKED(noop_qdisc.skb_bad_txq.lock), }, - .owner = -1, }; EXPORT_SYMBOL(noop_qdisc); @@ -985,7 +984,6 @@ struct Qdisc *qdisc_alloc(struct netdev_queue *dev_queue, sch->enqueue = ops->enqueue; sch->dequeue = ops->dequeue; sch->dev_queue = dev_queue; - sch->owner = -1; netdev_hold(dev, &sch->dev_tracker, GFP_KERNEL); refcount_set(&sch->refcnt, 1); From 526f5fb112f7c89c5a9b8b2f9870c8cb76ca4e42 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 14 Oct 2025 17:19:06 +0000 Subject: [PATCH 0066/1381] net: sched: claim one cache line in Qdisc MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Replace state2 field with a boolean. Move it to a hole between qstats and state so that we shrink Qdisc by a full cache line. Signed-off-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Reviewed-by: Toke Høiland-Jørgensen Tested-by: Jamal Hadi Salim Acked-by: Jamal Hadi Salim Link: https://patch.msgid.link/20251014171907.3554413-6-edumazet@google.com Signed-off-by: Jakub Kicinski --- include/net/sch_generic.h | 18 +++++++----------- 1 file changed, 7 insertions(+), 11 deletions(-) diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index 32e9961570b4..31561291bc92 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -41,13 +41,6 @@ enum qdisc_state_t { __QDISC_STATE_DRAINING, }; -enum qdisc_state2_t { - /* Only for !TCQ_F_NOLOCK qdisc. Never access it directly. - * Use qdisc_run_begin/end() or qdisc_is_running() instead. - */ - __QDISC_STATE2_RUNNING, -}; - #define QDISC_STATE_MISSED BIT(__QDISC_STATE_MISSED) #define QDISC_STATE_DRAINING BIT(__QDISC_STATE_DRAINING) @@ -117,8 +110,8 @@ struct Qdisc { struct qdisc_skb_head q; struct gnet_stats_basic_sync bstats; struct gnet_stats_queue qstats; + bool running; /* must be written under qdisc spinlock */ unsigned long state; - unsigned long state2; /* must be written under qdisc spinlock */ struct Qdisc *next_sched; struct sk_buff_head skb_bad_txq; @@ -167,7 +160,7 @@ static inline bool qdisc_is_running(struct Qdisc *qdisc) { if (qdisc->flags & TCQ_F_NOLOCK) return spin_is_locked(&qdisc->seqlock); - return test_bit(__QDISC_STATE2_RUNNING, &qdisc->state2); + return READ_ONCE(qdisc->running); } static inline bool nolock_qdisc_is_empty(const struct Qdisc *qdisc) @@ -210,7 +203,10 @@ static inline bool qdisc_run_begin(struct Qdisc *qdisc) */ return spin_trylock(&qdisc->seqlock); } - return !__test_and_set_bit(__QDISC_STATE2_RUNNING, &qdisc->state2); + if (READ_ONCE(qdisc->running)) + return false; + WRITE_ONCE(qdisc->running, true); + return true; } static inline void qdisc_run_end(struct Qdisc *qdisc) @@ -228,7 +224,7 @@ static inline void qdisc_run_end(struct Qdisc *qdisc) &qdisc->state))) __netif_schedule(qdisc); } else { - __clear_bit(__QDISC_STATE2_RUNNING, &qdisc->state2); + WRITE_ONCE(qdisc->running, false); } } From 100dfa74cad9d4665cdcf0cc8e673b123a3ea910 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 14 Oct 2025 17:19:07 +0000 Subject: [PATCH 0067/1381] net: dev_queue_xmit() llist adoption MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Remove busylock spinlock and use a lockless list (llist) to reduce spinlock contention to the minimum. Idea is that only one cpu might spin on the qdisc spinlock, while others simply add their skb in the llist. After this patch, we get a 300 % improvement on heavy TX workloads. - Sending twice the number of packets per second. - While consuming 50 % less cycles. Note that this also allows in the future to submit batches to various qdisc->enqueue() methods. Tested: - Dual Intel(R) Xeon(R) 6985P-C (480 hyper threads). - 100Gbit NIC, 30 TX queues with FQ packet scheduler. - echo 64 >/sys/kernel/slab/skbuff_small_head/cpu_partial (avoid contention in mm) - 240 concurrent "netperf -t UDP_STREAM -- -m 120 -n" Before: 16 Mpps (41 Mpps if each thread is pinned to a different cpu) vmstat 2 5 procs -----------memory---------- ---swap-- -----io---- -system-- ------cpu----- r b swpd free buff cache si so bi bo in cs us sy id wa st 243 0 0 2368988672 51036 1100852 0 0 146 1 242 60 0 9 91 0 0 244 0 0 2368988672 51036 1100852 0 0 536 10 487745 14718 0 52 48 0 0 244 0 0 2368988672 51036 1100852 0 0 512 0 503067 46033 0 52 48 0 0 244 0 0 2368988672 51036 1100852 0 0 512 0 494807 12107 0 52 48 0 0 244 0 0 2368988672 51036 1100852 0 0 702 26 492845 10110 0 52 48 0 0 Lock contention (1 second sample taken on 8 cores) perf lock record -C0-7 sleep 1; perf lock contention contended total wait max wait avg wait type caller 442111 6.79 s 162.47 ms 15.35 us spinlock dev_hard_start_xmit+0xcd 5961 9.57 ms 8.12 us 1.60 us spinlock __dev_queue_xmit+0x3a0 244 560.63 us 7.63 us 2.30 us spinlock do_softirq+0x5b 13 25.09 us 3.21 us 1.93 us spinlock net_tx_action+0xf8 If netperf threads are pinned, spinlock stress is very high. perf lock record -C0-7 sleep 1; perf lock contention contended total wait max wait avg wait type caller 964508 7.10 s 147.25 ms 7.36 us spinlock dev_hard_start_xmit+0xcd 201 268.05 us 4.65 us 1.33 us spinlock __dev_queue_xmit+0x3a0 12 26.05 us 3.84 us 2.17 us spinlock do_softirq+0x5b @__dev_queue_xmit_ns: [256, 512) 21 | | [512, 1K) 631 | | [1K, 2K) 27328 |@ | [2K, 4K) 265392 |@@@@@@@@@@@@@@@@ | [4K, 8K) 417543 |@@@@@@@@@@@@@@@@@@@@@@@@@@ | [8K, 16K) 826292 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@| [16K, 32K) 733822 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ | [32K, 64K) 19055 |@ | [64K, 128K) 17240 |@ | [128K, 256K) 25633 |@ | [256K, 512K) 4 | | After: 29 Mpps (57 Mpps if each thread is pinned to a different cpu) vmstat 2 5 procs -----------memory---------- ---swap-- -----io---- -system-- ------cpu----- r b swpd free buff cache si so bi bo in cs us sy id wa st 78 0 0 2369573632 32896 1350988 0 0 22 0 331 254 0 8 92 0 0 75 0 0 2369573632 32896 1350988 0 0 22 50 425713 280199 0 23 76 0 0 104 0 0 2369573632 32896 1350988 0 0 290 0 430238 298247 0 23 76 0 0 86 0 0 2369573632 32896 1350988 0 0 132 0 428019 291865 0 24 76 0 0 90 0 0 2369573632 32896 1350988 0 0 502 0 422498 278672 0 23 76 0 0 perf lock record -C0-7 sleep 1; perf lock contention contended total wait max wait avg wait type caller 2524 116.15 ms 486.61 us 46.02 us spinlock __dev_queue_xmit+0x55b 5821 107.18 ms 371.67 us 18.41 us spinlock dev_hard_start_xmit+0xcd 2377 9.73 ms 35.86 us 4.09 us spinlock ___slab_alloc+0x4e0 923 5.74 ms 20.91 us 6.22 us spinlock ___slab_alloc+0x5c9 121 3.42 ms 193.05 us 28.24 us spinlock net_tx_action+0xf8 6 564.33 us 167.60 us 94.05 us spinlock do_softirq+0x5b If netperf threads are pinned (~54 Mpps) perf lock record -C0-7 sleep 1; perf lock contention 32907 316.98 ms 195.98 us 9.63 us spinlock dev_hard_start_xmit+0xcd 4507 61.83 ms 212.73 us 13.72 us spinlock __dev_queue_xmit+0x554 2781 23.53 ms 40.03 us 8.46 us spinlock ___slab_alloc+0x5c9 3554 18.94 ms 34.69 us 5.33 us spinlock ___slab_alloc+0x4e0 233 9.09 ms 215.70 us 38.99 us spinlock do_softirq+0x5b 153 930.66 us 48.67 us 6.08 us spinlock net_tx_action+0xfd 84 331.10 us 14.22 us 3.94 us spinlock ___slab_alloc+0x5c9 140 323.71 us 9.94 us 2.31 us spinlock ___slab_alloc+0x4e0 @__dev_queue_xmit_ns: [128, 256) 1539830 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ | [256, 512) 2299558 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@| [512, 1K) 483936 |@@@@@@@@@@ | [1K, 2K) 265345 |@@@@@@ | [2K, 4K) 145463 |@@@ | [4K, 8K) 54571 |@ | [8K, 16K) 10270 | | [16K, 32K) 9385 | | [32K, 64K) 7749 | | [64K, 128K) 26799 | | [128K, 256K) 2665 | | [256K, 512K) 665 | | Signed-off-by: Eric Dumazet Reviewed-by: Toke Høiland-Jørgensen Reviewed-by: Kuniyuki Iwashima Tested-by: Jamal Hadi Salim Acked-by: Jamal Hadi Salim Link: https://patch.msgid.link/20251014171907.3554413-7-edumazet@google.com Signed-off-by: Jakub Kicinski --- include/net/sch_generic.h | 4 +- net/core/dev.c | 91 ++++++++++++++++++++++++--------------- net/sched/sch_generic.c | 5 --- 3 files changed, 59 insertions(+), 41 deletions(-) diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index 31561291bc92..94966692ccdf 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -115,7 +115,9 @@ struct Qdisc { struct Qdisc *next_sched; struct sk_buff_head skb_bad_txq; - spinlock_t busylock ____cacheline_aligned_in_smp; + atomic_long_t defer_count ____cacheline_aligned_in_smp; + struct llist_head defer_list; + spinlock_t seqlock; struct rcu_head rcu; diff --git a/net/core/dev.c b/net/core/dev.c index 1d8e7a76d83b..821e7c718924 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4125,9 +4125,10 @@ static inline int __dev_xmit_skb(struct sk_buff *skb, struct Qdisc *q, struct net_device *dev, struct netdev_queue *txq) { + struct sk_buff *next, *to_free = NULL; spinlock_t *root_lock = qdisc_lock(q); - struct sk_buff *to_free = NULL; - bool contended; + struct llist_node *ll_list, *first_n; + unsigned long defer_count = 0; int rc; qdisc_calculate_pkt_len(skb, q); @@ -4167,61 +4168,81 @@ static inline int __dev_xmit_skb(struct sk_buff *skb, struct Qdisc *q, return rc; } - /* - * Heuristic to force contended enqueues to serialize on a - * separate lock before trying to get qdisc main lock. - * This permits qdisc->running owner to get the lock more - * often and dequeue packets faster. - * On PREEMPT_RT it is possible to preempt the qdisc owner during xmit - * and then other tasks will only enqueue packets. The packets will be - * sent after the qdisc owner is scheduled again. To prevent this - * scenario the task always serialize on the lock. + /* Open code llist_add(&skb->ll_node, &q->defer_list) + queue limit. + * In the try_cmpxchg() loop, we want to increment q->defer_count + * at most once to limit the number of skbs in defer_list. + * We perform the defer_count increment only if the list is not empty, + * because some arches have slow atomic_long_inc_return(). */ - contended = qdisc_is_running(q) || IS_ENABLED(CONFIG_PREEMPT_RT); - if (unlikely(contended)) - spin_lock(&q->busylock); + first_n = READ_ONCE(q->defer_list.first); + do { + if (first_n && !defer_count) { + defer_count = atomic_long_inc_return(&q->defer_count); + if (unlikely(defer_count > q->limit)) { + kfree_skb_reason(skb, SKB_DROP_REASON_QDISC_DROP); + return NET_XMIT_DROP; + } + } + skb->ll_node.next = first_n; + } while (!try_cmpxchg(&q->defer_list.first, &first_n, &skb->ll_node)); + + /* If defer_list was not empty, we know the cpu which queued + * the first skb will process the whole list for us. + */ + if (first_n) + return NET_XMIT_SUCCESS; spin_lock(root_lock); + + ll_list = llist_del_all(&q->defer_list); + /* There is a small race because we clear defer_count not atomically + * with the prior llist_del_all(). This means defer_list could grow + * over q->limit. + */ + atomic_long_set(&q->defer_count, 0); + + ll_list = llist_reverse_order(ll_list); + if (unlikely(test_bit(__QDISC_STATE_DEACTIVATED, &q->state))) { - __qdisc_drop(skb, &to_free); + llist_for_each_entry_safe(skb, next, ll_list, ll_node) + __qdisc_drop(skb, &to_free); rc = NET_XMIT_DROP; - } else if ((q->flags & TCQ_F_CAN_BYPASS) && !qdisc_qlen(q) && - qdisc_run_begin(q)) { + goto unlock; + } + if ((q->flags & TCQ_F_CAN_BYPASS) && !qdisc_qlen(q) && + !llist_next(ll_list) && qdisc_run_begin(q)) { /* * This is a work-conserving queue; there are no old skbs * waiting to be sent out; and the qdisc is not running - * xmit the skb directly. */ + DEBUG_NET_WARN_ON_ONCE(skb != llist_entry(ll_list, + struct sk_buff, + ll_node)); qdisc_bstats_update(q, skb); - - if (sch_direct_xmit(skb, q, dev, txq, root_lock, true)) { - if (unlikely(contended)) { - spin_unlock(&q->busylock); - contended = false; - } + if (sch_direct_xmit(skb, q, dev, txq, root_lock, true)) __qdisc_run(q); - } - qdisc_run_end(q); rc = NET_XMIT_SUCCESS; } else { - rc = dev_qdisc_enqueue(skb, q, &to_free, txq); - if (qdisc_run_begin(q)) { - if (unlikely(contended)) { - spin_unlock(&q->busylock); - contended = false; - } - __qdisc_run(q); - qdisc_run_end(q); + int count = 0; + + llist_for_each_entry_safe(skb, next, ll_list, ll_node) { + prefetch(next); + skb_mark_not_on_list(skb); + rc = dev_qdisc_enqueue(skb, q, &to_free, txq); + count++; } + qdisc_run(q); + if (count != 1) + rc = NET_XMIT_SUCCESS; } +unlock: spin_unlock(root_lock); if (unlikely(to_free)) kfree_skb_list_reason(to_free, tcf_get_drop_reason(to_free)); - if (unlikely(contended)) - spin_unlock(&q->busylock); return rc; } diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index dfa8e8e667d2..d9a98d02a55f 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -666,7 +666,6 @@ struct Qdisc noop_qdisc = { .ops = &noop_qdisc_ops, .q.lock = __SPIN_LOCK_UNLOCKED(noop_qdisc.q.lock), .dev_queue = &noop_netdev_queue, - .busylock = __SPIN_LOCK_UNLOCKED(noop_qdisc.busylock), .gso_skb = { .next = (struct sk_buff *)&noop_qdisc.gso_skb, .prev = (struct sk_buff *)&noop_qdisc.gso_skb, @@ -970,10 +969,6 @@ struct Qdisc *qdisc_alloc(struct netdev_queue *dev_queue, } } - spin_lock_init(&sch->busylock); - lockdep_set_class(&sch->busylock, - dev->qdisc_tx_busylock ?: &qdisc_tx_busylock); - /* seqlock has the same scope of busylock, for NOLOCK qdisc */ spin_lock_init(&sch->seqlock); lockdep_set_class(&sch->seqlock, From f1150b779571da48031d1e643aeaa9d416cfeb60 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Th=C3=A9o=20Lebrun?= Date: Tue, 14 Oct 2025 17:25:02 +0200 Subject: [PATCH 0068/1381] dt-bindings: net: cdns,macb: sort compatibles MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Compatibles inside this enum are sorted-ish. Make it sorted. Reviewed-by: Krzysztof Kozlowski Signed-off-by: Théo Lebrun Link: https://patch.msgid.link/20251014-macb-cleanup-v1-1-31cd266e22cd@bootlin.com Signed-off-by: Jakub Kicinski --- Documentation/devicetree/bindings/net/cdns,macb.yaml | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/Documentation/devicetree/bindings/net/cdns,macb.yaml b/Documentation/devicetree/bindings/net/cdns,macb.yaml index 1029786a855c..02f14a0b72f9 100644 --- a/Documentation/devicetree/bindings/net/cdns,macb.yaml +++ b/Documentation/devicetree/bindings/net/cdns,macb.yaml @@ -47,18 +47,18 @@ properties: - const: cdns,macb # Generic - enum: - - atmel,sama5d29-gem # GEM XL IP (10/100) on Atmel sama5d29 SoCs - atmel,sama5d2-gem # GEM IP (10/100) on Atmel sama5d2 SoCs + - atmel,sama5d29-gem # GEM XL IP (10/100) on Atmel sama5d29 SoCs - atmel,sama5d3-gem # Gigabit IP on Atmel sama5d3 SoCs - atmel,sama5d4-gem # GEM IP (10/100) on Atmel sama5d4 SoCs + - cdns,emac # Generic + - cdns,gem # Generic + - cdns,macb # Generic - cdns,np4-macb # NP4 SoC devices - microchip,sama7g5-emac # Microchip SAMA7G5 ethernet interface - microchip,sama7g5-gem # Microchip SAMA7G5 gigabit ethernet interface - raspberrypi,rp1-gem # Raspberry Pi RP1 gigabit ethernet interface - sifive,fu540-c000-gem # SiFive FU540-C000 SoC - - cdns,emac # Generic - - cdns,gem # Generic - - cdns,macb # Generic - items: - enum: From a23b0b79e974c921e0d642a8a5f37e76810eac25 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Th=C3=A9o=20Lebrun?= Date: Tue, 14 Oct 2025 17:25:03 +0200 Subject: [PATCH 0069/1381] net: macb: use BIT() macro for capability definitions MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Replace all capabilities values by calls to the BIT() macro. Reviewed-by: Andrew Lunn Reviewed-by: Sean Anderson Signed-off-by: Théo Lebrun Link: https://patch.msgid.link/20251014-macb-cleanup-v1-2-31cd266e22cd@bootlin.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/cadence/macb.h | 42 ++++++++++++++--------------- 1 file changed, 21 insertions(+), 21 deletions(-) diff --git a/drivers/net/ethernet/cadence/macb.h b/drivers/net/ethernet/cadence/macb.h index 0830c48973aa..869d02284707 100644 --- a/drivers/net/ethernet/cadence/macb.h +++ b/drivers/net/ethernet/cadence/macb.h @@ -756,27 +756,27 @@ #define MACB_MAN_C45_CODE 2 /* Capability mask bits */ -#define MACB_CAPS_ISR_CLEAR_ON_WRITE 0x00000001 -#define MACB_CAPS_USRIO_HAS_CLKEN 0x00000002 -#define MACB_CAPS_USRIO_DEFAULT_IS_MII_GMII 0x00000004 -#define MACB_CAPS_NO_GIGABIT_HALF 0x00000008 -#define MACB_CAPS_USRIO_DISABLED 0x00000010 -#define MACB_CAPS_JUMBO 0x00000020 -#define MACB_CAPS_GEM_HAS_PTP 0x00000040 -#define MACB_CAPS_BD_RD_PREFETCH 0x00000080 -#define MACB_CAPS_NEEDS_RSTONUBR 0x00000100 -#define MACB_CAPS_MIIONRGMII 0x00000200 -#define MACB_CAPS_NEED_TSUCLK 0x00000400 -#define MACB_CAPS_QUEUE_DISABLE 0x00000800 -#define MACB_CAPS_QBV 0x00001000 -#define MACB_CAPS_PCS 0x01000000 -#define MACB_CAPS_HIGH_SPEED 0x02000000 -#define MACB_CAPS_CLK_HW_CHG 0x04000000 -#define MACB_CAPS_MACB_IS_EMAC 0x08000000 -#define MACB_CAPS_FIFO_MODE 0x10000000 -#define MACB_CAPS_GIGABIT_MODE_AVAILABLE 0x20000000 -#define MACB_CAPS_SG_DISABLED 0x40000000 -#define MACB_CAPS_MACB_IS_GEM 0x80000000 +#define MACB_CAPS_ISR_CLEAR_ON_WRITE BIT(0) +#define MACB_CAPS_USRIO_HAS_CLKEN BIT(1) +#define MACB_CAPS_USRIO_DEFAULT_IS_MII_GMII BIT(2) +#define MACB_CAPS_NO_GIGABIT_HALF BIT(3) +#define MACB_CAPS_USRIO_DISABLED BIT(4) +#define MACB_CAPS_JUMBO BIT(5) +#define MACB_CAPS_GEM_HAS_PTP BIT(6) +#define MACB_CAPS_BD_RD_PREFETCH BIT(7) +#define MACB_CAPS_NEEDS_RSTONUBR BIT(8) +#define MACB_CAPS_MIIONRGMII BIT(9) +#define MACB_CAPS_NEED_TSUCLK BIT(10) +#define MACB_CAPS_QUEUE_DISABLE BIT(11) +#define MACB_CAPS_QBV BIT(12) +#define MACB_CAPS_PCS BIT(24) +#define MACB_CAPS_HIGH_SPEED BIT(25) +#define MACB_CAPS_CLK_HW_CHG BIT(26) +#define MACB_CAPS_MACB_IS_EMAC BIT(27) +#define MACB_CAPS_FIFO_MODE BIT(28) +#define MACB_CAPS_GIGABIT_MODE_AVAILABLE BIT(29) +#define MACB_CAPS_SG_DISABLED BIT(30) +#define MACB_CAPS_MACB_IS_GEM BIT(31) /* LSO settings */ #define MACB_LSO_UFO_ENABLE 0x01 From bd0b35ec835a177539922834fd662be6280a65fd Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Th=C3=A9o=20Lebrun?= Date: Tue, 14 Oct 2025 17:25:04 +0200 Subject: [PATCH 0070/1381] net: macb: remove gap in MACB_CAPS_* flags MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit MACB_CAPS_* are bit constants that get used in bp->caps. They occupy bits 0..12 + 24..31. Remove 11..23 gap by moving bits 24..31 to 13..20. Occupation bitfields: 31 29 27 25 23 21 19 17 15 13 11 09 07 05 03 01 30 28 26 24 22 20 18 16 14 12 10 08 06 04 02 00 -- Before ------------------------------------------------------ 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 -- After ------------------------------------------------------- 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 Signed-off-by: Théo Lebrun Link: https://patch.msgid.link/20251014-macb-cleanup-v1-3-31cd266e22cd@bootlin.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/cadence/macb.h | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/drivers/net/ethernet/cadence/macb.h b/drivers/net/ethernet/cadence/macb.h index 869d02284707..9d21ec482c8c 100644 --- a/drivers/net/ethernet/cadence/macb.h +++ b/drivers/net/ethernet/cadence/macb.h @@ -769,14 +769,14 @@ #define MACB_CAPS_NEED_TSUCLK BIT(10) #define MACB_CAPS_QUEUE_DISABLE BIT(11) #define MACB_CAPS_QBV BIT(12) -#define MACB_CAPS_PCS BIT(24) -#define MACB_CAPS_HIGH_SPEED BIT(25) -#define MACB_CAPS_CLK_HW_CHG BIT(26) -#define MACB_CAPS_MACB_IS_EMAC BIT(27) -#define MACB_CAPS_FIFO_MODE BIT(28) -#define MACB_CAPS_GIGABIT_MODE_AVAILABLE BIT(29) -#define MACB_CAPS_SG_DISABLED BIT(30) -#define MACB_CAPS_MACB_IS_GEM BIT(31) +#define MACB_CAPS_PCS BIT(13) +#define MACB_CAPS_HIGH_SPEED BIT(14) +#define MACB_CAPS_CLK_HW_CHG BIT(15) +#define MACB_CAPS_MACB_IS_EMAC BIT(16) +#define MACB_CAPS_FIFO_MODE BIT(17) +#define MACB_CAPS_GIGABIT_MODE_AVAILABLE BIT(18) +#define MACB_CAPS_SG_DISABLED BIT(19) +#define MACB_CAPS_MACB_IS_GEM BIT(20) /* LSO settings */ #define MACB_LSO_UFO_ENABLE 0x01 From 80cf78c59a1a9640acf56e22ad18ca17521dea51 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Th=C3=A9o=20Lebrun?= Date: Tue, 14 Oct 2025 17:25:05 +0200 Subject: [PATCH 0071/1381] net: macb: Remove local variables clk_init and init in macb_probe() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Remove local variables clk_init and init. Those function pointers are always equivalent to macb_config->clk_init and macb_config->init. Reviewed-by: Sean Anderson Signed-off-by: Théo Lebrun Link: https://patch.msgid.link/20251014-macb-cleanup-v1-4-31cd266e22cd@bootlin.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/cadence/macb_main.c | 13 +++---------- 1 file changed, 3 insertions(+), 10 deletions(-) diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c index ca2386b83473..dad1188ef9d8 100644 --- a/drivers/net/ethernet/cadence/macb_main.c +++ b/drivers/net/ethernet/cadence/macb_main.c @@ -5424,10 +5424,6 @@ static const struct macb_config default_gem_config = { static int macb_probe(struct platform_device *pdev) { const struct macb_config *macb_config = &default_gem_config; - int (*clk_init)(struct platform_device *, struct clk **, - struct clk **, struct clk **, struct clk **, - struct clk **) = macb_config->clk_init; - int (*init)(struct platform_device *) = macb_config->init; struct device_node *np = pdev->dev.of_node; struct clk *pclk, *hclk = NULL, *tx_clk = NULL, *rx_clk = NULL; struct clk *tsu_clk = NULL; @@ -5449,14 +5445,11 @@ static int macb_probe(struct platform_device *pdev) const struct of_device_id *match; match = of_match_node(macb_dt_ids, np); - if (match && match->data) { + if (match && match->data) macb_config = match->data; - clk_init = macb_config->clk_init; - init = macb_config->init; - } } - err = clk_init(pdev, &pclk, &hclk, &tx_clk, &rx_clk, &tsu_clk); + err = macb_config->clk_init(pdev, &pclk, &hclk, &tx_clk, &rx_clk, &tsu_clk); if (err) return err; @@ -5594,7 +5587,7 @@ static int macb_probe(struct platform_device *pdev) bp->phy_interface = interface; /* IP specific init */ - err = init(pdev); + err = macb_config->init(pdev); if (err) goto err_out_free_netdev; From d7a4a20abe25bb5a15358c8d20a0ab12b7a52f79 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Th=C3=A9o=20Lebrun?= Date: Tue, 14 Oct 2025 17:25:06 +0200 Subject: [PATCH 0072/1381] net: macb: drop macb_config NULL checking MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Remove NULL checks on macb_config as it is always valid: - either it is its default value &default_gem_config, - or it got overridden using match data. Reviewed-by: Sean Anderson Signed-off-by: Théo Lebrun Link: https://patch.msgid.link/20251014-macb-cleanup-v1-5-31cd266e22cd@bootlin.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/cadence/macb_main.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c index dad1188ef9d8..33e99aab1dcb 100644 --- a/drivers/net/ethernet/cadence/macb_main.c +++ b/drivers/net/ethernet/cadence/macb_main.c @@ -5485,15 +5485,13 @@ static int macb_probe(struct platform_device *pdev) } bp->num_queues = num_queues; bp->queue_mask = queue_mask; - if (macb_config) - bp->dma_burst_length = macb_config->dma_burst_length; + bp->dma_burst_length = macb_config->dma_burst_length; bp->pclk = pclk; bp->hclk = hclk; bp->tx_clk = tx_clk; bp->rx_clk = rx_clk; bp->tsu_clk = tsu_clk; - if (macb_config) - bp->jumbo_max_len = macb_config->jumbo_max_len; + bp->jumbo_max_len = macb_config->jumbo_max_len; if (!hw_is_gem(bp->regs, bp->native_io)) bp->max_tx_length = MACB_MAX_TX_LEN; From 94a164598d833aaee98a81b2910ac7bfea09e60f Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Th=C3=A9o=20Lebrun?= Date: Tue, 14 Oct 2025 17:25:07 +0200 Subject: [PATCH 0073/1381] net: macb: simplify macb_dma_desc_get_size() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit macb_dma_desc_get_size() does a switch on bp->hw_dma_cap and covers all four cases: 0, 64B, PTP, 64B+PTP. It also covers the #ifndef MACB_EXT_DESC separately, making it four codepaths. Instead, notice the descriptor size grows with enabled features and use plain if-statements on 64B and PTP flags. Signed-off-by: Théo Lebrun Link: https://patch.msgid.link/20251014-macb-cleanup-v1-6-31cd266e22cd@bootlin.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/cadence/macb_main.c | 29 +++++++----------------- 1 file changed, 8 insertions(+), 21 deletions(-) diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c index 33e99aab1dcb..7f74e280a335 100644 --- a/drivers/net/ethernet/cadence/macb_main.c +++ b/drivers/net/ethernet/cadence/macb_main.c @@ -121,29 +121,16 @@ struct sifive_fu540_macb_mgmt { */ static unsigned int macb_dma_desc_get_size(struct macb *bp) { -#ifdef MACB_EXT_DESC - unsigned int desc_size; + unsigned int desc_size = sizeof(struct macb_dma_desc); - switch (bp->hw_dma_cap) { - case HW_DMA_CAP_64B: - desc_size = sizeof(struct macb_dma_desc) - + sizeof(struct macb_dma_desc_64); - break; - case HW_DMA_CAP_PTP: - desc_size = sizeof(struct macb_dma_desc) - + sizeof(struct macb_dma_desc_ptp); - break; - case HW_DMA_CAP_64B_PTP: - desc_size = sizeof(struct macb_dma_desc) - + sizeof(struct macb_dma_desc_64) - + sizeof(struct macb_dma_desc_ptp); - break; - default: - desc_size = sizeof(struct macb_dma_desc); - } - return desc_size; +#ifdef MACB_EXT_DESC + if (bp->hw_dma_cap & HW_DMA_CAP_64B) + desc_size += sizeof(struct macb_dma_desc_64); + if (bp->hw_dma_cap & HW_DMA_CAP_PTP) + desc_size += sizeof(struct macb_dma_desc_ptp); #endif - return sizeof(struct macb_dma_desc); + + return desc_size; } static unsigned int macb_adj_dma_desc_idx(struct macb *bp, unsigned int desc_idx) From 62e6c17463a7b361457193b4bf4e87de78534130 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Th=C3=A9o=20Lebrun?= Date: Tue, 14 Oct 2025 17:25:08 +0200 Subject: [PATCH 0074/1381] net: macb: simplify macb_adj_dma_desc_idx() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The function body uses a switch statement on bp->hw_dma_cap and handles its four possible values: 0, is_64b, is_ptp, is_64b && is_ptp. Instead, refactor by noticing that the return value is: desc_size * MULT with MULT = 3 if is_64b && is_ptp, 2 if is_64b || is_ptp, 1 otherwise. MULT can be expressed as: 1 + is_64b + is_ptp Signed-off-by: Théo Lebrun Link: https://patch.msgid.link/20251014-macb-cleanup-v1-7-31cd266e22cd@bootlin.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/cadence/macb_main.c | 18 ++++++------------ 1 file changed, 6 insertions(+), 12 deletions(-) diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c index 7f74e280a335..44a411662786 100644 --- a/drivers/net/ethernet/cadence/macb_main.c +++ b/drivers/net/ethernet/cadence/macb_main.c @@ -136,19 +136,13 @@ static unsigned int macb_dma_desc_get_size(struct macb *bp) static unsigned int macb_adj_dma_desc_idx(struct macb *bp, unsigned int desc_idx) { #ifdef MACB_EXT_DESC - switch (bp->hw_dma_cap) { - case HW_DMA_CAP_64B: - case HW_DMA_CAP_PTP: - desc_idx <<= 1; - break; - case HW_DMA_CAP_64B_PTP: - desc_idx *= 3; - break; - default: - break; - } -#endif + bool is_ptp = bp->hw_dma_cap & HW_DMA_CAP_PTP; + bool is_64b = bp->hw_dma_cap & HW_DMA_CAP_64B; + + return desc_idx * (1 + is_64b + is_ptp); +#else return desc_idx; +#endif } #ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT From 731e991afb75c3b49da1e781e68a7c710e1d9e92 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Th=C3=A9o=20Lebrun?= Date: Tue, 14 Oct 2025 17:25:09 +0200 Subject: [PATCH 0075/1381] net: macb: move bp->hw_dma_cap flags to bp->caps MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Drop bp->hw_dma_cap field and put its two flags into bp->caps. On my specific config (eyeq5_defconfig), bloat-o-meter indicates: - macb_main.o: Before=56251, After=56359, chg +0.19% - macb_ptp.o: Before= 3976, After= 3952, chg -0.60% Signed-off-by: Théo Lebrun Link: https://patch.msgid.link/20251014-macb-cleanup-v1-8-31cd266e22cd@bootlin.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/cadence/macb.h | 10 ++------- drivers/net/ethernet/cadence/macb_main.c | 28 ++++++++++++------------ drivers/net/ethernet/cadence/macb_ptp.c | 16 ++++++++------ 3 files changed, 25 insertions(+), 29 deletions(-) diff --git a/drivers/net/ethernet/cadence/macb.h b/drivers/net/ethernet/cadence/macb.h index 9d21ec482c8c..b3a4c9534240 100644 --- a/drivers/net/ethernet/cadence/macb.h +++ b/drivers/net/ethernet/cadence/macb.h @@ -777,6 +777,8 @@ #define MACB_CAPS_GIGABIT_MODE_AVAILABLE BIT(18) #define MACB_CAPS_SG_DISABLED BIT(19) #define MACB_CAPS_MACB_IS_GEM BIT(20) +#define MACB_CAPS_DMA_64B BIT(21) +#define MACB_CAPS_DMA_PTP BIT(22) /* LSO settings */ #define MACB_LSO_UFO_ENABLE 0x01 @@ -854,11 +856,6 @@ struct macb_dma_desc { }; #ifdef MACB_EXT_DESC -#define HW_DMA_CAP_32B 0 -#define HW_DMA_CAP_64B (1 << 0) -#define HW_DMA_CAP_PTP (1 << 1) -#define HW_DMA_CAP_64B_PTP (HW_DMA_CAP_64B | HW_DMA_CAP_PTP) - struct macb_dma_desc_64 { u32 addrh; u32 resvd; @@ -1349,9 +1346,6 @@ struct macb { struct phy *sgmii_phy; /* for ZynqMP SGMII mode */ -#ifdef MACB_EXT_DESC - uint8_t hw_dma_cap; -#endif spinlock_t tsu_clk_lock; /* gem tsu clock locking */ unsigned int tsu_rate; struct ptp_clock *ptp_clock; diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c index 44a411662786..44b96bf53ff6 100644 --- a/drivers/net/ethernet/cadence/macb_main.c +++ b/drivers/net/ethernet/cadence/macb_main.c @@ -124,9 +124,9 @@ static unsigned int macb_dma_desc_get_size(struct macb *bp) unsigned int desc_size = sizeof(struct macb_dma_desc); #ifdef MACB_EXT_DESC - if (bp->hw_dma_cap & HW_DMA_CAP_64B) + if (bp->caps & MACB_CAPS_DMA_64B) desc_size += sizeof(struct macb_dma_desc_64); - if (bp->hw_dma_cap & HW_DMA_CAP_PTP) + if (bp->caps & MACB_CAPS_DMA_PTP) desc_size += sizeof(struct macb_dma_desc_ptp); #endif @@ -136,8 +136,8 @@ static unsigned int macb_dma_desc_get_size(struct macb *bp) static unsigned int macb_adj_dma_desc_idx(struct macb *bp, unsigned int desc_idx) { #ifdef MACB_EXT_DESC - bool is_ptp = bp->hw_dma_cap & HW_DMA_CAP_PTP; - bool is_64b = bp->hw_dma_cap & HW_DMA_CAP_64B; + bool is_ptp = bp->caps & MACB_CAPS_DMA_PTP; + bool is_64b = bp->caps & MACB_CAPS_DMA_64B; return desc_idx * (1 + is_64b + is_ptp); #else @@ -475,7 +475,7 @@ static void macb_init_buffers(struct macb *bp) #ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT /* Single register for all queues' high 32 bits. */ - if (bp->hw_dma_cap & HW_DMA_CAP_64B) { + if (bp->caps & MACB_CAPS_DMA_64B) { macb_writel(bp, RBQPH, upper_32_bits(bp->queues[0].rx_ring_dma)); macb_writel(bp, TBQPH, @@ -1009,7 +1009,7 @@ static void macb_set_addr(struct macb *bp, struct macb_dma_desc *desc, dma_addr_ #ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT struct macb_dma_desc_64 *desc_64; - if (bp->hw_dma_cap & HW_DMA_CAP_64B) { + if (bp->caps & MACB_CAPS_DMA_64B) { desc_64 = macb_64b_desc(bp, desc); desc_64->addrh = upper_32_bits(addr); /* The low bits of RX address contain the RX_USED bit, clearing @@ -1028,14 +1028,14 @@ static dma_addr_t macb_get_addr(struct macb *bp, struct macb_dma_desc *desc) #ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT struct macb_dma_desc_64 *desc_64; - if (bp->hw_dma_cap & HW_DMA_CAP_64B) { + if (bp->caps & MACB_CAPS_DMA_64B) { desc_64 = macb_64b_desc(bp, desc); addr = ((u64)(desc_64->addrh) << 32); } #endif addr |= MACB_BF(RX_WADDR, MACB_BFEXT(RX_WADDR, desc->addr)); #ifdef CONFIG_MACB_USE_HWSTAMP - if (bp->hw_dma_cap & HW_DMA_CAP_PTP) + if (bp->caps & MACB_CAPS_DMA_PTP) addr &= ~GEM_BIT(DMA_RXVALID); #endif return addr; @@ -2301,7 +2301,7 @@ static netdev_tx_t macb_start_xmit(struct sk_buff *skb, struct net_device *dev) #ifdef CONFIG_MACB_USE_HWSTAMP if ((skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) && - (bp->hw_dma_cap & HW_DMA_CAP_PTP)) + (bp->caps & MACB_CAPS_DMA_PTP)) skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS; #endif @@ -2781,11 +2781,11 @@ static void macb_configure_dma(struct macb *bp) dmacfg &= ~GEM_BIT(ADDR64); #ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT - if (bp->hw_dma_cap & HW_DMA_CAP_64B) + if (bp->caps & MACB_CAPS_DMA_64B) dmacfg |= GEM_BIT(ADDR64); #endif #ifdef CONFIG_MACB_USE_HWSTAMP - if (bp->hw_dma_cap & HW_DMA_CAP_PTP) + if (bp->caps & MACB_CAPS_DMA_PTP) dmacfg |= GEM_BIT(RXEXT) | GEM_BIT(TXEXT); #endif netdev_dbg(bp->dev, "Cadence configure DMA with 0x%08x\n", @@ -3563,7 +3563,7 @@ static int gem_get_ts_info(struct net_device *dev, { struct macb *bp = netdev_priv(dev); - if ((bp->hw_dma_cap & HW_DMA_CAP_PTP) == 0) { + if (!(bp->caps & MACB_CAPS_DMA_PTP)) { ethtool_op_get_ts_info(dev, info); return 0; } @@ -4351,7 +4351,7 @@ static void macb_configure_caps(struct macb *bp, "GEM doesn't support hardware ptp.\n"); else { #ifdef CONFIG_MACB_USE_HWSTAMP - bp->hw_dma_cap |= HW_DMA_CAP_PTP; + bp->caps |= MACB_CAPS_DMA_PTP; bp->ptp_info = &gem_ptp_info; #endif } @@ -5518,7 +5518,7 @@ static int macb_probe(struct platform_device *pdev) dev_err(&pdev->dev, "failed to set DMA mask\n"); goto err_out_free_netdev; } - bp->hw_dma_cap |= HW_DMA_CAP_64B; + bp->caps |= MACB_CAPS_DMA_64B; } #endif platform_set_drvdata(pdev, dev); diff --git a/drivers/net/ethernet/cadence/macb_ptp.c b/drivers/net/ethernet/cadence/macb_ptp.c index a63bf29c4fa8..f4ab379f2849 100644 --- a/drivers/net/ethernet/cadence/macb_ptp.c +++ b/drivers/net/ethernet/cadence/macb_ptp.c @@ -28,14 +28,16 @@ static struct macb_dma_desc_ptp *macb_ptp_desc(struct macb *bp, struct macb_dma_desc *desc) { - if (bp->hw_dma_cap == HW_DMA_CAP_PTP) - return (struct macb_dma_desc_ptp *) - ((u8 *)desc + sizeof(struct macb_dma_desc)); - if (bp->hw_dma_cap == HW_DMA_CAP_64B_PTP) + if (!(bp->caps & MACB_CAPS_DMA_PTP)) + return NULL; + + if (bp->caps & MACB_CAPS_DMA_64B) return (struct macb_dma_desc_ptp *) ((u8 *)desc + sizeof(struct macb_dma_desc) + sizeof(struct macb_dma_desc_64)); - return NULL; + else + return (struct macb_dma_desc_ptp *) + ((u8 *)desc + sizeof(struct macb_dma_desc)); } static int gem_tsu_get_time(struct ptp_clock_info *ptp, struct timespec64 *ts, @@ -380,7 +382,7 @@ int gem_get_hwtst(struct net_device *dev, struct macb *bp = netdev_priv(dev); *tstamp_config = bp->tstamp_config; - if ((bp->hw_dma_cap & HW_DMA_CAP_PTP) == 0) + if (!(bp->caps & MACB_CAPS_DMA_PTP)) return -EOPNOTSUPP; return 0; @@ -407,7 +409,7 @@ int gem_set_hwtst(struct net_device *dev, struct macb *bp = netdev_priv(dev); u32 regval; - if ((bp->hw_dma_cap & HW_DMA_CAP_PTP) == 0) + if (!(bp->caps & MACB_CAPS_DMA_PTP)) return -EOPNOTSUPP; switch (tstamp_config->tx_type) { From 02d11c610555657f8524a56bab6856fa8d6ace71 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Th=C3=A9o=20Lebrun?= Date: Tue, 14 Oct 2025 17:25:10 +0200 Subject: [PATCH 0076/1381] net: macb: introduce DMA descriptor helpers (is 64bit? is PTP?) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Introduce macb_dma64() and macb_dma_ptp() helper functions. Many codepaths are made simpler by dropping conditional compilation. This implies two additional changes: - Always compile related structure definitions inside . - MACB_EXT_DESC can be dropped as it is useless now. The common case: #ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT struct macb_dma_desc_64 *desc_64; if (bp->hw_dma_cap & HW_DMA_CAP_64B) { desc_64 = macb_64b_desc(bp, desc); // ... } #endif Is replaced by: if (macb_dma64(bp)) { struct macb_dma_desc_64 *desc_64 = macb_64b_desc(bp, desc); // ... } Signed-off-by: Théo Lebrun Link: https://patch.msgid.link/20251014-macb-cleanup-v1-9-31cd266e22cd@bootlin.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/cadence/macb.h | 18 +++++--- drivers/net/ethernet/cadence/macb_main.c | 55 +++++++----------------- drivers/net/ethernet/cadence/macb_ptp.c | 8 ++-- 3 files changed, 32 insertions(+), 49 deletions(-) diff --git a/drivers/net/ethernet/cadence/macb.h b/drivers/net/ethernet/cadence/macb.h index b3a4c9534240..f696b2ddc412 100644 --- a/drivers/net/ethernet/cadence/macb.h +++ b/drivers/net/ethernet/cadence/macb.h @@ -15,10 +15,6 @@ #include #include -#if defined(CONFIG_ARCH_DMA_ADDR_T_64BIT) || defined(CONFIG_MACB_USE_HWSTAMP) -#define MACB_EXT_DESC -#endif - #define MACB_GREGS_NBR 16 #define MACB_GREGS_VERSION 2 #define MACB_MAX_QUEUES 8 @@ -855,7 +851,6 @@ struct macb_dma_desc { u32 ctrl; }; -#ifdef MACB_EXT_DESC struct macb_dma_desc_64 { u32 addrh; u32 resvd; @@ -865,7 +860,6 @@ struct macb_dma_desc_ptp { u32 ts_1; u32 ts_2; }; -#endif /* DMA descriptor bitfields */ #define MACB_RX_USED_OFFSET 0 @@ -1437,6 +1431,18 @@ static inline u64 enst_max_hw_interval(u32 speed_mbps) ENST_TIME_GRANULARITY_NS * 1000, (speed_mbps)); } +static inline bool macb_dma64(struct macb *bp) +{ + return IS_ENABLED(CONFIG_ARCH_DMA_ADDR_T_64BIT) && + bp->caps & MACB_CAPS_DMA_64B; +} + +static inline bool macb_dma_ptp(struct macb *bp) +{ + return IS_ENABLED(CONFIG_MACB_USE_HWSTAMP) && + bp->caps & MACB_CAPS_DMA_PTP; +} + /** * struct macb_platform_data - platform data for MACB Ethernet used for PCI registration * @pclk: platform clock diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c index 44b96bf53ff6..9db419b94d0b 100644 --- a/drivers/net/ethernet/cadence/macb_main.c +++ b/drivers/net/ethernet/cadence/macb_main.c @@ -123,35 +123,24 @@ static unsigned int macb_dma_desc_get_size(struct macb *bp) { unsigned int desc_size = sizeof(struct macb_dma_desc); -#ifdef MACB_EXT_DESC - if (bp->caps & MACB_CAPS_DMA_64B) + if (macb_dma64(bp)) desc_size += sizeof(struct macb_dma_desc_64); - if (bp->caps & MACB_CAPS_DMA_PTP) + if (macb_dma_ptp(bp)) desc_size += sizeof(struct macb_dma_desc_ptp); -#endif return desc_size; } static unsigned int macb_adj_dma_desc_idx(struct macb *bp, unsigned int desc_idx) { -#ifdef MACB_EXT_DESC - bool is_ptp = bp->caps & MACB_CAPS_DMA_PTP; - bool is_64b = bp->caps & MACB_CAPS_DMA_64B; - - return desc_idx * (1 + is_64b + is_ptp); -#else - return desc_idx; -#endif + return desc_idx * (1 + macb_dma64(bp) + macb_dma_ptp(bp)); } -#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT static struct macb_dma_desc_64 *macb_64b_desc(struct macb *bp, struct macb_dma_desc *desc) { return (struct macb_dma_desc_64 *)((void *)desc + sizeof(struct macb_dma_desc)); } -#endif /* Ring buffer accessors */ static unsigned int macb_tx_ring_wrap(struct macb *bp, unsigned int index) @@ -473,15 +462,13 @@ static void macb_init_buffers(struct macb *bp) struct macb_queue *queue; unsigned int q; -#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT /* Single register for all queues' high 32 bits. */ - if (bp->caps & MACB_CAPS_DMA_64B) { + if (macb_dma64(bp)) { macb_writel(bp, RBQPH, upper_32_bits(bp->queues[0].rx_ring_dma)); macb_writel(bp, TBQPH, upper_32_bits(bp->queues[0].tx_ring_dma)); } -#endif for (q = 0, queue = bp->queues; q < bp->num_queues; ++q, ++queue) { queue_writel(queue, RBQP, lower_32_bits(queue->rx_ring_dma)); @@ -1006,10 +993,9 @@ static void macb_tx_unmap(struct macb *bp, struct macb_tx_skb *tx_skb, int budge static void macb_set_addr(struct macb *bp, struct macb_dma_desc *desc, dma_addr_t addr) { -#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT - struct macb_dma_desc_64 *desc_64; + if (macb_dma64(bp)) { + struct macb_dma_desc_64 *desc_64; - if (bp->caps & MACB_CAPS_DMA_64B) { desc_64 = macb_64b_desc(bp, desc); desc_64->addrh = upper_32_bits(addr); /* The low bits of RX address contain the RX_USED bit, clearing @@ -1018,26 +1004,23 @@ static void macb_set_addr(struct macb *bp, struct macb_dma_desc *desc, dma_addr_ */ dma_wmb(); } -#endif + desc->addr = lower_32_bits(addr); } static dma_addr_t macb_get_addr(struct macb *bp, struct macb_dma_desc *desc) { dma_addr_t addr = 0; -#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT - struct macb_dma_desc_64 *desc_64; - if (bp->caps & MACB_CAPS_DMA_64B) { + if (macb_dma64(bp)) { + struct macb_dma_desc_64 *desc_64; + desc_64 = macb_64b_desc(bp, desc); addr = ((u64)(desc_64->addrh) << 32); } -#endif addr |= MACB_BF(RX_WADDR, MACB_BFEXT(RX_WADDR, desc->addr)); -#ifdef CONFIG_MACB_USE_HWSTAMP - if (bp->caps & MACB_CAPS_DMA_PTP) + if (macb_dma_ptp(bp)) addr &= ~GEM_BIT(DMA_RXVALID); -#endif return addr; } @@ -2299,11 +2282,9 @@ static netdev_tx_t macb_start_xmit(struct sk_buff *skb, struct net_device *dev) return ret; } -#ifdef CONFIG_MACB_USE_HWSTAMP - if ((skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) && - (bp->caps & MACB_CAPS_DMA_PTP)) + if (macb_dma_ptp(bp) && + (skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP)) skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS; -#endif is_lso = (skb_shinfo(skb)->gso_size != 0); @@ -2780,14 +2761,10 @@ static void macb_configure_dma(struct macb *bp) dmacfg &= ~GEM_BIT(TXCOEN); dmacfg &= ~GEM_BIT(ADDR64); -#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT - if (bp->caps & MACB_CAPS_DMA_64B) + if (macb_dma64(bp)) dmacfg |= GEM_BIT(ADDR64); -#endif -#ifdef CONFIG_MACB_USE_HWSTAMP - if (bp->caps & MACB_CAPS_DMA_PTP) + if (macb_dma_ptp(bp)) dmacfg |= GEM_BIT(RXEXT) | GEM_BIT(TXEXT); -#endif netdev_dbg(bp->dev, "Cadence configure DMA with 0x%08x\n", dmacfg); gem_writel(bp, DMACFG, dmacfg); @@ -3563,7 +3540,7 @@ static int gem_get_ts_info(struct net_device *dev, { struct macb *bp = netdev_priv(dev); - if (!(bp->caps & MACB_CAPS_DMA_PTP)) { + if (!macb_dma_ptp(bp)) { ethtool_op_get_ts_info(dev, info); return 0; } diff --git a/drivers/net/ethernet/cadence/macb_ptp.c b/drivers/net/ethernet/cadence/macb_ptp.c index f4ab379f2849..c9e77819196e 100644 --- a/drivers/net/ethernet/cadence/macb_ptp.c +++ b/drivers/net/ethernet/cadence/macb_ptp.c @@ -28,10 +28,10 @@ static struct macb_dma_desc_ptp *macb_ptp_desc(struct macb *bp, struct macb_dma_desc *desc) { - if (!(bp->caps & MACB_CAPS_DMA_PTP)) + if (!macb_dma_ptp(bp)) return NULL; - if (bp->caps & MACB_CAPS_DMA_64B) + if (macb_dma64(bp)) return (struct macb_dma_desc_ptp *) ((u8 *)desc + sizeof(struct macb_dma_desc) + sizeof(struct macb_dma_desc_64)); @@ -382,7 +382,7 @@ int gem_get_hwtst(struct net_device *dev, struct macb *bp = netdev_priv(dev); *tstamp_config = bp->tstamp_config; - if (!(bp->caps & MACB_CAPS_DMA_PTP)) + if (!macb_dma_ptp(bp)) return -EOPNOTSUPP; return 0; @@ -409,7 +409,7 @@ int gem_set_hwtst(struct net_device *dev, struct macb *bp = netdev_priv(dev); u32 regval; - if (!(bp->caps & MACB_CAPS_DMA_PTP)) + if (!macb_dma_ptp(bp)) return -EOPNOTSUPP; switch (tstamp_config->tx_type) { From 39a913db6a47fd988353cf83f0ccd6a1ee6e2db3 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Th=C3=A9o=20Lebrun?= Date: Tue, 14 Oct 2025 17:25:11 +0200 Subject: [PATCH 0077/1381] net: macb: remove bp->queue_mask MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The low 16 bits of GEM_DCFG6 tell us which queues are enabled in HW. In theory, there could be holes in the bitfield. In practice, the macb driver would fail if there were holes as most loops iterate upon bp->num_queues. Only macb_init() iterated correctly. - Drop bp->queue_mask field. - Error out at probe if a hole is in the queue mask. - Rely upon bp->num_queues for iteration. - As we drop the queue_mask probe local variable, fix RCT. - Compute queue_mask on the fly for TAPRIO using bp->num_queues. Signed-off-by: Théo Lebrun Link: https://patch.msgid.link/20251014-macb-cleanup-v1-10-31cd266e22cd@bootlin.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/cadence/macb.h | 1 - drivers/net/ethernet/cadence/macb_main.c | 69 +++++++++++++----------- 2 files changed, 37 insertions(+), 33 deletions(-) diff --git a/drivers/net/ethernet/cadence/macb.h b/drivers/net/ethernet/cadence/macb.h index f696b2ddc412..5b7d4cdb204d 100644 --- a/drivers/net/ethernet/cadence/macb.h +++ b/drivers/net/ethernet/cadence/macb.h @@ -1290,7 +1290,6 @@ struct macb { unsigned int tx_ring_size; unsigned int num_queues; - unsigned int queue_mask; struct macb_queue queues[MACB_MAX_QUEUES]; spinlock_t lock; diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c index 9db419b94d0b..98e28d51a6e1 100644 --- a/drivers/net/ethernet/cadence/macb_main.c +++ b/drivers/net/ethernet/cadence/macb_main.c @@ -4066,6 +4066,8 @@ static int macb_taprio_setup_replace(struct net_device *ndev, struct macb *bp = netdev_priv(ndev); struct ethtool_link_ksettings kset; struct macb_queue *queue; + u32 queue_mask; + u8 queue_id; size_t i; int err; @@ -4117,8 +4119,9 @@ static int macb_taprio_setup_replace(struct net_device *ndev, goto cleanup; } - /* gate_mask must not select queues outside the valid queue_mask */ - if (entry->gate_mask & ~bp->queue_mask) { + /* gate_mask must not select queues outside the valid queues */ + queue_id = order_base_2(entry->gate_mask); + if (queue_id >= bp->num_queues) { netdev_err(ndev, "Entry %zu: gate_mask 0x%x exceeds queue range (max_queues=%d)\n", i, entry->gate_mask, bp->num_queues); err = -EINVAL; @@ -4152,7 +4155,7 @@ static int macb_taprio_setup_replace(struct net_device *ndev, goto cleanup; } - enst_queue[i].queue_id = order_base_2(entry->gate_mask); + enst_queue[i].queue_id = queue_id; enst_queue[i].start_time_mask = (start_time_sec << GEM_START_TIME_SEC_OFFSET) | start_time_nsec; @@ -4180,8 +4183,9 @@ static int macb_taprio_setup_replace(struct net_device *ndev, /* All validations passed - proceed with hardware configuration */ scoped_guard(spinlock_irqsave, &bp->lock) { /* Disable ENST queues if running before configuring */ + queue_mask = BIT_U32(bp->num_queues) - 1; gem_writel(bp, ENST_CONTROL, - bp->queue_mask << GEM_ENST_DISABLE_QUEUE_OFFSET); + queue_mask << GEM_ENST_DISABLE_QUEUE_OFFSET); for (i = 0; i < conf->num_entries; i++) { queue = &bp->queues[enst_queue[i].queue_id]; @@ -4210,15 +4214,16 @@ static void macb_taprio_destroy(struct net_device *ndev) { struct macb *bp = netdev_priv(ndev); struct macb_queue *queue; - u32 enst_disable_mask; + u32 queue_mask; unsigned int q; netdev_reset_tc(ndev); - enst_disable_mask = bp->queue_mask << GEM_ENST_DISABLE_QUEUE_OFFSET; + queue_mask = BIT_U32(bp->num_queues) - 1; scoped_guard(spinlock_irqsave, &bp->lock) { /* Single disable command for all queues */ - gem_writel(bp, ENST_CONTROL, enst_disable_mask); + gem_writel(bp, ENST_CONTROL, + queue_mask << GEM_ENST_DISABLE_QUEUE_OFFSET); /* Clear all queue ENST registers in batch */ for (q = 0, queue = bp->queues; q < bp->num_queues; ++q, ++queue) { @@ -4341,26 +4346,25 @@ static void macb_configure_caps(struct macb *bp, dev_dbg(&bp->pdev->dev, "Cadence caps 0x%08x\n", bp->caps); } -static void macb_probe_queues(void __iomem *mem, - bool native_io, - unsigned int *queue_mask, - unsigned int *num_queues) +static int macb_probe_queues(struct device *dev, void __iomem *mem, bool native_io) { - *queue_mask = 0x1; - *num_queues = 1; + /* BIT(0) is never set but queue 0 always exists. */ + unsigned int queue_mask = 0x1; - /* is it macb or gem ? - * - * We need to read directly from the hardware here because - * we are early in the probe process and don't have the - * MACB_CAPS_MACB_IS_GEM flag positioned - */ - if (!hw_is_gem(mem, native_io)) - return; + /* Use hw_is_gem() as MACB_CAPS_MACB_IS_GEM is not yet positioned. */ + if (hw_is_gem(mem, native_io)) { + if (native_io) + queue_mask |= __raw_readl(mem + GEM_DCFG6) & 0xFF; + else + queue_mask |= readl_relaxed(mem + GEM_DCFG6) & 0xFF; - /* bit 0 is never set but queue 0 always exists */ - *queue_mask |= readl_relaxed(mem + GEM_DCFG6) & 0xff; - *num_queues = hweight32(*queue_mask); + if (fls(queue_mask) != ffz(queue_mask)) { + dev_err(dev, "queue mask %#x has a hole\n", queue_mask); + return -EINVAL; + } + } + + return hweight32(queue_mask); } static void macb_clks_disable(struct clk *pclk, struct clk *hclk, struct clk *tx_clk, @@ -4478,10 +4482,7 @@ static int macb_init(struct platform_device *pdev) * register mapping but we don't want to test the queue index then * compute the corresponding register offset at run time. */ - for (hw_q = 0, q = 0; hw_q < MACB_MAX_QUEUES; ++hw_q) { - if (!(bp->queue_mask & (1 << hw_q))) - continue; - + for (hw_q = 0, q = 0; hw_q < bp->num_queues; ++hw_q) { queue = &bp->queues[q]; queue->bp = bp; spin_lock_init(&queue->tx_ptr_lock); @@ -5385,14 +5386,14 @@ static int macb_probe(struct platform_device *pdev) struct device_node *np = pdev->dev.of_node; struct clk *pclk, *hclk = NULL, *tx_clk = NULL, *rx_clk = NULL; struct clk *tsu_clk = NULL; - unsigned int queue_mask, num_queues; - bool native_io; phy_interface_t interface; struct net_device *dev; struct resource *regs; u32 wtrmrk_rst_val; void __iomem *mem; struct macb *bp; + int num_queues; + bool native_io; int err, val; mem = devm_platform_get_and_ioremap_resource(pdev, 0, ®s); @@ -5418,7 +5419,12 @@ static int macb_probe(struct platform_device *pdev) pm_runtime_enable(&pdev->dev); native_io = hw_is_native_io(mem); - macb_probe_queues(mem, native_io, &queue_mask, &num_queues); + num_queues = macb_probe_queues(&pdev->dev, mem, native_io); + if (num_queues < 0) { + err = num_queues; + goto err_disable_clocks; + } + dev = alloc_etherdev_mq(sizeof(*bp), num_queues); if (!dev) { err = -ENOMEM; @@ -5442,7 +5448,6 @@ static int macb_probe(struct platform_device *pdev) bp->macb_reg_writel = hw_writel; } bp->num_queues = num_queues; - bp->queue_mask = queue_mask; bp->dma_burst_length = macb_config->dma_burst_length; bp->pclk = pclk; bp->hclk = hclk; From f26c6438a285157e24973b4188d75d4674def46c Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Th=C3=A9o=20Lebrun?= Date: Tue, 14 Oct 2025 17:25:12 +0200 Subject: [PATCH 0078/1381] net: macb: replace min() with umin() calls MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Whenever min(a, b) is used with a and b unsigned variables or literals, `make W=2` complains. Change four min() calls into umin(). stderr extract (GCC 11.2.0, MIPS Codescape): ./include/linux/minmax.h:68:57: warning: comparison is always true due to limited range of data type [-Wtype-limits] 68 | #define __is_nonneg(ux) statically_true((long long)(ux) >= 0) | ^~ drivers/net/ethernet/cadence/macb_main.c:2299:26: note: in expansion of macro ‘min’ 2299 | hdrlen = min(skb_headlen(skb), bp->max_tx_length); | ^~~ Signed-off-by: Théo Lebrun Link: https://patch.msgid.link/20251014-macb-cleanup-v1-11-31cd266e22cd@bootlin.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/cadence/macb_main.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c index 98e28d51a6e1..6c6bc6aa23c7 100644 --- a/drivers/net/ethernet/cadence/macb_main.c +++ b/drivers/net/ethernet/cadence/macb_main.c @@ -2035,7 +2035,7 @@ static unsigned int macb_tx_map(struct macb *bp, count++; tx_head++; - size = min(len, bp->max_tx_length); + size = umin(len, bp->max_tx_length); } /* Then, map paged data from fragments */ @@ -2045,7 +2045,7 @@ static unsigned int macb_tx_map(struct macb *bp, len = skb_frag_size(frag); offset = 0; while (len) { - size = min(len, bp->max_tx_length); + size = umin(len, bp->max_tx_length); entry = macb_tx_ring_wrap(bp, tx_head); tx_skb = &queue->tx_skb[entry]; @@ -2301,7 +2301,7 @@ static netdev_tx_t macb_start_xmit(struct sk_buff *skb, struct net_device *dev) return NETDEV_TX_BUSY; } } else - hdrlen = min(skb_headlen(skb), bp->max_tx_length); + hdrlen = umin(skb_headlen(skb), bp->max_tx_length); #if defined(DEBUG) && defined(VERBOSE_DEBUG) netdev_vdbg(bp->dev, @@ -4573,8 +4573,8 @@ static int macb_init(struct platform_device *pdev) * each 4-tuple define requires 1 T2 screener reg + 3 compare regs */ reg = gem_readl(bp, DCFG8); - bp->max_tuples = min((GEM_BFEXT(SCR2CMP, reg) / 3), - GEM_BFEXT(T2SCR, reg)); + bp->max_tuples = umin((GEM_BFEXT(SCR2CMP, reg) / 3), + GEM_BFEXT(T2SCR, reg)); INIT_LIST_HEAD(&bp->rx_fs_list.list); if (bp->max_tuples > 0) { /* also needs one ethtype match to check IPv4 */ From 027202adf07952d9b4d7a3199fe1014a537e7a81 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Th=C3=A9o=20Lebrun?= Date: Tue, 14 Oct 2025 17:25:13 +0200 Subject: [PATCH 0079/1381] net: macb: drop `entry` local variable in macb_tx_map() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The pattern: entry = macb_tx_ring_wrap(bp, i); tx_skb = &queue->tx_skb[entry]; is the exact definition of: macb_tx_skb(queue, i); The pattern: entry = macb_tx_ring_wrap(bp, i); desc = macb_tx_desc(queue, entry); is redundant because macb_tx_desc() calls macb_tx_ring_wrap(). One explicit call to macb_tx_ring_wrap() is still required for checking if it is the last buffer (TX_WRAP case). Signed-off-by: Théo Lebrun Link: https://patch.msgid.link/20251014-macb-cleanup-v1-12-31cd266e22cd@bootlin.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/cadence/macb_main.c | 18 +++++++----------- 1 file changed, 7 insertions(+), 11 deletions(-) diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c index 6c6bc6aa23c7..08e541d8f8e6 100644 --- a/drivers/net/ethernet/cadence/macb_main.c +++ b/drivers/net/ethernet/cadence/macb_main.c @@ -1989,7 +1989,7 @@ static unsigned int macb_tx_map(struct macb *bp, unsigned int hdrlen) { dma_addr_t mapping; - unsigned int len, entry, i, tx_head = queue->tx_head; + unsigned int len, i, tx_head = queue->tx_head; struct macb_tx_skb *tx_skb = NULL; struct macb_dma_desc *desc; unsigned int offset, size, count = 0; @@ -2015,8 +2015,7 @@ static unsigned int macb_tx_map(struct macb *bp, offset = 0; while (len) { - entry = macb_tx_ring_wrap(bp, tx_head); - tx_skb = &queue->tx_skb[entry]; + tx_skb = macb_tx_skb(queue, tx_head); mapping = dma_map_single(&bp->pdev->dev, skb->data + offset, @@ -2046,8 +2045,7 @@ static unsigned int macb_tx_map(struct macb *bp, offset = 0; while (len) { size = umin(len, bp->max_tx_length); - entry = macb_tx_ring_wrap(bp, tx_head); - tx_skb = &queue->tx_skb[entry]; + tx_skb = macb_tx_skb(queue, tx_head); mapping = skb_frag_dma_map(&bp->pdev->dev, frag, offset, size, DMA_TO_DEVICE); @@ -2084,9 +2082,8 @@ static unsigned int macb_tx_map(struct macb *bp, * to set the end of TX queue */ i = tx_head; - entry = macb_tx_ring_wrap(bp, i); ctrl = MACB_BIT(TX_USED); - desc = macb_tx_desc(queue, entry); + desc = macb_tx_desc(queue, i); desc->ctrl = ctrl; if (lso_ctrl) { @@ -2106,16 +2103,15 @@ static unsigned int macb_tx_map(struct macb *bp, do { i--; - entry = macb_tx_ring_wrap(bp, i); - tx_skb = &queue->tx_skb[entry]; - desc = macb_tx_desc(queue, entry); + tx_skb = macb_tx_skb(queue, i); + desc = macb_tx_desc(queue, i); ctrl = (u32)tx_skb->size; if (eof) { ctrl |= MACB_BIT(TX_LAST); eof = 0; } - if (unlikely(entry == (bp->tx_ring_size - 1))) + if (unlikely(macb_tx_ring_wrap(bp, i) == bp->tx_ring_size - 1)) ctrl |= MACB_BIT(TX_WRAP); /* First descriptor is header descriptor */ From b5fe4f3e5912d79832bc37b207234aaf84c1b789 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Th=C3=A9o=20Lebrun?= Date: Tue, 14 Oct 2025 17:25:14 +0200 Subject: [PATCH 0080/1381] net: macb: drop `count` local variable in macb_tx_map() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Local variable `count` is useless: it counts number of DMA descriptors used and returns it. But the return value is only checked for error. Drop counting the number of DMA descriptors and return a usual negative-if-error integer. Signed-off-by: Théo Lebrun Link: https://patch.msgid.link/20251014-macb-cleanup-v1-13-31cd266e22cd@bootlin.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/cadence/macb_main.c | 10 ++++------ 1 file changed, 4 insertions(+), 6 deletions(-) diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c index 08e541d8f8e6..dd3b13fa3047 100644 --- a/drivers/net/ethernet/cadence/macb_main.c +++ b/drivers/net/ethernet/cadence/macb_main.c @@ -1992,7 +1992,7 @@ static unsigned int macb_tx_map(struct macb *bp, unsigned int len, i, tx_head = queue->tx_head; struct macb_tx_skb *tx_skb = NULL; struct macb_dma_desc *desc; - unsigned int offset, size, count = 0; + unsigned int offset, size; unsigned int f, nr_frags = skb_shinfo(skb)->nr_frags; unsigned int eof = 1, mss_mfs = 0; u32 ctrl, lso_ctrl = 0, seq_ctrl = 0; @@ -2031,7 +2031,6 @@ static unsigned int macb_tx_map(struct macb *bp, len -= size; offset += size; - count++; tx_head++; size = umin(len, bp->max_tx_length); @@ -2060,7 +2059,6 @@ static unsigned int macb_tx_map(struct macb *bp, len -= size; offset += size; - count++; tx_head++; } } @@ -2139,7 +2137,7 @@ static unsigned int macb_tx_map(struct macb *bp, queue->tx_head = tx_head; - return count; + return 0; dma_error: netdev_err(bp->dev, "TX DMA map failed\n"); @@ -2150,7 +2148,7 @@ static unsigned int macb_tx_map(struct macb *bp, macb_tx_unmap(bp, tx_skb, 0); } - return 0; + return -ENOMEM; } static netdev_features_t macb_features_check(struct sk_buff *skb, @@ -2336,7 +2334,7 @@ static netdev_tx_t macb_start_xmit(struct sk_buff *skb, struct net_device *dev) } /* Map socket buffer for DMA transfer */ - if (!macb_tx_map(bp, queue, skb, hdrlen)) { + if (macb_tx_map(bp, queue, skb, hdrlen)) { dev_kfree_skb_any(skb); goto unlock; } From 1ce9662e31fdae96df543b22ecbc1b48f6db753a Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Th=C3=A9o=20Lebrun?= Date: Tue, 14 Oct 2025 17:25:15 +0200 Subject: [PATCH 0081/1381] net: macb: apply reverse christmas tree in macb_tx_map() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The arguments grew over time; follow conventions and apply reverse christmas tree (RCT). Signed-off-by: Théo Lebrun Link: https://patch.msgid.link/20251014-macb-cleanup-v1-14-31cd266e22cd@bootlin.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/cadence/macb_main.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c index dd3b13fa3047..e15fcdd43d77 100644 --- a/drivers/net/ethernet/cadence/macb_main.c +++ b/drivers/net/ethernet/cadence/macb_main.c @@ -1988,14 +1988,14 @@ static unsigned int macb_tx_map(struct macb *bp, struct sk_buff *skb, unsigned int hdrlen) { - dma_addr_t mapping; + unsigned int f, nr_frags = skb_shinfo(skb)->nr_frags; unsigned int len, i, tx_head = queue->tx_head; + u32 ctrl, lso_ctrl = 0, seq_ctrl = 0; + unsigned int eof = 1, mss_mfs = 0; struct macb_tx_skb *tx_skb = NULL; struct macb_dma_desc *desc; unsigned int offset, size; - unsigned int f, nr_frags = skb_shinfo(skb)->nr_frags; - unsigned int eof = 1, mss_mfs = 0; - u32 ctrl, lso_ctrl = 0, seq_ctrl = 0; + dma_addr_t mapping; /* LSO */ if (skb_shinfo(skb)->gso_size != 0) { From 8ebeef3d01c8b9e5807afdf1d38547f4625d0e4e Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Th=C3=A9o=20Lebrun?= Date: Tue, 14 Oct 2025 17:25:16 +0200 Subject: [PATCH 0082/1381] net: macb: sort #includes MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Sort #include preprocessor directives. Reviewed-by: Andrew Lunn Reviewed-by: Sean Anderson Signed-off-by: Théo Lebrun Link: https://patch.msgid.link/20251014-macb-cleanup-v1-15-31cd266e22cd@bootlin.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/cadence/macb_main.c | 37 ++++++++++++------------ 1 file changed, 19 insertions(+), 18 deletions(-) diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c index e15fcdd43d77..214f543af3b8 100644 --- a/drivers/net/ethernet/cadence/macb_main.c +++ b/drivers/net/ethernet/cadence/macb_main.c @@ -6,36 +6,37 @@ */ #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt -#include +#include #include +#include #include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include #include #include -#include -#include -#include -#include -#include -#include -#include #include -#include -#include -#include -#include #include #include #include -#include -#include -#include -#include #include +#include +#include #include #include #include -#include -#include +#include +#include +#include +#include #include #include "macb.h" From 9271d0ea07c27e6f482f20b615c2a4bba991e68c Mon Sep 17 00:00:00 2001 From: Marc Kleine-Budde Date: Wed, 6 Aug 2025 11:38:07 +0200 Subject: [PATCH 0083/1381] can: m_can: add support for optional reset This patch has been split from the original series [1]. In some SoCs (observed on the STM32MP15) the M_CAN IP core keeps the CAN state and CAN error counters over an internal reset cycle. The STM32MP15 SoC provides an external reset, which is shared between both M_CAN cores. Add support for an optional external reset. Take care of shared resets, de-assert reset during the probe phase in m_can_class_register() and while the interface is up, assert the reset otherwise. [1] https://lore.kernel.org/all/20250923-m_can-fix-state-handling-v3-0-06d8baccadbf@pengutronix.de Reviewed-by: Philipp Zabel Reviewed-by: Markus Schneider-Pargmann Link: https://patch.msgid.link/20251008-m_can-add-reset-v1-1-49f0bbf820c4@pengutronix.de Signed-off-by: Marc Kleine-Budde --- drivers/net/can/m_can/m_can.c | 27 ++++++++++++++++++++++++--- drivers/net/can/m_can/m_can.h | 1 + 2 files changed, 25 insertions(+), 3 deletions(-) diff --git a/drivers/net/can/m_can/m_can.c b/drivers/net/can/m_can/m_can.c index ad4f577c1ef7..48b7a67336b5 100644 --- a/drivers/net/can/m_can/m_can.c +++ b/drivers/net/can/m_can/m_can.c @@ -23,6 +23,7 @@ #include #include #include +#include #include "m_can.h" @@ -1827,6 +1828,7 @@ static int m_can_close(struct net_device *dev) close_candev(dev); + reset_control_assert(cdev->rst); m_can_clk_stop(cdev); phy_power_off(cdev->transceiver); @@ -2069,11 +2071,15 @@ static int m_can_open(struct net_device *dev) if (err) goto out_phy_power_off; + err = reset_control_deassert(cdev->rst); + if (err) + goto exit_disable_clks; + /* open the can device */ err = open_candev(dev); if (err) { netdev_err(dev, "failed to open can device\n"); - goto exit_disable_clks; + goto out_reset_control_assert; } if (cdev->is_peripheral) @@ -2129,6 +2135,8 @@ static int m_can_open(struct net_device *dev) else napi_disable(&cdev->napi); close_candev(dev); +out_reset_control_assert: + reset_control_assert(cdev->rst); exit_disable_clks: m_can_clk_stop(cdev); out_phy_power_off: @@ -2417,15 +2425,24 @@ int m_can_class_register(struct m_can_classdev *cdev) } } + cdev->rst = devm_reset_control_get_optional_shared(cdev->dev, NULL); + if (IS_ERR(cdev->rst)) + return dev_err_probe(cdev->dev, PTR_ERR(cdev->rst), + "Failed to get reset line\n"); + ret = m_can_clk_start(cdev); if (ret) return ret; + ret = reset_control_deassert(cdev->rst); + if (ret) + goto clk_disable; + if (cdev->is_peripheral) { ret = can_rx_offload_add_manual(cdev->net, &cdev->offload, NAPI_POLL_WEIGHT); if (ret) - goto clk_disable; + goto out_reset_control_assert; } if (!cdev->net->irq) { @@ -2454,8 +2471,10 @@ int m_can_class_register(struct m_can_classdev *cdev) KBUILD_MODNAME, cdev->net->irq, cdev->version); /* Probe finished - * Stop clocks. They will be reactivated once the M_CAN device is opened + * Assert reset and stop clocks. + * They will be reactivated once the M_CAN device is opened */ + reset_control_assert(cdev->rst); m_can_clk_stop(cdev); return 0; @@ -2463,6 +2482,8 @@ int m_can_class_register(struct m_can_classdev *cdev) rx_offload_del: if (cdev->is_peripheral) can_rx_offload_del(&cdev->offload); +out_reset_control_assert: + reset_control_assert(cdev->rst); clk_disable: m_can_clk_stop(cdev); diff --git a/drivers/net/can/m_can/m_can.h b/drivers/net/can/m_can/m_can.h index bd4746c63af3..7b7600697c6b 100644 --- a/drivers/net/can/m_can/m_can.h +++ b/drivers/net/can/m_can/m_can.h @@ -86,6 +86,7 @@ struct m_can_classdev { struct device *dev; struct clk *hclk; struct clk *cclk; + struct reset_control *rst; struct workqueue_struct *tx_wq; struct phy *transceiver; From f968a24cad3da72fdff12a0ae5ac0b679439cca1 Mon Sep 17 00:00:00 2001 From: Vincent Mailhol Date: Fri, 3 Oct 2025 12:16:38 +0900 Subject: [PATCH 0084/1381] can: treewide: remove can_change_mtu() can_change_mtu() became obsolete by commit 23049938605b ("can: populate the minimum and maximum MTU values"). Now that net_device->min_mtu and net_device->max_mtu are populated, all the checks are already done by dev_validate_mtu() in net/core/dev.c. Remove the net_device_ops->ndo_change_mtu() callback of all the physical interfaces, then remove can_change_mtu(). Only keep the vcan_change_mtu() and vxcan_change_mtu() because the virtual interfaces use their own different MTU logic. The only functional change this patch introduces is that now the user will be able to change the MTU even if the interface is up. This does not matter for Classical CAN and CAN FD because their MTU range is composed of only one value, respectively CAN_MTU and CANFD_MTU. For the upcoming CAN XL, the MTU will be configurable within the CANXL_MIN_MTU to CANXL_MAX_MTU range at any time, even if the interface is up. This is consistent with the other net protocols and does not contradict ISO 11898-1:2024 as having a modifiable MTU is a kernel extension. Signed-off-by: Vincent Mailhol Link: https://patch.msgid.link/20251003-remove-can_change_mtu-v1-1-337f8bc21181@kernel.org Signed-off-by: Marc Kleine-Budde --- drivers/net/can/at91_can.c | 1 - drivers/net/can/bxcan.c | 1 - drivers/net/can/c_can/c_can_main.c | 1 - drivers/net/can/can327.c | 1 - drivers/net/can/cc770/cc770.c | 1 - drivers/net/can/ctucanfd/ctucanfd_base.c | 1 - drivers/net/can/dev/dev.c | 38 ------------------- drivers/net/can/esd/esd_402_pci-core.c | 1 - drivers/net/can/flexcan/flexcan-core.c | 1 - drivers/net/can/grcan.c | 1 - drivers/net/can/ifi_canfd/ifi_canfd.c | 1 - drivers/net/can/janz-ican3.c | 1 - .../can/kvaser_pciefd/kvaser_pciefd_core.c | 1 - drivers/net/can/m_can/m_can.c | 1 - drivers/net/can/mscan/mscan.c | 1 - drivers/net/can/peak_canfd/peak_canfd.c | 1 - drivers/net/can/rcar/rcar_can.c | 1 - drivers/net/can/rcar/rcar_canfd.c | 1 - .../net/can/rockchip/rockchip_canfd-core.c | 1 - drivers/net/can/sja1000/sja1000.c | 1 - drivers/net/can/slcan/slcan-core.c | 1 - drivers/net/can/softing/softing_main.c | 1 - drivers/net/can/spi/hi311x.c | 1 - drivers/net/can/spi/mcp251x.c | 1 - .../net/can/spi/mcp251xfd/mcp251xfd-core.c | 1 - drivers/net/can/sun4i_can.c | 1 - drivers/net/can/ti_hecc.c | 1 - drivers/net/can/usb/ems_usb.c | 1 - drivers/net/can/usb/esd_usb.c | 1 - drivers/net/can/usb/etas_es58x/es58x_core.c | 1 - drivers/net/can/usb/f81604.c | 1 - drivers/net/can/usb/gs_usb.c | 1 - .../net/can/usb/kvaser_usb/kvaser_usb_core.c | 1 - drivers/net/can/usb/mcba_usb.c | 1 - drivers/net/can/usb/nct6694_canfd.c | 1 - drivers/net/can/usb/peak_usb/pcan_usb_core.c | 1 - drivers/net/can/usb/ucan.c | 1 - drivers/net/can/usb/usb_8dev.c | 1 - drivers/net/can/xilinx_can.c | 1 - include/linux/can/dev.h | 1 - 40 files changed, 77 deletions(-) diff --git a/drivers/net/can/at91_can.c b/drivers/net/can/at91_can.c index 191707d7e3da..c2a3a4eef5b2 100644 --- a/drivers/net/can/at91_can.c +++ b/drivers/net/can/at91_can.c @@ -948,7 +948,6 @@ static const struct net_device_ops at91_netdev_ops = { .ndo_open = at91_open, .ndo_stop = at91_close, .ndo_start_xmit = at91_start_xmit, - .ndo_change_mtu = can_change_mtu, }; static const struct ethtool_ops at91_ethtool_ops = { diff --git a/drivers/net/can/bxcan.c b/drivers/net/can/bxcan.c index bfc60eb33dc3..9c3af7049814 100644 --- a/drivers/net/can/bxcan.c +++ b/drivers/net/can/bxcan.c @@ -881,7 +881,6 @@ static const struct net_device_ops bxcan_netdev_ops = { .ndo_open = bxcan_open, .ndo_stop = bxcan_stop, .ndo_start_xmit = bxcan_start_xmit, - .ndo_change_mtu = can_change_mtu, }; static const struct ethtool_ops bxcan_ethtool_ops = { diff --git a/drivers/net/can/c_can/c_can_main.c b/drivers/net/can/c_can/c_can_main.c index cc371d0c9f3c..3702cac7fbf0 100644 --- a/drivers/net/can/c_can/c_can_main.c +++ b/drivers/net/can/c_can/c_can_main.c @@ -1362,7 +1362,6 @@ static const struct net_device_ops c_can_netdev_ops = { .ndo_open = c_can_open, .ndo_stop = c_can_close, .ndo_start_xmit = c_can_start_xmit, - .ndo_change_mtu = can_change_mtu, }; int register_c_can_dev(struct net_device *dev) diff --git a/drivers/net/can/can327.c b/drivers/net/can/can327.c index 24af63961030..b66fc16aedd2 100644 --- a/drivers/net/can/can327.c +++ b/drivers/net/can/can327.c @@ -849,7 +849,6 @@ static const struct net_device_ops can327_netdev_ops = { .ndo_open = can327_netdev_open, .ndo_stop = can327_netdev_close, .ndo_start_xmit = can327_netdev_start_xmit, - .ndo_change_mtu = can_change_mtu, }; static const struct ethtool_ops can327_ethtool_ops = { diff --git a/drivers/net/can/cc770/cc770.c b/drivers/net/can/cc770/cc770.c index 30909f3aab57..8d5abd643c06 100644 --- a/drivers/net/can/cc770/cc770.c +++ b/drivers/net/can/cc770/cc770.c @@ -834,7 +834,6 @@ static const struct net_device_ops cc770_netdev_ops = { .ndo_open = cc770_open, .ndo_stop = cc770_close, .ndo_start_xmit = cc770_start_xmit, - .ndo_change_mtu = can_change_mtu, }; static const struct ethtool_ops cc770_ethtool_ops = { diff --git a/drivers/net/can/ctucanfd/ctucanfd_base.c b/drivers/net/can/ctucanfd/ctucanfd_base.c index 8bd3f0fc385c..1e6b9e3dc2fe 100644 --- a/drivers/net/can/ctucanfd/ctucanfd_base.c +++ b/drivers/net/can/ctucanfd/ctucanfd_base.c @@ -1301,7 +1301,6 @@ static const struct net_device_ops ctucan_netdev_ops = { .ndo_open = ctucan_open, .ndo_stop = ctucan_close, .ndo_start_xmit = ctucan_start_xmit, - .ndo_change_mtu = can_change_mtu, }; static const struct ethtool_ops ctucan_ethtool_ops = { diff --git a/drivers/net/can/dev/dev.c b/drivers/net/can/dev/dev.c index 15ccedbb3f8d..0cc3d008adb3 100644 --- a/drivers/net/can/dev/dev.c +++ b/drivers/net/can/dev/dev.c @@ -359,44 +359,6 @@ void can_set_default_mtu(struct net_device *dev) } } -/* changing MTU and control mode for CAN/CANFD devices */ -int can_change_mtu(struct net_device *dev, int new_mtu) -{ - struct can_priv *priv = netdev_priv(dev); - u32 ctrlmode_static = can_get_static_ctrlmode(priv); - - /* Do not allow changing the MTU while running */ - if (dev->flags & IFF_UP) - return -EBUSY; - - /* allow change of MTU according to the CANFD ability of the device */ - switch (new_mtu) { - case CAN_MTU: - /* 'CANFD-only' controllers can not switch to CAN_MTU */ - if (ctrlmode_static & CAN_CTRLMODE_FD) - return -EINVAL; - - priv->ctrlmode &= ~CAN_CTRLMODE_FD; - break; - - case CANFD_MTU: - /* check for potential CANFD ability */ - if (!(priv->ctrlmode_supported & CAN_CTRLMODE_FD) && - !(ctrlmode_static & CAN_CTRLMODE_FD)) - return -EINVAL; - - priv->ctrlmode |= CAN_CTRLMODE_FD; - break; - - default: - return -EINVAL; - } - - WRITE_ONCE(dev->mtu, new_mtu); - return 0; -} -EXPORT_SYMBOL_GPL(can_change_mtu); - /* helper to define static CAN controller features at device creation time */ int can_set_static_ctrlmode(struct net_device *dev, u32 static_mode) { diff --git a/drivers/net/can/esd/esd_402_pci-core.c b/drivers/net/can/esd/esd_402_pci-core.c index 5d6d2828cd04..05adecae6375 100644 --- a/drivers/net/can/esd/esd_402_pci-core.c +++ b/drivers/net/can/esd/esd_402_pci-core.c @@ -86,7 +86,6 @@ static const struct net_device_ops pci402_acc_netdev_ops = { .ndo_open = acc_open, .ndo_stop = acc_close, .ndo_start_xmit = acc_start_xmit, - .ndo_change_mtu = can_change_mtu, .ndo_eth_ioctl = can_eth_ioctl_hwts, }; diff --git a/drivers/net/can/flexcan/flexcan-core.c b/drivers/net/can/flexcan/flexcan-core.c index 06d5d35fc1b5..f5d22c61503f 100644 --- a/drivers/net/can/flexcan/flexcan-core.c +++ b/drivers/net/can/flexcan/flexcan-core.c @@ -1867,7 +1867,6 @@ static const struct net_device_ops flexcan_netdev_ops = { .ndo_open = flexcan_open, .ndo_stop = flexcan_close, .ndo_start_xmit = flexcan_start_xmit, - .ndo_change_mtu = can_change_mtu, }; static int register_flexcandev(struct net_device *dev) diff --git a/drivers/net/can/grcan.c b/drivers/net/can/grcan.c index c5784d9779ef..3b1b09943436 100644 --- a/drivers/net/can/grcan.c +++ b/drivers/net/can/grcan.c @@ -1561,7 +1561,6 @@ static const struct net_device_ops grcan_netdev_ops = { .ndo_open = grcan_open, .ndo_stop = grcan_close, .ndo_start_xmit = grcan_start_xmit, - .ndo_change_mtu = can_change_mtu, }; static const struct ethtool_ops grcan_ethtool_ops = { diff --git a/drivers/net/can/ifi_canfd/ifi_canfd.c b/drivers/net/can/ifi_canfd/ifi_canfd.c index 2eeee65f606f..0f83335e4d07 100644 --- a/drivers/net/can/ifi_canfd/ifi_canfd.c +++ b/drivers/net/can/ifi_canfd/ifi_canfd.c @@ -944,7 +944,6 @@ static const struct net_device_ops ifi_canfd_netdev_ops = { .ndo_open = ifi_canfd_open, .ndo_stop = ifi_canfd_close, .ndo_start_xmit = ifi_canfd_start_xmit, - .ndo_change_mtu = can_change_mtu, }; static const struct ethtool_ops ifi_canfd_ethtool_ops = { diff --git a/drivers/net/can/janz-ican3.c b/drivers/net/can/janz-ican3.c index bfa5cbe88017..1efdd1fd8caa 100644 --- a/drivers/net/can/janz-ican3.c +++ b/drivers/net/can/janz-ican3.c @@ -1752,7 +1752,6 @@ static const struct net_device_ops ican3_netdev_ops = { .ndo_open = ican3_open, .ndo_stop = ican3_stop, .ndo_start_xmit = ican3_xmit, - .ndo_change_mtu = can_change_mtu, }; static const struct ethtool_ops ican3_ethtool_ops = { diff --git a/drivers/net/can/kvaser_pciefd/kvaser_pciefd_core.c b/drivers/net/can/kvaser_pciefd/kvaser_pciefd_core.c index 0880023611be..705f9bb74cd2 100644 --- a/drivers/net/can/kvaser_pciefd/kvaser_pciefd_core.c +++ b/drivers/net/can/kvaser_pciefd/kvaser_pciefd_core.c @@ -904,7 +904,6 @@ static const struct net_device_ops kvaser_pciefd_netdev_ops = { .ndo_stop = kvaser_pciefd_stop, .ndo_eth_ioctl = can_eth_ioctl_hwts, .ndo_start_xmit = kvaser_pciefd_start_xmit, - .ndo_change_mtu = can_change_mtu, }; static int kvaser_pciefd_set_phys_id(struct net_device *netdev, diff --git a/drivers/net/can/m_can/m_can.c b/drivers/net/can/m_can/m_can.c index 48b7a67336b5..873f5991fc5a 100644 --- a/drivers/net/can/m_can/m_can.c +++ b/drivers/net/can/m_can/m_can.c @@ -2148,7 +2148,6 @@ static const struct net_device_ops m_can_netdev_ops = { .ndo_open = m_can_open, .ndo_stop = m_can_close, .ndo_start_xmit = m_can_start_xmit, - .ndo_change_mtu = can_change_mtu, }; static int m_can_get_coalesce(struct net_device *dev, diff --git a/drivers/net/can/mscan/mscan.c b/drivers/net/can/mscan/mscan.c index 8c2a7bc64d3d..39c7aa2a0b2f 100644 --- a/drivers/net/can/mscan/mscan.c +++ b/drivers/net/can/mscan/mscan.c @@ -607,7 +607,6 @@ static const struct net_device_ops mscan_netdev_ops = { .ndo_open = mscan_open, .ndo_stop = mscan_close, .ndo_start_xmit = mscan_start_xmit, - .ndo_change_mtu = can_change_mtu, }; static const struct ethtool_ops mscan_ethtool_ops = { diff --git a/drivers/net/can/peak_canfd/peak_canfd.c b/drivers/net/can/peak_canfd/peak_canfd.c index b5bc80ac7876..a53c9d347b7b 100644 --- a/drivers/net/can/peak_canfd/peak_canfd.c +++ b/drivers/net/can/peak_canfd/peak_canfd.c @@ -773,7 +773,6 @@ static const struct net_device_ops peak_canfd_netdev_ops = { .ndo_stop = peak_canfd_close, .ndo_eth_ioctl = peak_eth_ioctl, .ndo_start_xmit = peak_canfd_start_xmit, - .ndo_change_mtu = can_change_mtu, }; static int peak_get_ts_info(struct net_device *dev, diff --git a/drivers/net/can/rcar/rcar_can.c b/drivers/net/can/rcar/rcar_can.c index 5f85f4e27205..fc3df328e877 100644 --- a/drivers/net/can/rcar/rcar_can.c +++ b/drivers/net/can/rcar/rcar_can.c @@ -635,7 +635,6 @@ static const struct net_device_ops rcar_can_netdev_ops = { .ndo_open = rcar_can_open, .ndo_stop = rcar_can_close, .ndo_start_xmit = rcar_can_start_xmit, - .ndo_change_mtu = can_change_mtu, }; static const struct ethtool_ops rcar_can_ethtool_ops = { diff --git a/drivers/net/can/rcar/rcar_canfd.c b/drivers/net/can/rcar/rcar_canfd.c index 45d36adb51b7..49ab65274b51 100644 --- a/drivers/net/can/rcar/rcar_canfd.c +++ b/drivers/net/can/rcar/rcar_canfd.c @@ -1818,7 +1818,6 @@ static const struct net_device_ops rcar_canfd_netdev_ops = { .ndo_open = rcar_canfd_open, .ndo_stop = rcar_canfd_close, .ndo_start_xmit = rcar_canfd_start_xmit, - .ndo_change_mtu = can_change_mtu, }; static const struct ethtool_ops rcar_canfd_ethtool_ops = { diff --git a/drivers/net/can/rockchip/rockchip_canfd-core.c b/drivers/net/can/rockchip/rockchip_canfd-core.c index 046f0a0ae4d4..29de0c01e4ed 100644 --- a/drivers/net/can/rockchip/rockchip_canfd-core.c +++ b/drivers/net/can/rockchip/rockchip_canfd-core.c @@ -761,7 +761,6 @@ static const struct net_device_ops rkcanfd_netdev_ops = { .ndo_open = rkcanfd_open, .ndo_stop = rkcanfd_stop, .ndo_start_xmit = rkcanfd_start_xmit, - .ndo_change_mtu = can_change_mtu, }; static int __maybe_unused rkcanfd_runtime_suspend(struct device *dev) diff --git a/drivers/net/can/sja1000/sja1000.c b/drivers/net/can/sja1000/sja1000.c index 4d245857ef1c..acfa49db3907 100644 --- a/drivers/net/can/sja1000/sja1000.c +++ b/drivers/net/can/sja1000/sja1000.c @@ -697,7 +697,6 @@ static const struct net_device_ops sja1000_netdev_ops = { .ndo_open = sja1000_open, .ndo_stop = sja1000_close, .ndo_start_xmit = sja1000_start_xmit, - .ndo_change_mtu = can_change_mtu, }; static const struct ethtool_ops sja1000_ethtool_ops = { diff --git a/drivers/net/can/slcan/slcan-core.c b/drivers/net/can/slcan/slcan-core.c index 58ff2ec1d975..cd789e178d34 100644 --- a/drivers/net/can/slcan/slcan-core.c +++ b/drivers/net/can/slcan/slcan-core.c @@ -774,7 +774,6 @@ static const struct net_device_ops slcan_netdev_ops = { .ndo_open = slcan_netdev_open, .ndo_stop = slcan_netdev_close, .ndo_start_xmit = slcan_netdev_xmit, - .ndo_change_mtu = can_change_mtu, }; /****************************************** diff --git a/drivers/net/can/softing/softing_main.c b/drivers/net/can/softing/softing_main.c index 278ee8722770..79bc64395ac4 100644 --- a/drivers/net/can/softing/softing_main.c +++ b/drivers/net/can/softing/softing_main.c @@ -609,7 +609,6 @@ static const struct net_device_ops softing_netdev_ops = { .ndo_open = softing_netdev_open, .ndo_stop = softing_netdev_stop, .ndo_start_xmit = softing_netdev_start_xmit, - .ndo_change_mtu = can_change_mtu, }; static const struct ethtool_ops softing_ethtool_ops = { diff --git a/drivers/net/can/spi/hi311x.c b/drivers/net/can/spi/hi311x.c index 6d4b643e135f..e00d3dbc4cf4 100644 --- a/drivers/net/can/spi/hi311x.c +++ b/drivers/net/can/spi/hi311x.c @@ -799,7 +799,6 @@ static const struct net_device_ops hi3110_netdev_ops = { .ndo_open = hi3110_open, .ndo_stop = hi3110_stop, .ndo_start_xmit = hi3110_hard_start_xmit, - .ndo_change_mtu = can_change_mtu, }; static const struct ethtool_ops hi3110_ethtool_ops = { diff --git a/drivers/net/can/spi/mcp251x.c b/drivers/net/can/spi/mcp251x.c index b797e08499d7..1e54e1a22702 100644 --- a/drivers/net/can/spi/mcp251x.c +++ b/drivers/net/can/spi/mcp251x.c @@ -1270,7 +1270,6 @@ static const struct net_device_ops mcp251x_netdev_ops = { .ndo_open = mcp251x_open, .ndo_stop = mcp251x_stop, .ndo_start_xmit = mcp251x_hard_start_xmit, - .ndo_change_mtu = can_change_mtu, }; static const struct ethtool_ops mcp251x_ethtool_ops = { diff --git a/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c b/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c index 7450ea42c1ea..9402530ba3d4 100644 --- a/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c +++ b/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c @@ -1715,7 +1715,6 @@ static const struct net_device_ops mcp251xfd_netdev_ops = { .ndo_stop = mcp251xfd_stop, .ndo_start_xmit = mcp251xfd_start_xmit, .ndo_eth_ioctl = can_eth_ioctl_hwts, - .ndo_change_mtu = can_change_mtu, }; static void diff --git a/drivers/net/can/sun4i_can.c b/drivers/net/can/sun4i_can.c index 53bfd873de9b..6fcb301ef611 100644 --- a/drivers/net/can/sun4i_can.c +++ b/drivers/net/can/sun4i_can.c @@ -768,7 +768,6 @@ static const struct net_device_ops sun4ican_netdev_ops = { .ndo_open = sun4ican_open, .ndo_stop = sun4ican_close, .ndo_start_xmit = sun4ican_start_xmit, - .ndo_change_mtu = can_change_mtu, }; static const struct ethtool_ops sun4ican_ethtool_ops = { diff --git a/drivers/net/can/ti_hecc.c b/drivers/net/can/ti_hecc.c index e6d6661a908a..1d3dbf28b105 100644 --- a/drivers/net/can/ti_hecc.c +++ b/drivers/net/can/ti_hecc.c @@ -829,7 +829,6 @@ static const struct net_device_ops ti_hecc_netdev_ops = { .ndo_open = ti_hecc_open, .ndo_stop = ti_hecc_close, .ndo_start_xmit = ti_hecc_xmit, - .ndo_change_mtu = can_change_mtu, }; static const struct ethtool_ops ti_hecc_ethtool_ops = { diff --git a/drivers/net/can/usb/ems_usb.c b/drivers/net/can/usb/ems_usb.c index 5355bac4dccb..de8e212a1366 100644 --- a/drivers/net/can/usb/ems_usb.c +++ b/drivers/net/can/usb/ems_usb.c @@ -885,7 +885,6 @@ static const struct net_device_ops ems_usb_netdev_ops = { .ndo_open = ems_usb_open, .ndo_stop = ems_usb_close, .ndo_start_xmit = ems_usb_start_xmit, - .ndo_change_mtu = can_change_mtu, }; static const struct ethtool_ops ems_usb_ethtool_ops = { diff --git a/drivers/net/can/usb/esd_usb.c b/drivers/net/can/usb/esd_usb.c index 9bc1824d7be6..08da507faef4 100644 --- a/drivers/net/can/usb/esd_usb.c +++ b/drivers/net/can/usb/esd_usb.c @@ -1011,7 +1011,6 @@ static const struct net_device_ops esd_usb_netdev_ops = { .ndo_open = esd_usb_open, .ndo_stop = esd_usb_close, .ndo_start_xmit = esd_usb_start_xmit, - .ndo_change_mtu = can_change_mtu, }; static const struct ethtool_ops esd_usb_ethtool_ops = { diff --git a/drivers/net/can/usb/etas_es58x/es58x_core.c b/drivers/net/can/usb/etas_es58x/es58x_core.c index adc91873c083..47d9e03f3044 100644 --- a/drivers/net/can/usb/etas_es58x/es58x_core.c +++ b/drivers/net/can/usb/etas_es58x/es58x_core.c @@ -1977,7 +1977,6 @@ static const struct net_device_ops es58x_netdev_ops = { .ndo_stop = es58x_stop, .ndo_start_xmit = es58x_start_xmit, .ndo_eth_ioctl = can_eth_ioctl_hwts, - .ndo_change_mtu = can_change_mtu, }; static const struct ethtool_ops es58x_ethtool_ops = { diff --git a/drivers/net/can/usb/f81604.c b/drivers/net/can/usb/f81604.c index e0cfa1460b0b..efe61ece79ea 100644 --- a/drivers/net/can/usb/f81604.c +++ b/drivers/net/can/usb/f81604.c @@ -1052,7 +1052,6 @@ static const struct net_device_ops f81604_netdev_ops = { .ndo_open = f81604_open, .ndo_stop = f81604_close, .ndo_start_xmit = f81604_start_xmit, - .ndo_change_mtu = can_change_mtu, }; static const struct can_bittiming_const f81604_bittiming_const = { diff --git a/drivers/net/can/usb/gs_usb.c b/drivers/net/can/usb/gs_usb.c index 69b8d6da651b..30608901a974 100644 --- a/drivers/net/can/usb/gs_usb.c +++ b/drivers/net/can/usb/gs_usb.c @@ -1101,7 +1101,6 @@ static const struct net_device_ops gs_usb_netdev_ops = { .ndo_open = gs_can_open, .ndo_stop = gs_can_close, .ndo_start_xmit = gs_can_start_xmit, - .ndo_change_mtu = can_change_mtu, .ndo_eth_ioctl = gs_can_eth_ioctl, }; diff --git a/drivers/net/can/usb/kvaser_usb/kvaser_usb_core.c b/drivers/net/can/usb/kvaser_usb/kvaser_usb_core.c index 90e77fa0ff4a..89e22b66f919 100644 --- a/drivers/net/can/usb/kvaser_usb/kvaser_usb_core.c +++ b/drivers/net/can/usb/kvaser_usb/kvaser_usb_core.c @@ -786,7 +786,6 @@ static const struct net_device_ops kvaser_usb_netdev_ops = { .ndo_stop = kvaser_usb_close, .ndo_eth_ioctl = can_eth_ioctl_hwts, .ndo_start_xmit = kvaser_usb_start_xmit, - .ndo_change_mtu = can_change_mtu, }; static const struct ethtool_ops kvaser_usb_ethtool_ops = { diff --git a/drivers/net/can/usb/mcba_usb.c b/drivers/net/can/usb/mcba_usb.c index 1f9b915094e6..41c0a1c399bf 100644 --- a/drivers/net/can/usb/mcba_usb.c +++ b/drivers/net/can/usb/mcba_usb.c @@ -761,7 +761,6 @@ static const struct net_device_ops mcba_netdev_ops = { .ndo_open = mcba_usb_open, .ndo_stop = mcba_usb_close, .ndo_start_xmit = mcba_usb_start_xmit, - .ndo_change_mtu = can_change_mtu, }; static const struct ethtool_ops mcba_ethtool_ops = { diff --git a/drivers/net/can/usb/nct6694_canfd.c b/drivers/net/can/usb/nct6694_canfd.c index 8deff16491a1..dd6df2ec3742 100644 --- a/drivers/net/can/usb/nct6694_canfd.c +++ b/drivers/net/can/usb/nct6694_canfd.c @@ -690,7 +690,6 @@ static const struct net_device_ops nct6694_canfd_netdev_ops = { .ndo_open = nct6694_canfd_open, .ndo_stop = nct6694_canfd_close, .ndo_start_xmit = nct6694_canfd_start_xmit, - .ndo_change_mtu = can_change_mtu, }; static const struct ethtool_ops nct6694_canfd_ethtool_ops = { diff --git a/drivers/net/can/usb/peak_usb/pcan_usb_core.c b/drivers/net/can/usb/peak_usb/pcan_usb_core.c index c74302ca7cee..94b1d7f15d27 100644 --- a/drivers/net/can/usb/peak_usb/pcan_usb_core.c +++ b/drivers/net/can/usb/peak_usb/pcan_usb_core.c @@ -814,7 +814,6 @@ static const struct net_device_ops peak_usb_netdev_ops = { .ndo_stop = peak_usb_ndo_stop, .ndo_eth_ioctl = peak_eth_ioctl, .ndo_start_xmit = peak_usb_ndo_start_xmit, - .ndo_change_mtu = can_change_mtu, }; /* CAN-USB devices generally handle 32-bit CAN channel IDs. diff --git a/drivers/net/can/usb/ucan.c b/drivers/net/can/usb/ucan.c index 07406daf7c88..de61d9da99e3 100644 --- a/drivers/net/can/usb/ucan.c +++ b/drivers/net/can/usb/ucan.c @@ -1233,7 +1233,6 @@ static const struct net_device_ops ucan_netdev_ops = { .ndo_open = ucan_open, .ndo_stop = ucan_close, .ndo_start_xmit = ucan_start_xmit, - .ndo_change_mtu = can_change_mtu, }; static const struct ethtool_ops ucan_ethtool_ops = { diff --git a/drivers/net/can/usb/usb_8dev.c b/drivers/net/can/usb/usb_8dev.c index 8a5596ce4e46..7449328f7cd7 100644 --- a/drivers/net/can/usb/usb_8dev.c +++ b/drivers/net/can/usb/usb_8dev.c @@ -868,7 +868,6 @@ static const struct net_device_ops usb_8dev_netdev_ops = { .ndo_open = usb_8dev_open, .ndo_stop = usb_8dev_close, .ndo_start_xmit = usb_8dev_start_xmit, - .ndo_change_mtu = can_change_mtu, }; static const struct ethtool_ops usb_8dev_ethtool_ops = { diff --git a/drivers/net/can/xilinx_can.c b/drivers/net/can/xilinx_can.c index a25a3ca62c12..43d7f22820b8 100644 --- a/drivers/net/can/xilinx_can.c +++ b/drivers/net/can/xilinx_can.c @@ -1702,7 +1702,6 @@ static const struct net_device_ops xcan_netdev_ops = { .ndo_open = xcan_open, .ndo_stop = xcan_close, .ndo_start_xmit = xcan_start_xmit, - .ndo_change_mtu = can_change_mtu, }; static const struct ethtool_ops xcan_ethtool_ops = { diff --git a/include/linux/can/dev.h b/include/linux/can/dev.h index a2229a61ccde..0fe8f80f223e 100644 --- a/include/linux/can/dev.h +++ b/include/linux/can/dev.h @@ -127,7 +127,6 @@ struct can_priv *safe_candev_priv(struct net_device *dev); int open_candev(struct net_device *dev); void close_candev(struct net_device *dev); void can_set_default_mtu(struct net_device *dev); -int can_change_mtu(struct net_device *dev, int new_mtu); int __must_check can_set_static_ctrlmode(struct net_device *dev, u32 static_mode); int can_eth_ioctl_hwts(struct net_device *netdev, struct ifreq *ifr, int cmd); From 73cc2882b644e5302237cc6ee1f8885cfb387245 Mon Sep 17 00:00:00 2001 From: "Markus Schneider-Pargmann (TI.com)" Date: Wed, 1 Oct 2025 16:30:19 +0200 Subject: [PATCH 0085/1381] dt-bindings: can: m_can: Add wakeup properties The pins associated with m_can have to have a special configuration to be able to wakeup the SoC from some system states. This configuration is described in the wakeup pinctrl state while the default state describes the default configuration. Also add the sleep state which is already in use by some devicetrees. Also m_can can be a wakeup-source if capable of wakeup. Signed-off-by: Markus Schneider-Pargmann (TI.com) Reviewed-by: Dhruva Gole Reviewed-by: Rob Herring (Arm) Link: https://patch.msgid.link/20251001-topic-mcan-wakeup-source-v6-12-v10-1-4ab508ac5d1e@baylibre.com Signed-off-by: Marc Kleine-Budde --- .../bindings/net/can/bosch,m_can.yaml | 25 +++++++++++++++++++ 1 file changed, 25 insertions(+) diff --git a/Documentation/devicetree/bindings/net/can/bosch,m_can.yaml b/Documentation/devicetree/bindings/net/can/bosch,m_can.yaml index 61ef60d8f1c7..2c9d37975bed 100644 --- a/Documentation/devicetree/bindings/net/can/bosch,m_can.yaml +++ b/Documentation/devicetree/bindings/net/can/bosch,m_can.yaml @@ -109,6 +109,26 @@ properties: maximum: 32 minItems: 1 + pinctrl-0: + description: Default pinctrl state + + pinctrl-1: + description: Can be "sleep" or "wakeup" pinctrl state + + pinctrl-2: + description: Can be "sleep" or "wakeup" pinctrl state + + pinctrl-names: + description: + When present should contain at least "default" describing the default pin + states. Other states are "sleep" which describes the pinstate when + sleeping and "wakeup" describing the pins if wakeup is enabled. + minItems: 1 + items: + - const: default + - enum: [ sleep, wakeup ] + - const: wakeup + power-domains: description: Power domain provider node and an args specifier containing @@ -125,6 +145,11 @@ properties: minItems: 1 maxItems: 2 + wakeup-source: + $ref: /schemas/types.yaml#/definitions/phandle-array + description: + List of phandles to system idle states in which mcan can wakeup the system. + required: - compatible - reg From 04d5826b074e09c69895fc247ce13925bd7a7284 Mon Sep 17 00:00:00 2001 From: "Markus Schneider-Pargmann (TI.com)" Date: Wed, 1 Oct 2025 16:30:20 +0200 Subject: [PATCH 0086/1381] can: m_can: Map WoL to device_set_wakeup_enable In some devices the pins of the m_can module can act as a wakeup source. This patch helps do that by connecting the PHY_WAKE WoL option to device_set_wakeup_enable. By marking this device as being wakeup enabled, this setting can be used by platform code to decide which sleep or poweroff mode to use. Also this prepares the driver for the next patch in which the pinctrl settings are changed depending on the desired wakeup source. Reviewed-by: Vincent Mailhol Reviewed-by: Kendall Willis Signed-off-by: Markus Schneider-Pargmann (TI.com) Link: https://patch.msgid.link/20251001-topic-mcan-wakeup-source-v6-12-v10-2-4ab508ac5d1e@baylibre.com Signed-off-by: Marc Kleine-Budde --- drivers/net/can/m_can/m_can.c | 37 +++++++++++++++++++++++++++++++++++ 1 file changed, 37 insertions(+) diff --git a/drivers/net/can/m_can/m_can.c b/drivers/net/can/m_can/m_can.c index 873f5991fc5a..f00bdec3246a 100644 --- a/drivers/net/can/m_can/m_can.c +++ b/drivers/net/can/m_can/m_can.c @@ -2238,6 +2238,36 @@ static int m_can_set_coalesce(struct net_device *dev, return 0; } +static void m_can_get_wol(struct net_device *dev, struct ethtool_wolinfo *wol) +{ + struct m_can_classdev *cdev = netdev_priv(dev); + + wol->supported = device_can_wakeup(cdev->dev) ? WAKE_PHY : 0; + wol->wolopts = device_may_wakeup(cdev->dev) ? WAKE_PHY : 0; +} + +static int m_can_set_wol(struct net_device *dev, struct ethtool_wolinfo *wol) +{ + struct m_can_classdev *cdev = netdev_priv(dev); + bool wol_enable = !!(wol->wolopts & WAKE_PHY); + int ret; + + if (wol->wolopts & ~WAKE_PHY) + return -EINVAL; + + if (wol_enable == device_may_wakeup(cdev->dev)) + return 0; + + ret = device_set_wakeup_enable(cdev->dev, wol_enable); + if (ret) { + netdev_err(cdev->net, "Failed to set wakeup enable %pE\n", + ERR_PTR(ret)); + return ret; + } + + return 0; +} + static const struct ethtool_ops m_can_ethtool_ops_coalescing = { .supported_coalesce_params = ETHTOOL_COALESCE_RX_USECS_IRQ | ETHTOOL_COALESCE_RX_MAX_FRAMES_IRQ | @@ -2247,10 +2277,14 @@ static const struct ethtool_ops m_can_ethtool_ops_coalescing = { .get_ts_info = ethtool_op_get_ts_info, .get_coalesce = m_can_get_coalesce, .set_coalesce = m_can_set_coalesce, + .get_wol = m_can_get_wol, + .set_wol = m_can_set_wol, }; static const struct ethtool_ops m_can_ethtool_ops = { .get_ts_info = ethtool_op_get_ts_info, + .get_wol = m_can_get_wol, + .set_wol = m_can_set_wol, }; static int register_m_can_dev(struct m_can_classdev *cdev) @@ -2377,6 +2411,9 @@ struct m_can_classdev *m_can_class_allocate_dev(struct device *dev, goto out; } + if (dev->of_node && of_property_read_bool(dev->of_node, "wakeup-source")) + device_set_wakeup_capable(dev, true); + /* Get TX FIFO size * Defines the total amount of echo buffers for loopback */ From 148e125d4e6fa8ff03f91d9070ad87f7b4f5c610 Mon Sep 17 00:00:00 2001 From: "Markus Schneider-Pargmann (TI.com)" Date: Wed, 1 Oct 2025 16:30:21 +0200 Subject: [PATCH 0087/1381] can: m_can: Return ERR_PTR on error in allocation We have more detailed error values available, return them in the core driver and the calling drivers to return proper errors to callers. Reviewed-by: Vincent Mailhol Reviewed-by: Dhruva Gole Reviewed-by: Kendall Willis Signed-off-by: Markus Schneider-Pargmann (TI.com) Link: https://patch.msgid.link/20251001-topic-mcan-wakeup-source-v6-12-v10-3-4ab508ac5d1e@baylibre.com Signed-off-by: Marc Kleine-Budde --- drivers/net/can/m_can/m_can.c | 6 +++--- drivers/net/can/m_can/m_can_pci.c | 4 ++-- drivers/net/can/m_can/m_can_platform.c | 4 ++-- drivers/net/can/m_can/tcan4x5x-core.c | 4 ++-- 4 files changed, 9 insertions(+), 9 deletions(-) diff --git a/drivers/net/can/m_can/m_can.c b/drivers/net/can/m_can/m_can.c index f00bdec3246a..10b5862b4880 100644 --- a/drivers/net/can/m_can/m_can.c +++ b/drivers/net/can/m_can/m_can.c @@ -2408,7 +2408,7 @@ struct m_can_classdev *m_can_class_allocate_dev(struct device *dev, sizeof(mram_config_vals) / 4); if (ret) { dev_err(dev, "Could not get Message RAM configuration."); - goto out; + return ERR_PTR(ret); } if (dev->of_node && of_property_read_bool(dev->of_node, "wakeup-source")) @@ -2423,7 +2423,7 @@ struct m_can_classdev *m_can_class_allocate_dev(struct device *dev, net_dev = alloc_candev(sizeof_priv, tx_fifo_size); if (!net_dev) { dev_err(dev, "Failed to allocate CAN device"); - goto out; + return ERR_PTR(-ENOMEM); } class_dev = netdev_priv(net_dev); @@ -2433,7 +2433,7 @@ struct m_can_classdev *m_can_class_allocate_dev(struct device *dev, m_can_of_parse_mram(class_dev, mram_config_vals); spin_lock_init(&class_dev->tx_handling_spinlock); -out: + return class_dev; } EXPORT_SYMBOL_GPL(m_can_class_allocate_dev); diff --git a/drivers/net/can/m_can/m_can_pci.c b/drivers/net/can/m_can/m_can_pci.c index 9ad7419f88f8..eb31ed1f9644 100644 --- a/drivers/net/can/m_can/m_can_pci.c +++ b/drivers/net/can/m_can/m_can_pci.c @@ -111,8 +111,8 @@ static int m_can_pci_probe(struct pci_dev *pci, const struct pci_device_id *id) mcan_class = m_can_class_allocate_dev(&pci->dev, sizeof(struct m_can_pci_priv)); - if (!mcan_class) - return -ENOMEM; + if (IS_ERR(mcan_class)) + return PTR_ERR(mcan_class); priv = cdev_to_priv(mcan_class); diff --git a/drivers/net/can/m_can/m_can_platform.c b/drivers/net/can/m_can/m_can_platform.c index 4a412add2b8d..56da411878af 100644 --- a/drivers/net/can/m_can/m_can_platform.c +++ b/drivers/net/can/m_can/m_can_platform.c @@ -87,8 +87,8 @@ static int m_can_plat_probe(struct platform_device *pdev) mcan_class = m_can_class_allocate_dev(&pdev->dev, sizeof(struct m_can_plat_priv)); - if (!mcan_class) - return -ENOMEM; + if (IS_ERR(mcan_class)) + return PTR_ERR(mcan_class); priv = cdev_to_priv(mcan_class); diff --git a/drivers/net/can/m_can/tcan4x5x-core.c b/drivers/net/can/m_can/tcan4x5x-core.c index 39b0b5277b11..31cc9d0abd45 100644 --- a/drivers/net/can/m_can/tcan4x5x-core.c +++ b/drivers/net/can/m_can/tcan4x5x-core.c @@ -416,8 +416,8 @@ static int tcan4x5x_can_probe(struct spi_device *spi) mcan_class = m_can_class_allocate_dev(&spi->dev, sizeof(struct tcan4x5x_priv)); - if (!mcan_class) - return -ENOMEM; + if (IS_ERR(mcan_class)) + return PTR_ERR(mcan_class); ret = m_can_check_mram_cfg(mcan_class, TCAN4X5X_MRAM_SIZE); if (ret) From a77a29775373a91ad1ca0375e0ab1b1b294d53bb Mon Sep 17 00:00:00 2001 From: "Markus Schneider-Pargmann (TI.com)" Date: Wed, 1 Oct 2025 16:30:22 +0200 Subject: [PATCH 0088/1381] can: m_can: Support pinctrl wakeup state TI AM62x SoC requires a wakeup flag being set in pinctrl when mcan pins act as a wakeup source. Add support to select the wakeup state if WOL is enabled. Signed-off-by: Markus Schneider-Pargmann (TI.com) Link: https://patch.msgid.link/20251001-topic-mcan-wakeup-source-v6-12-v10-4-4ab508ac5d1e@baylibre.com Signed-off-by: Marc Kleine-Budde --- drivers/net/can/m_can/m_can.c | 69 ++++++++++++++++++++++++++++++++++- drivers/net/can/m_can/m_can.h | 3 ++ 2 files changed, 70 insertions(+), 2 deletions(-) diff --git a/drivers/net/can/m_can/m_can.c b/drivers/net/can/m_can/m_can.c index 10b5862b4880..8569596ae830 100644 --- a/drivers/net/can/m_can/m_can.c +++ b/drivers/net/can/m_can/m_can.c @@ -2265,7 +2265,26 @@ static int m_can_set_wol(struct net_device *dev, struct ethtool_wolinfo *wol) return ret; } + if (!IS_ERR_OR_NULL(cdev->pinctrl_state_wakeup)) { + if (wol_enable) + ret = pinctrl_select_state(cdev->pinctrl, cdev->pinctrl_state_wakeup); + else + ret = pinctrl_pm_select_default_state(cdev->dev); + + if (ret) { + netdev_err(cdev->net, "Failed to select pinctrl state %pE\n", + ERR_PTR(ret)); + goto err_wakeup_enable; + } + } + return 0; + +err_wakeup_enable: + /* Revert wakeup enable */ + device_set_wakeup_enable(cdev->dev, !wol_enable); + + return ret; } static const struct ethtool_ops m_can_ethtool_ops_coalescing = { @@ -2393,6 +2412,42 @@ int m_can_class_get_clocks(struct m_can_classdev *cdev) } EXPORT_SYMBOL_GPL(m_can_class_get_clocks); +static bool m_can_class_wakeup_pinctrl_enabled(struct m_can_classdev *class_dev) +{ + return device_may_wakeup(class_dev->dev) && class_dev->pinctrl_state_wakeup; +} + +static int m_can_class_parse_pinctrl(struct m_can_classdev *class_dev) +{ + struct device *dev = class_dev->dev; + int ret; + + class_dev->pinctrl = devm_pinctrl_get(dev); + if (IS_ERR(class_dev->pinctrl)) { + ret = PTR_ERR(class_dev->pinctrl); + class_dev->pinctrl = NULL; + + if (ret == -ENODEV) + return 0; + + return dev_err_probe(dev, ret, "Failed to get pinctrl\n"); + } + + class_dev->pinctrl_state_wakeup = + pinctrl_lookup_state(class_dev->pinctrl, "wakeup"); + if (IS_ERR(class_dev->pinctrl_state_wakeup)) { + ret = PTR_ERR(class_dev->pinctrl_state_wakeup); + class_dev->pinctrl_state_wakeup = NULL; + + if (ret == -ENODEV) + return 0; + + return dev_err_probe(dev, ret, "Failed to lookup pinctrl wakeup state\n"); + } + + return 0; +} + struct m_can_classdev *m_can_class_allocate_dev(struct device *dev, int sizeof_priv) { @@ -2434,7 +2489,15 @@ struct m_can_classdev *m_can_class_allocate_dev(struct device *dev, m_can_of_parse_mram(class_dev, mram_config_vals); spin_lock_init(&class_dev->tx_handling_spinlock); + ret = m_can_class_parse_pinctrl(class_dev); + if (ret) + goto err_free_candev; + return class_dev; + +err_free_candev: + free_candev(net_dev); + return ERR_PTR(ret); } EXPORT_SYMBOL_GPL(m_can_class_allocate_dev); @@ -2563,7 +2626,8 @@ int m_can_class_suspend(struct device *dev) cdev->can.state = CAN_STATE_SLEEPING; } - pinctrl_pm_select_sleep_state(dev); + if (!m_can_class_wakeup_pinctrl_enabled(cdev)) + pinctrl_pm_select_sleep_state(dev); return ret; } @@ -2575,7 +2639,8 @@ int m_can_class_resume(struct device *dev) struct net_device *ndev = cdev->net; int ret = 0; - pinctrl_pm_select_default_state(dev); + if (!m_can_class_wakeup_pinctrl_enabled(cdev)) + pinctrl_pm_select_default_state(dev); if (netif_running(ndev)) { ret = m_can_clk_start(cdev); diff --git a/drivers/net/can/m_can/m_can.h b/drivers/net/can/m_can/m_can.h index 7b7600697c6b..f2f89687bbd2 100644 --- a/drivers/net/can/m_can/m_can.h +++ b/drivers/net/can/m_can/m_can.h @@ -129,6 +129,9 @@ struct m_can_classdev { struct mram_cfg mcfg[MRAM_CFG_NUM]; struct hrtimer hrtimer; + + struct pinctrl *pinctrl; + struct pinctrl_state *pinctrl_state_wakeup; }; struct m_can_classdev *m_can_class_allocate_dev(struct device *dev, int sizeof_priv); From c6dcc2b321cccf1fc2a18480bb547956fdb2b2e0 Mon Sep 17 00:00:00 2001 From: Marc Kleine-Budde Date: Fri, 8 Aug 2025 15:29:34 +0200 Subject: [PATCH 0089/1381] can: m_can: m_can_init_ram(): make static Since commit eaacfeaca7ad ("can: m_can: Call the RAM init directly from m_can_chip_config") m_can_init_ram() is not used outside of m_can.c. Mark as static and remove the EXPORT_SYMBOL_GPL(). Link: https://patch.msgid.link/20251008-m_can-cleanups-v1-1-1784a18eaa84@pengutronix.de Signed-off-by: Marc Kleine-Budde --- drivers/net/can/m_can/m_can.c | 43 +++++++++++++++++------------------ drivers/net/can/m_can/m_can.h | 1 - 2 files changed, 21 insertions(+), 23 deletions(-) diff --git a/drivers/net/can/m_can/m_can.c b/drivers/net/can/m_can/m_can.c index 8569596ae830..9f4002f3481e 100644 --- a/drivers/net/can/m_can/m_can.c +++ b/drivers/net/can/m_can/m_can.c @@ -1380,6 +1380,27 @@ static const struct can_bittiming_const m_can_data_bittiming_const_31X = { .brp_inc = 1, }; +static int m_can_init_ram(struct m_can_classdev *cdev) +{ + int end, i, start; + int err = 0; + + /* initialize the entire Message RAM in use to avoid possible + * ECC/parity checksum errors when reading an uninitialized buffer + */ + start = cdev->mcfg[MRAM_SIDF].off; + end = cdev->mcfg[MRAM_TXB].off + + cdev->mcfg[MRAM_TXB].num * TXB_ELEMENT_SIZE; + + for (i = start; i < end; i += 4) { + err = m_can_fifo_write_no_off(cdev, i, 0x0); + if (err) + break; + } + + return err; +} + static int m_can_set_bittiming(struct net_device *dev) { struct m_can_classdev *cdev = netdev_priv(dev); @@ -2374,28 +2395,6 @@ static void m_can_of_parse_mram(struct m_can_classdev *cdev, cdev->mcfg[MRAM_TXB].off, cdev->mcfg[MRAM_TXB].num); } -int m_can_init_ram(struct m_can_classdev *cdev) -{ - int end, i, start; - int err = 0; - - /* initialize the entire Message RAM in use to avoid possible - * ECC/parity checksum errors when reading an uninitialized buffer - */ - start = cdev->mcfg[MRAM_SIDF].off; - end = cdev->mcfg[MRAM_TXB].off + - cdev->mcfg[MRAM_TXB].num * TXB_ELEMENT_SIZE; - - for (i = start; i < end; i += 4) { - err = m_can_fifo_write_no_off(cdev, i, 0x0); - if (err) - break; - } - - return err; -} -EXPORT_SYMBOL_GPL(m_can_init_ram); - int m_can_class_get_clocks(struct m_can_classdev *cdev) { int ret = 0; diff --git a/drivers/net/can/m_can/m_can.h b/drivers/net/can/m_can/m_can.h index f2f89687bbd2..4743342b2fba 100644 --- a/drivers/net/can/m_can/m_can.h +++ b/drivers/net/can/m_can/m_can.h @@ -139,7 +139,6 @@ void m_can_class_free_dev(struct net_device *net); int m_can_class_register(struct m_can_classdev *cdev); void m_can_class_unregister(struct m_can_classdev *cdev); int m_can_class_get_clocks(struct m_can_classdev *cdev); -int m_can_init_ram(struct m_can_classdev *priv); int m_can_check_mram_cfg(struct m_can_classdev *cdev, u32 mram_max_size); int m_can_class_suspend(struct device *dev); From 60af9dbb63fb077cc71496cf5ada82261be0bb25 Mon Sep 17 00:00:00 2001 From: Marc Kleine-Budde Date: Thu, 7 Aug 2025 20:14:55 +0200 Subject: [PATCH 0090/1381] can: m_can: hrtimer_callback(): rename to m_can_polling_timer() The original use of struct m_can_classdev::hrtimer was to support polling for devices without IRQ, with the timer function called hrtimer_callback(). Commit 07f25091ca02 ("can: m_can: Implement receive coalescing") uses the hrtimer for software-supported IRQ coalescence, with the timer function called m_can_coalescing_timer(). To improve the readability of the driver, rename hrtimer_callback() to m_can_polling_timer(), which better describes the functionality. Link: https://patch.msgid.link/20251008-m_can-cleanups-v1-2-1784a18eaa84@pengutronix.de Signed-off-by: Marc Kleine-Budde --- drivers/net/can/m_can/m_can.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/can/m_can/m_can.c b/drivers/net/can/m_can/m_can.c index 9f4002f3481e..110cfd54b669 100644 --- a/drivers/net/can/m_can/m_can.c +++ b/drivers/net/can/m_can/m_can.c @@ -2058,7 +2058,7 @@ static netdev_tx_t m_can_start_xmit(struct sk_buff *skb, return ret; } -static enum hrtimer_restart hrtimer_callback(struct hrtimer *timer) +static enum hrtimer_restart m_can_polling_timer(struct hrtimer *timer) { struct m_can_classdev *cdev = container_of(timer, struct m_can_classdev, hrtimer); @@ -2545,7 +2545,7 @@ int m_can_class_register(struct m_can_classdev *cdev) if (!cdev->net->irq) { dev_dbg(cdev->dev, "Polling enabled, initialize hrtimer"); - hrtimer_setup(&cdev->hrtimer, &hrtimer_callback, CLOCK_MONOTONIC, + hrtimer_setup(&cdev->hrtimer, m_can_polling_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL_PINNED); } else { hrtimer_setup(&cdev->hrtimer, m_can_coalescing_timer, CLOCK_MONOTONIC, From 293735053eaa7a15e0debacdcbe3cdcd006436be Mon Sep 17 00:00:00 2001 From: Marc Kleine-Budde Date: Fri, 8 Aug 2025 14:14:55 +0200 Subject: [PATCH 0091/1381] net: m_can: convert dev_{dbg,info,err} -> netdev_{dbg,info,err} To ease debugging use the netdev_{dbg,info,err}() functions instead of dev_{dbg,info,err}. Link: https://patch.msgid.link/20251008-m_can-cleanups-v1-3-1784a18eaa84@pengutronix.de Signed-off-by: Marc Kleine-Budde --- drivers/net/can/m_can/m_can.c | 54 +++++++++++++++++------------------ 1 file changed, 27 insertions(+), 27 deletions(-) diff --git a/drivers/net/can/m_can/m_can.c b/drivers/net/can/m_can/m_can.c index 110cfd54b669..6aef5e771fc3 100644 --- a/drivers/net/can/m_can/m_can.c +++ b/drivers/net/can/m_can/m_can.c @@ -387,8 +387,8 @@ static int m_can_cccr_update_bits(struct m_can_classdev *cdev, u32 mask, u32 val size_t tries = 10; if (!(mask & CCCR_INIT) && !(val_before & CCCR_INIT)) { - dev_err(cdev->dev, - "refusing to configure device when in normal mode\n"); + netdev_err(cdev->net, + "refusing to configure device when in normal mode\n"); return -EBUSY; } @@ -470,7 +470,7 @@ static void m_can_coalescing_disable(struct m_can_classdev *cdev) static inline void m_can_enable_all_interrupts(struct m_can_classdev *cdev) { if (!cdev->net->irq) { - dev_dbg(cdev->dev, "Start hrtimer\n"); + netdev_dbg(cdev->net, "Start hrtimer\n"); hrtimer_start(&cdev->hrtimer, ms_to_ktime(HRTIMER_POLL_INTERVAL_MS), HRTIMER_MODE_REL_PINNED); @@ -486,7 +486,7 @@ static inline void m_can_disable_all_interrupts(struct m_can_classdev *cdev) m_can_write(cdev, M_CAN_ILE, 0x0); if (!cdev->net->irq) { - dev_dbg(cdev->dev, "Stop hrtimer\n"); + netdev_dbg(cdev->net, "Stop hrtimer\n"); hrtimer_try_to_cancel(&cdev->hrtimer); } } @@ -1486,7 +1486,7 @@ static int m_can_chip_config(struct net_device *dev) err = m_can_init_ram(cdev); if (err) { - dev_err(cdev->dev, "Message RAM configuration failed\n"); + netdev_err(dev, "Message RAM configuration failed\n"); return err; } @@ -1716,7 +1716,7 @@ static int m_can_niso_supported(struct m_can_classdev *cdev) /* Then clear the it again. */ ret = m_can_cccr_update_bits(cdev, CCCR_NISO, 0); if (ret) { - dev_err(cdev->dev, "failed to revert the NON-ISO bit in CCCR\n"); + netdev_err(cdev->net, "failed to revert the NON-ISO bit in CCCR\n"); return ret; } @@ -1735,8 +1735,8 @@ static int m_can_dev_setup(struct m_can_classdev *cdev) m_can_version = m_can_check_core_release(cdev); /* return if unsupported version */ if (!m_can_version) { - dev_err(cdev->dev, "Unsupported version number: %2d", - m_can_version); + netdev_err(cdev->net, "Unsupported version number: %2d", + m_can_version); return -EINVAL; } @@ -1794,8 +1794,8 @@ static int m_can_dev_setup(struct m_can_classdev *cdev) cdev->can.ctrlmode_supported |= CAN_CTRLMODE_FD_NON_ISO; break; default: - dev_err(cdev->dev, "Unsupported version number: %2d", - cdev->version); + netdev_err(cdev->net, "Unsupported version number: %2d", + cdev->version); return -EINVAL; } @@ -2348,8 +2348,8 @@ int m_can_check_mram_cfg(struct m_can_classdev *cdev, u32 mram_max_size) total_size = cdev->mcfg[MRAM_TXB].off - cdev->mcfg[MRAM_SIDF].off + cdev->mcfg[MRAM_TXB].num * TXB_ELEMENT_SIZE; if (total_size > mram_max_size) { - dev_err(cdev->dev, "Total size of mram config(%u) exceeds mram(%u)\n", - total_size, mram_max_size); + netdev_err(cdev->net, "Total size of mram config(%u) exceeds mram(%u)\n", + total_size, mram_max_size); return -EINVAL; } @@ -2384,15 +2384,15 @@ static void m_can_of_parse_mram(struct m_can_classdev *cdev, cdev->mcfg[MRAM_TXB].num = mram_config_vals[7] & FIELD_MAX(TXBC_NDTB_MASK); - dev_dbg(cdev->dev, - "sidf 0x%x %d xidf 0x%x %d rxf0 0x%x %d rxf1 0x%x %d rxb 0x%x %d txe 0x%x %d txb 0x%x %d\n", - cdev->mcfg[MRAM_SIDF].off, cdev->mcfg[MRAM_SIDF].num, - cdev->mcfg[MRAM_XIDF].off, cdev->mcfg[MRAM_XIDF].num, - cdev->mcfg[MRAM_RXF0].off, cdev->mcfg[MRAM_RXF0].num, - cdev->mcfg[MRAM_RXF1].off, cdev->mcfg[MRAM_RXF1].num, - cdev->mcfg[MRAM_RXB].off, cdev->mcfg[MRAM_RXB].num, - cdev->mcfg[MRAM_TXE].off, cdev->mcfg[MRAM_TXE].num, - cdev->mcfg[MRAM_TXB].off, cdev->mcfg[MRAM_TXB].num); + netdev_dbg(cdev->net, + "sidf 0x%x %d xidf 0x%x %d rxf0 0x%x %d rxf1 0x%x %d rxb 0x%x %d txe 0x%x %d txb 0x%x %d\n", + cdev->mcfg[MRAM_SIDF].off, cdev->mcfg[MRAM_SIDF].num, + cdev->mcfg[MRAM_XIDF].off, cdev->mcfg[MRAM_XIDF].num, + cdev->mcfg[MRAM_RXF0].off, cdev->mcfg[MRAM_RXF0].num, + cdev->mcfg[MRAM_RXF1].off, cdev->mcfg[MRAM_RXF1].num, + cdev->mcfg[MRAM_RXB].off, cdev->mcfg[MRAM_RXB].num, + cdev->mcfg[MRAM_TXE].off, cdev->mcfg[MRAM_TXE].num, + cdev->mcfg[MRAM_TXB].off, cdev->mcfg[MRAM_TXB].num); } int m_can_class_get_clocks(struct m_can_classdev *cdev) @@ -2403,7 +2403,7 @@ int m_can_class_get_clocks(struct m_can_classdev *cdev) cdev->cclk = devm_clk_get(cdev->dev, "cclk"); if (IS_ERR(cdev->hclk) || IS_ERR(cdev->cclk)) { - dev_err(cdev->dev, "no clock found\n"); + netdev_err(cdev->net, "no clock found\n"); ret = -ENODEV; } @@ -2544,7 +2544,7 @@ int m_can_class_register(struct m_can_classdev *cdev) } if (!cdev->net->irq) { - dev_dbg(cdev->dev, "Polling enabled, initialize hrtimer"); + netdev_dbg(cdev->net, "Polling enabled, initialize hrtimer"); hrtimer_setup(&cdev->hrtimer, m_can_polling_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL_PINNED); } else { @@ -2558,15 +2558,15 @@ int m_can_class_register(struct m_can_classdev *cdev) ret = register_m_can_dev(cdev); if (ret) { - dev_err(cdev->dev, "registering %s failed (err=%d)\n", - cdev->net->name, ret); + netdev_err(cdev->net, "registering %s failed (err=%d)\n", + cdev->net->name, ret); goto rx_offload_del; } of_can_transceiver(cdev->net); - dev_info(cdev->dev, "%s device registered (irq=%d, version=%d)\n", - KBUILD_MODNAME, cdev->net->irq, cdev->version); + netdev_info(cdev->net, "device registered (irq=%d, version=%d)\n", + cdev->net->irq, cdev->version); /* Probe finished * Assert reset and stop clocks. From c6cbd24f65f1bbdae15cf8450a8298255c231e2e Mon Sep 17 00:00:00 2001 From: Marc Kleine-Budde Date: Thu, 7 Aug 2025 16:27:45 +0200 Subject: [PATCH 0092/1381] can: m_can: m_can_interrupt_enable(): use m_can_write() instead of open coding it As everywhere else in the driver, use m_can_write() instead of open coding it. Link: https://patch.msgid.link/20251008-m_can-cleanups-v1-4-1784a18eaa84@pengutronix.de Signed-off-by: Marc Kleine-Budde --- drivers/net/can/m_can/m_can.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/can/m_can/m_can.c b/drivers/net/can/m_can/m_can.c index 6aef5e771fc3..98e7ab612bba 100644 --- a/drivers/net/can/m_can/m_can.c +++ b/drivers/net/can/m_can/m_can.c @@ -452,7 +452,7 @@ static void m_can_interrupt_enable(struct m_can_classdev *cdev, u32 interrupts) { if (cdev->active_interrupts == interrupts) return; - cdev->ops->write_reg(cdev, M_CAN_IE, interrupts); + m_can_write(cdev, M_CAN_IE, interrupts); cdev->active_interrupts = interrupts; } From 6218391758b53e6cadf2aa80093457eeb485f06e Mon Sep 17 00:00:00 2001 From: Marc Kleine-Budde Date: Thu, 7 Aug 2025 17:19:31 +0200 Subject: [PATCH 0093/1381] can: m_can: m_can_class_register(): remove error message in case devm_kzalloc() fails If devm_kzalloc() fails, it already outputs an error message. Remove the error message from m_can_class_register() accordingly. Link: https://patch.msgid.link/20251008-m_can-cleanups-v1-5-1784a18eaa84@pengutronix.de Signed-off-by: Marc Kleine-Budde --- drivers/net/can/m_can/m_can.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/drivers/net/can/m_can/m_can.c b/drivers/net/can/m_can/m_can.c index 98e7ab612bba..8013e8835027 100644 --- a/drivers/net/can/m_can/m_can.c +++ b/drivers/net/can/m_can/m_can.c @@ -2517,10 +2517,8 @@ int m_can_class_register(struct m_can_classdev *cdev) devm_kzalloc(cdev->dev, cdev->tx_fifo_size * sizeof(*cdev->tx_ops), GFP_KERNEL); - if (!cdev->tx_ops) { - dev_err(cdev->dev, "Failed to allocate tx_ops for workqueue\n"); + if (!cdev->tx_ops) return -ENOMEM; - } } cdev->rst = devm_reset_control_get_optional_shared(cdev->dev, NULL); From b24b43522eb33c32ca2348354e2555685c8c4644 Mon Sep 17 00:00:00 2001 From: Marc Kleine-Budde Date: Thu, 7 Aug 2025 17:20:08 +0200 Subject: [PATCH 0094/1381] can: m_can: m_can_tx_submit(): remove unneeded sanity checks m_can_tx_submit() is only called for peripheral devices. So remove the sanity check. Link: https://patch.msgid.link/20251008-m_can-cleanups-v1-6-1784a18eaa84@pengutronix.de Signed-off-by: Marc Kleine-Budde --- drivers/net/can/m_can/m_can.c | 5 ----- 1 file changed, 5 deletions(-) diff --git a/drivers/net/can/m_can/m_can.c b/drivers/net/can/m_can/m_can.c index 8013e8835027..713d3df296df 100644 --- a/drivers/net/can/m_can/m_can.c +++ b/drivers/net/can/m_can/m_can.c @@ -1973,11 +1973,6 @@ static netdev_tx_t m_can_tx_handler(struct m_can_classdev *cdev, static void m_can_tx_submit(struct m_can_classdev *cdev) { - if (cdev->version == 30) - return; - if (!cdev->is_peripheral) - return; - m_can_write(cdev, M_CAN_TXBAR, cdev->tx_peripheral_submit); cdev->tx_peripheral_submit = 0; } From 91a55c72a821d106a588af4c28aec129dcbea5af Mon Sep 17 00:00:00 2001 From: Marc Kleine-Budde Date: Wed, 6 Aug 2025 12:20:44 +0200 Subject: [PATCH 0095/1381] can: m_can: m_can_get_berr_counter(): don't wake up controller if interface is down If the interface is down, the CAN controller might be powered down, the clock disabled, and/or it's external reset asserted. Don't wake up the controller to read the CAN bus error counters, if the interface is down. Reviewed-by: Markus Schneider-Pargmann Link: https://patch.msgid.link/20251008-m_can-cleanups-v1-7-1784a18eaa84@pengutronix.de Signed-off-by: Marc Kleine-Budde --- drivers/net/can/m_can/m_can.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/net/can/m_can/m_can.c b/drivers/net/can/m_can/m_can.c index 713d3df296df..eb856547ae7d 100644 --- a/drivers/net/can/m_can/m_can.c +++ b/drivers/net/can/m_can/m_can.c @@ -791,6 +791,10 @@ static int m_can_get_berr_counter(const struct net_device *dev, struct m_can_classdev *cdev = netdev_priv(dev); int err; + /* Avoid waking up the controller if the interface is down */ + if (!(dev->flags & IFF_UP)) + return 0; + err = m_can_clk_start(cdev); if (err) return err; From e5ae07b2ef86ee5bcb90a1c933c9d5ac1a0e33be Mon Sep 17 00:00:00 2001 From: Simon Wunderlich Date: Sun, 28 Sep 2025 20:28:19 +0200 Subject: [PATCH 0096/1381] batman-adv: Start new development cycle This version will contain all the (major or even only minor) changes for Linux 6.19. The version number isn't a semantic version number with major and minor information. It is just encoding the year of the expected publishing as Linux -rc1 and the number of published versions this year (starting at 0). Signed-off-by: Simon Wunderlich --- net/batman-adv/main.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h index 2be1ac17acaa..af230b017bc1 100644 --- a/net/batman-adv/main.h +++ b/net/batman-adv/main.h @@ -13,7 +13,7 @@ #define BATADV_DRIVER_DEVICE "batman-adv" #ifndef BATADV_SOURCE_VERSION -#define BATADV_SOURCE_VERSION "2025.4" +#define BATADV_SOURCE_VERSION "2025.5" #endif /* B.A.T.M.A.N. parameters */ From ed5730f3f733659a4a023a5f1e767365fe341648 Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Sun, 28 Sep 2025 10:29:32 +0200 Subject: [PATCH 0097/1381] batman-adv: use skb_crc32c() instead of skb_seq_read() Make batadv_bla_check_duplist() just use the new function skb_crc32c(), instead of calling skb_seq_read() with crc32c(). This is faster and simpler. Suggested-by: Eric Biggers Signed-off-by: Sven Eckelmann Signed-off-by: Simon Wunderlich --- net/batman-adv/Kconfig | 1 + net/batman-adv/bridge_loop_avoidance.c | 51 ++++---------------------- net/batman-adv/types.h | 2 +- 3 files changed, 10 insertions(+), 44 deletions(-) diff --git a/net/batman-adv/Kconfig b/net/batman-adv/Kconfig index c299e2bc87ed..58c408b7a7d9 100644 --- a/net/batman-adv/Kconfig +++ b/net/batman-adv/Kconfig @@ -35,6 +35,7 @@ config BATMAN_ADV_BLA bool "Bridge Loop Avoidance" depends on BATMAN_ADV && INET select CRC16 + select NET_CRC32C default y help This option enables BLA (Bridge Loop Avoidance), a mechanism diff --git a/net/batman-adv/bridge_loop_avoidance.c b/net/batman-adv/bridge_loop_avoidance.c index b992ba12aa24..3dc791c15bf7 100644 --- a/net/batman-adv/bridge_loop_avoidance.c +++ b/net/batman-adv/bridge_loop_avoidance.c @@ -12,7 +12,6 @@ #include #include #include -#include #include #include #include @@ -1585,45 +1584,11 @@ int batadv_bla_init(struct batadv_priv *bat_priv) return 0; } -/** - * batadv_skb_crc32() - calculate CRC32 of the whole packet and skip bytes in - * the header - * @skb: skb pointing to fragmented socket buffers - * @payload_ptr: Pointer to position inside the head buffer of the skb - * marking the start of the data to be CRC'ed - * - * payload_ptr must always point to an address in the skb head buffer and not to - * a fragment. - * - * Return: big endian crc32c of the checksummed data - */ -static __be32 batadv_skb_crc32(struct sk_buff *skb, u8 *payload_ptr) -{ - unsigned int to = skb->len; - unsigned int consumed = 0; - struct skb_seq_state st; - unsigned int from; - unsigned int len; - const u8 *data; - u32 crc = 0; - - from = (unsigned int)(payload_ptr - skb->data); - - skb_prepare_seq_read(skb, from, to, &st); - while ((len = skb_seq_read(consumed, &data, &st)) != 0) { - crc = crc32c(crc, data, len); - consumed += len; - } - - return htonl(crc); -} - /** * batadv_bla_check_duplist() - Check if a frame is in the broadcast dup. * @bat_priv: the bat priv with all the mesh interface information * @skb: contains the multicast packet to be checked - * @payload_ptr: pointer to position inside the head buffer of the skb - * marking the start of the data to be CRC'ed + * @payload_offset: offset in the skb, marking the start of the data to be CRC'ed * @orig: originator mac address, NULL if unknown * * Check if it is on our broadcast list. Another gateway might have sent the @@ -1638,16 +1603,18 @@ static __be32 batadv_skb_crc32(struct sk_buff *skb, u8 *payload_ptr) * Return: true if a packet is in the duplicate list, false otherwise. */ static bool batadv_bla_check_duplist(struct batadv_priv *bat_priv, - struct sk_buff *skb, u8 *payload_ptr, + struct sk_buff *skb, int payload_offset, const u8 *orig) { struct batadv_bcast_duplist_entry *entry; bool ret = false; + int payload_len; int i, curr; - __be32 crc; + u32 crc; /* calculate the crc ... */ - crc = batadv_skb_crc32(skb, payload_ptr); + payload_len = skb->len - payload_offset; + crc = skb_crc32c(skb, payload_offset, payload_len, 0); spin_lock_bh(&bat_priv->bla.bcast_duplist_lock); @@ -1727,7 +1694,7 @@ static bool batadv_bla_check_duplist(struct batadv_priv *bat_priv, static bool batadv_bla_check_ucast_duplist(struct batadv_priv *bat_priv, struct sk_buff *skb) { - return batadv_bla_check_duplist(bat_priv, skb, (u8 *)skb->data, NULL); + return batadv_bla_check_duplist(bat_priv, skb, 0, NULL); } /** @@ -1745,12 +1712,10 @@ bool batadv_bla_check_bcast_duplist(struct batadv_priv *bat_priv, struct sk_buff *skb) { struct batadv_bcast_packet *bcast_packet; - u8 *payload_ptr; bcast_packet = (struct batadv_bcast_packet *)skb->data; - payload_ptr = (u8 *)(bcast_packet + 1); - return batadv_bla_check_duplist(bat_priv, skb, payload_ptr, + return batadv_bla_check_duplist(bat_priv, skb, sizeof(*bcast_packet), bcast_packet->orig); } diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h index ae1d7a8dc480..8fc5fe0e9b05 100644 --- a/net/batman-adv/types.h +++ b/net/batman-adv/types.h @@ -734,7 +734,7 @@ struct batadv_bcast_duplist_entry { u8 orig[ETH_ALEN]; /** @crc: crc32 checksum of broadcast payload */ - __be32 crc; + u32 crc; /** @entrytime: time when the broadcast packet was received */ unsigned long entrytime; From 0746da01767e8a0df97ae5d031d852e932e03682 Mon Sep 17 00:00:00 2001 From: Jijie Shao Date: Tue, 14 Oct 2025 21:40:18 +0800 Subject: [PATCH 0098/1381] net: hibmcge: support pci_driver.shutdown() support pci_driver.shutdown() for hibmcge driver. Signed-off-by: Jijie Shao Link: https://patch.msgid.link/20251014134018.1178385-1-shaojijie@huawei.com Signed-off-by: Jakub Kicinski --- .../net/ethernet/hisilicon/hibmcge/hbg_main.c | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) diff --git a/drivers/net/ethernet/hisilicon/hibmcge/hbg_main.c b/drivers/net/ethernet/hisilicon/hibmcge/hbg_main.c index 0b92a2e5e986..068da2fd1fea 100644 --- a/drivers/net/ethernet/hisilicon/hibmcge/hbg_main.c +++ b/drivers/net/ethernet/hisilicon/hibmcge/hbg_main.c @@ -472,6 +472,22 @@ static int hbg_probe(struct pci_dev *pdev, const struct pci_device_id *ent) return 0; } +static void hbg_shutdown(struct pci_dev *pdev) +{ + struct net_device *netdev = pci_get_drvdata(pdev); + + rtnl_lock(); + if (netif_running(netdev)) + dev_close(netdev); + rtnl_unlock(); + + pci_disable_device(pdev); + pci_set_drvdata(pdev, NULL); + + if (system_state == SYSTEM_POWER_OFF) + pci_set_power_state(pdev, PCI_D3hot); +} + static const struct pci_device_id hbg_pci_tbl[] = { {PCI_VDEVICE(HUAWEI, 0x3730), 0}, { } @@ -482,6 +498,7 @@ static struct pci_driver hbg_driver = { .name = "hibmcge", .id_table = hbg_pci_tbl, .probe = hbg_probe, + .shutdown = hbg_shutdown, }; static int __init hbg_module_init(void) From 1c17f4373d4db1e1f0ebd3ddcd8e7a642927a826 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Tue, 14 Oct 2025 22:42:07 +0000 Subject: [PATCH 0099/1381] ipv6: Move ipv6_fl_list from ipv6_pinfo to inet_sock. In {tcp6,udp6,raw6}_sock, struct ipv6_pinfo is always placed at the beginning of a new cache line because 1. __alignof__(struct tcp_sock) is 64 due to ____cacheline_aligned of __cacheline_group_begin(tcp_sock_write_tx) 2. __alignof__(struct udp_sock) is 64 due to ____cacheline_aligned of struct numa_drop_counters 3. in raw6_sock, struct numa_drop_counters is placed before struct ipv6_pinfo . struct ipv6_pinfo is 136 bytes, but the last cache line is only used by ipv6_fl_list: $ pahole -C ipv6_pinfo vmlinux struct ipv6_pinfo { ... /* --- cacheline 2 boundary (128 bytes) --- */ struct ipv6_fl_socklist * ipv6_fl_list; /* 128 8 */ /* size: 136, cachelines: 3, members: 23 */ Let's move ipv6_fl_list from struct ipv6_pinfo to struct inet_sock to save a full cache line for {tcp6,udp6,raw6}_sock. Now, struct ipv6_pinfo is 128 bytes, and {tcp6,udp6,raw6}_sock have 64 bytes less, while {tcp,udp,raw}_sock retain the same size. Before: # grep -E "^(RAW|UDP[^L\-]|TCP)" /proc/slabinfo | awk '{print $1, "\t", $4}' RAWv6 1408 UDPv6 1472 TCPv6 2560 RAW 1152 UDP 1280 TCP 2368 After: # grep -E "^(RAW|UDP[^L\-]|TCP)" /proc/slabinfo | awk '{print $1, "\t", $4}' RAWv6 1344 UDPv6 1408 TCPv6 2496 RAW 1152 UDP 1280 TCP 2368 Also, ipv6_fl_list and inet_flags (SNDFLOW bit) are placed in the same cache line. $ pahole -C inet_sock vmlinux ... /* --- cacheline 11 boundary (704 bytes) was 56 bytes ago --- */ struct ipv6_pinfo * pinet6; /* 760 8 */ /* --- cacheline 12 boundary (768 bytes) --- */ struct ipv6_fl_socklist * ipv6_fl_list; /* 768 8 */ unsigned long inet_flags; /* 776 8 */ Doc churn is due to the insufficient Type column (only 1 space short). Suggested-by: Eric Dumazet Signed-off-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20251014224210.2964778-1-kuniyu@google.com Signed-off-by: Jakub Kicinski --- .../networking/net_cachelines/inet_sock.rst | 79 ++++++++++--------- .../chelsio/inline_crypto/chtls/chtls_cm.c | 4 +- include/linux/ipv6.h | 1 - include/net/inet_sock.h | 1 + net/ipv6/ip6_flowlabel.c | 44 +++++------ net/ipv6/tcp_ipv6.c | 13 +-- net/sctp/ipv6.c | 8 +- 7 files changed, 76 insertions(+), 74 deletions(-) diff --git a/Documentation/networking/net_cachelines/inet_sock.rst b/Documentation/networking/net_cachelines/inet_sock.rst index b11bf48fa2b3..4c72a28a7012 100644 --- a/Documentation/networking/net_cachelines/inet_sock.rst +++ b/Documentation/networking/net_cachelines/inet_sock.rst @@ -5,42 +5,43 @@ inet_sock struct fast path usage breakdown ========================================== -======================= ===================== =================== =================== ====================================================================================================== -Type Name fastpath_tx_access fastpath_rx_access comment -======================= ===================== =================== =================== ====================================================================================================== -struct sock sk read_mostly read_mostly tcp_init_buffer_space,tcp_init_transfer,tcp_finish_connect,tcp_connect,tcp_send_rcvq,tcp_send_syn_data -struct ipv6_pinfo* pinet6 -be16 inet_sport read_mostly __tcp_transmit_skb -be32 inet_daddr read_mostly ip_select_ident_segs -be32 inet_rcv_saddr -be16 inet_dport read_mostly __tcp_transmit_skb -u16 inet_num -be32 inet_saddr -s16 uc_ttl read_mostly __ip_queue_xmit/ip_select_ttl -u16 cmsg_flags -struct ip_options_rcu* inet_opt read_mostly __ip_queue_xmit -u16 inet_id read_mostly ip_select_ident_segs -u8 tos read_mostly ip_queue_xmit -u8 min_ttl -u8 mc_ttl -u8 pmtudisc -u8:1 recverr -u8:1 is_icsk -u8:1 freebind -u8:1 hdrincl -u8:1 mc_loop -u8:1 transparent -u8:1 mc_all -u8:1 nodefrag -u8:1 bind_address_no_port -u8:1 recverr_rfc4884 -u8:1 defer_connect read_mostly tcp_sendmsg_fastopen -u8 rcv_tos -u8 convert_csum -int uc_index -int mc_index -be32 mc_addr -struct ip_mc_socklist* mc_list -struct inet_cork_full cork read_mostly __tcp_transmit_skb -struct local_port_range -======================= ===================== =================== =================== ====================================================================================================== +======================== ===================== =================== =================== ====================================================================================================== +Type Name fastpath_tx_access fastpath_rx_access comment +======================== ===================== =================== =================== ====================================================================================================== +struct sock sk read_mostly read_mostly tcp_init_buffer_space,tcp_init_transfer,tcp_finish_connect,tcp_connect,tcp_send_rcvq,tcp_send_syn_data +struct ipv6_pinfo* pinet6 +struct ipv6_fl_socklist* ipv6_fl_list read_mostly tcp_v6_connect,__ip6_datagram_connect,udpv6_sendmsg,rawv6_sendmsg +be16 inet_sport read_mostly __tcp_transmit_skb +be32 inet_daddr read_mostly ip_select_ident_segs +be32 inet_rcv_saddr +be16 inet_dport read_mostly __tcp_transmit_skb +u16 inet_num +be32 inet_saddr +s16 uc_ttl read_mostly __ip_queue_xmit/ip_select_ttl +u16 cmsg_flags +struct ip_options_rcu* inet_opt read_mostly __ip_queue_xmit +u16 inet_id read_mostly ip_select_ident_segs +u8 tos read_mostly ip_queue_xmit +u8 min_ttl +u8 mc_ttl +u8 pmtudisc +u8:1 recverr +u8:1 is_icsk +u8:1 freebind +u8:1 hdrincl +u8:1 mc_loop +u8:1 transparent +u8:1 mc_all +u8:1 nodefrag +u8:1 bind_address_no_port +u8:1 recverr_rfc4884 +u8:1 defer_connect read_mostly tcp_sendmsg_fastopen +u8 rcv_tos +u8 convert_csum +int uc_index +int mc_index +be32 mc_addr +struct ip_mc_socklist* mc_list +struct inet_cork_full cork read_mostly __tcp_transmit_skb +struct local_port_range +======================== ===================== =================== =================== ====================================================================================================== diff --git a/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_cm.c b/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_cm.c index 4ee970f3bad6..ee0154337a9c 100644 --- a/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_cm.c +++ b/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_cm.c @@ -1199,12 +1199,12 @@ static struct sock *chtls_recv_sock(struct sock *lsk, struct ipv6_pinfo *newnp = inet6_sk(newsk); struct ipv6_pinfo *np = inet6_sk(lsk); - inet_sk(newsk)->pinet6 = &newtcp6sk->inet6; + newinet->pinet6 = &newtcp6sk->inet6; + newinet->ipv6_fl_list = NULL; memcpy(newnp, np, sizeof(struct ipv6_pinfo)); newsk->sk_v6_daddr = treq->ir_v6_rmt_addr; newsk->sk_v6_rcv_saddr = treq->ir_v6_loc_addr; inet6_sk(newsk)->saddr = treq->ir_v6_loc_addr; - newnp->ipv6_fl_list = NULL; newnp->pktoptions = NULL; newsk->sk_bound_dev_if = treq->ir_iif; newinet->inet_opt = NULL; diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h index 43b7bb828738..7294e4e89b79 100644 --- a/include/linux/ipv6.h +++ b/include/linux/ipv6.h @@ -271,7 +271,6 @@ struct ipv6_pinfo { struct ipv6_mc_socklist __rcu *ipv6_mc_list; struct ipv6_ac_socklist *ipv6_ac_list; - struct ipv6_fl_socklist __rcu *ipv6_fl_list; }; /* We currently use available bits from inet_sk(sk)->inet_flags, diff --git a/include/net/inet_sock.h b/include/net/inet_sock.h index 1086256549fa..b6ec08072533 100644 --- a/include/net/inet_sock.h +++ b/include/net/inet_sock.h @@ -214,6 +214,7 @@ struct inet_sock { struct sock sk; #if IS_ENABLED(CONFIG_IPV6) struct ipv6_pinfo *pinet6; + struct ipv6_fl_socklist __rcu *ipv6_fl_list; #endif /* Socket demultiplex comparisons on incoming packets. */ #define inet_daddr sk.__sk_common.skc_daddr diff --git a/net/ipv6/ip6_flowlabel.c b/net/ipv6/ip6_flowlabel.c index a3ff575798dd..60d0be47a9f3 100644 --- a/net/ipv6/ip6_flowlabel.c +++ b/net/ipv6/ip6_flowlabel.c @@ -66,8 +66,8 @@ EXPORT_SYMBOL(ipv6_flowlabel_exclusive); fl != NULL; \ fl = rcu_dereference(fl->next)) -#define for_each_sk_fl_rcu(np, sfl) \ - for (sfl = rcu_dereference(np->ipv6_fl_list); \ +#define for_each_sk_fl_rcu(sk, sfl) \ + for (sfl = rcu_dereference(inet_sk(sk)->ipv6_fl_list); \ sfl != NULL; \ sfl = rcu_dereference(sfl->next)) @@ -262,12 +262,11 @@ static struct ip6_flowlabel *fl_intern(struct net *net, struct ip6_flowlabel *__fl6_sock_lookup(struct sock *sk, __be32 label) { struct ipv6_fl_socklist *sfl; - struct ipv6_pinfo *np = inet6_sk(sk); label &= IPV6_FLOWLABEL_MASK; rcu_read_lock(); - for_each_sk_fl_rcu(np, sfl) { + for_each_sk_fl_rcu(sk, sfl) { struct ip6_flowlabel *fl = sfl->fl; if (fl->label == label && atomic_inc_not_zero(&fl->users)) { @@ -283,16 +282,16 @@ EXPORT_SYMBOL_GPL(__fl6_sock_lookup); void fl6_free_socklist(struct sock *sk) { - struct ipv6_pinfo *np = inet6_sk(sk); + struct inet_sock *inet = inet_sk(sk); struct ipv6_fl_socklist *sfl; - if (!rcu_access_pointer(np->ipv6_fl_list)) + if (!rcu_access_pointer(inet->ipv6_fl_list)) return; spin_lock_bh(&ip6_sk_fl_lock); - while ((sfl = rcu_dereference_protected(np->ipv6_fl_list, + while ((sfl = rcu_dereference_protected(inet->ipv6_fl_list, lockdep_is_held(&ip6_sk_fl_lock))) != NULL) { - np->ipv6_fl_list = sfl->next; + inet->ipv6_fl_list = sfl->next; spin_unlock_bh(&ip6_sk_fl_lock); fl_release(sfl->fl); @@ -470,16 +469,15 @@ fl_create(struct net *net, struct sock *sk, struct in6_flowlabel_req *freq, static int mem_check(struct sock *sk) { - struct ipv6_pinfo *np = inet6_sk(sk); - struct ipv6_fl_socklist *sfl; int room = FL_MAX_SIZE - atomic_read(&fl_size); + struct ipv6_fl_socklist *sfl; int count = 0; if (room > FL_MAX_SIZE - FL_MAX_PER_SOCK) return 0; rcu_read_lock(); - for_each_sk_fl_rcu(np, sfl) + for_each_sk_fl_rcu(sk, sfl) count++; rcu_read_unlock(); @@ -492,13 +490,15 @@ static int mem_check(struct sock *sk) return 0; } -static inline void fl_link(struct ipv6_pinfo *np, struct ipv6_fl_socklist *sfl, - struct ip6_flowlabel *fl) +static inline void fl_link(struct sock *sk, struct ipv6_fl_socklist *sfl, + struct ip6_flowlabel *fl) { + struct inet_sock *inet = inet_sk(sk); + spin_lock_bh(&ip6_sk_fl_lock); sfl->fl = fl; - sfl->next = np->ipv6_fl_list; - rcu_assign_pointer(np->ipv6_fl_list, sfl); + sfl->next = inet->ipv6_fl_list; + rcu_assign_pointer(inet->ipv6_fl_list, sfl); spin_unlock_bh(&ip6_sk_fl_lock); } @@ -520,7 +520,7 @@ int ipv6_flowlabel_opt_get(struct sock *sk, struct in6_flowlabel_req *freq, rcu_read_lock(); - for_each_sk_fl_rcu(np, sfl) { + for_each_sk_fl_rcu(sk, sfl) { if (sfl->fl->label == (np->flow_label & IPV6_FLOWLABEL_MASK)) { spin_lock_bh(&ip6_fl_lock); freq->flr_label = sfl->fl->label; @@ -559,7 +559,7 @@ static int ipv6_flowlabel_put(struct sock *sk, struct in6_flowlabel_req *freq) } spin_lock_bh(&ip6_sk_fl_lock); - for (sflp = &np->ipv6_fl_list; + for (sflp = &inet_sk(sk)->ipv6_fl_list; (sfl = socklist_dereference(*sflp)) != NULL; sflp = &sfl->next) { if (sfl->fl->label == freq->flr_label) @@ -579,13 +579,12 @@ static int ipv6_flowlabel_put(struct sock *sk, struct in6_flowlabel_req *freq) static int ipv6_flowlabel_renew(struct sock *sk, struct in6_flowlabel_req *freq) { - struct ipv6_pinfo *np = inet6_sk(sk); struct net *net = sock_net(sk); struct ipv6_fl_socklist *sfl; int err; rcu_read_lock(); - for_each_sk_fl_rcu(np, sfl) { + for_each_sk_fl_rcu(sk, sfl) { if (sfl->fl->label == freq->flr_label) { err = fl6_renew(sfl->fl, freq->flr_linger, freq->flr_expires); @@ -614,7 +613,6 @@ static int ipv6_flowlabel_get(struct sock *sk, struct in6_flowlabel_req *freq, { struct ipv6_fl_socklist *sfl, *sfl1 = NULL; struct ip6_flowlabel *fl, *fl1 = NULL; - struct ipv6_pinfo *np = inet6_sk(sk); struct net *net = sock_net(sk); int err; @@ -645,7 +643,7 @@ static int ipv6_flowlabel_get(struct sock *sk, struct in6_flowlabel_req *freq, if (freq->flr_label) { err = -EEXIST; rcu_read_lock(); - for_each_sk_fl_rcu(np, sfl) { + for_each_sk_fl_rcu(sk, sfl) { if (sfl->fl->label == freq->flr_label) { if (freq->flr_flags & IPV6_FL_F_EXCL) { rcu_read_unlock(); @@ -682,7 +680,7 @@ static int ipv6_flowlabel_get(struct sock *sk, struct in6_flowlabel_req *freq, fl1->linger = fl->linger; if ((long)(fl->expires - fl1->expires) > 0) fl1->expires = fl->expires; - fl_link(np, sfl1, fl1); + fl_link(sk, sfl1, fl1); fl_free(fl); return 0; @@ -716,7 +714,7 @@ static int ipv6_flowlabel_get(struct sock *sk, struct in6_flowlabel_req *freq, } } - fl_link(np, sfl1, fl); + fl_link(sk, sfl1, fl); return 0; done: fl_free(fl); diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 59c4977a811a..6197dd4e6261 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -1386,7 +1386,9 @@ static struct sock *tcp_v6_syn_recv_sock(const struct sock *sk, struct sk_buff * if (!newsk) return NULL; - inet_sk(newsk)->pinet6 = tcp_inet6_sk(newsk); + newinet = inet_sk(newsk); + newinet->pinet6 = tcp_inet6_sk(newsk); + newinet->ipv6_fl_list = NULL; newnp = tcp_inet6_sk(newsk); newtp = tcp_sk(newsk); @@ -1405,7 +1407,6 @@ static struct sock *tcp_v6_syn_recv_sock(const struct sock *sk, struct sk_buff * newnp->ipv6_mc_list = NULL; newnp->ipv6_ac_list = NULL; - newnp->ipv6_fl_list = NULL; newnp->pktoptions = NULL; newnp->opt = NULL; newnp->mcast_oif = inet_iif(skb); @@ -1453,10 +1454,12 @@ static struct sock *tcp_v6_syn_recv_sock(const struct sock *sk, struct sk_buff * newsk->sk_gso_type = SKB_GSO_TCPV6; inet6_sk_rx_dst_set(newsk, skb); - inet_sk(newsk)->pinet6 = tcp_inet6_sk(newsk); + newinet = inet_sk(newsk); + newinet->pinet6 = tcp_inet6_sk(newsk); + newinet->ipv6_fl_list = NULL; + newinet->inet_opt = NULL; newtp = tcp_sk(newsk); - newinet = inet_sk(newsk); newnp = tcp_inet6_sk(newsk); memcpy(newnp, np, sizeof(struct ipv6_pinfo)); @@ -1469,10 +1472,8 @@ static struct sock *tcp_v6_syn_recv_sock(const struct sock *sk, struct sk_buff * First: no IPv4 options. */ - newinet->inet_opt = NULL; newnp->ipv6_mc_list = NULL; newnp->ipv6_ac_list = NULL; - newnp->ipv6_fl_list = NULL; /* Clone RX bits */ newnp->rxopt.all = np->rxopt.all; diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c index 568ff8797c39..d725b2158758 100644 --- a/net/sctp/ipv6.c +++ b/net/sctp/ipv6.c @@ -782,9 +782,10 @@ static struct sock *sctp_v6_create_accept_sk(struct sock *sk, struct sctp_association *asoc, bool kern) { - struct sock *newsk; struct ipv6_pinfo *newnp, *np = inet6_sk(sk); struct sctp6_sock *newsctp6sk; + struct inet_sock *newinet; + struct sock *newsk; newsk = sk_alloc(sock_net(sk), PF_INET6, GFP_KERNEL, sk->sk_prot, kern); if (!newsk) @@ -796,7 +797,9 @@ static struct sock *sctp_v6_create_accept_sk(struct sock *sk, sock_reset_flag(sk, SOCK_ZAPPED); newsctp6sk = (struct sctp6_sock *)newsk; - inet_sk(newsk)->pinet6 = &newsctp6sk->inet6; + newinet = inet_sk(newsk); + newinet->pinet6 = &newsctp6sk->inet6; + newinet->ipv6_fl_list = NULL; sctp_sk(newsk)->v4mapped = sctp_sk(sk)->v4mapped; @@ -805,7 +808,6 @@ static struct sock *sctp_v6_create_accept_sk(struct sock *sk, memcpy(newnp, np, sizeof(struct ipv6_pinfo)); newnp->ipv6_mc_list = NULL; newnp->ipv6_ac_list = NULL; - newnp->ipv6_fl_list = NULL; sctp_v6_copy_ip_options(sk, newsk); From 9c4609225ec1cb551006d6a03c7c4ad8cb5584c0 Mon Sep 17 00:00:00 2001 From: Xuanqiang Luo Date: Wed, 15 Oct 2025 10:02:34 +0800 Subject: [PATCH 0100/1381] rculist: Add hlist_nulls_replace_rcu() and hlist_nulls_replace_init_rcu() Add two functions to atomically replace RCU-protected hlist_nulls entries. Keep using WRITE_ONCE() to assign values to ->next and ->pprev, as mentioned in the patch below: commit efd04f8a8b45 ("rcu: Use WRITE_ONCE() for assignments to ->next for rculist_nulls") commit 860c8802ace1 ("rcu: Use WRITE_ONCE() for assignments to ->pprev for hlist_nulls") Reviewed-by: Kuniyuki Iwashima Reviewed-by: Frederic Weisbecker Reviewed-by: Eric Dumazet Signed-off-by: Xuanqiang Luo Link: https://patch.msgid.link/20251015020236.431822-2-xuanqiang.luo@linux.dev Signed-off-by: Jakub Kicinski --- include/linux/rculist_nulls.h | 59 +++++++++++++++++++++++++++++++++++ 1 file changed, 59 insertions(+) diff --git a/include/linux/rculist_nulls.h b/include/linux/rculist_nulls.h index 89186c499dd4..c26cb83ca071 100644 --- a/include/linux/rculist_nulls.h +++ b/include/linux/rculist_nulls.h @@ -52,6 +52,13 @@ static inline void hlist_nulls_del_init_rcu(struct hlist_nulls_node *n) #define hlist_nulls_next_rcu(node) \ (*((struct hlist_nulls_node __rcu __force **)&(node)->next)) +/** + * hlist_nulls_pprev_rcu - returns the dereferenced pprev of @node. + * @node: element of the list. + */ +#define hlist_nulls_pprev_rcu(node) \ + (*((struct hlist_nulls_node __rcu __force **)(node)->pprev)) + /** * hlist_nulls_del_rcu - deletes entry from hash list without re-initialization * @n: the element to delete from the hash list. @@ -152,6 +159,58 @@ static inline void hlist_nulls_add_fake(struct hlist_nulls_node *n) n->next = (struct hlist_nulls_node *)NULLS_MARKER(NULL); } +/** + * hlist_nulls_replace_rcu - replace an old entry by a new one + * @old: the element to be replaced + * @new: the new element to insert + * + * Description: + * Replace the old entry with the new one in a RCU-protected hlist_nulls, while + * permitting racing traversals. + * + * The caller must take whatever precautions are necessary (such as holding + * appropriate locks) to avoid racing with another list-mutation primitive, such + * as hlist_nulls_add_head_rcu() or hlist_nulls_del_rcu(), running on this same + * list. However, it is perfectly legal to run concurrently with the _rcu + * list-traversal primitives, such as hlist_nulls_for_each_entry_rcu(). + */ +static inline void hlist_nulls_replace_rcu(struct hlist_nulls_node *old, + struct hlist_nulls_node *new) +{ + struct hlist_nulls_node *next = old->next; + + WRITE_ONCE(new->next, next); + WRITE_ONCE(new->pprev, old->pprev); + rcu_assign_pointer(hlist_nulls_pprev_rcu(new), new); + if (!is_a_nulls(next)) + WRITE_ONCE(next->pprev, &new->next); +} + +/** + * hlist_nulls_replace_init_rcu - replace an old entry by a new one and + * initialize the old + * @old: the element to be replaced + * @new: the new element to insert + * + * Description: + * Replace the old entry with the new one in a RCU-protected hlist_nulls, while + * permitting racing traversals, and reinitialize the old entry. + * + * Note: @old must be hashed. + * + * The caller must take whatever precautions are necessary (such as holding + * appropriate locks) to avoid racing with another list-mutation primitive, such + * as hlist_nulls_add_head_rcu() or hlist_nulls_del_rcu(), running on this same + * list. However, it is perfectly legal to run concurrently with the _rcu + * list-traversal primitives, such as hlist_nulls_for_each_entry_rcu(). + */ +static inline void hlist_nulls_replace_init_rcu(struct hlist_nulls_node *old, + struct hlist_nulls_node *new) +{ + hlist_nulls_replace_rcu(old, new); + WRITE_ONCE(old->pprev, NULL); +} + /** * hlist_nulls_for_each_entry_rcu - iterate over rcu list of given type * @tpos: the type * to use as a loop cursor. From 1532ed0d0753c83e72595f785f82b48c28bbe5dc Mon Sep 17 00:00:00 2001 From: Xuanqiang Luo Date: Wed, 15 Oct 2025 10:02:35 +0800 Subject: [PATCH 0101/1381] inet: Avoid ehash lookup race in inet_ehash_insert() Since ehash lookups are lockless, if one CPU performs a lookup while another concurrently deletes and inserts (removing reqsk and inserting sk), the lookup may fail to find the socket, an RST may be sent. The call trace map is drawn as follows: CPU 0 CPU 1 ----- ----- inet_ehash_insert() spin_lock() sk_nulls_del_node_init_rcu(osk) __inet_lookup_established() (lookup failed) __sk_nulls_add_node_rcu(sk, list) spin_unlock() As both deletion and insertion operate on the same ehash chain, this patch introduces a new sk_nulls_replace_node_init_rcu() helper functions to implement atomic replacement. Fixes: 5e0724d027f0 ("tcp/dccp: fix hashdance race for passive sessions") Reviewed-by: Kuniyuki Iwashima Reviewed-by: Jiayuan Chen Signed-off-by: Xuanqiang Luo Reviewed-by: Eric Dumazet Link: https://patch.msgid.link/20251015020236.431822-3-xuanqiang.luo@linux.dev Signed-off-by: Jakub Kicinski --- include/net/sock.h | 13 +++++++++++++ net/ipv4/inet_hashtables.c | 8 ++++++-- 2 files changed, 19 insertions(+), 2 deletions(-) diff --git a/include/net/sock.h b/include/net/sock.h index 30ac2eb4ef9b..335d0da82d79 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -856,6 +856,19 @@ static inline bool sk_nulls_del_node_init_rcu(struct sock *sk) return rc; } +static inline bool sk_nulls_replace_node_init_rcu(struct sock *old, + struct sock *new) +{ + if (sk_hashed(old)) { + hlist_nulls_replace_init_rcu(&old->sk_nulls_node, + &new->sk_nulls_node); + __sock_put(old); + return true; + } + + return false; +} + static inline void __sk_add_node(struct sock *sk, struct hlist_head *list) { hlist_add_head(&sk->sk_node, list); diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c index b7024e3d9ac3..f5826ec4bcaa 100644 --- a/net/ipv4/inet_hashtables.c +++ b/net/ipv4/inet_hashtables.c @@ -720,8 +720,11 @@ bool inet_ehash_insert(struct sock *sk, struct sock *osk, bool *found_dup_sk) spin_lock(lock); if (osk) { WARN_ON_ONCE(sk->sk_hash != osk->sk_hash); - ret = sk_nulls_del_node_init_rcu(osk); - } else if (found_dup_sk) { + ret = sk_nulls_replace_node_init_rcu(osk, sk); + goto unlock; + } + + if (found_dup_sk) { *found_dup_sk = inet_ehash_lookup_by_sk(sk, list); if (*found_dup_sk) ret = false; @@ -730,6 +733,7 @@ bool inet_ehash_insert(struct sock *sk, struct sock *osk, bool *found_dup_sk) if (ret) __sk_nulls_add_node_rcu(sk, list); +unlock: spin_unlock(lock); return ret; From b8ec80b130211e7bf076ef72365952979d5f7a72 Mon Sep 17 00:00:00 2001 From: Xuanqiang Luo Date: Wed, 15 Oct 2025 10:02:36 +0800 Subject: [PATCH 0102/1381] inet: Avoid ehash lookup race in inet_twsk_hashdance_schedule() Since ehash lookups are lockless, if another CPU is converting sk to tw concurrently, fetching the newly inserted tw with tw->tw_refcnt == 0 cause lookup failure. The call trace map is drawn as follows: CPU 0 CPU 1 ----- ----- inet_twsk_hashdance_schedule() spin_lock() inet_twsk_add_node_rcu(tw, ...) __inet_lookup_established() (find tw, failure due to tw_refcnt = 0) __sk_nulls_del_node_init_rcu(sk) refcount_set(&tw->tw_refcnt, 3) spin_unlock() By replacing sk with tw atomically via hlist_nulls_replace_init_rcu() after setting tw_refcnt, we ensure that tw is either fully initialized or not visible to other CPUs, eliminating the race. It's worth noting that we held lock_sock() before the replacement, so there's no need to check if sk is hashed. Thanks to Kuniyuki Iwashima! Fixes: 3ab5aee7fe84 ("net: Convert TCP & DCCP hash tables to use RCU / hlist_nulls") Reviewed-by: Kuniyuki Iwashima Reviewed-by: Jiayuan Chen Signed-off-by: Xuanqiang Luo Reviewed-by: Eric Dumazet Link: https://patch.msgid.link/20251015020236.431822-4-xuanqiang.luo@linux.dev Signed-off-by: Jakub Kicinski --- net/ipv4/inet_timewait_sock.c | 35 ++++++++++++----------------------- 1 file changed, 12 insertions(+), 23 deletions(-) diff --git a/net/ipv4/inet_timewait_sock.c b/net/ipv4/inet_timewait_sock.c index c96d61d08854..d4c781a0667f 100644 --- a/net/ipv4/inet_timewait_sock.c +++ b/net/ipv4/inet_timewait_sock.c @@ -88,12 +88,6 @@ void inet_twsk_put(struct inet_timewait_sock *tw) } EXPORT_SYMBOL_GPL(inet_twsk_put); -static void inet_twsk_add_node_rcu(struct inet_timewait_sock *tw, - struct hlist_nulls_head *list) -{ - hlist_nulls_add_head_rcu(&tw->tw_node, list); -} - static void inet_twsk_schedule(struct inet_timewait_sock *tw, int timeo) { __inet_twsk_schedule(tw, timeo, false); @@ -113,13 +107,12 @@ void inet_twsk_hashdance_schedule(struct inet_timewait_sock *tw, { const struct inet_sock *inet = inet_sk(sk); const struct inet_connection_sock *icsk = inet_csk(sk); - struct inet_ehash_bucket *ehead = inet_ehash_bucket(hashinfo, sk->sk_hash); spinlock_t *lock = inet_ehash_lockp(hashinfo, sk->sk_hash); struct inet_bind_hashbucket *bhead, *bhead2; - /* Step 1: Put TW into bind hash. Original socket stays there too. - Note, that any socket with inet->num != 0 MUST be bound in - binding cache, even if it is closed. + /* Put TW into bind hash. Original socket stays there too. + * Note, that any socket with inet->num != 0 MUST be bound in + * binding cache, even if it is closed. */ bhead = &hashinfo->bhash[inet_bhashfn(twsk_net(tw), inet->inet_num, hashinfo->bhash_size)]; @@ -141,19 +134,6 @@ void inet_twsk_hashdance_schedule(struct inet_timewait_sock *tw, spin_lock(lock); - /* Step 2: Hash TW into tcp ehash chain */ - inet_twsk_add_node_rcu(tw, &ehead->chain); - - /* Step 3: Remove SK from hash chain */ - if (__sk_nulls_del_node_init_rcu(sk)) - sock_prot_inuse_add(sock_net(sk), sk->sk_prot, -1); - - - /* Ensure above writes are committed into memory before updating the - * refcount. - * Provides ordering vs later refcount_inc(). - */ - smp_wmb(); /* tw_refcnt is set to 3 because we have : * - one reference for bhash chain. * - one reference for ehash chain. @@ -163,6 +143,15 @@ void inet_twsk_hashdance_schedule(struct inet_timewait_sock *tw, */ refcount_set(&tw->tw_refcnt, 3); + /* Ensure tw_refcnt has been set before tw is published. + * smp_wmb() provides the necessary memory barrier to enforce this + * ordering. + */ + smp_wmb(); + + hlist_nulls_replace_init_rcu(&sk->sk_nulls_node, &tw->tw_node); + sock_prot_inuse_add(sock_net(sk), sk->sk_prot, -1); + inet_twsk_schedule(tw, timeo); spin_unlock(lock); From 2af8ff1e472e9862983303890e98d45f40863351 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 16 Oct 2025 13:51:47 +0200 Subject: [PATCH 0103/1381] net: Kconfig: discourage drop_monitor enablement Quoting Eric Dumazet: "I do not understand the fascination with net/core/drop_monitor.c [..] misses all the features, flexibility, scalability that 'perf', eBPF tracing, bpftrace, .... have today." Reword DROP_MONITOR kconfig help text to clearly state that its not related to perf-based drop monitoring and that its safe to disable this unless support for the older netlink-based tools is needed. Signed-off-by: Florian Westphal Reviewed-by: Simon Horman Reviewed-by: Eric Dumazet Link: https://patch.msgid.link/20251016115147.18503-1-fw@strlen.de Signed-off-by: Jakub Kicinski --- net/Kconfig | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/net/Kconfig b/net/Kconfig index 1d3f757d4b07..62266eaf0e95 100644 --- a/net/Kconfig +++ b/net/Kconfig @@ -400,15 +400,15 @@ config NET_PKTGEN module will be called pktgen. config NET_DROP_MONITOR - tristate "Network packet drop alerting service" + tristate "Legacy network packet drop alerting service" depends on INET && TRACEPOINTS help This feature provides an alerting service to userspace in the event that packets are discarded in the network stack. Alerts are broadcast via netlink socket to any listening user space - process. If you don't need network drop alerts, or if you are ok - just checking the various proc files and other utilities for - drop statistics, say N here. + process. This feature is NOT related to "perf" based drop monitoring. + Say N here unless you need to support older userspace tools like + "dropwatch". endmenu # Network testing From 3dc2a17efc5f9d0a51caa9f63600e6f02e0feba4 Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Thu, 16 Oct 2025 21:25:28 +0200 Subject: [PATCH 0104/1381] r8169: reconfigure rx unconditionally before chip reset when resuming There's a good chance that more chip versions suffer from the same hw issue. So let's reconfigure rx unconditionally before the chip reset when resuming. This shouldn't have any side effect on unaffected chip versions. Reviewed-by: Simon Horman Signed-off-by: Heiner Kallweit Link: https://patch.msgid.link/a5c2e2d2-226f-4896-b8f6-45e2d91f0e24@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/realtek/r8169_main.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c index d18734fe12e4..2a4d9b548810 100644 --- a/drivers/net/ethernet/realtek/r8169_main.c +++ b/drivers/net/ethernet/realtek/r8169_main.c @@ -4995,9 +4995,7 @@ static int rtl8169_resume(struct device *device) clk_prepare_enable(tp->clk); /* Some chip versions may truncate packets without this initialization */ - if (tp->mac_version == RTL_GIGA_MAC_VER_37 || - tp->mac_version == RTL_GIGA_MAC_VER_46) - rtl_init_rxcfg(tp); + rtl_init_rxcfg(tp); return rtl8169_runtime_resume(device); } From f578ff4c53889cb9bc15a5b2acc7274d46bb38cb Mon Sep 17 00:00:00 2001 From: Carlos Llamas Date: Thu, 16 Oct 2025 18:25:37 +0000 Subject: [PATCH 0105/1381] selftests/net: io_uring: fix unknown errnum values The io_uring functions return negative error values, but error() expects these to be positive to properly match them to an errno string. Fix this to make sure the correct error descriptions are displayed upon failure. Signed-off-by: Carlos Llamas Link: https://patch.msgid.link/20251016182538.3790567-1-cmllamas@google.com Signed-off-by: Jakub Kicinski --- .../selftests/net/io_uring_zerocopy_tx.c | 24 +++++++++---------- 1 file changed, 12 insertions(+), 12 deletions(-) diff --git a/tools/testing/selftests/net/io_uring_zerocopy_tx.c b/tools/testing/selftests/net/io_uring_zerocopy_tx.c index 76e604e4810e..7bfeeb133705 100644 --- a/tools/testing/selftests/net/io_uring_zerocopy_tx.c +++ b/tools/testing/selftests/net/io_uring_zerocopy_tx.c @@ -106,14 +106,14 @@ static void do_tx(int domain, int type, int protocol) ret = io_uring_queue_init(512, &ring, 0); if (ret) - error(1, ret, "io_uring: queue init"); + error(1, -ret, "io_uring: queue init"); iov.iov_base = payload; iov.iov_len = cfg_payload_len; ret = io_uring_register_buffers(&ring, &iov, 1); if (ret) - error(1, ret, "io_uring: buffer registration"); + error(1, -ret, "io_uring: buffer registration"); tstop = gettimeofday_ms() + cfg_runtime_ms; do { @@ -149,24 +149,24 @@ static void do_tx(int domain, int type, int protocol) ret = io_uring_submit(&ring); if (ret != cfg_nr_reqs) - error(1, ret, "submit"); + error(1, -ret, "submit"); if (cfg_cork) do_setsockopt(fd, IPPROTO_UDP, UDP_CORK, 0); for (i = 0; i < cfg_nr_reqs; i++) { ret = io_uring_wait_cqe(&ring, &cqe); if (ret) - error(1, ret, "wait cqe"); + error(1, -ret, "wait cqe"); if (cqe->user_data != NONZC_TAG && cqe->user_data != ZC_TAG) - error(1, -EINVAL, "invalid cqe->user_data"); + error(1, EINVAL, "invalid cqe->user_data"); if (cqe->flags & IORING_CQE_F_NOTIF) { if (cqe->flags & IORING_CQE_F_MORE) - error(1, -EINVAL, "invalid notif flags"); + error(1, EINVAL, "invalid notif flags"); if (compl_cqes <= 0) - error(1, -EINVAL, "notification mismatch"); + error(1, EINVAL, "notification mismatch"); compl_cqes--; i--; io_uring_cqe_seen(&ring); @@ -174,14 +174,14 @@ static void do_tx(int domain, int type, int protocol) } if (cqe->flags & IORING_CQE_F_MORE) { if (cqe->user_data != ZC_TAG) - error(1, cqe->res, "unexpected F_MORE"); + error(1, -cqe->res, "unexpected F_MORE"); compl_cqes++; } if (cqe->res >= 0) { packets++; bytes += cqe->res; } else if (cqe->res != -EAGAIN) { - error(1, cqe->res, "send failed"); + error(1, -cqe->res, "send failed"); } io_uring_cqe_seen(&ring); } @@ -190,11 +190,11 @@ static void do_tx(int domain, int type, int protocol) while (compl_cqes) { ret = io_uring_wait_cqe(&ring, &cqe); if (ret) - error(1, ret, "wait cqe"); + error(1, -ret, "wait cqe"); if (cqe->flags & IORING_CQE_F_MORE) - error(1, -EINVAL, "invalid notif flags"); + error(1, EINVAL, "invalid notif flags"); if (!(cqe->flags & IORING_CQE_F_NOTIF)) - error(1, -EINVAL, "missing notif flag"); + error(1, EINVAL, "missing notif flag"); io_uring_cqe_seen(&ring); compl_cqes--; From 37a183d3b7cdb873e7f5f9daef1ad6d8f7c95fb7 Mon Sep 17 00:00:00 2001 From: Eric Biggers Date: Tue, 14 Oct 2025 14:58:36 -0700 Subject: [PATCH 0106/1381] tcp: Convert tcp-md5 to use MD5 library instead of crypto_ahash Make tcp-md5 use the MD5 library API (added in 6.18) instead of the crypto_ahash API. This is much simpler and also more efficient: - The library API just operates on struct md5_ctx. Just allocate this struct on the stack instead of using a pool of pre-allocated crypto_ahash and ahash_request objects. - The library API accepts standard pointers and doesn't require scatterlists. So, for hashing the headers just use an on-stack buffer instead of a pool of pre-allocated kmalloc'ed scratch buffers. - The library API never fails. Therefore, checking for MD5 hashing errors is no longer necessary. Update tcp_v4_md5_hash_skb(), tcp_v6_md5_hash_skb(), tcp_v4_md5_hash_hdr(), tcp_v6_md5_hash_hdr(), tcp_md5_hash_key(), tcp_sock_af_ops::calc_md5_hash, and tcp_request_sock_ops::calc_md5_hash to return void instead of int. - The library API provides direct access to the MD5 code, eliminating unnecessary overhead such as indirect function calls and scatterlist management. Microbenchmarks of tcp_v4_md5_hash_skb() on x86_64 show a speedup from 7518 to 7041 cycles (6% fewer) with skb->len == 1440, or from 1020 to 678 cycles (33% fewer) with skb->len == 140. Since tcp_sigpool_hash_skb_data() can no longer be used, add a function tcp_md5_hash_skb_data() which is specialized to MD5. Of course, to the extent that this duplicates any code, it's well worth it. To preserve the existing behavior of TCP-MD5 support being disabled when the kernel is booted with "fips=1", make tcp_md5_do_add() check fips_enabled itself. Previously it relied on the error from crypto_alloc_ahash("md5") being bubbled up. I don't know for sure that this is actually needed, but this preserves the existing behavior. Tested with bidirectional TCP-MD5, both IPv4 and IPv6, between a kernel that includes this commit and a kernel that doesn't include this commit. (Side note: please don't use TCP-MD5! It's cryptographically weak. But as long as Linux supports it, it might as well be implemented properly.) Signed-off-by: Eric Biggers Link: https://patch.msgid.link/20251014215836.115616-1-ebiggers@kernel.org Signed-off-by: Jakub Kicinski --- include/net/tcp.h | 26 +++----- net/ipv4/Kconfig | 4 +- net/ipv4/tcp.c | 73 +++++++++----------- net/ipv4/tcp_ipv4.c | 139 +++++++++++++-------------------------- net/ipv4/tcp_minisocks.c | 2 - net/ipv6/tcp_ipv6.c | 117 ++++++++++---------------------- 6 files changed, 121 insertions(+), 240 deletions(-) diff --git a/include/net/tcp.h b/include/net/tcp.h index 1e547138f4fb..67fdd2523d92 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1898,13 +1898,6 @@ struct tcp6_pseudohdr { __be32 protocol; /* including padding */ }; -union tcp_md5sum_block { - struct tcp4_pseudohdr ip4; -#if IS_ENABLED(CONFIG_IPV6) - struct tcp6_pseudohdr ip6; -#endif -}; - /* * struct tcp_sigpool - per-CPU pool of ahash_requests * @scratch: per-CPU temporary area, that can be used between @@ -1939,8 +1932,8 @@ int tcp_sigpool_start(unsigned int id, struct tcp_sigpool *c); void tcp_sigpool_end(struct tcp_sigpool *c); size_t tcp_sigpool_algo(unsigned int id, char *buf, size_t buf_len); /* - functions */ -int tcp_v4_md5_hash_skb(char *md5_hash, const struct tcp_md5sig_key *key, - const struct sock *sk, const struct sk_buff *skb); +void tcp_v4_md5_hash_skb(char *md5_hash, const struct tcp_md5sig_key *key, + const struct sock *sk, const struct sk_buff *skb); int tcp_md5_do_add(struct sock *sk, const union tcp_md5_addr *addr, int family, u8 prefixlen, int l3index, u8 flags, const u8 *newkey, u8 newkeylen); @@ -1999,13 +1992,10 @@ static inline void tcp_md5_destruct_sock(struct sock *sk) } #endif -int tcp_md5_alloc_sigpool(void); -void tcp_md5_release_sigpool(void); -void tcp_md5_add_sigpool(void); -extern int tcp_md5_sigpool_id; - -int tcp_md5_hash_key(struct tcp_sigpool *hp, - const struct tcp_md5sig_key *key); +struct md5_ctx; +void tcp_md5_hash_skb_data(struct md5_ctx *ctx, const struct sk_buff *skb, + unsigned int header_len); +void tcp_md5_hash_key(struct md5_ctx *ctx, const struct tcp_md5sig_key *key); /* From tcp_fastopen.c */ void tcp_fastopen_cache_get(struct sock *sk, u16 *mss, @@ -2355,7 +2345,7 @@ struct tcp_sock_af_ops { #ifdef CONFIG_TCP_MD5SIG struct tcp_md5sig_key *(*md5_lookup) (const struct sock *sk, const struct sock *addr_sk); - int (*calc_md5_hash)(char *location, + void (*calc_md5_hash)(char *location, const struct tcp_md5sig_key *md5, const struct sock *sk, const struct sk_buff *skb); @@ -2383,7 +2373,7 @@ struct tcp_request_sock_ops { #ifdef CONFIG_TCP_MD5SIG struct tcp_md5sig_key *(*req_md5_lookup)(const struct sock *sk, const struct sock *addr_sk); - int (*calc_md5_hash) (char *location, + void (*calc_md5_hash) (char *location, const struct tcp_md5sig_key *md5, const struct sock *sk, const struct sk_buff *skb); diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig index 12850a277251..b71c22475c51 100644 --- a/net/ipv4/Kconfig +++ b/net/ipv4/Kconfig @@ -760,9 +760,7 @@ config TCP_AO config TCP_MD5SIG bool "TCP: MD5 Signature Option support (RFC2385)" - select CRYPTO - select CRYPTO_MD5 - select TCP_SIGPOOL + select CRYPTO_LIB_MD5 help RFC2385 specifies a method of giving MD5 protection to TCP sessions. Its main (only?) use is to protect BGP sessions between core routers diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 4d720aa09a4c..0ccc5405e740 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -243,7 +243,7 @@ #define pr_fmt(fmt) "TCP: " fmt -#include +#include #include #include #include @@ -253,7 +253,6 @@ #include #include #include -#include #include #include #include @@ -425,7 +424,6 @@ void tcp_md5_destruct_sock(struct sock *sk) tcp_clear_md5_list(sk); kfree(rcu_replace_pointer(tp->md5sig_info, NULL, 1)); static_branch_slow_dec_deferred(&tcp_md5_needed); - tcp_md5_release_sigpool(); } } EXPORT_IPV6_MOD_GPL(tcp_md5_destruct_sock); @@ -4838,52 +4836,45 @@ int tcp_getsockopt(struct sock *sk, int level, int optname, char __user *optval, EXPORT_IPV6_MOD(tcp_getsockopt); #ifdef CONFIG_TCP_MD5SIG -int tcp_md5_sigpool_id = -1; -EXPORT_IPV6_MOD_GPL(tcp_md5_sigpool_id); - -int tcp_md5_alloc_sigpool(void) +void tcp_md5_hash_skb_data(struct md5_ctx *ctx, const struct sk_buff *skb, + unsigned int header_len) { - size_t scratch_size; - int ret; + const unsigned int head_data_len = skb_headlen(skb) > header_len ? + skb_headlen(skb) - header_len : 0; + const struct skb_shared_info *shi = skb_shinfo(skb); + struct sk_buff *frag_iter; + unsigned int i; - scratch_size = sizeof(union tcp_md5sum_block) + sizeof(struct tcphdr); - ret = tcp_sigpool_alloc_ahash("md5", scratch_size); - if (ret >= 0) { - /* As long as any md5 sigpool was allocated, the return - * id would stay the same. Re-write the id only for the case - * when previously all MD5 keys were deleted and this call - * allocates the first MD5 key, which may return a different - * sigpool id than was used previously. - */ - WRITE_ONCE(tcp_md5_sigpool_id, ret); /* Avoids the compiler potentially being smart here */ - return 0; + md5_update(ctx, (const u8 *)tcp_hdr(skb) + header_len, head_data_len); + + for (i = 0; i < shi->nr_frags; ++i) { + const skb_frag_t *f = &shi->frags[i]; + u32 p_off, p_len, copied; + const void *vaddr; + struct page *p; + + skb_frag_foreach_page(f, skb_frag_off(f), skb_frag_size(f), + p, p_off, p_len, copied) { + vaddr = kmap_local_page(p); + md5_update(ctx, vaddr + p_off, p_len); + kunmap_local(vaddr); + } } - return ret; -} -void tcp_md5_release_sigpool(void) -{ - tcp_sigpool_release(READ_ONCE(tcp_md5_sigpool_id)); + skb_walk_frags(skb, frag_iter) + tcp_md5_hash_skb_data(ctx, frag_iter, 0); } +EXPORT_IPV6_MOD(tcp_md5_hash_skb_data); -void tcp_md5_add_sigpool(void) -{ - tcp_sigpool_get(READ_ONCE(tcp_md5_sigpool_id)); -} - -int tcp_md5_hash_key(struct tcp_sigpool *hp, - const struct tcp_md5sig_key *key) +void tcp_md5_hash_key(struct md5_ctx *ctx, + const struct tcp_md5sig_key *key) { u8 keylen = READ_ONCE(key->keylen); /* paired with WRITE_ONCE() in tcp_md5_do_add */ - struct scatterlist sg; - - sg_init_one(&sg, key->key, keylen); - ahash_request_set_crypt(hp->req, &sg, NULL, keylen); /* We use data_race() because tcp_md5_do_add() might change * key->key under us */ - return data_race(crypto_ahash_update(hp->req)); + data_race(({ md5_update(ctx, key->key, keylen), 0; })); } EXPORT_IPV6_MOD(tcp_md5_hash_key); @@ -4902,7 +4893,6 @@ tcp_inbound_md5_hash(const struct sock *sk, const struct sk_buff *skb, const struct tcp_sock *tp = tcp_sk(sk); struct tcp_md5sig_key *key; u8 newhash[16]; - int genhash; key = tcp_md5_do_lookup(sk, l3index, saddr, family); @@ -4917,11 +4907,10 @@ tcp_inbound_md5_hash(const struct sock *sk, const struct sk_buff *skb, * IPv4-mapped case. */ if (family == AF_INET) - genhash = tcp_v4_md5_hash_skb(newhash, key, NULL, skb); + tcp_v4_md5_hash_skb(newhash, key, NULL, skb); else - genhash = tp->af_specific->calc_md5_hash(newhash, key, - NULL, skb); - if (genhash || memcmp(hash_location, newhash, 16) != 0) { + tp->af_specific->calc_md5_hash(newhash, key, NULL, skb); + if (memcmp(hash_location, newhash, 16) != 0) { NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPMD5FAILURE); trace_tcp_hash_md5_mismatch(sk, skb); return SKB_DROP_REASON_TCP_MD5FAILURE; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index b1fcf3e4e1ce..40a76da5364a 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -53,6 +53,7 @@ #include #include #include +#include #include #include #include @@ -86,14 +87,13 @@ #include #include -#include -#include +#include #include #ifdef CONFIG_TCP_MD5SIG -static int tcp_v4_md5_hash_hdr(char *md5_hash, const struct tcp_md5sig_key *key, - __be32 daddr, __be32 saddr, const struct tcphdr *th); +static void tcp_v4_md5_hash_hdr(char *md5_hash, const struct tcp_md5sig_key *key, + __be32 daddr, __be32 saddr, const struct tcphdr *th); #endif struct inet_hashinfo tcp_hashinfo; @@ -754,7 +754,6 @@ static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb, struct tcp_md5sig_key *key = NULL; unsigned char newhash[16]; struct sock *sk1 = NULL; - int genhash; #endif u64 transmit_time = 0; struct sock *ctl_sk; @@ -840,11 +839,9 @@ static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb, if (!key) goto out; - - genhash = tcp_v4_md5_hash_skb(newhash, key, NULL, skb); - if (genhash || memcmp(md5_hash_location, newhash, 16) != 0) + tcp_v4_md5_hash_skb(newhash, key, NULL, skb); + if (memcmp(md5_hash_location, newhash, 16) != 0) goto out; - } if (key) { @@ -1425,21 +1422,20 @@ int tcp_md5_do_add(struct sock *sk, const union tcp_md5_addr *addr, struct tcp_sock *tp = tcp_sk(sk); if (!rcu_dereference_protected(tp->md5sig_info, lockdep_sock_is_held(sk))) { - if (tcp_md5_alloc_sigpool()) - return -ENOMEM; - - if (tcp_md5sig_info_add(sk, GFP_KERNEL)) { - tcp_md5_release_sigpool(); - return -ENOMEM; + if (fips_enabled) { + pr_warn_once("TCP-MD5 support is disabled due to FIPS\n"); + return -EOPNOTSUPP; } + if (tcp_md5sig_info_add(sk, GFP_KERNEL)) + return -ENOMEM; + if (!static_branch_inc(&tcp_md5_needed.key)) { struct tcp_md5sig_info *md5sig; md5sig = rcu_dereference_protected(tp->md5sig_info, lockdep_sock_is_held(sk)); rcu_assign_pointer(tp->md5sig_info, NULL); kfree_rcu(md5sig, rcu); - tcp_md5_release_sigpool(); return -EUSERS; } } @@ -1456,12 +1452,9 @@ int tcp_md5_key_copy(struct sock *sk, const union tcp_md5_addr *addr, struct tcp_sock *tp = tcp_sk(sk); if (!rcu_dereference_protected(tp->md5sig_info, lockdep_sock_is_held(sk))) { - tcp_md5_add_sigpool(); - if (tcp_md5sig_info_add(sk, sk_gfp_mask(sk, GFP_ATOMIC))) { - tcp_md5_release_sigpool(); + if (tcp_md5sig_info_add(sk, sk_gfp_mask(sk, GFP_ATOMIC))) return -ENOMEM; - } if (!static_key_fast_inc_not_disabled(&tcp_md5_needed.key.key)) { struct tcp_md5sig_info *md5sig; @@ -1470,7 +1463,6 @@ int tcp_md5_key_copy(struct sock *sk, const union tcp_md5_addr *addr, net_warn_ratelimited("Too many TCP-MD5 keys in the system\n"); rcu_assign_pointer(tp->md5sig_info, NULL); kfree_rcu(md5sig, rcu); - tcp_md5_release_sigpool(); return -EUSERS; } } @@ -1578,66 +1570,44 @@ static int tcp_v4_parse_md5_keys(struct sock *sk, int optname, cmd.tcpm_key, cmd.tcpm_keylen); } -static int tcp_v4_md5_hash_headers(struct tcp_sigpool *hp, - __be32 daddr, __be32 saddr, - const struct tcphdr *th, int nbytes) +static void tcp_v4_md5_hash_headers(struct md5_ctx *ctx, + __be32 daddr, __be32 saddr, + const struct tcphdr *th, int nbytes) { - struct tcp4_pseudohdr *bp; - struct scatterlist sg; - struct tcphdr *_th; + struct { + struct tcp4_pseudohdr ip; + struct tcphdr tcp; + } h; - bp = hp->scratch; - bp->saddr = saddr; - bp->daddr = daddr; - bp->pad = 0; - bp->protocol = IPPROTO_TCP; - bp->len = cpu_to_be16(nbytes); - - _th = (struct tcphdr *)(bp + 1); - memcpy(_th, th, sizeof(*th)); - _th->check = 0; - - sg_init_one(&sg, bp, sizeof(*bp) + sizeof(*th)); - ahash_request_set_crypt(hp->req, &sg, NULL, - sizeof(*bp) + sizeof(*th)); - return crypto_ahash_update(hp->req); + h.ip.saddr = saddr; + h.ip.daddr = daddr; + h.ip.pad = 0; + h.ip.protocol = IPPROTO_TCP; + h.ip.len = cpu_to_be16(nbytes); + h.tcp = *th; + h.tcp.check = 0; + md5_update(ctx, (const u8 *)&h, sizeof(h.ip) + sizeof(h.tcp)); } -static int tcp_v4_md5_hash_hdr(char *md5_hash, const struct tcp_md5sig_key *key, - __be32 daddr, __be32 saddr, const struct tcphdr *th) +static noinline_for_stack void +tcp_v4_md5_hash_hdr(char *md5_hash, const struct tcp_md5sig_key *key, + __be32 daddr, __be32 saddr, const struct tcphdr *th) { - struct tcp_sigpool hp; + struct md5_ctx ctx; - if (tcp_sigpool_start(tcp_md5_sigpool_id, &hp)) - goto clear_hash_nostart; - - if (crypto_ahash_init(hp.req)) - goto clear_hash; - if (tcp_v4_md5_hash_headers(&hp, daddr, saddr, th, th->doff << 2)) - goto clear_hash; - if (tcp_md5_hash_key(&hp, key)) - goto clear_hash; - ahash_request_set_crypt(hp.req, NULL, md5_hash, 0); - if (crypto_ahash_final(hp.req)) - goto clear_hash; - - tcp_sigpool_end(&hp); - return 0; - -clear_hash: - tcp_sigpool_end(&hp); -clear_hash_nostart: - memset(md5_hash, 0, 16); - return 1; + md5_init(&ctx); + tcp_v4_md5_hash_headers(&ctx, daddr, saddr, th, th->doff << 2); + tcp_md5_hash_key(&ctx, key); + md5_final(&ctx, md5_hash); } -int tcp_v4_md5_hash_skb(char *md5_hash, const struct tcp_md5sig_key *key, - const struct sock *sk, - const struct sk_buff *skb) +noinline_for_stack void +tcp_v4_md5_hash_skb(char *md5_hash, const struct tcp_md5sig_key *key, + const struct sock *sk, const struct sk_buff *skb) { const struct tcphdr *th = tcp_hdr(skb); - struct tcp_sigpool hp; __be32 saddr, daddr; + struct md5_ctx ctx; if (sk) { /* valid for establish/request sockets */ saddr = sk->sk_rcv_saddr; @@ -1648,30 +1618,11 @@ int tcp_v4_md5_hash_skb(char *md5_hash, const struct tcp_md5sig_key *key, daddr = iph->daddr; } - if (tcp_sigpool_start(tcp_md5_sigpool_id, &hp)) - goto clear_hash_nostart; - - if (crypto_ahash_init(hp.req)) - goto clear_hash; - - if (tcp_v4_md5_hash_headers(&hp, daddr, saddr, th, skb->len)) - goto clear_hash; - if (tcp_sigpool_hash_skb_data(&hp, skb, th->doff << 2)) - goto clear_hash; - if (tcp_md5_hash_key(&hp, key)) - goto clear_hash; - ahash_request_set_crypt(hp.req, NULL, md5_hash, 0); - if (crypto_ahash_final(hp.req)) - goto clear_hash; - - tcp_sigpool_end(&hp); - return 0; - -clear_hash: - tcp_sigpool_end(&hp); -clear_hash_nostart: - memset(md5_hash, 0, 16); - return 1; + md5_init(&ctx); + tcp_v4_md5_hash_headers(&ctx, daddr, saddr, th, skb->len); + tcp_md5_hash_skb_data(&ctx, skb, th->doff << 2); + tcp_md5_hash_key(&ctx, key); + md5_final(&ctx, md5_hash); } EXPORT_IPV6_MOD(tcp_v4_md5_hash_skb); diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index 2ec8c6f1cdcc..ded2cf1f6006 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -312,7 +312,6 @@ static void tcp_time_wait_init(struct sock *sk, struct tcp_timewait_sock *tcptw) return; if (!static_key_fast_inc_not_disabled(&tcp_md5_needed.key.key)) goto out_free; - tcp_md5_add_sigpool(); } return; out_free: @@ -406,7 +405,6 @@ void tcp_twsk_destructor(struct sock *sk) if (twsk->tw_md5_key) { kfree(twsk->tw_md5_key); static_branch_slow_dec_deferred(&tcp_md5_needed); - tcp_md5_release_sigpool(); } } #endif diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 6197dd4e6261..06eb90e4078e 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -67,8 +67,7 @@ #include #include -#include -#include +#include #include @@ -691,69 +690,45 @@ static int tcp_v6_parse_md5_keys(struct sock *sk, int optname, cmd.tcpm_key, cmd.tcpm_keylen); } -static int tcp_v6_md5_hash_headers(struct tcp_sigpool *hp, - const struct in6_addr *daddr, - const struct in6_addr *saddr, - const struct tcphdr *th, int nbytes) +static void tcp_v6_md5_hash_headers(struct md5_ctx *ctx, + const struct in6_addr *daddr, + const struct in6_addr *saddr, + const struct tcphdr *th, int nbytes) { - struct tcp6_pseudohdr *bp; - struct scatterlist sg; - struct tcphdr *_th; + struct { + struct tcp6_pseudohdr ip; /* TCP pseudo-header (RFC2460) */ + struct tcphdr tcp; + } h; - bp = hp->scratch; - /* 1. TCP pseudo-header (RFC2460) */ - bp->saddr = *saddr; - bp->daddr = *daddr; - bp->protocol = cpu_to_be32(IPPROTO_TCP); - bp->len = cpu_to_be32(nbytes); - - _th = (struct tcphdr *)(bp + 1); - memcpy(_th, th, sizeof(*th)); - _th->check = 0; - - sg_init_one(&sg, bp, sizeof(*bp) + sizeof(*th)); - ahash_request_set_crypt(hp->req, &sg, NULL, - sizeof(*bp) + sizeof(*th)); - return crypto_ahash_update(hp->req); + h.ip.saddr = *saddr; + h.ip.daddr = *daddr; + h.ip.protocol = cpu_to_be32(IPPROTO_TCP); + h.ip.len = cpu_to_be32(nbytes); + h.tcp = *th; + h.tcp.check = 0; + md5_update(ctx, (const u8 *)&h, sizeof(h.ip) + sizeof(h.tcp)); } -static int tcp_v6_md5_hash_hdr(char *md5_hash, const struct tcp_md5sig_key *key, - const struct in6_addr *daddr, struct in6_addr *saddr, - const struct tcphdr *th) +static noinline_for_stack void +tcp_v6_md5_hash_hdr(char *md5_hash, const struct tcp_md5sig_key *key, + const struct in6_addr *daddr, struct in6_addr *saddr, + const struct tcphdr *th) { - struct tcp_sigpool hp; + struct md5_ctx ctx; - if (tcp_sigpool_start(tcp_md5_sigpool_id, &hp)) - goto clear_hash_nostart; - - if (crypto_ahash_init(hp.req)) - goto clear_hash; - if (tcp_v6_md5_hash_headers(&hp, daddr, saddr, th, th->doff << 2)) - goto clear_hash; - if (tcp_md5_hash_key(&hp, key)) - goto clear_hash; - ahash_request_set_crypt(hp.req, NULL, md5_hash, 0); - if (crypto_ahash_final(hp.req)) - goto clear_hash; - - tcp_sigpool_end(&hp); - return 0; - -clear_hash: - tcp_sigpool_end(&hp); -clear_hash_nostart: - memset(md5_hash, 0, 16); - return 1; + md5_init(&ctx); + tcp_v6_md5_hash_headers(&ctx, daddr, saddr, th, th->doff << 2); + tcp_md5_hash_key(&ctx, key); + md5_final(&ctx, md5_hash); } -static int tcp_v6_md5_hash_skb(char *md5_hash, - const struct tcp_md5sig_key *key, - const struct sock *sk, - const struct sk_buff *skb) +static noinline_for_stack void +tcp_v6_md5_hash_skb(char *md5_hash, const struct tcp_md5sig_key *key, + const struct sock *sk, const struct sk_buff *skb) { const struct tcphdr *th = tcp_hdr(skb); const struct in6_addr *saddr, *daddr; - struct tcp_sigpool hp; + struct md5_ctx ctx; if (sk) { /* valid for establish/request sockets */ saddr = &sk->sk_v6_rcv_saddr; @@ -764,30 +739,11 @@ static int tcp_v6_md5_hash_skb(char *md5_hash, daddr = &ip6h->daddr; } - if (tcp_sigpool_start(tcp_md5_sigpool_id, &hp)) - goto clear_hash_nostart; - - if (crypto_ahash_init(hp.req)) - goto clear_hash; - - if (tcp_v6_md5_hash_headers(&hp, daddr, saddr, th, skb->len)) - goto clear_hash; - if (tcp_sigpool_hash_skb_data(&hp, skb, th->doff << 2)) - goto clear_hash; - if (tcp_md5_hash_key(&hp, key)) - goto clear_hash; - ahash_request_set_crypt(hp.req, NULL, md5_hash, 0); - if (crypto_ahash_final(hp.req)) - goto clear_hash; - - tcp_sigpool_end(&hp); - return 0; - -clear_hash: - tcp_sigpool_end(&hp); -clear_hash_nostart: - memset(md5_hash, 0, 16); - return 1; + md5_init(&ctx); + tcp_v6_md5_hash_headers(&ctx, daddr, saddr, th, skb->len); + tcp_md5_hash_skb_data(&ctx, skb, th->doff << 2); + tcp_md5_hash_key(&ctx, key); + md5_final(&ctx, md5_hash); } #endif @@ -1032,7 +988,6 @@ static void tcp_v6_send_reset(const struct sock *sk, struct sk_buff *skb, int oif = 0; #ifdef CONFIG_TCP_MD5SIG unsigned char newhash[16]; - int genhash; struct sock *sk1 = NULL; #endif @@ -1091,8 +1046,8 @@ static void tcp_v6_send_reset(const struct sock *sk, struct sk_buff *skb, goto out; key.type = TCP_KEY_MD5; - genhash = tcp_v6_md5_hash_skb(newhash, key.md5_key, NULL, skb); - if (genhash || memcmp(md5_hash_location, newhash, 16) != 0) + tcp_v6_md5_hash_skb(newhash, key.md5_key, NULL, skb); + if (memcmp(md5_hash_location, newhash, 16) != 0) goto out; } #endif From e29bbd73ad7129adfeba56d28871e56e637aa7a4 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Wed, 15 Oct 2025 23:32:05 +0100 Subject: [PATCH 0107/1381] net: dsa: lantiq_gswip: support bridge FDB entries on the CPU port Currently, the driver takes the bridge from dsa_port_bridge_dev_get(), which only works for user ports. This is why it has to ignore FDB entries installed on the CPU port. Commit c26933639b54 ("net: dsa: request drivers to perform FDB isolation") introduced the possibility of getting the originating bridge from the passed dsa_db argument, so let's do that instead. This way, we can act on the local FDB entries coming from the bridge. Note that we do not expect FDB events for the DSA_DB_PORT database, because this driver doesn't fulfill the dsa_switch_supports_uc_filtering() requirements. So we can just return -EOPNOTSUPP and expect it will never be triggered. Signed-off-by: Vladimir Oltean Signed-off-by: Daniel Golle Link: https://patch.msgid.link/ed9d847c0356f0fec81422bdad9ebdcc6a59da79.1760566491.git.daniel@makrotopia.org Signed-off-by: Jakub Kicinski --- drivers/net/dsa/lantiq/lantiq_gswip.c | 18 ++++++++++-------- 1 file changed, 10 insertions(+), 8 deletions(-) diff --git a/drivers/net/dsa/lantiq/lantiq_gswip.c b/drivers/net/dsa/lantiq/lantiq_gswip.c index 2169c0814a48..91755a5972fa 100644 --- a/drivers/net/dsa/lantiq/lantiq_gswip.c +++ b/drivers/net/dsa/lantiq/lantiq_gswip.c @@ -1140,9 +1140,9 @@ static void gswip_port_stp_state_set(struct dsa_switch *ds, int port, u8 state) } static int gswip_port_fdb(struct dsa_switch *ds, int port, - const unsigned char *addr, u16 vid, bool add) + struct net_device *bridge, const unsigned char *addr, + u16 vid, bool add) { - struct net_device *bridge = dsa_port_bridge_dev_get(dsa_to_port(ds, port)); struct gswip_priv *priv = ds->priv; struct gswip_pce_table_entry mac_bridge = {0,}; unsigned int max_ports = priv->hw_info->max_ports; @@ -1150,10 +1150,6 @@ static int gswip_port_fdb(struct dsa_switch *ds, int port, int i; int err; - /* Operation not supported on the CPU port, don't throw errors */ - if (!bridge) - return 0; - for (i = max_ports; i < ARRAY_SIZE(priv->vlans); i++) { if (priv->vlans[i].bridge == bridge) { fid = priv->vlans[i].fid; @@ -1188,14 +1184,20 @@ static int gswip_port_fdb_add(struct dsa_switch *ds, int port, const unsigned char *addr, u16 vid, struct dsa_db db) { - return gswip_port_fdb(ds, port, addr, vid, true); + if (db.type != DSA_DB_BRIDGE) + return -EOPNOTSUPP; + + return gswip_port_fdb(ds, port, db.bridge.dev, addr, vid, true); } static int gswip_port_fdb_del(struct dsa_switch *ds, int port, const unsigned char *addr, u16 vid, struct dsa_db db) { - return gswip_port_fdb(ds, port, addr, vid, false); + if (db.type != DSA_DB_BRIDGE) + return -EOPNOTSUPP; + + return gswip_port_fdb(ds, port, db.bridge.dev, addr, vid, false); } static int gswip_port_fdb_dump(struct dsa_switch *ds, int port, From 92790e6c11a882cc8d49e05da5ed39fb5475a5ae Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Wed, 15 Oct 2025 23:32:15 +0100 Subject: [PATCH 0108/1381] net: dsa: lantiq_gswip: define VLAN ID 0 constant This patch adds an explicit definition for VID 0 to the Lantiq GSWIP DSA driver, clarifying its special meaning. Signed-off-by: Vladimir Oltean Signed-off-by: Daniel Golle Link: https://patch.msgid.link/e8862239d0bb727723cf60947d2262473b46c96d.1760566491.git.daniel@makrotopia.org Signed-off-by: Jakub Kicinski --- drivers/net/dsa/lantiq/lantiq_gswip.c | 12 +++++++----- drivers/net/dsa/lantiq/lantiq_gswip.h | 2 ++ 2 files changed, 9 insertions(+), 5 deletions(-) diff --git a/drivers/net/dsa/lantiq/lantiq_gswip.c b/drivers/net/dsa/lantiq/lantiq_gswip.c index 91755a5972fa..9526317443a1 100644 --- a/drivers/net/dsa/lantiq/lantiq_gswip.c +++ b/drivers/net/dsa/lantiq/lantiq_gswip.c @@ -432,7 +432,7 @@ static int gswip_add_single_port_br(struct gswip_priv *priv, int port, bool add) vlan_active.index = port + 1; vlan_active.table = GSWIP_TABLE_ACTIVE_VLAN; - vlan_active.key[0] = 0; /* vid */ + vlan_active.key[0] = GSWIP_VLAN_UNAWARE_PVID; vlan_active.val[0] = port + 1 /* fid */; vlan_active.valid = add; err = gswip_pce_table_entry_write(priv, &vlan_active); @@ -446,7 +446,7 @@ static int gswip_add_single_port_br(struct gswip_priv *priv, int port, bool add) vlan_mapping.index = port + 1; vlan_mapping.table = GSWIP_TABLE_VLAN_MAPPING; - vlan_mapping.val[0] = 0 /* vid */; + vlan_mapping.val[0] = GSWIP_VLAN_UNAWARE_PVID; vlan_mapping.val[1] = BIT(port) | dsa_cpu_ports(priv->ds); vlan_mapping.val[2] = 0; err = gswip_pce_table_entry_write(priv, &vlan_mapping); @@ -772,7 +772,8 @@ static int gswip_vlan_add_unaware(struct gswip_priv *priv, * entry in a free slot and prepare the VLAN mapping table entry. */ if (idx == -1) { - idx = gswip_vlan_active_create(priv, bridge, -1, 0); + idx = gswip_vlan_active_create(priv, bridge, -1, + GSWIP_VLAN_UNAWARE_PVID); if (idx < 0) return idx; active_vlan_created = true; @@ -780,7 +781,7 @@ static int gswip_vlan_add_unaware(struct gswip_priv *priv, vlan_mapping.index = idx; vlan_mapping.table = GSWIP_TABLE_VLAN_MAPPING; /* VLAN ID byte, maps to the VLAN ID of vlan active table */ - vlan_mapping.val[0] = 0; + vlan_mapping.val[0] = GSWIP_VLAN_UNAWARE_PVID; } else { /* Read the existing VLAN mapping entry from the switch */ vlan_mapping.index = idx; @@ -977,7 +978,8 @@ static void gswip_port_bridge_leave(struct dsa_switch *ds, int port, * specific bridges. No bridge is configured here. */ if (!br_vlan_enabled(br)) - gswip_vlan_remove(priv, br, port, 0, true, false); + gswip_vlan_remove(priv, br, port, GSWIP_VLAN_UNAWARE_PVID, true, + false); } static int gswip_port_vlan_prepare(struct dsa_switch *ds, int port, diff --git a/drivers/net/dsa/lantiq/lantiq_gswip.h b/drivers/net/dsa/lantiq/lantiq_gswip.h index 2df9c8e8cfd0..6aae1ff2f130 100644 --- a/drivers/net/dsa/lantiq/lantiq_gswip.h +++ b/drivers/net/dsa/lantiq/lantiq_gswip.h @@ -222,6 +222,8 @@ */ #define GSWIP_MAX_PACKET_LENGTH 2400 +#define GSWIP_VLAN_UNAWARE_PVID 0 + struct gswip_pce_microcode { u16 val_3; u16 val_2; From 8f5c71e44413ac1c8dcc162cd4769645eff80590 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Wed, 15 Oct 2025 23:32:30 +0100 Subject: [PATCH 0109/1381] net: dsa: lantiq_gswip: remove duplicate assignment to vlan_mapping.val[0] When idx == -1 in gswip_vlan_add(), we set vlan_mapping.val[0] = vid, even though we do the exact same thing again outside the if/else block. Remove the duplicate assignment. Signed-off-by: Vladimir Oltean Signed-off-by: Daniel Golle Link: https://patch.msgid.link/039ecb48e038cea856a9a6230ad1543db2bc382d.1760566491.git.daniel@makrotopia.org Signed-off-by: Jakub Kicinski --- drivers/net/dsa/lantiq/lantiq_gswip.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/net/dsa/lantiq/lantiq_gswip.c b/drivers/net/dsa/lantiq/lantiq_gswip.c index 9526317443a1..e41d67ea89c5 100644 --- a/drivers/net/dsa/lantiq/lantiq_gswip.c +++ b/drivers/net/dsa/lantiq/lantiq_gswip.c @@ -848,8 +848,6 @@ static int gswip_vlan_add_aware(struct gswip_priv *priv, vlan_mapping.index = idx; vlan_mapping.table = GSWIP_TABLE_VLAN_MAPPING; - /* VLAN ID byte, maps to the VLAN ID of vlan active table */ - vlan_mapping.val[0] = vid; } else { /* Read the existing VLAN mapping entry from the switch */ vlan_mapping.index = idx; @@ -862,6 +860,7 @@ static int gswip_vlan_add_aware(struct gswip_priv *priv, } } + /* VLAN ID byte, maps to the VLAN ID of vlan active table */ vlan_mapping.val[0] = vid; /* Update the VLAN mapping entry and write it to the switch */ vlan_mapping.val[1] |= cpu_ports; From b92068755ee0359eb009ae7584145fa3bab51cce Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Wed, 15 Oct 2025 23:32:41 +0100 Subject: [PATCH 0110/1381] net: dsa: lantiq_gswip: merge gswip_vlan_add_unaware() and gswip_vlan_add_aware() The two functions largely duplicate functionality. The differences consist in: - the "fid" passed to gswip_vlan_active_create(). The unaware variant always passes -1, the aware variant passes fid = priv->vlans[i].fid, where i is an index into priv->vlans[] for which priv->vlans[i].bridge is equal to the given bridge. - the "vid" is not passed to gswip_vlan_add_unaware(). It is implicitly GSWIP_VLAN_UNAWARE_PVID (zero). - The "untagged" is not passed to gswip_vlan_add_unaware(). It is implicitly true. Also, the CPU port must not be a tag member of the PVID used for VLAN-unaware bridging. - The "pvid" is not passed to gswip_vlan_add_unaware(). It is implicitly true. - The GSWIP_PCE_DEFPVID(port) register is written by the aware variant with an "idx", but with a hardcoded 0 by the unaware variant. Merge the two functions into a single unified function without any functional changes. Signed-off-by: Vladimir Oltean Signed-off-by: Daniel Golle Link: https://patch.msgid.link/2be190701d4c17038ce4b8047f9fb0bdf8abdf6e.1760566491.git.daniel@makrotopia.org Signed-off-by: Jakub Kicinski --- drivers/net/dsa/lantiq/lantiq_gswip.c | 94 +++++---------------------- 1 file changed, 17 insertions(+), 77 deletions(-) diff --git a/drivers/net/dsa/lantiq/lantiq_gswip.c b/drivers/net/dsa/lantiq/lantiq_gswip.c index e41d67ea89c5..6cbcb54a5ed0 100644 --- a/drivers/net/dsa/lantiq/lantiq_gswip.c +++ b/drivers/net/dsa/lantiq/lantiq_gswip.c @@ -750,86 +750,25 @@ static int gswip_vlan_active_remove(struct gswip_priv *priv, int idx) return err; } -static int gswip_vlan_add_unaware(struct gswip_priv *priv, - struct net_device *bridge, int port) -{ - struct gswip_pce_table_entry vlan_mapping = {0,}; - unsigned int max_ports = priv->hw_info->max_ports; - bool active_vlan_created = false; - int idx = -1; - int i; - int err; - - /* Check if there is already a page for this bridge */ - for (i = max_ports; i < ARRAY_SIZE(priv->vlans); i++) { - if (priv->vlans[i].bridge == bridge) { - idx = i; - break; - } - } - - /* If this bridge is not programmed yet, add a Active VLAN table - * entry in a free slot and prepare the VLAN mapping table entry. - */ - if (idx == -1) { - idx = gswip_vlan_active_create(priv, bridge, -1, - GSWIP_VLAN_UNAWARE_PVID); - if (idx < 0) - return idx; - active_vlan_created = true; - - vlan_mapping.index = idx; - vlan_mapping.table = GSWIP_TABLE_VLAN_MAPPING; - /* VLAN ID byte, maps to the VLAN ID of vlan active table */ - vlan_mapping.val[0] = GSWIP_VLAN_UNAWARE_PVID; - } else { - /* Read the existing VLAN mapping entry from the switch */ - vlan_mapping.index = idx; - vlan_mapping.table = GSWIP_TABLE_VLAN_MAPPING; - err = gswip_pce_table_entry_read(priv, &vlan_mapping); - if (err) { - dev_err(priv->dev, "failed to read VLAN mapping: %d\n", - err); - return err; - } - } - - /* Update the VLAN mapping entry and write it to the switch */ - vlan_mapping.val[1] |= dsa_cpu_ports(priv->ds); - vlan_mapping.val[1] |= BIT(port); - err = gswip_pce_table_entry_write(priv, &vlan_mapping); - if (err) { - dev_err(priv->dev, "failed to write VLAN mapping: %d\n", err); - /* In case an Active VLAN was creaetd delete it again */ - if (active_vlan_created) - gswip_vlan_active_remove(priv, idx); - return err; - } - - gswip_switch_w(priv, 0, GSWIP_PCE_DEFPVID(port)); - return 0; -} - -static int gswip_vlan_add_aware(struct gswip_priv *priv, - struct net_device *bridge, int port, - u16 vid, bool untagged, - bool pvid) +static int gswip_vlan_add(struct gswip_priv *priv, struct net_device *bridge, + int port, u16 vid, bool untagged, bool pvid, + bool vlan_aware) { struct gswip_pce_table_entry vlan_mapping = {0,}; unsigned int max_ports = priv->hw_info->max_ports; unsigned int cpu_ports = dsa_cpu_ports(priv->ds); bool active_vlan_created = false; - int idx = -1; - int fid = -1; - int i; - int err; + int fid = -1, idx = -1; + int i, err; /* Check if there is already a page for this bridge */ for (i = max_ports; i < ARRAY_SIZE(priv->vlans); i++) { if (priv->vlans[i].bridge == bridge) { - if (fid != -1 && fid != priv->vlans[i].fid) - dev_err(priv->dev, "one bridge with multiple flow ids\n"); - fid = priv->vlans[i].fid; + if (vlan_aware) { + if (fid != -1 && fid != priv->vlans[i].fid) + dev_err(priv->dev, "one bridge with multiple flow ids\n"); + fid = priv->vlans[i].fid; + } if (priv->vlans[i].vid == vid) { idx = i; break; @@ -864,8 +803,9 @@ static int gswip_vlan_add_aware(struct gswip_priv *priv, vlan_mapping.val[0] = vid; /* Update the VLAN mapping entry and write it to the switch */ vlan_mapping.val[1] |= cpu_ports; - vlan_mapping.val[2] |= cpu_ports; vlan_mapping.val[1] |= BIT(port); + if (vlan_aware) + vlan_mapping.val[2] |= cpu_ports; if (untagged) vlan_mapping.val[2] &= ~BIT(port); else @@ -879,8 +819,7 @@ static int gswip_vlan_add_aware(struct gswip_priv *priv, return err; } - if (pvid) - gswip_switch_w(priv, idx, GSWIP_PCE_DEFPVID(port)); + gswip_switch_w(priv, vlan_aware ? idx : 0, GSWIP_PCE_DEFPVID(port)); return 0; } @@ -955,7 +894,8 @@ static int gswip_port_bridge_join(struct dsa_switch *ds, int port, * specific bridges. No bridge is configured here. */ if (!br_vlan_enabled(br)) { - err = gswip_vlan_add_unaware(priv, br, port); + err = gswip_vlan_add(priv, br, port, GSWIP_VLAN_UNAWARE_PVID, + true, true, false); if (err) return err; priv->port_vlan_filter &= ~BIT(port); @@ -1049,8 +989,8 @@ static int gswip_port_vlan_add(struct dsa_switch *ds, int port, if (dsa_is_cpu_port(ds, port)) return 0; - return gswip_vlan_add_aware(priv, bridge, port, vlan->vid, - untagged, pvid); + return gswip_vlan_add(priv, bridge, port, vlan->vid, untagged, pvid, + true); } static int gswip_port_vlan_del(struct dsa_switch *ds, int port, From 21c3237c60c3694ba6afc582c6b0568905d1dca8 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Wed, 15 Oct 2025 23:32:50 +0100 Subject: [PATCH 0111/1381] net: dsa: lantiq_gswip: remove legacy configure_vlan_while_not_filtering option This driver doesn't support dynamic VLAN filtering changes, for simplicity. It expects that on a port, either gswip_vlan_add_unaware() or gswip_vlan_add_aware() is called, but not both. When !br_vlan_enabled(), the configure_vlan_while_not_filtering = false option is exactly what will prevent calls to gswip_port_vlan_add() from being issued by DSA. In fact, at the time these features were submitted: https://patchwork.ozlabs.org/project/netdev/patch/20190501204506.21579-3-hauke@hauke-m.de/ "configure_vlan_while_not_filtering = false" did not even have a name, it was implicit behaviour. It only became legacy in commit 54a0ed0df496 ("net: dsa: provide an option for drivers to always receive bridge VLANs"). Section "Bridge VLAN filtering" of Documentation/networking/switchdev.rst describes the exact set of rules. Notably, the PVID of the port must follow the VLAN awareness state of the bridge port. A VLAN-unaware bridge port should not respond to the addition of a bridge VLAN with the PVID flag. In fact, the pvid_change() test in tools/testing/selftests/net/forwarding/bridge_vlan_unaware.sh tests exactly this. The lantiq_gswip driver indeed does not respond to the addition of PVID VLANs while VLAN-unaware in the way described above, but only because of configure_vlan_while_not_filtering. Our purpose here is to get rid of configure_vlan_while_not_filtering, so we must add more complex logic which follows the VLAN awareness state and walks through the Active VLAN table entries, to find the index of the PVID register that should be committed to hardware on each port. As a side-effect of now having a proper implementation to assign the PVID all the "VLAN upper: ..." tests of the local_termination.sh self- tests which would previously all FAIL now all PASS (or XFAIL, but that's ok). Signed-off-by: Vladimir Oltean Signed-off-by: Daniel Golle Tested-by: Daniel Golle Link: https://patch.msgid.link/47dab8a8b69ebb92624b9795b723114475d3fe4e.1760566491.git.daniel@makrotopia.org Signed-off-by: Jakub Kicinski --- drivers/net/dsa/lantiq/lantiq_gswip.c | 49 +++++++++++++++++++++++---- 1 file changed, 43 insertions(+), 6 deletions(-) diff --git a/drivers/net/dsa/lantiq/lantiq_gswip.c b/drivers/net/dsa/lantiq/lantiq_gswip.c index 6cbcb54a5ed0..30cff623bec0 100644 --- a/drivers/net/dsa/lantiq/lantiq_gswip.c +++ b/drivers/net/dsa/lantiq/lantiq_gswip.c @@ -547,6 +547,45 @@ static int gswip_pce_load_microcode(struct gswip_priv *priv) return 0; } +static void gswip_port_commit_pvid(struct gswip_priv *priv, int port) +{ + struct dsa_port *dp = dsa_to_port(priv->ds, port); + struct net_device *br = dsa_port_bridge_dev_get(dp); + int idx; + + if (!dsa_port_is_user(dp)) + return; + + if (br) { + u16 pvid = GSWIP_VLAN_UNAWARE_PVID; + + if (br_vlan_enabled(br)) + br_vlan_get_pvid(br, &pvid); + + /* VLAN-aware bridge ports with no PVID will use Active VLAN + * index 0. The expectation is that this drops all untagged and + * VID-0 tagged ingress traffic. + */ + idx = 0; + for (int i = priv->hw_info->max_ports; + i < ARRAY_SIZE(priv->vlans); i++) { + if (priv->vlans[i].bridge == br && + priv->vlans[i].vid == pvid) { + idx = i; + break; + } + } + } else { + /* The Active VLAN table index as configured by + * gswip_add_single_port_br() + */ + idx = port + 1; + } + + /* GSWIP 2.2 (GRX300) and later program here the VID directly. */ + gswip_switch_w(priv, idx, GSWIP_PCE_DEFPVID(port)); +} + static int gswip_port_vlan_filtering(struct dsa_switch *ds, int port, bool vlan_filtering, struct netlink_ext_ack *extack) @@ -581,6 +620,8 @@ static int gswip_port_vlan_filtering(struct dsa_switch *ds, int port, GSWIP_PCE_PCTRL_0p(port)); } + gswip_port_commit_pvid(priv, port); + return 0; } @@ -677,8 +718,6 @@ static int gswip_setup(struct dsa_switch *ds) ds->mtu_enforcement_ingress = true; - ds->configure_vlan_while_not_filtering = false; - return 0; } @@ -819,7 +858,7 @@ static int gswip_vlan_add(struct gswip_priv *priv, struct net_device *bridge, return err; } - gswip_switch_w(priv, vlan_aware ? idx : 0, GSWIP_PCE_DEFPVID(port)); + gswip_port_commit_pvid(priv, port); return 0; } @@ -874,9 +913,7 @@ static int gswip_vlan_remove(struct gswip_priv *priv, } } - /* GSWIP 2.2 (GRX300) and later program here the VID directly. */ - if (pvid) - gswip_switch_w(priv, 0, GSWIP_PCE_DEFPVID(port)); + gswip_port_commit_pvid(priv, port); return 0; } From ab3ce58559d6227a8a54dc78650f5763768a1cfc Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Wed, 15 Oct 2025 23:32:58 +0100 Subject: [PATCH 0112/1381] net: dsa: lantiq_gswip: permit dynamic changes to VLAN filtering state The driver should now tolerate these changes, now that the PVID is automatically recalculated on a VLAN awareness state change. The VLAN-unaware PVID must be installed to hardware even if the joined bridge is currently VLAN-aware. Otherwise, when the bridge VLAN filtering state dynamically changes to VLAN-unaware later, this PVID will be missing. Signed-off-by: Vladimir Oltean Signed-off-by: Daniel Golle Link: https://patch.msgid.link/c58759074fb699581336dc2c2c6bf106257b134e.1760566491.git.daniel@makrotopia.org Signed-off-by: Jakub Kicinski --- drivers/net/dsa/lantiq/lantiq_gswip.c | 38 +++++++++------------------ drivers/net/dsa/lantiq/lantiq_gswip.h | 1 - 2 files changed, 12 insertions(+), 27 deletions(-) diff --git a/drivers/net/dsa/lantiq/lantiq_gswip.c b/drivers/net/dsa/lantiq/lantiq_gswip.c index 30cff623bec0..58fdd54094d6 100644 --- a/drivers/net/dsa/lantiq/lantiq_gswip.c +++ b/drivers/net/dsa/lantiq/lantiq_gswip.c @@ -590,16 +590,8 @@ static int gswip_port_vlan_filtering(struct dsa_switch *ds, int port, bool vlan_filtering, struct netlink_ext_ack *extack) { - struct net_device *bridge = dsa_port_bridge_dev_get(dsa_to_port(ds, port)); struct gswip_priv *priv = ds->priv; - /* Do not allow changing the VLAN filtering options while in bridge */ - if (bridge && !!(priv->port_vlan_filter & BIT(port)) != vlan_filtering) { - NL_SET_ERR_MSG_MOD(extack, - "Dynamic toggling of vlan_filtering not supported"); - return -EIO; - } - if (vlan_filtering) { /* Use tag based VLAN */ gswip_switch_mask(priv, @@ -927,18 +919,15 @@ static int gswip_port_bridge_join(struct dsa_switch *ds, int port, struct gswip_priv *priv = ds->priv; int err; - /* When the bridge uses VLAN filtering we have to configure VLAN - * specific bridges. No bridge is configured here. + /* Set up the VLAN for VLAN-unaware bridging for this port, and remove + * it from the "single-port bridge" through which it was operating as + * standalone. */ - if (!br_vlan_enabled(br)) { - err = gswip_vlan_add(priv, br, port, GSWIP_VLAN_UNAWARE_PVID, - true, true, false); - if (err) - return err; - priv->port_vlan_filter &= ~BIT(port); - } else { - priv->port_vlan_filter |= BIT(port); - } + err = gswip_vlan_add(priv, br, port, GSWIP_VLAN_UNAWARE_PVID, + true, true, false); + if (err) + return err; + return gswip_add_single_port_br(priv, port, false); } @@ -948,14 +937,11 @@ static void gswip_port_bridge_leave(struct dsa_switch *ds, int port, struct net_device *br = bridge.dev; struct gswip_priv *priv = ds->priv; - gswip_add_single_port_br(priv, port, true); - - /* When the bridge uses VLAN filtering we have to configure VLAN - * specific bridges. No bridge is configured here. + /* Add the port back to the "single-port bridge", and remove it from + * the VLAN-unaware PVID created for this bridge. */ - if (!br_vlan_enabled(br)) - gswip_vlan_remove(priv, br, port, GSWIP_VLAN_UNAWARE_PVID, true, - false); + gswip_add_single_port_br(priv, port, true); + gswip_vlan_remove(priv, br, port, GSWIP_VLAN_UNAWARE_PVID, true, false); } static int gswip_port_vlan_prepare(struct dsa_switch *ds, int port, diff --git a/drivers/net/dsa/lantiq/lantiq_gswip.h b/drivers/net/dsa/lantiq/lantiq_gswip.h index 6aae1ff2f130..4590a1a7dbd9 100644 --- a/drivers/net/dsa/lantiq/lantiq_gswip.h +++ b/drivers/net/dsa/lantiq/lantiq_gswip.h @@ -270,7 +270,6 @@ struct gswip_priv { struct gswip_vlan vlans[64]; int num_gphy_fw; struct gswip_gphy_fw *gphy_fw; - u32 port_vlan_filter; struct mutex pce_table_lock; u16 version; }; From 96a91e6eeb4d7881454071ecd1443f025cc21c3b Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Wed, 15 Oct 2025 23:33:25 +0100 Subject: [PATCH 0113/1381] net: dsa: lantiq_gswip: disallow changes to privately set up VID 0 User space can force the altering of VID 0 as it was privately set up by this driver. For example, when the port joins a VLAN-aware bridge, dsa_user_manage_vlan_filtering() will set NETIF_F_HW_VLAN_CTAG_FILTER. If the port is subsequently brought up and CONFIG_VLAN_8021Q is enabled, the vlan_vid0_add() function will want to make sure we are capable of accepting packets tagged with VID 0. Generally, DSA/switchdev drivers want to suppress that bit of help from the 8021q layer, and handle VID 0 filters themselves. The 8021q layer might actually be even detrimential, because VLANs added through vlan_vid_add() pass through dsa_user_vlan_rx_add_vid(), which is documented as this: /* This API only allows programming tagged, non-PVID VIDs */ .flags = 0, so it will force VID 0 to be reconfigured as egress-tagged, non-PVID. Whereas the driver configures it as PVID and egress-untagged, the exact opposite. Signed-off-by: Vladimir Oltean Signed-off-by: Daniel Golle Link: https://patch.msgid.link/9f68340c34b5312c3b8c6c7ecf3cfce574a3f65d.1760566491.git.daniel@makrotopia.org Signed-off-by: Jakub Kicinski --- drivers/net/dsa/lantiq/lantiq_gswip.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/net/dsa/lantiq/lantiq_gswip.c b/drivers/net/dsa/lantiq/lantiq_gswip.c index 58fdd54094d6..26e963840f3b 100644 --- a/drivers/net/dsa/lantiq/lantiq_gswip.c +++ b/drivers/net/dsa/lantiq/lantiq_gswip.c @@ -1000,6 +1000,9 @@ static int gswip_port_vlan_add(struct dsa_switch *ds, int port, bool pvid = vlan->flags & BRIDGE_VLAN_INFO_PVID; int err; + if (vlan->vid == GSWIP_VLAN_UNAWARE_PVID) + return 0; + err = gswip_port_vlan_prepare(ds, port, vlan, extack); if (err) return err; @@ -1023,6 +1026,9 @@ static int gswip_port_vlan_del(struct dsa_switch *ds, int port, struct gswip_priv *priv = ds->priv; bool pvid = vlan->flags & BRIDGE_VLAN_INFO_PVID; + if (vlan->vid == GSWIP_VLAN_UNAWARE_PVID) + return 0; + /* We have to receive all packets on the CPU port and should not * do any VLAN filtering here. This is also called with bridge * NULL and then we do not know for which bridge to configure From 7ed1965f10100a0928e8d88b205273240d597d95 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Wed, 15 Oct 2025 23:33:33 +0100 Subject: [PATCH 0114/1381] net: dsa: lantiq_gswip: remove vlan_aware and pvid arguments from gswip_vlan_remove() "bool pvid" is unused since commit "net: dsa: lantiq_gswip: remove legacy configure_vlan_while_not_filtering option". "bool vlan_aware" shouldn't have a role in finding the bridge VLAN. It should be identified by VID regardless of VLAN-aware or VLAN-unaware. The driver sets up VID 0 for the VLAN-unaware PVID. Signed-off-by: Vladimir Oltean Signed-off-by: Daniel Golle Link: https://patch.msgid.link/c63f89ca19269ef6c8bf00a62cacc739164b4441.1760566491.git.daniel@makrotopia.org Signed-off-by: Jakub Kicinski --- drivers/net/dsa/lantiq/lantiq_gswip.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) diff --git a/drivers/net/dsa/lantiq/lantiq_gswip.c b/drivers/net/dsa/lantiq/lantiq_gswip.c index 26e963840f3b..d9a7a004f9eb 100644 --- a/drivers/net/dsa/lantiq/lantiq_gswip.c +++ b/drivers/net/dsa/lantiq/lantiq_gswip.c @@ -857,7 +857,7 @@ static int gswip_vlan_add(struct gswip_priv *priv, struct net_device *bridge, static int gswip_vlan_remove(struct gswip_priv *priv, struct net_device *bridge, int port, - u16 vid, bool pvid, bool vlan_aware) + u16 vid) { struct gswip_pce_table_entry vlan_mapping = {0,}; unsigned int max_ports = priv->hw_info->max_ports; @@ -868,7 +868,7 @@ static int gswip_vlan_remove(struct gswip_priv *priv, /* Check if there is already a page for this bridge */ for (i = max_ports; i < ARRAY_SIZE(priv->vlans); i++) { if (priv->vlans[i].bridge == bridge && - (!vlan_aware || priv->vlans[i].vid == vid)) { + priv->vlans[i].vid == vid) { idx = i; break; } @@ -941,7 +941,7 @@ static void gswip_port_bridge_leave(struct dsa_switch *ds, int port, * the VLAN-unaware PVID created for this bridge. */ gswip_add_single_port_br(priv, port, true); - gswip_vlan_remove(priv, br, port, GSWIP_VLAN_UNAWARE_PVID, true, false); + gswip_vlan_remove(priv, br, port, GSWIP_VLAN_UNAWARE_PVID); } static int gswip_port_vlan_prepare(struct dsa_switch *ds, int port, @@ -1024,7 +1024,6 @@ static int gswip_port_vlan_del(struct dsa_switch *ds, int port, { struct net_device *bridge = dsa_port_bridge_dev_get(dsa_to_port(ds, port)); struct gswip_priv *priv = ds->priv; - bool pvid = vlan->flags & BRIDGE_VLAN_INFO_PVID; if (vlan->vid == GSWIP_VLAN_UNAWARE_PVID) return 0; @@ -1037,7 +1036,7 @@ static int gswip_port_vlan_del(struct dsa_switch *ds, int port, if (dsa_is_cpu_port(ds, port)) return 0; - return gswip_vlan_remove(priv, bridge, port, vlan->vid, pvid, true); + return gswip_vlan_remove(priv, bridge, port, vlan->vid); } static void gswip_port_fast_age(struct dsa_switch *ds, int port) From a57627626636c20399f0c45ea1d16e6283affc91 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Wed, 15 Oct 2025 23:33:41 +0100 Subject: [PATCH 0115/1381] net: dsa: lantiq_gswip: put a more descriptive error print in gswip_vlan_remove() Improve the error message printed in case of a port VLAN entry not being found upon removal. Signed-off-by: Vladimir Oltean Signed-off-by: Daniel Golle Link: https://patch.msgid.link/abd4ec58e0f0f53eb3d7027097a20af0bd7b1d6d.1760566491.git.daniel@makrotopia.org Signed-off-by: Jakub Kicinski --- drivers/net/dsa/lantiq/lantiq_gswip.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/dsa/lantiq/lantiq_gswip.c b/drivers/net/dsa/lantiq/lantiq_gswip.c index d9a7a004f9eb..cfdeb8148500 100644 --- a/drivers/net/dsa/lantiq/lantiq_gswip.c +++ b/drivers/net/dsa/lantiq/lantiq_gswip.c @@ -875,7 +875,8 @@ static int gswip_vlan_remove(struct gswip_priv *priv, } if (idx == -1) { - dev_err(priv->dev, "bridge to leave does not exists\n"); + dev_err(priv->dev, "Port %d cannot find VID %u of bridge %s\n", + port, vid, bridge ? bridge->name : "(null)"); return -ENOENT; } From 3bb500caf656b918bddd7e32dba7ed0e5c1c9598 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Wed, 15 Oct 2025 23:33:50 +0100 Subject: [PATCH 0116/1381] net: dsa: lantiq_gswip: drop untagged on VLAN-aware bridge ports with no PVID Implement the required functionality, as written in Documentation/networking/switchdev.rst section "Bridge VLAN filtering", by using the "VLAN Ingress Tag Rule" feature of the switch. The bit field definitions for this were found while browsing the Intel dual BSD/GPLv2 licensed drivers for this switch IP. Signed-off-by: Vladimir Oltean Signed-off-by: Daniel Golle Link: https://patch.msgid.link/787aa807d00b726d75db2a40add215c8b8ba7466.1760566491.git.daniel@makrotopia.org Signed-off-by: Jakub Kicinski --- drivers/net/dsa/lantiq/lantiq_gswip.c | 6 ++++++ drivers/net/dsa/lantiq/lantiq_gswip.h | 4 ++++ 2 files changed, 10 insertions(+) diff --git a/drivers/net/dsa/lantiq/lantiq_gswip.c b/drivers/net/dsa/lantiq/lantiq_gswip.c index cfdeb8148500..1ff0932dae31 100644 --- a/drivers/net/dsa/lantiq/lantiq_gswip.c +++ b/drivers/net/dsa/lantiq/lantiq_gswip.c @@ -551,6 +551,7 @@ static void gswip_port_commit_pvid(struct gswip_priv *priv, int port) { struct dsa_port *dp = dsa_to_port(priv->ds, port); struct net_device *br = dsa_port_bridge_dev_get(dp); + u32 vinr; int idx; if (!dsa_port_is_user(dp)) @@ -582,6 +583,11 @@ static void gswip_port_commit_pvid(struct gswip_priv *priv, int port) idx = port + 1; } + vinr = idx ? GSWIP_PCE_VCTRL_VINR_ALL : GSWIP_PCE_VCTRL_VINR_TAGGED; + gswip_switch_mask(priv, GSWIP_PCE_VCTRL_VINR, + FIELD_PREP(GSWIP_PCE_VCTRL_VINR, vinr), + GSWIP_PCE_VCTRL(port)); + /* GSWIP 2.2 (GRX300) and later program here the VID directly. */ gswip_switch_w(priv, idx, GSWIP_PCE_DEFPVID(port)); } diff --git a/drivers/net/dsa/lantiq/lantiq_gswip.h b/drivers/net/dsa/lantiq/lantiq_gswip.h index 4590a1a7dbd9..69c8d2deff2d 100644 --- a/drivers/net/dsa/lantiq/lantiq_gswip.h +++ b/drivers/net/dsa/lantiq/lantiq_gswip.h @@ -159,6 +159,10 @@ #define GSWIP_PCE_PCTRL_0_PSTATE_MASK GENMASK(2, 0) #define GSWIP_PCE_VCTRL(p) (0x485 + ((p) * 0xA)) #define GSWIP_PCE_VCTRL_UVR BIT(0) /* Unknown VLAN Rule */ +#define GSWIP_PCE_VCTRL_VINR GENMASK(2, 1) /* VLAN Ingress Tag Rule */ +#define GSWIP_PCE_VCTRL_VINR_ALL 0 /* Admit tagged and untagged packets */ +#define GSWIP_PCE_VCTRL_VINR_TAGGED 1 /* Admit only tagged packets */ +#define GSWIP_PCE_VCTRL_VINR_UNTAGGED 2 /* Admit only untagged packets */ #define GSWIP_PCE_VCTRL_VIMR BIT(3) /* VLAN Ingress Member violation rule */ #define GSWIP_PCE_VCTRL_VEMR BIT(4) /* VLAN Egress Member violation rule */ #define GSWIP_PCE_VCTRL_VSR BIT(5) /* VLAN Security */ From 1f89ed0ebf2696d1d8fa7625e26c692aa153774a Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Wed, 15 Oct 2025 23:34:01 +0100 Subject: [PATCH 0117/1381] net: dsa: lantiq_gswip: treat VID 0 like the PVID Documentation/networking/switchdev.rst says that VLAN-aware bridges must treat packets tagged with VID 0 the same as untagged. It appears from the documentation that setting the GSWIP_PCE_VCTRL_VID0 flag (which this driver already had defined) might achieve this. Signed-off-by: Vladimir Oltean Signed-off-by: Daniel Golle Link: https://patch.msgid.link/b220ac149922839a261b754202c05df5bb253c98.1760566491.git.daniel@makrotopia.org Signed-off-by: Jakub Kicinski --- drivers/net/dsa/lantiq/lantiq_gswip.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/dsa/lantiq/lantiq_gswip.c b/drivers/net/dsa/lantiq/lantiq_gswip.c index 1ff0932dae31..25f6b46957a0 100644 --- a/drivers/net/dsa/lantiq/lantiq_gswip.c +++ b/drivers/net/dsa/lantiq/lantiq_gswip.c @@ -603,7 +603,7 @@ static int gswip_port_vlan_filtering(struct dsa_switch *ds, int port, gswip_switch_mask(priv, GSWIP_PCE_VCTRL_VSR, GSWIP_PCE_VCTRL_UVR | GSWIP_PCE_VCTRL_VIMR | - GSWIP_PCE_VCTRL_VEMR, + GSWIP_PCE_VCTRL_VEMR | GSWIP_PCE_VCTRL_VID0, GSWIP_PCE_VCTRL(port)); gswip_switch_mask(priv, GSWIP_PCE_PCTRL_0_TVM, 0, GSWIP_PCE_PCTRL_0p(port)); @@ -611,7 +611,7 @@ static int gswip_port_vlan_filtering(struct dsa_switch *ds, int port, /* Use port based VLAN */ gswip_switch_mask(priv, GSWIP_PCE_VCTRL_UVR | GSWIP_PCE_VCTRL_VIMR | - GSWIP_PCE_VCTRL_VEMR, + GSWIP_PCE_VCTRL_VEMR | GSWIP_PCE_VCTRL_VID0, GSWIP_PCE_VCTRL_VSR, GSWIP_PCE_VCTRL(port)); gswip_switch_mask(priv, 0, GSWIP_PCE_PCTRL_0_TVM, From 38e3a9408496540f3a1dbbfc2ea7e495e14e03d7 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Fri, 3 Oct 2025 10:29:19 +0200 Subject: [PATCH 0118/1381] wifi: ath12k: Add MODULE_FIRMWARE() entries Some systems such as live-image or installer require the firmware information for each module declared by MODULE_FIRMWARE(), which is currently missing in ath12k driver. For addressing it, this patch adds the MODULE_FIRMWARE() entries. Like ath11k driver, we can just put the currently used firmware entries for QCN9274 and WCN7850 with wildcards. Signed-off-by: Takashi Iwai Reviewed-by: Baochen Qiang Reviewed-by: Vasanthakumar Thiagarajan Link: https://patch.msgid.link/20251003082955.11436-1-tiwai@suse.de Signed-off-by: Jeff Johnson --- drivers/net/wireless/ath/ath12k/pci.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/net/wireless/ath/ath12k/pci.c b/drivers/net/wireless/ath/ath12k/pci.c index c729d5526c75..48161db6af57 100644 --- a/drivers/net/wireless/ath/ath12k/pci.c +++ b/drivers/net/wireless/ath/ath12k/pci.c @@ -1871,3 +1871,7 @@ void ath12k_pci_exit(void) { pci_unregister_driver(&ath12k_pci_driver); } + +/* firmware files */ +MODULE_FIRMWARE(ATH12K_FW_DIR "/QCN9274/hw2.0/*"); +MODULE_FIRMWARE(ATH12K_FW_DIR "/WCN7850/hw2.0/*"); From 3dfdc98d1dc29d6339fd0551348df0d968fd9ad4 Mon Sep 17 00:00:00 2001 From: Alok Tiwari Date: Fri, 17 Oct 2025 12:35:20 -0700 Subject: [PATCH 0119/1381] net: phy: micrel: simplify return in ksz9477_phy_errata() ksz9477_phy_errata function currently assigns the return value of genphy_restart_aneg() to a variable and then immediately returns it err = genphy_restart_aneg(phydev); if (err) return err; return err; This can be simplified by directly returning the function call result, as the intermediate variable and conditional are redundant. Signed-off-by: Alok Tiwari Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20251017193525.1457064-1-alok.a.tiwari@oracle.com Signed-off-by: Jakub Kicinski --- drivers/net/phy/micrel.c | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) diff --git a/drivers/net/phy/micrel.c b/drivers/net/phy/micrel.c index 79ce3eb6752b..65994d97c403 100644 --- a/drivers/net/phy/micrel.c +++ b/drivers/net/phy/micrel.c @@ -2095,11 +2095,7 @@ static int ksz9477_phy_errata(struct phy_device *phydev) return err; } - err = genphy_restart_aneg(phydev); - if (err) - return err; - - return err; + return genphy_restart_aneg(phydev); } static int ksz9477_config_init(struct phy_device *phydev) From ba397fde5e998478e4962b009321c184b0060067 Mon Sep 17 00:00:00 2001 From: Alok Tiwari Date: Fri, 17 Oct 2025 12:35:21 -0700 Subject: [PATCH 0120/1381] net: phy: micrel: fix typos in comments Fix several spelling and grammatical errors in comments across micrel PHY drivers. Corrections include: - "dealy" -> "delay" - "autonegotation" -> "autonegotiation" - "recheas" -> "reaches" - "one" -> "on" - "improvenent" -> "improvement" - "intput" -> "input" Signed-off-by: Alok Tiwari Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20251017193525.1457064-2-alok.a.tiwari@oracle.com Signed-off-by: Jakub Kicinski --- drivers/net/phy/micrel.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/drivers/net/phy/micrel.c b/drivers/net/phy/micrel.c index 65994d97c403..5f2c7e5c314f 100644 --- a/drivers/net/phy/micrel.c +++ b/drivers/net/phy/micrel.c @@ -1050,7 +1050,7 @@ static int ksz9021_config_init(struct phy_device *phydev) #define TX_CLK_ID 0x1f /* set tx and tx_clk to "No delay adjustment" to keep 0ns - * dealy + * delay */ #define TX_ND 0x7 #define TX_CLK_ND 0xf @@ -1913,7 +1913,7 @@ static int ksz886x_config_aneg(struct phy_device *phydev) return ret; if (phydev->autoneg != AUTONEG_ENABLE) { - /* When autonegotation is disabled, we need to manually force + /* When autonegotiation is disabled, we need to manually force * the link state. If we don't do this, the PHY will keep * sending Fast Link Pulses (FLPs) which are part of the * autonegotiation process. This is not desired when @@ -3533,7 +3533,7 @@ static void lan8814_ptp_disable_event(struct phy_device *phydev, int event) /* Set target to too far in the future, effectively disabling it */ lan8814_ptp_set_target(phydev, event, 0xFFFFFFFF, 0); - /* And then reload once it recheas the target */ + /* And then reload once it reaches the target */ lanphy_modify_page_reg(phydev, LAN8814_PAGE_COMMON_REGS, LAN8814_PTP_GENERAL_CONFIG, LAN8814_PTP_GENERAL_CONFIG_RELOAD_ADD_X(event), LAN8814_PTP_GENERAL_CONFIG_RELOAD_ADD_X(event)); @@ -4403,7 +4403,7 @@ static int lan8814_release_coma_mode(struct phy_device *phydev) static void lan8814_clear_2psp_bit(struct phy_device *phydev) { /* It was noticed that when traffic is passing through the PHY and the - * cable is removed then the LED was still one even though there is no + * cable is removed then the LED was still on even though there is no * link */ lanphy_modify_page_reg(phydev, LAN8814_PAGE_PCS_DIGITAL, LAN8814_EEE_STATE, @@ -4543,7 +4543,7 @@ static int lan8841_config_init(struct phy_device *phydev) phy_write_mmd(phydev, KSZ9131RN_MMD_COMMON_CTRL_REG, LAN8841_PTP_TX_VERSION, 0xff00); - /* 100BT Clause 40 improvenent errata */ + /* 100BT Clause 40 improvement errata */ phy_write_mmd(phydev, LAN8841_MMD_ANALOG_REG, LAN8841_ANALOG_CONTROL_1, LAN8841_ANALOG_CONTROL_1_PLL_TRIM(0x2)); @@ -5563,7 +5563,7 @@ static int lan8841_ptp_extts_on(struct kszphy_ptp_priv *ptp_priv, int pin, u16 tmp = 0; int ret; - /* Set GPIO to be intput */ + /* Set GPIO to be input */ ret = phy_set_bits_mmd(phydev, 2, LAN8841_GPIO_EN, BIT(pin)); if (ret) return ret; From f8a55d5e71e6bdf4793080e622b92cb2646171b3 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 17 Oct 2025 14:53:34 +0000 Subject: [PATCH 0121/1381] net: add a fast path in __netif_schedule() Cpus serving NIC interrupts and specifically TX completions are often trapped in also restarting a busy qdisc (because qdisc was stopped by BQL or the driver's own flow control). When they call netdev_tx_completed_queue() or netif_tx_wake_queue(), they call __netif_schedule() so that the queue can be run later from net_tx_action() (involving NET_TX_SOFTIRQ) Quite often, by the time the cpu reaches net_tx_action(), another cpu grabbed the qdisc spinlock from __dev_xmit_skb(), and we spend too much time spinning on this lock. We can detect in __netif_schedule() if a cpu is already at a specific point in __dev_xmit_skb() where we have the guarantee the queue will be run. This patch gives a 13 % increase of throughput on an IDPF NIC (200Gbit), 32 TX qeues, sending UDP packets of 120 bytes. This also helps __qdisc_run() to not force a NET_TX_SOFTIRQ if another thread is waiting in __dev_xmit_skb() Before: sar -n DEV 5 5|grep eth1|grep Average Average: eth1 1496.44 52191462.56 210.00 13369396.90 0.00 0.00 0.00 54.76 After: sar -n DEV 5 5|grep eth1|grep Average Average: eth1 1457.88 59363099.96 205.08 15206384.35 0.00 0.00 0.00 62.29 Signed-off-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20251017145334.3016097-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/core/dev.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/net/core/dev.c b/net/core/dev.c index 821e7c718924..9482b905c66a 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3373,6 +3373,13 @@ static void __netif_reschedule(struct Qdisc *q) void __netif_schedule(struct Qdisc *q) { + /* If q->defer_list is not empty, at least one thread is + * in __dev_xmit_skb() before llist_del_all(&q->defer_list). + * This thread will attempt to run the queue. + */ + if (!llist_empty(&q->defer_list)) + return; + if (!test_and_set_bit(__QDISC_STATE_SCHED, &q->state)) __netif_reschedule(q); } From 370157293175a702036203faec3e0495b081f135 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Wed, 15 Oct 2025 20:59:17 -0700 Subject: [PATCH 0122/1381] nl802154: fix some kernel-doc warnings Correct multiple kernel-doc warnings in nl802154.h: - Fix a typo on one enum name to avoid a kernel-doc warning. - Drop 2 enum descriptions that are no longer needed. - Mark 2 internal enums as "private:" so that kernel-doc is not needed for them. Warning: nl802154.h:239 Enum value 'NL802154_CAP_ATTR_MAX_MAXBE' not described in enum 'nl802154_wpan_phy_capability_attr' Warning: nl802154.h:239 Excess enum value '%NL802154_CAP_ATTR_MIN_CCA_ED_LEVEL' description in 'nl802154_wpan_phy_capability_attr' Warning: nl802154.h:239 Excess enum value '%NL802154_CAP_ATTR_MAX_CCA_ED_LEVEL' description in 'nl802154_wpan_phy_capability_attr' Warning: nl802154.h:369 Enum value '__NL802154_CCA_OPT_ATTR_AFTER_LAST' not described in enum 'nl802154_cca_opts' Warning: nl802154.h:369 Enum value 'NL802154_CCA_OPT_ATTR_MAX' not described in enum 'nl802154_cca_opts' Signed-off-by: Randy Dunlap Reviewed-by: Simon Horman Link: https://patch.msgid.link/20251016035917.1148012-1-rdunlap@infradead.org Signed-off-by: Jakub Kicinski --- include/net/nl802154.h | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/include/net/nl802154.h b/include/net/nl802154.h index a994dea74596..442822746e92 100644 --- a/include/net/nl802154.h +++ b/include/net/nl802154.h @@ -191,14 +191,12 @@ enum nl802154_iftype { * @NL802154_CAP_ATTR_CHANNELS: a nested attribute for nl802154_channel_attr * @NL802154_CAP_ATTR_TX_POWERS: a nested attribute for * nl802154_wpan_phy_tx_power - * @NL802154_CAP_ATTR_MIN_CCA_ED_LEVEL: minimum value for cca_ed_level - * @NL802154_CAP_ATTR_MAX_CCA_ED_LEVEL: maximum value for cca_ed_level * @NL802154_CAP_ATTR_CCA_MODES: nl802154_cca_modes flags * @NL802154_CAP_ATTR_CCA_OPTS: nl802154_cca_opts flags * @NL802154_CAP_ATTR_MIN_MINBE: minimum of minbe value * @NL802154_CAP_ATTR_MAX_MINBE: maximum of minbe value * @NL802154_CAP_ATTR_MIN_MAXBE: minimum of maxbe value - * @NL802154_CAP_ATTR_MAX_MINBE: maximum of maxbe value + * @NL802154_CAP_ATTR_MAX_MAXBE: maximum of maxbe value * @NL802154_CAP_ATTR_MIN_CSMA_BACKOFFS: minimum of csma backoff value * @NL802154_CAP_ATTR_MAX_CSMA_BACKOFFS: maximum of csma backoffs value * @NL802154_CAP_ATTR_MIN_FRAME_RETRIES: minimum of frame retries value @@ -364,6 +362,7 @@ enum nl802154_cca_opts { NL802154_CCA_OPT_ENERGY_CARRIER_AND, NL802154_CCA_OPT_ENERGY_CARRIER_OR, + /* private: */ /* keep last */ __NL802154_CCA_OPT_ATTR_AFTER_LAST, NL802154_CCA_OPT_ATTR_MAX = __NL802154_CCA_OPT_ATTR_AFTER_LAST - 1 From 813882ae22756bcf9645d405e045c60e5aab0a93 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Thu, 16 Oct 2025 15:36:46 +0100 Subject: [PATCH 0123/1381] net: stmmac: remove broken PCS code Changing the netif_carrier_*() state behind phylink's back has always been prohibited because it messes up with phylinks state tracking, and means that phylink no longer guarantees to call the mac_link_down() and mac_link_up() methods at the appropriate times. This was later documented in the sfp-phylink network driver conversion guide. stmmac was converted to phylink in 2019, but nothing was done with the "PCS" code. Since then, apart from the updates as part of phylink development, nothing has happened with stmmac to improve its use of phylink, or even to address this point. A couple of years ago, a has_integrated_pcs boolean was added by Bart, which later became the STMMAC_FLAG_HAS_INTEGRATED_PCS flag, to avoid manipulating the netif_carrier_*() state. This flag is mis-named, because whenever the stmmac is synthesized for its native SGMII, TBI or RTBI interfaces, it has an "integrated PCS". This boolean/flag actually means "ignore the status from the integrated PCS". Discussing with Bart, the reasons for this are lost to the winds of time (which is why we should always document the reasons in the commit message.) RGMII also has in-band status, and the dwmac cores and stmmac code supports this but with one bug that saves the day. When dwmac cores are synthesised for RGMII only, they do not contain an integrated PCS, and so priv->dma_cap.pcs is clear, which prevents (incorrectly) the "RGMII PCS" being used, meaning we don't read the in-band status. However, a core synthesised for RGMII and also SGMII, TBI or RTBI will have this capability bit set, thus making these code paths reachable. The Jetson Xavier NX uses RGMII mode to talk to its PHY, and removing the incorrect check for priv->dma_cap.pcs reveals the theortical issue with netif_carrier_*() manipulation is real: dwc-eth-dwmac 2490000.ethernet eth0: Register MEM_TYPE_PAGE_POOL RxQ-0 dwc-eth-dwmac 2490000.ethernet eth0: PHY [stmmac-0:00] driver [RTL8211F Gigabit Ethernet] (irq=141) dwc-eth-dwmac 2490000.ethernet eth0: No Safety Features support found dwc-eth-dwmac 2490000.ethernet eth0: IEEE 1588-2008 Advanced Timestamp supported dwc-eth-dwmac 2490000.ethernet eth0: registered PTP clock dwc-eth-dwmac 2490000.ethernet eth0: configuring for phy/rgmii-id link mode 8021q: adding VLAN 0 to HW filter on device eth0 dwc-eth-dwmac 2490000.ethernet eth0: Adding VLAN ID 0 is not supported Link is Up - 1000/Full Link is Down Link is Up - 1000/Full This looks good until one realises that the phylink "Link" status messages are missing, even when the RJ45 cable is reconnected. Nothing one can do results in the interface working. The interrupt handler (which prints those "Link is" messages) always wins over phylink's resolve worker, meaning phylink never calls the mac_link_up() nor mac_link_down() methods. eth0 also sees no traffic received, and is unable to obtain a DHCP address: 3: eth0: mtu 1500 qdisc mq state UP group defa ult qlen 1000 link/ether e6:d3:6a:e6:92:de brd ff:ff:ff:ff:ff:ff RX: bytes packets errors dropped overrun mcast 0 0 0 0 0 0 TX: bytes packets errors dropped carrier collsns 27686 149 0 0 0 0 With the STMMAC_FLAG_HAS_INTEGRATED_PCS flag set, which disables the netif_carrier_*() manipulation then stmmac works normally: dwc-eth-dwmac 2490000.ethernet eth0: Register MEM_TYPE_PAGE_POOL RxQ-0 dwc-eth-dwmac 2490000.ethernet eth0: PHY [stmmac-0:00] driver [RTL8211F Gigabit Ethernet] (irq=141) dwc-eth-dwmac 2490000.ethernet eth0: No Safety Features support found dwc-eth-dwmac 2490000.ethernet eth0: IEEE 1588-2008 Advanced Timestamp supported dwc-eth-dwmac 2490000.ethernet eth0: registered PTP clock dwc-eth-dwmac 2490000.ethernet eth0: configuring for phy/rgmii-id link mode 8021q: adding VLAN 0 to HW filter on device eth0 dwc-eth-dwmac 2490000.ethernet eth0: Adding VLAN ID 0 is not supported Link is Up - 1000/Full dwc-eth-dwmac 2490000.ethernet eth0: Link is Up - 1Gbps/Full - flow control rx/tx and packets can be transferred. This clearly shows that when priv->hw->pcs is set, but STMMAC_FLAG_HAS_INTEGRATED_PCS is clear, the driver reliably fails. Discovering whether a platform falls into this is impossible as parsing all the dtsi and dts files to find out which use the stmmac driver, whether any of them use RGMII or SGMII and also depends whether an external interface is being used. The kernel likely doesn't contain all dts files either. The only driver that sets this flag uses the qcom,sa8775p-ethqos compatible, and uses SGMII or 2500BASE-X. but these are saved from this problem by the incorrect check for priv->dma_cap.pcs. So, we have to assume that for every other platform that uses SGMII with stmmac is using an external PCS. Moreover, ethtool output can be incorrect. With the full-duplex link negotiated, ethtool reports: Speed: 1000Mb/s Duplex: Half because with dwmac4, the full-duplex bit is in bit 16 of the status, priv->xstats.pcs_duplex becomes BIT(16) for full duplex, but the ethtool ksettings duplex member is u8 - so becomes zero. Moreover, the supported, advertised and link partner modes are all "not reported". Finally, ksettings_set() won't be able to set the advertisement on a PHY if this PCS code is activated, which is incorrect when SGMII is used with a PHY. Thus, remove: 1. the incorrect netif_carrier_*() manipulation. 2. the broken ethtool ksettings code. Given that all uses of STMMAC_FLAG_HAS_INTEGRATED_PCS are now gone, remove the flag from stmmac.h and dwmac-qcom-ethqos.c. Reviewed-by: Andrew Lunn Signed-off-by: Russell King (Oracle) Tested-by: Maxime Chevallier Tested-by: Lad Prabhakar Link: https://patch.msgid.link/E1v9P5y-0000000AolC-1QWH@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../stmicro/stmmac/dwmac-qcom-ethqos.c | 4 -- .../ethernet/stmicro/stmmac/stmmac_ethtool.c | 55 ------------------- .../net/ethernet/stmicro/stmmac/stmmac_main.c | 9 --- include/linux/stmmac.h | 1 - 4 files changed, 69 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c index d8fd4d8f6ced..f62825220cf7 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c @@ -96,7 +96,6 @@ struct ethqos_emac_driver_data { bool rgmii_config_loopback_en; bool has_emac_ge_3; const char *link_clk_name; - bool has_integrated_pcs; u32 dma_addr_width; struct dwmac4_addrs dwmac4_addrs; bool needs_sgmii_loopback; @@ -282,7 +281,6 @@ static const struct ethqos_emac_driver_data emac_v4_0_0_data = { .rgmii_config_loopback_en = false, .has_emac_ge_3 = true, .link_clk_name = "phyaux", - .has_integrated_pcs = true, .needs_sgmii_loopback = true, .dma_addr_width = 36, .dwmac4_addrs = { @@ -856,8 +854,6 @@ static int qcom_ethqos_probe(struct platform_device *pdev) plat_dat->flags |= STMMAC_FLAG_TSO_EN; if (of_device_is_compatible(np, "qcom,qcs404-ethqos")) plat_dat->flags |= STMMAC_FLAG_RX_CLK_RUNS_IN_LPI; - if (data->has_integrated_pcs) - plat_dat->flags |= STMMAC_FLAG_HAS_INTEGRATED_PCS; if (data->dma_addr_width) plat_dat->host_dma_width = data->dma_addr_width; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c index 39fa1ec92f82..d89662b48087 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c @@ -322,47 +322,6 @@ static int stmmac_ethtool_get_link_ksettings(struct net_device *dev, { struct stmmac_priv *priv = netdev_priv(dev); - if (!(priv->plat->flags & STMMAC_FLAG_HAS_INTEGRATED_PCS) && - (priv->hw->pcs & STMMAC_PCS_RGMII || - priv->hw->pcs & STMMAC_PCS_SGMII)) { - u32 supported, advertising, lp_advertising; - - if (!priv->xstats.pcs_link) { - cmd->base.speed = SPEED_UNKNOWN; - cmd->base.duplex = DUPLEX_UNKNOWN; - return 0; - } - cmd->base.duplex = priv->xstats.pcs_duplex; - - cmd->base.speed = priv->xstats.pcs_speed; - - /* Encoding of PSE bits is defined in 802.3z, 37.2.1.4 */ - - ethtool_convert_link_mode_to_legacy_u32( - &supported, cmd->link_modes.supported); - ethtool_convert_link_mode_to_legacy_u32( - &advertising, cmd->link_modes.advertising); - ethtool_convert_link_mode_to_legacy_u32( - &lp_advertising, cmd->link_modes.lp_advertising); - - /* Reg49[3] always set because ANE is always supported */ - cmd->base.autoneg = ADVERTISED_Autoneg; - supported |= SUPPORTED_Autoneg; - advertising |= ADVERTISED_Autoneg; - lp_advertising |= ADVERTISED_Autoneg; - - cmd->base.port = PORT_OTHER; - - ethtool_convert_legacy_u32_to_link_mode( - cmd->link_modes.supported, supported); - ethtool_convert_legacy_u32_to_link_mode( - cmd->link_modes.advertising, advertising); - ethtool_convert_legacy_u32_to_link_mode( - cmd->link_modes.lp_advertising, lp_advertising); - - return 0; - } - return phylink_ethtool_ksettings_get(priv->phylink, cmd); } @@ -372,20 +331,6 @@ stmmac_ethtool_set_link_ksettings(struct net_device *dev, { struct stmmac_priv *priv = netdev_priv(dev); - if (!(priv->plat->flags & STMMAC_FLAG_HAS_INTEGRATED_PCS) && - (priv->hw->pcs & STMMAC_PCS_RGMII || - priv->hw->pcs & STMMAC_PCS_SGMII)) { - /* Only support ANE */ - if (cmd->base.autoneg != AUTONEG_ENABLE) - return -EINVAL; - - mutex_lock(&priv->lock); - stmmac_pcs_ctrl_ane(priv, 1, priv->hw->ps, 0); - mutex_unlock(&priv->lock); - - return 0; - } - return phylink_ethtool_ksettings_set(priv->phylink, cmd); } diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index c9fa965c8566..867d0ca3b45e 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -6001,15 +6001,6 @@ static void stmmac_common_interrupt(struct stmmac_priv *priv) for (queue = 0; queue < queues_count; queue++) stmmac_host_mtl_irq_status(priv, priv->hw, queue); - /* PCS link status */ - if (priv->hw->pcs && - !(priv->plat->flags & STMMAC_FLAG_HAS_INTEGRATED_PCS)) { - if (priv->xstats.pcs_link) - netif_carrier_on(priv->dev); - else - netif_carrier_off(priv->dev); - } - stmmac_timestamp_interrupt(priv, priv); } } diff --git a/include/linux/stmmac.h b/include/linux/stmmac.h index fa1318bac06c..99022620457a 100644 --- a/include/linux/stmmac.h +++ b/include/linux/stmmac.h @@ -171,7 +171,6 @@ struct dwmac4_addrs { u32 mtl_low_cred_offset; }; -#define STMMAC_FLAG_HAS_INTEGRATED_PCS BIT(0) #define STMMAC_FLAG_SPH_DISABLE BIT(1) #define STMMAC_FLAG_USE_PHY_WOL BIT(2) #define STMMAC_FLAG_HAS_SUN8I BIT(3) From 14f74bc6dc699f63b5a6dfa9b22229f0caea89f3 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Thu, 16 Oct 2025 15:36:51 +0100 Subject: [PATCH 0124/1381] net: stmmac: remove xstats.pcs_* members As a result of the previous commit, the pcs_link, pcs_duplex and pcs_speed members are not used outside of the interrupt handling code, and are only used to print their status using the misleading "Link is" messages that bear no relation to the actual status of the link. Remove the printing of these messages, these members, and the code that decodes them from the hardware. Reviewed-by: Andrew Lunn Signed-off-by: Russell King (Oracle) Tested-by: Maxime Chevallier Tested-by: Lad Prabhakar Link: https://patch.msgid.link/E1v9P63-0000000AolI-23Kf@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/common.h | 3 -- .../ethernet/stmicro/stmmac/dwmac1000_core.c | 28 +------------------ .../net/ethernet/stmicro/stmmac/dwmac4_core.c | 28 +------------------ 3 files changed, 2 insertions(+), 57 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/common.h b/drivers/net/ethernet/stmicro/stmmac/common.h index 8f34c9ad457f..33aeac5666f4 100644 --- a/drivers/net/ethernet/stmicro/stmmac/common.h +++ b/drivers/net/ethernet/stmicro/stmmac/common.h @@ -192,9 +192,6 @@ struct stmmac_extra_stats { unsigned long irq_pcs_ane_n; unsigned long irq_pcs_link_n; unsigned long irq_rgmii_n; - unsigned long pcs_link; - unsigned long pcs_duplex; - unsigned long pcs_speed; /* debug register */ unsigned long mtl_tx_status_fifo_full; unsigned long mtl_tx_fifo_not_empty; diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c index fe776ddf6889..2c5ee59c3208 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c @@ -266,34 +266,8 @@ static void dwmac1000_pmt(struct mac_device_info *hw, unsigned long mode) /* RGMII or SMII interface */ static void dwmac1000_rgsmii(void __iomem *ioaddr, struct stmmac_extra_stats *x) { - u32 status; - - status = readl(ioaddr + GMAC_RGSMIIIS); + readl(ioaddr + GMAC_RGSMIIIS); x->irq_rgmii_n++; - - /* Check the link status */ - if (status & GMAC_RGSMIIIS_LNKSTS) { - int speed_value; - - x->pcs_link = 1; - - speed_value = ((status & GMAC_RGSMIIIS_SPEED) >> - GMAC_RGSMIIIS_SPEED_SHIFT); - if (speed_value == GMAC_RGSMIIIS_SPEED_125) - x->pcs_speed = SPEED_1000; - else if (speed_value == GMAC_RGSMIIIS_SPEED_25) - x->pcs_speed = SPEED_100; - else - x->pcs_speed = SPEED_10; - - x->pcs_duplex = (status & GMAC_RGSMIIIS_LNKMOD_MASK); - - pr_info("Link is Up - %d/%s\n", (int)x->pcs_speed, - x->pcs_duplex ? "Full" : "Half"); - } else { - x->pcs_link = 0; - pr_info("Link is Down\n"); - } } static int dwmac1000_irq_status(struct mac_device_info *hw, diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c index d85bc0bb5c3c..8a19df7b0577 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c @@ -592,34 +592,8 @@ static void dwmac4_ctrl_ane(struct stmmac_priv *priv, bool ane, bool srgmi_ral, /* RGMII or SMII interface */ static void dwmac4_phystatus(void __iomem *ioaddr, struct stmmac_extra_stats *x) { - u32 status; - - status = readl(ioaddr + GMAC_PHYIF_CONTROL_STATUS); + readl(ioaddr + GMAC_PHYIF_CONTROL_STATUS); x->irq_rgmii_n++; - - /* Check the link status */ - if (status & GMAC_PHYIF_CTRLSTATUS_LNKSTS) { - int speed_value; - - x->pcs_link = 1; - - speed_value = ((status & GMAC_PHYIF_CTRLSTATUS_SPEED) >> - GMAC_PHYIF_CTRLSTATUS_SPEED_SHIFT); - if (speed_value == GMAC_PHYIF_CTRLSTATUS_SPEED_125) - x->pcs_speed = SPEED_1000; - else if (speed_value == GMAC_PHYIF_CTRLSTATUS_SPEED_25) - x->pcs_speed = SPEED_100; - else - x->pcs_speed = SPEED_10; - - x->pcs_duplex = (status & GMAC_PHYIF_CTRLSTATUS_LNKMOD); - - pr_info("Link is Up - %d/%s\n", (int)x->pcs_speed, - x->pcs_duplex ? "Full" : "Half"); - } else { - x->pcs_link = 0; - pr_info("Link is Down\n"); - } } static int dwmac4_irq_mtl_status(struct stmmac_priv *priv, From 2e2c878a314175a4c3b76ee3be455f543438f472 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Thu, 16 Oct 2025 15:36:56 +0100 Subject: [PATCH 0125/1381] net: stmmac: remove SGMII/RGMII/SMII interrupt handling Now that the only use for the interrupt is to clear it and increment a statistic counter (which is not that relevant anymore) remove all this code and ensure that the interrupt remains disabled to avoid a stuck interrupt. dwmac-sun8i still uses this statistic counter, so it is inappropriate for this patch to remove it. Reviewed-by: Andrew Lunn Signed-off-by: Russell King (Oracle) Tested-by: Maxime Chevallier Tested-by: Lad Prabhakar Link: https://patch.msgid.link/E1v9P68-0000000AolO-2W5s@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/dwmac1000.h | 6 +++--- drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c | 10 ---------- drivers/net/ethernet/stmicro/stmmac/dwmac4.h | 3 +-- drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c | 9 --------- 4 files changed, 4 insertions(+), 24 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac1000.h b/drivers/net/ethernet/stmicro/stmmac/dwmac1000.h index 0c011a47d5a3..8f3002d9de78 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac1000.h +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac1000.h @@ -38,10 +38,10 @@ #define GMAC_INT_DISABLE_PCSAN BIT(2) #define GMAC_INT_DISABLE_PMT BIT(3) #define GMAC_INT_DISABLE_TIMESTAMP BIT(9) -#define GMAC_INT_DISABLE_PCS (GMAC_INT_DISABLE_RGMII | \ - GMAC_INT_DISABLE_PCSLINK | \ +#define GMAC_INT_DISABLE_PCS (GMAC_INT_DISABLE_PCSLINK | \ GMAC_INT_DISABLE_PCSAN) -#define GMAC_INT_DEFAULT_MASK (GMAC_INT_DISABLE_TIMESTAMP | \ +#define GMAC_INT_DEFAULT_MASK (GMAC_INT_DISABLE_RGMII | \ + GMAC_INT_DISABLE_TIMESTAMP | \ GMAC_INT_DISABLE_PCS) /* PMT Control and Status */ diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c index 2c5ee59c3208..654331b411f4 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c @@ -263,13 +263,6 @@ static void dwmac1000_pmt(struct mac_device_info *hw, unsigned long mode) writel(pmt, ioaddr + GMAC_PMT); } -/* RGMII or SMII interface */ -static void dwmac1000_rgsmii(void __iomem *ioaddr, struct stmmac_extra_stats *x) -{ - readl(ioaddr + GMAC_RGSMIIIS); - x->irq_rgmii_n++; -} - static int dwmac1000_irq_status(struct mac_device_info *hw, struct stmmac_extra_stats *x) { @@ -311,9 +304,6 @@ static int dwmac1000_irq_status(struct mac_device_info *hw, dwmac_pcs_isr(ioaddr, GMAC_PCS_BASE, intr_status, x); - if (intr_status & PCS_RGSMIIIS_IRQ) - dwmac1000_rgsmii(ioaddr, x); - return ret; } diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4.h b/drivers/net/ethernet/stmicro/stmmac/dwmac4.h index 3dec1a264cf6..6dd84b6544cc 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac4.h +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4.h @@ -106,8 +106,7 @@ #define GMAC_INT_LPI_EN BIT(5) #define GMAC_INT_TSIE BIT(12) -#define GMAC_PCS_IRQ_DEFAULT (GMAC_INT_RGSMIIS | GMAC_INT_PCS_LINK | \ - GMAC_INT_PCS_ANE) +#define GMAC_PCS_IRQ_DEFAULT (GMAC_INT_PCS_LINK | GMAC_INT_PCS_ANE) #define GMAC_INT_DEFAULT_ENABLE (GMAC_INT_PMT_EN | GMAC_INT_LPI_EN | \ GMAC_INT_TSIE) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c index 8a19df7b0577..bff4c371c1d2 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c @@ -589,13 +589,6 @@ static void dwmac4_ctrl_ane(struct stmmac_priv *priv, bool ane, bool srgmi_ral, dwmac_ctrl_ane(priv->ioaddr, GMAC_PCS_BASE, ane, srgmi_ral, loopback); } -/* RGMII or SMII interface */ -static void dwmac4_phystatus(void __iomem *ioaddr, struct stmmac_extra_stats *x) -{ - readl(ioaddr + GMAC_PHYIF_CONTROL_STATUS); - x->irq_rgmii_n++; -} - static int dwmac4_irq_mtl_status(struct stmmac_priv *priv, struct mac_device_info *hw, u32 chan) { @@ -667,8 +660,6 @@ static int dwmac4_irq_status(struct mac_device_info *hw, } dwmac_pcs_isr(ioaddr, GMAC_PCS_BASE, intr_status, x); - if (intr_status & PCS_RGSMIIIS_IRQ) - dwmac4_phystatus(ioaddr, x); return ret; } From ebc5d656b78cbcc6314f3af12af2d556f795d280 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Thu, 16 Oct 2025 15:37:01 +0100 Subject: [PATCH 0126/1381] net: stmmac: remove PCS "mode" pause handling Remove the "we always autoneg pause" forcing when the stmmac driver decides that a "PCS" is present, which blocks passing the ethtool pause calls to phylink when using SGMII mode. This prevents the pause results being reported when a PHY is attached using SGMII mode, or the pause settings being changed in SGMII mode. There is no reason to prevent this. Reviewed-by: Andrew Lunn Signed-off-by: Russell King (Oracle) Tested-by: Maxime Chevallier Tested-by: Lad Prabhakar Link: https://patch.msgid.link/E1v9P6D-0000000AolU-2zjv@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../net/ethernet/stmicro/stmmac/stmmac_ethtool.c | 13 ++----------- 1 file changed, 2 insertions(+), 11 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c index d89662b48087..c60cd948311e 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c @@ -424,11 +424,7 @@ stmmac_get_pauseparam(struct net_device *netdev, { struct stmmac_priv *priv = netdev_priv(netdev); - if (priv->hw->pcs) { - pause->autoneg = 1; - } else { - phylink_ethtool_get_pauseparam(priv->phylink, pause); - } + phylink_ethtool_get_pauseparam(priv->phylink, pause); } static int @@ -437,12 +433,7 @@ stmmac_set_pauseparam(struct net_device *netdev, { struct stmmac_priv *priv = netdev_priv(netdev); - if (priv->hw->pcs) { - pause->autoneg = 1; - return 0; - } else { - return phylink_ethtool_set_pauseparam(priv->phylink, pause); - } + return phylink_ethtool_set_pauseparam(priv->phylink, pause); } static u64 stmmac_get_rx_normal_irq_n(struct stmmac_priv *priv, int q) From 19064a58bd3c38dd1f93eb6c687a5b268cea03ba Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Thu, 16 Oct 2025 15:37:06 +0100 Subject: [PATCH 0127/1381] net: stmmac: remove unused PCS loopback support Nothing calls stmmac_pcs_ctrl_ane() with the "loopback" argument set to anything except zero, so this serves no useful purpose. Remove the argument to reduce the code complexity. Reviewed-by: Andrew Lunn Signed-off-by: Russell King (Oracle) Tested-by: Maxime Chevallier Tested-by: Lad Prabhakar Link: https://patch.msgid.link/E1v9P6I-0000000Aola-3Sih@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c | 2 +- drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c | 4 ++-- drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c | 5 ++--- drivers/net/ethernet/stmicro/stmmac/hwif.h | 4 ++-- drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 2 +- drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.h | 6 +----- 6 files changed, 9 insertions(+), 14 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c index f62825220cf7..32244217d952 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c @@ -622,7 +622,7 @@ static void ethqos_set_serdes_speed(struct qcom_ethqos *ethqos, int speed) static void ethqos_pcs_set_inband(struct stmmac_priv *priv, bool enable) { - stmmac_pcs_ctrl_ane(priv, enable, 0, 0); + stmmac_pcs_ctrl_ane(priv, enable, 0); } /* On interface toggle MAC registers gets reset. diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c index 654331b411f4..5c653be3d453 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c @@ -358,9 +358,9 @@ static void dwmac1000_set_eee_timer(struct mac_device_info *hw, int ls, int tw) } static void dwmac1000_ctrl_ane(struct stmmac_priv *priv, bool ane, - bool srgmi_ral, bool loopback) + bool srgmi_ral) { - dwmac_ctrl_ane(priv->ioaddr, GMAC_PCS_BASE, ane, srgmi_ral, loopback); + dwmac_ctrl_ane(priv->ioaddr, GMAC_PCS_BASE, ane, srgmi_ral); } static void dwmac1000_debug(struct stmmac_priv *priv, void __iomem *ioaddr, diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c index bff4c371c1d2..21e4461db937 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c @@ -583,10 +583,9 @@ static void dwmac4_flow_ctrl(struct mac_device_info *hw, unsigned int duplex, } } -static void dwmac4_ctrl_ane(struct stmmac_priv *priv, bool ane, bool srgmi_ral, - bool loopback) +static void dwmac4_ctrl_ane(struct stmmac_priv *priv, bool ane, bool srgmi_ral) { - dwmac_ctrl_ane(priv->ioaddr, GMAC_PCS_BASE, ane, srgmi_ral, loopback); + dwmac_ctrl_ane(priv->ioaddr, GMAC_PCS_BASE, ane, srgmi_ral); } static int dwmac4_irq_mtl_status(struct stmmac_priv *priv, diff --git a/drivers/net/ethernet/stmicro/stmmac/hwif.h b/drivers/net/ethernet/stmicro/stmmac/hwif.h index 14dbe0685997..7796f5f3c96f 100644 --- a/drivers/net/ethernet/stmicro/stmmac/hwif.h +++ b/drivers/net/ethernet/stmicro/stmmac/hwif.h @@ -374,8 +374,8 @@ struct stmmac_ops { struct stmmac_extra_stats *x, u32 rx_queues, u32 tx_queues); /* PCS calls */ - void (*pcs_ctrl_ane)(struct stmmac_priv *priv, bool ane, bool srgmi_ral, - bool loopback); + void (*pcs_ctrl_ane)(struct stmmac_priv *priv, bool ane, + bool srgmi_ral); /* Safety Features */ int (*safety_feat_config)(void __iomem *ioaddr, unsigned int asp, struct stmmac_safety_feature_cfg *safety_cfg); diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index 867d0ca3b45e..e21d96933408 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -3493,7 +3493,7 @@ static int stmmac_hw_setup(struct net_device *dev) } if (priv->hw->pcs) - stmmac_pcs_ctrl_ane(priv, 1, priv->hw->ps, 0); + stmmac_pcs_ctrl_ane(priv, 1, priv->hw->ps); /* set TX and RX rings length */ stmmac_set_rings_length(priv); diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.h b/drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.h index 4a684c97dfae..5778f5b2f313 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.h +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.h @@ -82,13 +82,12 @@ static inline void dwmac_pcs_isr(void __iomem *ioaddr, u32 reg, * @reg: Base address of the AN Control Register. * @ane: to enable the auto-negotiation * @srgmi_ral: to manage MAC-2-MAC SGMII connections. - * @loopback: to cause the PHY to loopback tx data into rx path. * Description: this is the main function to configure the AN control register * and init the ANE, select loopback (usually for debugging purpose) and * configure SGMII RAL. */ static inline void dwmac_ctrl_ane(void __iomem *ioaddr, u32 reg, bool ane, - bool srgmi_ral, bool loopback) + bool srgmi_ral) { u32 value = readl(ioaddr + GMAC_AN_CTRL(reg)); @@ -104,9 +103,6 @@ static inline void dwmac_ctrl_ane(void __iomem *ioaddr, u32 reg, bool ane, if (srgmi_ral) value |= GMAC_AN_CTRL_SGMRAL; - if (loopback) - value |= GMAC_AN_CTRL_ELE; - writel(value, ioaddr + GMAC_AN_CTRL(reg)); } #endif /* __STMMAC_PCS_H__ */ From aa1b6775aef74b2177aa72f912564cd064a83b70 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Thu, 16 Oct 2025 15:37:11 +0100 Subject: [PATCH 0128/1381] net: stmmac: remove hw->ps xxx_core_init() hardware setup After a lot of digging, it seems that the oddly named hw->ps member is all about configuring the core for reverse SGMII. This member is set to one of 0, SPEED_10, SPEED_100 or SPEED_1000 depending on priv->plat->mac_port_sel_speed. On DT systems, this comes from the "snps,ps-speed" DT property. When set to a non-zero value, it: 1. Configures the MAC at initialisation time to operate at a specific speed. However, this will be overwritten by mac_link_up() when the link comes up (e.g. with the fixed-link parameters.) Note that dwxgmac2 wants to also support SPEED_2500 and SPEED_10000, but both these values are impossible. 2. It _incorrectly_ enables the transmitter (GMAC_CONFIG_TE) which makes no sense, rather than enabling the "transmit configuration" bit (GMAC_CONFIG_TC). Likely a typo. 3. It configures the SGMII rate adapter layer to retrieve its speed setting from the MAC configuration register rather than the PHY. There are two ways forward here: a) fixing (2) so that we set GMAC_CONFIG_TC. However, we have platform that set the "snps,ps-speed" property and that work today. Fixing this will cause the RGMII, SGMII or SMII inband configuration to be transmitted, which will be a functional change which could cause a regression. b) ripping out (1) and (2) as they are ineffective. This also has the possibility of regressions, but the patch author believes this risk is much lower than (a). Therefore, this commit takes the approach in (b). Signed-off-by: Russell King (Oracle) Tested-by: Maxime Chevallier Tested-by: Lad Prabhakar Link: https://patch.msgid.link/E1v9P6N-0000000Aolg-3y0a@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../ethernet/stmicro/stmmac/dwmac1000_core.c | 23 +++-------------- .../net/ethernet/stmicro/stmmac/dwmac4_core.c | 24 +++--------------- .../ethernet/stmicro/stmmac/dwxgmac2_core.c | 25 ++----------------- 3 files changed, 8 insertions(+), 64 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c index 5c653be3d453..d35db8958be1 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c @@ -26,35 +26,18 @@ static void dwmac1000_core_init(struct mac_device_info *hw, struct net_device *dev) { void __iomem *ioaddr = hw->pcsr; - u32 value = readl(ioaddr + GMAC_CONTROL); int mtu = dev->mtu; + u32 value; /* Configure GMAC core */ - value |= GMAC_CORE_INIT; + value = readl(ioaddr + GMAC_CONTROL); if (mtu > 1500) value |= GMAC_CONTROL_2K; if (mtu > 2000) value |= GMAC_CONTROL_JE; - if (hw->ps) { - value |= GMAC_CONTROL_TE; - - value &= ~hw->link.speed_mask; - switch (hw->ps) { - case SPEED_1000: - value |= hw->link.speed1000; - break; - case SPEED_100: - value |= hw->link.speed100; - break; - case SPEED_10: - value |= hw->link.speed10; - break; - } - } - - writel(value, ioaddr + GMAC_CONTROL); + writel(value | GMAC_CORE_INIT, ioaddr + GMAC_CONTROL); /* Mask GMAC interrupts */ value = GMAC_INT_DEFAULT_MASK; diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c index 21e4461db937..d855ab6b9145 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c @@ -27,29 +27,11 @@ static void dwmac4_core_init(struct mac_device_info *hw, { struct stmmac_priv *priv = netdev_priv(dev); void __iomem *ioaddr = hw->pcsr; - u32 value = readl(ioaddr + GMAC_CONFIG); unsigned long clk_rate; + u32 value; - value |= GMAC_CORE_INIT; - - if (hw->ps) { - value |= GMAC_CONFIG_TE; - - value &= hw->link.speed_mask; - switch (hw->ps) { - case SPEED_1000: - value |= hw->link.speed1000; - break; - case SPEED_100: - value |= hw->link.speed100; - break; - case SPEED_10: - value |= hw->link.speed10; - break; - } - } - - writel(value, ioaddr + GMAC_CONFIG); + value = readl(ioaddr + GMAC_CONFIG); + writel(value | GMAC_CORE_INIT, ioaddr + GMAC_CONFIG); /* Configure LPI 1us counter to number of CSR clock ticks in 1us - 1 */ clk_rate = clk_get_rate(priv->plat->stmmac_clk); diff --git a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c index 00e929bf280b..0430af27da40 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c @@ -23,29 +23,8 @@ static void dwxgmac2_core_init(struct mac_device_info *hw, tx = readl(ioaddr + XGMAC_TX_CONFIG); rx = readl(ioaddr + XGMAC_RX_CONFIG); - tx |= XGMAC_CORE_INIT_TX; - rx |= XGMAC_CORE_INIT_RX; - - if (hw->ps) { - tx |= XGMAC_CONFIG_TE; - tx &= ~hw->link.speed_mask; - - switch (hw->ps) { - case SPEED_10000: - tx |= hw->link.xgmii.speed10000; - break; - case SPEED_2500: - tx |= hw->link.speed2500; - break; - case SPEED_1000: - default: - tx |= hw->link.speed1000; - break; - } - } - - writel(tx, ioaddr + XGMAC_TX_CONFIG); - writel(rx, ioaddr + XGMAC_RX_CONFIG); + writel(tx | XGMAC_CORE_INIT_TX, ioaddr + XGMAC_TX_CONFIG); + writel(rx | XGMAC_CORE_INIT_RX, ioaddr + XGMAC_RX_CONFIG); writel(XGMAC_INT_DEFAULT_EN, ioaddr + XGMAC_INT_EN); } From 70589b05a03e853367c6f887191a7527bcf08163 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Thu, 16 Oct 2025 15:37:17 +0100 Subject: [PATCH 0129/1381] net: stmmac: remove RGMII "pcs" mode Remove the RGMII "pcs" code in stmmac_check_pcs_mode() due to: 1) This should never have been conditional on a PCS being present, as when a core is synthesised using only RGMII, the PCS won't be present and priv->dma_cap.pcs will be false. Only multi-interface cores which have a PCS present would have detected RGMII. 2) STMMAC_PCS_RGMII has no effect since the broken netif_carrier and ethtool code was removed. Reviewed-by: Andrew Lunn Signed-off-by: Russell King (Oracle) Tested-by: Maxime Chevallier Tested-by: Lad Prabhakar Link: https://patch.msgid.link/E1v9P6T-0000000Aoll-0Ify@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/common.h | 1 - drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 14 +++----------- 2 files changed, 3 insertions(+), 12 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/common.h b/drivers/net/ethernet/stmicro/stmmac/common.h index 33aeac5666f4..ed5e207ffdba 100644 --- a/drivers/net/ethernet/stmicro/stmmac/common.h +++ b/drivers/net/ethernet/stmicro/stmmac/common.h @@ -270,7 +270,6 @@ struct stmmac_safety_stats { #define FLOW_AUTO (FLOW_TX | FLOW_RX) /* PCS defines */ -#define STMMAC_PCS_RGMII (1 << 0) #define STMMAC_PCS_SGMII (1 << 1) #define SF_DMA_MODE 1 /* DMA STORE-AND-FORWARD Operation Mode */ diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index e21d96933408..f160c5bd25b2 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -1087,17 +1087,9 @@ static void stmmac_check_pcs_mode(struct stmmac_priv *priv) { int interface = priv->plat->phy_interface; - if (priv->dma_cap.pcs) { - if ((interface == PHY_INTERFACE_MODE_RGMII) || - (interface == PHY_INTERFACE_MODE_RGMII_ID) || - (interface == PHY_INTERFACE_MODE_RGMII_RXID) || - (interface == PHY_INTERFACE_MODE_RGMII_TXID)) { - netdev_dbg(priv->dev, "PCS RGMII support enabled\n"); - priv->hw->pcs = STMMAC_PCS_RGMII; - } else if (interface == PHY_INTERFACE_MODE_SGMII) { - netdev_dbg(priv->dev, "PCS SGMII support enabled\n"); - priv->hw->pcs = STMMAC_PCS_SGMII; - } + if (priv->dma_cap.pcs && interface == PHY_INTERFACE_MODE_SGMII) { + netdev_dbg(priv->dev, "PCS SGMII support enabled\n"); + priv->hw->pcs = STMMAC_PCS_SGMII; } } From c7b0d7874de0db6ce24bae3530ee91309aa2ff4e Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Thu, 16 Oct 2025 15:37:22 +0100 Subject: [PATCH 0130/1381] net: stmmac: move reverse-"pcs" mode setup to stmmac_check_pcs_mode() The broken reverse-mode, selected by snps,ps-speed, is configured when the platform provides a valid port speed and a PCS is being used. Both these remain constant after the driver has probed, so the software state doesn't need to be re-initialised each time stmmac_hw_setup() is called (which is called at open and resume time.) Move the software setup of reverse-mode to stmmac_check_pcs_mode() which is called from the driver probe function. Reviewed-by: Andrew Lunn Signed-off-by: Russell King (Oracle) Tested-by: Maxime Chevallier Tested-by: Lad Prabhakar Link: https://patch.msgid.link/E1v9P6Y-0000000Aolr-0vLH@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../net/ethernet/stmicro/stmmac/stmmac_main.c | 26 +++++++++---------- 1 file changed, 13 insertions(+), 13 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index f160c5bd25b2..9a8dacf18c14 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -1091,6 +1091,19 @@ static void stmmac_check_pcs_mode(struct stmmac_priv *priv) netdev_dbg(priv->dev, "PCS SGMII support enabled\n"); priv->hw->pcs = STMMAC_PCS_SGMII; } + + /* PS and related bits will be programmed according to the speed */ + if (priv->hw->pcs) { + int speed = priv->plat->mac_port_sel_speed; + + if ((speed == SPEED_10) || (speed == SPEED_100) || + (speed == SPEED_1000)) { + priv->hw->ps = speed; + } else { + dev_warn(priv->device, "invalid port speed\n"); + priv->hw->ps = 0; + } + } } /** @@ -3435,19 +3448,6 @@ static int stmmac_hw_setup(struct net_device *dev) stmmac_set_umac_addr(priv, priv->hw, dev->dev_addr, 0); phylink_rx_clk_stop_unblock(priv->phylink); - /* PS and related bits will be programmed according to the speed */ - if (priv->hw->pcs) { - int speed = priv->plat->mac_port_sel_speed; - - if ((speed == SPEED_10) || (speed == SPEED_100) || - (speed == SPEED_1000)) { - priv->hw->ps = speed; - } else { - dev_warn(priv->device, "invalid port speed\n"); - priv->hw->ps = 0; - } - } - /* Initialize the MAC Core */ stmmac_core_init(priv, priv->hw, dev); From 412d5f32cb36110475d1c77e291b73e97f2f000a Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Thu, 16 Oct 2025 15:37:27 +0100 Subject: [PATCH 0131/1381] net: stmmac: simplify stmmac_check_pcs_mode() Now that we only support one mode, simplify stmmac_check_pcs_mode(). Reviewed-by: Andrew Lunn Signed-off-by: Russell King (Oracle) Tested-by: Maxime Chevallier Tested-by: Lad Prabhakar Link: https://patch.msgid.link/E1v9P6d-0000000Aolw-1T7d@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../net/ethernet/stmicro/stmmac/stmmac_main.c | 17 +++++++++-------- 1 file changed, 9 insertions(+), 8 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index 9a8dacf18c14..832062b2f4f3 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -1086,22 +1086,23 @@ static const struct phylink_mac_ops stmmac_phylink_mac_ops = { static void stmmac_check_pcs_mode(struct stmmac_priv *priv) { int interface = priv->plat->phy_interface; + int speed = priv->plat->mac_port_sel_speed; if (priv->dma_cap.pcs && interface == PHY_INTERFACE_MODE_SGMII) { netdev_dbg(priv->dev, "PCS SGMII support enabled\n"); priv->hw->pcs = STMMAC_PCS_SGMII; - } - /* PS and related bits will be programmed according to the speed */ - if (priv->hw->pcs) { - int speed = priv->plat->mac_port_sel_speed; - - if ((speed == SPEED_10) || (speed == SPEED_100) || - (speed == SPEED_1000)) { + switch (speed) { + case SPEED_10: + case SPEED_100: + case SPEED_1000: priv->hw->ps = speed; - } else { + break; + + default: dev_warn(priv->device, "invalid port speed\n"); priv->hw->ps = 0; + break; } } } From 5d1e7621f869a4d9d1da1652eb2e944784d868d8 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Thu, 16 Oct 2025 15:37:32 +0100 Subject: [PATCH 0132/1381] net: stmmac: hw->ps becomes hw->reverse_sgmii_enable After a lot of digging, it seems that the oddly named hw->ps member is all about setting the core into reverse SGMII speed. When set to a non-zero value, it: 1. Configures the MAC at initialisation time to operate at a specific speed. 2. It _incorrectly_ enables the transmitter (GMAC_CONFIG_TE) which makes no sense, rather than enabling the "transmit configuration" bit (GMAC_CONFIG_TC). 3. It configures the SGMII rate adapter layer to retrieve its speed setting from the MAC configuration register rather than the PHY. In the previous commit, we removed (1) and (2) as phylink overwrites the configuration set at that step. Thus, the only functional aspect is (3), which is a boolean operation. This means there is no need to store the actual speed, and just have a boolean flag. Convert the priv->ps member to a boolean, and rename it to priv->reverse_sgmii_enable to make it more understandable. Reviewed-by: Andrew Lunn Signed-off-by: Russell King (Oracle) Tested-by: Maxime Chevallier Tested-by: Lad Prabhakar Link: https://patch.msgid.link/E1v9P6i-0000000Aom3-1y2y@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/common.h | 2 +- drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 6 +++--- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/common.h b/drivers/net/ethernet/stmicro/stmmac/common.h index ed5e207ffdba..fee7021246b1 100644 --- a/drivers/net/ethernet/stmicro/stmmac/common.h +++ b/drivers/net/ethernet/stmicro/stmmac/common.h @@ -599,13 +599,13 @@ struct mac_device_info { unsigned int mcast_bits_log2; unsigned int rx_csum; unsigned int pcs; - unsigned int ps; unsigned int xlgmac; unsigned int num_vlan; u32 vlan_filter[32]; bool vlan_fail_q_en; u8 vlan_fail_q; bool hw_vlan_en; + bool reverse_sgmii_enable; }; struct stmmac_rx_routing { diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index 832062b2f4f3..c4d749396b53 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -1096,12 +1096,12 @@ static void stmmac_check_pcs_mode(struct stmmac_priv *priv) case SPEED_10: case SPEED_100: case SPEED_1000: - priv->hw->ps = speed; + priv->hw->reverse_sgmii_enable = true; break; default: dev_warn(priv->device, "invalid port speed\n"); - priv->hw->ps = 0; + priv->hw->reverse_sgmii_enable = false; break; } } @@ -3486,7 +3486,7 @@ static int stmmac_hw_setup(struct net_device *dev) } if (priv->hw->pcs) - stmmac_pcs_ctrl_ane(priv, 1, priv->hw->ps); + stmmac_pcs_ctrl_ane(priv, 1, priv->hw->reverse_sgmii_enable); /* set TX and RX rings length */ stmmac_set_rings_length(priv); From 5c61db08d9aeb960691fae16cfdb23df1355987e Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Thu, 16 Oct 2025 15:37:37 +0100 Subject: [PATCH 0133/1381] net: stmmac: do not require snps,ps-speed for SGMII SGMII mode does not require port-speed to be specified; this only switches SGMII to use the MAC configuration register speed settings and the actual value is irrelevant when the link comes up. As it seems the intention was to support "reverse SGMII" with this setting, but the code didn't actually configure that due to a typo, the warning and bad DT binding documentation has led people to specify snps,ps-speed in their DT files inappropriately. If mac_port_sel_speed is zero, then don't complain that the speed is invalid, as this means we're using "normal" SGMII. This does _not_ obsolete snps,ps-speed, nor does it change the behaviour of that property, with the exception of not making people mistakenly believe that they need to specify this option to use normal SGMII. There is no need to modify the binding. Reviewed-by: Andrew Lunn Signed-off-by: Russell King (Oracle) Tested-by: Maxime Chevallier Tested-by: Lad Prabhakar Link: https://patch.msgid.link/E1v9P6n-0000000Aom9-2LuZ@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index c4d749396b53..c643817f30da 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -1101,6 +1101,8 @@ static void stmmac_check_pcs_mode(struct stmmac_priv *priv) default: dev_warn(priv->device, "invalid port speed\n"); + fallthrough; + case 0: priv->hw->reverse_sgmii_enable = false; break; } From 045d7e5727c451f310f0de72ec1452274167431f Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Thu, 16 Oct 2025 15:37:42 +0100 Subject: [PATCH 0134/1381] net: stmmac: only call stmmac_pcs_ctrl_ane() for integrated SGMII PCS The internal PCS registers only exist if the core is synthesized with SGMII, TBI or RTBI support. They have no relevance for RGMII. However, priv->hw->pcs contains a STMMAC_PCS_RGMII flag, which is set if a PCS has been synthesized but we are operating in RGMII mode. As the register has no effect for RGMII, there is no point calling stmmac_pcs_ctrl_ane() in this case. Add a comment describing this and make it conditional on STMMAC_PCS_SGMII. Reviewed-by: Andrew Lunn Signed-off-by: Russell King (Oracle) Tested-by: Maxime Chevallier Tested-by: Lad Prabhakar Link: https://patch.msgid.link/E1v9P6s-0000000AomE-2pAa@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index c643817f30da..02dcbfa7d23e 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -3487,7 +3487,11 @@ static int stmmac_hw_setup(struct net_device *dev) } } - if (priv->hw->pcs) + /* The PCS control register is only relevant for SGMII, TBI and RTBI + * modes. We no longer support TBI or RTBI, so only configure this + * register when operating in SGMII mode with the integrated PCS. + */ + if (priv->hw->pcs & STMMAC_PCS_SGMII) stmmac_pcs_ctrl_ane(priv, 1, priv->hw->reverse_sgmii_enable); /* set TX and RX rings length */ From 237e54caeaeff5693c1fc2c04737ee5fe77a8bcd Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Thu, 16 Oct 2025 15:37:47 +0100 Subject: [PATCH 0135/1381] net: stmmac: provide PCS initialisation hook dwmac cores provide a feature bit to indicate when the PCS block is present, but features are only read after the core's setup() function has been called, meaning we can't decide whether to initialise the integrated PCS in the setup function. Provide a new MAC core hook for PCS initialisation, which will be called after the feature registers have been read. Reviewed-by: Andrew Lunn Signed-off-by: Russell King (Oracle) Tested-by: Maxime Chevallier Tested-by: Lad Prabhakar Link: https://patch.msgid.link/E1v9P6x-0000000AomL-3OKd@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/hwif.h | 4 ++++ drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 7 +++++++ 2 files changed, 11 insertions(+) diff --git a/drivers/net/ethernet/stmicro/stmmac/hwif.h b/drivers/net/ethernet/stmicro/stmmac/hwif.h index 7796f5f3c96f..82cfb6bec334 100644 --- a/drivers/net/ethernet/stmicro/stmmac/hwif.h +++ b/drivers/net/ethernet/stmicro/stmmac/hwif.h @@ -313,6 +313,8 @@ enum stmmac_lpi_mode { /* Helpers to program the MAC core */ struct stmmac_ops { + /* Initialise any PCS instances */ + int (*pcs_init)(struct stmmac_priv *priv); /* MAC core initialization */ void (*core_init)(struct mac_device_info *hw, struct net_device *dev); /* Update MAC capabilities */ @@ -413,6 +415,8 @@ struct stmmac_ops { u32 pclass); }; +#define stmmac_mac_pcs_init(__priv) \ + stmmac_do_callback(__priv, mac, pcs_init, __priv) #define stmmac_core_init(__priv, __args...) \ stmmac_do_void_callback(__priv, mac, core_init, __args) #define stmmac_mac_update_caps(__priv) \ diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index 02dcbfa7d23e..c041268d3a8d 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -7239,6 +7239,13 @@ static int stmmac_hw_init(struct stmmac_priv *priv) "Enable RX Mitigation via HW Watchdog Timer\n"); } + /* Unimplemented PCS init (as indicated by stmmac_do_callback() + * perversely returning -EINVAL) is non-fatal. + */ + ret = stmmac_mac_pcs_init(priv); + if (ret != -EINVAL) + return ret; + return 0; } From 2c81f3357136a0ece9b36be991c78a384399eaac Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Thu, 16 Oct 2025 15:37:52 +0100 Subject: [PATCH 0136/1381] net: stmmac: convert to phylink PCS support Now that stmmac's PCS support is much more simple - just a matter of configuring the control register - the basic conversion to phylink PCS support becomes straight forward. Create the infrastructure to setup a phylink_pcs instance for the integrated PCS: - add a struct stmmac_pcs to encapsulate the phylink_pcs structure, pointer to stmmac_priv, and the core-specific base address of the PCS registers. - modify stmmac_priv and stmmac_mac_select_pcs() to return the embedded phylink_pcs structure when setup and STMMAC_PCS_SGMII is in use, and move the comment from stmmac_hw_setup() to here. - create stmmac_pcs.c, which contains the phylink_pcs_ops structure, a dummy .pcs_get_state() method which always reports link-down, and .pcs_config() method, moving the call to stmmac_pcs_ctrl_ane() here, but without indirecting through the dwmac specific core code. The link-down behaviour mentioned above maintains the current behaviour when phylink is used with inband but without a PCS. This will ensure that the PCS control register is configured to the same settings as before, but will now happen when the netdev is opened or reusmed rather than only during probe time. However, this will be before the .fix_mac_speed() method is called, which is critical for the behaviour in dwmac-qcom-ethqos's ethqos_configure_sgmii() function to be maintained. Signed-off-by: Russell King (Oracle) Tested-by: Maxime Chevallier Tested-by: Lad Prabhakar Link: https://patch.msgid.link/E1v9P72-0000000AomR-3ro4@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/Makefile | 2 +- .../ethernet/stmicro/stmmac/dwmac1000_core.c | 9 ++++ .../net/ethernet/stmicro/stmmac/dwmac4_core.c | 11 +++++ drivers/net/ethernet/stmicro/stmmac/stmmac.h | 4 ++ .../net/ethernet/stmicro/stmmac/stmmac_main.c | 15 +++--- .../net/ethernet/stmicro/stmmac/stmmac_pcs.c | 47 +++++++++++++++++++ .../net/ethernet/stmicro/stmmac/stmmac_pcs.h | 17 +++++++ 7 files changed, 97 insertions(+), 8 deletions(-) create mode 100644 drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.c diff --git a/drivers/net/ethernet/stmicro/stmmac/Makefile b/drivers/net/ethernet/stmicro/stmmac/Makefile index ec56bcf2db62..1681a8a28313 100644 --- a/drivers/net/ethernet/stmicro/stmmac/Makefile +++ b/drivers/net/ethernet/stmicro/stmmac/Makefile @@ -7,7 +7,7 @@ stmmac-objs:= stmmac_main.o stmmac_ethtool.o stmmac_mdio.o ring_mode.o \ dwmac4_dma.o dwmac4_lib.o dwmac4_core.o dwmac5.o hwif.o \ stmmac_tc.o dwxgmac2_core.o dwxgmac2_dma.o dwxgmac2_descs.o \ stmmac_xdp.o stmmac_est.o stmmac_fpe.o stmmac_vlan.o \ - $(stmmac-y) + stmmac_pcs.o $(stmmac-y) stmmac-$(CONFIG_STMMAC_SELFTESTS) += stmmac_selftests.o diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c index d35db8958be1..571e48362444 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c @@ -22,6 +22,14 @@ #include "stmmac_ptp.h" #include "dwmac1000.h" +static int dwmac1000_pcs_init(struct stmmac_priv *priv) +{ + if (!priv->dma_cap.pcs) + return 0; + + return stmmac_integrated_pcs_init(priv, GMAC_PCS_BASE); +} + static void dwmac1000_core_init(struct mac_device_info *hw, struct net_device *dev) { @@ -435,6 +443,7 @@ static void dwmac1000_set_mac_loopback(void __iomem *ioaddr, bool enable) } const struct stmmac_ops dwmac1000_ops = { + .pcs_init = dwmac1000_pcs_init, .core_init = dwmac1000_core_init, .set_mac = stmmac_set_mac, .rx_ipc = dwmac1000_rx_ipc_enable, diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c index d855ab6b9145..0b785389b7ef 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c @@ -22,6 +22,14 @@ #include "dwmac4.h" #include "dwmac5.h" +static int dwmac4_pcs_init(struct stmmac_priv *priv) +{ + if (!priv->dma_cap.pcs) + return 0; + + return stmmac_integrated_pcs_init(priv, GMAC_PCS_BASE); +} + static void dwmac4_core_init(struct mac_device_info *hw, struct net_device *dev) { @@ -875,6 +883,7 @@ static int dwmac4_config_l4_filter(struct mac_device_info *hw, u32 filter_no, } const struct stmmac_ops dwmac4_ops = { + .pcs_init = dwmac4_pcs_init, .core_init = dwmac4_core_init, .update_caps = dwmac4_update_caps, .set_mac = stmmac_set_mac, @@ -909,6 +918,7 @@ const struct stmmac_ops dwmac4_ops = { }; const struct stmmac_ops dwmac410_ops = { + .pcs_init = dwmac4_pcs_init, .core_init = dwmac4_core_init, .update_caps = dwmac4_update_caps, .set_mac = stmmac_dwmac4_set_mac, @@ -945,6 +955,7 @@ const struct stmmac_ops dwmac410_ops = { }; const struct stmmac_ops dwmac510_ops = { + .pcs_init = dwmac4_pcs_init, .core_init = dwmac4_core_init, .update_caps = dwmac4_update_caps, .set_mac = stmmac_dwmac4_set_mac, diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac.h b/drivers/net/ethernet/stmicro/stmmac/stmmac.h index ec8bddc1c37f..aaeaf42084f0 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac.h +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac.h @@ -25,6 +25,8 @@ #include #include +struct stmmac_pcs; + struct stmmac_resources { void __iomem *addr; u8 mac[ETH_ALEN]; @@ -273,6 +275,8 @@ struct stmmac_priv { unsigned int pause_time; struct mii_bus *mii; + struct stmmac_pcs *integrated_pcs; + struct phylink_config phylink_config; struct phylink *phylink; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index c041268d3a8d..5e6aaead5894 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -46,6 +46,7 @@ #include "stmmac_ptp.h" #include "stmmac_fpe.h" #include "stmmac.h" +#include "stmmac_pcs.h" #include "stmmac_xdp.h" #include #include @@ -850,6 +851,13 @@ static struct phylink_pcs *stmmac_mac_select_pcs(struct phylink_config *config, return pcs; } + /* The PCS control register is only relevant for SGMII, TBI and RTBI + * modes. We no longer support TBI or RTBI, so only configure this + * register when operating in SGMII mode with the integrated PCS. + */ + if (priv->hw->pcs & STMMAC_PCS_SGMII && priv->integrated_pcs) + return &priv->integrated_pcs->pcs; + return NULL; } @@ -3487,13 +3495,6 @@ static int stmmac_hw_setup(struct net_device *dev) } } - /* The PCS control register is only relevant for SGMII, TBI and RTBI - * modes. We no longer support TBI or RTBI, so only configure this - * register when operating in SGMII mode with the integrated PCS. - */ - if (priv->hw->pcs & STMMAC_PCS_SGMII) - stmmac_pcs_ctrl_ane(priv, 1, priv->hw->reverse_sgmii_enable); - /* set TX and RX rings length */ stmmac_set_rings_length(priv); diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.c new file mode 100644 index 000000000000..50ea51d7a1cc --- /dev/null +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.c @@ -0,0 +1,47 @@ +// SPDX-License-Identifier: GPL-2.0-only +#include "stmmac.h" +#include "stmmac_pcs.h" + +static void dwmac_integrated_pcs_get_state(struct phylink_pcs *pcs, + unsigned int neg_mode, + struct phylink_link_state *state) +{ + state->link = false; +} + +static int dwmac_integrated_pcs_config(struct phylink_pcs *pcs, + unsigned int neg_mode, + phy_interface_t interface, + const unsigned long *advertising, + bool permit_pause_to_mac) +{ + struct stmmac_pcs *spcs = phylink_pcs_to_stmmac_pcs(pcs); + + dwmac_ctrl_ane(spcs->base, 0, 1, spcs->priv->hw->reverse_sgmii_enable); + + return 0; +} + +static const struct phylink_pcs_ops dwmac_integrated_pcs_ops = { + .pcs_get_state = dwmac_integrated_pcs_get_state, + .pcs_config = dwmac_integrated_pcs_config, +}; + +int stmmac_integrated_pcs_init(struct stmmac_priv *priv, unsigned int offset) +{ + struct stmmac_pcs *spcs; + + spcs = devm_kzalloc(priv->device, sizeof(*spcs), GFP_KERNEL); + if (!spcs) + return -ENOMEM; + + spcs->priv = priv; + spcs->base = priv->ioaddr + offset; + spcs->pcs.ops = &dwmac_integrated_pcs_ops; + + __set_bit(PHY_INTERFACE_MODE_SGMII, spcs->pcs.supported_interfaces); + + priv->integrated_pcs = spcs; + + return 0; +} diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.h b/drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.h index 5778f5b2f313..64397ac8ecab 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.h +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.h @@ -9,6 +9,7 @@ #ifndef __STMMAC_PCS_H__ #define __STMMAC_PCS_H__ +#include #include #include #include "common.h" @@ -46,6 +47,22 @@ #define GMAC_ANE_RFE_SHIFT 12 #define GMAC_ANE_ACK BIT(14) +struct stmmac_priv; + +struct stmmac_pcs { + struct stmmac_priv *priv; + void __iomem *base; + struct phylink_pcs pcs; +}; + +static inline struct stmmac_pcs * +phylink_pcs_to_stmmac_pcs(struct phylink_pcs *pcs) +{ + return container_of(pcs, struct stmmac_pcs, pcs); +} + +int stmmac_integrated_pcs_init(struct stmmac_priv *priv, unsigned int offset); + /** * dwmac_pcs_isr - TBI, RTBI, or SGMII PHY ISR * @ioaddr: IO registers pointer From ed5d5928bd54f66af19b71ad342ebf0947d50674 Mon Sep 17 00:00:00 2001 From: Vadim Fedorenko Date: Thu, 16 Oct 2025 15:25:09 +0000 Subject: [PATCH 0137/1381] net: ti: am65-cpsw: move hw timestamping to ndo callback Migrate driver to new API for HW timestamping. Reviewed-by: Simon Horman Reviewed-by: Jacob Keller Signed-off-by: Vadim Fedorenko Link: https://patch.msgid.link/20251016152515.3510991-2-vadim.fedorenko@linux.dev Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/ti/am65-cpsw-nuss.c | 44 +++++++++++------------- 1 file changed, 20 insertions(+), 24 deletions(-) diff --git a/drivers/net/ethernet/ti/am65-cpsw-nuss.c b/drivers/net/ethernet/ti/am65-cpsw-nuss.c index 110eb2da8dbc..d5f358ec9820 100644 --- a/drivers/net/ethernet/ti/am65-cpsw-nuss.c +++ b/drivers/net/ethernet/ti/am65-cpsw-nuss.c @@ -1788,28 +1788,28 @@ static int am65_cpsw_nuss_ndo_slave_set_mac_address(struct net_device *ndev, } static int am65_cpsw_nuss_hwtstamp_set(struct net_device *ndev, - struct ifreq *ifr) + struct kernel_hwtstamp_config *cfg, + struct netlink_ext_ack *extack) { struct am65_cpsw_port *port = am65_ndev_to_port(ndev); u32 ts_ctrl, seq_id, ts_ctrl_ltype2, ts_vlan_ltype; - struct hwtstamp_config cfg; - if (!IS_ENABLED(CONFIG_TI_K3_AM65_CPTS)) + if (!IS_ENABLED(CONFIG_TI_K3_AM65_CPTS)) { + NL_SET_ERR_MSG(extack, "Time stamping is not supported"); return -EOPNOTSUPP; - - if (copy_from_user(&cfg, ifr->ifr_data, sizeof(cfg))) - return -EFAULT; + } /* TX HW timestamp */ - switch (cfg.tx_type) { + switch (cfg->tx_type) { case HWTSTAMP_TX_OFF: case HWTSTAMP_TX_ON: break; default: + NL_SET_ERR_MSG(extack, "TX mode is not supported"); return -ERANGE; } - switch (cfg.rx_filter) { + switch (cfg->rx_filter) { case HWTSTAMP_FILTER_NONE: port->rx_ts_enabled = false; break; @@ -1826,17 +1826,19 @@ static int am65_cpsw_nuss_hwtstamp_set(struct net_device *ndev, case HWTSTAMP_FILTER_PTP_V2_SYNC: case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ: port->rx_ts_enabled = true; - cfg.rx_filter = HWTSTAMP_FILTER_PTP_V2_EVENT | HWTSTAMP_FILTER_PTP_V1_L4_EVENT; + cfg->rx_filter = HWTSTAMP_FILTER_PTP_V2_EVENT | HWTSTAMP_FILTER_PTP_V1_L4_EVENT; break; case HWTSTAMP_FILTER_ALL: case HWTSTAMP_FILTER_SOME: case HWTSTAMP_FILTER_NTP_ALL: + NL_SET_ERR_MSG(extack, "RX filter is not supported"); return -EOPNOTSUPP; default: + NL_SET_ERR_MSG(extack, "RX filter is not supported"); return -ERANGE; } - port->tx_ts_enabled = (cfg.tx_type == HWTSTAMP_TX_ON); + port->tx_ts_enabled = (cfg->tx_type == HWTSTAMP_TX_ON); /* cfg TX timestamp */ seq_id = (AM65_CPSW_TS_SEQ_ID_OFFSET << @@ -1872,25 +1874,24 @@ static int am65_cpsw_nuss_hwtstamp_set(struct net_device *ndev, AM65_CPSW_PORTN_REG_TS_CTL_LTYPE2); writel(ts_ctrl, port->port_base + AM65_CPSW_PORTN_REG_TS_CTL); - return copy_to_user(ifr->ifr_data, &cfg, sizeof(cfg)) ? -EFAULT : 0; + return 0; } static int am65_cpsw_nuss_hwtstamp_get(struct net_device *ndev, - struct ifreq *ifr) + struct kernel_hwtstamp_config *cfg) { struct am65_cpsw_port *port = am65_ndev_to_port(ndev); - struct hwtstamp_config cfg; if (!IS_ENABLED(CONFIG_TI_K3_AM65_CPTS)) return -EOPNOTSUPP; - cfg.flags = 0; - cfg.tx_type = port->tx_ts_enabled ? + cfg->flags = 0; + cfg->tx_type = port->tx_ts_enabled ? HWTSTAMP_TX_ON : HWTSTAMP_TX_OFF; - cfg.rx_filter = port->rx_ts_enabled ? HWTSTAMP_FILTER_PTP_V2_EVENT | + cfg->rx_filter = port->rx_ts_enabled ? HWTSTAMP_FILTER_PTP_V2_EVENT | HWTSTAMP_FILTER_PTP_V1_L4_EVENT : HWTSTAMP_FILTER_NONE; - return copy_to_user(ifr->ifr_data, &cfg, sizeof(cfg)) ? -EFAULT : 0; + return 0; } static int am65_cpsw_nuss_ndo_slave_ioctl(struct net_device *ndev, @@ -1901,13 +1902,6 @@ static int am65_cpsw_nuss_ndo_slave_ioctl(struct net_device *ndev, if (!netif_running(ndev)) return -EINVAL; - switch (cmd) { - case SIOCSHWTSTAMP: - return am65_cpsw_nuss_hwtstamp_set(ndev, req); - case SIOCGHWTSTAMP: - return am65_cpsw_nuss_hwtstamp_get(ndev, req); - } - return phylink_mii_ioctl(port->slave.phylink, req, cmd); } @@ -1991,6 +1985,8 @@ static const struct net_device_ops am65_cpsw_nuss_netdev_ops = { .ndo_set_tx_maxrate = am65_cpsw_qos_ndo_tx_p0_set_maxrate, .ndo_bpf = am65_cpsw_ndo_bpf, .ndo_xdp_xmit = am65_cpsw_ndo_xdp_xmit, + .ndo_hwtstamp_get = am65_cpsw_nuss_hwtstamp_get, + .ndo_hwtstamp_set = am65_cpsw_nuss_hwtstamp_set, }; static void am65_cpsw_disable_phy(struct phy *phy) From b8fa98ea4a22d949e92ebc755abc5e633db88bf9 Mon Sep 17 00:00:00 2001 From: Vadim Fedorenko Date: Thu, 16 Oct 2025 15:25:10 +0000 Subject: [PATCH 0138/1381] ti: icssg: convert to ndo_hwtstamp API Convert driver to use .ndo_hwtstamp_get()/.ndo_hwtstamp_set() API. .ndo_eth_ioctl() implementation becomes pure phy_do_ioctl(), remove it from common module, remove exported symbol and replace ndo callback. Reviewed-by: Simon Horman Reviewed-by: Jacob Keller Signed-off-by: Vadim Fedorenko Link: https://patch.msgid.link/20251016152515.3510991-3-vadim.fedorenko@linux.dev Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/ti/icssg/icssg_common.c | 47 ++++++------------- drivers/net/ethernet/ti/icssg/icssg_prueth.c | 4 +- drivers/net/ethernet/ti/icssg/icssg_prueth.h | 6 ++- .../net/ethernet/ti/icssg/icssg_prueth_sr1.c | 4 +- 4 files changed, 26 insertions(+), 35 deletions(-) diff --git a/drivers/net/ethernet/ti/icssg/icssg_common.c b/drivers/net/ethernet/ti/icssg/icssg_common.c index 57e5f1c88f50..0eed29d6187a 100644 --- a/drivers/net/ethernet/ti/icssg/icssg_common.c +++ b/drivers/net/ethernet/ti/icssg/icssg_common.c @@ -1223,15 +1223,13 @@ void icssg_ndo_tx_timeout(struct net_device *ndev, unsigned int txqueue) } EXPORT_SYMBOL_GPL(icssg_ndo_tx_timeout); -static int emac_set_ts_config(struct net_device *ndev, struct ifreq *ifr) +int icssg_ndo_set_ts_config(struct net_device *ndev, + struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack) { struct prueth_emac *emac = netdev_priv(ndev); - struct hwtstamp_config config; - if (copy_from_user(&config, ifr->ifr_data, sizeof(config))) - return -EFAULT; - - switch (config.tx_type) { + switch (config->tx_type) { case HWTSTAMP_TX_OFF: emac->tx_ts_enabled = 0; break; @@ -1242,7 +1240,7 @@ static int emac_set_ts_config(struct net_device *ndev, struct ifreq *ifr) return -ERANGE; } - switch (config.rx_filter) { + switch (config->rx_filter) { case HWTSTAMP_FILTER_NONE: emac->rx_ts_enabled = 0; break; @@ -1262,43 +1260,28 @@ static int emac_set_ts_config(struct net_device *ndev, struct ifreq *ifr) case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ: case HWTSTAMP_FILTER_NTP_ALL: emac->rx_ts_enabled = 1; - config.rx_filter = HWTSTAMP_FILTER_ALL; + config->rx_filter = HWTSTAMP_FILTER_ALL; break; default: return -ERANGE; } - return copy_to_user(ifr->ifr_data, &config, sizeof(config)) ? - -EFAULT : 0; + return 0; } +EXPORT_SYMBOL_GPL(icssg_ndo_set_ts_config); -static int emac_get_ts_config(struct net_device *ndev, struct ifreq *ifr) +int icssg_ndo_get_ts_config(struct net_device *ndev, + struct kernel_hwtstamp_config *config) { struct prueth_emac *emac = netdev_priv(ndev); - struct hwtstamp_config config; - config.flags = 0; - config.tx_type = emac->tx_ts_enabled ? HWTSTAMP_TX_ON : HWTSTAMP_TX_OFF; - config.rx_filter = emac->rx_ts_enabled ? HWTSTAMP_FILTER_ALL : HWTSTAMP_FILTER_NONE; + config->flags = 0; + config->tx_type = emac->tx_ts_enabled ? HWTSTAMP_TX_ON : HWTSTAMP_TX_OFF; + config->rx_filter = emac->rx_ts_enabled ? HWTSTAMP_FILTER_ALL : HWTSTAMP_FILTER_NONE; - return copy_to_user(ifr->ifr_data, &config, sizeof(config)) ? - -EFAULT : 0; + return 0; } - -int icssg_ndo_ioctl(struct net_device *ndev, struct ifreq *ifr, int cmd) -{ - switch (cmd) { - case SIOCGHWTSTAMP: - return emac_get_ts_config(ndev, ifr); - case SIOCSHWTSTAMP: - return emac_set_ts_config(ndev, ifr); - default: - break; - } - - return phy_do_ioctl(ndev, ifr, cmd); -} -EXPORT_SYMBOL_GPL(icssg_ndo_ioctl); +EXPORT_SYMBOL_GPL(icssg_ndo_get_ts_config); void icssg_ndo_get_stats64(struct net_device *ndev, struct rtnl_link_stats64 *stats) diff --git a/drivers/net/ethernet/ti/icssg/icssg_prueth.c b/drivers/net/ethernet/ti/icssg/icssg_prueth.c index e42d0fdefee1..1c1f4394ff1f 100644 --- a/drivers/net/ethernet/ti/icssg/icssg_prueth.c +++ b/drivers/net/ethernet/ti/icssg/icssg_prueth.c @@ -1168,7 +1168,7 @@ static const struct net_device_ops emac_netdev_ops = { .ndo_validate_addr = eth_validate_addr, .ndo_tx_timeout = icssg_ndo_tx_timeout, .ndo_set_rx_mode = emac_ndo_set_rx_mode, - .ndo_eth_ioctl = icssg_ndo_ioctl, + .ndo_eth_ioctl = phy_do_ioctl, .ndo_get_stats64 = icssg_ndo_get_stats64, .ndo_get_phys_port_name = icssg_ndo_get_phys_port_name, .ndo_fix_features = emac_ndo_fix_features, @@ -1176,6 +1176,8 @@ static const struct net_device_ops emac_netdev_ops = { .ndo_vlan_rx_kill_vid = emac_ndo_vlan_rx_del_vid, .ndo_bpf = emac_ndo_bpf, .ndo_xdp_xmit = emac_xdp_xmit, + .ndo_hwtstamp_get = icssg_ndo_get_ts_config, + .ndo_hwtstamp_set = icssg_ndo_set_ts_config, }; static int prueth_netdev_init(struct prueth *prueth, diff --git a/drivers/net/ethernet/ti/icssg/icssg_prueth.h b/drivers/net/ethernet/ti/icssg/icssg_prueth.h index ca8a22a4a5da..f0fa9688d9a0 100644 --- a/drivers/net/ethernet/ti/icssg/icssg_prueth.h +++ b/drivers/net/ethernet/ti/icssg/icssg_prueth.h @@ -479,7 +479,11 @@ void prueth_reset_tx_chan(struct prueth_emac *emac, int ch_num, void prueth_reset_rx_chan(struct prueth_rx_chn *chn, int num_flows, bool disable); void icssg_ndo_tx_timeout(struct net_device *ndev, unsigned int txqueue); -int icssg_ndo_ioctl(struct net_device *ndev, struct ifreq *ifr, int cmd); +int icssg_ndo_get_ts_config(struct net_device *ndev, + struct kernel_hwtstamp_config *config); +int icssg_ndo_set_ts_config(struct net_device *ndev, + struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack); void icssg_ndo_get_stats64(struct net_device *ndev, struct rtnl_link_stats64 *stats); int icssg_ndo_get_phys_port_name(struct net_device *ndev, char *name, diff --git a/drivers/net/ethernet/ti/icssg/icssg_prueth_sr1.c b/drivers/net/ethernet/ti/icssg/icssg_prueth_sr1.c index 5e225310c9de..2a8c8847a6bd 100644 --- a/drivers/net/ethernet/ti/icssg/icssg_prueth_sr1.c +++ b/drivers/net/ethernet/ti/icssg/icssg_prueth_sr1.c @@ -747,9 +747,11 @@ static const struct net_device_ops emac_netdev_ops = { .ndo_validate_addr = eth_validate_addr, .ndo_tx_timeout = icssg_ndo_tx_timeout, .ndo_set_rx_mode = emac_ndo_set_rx_mode_sr1, - .ndo_eth_ioctl = icssg_ndo_ioctl, + .ndo_eth_ioctl = phy_do_ioctl, .ndo_get_stats64 = icssg_ndo_get_stats64, .ndo_get_phys_port_name = icssg_ndo_get_phys_port_name, + .ndo_hwtstamp_get = icssg_ndo_get_ts_config, + .ndo_hwtstamp_set = icssg_ndo_set_ts_config, }; static int prueth_netdev_init(struct prueth *prueth, From 149cfae71166db1051e1160f4f86ad3ef5a458ae Mon Sep 17 00:00:00 2001 From: Vadim Fedorenko Date: Thu, 16 Oct 2025 15:25:11 +0000 Subject: [PATCH 0139/1381] amd-xgbe: convert to ndo_hwtstamp callbacks Convert driver to use .ndo_hwtstamp_get()/.ndo_hwtstamp_set() callbacks. .ndo_eth_ioctl() becomes empty function, remove it. Reviewed-by: Simon Horman Reviewed-by: Jacob Keller Signed-off-by: Vadim Fedorenko Link: https://patch.msgid.link/20251016152515.3510991-4-vadim.fedorenko@linux.dev Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/amd/xgbe/xgbe-drv.c | 24 ++-------------- drivers/net/ethernet/amd/xgbe/xgbe-hwtstamp.c | 28 +++++++++---------- drivers/net/ethernet/amd/xgbe/xgbe.h | 11 ++++---- 3 files changed, 21 insertions(+), 42 deletions(-) diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c index 4dc631af7933..f3adf29b222b 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c @@ -1754,27 +1754,6 @@ static int xgbe_set_mac_address(struct net_device *netdev, void *addr) return 0; } -static int xgbe_ioctl(struct net_device *netdev, struct ifreq *ifreq, int cmd) -{ - struct xgbe_prv_data *pdata = netdev_priv(netdev); - int ret; - - switch (cmd) { - case SIOCGHWTSTAMP: - ret = xgbe_get_hwtstamp_settings(pdata, ifreq); - break; - - case SIOCSHWTSTAMP: - ret = xgbe_set_hwtstamp_settings(pdata, ifreq); - break; - - default: - ret = -EOPNOTSUPP; - } - - return ret; -} - static int xgbe_change_mtu(struct net_device *netdev, int mtu) { struct xgbe_prv_data *pdata = netdev_priv(netdev); @@ -2020,7 +1999,6 @@ static const struct net_device_ops xgbe_netdev_ops = { .ndo_set_rx_mode = xgbe_set_rx_mode, .ndo_set_mac_address = xgbe_set_mac_address, .ndo_validate_addr = eth_validate_addr, - .ndo_eth_ioctl = xgbe_ioctl, .ndo_change_mtu = xgbe_change_mtu, .ndo_tx_timeout = xgbe_tx_timeout, .ndo_get_stats64 = xgbe_get_stats64, @@ -2033,6 +2011,8 @@ static const struct net_device_ops xgbe_netdev_ops = { .ndo_fix_features = xgbe_fix_features, .ndo_set_features = xgbe_set_features, .ndo_features_check = xgbe_features_check, + .ndo_hwtstamp_get = xgbe_get_hwtstamp_settings, + .ndo_hwtstamp_set = xgbe_set_hwtstamp_settings, }; const struct net_device_ops *xgbe_get_netdev_ops(void) diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-hwtstamp.c b/drivers/net/ethernet/amd/xgbe/xgbe-hwtstamp.c index bc52e5ec6420..0127988e10be 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-hwtstamp.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-hwtstamp.c @@ -157,26 +157,24 @@ void xgbe_tx_tstamp(struct work_struct *work) spin_unlock_irqrestore(&pdata->tstamp_lock, flags); } -int xgbe_get_hwtstamp_settings(struct xgbe_prv_data *pdata, struct ifreq *ifreq) +int xgbe_get_hwtstamp_settings(struct net_device *netdev, + struct kernel_hwtstamp_config *config) { - if (copy_to_user(ifreq->ifr_data, &pdata->tstamp_config, - sizeof(pdata->tstamp_config))) - return -EFAULT; + struct xgbe_prv_data *pdata = netdev_priv(netdev); + + *config = pdata->tstamp_config; return 0; } -int xgbe_set_hwtstamp_settings(struct xgbe_prv_data *pdata, struct ifreq *ifreq) +int xgbe_set_hwtstamp_settings(struct net_device *netdev, + struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack) { - struct hwtstamp_config config; - unsigned int mac_tscr; + struct xgbe_prv_data *pdata = netdev_priv(netdev); + unsigned int mac_tscr = 0; - if (copy_from_user(&config, ifreq->ifr_data, sizeof(config))) - return -EFAULT; - - mac_tscr = 0; - - switch (config.tx_type) { + switch (config->tx_type) { case HWTSTAMP_TX_OFF: break; @@ -188,7 +186,7 @@ int xgbe_set_hwtstamp_settings(struct xgbe_prv_data *pdata, struct ifreq *ifreq) return -ERANGE; } - switch (config.rx_filter) { + switch (config->rx_filter) { case HWTSTAMP_FILTER_NONE: break; @@ -290,7 +288,7 @@ int xgbe_set_hwtstamp_settings(struct xgbe_prv_data *pdata, struct ifreq *ifreq) xgbe_config_tstamp(pdata, mac_tscr); - memcpy(&pdata->tstamp_config, &config, sizeof(config)); + pdata->tstamp_config = *config; return 0; } diff --git a/drivers/net/ethernet/amd/xgbe/xgbe.h b/drivers/net/ethernet/amd/xgbe/xgbe.h index e8bbb6805901..381f72a33d1a 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe.h +++ b/drivers/net/ethernet/amd/xgbe/xgbe.h @@ -1146,7 +1146,7 @@ struct xgbe_prv_data { spinlock_t tstamp_lock; struct ptp_clock_info ptp_clock_info; struct ptp_clock *ptp_clock; - struct hwtstamp_config tstamp_config; + struct kernel_hwtstamp_config tstamp_config; unsigned int tstamp_addend; struct work_struct tx_tstamp_work; struct sk_buff *tx_tstamp_skb; @@ -1307,10 +1307,11 @@ void xgbe_update_tstamp_addend(struct xgbe_prv_data *pdata, void xgbe_set_tstamp_time(struct xgbe_prv_data *pdata, unsigned int sec, unsigned int nsec); void xgbe_tx_tstamp(struct work_struct *work); -int xgbe_get_hwtstamp_settings(struct xgbe_prv_data *pdata, - struct ifreq *ifreq); -int xgbe_set_hwtstamp_settings(struct xgbe_prv_data *pdata, - struct ifreq *ifreq); +int xgbe_get_hwtstamp_settings(struct net_device *netdev, + struct kernel_hwtstamp_config *config); +int xgbe_set_hwtstamp_settings(struct net_device *netdev, + struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack); void xgbe_prep_tx_tstamp(struct xgbe_prv_data *pdata, struct sk_buff *skb, struct xgbe_packet_data *packet); From 8a15a84e80dcf838062971f32fae653906b131f4 Mon Sep 17 00:00:00 2001 From: Vadim Fedorenko Date: Thu, 16 Oct 2025 15:25:12 +0000 Subject: [PATCH 0140/1381] net: atlantic: convert to ndo_hwtstamp API Convert driver to .ndo_hwtstamp_get()/.ndo_hwtstamp_set() callbacks. .ndo_eth_ioctl() becomes empty so remove it. Also simplify code with no functional changes. Reviewed-by: Simon Horman Reviewed-by: Jacob Keller Signed-off-by: Vadim Fedorenko Link: https://patch.msgid.link/20251016152515.3510991-5-vadim.fedorenko@linux.dev Signed-off-by: Jakub Kicinski --- .../net/ethernet/aquantia/atlantic/aq_main.c | 68 +++++-------------- .../net/ethernet/aquantia/atlantic/aq_ptp.c | 6 +- .../net/ethernet/aquantia/atlantic/aq_ptp.h | 8 +-- 3 files changed, 23 insertions(+), 59 deletions(-) diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_main.c b/drivers/net/ethernet/aquantia/atlantic/aq_main.c index b565189e5913..4ef4fe64b8ac 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_main.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_main.c @@ -258,10 +258,15 @@ static void aq_ndev_set_multicast_settings(struct net_device *ndev) (void)aq_nic_set_multicast_list(aq_nic, ndev); } -#if IS_REACHABLE(CONFIG_PTP_1588_CLOCK) -static int aq_ndev_config_hwtstamp(struct aq_nic_s *aq_nic, - struct hwtstamp_config *config) +static int aq_ndev_hwtstamp_set(struct net_device *netdev, + struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack) { + struct aq_nic_s *aq_nic = netdev_priv(netdev); + + if (!IS_REACHABLE(CONFIG_PTP_1588_CLOCK) || !aq_nic->aq_ptp) + return -EOPNOTSUPP; + switch (config->tx_type) { case HWTSTAMP_TX_OFF: case HWTSTAMP_TX_ON: @@ -290,59 +295,17 @@ static int aq_ndev_config_hwtstamp(struct aq_nic_s *aq_nic, return aq_ptp_hwtstamp_config_set(aq_nic->aq_ptp, config); } -#endif -static int aq_ndev_hwtstamp_set(struct aq_nic_s *aq_nic, struct ifreq *ifr) -{ - struct hwtstamp_config config; -#if IS_REACHABLE(CONFIG_PTP_1588_CLOCK) - int ret_val; -#endif - - if (!aq_nic->aq_ptp) - return -EOPNOTSUPP; - - if (copy_from_user(&config, ifr->ifr_data, sizeof(config))) - return -EFAULT; -#if IS_REACHABLE(CONFIG_PTP_1588_CLOCK) - ret_val = aq_ndev_config_hwtstamp(aq_nic, &config); - if (ret_val) - return ret_val; -#endif - - return copy_to_user(ifr->ifr_data, &config, sizeof(config)) ? - -EFAULT : 0; -} - -#if IS_REACHABLE(CONFIG_PTP_1588_CLOCK) -static int aq_ndev_hwtstamp_get(struct aq_nic_s *aq_nic, struct ifreq *ifr) -{ - struct hwtstamp_config config; - - if (!aq_nic->aq_ptp) - return -EOPNOTSUPP; - - aq_ptp_hwtstamp_config_get(aq_nic->aq_ptp, &config); - return copy_to_user(ifr->ifr_data, &config, sizeof(config)) ? - -EFAULT : 0; -} -#endif - -static int aq_ndev_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) +static int aq_ndev_hwtstamp_get(struct net_device *netdev, + struct kernel_hwtstamp_config *config) { struct aq_nic_s *aq_nic = netdev_priv(netdev); - switch (cmd) { - case SIOCSHWTSTAMP: - return aq_ndev_hwtstamp_set(aq_nic, ifr); + if (!aq_nic->aq_ptp) + return -EOPNOTSUPP; -#if IS_REACHABLE(CONFIG_PTP_1588_CLOCK) - case SIOCGHWTSTAMP: - return aq_ndev_hwtstamp_get(aq_nic, ifr); -#endif - } - - return -EOPNOTSUPP; + aq_ptp_hwtstamp_config_get(aq_nic->aq_ptp, config); + return 0; } static int aq_ndo_vlan_rx_add_vid(struct net_device *ndev, __be16 proto, @@ -500,12 +463,13 @@ static const struct net_device_ops aq_ndev_ops = { .ndo_set_mac_address = aq_ndev_set_mac_address, .ndo_set_features = aq_ndev_set_features, .ndo_fix_features = aq_ndev_fix_features, - .ndo_eth_ioctl = aq_ndev_ioctl, .ndo_vlan_rx_add_vid = aq_ndo_vlan_rx_add_vid, .ndo_vlan_rx_kill_vid = aq_ndo_vlan_rx_kill_vid, .ndo_setup_tc = aq_ndo_setup_tc, .ndo_bpf = aq_xdp, .ndo_xdp_xmit = aq_xdp_xmit, + .ndo_hwtstamp_get = aq_ndev_hwtstamp_get, + .ndo_hwtstamp_set = aq_ndev_hwtstamp_set, }; static int __init aq_ndev_init_module(void) diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_ptp.c b/drivers/net/ethernet/aquantia/atlantic/aq_ptp.c index 5acb3e16b567..0fa0f891c0e0 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_ptp.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_ptp.c @@ -51,7 +51,7 @@ struct ptp_tx_timeout { struct aq_ptp_s { struct aq_nic_s *aq_nic; - struct hwtstamp_config hwtstamp_config; + struct kernel_hwtstamp_config hwtstamp_config; spinlock_t ptp_lock; spinlock_t ptp_ring_lock; struct ptp_clock *ptp_clock; @@ -567,7 +567,7 @@ static void aq_ptp_rx_hwtstamp(struct aq_ptp_s *aq_ptp, struct skb_shared_hwtsta } void aq_ptp_hwtstamp_config_get(struct aq_ptp_s *aq_ptp, - struct hwtstamp_config *config) + struct kernel_hwtstamp_config *config) { *config = aq_ptp->hwtstamp_config; } @@ -588,7 +588,7 @@ static void aq_ptp_prepare_filters(struct aq_ptp_s *aq_ptp) } int aq_ptp_hwtstamp_config_set(struct aq_ptp_s *aq_ptp, - struct hwtstamp_config *config) + struct kernel_hwtstamp_config *config) { struct aq_nic_s *aq_nic = aq_ptp->aq_nic; const struct aq_hw_ops *hw_ops; diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_ptp.h b/drivers/net/ethernet/aquantia/atlantic/aq_ptp.h index 210b723f2207..5e643ec7cc06 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_ptp.h +++ b/drivers/net/ethernet/aquantia/atlantic/aq_ptp.h @@ -60,9 +60,9 @@ void aq_ptp_tx_hwtstamp(struct aq_nic_s *aq_nic, u64 timestamp); /* Must be to check available of PTP before call */ void aq_ptp_hwtstamp_config_get(struct aq_ptp_s *aq_ptp, - struct hwtstamp_config *config); + struct kernel_hwtstamp_config *config); int aq_ptp_hwtstamp_config_set(struct aq_ptp_s *aq_ptp, - struct hwtstamp_config *config); + struct kernel_hwtstamp_config *config); /* Return either ring is belong to PTP or not*/ bool aq_ptp_ring(struct aq_nic_s *aq_nic, struct aq_ring_s *ring); @@ -130,9 +130,9 @@ static inline int aq_ptp_xmit(struct aq_nic_s *aq_nic, struct sk_buff *skb) static inline void aq_ptp_tx_hwtstamp(struct aq_nic_s *aq_nic, u64 timestamp) {} static inline void aq_ptp_hwtstamp_config_get(struct aq_ptp_s *aq_ptp, - struct hwtstamp_config *config) {} + struct kernel_hwtstamp_config *config) {} static inline int aq_ptp_hwtstamp_config_set(struct aq_ptp_s *aq_ptp, - struct hwtstamp_config *config) + struct kernel_hwtstamp_config *config) { return 0; } From a6a64bb4115f15367573235292ef23a8c60f5f17 Mon Sep 17 00:00:00 2001 From: Vadim Fedorenko Date: Thu, 16 Oct 2025 15:25:13 +0000 Subject: [PATCH 0141/1381] cxgb4: convert to ndo_hwtstamp API Convert to use .ndo_hwtstamp_get()/.ndo_hwtstamp_set() callbacks. There is some change in the logic as well. Previously, the driver was storing newly requested configuration regardless of whether it was applied or not. In case of request validation error, inconsistent configuration would be returned by the driver. New logic stores configuration only if was properly validated and applied. It brings the consistency between reported and actual configuration. Signed-off-by: Vadim Fedorenko Reviewed-by: Jacob Keller Link: https://patch.msgid.link/20251016152515.3510991-6-vadim.fedorenko@linux.dev Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/chelsio/cxgb4/cxgb4.h | 2 +- .../net/ethernet/chelsio/cxgb4/cxgb4_main.c | 154 +++++++++--------- 2 files changed, 79 insertions(+), 77 deletions(-) diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h index 0d85198fb03d..f20f4bc58492 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h @@ -674,7 +674,7 @@ struct port_info { struct cxgb_fcoe fcoe; #endif /* CONFIG_CHELSIO_T4_FCOE */ bool rxtstamp; /* Enable TS */ - struct hwtstamp_config tstamp_config; + struct kernel_hwtstamp_config tstamp_config; bool ptp_enable; struct sched_table *sched_tbl; u32 eth_flags; diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c index 392723ef14e5..7e2283c95b97 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c @@ -3042,12 +3042,87 @@ static void cxgb_get_stats(struct net_device *dev, ns->rx_length_errors + stats.rx_len_err + ns->rx_fifo_errors; } +static int cxgb_hwtstamp_get(struct net_device *dev, + struct kernel_hwtstamp_config *config) +{ + struct port_info *pi = netdev_priv(dev); + + *config = pi->tstamp_config; + return 0; +} + +static int cxgb_hwtstamp_set(struct net_device *dev, + struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack) +{ + struct port_info *pi = netdev_priv(dev); + struct adapter *adapter = pi->adapter; + + if (is_t4(adapter->params.chip)) { + /* For T4 Adapters */ + switch (config->rx_filter) { + case HWTSTAMP_FILTER_NONE: + pi->rxtstamp = false; + break; + case HWTSTAMP_FILTER_ALL: + pi->rxtstamp = true; + break; + default: + return -ERANGE; + } + pi->tstamp_config = *config; + return 0; + } + + switch (config->tx_type) { + case HWTSTAMP_TX_OFF: + case HWTSTAMP_TX_ON: + break; + default: + return -ERANGE; + } + + switch (config->rx_filter) { + case HWTSTAMP_FILTER_NONE: + pi->rxtstamp = false; + break; + case HWTSTAMP_FILTER_PTP_V1_L4_EVENT: + case HWTSTAMP_FILTER_PTP_V2_L4_EVENT: + cxgb4_ptprx_timestamping(pi, pi->port_id, PTP_TS_L4); + break; + case HWTSTAMP_FILTER_PTP_V2_EVENT: + cxgb4_ptprx_timestamping(pi, pi->port_id, PTP_TS_L2_L4); + break; + case HWTSTAMP_FILTER_ALL: + case HWTSTAMP_FILTER_PTP_V1_L4_SYNC: + case HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ: + case HWTSTAMP_FILTER_PTP_V2_L4_SYNC: + case HWTSTAMP_FILTER_PTP_V2_L4_DELAY_REQ: + pi->rxtstamp = true; + break; + default: + return -ERANGE; + } + + if (config->tx_type == HWTSTAMP_TX_OFF && + config->rx_filter == HWTSTAMP_FILTER_NONE) { + if (cxgb4_ptp_txtype(adapter, pi->port_id) >= 0) + pi->ptp_enable = false; + } + + if (config->rx_filter != HWTSTAMP_FILTER_NONE) { + if (cxgb4_ptp_redirect_rx_packet(adapter, pi) >= 0) + pi->ptp_enable = true; + } + pi->tstamp_config = *config; + return 0; +} + static int cxgb_ioctl(struct net_device *dev, struct ifreq *req, int cmd) { unsigned int mbox; int ret = 0, prtad, devad; struct port_info *pi = netdev_priv(dev); - struct adapter *adapter = pi->adapter; struct mii_ioctl_data *data = (struct mii_ioctl_data *)&req->ifr_data; switch (cmd) { @@ -3076,81 +3151,6 @@ static int cxgb_ioctl(struct net_device *dev, struct ifreq *req, int cmd) ret = t4_mdio_wr(pi->adapter, mbox, prtad, devad, data->reg_num, data->val_in); break; - case SIOCGHWTSTAMP: - return copy_to_user(req->ifr_data, &pi->tstamp_config, - sizeof(pi->tstamp_config)) ? - -EFAULT : 0; - case SIOCSHWTSTAMP: - if (copy_from_user(&pi->tstamp_config, req->ifr_data, - sizeof(pi->tstamp_config))) - return -EFAULT; - - if (!is_t4(adapter->params.chip)) { - switch (pi->tstamp_config.tx_type) { - case HWTSTAMP_TX_OFF: - case HWTSTAMP_TX_ON: - break; - default: - return -ERANGE; - } - - switch (pi->tstamp_config.rx_filter) { - case HWTSTAMP_FILTER_NONE: - pi->rxtstamp = false; - break; - case HWTSTAMP_FILTER_PTP_V1_L4_EVENT: - case HWTSTAMP_FILTER_PTP_V2_L4_EVENT: - cxgb4_ptprx_timestamping(pi, pi->port_id, - PTP_TS_L4); - break; - case HWTSTAMP_FILTER_PTP_V2_EVENT: - cxgb4_ptprx_timestamping(pi, pi->port_id, - PTP_TS_L2_L4); - break; - case HWTSTAMP_FILTER_ALL: - case HWTSTAMP_FILTER_PTP_V1_L4_SYNC: - case HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ: - case HWTSTAMP_FILTER_PTP_V2_L4_SYNC: - case HWTSTAMP_FILTER_PTP_V2_L4_DELAY_REQ: - pi->rxtstamp = true; - break; - default: - pi->tstamp_config.rx_filter = - HWTSTAMP_FILTER_NONE; - return -ERANGE; - } - - if ((pi->tstamp_config.tx_type == HWTSTAMP_TX_OFF) && - (pi->tstamp_config.rx_filter == - HWTSTAMP_FILTER_NONE)) { - if (cxgb4_ptp_txtype(adapter, pi->port_id) >= 0) - pi->ptp_enable = false; - } - - if (pi->tstamp_config.rx_filter != - HWTSTAMP_FILTER_NONE) { - if (cxgb4_ptp_redirect_rx_packet(adapter, - pi) >= 0) - pi->ptp_enable = true; - } - } else { - /* For T4 Adapters */ - switch (pi->tstamp_config.rx_filter) { - case HWTSTAMP_FILTER_NONE: - pi->rxtstamp = false; - break; - case HWTSTAMP_FILTER_ALL: - pi->rxtstamp = true; - break; - default: - pi->tstamp_config.rx_filter = - HWTSTAMP_FILTER_NONE; - return -ERANGE; - } - } - return copy_to_user(req->ifr_data, &pi->tstamp_config, - sizeof(pi->tstamp_config)) ? - -EFAULT : 0; default: return -EOPNOTSUPP; } @@ -3875,6 +3875,8 @@ static const struct net_device_ops cxgb4_netdev_ops = { .ndo_setup_tc = cxgb_setup_tc, .ndo_features_check = cxgb_features_check, .ndo_fix_features = cxgb_fix_features, + .ndo_hwtstamp_get = cxgb_hwtstamp_get, + .ndo_hwtstamp_set = cxgb_hwtstamp_set, }; #ifdef CONFIG_PCI_IOV From d8db98db0d463299b891f9144b6e8924da051150 Mon Sep 17 00:00:00 2001 From: Vadim Fedorenko Date: Thu, 16 Oct 2025 15:25:14 +0000 Subject: [PATCH 0142/1381] tsnep: convert to ndo_hwtstatmp API Convert to .ndo_hwtstamp_get()/.ndo_hwtstamp_set() callbacks. After conversions the rest of tsnep_netdev_ioctl() becomes pure phy_do_ioctl_running(), so remove tsnep_netdev_ioctl() and replace it with phy_do_ioctl_running() in .ndo_eth_ioctl. Reviewed-by: Simon Horman Signed-off-by: Vadim Fedorenko Reviewed-by: Jacob Keller Reviewed-by: Gerhard Engleder Link: https://patch.msgid.link/20251016152515.3510991-7-vadim.fedorenko@linux.dev Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/engleder/tsnep.h | 8 ++- drivers/net/ethernet/engleder/tsnep_main.c | 14 +--- drivers/net/ethernet/engleder/tsnep_ptp.c | 82 ++++++++++------------ 3 files changed, 48 insertions(+), 56 deletions(-) diff --git a/drivers/net/ethernet/engleder/tsnep.h b/drivers/net/ethernet/engleder/tsnep.h index f188fba021a6..03e19aea9ea4 100644 --- a/drivers/net/ethernet/engleder/tsnep.h +++ b/drivers/net/ethernet/engleder/tsnep.h @@ -176,7 +176,7 @@ struct tsnep_adapter { struct tsnep_gcl gcl[2]; int next_gcl; - struct hwtstamp_config hwtstamp_config; + struct kernel_hwtstamp_config hwtstamp_config; struct ptp_clock *ptp_clock; struct ptp_clock_info ptp_clock_info; /* ptp clock lock */ @@ -203,7 +203,11 @@ extern const struct ethtool_ops tsnep_ethtool_ops; int tsnep_ptp_init(struct tsnep_adapter *adapter); void tsnep_ptp_cleanup(struct tsnep_adapter *adapter); -int tsnep_ptp_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd); +int tsnep_ptp_hwtstamp_get(struct net_device *netdev, + struct kernel_hwtstamp_config *config); +int tsnep_ptp_hwtstamp_set(struct net_device *netdev, + struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack); int tsnep_tc_init(struct tsnep_adapter *adapter); void tsnep_tc_cleanup(struct tsnep_adapter *adapter); diff --git a/drivers/net/ethernet/engleder/tsnep_main.c b/drivers/net/ethernet/engleder/tsnep_main.c index eba73246f986..b118407c30e8 100644 --- a/drivers/net/ethernet/engleder/tsnep_main.c +++ b/drivers/net/ethernet/engleder/tsnep_main.c @@ -2168,16 +2168,6 @@ static netdev_tx_t tsnep_netdev_xmit_frame(struct sk_buff *skb, return tsnep_xmit_frame_ring(skb, &adapter->tx[queue_mapping]); } -static int tsnep_netdev_ioctl(struct net_device *netdev, struct ifreq *ifr, - int cmd) -{ - if (!netif_running(netdev)) - return -EINVAL; - if (cmd == SIOCSHWTSTAMP || cmd == SIOCGHWTSTAMP) - return tsnep_ptp_ioctl(netdev, ifr, cmd); - return phy_mii_ioctl(netdev->phydev, ifr, cmd); -} - static void tsnep_netdev_set_multicast(struct net_device *netdev) { struct tsnep_adapter *adapter = netdev_priv(netdev); @@ -2384,7 +2374,7 @@ static const struct net_device_ops tsnep_netdev_ops = { .ndo_open = tsnep_netdev_open, .ndo_stop = tsnep_netdev_close, .ndo_start_xmit = tsnep_netdev_xmit_frame, - .ndo_eth_ioctl = tsnep_netdev_ioctl, + .ndo_eth_ioctl = phy_do_ioctl_running, .ndo_set_rx_mode = tsnep_netdev_set_multicast, .ndo_get_stats64 = tsnep_netdev_get_stats64, .ndo_set_mac_address = tsnep_netdev_set_mac_address, @@ -2394,6 +2384,8 @@ static const struct net_device_ops tsnep_netdev_ops = { .ndo_bpf = tsnep_netdev_bpf, .ndo_xdp_xmit = tsnep_netdev_xdp_xmit, .ndo_xsk_wakeup = tsnep_netdev_xsk_wakeup, + .ndo_hwtstamp_get = tsnep_ptp_hwtstamp_get, + .ndo_hwtstamp_set = tsnep_ptp_hwtstamp_set, }; static int tsnep_mac_init(struct tsnep_adapter *adapter) diff --git a/drivers/net/ethernet/engleder/tsnep_ptp.c b/drivers/net/ethernet/engleder/tsnep_ptp.c index 54fbf0126815..ae1308eb813d 100644 --- a/drivers/net/ethernet/engleder/tsnep_ptp.c +++ b/drivers/net/ethernet/engleder/tsnep_ptp.c @@ -19,57 +19,53 @@ void tsnep_get_system_time(struct tsnep_adapter *adapter, u64 *time) *time = (((u64)high) << 32) | ((u64)low); } -int tsnep_ptp_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) +int tsnep_ptp_hwtstamp_get(struct net_device *netdev, + struct kernel_hwtstamp_config *config) { struct tsnep_adapter *adapter = netdev_priv(netdev); - struct hwtstamp_config config; - if (!ifr) - return -EINVAL; + *config = adapter->hwtstamp_config; + return 0; +} - if (cmd == SIOCSHWTSTAMP) { - if (copy_from_user(&config, ifr->ifr_data, sizeof(config))) - return -EFAULT; +int tsnep_ptp_hwtstamp_set(struct net_device *netdev, + struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack) +{ + struct tsnep_adapter *adapter = netdev_priv(netdev); - switch (config.tx_type) { - case HWTSTAMP_TX_OFF: - case HWTSTAMP_TX_ON: - break; - default: - return -ERANGE; - } - - switch (config.rx_filter) { - case HWTSTAMP_FILTER_NONE: - break; - case HWTSTAMP_FILTER_ALL: - case HWTSTAMP_FILTER_PTP_V1_L4_EVENT: - case HWTSTAMP_FILTER_PTP_V1_L4_SYNC: - case HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ: - case HWTSTAMP_FILTER_PTP_V2_L4_EVENT: - case HWTSTAMP_FILTER_PTP_V2_L4_SYNC: - case HWTSTAMP_FILTER_PTP_V2_L4_DELAY_REQ: - case HWTSTAMP_FILTER_PTP_V2_L2_EVENT: - case HWTSTAMP_FILTER_PTP_V2_L2_SYNC: - case HWTSTAMP_FILTER_PTP_V2_L2_DELAY_REQ: - case HWTSTAMP_FILTER_PTP_V2_EVENT: - case HWTSTAMP_FILTER_PTP_V2_SYNC: - case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ: - case HWTSTAMP_FILTER_NTP_ALL: - config.rx_filter = HWTSTAMP_FILTER_ALL; - break; - default: - return -ERANGE; - } - - memcpy(&adapter->hwtstamp_config, &config, - sizeof(adapter->hwtstamp_config)); + switch (config->tx_type) { + case HWTSTAMP_TX_OFF: + case HWTSTAMP_TX_ON: + break; + default: + return -ERANGE; } - if (copy_to_user(ifr->ifr_data, &adapter->hwtstamp_config, - sizeof(adapter->hwtstamp_config))) - return -EFAULT; + switch (config->rx_filter) { + case HWTSTAMP_FILTER_NONE: + break; + case HWTSTAMP_FILTER_ALL: + case HWTSTAMP_FILTER_PTP_V1_L4_EVENT: + case HWTSTAMP_FILTER_PTP_V1_L4_SYNC: + case HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ: + case HWTSTAMP_FILTER_PTP_V2_L4_EVENT: + case HWTSTAMP_FILTER_PTP_V2_L4_SYNC: + case HWTSTAMP_FILTER_PTP_V2_L4_DELAY_REQ: + case HWTSTAMP_FILTER_PTP_V2_L2_EVENT: + case HWTSTAMP_FILTER_PTP_V2_L2_SYNC: + case HWTSTAMP_FILTER_PTP_V2_L2_DELAY_REQ: + case HWTSTAMP_FILTER_PTP_V2_EVENT: + case HWTSTAMP_FILTER_PTP_V2_SYNC: + case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ: + case HWTSTAMP_FILTER_NTP_ALL: + config->rx_filter = HWTSTAMP_FILTER_ALL; + break; + default: + return -ERANGE; + } + adapter->hwtstamp_config = *config; return 0; } From dc34040654e455564e676006faa89ff1ab2adfd7 Mon Sep 17 00:00:00 2001 From: Vadim Fedorenko Date: Thu, 16 Oct 2025 15:25:15 +0000 Subject: [PATCH 0143/1381] funeth: convert to ndo_hwtstamp API Convert driver to use .ndo_hwtstamp_get()/.ndo_hwtstamp_set() callbacks. .ndo_eth_ioctl() implementation becomes empty, remove it. Reviewed-by: Simon Horman Reviewed-by: Jacob Keller Signed-off-by: Vadim Fedorenko Link: https://patch.msgid.link/20251016152515.3510991-8-vadim.fedorenko@linux.dev Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/fungible/funeth/funeth.h | 4 +- .../ethernet/fungible/funeth/funeth_main.c | 40 +++++++------------ 2 files changed, 16 insertions(+), 28 deletions(-) diff --git a/drivers/net/ethernet/fungible/funeth/funeth.h b/drivers/net/ethernet/fungible/funeth/funeth.h index 1250e10d21db..55e705e239f8 100644 --- a/drivers/net/ethernet/fungible/funeth/funeth.h +++ b/drivers/net/ethernet/fungible/funeth/funeth.h @@ -4,7 +4,7 @@ #define _FUNETH_H #include -#include +#include #include #include #include @@ -121,7 +121,7 @@ struct funeth_priv { u8 rx_coal_usec; u8 rx_coal_count; - struct hwtstamp_config hwtstamp_cfg; + struct kernel_hwtstamp_config hwtstamp_cfg; /* cumulative queue stats from earlier queue instances */ u64 tx_packets; diff --git a/drivers/net/ethernet/fungible/funeth/funeth_main.c b/drivers/net/ethernet/fungible/funeth/funeth_main.c index ac86179a0a81..792cddac6f1b 100644 --- a/drivers/net/ethernet/fungible/funeth/funeth_main.c +++ b/drivers/net/ethernet/fungible/funeth/funeth_main.c @@ -1014,26 +1014,25 @@ static int fun_get_port_attributes(struct net_device *netdev) return 0; } -static int fun_hwtstamp_get(struct net_device *dev, struct ifreq *ifr) +static int fun_hwtstamp_get(struct net_device *dev, + struct kernel_hwtstamp_config *config) { const struct funeth_priv *fp = netdev_priv(dev); - return copy_to_user(ifr->ifr_data, &fp->hwtstamp_cfg, - sizeof(fp->hwtstamp_cfg)) ? -EFAULT : 0; + *config = fp->hwtstamp_cfg; + return 0; } -static int fun_hwtstamp_set(struct net_device *dev, struct ifreq *ifr) +static int fun_hwtstamp_set(struct net_device *dev, + struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack) { struct funeth_priv *fp = netdev_priv(dev); - struct hwtstamp_config cfg; - - if (copy_from_user(&cfg, ifr->ifr_data, sizeof(cfg))) - return -EFAULT; /* no TX HW timestamps */ - cfg.tx_type = HWTSTAMP_TX_OFF; + config->tx_type = HWTSTAMP_TX_OFF; - switch (cfg.rx_filter) { + switch (config->rx_filter) { case HWTSTAMP_FILTER_NONE: break; case HWTSTAMP_FILTER_ALL: @@ -1051,26 +1050,14 @@ static int fun_hwtstamp_set(struct net_device *dev, struct ifreq *ifr) case HWTSTAMP_FILTER_PTP_V2_SYNC: case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ: case HWTSTAMP_FILTER_NTP_ALL: - cfg.rx_filter = HWTSTAMP_FILTER_ALL; + config->rx_filter = HWTSTAMP_FILTER_ALL; break; default: return -ERANGE; } - fp->hwtstamp_cfg = cfg; - return copy_to_user(ifr->ifr_data, &cfg, sizeof(cfg)) ? -EFAULT : 0; -} - -static int fun_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) -{ - switch (cmd) { - case SIOCSHWTSTAMP: - return fun_hwtstamp_set(dev, ifr); - case SIOCGHWTSTAMP: - return fun_hwtstamp_get(dev, ifr); - default: - return -EOPNOTSUPP; - } + fp->hwtstamp_cfg = *config; + return 0; } /* Prepare the queues for XDP. */ @@ -1340,7 +1327,6 @@ static const struct net_device_ops fun_netdev_ops = { .ndo_change_mtu = fun_change_mtu, .ndo_set_mac_address = fun_set_macaddr, .ndo_validate_addr = eth_validate_addr, - .ndo_eth_ioctl = fun_ioctl, .ndo_uninit = fun_uninit, .ndo_bpf = fun_xdp, .ndo_xdp_xmit = fun_xdp_xmit_frames, @@ -1348,6 +1334,8 @@ static const struct net_device_ops fun_netdev_ops = { .ndo_set_vf_vlan = fun_set_vf_vlan, .ndo_set_vf_rate = fun_set_vf_rate, .ndo_get_vf_config = fun_get_vf_config, + .ndo_hwtstamp_get = fun_hwtstamp_get, + .ndo_hwtstamp_set = fun_hwtstamp_set, }; #define GSO_ENCAP_FLAGS (NETIF_F_GSO_GRE | NETIF_F_GSO_IPXIP4 | \ From a5cd3a60aa1d0366265638c43012e1ac1b3f6d6a Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 16 Oct 2025 18:29:11 +0000 Subject: [PATCH 0144/1381] net: shrink napi_skb_cache_{put,get}() and napi_skb_cache_get_bulk() Following loop in napi_skb_cache_put() is unrolled by the compiler even if CONFIG_KASAN is not enabled: for (i = NAPI_SKB_CACHE_HALF; i < NAPI_SKB_CACHE_SIZE; i++) kasan_mempool_unpoison_object(nc->skb_cache[i], kmem_cache_size(net_hotdata.skbuff_cache)); We have 32 times this sequence, for a total of 384 bytes. 48 8b 3d 00 00 00 00 net_hotdata.skbuff_cache,%rdi e8 00 00 00 00 call kmem_cache_size This is because kmem_cache_size() is not an inline and not const, and kasan_unpoison_object_data() is an inline function. Cache kmem_cache_size() result in a variable, so that the compiler can remove dead code (and variable) when/if CONFIG_KASAN is unset. After this patch, napi_skb_cache_put() is inlined in its callers, and we avoid one kmem_cache_size() call in napi_skb_cache_get() and napi_skb_cache_get_bulk(). Signed-off-by: Eric Dumazet Reviewed-by: Jacob Keller Reviewed-by: Kuniyuki Iwashima Reviewed-by: Alexander Lobakin Link: https://patch.msgid.link/20251016182911.1132792-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/core/skbuff.c | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-) diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 8eb3c5820724..5b4bc8b1c7d5 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -274,6 +274,11 @@ void *__netdev_alloc_frag_align(unsigned int fragsz, unsigned int align_mask) } EXPORT_SYMBOL(__netdev_alloc_frag_align); +/* Cache kmem_cache_size(net_hotdata.skbuff_cache) to help the compiler + * remove dead code (and skbuff_cache_size) when CONFIG_KASAN is unset. + */ +static u32 skbuff_cache_size __read_mostly; + static struct sk_buff *napi_skb_cache_get(void) { struct napi_alloc_cache *nc = this_cpu_ptr(&napi_alloc_cache); @@ -293,7 +298,7 @@ static struct sk_buff *napi_skb_cache_get(void) skb = nc->skb_cache[--nc->skb_count]; local_unlock_nested_bh(&napi_alloc_cache.bh_lock); - kasan_mempool_unpoison_object(skb, kmem_cache_size(net_hotdata.skbuff_cache)); + kasan_mempool_unpoison_object(skb, skbuff_cache_size); return skb; } @@ -345,11 +350,9 @@ u32 napi_skb_cache_get_bulk(void **skbs, u32 n) get: for (u32 base = nc->skb_count - n, i = 0; i < n; i++) { - u32 cache_size = kmem_cache_size(net_hotdata.skbuff_cache); - skbs[i] = nc->skb_cache[base + i]; - kasan_mempool_unpoison_object(skbs[i], cache_size); + kasan_mempool_unpoison_object(skbs[i], skbuff_cache_size); memset(skbs[i], 0, offsetof(struct sk_buff, tail)); } @@ -1437,7 +1440,7 @@ static void napi_skb_cache_put(struct sk_buff *skb) if (unlikely(nc->skb_count == NAPI_SKB_CACHE_SIZE)) { for (i = NAPI_SKB_CACHE_HALF; i < NAPI_SKB_CACHE_SIZE; i++) kasan_mempool_unpoison_object(nc->skb_cache[i], - kmem_cache_size(net_hotdata.skbuff_cache)); + skbuff_cache_size); kmem_cache_free_bulk(net_hotdata.skbuff_cache, NAPI_SKB_CACHE_HALF, nc->skb_cache + NAPI_SKB_CACHE_HALF); @@ -5125,6 +5128,8 @@ void __init skb_init(void) offsetof(struct sk_buff, cb), sizeof_field(struct sk_buff, cb), NULL); + skbuff_cache_size = kmem_cache_size(net_hotdata.skbuff_cache); + net_hotdata.skbuff_fclone_cache = kmem_cache_create("skbuff_fclone_cache", sizeof(struct sk_buff_fclones), 0, From c30fd916c4d7760ccf654e52370b0a31be885789 Mon Sep 17 00:00:00 2001 From: Ankit Garg Date: Thu, 16 Oct 2025 18:25:42 -0700 Subject: [PATCH 0145/1381] gve: Consolidate and persist ethtool ring changes Refactor the ethtool ring parameter configuration logic to address two issues: unnecessary queue resets and lost configuration changes when the interface is down. Previously, `gve_set_ringparam` could trigger multiple queue destructions and recreations for a single command, as different settings (e.g., header split, ring sizes) were applied one by one. Furthermore, if the interface was down, any changes made via ethtool were discarded instead of being saved for the next time the interface was brought up. This patch centralizes the configuration logic. Individual functions like `gve_set_hsplit_config` are modified to only validate and stage changes in a temporary config struct. The main `gve_set_ringparam` function now gathers all staged changes and applies them as a single, combined configuration: 1. If the interface is up, it calls `gve_adjust_config` once. 2. If the interface is down, it saves the settings directly to the driver's private struct, ensuring they persist and are used when the interface is brought back up. Signed-off-by: Ankit Garg Reviewed-by: Harshitha Ramamurthy Reviewed-by: Jordan Rhee Reviewed-by: Willem de Bruijn Signed-off-by: Joshua Washington Link: https://patch.msgid.link/20251017012614.3631351-1-joshwash@google.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/google/gve/gve.h | 3 +- drivers/net/ethernet/google/gve/gve_ethtool.c | 92 +++++++++---------- drivers/net/ethernet/google/gve/gve_main.c | 17 ++-- 3 files changed, 54 insertions(+), 58 deletions(-) diff --git a/drivers/net/ethernet/google/gve/gve.h b/drivers/net/ethernet/google/gve/gve.h index 4cc6dcbfd367..cf95ec25b11a 100644 --- a/drivers/net/ethernet/google/gve/gve.h +++ b/drivers/net/ethernet/google/gve/gve.h @@ -1251,7 +1251,8 @@ void gve_rx_start_ring_gqi(struct gve_priv *priv, int idx); void gve_rx_stop_ring_gqi(struct gve_priv *priv, int idx); u16 gve_get_pkt_buf_size(const struct gve_priv *priv, bool enable_hplit); bool gve_header_split_supported(const struct gve_priv *priv); -int gve_set_hsplit_config(struct gve_priv *priv, u8 tcp_data_split); +int gve_set_hsplit_config(struct gve_priv *priv, u8 tcp_data_split, + struct gve_rx_alloc_rings_cfg *rx_alloc_cfg); /* rx buffer handling */ int gve_buf_ref_cnt(struct gve_rx_buf_state_dqo *bs); void gve_free_page_dqo(struct gve_priv *priv, struct gve_rx_buf_state_dqo *bs, diff --git a/drivers/net/ethernet/google/gve/gve_ethtool.c b/drivers/net/ethernet/google/gve/gve_ethtool.c index d0a223250845..b030a84b678c 100644 --- a/drivers/net/ethernet/google/gve/gve_ethtool.c +++ b/drivers/net/ethernet/google/gve/gve_ethtool.c @@ -537,34 +537,6 @@ static void gve_get_ringparam(struct net_device *netdev, kernel_cmd->tcp_data_split = ETHTOOL_TCP_DATA_SPLIT_DISABLED; } -static int gve_adjust_ring_sizes(struct gve_priv *priv, - u16 new_tx_desc_cnt, - u16 new_rx_desc_cnt) -{ - struct gve_tx_alloc_rings_cfg tx_alloc_cfg = {0}; - struct gve_rx_alloc_rings_cfg rx_alloc_cfg = {0}; - int err; - - /* get current queue configuration */ - gve_get_curr_alloc_cfgs(priv, &tx_alloc_cfg, &rx_alloc_cfg); - - /* copy over the new ring_size from ethtool */ - tx_alloc_cfg.ring_size = new_tx_desc_cnt; - rx_alloc_cfg.ring_size = new_rx_desc_cnt; - - if (netif_running(priv->dev)) { - err = gve_adjust_config(priv, &tx_alloc_cfg, &rx_alloc_cfg); - if (err) - return err; - } - - /* Set new ring_size for the next up */ - priv->tx_desc_cnt = new_tx_desc_cnt; - priv->rx_desc_cnt = new_rx_desc_cnt; - - return 0; -} - static int gve_validate_req_ring_size(struct gve_priv *priv, u16 new_tx_desc_cnt, u16 new_rx_desc_cnt) { @@ -584,20 +556,13 @@ static int gve_validate_req_ring_size(struct gve_priv *priv, u16 new_tx_desc_cnt return 0; } -static int gve_set_ringparam(struct net_device *netdev, - struct ethtool_ringparam *cmd, - struct kernel_ethtool_ringparam *kernel_cmd, - struct netlink_ext_ack *extack) +static int gve_set_ring_sizes_config(struct gve_priv *priv, u16 new_tx_desc_cnt, + u16 new_rx_desc_cnt, + struct gve_tx_alloc_rings_cfg *tx_alloc_cfg, + struct gve_rx_alloc_rings_cfg *rx_alloc_cfg) { - struct gve_priv *priv = netdev_priv(netdev); - u16 new_tx_cnt, new_rx_cnt; - int err; - - err = gve_set_hsplit_config(priv, kernel_cmd->tcp_data_split); - if (err) - return err; - - if (cmd->tx_pending == priv->tx_desc_cnt && cmd->rx_pending == priv->rx_desc_cnt) + if (new_tx_desc_cnt == priv->tx_desc_cnt && + new_rx_desc_cnt == priv->rx_desc_cnt) return 0; if (!priv->modify_ring_size_enabled) { @@ -605,13 +570,48 @@ static int gve_set_ringparam(struct net_device *netdev, return -EOPNOTSUPP; } - new_tx_cnt = cmd->tx_pending; - new_rx_cnt = cmd->rx_pending; - - if (gve_validate_req_ring_size(priv, new_tx_cnt, new_rx_cnt)) + if (gve_validate_req_ring_size(priv, new_tx_desc_cnt, new_rx_desc_cnt)) return -EINVAL; - return gve_adjust_ring_sizes(priv, new_tx_cnt, new_rx_cnt); + tx_alloc_cfg->ring_size = new_tx_desc_cnt; + rx_alloc_cfg->ring_size = new_rx_desc_cnt; + return 0; +} + +static int gve_set_ringparam(struct net_device *netdev, + struct ethtool_ringparam *cmd, + struct kernel_ethtool_ringparam *kernel_cmd, + struct netlink_ext_ack *extack) +{ + struct gve_tx_alloc_rings_cfg tx_alloc_cfg = {0}; + struct gve_rx_alloc_rings_cfg rx_alloc_cfg = {0}; + struct gve_priv *priv = netdev_priv(netdev); + int err; + + gve_get_curr_alloc_cfgs(priv, &tx_alloc_cfg, &rx_alloc_cfg); + err = gve_set_hsplit_config(priv, kernel_cmd->tcp_data_split, + &rx_alloc_cfg); + if (err) + return err; + + err = gve_set_ring_sizes_config(priv, cmd->tx_pending, cmd->rx_pending, + &tx_alloc_cfg, &rx_alloc_cfg); + if (err) + return err; + + if (netif_running(priv->dev)) { + err = gve_adjust_config(priv, &tx_alloc_cfg, &rx_alloc_cfg); + if (err) + return err; + } else { + /* Set ring params for the next up */ + priv->header_split_enabled = rx_alloc_cfg.enable_header_split; + priv->rx_cfg.packet_buffer_size = + rx_alloc_cfg.packet_buffer_size; + priv->tx_desc_cnt = tx_alloc_cfg.ring_size; + priv->rx_desc_cnt = rx_alloc_cfg.ring_size; + } + return 0; } static int gve_user_reset(struct net_device *netdev, u32 *flags) diff --git a/drivers/net/ethernet/google/gve/gve_main.c b/drivers/net/ethernet/google/gve/gve_main.c index 1be1b1ef31ee..29845e8f3c0d 100644 --- a/drivers/net/ethernet/google/gve/gve_main.c +++ b/drivers/net/ethernet/google/gve/gve_main.c @@ -2058,12 +2058,10 @@ bool gve_header_split_supported(const struct gve_priv *priv) priv->queue_format == GVE_DQO_RDA_FORMAT && !priv->xdp_prog; } -int gve_set_hsplit_config(struct gve_priv *priv, u8 tcp_data_split) +int gve_set_hsplit_config(struct gve_priv *priv, u8 tcp_data_split, + struct gve_rx_alloc_rings_cfg *rx_alloc_cfg) { - struct gve_tx_alloc_rings_cfg tx_alloc_cfg = {0}; - struct gve_rx_alloc_rings_cfg rx_alloc_cfg = {0}; bool enable_hdr_split; - int err = 0; if (tcp_data_split == ETHTOOL_TCP_DATA_SPLIT_UNKNOWN) return 0; @@ -2081,14 +2079,11 @@ int gve_set_hsplit_config(struct gve_priv *priv, u8 tcp_data_split) if (enable_hdr_split == priv->header_split_enabled) return 0; - gve_get_curr_alloc_cfgs(priv, &tx_alloc_cfg, &rx_alloc_cfg); + rx_alloc_cfg->enable_header_split = enable_hdr_split; + rx_alloc_cfg->packet_buffer_size = + gve_get_pkt_buf_size(priv, enable_hdr_split); - rx_alloc_cfg.enable_header_split = enable_hdr_split; - rx_alloc_cfg.packet_buffer_size = gve_get_pkt_buf_size(priv, enable_hdr_split); - - if (netif_running(priv->dev)) - err = gve_adjust_config(priv, &tx_alloc_cfg, &rx_alloc_cfg); - return err; + return 0; } static int gve_set_features(struct net_device *netdev, From 98c2f0b42eead51282fa722f4aedccd2987e0374 Mon Sep 17 00:00:00 2001 From: Jesse Brandeburg Date: Thu, 16 Oct 2025 23:08:37 -0700 Subject: [PATCH 0146/1381] net: docs: add missing features that can have stats While trying to figure out ethtool -I | --include-statistics, I noticed some docs got missed when implementing commit 0e9c127729be ("ethtool: add interface to read Tx hardware timestamping statistics"). Fix up the docs to match the kernel code, and while there, sort them in alphabetical order. Reviewed-by: Jacob Keller Reviewed-by: Rahul Rameshbabu Reviewed-by: Jakub Kicinski Reviewed-by: Simon Horman Signed-off-by: Jesse Brandeburg Reviewed-by: Paul Menzel Reviewed-by: Aleksandr Loktionov Signed-off-by: Jacob Keller Link: https://patch.msgid.link/20251016-jk-iwl-next-2025-10-15-v2-8-ff3a390d9fc6@intel.com Signed-off-by: Jakub Kicinski --- Documentation/networking/statistics.rst | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/Documentation/networking/statistics.rst b/Documentation/networking/statistics.rst index 518284e287b0..66b0ef941457 100644 --- a/Documentation/networking/statistics.rst +++ b/Documentation/networking/statistics.rst @@ -184,9 +184,11 @@ Protocol-related statistics can be requested in get commands by setting the `ETHTOOL_FLAG_STATS` flag in `ETHTOOL_A_HEADER_FLAGS`. Currently statistics are supported in the following commands: - - `ETHTOOL_MSG_PAUSE_GET` - `ETHTOOL_MSG_FEC_GET` + - `ETHTOOL_MSG_LINKSTATE_GET` - `ETHTOOL_MSG_MM_GET` + - `ETHTOOL_MSG_PAUSE_GET` + - `ETHTOOL_MSG_TSINFO_GET` debugfs ------- From 20ae87514ad5159f821951761fc7362bceeca265 Mon Sep 17 00:00:00 2001 From: Jesse Brandeburg Date: Thu, 16 Oct 2025 23:08:38 -0700 Subject: [PATCH 0147/1381] ice: implement ethtool standard stats Add support for MAC/pause/RMON stats. This enables reporting hardware statistics in a common way via: ethtool -S eth0 --all-groups and ethtool --include-statistics --show-pause eth0 While doing so, add support for one new stat, receive length error (RLEC), which is extremely unlikely to happen since most L2 frames have a type/length field specifying a "type", and raw ethernet frames aren't used much any longer. NOTE: I didn't implement Ctrl aka control frame stats because the hardware doesn't seem to implement support. Reviewed-by: Marcin Szycik Reviewed-by: Jacob Keller Reviewed-by: Jakub Kicinski Reviewed-by: Simon Horman Signed-off-by: Jesse Brandeburg Tested-by: Rinitha S Reviewed-by: Aleksandr Loktionov Signed-off-by: Jacob Keller Link: https://patch.msgid.link/20251016-jk-iwl-next-2025-10-15-v2-9-ff3a390d9fc6@intel.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/intel/ice/ice_ethtool.c | 78 ++++++++++++++++++++ drivers/net/ethernet/intel/ice/ice_main.c | 3 + drivers/net/ethernet/intel/ice/ice_type.h | 1 + 3 files changed, 82 insertions(+) diff --git a/drivers/net/ethernet/intel/ice/ice_ethtool.c b/drivers/net/ethernet/intel/ice/ice_ethtool.c index dc131779d426..d1ec7e6f12bf 100644 --- a/drivers/net/ethernet/intel/ice/ice_ethtool.c +++ b/drivers/net/ethernet/intel/ice/ice_ethtool.c @@ -4661,6 +4661,81 @@ static void ice_get_fec_stats(struct net_device *netdev, pi->lport, err); } +static void ice_get_eth_mac_stats(struct net_device *netdev, + struct ethtool_eth_mac_stats *mac_stats) +{ + struct ice_pf *pf = ice_netdev_to_pf(netdev); + struct ice_hw_port_stats *ps = &pf->stats; + + mac_stats->FramesTransmittedOK = ps->eth.tx_unicast + + ps->eth.tx_multicast + + ps->eth.tx_broadcast; + mac_stats->FramesReceivedOK = ps->eth.rx_unicast + + ps->eth.rx_multicast + + ps->eth.rx_broadcast; + mac_stats->FrameCheckSequenceErrors = ps->crc_errors; + mac_stats->OctetsTransmittedOK = ps->eth.tx_bytes; + mac_stats->OctetsReceivedOK = ps->eth.rx_bytes; + mac_stats->MulticastFramesXmittedOK = ps->eth.tx_multicast; + mac_stats->BroadcastFramesXmittedOK = ps->eth.tx_broadcast; + mac_stats->MulticastFramesReceivedOK = ps->eth.rx_multicast; + mac_stats->BroadcastFramesReceivedOK = ps->eth.rx_broadcast; + mac_stats->InRangeLengthErrors = ps->rx_len_errors; + mac_stats->FrameTooLongErrors = ps->rx_oversize; +} + +static void ice_get_pause_stats(struct net_device *netdev, + struct ethtool_pause_stats *pause_stats) +{ + struct ice_pf *pf = ice_netdev_to_pf(netdev); + struct ice_hw_port_stats *ps = &pf->stats; + + pause_stats->tx_pause_frames = ps->link_xon_tx + ps->link_xoff_tx; + pause_stats->rx_pause_frames = ps->link_xon_rx + ps->link_xoff_rx; +} + +static const struct ethtool_rmon_hist_range ice_rmon_ranges[] = { + { 0, 64 }, + { 65, 127 }, + { 128, 255 }, + { 256, 511 }, + { 512, 1023 }, + { 1024, 1522 }, + { 1523, 9522 }, + {} +}; + +static void ice_get_rmon_stats(struct net_device *netdev, + struct ethtool_rmon_stats *rmon, + const struct ethtool_rmon_hist_range **ranges) +{ + struct ice_pf *pf = ice_netdev_to_pf(netdev); + struct ice_hw_port_stats *ps = &pf->stats; + + rmon->undersize_pkts = ps->rx_undersize; + rmon->oversize_pkts = ps->rx_oversize; + rmon->fragments = ps->rx_fragments; + rmon->jabbers = ps->rx_jabber; + + rmon->hist[0] = ps->rx_size_64; + rmon->hist[1] = ps->rx_size_127; + rmon->hist[2] = ps->rx_size_255; + rmon->hist[3] = ps->rx_size_511; + rmon->hist[4] = ps->rx_size_1023; + rmon->hist[5] = ps->rx_size_1522; + rmon->hist[6] = ps->rx_size_big; + + rmon->hist_tx[0] = ps->tx_size_64; + rmon->hist_tx[1] = ps->tx_size_127; + rmon->hist_tx[2] = ps->tx_size_255; + rmon->hist_tx[3] = ps->tx_size_511; + rmon->hist_tx[4] = ps->tx_size_1023; + rmon->hist_tx[5] = ps->tx_size_1522; + rmon->hist_tx[6] = ps->tx_size_big; + + *ranges = ice_rmon_ranges; +} + #define ICE_ETHTOOL_PFR (ETH_RESET_IRQ | ETH_RESET_DMA | \ ETH_RESET_FILTER | ETH_RESET_OFFLOAD) @@ -4744,6 +4819,9 @@ static const struct ethtool_ops ice_ethtool_ops = { .get_link_ksettings = ice_get_link_ksettings, .set_link_ksettings = ice_set_link_ksettings, .get_fec_stats = ice_get_fec_stats, + .get_eth_mac_stats = ice_get_eth_mac_stats, + .get_pause_stats = ice_get_pause_stats, + .get_rmon_stats = ice_get_rmon_stats, .get_drvinfo = ice_get_drvinfo, .get_regs_len = ice_get_regs_len, .get_regs = ice_get_regs, diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c index 86f5859e88ef..3d5615caf6d1 100644 --- a/drivers/net/ethernet/intel/ice/ice_main.c +++ b/drivers/net/ethernet/intel/ice/ice_main.c @@ -7138,6 +7138,9 @@ void ice_update_pf_stats(struct ice_pf *pf) &prev_ps->mac_remote_faults, &cur_ps->mac_remote_faults); + ice_stat_update32(hw, GLPRT_RLEC(port), pf->stat_prev_loaded, + &prev_ps->rx_len_errors, &cur_ps->rx_len_errors); + ice_stat_update32(hw, GLPRT_RUC(port), pf->stat_prev_loaded, &prev_ps->rx_undersize, &cur_ps->rx_undersize); diff --git a/drivers/net/ethernet/intel/ice/ice_type.h b/drivers/net/ethernet/intel/ice/ice_type.h index b0a1b67071c5..6a2ec8389a8f 100644 --- a/drivers/net/ethernet/intel/ice/ice_type.h +++ b/drivers/net/ethernet/intel/ice/ice_type.h @@ -1063,6 +1063,7 @@ struct ice_hw_port_stats { u64 error_bytes; /* errbc */ u64 mac_local_faults; /* mlfc */ u64 mac_remote_faults; /* mrfc */ + u64 rx_len_errors; /* rlec */ u64 link_xon_rx; /* lxonrxc */ u64 link_xoff_rx; /* lxoffrxc */ u64 link_xon_tx; /* lxontxc */ From 4368d5fe02f6d2cf36bb3b261a7ca25a2dd9a3ca Mon Sep 17 00:00:00 2001 From: Jesse Brandeburg Date: Thu, 16 Oct 2025 23:08:39 -0700 Subject: [PATCH 0148/1381] ice: add tracking of good transmit timestamps As a pre-requisite to implementing timestamp statistics, start tracking successful PTP timestamps. There already existed a trace event, but add a counter as well so it can be displayed by the next patch. Good count is a u64 as it is much more likely to be incremented. The existing error stats are all u32 as before, and are less likely so will wrap less. Reviewed-by: Jacob Keller Reviewed-by: Jakub Kicinski Reviewed-by: Simon Horman Signed-off-by: Jesse Brandeburg Tested-by: Rinitha S Reviewed-by: Aleksandr Loktionov Signed-off-by: Jacob Keller Link: https://patch.msgid.link/20251016-jk-iwl-next-2025-10-15-v2-10-ff3a390d9fc6@intel.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/intel/ice/ice_ptp.c | 9 +++++++++ drivers/net/ethernet/intel/ice/ice_ptp.h | 2 ++ 2 files changed, 11 insertions(+) diff --git a/drivers/net/ethernet/intel/ice/ice_ptp.c b/drivers/net/ethernet/intel/ice/ice_ptp.c index fb0f6365a6d6..4f50e952bfb5 100644 --- a/drivers/net/ethernet/intel/ice/ice_ptp.c +++ b/drivers/net/ethernet/intel/ice/ice_ptp.c @@ -500,6 +500,9 @@ void ice_ptp_complete_tx_single_tstamp(struct ice_ptp_tx *tx) if (tstamp) { shhwtstamps.hwtstamp = ns_to_ktime(tstamp); ice_trace(tx_tstamp_complete, skb, idx); + + /* Count the number of Tx timestamps that succeeded */ + pf->ptp.tx_hwtstamp_good++; } skb_tstamp_tx(skb, &shhwtstamps); @@ -558,6 +561,7 @@ static void ice_ptp_process_tx_tstamp(struct ice_ptp_tx *tx) { struct ice_ptp_port *ptp_port; unsigned long flags; + u32 tstamp_good = 0; struct ice_pf *pf; struct ice_hw *hw; u64 tstamp_ready; @@ -658,11 +662,16 @@ static void ice_ptp_process_tx_tstamp(struct ice_ptp_tx *tx) if (tstamp) { shhwtstamps.hwtstamp = ns_to_ktime(tstamp); ice_trace(tx_tstamp_complete, skb, idx); + + /* Count the number of Tx timestamps that succeeded */ + tstamp_good++; } skb_tstamp_tx(skb, &shhwtstamps); dev_kfree_skb_any(skb); } + + pf->ptp.tx_hwtstamp_good += tstamp_good; } /** diff --git a/drivers/net/ethernet/intel/ice/ice_ptp.h b/drivers/net/ethernet/intel/ice/ice_ptp.h index 137f2070a2d9..27016aac4f1e 100644 --- a/drivers/net/ethernet/intel/ice/ice_ptp.h +++ b/drivers/net/ethernet/intel/ice/ice_ptp.h @@ -237,6 +237,7 @@ struct ice_ptp_pin_desc { * @clock: pointer to registered PTP clock device * @tstamp_config: hardware timestamping configuration * @reset_time: kernel time after clock stop on reset + * @tx_hwtstamp_good: number of completed Tx timestamp requests * @tx_hwtstamp_skipped: number of Tx time stamp requests skipped * @tx_hwtstamp_timeouts: number of Tx skbs discarded with no time stamp * @tx_hwtstamp_flushed: number of Tx skbs flushed due to interface closed @@ -261,6 +262,7 @@ struct ice_ptp { struct ptp_clock *clock; struct kernel_hwtstamp_config tstamp_config; u64 reset_time; + u64 tx_hwtstamp_good; u32 tx_hwtstamp_skipped; u32 tx_hwtstamp_timeouts; u32 tx_hwtstamp_flushed; From 71462475d0020b4001066d5b591ab09ea143002d Mon Sep 17 00:00:00 2001 From: Jesse Brandeburg Date: Thu, 16 Oct 2025 23:08:40 -0700 Subject: [PATCH 0149/1381] ice: implement transmit hardware timestamp statistics The kernel now has common statistics for transmit timestamps, so implement them in the ice driver. use via ethtool -I -T eth0 Reviewed-by: Jacob Keller Reviewed-by: Jakub Kicinski Reviewed-by: Hariprasad Kelam Reviewed-by: Simon Horman Signed-off-by: Jesse Brandeburg Tested-by: Rinitha S Reviewed-by: Aleksandr Loktionov Signed-off-by: Jacob Keller Link: https://patch.msgid.link/20251016-jk-iwl-next-2025-10-15-v2-11-ff3a390d9fc6@intel.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/intel/ice/ice_ethtool.c | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+) diff --git a/drivers/net/ethernet/intel/ice/ice_ethtool.c b/drivers/net/ethernet/intel/ice/ice_ethtool.c index d1ec7e6f12bf..75492a720c68 100644 --- a/drivers/net/ethernet/intel/ice/ice_ethtool.c +++ b/drivers/net/ethernet/intel/ice/ice_ethtool.c @@ -4736,6 +4736,23 @@ static void ice_get_rmon_stats(struct net_device *netdev, *ranges = ice_rmon_ranges; } +/* ice_get_ts_stats - provide timestamping stats + * @netdev: the netdevice pointer from ethtool + * @ts_stats: the ethtool data structure to fill in + */ +static void ice_get_ts_stats(struct net_device *netdev, + struct ethtool_ts_stats *ts_stats) +{ + struct ice_pf *pf = ice_netdev_to_pf(netdev); + struct ice_ptp *ptp = &pf->ptp; + + ts_stats->pkts = ptp->tx_hwtstamp_good; + ts_stats->err = ptp->tx_hwtstamp_skipped + + ptp->tx_hwtstamp_flushed + + ptp->tx_hwtstamp_discarded; + ts_stats->lost = ptp->tx_hwtstamp_timeouts; +} + #define ICE_ETHTOOL_PFR (ETH_RESET_IRQ | ETH_RESET_DMA | \ ETH_RESET_FILTER | ETH_RESET_OFFLOAD) @@ -4822,6 +4839,7 @@ static const struct ethtool_ops ice_ethtool_ops = { .get_eth_mac_stats = ice_get_eth_mac_stats, .get_pause_stats = ice_get_pause_stats, .get_rmon_stats = ice_get_rmon_stats, + .get_ts_stats = ice_get_ts_stats, .get_drvinfo = ice_get_drvinfo, .get_regs_len = ice_get_regs_len, .get_regs = ice_get_regs, From a308ea9721122d632d1ce2af4cec1c008aff512a Mon Sep 17 00:00:00 2001 From: Jesse Brandeburg Date: Thu, 16 Oct 2025 23:08:41 -0700 Subject: [PATCH 0150/1381] ice: refactor to use helpers Use the ice_netdev_to_pf() helper in more places and remove a bunch of boilerplate code. Not every instance could be replaced due to use of the netdev_priv() output or the vsi variable within a bunch of functions. Signed-off-by: Jesse Brandeburg Reviewed-by: Simon Horman Reviewed-by: Przemek Kitszel Reviewed-by: Aleksandr Loktionov Tested-by: Rinitha S Signed-off-by: Jacob Keller Link: https://patch.msgid.link/20251016-jk-iwl-next-2025-10-15-v2-12-ff3a390d9fc6@intel.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/intel/ice/ice_ethtool.c | 48 ++++++------------- .../net/ethernet/intel/ice/ice_flex_pipe.c | 8 +--- drivers/net/ethernet/intel/ice/ice_lag.c | 3 +- drivers/net/ethernet/intel/ice/ice_main.c | 10 ++-- drivers/net/ethernet/intel/ice/ice_ptp.c | 6 +-- drivers/net/ethernet/intel/ice/ice_sriov.c | 3 +- 6 files changed, 24 insertions(+), 54 deletions(-) diff --git a/drivers/net/ethernet/intel/ice/ice_ethtool.c b/drivers/net/ethernet/intel/ice/ice_ethtool.c index 75492a720c68..cb34d4675a78 100644 --- a/drivers/net/ethernet/intel/ice/ice_ethtool.c +++ b/drivers/net/ethernet/intel/ice/ice_ethtool.c @@ -794,8 +794,7 @@ static int ice_get_extended_regs(struct net_device *netdev, void *p) static void ice_get_regs(struct net_device *netdev, struct ethtool_regs *regs, void *p) { - struct ice_netdev_priv *np = netdev_priv(netdev); - struct ice_pf *pf = np->vsi->back; + struct ice_pf *pf = ice_netdev_to_pf(netdev); struct ice_hw *hw = &pf->hw; u32 *regs_buf = (u32 *)p; unsigned int i; @@ -810,8 +809,7 @@ ice_get_regs(struct net_device *netdev, struct ethtool_regs *regs, void *p) static u32 ice_get_msglevel(struct net_device *netdev) { - struct ice_netdev_priv *np = netdev_priv(netdev); - struct ice_pf *pf = np->vsi->back; + struct ice_pf *pf = ice_netdev_to_pf(netdev); #ifndef CONFIG_DYNAMIC_DEBUG if (pf->hw.debug_mask) @@ -824,8 +822,7 @@ static u32 ice_get_msglevel(struct net_device *netdev) static void ice_set_msglevel(struct net_device *netdev, u32 data) { - struct ice_netdev_priv *np = netdev_priv(netdev); - struct ice_pf *pf = np->vsi->back; + struct ice_pf *pf = ice_netdev_to_pf(netdev); #ifndef CONFIG_DYNAMIC_DEBUG if (ICE_DBG_USER & data) @@ -840,16 +837,14 @@ static void ice_set_msglevel(struct net_device *netdev, u32 data) static void ice_get_link_ext_stats(struct net_device *netdev, struct ethtool_link_ext_stats *stats) { - struct ice_netdev_priv *np = netdev_priv(netdev); - struct ice_pf *pf = np->vsi->back; + struct ice_pf *pf = ice_netdev_to_pf(netdev); stats->link_down_events = pf->link_down_events; } static int ice_get_eeprom_len(struct net_device *netdev) { - struct ice_netdev_priv *np = netdev_priv(netdev); - struct ice_pf *pf = np->vsi->back; + struct ice_pf *pf = ice_netdev_to_pf(netdev); return (int)pf->hw.flash.flash_size; } @@ -858,9 +853,7 @@ static int ice_get_eeprom(struct net_device *netdev, struct ethtool_eeprom *eeprom, u8 *bytes) { - struct ice_netdev_priv *np = netdev_priv(netdev); - struct ice_vsi *vsi = np->vsi; - struct ice_pf *pf = vsi->back; + struct ice_pf *pf = ice_netdev_to_pf(netdev); struct ice_hw *hw = &pf->hw; struct device *dev; int ret; @@ -959,8 +952,7 @@ static u64 ice_link_test(struct net_device *netdev) */ static u64 ice_eeprom_test(struct net_device *netdev) { - struct ice_netdev_priv *np = netdev_priv(netdev); - struct ice_pf *pf = np->vsi->back; + struct ice_pf *pf = ice_netdev_to_pf(netdev); netdev_info(netdev, "EEPROM test\n"); return !!(ice_nvm_validate_checksum(&pf->hw)); @@ -1274,9 +1266,8 @@ static int ice_lbtest_receive_frames(struct ice_rx_ring *rx_ring) */ static u64 ice_loopback_test(struct net_device *netdev) { - struct ice_netdev_priv *np = netdev_priv(netdev); - struct ice_vsi *orig_vsi = np->vsi, *test_vsi; - struct ice_pf *pf = orig_vsi->back; + struct ice_pf *pf = ice_netdev_to_pf(netdev); + struct ice_vsi *test_vsi; u8 *tx_frame __free(kfree) = NULL; u8 broadcast[ETH_ALEN], ret = 0; int num_frames, valid_frames; @@ -1365,8 +1356,7 @@ static u64 ice_loopback_test(struct net_device *netdev) */ static u64 ice_intr_test(struct net_device *netdev) { - struct ice_netdev_priv *np = netdev_priv(netdev); - struct ice_pf *pf = np->vsi->back; + struct ice_pf *pf = ice_netdev_to_pf(netdev); u16 swic_old = pf->sw_int_count; netdev_info(netdev, "interrupt test\n"); @@ -1394,9 +1384,8 @@ static void ice_self_test(struct net_device *netdev, struct ethtool_test *eth_test, u64 *data) { - struct ice_netdev_priv *np = netdev_priv(netdev); + struct ice_pf *pf = ice_netdev_to_pf(netdev); bool if_running = netif_running(netdev); - struct ice_pf *pf = np->vsi->back; struct device *dev; dev = ice_pf_to_dev(pf); @@ -1720,9 +1709,7 @@ static int ice_nway_reset(struct net_device *netdev) */ static u32 ice_get_priv_flags(struct net_device *netdev) { - struct ice_netdev_priv *np = netdev_priv(netdev); - struct ice_vsi *vsi = np->vsi; - struct ice_pf *pf = vsi->back; + struct ice_pf *pf = ice_netdev_to_pf(netdev); u32 i, ret_flags = 0; for (i = 0; i < ICE_PRIV_FLAG_ARRAY_SIZE; i++) { @@ -4417,9 +4404,7 @@ static int ice_get_module_info(struct net_device *netdev, struct ethtool_modinfo *modinfo) { - struct ice_netdev_priv *np = netdev_priv(netdev); - struct ice_vsi *vsi = np->vsi; - struct ice_pf *pf = vsi->back; + struct ice_pf *pf = ice_netdev_to_pf(netdev); struct ice_hw *hw = &pf->hw; u8 sff8472_comp = 0; u8 sff8472_swap = 0; @@ -4491,12 +4476,10 @@ static int ice_get_module_eeprom(struct net_device *netdev, struct ethtool_eeprom *ee, u8 *data) { - struct ice_netdev_priv *np = netdev_priv(netdev); + struct ice_pf *pf = ice_netdev_to_pf(netdev); #define SFF_READ_BLOCK_SIZE 8 u8 value[SFF_READ_BLOCK_SIZE] = { 0 }; u8 addr = ICE_I2C_EEPROM_DEV_ADDR; - struct ice_vsi *vsi = np->vsi; - struct ice_pf *pf = vsi->back; struct ice_hw *hw = &pf->hw; bool is_sfp = false; unsigned int i, j; @@ -4774,8 +4757,7 @@ static void ice_get_ts_stats(struct net_device *netdev, */ static int ice_ethtool_reset(struct net_device *dev, u32 *flags) { - struct ice_netdev_priv *np = netdev_priv(dev); - struct ice_pf *pf = np->vsi->back; + struct ice_pf *pf = ice_netdev_to_pf(dev); enum ice_reset_req reset; switch (*flags) { diff --git a/drivers/net/ethernet/intel/ice/ice_flex_pipe.c b/drivers/net/ethernet/intel/ice/ice_flex_pipe.c index 363ae79a3620..06b8786ae3ab 100644 --- a/drivers/net/ethernet/intel/ice/ice_flex_pipe.c +++ b/drivers/net/ethernet/intel/ice/ice_flex_pipe.c @@ -574,9 +574,7 @@ ice_destroy_tunnel(struct ice_hw *hw, u16 index, enum ice_tunnel_type type, int ice_udp_tunnel_set_port(struct net_device *netdev, unsigned int table, unsigned int idx, struct udp_tunnel_info *ti) { - struct ice_netdev_priv *np = netdev_priv(netdev); - struct ice_vsi *vsi = np->vsi; - struct ice_pf *pf = vsi->back; + struct ice_pf *pf = ice_netdev_to_pf(netdev); enum ice_tunnel_type tnl_type; int status; u16 index; @@ -598,9 +596,7 @@ int ice_udp_tunnel_set_port(struct net_device *netdev, unsigned int table, int ice_udp_tunnel_unset_port(struct net_device *netdev, unsigned int table, unsigned int idx, struct udp_tunnel_info *ti) { - struct ice_netdev_priv *np = netdev_priv(netdev); - struct ice_vsi *vsi = np->vsi; - struct ice_pf *pf = vsi->back; + struct ice_pf *pf = ice_netdev_to_pf(netdev); enum ice_tunnel_type tnl_type; int status; diff --git a/drivers/net/ethernet/intel/ice/ice_lag.c b/drivers/net/ethernet/intel/ice/ice_lag.c index aebf8e08a297..d2576d606e10 100644 --- a/drivers/net/ethernet/intel/ice/ice_lag.c +++ b/drivers/net/ethernet/intel/ice/ice_lag.c @@ -2177,8 +2177,7 @@ static void ice_lag_chk_disabled_bond(struct ice_lag *lag, void *ptr) */ static void ice_lag_disable_sriov_bond(struct ice_lag *lag) { - struct ice_netdev_priv *np = netdev_priv(lag->netdev); - struct ice_pf *pf = np->vsi->back; + struct ice_pf *pf = ice_netdev_to_pf(lag->netdev); ice_clear_feature_support(pf, ICE_F_SRIOV_LAG); ice_clear_feature_support(pf, ICE_F_SRIOV_AA_LAG); diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c index 3d5615caf6d1..ca95b8800bb3 100644 --- a/drivers/net/ethernet/intel/ice/ice_main.c +++ b/drivers/net/ethernet/intel/ice/ice_main.c @@ -8074,9 +8074,7 @@ static int ice_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq, struct net_device *dev, u32 filter_mask, int nlflags) { - struct ice_netdev_priv *np = netdev_priv(dev); - struct ice_vsi *vsi = np->vsi; - struct ice_pf *pf = vsi->back; + struct ice_pf *pf = ice_netdev_to_pf(dev); u16 bmode; bmode = pf->first_sw->bridge_mode; @@ -8146,8 +8144,7 @@ ice_bridge_setlink(struct net_device *dev, struct nlmsghdr *nlh, u16 __always_unused flags, struct netlink_ext_ack __always_unused *extack) { - struct ice_netdev_priv *np = netdev_priv(dev); - struct ice_pf *pf = np->vsi->back; + struct ice_pf *pf = ice_netdev_to_pf(dev); struct nlattr *attr, *br_spec; struct ice_hw *hw = &pf->hw; struct ice_sw *pf_sw; @@ -9581,8 +9578,7 @@ ice_indr_setup_tc_cb(struct net_device *netdev, struct Qdisc *sch, */ int ice_open(struct net_device *netdev) { - struct ice_netdev_priv *np = netdev_priv(netdev); - struct ice_pf *pf = np->vsi->back; + struct ice_pf *pf = ice_netdev_to_pf(netdev); if (ice_is_reset_in_progress(pf->state)) { netdev_err(netdev, "can't open net device while reset is in progress"); diff --git a/drivers/net/ethernet/intel/ice/ice_ptp.c b/drivers/net/ethernet/intel/ice/ice_ptp.c index 4f50e952bfb5..985b3e79b312 100644 --- a/drivers/net/ethernet/intel/ice/ice_ptp.c +++ b/drivers/net/ethernet/intel/ice/ice_ptp.c @@ -2215,8 +2215,7 @@ static int ice_ptp_getcrosststamp(struct ptp_clock_info *info, int ice_ptp_hwtstamp_get(struct net_device *netdev, struct kernel_hwtstamp_config *config) { - struct ice_netdev_priv *np = netdev_priv(netdev); - struct ice_pf *pf = np->vsi->back; + struct ice_pf *pf = ice_netdev_to_pf(netdev); if (pf->ptp.state != ICE_PTP_READY) return -EIO; @@ -2287,8 +2286,7 @@ int ice_ptp_hwtstamp_set(struct net_device *netdev, struct kernel_hwtstamp_config *config, struct netlink_ext_ack *extack) { - struct ice_netdev_priv *np = netdev_priv(netdev); - struct ice_pf *pf = np->vsi->back; + struct ice_pf *pf = ice_netdev_to_pf(netdev); int err; if (pf->ptp.state != ICE_PTP_READY) diff --git a/drivers/net/ethernet/intel/ice/ice_sriov.c b/drivers/net/ethernet/intel/ice/ice_sriov.c index 843e82fd3bf9..6b1126ddb561 100644 --- a/drivers/net/ethernet/intel/ice/ice_sriov.c +++ b/drivers/net/ethernet/intel/ice/ice_sriov.c @@ -1190,8 +1190,7 @@ ice_vf_lan_overflow_event(struct ice_pf *pf, struct ice_rq_event_info *event) */ int ice_set_vf_spoofchk(struct net_device *netdev, int vf_id, bool ena) { - struct ice_netdev_priv *np = netdev_priv(netdev); - struct ice_pf *pf = np->vsi->back; + struct ice_pf *pf = ice_netdev_to_pf(netdev); struct ice_vsi *vf_vsi; struct device *dev; struct ice_vf *vf; From 3c7bf5af2196087f394f9099b53e37569636b259 Mon Sep 17 00:00:00 2001 From: Vitaly Lifshits Date: Thu, 16 Oct 2025 23:08:43 -0700 Subject: [PATCH 0151/1381] e1000e: Introduce private flag to disable K1 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The K1 state reduces power consumption on ICH family network controllers during idle periods, similarly to L1 state on PCI Express NICs. Therefore, it is recommended and enabled by default. However, on some systems it has been observed to have adverse side effects, such as packet loss. It has been established through debug that the problem may be due to firmware misconfiguration of specific systems, interoperability with certain link partners, or marginal electrical conditions of specific units. These problems typically cannot be fixed in the field, and generic workarounds to resolve the side effects on all systems, while keeping K1 enabled, were found infeasible. Therefore, add the option for users to globally disable K1 idle state on the adapter. Additionally, disable K1 by default for MTL and later platforms, due to issues reported with the current configuration. Link: https://lore.kernel.org/intel-wired-lan/CAMqyJG3LVqfgqMcTxeaPur_Jq0oQH7GgdxRuVtRX_6TTH2mX5Q@mail.gmail.com/ Link: https://lore.kernel.org/intel-wired-lan/20250626153544.1853d106@onyx.my.domain/ Link: https://lore.kernel.org/intel-wired-lan/Z_z9EjcKtwHCQcZR@mail-itl/ Link: https://github.com/QubesOS/qubes-issues/issues/9896 Link: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/2115393 Signed-off-by: Vitaly Lifshits Reviewed-by: Timo Teräs Tested-by: Timo Teräs Reviewed-by: Aleksandr Loktionov Reviewed-by: Dima Ruinskiy Tested-by: Avraham Koren Signed-off-by: Jacob Keller Link: https://patch.msgid.link/20251016-jk-iwl-next-2025-10-15-v2-14-ff3a390d9fc6@intel.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/intel/e1000e/e1000.h | 1 + drivers/net/ethernet/intel/e1000e/ethtool.c | 45 ++++++++++++++++++--- drivers/net/ethernet/intel/e1000e/ich8lan.c | 41 ++++++++++--------- drivers/net/ethernet/intel/e1000e/netdev.c | 3 ++ 4 files changed, 67 insertions(+), 23 deletions(-) diff --git a/drivers/net/ethernet/intel/e1000e/e1000.h b/drivers/net/ethernet/intel/e1000e/e1000.h index 018e61aea787..aa08f397988e 100644 --- a/drivers/net/ethernet/intel/e1000e/e1000.h +++ b/drivers/net/ethernet/intel/e1000e/e1000.h @@ -461,6 +461,7 @@ s32 e1000e_get_base_timinca(struct e1000_adapter *adapter, u32 *timinca); #define FLAG2_CHECK_RX_HWTSTAMP BIT(13) #define FLAG2_CHECK_SYSTIM_OVERFLOW BIT(14) #define FLAG2_ENABLE_S0IX_FLOWS BIT(15) +#define FLAG2_DISABLE_K1 BIT(16) #define E1000_RX_DESC_PS(R, i) \ (&(((union e1000_rx_desc_packet_split *)((R).desc))[i])) diff --git a/drivers/net/ethernet/intel/e1000e/ethtool.c b/drivers/net/ethernet/intel/e1000e/ethtool.c index 8e40bb50a01e..cee57a2149ab 100644 --- a/drivers/net/ethernet/intel/e1000e/ethtool.c +++ b/drivers/net/ethernet/intel/e1000e/ethtool.c @@ -26,6 +26,8 @@ struct e1000_stats { static const char e1000e_priv_flags_strings[][ETH_GSTRING_LEN] = { #define E1000E_PRIV_FLAGS_S0IX_ENABLED BIT(0) "s0ix-enabled", +#define E1000E_PRIV_FLAGS_DISABLE_K1 BIT(1) + "disable-k1", }; #define E1000E_PRIV_FLAGS_STR_LEN ARRAY_SIZE(e1000e_priv_flags_strings) @@ -2301,26 +2303,59 @@ static u32 e1000e_get_priv_flags(struct net_device *netdev) if (adapter->flags2 & FLAG2_ENABLE_S0IX_FLOWS) priv_flags |= E1000E_PRIV_FLAGS_S0IX_ENABLED; + if (adapter->flags2 & FLAG2_DISABLE_K1) + priv_flags |= E1000E_PRIV_FLAGS_DISABLE_K1; + return priv_flags; } static int e1000e_set_priv_flags(struct net_device *netdev, u32 priv_flags) { struct e1000_adapter *adapter = netdev_priv(netdev); + struct e1000_hw *hw = &adapter->hw; unsigned int flags2 = adapter->flags2; + unsigned int changed; + + flags2 &= ~(FLAG2_ENABLE_S0IX_FLOWS | FLAG2_DISABLE_K1); - flags2 &= ~FLAG2_ENABLE_S0IX_FLOWS; if (priv_flags & E1000E_PRIV_FLAGS_S0IX_ENABLED) { - struct e1000_hw *hw = &adapter->hw; - - if (hw->mac.type < e1000_pch_cnp) + if (hw->mac.type < e1000_pch_cnp) { + e_err("S0ix is not supported on this device\n"); return -EINVAL; + } + flags2 |= FLAG2_ENABLE_S0IX_FLOWS; } - if (flags2 != adapter->flags2) + if (priv_flags & E1000E_PRIV_FLAGS_DISABLE_K1) { + if (hw->mac.type < e1000_ich8lan) { + e_err("Disabling K1 is not supported on this device\n"); + return -EINVAL; + } + + flags2 |= FLAG2_DISABLE_K1; + } + + changed = adapter->flags2 ^ flags2; + if (changed) adapter->flags2 = flags2; + if (changed & FLAG2_DISABLE_K1) { + /* reset the hardware to apply the changes */ + while (test_and_set_bit(__E1000_RESETTING, + &adapter->state)) + usleep_range(1000, 2000); + + if (netif_running(adapter->netdev)) { + e1000e_down(adapter, true); + e1000e_up(adapter); + } else { + e1000e_reset(adapter); + } + + clear_bit(__E1000_RESETTING, &adapter->state); + } + return 0; } diff --git a/drivers/net/ethernet/intel/e1000e/ich8lan.c b/drivers/net/ethernet/intel/e1000e/ich8lan.c index df4e7d781cb1..0ff8688ac3b8 100644 --- a/drivers/net/ethernet/intel/e1000e/ich8lan.c +++ b/drivers/net/ethernet/intel/e1000e/ich8lan.c @@ -286,21 +286,26 @@ static void e1000_toggle_lanphypc_pch_lpt(struct e1000_hw *hw) } /** - * e1000_reconfigure_k1_exit_timeout - reconfigure K1 exit timeout to - * align to MTP and later platform requirements. + * e1000_reconfigure_k1_params - reconfigure Kumeran K1 parameters. * @hw: pointer to the HW structure * + * By default K1 is enabled after MAC reset, so this function only + * disables it. + * * Context: PHY semaphore must be held by caller. * Return: 0 on success, negative on failure */ -static s32 e1000_reconfigure_k1_exit_timeout(struct e1000_hw *hw) +static s32 e1000_reconfigure_k1_params(struct e1000_hw *hw) { u16 phy_timeout; u32 fextnvm12; s32 ret_val; - if (hw->mac.type < e1000_pch_mtp) + if (hw->mac.type < e1000_pch_mtp) { + if (hw->adapter->flags2 & FLAG2_DISABLE_K1) + return e1000_configure_k1_ich8lan(hw, false); return 0; + } /* Change Kumeran K1 power down state from P0s to P1 */ fextnvm12 = er32(FEXTNVM12); @@ -310,6 +315,8 @@ static s32 e1000_reconfigure_k1_exit_timeout(struct e1000_hw *hw) /* Wait for the interface the settle */ usleep_range(1000, 1100); + if (hw->adapter->flags2 & FLAG2_DISABLE_K1) + return e1000_configure_k1_ich8lan(hw, false); /* Change K1 exit timeout */ ret_val = e1e_rphy_locked(hw, I217_PHY_TIMEOUTS_REG, @@ -373,8 +380,8 @@ static s32 e1000_init_phy_workarounds_pchlan(struct e1000_hw *hw) /* At this point the PHY might be inaccessible so don't * propagate the failure */ - if (e1000_reconfigure_k1_exit_timeout(hw)) - e_dbg("Failed to reconfigure K1 exit timeout\n"); + if (e1000_reconfigure_k1_params(hw)) + e_dbg("Failed to reconfigure K1 parameters\n"); fallthrough; case e1000_pch_lpt: @@ -473,10 +480,10 @@ static s32 e1000_init_phy_workarounds_pchlan(struct e1000_hw *hw) if (hw->mac.type >= e1000_pch_mtp) { ret_val = hw->phy.ops.acquire(hw); if (ret_val) { - e_err("Failed to reconfigure K1 exit timeout\n"); + e_err("Failed to reconfigure K1 parameters\n"); goto out; } - ret_val = e1000_reconfigure_k1_exit_timeout(hw); + ret_val = e1000_reconfigure_k1_params(hw); hw->phy.ops.release(hw); } } @@ -4948,17 +4955,15 @@ static s32 e1000_init_hw_ich8lan(struct e1000_hw *hw) u16 i; e1000_initialize_hw_bits_ich8lan(hw); - if (hw->mac.type >= e1000_pch_mtp) { - ret_val = hw->phy.ops.acquire(hw); - if (ret_val) - return ret_val; + ret_val = hw->phy.ops.acquire(hw); + if (ret_val) + return ret_val; - ret_val = e1000_reconfigure_k1_exit_timeout(hw); - hw->phy.ops.release(hw); - if (ret_val) { - e_dbg("Error failed to reconfigure K1 exit timeout\n"); - return ret_val; - } + ret_val = e1000_reconfigure_k1_params(hw); + hw->phy.ops.release(hw); + if (ret_val) { + e_dbg("Error failed to reconfigure K1 parameters\n"); + return ret_val; } /* Initialize identification LED */ diff --git a/drivers/net/ethernet/intel/e1000e/netdev.c b/drivers/net/ethernet/intel/e1000e/netdev.c index 201322dac233..116f3c92b5bc 100644 --- a/drivers/net/ethernet/intel/e1000e/netdev.c +++ b/drivers/net/ethernet/intel/e1000e/netdev.c @@ -7675,6 +7675,9 @@ static int e1000_probe(struct pci_dev *pdev, const struct pci_device_id *ent) /* init PTP hardware clock */ e1000e_ptp_init(adapter); + if (hw->mac.type >= e1000_pch_mtp) + adapter->flags2 |= FLAG2_DISABLE_K1; + /* reset the hardware with the new settings */ e1000e_reset(adapter); From 38516e3fa4ca2c8f06a25716a7286b48824b386b Mon Sep 17 00:00:00 2001 From: Aswin Karuvally Date: Fri, 17 Oct 2025 11:49:54 +0200 Subject: [PATCH 0152/1381] s390/iucv: Convert sprintf/snprintf to scnprintf Convert sprintf/snprintf calls to scnprintf to better align with the kernel development community practices [1]. Link: https://lwn.net/Articles/69419 [1] Reviewed-by: Alexandra Winter Signed-off-by: Aswin Karuvally Signed-off-by: Alexandra Winter Reviewed-by: Simon Horman Link: https://patch.msgid.link/20251017094954.1402684-1-wintera@linux.ibm.com Signed-off-by: Jakub Kicinski --- drivers/s390/net/smsgiucv_app.c | 7 ++++--- net/iucv/af_iucv.c | 7 ++++--- net/iucv/iucv.c | 2 +- 3 files changed, 9 insertions(+), 7 deletions(-) diff --git a/drivers/s390/net/smsgiucv_app.c b/drivers/s390/net/smsgiucv_app.c index 4bd4d6bfc126..768108c90b32 100644 --- a/drivers/s390/net/smsgiucv_app.c +++ b/drivers/s390/net/smsgiucv_app.c @@ -88,9 +88,10 @@ static struct smsg_app_event *smsg_app_event_alloc(const char *from, ev->envp[3] = NULL; /* setting up environment: sender, prefix name, and message text */ - snprintf(ev->envp[0], ENV_SENDER_LEN, ENV_SENDER_STR "%s", from); - snprintf(ev->envp[1], ENV_PREFIX_LEN, ENV_PREFIX_STR "%s", SMSG_PREFIX); - snprintf(ev->envp[2], ENV_TEXT_LEN(msg), ENV_TEXT_STR "%s", msg); + scnprintf(ev->envp[0], ENV_SENDER_LEN, ENV_SENDER_STR "%s", from); + scnprintf(ev->envp[1], ENV_PREFIX_LEN, ENV_PREFIX_STR "%s", + SMSG_PREFIX); + scnprintf(ev->envp[2], ENV_TEXT_LEN(msg), ENV_TEXT_STR "%s", msg); return ev; } diff --git a/net/iucv/af_iucv.c b/net/iucv/af_iucv.c index 6c717a7ef292..4ddfc633d30c 100644 --- a/net/iucv/af_iucv.c +++ b/net/iucv/af_iucv.c @@ -553,10 +553,11 @@ static void __iucv_auto_name(struct iucv_sock *iucv) { char name[12]; - sprintf(name, "%08x", atomic_inc_return(&iucv_sk_list.autobind_name)); + scnprintf(name, sizeof(name), + "%08x", atomic_inc_return(&iucv_sk_list.autobind_name)); while (__iucv_get_sock_by_name(name)) { - sprintf(name, "%08x", - atomic_inc_return(&iucv_sk_list.autobind_name)); + scnprintf(name, sizeof(name), "%08x", + atomic_inc_return(&iucv_sk_list.autobind_name)); } memcpy(iucv->src_name, name, 8); } diff --git a/net/iucv/iucv.c b/net/iucv/iucv.c index 473a7847d80b..008be0abe3a5 100644 --- a/net/iucv/iucv.c +++ b/net/iucv/iucv.c @@ -95,7 +95,7 @@ struct device *iucv_alloc_device(const struct attribute_group **attrs, if (!dev) goto out_error; va_start(vargs, fmt); - vsnprintf(buf, sizeof(buf), fmt, vargs); + vscnprintf(buf, sizeof(buf), fmt, vargs); rc = dev_set_name(dev, "%s", buf); va_end(vargs); if (rc) From 3f9553f65d0b77b724565bbe42c4daa3fab57d5c Mon Sep 17 00:00:00 2001 From: Zenm Chen Date: Mon, 29 Sep 2025 11:57:18 +0800 Subject: [PATCH 0153/1381] wifi: rtl8xxxu: Add USB ID 2001:3328 for D-Link AN3U rev. A1 Add USB ID 2001:3328 for D-Link AN3U rev. A1 which is a RTL8192FU-based Wi-Fi adapter. Compile tested only. Cc: stable@vger.kernel.org # 6.6.x Signed-off-by: Zenm Chen Reviewed-by: Ping-Ke Shih Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20250929035719.6172-1-zenmchen@gmail.com --- drivers/net/wireless/realtek/rtl8xxxu/core.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/net/wireless/realtek/rtl8xxxu/core.c b/drivers/net/wireless/realtek/rtl8xxxu/core.c index 3ded5952729f..be39463bd6c4 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/core.c +++ b/drivers/net/wireless/realtek/rtl8xxxu/core.c @@ -8136,6 +8136,9 @@ static const struct usb_device_id dev_table[] = { /* TP-Link TL-WN823N V2 */ {USB_DEVICE_AND_INTERFACE_INFO(0x2357, 0x0135, 0xff, 0xff, 0xff), .driver_info = (unsigned long)&rtl8192fu_fops}, +/* D-Link AN3U rev. A1 */ +{USB_DEVICE_AND_INTERFACE_INFO(0x2001, 0x3328, 0xff, 0xff, 0xff), + .driver_info = (unsigned long)&rtl8192fu_fops}, #ifdef CONFIG_RTL8XXXU_UNTESTED /* Still supported by rtlwifi */ {USB_DEVICE_AND_INTERFACE_INFO(USB_VENDOR_ID_REALTEK, 0x8176, 0xff, 0xff, 0xff), From b377dcd9a286a6f81922ae442cd1c743bc4a2b35 Mon Sep 17 00:00:00 2001 From: Zenm Chen Date: Mon, 29 Sep 2025 11:57:19 +0800 Subject: [PATCH 0154/1381] wifi: rtw88: Add USB ID 2001:3329 for D-Link AC13U rev. A1 Add USB ID 2001:3329 for D-Link AC13U rev. A1 which is a RTL8812CU-based Wi-Fi adapter. Compile tested only. Cc: stable@vger.kernel.org # 6.6.x Signed-off-by: Zenm Chen Acked-by: Ping-Ke Shih Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20250929035719.6172-2-zenmchen@gmail.com --- drivers/net/wireless/realtek/rtw88/rtw8822cu.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/wireless/realtek/rtw88/rtw8822cu.c b/drivers/net/wireless/realtek/rtw88/rtw8822cu.c index 324fd5c8bfd4..755f76840b12 100644 --- a/drivers/net/wireless/realtek/rtw88/rtw8822cu.c +++ b/drivers/net/wireless/realtek/rtw88/rtw8822cu.c @@ -21,6 +21,8 @@ static const struct usb_device_id rtw_8822cu_id_table[] = { .driver_info = (kernel_ulong_t)&(rtw8822c_hw_spec) }, { USB_DEVICE_AND_INTERFACE_INFO(0x13b1, 0x0043, 0xff, 0xff, 0xff), .driver_info = (kernel_ulong_t)&(rtw8822c_hw_spec) }, /* Alpha - Alpha */ + { USB_DEVICE_AND_INTERFACE_INFO(0x2001, 0x3329, 0xff, 0xff, 0xff), + .driver_info = (kernel_ulong_t)&(rtw8822c_hw_spec) }, /* D-Link AC13U rev. A1 */ {}, }; MODULE_DEVICE_TABLE(usb, rtw_8822cu_id_table); From 8b795fe42927c2df83b5dbc528e610ab662b1024 Mon Sep 17 00:00:00 2001 From: Yu-Chun Lin Date: Tue, 30 Sep 2025 08:05:45 +0800 Subject: [PATCH 0155/1381] wifi: rtw89: Replace hardcoded strings with helper functions Replace hardcoded strings with 'str_on_off()', 'str_enable_disable()', and 'str_read_write()'. The change improves readability. Signed-off-by: Yu-Chun Lin Acked-by: Ping-Ke Shih Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20250930000545.2192692-1-eleanor15x@gmail.com --- drivers/net/wireless/realtek/rtw89/rtw8851b_rfk.c | 8 ++++---- drivers/net/wireless/realtek/rtw89/rtw8852a_rfk.c | 8 ++++---- drivers/net/wireless/realtek/rtw89/rtw8852b_rfk.c | 6 +++--- drivers/net/wireless/realtek/rtw89/rtw8852c_rfk.c | 10 +++++----- drivers/net/wireless/realtek/rtw89/usb.c | 2 +- drivers/net/wireless/realtek/rtw89/wow.c | 2 +- 6 files changed, 18 insertions(+), 18 deletions(-) diff --git a/drivers/net/wireless/realtek/rtw89/rtw8851b_rfk.c b/drivers/net/wireless/realtek/rtw89/rtw8851b_rfk.c index 84c46d2f4d85..e574a9950a3b 100644 --- a/drivers/net/wireless/realtek/rtw89/rtw8851b_rfk.c +++ b/drivers/net/wireless/realtek/rtw89/rtw8851b_rfk.c @@ -1626,7 +1626,7 @@ static void _iqk_get_ch_info(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy, iqk_info->iqk_table_idx[path] = idx; rtw89_debug(rtwdev, RTW89_DBG_RFK, "[IQK]S%d (PHY%d): / DBCC %s/ %s/ CH%d/ %s\n", - path, phy, rtwdev->dbcc_en ? "on" : "off", + path, phy, str_on_off(rtwdev->dbcc_en), iqk_info->iqk_band[path] == 0 ? "2G" : iqk_info->iqk_band[path] == 1 ? "5G" : "6G", iqk_info->iqk_ch[path], @@ -1901,8 +1901,8 @@ static void _dpk_information(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy, rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DPK] S%d[%d] (PHY%d): TSSI %s/ DBCC %s/ %s/ CH%d/ %s\n", path, dpk->cur_idx[path], phy, - rtwdev->is_tssi_mode[path] ? "on" : "off", - rtwdev->dbcc_en ? "on" : "off", + str_on_off(rtwdev->is_tssi_mode[path]), + str_on_off(rtwdev->dbcc_en), dpk->bp[path][kidx].band == 0 ? "2G" : dpk->bp[path][kidx].band == 1 ? "5G" : "6G", dpk->bp[path][kidx].ch, @@ -2016,7 +2016,7 @@ static void _dpk_txpwr_bb_force(struct rtw89_dev *rtwdev, rtw89_phy_write32_mask(rtwdev, R_TXPWRB_H + (path << 13), B_TXPWRB_RDY, force); rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DPK] S%d txpwr_bb_force %s\n", - path, force ? "on" : "off"); + path, str_on_off(force)); } static void _dpk_kip_pwr_clk_onoff(struct rtw89_dev *rtwdev, bool turn_on) diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852a_rfk.c b/drivers/net/wireless/realtek/rtw89/rtw8852a_rfk.c index 9db8713ac99b..e74257d19412 100644 --- a/drivers/net/wireless/realtek/rtw89/rtw8852a_rfk.c +++ b/drivers/net/wireless/realtek/rtw89/rtw8852a_rfk.c @@ -1403,7 +1403,7 @@ static void _iqk_get_ch_info(struct rtw89_dev *rtwdev, path, iqk_info->iqk_ch[path]); rtw89_debug(rtwdev, RTW89_DBG_RFK, "[IQK]S%d (PHY%d): / DBCC %s/ %s/ CH%d/ %s\n", path, phy, - rtwdev->dbcc_en ? "on" : "off", + str_on_off(rtwdev->dbcc_en), iqk_info->iqk_band[path] == 0 ? "2G" : iqk_info->iqk_band[path] == 1 ? "5G" : "6G", iqk_info->iqk_ch[path], @@ -1881,8 +1881,8 @@ static void _dpk_information(struct rtw89_dev *rtwdev, rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DPK] S%d[%d] (PHY%d): TSSI %s/ DBCC %s/ %s/ CH%d/ %s\n", path, dpk->cur_idx[path], phy, - rtwdev->is_tssi_mode[path] ? "on" : "off", - rtwdev->dbcc_en ? "on" : "off", + str_on_off(rtwdev->is_tssi_mode[path]), + str_on_off(rtwdev->dbcc_en), dpk->bp[path][kidx].band == 0 ? "2G" : dpk->bp[path][kidx].band == 1 ? "5G" : "6G", dpk->bp[path][kidx].ch, @@ -2736,7 +2736,7 @@ static void _dpk_onoff(struct rtw89_dev *rtwdev, MASKBYTE3, 0x6 | val); rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DPK] S%d[%d] DPK %s !!!\n", path, - kidx, dpk->is_dpk_enable && !off ? "enable" : "disable"); + kidx, str_enable_disable(dpk->is_dpk_enable && !off)); } static void _dpk_track(struct rtw89_dev *rtwdev) diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852b_rfk.c b/drivers/net/wireless/realtek/rtw89/rtw8852b_rfk.c index 4796588c0256..70b1515c00fa 100644 --- a/drivers/net/wireless/realtek/rtw89/rtw8852b_rfk.c +++ b/drivers/net/wireless/realtek/rtw89/rtw8852b_rfk.c @@ -1696,7 +1696,7 @@ static void _dpk_onoff(struct rtw89_dev *rtwdev, enum rtw89_rf_path path, bool o MASKBYTE3, _dpk_order_convert(rtwdev) << 1 | val); rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DPK] S%d[%d] DPK %s !!!\n", path, - kidx, dpk->is_dpk_enable && !off ? "enable" : "disable"); + kidx, str_enable_disable(dpk->is_dpk_enable && !off)); } static void _dpk_one_shot(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy, @@ -1763,8 +1763,8 @@ static void _dpk_information(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy, rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DPK] S%d[%d] (PHY%d): TSSI %s/ DBCC %s/ %s/ CH%d/ %s\n", path, dpk->cur_idx[path], phy, - rtwdev->is_tssi_mode[path] ? "on" : "off", - rtwdev->dbcc_en ? "on" : "off", + str_on_off(rtwdev->is_tssi_mode[path]), + str_on_off(rtwdev->dbcc_en), dpk->bp[path][kidx].band == 0 ? "2G" : dpk->bp[path][kidx].band == 1 ? "5G" : "6G", dpk->bp[path][kidx].ch, diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852c_rfk.c b/drivers/net/wireless/realtek/rtw89/rtw8852c_rfk.c index b92e2ce4f4ad..e7fd028c5e82 100644 --- a/drivers/net/wireless/realtek/rtw89/rtw8852c_rfk.c +++ b/drivers/net/wireless/realtek/rtw89/rtw8852c_rfk.c @@ -1344,7 +1344,7 @@ static void _iqk_get_ch_info(struct rtw89_dev *rtwdev, path, iqk_info->iqk_ch[path]); rtw89_debug(rtwdev, RTW89_DBG_RFK, "[IQK]S%d (PHY%d): / DBCC %s/ %s/ CH%d/ %s\n", path, phy, - rtwdev->dbcc_en ? "on" : "off", + str_on_off(rtwdev->dbcc_en), iqk_info->iqk_band[path] == 0 ? "2G" : iqk_info->iqk_band[path] == 1 ? "5G" : "6G", iqk_info->iqk_ch[path], @@ -1920,8 +1920,8 @@ static void _dpk_information(struct rtw89_dev *rtwdev, rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DPK] S%d[%d] (PHY%d): TSSI %s/ DBCC %s/ %s/ CH%d/ %s\n", path, dpk->cur_idx[path], phy, - rtwdev->is_tssi_mode[path] ? "on" : "off", - rtwdev->dbcc_en ? "on" : "off", + str_on_off(rtwdev->is_tssi_mode[path]), + str_on_off(rtwdev->dbcc_en), dpk->bp[path][kidx].band == 0 ? "2G" : dpk->bp[path][kidx].band == 1 ? "5G" : "6G", dpk->bp[path][kidx].ch, @@ -2000,7 +2000,7 @@ static void _dpk_txpwr_bb_force(struct rtw89_dev *rtwdev, u8 path, bool force) rtw89_phy_write32_mask(rtwdev, R_TXPWRB_H + (path << 13), B_TXPWRB_RDY, force); rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DPK] S%d txpwr_bb_force %s\n", - path, force ? "on" : "off"); + path, str_on_off(force)); } static void _dpk_kip_restore(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy, @@ -2828,7 +2828,7 @@ static void _dpk_onoff(struct rtw89_dev *rtwdev, B_DPD_MEN, val); rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DPK] S%d[%d] DPK %s !!!\n", path, - kidx, dpk->is_dpk_enable && !off ? "enable" : "disable"); + kidx, str_enable_disable(dpk->is_dpk_enable && !off)); } static void _dpk_track(struct rtw89_dev *rtwdev) diff --git a/drivers/net/wireless/realtek/rtw89/usb.c b/drivers/net/wireless/realtek/rtw89/usb.c index 6cf89aee252e..afbd1fb79021 100644 --- a/drivers/net/wireless/realtek/rtw89/usb.c +++ b/drivers/net/wireless/realtek/rtw89/usb.c @@ -55,7 +55,7 @@ static void rtw89_usb_vendorreq(struct rtw89_dev *rtwdev, u32 addr, else if (ret < 0) rtw89_warn(rtwdev, "usb %s%u 0x%x fail ret=%d value=0x%x attempt=%d\n", - reqtype == RTW89_USB_VENQT_READ ? "read" : "write", + str_read_write(reqtype == RTW89_USB_VENQT_READ), len * 8, addr, ret, le32_to_cpup(rtwusb->vendor_req_buf), attempt); diff --git a/drivers/net/wireless/realtek/rtw89/wow.c b/drivers/net/wireless/realtek/rtw89/wow.c index 5faa51ad896a..b9fce0be3ab7 100644 --- a/drivers/net/wireless/realtek/rtw89/wow.c +++ b/drivers/net/wireless/realtek/rtw89/wow.c @@ -1248,7 +1248,7 @@ static int rtw89_wow_check_fw_status(struct rtw89_dev *rtwdev, bool wow_enable) mac->wow_ctrl.addr, mac->wow_ctrl.mask); if (ret) rtw89_err(rtwdev, "failed to check wow status %s\n", - wow_enable ? "enabled" : "disabled"); + str_enabled_disabled(wow_enable)); return ret; } From 7aeb8c118ef9c7a19ca9c22a2984f1d6653eff84 Mon Sep 17 00:00:00 2001 From: Bitterblue Smith Date: Thu, 9 Oct 2025 14:52:39 +0300 Subject: [PATCH 0156/1381] wifi: rtl8xxxu: Report the signal strength only if it's known These chips don't report the signal strength for many (any?) data frames. When the signal strength is not known, set RX_FLAG_NO_SIGNAL_VAL in order to avoid reporting a signal strength of 0. Tested with RTL8192FU and RTL8192CU. Signed-off-by: Bitterblue Smith Reviewed-by: Ping-Ke Shih Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/cdbe1a18-f6f1-4575-bad9-e85db8101247@gmail.com --- drivers/net/wireless/realtek/rtl8xxxu/core.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/net/wireless/realtek/rtl8xxxu/core.c b/drivers/net/wireless/realtek/rtl8xxxu/core.c index be39463bd6c4..2d88b7fba317 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/core.c +++ b/drivers/net/wireless/realtek/rtl8xxxu/core.c @@ -6454,6 +6454,8 @@ int rtl8xxxu_parse_rxdesc16(struct rtl8xxxu_priv *priv, struct sk_buff *skb) rtl8xxxu_rx_update_rssi(priv, rx_status, hdr); + } else { + rx_status->flag |= RX_FLAG_NO_SIGNAL_VAL; } rx_status->mactime = rx_desc->tsfl; @@ -6560,6 +6562,8 @@ int rtl8xxxu_parse_rxdesc24(struct rtl8xxxu_priv *priv, struct sk_buff *skb) rtl8xxxu_rx_update_rssi(priv, rx_status, hdr); + } else { + rx_status->flag |= RX_FLAG_NO_SIGNAL_VAL; } rx_status->mactime = rx_desc->tsfl; From fb5183aa65f6af7bc63869ad7f7d0e6505a80a58 Mon Sep 17 00:00:00 2001 From: Bitterblue Smith Date: Thu, 9 Oct 2025 14:53:37 +0300 Subject: [PATCH 0157/1381] wifi: rtl8xxxu: Dump the efuse right after reading it Someone reported a problem with their TP-Link WN722N (RTL8188EUS): usb 1-1: This Realtek USB WiFi dongle (0x2357:0x010c) is untested! usb 1-1: Please report results to Jes.Sorensen@gmail.com usb 1-1: Fatal - failed to parse EFuse rtl8xxxu: probe of 1-1:1.0 failed with error -22 This error can happen when the first two bytes of the efuse don't have the expected magic value of 0x8129. In a situation like this it could be useful to see the contents of the efuse. Dump the efuse right after reading it, before trying to parse it. Signed-off-by: Bitterblue Smith Reviewed-by: Ping-Ke Shih Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/1f544dc4-17f6-4401-995d-5cb4567b82fa@gmail.com --- drivers/net/wireless/realtek/rtl8xxxu/core.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/net/wireless/realtek/rtl8xxxu/core.c b/drivers/net/wireless/realtek/rtl8xxxu/core.c index 2d88b7fba317..cf631df54fa5 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/core.c +++ b/drivers/net/wireless/realtek/rtl8xxxu/core.c @@ -7910,15 +7910,15 @@ static int rtl8xxxu_probe(struct usb_interface *interface, goto err_set_intfdata; } + if (rtl8xxxu_debug & RTL8XXXU_DEBUG_EFUSE) + rtl8xxxu_dump_efuse(priv); + ret = priv->fops->parse_efuse(priv); if (ret) { dev_err(&udev->dev, "Fatal - failed to parse EFuse\n"); goto err_set_intfdata; } - if (rtl8xxxu_debug & RTL8XXXU_DEBUG_EFUSE) - rtl8xxxu_dump_efuse(priv); - rtl8xxxu_print_chipinfo(priv); ret = priv->fops->load_firmware(priv); From 9b567039e540705c5cf0f39a3344086b6f51b3dd Mon Sep 17 00:00:00 2001 From: Bitterblue Smith Date: Thu, 9 Oct 2025 14:55:08 +0300 Subject: [PATCH 0158/1381] wifi: rtl8xxxu: Use correct power off sequence for RTL8192CU RTL8192CU disappears and reappears when rtl8xxxu is unloaded: usbcore: deregistering interface driver rtl8xxxu wlp3s0f3u2: deauthenticating from ... by local choice (Reason: 3=DEAUTH_LEAVING) usb 1-2: rtl8xxxu_active_to_emu: Disabling MAC timed out usb 1-2: USB disconnect, device number 7 usb 1-2: disconnecting usb 1-2: new high-speed USB device number 8 using xhci_hcd usb 1-2: New USB device found, idVendor=0bda, idProduct=8178, bcdDevice= 2.00 usb 1-2: New USB device strings: Mfr=1, Product=2, SerialNumber=3 usb 1-2: Product: 802.11n WLAN Adapter usb 1-2: Manufacturer: Realtek usb 1-2: SerialNumber: 00e04c000001 This is because rtl8xxxu is using the power off sequence for RTL8723AU. Add the correct power off sequence for RTL8192CU. rtl8xxxu_power_off(), rtl8xxxu_active_to_emu(), and rtl8xxxu_emu_to_disabled() are now only used for RTL8723AU, so move them to 8723a.c and rename them to have the "rtl8723au" prefix. Tested only with RTL8192CU. Signed-off-by: Bitterblue Smith Reviewed-by: Ping-Ke Shih Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/b9d3f137-12ce-4bd9-8ada-3b8874bc3615@gmail.com --- drivers/net/wireless/realtek/rtl8xxxu/8192c.c | 80 +++++++++++- drivers/net/wireless/realtek/rtl8xxxu/8723a.c | 115 +++++++++++++++- drivers/net/wireless/realtek/rtl8xxxu/core.c | 123 ------------------ drivers/net/wireless/realtek/rtl8xxxu/regs.h | 1 + .../net/wireless/realtek/rtl8xxxu/rtl8xxxu.h | 1 - 5 files changed, 194 insertions(+), 126 deletions(-) diff --git a/drivers/net/wireless/realtek/rtl8xxxu/8192c.c b/drivers/net/wireless/realtek/rtl8xxxu/8192c.c index 73034e7e41d1..444872131c66 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/8192c.c +++ b/drivers/net/wireless/realtek/rtl8xxxu/8192c.c @@ -593,6 +593,84 @@ static int rtl8192cu_power_on(struct rtl8xxxu_priv *priv) return 0; } +static void rtl8192cu_power_off(struct rtl8xxxu_priv *priv) +{ + u32 val32; + u16 val16; + u8 val8; + int i; + + /* + * Workaround for 8188RU LNA power leakage problem. + */ + if (priv->rtl_chip == RTL8188R) { + val32 = rtl8xxxu_read32(priv, REG_FPGA0_XCD_RF_PARM); + val32 |= BIT(1); + rtl8xxxu_write32(priv, REG_FPGA0_XCD_RF_PARM, val32); + } + + /* _DisableRFAFEAndResetBB */ + rtl8xxxu_write8(priv, REG_TXPAUSE, 0xff); + rtl8xxxu_write_rfreg_mask(priv, RF_A, RF6052_REG_AC, 0xff, 0); + + rtl8xxxu_write8_set(priv, REG_APSD_CTRL, APSD_CTRL_OFF); + rtl8xxxu_write32_set(priv, REG_FPGA0_XCD_RF_PARM, FPGA0_RF_PARM_CLK_GATE); + + rtl8xxxu_write8(priv, REG_SYS_FUNC, + SYS_FUNC_USBA | SYS_FUNC_USBD | SYS_FUNC_BB_GLB_RSTN); + rtl8xxxu_write8(priv, REG_SYS_FUNC, SYS_FUNC_USBA | SYS_FUNC_USBD); + + /* _ResetDigitalProcedure1 */ + if (rtl8xxxu_read8(priv, REG_MCU_FW_DL) & MCU_FW_DL_READY) { + rtl8xxxu_write8(priv, REG_MCU_FW_DL, 0x00); + + rtl8xxxu_write8(priv, REG_FWIMR, 0x20); + + rtl8xxxu_write8(priv, REG_HMTFR + 3, 0x20); + + for (i = 0; i < 100; i++) { + val16 = rtl8xxxu_read16(priv, REG_SYS_FUNC); + if (!(val16 & SYS_FUNC_CPU_ENABLE)) + break; + + fsleep(50); + } + + if (i == 100) { + rtl8xxxu_write8(priv, REG_SYS_FUNC + 1, + (SYS_FUNC_HWPDN | SYS_FUNC_ELDR) >> 8); + msleep(10); + } + } + + val8 = (SYS_FUNC_HWPDN | SYS_FUNC_ELDR | SYS_FUNC_CPU_ENABLE) >> 8; + rtl8xxxu_write8(priv, REG_SYS_FUNC + 1, val8); + + /* _DisableGPIO */ + rtl8xxxu_write16(priv, REG_GPIO_PIN_CTRL + 2, 0); + val32 = rtl8xxxu_read32(priv, REG_GPIO_PIN_CTRL) & 0xffff00ff; + val32 |= (val32 & 0xff) << 8; + val32 |= 0x00ff0000; + rtl8xxxu_write32(priv, REG_GPIO_PIN_CTRL, val32); + + rtl8xxxu_write8(priv, REG_GPIO_MUXCFG + 3, 0); + val16 = rtl8xxxu_read16(priv, REG_GPIO_MUXCFG + 2) & 0xff0f; + val16 |= (val16 & 0xf) << 4; + val16 |= 0x0780; + rtl8xxxu_write16(priv, REG_GPIO_MUXCFG + 2, val16); + + /* _DisableAnalog */ + val8 = 0x23; + if (priv->vendor_umc && priv->chip_cut == 1) + val8 |= BIT(3); + rtl8xxxu_write8(priv, REG_SPS0_CTRL, val8); + + val16 = APS_FSMCO_HOST | APS_FSMCO_HW_SUSPEND | APS_FSMCO_PFM_ALDN; + rtl8xxxu_write16(priv, REG_APS_FSMCO, val16); + + rtl8xxxu_write8(priv, REG_RSV_CTRL, 0x0e); +} + static int rtl8192cu_led_brightness_set(struct led_classdev *led_cdev, enum led_brightness brightness) { @@ -618,7 +696,7 @@ struct rtl8xxxu_fileops rtl8192cu_fops = { .parse_efuse = rtl8192cu_parse_efuse, .load_firmware = rtl8192cu_load_firmware, .power_on = rtl8192cu_power_on, - .power_off = rtl8xxxu_power_off, + .power_off = rtl8192cu_power_off, .read_efuse = rtl8xxxu_read_efuse, .reset_8051 = rtl8xxxu_reset_8051, .llt_init = rtl8xxxu_init_llt_table, diff --git a/drivers/net/wireless/realtek/rtl8xxxu/8723a.c b/drivers/net/wireless/realtek/rtl8xxxu/8723a.c index ecbc324e4609..4f4493d0bfc2 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/8723a.c +++ b/drivers/net/wireless/realtek/rtl8xxxu/8723a.c @@ -411,6 +411,119 @@ static int rtl8723au_power_on(struct rtl8xxxu_priv *priv) return ret; } +static int rtl8723au_active_to_emu(struct rtl8xxxu_priv *priv) +{ + u8 val8; + int count, ret = 0; + + /* Start of rtl8723AU_card_enable_flow */ + /* Act to Cardemu sequence*/ + /* Turn off RF */ + rtl8xxxu_write8(priv, REG_RF_CTRL, 0); + + /* 0x004E[7] = 0, switch DPDT_SEL_P output from register 0x0065[2] */ + val8 = rtl8xxxu_read8(priv, REG_LEDCFG2); + val8 &= ~LEDCFG2_DPDT_SELECT; + rtl8xxxu_write8(priv, REG_LEDCFG2, val8); + + /* 0x0005[1] = 1 turn off MAC by HW state machine*/ + val8 = rtl8xxxu_read8(priv, REG_APS_FSMCO + 1); + val8 |= BIT(1); + rtl8xxxu_write8(priv, REG_APS_FSMCO + 1, val8); + + for (count = RTL8XXXU_MAX_REG_POLL; count; count--) { + val8 = rtl8xxxu_read8(priv, REG_APS_FSMCO + 1); + if ((val8 & BIT(1)) == 0) + break; + udelay(10); + } + + if (!count) { + dev_warn(&priv->udev->dev, "%s: Disabling MAC timed out\n", + __func__); + ret = -EBUSY; + goto exit; + } + + /* 0x0000[5] = 1 analog Ips to digital, 1:isolation */ + val8 = rtl8xxxu_read8(priv, REG_SYS_ISO_CTRL); + val8 |= SYS_ISO_ANALOG_IPS; + rtl8xxxu_write8(priv, REG_SYS_ISO_CTRL, val8); + + /* 0x0020[0] = 0 disable LDOA12 MACRO block*/ + val8 = rtl8xxxu_read8(priv, REG_LDOA15_CTRL); + val8 &= ~LDOA15_ENABLE; + rtl8xxxu_write8(priv, REG_LDOA15_CTRL, val8); + +exit: + return ret; +} + +static int rtl8723au_emu_to_disabled(struct rtl8xxxu_priv *priv) +{ + u8 val8; + + /* 0x0007[7:0] = 0x20 SOP option to disable BG/MB */ + rtl8xxxu_write8(priv, REG_APS_FSMCO + 3, 0x20); + + /* 0x04[12:11] = 01 enable WL suspend */ + val8 = rtl8xxxu_read8(priv, REG_APS_FSMCO + 1); + val8 &= ~BIT(4); + val8 |= BIT(3); + rtl8xxxu_write8(priv, REG_APS_FSMCO + 1, val8); + + val8 = rtl8xxxu_read8(priv, REG_APS_FSMCO + 1); + val8 |= BIT(7); + rtl8xxxu_write8(priv, REG_APS_FSMCO + 1, val8); + + /* 0x48[16] = 1 to enable GPIO9 as EXT wakeup */ + val8 = rtl8xxxu_read8(priv, REG_GPIO_INTM + 2); + val8 |= BIT(0); + rtl8xxxu_write8(priv, REG_GPIO_INTM + 2, val8); + + return 0; +} + +static void rtl8723au_power_off(struct rtl8xxxu_priv *priv) +{ + u8 val8; + u16 val16; + + rtl8xxxu_flush_fifo(priv); + + rtl8xxxu_active_to_lps(priv); + + /* Turn off RF */ + rtl8xxxu_write8(priv, REG_RF_CTRL, 0x00); + + /* Reset Firmware if running in RAM */ + if (rtl8xxxu_read8(priv, REG_MCU_FW_DL) & MCU_FW_RAM_SEL) + rtl8xxxu_firmware_self_reset(priv); + + /* Reset MCU */ + val16 = rtl8xxxu_read16(priv, REG_SYS_FUNC); + val16 &= ~SYS_FUNC_CPU_ENABLE; + rtl8xxxu_write16(priv, REG_SYS_FUNC, val16); + + /* Reset MCU ready status */ + rtl8xxxu_write8(priv, REG_MCU_FW_DL, 0x00); + + rtl8723au_active_to_emu(priv); + rtl8723au_emu_to_disabled(priv); + + /* Reset MCU IO Wrapper */ + val8 = rtl8xxxu_read8(priv, REG_RSV_CTRL + 1); + val8 &= ~BIT(0); + rtl8xxxu_write8(priv, REG_RSV_CTRL + 1, val8); + + val8 = rtl8xxxu_read8(priv, REG_RSV_CTRL + 1); + val8 |= BIT(0); + rtl8xxxu_write8(priv, REG_RSV_CTRL + 1, val8); + + /* RSV_CTRL 0x1C[7:0] = 0x0e lock ISO/CLK/Power control register */ + rtl8xxxu_write8(priv, REG_RSV_CTRL, 0x0e); +} + #define XTAL1 GENMASK(23, 18) #define XTAL0 GENMASK(17, 12) @@ -492,7 +605,7 @@ struct rtl8xxxu_fileops rtl8723au_fops = { .parse_efuse = rtl8723au_parse_efuse, .load_firmware = rtl8723au_load_firmware, .power_on = rtl8723au_power_on, - .power_off = rtl8xxxu_power_off, + .power_off = rtl8723au_power_off, .read_efuse = rtl8xxxu_read_efuse, .reset_8051 = rtl8xxxu_reset_8051, .llt_init = rtl8xxxu_init_llt_table, diff --git a/drivers/net/wireless/realtek/rtl8xxxu/core.c b/drivers/net/wireless/realtek/rtl8xxxu/core.c index cf631df54fa5..d82cb7714314 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/core.c +++ b/drivers/net/wireless/realtek/rtl8xxxu/core.c @@ -3637,54 +3637,6 @@ static void rtl8xxxu_set_ampdu_min_space(struct rtl8xxxu_priv *priv, u8 density) rtl8xxxu_write8(priv, REG_AMPDU_MIN_SPACE, val8); } -static int rtl8xxxu_active_to_emu(struct rtl8xxxu_priv *priv) -{ - u8 val8; - int count, ret = 0; - - /* Start of rtl8723AU_card_enable_flow */ - /* Act to Cardemu sequence*/ - /* Turn off RF */ - rtl8xxxu_write8(priv, REG_RF_CTRL, 0); - - /* 0x004E[7] = 0, switch DPDT_SEL_P output from register 0x0065[2] */ - val8 = rtl8xxxu_read8(priv, REG_LEDCFG2); - val8 &= ~LEDCFG2_DPDT_SELECT; - rtl8xxxu_write8(priv, REG_LEDCFG2, val8); - - /* 0x0005[1] = 1 turn off MAC by HW state machine*/ - val8 = rtl8xxxu_read8(priv, REG_APS_FSMCO + 1); - val8 |= BIT(1); - rtl8xxxu_write8(priv, REG_APS_FSMCO + 1, val8); - - for (count = RTL8XXXU_MAX_REG_POLL; count; count--) { - val8 = rtl8xxxu_read8(priv, REG_APS_FSMCO + 1); - if ((val8 & BIT(1)) == 0) - break; - udelay(10); - } - - if (!count) { - dev_warn(&priv->udev->dev, "%s: Disabling MAC timed out\n", - __func__); - ret = -EBUSY; - goto exit; - } - - /* 0x0000[5] = 1 analog Ips to digital, 1:isolation */ - val8 = rtl8xxxu_read8(priv, REG_SYS_ISO_CTRL); - val8 |= SYS_ISO_ANALOG_IPS; - rtl8xxxu_write8(priv, REG_SYS_ISO_CTRL, val8); - - /* 0x0020[0] = 0 disable LDOA12 MACRO block*/ - val8 = rtl8xxxu_read8(priv, REG_LDOA15_CTRL); - val8 &= ~LDOA15_ENABLE; - rtl8xxxu_write8(priv, REG_LDOA15_CTRL, val8); - -exit: - return ret; -} - int rtl8xxxu_active_to_lps(struct rtl8xxxu_priv *priv) { u8 val8; @@ -3761,31 +3713,6 @@ void rtl8xxxu_disabled_to_emu(struct rtl8xxxu_priv *priv) rtl8xxxu_write8(priv, REG_APS_FSMCO + 1, val8); } -static int rtl8xxxu_emu_to_disabled(struct rtl8xxxu_priv *priv) -{ - u8 val8; - - /* 0x0007[7:0] = 0x20 SOP option to disable BG/MB */ - rtl8xxxu_write8(priv, REG_APS_FSMCO + 3, 0x20); - - /* 0x04[12:11] = 01 enable WL suspend */ - val8 = rtl8xxxu_read8(priv, REG_APS_FSMCO + 1); - val8 &= ~BIT(4); - val8 |= BIT(3); - rtl8xxxu_write8(priv, REG_APS_FSMCO + 1, val8); - - val8 = rtl8xxxu_read8(priv, REG_APS_FSMCO + 1); - val8 |= BIT(7); - rtl8xxxu_write8(priv, REG_APS_FSMCO + 1, val8); - - /* 0x48[16] = 1 to enable GPIO9 as EXT wakeup */ - val8 = rtl8xxxu_read8(priv, REG_GPIO_INTM + 2); - val8 |= BIT(0); - rtl8xxxu_write8(priv, REG_GPIO_INTM + 2, val8); - - return 0; -} - int rtl8xxxu_flush_fifo(struct rtl8xxxu_priv *priv) { struct device *dev = &priv->udev->dev; @@ -3863,56 +3790,6 @@ void rtl8xxxu_gen2_usb_quirks(struct rtl8xxxu_priv *priv) rtl8xxxu_write32(priv, REG_TXDMA_OFFSET_CHK, val32); } -void rtl8xxxu_power_off(struct rtl8xxxu_priv *priv) -{ - u8 val8; - u16 val16; - u32 val32; - - /* - * Workaround for 8188RU LNA power leakage problem. - */ - if (priv->rtl_chip == RTL8188R) { - val32 = rtl8xxxu_read32(priv, REG_FPGA0_XCD_RF_PARM); - val32 |= BIT(1); - rtl8xxxu_write32(priv, REG_FPGA0_XCD_RF_PARM, val32); - } - - rtl8xxxu_flush_fifo(priv); - - rtl8xxxu_active_to_lps(priv); - - /* Turn off RF */ - rtl8xxxu_write8(priv, REG_RF_CTRL, 0x00); - - /* Reset Firmware if running in RAM */ - if (rtl8xxxu_read8(priv, REG_MCU_FW_DL) & MCU_FW_RAM_SEL) - rtl8xxxu_firmware_self_reset(priv); - - /* Reset MCU */ - val16 = rtl8xxxu_read16(priv, REG_SYS_FUNC); - val16 &= ~SYS_FUNC_CPU_ENABLE; - rtl8xxxu_write16(priv, REG_SYS_FUNC, val16); - - /* Reset MCU ready status */ - rtl8xxxu_write8(priv, REG_MCU_FW_DL, 0x00); - - rtl8xxxu_active_to_emu(priv); - rtl8xxxu_emu_to_disabled(priv); - - /* Reset MCU IO Wrapper */ - val8 = rtl8xxxu_read8(priv, REG_RSV_CTRL + 1); - val8 &= ~BIT(0); - rtl8xxxu_write8(priv, REG_RSV_CTRL + 1, val8); - - val8 = rtl8xxxu_read8(priv, REG_RSV_CTRL + 1); - val8 |= BIT(0); - rtl8xxxu_write8(priv, REG_RSV_CTRL + 1, val8); - - /* RSV_CTRL 0x1C[7:0] = 0x0e lock ISO/CLK/Power control register */ - rtl8xxxu_write8(priv, REG_RSV_CTRL, 0x0e); -} - void rtl8723bu_set_ps_tdma(struct rtl8xxxu_priv *priv, u8 arg1, u8 arg2, u8 arg3, u8 arg4, u8 arg5) { diff --git a/drivers/net/wireless/realtek/rtl8xxxu/regs.h b/drivers/net/wireless/realtek/rtl8xxxu/regs.h index 61c0c0ec07b3..0741db8d08bf 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/regs.h +++ b/drivers/net/wireless/realtek/rtl8xxxu/regs.h @@ -40,6 +40,7 @@ #define APS_FSMCO_SW_LPS BIT(10) #define APS_FSMCO_HW_SUSPEND BIT(11) #define APS_FSMCO_PCIE BIT(12) +#define APS_FSMCO_HOST BIT(14) #define APS_FSMCO_HW_POWERDOWN BIT(15) #define APS_FSMCO_WLON_RESET BIT(16) diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h index f42463e595cc..9fb2583ffffc 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h +++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h @@ -2078,7 +2078,6 @@ int rtl8xxxu_init_phy_regs(struct rtl8xxxu_priv *priv, const struct rtl8xxxu_reg32val *array); int rtl8xxxu_load_firmware(struct rtl8xxxu_priv *priv, const char *fw_name); void rtl8xxxu_firmware_self_reset(struct rtl8xxxu_priv *priv); -void rtl8xxxu_power_off(struct rtl8xxxu_priv *priv); void rtl8xxxu_identify_vendor_1bit(struct rtl8xxxu_priv *priv, u32 vendor); void rtl8xxxu_identify_vendor_2bits(struct rtl8xxxu_priv *priv, u32 vendor); void rtl8xxxu_config_endpoints_sie(struct rtl8xxxu_priv *priv); From 122d696c1789621a367d700d777c424f8494a5fa Mon Sep 17 00:00:00 2001 From: Bagas Sanjaya Date: Fri, 17 Oct 2025 13:45:25 +0700 Subject: [PATCH 0159/1381] net: nfc: Format userspace interface subsection headings Subsection headings of "Userspace interface" is written in normal paragraph, all-capped. Properly format them as reST section headings. Signed-off-by: Bagas Sanjaya Link: https://patch.msgid.link/20251017064525.28836-3-bagasdotme@gmail.com Signed-off-by: Paolo Abeni --- Documentation/networking/nfc.rst | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/Documentation/networking/nfc.rst b/Documentation/networking/nfc.rst index 9aab3a88c9b2..401735006143 100644 --- a/Documentation/networking/nfc.rst +++ b/Documentation/networking/nfc.rst @@ -71,7 +71,8 @@ Userspace interface The userspace interface is divided in control operations and low-level data exchange operation. -CONTROL OPERATIONS: +Control operations +------------------ Generic netlink is used to implement the interface to the control operations. The operations are composed by commands and events, all listed below: @@ -100,7 +101,8 @@ relevant information such as the supported NFC protocols. All polling operations requested through one netlink socket are stopped when it's closed. -LOW-LEVEL DATA EXCHANGE: +Low-level data exchange +----------------------- The userspace must use PF_NFC sockets to perform any data communication with targets. All NFC sockets use AF_NFC:: From 97aa8ecb573213b79667a40b5fe5345417dfe961 Mon Sep 17 00:00:00 2001 From: Bagas Sanjaya Date: Fri, 17 Oct 2025 13:45:26 +0700 Subject: [PATCH 0160/1381] net: 6pack: Demote "How to turn on 6pack support" section heading "How to turn on 6pack support" is a subsection of "Building and installing the 6pack driver". Yet, the former is in the same heading level as the latter as sections, making it listed in networking docs toctree. Demote it to subsection. Signed-off-by: Bagas Sanjaya Link: https://patch.msgid.link/20251017064525.28836-4-bagasdotme@gmail.com Signed-off-by: Paolo Abeni --- Documentation/networking/6pack.rst | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Documentation/networking/6pack.rst b/Documentation/networking/6pack.rst index bc5bf1f1a98f..66d5fd4fc821 100644 --- a/Documentation/networking/6pack.rst +++ b/Documentation/networking/6pack.rst @@ -94,7 +94,7 @@ kernels may lead to a compilation error because the interface to a kernel function has been changed in the 2.1.8x kernels. How to turn on 6pack support: -============================= +----------------------------- - In the linux kernel configuration program, select the code maturity level options menu and turn on the prompting for development drivers. From 51538c0c9d8cd59bac83cce51b77ccc66ca9fdb8 Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Fri, 17 Oct 2025 11:06:10 +0200 Subject: [PATCH 0161/1381] dt-bindings: net: airoha: Add AN7583 support Introduce AN7583 ethernet controller support to Airoha EN7581 device-tree bindings. The main difference between EN7581 and AN7583 is the number of reset lines required by the controller (AN7583 does not require hsi-mac). Signed-off-by: Lorenzo Bianconi Reviewed-by: Conor Dooley Link: https://patch.msgid.link/20251017-an7583-eth-support-v3-1-f28319666667@kernel.org Signed-off-by: Paolo Abeni --- .../bindings/net/airoha,en7581-eth.yaml | 35 ++++++++++++++++++- 1 file changed, 34 insertions(+), 1 deletion(-) diff --git a/Documentation/devicetree/bindings/net/airoha,en7581-eth.yaml b/Documentation/devicetree/bindings/net/airoha,en7581-eth.yaml index 6d22131ac2f9..fbe2ddcdd909 100644 --- a/Documentation/devicetree/bindings/net/airoha,en7581-eth.yaml +++ b/Documentation/devicetree/bindings/net/airoha,en7581-eth.yaml @@ -17,6 +17,7 @@ properties: compatible: enum: - airoha,en7581-eth + - airoha,an7583-eth reg: items: @@ -44,6 +45,7 @@ properties: - description: PDMA irq resets: + minItems: 7 maxItems: 8 reset-names: @@ -54,8 +56,9 @@ properties: - const: xsi-mac - const: hsi0-mac - const: hsi1-mac - - const: hsi-mac + - enum: [ hsi-mac, xfp-mac ] - const: xfp-mac + minItems: 7 memory-region: items: @@ -81,6 +84,36 @@ properties: interface to implement hardware flow offloading programming Packet Processor Engine (PPE) flow table. +allOf: + - $ref: ethernet-controller.yaml# + - if: + properties: + compatible: + contains: + enum: + - airoha,en7581-eth + then: + properties: + resets: + minItems: 8 + + reset-names: + minItems: 8 + + - if: + properties: + compatible: + contains: + enum: + - airoha,an7583-eth + then: + properties: + resets: + maxItems: 7 + + reset-names: + maxItems: 7 + patternProperties: "^ethernet@[1-4]$": type: object From 6d5b601d52a27aafff555b480e538507901c672c Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Fri, 17 Oct 2025 11:06:11 +0200 Subject: [PATCH 0162/1381] net: airoha: ppe: Dynamically allocate foe_check_time array in airoha_ppe struct This is a preliminary patch to properly enable PPE support for AN7583 SoC. Reviewed-by: Simon Horman Signed-off-by: Lorenzo Bianconi Link: https://patch.msgid.link/20251017-an7583-eth-support-v3-2-f28319666667@kernel.org Signed-off-by: Paolo Abeni --- drivers/net/ethernet/airoha/airoha_eth.h | 2 +- drivers/net/ethernet/airoha/airoha_ppe.c | 5 +++++ 2 files changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/airoha/airoha_eth.h b/drivers/net/ethernet/airoha/airoha_eth.h index cd13c1c1224f..4330b672d99e 100644 --- a/drivers/net/ethernet/airoha/airoha_eth.h +++ b/drivers/net/ethernet/airoha/airoha_eth.h @@ -554,7 +554,7 @@ struct airoha_ppe { struct rhashtable l2_flows; struct hlist_head *foe_flow; - u16 foe_check_time[PPE_NUM_ENTRIES]; + u16 *foe_check_time; struct airoha_foe_stats *foe_stats; dma_addr_t foe_stats_dma; diff --git a/drivers/net/ethernet/airoha/airoha_ppe.c b/drivers/net/ethernet/airoha/airoha_ppe.c index 691361b25407..8d1dceadce0b 100644 --- a/drivers/net/ethernet/airoha/airoha_ppe.c +++ b/drivers/net/ethernet/airoha/airoha_ppe.c @@ -1440,6 +1440,11 @@ int airoha_ppe_init(struct airoha_eth *eth) return -ENOMEM; } + ppe->foe_check_time = devm_kzalloc(eth->dev, PPE_NUM_ENTRIES, + GFP_KERNEL); + if (!ppe->foe_check_time) + return -ENOMEM; + err = rhashtable_init(ð->flow_table, &airoha_flow_table_params); if (err) return err; From 15f357cd4581ce6e02e5e97719320600783140ec Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Fri, 17 Oct 2025 11:06:12 +0200 Subject: [PATCH 0163/1381] net: airoha: Add airoha_ppe_get_num_stats_entries() and airoha_ppe_get_num_total_stats_entries() Introduce airoha_ppe_get_num_stats_entries and airoha_ppe_get_num_total_stats_entries routines in order to make the code more readable controlling if CONFIG_NET_AIROHA_FLOW_STATS is enabled or disabled. Modify airoha_ppe_foe_get_flow_stats_index routine signature relying on airoha_ppe_get_num_total_stats_entries(). Reviewed-by: Simon Horman Signed-off-by: Lorenzo Bianconi Link: https://patch.msgid.link/20251017-an7583-eth-support-v3-3-f28319666667@kernel.org Signed-off-by: Paolo Abeni --- drivers/net/ethernet/airoha/airoha_eth.h | 10 +-- drivers/net/ethernet/airoha/airoha_ppe.c | 101 ++++++++++++++++++----- 2 files changed, 81 insertions(+), 30 deletions(-) diff --git a/drivers/net/ethernet/airoha/airoha_eth.h b/drivers/net/ethernet/airoha/airoha_eth.h index 4330b672d99e..1f7e34a5f457 100644 --- a/drivers/net/ethernet/airoha/airoha_eth.h +++ b/drivers/net/ethernet/airoha/airoha_eth.h @@ -50,15 +50,9 @@ #define PPE_NUM 2 #define PPE1_SRAM_NUM_ENTRIES (8 * 1024) -#define PPE_SRAM_NUM_ENTRIES (2 * PPE1_SRAM_NUM_ENTRIES) -#ifdef CONFIG_NET_AIROHA_FLOW_STATS +#define PPE_SRAM_NUM_ENTRIES (PPE_NUM * PPE1_SRAM_NUM_ENTRIES) #define PPE1_STATS_NUM_ENTRIES (4 * 1024) -#else -#define PPE1_STATS_NUM_ENTRIES 0 -#endif /* CONFIG_NET_AIROHA_FLOW_STATS */ -#define PPE_STATS_NUM_ENTRIES (2 * PPE1_STATS_NUM_ENTRIES) -#define PPE1_SRAM_NUM_DATA_ENTRIES (PPE1_SRAM_NUM_ENTRIES - PPE1_STATS_NUM_ENTRIES) -#define PPE_SRAM_NUM_DATA_ENTRIES (2 * PPE1_SRAM_NUM_DATA_ENTRIES) +#define PPE_STATS_NUM_ENTRIES (PPE_NUM * PPE1_STATS_NUM_ENTRIES) #define PPE_DRAM_NUM_ENTRIES (16 * 1024) #define PPE_NUM_ENTRIES (PPE_SRAM_NUM_ENTRIES + PPE_DRAM_NUM_ENTRIES) #define PPE_HASH_MASK (PPE_NUM_ENTRIES - 1) diff --git a/drivers/net/ethernet/airoha/airoha_ppe.c b/drivers/net/ethernet/airoha/airoha_ppe.c index 8d1dceadce0b..22ecece0e33e 100644 --- a/drivers/net/ethernet/airoha/airoha_ppe.c +++ b/drivers/net/ethernet/airoha/airoha_ppe.c @@ -32,6 +32,24 @@ static const struct rhashtable_params airoha_l2_flow_table_params = { .automatic_shrinking = true, }; +static int airoha_ppe_get_num_stats_entries(struct airoha_ppe *ppe) +{ + if (!IS_ENABLED(CONFIG_NET_AIROHA_FLOW_STATS)) + return -EOPNOTSUPP; + + return PPE1_STATS_NUM_ENTRIES; +} + +static int airoha_ppe_get_total_num_stats_entries(struct airoha_ppe *ppe) +{ + int num_stats = airoha_ppe_get_num_stats_entries(ppe); + + if (num_stats > 0) + num_stats = num_stats * PPE_NUM; + + return num_stats; +} + static bool airoha_ppe2_is_enabled(struct airoha_eth *eth) { return airoha_fe_rr(eth, REG_PPE_GLO_CFG(1)) & PPE_GLO_CFG_EN_MASK; @@ -48,7 +66,7 @@ static void airoha_ppe_hw_init(struct airoha_ppe *ppe) { u32 sram_tb_size, sram_num_entries, dram_num_entries; struct airoha_eth *eth = ppe->eth; - int i; + int i, sram_num_stats_entries; sram_tb_size = PPE_SRAM_NUM_ENTRIES * sizeof(struct airoha_foe_entry); dram_num_entries = PPE_RAM_NUM_ENTRIES_SHIFT(PPE_DRAM_NUM_ENTRIES); @@ -103,8 +121,13 @@ static void airoha_ppe_hw_init(struct airoha_ppe *ppe) } if (airoha_ppe2_is_enabled(eth)) { - sram_num_entries = - PPE_RAM_NUM_ENTRIES_SHIFT(PPE1_SRAM_NUM_DATA_ENTRIES); + sram_num_entries = PPE1_SRAM_NUM_ENTRIES; + sram_num_stats_entries = + airoha_ppe_get_num_stats_entries(ppe); + if (sram_num_stats_entries > 0) + sram_num_entries -= sram_num_stats_entries; + sram_num_entries = PPE_RAM_NUM_ENTRIES_SHIFT(sram_num_entries); + airoha_fe_rmw(eth, REG_PPE_TB_CFG(0), PPE_SRAM_TB_NUM_ENTRY_MASK | PPE_DRAM_TB_NUM_ENTRY_MASK, @@ -120,8 +143,13 @@ static void airoha_ppe_hw_init(struct airoha_ppe *ppe) FIELD_PREP(PPE_DRAM_TB_NUM_ENTRY_MASK, dram_num_entries)); } else { - sram_num_entries = - PPE_RAM_NUM_ENTRIES_SHIFT(PPE_SRAM_NUM_DATA_ENTRIES); + sram_num_entries = PPE_SRAM_NUM_ENTRIES; + sram_num_stats_entries = + airoha_ppe_get_total_num_stats_entries(ppe); + if (sram_num_stats_entries > 0) + sram_num_entries -= sram_num_stats_entries; + sram_num_entries = PPE_RAM_NUM_ENTRIES_SHIFT(sram_num_entries); + airoha_fe_rmw(eth, REG_PPE_TB_CFG(0), PPE_SRAM_TB_NUM_ENTRY_MASK | PPE_DRAM_TB_NUM_ENTRY_MASK, @@ -480,13 +508,21 @@ static u32 airoha_ppe_foe_get_entry_hash(struct airoha_foe_entry *hwe) return hash; } -static u32 airoha_ppe_foe_get_flow_stats_index(struct airoha_ppe *ppe, u32 hash) +static int airoha_ppe_foe_get_flow_stats_index(struct airoha_ppe *ppe, + u32 hash, u32 *index) { - if (!airoha_ppe2_is_enabled(ppe->eth)) - return hash; + int ppe_num_stats_entries; - return hash >= PPE_STATS_NUM_ENTRIES ? hash - PPE1_STATS_NUM_ENTRIES - : hash; + ppe_num_stats_entries = airoha_ppe_get_total_num_stats_entries(ppe); + if (ppe_num_stats_entries < 0) + return ppe_num_stats_entries; + + *index = hash; + if (airoha_ppe2_is_enabled(ppe->eth) && + hash >= ppe_num_stats_entries) + *index = *index - PPE_STATS_NUM_ENTRIES; + + return 0; } static void airoha_ppe_foe_flow_stat_entry_reset(struct airoha_ppe *ppe, @@ -500,9 +536,13 @@ static void airoha_ppe_foe_flow_stat_entry_reset(struct airoha_ppe *ppe, static void airoha_ppe_foe_flow_stats_reset(struct airoha_ppe *ppe, struct airoha_npu *npu) { - int i; + int i, ppe_num_stats_entries; - for (i = 0; i < PPE_STATS_NUM_ENTRIES; i++) + ppe_num_stats_entries = airoha_ppe_get_total_num_stats_entries(ppe); + if (ppe_num_stats_entries < 0) + return; + + for (i = 0; i < ppe_num_stats_entries; i++) airoha_ppe_foe_flow_stat_entry_reset(ppe, npu, i); } @@ -513,10 +553,17 @@ static void airoha_ppe_foe_flow_stats_update(struct airoha_ppe *ppe, { int type = FIELD_GET(AIROHA_FOE_IB1_BIND_PACKET_TYPE, hwe->ib1); u32 index, pse_port, val, *data, *ib2, *meter; + int ppe_num_stats_entries; u8 nbq; - index = airoha_ppe_foe_get_flow_stats_index(ppe, hash); - if (index >= PPE_STATS_NUM_ENTRIES) + ppe_num_stats_entries = airoha_ppe_get_total_num_stats_entries(ppe); + if (ppe_num_stats_entries < 0) + return; + + if (airoha_ppe_foe_get_flow_stats_index(ppe, hash, &index)) + return; + + if (index >= ppe_num_stats_entries) return; if (type == PPE_PKT_TYPE_BRIDGE) { @@ -1158,11 +1205,19 @@ static int airoha_ppe_flow_offload_destroy(struct airoha_eth *eth, void airoha_ppe_foe_entry_get_stats(struct airoha_ppe *ppe, u32 hash, struct airoha_foe_stats64 *stats) { - u32 index = airoha_ppe_foe_get_flow_stats_index(ppe, hash); struct airoha_eth *eth = ppe->eth; + int ppe_num_stats_entries; struct airoha_npu *npu; + u32 index; - if (index >= PPE_STATS_NUM_ENTRIES) + ppe_num_stats_entries = airoha_ppe_get_total_num_stats_entries(ppe); + if (ppe_num_stats_entries < 0) + return; + + if (airoha_ppe_foe_get_flow_stats_index(ppe, hash, &index)) + return; + + if (index >= ppe_num_stats_entries) return; rcu_read_lock(); @@ -1257,7 +1312,7 @@ static int airoha_ppe_offload_setup(struct airoha_eth *eth) { struct airoha_npu *npu = airoha_ppe_npu_get(eth); struct airoha_ppe *ppe = eth->ppe; - int err; + int err, ppe_num_stats_entries; if (IS_ERR(npu)) return PTR_ERR(npu); @@ -1266,9 +1321,10 @@ static int airoha_ppe_offload_setup(struct airoha_eth *eth) if (err) goto error_npu_put; - if (PPE_STATS_NUM_ENTRIES) { + ppe_num_stats_entries = airoha_ppe_get_total_num_stats_entries(ppe); + if (ppe_num_stats_entries > 0) { err = npu->ops.ppe_init_stats(npu, ppe->foe_stats_dma, - PPE_STATS_NUM_ENTRIES); + ppe_num_stats_entries); if (err) goto error_npu_put; } @@ -1405,8 +1461,8 @@ EXPORT_SYMBOL_GPL(airoha_ppe_put_dev); int airoha_ppe_init(struct airoha_eth *eth) { + int foe_size, err, ppe_num_stats_entries; struct airoha_ppe *ppe; - int foe_size, err; ppe = devm_kzalloc(eth->dev, sizeof(*ppe), GFP_KERNEL); if (!ppe) @@ -1431,8 +1487,9 @@ int airoha_ppe_init(struct airoha_eth *eth) if (!ppe->foe_flow) return -ENOMEM; - foe_size = PPE_STATS_NUM_ENTRIES * sizeof(*ppe->foe_stats); - if (foe_size) { + ppe_num_stats_entries = airoha_ppe_get_total_num_stats_entries(ppe); + if (ppe_num_stats_entries > 0) { + foe_size = ppe_num_stats_entries * sizeof(*ppe->foe_stats); ppe->foe_stats = dmam_alloc_coherent(eth->dev, foe_size, &ppe->foe_stats_dma, GFP_KERNEL); From 5863b4e065e2253ef05684f728a04e4972046bcb Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Fri, 17 Oct 2025 11:06:13 +0200 Subject: [PATCH 0164/1381] net: airoha: Add airoha_eth_soc_data struct Introduce airoha_eth_soc_data struct to contain differences between various SoC. Move XSI reset names in airoha_eth_soc_data. This is a preliminary patch to enable AN7583 ethernet controller support in airoha-eth driver. Co-developed-by: Christian Marangi Signed-off-by: Christian Marangi Reviewed-by: Simon Horman Signed-off-by: Lorenzo Bianconi Link: https://patch.msgid.link/20251017-an7583-eth-support-v3-4-f28319666667@kernel.org Signed-off-by: Paolo Abeni --- drivers/net/ethernet/airoha/airoha_eth.c | 42 +++++++++++++++++++----- drivers/net/ethernet/airoha/airoha_eth.h | 17 ++++++++-- 2 files changed, 48 insertions(+), 11 deletions(-) diff --git a/drivers/net/ethernet/airoha/airoha_eth.c b/drivers/net/ethernet/airoha/airoha_eth.c index 5825f6f29a92..c9cebe6752eb 100644 --- a/drivers/net/ethernet/airoha/airoha_eth.c +++ b/drivers/net/ethernet/airoha/airoha_eth.c @@ -1387,8 +1387,7 @@ static int airoha_hw_init(struct platform_device *pdev, int err, i; /* disable xsi */ - err = reset_control_bulk_assert(ARRAY_SIZE(eth->xsi_rsts), - eth->xsi_rsts); + err = reset_control_bulk_assert(eth->soc->num_xsi_rsts, eth->xsi_rsts); if (err) return err; @@ -2922,6 +2921,7 @@ static int airoha_alloc_gdm_port(struct airoha_eth *eth, static int airoha_probe(struct platform_device *pdev) { + struct reset_control_bulk_data *xsi_rsts; struct device_node *np; struct airoha_eth *eth; int i, err; @@ -2930,6 +2930,10 @@ static int airoha_probe(struct platform_device *pdev) if (!eth) return -ENOMEM; + eth->soc = of_device_get_match_data(&pdev->dev); + if (!eth->soc) + return -EINVAL; + eth->dev = &pdev->dev; err = dma_set_mask_and_coherent(eth->dev, DMA_BIT_MASK(32)); @@ -2954,13 +2958,18 @@ static int airoha_probe(struct platform_device *pdev) return err; } - eth->xsi_rsts[0].id = "xsi-mac"; - eth->xsi_rsts[1].id = "hsi0-mac"; - eth->xsi_rsts[2].id = "hsi1-mac"; - eth->xsi_rsts[3].id = "hsi-mac"; - eth->xsi_rsts[4].id = "xfp-mac"; + xsi_rsts = devm_kzalloc(eth->dev, + eth->soc->num_xsi_rsts * sizeof(*xsi_rsts), + GFP_KERNEL); + if (err) + return err; + + eth->xsi_rsts = xsi_rsts; + for (i = 0; i < eth->soc->num_xsi_rsts; i++) + eth->xsi_rsts[i].id = eth->soc->xsi_rsts_names[i]; + err = devm_reset_control_bulk_get_exclusive(eth->dev, - ARRAY_SIZE(eth->xsi_rsts), + eth->soc->num_xsi_rsts, eth->xsi_rsts); if (err) { dev_err(eth->dev, "failed to get bulk xsi reset lines\n"); @@ -3048,8 +3057,23 @@ static void airoha_remove(struct platform_device *pdev) platform_set_drvdata(pdev, NULL); } +static const char * const en7581_xsi_rsts_names[] = { + "xsi-mac", + "hsi0-mac", + "hsi1-mac", + "hsi-mac", + "xfp-mac", +}; + +static const struct airoha_eth_soc_data en7581_soc_data = { + .version = 0x7581, + .xsi_rsts_names = en7581_xsi_rsts_names, + .num_xsi_rsts = ARRAY_SIZE(en7581_xsi_rsts_names), + .num_ppe = 2, +}; + static const struct of_device_id of_airoha_match[] = { - { .compatible = "airoha,en7581-eth" }, + { .compatible = "airoha,en7581-eth", .data = &en7581_soc_data }, { /* sentinel */ } }; MODULE_DEVICE_TABLE(of, of_airoha_match); diff --git a/drivers/net/ethernet/airoha/airoha_eth.h b/drivers/net/ethernet/airoha/airoha_eth.h index 1f7e34a5f457..cb7e198e40ee 100644 --- a/drivers/net/ethernet/airoha/airoha_eth.h +++ b/drivers/net/ethernet/airoha/airoha_eth.h @@ -21,7 +21,6 @@ #define AIROHA_MAX_NUM_IRQ_BANKS 4 #define AIROHA_MAX_DSA_PORTS 7 #define AIROHA_MAX_NUM_RSTS 3 -#define AIROHA_MAX_NUM_XSI_RSTS 5 #define AIROHA_MAX_MTU 9216 #define AIROHA_MAX_PACKET_SIZE 2048 #define AIROHA_NUM_QOS_CHANNELS 4 @@ -556,9 +555,18 @@ struct airoha_ppe { struct dentry *debugfs_dir; }; +struct airoha_eth_soc_data { + u16 version; + const char * const *xsi_rsts_names; + int num_xsi_rsts; + int num_ppe; +}; + struct airoha_eth { struct device *dev; + const struct airoha_eth_soc_data *soc; + unsigned long state; void __iomem *fe_regs; @@ -568,7 +576,7 @@ struct airoha_eth { struct rhashtable flow_table; struct reset_control_bulk_data rsts[AIROHA_MAX_NUM_RSTS]; - struct reset_control_bulk_data xsi_rsts[AIROHA_MAX_NUM_XSI_RSTS]; + struct reset_control_bulk_data *xsi_rsts; struct net_device *napi_dev; @@ -611,6 +619,11 @@ static inline bool airhoa_is_lan_gdm_port(struct airoha_gdm_port *port) return port->id == 1; } +static inline bool airoha_is_7581(struct airoha_eth *eth) +{ + return eth->soc->version == 0x7581; +} + bool airoha_is_valid_gdm_port(struct airoha_eth *eth, struct airoha_gdm_port *port); From ef9449f080b61920cdc3d3106f8ffc2d9ba8b861 Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Fri, 17 Oct 2025 11:06:14 +0200 Subject: [PATCH 0165/1381] net: airoha: Generalize airoha_ppe2_is_enabled routine Rename airoha_ppe2_is_enabled() in airoha_ppe_is_enabled() and generalize it in order to check if each PPE module is enabled. Rely on airoha_ppe_is_enabled routine to properly initialize PPE for AN7583 SoC since AN7583 does not support PPE2. Reviewed-by: Simon Horman Signed-off-by: Lorenzo Bianconi Link: https://patch.msgid.link/20251017-an7583-eth-support-v3-5-f28319666667@kernel.org Signed-off-by: Paolo Abeni --- drivers/net/ethernet/airoha/airoha_eth.c | 32 ++++++++++++++++-------- drivers/net/ethernet/airoha/airoha_eth.h | 1 + drivers/net/ethernet/airoha/airoha_ppe.c | 17 +++++++------ 3 files changed, 32 insertions(+), 18 deletions(-) diff --git a/drivers/net/ethernet/airoha/airoha_eth.c b/drivers/net/ethernet/airoha/airoha_eth.c index c9cebe6752eb..dea856ddf242 100644 --- a/drivers/net/ethernet/airoha/airoha_eth.c +++ b/drivers/net/ethernet/airoha/airoha_eth.c @@ -297,8 +297,11 @@ static void airoha_fe_pse_ports_init(struct airoha_eth *eth) int q; all_rsv = airoha_fe_get_pse_all_rsv(eth); - /* hw misses PPE2 oq rsv */ - all_rsv += PSE_RSV_PAGES * pse_port_num_queues[FE_PSE_PORT_PPE2]; + if (airoha_ppe_is_enabled(eth, 1)) { + /* hw misses PPE2 oq rsv */ + all_rsv += PSE_RSV_PAGES * + pse_port_num_queues[FE_PSE_PORT_PPE2]; + } airoha_fe_set(eth, REG_FE_PSE_BUF_SET, all_rsv); /* CMD1 */ @@ -335,13 +338,17 @@ static void airoha_fe_pse_ports_init(struct airoha_eth *eth) for (q = 4; q < pse_port_num_queues[FE_PSE_PORT_CDM4]; q++) airoha_fe_set_pse_oq_rsv(eth, FE_PSE_PORT_CDM4, q, PSE_QUEUE_RSV_PAGES); - /* PPE2 */ - for (q = 0; q < pse_port_num_queues[FE_PSE_PORT_PPE2]; q++) { - if (q < pse_port_num_queues[FE_PSE_PORT_PPE2] / 2) - airoha_fe_set_pse_oq_rsv(eth, FE_PSE_PORT_PPE2, q, - PSE_QUEUE_RSV_PAGES); - else - airoha_fe_set_pse_oq_rsv(eth, FE_PSE_PORT_PPE2, q, 0); + if (airoha_ppe_is_enabled(eth, 1)) { + /* PPE2 */ + for (q = 0; q < pse_port_num_queues[FE_PSE_PORT_PPE2]; q++) { + if (q < pse_port_num_queues[FE_PSE_PORT_PPE2] / 2) + airoha_fe_set_pse_oq_rsv(eth, FE_PSE_PORT_PPE2, + q, + PSE_QUEUE_RSV_PAGES); + else + airoha_fe_set_pse_oq_rsv(eth, FE_PSE_PORT_PPE2, + q, 0); + } } /* GMD4 */ for (q = 0; q < pse_port_num_queues[FE_PSE_PORT_GDM4]; q++) @@ -1762,8 +1769,11 @@ static int airoha_dev_init(struct net_device *dev) airhoha_set_gdm2_loopback(port); fallthrough; case 2: - pse_port = FE_PSE_PORT_PPE2; - break; + if (airoha_ppe_is_enabled(eth, 1)) { + pse_port = FE_PSE_PORT_PPE2; + break; + } + fallthrough; default: pse_port = FE_PSE_PORT_PPE1; break; diff --git a/drivers/net/ethernet/airoha/airoha_eth.h b/drivers/net/ethernet/airoha/airoha_eth.h index cb7e198e40ee..81b1e5f273df 100644 --- a/drivers/net/ethernet/airoha/airoha_eth.h +++ b/drivers/net/ethernet/airoha/airoha_eth.h @@ -627,6 +627,7 @@ static inline bool airoha_is_7581(struct airoha_eth *eth) bool airoha_is_valid_gdm_port(struct airoha_eth *eth, struct airoha_gdm_port *port); +bool airoha_ppe_is_enabled(struct airoha_eth *eth, int index); void airoha_ppe_check_skb(struct airoha_ppe_dev *dev, struct sk_buff *skb, u16 hash, bool rx_wlan); int airoha_ppe_setup_tc_block_cb(struct airoha_ppe_dev *dev, void *type_data); diff --git a/drivers/net/ethernet/airoha/airoha_ppe.c b/drivers/net/ethernet/airoha/airoha_ppe.c index 22ecece0e33e..505a3005f7db 100644 --- a/drivers/net/ethernet/airoha/airoha_ppe.c +++ b/drivers/net/ethernet/airoha/airoha_ppe.c @@ -50,9 +50,12 @@ static int airoha_ppe_get_total_num_stats_entries(struct airoha_ppe *ppe) return num_stats; } -static bool airoha_ppe2_is_enabled(struct airoha_eth *eth) +bool airoha_ppe_is_enabled(struct airoha_eth *eth, int index) { - return airoha_fe_rr(eth, REG_PPE_GLO_CFG(1)) & PPE_GLO_CFG_EN_MASK; + if (index >= eth->soc->num_ppe) + return false; + + return airoha_fe_rr(eth, REG_PPE_GLO_CFG(index)) & PPE_GLO_CFG_EN_MASK; } static u32 airoha_ppe_get_timestamp(struct airoha_ppe *ppe) @@ -120,7 +123,7 @@ static void airoha_ppe_hw_init(struct airoha_ppe *ppe) AIROHA_MAX_MTU)); } - if (airoha_ppe2_is_enabled(eth)) { + if (airoha_ppe_is_enabled(eth, 1)) { sram_num_entries = PPE1_SRAM_NUM_ENTRIES; sram_num_stats_entries = airoha_ppe_get_num_stats_entries(ppe); @@ -518,7 +521,7 @@ static int airoha_ppe_foe_get_flow_stats_index(struct airoha_ppe *ppe, return ppe_num_stats_entries; *index = hash; - if (airoha_ppe2_is_enabled(ppe->eth) && + if (airoha_ppe_is_enabled(ppe->eth, 1) && hash >= ppe_num_stats_entries) *index = *index - PPE_STATS_NUM_ENTRIES; @@ -613,7 +616,7 @@ airoha_ppe_foe_get_entry_locked(struct airoha_ppe *ppe, u32 hash) u32 val; int i; - ppe2 = airoha_ppe2_is_enabled(ppe->eth) && + ppe2 = airoha_ppe_is_enabled(ppe->eth, 1) && hash >= PPE1_SRAM_NUM_ENTRIES; airoha_fe_wr(ppe->eth, REG_PPE_RAM_CTRL(ppe2), FIELD_PREP(PPE_SRAM_CTRL_ENTRY_MASK, hash) | @@ -691,7 +694,7 @@ static int airoha_ppe_foe_commit_entry(struct airoha_ppe *ppe, if (hash < PPE_SRAM_NUM_ENTRIES) { dma_addr_t addr = ppe->foe_dma + hash * sizeof(*hwe); - bool ppe2 = airoha_ppe2_is_enabled(eth) && + bool ppe2 = airoha_ppe_is_enabled(eth, 1) && hash >= PPE1_SRAM_NUM_ENTRIES; err = npu->ops.ppe_foe_commit_entry(npu, addr, sizeof(*hwe), @@ -1286,7 +1289,7 @@ static int airoha_ppe_flush_sram_entries(struct airoha_ppe *ppe, int i, sram_num_entries = PPE_SRAM_NUM_ENTRIES; struct airoha_foe_entry *hwe = ppe->foe; - if (airoha_ppe2_is_enabled(ppe->eth)) + if (airoha_ppe_is_enabled(ppe->eth, 1)) sram_num_entries = sram_num_entries / 2; for (i = 0; i < sram_num_entries; i++) From 5bd1d1fd48ea9f8300b211540d946899c7f96480 Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Fri, 17 Oct 2025 11:06:15 +0200 Subject: [PATCH 0166/1381] net: airoha: ppe: Move PPE memory info in airoha_eth_soc_data struct AN7583 SoC runs a single PPE device while EN7581 runs two of them. Moreover PPE SRAM in AN7583 SoC is reduced to 8K (while SRAM is 16K on EN7581). Take into account PPE memory layout during PPE configuration. Reviewed-by: Simon Horman Signed-off-by: Lorenzo Bianconi Link: https://patch.msgid.link/20251017-an7583-eth-support-v3-6-f28319666667@kernel.org Signed-off-by: Paolo Abeni --- drivers/net/ethernet/airoha/airoha_eth.h | 10 +- drivers/net/ethernet/airoha/airoha_ppe.c | 133 +++++++++--------- .../net/ethernet/airoha/airoha_ppe_debugfs.c | 3 +- 3 files changed, 70 insertions(+), 76 deletions(-) diff --git a/drivers/net/ethernet/airoha/airoha_eth.h b/drivers/net/ethernet/airoha/airoha_eth.h index 81b1e5f273df..df168d798699 100644 --- a/drivers/net/ethernet/airoha/airoha_eth.h +++ b/drivers/net/ethernet/airoha/airoha_eth.h @@ -47,14 +47,9 @@ #define QDMA_METER_IDX(_n) ((_n) & 0xff) #define QDMA_METER_GROUP(_n) (((_n) >> 8) & 0x3) -#define PPE_NUM 2 -#define PPE1_SRAM_NUM_ENTRIES (8 * 1024) -#define PPE_SRAM_NUM_ENTRIES (PPE_NUM * PPE1_SRAM_NUM_ENTRIES) -#define PPE1_STATS_NUM_ENTRIES (4 * 1024) -#define PPE_STATS_NUM_ENTRIES (PPE_NUM * PPE1_STATS_NUM_ENTRIES) +#define PPE_SRAM_NUM_ENTRIES (8 * 1024) +#define PPE_STATS_NUM_ENTRIES (4 * 1024) #define PPE_DRAM_NUM_ENTRIES (16 * 1024) -#define PPE_NUM_ENTRIES (PPE_SRAM_NUM_ENTRIES + PPE_DRAM_NUM_ENTRIES) -#define PPE_HASH_MASK (PPE_NUM_ENTRIES - 1) #define PPE_ENTRY_SIZE 80 #define PPE_RAM_NUM_ENTRIES_SHIFT(_n) (__ffs((_n) >> 10)) @@ -634,6 +629,7 @@ int airoha_ppe_setup_tc_block_cb(struct airoha_ppe_dev *dev, void *type_data); int airoha_ppe_init(struct airoha_eth *eth); void airoha_ppe_deinit(struct airoha_eth *eth); void airoha_ppe_init_upd_mem(struct airoha_gdm_port *port); +u32 airoha_ppe_get_total_num_entries(struct airoha_ppe *ppe); struct airoha_foe_entry *airoha_ppe_foe_get_entry(struct airoha_ppe *ppe, u32 hash); void airoha_ppe_foe_entry_get_stats(struct airoha_ppe *ppe, u32 hash, diff --git a/drivers/net/ethernet/airoha/airoha_ppe.c b/drivers/net/ethernet/airoha/airoha_ppe.c index 505a3005f7db..d142660e7910 100644 --- a/drivers/net/ethernet/airoha/airoha_ppe.c +++ b/drivers/net/ethernet/airoha/airoha_ppe.c @@ -37,19 +37,36 @@ static int airoha_ppe_get_num_stats_entries(struct airoha_ppe *ppe) if (!IS_ENABLED(CONFIG_NET_AIROHA_FLOW_STATS)) return -EOPNOTSUPP; - return PPE1_STATS_NUM_ENTRIES; + return PPE_STATS_NUM_ENTRIES; } static int airoha_ppe_get_total_num_stats_entries(struct airoha_ppe *ppe) { int num_stats = airoha_ppe_get_num_stats_entries(ppe); - if (num_stats > 0) - num_stats = num_stats * PPE_NUM; + if (num_stats > 0) { + struct airoha_eth *eth = ppe->eth; + + num_stats = num_stats * eth->soc->num_ppe; + } return num_stats; } +static u32 airoha_ppe_get_total_sram_num_entries(struct airoha_ppe *ppe) +{ + struct airoha_eth *eth = ppe->eth; + + return PPE_SRAM_NUM_ENTRIES * eth->soc->num_ppe; +} + +u32 airoha_ppe_get_total_num_entries(struct airoha_ppe *ppe) +{ + u32 sram_num_entries = airoha_ppe_get_total_sram_num_entries(ppe); + + return sram_num_entries + PPE_DRAM_NUM_ENTRIES; +} + bool airoha_ppe_is_enabled(struct airoha_eth *eth, int index) { if (index >= eth->soc->num_ppe) @@ -67,14 +84,22 @@ static u32 airoha_ppe_get_timestamp(struct airoha_ppe *ppe) static void airoha_ppe_hw_init(struct airoha_ppe *ppe) { - u32 sram_tb_size, sram_num_entries, dram_num_entries; + u32 sram_ppe_num_data_entries = PPE_SRAM_NUM_ENTRIES, sram_num_entries; + u32 sram_tb_size, dram_num_entries; struct airoha_eth *eth = ppe->eth; int i, sram_num_stats_entries; - sram_tb_size = PPE_SRAM_NUM_ENTRIES * sizeof(struct airoha_foe_entry); + sram_num_entries = airoha_ppe_get_total_sram_num_entries(ppe); + sram_tb_size = sram_num_entries * sizeof(struct airoha_foe_entry); dram_num_entries = PPE_RAM_NUM_ENTRIES_SHIFT(PPE_DRAM_NUM_ENTRIES); - for (i = 0; i < PPE_NUM; i++) { + sram_num_stats_entries = airoha_ppe_get_num_stats_entries(ppe); + if (sram_num_stats_entries > 0) + sram_ppe_num_data_entries -= sram_num_stats_entries; + sram_ppe_num_data_entries = + PPE_RAM_NUM_ENTRIES_SHIFT(sram_ppe_num_data_entries); + + for (i = 0; i < eth->soc->num_ppe; i++) { int p; airoha_fe_wr(eth, REG_PPE_TB_BASE(i), @@ -106,10 +131,16 @@ static void airoha_ppe_hw_init(struct airoha_ppe *ppe) airoha_fe_rmw(eth, REG_PPE_TB_CFG(i), PPE_TB_CFG_SEARCH_MISS_MASK | + PPE_SRAM_TB_NUM_ENTRY_MASK | + PPE_DRAM_TB_NUM_ENTRY_MASK | PPE_TB_CFG_KEEPALIVE_MASK | PPE_TB_ENTRY_SIZE_MASK, FIELD_PREP(PPE_TB_CFG_SEARCH_MISS_MASK, 3) | - FIELD_PREP(PPE_TB_ENTRY_SIZE_MASK, 0)); + FIELD_PREP(PPE_TB_ENTRY_SIZE_MASK, 0) | + FIELD_PREP(PPE_SRAM_TB_NUM_ENTRY_MASK, + sram_ppe_num_data_entries) | + FIELD_PREP(PPE_DRAM_TB_NUM_ENTRY_MASK, + dram_num_entries)); airoha_fe_wr(eth, REG_PPE_HASH_SEED(i), PPE_HASH_SEED); @@ -122,45 +153,6 @@ static void airoha_ppe_hw_init(struct airoha_ppe *ppe) FIELD_PREP(FP1_EGRESS_MTU_MASK, AIROHA_MAX_MTU)); } - - if (airoha_ppe_is_enabled(eth, 1)) { - sram_num_entries = PPE1_SRAM_NUM_ENTRIES; - sram_num_stats_entries = - airoha_ppe_get_num_stats_entries(ppe); - if (sram_num_stats_entries > 0) - sram_num_entries -= sram_num_stats_entries; - sram_num_entries = PPE_RAM_NUM_ENTRIES_SHIFT(sram_num_entries); - - airoha_fe_rmw(eth, REG_PPE_TB_CFG(0), - PPE_SRAM_TB_NUM_ENTRY_MASK | - PPE_DRAM_TB_NUM_ENTRY_MASK, - FIELD_PREP(PPE_SRAM_TB_NUM_ENTRY_MASK, - sram_num_entries) | - FIELD_PREP(PPE_DRAM_TB_NUM_ENTRY_MASK, - dram_num_entries)); - airoha_fe_rmw(eth, REG_PPE_TB_CFG(1), - PPE_SRAM_TB_NUM_ENTRY_MASK | - PPE_DRAM_TB_NUM_ENTRY_MASK, - FIELD_PREP(PPE_SRAM_TB_NUM_ENTRY_MASK, - sram_num_entries) | - FIELD_PREP(PPE_DRAM_TB_NUM_ENTRY_MASK, - dram_num_entries)); - } else { - sram_num_entries = PPE_SRAM_NUM_ENTRIES; - sram_num_stats_entries = - airoha_ppe_get_total_num_stats_entries(ppe); - if (sram_num_stats_entries > 0) - sram_num_entries -= sram_num_stats_entries; - sram_num_entries = PPE_RAM_NUM_ENTRIES_SHIFT(sram_num_entries); - - airoha_fe_rmw(eth, REG_PPE_TB_CFG(0), - PPE_SRAM_TB_NUM_ENTRY_MASK | - PPE_DRAM_TB_NUM_ENTRY_MASK, - FIELD_PREP(PPE_SRAM_TB_NUM_ENTRY_MASK, - sram_num_entries) | - FIELD_PREP(PPE_DRAM_TB_NUM_ENTRY_MASK, - dram_num_entries)); - } } static void airoha_ppe_flow_mangle_eth(const struct flow_action_entry *act, void *eth) @@ -459,9 +451,11 @@ static int airoha_ppe_foe_entry_set_ipv6_tuple(struct airoha_foe_entry *hwe, return 0; } -static u32 airoha_ppe_foe_get_entry_hash(struct airoha_foe_entry *hwe) +static u32 airoha_ppe_foe_get_entry_hash(struct airoha_ppe *ppe, + struct airoha_foe_entry *hwe) { int type = FIELD_GET(AIROHA_FOE_IB1_BIND_PACKET_TYPE, hwe->ib1); + u32 ppe_hash_mask = airoha_ppe_get_total_num_entries(ppe) - 1; u32 hash, hv1, hv2, hv3; switch (type) { @@ -499,14 +493,14 @@ static u32 airoha_ppe_foe_get_entry_hash(struct airoha_foe_entry *hwe) case PPE_PKT_TYPE_IPV6_6RD: default: WARN_ON_ONCE(1); - return PPE_HASH_MASK; + return ppe_hash_mask; } hash = (hv1 & hv2) | ((~hv1) & hv3); hash = (hash >> 24) | ((hash & 0xffffff) << 8); hash ^= hv1 ^ hv2 ^ hv3; hash ^= hash >> 16; - hash &= PPE_NUM_ENTRIES - 1; + hash &= ppe_hash_mask; return hash; } @@ -607,9 +601,11 @@ static void airoha_ppe_foe_flow_stats_update(struct airoha_ppe *ppe, static struct airoha_foe_entry * airoha_ppe_foe_get_entry_locked(struct airoha_ppe *ppe, u32 hash) { + u32 sram_num_entries = airoha_ppe_get_total_sram_num_entries(ppe); + lockdep_assert_held(&ppe_lock); - if (hash < PPE_SRAM_NUM_ENTRIES) { + if (hash < sram_num_entries) { u32 *hwe = ppe->foe + hash * sizeof(struct airoha_foe_entry); struct airoha_eth *eth = ppe->eth; bool ppe2; @@ -617,7 +613,7 @@ airoha_ppe_foe_get_entry_locked(struct airoha_ppe *ppe, u32 hash) int i; ppe2 = airoha_ppe_is_enabled(ppe->eth, 1) && - hash >= PPE1_SRAM_NUM_ENTRIES; + hash >= PPE_SRAM_NUM_ENTRIES; airoha_fe_wr(ppe->eth, REG_PPE_RAM_CTRL(ppe2), FIELD_PREP(PPE_SRAM_CTRL_ENTRY_MASK, hash) | PPE_SRAM_CTRL_REQ_MASK); @@ -668,6 +664,7 @@ static int airoha_ppe_foe_commit_entry(struct airoha_ppe *ppe, struct airoha_foe_entry *e, u32 hash, bool rx_wlan) { + u32 sram_num_entries = airoha_ppe_get_total_sram_num_entries(ppe); struct airoha_foe_entry *hwe = ppe->foe + hash * sizeof(*hwe); u32 ts = airoha_ppe_get_timestamp(ppe); struct airoha_eth *eth = ppe->eth; @@ -692,10 +689,10 @@ static int airoha_ppe_foe_commit_entry(struct airoha_ppe *ppe, if (!rx_wlan) airoha_ppe_foe_flow_stats_update(ppe, npu, hwe, hash); - if (hash < PPE_SRAM_NUM_ENTRIES) { + if (hash < sram_num_entries) { dma_addr_t addr = ppe->foe_dma + hash * sizeof(*hwe); bool ppe2 = airoha_ppe_is_enabled(eth, 1) && - hash >= PPE1_SRAM_NUM_ENTRIES; + hash >= PPE_SRAM_NUM_ENTRIES; err = npu->ops.ppe_foe_commit_entry(npu, addr, sizeof(*hwe), hash, ppe2); @@ -822,7 +819,7 @@ static void airoha_ppe_foe_insert_entry(struct airoha_ppe *ppe, if (state == AIROHA_FOE_STATE_BIND) goto unlock; - index = airoha_ppe_foe_get_entry_hash(hwe); + index = airoha_ppe_foe_get_entry_hash(ppe, hwe); hlist_for_each_entry_safe(e, n, &ppe->foe_flow[index], list) { if (e->type == FLOW_TYPE_L2_SUBFLOW) { state = FIELD_GET(AIROHA_FOE_IB1_BIND_STATE, hwe->ib1); @@ -882,7 +879,7 @@ static int airoha_ppe_foe_flow_commit_entry(struct airoha_ppe *ppe, if (type == PPE_PKT_TYPE_BRIDGE) return airoha_ppe_foe_l2_flow_commit_entry(ppe, e); - hash = airoha_ppe_foe_get_entry_hash(&e->data); + hash = airoha_ppe_foe_get_entry_hash(ppe, &e->data); e->type = FLOW_TYPE_L4; e->hash = 0xffff; @@ -1286,17 +1283,15 @@ static int airoha_ppe_flow_offload_cmd(struct airoha_eth *eth, static int airoha_ppe_flush_sram_entries(struct airoha_ppe *ppe, struct airoha_npu *npu) { - int i, sram_num_entries = PPE_SRAM_NUM_ENTRIES; + u32 sram_num_entries = airoha_ppe_get_total_sram_num_entries(ppe); struct airoha_foe_entry *hwe = ppe->foe; + int i; - if (airoha_ppe_is_enabled(ppe->eth, 1)) - sram_num_entries = sram_num_entries / 2; - - for (i = 0; i < sram_num_entries; i++) + for (i = 0; i < PPE_SRAM_NUM_ENTRIES; i++) memset(&hwe[i], 0, sizeof(*hwe)); return npu->ops.ppe_flush_sram_entries(npu, ppe->foe_dma, - PPE_SRAM_NUM_ENTRIES); + sram_num_entries); } static struct airoha_npu *airoha_ppe_npu_get(struct airoha_eth *eth) @@ -1372,9 +1367,10 @@ void airoha_ppe_check_skb(struct airoha_ppe_dev *dev, struct sk_buff *skb, u16 hash, bool rx_wlan) { struct airoha_ppe *ppe = dev->priv; + u32 ppe_hash_mask = airoha_ppe_get_total_num_entries(ppe) - 1; u16 now, diff; - if (hash > PPE_HASH_MASK) + if (hash > ppe_hash_mask) return; now = (u16)jiffies; @@ -1465,6 +1461,7 @@ EXPORT_SYMBOL_GPL(airoha_ppe_put_dev); int airoha_ppe_init(struct airoha_eth *eth) { int foe_size, err, ppe_num_stats_entries; + u32 ppe_num_entries; struct airoha_ppe *ppe; ppe = devm_kzalloc(eth->dev, sizeof(*ppe), GFP_KERNEL); @@ -1474,18 +1471,18 @@ int airoha_ppe_init(struct airoha_eth *eth) ppe->dev.ops.setup_tc_block_cb = airoha_ppe_setup_tc_block_cb; ppe->dev.ops.check_skb = airoha_ppe_check_skb; ppe->dev.priv = ppe; + ppe->eth = eth; + eth->ppe = ppe; - foe_size = PPE_NUM_ENTRIES * sizeof(struct airoha_foe_entry); + ppe_num_entries = airoha_ppe_get_total_num_entries(ppe); + foe_size = ppe_num_entries * sizeof(struct airoha_foe_entry); ppe->foe = dmam_alloc_coherent(eth->dev, foe_size, &ppe->foe_dma, GFP_KERNEL); if (!ppe->foe) return -ENOMEM; - ppe->eth = eth; - eth->ppe = ppe; - ppe->foe_flow = devm_kzalloc(eth->dev, - PPE_NUM_ENTRIES * sizeof(*ppe->foe_flow), + ppe_num_entries * sizeof(*ppe->foe_flow), GFP_KERNEL); if (!ppe->foe_flow) return -ENOMEM; @@ -1500,7 +1497,7 @@ int airoha_ppe_init(struct airoha_eth *eth) return -ENOMEM; } - ppe->foe_check_time = devm_kzalloc(eth->dev, PPE_NUM_ENTRIES, + ppe->foe_check_time = devm_kzalloc(eth->dev, ppe_num_entries, GFP_KERNEL); if (!ppe->foe_check_time) return -ENOMEM; diff --git a/drivers/net/ethernet/airoha/airoha_ppe_debugfs.c b/drivers/net/ethernet/airoha/airoha_ppe_debugfs.c index 05a756233f6a..0112c41150bb 100644 --- a/drivers/net/ethernet/airoha/airoha_ppe_debugfs.c +++ b/drivers/net/ethernet/airoha/airoha_ppe_debugfs.c @@ -53,9 +53,10 @@ static int airoha_ppe_debugfs_foe_show(struct seq_file *m, void *private, [AIROHA_FOE_STATE_FIN] = "FIN", }; struct airoha_ppe *ppe = m->private; + u32 ppe_num_entries = airoha_ppe_get_total_num_entries(ppe); int i; - for (i = 0; i < PPE_NUM_ENTRIES; i++) { + for (i = 0; i < ppe_num_entries; i++) { const char *state_str, *type_str = "UNKNOWN"; void *src_addr = NULL, *dest_addr = NULL; u16 *src_port = NULL, *dest_port = NULL; From 41139125f5c70e0f66f0cc4ac1b3a62f5801ab42 Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Fri, 17 Oct 2025 11:06:16 +0200 Subject: [PATCH 0167/1381] net: airoha: ppe: Remove airoha_ppe_is_enabled() where not necessary Now each PPE has always PPE_STATS_NUM_ENTRIES entries so we do not need to run airoha_ppe_is_enabled routine to check if the hash refers to PPE1 or PPE2. Reviewed-by: Simon Horman Signed-off-by: Lorenzo Bianconi Link: https://patch.msgid.link/20251017-an7583-eth-support-v3-7-f28319666667@kernel.org Signed-off-by: Paolo Abeni --- drivers/net/ethernet/airoha/airoha_ppe.c | 13 ++++--------- 1 file changed, 4 insertions(+), 9 deletions(-) diff --git a/drivers/net/ethernet/airoha/airoha_ppe.c b/drivers/net/ethernet/airoha/airoha_ppe.c index d142660e7910..195d97e61197 100644 --- a/drivers/net/ethernet/airoha/airoha_ppe.c +++ b/drivers/net/ethernet/airoha/airoha_ppe.c @@ -514,10 +514,8 @@ static int airoha_ppe_foe_get_flow_stats_index(struct airoha_ppe *ppe, if (ppe_num_stats_entries < 0) return ppe_num_stats_entries; - *index = hash; - if (airoha_ppe_is_enabled(ppe->eth, 1) && - hash >= ppe_num_stats_entries) - *index = *index - PPE_STATS_NUM_ENTRIES; + *index = hash >= ppe_num_stats_entries ? hash - PPE_STATS_NUM_ENTRIES + : hash; return 0; } @@ -607,13 +605,11 @@ airoha_ppe_foe_get_entry_locked(struct airoha_ppe *ppe, u32 hash) if (hash < sram_num_entries) { u32 *hwe = ppe->foe + hash * sizeof(struct airoha_foe_entry); + bool ppe2 = hash >= PPE_SRAM_NUM_ENTRIES; struct airoha_eth *eth = ppe->eth; - bool ppe2; u32 val; int i; - ppe2 = airoha_ppe_is_enabled(ppe->eth, 1) && - hash >= PPE_SRAM_NUM_ENTRIES; airoha_fe_wr(ppe->eth, REG_PPE_RAM_CTRL(ppe2), FIELD_PREP(PPE_SRAM_CTRL_ENTRY_MASK, hash) | PPE_SRAM_CTRL_REQ_MASK); @@ -691,8 +687,7 @@ static int airoha_ppe_foe_commit_entry(struct airoha_ppe *ppe, if (hash < sram_num_entries) { dma_addr_t addr = ppe->foe_dma + hash * sizeof(*hwe); - bool ppe2 = airoha_ppe_is_enabled(eth, 1) && - hash >= PPE_SRAM_NUM_ENTRIES; + bool ppe2 = hash >= PPE_SRAM_NUM_ENTRIES; err = npu->ops.ppe_foe_commit_entry(npu, addr, sizeof(*hwe), hash, ppe2); From 306b78f5035af4bd011753c5a6b12812515caa6c Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Fri, 17 Oct 2025 11:06:17 +0200 Subject: [PATCH 0168/1381] net: airoha: ppe: Configure SRAM PPE entries via the cpu Introduce airoha_ppe_foe_commit_sram_entry routine in order to configure the SRAM PPE entries directly via the CPU instead of using the NPU APIs. This is a preliminary patch to enable netfilter flowtable hw offload for AN7583 SoC. Reviewed-by: Simon Horman Signed-off-by: Lorenzo Bianconi Link: https://patch.msgid.link/20251017-an7583-eth-support-v3-8-f28319666667@kernel.org Signed-off-by: Paolo Abeni --- drivers/net/ethernet/airoha/airoha_ppe.c | 30 ++++++++++++++++++------ 1 file changed, 23 insertions(+), 7 deletions(-) diff --git a/drivers/net/ethernet/airoha/airoha_ppe.c b/drivers/net/ethernet/airoha/airoha_ppe.c index 195d97e61197..46755bc60a8e 100644 --- a/drivers/net/ethernet/airoha/airoha_ppe.c +++ b/drivers/net/ethernet/airoha/airoha_ppe.c @@ -656,6 +656,27 @@ static bool airoha_ppe_foe_compare_entry(struct airoha_flow_table_entry *e, return !memcmp(&e->data.d, &hwe->d, len - sizeof(hwe->ib1)); } +static int airoha_ppe_foe_commit_sram_entry(struct airoha_ppe *ppe, u32 hash) +{ + struct airoha_foe_entry *hwe = ppe->foe + hash * sizeof(*hwe); + bool ppe2 = hash >= PPE_SRAM_NUM_ENTRIES; + u32 *ptr = (u32 *)hwe, val; + int i; + + for (i = 0; i < sizeof(*hwe) / sizeof(*ptr); i++) + airoha_fe_wr(ppe->eth, REG_PPE_RAM_ENTRY(ppe2, i), ptr[i]); + + wmb(); + airoha_fe_wr(ppe->eth, REG_PPE_RAM_CTRL(ppe2), + FIELD_PREP(PPE_SRAM_CTRL_ENTRY_MASK, hash) | + PPE_SRAM_CTRL_WR_MASK | PPE_SRAM_CTRL_REQ_MASK); + + return read_poll_timeout_atomic(airoha_fe_rr, val, + val & PPE_SRAM_CTRL_ACK_MASK, + 10, 100, false, ppe->eth, + REG_PPE_RAM_CTRL(ppe2)); +} + static int airoha_ppe_foe_commit_entry(struct airoha_ppe *ppe, struct airoha_foe_entry *e, u32 hash, bool rx_wlan) @@ -685,13 +706,8 @@ static int airoha_ppe_foe_commit_entry(struct airoha_ppe *ppe, if (!rx_wlan) airoha_ppe_foe_flow_stats_update(ppe, npu, hwe, hash); - if (hash < sram_num_entries) { - dma_addr_t addr = ppe->foe_dma + hash * sizeof(*hwe); - bool ppe2 = hash >= PPE_SRAM_NUM_ENTRIES; - - err = npu->ops.ppe_foe_commit_entry(npu, addr, sizeof(*hwe), - hash, ppe2); - } + if (hash < sram_num_entries) + err = airoha_ppe_foe_commit_sram_entry(ppe, hash); unlock: rcu_read_unlock(); From 620d7b91aadbd4eb930895b07e34f0a155a9d3c1 Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Fri, 17 Oct 2025 11:06:18 +0200 Subject: [PATCH 0169/1381] net: airoha: ppe: Flush PPE SRAM table during PPE setup Rely on airoha_ppe_foe_commit_sram_entry routine to flush SRAM PPE table entries. This patch allow moving PPE SRAM flush during PPE setup and avoid dumping uninitialized values via the debugfs if no entries are offloaded yet. Reviewed-by: Simon Horman Signed-off-by: Lorenzo Bianconi Link: https://patch.msgid.link/20251017-an7583-eth-support-v3-9-f28319666667@kernel.org Signed-off-by: Paolo Abeni --- drivers/net/ethernet/airoha/airoha_ppe.c | 24 ++++++++++++++---------- 1 file changed, 14 insertions(+), 10 deletions(-) diff --git a/drivers/net/ethernet/airoha/airoha_ppe.c b/drivers/net/ethernet/airoha/airoha_ppe.c index 46755bc60a8e..4b038673cefe 100644 --- a/drivers/net/ethernet/airoha/airoha_ppe.c +++ b/drivers/net/ethernet/airoha/airoha_ppe.c @@ -1291,18 +1291,22 @@ static int airoha_ppe_flow_offload_cmd(struct airoha_eth *eth, return -EOPNOTSUPP; } -static int airoha_ppe_flush_sram_entries(struct airoha_ppe *ppe, - struct airoha_npu *npu) +static int airoha_ppe_flush_sram_entries(struct airoha_ppe *ppe) { u32 sram_num_entries = airoha_ppe_get_total_sram_num_entries(ppe); struct airoha_foe_entry *hwe = ppe->foe; - int i; + int i, err = 0; + + for (i = 0; i < sram_num_entries; i++) { + int err; - for (i = 0; i < PPE_SRAM_NUM_ENTRIES; i++) memset(&hwe[i], 0, sizeof(*hwe)); + err = airoha_ppe_foe_commit_sram_entry(ppe, i); + if (err) + break; + } - return npu->ops.ppe_flush_sram_entries(npu, ppe->foe_dma, - sram_num_entries); + return err; } static struct airoha_npu *airoha_ppe_npu_get(struct airoha_eth *eth) @@ -1339,10 +1343,6 @@ static int airoha_ppe_offload_setup(struct airoha_eth *eth) } airoha_ppe_hw_init(ppe); - err = airoha_ppe_flush_sram_entries(ppe, npu); - if (err) - goto error_npu_put; - airoha_ppe_foe_flow_stats_reset(ppe, npu); rcu_assign_pointer(eth->npu, npu); @@ -1513,6 +1513,10 @@ int airoha_ppe_init(struct airoha_eth *eth) if (!ppe->foe_check_time) return -ENOMEM; + err = airoha_ppe_flush_sram_entries(ppe); + if (err) + return err; + err = rhashtable_init(ð->flow_table, &airoha_flow_table_params); if (err) return err; From c71a7a861ef02aa2bebb18c2f3385aa3f19094e0 Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Fri, 17 Oct 2025 11:06:19 +0200 Subject: [PATCH 0170/1381] net: airoha: Select default ppe cpu port in airoha_dev_init() Select the PPE default cpu port in airoha_dev_init routine. This patch allows to distribute the load between the two available cpu ports (FE_PSE_PORT_CDM1 and FE_PSE_PORT_CDM2) if the device is running a single PPE module (e.g. 7583) selecting the cpu port based on the use QDMA device. For multi-PPE device (e.g. 7581) assign FE_PSE_PORT_CDM1 to PPE1 and FE_PSE_PORT_CDM2 to PPE2. Reviewed-by: Simon Horman Signed-off-by: Lorenzo Bianconi Link: https://patch.msgid.link/20251017-an7583-eth-support-v3-10-f28319666667@kernel.org Signed-off-by: Paolo Abeni --- drivers/net/ethernet/airoha/airoha_eth.c | 38 ++++++++++-------------- 1 file changed, 16 insertions(+), 22 deletions(-) diff --git a/drivers/net/ethernet/airoha/airoha_eth.c b/drivers/net/ethernet/airoha/airoha_eth.c index dea856ddf242..4e338c126dd3 100644 --- a/drivers/net/ethernet/airoha/airoha_eth.c +++ b/drivers/net/ethernet/airoha/airoha_eth.c @@ -531,25 +531,6 @@ static int airoha_fe_init(struct airoha_eth *eth) /* disable IFC by default */ airoha_fe_clear(eth, REG_FE_CSR_IFC_CFG, FE_IFC_EN_MASK); - airoha_fe_wr(eth, REG_PPE_DFT_CPORT0(0), - FIELD_PREP(DFT_CPORT_MASK(7), FE_PSE_PORT_CDM1) | - FIELD_PREP(DFT_CPORT_MASK(6), FE_PSE_PORT_CDM1) | - FIELD_PREP(DFT_CPORT_MASK(5), FE_PSE_PORT_CDM1) | - FIELD_PREP(DFT_CPORT_MASK(4), FE_PSE_PORT_CDM1) | - FIELD_PREP(DFT_CPORT_MASK(3), FE_PSE_PORT_CDM1) | - FIELD_PREP(DFT_CPORT_MASK(2), FE_PSE_PORT_CDM1) | - FIELD_PREP(DFT_CPORT_MASK(1), FE_PSE_PORT_CDM1) | - FIELD_PREP(DFT_CPORT_MASK(0), FE_PSE_PORT_CDM1)); - airoha_fe_wr(eth, REG_PPE_DFT_CPORT0(1), - FIELD_PREP(DFT_CPORT_MASK(7), FE_PSE_PORT_CDM2) | - FIELD_PREP(DFT_CPORT_MASK(6), FE_PSE_PORT_CDM2) | - FIELD_PREP(DFT_CPORT_MASK(5), FE_PSE_PORT_CDM2) | - FIELD_PREP(DFT_CPORT_MASK(4), FE_PSE_PORT_CDM2) | - FIELD_PREP(DFT_CPORT_MASK(3), FE_PSE_PORT_CDM2) | - FIELD_PREP(DFT_CPORT_MASK(2), FE_PSE_PORT_CDM2) | - FIELD_PREP(DFT_CPORT_MASK(1), FE_PSE_PORT_CDM2) | - FIELD_PREP(DFT_CPORT_MASK(0), FE_PSE_PORT_CDM2)); - /* enable 1:N vlan action, init vlan table */ airoha_fe_set(eth, REG_MC_VLAN_EN, MC_VLAN_EN_MASK); @@ -1756,8 +1737,10 @@ static void airhoha_set_gdm2_loopback(struct airoha_gdm_port *port) static int airoha_dev_init(struct net_device *dev) { struct airoha_gdm_port *port = netdev_priv(dev); - struct airoha_eth *eth = port->qdma->eth; - u32 pse_port; + struct airoha_qdma *qdma = port->qdma; + struct airoha_eth *eth = qdma->eth; + u32 pse_port, fe_cpu_port; + u8 ppe_id; airoha_set_macaddr(port, dev->dev_addr); @@ -1770,16 +1753,27 @@ static int airoha_dev_init(struct net_device *dev) fallthrough; case 2: if (airoha_ppe_is_enabled(eth, 1)) { + /* For PPE2 always use secondary cpu port. */ + fe_cpu_port = FE_PSE_PORT_CDM2; pse_port = FE_PSE_PORT_PPE2; break; } fallthrough; - default: + default: { + u8 qdma_id = qdma - ð->qdma[0]; + + /* For PPE1 select cpu port according to the running QDMA. */ + fe_cpu_port = qdma_id ? FE_PSE_PORT_CDM2 : FE_PSE_PORT_CDM1; pse_port = FE_PSE_PORT_PPE1; break; } + } airoha_set_gdm_port_fwd_cfg(eth, REG_GDM_FWD_CFG(port->id), pse_port); + ppe_id = pse_port == FE_PSE_PORT_PPE2 ? 1 : 0; + airoha_fe_rmw(eth, REG_PPE_DFT_CPORT0(ppe_id), + DFT_CPORT_MASK(port->id), + fe_cpu_port << __ffs(DFT_CPORT_MASK(port->id))); return 0; } From 9d5b5219f672c80bed4d4e15f0068e648cdca43b Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Fri, 17 Oct 2025 11:06:20 +0200 Subject: [PATCH 0171/1381] net: airoha: Refactor src port configuration in airhoha_set_gdm2_loopback AN7583 chipset relies on different definitions for source-port identifier used for hw offloading. In order to support hw offloading in AN7583 controller, refactor src port configuration in airhoha_set_gdm2_loopback routine and introduce get_src_port_id callback. Reviewed-by: Simon Horman Signed-off-by: Lorenzo Bianconi Link: https://patch.msgid.link/20251017-an7583-eth-support-v3-11-f28319666667@kernel.org Signed-off-by: Paolo Abeni --- drivers/net/ethernet/airoha/airoha_eth.c | 82 ++++++++++++++++------- drivers/net/ethernet/airoha/airoha_eth.h | 18 +++-- drivers/net/ethernet/airoha/airoha_regs.h | 6 +- 3 files changed, 73 insertions(+), 33 deletions(-) diff --git a/drivers/net/ethernet/airoha/airoha_eth.c b/drivers/net/ethernet/airoha/airoha_eth.c index 4e338c126dd3..4671f906a68c 100644 --- a/drivers/net/ethernet/airoha/airoha_eth.c +++ b/drivers/net/ethernet/airoha/airoha_eth.c @@ -1682,13 +1682,17 @@ static int airoha_dev_set_macaddr(struct net_device *dev, void *p) return 0; } -static void airhoha_set_gdm2_loopback(struct airoha_gdm_port *port) +static int airhoha_set_gdm2_loopback(struct airoha_gdm_port *port) { - u32 pse_port = port->id == 3 ? FE_PSE_PORT_GDM3 : FE_PSE_PORT_GDM4; + u32 val, pse_port, chan = port->id == AIROHA_GDM3_IDX ? 4 : 0; struct airoha_eth *eth = port->qdma->eth; - u32 chan = port->id == 3 ? 4 : 0; + /* XXX: handle XSI_USB_PORT and XSI_PCE1_PORT */ + u32 nbq = port->id == AIROHA_GDM3_IDX ? 4 : 0; + int src_port; /* Forward the traffic to the proper GDM port */ + pse_port = port->id == AIROHA_GDM3_IDX ? FE_PSE_PORT_GDM3 + : FE_PSE_PORT_GDM4; airoha_set_gdm_port_fwd_cfg(eth, REG_GDM_FWD_CFG(2), pse_port); airoha_fe_clear(eth, REG_GDM_FWD_CFG(2), GDM_STRIP_CRC); @@ -1709,29 +1713,25 @@ static void airhoha_set_gdm2_loopback(struct airoha_gdm_port *port) airoha_fe_clear(eth, REG_FE_VIP_PORT_EN, BIT(2)); airoha_fe_clear(eth, REG_FE_IFC_PORT_EN, BIT(2)); - if (port->id == 3) { - /* FIXME: handle XSI_PCE1_PORT */ - airoha_fe_rmw(eth, REG_FE_WAN_PORT, - WAN1_EN_MASK | WAN1_MASK | WAN0_MASK, - FIELD_PREP(WAN0_MASK, HSGMII_LAN_PCIE0_SRCPORT)); - airoha_fe_rmw(eth, - REG_SP_DFT_CPORT(HSGMII_LAN_PCIE0_SRCPORT >> 3), - SP_CPORT_PCIE0_MASK, - FIELD_PREP(SP_CPORT_PCIE0_MASK, - FE_PSE_PORT_CDM2)); - } else { - /* FIXME: handle XSI_USB_PORT */ + src_port = eth->soc->ops.get_src_port_id(port, nbq); + if (src_port < 0) + return src_port; + + airoha_fe_rmw(eth, REG_FE_WAN_PORT, + WAN1_EN_MASK | WAN1_MASK | WAN0_MASK, + FIELD_PREP(WAN0_MASK, src_port)); + val = src_port & SP_CPORT_DFT_MASK; + airoha_fe_rmw(eth, + REG_SP_DFT_CPORT(src_port >> fls(SP_CPORT_DFT_MASK)), + SP_CPORT_MASK(val), + FE_PSE_PORT_CDM2 << __ffs(SP_CPORT_MASK(val))); + + if (port->id != AIROHA_GDM3_IDX) airoha_fe_rmw(eth, REG_SRC_PORT_FC_MAP6, FC_ID_OF_SRC_PORT24_MASK, FIELD_PREP(FC_ID_OF_SRC_PORT24_MASK, 2)); - airoha_fe_rmw(eth, REG_FE_WAN_PORT, - WAN1_EN_MASK | WAN1_MASK | WAN0_MASK, - FIELD_PREP(WAN0_MASK, HSGMII_LAN_ETH_SRCPORT)); - airoha_fe_rmw(eth, - REG_SP_DFT_CPORT(HSGMII_LAN_ETH_SRCPORT >> 3), - SP_CPORT_ETH_MASK, - FIELD_PREP(SP_CPORT_ETH_MASK, FE_PSE_PORT_CDM2)); - } + + return 0; } static int airoha_dev_init(struct net_device *dev) @@ -1748,8 +1748,13 @@ static int airoha_dev_init(struct net_device *dev) case 3: case 4: /* If GDM2 is active we can't enable loopback */ - if (!eth->ports[1]) - airhoha_set_gdm2_loopback(port); + if (!eth->ports[1]) { + int err; + + err = airhoha_set_gdm2_loopback(port); + if (err) + return err; + } fallthrough; case 2: if (airoha_ppe_is_enabled(eth, 1)) { @@ -3069,11 +3074,38 @@ static const char * const en7581_xsi_rsts_names[] = { "xfp-mac", }; +static int airoha_en7581_get_src_port_id(struct airoha_gdm_port *port, int nbq) +{ + switch (port->id) { + case 3: + /* 7581 SoC supports PCIe serdes on GDM3 port */ + if (nbq == 4) + return HSGMII_LAN_7581_PCIE0_SRCPORT; + if (nbq == 5) + return HSGMII_LAN_7581_PCIE1_SRCPORT; + break; + case 4: + /* 7581 SoC supports eth and usb serdes on GDM4 port */ + if (!nbq) + return HSGMII_LAN_7581_ETH_SRCPORT; + if (nbq == 1) + return HSGMII_LAN_7581_USB_SRCPORT; + break; + default: + break; + } + + return -EINVAL; +} + static const struct airoha_eth_soc_data en7581_soc_data = { .version = 0x7581, .xsi_rsts_names = en7581_xsi_rsts_names, .num_xsi_rsts = ARRAY_SIZE(en7581_xsi_rsts_names), .num_ppe = 2, + .ops = { + .get_src_port_id = airoha_en7581_get_src_port_id, + }, }; static const struct of_device_id of_airoha_match[] = { diff --git a/drivers/net/ethernet/airoha/airoha_eth.h b/drivers/net/ethernet/airoha/airoha_eth.h index df168d798699..e09579da8c78 100644 --- a/drivers/net/ethernet/airoha/airoha_eth.h +++ b/drivers/net/ethernet/airoha/airoha_eth.h @@ -67,10 +67,10 @@ enum { }; enum { - HSGMII_LAN_PCIE0_SRCPORT = 0x16, - HSGMII_LAN_PCIE1_SRCPORT, - HSGMII_LAN_ETH_SRCPORT, - HSGMII_LAN_USB_SRCPORT, + HSGMII_LAN_7581_PCIE0_SRCPORT = 0x16, + HSGMII_LAN_7581_PCIE1_SRCPORT, + HSGMII_LAN_7581_ETH_SRCPORT, + HSGMII_LAN_7581_USB_SRCPORT, }; enum { @@ -99,6 +99,13 @@ enum { CRSN_25 = 0x19, }; +enum airoha_gdm_index { + AIROHA_GDM1_IDX = 1, + AIROHA_GDM2_IDX = 2, + AIROHA_GDM3_IDX = 3, + AIROHA_GDM4_IDX = 4, +}; + enum { FE_PSE_PORT_CDM1, FE_PSE_PORT_GDM1, @@ -555,6 +562,9 @@ struct airoha_eth_soc_data { const char * const *xsi_rsts_names; int num_xsi_rsts; int num_ppe; + struct { + int (*get_src_port_id)(struct airoha_gdm_port *port, int nbq); + } ops; }; struct airoha_eth { diff --git a/drivers/net/ethernet/airoha/airoha_regs.h b/drivers/net/ethernet/airoha/airoha_regs.h index 69c5a143db8c..ebcce00d9bc6 100644 --- a/drivers/net/ethernet/airoha/airoha_regs.h +++ b/drivers/net/ethernet/airoha/airoha_regs.h @@ -383,10 +383,8 @@ #define REG_MC_VLAN_DATA 0x2108 #define REG_SP_DFT_CPORT(_n) (0x20e0 + ((_n) << 2)) -#define SP_CPORT_PCIE1_MASK GENMASK(31, 28) -#define SP_CPORT_PCIE0_MASK GENMASK(27, 24) -#define SP_CPORT_USB_MASK GENMASK(7, 4) -#define SP_CPORT_ETH_MASK GENMASK(7, 4) +#define SP_CPORT_DFT_MASK GENMASK(2, 0) +#define SP_CPORT_MASK(_n) GENMASK(3 + ((_n) << 2), ((_n) << 2)) #define REG_SRC_PORT_FC_MAP6 0x2298 #define FC_ID_OF_SRC_PORT27_MASK GENMASK(28, 24) From 63f283d36b1fb06b55ae609a1f679544f5f66057 Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Fri, 17 Oct 2025 11:06:21 +0200 Subject: [PATCH 0172/1381] net: airoha: ppe: Do not use magic numbers in airoha_ppe_foe_get_entry_locked() Explicit the size of entries pointed by hwe pointer in airoha_ppe_foe_get_entry_locked routine Reviewed-by: Simon Horman Signed-off-by: Lorenzo Bianconi Link: https://patch.msgid.link/20251017-an7583-eth-support-v3-12-f28319666667@kernel.org Signed-off-by: Paolo Abeni --- drivers/net/ethernet/airoha/airoha_ppe.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/airoha/airoha_ppe.c b/drivers/net/ethernet/airoha/airoha_ppe.c index 4b038673cefe..eda95107cd1d 100644 --- a/drivers/net/ethernet/airoha/airoha_ppe.c +++ b/drivers/net/ethernet/airoha/airoha_ppe.c @@ -619,7 +619,8 @@ airoha_ppe_foe_get_entry_locked(struct airoha_ppe *ppe, u32 hash) REG_PPE_RAM_CTRL(ppe2))) return NULL; - for (i = 0; i < sizeof(struct airoha_foe_entry) / 4; i++) + for (i = 0; i < sizeof(struct airoha_foe_entry) / sizeof(*hwe); + i++) hwe[i] = airoha_fe_rr(eth, REG_PPE_RAM_ENTRY(ppe2, i)); } From e4e5ce823bdd4601bd75ae7c206ae35e7c2fa60b Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Fri, 17 Oct 2025 11:06:22 +0200 Subject: [PATCH 0173/1381] net: airoha: Add AN7583 SoC support Introduce support for AN7583 ethernet controller to airoha-eth dirver. Reviewed-by: Simon Horman Signed-off-by: Lorenzo Bianconi Link: https://patch.msgid.link/20251017-an7583-eth-support-v3-13-f28319666667@kernel.org Signed-off-by: Paolo Abeni --- drivers/net/ethernet/airoha/airoha_eth.c | 68 ++++++++++++++++++++++-- drivers/net/ethernet/airoha/airoha_eth.h | 11 ++++ drivers/net/ethernet/airoha/airoha_ppe.c | 3 ++ 3 files changed, 77 insertions(+), 5 deletions(-) diff --git a/drivers/net/ethernet/airoha/airoha_eth.c b/drivers/net/ethernet/airoha/airoha_eth.c index 4671f906a68c..8483ea02603e 100644 --- a/drivers/net/ethernet/airoha/airoha_eth.c +++ b/drivers/net/ethernet/airoha/airoha_eth.c @@ -1684,10 +1684,8 @@ static int airoha_dev_set_macaddr(struct net_device *dev, void *p) static int airhoha_set_gdm2_loopback(struct airoha_gdm_port *port) { - u32 val, pse_port, chan = port->id == AIROHA_GDM3_IDX ? 4 : 0; struct airoha_eth *eth = port->qdma->eth; - /* XXX: handle XSI_USB_PORT and XSI_PCE1_PORT */ - u32 nbq = port->id == AIROHA_GDM3_IDX ? 4 : 0; + u32 val, pse_port, chan, nbq; int src_port; /* Forward the traffic to the proper GDM port */ @@ -1699,6 +1697,8 @@ static int airhoha_set_gdm2_loopback(struct airoha_gdm_port *port) /* Enable GDM2 loopback */ airoha_fe_wr(eth, REG_GDM_TXCHN_EN(2), 0xffffffff); airoha_fe_wr(eth, REG_GDM_RXCHN_EN(2), 0xffff); + + chan = port->id == AIROHA_GDM3_IDX ? airoha_is_7581(eth) ? 4 : 3 : 0; airoha_fe_rmw(eth, REG_GDM_LPBK_CFG(2), LPBK_CHAN_MASK | LPBK_MODE_MASK | LPBK_EN_MASK, FIELD_PREP(LPBK_CHAN_MASK, chan) | @@ -1713,6 +1713,8 @@ static int airhoha_set_gdm2_loopback(struct airoha_gdm_port *port) airoha_fe_clear(eth, REG_FE_VIP_PORT_EN, BIT(2)); airoha_fe_clear(eth, REG_FE_IFC_PORT_EN, BIT(2)); + /* XXX: handle XSI_USB_PORT and XSI_PCE1_PORT */ + nbq = port->id == AIROHA_GDM3_IDX && airoha_is_7581(eth) ? 4 : 0; src_port = eth->soc->ops.get_src_port_id(port, nbq); if (src_port < 0) return src_port; @@ -1726,7 +1728,7 @@ static int airhoha_set_gdm2_loopback(struct airoha_gdm_port *port) SP_CPORT_MASK(val), FE_PSE_PORT_CDM2 << __ffs(SP_CPORT_MASK(val))); - if (port->id != AIROHA_GDM3_IDX) + if (port->id != AIROHA_GDM3_IDX && airoha_is_7581(eth)) airoha_fe_rmw(eth, REG_SRC_PORT_FC_MAP6, FC_ID_OF_SRC_PORT24_MASK, FIELD_PREP(FC_ID_OF_SRC_PORT24_MASK, 2)); @@ -1895,6 +1897,22 @@ static bool airoha_dev_tx_queue_busy(struct airoha_queue *q, u32 nr_frags) return index >= tail; } +static int airoha_get_fe_port(struct airoha_gdm_port *port) +{ + struct airoha_qdma *qdma = port->qdma; + struct airoha_eth *eth = qdma->eth; + + switch (eth->soc->version) { + case 0x7583: + return port->id == AIROHA_GDM3_IDX ? FE_PSE_PORT_GDM3 + : port->id; + case 0x7581: + default: + return port->id == AIROHA_GDM4_IDX ? FE_PSE_PORT_GDM4 + : port->id; + } +} + static netdev_tx_t airoha_dev_xmit(struct sk_buff *skb, struct net_device *dev) { @@ -1935,7 +1953,7 @@ static netdev_tx_t airoha_dev_xmit(struct sk_buff *skb, } } - fport = port->id == 4 ? FE_PSE_PORT_GDM4 : port->id; + fport = airoha_get_fe_port(port); msg1 = FIELD_PREP(QDMA_ETH_TXMSG_FPORT_MASK, fport) | FIELD_PREP(QDMA_ETH_TXMSG_METER_MASK, 0x7f); @@ -3098,6 +3116,35 @@ static int airoha_en7581_get_src_port_id(struct airoha_gdm_port *port, int nbq) return -EINVAL; } +static const char * const an7583_xsi_rsts_names[] = { + "xsi-mac", + "hsi0-mac", + "hsi1-mac", + "xfp-mac", +}; + +static int airoha_an7583_get_src_port_id(struct airoha_gdm_port *port, int nbq) +{ + switch (port->id) { + case 3: + /* 7583 SoC supports eth serdes on GDM3 port */ + if (!nbq) + return HSGMII_LAN_7583_ETH_SRCPORT; + break; + case 4: + /* 7583 SoC supports PCIe and USB serdes on GDM4 port */ + if (!nbq) + return HSGMII_LAN_7583_PCIE_SRCPORT; + if (nbq == 1) + return HSGMII_LAN_7583_USB_SRCPORT; + break; + default: + break; + } + + return -EINVAL; +} + static const struct airoha_eth_soc_data en7581_soc_data = { .version = 0x7581, .xsi_rsts_names = en7581_xsi_rsts_names, @@ -3108,8 +3155,19 @@ static const struct airoha_eth_soc_data en7581_soc_data = { }, }; +static const struct airoha_eth_soc_data an7583_soc_data = { + .version = 0x7583, + .xsi_rsts_names = an7583_xsi_rsts_names, + .num_xsi_rsts = ARRAY_SIZE(an7583_xsi_rsts_names), + .num_ppe = 1, + .ops = { + .get_src_port_id = airoha_an7583_get_src_port_id, + }, +}; + static const struct of_device_id of_airoha_match[] = { { .compatible = "airoha,en7581-eth", .data = &en7581_soc_data }, + { .compatible = "airoha,an7583-eth", .data = &an7583_soc_data }, { /* sentinel */ } }; MODULE_DEVICE_TABLE(of, of_airoha_match); diff --git a/drivers/net/ethernet/airoha/airoha_eth.h b/drivers/net/ethernet/airoha/airoha_eth.h index e09579da8c78..eb27a4ff5198 100644 --- a/drivers/net/ethernet/airoha/airoha_eth.h +++ b/drivers/net/ethernet/airoha/airoha_eth.h @@ -73,6 +73,12 @@ enum { HSGMII_LAN_7581_USB_SRCPORT, }; +enum { + HSGMII_LAN_7583_ETH_SRCPORT = 0x16, + HSGMII_LAN_7583_PCIE_SRCPORT = 0x18, + HSGMII_LAN_7583_USB_SRCPORT, +}; + enum { XSI_PCIE0_VIP_PORT_MASK = BIT(22), XSI_PCIE1_VIP_PORT_MASK = BIT(23), @@ -629,6 +635,11 @@ static inline bool airoha_is_7581(struct airoha_eth *eth) return eth->soc->version == 0x7581; } +static inline bool airoha_is_7583(struct airoha_eth *eth) +{ + return eth->soc->version == 0x7583; +} + bool airoha_is_valid_gdm_port(struct airoha_eth *eth, struct airoha_gdm_port *port); diff --git a/drivers/net/ethernet/airoha/airoha_ppe.c b/drivers/net/ethernet/airoha/airoha_ppe.c index eda95107cd1d..c373f21d95f5 100644 --- a/drivers/net/ethernet/airoha/airoha_ppe.c +++ b/drivers/net/ethernet/airoha/airoha_ppe.c @@ -37,6 +37,9 @@ static int airoha_ppe_get_num_stats_entries(struct airoha_ppe *ppe) if (!IS_ENABLED(CONFIG_NET_AIROHA_FLOW_STATS)) return -EOPNOTSUPP; + if (airoha_is_7583(ppe->eth)) + return -EOPNOTSUPP; + return PPE_STATS_NUM_ENTRIES; } From 3ff9bcecce83f12169ab3e42671bd76554ca521a Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 17 Oct 2025 13:37:12 +0000 Subject: [PATCH 0174/1381] net: avoid extra access to sk->sk_wmem_alloc in sock_wfree() UDP TX packets destructor is sock_wfree(). It suffers from a cache line bouncing in sock_def_write_space_wfree(). Instead of reading sk->sk_wmem_alloc after we just did an atomic RMW on it, use __refcount_sub_and_test() to get the old value for free, and pass the new value to sock_def_write_space_wfree(). Add __sock_writeable() helper. Signed-off-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20251017133712.2842665-1-edumazet@google.com Signed-off-by: Paolo Abeni --- include/net/sock.h | 6 +++++- net/core/sock.c | 14 ++++++++------ 2 files changed, 13 insertions(+), 7 deletions(-) diff --git a/include/net/sock.h b/include/net/sock.h index 5c564f114ae9..01ce231603db 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -2607,12 +2607,16 @@ static inline struct page_frag *sk_page_frag(struct sock *sk) bool sk_page_frag_refill(struct sock *sk, struct page_frag *pfrag); +static inline bool __sock_writeable(const struct sock *sk, int wmem_alloc) +{ + return wmem_alloc < (READ_ONCE(sk->sk_sndbuf) >> 1); +} /* * Default write policy as shown to user space via poll/select/SIGIO */ static inline bool sock_writeable(const struct sock *sk) { - return refcount_read(&sk->sk_wmem_alloc) < (READ_ONCE(sk->sk_sndbuf) >> 1); + return __sock_writeable(sk, refcount_read(&sk->sk_wmem_alloc)); } static inline gfp_t gfp_any(void) diff --git a/net/core/sock.c b/net/core/sock.c index b78533fb9268..a99132cc0965 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -155,7 +155,7 @@ static DEFINE_MUTEX(proto_list_mutex); static LIST_HEAD(proto_list); -static void sock_def_write_space_wfree(struct sock *sk); +static void sock_def_write_space_wfree(struct sock *sk, int wmem_alloc); static void sock_def_write_space(struct sock *sk); /** @@ -2659,16 +2659,18 @@ EXPORT_SYMBOL_GPL(sk_setup_caps); */ void sock_wfree(struct sk_buff *skb) { - struct sock *sk = skb->sk; unsigned int len = skb->truesize; + struct sock *sk = skb->sk; bool free; + int old; if (!sock_flag(sk, SOCK_USE_WRITE_QUEUE)) { if (sock_flag(sk, SOCK_RCU_FREE) && sk->sk_write_space == sock_def_write_space) { rcu_read_lock(); - free = refcount_sub_and_test(len, &sk->sk_wmem_alloc); - sock_def_write_space_wfree(sk); + free = __refcount_sub_and_test(len, &sk->sk_wmem_alloc, + &old); + sock_def_write_space_wfree(sk, old - len); rcu_read_unlock(); if (unlikely(free)) __sk_free(sk); @@ -3612,12 +3614,12 @@ static void sock_def_write_space(struct sock *sk) * for SOCK_RCU_FREE sockets under RCU read section and after putting * ->sk_wmem_alloc. */ -static void sock_def_write_space_wfree(struct sock *sk) +static void sock_def_write_space_wfree(struct sock *sk, int wmem_alloc) { /* Do not wake up a writer until he can make "significant" * progress. --DaveM */ - if (sock_writeable(sk)) { + if (__sock_writeable(sk, wmem_alloc)) { struct socket_wq *wq = rcu_dereference(sk->sk_wq); /* rely on refcount_sub from sock_wfree() */ From 0364ca33097da247ef6539c765405657324f243e Mon Sep 17 00:00:00 2001 From: Alok Tiwari Date: Mon, 20 Oct 2025 10:09:13 -0700 Subject: [PATCH 0175/1381] devlink: region: correct port region lookup to use port_ops The function devlink_port_region_get_by_name() incorrectly uses region->ops->name to compare the region name. as it is not any critical impact as ops and port_ops define as union for devlink_region but as per code logic it should refer port_ops here. No functional impact as ops and port_ops are part of same union, and name is the first member of both. Update it to use region->port_ops->name to properly reference the name of the devlink port region. Signed-off-by: Alok Tiwari Reviewed-by: Jiri Pirko Link: https://patch.msgid.link/20251020170916.1741808-1-alok.a.tiwari@oracle.com Signed-off-by: Jakub Kicinski --- net/devlink/region.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/devlink/region.c b/net/devlink/region.c index 63fb297f6d67..d6e5805cf3a0 100644 --- a/net/devlink/region.c +++ b/net/devlink/region.c @@ -50,7 +50,7 @@ devlink_port_region_get_by_name(struct devlink_port *port, struct devlink_region *region; list_for_each_entry(region, &port->region_list, list) - if (!strcmp(region->ops->name, region_name)) + if (!strcmp(region->port_ops->name, region_name)) return region; return NULL; From 28098defc79fe7d29e6bfe4eb6312991f6bdc3d3 Mon Sep 17 00:00:00 2001 From: Hangbin Liu Date: Fri, 17 Oct 2025 03:41:52 +0000 Subject: [PATCH 0176/1381] net: add a common function to compute features for upper devices Some high level software drivers need to compute features from lower devices. But each has their own implementations and may lost some feature compute. Let's use one common function to compute features for kinds of these devices. The new helper uses the current bond implementation as the reference one, as the latter already handles all the relevant aspects: netdev features, TSO limits and dst retention. Suggested-by: Paolo Abeni Signed-off-by: Hangbin Liu Reviewed-by: Sabrina Dubroca Reviewed-by: Jiri Pirko Link: https://patch.msgid.link/20251017034155.61990-2-liuhangbin@gmail.com Signed-off-by: Jakub Kicinski --- include/linux/netdev_features.h | 18 +++++++ include/linux/netdevice.h | 1 + net/core/dev.c | 88 +++++++++++++++++++++++++++++++++ 3 files changed, 107 insertions(+) diff --git a/include/linux/netdev_features.h b/include/linux/netdev_features.h index 7a01c518e573..93e4da7046a1 100644 --- a/include/linux/netdev_features.h +++ b/include/linux/netdev_features.h @@ -255,6 +255,24 @@ static inline int find_next_netdev_feature(u64 feature, unsigned long start) NETIF_F_GSO_UDP_TUNNEL | \ NETIF_F_GSO_UDP_TUNNEL_CSUM) +/* virtual device features */ +#define MASTER_UPPER_DEV_VLAN_FEATURES (NETIF_F_HW_CSUM | NETIF_F_SG | \ + NETIF_F_FRAGLIST | NETIF_F_GSO_SOFTWARE | \ + NETIF_F_GSO_ENCAP_ALL | \ + NETIF_F_HIGHDMA | NETIF_F_LRO) + +#define MASTER_UPPER_DEV_ENC_FEATURES (NETIF_F_HW_CSUM | NETIF_F_SG | \ + NETIF_F_RXCSUM | NETIF_F_GSO_SOFTWARE | \ + NETIF_F_GSO_PARTIAL) + +#define MASTER_UPPER_DEV_MPLS_FEATURES (NETIF_F_HW_CSUM | NETIF_F_SG | \ + NETIF_F_GSO_SOFTWARE) + +#define MASTER_UPPER_DEV_XFRM_FEATURES (NETIF_F_HW_ESP | NETIF_F_HW_ESP_TX_CSUM | \ + NETIF_F_GSO_ESP) + +#define MASTER_UPPER_DEV_GSO_PARTIAL_FEATURES (NETIF_F_GSO_ESP) + static inline netdev_features_t netdev_base_features(netdev_features_t features) { features &= ~NETIF_F_ONE_FOR_ALL; diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index d1a687444b27..7f5aad5cc9a1 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -5304,6 +5304,7 @@ static inline netdev_features_t netdev_add_tso_features(netdev_features_t featur int __netdev_update_features(struct net_device *dev); void netdev_update_features(struct net_device *dev); void netdev_change_features(struct net_device *dev); +void netdev_compute_master_upper_features(struct net_device *dev, bool update_header); void netif_stacked_transfer_operstate(const struct net_device *rootdev, struct net_device *dev); diff --git a/net/core/dev.c b/net/core/dev.c index 9482b905c66a..378c2d010faf 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -12693,6 +12693,94 @@ netdev_features_t netdev_increment_features(netdev_features_t all, } EXPORT_SYMBOL(netdev_increment_features); +/** + * netdev_compute_master_upper_features - compute feature from lowers + * @dev: the upper device + * @update_header: whether to update upper device's header_len/headroom/tailroom + * + * Recompute the upper device's feature based on all lower devices. + */ +void netdev_compute_master_upper_features(struct net_device *dev, bool update_header) +{ + unsigned int dst_release_flag = IFF_XMIT_DST_RELEASE | IFF_XMIT_DST_RELEASE_PERM; + netdev_features_t gso_partial_features = MASTER_UPPER_DEV_GSO_PARTIAL_FEATURES; + netdev_features_t xfrm_features = MASTER_UPPER_DEV_XFRM_FEATURES; + netdev_features_t mpls_features = MASTER_UPPER_DEV_MPLS_FEATURES; + netdev_features_t vlan_features = MASTER_UPPER_DEV_VLAN_FEATURES; + netdev_features_t enc_features = MASTER_UPPER_DEV_ENC_FEATURES; + unsigned short max_header_len = ETH_HLEN; + unsigned int tso_max_size = TSO_MAX_SIZE; + unsigned short max_headroom = 0; + unsigned short max_tailroom = 0; + u16 tso_max_segs = TSO_MAX_SEGS; + struct net_device *lower_dev; + struct list_head *iter; + + mpls_features = netdev_base_features(mpls_features); + vlan_features = netdev_base_features(vlan_features); + enc_features = netdev_base_features(enc_features); + + netdev_for_each_lower_dev(dev, lower_dev, iter) { + gso_partial_features = netdev_increment_features(gso_partial_features, + lower_dev->gso_partial_features, + MASTER_UPPER_DEV_GSO_PARTIAL_FEATURES); + + vlan_features = netdev_increment_features(vlan_features, + lower_dev->vlan_features, + MASTER_UPPER_DEV_VLAN_FEATURES); + + enc_features = netdev_increment_features(enc_features, + lower_dev->hw_enc_features, + MASTER_UPPER_DEV_ENC_FEATURES); + + if (IS_ENABLED(CONFIG_XFRM_OFFLOAD)) + xfrm_features = netdev_increment_features(xfrm_features, + lower_dev->hw_enc_features, + MASTER_UPPER_DEV_XFRM_FEATURES); + + mpls_features = netdev_increment_features(mpls_features, + lower_dev->mpls_features, + MASTER_UPPER_DEV_MPLS_FEATURES); + + dst_release_flag &= lower_dev->priv_flags; + + if (update_header) { + max_header_len = max(max_header_len, lower_dev->hard_header_len); + max_headroom = max(max_headroom, lower_dev->needed_headroom); + max_tailroom = max(max_tailroom, lower_dev->needed_tailroom); + } + + tso_max_size = min(tso_max_size, lower_dev->tso_max_size); + tso_max_segs = min(tso_max_segs, lower_dev->tso_max_segs); + } + + dev->gso_partial_features = gso_partial_features; + dev->vlan_features = vlan_features; + dev->hw_enc_features = enc_features | NETIF_F_GSO_ENCAP_ALL | + NETIF_F_HW_VLAN_CTAG_TX | + NETIF_F_HW_VLAN_STAG_TX; + if (IS_ENABLED(CONFIG_XFRM_OFFLOAD)) + dev->hw_enc_features |= xfrm_features; + dev->mpls_features = mpls_features; + + dev->priv_flags &= ~IFF_XMIT_DST_RELEASE; + if ((dev->priv_flags & IFF_XMIT_DST_RELEASE_PERM) && + dst_release_flag == (IFF_XMIT_DST_RELEASE | IFF_XMIT_DST_RELEASE_PERM)) + dev->priv_flags |= IFF_XMIT_DST_RELEASE; + + if (update_header) { + dev->hard_header_len = max_header_len; + dev->needed_headroom = max_headroom; + dev->needed_tailroom = max_tailroom; + } + + netif_set_tso_max_segs(dev, tso_max_segs); + netif_set_tso_max_size(dev, tso_max_size); + + netdev_change_features(dev); +} +EXPORT_SYMBOL(netdev_compute_master_upper_features); + static struct hlist_head * __net_init netdev_create_hash(void) { int i; From d4fde269a970666a30dd3abd0413273a06dd972d Mon Sep 17 00:00:00 2001 From: Hangbin Liu Date: Fri, 17 Oct 2025 03:41:53 +0000 Subject: [PATCH 0177/1381] bonding: use common function to compute the features Use the new functon netdev_compute_master_upper_features() to compute the bonding features. Note that bond_compute_features() currently uses bond_for_each_slave() to traverse the lower devices list, and that is just a macro wrapper of netdev_for_each_lower_private(). We use similar helper netdev_for_each_lower_dev() in netdev_compute_master_upper_features() to iterate the slave device, as there is not need to get the private data. No functional change intended. Signed-off-by: Hangbin Liu Reviewed-by: Sabrina Dubroca Reviewed-by: Jiri Pirko Link: https://patch.msgid.link/20251017034155.61990-3-liuhangbin@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/bonding/bond_main.c | 99 ++------------------------------- 1 file changed, 4 insertions(+), 95 deletions(-) diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 4da619210c1f..cd7da6ed8c6b 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -1468,97 +1468,6 @@ static netdev_features_t bond_fix_features(struct net_device *dev, return features; } -#define BOND_VLAN_FEATURES (NETIF_F_HW_CSUM | NETIF_F_SG | \ - NETIF_F_FRAGLIST | NETIF_F_GSO_SOFTWARE | \ - NETIF_F_GSO_ENCAP_ALL | \ - NETIF_F_HIGHDMA | NETIF_F_LRO) - -#define BOND_ENC_FEATURES (NETIF_F_HW_CSUM | NETIF_F_SG | \ - NETIF_F_RXCSUM | NETIF_F_GSO_SOFTWARE | \ - NETIF_F_GSO_PARTIAL) - -#define BOND_MPLS_FEATURES (NETIF_F_HW_CSUM | NETIF_F_SG | \ - NETIF_F_GSO_SOFTWARE) - -#define BOND_GSO_PARTIAL_FEATURES (NETIF_F_GSO_ESP) - - -static void bond_compute_features(struct bonding *bond) -{ - netdev_features_t gso_partial_features = BOND_GSO_PARTIAL_FEATURES; - unsigned int dst_release_flag = IFF_XMIT_DST_RELEASE | - IFF_XMIT_DST_RELEASE_PERM; - netdev_features_t vlan_features = BOND_VLAN_FEATURES; - netdev_features_t enc_features = BOND_ENC_FEATURES; -#ifdef CONFIG_XFRM_OFFLOAD - netdev_features_t xfrm_features = BOND_XFRM_FEATURES; -#endif /* CONFIG_XFRM_OFFLOAD */ - netdev_features_t mpls_features = BOND_MPLS_FEATURES; - struct net_device *bond_dev = bond->dev; - struct list_head *iter; - struct slave *slave; - unsigned short max_hard_header_len = ETH_HLEN; - unsigned int tso_max_size = TSO_MAX_SIZE; - u16 tso_max_segs = TSO_MAX_SEGS; - - if (!bond_has_slaves(bond)) - goto done; - - vlan_features = netdev_base_features(vlan_features); - mpls_features = netdev_base_features(mpls_features); - - bond_for_each_slave(bond, slave, iter) { - vlan_features = netdev_increment_features(vlan_features, - slave->dev->vlan_features, BOND_VLAN_FEATURES); - - enc_features = netdev_increment_features(enc_features, - slave->dev->hw_enc_features, - BOND_ENC_FEATURES); - -#ifdef CONFIG_XFRM_OFFLOAD - xfrm_features = netdev_increment_features(xfrm_features, - slave->dev->hw_enc_features, - BOND_XFRM_FEATURES); -#endif /* CONFIG_XFRM_OFFLOAD */ - - gso_partial_features = netdev_increment_features(gso_partial_features, - slave->dev->gso_partial_features, - BOND_GSO_PARTIAL_FEATURES); - - mpls_features = netdev_increment_features(mpls_features, - slave->dev->mpls_features, - BOND_MPLS_FEATURES); - - dst_release_flag &= slave->dev->priv_flags; - if (slave->dev->hard_header_len > max_hard_header_len) - max_hard_header_len = slave->dev->hard_header_len; - - tso_max_size = min(tso_max_size, slave->dev->tso_max_size); - tso_max_segs = min(tso_max_segs, slave->dev->tso_max_segs); - } - bond_dev->hard_header_len = max_hard_header_len; - -done: - bond_dev->gso_partial_features = gso_partial_features; - bond_dev->vlan_features = vlan_features; - bond_dev->hw_enc_features = enc_features | NETIF_F_GSO_ENCAP_ALL | - NETIF_F_HW_VLAN_CTAG_TX | - NETIF_F_HW_VLAN_STAG_TX; -#ifdef CONFIG_XFRM_OFFLOAD - bond_dev->hw_enc_features |= xfrm_features; -#endif /* CONFIG_XFRM_OFFLOAD */ - bond_dev->mpls_features = mpls_features; - netif_set_tso_max_segs(bond_dev, tso_max_segs); - netif_set_tso_max_size(bond_dev, tso_max_size); - - bond_dev->priv_flags &= ~IFF_XMIT_DST_RELEASE; - if ((bond_dev->priv_flags & IFF_XMIT_DST_RELEASE_PERM) && - dst_release_flag == (IFF_XMIT_DST_RELEASE | IFF_XMIT_DST_RELEASE_PERM)) - bond_dev->priv_flags |= IFF_XMIT_DST_RELEASE; - - netdev_change_features(bond_dev); -} - static void bond_setup_by_slave(struct net_device *bond_dev, struct net_device *slave_dev) { @@ -2273,7 +2182,7 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev, } bond->slave_cnt++; - bond_compute_features(bond); + netdev_compute_master_upper_features(bond->dev, true); bond_set_carrier(bond); /* Needs to be called before bond_select_active_slave(), which will @@ -2525,7 +2434,7 @@ static int __bond_release_one(struct net_device *bond_dev, call_netdevice_notifiers(NETDEV_RELEASE, bond->dev); } - bond_compute_features(bond); + netdev_compute_master_upper_features(bond->dev, true); if (!(bond_dev->features & NETIF_F_VLAN_CHALLENGED) && (old_features & NETIF_F_VLAN_CHALLENGED)) slave_info(bond_dev, slave_dev, "last VLAN challenged slave left bond - VLAN blocking is removed\n"); @@ -4028,7 +3937,7 @@ static int bond_slave_netdev_event(unsigned long event, case NETDEV_FEAT_CHANGE: if (!bond->notifier_ctx) { bond->notifier_ctx = true; - bond_compute_features(bond); + netdev_compute_master_upper_features(bond->dev, true); bond->notifier_ctx = false; } break; @@ -6011,7 +5920,7 @@ void bond_setup(struct net_device *bond_dev) * capable */ - bond_dev->hw_features = BOND_VLAN_FEATURES | + bond_dev->hw_features = MASTER_UPPER_DEV_VLAN_FEATURES | NETIF_F_HW_VLAN_CTAG_RX | NETIF_F_HW_VLAN_CTAG_FILTER | NETIF_F_HW_VLAN_STAG_RX | From 745cd46c2a47144dd656185b9be0a1e5e9b02d2d Mon Sep 17 00:00:00 2001 From: Hangbin Liu Date: Fri, 17 Oct 2025 03:41:54 +0000 Subject: [PATCH 0178/1381] team: use common function to compute the features Use the new helper netdev_compute_master_upper_features() to compute the team device features. This helper performs both the feature computation and the netdev_change_features() call. Note that such change replace the lower layer traversing currently done using team->port_list with netdev_for_each_lower_dev(). Such change is safe as `port_list` contains exactly the same elements as `team->dev->adj_list.lower` and the helper is always invoked under the RTNL lock. With this change, the explicit netdev_change_features() in team_add_slave() can be safely removed, as team_port_add() already takes care of the notification via netdev_compute_master_upper_features(), and same thing for team_del_slave() This also fixes missing computations for MPLS, XFRM, and TSO/GSO partial features. Signed-off-by: Hangbin Liu Reviewed-by: Sabrina Dubroca Reviewed-by: Jiri Pirko Link: https://patch.msgid.link/20251017034155.61990-4-liuhangbin@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/team/team_core.c | 83 +++--------------------------------- 1 file changed, 6 insertions(+), 77 deletions(-) diff --git a/drivers/net/team/team_core.c b/drivers/net/team/team_core.c index 17f07eb0ee52..29dc04c299a3 100644 --- a/drivers/net/team/team_core.c +++ b/drivers/net/team/team_core.c @@ -982,63 +982,6 @@ static void team_port_disable(struct team *team, team_lower_state_changed(port); } -#define TEAM_VLAN_FEATURES (NETIF_F_HW_CSUM | NETIF_F_SG | \ - NETIF_F_FRAGLIST | NETIF_F_GSO_SOFTWARE | \ - NETIF_F_HIGHDMA | NETIF_F_LRO | \ - NETIF_F_GSO_ENCAP_ALL) - -#define TEAM_ENC_FEATURES (NETIF_F_HW_CSUM | NETIF_F_SG | \ - NETIF_F_RXCSUM | NETIF_F_GSO_SOFTWARE) - -static void __team_compute_features(struct team *team) -{ - struct team_port *port; - netdev_features_t vlan_features = TEAM_VLAN_FEATURES; - netdev_features_t enc_features = TEAM_ENC_FEATURES; - unsigned short max_hard_header_len = ETH_HLEN; - unsigned int dst_release_flag = IFF_XMIT_DST_RELEASE | - IFF_XMIT_DST_RELEASE_PERM; - - rcu_read_lock(); - if (list_empty(&team->port_list)) - goto done; - - vlan_features = netdev_base_features(vlan_features); - enc_features = netdev_base_features(enc_features); - - list_for_each_entry_rcu(port, &team->port_list, list) { - vlan_features = netdev_increment_features(vlan_features, - port->dev->vlan_features, - TEAM_VLAN_FEATURES); - enc_features = - netdev_increment_features(enc_features, - port->dev->hw_enc_features, - TEAM_ENC_FEATURES); - - dst_release_flag &= port->dev->priv_flags; - if (port->dev->hard_header_len > max_hard_header_len) - max_hard_header_len = port->dev->hard_header_len; - } -done: - rcu_read_unlock(); - - team->dev->vlan_features = vlan_features; - team->dev->hw_enc_features = enc_features | NETIF_F_GSO_ENCAP_ALL | - NETIF_F_HW_VLAN_CTAG_TX | - NETIF_F_HW_VLAN_STAG_TX; - team->dev->hard_header_len = max_hard_header_len; - - team->dev->priv_flags &= ~IFF_XMIT_DST_RELEASE; - if (dst_release_flag == (IFF_XMIT_DST_RELEASE | IFF_XMIT_DST_RELEASE_PERM)) - team->dev->priv_flags |= IFF_XMIT_DST_RELEASE; -} - -static void team_compute_features(struct team *team) -{ - __team_compute_features(team); - netdev_change_features(team->dev); -} - static int team_port_enter(struct team *team, struct team_port *port) { int err = 0; @@ -1300,7 +1243,7 @@ static int team_port_add(struct team *team, struct net_device *port_dev, port->index = -1; list_add_tail_rcu(&port->list, &team->port_list); team_port_enable(team, port); - __team_compute_features(team); + netdev_compute_master_upper_features(team->dev, true); __team_port_change_port_added(port, !!netif_oper_up(port_dev)); __team_options_change_check(team); @@ -1382,7 +1325,7 @@ static int team_port_del(struct team *team, struct net_device *port_dev) dev_set_mtu(port_dev, port->orig.mtu); kfree_rcu(port, rcu); netdev_info(dev, "Port device %s removed\n", portname); - __team_compute_features(team); + netdev_compute_master_upper_features(team->dev, true); return 0; } @@ -1970,33 +1913,19 @@ static int team_add_slave(struct net_device *dev, struct net_device *port_dev, struct netlink_ext_ack *extack) { struct team *team = netdev_priv(dev); - int err; ASSERT_RTNL(); - err = team_port_add(team, port_dev, extack); - - if (!err) - netdev_change_features(dev); - - return err; + return team_port_add(team, port_dev, extack); } static int team_del_slave(struct net_device *dev, struct net_device *port_dev) { struct team *team = netdev_priv(dev); - int err; ASSERT_RTNL(); - err = team_port_del(team, port_dev); - - if (err) - return err; - - netdev_change_features(dev); - - return err; + return team_port_del(team, port_dev); } static netdev_features_t team_fix_features(struct net_device *dev, @@ -2190,7 +2119,7 @@ static void team_setup(struct net_device *dev) dev->features |= NETIF_F_GRO; - dev->hw_features = TEAM_VLAN_FEATURES | + dev->hw_features = MASTER_UPPER_DEV_VLAN_FEATURES | NETIF_F_HW_VLAN_CTAG_RX | NETIF_F_HW_VLAN_CTAG_FILTER | NETIF_F_HW_VLAN_STAG_RX | @@ -2994,7 +2923,7 @@ static int team_device_event(struct notifier_block *unused, case NETDEV_FEAT_CHANGE: if (!port->team->notifier_ctx) { port->team->notifier_ctx = true; - team_compute_features(port->team); + netdev_compute_master_upper_features(port->team->dev, true); port->team->notifier_ctx = false; } break; From 0152747a528a185182bdb2cab973848a52d715ac Mon Sep 17 00:00:00 2001 From: Hangbin Liu Date: Fri, 17 Oct 2025 03:41:55 +0000 Subject: [PATCH 0179/1381] net: bridge: use common function to compute the features Previously, bridge ignored all features propagation and DST retention, only handling explicitly the GSO limits. By switching to the new helper netdev_compute_master_upper_features(), the bridge now expose additional features, depending on the lowers capabilities. Since br_set_gso_limits() is already covered by the helper, it can be removed safely. Bridge has it's own way to update needed_headroom. So we don't need to update it in the helper. Signed-off-by: Hangbin Liu Reviewed-by: Sabrina Dubroca Reviewed-by: Jiri Pirko Link: https://patch.msgid.link/20251017034155.61990-5-liuhangbin@gmail.com Signed-off-by: Jakub Kicinski --- net/bridge/br_if.c | 22 +++------------------- 1 file changed, 3 insertions(+), 19 deletions(-) diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c index 98c5b9c3145f..a6d4c44890fd 100644 --- a/net/bridge/br_if.c +++ b/net/bridge/br_if.c @@ -525,20 +525,6 @@ void br_mtu_auto_adjust(struct net_bridge *br) br_opt_toggle(br, BROPT_MTU_SET_BY_USER, false); } -static void br_set_gso_limits(struct net_bridge *br) -{ - unsigned int tso_max_size = TSO_MAX_SIZE; - const struct net_bridge_port *p; - u16 tso_max_segs = TSO_MAX_SEGS; - - list_for_each_entry(p, &br->port_list, list) { - tso_max_size = min(tso_max_size, p->dev->tso_max_size); - tso_max_segs = min(tso_max_segs, p->dev->tso_max_segs); - } - netif_set_tso_max_size(br->dev, tso_max_size); - netif_set_tso_max_segs(br->dev, tso_max_segs); -} - /* * Recomputes features using slave's features */ @@ -652,8 +638,6 @@ int br_add_if(struct net_bridge *br, struct net_device *dev, netdev_err(dev, "failed to sync bridge static fdb addresses to this port\n"); } - netdev_update_features(br->dev); - br_hr = br->dev->needed_headroom; dev_hr = netdev_get_fwd_headroom(dev); if (br_hr < dev_hr) @@ -694,7 +678,8 @@ int br_add_if(struct net_bridge *br, struct net_device *dev, call_netdevice_notifiers(NETDEV_CHANGEADDR, br->dev); br_mtu_auto_adjust(br); - br_set_gso_limits(br); + + netdev_compute_master_upper_features(br->dev, false); kobject_uevent(&p->kobj, KOBJ_ADD); @@ -740,7 +725,6 @@ int br_del_if(struct net_bridge *br, struct net_device *dev) del_nbp(p); br_mtu_auto_adjust(br); - br_set_gso_limits(br); spin_lock_bh(&br->lock); changed_addr = br_stp_recalculate_bridge_id(br); @@ -749,7 +733,7 @@ int br_del_if(struct net_bridge *br, struct net_device *dev) if (changed_addr) call_netdevice_notifiers(NETDEV_CHANGEADDR, br->dev); - netdev_update_features(br->dev); + netdev_compute_master_upper_features(br->dev, false); return 0; } From a9dff2b5f72b5ec21cec8e754a4fc2d1c03fd0a1 Mon Sep 17 00:00:00 2001 From: David Yang Date: Fri, 17 Oct 2025 14:08:53 +0800 Subject: [PATCH 0180/1381] dt-bindings: net: dsa: yt921x: Add Motorcomm YT921x switch support The Motorcomm YT921x series is a family of Ethernet switches with up to 8 internal GbE PHYs and up to 2 GMACs. Signed-off-by: David Yang Reviewed-by: Krzysztof Kozlowski Link: https://patch.msgid.link/20251017060859.326450-2-mmyangfl@gmail.com Signed-off-by: Jakub Kicinski --- .../bindings/net/dsa/motorcomm,yt921x.yaml | 167 ++++++++++++++++++ 1 file changed, 167 insertions(+) create mode 100644 Documentation/devicetree/bindings/net/dsa/motorcomm,yt921x.yaml diff --git a/Documentation/devicetree/bindings/net/dsa/motorcomm,yt921x.yaml b/Documentation/devicetree/bindings/net/dsa/motorcomm,yt921x.yaml new file mode 100644 index 000000000000..33a6552e46fc --- /dev/null +++ b/Documentation/devicetree/bindings/net/dsa/motorcomm,yt921x.yaml @@ -0,0 +1,167 @@ +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/net/dsa/motorcomm,yt921x.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Motorcomm YT921x Ethernet switch family + +maintainers: + - David Yang + +description: | + The Motorcomm YT921x series is a family of Ethernet switches with up to 8 + internal GbE PHYs and up to 2 GMACs, including: + + - YT9215S / YT9215RB / YT9215SC: 5 GbE PHYs (Port 0-4) + 2 GMACs (Port 8-9) + - YT9213NB: 2 GbE PHYs (Port 1/3) + 1 GMAC (Port 9) + - YT9214NB: 2 GbE PHYs (Port 1/3) + 2 GMACs (Port 8-9) + - YT9218N: 8 GbE PHYs (Port 0-7) + - YT9218MB: 8 GbE PHYs (Port 0-7) + 2 GMACs (Port 8-9) + + Any port can be used as the CPU port. + +properties: + compatible: + const: motorcomm,yt9215 + + reg: + enum: [0x0, 0x1d] + + reset-gpios: + maxItems: 1 + + mdio: + $ref: /schemas/net/mdio.yaml# + unevaluatedProperties: false + description: + Internal MDIO bus for the internal GbE PHYs. PHY 0-7 are used for Port + 0-7 respectively. + + mdio-external: + $ref: /schemas/net/mdio.yaml# + unevaluatedProperties: false + description: + External MDIO bus to access external components. External PHYs for GMACs + (Port 8-9) are expected to be connected to the external MDIO bus in + vendor's reference design, but that is not a hard limitation from the + chip. + +required: + - compatible + - reg + +allOf: + - $ref: dsa.yaml#/$defs/ethernet-ports + +unevaluatedProperties: false + +examples: + - | + #include + + mdio { + #address-cells = <1>; + #size-cells = <0>; + + switch@1d { + compatible = "motorcomm,yt9215"; + /* default 0x1d, alternate 0x0 */ + reg = <0x1d>; + reset-gpios = <&tlmm 39 GPIO_ACTIVE_LOW>; + + mdio { + #address-cells = <1>; + #size-cells = <0>; + + sw_phy0: phy@0 { + reg = <0x0>; + }; + + sw_phy1: phy@1 { + reg = <0x1>; + }; + + sw_phy2: phy@2 { + reg = <0x2>; + }; + + sw_phy3: phy@3 { + reg = <0x3>; + }; + + sw_phy4: phy@4 { + reg = <0x4>; + }; + }; + + mdio-external { + #address-cells = <1>; + #size-cells = <0>; + + phy1: phy@b { + reg = <0xb>; + }; + }; + + ethernet-ports { + #address-cells = <1>; + #size-cells = <0>; + + ethernet-port@0 { + reg = <0>; + label = "lan1"; + phy-mode = "internal"; + phy-handle = <&sw_phy0>; + }; + + ethernet-port@1 { + reg = <1>; + label = "lan2"; + phy-mode = "internal"; + phy-handle = <&sw_phy1>; + }; + + ethernet-port@2 { + reg = <2>; + label = "lan3"; + phy-mode = "internal"; + phy-handle = <&sw_phy2>; + }; + + ethernet-port@3 { + reg = <3>; + label = "lan4"; + phy-mode = "internal"; + phy-handle = <&sw_phy3>; + }; + + ethernet-port@4 { + reg = <4>; + label = "lan5"; + phy-mode = "internal"; + phy-handle = <&sw_phy4>; + }; + + /* CPU port */ + ethernet-port@8 { + reg = <8>; + phy-mode = "2500base-x"; + ethernet = <ð0>; + + fixed-link { + speed = <2500>; + full-duplex; + }; + }; + + /* if external phy is connected to a MAC */ + ethernet-port@9 { + reg = <9>; + label = "wan"; + phy-mode = "rgmii-id"; + phy-handle = <&phy1>; + }; + }; + }; + }; From ca4709843b7e72f96976cd6b35bca148a4071673 Mon Sep 17 00:00:00 2001 From: David Yang Date: Fri, 17 Oct 2025 14:08:54 +0800 Subject: [PATCH 0181/1381] net: dsa: tag_yt921x: add support for Motorcomm YT921x tags Add support for Motorcomm YT921x tags, which includes a proper configurable ethertype field (default to 0x9988). Signed-off-by: David Yang Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20251017060859.326450-3-mmyangfl@gmail.com Signed-off-by: Jakub Kicinski --- include/net/dsa.h | 2 + include/uapi/linux/if_ether.h | 1 + net/dsa/Kconfig | 6 ++ net/dsa/Makefile | 1 + net/dsa/tag_yt921x.c | 141 ++++++++++++++++++++++++++++++++++ 5 files changed, 151 insertions(+) create mode 100644 net/dsa/tag_yt921x.c diff --git a/include/net/dsa.h b/include/net/dsa.h index d73ea0880066..67762fdaf3c7 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -55,6 +55,7 @@ struct tc_action; #define DSA_TAG_PROTO_LAN937X_VALUE 27 #define DSA_TAG_PROTO_VSC73XX_8021Q_VALUE 28 #define DSA_TAG_PROTO_BRCM_LEGACY_FCS_VALUE 29 +#define DSA_TAG_PROTO_YT921X_VALUE 30 enum dsa_tag_protocol { DSA_TAG_PROTO_NONE = DSA_TAG_PROTO_NONE_VALUE, @@ -87,6 +88,7 @@ enum dsa_tag_protocol { DSA_TAG_PROTO_RZN1_A5PSW = DSA_TAG_PROTO_RZN1_A5PSW_VALUE, DSA_TAG_PROTO_LAN937X = DSA_TAG_PROTO_LAN937X_VALUE, DSA_TAG_PROTO_VSC73XX_8021Q = DSA_TAG_PROTO_VSC73XX_8021Q_VALUE, + DSA_TAG_PROTO_YT921X = DSA_TAG_PROTO_YT921X_VALUE, }; struct dsa_switch; diff --git a/include/uapi/linux/if_ether.h b/include/uapi/linux/if_ether.h index 69e0457eb200..cfd200c87e5e 100644 --- a/include/uapi/linux/if_ether.h +++ b/include/uapi/linux/if_ether.h @@ -114,6 +114,7 @@ #define ETH_P_QINQ1 0x9100 /* deprecated QinQ VLAN [ NOT AN OFFICIALLY REGISTERED ID ] */ #define ETH_P_QINQ2 0x9200 /* deprecated QinQ VLAN [ NOT AN OFFICIALLY REGISTERED ID ] */ #define ETH_P_QINQ3 0x9300 /* deprecated QinQ VLAN [ NOT AN OFFICIALLY REGISTERED ID ] */ +#define ETH_P_YT921X 0x9988 /* Motorcomm YT921x DSA [ NOT AN OFFICIALLY REGISTERED ID ] */ #define ETH_P_EDSA 0xDADA /* Ethertype DSA [ NOT AN OFFICIALLY REGISTERED ID ] */ #define ETH_P_DSA_8021Q 0xDADB /* Fake VLAN Header for DSA [ NOT AN OFFICIALLY REGISTERED ID ] */ #define ETH_P_DSA_A5PSW 0xE001 /* A5PSW Tag Value [ NOT AN OFFICIALLY REGISTERED ID ] */ diff --git a/net/dsa/Kconfig b/net/dsa/Kconfig index 869cbe57162f..6b94028b1fcc 100644 --- a/net/dsa/Kconfig +++ b/net/dsa/Kconfig @@ -190,4 +190,10 @@ config NET_DSA_TAG_XRS700X Say Y or M if you want to enable support for tagging frames for Arrow SpeedChips XRS700x switches that use a single byte tag trailer. +config NET_DSA_TAG_YT921X + tristate "Tag driver for Motorcomm YT921x switches" + help + Say Y or M if you want to enable support for tagging frames for + Motorcomm YT921x switches. + endif diff --git a/net/dsa/Makefile b/net/dsa/Makefile index 555c07cfeb71..4b011a1d5c87 100644 --- a/net/dsa/Makefile +++ b/net/dsa/Makefile @@ -39,6 +39,7 @@ obj-$(CONFIG_NET_DSA_TAG_SJA1105) += tag_sja1105.o obj-$(CONFIG_NET_DSA_TAG_TRAILER) += tag_trailer.o obj-$(CONFIG_NET_DSA_TAG_VSC73XX_8021Q) += tag_vsc73xx_8021q.o obj-$(CONFIG_NET_DSA_TAG_XRS700X) += tag_xrs700x.o +obj-$(CONFIG_NET_DSA_TAG_YT921X) += tag_yt921x.o # for tracing framework to find trace.h CFLAGS_trace.o := -I$(src) diff --git a/net/dsa/tag_yt921x.c b/net/dsa/tag_yt921x.c new file mode 100644 index 000000000000..995da44f0a2a --- /dev/null +++ b/net/dsa/tag_yt921x.c @@ -0,0 +1,141 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* + * Motorcomm YT921x Switch Extended CPU Port Tagging + * + * Copyright (c) 2025 David Yang + * + * +----+----+-------+-----+----+--------- + * | DA | SA | TagET | Tag | ET | Payload ... + * +----+----+-------+-----+----+--------- + * 6 6 2 6 2 N + * + * Tag Ethertype: CPU_TAG_TPID_TPID (default: ETH_P_YT921X = 0x9988) + * * Hardcoded for the moment, but still configurable. Discuss it if there + * are conflicts somewhere and/or you want to change it for some reason. + * Tag: + * 2: VLAN Tag + * 2: Rx Port + * 15b: Rx Port Valid + * 14b-11b: Rx Port + * 10b-0b: Cmd? + * 2: Tx Port(s) + * 15b: Tx Port(s) Valid + * 10b-0b: Tx Port(s) Mask + */ + +#include + +#include "tag.h" + +#define YT921X_TAG_NAME "yt921x" + +#define YT921X_TAG_LEN 8 + +#define YT921X_TAG_PORT_EN BIT(15) +#define YT921X_TAG_RX_PORT_M GENMASK(14, 11) +#define YT921X_TAG_RX_CMD_M GENMASK(10, 0) +#define YT921X_TAG_RX_CMD(x) FIELD_PREP(YT921X_TAG_RX_CMD_M, (x)) +#define YT921X_TAG_RX_CMD_FORWARDED 0x80 +#define YT921X_TAG_RX_CMD_UNK_UCAST 0xb2 +#define YT921X_TAG_RX_CMD_UNK_MCAST 0xb4 +#define YT921X_TAG_TX_PORTS_M GENMASK(10, 0) +#define YT921X_TAG_TX_PORTn(port) BIT(port) + +static struct sk_buff * +yt921x_tag_xmit(struct sk_buff *skb, struct net_device *netdev) +{ + struct dsa_port *dp = dsa_user_to_port(netdev); + unsigned int port = dp->index; + __be16 *tag; + u16 tx; + + skb_push(skb, YT921X_TAG_LEN); + dsa_alloc_etype_header(skb, YT921X_TAG_LEN); + + tag = dsa_etype_header_pos_tx(skb); + + tag[0] = htons(ETH_P_YT921X); + /* VLAN tag unrelated when TX */ + tag[1] = 0; + tag[2] = 0; + tx = YT921X_TAG_PORT_EN | YT921X_TAG_TX_PORTn(port); + tag[3] = htons(tx); + + return skb; +} + +static struct sk_buff * +yt921x_tag_rcv(struct sk_buff *skb, struct net_device *netdev) +{ + unsigned int port; + __be16 *tag; + u16 cmd; + u16 rx; + + if (unlikely(!pskb_may_pull(skb, YT921X_TAG_LEN))) + return NULL; + + tag = dsa_etype_header_pos_rx(skb); + + if (unlikely(tag[0] != htons(ETH_P_YT921X))) { + dev_warn_ratelimited(&netdev->dev, + "Unexpected EtherType 0x%04x\n", + ntohs(tag[0])); + return NULL; + } + + /* Locate which port this is coming from */ + rx = ntohs(tag[2]); + if (unlikely((rx & YT921X_TAG_PORT_EN) == 0)) { + dev_warn_ratelimited(&netdev->dev, + "Unexpected rx tag 0x%04x\n", rx); + return NULL; + } + + port = FIELD_GET(YT921X_TAG_RX_PORT_M, rx); + skb->dev = dsa_conduit_find_user(netdev, 0, port); + if (unlikely(!skb->dev)) { + dev_warn_ratelimited(&netdev->dev, + "Couldn't decode source port %u\n", port); + return NULL; + } + + cmd = FIELD_GET(YT921X_TAG_RX_CMD_M, rx); + switch (cmd) { + case YT921X_TAG_RX_CMD_FORWARDED: + /* Already forwarded by hardware */ + dsa_default_offload_fwd_mark(skb); + break; + case YT921X_TAG_RX_CMD_UNK_UCAST: + case YT921X_TAG_RX_CMD_UNK_MCAST: + /* NOTE: hardware doesn't distinguish between TRAP (copy to CPU + * only) and COPY (forward and copy to CPU). In order to perform + * a soft switch, NEVER use COPY action in the switch driver. + */ + break; + default: + dev_warn_ratelimited(&netdev->dev, + "Unexpected rx cmd 0x%02x\n", cmd); + break; + } + + /* Remove YT921x tag and update checksum */ + skb_pull_rcsum(skb, YT921X_TAG_LEN); + dsa_strip_etype_header(skb, YT921X_TAG_LEN); + + return skb; +} + +static const struct dsa_device_ops yt921x_netdev_ops = { + .name = YT921X_TAG_NAME, + .proto = DSA_TAG_PROTO_YT921X, + .xmit = yt921x_tag_xmit, + .rcv = yt921x_tag_rcv, + .needed_headroom = YT921X_TAG_LEN, +}; + +MODULE_DESCRIPTION("DSA tag driver for Motorcomm YT921x switches"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS_DSA_TAG_DRIVER(DSA_TAG_PROTO_YT921X, YT921X_TAG_NAME); + +module_dsa_tag_driver(yt921x_netdev_ops); From 186623f4aa724c46cbb4dbd5235cf6942215f5b5 Mon Sep 17 00:00:00 2001 From: David Yang Date: Fri, 17 Oct 2025 14:08:55 +0800 Subject: [PATCH 0182/1381] net: dsa: yt921x: Add support for Motorcomm YT921x Motorcomm YT921x is a series of ethernet switches developed by Shanghai Motorcomm Electronic Technology, including: - YT9215S / YT9215RB / YT9215SC: 5 GbE PHYs - YT9213NB / YT9214NB: 2 GbE PHYs - YT9218N / YT9218MB: 8 GbE PHYs and up to 2 GMACs. Driver verified on a stock wireless router with IPQ5018 + YT9215S. Signed-off-by: David Yang Link: https://patch.msgid.link/20251017060859.326450-4-mmyangfl@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/Kconfig | 7 + drivers/net/dsa/Makefile | 1 + drivers/net/dsa/yt921x.c | 2891 ++++++++++++++++++++++++++++++++++++++ drivers/net/dsa/yt921x.h | 504 +++++++ 4 files changed, 3403 insertions(+) create mode 100644 drivers/net/dsa/yt921x.c create mode 100644 drivers/net/dsa/yt921x.h diff --git a/drivers/net/dsa/Kconfig b/drivers/net/dsa/Kconfig index 4d9af691b989..7eb301fd987d 100644 --- a/drivers/net/dsa/Kconfig +++ b/drivers/net/dsa/Kconfig @@ -154,4 +154,11 @@ config NET_DSA_VITESSE_VSC73XX_PLATFORM This enables support for the Vitesse VSC7385, VSC7388, VSC7395 and VSC7398 SparX integrated ethernet switches, connected over a CPU-attached address bus and work in memory-mapped I/O mode. + +config NET_DSA_YT921X + tristate "Motorcomm YT9215 ethernet switch chip support" + select NET_DSA_TAG_YT921X + help + This enables support for the Motorcomm YT9215 ethernet switch + chip. endmenu diff --git a/drivers/net/dsa/Makefile b/drivers/net/dsa/Makefile index 0f8ff4a1a313..16de4ba3fa38 100644 --- a/drivers/net/dsa/Makefile +++ b/drivers/net/dsa/Makefile @@ -14,6 +14,7 @@ obj-$(CONFIG_NET_DSA_SMSC_LAN9303_MDIO) += lan9303_mdio.o obj-$(CONFIG_NET_DSA_VITESSE_VSC73XX) += vitesse-vsc73xx-core.o obj-$(CONFIG_NET_DSA_VITESSE_VSC73XX_PLATFORM) += vitesse-vsc73xx-platform.o obj-$(CONFIG_NET_DSA_VITESSE_VSC73XX_SPI) += vitesse-vsc73xx-spi.o +obj-$(CONFIG_NET_DSA_YT921X) += yt921x.o obj-y += b53/ obj-y += hirschmann/ obj-y += lantiq/ diff --git a/drivers/net/dsa/yt921x.c b/drivers/net/dsa/yt921x.c new file mode 100644 index 000000000000..ab762ffc4661 --- /dev/null +++ b/drivers/net/dsa/yt921x.c @@ -0,0 +1,2891 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* + * Driver for Motorcomm YT921x Switch + * + * Should work on YT9213/YT9214/YT9215/YT9218, but only tested on YT9215+SGMII, + * be sure to do your own checks before porting to another chip. + * + * Copyright (c) 2025 David Yang + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include + +#include "yt921x.h" + +struct yt921x_mib_desc { + unsigned int size; + unsigned int offset; + const char *name; +}; + +#define MIB_DESC(_size, _offset, _name) \ + {_size, _offset, _name} + +/* Must agree with yt921x_mib + * + * Unstructured fields (name != NULL) will appear in get_ethtool_stats(), + * structured go to their *_stats() methods, but we need their sizes and offsets + * to perform 32bit MIB overflow wraparound. + */ +static const struct yt921x_mib_desc yt921x_mib_descs[] = { + MIB_DESC(1, 0x00, NULL), /* RxBroadcast */ + MIB_DESC(1, 0x04, NULL), /* RxPause */ + MIB_DESC(1, 0x08, NULL), /* RxMulticast */ + MIB_DESC(1, 0x0c, NULL), /* RxCrcErr */ + + MIB_DESC(1, 0x10, NULL), /* RxAlignErr */ + MIB_DESC(1, 0x14, NULL), /* RxUnderSizeErr */ + MIB_DESC(1, 0x18, NULL), /* RxFragErr */ + MIB_DESC(1, 0x1c, NULL), /* RxPktSz64 */ + + MIB_DESC(1, 0x20, NULL), /* RxPktSz65To127 */ + MIB_DESC(1, 0x24, NULL), /* RxPktSz128To255 */ + MIB_DESC(1, 0x28, NULL), /* RxPktSz256To511 */ + MIB_DESC(1, 0x2c, NULL), /* RxPktSz512To1023 */ + + MIB_DESC(1, 0x30, NULL), /* RxPktSz1024To1518 */ + MIB_DESC(1, 0x34, NULL), /* RxPktSz1519ToMax */ + MIB_DESC(2, 0x38, NULL), /* RxGoodBytes */ + /* 0x3c */ + + MIB_DESC(2, 0x40, "RxBadBytes"), + /* 0x44 */ + MIB_DESC(2, 0x48, NULL), /* RxOverSzErr */ + /* 0x4c */ + + MIB_DESC(1, 0x50, NULL), /* RxDropped */ + MIB_DESC(1, 0x54, NULL), /* TxBroadcast */ + MIB_DESC(1, 0x58, NULL), /* TxPause */ + MIB_DESC(1, 0x5c, NULL), /* TxMulticast */ + + MIB_DESC(1, 0x60, NULL), /* TxUnderSizeErr */ + MIB_DESC(1, 0x64, NULL), /* TxPktSz64 */ + MIB_DESC(1, 0x68, NULL), /* TxPktSz65To127 */ + MIB_DESC(1, 0x6c, NULL), /* TxPktSz128To255 */ + + MIB_DESC(1, 0x70, NULL), /* TxPktSz256To511 */ + MIB_DESC(1, 0x74, NULL), /* TxPktSz512To1023 */ + MIB_DESC(1, 0x78, NULL), /* TxPktSz1024To1518 */ + MIB_DESC(1, 0x7c, NULL), /* TxPktSz1519ToMax */ + + MIB_DESC(2, 0x80, NULL), /* TxGoodBytes */ + /* 0x84 */ + MIB_DESC(2, 0x88, NULL), /* TxCollision */ + /* 0x8c */ + + MIB_DESC(1, 0x90, NULL), /* TxExcessiveCollistion */ + MIB_DESC(1, 0x94, NULL), /* TxMultipleCollision */ + MIB_DESC(1, 0x98, NULL), /* TxSingleCollision */ + MIB_DESC(1, 0x9c, NULL), /* TxPkt */ + + MIB_DESC(1, 0xa0, NULL), /* TxDeferred */ + MIB_DESC(1, 0xa4, NULL), /* TxLateCollision */ + MIB_DESC(1, 0xa8, "RxOAM"), + MIB_DESC(1, 0xac, "TxOAM"), +}; + +struct yt921x_info { + const char *name; + u16 major; + /* Unknown, seems to be plain enumeration */ + u8 mode; + u8 extmode; + /* Ports with integral GbE PHYs, not including MCU Port 10 */ + u16 internal_mask; + /* TODO: see comments in yt921x_dsa_phylink_get_caps() */ + u16 external_mask; +}; + +#define YT921X_PORT_MASK_INTn(port) BIT(port) +#define YT921X_PORT_MASK_INT0_n(n) GENMASK((n) - 1, 0) +#define YT921X_PORT_MASK_EXT0 BIT(8) +#define YT921X_PORT_MASK_EXT1 BIT(9) + +static const struct yt921x_info yt921x_infos[] = { + { + "YT9215SC", YT9215_MAJOR, 1, 0, + YT921X_PORT_MASK_INT0_n(5), + YT921X_PORT_MASK_EXT0 | YT921X_PORT_MASK_EXT1, + }, + { + "YT9215S", YT9215_MAJOR, 2, 0, + YT921X_PORT_MASK_INT0_n(5), + YT921X_PORT_MASK_EXT0 | YT921X_PORT_MASK_EXT1, + }, + { + "YT9215RB", YT9215_MAJOR, 3, 0, + YT921X_PORT_MASK_INT0_n(5), + YT921X_PORT_MASK_EXT0 | YT921X_PORT_MASK_EXT1, + }, + { + "YT9214NB", YT9215_MAJOR, 3, 2, + YT921X_PORT_MASK_INTn(1) | YT921X_PORT_MASK_INTn(3), + YT921X_PORT_MASK_EXT0 | YT921X_PORT_MASK_EXT1, + }, + { + "YT9213NB", YT9215_MAJOR, 3, 3, + YT921X_PORT_MASK_INTn(1) | YT921X_PORT_MASK_INTn(3), + YT921X_PORT_MASK_EXT1, + }, + { + "YT9218N", YT9218_MAJOR, 0, 0, + YT921X_PORT_MASK_INT0_n(8), + 0, + }, + { + "YT9218MB", YT9218_MAJOR, 1, 0, + YT921X_PORT_MASK_INT0_n(8), + YT921X_PORT_MASK_EXT0 | YT921X_PORT_MASK_EXT1, + }, + {} +}; + +#define YT921X_NAME "yt921x" + +#define YT921X_VID_UNWARE 4095 + +#define YT921X_POLL_SLEEP_US 10000 +#define YT921X_POLL_TIMEOUT_US 100000 + +/* The interval should be small enough to avoid overflow of 32bit MIBs. + * + * Until we can read MIBs from stats64 call directly (i.e. sleep + * there), we have to poll stats more frequently then it is actually needed. + * For overflow protection, normally, 100 sec interval should have been OK. + */ +#define YT921X_STATS_INTERVAL_JIFFIES (3 * HZ) + +struct yt921x_reg_mdio { + struct mii_bus *bus; + int addr; + /* SWITCH_ID_1 / SWITCH_ID_0 of the device + * + * This is a way to multiplex multiple devices on the same MII phyaddr + * and should be configurable in DT. However, MDIO core simply doesn't + * allow multiple devices over one reg addr, so this is a fixed value + * for now until a solution is found. + * + * Keep this because we need switchid to form MII regaddrs anyway. + */ + unsigned char switchid; +}; + +/* TODO: SPI/I2C */ + +#define to_yt921x_priv(_ds) container_of_const(_ds, struct yt921x_priv, ds) +#define to_device(priv) ((priv)->ds.dev) + +static int yt921x_reg_read(struct yt921x_priv *priv, u32 reg, u32 *valp) +{ + WARN_ON(!mutex_is_locked(&priv->reg_lock)); + + return priv->reg_ops->read(priv->reg_ctx, reg, valp); +} + +static int yt921x_reg_write(struct yt921x_priv *priv, u32 reg, u32 val) +{ + WARN_ON(!mutex_is_locked(&priv->reg_lock)); + + return priv->reg_ops->write(priv->reg_ctx, reg, val); +} + +static int +yt921x_reg_wait(struct yt921x_priv *priv, u32 reg, u32 mask, u32 *valp) +{ + u32 val; + int res; + int ret; + + ret = read_poll_timeout(yt921x_reg_read, res, + res || (val & mask) == *valp, + YT921X_POLL_SLEEP_US, YT921X_POLL_TIMEOUT_US, + false, priv, reg, &val); + if (ret) + return ret; + if (res) + return res; + + *valp = val; + return 0; +} + +static int +yt921x_reg_update_bits(struct yt921x_priv *priv, u32 reg, u32 mask, u32 val) +{ + int res; + u32 v; + u32 u; + + res = yt921x_reg_read(priv, reg, &v); + if (res) + return res; + + u = v; + u &= ~mask; + u |= val; + if (u == v) + return 0; + + return yt921x_reg_write(priv, reg, u); +} + +static int yt921x_reg_set_bits(struct yt921x_priv *priv, u32 reg, u32 mask) +{ + return yt921x_reg_update_bits(priv, reg, 0, mask); +} + +static int yt921x_reg_clear_bits(struct yt921x_priv *priv, u32 reg, u32 mask) +{ + return yt921x_reg_update_bits(priv, reg, mask, 0); +} + +static int +yt921x_reg_toggle_bits(struct yt921x_priv *priv, u32 reg, u32 mask, bool set) +{ + return yt921x_reg_update_bits(priv, reg, mask, !set ? 0 : mask); +} + +/* Some registers, like VLANn_CTRL, should always be written in 64-bit, even if + * you are to write only the lower / upper 32 bits. + * + * There is no such restriction for reading, but we still provide 64-bit read + * wrappers so that we always handle u64 values. + */ + +static int yt921x_reg64_read(struct yt921x_priv *priv, u32 reg, u64 *valp) +{ + u32 lo; + u32 hi; + int res; + + res = yt921x_reg_read(priv, reg, &lo); + if (res) + return res; + res = yt921x_reg_read(priv, reg + 4, &hi); + if (res) + return res; + + *valp = ((u64)hi << 32) | lo; + return 0; +} + +static int yt921x_reg64_write(struct yt921x_priv *priv, u32 reg, u64 val) +{ + int res; + + res = yt921x_reg_write(priv, reg, (u32)val); + if (res) + return res; + return yt921x_reg_write(priv, reg + 4, (u32)(val >> 32)); +} + +static int +yt921x_reg64_update_bits(struct yt921x_priv *priv, u32 reg, u64 mask, u64 val) +{ + int res; + u64 v; + u64 u; + + res = yt921x_reg64_read(priv, reg, &v); + if (res) + return res; + + u = v; + u &= ~mask; + u |= val; + if (u == v) + return 0; + + return yt921x_reg64_write(priv, reg, u); +} + +static int yt921x_reg64_clear_bits(struct yt921x_priv *priv, u32 reg, u64 mask) +{ + return yt921x_reg64_update_bits(priv, reg, mask, 0); +} + +static int yt921x_reg_mdio_read(void *context, u32 reg, u32 *valp) +{ + struct yt921x_reg_mdio *mdio = context; + struct mii_bus *bus = mdio->bus; + int addr = mdio->addr; + u32 reg_addr; + u32 reg_data; + u32 val; + int res; + + /* Hold the mdio bus lock to avoid (un)locking for 4 times */ + mutex_lock_nested(&bus->mdio_lock, MDIO_MUTEX_NESTED); + + reg_addr = YT921X_SMI_SWITCHID(mdio->switchid) | YT921X_SMI_ADDR | + YT921X_SMI_READ; + res = __mdiobus_write(bus, addr, reg_addr, (u16)(reg >> 16)); + if (res) + goto end; + res = __mdiobus_write(bus, addr, reg_addr, (u16)reg); + if (res) + goto end; + + reg_data = YT921X_SMI_SWITCHID(mdio->switchid) | YT921X_SMI_DATA | + YT921X_SMI_READ; + res = __mdiobus_read(bus, addr, reg_data); + if (res < 0) + goto end; + val = (u16)res; + res = __mdiobus_read(bus, addr, reg_data); + if (res < 0) + goto end; + val = (val << 16) | (u16)res; + + *valp = val; + res = 0; + +end: + mutex_unlock(&bus->mdio_lock); + return res; +} + +static int yt921x_reg_mdio_write(void *context, u32 reg, u32 val) +{ + struct yt921x_reg_mdio *mdio = context; + struct mii_bus *bus = mdio->bus; + int addr = mdio->addr; + u32 reg_addr; + u32 reg_data; + int res; + + mutex_lock_nested(&bus->mdio_lock, MDIO_MUTEX_NESTED); + + reg_addr = YT921X_SMI_SWITCHID(mdio->switchid) | YT921X_SMI_ADDR | + YT921X_SMI_WRITE; + res = __mdiobus_write(bus, addr, reg_addr, (u16)(reg >> 16)); + if (res) + goto end; + res = __mdiobus_write(bus, addr, reg_addr, (u16)reg); + if (res) + goto end; + + reg_data = YT921X_SMI_SWITCHID(mdio->switchid) | YT921X_SMI_DATA | + YT921X_SMI_WRITE; + res = __mdiobus_write(bus, addr, reg_data, (u16)(val >> 16)); + if (res) + goto end; + res = __mdiobus_write(bus, addr, reg_data, (u16)val); + if (res) + goto end; + + res = 0; + +end: + mutex_unlock(&bus->mdio_lock); + return res; +} + +static const struct yt921x_reg_ops yt921x_reg_ops_mdio = { + .read = yt921x_reg_mdio_read, + .write = yt921x_reg_mdio_write, +}; + +/* TODO: SPI/I2C */ + +static int yt921x_intif_wait(struct yt921x_priv *priv) +{ + u32 val = 0; + + return yt921x_reg_wait(priv, YT921X_INT_MBUS_OP, YT921X_MBUS_OP_START, + &val); +} + +static int +yt921x_intif_read(struct yt921x_priv *priv, int port, int reg, u16 *valp) +{ + struct device *dev = to_device(priv); + u32 mask; + u32 ctrl; + u32 val; + int res; + + res = yt921x_intif_wait(priv); + if (res) + return res; + + mask = YT921X_MBUS_CTRL_PORT_M | YT921X_MBUS_CTRL_REG_M | + YT921X_MBUS_CTRL_OP_M; + ctrl = YT921X_MBUS_CTRL_PORT(port) | YT921X_MBUS_CTRL_REG(reg) | + YT921X_MBUS_CTRL_READ; + res = yt921x_reg_update_bits(priv, YT921X_INT_MBUS_CTRL, mask, ctrl); + if (res) + return res; + res = yt921x_reg_write(priv, YT921X_INT_MBUS_OP, YT921X_MBUS_OP_START); + if (res) + return res; + + res = yt921x_intif_wait(priv); + if (res) + return res; + res = yt921x_reg_read(priv, YT921X_INT_MBUS_DIN, &val); + if (res) + return res; + + if ((u16)val != val) + dev_info(dev, + "%s: port %d, reg 0x%x: Expected u16, got 0x%08x\n", + __func__, port, reg, val); + *valp = (u16)val; + return 0; +} + +static int +yt921x_intif_write(struct yt921x_priv *priv, int port, int reg, u16 val) +{ + u32 mask; + u32 ctrl; + int res; + + res = yt921x_intif_wait(priv); + if (res) + return res; + + mask = YT921X_MBUS_CTRL_PORT_M | YT921X_MBUS_CTRL_REG_M | + YT921X_MBUS_CTRL_OP_M; + ctrl = YT921X_MBUS_CTRL_PORT(port) | YT921X_MBUS_CTRL_REG(reg) | + YT921X_MBUS_CTRL_WRITE; + res = yt921x_reg_update_bits(priv, YT921X_INT_MBUS_CTRL, mask, ctrl); + if (res) + return res; + res = yt921x_reg_write(priv, YT921X_INT_MBUS_DOUT, val); + if (res) + return res; + res = yt921x_reg_write(priv, YT921X_INT_MBUS_OP, YT921X_MBUS_OP_START); + if (res) + return res; + + return yt921x_intif_wait(priv); +} + +static int yt921x_mbus_int_read(struct mii_bus *mbus, int port, int reg) +{ + struct yt921x_priv *priv = mbus->priv; + u16 val; + int res; + + if (port >= YT921X_PORT_NUM) + return U16_MAX; + + mutex_lock(&priv->reg_lock); + res = yt921x_intif_read(priv, port, reg, &val); + mutex_unlock(&priv->reg_lock); + + if (res) + return res; + return val; +} + +static int +yt921x_mbus_int_write(struct mii_bus *mbus, int port, int reg, u16 data) +{ + struct yt921x_priv *priv = mbus->priv; + int res; + + if (port >= YT921X_PORT_NUM) + return -ENODEV; + + mutex_lock(&priv->reg_lock); + res = yt921x_intif_write(priv, port, reg, data); + mutex_unlock(&priv->reg_lock); + + return res; +} + +static int +yt921x_mbus_int_init(struct yt921x_priv *priv, struct device_node *mnp) +{ + struct device *dev = to_device(priv); + struct mii_bus *mbus; + int res; + + mbus = devm_mdiobus_alloc(dev); + if (!mbus) + return -ENOMEM; + + mbus->name = "YT921x internal MDIO bus"; + snprintf(mbus->id, MII_BUS_ID_SIZE, "%s", dev_name(dev)); + mbus->priv = priv; + mbus->read = yt921x_mbus_int_read; + mbus->write = yt921x_mbus_int_write; + mbus->parent = dev; + mbus->phy_mask = (u32)~GENMASK(YT921X_PORT_NUM - 1, 0); + + res = devm_of_mdiobus_register(dev, mbus, mnp); + if (res) + return res; + + priv->mbus_int = mbus; + + return 0; +} + +static int yt921x_extif_wait(struct yt921x_priv *priv) +{ + u32 val = 0; + + return yt921x_reg_wait(priv, YT921X_EXT_MBUS_OP, YT921X_MBUS_OP_START, + &val); +} + +static int +yt921x_extif_read(struct yt921x_priv *priv, int port, int reg, u16 *valp) +{ + struct device *dev = to_device(priv); + u32 mask; + u32 ctrl; + u32 val; + int res; + + res = yt921x_extif_wait(priv); + if (res) + return res; + + mask = YT921X_MBUS_CTRL_PORT_M | YT921X_MBUS_CTRL_REG_M | + YT921X_MBUS_CTRL_TYPE_M | YT921X_MBUS_CTRL_OP_M; + ctrl = YT921X_MBUS_CTRL_PORT(port) | YT921X_MBUS_CTRL_REG(reg) | + YT921X_MBUS_CTRL_TYPE_C22 | YT921X_MBUS_CTRL_READ; + res = yt921x_reg_update_bits(priv, YT921X_EXT_MBUS_CTRL, mask, ctrl); + if (res) + return res; + res = yt921x_reg_write(priv, YT921X_EXT_MBUS_OP, YT921X_MBUS_OP_START); + if (res) + return res; + + res = yt921x_extif_wait(priv); + if (res) + return res; + res = yt921x_reg_read(priv, YT921X_EXT_MBUS_DIN, &val); + if (res) + return res; + + if ((u16)val != val) + dev_info(dev, + "%s: port %d, reg 0x%x: Expected u16, got 0x%08x\n", + __func__, port, reg, val); + *valp = (u16)val; + return 0; +} + +static int +yt921x_extif_write(struct yt921x_priv *priv, int port, int reg, u16 val) +{ + u32 mask; + u32 ctrl; + int res; + + res = yt921x_extif_wait(priv); + if (res) + return res; + + mask = YT921X_MBUS_CTRL_PORT_M | YT921X_MBUS_CTRL_REG_M | + YT921X_MBUS_CTRL_TYPE_M | YT921X_MBUS_CTRL_OP_M; + ctrl = YT921X_MBUS_CTRL_PORT(port) | YT921X_MBUS_CTRL_REG(reg) | + YT921X_MBUS_CTRL_TYPE_C22 | YT921X_MBUS_CTRL_WRITE; + res = yt921x_reg_update_bits(priv, YT921X_EXT_MBUS_CTRL, mask, ctrl); + if (res) + return res; + res = yt921x_reg_write(priv, YT921X_EXT_MBUS_DOUT, val); + if (res) + return res; + res = yt921x_reg_write(priv, YT921X_EXT_MBUS_OP, YT921X_MBUS_OP_START); + if (res) + return res; + + return yt921x_extif_wait(priv); +} + +static int yt921x_mbus_ext_read(struct mii_bus *mbus, int port, int reg) +{ + struct yt921x_priv *priv = mbus->priv; + u16 val; + int res; + + mutex_lock(&priv->reg_lock); + res = yt921x_extif_read(priv, port, reg, &val); + mutex_unlock(&priv->reg_lock); + + if (res) + return res; + return val; +} + +static int +yt921x_mbus_ext_write(struct mii_bus *mbus, int port, int reg, u16 data) +{ + struct yt921x_priv *priv = mbus->priv; + int res; + + mutex_lock(&priv->reg_lock); + res = yt921x_extif_write(priv, port, reg, data); + mutex_unlock(&priv->reg_lock); + + return res; +} + +static int +yt921x_mbus_ext_init(struct yt921x_priv *priv, struct device_node *mnp) +{ + struct device *dev = to_device(priv); + struct mii_bus *mbus; + int res; + + mbus = devm_mdiobus_alloc(dev); + if (!mbus) + return -ENOMEM; + + mbus->name = "YT921x external MDIO bus"; + snprintf(mbus->id, MII_BUS_ID_SIZE, "%s@ext", dev_name(dev)); + mbus->priv = priv; + /* TODO: c45? */ + mbus->read = yt921x_mbus_ext_read; + mbus->write = yt921x_mbus_ext_write; + mbus->parent = dev; + + res = devm_of_mdiobus_register(dev, mbus, mnp); + if (res) + return res; + + priv->mbus_ext = mbus; + + return 0; +} + +/* Read and handle overflow of 32bit MIBs. MIB buffer must be zeroed before. */ +static int yt921x_read_mib(struct yt921x_priv *priv, int port) +{ + struct yt921x_port *pp = &priv->ports[port]; + struct device *dev = to_device(priv); + struct yt921x_mib *mib = &pp->mib; + int res = 0; + + /* Reading of yt921x_port::mib is not protected by a lock and it's vain + * to keep its consistency, since we have to read registers one by one + * and there is no way to make a snapshot of MIB stats. + * + * Writing (by this function only) is and should be protected by + * reg_lock. + */ + + for (size_t i = 0; i < ARRAY_SIZE(yt921x_mib_descs); i++) { + const struct yt921x_mib_desc *desc = &yt921x_mib_descs[i]; + u32 reg = YT921X_MIBn_DATA0(port) + desc->offset; + u64 *valp = &((u64 *)mib)[i]; + u64 val = *valp; + u32 val0; + u32 val1; + + res = yt921x_reg_read(priv, reg, &val0); + if (res) + break; + + if (desc->size <= 1) { + if (val < (u32)val) + /* overflow */ + val += (u64)U32_MAX + 1; + val &= ~U32_MAX; + val |= val0; + } else { + res = yt921x_reg_read(priv, reg + 4, &val1); + if (res) + break; + val = ((u64)val0 << 32) | val1; + } + + WRITE_ONCE(*valp, val); + } + + pp->rx_frames = mib->rx_64byte + mib->rx_65_127byte + + mib->rx_128_255byte + mib->rx_256_511byte + + mib->rx_512_1023byte + mib->rx_1024_1518byte + + mib->rx_jumbo; + pp->tx_frames = mib->tx_64byte + mib->tx_65_127byte + + mib->tx_128_255byte + mib->tx_256_511byte + + mib->tx_512_1023byte + mib->tx_1024_1518byte + + mib->tx_jumbo; + + if (res) + dev_err(dev, "Failed to %s port %d: %i\n", "read stats for", + port, res); + return res; +} + +static void yt921x_poll_mib(struct work_struct *work) +{ + struct yt921x_port *pp = container_of_const(work, struct yt921x_port, + mib_read.work); + struct yt921x_priv *priv = (void *)(pp - pp->index) - + offsetof(struct yt921x_priv, ports); + unsigned long delay = YT921X_STATS_INTERVAL_JIFFIES; + int port = pp->index; + int res; + + mutex_lock(&priv->reg_lock); + res = yt921x_read_mib(priv, port); + mutex_unlock(&priv->reg_lock); + if (res) + delay *= 4; + + schedule_delayed_work(&pp->mib_read, delay); +} + +static void +yt921x_dsa_get_strings(struct dsa_switch *ds, int port, u32 stringset, + uint8_t *data) +{ + if (stringset != ETH_SS_STATS) + return; + + for (size_t i = 0; i < ARRAY_SIZE(yt921x_mib_descs); i++) { + const struct yt921x_mib_desc *desc = &yt921x_mib_descs[i]; + + if (desc->name) + ethtool_puts(&data, desc->name); + } +} + +static void +yt921x_dsa_get_ethtool_stats(struct dsa_switch *ds, int port, uint64_t *data) +{ + struct yt921x_priv *priv = to_yt921x_priv(ds); + struct yt921x_port *pp = &priv->ports[port]; + struct yt921x_mib *mib = &pp->mib; + size_t j; + + mutex_lock(&priv->reg_lock); + yt921x_read_mib(priv, port); + mutex_unlock(&priv->reg_lock); + + j = 0; + for (size_t i = 0; i < ARRAY_SIZE(yt921x_mib_descs); i++) { + const struct yt921x_mib_desc *desc = &yt921x_mib_descs[i]; + + if (!desc->name) + continue; + + data[j] = ((u64 *)mib)[i]; + j++; + } +} + +static int yt921x_dsa_get_sset_count(struct dsa_switch *ds, int port, int sset) +{ + int cnt = 0; + + if (sset != ETH_SS_STATS) + return 0; + + for (size_t i = 0; i < ARRAY_SIZE(yt921x_mib_descs); i++) { + const struct yt921x_mib_desc *desc = &yt921x_mib_descs[i]; + + if (desc->name) + cnt++; + } + + return cnt; +} + +static void +yt921x_dsa_get_eth_mac_stats(struct dsa_switch *ds, int port, + struct ethtool_eth_mac_stats *mac_stats) +{ + struct yt921x_priv *priv = to_yt921x_priv(ds); + struct yt921x_port *pp = &priv->ports[port]; + struct yt921x_mib *mib = &pp->mib; + + mutex_lock(&priv->reg_lock); + yt921x_read_mib(priv, port); + mutex_unlock(&priv->reg_lock); + + mac_stats->FramesTransmittedOK = pp->tx_frames; + mac_stats->SingleCollisionFrames = mib->tx_single_collisions; + mac_stats->MultipleCollisionFrames = mib->tx_multiple_collisions; + mac_stats->FramesReceivedOK = pp->rx_frames; + mac_stats->FrameCheckSequenceErrors = mib->rx_crc_errors; + mac_stats->AlignmentErrors = mib->rx_alignment_errors; + mac_stats->OctetsTransmittedOK = mib->tx_good_bytes; + mac_stats->FramesWithDeferredXmissions = mib->tx_deferred; + mac_stats->LateCollisions = mib->tx_late_collisions; + mac_stats->FramesAbortedDueToXSColls = mib->tx_aborted_errors; + /* mac_stats->FramesLostDueToIntMACXmitError */ + /* mac_stats->CarrierSenseErrors */ + mac_stats->OctetsReceivedOK = mib->rx_good_bytes; + /* mac_stats->FramesLostDueToIntMACRcvError */ + mac_stats->MulticastFramesXmittedOK = mib->tx_multicast; + mac_stats->BroadcastFramesXmittedOK = mib->tx_broadcast; + /* mac_stats->FramesWithExcessiveDeferral */ + mac_stats->MulticastFramesReceivedOK = mib->rx_multicast; + mac_stats->BroadcastFramesReceivedOK = mib->rx_broadcast; + /* mac_stats->InRangeLengthErrors */ + /* mac_stats->OutOfRangeLengthField */ + mac_stats->FrameTooLongErrors = mib->rx_oversize_errors; +} + +static void +yt921x_dsa_get_eth_ctrl_stats(struct dsa_switch *ds, int port, + struct ethtool_eth_ctrl_stats *ctrl_stats) +{ + struct yt921x_priv *priv = to_yt921x_priv(ds); + struct yt921x_port *pp = &priv->ports[port]; + struct yt921x_mib *mib = &pp->mib; + + mutex_lock(&priv->reg_lock); + yt921x_read_mib(priv, port); + mutex_unlock(&priv->reg_lock); + + ctrl_stats->MACControlFramesTransmitted = mib->tx_pause; + ctrl_stats->MACControlFramesReceived = mib->rx_pause; + /* ctrl_stats->UnsupportedOpcodesReceived */ +} + +static const struct ethtool_rmon_hist_range yt921x_rmon_ranges[] = { + { 0, 64 }, + { 65, 127 }, + { 128, 255 }, + { 256, 511 }, + { 512, 1023 }, + { 1024, 1518 }, + { 1519, YT921X_FRAME_SIZE_MAX }, + {} +}; + +static void +yt921x_dsa_get_rmon_stats(struct dsa_switch *ds, int port, + struct ethtool_rmon_stats *rmon_stats, + const struct ethtool_rmon_hist_range **ranges) +{ + struct yt921x_priv *priv = to_yt921x_priv(ds); + struct yt921x_port *pp = &priv->ports[port]; + struct yt921x_mib *mib = &pp->mib; + + mutex_lock(&priv->reg_lock); + yt921x_read_mib(priv, port); + mutex_unlock(&priv->reg_lock); + + *ranges = yt921x_rmon_ranges; + + rmon_stats->undersize_pkts = mib->rx_undersize_errors; + rmon_stats->oversize_pkts = mib->rx_oversize_errors; + rmon_stats->fragments = mib->rx_alignment_errors; + /* rmon_stats->jabbers */ + + rmon_stats->hist[0] = mib->rx_64byte; + rmon_stats->hist[1] = mib->rx_65_127byte; + rmon_stats->hist[2] = mib->rx_128_255byte; + rmon_stats->hist[3] = mib->rx_256_511byte; + rmon_stats->hist[4] = mib->rx_512_1023byte; + rmon_stats->hist[5] = mib->rx_1024_1518byte; + rmon_stats->hist[6] = mib->rx_jumbo; + + rmon_stats->hist_tx[0] = mib->tx_64byte; + rmon_stats->hist_tx[1] = mib->tx_65_127byte; + rmon_stats->hist_tx[2] = mib->tx_128_255byte; + rmon_stats->hist_tx[3] = mib->tx_256_511byte; + rmon_stats->hist_tx[4] = mib->tx_512_1023byte; + rmon_stats->hist_tx[5] = mib->tx_1024_1518byte; + rmon_stats->hist_tx[6] = mib->tx_jumbo; +} + +static void +yt921x_dsa_get_stats64(struct dsa_switch *ds, int port, + struct rtnl_link_stats64 *stats) +{ + struct yt921x_priv *priv = to_yt921x_priv(ds); + struct yt921x_port *pp = &priv->ports[port]; + struct yt921x_mib *mib = &pp->mib; + + stats->rx_length_errors = mib->rx_undersize_errors + + mib->rx_fragment_errors; + stats->rx_over_errors = mib->rx_oversize_errors; + stats->rx_crc_errors = mib->rx_crc_errors; + stats->rx_frame_errors = mib->rx_alignment_errors; + /* stats->rx_fifo_errors */ + /* stats->rx_missed_errors */ + + stats->tx_aborted_errors = mib->tx_aborted_errors; + /* stats->tx_carrier_errors */ + stats->tx_fifo_errors = mib->tx_undersize_errors; + /* stats->tx_heartbeat_errors */ + stats->tx_window_errors = mib->tx_late_collisions; + + stats->rx_packets = pp->rx_frames; + stats->tx_packets = pp->tx_frames; + stats->rx_bytes = mib->rx_good_bytes - ETH_FCS_LEN * stats->rx_packets; + stats->tx_bytes = mib->tx_good_bytes - ETH_FCS_LEN * stats->tx_packets; + stats->rx_errors = stats->rx_length_errors + stats->rx_over_errors + + stats->rx_crc_errors + stats->rx_frame_errors; + stats->tx_errors = stats->tx_aborted_errors + stats->tx_fifo_errors + + stats->tx_window_errors; + stats->rx_dropped = mib->rx_dropped; + /* stats->tx_dropped */ + stats->multicast = mib->rx_multicast; + stats->collisions = mib->tx_collisions; +} + +static void +yt921x_dsa_get_pause_stats(struct dsa_switch *ds, int port, + struct ethtool_pause_stats *pause_stats) +{ + struct yt921x_priv *priv = to_yt921x_priv(ds); + struct yt921x_port *pp = &priv->ports[port]; + struct yt921x_mib *mib = &pp->mib; + + mutex_lock(&priv->reg_lock); + yt921x_read_mib(priv, port); + mutex_unlock(&priv->reg_lock); + + pause_stats->tx_pause_frames = mib->tx_pause; + pause_stats->rx_pause_frames = mib->rx_pause; +} + +static int +yt921x_set_eee(struct yt921x_priv *priv, int port, struct ethtool_keee *e) +{ + /* Poor datasheet for EEE operations; don't ask if you are confused */ + + bool enable = e->eee_enabled; + u16 new_mask; + int res; + + /* Enable / disable global EEE */ + new_mask = priv->eee_ports_mask; + new_mask &= ~BIT(port); + new_mask |= !enable ? 0 : BIT(port); + + if (!!new_mask != !!priv->eee_ports_mask) { + res = yt921x_reg_toggle_bits(priv, YT921X_PON_STRAP_FUNC, + YT921X_PON_STRAP_EEE, !!new_mask); + if (res) + return res; + res = yt921x_reg_toggle_bits(priv, YT921X_PON_STRAP_VAL, + YT921X_PON_STRAP_EEE, !!new_mask); + if (res) + return res; + } + + priv->eee_ports_mask = new_mask; + + /* Enable / disable port EEE */ + res = yt921x_reg_toggle_bits(priv, YT921X_EEE_CTRL, + YT921X_EEE_CTRL_ENn(port), enable); + if (res) + return res; + res = yt921x_reg_toggle_bits(priv, YT921X_EEEn_VAL(port), + YT921X_EEE_VAL_DATA, enable); + if (res) + return res; + + return 0; +} + +static int +yt921x_dsa_set_mac_eee(struct dsa_switch *ds, int port, struct ethtool_keee *e) +{ + struct yt921x_priv *priv = to_yt921x_priv(ds); + int res; + + mutex_lock(&priv->reg_lock); + res = yt921x_set_eee(priv, port, e); + mutex_unlock(&priv->reg_lock); + + return res; +} + +static int +yt921x_dsa_port_change_mtu(struct dsa_switch *ds, int port, int new_mtu) +{ + /* Only serves as packet filter, since the frame size is always set to + * maximum after reset + */ + + struct yt921x_priv *priv = to_yt921x_priv(ds); + struct dsa_port *dp = dsa_to_port(ds, port); + int frame_size; + int res; + + frame_size = new_mtu + ETH_HLEN + ETH_FCS_LEN; + if (dsa_port_is_cpu(dp)) + frame_size += YT921X_TAG_LEN; + + mutex_lock(&priv->reg_lock); + res = yt921x_reg_update_bits(priv, YT921X_MACn_FRAME(port), + YT921X_MAC_FRAME_SIZE_M, + YT921X_MAC_FRAME_SIZE(frame_size)); + mutex_unlock(&priv->reg_lock); + + return res; +} + +static int yt921x_dsa_port_max_mtu(struct dsa_switch *ds, int port) +{ + /* Only called for user ports, exclude tag len here */ + return YT921X_FRAME_SIZE_MAX - ETH_HLEN - ETH_FCS_LEN - YT921X_TAG_LEN; +} + +static int +yt921x_mirror_del(struct yt921x_priv *priv, int port, bool ingress) +{ + u32 mask; + + if (ingress) + mask = YT921X_MIRROR_IGR_PORTn(port); + else + mask = YT921X_MIRROR_EGR_PORTn(port); + return yt921x_reg_clear_bits(priv, YT921X_MIRROR, mask); +} + +static int +yt921x_mirror_add(struct yt921x_priv *priv, int port, bool ingress, + int to_local_port, struct netlink_ext_ack *extack) +{ + u32 srcs; + u32 ctrl; + u32 val; + u32 dst; + int res; + + if (ingress) + srcs = YT921X_MIRROR_IGR_PORTn(port); + else + srcs = YT921X_MIRROR_EGR_PORTn(port); + dst = YT921X_MIRROR_PORT(to_local_port); + + res = yt921x_reg_read(priv, YT921X_MIRROR, &val); + if (res) + return res; + + /* other mirror tasks & different dst port -> conflict */ + if ((val & ~srcs & (YT921X_MIRROR_EGR_PORTS_M | + YT921X_MIRROR_IGR_PORTS_M)) && + (val & YT921X_MIRROR_PORT_M) != dst) { + NL_SET_ERR_MSG_MOD(extack, + "Sniffer port is already configured, delete existing rules & retry"); + return -EBUSY; + } + + ctrl = val & ~YT921X_MIRROR_PORT_M; + ctrl |= srcs; + ctrl |= dst; + + if (ctrl == val) + return 0; + + return yt921x_reg_write(priv, YT921X_MIRROR, ctrl); +} + +static void +yt921x_dsa_port_mirror_del(struct dsa_switch *ds, int port, + struct dsa_mall_mirror_tc_entry *mirror) +{ + struct yt921x_priv *priv = to_yt921x_priv(ds); + struct device *dev = to_device(priv); + int res; + + mutex_lock(&priv->reg_lock); + res = yt921x_mirror_del(priv, port, mirror->ingress); + mutex_unlock(&priv->reg_lock); + + if (res) + dev_err(dev, "Failed to %s port %d: %i\n", "unmirror", + port, res); +} + +static int +yt921x_dsa_port_mirror_add(struct dsa_switch *ds, int port, + struct dsa_mall_mirror_tc_entry *mirror, + bool ingress, struct netlink_ext_ack *extack) +{ + struct yt921x_priv *priv = to_yt921x_priv(ds); + int res; + + mutex_lock(&priv->reg_lock); + res = yt921x_mirror_add(priv, port, ingress, + mirror->to_local_port, extack); + mutex_unlock(&priv->reg_lock); + + return res; +} + +static int yt921x_fdb_wait(struct yt921x_priv *priv, u32 *valp) +{ + struct device *dev = to_device(priv); + u32 val = YT921X_FDB_RESULT_DONE; + int res; + + res = yt921x_reg_wait(priv, YT921X_FDB_RESULT, YT921X_FDB_RESULT_DONE, + &val); + if (res) { + dev_err(dev, "FDB probably stucked\n"); + return res; + } + + *valp = val; + return 0; +} + +static int +yt921x_fdb_in01(struct yt921x_priv *priv, const unsigned char *addr, + u16 vid, u32 ctrl1) +{ + u32 ctrl; + int res; + + ctrl = (addr[0] << 24) | (addr[1] << 16) | (addr[2] << 8) | addr[3]; + res = yt921x_reg_write(priv, YT921X_FDB_IN0, ctrl); + if (res) + return res; + + ctrl = ctrl1 | YT921X_FDB_IO1_FID(vid) | (addr[4] << 8) | addr[5]; + return yt921x_reg_write(priv, YT921X_FDB_IN1, ctrl); +} + +static int +yt921x_fdb_has(struct yt921x_priv *priv, const unsigned char *addr, u16 vid, + u16 *indexp) +{ + u32 ctrl; + u32 val; + int res; + + res = yt921x_fdb_in01(priv, addr, vid, 0); + if (res) + return res; + + ctrl = 0; + res = yt921x_reg_write(priv, YT921X_FDB_IN2, ctrl); + if (res) + return res; + + ctrl = YT921X_FDB_OP_OP_GET_ONE | YT921X_FDB_OP_START; + res = yt921x_reg_write(priv, YT921X_FDB_OP, ctrl); + if (res) + return res; + + res = yt921x_fdb_wait(priv, &val); + if (res) + return res; + if (val & YT921X_FDB_RESULT_NOTFOUND) { + *indexp = YT921X_FDB_NUM; + return 0; + } + + *indexp = FIELD_GET(YT921X_FDB_RESULT_INDEX_M, val); + return 0; +} + +static int +yt921x_fdb_read(struct yt921x_priv *priv, unsigned char *addr, u16 *vidp, + u16 *ports_maskp, u16 *indexp, u8 *statusp) +{ + struct device *dev = to_device(priv); + u16 index; + u32 data0; + u32 data1; + u32 data2; + u32 val; + int res; + + res = yt921x_fdb_wait(priv, &val); + if (res) + return res; + if (val & YT921X_FDB_RESULT_NOTFOUND) { + *ports_maskp = 0; + return 0; + } + index = FIELD_GET(YT921X_FDB_RESULT_INDEX_M, val); + + res = yt921x_reg_read(priv, YT921X_FDB_OUT1, &data1); + if (res) + return res; + if ((data1 & YT921X_FDB_IO1_STATUS_M) == + YT921X_FDB_IO1_STATUS_INVALID) { + *ports_maskp = 0; + return 0; + } + + res = yt921x_reg_read(priv, YT921X_FDB_OUT0, &data0); + if (res) + return res; + res = yt921x_reg_read(priv, YT921X_FDB_OUT2, &data2); + if (res) + return res; + + addr[0] = data0 >> 24; + addr[1] = data0 >> 16; + addr[2] = data0 >> 8; + addr[3] = data0; + addr[4] = data1 >> 8; + addr[5] = data1; + *vidp = FIELD_GET(YT921X_FDB_IO1_FID_M, data1); + *indexp = index; + *ports_maskp = FIELD_GET(YT921X_FDB_IO2_EGR_PORTS_M, data2); + *statusp = FIELD_GET(YT921X_FDB_IO1_STATUS_M, data1); + + dev_dbg(dev, + "%s: index 0x%x, mac %02x:%02x:%02x:%02x:%02x:%02x, vid %d, ports 0x%x, status %d\n", + __func__, *indexp, addr[0], addr[1], addr[2], addr[3], + addr[4], addr[5], *vidp, *ports_maskp, *statusp); + return 0; +} + +static int +yt921x_fdb_dump(struct yt921x_priv *priv, u16 ports_mask, + dsa_fdb_dump_cb_t *cb, void *data) +{ + unsigned char addr[ETH_ALEN]; + u8 status; + u16 pmask; + u16 index; + u32 ctrl; + u16 vid; + int res; + + ctrl = YT921X_FDB_OP_INDEX(0) | YT921X_FDB_OP_MODE_INDEX | + YT921X_FDB_OP_OP_GET_ONE | YT921X_FDB_OP_START; + res = yt921x_reg_write(priv, YT921X_FDB_OP, ctrl); + if (res) + return res; + res = yt921x_fdb_read(priv, addr, &vid, &pmask, &index, &status); + if (res) + return res; + if ((pmask & ports_mask) && !is_multicast_ether_addr(addr)) { + res = cb(addr, vid, + status == YT921X_FDB_ENTRY_STATUS_STATIC, data); + if (res) + return res; + } + + ctrl = YT921X_FDB_IO2_EGR_PORTS(ports_mask); + res = yt921x_reg_write(priv, YT921X_FDB_IN2, ctrl); + if (res) + return res; + + index = 0; + do { + ctrl = YT921X_FDB_OP_INDEX(index) | YT921X_FDB_OP_MODE_INDEX | + YT921X_FDB_OP_NEXT_TYPE_UCAST_PORT | + YT921X_FDB_OP_OP_GET_NEXT | YT921X_FDB_OP_START; + res = yt921x_reg_write(priv, YT921X_FDB_OP, ctrl); + if (res) + return res; + + res = yt921x_fdb_read(priv, addr, &vid, &pmask, &index, + &status); + if (res) + return res; + if (!pmask) + break; + + if ((pmask & ports_mask) && !is_multicast_ether_addr(addr)) { + res = cb(addr, vid, + status == YT921X_FDB_ENTRY_STATUS_STATIC, + data); + if (res) + return res; + } + + /* Never call GET_NEXT with 4095, otherwise it will hang + * forever until a reset! + */ + } while (index < YT921X_FDB_NUM - 1); + + return 0; +} + +static int +yt921x_fdb_flush_raw(struct yt921x_priv *priv, u16 ports_mask, u16 vid, + bool flush_static) +{ + u32 ctrl; + u32 val; + int res; + + if (vid < 4096) { + ctrl = YT921X_FDB_IO1_FID(vid); + res = yt921x_reg_write(priv, YT921X_FDB_IN1, ctrl); + if (res) + return res; + } + + ctrl = YT921X_FDB_IO2_EGR_PORTS(ports_mask); + res = yt921x_reg_write(priv, YT921X_FDB_IN2, ctrl); + if (res) + return res; + + ctrl = YT921X_FDB_OP_OP_FLUSH | YT921X_FDB_OP_START; + if (vid >= 4096) + ctrl |= YT921X_FDB_OP_FLUSH_PORT; + else + ctrl |= YT921X_FDB_OP_FLUSH_PORT_VID; + if (flush_static) + ctrl |= YT921X_FDB_OP_FLUSH_STATIC; + res = yt921x_reg_write(priv, YT921X_FDB_OP, ctrl); + if (res) + return res; + + res = yt921x_fdb_wait(priv, &val); + if (res) + return res; + + return 0; +} + +static int +yt921x_fdb_flush_port(struct yt921x_priv *priv, int port, bool flush_static) +{ + return yt921x_fdb_flush_raw(priv, BIT(port), 4096, flush_static); +} + +static int +yt921x_fdb_add_index_in12(struct yt921x_priv *priv, u16 index, u16 ctrl1, + u16 ctrl2) +{ + u32 ctrl; + u32 val; + int res; + + res = yt921x_reg_write(priv, YT921X_FDB_IN1, ctrl1); + if (res) + return res; + res = yt921x_reg_write(priv, YT921X_FDB_IN2, ctrl2); + if (res) + return res; + + ctrl = YT921X_FDB_OP_INDEX(index) | YT921X_FDB_OP_MODE_INDEX | + YT921X_FDB_OP_OP_ADD | YT921X_FDB_OP_START; + res = yt921x_reg_write(priv, YT921X_FDB_OP, ctrl); + if (res) + return res; + + return yt921x_fdb_wait(priv, &val); +} + +static int +yt921x_fdb_add(struct yt921x_priv *priv, const unsigned char *addr, u16 vid, + u16 ports_mask) +{ + u32 ctrl; + u32 val; + int res; + + ctrl = YT921X_FDB_IO1_STATUS_STATIC; + res = yt921x_fdb_in01(priv, addr, vid, ctrl); + if (res) + return res; + + ctrl = YT921X_FDB_IO2_EGR_PORTS(ports_mask); + res = yt921x_reg_write(priv, YT921X_FDB_IN2, ctrl); + if (res) + return res; + + ctrl = YT921X_FDB_OP_OP_ADD | YT921X_FDB_OP_START; + res = yt921x_reg_write(priv, YT921X_FDB_OP, ctrl); + if (res) + return res; + + return yt921x_fdb_wait(priv, &val); +} + +static int +yt921x_fdb_leave(struct yt921x_priv *priv, const unsigned char *addr, + u16 vid, u16 ports_mask) +{ + u16 index; + u32 ctrl1; + u32 ctrl2; + u32 ctrl; + u32 val2; + u32 val; + int res; + + /* Check for presence */ + res = yt921x_fdb_has(priv, addr, vid, &index); + if (res) + return res; + if (index >= YT921X_FDB_NUM) + return 0; + + /* Check if action required */ + res = yt921x_reg_read(priv, YT921X_FDB_OUT2, &val2); + if (res) + return res; + + ctrl2 = val2 & ~YT921X_FDB_IO2_EGR_PORTS(ports_mask); + if (ctrl2 == val2) + return 0; + if (!(ctrl2 & YT921X_FDB_IO2_EGR_PORTS_M)) { + ctrl = YT921X_FDB_OP_OP_DEL | YT921X_FDB_OP_START; + res = yt921x_reg_write(priv, YT921X_FDB_OP, ctrl); + if (res) + return res; + + return yt921x_fdb_wait(priv, &val); + } + + res = yt921x_reg_read(priv, YT921X_FDB_OUT1, &ctrl1); + if (res) + return res; + + return yt921x_fdb_add_index_in12(priv, index, ctrl1, ctrl2); +} + +static int +yt921x_fdb_join(struct yt921x_priv *priv, const unsigned char *addr, u16 vid, + u16 ports_mask) +{ + u16 index; + u32 ctrl1; + u32 ctrl2; + u32 val1; + u32 val2; + int res; + + /* Check for presence */ + res = yt921x_fdb_has(priv, addr, vid, &index); + if (res) + return res; + if (index >= YT921X_FDB_NUM) + return yt921x_fdb_add(priv, addr, vid, ports_mask); + + /* Check if action required */ + res = yt921x_reg_read(priv, YT921X_FDB_OUT1, &val1); + if (res) + return res; + res = yt921x_reg_read(priv, YT921X_FDB_OUT2, &val2); + if (res) + return res; + + ctrl1 = val1 & ~YT921X_FDB_IO1_STATUS_M; + ctrl1 |= YT921X_FDB_IO1_STATUS_STATIC; + ctrl2 = val2 | YT921X_FDB_IO2_EGR_PORTS(ports_mask); + if (ctrl1 == val1 && ctrl2 == val2) + return 0; + + return yt921x_fdb_add_index_in12(priv, index, ctrl1, ctrl2); +} + +static int +yt921x_dsa_port_fdb_dump(struct dsa_switch *ds, int port, + dsa_fdb_dump_cb_t *cb, void *data) +{ + struct yt921x_priv *priv = to_yt921x_priv(ds); + int res; + + mutex_lock(&priv->reg_lock); + /* Hardware FDB is shared for fdb and mdb, "bridge fdb show" + * only wants to see unicast + */ + res = yt921x_fdb_dump(priv, BIT(port), cb, data); + mutex_unlock(&priv->reg_lock); + + return res; +} + +static void yt921x_dsa_port_fast_age(struct dsa_switch *ds, int port) +{ + struct yt921x_priv *priv = to_yt921x_priv(ds); + struct device *dev = to_device(priv); + int res; + + mutex_lock(&priv->reg_lock); + res = yt921x_fdb_flush_port(priv, port, false); + mutex_unlock(&priv->reg_lock); + + if (res) + dev_err(dev, "Failed to %s port %d: %i\n", "clear FDB for", + port, res); +} + +static int +yt921x_dsa_set_ageing_time(struct dsa_switch *ds, unsigned int msecs) +{ + struct yt921x_priv *priv = to_yt921x_priv(ds); + u32 ctrl; + int res; + + /* AGEING reg is set in 5s step */ + ctrl = clamp(msecs / 5000, 1, U16_MAX); + + mutex_lock(&priv->reg_lock); + res = yt921x_reg_write(priv, YT921X_AGEING, ctrl); + mutex_unlock(&priv->reg_lock); + + return res; +} + +static int +yt921x_dsa_port_fdb_del(struct dsa_switch *ds, int port, + const unsigned char *addr, u16 vid, struct dsa_db db) +{ + struct yt921x_priv *priv = to_yt921x_priv(ds); + int res; + + mutex_lock(&priv->reg_lock); + res = yt921x_fdb_leave(priv, addr, vid, BIT(port)); + mutex_unlock(&priv->reg_lock); + + return res; +} + +static int +yt921x_dsa_port_fdb_add(struct dsa_switch *ds, int port, + const unsigned char *addr, u16 vid, struct dsa_db db) +{ + struct yt921x_priv *priv = to_yt921x_priv(ds); + int res; + + mutex_lock(&priv->reg_lock); + res = yt921x_fdb_join(priv, addr, vid, BIT(port)); + mutex_unlock(&priv->reg_lock); + + return res; +} + +static int +yt921x_dsa_port_mdb_del(struct dsa_switch *ds, int port, + const struct switchdev_obj_port_mdb *mdb, + struct dsa_db db) +{ + struct yt921x_priv *priv = to_yt921x_priv(ds); + const unsigned char *addr = mdb->addr; + u16 vid = mdb->vid; + int res; + + mutex_lock(&priv->reg_lock); + res = yt921x_fdb_leave(priv, addr, vid, BIT(port)); + mutex_unlock(&priv->reg_lock); + + return res; +} + +static int +yt921x_dsa_port_mdb_add(struct dsa_switch *ds, int port, + const struct switchdev_obj_port_mdb *mdb, + struct dsa_db db) +{ + struct yt921x_priv *priv = to_yt921x_priv(ds); + const unsigned char *addr = mdb->addr; + u16 vid = mdb->vid; + int res; + + mutex_lock(&priv->reg_lock); + res = yt921x_fdb_join(priv, addr, vid, BIT(port)); + mutex_unlock(&priv->reg_lock); + + return res; +} + +static int +yt921x_port_set_pvid(struct yt921x_priv *priv, int port, u16 vid) +{ + u32 mask; + u32 ctrl; + + mask = YT921X_PORT_VLAN_CTRL_CVID_M; + ctrl = YT921X_PORT_VLAN_CTRL_CVID(vid); + return yt921x_reg_update_bits(priv, YT921X_PORTn_VLAN_CTRL(port), + mask, ctrl); +} + +static int +yt921x_vlan_filtering(struct yt921x_priv *priv, int port, bool vlan_filtering) +{ + struct dsa_port *dp = dsa_to_port(&priv->ds, port); + struct net_device *bdev; + u16 pvid; + u32 mask; + u32 ctrl; + int res; + + bdev = dsa_port_bridge_dev_get(dp); + + if (!bdev || !vlan_filtering) + pvid = YT921X_VID_UNWARE; + else + br_vlan_get_pvid(bdev, &pvid); + res = yt921x_port_set_pvid(priv, port, pvid); + if (res) + return res; + + mask = YT921X_PORT_VLAN_CTRL1_CVLAN_DROP_TAGGED | + YT921X_PORT_VLAN_CTRL1_CVLAN_DROP_UNTAGGED; + ctrl = 0; + /* Do not drop tagged frames here; let VLAN_IGR_FILTER do it */ + if (vlan_filtering && !pvid) + ctrl |= YT921X_PORT_VLAN_CTRL1_CVLAN_DROP_UNTAGGED; + res = yt921x_reg_update_bits(priv, YT921X_PORTn_VLAN_CTRL1(port), + mask, ctrl); + if (res) + return res; + + res = yt921x_reg_toggle_bits(priv, YT921X_VLAN_IGR_FILTER, + YT921X_VLAN_IGR_FILTER_PORTn(port), + vlan_filtering); + if (res) + return res; + + /* Turn on / off VLAN awareness */ + mask = YT921X_PORT_IGR_TPIDn_CTAG_M; + if (!vlan_filtering) + ctrl = 0; + else + ctrl = YT921X_PORT_IGR_TPIDn_CTAG(0); + res = yt921x_reg_update_bits(priv, YT921X_PORTn_IGR_TPID(port), + mask, ctrl); + if (res) + return res; + + return 0; +} + +static int +yt921x_vlan_del(struct yt921x_priv *priv, int port, u16 vid) +{ + u64 mask64; + + mask64 = YT921X_VLAN_CTRL_PORTS(port) | + YT921X_VLAN_CTRL_UNTAG_PORTn(port); + + return yt921x_reg64_clear_bits(priv, YT921X_VLANn_CTRL(vid), mask64); +} + +static int +yt921x_vlan_add(struct yt921x_priv *priv, int port, u16 vid, bool untagged) +{ + u64 mask64; + u64 ctrl64; + + mask64 = YT921X_VLAN_CTRL_PORTn(port) | + YT921X_VLAN_CTRL_PORTS(priv->cpu_ports_mask); + ctrl64 = mask64; + + mask64 |= YT921X_VLAN_CTRL_UNTAG_PORTn(port); + if (untagged) + ctrl64 |= YT921X_VLAN_CTRL_UNTAG_PORTn(port); + + return yt921x_reg64_update_bits(priv, YT921X_VLANn_CTRL(vid), + mask64, ctrl64); +} + +static int +yt921x_pvid_clear(struct yt921x_priv *priv, int port) +{ + struct dsa_port *dp = dsa_to_port(&priv->ds, port); + bool vlan_filtering; + u32 mask; + int res; + + vlan_filtering = dsa_port_is_vlan_filtering(dp); + + res = yt921x_port_set_pvid(priv, port, + vlan_filtering ? 0 : YT921X_VID_UNWARE); + if (res) + return res; + + if (vlan_filtering) { + mask = YT921X_PORT_VLAN_CTRL1_CVLAN_DROP_UNTAGGED; + res = yt921x_reg_set_bits(priv, YT921X_PORTn_VLAN_CTRL1(port), + mask); + if (res) + return res; + } + + return 0; +} + +static int +yt921x_pvid_set(struct yt921x_priv *priv, int port, u16 vid) +{ + struct dsa_port *dp = dsa_to_port(&priv->ds, port); + bool vlan_filtering; + u32 mask; + int res; + + vlan_filtering = dsa_port_is_vlan_filtering(dp); + + if (vlan_filtering) { + res = yt921x_port_set_pvid(priv, port, vid); + if (res) + return res; + } + + mask = YT921X_PORT_VLAN_CTRL1_CVLAN_DROP_UNTAGGED; + res = yt921x_reg_clear_bits(priv, YT921X_PORTn_VLAN_CTRL1(port), mask); + if (res) + return res; + + return 0; +} + +static int +yt921x_dsa_port_vlan_filtering(struct dsa_switch *ds, int port, + bool vlan_filtering, + struct netlink_ext_ack *extack) +{ + struct yt921x_priv *priv = to_yt921x_priv(ds); + int res; + + if (dsa_is_cpu_port(ds, port)) + return 0; + + mutex_lock(&priv->reg_lock); + res = yt921x_vlan_filtering(priv, port, vlan_filtering); + mutex_unlock(&priv->reg_lock); + + return res; +} + +static int +yt921x_dsa_port_vlan_del(struct dsa_switch *ds, int port, + const struct switchdev_obj_port_vlan *vlan) +{ + struct yt921x_priv *priv = to_yt921x_priv(ds); + u16 vid = vlan->vid; + u16 pvid; + int res; + + if (dsa_is_cpu_port(ds, port)) + return 0; + + mutex_lock(&priv->reg_lock); + do { + struct dsa_port *dp = dsa_to_port(ds, port); + struct net_device *bdev; + + res = yt921x_vlan_del(priv, port, vid); + if (res) + break; + + bdev = dsa_port_bridge_dev_get(dp); + if (bdev) { + br_vlan_get_pvid(bdev, &pvid); + if (pvid == vid) + res = yt921x_pvid_clear(priv, port); + } + } while (0); + mutex_unlock(&priv->reg_lock); + + return res; +} + +static int +yt921x_dsa_port_vlan_add(struct dsa_switch *ds, int port, + const struct switchdev_obj_port_vlan *vlan, + struct netlink_ext_ack *extack) +{ + struct yt921x_priv *priv = to_yt921x_priv(ds); + u16 vid = vlan->vid; + u16 pvid; + int res; + + /* CPU port is supposed to be a member of every VLAN; see + * yt921x_vlan_add() and yt921x_port_setup() + */ + if (dsa_is_cpu_port(ds, port)) + return 0; + + mutex_lock(&priv->reg_lock); + do { + struct dsa_port *dp = dsa_to_port(ds, port); + struct net_device *bdev; + + res = yt921x_vlan_add(priv, port, vid, + vlan->flags & BRIDGE_VLAN_INFO_UNTAGGED); + if (res) + break; + + bdev = dsa_port_bridge_dev_get(dp); + if (bdev) { + if (vlan->flags & BRIDGE_VLAN_INFO_PVID) { + res = yt921x_pvid_set(priv, port, vid); + } else { + br_vlan_get_pvid(bdev, &pvid); + if (pvid == vid) + res = yt921x_pvid_clear(priv, port); + } + } + } while (0); + mutex_unlock(&priv->reg_lock); + + return res; +} + +static int yt921x_userport_standalone(struct yt921x_priv *priv, int port) +{ + u32 mask; + u32 ctrl; + int res; + + ctrl = ~priv->cpu_ports_mask; + res = yt921x_reg_write(priv, YT921X_PORTn_ISOLATION(port), ctrl); + if (res) + return res; + + /* Turn off FDB learning to prevent FDB pollution */ + mask = YT921X_PORT_LEARN_DIS; + res = yt921x_reg_set_bits(priv, YT921X_PORTn_LEARN(port), mask); + if (res) + return res; + + /* Turn off VLAN awareness */ + mask = YT921X_PORT_IGR_TPIDn_CTAG_M; + res = yt921x_reg_clear_bits(priv, YT921X_PORTn_IGR_TPID(port), mask); + if (res) + return res; + + /* Unrelated since learning is off and all packets are trapped; + * set it anyway + */ + res = yt921x_port_set_pvid(priv, port, YT921X_VID_UNWARE); + if (res) + return res; + + return 0; +} + +static int yt921x_userport_bridge(struct yt921x_priv *priv, int port) +{ + u32 mask; + int res; + + mask = YT921X_PORT_LEARN_DIS; + res = yt921x_reg_clear_bits(priv, YT921X_PORTn_LEARN(port), mask); + if (res) + return res; + + return 0; +} + +static int yt921x_isolate(struct yt921x_priv *priv, int port) +{ + u32 mask; + int res; + + mask = BIT(port); + for (int i = 0; i < YT921X_PORT_NUM; i++) { + if ((BIT(i) & priv->cpu_ports_mask) || i == port) + continue; + + res = yt921x_reg_set_bits(priv, YT921X_PORTn_ISOLATION(i), + mask); + if (res) + return res; + } + + return 0; +} + +/* Make sure to include the CPU port in ports_mask, or your bridge will + * not have it. + */ +static int yt921x_bridge(struct yt921x_priv *priv, u16 ports_mask) +{ + unsigned long targets_mask = ports_mask & ~priv->cpu_ports_mask; + u32 isolated_mask; + u32 ctrl; + int port; + int res; + + isolated_mask = 0; + for_each_set_bit(port, &targets_mask, YT921X_PORT_NUM) { + struct yt921x_port *pp = &priv->ports[port]; + + if (pp->isolated) + isolated_mask |= BIT(port); + } + + /* Block from non-cpu bridge ports ... */ + for_each_set_bit(port, &targets_mask, YT921X_PORT_NUM) { + struct yt921x_port *pp = &priv->ports[port]; + + /* to non-bridge ports */ + ctrl = ~ports_mask; + /* to isolated ports when isolated */ + if (pp->isolated) + ctrl |= isolated_mask; + /* to itself when non-hairpin */ + if (!pp->hairpin) + ctrl |= BIT(port); + else + ctrl &= ~BIT(port); + + res = yt921x_reg_write(priv, YT921X_PORTn_ISOLATION(port), + ctrl); + if (res) + return res; + } + + return 0; +} + +static int yt921x_bridge_leave(struct yt921x_priv *priv, int port) +{ + int res; + + res = yt921x_userport_standalone(priv, port); + if (res) + return res; + + res = yt921x_isolate(priv, port); + if (res) + return res; + + return 0; +} + +static int +yt921x_bridge_join(struct yt921x_priv *priv, int port, u16 ports_mask) +{ + int res; + + res = yt921x_userport_bridge(priv, port); + if (res) + return res; + + res = yt921x_bridge(priv, ports_mask); + if (res) + return res; + + return 0; +} + +static u32 +dsa_bridge_ports(struct dsa_switch *ds, const struct net_device *bdev) +{ + struct dsa_port *dp; + u32 mask = 0; + + dsa_switch_for_each_user_port(dp, ds) + if (dsa_port_offloads_bridge_dev(dp, bdev)) + mask |= BIT(dp->index); + + return mask; +} + +static int +yt921x_bridge_flags(struct yt921x_priv *priv, int port, + struct switchdev_brport_flags flags) +{ + struct yt921x_port *pp = &priv->ports[port]; + bool do_flush; + u32 mask; + int res; + + if (flags.mask & BR_LEARNING) { + bool learning = flags.val & BR_LEARNING; + + mask = YT921X_PORT_LEARN_DIS; + res = yt921x_reg_toggle_bits(priv, YT921X_PORTn_LEARN(port), + mask, !learning); + if (res) + return res; + } + + /* BR_FLOOD, BR_MCAST_FLOOD: see the comment where ACT_UNK_ACTn_TRAP + * is set + */ + + /* BR_BCAST_FLOOD: we can filter bcast, but cannot trap them */ + + do_flush = false; + if (flags.mask & BR_HAIRPIN_MODE) { + pp->hairpin = flags.val & BR_HAIRPIN_MODE; + do_flush = true; + } + if (flags.mask & BR_ISOLATED) { + pp->isolated = flags.val & BR_ISOLATED; + do_flush = true; + } + if (do_flush) { + struct dsa_switch *ds = &priv->ds; + struct dsa_port *dp = dsa_to_port(ds, port); + struct net_device *bdev; + + bdev = dsa_port_bridge_dev_get(dp); + if (bdev) { + u32 ports_mask; + + ports_mask = dsa_bridge_ports(ds, bdev); + ports_mask |= priv->cpu_ports_mask; + res = yt921x_bridge(priv, ports_mask); + if (res) + return res; + } + } + + return 0; +} + +static int +yt921x_dsa_port_pre_bridge_flags(struct dsa_switch *ds, int port, + struct switchdev_brport_flags flags, + struct netlink_ext_ack *extack) +{ + if (flags.mask & ~(BR_HAIRPIN_MODE | BR_LEARNING | BR_FLOOD | + BR_MCAST_FLOOD | BR_ISOLATED)) + return -EINVAL; + return 0; +} + +static int +yt921x_dsa_port_bridge_flags(struct dsa_switch *ds, int port, + struct switchdev_brport_flags flags, + struct netlink_ext_ack *extack) +{ + struct yt921x_priv *priv = to_yt921x_priv(ds); + int res; + + if (dsa_is_cpu_port(ds, port)) + return 0; + + mutex_lock(&priv->reg_lock); + res = yt921x_bridge_flags(priv, port, flags); + mutex_unlock(&priv->reg_lock); + + return res; +} + +static void +yt921x_dsa_port_bridge_leave(struct dsa_switch *ds, int port, + struct dsa_bridge bridge) +{ + struct yt921x_priv *priv = to_yt921x_priv(ds); + struct device *dev = to_device(priv); + int res; + + if (dsa_is_cpu_port(ds, port)) + return; + + mutex_lock(&priv->reg_lock); + res = yt921x_bridge_leave(priv, port); + mutex_unlock(&priv->reg_lock); + + if (res) + dev_err(dev, "Failed to %s port %d: %i\n", "unbridge", + port, res); +} + +static int +yt921x_dsa_port_bridge_join(struct dsa_switch *ds, int port, + struct dsa_bridge bridge, bool *tx_fwd_offload, + struct netlink_ext_ack *extack) +{ + struct yt921x_priv *priv = to_yt921x_priv(ds); + u16 ports_mask; + int res; + + if (dsa_is_cpu_port(ds, port)) + return 0; + + ports_mask = dsa_bridge_ports(ds, bridge.dev); + ports_mask |= priv->cpu_ports_mask; + + mutex_lock(&priv->reg_lock); + res = yt921x_bridge_join(priv, port, ports_mask); + mutex_unlock(&priv->reg_lock); + + return res; +} + +static int yt921x_port_down(struct yt921x_priv *priv, int port) +{ + u32 mask; + int res; + + mask = YT921X_PORT_LINK | YT921X_PORT_RX_MAC_EN | YT921X_PORT_TX_MAC_EN; + res = yt921x_reg_clear_bits(priv, YT921X_PORTn_CTRL(port), mask); + if (res) + return res; + + if (yt921x_port_is_external(port)) { + mask = YT921X_SERDES_LINK; + res = yt921x_reg_clear_bits(priv, YT921X_SERDESn(port), mask); + if (res) + return res; + + mask = YT921X_XMII_LINK; + res = yt921x_reg_clear_bits(priv, YT921X_XMIIn(port), mask); + if (res) + return res; + } + + return 0; +} + +static int +yt921x_port_up(struct yt921x_priv *priv, int port, unsigned int mode, + phy_interface_t interface, int speed, int duplex, + bool tx_pause, bool rx_pause) +{ + u32 mask; + u32 ctrl; + int res; + + switch (speed) { + case SPEED_10: + ctrl = YT921X_PORT_SPEED_10; + break; + case SPEED_100: + ctrl = YT921X_PORT_SPEED_100; + break; + case SPEED_1000: + ctrl = YT921X_PORT_SPEED_1000; + break; + case SPEED_2500: + ctrl = YT921X_PORT_SPEED_2500; + break; + case SPEED_10000: + ctrl = YT921X_PORT_SPEED_10000; + break; + default: + return -EINVAL; + } + if (duplex == DUPLEX_FULL) + ctrl |= YT921X_PORT_DUPLEX_FULL; + if (tx_pause) + ctrl |= YT921X_PORT_TX_PAUSE; + if (rx_pause) + ctrl |= YT921X_PORT_RX_PAUSE; + ctrl |= YT921X_PORT_RX_MAC_EN | YT921X_PORT_TX_MAC_EN; + res = yt921x_reg_write(priv, YT921X_PORTn_CTRL(port), ctrl); + if (res) + return res; + + if (yt921x_port_is_external(port)) { + mask = YT921X_SERDES_SPEED_M; + switch (speed) { + case SPEED_10: + ctrl = YT921X_SERDES_SPEED_10; + break; + case SPEED_100: + ctrl = YT921X_SERDES_SPEED_100; + break; + case SPEED_1000: + ctrl = YT921X_SERDES_SPEED_1000; + break; + case SPEED_2500: + ctrl = YT921X_SERDES_SPEED_2500; + break; + case SPEED_10000: + ctrl = YT921X_SERDES_SPEED_10000; + break; + default: + return -EINVAL; + } + mask |= YT921X_SERDES_DUPLEX_FULL; + if (duplex == DUPLEX_FULL) + ctrl |= YT921X_SERDES_DUPLEX_FULL; + mask |= YT921X_SERDES_TX_PAUSE; + if (tx_pause) + ctrl |= YT921X_SERDES_TX_PAUSE; + mask |= YT921X_SERDES_RX_PAUSE; + if (rx_pause) + ctrl |= YT921X_SERDES_RX_PAUSE; + mask |= YT921X_SERDES_LINK; + ctrl |= YT921X_SERDES_LINK; + res = yt921x_reg_update_bits(priv, YT921X_SERDESn(port), + mask, ctrl); + if (res) + return res; + + mask = YT921X_XMII_LINK; + res = yt921x_reg_set_bits(priv, YT921X_XMIIn(port), mask); + if (res) + return res; + + switch (speed) { + case SPEED_10: + ctrl = YT921X_MDIO_POLLING_SPEED_10; + break; + case SPEED_100: + ctrl = YT921X_MDIO_POLLING_SPEED_100; + break; + case SPEED_1000: + ctrl = YT921X_MDIO_POLLING_SPEED_1000; + break; + case SPEED_2500: + ctrl = YT921X_MDIO_POLLING_SPEED_2500; + break; + case SPEED_10000: + ctrl = YT921X_MDIO_POLLING_SPEED_10000; + break; + default: + return -EINVAL; + } + if (duplex == DUPLEX_FULL) + ctrl |= YT921X_MDIO_POLLING_DUPLEX_FULL; + ctrl |= YT921X_MDIO_POLLING_LINK; + res = yt921x_reg_write(priv, YT921X_MDIO_POLLINGn(port), ctrl); + if (res) + return res; + } + + return 0; +} + +static int +yt921x_port_config(struct yt921x_priv *priv, int port, unsigned int mode, + phy_interface_t interface) +{ + struct device *dev = to_device(priv); + u32 mask; + u32 ctrl; + int res; + + if (!yt921x_port_is_external(port)) { + if (interface != PHY_INTERFACE_MODE_INTERNAL) { + dev_err(dev, "Wrong mode %d on port %d\n", + interface, port); + return -EINVAL; + } + return 0; + } + + switch (interface) { + /* SERDES */ + case PHY_INTERFACE_MODE_SGMII: + case PHY_INTERFACE_MODE_100BASEX: + case PHY_INTERFACE_MODE_1000BASEX: + case PHY_INTERFACE_MODE_2500BASEX: + mask = YT921X_SERDES_CTRL_PORTn(port); + res = yt921x_reg_set_bits(priv, YT921X_SERDES_CTRL, mask); + if (res) + return res; + + mask = YT921X_XMII_CTRL_PORTn(port); + res = yt921x_reg_clear_bits(priv, YT921X_XMII_CTRL, mask); + if (res) + return res; + + mask = YT921X_SERDES_MODE_M; + switch (interface) { + case PHY_INTERFACE_MODE_SGMII: + ctrl = YT921X_SERDES_MODE_SGMII; + break; + case PHY_INTERFACE_MODE_100BASEX: + ctrl = YT921X_SERDES_MODE_100BASEX; + break; + case PHY_INTERFACE_MODE_1000BASEX: + ctrl = YT921X_SERDES_MODE_1000BASEX; + break; + case PHY_INTERFACE_MODE_2500BASEX: + ctrl = YT921X_SERDES_MODE_2500BASEX; + break; + default: + return -EINVAL; + } + res = yt921x_reg_update_bits(priv, YT921X_SERDESn(port), + mask, ctrl); + if (res) + return res; + + break; + /* add XMII support here */ + default: + return -EINVAL; + } + + return 0; +} + +static void +yt921x_phylink_mac_link_down(struct phylink_config *config, unsigned int mode, + phy_interface_t interface) +{ + struct dsa_port *dp = dsa_phylink_to_port(config); + struct yt921x_priv *priv = to_yt921x_priv(dp->ds); + int port = dp->index; + int res; + + /* No need to sync; port control block is hold until device remove */ + cancel_delayed_work(&priv->ports[port].mib_read); + + mutex_lock(&priv->reg_lock); + res = yt921x_port_down(priv, port); + mutex_unlock(&priv->reg_lock); + + if (res) + dev_err(dp->ds->dev, "Failed to %s port %d: %i\n", "bring down", + port, res); +} + +static void +yt921x_phylink_mac_link_up(struct phylink_config *config, + struct phy_device *phydev, unsigned int mode, + phy_interface_t interface, int speed, int duplex, + bool tx_pause, bool rx_pause) +{ + struct dsa_port *dp = dsa_phylink_to_port(config); + struct yt921x_priv *priv = to_yt921x_priv(dp->ds); + int port = dp->index; + int res; + + mutex_lock(&priv->reg_lock); + res = yt921x_port_up(priv, port, mode, interface, speed, duplex, + tx_pause, rx_pause); + mutex_unlock(&priv->reg_lock); + + if (res) + dev_err(dp->ds->dev, "Failed to %s port %d: %i\n", "bring up", + port, res); + + schedule_delayed_work(&priv->ports[port].mib_read, 0); +} + +static void +yt921x_phylink_mac_config(struct phylink_config *config, unsigned int mode, + const struct phylink_link_state *state) +{ + struct dsa_port *dp = dsa_phylink_to_port(config); + struct yt921x_priv *priv = to_yt921x_priv(dp->ds); + int port = dp->index; + int res; + + mutex_lock(&priv->reg_lock); + res = yt921x_port_config(priv, port, mode, state->interface); + mutex_unlock(&priv->reg_lock); + + if (res) + dev_err(dp->ds->dev, "Failed to %s port %d: %i\n", "config", + port, res); +} + +static void +yt921x_dsa_phylink_get_caps(struct dsa_switch *ds, int port, + struct phylink_config *config) +{ + struct yt921x_priv *priv = to_yt921x_priv(ds); + const struct yt921x_info *info = priv->info; + + config->mac_capabilities = MAC_ASYM_PAUSE | MAC_SYM_PAUSE | + MAC_10 | MAC_100 | MAC_1000; + + if (info->internal_mask & BIT(port)) { + /* Port 10 for MCU should probably go here too. But since that + * is untested yet, turn it down for the moment by letting it + * fall to the default branch. + */ + __set_bit(PHY_INTERFACE_MODE_INTERNAL, + config->supported_interfaces); + } else if (info->external_mask & BIT(port)) { + /* TODO: external ports may support SERDES only, XMII only, or + * SERDES + XMII depending on the chip. However, we can't get + * the accurate config table due to lack of document, thus + * we simply declare SERDES + XMII and rely on the correctness + * of devicetree for now. + */ + + /* SERDES */ + __set_bit(PHY_INTERFACE_MODE_SGMII, + config->supported_interfaces); + /* REVSGMII (SGMII in PHY role) should go here, once + * PHY_INTERFACE_MODE_REVSGMII is introduced. + */ + __set_bit(PHY_INTERFACE_MODE_100BASEX, + config->supported_interfaces); + __set_bit(PHY_INTERFACE_MODE_1000BASEX, + config->supported_interfaces); + __set_bit(PHY_INTERFACE_MODE_2500BASEX, + config->supported_interfaces); + config->mac_capabilities |= MAC_2500FD; + + /* XMII */ + + /* Not tested. To add support for XMII: + * - Add proper interface modes below + * - Handle them in yt921x_port_config() + */ + } + /* no such port: empty supported_interfaces causes phylink to turn it + * down + */ +} + +static int yt921x_port_setup(struct yt921x_priv *priv, int port) +{ + struct dsa_switch *ds = &priv->ds; + u32 ctrl; + int res; + + res = yt921x_userport_standalone(priv, port); + if (res) + return res; + + if (dsa_is_cpu_port(ds, port)) { + /* Egress of CPU port is supposed to be completely controlled + * via tagging, so set to oneway isolated (drop all packets + * without tag). + */ + ctrl = ~(u32)0; + res = yt921x_reg_write(priv, YT921X_PORTn_ISOLATION(port), + ctrl); + if (res) + return res; + + /* To simplify FDB "isolation" simulation, we also disable + * learning on the CPU port, and let software identify packets + * towarding CPU (either trapped or a static FDB entry is + * matched, no matter which bridge that entry is for), which is + * already done by yt921x_userport_standalone(). As a result, + * VLAN-awareness becomes unrelated on the CPU port (set to + * VLAN-unaware by the way). + */ + } + + return 0; +} + +static enum dsa_tag_protocol +yt921x_dsa_get_tag_protocol(struct dsa_switch *ds, int port, + enum dsa_tag_protocol m) +{ + return DSA_TAG_PROTO_YT921X; +} + +static int yt921x_dsa_port_setup(struct dsa_switch *ds, int port) +{ + struct yt921x_priv *priv = to_yt921x_priv(ds); + int res; + + mutex_lock(&priv->reg_lock); + res = yt921x_port_setup(priv, port); + mutex_unlock(&priv->reg_lock); + + return res; +} + +static int yt921x_edata_wait(struct yt921x_priv *priv, u32 *valp) +{ + u32 val = YT921X_EDATA_DATA_IDLE; + int res; + + res = yt921x_reg_wait(priv, YT921X_EDATA_DATA, + YT921X_EDATA_DATA_STATUS_M, &val); + if (res) + return res; + + *valp = val; + return 0; +} + +static int +yt921x_edata_read_cont(struct yt921x_priv *priv, u8 addr, u8 *valp) +{ + u32 ctrl; + u32 val; + int res; + + ctrl = YT921X_EDATA_CTRL_ADDR(addr) | YT921X_EDATA_CTRL_READ; + res = yt921x_reg_write(priv, YT921X_EDATA_CTRL, ctrl); + if (res) + return res; + res = yt921x_edata_wait(priv, &val); + if (res) + return res; + + *valp = FIELD_GET(YT921X_EDATA_DATA_DATA_M, val); + return 0; +} + +static int yt921x_edata_read(struct yt921x_priv *priv, u8 addr, u8 *valp) +{ + u32 val; + int res; + + res = yt921x_edata_wait(priv, &val); + if (res) + return res; + return yt921x_edata_read_cont(priv, addr, valp); +} + +static int yt921x_chip_detect(struct yt921x_priv *priv) +{ + struct device *dev = to_device(priv); + const struct yt921x_info *info; + u8 extmode; + u32 chipid; + u32 major; + u32 mode; + int res; + + res = yt921x_reg_read(priv, YT921X_CHIP_ID, &chipid); + if (res) + return res; + + major = FIELD_GET(YT921X_CHIP_ID_MAJOR, chipid); + + for (info = yt921x_infos; info->name; info++) + if (info->major == major) + break; + if (!info->name) { + dev_err(dev, "Unexpected chipid 0x%x\n", chipid); + return -ENODEV; + } + + res = yt921x_reg_read(priv, YT921X_CHIP_MODE, &mode); + if (res) + return res; + res = yt921x_edata_read(priv, YT921X_EDATA_EXTMODE, &extmode); + if (res) + return res; + + for (; info->name; info++) + if (info->major == major && info->mode == mode && + info->extmode == extmode) + break; + if (!info->name) { + dev_err(dev, + "Unsupported chipid 0x%x with chipmode 0x%x 0x%x\n", + chipid, mode, extmode); + return -ENODEV; + } + + /* Print chipid here since we are interested in lower 16 bits */ + dev_info(dev, + "Motorcomm %s ethernet switch, chipid: 0x%x, chipmode: 0x%x 0x%x\n", + info->name, chipid, mode, extmode); + + priv->info = info; + return 0; +} + +static int yt921x_chip_reset(struct yt921x_priv *priv) +{ + struct device *dev = to_device(priv); + u16 eth_p_tag; + u32 val; + int res; + + res = yt921x_chip_detect(priv); + if (res) + return res; + + /* Reset */ + res = yt921x_reg_write(priv, YT921X_RST, YT921X_RST_HW); + if (res) + return res; + + /* RST_HW is almost same as GPIO hard reset, so we need this delay. */ + fsleep(YT921X_RST_DELAY_US); + + val = 0; + res = yt921x_reg_wait(priv, YT921X_RST, ~0, &val); + if (res) + return res; + + /* Check for tag EtherType; do it after reset in case you messed it up + * before. + */ + res = yt921x_reg_read(priv, YT921X_CPU_TAG_TPID, &val); + if (res) + return res; + eth_p_tag = FIELD_GET(YT921X_CPU_TAG_TPID_TPID_M, val); + if (eth_p_tag != ETH_P_YT921X) { + dev_err(dev, "Tag type 0x%x != 0x%x\n", eth_p_tag, + ETH_P_YT921X); + /* Despite being possible, we choose not to set CPU_TAG_TPID, + * since there is no way it can be different unless you have the + * wrong chip. + */ + return -EINVAL; + } + + return 0; +} + +static int yt921x_chip_setup(struct yt921x_priv *priv) +{ + struct dsa_switch *ds = &priv->ds; + unsigned long cpu_ports_mask; + u64 ctrl64; + u32 ctrl; + int port; + int res; + + /* Enable DSA */ + priv->cpu_ports_mask = dsa_cpu_ports(ds); + + ctrl = YT921X_EXT_CPU_PORT_TAG_EN | YT921X_EXT_CPU_PORT_PORT_EN | + YT921X_EXT_CPU_PORT_PORT(__ffs(priv->cpu_ports_mask)); + res = yt921x_reg_write(priv, YT921X_EXT_CPU_PORT, ctrl); + if (res) + return res; + + /* Enable and clear MIB */ + res = yt921x_reg_set_bits(priv, YT921X_FUNC, YT921X_FUNC_MIB); + if (res) + return res; + + ctrl = YT921X_MIB_CTRL_CLEAN | YT921X_MIB_CTRL_ALL_PORT; + res = yt921x_reg_write(priv, YT921X_MIB_CTRL, ctrl); + if (res) + return res; + + /* Setup software switch */ + ctrl = YT921X_CPU_COPY_TO_EXT_CPU; + res = yt921x_reg_write(priv, YT921X_CPU_COPY, ctrl); + if (res) + return res; + + ctrl = GENMASK(10, 0); + res = yt921x_reg_write(priv, YT921X_FILTER_UNK_UCAST, ctrl); + if (res) + return res; + res = yt921x_reg_write(priv, YT921X_FILTER_UNK_MCAST, ctrl); + if (res) + return res; + + /* YT921x does not support native DSA port bridging, so we use port + * isolation to emulate it. However, be especially careful that port + * isolation takes _after_ FDB lookups, i.e. if an FDB entry (from + * another bridge) is matched and the destination port (in another + * bridge) is blocked, the packet will be dropped instead of flooding to + * the "bridged" ports, thus we need to trap and handle those packets by + * software. + * + * If there is no more than one bridge, we might be able to drop them + * directly given some conditions are met, but we trap them in all cases + * for now. + */ + ctrl = 0; + for (int i = 0; i < YT921X_PORT_NUM; i++) + ctrl |= YT921X_ACT_UNK_ACTn_TRAP(i); + /* Except for CPU ports, if any packets are sent via CPU ports without + * tag, they should be dropped. + */ + cpu_ports_mask = priv->cpu_ports_mask; + for_each_set_bit(port, &cpu_ports_mask, YT921X_PORT_NUM) { + ctrl &= ~YT921X_ACT_UNK_ACTn_M(port); + ctrl |= YT921X_ACT_UNK_ACTn_DROP(port); + } + res = yt921x_reg_write(priv, YT921X_ACT_UNK_UCAST, ctrl); + if (res) + return res; + res = yt921x_reg_write(priv, YT921X_ACT_UNK_MCAST, ctrl); + if (res) + return res; + + /* Tagged VID 0 should be treated as untagged, which confuses the + * hardware a lot + */ + ctrl64 = YT921X_VLAN_CTRL_LEARN_DIS | YT921X_VLAN_CTRL_PORTS_M; + res = yt921x_reg64_write(priv, YT921X_VLANn_CTRL(0), ctrl64); + if (res) + return res; + + /* Miscellaneous */ + res = yt921x_reg_set_bits(priv, YT921X_SENSOR, YT921X_SENSOR_TEMP); + if (res) + return res; + + return 0; +} + +static int yt921x_dsa_setup(struct dsa_switch *ds) +{ + struct yt921x_priv *priv = to_yt921x_priv(ds); + struct device *dev = to_device(priv); + struct device_node *np = dev->of_node; + struct device_node *child; + int res; + + mutex_lock(&priv->reg_lock); + res = yt921x_chip_reset(priv); + mutex_unlock(&priv->reg_lock); + + if (res) + return res; + + /* Register the internal mdio bus. Nodes for internal ports should have + * proper phy-handle pointing to their PHYs. Not enabling the internal + * bus is possible, though pretty wired, if internal ports are not used. + */ + child = of_get_child_by_name(np, "mdio"); + if (child) { + res = yt921x_mbus_int_init(priv, child); + of_node_put(child); + if (res) + return res; + } + + /* External mdio bus is optional */ + child = of_get_child_by_name(np, "mdio-external"); + if (child) { + res = yt921x_mbus_ext_init(priv, child); + of_node_put(child); + if (res) + return res; + + dev_err(dev, "Untested external mdio bus\n"); + return -ENODEV; + } + + mutex_lock(&priv->reg_lock); + res = yt921x_chip_setup(priv); + mutex_unlock(&priv->reg_lock); + + if (res) + return res; + + return 0; +} + +static const struct phylink_mac_ops yt921x_phylink_mac_ops = { + .mac_link_down = yt921x_phylink_mac_link_down, + .mac_link_up = yt921x_phylink_mac_link_up, + .mac_config = yt921x_phylink_mac_config, +}; + +static const struct dsa_switch_ops yt921x_dsa_switch_ops = { + /* mib */ + .get_strings = yt921x_dsa_get_strings, + .get_ethtool_stats = yt921x_dsa_get_ethtool_stats, + .get_sset_count = yt921x_dsa_get_sset_count, + .get_eth_mac_stats = yt921x_dsa_get_eth_mac_stats, + .get_eth_ctrl_stats = yt921x_dsa_get_eth_ctrl_stats, + .get_rmon_stats = yt921x_dsa_get_rmon_stats, + .get_stats64 = yt921x_dsa_get_stats64, + .get_pause_stats = yt921x_dsa_get_pause_stats, + /* eee */ + .support_eee = dsa_supports_eee, + .set_mac_eee = yt921x_dsa_set_mac_eee, + /* mtu */ + .port_change_mtu = yt921x_dsa_port_change_mtu, + .port_max_mtu = yt921x_dsa_port_max_mtu, + /* mirror */ + .port_mirror_del = yt921x_dsa_port_mirror_del, + .port_mirror_add = yt921x_dsa_port_mirror_add, + /* fdb */ + .port_fdb_dump = yt921x_dsa_port_fdb_dump, + .port_fast_age = yt921x_dsa_port_fast_age, + .set_ageing_time = yt921x_dsa_set_ageing_time, + .port_fdb_del = yt921x_dsa_port_fdb_del, + .port_fdb_add = yt921x_dsa_port_fdb_add, + .port_mdb_del = yt921x_dsa_port_mdb_del, + .port_mdb_add = yt921x_dsa_port_mdb_add, + /* vlan */ + .port_vlan_filtering = yt921x_dsa_port_vlan_filtering, + .port_vlan_del = yt921x_dsa_port_vlan_del, + .port_vlan_add = yt921x_dsa_port_vlan_add, + /* bridge */ + .port_pre_bridge_flags = yt921x_dsa_port_pre_bridge_flags, + .port_bridge_flags = yt921x_dsa_port_bridge_flags, + .port_bridge_leave = yt921x_dsa_port_bridge_leave, + .port_bridge_join = yt921x_dsa_port_bridge_join, + /* port */ + .get_tag_protocol = yt921x_dsa_get_tag_protocol, + .phylink_get_caps = yt921x_dsa_phylink_get_caps, + .port_setup = yt921x_dsa_port_setup, + /* chip */ + .setup = yt921x_dsa_setup, +}; + +static void yt921x_mdio_shutdown(struct mdio_device *mdiodev) +{ + struct yt921x_priv *priv = mdiodev_get_drvdata(mdiodev); + + if (!priv) + return; + + dsa_switch_shutdown(&priv->ds); +} + +static void yt921x_mdio_remove(struct mdio_device *mdiodev) +{ + struct yt921x_priv *priv = mdiodev_get_drvdata(mdiodev); + + if (!priv) + return; + + for (size_t i = ARRAY_SIZE(priv->ports); i-- > 0; ) { + struct yt921x_port *pp = &priv->ports[i]; + + disable_delayed_work_sync(&pp->mib_read); + } + + dsa_unregister_switch(&priv->ds); + + mutex_destroy(&priv->reg_lock); +} + +static int yt921x_mdio_probe(struct mdio_device *mdiodev) +{ + struct device *dev = &mdiodev->dev; + struct yt921x_reg_mdio *mdio; + struct yt921x_priv *priv; + struct dsa_switch *ds; + + priv = devm_kzalloc(dev, sizeof(*priv), GFP_KERNEL); + if (!priv) + return -ENOMEM; + + mdio = devm_kzalloc(dev, sizeof(*mdio), GFP_KERNEL); + if (!mdio) + return -ENOMEM; + + mdio->bus = mdiodev->bus; + mdio->addr = mdiodev->addr; + mdio->switchid = 0; + + mutex_init(&priv->reg_lock); + + priv->reg_ops = &yt921x_reg_ops_mdio; + priv->reg_ctx = mdio; + + for (size_t i = 0; i < ARRAY_SIZE(priv->ports); i++) { + struct yt921x_port *pp = &priv->ports[i]; + + pp->index = i; + INIT_DELAYED_WORK(&pp->mib_read, yt921x_poll_mib); + } + + ds = &priv->ds; + ds->dev = dev; + ds->assisted_learning_on_cpu_port = true; + ds->priv = priv; + ds->ops = &yt921x_dsa_switch_ops; + ds->phylink_mac_ops = &yt921x_phylink_mac_ops; + ds->num_ports = YT921X_PORT_NUM; + + mdiodev_set_drvdata(mdiodev, priv); + + return dsa_register_switch(ds); +} + +static const struct of_device_id yt921x_of_match[] = { + { .compatible = "motorcomm,yt9215" }, + {} +}; +MODULE_DEVICE_TABLE(of, yt921x_of_match); + +static struct mdio_driver yt921x_mdio_driver = { + .probe = yt921x_mdio_probe, + .remove = yt921x_mdio_remove, + .shutdown = yt921x_mdio_shutdown, + .mdiodrv.driver = { + .name = YT921X_NAME, + .of_match_table = yt921x_of_match, + }, +}; + +mdio_module_driver(yt921x_mdio_driver); + +MODULE_AUTHOR("David Yang "); +MODULE_DESCRIPTION("Driver for Motorcomm YT921x Switch"); +MODULE_LICENSE("GPL"); diff --git a/drivers/net/dsa/yt921x.h b/drivers/net/dsa/yt921x.h new file mode 100644 index 000000000000..3e85d90826fb --- /dev/null +++ b/drivers/net/dsa/yt921x.h @@ -0,0 +1,504 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ +/* + * Copyright (c) 2025 David Yang + */ + +#ifndef __YT921X_H +#define __YT921X_H + +#include + +#define YT921X_SMI_SWITCHID_M GENMASK(3, 2) +#define YT921X_SMI_SWITCHID(x) FIELD_PREP(YT921X_SMI_SWITCHID_M, (x)) +#define YT921X_SMI_AD BIT(1) +#define YT921X_SMI_ADDR 0 +#define YT921X_SMI_DATA YT921X_SMI_AD +#define YT921X_SMI_RW BIT(0) +#define YT921X_SMI_WRITE 0 +#define YT921X_SMI_READ YT921X_SMI_RW + +#define YT921X_SWITCHID_NUM 4 + +#define YT921X_RST 0x80000 +#define YT921X_RST_HW BIT(31) +#define YT921X_RST_SW BIT(1) +#define YT921X_FUNC 0x80004 +#define YT921X_FUNC_MIB BIT(1) +#define YT921X_CHIP_ID 0x80008 +#define YT921X_CHIP_ID_MAJOR GENMASK(31, 16) +#define YT921X_EXT_CPU_PORT 0x8000c +#define YT921X_EXT_CPU_PORT_TAG_EN BIT(15) +#define YT921X_EXT_CPU_PORT_PORT_EN BIT(14) +#define YT921X_EXT_CPU_PORT_PORT_M GENMASK(3, 0) +#define YT921X_EXT_CPU_PORT_PORT(x) FIELD_PREP(YT921X_EXT_CPU_PORT_PORT_M, (x)) +#define YT921X_CPU_TAG_TPID 0x80010 +#define YT921X_CPU_TAG_TPID_TPID_M GENMASK(15, 0) +/* Same as ETH_P_YT921X, but this represents the true HW default, while the + * former is a local convention chosen by us. + */ +#define YT921X_CPU_TAG_TPID_TPID_DEFAULT 0x9988 +#define YT921X_PVID_SEL 0x80014 +#define YT921X_PVID_SEL_SVID_PORTn(port) BIT(port) +#define YT921X_SERDES_CTRL 0x80028 +#define YT921X_SERDES_CTRL_PORTn_TEST(port) BIT((port) - 3) +#define YT921X_SERDES_CTRL_PORTn(port) BIT((port) - 8) +#define YT921X_IO_LEVEL 0x80030 +#define YT9215_IO_LEVEL_NORMAL_M GENMASK(5, 4) +#define YT9215_IO_LEVEL_NORMAL(x) FIELD_PREP(YT9215_IO_LEVEL_NORMAL_M, (x)) +#define YT9215_IO_LEVEL_NORMAL_3V3 YT9215_IO_LEVEL_NORMAL(0) +#define YT9215_IO_LEVEL_NORMAL_1V8 YT9215_IO_LEVEL_NORMAL(3) +#define YT9215_IO_LEVEL_RGMII1_M GENMASK(3, 2) +#define YT9215_IO_LEVEL_RGMII1(x) FIELD_PREP(YT9215_IO_LEVEL_RGMII1_M, (x)) +#define YT9215_IO_LEVEL_RGMII1_3V3 YT9215_IO_LEVEL_RGMII1(0) +#define YT9215_IO_LEVEL_RGMII1_2V5 YT9215_IO_LEVEL_RGMII1(1) +#define YT9215_IO_LEVEL_RGMII1_1V8 YT9215_IO_LEVEL_RGMII1(2) +#define YT9215_IO_LEVEL_RGMII0_M GENMASK(1, 0) +#define YT9215_IO_LEVEL_RGMII0(x) FIELD_PREP(YT9215_IO_LEVEL_RGMII0_M, (x)) +#define YT9215_IO_LEVEL_RGMII0_3V3 YT9215_IO_LEVEL_RGMII0(0) +#define YT9215_IO_LEVEL_RGMII0_2V5 YT9215_IO_LEVEL_RGMII0(1) +#define YT9215_IO_LEVEL_RGMII0_1V8 YT9215_IO_LEVEL_RGMII0(2) +#define YT9218_IO_LEVEL_RGMII1_M GENMASK(5, 4) +#define YT9218_IO_LEVEL_RGMII1(x) FIELD_PREP(YT9218_IO_LEVEL_RGMII1_M, (x)) +#define YT9218_IO_LEVEL_RGMII1_3V3 YT9218_IO_LEVEL_RGMII1(0) +#define YT9218_IO_LEVEL_RGMII1_2V5 YT9218_IO_LEVEL_RGMII1(1) +#define YT9218_IO_LEVEL_RGMII1_1V8 YT9218_IO_LEVEL_RGMII1(2) +#define YT9218_IO_LEVEL_RGMII0_M GENMASK(3, 2) +#define YT9218_IO_LEVEL_RGMII0(x) FIELD_PREP(YT9218_IO_LEVEL_RGMII0_M, (x)) +#define YT9218_IO_LEVEL_RGMII0_3V3 YT9218_IO_LEVEL_RGMII0(0) +#define YT9218_IO_LEVEL_RGMII0_2V5 YT9218_IO_LEVEL_RGMII0(1) +#define YT9218_IO_LEVEL_RGMII0_1V8 YT9218_IO_LEVEL_RGMII0(2) +#define YT9218_IO_LEVEL_NORMAL_M GENMASK(1, 0) +#define YT9218_IO_LEVEL_NORMAL(x) FIELD_PREP(YT9218_IO_LEVEL_NORMAL_M, (x)) +#define YT9218_IO_LEVEL_NORMAL_3V3 YT9218_IO_LEVEL_NORMAL(0) +#define YT9218_IO_LEVEL_NORMAL_1V8 YT9218_IO_LEVEL_NORMAL(3) +#define YT921X_MAC_ADDR_HI2 0x80080 +#define YT921X_MAC_ADDR_LO4 0x80084 +#define YT921X_SERDESn(port) (0x8008c + 4 * ((port) - 8)) +#define YT921X_SERDES_MODE_M GENMASK(9, 7) +#define YT921X_SERDES_MODE(x) FIELD_PREP(YT921X_SERDES_MODE_M, (x)) +#define YT921X_SERDES_MODE_SGMII YT921X_SERDES_MODE(0) +#define YT921X_SERDES_MODE_REVSGMII YT921X_SERDES_MODE(1) +#define YT921X_SERDES_MODE_1000BASEX YT921X_SERDES_MODE(2) +#define YT921X_SERDES_MODE_100BASEX YT921X_SERDES_MODE(3) +#define YT921X_SERDES_MODE_2500BASEX YT921X_SERDES_MODE(4) +#define YT921X_SERDES_RX_PAUSE BIT(6) +#define YT921X_SERDES_TX_PAUSE BIT(5) +#define YT921X_SERDES_LINK BIT(4) /* force link */ +#define YT921X_SERDES_DUPLEX_FULL BIT(3) +#define YT921X_SERDES_SPEED_M GENMASK(2, 0) +#define YT921X_SERDES_SPEED(x) FIELD_PREP(YT921X_SERDES_SPEED_M, (x)) +#define YT921X_SERDES_SPEED_10 YT921X_SERDES_SPEED(0) +#define YT921X_SERDES_SPEED_100 YT921X_SERDES_SPEED(1) +#define YT921X_SERDES_SPEED_1000 YT921X_SERDES_SPEED(2) +#define YT921X_SERDES_SPEED_10000 YT921X_SERDES_SPEED(3) +#define YT921X_SERDES_SPEED_2500 YT921X_SERDES_SPEED(4) +#define YT921X_PORTn_CTRL(port) (0x80100 + 4 * (port)) +#define YT921X_PORT_CTRL_PAUSE_AN BIT(10) +#define YT921X_PORTn_STATUS(port) (0x80200 + 4 * (port)) +#define YT921X_PORT_LINK BIT(9) /* CTRL: auto negotiation */ +#define YT921X_PORT_HALF_PAUSE BIT(8) /* Half-duplex back pressure mode */ +#define YT921X_PORT_DUPLEX_FULL BIT(7) +#define YT921X_PORT_RX_PAUSE BIT(6) +#define YT921X_PORT_TX_PAUSE BIT(5) +#define YT921X_PORT_RX_MAC_EN BIT(4) +#define YT921X_PORT_TX_MAC_EN BIT(3) +#define YT921X_PORT_SPEED_M GENMASK(2, 0) +#define YT921X_PORT_SPEED(x) FIELD_PREP(YT921X_PORT_SPEED_M, (x)) +#define YT921X_PORT_SPEED_10 YT921X_PORT_SPEED(0) +#define YT921X_PORT_SPEED_100 YT921X_PORT_SPEED(1) +#define YT921X_PORT_SPEED_1000 YT921X_PORT_SPEED(2) +#define YT921X_PORT_SPEED_10000 YT921X_PORT_SPEED(3) +#define YT921X_PORT_SPEED_2500 YT921X_PORT_SPEED(4) +#define YT921X_PON_STRAP_FUNC 0x80320 +#define YT921X_PON_STRAP_VAL 0x80324 +#define YT921X_PON_STRAP_CAP 0x80328 +#define YT921X_PON_STRAP_EEE BIT(16) +#define YT921X_PON_STRAP_LOOP_DETECT BIT(7) +#define YT921X_MDIO_POLLINGn(port) (0x80364 + 4 * ((port) - 8)) +#define YT921X_MDIO_POLLING_DUPLEX_FULL BIT(4) +#define YT921X_MDIO_POLLING_LINK BIT(3) +#define YT921X_MDIO_POLLING_SPEED_M GENMASK(2, 0) +#define YT921X_MDIO_POLLING_SPEED(x) FIELD_PREP(YT921X_MDIO_POLLING_SPEED_M, (x)) +#define YT921X_MDIO_POLLING_SPEED_10 YT921X_MDIO_POLLING_SPEED(0) +#define YT921X_MDIO_POLLING_SPEED_100 YT921X_MDIO_POLLING_SPEED(1) +#define YT921X_MDIO_POLLING_SPEED_1000 YT921X_MDIO_POLLING_SPEED(2) +#define YT921X_MDIO_POLLING_SPEED_10000 YT921X_MDIO_POLLING_SPEED(3) +#define YT921X_MDIO_POLLING_SPEED_2500 YT921X_MDIO_POLLING_SPEED(4) +#define YT921X_SENSOR 0x8036c +#define YT921X_SENSOR_TEMP BIT(18) +#define YT921X_TEMP 0x80374 +#define YT921X_CHIP_MODE 0x80388 +#define YT921X_CHIP_MODE_MODE GENMASK(1, 0) +#define YT921X_XMII_CTRL 0x80394 +#define YT921X_XMII_CTRL_PORTn(port) BIT(9 - (port)) /* Yes, it's reversed */ +#define YT921X_XMIIn(port) (0x80400 + 8 * ((port) - 8)) +#define YT921X_XMII_MODE_M GENMASK(31, 29) +#define YT921X_XMII_MODE(x) FIELD_PREP(YT921X_XMII_MODE_M, (x)) +#define YT921X_XMII_MODE_MII YT921X_XMII_MODE(0) +#define YT921X_XMII_MODE_REVMII YT921X_XMII_MODE(1) +#define YT921X_XMII_MODE_RMII YT921X_XMII_MODE(2) +#define YT921X_XMII_MODE_REVRMII YT921X_XMII_MODE(3) +#define YT921X_XMII_MODE_RGMII YT921X_XMII_MODE(4) +#define YT921X_XMII_MODE_DISABLE YT921X_XMII_MODE(5) +#define YT921X_XMII_LINK BIT(19) /* force link */ +#define YT921X_XMII_EN BIT(18) +#define YT921X_XMII_SOFT_RST BIT(17) +#define YT921X_XMII_RGMII_TX_DELAY_150PS_M GENMASK(16, 13) +#define YT921X_XMII_RGMII_TX_DELAY_150PS(x) FIELD_PREP(YT921X_XMII_RGMII_TX_DELAY_150PS_M, (x)) +#define YT921X_XMII_TX_CLK_IN BIT(11) +#define YT921X_XMII_RX_CLK_IN BIT(10) +#define YT921X_XMII_RGMII_TX_DELAY_2NS BIT(8) +#define YT921X_XMII_RGMII_TX_CLK_OUT BIT(7) +#define YT921X_XMII_RGMII_RX_DELAY_150PS_M GENMASK(6, 3) +#define YT921X_XMII_RGMII_RX_DELAY_150PS(x) FIELD_PREP(YT921X_XMII_RGMII_RX_DELAY_150PS_M, (x)) +#define YT921X_XMII_RMII_PHY_TX_CLK_OUT BIT(2) +#define YT921X_XMII_REVMII_TX_CLK_OUT BIT(1) +#define YT921X_XMII_REVMII_RX_CLK_OUT BIT(0) + +#define YT921X_MACn_FRAME(port) (0x81008 + 0x1000 * (port)) +#define YT921X_MAC_FRAME_SIZE_M GENMASK(21, 8) +#define YT921X_MAC_FRAME_SIZE(x) FIELD_PREP(YT921X_MAC_FRAME_SIZE_M, (x)) + +#define YT921X_EEEn_VAL(port) (0xa0000 + 0x40 * (port)) +#define YT921X_EEE_VAL_DATA BIT(1) + +#define YT921X_EEE_CTRL 0xb0000 +#define YT921X_EEE_CTRL_ENn(port) BIT(port) + +#define YT921X_MIB_CTRL 0xc0004 +#define YT921X_MIB_CTRL_CLEAN BIT(30) +#define YT921X_MIB_CTRL_PORT_M GENMASK(6, 3) +#define YT921X_MIB_CTRL_PORT(x) FIELD_PREP(YT921X_MIB_CTRL_PORT_M, (x)) +#define YT921X_MIB_CTRL_ONE_PORT BIT(1) +#define YT921X_MIB_CTRL_ALL_PORT BIT(0) +#define YT921X_MIBn_DATA0(port) (0xc0100 + 0x100 * (port)) +#define YT921X_MIBn_DATAm(port, x) (YT921X_MIBn_DATA0(port) + 4 * (x)) + +#define YT921X_EDATA_CTRL 0xe0000 +#define YT921X_EDATA_CTRL_ADDR_M GENMASK(15, 8) +#define YT921X_EDATA_CTRL_ADDR(x) FIELD_PREP(YT921X_EDATA_CTRL_ADDR_M, (x)) +#define YT921X_EDATA_CTRL_OP_M GENMASK(3, 0) +#define YT921X_EDATA_CTRL_OP(x) FIELD_PREP(YT921X_EDATA_CTRL_OP_M, (x)) +#define YT921X_EDATA_CTRL_READ YT921X_EDATA_CTRL_OP(5) +#define YT921X_EDATA_DATA 0xe0004 +#define YT921X_EDATA_DATA_DATA_M GENMASK(31, 24) +#define YT921X_EDATA_DATA_STATUS_M GENMASK(3, 0) +#define YT921X_EDATA_DATA_STATUS(x) FIELD_PREP(YT921X_EDATA_DATA_STATUS_M, (x)) +#define YT921X_EDATA_DATA_IDLE YT921X_EDATA_DATA_STATUS(3) + +#define YT921X_EXT_MBUS_OP 0x6a000 +#define YT921X_INT_MBUS_OP 0xf0000 +#define YT921X_MBUS_OP_START BIT(0) +#define YT921X_EXT_MBUS_CTRL 0x6a004 +#define YT921X_INT_MBUS_CTRL 0xf0004 +#define YT921X_MBUS_CTRL_PORT_M GENMASK(25, 21) +#define YT921X_MBUS_CTRL_PORT(x) FIELD_PREP(YT921X_MBUS_CTRL_PORT_M, (x)) +#define YT921X_MBUS_CTRL_REG_M GENMASK(20, 16) +#define YT921X_MBUS_CTRL_REG(x) FIELD_PREP(YT921X_MBUS_CTRL_REG_M, (x)) +#define YT921X_MBUS_CTRL_TYPE_M GENMASK(11, 8) /* wild guess */ +#define YT921X_MBUS_CTRL_TYPE(x) FIELD_PREP(YT921X_MBUS_CTRL_TYPE_M, (x)) +#define YT921X_MBUS_CTRL_TYPE_C22 YT921X_MBUS_CTRL_TYPE(4) +#define YT921X_MBUS_CTRL_OP_M GENMASK(3, 2) /* wild guess */ +#define YT921X_MBUS_CTRL_OP(x) FIELD_PREP(YT921X_MBUS_CTRL_OP_M, (x)) +#define YT921X_MBUS_CTRL_WRITE YT921X_MBUS_CTRL_OP(1) +#define YT921X_MBUS_CTRL_READ YT921X_MBUS_CTRL_OP(2) +#define YT921X_EXT_MBUS_DOUT 0x6a008 +#define YT921X_INT_MBUS_DOUT 0xf0008 +#define YT921X_EXT_MBUS_DIN 0x6a00c +#define YT921X_INT_MBUS_DIN 0xf000c + +#define YT921X_PORTn_EGR(port) (0x100000 + 4 * (port)) +#define YT921X_PORT_EGR_TPID_CTAG_M GENMASK(5, 4) +#define YT921X_PORT_EGR_TPID_CTAG(x) FIELD_PREP(YT921X_PORT_EGR_TPID_CTAG_M, (x)) +#define YT921X_PORT_EGR_TPID_STAG_M GENMASK(3, 2) +#define YT921X_PORT_EGR_TPID_STAG(x) FIELD_PREP(YT921X_PORT_EGR_TPID_STAG_M, (x)) +#define YT921X_TPID_EGRn(x) (0x100300 + 4 * (x)) /* [0, 3] */ +#define YT921X_TPID_EGR_TPID_M GENMASK(15, 0) + +#define YT921X_VLAN_IGR_FILTER 0x180280 +#define YT921X_VLAN_IGR_FILTER_PORTn_BYPASS_IGMP(port) BIT((port) + 11) +#define YT921X_VLAN_IGR_FILTER_PORTn(port) BIT(port) +#define YT921X_PORTn_ISOLATION(port) (0x180294 + 4 * (port)) +#define YT921X_PORT_ISOLATION_BLOCKn(port) BIT(port) +#define YT921X_PORTn_LEARN(port) (0x1803d0 + 4 * (port)) +#define YT921X_PORT_LEARN_VID_LEARN_MULTI_EN BIT(22) +#define YT921X_PORT_LEARN_VID_LEARN_MODE BIT(21) +#define YT921X_PORT_LEARN_VID_LEARN_EN BIT(20) +#define YT921X_PORT_LEARN_SUSPEND_COPY_EN BIT(19) +#define YT921X_PORT_LEARN_SUSPEND_DROP_EN BIT(18) +#define YT921X_PORT_LEARN_DIS BIT(17) +#define YT921X_PORT_LEARN_LIMIT_EN BIT(16) +#define YT921X_PORT_LEARN_LIMIT_M GENMASK(15, 8) +#define YT921X_PORT_LEARN_LIMIT(x) FIELD_PREP(YT921X_PORT_LEARN_LIMIT_M, (x)) +#define YT921X_PORT_LEARN_DROP_ON_EXCEEDED BIT(2) +#define YT921X_PORT_LEARN_MODE_M GENMASK(1, 0) +#define YT921X_PORT_LEARN_MODE(x) FIELD_PREP(YT921X_PORT_LEARN_MODE_M, (x)) +#define YT921X_PORT_LEARN_MODE_AUTO YT921X_PORT_LEARN_MODE(0) +#define YT921X_PORT_LEARN_MODE_AUTO_AND_COPY YT921X_PORT_LEARN_MODE(1) +#define YT921X_PORT_LEARN_MODE_CPU_CONTROL YT921X_PORT_LEARN_MODE(2) +#define YT921X_AGEING 0x180440 +#define YT921X_AGEING_INTERVAL_M GENMASK(15, 0) +#define YT921X_FDB_IN0 0x180454 +#define YT921X_FDB_IN1 0x180458 +#define YT921X_FDB_IN2 0x18045c +#define YT921X_FDB_OP 0x180460 +#define YT921X_FDB_OP_INDEX_M GENMASK(22, 11) +#define YT921X_FDB_OP_INDEX(x) FIELD_PREP(YT921X_FDB_OP_INDEX_M, (x)) +#define YT921X_FDB_OP_MODE_INDEX BIT(10) /* mac+fid / index */ +#define YT921X_FDB_OP_FLUSH_MCAST BIT(9) /* ucast / mcast */ +#define YT921X_FDB_OP_FLUSH_M GENMASK(8, 7) +#define YT921X_FDB_OP_FLUSH(x) FIELD_PREP(YT921X_FDB_OP_FLUSH_M, (x)) +#define YT921X_FDB_OP_FLUSH_ALL YT921X_FDB_OP_FLUSH(0) +#define YT921X_FDB_OP_FLUSH_PORT YT921X_FDB_OP_FLUSH(1) +#define YT921X_FDB_OP_FLUSH_PORT_VID YT921X_FDB_OP_FLUSH(2) +#define YT921X_FDB_OP_FLUSH_VID YT921X_FDB_OP_FLUSH(3) +#define YT921X_FDB_OP_FLUSH_STATIC BIT(6) +#define YT921X_FDB_OP_NEXT_TYPE_M GENMASK(5, 4) +#define YT921X_FDB_OP_NEXT_TYPE(x) FIELD_PREP(YT921X_FDB_OP_NEXT_TYPE_M, (x)) +#define YT921X_FDB_OP_NEXT_TYPE_UCAST_PORT YT921X_FDB_OP_NEXT_TYPE(0) +#define YT921X_FDB_OP_NEXT_TYPE_UCAST_VID YT921X_FDB_OP_NEXT_TYPE(1) +#define YT921X_FDB_OP_NEXT_TYPE_UCAST YT921X_FDB_OP_NEXT_TYPE(2) +#define YT921X_FDB_OP_NEXT_TYPE_MCAST YT921X_FDB_OP_NEXT_TYPE(3) +#define YT921X_FDB_OP_OP_M GENMASK(3, 1) +#define YT921X_FDB_OP_OP(x) FIELD_PREP(YT921X_FDB_OP_OP_M, (x)) +#define YT921X_FDB_OP_OP_ADD YT921X_FDB_OP_OP(0) +#define YT921X_FDB_OP_OP_DEL YT921X_FDB_OP_OP(1) +#define YT921X_FDB_OP_OP_GET_ONE YT921X_FDB_OP_OP(2) +#define YT921X_FDB_OP_OP_GET_NEXT YT921X_FDB_OP_OP(3) +#define YT921X_FDB_OP_OP_FLUSH YT921X_FDB_OP_OP(4) +#define YT921X_FDB_OP_START BIT(0) +#define YT921X_FDB_RESULT 0x180464 +#define YT921X_FDB_RESULT_DONE BIT(15) +#define YT921X_FDB_RESULT_NOTFOUND BIT(14) +#define YT921X_FDB_RESULT_OVERWRITED BIT(13) +#define YT921X_FDB_RESULT_INDEX_M GENMASK(11, 0) +#define YT921X_FDB_RESULT_INDEX(x) FIELD_PREP(YT921X_FDB_RESULT_INDEX_M, (x)) +#define YT921X_FDB_OUT0 0x1804b0 +#define YT921X_FDB_IO0_ADDR_HI4_M GENMASK(31, 0) +#define YT921X_FDB_OUT1 0x1804b4 +#define YT921X_FDB_IO1_EGR_INT_PRI_EN BIT(31) +#define YT921X_FDB_IO1_STATUS_M GENMASK(30, 28) +#define YT921X_FDB_IO1_STATUS(x) FIELD_PREP(YT921X_FDB_IO1_STATUS_M, (x)) +#define YT921X_FDB_IO1_STATUS_INVALID YT921X_FDB_IO1_STATUS(0) +#define YT921X_FDB_IO1_STATUS_MIN_TIME YT921X_FDB_IO1_STATUS(1) +#define YT921X_FDB_IO1_STATUS_MOVE_AGING_MAX_TIME YT921X_FDB_IO1_STATUS(3) +#define YT921X_FDB_IO1_STATUS_MAX_TIME YT921X_FDB_IO1_STATUS(5) +#define YT921X_FDB_IO1_STATUS_PENDING YT921X_FDB_IO1_STATUS(6) +#define YT921X_FDB_IO1_STATUS_STATIC YT921X_FDB_IO1_STATUS(7) +#define YT921X_FDB_IO1_FID_M GENMASK(27, 16) /* filtering ID (VID) */ +#define YT921X_FDB_IO1_FID(x) FIELD_PREP(YT921X_FDB_IO1_FID_M, (x)) +#define YT921X_FDB_IO1_ADDR_LO2_M GENMASK(15, 0) +#define YT921X_FDB_OUT2 0x1804b8 +#define YT921X_FDB_IO2_MOVE_AGING_STATUS_M GENMASK(31, 30) +#define YT921X_FDB_IO2_IGR_DROP BIT(29) +#define YT921X_FDB_IO2_EGR_PORTS_M GENMASK(28, 18) +#define YT921X_FDB_IO2_EGR_PORTS(x) FIELD_PREP(YT921X_FDB_IO2_EGR_PORTS_M, (x)) +#define YT921X_FDB_IO2_EGR_DROP BIT(17) +#define YT921X_FDB_IO2_COPY_TO_CPU BIT(16) +#define YT921X_FDB_IO2_IGR_INT_PRI_EN BIT(15) +#define YT921X_FDB_IO2_INT_PRI_M GENMASK(14, 12) +#define YT921X_FDB_IO2_INT_PRI(x) FIELD_PREP(YT921X_FDB_IO2_INT_PRI_M, (x)) +#define YT921X_FDB_IO2_NEW_VID_M GENMASK(11, 0) +#define YT921X_FDB_IO2_NEW_VID(x) FIELD_PREP(YT921X_FDB_IO2_NEW_VID_M, (x)) +#define YT921X_FILTER_UNK_UCAST 0x180508 +#define YT921X_FILTER_UNK_MCAST 0x18050c +#define YT921X_FILTER_MCAST 0x180510 +#define YT921X_FILTER_BCAST 0x180514 +#define YT921X_FILTER_PORTS_M GENMASK(10, 0) +#define YT921X_FILTER_PORTS(x) FIELD_PREP(YT921X_FILTER_PORTS_M, (x)) +#define YT921X_FILTER_PORTn(port) BIT(port) +#define YT921X_VLAN_EGR_FILTER 0x180598 +#define YT921X_VLAN_EGR_FILTER_PORTn(port) BIT(port) +#define YT921X_CPU_COPY 0x180690 +#define YT921X_CPU_COPY_FORCE_INT_PORT BIT(2) +#define YT921X_CPU_COPY_TO_INT_CPU BIT(1) +#define YT921X_CPU_COPY_TO_EXT_CPU BIT(0) +#define YT921X_ACT_UNK_UCAST 0x180734 +#define YT921X_ACT_UNK_MCAST 0x180738 +#define YT921X_ACT_UNK_MCAST_BYPASS_DROP_RMA BIT(23) +#define YT921X_ACT_UNK_MCAST_BYPASS_DROP_IGMP BIT(22) +#define YT921X_ACT_UNK_ACTn_M(port) GENMASK(2 * (port) + 1, 2 * (port)) +#define YT921X_ACT_UNK_ACTn(port, x) ((x) << (2 * (port))) +#define YT921X_ACT_UNK_ACTn_FORWARD(port) YT921X_ACT_UNK_ACTn(port, 0) /* flood */ +#define YT921X_ACT_UNK_ACTn_TRAP(port) YT921X_ACT_UNK_ACTn(port, 1) /* steer to CPU */ +#define YT921X_ACT_UNK_ACTn_DROP(port) YT921X_ACT_UNK_ACTn(port, 2) /* discard */ +/* NEVER use this action; see comments in the tag driver */ +#define YT921X_ACT_UNK_ACTn_COPY(port) YT921X_ACT_UNK_ACTn(port, 3) /* flood and copy */ +#define YT921X_FDB_HW_FLUSH 0x180958 +#define YT921X_FDB_HW_FLUSH_ON_LINKDOWN BIT(0) + +#define YT921X_VLANn_CTRL(vlan) (0x188000 + 8 * (vlan)) +#define YT921X_VLAN_CTRL_UNTAG_PORTS_M GENMASK(50, 40) +#define YT921X_VLAN_CTRL_UNTAG_PORTS(x) FIELD_PREP(YT921X_VLAN_CTRL_UNTAG_PORTS_M, (x)) +#define YT921X_VLAN_CTRL_UNTAG_PORTn(port) BIT((port) + 40) +#define YT921X_VLAN_CTRL_STP_ID_M GENMASK(39, 36) +#define YT921X_VLAN_CTRL_STP_ID(x) FIELD_PREP(YT921X_VLAN_CTRL_STP_ID_M, (x)) +#define YT921X_VLAN_CTRL_SVLAN_EN BIT(35) +#define YT921X_VLAN_CTRL_FID_M GENMASK(34, 23) +#define YT921X_VLAN_CTRL_FID(x) FIELD_PREP(YT921X_VLAN_CTRL_FID_M, (x)) +#define YT921X_VLAN_CTRL_LEARN_DIS BIT(22) +#define YT921X_VLAN_CTRL_INT_PRI_EN BIT(21) +#define YT921X_VLAN_CTRL_INT_PRI_M GENMASK(20, 18) +#define YT921X_VLAN_CTRL_PORTS_M GENMASK(17, 7) +#define YT921X_VLAN_CTRL_PORTS(x) FIELD_PREP(YT921X_VLAN_CTRL_PORTS_M, (x)) +#define YT921X_VLAN_CTRL_PORTn(port) BIT((port) + 7) +#define YT921X_VLAN_CTRL_BYPASS_1X_AC BIT(6) +#define YT921X_VLAN_CTRL_METER_EN BIT(5) +#define YT921X_VLAN_CTRL_METER_ID_M GENMASK(4, 0) + +#define YT921X_TPID_IGRn(x) (0x210000 + 4 * (x)) /* [0, 3] */ +#define YT921X_TPID_IGR_TPID_M GENMASK(15, 0) +#define YT921X_PORTn_IGR_TPID(port) (0x210010 + 4 * (port)) +#define YT921X_PORT_IGR_TPIDn_STAG_M GENMASK(7, 4) +#define YT921X_PORT_IGR_TPIDn_STAG(x) BIT((x) + 4) +#define YT921X_PORT_IGR_TPIDn_CTAG_M GENMASK(3, 0) +#define YT921X_PORT_IGR_TPIDn_CTAG(x) BIT(x) + +#define YT921X_PORTn_VLAN_CTRL(port) (0x230010 + 4 * (port)) +#define YT921X_PORT_VLAN_CTRL_SVLAN_PRI_EN BIT(31) +#define YT921X_PORT_VLAN_CTRL_CVLAN_PRI_EN BIT(30) +#define YT921X_PORT_VLAN_CTRL_SVID_M GENMASK(29, 18) +#define YT921X_PORT_VLAN_CTRL_SVID(x) FIELD_PREP(YT921X_PORT_VLAN_CTRL_SVID_M, (x)) +#define YT921X_PORT_VLAN_CTRL_CVID_M GENMASK(17, 6) +#define YT921X_PORT_VLAN_CTRL_CVID(x) FIELD_PREP(YT921X_PORT_VLAN_CTRL_CVID_M, (x)) +#define YT921X_PORT_VLAN_CTRL_SVLAN_PRI_M GENMASK(5, 3) +#define YT921X_PORT_VLAN_CTRL_CVLAN_PRI_M GENMASK(2, 0) +#define YT921X_PORTn_VLAN_CTRL1(port) (0x230080 + 4 * (port)) +#define YT921X_PORT_VLAN_CTRL1_VLAN_RANGE_EN BIT(8) +#define YT921X_PORT_VLAN_CTRL1_VLAN_RANGE_PROFILE_ID_M GENMASK(7, 4) +#define YT921X_PORT_VLAN_CTRL1_SVLAN_DROP_TAGGED BIT(3) +#define YT921X_PORT_VLAN_CTRL1_SVLAN_DROP_UNTAGGED BIT(2) +#define YT921X_PORT_VLAN_CTRL1_CVLAN_DROP_TAGGED BIT(1) +#define YT921X_PORT_VLAN_CTRL1_CVLAN_DROP_UNTAGGED BIT(0) + +#define YT921X_MIRROR 0x300300 +#define YT921X_MIRROR_IGR_PORTS_M GENMASK(26, 16) +#define YT921X_MIRROR_IGR_PORTS(x) FIELD_PREP(YT921X_MIRROR_IGR_PORTS_M, (x)) +#define YT921X_MIRROR_IGR_PORTn(port) BIT((port) + 16) +#define YT921X_MIRROR_EGR_PORTS_M GENMASK(14, 4) +#define YT921X_MIRROR_EGR_PORTS(x) FIELD_PREP(YT921X_MIRROR_EGR_PORTS_M, (x)) +#define YT921X_MIRROR_EGR_PORTn(port) BIT((port) + 4) +#define YT921X_MIRROR_PORT_M GENMASK(3, 0) +#define YT921X_MIRROR_PORT(x) FIELD_PREP(YT921X_MIRROR_PORT_M, (x)) + +#define YT921X_EDATA_EXTMODE 0xfb +#define YT921X_EDATA_LEN 0x100 + +#define YT921X_FDB_NUM 4096 + +enum yt921x_fdb_entry_status { + YT921X_FDB_ENTRY_STATUS_INVALID = 0, + YT921X_FDB_ENTRY_STATUS_MIN_TIME = 1, + YT921X_FDB_ENTRY_STATUS_MOVE_AGING_MAX_TIME = 3, + YT921X_FDB_ENTRY_STATUS_MAX_TIME = 5, + YT921X_FDB_ENTRY_STATUS_PENDING = 6, + YT921X_FDB_ENTRY_STATUS_STATIC = 7, +}; + +#define YT9215_MAJOR 0x9002 +#define YT9218_MAJOR 0x9001 + +/* required for a hard reset */ +#define YT921X_RST_DELAY_US 10000 + +#define YT921X_FRAME_SIZE_MAX 0x2400 /* 9216 */ + +#define YT921X_TAG_LEN 8 + +/* 8 internal + 2 external + 1 mcu */ +#define YT921X_PORT_NUM 11 + +#define yt921x_port_is_internal(port) ((port) < 8) +#define yt921x_port_is_external(port) (8 <= (port) && (port) < 9) + +struct yt921x_mib { + u64 rx_broadcast; + u64 rx_pause; + u64 rx_multicast; + u64 rx_crc_errors; + + u64 rx_alignment_errors; + u64 rx_undersize_errors; + u64 rx_fragment_errors; + u64 rx_64byte; + + u64 rx_65_127byte; + u64 rx_128_255byte; + u64 rx_256_511byte; + u64 rx_512_1023byte; + + u64 rx_1024_1518byte; + u64 rx_jumbo; + u64 rx_good_bytes; + + u64 rx_bad_bytes; + u64 rx_oversize_errors; + + u64 rx_dropped; + u64 tx_broadcast; + u64 tx_pause; + u64 tx_multicast; + + u64 tx_undersize_errors; + u64 tx_64byte; + u64 tx_65_127byte; + u64 tx_128_255byte; + + u64 tx_256_511byte; + u64 tx_512_1023byte; + u64 tx_1024_1518byte; + u64 tx_jumbo; + + u64 tx_good_bytes; + u64 tx_collisions; + + u64 tx_aborted_errors; + u64 tx_multiple_collisions; + u64 tx_single_collisions; + u64 tx_good; + + u64 tx_deferred; + u64 tx_late_collisions; + u64 rx_oam; + u64 tx_oam; +}; + +struct yt921x_port { + unsigned char index; + + bool hairpin; + bool isolated; + + struct delayed_work mib_read; + struct yt921x_mib mib; + u64 rx_frames; + u64 tx_frames; +}; + +struct yt921x_reg_ops { + int (*read)(void *context, u32 reg, u32 *valp); + int (*write)(void *context, u32 reg, u32 val); +}; + +struct yt921x_priv { + struct dsa_switch ds; + + const struct yt921x_info *info; + /* cache of dsa_cpu_ports(ds) */ + u16 cpu_ports_mask; + + /* protect the access to the switch registers */ + struct mutex reg_lock; + const struct yt921x_reg_ops *reg_ops; + void *reg_ctx; + + /* mdio master bus */ + struct mii_bus *mbus_int; + struct mii_bus *mbus_ext; + + struct yt921x_port ports[YT921X_PORT_NUM]; + + u16 eee_ports_mask; +}; + +#endif From 0c5480ac96a4b7079f6cf6877561d7f172c5c8e5 Mon Sep 17 00:00:00 2001 From: David Yang Date: Fri, 17 Oct 2025 14:08:56 +0800 Subject: [PATCH 0183/1381] MAINTAINERS: add entry for Motorcomm YT921x ethernet switch driver Add a MAINTAINERS entry for the Motorcomm YT921x ethernet switch driver and its DT binding. Signed-off-by: David Yang Link: https://patch.msgid.link/20251017060859.326450-5-mmyangfl@gmail.com Signed-off-by: Jakub Kicinski --- MAINTAINERS | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/MAINTAINERS b/MAINTAINERS index 4faa7719bf86..ea72b3bd2248 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -17435,6 +17435,14 @@ S: Maintained F: Documentation/devicetree/bindings/net/motorcomm,yt8xxx.yaml F: drivers/net/phy/motorcomm.c +MOTORCOMM YT921X ETHERNET SWITCH DRIVER +M: David Yang +L: netdev@vger.kernel.org +S: Maintained +F: Documentation/devicetree/bindings/net/dsa/motorcomm,yt921x.yaml +F: drivers/net/dsa/yt921x.* +F: net/dsa/tag_yt921x.c + MOXA SMARTIO/INDUSTIO/INTELLIO SERIAL CARD M: Jiri Slaby S: Maintained From 1471a274b76d1469a06e32752d49b25cb6db2406 Mon Sep 17 00:00:00 2001 From: Shi Hao Date: Sat, 18 Oct 2025 10:55:41 +0530 Subject: [PATCH 0184/1381] eth: 3c515: replace cleanup_module with __exit update old legacy cleanup_module from the file with __exit module as per kernel code practices and restore the #ifdef MODULE condition to allow successful compilation as a built -in driver. The file had an old cleanup_module still in use which could be updated with __exit module function although its init_module is indeed newer however the cleanup_module was still using the older version of exit. To set proper exit module function replace cleanup_module with __exit corkscrew_exit_module to align it to the kernel code consistency. Signed-off-by: Shi Hao Reviewed-by: Simon Horman Link: https://patch.msgid.link/20251018052541.124365-1-i.shihao.999@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/3com/3c515.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/3com/3c515.c b/drivers/net/ethernet/3com/3c515.c index ecdea58e6a21..2227c83a4862 100644 --- a/drivers/net/ethernet/3com/3c515.c +++ b/drivers/net/ethernet/3com/3c515.c @@ -1547,9 +1547,8 @@ static const struct ethtool_ops netdev_ethtool_ops = { .set_msglevel = netdev_set_msglevel, }; - #ifdef MODULE -void cleanup_module(void) +static void __exit corkscrew_exit_module(void) { while (!list_empty(&root_corkscrew_dev)) { struct net_device *dev; @@ -1563,4 +1562,5 @@ void cleanup_module(void) free_netdev(dev); } } +module_exit(corkscrew_exit_module); #endif /* MODULE */ From 4a107a0e836177359b861e9849a2a24ad35e9919 Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Sat, 18 Oct 2025 20:48:07 +0200 Subject: [PATCH 0185/1381] net: stmmac: mdio: use phy_find_first to simplify stmmac_mdio_register Simplify the code by using phy_find_first(). Signed-off-by: Heiner Kallweit Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20ca4962-9588-40b8-b021-fb349a92e9e5@gmail.com Signed-off-by: Jakub Kicinski --- .../net/ethernet/stmicro/stmmac/stmmac_mdio.c | 53 ++++++++----------- 1 file changed, 22 insertions(+), 31 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c index d62b2870899d..65db43e9c85e 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c @@ -584,7 +584,8 @@ int stmmac_mdio_register(struct net_device *ndev) struct device *dev = ndev->dev.parent; struct fwnode_handle *fixed_node; struct fwnode_handle *fwnode; - int addr, found, max_addr; + struct phy_device *phydev; + int max_addr; if (!mdio_bus_data) return 0; @@ -668,41 +669,31 @@ int stmmac_mdio_register(struct net_device *ndev) if (priv->plat->phy_node || mdio_node) goto bus_register_done; - found = 0; - for (addr = 0; addr < max_addr; addr++) { - struct phy_device *phydev = mdiobus_get_phy(new_bus, addr); - - if (!phydev) - continue; - - /* - * If an IRQ was provided to be assigned after - * the bus probe, do it here. - */ - if (!mdio_bus_data->irqs && - (mdio_bus_data->probed_phy_irq > 0)) { - new_bus->irq[addr] = mdio_bus_data->probed_phy_irq; - phydev->irq = mdio_bus_data->probed_phy_irq; - } - - /* - * If we're going to bind the MAC to this PHY bus, - * and no PHY number was provided to the MAC, - * use the one probed here. - */ - if (priv->plat->phy_addr == -1) - priv->plat->phy_addr = addr; - - phy_attached_info(phydev); - found = 1; - } - - if (!found && !mdio_node) { + phydev = phy_find_first(new_bus); + if (!phydev || phydev->mdio.addr > max_addr) { dev_warn(dev, "No PHY found\n"); err = -ENODEV; goto no_phy_found; } + /* + * If an IRQ was provided to be assigned after + * the bus probe, do it here. + */ + if (!mdio_bus_data->irqs && mdio_bus_data->probed_phy_irq > 0) { + new_bus->irq[phydev->mdio.addr] = mdio_bus_data->probed_phy_irq; + phydev->irq = mdio_bus_data->probed_phy_irq; + } + + /* + * If we're going to bind the MAC to this PHY bus, and no PHY number + * was provided to the MAC, use the one probed here. + */ + if (priv->plat->phy_addr == -1) + priv->plat->phy_addr = phydev->mdio.addr; + + phy_attached_info(phydev); + bus_register_done: priv->mii = new_bus; From 91f76771dba0e5ec9ef74e55eb50f2855ce583a3 Mon Sep 17 00:00:00 2001 From: Vadim Fedorenko Date: Sun, 19 Oct 2025 22:57:20 +0000 Subject: [PATCH 0186/1381] bnxt_en: support PPS in/out on all pins Add supported_extts_flags and supported_perout_flags configuration to make the driver complaint with the latest API. Initialize channel information to 0 to avoid confusing users, because HW doesn't actually care about channels. Signed-off-by: Vadim Fedorenko Reviewed-by: Pavan Chebbi Link: https://patch.msgid.link/20251019225720.898550-1-vadim.fedorenko@linux.dev Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c index db81cf6d5289..1425a75de9a1 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c @@ -952,7 +952,6 @@ static int bnxt_ptp_pps_init(struct bnxt *bp) snprintf(ptp_info->pin_config[i].name, sizeof(ptp_info->pin_config[i].name), "bnxt_pps%d", i); ptp_info->pin_config[i].index = i; - ptp_info->pin_config[i].chan = i; if (*pin_usg == BNXT_PPS_PIN_PPS_IN) ptp_info->pin_config[i].func = PTP_PF_EXTTS; else if (*pin_usg == BNXT_PPS_PIN_PPS_OUT) @@ -969,6 +968,8 @@ static int bnxt_ptp_pps_init(struct bnxt *bp) ptp_info->n_per_out = 1; ptp_info->pps = 1; ptp_info->verify = bnxt_ptp_verify; + ptp_info->supported_extts_flags = PTP_RISING_EDGE | PTP_STRICT_FLAGS; + ptp_info->supported_perout_flags = PTP_PEROUT_DUTY_CYCLE; return 0; } From 962ac5ca99a5c3e7469215bf47572440402dfd59 Mon Sep 17 00:00:00 2001 From: Jiapeng Chong Date: Mon, 20 Oct 2025 09:44:41 +0800 Subject: [PATCH 0187/1381] net: macb: Remove duplicate linux/inetdevice.h header ./drivers/net/ethernet/cadence/macb_main.c: linux/inetdevice.h is included more than once. Reported-by: Abaci Robot Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=26474 Signed-off-by: Jiapeng Chong Acked-by: Nicolas Ferre Link: https://patch.msgid.link/20251020014441.2070356-1-jiapeng.chong@linux.alibaba.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/cadence/macb_main.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c index 214f543af3b8..39673f5c3337 100644 --- a/drivers/net/ethernet/cadence/macb_main.c +++ b/drivers/net/ethernet/cadence/macb_main.c @@ -14,7 +14,6 @@ #include #include #include -#include #include #include #include From 54be197109762d2a3c2b23d44e960ce7a43a3798 Mon Sep 17 00:00:00 2001 From: Loic Poulain Date: Mon, 20 Oct 2025 17:37:59 +0200 Subject: [PATCH 0188/1381] wifi: ath10k: Support for FTM TLV test commands Existing tools like myftm use 'legacy' test command API. Similarly to ath11k and ath12k, we want to support raw TLV payload submitted from the test tool. This requires segmenting the TLV payload and encapsulating it within a WMI command. The opposite operation needs to be done upon corresponding event receiving. Tested-on: WCN3990 hw1.0 WLAN.HL.3.3.7.c2-00931-QCAHLSWMTPLZ-1 Signed-off-by: Loic Poulain Link: https://patch.msgid.link/20251020153759.407516-1-loic.poulain@oss.qualcomm.com Signed-off-by: Jeff Johnson --- drivers/net/wireless/ath/ath10k/core.h | 4 + drivers/net/wireless/ath/ath10k/testmode.c | 275 ++++++++++++++++--- drivers/net/wireless/ath/ath10k/testmode_i.h | 15 + drivers/net/wireless/ath/ath10k/wmi.h | 19 +- 4 files changed, 271 insertions(+), 42 deletions(-) diff --git a/drivers/net/wireless/ath/ath10k/core.h b/drivers/net/wireless/ath/ath10k/core.h index 859176fcb5a2..73a9db302245 100644 --- a/drivers/net/wireless/ath/ath10k/core.h +++ b/drivers/net/wireless/ath/ath10k/core.h @@ -1259,9 +1259,13 @@ struct ath10k { struct { /* protected by conf_mutex */ struct ath10k_fw_components utf_mode_fw; + u8 ftm_msgref; /* protected by data_lock */ bool utf_monitor; + u32 data_pos; + u32 expected_seq; + u8 *eventdata; } testmode; struct { diff --git a/drivers/net/wireless/ath/ath10k/testmode.c b/drivers/net/wireless/ath/ath10k/testmode.c index 3fcefc55b74f..d3bd385694d6 100644 --- a/drivers/net/wireless/ath/ath10k/testmode.c +++ b/drivers/net/wireless/ath/ath10k/testmode.c @@ -1,6 +1,7 @@ // SPDX-License-Identifier: ISC /* * Copyright (c) 2014-2017 Qualcomm Atheros, Inc. + * Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. */ #include "testmode.h" @@ -10,12 +11,17 @@ #include "debug.h" #include "wmi.h" +#include "wmi-tlv.h" #include "hif.h" #include "hw.h" #include "core.h" #include "testmode_i.h" +#define ATH10K_FTM_SEG_NONE ((u32)-1) +#define ATH10K_FTM_SEGHDR_CURRENT_SEQ GENMASK(3, 0) +#define ATH10K_FTM_SEGHDR_TOTAL_SEGMENTS GENMASK(7, 4) + static const struct nla_policy ath10k_tm_policy[ATH10K_TM_ATTR_MAX + 1] = { [ATH10K_TM_ATTR_CMD] = { .type = NLA_U32 }, [ATH10K_TM_ATTR_DATA] = { .type = NLA_BINARY, @@ -25,14 +31,137 @@ static const struct nla_policy ath10k_tm_policy[ATH10K_TM_ATTR_MAX + 1] = { [ATH10K_TM_ATTR_VERSION_MINOR] = { .type = NLA_U32 }, }; +static void ath10k_tm_event_unsegmented(struct ath10k *ar, u32 cmd_id, + struct sk_buff *skb) +{ + struct sk_buff *nl_skb; + int ret; + + nl_skb = cfg80211_testmode_alloc_event_skb(ar->hw->wiphy, + 2 * sizeof(u32) + skb->len, + GFP_ATOMIC); + if (!nl_skb) { + ath10k_warn(ar, + "failed to allocate skb for testmode wmi event\n"); + return; + } + + ret = nla_put_u32(nl_skb, ATH10K_TM_ATTR_CMD, ATH10K_TM_CMD_WMI); + if (ret) { + ath10k_warn(ar, + "failed to put testmode wmi event cmd attribute: %d\n", + ret); + kfree_skb(nl_skb); + return; + } + + ret = nla_put_u32(nl_skb, ATH10K_TM_ATTR_WMI_CMDID, cmd_id); + if (ret) { + ath10k_warn(ar, + "failed to put testmode wmi event cmd_id: %d\n", + ret); + kfree_skb(nl_skb); + return; + } + + ret = nla_put(nl_skb, ATH10K_TM_ATTR_DATA, skb->len, skb->data); + if (ret) { + ath10k_warn(ar, + "failed to copy skb to testmode wmi event: %d\n", + ret); + kfree_skb(nl_skb); + return; + } + + cfg80211_testmode_event(nl_skb, GFP_ATOMIC); +} + +static void ath10k_tm_event_segmented(struct ath10k *ar, u32 cmd_id, struct sk_buff *skb) +{ + struct wmi_ftm_cmd *ftm = (struct wmi_ftm_cmd *)skb->data; + u8 total_segments, current_seq; + struct sk_buff *nl_skb; + u8 const *buf_pos; + u16 datalen; + u32 data_pos; + int ret; + + if (skb->len < sizeof(*ftm)) { + ath10k_warn(ar, "Invalid ftm event length: %d\n", skb->len); + return; + } + + current_seq = FIELD_GET(ATH10K_FTM_SEGHDR_CURRENT_SEQ, + __le32_to_cpu(ftm->seg_hdr.segmentinfo)); + total_segments = FIELD_GET(ATH10K_FTM_SEGHDR_TOTAL_SEGMENTS, + __le32_to_cpu(ftm->seg_hdr.segmentinfo)); + datalen = skb->len - sizeof(*ftm); + buf_pos = ftm->data; + + if (current_seq == 0) { + ar->testmode.expected_seq = 0; + ar->testmode.data_pos = 0; + } + + data_pos = ar->testmode.data_pos; + + if ((data_pos + datalen) > ATH_FTM_EVENT_MAX_BUF_LENGTH) { + ath10k_warn(ar, "Invalid ftm event length at %u: %u\n", + data_pos, datalen); + ret = -EINVAL; + return; + } + + memcpy(&ar->testmode.eventdata[data_pos], buf_pos, datalen); + data_pos += datalen; + + if (++ar->testmode.expected_seq != total_segments) { + ar->testmode.data_pos = data_pos; + ath10k_dbg(ar, ATH10K_DBG_TESTMODE, "partial data received %u/%u\n", + current_seq + 1, total_segments); + return; + } + + ath10k_dbg(ar, ATH10K_DBG_TESTMODE, "total data length %u\n", data_pos); + + nl_skb = cfg80211_testmode_alloc_event_skb(ar->hw->wiphy, + 2 * sizeof(u32) + data_pos, + GFP_ATOMIC); + if (!nl_skb) { + ath10k_warn(ar, "failed to allocate skb for testmode wmi event\n"); + return; + } + + ret = nla_put_u32(nl_skb, ATH10K_TM_ATTR_CMD, ATH10K_TM_CMD_TLV); + if (ret) { + ath10k_warn(ar, "failed to put testmode wmi event attribute: %d\n", ret); + kfree_skb(nl_skb); + return; + } + + ret = nla_put_u32(nl_skb, ATH10K_TM_ATTR_WMI_CMDID, cmd_id); + if (ret) { + ath10k_warn(ar, "failed to put testmode wmi event cmd_id: %d\n", ret); + kfree_skb(nl_skb); + return; + } + + ret = nla_put(nl_skb, ATH10K_TM_ATTR_DATA, data_pos, &ar->testmode.eventdata[0]); + if (ret) { + ath10k_warn(ar, "failed to copy skb to testmode wmi event: %d\n", ret); + kfree_skb(nl_skb); + return; + } + + cfg80211_testmode_event(nl_skb, GFP_ATOMIC); +} + /* Returns true if callee consumes the skb and the skb should be discarded. * Returns false if skb is not used. Does not sleep. */ bool ath10k_tm_event_wmi(struct ath10k *ar, u32 cmd_id, struct sk_buff *skb) { - struct sk_buff *nl_skb; bool consumed; - int ret; ath10k_dbg(ar, ATH10K_DBG_TESTMODE, "testmode event wmi cmd_id %d skb %p skb->len %d\n", @@ -53,43 +182,10 @@ bool ath10k_tm_event_wmi(struct ath10k *ar, u32 cmd_id, struct sk_buff *skb) */ consumed = true; - nl_skb = cfg80211_testmode_alloc_event_skb(ar->hw->wiphy, - 2 * sizeof(u32) + skb->len, - GFP_ATOMIC); - if (!nl_skb) { - ath10k_warn(ar, - "failed to allocate skb for testmode wmi event\n"); - goto out; - } - - ret = nla_put_u32(nl_skb, ATH10K_TM_ATTR_CMD, ATH10K_TM_CMD_WMI); - if (ret) { - ath10k_warn(ar, - "failed to put testmode wmi event cmd attribute: %d\n", - ret); - kfree_skb(nl_skb); - goto out; - } - - ret = nla_put_u32(nl_skb, ATH10K_TM_ATTR_WMI_CMDID, cmd_id); - if (ret) { - ath10k_warn(ar, - "failed to put testmode wmi event cmd_id: %d\n", - ret); - kfree_skb(nl_skb); - goto out; - } - - ret = nla_put(nl_skb, ATH10K_TM_ATTR_DATA, skb->len, skb->data); - if (ret) { - ath10k_warn(ar, - "failed to copy skb to testmode wmi event: %d\n", - ret); - kfree_skb(nl_skb); - goto out; - } - - cfg80211_testmode_event(nl_skb, GFP_ATOMIC); + if (ar->testmode.expected_seq != ATH10K_FTM_SEG_NONE) + ath10k_tm_event_segmented(ar, cmd_id, skb); + else + ath10k_tm_event_unsegmented(ar, cmd_id, skb); out: spin_unlock_bh(&ar->data_lock); @@ -281,12 +377,18 @@ static int ath10k_tm_cmd_utf_start(struct ath10k *ar, struct nlattr *tb[]) goto err_release_utf_mode_fw; } + ar->testmode.eventdata = kzalloc(ATH_FTM_EVENT_MAX_BUF_LENGTH, GFP_KERNEL); + if (!ar->testmode.eventdata) { + ret = -ENOMEM; + goto err_power_down; + } + ret = ath10k_core_start(ar, ATH10K_FIRMWARE_MODE_UTF, &ar->testmode.utf_mode_fw); if (ret) { ath10k_err(ar, "failed to start core (testmode): %d\n", ret); ar->state = ATH10K_STATE_OFF; - goto err_power_down; + goto err_release_eventdata; } ar->state = ATH10K_STATE_UTF; @@ -302,6 +404,10 @@ static int ath10k_tm_cmd_utf_start(struct ath10k *ar, struct nlattr *tb[]) return 0; +err_release_eventdata: + kfree(ar->testmode.eventdata); + ar->testmode.eventdata = NULL; + err_power_down: ath10k_hif_power_down(ar); @@ -341,6 +447,9 @@ static void __ath10k_tm_cmd_utf_stop(struct ath10k *ar) release_firmware(ar->testmode.utf_mode_fw.fw_file.firmware); ar->testmode.utf_mode_fw.fw_file.firmware = NULL; + kfree(ar->testmode.eventdata); + ar->testmode.eventdata = NULL; + ar->state = ATH10K_STATE_OFF; } @@ -424,6 +533,85 @@ static int ath10k_tm_cmd_wmi(struct ath10k *ar, struct nlattr *tb[]) return ret; } +static int ath10k_tm_cmd_tlv(struct ath10k *ar, struct nlattr *tb[]) +{ + u16 total_bytes, num_segments; + u32 cmd_id, buf_len; + u8 segnumber = 0; + u8 *bufpos; + void *buf; + int ret; + + mutex_lock(&ar->conf_mutex); + + if (ar->state != ATH10K_STATE_UTF) { + ret = -ENETDOWN; + goto out; + } + + buf = nla_data(tb[ATH10K_TM_ATTR_DATA]); + buf_len = nla_len(tb[ATH10K_TM_ATTR_DATA]); + cmd_id = WMI_PDEV_UTF_CMDID; + + ath10k_dbg(ar, ATH10K_DBG_TESTMODE, + "cmd wmi ftm cmd_id %d buffer length %d\n", + cmd_id, buf_len); + ath10k_dbg_dump(ar, ATH10K_DBG_TESTMODE, NULL, "", buf, buf_len); + + bufpos = buf; + total_bytes = buf_len; + num_segments = total_bytes / MAX_WMI_UTF_LEN; + ar->testmode.expected_seq = 0; + + if (buf_len - (num_segments * MAX_WMI_UTF_LEN)) + num_segments++; + + while (buf_len) { + u16 chunk_len = min_t(u16, buf_len, MAX_WMI_UTF_LEN); + struct wmi_ftm_cmd *ftm_cmd; + struct sk_buff *skb; + u32 hdr_info; + u8 seginfo; + + skb = ath10k_wmi_alloc_skb(ar, (chunk_len + + sizeof(struct wmi_ftm_cmd))); + if (!skb) { + ret = -ENOMEM; + goto out; + } + + ftm_cmd = (struct wmi_ftm_cmd *)skb->data; + hdr_info = FIELD_PREP(WMI_TLV_TAG, WMI_TLV_TAG_ARRAY_BYTE) | + FIELD_PREP(WMI_TLV_LEN, (chunk_len + + sizeof(struct wmi_ftm_seg_hdr))); + ftm_cmd->tlv_header = __cpu_to_le32(hdr_info); + ftm_cmd->seg_hdr.len = __cpu_to_le32(total_bytes); + ftm_cmd->seg_hdr.msgref = __cpu_to_le32(ar->testmode.ftm_msgref); + seginfo = FIELD_PREP(ATH10K_FTM_SEGHDR_TOTAL_SEGMENTS, num_segments) | + FIELD_PREP(ATH10K_FTM_SEGHDR_CURRENT_SEQ, segnumber); + ftm_cmd->seg_hdr.segmentinfo = __cpu_to_le32(seginfo); + segnumber++; + + memcpy(&ftm_cmd->data, bufpos, chunk_len); + + ret = ath10k_wmi_cmd_send(ar, skb, cmd_id); + if (ret) { + ath10k_warn(ar, "failed to send wmi ftm command: %d\n", ret); + goto out; + } + + buf_len -= chunk_len; + bufpos += chunk_len; + } + + ar->testmode.ftm_msgref++; + ret = 0; + +out: + mutex_unlock(&ar->conf_mutex); + return ret; +} + int ath10k_tm_cmd(struct ieee80211_hw *hw, struct ieee80211_vif *vif, void *data, int len) { @@ -439,9 +627,14 @@ int ath10k_tm_cmd(struct ieee80211_hw *hw, struct ieee80211_vif *vif, if (!tb[ATH10K_TM_ATTR_CMD]) return -EINVAL; + ar->testmode.expected_seq = ATH10K_FTM_SEG_NONE; + switch (nla_get_u32(tb[ATH10K_TM_ATTR_CMD])) { case ATH10K_TM_CMD_GET_VERSION: - return ath10k_tm_cmd_get_version(ar, tb); + if (!tb[ATH10K_TM_ATTR_DATA]) + return ath10k_tm_cmd_get_version(ar, tb); + else /* ATH10K_TM_CMD_TLV */ + return ath10k_tm_cmd_tlv(ar, tb); case ATH10K_TM_CMD_UTF_START: return ath10k_tm_cmd_utf_start(ar, tb); case ATH10K_TM_CMD_UTF_STOP: diff --git a/drivers/net/wireless/ath/ath10k/testmode_i.h b/drivers/net/wireless/ath/ath10k/testmode_i.h index ee1cb27c1d60..1603f5276682 100644 --- a/drivers/net/wireless/ath/ath10k/testmode_i.h +++ b/drivers/net/wireless/ath/ath10k/testmode_i.h @@ -1,6 +1,7 @@ /* SPDX-License-Identifier: ISC */ /* * Copyright (c) 2014,2017 Qualcomm Atheros, Inc. + * Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. */ /* "API" level of the ath10k testmode interface. Bump it after every @@ -14,6 +15,7 @@ #define ATH10K_TESTMODE_VERSION_MINOR 0 #define ATH10K_TM_DATA_MAX_LEN 5000 +#define ATH_FTM_EVENT_MAX_BUF_LENGTH 2048 enum ath10k_tm_attr { __ATH10K_TM_ATTR_INVALID = 0, @@ -57,4 +59,17 @@ enum ath10k_tm_cmd { * ATH10K_TM_ATTR_DATA. */ ATH10K_TM_CMD_WMI = 3, + + /* The command used to transmit a test command to the firmware + * and the event to receive test events from the firmware. The data + * received only contain the TLV payload, need to add the tlv header + * and send the cmd to firmware with command id WMI_PDEV_UTF_CMDID. + * The data payload size could be large and the driver needs to + * send segmented data to firmware. + * + * This legacy testmode command shares the same value as the get-version + * command. To distinguish between them, we check whether the data attribute + * is present. + */ + ATH10K_TM_CMD_TLV = ATH10K_TM_CMD_GET_VERSION, }; diff --git a/drivers/net/wireless/ath/ath10k/wmi.h b/drivers/net/wireless/ath/ath10k/wmi.h index 0faefc0a9a40..7f50a1de6b97 100644 --- a/drivers/net/wireless/ath/ath10k/wmi.h +++ b/drivers/net/wireless/ath/ath10k/wmi.h @@ -3,7 +3,7 @@ * Copyright (c) 2005-2011 Atheros Communications Inc. * Copyright (c) 2011-2017 Qualcomm Atheros, Inc. * Copyright (c) 2018-2019, The Linux Foundation. All rights reserved. - * Copyright (c) 2021-2024 Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. */ #ifndef _WMI_H_ @@ -7418,6 +7418,23 @@ struct wmi_pdev_bb_timing_cfg_cmd { __le32 bb_xpa_timing; } __packed; +struct wmi_ftm_seg_hdr { + __le32 len; + __le32 msgref; + __le32 segmentinfo; + __le32 pdev_id; +} __packed; + +struct wmi_ftm_cmd { + __le32 tlv_header; + struct wmi_ftm_seg_hdr seg_hdr; + u8 data[]; +} __packed; + +#define WMI_TLV_LEN GENMASK(15, 0) +#define WMI_TLV_TAG GENMASK(31, 16) +#define MAX_WMI_UTF_LEN 252 + struct ath10k; struct ath10k_vif; struct ath10k_fw_stats_pdev; From 26ab9830beabda863766be4a79dc590c7645f4d9 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Tue, 21 Oct 2025 08:26:49 +0100 Subject: [PATCH 0189/1381] net: stmmac: replace has_xxxx with core_type Replace the has_gmac, has_gmac4 and has_xgmac ints, of which only one can be set when matching a core to its driver backend, with an enumerated type carrying the DWMAC core type. Tested-by: Maxime Chevallier Signed-off-by: Russell King (Oracle) Acked-by: Chen-Yu Tsai Reviewed-by: Maxime Chevallier Tested-by: Mohd Ayaan Anwar Reviewed-by: Bartosz Golaszewski Link: https://patch.msgid.link/E1vB6ld-0000000BIPy-2Qi4@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/common.h | 5 ++ .../stmicro/stmmac/dwmac-dwc-qos-eth.c | 2 +- .../net/ethernet/stmicro/stmmac/dwmac-intel.c | 5 +- .../ethernet/stmicro/stmmac/dwmac-ipq806x.c | 2 +- .../ethernet/stmicro/stmmac/dwmac-loongson.c | 2 +- .../ethernet/stmicro/stmmac/dwmac-lpc18xx.c | 2 +- .../stmicro/stmmac/dwmac-qcom-ethqos.c | 2 +- .../net/ethernet/stmicro/stmmac/dwmac-rk.c | 4 +- .../net/ethernet/stmicro/stmmac/dwmac-s32.c | 2 +- .../ethernet/stmicro/stmmac/dwmac-socfpga.c | 2 +- .../net/ethernet/stmicro/stmmac/dwmac-sunxi.c | 2 +- .../net/ethernet/stmicro/stmmac/dwmac-tegra.c | 2 +- drivers/net/ethernet/stmicro/stmmac/hwif.c | 73 +++++++------------ .../net/ethernet/stmicro/stmmac/stmmac_est.c | 4 +- .../ethernet/stmicro/stmmac/stmmac_ethtool.c | 13 ++-- .../net/ethernet/stmicro/stmmac/stmmac_main.c | 34 +++++---- .../net/ethernet/stmicro/stmmac/stmmac_mdio.c | 14 ++-- .../net/ethernet/stmicro/stmmac/stmmac_pci.c | 4 +- .../ethernet/stmicro/stmmac/stmmac_platform.c | 9 +-- .../net/ethernet/stmicro/stmmac/stmmac_ptp.c | 4 +- include/linux/stmmac.h | 11 ++- 21 files changed, 94 insertions(+), 104 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/common.h b/drivers/net/ethernet/stmicro/stmmac/common.h index fee7021246b1..31254ba525d5 100644 --- a/drivers/net/ethernet/stmicro/stmmac/common.h +++ b/drivers/net/ethernet/stmicro/stmmac/common.h @@ -43,6 +43,11 @@ #define DWXGMAC_ID 0x76 #define DWXLGMAC_ID 0x27 +static inline bool dwmac_is_xmac(enum dwmac_core_type core_type) +{ + return core_type == DWMAC_CORE_GMAC4 || core_type == DWMAC_CORE_XGMAC; +} + #define STMMAC_CHAN0 0 /* Always supported and default for all chips */ /* TX and RX Descriptor Length, these need to be power of two. diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-dwc-qos-eth.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-dwc-qos-eth.c index f1c2e35badf7..c7cd6497d42d 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-dwc-qos-eth.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-dwc-qos-eth.c @@ -109,7 +109,7 @@ static int dwc_eth_dwmac_config_dt(struct platform_device *pdev, } /* dwc-qos needs GMAC4, AAL, TSO and PMT */ - plat_dat->has_gmac4 = 1; + plat_dat->core_type = DWMAC_CORE_GMAC4; plat_dat->dma_cfg->aal = 1; plat_dat->flags |= STMMAC_FLAG_TSO_EN; plat_dat->pmt = 1; diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c index e74d00984b88..b2194e414ec1 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c @@ -565,7 +565,7 @@ static void common_default_data(struct plat_stmmacenet_data *plat) { /* clk_csr_i = 20-35MHz & MDC = clk_csr_i/16 */ plat->clk_csr = STMMAC_CSR_20_35M; - plat->has_gmac = 1; + plat->core_type = DWMAC_CORE_GMAC; plat->force_sf_dma_mode = 1; plat->mdio_bus_data->needs_reset = true; @@ -612,8 +612,7 @@ static int intel_mgbe_common_data(struct pci_dev *pdev, plat->pdev = pdev; plat->phy_addr = -1; plat->clk_csr = STMMAC_CSR_250_300M; - plat->has_gmac = 0; - plat->has_gmac4 = 1; + plat->core_type = DWMAC_CORE_GMAC4; plat->force_sf_dma_mode = 0; plat->flags |= (STMMAC_FLAG_TSO_EN | STMMAC_FLAG_SPH_DISABLE); diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-ipq806x.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-ipq806x.c index ca4035cbb55b..c05f85534f0c 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-ipq806x.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-ipq806x.c @@ -473,7 +473,7 @@ static int ipq806x_gmac_probe(struct platform_device *pdev) return err; } - plat_dat->has_gmac = true; + plat_dat->core_type = DWMAC_CORE_GMAC; plat_dat->bsp_priv = gmac; plat_dat->set_clk_tx_rate = ipq806x_gmac_set_clk_tx_rate; plat_dat->multicast_filter_bins = 0; diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c index 592aa9d636e5..2a3ac0136cdb 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c @@ -92,7 +92,7 @@ static void loongson_default_data(struct pci_dev *pdev, /* clk_csr_i = 20-35MHz & MDC = clk_csr_i/16 */ plat->clk_csr = STMMAC_CSR_20_35M; - plat->has_gmac = 1; + plat->core_type = DWMAC_CORE_GMAC; plat->force_sf_dma_mode = 1; /* Set default value for multicast hash bins */ diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-lpc18xx.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-lpc18xx.c index 2562a6d036a2..6fffc9dfbae5 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-lpc18xx.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-lpc18xx.c @@ -41,7 +41,7 @@ static int lpc18xx_dwmac_probe(struct platform_device *pdev) if (IS_ERR(plat_dat)) return PTR_ERR(plat_dat); - plat_dat->has_gmac = true; + plat_dat->core_type = DWMAC_CORE_GMAC; reg = syscon_regmap_lookup_by_compatible("nxp,lpc1850-creg"); if (IS_ERR(reg)) { diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c index 32244217d952..d1e48b524d7a 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c @@ -846,7 +846,7 @@ static int qcom_ethqos_probe(struct platform_device *pdev) plat_dat->fix_mac_speed = ethqos_fix_mac_speed; plat_dat->dump_debug_regs = rgmii_dump; plat_dat->ptp_clk_freq_config = ethqos_ptp_clk_freq_config; - plat_dat->has_gmac4 = 1; + plat_dat->core_type = DWMAC_CORE_GMAC4; if (ethqos->has_emac_ge_3) plat_dat->dwmac4_addrs = &data->dwmac4_addrs; plat_dat->pmt = 1; diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c index 51ea0caf16c1..9b92f4d335cc 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c @@ -1750,8 +1750,8 @@ static int rk_gmac_probe(struct platform_device *pdev) /* If the stmmac is not already selected as gmac4, * then make sure we fallback to gmac. */ - if (!plat_dat->has_gmac4) { - plat_dat->has_gmac = true; + if (plat_dat->core_type != DWMAC_CORE_GMAC4) { + plat_dat->core_type = DWMAC_CORE_GMAC; plat_dat->rx_fifo_size = 4096; plat_dat->tx_fifo_size = 2048; } diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-s32.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-s32.c index 221539d760bc..ee095ac13203 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-s32.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-s32.c @@ -146,7 +146,7 @@ static int s32_dwmac_probe(struct platform_device *pdev) gmac->ioaddr = res.addr; /* S32CC core feature set */ - plat->has_gmac4 = true; + plat->core_type = DWMAC_CORE_GMAC4; plat->pmt = 1; plat->flags |= STMMAC_FLAG_SPH_DISABLE; plat->rx_fifo_size = 20480; diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c index 354f01184e6c..2ff5db6d41ca 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c @@ -497,7 +497,7 @@ static int socfpga_dwmac_probe(struct platform_device *pdev) plat_dat->pcs_init = socfpga_dwmac_pcs_init; plat_dat->pcs_exit = socfpga_dwmac_pcs_exit; plat_dat->select_pcs = socfpga_dwmac_select_pcs; - plat_dat->has_gmac = true; + plat_dat->core_type = DWMAC_CORE_GMAC; plat_dat->riwt_off = 1; diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-sunxi.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-sunxi.c index 1eadcf5d1ad6..7f560d78209d 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-sunxi.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-sunxi.c @@ -136,7 +136,7 @@ static int sun7i_gmac_probe(struct platform_device *pdev) /* platform data specifying hardware features and callbacks. * hardware features were copied from Allwinner drivers. */ plat_dat->tx_coe = 1; - plat_dat->has_gmac = true; + plat_dat->core_type = DWMAC_CORE_GMAC; plat_dat->bsp_priv = gmac; plat_dat->init = sun7i_gmac_init; plat_dat->exit = sun7i_gmac_exit; diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-tegra.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-tegra.c index dc903b846b1b..d765acbe3754 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-tegra.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-tegra.c @@ -308,7 +308,7 @@ static int tegra_mgbe_probe(struct platform_device *pdev) goto disable_clks; } - plat->has_xgmac = 1; + plat->core_type = DWMAC_CORE_XGMAC; plat->flags |= STMMAC_FLAG_TSO_EN; plat->pmt = 1; plat->bsp_priv = mgbe; diff --git a/drivers/net/ethernet/stmicro/stmmac/hwif.c b/drivers/net/ethernet/stmicro/stmmac/hwif.c index 3f7c765dcb79..00083ce52549 100644 --- a/drivers/net/ethernet/stmicro/stmmac/hwif.c +++ b/drivers/net/ethernet/stmicro/stmmac/hwif.c @@ -106,9 +106,7 @@ int stmmac_reset(struct stmmac_priv *priv, void __iomem *ioaddr) } static const struct stmmac_hwif_entry { - bool gmac; - bool gmac4; - bool xgmac; + enum dwmac_core_type core_type; u32 min_id; u32 dev_id; const struct stmmac_regs_off regs; @@ -127,9 +125,7 @@ static const struct stmmac_hwif_entry { } stmmac_hw[] = { /* NOTE: New HW versions shall go to the end of this table */ { - .gmac = false, - .gmac4 = false, - .xgmac = false, + .core_type = DWMAC_CORE_MAC100, .min_id = 0, .regs = { .ptp_off = PTP_GMAC3_X_OFFSET, @@ -146,9 +142,7 @@ static const struct stmmac_hwif_entry { .setup = dwmac100_setup, .quirks = stmmac_dwmac1_quirks, }, { - .gmac = true, - .gmac4 = false, - .xgmac = false, + .core_type = DWMAC_CORE_GMAC, .min_id = 0, .regs = { .ptp_off = PTP_GMAC3_X_OFFSET, @@ -165,9 +159,7 @@ static const struct stmmac_hwif_entry { .setup = dwmac1000_setup, .quirks = stmmac_dwmac1_quirks, }, { - .gmac = false, - .gmac4 = true, - .xgmac = false, + .core_type = DWMAC_CORE_GMAC4, .min_id = 0, .regs = { .ptp_off = PTP_GMAC4_OFFSET, @@ -187,9 +179,7 @@ static const struct stmmac_hwif_entry { .setup = dwmac4_setup, .quirks = stmmac_dwmac4_quirks, }, { - .gmac = false, - .gmac4 = true, - .xgmac = false, + .core_type = DWMAC_CORE_GMAC4, .min_id = DWMAC_CORE_4_00, .regs = { .ptp_off = PTP_GMAC4_OFFSET, @@ -210,9 +200,7 @@ static const struct stmmac_hwif_entry { .setup = dwmac4_setup, .quirks = NULL, }, { - .gmac = false, - .gmac4 = true, - .xgmac = false, + .core_type = DWMAC_CORE_GMAC4, .min_id = DWMAC_CORE_4_10, .regs = { .ptp_off = PTP_GMAC4_OFFSET, @@ -233,9 +221,7 @@ static const struct stmmac_hwif_entry { .setup = dwmac4_setup, .quirks = NULL, }, { - .gmac = false, - .gmac4 = true, - .xgmac = false, + .core_type = DWMAC_CORE_GMAC4, .min_id = DWMAC_CORE_5_10, .regs = { .ptp_off = PTP_GMAC4_OFFSET, @@ -256,9 +242,7 @@ static const struct stmmac_hwif_entry { .setup = dwmac4_setup, .quirks = NULL, }, { - .gmac = false, - .gmac4 = false, - .xgmac = true, + .core_type = DWMAC_CORE_XGMAC, .min_id = DWXGMAC_CORE_2_10, .dev_id = DWXGMAC_ID, .regs = { @@ -280,9 +264,7 @@ static const struct stmmac_hwif_entry { .setup = dwxgmac2_setup, .quirks = NULL, }, { - .gmac = false, - .gmac4 = false, - .xgmac = true, + .core_type = DWMAC_CORE_XGMAC, .min_id = DWXLGMAC_CORE_2_00, .dev_id = DWXLGMAC_ID, .regs = { @@ -308,20 +290,18 @@ static const struct stmmac_hwif_entry { int stmmac_hwif_init(struct stmmac_priv *priv) { - bool needs_xgmac = priv->plat->has_xgmac; - bool needs_gmac4 = priv->plat->has_gmac4; - bool needs_gmac = priv->plat->has_gmac; + enum dwmac_core_type core_type = priv->plat->core_type; const struct stmmac_hwif_entry *entry; struct mac_device_info *mac; bool needs_setup = true; u32 id, dev_id = 0; int i, ret; - if (needs_gmac) { + if (core_type == DWMAC_CORE_GMAC) { id = stmmac_get_id(priv, GMAC_VERSION); - } else if (needs_gmac4 || needs_xgmac) { + } else if (dwmac_is_xmac(core_type)) { id = stmmac_get_id(priv, GMAC4_VERSION); - if (needs_xgmac) + if (core_type == DWMAC_CORE_XGMAC) dev_id = stmmac_get_dev_id(priv, GMAC4_VERSION); } else { id = 0; @@ -331,14 +311,16 @@ int stmmac_hwif_init(struct stmmac_priv *priv) priv->synopsys_id = id; /* Lets assume some safe values first */ - priv->ptpaddr = priv->ioaddr + - (needs_gmac4 ? PTP_GMAC4_OFFSET : PTP_GMAC3_X_OFFSET); - priv->mmcaddr = priv->ioaddr + - (needs_gmac4 ? MMC_GMAC4_OFFSET : MMC_GMAC3_X_OFFSET); - if (needs_gmac4) + if (core_type == DWMAC_CORE_GMAC4) { + priv->ptpaddr = priv->ioaddr + PTP_GMAC4_OFFSET; + priv->mmcaddr = priv->ioaddr + MMC_GMAC4_OFFSET; priv->estaddr = priv->ioaddr + EST_GMAC4_OFFSET; - else if (needs_xgmac) - priv->estaddr = priv->ioaddr + EST_XGMAC_OFFSET; + } else { + priv->ptpaddr = priv->ioaddr + PTP_GMAC3_X_OFFSET; + priv->mmcaddr = priv->ioaddr + MMC_GMAC3_X_OFFSET; + if (core_type == DWMAC_CORE_XGMAC) + priv->estaddr = priv->ioaddr + EST_XGMAC_OFFSET; + } /* Check for HW specific setup first */ if (priv->plat->setup) { @@ -355,16 +337,12 @@ int stmmac_hwif_init(struct stmmac_priv *priv) for (i = ARRAY_SIZE(stmmac_hw) - 1; i >= 0; i--) { entry = &stmmac_hw[i]; - if (needs_gmac ^ entry->gmac) - continue; - if (needs_gmac4 ^ entry->gmac4) - continue; - if (needs_xgmac ^ entry->xgmac) + if (core_type != entry->core_type) continue; /* Use synopsys_id var because some setups can override this */ if (priv->synopsys_id < entry->min_id) continue; - if (needs_xgmac && (dev_id ^ entry->dev_id)) + if (core_type == DWMAC_CORE_XGMAC && (dev_id ^ entry->dev_id)) continue; /* Only use generic HW helpers if needed */ @@ -400,6 +378,7 @@ int stmmac_hwif_init(struct stmmac_priv *priv) } dev_err(priv->device, "Failed to find HW IF (id=0x%x, gmac=%d/%d)\n", - id, needs_gmac, needs_gmac4); + id, core_type == DWMAC_CORE_GMAC, + core_type == DWMAC_CORE_GMAC4); return -EINVAL; } diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_est.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_est.c index 4b513d27a988..afc516059b89 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_est.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_est.c @@ -53,7 +53,7 @@ static int est_configure(struct stmmac_priv *priv, struct stmmac_est *cfg, } ctrl = readl(est_addr + EST_CONTROL); - if (priv->plat->has_xgmac) { + if (priv->plat->core_type == DWMAC_CORE_XGMAC) { ctrl &= ~EST_XGMAC_PTOV; ctrl |= ((NSEC_PER_SEC / ptp_rate) * EST_XGMAC_PTOV_MUL) << EST_XGMAC_PTOV_SHIFT; @@ -148,7 +148,7 @@ static void est_irq_status(struct stmmac_priv *priv, struct net_device *dev, } if (status & EST_BTRE) { - if (priv->plat->has_xgmac) { + if (priv->plat->core_type == DWMAC_CORE_XGMAC) { btrl = FIELD_GET(EST_XGMAC_BTRL, status); btrl_max = FIELD_MAX(EST_XGMAC_BTRL); } else { diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c index c60cd948311e..df016c4eb710 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c @@ -303,9 +303,10 @@ static void stmmac_ethtool_getdrvinfo(struct net_device *dev, { struct stmmac_priv *priv = netdev_priv(dev); - if (priv->plat->has_gmac || priv->plat->has_gmac4) + if (priv->plat->core_type == DWMAC_CORE_GMAC || + priv->plat->core_type == DWMAC_CORE_GMAC4) strscpy(info->driver, GMAC_ETHTOOL_NAME, sizeof(info->driver)); - else if (priv->plat->has_xgmac) + else if (priv->plat->core_type == DWMAC_CORE_XGMAC) strscpy(info->driver, XGMAC_ETHTOOL_NAME, sizeof(info->driver)); else strscpy(info->driver, MAC100_ETHTOOL_NAME, @@ -351,9 +352,9 @@ static int stmmac_ethtool_get_regs_len(struct net_device *dev) { struct stmmac_priv *priv = netdev_priv(dev); - if (priv->plat->has_xgmac) + if (priv->plat->core_type == DWMAC_CORE_XGMAC) return XGMAC_REGSIZE * 4; - else if (priv->plat->has_gmac4) + else if (priv->plat->core_type == DWMAC_CORE_GMAC4) return GMAC4_REG_SPACE_SIZE; return REG_SPACE_SIZE; } @@ -368,12 +369,12 @@ static void stmmac_ethtool_gregs(struct net_device *dev, stmmac_dump_dma_regs(priv, priv->ioaddr, reg_space); /* Copy DMA registers to where ethtool expects them */ - if (priv->plat->has_gmac4) { + if (priv->plat->core_type == DWMAC_CORE_GMAC4) { /* GMAC4 dumps its DMA registers at its DMA_CHAN_BASE_ADDR */ memcpy(®_space[ETHTOOL_DMA_OFFSET], ®_space[GMAC4_DMA_CHAN_BASE_ADDR / 4], NUM_DWMAC4_DMA_REGS * 4); - } else if (!priv->plat->has_xgmac) { + } else if (priv->plat->core_type != DWMAC_CORE_XGMAC) { memcpy(®_space[ETHTOOL_DMA_OFFSET], ®_space[DMA_BUS_MODE / 4], NUM_DWMAC1000_DMA_REGS * 4); diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index 5e6aaead5894..9fa3c221a0c3 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -446,7 +446,7 @@ static void stmmac_get_rx_hwtstamp(struct stmmac_priv *priv, struct dma_desc *p, if (!priv->hwts_rx_en) return; /* For GMAC4, the valid timestamp is from CTX next desc. */ - if (priv->plat->has_gmac4 || priv->plat->has_xgmac) + if (dwmac_is_xmac(priv->plat->core_type)) desc = np; /* Check if timestamp is available */ @@ -697,7 +697,7 @@ static int stmmac_hwtstamp_get(struct net_device *dev, static int stmmac_init_tstamp_counter(struct stmmac_priv *priv, u32 systime_flags) { - bool xmac = priv->plat->has_gmac4 || priv->plat->has_xgmac; + bool xmac = dwmac_is_xmac(priv->plat->core_type); struct timespec64 now; u32 sec_inc = 0; u64 temp = 0; @@ -746,7 +746,7 @@ static int stmmac_init_tstamp_counter(struct stmmac_priv *priv, */ static int stmmac_init_timestamping(struct stmmac_priv *priv) { - bool xmac = priv->plat->has_gmac4 || priv->plat->has_xgmac; + bool xmac = dwmac_is_xmac(priv->plat->core_type); int ret; if (priv->plat->ptp_clk_freq_config) @@ -2413,7 +2413,7 @@ static void stmmac_dma_operation_mode(struct stmmac_priv *priv) txfifosz = priv->dma_cap.tx_fifo_size; /* Split up the shared Tx/Rx FIFO memory on DW QoS Eth and DW XGMAC */ - if (priv->plat->has_gmac4 || priv->plat->has_xgmac) { + if (dwmac_is_xmac(priv->plat->core_type)) { rxfifosz /= rx_channels_count; txfifosz /= tx_channels_count; } @@ -4520,7 +4520,8 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev) if (skb_is_gso(skb) && priv->tso) { if (gso & (SKB_GSO_TCPV4 | SKB_GSO_TCPV6)) return stmmac_tso_xmit(skb, dev); - if (priv->plat->has_gmac4 && (gso & SKB_GSO_UDP_L4)) + if (priv->plat->core_type == DWMAC_CORE_GMAC4 && + (gso & SKB_GSO_UDP_L4)) return stmmac_tso_xmit(skb, dev); } @@ -5973,7 +5974,7 @@ static void stmmac_common_interrupt(struct stmmac_priv *priv) u32 queue; bool xmac; - xmac = priv->plat->has_gmac4 || priv->plat->has_xgmac; + xmac = dwmac_is_xmac(priv->plat->core_type); queues_count = (rx_cnt > tx_cnt) ? rx_cnt : tx_cnt; if (priv->irq_wake) @@ -5987,7 +5988,7 @@ static void stmmac_common_interrupt(struct stmmac_priv *priv) stmmac_fpe_irq_status(priv); /* To handle GMAC own interrupts */ - if ((priv->plat->has_gmac) || xmac) { + if (priv->plat->core_type == DWMAC_CORE_GMAC || xmac) { int status = stmmac_host_irq_status(priv, priv->hw, &priv->xstats); if (unlikely(status)) { @@ -6348,7 +6349,7 @@ static int stmmac_dma_cap_show(struct seq_file *seq, void *v) (priv->dma_cap.mbps_1000) ? "Y" : "N"); seq_printf(seq, "\tHalf duplex: %s\n", (priv->dma_cap.half_duplex) ? "Y" : "N"); - if (priv->plat->has_xgmac) { + if (priv->plat->core_type == DWMAC_CORE_XGMAC) { seq_printf(seq, "\tNumber of Additional MAC address registers: %d\n", priv->dma_cap.multi_addr); @@ -6372,7 +6373,7 @@ static int stmmac_dma_cap_show(struct seq_file *seq, void *v) (priv->dma_cap.time_stamp) ? "Y" : "N"); seq_printf(seq, "\tIEEE 1588-2008 Advanced Time Stamp: %s\n", (priv->dma_cap.atime_stamp) ? "Y" : "N"); - if (priv->plat->has_xgmac) + if (priv->plat->core_type == DWMAC_CORE_XGMAC) seq_printf(seq, "\tTimestamp System Time Source: %s\n", dwxgmac_timestamp_source[priv->dma_cap.tssrc]); seq_printf(seq, "\t802.3az - Energy-Efficient Ethernet (EEE): %s\n", @@ -6381,7 +6382,7 @@ static int stmmac_dma_cap_show(struct seq_file *seq, void *v) seq_printf(seq, "\tChecksum Offload in TX: %s\n", (priv->dma_cap.tx_coe) ? "Y" : "N"); if (priv->synopsys_id >= DWMAC_CORE_4_00 || - priv->plat->has_xgmac) { + priv->plat->core_type == DWMAC_CORE_XGMAC) { seq_printf(seq, "\tIP Checksum Offload in RX: %s\n", (priv->dma_cap.rx_coe) ? "Y" : "N"); } else { @@ -7233,8 +7234,9 @@ static int stmmac_hw_init(struct stmmac_priv *priv) * has to be disable and this can be done by passing the * riwt_off field from the platform. */ - if (((priv->synopsys_id >= DWMAC_CORE_3_50) || - (priv->plat->has_xgmac)) && (!priv->plat->riwt_off)) { + if ((priv->synopsys_id >= DWMAC_CORE_3_50 || + priv->plat->core_type == DWMAC_CORE_XGMAC) && + !priv->plat->riwt_off) { priv->use_riwt = 1; dev_info(priv->device, "Enable RX Mitigation via HW Watchdog Timer\n"); @@ -7355,7 +7357,7 @@ static int stmmac_xdp_rx_timestamp(const struct xdp_md *_ctx, u64 *timestamp) return -ENODATA; /* For GMAC4, the valid timestamp is from CTX next desc. */ - if (priv->plat->has_gmac4 || priv->plat->has_xgmac) + if (dwmac_is_xmac(priv->plat->core_type)) desc_contains_ts = ndesc; /* Check if timestamp is available */ @@ -7511,7 +7513,7 @@ int stmmac_dvr_probe(struct device *device, if ((priv->plat->flags & STMMAC_FLAG_TSO_EN) && (priv->dma_cap.tsoen)) { ndev->hw_features |= NETIF_F_TSO | NETIF_F_TSO6; - if (priv->plat->has_gmac4) + if (priv->plat->core_type == DWMAC_CORE_GMAC4) ndev->hw_features |= NETIF_F_GSO_UDP_L4; priv->tso = true; dev_info(priv->device, "TSO feature enabled\n"); @@ -7564,7 +7566,7 @@ int stmmac_dvr_probe(struct device *device, #ifdef STMMAC_VLAN_TAG_USED /* Both mac100 and gmac support receive VLAN tag detection */ ndev->features |= NETIF_F_HW_VLAN_CTAG_RX | NETIF_F_HW_VLAN_STAG_RX; - if (priv->plat->has_gmac4 || priv->plat->has_xgmac) { + if (dwmac_is_xmac(priv->plat->core_type)) { ndev->hw_features |= NETIF_F_HW_VLAN_CTAG_RX; priv->hw->hw_vlan_en = true; } @@ -7595,7 +7597,7 @@ int stmmac_dvr_probe(struct device *device, /* MTU range: 46 - hw-specific max */ ndev->min_mtu = ETH_ZLEN - ETH_HLEN; - if (priv->plat->has_xgmac) + if (priv->plat->core_type == DWMAC_CORE_XGMAC) ndev->max_mtu = XGMAC_JUMBO_LEN; else if ((priv->plat->enh_desc) || (priv->synopsys_id >= DWMAC_CORE_4_00)) ndev->max_mtu = JUMBO_LEN; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c index 65db43e9c85e..3f8cc3293964 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c @@ -301,7 +301,7 @@ static int stmmac_mdio_read_c22(struct mii_bus *bus, int phyaddr, int phyreg) struct stmmac_priv *priv = netdev_priv(bus->priv); u32 cmd; - if (priv->plat->has_gmac4) + if (priv->plat->core_type == DWMAC_CORE_GMAC4) cmd = MII_GMAC4_READ; else cmd = 0; @@ -344,7 +344,7 @@ static int stmmac_mdio_write_c22(struct mii_bus *bus, int phyaddr, int phyreg, struct stmmac_priv *priv = netdev_priv(bus->priv); u32 cmd; - if (priv->plat->has_gmac4) + if (priv->plat->core_type == DWMAC_CORE_GMAC4) cmd = MII_GMAC4_WRITE; else cmd = MII_ADDR_GWRITE; @@ -417,7 +417,7 @@ int stmmac_mdio_reset(struct mii_bus *bus) * on MDC, so perform a dummy mdio read. To be updated for GMAC4 * if needed. */ - if (!priv->plat->has_gmac4) + if (priv->plat->core_type != DWMAC_CORE_GMAC4) writel(0, priv->ioaddr + mii_address); #endif return 0; @@ -528,7 +528,7 @@ static u32 stmmac_clk_csr_set(struct stmmac_priv *priv) value = 0; } - if (priv->plat->has_xgmac) { + if (priv->plat->core_type == DWMAC_CORE_XGMAC) { if (clk_rate > 400000000) value = 0x5; else if (clk_rate > 350000000) @@ -601,7 +601,7 @@ int stmmac_mdio_register(struct net_device *ndev) new_bus->name = "stmmac"; - if (priv->plat->has_xgmac) { + if (priv->plat->core_type == DWMAC_CORE_XGMAC) { new_bus->read = &stmmac_xgmac2_mdio_read_c22; new_bus->write = &stmmac_xgmac2_mdio_write_c22; new_bus->read_c45 = &stmmac_xgmac2_mdio_read_c45; @@ -622,7 +622,7 @@ int stmmac_mdio_register(struct net_device *ndev) } else { new_bus->read = &stmmac_mdio_read_c22; new_bus->write = &stmmac_mdio_write_c22; - if (priv->plat->has_gmac4) { + if (priv->plat->core_type == DWMAC_CORE_GMAC4) { new_bus->read_c45 = &stmmac_mdio_read_c45; new_bus->write_c45 = &stmmac_mdio_write_c45; } @@ -650,7 +650,7 @@ int stmmac_mdio_register(struct net_device *ndev) } /* Looks like we need a dummy read for XGMAC only and C45 PHYs */ - if (priv->plat->has_xgmac) + if (priv->plat->core_type == DWMAC_CORE_XGMAC) stmmac_xgmac2_mdio_read_c45(new_bus, 0, 0, 0); /* If fixed-link is set, skip PHY scanning */ diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c index 4e3aa611fda8..94b3a3b27270 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c @@ -23,7 +23,7 @@ static void common_default_data(struct plat_stmmacenet_data *plat) { /* clk_csr_i = 20-35MHz & MDC = clk_csr_i/16 */ plat->clk_csr = STMMAC_CSR_20_35M; - plat->has_gmac = 1; + plat->core_type = DWMAC_CORE_GMAC; plat->force_sf_dma_mode = 1; plat->mdio_bus_data->needs_reset = true; @@ -76,7 +76,7 @@ static int snps_gmac5_default_data(struct pci_dev *pdev, int i; plat->clk_csr = STMMAC_CSR_250_300M; - plat->has_gmac4 = 1; + plat->core_type = DWMAC_CORE_GMAC4; plat->force_sf_dma_mode = 1; plat->flags |= STMMAC_FLAG_TSO_EN; plat->pmt = 1; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c index 27bcaae07a7f..fbb92cc6ab59 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c @@ -552,12 +552,12 @@ stmmac_probe_config_dt(struct platform_device *pdev, u8 *mac) &pdev->dev, plat->unicast_filter_entries); plat->multicast_filter_bins = dwmac1000_validate_mcast_bins( &pdev->dev, plat->multicast_filter_bins); - plat->has_gmac = 1; + plat->core_type = DWMAC_CORE_GMAC; plat->pmt = 1; } if (of_device_is_compatible(np, "snps,dwmac-3.40a")) { - plat->has_gmac = 1; + plat->core_type = DWMAC_CORE_GMAC; plat->enh_desc = 1; plat->tx_coe = 1; plat->bugged_jumbo = 1; @@ -565,8 +565,7 @@ stmmac_probe_config_dt(struct platform_device *pdev, u8 *mac) } if (of_device_compatible_match(np, stmmac_gmac4_compats)) { - plat->has_gmac4 = 1; - plat->has_gmac = 0; + plat->core_type = DWMAC_CORE_GMAC4; plat->pmt = 1; if (of_property_read_bool(np, "snps,tso")) plat->flags |= STMMAC_FLAG_TSO_EN; @@ -580,7 +579,7 @@ stmmac_probe_config_dt(struct platform_device *pdev, u8 *mac) } if (of_device_is_compatible(np, "snps,dwxgmac")) { - plat->has_xgmac = 1; + plat->core_type = DWMAC_CORE_XGMAC; plat->pmt = 1; if (of_property_read_bool(np, "snps,tso")) plat->flags |= STMMAC_FLAG_TSO_EN; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_ptp.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_ptp.c index 993ff4e87e55..3e30172fa129 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_ptp.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_ptp.c @@ -57,7 +57,7 @@ static int stmmac_adjust_time(struct ptp_clock_info *ptp, s64 delta) bool xmac, est_rst = false; int ret; - xmac = priv->plat->has_gmac4 || priv->plat->has_xgmac; + xmac = dwmac_is_xmac(priv->plat->core_type); if (delta < 0) { neg_adj = 1; @@ -344,7 +344,7 @@ void stmmac_ptp_register(struct stmmac_priv *priv) /* Calculate the clock domain crossing (CDC) error if necessary */ priv->plat->cdc_error_adj = 0; - if (priv->plat->has_gmac4) + if (priv->plat->core_type == DWMAC_CORE_GMAC4) priv->plat->cdc_error_adj = (2 * NSEC_PER_SEC) / priv->plat->clk_ptp_rate; /* Update the ptp clock parameters based on feature discovery, when diff --git a/include/linux/stmmac.h b/include/linux/stmmac.h index 99022620457a..151c81c560c8 100644 --- a/include/linux/stmmac.h +++ b/include/linux/stmmac.h @@ -171,6 +171,13 @@ struct dwmac4_addrs { u32 mtl_low_cred_offset; }; +enum dwmac_core_type { + DWMAC_CORE_MAC100, + DWMAC_CORE_GMAC, + DWMAC_CORE_GMAC4, + DWMAC_CORE_XGMAC, +}; + #define STMMAC_FLAG_SPH_DISABLE BIT(1) #define STMMAC_FLAG_USE_PHY_WOL BIT(2) #define STMMAC_FLAG_HAS_SUN8I BIT(3) @@ -186,6 +193,7 @@ struct dwmac4_addrs { #define STMMAC_FLAG_HWTSTAMP_CORRECT_LATENCY BIT(13) struct plat_stmmacenet_data { + enum dwmac_core_type core_type; int bus_id; int phy_addr; /* MAC ----- optional PCS ----- SerDes ----- optional PHY ----- Media @@ -219,7 +227,6 @@ struct plat_stmmacenet_data { struct stmmac_dma_cfg *dma_cfg; struct stmmac_safety_feature_cfg *safety_feat_cfg; int clk_csr; - int has_gmac; int enh_desc; int tx_coe; int rx_coe; @@ -282,10 +289,8 @@ struct plat_stmmacenet_data { struct reset_control *stmmac_rst; struct reset_control *stmmac_ahb_rst; struct stmmac_axi *axi; - int has_gmac4; int rss_en; int mac_port_sel_speed; - int has_xgmac; u8 vlan_fail_q; struct pci_dev *pdev; int int_snapshot_num; From 10e0378f05d2b89c4f88f04ec56b2a5b9a116648 Mon Sep 17 00:00:00 2001 From: "Gustavo A. R. Silva" Date: Tue, 21 Oct 2025 12:54:10 +0100 Subject: [PATCH 0190/1381] net: spacemit: Avoid -Wflex-array-member-not-at-end warnings -Wflex-array-member-not-at-end was introduced in GCC-14, and we are getting ready to enable it, globally. Use regular arrays instead of flexible-array members (they're not really needed in this case) in a couple of unions, and fix the following warnings: 1 drivers/net/ethernet/spacemit/k1_emac.c:122:42: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end] 1 drivers/net/ethernet/spacemit/k1_emac.c:122:32: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end] 1 drivers/net/ethernet/spacemit/k1_emac.c:121:42: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end] 1 drivers/net/ethernet/spacemit/k1_emac.c:121:32: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end] Signed-off-by: Gustavo A. R. Silva Reviewed-by: Simon Horman Acked-by: Vivian Wang Link: https://patch.msgid.link/aPd0YjO-oP60Lgvj@kspp Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/spacemit/k1_emac.h | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/spacemit/k1_emac.h b/drivers/net/ethernet/spacemit/k1_emac.h index 5a09e946a276..577efe66573e 100644 --- a/drivers/net/ethernet/spacemit/k1_emac.h +++ b/drivers/net/ethernet/spacemit/k1_emac.h @@ -363,7 +363,7 @@ struct emac_desc { /* Keep stats in this order, index used for accessing hardware */ union emac_hw_tx_stats { - struct { + struct individual_tx_stats { u64 tx_ok_pkts; u64 tx_total_pkts; u64 tx_ok_bytes; @@ -378,11 +378,11 @@ union emac_hw_tx_stats { u64 tx_pause_pkts; } stats; - DECLARE_FLEX_ARRAY(u64, array); + u64 array[sizeof(struct individual_tx_stats) / sizeof(u64)]; }; union emac_hw_rx_stats { - struct { + struct individual_rx_stats { u64 rx_ok_pkts; u64 rx_total_pkts; u64 rx_crc_err_pkts; @@ -410,7 +410,7 @@ union emac_hw_rx_stats { u64 rx_truncate_fifo_full_pkts; } stats; - DECLARE_FLEX_ARRAY(u64, array); + u64 array[sizeof(struct individual_rx_stats) / sizeof(u64)]; }; #endif /* _K1_EMAC_H_ */ From 114573962a68a527835f2f1433a89bc2f9feac1b Mon Sep 17 00:00:00 2001 From: Yue Haibing Date: Tue, 21 Oct 2025 19:46:26 +0800 Subject: [PATCH 0191/1381] net/sched: Remove unused inline helper qdisc_from_priv() Since commit fb38306ceb9e ("net/sched: Retire ATM qdisc"), this is not used and can be removed. Signed-off-by: Yue Haibing Link: https://patch.msgid.link/20251021114626.3148894-1-yuehaibing@huawei.com Signed-off-by: Jakub Kicinski --- include/net/pkt_sched.h | 5 ----- 1 file changed, 5 deletions(-) diff --git a/include/net/pkt_sched.h b/include/net/pkt_sched.h index 8a75c73fc555..c660ac871083 100644 --- a/include/net/pkt_sched.h +++ b/include/net/pkt_sched.h @@ -25,11 +25,6 @@ struct qdisc_walker { const struct Qdisc * : (const void *)&q->privdata, \ struct Qdisc * : (void *)&q->privdata) -static inline struct Qdisc *qdisc_from_priv(void *priv) -{ - return container_of(priv, struct Qdisc, privdata); -} - /* Timer resolution MUST BE < 10% of min_schedulable_packet_size/bandwidth From d550d63d0082268a31e93a10c64cbc2476b98b24 Mon Sep 17 00:00:00 2001 From: Pei Xiao Date: Tue, 21 Oct 2025 17:42:27 +0800 Subject: [PATCH 0192/1381] eth: fbnic: fix integer overflow warning in TLV_MAX_DATA definition The TLV_MAX_DATA macro calculates (PAGE_SIZE - 512) which can exceed the maximum value of a 16-bit unsigned integer on architectures with large page sizes, causing compiler warnings: drivers/net/ethernet/meta/fbnic/fbnic_tlv.h:83:24: warning: conversion from 'long unsigned int' to 'short unsigned int' changes value from '261632' to '65024' [-Woverflow] Fix this by explicitly masking the result to 16 bits using bitwise AND with 0xFFFF, ensuring the value fits within the expected data type while maintaining the intended behavior for normal page sizes. This preserves the existing functionality while eliminating the compiler warning and potential undefined behavior from integer truncation. Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-kbuild-all/202510190832.3SQkTCHe-lkp@intel.com/ Signed-off-by: Pei Xiao Link: https://patch.msgid.link/182b9d0235d044d69d7a57c1296cc6f46e395beb.1761039651.git.xiaopei01@kylinos.cn Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/meta/fbnic/fbnic_tlv.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_tlv.h b/drivers/net/ethernet/meta/fbnic/fbnic_tlv.h index c34bf87eeec9..3508b46ebdd0 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_tlv.h +++ b/drivers/net/ethernet/meta/fbnic/fbnic_tlv.h @@ -80,7 +80,7 @@ struct fbnic_tlv_index { enum fbnic_tlv_type type; }; -#define TLV_MAX_DATA (PAGE_SIZE - 512) +#define TLV_MAX_DATA ((PAGE_SIZE - 512) & 0xFFFF) #define FBNIC_TLV_ATTR_ID_UNKNOWN USHRT_MAX #define FBNIC_TLV_ATTR_STRING(id, len) { id, len, FBNIC_TLV_STRING } #define FBNIC_TLV_ATTR_FLAG(id) { id, 0, FBNIC_TLV_FLAG } From e0665df8c501829f09e6a2a7c06b9860209d3a70 Mon Sep 17 00:00:00 2001 From: Markus Elfring Date: Mon, 20 Oct 2025 15:46:11 +0200 Subject: [PATCH 0193/1381] net: ti: icssg-prueth: Omit a variable reassignment in prueth_netdev_init() An error code was assigned to a variable and checked accordingly. This value was passed to a dev_err_probe() call in an if branch. This function is documented in the way that the same value is returned. Thus delete two redundant variable reassignments. The source code was transformed by using the Coccinelle software. Signed-off-by: Markus Elfring Link: https://patch.msgid.link/71f7daa3-d4f4-4753-aae8-67040fc8297d@web.de Signed-off-by: Paolo Abeni --- drivers/net/ethernet/ti/icssg/icssg_prueth.c | 3 +-- drivers/net/ethernet/ti/icssg/icssg_prueth_sr1.c | 3 +-- 2 files changed, 2 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/ti/icssg/icssg_prueth.c b/drivers/net/ethernet/ti/icssg/icssg_prueth.c index 1c1f4394ff1f..57a7d1ceab08 100644 --- a/drivers/net/ethernet/ti/icssg/icssg_prueth.c +++ b/drivers/net/ethernet/ti/icssg/icssg_prueth.c @@ -1250,8 +1250,7 @@ static int prueth_netdev_init(struct prueth *prueth, } else if (of_phy_is_fixed_link(eth_node)) { ret = of_phy_register_fixed_link(eth_node); if (ret) { - ret = dev_err_probe(prueth->dev, ret, - "failed to register fixed-link phy\n"); + dev_err_probe(prueth->dev, ret, "failed to register fixed-link phy\n"); goto free; } diff --git a/drivers/net/ethernet/ti/icssg/icssg_prueth_sr1.c b/drivers/net/ethernet/ti/icssg/icssg_prueth_sr1.c index 2a8c8847a6bd..7bb4f0d850cc 100644 --- a/drivers/net/ethernet/ti/icssg/icssg_prueth_sr1.c +++ b/drivers/net/ethernet/ti/icssg/icssg_prueth_sr1.c @@ -818,8 +818,7 @@ static int prueth_netdev_init(struct prueth *prueth, } else if (of_phy_is_fixed_link(eth_node)) { ret = of_phy_register_fixed_link(eth_node); if (ret) { - ret = dev_err_probe(prueth->dev, ret, - "failed to register fixed-link phy\n"); + dev_err_probe(prueth->dev, ret, "failed to register fixed-link phy\n"); goto free; } From 61b7ade9ba8c3b16867e25411b5f7cf1abe35879 Mon Sep 17 00:00:00 2001 From: Horatiu Vultur Date: Tue, 21 Oct 2025 09:07:26 +0200 Subject: [PATCH 0194/1381] net: phy: micrel: Add support for non PTP SKUs for lan8814 The lan8814 has 4 different SKUs and for 2 of these SKUs the PTP is disabled. All these SKUs have the same value in the register 2 and 3. Meaning that we can't differentiate them based on device id, therefore check the SKU register and based on this allow or not to create a PTP device. Signed-off-by: Horatiu Vultur Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20251021070726.3690685-1-horatiu.vultur@microchip.com Signed-off-by: Paolo Abeni --- drivers/net/phy/micrel.c | 38 ++++++++++++++++++++++++++++++++++++++ 1 file changed, 38 insertions(+) diff --git a/drivers/net/phy/micrel.c b/drivers/net/phy/micrel.c index 5f2c7e5c314f..a47e55c22815 100644 --- a/drivers/net/phy/micrel.c +++ b/drivers/net/phy/micrel.c @@ -101,6 +101,8 @@ #define LAN8814_CABLE_DIAG_VCT_DATA_MASK GENMASK(7, 0) #define LAN8814_PAIR_BIT_SHIFT 12 +#define LAN8814_SKUS 0xB + #define LAN8814_WIRE_PAIR_MASK 0xF /* Lan8814 general Interrupt control/status reg in GPHY specific block. */ @@ -367,6 +369,9 @@ #define LAN8842_REV_8832 0x8832 +#define LAN8814_REV_LAN8814 0x8814 +#define LAN8814_REV_LAN8818 0x8818 + struct kszphy_hw_stat { const char *string; u8 reg; @@ -449,6 +454,7 @@ struct kszphy_priv { bool rmii_ref_clk_sel; bool rmii_ref_clk_sel_val; bool clk_enable; + bool is_ptp_available; u64 stats[ARRAY_SIZE(kszphy_hw_stats)]; struct kszphy_phy_stats phy_stats; }; @@ -4126,6 +4132,17 @@ static int lan8804_config_intr(struct phy_device *phydev) return 0; } +/* Check if the PHY has 1588 support. There are multiple skus of the PHY and + * some of them support PTP while others don't support it. This function will + * return true is the sku supports it, otherwise will return false. + */ +static bool lan8814_has_ptp(struct phy_device *phydev) +{ + struct kszphy_priv *priv = phydev->priv; + + return priv->is_ptp_available; +} + static irqreturn_t lan8814_handle_interrupt(struct phy_device *phydev) { int ret = IRQ_NONE; @@ -4142,6 +4159,9 @@ static irqreturn_t lan8814_handle_interrupt(struct phy_device *phydev) ret = IRQ_HANDLED; } + if (!lan8814_has_ptp(phydev)) + return ret; + while (true) { irq_status = lanphy_read_page_reg(phydev, LAN8814_PAGE_PORT_REGS, PTP_TSU_INT_STS); @@ -4203,6 +4223,9 @@ static void lan8814_ptp_init(struct phy_device *phydev) !IS_ENABLED(CONFIG_NETWORK_PHY_TIMESTAMPING)) return; + if (!lan8814_has_ptp(phydev)) + return; + lanphy_write_page_reg(phydev, LAN8814_PAGE_PORT_REGS, TSU_HARD_RESET, TSU_HARD_RESET_); @@ -4332,6 +4355,9 @@ static int __lan8814_ptp_probe_once(struct phy_device *phydev, char *pin_name, static int lan8814_ptp_probe_once(struct phy_device *phydev) { + if (!lan8814_has_ptp(phydev)) + return 0; + return __lan8814_ptp_probe_once(phydev, "lan8814_ptp_pin", LAN8814_PTP_GPIO_NUM); } @@ -4446,6 +4472,18 @@ static int lan8814_probe(struct phy_device *phydev) devm_phy_package_join(&phydev->mdio.dev, phydev, addr, sizeof(struct lan8814_shared_priv)); + /* There are lan8814 SKUs that don't support PTP. Make sure that for + * those skus no PTP device is created. Here we check if the SKU + * supports PTP. + */ + err = lanphy_read_page_reg(phydev, LAN8814_PAGE_COMMON_REGS, + LAN8814_SKUS); + if (err < 0) + return err; + + priv->is_ptp_available = err == LAN8814_REV_LAN8814 || + err == LAN8814_REV_LAN8818; + if (phy_package_init_once(phydev)) { err = lan8814_release_coma_mode(phydev); if (err) From 47d0cd6bccb4604192633cc8d29511e85d811fc0 Mon Sep 17 00:00:00 2001 From: Baochen Qiang Date: Fri, 17 Oct 2025 09:48:59 +0800 Subject: [PATCH 0195/1381] wifi: ath11k: fix VHT MCS assignment While associating, firmware needs to know peer's receive capability to calculate its own VHT transmit MCS, currently host sends this information to firmware via mcs->rx_mcs_set field, this is wrong as firmware actually takes it from mcs->tx_mcs_set field. Till now there is no failure seen due to this, most likely because almost all peers are advertising the same capability for both transmit and receive. Swap the assignment to fix it. Besides, rate control mask is meant to limit our own transmit MCS, hence need to go via mcs->tx_mcs_set field. With the aforementioned swapping done, change is needed as well to apply it to the peer's receive capability rather than transmit capability. Tested-on: WCN6855 hw2.1 PCI WLAN.HSP.1.1-03125-QCAHSPSWPL_V1_V2_SILICONZ_LITE-3.6510.41 Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.4.1-00199-QCAHKSWPL_SILICONZ-1 Fixes: d5c65159f289 ("ath11k: driver for Qualcomm IEEE 802.11ax devices") Signed-off-by: Baochen Qiang Reviewed-by: Vasanthakumar Thiagarajan Link: https://patch.msgid.link/20251017-ath11k-mcs-assignment-v1-1-da40825c1783@oss.qualcomm.com Signed-off-by: Jeff Johnson --- drivers/net/wireless/ath/ath11k/mac.c | 4 ++-- drivers/net/wireless/ath/ath11k/wmi.c | 13 ++++++++----- drivers/net/wireless/ath/ath11k/wmi.h | 2 ++ 3 files changed, 12 insertions(+), 7 deletions(-) diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c index 106e2530b64e..367ca0c90167 100644 --- a/drivers/net/wireless/ath/ath11k/mac.c +++ b/drivers/net/wireless/ath/ath11k/mac.c @@ -2235,9 +2235,9 @@ static void ath11k_peer_assoc_h_vht(struct ath11k *ar, arg->peer_nss = min(sta->deflink.rx_nss, max_nss); arg->rx_max_rate = __le16_to_cpu(vht_cap->vht_mcs.rx_highest); arg->rx_mcs_set = __le16_to_cpu(vht_cap->vht_mcs.rx_mcs_map); + arg->rx_mcs_set = ath11k_peer_assoc_h_vht_limit(arg->rx_mcs_set, vht_mcs_mask); arg->tx_max_rate = __le16_to_cpu(vht_cap->vht_mcs.tx_highest); - arg->tx_mcs_set = ath11k_peer_assoc_h_vht_limit( - __le16_to_cpu(vht_cap->vht_mcs.tx_mcs_map), vht_mcs_mask); + arg->tx_mcs_set = __le16_to_cpu(vht_cap->vht_mcs.tx_mcs_map); /* In IPQ8074 platform, VHT mcs rate 10 and 11 is enabled by default. * VHT mcs rate 10 and 11 is not supported in 11ac standard. diff --git a/drivers/net/wireless/ath/ath11k/wmi.c b/drivers/net/wireless/ath/ath11k/wmi.c index 0491e3fd6b5e..942dfeb8b1af 100644 --- a/drivers/net/wireless/ath/ath11k/wmi.c +++ b/drivers/net/wireless/ath/ath11k/wmi.c @@ -1,7 +1,7 @@ // SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved. - * Copyright (c) 2021-2025 Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. */ #include #include @@ -2061,10 +2061,13 @@ int ath11k_wmi_send_peer_assoc_cmd(struct ath11k *ar, cmd->peer_bw_rxnss_override |= param->peer_bw_rxnss_override; if (param->vht_capable) { - mcs->rx_max_rate = param->rx_max_rate; - mcs->rx_mcs_set = param->rx_mcs_set; - mcs->tx_max_rate = param->tx_max_rate; - mcs->tx_mcs_set = param->tx_mcs_set; + /* firmware interprets mcs->tx_mcs_set field as peer's + * RX capability + */ + mcs->tx_max_rate = param->rx_max_rate; + mcs->tx_mcs_set = param->rx_mcs_set; + mcs->rx_max_rate = param->tx_max_rate; + mcs->rx_mcs_set = param->tx_mcs_set; } /* HE Rates */ diff --git a/drivers/net/wireless/ath/ath11k/wmi.h b/drivers/net/wireless/ath/ath11k/wmi.h index 3c4885a12855..0f0de24a3840 100644 --- a/drivers/net/wireless/ath/ath11k/wmi.h +++ b/drivers/net/wireless/ath/ath11k/wmi.h @@ -4119,8 +4119,10 @@ struct wmi_rate_set { struct wmi_vht_rate_set { u32 tlv_header; u32 rx_max_rate; + /* MCS at which the peer can transmit */ u32 rx_mcs_set; u32 tx_max_rate; + /* MCS at which the peer can receive */ u32 tx_mcs_set; u32 tx_max_mcs_nss; } __packed; From 4a013ca2d490c73c40588d62712ffaa432046a04 Mon Sep 17 00:00:00 2001 From: Baochen Qiang Date: Fri, 17 Oct 2025 09:49:00 +0800 Subject: [PATCH 0196/1381] wifi: ath11k: fix peer HE MCS assignment In ath11k_wmi_send_peer_assoc_cmd(), peer's transmit MCS is sent to firmware as receive MCS while peer's receive MCS sent as transmit MCS, which goes against firmwire's definition. While connecting to a misbehaved AP that advertises 0xffff (meaning not supported) for 160 MHz transmit MCS map, firmware crashes due to 0xffff is assigned to he_mcs->rx_mcs_set field. Ext Tag: HE Capabilities [...] Supported HE-MCS and NSS Set [...] Rx and Tx MCS Maps 160 MHz [...] Tx HE-MCS Map 160 MHz: 0xffff Swap the assignment to fix this issue. As the HE rate control mask is meant to limit our own transmit MCS, it needs to go via he_mcs->rx_mcs_set field. With the aforementioned swapping done, change is needed as well to apply it to the peer's receive MCS. Tested-on: WCN6855 hw2.1 PCI WLAN.HSP.1.1-03125-QCAHSPSWPL_V1_V2_SILICONZ_LITE-3.6510.41 Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.4.1-00199-QCAHKSWPL_SILICONZ-1 Fixes: 61fe43e7216d ("ath11k: add support for setting fixed HE rate/gi/ltf") Signed-off-by: Baochen Qiang Reviewed-by: Vasanthakumar Thiagarajan Link: https://patch.msgid.link/20251017-ath11k-mcs-assignment-v1-2-da40825c1783@oss.qualcomm.com Signed-off-by: Jeff Johnson --- drivers/net/wireless/ath/ath11k/mac.c | 4 ++-- drivers/net/wireless/ath/ath11k/wmi.c | 7 +++++-- 2 files changed, 7 insertions(+), 4 deletions(-) diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c index 367ca0c90167..ebde4208791c 100644 --- a/drivers/net/wireless/ath/ath11k/mac.c +++ b/drivers/net/wireless/ath/ath11k/mac.c @@ -2522,10 +2522,10 @@ static void ath11k_peer_assoc_h_he(struct ath11k *ar, he_tx_mcs = v; } v = le16_to_cpu(he_cap->he_mcs_nss_supp.rx_mcs_160); + v = ath11k_peer_assoc_h_he_limit(v, he_mcs_mask); arg->peer_he_rx_mcs_set[WMI_HECAP_TXRX_MCS_NSS_IDX_160] = v; v = le16_to_cpu(he_cap->he_mcs_nss_supp.tx_mcs_160); - v = ath11k_peer_assoc_h_he_limit(v, he_mcs_mask); arg->peer_he_tx_mcs_set[WMI_HECAP_TXRX_MCS_NSS_IDX_160] = v; arg->peer_he_mcs_count++; @@ -2535,10 +2535,10 @@ static void ath11k_peer_assoc_h_he(struct ath11k *ar, default: v = le16_to_cpu(he_cap->he_mcs_nss_supp.rx_mcs_80); + v = ath11k_peer_assoc_h_he_limit(v, he_mcs_mask); arg->peer_he_rx_mcs_set[WMI_HECAP_TXRX_MCS_NSS_IDX_80] = v; v = le16_to_cpu(he_cap->he_mcs_nss_supp.tx_mcs_80); - v = ath11k_peer_assoc_h_he_limit(v, he_mcs_mask); arg->peer_he_tx_mcs_set[WMI_HECAP_TXRX_MCS_NSS_IDX_80] = v; arg->peer_he_mcs_count++; diff --git a/drivers/net/wireless/ath/ath11k/wmi.c b/drivers/net/wireless/ath/ath11k/wmi.c index 942dfeb8b1af..edff6fb61344 100644 --- a/drivers/net/wireless/ath/ath11k/wmi.c +++ b/drivers/net/wireless/ath/ath11k/wmi.c @@ -2091,8 +2091,11 @@ int ath11k_wmi_send_peer_assoc_cmd(struct ath11k *ar, FIELD_PREP(WMI_TLV_LEN, sizeof(*he_mcs) - TLV_HDR_SIZE); - he_mcs->rx_mcs_set = param->peer_he_tx_mcs_set[i]; - he_mcs->tx_mcs_set = param->peer_he_rx_mcs_set[i]; + /* firmware interprets mcs->rx_mcs_set field as peer's + * RX capability + */ + he_mcs->rx_mcs_set = param->peer_he_rx_mcs_set[i]; + he_mcs->tx_mcs_set = param->peer_he_tx_mcs_set[i]; ptr += sizeof(*he_mcs); } From 66887282233d281cd9109dabfdad5d86b709acc0 Mon Sep 17 00:00:00 2001 From: Aditya Kumar Singh Date: Fri, 17 Oct 2025 09:37:57 +0530 Subject: [PATCH 0197/1381] wifi: ath11k: relocate some Tx power related functions in mac.c A forthcoming change necessitates that these functions be defined prior to their usage. Therefore, relocate them now as a preparatory step for the upcoming modifications. Compile tested only. Signed-off-by: Aditya Kumar Singh Reviewed-by: Vasanthakumar Thiagarajan Reviewed-by: Baochen Qiang Link: https://patch.msgid.link/20251017-add_tx_power_insertion_support-v1-1-f08feacfca93@oss.qualcomm.com Signed-off-by: Jeff Johnson --- drivers/net/wireless/ath/ath11k/mac.c | 270 +++++++++++++------------- 1 file changed, 135 insertions(+), 135 deletions(-) diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c index ebde4208791c..99dea865e31e 100644 --- a/drivers/net/wireless/ath/ath11k/mac.c +++ b/drivers/net/wireless/ath/ath11k/mac.c @@ -4028,6 +4028,141 @@ static int ath11k_start_scan(struct ath11k *ar, return 0; } +static void ath11k_mac_fw_stats_reset(struct ath11k *ar) +{ + spin_lock_bh(&ar->data_lock); + ath11k_fw_stats_pdevs_free(&ar->fw_stats.pdevs); + ath11k_fw_stats_vdevs_free(&ar->fw_stats.vdevs); + ar->fw_stats.num_vdev_recvd = 0; + ar->fw_stats.num_bcn_recvd = 0; + spin_unlock_bh(&ar->data_lock); +} + +int ath11k_mac_fw_stats_request(struct ath11k *ar, + struct stats_request_params *req_param) +{ + struct ath11k_base *ab = ar->ab; + unsigned long time_left; + int ret; + + lockdep_assert_held(&ar->conf_mutex); + + ath11k_mac_fw_stats_reset(ar); + + reinit_completion(&ar->fw_stats_complete); + reinit_completion(&ar->fw_stats_done); + + ret = ath11k_wmi_send_stats_request_cmd(ar, req_param); + + if (ret) { + ath11k_warn(ab, "could not request fw stats (%d)\n", + ret); + return ret; + } + + time_left = wait_for_completion_timeout(&ar->fw_stats_complete, 1 * HZ); + if (!time_left) + return -ETIMEDOUT; + + /* FW stats can get split when exceeding the stats data buffer limit. + * In that case, since there is no end marking for the back-to-back + * received 'update stats' event, we keep a 3 seconds timeout in case, + * fw_stats_done is not marked yet + */ + time_left = wait_for_completion_timeout(&ar->fw_stats_done, 3 * HZ); + if (!time_left) + return -ETIMEDOUT; + + return 0; +} + +static int ath11k_mac_get_fw_stats(struct ath11k *ar, u32 pdev_id, + u32 vdev_id, u32 stats_id) +{ + struct ath11k_base *ab = ar->ab; + struct stats_request_params req_param; + int ret; + + lockdep_assert_held(&ar->conf_mutex); + + if (ar->state != ATH11K_STATE_ON) + return -ENETDOWN; + + req_param.pdev_id = pdev_id; + req_param.vdev_id = vdev_id; + req_param.stats_id = stats_id; + + ret = ath11k_mac_fw_stats_request(ar, &req_param); + if (ret) + ath11k_warn(ab, "failed to request fw stats: %d\n", ret); + + ath11k_dbg(ab, ATH11K_DBG_WMI, + "debug get fw stat pdev id %d vdev id %d stats id 0x%x\n", + pdev_id, vdev_id, stats_id); + + return ret; +} + +static int ath11k_mac_op_get_txpower(struct ieee80211_hw *hw, + struct ieee80211_vif *vif, + unsigned int link_id, + int *dbm) +{ + struct ath11k *ar = hw->priv; + struct ath11k_base *ab = ar->ab; + struct ath11k_fw_stats_pdev *pdev; + int ret; + + /* Final Tx power is minimum of Target Power, CTL power, Regulatory + * Power, PSD EIRP Power. We just know the Regulatory power from the + * regulatory rules obtained. FW knows all these power and sets the min + * of these. Hence, we request the FW pdev stats in which FW reports + * the minimum of all vdev's channel Tx power. + */ + mutex_lock(&ar->conf_mutex); + + /* Firmware doesn't provide Tx power during CAC hence no need to fetch + * the stats. + */ + if (test_bit(ATH11K_CAC_RUNNING, &ar->dev_flags)) { + mutex_unlock(&ar->conf_mutex); + return -EAGAIN; + } + + ret = ath11k_mac_get_fw_stats(ar, ar->pdev->pdev_id, 0, + WMI_REQUEST_PDEV_STAT); + if (ret) { + ath11k_warn(ab, "failed to request fw pdev stats: %d\n", ret); + goto err_fallback; + } + + spin_lock_bh(&ar->data_lock); + pdev = list_first_entry_or_null(&ar->fw_stats.pdevs, + struct ath11k_fw_stats_pdev, list); + if (!pdev) { + spin_unlock_bh(&ar->data_lock); + goto err_fallback; + } + + /* tx power is set as 2 units per dBm in FW. */ + *dbm = pdev->chan_tx_power / 2; + + spin_unlock_bh(&ar->data_lock); + mutex_unlock(&ar->conf_mutex); + + ath11k_dbg(ar->ab, ATH11K_DBG_MAC, "txpower from firmware %d, reported %d dBm\n", + pdev->chan_tx_power, *dbm); + return 0; + +err_fallback: + mutex_unlock(&ar->conf_mutex); + /* We didn't get txpower from FW. Hence, relying on vif->bss_conf.txpower */ + *dbm = vif->bss_conf.txpower; + ath11k_dbg(ar->ab, ATH11K_DBG_MAC, "txpower from firmware NaN, reported %d dBm\n", + *dbm); + return 0; +} + static int ath11k_mac_op_hw_scan(struct ieee80211_hw *hw, struct ieee80211_vif *vif, struct ieee80211_scan_request *hw_req) @@ -9079,81 +9214,6 @@ static void ath11k_mac_put_chain_rssi(struct station_info *sinfo, } } -static void ath11k_mac_fw_stats_reset(struct ath11k *ar) -{ - spin_lock_bh(&ar->data_lock); - ath11k_fw_stats_pdevs_free(&ar->fw_stats.pdevs); - ath11k_fw_stats_vdevs_free(&ar->fw_stats.vdevs); - ar->fw_stats.num_vdev_recvd = 0; - ar->fw_stats.num_bcn_recvd = 0; - spin_unlock_bh(&ar->data_lock); -} - -int ath11k_mac_fw_stats_request(struct ath11k *ar, - struct stats_request_params *req_param) -{ - struct ath11k_base *ab = ar->ab; - unsigned long time_left; - int ret; - - lockdep_assert_held(&ar->conf_mutex); - - ath11k_mac_fw_stats_reset(ar); - - reinit_completion(&ar->fw_stats_complete); - reinit_completion(&ar->fw_stats_done); - - ret = ath11k_wmi_send_stats_request_cmd(ar, req_param); - - if (ret) { - ath11k_warn(ab, "could not request fw stats (%d)\n", - ret); - return ret; - } - - time_left = wait_for_completion_timeout(&ar->fw_stats_complete, 1 * HZ); - if (!time_left) - return -ETIMEDOUT; - - /* FW stats can get split when exceeding the stats data buffer limit. - * In that case, since there is no end marking for the back-to-back - * received 'update stats' event, we keep a 3 seconds timeout in case, - * fw_stats_done is not marked yet - */ - time_left = wait_for_completion_timeout(&ar->fw_stats_done, 3 * HZ); - if (!time_left) - return -ETIMEDOUT; - - return 0; -} - -static int ath11k_mac_get_fw_stats(struct ath11k *ar, u32 pdev_id, - u32 vdev_id, u32 stats_id) -{ - struct ath11k_base *ab = ar->ab; - struct stats_request_params req_param; - int ret; - - lockdep_assert_held(&ar->conf_mutex); - - if (ar->state != ATH11K_STATE_ON) - return -ENETDOWN; - - req_param.pdev_id = pdev_id; - req_param.vdev_id = vdev_id; - req_param.stats_id = stats_id; - - ret = ath11k_mac_fw_stats_request(ar, &req_param); - if (ret) - ath11k_warn(ab, "failed to request fw stats: %d\n", ret); - - ath11k_dbg(ab, ATH11K_DBG_WMI, - "debug get fw stat pdev id %d vdev id %d stats id 0x%x\n", - pdev_id, vdev_id, stats_id); - - return ret; -} - static void ath11k_mac_op_sta_statistics(struct ieee80211_hw *hw, struct ieee80211_vif *vif, struct ieee80211_sta *sta, @@ -9539,66 +9599,6 @@ static int ath11k_mac_op_remain_on_channel(struct ieee80211_hw *hw, return ret; } -static int ath11k_mac_op_get_txpower(struct ieee80211_hw *hw, - struct ieee80211_vif *vif, - unsigned int link_id, - int *dbm) -{ - struct ath11k *ar = hw->priv; - struct ath11k_base *ab = ar->ab; - struct ath11k_fw_stats_pdev *pdev; - int ret; - - /* Final Tx power is minimum of Target Power, CTL power, Regulatory - * Power, PSD EIRP Power. We just know the Regulatory power from the - * regulatory rules obtained. FW knows all these power and sets the min - * of these. Hence, we request the FW pdev stats in which FW reports - * the minimum of all vdev's channel Tx power. - */ - mutex_lock(&ar->conf_mutex); - - /* Firmware doesn't provide Tx power during CAC hence no need to fetch - * the stats. - */ - if (test_bit(ATH11K_CAC_RUNNING, &ar->dev_flags)) { - mutex_unlock(&ar->conf_mutex); - return -EAGAIN; - } - - ret = ath11k_mac_get_fw_stats(ar, ar->pdev->pdev_id, 0, - WMI_REQUEST_PDEV_STAT); - if (ret) { - ath11k_warn(ab, "failed to request fw pdev stats: %d\n", ret); - goto err_fallback; - } - - spin_lock_bh(&ar->data_lock); - pdev = list_first_entry_or_null(&ar->fw_stats.pdevs, - struct ath11k_fw_stats_pdev, list); - if (!pdev) { - spin_unlock_bh(&ar->data_lock); - goto err_fallback; - } - - /* tx power is set as 2 units per dBm in FW. */ - *dbm = pdev->chan_tx_power / 2; - - spin_unlock_bh(&ar->data_lock); - mutex_unlock(&ar->conf_mutex); - - ath11k_dbg(ar->ab, ATH11K_DBG_MAC, "txpower from firmware %d, reported %d dBm\n", - pdev->chan_tx_power, *dbm); - return 0; - -err_fallback: - mutex_unlock(&ar->conf_mutex); - /* We didn't get txpower from FW. Hence, relying on vif->bss_conf.txpower */ - *dbm = vif->bss_conf.txpower; - ath11k_dbg(ar->ab, ATH11K_DBG_MAC, "txpower from firmware NaN, reported %d dBm\n", - *dbm); - return 0; -} - static int ath11k_mac_station_add(struct ath11k *ar, struct ieee80211_vif *vif, struct ieee80211_sta *sta) From c243d5e44f6ecbb29bf55b82e6dd92bca4fde0b1 Mon Sep 17 00:00:00 2001 From: Aditya Kumar Singh Date: Fri, 17 Oct 2025 09:37:58 +0530 Subject: [PATCH 0198/1381] wifi: ath11k: wrap ath11k_mac_op_get_txpower() with lock-aware internal helper Refactor ath11k_mac_op_get_txpower() by introducing a new internal function ath11k_mac_handle_get_txpower(), which assumes the caller holds the appropriate lock. This prepares the codebase for future change where the internal function may be invoked directly with the lock already acquired, improving modularity and lock handling consistency. No functional change intended. Tested-on: QCN9074 hw1.0 PCI WLAN.HK.2.9.0.1-02146-QCAHKSWPL_SILICONZ-1 Signed-off-by: Aditya Kumar Singh Reviewed-by: Vasanthakumar Thiagarajan Reviewed-by: Baochen Qiang Link: https://patch.msgid.link/20251017-add_tx_power_insertion_support-v1-2-f08feacfca93@oss.qualcomm.com Signed-off-by: Jeff Johnson --- drivers/net/wireless/ath/ath11k/mac.c | 31 +++++++++++++++++---------- 1 file changed, 20 insertions(+), 11 deletions(-) diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c index 99dea865e31e..319b4cfeb368 100644 --- a/drivers/net/wireless/ath/ath11k/mac.c +++ b/drivers/net/wireless/ath/ath11k/mac.c @@ -4103,12 +4103,10 @@ static int ath11k_mac_get_fw_stats(struct ath11k *ar, u32 pdev_id, return ret; } -static int ath11k_mac_op_get_txpower(struct ieee80211_hw *hw, - struct ieee80211_vif *vif, - unsigned int link_id, - int *dbm) +static int ath11k_mac_handle_get_txpower(struct ath11k *ar, + struct ieee80211_vif *vif, + int *dbm) { - struct ath11k *ar = hw->priv; struct ath11k_base *ab = ar->ab; struct ath11k_fw_stats_pdev *pdev; int ret; @@ -4119,15 +4117,13 @@ static int ath11k_mac_op_get_txpower(struct ieee80211_hw *hw, * of these. Hence, we request the FW pdev stats in which FW reports * the minimum of all vdev's channel Tx power. */ - mutex_lock(&ar->conf_mutex); + lockdep_assert_held(&ar->conf_mutex); /* Firmware doesn't provide Tx power during CAC hence no need to fetch * the stats. */ - if (test_bit(ATH11K_CAC_RUNNING, &ar->dev_flags)) { - mutex_unlock(&ar->conf_mutex); + if (test_bit(ATH11K_CAC_RUNNING, &ar->dev_flags)) return -EAGAIN; - } ret = ath11k_mac_get_fw_stats(ar, ar->pdev->pdev_id, 0, WMI_REQUEST_PDEV_STAT); @@ -4148,14 +4144,12 @@ static int ath11k_mac_op_get_txpower(struct ieee80211_hw *hw, *dbm = pdev->chan_tx_power / 2; spin_unlock_bh(&ar->data_lock); - mutex_unlock(&ar->conf_mutex); ath11k_dbg(ar->ab, ATH11K_DBG_MAC, "txpower from firmware %d, reported %d dBm\n", pdev->chan_tx_power, *dbm); return 0; err_fallback: - mutex_unlock(&ar->conf_mutex); /* We didn't get txpower from FW. Hence, relying on vif->bss_conf.txpower */ *dbm = vif->bss_conf.txpower; ath11k_dbg(ar->ab, ATH11K_DBG_MAC, "txpower from firmware NaN, reported %d dBm\n", @@ -4163,6 +4157,21 @@ static int ath11k_mac_op_get_txpower(struct ieee80211_hw *hw, return 0; } +static int ath11k_mac_op_get_txpower(struct ieee80211_hw *hw, + struct ieee80211_vif *vif, + unsigned int link_id, + int *dbm) +{ + struct ath11k *ar = hw->priv; + int ret; + + mutex_lock(&ar->conf_mutex); + ret = ath11k_mac_handle_get_txpower(ar, vif, dbm); + mutex_unlock(&ar->conf_mutex); + + return ret; +} + static int ath11k_mac_op_hw_scan(struct ieee80211_hw *hw, struct ieee80211_vif *vif, struct ieee80211_scan_request *hw_req) From 722015690f52d046ae609e1b90cd3f018644d93d Mon Sep 17 00:00:00 2001 From: Aditya Kumar Singh Date: Fri, 17 Oct 2025 09:37:59 +0530 Subject: [PATCH 0199/1381] wifi: ath11k: add support for Tx Power insertion in RRM action frame For certain action frames like the TPC Report IE in the spectrum management TPC Report action frame, and in the Radio Measurement Link Measurement Report action frame there is a requirement to fill in the current and max Tx power of the device in the packet. Add support to populate these fields in the relevant packets. In software-encrypted cases such as PMF, skip insertion since the packets are already encrypted and cannot be modified. Tested-on: QCN9074 hw1.0 PCI WLAN.HK.2.9.0.1-02146-QCAHKSWPL_SILICONZ-1 Signed-off-by: Aditya Kumar Singh Reviewed-by: Vasanthakumar Thiagarajan Reviewed-by: Baochen Qiang Link: https://patch.msgid.link/20251017-add_tx_power_insertion_support-v1-3-f08feacfca93@oss.qualcomm.com Signed-off-by: Jeff Johnson --- drivers/net/wireless/ath/ath11k/mac.c | 166 ++++++++++++++++++++++++++ 1 file changed, 166 insertions(+) diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c index 319b4cfeb368..d15e3b67e625 100644 --- a/drivers/net/wireless/ath/ath11k/mac.c +++ b/drivers/net/wireless/ath/ath11k/mac.c @@ -6251,6 +6251,159 @@ static void ath11k_mgmt_over_wmi_tx_purge(struct ath11k *ar) ath11k_mgmt_over_wmi_tx_drop(ar, skb); } +static int ath11k_mac_mgmt_action_frame_fill_elem_data(struct ath11k_vif *arvif, + struct sk_buff *skb) +{ + struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)skb->data; + u8 category, *buf, iv_len, action_code, dialog_token; + int cur_tx_power, max_tx_power; + struct ath11k *ar = arvif->ar; + struct cfg80211_chan_def def; + struct ath11k_skb_cb *skb_cb; + struct ieee80211_mgmt *mgmt; + unsigned int remaining_len; + bool has_protected; + + lockdep_assert_held(&ar->conf_mutex); + + /* make sure category field is present */ + if (skb->len < IEEE80211_MIN_ACTION_SIZE) + return -EINVAL; + + remaining_len = skb->len - IEEE80211_MIN_ACTION_SIZE; + has_protected = ieee80211_has_protected(hdr->frame_control); + + /* In case of SW crypto and hdr protected (PMF), packet will already be encrypted, + * we can't put in data in this case + */ + if (test_bit(ATH11K_FLAG_HW_CRYPTO_DISABLED, &ar->ab->dev_flags) && + has_protected) + return 0; + + mgmt = (struct ieee80211_mgmt *)hdr; + buf = (u8 *)&mgmt->u.action; + + /* FCTL_PROTECTED frame might have extra space added for HDR_LEN. Offset that + * many bytes if it is there + */ + if (has_protected) { + skb_cb = ATH11K_SKB_CB(skb); + + switch (skb_cb->cipher) { + /* Cipher suite having flag %IEEE80211_KEY_FLAG_GENERATE_IV_MGMT set in + * key needs to be processed. See ath11k_install_key() + */ + case WLAN_CIPHER_SUITE_CCMP: + case WLAN_CIPHER_SUITE_CCMP_256: + case WLAN_CIPHER_SUITE_GCMP: + case WLAN_CIPHER_SUITE_GCMP_256: + iv_len = IEEE80211_CCMP_HDR_LEN; + break; + case WLAN_CIPHER_SUITE_TKIP: + iv_len = 0; + break; + default: + return -EINVAL; + } + + if (remaining_len < iv_len) + return -EINVAL; + + buf += iv_len; + remaining_len -= iv_len; + } + + category = *buf++; + /* category code is already taken care in %IEEE80211_MIN_ACTION_SIZE hence + * no need to adjust remaining_len + */ + + switch (category) { + case WLAN_CATEGORY_RADIO_MEASUREMENT: + /* need action code and dialog token */ + if (remaining_len < 2) + return -EINVAL; + + /* Packet Format: + * Action Code | Dialog Token | Variable Len (based on Action Code) + */ + action_code = *buf++; + dialog_token = *buf++; + remaining_len -= 2; + + if (ath11k_mac_vif_chan(arvif->vif, &def)) + return -ENOENT; + + cur_tx_power = arvif->vif->bss_conf.txpower; + max_tx_power = min(def.chan->max_reg_power, (int)ar->max_tx_power / 2); + ath11k_mac_handle_get_txpower(ar, arvif->vif, &cur_tx_power); + + switch (action_code) { + case WLAN_RM_ACTION_LINK_MEASUREMENT_REQUEST: + /* need variable fields to be present in len */ + if (remaining_len < 2) + return -EINVAL; + + /* Variable length format as defined in IEEE 802.11-2024, + * Figure 9-1187-Link Measurement Request frame Action field + * format. + * Transmit Power | Max Tx Power + * We fill both of these. + */ + *buf++ = cur_tx_power; + *buf = max_tx_power; + + ath11k_dbg(ar->ab, ATH11K_DBG_MAC, + "RRM: Link Measurement Req dialog_token %u cur_tx_power %d max_tx_power %d\n", + dialog_token, cur_tx_power, max_tx_power); + break; + case WLAN_RM_ACTION_LINK_MEASUREMENT_REPORT: + /* need variable fields to be present in len */ + if (remaining_len < 3) + return -EINVAL; + + /* Variable length format as defined in IEEE 802.11-2024, + * Figure 9-1188-Link Measurement Report frame Action field format + * TPC Report | Variable Fields + * + * TPC Report Format: + * Element ID | Len | Tx Power | Link Margin + * + * We fill Tx power in the TPC Report (2nd index) + */ + buf[2] = cur_tx_power; + + /* TODO: At present, Link margin data is not present so can't + * really fill it now. Once it is available, it can be added + * here + */ + ath11k_dbg(ar->ab, ATH11K_DBG_MAC, + "RRM: Link Measurement Report dialog_token %u cur_tx_power %d\n", + dialog_token, cur_tx_power); + break; + default: + return -EINVAL; + } + break; + default: + /* nothing to fill */ + return 0; + } + + return 0; +} + +static int ath11k_mac_mgmt_frame_fill_elem_data(struct ath11k_vif *arvif, + struct sk_buff *skb) +{ + struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)skb->data; + + if (!ieee80211_is_action(hdr->frame_control)) + return 0; + + return ath11k_mac_mgmt_action_frame_fill_elem_data(arvif, skb); +} + static void ath11k_mgmt_over_wmi_tx_work(struct work_struct *work) { struct ath11k *ar = container_of(work, struct ath11k, wmi_mgmt_tx_work); @@ -6270,6 +6423,19 @@ static void ath11k_mgmt_over_wmi_tx_work(struct work_struct *work) arvif = ath11k_vif_to_arvif(skb_cb->vif); mutex_lock(&ar->conf_mutex); if (ar->allocated_vdev_map & (1LL << arvif->vdev_id)) { + /* Fill in the data which is required to be filled by the driver + * For example: Max Tx power in Link Measurement Request/Report + */ + ret = ath11k_mac_mgmt_frame_fill_elem_data(arvif, skb); + if (ret) { + /* If we couldn't fill the data due to any reason, + * let's not discard transmitting the packet. + */ + ath11k_dbg(ar->ab, ATH11K_DBG_MAC, + "Failed to fill the required data for the mgmt packet err %d\n", + ret); + } + ret = ath11k_mac_mgmt_tx_wmi(ar, arvif, skb); if (ret) { ath11k_warn(ar->ab, "failed to tx mgmt frame, vdev_id %d :%d\n", From 998c68e96c03f10dec19b65279ade9d4000d1ae9 Mon Sep 17 00:00:00 2001 From: Aditya Kumar Singh Date: Fri, 17 Oct 2025 09:38:00 +0530 Subject: [PATCH 0200/1381] wifi: ath11k: advertise NL80211_FEATURE_TX_POWER_INSERTION Now that driver is capable of inserting Tx power, advertise the support for the same to upper layers. Tested-on: QCN9074 hw1.0 PCI WLAN.HK.2.9.0.1-02146-QCAHKSWPL_SILICONZ-1 Signed-off-by: Aditya Kumar Singh Reviewed-by: Vasanthakumar Thiagarajan Reviewed-by: Baochen Qiang Link: https://patch.msgid.link/20251017-add_tx_power_insertion_support-v1-4-f08feacfca93@oss.qualcomm.com Signed-off-by: Jeff Johnson --- drivers/net/wireless/ath/ath11k/mac.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c index d15e3b67e625..c6360d6a72bd 100644 --- a/drivers/net/wireless/ath/ath11k/mac.c +++ b/drivers/net/wireless/ath/ath11k/mac.c @@ -10543,6 +10543,8 @@ static int __ath11k_mac_register(struct ath11k *ar) ar->hw->wiphy->features |= NL80211_FEATURE_AP_MODE_CHAN_WIDTH_CHANGE | NL80211_FEATURE_AP_SCAN; + ar->hw->wiphy->features |= NL80211_FEATURE_TX_POWER_INSERTION; + ar->max_num_stations = TARGET_NUM_STATIONS(ab); ar->max_num_peers = TARGET_NUM_PEERS_PDEV(ab); From 50cb7ccab8176cbce4e32420f9fc2d6d80e69a09 Mon Sep 17 00:00:00 2001 From: Jeff Johnson Date: Wed, 22 Oct 2025 08:24:46 -0700 Subject: [PATCH 0201/1381] wifi: ath11k: Correctly use "ab" macro parameter The checkpatch script is reporting multiple instances of: Argument 'x' is not used in function-like macro Fix these by renaming the argument to match the usage. In the process, also add parenthesis to the usage to avoid MACRO_ARG_PRECEDENCE issues. Compile tested only. Link: https://patch.msgid.link/20251022-ath11k-bad-macro-arg-v1-1-93a8eadb6191@oss.qualcomm.com Signed-off-by: Jeff Johnson --- drivers/net/wireless/ath/ath11k/hal.h | 38 +++++++++++++-------------- drivers/net/wireless/ath/ath11k/pci.h | 18 ++++++------- 2 files changed, 28 insertions(+), 28 deletions(-) diff --git a/drivers/net/wireless/ath/ath11k/hal.h b/drivers/net/wireless/ath/ath11k/hal.h index 839095af9267..82603a389bb9 100644 --- a/drivers/net/wireless/ath/ath11k/hal.h +++ b/drivers/net/wireless/ath/ath11k/hal.h @@ -1,7 +1,7 @@ /* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved. - * Copyright (c) 2021-2022, 2024 Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. */ #ifndef ATH11K_HAL_H @@ -43,14 +43,14 @@ struct ath11k_base; #define HAL_SEQ_WCSS_UMAC_OFFSET 0x00a00000 #define HAL_SEQ_WCSS_UMAC_REO_REG 0x00a38000 #define HAL_SEQ_WCSS_UMAC_TCL_REG 0x00a44000 -#define HAL_SEQ_WCSS_UMAC_CE0_SRC_REG(x) \ - (ab->hw_params.regs->hal_seq_wcss_umac_ce0_src_reg) -#define HAL_SEQ_WCSS_UMAC_CE0_DST_REG(x) \ - (ab->hw_params.regs->hal_seq_wcss_umac_ce0_dst_reg) -#define HAL_SEQ_WCSS_UMAC_CE1_SRC_REG(x) \ - (ab->hw_params.regs->hal_seq_wcss_umac_ce1_src_reg) -#define HAL_SEQ_WCSS_UMAC_CE1_DST_REG(x) \ - (ab->hw_params.regs->hal_seq_wcss_umac_ce1_dst_reg) +#define HAL_SEQ_WCSS_UMAC_CE0_SRC_REG(ab) \ + ((ab)->hw_params.regs->hal_seq_wcss_umac_ce0_src_reg) +#define HAL_SEQ_WCSS_UMAC_CE0_DST_REG(ab) \ + ((ab)->hw_params.regs->hal_seq_wcss_umac_ce0_dst_reg) +#define HAL_SEQ_WCSS_UMAC_CE1_SRC_REG(ab) \ + ((ab)->hw_params.regs->hal_seq_wcss_umac_ce1_src_reg) +#define HAL_SEQ_WCSS_UMAC_CE1_DST_REG(ab) \ + ((ab)->hw_params.regs->hal_seq_wcss_umac_ce1_dst_reg) #define HAL_SEQ_WCSS_UMAC_WBM_REG 0x00a34000 #define HAL_CE_WFSS_CE_REG_BASE 0x01b80000 @@ -209,10 +209,10 @@ struct ath11k_base; #define HAL_REO_STATUS_HP(ab) ab->hw_params.regs->hal_reo_status_hp /* WBM Idle R0 address */ -#define HAL_WBM_IDLE_LINK_RING_BASE_LSB(x) \ - (ab->hw_params.regs->hal_wbm_idle_link_ring_base_lsb) -#define HAL_WBM_IDLE_LINK_RING_MISC_ADDR(x) \ - (ab->hw_params.regs->hal_wbm_idle_link_ring_misc) +#define HAL_WBM_IDLE_LINK_RING_BASE_LSB(ab) \ + ((ab)->hw_params.regs->hal_wbm_idle_link_ring_base_lsb) +#define HAL_WBM_IDLE_LINK_RING_MISC_ADDR(ab) \ + ((ab)->hw_params.regs->hal_wbm_idle_link_ring_misc) #define HAL_WBM_R0_IDLE_LIST_CONTROL_ADDR 0x00000048 #define HAL_WBM_R0_IDLE_LIST_SIZE_ADDR 0x0000004c #define HAL_WBM_SCATTERED_RING_BASE_LSB 0x00000058 @@ -227,17 +227,17 @@ struct ath11k_base; #define HAL_WBM_IDLE_LINK_RING_HP 0x000030b0 /* SW2WBM R0 release address */ -#define HAL_WBM_RELEASE_RING_BASE_LSB(x) \ - (ab->hw_params.regs->hal_wbm_release_ring_base_lsb) +#define HAL_WBM_RELEASE_RING_BASE_LSB(ab) \ + ((ab)->hw_params.regs->hal_wbm_release_ring_base_lsb) /* SW2WBM R2 release address */ #define HAL_WBM_RELEASE_RING_HP 0x00003018 /* WBM2SW R0 release address */ -#define HAL_WBM0_RELEASE_RING_BASE_LSB(x) \ - (ab->hw_params.regs->hal_wbm0_release_ring_base_lsb) -#define HAL_WBM1_RELEASE_RING_BASE_LSB(x) \ - (ab->hw_params.regs->hal_wbm1_release_ring_base_lsb) +#define HAL_WBM0_RELEASE_RING_BASE_LSB(ab) \ + ((ab)->hw_params.regs->hal_wbm0_release_ring_base_lsb) +#define HAL_WBM1_RELEASE_RING_BASE_LSB(ab) \ + ((ab)->hw_params.regs->hal_wbm1_release_ring_base_lsb) /* WBM2SW R2 release address */ #define HAL_WBM0_RELEASE_RING_HP 0x000030c0 diff --git a/drivers/net/wireless/ath/ath11k/pci.h b/drivers/net/wireless/ath/ath11k/pci.h index c33c7865145c..1e3005a4b64c 100644 --- a/drivers/net/wireless/ath/ath11k/pci.h +++ b/drivers/net/wireless/ath/ath11k/pci.h @@ -1,7 +1,7 @@ /* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* * Copyright (c) 2019-2020 The Linux Foundation. All rights reserved. - * Copyright (c) 2021-2022,2024 Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. */ #ifndef _ATH11K_PCI_H #define _ATH11K_PCI_H @@ -35,18 +35,18 @@ #define PCIE_SMLH_REQ_RST_LINK_DOWN 0x2 #define PCIE_INT_CLEAR_ALL 0xffffffff -#define PCIE_QSERDES_COM_SYSCLK_EN_SEL_REG(x) \ - (ab->hw_params.regs->pcie_qserdes_sysclk_en_sel) +#define PCIE_QSERDES_COM_SYSCLK_EN_SEL_REG(ab) \ + ((ab)->hw_params.regs->pcie_qserdes_sysclk_en_sel) #define PCIE_QSERDES_COM_SYSCLK_EN_SEL_VAL 0x10 #define PCIE_QSERDES_COM_SYSCLK_EN_SEL_MSK 0xffffffff -#define PCIE_PCS_OSC_DTCT_CONFIG1_REG(x) \ - (ab->hw_params.regs->pcie_pcs_osc_dtct_config_base) +#define PCIE_PCS_OSC_DTCT_CONFIG1_REG(ab) \ + ((ab)->hw_params.regs->pcie_pcs_osc_dtct_config_base) #define PCIE_PCS_OSC_DTCT_CONFIG1_VAL 0x02 -#define PCIE_PCS_OSC_DTCT_CONFIG2_REG(x) \ - (ab->hw_params.regs->pcie_pcs_osc_dtct_config_base + 0x4) +#define PCIE_PCS_OSC_DTCT_CONFIG2_REG(ab) \ + ((ab)->hw_params.regs->pcie_pcs_osc_dtct_config_base + 0x4) #define PCIE_PCS_OSC_DTCT_CONFIG2_VAL 0x52 -#define PCIE_PCS_OSC_DTCT_CONFIG4_REG(x) \ - (ab->hw_params.regs->pcie_pcs_osc_dtct_config_base + 0xc) +#define PCIE_PCS_OSC_DTCT_CONFIG4_REG(ab) \ + ((ab)->hw_params.regs->pcie_pcs_osc_dtct_config_base + 0xc) #define PCIE_PCS_OSC_DTCT_CONFIG4_VAL 0xff #define PCIE_PCS_OSC_DTCT_CONFIG_MSK 0x000000ff From 99ad2b6815f41acbec15ab051ccc79b11b05710a Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Wed, 22 Oct 2025 09:11:12 +0200 Subject: [PATCH 0202/1381] net: airoha: Remove code duplication in airoha_regs.h This patch does not introduce any logical change, it just removes duplicated code in airoha_regs.h. Fix naming conventions in airoha_regs.h. Reviewed-by: Simon Horman Signed-off-by: Lorenzo Bianconi Link: https://patch.msgid.link/20251022-airoha-regs-cosmetics-v2-1-e0425b3f2c2c@kernel.org Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/airoha/airoha_eth.c | 102 ++++++++++---------- drivers/net/ethernet/airoha/airoha_regs.h | 109 ++++++++++------------ 2 files changed, 100 insertions(+), 111 deletions(-) diff --git a/drivers/net/ethernet/airoha/airoha_eth.c b/drivers/net/ethernet/airoha/airoha_eth.c index 8483ea02603e..e17a285a9e8f 100644 --- a/drivers/net/ethernet/airoha/airoha_eth.c +++ b/drivers/net/ethernet/airoha/airoha_eth.c @@ -137,11 +137,11 @@ static void airoha_fe_maccr_init(struct airoha_eth *eth) for (p = 1; p <= ARRAY_SIZE(eth->ports); p++) airoha_fe_set(eth, REG_GDM_FWD_CFG(p), - GDM_TCP_CKSUM | GDM_UDP_CKSUM | GDM_IP4_CKSUM | - GDM_DROP_CRC_ERR); + GDM_TCP_CKSUM_MASK | GDM_UDP_CKSUM_MASK | + GDM_IP4_CKSUM_MASK | GDM_DROP_CRC_ERR_MASK); - airoha_fe_rmw(eth, REG_CDM1_VLAN_CTRL, CDM1_VLAN_MASK, - FIELD_PREP(CDM1_VLAN_MASK, 0x8100)); + airoha_fe_rmw(eth, REG_CDM_VLAN_CTRL(1), CDM_VLAN_MASK, + FIELD_PREP(CDM_VLAN_MASK, 0x8100)); airoha_fe_set(eth, REG_FE_CPORT_CFG, FE_CPORT_PAD); } @@ -403,46 +403,46 @@ static int airoha_fe_mc_vlan_clear(struct airoha_eth *eth) static void airoha_fe_crsn_qsel_init(struct airoha_eth *eth) { /* CDM1_CRSN_QSEL */ - airoha_fe_rmw(eth, REG_CDM1_CRSN_QSEL(CRSN_22 >> 2), - CDM1_CRSN_QSEL_REASON_MASK(CRSN_22), - FIELD_PREP(CDM1_CRSN_QSEL_REASON_MASK(CRSN_22), + airoha_fe_rmw(eth, REG_CDM_CRSN_QSEL(1, CRSN_22 >> 2), + CDM_CRSN_QSEL_REASON_MASK(CRSN_22), + FIELD_PREP(CDM_CRSN_QSEL_REASON_MASK(CRSN_22), CDM_CRSN_QSEL_Q1)); - airoha_fe_rmw(eth, REG_CDM1_CRSN_QSEL(CRSN_08 >> 2), - CDM1_CRSN_QSEL_REASON_MASK(CRSN_08), - FIELD_PREP(CDM1_CRSN_QSEL_REASON_MASK(CRSN_08), + airoha_fe_rmw(eth, REG_CDM_CRSN_QSEL(1, CRSN_08 >> 2), + CDM_CRSN_QSEL_REASON_MASK(CRSN_08), + FIELD_PREP(CDM_CRSN_QSEL_REASON_MASK(CRSN_08), CDM_CRSN_QSEL_Q1)); - airoha_fe_rmw(eth, REG_CDM1_CRSN_QSEL(CRSN_21 >> 2), - CDM1_CRSN_QSEL_REASON_MASK(CRSN_21), - FIELD_PREP(CDM1_CRSN_QSEL_REASON_MASK(CRSN_21), + airoha_fe_rmw(eth, REG_CDM_CRSN_QSEL(1, CRSN_21 >> 2), + CDM_CRSN_QSEL_REASON_MASK(CRSN_21), + FIELD_PREP(CDM_CRSN_QSEL_REASON_MASK(CRSN_21), CDM_CRSN_QSEL_Q1)); - airoha_fe_rmw(eth, REG_CDM1_CRSN_QSEL(CRSN_24 >> 2), - CDM1_CRSN_QSEL_REASON_MASK(CRSN_24), - FIELD_PREP(CDM1_CRSN_QSEL_REASON_MASK(CRSN_24), + airoha_fe_rmw(eth, REG_CDM_CRSN_QSEL(1, CRSN_24 >> 2), + CDM_CRSN_QSEL_REASON_MASK(CRSN_24), + FIELD_PREP(CDM_CRSN_QSEL_REASON_MASK(CRSN_24), CDM_CRSN_QSEL_Q6)); - airoha_fe_rmw(eth, REG_CDM1_CRSN_QSEL(CRSN_25 >> 2), - CDM1_CRSN_QSEL_REASON_MASK(CRSN_25), - FIELD_PREP(CDM1_CRSN_QSEL_REASON_MASK(CRSN_25), + airoha_fe_rmw(eth, REG_CDM_CRSN_QSEL(1, CRSN_25 >> 2), + CDM_CRSN_QSEL_REASON_MASK(CRSN_25), + FIELD_PREP(CDM_CRSN_QSEL_REASON_MASK(CRSN_25), CDM_CRSN_QSEL_Q1)); /* CDM2_CRSN_QSEL */ - airoha_fe_rmw(eth, REG_CDM2_CRSN_QSEL(CRSN_08 >> 2), - CDM2_CRSN_QSEL_REASON_MASK(CRSN_08), - FIELD_PREP(CDM2_CRSN_QSEL_REASON_MASK(CRSN_08), + airoha_fe_rmw(eth, REG_CDM_CRSN_QSEL(2, CRSN_08 >> 2), + CDM_CRSN_QSEL_REASON_MASK(CRSN_08), + FIELD_PREP(CDM_CRSN_QSEL_REASON_MASK(CRSN_08), CDM_CRSN_QSEL_Q1)); - airoha_fe_rmw(eth, REG_CDM2_CRSN_QSEL(CRSN_21 >> 2), - CDM2_CRSN_QSEL_REASON_MASK(CRSN_21), - FIELD_PREP(CDM2_CRSN_QSEL_REASON_MASK(CRSN_21), + airoha_fe_rmw(eth, REG_CDM_CRSN_QSEL(2, CRSN_21 >> 2), + CDM_CRSN_QSEL_REASON_MASK(CRSN_21), + FIELD_PREP(CDM_CRSN_QSEL_REASON_MASK(CRSN_21), CDM_CRSN_QSEL_Q1)); - airoha_fe_rmw(eth, REG_CDM2_CRSN_QSEL(CRSN_22 >> 2), - CDM2_CRSN_QSEL_REASON_MASK(CRSN_22), - FIELD_PREP(CDM2_CRSN_QSEL_REASON_MASK(CRSN_22), + airoha_fe_rmw(eth, REG_CDM_CRSN_QSEL(2, CRSN_22 >> 2), + CDM_CRSN_QSEL_REASON_MASK(CRSN_22), + FIELD_PREP(CDM_CRSN_QSEL_REASON_MASK(CRSN_22), CDM_CRSN_QSEL_Q1)); - airoha_fe_rmw(eth, REG_CDM2_CRSN_QSEL(CRSN_24 >> 2), - CDM2_CRSN_QSEL_REASON_MASK(CRSN_24), - FIELD_PREP(CDM2_CRSN_QSEL_REASON_MASK(CRSN_24), + airoha_fe_rmw(eth, REG_CDM_CRSN_QSEL(2, CRSN_24 >> 2), + CDM_CRSN_QSEL_REASON_MASK(CRSN_24), + FIELD_PREP(CDM_CRSN_QSEL_REASON_MASK(CRSN_24), CDM_CRSN_QSEL_Q6)); - airoha_fe_rmw(eth, REG_CDM2_CRSN_QSEL(CRSN_25 >> 2), - CDM2_CRSN_QSEL_REASON_MASK(CRSN_25), - FIELD_PREP(CDM2_CRSN_QSEL_REASON_MASK(CRSN_25), + airoha_fe_rmw(eth, REG_CDM_CRSN_QSEL(2, CRSN_25 >> 2), + CDM_CRSN_QSEL_REASON_MASK(CRSN_25), + FIELD_PREP(CDM_CRSN_QSEL_REASON_MASK(CRSN_25), CDM_CRSN_QSEL_Q1)); } @@ -462,18 +462,18 @@ static int airoha_fe_init(struct airoha_eth *eth) airoha_fe_wr(eth, REG_FE_PCE_CFG, PCE_DPI_EN_MASK | PCE_KA_EN_MASK | PCE_MC_EN_MASK); /* set vip queue selection to ring 1 */ - airoha_fe_rmw(eth, REG_CDM1_FWD_CFG, CDM1_VIP_QSEL_MASK, - FIELD_PREP(CDM1_VIP_QSEL_MASK, 0x4)); - airoha_fe_rmw(eth, REG_CDM2_FWD_CFG, CDM2_VIP_QSEL_MASK, - FIELD_PREP(CDM2_VIP_QSEL_MASK, 0x4)); + airoha_fe_rmw(eth, REG_CDM_FWD_CFG(1), CDM_VIP_QSEL_MASK, + FIELD_PREP(CDM_VIP_QSEL_MASK, 0x4)); + airoha_fe_rmw(eth, REG_CDM_FWD_CFG(2), CDM_VIP_QSEL_MASK, + FIELD_PREP(CDM_VIP_QSEL_MASK, 0x4)); /* set GDM4 source interface offset to 8 */ - airoha_fe_rmw(eth, REG_GDM4_SRC_PORT_SET, - GDM4_SPORT_OFF2_MASK | - GDM4_SPORT_OFF1_MASK | - GDM4_SPORT_OFF0_MASK, - FIELD_PREP(GDM4_SPORT_OFF2_MASK, 8) | - FIELD_PREP(GDM4_SPORT_OFF1_MASK, 8) | - FIELD_PREP(GDM4_SPORT_OFF0_MASK, 8)); + airoha_fe_rmw(eth, REG_GDM_SRC_PORT_SET(4), + GDM_SPORT_OFF2_MASK | + GDM_SPORT_OFF1_MASK | + GDM_SPORT_OFF0_MASK, + FIELD_PREP(GDM_SPORT_OFF2_MASK, 8) | + FIELD_PREP(GDM_SPORT_OFF1_MASK, 8) | + FIELD_PREP(GDM_SPORT_OFF0_MASK, 8)); /* set PSE Page as 128B */ airoha_fe_rmw(eth, REG_FE_DMA_GLO_CFG, @@ -499,8 +499,8 @@ static int airoha_fe_init(struct airoha_eth *eth) airoha_fe_set(eth, REG_GDM_MISC_CFG, GDM2_RDM_ACK_WAIT_PREF_MASK | GDM2_CHN_VLD_MODE_MASK); - airoha_fe_rmw(eth, REG_CDM2_FWD_CFG, CDM2_OAM_QSEL_MASK, - FIELD_PREP(CDM2_OAM_QSEL_MASK, 15)); + airoha_fe_rmw(eth, REG_CDM_FWD_CFG(2), CDM_OAM_QSEL_MASK, + FIELD_PREP(CDM_OAM_QSEL_MASK, 15)); /* init fragment and assemble Force Port */ /* NPU Core-3, NPU Bridge Channel-3 */ @@ -514,8 +514,8 @@ static int airoha_fe_init(struct airoha_eth *eth) FIELD_PREP(IP_ASSEMBLE_PORT_MASK, 0) | FIELD_PREP(IP_ASSEMBLE_NBQ_MASK, 22)); - airoha_fe_set(eth, REG_GDM3_FWD_CFG, GDM3_PAD_EN_MASK); - airoha_fe_set(eth, REG_GDM4_FWD_CFG, GDM4_PAD_EN_MASK); + airoha_fe_set(eth, REG_GDM_FWD_CFG(3), GDM_PAD_EN_MASK); + airoha_fe_set(eth, REG_GDM_FWD_CFG(4), GDM_PAD_EN_MASK); airoha_fe_crsn_qsel_init(eth); @@ -523,7 +523,7 @@ static int airoha_fe_init(struct airoha_eth *eth) airoha_fe_set(eth, REG_FE_CPORT_CFG, FE_CPORT_PORT_XFC_MASK); /* default aging mode for mbi unlock issue */ - airoha_fe_rmw(eth, REG_GDM2_CHN_RLS, + airoha_fe_rmw(eth, REG_GDM_CHN_RLS(2), MBI_RX_AGE_SEL_MASK | MBI_TX_AGE_SEL_MASK, FIELD_PREP(MBI_RX_AGE_SEL_MASK, 3) | FIELD_PREP(MBI_TX_AGE_SEL_MASK, 3)); @@ -1692,7 +1692,7 @@ static int airhoha_set_gdm2_loopback(struct airoha_gdm_port *port) pse_port = port->id == AIROHA_GDM3_IDX ? FE_PSE_PORT_GDM3 : FE_PSE_PORT_GDM4; airoha_set_gdm_port_fwd_cfg(eth, REG_GDM_FWD_CFG(2), pse_port); - airoha_fe_clear(eth, REG_GDM_FWD_CFG(2), GDM_STRIP_CRC); + airoha_fe_clear(eth, REG_GDM_FWD_CFG(2), GDM_STRIP_CRC_MASK); /* Enable GDM2 loopback */ airoha_fe_wr(eth, REG_GDM_TXCHN_EN(2), 0xffffffff); diff --git a/drivers/net/ethernet/airoha/airoha_regs.h b/drivers/net/ethernet/airoha/airoha_regs.h index ebcce00d9bc6..ed4e3407f4a0 100644 --- a/drivers/net/ethernet/airoha/airoha_regs.h +++ b/drivers/net/ethernet/airoha/airoha_regs.h @@ -23,6 +23,8 @@ #define GDM3_BASE 0x1100 #define GDM4_BASE 0x2500 +#define CDM_BASE(_n) \ + ((_n) == 2 ? CDM2_BASE : CDM1_BASE) #define GDM_BASE(_n) \ ((_n) == 4 ? GDM4_BASE : \ (_n) == 3 ? GDM3_BASE : \ @@ -109,30 +111,24 @@ #define PATN_DP_MASK GENMASK(31, 16) #define PATN_SP_MASK GENMASK(15, 0) -#define REG_CDM1_VLAN_CTRL CDM1_BASE -#define CDM1_VLAN_MASK GENMASK(31, 16) +#define REG_CDM_VLAN_CTRL(_n) CDM_BASE(_n) +#define CDM_VLAN_MASK GENMASK(31, 16) -#define REG_CDM1_FWD_CFG (CDM1_BASE + 0x08) -#define CDM1_VIP_QSEL_MASK GENMASK(24, 20) +#define REG_CDM_FWD_CFG(_n) (CDM_BASE(_n) + 0x08) +#define CDM_OAM_QSEL_MASK GENMASK(31, 27) +#define CDM_VIP_QSEL_MASK GENMASK(24, 20) -#define REG_CDM1_CRSN_QSEL(_n) (CDM1_BASE + 0x10 + ((_n) << 2)) -#define CDM1_CRSN_QSEL_REASON_MASK(_n) \ - GENMASK(4 + (((_n) % 4) << 3), (((_n) % 4) << 3)) - -#define REG_CDM2_FWD_CFG (CDM2_BASE + 0x08) -#define CDM2_OAM_QSEL_MASK GENMASK(31, 27) -#define CDM2_VIP_QSEL_MASK GENMASK(24, 20) - -#define REG_CDM2_CRSN_QSEL(_n) (CDM2_BASE + 0x10 + ((_n) << 2)) -#define CDM2_CRSN_QSEL_REASON_MASK(_n) \ +#define REG_CDM_CRSN_QSEL(_n, _m) (CDM_BASE(_n) + 0x10 + ((_m) << 2)) +#define CDM_CRSN_QSEL_REASON_MASK(_n) \ GENMASK(4 + (((_n) % 4) << 3), (((_n) % 4) << 3)) #define REG_GDM_FWD_CFG(_n) GDM_BASE(_n) -#define GDM_DROP_CRC_ERR BIT(23) -#define GDM_IP4_CKSUM BIT(22) -#define GDM_TCP_CKSUM BIT(21) -#define GDM_UDP_CKSUM BIT(20) -#define GDM_STRIP_CRC BIT(16) +#define GDM_PAD_EN_MASK BIT(28) +#define GDM_DROP_CRC_ERR_MASK BIT(23) +#define GDM_IP4_CKSUM_MASK BIT(22) +#define GDM_TCP_CKSUM_MASK BIT(21) +#define GDM_UDP_CKSUM_MASK BIT(20) +#define GDM_STRIP_CRC_MASK BIT(16) #define GDM_UCFQ_MASK GENMASK(15, 12) #define GDM_BCFQ_MASK GENMASK(11, 8) #define GDM_MCFQ_MASK GENMASK(7, 4) @@ -156,6 +152,10 @@ #define LBK_CHAN_MODE_MASK BIT(1) #define LPBK_EN_MASK BIT(0) +#define REG_GDM_CHN_RLS(_n) (GDM_BASE(_n) + 0x20) +#define MBI_RX_AGE_SEL_MASK GENMASK(26, 25) +#define MBI_TX_AGE_SEL_MASK GENMASK(18, 17) + #define REG_GDM_TXCHN_EN(_n) (GDM_BASE(_n) + 0x24) #define REG_GDM_RXCHN_EN(_n) (GDM_BASE(_n) + 0x28) @@ -168,10 +168,10 @@ #define FE_GDM_MIB_RX_CLEAR_MASK BIT(1) #define FE_GDM_MIB_TX_CLEAR_MASK BIT(0) -#define REG_FE_GDM1_MIB_CFG (GDM1_BASE + 0xf4) +#define REG_FE_GDM_MIB_CFG(_n) (GDM_BASE(_n) + 0xf4) #define FE_STRICT_RFC2819_MODE_MASK BIT(31) -#define FE_GDM1_TX_MIB_SPLIT_EN_MASK BIT(17) -#define FE_GDM1_RX_MIB_SPLIT_EN_MASK BIT(16) +#define FE_GDM_TX_MIB_SPLIT_EN_MASK BIT(17) +#define FE_GDM_RX_MIB_SPLIT_EN_MASK BIT(16) #define FE_TX_MIB_ID_MASK GENMASK(15, 8) #define FE_RX_MIB_ID_MASK GENMASK(7, 0) @@ -214,6 +214,33 @@ #define REG_FE_GDM_RX_ETH_L511_CNT_L(_n) (GDM_BASE(_n) + 0x198) #define REG_FE_GDM_RX_ETH_L1023_CNT_L(_n) (GDM_BASE(_n) + 0x19c) +#define REG_GDM_SRC_PORT_SET(_n) (GDM_BASE(_n) + 0x23c) +#define GDM_SPORT_OFF2_MASK GENMASK(19, 16) +#define GDM_SPORT_OFF1_MASK GENMASK(15, 12) +#define GDM_SPORT_OFF0_MASK GENMASK(11, 8) + +#define REG_FE_GDM_TX_OK_PKT_CNT_H(_n) (GDM_BASE(_n) + 0x280) +#define REG_FE_GDM_TX_OK_BYTE_CNT_H(_n) (GDM_BASE(_n) + 0x284) +#define REG_FE_GDM_TX_ETH_PKT_CNT_H(_n) (GDM_BASE(_n) + 0x288) +#define REG_FE_GDM_TX_ETH_BYTE_CNT_H(_n) (GDM_BASE(_n) + 0x28c) + +#define REG_FE_GDM_RX_OK_PKT_CNT_H(_n) (GDM_BASE(_n) + 0x290) +#define REG_FE_GDM_RX_OK_BYTE_CNT_H(_n) (GDM_BASE(_n) + 0x294) +#define REG_FE_GDM_RX_ETH_PKT_CNT_H(_n) (GDM_BASE(_n) + 0x298) +#define REG_FE_GDM_RX_ETH_BYTE_CNT_H(_n) (GDM_BASE(_n) + 0x29c) +#define REG_FE_GDM_TX_ETH_E64_CNT_H(_n) (GDM_BASE(_n) + 0x2b8) +#define REG_FE_GDM_TX_ETH_L64_CNT_H(_n) (GDM_BASE(_n) + 0x2bc) +#define REG_FE_GDM_TX_ETH_L127_CNT_H(_n) (GDM_BASE(_n) + 0x2c0) +#define REG_FE_GDM_TX_ETH_L255_CNT_H(_n) (GDM_BASE(_n) + 0x2c4) +#define REG_FE_GDM_TX_ETH_L511_CNT_H(_n) (GDM_BASE(_n) + 0x2c8) +#define REG_FE_GDM_TX_ETH_L1023_CNT_H(_n) (GDM_BASE(_n) + 0x2cc) +#define REG_FE_GDM_RX_ETH_E64_CNT_H(_n) (GDM_BASE(_n) + 0x2e8) +#define REG_FE_GDM_RX_ETH_L64_CNT_H(_n) (GDM_BASE(_n) + 0x2ec) +#define REG_FE_GDM_RX_ETH_L127_CNT_H(_n) (GDM_BASE(_n) + 0x2f0) +#define REG_FE_GDM_RX_ETH_L255_CNT_H(_n) (GDM_BASE(_n) + 0x2f4) +#define REG_FE_GDM_RX_ETH_L511_CNT_H(_n) (GDM_BASE(_n) + 0x2f8) +#define REG_FE_GDM_RX_ETH_L1023_CNT_H(_n) (GDM_BASE(_n) + 0x2fc) + #define REG_PPE_GLO_CFG(_n) (((_n) ? PPE2_BASE : PPE1_BASE) + 0x200) #define PPE_GLO_CFG_BUSY_MASK BIT(31) #define PPE_GLO_CFG_FLOW_DROP_UPDATE_MASK BIT(9) @@ -326,44 +353,6 @@ #define REG_UPDMEM_DATA(_n) (((_n) ? PPE2_BASE : PPE1_BASE) + 0x374) -#define REG_FE_GDM_TX_OK_PKT_CNT_H(_n) (GDM_BASE(_n) + 0x280) -#define REG_FE_GDM_TX_OK_BYTE_CNT_H(_n) (GDM_BASE(_n) + 0x284) -#define REG_FE_GDM_TX_ETH_PKT_CNT_H(_n) (GDM_BASE(_n) + 0x288) -#define REG_FE_GDM_TX_ETH_BYTE_CNT_H(_n) (GDM_BASE(_n) + 0x28c) - -#define REG_FE_GDM_RX_OK_PKT_CNT_H(_n) (GDM_BASE(_n) + 0x290) -#define REG_FE_GDM_RX_OK_BYTE_CNT_H(_n) (GDM_BASE(_n) + 0x294) -#define REG_FE_GDM_RX_ETH_PKT_CNT_H(_n) (GDM_BASE(_n) + 0x298) -#define REG_FE_GDM_RX_ETH_BYTE_CNT_H(_n) (GDM_BASE(_n) + 0x29c) -#define REG_FE_GDM_TX_ETH_E64_CNT_H(_n) (GDM_BASE(_n) + 0x2b8) -#define REG_FE_GDM_TX_ETH_L64_CNT_H(_n) (GDM_BASE(_n) + 0x2bc) -#define REG_FE_GDM_TX_ETH_L127_CNT_H(_n) (GDM_BASE(_n) + 0x2c0) -#define REG_FE_GDM_TX_ETH_L255_CNT_H(_n) (GDM_BASE(_n) + 0x2c4) -#define REG_FE_GDM_TX_ETH_L511_CNT_H(_n) (GDM_BASE(_n) + 0x2c8) -#define REG_FE_GDM_TX_ETH_L1023_CNT_H(_n) (GDM_BASE(_n) + 0x2cc) -#define REG_FE_GDM_RX_ETH_E64_CNT_H(_n) (GDM_BASE(_n) + 0x2e8) -#define REG_FE_GDM_RX_ETH_L64_CNT_H(_n) (GDM_BASE(_n) + 0x2ec) -#define REG_FE_GDM_RX_ETH_L127_CNT_H(_n) (GDM_BASE(_n) + 0x2f0) -#define REG_FE_GDM_RX_ETH_L255_CNT_H(_n) (GDM_BASE(_n) + 0x2f4) -#define REG_FE_GDM_RX_ETH_L511_CNT_H(_n) (GDM_BASE(_n) + 0x2f8) -#define REG_FE_GDM_RX_ETH_L1023_CNT_H(_n) (GDM_BASE(_n) + 0x2fc) - -#define REG_GDM2_CHN_RLS (GDM2_BASE + 0x20) -#define MBI_RX_AGE_SEL_MASK GENMASK(26, 25) -#define MBI_TX_AGE_SEL_MASK GENMASK(18, 17) - -#define REG_GDM3_FWD_CFG GDM3_BASE -#define GDM3_PAD_EN_MASK BIT(28) - -#define REG_GDM4_FWD_CFG GDM4_BASE -#define GDM4_PAD_EN_MASK BIT(28) -#define GDM4_SPORT_OFFSET0_MASK GENMASK(11, 8) - -#define REG_GDM4_SRC_PORT_SET (GDM4_BASE + 0x23c) -#define GDM4_SPORT_OFF2_MASK GENMASK(19, 16) -#define GDM4_SPORT_OFF1_MASK GENMASK(15, 12) -#define GDM4_SPORT_OFF0_MASK GENMASK(11, 8) - #define REG_IP_FRAG_FP 0x2010 #define IP_ASSEMBLE_PORT_MASK GENMASK(24, 21) #define IP_ASSEMBLE_NBQ_MASK GENMASK(20, 16) From ec538867a376c10161decc24318ab25a503622c7 Mon Sep 17 00:00:00 2001 From: Sunday Adelodun Date: Tue, 21 Oct 2025 20:59:06 +0100 Subject: [PATCH 0203/1381] net: unix: remove outdated BSD behavior comment in unix_release_sock() Remove the long-standing comment in unix_release_sock() that described a behavioral difference between Linux and BSD regarding when ECONNRESET is sent to connected UNIX sockets upon closure. As confirmed by testing on macOS (similar to BSD behavior), ECONNRESET is only observed for SOCK_DGRAM sockets, not for SOCK_STREAM. Meanwhile, Linux already returns ECONNRESET in cases where a socket is closed with unread data or is not yet accept()ed. This means the previous comment no longer accurately describes current behavior and is misleading. Suggested-by: Kuniyuki Iwashima Signed-off-by: Sunday Adelodun Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20251021195906.20389-1-adelodunolaoluwa@yahoo.com Signed-off-by: Jakub Kicinski --- net/unix/af_unix.c | 11 ----------- 1 file changed, 11 deletions(-) diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index 768098dec231..54177caa9c12 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -733,17 +733,6 @@ static void unix_release_sock(struct sock *sk, int embrion) /* ---- Socket is dead now and most probably destroyed ---- */ - /* - * Fixme: BSD difference: In BSD all sockets connected to us get - * ECONNRESET and we die on the spot. In Linux we behave - * like files and pipes do and wait for the last - * dereference. - * - * Can't we simply set sock->err? - * - * What the above comment does talk about? --ANK(980817) - */ - if (READ_ONCE(unix_tot_inflight)) unix_gc(); /* Garbage collect fds */ } From 9ff86092655f7a22cc115188e3bf1b88ef49ec9e Mon Sep 17 00:00:00 2001 From: Bagas Sanjaya Date: Wed, 22 Oct 2025 09:54:57 +0700 Subject: [PATCH 0204/1381] net: rmnet: Use section heading markup for packet format subsections Format subsections of "Packet format" section as reST subsections. Link: https://lore.kernel.org/linux-doc/aO_MefPIlQQrCU3j@horms.kernel.org/ Suggested-by: Simon Horman Signed-off-by: Bagas Sanjaya Reviewed-by: Randy Dunlap Tested-by: Randy Dunlap Link: https://patch.msgid.link/20251022025456.19004-2-bagasdotme@gmail.com Signed-off-by: Jakub Kicinski --- .../device_drivers/cellular/qualcomm/rmnet.rst | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/Documentation/networking/device_drivers/cellular/qualcomm/rmnet.rst b/Documentation/networking/device_drivers/cellular/qualcomm/rmnet.rst index 6877a3260582..5aedbabb7382 100644 --- a/Documentation/networking/device_drivers/cellular/qualcomm/rmnet.rst +++ b/Documentation/networking/device_drivers/cellular/qualcomm/rmnet.rst @@ -28,6 +28,7 @@ these MAP frames and send them to appropriate PDN's. ================ a. MAP packet v1 (data / control) +--------------------------------- MAP header fields are in big endian format. @@ -54,6 +55,7 @@ Payload length includes the padding length but does not include MAP header length. b. Map packet v4 (data / control) +--------------------------------- MAP header fields are in big endian format. @@ -107,6 +109,7 @@ over which checksum is computed. Checksum value, indicates the checksum computed. c. MAP packet v5 (data / control) +--------------------------------- MAP header fields are in big endian format. @@ -134,6 +137,7 @@ Payload length includes the padding length but does not include MAP header length. d. Checksum offload header v5 +----------------------------- Checksum offload header fields are in big endian format. @@ -158,7 +162,10 @@ indicates that the calculated packet checksum is invalid. Reserved bits must be zero when sent and ignored when received. -e. MAP packet v1/v5 (command specific):: +e. MAP packet v1/v5 (command specific) +-------------------------------------- + +Packet format:: Bit 0 1 2-7 8 - 15 16 - 31 Function Command Reserved Pad Multiplexer ID Payload length @@ -181,6 +188,7 @@ Command types = ========================================== f. Aggregation +-------------- Aggregation is multiple MAP packets (can be data or command) delivered to rmnet in a single linear skb. rmnet will process the individual From 05774d7e4201b673b415d65a1344dc396e7c00e1 Mon Sep 17 00:00:00 2001 From: Eric Biggers Date: Wed, 22 Oct 2025 15:12:09 -0700 Subject: [PATCH 0205/1381] tcp: Remove unnecessary null check in tcp_inbound_md5_hash() The 'if (!key && hash_location)' check in tcp_inbound_md5_hash() implies that hash_location might be null. However, later code in the function dereferences hash_location anyway, without checking for null first. Fortunately, there is no real bug, since tcp_inbound_md5_hash() is called only with non-null values of hash_location. Therefore, remove the unnecessary and misleading null check of hash_location. This silences a Smatch static checker warning (https://lore.kernel.org/netdev/aPi4b6aWBbBR52P1@stanley.mountain/) Also fix the related comment at the beginning of the function. Signed-off-by: Eric Biggers Reviewed-by: Kuniyuki Iwashima Reviewed-by: Dmitry Safonov <0x7f454c46@gmail.com> Link: https://patch.msgid.link/20251022221209.19716-1-ebiggers@kernel.org Signed-off-by: Jakub Kicinski --- net/ipv4/tcp.c | 10 ++++------ 1 file changed, 4 insertions(+), 6 deletions(-) diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index e15b38f6bd2d..b79da6d39392 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -4886,18 +4886,16 @@ tcp_inbound_md5_hash(const struct sock *sk, const struct sk_buff *skb, int family, int l3index, const __u8 *hash_location) { /* This gets called for each TCP segment that has TCP-MD5 option. - * We have 3 drop cases: - * o No MD5 hash and one expected. - * o MD5 hash and we're not expecting one. - * o MD5 hash and its wrong. + * We have 2 drop cases: + * o An MD5 signature is present, but we're not expecting one. + * o The MD5 signature is wrong. */ const struct tcp_sock *tp = tcp_sk(sk); struct tcp_md5sig_key *key; u8 newhash[16]; key = tcp_md5_do_lookup(sk, l3index, saddr, family); - - if (!key && hash_location) { + if (!key) { NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPMD5UNEXPECTED); trace_tcp_hash_md5_unexpected(sk, skb); return SKB_DROP_REASON_TCP_MD5UNEXPECTED; From 41b66240e90b0dceb488ce5a4a285a97e6ca63c7 Mon Sep 17 00:00:00 2001 From: Daniel Golle Date: Tue, 21 Oct 2025 12:16:30 +0100 Subject: [PATCH 0206/1381] net: dsa: lantiq_gswip: clarify GSWIP 2.2 VLAN mode in comment The comment above writing the default PVID incorrectly states that "GSWIP 2.2 (GRX300) and later program here the VID directly." The truth is that even GSWIP 2.2 and newer maintain the behavior of GSWIP 2.1 unless the VLANMD bit in PCE Global Control Register 1 is set ("GSWIP2.2 VLAN Mode"). Fix the misleading comment accordingly. Signed-off-by: Daniel Golle Acked-by; Hauke Mehrtens : Acked-by; Hauke Mehrtens : Link: https://patch.msgid.link/018056a575503d9797f3222f71a988e825316be0.1761045000.git.daniel@makrotopia.org Signed-off-by: Jakub Kicinski --- drivers/net/dsa/lantiq/lantiq_gswip.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/drivers/net/dsa/lantiq/lantiq_gswip.c b/drivers/net/dsa/lantiq/lantiq_gswip.c index 25f6b46957a0..86b410a40d32 100644 --- a/drivers/net/dsa/lantiq/lantiq_gswip.c +++ b/drivers/net/dsa/lantiq/lantiq_gswip.c @@ -588,7 +588,11 @@ static void gswip_port_commit_pvid(struct gswip_priv *priv, int port) FIELD_PREP(GSWIP_PCE_VCTRL_VINR, vinr), GSWIP_PCE_VCTRL(port)); - /* GSWIP 2.2 (GRX300) and later program here the VID directly. */ + /* Note that in GSWIP 2.2 VLAN mode the VID needs to be programmed + * directly instead of referencing the index in the Active VLAN Tablet. + * However, without the VLANMD bit (9) in PCE_GCTRL_1 (0x457) even + * GSWIP 2.2 and newer hardware maintain the GSWIP 2.1 behavior. + */ gswip_switch_w(priv, idx, GSWIP_PCE_DEFPVID(port)); } From 70535979738937d8397e76f850bd00e97115d02d Mon Sep 17 00:00:00 2001 From: Daniel Golle Date: Tue, 21 Oct 2025 12:16:37 +0100 Subject: [PATCH 0207/1381] net: dsa: lantiq_gswip: convert accessors to use regmap Use regmap for register access in preparation for supporting the MaxLinear GSW1xx family of switches connected via MDIO or SPI. Rewrite the existing accessor read-poll-timeout functions to use calls to the regmap API for now. Signed-off-by: Daniel Golle Reviewed-by: Maxime Chevallier Acked-by; Hauke Mehrtens : Acked-by; Hauke Mehrtens : Link: https://patch.msgid.link/535d968bc6319a74bdf76166ef19364ee659285f.1761045000.git.daniel@makrotopia.org Signed-off-by: Jakub Kicinski --- drivers/net/dsa/lantiq/Kconfig | 1 + drivers/net/dsa/lantiq/lantiq_gswip.c | 107 +++++++++++++++----------- drivers/net/dsa/lantiq/lantiq_gswip.h | 6 +- 3 files changed, 67 insertions(+), 47 deletions(-) diff --git a/drivers/net/dsa/lantiq/Kconfig b/drivers/net/dsa/lantiq/Kconfig index 1cb053c823f7..3cfa16840cf5 100644 --- a/drivers/net/dsa/lantiq/Kconfig +++ b/drivers/net/dsa/lantiq/Kconfig @@ -2,6 +2,7 @@ config NET_DSA_LANTIQ_GSWIP tristate "Lantiq / Intel GSWIP" depends on HAS_IOMEM select NET_DSA_TAG_GSWIP + select REGMAP help This enables support for the Lantiq / Intel GSWIP 2.1 found in the xrx200 / VR9 SoC. diff --git a/drivers/net/dsa/lantiq/lantiq_gswip.c b/drivers/net/dsa/lantiq/lantiq_gswip.c index 86b410a40d32..8a448e1f5eef 100644 --- a/drivers/net/dsa/lantiq/lantiq_gswip.c +++ b/drivers/net/dsa/lantiq/lantiq_gswip.c @@ -113,22 +113,22 @@ static const struct gswip_rmon_cnt_desc gswip_rmon_cnt[] = { static u32 gswip_switch_r(struct gswip_priv *priv, u32 offset) { - return __raw_readl(priv->gswip + (offset * 4)); + u32 val; + + regmap_read(priv->gswip, offset, &val); + + return val; } static void gswip_switch_w(struct gswip_priv *priv, u32 val, u32 offset) { - __raw_writel(val, priv->gswip + (offset * 4)); + regmap_write(priv->gswip, offset, val); } static void gswip_switch_mask(struct gswip_priv *priv, u32 clear, u32 set, u32 offset) { - u32 val = gswip_switch_r(priv, offset); - - val &= ~(clear); - val |= set; - gswip_switch_w(priv, val, offset); + regmap_write_bits(priv->gswip, offset, clear | set, set); } static u32 gswip_switch_r_timeout(struct gswip_priv *priv, u32 offset, @@ -136,48 +136,34 @@ static u32 gswip_switch_r_timeout(struct gswip_priv *priv, u32 offset, { u32 val; - return readx_poll_timeout(__raw_readl, priv->gswip + (offset * 4), val, - (val & cleared) == 0, 20, 50000); + return regmap_read_poll_timeout(priv->gswip, offset, val, + !(val & cleared), 20, 50000); } static u32 gswip_mdio_r(struct gswip_priv *priv, u32 offset) { - return __raw_readl(priv->mdio + (offset * 4)); + u32 val; + + regmap_read(priv->mdio, offset, &val); + + return val; } static void gswip_mdio_w(struct gswip_priv *priv, u32 val, u32 offset) { - __raw_writel(val, priv->mdio + (offset * 4)); + regmap_write(priv->mdio, offset, val); } static void gswip_mdio_mask(struct gswip_priv *priv, u32 clear, u32 set, u32 offset) { - u32 val = gswip_mdio_r(priv, offset); - - val &= ~(clear); - val |= set; - gswip_mdio_w(priv, val, offset); -} - -static u32 gswip_mii_r(struct gswip_priv *priv, u32 offset) -{ - return __raw_readl(priv->mii + (offset * 4)); -} - -static void gswip_mii_w(struct gswip_priv *priv, u32 val, u32 offset) -{ - __raw_writel(val, priv->mii + (offset * 4)); + regmap_write_bits(priv->mdio, offset, clear | set, set); } static void gswip_mii_mask(struct gswip_priv *priv, u32 clear, u32 set, u32 offset) { - u32 val = gswip_mii_r(priv, offset); - - val &= ~(clear); - val |= set; - gswip_mii_w(priv, val, offset); + regmap_write_bits(priv->mii, offset, clear | set, set); } static void gswip_mii_mask_cfg(struct gswip_priv *priv, u32 clear, u32 set, @@ -220,17 +206,10 @@ static void gswip_mii_mask_pcdu(struct gswip_priv *priv, u32 clear, u32 set, static int gswip_mdio_poll(struct gswip_priv *priv) { - int cnt = 100; + u32 ctrl; - while (likely(cnt--)) { - u32 ctrl = gswip_mdio_r(priv, GSWIP_MDIO_CTRL); - - if ((ctrl & GSWIP_MDIO_CTRL_BUSY) == 0) - return 0; - usleep_range(20, 40); - } - - return -ETIMEDOUT; + return regmap_read_poll_timeout(priv->mdio, GSWIP_MDIO_CTRL, ctrl, + !(ctrl & GSWIP_MDIO_CTRL_BUSY), 40, 4000); } static int gswip_mdio_wr(struct mii_bus *bus, int addr, int reg, u16 val) @@ -1893,9 +1872,37 @@ static int gswip_validate_cpu_port(struct dsa_switch *ds) return 0; } +static const struct regmap_config sw_regmap_config = { + .name = "switch", + .reg_bits = 32, + .val_bits = 32, + .reg_shift = REGMAP_UPSHIFT(2), + .val_format_endian = REGMAP_ENDIAN_NATIVE, + .max_register = GSWIP_SDMA_PCTRLp(6), +}; + +static const struct regmap_config mdio_regmap_config = { + .name = "mdio", + .reg_bits = 32, + .val_bits = 32, + .reg_shift = REGMAP_UPSHIFT(2), + .val_format_endian = REGMAP_ENDIAN_NATIVE, + .max_register = GSWIP_MDIO_PHYp(0), +}; + +static const struct regmap_config mii_regmap_config = { + .name = "mii", + .reg_bits = 32, + .val_bits = 32, + .reg_shift = REGMAP_UPSHIFT(2), + .val_format_endian = REGMAP_ENDIAN_NATIVE, + .max_register = GSWIP_MII_CFGp(6), +}; + static int gswip_probe(struct platform_device *pdev) { struct device_node *np, *gphy_fw_np; + __iomem void *gswip, *mdio, *mii; struct device *dev = &pdev->dev; struct gswip_priv *priv; int err; @@ -1906,15 +1913,27 @@ static int gswip_probe(struct platform_device *pdev) if (!priv) return -ENOMEM; - priv->gswip = devm_platform_ioremap_resource(pdev, 0); + gswip = devm_platform_ioremap_resource(pdev, 0); + if (IS_ERR(gswip)) + return PTR_ERR(gswip); + + mdio = devm_platform_ioremap_resource(pdev, 1); + if (IS_ERR(mdio)) + return PTR_ERR(mdio); + + mii = devm_platform_ioremap_resource(pdev, 2); + if (IS_ERR(mii)) + return PTR_ERR(mii); + + priv->gswip = devm_regmap_init_mmio(dev, gswip, &sw_regmap_config); if (IS_ERR(priv->gswip)) return PTR_ERR(priv->gswip); - priv->mdio = devm_platform_ioremap_resource(pdev, 1); + priv->mdio = devm_regmap_init_mmio(dev, mdio, &mdio_regmap_config); if (IS_ERR(priv->mdio)) return PTR_ERR(priv->mdio); - priv->mii = devm_platform_ioremap_resource(pdev, 2); + priv->mii = devm_regmap_init_mmio(dev, mii, &mii_regmap_config); if (IS_ERR(priv->mii)) return PTR_ERR(priv->mii); diff --git a/drivers/net/dsa/lantiq/lantiq_gswip.h b/drivers/net/dsa/lantiq/lantiq_gswip.h index 69c8d2deff2d..24d759e06e15 100644 --- a/drivers/net/dsa/lantiq/lantiq_gswip.h +++ b/drivers/net/dsa/lantiq/lantiq_gswip.h @@ -263,9 +263,9 @@ struct gswip_vlan { }; struct gswip_priv { - __iomem void *gswip; - __iomem void *mdio; - __iomem void *mii; + struct regmap *gswip; + struct regmap *mdio; + struct regmap *mii; const struct gswip_hw_info *hw_info; const struct xway_gphy_match_data *gphy_fw_name_cfg; struct dsa_switch *ds; From 128f5cf40fa5146267e2acfa29ad0bceaef763a3 Mon Sep 17 00:00:00 2001 From: Daniel Golle Date: Tue, 21 Oct 2025 12:16:50 +0100 Subject: [PATCH 0208/1381] net: dsa: lantiq_gswip: convert trivial accessor uses to regmap Use coccinelle semantic patch to convert all trivial uses of the register accessor functions to use the regmap API directly. // Replace gswip_switch_w with regmap_write @@ expression priv, val, offset; @@ - gswip_switch_w(priv, val, offset) + regmap_write(priv->gswip, offset, val) // Replace gswip_mdio_w with regmap_write @@ expression priv, val, offset; @@ - gswip_mdio_w(priv, val, offset) + regmap_write(priv->mdio, offset, val) // Replace gswip_switch_r in simple assignment - only for u32 @@ expression priv, offset; u32 var; @@ - var = gswip_switch_r(priv, offset) + regmap_read(priv->gswip, offset, &var) // Replace gswip_switch_mask with regmap_set_bits when clear is 0 @@ expression priv, set, offset; @@ - gswip_switch_mask(priv, 0, set, offset) + regmap_set_bits(priv->gswip, offset, set) // Replace gswip_mdio_mask with regmap_set_bits when clear is 0 @@ expression priv, set, offset; @@ - gswip_mdio_mask(priv, 0, set, offset) + regmap_set_bits(priv->mdio, offset, set) // Replace gswip_switch_mask with regmap_clear_bits when set is 0 @@ expression priv, clear, offset; @@ - gswip_switch_mask(priv, clear, 0, offset) + regmap_clear_bits(priv->gswip, offset, clear) // Replace gswip_mdio_mask with regmap_clear_bits when set is 0 @@ expression priv, clear, offset; @@ - gswip_mdio_mask(priv, clear, 0, offset) + regmap_clear_bits(priv->mdio, offset, clear) Remove gswip_switch_w() and gswip_mdio_w() functions as they now no longer have any users. Signed-off-by: Daniel Golle Acked-by; Hauke Mehrtens : Acked-by; Hauke Mehrtens : Link: https://patch.msgid.link/48a60f386b1bd487c410b1f5fb25ba50ceddc6f7.1761045000.git.daniel@makrotopia.org Signed-off-by: Jakub Kicinski --- drivers/net/dsa/lantiq/lantiq_gswip.c | 160 ++++++++++++-------------- 1 file changed, 76 insertions(+), 84 deletions(-) diff --git a/drivers/net/dsa/lantiq/lantiq_gswip.c b/drivers/net/dsa/lantiq/lantiq_gswip.c index 8a448e1f5eef..e58320eaf9da 100644 --- a/drivers/net/dsa/lantiq/lantiq_gswip.c +++ b/drivers/net/dsa/lantiq/lantiq_gswip.c @@ -120,11 +120,6 @@ static u32 gswip_switch_r(struct gswip_priv *priv, u32 offset) return val; } -static void gswip_switch_w(struct gswip_priv *priv, u32 val, u32 offset) -{ - regmap_write(priv->gswip, offset, val); -} - static void gswip_switch_mask(struct gswip_priv *priv, u32 clear, u32 set, u32 offset) { @@ -149,11 +144,6 @@ static u32 gswip_mdio_r(struct gswip_priv *priv, u32 offset) return val; } -static void gswip_mdio_w(struct gswip_priv *priv, u32 val, u32 offset) -{ - regmap_write(priv->mdio, offset, val); -} - static void gswip_mdio_mask(struct gswip_priv *priv, u32 clear, u32 set, u32 offset) { @@ -223,11 +213,11 @@ static int gswip_mdio_wr(struct mii_bus *bus, int addr, int reg, u16 val) return err; } - gswip_mdio_w(priv, val, GSWIP_MDIO_WRITE); - gswip_mdio_w(priv, GSWIP_MDIO_CTRL_BUSY | GSWIP_MDIO_CTRL_WR | - ((addr & GSWIP_MDIO_CTRL_PHYAD_MASK) << GSWIP_MDIO_CTRL_PHYAD_SHIFT) | - (reg & GSWIP_MDIO_CTRL_REGAD_MASK), - GSWIP_MDIO_CTRL); + regmap_write(priv->mdio, GSWIP_MDIO_WRITE, val); + regmap_write(priv->mdio, GSWIP_MDIO_CTRL, + GSWIP_MDIO_CTRL_BUSY | GSWIP_MDIO_CTRL_WR | + ((addr & GSWIP_MDIO_CTRL_PHYAD_MASK) << GSWIP_MDIO_CTRL_PHYAD_SHIFT) | + (reg & GSWIP_MDIO_CTRL_REGAD_MASK)); return 0; } @@ -243,10 +233,10 @@ static int gswip_mdio_rd(struct mii_bus *bus, int addr, int reg) return err; } - gswip_mdio_w(priv, GSWIP_MDIO_CTRL_BUSY | GSWIP_MDIO_CTRL_RD | - ((addr & GSWIP_MDIO_CTRL_PHYAD_MASK) << GSWIP_MDIO_CTRL_PHYAD_SHIFT) | - (reg & GSWIP_MDIO_CTRL_REGAD_MASK), - GSWIP_MDIO_CTRL); + regmap_write(priv->mdio, GSWIP_MDIO_CTRL, + GSWIP_MDIO_CTRL_BUSY | GSWIP_MDIO_CTRL_RD | + ((addr & GSWIP_MDIO_CTRL_PHYAD_MASK) << GSWIP_MDIO_CTRL_PHYAD_SHIFT) | + (reg & GSWIP_MDIO_CTRL_REGAD_MASK)); err = gswip_mdio_poll(priv); if (err) { @@ -310,7 +300,7 @@ static int gswip_pce_table_entry_read(struct gswip_priv *priv, return err; } - gswip_switch_w(priv, tbl->index, GSWIP_PCE_TBL_ADDR); + regmap_write(priv->gswip, GSWIP_PCE_TBL_ADDR, tbl->index); gswip_switch_mask(priv, GSWIP_PCE_TBL_CTRL_ADDR_MASK | GSWIP_PCE_TBL_CTRL_OPMOD_MASK, tbl->table | addr_mode | GSWIP_PCE_TBL_CTRL_BAS, @@ -360,24 +350,24 @@ static int gswip_pce_table_entry_write(struct gswip_priv *priv, return err; } - gswip_switch_w(priv, tbl->index, GSWIP_PCE_TBL_ADDR); + regmap_write(priv->gswip, GSWIP_PCE_TBL_ADDR, tbl->index); gswip_switch_mask(priv, GSWIP_PCE_TBL_CTRL_ADDR_MASK | GSWIP_PCE_TBL_CTRL_OPMOD_MASK, tbl->table | addr_mode, GSWIP_PCE_TBL_CTRL); for (i = 0; i < ARRAY_SIZE(tbl->key); i++) - gswip_switch_w(priv, tbl->key[i], GSWIP_PCE_TBL_KEY(i)); + regmap_write(priv->gswip, GSWIP_PCE_TBL_KEY(i), tbl->key[i]); for (i = 0; i < ARRAY_SIZE(tbl->val); i++) - gswip_switch_w(priv, tbl->val[i], GSWIP_PCE_TBL_VAL(i)); + regmap_write(priv->gswip, GSWIP_PCE_TBL_VAL(i), tbl->val[i]); gswip_switch_mask(priv, GSWIP_PCE_TBL_CTRL_ADDR_MASK | GSWIP_PCE_TBL_CTRL_OPMOD_MASK, tbl->table | addr_mode, GSWIP_PCE_TBL_CTRL); - gswip_switch_w(priv, tbl->mask, GSWIP_PCE_TBL_MASK); + regmap_write(priv->gswip, GSWIP_PCE_TBL_MASK, tbl->mask); crtl = gswip_switch_r(priv, GSWIP_PCE_TBL_CTRL); crtl &= ~(GSWIP_PCE_TBL_CTRL_TYPE | GSWIP_PCE_TBL_CTRL_VLD | @@ -388,7 +378,7 @@ static int gswip_pce_table_entry_write(struct gswip_priv *priv, crtl |= GSWIP_PCE_TBL_CTRL_VLD; crtl |= (tbl->gmap << 7) & GSWIP_PCE_TBL_CTRL_GMAP_MASK; crtl |= GSWIP_PCE_TBL_CTRL_BAS; - gswip_switch_w(priv, crtl, GSWIP_PCE_TBL_CTRL); + regmap_write(priv->gswip, GSWIP_PCE_TBL_CTRL, crtl); err = gswip_switch_r_timeout(priv, GSWIP_PCE_TBL_CTRL, GSWIP_PCE_TBL_CTRL_BAS); @@ -467,14 +457,13 @@ static int gswip_port_enable(struct dsa_switch *ds, int port, } /* RMON Counter Enable for port */ - gswip_switch_w(priv, GSWIP_BM_PCFG_CNTEN, GSWIP_BM_PCFGp(port)); + regmap_write(priv->gswip, GSWIP_BM_PCFGp(port), GSWIP_BM_PCFG_CNTEN); /* enable port fetch/store dma & VLAN Modification */ - gswip_switch_mask(priv, 0, GSWIP_FDMA_PCTRL_EN | - GSWIP_FDMA_PCTRL_VLANMOD_BOTH, - GSWIP_FDMA_PCTRLp(port)); - gswip_switch_mask(priv, 0, GSWIP_SDMA_PCTRL_EN, - GSWIP_SDMA_PCTRLp(port)); + regmap_set_bits(priv->gswip, GSWIP_FDMA_PCTRLp(port), + GSWIP_FDMA_PCTRL_EN | GSWIP_FDMA_PCTRL_VLANMOD_BOTH); + regmap_set_bits(priv->gswip, GSWIP_SDMA_PCTRLp(port), + GSWIP_SDMA_PCTRL_EN); return 0; } @@ -483,10 +472,10 @@ static void gswip_port_disable(struct dsa_switch *ds, int port) { struct gswip_priv *priv = ds->priv; - gswip_switch_mask(priv, GSWIP_FDMA_PCTRL_EN, 0, - GSWIP_FDMA_PCTRLp(port)); - gswip_switch_mask(priv, GSWIP_SDMA_PCTRL_EN, 0, - GSWIP_SDMA_PCTRLp(port)); + regmap_clear_bits(priv->gswip, GSWIP_FDMA_PCTRLp(port), + GSWIP_FDMA_PCTRL_EN); + regmap_clear_bits(priv->gswip, GSWIP_SDMA_PCTRLp(port), + GSWIP_SDMA_PCTRL_EN); } static int gswip_pce_load_microcode(struct gswip_priv *priv) @@ -497,22 +486,22 @@ static int gswip_pce_load_microcode(struct gswip_priv *priv) gswip_switch_mask(priv, GSWIP_PCE_TBL_CTRL_ADDR_MASK | GSWIP_PCE_TBL_CTRL_OPMOD_MASK, GSWIP_PCE_TBL_CTRL_OPMOD_ADWR, GSWIP_PCE_TBL_CTRL); - gswip_switch_w(priv, 0, GSWIP_PCE_TBL_MASK); + regmap_write(priv->gswip, GSWIP_PCE_TBL_MASK, 0); for (i = 0; i < priv->hw_info->pce_microcode_size; i++) { - gswip_switch_w(priv, i, GSWIP_PCE_TBL_ADDR); - gswip_switch_w(priv, (*priv->hw_info->pce_microcode)[i].val_0, - GSWIP_PCE_TBL_VAL(0)); - gswip_switch_w(priv, (*priv->hw_info->pce_microcode)[i].val_1, - GSWIP_PCE_TBL_VAL(1)); - gswip_switch_w(priv, (*priv->hw_info->pce_microcode)[i].val_2, - GSWIP_PCE_TBL_VAL(2)); - gswip_switch_w(priv, (*priv->hw_info->pce_microcode)[i].val_3, - GSWIP_PCE_TBL_VAL(3)); + regmap_write(priv->gswip, GSWIP_PCE_TBL_ADDR, i); + regmap_write(priv->gswip, GSWIP_PCE_TBL_VAL(0), + (*priv->hw_info->pce_microcode)[i].val_0); + regmap_write(priv->gswip, GSWIP_PCE_TBL_VAL(1), + (*priv->hw_info->pce_microcode)[i].val_1); + regmap_write(priv->gswip, GSWIP_PCE_TBL_VAL(2), + (*priv->hw_info->pce_microcode)[i].val_2); + regmap_write(priv->gswip, GSWIP_PCE_TBL_VAL(3), + (*priv->hw_info->pce_microcode)[i].val_3); /* start the table access: */ - gswip_switch_mask(priv, 0, GSWIP_PCE_TBL_CTRL_BAS, - GSWIP_PCE_TBL_CTRL); + regmap_set_bits(priv->gswip, GSWIP_PCE_TBL_CTRL, + GSWIP_PCE_TBL_CTRL_BAS); err = gswip_switch_r_timeout(priv, GSWIP_PCE_TBL_CTRL, GSWIP_PCE_TBL_CTRL_BAS); if (err) @@ -520,8 +509,8 @@ static int gswip_pce_load_microcode(struct gswip_priv *priv) } /* tell the switch that the microcode is loaded */ - gswip_switch_mask(priv, 0, GSWIP_PCE_GCTRL_0_MC_VALID, - GSWIP_PCE_GCTRL_0); + regmap_set_bits(priv->gswip, GSWIP_PCE_GCTRL_0, + GSWIP_PCE_GCTRL_0_MC_VALID); return 0; } @@ -572,7 +561,7 @@ static void gswip_port_commit_pvid(struct gswip_priv *priv, int port) * However, without the VLANMD bit (9) in PCE_GCTRL_1 (0x457) even * GSWIP 2.2 and newer hardware maintain the GSWIP 2.1 behavior. */ - gswip_switch_w(priv, idx, GSWIP_PCE_DEFPVID(port)); + regmap_write(priv->gswip, GSWIP_PCE_DEFPVID(port), idx); } static int gswip_port_vlan_filtering(struct dsa_switch *ds, int port, @@ -588,8 +577,8 @@ static int gswip_port_vlan_filtering(struct dsa_switch *ds, int port, GSWIP_PCE_VCTRL_UVR | GSWIP_PCE_VCTRL_VIMR | GSWIP_PCE_VCTRL_VEMR | GSWIP_PCE_VCTRL_VID0, GSWIP_PCE_VCTRL(port)); - gswip_switch_mask(priv, GSWIP_PCE_PCTRL_0_TVM, 0, - GSWIP_PCE_PCTRL_0p(port)); + regmap_clear_bits(priv->gswip, GSWIP_PCE_PCTRL_0p(port), + GSWIP_PCE_PCTRL_0_TVM); } else { /* Use port based VLAN */ gswip_switch_mask(priv, @@ -597,8 +586,8 @@ static int gswip_port_vlan_filtering(struct dsa_switch *ds, int port, GSWIP_PCE_VCTRL_VEMR | GSWIP_PCE_VCTRL_VID0, GSWIP_PCE_VCTRL_VSR, GSWIP_PCE_VCTRL(port)); - gswip_switch_mask(priv, 0, GSWIP_PCE_PCTRL_0_TVM, - GSWIP_PCE_PCTRL_0p(port)); + regmap_set_bits(priv->gswip, GSWIP_PCE_PCTRL_0p(port), + GSWIP_PCE_PCTRL_0_TVM); } gswip_port_commit_pvid(priv, port); @@ -613,9 +602,9 @@ static int gswip_setup(struct dsa_switch *ds) struct dsa_port *cpu_dp; int err, i; - gswip_switch_w(priv, GSWIP_SWRES_R0, GSWIP_SWRES); + regmap_write(priv->gswip, GSWIP_SWRES, GSWIP_SWRES_R0); usleep_range(5000, 10000); - gswip_switch_w(priv, 0, GSWIP_SWRES); + regmap_write(priv->gswip, GSWIP_SWRES, 0); /* disable port fetch/store dma on all ports */ for (i = 0; i < priv->hw_info->max_ports; i++) { @@ -624,7 +613,7 @@ static int gswip_setup(struct dsa_switch *ds) } /* enable Switch */ - gswip_mdio_mask(priv, 0, GSWIP_MDIO_GLOB_ENABLE, GSWIP_MDIO_GLOB); + regmap_set_bits(priv->mdio, GSWIP_MDIO_GLOB, GSWIP_MDIO_GLOB_ENABLE); err = gswip_pce_load_microcode(priv); if (err) { @@ -633,9 +622,9 @@ static int gswip_setup(struct dsa_switch *ds) } /* Default unknown Broadcast/Multicast/Unicast port maps */ - gswip_switch_w(priv, cpu_ports, GSWIP_PCE_PMAP1); - gswip_switch_w(priv, cpu_ports, GSWIP_PCE_PMAP2); - gswip_switch_w(priv, cpu_ports, GSWIP_PCE_PMAP3); + regmap_write(priv->gswip, GSWIP_PCE_PMAP1, cpu_ports); + regmap_write(priv->gswip, GSWIP_PCE_PMAP2, cpu_ports); + regmap_write(priv->gswip, GSWIP_PCE_PMAP3, cpu_ports); /* Deactivate MDIO PHY auto polling. Some PHYs as the AR8030 have an * interoperability problem with this auto polling mechanism because @@ -653,7 +642,7 @@ static int gswip_setup(struct dsa_switch *ds) * Testing shows that when PHY auto polling is disabled these problems * go away. */ - gswip_mdio_w(priv, 0x0, GSWIP_MDIO_MDC_CFG0); + regmap_write(priv->mdio, GSWIP_MDIO_MDC_CFG0, 0x0); /* Configure the MDIO Clock 2.5 MHz */ gswip_mdio_mask(priv, 0xff, 0x09, GSWIP_MDIO_MDC_CFG1); @@ -673,22 +662,25 @@ static int gswip_setup(struct dsa_switch *ds) dsa_switch_for_each_cpu_port(cpu_dp, ds) { /* enable special tag insertion on cpu port */ - gswip_switch_mask(priv, 0, GSWIP_FDMA_PCTRL_STEN, - GSWIP_FDMA_PCTRLp(cpu_dp->index)); + regmap_set_bits(priv->gswip, GSWIP_FDMA_PCTRLp(cpu_dp->index), + GSWIP_FDMA_PCTRL_STEN); /* accept special tag in ingress direction */ - gswip_switch_mask(priv, 0, GSWIP_PCE_PCTRL_0_INGRESS, - GSWIP_PCE_PCTRL_0p(cpu_dp->index)); + regmap_set_bits(priv->gswip, + GSWIP_PCE_PCTRL_0p(cpu_dp->index), + GSWIP_PCE_PCTRL_0_INGRESS); } - gswip_switch_mask(priv, 0, GSWIP_BM_QUEUE_GCTRL_GL_MOD, - GSWIP_BM_QUEUE_GCTRL); + regmap_set_bits(priv->gswip, GSWIP_BM_QUEUE_GCTRL, + GSWIP_BM_QUEUE_GCTRL_GL_MOD); /* VLAN aware Switching */ - gswip_switch_mask(priv, 0, GSWIP_PCE_GCTRL_0_VLAN, GSWIP_PCE_GCTRL_0); + regmap_set_bits(priv->gswip, GSWIP_PCE_GCTRL_0, + GSWIP_PCE_GCTRL_0_VLAN); /* Flush MAC Table */ - gswip_switch_mask(priv, 0, GSWIP_PCE_GCTRL_0_MTFL, GSWIP_PCE_GCTRL_0); + regmap_set_bits(priv->gswip, GSWIP_PCE_GCTRL_0, + GSWIP_PCE_GCTRL_0_MTFL); err = gswip_switch_r_timeout(priv, GSWIP_PCE_GCTRL_0, GSWIP_PCE_GCTRL_0_MTFL); @@ -1074,8 +1066,8 @@ static void gswip_port_stp_state_set(struct dsa_switch *ds, int port, u8 state) switch (state) { case BR_STATE_DISABLED: - gswip_switch_mask(priv, GSWIP_SDMA_PCTRL_EN, 0, - GSWIP_SDMA_PCTRLp(port)); + regmap_clear_bits(priv->gswip, GSWIP_SDMA_PCTRLp(port), + GSWIP_SDMA_PCTRL_EN); return; case BR_STATE_BLOCKING: case BR_STATE_LISTENING: @@ -1092,8 +1084,8 @@ static void gswip_port_stp_state_set(struct dsa_switch *ds, int port, u8 state) return; } - gswip_switch_mask(priv, 0, GSWIP_SDMA_PCTRL_EN, - GSWIP_SDMA_PCTRLp(port)); + regmap_set_bits(priv->gswip, GSWIP_SDMA_PCTRLp(port), + GSWIP_SDMA_PCTRL_EN); gswip_switch_mask(priv, GSWIP_PCE_PCTRL_0_PSTATE_MASK, stp_state, GSWIP_PCE_PCTRL_0p(port)); } @@ -1222,19 +1214,19 @@ static int gswip_port_change_mtu(struct dsa_switch *ds, int port, int new_mtu) */ if (dsa_is_cpu_port(ds, port)) { new_mtu += 8; - gswip_switch_w(priv, VLAN_ETH_HLEN + new_mtu + ETH_FCS_LEN, - GSWIP_MAC_FLEN); + regmap_write(priv->gswip, GSWIP_MAC_FLEN, + VLAN_ETH_HLEN + new_mtu + ETH_FCS_LEN); } /* Enable MLEN for ports with non-standard MTUs, including the special * header on the CPU port added above. */ if (new_mtu != ETH_DATA_LEN) - gswip_switch_mask(priv, 0, GSWIP_MAC_CTRL_2_MLEN, - GSWIP_MAC_CTRL_2p(port)); + regmap_set_bits(priv->gswip, GSWIP_MAC_CTRL_2p(port), + GSWIP_MAC_CTRL_2_MLEN); else - gswip_switch_mask(priv, GSWIP_MAC_CTRL_2_MLEN, 0, - GSWIP_MAC_CTRL_2p(port)); + regmap_clear_bits(priv->gswip, GSWIP_MAC_CTRL_2p(port), + GSWIP_MAC_CTRL_2_MLEN); return 0; } @@ -1536,7 +1528,7 @@ static u32 gswip_bcm_ram_entry_read(struct gswip_priv *priv, u32 table, u32 result; int err; - gswip_switch_w(priv, index, GSWIP_BM_RAM_ADDR); + regmap_write(priv->gswip, GSWIP_BM_RAM_ADDR, index); gswip_switch_mask(priv, GSWIP_BM_RAM_CTRL_ADDR_MASK | GSWIP_BM_RAM_CTRL_OPMOD, table | GSWIP_BM_RAM_CTRL_BAS, @@ -1550,7 +1542,7 @@ static u32 gswip_bcm_ram_entry_read(struct gswip_priv *priv, u32 table, return 0; } - result = gswip_switch_r(priv, GSWIP_BM_RAM_VAL(0)); + regmap_read(priv->gswip, GSWIP_BM_RAM_VAL(0), &result); result |= gswip_switch_r(priv, GSWIP_BM_RAM_VAL(1)) << 16; return result; @@ -1952,7 +1944,7 @@ static int gswip_probe(struct platform_device *pdev) priv->ds->phylink_mac_ops = &gswip_phylink_mac_ops; priv->dev = dev; mutex_init(&priv->pce_table_lock); - version = gswip_switch_r(priv, GSWIP_VERSION); + regmap_read(priv->gswip, GSWIP_VERSION, &version); /* The hardware has the 'major/minor' version bytes in the wrong order * preventing numerical comparisons. Construct a 16-bit unsigned integer @@ -2009,7 +2001,7 @@ static int gswip_probe(struct platform_device *pdev) return 0; disable_switch: - gswip_mdio_mask(priv, GSWIP_MDIO_GLOB_ENABLE, 0, GSWIP_MDIO_GLOB); + regmap_clear_bits(priv->mdio, GSWIP_MDIO_GLOB, GSWIP_MDIO_GLOB_ENABLE); dsa_unregister_switch(priv->ds); gphy_fw_remove: for (i = 0; i < priv->num_gphy_fw; i++) @@ -2026,7 +2018,7 @@ static void gswip_remove(struct platform_device *pdev) return; /* disable the switch */ - gswip_mdio_mask(priv, GSWIP_MDIO_GLOB_ENABLE, 0, GSWIP_MDIO_GLOB); + regmap_clear_bits(priv->mdio, GSWIP_MDIO_GLOB, GSWIP_MDIO_GLOB_ENABLE); dsa_unregister_switch(priv->ds); From 4cc06901ef34be259e8af850bfc3c3a17178486a Mon Sep 17 00:00:00 2001 From: Daniel Golle Date: Tue, 21 Oct 2025 12:16:58 +0100 Subject: [PATCH 0209/1381] net: dsa: lantiq_gswip: manually convert remaining uses of read accessors Manually convert the remaining uses of the read accessor functions and remove them now that they are unused. Signed-off-by: Daniel Golle Acked-by; Hauke Mehrtens : Acked-by; Hauke Mehrtens : Link: https://patch.msgid.link/0e2a44b83131b40fc1ee558ed1f536c26e1232ba.1761045000.git.daniel@makrotopia.org Signed-off-by: Jakub Kicinski --- drivers/net/dsa/lantiq/lantiq_gswip.c | 78 +++++++++++++-------------- 1 file changed, 38 insertions(+), 40 deletions(-) diff --git a/drivers/net/dsa/lantiq/lantiq_gswip.c b/drivers/net/dsa/lantiq/lantiq_gswip.c index e58320eaf9da..46fdc9d9c2c9 100644 --- a/drivers/net/dsa/lantiq/lantiq_gswip.c +++ b/drivers/net/dsa/lantiq/lantiq_gswip.c @@ -111,15 +111,6 @@ static const struct gswip_rmon_cnt_desc gswip_rmon_cnt[] = { MIB_DESC(2, 0x0E, "TxGoodBytes"), }; -static u32 gswip_switch_r(struct gswip_priv *priv, u32 offset) -{ - u32 val; - - regmap_read(priv->gswip, offset, &val); - - return val; -} - static void gswip_switch_mask(struct gswip_priv *priv, u32 clear, u32 set, u32 offset) { @@ -135,15 +126,6 @@ static u32 gswip_switch_r_timeout(struct gswip_priv *priv, u32 offset, !(val & cleared), 20, 50000); } -static u32 gswip_mdio_r(struct gswip_priv *priv, u32 offset) -{ - u32 val; - - regmap_read(priv->mdio, offset, &val); - - return val; -} - static void gswip_mdio_mask(struct gswip_priv *priv, u32 clear, u32 set, u32 offset) { @@ -225,6 +207,7 @@ static int gswip_mdio_wr(struct mii_bus *bus, int addr, int reg, u16 val) static int gswip_mdio_rd(struct mii_bus *bus, int addr, int reg) { struct gswip_priv *priv = bus->priv; + u32 val; int err; err = gswip_mdio_poll(priv); @@ -244,7 +227,11 @@ static int gswip_mdio_rd(struct mii_bus *bus, int addr, int reg) return err; } - return gswip_mdio_r(priv, GSWIP_MDIO_READ); + err = regmap_read(priv->mdio, GSWIP_MDIO_READ, &val); + if (err) + return err; + + return val; } static int gswip_mdio(struct gswip_priv *priv) @@ -287,7 +274,8 @@ static int gswip_pce_table_entry_read(struct gswip_priv *priv, { int i; int err; - u16 crtl; + u32 crtl; + u32 tmp; u16 addr_mode = tbl->key_mode ? GSWIP_PCE_TBL_CTRL_OPMOD_KSRD : GSWIP_PCE_TBL_CTRL_OPMOD_ADRD; @@ -295,10 +283,8 @@ static int gswip_pce_table_entry_read(struct gswip_priv *priv, err = gswip_switch_r_timeout(priv, GSWIP_PCE_TBL_CTRL, GSWIP_PCE_TBL_CTRL_BAS); - if (err) { - mutex_unlock(&priv->pce_table_lock); - return err; - } + if (err) + goto out_unlock; regmap_write(priv->gswip, GSWIP_PCE_TBL_ADDR, tbl->index); gswip_switch_mask(priv, GSWIP_PCE_TBL_CTRL_ADDR_MASK | @@ -308,28 +294,39 @@ static int gswip_pce_table_entry_read(struct gswip_priv *priv, err = gswip_switch_r_timeout(priv, GSWIP_PCE_TBL_CTRL, GSWIP_PCE_TBL_CTRL_BAS); - if (err) { - mutex_unlock(&priv->pce_table_lock); - return err; + if (err) + goto out_unlock; + + for (i = 0; i < ARRAY_SIZE(tbl->key); i++) { + err = regmap_read(priv->gswip, GSWIP_PCE_TBL_KEY(i), &tmp); + if (err) + goto out_unlock; + tbl->key[i] = tmp; + } + for (i = 0; i < ARRAY_SIZE(tbl->val); i++) { + err = regmap_read(priv->gswip, GSWIP_PCE_TBL_VAL(i), &tmp); + if (err) + goto out_unlock; + tbl->val[i] = tmp; } - for (i = 0; i < ARRAY_SIZE(tbl->key); i++) - tbl->key[i] = gswip_switch_r(priv, GSWIP_PCE_TBL_KEY(i)); + err = regmap_read(priv->gswip, GSWIP_PCE_TBL_MASK, &tmp); + if (err) + goto out_unlock; - for (i = 0; i < ARRAY_SIZE(tbl->val); i++) - tbl->val[i] = gswip_switch_r(priv, GSWIP_PCE_TBL_VAL(i)); - - tbl->mask = gswip_switch_r(priv, GSWIP_PCE_TBL_MASK); - - crtl = gswip_switch_r(priv, GSWIP_PCE_TBL_CTRL); + tbl->mask = tmp; + err = regmap_read(priv->gswip, GSWIP_PCE_TBL_CTRL, &crtl); + if (err) + goto out_unlock; tbl->type = !!(crtl & GSWIP_PCE_TBL_CTRL_TYPE); tbl->valid = !!(crtl & GSWIP_PCE_TBL_CTRL_VLD); tbl->gmap = (crtl & GSWIP_PCE_TBL_CTRL_GMAP_MASK) >> 7; +out_unlock: mutex_unlock(&priv->pce_table_lock); - return 0; + return err; } static int gswip_pce_table_entry_write(struct gswip_priv *priv, @@ -337,7 +334,7 @@ static int gswip_pce_table_entry_write(struct gswip_priv *priv, { int i; int err; - u16 crtl; + u32 crtl; u16 addr_mode = tbl->key_mode ? GSWIP_PCE_TBL_CTRL_OPMOD_KSWR : GSWIP_PCE_TBL_CTRL_OPMOD_ADWR; @@ -369,7 +366,7 @@ static int gswip_pce_table_entry_write(struct gswip_priv *priv, regmap_write(priv->gswip, GSWIP_PCE_TBL_MASK, tbl->mask); - crtl = gswip_switch_r(priv, GSWIP_PCE_TBL_CTRL); + regmap_read(priv->gswip, GSWIP_PCE_TBL_CTRL, &crtl); crtl &= ~(GSWIP_PCE_TBL_CTRL_TYPE | GSWIP_PCE_TBL_CTRL_VLD | GSWIP_PCE_TBL_CTRL_GMAP_MASK); if (tbl->type) @@ -1525,7 +1522,7 @@ static void gswip_get_strings(struct dsa_switch *ds, int port, u32 stringset, static u32 gswip_bcm_ram_entry_read(struct gswip_priv *priv, u32 table, u32 index) { - u32 result; + u32 result, val; int err; regmap_write(priv->gswip, GSWIP_BM_RAM_ADDR, index); @@ -1543,7 +1540,8 @@ static u32 gswip_bcm_ram_entry_read(struct gswip_priv *priv, u32 table, } regmap_read(priv->gswip, GSWIP_BM_RAM_VAL(0), &result); - result |= gswip_switch_r(priv, GSWIP_BM_RAM_VAL(1)) << 16; + regmap_read(priv->gswip, GSWIP_BM_RAM_VAL(1), &val); + result |= val << 16; return result; } From 748b0aebd48f77f147418a6280a3792a487adcc0 Mon Sep 17 00:00:00 2001 From: Daniel Golle Date: Tue, 21 Oct 2025 12:17:06 +0100 Subject: [PATCH 0210/1381] net: dsa: lantiq_gswip: replace *_mask() functions with regmap API Use coccinelle to replace all uses of *_mask() with an equivalent call to regmap_write_bits(). // Replace gswip_switch_mask with regmap_write_bits @@ expression priv, clear, set, offset; @@ - gswip_switch_mask(priv, clear, set, offset) + regmap_write_bits(priv->gswip, offset, clear | set, set) // Replace gswip_mdio_mask with regmap_write_bits @@ expression priv, clear, set, offset; @@ - gswip_mdio_mask(priv, clear, set, offset) + regmap_write_bits(priv->mdio, offset, clear | set, set) // Replace gswip_mii_mask with regmap_write_bits @@ expression priv, clear, set, offset; @@ - gswip_mii_mask(priv, clear, set, offset) + regmap_write_bits(priv->mii, offset, clear | set, set) Remove the new unused *_mask() functions. This naive approach will be further optmized manually in the next commit. Signed-off-by: Daniel Golle Acked-by; Hauke Mehrtens : Acked-by; Hauke Mehrtens : Link: https://patch.msgid.link/258d931386a512b7089924c70073ca7acba71168.1761045000.git.daniel@makrotopia.org Signed-off-by: Jakub Kicinski --- drivers/net/dsa/lantiq/lantiq_gswip.c | 140 +++++++++++++------------- 1 file changed, 70 insertions(+), 70 deletions(-) diff --git a/drivers/net/dsa/lantiq/lantiq_gswip.c b/drivers/net/dsa/lantiq/lantiq_gswip.c index 46fdc9d9c2c9..71dfddd62d9f 100644 --- a/drivers/net/dsa/lantiq/lantiq_gswip.c +++ b/drivers/net/dsa/lantiq/lantiq_gswip.c @@ -111,12 +111,6 @@ static const struct gswip_rmon_cnt_desc gswip_rmon_cnt[] = { MIB_DESC(2, 0x0E, "TxGoodBytes"), }; -static void gswip_switch_mask(struct gswip_priv *priv, u32 clear, u32 set, - u32 offset) -{ - regmap_write_bits(priv->gswip, offset, clear | set, set); -} - static u32 gswip_switch_r_timeout(struct gswip_priv *priv, u32 offset, u32 cleared) { @@ -126,18 +120,6 @@ static u32 gswip_switch_r_timeout(struct gswip_priv *priv, u32 offset, !(val & cleared), 20, 50000); } -static void gswip_mdio_mask(struct gswip_priv *priv, u32 clear, u32 set, - u32 offset) -{ - regmap_write_bits(priv->mdio, offset, clear | set, set); -} - -static void gswip_mii_mask(struct gswip_priv *priv, u32 clear, u32 set, - u32 offset) -{ - regmap_write_bits(priv->mii, offset, clear | set, set); -} - static void gswip_mii_mask_cfg(struct gswip_priv *priv, u32 clear, u32 set, int port) { @@ -149,7 +131,8 @@ static void gswip_mii_mask_cfg(struct gswip_priv *priv, u32 clear, u32 set, reg_port = port + priv->hw_info->mii_port_reg_offset; - gswip_mii_mask(priv, clear, set, GSWIP_MII_CFGp(reg_port)); + regmap_write_bits(priv->mii, GSWIP_MII_CFGp(reg_port), clear | set, + set); } static void gswip_mii_mask_pcdu(struct gswip_priv *priv, u32 clear, u32 set, @@ -165,13 +148,16 @@ static void gswip_mii_mask_pcdu(struct gswip_priv *priv, u32 clear, u32 set, switch (reg_port) { case 0: - gswip_mii_mask(priv, clear, set, GSWIP_MII_PCDU0); + regmap_write_bits(priv->mii, GSWIP_MII_PCDU0, clear | set, + set); break; case 1: - gswip_mii_mask(priv, clear, set, GSWIP_MII_PCDU1); + regmap_write_bits(priv->mii, GSWIP_MII_PCDU1, clear | set, + set); break; case 5: - gswip_mii_mask(priv, clear, set, GSWIP_MII_PCDU5); + regmap_write_bits(priv->mii, GSWIP_MII_PCDU5, clear | set, + set); break; } } @@ -287,10 +273,11 @@ static int gswip_pce_table_entry_read(struct gswip_priv *priv, goto out_unlock; regmap_write(priv->gswip, GSWIP_PCE_TBL_ADDR, tbl->index); - gswip_switch_mask(priv, GSWIP_PCE_TBL_CTRL_ADDR_MASK | - GSWIP_PCE_TBL_CTRL_OPMOD_MASK, + regmap_write_bits(priv->gswip, GSWIP_PCE_TBL_CTRL, + GSWIP_PCE_TBL_CTRL_ADDR_MASK | + GSWIP_PCE_TBL_CTRL_OPMOD_MASK | tbl->table | addr_mode | GSWIP_PCE_TBL_CTRL_BAS, - GSWIP_PCE_TBL_CTRL); + tbl->table | addr_mode | GSWIP_PCE_TBL_CTRL_BAS); err = gswip_switch_r_timeout(priv, GSWIP_PCE_TBL_CTRL, GSWIP_PCE_TBL_CTRL_BAS); @@ -348,10 +335,11 @@ static int gswip_pce_table_entry_write(struct gswip_priv *priv, } regmap_write(priv->gswip, GSWIP_PCE_TBL_ADDR, tbl->index); - gswip_switch_mask(priv, GSWIP_PCE_TBL_CTRL_ADDR_MASK | - GSWIP_PCE_TBL_CTRL_OPMOD_MASK, + regmap_write_bits(priv->gswip, GSWIP_PCE_TBL_CTRL, + GSWIP_PCE_TBL_CTRL_ADDR_MASK | + GSWIP_PCE_TBL_CTRL_OPMOD_MASK | tbl->table | addr_mode, - GSWIP_PCE_TBL_CTRL); + tbl->table | addr_mode); for (i = 0; i < ARRAY_SIZE(tbl->key); i++) regmap_write(priv->gswip, GSWIP_PCE_TBL_KEY(i), tbl->key[i]); @@ -359,10 +347,11 @@ static int gswip_pce_table_entry_write(struct gswip_priv *priv, for (i = 0; i < ARRAY_SIZE(tbl->val); i++) regmap_write(priv->gswip, GSWIP_PCE_TBL_VAL(i), tbl->val[i]); - gswip_switch_mask(priv, GSWIP_PCE_TBL_CTRL_ADDR_MASK | - GSWIP_PCE_TBL_CTRL_OPMOD_MASK, + regmap_write_bits(priv->gswip, GSWIP_PCE_TBL_CTRL, + GSWIP_PCE_TBL_CTRL_ADDR_MASK | + GSWIP_PCE_TBL_CTRL_OPMOD_MASK | tbl->table | addr_mode, - GSWIP_PCE_TBL_CTRL); + tbl->table | addr_mode); regmap_write(priv->gswip, GSWIP_PCE_TBL_MASK, tbl->mask); @@ -449,8 +438,9 @@ static int gswip_port_enable(struct dsa_switch *ds, int port, if (phydev) mdio_phy = phydev->mdio.addr & GSWIP_MDIO_PHY_ADDR_MASK; - gswip_mdio_mask(priv, GSWIP_MDIO_PHY_ADDR_MASK, mdio_phy, - GSWIP_MDIO_PHYp(port)); + regmap_write_bits(priv->mdio, GSWIP_MDIO_PHYp(port), + GSWIP_MDIO_PHY_ADDR_MASK | mdio_phy, + mdio_phy); } /* RMON Counter Enable for port */ @@ -480,9 +470,11 @@ static int gswip_pce_load_microcode(struct gswip_priv *priv) int i; int err; - gswip_switch_mask(priv, GSWIP_PCE_TBL_CTRL_ADDR_MASK | - GSWIP_PCE_TBL_CTRL_OPMOD_MASK, - GSWIP_PCE_TBL_CTRL_OPMOD_ADWR, GSWIP_PCE_TBL_CTRL); + regmap_write_bits(priv->gswip, GSWIP_PCE_TBL_CTRL, + GSWIP_PCE_TBL_CTRL_ADDR_MASK | + GSWIP_PCE_TBL_CTRL_OPMOD_MASK | + GSWIP_PCE_TBL_CTRL_OPMOD_ADWR, + GSWIP_PCE_TBL_CTRL_OPMOD_ADWR); regmap_write(priv->gswip, GSWIP_PCE_TBL_MASK, 0); for (i = 0; i < priv->hw_info->pce_microcode_size; i++) { @@ -549,9 +541,10 @@ static void gswip_port_commit_pvid(struct gswip_priv *priv, int port) } vinr = idx ? GSWIP_PCE_VCTRL_VINR_ALL : GSWIP_PCE_VCTRL_VINR_TAGGED; - gswip_switch_mask(priv, GSWIP_PCE_VCTRL_VINR, + regmap_write_bits(priv->gswip, GSWIP_PCE_VCTRL(port), + GSWIP_PCE_VCTRL_VINR | FIELD_PREP(GSWIP_PCE_VCTRL_VINR, vinr), - GSWIP_PCE_VCTRL(port)); + FIELD_PREP(GSWIP_PCE_VCTRL_VINR, vinr)); /* Note that in GSWIP 2.2 VLAN mode the VID needs to be programmed * directly instead of referencing the index in the Active VLAN Tablet. @@ -569,20 +562,27 @@ static int gswip_port_vlan_filtering(struct dsa_switch *ds, int port, if (vlan_filtering) { /* Use tag based VLAN */ - gswip_switch_mask(priv, - GSWIP_PCE_VCTRL_VSR, - GSWIP_PCE_VCTRL_UVR | GSWIP_PCE_VCTRL_VIMR | - GSWIP_PCE_VCTRL_VEMR | GSWIP_PCE_VCTRL_VID0, - GSWIP_PCE_VCTRL(port)); + regmap_write_bits(priv->gswip, GSWIP_PCE_VCTRL(port), + GSWIP_PCE_VCTRL_VSR | + GSWIP_PCE_VCTRL_UVR | + GSWIP_PCE_VCTRL_VIMR | + GSWIP_PCE_VCTRL_VEMR | + GSWIP_PCE_VCTRL_VID0, + GSWIP_PCE_VCTRL_UVR | + GSWIP_PCE_VCTRL_VIMR | + GSWIP_PCE_VCTRL_VEMR | + GSWIP_PCE_VCTRL_VID0); regmap_clear_bits(priv->gswip, GSWIP_PCE_PCTRL_0p(port), GSWIP_PCE_PCTRL_0_TVM); } else { /* Use port based VLAN */ - gswip_switch_mask(priv, - GSWIP_PCE_VCTRL_UVR | GSWIP_PCE_VCTRL_VIMR | - GSWIP_PCE_VCTRL_VEMR | GSWIP_PCE_VCTRL_VID0, + regmap_write_bits(priv->gswip, GSWIP_PCE_VCTRL(port), + GSWIP_PCE_VCTRL_UVR | + GSWIP_PCE_VCTRL_VIMR | + GSWIP_PCE_VCTRL_VEMR | + GSWIP_PCE_VCTRL_VID0 | GSWIP_PCE_VCTRL_VSR, - GSWIP_PCE_VCTRL(port)); + GSWIP_PCE_VCTRL_VSR); regmap_set_bits(priv->gswip, GSWIP_PCE_PCTRL_0p(port), GSWIP_PCE_PCTRL_0_TVM); } @@ -642,7 +642,7 @@ static int gswip_setup(struct dsa_switch *ds) regmap_write(priv->mdio, GSWIP_MDIO_MDC_CFG0, 0x0); /* Configure the MDIO Clock 2.5 MHz */ - gswip_mdio_mask(priv, 0xff, 0x09, GSWIP_MDIO_MDC_CFG1); + regmap_write_bits(priv->mdio, GSWIP_MDIO_MDC_CFG1, 0xff | 0x09, 0x09); /* bring up the mdio bus */ err = gswip_mdio(priv); @@ -1083,8 +1083,9 @@ static void gswip_port_stp_state_set(struct dsa_switch *ds, int port, u8 state) regmap_set_bits(priv->gswip, GSWIP_SDMA_PCTRLp(port), GSWIP_SDMA_PCTRL_EN); - gswip_switch_mask(priv, GSWIP_PCE_PCTRL_0_PSTATE_MASK, stp_state, - GSWIP_PCE_PCTRL_0p(port)); + regmap_write_bits(priv->gswip, GSWIP_PCE_PCTRL_0p(port), + GSWIP_PCE_PCTRL_0_PSTATE_MASK | stp_state, + stp_state); } static int gswip_port_fdb(struct dsa_switch *ds, int port, @@ -1313,8 +1314,8 @@ static void gswip_port_set_link(struct gswip_priv *priv, int port, bool link) else mdio_phy = GSWIP_MDIO_PHY_LINK_DOWN; - gswip_mdio_mask(priv, GSWIP_MDIO_PHY_LINK_MASK, mdio_phy, - GSWIP_MDIO_PHYp(port)); + regmap_write_bits(priv->mdio, GSWIP_MDIO_PHYp(port), + GSWIP_MDIO_PHY_LINK_MASK | mdio_phy, mdio_phy); } static void gswip_port_set_speed(struct gswip_priv *priv, int port, int speed, @@ -1354,11 +1355,11 @@ static void gswip_port_set_speed(struct gswip_priv *priv, int port, int speed, break; } - gswip_mdio_mask(priv, GSWIP_MDIO_PHY_SPEED_MASK, mdio_phy, - GSWIP_MDIO_PHYp(port)); + regmap_write_bits(priv->mdio, GSWIP_MDIO_PHYp(port), + GSWIP_MDIO_PHY_SPEED_MASK | mdio_phy, mdio_phy); gswip_mii_mask_cfg(priv, GSWIP_MII_CFG_RATE_MASK, mii_cfg, port); - gswip_switch_mask(priv, GSWIP_MAC_CTRL_0_GMII_MASK, mac_ctrl_0, - GSWIP_MAC_CTRL_0p(port)); + regmap_write_bits(priv->gswip, GSWIP_MAC_CTRL_0p(port), + GSWIP_MAC_CTRL_0_GMII_MASK | mac_ctrl_0, mac_ctrl_0); } static void gswip_port_set_duplex(struct gswip_priv *priv, int port, int duplex) @@ -1373,10 +1374,10 @@ static void gswip_port_set_duplex(struct gswip_priv *priv, int port, int duplex) mdio_phy = GSWIP_MDIO_PHY_FDUP_DIS; } - gswip_switch_mask(priv, GSWIP_MAC_CTRL_0_FDUP_MASK, mac_ctrl_0, - GSWIP_MAC_CTRL_0p(port)); - gswip_mdio_mask(priv, GSWIP_MDIO_PHY_FDUP_MASK, mdio_phy, - GSWIP_MDIO_PHYp(port)); + regmap_write_bits(priv->gswip, GSWIP_MAC_CTRL_0p(port), + GSWIP_MAC_CTRL_0_FDUP_MASK | mac_ctrl_0, mac_ctrl_0); + regmap_write_bits(priv->mdio, GSWIP_MDIO_PHYp(port), + GSWIP_MDIO_PHY_FDUP_MASK | mdio_phy, mdio_phy); } static void gswip_port_set_pause(struct gswip_priv *priv, int port, @@ -1402,12 +1403,11 @@ static void gswip_port_set_pause(struct gswip_priv *priv, int port, GSWIP_MDIO_PHY_FCONRX_DIS; } - gswip_switch_mask(priv, GSWIP_MAC_CTRL_0_FCON_MASK, - mac_ctrl_0, GSWIP_MAC_CTRL_0p(port)); - gswip_mdio_mask(priv, - GSWIP_MDIO_PHY_FCONTX_MASK | - GSWIP_MDIO_PHY_FCONRX_MASK, - mdio_phy, GSWIP_MDIO_PHYp(port)); + regmap_write_bits(priv->gswip, GSWIP_MAC_CTRL_0p(port), + GSWIP_MAC_CTRL_0_FCON_MASK | mac_ctrl_0, mac_ctrl_0); + regmap_write_bits(priv->mdio, GSWIP_MDIO_PHYp(port), + GSWIP_MDIO_PHY_FCONTX_MASK | GSWIP_MDIO_PHY_FCONRX_MASK | mdio_phy, + mdio_phy); } static void gswip_phylink_mac_config(struct phylink_config *config, @@ -1526,10 +1526,10 @@ static u32 gswip_bcm_ram_entry_read(struct gswip_priv *priv, u32 table, int err; regmap_write(priv->gswip, GSWIP_BM_RAM_ADDR, index); - gswip_switch_mask(priv, GSWIP_BM_RAM_CTRL_ADDR_MASK | - GSWIP_BM_RAM_CTRL_OPMOD, - table | GSWIP_BM_RAM_CTRL_BAS, - GSWIP_BM_RAM_CTRL); + regmap_write_bits(priv->gswip, GSWIP_BM_RAM_CTRL, + GSWIP_BM_RAM_CTRL_ADDR_MASK | GSWIP_BM_RAM_CTRL_OPMOD | + table | GSWIP_BM_RAM_CTRL_BAS, + table | GSWIP_BM_RAM_CTRL_BAS); err = gswip_switch_r_timeout(priv, GSWIP_BM_RAM_CTRL, GSWIP_BM_RAM_CTRL_BAS); From 1d88358303fc81896b92e50eb732b726e7b88ee6 Mon Sep 17 00:00:00 2001 From: Daniel Golle Date: Tue, 21 Oct 2025 12:17:14 +0100 Subject: [PATCH 0211/1381] net: dsa: lantiq_gswip: optimize regmap_write_bits() statements Further optimize the previous naive conversion of the *_mask() accessor functions to regmap_write_bits by manually removing redundant mask operands. Signed-off-by: Daniel Golle Acked-by; Hauke Mehrtens : Acked-by; Hauke Mehrtens : Link: https://patch.msgid.link/fce2f964b22fe3efc234c664b1e50de28dddf512.1761045000.git.daniel@makrotopia.org Signed-off-by: Jakub Kicinski --- drivers/net/dsa/lantiq/lantiq_gswip.c | 33 ++++++++++++--------------- 1 file changed, 15 insertions(+), 18 deletions(-) diff --git a/drivers/net/dsa/lantiq/lantiq_gswip.c b/drivers/net/dsa/lantiq/lantiq_gswip.c index 71dfddd62d9f..248323524166 100644 --- a/drivers/net/dsa/lantiq/lantiq_gswip.c +++ b/drivers/net/dsa/lantiq/lantiq_gswip.c @@ -276,7 +276,7 @@ static int gswip_pce_table_entry_read(struct gswip_priv *priv, regmap_write_bits(priv->gswip, GSWIP_PCE_TBL_CTRL, GSWIP_PCE_TBL_CTRL_ADDR_MASK | GSWIP_PCE_TBL_CTRL_OPMOD_MASK | - tbl->table | addr_mode | GSWIP_PCE_TBL_CTRL_BAS, + GSWIP_PCE_TBL_CTRL_BAS, tbl->table | addr_mode | GSWIP_PCE_TBL_CTRL_BAS); err = gswip_switch_r_timeout(priv, GSWIP_PCE_TBL_CTRL, @@ -337,8 +337,7 @@ static int gswip_pce_table_entry_write(struct gswip_priv *priv, regmap_write(priv->gswip, GSWIP_PCE_TBL_ADDR, tbl->index); regmap_write_bits(priv->gswip, GSWIP_PCE_TBL_CTRL, GSWIP_PCE_TBL_CTRL_ADDR_MASK | - GSWIP_PCE_TBL_CTRL_OPMOD_MASK | - tbl->table | addr_mode, + GSWIP_PCE_TBL_CTRL_OPMOD_MASK, tbl->table | addr_mode); for (i = 0; i < ARRAY_SIZE(tbl->key); i++) @@ -349,8 +348,7 @@ static int gswip_pce_table_entry_write(struct gswip_priv *priv, regmap_write_bits(priv->gswip, GSWIP_PCE_TBL_CTRL, GSWIP_PCE_TBL_CTRL_ADDR_MASK | - GSWIP_PCE_TBL_CTRL_OPMOD_MASK | - tbl->table | addr_mode, + GSWIP_PCE_TBL_CTRL_OPMOD_MASK, tbl->table | addr_mode); regmap_write(priv->gswip, GSWIP_PCE_TBL_MASK, tbl->mask); @@ -439,7 +437,7 @@ static int gswip_port_enable(struct dsa_switch *ds, int port, mdio_phy = phydev->mdio.addr & GSWIP_MDIO_PHY_ADDR_MASK; regmap_write_bits(priv->mdio, GSWIP_MDIO_PHYp(port), - GSWIP_MDIO_PHY_ADDR_MASK | mdio_phy, + GSWIP_MDIO_PHY_ADDR_MASK, mdio_phy); } @@ -542,8 +540,7 @@ static void gswip_port_commit_pvid(struct gswip_priv *priv, int port) vinr = idx ? GSWIP_PCE_VCTRL_VINR_ALL : GSWIP_PCE_VCTRL_VINR_TAGGED; regmap_write_bits(priv->gswip, GSWIP_PCE_VCTRL(port), - GSWIP_PCE_VCTRL_VINR | - FIELD_PREP(GSWIP_PCE_VCTRL_VINR, vinr), + GSWIP_PCE_VCTRL_VINR, FIELD_PREP(GSWIP_PCE_VCTRL_VINR, vinr)); /* Note that in GSWIP 2.2 VLAN mode the VID needs to be programmed @@ -642,7 +639,7 @@ static int gswip_setup(struct dsa_switch *ds) regmap_write(priv->mdio, GSWIP_MDIO_MDC_CFG0, 0x0); /* Configure the MDIO Clock 2.5 MHz */ - regmap_write_bits(priv->mdio, GSWIP_MDIO_MDC_CFG1, 0xff | 0x09, 0x09); + regmap_write_bits(priv->mdio, GSWIP_MDIO_MDC_CFG1, 0xff, 0x09); /* bring up the mdio bus */ err = gswip_mdio(priv); @@ -1084,7 +1081,7 @@ static void gswip_port_stp_state_set(struct dsa_switch *ds, int port, u8 state) regmap_set_bits(priv->gswip, GSWIP_SDMA_PCTRLp(port), GSWIP_SDMA_PCTRL_EN); regmap_write_bits(priv->gswip, GSWIP_PCE_PCTRL_0p(port), - GSWIP_PCE_PCTRL_0_PSTATE_MASK | stp_state, + GSWIP_PCE_PCTRL_0_PSTATE_MASK, stp_state); } @@ -1315,7 +1312,7 @@ static void gswip_port_set_link(struct gswip_priv *priv, int port, bool link) mdio_phy = GSWIP_MDIO_PHY_LINK_DOWN; regmap_write_bits(priv->mdio, GSWIP_MDIO_PHYp(port), - GSWIP_MDIO_PHY_LINK_MASK | mdio_phy, mdio_phy); + GSWIP_MDIO_PHY_LINK_MASK, mdio_phy); } static void gswip_port_set_speed(struct gswip_priv *priv, int port, int speed, @@ -1356,10 +1353,10 @@ static void gswip_port_set_speed(struct gswip_priv *priv, int port, int speed, } regmap_write_bits(priv->mdio, GSWIP_MDIO_PHYp(port), - GSWIP_MDIO_PHY_SPEED_MASK | mdio_phy, mdio_phy); + GSWIP_MDIO_PHY_SPEED_MASK, mdio_phy); gswip_mii_mask_cfg(priv, GSWIP_MII_CFG_RATE_MASK, mii_cfg, port); regmap_write_bits(priv->gswip, GSWIP_MAC_CTRL_0p(port), - GSWIP_MAC_CTRL_0_GMII_MASK | mac_ctrl_0, mac_ctrl_0); + GSWIP_MAC_CTRL_0_GMII_MASK, mac_ctrl_0); } static void gswip_port_set_duplex(struct gswip_priv *priv, int port, int duplex) @@ -1375,9 +1372,9 @@ static void gswip_port_set_duplex(struct gswip_priv *priv, int port, int duplex) } regmap_write_bits(priv->gswip, GSWIP_MAC_CTRL_0p(port), - GSWIP_MAC_CTRL_0_FDUP_MASK | mac_ctrl_0, mac_ctrl_0); + GSWIP_MAC_CTRL_0_FDUP_MASK, mac_ctrl_0); regmap_write_bits(priv->mdio, GSWIP_MDIO_PHYp(port), - GSWIP_MDIO_PHY_FDUP_MASK | mdio_phy, mdio_phy); + GSWIP_MDIO_PHY_FDUP_MASK, mdio_phy); } static void gswip_port_set_pause(struct gswip_priv *priv, int port, @@ -1404,9 +1401,9 @@ static void gswip_port_set_pause(struct gswip_priv *priv, int port, } regmap_write_bits(priv->gswip, GSWIP_MAC_CTRL_0p(port), - GSWIP_MAC_CTRL_0_FCON_MASK | mac_ctrl_0, mac_ctrl_0); + GSWIP_MAC_CTRL_0_FCON_MASK, mac_ctrl_0); regmap_write_bits(priv->mdio, GSWIP_MDIO_PHYp(port), - GSWIP_MDIO_PHY_FCONTX_MASK | GSWIP_MDIO_PHY_FCONRX_MASK | mdio_phy, + GSWIP_MDIO_PHY_FCONTX_MASK | GSWIP_MDIO_PHY_FCONRX_MASK, mdio_phy); } @@ -1528,7 +1525,7 @@ static u32 gswip_bcm_ram_entry_read(struct gswip_priv *priv, u32 table, regmap_write(priv->gswip, GSWIP_BM_RAM_ADDR, index); regmap_write_bits(priv->gswip, GSWIP_BM_RAM_CTRL, GSWIP_BM_RAM_CTRL_ADDR_MASK | GSWIP_BM_RAM_CTRL_OPMOD | - table | GSWIP_BM_RAM_CTRL_BAS, + GSWIP_BM_RAM_CTRL_BAS, table | GSWIP_BM_RAM_CTRL_BAS); err = gswip_switch_r_timeout(priv, GSWIP_BM_RAM_CTRL, From b0911b9e014004d3e617e8bba27000f521a26422 Mon Sep 17 00:00:00 2001 From: Daniel Golle Date: Tue, 21 Oct 2025 12:17:33 +0100 Subject: [PATCH 0212/1381] net: dsa: lantiq_gswip: harmonize gswip_mii_mask_*() parameters The 'clear' parameter of gswip_mii_mask_cfg() and gswip_mii_mask_pcdu() is inconsistent with the semantics of regmap_write_bits() which also applies the mask to the value to be written. Change the semantic mask/set of the functions gswip_mii_mask_cfg() and gswip_mii_mask_pcdu() to follow the regmap_write_bits() pattern. Signed-off-by: Daniel Golle Acked-by; Hauke Mehrtens : Acked-by; Hauke Mehrtens : Link: https://patch.msgid.link/218854236c97a152af071852bda83d02ff2dd918.1761045000.git.daniel@makrotopia.org Signed-off-by: Jakub Kicinski --- drivers/net/dsa/lantiq/lantiq_gswip.c | 17 +++++++---------- 1 file changed, 7 insertions(+), 10 deletions(-) diff --git a/drivers/net/dsa/lantiq/lantiq_gswip.c b/drivers/net/dsa/lantiq/lantiq_gswip.c index 248323524166..38f7f6352e8d 100644 --- a/drivers/net/dsa/lantiq/lantiq_gswip.c +++ b/drivers/net/dsa/lantiq/lantiq_gswip.c @@ -120,7 +120,7 @@ static u32 gswip_switch_r_timeout(struct gswip_priv *priv, u32 offset, !(val & cleared), 20, 50000); } -static void gswip_mii_mask_cfg(struct gswip_priv *priv, u32 clear, u32 set, +static void gswip_mii_mask_cfg(struct gswip_priv *priv, u32 mask, u32 set, int port) { int reg_port; @@ -131,11 +131,11 @@ static void gswip_mii_mask_cfg(struct gswip_priv *priv, u32 clear, u32 set, reg_port = port + priv->hw_info->mii_port_reg_offset; - regmap_write_bits(priv->mii, GSWIP_MII_CFGp(reg_port), clear | set, + regmap_write_bits(priv->mii, GSWIP_MII_CFGp(reg_port), mask, set); } -static void gswip_mii_mask_pcdu(struct gswip_priv *priv, u32 clear, u32 set, +static void gswip_mii_mask_pcdu(struct gswip_priv *priv, u32 mask, u32 set, int port) { int reg_port; @@ -148,16 +148,13 @@ static void gswip_mii_mask_pcdu(struct gswip_priv *priv, u32 clear, u32 set, switch (reg_port) { case 0: - regmap_write_bits(priv->mii, GSWIP_MII_PCDU0, clear | set, - set); + regmap_write_bits(priv->mii, GSWIP_MII_PCDU0, mask, set); break; case 1: - regmap_write_bits(priv->mii, GSWIP_MII_PCDU1, clear | set, - set); + regmap_write_bits(priv->mii, GSWIP_MII_PCDU1, mask, set); break; case 5: - regmap_write_bits(priv->mii, GSWIP_MII_PCDU5, clear | set, - set); + regmap_write_bits(priv->mii, GSWIP_MII_PCDU5, mask, set); break; } } @@ -1501,7 +1498,7 @@ static void gswip_phylink_mac_link_up(struct phylink_config *config, gswip_port_set_pause(priv, port, tx_pause, rx_pause); } - gswip_mii_mask_cfg(priv, 0, GSWIP_MII_CFG_EN, port); + gswip_mii_mask_cfg(priv, GSWIP_MII_CFG_EN, GSWIP_MII_CFG_EN, port); } static void gswip_get_strings(struct dsa_switch *ds, int port, u32 stringset, From f7746cfcdbc5d3af9c5059deed30e277d926098b Mon Sep 17 00:00:00 2001 From: Wei Zhang Date: Thu, 16 Oct 2025 23:01:00 -0700 Subject: [PATCH 0213/1381] wifi: ath12k: add support for BSS color change Add support for handling BSS color collision events reported by firmware. There are two scenarios where a BSS color collision may be detected: 1. The AP's MAC detects the collision directly, and firmware reports a BSS color collision event to the host. 2. A STA associated with the AP detects the collision. The notification frame from the peer is routed directly to the AP firmware, which handles it and sends the BSS color collision event to the host. Add logic to parse and handle such events, and pass the data up to mac80211. Unlike CSA, firmware does not provide an offload mechanism for BSS color change. Instead, the color change process is triggered via beacon offload TX completion events sent by firmware. BSS color feature is enabled depending on service flag advertised by firmware, based on which color change functionality is invoked. This change builds upon the following ath11k patch. commit 886433a98425 ("ath11k: add support for BSS color change") Tested-on: WCN7850 hw2.0 PCI WLAN.IOE_HMT.1.1-00011-QCAHMTSWPL_V1.0_V2.0_SILICONZ-1 Signed-off-by: Wei Zhang Reviewed-by: Vasanthakumar Thiagarajan Reviewed-by: Baochen Qiang Link: https://patch.msgid.link/20251017060100.1751692-1-wei.zhang@oss.qualcomm.com Signed-off-by: Jeff Johnson --- drivers/net/wireless/ath/ath12k/core.h | 1 + drivers/net/wireless/ath/ath12k/mac.c | 58 +++++++++++++++++++- drivers/net/wireless/ath/ath12k/wmi.c | 73 +++++++++++++++++++++++++- drivers/net/wireless/ath/ath12k/wmi.h | 24 +++++++++ 4 files changed, 154 insertions(+), 2 deletions(-) diff --git a/drivers/net/wireless/ath/ath12k/core.h b/drivers/net/wireless/ath/ath12k/core.h index 02a32b9f3ac2..41da0efaa854 100644 --- a/drivers/net/wireless/ath/ath12k/core.h +++ b/drivers/net/wireless/ath/ath12k/core.h @@ -356,6 +356,7 @@ struct ath12k_link_vif { bool pairwise_key_done; u16 num_stations; bool is_csa_in_progress; + struct wiphy_work bcn_tx_work; }; struct ath12k_vif { diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c index d7bc19cea2a6..e79d457e3c03 100644 --- a/drivers/net/wireless/ath/ath12k/mac.c +++ b/drivers/net/wireless/ath/ath12k/mac.c @@ -3834,6 +3834,38 @@ static void ath12k_recalculate_mgmt_rate(struct ath12k *ar, ath12k_warn(ar->ab, "failed to set beacon tx rate %d\n", ret); } +static void ath12k_mac_bcn_tx_event(struct ath12k_link_vif *arvif) +{ + struct ieee80211_vif *vif = ath12k_ahvif_to_vif(arvif->ahvif); + struct ieee80211_bss_conf *link_conf; + + link_conf = ath12k_mac_get_link_bss_conf(arvif); + if (!link_conf) { + ath12k_warn(arvif->ar->ab, "failed to get link conf for vdev %u\n", + arvif->vdev_id); + return; + } + + if (link_conf->color_change_active) { + if (ieee80211_beacon_cntdwn_is_complete(vif, arvif->link_id)) { + ieee80211_color_change_finish(vif, arvif->link_id); + return; + } + + ieee80211_beacon_update_cntdwn(vif, arvif->link_id); + ath12k_mac_setup_bcn_tmpl(arvif); + } +} + +static void ath12k_mac_bcn_tx_work(struct wiphy *wiphy, struct wiphy_work *work) +{ + struct ath12k_link_vif *arvif = container_of(work, struct ath12k_link_vif, + bcn_tx_work); + + lockdep_assert_wiphy(wiphy); + ath12k_mac_bcn_tx_event(arvif); +} + static void ath12k_mac_init_arvif(struct ath12k_vif *ahvif, struct ath12k_link_vif *arvif, int link_id) { @@ -3863,6 +3895,7 @@ static void ath12k_mac_init_arvif(struct ath12k_vif *ahvif, INIT_LIST_HEAD(&arvif->list); INIT_DELAYED_WORK(&arvif->connection_loss_work, ath12k_mac_vif_sta_connection_loss_work); + wiphy_work_init(&arvif->bcn_tx_work, ath12k_mac_bcn_tx_work); arvif->num_stations = 0; @@ -3900,6 +3933,7 @@ static void ath12k_mac_remove_link_interface(struct ieee80211_hw *hw, lockdep_assert_wiphy(ah->hw->wiphy); cancel_delayed_work_sync(&arvif->connection_loss_work); + wiphy_work_cancel(ath12k_ar_to_hw(ar)->wiphy, &arvif->bcn_tx_work); ath12k_dbg(ar->ab, ATH12K_DBG_MAC, "mac remove link interface (vdev %d link id %d)", arvif->vdev_id, arvif->link_id); @@ -4547,8 +4581,25 @@ static void ath12k_mac_bss_info_changed(struct ath12k *ar, ATH12K_BSS_COLOR_AP_PERIODS, info->he_bss_color.enabled); if (ret) - ath12k_warn(ar->ab, "failed to set bss color collision on vdev %i: %d\n", + ath12k_warn(ar->ab, "failed to set bss color collision on vdev %u: %d\n", arvif->vdev_id, ret); + + param_id = WMI_VDEV_PARAM_BSS_COLOR; + if (info->he_bss_color.enabled) + param_value = info->he_bss_color.color << + IEEE80211_HE_OPERATION_BSS_COLOR_OFFSET; + else + param_value = IEEE80211_HE_OPERATION_BSS_COLOR_DISABLED; + + ret = ath12k_wmi_vdev_set_param_cmd(ar, arvif->vdev_id, + param_id, + param_value); + if (ret) + ath12k_warn(ar->ab, "failed to set bss color param on vdev %u: %d\n", + arvif->vdev_id, ret); + else + ath12k_dbg(ar->ab, ATH12K_DBG_MAC, "bss color param 0x%x set on vdev %u\n", + param_value, arvif->vdev_id); } else if (vif->type == NL80211_IFTYPE_STATION) { ret = ath12k_wmi_send_bss_color_change_enable_cmd(ar, arvif->vdev_id, @@ -13970,6 +14021,11 @@ static int ath12k_mac_hw_register(struct ath12k_hw *ah) wiphy_ext_feature_set(wiphy, NL80211_EXT_FEATURE_CQM_RSSI_LIST); wiphy_ext_feature_set(wiphy, NL80211_EXT_FEATURE_STA_TX_PWR); wiphy_ext_feature_set(wiphy, NL80211_EXT_FEATURE_ACK_SIGNAL_SUPPORT); + if (test_bit(WMI_TLV_SERVICE_BSS_COLOR_OFFLOAD, + ab->wmi_ab.svc_map)) { + wiphy_ext_feature_set(wiphy, NL80211_EXT_FEATURE_BSS_COLOR); + ieee80211_hw_set(hw, DETECTS_COLOR_COLLISION); + } wiphy->cipher_suites = cipher_suites; wiphy->n_cipher_suites = ARRAY_SIZE(cipher_suites); diff --git a/drivers/net/wireless/ath/ath12k/wmi.c b/drivers/net/wireless/ath/ath12k/wmi.c index e76275bd6916..5075d86df36f 100644 --- a/drivers/net/wireless/ath/ath12k/wmi.c +++ b/drivers/net/wireless/ath/ath12k/wmi.c @@ -14,6 +14,7 @@ #include #include #include +#include #include "core.h" #include "debugfs.h" #include "debug.h" @@ -190,6 +191,8 @@ static const struct ath12k_wmi_tlv_policy ath12k_wmi_tlv_policies[] = { .min_len = sizeof(struct wmi_11d_new_cc_event) }, [WMI_TAG_PER_CHAIN_RSSI_STATS] = { .min_len = sizeof(struct wmi_per_chain_rssi_stat_params) }, + [WMI_TAG_OBSS_COLOR_COLLISION_EVT] = { + .min_len = sizeof(struct wmi_obss_color_collision_event) }, }; __le32 ath12k_wmi_tlv_hdr(u32 cmd, u32 len) @@ -3850,6 +3853,58 @@ int ath12k_wmi_fils_discovery(struct ath12k *ar, u32 vdev_id, u32 interval, return ret; } +static void +ath12k_wmi_obss_color_collision_event(struct ath12k_base *ab, struct sk_buff *skb) +{ + const struct wmi_obss_color_collision_event *ev; + struct ath12k_link_vif *arvif; + u32 vdev_id, evt_type; + u64 bitmap; + + const void **tb __free(kfree) = ath12k_wmi_tlv_parse_alloc(ab, skb, GFP_ATOMIC); + if (IS_ERR(tb)) { + ath12k_warn(ab, "failed to parse OBSS color collision tlv %ld\n", + PTR_ERR(tb)); + return; + } + + ev = tb[WMI_TAG_OBSS_COLOR_COLLISION_EVT]; + if (!ev) { + ath12k_warn(ab, "failed to fetch OBSS color collision event\n"); + return; + } + + vdev_id = le32_to_cpu(ev->vdev_id); + evt_type = le32_to_cpu(ev->evt_type); + bitmap = le64_to_cpu(ev->obss_color_bitmap); + + guard(rcu)(); + + arvif = ath12k_mac_get_arvif_by_vdev_id(ab, vdev_id); + if (!arvif) { + ath12k_warn(ab, "no arvif found for vdev %u in OBSS color collision event\n", + vdev_id); + return; + } + + switch (evt_type) { + case WMI_BSS_COLOR_COLLISION_DETECTION: + ieee80211_obss_color_collision_notify(arvif->ahvif->vif, + bitmap, + arvif->link_id); + ath12k_dbg(ab, ATH12K_DBG_WMI, + "obss color collision detected vdev %u event %d bitmap %016llx\n", + vdev_id, evt_type, bitmap); + break; + case WMI_BSS_COLOR_COLLISION_DISABLE: + case WMI_BSS_COLOR_FREE_SLOT_TIMER_EXPIRY: + case WMI_BSS_COLOR_FREE_SLOT_AVAILABLE: + break; + default: + ath12k_warn(ab, "unknown OBSS color collision event type %d\n", evt_type); + } +} + static void ath12k_fill_band_to_mac_param(struct ath12k_base *soc, struct ath12k_wmi_pdev_band_arg *arg) @@ -7014,12 +7069,26 @@ static void ath12k_vdev_start_resp_event(struct ath12k_base *ab, struct sk_buff static void ath12k_bcn_tx_status_event(struct ath12k_base *ab, struct sk_buff *skb) { + struct ath12k_link_vif *arvif; + struct ath12k *ar; u32 vdev_id, tx_status; if (ath12k_pull_bcn_tx_status_ev(ab, skb, &vdev_id, &tx_status) != 0) { ath12k_warn(ab, "failed to extract bcn tx status"); return; } + + guard(rcu)(); + + arvif = ath12k_mac_get_arvif_by_vdev_id(ab, vdev_id); + if (!arvif) { + ath12k_warn(ab, "invalid vdev %u in bcn tx status\n", + vdev_id); + return; + } + + ar = arvif->ar; + wiphy_work_queue(ath12k_ar_to_hw(ar)->wiphy, &arvif->bcn_tx_work); } static void ath12k_vdev_stopped_event(struct ath12k_base *ab, struct sk_buff *skb) @@ -9877,6 +9946,9 @@ static void ath12k_wmi_op_rx(struct ath12k_base *ab, struct sk_buff *skb) case WMI_PDEV_RSSI_DBM_CONVERSION_PARAMS_INFO_EVENTID: ath12k_wmi_rssi_dbm_conversion_params_info_event(ab, skb); break; + case WMI_OBSS_COLOR_COLLISION_DETECTION_EVENTID: + ath12k_wmi_obss_color_collision_event(ab, skb); + break; /* add Unsupported events (rare) here */ case WMI_TBTTOFFSET_EXT_UPDATE_EVENTID: case WMI_PEER_OPER_MODE_CHANGE_EVENTID: @@ -9887,7 +9959,6 @@ static void ath12k_wmi_op_rx(struct ath12k_base *ab, struct sk_buff *skb) /* add Unsupported events (frequent) here */ case WMI_PDEV_GET_HALPHY_CAL_STATUS_EVENTID: case WMI_MGMT_RX_FW_CONSUMED_EVENTID: - case WMI_OBSS_COLOR_COLLISION_DETECTION_EVENTID: /* debug might flood hence silently ignore (no-op) */ break; case WMI_PDEV_UTF_EVENTID: diff --git a/drivers/net/wireless/ath/ath12k/wmi.h b/drivers/net/wireless/ath/ath12k/wmi.h index 64bd968989c8..911ef9d52817 100644 --- a/drivers/net/wireless/ath/ath12k/wmi.h +++ b/drivers/net/wireless/ath/ath12k/wmi.h @@ -4928,6 +4928,24 @@ struct wmi_obss_spatial_reuse_params_cmd { #define ATH12K_BSS_COLOR_STA_PERIODS 10000 #define ATH12K_BSS_COLOR_AP_PERIODS 5000 +/** + * enum wmi_bss_color_collision - Event types for BSS color collision handling + * @WMI_BSS_COLOR_COLLISION_DISABLE: Indicates that BSS color collision detection + * is disabled. + * @WMI_BSS_COLOR_COLLISION_DETECTION: Event triggered when a BSS color collision + * is detected. + * @WMI_BSS_COLOR_FREE_SLOT_TIMER_EXPIRY: Event indicating that the timer for waiting + * on a free BSS color slot has expired. + * @WMI_BSS_COLOR_FREE_SLOT_AVAILABLE: Event indicating that a free BSS color slot + * has become available. + */ +enum wmi_bss_color_collision { + WMI_BSS_COLOR_COLLISION_DISABLE = 0, + WMI_BSS_COLOR_COLLISION_DETECTION, + WMI_BSS_COLOR_FREE_SLOT_TIMER_EXPIRY, + WMI_BSS_COLOR_FREE_SLOT_AVAILABLE, +}; + struct wmi_obss_color_collision_cfg_params_cmd { __le32 tlv_header; __le32 vdev_id; @@ -4945,6 +4963,12 @@ struct wmi_bss_color_change_enable_params_cmd { __le32 enable; } __packed; +struct wmi_obss_color_collision_event { + __le32 vdev_id; + __le32 evt_type; + __le64 obss_color_bitmap; +} __packed; + #define ATH12K_IPV4_TH_SEED_SIZE 5 #define ATH12K_IPV6_TH_SEED_SIZE 11 From a41281f6518e485220d180a6031d302a736fc463 Mon Sep 17 00:00:00 2001 From: Baochen Qiang Date: Fri, 17 Oct 2025 10:36:36 +0800 Subject: [PATCH 0214/1381] wifi: ath12k: restore register window after global reset Hardware target implements an address space larger than that PCI BAR can map. In order to be able to access the whole target address space, the BAR space is split into 4 segments, of which the last 3, called windows, can be dynamically mapped to the desired area. This is achieved by updating WINDOW_REG_ADDRESS register with appropriate window value. Currently each time when accessing a register that beyond WINDOW_START, host calculates the window value and caches it after window update, this way next time when accessing a register falling in the same window, host knows that the window is already good hence no additional update needed. However this mechanism breaks after global reset is triggered in ath12k_pci_soc_global_reset(), because with global reset hardware resets WINDOW_REG_ADDRESS register hence the window is not properly mapped any more. Current host does nothing about this, as a result a subsequent register access may not work as expected if it falls in a window same as before. Although there is no obvious issue seen now, better to fix it to avoid future problem. The fix is done by restoring the window register after global reset. Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.1.c5-00284.1-QCAHMTSWPL_V1.0_V2.0_SILICONZ-3 Fixes: d889913205cf ("wifi: ath12k: driver for Qualcomm Wi-Fi 7 devices") Signed-off-by: Baochen Qiang Reviewed-by: Vasanthakumar Thiagarajan Link: https://patch.msgid.link/20251017-ath12k-reset-window-cache-v1-1-29e0e751deed@oss.qualcomm.com Signed-off-by: Jeff Johnson --- drivers/net/wireless/ath/ath12k/pci.c | 20 +++++++++++++++++++- 1 file changed, 19 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/ath/ath12k/pci.c b/drivers/net/wireless/ath/ath12k/pci.c index 48161db6af57..a12c8379cb46 100644 --- a/drivers/net/wireless/ath/ath12k/pci.c +++ b/drivers/net/wireless/ath/ath12k/pci.c @@ -1,7 +1,7 @@ // SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (c) 2019-2021 The Linux Foundation. All rights reserved. - * Copyright (c) 2021-2025 Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. */ #include @@ -218,6 +218,19 @@ static inline bool ath12k_pci_is_offset_within_mhi_region(u32 offset) return (offset >= PCI_MHIREGLEN_REG && offset <= PCI_MHI_REGION_END); } +static void ath12k_pci_restore_window(struct ath12k_base *ab) +{ + struct ath12k_pci *ab_pci = ath12k_pci_priv(ab); + + spin_lock_bh(&ab_pci->window_lock); + + iowrite32(WINDOW_ENABLE_BIT | ab_pci->register_window, + ab->mem + WINDOW_REG_ADDRESS); + ioread32(ab->mem + WINDOW_REG_ADDRESS); + + spin_unlock_bh(&ab_pci->window_lock); +} + static void ath12k_pci_soc_global_reset(struct ath12k_base *ab) { u32 val, delay; @@ -242,6 +255,11 @@ static void ath12k_pci_soc_global_reset(struct ath12k_base *ab) val = ath12k_pci_read32(ab, PCIE_SOC_GLOBAL_RESET); if (val == 0xffffffff) ath12k_warn(ab, "link down error during global reset\n"); + + /* Restore window register as its content is cleared during + * hardware global reset, such that it aligns with host cache. + */ + ath12k_pci_restore_window(ab); } static void ath12k_pci_clear_dbg_registers(struct ath12k_base *ab) From 197498315de711140bcc4722fdeb7c1761777100 Mon Sep 17 00:00:00 2001 From: Sarika Sharma Date: Tue, 21 Oct 2025 16:52:04 +0530 Subject: [PATCH 0215/1381] wifi: ath12k: Assert base_lock is held before allocating REO update element Add a lockdep assertion to verify that ab->base_lock is held prior to allocating a REO update element in ath12k_dp_prepare_reo_update_elem(). This helps detect potential concurrency issues during development and improves code robustness. Compiled tested only. Signed-off-by: Sarika Sharma Reviewed-by: Vasanthakumar Thiagarajan Reviewed-by: Baochen Qiang Link: https://patch.msgid.link/20251021112204.323242-1-sarika.sharma@oss.qualcomm.com Signed-off-by: Jeff Johnson --- drivers/net/wireless/ath/ath12k/dp_rx.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/wireless/ath/ath12k/dp_rx.c b/drivers/net/wireless/ath/ath12k/dp_rx.c index 6c9f0839c83a..d28d8ffec0f8 100644 --- a/drivers/net/wireless/ath/ath12k/dp_rx.c +++ b/drivers/net/wireless/ath/ath12k/dp_rx.c @@ -1089,6 +1089,8 @@ static int ath12k_dp_prepare_reo_update_elem(struct ath12k_dp *dp, { struct dp_reo_update_rx_queue_elem *elem; + lockdep_assert_held(&dp->ab->base_lock); + elem = kzalloc(sizeof(*elem), GFP_ATOMIC); if (!elem) return -ENOMEM; From c35c178fcdffa5f3bedf261a628769b9a52c2436 Mon Sep 17 00:00:00 2001 From: Przemek Kitszel Date: Fri, 12 Sep 2025 15:06:19 +0200 Subject: [PATCH 0216/1381] ice: enforce RTNL assumption of queue NAPI manipulation Instead of making assumptions in comments move them into code. Be also more precise, RTNL must be locked only when there is NAPI, and we have VSIs w/o NAPI that call ice_vsi_clear_napi_queues() during rmmod. Signed-off-by: Przemek Kitszel Reviewed-by: Paul Menzel Reviewed-by: Aleksandr Loktionov Tested-by: Rinitha S (A Contingent worker at Intel) Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_lib.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/intel/ice/ice_lib.c b/drivers/net/ethernet/intel/ice/ice_lib.c index 4479c824561e..69cb0381c460 100644 --- a/drivers/net/ethernet/intel/ice/ice_lib.c +++ b/drivers/net/ethernet/intel/ice/ice_lib.c @@ -2769,7 +2769,6 @@ void ice_dis_vsi(struct ice_vsi *vsi, bool locked) * @vsi: VSI pointer * * Associate queue[s] with napi for all vectors. - * The caller must hold rtnl_lock. */ void ice_vsi_set_napi_queues(struct ice_vsi *vsi) { @@ -2779,6 +2778,7 @@ void ice_vsi_set_napi_queues(struct ice_vsi *vsi) if (!netdev) return; + ASSERT_RTNL(); ice_for_each_rxq(vsi, q_idx) netif_queue_set_napi(netdev, q_idx, NETDEV_QUEUE_TYPE_RX, &vsi->rx_rings[q_idx]->q_vector->napi); @@ -2799,7 +2799,6 @@ void ice_vsi_set_napi_queues(struct ice_vsi *vsi) * @vsi: VSI pointer * * Clear the association between all VSI queues queue[s] and napi. - * The caller must hold rtnl_lock. */ void ice_vsi_clear_napi_queues(struct ice_vsi *vsi) { @@ -2809,6 +2808,7 @@ void ice_vsi_clear_napi_queues(struct ice_vsi *vsi) if (!netdev) return; + ASSERT_RTNL(); /* Clear the NAPI's interrupt number */ ice_for_each_q_vector(vsi, v_idx) { struct ice_q_vector *q_vector = vsi->q_vectors[v_idx]; From 806c4f32a80627f8977fda53520afc41491d162f Mon Sep 17 00:00:00 2001 From: Przemek Kitszel Date: Fri, 12 Sep 2025 15:06:20 +0200 Subject: [PATCH 0217/1381] ice: move service task start out of ice_init_pf() Move service task start out of ice_init_pf(). Do analogous with deinit. Service task is needed up to the very end of driver removal, later commit of the series will move it later on execution timeline. Signed-off-by: Przemek Kitszel Tested-by: Rinitha S (A Contingent worker at Intel) Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice.h | 1 + drivers/net/ethernet/intel/ice/ice_main.c | 18 +++++++++++------- 2 files changed, 12 insertions(+), 7 deletions(-) diff --git a/drivers/net/ethernet/intel/ice/ice.h b/drivers/net/ethernet/intel/ice/ice.h index 22b8323ff0d0..0e58a58c23eb 100644 --- a/drivers/net/ethernet/intel/ice/ice.h +++ b/drivers/net/ethernet/intel/ice/ice.h @@ -1029,6 +1029,7 @@ int ice_open(struct net_device *netdev); int ice_open_internal(struct net_device *netdev); int ice_stop(struct net_device *netdev); void ice_service_task_schedule(struct ice_pf *pf); +void ice_start_service_task(struct ice_pf *pf); int ice_load(struct ice_pf *pf); void ice_unload(struct ice_pf *pf); void ice_adv_lnk_speed_maps_init(void); diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c index ca95b8800bb3..f9e464b79bca 100644 --- a/drivers/net/ethernet/intel/ice/ice_main.c +++ b/drivers/net/ethernet/intel/ice/ice_main.c @@ -3951,7 +3951,6 @@ u16 ice_get_avail_rxq_count(struct ice_pf *pf) */ static void ice_deinit_pf(struct ice_pf *pf) { - ice_service_task_stop(pf); mutex_destroy(&pf->lag_mutex); mutex_destroy(&pf->adev_mutex); mutex_destroy(&pf->sw_mutex); @@ -4030,6 +4029,14 @@ static void ice_set_pf_caps(struct ice_pf *pf) pf->max_pf_rxqs = func_caps->common_cap.num_rxq; } +void ice_start_service_task(struct ice_pf *pf) +{ + timer_setup(&pf->serv_tmr, ice_service_timer, 0); + pf->serv_tmr_period = HZ; + INIT_WORK(&pf->serv_task, ice_service_task); + clear_bit(ICE_SERVICE_SCHED, pf->state); +} + /** * ice_init_pf - Initialize general software structures (struct ice_pf) * @pf: board private structure to initialize @@ -4049,12 +4056,6 @@ static int ice_init_pf(struct ice_pf *pf) init_waitqueue_head(&pf->reset_wait_queue); - /* setup service timer and periodic service task */ - timer_setup(&pf->serv_tmr, ice_service_timer, 0); - pf->serv_tmr_period = HZ; - INIT_WORK(&pf->serv_task, ice_service_task); - clear_bit(ICE_SERVICE_SCHED, pf->state); - mutex_init(&pf->avail_q_mutex); pf->avail_txqs = bitmap_zalloc(pf->max_pf_txqs, GFP_KERNEL); if (!pf->avail_txqs) @@ -4745,6 +4746,7 @@ int ice_init_dev(struct ice_pf *pf) ice_set_safe_mode_caps(hw); } + ice_start_service_task(pf); err = ice_init_pf(pf); if (err) { dev_err(dev, "ice_init_pf failed: %d\n", err); @@ -4791,6 +4793,7 @@ int ice_init_dev(struct ice_pf *pf) ice_clear_interrupt_scheme(pf); unroll_pf_init: ice_deinit_pf(pf); + ice_service_task_stop(pf); return err; } @@ -4799,6 +4802,7 @@ void ice_deinit_dev(struct ice_pf *pf) ice_free_irq_msix_misc(pf); ice_deinit_pf(pf); ice_deinit_hw(&pf->hw); + ice_service_task_stop(pf); /* Service task is already stopped, so call reset directly. */ ice_reset(&pf->hw, ICE_RESET_PFR); From 2fe18288fce6b421f1dc585bcb9dd3afa32d3ad9 Mon Sep 17 00:00:00 2001 From: Przemek Kitszel Date: Fri, 12 Sep 2025 15:06:21 +0200 Subject: [PATCH 0218/1381] ice: move ice_init_interrupt_scheme() prior ice_init_pf() Move ice_init_interrupt_scheme() prior ice_init_pf(). To enable the move ice_set_pf_caps() was moved out from ice_init_pf() to the caller (ice_init_dev()), and placed prior to the irq scheme init. The move makes deinit order of ice_deinit_dev() and failure-path of ice_init_pf() match (at least in terms of not calling ice_clear_interrupt_scheme() and ice_deinit_pf() in opposite ways). The new order aligns with findings made by Jakub Buchocki in the commit 24b454bc354a ("ice: Fix ice module unload"). Signed-off-by: Przemek Kitszel Reviewed-by: Aleksandr Loktionov Tested-by: Rinitha S (A Contingent worker at Intel) Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_main.c | 25 ++++++++++------------- 1 file changed, 11 insertions(+), 14 deletions(-) diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c index f9e464b79bca..e00c282a8c18 100644 --- a/drivers/net/ethernet/intel/ice/ice_main.c +++ b/drivers/net/ethernet/intel/ice/ice_main.c @@ -4043,8 +4043,6 @@ void ice_start_service_task(struct ice_pf *pf) */ static int ice_init_pf(struct ice_pf *pf) { - ice_set_pf_caps(pf); - mutex_init(&pf->sw_mutex); mutex_init(&pf->tc_mutex); mutex_init(&pf->adev_mutex); @@ -4746,11 +4744,18 @@ int ice_init_dev(struct ice_pf *pf) ice_set_safe_mode_caps(hw); } + ice_set_pf_caps(pf); + err = ice_init_interrupt_scheme(pf); + if (err) { + dev_err(dev, "ice_init_interrupt_scheme failed: %d\n", err); + return -EIO; + } + ice_start_service_task(pf); err = ice_init_pf(pf); if (err) { dev_err(dev, "ice_init_pf failed: %d\n", err); - return err; + goto unroll_irq_scheme_init; } pf->hw.udp_tunnel_nic.set_port = ice_udp_tunnel_set_port; @@ -4768,14 +4773,6 @@ int ice_init_dev(struct ice_pf *pf) pf->hw.udp_tunnel_nic.tables[1].tunnel_types = UDP_TUNNEL_TYPE_GENEVE; } - - err = ice_init_interrupt_scheme(pf); - if (err) { - dev_err(dev, "ice_init_interrupt_scheme failed: %d\n", err); - err = -EIO; - goto unroll_pf_init; - } - /* In case of MSIX we are going to setup the misc vector right here * to handle admin queue events etc. In case of legacy and MSI * the misc functionality and queue processing is combined in @@ -4784,16 +4781,16 @@ int ice_init_dev(struct ice_pf *pf) err = ice_req_irq_msix_misc(pf); if (err) { dev_err(dev, "setup of misc vector failed: %d\n", err); - goto unroll_irq_scheme_init; + goto unroll_pf_init; } return 0; -unroll_irq_scheme_init: - ice_clear_interrupt_scheme(pf); unroll_pf_init: ice_deinit_pf(pf); +unroll_irq_scheme_init: ice_service_task_stop(pf); + ice_clear_interrupt_scheme(pf); return err; } From 71430451f81bd6550e46d89b69103a111fc42982 Mon Sep 17 00:00:00 2001 From: Przemek Kitszel Date: Fri, 12 Sep 2025 15:06:22 +0200 Subject: [PATCH 0219/1381] ice: ice_init_pf: destroy mutexes and xarrays on memory alloc failure Unroll actions of ice_init_pf() when it fails. ice_deinit_pf() happens to be perfect to call here. Signed-off-by: Przemek Kitszel Reviewed-by: Aleksandr Loktionov Tested-by: Rinitha S (A Contingent worker at Intel) Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_main.c | 31 +++++++++-------------- 1 file changed, 12 insertions(+), 19 deletions(-) diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c index e00c282a8c18..09dee43e48aa 100644 --- a/drivers/net/ethernet/intel/ice/ice_main.c +++ b/drivers/net/ethernet/intel/ice/ice_main.c @@ -3951,6 +3951,8 @@ u16 ice_get_avail_rxq_count(struct ice_pf *pf) */ static void ice_deinit_pf(struct ice_pf *pf) { + /* note that we unroll also on ice_init_pf() failure here */ + mutex_destroy(&pf->lag_mutex); mutex_destroy(&pf->adev_mutex); mutex_destroy(&pf->sw_mutex); @@ -4055,25 +4057,6 @@ static int ice_init_pf(struct ice_pf *pf) init_waitqueue_head(&pf->reset_wait_queue); mutex_init(&pf->avail_q_mutex); - pf->avail_txqs = bitmap_zalloc(pf->max_pf_txqs, GFP_KERNEL); - if (!pf->avail_txqs) - return -ENOMEM; - - pf->avail_rxqs = bitmap_zalloc(pf->max_pf_rxqs, GFP_KERNEL); - if (!pf->avail_rxqs) { - bitmap_free(pf->avail_txqs); - pf->avail_txqs = NULL; - return -ENOMEM; - } - - pf->txtime_txqs = bitmap_zalloc(pf->max_pf_txqs, GFP_KERNEL); - if (!pf->txtime_txqs) { - bitmap_free(pf->avail_txqs); - pf->avail_txqs = NULL; - bitmap_free(pf->avail_rxqs); - pf->avail_rxqs = NULL; - return -ENOMEM; - } mutex_init(&pf->vfs.table_lock); hash_init(pf->vfs.table); @@ -4086,7 +4069,17 @@ static int ice_init_pf(struct ice_pf *pf) xa_init(&pf->dyn_ports); xa_init(&pf->sf_nums); + pf->avail_txqs = bitmap_zalloc(pf->max_pf_txqs, GFP_KERNEL); + pf->avail_rxqs = bitmap_zalloc(pf->max_pf_rxqs, GFP_KERNEL); + pf->txtime_txqs = bitmap_zalloc(pf->max_pf_txqs, GFP_KERNEL); + if (!pf->avail_txqs || !pf->avail_rxqs || !pf->txtime_txqs) + goto undo_init; + return 0; +undo_init: + /* deinit handles half-initialized pf just fine */ + ice_deinit_pf(pf); + return -ENOMEM; } /** From e3bf1cdde7471bab7fc20dd1a37c2cdb82d3f76b Mon Sep 17 00:00:00 2001 From: Przemek Kitszel Date: Fri, 12 Sep 2025 15:06:23 +0200 Subject: [PATCH 0220/1381] ice: move udp_tunnel_nic and misc IRQ setup into ice_init_pf() Move udp_tunnel_nic setup and ice_req_irq_msix_misc() call into ice_init_pf(), remove some redundancy in the former while moving. Move ice_free_irq_msix_misc() call into ice_deinit_pf(), to mimic the above in terms of needed cleanup. Guard it via emptiness check, to keep the allowance of half-initialized pf being cleaned up. Signed-off-by: Przemek Kitszel Reviewed-by: Aleksandr Loktionov Tested-by: Rinitha S (A Contingent worker at Intel) Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_main.c | 58 +++++++++++------------ 1 file changed, 28 insertions(+), 30 deletions(-) diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c index 09dee43e48aa..1691dda1067e 100644 --- a/drivers/net/ethernet/intel/ice/ice_main.c +++ b/drivers/net/ethernet/intel/ice/ice_main.c @@ -3978,6 +3978,9 @@ static void ice_deinit_pf(struct ice_pf *pf) if (pf->ptp.clock) ptp_clock_unregister(pf->ptp.clock); + if (!xa_empty(&pf->irq_tracker.entries)) + ice_free_irq_msix_misc(pf); + xa_destroy(&pf->dyn_ports); xa_destroy(&pf->sf_nums); } @@ -4045,6 +4048,11 @@ void ice_start_service_task(struct ice_pf *pf) */ static int ice_init_pf(struct ice_pf *pf) { + struct udp_tunnel_nic_info *udp_tunnel_nic = &pf->hw.udp_tunnel_nic; + struct device *dev = ice_pf_to_dev(pf); + struct ice_hw *hw = &pf->hw; + int err = -ENOMEM; + mutex_init(&pf->sw_mutex); mutex_init(&pf->tc_mutex); mutex_init(&pf->adev_mutex); @@ -4075,11 +4083,30 @@ static int ice_init_pf(struct ice_pf *pf) if (!pf->avail_txqs || !pf->avail_rxqs || !pf->txtime_txqs) goto undo_init; + udp_tunnel_nic->set_port = ice_udp_tunnel_set_port; + udp_tunnel_nic->unset_port = ice_udp_tunnel_unset_port; + udp_tunnel_nic->shared = &hw->udp_tunnel_shared; + udp_tunnel_nic->tables[0].n_entries = hw->tnl.valid_count[TNL_VXLAN]; + udp_tunnel_nic->tables[0].tunnel_types = UDP_TUNNEL_TYPE_VXLAN; + udp_tunnel_nic->tables[1].n_entries = hw->tnl.valid_count[TNL_GENEVE]; + udp_tunnel_nic->tables[1].tunnel_types = UDP_TUNNEL_TYPE_GENEVE; + + /* In case of MSIX we are going to setup the misc vector right here + * to handle admin queue events etc. In case of legacy and MSI + * the misc functionality and queue processing is combined in + * the same vector and that gets setup at open. + */ + err = ice_req_irq_msix_misc(pf); + if (err) { + dev_err(dev, "setup of misc vector failed: %d\n", err); + goto undo_init; + } + return 0; undo_init: /* deinit handles half-initialized pf just fine */ ice_deinit_pf(pf); - return -ENOMEM; + return err; } /** @@ -4751,36 +4778,8 @@ int ice_init_dev(struct ice_pf *pf) goto unroll_irq_scheme_init; } - pf->hw.udp_tunnel_nic.set_port = ice_udp_tunnel_set_port; - pf->hw.udp_tunnel_nic.unset_port = ice_udp_tunnel_unset_port; - pf->hw.udp_tunnel_nic.shared = &pf->hw.udp_tunnel_shared; - if (pf->hw.tnl.valid_count[TNL_VXLAN]) { - pf->hw.udp_tunnel_nic.tables[0].n_entries = - pf->hw.tnl.valid_count[TNL_VXLAN]; - pf->hw.udp_tunnel_nic.tables[0].tunnel_types = - UDP_TUNNEL_TYPE_VXLAN; - } - if (pf->hw.tnl.valid_count[TNL_GENEVE]) { - pf->hw.udp_tunnel_nic.tables[1].n_entries = - pf->hw.tnl.valid_count[TNL_GENEVE]; - pf->hw.udp_tunnel_nic.tables[1].tunnel_types = - UDP_TUNNEL_TYPE_GENEVE; - } - /* In case of MSIX we are going to setup the misc vector right here - * to handle admin queue events etc. In case of legacy and MSI - * the misc functionality and queue processing is combined in - * the same vector and that gets setup at open. - */ - err = ice_req_irq_msix_misc(pf); - if (err) { - dev_err(dev, "setup of misc vector failed: %d\n", err); - goto unroll_pf_init; - } - return 0; -unroll_pf_init: - ice_deinit_pf(pf); unroll_irq_scheme_init: ice_service_task_stop(pf); ice_clear_interrupt_scheme(pf); @@ -4789,7 +4788,6 @@ int ice_init_dev(struct ice_pf *pf) void ice_deinit_dev(struct ice_pf *pf) { - ice_free_irq_msix_misc(pf); ice_deinit_pf(pf); ice_deinit_hw(&pf->hw); ice_service_task_stop(pf); From ef825bdb4605742c4efc03f67d930e80c42f33cb Mon Sep 17 00:00:00 2001 From: Przemek Kitszel Date: Fri, 12 Sep 2025 15:06:24 +0200 Subject: [PATCH 0221/1381] ice: move ice_init_pf() out of ice_init_dev() Move ice_init_pf() out of ice_init_dev(). Do the same for deinit counterpart. Signed-off-by: Przemek Kitszel Reviewed-by: Aleksandr Loktionov Tested-by: Rinitha S (A Contingent worker at Intel) Signed-off-by: Tony Nguyen --- .../net/ethernet/intel/ice/devlink/devlink.c | 16 ++++++++-- drivers/net/ethernet/intel/ice/ice.h | 2 ++ drivers/net/ethernet/intel/ice/ice_main.c | 32 +++++++++---------- 3 files changed, 31 insertions(+), 19 deletions(-) diff --git a/drivers/net/ethernet/intel/ice/devlink/devlink.c b/drivers/net/ethernet/intel/ice/devlink/devlink.c index fb2de521731a..c354a03c950c 100644 --- a/drivers/net/ethernet/intel/ice/devlink/devlink.c +++ b/drivers/net/ethernet/intel/ice/devlink/devlink.c @@ -459,6 +459,7 @@ static void ice_devlink_reinit_down(struct ice_pf *pf) rtnl_lock(); ice_vsi_decfg(ice_get_main_vsi(pf)); rtnl_unlock(); + ice_deinit_pf(pf); ice_deinit_dev(pf); } @@ -1231,11 +1232,12 @@ static void ice_set_min_max_msix(struct ice_pf *pf) static int ice_devlink_reinit_up(struct ice_pf *pf) { struct ice_vsi *vsi = ice_get_main_vsi(pf); + struct device *dev = ice_pf_to_dev(pf); int err; err = ice_init_hw(&pf->hw); if (err) { - dev_err(ice_pf_to_dev(pf), "ice_init_hw failed: %d\n", err); + dev_err(dev, "ice_init_hw failed: %d\n", err); return err; } @@ -1246,13 +1248,19 @@ static int ice_devlink_reinit_up(struct ice_pf *pf) if (err) goto unroll_hw_init; + err = ice_init_pf(pf); + if (err) { + dev_err(dev, "ice_init_pf failed: %d\n", err); + goto unroll_dev_init; + } + vsi->flags = ICE_VSI_FLAG_INIT; rtnl_lock(); err = ice_vsi_cfg(vsi); rtnl_unlock(); if (err) - goto err_vsi_cfg; + goto unroll_pf_init; /* No need to take devl_lock, it's already taken by devlink API */ err = ice_load(pf); @@ -1265,7 +1273,9 @@ static int ice_devlink_reinit_up(struct ice_pf *pf) rtnl_lock(); ice_vsi_decfg(vsi); rtnl_unlock(); -err_vsi_cfg: +unroll_pf_init: + ice_deinit_pf(pf); +unroll_dev_init: ice_deinit_dev(pf); unroll_hw_init: ice_deinit_hw(&pf->hw); diff --git a/drivers/net/ethernet/intel/ice/ice.h b/drivers/net/ethernet/intel/ice/ice.h index 0e58a58c23eb..9a1abd457337 100644 --- a/drivers/net/ethernet/intel/ice/ice.h +++ b/drivers/net/ethernet/intel/ice/ice.h @@ -1035,6 +1035,8 @@ void ice_unload(struct ice_pf *pf); void ice_adv_lnk_speed_maps_init(void); int ice_init_dev(struct ice_pf *pf); void ice_deinit_dev(struct ice_pf *pf); +int ice_init_pf(struct ice_pf *pf); +void ice_deinit_pf(struct ice_pf *pf); int ice_change_mtu(struct net_device *netdev, int new_mtu); void ice_tx_timeout(struct net_device *netdev, unsigned int txqueue); int ice_xdp(struct net_device *dev, struct netdev_bpf *xdp); diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c index 1691dda1067e..a4acc42fabab 100644 --- a/drivers/net/ethernet/intel/ice/ice_main.c +++ b/drivers/net/ethernet/intel/ice/ice_main.c @@ -3949,7 +3949,7 @@ u16 ice_get_avail_rxq_count(struct ice_pf *pf) * ice_deinit_pf - Unrolls initialziations done by ice_init_pf * @pf: board private structure to initialize */ -static void ice_deinit_pf(struct ice_pf *pf) +void ice_deinit_pf(struct ice_pf *pf) { /* note that we unroll also on ice_init_pf() failure here */ @@ -4045,8 +4045,9 @@ void ice_start_service_task(struct ice_pf *pf) /** * ice_init_pf - Initialize general software structures (struct ice_pf) * @pf: board private structure to initialize + * Return: 0 on success, negative errno otherwise. */ -static int ice_init_pf(struct ice_pf *pf) +int ice_init_pf(struct ice_pf *pf) { struct udp_tunnel_nic_info *udp_tunnel_nic = &pf->hw.udp_tunnel_nic; struct device *dev = ice_pf_to_dev(pf); @@ -4772,23 +4773,12 @@ int ice_init_dev(struct ice_pf *pf) } ice_start_service_task(pf); - err = ice_init_pf(pf); - if (err) { - dev_err(dev, "ice_init_pf failed: %d\n", err); - goto unroll_irq_scheme_init; - } return 0; - -unroll_irq_scheme_init: - ice_service_task_stop(pf); - ice_clear_interrupt_scheme(pf); - return err; } void ice_deinit_dev(struct ice_pf *pf) { - ice_deinit_pf(pf); ice_deinit_hw(&pf->hw); ice_service_task_stop(pf); @@ -5030,21 +5020,28 @@ static void ice_deinit_devlink(struct ice_pf *pf) static int ice_init(struct ice_pf *pf) { + struct device *dev = ice_pf_to_dev(pf); int err; err = ice_init_dev(pf); if (err) return err; + err = ice_init_pf(pf); + if (err) { + dev_err(dev, "ice_init_pf failed: %d\n", err); + goto unroll_dev_init; + } + if (pf->hw.mac_type == ICE_MAC_E830) { err = pci_enable_ptm(pf->pdev, NULL); if (err) - dev_dbg(ice_pf_to_dev(pf), "PCIe PTM not supported by PCIe bus/controller\n"); + dev_dbg(dev, "PCIe PTM not supported by PCIe bus/controller\n"); } err = ice_alloc_vsis(pf); if (err) - goto err_alloc_vsis; + goto unroll_pf_init; err = ice_init_pf_sw(pf); if (err) @@ -5081,7 +5078,9 @@ static int ice_init(struct ice_pf *pf) ice_deinit_pf_sw(pf); err_init_pf_sw: ice_dealloc_vsis(pf); -err_alloc_vsis: +unroll_pf_init: + ice_deinit_pf(pf); +unroll_dev_init: ice_deinit_dev(pf); return err; } @@ -5093,6 +5092,7 @@ static void ice_deinit(struct ice_pf *pf) ice_deinit_pf_sw(pf); ice_dealloc_vsis(pf); + ice_deinit_pf(pf); ice_deinit_dev(pf); } From c2fb9398f73d41cb2b5da74ff505578525ee3fd8 Mon Sep 17 00:00:00 2001 From: Przemek Kitszel Date: Fri, 12 Sep 2025 15:06:25 +0200 Subject: [PATCH 0222/1381] ice: extract ice_init_dev() from ice_init() Extract ice_init_dev() from ice_init(), to allow service task and IRQ scheme teardown to be put after clearing SW constructs in the subsequent commit. Signed-off-by: Przemek Kitszel Reviewed-by: Aleksandr Loktionov Tested-by: Rinitha S (A Contingent worker at Intel) Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_main.c | 18 +++++++++--------- 1 file changed, 9 insertions(+), 9 deletions(-) diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c index a4acc42fabab..9a817c3c8b99 100644 --- a/drivers/net/ethernet/intel/ice/ice_main.c +++ b/drivers/net/ethernet/intel/ice/ice_main.c @@ -5023,14 +5023,10 @@ static int ice_init(struct ice_pf *pf) struct device *dev = ice_pf_to_dev(pf); int err; - err = ice_init_dev(pf); - if (err) - return err; - err = ice_init_pf(pf); if (err) { dev_err(dev, "ice_init_pf failed: %d\n", err); - goto unroll_dev_init; + return err; } if (pf->hw.mac_type == ICE_MAC_E830) { @@ -5080,8 +5076,6 @@ static int ice_init(struct ice_pf *pf) ice_dealloc_vsis(pf); unroll_pf_init: ice_deinit_pf(pf); -unroll_dev_init: - ice_deinit_dev(pf); return err; } @@ -5093,7 +5087,6 @@ static void ice_deinit(struct ice_pf *pf) ice_deinit_pf_sw(pf); ice_dealloc_vsis(pf); ice_deinit_pf(pf); - ice_deinit_dev(pf); } /** @@ -5323,10 +5316,14 @@ ice_probe(struct pci_dev *pdev, const struct pci_device_id __always_unused *ent) } pf->adapter = adapter; - err = ice_init(pf); + err = ice_init_dev(pf); if (err) goto unroll_adapter; + err = ice_init(pf); + if (err) + goto unroll_dev_init; + devl_lock(priv_to_devlink(pf)); err = ice_load(pf); if (err) @@ -5344,6 +5341,8 @@ ice_probe(struct pci_dev *pdev, const struct pci_device_id __always_unused *ent) unroll_init: devl_unlock(priv_to_devlink(pf)); ice_deinit(pf); +unroll_dev_init: + ice_deinit_dev(pf); unroll_adapter: ice_adapter_put(pdev); unroll_hw_init: @@ -5457,6 +5456,7 @@ static void ice_remove(struct pci_dev *pdev) devl_unlock(priv_to_devlink(pf)); ice_deinit(pf); + ice_deinit_dev(pf); ice_vsi_release_all(pf); ice_setup_mc_magic_wake(pf); From 8a37f9e2ff40f4a4fa8def22febefe4daf58e573 Mon Sep 17 00:00:00 2001 From: Przemek Kitszel Date: Fri, 12 Sep 2025 15:06:26 +0200 Subject: [PATCH 0223/1381] ice: move ice_deinit_dev() to the end of deinit paths ice_deinit_dev() takes care of turning off adminq processing, which is much needed during driver teardown (remove, reset, error path). Move it to the very end where applicable. For example, ice_deinit_hw() called after adminq deinit slows rmmod on my two-card setup by about 60 seconds. ice_init_dev() and ice_deinit_dev() scopes were reduced by previous commits of the series, with a final touch of extracting ice_init_dev_hw() out now (there is no deinit counterpart). Note that removed ice_service_task_stop() call from ice_remove() is placed in the ice_deinit_dev() (and stopping twice makes no sense). Signed-off-by: Przemek Kitszel Reviewed-by: Aleksandr Loktionov Tested-by: Rinitha S (A Contingent worker at Intel) Signed-off-by: Tony Nguyen --- .../net/ethernet/intel/ice/devlink/devlink.c | 5 +++- drivers/net/ethernet/intel/ice/ice.h | 1 + drivers/net/ethernet/intel/ice/ice_common.c | 3 +++ drivers/net/ethernet/intel/ice/ice_main.c | 23 ++++++++++++------- 4 files changed, 23 insertions(+), 9 deletions(-) diff --git a/drivers/net/ethernet/intel/ice/devlink/devlink.c b/drivers/net/ethernet/intel/ice/devlink/devlink.c index c354a03c950c..938914abbe06 100644 --- a/drivers/net/ethernet/intel/ice/devlink/devlink.c +++ b/drivers/net/ethernet/intel/ice/devlink/devlink.c @@ -1233,6 +1233,7 @@ static int ice_devlink_reinit_up(struct ice_pf *pf) { struct ice_vsi *vsi = ice_get_main_vsi(pf); struct device *dev = ice_pf_to_dev(pf); + bool need_dev_deinit = false; int err; err = ice_init_hw(&pf->hw); @@ -1276,9 +1277,11 @@ static int ice_devlink_reinit_up(struct ice_pf *pf) unroll_pf_init: ice_deinit_pf(pf); unroll_dev_init: - ice_deinit_dev(pf); + need_dev_deinit = true; unroll_hw_init: ice_deinit_hw(&pf->hw); + if (need_dev_deinit) + ice_deinit_dev(pf); return err; } diff --git a/drivers/net/ethernet/intel/ice/ice.h b/drivers/net/ethernet/intel/ice/ice.h index 9a1abd457337..9ee596773f34 100644 --- a/drivers/net/ethernet/intel/ice/ice.h +++ b/drivers/net/ethernet/intel/ice/ice.h @@ -1033,6 +1033,7 @@ void ice_start_service_task(struct ice_pf *pf); int ice_load(struct ice_pf *pf); void ice_unload(struct ice_pf *pf); void ice_adv_lnk_speed_maps_init(void); +void ice_init_dev_hw(struct ice_pf *pf); int ice_init_dev(struct ice_pf *pf); void ice_deinit_dev(struct ice_pf *pf); int ice_init_pf(struct ice_pf *pf); diff --git a/drivers/net/ethernet/intel/ice/ice_common.c b/drivers/net/ethernet/intel/ice/ice_common.c index 2250426ec91b..b097cc8b175c 100644 --- a/drivers/net/ethernet/intel/ice/ice_common.c +++ b/drivers/net/ethernet/intel/ice/ice_common.c @@ -1161,6 +1161,9 @@ int ice_init_hw(struct ice_hw *hw) status = ice_init_hw_tbls(hw); if (status) goto err_unroll_fltr_mgmt_struct; + + ice_init_dev_hw(hw->back); + mutex_init(&hw->tnl_lock); ice_init_chk_recipe_reuse_support(hw); diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c index 9a817c3c8b99..a1fe2d363adb 100644 --- a/drivers/net/ethernet/intel/ice/ice_main.c +++ b/drivers/net/ethernet/intel/ice/ice_main.c @@ -4742,9 +4742,8 @@ static void ice_decfg_netdev(struct ice_vsi *vsi) vsi->netdev = NULL; } -int ice_init_dev(struct ice_pf *pf) +void ice_init_dev_hw(struct ice_pf *pf) { - struct device *dev = ice_pf_to_dev(pf); struct ice_hw *hw = &pf->hw; int err; @@ -4764,6 +4763,12 @@ int ice_init_dev(struct ice_pf *pf) */ ice_set_safe_mode_caps(hw); } +} + +int ice_init_dev(struct ice_pf *pf) +{ + struct device *dev = ice_pf_to_dev(pf); + int err; ice_set_pf_caps(pf); err = ice_init_interrupt_scheme(pf); @@ -5220,6 +5225,7 @@ static int ice_probe(struct pci_dev *pdev, const struct pci_device_id __always_unused *ent) { struct device *dev = &pdev->dev; + bool need_dev_deinit = false; struct ice_adapter *adapter; struct ice_pf *pf; struct ice_hw *hw; @@ -5342,11 +5348,13 @@ ice_probe(struct pci_dev *pdev, const struct pci_device_id __always_unused *ent) devl_unlock(priv_to_devlink(pf)); ice_deinit(pf); unroll_dev_init: - ice_deinit_dev(pf); + need_dev_deinit = true; unroll_adapter: ice_adapter_put(pdev); unroll_hw_init: ice_deinit_hw(hw); + if (need_dev_deinit) + ice_deinit_dev(pf); return err; } @@ -5441,10 +5449,6 @@ static void ice_remove(struct pci_dev *pdev) ice_hwmon_exit(pf); - ice_service_task_stop(pf); - ice_aq_cancel_waiting_tasks(pf); - set_bit(ICE_DOWN, pf->state); - if (!ice_is_safe_mode(pf)) ice_remove_arfs(pf); @@ -5456,13 +5460,16 @@ static void ice_remove(struct pci_dev *pdev) devl_unlock(priv_to_devlink(pf)); ice_deinit(pf); - ice_deinit_dev(pf); ice_vsi_release_all(pf); ice_setup_mc_magic_wake(pf); ice_set_wake(pf); ice_adapter_put(pdev); + + ice_deinit_dev(pf); + ice_aq_cancel_waiting_tasks(pf); + set_bit(ICE_DOWN, pf->state); } /** From 1390b8b3d2bef9bfbb852fc735430798bfca36e7 Mon Sep 17 00:00:00 2001 From: Przemek Kitszel Date: Fri, 12 Sep 2025 15:06:27 +0200 Subject: [PATCH 0224/1381] ice: remove duplicate call to ice_deinit_hw() on error paths Current unwinding code on error paths of ice_devlink_reinit_up() and ice_probe() have manual call to ice_deinit_hw() (which is good, as there is also manual call to ice_hw_init() there), which is then duplicated (and was prior current series) in ice_deinit_dev(). Fix the above by removing ice_deinit_hw() from ice_deinit_dev(). Add a (now missing) call in ice_remove(). Reported-by: Jacob Keller Link: https://lore.kernel.org/intel-wired-lan/20250717-jk-ddp-safe-mode-issue-v1-1-e113b2baed79@intel.com/ Fixes: 4d3f59bfa2cd ("ice: split ice_init_hw() out from ice_init_dev()") Signed-off-by: Przemek Kitszel Reviewed-by: Aleksandr Loktionov Tested-by: Rinitha S (A Contingent worker at Intel) Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c index a1fe2d363adb..1de3da7b3907 100644 --- a/drivers/net/ethernet/intel/ice/ice_main.c +++ b/drivers/net/ethernet/intel/ice/ice_main.c @@ -4784,7 +4784,6 @@ int ice_init_dev(struct ice_pf *pf) void ice_deinit_dev(struct ice_pf *pf) { - ice_deinit_hw(&pf->hw); ice_service_task_stop(pf); /* Service task is already stopped, so call reset directly. */ @@ -5466,6 +5465,7 @@ static void ice_remove(struct pci_dev *pdev) ice_set_wake(pf); ice_adapter_put(pdev); + ice_deinit_hw(&pf->hw); ice_deinit_dev(pf); ice_aq_cancel_waiting_tasks(pf); From 06d6322280d95757cef1e3ee0fc62c644629523e Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Wed, 22 Oct 2025 05:39:45 +0000 Subject: [PATCH 0225/1381] neighbour: Use RCU list helpers for neigh_parms.list writers. We will convert RTM_GETNEIGHTBL to RCU soon, where we traverse tbl->parms_list under RCU in neightbl_dump_info(). Let's use RCU list helper for neigh_parms in neigh_parms_alloc() and neigh_parms_release(). neigh_table_init() uses the plain list_add() for the default neigh_parm that is embedded in the table and not yet published. Note that neigh_parms_release() already uses call_rcu() to free neigh_parms. Signed-off-by: Kuniyuki Iwashima Reviewed-by: Eric Dumazet Link: https://patch.msgid.link/20251022054004.2514876-2-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/core/neighbour.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/core/neighbour.c b/net/core/neighbour.c index bddfa389effa..98428f60731b 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -1764,7 +1764,7 @@ struct neigh_parms *neigh_parms_alloc(struct net_device *dev, } write_lock_bh(&tbl->lock); - list_add(&p->list, &tbl->parms.list); + list_add_rcu(&p->list, &tbl->parms.list); write_unlock_bh(&tbl->lock); neigh_parms_data_state_cleanall(p); @@ -1786,7 +1786,7 @@ void neigh_parms_release(struct neigh_table *tbl, struct neigh_parms *parms) if (!parms || parms == &tbl->parms) return; write_lock_bh(&tbl->lock); - list_del(&parms->list); + list_del_rcu(&parms->list); parms->dead = 1; write_unlock_bh(&tbl->lock); netdev_put(parms->dev, &parms->dev_tracker); From 35d7c70870338aa6a367b9e4ed528914320b0be0 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Wed, 22 Oct 2025 05:39:46 +0000 Subject: [PATCH 0226/1381] neighbour: Annotate access to neigh_parms fields. NEIGH_VAR() is read locklessly in the fast path, and IPv6 ndisc uses NEIGH_VAR_SET() locklessly. The next patch will convert neightbl_dump_info() to RCU. Let's annotate accesses to neigh_param with READ_ONCE() and WRITE_ONCE(). Note that ndisc_ifinfo_sysctl_change() uses &NEIGH_VAR() and we cannot use '&' with READ_ONCE(), so NEIGH_VAR_PTR() is introduced. Note also that NEIGH_VAR_INIT() does not need WRITE_ONCE() as it is before parms is published. Also, the only user hippi_neigh_setup_dev() is no longer called since commit e3804cbebb67 ("net: remove COMPAT_NET_DEV_OPS"), which looks wrong, but probably no one uses HIPPI and RoadRunner. Signed-off-by: Kuniyuki Iwashima Reviewed-by: Eric Dumazet Link: https://patch.msgid.link/20251022054004.2514876-3-kuniyu@google.com Signed-off-by: Jakub Kicinski --- include/net/neighbour.h | 15 ++++++++++++--- net/core/neighbour.c | 17 ++++++----------- net/ipv6/ndisc.c | 8 ++++---- 3 files changed, 22 insertions(+), 18 deletions(-) diff --git a/include/net/neighbour.h b/include/net/neighbour.h index 4a30bd458c5a..998ff9eccebb 100644 --- a/include/net/neighbour.h +++ b/include/net/neighbour.h @@ -92,15 +92,17 @@ struct neigh_parms { static inline void neigh_var_set(struct neigh_parms *p, int index, int val) { set_bit(index, p->data_state); - p->data[index] = val; + WRITE_ONCE(p->data[index], val); } -#define NEIGH_VAR(p, attr) ((p)->data[NEIGH_VAR_ ## attr]) +#define __NEIGH_VAR(p, attr) ((p)->data[NEIGH_VAR_ ## attr]) +#define NEIGH_VAR(p, attr) READ_ONCE(__NEIGH_VAR(p, attr)) +#define NEIGH_VAR_PTR(p, attr) (&(__NEIGH_VAR(p, attr))) /* In ndo_neigh_setup, NEIGH_VAR_INIT should be used. * In other cases, NEIGH_VAR_SET should be used. */ -#define NEIGH_VAR_INIT(p, attr, val) (NEIGH_VAR(p, attr) = val) +#define NEIGH_VAR_INIT(p, attr, val) (__NEIGH_VAR(p, attr) = val) #define NEIGH_VAR_SET(p, attr, val) neigh_var_set(p, NEIGH_VAR_ ## attr, val) static inline void neigh_parms_data_state_setall(struct neigh_parms *p) @@ -378,6 +380,13 @@ struct net *neigh_parms_net(const struct neigh_parms *parms) unsigned long neigh_rand_reach_time(unsigned long base); +static inline void neigh_set_reach_time(struct neigh_parms *p) +{ + unsigned long base = NEIGH_VAR(p, BASE_REACHABLE_TIME); + + WRITE_ONCE(p->reachable_time, neigh_rand_reach_time(base)); +} + void pneigh_enqueue(struct neigh_table *tbl, struct neigh_parms *p, struct sk_buff *skb); struct pneigh_entry *pneigh_lookup(struct neigh_table *tbl, struct net *net, diff --git a/net/core/neighbour.c b/net/core/neighbour.c index 98428f60731b..5bbebbfcba43 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -995,8 +995,7 @@ static void neigh_periodic_work(struct work_struct *work) WRITE_ONCE(tbl->last_rand, jiffies); list_for_each_entry(p, &tbl->parms_list, list) - p->reachable_time = - neigh_rand_reach_time(NEIGH_VAR(p, BASE_REACHABLE_TIME)); + neigh_set_reach_time(p); } if (atomic_read(&tbl->entries) < READ_ONCE(tbl->gc_thresh1)) @@ -1749,8 +1748,7 @@ struct neigh_parms *neigh_parms_alloc(struct net_device *dev, if (p) { p->tbl = tbl; refcount_set(&p->refcnt, 1); - p->reachable_time = - neigh_rand_reach_time(NEIGH_VAR(p, BASE_REACHABLE_TIME)); + neigh_set_reach_time(p); p->qlen = 0; netdev_hold(dev, &p->dev_tracker, GFP_KERNEL); p->dev = dev; @@ -1810,8 +1808,7 @@ void neigh_table_init(int index, struct neigh_table *tbl) list_add(&tbl->parms.list, &tbl->parms_list); write_pnet(&tbl->parms.net, &init_net); refcount_set(&tbl->parms.refcnt, 1); - tbl->parms.reachable_time = - neigh_rand_reach_time(NEIGH_VAR(&tbl->parms, BASE_REACHABLE_TIME)); + neigh_set_reach_time(&tbl->parms); tbl->parms.qlen = 0; tbl->stats = alloc_percpu(struct neigh_statistics); @@ -2194,7 +2191,7 @@ static int neightbl_fill_parms(struct sk_buff *skb, struct neigh_parms *parms) NEIGH_VAR(parms, MCAST_PROBES)) || nla_put_u32(skb, NDTPA_MCAST_REPROBES, NEIGH_VAR(parms, MCAST_REPROBES)) || - nla_put_msecs(skb, NDTPA_REACHABLE_TIME, parms->reachable_time, + nla_put_msecs(skb, NDTPA_REACHABLE_TIME, READ_ONCE(parms->reachable_time), NDTPA_PAD) || nla_put_msecs(skb, NDTPA_BASE_REACHABLE_TIME, NEIGH_VAR(parms, BASE_REACHABLE_TIME), NDTPA_PAD) || @@ -2475,8 +2472,7 @@ static int neightbl_set(struct sk_buff *skb, struct nlmsghdr *nlh, * only be effective after the next time neigh_periodic_work * decides to recompute it (can be multiple minutes) */ - p->reachable_time = - neigh_rand_reach_time(NEIGH_VAR(p, BASE_REACHABLE_TIME)); + neigh_set_reach_time(p); break; case NDTPA_GC_STALETIME: NEIGH_VAR_SET(p, GC_STALETIME, @@ -3721,8 +3717,7 @@ static int neigh_proc_base_reachable_time(const struct ctl_table *ctl, int write * only be effective after the next time neigh_periodic_work * decides to recompute it */ - p->reachable_time = - neigh_rand_reach_time(NEIGH_VAR(p, BASE_REACHABLE_TIME)); + neigh_set_reach_time(p); } return ret; } diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index f427e41e9c49..59d17b6f06bf 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -1449,7 +1449,7 @@ static enum skb_drop_reason ndisc_router_discovery(struct sk_buff *skb) BASE_REACHABLE_TIME, rtime); NEIGH_VAR_SET(in6_dev->nd_parms, GC_STALETIME, 3 * rtime); - in6_dev->nd_parms->reachable_time = neigh_rand_reach_time(rtime); + neigh_set_reach_time(in6_dev->nd_parms); in6_dev->tstamp = jiffies; send_ifinfo_notify = true; } @@ -1948,9 +1948,9 @@ int ndisc_ifinfo_sysctl_change(const struct ctl_table *ctl, int write, void *buf ret = -1; if (write && ret == 0 && dev && (idev = in6_dev_get(dev)) != NULL) { - if (ctl->data == &NEIGH_VAR(idev->nd_parms, BASE_REACHABLE_TIME)) - idev->nd_parms->reachable_time = - neigh_rand_reach_time(NEIGH_VAR(idev->nd_parms, BASE_REACHABLE_TIME)); + if (ctl->data == NEIGH_VAR_PTR(idev->nd_parms, BASE_REACHABLE_TIME)) + neigh_set_reach_time(idev->nd_parms); + WRITE_ONCE(idev->tstamp, jiffies); inet6_ifinfo_notify(RTM_NEWLINK, idev); in6_dev_put(idev); From 4ae34be500649ec452ac1fc2748958683ad9b55d Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Wed, 22 Oct 2025 05:39:47 +0000 Subject: [PATCH 0227/1381] neighbour: Convert RTM_GETNEIGHTBL to RCU. neightbl_dump_info() calls these functions for each neigh_tables[] entry: 1. neightbl_fill_info() for tbl->parms 2. neightbl_fill_param_info() for tbl->parms_list (except tbl->parms) Both functions rely on the table lock (read_lock_bh(&tbl->lock)) and RTNL is not needed. Let's fetch the table under RCU and convert RTM_GETNEIGHTBL to RCU. Note that the first entry of tbl->parms_list is tbl->parms.list and embedded in neigh_table, so list_next_entry() is safe. Signed-off-by: Kuniyuki Iwashima Reviewed-by: Eric Dumazet Link: https://patch.msgid.link/20251022054004.2514876-4-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/core/neighbour.c | 23 +++++++++-------------- 1 file changed, 9 insertions(+), 14 deletions(-) diff --git a/net/core/neighbour.c b/net/core/neighbour.c index 5bbebbfcba43..a660723476fd 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -2176,7 +2176,7 @@ static int neightbl_fill_parms(struct sk_buff *skb, struct neigh_parms *parms) return -ENOBUFS; if ((parms->dev && - nla_put_u32(skb, NDTPA_IFINDEX, parms->dev->ifindex)) || + nla_put_u32(skb, NDTPA_IFINDEX, READ_ONCE(parms->dev->ifindex))) || nla_put_u32(skb, NDTPA_REFCNT, refcount_read(&parms->refcnt)) || nla_put_u32(skb, NDTPA_QUEUE_LENBYTES, NEIGH_VAR(parms, QUEUE_LEN_BYTES)) || @@ -2228,8 +2228,6 @@ static int neightbl_fill_info(struct sk_buff *skb, struct neigh_table *tbl, return -EMSGSIZE; ndtmsg = nlmsg_data(nlh); - - read_lock_bh(&tbl->lock); ndtmsg->ndtm_family = tbl->family; ndtmsg->ndtm_pad1 = 0; ndtmsg->ndtm_pad2 = 0; @@ -2255,11 +2253,9 @@ static int neightbl_fill_info(struct sk_buff *skb, struct neigh_table *tbl, .ndtc_proxy_qlen = READ_ONCE(tbl->proxy_queue.qlen), }; - rcu_read_lock(); nht = rcu_dereference(tbl->nht); ndc.ndtc_hash_rnd = nht->hash_rnd[0]; ndc.ndtc_hash_mask = ((1 << nht->hash_shift) - 1); - rcu_read_unlock(); if (nla_put(skb, NDTA_CONFIG, sizeof(ndc), &ndc)) goto nla_put_failure; @@ -2297,12 +2293,10 @@ static int neightbl_fill_info(struct sk_buff *skb, struct neigh_table *tbl, if (neightbl_fill_parms(skb, &tbl->parms) < 0) goto nla_put_failure; - read_unlock_bh(&tbl->lock); nlmsg_end(skb, nlh); return 0; nla_put_failure: - read_unlock_bh(&tbl->lock); nlmsg_cancel(skb, nlh); return -EMSGSIZE; } @@ -2321,8 +2315,6 @@ static int neightbl_fill_param_info(struct sk_buff *skb, return -EMSGSIZE; ndtmsg = nlmsg_data(nlh); - - read_lock_bh(&tbl->lock); ndtmsg->ndtm_family = tbl->family; ndtmsg->ndtm_pad1 = 0; ndtmsg->ndtm_pad2 = 0; @@ -2331,11 +2323,9 @@ static int neightbl_fill_param_info(struct sk_buff *skb, neightbl_fill_parms(skb, parms) < 0) goto errout; - read_unlock_bh(&tbl->lock); nlmsg_end(skb, nlh); return 0; errout: - read_unlock_bh(&tbl->lock); nlmsg_cancel(skb, nlh); return -EMSGSIZE; } @@ -2575,10 +2565,12 @@ static int neightbl_dump_info(struct sk_buff *skb, struct netlink_callback *cb) family = ((struct rtgenmsg *)nlmsg_data(nlh))->rtgen_family; + rcu_read_lock(); + for (tidx = 0; tidx < NEIGH_NR_TABLES; tidx++) { struct neigh_parms *p; - tbl = rcu_dereference_rtnl(neigh_tables[tidx]); + tbl = rcu_dereference(neigh_tables[tidx]); if (!tbl) continue; @@ -2592,7 +2584,7 @@ static int neightbl_dump_info(struct sk_buff *skb, struct netlink_callback *cb) nidx = 0; p = list_next_entry(&tbl->parms, list); - list_for_each_entry_from(p, &tbl->parms_list, list) { + list_for_each_entry_from_rcu(p, &tbl->parms_list, list) { if (!net_eq(neigh_parms_net(p), net)) continue; @@ -2612,6 +2604,8 @@ static int neightbl_dump_info(struct sk_buff *skb, struct netlink_callback *cb) neigh_skip = 0; } out: + rcu_read_unlock(); + cb->args[0] = tidx; cb->args[1] = nidx; @@ -3913,7 +3907,8 @@ static const struct rtnl_msg_handler neigh_rtnl_msg_handlers[] __initconst = { {.msgtype = RTM_DELNEIGH, .doit = neigh_delete}, {.msgtype = RTM_GETNEIGH, .doit = neigh_get, .dumpit = neigh_dump_info, .flags = RTNL_FLAG_DOIT_UNLOCKED | RTNL_FLAG_DUMP_UNLOCKED}, - {.msgtype = RTM_GETNEIGHTBL, .dumpit = neightbl_dump_info}, + {.msgtype = RTM_GETNEIGHTBL, .dumpit = neightbl_dump_info, + .flags = RTNL_FLAG_DUMP_UNLOCKED}, {.msgtype = RTM_SETNEIGHTBL, .doit = neightbl_set}, }; From 55a6046b48a8c7d3ea164d9b68902b4ea6930cf3 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Wed, 22 Oct 2025 05:39:48 +0000 Subject: [PATCH 0228/1381] neighbour: Convert RTM_SETNEIGHTBL to RCU. neightbl_set() fetches neigh_tables[] and updates attributes under write_lock_bh(&tbl->lock), so RTNL is not needed. neigh_table_clear() synchronises RCU only, and rcu_dereference_rtnl() protects nothing here. If we released RCU after fetching neigh_tables[], there would be no synchronisation to block neigh_table_clear() further, so RCU is held until the end of the function. Another option would be to protect neigh_tables[] user with SRCU and add synchronize_srcu() in neigh_table_clear(). But, holding RCU should be fine as we hold write_lock_bh() for the rest of neightbl_set() anyway. Let's perform RTM_SETNEIGHTBL under RCU and drop RTNL. Signed-off-by: Kuniyuki Iwashima Reviewed-by: Eric Dumazet Link: https://patch.msgid.link/20251022054004.2514876-5-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/core/neighbour.c | 19 ++++++++++++++----- 1 file changed, 14 insertions(+), 5 deletions(-) diff --git a/net/core/neighbour.c b/net/core/neighbour.c index a660723476fd..6d2164b4d999 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -2362,9 +2362,9 @@ static int neightbl_set(struct sk_buff *skb, struct nlmsghdr *nlh, struct netlink_ext_ack *extack) { struct net *net = sock_net(skb->sk); + struct nlattr *tb[NDTA_MAX + 1]; struct neigh_table *tbl; struct ndtmsg *ndtmsg; - struct nlattr *tb[NDTA_MAX+1]; bool found = false; int err, tidx; @@ -2380,20 +2380,27 @@ static int neightbl_set(struct sk_buff *skb, struct nlmsghdr *nlh, ndtmsg = nlmsg_data(nlh); + rcu_read_lock(); + for (tidx = 0; tidx < NEIGH_NR_TABLES; tidx++) { - tbl = rcu_dereference_rtnl(neigh_tables[tidx]); + tbl = rcu_dereference(neigh_tables[tidx]); if (!tbl) continue; + if (ndtmsg->ndtm_family && tbl->family != ndtmsg->ndtm_family) continue; + if (nla_strcmp(tb[NDTA_NAME], tbl->id) == 0) { found = true; break; } } - if (!found) - return -ENOENT; + if (!found) { + rcu_read_unlock(); + err = -ENOENT; + goto errout; + } /* * We acquire tbl->lock to be nice to the periodic timers and @@ -2519,6 +2526,7 @@ static int neightbl_set(struct sk_buff *skb, struct nlmsghdr *nlh, errout_tbl_lock: write_unlock_bh(&tbl->lock); + rcu_read_unlock(); errout: return err; } @@ -3909,7 +3917,8 @@ static const struct rtnl_msg_handler neigh_rtnl_msg_handlers[] __initconst = { .flags = RTNL_FLAG_DOIT_UNLOCKED | RTNL_FLAG_DUMP_UNLOCKED}, {.msgtype = RTM_GETNEIGHTBL, .dumpit = neightbl_dump_info, .flags = RTNL_FLAG_DUMP_UNLOCKED}, - {.msgtype = RTM_SETNEIGHTBL, .doit = neightbl_set}, + {.msgtype = RTM_SETNEIGHTBL, .doit = neightbl_set, + .flags = RTNL_FLAG_DOIT_UNLOCKED}, }; static int __init neigh_init(void) From 3064d0fe02af23a3956d2b690461abb44da88cf4 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Wed, 22 Oct 2025 05:39:49 +0000 Subject: [PATCH 0229/1381] neighbour: Convert rwlock of struct neigh_table to spinlock. Only neigh_for_each() and neigh_seq_start/stop() are on the reader side of neigh_table.lock. Let's convert rwlock to the plain spinlock. Signed-off-by: Kuniyuki Iwashima Reviewed-by: Eric Dumazet Link: https://patch.msgid.link/20251022054004.2514876-6-kuniyu@google.com Signed-off-by: Jakub Kicinski --- include/net/neighbour.h | 2 +- net/atm/clip.c | 4 +-- net/core/neighbour.c | 68 +++++++++++++++++++++-------------------- net/ipv4/arp.c | 4 +-- 4 files changed, 40 insertions(+), 38 deletions(-) diff --git a/include/net/neighbour.h b/include/net/neighbour.h index 998ff9eccebb..2dfee6d4258a 100644 --- a/include/net/neighbour.h +++ b/include/net/neighbour.h @@ -238,7 +238,7 @@ struct neigh_table { atomic_t gc_entries; struct list_head gc_list; struct list_head managed_list; - rwlock_t lock; + spinlock_t lock; unsigned long last_rand; struct neigh_statistics __percpu *stats; struct neigh_hash_table __rcu *nht; diff --git a/net/atm/clip.c b/net/atm/clip.c index f7a5565e794e..8f152e5fa659 100644 --- a/net/atm/clip.c +++ b/net/atm/clip.c @@ -168,10 +168,10 @@ static int neigh_check_cb(struct neighbour *n) static void idle_timer_check(struct timer_list *unused) { - write_lock(&arp_tbl.lock); + spin_lock(&arp_tbl.lock); __neigh_for_each_release(&arp_tbl, neigh_check_cb); mod_timer(&idle_timer, jiffies + CLIP_CHECK_INTERVAL * HZ); - write_unlock(&arp_tbl.lock); + spin_unlock(&arp_tbl.lock); } static int clip_arp_rcv(struct sk_buff *skb) diff --git a/net/core/neighbour.c b/net/core/neighbour.c index 6d2164b4d999..96a3b1a93252 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -81,7 +81,7 @@ static struct hlist_head *neigh_get_dev_table(struct net_device *dev, int family } /* - Neighbour hash table buckets are protected with rwlock tbl->lock. + Neighbour hash table buckets are protected with tbl->lock. - All the scans/updates to hash buckets MUST be made under this lock. - NOTHING clever should be made under this lock: no callbacks @@ -149,7 +149,7 @@ static void neigh_update_gc_list(struct neighbour *n) { bool on_gc_list, exempt_from_gc; - write_lock_bh(&n->tbl->lock); + spin_lock_bh(&n->tbl->lock); write_lock(&n->lock); if (n->dead) goto out; @@ -172,14 +172,14 @@ static void neigh_update_gc_list(struct neighbour *n) } out: write_unlock(&n->lock); - write_unlock_bh(&n->tbl->lock); + spin_unlock_bh(&n->tbl->lock); } static void neigh_update_managed_list(struct neighbour *n) { bool on_managed_list, add_to_managed; - write_lock_bh(&n->tbl->lock); + spin_lock_bh(&n->tbl->lock); write_lock(&n->lock); if (n->dead) goto out; @@ -193,7 +193,7 @@ static void neigh_update_managed_list(struct neighbour *n) list_add_tail(&n->managed_list, &n->tbl->managed_list); out: write_unlock(&n->lock); - write_unlock_bh(&n->tbl->lock); + spin_unlock_bh(&n->tbl->lock); } static void neigh_update_flags(struct neighbour *neigh, u32 flags, int *notify, @@ -263,7 +263,7 @@ static int neigh_forced_gc(struct neigh_table *tbl) NEIGH_CACHE_STAT_INC(tbl, forced_gc_runs); - write_lock_bh(&tbl->lock); + spin_lock_bh(&tbl->lock); list_for_each_entry_safe(n, tmp, &tbl->gc_list, gc_list) { if (refcount_read(&n->refcnt) == 1) { @@ -292,7 +292,7 @@ static int neigh_forced_gc(struct neigh_table *tbl) WRITE_ONCE(tbl->last_flush, jiffies); unlock: - write_unlock_bh(&tbl->lock); + spin_unlock_bh(&tbl->lock); return shrunk; } @@ -454,23 +454,23 @@ static void neigh_flush_table(struct neigh_table *tbl) void neigh_changeaddr(struct neigh_table *tbl, struct net_device *dev) { - write_lock_bh(&tbl->lock); + spin_lock_bh(&tbl->lock); neigh_flush_dev(tbl, dev, false); - write_unlock_bh(&tbl->lock); + spin_unlock_bh(&tbl->lock); } EXPORT_SYMBOL(neigh_changeaddr); static int __neigh_ifdown(struct neigh_table *tbl, struct net_device *dev, bool skip_perm) { - write_lock_bh(&tbl->lock); + spin_lock_bh(&tbl->lock); if (likely(dev)) { neigh_flush_dev(tbl, dev, skip_perm); } else { DEBUG_NET_WARN_ON_ONCE(skip_perm); neigh_flush_table(tbl); } - write_unlock_bh(&tbl->lock); + spin_unlock_bh(&tbl->lock); pneigh_ifdown(tbl, dev, skip_perm); pneigh_queue_purge(&tbl->proxy_queue, dev ? dev_net(dev) : NULL, @@ -687,7 +687,7 @@ ___neigh_create(struct neigh_table *tbl, const void *pkey, n->confirmed = jiffies - (NEIGH_VAR(n->parms, BASE_REACHABLE_TIME) << 1); - write_lock_bh(&tbl->lock); + spin_lock_bh(&tbl->lock); nht = rcu_dereference_protected(tbl->nht, lockdep_is_held(&tbl->lock)); @@ -722,13 +722,13 @@ ___neigh_create(struct neigh_table *tbl, const void *pkey, hlist_add_head_rcu(&n->dev_list, neigh_get_dev_table(dev, tbl->family)); - write_unlock_bh(&tbl->lock); + spin_unlock_bh(&tbl->lock); neigh_dbg(2, "neigh %p is created\n", n); rc = n; out: return rc; out_tbl_unlock: - write_unlock_bh(&tbl->lock); + spin_unlock_bh(&tbl->lock); out_neigh_release: if (!exempt_from_gc) atomic_dec(&tbl->gc_entries); @@ -982,7 +982,7 @@ static void neigh_periodic_work(struct work_struct *work) NEIGH_CACHE_STAT_INC(tbl, periodic_gc_runs); - write_lock_bh(&tbl->lock); + spin_lock_bh(&tbl->lock); nht = rcu_dereference_protected(tbl->nht, lockdep_is_held(&tbl->lock)); @@ -1036,9 +1036,9 @@ static void neigh_periodic_work(struct work_struct *work) * It's fine to release lock here, even if hash table * grows while we are preempted. */ - write_unlock_bh(&tbl->lock); + spin_unlock_bh(&tbl->lock); cond_resched(); - write_lock_bh(&tbl->lock); + spin_lock_bh(&tbl->lock); nht = rcu_dereference_protected(tbl->nht, lockdep_is_held(&tbl->lock)); } @@ -1049,7 +1049,7 @@ static void neigh_periodic_work(struct work_struct *work) */ queue_delayed_work(system_power_efficient_wq, &tbl->gc_work, NEIGH_VAR(&tbl->parms, BASE_REACHABLE_TIME) >> 1); - write_unlock_bh(&tbl->lock); + spin_unlock_bh(&tbl->lock); } static __inline__ int neigh_max_probes(struct neighbour *n) @@ -1641,12 +1641,12 @@ static void neigh_managed_work(struct work_struct *work) managed_work.work); struct neighbour *neigh; - write_lock_bh(&tbl->lock); + spin_lock_bh(&tbl->lock); list_for_each_entry(neigh, &tbl->managed_list, managed_list) neigh_event_send_probe(neigh, NULL, false); queue_delayed_work(system_power_efficient_wq, &tbl->managed_work, NEIGH_VAR(&tbl->parms, INTERVAL_PROBE_TIME_MS)); - write_unlock_bh(&tbl->lock); + spin_unlock_bh(&tbl->lock); } static void neigh_proxy_process(struct timer_list *t) @@ -1761,9 +1761,9 @@ struct neigh_parms *neigh_parms_alloc(struct net_device *dev, return NULL; } - write_lock_bh(&tbl->lock); + spin_lock_bh(&tbl->lock); list_add_rcu(&p->list, &tbl->parms.list); - write_unlock_bh(&tbl->lock); + spin_unlock_bh(&tbl->lock); neigh_parms_data_state_cleanall(p); } @@ -1783,10 +1783,12 @@ void neigh_parms_release(struct neigh_table *tbl, struct neigh_parms *parms) { if (!parms || parms == &tbl->parms) return; - write_lock_bh(&tbl->lock); + + spin_lock_bh(&tbl->lock); list_del_rcu(&parms->list); parms->dead = 1; - write_unlock_bh(&tbl->lock); + spin_unlock_bh(&tbl->lock); + netdev_put(parms->dev, &parms->dev_tracker); call_rcu(&parms->rcu_head, neigh_rcu_free_parms); } @@ -1835,7 +1837,7 @@ void neigh_table_init(int index, struct neigh_table *tbl) else WARN_ON(tbl->entry_size % NEIGH_PRIV_ALIGN); - rwlock_init(&tbl->lock); + spin_lock_init(&tbl->lock); mutex_init(&tbl->phash_lock); INIT_DEFERRABLE_WORK(&tbl->gc_work, neigh_periodic_work); @@ -1978,10 +1980,10 @@ static int neigh_delete(struct sk_buff *skb, struct nlmsghdr *nlh, err = __neigh_update(neigh, NULL, NUD_FAILED, NEIGH_UPDATE_F_OVERRIDE | NEIGH_UPDATE_F_ADMIN, NETLINK_CB(skb).portid, extack); - write_lock_bh(&tbl->lock); + spin_lock_bh(&tbl->lock); neigh_release(neigh); neigh_remove_one(neigh); - write_unlock_bh(&tbl->lock); + spin_unlock_bh(&tbl->lock); out: return err; @@ -2406,7 +2408,7 @@ static int neightbl_set(struct sk_buff *skb, struct nlmsghdr *nlh, * We acquire tbl->lock to be nice to the periodic timers and * make sure they always see a consistent set of values. */ - write_lock_bh(&tbl->lock); + spin_lock_bh(&tbl->lock); if (tb[NDTA_PARMS]) { struct nlattr *tbp[NDTPA_MAX+1]; @@ -2525,7 +2527,7 @@ static int neightbl_set(struct sk_buff *skb, struct nlmsghdr *nlh, err = 0; errout_tbl_lock: - write_unlock_bh(&tbl->lock); + spin_unlock_bh(&tbl->lock); rcu_read_unlock(); errout: return err; @@ -3125,14 +3127,14 @@ void neigh_for_each(struct neigh_table *tbl, void (*cb)(struct neighbour *, void rcu_read_lock(); nht = rcu_dereference(tbl->nht); - read_lock_bh(&tbl->lock); /* avoid resizes */ + spin_lock_bh(&tbl->lock); /* avoid resizes */ for (chain = 0; chain < (1 << nht->hash_shift); chain++) { struct neighbour *n; neigh_for_each_in_bucket(n, &nht->hash_heads[chain]) cb(n, cookie); } - read_unlock_bh(&tbl->lock); + spin_unlock_bh(&tbl->lock); rcu_read_unlock(); } EXPORT_SYMBOL(neigh_for_each); @@ -3402,7 +3404,7 @@ void *neigh_seq_start(struct seq_file *seq, loff_t *pos, struct neigh_table *tbl rcu_read_lock(); state->nht = rcu_dereference(tbl->nht); - read_lock_bh(&tbl->lock); + spin_lock_bh(&tbl->lock); return *pos ? neigh_get_idx_any(seq, pos) : SEQ_START_TOKEN; } @@ -3442,7 +3444,7 @@ void neigh_seq_stop(struct seq_file *seq, void *v) struct neigh_seq_state *state = seq->private; struct neigh_table *tbl = state->tbl; - read_unlock_bh(&tbl->lock); + spin_unlock_bh(&tbl->lock); rcu_read_unlock(); } EXPORT_SYMBOL(neigh_seq_stop); diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c index 833f2cf97178..f3bfecf8a234 100644 --- a/net/ipv4/arp.c +++ b/net/ipv4/arp.c @@ -1217,10 +1217,10 @@ int arp_invalidate(struct net_device *dev, __be32 ip, bool force) err = neigh_update(neigh, NULL, NUD_FAILED, NEIGH_UPDATE_F_OVERRIDE| NEIGH_UPDATE_F_ADMIN, 0); - write_lock_bh(&tbl->lock); + spin_lock_bh(&tbl->lock); neigh_release(neigh); neigh_remove_one(neigh); - write_unlock_bh(&tbl->lock); + spin_unlock_bh(&tbl->lock); } return err; From 13cb6ac5b506d510d5d57128ff9d99541aae886e Mon Sep 17 00:00:00 2001 From: Alessandro Zanni Date: Thu, 23 Oct 2025 22:53:52 +0200 Subject: [PATCH 0230/1381] selftest: net: prevent use of uninitialized variable MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Fix to avoid the usage of the `ret` variable uninitialized in the following macro expansions. It solves the following warning: In file included from netlink-dumps.c:21: netlink-dumps.c: In function ‘dump_extack’: ../kselftest_harness.h:788:35: warning: ‘ret’ may be used uninitialized [-Wmaybe-uninitialized] 788 | intmax_t __exp_print = (intmax_t)__exp; \ | ^~~~~~~~~~~ ../kselftest_harness.h:631:9: note: in expansion of macro ‘__EXPECT’ 631 | __EXPECT(expected, #expected, seen, #seen, ==, 0) | ^~~~~~~~ netlink-dumps.c:169:9: note: in expansion of macro ‘EXPECT_EQ’ 169 | EXPECT_EQ(ret, FOUND_EXTACK); | ^~~~~~~~~ The issue can be reproduced, building the tests, with the command: make -C tools/testing/selftests TARGETS=net Signed-off-by: Alessandro Zanni Link: https://patch.msgid.link/20251023205354.28249-1-alessandro.zanni87@gmail.com Signed-off-by: Jakub Kicinski --- tools/testing/selftests/net/netlink-dumps.c | 1 + 1 file changed, 1 insertion(+) diff --git a/tools/testing/selftests/net/netlink-dumps.c b/tools/testing/selftests/net/netlink-dumps.c index 7618ebe528a4..679b6c77ace7 100644 --- a/tools/testing/selftests/net/netlink-dumps.c +++ b/tools/testing/selftests/net/netlink-dumps.c @@ -143,6 +143,7 @@ TEST(dump_extack) EXPECT_EQ(n, -1); EXPECT_EQ(errno, ENOBUFS); + ret = NO_CTRL; for (i = 0; i < cnt; i++) { struct ext_ack ea = {}; From d0d2203b9ab71b69c22c8ee1d60c51f9561426a3 Mon Sep 17 00:00:00 2001 From: Julia Lawall Date: Thu, 23 Oct 2025 03:30:51 +0200 Subject: [PATCH 0231/1381] strparser: fix typo in comment The name frags_list doesn't appear in the kernel. It should be frag_list as in the next sentence. Signed-off-by: Julia Lawall Link: https://patch.msgid.link/20251023013051.1728388-1-Julia.Lawall@inria.fr Signed-off-by: Jakub Kicinski --- net/strparser/strparser.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/strparser/strparser.c b/net/strparser/strparser.c index 43b1f558b33d..b929c1cd85e0 100644 --- a/net/strparser/strparser.c +++ b/net/strparser/strparser.c @@ -127,7 +127,7 @@ static int __strp_recv(read_descriptor_t *desc, struct sk_buff *orig_skb, } if (!strp->skb_nextp) { - /* We are going to append to the frags_list of head. + /* We are going to append to the frag_list of head. * Need to unshare the frag_list. */ err = skb_unclone(head, GFP_ATOMIC); From f0773d0b41b492c2ffef92ae74edf65b1d84367e Mon Sep 17 00:00:00 2001 From: Dust Li Date: Thu, 23 Oct 2025 10:00:12 +0800 Subject: [PATCH 0232/1381] smc: rename smc_find_ism_store_rc to reflect broader usage The function smc_find_ism_store_rc() is used to record the reason why a suitable device (either ISM or RDMA) could not be found. However, its name suggests it is ISM-specific, which is misleading. Rename it to better reflect its actual usage. No functional changes. Signed-off-by: Dust Li Reviewed-by: Simon Horman Link: https://patch.msgid.link/20251023020012.69609-1-dust.li@linux.alibaba.com Signed-off-by: Jakub Kicinski --- net/smc/af_smc.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c index 77b99e8ef35a..e9d0e62e0b1b 100644 --- a/net/smc/af_smc.c +++ b/net/smc/af_smc.c @@ -2140,7 +2140,7 @@ static void smc_check_ism_v2_match(struct smc_init_info *ini, } } -static void smc_find_ism_store_rc(u32 rc, struct smc_init_info *ini) +static void smc_init_info_store_rc(u32 rc, struct smc_init_info *ini) { if (!ini->rc) ini->rc = rc; @@ -2203,7 +2203,7 @@ static void smc_find_ism_v2_device_serv(struct smc_sock *new_smc, mutex_unlock(&smcd_dev_list.mutex); if (!ini->ism_dev[0]) { - smc_find_ism_store_rc(SMC_CLC_DECL_NOSMCD2DEV, ini); + smc_init_info_store_rc(SMC_CLC_DECL_NOSMCD2DEV, ini); goto not_found; } @@ -2220,7 +2220,7 @@ static void smc_find_ism_v2_device_serv(struct smc_sock *new_smc, ini->ism_selected = i; rc = smc_listen_ism_init(new_smc, ini); if (rc) { - smc_find_ism_store_rc(rc, ini); + smc_init_info_store_rc(rc, ini); /* try next active ISM device */ continue; } @@ -2260,7 +2260,7 @@ static void smc_find_ism_v1_device_serv(struct smc_sock *new_smc, return; /* V1 ISM device found */ not_found: - smc_find_ism_store_rc(rc, ini); + smc_init_info_store_rc(rc, ini); ini->smcd_version &= ~SMC_V1; ini->ism_dev[0] = NULL; ini->is_smcd = false; @@ -2311,7 +2311,7 @@ static void smc_find_rdma_v2_device_serv(struct smc_sock *new_smc, ini->smcrv2.daddr = smc_ib_gid_to_ipv4(smc_v2_ext->roce); rc = smc_find_rdma_device(new_smc, ini); if (rc) { - smc_find_ism_store_rc(rc, ini); + smc_init_info_store_rc(rc, ini); goto not_found; } if (!ini->smcrv2.uses_gateway) @@ -2328,7 +2328,7 @@ static void smc_find_rdma_v2_device_serv(struct smc_sock *new_smc, if (!rc) return; ini->smcr_version = smcr_version; - smc_find_ism_store_rc(rc, ini); + smc_init_info_store_rc(rc, ini); not_found: ini->smcr_version &= ~SMC_V2; @@ -2375,7 +2375,7 @@ static int smc_listen_find_device(struct smc_sock *new_smc, /* check for matching IP prefix and subnet length (V1) */ prfx_rc = smc_listen_prfx_check(new_smc, pclc); if (prfx_rc) - smc_find_ism_store_rc(prfx_rc, ini); + smc_init_info_store_rc(prfx_rc, ini); /* get vlan id from IP device */ if (smc_vlan_by_tcpsk(new_smc->clcsock, ini)) @@ -2402,7 +2402,7 @@ static int smc_listen_find_device(struct smc_sock *new_smc, int rc; rc = smc_find_rdma_v1_device_serv(new_smc, pclc, ini); - smc_find_ism_store_rc(rc, ini); + smc_init_info_store_rc(rc, ini); return (!rc) ? 0 : ini->rc; } return prfx_rc; From 330ce8ffc1848cbfa3e06c2c22750cfffa115579 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Thu, 23 Oct 2025 10:16:30 +0100 Subject: [PATCH 0233/1381] net: phy: add phy_can_wakeup() Add phy_can_wakeup() to report whether the PHY driver has marked the PHY device as being wake-up capable as far as the driver model is concerned. Reviewed-by: Maxime Chevallier Reviewed-by: Florian Fainelli Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vBrQs-0000000BLzI-0w3U@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- include/linux/phy.h | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/include/linux/phy.h b/include/linux/phy.h index 3c7634482356..3eeeaec52832 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -1379,6 +1379,18 @@ static inline void phy_disable_eee_mode(struct phy_device *phydev, u32 link_mode linkmode_clear_bit(link_mode, phydev->advertising_eee); } +/** + * phy_can_wakeup() - indicate whether PHY has driver model wakeup capabilities + * @phydev: The phy_device struct + * + * Returns: true/false depending on the PHY driver's device_set_wakeup_capable() + * setting. + */ +static inline bool phy_can_wakeup(struct phy_device *phydev) +{ + return device_can_wakeup(&phydev->mdio.dev); +} + void phy_resolve_aneg_pause(struct phy_device *phydev); void phy_resolve_aneg_linkmode(struct phy_device *phydev); From b344bfacf1de2dd776a218ce8341b9c672745a01 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Thu, 23 Oct 2025 10:16:35 +0100 Subject: [PATCH 0234/1381] net: phy: add phy_may_wakeup() Add phy_may_wakeup() which uses the driver model's device_may_wakeup() when the PHY driver has marked the device as wakeup capable in the driver model, otherwise use phy_drv_wol_enabled(). Replace the sites that used to call phy_drv_wol_enabled() with this as checking the driver model will be more efficient than checking the WoL state. Export phy_may_wakeup() so that phylink can use it. Reviewed-by: Maxime Chevallier Reviewed-by: Florian Fainelli Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vBrQx-0000000BLzO-1RLt@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/phy/phy_device.c | 14 ++++++++++++-- include/linux/phy.h | 9 +++++++++ 2 files changed, 21 insertions(+), 2 deletions(-) diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c index 7a67c900e79a..b7feaf0cb1df 100644 --- a/drivers/net/phy/phy_device.c +++ b/drivers/net/phy/phy_device.c @@ -251,6 +251,16 @@ static bool phy_drv_wol_enabled(struct phy_device *phydev) return wol.wolopts != 0; } +bool phy_may_wakeup(struct phy_device *phydev) +{ + /* If the PHY is using driver-model based wakeup, use that state. */ + if (phy_can_wakeup(phydev)) + return device_may_wakeup(&phydev->mdio.dev); + + return phy_drv_wol_enabled(phydev); +} +EXPORT_SYMBOL_GPL(phy_may_wakeup); + static void phy_link_change(struct phy_device *phydev, bool up) { struct net_device *netdev = phydev->attached_dev; @@ -302,7 +312,7 @@ static bool mdio_bus_phy_may_suspend(struct phy_device *phydev) /* If the PHY on the mido bus is not attached but has WOL enabled * we cannot suspend the PHY. */ - if (!netdev && phy_drv_wol_enabled(phydev)) + if (!netdev && phy_may_wakeup(phydev)) return false; /* PHY not attached? May suspend if the PHY has not already been @@ -1909,7 +1919,7 @@ int phy_suspend(struct phy_device *phydev) if (phydev->suspended || !phydrv) return 0; - phydev->wol_enabled = phy_drv_wol_enabled(phydev) || + phydev->wol_enabled = phy_may_wakeup(phydev) || (netdev && netdev->ethtool->wol_enabled); /* If the device has WOL enabled, we cannot suspend the PHY */ if (phydev->wol_enabled && !(phydrv->flags & PHY_ALWAYS_CALL_SUSPEND)) diff --git a/include/linux/phy.h b/include/linux/phy.h index 3eeeaec52832..17a2cdc9f1a0 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -1391,6 +1391,15 @@ static inline bool phy_can_wakeup(struct phy_device *phydev) return device_can_wakeup(&phydev->mdio.dev); } +/** + * phy_may_wakeup() - indicate whether PHY has wakeup enabled + * @phydev: The phy_device struct + * + * Returns: true/false depending on the PHY driver's device_set_wakeup_enabled() + * setting if using the driver model, otherwise the legacy determination. + */ +bool phy_may_wakeup(struct phy_device *phydev); + void phy_resolve_aneg_pause(struct phy_device *phydev); void phy_resolve_aneg_linkmode(struct phy_device *phydev); From b79fbd86c84918790c128e6899b420de4667018e Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Thu, 23 Oct 2025 10:16:40 +0100 Subject: [PATCH 0235/1381] net: phylink: add phylink managed MAC Wake-on-Lan support Add core phylink managed Wake-on-Lan support, which is enabled when the MAC driver fills in the new .mac_wol_set() method that this commit creates. When this feature is disabled, phylink acts as it has in the past, merely passing the ethtool WoL calls to phylib whenever a PHY exists. No other new functionality provided by this commit is enabled. When this feature is enabled, a more inteligent approach is used. Phylink will first pass WoL options to the PHY, read them back, and attempt to set any options that were not set at the PHY at the MAC. Since we have PHY drivers that report they support WoL, and accept WoL configuration even though they aren't wired up to be capable of waking the system, we need a way to differentiate between PHYs that think they support WoL and those which actually do. As PHY drivers do not make use of the driver model's wake-up infrastructure, but could, we use this to determine whether PHY drivers can participate. This gives a path forward where, as MAC drivers are converted to this, it encourages PHY drivers to also be converted. Phylink will also ignore the mac_wol argument to phylink_suspend() as it now knows the WoL state at the MAC. MAC drivers are expected to record/configure the Wake-on-Lan state in their .mac_set_wol() method, and deal appropriately with it in their suspend/resume methods. The driver model provides assistance to set the IRQ wake support which may assist driver authors in achieving the necessary configuration. Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vBrR2-0000000BLzU-1xYL@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/phy/phylink.c | 80 +++++++++++++++++++++++++++++++++++++-- include/linux/phylink.h | 26 +++++++++++++ 2 files changed, 102 insertions(+), 4 deletions(-) diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c index 9d7799ea1c17..bec44ebdf80b 100644 --- a/drivers/net/phy/phylink.c +++ b/drivers/net/phy/phylink.c @@ -93,6 +93,9 @@ struct phylink { u8 sfp_port; struct eee_config eee_cfg; + + u32 wolopts_mac; + u8 wol_sopass[SOPASS_MAX]; }; #define phylink_printk(level, pl, fmt, ...) \ @@ -2562,6 +2565,17 @@ void phylink_rx_clk_stop_unblock(struct phylink *pl) } EXPORT_SYMBOL_GPL(phylink_rx_clk_stop_unblock); +static bool phylink_mac_supports_wol(struct phylink *pl) +{ + return !!pl->mac_ops->mac_wol_set; +} + +static bool phylink_phy_supports_wol(struct phylink *pl, + struct phy_device *phydev) +{ + return phydev && (pl->config->wol_phy_legacy || phy_can_wakeup(phydev)); +} + /** * phylink_suspend() - handle a network device suspend event * @pl: a pointer to a &struct phylink returned from phylink_create() @@ -2575,11 +2589,17 @@ EXPORT_SYMBOL_GPL(phylink_rx_clk_stop_unblock); * can also bring down the link between the MAC and PHY. * - If Wake-on-Lan is active, but being handled by the MAC, the MAC * still needs to receive packets, so we can not bring the link down. + * + * Note: when phylink managed Wake-on-Lan is in use, @mac_wol is ignored. + * (struct phylink_mac_ops.mac_set_wol populated.) */ void phylink_suspend(struct phylink *pl, bool mac_wol) { ASSERT_RTNL(); + if (phylink_mac_supports_wol(pl)) + mac_wol = !!pl->wolopts_mac; + if (mac_wol && (!pl->netdev || pl->netdev->ethtool->wol_enabled)) { /* Wake-on-Lan enabled, MAC handling */ mutex_lock(&pl->state_mutex); @@ -2689,8 +2709,24 @@ void phylink_ethtool_get_wol(struct phylink *pl, struct ethtool_wolinfo *wol) wol->supported = 0; wol->wolopts = 0; - if (pl->phydev) - phy_ethtool_get_wol(pl->phydev, wol); + if (phylink_mac_supports_wol(pl)) { + if (phylink_phy_supports_wol(pl, pl->phydev)) + phy_ethtool_get_wol(pl->phydev, wol); + + /* Where the MAC augments the WoL support, merge its support and + * current configuration. + */ + if (~wol->wolopts & pl->wolopts_mac & WAKE_MAGICSECURE) + memcpy(wol->sopass, pl->wol_sopass, + sizeof(wol->sopass)); + + wol->supported |= pl->config->wol_mac_support; + wol->wolopts |= pl->wolopts_mac; + } else { + /* Legacy */ + if (pl->phydev) + phy_ethtool_get_wol(pl->phydev, wol); + } } EXPORT_SYMBOL_GPL(phylink_ethtool_get_wol); @@ -2707,12 +2743,48 @@ EXPORT_SYMBOL_GPL(phylink_ethtool_get_wol); */ int phylink_ethtool_set_wol(struct phylink *pl, struct ethtool_wolinfo *wol) { + struct ethtool_wolinfo w = { .cmd = ETHTOOL_GWOL }; int ret = -EOPNOTSUPP; + bool changed; + u32 wolopts; ASSERT_RTNL(); - if (pl->phydev) - ret = phy_ethtool_set_wol(pl->phydev, wol); + if (phylink_mac_supports_wol(pl)) { + wolopts = wol->wolopts; + + if (phylink_phy_supports_wol(pl, pl->phydev)) { + ret = phy_ethtool_set_wol(pl->phydev, wol); + if (ret != 0 && ret != -EOPNOTSUPP) + return ret; + + phy_ethtool_get_wol(pl->phydev, &w); + + /* Any Wake-on-Lan modes which the PHY is handling + * should not be passed on to the MAC. + */ + wolopts &= ~w.wolopts; + } + + wolopts &= pl->config->wol_mac_support; + changed = pl->wolopts_mac != wolopts; + if (wolopts & WAKE_MAGICSECURE) + changed |= !!memcmp(wol->sopass, pl->wol_sopass, + sizeof(wol->sopass)); + memcpy(pl->wol_sopass, wol->sopass, sizeof(pl->wol_sopass)); + + if (changed) { + ret = pl->mac_ops->mac_wol_set(pl->config, wolopts, + wol->sopass); + if (!ret) + pl->wolopts_mac = wolopts; + } else { + ret = 0; + } + } else { + if (pl->phydev) + ret = phy_ethtool_set_wol(pl->phydev, wol); + } return ret; } diff --git a/include/linux/phylink.h b/include/linux/phylink.h index 9af0411761d7..59cb58b29d1d 100644 --- a/include/linux/phylink.h +++ b/include/linux/phylink.h @@ -156,6 +156,8 @@ enum phylink_op_type { * @lpi_capabilities: MAC speeds which can support LPI signalling * @lpi_timer_default: Default EEE LPI timer setting. * @eee_enabled_default: If set, EEE will be enabled by phylink at creation time + * @wol_phy_legacy: Use Wake-on-Lan with PHY even if phy_can_wakeup() is false + * @wol_mac_support: Bitmask of MAC supported %WAKE_* options */ struct phylink_config { struct device *dev; @@ -173,6 +175,10 @@ struct phylink_config { unsigned long lpi_capabilities; u32 lpi_timer_default; bool eee_enabled_default; + + /* Wake-on-Lan support */ + bool wol_phy_legacy; + u32 wol_mac_support; }; void phylink_limit_mac_speed(struct phylink_config *config, u32 max_speed); @@ -188,6 +194,7 @@ void phylink_limit_mac_speed(struct phylink_config *config, u32 max_speed); * @mac_link_up: allow the link to come up. * @mac_disable_tx_lpi: disable LPI. * @mac_enable_tx_lpi: enable and configure LPI. + * @mac_wol_set: configure Wake-on-Lan settings at the MAC. * * The individual methods are described more fully below. */ @@ -211,6 +218,9 @@ struct phylink_mac_ops { void (*mac_disable_tx_lpi)(struct phylink_config *config); int (*mac_enable_tx_lpi)(struct phylink_config *config, u32 timer, bool tx_clk_stop); + + int (*mac_wol_set)(struct phylink_config *config, u32 wolopts, + const u8 *sopass); }; #if 0 /* For kernel-doc purposes only. */ @@ -440,6 +450,22 @@ void mac_disable_tx_lpi(struct phylink_config *config); */ int mac_enable_tx_lpi(struct phylink_config *config, u32 timer, bool tx_clk_stop); + +/** + * mac_wol_set() - configure the Wake-on-Lan parameters + * @config: a pointer to a &struct phylink_config. + * @wolopts: Bitmask of %WAKE_* flags for enabled Wake-On-Lan modes. + * @sopass: SecureOn(tm) password; meaningful only for %WAKE_MAGICSECURE + * + * Enable the specified Wake-on-Lan options at the MAC. Options that the + * PHY can handle will have been removed from @wolopts. + * + * The presence of this method enables phylink-managed WoL support. + * + * Returns: 0 on success. + */ +int (*mac_wol_set)(struct phylink_config *config, u32 wolopts, + const u8 *sopass); #endif struct phylink_pcs_ops; From dc1a2a9ce5b2c80e02115ff6fb29b726ad9d7777 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Thu, 23 Oct 2025 10:16:45 +0100 Subject: [PATCH 0236/1381] net: phylink: add phylink managed wake-on-lan PHY speed control Some drivers, e.g. stmmac, use the speed_up()/speed_down() APIs to gain additional power saving during Wake-on-LAN where the PHY is managing the state. Add support to phylink for this, which can be enabled by the MAC driver. Only change the PHY speed if the PHY is configured for wake-up, but without any wake-up on the MAC side, as MAC side means changing the configuration once the negotiation has completed. Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vBrR7-0000000BLza-2PjK@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/phy/phylink.c | 12 ++++++++++++ include/linux/phylink.h | 2 ++ 2 files changed, 14 insertions(+) diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c index bec44ebdf80b..6e1243bf68aa 100644 --- a/drivers/net/phy/phylink.c +++ b/drivers/net/phy/phylink.c @@ -2576,6 +2576,12 @@ static bool phylink_phy_supports_wol(struct phylink *pl, return phydev && (pl->config->wol_phy_legacy || phy_can_wakeup(phydev)); } +static bool phylink_phy_pm_speed_ctrl(struct phylink *pl) +{ + return pl->config->wol_phy_speed_ctrl && !pl->wolopts_mac && + pl->phydev && phy_may_wakeup(pl->phydev); +} + /** * phylink_suspend() - handle a network device suspend event * @pl: a pointer to a &struct phylink returned from phylink_create() @@ -2625,6 +2631,9 @@ void phylink_suspend(struct phylink *pl, bool mac_wol) } else { phylink_stop(pl); } + + if (phylink_phy_pm_speed_ctrl(pl)) + phylink_speed_down(pl, false); } EXPORT_SYMBOL_GPL(phylink_suspend); @@ -2664,6 +2673,9 @@ void phylink_resume(struct phylink *pl) { ASSERT_RTNL(); + if (phylink_phy_pm_speed_ctrl(pl)) + phylink_speed_up(pl); + if (test_bit(PHYLINK_DISABLE_MAC_WOL, &pl->phylink_disable_state)) { /* Wake-on-Lan enabled, MAC handling */ diff --git a/include/linux/phylink.h b/include/linux/phylink.h index 59cb58b29d1d..38363e566ac3 100644 --- a/include/linux/phylink.h +++ b/include/linux/phylink.h @@ -157,6 +157,7 @@ enum phylink_op_type { * @lpi_timer_default: Default EEE LPI timer setting. * @eee_enabled_default: If set, EEE will be enabled by phylink at creation time * @wol_phy_legacy: Use Wake-on-Lan with PHY even if phy_can_wakeup() is false + * @wol_phy_speed_ctrl: Use phy speed control on suspend/resume * @wol_mac_support: Bitmask of MAC supported %WAKE_* options */ struct phylink_config { @@ -178,6 +179,7 @@ struct phylink_config { /* Wake-on-Lan support */ bool wol_phy_legacy; + bool wol_phy_speed_ctrl; u32 wol_mac_support; }; From 6911308d7d111a9c367293b52f2dc265819f2b60 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Thu, 23 Oct 2025 10:16:50 +0100 Subject: [PATCH 0237/1381] net: stmmac: convert to phylink-managed Wake-on-Lan Convert stmmac to use phylink-managed Wake-on-Lan support. To achieve this, we implement the .mac_wol_set() method, which simply configures the driver model's struct device wakeup for stmmac, and sets the priv->wolopts appropriately. When STMMAC_FLAG_USE_PHY_WOL is set, in the stmmac world this means to only use the PHY's WoL support and ignore the MAC's WoL capabilities. To preserve this behaviour, we enable phylink's legacy mode, and avoid telling phylink that the MAC has any WoL support. This achieves the same functionality for this case. When STMMAC_FLAG_USE_PHY_WOL is not set, we provide the MAC's WoL capabilities to phylink, which then allows phylink to choose between the PHY and MAC for WoL depending on their individual capabilities as described in the phylink commit. This only augments the WoL functionality with PHYs that declare to the driver model that they are wake-up capable. Currently, very few PHY drivers support this. Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vBrRC-0000000BLzg-2tA4@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/stmmac.h | 6 +--- .../ethernet/stmicro/stmmac/stmmac_ethtool.c | 34 ++++--------------- .../net/ethernet/stmicro/stmmac/stmmac_main.c | 30 ++++++++++++++-- .../ethernet/stmicro/stmmac/stmmac_platform.c | 4 +-- 4 files changed, 36 insertions(+), 38 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac.h b/drivers/net/ethernet/stmicro/stmmac/stmmac.h index aaeaf42084f0..f128d25346a9 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac.h +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac.h @@ -291,6 +291,7 @@ struct stmmac_priv { int hw_cap_support; int synopsys_id; u32 msg_enable; + /* Our MAC Wake-on-Lan options */ int wolopts; int wol_irq; u32 gmii_address_bus_config; @@ -379,11 +380,6 @@ enum stmmac_state { extern const struct dev_pm_ops stmmac_simple_pm_ops; -static inline bool stmmac_wol_enabled_mac(struct stmmac_priv *priv) -{ - return priv->plat->pmt && device_may_wakeup(priv->device); -} - static inline bool stmmac_wol_enabled_phy(struct stmmac_priv *priv) { return !priv->plat->pmt && device_may_wakeup(priv->device); diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c index df016c4eb710..08b570bc60c7 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c @@ -724,41 +724,19 @@ static void stmmac_get_wol(struct net_device *dev, struct ethtool_wolinfo *wol) { struct stmmac_priv *priv = netdev_priv(dev); - if (!priv->plat->pmt) - return phylink_ethtool_get_wol(priv->phylink, wol); - - mutex_lock(&priv->lock); - if (device_can_wakeup(priv->device)) { - wol->supported = WAKE_MAGIC | WAKE_UCAST; - if (priv->hw_cap_support && !priv->dma_cap.pmt_magic_frame) - wol->supported &= ~WAKE_MAGIC; - wol->wolopts = priv->wolopts; - } - mutex_unlock(&priv->lock); + return phylink_ethtool_get_wol(priv->phylink, wol); } static int stmmac_set_wol(struct net_device *dev, struct ethtool_wolinfo *wol) { struct stmmac_priv *priv = netdev_priv(dev); + int ret; - if (!device_can_wakeup(priv->device)) - return -EOPNOTSUPP; + ret = phylink_ethtool_set_wol(priv->phylink, wol); + if (!ret) + device_set_wakeup_enable(priv->device, !!wol->wolopts); - if (!priv->plat->pmt) { - int ret = phylink_ethtool_set_wol(priv->phylink, wol); - - if (!ret) - device_set_wakeup_enable(priv->device, !!wol->wolopts); - return ret; - } - - device_set_wakeup_enable(priv->device, !!wol->wolopts); - - mutex_lock(&priv->lock); - priv->wolopts = wol->wolopts; - mutex_unlock(&priv->lock); - - return 0; + return ret; } static int stmmac_ethtool_op_get_eee(struct net_device *dev, diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index 9fa3c221a0c3..af4eb94f0f4f 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -1073,6 +1073,20 @@ static int stmmac_mac_enable_tx_lpi(struct phylink_config *config, u32 timer, return 0; } +static int stmmac_mac_wol_set(struct phylink_config *config, u32 wolopts, + const u8 *sopass) +{ + struct stmmac_priv *priv = netdev_priv(to_net_dev(config->dev)); + + device_set_wakeup_enable(priv->device, !!wolopts); + + mutex_lock(&priv->lock); + priv->wolopts = wolopts; + mutex_unlock(&priv->lock); + + return 0; +} + static const struct phylink_mac_ops stmmac_phylink_mac_ops = { .mac_get_caps = stmmac_mac_get_caps, .mac_select_pcs = stmmac_mac_select_pcs, @@ -1082,6 +1096,7 @@ static const struct phylink_mac_ops stmmac_phylink_mac_ops = { .mac_link_up = stmmac_mac_link_up, .mac_disable_tx_lpi = stmmac_mac_disable_tx_lpi, .mac_enable_tx_lpi = stmmac_mac_enable_tx_lpi, + .mac_wol_set = stmmac_mac_wol_set, }; /** @@ -1266,6 +1281,15 @@ static int stmmac_phylink_setup(struct stmmac_priv *priv) config->eee_enabled_default = true; } + if (priv->plat->flags & STMMAC_FLAG_USE_PHY_WOL) { + config->wol_phy_legacy = true; + } else { + if (priv->dma_cap.pmt_remote_wake_up) + config->wol_mac_support |= WAKE_UCAST; + if (priv->dma_cap.pmt_magic_frame) + config->wol_mac_support |= WAKE_MAGIC; + } + fwnode = priv->plat->port_node; if (!fwnode) fwnode = dev_fwnode(priv->device); @@ -7760,7 +7784,7 @@ int stmmac_suspend(struct device *dev) priv->plat->serdes_powerdown(ndev, priv->plat->bsp_priv); /* Enable Power down mode by programming the PMT regs */ - if (stmmac_wol_enabled_mac(priv)) { + if (priv->wolopts) { stmmac_pmt(priv, priv->hw, priv->wolopts); priv->irq_wake = 1; } else { @@ -7774,7 +7798,7 @@ int stmmac_suspend(struct device *dev) if (stmmac_wol_enabled_phy(priv)) phylink_speed_down(priv->phylink, false); - phylink_suspend(priv->phylink, stmmac_wol_enabled_mac(priv)); + phylink_suspend(priv->phylink, !!priv->wolopts); rtnl_unlock(); if (stmmac_fpe_supported(priv)) @@ -7850,7 +7874,7 @@ int stmmac_resume(struct device *dev) * this bit because it can generate problems while resuming * from another devices (e.g. serial console). */ - if (stmmac_wol_enabled_mac(priv)) { + if (priv->wolopts) { mutex_lock(&priv->lock); stmmac_pmt(priv, priv->hw, 0); mutex_unlock(&priv->lock); diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c index fbb92cc6ab59..6483d52b4c0f 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c @@ -969,7 +969,7 @@ static int __maybe_unused stmmac_pltfr_noirq_suspend(struct device *dev) if (!netif_running(ndev)) return 0; - if (!stmmac_wol_enabled_mac(priv)) { + if (!priv->wolopts) { /* Disable clock in case of PWM is off */ clk_disable_unprepare(priv->plat->clk_ptp_ref); @@ -990,7 +990,7 @@ static int __maybe_unused stmmac_pltfr_noirq_resume(struct device *dev) if (!netif_running(ndev)) return 0; - if (!stmmac_wol_enabled_mac(priv)) { + if (!priv->wolopts) { /* enable the clk previously disabled */ ret = pm_runtime_force_resume(dev); if (ret) From d65cb2e27e6e18d036276905c1aa11828aac5b6a Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Thu, 23 Oct 2025 10:16:55 +0100 Subject: [PATCH 0238/1381] net: stmmac: convert to phylink managed WoL PHY speed Convert stmmac to use phylink's management of the PHY speed when Wake-on-Lan is enabled. Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vBrRH-0000000BLzm-3JjF@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/stmmac.h | 5 ----- .../net/ethernet/stmicro/stmmac/stmmac_ethtool.c | 7 +------ drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 15 +-------------- 3 files changed, 2 insertions(+), 25 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac.h b/drivers/net/ethernet/stmicro/stmmac/stmmac.h index f128d25346a9..d5af9344dfb0 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac.h +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac.h @@ -380,11 +380,6 @@ enum stmmac_state { extern const struct dev_pm_ops stmmac_simple_pm_ops; -static inline bool stmmac_wol_enabled_phy(struct stmmac_priv *priv) -{ - return !priv->plat->pmt && device_may_wakeup(priv->device); -} - int stmmac_mdio_unregister(struct net_device *ndev); int stmmac_mdio_register(struct net_device *ndev); int stmmac_mdio_reset(struct mii_bus *mii); diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c index 08b570bc60c7..b155e71aac51 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c @@ -730,13 +730,8 @@ static void stmmac_get_wol(struct net_device *dev, struct ethtool_wolinfo *wol) static int stmmac_set_wol(struct net_device *dev, struct ethtool_wolinfo *wol) { struct stmmac_priv *priv = netdev_priv(dev); - int ret; - ret = phylink_ethtool_set_wol(priv->phylink, wol); - if (!ret) - device_set_wakeup_enable(priv->device, !!wol->wolopts); - - return ret; + return phylink_ethtool_set_wol(priv->phylink, wol); } static int stmmac_ethtool_op_get_eee(struct net_device *dev, diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index af4eb94f0f4f..fd5106880192 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -1205,14 +1205,6 @@ static int stmmac_init_phy(struct net_device *dev) phylink_ethtool_set_eee(priv->phylink, &eee); } - if (!priv->plat->pmt) { - struct ethtool_wolinfo wol = { .cmd = ETHTOOL_GWOL }; - - phylink_ethtool_get_wol(priv->phylink, &wol); - device_set_wakeup_capable(priv->device, !!wol.supported); - device_set_wakeup_enable(priv->device, !!wol.wolopts); - } - return 0; } @@ -1281,6 +1273,7 @@ static int stmmac_phylink_setup(struct stmmac_priv *priv) config->eee_enabled_default = true; } + config->wol_phy_speed_ctrl = true; if (priv->plat->flags & STMMAC_FLAG_USE_PHY_WOL) { config->wol_phy_legacy = true; } else { @@ -7795,9 +7788,6 @@ int stmmac_suspend(struct device *dev) mutex_unlock(&priv->lock); rtnl_lock(); - if (stmmac_wol_enabled_phy(priv)) - phylink_speed_down(priv->phylink, false); - phylink_suspend(priv->phylink, !!priv->wolopts); rtnl_unlock(); @@ -7936,9 +7926,6 @@ int stmmac_resume(struct device *dev) * workqueue thread, which will race with initialisation. */ phylink_resume(priv->phylink); - if (stmmac_wol_enabled_phy(priv)) - phylink_speed_up(priv->phylink); - rtnl_unlock(); netif_device_attach(ndev); From 442a8c68f083f267c0f52526f1cd16988837ec0f Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Thu, 23 Oct 2025 10:46:20 +0100 Subject: [PATCH 0239/1381] net: stmmac: add stmmac_mac_irq_modify() Add a function to allow interrupts to be enabled and disabled in a core independent manner. Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vBrtk-0000000BMYm-3CV5@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/common.h | 5 +++++ .../ethernet/stmicro/stmmac/dwmac1000_core.c | 15 +++++++++++++++ .../net/ethernet/stmicro/stmmac/dwmac4_core.c | 17 +++++++++++++++++ .../net/ethernet/stmicro/stmmac/dwxgmac2_core.c | 16 ++++++++++++++++ drivers/net/ethernet/stmicro/stmmac/hwif.c | 2 ++ drivers/net/ethernet/stmicro/stmmac/hwif.h | 4 ++++ .../net/ethernet/stmicro/stmmac/stmmac_fpe.c | 3 +++ 7 files changed, 62 insertions(+) diff --git a/drivers/net/ethernet/stmicro/stmmac/common.h b/drivers/net/ethernet/stmicro/stmmac/common.h index 31254ba525d5..553a8897b005 100644 --- a/drivers/net/ethernet/stmicro/stmmac/common.h +++ b/drivers/net/ethernet/stmicro/stmmac/common.h @@ -611,6 +611,11 @@ struct mac_device_info { u8 vlan_fail_q; bool hw_vlan_en; bool reverse_sgmii_enable; + + /* This spinlock protects read-modify-write of the interrupt + * mask/enable registers. + */ + spinlock_t irq_ctrl_lock; }; struct stmmac_rx_routing { diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c index 571e48362444..2ca94bfd3f71 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c @@ -61,6 +61,20 @@ static void dwmac1000_core_init(struct mac_device_info *hw, #endif } +static void dwmac1000_irq_modify(struct mac_device_info *hw, u32 disable, + u32 enable) +{ + void __iomem *int_mask = hw->pcsr + GMAC_INT_MASK; + unsigned long flags; + u32 value; + + spin_lock_irqsave(&hw->irq_ctrl_lock, flags); + value = readl(int_mask) | disable; + value &= ~enable; + writel(value, int_mask); + spin_unlock_irqrestore(&hw->irq_ctrl_lock, flags); +} + static int dwmac1000_rx_ipc_enable(struct mac_device_info *hw) { void __iomem *ioaddr = hw->pcsr; @@ -445,6 +459,7 @@ static void dwmac1000_set_mac_loopback(void __iomem *ioaddr, bool enable) const struct stmmac_ops dwmac1000_ops = { .pcs_init = dwmac1000_pcs_init, .core_init = dwmac1000_core_init, + .irq_modify = dwmac1000_irq_modify, .set_mac = stmmac_set_mac, .rx_ipc = dwmac1000_rx_ipc_enable, .dump_regs = dwmac1000_dump_regs, diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c index 0b785389b7ef..6269407d70cd 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c @@ -57,6 +57,20 @@ static void dwmac4_core_init(struct mac_device_info *hw, init_waitqueue_head(&priv->tstamp_busy_wait); } +static void dwmac4_irq_modify(struct mac_device_info *hw, u32 disable, + u32 enable) +{ + void __iomem *int_mask = hw->pcsr + GMAC_INT_EN; + unsigned long flags; + u32 value; + + spin_lock_irqsave(&hw->irq_ctrl_lock, flags); + value = readl(int_mask) & ~disable; + value |= enable; + writel(value, int_mask); + spin_unlock_irqrestore(&hw->irq_ctrl_lock, flags); +} + static void dwmac4_update_caps(struct stmmac_priv *priv) { if (priv->plat->tx_queues_to_use > 1) @@ -885,6 +899,7 @@ static int dwmac4_config_l4_filter(struct mac_device_info *hw, u32 filter_no, const struct stmmac_ops dwmac4_ops = { .pcs_init = dwmac4_pcs_init, .core_init = dwmac4_core_init, + .irq_modify = dwmac4_irq_modify, .update_caps = dwmac4_update_caps, .set_mac = stmmac_set_mac, .rx_ipc = dwmac4_rx_ipc_enable, @@ -920,6 +935,7 @@ const struct stmmac_ops dwmac4_ops = { const struct stmmac_ops dwmac410_ops = { .pcs_init = dwmac4_pcs_init, .core_init = dwmac4_core_init, + .irq_modify = dwmac4_irq_modify, .update_caps = dwmac4_update_caps, .set_mac = stmmac_dwmac4_set_mac, .rx_ipc = dwmac4_rx_ipc_enable, @@ -957,6 +973,7 @@ const struct stmmac_ops dwmac410_ops = { const struct stmmac_ops dwmac510_ops = { .pcs_init = dwmac4_pcs_init, .core_init = dwmac4_core_init, + .irq_modify = dwmac4_irq_modify, .update_caps = dwmac4_update_caps, .set_mac = stmmac_dwmac4_set_mac, .rx_ipc = dwmac4_rx_ipc_enable, diff --git a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c index 0430af27da40..b40b3ea50e25 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c @@ -28,6 +28,20 @@ static void dwxgmac2_core_init(struct mac_device_info *hw, writel(XGMAC_INT_DEFAULT_EN, ioaddr + XGMAC_INT_EN); } +static void dwxgmac2_irq_modify(struct mac_device_info *hw, u32 disable, + u32 enable) +{ + void __iomem *int_mask = hw->pcsr + XGMAC_INT_EN; + unsigned long flags; + u32 value; + + spin_lock_irqsave(&hw->irq_ctrl_lock, flags); + value = readl(int_mask) & ~disable; + value |= enable; + writel(value, int_mask); + spin_unlock_irqrestore(&hw->irq_ctrl_lock, flags); +} + static void dwxgmac2_update_caps(struct stmmac_priv *priv) { if (!priv->dma_cap.mbps_10_100) @@ -1411,6 +1425,7 @@ static void dwxgmac2_set_arp_offload(struct mac_device_info *hw, bool en, const struct stmmac_ops dwxgmac210_ops = { .core_init = dwxgmac2_core_init, + .irq_modify = dwxgmac2_irq_modify, .update_caps = dwxgmac2_update_caps, .set_mac = dwxgmac2_set_mac, .rx_ipc = dwxgmac2_rx_ipc, @@ -1466,6 +1481,7 @@ static void dwxlgmac2_rx_queue_enable(struct mac_device_info *hw, u8 mode, const struct stmmac_ops dwxlgmac2_ops = { .core_init = dwxgmac2_core_init, + .irq_modify = dwxgmac2_irq_modify, .set_mac = dwxgmac2_set_mac, .rx_ipc = dwxgmac2_rx_ipc, .rx_queue_enable = dwxlgmac2_rx_queue_enable, diff --git a/drivers/net/ethernet/stmicro/stmmac/hwif.c b/drivers/net/ethernet/stmicro/stmmac/hwif.c index 00083ce52549..41a7e1841227 100644 --- a/drivers/net/ethernet/stmicro/stmmac/hwif.c +++ b/drivers/net/ethernet/stmicro/stmmac/hwif.c @@ -333,6 +333,8 @@ int stmmac_hwif_init(struct stmmac_priv *priv) if (!mac) return -ENOMEM; + spin_lock_init(&mac->irq_ctrl_lock); + /* Fallback to generic HW */ for (i = ARRAY_SIZE(stmmac_hw) - 1; i >= 0; i--) { entry = &stmmac_hw[i]; diff --git a/drivers/net/ethernet/stmicro/stmmac/hwif.h b/drivers/net/ethernet/stmicro/stmmac/hwif.h index 82cfb6bec334..cb8fc09caf86 100644 --- a/drivers/net/ethernet/stmicro/stmmac/hwif.h +++ b/drivers/net/ethernet/stmicro/stmmac/hwif.h @@ -319,6 +319,8 @@ struct stmmac_ops { void (*core_init)(struct mac_device_info *hw, struct net_device *dev); /* Update MAC capabilities */ void (*update_caps)(struct stmmac_priv *priv); + /* Change the interrupt enable setting. Enable takes precedence. */ + void (*irq_modify)(struct mac_device_info *hw, u32 disable, u32 enable); /* Enable the MAC RX/TX */ void (*set_mac)(void __iomem *ioaddr, bool enable); /* Enable and verify that the IPC module is supported */ @@ -421,6 +423,8 @@ struct stmmac_ops { stmmac_do_void_callback(__priv, mac, core_init, __args) #define stmmac_mac_update_caps(__priv) \ stmmac_do_void_callback(__priv, mac, update_caps, __priv) +#define stmmac_mac_irq_modify(__priv, __args...) \ + stmmac_do_void_callback(__priv, mac, irq_modify, (__priv)->hw, __args) #define stmmac_mac_set(__priv, __args...) \ stmmac_do_void_callback(__priv, mac, set_mac, __args) #define stmmac_rx_ipc(__priv, __args...) \ diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_fpe.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_fpe.c index 75b470ee621a..c54c70224351 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_fpe.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_fpe.c @@ -70,8 +70,10 @@ static void stmmac_fpe_configure_pmac(struct ethtool_mmsv *mmsv, bool pmac_enabl struct stmmac_priv *priv = container_of(cfg, struct stmmac_priv, fpe_cfg); const struct stmmac_fpe_reg *reg = cfg->reg; void __iomem *ioaddr = priv->ioaddr; + unsigned long flags; u32 value; + spin_lock_irqsave(&priv->hw->irq_ctrl_lock, flags); value = readl(ioaddr + reg->int_en_reg); if (pmac_enable) { @@ -86,6 +88,7 @@ static void stmmac_fpe_configure_pmac(struct ethtool_mmsv *mmsv, bool pmac_enabl } writel(value, ioaddr + reg->int_en_reg); + spin_unlock_irqrestore(&priv->hw->irq_ctrl_lock, flags); } static void stmmac_fpe_send_mpacket(struct ethtool_mmsv *mmsv, From eed68edac508fed36e3726f5b0b828a83efab7f8 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Thu, 23 Oct 2025 10:46:25 +0100 Subject: [PATCH 0240/1381] net: stmmac: add support for controlling PCS interrupts Add support to the PCS instance for controlling the PCS interrupts depending on whether the PCS is used. Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vBrtp-0000000BMYs-3bhI@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../net/ethernet/stmicro/stmmac/dwmac1000.h | 7 +++--- .../ethernet/stmicro/stmmac/dwmac1000_core.c | 11 ++++------ drivers/net/ethernet/stmicro/stmmac/dwmac4.h | 2 -- .../net/ethernet/stmicro/stmmac/dwmac4_core.c | 10 +++------ .../net/ethernet/stmicro/stmmac/stmmac_pcs.c | 22 ++++++++++++++++++- .../net/ethernet/stmicro/stmmac/stmmac_pcs.h | 4 +++- 6 files changed, 34 insertions(+), 22 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac1000.h b/drivers/net/ethernet/stmicro/stmmac/dwmac1000.h index 8f3002d9de78..697bba641e05 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac1000.h +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac1000.h @@ -38,11 +38,10 @@ #define GMAC_INT_DISABLE_PCSAN BIT(2) #define GMAC_INT_DISABLE_PMT BIT(3) #define GMAC_INT_DISABLE_TIMESTAMP BIT(9) -#define GMAC_INT_DISABLE_PCS (GMAC_INT_DISABLE_PCSLINK | \ - GMAC_INT_DISABLE_PCSAN) #define GMAC_INT_DEFAULT_MASK (GMAC_INT_DISABLE_RGMII | \ - GMAC_INT_DISABLE_TIMESTAMP | \ - GMAC_INT_DISABLE_PCS) + GMAC_INT_DISABLE_PCSLINK | \ + GMAC_INT_DISABLE_PCSAN | \ + GMAC_INT_DISABLE_TIMESTAMP) /* PMT Control and Status */ #define GMAC_PMT 0x0000002c diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c index 2ca94bfd3f71..a2ae136d2c0e 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c @@ -27,7 +27,9 @@ static int dwmac1000_pcs_init(struct stmmac_priv *priv) if (!priv->dma_cap.pcs) return 0; - return stmmac_integrated_pcs_init(priv, GMAC_PCS_BASE); + return stmmac_integrated_pcs_init(priv, GMAC_PCS_BASE, + GMAC_INT_DISABLE_PCSLINK | + GMAC_INT_DISABLE_PCSAN); } static void dwmac1000_core_init(struct mac_device_info *hw, @@ -48,12 +50,7 @@ static void dwmac1000_core_init(struct mac_device_info *hw, writel(value | GMAC_CORE_INIT, ioaddr + GMAC_CONTROL); /* Mask GMAC interrupts */ - value = GMAC_INT_DEFAULT_MASK; - - if (hw->pcs) - value &= ~GMAC_INT_DISABLE_PCS; - - writel(value, ioaddr + GMAC_INT_MASK); + writel(GMAC_INT_DEFAULT_MASK, ioaddr + GMAC_INT_MASK); #ifdef STMMAC_VLAN_TAG_USED /* Tag detection without filtering */ diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4.h b/drivers/net/ethernet/stmicro/stmmac/dwmac4.h index 6dd84b6544cc..3cb733781e1e 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac4.h +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4.h @@ -106,8 +106,6 @@ #define GMAC_INT_LPI_EN BIT(5) #define GMAC_INT_TSIE BIT(12) -#define GMAC_PCS_IRQ_DEFAULT (GMAC_INT_PCS_LINK | GMAC_INT_PCS_ANE) - #define GMAC_INT_DEFAULT_ENABLE (GMAC_INT_PMT_EN | GMAC_INT_LPI_EN | \ GMAC_INT_TSIE) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c index 6269407d70cd..a4282fd7c3c7 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c @@ -27,7 +27,8 @@ static int dwmac4_pcs_init(struct stmmac_priv *priv) if (!priv->dma_cap.pcs) return 0; - return stmmac_integrated_pcs_init(priv, GMAC_PCS_BASE); + return stmmac_integrated_pcs_init(priv, GMAC_PCS_BASE, + GMAC_INT_PCS_LINK | GMAC_INT_PCS_ANE); } static void dwmac4_core_init(struct mac_device_info *hw, @@ -46,12 +47,7 @@ static void dwmac4_core_init(struct mac_device_info *hw, writel((clk_rate / 1000000) - 1, ioaddr + GMAC4_MAC_ONEUS_TIC_COUNTER); /* Enable GMAC interrupts */ - value = GMAC_INT_DEFAULT_ENABLE; - - if (hw->pcs) - value |= GMAC_PCS_IRQ_DEFAULT; - - writel(value, ioaddr + GMAC_INT_EN); + writel(GMAC_INT_DEFAULT_ENABLE, ioaddr + GMAC_INT_EN); if (GMAC_INT_DEFAULT_ENABLE & GMAC_INT_TSIE) init_waitqueue_head(&priv->tstamp_busy_wait); diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.c index 50ea51d7a1cc..e2f531c11986 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.c @@ -2,6 +2,22 @@ #include "stmmac.h" #include "stmmac_pcs.h" +static int dwmac_integrated_pcs_enable(struct phylink_pcs *pcs) +{ + struct stmmac_pcs *spcs = phylink_pcs_to_stmmac_pcs(pcs); + + stmmac_mac_irq_modify(spcs->priv, 0, spcs->int_mask); + + return 0; +} + +static void dwmac_integrated_pcs_disable(struct phylink_pcs *pcs) +{ + struct stmmac_pcs *spcs = phylink_pcs_to_stmmac_pcs(pcs); + + stmmac_mac_irq_modify(spcs->priv, spcs->int_mask, 0); +} + static void dwmac_integrated_pcs_get_state(struct phylink_pcs *pcs, unsigned int neg_mode, struct phylink_link_state *state) @@ -23,11 +39,14 @@ static int dwmac_integrated_pcs_config(struct phylink_pcs *pcs, } static const struct phylink_pcs_ops dwmac_integrated_pcs_ops = { + .pcs_enable = dwmac_integrated_pcs_enable, + .pcs_disable = dwmac_integrated_pcs_disable, .pcs_get_state = dwmac_integrated_pcs_get_state, .pcs_config = dwmac_integrated_pcs_config, }; -int stmmac_integrated_pcs_init(struct stmmac_priv *priv, unsigned int offset) +int stmmac_integrated_pcs_init(struct stmmac_priv *priv, unsigned int offset, + u32 int_mask) { struct stmmac_pcs *spcs; @@ -37,6 +56,7 @@ int stmmac_integrated_pcs_init(struct stmmac_priv *priv, unsigned int offset) spcs->priv = priv; spcs->base = priv->ioaddr + offset; + spcs->int_mask = int_mask; spcs->pcs.ops = &dwmac_integrated_pcs_ops; __set_bit(PHY_INTERFACE_MODE_SGMII, spcs->pcs.supported_interfaces); diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.h b/drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.h index 64397ac8ecab..cda93894168e 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.h +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.h @@ -52,6 +52,7 @@ struct stmmac_priv; struct stmmac_pcs { struct stmmac_priv *priv; void __iomem *base; + u32 int_mask; struct phylink_pcs pcs; }; @@ -61,7 +62,8 @@ phylink_pcs_to_stmmac_pcs(struct phylink_pcs *pcs) return container_of(pcs, struct stmmac_pcs, pcs); } -int stmmac_integrated_pcs_init(struct stmmac_priv *priv, unsigned int offset); +int stmmac_integrated_pcs_init(struct stmmac_priv *priv, unsigned int offset, + u32 int_mask); /** * dwmac_pcs_isr - TBI, RTBI, or SGMII PHY ISR From c09b183dc14e0fda14bb99b31b9637ce2e1d3682 Mon Sep 17 00:00:00 2001 From: Oliver Neukum Date: Thu, 23 Oct 2025 12:00:19 +0200 Subject: [PATCH 0241/1381] net: usb: usbnet: coding style for functions Functions are not to have blanks between names and parameter lists. Remove them. Signed-off-by: Oliver Neukum Link: https://patch.msgid.link/20251023100136.909118-1-oneukum@suse.com Signed-off-by: Jakub Kicinski --- drivers/net/usb/usbnet.c | 49 ++++++++++++++++++++-------------------- 1 file changed, 24 insertions(+), 25 deletions(-) diff --git a/drivers/net/usb/usbnet.c b/drivers/net/usb/usbnet.c index bf01f2728531..62a85dbad31a 100644 --- a/drivers/net/usb/usbnet.c +++ b/drivers/net/usb/usbnet.c @@ -189,7 +189,7 @@ static bool usbnet_needs_usb_name_format(struct usbnet *dev, struct net_device * is_local_ether_addr(net->dev_addr)); } -static void intr_complete (struct urb *urb) +static void intr_complete(struct urb *urb) { struct usbnet *dev = urb->context; int status = urb->status; @@ -221,7 +221,7 @@ static void intr_complete (struct urb *urb) "intr resubmit --> %d\n", status); } -static int init_status (struct usbnet *dev, struct usb_interface *intf) +static int init_status(struct usbnet *dev, struct usb_interface *intf) { char *buf = NULL; unsigned pipe = 0; @@ -326,7 +326,7 @@ static void __usbnet_status_stop_force(struct usbnet *dev) * Some link protocols batch packets, so their rx_fixup paths * can return clones as well as just modify the original skb. */ -void usbnet_skb_return (struct usbnet *dev, struct sk_buff *skb) +void usbnet_skb_return(struct usbnet *dev, struct sk_buff *skb) { struct pcpu_sw_netstats *stats64 = this_cpu_ptr(dev->net->tstats); unsigned long flags; @@ -396,7 +396,7 @@ EXPORT_SYMBOL_GPL(usbnet_update_max_qlen); * *-------------------------------------------------------------------------*/ -int usbnet_change_mtu (struct net_device *net, int new_mtu) +int usbnet_change_mtu(struct net_device *net, int new_mtu) { struct usbnet *dev = netdev_priv(net); int ll_mtu = new_mtu + net->hard_header_len; @@ -472,7 +472,7 @@ static enum skb_state defer_bh(struct usbnet *dev, struct sk_buff *skb, * NOTE: annoying asymmetry: if it's active, schedule_work() fails, * but tasklet_schedule() doesn't. hope the failure is rare. */ -void usbnet_defer_kevent (struct usbnet *dev, int work) +void usbnet_defer_kevent(struct usbnet *dev, int work) { set_bit (work, &dev->flags); if (!usbnet_going_away(dev)) { @@ -489,9 +489,9 @@ EXPORT_SYMBOL_GPL(usbnet_defer_kevent); /*-------------------------------------------------------------------------*/ -static void rx_complete (struct urb *urb); +static void rx_complete(struct urb *urb); -static int rx_submit (struct usbnet *dev, struct urb *urb, gfp_t flags) +static int rx_submit(struct usbnet *dev, struct urb *urb, gfp_t flags) { struct sk_buff *skb; struct skb_data *entry; @@ -597,7 +597,7 @@ static inline int rx_process(struct usbnet *dev, struct sk_buff *skb) /*-------------------------------------------------------------------------*/ -static void rx_complete (struct urb *urb) +static void rx_complete(struct urb *urb) { struct sk_buff *skb = (struct sk_buff *) urb->context; struct skb_data *entry = (struct skb_data *) skb->cb; @@ -728,7 +728,7 @@ EXPORT_SYMBOL_GPL(usbnet_purge_paused_rxq); // unlink pending rx/tx; completion handlers do all other cleanup -static int unlink_urbs (struct usbnet *dev, struct sk_buff_head *q) +static int unlink_urbs(struct usbnet *dev, struct sk_buff_head *q) { unsigned long flags; struct sk_buff *skb; @@ -823,7 +823,7 @@ static void usbnet_terminate_urbs(struct usbnet *dev) remove_wait_queue(&dev->wait, &wait); } -int usbnet_stop (struct net_device *net) +int usbnet_stop(struct net_device *net) { struct usbnet *dev = netdev_priv(net); const struct driver_info *info = dev->driver_info; @@ -892,7 +892,7 @@ EXPORT_SYMBOL_GPL(usbnet_stop); // precondition: never called in_interrupt -int usbnet_open (struct net_device *net) +int usbnet_open(struct net_device *net) { struct usbnet *dev = netdev_priv(net); int retval; @@ -1048,7 +1048,7 @@ int usbnet_set_link_ksettings_mii(struct net_device *net, } EXPORT_SYMBOL_GPL(usbnet_set_link_ksettings_mii); -u32 usbnet_get_link (struct net_device *net) +u32 usbnet_get_link(struct net_device *net) { struct usbnet *dev = netdev_priv(net); @@ -1076,7 +1076,7 @@ int usbnet_nway_reset(struct net_device *net) } EXPORT_SYMBOL_GPL(usbnet_nway_reset); -void usbnet_get_drvinfo (struct net_device *net, struct ethtool_drvinfo *info) +void usbnet_get_drvinfo(struct net_device *net, struct ethtool_drvinfo *info) { struct usbnet *dev = netdev_priv(net); @@ -1087,7 +1087,7 @@ void usbnet_get_drvinfo (struct net_device *net, struct ethtool_drvinfo *info) } EXPORT_SYMBOL_GPL(usbnet_get_drvinfo); -u32 usbnet_get_msglevel (struct net_device *net) +u32 usbnet_get_msglevel(struct net_device *net) { struct usbnet *dev = netdev_priv(net); @@ -1095,7 +1095,7 @@ u32 usbnet_get_msglevel (struct net_device *net) } EXPORT_SYMBOL_GPL(usbnet_get_msglevel); -void usbnet_set_msglevel (struct net_device *net, u32 level) +void usbnet_set_msglevel(struct net_device *net, u32 level) { struct usbnet *dev = netdev_priv(net); @@ -1166,7 +1166,7 @@ static void __handle_set_rx_mode(struct usbnet *dev) * especially now that control transfers can be queued. */ static void -usbnet_deferred_kevent (struct work_struct *work) +usbnet_deferred_kevent(struct work_struct *work) { struct usbnet *dev = container_of(work, struct usbnet, kevent); @@ -1277,7 +1277,7 @@ usbnet_deferred_kevent (struct work_struct *work) /*-------------------------------------------------------------------------*/ -static void tx_complete (struct urb *urb) +static void tx_complete(struct urb *urb) { struct sk_buff *skb = (struct sk_buff *) urb->context; struct skb_data *entry = (struct skb_data *) skb->cb; @@ -1332,7 +1332,7 @@ static void tx_complete (struct urb *urb) /*-------------------------------------------------------------------------*/ -void usbnet_tx_timeout (struct net_device *net, unsigned int txqueue) +void usbnet_tx_timeout(struct net_device *net, unsigned int txqueue) { struct usbnet *dev = netdev_priv(net); @@ -1382,8 +1382,7 @@ static int build_dma_sg(const struct sk_buff *skb, struct urb *urb) return 1; } -netdev_tx_t usbnet_start_xmit (struct sk_buff *skb, - struct net_device *net) +netdev_tx_t usbnet_start_xmit(struct sk_buff *skb, struct net_device *net) { struct usbnet *dev = netdev_priv(net); unsigned int length; @@ -1561,7 +1560,7 @@ static inline void usb_free_skb(struct sk_buff *skb) // work (work deferred from completions, in_irq) or timer -static void usbnet_bh (struct timer_list *t) +static void usbnet_bh(struct timer_list *t) { struct usbnet *dev = timer_container_of(dev, t, delay); struct sk_buff *skb; @@ -1636,7 +1635,7 @@ static void usbnet_bh_work(struct work_struct *work) // precondition: never called in_interrupt -void usbnet_disconnect (struct usb_interface *intf) +void usbnet_disconnect(struct usb_interface *intf) { struct usbnet *dev; struct usb_device *xdev; @@ -1700,7 +1699,7 @@ static const struct device_type wwan_type = { }; int -usbnet_probe (struct usb_interface *udev, const struct usb_device_id *prod) +usbnet_probe(struct usb_interface *udev, const struct usb_device_id *prod) { struct usbnet *dev; struct net_device *net; @@ -1907,7 +1906,7 @@ EXPORT_SYMBOL_GPL(usbnet_probe); * resume only when the last interface is resumed */ -int usbnet_suspend (struct usb_interface *intf, pm_message_t message) +int usbnet_suspend(struct usb_interface *intf, pm_message_t message) { struct usbnet *dev = usb_get_intfdata(intf); @@ -1940,7 +1939,7 @@ int usbnet_suspend (struct usb_interface *intf, pm_message_t message) } EXPORT_SYMBOL_GPL(usbnet_suspend); -int usbnet_resume (struct usb_interface *intf) +int usbnet_resume(struct usb_interface *intf) { struct usbnet *dev = usb_get_intfdata(intf); struct sk_buff *skb; From 9078e6c5f1de342ae0c2322c999bbab9c2ad08b7 Mon Sep 17 00:00:00 2001 From: Lad Prabhakar Date: Thu, 23 Oct 2025 12:21:10 +0100 Subject: [PATCH 0242/1381] net: ravb: Make DBAT entry count configurable per-SoC MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Avoid wasting coherent DMA memory by allocating the descriptor base address table sized for the actual number of DBAT/CDARq entries supported by the SoC. Some platforms (for example GBETH) only provide two CDARq entries; previously the driver always allocated space for 22 entries which needlessly consumed memory on those systems. Pass the per-SoC dbat_entry_num via struct ravb_hw_info and use it for allocation and initialization in probe. This sizes the table correctly and removes the unnecessary memory overhead on SoCs with fewer DBAT entries. Signed-off-by: Lad Prabhakar Reviewed-by: Niklas Söderlund Link: https://patch.msgid.link/20251023112111.215198-2-prabhakar.mahadev-lad.rj@bp.renesas.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/renesas/ravb.h | 2 +- drivers/net/ethernet/renesas/ravb_main.c | 9 +++++++-- 2 files changed, 8 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/renesas/ravb.h b/drivers/net/ethernet/renesas/ravb.h index 7b48060c250b..d65cd83ddd16 100644 --- a/drivers/net/ethernet/renesas/ravb.h +++ b/drivers/net/ethernet/renesas/ravb.h @@ -1017,7 +1017,6 @@ enum CSR2_BIT { #define CSR2_CSUM_ENABLE (CSR2_RTCP4 | CSR2_RUDP4 | CSR2_RICMP4 | \ CSR2_RTCP6 | CSR2_RUDP6 | CSR2_RICMP6) -#define DBAT_ENTRY_NUM 22 #define RX_QUEUE_OFFSET 4 #define NUM_RX_QUEUE 2 #define NUM_TX_QUEUE 2 @@ -1062,6 +1061,7 @@ struct ravb_hw_info { u32 rx_max_frame_size; u32 rx_buffer_size; u32 rx_desc_size; + u32 dbat_entry_num; unsigned aligned_tx: 1; unsigned coalesce_irqs:1; /* Needs software IRQ coalescing */ diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c index e2d7ce1a85e8..cb5ae9f85252 100644 --- a/drivers/net/ethernet/renesas/ravb_main.c +++ b/drivers/net/ethernet/renesas/ravb_main.c @@ -2714,6 +2714,7 @@ static const struct ravb_hw_info ravb_gen2_hw_info = { .rx_buffer_size = SZ_2K + SKB_DATA_ALIGN(sizeof(struct skb_shared_info)), .rx_desc_size = sizeof(struct ravb_ex_rx_desc), + .dbat_entry_num = 22, .aligned_tx = 1, .gptp = 1, .nc_queues = 1, @@ -2737,6 +2738,7 @@ static const struct ravb_hw_info ravb_gen3_hw_info = { .rx_buffer_size = SZ_2K + SKB_DATA_ALIGN(sizeof(struct skb_shared_info)), .rx_desc_size = sizeof(struct ravb_ex_rx_desc), + .dbat_entry_num = 22, .internal_delay = 1, .tx_counters = 1, .multi_irqs = 1, @@ -2763,6 +2765,7 @@ static const struct ravb_hw_info ravb_gen4_hw_info = { .rx_buffer_size = SZ_2K + SKB_DATA_ALIGN(sizeof(struct skb_shared_info)), .rx_desc_size = sizeof(struct ravb_ex_rx_desc), + .dbat_entry_num = 22, .internal_delay = 1, .tx_counters = 1, .multi_irqs = 1, @@ -2789,6 +2792,7 @@ static const struct ravb_hw_info ravb_rzv2m_hw_info = { .rx_buffer_size = SZ_2K + SKB_DATA_ALIGN(sizeof(struct skb_shared_info)), .rx_desc_size = sizeof(struct ravb_ex_rx_desc), + .dbat_entry_num = 22, .multi_irqs = 1, .err_mgmt_irqs = 1, .gptp = 1, @@ -2814,6 +2818,7 @@ static const struct ravb_hw_info gbeth_hw_info = { .rx_max_frame_size = SZ_8K, .rx_buffer_size = SZ_2K, .rx_desc_size = sizeof(struct ravb_rx_desc), + .dbat_entry_num = 2, .aligned_tx = 1, .coalesce_irqs = 1, .tx_counters = 1, @@ -3045,7 +3050,7 @@ static int ravb_probe(struct platform_device *pdev) ravb_parse_delay_mode(np, ndev); /* Allocate descriptor base address table */ - priv->desc_bat_size = sizeof(struct ravb_desc) * DBAT_ENTRY_NUM; + priv->desc_bat_size = sizeof(struct ravb_desc) * info->dbat_entry_num; priv->desc_bat = dma_alloc_coherent(ndev->dev.parent, priv->desc_bat_size, &priv->desc_bat_dma, GFP_KERNEL); if (!priv->desc_bat) { @@ -3055,7 +3060,7 @@ static int ravb_probe(struct platform_device *pdev) error = -ENOMEM; goto out_rpm_put; } - for (q = RAVB_BE; q < DBAT_ENTRY_NUM; q++) + for (q = RAVB_BE; q < info->dbat_entry_num; q++) priv->desc_bat[q].die_dt = DT_EOS; /* Initialise HW timestamp list */ From 3912e804ff6a03693cc50d801ab840479f7b20ac Mon Sep 17 00:00:00 2001 From: Lad Prabhakar Date: Thu, 23 Oct 2025 12:21:11 +0100 Subject: [PATCH 0243/1381] net: ravb: Allocate correct number of queues based on SoC support MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Use the per-SoC match data flag `nc_queues` to decide how many TX/RX queues to allocate. If the SoC does not provide a network-control queue, fall back to a single TX/RX queue. Obtain the match data before calling alloc_etherdev_mqs() so the allocation is sized correctly. Signed-off-by: Lad Prabhakar Reviewed-by: Niklas Söderlund Link: https://patch.msgid.link/20251023112111.215198-3-prabhakar.mahadev-lad.rj@bp.renesas.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/renesas/ravb_main.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c index cb5ae9f85252..c3fc15f9ec85 100644 --- a/drivers/net/ethernet/renesas/ravb_main.c +++ b/drivers/net/ethernet/renesas/ravb_main.c @@ -2946,13 +2946,14 @@ static int ravb_probe(struct platform_device *pdev) return dev_err_probe(&pdev->dev, PTR_ERR(rstc), "failed to get cpg reset\n"); + info = of_device_get_match_data(&pdev->dev); + ndev = alloc_etherdev_mqs(sizeof(struct ravb_private), - NUM_TX_QUEUE, NUM_RX_QUEUE); + info->nc_queues ? NUM_TX_QUEUE : 1, + info->nc_queues ? NUM_RX_QUEUE : 1); if (!ndev) return -ENOMEM; - info = of_device_get_match_data(&pdev->dev); - ndev->features = info->net_features; ndev->hw_features = info->net_hw_features; ndev->vlan_features = info->vlan_features; From 32dd679b88d58e5245727974d1726f499f7f8f3d Mon Sep 17 00:00:00 2001 From: Heiko Stuebner Date: Thu, 23 Oct 2025 13:12:08 +0200 Subject: [PATCH 0244/1381] dt-bindings: net: snps,dwmac: move rk3399 line to its correct position Move the rk3399 compatible to its alphabetically correct position. Reviewed-by: Andrew Lunn Acked-by: Conor Dooley Signed-off-by: Heiko Stuebner Link: https://patch.msgid.link/20251023111213.298860-2-heiko@sntech.de Signed-off-by: Jakub Kicinski --- Documentation/devicetree/bindings/net/snps,dwmac.yaml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Documentation/devicetree/bindings/net/snps,dwmac.yaml b/Documentation/devicetree/bindings/net/snps,dwmac.yaml index 658c004e6a5c..28113ac5e11a 100644 --- a/Documentation/devicetree/bindings/net/snps,dwmac.yaml +++ b/Documentation/devicetree/bindings/net/snps,dwmac.yaml @@ -86,9 +86,9 @@ properties: - rockchip,rk3328-gmac - rockchip,rk3366-gmac - rockchip,rk3368-gmac + - rockchip,rk3399-gmac - rockchip,rk3576-gmac - rockchip,rk3588-gmac - - rockchip,rk3399-gmac - rockchip,rv1108-gmac - snps,dwmac - snps,dwmac-3.40a From e774c91dca451bcf6eb4ca05d6bef977f88ceff6 Mon Sep 17 00:00:00 2001 From: Heiko Stuebner Date: Thu, 23 Oct 2025 13:12:09 +0200 Subject: [PATCH 0245/1381] dt-bindings: net: snps,dwmac: Sync list of Rockchip compatibles A number of dwmac variants from Rockchip SoCs have turned up in the Rockchip-specific binding, but not in the main list in snps,dwmac.yaml which as the comment indicates is needed for accurate matching. So add the missing rk3528, rk3568 and rv1126 to the main list. Reviewed-by: Andrew Lunn Acked-by: Conor Dooley Signed-off-by: Heiko Stuebner Link: https://patch.msgid.link/20251023111213.298860-3-heiko@sntech.de Signed-off-by: Jakub Kicinski --- Documentation/devicetree/bindings/net/snps,dwmac.yaml | 3 +++ 1 file changed, 3 insertions(+) diff --git a/Documentation/devicetree/bindings/net/snps,dwmac.yaml b/Documentation/devicetree/bindings/net/snps,dwmac.yaml index 28113ac5e11a..1a0d6789a59b 100644 --- a/Documentation/devicetree/bindings/net/snps,dwmac.yaml +++ b/Documentation/devicetree/bindings/net/snps,dwmac.yaml @@ -87,9 +87,12 @@ properties: - rockchip,rk3366-gmac - rockchip,rk3368-gmac - rockchip,rk3399-gmac + - rockchip,rk3528-gmac + - rockchip,rk3568-gmac - rockchip,rk3576-gmac - rockchip,rk3588-gmac - rockchip,rv1108-gmac + - rockchip,rv1126-gmac - snps,dwmac - snps,dwmac-3.40a - snps,dwmac-3.50a From 4a667bec74b3c108729ae2db2196f6bd15d62f74 Mon Sep 17 00:00:00 2001 From: Heiko Stuebner Date: Thu, 23 Oct 2025 13:12:10 +0200 Subject: [PATCH 0246/1381] dt-bindings: net: rockchip-dwmac: Add compatible string for RK3506 Rockchip RK3506 has two Ethernet controllers based on Synopsys DWC Ethernet QoS IP. Add compatible string for the RK3506 variant. Reviewed-by: Andrew Lunn Acked-by: Conor Dooley Signed-off-by: Heiko Stuebner Link: https://patch.msgid.link/20251023111213.298860-4-heiko@sntech.de Signed-off-by: Jakub Kicinski --- Documentation/devicetree/bindings/net/rockchip-dwmac.yaml | 3 +++ Documentation/devicetree/bindings/net/snps,dwmac.yaml | 1 + 2 files changed, 4 insertions(+) diff --git a/Documentation/devicetree/bindings/net/rockchip-dwmac.yaml b/Documentation/devicetree/bindings/net/rockchip-dwmac.yaml index 0ac7c4b47d6b..d17112527dab 100644 --- a/Documentation/devicetree/bindings/net/rockchip-dwmac.yaml +++ b/Documentation/devicetree/bindings/net/rockchip-dwmac.yaml @@ -24,6 +24,7 @@ select: - rockchip,rk3366-gmac - rockchip,rk3368-gmac - rockchip,rk3399-gmac + - rockchip,rk3506-gmac - rockchip,rk3528-gmac - rockchip,rk3568-gmac - rockchip,rk3576-gmac @@ -50,6 +51,7 @@ properties: - rockchip,rv1108-gmac - items: - enum: + - rockchip,rk3506-gmac - rockchip,rk3528-gmac - rockchip,rk3568-gmac - rockchip,rk3576-gmac @@ -148,6 +150,7 @@ allOf: compatible: contains: enum: + - rockchip,rk3506-gmac - rockchip,rk3528-gmac then: properties: diff --git a/Documentation/devicetree/bindings/net/snps,dwmac.yaml b/Documentation/devicetree/bindings/net/snps,dwmac.yaml index 1a0d6789a59b..dd3c72e8363e 100644 --- a/Documentation/devicetree/bindings/net/snps,dwmac.yaml +++ b/Documentation/devicetree/bindings/net/snps,dwmac.yaml @@ -87,6 +87,7 @@ properties: - rockchip,rk3366-gmac - rockchip,rk3368-gmac - rockchip,rk3399-gmac + - rockchip,rk3506-gmac - rockchip,rk3528-gmac - rockchip,rk3568-gmac - rockchip,rk3576-gmac From 2010163a8ea4a19308ec6cf259fe68f014553efb Mon Sep 17 00:00:00 2001 From: David Wu Date: Thu, 23 Oct 2025 13:12:11 +0200 Subject: [PATCH 0247/1381] ethernet: stmmac: dwmac-rk: Add RK3506 GMAC support Add the needed glue blocks for the RK3506-specific setup. The RK3506 dwmac only supports up to 100MBit with a RMII PHY, but no RGMII. Signed-off-by: David Wu Signed-off-by: Heiko Stuebner Link: https://patch.msgid.link/20251023111213.298860-5-heiko@sntech.de Signed-off-by: Jakub Kicinski --- .../net/ethernet/stmicro/stmmac/dwmac-rk.c | 64 +++++++++++++++++++ 1 file changed, 64 insertions(+) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c index 643578266dfc..a5c7e03ebc63 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c @@ -827,6 +827,69 @@ static const struct rk_gmac_ops rk3399_ops = { .set_speed = rk3399_set_speed, }; +#define RK3506_GRF_SOC_CON8 0x0020 +#define RK3506_GRF_SOC_CON11 0x002c + +#define RK3506_GMAC_RMII_MODE GRF_BIT(1) + +#define RK3506_GMAC_CLK_RMII_DIV2 GRF_BIT(3) +#define RK3506_GMAC_CLK_RMII_DIV20 GRF_CLR_BIT(3) + +#define RK3506_GMAC_CLK_SELECT_CRU GRF_CLR_BIT(5) +#define RK3506_GMAC_CLK_SELECT_IO GRF_BIT(5) + +#define RK3506_GMAC_CLK_RMII_GATE GRF_BIT(2) +#define RK3506_GMAC_CLK_RMII_NOGATE GRF_CLR_BIT(2) + +static void rk3506_set_to_rmii(struct rk_priv_data *bsp_priv) +{ + unsigned int id = bsp_priv->id, offset; + + offset = (id == 1) ? RK3506_GRF_SOC_CON11 : RK3506_GRF_SOC_CON8; + regmap_write(bsp_priv->grf, offset, RK3506_GMAC_RMII_MODE); +} + +static const struct rk_reg_speed_data rk3506_reg_speed_data = { + .rmii_10 = RK3506_GMAC_CLK_RMII_DIV20, + .rmii_100 = RK3506_GMAC_CLK_RMII_DIV2, +}; + +static int rk3506_set_speed(struct rk_priv_data *bsp_priv, + phy_interface_t interface, int speed) +{ + unsigned int id = bsp_priv->id, offset; + + offset = (id == 1) ? RK3506_GRF_SOC_CON11 : RK3506_GRF_SOC_CON8; + return rk_set_reg_speed(bsp_priv, &rk3506_reg_speed_data, + offset, interface, speed); +} + +static void rk3506_set_clock_selection(struct rk_priv_data *bsp_priv, + bool input, bool enable) +{ + unsigned int value, offset, id = bsp_priv->id; + + offset = (id == 1) ? RK3506_GRF_SOC_CON11 : RK3506_GRF_SOC_CON8; + + value = input ? RK3506_GMAC_CLK_SELECT_IO : + RK3506_GMAC_CLK_SELECT_CRU; + value |= enable ? RK3506_GMAC_CLK_RMII_NOGATE : + RK3506_GMAC_CLK_RMII_GATE; + regmap_write(bsp_priv->grf, offset, value); +} + +static const struct rk_gmac_ops rk3506_ops = { + .set_to_rmii = rk3506_set_to_rmii, + .set_speed = rk3506_set_speed, + .set_clock_selection = rk3506_set_clock_selection, + .regs_valid = true, + .regs = { + 0xff4c8000, /* gmac0 */ + 0xff4d0000, /* gmac1 */ + 0x0, /* sentinel */ + }, +}; + #define RK3528_VO_GRF_GMAC_CON 0x0018 #define RK3528_VO_GRF_MACPHY_CON0 0x001c #define RK3528_VO_GRF_MACPHY_CON1 0x0020 @@ -1809,6 +1872,7 @@ static const struct of_device_id rk_gmac_dwmac_match[] = { { .compatible = "rockchip,rk3366-gmac", .data = &rk3366_ops }, { .compatible = "rockchip,rk3368-gmac", .data = &rk3368_ops }, { .compatible = "rockchip,rk3399-gmac", .data = &rk3399_ops }, + { .compatible = "rockchip,rk3506-gmac", .data = &rk3506_ops }, { .compatible = "rockchip,rk3528-gmac", .data = &rk3528_ops }, { .compatible = "rockchip,rk3568-gmac", .data = &rk3568_ops }, { .compatible = "rockchip,rk3576-gmac", .data = &rk3576_ops }, From 384d8426329531fe1952549266fa3a7444dc6c31 Mon Sep 17 00:00:00 2001 From: Heiko Stuebner Date: Thu, 23 Oct 2025 13:12:12 +0200 Subject: [PATCH 0248/1381] MAINTAINERS: add dwmac-rk glue driver to the main Rockchip entry The dwmac-rk glue driver is currently not caught by the general maintainer entry for Rockchip SoCs, so add it explicitly, similar to the i2c driver. The binding document in net/rockchip-dwmac.yaml already gets caught by the wildcard match. Signed-off-by: Heiko Stuebner Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20251023111213.298860-6-heiko@sntech.de Signed-off-by: Jakub Kicinski --- MAINTAINERS | 1 + 1 file changed, 1 insertion(+) diff --git a/MAINTAINERS b/MAINTAINERS index 3ed59823f7a4..d652f4f27756 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -3296,6 +3296,7 @@ F: drivers/*/*/*rockchip* F: drivers/*/*rockchip* F: drivers/clk/rockchip/ F: drivers/i2c/busses/i2c-rk3x.c +F: drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c F: sound/soc/rockchip/ N: rockchip From eea31f21dce10814e34dc7ef7ed5136269c7bb59 Mon Sep 17 00:00:00 2001 From: Adithya Jayachandran Date: Wed, 15 Oct 2025 18:40:55 -0700 Subject: [PATCH 0249/1381] {rdma,net}/mlx5: Query vports mac address from device Before this patch during either switchdev or legacy mode enablement we cleared the mac address of vports between changes. This change allows us to preserve the vports mac address between eswitch mode changes. Vports hold information for VFs/SFs such as the permanent mac address. VF/SF mac can be set either by iproute vf interface or devlink function interface. For no obvious reason we reset it to 0 on switchdev/legacy mode changes, this patch is fixing that, to align with other vport information that are never reset, e.g GUID,mtu,promisc mode, etc .. Signed-off-by: Adithya Jayachandran Signed-off-by: Saeed Mahameed Reviewed-by: Mark Bloch Acked-by: Leon Romanovsky # RDMA --- drivers/infiniband/hw/mlx5/main.c | 2 +- .../net/ethernet/mellanox/mlx5/core/eswitch.c | 20 ++++++---------- .../mellanox/mlx5/core/eswitch_offloads.c | 3 +++ .../net/ethernet/mellanox/mlx5/core/vport.c | 24 +++++++++---------- include/linux/mlx5/vport.h | 3 ++- 5 files changed, 25 insertions(+), 27 deletions(-) diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c index fc1e86f6c409..90daa58126f4 100644 --- a/drivers/infiniband/hw/mlx5/main.c +++ b/drivers/infiniband/hw/mlx5/main.c @@ -842,7 +842,7 @@ static int mlx5_query_node_guid(struct mlx5_ib_dev *dev, break; case MLX5_VPORT_ACCESS_METHOD_NIC: - err = mlx5_query_nic_vport_node_guid(dev->mdev, &tmp); + err = mlx5_query_nic_vport_node_guid(dev->mdev, 0, false, &tmp); break; default: diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c index e2ffb87b94cb..25af8bd7f077 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c @@ -875,13 +875,10 @@ static int esw_vport_setup(struct mlx5_eswitch *esw, struct mlx5_vport *vport) vport_num, 1, vport->info.link_state); - /* Host PF has its own mac/guid. */ - if (vport_num) { - mlx5_modify_nic_vport_mac_address(esw->dev, vport_num, - vport->info.mac); - mlx5_modify_nic_vport_node_guid(esw->dev, vport_num, - vport->info.node_guid); - } + mlx5_query_nic_vport_mac_address(esw->dev, vport_num, true, + vport->info.mac); + mlx5_query_nic_vport_node_guid(esw->dev, vport_num, true, + &vport->info.node_guid); flags = (vport->info.vlan || vport->info.qos) ? SET_VLAN_STRIP | SET_VLAN_INSERT : 0; @@ -947,12 +944,6 @@ int mlx5_esw_vport_enable(struct mlx5_eswitch *esw, struct mlx5_vport *vport, goto err_vhca_mapping; } - /* External controller host PF has factory programmed MAC. - * Read it from the device. - */ - if (mlx5_core_is_ecpf(esw->dev) && vport_num == MLX5_VPORT_PF) - mlx5_query_nic_vport_mac_address(esw->dev, vport_num, true, vport->info.mac); - esw_vport_change_handle_locked(vport); esw->enabled_vports++; @@ -2235,6 +2226,9 @@ int mlx5_eswitch_get_vport_config(struct mlx5_eswitch *esw, ivi->vf = vport - 1; mutex_lock(&esw->state_lock); + + mlx5_query_nic_vport_mac_address(esw->dev, vport, true, + evport->info.mac); ether_addr_copy(ivi->mac, evport->info.mac); ivi->linkstate = evport->info.link_state; ivi->vlan = evport->info.vlan; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c index 4cf995be127d..880e238497b1 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -4303,6 +4303,9 @@ int mlx5_devlink_port_fn_hw_addr_get(struct devlink_port *port, struct mlx5_vport *vport = mlx5_devlink_port_vport_get(port); mutex_lock(&esw->state_lock); + + mlx5_query_nic_vport_mac_address(esw->dev, vport->vport, true, + vport->info.mac); ether_addr_copy(hw_addr, vport->info.mac); *hw_addr_len = ETH_ALEN; mutex_unlock(&esw->state_lock); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/vport.c b/drivers/net/ethernet/mellanox/mlx5/core/vport.c index 2ed2e530b07d..d1483f66cd0c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/vport.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/vport.c @@ -78,15 +78,14 @@ int mlx5_modify_vport_admin_state(struct mlx5_core_dev *mdev, u8 opmod, } static int mlx5_query_nic_vport_context(struct mlx5_core_dev *mdev, u16 vport, - u32 *out) + bool other_vport, u32 *out) { u32 in[MLX5_ST_SZ_DW(query_nic_vport_context_in)] = {}; MLX5_SET(query_nic_vport_context_in, in, opcode, MLX5_CMD_OP_QUERY_NIC_VPORT_CONTEXT); MLX5_SET(query_nic_vport_context_in, in, vport_number, vport); - if (vport) - MLX5_SET(query_nic_vport_context_in, in, other_vport, 1); + MLX5_SET(query_nic_vport_context_in, in, other_vport, other_vport); return mlx5_cmd_exec_inout(mdev, query_nic_vport_context, in, out); } @@ -97,7 +96,7 @@ int mlx5_query_nic_vport_min_inline(struct mlx5_core_dev *mdev, u32 out[MLX5_ST_SZ_DW(query_nic_vport_context_out)] = {}; int err; - err = mlx5_query_nic_vport_context(mdev, vport, out); + err = mlx5_query_nic_vport_context(mdev, vport, vport > 0, out); if (!err) *min_inline = MLX5_GET(query_nic_vport_context_out, out, nic_vport_context.min_wqe_inline_mode); @@ -219,7 +218,7 @@ int mlx5_query_nic_vport_mtu(struct mlx5_core_dev *mdev, u16 *mtu) if (!out) return -ENOMEM; - err = mlx5_query_nic_vport_context(mdev, 0, out); + err = mlx5_query_nic_vport_context(mdev, 0, false, out); if (!err) *mtu = MLX5_GET(query_nic_vport_context_out, out, nic_vport_context.mtu); @@ -429,7 +428,7 @@ int mlx5_query_nic_vport_system_image_guid(struct mlx5_core_dev *mdev, if (!out) return -ENOMEM; - err = mlx5_query_nic_vport_context(mdev, 0, out); + err = mlx5_query_nic_vport_context(mdev, 0, false, out); if (err) goto out; @@ -451,7 +450,7 @@ int mlx5_query_nic_vport_sd_group(struct mlx5_core_dev *mdev, u8 *sd_group) if (!out) return -ENOMEM; - err = mlx5_query_nic_vport_context(mdev, 0, out); + err = mlx5_query_nic_vport_context(mdev, 0, false, out); if (err) goto out; @@ -462,7 +461,8 @@ int mlx5_query_nic_vport_sd_group(struct mlx5_core_dev *mdev, u8 *sd_group) return err; } -int mlx5_query_nic_vport_node_guid(struct mlx5_core_dev *mdev, u64 *node_guid) +int mlx5_query_nic_vport_node_guid(struct mlx5_core_dev *mdev, + u16 vport, bool other_vport, u64 *node_guid) { u32 *out; int outlen = MLX5_ST_SZ_BYTES(query_nic_vport_context_out); @@ -472,7 +472,7 @@ int mlx5_query_nic_vport_node_guid(struct mlx5_core_dev *mdev, u64 *node_guid) if (!out) return -ENOMEM; - err = mlx5_query_nic_vport_context(mdev, 0, out); + err = mlx5_query_nic_vport_context(mdev, vport, other_vport, out); if (err) goto out; @@ -529,7 +529,7 @@ int mlx5_query_nic_vport_qkey_viol_cntr(struct mlx5_core_dev *mdev, if (!out) return -ENOMEM; - err = mlx5_query_nic_vport_context(mdev, 0, out); + err = mlx5_query_nic_vport_context(mdev, 0, false, out); if (err) goto out; @@ -804,7 +804,7 @@ int mlx5_query_nic_vport_promisc(struct mlx5_core_dev *mdev, if (!out) return -ENOMEM; - err = mlx5_query_nic_vport_context(mdev, vport, out); + err = mlx5_query_nic_vport_context(mdev, vport, vport > 0, out); if (err) goto out; @@ -908,7 +908,7 @@ int mlx5_nic_vport_query_local_lb(struct mlx5_core_dev *mdev, bool *status) if (!out) return -ENOMEM; - err = mlx5_query_nic_vport_context(mdev, 0, out); + err = mlx5_query_nic_vport_context(mdev, 0, false, out); if (err) goto out; diff --git a/include/linux/mlx5/vport.h b/include/linux/mlx5/vport.h index c87b9507cfa1..f876bfc0669c 100644 --- a/include/linux/mlx5/vport.h +++ b/include/linux/mlx5/vport.h @@ -73,7 +73,8 @@ int mlx5_modify_nic_vport_mtu(struct mlx5_core_dev *mdev, u16 mtu); int mlx5_query_nic_vport_system_image_guid(struct mlx5_core_dev *mdev, u64 *system_image_guid); int mlx5_query_nic_vport_sd_group(struct mlx5_core_dev *mdev, u8 *sd_group); -int mlx5_query_nic_vport_node_guid(struct mlx5_core_dev *mdev, u64 *node_guid); +int mlx5_query_nic_vport_node_guid(struct mlx5_core_dev *mdev, + u16 vport, bool other_vport, u64 *node_guid); int mlx5_modify_nic_vport_node_guid(struct mlx5_core_dev *mdev, u16 vport, u64 node_guid); int mlx5_query_nic_vport_qkey_viol_cntr(struct mlx5_core_dev *mdev, From 428ea708b714ba09c96da5722fb1c8b86af509d1 Mon Sep 17 00:00:00 2001 From: Rosen Penev Date: Mon, 13 Oct 2025 22:08:33 -0700 Subject: [PATCH 0250/1381] wifi: rt2x00: check retval for of_get_mac_address of_get_mac_address can return -EPROBE_DEFER when nvmem is not probed yet for whatever reason. In this case, nvmem mac assignments will not work. Based on the function path, this change only has effect for rt2800soc.c and rt2800pci.c. The former tends to use nvmem for assignments. Signed-off-by: Rosen Penev Acked-by: Stanislaw Gruszka Link: https://patch.msgid.link/20251014050833.46377-1-rosenp@gmail.com Signed-off-by: Johannes Berg --- drivers/net/wireless/ralink/rt2x00/rt2800lib.c | 4 +++- drivers/net/wireless/ralink/rt2x00/rt2x00.h | 2 +- drivers/net/wireless/ralink/rt2x00/rt2x00dev.c | 10 ++++++++-- 3 files changed, 12 insertions(+), 4 deletions(-) diff --git a/drivers/net/wireless/ralink/rt2x00/rt2800lib.c b/drivers/net/wireless/ralink/rt2x00/rt2800lib.c index b264ed0af923..f07152fa3725 100644 --- a/drivers/net/wireless/ralink/rt2x00/rt2800lib.c +++ b/drivers/net/wireless/ralink/rt2x00/rt2800lib.c @@ -11011,7 +11011,9 @@ static int rt2800_validate_eeprom(struct rt2x00_dev *rt2x00dev) * Start validation of the data that has been read. */ mac = rt2800_eeprom_addr(rt2x00dev, EEPROM_MAC_ADDR_0); - rt2x00lib_set_mac_address(rt2x00dev, mac); + retval = rt2x00lib_set_mac_address(rt2x00dev, mac); + if (retval) + return retval; word = rt2800_eeprom_read(rt2x00dev, EEPROM_NIC_CONF0); if (word == 0xffff) { diff --git a/drivers/net/wireless/ralink/rt2x00/rt2x00.h b/drivers/net/wireless/ralink/rt2x00/rt2x00.h index 09b9d1f9f793..665887e9b118 100644 --- a/drivers/net/wireless/ralink/rt2x00/rt2x00.h +++ b/drivers/net/wireless/ralink/rt2x00/rt2x00.h @@ -1427,7 +1427,7 @@ static inline void rt2x00debug_dump_frame(struct rt2x00_dev *rt2x00dev, */ u32 rt2x00lib_get_bssidx(struct rt2x00_dev *rt2x00dev, struct ieee80211_vif *vif); -void rt2x00lib_set_mac_address(struct rt2x00_dev *rt2x00dev, u8 *eeprom_mac_addr); +int rt2x00lib_set_mac_address(struct rt2x00_dev *rt2x00dev, u8 *eeprom_mac_addr); /* * Interrupt context handlers. diff --git a/drivers/net/wireless/ralink/rt2x00/rt2x00dev.c b/drivers/net/wireless/ralink/rt2x00/rt2x00dev.c index f8a6f9c968a1..778a478ab53a 100644 --- a/drivers/net/wireless/ralink/rt2x00/rt2x00dev.c +++ b/drivers/net/wireless/ralink/rt2x00/rt2x00dev.c @@ -988,14 +988,20 @@ static void rt2x00lib_rate(struct ieee80211_rate *entry, entry->flags |= IEEE80211_RATE_SHORT_PREAMBLE; } -void rt2x00lib_set_mac_address(struct rt2x00_dev *rt2x00dev, u8 *eeprom_mac_addr) +int rt2x00lib_set_mac_address(struct rt2x00_dev *rt2x00dev, u8 *eeprom_mac_addr) { - of_get_mac_address(rt2x00dev->dev->of_node, eeprom_mac_addr); + int ret; + + ret = of_get_mac_address(rt2x00dev->dev->of_node, eeprom_mac_addr); + if (ret == -EPROBE_DEFER) + return ret; if (!is_valid_ether_addr(eeprom_mac_addr)) { eth_random_addr(eeprom_mac_addr); rt2x00_eeprom_dbg(rt2x00dev, "MAC: %pM\n", eeprom_mac_addr); } + + return 0; } EXPORT_SYMBOL_GPL(rt2x00lib_set_mac_address); From a392cde88d19af917740d27e13115447d3b21a06 Mon Sep 17 00:00:00 2001 From: Ryder Lee Date: Tue, 23 Sep 2025 17:23:22 +0000 Subject: [PATCH 0251/1381] wifi: cfg80211/mac80211: validate radio frequency range for monitor mode In multi-radio devices, it is possible to have an MLD AP and a monitor interface active at the same time. In such cases, monitor mode may not be able to specify a fixed channel and could end up capturing frames from all radios, including those outside the intended frequency bands. This patch adds frequency validation for monitor mode. Received frames are now only processed if their frequency fall within the allowed ranges of the radios specified by the interface's radio_mask. This prevents monitor mode from capturing frames outside the supported radio. Signed-off-by: Ryder Lee Link: https://patch.msgid.link/700b8284e845d96654eb98431f8eeb5a81503862.1758647858.git.ryder.lee@mediatek.com Signed-off-by: Johannes Berg --- include/net/cfg80211.h | 14 ++++++++++++ net/mac80211/rx.c | 49 ++++++++++++++++++++++++++++++++++++++++++ net/wireless/util.c | 6 +++--- 3 files changed, 66 insertions(+), 3 deletions(-) diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index 781624f5913a..3b6f48a783bb 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -1015,6 +1015,7 @@ const struct cfg80211_chan_def * cfg80211_chandef_compatible(const struct cfg80211_chan_def *chandef1, const struct cfg80211_chan_def *chandef2); + /** * nl80211_chan_width_to_mhz - get the channel width in MHz * @chan_width: the channel width from &enum nl80211_chan_width @@ -6882,6 +6883,19 @@ static inline bool cfg80211_channel_is_psc(struct ieee80211_channel *chan) return ieee80211_frequency_to_channel(chan->center_freq) % 16 == 5; } +/** + * ieee80211_radio_freq_range_valid - Check if the radio supports the + * specified frequency range + * + * @radio: wiphy radio + * @freq: the frequency (in KHz) to be queried + * @width: the bandwidth (in KHz) to be queried + * + * Return: whether or not the given frequency range is valid for the given radio + */ +bool ieee80211_radio_freq_range_valid(const struct wiphy_radio *radio, + u32 freq, u32 width); + /** * cfg80211_radio_chandef_valid - Check if the radio supports the chandef * diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c index 6af43dfefdd6..29175a0c9f68 100644 --- a/net/mac80211/rx.c +++ b/net/mac80211/rx.c @@ -763,6 +763,51 @@ ieee80211_make_monitor_skb(struct ieee80211_local *local, return skb; } +static bool +ieee80211_validate_monitor_radio(struct ieee80211_sub_if_data *sdata, + struct ieee80211_local *local, + struct ieee80211_rx_status *status) +{ + struct wiphy *wiphy = local->hw.wiphy; + int i, freq, bw; + + if (!wiphy->n_radio) + return true; + + switch (status->bw) { + case RATE_INFO_BW_20: + bw = 20000; + break; + case RATE_INFO_BW_40: + bw = 40000; + break; + case RATE_INFO_BW_80: + bw = 80000; + break; + case RATE_INFO_BW_160: + bw = 160000; + break; + case RATE_INFO_BW_320: + bw = 320000; + break; + default: + return false; + } + + freq = MHZ_TO_KHZ(status->freq); + + for (i = 0; i < wiphy->n_radio; i++) { + if (!(sdata->wdev.radio_mask & BIT(i))) + continue; + + if (!ieee80211_radio_freq_range_valid(&wiphy->radio[i], freq, bw)) + continue; + + return true; + } + return false; +} + /* * This function copies a received frame to all monitor interfaces and * returns a cleaned-up SKB that no longer includes the FCS nor the @@ -855,6 +900,10 @@ ieee80211_rx_monitor(struct ieee80211_local *local, struct sk_buff *origskb, chandef->chan->center_freq != status->freq) continue; + if (ieee80211_hw_check(&local->hw, NO_VIRTUAL_MONITOR) && + !ieee80211_validate_monitor_radio(sdata, local, status)) + continue; + if (!prev_sdata) { prev_sdata = sdata; continue; diff --git a/net/wireless/util.c b/net/wireless/util.c index 56724b33af04..97f40c6d1e9d 100644 --- a/net/wireless/util.c +++ b/net/wireless/util.c @@ -2942,9 +2942,8 @@ cfg80211_get_iftype_ext_capa(struct wiphy *wiphy, enum nl80211_iftype type) } EXPORT_SYMBOL(cfg80211_get_iftype_ext_capa); -static bool -ieee80211_radio_freq_range_valid(const struct wiphy_radio *radio, - u32 freq, u32 width) +bool ieee80211_radio_freq_range_valid(const struct wiphy_radio *radio, + u32 freq, u32 width) { const struct wiphy_radio_freq_range *r; int i; @@ -2958,6 +2957,7 @@ ieee80211_radio_freq_range_valid(const struct wiphy_radio *radio, return false; } +EXPORT_SYMBOL(ieee80211_radio_freq_range_valid); bool cfg80211_radio_chandef_valid(const struct wiphy_radio *radio, const struct cfg80211_chan_def *chandef) From a3b16dfe79eecafea0e058b038bb506ed9bd2c89 Mon Sep 17 00:00:00 2001 From: Aditya Kumar Singh Date: Fri, 17 Oct 2025 09:32:41 +0530 Subject: [PATCH 0252/1381] wifi: mac80211_hwsim: advertise puncturing feature support If userspace provides a puncturing bitmap via the NL80211_ATTR_PUNCT_BITMAP attribute, the kernel with mac80211_hwsim driver currently rejects the command with the error: "driver doesn't support puncturing", because the driver does not advertise support for this feature. At present, the following hwsim test cases utilize puncturing, but the bitmap is not sent to the kernel. Instead, the puncturing information is conveyed only through the beacon data: * eht_5ghz_80mhz_puncturing_override_1 * eht_5ghz_80mhz_puncturing_override_2 * eht_5ghz_80mhz_puncturing_override_3 A future change in hostapd will begin configuring the puncturing bitmap explicitly, which will cause these test cases to fail unless the driver advertises support. To address this, update mac80211_hwsim driver to advertise puncturing feature support. Signed-off-by: Aditya Kumar Singh Link: https://patch.msgid.link/20251017-hwsim_set_punct_feature_bit-v1-1-3be1bb3450c0@oss.qualcomm.com Signed-off-by: Johannes Berg --- drivers/net/wireless/virtual/mac80211_hwsim.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/wireless/virtual/mac80211_hwsim.c b/drivers/net/wireless/virtual/mac80211_hwsim.c index 9f856042a67a..cd84dfd5b47e 100644 --- a/drivers/net/wireless/virtual/mac80211_hwsim.c +++ b/drivers/net/wireless/virtual/mac80211_hwsim.c @@ -5793,6 +5793,7 @@ static int mac80211_hwsim_new_radio(struct genl_info *info, ieee80211_hw_set(hw, NO_AUTO_VIF); wiphy_ext_feature_set(hw->wiphy, NL80211_EXT_FEATURE_CQM_RSSI_LIST); + wiphy_ext_feature_set(hw->wiphy, NL80211_EXT_FEATURE_PUNCT); for (i = 0; i < ARRAY_SIZE(data->link_data); i++) { hrtimer_setup(&data->link_data[i].beacon_timer, mac80211_hwsim_beacon, From 8f24be708829854560e1db9f765c51305b046183 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Sun, 19 Oct 2025 11:50:49 +0300 Subject: [PATCH 0253/1381] wifi: mac80211: reset CRC valid after CSA While waiting for a beacon after CSA, reset the CRC valid so that the next beacon is handled even if it happens to be identical the last one on the old channel. This is an AP bug either way, but it's better to disconnect cleanly than to have lingering CSA state. In the iwlwifi instantiation of this problem, mac80211 is ignoring the beacon but the firmware creates a new CSA, and then crashes later because mac80211/driver didn't do anything about it. Signed-off-by: Johannes Berg Reviewed-by: Emmanuel Grumbach Signed-off-by: Miri Korenblit Link: https://patch.msgid.link/20251019115024.521ad9c6b87d.I86376900df3d3423185b75bf63358c29f33a5eb6@changeid Signed-off-by: Johannes Berg --- net/mac80211/mlme.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c index 3b5827ea438e..e699702fe5b1 100644 --- a/net/mac80211/mlme.c +++ b/net/mac80211/mlme.c @@ -2508,6 +2508,16 @@ static void ieee80211_csa_switch_work(struct wiphy *wiphy, link->u.mgd.csa.waiting_bcn = true; + /* + * The next beacon really should always be different, so this should + * have no effect whatsoever. However, some APs (we observed this in + * an Asus AXE11000), the beacon after the CSA might be identical to + * the last beacon on the old channel - in this case we'd ignore it. + * Resetting the CRC will lead us to handle it better (albeit with a + * disconnect, but clearly the AP is broken.) + */ + link->u.mgd.beacon_crc_valid = false; + /* apply new TPE restrictions immediately on the new channel */ if (link->u.mgd.csa.ap_chandef.chan->band == NL80211_BAND_6GHZ && link->u.mgd.conn.mode >= IEEE80211_CONN_MODE_HE) { From ad55aa3ad8f843b2600db322d312f9f28d79568e Mon Sep 17 00:00:00 2001 From: Lachlan Hodges Date: Tue, 21 Oct 2025 17:10:51 +1100 Subject: [PATCH 0254/1381] wifi: mac80211: get probe response chan via ieee80211_get_channel_khz Make use of ieee80211_get_channel_khz() rather then the MHz counterpart to ensure probe responses received on an S1G channel pass the check. Signed-off-by: Lachlan Hodges Link: https://patch.msgid.link/20251021061051.235258-1-lachlan.hodges@morsemicro.com [modify indentation] Signed-off-by: Johannes Berg --- net/mac80211/mlme.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c index e699702fe5b1..025210d50405 100644 --- a/net/mac80211/mlme.c +++ b/net/mac80211/mlme.c @@ -6620,8 +6620,8 @@ static void ieee80211_rx_mgmt_probe_resp(struct ieee80211_link_data *link, * Response frame shall be set to the broadcast address [..]" * So, on 6GHz band we should also accept broadcast responses. */ - channel = ieee80211_get_channel(sdata->local->hw.wiphy, - rx_status->freq); + channel = ieee80211_get_channel_khz(sdata->local->hw.wiphy, + ieee80211_rx_status_to_khz(rx_status)); if (!channel) return; From bca76b875d0530658f3ba1bc946dbae1974f14c3 Mon Sep 17 00:00:00 2001 From: Lachlan Hodges Date: Tue, 21 Oct 2025 17:12:01 +1100 Subject: [PATCH 0255/1381] wifi: cfg80211: default S1G chandef width to 1MHz When management frames are passed down to be transmitted by usermode, often times the NL80211_ATTR_CHANNEL_WIDTH is not used as its implied to be transmitted on the control width. This can lead to errors during chandef validation as the offsets from the channel center are wrong. Ensure we initialise S1G chandefs to a width of 1MHz rather then 20MHz. Signed-off-by: Lachlan Hodges Link: https://patch.msgid.link/20251021061201.235754-1-lachlan.hodges@morsemicro.com Signed-off-by: Johannes Berg --- net/wireless/nl80211.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index 346dfd2bd987..ceca47cd9e25 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -3544,6 +3544,9 @@ static int _nl80211_parse_chandef(struct cfg80211_registered_device *rdev, return -EINVAL; } + if (cfg80211_chandef_is_s1g(chandef)) + chandef->width = NL80211_CHAN_WIDTH_1; + if (attrs[NL80211_ATTR_WIPHY_CHANNEL_TYPE]) { enum nl80211_channel_type chantype; From cc18fffa3a51792637169872886df3407bd5bb84 Mon Sep 17 00:00:00 2001 From: Sarika Sharma Date: Fri, 24 Oct 2025 10:06:27 +0530 Subject: [PATCH 0256/1381] wifi: mac80211: fix missing RX bitrate update for mesh forwarding path Currently, RX bitrate statistics are not updated for packets received on the mesh forwarding path during fast RX processing. This results in incomplete RX rate tracking in station dump outputs for mesh scenarios. Update ieee80211_invoke_fast_rx() to record the RX rate using sta_stats_encode_rate() and store it in the last_rate field of ieee80211_sta_rx_stats when RX_QUEUED is returned from ieee80211_rx_mesh_data(). This ensures that RX bitrate is properly accounted for in both RSS and non-RSS paths. Signed-off-by: Sarika Sharma Link: https://patch.msgid.link/20251024043627.1640447-1-sarika.sharma@oss.qualcomm.com Signed-off-by: Johannes Berg --- net/mac80211/rx.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c index 29175a0c9f68..4641a2a80856 100644 --- a/net/mac80211/rx.c +++ b/net/mac80211/rx.c @@ -4952,6 +4952,11 @@ static bool ieee80211_invoke_fast_rx(struct ieee80211_rx_data *rx, /* after this point, don't punt to the slowpath! */ + if (fast_rx->uses_rss) + stats = this_cpu_ptr(rx->link_sta->pcpu_rx_stats); + else + stats = &rx->link_sta->rx_stats; + if (rx->key && !(status->flag & RX_FLAG_MIC_STRIPPED) && pskb_trim(skb, skb->len - fast_rx->icv_len)) goto drop; @@ -4986,6 +4991,8 @@ static bool ieee80211_invoke_fast_rx(struct ieee80211_rx_data *rx, res = ieee80211_rx_mesh_data(rx->sdata, rx->sta, rx->skb); switch (res) { case RX_QUEUED: + stats->last_rx = jiffies; + stats->last_rate = sta_stats_encode_rate(status); return true; case RX_CONTINUE: break; @@ -4999,11 +5006,6 @@ static bool ieee80211_invoke_fast_rx(struct ieee80211_rx_data *rx, drop: dev_kfree_skb(skb); - if (fast_rx->uses_rss) - stats = this_cpu_ptr(rx->link_sta->pcpu_rx_stats); - else - stats = &rx->link_sta->rx_stats; - stats->dropped++; return true; } From 7cc986c04a9b07d91684f7e326fa5b960215bc97 Mon Sep 17 00:00:00 2001 From: Roopni Devanathan Date: Fri, 24 Oct 2025 10:16:48 +0530 Subject: [PATCH 0257/1381] wifi: cfg80211: Add debugfs support for multi-radio wiphy In multi-radio wiphy architecture, where a single wiphy can have multiple radios tied to it, radio specific configuration parameters and global wiphy parameters are maintained for the entire physical device and common to all radios. But, each radio in a wiphy can have different values for each radio configuration parameter, like RTS threshold. With the current debugfs directory structure, the values of global wiphy configuration parameters can be viewed, but, values of individual radio configuration parameters cannot be viewed, as radio specific configuration parameters are not maintained, separately. To address this, in addition to maintaining global wiphy configuration parameters common to all radios, create separate debugfs directories for each radio in a wiphy to maintain parameters corresponding to that radio in this directory. In implementation, maintain a dentry structure in wiphy_radio_cfg, a structure containing radio configurations of a wiphy. This struct is maintained to denote per-radio configurations of a wiphy. Create separate directories representing each radio within phy#X directory in debugfs during wiphy registration. Sample directory structure with this change: ls /sys/kernel/debug/ieee80211/phy0/radio radio0/ radio1/ radio2/ Signed-off-by: Roopni Devanathan Link: https://patch.msgid.link/20251024044649.483557-2-quic_rdevanat@quicinc.com Signed-off-by: Johannes Berg --- include/net/cfg80211.h | 4 ++++ net/wireless/core.c | 15 +++++++++++++++ 2 files changed, 19 insertions(+) diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index 3b6f48a783bb..53490eb04e87 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -5684,9 +5684,13 @@ struct wiphy_iftype_akm_suites { * * @rts_threshold: RTS threshold (dot11RTSThreshold); * -1 (default) = RTS/CTS disabled + * @radio_debugfsdir: Pointer to debugfs directory containing the radio- + * specific parameters. + * NULL (default) = Debugfs directory not created */ struct wiphy_radio_cfg { u32 rts_threshold; + struct dentry *radio_debugfsdir; }; /** diff --git a/net/wireless/core.c b/net/wireless/core.c index 797f9f2004a6..f3568eb5e592 100644 --- a/net/wireless/core.c +++ b/net/wireless/core.c @@ -34,6 +34,9 @@ /* name for sysfs, %d is appended */ #define PHY_NAME "phy" +/* maximum length of radio debugfs directory name */ +#define RADIO_DEBUGFSDIR_MAX_LEN 8 + MODULE_AUTHOR("Johannes Berg"); MODULE_LICENSE("GPL"); MODULE_DESCRIPTION("wireless configuration support"); @@ -1042,6 +1045,18 @@ int wiphy_register(struct wiphy *wiphy) /* add to debugfs */ rdev->wiphy.debugfsdir = debugfs_create_dir(wiphy_name(&rdev->wiphy), ieee80211_debugfs_dir); + if (wiphy->n_radio > 0) { + int idx; + char radio_name[RADIO_DEBUGFSDIR_MAX_LEN]; + + for (idx = 0; idx < wiphy->n_radio; idx++) { + scnprintf(radio_name, sizeof(radio_name), "radio%d", + idx); + wiphy->radio_cfg[idx].radio_debugfsdir = + debugfs_create_dir(radio_name, + rdev->wiphy.debugfsdir); + } + } cfg80211_debugfs_rdev_add(rdev); nl80211_notify_wiphy(rdev, NL80211_CMD_NEW_WIPHY); From 88de08348af8ce15dc563e0ebb5553eddd821c06 Mon Sep 17 00:00:00 2001 From: Roopni Devanathan Date: Fri, 24 Oct 2025 10:16:49 +0530 Subject: [PATCH 0258/1381] wifi: cfg80211: Add parameters to radio-specific debugfs directories In multi-radio wiphy architecture, where a single wiphy can have multiple radios tied to it, radio specific configuration parameters and global wiphy parameters are maintained for the entire physical device and common to all radios. But, each radio in a wiphy can have different values for each radio configuration parameter like RTS threshold. With the current debugfs directory structure, the values of global wiphy configuration parameters can be viewed, but, values of individual radio configuration parameters cannot be viewed. To address this requirement, maintain separate entries of each radio configuration parameter i.e., RTS threshold in corresponding radio- specific debugfs directory. This way, radio-specific configuration parameters can be maintained along with global wiphy configuration parameters. Whenever the values are changed for one radio, the values for rest of the radios in the wiphy and the global wiphy parameter value will remain intact. Sample output: /# iw phy#0 set rts 100 radio 1 /# iw phy#0 set rts 468 radio 0 /# cat /sys/kernel/debug/ieee80211/phy0/rts_threshold -1 /# cat /sys/kernel/debug/ieee80211/phy0/radio0/radio_rts_threshold 468 /# cat /sys/kernel/debug/ieee80211/phy0/radio1/radio_rts_threshold 100 /# iw phy#0 set rts 500 /# cat /sys/kernel/debug/ieee80211/phy0/rts_threshold 500 /# cat /sys/kernel/debug/ieee80211/phy0/radio0/radio_rts_threshold 500 /# cat /sys/kernel/debug/ieee80211/phy0/radio1/radio_rts_threshold 500 Signed-off-by: Roopni Devanathan Link: https://patch.msgid.link/20251024044649.483557-3-quic_rdevanat@quicinc.com Signed-off-by: Johannes Berg --- net/wireless/debugfs.c | 33 +++++++++++++++++++++++++++++++++ 1 file changed, 33 insertions(+) diff --git a/net/wireless/debugfs.c b/net/wireless/debugfs.c index 40e49074e2ee..f9e7fff1ef25 100644 --- a/net/wireless/debugfs.c +++ b/net/wireless/debugfs.c @@ -29,6 +29,24 @@ static const struct file_operations name## _ops = { \ .llseek = generic_file_llseek, \ } +#define DEBUGFS_RADIO_READONLY_FILE(name, buflen, fmt, value...) \ +static ssize_t name## _read(struct file *file, char __user *userbuf, \ + size_t count, loff_t *ppos) \ +{ \ + struct wiphy_radio_cfg *radio_cfg = file->private_data; \ + char buf[buflen]; \ + int res; \ + \ + res = scnprintf(buf, buflen, fmt "\n", ##value); \ + return simple_read_from_buffer(userbuf, count, ppos, buf, res); \ +} \ + \ +static const struct file_operations name## _ops = { \ + .read = name## _read, \ + .open = simple_open, \ + .llseek = generic_file_llseek, \ +} + DEBUGFS_READONLY_FILE(rts_threshold, 20, "%d", wiphy->rts_threshold); DEBUGFS_READONLY_FILE(fragmentation_threshold, 20, "%d", @@ -38,6 +56,9 @@ DEBUGFS_READONLY_FILE(short_retry_limit, 20, "%d", DEBUGFS_READONLY_FILE(long_retry_limit, 20, "%d", wiphy->retry_long); +DEBUGFS_RADIO_READONLY_FILE(radio_rts_threshold, 20, "%d", + radio_cfg->rts_threshold); + static int ht_print_chan(struct ieee80211_channel *chan, char *buf, int buf_size, int offset) { @@ -100,15 +121,27 @@ static const struct file_operations ht40allow_map_ops = { #define DEBUGFS_ADD(name) \ debugfs_create_file(#name, 0444, phyd, &rdev->wiphy, &name## _ops) +#define DEBUGFS_RADIO_ADD(name, radio_idx) \ + debugfs_create_file(#name, 0444, radiod, \ + &rdev->wiphy.radio_cfg[radio_idx], \ + &name## _ops) + void cfg80211_debugfs_rdev_add(struct cfg80211_registered_device *rdev) { struct dentry *phyd = rdev->wiphy.debugfsdir; + struct dentry *radiod; + u8 i; DEBUGFS_ADD(rts_threshold); DEBUGFS_ADD(fragmentation_threshold); DEBUGFS_ADD(short_retry_limit); DEBUGFS_ADD(long_retry_limit); DEBUGFS_ADD(ht40allow_map); + + for (i = 0; i < rdev->wiphy.n_radio; i++) { + radiod = rdev->wiphy.radio_cfg[i].radio_debugfsdir; + DEBUGFS_RADIO_ADD(radio_rts_threshold, i); + } } struct debugfs_read_work { From 10a11861943902fda74f37f456b45183b2bca270 Mon Sep 17 00:00:00 2001 From: Jianbo Liu Date: Tue, 21 Oct 2025 04:35:42 +0300 Subject: [PATCH 0259/1381] xfrm: Refactor xfrm_input lock to reduce contention with RSS With newer NICs like mlx5 supporting RSS for IPsec crypto offload, packets for a single Security Association (SA) are scattered across multiple CPU cores for parallel processing. The xfrm_state spinlock (x->lock) is held for each packet during xfrm processing. When multiple connections or flows share the same SA, this parallelism causes high lock contention on x->lock, creating a performance bottleneck and limiting scalability. The original xfrm_input() function exacerbated this issue by releasing and immediately re-acquiring x->lock. For hardware crypto offload paths, this unlock/relock sequence is unnecessary and introduces significant overhead. This patch refactors the function to relocate the type_offload->input_tail call for the offload path, performing all necessary work while continuously holding the lock. This reordering is safe, since packets which don't pass the checks below will still fail them with the new code. Performance testing with iperf using multiple parallel streams over a single IPsec SA shows significant improvement in throughput as the number of queues (and thus CPU cores) increases: +-----------+---------------+--------------+-----------------+ | RX queues | Before (Gbps) | After (Gbps) | Improvement (%) | +-----------+---------------+--------------+-----------------+ | 2 | 32.3 | 34.4 | 6.5 | | 4 | 34.4 | 40.0 | 16.3 | | 6 | 24.5 | 38.3 | 56.3 | | 8 | 23.1 | 38.3 | 65.8 | | 12 | 18.1 | 29.9 | 65.2 | | 16 | 16.0 | 25.2 | 57.5 | +-----------+---------------+--------------+-----------------+ Signed-off-by: Jianbo Liu Reviewed-by: Cosmin Ratiu Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_input.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/net/xfrm/xfrm_input.c b/net/xfrm/xfrm_input.c index c9ddef869aa5..257935cbd221 100644 --- a/net/xfrm/xfrm_input.c +++ b/net/xfrm/xfrm_input.c @@ -505,6 +505,7 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type) async = 1; dev_put(skb->dev); seq = XFRM_SKB_CB(skb)->seq.input.low; + spin_lock(&x->lock); goto resume; } /* GRO call */ @@ -541,6 +542,8 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type) XFRM_INC_STATS(net, LINUX_MIB_XFRMINHDRERROR); goto drop; } + + nexthdr = x->type_offload->input_tail(x, skb); } goto lock; @@ -638,11 +641,9 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type) goto drop_unlock; } - spin_unlock(&x->lock); - if (xfrm_tunnel_check(skb, x, family)) { XFRM_INC_STATS(net, LINUX_MIB_XFRMINSTATEMODEERROR); - goto drop; + goto drop_unlock; } seq_hi = htonl(xfrm_replay_seqhi(x, seq)); @@ -650,9 +651,8 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type) XFRM_SKB_CB(skb)->seq.input.low = seq; XFRM_SKB_CB(skb)->seq.input.hi = seq_hi; - if (crypto_done) { - nexthdr = x->type_offload->input_tail(x, skb); - } else { + if (!crypto_done) { + spin_unlock(&x->lock); dev_hold(skb->dev); nexthdr = x->type->input(x, skb); @@ -660,9 +660,9 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type) return 0; dev_put(skb->dev); + spin_lock(&x->lock); } resume: - spin_lock(&x->lock); if (nexthdr < 0) { if (nexthdr == -EBADMSG) { xfrm_audit_state_icvfail(x, skb, From b427c0c3bc40cca268a5d54a1cdf6166cb1360e2 Mon Sep 17 00:00:00 2001 From: Jianbo Liu Date: Tue, 21 Oct 2025 04:35:43 +0300 Subject: [PATCH 0260/1381] xfrm: Skip redundant replay recheck for the hardware offload path The xfrm_replay_recheck() function was introduced to handle the issues arising from asynchronous crypto algorithms. The crypto offload path is now effectively synchronous, as it holds the state lock throughout its operation. This eliminates the race condition, making the recheck an unnecessary overhead. This patch improves performance by skipping the redundant call when crypto_done is true. Additionally, the sequence number assignment is moved to an earlier point in the function. This improves performance by reducing lock contention and places the logic at a more appropriate point, as the full sequence number (including the higher-order bits) can be determined as soon as the packet is received. Signed-off-by: Jianbo Liu Reviewed-by: Cosmin Ratiu Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_input.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/net/xfrm/xfrm_input.c b/net/xfrm/xfrm_input.c index 257935cbd221..4ed346e682c7 100644 --- a/net/xfrm/xfrm_input.c +++ b/net/xfrm/xfrm_input.c @@ -546,7 +546,7 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type) nexthdr = x->type_offload->input_tail(x, skb); } - goto lock; + goto process; } family = XFRM_SPI_SKB_CB(skb)->family; @@ -614,7 +614,12 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type) goto drop; } -lock: +process: + seq_hi = htonl(xfrm_replay_seqhi(x, seq)); + + XFRM_SKB_CB(skb)->seq.input.low = seq; + XFRM_SKB_CB(skb)->seq.input.hi = seq_hi; + spin_lock(&x->lock); if (unlikely(x->km.state != XFRM_STATE_VALID)) { @@ -646,11 +651,6 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type) goto drop_unlock; } - seq_hi = htonl(xfrm_replay_seqhi(x, seq)); - - XFRM_SKB_CB(skb)->seq.input.low = seq; - XFRM_SKB_CB(skb)->seq.input.hi = seq_hi; - if (!crypto_done) { spin_unlock(&x->lock); dev_hold(skb->dev); @@ -676,7 +676,7 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type) /* only the first xfrm gets the encap type */ encap_type = 0; - if (xfrm_replay_recheck(x, skb, seq)) { + if (!crypto_done && xfrm_replay_recheck(x, skb, seq)) { XFRM_INC_STATS(net, LINUX_MIB_XFRMINSTATESEQERROR); goto drop_unlock; } From 82cb5be6ad64198a3a028aeb49dcc7f6224d558a Mon Sep 17 00:00:00 2001 From: Wilfred Mallawa Date: Wed, 22 Oct 2025 10:19:36 +1000 Subject: [PATCH 0261/1381] net/tls: support setting the maximum payload size During a handshake, an endpoint may specify a maximum record size limit. Currently, the kernel defaults to TLS_MAX_PAYLOAD_SIZE (16KB) for the maximum record size. Meaning that, the outgoing records from the kernel can exceed a lower size negotiated during the handshake. In such a case, the TLS endpoint must send a fatal "record_overflow" alert [1], and thus the record is discarded. Upcoming Western Digital NVMe-TCP hardware controllers implement TLS support. For these devices, supporting TLS record size negotiation is necessary because the maximum TLS record size supported by the controller is less than the default 16KB currently used by the kernel. Currently, there is no way to inform the kernel of such a limit. This patch adds support to a new setsockopt() option `TLS_TX_MAX_PAYLOAD_LEN` that allows for setting the maximum plaintext fragment size. Once set, outgoing records are no larger than the size specified. This option can be used to specify the record size limit. [1] https://www.rfc-editor.org/rfc/rfc8449 Signed-off-by: Wilfred Mallawa Reviewed-by: Sabrina Dubroca Link: https://patch.msgid.link/20251022001937.20155-1-wilfred.opensource@gmail.com Signed-off-by: Jakub Kicinski --- Documentation/networking/tls.rst | 20 ++++++++++ include/net/tls.h | 3 ++ include/uapi/linux/tls.h | 2 + net/tls/tls_device.c | 2 +- net/tls/tls_main.c | 64 ++++++++++++++++++++++++++++++++ net/tls/tls_sw.c | 2 +- 6 files changed, 91 insertions(+), 2 deletions(-) diff --git a/Documentation/networking/tls.rst b/Documentation/networking/tls.rst index 36cc7afc2527..980c442d7161 100644 --- a/Documentation/networking/tls.rst +++ b/Documentation/networking/tls.rst @@ -280,6 +280,26 @@ If the record decrypted turns out to had been padded or is not a data record it will be decrypted again into a kernel buffer without zero copy. Such events are counted in the ``TlsDecryptRetry`` statistic. +TLS_TX_MAX_PAYLOAD_LEN +~~~~~~~~~~~~~~~~~~~~~~ + +Specifies the maximum size of the plaintext payload for transmitted TLS records. + +When this option is set, the kernel enforces the specified limit on all outgoing +TLS records. No plaintext fragment will exceed this size. This option can be used +to implement the TLS Record Size Limit extension [1]. + +* For TLS 1.2, the value corresponds directly to the record size limit. +* For TLS 1.3, the value should be set to record_size_limit - 1, since + the record size limit includes one additional byte for the ContentType + field. + +The valid range for this option is 64 to 16384 bytes for TLS 1.2, and 63 to +16384 bytes for TLS 1.3. The lower minimum for TLS 1.3 accounts for the +extra byte used by the ContentType field. + +[1] https://datatracker.ietf.org/doc/html/rfc8449 + Statistics ========== diff --git a/include/net/tls.h b/include/net/tls.h index 857340338b69..f2af113728aa 100644 --- a/include/net/tls.h +++ b/include/net/tls.h @@ -53,6 +53,8 @@ struct tls_rec; /* Maximum data size carried in a TLS record */ #define TLS_MAX_PAYLOAD_SIZE ((size_t)1 << 14) +/* Minimum record size limit as per RFC8449 */ +#define TLS_MIN_RECORD_SIZE_LIM ((size_t)1 << 6) #define TLS_HEADER_SIZE 5 #define TLS_NONCE_OFFSET TLS_HEADER_SIZE @@ -226,6 +228,7 @@ struct tls_context { u8 rx_conf:3; u8 zerocopy_sendfile:1; u8 rx_no_pad:1; + u16 tx_max_payload_len; int (*push_pending_record)(struct sock *sk, int flags); void (*sk_write_space)(struct sock *sk); diff --git a/include/uapi/linux/tls.h b/include/uapi/linux/tls.h index b66a800389cc..b8b9c42f848c 100644 --- a/include/uapi/linux/tls.h +++ b/include/uapi/linux/tls.h @@ -41,6 +41,7 @@ #define TLS_RX 2 /* Set receive parameters */ #define TLS_TX_ZEROCOPY_RO 3 /* TX zerocopy (only sendfile now) */ #define TLS_RX_EXPECT_NO_PAD 4 /* Attempt opportunistic zero-copy */ +#define TLS_TX_MAX_PAYLOAD_LEN 5 /* Maximum plaintext size */ /* Supported versions */ #define TLS_VERSION_MINOR(ver) ((ver) & 0xFF) @@ -194,6 +195,7 @@ enum { TLS_INFO_RXCONF, TLS_INFO_ZC_RO_TX, TLS_INFO_RX_NO_PAD, + TLS_INFO_TX_MAX_PAYLOAD_LEN, __TLS_INFO_MAX, }; #define TLS_INFO_MAX (__TLS_INFO_MAX - 1) diff --git a/net/tls/tls_device.c b/net/tls/tls_device.c index caa2b5d24622..4d29b390aed9 100644 --- a/net/tls/tls_device.c +++ b/net/tls/tls_device.c @@ -462,7 +462,7 @@ static int tls_push_data(struct sock *sk, /* TLS_HEADER_SIZE is not counted as part of the TLS record, and * we need to leave room for an authentication tag. */ - max_open_record_len = TLS_MAX_PAYLOAD_SIZE + + max_open_record_len = tls_ctx->tx_max_payload_len + prot->prepend_size; do { rc = tls_do_allocation(sk, ctx, pfrag, prot->prepend_size); diff --git a/net/tls/tls_main.c b/net/tls/tls_main.c index 39a2ab47fe72..56ce0bc8317b 100644 --- a/net/tls/tls_main.c +++ b/net/tls/tls_main.c @@ -541,6 +541,28 @@ static int do_tls_getsockopt_no_pad(struct sock *sk, char __user *optval, return 0; } +static int do_tls_getsockopt_tx_payload_len(struct sock *sk, char __user *optval, + int __user *optlen) +{ + struct tls_context *ctx = tls_get_ctx(sk); + u16 payload_len = ctx->tx_max_payload_len; + int len; + + if (get_user(len, optlen)) + return -EFAULT; + + if (len < sizeof(payload_len)) + return -EINVAL; + + if (put_user(sizeof(payload_len), optlen)) + return -EFAULT; + + if (copy_to_user(optval, &payload_len, sizeof(payload_len))) + return -EFAULT; + + return 0; +} + static int do_tls_getsockopt(struct sock *sk, int optname, char __user *optval, int __user *optlen) { @@ -560,6 +582,9 @@ static int do_tls_getsockopt(struct sock *sk, int optname, case TLS_RX_EXPECT_NO_PAD: rc = do_tls_getsockopt_no_pad(sk, optval, optlen); break; + case TLS_TX_MAX_PAYLOAD_LEN: + rc = do_tls_getsockopt_tx_payload_len(sk, optval, optlen); + break; default: rc = -ENOPROTOOPT; break; @@ -809,6 +834,32 @@ static int do_tls_setsockopt_no_pad(struct sock *sk, sockptr_t optval, return rc; } +static int do_tls_setsockopt_tx_payload_len(struct sock *sk, sockptr_t optval, + unsigned int optlen) +{ + struct tls_context *ctx = tls_get_ctx(sk); + struct tls_sw_context_tx *sw_ctx = tls_sw_ctx_tx(ctx); + u16 value; + bool tls_13 = ctx->prot_info.version == TLS_1_3_VERSION; + + if (sw_ctx && sw_ctx->open_rec) + return -EBUSY; + + if (sockptr_is_null(optval) || optlen != sizeof(value)) + return -EINVAL; + + if (copy_from_sockptr(&value, optval, sizeof(value))) + return -EFAULT; + + if (value < TLS_MIN_RECORD_SIZE_LIM - (tls_13 ? 1 : 0) || + value > TLS_MAX_PAYLOAD_SIZE) + return -EINVAL; + + ctx->tx_max_payload_len = value; + + return 0; +} + static int do_tls_setsockopt(struct sock *sk, int optname, sockptr_t optval, unsigned int optlen) { @@ -830,6 +881,11 @@ static int do_tls_setsockopt(struct sock *sk, int optname, sockptr_t optval, case TLS_RX_EXPECT_NO_PAD: rc = do_tls_setsockopt_no_pad(sk, optval, optlen); break; + case TLS_TX_MAX_PAYLOAD_LEN: + lock_sock(sk); + rc = do_tls_setsockopt_tx_payload_len(sk, optval, optlen); + release_sock(sk); + break; default: rc = -ENOPROTOOPT; break; @@ -1019,6 +1075,7 @@ static int tls_init(struct sock *sk) ctx->tx_conf = TLS_BASE; ctx->rx_conf = TLS_BASE; + ctx->tx_max_payload_len = TLS_MAX_PAYLOAD_SIZE; update_sk_prot(sk, ctx); out: write_unlock_bh(&sk->sk_callback_lock); @@ -1108,6 +1165,12 @@ static int tls_get_info(struct sock *sk, struct sk_buff *skb, bool net_admin) goto nla_failure; } + err = nla_put_u16(skb, TLS_INFO_TX_MAX_PAYLOAD_LEN, + ctx->tx_max_payload_len); + + if (err) + goto nla_failure; + rcu_read_unlock(); nla_nest_end(skb, start); return 0; @@ -1129,6 +1192,7 @@ static size_t tls_get_info_size(const struct sock *sk, bool net_admin) nla_total_size(sizeof(u16)) + /* TLS_INFO_TXCONF */ nla_total_size(0) + /* TLS_INFO_ZC_RO_TX */ nla_total_size(0) + /* TLS_INFO_RX_NO_PAD */ + nla_total_size(sizeof(u16)) + /* TLS_INFO_TX_MAX_PAYLOAD_LEN */ 0; return size; diff --git a/net/tls/tls_sw.c b/net/tls/tls_sw.c index d17135369980..9937d4c810f2 100644 --- a/net/tls/tls_sw.c +++ b/net/tls/tls_sw.c @@ -1079,7 +1079,7 @@ static int tls_sw_sendmsg_locked(struct sock *sk, struct msghdr *msg, orig_size = msg_pl->sg.size; full_record = false; try_to_copy = msg_data_left(msg); - record_room = TLS_MAX_PAYLOAD_SIZE - msg_pl->sg.size; + record_room = tls_ctx->tx_max_payload_len - msg_pl->sg.size; if (try_to_copy >= record_room) { try_to_copy = record_room; full_record = true; From 5f30bc470672f7b38a60d6641d519f308723085c Mon Sep 17 00:00:00 2001 From: Wilfred Mallawa Date: Wed, 22 Oct 2025 10:19:37 +1000 Subject: [PATCH 0262/1381] selftests: tls: add tls record_size_limit test Test that outgoing plaintext records respect the tls TLS_TX_MAX_PAYLOAD_LEN set using setsockopt(). The limit is set to be 128, thus, in all received records, the plaintext must not exceed this amount. Also test that setting a new record size limit whilst a pending open record exists is handled correctly by discarding the request. Suggested-by: Sabrina Dubroca Signed-off-by: Wilfred Mallawa Reviewed-by: Sabrina Dubroca Link: https://patch.msgid.link/20251022001937.20155-2-wilfred.opensource@gmail.com Signed-off-by: Jakub Kicinski --- tools/testing/selftests/net/tls.c | 141 ++++++++++++++++++++++++++++++ 1 file changed, 141 insertions(+) diff --git a/tools/testing/selftests/net/tls.c b/tools/testing/selftests/net/tls.c index 5c6d8215021c..da1b50b30719 100644 --- a/tools/testing/selftests/net/tls.c +++ b/tools/testing/selftests/net/tls.c @@ -2856,6 +2856,147 @@ TEST_F(tls_err, oob_pressure) EXPECT_EQ(send(self->fd2, buf, 5, MSG_OOB), 5); } +/* + * Parse a stream of TLS records and ensure that each record respects + * the specified @max_payload_len. + */ +static size_t parse_tls_records(struct __test_metadata *_metadata, + const __u8 *rx_buf, int rx_len, int overhead, + __u16 max_payload_len) +{ + const __u8 *rec = rx_buf; + size_t total_plaintext_rx = 0; + const __u8 rec_header_len = 5; + + while (rec < rx_buf + rx_len) { + __u16 record_payload_len; + __u16 plaintext_len; + + /* Sanity check that it's a TLS header for application data */ + ASSERT_EQ(rec[0], 23); + ASSERT_EQ(rec[1], 0x3); + ASSERT_EQ(rec[2], 0x3); + + memcpy(&record_payload_len, rec + 3, 2); + record_payload_len = ntohs(record_payload_len); + ASSERT_GE(record_payload_len, overhead); + + plaintext_len = record_payload_len - overhead; + total_plaintext_rx += plaintext_len; + + /* Plaintext must not exceed the specified limit */ + ASSERT_LE(plaintext_len, max_payload_len); + rec += rec_header_len + record_payload_len; + } + + return total_plaintext_rx; +} + +TEST(tls_12_tx_max_payload_len) +{ + struct tls_crypto_info_keys tls12; + int cfd, ret, fd, overhead; + size_t total_plaintext_rx = 0; + __u8 tx[1024], rx[2000]; + __u16 limit = 128; + __u16 opt = 0; + unsigned int optlen = sizeof(opt); + bool notls; + + tls_crypto_info_init(TLS_1_2_VERSION, TLS_CIPHER_AES_CCM_128, + &tls12, 0); + + ulp_sock_pair(_metadata, &fd, &cfd, ¬ls); + + if (notls) + exit(KSFT_SKIP); + + /* Don't install keys on fd, we'll parse raw records */ + ret = setsockopt(cfd, SOL_TLS, TLS_TX, &tls12, tls12.len); + ASSERT_EQ(ret, 0); + + ret = setsockopt(cfd, SOL_TLS, TLS_TX_MAX_PAYLOAD_LEN, &limit, + sizeof(limit)); + ASSERT_EQ(ret, 0); + + ret = getsockopt(cfd, SOL_TLS, TLS_TX_MAX_PAYLOAD_LEN, &opt, &optlen); + EXPECT_EQ(ret, 0); + EXPECT_EQ(limit, opt); + EXPECT_EQ(optlen, sizeof(limit)); + + memset(tx, 0, sizeof(tx)); + ASSERT_EQ(send(cfd, tx, sizeof(tx), 0), sizeof(tx)); + close(cfd); + + ret = recv(fd, rx, sizeof(rx), 0); + + /* + * 16B tag + 8B IV -- record header (5B) is not counted but we'll + * need it to walk the record stream + */ + overhead = 16 + 8; + total_plaintext_rx = parse_tls_records(_metadata, rx, ret, overhead, + limit); + + ASSERT_EQ(total_plaintext_rx, sizeof(tx)); + close(fd); +} + +TEST(tls_12_tx_max_payload_len_open_rec) +{ + struct tls_crypto_info_keys tls12; + int cfd, ret, fd, overhead; + size_t total_plaintext_rx = 0; + __u8 tx[1024], rx[2000]; + __u16 tx_partial = 256; + __u16 og_limit = 512, limit = 128; + bool notls; + + tls_crypto_info_init(TLS_1_2_VERSION, TLS_CIPHER_AES_CCM_128, + &tls12, 0); + + ulp_sock_pair(_metadata, &fd, &cfd, ¬ls); + + if (notls) + exit(KSFT_SKIP); + + /* Don't install keys on fd, we'll parse raw records */ + ret = setsockopt(cfd, SOL_TLS, TLS_TX, &tls12, tls12.len); + ASSERT_EQ(ret, 0); + + ret = setsockopt(cfd, SOL_TLS, TLS_TX_MAX_PAYLOAD_LEN, &og_limit, + sizeof(og_limit)); + ASSERT_EQ(ret, 0); + + memset(tx, 0, sizeof(tx)); + ASSERT_EQ(send(cfd, tx, tx_partial, MSG_MORE), tx_partial); + + /* + * Changing the payload limit with a pending open record should + * not be allowed. + */ + ret = setsockopt(cfd, SOL_TLS, TLS_TX_MAX_PAYLOAD_LEN, &limit, + sizeof(limit)); + ASSERT_EQ(ret, -1); + ASSERT_EQ(errno, EBUSY); + + ASSERT_EQ(send(cfd, tx + tx_partial, sizeof(tx) - tx_partial, MSG_EOR), + sizeof(tx) - tx_partial); + close(cfd); + + ret = recv(fd, rx, sizeof(rx), 0); + + /* + * 16B tag + 8B IV -- record header (5B) is not counted but we'll + * need it to walk the record stream + */ + overhead = 16 + 8; + total_plaintext_rx = parse_tls_records(_metadata, rx, ret, overhead, + og_limit); + ASSERT_EQ(total_plaintext_rx, sizeof(tx)); + close(fd); +} + TEST(non_established) { struct tls12_crypto_info_aes_gcm_256 tls12; struct sockaddr_in addr; From 68800bbf583f26f71491141e4b3c8582f9cfcbde Mon Sep 17 00:00:00 2001 From: Petr Machata Date: Thu, 23 Oct 2025 16:45:37 +0200 Subject: [PATCH 0263/1381] net: bridge: Flush multicast groups when snooping is disabled When forwarding multicast packets, the bridge takes MDB into account when IGMP / MLD snooping is enabled. Currently, when snooping is disabled, the MDB is retained, even though it is not used anymore. At the same time, during the time that snooping is disabled, the IGMP / MLD control packets are obviously ignored, and after the snooping is reenabled, the administrator has to assume it is out of sync. In particular, missed join and leave messages would lead to traffic being forwarded to wrong interfaces. Keeping the MDB entries around thus serves no purpose, and just takes memory. Note also that disabling per-VLAN snooping does actually flush the relevant MDB entries. This patch flushes non-permanent MDB entries as global snooping is disabled. Signed-off-by: Petr Machata Reviewed-by: Ido Schimmel Acked-by: Nikolay Aleksandrov Link: https://patch.msgid.link/5e992df1bb93b88e19c0ea5819e23b669e3dde5d.1761228273.git.petrm@nvidia.com Signed-off-by: Jakub Kicinski --- net/bridge/br_multicast.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/net/bridge/br_multicast.c b/net/bridge/br_multicast.c index 22d12e545966..d55a4ab87837 100644 --- a/net/bridge/br_multicast.c +++ b/net/bridge/br_multicast.c @@ -4649,6 +4649,14 @@ static void br_multicast_start_querier(struct net_bridge_mcast *brmctx, rcu_read_unlock(); } +static void br_multicast_del_grps(struct net_bridge *br) +{ + struct net_bridge_port *port; + + list_for_each_entry(port, &br->port_list, list) + __br_multicast_disable_port_ctx(&port->multicast_ctx); +} + int br_multicast_toggle(struct net_bridge *br, unsigned long val, struct netlink_ext_ack *extack) { @@ -4669,6 +4677,7 @@ int br_multicast_toggle(struct net_bridge *br, unsigned long val, br_opt_toggle(br, BROPT_MULTICAST_ENABLED, !!val); if (!br_opt_get(br, BROPT_MULTICAST_ENABLED)) { change_snoopers = true; + br_multicast_del_grps(br); goto unlock; } From d10920607ffedc6b8d27c7483d9dfe0cff0b2fa8 Mon Sep 17 00:00:00 2001 From: Petr Machata Date: Thu, 23 Oct 2025 16:45:38 +0200 Subject: [PATCH 0264/1381] selftests: bridge_mdb: Add a test for MDB flush on snooping disable Check that non-permanent MDB entries are removed as IGMP / MLD snooping is disabled. Signed-off-by: Petr Machata Reviewed-by: Ido Schimmel Acked-by: Nikolay Aleksandrov Link: https://patch.msgid.link/9420dfbcf26c8e1134d31244e9e7d6a49d677a69.1761228273.git.petrm@nvidia.com Signed-off-by: Jakub Kicinski --- .../selftests/net/forwarding/bridge_mdb.sh | 100 +++++++++++++++++- 1 file changed, 98 insertions(+), 2 deletions(-) diff --git a/tools/testing/selftests/net/forwarding/bridge_mdb.sh b/tools/testing/selftests/net/forwarding/bridge_mdb.sh index 8c1597ebc2d3..e86d77946585 100755 --- a/tools/testing/selftests/net/forwarding/bridge_mdb.sh +++ b/tools/testing/selftests/net/forwarding/bridge_mdb.sh @@ -28,6 +28,7 @@ ALL_TESTS=" cfg_test fwd_test ctrl_test + disable_test " NUM_NETIFS=4 @@ -64,7 +65,10 @@ h2_destroy() switch_create() { - ip link add name br0 type bridge vlan_filtering 1 vlan_default_pvid 0 \ + local vlan_filtering=$1; shift + + ip link add name br0 type bridge \ + vlan_filtering "$vlan_filtering" vlan_default_pvid 0 \ mcast_snooping 1 mcast_igmp_version 3 mcast_mld_version 2 bridge vlan add vid 10 dev br0 self bridge vlan add vid 20 dev br0 self @@ -118,7 +122,7 @@ setup_prepare() h1_create h2_create - switch_create + switch_create 1 } cleanup() @@ -1357,6 +1361,98 @@ ctrl_test() ctrl_mldv2_is_in_test } +check_group() +{ + local group=$1; shift + local vid=$1; shift + local should_fail=$1; shift + local when=$1; shift + local -a vidkws + + if ((vid)); then + vidkws=(vid "$vid") + fi + + bridge mdb get dev br0 grp "$group" "${vidkws[@]}" 2>/dev/null | + grep -q "port $swp1" + check_err_fail "$should_fail" $? "$group seen $when snooping disable:" +} + +__disable_test() +{ + local vid=$1; shift + local what=$1; shift + local -a vidkws + + if ((vid)); then + vidkws=(vid "$vid") + fi + + RET=0 + + bridge mdb add dev br0 port "$swp1" grp ff0e::1 permanent \ + "${vidkws[@]}" filter_mode include source_list 2001:db8:1::1 + bridge mdb add dev br0 port "$swp1" grp ff0e::2 permanent \ + "${vidkws[@]}" filter_mode exclude + + bridge mdb add dev br0 port "$swp1" grp ff0e::3 \ + "${vidkws[@]}" filter_mode include source_list 2001:db8:1::2 + bridge mdb add dev br0 port "$swp1" grp ff0e::4 \ + "${vidkws[@]}" filter_mode exclude + + bridge mdb add dev br0 port "$swp1" grp 239.1.1.1 permanent \ + "${vidkws[@]}" filter_mode include source_list 192.0.2.1 + bridge mdb add dev br0 port "$swp1" grp 239.1.1.2 permanent \ + "${vidkws[@]}" filter_mode exclude + + bridge mdb add dev br0 port "$swp1" grp 239.1.1.3 \ + "${vidkws[@]}" filter_mode include source_list 192.0.2.2 + bridge mdb add dev br0 port "$swp1" grp 239.1.1.4 \ + "${vidkws[@]}" filter_mode exclude + + check_group ff0e::1 "$vid" 0 "before" + check_group ff0e::2 "$vid" 0 "before" + check_group ff0e::3 "$vid" 0 "before" + check_group ff0e::4 "$vid" 0 "before" + + check_group 239.1.1.1 "$vid" 0 "before" + check_group 239.1.1.2 "$vid" 0 "before" + check_group 239.1.1.3 "$vid" 0 "before" + check_group 239.1.1.4 "$vid" 0 "before" + + ip link set dev br0 type bridge mcast_snooping 0 + + check_group ff0e::1 "$vid" 0 "after" + check_group ff0e::2 "$vid" 0 "after" + check_group ff0e::3 "$vid" 1 "after" + check_group ff0e::4 "$vid" 1 "after" + + check_group 239.1.1.1 "$vid" 0 "after" + check_group 239.1.1.2 "$vid" 0 "after" + check_group 239.1.1.3 "$vid" 1 "after" + check_group 239.1.1.4 "$vid" 1 "after" + + log_test "$what: Flush after disable" + + ip link set dev br0 type bridge mcast_snooping 1 + sleep 10 +} + +disable_test() +{ + __disable_test 10 802.1q + + switch_destroy + switch_create 0 + setup_wait + + __disable_test 0 802.1d + + switch_destroy + switch_create 1 + setup_wait +} + if ! bridge mdb help 2>&1 | grep -q "flush"; then echo "SKIP: iproute2 too old, missing bridge mdb flush support" exit $ksft_skip From 1bc80d673087e5704adbb3ee8e4b785c14899cce Mon Sep 17 00:00:00 2001 From: Horatiu Vultur Date: Thu, 23 Oct 2025 21:13:49 +0200 Subject: [PATCH 0265/1381] phy: mscc: Use PHY_ID_MATCH_EXACT for VSC8584, VSC8582, VSC8575, VSC856X As the PHYs VSC8584, VSC8582, VSC8575 and VSC856X exists only as rev B, we can use PHY_ID_MATCH_EXACT to match exactly on revision B of the PHY. Because of this change then there is not need the check if it is a different revision than rev B in the function vsc8584_probe() as we already know that this will never happen. These changes are a preparation for the next patch because in that patch we will make the PHYs VSC8574 and VSC8572 to use vsc8584_probe() and these PHYs have multiple revision. Reviewed-by: Maxime Chevallier Signed-off-by: Horatiu Vultur Link: https://patch.msgid.link/20251023191350.190940-2-horatiu.vultur@microchip.com Signed-off-by: Jakub Kicinski --- drivers/net/phy/mscc/mscc.h | 8 ++++---- drivers/net/phy/mscc/mscc_main.c | 23 ++++------------------- 2 files changed, 8 insertions(+), 23 deletions(-) diff --git a/drivers/net/phy/mscc/mscc.h b/drivers/net/phy/mscc/mscc.h index 2d8eca54c40a..2eef5956b9cc 100644 --- a/drivers/net/phy/mscc/mscc.h +++ b/drivers/net/phy/mscc/mscc.h @@ -289,12 +289,12 @@ enum rgmii_clock_delay { #define PHY_ID_VSC8540 0x00070760 #define PHY_ID_VSC8541 0x00070770 #define PHY_ID_VSC8552 0x000704e0 -#define PHY_ID_VSC856X 0x000707e0 +#define PHY_ID_VSC856X 0x000707e1 #define PHY_ID_VSC8572 0x000704d0 #define PHY_ID_VSC8574 0x000704a0 -#define PHY_ID_VSC8575 0x000707d0 -#define PHY_ID_VSC8582 0x000707b0 -#define PHY_ID_VSC8584 0x000707c0 +#define PHY_ID_VSC8575 0x000707d1 +#define PHY_ID_VSC8582 0x000707b1 +#define PHY_ID_VSC8584 0x000707c1 #define PHY_VENDOR_MSCC 0x00070400 #define MSCC_VDDMAC_1500 1500 diff --git a/drivers/net/phy/mscc/mscc_main.c b/drivers/net/phy/mscc/mscc_main.c index ef0ef1570d39..9343ed3b000d 100644 --- a/drivers/net/phy/mscc/mscc_main.c +++ b/drivers/net/phy/mscc/mscc_main.c @@ -1724,12 +1724,6 @@ static int vsc8584_config_init(struct phy_device *phydev) * in this pre-init function. */ if (phy_package_init_once(phydev)) { - /* The following switch statement assumes that the lowest - * nibble of the phy_id_mask is always 0. This works because - * the lowest nibble of the PHY_ID's below are also 0. - */ - WARN_ON(phydev->drv->phy_id_mask & 0xf); - switch (phydev->phy_id & phydev->drv->phy_id_mask) { case PHY_ID_VSC8504: case PHY_ID_VSC8552: @@ -2290,11 +2284,6 @@ static int vsc8584_probe(struct phy_device *phydev) VSC8531_DUPLEX_COLLISION}; int ret; - if ((phydev->phy_id & MSCC_DEV_REV_MASK) != VSC8584_REVB) { - dev_err(&phydev->mdio.dev, "Only VSC8584 revB is supported.\n"); - return -ENOTSUPP; - } - vsc8531 = devm_kzalloc(&phydev->mdio.dev, sizeof(*vsc8531), GFP_KERNEL); if (!vsc8531) return -ENOMEM; @@ -2587,9 +2576,8 @@ static struct phy_driver vsc85xx_driver[] = { .config_inband = vsc85xx_config_inband, }, { - .phy_id = PHY_ID_VSC856X, + PHY_ID_MATCH_EXACT(PHY_ID_VSC856X), .name = "Microsemi GE VSC856X SyncE", - .phy_id_mask = 0xfffffff0, /* PHY_GBIT_FEATURES */ .soft_reset = &genphy_soft_reset, .config_init = &vsc8584_config_init, @@ -2667,9 +2655,8 @@ static struct phy_driver vsc85xx_driver[] = { .config_inband = vsc85xx_config_inband, }, { - .phy_id = PHY_ID_VSC8575, + PHY_ID_MATCH_EXACT(PHY_ID_VSC8575), .name = "Microsemi GE VSC8575 SyncE", - .phy_id_mask = 0xfffffff0, /* PHY_GBIT_FEATURES */ .soft_reset = &genphy_soft_reset, .config_init = &vsc8584_config_init, @@ -2693,9 +2680,8 @@ static struct phy_driver vsc85xx_driver[] = { .config_inband = vsc85xx_config_inband, }, { - .phy_id = PHY_ID_VSC8582, + PHY_ID_MATCH_EXACT(PHY_ID_VSC8582), .name = "Microsemi GE VSC8582 SyncE", - .phy_id_mask = 0xfffffff0, /* PHY_GBIT_FEATURES */ .soft_reset = &genphy_soft_reset, .config_init = &vsc8584_config_init, @@ -2719,9 +2705,8 @@ static struct phy_driver vsc85xx_driver[] = { .config_inband = vsc85xx_config_inband, }, { - .phy_id = PHY_ID_VSC8584, + PHY_ID_MATCH_EXACT(PHY_ID_VSC8584), .name = "Microsemi GE VSC8584 SyncE", - .phy_id_mask = 0xfffffff0, /* PHY_GBIT_FEATURES */ .soft_reset = &genphy_soft_reset, .config_init = &vsc8584_config_init, From ea5df88aeca112aac69e6c32e3dd1433a113b0c9 Mon Sep 17 00:00:00 2001 From: Horatiu Vultur Date: Thu, 23 Oct 2025 21:13:50 +0200 Subject: [PATCH 0266/1381] phy: mscc: Fix PTP for VSC8574 and VSC8572 The PTP initialization is two-step. First part are the function vsc8584_ptp_probe_once() and vsc8584_ptp_probe() at probe time which initialize the locks, queues, creates the PTP device. The second part is the function vsc8584_ptp_init() at config_init() time which initialize PTP in the HW. For VSC8574 and VSC8572, the PTP initialization is incomplete. It is missing the first part but it makes the second part. Meaning that the ptp_clock_register() is never called. There is no crash without the first part when enabling PTP but this is unexpected because some PHys have PTP functionality exposed by the driver and some don't even though they share the same PTP clock PTP. Fixes: 774626fa440e ("net: phy: mscc: Add PTP support for 2 more VSC PHYs") Reviewed-by: Maxime Chevallier Signed-off-by: Horatiu Vultur Link: https://patch.msgid.link/20251023191350.190940-3-horatiu.vultur@microchip.com Signed-off-by: Jakub Kicinski --- drivers/net/phy/mscc/mscc_main.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/net/phy/mscc/mscc_main.c b/drivers/net/phy/mscc/mscc_main.c index 9343ed3b000d..8678ebf89cca 100644 --- a/drivers/net/phy/mscc/mscc_main.c +++ b/drivers/net/phy/mscc/mscc_main.c @@ -2613,7 +2613,7 @@ static struct phy_driver vsc85xx_driver[] = { .suspend = &genphy_suspend, .resume = &genphy_resume, .remove = &vsc85xx_remove, - .probe = &vsc8574_probe, + .probe = &vsc8584_probe, .set_wol = &vsc85xx_wol_set, .get_wol = &vsc85xx_wol_get, .get_tunable = &vsc85xx_get_tunable, @@ -2636,12 +2636,12 @@ static struct phy_driver vsc85xx_driver[] = { .config_aneg = &vsc85xx_config_aneg, .aneg_done = &genphy_aneg_done, .read_status = &vsc85xx_read_status, - .handle_interrupt = vsc85xx_handle_interrupt, + .handle_interrupt = vsc8584_handle_interrupt, .config_intr = &vsc85xx_config_intr, .suspend = &genphy_suspend, .resume = &genphy_resume, .remove = &vsc85xx_remove, - .probe = &vsc8574_probe, + .probe = &vsc8584_probe, .set_wol = &vsc85xx_wol_set, .get_wol = &vsc85xx_wol_get, .get_tunable = &vsc85xx_get_tunable, From 05e090620bacf317020f9591cfff8926093380bd Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Fri, 24 Oct 2025 14:23:35 +0300 Subject: [PATCH 0267/1381] net: airoha: Fix a copy and paste bug in probe() This code has a copy and paste bug where it accidentally checks "if (err)" instead of checking if "xsi_rsts" is NULL. Also, as a free bonus, I changed the allocation from kzalloc() to kcalloc() which is a kernel hardening measure to protect against integer overflows. Fixes: 5863b4e065e2 ("net: airoha: Add airoha_eth_soc_data struct") Signed-off-by: Dan Carpenter Acked-by: Lorenzo Bianconi Link: https://patch.msgid.link/aPtht6y5DRokn9zv@stanley.mountain Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/airoha/airoha_eth.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/airoha/airoha_eth.c b/drivers/net/ethernet/airoha/airoha_eth.c index e17a285a9e8f..688faf999e4c 100644 --- a/drivers/net/ethernet/airoha/airoha_eth.c +++ b/drivers/net/ethernet/airoha/airoha_eth.c @@ -2985,11 +2985,11 @@ static int airoha_probe(struct platform_device *pdev) return err; } - xsi_rsts = devm_kzalloc(eth->dev, - eth->soc->num_xsi_rsts * sizeof(*xsi_rsts), + xsi_rsts = devm_kcalloc(eth->dev, + eth->soc->num_xsi_rsts, sizeof(*xsi_rsts), GFP_KERNEL); - if (err) - return err; + if (!xsi_rsts) + return -ENOMEM; eth->xsi_rsts = xsi_rsts; for (i = 0; i < eth->soc->num_xsi_rsts; i++) From a5c12b060efe7a7830effcd233af6b2973176626 Mon Sep 17 00:00:00 2001 From: Vadim Fedorenko Date: Thu, 23 Oct 2025 22:04:52 +0000 Subject: [PATCH 0268/1381] octeontx2: convert to ndo_hwtstamp API Convert driver to use .ndo_hwtstamp_get()/.ndo_hwtstamp_set() callbacks. otx2_ioctl() becomes empty, remove it. Reviewed-by: Jacob Keller Signed-off-by: Vadim Fedorenko Link: https://patch.msgid.link/20251023220457.3201122-2-vadim.fedorenko@linux.dev Signed-off-by: Jakub Kicinski --- .../marvell/octeontx2/nic/otx2_common.h | 9 ++- .../ethernet/marvell/octeontx2/nic/otx2_pf.c | 56 +++++++++---------- .../ethernet/marvell/octeontx2/nic/otx2_vf.c | 3 +- 3 files changed, 33 insertions(+), 35 deletions(-) diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h index 1c8a3c078a64..ec26d1b6c789 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h @@ -527,7 +527,7 @@ struct otx2_nic { u32 nix_lmt_size; struct otx2_ptp *ptp; - struct hwtstamp_config tstamp; + struct kernel_hwtstamp_config tstamp; unsigned long rq_bmap; @@ -1098,8 +1098,11 @@ int otx2_open(struct net_device *netdev); int otx2_stop(struct net_device *netdev); int otx2_set_real_num_queues(struct net_device *netdev, int tx_queues, int rx_queues); -int otx2_ioctl(struct net_device *netdev, struct ifreq *req, int cmd); -int otx2_config_hwtstamp(struct net_device *netdev, struct ifreq *ifr); +int otx2_config_hwtstamp_get(struct net_device *netdev, + struct kernel_hwtstamp_config *config); +int otx2_config_hwtstamp_set(struct net_device *netdev, + struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack); /* MCAM filter related APIs */ int otx2_mcam_flow_init(struct otx2_nic *pf); diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c index e808995703cf..a7feb4c392b3 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c @@ -2445,18 +2445,26 @@ static int otx2_config_hw_tx_tstamp(struct otx2_nic *pfvf, bool enable) return 0; } -int otx2_config_hwtstamp(struct net_device *netdev, struct ifreq *ifr) +int otx2_config_hwtstamp_get(struct net_device *netdev, + struct kernel_hwtstamp_config *config) +{ + struct otx2_nic *pfvf = netdev_priv(netdev); + + *config = pfvf->tstamp; + return 0; +} +EXPORT_SYMBOL(otx2_config_hwtstamp_get); + +int otx2_config_hwtstamp_set(struct net_device *netdev, + struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack) { struct otx2_nic *pfvf = netdev_priv(netdev); - struct hwtstamp_config config; if (!pfvf->ptp) return -ENODEV; - if (copy_from_user(&config, ifr->ifr_data, sizeof(config))) - return -EFAULT; - - switch (config.tx_type) { + switch (config->tx_type) { case HWTSTAMP_TX_OFF: if (pfvf->flags & OTX2_FLAG_PTP_ONESTEP_SYNC) pfvf->flags &= ~OTX2_FLAG_PTP_ONESTEP_SYNC; @@ -2465,8 +2473,11 @@ int otx2_config_hwtstamp(struct net_device *netdev, struct ifreq *ifr) otx2_config_hw_tx_tstamp(pfvf, false); break; case HWTSTAMP_TX_ONESTEP_SYNC: - if (!test_bit(CN10K_PTP_ONESTEP, &pfvf->hw.cap_flag)) + if (!test_bit(CN10K_PTP_ONESTEP, &pfvf->hw.cap_flag)) { + NL_SET_ERR_MSG_MOD(extack, + "One-step time stamping is not supported"); return -ERANGE; + } pfvf->flags |= OTX2_FLAG_PTP_ONESTEP_SYNC; schedule_delayed_work(&pfvf->ptp->synctstamp_work, msecs_to_jiffies(500)); @@ -2478,7 +2489,7 @@ int otx2_config_hwtstamp(struct net_device *netdev, struct ifreq *ifr) return -ERANGE; } - switch (config.rx_filter) { + switch (config->rx_filter) { case HWTSTAMP_FILTER_NONE: otx2_config_hw_rx_tstamp(pfvf, false); break; @@ -2497,35 +2508,17 @@ int otx2_config_hwtstamp(struct net_device *netdev, struct ifreq *ifr) case HWTSTAMP_FILTER_PTP_V2_SYNC: case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ: otx2_config_hw_rx_tstamp(pfvf, true); - config.rx_filter = HWTSTAMP_FILTER_ALL; + config->rx_filter = HWTSTAMP_FILTER_ALL; break; default: return -ERANGE; } - memcpy(&pfvf->tstamp, &config, sizeof(config)); + pfvf->tstamp = *config; - return copy_to_user(ifr->ifr_data, &config, - sizeof(config)) ? -EFAULT : 0; + return 0; } -EXPORT_SYMBOL(otx2_config_hwtstamp); - -int otx2_ioctl(struct net_device *netdev, struct ifreq *req, int cmd) -{ - struct otx2_nic *pfvf = netdev_priv(netdev); - struct hwtstamp_config *cfg = &pfvf->tstamp; - - switch (cmd) { - case SIOCSHWTSTAMP: - return otx2_config_hwtstamp(netdev, req); - case SIOCGHWTSTAMP: - return copy_to_user(req->ifr_data, cfg, - sizeof(*cfg)) ? -EFAULT : 0; - default: - return -EOPNOTSUPP; - } -} -EXPORT_SYMBOL(otx2_ioctl); +EXPORT_SYMBOL(otx2_config_hwtstamp_set); static int otx2_do_set_vf_mac(struct otx2_nic *pf, int vf, const u8 *mac) { @@ -2942,7 +2935,6 @@ static const struct net_device_ops otx2_netdev_ops = { .ndo_set_features = otx2_set_features, .ndo_tx_timeout = otx2_tx_timeout, .ndo_get_stats64 = otx2_get_stats64, - .ndo_eth_ioctl = otx2_ioctl, .ndo_set_vf_mac = otx2_set_vf_mac, .ndo_set_vf_vlan = otx2_set_vf_vlan, .ndo_get_vf_config = otx2_get_vf_config, @@ -2951,6 +2943,8 @@ static const struct net_device_ops otx2_netdev_ops = { .ndo_xdp_xmit = otx2_xdp_xmit, .ndo_setup_tc = otx2_setup_tc, .ndo_set_vf_trust = otx2_ndo_set_vf_trust, + .ndo_hwtstamp_get = otx2_config_hwtstamp_get, + .ndo_hwtstamp_set = otx2_config_hwtstamp_set, }; int otx2_wq_init(struct otx2_nic *pf) diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_vf.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_vf.c index 25381f079b97..f4fdbfba8667 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_vf.c +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_vf.c @@ -534,8 +534,9 @@ static const struct net_device_ops otx2vf_netdev_ops = { .ndo_set_features = otx2vf_set_features, .ndo_get_stats64 = otx2_get_stats64, .ndo_tx_timeout = otx2_tx_timeout, - .ndo_eth_ioctl = otx2_ioctl, .ndo_setup_tc = otx2_setup_tc, + .ndo_hwtstamp_get = otx2_config_hwtstamp_get, + .ndo_hwtstamp_set = otx2_config_hwtstamp_set, }; static int otx2_vf_wq_init(struct otx2_nic *vf) From 7a07dc723fadadb1680f6353954a965ae6e75862 Mon Sep 17 00:00:00 2001 From: Vadim Fedorenko Date: Thu, 23 Oct 2025 22:04:53 +0000 Subject: [PATCH 0269/1381] mlx4: convert to ndo_hwtstamp API Convert driver to use .ndo_hwtstamp_get()/.ndo_hwtstamp_set() callbacks. mlx4_en_ioctl() becomes empty, remove it. Reviewed-by: Jacob Keller Reviewed-by: Tariq Toukan Signed-off-by: Vadim Fedorenko Link: https://patch.msgid.link/20251023220457.3201122-3-vadim.fedorenko@linux.dev Signed-off-by: Jakub Kicinski --- .../net/ethernet/mellanox/mlx4/en_netdev.c | 62 ++++++++----------- drivers/net/ethernet/mellanox/mlx4/mlx4_en.h | 6 +- 2 files changed, 29 insertions(+), 39 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c index 308b4458e0d4..81bf8908b897 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c @@ -2420,21 +2420,22 @@ static int mlx4_en_change_mtu(struct net_device *dev, int new_mtu) return 0; } -static int mlx4_en_hwtstamp_set(struct net_device *dev, struct ifreq *ifr) +static int mlx4_en_hwtstamp_set(struct net_device *dev, + struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack) { struct mlx4_en_priv *priv = netdev_priv(dev); struct mlx4_en_dev *mdev = priv->mdev; - struct hwtstamp_config config; - - if (copy_from_user(&config, ifr->ifr_data, sizeof(config))) - return -EFAULT; /* device doesn't support time stamping */ - if (!(mdev->dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_TS)) + if (!(mdev->dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_TS)) { + NL_SET_ERR_MSG_MOD(extack, + "device doesn't support time stamping"); return -EINVAL; + } /* TX HW timestamp */ - switch (config.tx_type) { + switch (config->tx_type) { case HWTSTAMP_TX_OFF: case HWTSTAMP_TX_ON: break; @@ -2443,7 +2444,7 @@ static int mlx4_en_hwtstamp_set(struct net_device *dev, struct ifreq *ifr) } /* RX HW timestamp */ - switch (config.rx_filter) { + switch (config->rx_filter) { case HWTSTAMP_FILTER_NONE: break; case HWTSTAMP_FILTER_ALL: @@ -2461,39 +2462,27 @@ static int mlx4_en_hwtstamp_set(struct net_device *dev, struct ifreq *ifr) case HWTSTAMP_FILTER_PTP_V2_SYNC: case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ: case HWTSTAMP_FILTER_NTP_ALL: - config.rx_filter = HWTSTAMP_FILTER_ALL; + config->rx_filter = HWTSTAMP_FILTER_ALL; break; default: return -ERANGE; } if (mlx4_en_reset_config(dev, config, dev->features)) { - config.tx_type = HWTSTAMP_TX_OFF; - config.rx_filter = HWTSTAMP_FILTER_NONE; + config->tx_type = HWTSTAMP_TX_OFF; + config->rx_filter = HWTSTAMP_FILTER_NONE; } - return copy_to_user(ifr->ifr_data, &config, - sizeof(config)) ? -EFAULT : 0; + return 0; } -static int mlx4_en_hwtstamp_get(struct net_device *dev, struct ifreq *ifr) +static int mlx4_en_hwtstamp_get(struct net_device *dev, + struct kernel_hwtstamp_config *config) { struct mlx4_en_priv *priv = netdev_priv(dev); - return copy_to_user(ifr->ifr_data, &priv->hwtstamp_config, - sizeof(priv->hwtstamp_config)) ? -EFAULT : 0; -} - -static int mlx4_en_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) -{ - switch (cmd) { - case SIOCSHWTSTAMP: - return mlx4_en_hwtstamp_set(dev, ifr); - case SIOCGHWTSTAMP: - return mlx4_en_hwtstamp_get(dev, ifr); - default: - return -EOPNOTSUPP; - } + *config = priv->hwtstamp_config; + return 0; } static netdev_features_t mlx4_en_fix_features(struct net_device *netdev, @@ -2560,7 +2549,7 @@ static int mlx4_en_set_features(struct net_device *netdev, } if (reset) { - ret = mlx4_en_reset_config(netdev, priv->hwtstamp_config, + ret = mlx4_en_reset_config(netdev, &priv->hwtstamp_config, features); if (ret) return ret; @@ -2844,7 +2833,6 @@ static const struct net_device_ops mlx4_netdev_ops = { .ndo_set_mac_address = mlx4_en_set_mac, .ndo_validate_addr = eth_validate_addr, .ndo_change_mtu = mlx4_en_change_mtu, - .ndo_eth_ioctl = mlx4_en_ioctl, .ndo_tx_timeout = mlx4_en_tx_timeout, .ndo_vlan_rx_add_vid = mlx4_en_vlan_rx_add_vid, .ndo_vlan_rx_kill_vid = mlx4_en_vlan_rx_kill_vid, @@ -2858,6 +2846,8 @@ static const struct net_device_ops mlx4_netdev_ops = { .ndo_features_check = mlx4_en_features_check, .ndo_set_tx_maxrate = mlx4_en_set_tx_maxrate, .ndo_bpf = mlx4_xdp, + .ndo_hwtstamp_get = mlx4_en_hwtstamp_get, + .ndo_hwtstamp_set = mlx4_en_hwtstamp_set, }; static const struct net_device_ops mlx4_netdev_ops_master = { @@ -3512,7 +3502,7 @@ int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port, } int mlx4_en_reset_config(struct net_device *dev, - struct hwtstamp_config ts_config, + struct kernel_hwtstamp_config *ts_config, netdev_features_t features) { struct mlx4_en_priv *priv = netdev_priv(dev); @@ -3522,8 +3512,8 @@ int mlx4_en_reset_config(struct net_device *dev, int port_up = 0; int err = 0; - if (priv->hwtstamp_config.tx_type == ts_config.tx_type && - priv->hwtstamp_config.rx_filter == ts_config.rx_filter && + if (priv->hwtstamp_config.tx_type == ts_config->tx_type && + priv->hwtstamp_config.rx_filter == ts_config->rx_filter && !DEV_FEATURE_CHANGED(dev, features, NETIF_F_HW_VLAN_CTAG_RX) && !DEV_FEATURE_CHANGED(dev, features, NETIF_F_RXFCS)) return 0; /* Nothing to change */ @@ -3542,7 +3532,7 @@ int mlx4_en_reset_config(struct net_device *dev, mutex_lock(&mdev->state_lock); memcpy(&new_prof, priv->prof, sizeof(struct mlx4_en_port_profile)); - memcpy(&new_prof.hwtstamp_config, &ts_config, sizeof(ts_config)); + memcpy(&new_prof.hwtstamp_config, ts_config, sizeof(*ts_config)); err = mlx4_en_try_alloc_resources(priv, tmp, &new_prof, true); if (err) @@ -3560,7 +3550,7 @@ int mlx4_en_reset_config(struct net_device *dev, dev->features |= NETIF_F_HW_VLAN_CTAG_RX; else dev->features &= ~NETIF_F_HW_VLAN_CTAG_RX; - } else if (ts_config.rx_filter == HWTSTAMP_FILTER_NONE) { + } else if (ts_config->rx_filter == HWTSTAMP_FILTER_NONE) { /* RX time-stamping is OFF, update the RX vlan offload * to the latest wanted state */ @@ -3581,7 +3571,7 @@ int mlx4_en_reset_config(struct net_device *dev, * Regardless of the caller's choice, * Turn Off RX vlan offload in case of time-stamping is ON */ - if (ts_config.rx_filter != HWTSTAMP_FILTER_NONE) { + if (ts_config->rx_filter != HWTSTAMP_FILTER_NONE) { if (dev->features & NETIF_F_HW_VLAN_CTAG_RX) en_warn(priv, "Turning off RX vlan offload since RX time-stamping is ON\n"); dev->features &= ~NETIF_F_HW_VLAN_CTAG_RX; diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h index ad0d91a75184..aab97694f86b 100644 --- a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h +++ b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h @@ -388,7 +388,7 @@ struct mlx4_en_port_profile { u8 num_up; int rss_rings; int inline_thold; - struct hwtstamp_config hwtstamp_config; + struct kernel_hwtstamp_config hwtstamp_config; }; struct mlx4_en_profile { @@ -612,7 +612,7 @@ struct mlx4_en_priv { bool wol; struct device *ddev; struct hlist_head mac_hash[MLX4_EN_MAC_HASH_SIZE]; - struct hwtstamp_config hwtstamp_config; + struct kernel_hwtstamp_config hwtstamp_config; u32 counter_index; #ifdef CONFIG_MLX4_EN_DCB @@ -780,7 +780,7 @@ void mlx4_en_ptp_overflow_check(struct mlx4_en_dev *mdev); int mlx4_en_moderation_update(struct mlx4_en_priv *priv); int mlx4_en_reset_config(struct net_device *dev, - struct hwtstamp_config ts_config, + struct kernel_hwtstamp_config *ts_config, netdev_features_t new_features); void mlx4_en_update_pfc_stats_bitmap(struct mlx4_dev *dev, struct mlx4_en_stats_bitmap *stats_bitmap, From 38efb0ba3cd07c3f144ef80895966debeee3f60e Mon Sep 17 00:00:00 2001 From: Vadim Fedorenko Date: Thu, 23 Oct 2025 22:04:54 +0000 Subject: [PATCH 0270/1381] ionic: convert to ndo_hwtstamp API Convert driver to use .ndo_hwtstamp_get()/.ndo_hwtstamp_set() callbacks. ionic_eth_ioctl() becomes empty, remove it. Reviewed-by: Jacob Keller Reviewed-by: Brett Creeley Signed-off-by: Vadim Fedorenko Link: https://patch.msgid.link/20251023220457.3201122-4-vadim.fedorenko@linux.dev Signed-off-by: Jakub Kicinski --- .../net/ethernet/pensando/ionic/ionic_lif.c | 17 +----- .../net/ethernet/pensando/ionic/ionic_lif.h | 18 ++++-- .../net/ethernet/pensando/ionic/ionic_phc.c | 61 ++++++++++++------- 3 files changed, 52 insertions(+), 44 deletions(-) diff --git a/drivers/net/ethernet/pensando/ionic/ionic_lif.c b/drivers/net/ethernet/pensando/ionic/ionic_lif.c index b28966ae50c2..058eea86e141 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_lif.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_lif.c @@ -2335,20 +2335,6 @@ static int ionic_stop(struct net_device *netdev) return 0; } -static int ionic_eth_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) -{ - struct ionic_lif *lif = netdev_priv(netdev); - - switch (cmd) { - case SIOCSHWTSTAMP: - return ionic_lif_hwstamp_set(lif, ifr); - case SIOCGHWTSTAMP: - return ionic_lif_hwstamp_get(lif, ifr); - default: - return -EOPNOTSUPP; - } -} - static int ionic_get_vf_config(struct net_device *netdev, int vf, struct ifla_vf_info *ivf) { @@ -2812,7 +2798,6 @@ static int ionic_xdp(struct net_device *netdev, struct netdev_bpf *bpf) static const struct net_device_ops ionic_netdev_ops = { .ndo_open = ionic_open, .ndo_stop = ionic_stop, - .ndo_eth_ioctl = ionic_eth_ioctl, .ndo_start_xmit = ionic_start_xmit, .ndo_bpf = ionic_xdp, .ndo_xdp_xmit = ionic_xdp_xmit, @@ -2833,6 +2818,8 @@ static const struct net_device_ops ionic_netdev_ops = { .ndo_get_vf_config = ionic_get_vf_config, .ndo_set_vf_link_state = ionic_set_vf_link_state, .ndo_get_vf_stats = ionic_get_vf_stats, + .ndo_hwtstamp_get = ionic_hwstamp_get, + .ndo_hwtstamp_set = ionic_hwstamp_set, }; static int ionic_cmb_reconfig(struct ionic_lif *lif, diff --git a/drivers/net/ethernet/pensando/ionic/ionic_lif.h b/drivers/net/ethernet/pensando/ionic/ionic_lif.h index 43bdd0fb8733..8e10f66dc50e 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_lif.h +++ b/drivers/net/ethernet/pensando/ionic/ionic_lif.h @@ -6,7 +6,7 @@ #include #include -#include +#include #include #include #include "ionic_rx_filter.h" @@ -254,7 +254,7 @@ struct ionic_phc { struct timecounter tc; struct mutex config_lock; /* lock for ts_config */ - struct hwtstamp_config ts_config; + struct kernel_hwtstamp_config ts_config; u64 ts_config_rx_filt; u32 ts_config_tx_mode; @@ -362,8 +362,11 @@ int ionic_lif_size(struct ionic *ionic); #if IS_ENABLED(CONFIG_PTP_1588_CLOCK) void ionic_lif_hwstamp_replay(struct ionic_lif *lif); void ionic_lif_hwstamp_recreate_queues(struct ionic_lif *lif); -int ionic_lif_hwstamp_set(struct ionic_lif *lif, struct ifreq *ifr); -int ionic_lif_hwstamp_get(struct ionic_lif *lif, struct ifreq *ifr); +int ionic_hwstamp_set(struct net_device *netdev, + struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack); +int ionic_hwstamp_get(struct net_device *netdev, + struct kernel_hwtstamp_config *config); ktime_t ionic_lif_phc_ktime(struct ionic_lif *lif, u64 counter); void ionic_lif_register_phc(struct ionic_lif *lif); void ionic_lif_unregister_phc(struct ionic_lif *lif); @@ -373,12 +376,15 @@ void ionic_lif_free_phc(struct ionic_lif *lif); static inline void ionic_lif_hwstamp_replay(struct ionic_lif *lif) {} static inline void ionic_lif_hwstamp_recreate_queues(struct ionic_lif *lif) {} -static inline int ionic_lif_hwstamp_set(struct ionic_lif *lif, struct ifreq *ifr) +static inline int ionic_hwstamp_set(struct net_device *netdev, + struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack) { return -EOPNOTSUPP; } -static inline int ionic_lif_hwstamp_get(struct ionic_lif *lif, struct ifreq *ifr) +static inline int ionic_hwstamp_get(struct net_device *netdev, + struct kernel_hwtstamp_config *config) { return -EOPNOTSUPP; } diff --git a/drivers/net/ethernet/pensando/ionic/ionic_phc.c b/drivers/net/ethernet/pensando/ionic/ionic_phc.c index 9f5c81d44f99..05b44fc482f8 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_phc.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_phc.c @@ -65,11 +65,12 @@ static u64 ionic_hwstamp_rx_filt(int config_rx_filter) } static int ionic_lif_hwstamp_set_ts_config(struct ionic_lif *lif, - struct hwtstamp_config *new_ts) + struct kernel_hwtstamp_config *new_ts, + struct netlink_ext_ack *extack) { + struct kernel_hwtstamp_config *config; + struct kernel_hwtstamp_config ts = {}; struct ionic *ionic = lif->ionic; - struct hwtstamp_config *config; - struct hwtstamp_config ts; int tx_mode = 0; u64 rx_filt = 0; int err, err2; @@ -99,12 +100,16 @@ static int ionic_lif_hwstamp_set_ts_config(struct ionic_lif *lif, tx_mode = ionic_hwstamp_tx_mode(config->tx_type); if (tx_mode < 0) { + NL_SET_ERR_MSG_MOD(extack, + "TX time stamping mode isn't supported"); err = tx_mode; goto err_queues; } mask = cpu_to_le64(BIT_ULL(tx_mode)); if ((ionic->ident.lif.eth.hwstamp_tx_modes & mask) != mask) { + NL_SET_ERR_MSG_MOD(extack, + "TX time stamping mode isn't supported"); err = -ERANGE; goto err_queues; } @@ -124,32 +129,47 @@ static int ionic_lif_hwstamp_set_ts_config(struct ionic_lif *lif, if (tx_mode) { err = ionic_lif_create_hwstamp_txq(lif); - if (err) + if (err) { + NL_SET_ERR_MSG_MOD(extack, + "Error creating TX timestamp queue"); goto err_queues; + } } if (rx_filt) { err = ionic_lif_create_hwstamp_rxq(lif); - if (err) + if (err) { + NL_SET_ERR_MSG_MOD(extack, + "Error creating RX timestamp queue"); goto err_queues; + } } if (tx_mode != lif->phc->ts_config_tx_mode) { err = ionic_lif_set_hwstamp_txmode(lif, tx_mode); - if (err) + if (err) { + NL_SET_ERR_MSG_MOD(extack, + "Error enabling TX timestamp mode"); goto err_txmode; + } } if (rx_filt != lif->phc->ts_config_rx_filt) { err = ionic_lif_set_hwstamp_rxfilt(lif, rx_filt); - if (err) + if (err) { + NL_SET_ERR_MSG_MOD(extack, + "Error enabling RX timestamp mode"); goto err_rxfilt; + } } if (rx_all != (lif->phc->ts_config.rx_filter == HWTSTAMP_FILTER_ALL)) { err = ionic_lif_config_hwstamp_rxq_all(lif, rx_all); - if (err) + if (err) { + NL_SET_ERR_MSG_MOD(extack, + "Error enabling RX timestamp mode"); goto err_rxall; + } } memcpy(&lif->phc->ts_config, config, sizeof(*config)); @@ -183,28 +203,24 @@ static int ionic_lif_hwstamp_set_ts_config(struct ionic_lif *lif, return err; } -int ionic_lif_hwstamp_set(struct ionic_lif *lif, struct ifreq *ifr) +int ionic_hwstamp_set(struct net_device *netdev, + struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack) { - struct hwtstamp_config config; + struct ionic_lif *lif = netdev_priv(netdev); int err; if (!lif->phc || !lif->phc->ptp) return -EOPNOTSUPP; - if (copy_from_user(&config, ifr->ifr_data, sizeof(config))) - return -EFAULT; - mutex_lock(&lif->queue_lock); - err = ionic_lif_hwstamp_set_ts_config(lif, &config); + err = ionic_lif_hwstamp_set_ts_config(lif, config, extack); mutex_unlock(&lif->queue_lock); if (err) { netdev_info(lif->netdev, "hwstamp set failed: %d\n", err); return err; } - if (copy_to_user(ifr->ifr_data, &config, sizeof(config))) - return -EFAULT; - return 0; } @@ -216,7 +232,7 @@ void ionic_lif_hwstamp_replay(struct ionic_lif *lif) return; mutex_lock(&lif->queue_lock); - err = ionic_lif_hwstamp_set_ts_config(lif, NULL); + err = ionic_lif_hwstamp_set_ts_config(lif, NULL, NULL); mutex_unlock(&lif->queue_lock); if (err) netdev_info(lif->netdev, "hwstamp replay failed: %d\n", err); @@ -246,19 +262,18 @@ void ionic_lif_hwstamp_recreate_queues(struct ionic_lif *lif) mutex_unlock(&lif->phc->config_lock); } -int ionic_lif_hwstamp_get(struct ionic_lif *lif, struct ifreq *ifr) +int ionic_hwstamp_get(struct net_device *netdev, + struct kernel_hwtstamp_config *config) { - struct hwtstamp_config config; + struct ionic_lif *lif = netdev_priv(netdev); if (!lif->phc || !lif->phc->ptp) return -EOPNOTSUPP; mutex_lock(&lif->phc->config_lock); - memcpy(&config, &lif->phc->ts_config, sizeof(config)); + memcpy(config, &lif->phc->ts_config, sizeof(*config)); mutex_unlock(&lif->phc->config_lock); - if (copy_to_user(ifr->ifr_data, &config, sizeof(config))) - return -EFAULT; return 0; } From faac57cddfc256f9cba35702b12bcc5cf793db99 Mon Sep 17 00:00:00 2001 From: Vadim Fedorenko Date: Thu, 23 Oct 2025 22:04:55 +0000 Subject: [PATCH 0271/1381] net: ravb: convert to ndo_hwtstamp API MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Convert driver to use .ndo_hwtstamp_set()/.ndo_hwtstamp_get callbacks. ravb_do_ioctl() becomes pure phy_do_ioctl_running(), remove it and replace in callbacks. Reviewed-by: Niklas Söderlund Reviewed-by: Jacob Keller Signed-off-by: Vadim Fedorenko Link: https://patch.msgid.link/20251023220457.3201122-5-vadim.fedorenko@linux.dev Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/renesas/ravb_main.c | 61 ++++++++---------------- 1 file changed, 19 insertions(+), 42 deletions(-) diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c index c3fc15f9ec85..edfaa5850527 100644 --- a/drivers/net/ethernet/renesas/ravb_main.c +++ b/drivers/net/ethernet/renesas/ravb_main.c @@ -2410,41 +2410,38 @@ static int ravb_close(struct net_device *ndev) return 0; } -static int ravb_hwtstamp_get(struct net_device *ndev, struct ifreq *req) +static int ravb_hwtstamp_get(struct net_device *ndev, + struct kernel_hwtstamp_config *config) { struct ravb_private *priv = netdev_priv(ndev); - struct hwtstamp_config config; - config.flags = 0; - config.tx_type = priv->tstamp_tx_ctrl ? HWTSTAMP_TX_ON : - HWTSTAMP_TX_OFF; + config->flags = 0; + config->tx_type = priv->tstamp_tx_ctrl ? HWTSTAMP_TX_ON : + HWTSTAMP_TX_OFF; switch (priv->tstamp_rx_ctrl & RAVB_RXTSTAMP_TYPE) { case RAVB_RXTSTAMP_TYPE_V2_L2_EVENT: - config.rx_filter = HWTSTAMP_FILTER_PTP_V2_L2_EVENT; + config->rx_filter = HWTSTAMP_FILTER_PTP_V2_L2_EVENT; break; case RAVB_RXTSTAMP_TYPE_ALL: - config.rx_filter = HWTSTAMP_FILTER_ALL; + config->rx_filter = HWTSTAMP_FILTER_ALL; break; default: - config.rx_filter = HWTSTAMP_FILTER_NONE; + config->rx_filter = HWTSTAMP_FILTER_NONE; } - return copy_to_user(req->ifr_data, &config, sizeof(config)) ? - -EFAULT : 0; + return 0; } /* Control hardware time stamping */ -static int ravb_hwtstamp_set(struct net_device *ndev, struct ifreq *req) +static int ravb_hwtstamp_set(struct net_device *ndev, + struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack) { struct ravb_private *priv = netdev_priv(ndev); - struct hwtstamp_config config; u32 tstamp_rx_ctrl = RAVB_RXTSTAMP_ENABLED; u32 tstamp_tx_ctrl; - if (copy_from_user(&config, req->ifr_data, sizeof(config))) - return -EFAULT; - - switch (config.tx_type) { + switch (config->tx_type) { case HWTSTAMP_TX_OFF: tstamp_tx_ctrl = 0; break; @@ -2455,7 +2452,7 @@ static int ravb_hwtstamp_set(struct net_device *ndev, struct ifreq *req) return -ERANGE; } - switch (config.rx_filter) { + switch (config->rx_filter) { case HWTSTAMP_FILTER_NONE: tstamp_rx_ctrl = 0; break; @@ -2463,36 +2460,14 @@ static int ravb_hwtstamp_set(struct net_device *ndev, struct ifreq *req) tstamp_rx_ctrl |= RAVB_RXTSTAMP_TYPE_V2_L2_EVENT; break; default: - config.rx_filter = HWTSTAMP_FILTER_ALL; + config->rx_filter = HWTSTAMP_FILTER_ALL; tstamp_rx_ctrl |= RAVB_RXTSTAMP_TYPE_ALL; } priv->tstamp_tx_ctrl = tstamp_tx_ctrl; priv->tstamp_rx_ctrl = tstamp_rx_ctrl; - return copy_to_user(req->ifr_data, &config, sizeof(config)) ? - -EFAULT : 0; -} - -/* ioctl to device function */ -static int ravb_do_ioctl(struct net_device *ndev, struct ifreq *req, int cmd) -{ - struct phy_device *phydev = ndev->phydev; - - if (!netif_running(ndev)) - return -EINVAL; - - if (!phydev) - return -ENODEV; - - switch (cmd) { - case SIOCGHWTSTAMP: - return ravb_hwtstamp_get(ndev, req); - case SIOCSHWTSTAMP: - return ravb_hwtstamp_set(ndev, req); - } - - return phy_mii_ioctl(phydev, req, cmd); + return 0; } static int ravb_change_mtu(struct net_device *ndev, int new_mtu) @@ -2628,11 +2603,13 @@ static const struct net_device_ops ravb_netdev_ops = { .ndo_get_stats = ravb_get_stats, .ndo_set_rx_mode = ravb_set_rx_mode, .ndo_tx_timeout = ravb_tx_timeout, - .ndo_eth_ioctl = ravb_do_ioctl, + .ndo_eth_ioctl = phy_do_ioctl_running, .ndo_change_mtu = ravb_change_mtu, .ndo_validate_addr = eth_validate_addr, .ndo_set_mac_address = eth_mac_addr, .ndo_set_features = ravb_set_features, + .ndo_hwtstamp_get = ravb_hwtstamp_get, + .ndo_hwtstamp_set = ravb_hwtstamp_set, }; /* MDIO bus init function */ From 87e1b590f776ec9c9f66362a6f818088560c645f Mon Sep 17 00:00:00 2001 From: Vadim Fedorenko Date: Thu, 23 Oct 2025 22:04:56 +0000 Subject: [PATCH 0272/1381] net: renesas: rswitch: convert to ndo_hwtstamp API Convert driver to use .ndo_hwtstamp_set()/.ndo_hwtstamp_get() callbacks. rswitch_eth_ioctl() becomes phy_do_ioctl_running(), remove it and replace .ndo_eth_ioctl callback with phy_do_ioctl_running(). Reviewed-by: Jacob Keller Signed-off-by: Vadim Fedorenko Link: https://patch.msgid.link/20251023220457.3201122-6-vadim.fedorenko@linux.dev Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/renesas/rswitch_main.c | 53 ++++++++------------- 1 file changed, 19 insertions(+), 34 deletions(-) diff --git a/drivers/net/ethernet/renesas/rswitch_main.c b/drivers/net/ethernet/renesas/rswitch_main.c index 8d8acc2124b8..f21a814aa9d1 100644 --- a/drivers/net/ethernet/renesas/rswitch_main.c +++ b/drivers/net/ethernet/renesas/rswitch_main.c @@ -1793,46 +1793,44 @@ static struct net_device_stats *rswitch_get_stats(struct net_device *ndev) return &ndev->stats; } -static int rswitch_hwstamp_get(struct net_device *ndev, struct ifreq *req) +static int rswitch_hwstamp_get(struct net_device *ndev, + struct kernel_hwtstamp_config *config) { struct rswitch_device *rdev = netdev_priv(ndev); struct rcar_gen4_ptp_private *ptp_priv; - struct hwtstamp_config config; ptp_priv = rdev->priv->ptp_priv; - config.flags = 0; - config.tx_type = ptp_priv->tstamp_tx_ctrl ? HWTSTAMP_TX_ON : + config->flags = 0; + config->tx_type = ptp_priv->tstamp_tx_ctrl ? HWTSTAMP_TX_ON : HWTSTAMP_TX_OFF; switch (ptp_priv->tstamp_rx_ctrl & RCAR_GEN4_RXTSTAMP_TYPE) { case RCAR_GEN4_RXTSTAMP_TYPE_V2_L2_EVENT: - config.rx_filter = HWTSTAMP_FILTER_PTP_V2_L2_EVENT; + config->rx_filter = HWTSTAMP_FILTER_PTP_V2_L2_EVENT; break; case RCAR_GEN4_RXTSTAMP_TYPE_ALL: - config.rx_filter = HWTSTAMP_FILTER_ALL; + config->rx_filter = HWTSTAMP_FILTER_ALL; break; default: - config.rx_filter = HWTSTAMP_FILTER_NONE; + config->rx_filter = HWTSTAMP_FILTER_NONE; break; } - return copy_to_user(req->ifr_data, &config, sizeof(config)) ? -EFAULT : 0; + return 0; } -static int rswitch_hwstamp_set(struct net_device *ndev, struct ifreq *req) +static int rswitch_hwstamp_set(struct net_device *ndev, + struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack) { struct rswitch_device *rdev = netdev_priv(ndev); u32 tstamp_rx_ctrl = RCAR_GEN4_RXTSTAMP_ENABLED; - struct hwtstamp_config config; u32 tstamp_tx_ctrl; - if (copy_from_user(&config, req->ifr_data, sizeof(config))) - return -EFAULT; - - if (config.flags) + if (config->flags) return -EINVAL; - switch (config.tx_type) { + switch (config->tx_type) { case HWTSTAMP_TX_OFF: tstamp_tx_ctrl = 0; break; @@ -1843,7 +1841,7 @@ static int rswitch_hwstamp_set(struct net_device *ndev, struct ifreq *req) return -ERANGE; } - switch (config.rx_filter) { + switch (config->rx_filter) { case HWTSTAMP_FILTER_NONE: tstamp_rx_ctrl = 0; break; @@ -1851,7 +1849,7 @@ static int rswitch_hwstamp_set(struct net_device *ndev, struct ifreq *req) tstamp_rx_ctrl |= RCAR_GEN4_RXTSTAMP_TYPE_V2_L2_EVENT; break; default: - config.rx_filter = HWTSTAMP_FILTER_ALL; + config->rx_filter = HWTSTAMP_FILTER_ALL; tstamp_rx_ctrl |= RCAR_GEN4_RXTSTAMP_TYPE_ALL; break; } @@ -1859,22 +1857,7 @@ static int rswitch_hwstamp_set(struct net_device *ndev, struct ifreq *req) rdev->priv->ptp_priv->tstamp_tx_ctrl = tstamp_tx_ctrl; rdev->priv->ptp_priv->tstamp_rx_ctrl = tstamp_rx_ctrl; - return copy_to_user(req->ifr_data, &config, sizeof(config)) ? -EFAULT : 0; -} - -static int rswitch_eth_ioctl(struct net_device *ndev, struct ifreq *req, int cmd) -{ - if (!netif_running(ndev)) - return -EINVAL; - - switch (cmd) { - case SIOCGHWTSTAMP: - return rswitch_hwstamp_get(ndev, req); - case SIOCSHWTSTAMP: - return rswitch_hwstamp_set(ndev, req); - default: - return phy_mii_ioctl(ndev->phydev, req, cmd); - } + return 0; } static int rswitch_get_port_parent_id(struct net_device *ndev, @@ -1905,11 +1888,13 @@ static const struct net_device_ops rswitch_netdev_ops = { .ndo_stop = rswitch_stop, .ndo_start_xmit = rswitch_start_xmit, .ndo_get_stats = rswitch_get_stats, - .ndo_eth_ioctl = rswitch_eth_ioctl, + .ndo_eth_ioctl = phy_do_ioctl_running, .ndo_get_port_parent_id = rswitch_get_port_parent_id, .ndo_get_phys_port_name = rswitch_get_phys_port_name, .ndo_validate_addr = eth_validate_addr, .ndo_set_mac_address = eth_mac_addr, + .ndo_hwtstamp_get = rswitch_hwstamp_get, + .ndo_hwtstamp_set = rswitch_hwstamp_set, }; bool is_rdev(const struct net_device *ndev) From 329021eeae035f169822c7575da49561b0bd7138 Mon Sep 17 00:00:00 2001 From: Vadim Fedorenko Date: Thu, 23 Oct 2025 22:04:57 +0000 Subject: [PATCH 0273/1381] net: hns3: add hwtstamp_get/hwtstamp_set ops And .ndo_hwtstamp_get()/.ndo_hwtstamp_set() callbacks to HNS3 framework to support HW timestamp configuration via netlink and adopt hns3pf to use .ndo_hwtstamp_get()/.ndo_hwtstamp_set() callbacks. Reviewed-by: Jacob Keller Reviewed-by: Jijie Shao Signed-off-by: Vadim Fedorenko Link: https://patch.msgid.link/20251023220457.3201122-7-vadim.fedorenko@linux.dev Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/hisilicon/hns3/hnae3.h | 5 +++ .../net/ethernet/hisilicon/hns3/hns3_enet.c | 31 ++++++++++++++++++ .../hisilicon/hns3/hns3pf/hclge_main.c | 13 +++----- .../hisilicon/hns3/hns3pf/hclge_ptp.c | 32 +++++++++++-------- .../hisilicon/hns3/hns3pf/hclge_ptp.h | 9 ++++-- 5 files changed, 64 insertions(+), 26 deletions(-) diff --git a/drivers/net/ethernet/hisilicon/hns3/hnae3.h b/drivers/net/ethernet/hisilicon/hns3/hnae3.h index 3b548f71fa8a..d7c3df1958f3 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hnae3.h +++ b/drivers/net/ethernet/hisilicon/hns3/hnae3.h @@ -804,6 +804,11 @@ struct hnae3_ae_ops { int (*dbg_get_read_func)(struct hnae3_handle *handle, enum hnae3_dbg_cmd cmd, read_func *func); + int (*hwtstamp_get)(struct hnae3_handle *handle, + struct kernel_hwtstamp_config *config); + int (*hwtstamp_set)(struct hnae3_handle *handle, + struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack); }; struct hnae3_dcb_ops { diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c index bfa5568baa92..7a0654e2d3dd 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c @@ -2419,6 +2419,35 @@ static int hns3_nic_do_ioctl(struct net_device *netdev, return h->ae_algo->ops->do_ioctl(h, ifr, cmd); } +static int hns3_nic_hwtstamp_get(struct net_device *netdev, + struct kernel_hwtstamp_config *config) +{ + struct hnae3_handle *h = hns3_get_handle(netdev); + + if (!netif_running(netdev)) + return -EINVAL; + + if (!h->ae_algo->ops->hwtstamp_get) + return -EOPNOTSUPP; + + return h->ae_algo->ops->hwtstamp_get(h, config); +} + +static int hns3_nic_hwtstamp_set(struct net_device *netdev, + struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack) +{ + struct hnae3_handle *h = hns3_get_handle(netdev); + + if (!netif_running(netdev)) + return -EINVAL; + + if (!h->ae_algo->ops->hwtstamp_set) + return -EOPNOTSUPP; + + return h->ae_algo->ops->hwtstamp_set(h, config, extack); +} + static int hns3_nic_set_features(struct net_device *netdev, netdev_features_t features) { @@ -3048,6 +3077,8 @@ static const struct net_device_ops hns3_nic_netdev_ops = { .ndo_set_vf_rate = hns3_nic_set_vf_rate, .ndo_set_vf_mac = hns3_nic_set_vf_mac, .ndo_select_queue = hns3_nic_select_queue, + .ndo_hwtstamp_get = hns3_nic_hwtstamp_get, + .ndo_hwtstamp_set = hns3_nic_hwtstamp_set, }; bool hns3_is_phys_func(struct pci_dev *pdev) diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c index 9d34d28ff168..81d3bdc098e6 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c @@ -9445,15 +9445,8 @@ static int hclge_do_ioctl(struct hnae3_handle *handle, struct ifreq *ifr, struct hclge_vport *vport = hclge_get_vport(handle); struct hclge_dev *hdev = vport->back; - switch (cmd) { - case SIOCGHWTSTAMP: - return hclge_ptp_get_cfg(hdev, ifr); - case SIOCSHWTSTAMP: - return hclge_ptp_set_cfg(hdev, ifr); - default: - if (!hdev->hw.mac.phydev) - return hclge_mii_ioctl(hdev, ifr, cmd); - } + if (!hdev->hw.mac.phydev) + return hclge_mii_ioctl(hdev, ifr, cmd); return phy_mii_ioctl(hdev->hw.mac.phydev, ifr, cmd); } @@ -12901,6 +12894,8 @@ static const struct hnae3_ae_ops hclge_ops = { .get_dscp_prio = hclge_get_dscp_prio, .get_wol = hclge_get_wol, .set_wol = hclge_set_wol, + .hwtstamp_get = hclge_ptp_get_cfg, + .hwtstamp_set = hclge_ptp_set_cfg, }; static struct hnae3_ae_algo ae_algo = { diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.c index 4bd52eab3914..0081c5281455 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.c @@ -204,13 +204,17 @@ static int hclge_ptp_adjtime(struct ptp_clock_info *ptp, s64 delta) return 0; } -int hclge_ptp_get_cfg(struct hclge_dev *hdev, struct ifreq *ifr) +int hclge_ptp_get_cfg(struct hnae3_handle *handle, + struct kernel_hwtstamp_config *config) { + struct hclge_vport *vport = hclge_get_vport(handle); + struct hclge_dev *hdev = vport->back; + if (!test_bit(HCLGE_STATE_PTP_EN, &hdev->state)) return -EOPNOTSUPP; - return copy_to_user(ifr->ifr_data, &hdev->ptp->ts_cfg, - sizeof(struct hwtstamp_config)) ? -EFAULT : 0; + *config = hdev->ptp->ts_cfg; + return 0; } static int hclge_ptp_int_en(struct hclge_dev *hdev, bool en) @@ -269,7 +273,7 @@ static int hclge_ptp_cfg(struct hclge_dev *hdev, u32 cfg) return ret; } -static int hclge_ptp_set_tx_mode(struct hwtstamp_config *cfg, +static int hclge_ptp_set_tx_mode(struct kernel_hwtstamp_config *cfg, unsigned long *flags, u32 *ptp_cfg) { switch (cfg->tx_type) { @@ -287,7 +291,7 @@ static int hclge_ptp_set_tx_mode(struct hwtstamp_config *cfg, return 0; } -static int hclge_ptp_set_rx_mode(struct hwtstamp_config *cfg, +static int hclge_ptp_set_rx_mode(struct kernel_hwtstamp_config *cfg, unsigned long *flags, u32 *ptp_cfg) { int rx_filter = cfg->rx_filter; @@ -332,7 +336,7 @@ static int hclge_ptp_set_rx_mode(struct hwtstamp_config *cfg, } static int hclge_ptp_set_ts_mode(struct hclge_dev *hdev, - struct hwtstamp_config *cfg) + struct kernel_hwtstamp_config *cfg) { unsigned long flags = hdev->ptp->flags; u32 ptp_cfg = 0; @@ -359,9 +363,12 @@ static int hclge_ptp_set_ts_mode(struct hclge_dev *hdev, return 0; } -int hclge_ptp_set_cfg(struct hclge_dev *hdev, struct ifreq *ifr) +int hclge_ptp_set_cfg(struct hnae3_handle *handle, + struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack) { - struct hwtstamp_config cfg; + struct hclge_vport *vport = hclge_get_vport(handle); + struct hclge_dev *hdev = vport->back; int ret; if (!test_bit(HCLGE_STATE_PTP_EN, &hdev->state)) { @@ -369,16 +376,13 @@ int hclge_ptp_set_cfg(struct hclge_dev *hdev, struct ifreq *ifr) return -EOPNOTSUPP; } - if (copy_from_user(&cfg, ifr->ifr_data, sizeof(cfg))) - return -EFAULT; - - ret = hclge_ptp_set_ts_mode(hdev, &cfg); + ret = hclge_ptp_set_ts_mode(hdev, config); if (ret) return ret; - hdev->ptp->ts_cfg = cfg; + hdev->ptp->ts_cfg = *config; - return copy_to_user(ifr->ifr_data, &cfg, sizeof(cfg)) ? -EFAULT : 0; + return 0; } int hclge_ptp_get_ts_info(struct hnae3_handle *handle, diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.h index 61faddcc3dd0..0162fa5ac146 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.h +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.h @@ -62,7 +62,7 @@ struct hclge_ptp { unsigned long flags; void __iomem *io_base; struct ptp_clock_info info; - struct hwtstamp_config ts_cfg; + struct kernel_hwtstamp_config ts_cfg; spinlock_t lock; /* protects ptp registers */ u32 ptp_cfg; u32 last_tx_seqid; @@ -133,8 +133,11 @@ bool hclge_ptp_set_tx_info(struct hnae3_handle *handle, struct sk_buff *skb); void hclge_ptp_clean_tx_hwts(struct hclge_dev *hdev); void hclge_ptp_get_rx_hwts(struct hnae3_handle *handle, struct sk_buff *skb, u32 nsec, u32 sec); -int hclge_ptp_get_cfg(struct hclge_dev *hdev, struct ifreq *ifr); -int hclge_ptp_set_cfg(struct hclge_dev *hdev, struct ifreq *ifr); +int hclge_ptp_get_cfg(struct hnae3_handle *handle, + struct kernel_hwtstamp_config *config); +int hclge_ptp_set_cfg(struct hnae3_handle *handle, + struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack); int hclge_ptp_init(struct hclge_dev *hdev); void hclge_ptp_uninit(struct hclge_dev *hdev); int hclge_ptp_get_ts_info(struct hnae3_handle *handle, From 622e8838a29845316668ec2e7648428878df7f9a Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Thu, 23 Oct 2025 23:16:50 +0000 Subject: [PATCH 0274/1381] sctp: Defer SCTP_DBG_OBJCNT_DEC() to sctp_destroy_sock(). SCTP_DBG_OBJCNT_INC() is called only when sctp_init_sock() returns 0 after successfully allocating sctp_sk(sk)->ep. OTOH, SCTP_DBG_OBJCNT_DEC() is called in sctp_close(). The code seems to expect that the socket is always exposed to userspace once SCTP_DBG_OBJCNT_INC() is incremented, but there is a path where the assumption is not true. In sctp_accept(), sctp_sock_migrate() could fail after sctp_init_sock(). Then, sk_common_release() does not call inet_release() nor sctp_close(). Instead, it calls sk->sk_prot->destroy(). Let's move SCTP_DBG_OBJCNT_DEC() from sctp_close() to sctp_destroy_sock(). Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Kuniyuki Iwashima Acked-by: Xin Long Link: https://patch.msgid.link/20251023231751.4168390-2-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/sctp/socket.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/net/sctp/socket.c b/net/sctp/socket.c index ed8293a34240..d190e75e4645 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -1553,8 +1553,6 @@ static void sctp_close(struct sock *sk, long timeout) spin_unlock_bh(&net->sctp.addr_wq_lock); sock_put(sk); - - SCTP_DBG_OBJCNT_DEC(sock); } /* Handle EPIPE error. */ @@ -5109,9 +5107,12 @@ static void sctp_destroy_sock(struct sock *sk) sp->do_auto_asconf = 0; list_del(&sp->auto_asconf_list); } + sctp_endpoint_free(sp->ep); + sk_sockets_allocated_dec(sk); sock_prot_inuse_add(sock_net(sk), sk->sk_prot, -1); + SCTP_DBG_OBJCNT_DEC(sock); } static void sctp_destruct_sock(struct sock *sk) From 2d4df59aae91340e777660cfe9862b7d8e15b077 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Thu, 23 Oct 2025 23:16:51 +0000 Subject: [PATCH 0275/1381] sctp: Don't copy sk_sndbuf and sk_rcvbuf in sctp_sock_migrate(). sctp_sock_migrate() is called from 2 places. 1) sctp_accept() calls sp->pf->create_accept_sk() before sctp_sock_migrate(), and sp->pf->create_accept_sk() calls sctp_copy_sock(). 2) sctp_do_peeloff() also calls sctp_copy_sock() before sctp_sock_migrate(). sctp_copy_sock() copies sk_sndbuf and sk_rcvbuf from the parent socket. Let's not copy the two fields in sctp_sock_migrate(). Signed-off-by: Kuniyuki Iwashima Acked-by: Xin Long Link: https://patch.msgid.link/20251023231751.4168390-3-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/sctp/socket.c | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) diff --git a/net/sctp/socket.c b/net/sctp/socket.c index d190e75e4645..735b1222af95 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -9523,12 +9523,9 @@ static int sctp_sock_migrate(struct sock *oldsk, struct sock *newsk, struct sctp_bind_hashbucket *head; int err; - /* Migrate socket buffer sizes and all the socket level options to the - * new socket. + /* Migrate all the socket level options to the new socket. + * Brute force copy old sctp opt. */ - newsk->sk_sndbuf = oldsk->sk_sndbuf; - newsk->sk_rcvbuf = oldsk->sk_rcvbuf; - /* Brute force copy old sctp opt. */ sctp_copy_descendant(newsk, oldsk); /* Restore the ep value that was overwritten with the above structure From b7185792f80a0069aa4eab87cb4263a1fb611a4e Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Thu, 23 Oct 2025 23:16:52 +0000 Subject: [PATCH 0276/1381] sctp: Don't call sk->sk_prot->init() in sctp_v[46]_create_accept_sk(). sctp_accept() calls sctp_v[46]_create_accept_sk() to allocate a new socket and calls sctp_sock_migrate() to copy fields from the parent socket to the new socket. sctp_v[46]_create_accept_sk() calls sctp_init_sock() to initialise sctp_sock, but most fields are overwritten by sctp_copy_descendant() called from sctp_sock_migrate(). Things done in sctp_init_sock() but not in sctp_sock_migrate() are the following: 1. Copy sk->sk_gso 2. Copy sk->sk_destruct (sctp_v6_init_sock()) 3. Allocate sctp_sock.ep 4. Initialise sctp_sock.pd_lobby 5. Count sk_sockets_allocated_inc(), sock_prot_inuse_add(), and SCTP_DBG_OBJCNT_INC() Let's do these in sctp_copy_sock() and sctp_sock_migrate() and avoid calling sk->sk_prot->init() in sctp_v[46]_create_accept_sk(). Note that sk->sk_destruct is already copied in sctp_copy_sock(). Signed-off-by: Kuniyuki Iwashima Acked-by: Xin Long Link: https://patch.msgid.link/20251023231751.4168390-4-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/sctp/ipv6.c | 8 +------- net/sctp/protocol.c | 8 +------- net/sctp/socket.c | 27 ++++++++++++++++++++++----- 3 files changed, 24 insertions(+), 19 deletions(-) diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c index d725b2158758..c0762424a854 100644 --- a/net/sctp/ipv6.c +++ b/net/sctp/ipv6.c @@ -789,7 +789,7 @@ static struct sock *sctp_v6_create_accept_sk(struct sock *sk, newsk = sk_alloc(sock_net(sk), PF_INET6, GFP_KERNEL, sk->sk_prot, kern); if (!newsk) - goto out; + return NULL; sock_init_data(NULL, newsk); @@ -818,12 +818,6 @@ static struct sock *sctp_v6_create_accept_sk(struct sock *sk, newsk->sk_v6_rcv_saddr = sk->sk_v6_rcv_saddr; - if (newsk->sk_prot->init(newsk)) { - sk_common_release(newsk); - newsk = NULL; - } - -out: return newsk; } diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index 9dbc24af749b..ad2722d1ec15 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -590,7 +590,7 @@ static struct sock *sctp_v4_create_accept_sk(struct sock *sk, struct inet_sock *newinet; if (!newsk) - goto out; + return NULL; sock_init_data(NULL, newsk); @@ -603,12 +603,6 @@ static struct sock *sctp_v4_create_accept_sk(struct sock *sk, newinet->inet_daddr = asoc->peer.primary_addr.v4.sin_addr.s_addr; - if (newsk->sk_prot->init(newsk)) { - sk_common_release(newsk); - newsk = NULL; - } - -out: return newsk; } diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 735b1222af95..70c75ac8da55 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -4851,7 +4851,7 @@ static int sctp_disconnect(struct sock *sk, int flags) */ static struct sock *sctp_accept(struct sock *sk, struct proto_accept_arg *arg) { - struct sctp_sock *sp; + struct sctp_sock *sp, *newsp; struct sctp_endpoint *ep; struct sock *newsk = NULL; struct sctp_association *asoc; @@ -4891,19 +4891,35 @@ static struct sock *sctp_accept(struct sock *sk, struct proto_accept_arg *arg) goto out; } + newsp = sctp_sk(newsk); + newsp->ep = sctp_endpoint_new(newsk, GFP_KERNEL); + if (!newsp->ep) { + error = -ENOMEM; + goto out_release; + } + + skb_queue_head_init(&newsp->pd_lobby); + + sk_sockets_allocated_inc(newsk); + sock_prot_inuse_add(sock_net(sk), newsk->sk_prot, 1); + SCTP_DBG_OBJCNT_INC(sock); + /* Populate the fields of the newsk from the oldsk and migrate the * asoc to the newsk. */ error = sctp_sock_migrate(sk, newsk, asoc, SCTP_SOCKET_TCP); - if (error) { - sk_common_release(newsk); - newsk = NULL; - } + if (error) + goto out_release; out: release_sock(sk); arg->err = error; return newsk; + +out_release: + sk_common_release(newsk); + newsk = NULL; + goto out; } /* The SCTP ioctl handler. */ @@ -9469,6 +9485,7 @@ void sctp_copy_sock(struct sock *newsk, struct sock *sk, newsk->sk_rcvtimeo = READ_ONCE(sk->sk_rcvtimeo); newsk->sk_sndtimeo = READ_ONCE(sk->sk_sndtimeo); newsk->sk_rxhash = sk->sk_rxhash; + newsk->sk_gso_type = sk->sk_gso_type; newinet = inet_sk(newsk); From 151b98d10ef7c3174465e409b99d8762e7e8de60 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Thu, 23 Oct 2025 23:16:53 +0000 Subject: [PATCH 0277/1381] net: Add sk_clone(). sctp_accept() will use sk_clone_lock(), but it will be called with the parent socket locked, and sctp_migrate() acquires the child lock later. Let's add no lock version of sk_clone_lock(). Note that lockdep complains if we simply use bh_lock_sock_nested(). Signed-off-by: Kuniyuki Iwashima Reviewed-by: Xin Long Link: https://patch.msgid.link/20251023231751.4168390-5-kuniyu@google.com Signed-off-by: Jakub Kicinski --- include/net/sock.h | 7 ++++++- net/core/sock.c | 24 ++++++++++++++++-------- 2 files changed, 22 insertions(+), 9 deletions(-) diff --git a/include/net/sock.h b/include/net/sock.h index 01ce231603db..c7e58b8e8a90 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -1822,7 +1822,12 @@ struct sock *sk_alloc(struct net *net, int family, gfp_t priority, void sk_free(struct sock *sk); void sk_net_refcnt_upgrade(struct sock *sk); void sk_destruct(struct sock *sk); -struct sock *sk_clone_lock(const struct sock *sk, const gfp_t priority); +struct sock *sk_clone(const struct sock *sk, const gfp_t priority, bool lock); + +static inline struct sock *sk_clone_lock(const struct sock *sk, const gfp_t priority) +{ + return sk_clone(sk, priority, true); +} struct sk_buff *sock_wmalloc(struct sock *sk, unsigned long size, int force, gfp_t priority); diff --git a/net/core/sock.c b/net/core/sock.c index a99132cc0965..7a9bbc2afcf0 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -2462,13 +2462,16 @@ static void sk_init_common(struct sock *sk) } /** - * sk_clone_lock - clone a socket, and lock its clone - * @sk: the socket to clone - * @priority: for allocation (%GFP_KERNEL, %GFP_ATOMIC, etc) + * sk_clone - clone a socket + * @sk: the socket to clone + * @priority: for allocation (%GFP_KERNEL, %GFP_ATOMIC, etc) + * @lock: if true, lock the cloned sk * - * Caller must unlock socket even in error path (bh_unlock_sock(newsk)) + * If @lock is true, the clone is locked by bh_lock_sock(), and + * caller must unlock socket even in error path by bh_unlock_sock(). */ -struct sock *sk_clone_lock(const struct sock *sk, const gfp_t priority) +struct sock *sk_clone(const struct sock *sk, const gfp_t priority, + bool lock) { struct proto *prot = READ_ONCE(sk->sk_prot); struct sk_filter *filter; @@ -2497,9 +2500,13 @@ struct sock *sk_clone_lock(const struct sock *sk, const gfp_t priority) __netns_tracker_alloc(sock_net(newsk), &newsk->ns_tracker, false, priority); } + sk_node_init(&newsk->sk_node); sock_lock_init(newsk); - bh_lock_sock(newsk); + + if (lock) + bh_lock_sock(newsk); + newsk->sk_backlog.head = newsk->sk_backlog.tail = NULL; newsk->sk_backlog.len = 0; @@ -2590,12 +2597,13 @@ struct sock *sk_clone_lock(const struct sock *sk, const gfp_t priority) * destructor and make plain sk_free() */ newsk->sk_destruct = NULL; - bh_unlock_sock(newsk); + if (lock) + bh_unlock_sock(newsk); sk_free(newsk); newsk = NULL; goto out; } -EXPORT_SYMBOL_GPL(sk_clone_lock); +EXPORT_SYMBOL_GPL(sk_clone); static u32 sk_dst_gso_max_size(struct sock *sk, const struct net_device *dev) { From 16942cf4d3e31b6246b7d000dd823f7b0b38bf8c Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Thu, 23 Oct 2025 23:16:54 +0000 Subject: [PATCH 0278/1381] sctp: Use sk_clone() in sctp_accept(). sctp_accept() calls sctp_v[46]_create_accept_sk() to allocate a new socket and calls sctp_sock_migrate() to copy fields from the parent socket to the new socket. sctp_v4_create_accept_sk() allocates sk by sk_alloc(), initialises it by sock_init_data(), and copy a bunch of fields from the parent socekt by sctp_copy_sock(). sctp_sock_migrate() calls sctp_copy_descendant() to copy most fields in sctp_sock from the parent socket by memcpy(). These can be simply replaced by sk_clone(). Let's consolidate sctp_v[46]_create_accept_sk() to sctp_clone_sock() with sk_clone(). We will reuse sctp_clone_sock() for sctp_do_peeloff() and then remove sctp_copy_descendant(). Note that sock_reset_flag(newsk, SOCK_ZAPPED) is not copied to sctp_clone_sock() as sctp does not use SOCK_ZAPPED at all. Signed-off-by: Kuniyuki Iwashima Acked-by: Xin Long Link: https://patch.msgid.link/20251023231751.4168390-6-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/af_inet.c | 4 +- net/sctp/socket.c | 113 ++++++++++++++++++++++++++++++--------------- 2 files changed, 77 insertions(+), 40 deletions(-) diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index e8771faa5bbf..77f6ae0fc231 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -755,9 +755,7 @@ EXPORT_SYMBOL(inet_stream_connect); void __inet_accept(struct socket *sock, struct socket *newsock, struct sock *newsk) { - /* TODO: use sk_clone_lock() in SCTP and remove protocol checks */ - if (mem_cgroup_sockets_enabled && - (!IS_ENABLED(CONFIG_IP_SCTP) || sk_is_tcp(newsk))) { + if (mem_cgroup_sockets_enabled) { gfp_t gfp = GFP_KERNEL | __GFP_NOFAIL; mem_cgroup_sk_alloc(newsk); diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 70c75ac8da55..826f17747f17 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -4842,6 +4842,74 @@ static int sctp_disconnect(struct sock *sk, int flags) return 0; } +static struct sock *sctp_clone_sock(struct sock *sk, + struct sctp_association *asoc, + enum sctp_socket_type type) +{ + struct sock *newsk = sk_clone(sk, GFP_KERNEL, false); + struct inet_sock *newinet; + struct sctp_sock *newsp; + int err = -ENOMEM; + + if (!newsk) + return ERR_PTR(err); + + /* sk_clone() sets refcnt to 2 */ + sock_put(newsk); + + newinet = inet_sk(newsk); + newsp = sctp_sk(newsk); + + newsp->pf->to_sk_daddr(&asoc->peer.primary_addr, newsk); + newinet->inet_dport = htons(asoc->peer.port); + + newsp->pf->copy_ip_options(sk, newsk); + atomic_set(&newinet->inet_id, get_random_u16()); + + inet_set_bit(MC_LOOP, newsk); + newinet->mc_ttl = 1; + newinet->mc_index = 0; + newinet->mc_list = NULL; + +#if IS_ENABLED(CONFIG_IPV6) + if (sk->sk_family == AF_INET6) { + struct ipv6_pinfo *newnp = inet6_sk(newsk); + + newinet->pinet6 = &((struct sctp6_sock *)newsk)->inet6; + newinet->ipv6_fl_list = NULL; + + memcpy(newnp, inet6_sk(sk), sizeof(struct ipv6_pinfo)); + newnp->ipv6_mc_list = NULL; + newnp->ipv6_ac_list = NULL; + } +#endif + + skb_queue_head_init(&newsp->pd_lobby); + + newsp->ep = sctp_endpoint_new(newsk, GFP_KERNEL); + if (!newsp->ep) + goto out_release; + + SCTP_DBG_OBJCNT_INC(sock); + sk_sockets_allocated_inc(newsk); + sock_prot_inuse_add(sock_net(sk), newsk->sk_prot, 1); + + err = sctp_sock_migrate(sk, newsk, asoc, type); + if (err) + goto out_release; + + /* Set newsk security attributes from original sk and connection + * security attribute from asoc. + */ + security_sctp_sk_clone(asoc, sk, newsk); + + return newsk; + +out_release: + sk_common_release(newsk); + return ERR_PTR(err); +} + /* 4.1.4 accept() - TCP Style Syntax * * Applications use accept() call to remove an established SCTP @@ -4851,18 +4919,13 @@ static int sctp_disconnect(struct sock *sk, int flags) */ static struct sock *sctp_accept(struct sock *sk, struct proto_accept_arg *arg) { - struct sctp_sock *sp, *newsp; - struct sctp_endpoint *ep; - struct sock *newsk = NULL; struct sctp_association *asoc; - long timeo; + struct sock *newsk = NULL; int error = 0; + long timeo; lock_sock(sk); - sp = sctp_sk(sk); - ep = sp->ep; - if (!sctp_style(sk, TCP)) { error = -EOPNOTSUPP; goto out; @@ -4883,43 +4946,19 @@ static struct sock *sctp_accept(struct sock *sk, struct proto_accept_arg *arg) /* We treat the list of associations on the endpoint as the accept * queue and pick the first association on the list. */ - asoc = list_entry(ep->asocs.next, struct sctp_association, asocs); + asoc = list_entry(sctp_sk(sk)->ep->asocs.next, + struct sctp_association, asocs); - newsk = sp->pf->create_accept_sk(sk, asoc, arg->kern); - if (!newsk) { - error = -ENOMEM; - goto out; + newsk = sctp_clone_sock(sk, asoc, SCTP_SOCKET_TCP); + if (IS_ERR(newsk)) { + error = PTR_ERR(newsk); + newsk = NULL; } - newsp = sctp_sk(newsk); - newsp->ep = sctp_endpoint_new(newsk, GFP_KERNEL); - if (!newsp->ep) { - error = -ENOMEM; - goto out_release; - } - - skb_queue_head_init(&newsp->pd_lobby); - - sk_sockets_allocated_inc(newsk); - sock_prot_inuse_add(sock_net(sk), newsk->sk_prot, 1); - SCTP_DBG_OBJCNT_INC(sock); - - /* Populate the fields of the newsk from the oldsk and migrate the - * asoc to the newsk. - */ - error = sctp_sock_migrate(sk, newsk, asoc, SCTP_SOCKET_TCP); - if (error) - goto out_release; - out: release_sock(sk); arg->err = error; return newsk; - -out_release: - sk_common_release(newsk); - newsk = NULL; - goto out; } /* The SCTP ioctl handler. */ From c49ed521f1772ca9203d22a1e5950f337fd5f930 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Thu, 23 Oct 2025 23:16:55 +0000 Subject: [PATCH 0279/1381] sctp: Remove sctp_pf.create_accept_sk(). sctp_v[46]_create_accept_sk() are no longer used. Let's remove sctp_pf.create_accept_sk(). Signed-off-by: Kuniyuki Iwashima Acked-by: Xin Long Link: https://patch.msgid.link/20251023231751.4168390-7-kuniyu@google.com Signed-off-by: Jakub Kicinski --- include/net/sctp/structs.h | 3 --- net/sctp/ipv6.c | 45 -------------------------------------- net/sctp/protocol.c | 27 ----------------------- 3 files changed, 75 deletions(-) diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 2ae390219efd..3dd304e411d0 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -497,9 +497,6 @@ struct sctp_pf { int (*bind_verify) (struct sctp_sock *, union sctp_addr *); int (*send_verify) (struct sctp_sock *, union sctp_addr *); int (*supported_addrs)(const struct sctp_sock *, __be16 *); - struct sock *(*create_accept_sk) (struct sock *sk, - struct sctp_association *asoc, - bool kern); int (*addr_to_user)(struct sctp_sock *sk, union sctp_addr *addr); void (*to_sk_saddr)(union sctp_addr *, struct sock *sk); void (*to_sk_daddr)(union sctp_addr *, struct sock *sk); diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c index c0762424a854..069b7e45d8bd 100644 --- a/net/sctp/ipv6.c +++ b/net/sctp/ipv6.c @@ -777,50 +777,6 @@ static enum sctp_scope sctp_v6_scope(union sctp_addr *addr) return retval; } -/* Create and initialize a new sk for the socket to be returned by accept(). */ -static struct sock *sctp_v6_create_accept_sk(struct sock *sk, - struct sctp_association *asoc, - bool kern) -{ - struct ipv6_pinfo *newnp, *np = inet6_sk(sk); - struct sctp6_sock *newsctp6sk; - struct inet_sock *newinet; - struct sock *newsk; - - newsk = sk_alloc(sock_net(sk), PF_INET6, GFP_KERNEL, sk->sk_prot, kern); - if (!newsk) - return NULL; - - sock_init_data(NULL, newsk); - - sctp_copy_sock(newsk, sk, asoc); - sock_reset_flag(sk, SOCK_ZAPPED); - - newsctp6sk = (struct sctp6_sock *)newsk; - newinet = inet_sk(newsk); - newinet->pinet6 = &newsctp6sk->inet6; - newinet->ipv6_fl_list = NULL; - - sctp_sk(newsk)->v4mapped = sctp_sk(sk)->v4mapped; - - newnp = inet6_sk(newsk); - - memcpy(newnp, np, sizeof(struct ipv6_pinfo)); - newnp->ipv6_mc_list = NULL; - newnp->ipv6_ac_list = NULL; - - sctp_v6_copy_ip_options(sk, newsk); - - /* Initialize sk's sport, dport, rcv_saddr and daddr for getsockname() - * and getpeername(). - */ - sctp_v6_to_sk_daddr(&asoc->peer.primary_addr, newsk); - - newsk->sk_v6_rcv_saddr = sk->sk_v6_rcv_saddr; - - return newsk; -} - /* Format a sockaddr for return to user space. This makes sure the return is * AF_INET or AF_INET6 depending on the SCTP_I_WANT_MAPPED_V4_ADDR option. */ @@ -1167,7 +1123,6 @@ static struct sctp_pf sctp_pf_inet6 = { .bind_verify = sctp_inet6_bind_verify, .send_verify = sctp_inet6_send_verify, .supported_addrs = sctp_inet6_supported_addrs, - .create_accept_sk = sctp_v6_create_accept_sk, .addr_to_user = sctp_v6_addr_to_user, .to_sk_saddr = sctp_v6_to_sk_saddr, .to_sk_daddr = sctp_v6_to_sk_daddr, diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index ad2722d1ec15..2c3398f75d76 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -580,32 +580,6 @@ static int sctp_v4_is_ce(const struct sk_buff *skb) return INET_ECN_is_ce(ip_hdr(skb)->tos); } -/* Create and initialize a new sk for the socket returned by accept(). */ -static struct sock *sctp_v4_create_accept_sk(struct sock *sk, - struct sctp_association *asoc, - bool kern) -{ - struct sock *newsk = sk_alloc(sock_net(sk), PF_INET, GFP_KERNEL, - sk->sk_prot, kern); - struct inet_sock *newinet; - - if (!newsk) - return NULL; - - sock_init_data(NULL, newsk); - - sctp_copy_sock(newsk, sk, asoc); - sock_reset_flag(newsk, SOCK_ZAPPED); - - sctp_v4_copy_ip_options(sk, newsk); - - newinet = inet_sk(newsk); - - newinet->inet_daddr = asoc->peer.primary_addr.v4.sin_addr.s_addr; - - return newsk; -} - static int sctp_v4_addr_to_user(struct sctp_sock *sp, union sctp_addr *addr) { /* No address mapping for V4 sockets */ @@ -1113,7 +1087,6 @@ static struct sctp_pf sctp_pf_inet = { .bind_verify = sctp_inet_bind_verify, .send_verify = sctp_inet_send_verify, .supported_addrs = sctp_inet_supported_addrs, - .create_accept_sk = sctp_v4_create_accept_sk, .addr_to_user = sctp_v4_addr_to_user, .to_sk_saddr = sctp_v4_to_sk_saddr, .to_sk_daddr = sctp_v4_to_sk_daddr, From b7ddb55f31279f4e59acde3395fc03c3d94b6e5f Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Thu, 23 Oct 2025 23:16:56 +0000 Subject: [PATCH 0280/1381] sctp: Use sctp_clone_sock() in sctp_do_peeloff(). sctp_do_peeloff() calls sock_create() to allocate and initialise struct sock, inet_sock, and sctp_sock, but later sctp_copy_sock() and sctp_sock_migrate() overwrite most fields. What sctp_do_peeloff() does is more like accept(). Let's use sock_create_lite() and sctp_clone_sock(). Signed-off-by: Kuniyuki Iwashima Acked-by: Xin Long Link: https://patch.msgid.link/20251023231751.4168390-8-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/af_inet.c | 1 + net/sctp/socket.c | 36 +++++++++++++++--------------------- 2 files changed, 16 insertions(+), 21 deletions(-) diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index 77f6ae0fc231..0784e2a873a1 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -788,6 +788,7 @@ void __inet_accept(struct socket *sock, struct socket *newsock, struct sock *new newsock->state = SS_CONNECTED; } +EXPORT_SYMBOL_GPL(__inet_accept); /* * Accept a pending connection. The TCP layer now gives BSD semantics. diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 826f17747f17..60d3e340dfed 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -5671,11 +5671,11 @@ static int sctp_getsockopt_autoclose(struct sock *sk, int len, char __user *optv /* Helper routine to branch off an association to a new socket. */ static int sctp_do_peeloff(struct sock *sk, sctp_assoc_t id, - struct socket **sockp) + struct socket **sockp) { struct sctp_association *asoc = sctp_id2assoc(sk, id); - struct sctp_sock *sp = sctp_sk(sk); struct socket *sock; + struct sock *newsk; int err = 0; /* Do not peel off from one netns to another one. */ @@ -5691,30 +5691,24 @@ static int sctp_do_peeloff(struct sock *sk, sctp_assoc_t id, if (!sctp_style(sk, UDP)) return -EINVAL; - /* Create a new socket. */ - err = sock_create(sk->sk_family, SOCK_SEQPACKET, IPPROTO_SCTP, &sock); - if (err < 0) + err = sock_create_lite(sk->sk_family, SOCK_SEQPACKET, IPPROTO_SCTP, &sock); + if (err) return err; - sctp_copy_sock(sock->sk, sk, asoc); - - /* Make peeled-off sockets more like 1-1 accepted sockets. - * Set the daddr and initialize id to something more random and also - * copy over any ip options. - */ - sp->pf->to_sk_daddr(&asoc->peer.primary_addr, sock->sk); - sp->pf->copy_ip_options(sk, sock->sk); - - /* Populate the fields of the newsk from the oldsk and migrate the - * asoc to the newsk. - */ - err = sctp_sock_migrate(sk, sock->sk, asoc, - SCTP_SOCKET_UDP_HIGH_BANDWIDTH); - if (err) { + newsk = sctp_clone_sock(sk, asoc, SCTP_SOCKET_UDP_HIGH_BANDWIDTH); + if (IS_ERR(newsk)) { sock_release(sock); - sock = NULL; + *sockp = NULL; + return PTR_ERR(newsk); } + lock_sock_nested(newsk, SINGLE_DEPTH_NESTING); + __inet_accept(sk->sk_socket, sock, newsk); + release_sock(newsk); + + sock->ops = sk->sk_socket->ops; + __module_get(sock->ops->owner); + *sockp = sock; return err; From 71068e2e1b6bd78f5599e5bc89e125a75149884b Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Thu, 23 Oct 2025 23:16:57 +0000 Subject: [PATCH 0281/1381] sctp: Remove sctp_copy_sock() and sctp_copy_descendant(). Now, sctp_accept() and sctp_do_peeloff() use sk_clone(), and we no longer need sctp_copy_sock() and sctp_copy_descendant(). Let's remove them. Signed-off-by: Kuniyuki Iwashima Acked-by: Xin Long Link: https://patch.msgid.link/20251023231751.4168390-9-kuniyu@google.com Signed-off-by: Jakub Kicinski --- include/net/inet_sock.h | 8 ----- include/net/sctp/sctp.h | 3 +- net/sctp/socket.c | 71 ----------------------------------------- 3 files changed, 1 insertion(+), 81 deletions(-) diff --git a/include/net/inet_sock.h b/include/net/inet_sock.h index b6ec08072533..ac1c75975908 100644 --- a/include/net/inet_sock.h +++ b/include/net/inet_sock.h @@ -355,14 +355,6 @@ static inline struct sock *skb_to_full_sk(const struct sk_buff *skb) #define inet_sk(ptr) container_of_const(ptr, struct inet_sock, sk) -static inline void __inet_sk_copy_descendant(struct sock *sk_to, - const struct sock *sk_from, - const int ancestor_size) -{ - memcpy(inet_sk(sk_to) + 1, inet_sk(sk_from) + 1, - sk_from->sk_prot->obj_size - ancestor_size); -} - int inet_sk_rebuild_header(struct sock *sk); /** diff --git a/include/net/sctp/sctp.h b/include/net/sctp/sctp.h index e96d1bd087f6..bb4b80c12541 100644 --- a/include/net/sctp/sctp.h +++ b/include/net/sctp/sctp.h @@ -94,8 +94,7 @@ void sctp_data_ready(struct sock *sk); __poll_t sctp_poll(struct file *file, struct socket *sock, poll_table *wait); void sctp_sock_rfree(struct sk_buff *skb); -void sctp_copy_sock(struct sock *newsk, struct sock *sk, - struct sctp_association *asoc); + extern struct percpu_counter sctp_sockets_allocated; int sctp_asconf_mgmt(struct sctp_sock *, struct sctp_sockaddr_entry *); struct sk_buff *sctp_skb_recv_datagram(struct sock *, int, int *); diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 60d3e340dfed..ac737e60829b 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -9491,72 +9491,6 @@ static void sctp_skb_set_owner_r_frag(struct sk_buff *skb, struct sock *sk) sctp_skb_set_owner_r(skb, sk); } -void sctp_copy_sock(struct sock *newsk, struct sock *sk, - struct sctp_association *asoc) -{ - struct inet_sock *inet = inet_sk(sk); - struct inet_sock *newinet; - struct sctp_sock *sp = sctp_sk(sk); - - newsk->sk_type = sk->sk_type; - newsk->sk_bound_dev_if = sk->sk_bound_dev_if; - newsk->sk_flags = sk->sk_flags; - newsk->sk_tsflags = sk->sk_tsflags; - newsk->sk_no_check_tx = sk->sk_no_check_tx; - newsk->sk_no_check_rx = sk->sk_no_check_rx; - newsk->sk_reuse = sk->sk_reuse; - sctp_sk(newsk)->reuse = sp->reuse; - - newsk->sk_shutdown = sk->sk_shutdown; - newsk->sk_destruct = sk->sk_destruct; - newsk->sk_family = sk->sk_family; - newsk->sk_protocol = IPPROTO_SCTP; - newsk->sk_backlog_rcv = sk->sk_prot->backlog_rcv; - newsk->sk_sndbuf = sk->sk_sndbuf; - newsk->sk_rcvbuf = sk->sk_rcvbuf; - newsk->sk_lingertime = sk->sk_lingertime; - newsk->sk_rcvtimeo = READ_ONCE(sk->sk_rcvtimeo); - newsk->sk_sndtimeo = READ_ONCE(sk->sk_sndtimeo); - newsk->sk_rxhash = sk->sk_rxhash; - newsk->sk_gso_type = sk->sk_gso_type; - - newinet = inet_sk(newsk); - - /* Initialize sk's sport, dport, rcv_saddr and daddr for - * getsockname() and getpeername() - */ - newinet->inet_sport = inet->inet_sport; - newinet->inet_saddr = inet->inet_saddr; - newinet->inet_rcv_saddr = inet->inet_rcv_saddr; - newinet->inet_dport = htons(asoc->peer.port); - newinet->pmtudisc = inet->pmtudisc; - atomic_set(&newinet->inet_id, get_random_u16()); - - newinet->uc_ttl = inet->uc_ttl; - inet_set_bit(MC_LOOP, newsk); - newinet->mc_ttl = 1; - newinet->mc_index = 0; - newinet->mc_list = NULL; - - if (newsk->sk_flags & SK_FLAGS_TIMESTAMP) - net_enable_timestamp(); - - /* Set newsk security attributes from original sk and connection - * security attribute from asoc. - */ - security_sctp_sk_clone(asoc, sk, newsk); -} - -static inline void sctp_copy_descendant(struct sock *sk_to, - const struct sock *sk_from) -{ - size_t ancestor_size = sizeof(struct inet_sock); - - ancestor_size += sk_from->sk_prot->obj_size; - ancestor_size -= offsetof(struct sctp_sock, pd_lobby); - __inet_sk_copy_descendant(sk_to, sk_from, ancestor_size); -} - /* Populate the fields of the newsk from the oldsk and migrate the assoc * and its messages to the newsk. */ @@ -9573,11 +9507,6 @@ static int sctp_sock_migrate(struct sock *oldsk, struct sock *newsk, struct sctp_bind_hashbucket *head; int err; - /* Migrate all the socket level options to the new socket. - * Brute force copy old sctp opt. - */ - sctp_copy_descendant(newsk, oldsk); - /* Restore the ep value that was overwritten with the above structure * copy. */ From 6f147c8328e045de3a35155ca7c883d88da9e916 Mon Sep 17 00:00:00 2001 From: Yue Haibing Date: Fri, 24 Oct 2025 10:51:45 +0800 Subject: [PATCH 0282/1381] net/sched: Remove unused typedef psched_tdiff_t Since commit 051d44209842 ("net/sched: Retire CBQ qdisc") this is not used anymore. Signed-off-by: Yue Haibing Link: https://patch.msgid.link/20251024025145.4069583-1-yuehaibing@huawei.com Signed-off-by: Jakub Kicinski --- include/net/pkt_sched.h | 1 - 1 file changed, 1 deletion(-) diff --git a/include/net/pkt_sched.h b/include/net/pkt_sched.h index c660ac871083..4678db45832a 100644 --- a/include/net/pkt_sched.h +++ b/include/net/pkt_sched.h @@ -43,7 +43,6 @@ struct qdisc_walker { */ typedef u64 psched_time_t; -typedef long psched_tdiff_t; /* Avoid doing 64 bit divide */ #define PSCHED_SHIFT 6 From 0ae1ac7335ca2328c42fbbe2b71bc9c3fc8e65d9 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 24 Oct 2025 12:07:07 +0000 Subject: [PATCH 0283/1381] tcp: remove one ktime_get() from recvmsg() fast path Each time some payload is consumed by user space (recvmsg() and friends), TCP calls tcp_rcv_space_adjust() to run DRS algorithm to check if an increase of sk->sk_rcvbuf is needed. This function is based on time sampling, and currently calls tcp_mstamp_refresh(tp), which is a wrapper around ktime_get_ns(). ktime_get_ns() has a high cost on some platforms. 100+ cycles for rdtscp on AMD EPYC Turin for instance. We do not have to refresh tp->tcp_mpstamp, using the last cached value is enough. We only need to refresh it from __tcp_cleanup_rbuf() if an ACK must be sent (this is a rare event). Signed-off-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20251024120707.3516550-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/tcp.c | 4 +++- net/ipv4/tcp_input.c | 10 ++++++++-- 2 files changed, 11 insertions(+), 3 deletions(-) diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index b79da6d39392..a9345aa5a2e5 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -1556,8 +1556,10 @@ void __tcp_cleanup_rbuf(struct sock *sk, int copied) time_to_ack = true; } } - if (time_to_ack) + if (time_to_ack) { + tcp_mstamp_refresh(tp); tcp_send_ack(sk); + } } void tcp_cleanup_rbuf(struct sock *sk, int copied) diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 8fc97f4d8a6b..ff19f6e54d55 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -928,9 +928,15 @@ void tcp_rcv_space_adjust(struct sock *sk) trace_tcp_rcv_space_adjust(sk); - tcp_mstamp_refresh(tp); + if (unlikely(!tp->rcv_rtt_est.rtt_us)) + return; + + /* We do not refresh tp->tcp_mstamp here. + * Some platforms have expensive ktime_get() implementations. + * Using the last cached value is enough for DRS. + */ time = tcp_stamp_us_delta(tp->tcp_mstamp, tp->rcvq_space.time); - if (time < (tp->rcv_rtt_est.rtt_us >> 3) || tp->rcv_rtt_est.rtt_us == 0) + if (time < (tp->rcv_rtt_est.rtt_us >> 3)) return; /* Number of bytes copied to user in last RTT */ From 19ab0a22efbd824f342a794a5c61bce7842daef5 Mon Sep 17 00:00:00 2001 From: Lad Prabhakar Date: Sat, 25 Oct 2025 07:48:50 +0100 Subject: [PATCH 0284/1381] dt-bindings: net: phy: vsc8531: Convert to DT schema Convert VSC8531 Gigabit ethernet phy binding to DT schema format. While at it add compatible string for VSC8541 PHY which is very much similar to the VSC8531 PHY and is already supported in the kernel. VSC8541 PHY is present on the Renesas RZ/T2H EVK. Signed-off-by: Lad Prabhakar Reviewed-by: Rob Herring (Arm) Link: https://patch.msgid.link/20251025064850.393797-1-prabhakar.mahadev-lad.rj@bp.renesas.com Signed-off-by: Jakub Kicinski --- .../bindings/net/mscc-phy-vsc8531.txt | 73 ---------- .../bindings/net/mscc-phy-vsc8531.yaml | 131 ++++++++++++++++++ .../devicetree/bindings/vendor-prefixes.yaml | 2 +- 3 files changed, 132 insertions(+), 74 deletions(-) delete mode 100644 Documentation/devicetree/bindings/net/mscc-phy-vsc8531.txt create mode 100644 Documentation/devicetree/bindings/net/mscc-phy-vsc8531.yaml diff --git a/Documentation/devicetree/bindings/net/mscc-phy-vsc8531.txt b/Documentation/devicetree/bindings/net/mscc-phy-vsc8531.txt deleted file mode 100644 index 0a3647fe331b..000000000000 --- a/Documentation/devicetree/bindings/net/mscc-phy-vsc8531.txt +++ /dev/null @@ -1,73 +0,0 @@ -* Microsemi - vsc8531 Giga bit ethernet phy - -Optional properties: -- vsc8531,vddmac : The vddmac in mV. Allowed values is listed - in the first row of Table 1 (below). - This property is only used in combination - with the 'edge-slowdown' property. - Default value is 3300. -- vsc8531,edge-slowdown : % the edge should be slowed down relative to - the fastest possible edge time. - Edge rate sets the drive strength of the MAC - interface output signals. Changing the - drive strength will affect the edge rate of - the output signal. The goal of this setting - is to help reduce electrical emission (EMI) - by being able to reprogram drive strength - and in effect slow down the edge rate if - desired. - To adjust the edge-slowdown, the 'vddmac' - must be specified. Table 1 lists the - supported edge-slowdown values for a given - 'vddmac'. - Default value is 0%. - Ref: Table:1 - Edge rate change (below). -- vsc8531,led-[N]-mode : LED mode. Specify how the LED[N] should behave. - N depends on the number of LEDs supported by a - PHY. - Allowed values are defined in - "include/dt-bindings/net/mscc-phy-vsc8531.h". - Default values are VSC8531_LINK_1000_ACTIVITY (1), - VSC8531_LINK_100_ACTIVITY (2), - VSC8531_LINK_ACTIVITY (0) and - VSC8531_DUPLEX_COLLISION (8). -- load-save-gpios : GPIO used for the load/save operation of the PTP - hardware clock (PHC). - - -Table: 1 - Edge rate change -----------------------------------------------------------------| -| Edge Rate Change (VDDMAC) | -| | -| 3300 mV 2500 mV 1800 mV 1500 mV | -|---------------------------------------------------------------| -| 0% 0% 0% 0% | -| (Fastest) (recommended) (recommended) | -|---------------------------------------------------------------| -| 2% 3% 5% 6% | -|---------------------------------------------------------------| -| 4% 6% 9% 14% | -|---------------------------------------------------------------| -| 7% 10% 16% 21% | -|(recommended) (recommended) | -|---------------------------------------------------------------| -| 10% 14% 23% 29% | -|---------------------------------------------------------------| -| 17% 23% 35% 42% | -|---------------------------------------------------------------| -| 29% 37% 52% 58% | -|---------------------------------------------------------------| -| 53% 63% 76% 77% | -| (slowest) | -|---------------------------------------------------------------| - -Example: - - vsc8531_0: ethernet-phy@0 { - compatible = "ethernet-phy-id0007.0570"; - vsc8531,vddmac = <3300>; - vsc8531,edge-slowdown = <7>; - vsc8531,led-0-mode = ; - vsc8531,led-1-mode = ; - load-save-gpios = <&gpio 10 GPIO_ACTIVE_HIGH>; - }; diff --git a/Documentation/devicetree/bindings/net/mscc-phy-vsc8531.yaml b/Documentation/devicetree/bindings/net/mscc-phy-vsc8531.yaml new file mode 100644 index 000000000000..0afbd0ff126f --- /dev/null +++ b/Documentation/devicetree/bindings/net/mscc-phy-vsc8531.yaml @@ -0,0 +1,131 @@ +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/net/mscc-phy-vsc8531.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Microsemi VSC8531 Gigabit Ethernet PHY + +maintainers: + - Lad Prabhakar + +description: + The VSC8531 is a Gigabit Ethernet PHY with configurable MAC interface + drive strength and LED modes. + +allOf: + - $ref: ethernet-phy.yaml# + +select: + properties: + compatible: + contains: + enum: + - ethernet-phy-id0007.0570 # VSC8531 + - ethernet-phy-id0007.0772 # VSC8541 + required: + - compatible + +properties: + compatible: + items: + - enum: + - ethernet-phy-id0007.0570 # VSC8531 + - ethernet-phy-id0007.0772 # VSC8541 + - const: ethernet-phy-ieee802.3-c22 + + vsc8531,vddmac: + $ref: /schemas/types.yaml#/definitions/uint32 + description: + The VDDMAC voltage in millivolts. This property is used in combination + with the edge-slowdown property to control the drive strength of the + MAC interface output signals. + enum: [3300, 2500, 1800, 1500] + default: 3300 + + vsc8531,edge-slowdown: + $ref: /schemas/types.yaml#/definitions/uint32 + description: > + Percentage by which the edge rate should be slowed down relative to + the fastest possible edge time. This setting helps reduce electromagnetic + interference (EMI) by adjusting the drive strength of the MAC interface + output signals. Valid values depend on the vddmac voltage setting + according to the edge rate change table in the datasheet. + + - When vsc8531,vddmac = 3300 mV: allowed values are 0, 2, 4, 7, 10, 17, 29, and 53. + (Recommended: 7) + - When vsc8531,vddmac = 2500 mV: allowed values are 0, 3, 6, 10, 14, 23, 37, and 63. + (Recommended: 10) + - When vsc8531,vddmac = 1800 mV: allowed values are 0, 5, 9, 16, 23, 35, 52, and 76. + (Recommended: 0) + - When vsc8531,vddmac = 1500 mV: allowed values are 0, 6, 14, 21, 29, 42, 58, and 77. + (Recommended: 0) + enum: [0, 2, 3, 4, 5, 6, 7, 9, 10, 14, 16, 17, 21, 23, 29, 35, 37, 42, 52, 53, 58, 63, 76, 77] + default: 0 + + vsc8531,led-0-mode: + $ref: /schemas/types.yaml#/definitions/uint32 + description: LED[0] behavior mode. See include/dt-bindings/net/mscc-phy-vsc8531.h + for available modes. + minimum: 0 + maximum: 15 + default: 1 + + vsc8531,led-1-mode: + $ref: /schemas/types.yaml#/definitions/uint32 + description: LED[1] behavior mode. See include/dt-bindings/net/mscc-phy-vsc8531.h + for available modes. + minimum: 0 + maximum: 15 + default: 2 + + vsc8531,led-2-mode: + $ref: /schemas/types.yaml#/definitions/uint32 + description: LED[2] behavior mode. See include/dt-bindings/net/mscc-phy-vsc8531.h + for available modes. + minimum: 0 + maximum: 15 + default: 0 + + vsc8531,led-3-mode: + $ref: /schemas/types.yaml#/definitions/uint32 + description: LED[3] behavior mode. See include/dt-bindings/net/mscc-phy-vsc8531.h + for available modes. + minimum: 0 + maximum: 15 + default: 8 + + load-save-gpios: + description: GPIO phandle used for the load/save operation of the PTP hardware + clock (PHC). + maxItems: 1 + +dependencies: + vsc8531,edge-slowdown: + - vsc8531,vddmac + +required: + - compatible + - reg + +unevaluatedProperties: false + +examples: + - | + #include + #include + + mdio { + #address-cells = <1>; + #size-cells = <0>; + + ethernet-phy@0 { + compatible = "ethernet-phy-id0007.0772", "ethernet-phy-ieee802.3-c22"; + reg = <0>; + vsc8531,vddmac = <3300>; + vsc8531,edge-slowdown = <7>; + vsc8531,led-0-mode = ; + vsc8531,led-1-mode = ; + load-save-gpios = <&gpio 10 GPIO_ACTIVE_HIGH>; + }; + }; diff --git a/Documentation/devicetree/bindings/vendor-prefixes.yaml b/Documentation/devicetree/bindings/vendor-prefixes.yaml index f1d1882009ba..424aa7b911a7 100644 --- a/Documentation/devicetree/bindings/vendor-prefixes.yaml +++ b/Documentation/devicetree/bindings/vendor-prefixes.yaml @@ -20,7 +20,7 @@ patternProperties: "^(keypad|m25p|max8952|max8997|max8998|mpmc),.*": true "^(pciclass|pinctrl-single|#pinctrl-single|PowerPC),.*": true "^(pl022|pxa-mmc|rcar_sound|rotary-encoder|s5m8767|sdhci),.*": true - "^(simple-audio-card|st-plgpio|st-spics|ts),.*": true + "^(simple-audio-card|st-plgpio|st-spics|ts|vsc8531),.*": true "^pool[0-3],.*": true # Keep list in alphabetical order. From b47d748110fd2c2911b59b169b10cc14fd5d3648 Mon Sep 17 00:00:00 2001 From: Ping-Ke Shih Date: Tue, 21 Oct 2025 21:33:55 +0800 Subject: [PATCH 0285/1381] wifi: rtw89: splice C2H events queue to local to prevent racing RX task enqueues C2H events and fork a C2H work to handle events, but the work uses skb_queue_walk_safe() without a lock causing potential racing. Use skb_queue_splice() and its friends with spin_lock to splice the queue to local, and then still use skb_queue_walk_safe() to iterate all events. Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251021133402.15467-2-pkshih@realtek.com --- drivers/net/wireless/realtek/rtw89/fw.c | 31 ++++++++++++++++++------- 1 file changed, 22 insertions(+), 9 deletions(-) diff --git a/drivers/net/wireless/realtek/rtw89/fw.c b/drivers/net/wireless/realtek/rtw89/fw.c index ab904a7def1b..9a926bb2cf00 100644 --- a/drivers/net/wireless/realtek/rtw89/fw.c +++ b/drivers/net/wireless/realtek/rtw89/fw.c @@ -6891,11 +6891,18 @@ void rtw89_fw_c2h_work(struct wiphy *wiphy, struct wiphy_work *work) struct rtw89_dev *rtwdev = container_of(work, struct rtw89_dev, c2h_work); struct sk_buff *skb, *tmp; + struct sk_buff_head c2hq; + unsigned long flags; lockdep_assert_wiphy(rtwdev->hw->wiphy); - skb_queue_walk_safe(&rtwdev->c2h_queue, skb, tmp) { - skb_unlink(skb, &rtwdev->c2h_queue); + __skb_queue_head_init(&c2hq); + + spin_lock_irqsave(&rtwdev->c2h_queue.lock, flags); + skb_queue_splice_init(&rtwdev->c2h_queue, &c2hq); + spin_unlock_irqrestore(&rtwdev->c2h_queue.lock, flags); + + skb_queue_walk_safe(&c2hq, skb, tmp) { rtw89_fw_c2h_cmd_handle(rtwdev, skb); dev_kfree_skb_any(skb); } @@ -6905,18 +6912,20 @@ void rtw89_fw_c2h_purge_obsoleted_scan_events(struct rtw89_dev *rtwdev) { struct rtw89_hw_scan_info *scan_info = &rtwdev->scan_info; struct sk_buff *skb, *tmp; - int limit; + struct sk_buff_head c2hq; + unsigned long flags; lockdep_assert_wiphy(rtwdev->hw->wiphy); - limit = skb_queue_len(&rtwdev->c2h_queue); + __skb_queue_head_init(&c2hq); - skb_queue_walk_safe(&rtwdev->c2h_queue, skb, tmp) { + spin_lock_irqsave(&rtwdev->c2h_queue.lock, flags); + skb_queue_splice_init(&rtwdev->c2h_queue, &c2hq); + spin_unlock_irqrestore(&rtwdev->c2h_queue.lock, flags); + + skb_queue_walk_safe(&c2hq, skb, tmp) { struct rtw89_fw_c2h_attr *attr = RTW89_SKB_C2H_CB(skb); - if (--limit < 0) - return; - if (!attr->is_scan_event || attr->scan_seq == scan_info->seq) continue; @@ -6924,9 +6933,13 @@ void rtw89_fw_c2h_purge_obsoleted_scan_events(struct rtw89_dev *rtwdev) "purge obsoleted scan event with seq=%d (cur=%d)\n", attr->scan_seq, scan_info->seq); - skb_unlink(skb, &rtwdev->c2h_queue); + __skb_unlink(skb, &c2hq); dev_kfree_skb_any(skb); } + + spin_lock_irqsave(&rtwdev->c2h_queue.lock, flags); + skb_queue_splice(&c2hq, &rtwdev->c2h_queue); + spin_unlock_irqrestore(&rtwdev->c2h_queue.lock, flags); } static int rtw89_fw_write_h2c_reg(struct rtw89_dev *rtwdev, From f44a9b14a78f1e2e2c57dbc1f1e93889ad2f6ac0 Mon Sep 17 00:00:00 2001 From: Ping-Ke Shih Date: Tue, 21 Oct 2025 21:33:56 +0800 Subject: [PATCH 0286/1381] wifi: rtw89: use skb_dequeue() for queued ROC packets to prevent racing TX task can enqueue ROC skb, but other tasks dequeue the skb. Using skb_queue_walk_safe() without locking will cause potential racing. Use skb_dequeue() with lock instead. Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251021133402.15467-3-pkshih@realtek.com --- drivers/net/wireless/realtek/rtw89/core.c | 12 ++++-------- 1 file changed, 4 insertions(+), 8 deletions(-) diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c index 917b2adede61..8b40cada4149 100644 --- a/drivers/net/wireless/realtek/rtw89/core.c +++ b/drivers/net/wireless/realtek/rtw89/core.c @@ -3632,12 +3632,10 @@ void rtw89_core_free_sta_pending_roc_tx(struct rtw89_dev *rtwdev, struct ieee80211_sta *sta) { struct rtw89_sta *rtwsta = sta_to_rtwsta(sta); - struct sk_buff *skb, *tmp; + struct sk_buff *skb; - skb_queue_walk_safe(&rtwsta->roc_queue, skb, tmp) { - skb_unlink(skb, &rtwsta->roc_queue); + while ((skb = skb_dequeue(&rtwsta->roc_queue))) dev_kfree_skb_any(skb); - } } static void rtw89_core_stop_tx_ba_session(struct rtw89_dev *rtwdev, @@ -3881,8 +3879,8 @@ static void rtw89_core_sta_pending_tx_iter(void *data, struct ieee80211_vif *vif = rtwvif_to_vif(rtwvif); struct rtw89_vif_link *target = data; struct rtw89_vif_link *rtwvif_link; - struct sk_buff *skb, *tmp; unsigned int link_id; + struct sk_buff *skb; int qsel, ret; rtw89_vif_for_each_link(rtwvif, rtwvif_link, link_id) @@ -3895,9 +3893,7 @@ static void rtw89_core_sta_pending_tx_iter(void *data, if (skb_queue_len(&rtwsta->roc_queue) == 0) return; - skb_queue_walk_safe(&rtwsta->roc_queue, skb, tmp) { - skb_unlink(skb, &rtwsta->roc_queue); - + while ((skb = skb_dequeue(&rtwsta->roc_queue))) { ret = rtw89_core_tx_write(rtwdev, vif, sta, skb, &qsel); if (ret) { rtw89_warn(rtwdev, "pending tx failed with %d\n", ret); From e79382ab0347b1d7f9f26d27cb4645e4f053bcc0 Mon Sep 17 00:00:00 2001 From: Zong-Zhe Yang Date: Tue, 21 Oct 2025 21:33:57 +0800 Subject: [PATCH 0287/1381] wifi: rtw89: mlo: handle needed H2C when link switching is requested by stack To switch link, FW needs H2C commands to indicate which link is on or off. Originally, these H2C commands are considered only when the link switching is initiated by driver. But, in some cases, e.g. ml_reconf or TTLM, link switching would be initiated by stack. Hence, plan these H2C commands into ieee80211_ops. Signed-off-by: Zong-Zhe Yang Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251021133402.15467-4-pkshih@realtek.com --- drivers/net/wireless/realtek/rtw89/core.c | 18 +----- drivers/net/wireless/realtek/rtw89/core.h | 8 +++ drivers/net/wireless/realtek/rtw89/mac80211.c | 64 ++++++++++++++++++- 3 files changed, 74 insertions(+), 16 deletions(-) diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c index 8b40cada4149..a3f156bf708e 100644 --- a/drivers/net/wireless/realtek/rtw89/core.c +++ b/drivers/net/wireless/realtek/rtw89/core.c @@ -6010,7 +6010,7 @@ int rtw89_core_mlsr_switch(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwvif, struct ieee80211_vif *vif = rtwvif_to_vif(rtwvif); u16 usable_links = ieee80211_vif_usable_links(vif); u16 active_links = vif->active_links; - struct rtw89_vif_link *target, *cur; + struct rtw89_vif_link *target; int ret; lockdep_assert_wiphy(rtwdev->hw->wiphy); @@ -6036,11 +6036,9 @@ int rtw89_core_mlsr_switch(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwvif, ieee80211_stop_queues(rtwdev->hw); flush_work(&rtwdev->txq_work); - cur = rtw89_get_designated_link(rtwvif); - - ret = ieee80211_set_active_links(vif, active_links | BIT(link_id)); + ret = ieee80211_set_active_links(vif, BIT(link_id)); if (ret) { - rtw89_err(rtwdev, "%s: failed to activate link id %u\n", + rtw89_err(rtwdev, "%s: failed to work on link id %u\n", __func__, link_id); goto wake_queue; } @@ -6055,16 +6053,6 @@ int rtw89_core_mlsr_switch(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwvif, goto wake_queue; } - if (likely(cur)) - rtw89_fw_h2c_mlo_link_cfg(rtwdev, cur, false); - - rtw89_fw_h2c_mlo_link_cfg(rtwdev, target, true); - - ret = ieee80211_set_active_links(vif, BIT(link_id)); - if (ret) - rtw89_err(rtwdev, "%s: failed to inactivate links 0x%x\n", - __func__, active_links); - rtw89_chip_rfk_channel(rtwdev, target); rtwvif->mlo_mode = RTW89_MLO_MODE_MLSR; diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h index 928c8c84c964..7c74ed3cfaf7 100644 --- a/drivers/net/wireless/realtek/rtw89/core.h +++ b/drivers/net/wireless/realtek/rtw89/core.h @@ -5933,6 +5933,7 @@ struct rtw89_mcc_info { enum rtw89_mlo_mode { RTW89_MLO_MODE_MLSR = 0, + RTW89_MLO_MODE_EMLSR = 1, NUM_OF_RTW89_MLO_MODE, }; @@ -6097,6 +6098,12 @@ struct rtw89_link_conf_container { struct ieee80211_bss_conf *link_conf[IEEE80211_MLD_MAX_NUM_LINKS]; }; +struct rtw89_vif_ml_trans { + u16 mediate_links; + u16 links_to_del; + u16 links_to_add; +}; + #define RTW89_VIF_IDLE_LINK_ID 0 struct rtw89_vif { @@ -6119,6 +6126,7 @@ struct rtw89_vif { bool offchan; enum rtw89_mlo_mode mlo_mode; + struct rtw89_vif_ml_trans ml_trans; struct list_head dlink_pool; u8 links_inst_valid_num; diff --git a/drivers/net/wireless/realtek/rtw89/mac80211.c b/drivers/net/wireless/realtek/rtw89/mac80211.c index 7b04183a3a5d..a19304ff8306 100644 --- a/drivers/net/wireless/realtek/rtw89/mac80211.c +++ b/drivers/net/wireless/realtek/rtw89/mac80211.c @@ -718,6 +718,17 @@ static void rtw89_ops_vif_cfg_changed(struct ieee80211_hw *hw, if (changed & BSS_CHANGED_ARP_FILTER) rtwvif->ip_addr = vif->cfg.arp_addr_list[0]; + + if (changed & BSS_CHANGED_MLD_VALID_LINKS) { + struct rtw89_vif_link *cur = rtw89_get_designated_link(rtwvif); + + rtw89_chip_rfk_channel(rtwdev, cur); + + if (hweight16(vif->active_links) == 1) + rtwvif->mlo_mode = RTW89_MLO_MODE_MLSR; + else + rtwvif->mlo_mode = RTW89_MLO_MODE_EMLSR; + } } static void rtw89_ops_link_info_changed(struct ieee80211_hw *hw, @@ -1531,10 +1542,29 @@ static bool rtw89_ops_can_activate_links(struct ieee80211_hw *hw, u16 active_links) { struct rtw89_dev *rtwdev = hw->priv; + struct rtw89_vif *rtwvif = vif_to_rtwvif(vif); + u16 current_links = vif->active_links; + struct rtw89_vif_ml_trans trans = { + .mediate_links = current_links | active_links, + .links_to_del = current_links & ~active_links, + .links_to_add = active_links & ~current_links, + }; lockdep_assert_wiphy(hw->wiphy); - return rtw89_can_work_on_links(rtwdev, vif, active_links); + if (!rtw89_can_work_on_links(rtwdev, vif, active_links)) + return false; + + /* + * Leave LPS at the beginning of ieee80211_set_active_links(). + * Because the entire process takes the same lock as our track + * work, LPS will not enter during ieee80211_set_active_links(). + */ + rtw89_leave_lps(rtwdev); + + rtwvif->ml_trans = trans; + + return true; } static void __rtw89_ops_clr_vif_links(struct rtw89_dev *rtwdev, @@ -1579,6 +1609,36 @@ static int __rtw89_ops_set_vif_links(struct rtw89_dev *rtwdev, return 0; } +static void rtw89_vif_cfg_fw_links(struct rtw89_dev *rtwdev, + struct rtw89_vif *rtwvif, + unsigned long links, bool en) +{ + struct rtw89_vif_link *rtwvif_link; + unsigned int link_id; + + for_each_set_bit(link_id, &links, IEEE80211_MLD_MAX_NUM_LINKS) { + rtwvif_link = rtwvif->links[link_id]; + if (unlikely(!rtwvif_link)) + continue; + + rtw89_fw_h2c_mlo_link_cfg(rtwdev, rtwvif_link, en); + } +} + +static void rtw89_vif_update_fw_links(struct rtw89_dev *rtwdev, + struct rtw89_vif *rtwvif, + u16 current_links) +{ + struct rtw89_vif_ml_trans *trans = &rtwvif->ml_trans; + + /* Do follow-up when all updating links exist. */ + if (current_links != trans->mediate_links) + return; + + rtw89_vif_cfg_fw_links(rtwdev, rtwvif, trans->links_to_del, false); + rtw89_vif_cfg_fw_links(rtwdev, rtwvif, trans->links_to_add, true); +} + static int rtw89_ops_change_vif_links(struct ieee80211_hw *hw, struct ieee80211_vif *vif, @@ -1620,6 +1680,8 @@ int rtw89_ops_change_vif_links(struct ieee80211_hw *hw, if (rtwdev->scanning) rtw89_hw_scan_abort(rtwdev, rtwdev->scan_info.scanning_vif); + rtw89_vif_update_fw_links(rtwdev, rtwvif, old_links); + if (!old_links) __rtw89_ops_clr_vif_links(rtwdev, rtwvif, BIT(RTW89_VIF_IDLE_LINK_ID)); From 438c9178cd8ab629025374ea598b060024a410e7 Mon Sep 17 00:00:00 2001 From: Zong-Zhe Yang Date: Tue, 21 Oct 2025 21:33:58 +0800 Subject: [PATCH 0288/1381] wifi: rtw89: support EHT rate pattern via bitrate mask When setting bitrate mask, e.g. using iw set bitrates, there are some designated patterns to be recognized, called rate pattern. When a rate pattern is matched, treat the setting as asking a fixed rate. Now, add support to recognize EHT rates. Signed-off-by: Zong-Zhe Yang Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251021133402.15467-5-pkshih@realtek.com --- drivers/net/wireless/realtek/rtw89/phy.c | 39 +++++++++++++++++++++++- 1 file changed, 38 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/realtek/rtw89/phy.c b/drivers/net/wireless/realtek/rtw89/phy.c index ba7feadd7582..23892c1359a5 100644 --- a/drivers/net/wireless/realtek/rtw89/phy.c +++ b/drivers/net/wireless/realtek/rtw89/phy.c @@ -231,7 +231,12 @@ static u64 rtw89_phy_ra_mask_cfg(struct rtw89_dev *rtwdev, return -1; } - if (link_sta->he_cap.has_he) { + if (link_sta->eht_cap.has_eht) { + cfg_mask |= u64_encode_bits(mask->control[band].eht_mcs[0], + RA_MASK_EHT_1SS_RATES); + cfg_mask |= u64_encode_bits(mask->control[band].eht_mcs[1], + RA_MASK_EHT_2SS_RATES); + } else if (link_sta->he_cap.has_he) { cfg_mask |= u64_encode_bits(mask->control[band].he_mcs[0], RA_MASK_HE_1SS_RATES); cfg_mask |= u64_encode_bits(mask->control[band].he_mcs[1], @@ -557,6 +562,14 @@ static bool __check_rate_pattern(struct rtw89_phy_rate_pattern *next, return true; } +enum __rtw89_hw_rate_invalid_bases { + /* no EHT rate for ax chip */ + RTW89_HW_RATE_EHT_NSS1_MCS0 = RTW89_HW_RATE_INVAL, + RTW89_HW_RATE_EHT_NSS2_MCS0 = RTW89_HW_RATE_INVAL, + RTW89_HW_RATE_EHT_NSS3_MCS0 = RTW89_HW_RATE_INVAL, + RTW89_HW_RATE_EHT_NSS4_MCS0 = RTW89_HW_RATE_INVAL, +}; + #define RTW89_HW_RATE_BY_CHIP_GEN(rate) \ { \ [RTW89_CHIP_AX] = RTW89_HW_RATE_ ## rate, \ @@ -572,6 +585,12 @@ void __rtw89_phy_rate_pattern_vif(struct rtw89_dev *rtwdev, struct rtw89_phy_rate_pattern next_pattern = {0}; const struct rtw89_chan *chan = rtw89_chan_get(rtwdev, rtwvif_link->chanctx_idx); + static const u16 hw_rate_eht[][RTW89_CHIP_GEN_NUM] = { + RTW89_HW_RATE_BY_CHIP_GEN(EHT_NSS1_MCS0), + RTW89_HW_RATE_BY_CHIP_GEN(EHT_NSS2_MCS0), + RTW89_HW_RATE_BY_CHIP_GEN(EHT_NSS3_MCS0), + RTW89_HW_RATE_BY_CHIP_GEN(EHT_NSS4_MCS0), + }; static const u16 hw_rate_he[][RTW89_CHIP_GEN_NUM] = { RTW89_HW_RATE_BY_CHIP_GEN(HE_NSS1_MCS0), RTW89_HW_RATE_BY_CHIP_GEN(HE_NSS2_MCS0), @@ -596,6 +615,17 @@ void __rtw89_phy_rate_pattern_vif(struct rtw89_dev *rtwdev, u8 tx_nss = rtwdev->hal.tx_nss; u8 i; + if (chip_gen == RTW89_CHIP_AX) + goto rs_11ax; + + for (i = 0; i < tx_nss; i++) + if (!__check_rate_pattern(&next_pattern, hw_rate_eht[i][chip_gen], + RA_MASK_EHT_RATES, RTW89_RA_MODE_EHT, + mask->control[nl_band].eht_mcs[i], + 0, true)) + goto out; + +rs_11ax: for (i = 0; i < tx_nss; i++) if (!__check_rate_pattern(&next_pattern, hw_rate_he[i][chip_gen], RA_MASK_HE_RATES, RTW89_RA_MODE_HE, @@ -640,6 +670,13 @@ void __rtw89_phy_rate_pattern_vif(struct rtw89_dev *rtwdev, if (!next_pattern.enable) goto out; + if (unlikely(next_pattern.rate >= RTW89_HW_RATE_INVAL)) { + rtw89_debug(rtwdev, RTW89_DBG_RA, + "pattern invalid target: chip_gen %d, mode 0x%x\n", + chip_gen, next_pattern.ra_mode); + goto out; + } + rtwvif_link->rate_pattern = next_pattern; rtw89_debug(rtwdev, RTW89_DBG_RA, "configure pattern: rate 0x%x, mask 0x%llx, mode 0x%x\n", From 0ac5ead00bf90f3280b8f5c0a0ad12e9800da8ec Mon Sep 17 00:00:00 2001 From: Zong-Zhe Yang Date: Tue, 21 Oct 2025 21:33:59 +0800 Subject: [PATCH 0289/1381] wifi: rtw89: regd: apply ACPI policy even if country code is programmed There are regulatory related policy according to BIOS/ACPI configuration, e.g. distro decides to disable some bands or some channels. Even if chip has a programmed country code, these policy should still be applied within regulatory notifier. Signed-off-by: Zong-Zhe Yang Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251021133402.15467-6-pkshih@realtek.com --- drivers/net/wireless/realtek/rtw89/core.h | 2 ++ drivers/net/wireless/realtek/rtw89/regd.c | 22 +++++++++++----------- 2 files changed, 13 insertions(+), 11 deletions(-) diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h index 7c74ed3cfaf7..ae98d6866b30 100644 --- a/drivers/net/wireless/realtek/rtw89/core.h +++ b/drivers/net/wireless/realtek/rtw89/core.h @@ -5449,6 +5449,8 @@ struct rtw89_regd_ctrl { struct rtw89_regulatory_info { struct rtw89_regd_ctrl ctrl; const struct rtw89_regd *regd; + bool programmed; + enum rtw89_reg_6ghz_power reg_6ghz_power; struct rtw89_reg_6ghz_tpe reg_6ghz_tpe; bool txpwr_uk_follow_etsi; diff --git a/drivers/net/wireless/realtek/rtw89/regd.c b/drivers/net/wireless/realtek/rtw89/regd.c index 58582f8d2b74..209d84909f88 100644 --- a/drivers/net/wireless/realtek/rtw89/regd.c +++ b/drivers/net/wireless/realtek/rtw89/regd.c @@ -723,6 +723,8 @@ int rtw89_regd_init_hint(struct rtw89_dev *rtwdev) chip_regd = rtw89_regd_find_reg_by_name(rtwdev, rtwdev->efuse.country_code); if (!rtw89_regd_is_ww(chip_regd)) { rtwdev->regulatory.regd = chip_regd; + rtwdev->regulatory.programmed = true; + /* Ignore country ie if there is a country domain programmed in chip */ wiphy->regulatory_flags |= REGULATORY_COUNTRY_IE_IGNORE; wiphy->regulatory_flags |= REGULATORY_STRICT_REG; @@ -867,11 +869,6 @@ static void rtw89_regd_notifier_apply(struct rtw89_dev *rtwdev, wiphy->regulatory_flags |= REGULATORY_COUNTRY_IE_IGNORE; else wiphy->regulatory_flags &= ~REGULATORY_COUNTRY_IE_IGNORE; - - rtw89_regd_apply_policy_unii4(rtwdev, wiphy); - rtw89_regd_apply_policy_6ghz(rtwdev, wiphy); - rtw89_regd_apply_policy_tas(rtwdev); - rtw89_regd_apply_policy_ant_gain(rtwdev); } static @@ -883,19 +880,22 @@ void rtw89_regd_notifier(struct wiphy *wiphy, struct regulatory_request *request wiphy_lock(wiphy); rtw89_leave_ps_mode(rtwdev); - if (wiphy->regd) { - rtw89_debug(rtwdev, RTW89_DBG_REGD, - "There is a country domain programmed in chip, ignore notifications\n"); - goto exit; - } + if (rtwdev->regulatory.programmed) + goto policy; + rtw89_regd_notifier_apply(rtwdev, wiphy, request); rtw89_debug_regd(rtwdev, rtwdev->regulatory.regd, "get from initiator %d, alpha2", request->initiator); +policy: + rtw89_regd_apply_policy_unii4(rtwdev, wiphy); + rtw89_regd_apply_policy_6ghz(rtwdev, wiphy); + rtw89_regd_apply_policy_tas(rtwdev); + rtw89_regd_apply_policy_ant_gain(rtwdev); + rtw89_core_set_chip_txpwr(rtwdev); -exit: wiphy_unlock(wiphy); } From a48ae54a67f7a3c05d51ca87877832d4d6ea4639 Mon Sep 17 00:00:00 2001 From: Kuan-Chung Chen Date: Tue, 21 Oct 2025 21:34:00 +0800 Subject: [PATCH 0290/1381] wifi: rtw89: 8852c: fix ADC oscillation in 160MHz affecting RX performance When operating in 160 MHz, the ADC may oscillate and affect AGC, leading to unstable RX quality. This issue can be resolved by ensuring proper RF filter bandwidth switching to avoid ADC oscillation. Signed-off-by: Kuan-Chung Chen Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251021133402.15467-7-pkshih@realtek.com --- .../net/wireless/realtek/rtw89/rtw8852c_rfk.c | 59 ++++++++++++------- 1 file changed, 39 insertions(+), 20 deletions(-) diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852c_rfk.c b/drivers/net/wireless/realtek/rtw89/rtw8852c_rfk.c index e7fd028c5e82..cbee484dee30 100644 --- a/drivers/net/wireless/realtek/rtw89/rtw8852c_rfk.c +++ b/drivers/net/wireless/realtek/rtw89/rtw8852c_rfk.c @@ -3987,37 +3987,56 @@ static void _ctrl_ch(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy, } } +static void _set_rxbb_bw(struct rtw89_dev *rtwdev, enum rtw89_rf_path path, + enum rtw89_bandwidth bw) +{ + u32 val; + + rtw89_write_rf(rtwdev, path, RR_LUTWE2, RR_LUTWE2_RTXBW, 0x1); + rtw89_write_rf(rtwdev, path, RR_LUTWA, RR_LUTWA_M2, 0xa); + + switch (bw) { + case RTW89_CHANNEL_WIDTH_20: + val = 0x1b; + break; + case RTW89_CHANNEL_WIDTH_40: + val = 0x13; + break; + case RTW89_CHANNEL_WIDTH_80: + val = 0xb; + break; + case RTW89_CHANNEL_WIDTH_160: + default: + val = 0x3; + break; + } + + rtw89_write_rf(rtwdev, path, RR_LUTWD0, RR_LUTWD0_LB, val); + rtw89_write_rf(rtwdev, path, RR_LUTWE2, RR_LUTWE2_RTXBW, 0x0); +} + +static void _set_tia_bw(struct rtw89_dev *rtwdev, enum rtw89_rf_path path, + enum rtw89_bandwidth bw) +{ + if (bw == RTW89_CHANNEL_WIDTH_160) + rtw89_write_rf(rtwdev, path, RR_RXBB2, RR_RXBB2_EBW, 0x0); + else + rtw89_write_rf(rtwdev, path, RR_RXBB2, RR_RXBB2_EBW, 0x2); +} + static void _rxbb_bw(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy, enum rtw89_bandwidth bw) { u8 kpath; u8 path; - u32 val; kpath = _kpath(rtwdev, phy); for (path = 0; path < 2; path++) { if (!(kpath & BIT(path))) continue; - rtw89_write_rf(rtwdev, path, RR_LUTWE2, RR_LUTWE2_RTXBW, 0x1); - rtw89_write_rf(rtwdev, path, RR_LUTWA, RR_LUTWA_M2, 0xa); - switch (bw) { - case RTW89_CHANNEL_WIDTH_20: - val = 0x1b; - break; - case RTW89_CHANNEL_WIDTH_40: - val = 0x13; - break; - case RTW89_CHANNEL_WIDTH_80: - val = 0xb; - break; - case RTW89_CHANNEL_WIDTH_160: - default: - val = 0x3; - break; - } - rtw89_write_rf(rtwdev, path, RR_LUTWD0, RR_LUTWD0_LB, val); - rtw89_write_rf(rtwdev, path, RR_LUTWE2, RR_LUTWE2_RTXBW, 0x0); + _set_rxbb_bw(rtwdev, path, bw); + _set_tia_bw(rtwdev, path, bw); } } From e139b1c1f01a8874df02894b3aedf0b64f5b839b Mon Sep 17 00:00:00 2001 From: Chin-Yen Lee Date: Tue, 21 Oct 2025 21:34:01 +0800 Subject: [PATCH 0291/1381] wifi: rtw89: restart hardware to recover firmware if power-save becomes abnormal Somehow power-save related functions get failure, such as failed to send null packet, or no response form firmware, and then WiFi will become unstable. Trigger SER function actively to reset firmware/driver to recover from abnormal states, including - firmware failed to ACK for entering PS mode - firmware failed to ACK for leaving PS mode - check PS H2C command received by firmware fail - failed to leave PS state Signed-off-by: Chin-Yen Lee Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251021133402.15467-8-pkshih@realtek.com --- drivers/net/wireless/realtek/rtw89/core.c | 1 + drivers/net/wireless/realtek/rtw89/core.h | 2 ++ drivers/net/wireless/realtek/rtw89/mac.c | 7 +++++-- drivers/net/wireless/realtek/rtw89/ps.c | 23 +++++++++++++++++++++-- 4 files changed, 29 insertions(+), 4 deletions(-) diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c index a3f156bf708e..1b5a40e9821c 100644 --- a/drivers/net/wireless/realtek/rtw89/core.c +++ b/drivers/net/wireless/realtek/rtw89/core.c @@ -5538,6 +5538,7 @@ int rtw89_core_start(struct rtw89_dev *rtwdev) rtw89_fw_h2c_fw_log(rtwdev, rtwdev->fw.log.enable); rtw89_fw_h2c_init_ba_cam(rtwdev); rtw89_tas_fw_timer_enable(rtwdev, true); + rtwdev->ps_hang_cnt = 0; return 0; } diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h index ae98d6866b30..f8b443894db9 100644 --- a/drivers/net/wireless/realtek/rtw89/core.h +++ b/drivers/net/wireless/realtek/rtw89/core.h @@ -42,6 +42,7 @@ extern const struct ieee80211_ops rtw89_ops; #define RTW89_TRACK_WORK_PERIOD round_jiffies_relative(HZ * 2) #define RTW89_TRACK_PS_WORK_PERIOD msecs_to_jiffies(100) #define RTW89_FORBID_BA_TIMER round_jiffies_relative(HZ * 4) +#define RTW89_PS_HANG_MAX_CNT 3 #define CFO_TRACK_MAX_USER 64 #define MAX_RSSI 110 #define RSSI_FACTOR 1 @@ -6082,6 +6083,7 @@ struct rtw89_dev { struct rtw89_btc btc; enum rtw89_ps_mode ps_mode; bool lps_enabled; + u8 ps_hang_cnt; struct rtw89_wow_param wow; diff --git a/drivers/net/wireless/realtek/rtw89/mac.c b/drivers/net/wireless/realtek/rtw89/mac.c index fd11b8fb3c89..d837513f4e92 100644 --- a/drivers/net/wireless/realtek/rtw89/mac.c +++ b/drivers/net/wireless/realtek/rtw89/mac.c @@ -12,6 +12,7 @@ #include "phy.h" #include "ps.h" #include "reg.h" +#include "ser.h" #include "util.h" static const u32 rtw89_mac_mem_base_addrs_ax[RTW89_MAC_MEM_NUM] = { @@ -1423,13 +1424,15 @@ void rtw89_mac_power_mode_change(struct rtw89_dev *rtwdev, bool enter) if (!ret) break; - if (i == RPWM_TRY_CNT - 1) + if (i == RPWM_TRY_CNT - 1) { rtw89_err(rtwdev, "firmware failed to ack for %s ps mode\n", enter ? "entering" : "leaving"); - else + rtw89_ser_notify(rtwdev, MAC_AX_ERR_ASSERTION); + } else { rtw89_debug(rtwdev, RTW89_DBG_UNEXP, "%d time firmware failed to ack for %s ps mode\n", i + 1, enter ? "entering" : "leaving"); + } } } diff --git a/drivers/net/wireless/realtek/rtw89/ps.c b/drivers/net/wireless/realtek/rtw89/ps.c index cf58121eb541..3f69dd4361c3 100644 --- a/drivers/net/wireless/realtek/rtw89/ps.c +++ b/drivers/net/wireless/realtek/rtw89/ps.c @@ -11,6 +11,7 @@ #include "phy.h" #include "ps.h" #include "reg.h" +#include "ser.h" #include "util.h" static int rtw89_fw_receive_lps_h2c_check(struct rtw89_dev *rtwdev, u8 macid) @@ -26,16 +27,27 @@ static int rtw89_fw_receive_lps_h2c_check(struct rtw89_dev *rtwdev, u8 macid) c2h_info.id = RTW89_FWCMD_C2HREG_FUNC_PS_LEAVE_ACK; ret = rtw89_fw_msg_reg(rtwdev, NULL, &c2h_info); if (ret) - return ret; + goto fw_fail; c2hreg_macid = u32_get_bits(c2h_info.u.c2hreg[0], RTW89_C2HREG_PS_LEAVE_ACK_MACID); c2hreg_ret = u32_get_bits(c2h_info.u.c2hreg[1], RTW89_C2HREG_PS_LEAVE_ACK_RET); - if (macid != c2hreg_macid || c2hreg_ret) + if (macid != c2hreg_macid || c2hreg_ret) { rtw89_warn(rtwdev, "rtw89: check lps h2c received by firmware fail\n"); + ret = -EINVAL; + goto fw_fail; + } + rtwdev->ps_hang_cnt = 0; return 0; + +fw_fail: + rtwdev->ps_hang_cnt++; + if (rtwdev->ps_hang_cnt >= RTW89_PS_HANG_MAX_CNT) + rtw89_ser_notify(rtwdev, MAC_AX_ERR_ASSERTION); + + return ret; } static int rtw89_fw_leave_lps_check(struct rtw89_dev *rtwdev, u8 macid) @@ -51,9 +63,16 @@ static int rtw89_fw_leave_lps_check(struct rtw89_dev *rtwdev, u8 macid) mac->ps_status, chk_msk); if (ret) { rtw89_info(rtwdev, "rtw89: failed to leave lps state\n"); + + rtwdev->ps_hang_cnt++; + if (rtwdev->ps_hang_cnt >= RTW89_PS_HANG_MAX_CNT) + rtw89_ser_notify(rtwdev, MAC_AX_ERR_ASSERTION); + return -EBUSY; } + rtwdev->ps_hang_cnt = 0; + return 0; } From a62b654125068a69b7c459ba59526ebd9882f405 Mon Sep 17 00:00:00 2001 From: Jui-Peng Tsai Date: Tue, 21 Oct 2025 21:34:02 +0800 Subject: [PATCH 0292/1381] wifi: rtw89: improve scan time on 6 GHz band Reduce scan time for all supported channels from 4.4s to 3.5s. If NL80211_SCAN_FLAG_COLOCATED_6GHZ is set in scan request, only scan PSC channels and the channels from the RNR element found on the 2.4/5 GHz channels. When firmware support parsing RNR element from received beacon or probe response, offload the decision about non-PSC channels to firmware. Driver do not need to fill non-PSC channels to scan list. If NL80211_SCAN_FLAG_COLOCATED_6GHZ is not set, scan all supported channels. Signed-off-by: Jui-Peng Tsai Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251021133402.15467-9-pkshih@realtek.com --- drivers/net/wireless/realtek/rtw89/fw.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/net/wireless/realtek/rtw89/fw.c b/drivers/net/wireless/realtek/rtw89/fw.c index 9a926bb2cf00..cb431c8a65ac 100644 --- a/drivers/net/wireless/realtek/rtw89/fw.c +++ b/drivers/net/wireless/realtek/rtw89/fw.c @@ -7787,15 +7787,23 @@ int rtw89_hw_scan_prep_chan_list_be(struct rtw89_dev *rtwdev, struct ieee80211_channel *channel; struct list_head chan_list; enum rtw89_chan_type type; + bool chan_by_rnr; bool random_seq; int ret; u32 idx; random_seq = !!(req->flags & NL80211_SCAN_FLAG_RANDOM_SN); + chan_by_rnr = rtwdev->chip->support_rnr && + (req->flags & NL80211_SCAN_FLAG_COLOCATED_6GHZ); INIT_LIST_HEAD(&chan_list); for (idx = 0; idx < req->n_channels; idx++) { channel = req->channels[idx]; + + if (channel->band == NL80211_BAND_6GHZ && + !cfg80211_channel_is_psc(channel) && chan_by_rnr) + continue; + ch_info = kzalloc(sizeof(*ch_info), GFP_KERNEL); if (!ch_info) { ret = -ENOMEM; From a71e367773482a78566abd862dfee9cc3bb9332e Mon Sep 17 00:00:00 2001 From: Jiawen Wu Date: Thu, 23 Oct 2025 09:45:36 +0800 Subject: [PATCH 0293/1381] net: txgbe: support RX desc merge mode RX descriptor merge mode is supported on AML devices. When it is enabled, the hardware process the RX descriptors in batches. Signed-off-by: Jiawen Wu Reviewed-by: Jacob Keller Link: https://patch.msgid.link/20251023014538.12644-2-jiawenwu@trustnetic.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/wangxun/libwx/wx_hw.c | 10 ++++++++++ drivers/net/ethernet/wangxun/libwx/wx_type.h | 7 +++++++ drivers/net/ethernet/wangxun/libwx/wx_vf.h | 1 + drivers/net/ethernet/wangxun/libwx/wx_vf_lib.c | 3 +++ drivers/net/ethernet/wangxun/txgbe/txgbe_main.c | 1 + drivers/net/ethernet/wangxun/txgbevf/txgbevf_main.c | 11 +++++++++++ 6 files changed, 33 insertions(+) diff --git a/drivers/net/ethernet/wangxun/libwx/wx_hw.c b/drivers/net/ethernet/wangxun/libwx/wx_hw.c index 1e2713f0c921..2dbbb42aa9c0 100644 --- a/drivers/net/ethernet/wangxun/libwx/wx_hw.c +++ b/drivers/net/ethernet/wangxun/libwx/wx_hw.c @@ -1935,6 +1935,10 @@ static void wx_configure_rx_ring(struct wx *wx, rxdctl |= (ring->count / 128) << WX_PX_RR_CFG_RR_SIZE_SHIFT; rxdctl |= 0x1 << WX_PX_RR_CFG_RR_THER_SHIFT; + + if (test_bit(WX_FLAG_RX_MERGE_ENABLED, wx->flags)) + rxdctl |= WX_PX_RR_CFG_DESC_MERGE; + wr32(wx, WX_PX_RR_CFG(reg_idx), rxdctl); /* reset head and tail pointers */ @@ -2190,6 +2194,12 @@ void wx_configure_rx(struct wx *wx) /* set_rx_buffer_len must be called before ring initialization */ wx_set_rx_buffer_len(wx); + if (test_bit(WX_FLAG_RX_MERGE_ENABLED, wx->flags)) { + wr32(wx, WX_RDM_DCACHE_CTL, WX_RDM_DCACHE_CTL_EN); + wr32m(wx, WX_RDM_RSC_CTL, + WX_RDM_RSC_CTL_FREE_CTL | WX_RDM_RSC_CTL_FREE_CNT_DIS, + WX_RDM_RSC_CTL_FREE_CTL); + } /* Setup the HW Rx Head and Tail Descriptor Pointers and * the Base and Length of the Rx Descriptor Ring */ diff --git a/drivers/net/ethernet/wangxun/libwx/wx_type.h b/drivers/net/ethernet/wangxun/libwx/wx_type.h index 4880268b620e..eb3f32551c14 100644 --- a/drivers/net/ethernet/wangxun/libwx/wx_type.h +++ b/drivers/net/ethernet/wangxun/libwx/wx_type.h @@ -83,8 +83,13 @@ /*********************** Receive DMA registers **************************/ #define WX_RDM_VF_RE(_i) (0x12004 + ((_i) * 4)) +#define WX_RDM_RSC_CTL 0x1200C +#define WX_RDM_RSC_CTL_FREE_CNT_DIS BIT(8) +#define WX_RDM_RSC_CTL_FREE_CTL BIT(7) #define WX_RDM_PF_QDE(_i) (0x12080 + ((_i) * 4)) #define WX_RDM_VFRE_CLR(_i) (0x120A0 + ((_i) * 4)) +#define WX_RDM_DCACHE_CTL 0x120A8 +#define WX_RDM_DCACHE_CTL_EN BIT(0) #define WX_RDM_DRP_PKT 0x12500 #define WX_RDM_PKT_CNT 0x12504 #define WX_RDM_BYTE_CNT_LSB 0x12508 @@ -447,6 +452,7 @@ enum WX_MSCA_CMD_value { #define WX_PX_RR_CFG_VLAN BIT(31) #define WX_PX_RR_CFG_DROP_EN BIT(30) #define WX_PX_RR_CFG_SPLIT_MODE BIT(26) +#define WX_PX_RR_CFG_DESC_MERGE BIT(19) #define WX_PX_RR_CFG_RR_THER_SHIFT 16 #define WX_PX_RR_CFG_RR_HDR_SZ GENMASK(15, 12) #define WX_PX_RR_CFG_RR_BUF_SZ GENMASK(11, 8) @@ -1232,6 +1238,7 @@ enum wx_pf_flags { WX_FLAG_NEED_SFP_RESET, WX_FLAG_NEED_UPDATE_LINK, WX_FLAG_NEED_DO_RESET, + WX_FLAG_RX_MERGE_ENABLED, WX_PF_FLAGS_NBITS /* must be last */ }; diff --git a/drivers/net/ethernet/wangxun/libwx/wx_vf.h b/drivers/net/ethernet/wangxun/libwx/wx_vf.h index 3f16de0fa427..ecb198592393 100644 --- a/drivers/net/ethernet/wangxun/libwx/wx_vf.h +++ b/drivers/net/ethernet/wangxun/libwx/wx_vf.h @@ -74,6 +74,7 @@ #define WX_VXRXDCTL_BUFSZ(f) FIELD_PREP(GENMASK(11, 8), f) #define WX_VXRXDCTL_HDRSZ_MASK GENMASK(15, 12) #define WX_VXRXDCTL_HDRSZ(f) FIELD_PREP(GENMASK(15, 12), f) +#define WX_VXRXDCTL_DESC_MERGE BIT(19) #define WX_VXRXDCTL_RSCMAX_MASK GENMASK(24, 23) #define WX_VXRXDCTL_RSCMAX(f) FIELD_PREP(GENMASK(24, 23), f) #define WX_VXRXDCTL_RSCEN BIT(29) diff --git a/drivers/net/ethernet/wangxun/libwx/wx_vf_lib.c b/drivers/net/ethernet/wangxun/libwx/wx_vf_lib.c index a87887b9f8ee..f54107f3c6d7 100644 --- a/drivers/net/ethernet/wangxun/libwx/wx_vf_lib.c +++ b/drivers/net/ethernet/wangxun/libwx/wx_vf_lib.c @@ -272,6 +272,9 @@ void wx_configure_rx_ring_vf(struct wx *wx, struct wx_ring *ring) rxdctl |= WX_VXRXDCTL_RSCMAX(0); rxdctl |= WX_VXRXDCTL_RSCEN; + if (test_bit(WX_FLAG_RX_MERGE_ENABLED, wx->flags)) + rxdctl |= WX_VXRXDCTL_DESC_MERGE; + wr32(wx, WX_VXRXDCTL(reg_idx), rxdctl); /* pf/vf reuse */ diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c b/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c index c4c4d70d8466..60a04c5a7678 100644 --- a/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c +++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c @@ -423,6 +423,7 @@ static int txgbe_sw_init(struct wx *wx) break; case wx_mac_aml: case wx_mac_aml40: + set_bit(WX_FLAG_RX_MERGE_ENABLED, wx->flags); set_bit(WX_FLAG_SWFW_RING, wx->flags); wx->swfw_index = 0; break; diff --git a/drivers/net/ethernet/wangxun/txgbevf/txgbevf_main.c b/drivers/net/ethernet/wangxun/txgbevf/txgbevf_main.c index 72663e3c4205..52c1e223bbd7 100644 --- a/drivers/net/ethernet/wangxun/txgbevf/txgbevf_main.c +++ b/drivers/net/ethernet/wangxun/txgbevf/txgbevf_main.c @@ -157,6 +157,17 @@ static int txgbevf_sw_init(struct wx *wx) wx->set_num_queues = txgbevf_set_num_queues; + switch (wx->mac.type) { + case wx_mac_sp: + break; + case wx_mac_aml: + case wx_mac_aml40: + set_bit(WX_FLAG_RX_MERGE_ENABLED, wx->flags); + break; + default: + break; + } + return 0; err_reset_hw: kfree(wx->vfinfo); From eb57b16d90d3ec9eee17b0e66026507cd09a2734 Mon Sep 17 00:00:00 2001 From: Jiawen Wu Date: Thu, 23 Oct 2025 09:45:37 +0800 Subject: [PATCH 0294/1381] net: txgbe: support TX head write-back mode TX head write-back mode is supported on AML devices. When it is enabled, the hardware no longer writes the descriptors DD one by one, but write back pointer of completion descriptor to the head_wb address. Signed-off-by: Jiawen Wu Reviewed-by: Jacob Keller Link: https://patch.msgid.link/20251023014538.12644-3-jiawenwu@trustnetic.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/wangxun/libwx/wx_hw.c | 9 ++++ drivers/net/ethernet/wangxun/libwx/wx_lib.c | 53 ++++++++++++++++++- drivers/net/ethernet/wangxun/libwx/wx_type.h | 7 +++ drivers/net/ethernet/wangxun/libwx/wx_vf.h | 3 ++ .../net/ethernet/wangxun/libwx/wx_vf_lib.c | 9 ++++ .../net/ethernet/wangxun/txgbe/txgbe_main.c | 1 + .../ethernet/wangxun/txgbevf/txgbevf_main.c | 1 + 7 files changed, 81 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/wangxun/libwx/wx_hw.c b/drivers/net/ethernet/wangxun/libwx/wx_hw.c index 2dbbb42aa9c0..986bc5acc472 100644 --- a/drivers/net/ethernet/wangxun/libwx/wx_hw.c +++ b/drivers/net/ethernet/wangxun/libwx/wx_hw.c @@ -1905,6 +1905,15 @@ static void wx_configure_tx_ring(struct wx *wx, memset(ring->tx_buffer_info, 0, sizeof(struct wx_tx_buffer) * ring->count); + if (ring->headwb_mem) { + wr32(wx, WX_PX_TR_HEAD_ADDRL(reg_idx), + ring->headwb_dma & DMA_BIT_MASK(32)); + wr32(wx, WX_PX_TR_HEAD_ADDRH(reg_idx), + upper_32_bits(ring->headwb_dma)); + + txdctl |= WX_PX_TR_CFG_HEAD_WB; + } + /* enable queue */ wr32(wx, WX_PX_TR_CFG(reg_idx), txdctl); diff --git a/drivers/net/ethernet/wangxun/libwx/wx_lib.c b/drivers/net/ethernet/wangxun/libwx/wx_lib.c index 3adf7048320a..622213fe6790 100644 --- a/drivers/net/ethernet/wangxun/libwx/wx_lib.c +++ b/drivers/net/ethernet/wangxun/libwx/wx_lib.c @@ -735,9 +735,22 @@ static bool wx_clean_tx_irq(struct wx_q_vector *q_vector, /* prevent any other reads prior to eop_desc */ smp_rmb(); - /* if DD is not set pending work has not been completed */ - if (!(eop_desc->wb.status & cpu_to_le32(WX_TXD_STAT_DD))) + if (tx_ring->headwb_mem) { + u32 head = *tx_ring->headwb_mem; + + if (head == tx_ring->next_to_clean) + break; + else if (head > tx_ring->next_to_clean && + !(tx_buffer->next_eop >= tx_ring->next_to_clean && + tx_buffer->next_eop < head)) + break; + else if (!(tx_buffer->next_eop >= tx_ring->next_to_clean || + tx_buffer->next_eop < head)) + break; + } else if (!(eop_desc->wb.status & cpu_to_le32(WX_TXD_STAT_DD))) { + /* if DD is not set pending work has not been completed */ break; + } /* clear next_to_watch to prevent false hangs */ tx_buffer->next_to_watch = NULL; @@ -1075,6 +1088,10 @@ static int wx_tx_map(struct wx_ring *tx_ring, /* set next_to_watch value indicating a packet is present */ first->next_to_watch = tx_desc; + /* set next_eop for amlite tx head wb */ + if (tx_ring->headwb_mem) + first->next_eop = i; + i++; if (i == tx_ring->count) i = 0; @@ -2683,6 +2700,16 @@ void wx_clean_all_tx_rings(struct wx *wx) } EXPORT_SYMBOL(wx_clean_all_tx_rings); +static void wx_free_headwb_resources(struct wx_ring *tx_ring) +{ + if (!tx_ring->headwb_mem) + return; + + dma_free_coherent(tx_ring->dev, sizeof(u32), + tx_ring->headwb_mem, tx_ring->headwb_dma); + tx_ring->headwb_mem = NULL; +} + /** * wx_free_tx_resources - Free Tx Resources per Queue * @tx_ring: Tx descriptor ring for a specific queue @@ -2702,6 +2729,8 @@ static void wx_free_tx_resources(struct wx_ring *tx_ring) dma_free_coherent(tx_ring->dev, tx_ring->size, tx_ring->desc, tx_ring->dma); tx_ring->desc = NULL; + + wx_free_headwb_resources(tx_ring); } /** @@ -2840,6 +2869,24 @@ static int wx_setup_all_rx_resources(struct wx *wx) return err; } +static void wx_setup_headwb_resources(struct wx_ring *tx_ring) +{ + struct wx *wx = netdev_priv(tx_ring->netdev); + + if (!test_bit(WX_FLAG_TXHEAD_WB_ENABLED, wx->flags)) + return; + + if (!tx_ring->q_vector) + return; + + tx_ring->headwb_mem = dma_alloc_coherent(tx_ring->dev, + sizeof(u32), + &tx_ring->headwb_dma, + GFP_KERNEL); + if (!tx_ring->headwb_mem) + dev_info(tx_ring->dev, "Allocate headwb memory failed, disable it\n"); +} + /** * wx_setup_tx_resources - allocate Tx resources (Descriptors) * @tx_ring: tx descriptor ring (for a specific queue) to setup @@ -2880,6 +2927,8 @@ static int wx_setup_tx_resources(struct wx_ring *tx_ring) if (!tx_ring->desc) goto err; + wx_setup_headwb_resources(tx_ring); + tx_ring->next_to_use = 0; tx_ring->next_to_clean = 0; diff --git a/drivers/net/ethernet/wangxun/libwx/wx_type.h b/drivers/net/ethernet/wangxun/libwx/wx_type.h index eb3f32551c14..8b3c39945c0b 100644 --- a/drivers/net/ethernet/wangxun/libwx/wx_type.h +++ b/drivers/net/ethernet/wangxun/libwx/wx_type.h @@ -434,12 +434,15 @@ enum WX_MSCA_CMD_value { #define WX_PX_TR_WP(_i) (0x03008 + ((_i) * 0x40)) #define WX_PX_TR_RP(_i) (0x0300C + ((_i) * 0x40)) #define WX_PX_TR_CFG(_i) (0x03010 + ((_i) * 0x40)) +#define WX_PX_TR_HEAD_ADDRL(_i) (0x03028 + ((_i) * 0x40)) +#define WX_PX_TR_HEAD_ADDRH(_i) (0x0302C + ((_i) * 0x40)) /* Transmit Config masks */ #define WX_PX_TR_CFG_ENABLE BIT(0) /* Ena specific Tx Queue */ #define WX_PX_TR_CFG_TR_SIZE_SHIFT 1 /* tx desc number per ring */ #define WX_PX_TR_CFG_SWFLSH BIT(26) /* Tx Desc. wr-bk flushing */ #define WX_PX_TR_CFG_WTHRESH_SHIFT 16 /* shift to WTHRESH bits */ #define WX_PX_TR_CFG_THRE_SHIFT 8 +#define WX_PX_TR_CFG_HEAD_WB BIT(27) /* Receive DMA Registers */ #define WX_PX_RR_BAL(_i) (0x01000 + ((_i) * 0x40)) @@ -1011,6 +1014,7 @@ struct wx_tx_buffer { DEFINE_DMA_UNMAP_LEN(len); __be16 protocol; u32 tx_flags; + u32 next_eop; }; struct wx_rx_buffer { @@ -1062,6 +1066,8 @@ struct wx_ring { }; u8 __iomem *tail; dma_addr_t dma; /* phys. address of descriptor ring */ + dma_addr_t headwb_dma; + u32 *headwb_mem; unsigned int size; /* length in bytes */ u16 count; /* amount of descriptors */ @@ -1239,6 +1245,7 @@ enum wx_pf_flags { WX_FLAG_NEED_UPDATE_LINK, WX_FLAG_NEED_DO_RESET, WX_FLAG_RX_MERGE_ENABLED, + WX_FLAG_TXHEAD_WB_ENABLED, WX_PF_FLAGS_NBITS /* must be last */ }; diff --git a/drivers/net/ethernet/wangxun/libwx/wx_vf.h b/drivers/net/ethernet/wangxun/libwx/wx_vf.h index ecb198592393..eb6ca3fe4e97 100644 --- a/drivers/net/ethernet/wangxun/libwx/wx_vf.h +++ b/drivers/net/ethernet/wangxun/libwx/wx_vf.h @@ -92,6 +92,9 @@ #define WX_VXTXDCTL_PTHRESH(f) FIELD_PREP(GENMASK(11, 8), f) #define WX_VXTXDCTL_WTHRESH(f) FIELD_PREP(GENMASK(22, 16), f) #define WX_VXTXDCTL_FLUSH BIT(26) +#define WX_VXTXDCTL_HEAD_WB BIT(27) +#define WX_VXTXD_HEAD_ADDRL(r) (0x3028 + (0x40 * (r))) +#define WX_VXTXD_HEAD_ADDRH(r) (0x302C + (0x40 * (r))) #define WX_PFLINK_STATUS(g) FIELD_GET(BIT(0), g) #define WX_PFLINK_SPEED(g) FIELD_GET(GENMASK(31, 1), g) diff --git a/drivers/net/ethernet/wangxun/libwx/wx_vf_lib.c b/drivers/net/ethernet/wangxun/libwx/wx_vf_lib.c index f54107f3c6d7..aa8be036956c 100644 --- a/drivers/net/ethernet/wangxun/libwx/wx_vf_lib.c +++ b/drivers/net/ethernet/wangxun/libwx/wx_vf_lib.c @@ -132,6 +132,15 @@ static void wx_configure_tx_ring_vf(struct wx *wx, struct wx_ring *ring) txdctl |= WX_VXTXDCTL_BUFLEN(wx_buf_len(ring->count)); txdctl |= WX_VXTXDCTL_ENABLE; + if (ring->headwb_mem) { + wr32(wx, WX_VXTXD_HEAD_ADDRL(reg_idx), + ring->headwb_dma & DMA_BIT_MASK(32)); + wr32(wx, WX_VXTXD_HEAD_ADDRH(reg_idx), + upper_32_bits(ring->headwb_dma)); + + txdctl |= WX_VXTXDCTL_HEAD_WB; + } + /* reinitialize tx_buffer_info */ memset(ring->tx_buffer_info, 0, sizeof(struct wx_tx_buffer) * ring->count); diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c b/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c index 60a04c5a7678..ff690e9a075a 100644 --- a/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c +++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c @@ -424,6 +424,7 @@ static int txgbe_sw_init(struct wx *wx) case wx_mac_aml: case wx_mac_aml40: set_bit(WX_FLAG_RX_MERGE_ENABLED, wx->flags); + set_bit(WX_FLAG_TXHEAD_WB_ENABLED, wx->flags); set_bit(WX_FLAG_SWFW_RING, wx->flags); wx->swfw_index = 0; break; diff --git a/drivers/net/ethernet/wangxun/txgbevf/txgbevf_main.c b/drivers/net/ethernet/wangxun/txgbevf/txgbevf_main.c index 52c1e223bbd7..37e4ec487afd 100644 --- a/drivers/net/ethernet/wangxun/txgbevf/txgbevf_main.c +++ b/drivers/net/ethernet/wangxun/txgbevf/txgbevf_main.c @@ -163,6 +163,7 @@ static int txgbevf_sw_init(struct wx *wx) case wx_mac_aml: case wx_mac_aml40: set_bit(WX_FLAG_RX_MERGE_ENABLED, wx->flags); + set_bit(WX_FLAG_TXHEAD_WB_ENABLED, wx->flags); break; default: break; From eaed17770637af6d35d9b5465d91f1256a5eaacf Mon Sep 17 00:00:00 2001 From: Jiawen Wu Date: Thu, 23 Oct 2025 09:45:38 +0800 Subject: [PATCH 0295/1381] net: txgbe: support RSC offload Support to enable and disable RSC for txgbe devices. Signed-off-by: Jiawen Wu Reviewed-by: Jacob Keller Link: https://patch.msgid.link/20251023014538.12644-4-jiawenwu@trustnetic.com Signed-off-by: Paolo Abeni --- .../net/ethernet/wangxun/libwx/wx_ethtool.c | 61 ++++++++++++- drivers/net/ethernet/wangxun/libwx/wx_hw.c | 50 ++++++++++- drivers/net/ethernet/wangxun/libwx/wx_lib.c | 90 +++++++++++++++++-- drivers/net/ethernet/wangxun/libwx/wx_sriov.c | 4 + drivers/net/ethernet/wangxun/libwx/wx_type.h | 33 +++++-- .../net/ethernet/wangxun/txgbe/txgbe_main.c | 3 + 6 files changed, 224 insertions(+), 17 deletions(-) diff --git a/drivers/net/ethernet/wangxun/libwx/wx_ethtool.c b/drivers/net/ethernet/wangxun/libwx/wx_ethtool.c index 06f401bd975c..9aa3964187e1 100644 --- a/drivers/net/ethernet/wangxun/libwx/wx_ethtool.c +++ b/drivers/net/ethernet/wangxun/libwx/wx_ethtool.c @@ -51,6 +51,11 @@ static const struct wx_stats wx_gstrings_fdir_stats[] = { WX_STAT("fdir_miss", stats.fdirmiss), }; +static const struct wx_stats wx_gstrings_rsc_stats[] = { + WX_STAT("rsc_aggregated", rsc_count), + WX_STAT("rsc_flushed", rsc_flush), +}; + /* drivers allocates num_tx_queues and num_rx_queues symmetrically so * we set the num_rx_queues to evaluate to num_tx_queues. This is * used because we do not have a good way to get the max number of @@ -64,16 +69,21 @@ static const struct wx_stats wx_gstrings_fdir_stats[] = { (sizeof(struct wx_queue_stats) / sizeof(u64))) #define WX_GLOBAL_STATS_LEN ARRAY_SIZE(wx_gstrings_stats) #define WX_FDIR_STATS_LEN ARRAY_SIZE(wx_gstrings_fdir_stats) +#define WX_RSC_STATS_LEN ARRAY_SIZE(wx_gstrings_rsc_stats) #define WX_STATS_LEN (WX_GLOBAL_STATS_LEN + WX_QUEUE_STATS_LEN) int wx_get_sset_count(struct net_device *netdev, int sset) { struct wx *wx = netdev_priv(netdev); + int len = WX_STATS_LEN; switch (sset) { case ETH_SS_STATS: - return (test_bit(WX_FLAG_FDIR_CAPABLE, wx->flags)) ? - WX_STATS_LEN + WX_FDIR_STATS_LEN : WX_STATS_LEN; + if (test_bit(WX_FLAG_FDIR_CAPABLE, wx->flags)) + len += WX_FDIR_STATS_LEN; + if (test_bit(WX_FLAG_RSC_CAPABLE, wx->flags)) + len += WX_RSC_STATS_LEN; + return len; default: return -EOPNOTSUPP; } @@ -94,6 +104,10 @@ void wx_get_strings(struct net_device *netdev, u32 stringset, u8 *data) for (i = 0; i < WX_FDIR_STATS_LEN; i++) ethtool_puts(&p, wx_gstrings_fdir_stats[i].stat_string); } + if (test_bit(WX_FLAG_RSC_CAPABLE, wx->flags)) { + for (i = 0; i < WX_RSC_STATS_LEN; i++) + ethtool_puts(&p, wx_gstrings_rsc_stats[i].stat_string); + } for (i = 0; i < netdev->num_tx_queues; i++) { ethtool_sprintf(&p, "tx_queue_%u_packets", i); ethtool_sprintf(&p, "tx_queue_%u_bytes", i); @@ -131,6 +145,13 @@ void wx_get_ethtool_stats(struct net_device *netdev, } } + if (test_bit(WX_FLAG_RSC_CAPABLE, wx->flags)) { + for (k = 0; k < WX_RSC_STATS_LEN; k++) { + p = (char *)wx + wx_gstrings_rsc_stats[k].stat_offset; + data[i++] = *(u64 *)p; + } + } + for (j = 0; j < netdev->num_tx_queues; j++) { ring = wx->tx_ring[j]; if (!ring) { @@ -322,6 +343,40 @@ int wx_get_coalesce(struct net_device *netdev, } EXPORT_SYMBOL(wx_get_coalesce); +static void wx_update_rsc(struct wx *wx) +{ + struct net_device *netdev = wx->netdev; + bool need_reset = false; + + /* nothing to do if LRO or RSC are not enabled */ + if (!test_bit(WX_FLAG_RSC_CAPABLE, wx->flags) || + !(netdev->features & NETIF_F_LRO)) + return; + + /* check the feature flag value and enable RSC if necessary */ + if (wx->rx_itr_setting == 1 || + wx->rx_itr_setting > WX_MIN_RSC_ITR) { + if (!test_bit(WX_FLAG_RSC_ENABLED, wx->flags)) { + set_bit(WX_FLAG_RSC_ENABLED, wx->flags); + dev_info(&wx->pdev->dev, + "rx-usecs value high enough to re-enable RSC\n"); + + need_reset = true; + } + /* if interrupt rate is too high then disable RSC */ + } else if (test_bit(WX_FLAG_RSC_ENABLED, wx->flags)) { + clear_bit(WX_FLAG_RSC_ENABLED, wx->flags); + dev_info(&wx->pdev->dev, + "rx-usecs set too low, disabling RSC\n"); + + need_reset = true; + } + + /* reset the device to apply the new RSC setting */ + if (need_reset && wx->do_reset) + wx->do_reset(netdev); +} + int wx_set_coalesce(struct net_device *netdev, struct ethtool_coalesce *ec, struct kernel_ethtool_coalesce *kernel_coal, @@ -414,6 +469,8 @@ int wx_set_coalesce(struct net_device *netdev, wx_write_eitr(q_vector); } + wx_update_rsc(wx); + return 0; } EXPORT_SYMBOL(wx_set_coalesce); diff --git a/drivers/net/ethernet/wangxun/libwx/wx_hw.c b/drivers/net/ethernet/wangxun/libwx/wx_hw.c index 986bc5acc472..814164459707 100644 --- a/drivers/net/ethernet/wangxun/libwx/wx_hw.c +++ b/drivers/net/ethernet/wangxun/libwx/wx_hw.c @@ -1779,7 +1779,9 @@ EXPORT_SYMBOL(wx_set_rx_mode); static void wx_set_rx_buffer_len(struct wx *wx) { struct net_device *netdev = wx->netdev; + struct wx_ring *rx_ring; u32 mhadd, max_frame; + int i; max_frame = netdev->mtu + ETH_HLEN + ETH_FCS_LEN + VLAN_HLEN; /* adjust max frame to be at least the size of a standard frame */ @@ -1789,6 +1791,19 @@ static void wx_set_rx_buffer_len(struct wx *wx) mhadd = rd32(wx, WX_PSR_MAX_SZ); if (max_frame != mhadd) wr32(wx, WX_PSR_MAX_SZ, max_frame); + + /* + * Setup the HW Rx Head and Tail Descriptor Pointers and + * the Base and Length of the Rx Descriptor Ring + */ + for (i = 0; i < wx->num_rx_queues; i++) { + rx_ring = wx->rx_ring[i]; + rx_ring->rx_buf_len = WX_RXBUFFER_2K; +#if (PAGE_SIZE < 8192) + if (test_bit(WX_FLAG_RSC_ENABLED, wx->flags)) + rx_ring->rx_buf_len = WX_RXBUFFER_3K; +#endif + } } /** @@ -1865,11 +1880,27 @@ static void wx_configure_srrctl(struct wx *wx, srrctl |= WX_RXBUFFER_256 << WX_PX_RR_CFG_BHDRSIZE_SHIFT; /* configure the packet buffer length */ - srrctl |= WX_RX_BUFSZ >> WX_PX_RR_CFG_BSIZEPKT_SHIFT; + srrctl |= rx_ring->rx_buf_len >> WX_PX_RR_CFG_BSIZEPKT_SHIFT; wr32(wx, WX_PX_RR_CFG(reg_idx), srrctl); } +static void wx_configure_rscctl(struct wx *wx, + struct wx_ring *ring) +{ + u8 reg_idx = ring->reg_idx; + u32 rscctrl; + + if (!test_bit(WX_FLAG_RSC_ENABLED, wx->flags)) + return; + + rscctrl = rd32(wx, WX_PX_RR_CFG(reg_idx)); + rscctrl |= WX_PX_RR_CFG_RSC; + rscctrl |= WX_PX_RR_CFG_MAX_RSCBUF_16; + + wr32(wx, WX_PX_RR_CFG(reg_idx), rscctrl); +} + static void wx_configure_tx_ring(struct wx *wx, struct wx_ring *ring) { @@ -1956,6 +1987,7 @@ static void wx_configure_rx_ring(struct wx *wx, ring->tail = wx->hw_addr + WX_PX_RR_WP(reg_idx); wx_configure_srrctl(wx, ring); + wx_configure_rscctl(wx, ring); /* initialize rx_buffer_info */ memset(ring->rx_buffer_info, 0, @@ -2194,7 +2226,9 @@ void wx_configure_rx(struct wx *wx) /* RSC Setup */ psrctl = rd32(wx, WX_PSR_CTL); psrctl |= WX_PSR_CTL_RSC_ACK; /* Disable RSC for ACK packets */ - psrctl |= WX_PSR_CTL_RSC_DIS; + psrctl &= ~WX_PSR_CTL_RSC_DIS; + if (!test_bit(WX_FLAG_RSC_ENABLED, wx->flags)) + psrctl |= WX_PSR_CTL_RSC_DIS; wr32(wx, WX_PSR_CTL, psrctl); } @@ -2824,6 +2858,18 @@ void wx_update_stats(struct wx *wx) wx->hw_csum_rx_error = hw_csum_rx_error; wx->hw_csum_rx_good = hw_csum_rx_good; + if (test_bit(WX_FLAG_RSC_ENABLED, wx->flags)) { + u64 rsc_count = 0; + u64 rsc_flush = 0; + + for (i = 0; i < wx->num_rx_queues; i++) { + rsc_count += wx->rx_ring[i]->rx_stats.rsc_count; + rsc_flush += wx->rx_ring[i]->rx_stats.rsc_flush; + } + wx->rsc_count = rsc_count; + wx->rsc_flush = rsc_flush; + } + for (i = 0; i < wx->num_tx_queues; i++) { struct wx_ring *tx_ring = wx->tx_ring[i]; diff --git a/drivers/net/ethernet/wangxun/libwx/wx_lib.c b/drivers/net/ethernet/wangxun/libwx/wx_lib.c index 622213fe6790..32cadafa4b3b 100644 --- a/drivers/net/ethernet/wangxun/libwx/wx_lib.c +++ b/drivers/net/ethernet/wangxun/libwx/wx_lib.c @@ -235,7 +235,7 @@ static struct sk_buff *wx_build_skb(struct wx_ring *rx_ring, { unsigned int size = le16_to_cpu(rx_desc->wb.upper.length); #if (PAGE_SIZE < 8192) - unsigned int truesize = WX_RX_BUFSZ; + unsigned int truesize = wx_rx_pg_size(rx_ring) / 2; #else unsigned int truesize = ALIGN(size, L1_CACHE_BYTES); #endif @@ -341,7 +341,7 @@ void wx_alloc_rx_buffers(struct wx_ring *rx_ring, u16 cleaned_count) /* sync the buffer for use by the device */ dma_sync_single_range_for_device(rx_ring->dev, bi->dma, bi->page_offset, - WX_RX_BUFSZ, + rx_ring->rx_buf_len, DMA_FROM_DEVICE); rx_desc->read.pkt_addr = @@ -404,6 +404,7 @@ static bool wx_is_non_eop(struct wx_ring *rx_ring, union wx_rx_desc *rx_desc, struct sk_buff *skb) { + struct wx *wx = rx_ring->q_vector->wx; u32 ntc = rx_ring->next_to_clean + 1; /* fetch, update, and store next to clean */ @@ -412,6 +413,24 @@ static bool wx_is_non_eop(struct wx_ring *rx_ring, prefetch(WX_RX_DESC(rx_ring, ntc)); + /* update RSC append count if present */ + if (test_bit(WX_FLAG_RSC_ENABLED, wx->flags)) { + __le32 rsc_enabled = rx_desc->wb.lower.lo_dword.data & + cpu_to_le32(WX_RXD_RSCCNT_MASK); + + if (unlikely(rsc_enabled)) { + u32 rsc_cnt = le32_to_cpu(rsc_enabled); + + rsc_cnt >>= WX_RXD_RSCCNT_SHIFT; + WX_CB(skb)->append_cnt += rsc_cnt - 1; + + /* update ntc based on RSC value */ + ntc = le32_to_cpu(rx_desc->wb.upper.status_error); + ntc &= WX_RXD_NEXTP_MASK; + ntc >>= WX_RXD_NEXTP_SHIFT; + } + } + /* if we are the last buffer then there is nothing else to do */ if (likely(wx_test_staterr(rx_desc, WX_RXD_STAT_EOP))) return false; @@ -582,6 +601,33 @@ static void wx_rx_vlan(struct wx_ring *ring, union wx_rx_desc *rx_desc, } } +static void wx_set_rsc_gso_size(struct wx_ring *ring, + struct sk_buff *skb) +{ + u16 hdr_len = skb_headlen(skb); + + /* set gso_size to avoid messing up TCP MSS */ + skb_shinfo(skb)->gso_size = DIV_ROUND_UP((skb->len - hdr_len), + WX_CB(skb)->append_cnt); + skb_shinfo(skb)->gso_type = SKB_GSO_TCPV4; +} + +static void wx_update_rsc_stats(struct wx_ring *rx_ring, + struct sk_buff *skb) +{ + /* if append_cnt is 0 then frame is not RSC */ + if (!WX_CB(skb)->append_cnt) + return; + + rx_ring->rx_stats.rsc_count += WX_CB(skb)->append_cnt; + rx_ring->rx_stats.rsc_flush++; + + wx_set_rsc_gso_size(rx_ring, skb); + + /* gso_size is computed using append_cnt so always clear it last */ + WX_CB(skb)->append_cnt = 0; +} + /** * wx_process_skb_fields - Populate skb header fields from Rx descriptor * @rx_ring: rx descriptor ring packet is being transacted on @@ -598,6 +644,9 @@ static void wx_process_skb_fields(struct wx_ring *rx_ring, { struct wx *wx = netdev_priv(rx_ring->netdev); + if (test_bit(WX_FLAG_RSC_CAPABLE, wx->flags)) + wx_update_rsc_stats(rx_ring, skb); + wx_rx_hash(rx_ring, rx_desc, skb); wx_rx_checksum(rx_ring, rx_desc, skb); @@ -2549,7 +2598,7 @@ static void wx_clean_rx_ring(struct wx_ring *rx_ring) dma_sync_single_range_for_cpu(rx_ring->dev, rx_buffer->dma, rx_buffer->page_offset, - WX_RX_BUFSZ, + rx_ring->rx_buf_len, DMA_FROM_DEVICE); /* free resources associated with mapping */ @@ -2760,13 +2809,14 @@ static int wx_alloc_page_pool(struct wx_ring *rx_ring) struct page_pool_params pp_params = { .flags = PP_FLAG_DMA_MAP | PP_FLAG_DMA_SYNC_DEV, - .order = 0, - .pool_size = rx_ring->count, + .order = wx_rx_pg_order(rx_ring), + .pool_size = rx_ring->count * rx_ring->rx_buf_len / + wx_rx_pg_size(rx_ring), .nid = dev_to_node(rx_ring->dev), .dev = rx_ring->dev, .dma_dir = DMA_FROM_DEVICE, .offset = 0, - .max_len = PAGE_SIZE, + .max_len = wx_rx_pg_size(rx_ring), }; rx_ring->page_pool = page_pool_create(&pp_params); @@ -3075,8 +3125,25 @@ int wx_set_features(struct net_device *netdev, netdev_features_t features) else if (changed & (NETIF_F_HW_VLAN_CTAG_RX | NETIF_F_HW_VLAN_CTAG_FILTER)) wx_set_rx_mode(netdev); + if (test_bit(WX_FLAG_RSC_CAPABLE, wx->flags)) { + if (!(features & NETIF_F_LRO)) { + if (test_bit(WX_FLAG_RSC_ENABLED, wx->flags)) + need_reset = true; + clear_bit(WX_FLAG_RSC_ENABLED, wx->flags); + } else if (!(test_bit(WX_FLAG_RSC_ENABLED, wx->flags))) { + if (wx->rx_itr_setting == 1 || + wx->rx_itr_setting > WX_MIN_RSC_ITR) { + set_bit(WX_FLAG_RSC_ENABLED, wx->flags); + need_reset = true; + } else if (changed & NETIF_F_LRO) { + dev_info(&wx->pdev->dev, + "rx-usecs set too low, disable RSC\n"); + } + } + } + if (!(test_bit(WX_FLAG_FDIR_CAPABLE, wx->flags))) - return 0; + goto out; /* Check if Flow Director n-tuple support was enabled or disabled. If * the state changed, we need to reset. @@ -3102,6 +3169,7 @@ int wx_set_features(struct net_device *netdev, netdev_features_t features) break; } +out: if (need_reset && wx->do_reset) wx->do_reset(netdev); @@ -3151,6 +3219,14 @@ netdev_features_t wx_fix_features(struct net_device *netdev, } } + /* If Rx checksum is disabled, then RSC/LRO should also be disabled */ + if (!(features & NETIF_F_RXCSUM)) + features &= ~NETIF_F_LRO; + + /* Turn off LRO if not RSC capable */ + if (!test_bit(WX_FLAG_RSC_CAPABLE, wx->flags)) + features &= ~NETIF_F_LRO; + return features; } EXPORT_SYMBOL(wx_fix_features); diff --git a/drivers/net/ethernet/wangxun/libwx/wx_sriov.c b/drivers/net/ethernet/wangxun/libwx/wx_sriov.c index c6d158cd70da..493da5fffdb6 100644 --- a/drivers/net/ethernet/wangxun/libwx/wx_sriov.c +++ b/drivers/net/ethernet/wangxun/libwx/wx_sriov.c @@ -122,6 +122,10 @@ static int __wx_enable_sriov(struct wx *wx, u8 num_vfs) WX_CFG_PORT_CTL_NUM_VT_MASK, value); + /* Disable RSC when in SR-IOV mode */ + clear_bit(WX_FLAG_RSC_CAPABLE, wx->flags); + clear_bit(WX_FLAG_RSC_ENABLED, wx->flags); + return ret; } diff --git a/drivers/net/ethernet/wangxun/libwx/wx_type.h b/drivers/net/ethernet/wangxun/libwx/wx_type.h index 8b3c39945c0b..d0cbcded1dd4 100644 --- a/drivers/net/ethernet/wangxun/libwx/wx_type.h +++ b/drivers/net/ethernet/wangxun/libwx/wx_type.h @@ -424,6 +424,7 @@ enum WX_MSCA_CMD_value { #define WX_7K_ITR 595 #define WX_12K_ITR 336 #define WX_20K_ITR 200 +#define WX_MIN_RSC_ITR 24 #define WX_SP_MAX_EITR 0x00000FF8U #define WX_AML_MAX_EITR 0x00000FFFU #define WX_EM_MAX_EITR 0x00007FFCU @@ -454,7 +455,9 @@ enum WX_MSCA_CMD_value { /* PX_RR_CFG bit definitions */ #define WX_PX_RR_CFG_VLAN BIT(31) #define WX_PX_RR_CFG_DROP_EN BIT(30) +#define WX_PX_RR_CFG_RSC BIT(29) #define WX_PX_RR_CFG_SPLIT_MODE BIT(26) +#define WX_PX_RR_CFG_MAX_RSCBUF_16 FIELD_PREP(GENMASK(24, 23), 3) #define WX_PX_RR_CFG_DESC_MERGE BIT(19) #define WX_PX_RR_CFG_RR_THER_SHIFT 16 #define WX_PX_RR_CFG_RR_HDR_SZ GENMASK(15, 12) @@ -551,14 +554,9 @@ enum WX_MSCA_CMD_value { /* Supported Rx Buffer Sizes */ #define WX_RXBUFFER_256 256 /* Used for skb receive header */ #define WX_RXBUFFER_2K 2048 +#define WX_RXBUFFER_3K 3072 #define WX_MAX_RXBUFFER 16384 /* largest size for single descriptor */ -#if MAX_SKB_FRAGS < 8 -#define WX_RX_BUFSZ ALIGN(WX_MAX_RXBUFFER / MAX_SKB_FRAGS, 1024) -#else -#define WX_RX_BUFSZ WX_RXBUFFER_2K -#endif - #define WX_RX_BUFFER_WRITE 16 /* Must be power of 2 */ #define WX_MAX_DATA_PER_TXD BIT(14) @@ -652,6 +650,12 @@ enum wx_l2_ptypes { #define WX_RXD_PKTTYPE(_rxd) \ ((le32_to_cpu((_rxd)->wb.lower.lo_dword.data) >> 9) & 0xFF) + +#define WX_RXD_RSCCNT_MASK GENMASK(20, 17) +#define WX_RXD_RSCCNT_SHIFT 17 +#define WX_RXD_NEXTP_MASK GENMASK(19, 4) +#define WX_RXD_NEXTP_SHIFT 4 + /*********************** Transmit Descriptor Config Masks ****************/ #define WX_TXD_STAT_DD BIT(0) /* Descriptor Done */ #define WX_TXD_DTYP_DATA 0 /* Adv Data Descriptor */ @@ -1039,6 +1043,8 @@ struct wx_rx_queue_stats { u64 csum_good_cnt; u64 csum_err; u64 alloc_rx_buff_failed; + u64 rsc_count; + u64 rsc_flush; }; /* iterator for handling rings in ring container */ @@ -1081,6 +1087,7 @@ struct wx_ring { */ u16 next_to_use; u16 next_to_clean; + u16 rx_buf_len; union { u16 next_to_alloc; struct { @@ -1237,6 +1244,7 @@ enum wx_pf_flags { WX_FLAG_FDIR_HASH, WX_FLAG_FDIR_PERFECT, WX_FLAG_RSC_CAPABLE, + WX_FLAG_RSC_ENABLED, WX_FLAG_RX_HWTSTAMP_ENABLED, WX_FLAG_RX_HWTSTAMP_IN_REGISTER, WX_FLAG_PTP_PPS_ENABLED, @@ -1352,6 +1360,8 @@ struct wx { u64 hw_csum_rx_good; u64 hw_csum_rx_error; u64 alloc_rx_buff_failed; + u64 rsc_count; + u64 rsc_flush; unsigned int num_vfs; struct vf_data_storage *vfinfo; struct vf_macvlans vf_mvs; @@ -1483,4 +1493,15 @@ static inline int wx_set_state_reset(struct wx *wx) return 0; } +static inline unsigned int wx_rx_pg_order(struct wx_ring *ring) +{ +#if (PAGE_SIZE < 8192) + if (ring->rx_buf_len == WX_RXBUFFER_3K) + return 1; +#endif + return 0; +} + +#define wx_rx_pg_size(_ring) (PAGE_SIZE << wx_rx_pg_order(_ring)) + #endif /* _WX_TYPE_H_ */ diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c b/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c index ff690e9a075a..daa761e48f9d 100644 --- a/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c +++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c @@ -398,6 +398,7 @@ static int txgbe_sw_init(struct wx *wx) wx->configure_fdir = txgbe_configure_fdir; set_bit(WX_FLAG_RSC_CAPABLE, wx->flags); + set_bit(WX_FLAG_RSC_ENABLED, wx->flags); set_bit(WX_FLAG_MULTI_64_FUNC, wx->flags); /* enable itr by default in dynamic mode */ @@ -803,6 +804,8 @@ static int txgbe_probe(struct pci_dev *pdev, netdev->features |= NETIF_F_HIGHDMA; netdev->hw_features |= NETIF_F_GRO; netdev->features |= NETIF_F_GRO; + netdev->hw_features |= NETIF_F_LRO; + netdev->features |= NETIF_F_LRO; netdev->features |= NETIF_F_RX_UDP_TUNNEL_PORT; netdev->priv_flags |= IFF_UNICAST_FLT; From 211de28b1caf51581ba5e0978e83213db4f488c6 Mon Sep 17 00:00:00 2001 From: Mark Bloch Date: Thu, 23 Oct 2025 12:16:56 +0300 Subject: [PATCH 0296/1381] net/mlx5: Use common mlx5_same_hw_devs function Refactor duplicate hardware device comparison code to use the common mlx5_same_hw_devs() function instead of reimplementing system GUID comparison logic in multiple places. This cleanup eliminates code duplication in: - Bridge representor device comparison. - TC hardware device comparison. Using the centralized function improves maintainability and ensures consistent behavior across the driver. Signed-off-by: Mark Bloch Reviewed-by: Shay Drori Signed-off-by: Tariq Toukan Link: https://patch.msgid.link/1761211020-925651-2-git-send-email-tariqt@nvidia.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/mellanox/mlx5/core/en/rep/bridge.c | 6 +----- drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 6 +----- 2 files changed, 2 insertions(+), 10 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/rep/bridge.c b/drivers/net/ethernet/mellanox/mlx5/core/en/rep/bridge.c index 9d1c677814e0..87a2ad69526d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/rep/bridge.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/rep/bridge.c @@ -30,15 +30,11 @@ static bool mlx5_esw_bridge_dev_same_hw(struct net_device *dev, struct mlx5_eswi { struct mlx5e_priv *priv = netdev_priv(dev); struct mlx5_core_dev *mdev, *esw_mdev; - u64 system_guid, esw_system_guid; mdev = priv->mdev; esw_mdev = esw->dev; - system_guid = mlx5_query_nic_system_image_guid(mdev); - esw_system_guid = mlx5_query_nic_system_image_guid(esw_mdev); - - return system_guid == esw_system_guid; + return mlx5_same_hw_devs(mdev, esw_mdev); } static struct net_device * diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c index 00c2763e57ca..54ccfb4e6c4e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c @@ -3614,15 +3614,11 @@ static bool same_port_devs(struct mlx5e_priv *priv, struct mlx5e_priv *peer_priv bool mlx5e_same_hw_devs(struct mlx5e_priv *priv, struct mlx5e_priv *peer_priv) { struct mlx5_core_dev *fmdev, *pmdev; - u64 fsystem_guid, psystem_guid; fmdev = priv->mdev; pmdev = peer_priv->mdev; - fsystem_guid = mlx5_query_nic_system_image_guid(fmdev); - psystem_guid = mlx5_query_nic_system_image_guid(pmdev); - - return (fsystem_guid == psystem_guid); + return mlx5_same_hw_devs(fmdev, pmdev); } static int From 7718f2a8b87af7363d60819ac0ac0da8b2f8ff00 Mon Sep 17 00:00:00 2001 From: Mark Bloch Date: Thu, 23 Oct 2025 12:16:57 +0300 Subject: [PATCH 0297/1381] net/mlx5: Add software system image GUID infrastructure Replace direct hardware system image GUID usage with a new software system image GUID function that supports variable-length identifiers. Key changes: - Add mlx5_query_nic_sw_system_image_guid() function with length parameter. - Update all callsites to use the new function and buffer/length approach. - Modify mapping contexts to use byte arrays instead of u64 keys. - Update devcom matching to support variable-length keys. - Change mlx5_same_hw_devs() to use buffer comparison instead of u64. This refactoring prepares the infrastructure for balance ID support, which requires extending the system image GUID with additional data. The change maintains backward compatibility while enabling future enhancements. Signed-off-by: Mark Bloch Reviewed-by: Shay Drori Signed-off-by: Tariq Toukan Link: https://patch.msgid.link/1761211020-925651-3-git-send-email-tariqt@nvidia.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/mellanox/mlx5/core/dev.c | 12 ++++++--- .../ethernet/mellanox/mlx5/core/en/devlink.c | 7 ++--- .../ethernet/mellanox/mlx5/core/en/mapping.c | 13 +++++++--- .../ethernet/mellanox/mlx5/core/en/mapping.h | 3 ++- .../mellanox/mlx5/core/en/tc/int_port.c | 8 +++--- .../ethernet/mellanox/mlx5/core/en/tc_ct.c | 11 +++++--- .../net/ethernet/mellanox/mlx5/core/en_tc.c | 26 ++++++++++++------- .../mellanox/mlx5/core/esw/devlink_port.c | 6 +---- .../mellanox/mlx5/core/eswitch_offloads.c | 8 +++--- .../net/ethernet/mellanox/mlx5/core/lag/lag.c | 4 ++- .../ethernet/mellanox/mlx5/core/lib/devcom.h | 2 ++ .../ethernet/mellanox/mlx5/core/mlx5_core.h | 2 ++ .../net/ethernet/mellanox/mlx5/core/vport.c | 15 +++++++++++ include/linux/mlx5/driver.h | 3 +++ 14 files changed, 80 insertions(+), 40 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/dev.c b/drivers/net/ethernet/mellanox/mlx5/core/dev.c index 891bbbbfbbf1..64c04f52990f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/dev.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/dev.c @@ -564,10 +564,14 @@ int mlx5_rescan_drivers_locked(struct mlx5_core_dev *dev) bool mlx5_same_hw_devs(struct mlx5_core_dev *dev, struct mlx5_core_dev *peer_dev) { - u64 fsystem_guid, psystem_guid; + u8 fsystem_guid[MLX5_SW_IMAGE_GUID_MAX_BYTES]; + u8 psystem_guid[MLX5_SW_IMAGE_GUID_MAX_BYTES]; + u8 flen; + u8 plen; - fsystem_guid = mlx5_query_nic_system_image_guid(dev); - psystem_guid = mlx5_query_nic_system_image_guid(peer_dev); + mlx5_query_nic_sw_system_image_guid(dev, fsystem_guid, &flen); + mlx5_query_nic_sw_system_image_guid(peer_dev, psystem_guid, &plen); - return (fsystem_guid && psystem_guid && fsystem_guid == psystem_guid); + return plen && flen && flen == plen && + !memcmp(fsystem_guid, psystem_guid, flen); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/devlink.c b/drivers/net/ethernet/mellanox/mlx5/core/en/devlink.c index 0b1ac6e5c890..8818f65d1fbc 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/devlink.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/devlink.c @@ -40,11 +40,8 @@ void mlx5e_destroy_devlink(struct mlx5e_dev *mlx5e_dev) static void mlx5e_devlink_get_port_parent_id(struct mlx5_core_dev *dev, struct netdev_phys_item_id *ppid) { - u64 parent_id; - - parent_id = mlx5_query_nic_system_image_guid(dev); - ppid->id_len = sizeof(parent_id); - memcpy(ppid->id, &parent_id, sizeof(parent_id)); + BUILD_BUG_ON(MLX5_SW_IMAGE_GUID_MAX_BYTES > MAX_PHYS_ITEM_ID_LEN); + mlx5_query_nic_sw_system_image_guid(dev, ppid->id, &ppid->id_len); } int mlx5e_devlink_port_register(struct mlx5e_dev *mlx5e_dev, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/mapping.c b/drivers/net/ethernet/mellanox/mlx5/core/en/mapping.c index 4e72ca8070e2..1de18c7e96ec 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/mapping.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/mapping.c @@ -6,6 +6,7 @@ #include #include #include +#include #include "mapping.h" @@ -24,7 +25,8 @@ struct mapping_ctx { struct delayed_work dwork; struct list_head pending_list; spinlock_t pending_list_lock; /* Guards pending list */ - u64 id; + u8 id[MLX5_SW_IMAGE_GUID_MAX_BYTES]; + u8 id_len; u8 type; struct list_head list; refcount_t refcount; @@ -220,13 +222,15 @@ mapping_create(size_t data_size, u32 max_id, bool delayed_removal) } struct mapping_ctx * -mapping_create_for_id(u64 id, u8 type, size_t data_size, u32 max_id, bool delayed_removal) +mapping_create_for_id(u8 *id, u8 id_len, u8 type, size_t data_size, u32 max_id, + bool delayed_removal) { struct mapping_ctx *ctx; mutex_lock(&shared_ctx_lock); list_for_each_entry(ctx, &shared_ctx_list, list) { - if (ctx->id == id && ctx->type == type) { + if (ctx->type == type && ctx->id_len == id_len && + !memcmp(id, ctx->id, id_len)) { if (refcount_inc_not_zero(&ctx->refcount)) goto unlock; break; @@ -237,7 +241,8 @@ mapping_create_for_id(u64 id, u8 type, size_t data_size, u32 max_id, bool delaye if (IS_ERR(ctx)) goto unlock; - ctx->id = id; + memcpy(ctx->id, id, id_len); + ctx->id_len = id_len; ctx->type = type; list_add(&ctx->list, &shared_ctx_list); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/mapping.h b/drivers/net/ethernet/mellanox/mlx5/core/en/mapping.h index 4e2119f0f4c1..e86a103d58b9 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/mapping.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/mapping.h @@ -27,6 +27,7 @@ void mapping_destroy(struct mapping_ctx *ctx); /* adds mapping with an id or get an existing mapping with the same id */ struct mapping_ctx * -mapping_create_for_id(u64 id, u8 type, size_t data_size, u32 max_id, bool delayed_removal); +mapping_create_for_id(u8 *id, u8 id_len, u8 type, size_t data_size, u32 max_id, + bool delayed_removal); #endif /* __MLX5_MAPPING_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/int_port.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/int_port.c index 896f718483c3..991f47050643 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/int_port.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/int_port.c @@ -307,7 +307,8 @@ mlx5e_tc_int_port_init(struct mlx5e_priv *priv) { struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; struct mlx5e_tc_int_port_priv *int_port_priv; - u64 mapping_id; + u8 mapping_id[MLX5_SW_IMAGE_GUID_MAX_BYTES]; + u8 id_len; if (!mlx5e_tc_int_port_supported(esw)) return NULL; @@ -316,9 +317,10 @@ mlx5e_tc_int_port_init(struct mlx5e_priv *priv) if (!int_port_priv) return NULL; - mapping_id = mlx5_query_nic_system_image_guid(priv->mdev); + mlx5_query_nic_sw_system_image_guid(priv->mdev, mapping_id, &id_len); - int_port_priv->metadata_mapping = mapping_create_for_id(mapping_id, MAPPING_TYPE_INT_PORT, + int_port_priv->metadata_mapping = mapping_create_for_id(mapping_id, id_len, + MAPPING_TYPE_INT_PORT, sizeof(u32) * 2, (1 << ESW_VPORT_BITS) - 1, true); if (IS_ERR(int_port_priv->metadata_mapping)) { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c index 870d12364f99..fc0e57403d25 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c @@ -2287,9 +2287,10 @@ mlx5_tc_ct_init(struct mlx5e_priv *priv, struct mlx5_fs_chains *chains, enum mlx5_flow_namespace_type ns_type, struct mlx5e_post_act *post_act) { + u8 mapping_id[MLX5_SW_IMAGE_GUID_MAX_BYTES]; struct mlx5_tc_ct_priv *ct_priv; struct mlx5_core_dev *dev; - u64 mapping_id; + u8 id_len; int err; dev = priv->mdev; @@ -2301,16 +2302,18 @@ mlx5_tc_ct_init(struct mlx5e_priv *priv, struct mlx5_fs_chains *chains, if (!ct_priv) goto err_alloc; - mapping_id = mlx5_query_nic_system_image_guid(dev); + mlx5_query_nic_sw_system_image_guid(dev, mapping_id, &id_len); - ct_priv->zone_mapping = mapping_create_for_id(mapping_id, MAPPING_TYPE_ZONE, + ct_priv->zone_mapping = mapping_create_for_id(mapping_id, id_len, + MAPPING_TYPE_ZONE, sizeof(u16), 0, true); if (IS_ERR(ct_priv->zone_mapping)) { err = PTR_ERR(ct_priv->zone_mapping); goto err_mapping_zone; } - ct_priv->labels_mapping = mapping_create_for_id(mapping_id, MAPPING_TYPE_LABELS, + ct_priv->labels_mapping = mapping_create_for_id(mapping_id, id_len, + MAPPING_TYPE_LABELS, sizeof(u32) * 4, 0, true); if (IS_ERR(ct_priv->labels_mapping)) { err = PTR_ERR(ct_priv->labels_mapping); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c index 54ccfb4e6c4e..a8773b2342c2 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c @@ -5233,10 +5233,11 @@ static void mlx5e_tc_nic_destroy_miss_table(struct mlx5e_priv *priv) int mlx5e_tc_nic_init(struct mlx5e_priv *priv) { struct mlx5e_tc_table *tc = mlx5e_fs_get_tc(priv->fs); + u8 mapping_id[MLX5_SW_IMAGE_GUID_MAX_BYTES]; struct mlx5_core_dev *dev = priv->mdev; struct mapping_ctx *chains_mapping; struct mlx5_chains_attr attr = {}; - u64 mapping_id; + u8 id_len; int err; mlx5e_mod_hdr_tbl_init(&tc->mod_hdr); @@ -5252,11 +5253,13 @@ int mlx5e_tc_nic_init(struct mlx5e_priv *priv) lockdep_set_class(&tc->ht.mutex, &tc_ht_lock_key); lockdep_init_map(&tc->ht.run_work.lockdep_map, "tc_ht_wq_key", &tc_ht_wq_key, 0); - mapping_id = mlx5_query_nic_system_image_guid(dev); + mlx5_query_nic_sw_system_image_guid(dev, mapping_id, &id_len); - chains_mapping = mapping_create_for_id(mapping_id, MAPPING_TYPE_CHAIN, + chains_mapping = mapping_create_for_id(mapping_id, id_len, + MAPPING_TYPE_CHAIN, sizeof(struct mlx5_mapped_obj), - MLX5E_TC_TABLE_CHAIN_TAG_MASK, true); + MLX5E_TC_TABLE_CHAIN_TAG_MASK, + true); if (IS_ERR(chains_mapping)) { err = PTR_ERR(chains_mapping); @@ -5387,14 +5390,15 @@ void mlx5e_tc_ht_cleanup(struct rhashtable *tc_ht) int mlx5e_tc_esw_init(struct mlx5_rep_uplink_priv *uplink_priv) { const size_t sz_enc_opts = sizeof(struct tunnel_match_enc_opts); + u8 mapping_id[MLX5_SW_IMAGE_GUID_MAX_BYTES]; struct mlx5_devcom_match_attr attr = {}; struct netdev_phys_item_id ppid; struct mlx5e_rep_priv *rpriv; struct mapping_ctx *mapping; struct mlx5_eswitch *esw; struct mlx5e_priv *priv; - u64 mapping_id; int err = 0; + u8 id_len; rpriv = container_of(uplink_priv, struct mlx5e_rep_priv, uplink_priv); priv = netdev_priv(rpriv->netdev); @@ -5412,9 +5416,9 @@ int mlx5e_tc_esw_init(struct mlx5_rep_uplink_priv *uplink_priv) uplink_priv->tc_psample = mlx5e_tc_sample_init(esw, uplink_priv->post_act); - mapping_id = mlx5_query_nic_system_image_guid(esw->dev); + mlx5_query_nic_sw_system_image_guid(esw->dev, mapping_id, &id_len); - mapping = mapping_create_for_id(mapping_id, MAPPING_TYPE_TUNNEL, + mapping = mapping_create_for_id(mapping_id, id_len, MAPPING_TYPE_TUNNEL, sizeof(struct tunnel_match_key), TUNNEL_INFO_BITS_MASK, true); @@ -5427,8 +5431,10 @@ int mlx5e_tc_esw_init(struct mlx5_rep_uplink_priv *uplink_priv) /* Two last values are reserved for stack devices slow path table mark * and bridge ingress push mark. */ - mapping = mapping_create_for_id(mapping_id, MAPPING_TYPE_TUNNEL_ENC_OPTS, - sz_enc_opts, ENC_OPTS_BITS_MASK - 2, true); + mapping = mapping_create_for_id(mapping_id, id_len, + MAPPING_TYPE_TUNNEL_ENC_OPTS, + sz_enc_opts, ENC_OPTS_BITS_MASK - 2, + true); if (IS_ERR(mapping)) { err = PTR_ERR(mapping); goto err_enc_opts_mapping; @@ -5449,7 +5455,7 @@ int mlx5e_tc_esw_init(struct mlx5_rep_uplink_priv *uplink_priv) err = netif_get_port_parent_id(priv->netdev, &ppid, false); if (!err) { - memcpy(&attr.key.val, &ppid.id, sizeof(attr.key.val)); + memcpy(&attr.key.buf, &ppid.id, ppid.id_len); attr.flags = MLX5_DEVCOM_MATCH_FLAGS_NS; attr.net = mlx5_core_net(esw->dev); mlx5_esw_offloads_devcom_init(esw, &attr); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/devlink_port.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/devlink_port.c index cf88a106d80d..89a58dee50b3 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/esw/devlink_port.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/devlink_port.c @@ -7,11 +7,7 @@ static void mlx5_esw_get_port_parent_id(struct mlx5_core_dev *dev, struct netdev_phys_item_id *ppid) { - u64 parent_id; - - parent_id = mlx5_query_nic_system_image_guid(dev); - ppid->id_len = sizeof(parent_id); - memcpy(ppid->id, &parent_id, sizeof(parent_id)); + mlx5_query_nic_sw_system_image_guid(dev, ppid->id, &ppid->id_len); } static bool mlx5_esw_devlink_port_supported(struct mlx5_eswitch *esw, u16 vport_num) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c index 34749814f19b..9735a75732cf 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -3557,10 +3557,11 @@ bool mlx5_esw_offloads_controller_valid(const struct mlx5_eswitch *esw, u32 cont int esw_offloads_enable(struct mlx5_eswitch *esw) { + u8 mapping_id[MLX5_SW_IMAGE_GUID_MAX_BYTES]; struct mapping_ctx *reg_c0_obj_pool; struct mlx5_vport *vport; unsigned long i; - u64 mapping_id; + u8 id_len; int err; mutex_init(&esw->offloads.termtbl_mutex); @@ -3582,9 +3583,10 @@ int esw_offloads_enable(struct mlx5_eswitch *esw) if (err) goto err_vport_metadata; - mapping_id = mlx5_query_nic_system_image_guid(esw->dev); + mlx5_query_nic_sw_system_image_guid(esw->dev, mapping_id, &id_len); - reg_c0_obj_pool = mapping_create_for_id(mapping_id, MAPPING_TYPE_CHAIN, + reg_c0_obj_pool = mapping_create_for_id(mapping_id, id_len, + MAPPING_TYPE_CHAIN, sizeof(struct mlx5_mapped_obj), ESW_REG_C0_USER_DATA_METADATA_MASK, true); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.c b/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.c index 3db0387bf6dc..1ac933cd8f02 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.c @@ -1418,10 +1418,12 @@ static void mlx5_lag_unregister_hca_devcom_comp(struct mlx5_core_dev *dev) static int mlx5_lag_register_hca_devcom_comp(struct mlx5_core_dev *dev) { struct mlx5_devcom_match_attr attr = { - .key.val = mlx5_query_nic_system_image_guid(dev), .flags = MLX5_DEVCOM_MATCH_FLAGS_NS, .net = mlx5_core_net(dev), }; + u8 len __always_unused; + + mlx5_query_nic_sw_system_image_guid(dev, attr.key.buf, &len); /* This component is use to sync adding core_dev to lag_dev and to sync * changes of mlx5_adev_devices between LAG layer and other layers. diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.h b/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.h index 609c85f47917..91e5ae529d5c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.h @@ -10,8 +10,10 @@ enum mlx5_devom_match_flags { MLX5_DEVCOM_MATCH_FLAGS_NS = BIT(0), }; +#define MLX5_DEVCOM_MATCH_KEY_MAX 32 union mlx5_devcom_match_key { u64 val; + u8 buf[MLX5_DEVCOM_MATCH_KEY_MAX]; }; struct mlx5_devcom_match_attr { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h index 082259b56816..acef7d0ffa09 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h @@ -444,6 +444,8 @@ int mlx5_init_one_light(struct mlx5_core_dev *dev); void mlx5_uninit_one_light(struct mlx5_core_dev *dev); void mlx5_unload_one_light(struct mlx5_core_dev *dev); +void mlx5_query_nic_sw_system_image_guid(struct mlx5_core_dev *mdev, u8 *buf, + u8 *len); int mlx5_vport_set_other_func_cap(struct mlx5_core_dev *dev, const void *hca_cap, u16 vport, u16 opmod); #define mlx5_vport_get_other_func_general_cap(dev, vport, out) \ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/vport.c b/drivers/net/ethernet/mellanox/mlx5/core/vport.c index 2ed2e530b07d..4224e2750865 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/vport.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/vport.c @@ -1190,6 +1190,21 @@ u64 mlx5_query_nic_system_image_guid(struct mlx5_core_dev *mdev) } EXPORT_SYMBOL_GPL(mlx5_query_nic_system_image_guid); +void mlx5_query_nic_sw_system_image_guid(struct mlx5_core_dev *mdev, u8 *buf, + u8 *len) +{ + u64 fw_system_image_guid; + + *len = 0; + + fw_system_image_guid = mlx5_query_nic_system_image_guid(mdev); + if (!fw_system_image_guid) + return; + + memcpy(buf, &fw_system_image_guid, sizeof(fw_system_image_guid)); + *len += sizeof(fw_system_image_guid); +} + static bool mlx5_vport_use_vhca_id_as_func_id(struct mlx5_core_dev *dev, u16 vport_num, u16 *vhca_id) { diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index 5405ca1038f9..dcf262aa9ea6 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -1379,4 +1379,7 @@ static inline struct net *mlx5_core_net(struct mlx5_core_dev *dev) { return devlink_net(priv_to_devlink(dev)); } + +#define MLX5_SW_IMAGE_GUID_MAX_BYTES 8 + #endif /* MLX5_DRIVER_H */ From cd36818c34ac5ff7f6a50ce88822c7bbb5ac9e0d Mon Sep 17 00:00:00 2001 From: Mark Bloch Date: Thu, 23 Oct 2025 12:16:58 +0300 Subject: [PATCH 0298/1381] net/mlx5: Refactor PTP clock devcom pairing Refactor PTP clock device component pairing to use the clock identity buffer instead of casting it to a u64 key. This change leverages the new software system image GUID infrastructure. Changes include: - Pass identity buffer to mlx5_shared_clock_register(). - Use memcpy for identity buffer in devcom matching attributes. - Remove intermediate u64 key conversion. - Add BUILD_BUG_ON to ensure identity size fits in match key. Signed-off-by: Mark Bloch Reviewed-by: Shay Drori Signed-off-by: Tariq Toukan Link: https://patch.msgid.link/1761211020-925651-4-git-send-email-tariqt@nvidia.com Signed-off-by: Paolo Abeni --- .../ethernet/mellanox/mlx5/core/lib/clock.c | 19 +++++++++---------- 1 file changed, 9 insertions(+), 10 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c index 29e7fa09c32c..0ba0ef8bae42 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c @@ -1432,15 +1432,17 @@ static int mlx5_clock_alloc(struct mlx5_core_dev *mdev, bool shared) return 0; } -static void mlx5_shared_clock_register(struct mlx5_core_dev *mdev, u64 key) +static void mlx5_shared_clock_register(struct mlx5_core_dev *mdev, + u8 identity[MLX5_RT_CLOCK_IDENTITY_SIZE]) { struct mlx5_core_dev *peer_dev, *next = NULL; - struct mlx5_devcom_match_attr attr = { - .key.val = key, - }; + struct mlx5_devcom_match_attr attr = {}; struct mlx5_devcom_comp_dev *compd; struct mlx5_devcom_comp_dev *pos; + BUILD_BUG_ON(MLX5_RT_CLOCK_IDENTITY_SIZE > MLX5_DEVCOM_MATCH_KEY_MAX); + memcpy(attr.key.buf, identity, MLX5_RT_CLOCK_IDENTITY_SIZE); + compd = mlx5_devcom_register_component(mdev->priv.devc, MLX5_DEVCOM_SHARED_CLOCK, &attr, NULL, mdev); @@ -1594,7 +1596,6 @@ int mlx5_init_clock(struct mlx5_core_dev *mdev) { u8 identity[MLX5_RT_CLOCK_IDENTITY_SIZE]; struct mlx5_clock_dev_state *clock_state; - u64 key; int err; if (!MLX5_CAP_GEN(mdev, device_frequency_khz)) { @@ -1610,12 +1611,10 @@ int mlx5_init_clock(struct mlx5_core_dev *mdev) mdev->clock_state = clock_state; if (MLX5_CAP_MCAM_REG3(mdev, mrtcq) && mlx5_real_time_mode(mdev)) { - if (mlx5_clock_identity_get(mdev, identity)) { + if (mlx5_clock_identity_get(mdev, identity)) mlx5_core_warn(mdev, "failed to get rt clock identity, create ptp dev per function\n"); - } else { - memcpy(&key, &identity, sizeof(key)); - mlx5_shared_clock_register(mdev, key); - } + else + mlx5_shared_clock_register(mdev, identity); } if (!mdev->clock) { From 075e85a1261e4653c2068e68a8c91da6c7bc4e60 Mon Sep 17 00:00:00 2001 From: Mark Bloch Date: Thu, 23 Oct 2025 12:16:59 +0300 Subject: [PATCH 0299/1381] net/mlx5: Refactor HCA cap 2 setting Refactor HCA capability 2 setting logic to be more structured and conditional. Move the sw_vhca_id_valid setting inside proper conditional checks and prepare the function for additional capability settings. The refactoring: - Always copy current capabilities to set_hca_cap buffer. - Apply sw_vhca_id_valid setting only when conditions are met. - Improve code readability and maintainability. This cleanup prepares the handle_hca_cap_2() function for the upcoming balance ID capability setting. Signed-off-by: Mark Bloch Reviewed-by: Moshe Shemesh Reviewed-by: Shay Drori Signed-off-by: Tariq Toukan Link: https://patch.msgid.link/1761211020-925651-5-git-send-email-tariqt@nvidia.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/mellanox/mlx5/core/main.c | 18 ++++++++++++------ 1 file changed, 12 insertions(+), 6 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c index 70c156591b0b..563267acf386 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c @@ -553,6 +553,7 @@ EXPORT_SYMBOL(mlx5_is_roce_on); static int handle_hca_cap_2(struct mlx5_core_dev *dev, void *set_ctx) { + bool do_set = false; void *set_hca_cap; int err; @@ -563,17 +564,22 @@ static int handle_hca_cap_2(struct mlx5_core_dev *dev, void *set_ctx) if (err) return err; - if (!MLX5_CAP_GEN_2_MAX(dev, sw_vhca_id_valid) || - !(dev->priv.sw_vhca_id > 0)) - return 0; - set_hca_cap = MLX5_ADDR_OF(set_hca_cap_in, set_ctx, capability); memcpy(set_hca_cap, dev->caps.hca[MLX5_CAP_GENERAL_2]->cur, MLX5_ST_SZ_BYTES(cmd_hca_cap_2)); - MLX5_SET(cmd_hca_cap_2, set_hca_cap, sw_vhca_id_valid, 1); - return set_caps(dev, set_ctx, MLX5_CAP_GENERAL_2); + if (MLX5_CAP_GEN_2_MAX(dev, sw_vhca_id_valid) && + dev->priv.sw_vhca_id > 0) { + MLX5_SET(cmd_hca_cap_2, set_hca_cap, sw_vhca_id_valid, 1); + do_set = true; + } + + /* some FW versions that support querying MLX5_CAP_GENERAL_2 + * capabilities but don't support setting them. + * Skip unnecessary update to hca_cap_2 when no changes were introduced + */ + return do_set ? set_caps(dev, set_ctx, MLX5_CAP_GENERAL_2) : 0; } static int handle_hca_cap(struct mlx5_core_dev *dev, void *set_ctx) From 20d78ead947783b039b02ca4b8c551b4d1894759 Mon Sep 17 00:00:00 2001 From: Mark Bloch Date: Thu, 23 Oct 2025 12:17:00 +0300 Subject: [PATCH 0300/1381] net/mlx5: Add balance ID support for LAG multiplane groups Implement balance ID support for multiplane LAG configurations. This feature enables per-multiplane group load balancing by extending the software system image GUID with a balance ID component. Key implementations: - Enable lag_per_mp_group capability when supported by hardware. - Append load_balance_id to software system image GUID when conditions are met. - Increase MLX5_SW_IMAGE_GUID_MAX_BYTES from 8 to 9 to accommodate the extra byte. The balance ID is appended to the system image GUID only when both load_balance_id and lag_per_mp_group capabilities are available, ensuring backward compatibility while enabling enhanced LAG functionality. This enhancement allows for more granular load balancing control in complex multi-plane LAG deployments, improving network performance and flexibility. Signed-off-by: Mark Bloch Reviewed-by: Moshe Shemesh Reviewed-by: Shay Drori Signed-off-by: Tariq Toukan Link: https://patch.msgid.link/1761211020-925651-6-git-send-email-tariqt@nvidia.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/mellanox/mlx5/core/main.c | 5 +++++ drivers/net/ethernet/mellanox/mlx5/core/vport.c | 4 ++++ include/linux/mlx5/driver.h | 2 +- 3 files changed, 10 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c index 563267acf386..c904696cbc3a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c @@ -575,6 +575,11 @@ static int handle_hca_cap_2(struct mlx5_core_dev *dev, void *set_ctx) do_set = true; } + if (MLX5_CAP_GEN_2_MAX(dev, lag_per_mp_group)) { + MLX5_SET(cmd_hca_cap_2, set_hca_cap, lag_per_mp_group, 1); + do_set = true; + } + /* some FW versions that support querying MLX5_CAP_GENERAL_2 * capabilities but don't support setting them. * Skip unnecessary update to hca_cap_2 when no changes were introduced diff --git a/drivers/net/ethernet/mellanox/mlx5/core/vport.c b/drivers/net/ethernet/mellanox/mlx5/core/vport.c index 4224e2750865..992873536c1b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/vport.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/vport.c @@ -1203,6 +1203,10 @@ void mlx5_query_nic_sw_system_image_guid(struct mlx5_core_dev *mdev, u8 *buf, memcpy(buf, &fw_system_image_guid, sizeof(fw_system_image_guid)); *len += sizeof(fw_system_image_guid); + + if (MLX5_CAP_GEN_2(mdev, load_balance_id) && + MLX5_CAP_GEN_2(mdev, lag_per_mp_group)) + buf[(*len)++] = MLX5_CAP_GEN_2(mdev, load_balance_id); } static bool mlx5_vport_use_vhca_id_as_func_id(struct mlx5_core_dev *dev, diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index dcf262aa9ea6..046396269ccf 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -1380,6 +1380,6 @@ static inline struct net *mlx5_core_net(struct mlx5_core_dev *dev) return devlink_net(priv_to_devlink(dev)); } -#define MLX5_SW_IMAGE_GUID_MAX_BYTES 8 +#define MLX5_SW_IMAGE_GUID_MAX_BYTES 9 #endif /* MLX5_DRIVER_H */ From 182663bbff784e88cf7acbae7c53c6c5e0cb9ca4 Mon Sep 17 00:00:00 2001 From: Alexandra Winter Date: Thu, 23 Oct 2025 17:06:35 +0200 Subject: [PATCH 0301/1381] dibs: Remove reset of static vars in dibs_init() 'clients' and 'max_client' are static variables and therefore don't need to be initialized. Reported-by: Mete Durlu Signed-off-by: Alexandra Winter Reviewed-by: Alexander Lobakin Reviewed-by: Dust Li Link: https://patch.msgid.link/20251023150636.3995476-1-wintera@linux.ibm.com Signed-off-by: Paolo Abeni --- drivers/dibs/dibs_main.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/drivers/dibs/dibs_main.c b/drivers/dibs/dibs_main.c index 0374f8350ff7..b015578b4d2e 100644 --- a/drivers/dibs/dibs_main.c +++ b/drivers/dibs/dibs_main.c @@ -254,9 +254,6 @@ static int __init dibs_init(void) { int rc; - memset(clients, 0, sizeof(clients)); - max_client = 0; - dibs_class = class_create("dibs"); if (IS_ERR(dibs_class)) return PTR_ERR(dibs_class); From 968822086b74dd0a3df693f9d179bd4fe508faf9 Mon Sep 17 00:00:00 2001 From: Alexandra Winter Date: Thu, 23 Oct 2025 17:06:36 +0200 Subject: [PATCH 0302/1381] dibs: Use subsys_initcall() In the case of built-in modules, the order of module_init() calls are derived from the Makefiles. Use subsys_initcall() for the dibs module, to make sure dibs_init() is executed before dibs clients like smc and dibs devices like ism are initialized. So future dibs client or dibs device modules can use module_init() without the risk of getting the order in the Makefiles wrong. Reported-by: Mete Durlu Signed-off-by: Alexandra Winter Reviewed-by: Alexander Lobakin Link: https://patch.msgid.link/20251023150636.3995476-2-wintera@linux.ibm.com Signed-off-by: Paolo Abeni --- drivers/dibs/dibs_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/dibs/dibs_main.c b/drivers/dibs/dibs_main.c index b015578b4d2e..dac14d843af7 100644 --- a/drivers/dibs/dibs_main.c +++ b/drivers/dibs/dibs_main.c @@ -271,5 +271,5 @@ static void __exit dibs_exit(void) class_destroy(dibs_class); } -module_init(dibs_init); +subsys_initcall(dibs_init); module_exit(dibs_exit); From 37ff03b356ef9d2e960c07596e6fa9276198206d Mon Sep 17 00:00:00 2001 From: Miri Korenblit Date: Mon, 15 Sep 2025 11:34:22 +0300 Subject: [PATCH 0303/1381] wifi: iwlwifi: align the name of iwl_alive_ntf_v6 to the convention This struct rperesent versions 6 and 7. The convention is to name an API struct with the last version it represent, so rename to iwl_alive_ntf_v7. Reviewed-by: Emmanuel Grumbach Signed-off-by: Miri Korenblit Link: https://patch.msgid.link/20250915113137.81240e1d4df3.I2c1264a49b9f0fc160f960cf3c5dc4cedf6ceb6d@changeid --- drivers/net/wireless/intel/iwlwifi/fw/api/alive.h | 2 +- drivers/net/wireless/intel/iwlwifi/fw/api/commands.h | 2 +- drivers/net/wireless/intel/iwlwifi/mld/fw.c | 2 +- drivers/net/wireless/intel/iwlwifi/mvm/fw.c | 2 +- 4 files changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/alive.h b/drivers/net/wireless/intel/iwlwifi/fw/api/alive.h index ad5b95cad0bf..ea2ba4b4cb7b 100644 --- a/drivers/net/wireless/intel/iwlwifi/fw/api/alive.h +++ b/drivers/net/wireless/intel/iwlwifi/fw/api/alive.h @@ -88,7 +88,7 @@ struct iwl_imr_alive_info { __le32 enabled; } __packed; /* IMR_ALIVE_INFO_API_S_VER_1 */ -struct iwl_alive_ntf_v6 { +struct iwl_alive_ntf_v7 { __le16 status; __le16 flags; struct iwl_lmac_alive lmac_data[2]; diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/commands.h b/drivers/net/wireless/intel/iwlwifi/fw/api/commands.h index 997b0c9ce984..8d64a271bb94 100644 --- a/drivers/net/wireless/intel/iwlwifi/fw/api/commands.h +++ b/drivers/net/wireless/intel/iwlwifi/fw/api/commands.h @@ -60,7 +60,7 @@ enum iwl_legacy_cmds { * @UCODE_ALIVE_NTFY: * Alive data from the firmware, as described in * &struct iwl_alive_ntf_v3 or &struct iwl_alive_ntf_v4 or - * &struct iwl_alive_ntf_v5 or &struct iwl_alive_ntf_v6. + * &struct iwl_alive_ntf_v5 or &struct iwl_alive_ntf_v7. */ UCODE_ALIVE_NTFY = 0x1, diff --git a/drivers/net/wireless/intel/iwlwifi/mld/fw.c b/drivers/net/wireless/intel/iwlwifi/mld/fw.c index b372173c4a79..bdb69c098fd1 100644 --- a/drivers/net/wireless/intel/iwlwifi/mld/fw.c +++ b/drivers/net/wireless/intel/iwlwifi/mld/fw.c @@ -126,7 +126,7 @@ static bool iwl_alive_fn(struct iwl_notif_wait_data *notif_wait, switch (version) { case 6: case 7: - expected_sz = sizeof(struct iwl_alive_ntf_v6); + expected_sz = sizeof(struct iwl_alive_ntf_v7); break; case 8: expected_sz = sizeof(struct iwl_alive_ntf); diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c index 865f973f677d..6b76ce35443d 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c @@ -115,7 +115,7 @@ static bool iwl_alive_fn(struct iwl_notif_wait_data *notif_wait, if (version >= 6) { - struct iwl_alive_ntf_v6 *palive; + struct iwl_alive_ntf_v7 *palive; if (pkt_len < sizeof(*palive)) return false; From 9dc6e9dfdf9d9efe5afd2322e450b83a449993db Mon Sep 17 00:00:00 2001 From: Miri Korenblit Date: Mon, 15 Sep 2025 11:34:23 +0300 Subject: [PATCH 0304/1381] wifi: iwlwifi: mld: remove support from of alive notif version 6 The last FW API that supports version 6 is 97. Since this API is no longer supported on any device that loads iwlmld, we can remove support of it. Reviewed-by: Emmanuel Grumbach Signed-off-by: Miri Korenblit Link: https://patch.msgid.link/20250915113137.4923c981b0bf.Iff598c6d109fdbf0d5a5bab59d53485478ecc125@changeid --- drivers/net/wireless/intel/iwlwifi/mld/fw.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/drivers/net/wireless/intel/iwlwifi/mld/fw.c b/drivers/net/wireless/intel/iwlwifi/mld/fw.c index bdb69c098fd1..b3abfa1ec810 100644 --- a/drivers/net/wireless/intel/iwlwifi/mld/fw.c +++ b/drivers/net/wireless/intel/iwlwifi/mld/fw.c @@ -124,7 +124,6 @@ static bool iwl_alive_fn(struct iwl_notif_wait_data *notif_wait, u16 status; switch (version) { - case 6: case 7: expected_sz = sizeof(struct iwl_alive_ntf_v7); break; @@ -188,9 +187,8 @@ static bool iwl_alive_fn(struct iwl_notif_wait_data *notif_wait, le32_to_cpu(umac->umac_major), le32_to_cpu(umac->umac_minor)); - if (version >= 7) - IWL_DEBUG_FW(mld, "FW alive flags 0x%x\n", - le16_to_cpu(palive->flags)); + IWL_DEBUG_FW(mld, "FW alive flags 0x%x\n", + le16_to_cpu(palive->flags)); if (version >= 8) IWL_DEBUG_FW(mld, "platform_id 0x%llx\n", From 5c0251598f29646a19a7232c9437b6faa22ab4c2 Mon Sep 17 00:00:00 2001 From: Miri Korenblit Date: Mon, 15 Sep 2025 11:34:24 +0300 Subject: [PATCH 0305/1381] wifi: iwlwifi: mld: reschedule check_tpt_wk also not in EMLSR When the throughput count reaches the threshold, EMLSR is no longer blocked by throughput. This doesn't mean that EMLSR will be activated immediately, since there might be other reasons that block EMLSR. When the throughput blocker is not set, check_tpt_wk should run every 5 seconds and check if the throughput blocker should be set (if the throughtput counter dropped). If not, it should reschedule itself. In the current code, the worker will reschedule itself only if we are in EMLSR. This is wrong, since we might be in a case where the throughput blocker is not set but we are not in EMLSR, and then we will never check again the throughput counters (and block EMLSR if needed). Fix this by rescheduling the worker also when EMLSR is not active. Reviewed-by: Johannes Berg Signed-off-by: Miri Korenblit Link: https://patch.msgid.link/20250915113137.2a9cf2b2529d.I8284c0da9597e4c963e38ae133384f6f42044499@changeid --- drivers/net/wireless/intel/iwlwifi/mld/mlo.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/intel/iwlwifi/mld/mlo.c b/drivers/net/wireless/intel/iwlwifi/mld/mlo.c index 241a6271d13d..fa04fbe06656 100644 --- a/drivers/net/wireless/intel/iwlwifi/mld/mlo.c +++ b/drivers/net/wireless/intel/iwlwifi/mld/mlo.c @@ -603,7 +603,7 @@ void iwl_mld_emlsr_check_tpt(struct wiphy *wiphy, struct wiphy_work *wk) /* EMLSR is not active */ if (sec_link_id == -1) - return; + goto schedule; IWL_DEBUG_INFO(mld, "Secondary Link %d: Tx MPDUs: %ld. Rx MPDUs: %ld\n", sec_link_id, sec_link_tx, sec_link_rx); @@ -625,6 +625,7 @@ void iwl_mld_emlsr_check_tpt(struct wiphy *wiphy, struct wiphy_work *wk) return; } +schedule: /* Check again when the next window ends */ wiphy_delayed_work_queue(mld_vif->mld->wiphy, &mld_vif->emlsr.check_tpt_wk, From eade5cacc95c3d8b2e88a694a6a0a42dc0dc1819 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Mon, 15 Sep 2025 11:34:25 +0300 Subject: [PATCH 0306/1381] wifi: iwlwifi: mvm: move rate conversions to utils.c These aren't really related to rate scaling, they're just firmware API functions. Try to keep rs.c more for scaling and move these out. Signed-off-by: Johannes Berg Signed-off-by: Miri Korenblit Link: https://patch.msgid.link/20250915113137.547129c7732e.I12c40876537722680d069b4bb5fc058206ba63d4@changeid --- drivers/net/wireless/intel/iwlwifi/mvm/mvm.h | 5 + drivers/net/wireless/intel/iwlwifi/mvm/rs.c | 164 ------------------ drivers/net/wireless/intel/iwlwifi/mvm/rs.h | 3 - .../net/wireless/intel/iwlwifi/mvm/utils.c | 164 ++++++++++++++++++ 4 files changed, 169 insertions(+), 167 deletions(-) diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h index b515028adc8f..301d590fe0bd 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h +++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h @@ -2894,4 +2894,9 @@ iwl_mvm_send_ap_tx_power_constraint_cmd(struct iwl_mvm *mvm, void iwl_mvm_smps_workaround(struct iwl_mvm *mvm, struct ieee80211_vif *vif, bool update); + +/* rate_n_flags conversion */ +u32 iwl_mvm_v3_rate_from_fw(__le32 rate, u8 rate_ver); +__le32 iwl_mvm_v3_rate_to_fw(u32 rate, u8 rate_ver); + #endif /* __IWL_MVM_H__ */ diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rs.c b/drivers/net/wireless/intel/iwlwifi/mvm/rs.c index 5802ed80a9ca..d1619a229d8f 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/rs.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/rs.c @@ -4178,167 +4178,3 @@ int iwl_mvm_tx_protection(struct iwl_mvm *mvm, struct iwl_mvm_sta *mvmsta, else return rs_drv_tx_protection(mvm, mvmsta, enable); } - -static u32 iwl_legacy_rate_to_fw_idx(u32 rate_n_flags) -{ - int rate = rate_n_flags & RATE_LEGACY_RATE_MSK_V1; - int idx; - bool ofdm = !(rate_n_flags & RATE_MCS_CCK_MSK_V1); - int offset = ofdm ? IWL_FIRST_OFDM_RATE : 0; - int last = ofdm ? IWL_RATE_COUNT_LEGACY : IWL_FIRST_OFDM_RATE; - - for (idx = offset; idx < last; idx++) - if (iwl_fw_rate_idx_to_plcp(idx) == rate) - return idx - offset; - return IWL_RATE_INVALID; -} - -u32 iwl_mvm_v3_rate_from_fw(__le32 rate, u8 rate_ver) -{ - u32 rate_v3 = 0, rate_v1; - u32 dup = 0; - - if (rate_ver > 1) - return iwl_v3_rate_from_v2_v3(rate, rate_ver >= 3); - - rate_v1 = le32_to_cpu(rate); - if (rate_v1 == 0) - return rate_v1; - /* convert rate */ - if (rate_v1 & RATE_MCS_HT_MSK_V1) { - u32 nss; - - rate_v3 |= RATE_MCS_MOD_TYPE_HT; - rate_v3 |= - rate_v1 & RATE_HT_MCS_RATE_CODE_MSK_V1; - nss = u32_get_bits(rate_v1, RATE_HT_MCS_MIMO2_MSK); - rate_v3 |= u32_encode_bits(nss, RATE_MCS_NSS_MSK); - } else if (rate_v1 & RATE_MCS_VHT_MSK_V1 || - rate_v1 & RATE_MCS_HE_MSK_V1) { - u32 nss = u32_get_bits(rate_v1, RATE_VHT_MCS_NSS_MSK); - - rate_v3 |= rate_v1 & RATE_VHT_MCS_RATE_CODE_MSK; - - rate_v3 |= u32_encode_bits(nss, RATE_MCS_NSS_MSK); - - if (rate_v1 & RATE_MCS_HE_MSK_V1) { - u32 he_type_bits = rate_v1 & RATE_MCS_HE_TYPE_MSK_V1; - u32 he_type = he_type_bits >> RATE_MCS_HE_TYPE_POS_V1; - u32 he_106t = (rate_v1 & RATE_MCS_HE_106T_MSK_V1) >> - RATE_MCS_HE_106T_POS_V1; - u32 he_gi_ltf = (rate_v1 & RATE_MCS_HE_GI_LTF_MSK_V1) >> - RATE_MCS_HE_GI_LTF_POS; - - if ((he_type_bits == RATE_MCS_HE_TYPE_SU || - he_type_bits == RATE_MCS_HE_TYPE_EXT_SU) && - he_gi_ltf == RATE_MCS_HE_SU_4_LTF) - /* the new rate have an additional bit to - * represent the value 4 rather then using SGI - * bit for this purpose - as it was done in the - * old rate - */ - he_gi_ltf += (rate_v1 & RATE_MCS_SGI_MSK_V1) >> - RATE_MCS_SGI_POS_V1; - - rate_v3 |= he_gi_ltf << RATE_MCS_HE_GI_LTF_POS; - rate_v3 |= he_type << RATE_MCS_HE_TYPE_POS; - rate_v3 |= he_106t << RATE_MCS_HE_106T_POS; - rate_v3 |= rate_v1 & RATE_HE_DUAL_CARRIER_MODE_MSK; - rate_v3 |= RATE_MCS_MOD_TYPE_HE; - } else { - rate_v3 |= RATE_MCS_MOD_TYPE_VHT; - } - /* if legacy format */ - } else { - u32 legacy_rate = iwl_legacy_rate_to_fw_idx(rate_v1); - - if (WARN_ON_ONCE(legacy_rate == IWL_RATE_INVALID)) - legacy_rate = (rate_v1 & RATE_MCS_CCK_MSK_V1) ? - IWL_FIRST_CCK_RATE : IWL_FIRST_OFDM_RATE; - - rate_v3 |= legacy_rate; - if (!(rate_v1 & RATE_MCS_CCK_MSK_V1)) - rate_v3 |= RATE_MCS_MOD_TYPE_LEGACY_OFDM; - } - - /* convert flags */ - if (rate_v1 & RATE_MCS_LDPC_MSK_V1) - rate_v3 |= RATE_MCS_LDPC_MSK; - rate_v3 |= (rate_v1 & RATE_MCS_CHAN_WIDTH_MSK_V1) | - (rate_v1 & RATE_MCS_ANT_AB_MSK) | - (rate_v1 & RATE_MCS_STBC_MSK) | - (rate_v1 & RATE_MCS_BF_MSK); - - dup = (rate_v1 & RATE_MCS_DUP_MSK_V1) >> RATE_MCS_DUP_POS_V1; - if (dup) { - rate_v3 |= RATE_MCS_DUP_MSK; - rate_v3 |= dup << RATE_MCS_CHAN_WIDTH_POS; - } - - if ((!(rate_v1 & RATE_MCS_HE_MSK_V1)) && - (rate_v1 & RATE_MCS_SGI_MSK_V1)) - rate_v3 |= RATE_MCS_SGI_MSK; - - return rate_v3; -} - -__le32 iwl_mvm_v3_rate_to_fw(u32 rate, u8 rate_ver) -{ - u32 result = 0; - int rate_idx; - - if (rate_ver > 1) - return iwl_v3_rate_to_v2_v3(rate, rate_ver > 2); - - switch (rate & RATE_MCS_MOD_TYPE_MSK) { - case RATE_MCS_MOD_TYPE_CCK: - result = RATE_MCS_CCK_MSK_V1; - fallthrough; - case RATE_MCS_MOD_TYPE_LEGACY_OFDM: - rate_idx = u32_get_bits(rate, RATE_LEGACY_RATE_MSK); - if (!(result & RATE_MCS_CCK_MSK_V1)) - rate_idx += IWL_FIRST_OFDM_RATE; - result |= u32_encode_bits(iwl_fw_rate_idx_to_plcp(rate_idx), - RATE_LEGACY_RATE_MSK_V1); - break; - case RATE_MCS_MOD_TYPE_HT: - result = RATE_MCS_HT_MSK_V1; - result |= u32_encode_bits(u32_get_bits(rate, - RATE_HT_MCS_CODE_MSK), - RATE_HT_MCS_RATE_CODE_MSK_V1); - result |= u32_encode_bits(u32_get_bits(rate, - RATE_MCS_NSS_MSK), - RATE_HT_MCS_MIMO2_MSK); - break; - case RATE_MCS_MOD_TYPE_VHT: - result = RATE_MCS_VHT_MSK_V1; - result |= u32_encode_bits(u32_get_bits(rate, - RATE_VHT_MCS_NSS_MSK), - RATE_MCS_CODE_MSK); - result |= u32_encode_bits(u32_get_bits(rate, RATE_MCS_NSS_MSK), - RATE_VHT_MCS_NSS_MSK); - break; - case RATE_MCS_MOD_TYPE_HE: /* not generated */ - default: - WARN_ONCE(1, "bad modulation type %d\n", - u32_get_bits(rate, RATE_MCS_MOD_TYPE_MSK)); - return 0; - } - - if (rate & RATE_MCS_LDPC_MSK) - result |= RATE_MCS_LDPC_MSK_V1; - WARN_ON_ONCE(u32_get_bits(rate, RATE_MCS_CHAN_WIDTH_MSK) > - RATE_MCS_CHAN_WIDTH_160_VAL); - result |= (rate & RATE_MCS_CHAN_WIDTH_MSK_V1) | - (rate & RATE_MCS_ANT_AB_MSK) | - (rate & RATE_MCS_STBC_MSK) | - (rate & RATE_MCS_BF_MSK); - - /* not handling DUP since we don't use it */ - WARN_ON_ONCE(rate & RATE_MCS_DUP_MSK); - - if (rate & RATE_MCS_SGI_MSK) - result |= RATE_MCS_SGI_MSK_V1; - - return cpu_to_le32(result); -} diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rs.h b/drivers/net/wireless/intel/iwlwifi/mvm/rs.h index dfb062b7c5c2..34c957bef6f8 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/rs.h +++ b/drivers/net/wireless/intel/iwlwifi/mvm/rs.h @@ -425,9 +425,6 @@ void iwl_mvm_rate_control_unregister(void); struct iwl_mvm_sta; -u32 iwl_mvm_v3_rate_from_fw(__le32 rate, u8 rate_ver); -__le32 iwl_mvm_v3_rate_to_fw(u32 rate, u8 rate_ver); - int iwl_mvm_tx_protection(struct iwl_mvm *mvm, struct iwl_mvm_sta *mvmsta, bool enable); diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/utils.c b/drivers/net/wireless/intel/iwlwifi/mvm/utils.c index 22602c32faa5..aa7af04d914e 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/utils.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/utils.c @@ -1237,3 +1237,167 @@ bool iwl_mvm_vif_is_active(struct iwl_mvm_vif *mvmvif) return false; } + +static u32 iwl_legacy_rate_to_fw_idx(u32 rate_n_flags) +{ + int rate = rate_n_flags & RATE_LEGACY_RATE_MSK_V1; + int idx; + bool ofdm = !(rate_n_flags & RATE_MCS_CCK_MSK_V1); + int offset = ofdm ? IWL_FIRST_OFDM_RATE : 0; + int last = ofdm ? IWL_RATE_COUNT_LEGACY : IWL_FIRST_OFDM_RATE; + + for (idx = offset; idx < last; idx++) + if (iwl_fw_rate_idx_to_plcp(idx) == rate) + return idx - offset; + return IWL_RATE_INVALID; +} + +u32 iwl_mvm_v3_rate_from_fw(__le32 rate, u8 rate_ver) +{ + u32 rate_v3 = 0, rate_v1; + u32 dup = 0; + + if (rate_ver > 1) + return iwl_v3_rate_from_v2_v3(rate, rate_ver >= 3); + + rate_v1 = le32_to_cpu(rate); + if (rate_v1 == 0) + return rate_v1; + /* convert rate */ + if (rate_v1 & RATE_MCS_HT_MSK_V1) { + u32 nss; + + rate_v3 |= RATE_MCS_MOD_TYPE_HT; + rate_v3 |= + rate_v1 & RATE_HT_MCS_RATE_CODE_MSK_V1; + nss = u32_get_bits(rate_v1, RATE_HT_MCS_MIMO2_MSK); + rate_v3 |= u32_encode_bits(nss, RATE_MCS_NSS_MSK); + } else if (rate_v1 & RATE_MCS_VHT_MSK_V1 || + rate_v1 & RATE_MCS_HE_MSK_V1) { + u32 nss = u32_get_bits(rate_v1, RATE_VHT_MCS_NSS_MSK); + + rate_v3 |= rate_v1 & RATE_VHT_MCS_RATE_CODE_MSK; + + rate_v3 |= u32_encode_bits(nss, RATE_MCS_NSS_MSK); + + if (rate_v1 & RATE_MCS_HE_MSK_V1) { + u32 he_type_bits = rate_v1 & RATE_MCS_HE_TYPE_MSK_V1; + u32 he_type = he_type_bits >> RATE_MCS_HE_TYPE_POS_V1; + u32 he_106t = (rate_v1 & RATE_MCS_HE_106T_MSK_V1) >> + RATE_MCS_HE_106T_POS_V1; + u32 he_gi_ltf = (rate_v1 & RATE_MCS_HE_GI_LTF_MSK_V1) >> + RATE_MCS_HE_GI_LTF_POS; + + if ((he_type_bits == RATE_MCS_HE_TYPE_SU || + he_type_bits == RATE_MCS_HE_TYPE_EXT_SU) && + he_gi_ltf == RATE_MCS_HE_SU_4_LTF) + /* the new rate have an additional bit to + * represent the value 4 rather then using SGI + * bit for this purpose - as it was done in the + * old rate + */ + he_gi_ltf += (rate_v1 & RATE_MCS_SGI_MSK_V1) >> + RATE_MCS_SGI_POS_V1; + + rate_v3 |= he_gi_ltf << RATE_MCS_HE_GI_LTF_POS; + rate_v3 |= he_type << RATE_MCS_HE_TYPE_POS; + rate_v3 |= he_106t << RATE_MCS_HE_106T_POS; + rate_v3 |= rate_v1 & RATE_HE_DUAL_CARRIER_MODE_MSK; + rate_v3 |= RATE_MCS_MOD_TYPE_HE; + } else { + rate_v3 |= RATE_MCS_MOD_TYPE_VHT; + } + /* if legacy format */ + } else { + u32 legacy_rate = iwl_legacy_rate_to_fw_idx(rate_v1); + + if (WARN_ON_ONCE(legacy_rate == IWL_RATE_INVALID)) + legacy_rate = (rate_v1 & RATE_MCS_CCK_MSK_V1) ? + IWL_FIRST_CCK_RATE : IWL_FIRST_OFDM_RATE; + + rate_v3 |= legacy_rate; + if (!(rate_v1 & RATE_MCS_CCK_MSK_V1)) + rate_v3 |= RATE_MCS_MOD_TYPE_LEGACY_OFDM; + } + + /* convert flags */ + if (rate_v1 & RATE_MCS_LDPC_MSK_V1) + rate_v3 |= RATE_MCS_LDPC_MSK; + rate_v3 |= (rate_v1 & RATE_MCS_CHAN_WIDTH_MSK_V1) | + (rate_v1 & RATE_MCS_ANT_AB_MSK) | + (rate_v1 & RATE_MCS_STBC_MSK) | + (rate_v1 & RATE_MCS_BF_MSK); + + dup = (rate_v1 & RATE_MCS_DUP_MSK_V1) >> RATE_MCS_DUP_POS_V1; + if (dup) { + rate_v3 |= RATE_MCS_DUP_MSK; + rate_v3 |= dup << RATE_MCS_CHAN_WIDTH_POS; + } + + if ((!(rate_v1 & RATE_MCS_HE_MSK_V1)) && + (rate_v1 & RATE_MCS_SGI_MSK_V1)) + rate_v3 |= RATE_MCS_SGI_MSK; + + return rate_v3; +} + +__le32 iwl_mvm_v3_rate_to_fw(u32 rate, u8 rate_ver) +{ + u32 result = 0; + int rate_idx; + + if (rate_ver > 1) + return iwl_v3_rate_to_v2_v3(rate, rate_ver > 2); + + switch (rate & RATE_MCS_MOD_TYPE_MSK) { + case RATE_MCS_MOD_TYPE_CCK: + result = RATE_MCS_CCK_MSK_V1; + fallthrough; + case RATE_MCS_MOD_TYPE_LEGACY_OFDM: + rate_idx = u32_get_bits(rate, RATE_LEGACY_RATE_MSK); + if (!(result & RATE_MCS_CCK_MSK_V1)) + rate_idx += IWL_FIRST_OFDM_RATE; + result |= u32_encode_bits(iwl_fw_rate_idx_to_plcp(rate_idx), + RATE_LEGACY_RATE_MSK_V1); + break; + case RATE_MCS_MOD_TYPE_HT: + result = RATE_MCS_HT_MSK_V1; + result |= u32_encode_bits(u32_get_bits(rate, + RATE_HT_MCS_CODE_MSK), + RATE_HT_MCS_RATE_CODE_MSK_V1); + result |= u32_encode_bits(u32_get_bits(rate, + RATE_MCS_NSS_MSK), + RATE_HT_MCS_MIMO2_MSK); + break; + case RATE_MCS_MOD_TYPE_VHT: + result = RATE_MCS_VHT_MSK_V1; + result |= u32_encode_bits(u32_get_bits(rate, + RATE_VHT_MCS_NSS_MSK), + RATE_MCS_CODE_MSK); + result |= u32_encode_bits(u32_get_bits(rate, RATE_MCS_NSS_MSK), + RATE_VHT_MCS_NSS_MSK); + break; + case RATE_MCS_MOD_TYPE_HE: /* not generated */ + default: + WARN_ONCE(1, "bad modulation type %d\n", + u32_get_bits(rate, RATE_MCS_MOD_TYPE_MSK)); + return 0; + } + + if (rate & RATE_MCS_LDPC_MSK) + result |= RATE_MCS_LDPC_MSK_V1; + WARN_ON_ONCE(u32_get_bits(rate, RATE_MCS_CHAN_WIDTH_MSK) > + RATE_MCS_CHAN_WIDTH_160_VAL); + result |= (rate & RATE_MCS_CHAN_WIDTH_MSK_V1) | + (rate & RATE_MCS_ANT_AB_MSK) | + (rate & RATE_MCS_STBC_MSK) | + (rate & RATE_MCS_BF_MSK); + + /* not handling DUP since we don't use it */ + WARN_ON_ONCE(rate & RATE_MCS_DUP_MSK); + + if (rate & RATE_MCS_SGI_MSK) + result |= RATE_MCS_SGI_MSK_V1; + + return cpu_to_le32(result); +} From ba85816979360b78a40a3dd0cb8e6f57d6b22456 Mon Sep 17 00:00:00 2001 From: Miri Korenblit Date: Mon, 15 Sep 2025 11:34:26 +0300 Subject: [PATCH 0307/1381] wifi: iwlwifi: iwlmld is always used for wifi7 devices iwlmld is used since API 97 and for wifi7 devices. Since APIs < 97 are no longer supported on such devices, we can remove the API check and always load iwlmld for the wifi7 devices. Reviewed-by: Emmanuel Grumbach Signed-off-by: Miri Korenblit Link: https://patch.msgid.link/20250915113137.45ab33fcdc00.Ia3a40b687b75c872cf7e7a19331a014bccf5f2d6@changeid --- drivers/net/wireless/intel/iwlwifi/iwl-drv.c | 18 +++++++----------- 1 file changed, 7 insertions(+), 11 deletions(-) diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-drv.c b/drivers/net/wireless/intel/iwlwifi/iwl-drv.c index 607fcea6f4ef..7d58e294618d 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-drv.c +++ b/drivers/net/wireless/intel/iwlwifi/iwl-drv.c @@ -179,7 +179,8 @@ static inline char iwl_drv_get_step(int step) static bool iwl_drv_is_wifi7_supported(struct iwl_trans *trans) { - return CSR_HW_RFID_TYPE(trans->info.hw_rf_id) >= IWL_CFG_RF_TYPE_FM; + return trans->mac_cfg->device_family >= IWL_DEVICE_FAMILY_BZ && + CSR_HW_RFID_TYPE(trans->info.hw_rf_id) >= IWL_CFG_RF_TYPE_FM; } const char *iwl_drv_get_fwname_pre(struct iwl_trans *trans, char *buf) @@ -427,7 +428,6 @@ struct iwl_firmware_pieces { size_t dbg_trigger_tlv_len[FW_DBG_TRIGGER_MAX]; struct iwl_fw_dbg_mem_seg_tlv *dbg_mem_tlv; size_t n_mem_tlv; - u32 major; }; static void alloc_sec_data(struct iwl_firmware_pieces *pieces, @@ -1069,19 +1069,19 @@ static int iwl_parse_tlv_firmware(struct iwl_drv *drv, break; case IWL_UCODE_TLV_FW_VERSION: { const __le32 *ptr = (const void *)tlv_data; - u32 minor; + u32 major, minor; u8 local_comp; if (tlv_len != sizeof(u32) * 3) goto invalid_tlv_len; - pieces->major = le32_to_cpup(ptr++); + major = le32_to_cpup(ptr++); minor = le32_to_cpup(ptr++); local_comp = le32_to_cpup(ptr); snprintf(drv->fw.fw_version, sizeof(drv->fw.fw_version), - "%u.%08x.%u %s", pieces->major, minor, + "%u.%08x.%u %s", major, minor, local_comp, iwl_reduced_fw_name(drv)); break; } @@ -1589,8 +1589,6 @@ static void _iwl_op_mode_stop(struct iwl_drv *drv) } } -#define IWL_MLD_SUPPORTED_FW_VERSION 97 - /* * iwl_req_fw_callback - callback when firmware was loaded * @@ -1859,12 +1857,10 @@ static void iwl_req_fw_callback(const struct firmware *ucode_raw, void *context) } #if IS_ENABLED(CONFIG_IWLMLD) - if (pieces->major >= IWL_MLD_SUPPORTED_FW_VERSION && - iwl_drv_is_wifi7_supported(drv->trans)) + if (iwl_drv_is_wifi7_supported(drv->trans)) op = &iwlwifi_opmode_table[MLD_OP_MODE]; #else - if (pieces->major >= IWL_MLD_SUPPORTED_FW_VERSION && - iwl_drv_is_wifi7_supported(drv->trans)) { + if (iwl_drv_is_wifi7_supported(drv->trans)) { IWL_ERR(drv, "IWLMLD needs to be compiled to support this firmware\n"); mutex_unlock(&iwlwifi_opmode_table_mtx); From 30d47d8fe781469ebd4e38240999767f139effb2 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Mon, 15 Sep 2025 11:34:27 +0300 Subject: [PATCH 0308/1381] wifi: iwlwifi: cfg: add new device names Add a couple of device names so that these new devices will be shown correctly. Signed-off-by: Johannes Berg Signed-off-by: Miri Korenblit Link: https://patch.msgid.link/20250915113137.1cbc0251532f.I6183a6a08a7998e598042a50c7d7a6b82f9fa58e@changeid --- drivers/net/wireless/intel/iwlwifi/cfg/rf-wh.c | 1 + drivers/net/wireless/intel/iwlwifi/iwl-config.h | 1 + drivers/net/wireless/intel/iwlwifi/pcie/drv.c | 3 +++ 3 files changed, 5 insertions(+) diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/rf-wh.c b/drivers/net/wireless/intel/iwlwifi/cfg/rf-wh.c index 97735175cb0e..b8c6b06e7099 100644 --- a/drivers/net/wireless/intel/iwlwifi/cfg/rf-wh.c +++ b/drivers/net/wireless/intel/iwlwifi/cfg/rf-wh.c @@ -13,3 +13,4 @@ const char iwl_killer_be1775i_name[] = const char iwl_be211_name[] = "Intel(R) Wi-Fi 7 BE211 320MHz"; const char iwl_be213_name[] = "Intel(R) Wi-Fi 7 BE213 160MHz"; +const char iwl_ax221_name[] = "Intel(R) Wi-Fi 6E AX221 160MHz"; diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-config.h b/drivers/net/wireless/intel/iwlwifi/iwl-config.h index a607e7ab914b..cfd1629f0016 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-config.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-config.h @@ -688,6 +688,7 @@ extern const char iwl_killer_bn1850i_name[]; extern const char iwl_bn201_name[]; extern const char iwl_be221_name[]; extern const char iwl_be223_name[]; +extern const char iwl_ax221_name[]; #if IS_ENABLED(CONFIG_IWLDVM) extern const struct iwl_rf_cfg iwl5300_agn_cfg; extern const struct iwl_rf_cfg iwl5350_agn_cfg; diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c index b21a4d8eb105..de04a84def0d 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c @@ -1061,11 +1061,14 @@ VISIBLE_IF_IWLWIFI_KUNIT const struct iwl_dev_info iwl_dev_info_table[] = { /* WH RF */ IWL_DEV_INFO(iwl_rf_wh, iwl_be211_name, RF_TYPE(WH)), + IWL_DEV_INFO(iwl_rf_wh, iwl_ax221_name, RF_TYPE(WH), SUBDEV(0x0514)), + IWL_DEV_INFO(iwl_rf_wh, iwl_ax221_name, RF_TYPE(WH), SUBDEV(0x4514)), IWL_DEV_INFO(iwl_rf_wh_160mhz, iwl_be213_name, RF_TYPE(WH), BW_LIMITED), /* PE RF */ IWL_DEV_INFO(iwl_rf_pe, iwl_bn201_name, RF_TYPE(PE)), IWL_DEV_INFO(iwl_rf_pe, iwl_be223_name, RF_TYPE(PE), SUBDEV(0x0524)), + IWL_DEV_INFO(iwl_rf_pe, iwl_be223_name, RF_TYPE(PE), SUBDEV(0x4524)), IWL_DEV_INFO(iwl_rf_pe, iwl_be221_name, RF_TYPE(PE), SUBDEV(0x0324)), /* Killer */ From ef56bbed4c03fc272c65da821d31094b63a8461e Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Mon, 15 Sep 2025 11:34:28 +0300 Subject: [PATCH 0309/1381] wifi: iwlwifi: tests: check listed PCI IDs have configs Add a test that checks, for the old pre-CNVI devices, that PCI IDs listed in the PCI IDs table will also match in the config table. Newer ones we test against our database of devices, but the current database doesn't go back that far, so at least this checks against the PCI IDs the driver has. Signed-off-by: Johannes Berg Signed-off-by: Miri Korenblit Link: https://patch.msgid.link/20250915113137.eb728b270d46.Ie5754f4201b807eb0d55feb339a728fc0b42e8bf@changeid --- drivers/net/wireless/intel/iwlwifi/cfg/bz.c | 1 + .../wireless/intel/iwlwifi/tests/devinfo.c | 27 +++++++++++++++++++ 2 files changed, 28 insertions(+) diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/bz.c b/drivers/net/wireless/intel/iwlwifi/cfg/bz.c index 3e6206e739f6..18c9244ee8ef 100644 --- a/drivers/net/wireless/intel/iwlwifi/cfg/bz.c +++ b/drivers/net/wireless/intel/iwlwifi/cfg/bz.c @@ -90,6 +90,7 @@ const struct iwl_mac_cfg iwl_bz_mac_cfg = { .low_latency_xtal = true, .ltr_delay = IWL_CFG_TRANS_LTR_DELAY_2500US, }; +EXPORT_SYMBOL_IF_IWLWIFI_KUNIT(iwl_bz_mac_cfg); const struct iwl_mac_cfg iwl_gl_mac_cfg = { .device_family = IWL_DEVICE_FAMILY_BZ, diff --git a/drivers/net/wireless/intel/iwlwifi/tests/devinfo.c b/drivers/net/wireless/intel/iwlwifi/tests/devinfo.c index c31bbd4e7a4a..3054bc0a9c67 100644 --- a/drivers/net/wireless/intel/iwlwifi/tests/devinfo.c +++ b/drivers/net/wireless/intel/iwlwifi/tests/devinfo.c @@ -265,6 +265,32 @@ static void devinfo_api_range(struct kunit *test) } } +static void devinfo_pci_ids_config(struct kunit *test) +{ + for (int i = 0; iwl_hw_card_ids[i].vendor; i++) { + const struct pci_device_id *s = &iwl_hw_card_ids[i]; + const struct iwl_dev_info *di; + + if (s->device == PCI_ANY_ID || s->subdevice == PCI_ANY_ID) + continue; + + /* + * The check below only works for old (pre-CNVI) devices. Most + * new have subdevice==ANY, so are already skipped, but for some + * Bz platform(s) we list all the RF PCI IDs. Skip those too. + */ + if (s->driver_data == (kernel_ulong_t)&iwl_bz_mac_cfg) + continue; + + di = iwl_pci_find_dev_info(s->device, s->subdevice, + 0, 0, 0, 0, true); + + KUNIT_EXPECT_PTR_NE_MSG(test, di, NULL, + "PCI ID %04x:%04x not found\n", + s->device, s->subdevice); + } +} + static struct kunit_case devinfo_test_cases[] = { KUNIT_CASE(devinfo_table_order), KUNIT_CASE(devinfo_discrete_match), @@ -276,6 +302,7 @@ static struct kunit_case devinfo_test_cases[] = { KUNIT_CASE(devinfo_pci_ids), KUNIT_CASE(devinfo_no_mac_cfg_dups), KUNIT_CASE(devinfo_api_range), + KUNIT_CASE(devinfo_pci_ids_config), {} }; From 58a4ebe3168813a04bef08f7858a63b199e866e1 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Mon, 15 Sep 2025 11:34:29 +0300 Subject: [PATCH 0310/1381] wifi: iwlwifi: fix remaining kernel-doc warnings Fix the remaining kernel-doc warnings across the driver. Signed-off-by: Johannes Berg Signed-off-by: Miri Korenblit Link: https://patch.msgid.link/20250915113137.f94b6d4ef142.I91007eed4cf37125ca7a012f2021615b4fa9eb66@changeid --- drivers/net/wireless/intel/iwlwifi/fw/acpi.h | 1 + .../wireless/intel/iwlwifi/fw/api/cmdhdr.h | 4 +- .../net/wireless/intel/iwlwifi/fw/api/coex.h | 4 +- .../wireless/intel/iwlwifi/fw/api/dbg-tlv.h | 14 ++-- .../net/wireless/intel/iwlwifi/fw/api/debug.h | 2 +- .../wireless/intel/iwlwifi/fw/api/location.h | 8 +- .../net/wireless/intel/iwlwifi/fw/api/power.h | 5 +- .../net/wireless/intel/iwlwifi/fw/api/scan.h | 78 ++++++++++--------- .../net/wireless/intel/iwlwifi/fw/api/sta.h | 6 +- .../net/wireless/intel/iwlwifi/fw/api/stats.h | 39 +++++----- .../net/wireless/intel/iwlwifi/fw/api/tx.h | 2 +- .../wireless/intel/iwlwifi/fw/error-dump.h | 4 +- drivers/net/wireless/intel/iwlwifi/fw/file.h | 74 +++++++++++++----- drivers/net/wireless/intel/iwlwifi/fw/img.h | 12 ++- .../net/wireless/intel/iwlwifi/fw/runtime.h | 22 +++++- .../net/wireless/intel/iwlwifi/iwl-config.h | 2 +- .../net/wireless/intel/iwlwifi/iwl-dbg-tlv.h | 4 +- drivers/net/wireless/intel/iwlwifi/iwl-drv.h | 3 +- .../wireless/intel/iwlwifi/iwl-modparams.h | 4 +- .../wireless/intel/iwlwifi/iwl-nvm-parse.h | 17 ++-- .../net/wireless/intel/iwlwifi/iwl-op-mode.h | 1 + .../net/wireless/intel/iwlwifi/iwl-trans.h | 6 +- 22 files changed, 194 insertions(+), 118 deletions(-) diff --git a/drivers/net/wireless/intel/iwlwifi/fw/acpi.h b/drivers/net/wireless/intel/iwlwifi/fw/acpi.h index 20bc6671f4eb..06cece4ea6d9 100644 --- a/drivers/net/wireless/intel/iwlwifi/fw/acpi.h +++ b/drivers/net/wireless/intel/iwlwifi/fw/acpi.h @@ -151,6 +151,7 @@ union acpi_object *iwl_acpi_get_dsm_object(struct device *dev, int rev, * @mcc: output buffer (3 bytes) that will get the MCC * * This function tries to read the current MCC from ACPI if available. + * Return: 0 on success, or a negative error code */ int iwl_acpi_get_mcc(struct iwl_fw_runtime *fwrt, char *mcc); diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/cmdhdr.h b/drivers/net/wireless/intel/iwlwifi/fw/api/cmdhdr.h index d130d4f85444..073f003bdc5d 100644 --- a/drivers/net/wireless/intel/iwlwifi/fw/api/cmdhdr.h +++ b/drivers/net/wireless/intel/iwlwifi/fw/api/cmdhdr.h @@ -1,6 +1,6 @@ /* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */ /* - * Copyright (C) 2005-2014 Intel Corporation + * Copyright (C) 2005-2014, 2025 Intel Corporation * Copyright (C) 2013-2015 Intel Mobile Communications GmbH * Copyright (C) 2016-2017 Intel Deutschland GmbH */ @@ -98,7 +98,7 @@ struct iwl_cmd_header { } __packed; /** - * struct iwl_cmd_header_wide + * struct iwl_cmd_header_wide - wide command header * * This header format appears in the beginning of each command sent from the * driver, and each response/notification received from uCode. diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/coex.h b/drivers/net/wireless/intel/iwlwifi/fw/api/coex.h index ddc84430d895..616f00a8b603 100644 --- a/drivers/net/wireless/intel/iwlwifi/fw/api/coex.h +++ b/drivers/net/wireless/intel/iwlwifi/fw/api/coex.h @@ -1,6 +1,6 @@ /* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */ /* - * Copyright (C) 2023-2024 Intel Corporation + * Copyright (C) 2023-2025 Intel Corporation * Copyright (C) 2013-2014, 2018-2019 Intel Corporation * Copyright (C) 2013-2014 Intel Mobile Communications GmbH * Copyright (C) 2017 Intel Deutschland GmbH @@ -52,7 +52,7 @@ struct iwl_bt_coex_cmd { } __packed; /* BT_COEX_CMD_API_S_VER_6 */ /** - * struct iwl_bt_coex_reduced_txp_update_cmd + * struct iwl_bt_coex_reduced_txp_update_cmd - reduced TX power command * @reduced_txp: bit BT_REDUCED_TX_POWER_BIT to enable / disable, rest of the * bits are the sta_id (value) */ diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/dbg-tlv.h b/drivers/net/wireless/intel/iwlwifi/fw/api/dbg-tlv.h index 3173fa96cb48..b62f0687327a 100644 --- a/drivers/net/wireless/intel/iwlwifi/fw/api/dbg-tlv.h +++ b/drivers/net/wireless/intel/iwlwifi/fw/api/dbg-tlv.h @@ -16,7 +16,7 @@ #define IWL_FW_INI_PRESET_DISABLE 0xff /** - * struct iwl_fw_ini_hcmd + * struct iwl_fw_ini_hcmd - debug configuration host command * * @id: the debug configuration command type for instance: 0xf6 / 0xf5 / DHC * @group: the desired cmd group @@ -199,7 +199,7 @@ struct iwl_fw_ini_region_tlv { } __packed; /* FW_TLV_DEBUG_REGION_API_S_VER_1 */ /** - * struct iwl_fw_ini_debug_info_tlv + * struct iwl_fw_ini_debug_info_tlv - debug info TLV * * debug configuration name for a specific image * @@ -311,7 +311,7 @@ struct iwl_fw_ini_conf_set_tlv { } __packed; /* FW_TLV_DEBUG_CONFIG_SET_API_S_VER_1 */ /** - * enum iwl_fw_ini_config_set_type + * enum iwl_fw_ini_config_set_type - configuration set type * * @IWL_FW_INI_CONFIG_SET_TYPE_INVALID: invalid config set * @IWL_FW_INI_CONFIG_SET_TYPE_DEVICE_PERIPHERY_MAC: for PERIPHERY MAC configuration @@ -337,7 +337,7 @@ enum iwl_fw_ini_config_set_type { } __packed; /** - * enum iwl_fw_ini_allocation_id + * enum iwl_fw_ini_allocation_id - allocation ID * * @IWL_FW_INI_ALLOCATION_INVALID: invalid * @IWL_FW_INI_ALLOCATION_ID_DBGC1: allocation meant for DBGC1 configuration @@ -356,7 +356,7 @@ enum iwl_fw_ini_allocation_id { }; /* FW_DEBUG_TLV_ALLOCATION_ID_E_VER_1 */ /** - * enum iwl_fw_ini_buffer_location + * enum iwl_fw_ini_buffer_location - buffer location * * @IWL_FW_INI_LOCATION_INVALID: invalid * @IWL_FW_INI_LOCATION_SRAM_PATH: SRAM location @@ -373,7 +373,7 @@ enum iwl_fw_ini_buffer_location { }; /* FW_DEBUG_TLV_BUFFER_LOCATION_E_VER_1 */ /** - * enum iwl_fw_ini_region_type + * enum iwl_fw_ini_region_type - region type * * @IWL_FW_INI_REGION_INVALID: invalid * @IWL_FW_INI_REGION_TLV: uCode and debug TLVs @@ -437,7 +437,7 @@ enum iwl_fw_ini_region_device_memory_subtype { }; /* FW_TLV_DEBUG_REGION_DEVICE_MEMORY_SUBTYPE_API_E */ /** - * enum iwl_fw_ini_time_point + * enum iwl_fw_ini_time_point - time point type * * Hard coded time points in which the driver can send hcmd or perform dump * collection diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/debug.h b/drivers/net/wireless/intel/iwlwifi/fw/api/debug.h index 0cf1e5124fba..61a850de26fc 100644 --- a/drivers/net/wireless/intel/iwlwifi/fw/api/debug.h +++ b/drivers/net/wireless/intel/iwlwifi/fw/api/debug.h @@ -421,7 +421,7 @@ struct iwl_dbgc1_info { } __packed; /* INIT_DRAM_FRAGS_ALLOCATIONS_S_VER_1 */ /** - * struct iwl_dbg_host_event_cfg_cmd + * struct iwl_dbg_host_event_cfg_cmd - host event config command * @enabled_severities: enabled severities */ struct iwl_dbg_host_event_cfg_cmd { diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/location.h b/drivers/net/wireless/intel/iwlwifi/fw/api/location.h index 33541f92c7c7..2ee3a48aa5df 100644 --- a/drivers/net/wireless/intel/iwlwifi/fw/api/location.h +++ b/drivers/net/wireless/intel/iwlwifi/fw/api/location.h @@ -1092,7 +1092,7 @@ struct iwl_tof_range_req_ap_entry { } __packed; /* LOCATION_RANGE_REQ_AP_ENTRY_CMD_API_S_VER_9 */ /** - * enum iwl_tof_response_mode + * enum iwl_tof_response_mode - TOF response mode * @IWL_MVM_TOF_RESPONSE_ASAP: report each AP measurement separately as soon as * possible (not supported for this release) * @IWL_MVM_TOF_RESPONSE_TIMEOUT: report all AP measurements as a batch upon @@ -1108,7 +1108,7 @@ enum iwl_tof_response_mode { }; /** - * enum iwl_tof_initiator_flags + * enum iwl_tof_initiator_flags - TOF initiator flags * * @IWL_TOF_INITIATOR_FLAGS_FAST_ALGO_DISABLED: disable fast algo, meaning run * the algo on ant A+B, instead of only one of them. @@ -1409,7 +1409,7 @@ enum iwl_tof_range_request_status { }; /** - * enum iwl_tof_entry_status + * enum iwl_tof_entry_status - TOF entry status * * @IWL_TOF_ENTRY_SUCCESS: successful measurement. * @IWL_TOF_ENTRY_GENERAL_FAILURE: General failure. @@ -1856,7 +1856,7 @@ struct iwl_tof_mcsi_notif { } __packed; /** - * struct iwl_tof_range_abort_cmd + * struct iwl_tof_range_abort_cmd - TOF range abort command * @request_id: corresponds to a range request * @reserved: reserved */ diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/power.h b/drivers/net/wireless/intel/iwlwifi/fw/api/power.h index 5eb8d10678fd..535864e22626 100644 --- a/drivers/net/wireless/intel/iwlwifi/fw/api/power.h +++ b/drivers/net/wireless/intel/iwlwifi/fw/api/power.h @@ -620,7 +620,7 @@ struct iwl_sar_offset_mapping_cmd { } __packed; /*SAR_OFFSET_MAPPING_TABLE_CMD_API_S*/ /** - * struct iwl_beacon_filter_cmd + * struct iwl_beacon_filter_cmd - beacon filter command * REPLY_BEACON_FILTERING_CMD = 0xd2 (command) * @bf_energy_delta: Used for RSSI filtering, if in 'normal' state. Send beacon * to driver if delta in Energy values calculated for this and last @@ -762,7 +762,7 @@ enum iwl_6ghz_ap_type { }; /* PHY_AP_TYPE_API_E_VER_1 */ /** - * struct iwl_txpower_constraints_cmd + * struct iwl_txpower_constraints_cmd - TX power constraints command * AP_TX_POWER_CONSTRAINTS_CMD * Used for VLP/LPI/AFC Access Point power constraints for 6GHz channels * @link_id: linkId @@ -786,4 +786,5 @@ struct iwl_txpower_constraints_cmd { __s8 psd_pwr[IWL_MAX_TX_EIRP_PSD_PWR_MAX_SIZE]; u8 reserved[3]; } __packed; /* PHY_AP_TX_POWER_CONSTRAINTS_CMD_API_S_VER_1 */ + #endif /* __iwl_fw_api_power_h__ */ diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/scan.h b/drivers/net/wireless/intel/iwlwifi/fw/api/scan.h index f486d624500b..60f0a4924ddf 100644 --- a/drivers/net/wireless/intel/iwlwifi/fw/api/scan.h +++ b/drivers/net/wireless/intel/iwlwifi/fw/api/scan.h @@ -1,6 +1,6 @@ /* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */ /* - * Copyright (C) 2012-2014, 2018-2024 Intel Corporation + * Copyright (C) 2012-2014, 2018-2025 Intel Corporation * Copyright (C) 2013-2015 Intel Mobile Communications GmbH * Copyright (C) 2016-2017 Intel Deutschland GmbH */ @@ -129,7 +129,7 @@ struct iwl_scan_offload_profile { } __packed; /** - * struct iwl_scan_offload_profile_cfg_data + * struct iwl_scan_offload_profile_cfg_data - scan offload profile configs * @blocklist_len: length of blocklist * @num_profiles: num of profiles in the list * @match_notify: clients waiting for match found notification @@ -159,7 +159,7 @@ struct iwl_scan_offload_profile_cfg_v1 { } __packed; /* SCAN_OFFLOAD_PROFILES_CFG_API_S_VER_1-2*/ /** - * struct iwl_scan_offload_profile_cfg + * struct iwl_scan_offload_profile_cfg - scan offload profile config * @profiles: profiles to search for match * @data: the rest of the data for profile_cfg */ @@ -507,7 +507,7 @@ enum iwl_uhb_chan_cfg_flags { IWL_UHB_CHAN_CFG_FLAG_FORCE_PASSIVE = BIT(26), }; /** - * struct iwl_scan_dwell + * struct iwl_scan_dwell - scan dwell configuration * @active: default dwell time for active scan * @passive: default dwell time for passive scan * @fragmented: default dwell time for fragmented scan @@ -728,7 +728,7 @@ enum iwl_umac_scan_general_params_flags2 { }; /** - * struct iwl_scan_channel_cfg_umac + * struct iwl_scan_channel_cfg_umac - scan channel config * @flags: bitmap - 0-19: directed scan to i'th ssid. * @channel_num: channel number 1-13 etc. * @v1: command version 1 @@ -774,7 +774,7 @@ struct iwl_scan_channel_cfg_umac { } __packed; /** - * struct iwl_scan_umac_schedule + * struct iwl_scan_umac_schedule - scan schedule parameters * @interval: interval in seconds between scan iterations * @iter_count: num of scan iterations for schedule plan, 0xff for infinite loop * @reserved: for alignment and future use @@ -815,7 +815,7 @@ struct iwl_scan_req_umac_tail_v2 { } __packed; /** - * struct iwl_scan_umac_chan_param + * struct iwl_scan_umac_chan_param - scan channel parameters * @flags: channel flags &enum iwl_scan_channel_flags * @count: num of channels in scan request * @reserved: for future use and alignment @@ -827,33 +827,37 @@ struct iwl_scan_umac_chan_param { } __packed; /*SCAN_CHANNEL_PARAMS_API_S_VER_1 */ /** - * struct iwl_scan_req_umac + * struct iwl_scan_req_umac - scan request command * @flags: &enum iwl_umac_scan_flags * @uid: scan id, &enum iwl_umac_scan_uid_offsets * @ooc_priority: out of channel priority - &enum iwl_scan_priority * @general_flags: &enum iwl_umac_scan_general_flags + * @reserved: reserved * @scan_start_mac_id: report the scan start TSF time according to this mac TSF - * @extended_dwell: dwell time for channels 1, 6 and 11 - * @active_dwell: dwell time for active scan per LMAC - * @passive_dwell: dwell time for passive scan per LMAC - * @fragmented_dwell: dwell time for fragmented passive scan - * @adwell_default_n_aps: for adaptive dwell the default number of APs + * @v1: version 1 command data + * @v6: version 6 command data + * @v7: version 7 command data + * @v8: version 8 command data + * @v9: version 9 command data + * @v1.extended_dwell: dwell time for channels 1, 6 and 11 + * @v1.active_dwell: dwell time for active scan per LMAC + * @v1.passive_dwell: dwell time for passive scan per LMAC + * @v1.fragmented_dwell: dwell time for fragmented passive scan + * @v7.adwell_default_n_aps: for adaptive dwell the default number of APs * per channel - * @adwell_default_n_aps_social: for adaptive dwell the default + * @v7.adwell_default_n_aps_social: for adaptive dwell the default * number of APs per social (1,6,11) channel - * @general_flags2: &enum iwl_umac_scan_general_flags2 - * @adwell_max_budget: for adaptive dwell the maximal budget of TU to be added - * to total scan time - * @max_out_time: max out of serving channel time, per LMAC - for CDB there - * are 2 LMACs - * @suspend_time: max suspend time, per LMAC - for CDB there are 2 LMACs - * @scan_priority: scan internal prioritization &enum iwl_scan_priority - * @num_of_fragments: Number of fragments needed for full coverage per band. + * @v8.general_flags2: &enum iwl_umac_scan_general_flags2 + * @v7.adwell_max_budget: for adaptive dwell the maximal budget of TU to be + * added to total scan time + * @v1.max_out_time: max out of serving channel time, per LMAC - for CDB + * there are 2 LMACs + * @v1.suspend_time: max suspend time, per LMAC - for CDB there are 2 LMACs + * @v1.scan_priority: scan internal prioritization &enum iwl_scan_priority + * @v8.num_of_fragments: Number of fragments needed for full coverage per band. * Relevant only for fragmented scan. - * @channel: &struct iwl_scan_umac_chan_param - * @reserved: for future use and alignment - * @reserved3: for future use and alignment - * @data: &struct iwl_scan_channel_cfg_umac and + * @v1.channel: &struct iwl_scan_umac_chan_param + * @v1.data: &struct iwl_scan_channel_cfg_umac and * &struct iwl_scan_req_umac_tail */ struct iwl_scan_req_umac { @@ -939,7 +943,7 @@ struct iwl_scan_req_umac { #define IWL_SCAN_REQ_UMAC_SIZE_V1 36 /** - * struct iwl_scan_probe_params_v3 + * struct iwl_scan_probe_params_v3 - scan probe parameters * @preq: scan probe request params * @ssid_num: number of valid SSIDs in direct scan array * @short_ssid_num: number of valid short SSIDs in short ssid array @@ -961,7 +965,7 @@ struct iwl_scan_probe_params_v3 { } __packed; /* SCAN_PROBE_PARAMS_API_S_VER_3 */ /** - * struct iwl_scan_probe_params_v4 + * struct iwl_scan_probe_params_v4 - scan probe parameters * @preq: scan probe request params * @short_ssid_num: number of valid short SSIDs in short ssid array * @bssid_num: number of valid bssid in bssids array @@ -983,7 +987,7 @@ struct iwl_scan_probe_params_v4 { #define SCAN_MAX_NUM_CHANS_V3 67 /** - * struct iwl_scan_channel_params_v4 + * struct iwl_scan_channel_params_v4 - channel params * @flags: channel flags &enum iwl_scan_channel_flags * @count: num of channels in scan request * @num_of_aps_override: override the number of APs the FW uses to calculate @@ -1006,7 +1010,7 @@ struct iwl_scan_channel_params_v4 { SCAN_CHANNEL_PARAMS_API_S_VER_5 */ /** - * struct iwl_scan_channel_params_v7 + * struct iwl_scan_channel_params_v7 - channel params * @flags: channel flags &enum iwl_scan_channel_flags * @count: num of channels in scan request * @n_aps_override: override the number of APs the FW uses to calculate dwell @@ -1024,7 +1028,7 @@ struct iwl_scan_channel_params_v7 { } __packed; /* SCAN_CHANNEL_PARAMS_API_S_VER_6 */ /** - * struct iwl_scan_general_params_v11 + * struct iwl_scan_general_params_v11 - channel params * @flags: &enum iwl_umac_scan_general_flags_v2 * @reserved: reserved for future * @scan_start_mac_or_link_id: report the scan start TSF time according to this @@ -1066,7 +1070,7 @@ struct iwl_scan_general_params_v11 { } __packed; /* SCAN_GENERAL_PARAMS_API_S_VER_12, *_VER_11 and *_VER_10 */ /** - * struct iwl_scan_periodic_parms_v1 + * struct iwl_scan_periodic_parms_v1 - periodicity parameters * @schedule: can scheduling parameter * @delay: initial delay of the periodic scan in seconds * @reserved: reserved for future @@ -1078,7 +1082,7 @@ struct iwl_scan_periodic_parms_v1 { } __packed; /* SCAN_PERIODIC_PARAMS_API_S_VER_1 */ /** - * struct iwl_scan_req_params_v12 + * struct iwl_scan_req_params_v12 - scan request parameters (v12) * @general_params: &struct iwl_scan_general_params_v11 * @channel_params: &struct iwl_scan_channel_params_v4 * @periodic_params: &struct iwl_scan_periodic_parms_v1 @@ -1106,7 +1110,7 @@ struct iwl_scan_req_params_v17 { } __packed; /* SCAN_REQUEST_PARAMS_API_S_VER_17 - 14 */ /** - * struct iwl_scan_req_umac_v12 + * struct iwl_scan_req_umac_v12 - scan request command (v12) * @uid: scan id, &enum iwl_umac_scan_uid_offsets * @ooc_priority: out of channel priority - &enum iwl_scan_priority * @scan_params: scan parameters @@ -1130,7 +1134,7 @@ struct iwl_scan_req_umac_v17 { } __packed; /* SCAN_REQUEST_CMD_UMAC_API_S_VER_17 - 14 */ /** - * struct iwl_umac_scan_abort + * struct iwl_umac_scan_abort - scan abort command * @uid: scan id, &enum iwl_umac_scan_uid_offsets * @flags: reserved */ @@ -1140,7 +1144,7 @@ struct iwl_umac_scan_abort { } __packed; /* SCAN_ABORT_CMD_UMAC_API_S_VER_1 */ /** - * enum iwl_umac_scan_abort_status + * enum iwl_umac_scan_abort_status - scan abort status * * @IWL_UMAC_SCAN_ABORT_STATUS_SUCCESS: scan was successfully aborted * @IWL_UMAC_SCAN_ABORT_STATUS_IN_PROGRESS: scan abort is in progress @@ -1153,7 +1157,7 @@ enum iwl_umac_scan_abort_status { }; /** - * struct iwl_umac_scan_complete + * struct iwl_umac_scan_complete - scan complete notification * @uid: scan id, &enum iwl_umac_scan_uid_offsets * @last_schedule: last scheduling line * @last_iter: last scan iteration number diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/sta.h b/drivers/net/wireless/intel/iwlwifi/fw/api/sta.h index ecbcd5084cd8..e6f9abdfa546 100644 --- a/drivers/net/wireless/intel/iwlwifi/fw/api/sta.h +++ b/drivers/net/wireless/intel/iwlwifi/fw/api/sta.h @@ -1,6 +1,6 @@ /* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */ /* - * Copyright (C) 2012-2014, 2018-2021, 2023 Intel Corporation + * Copyright (C) 2012-2014, 2018-2021, 2023, 2025 Intel Corporation * Copyright (C) 2013-2014 Intel Mobile Communications GmbH * Copyright (C) 2016-2017 Intel Deutschland GmbH */ @@ -428,7 +428,7 @@ struct iwl_mvm_rm_sta_cmd { } __packed; /* REMOVE_STA_CMD_API_S_VER_2 */ /** - * struct iwl_mvm_mgmt_mcast_key_cmd_v1 + * struct iwl_mvm_mgmt_mcast_key_cmd_v1 - IGTK command * ( MGMT_MCAST_KEY = 0x1f ) * @ctrl_flags: &enum iwl_sta_key_flag * @igtk: IGTK key material @@ -449,7 +449,7 @@ struct iwl_mvm_mgmt_mcast_key_cmd_v1 { } __packed; /* SEC_MGMT_MULTICAST_KEY_CMD_API_S_VER_1 */ /** - * struct iwl_mvm_mgmt_mcast_key_cmd + * struct iwl_mvm_mgmt_mcast_key_cmd - IGTK command * ( MGMT_MCAST_KEY = 0x1f ) * @ctrl_flags: &enum iwl_sta_key_flag * @igtk: IGTK master key diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/stats.h b/drivers/net/wireless/intel/iwlwifi/fw/api/stats.h index 00713a991879..8d9a5058d5a5 100644 --- a/drivers/net/wireless/intel/iwlwifi/fw/api/stats.h +++ b/drivers/net/wireless/intel/iwlwifi/fw/api/stats.h @@ -26,7 +26,7 @@ struct mvm_statistics_div { } __packed; /* STATISTICS_SLOW_DIV_API_S_VER_2 */ /** - * struct mvm_statistics_rx_non_phy + * struct mvm_statistics_rx_non_phy - non-PHY RX statistics * @bogus_cts: CTS received when not expecting CTS * @bogus_ack: ACK received when not expecting ACK * @non_channel_beacons: beacons with our bss id but not on our serving channel @@ -456,7 +456,7 @@ struct iwl_system_statistics_cmd { } __packed; /* STATISTICS_FW_CMD_API_S_VER_1 */ /** - * enum iwl_fw_statistics_type + * enum iwl_fw_statistics_type - statistics type * * @FW_STATISTICS_OPERATIONAL: operational statistics * @FW_STATISTICS_PHY: phy statistics @@ -478,7 +478,7 @@ enum iwl_fw_statistics_type { #define IWL_STATISTICS_TYPE_MSK 0x7f /** - * struct iwl_statistics_ntfy_hdr + * struct iwl_statistics_ntfy_hdr - statistics notification header * * @type: struct type * @version: version of the struct @@ -491,7 +491,7 @@ struct iwl_statistics_ntfy_hdr { }; /* STATISTICS_NTFY_HDR_API_S_VER_1 */ /** - * struct iwl_stats_ntfy_per_link + * struct iwl_stats_ntfy_per_link - per-link statistics * * @beacon_filter_average_energy: Average energy [-dBm] of the 2 * antennas. @@ -514,7 +514,7 @@ struct iwl_stats_ntfy_per_link { } __packed; /* STATISTICS_NTFY_PER_LINK_API_S_VER_1 */ /** - * struct iwl_stats_ntfy_part1_per_link + * struct iwl_stats_ntfy_part1_per_link - part1 per link statistics * * @rx_time: rx time * @tx_time: tx time @@ -533,7 +533,7 @@ struct iwl_stats_ntfy_part1_per_link { } __packed; /* STATISTICS_FW_NTFY_OPERATIONAL_PART1_PER_LINK_API_S_VER_1 */ /** - * struct iwl_stats_ntfy_per_mac + * struct iwl_stats_ntfy_per_mac - per MAC statistics * * @beacon_filter_average_energy: Average energy [-dBm] of the 2 * antennas. @@ -556,7 +556,8 @@ struct iwl_stats_ntfy_per_mac { } __packed; /* STATISTICS_NTFY_PER_MAC_API_S_VER_1 */ #define IWL_STATS_MAX_BW_INDEX 5 -/** struct iwl_stats_ntfy_per_phy +/** + * struct iwl_stats_ntfy_per_phy - per PHY statistics * @channel_load: channel load * @channel_load_by_us: device contribution to MCLM * @channel_load_not_by_us: other devices' contribution to MCLM @@ -588,7 +589,7 @@ struct iwl_stats_ntfy_per_phy { #define IWL_STATS_UNKNOWN_CHANNEL_LOAD 0xffffffff /** - * struct iwl_stats_ntfy_per_sta + * struct iwl_stats_ntfy_per_sta - per STA statistics * * @average_energy: in fact it is minus the energy.. */ @@ -600,7 +601,7 @@ struct iwl_stats_ntfy_per_sta { #define IWL_STATS_MAX_FW_LINKS (IWL_FW_MAX_LINK_ID + 1) /** - * struct iwl_system_statistics_notif_oper + * struct iwl_system_statistics_notif_oper - statistics notification * * @time_stamp: time when the notification is sent from firmware * @per_link: per link statistics, &struct iwl_stats_ntfy_per_link @@ -615,7 +616,7 @@ struct iwl_system_statistics_notif_oper { } __packed; /* STATISTICS_FW_NTFY_OPERATIONAL_API_S_VER_3 */ /** - * struct iwl_system_statistics_part1_notif_oper + * struct iwl_system_statistics_part1_notif_oper - part1 stats notification * * @time_stamp: time when the notification is sent from firmware * @per_link: per link statistics &struct iwl_stats_ntfy_part1_per_link @@ -628,7 +629,7 @@ struct iwl_system_statistics_part1_notif_oper { } __packed; /* STATISTICS_FW_NTFY_OPERATIONAL_PART1_API_S_VER_4 */ /** - * struct iwl_system_statistics_end_notif + * struct iwl_system_statistics_end_notif - statistics end notification * * @time_stamp: time when the notification is sent from firmware */ @@ -637,7 +638,7 @@ struct iwl_system_statistics_end_notif { } __packed; /* STATISTICS_FW_NTFY_END_API_S_VER_1 */ /** - * struct iwl_statistics_operational_ntfy + * struct iwl_statistics_operational_ntfy - operational stats notification * * @hdr: general statistics header * @flags: bitmap of possible notification structures @@ -662,7 +663,7 @@ struct iwl_statistics_operational_ntfy { } __packed; /* STATISTICS_OPERATIONAL_NTFY_API_S_VER_15 */ /** - * struct iwl_statistics_operational_ntfy_ver_14 + * struct iwl_statistics_operational_ntfy_ver_14 - operational stats notification * * @hdr: general statistics header * @flags: bitmap of possible notification structures @@ -707,7 +708,7 @@ struct iwl_statistics_operational_ntfy_ver_14 { } __packed; /* STATISTICS_OPERATIONAL_NTFY_API_S_VER_14 */ /** - * struct iwl_statistics_phy_ntfy + * struct iwl_statistics_phy_ntfy - PHY statistics notification * * @hdr: general statistics header * RX PHY related statistics @@ -808,7 +809,7 @@ struct iwl_statistics_phy_ntfy { } __packed; /* STATISTICS_PHY_NTFY_API_S_VER_1 */ /** - * struct iwl_statistics_mac_ntfy + * struct iwl_statistics_mac_ntfy - MAC statistics notification * * @hdr: general statistics header * @bcast_filter_passed_per_mac: bcast filter passed per mac @@ -827,7 +828,7 @@ struct iwl_statistics_mac_ntfy { } __packed; /* STATISTICS_MAC_NTFY_API_S_VER_1 */ /** - * struct iwl_statistics_rx_ntfy + * struct iwl_statistics_rx_ntfy - RX statistics notification * * @hdr: general statistics header * @rx_agg_mpdu_cnt: aggregation frame count (number of @@ -867,7 +868,7 @@ struct iwl_statistics_rx_ntfy { } __packed; /* STATISTICS_RX_NTFY_API_S_VER_1 */ /** - * struct iwl_statistics_tx_ntfy + * struct iwl_statistics_tx_ntfy - TX statistics notification * * @hdr: general statistics header * @cts_timeout: timeout when waiting for CTS @@ -976,7 +977,7 @@ struct iwl_statistics_tx_ntfy { } __packed; /* STATISTICS_TX_NTFY_API_S_VER_1 */ /** - * struct iwl_statistics_duration_ntfy + * struct iwl_statistics_duration_ntfy - burst/duration statistics * * @hdr: general statistics header * @cont_burst_chk_cnt: number of times continuation or @@ -995,7 +996,7 @@ struct iwl_statistics_duration_ntfy { } __packed; /* STATISTICS_DURATION_NTFY_API_S_VER_1 */ /** - * struct iwl_statistics_he_ntfy + * struct iwl_statistics_he_ntfy - HE statistics * * @hdr: general statistics header * received HE frames diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/tx.h b/drivers/net/wireless/intel/iwlwifi/fw/api/tx.h index 26d2013905ed..31d3336726b4 100644 --- a/drivers/net/wireless/intel/iwlwifi/fw/api/tx.h +++ b/drivers/net/wireless/intel/iwlwifi/fw/api/tx.h @@ -963,7 +963,7 @@ struct iwl_scd_txq_cfg_cmd { } __packed; /* SCD_QUEUE_CFG_CMD_API_S_VER_1 */ /** - * struct iwl_scd_txq_cfg_rsp + * struct iwl_scd_txq_cfg_rsp - scheduler TXQ configuration response * @token: taken from the command * @sta_id: station id from the command * @tid: tid from the command diff --git a/drivers/net/wireless/intel/iwlwifi/fw/error-dump.h b/drivers/net/wireless/intel/iwlwifi/fw/error-dump.h index c2a73cc85eff..525a82030daa 100644 --- a/drivers/net/wireless/intel/iwlwifi/fw/error-dump.h +++ b/drivers/net/wireless/intel/iwlwifi/fw/error-dump.h @@ -266,7 +266,7 @@ struct iwl_fw_ini_error_dump_data { } __packed; /** - * struct iwl_fw_ini_dump_entry + * struct iwl_fw_ini_dump_entry - dump entry descriptor * @list: list of dump entries * @size: size of the data * @data: entry data @@ -305,7 +305,7 @@ struct iwl_fw_ini_fifo_hdr { * @dram_base_addr: base address of dram monitor range * @page_num: page number of memory range * @fifo_hdr: fifo header of memory range - * @fw_pkt: FW packet header of memory range + * @fw_pkt_hdr: FW packet header of memory range * @data: the actual memory */ struct iwl_fw_ini_error_dump_range { diff --git a/drivers/net/wireless/intel/iwlwifi/fw/file.h b/drivers/net/wireless/intel/iwlwifi/fw/file.h index b7c1ab7a3006..b9e0b69c6680 100644 --- a/drivers/net/wireless/intel/iwlwifi/fw/file.h +++ b/drivers/net/wireless/intel/iwlwifi/fw/file.h @@ -222,7 +222,10 @@ typedef unsigned int __bitwise iwl_ucode_tlv_api_t; * @IWL_UCODE_TLV_API_STA_TYPE: This ucode supports station type assignement. * @IWL_UCODE_TLV_API_NAN2_VER2: This ucode supports NAN API version 2 * @IWL_UCODE_TLV_API_ADAPTIVE_DWELL: support for adaptive dwell in scanning + * @IWL_UCODE_TLV_API_OCE: support for OCE + * @IWL_UCODE_TLV_API_NEW_BEACON_TEMPLATE: new beacon template * @IWL_UCODE_TLV_API_NEW_RX_STATS: should new RX STATISTICS API be used + * @IWL_UCODE_TLV_API_WOWLAN_KEY_MATERIAL: WoWLAN key material support * @IWL_UCODE_TLV_API_QUOTA_LOW_LATENCY: Quota command includes a field * indicating low latency direction. * @IWL_UCODE_TLV_API_DEPRECATE_TTAK: RX status flag TTAK ok (bit 7) is @@ -245,6 +248,7 @@ typedef unsigned int __bitwise iwl_ucode_tlv_api_t; * SCAN_OFFLOAD_PROFILES_QUERY_RSP_S. * @IWL_UCODE_TLV_API_MBSSID_HE: This ucode supports v2 of * STA_CONTEXT_DOT11AX_API_S + * @IWL_UCODE_TLV_API_WOWLAN_TCP_SYN_WAKE: WoWLAN TCP-SYN wake support * @IWL_UCODE_TLV_API_FTM_RTT_ACCURACY: version 7 of the range response API * is supported by FW, this indicates the RTT confidence value * @IWL_UCODE_TLV_API_SAR_TABLE_VER: This ucode supports different sar @@ -253,6 +257,7 @@ typedef unsigned int __bitwise iwl_ucode_tlv_api_t; * SCAN_CONFIG_DB_CMD_API_S. * @IWL_UCODE_TLV_API_ADWELL_HB_DEF_N_AP: support for setting adaptive dwell * number of APs in the 5 GHz band + * @IWL_UCODE_TLV_API_SCAN_EXT_CHAN_VER: extended channel config in scan * @IWL_UCODE_TLV_API_BAND_IN_RX_DATA: FW reports band number in RX notification * @IWL_UCODE_TLV_API_NO_HOST_DISABLE_TX: Firmware offloaded the station disable tx * logic. @@ -352,16 +357,24 @@ typedef unsigned int __bitwise iwl_ucode_tlv_capa_t; * @IWL_UCODE_TLV_CAPA_SOC_LATENCY_SUPPORT: the firmware supports setting * stabilization latency for SoCs. * @IWL_UCODE_TLV_CAPA_STA_PM_NOTIF: firmware will send STA PM notification + * @IWL_UCODE_TLV_CAPA_BINDING_CDB_SUPPORT: binding CDB support + * @IWL_UCODE_TLV_CAPA_CDB_SUPPORT: CDB support + * @IWL_UCODE_TLV_CAPA_D0I3_END_FIRST: D0I3 end command comes first * @IWL_UCODE_TLV_CAPA_TLC_OFFLOAD: firmware implements rate scaling algorithm * @IWL_UCODE_TLV_CAPA_DYNAMIC_QUOTA: firmware implements quota related * @IWL_UCODE_TLV_CAPA_COEX_SCHEMA_2: firmware implements Coex Schema 2 - * IWL_UCODE_TLV_CAPA_CHANNEL_SWITCH_CMD: firmware supports CSA command + * @IWL_UCODE_TLV_CAPA_CHANNEL_SWITCH_CMD: firmware supports CSA command * @IWL_UCODE_TLV_CAPA_ULTRA_HB_CHANNELS: firmware supports ultra high band * (6 GHz). * @IWL_UCODE_TLV_CAPA_CS_MODIFY: firmware supports modify action CSA command + * @IWL_UCODE_TLV_CAPA_SET_LTR_GEN2: LTR gen2 support + * @IWL_UCODE_TLV_CAPA_TAS_CFG: TAS configuration support + * @IWL_UCODE_TLV_CAPA_SESSION_PROT_CMD: session protection command + * @IWL_UCODE_TLV_CAPA_SET_PPAG: PPAG support * @IWL_UCODE_TLV_CAPA_EXTENDED_DTS_MEASURE: extended DTS measurement * @IWL_UCODE_TLV_CAPA_SHORT_PM_TIMEOUTS: supports short PM timeouts * @IWL_UCODE_TLV_CAPA_BT_MPLUT_SUPPORT: supports bt-coex Multi-priority LUT + * @IWL_UCODE_TLV_CAPA_MULTI_QUEUE_RX_SUPPORT: MQ RX support * @IWL_UCODE_TLV_CAPA_CSA_AND_TBTT_OFFLOAD: the firmware supports CSA * countdown offloading. Beacon notifications are not sent to the host. * The fw also offloads TBTT alignment. @@ -383,23 +396,46 @@ typedef unsigned int __bitwise iwl_ucode_tlv_capa_t; * command size (command version 4) that supports toggling ACK TX * power reduction. * @IWL_UCODE_TLV_CAPA_D3_DEBUG: supports debug recording during D3 + * @IWL_UCODE_TLV_CAPA_LED_CMD_SUPPORT: LED command support * @IWL_UCODE_TLV_CAPA_MCC_UPDATE_11AX_SUPPORT: MCC response support 11ax * capability. * @IWL_UCODE_TLV_CAPA_CSI_REPORTING: firmware is capable of being configured * to report the CSI information with (certain) RX frames + * @IWL_UCODE_TLV_CAPA_DBG_SUSPEND_RESUME_CMD_SUPP: suspend/resume command + * @IWL_UCODE_TLV_CAPA_DBG_BUF_ALLOC_CMD_SUPP: support for DBGC + * buffer allocation command * @IWL_UCODE_TLV_CAPA_FTM_CALIBRATED: has FTM calibrated and thus supports both * initiator and responder * @IWL_UCODE_TLV_CAPA_BIOS_OVERRIDE_UNII4_US_CA: supports (de)activating UNII-4 * for US/CA/WW from BIOS + * @IWL_UCODE_TLV_CAPA_PSC_CHAN_SUPPORT: supports PSC channels + * @IWL_UCODE_TLV_CAPA_BIGTK_SUPPORT: BIGTK support * @IWL_UCODE_TLV_CAPA_PROTECTED_TWT: Supports protection of TWT action frames * @IWL_UCODE_TLV_CAPA_FW_RESET_HANDSHAKE: Supports the firmware handshake in * reset flow * @IWL_UCODE_TLV_CAPA_PASSIVE_6GHZ_SCAN: Support for passive scan on 6GHz PSC * channels even when these are not enabled. + * @IWL_UCODE_TLV_CAPA_HIDDEN_6GHZ_SCAN: hidden SSID 6 GHz scan support + * @IWL_UCODE_TLV_CAPA_BROADCAST_TWT: broadcast TWT support + * @IWL_UCODE_TLV_CAPA_COEX_HIGH_PRIO: support for BT-coex high + * priority for 802.1X/4-way-HS + * @IWL_UCODE_TLV_CAPA_BAID_ML_SUPPORT: multi-link BAID support + * @IWL_UCODE_TLV_CAPA_SYNCED_TIME: synced time command support + * @IWL_UCODE_TLV_CAPA_TIME_SYNC_BOTH_FTM_TM: time sync support + * @IWL_UCODE_TLV_CAPA_BIGTK_TX_SUPPORT: BIGTK TX support + * @IWL_UCODE_TLV_CAPA_MLD_API_SUPPORT: MLD API support + * @IWL_UCODE_TLV_CAPA_SCAN_DONT_TOGGLE_ANT: fixed antenna scan support + * @IWL_UCODE_TLV_CAPA_PPAG_CHINA_BIOS_SUPPORT: PPAG China BIOS support + * @IWL_UCODE_TLV_CAPA_OFFLOAD_BTM_SUPPORT: BTM protocol offload support + * @IWL_UCODE_TLV_CAPA_STA_EXP_MFP_SUPPORT: STA command MFP support + * @IWL_UCODE_TLV_CAPA_SNIFF_VALIDATE_SUPPORT: sniffer validate bits support + * @IWL_UCODE_TLV_CAPA_CHINA_22_REG_SUPPORT: China 2022 regulator support * @IWL_UCODE_TLV_CAPA_DUMP_COMPLETE_SUPPORT: Support for indicating dump collection * complete to FW. * @IWL_UCODE_TLV_CAPA_SPP_AMSDU_SUPPORT: Support SPP (signaling and payload * protected) A-MSDU. + * @IWL_UCODE_TLV_CAPA_DRAM_FRAG_SUPPORT: support for DBGC fragmented + * DRAM buffers * @IWL_UCODE_TLV_CAPA_SECURE_LTF_SUPPORT: Support secure LTF measurement. * @IWL_UCODE_TLV_CAPA_MONITOR_PASSIVE_CHANS: Support monitor mode on otherwise * passive channels @@ -407,6 +443,8 @@ typedef unsigned int __bitwise iwl_ucode_tlv_capa_t; * for CA from BIOS. * @IWL_UCODE_TLV_CAPA_UHB_CANADA_TAS_SUPPORT: supports %TAS_UHB_ALLOWED_CANADA * @IWL_UCODE_TLV_CAPA_EXT_FSEQ_IMAGE_SUPPORT: external FSEQ image support + * @IWL_UCODE_TLV_CAPA_RESET_DURING_ASSERT: FW reset handshake is needed + * during assert handling even if the dump isn't split * @IWL_UCODE_TLV_CAPA_FW_ACCEPTS_RAW_DSM_TABLE: Firmware has capability of * handling raw DSM table data. * @@ -487,12 +525,7 @@ enum iwl_ucode_tlv_capa { /* set 3 */ IWL_UCODE_TLV_CAPA_BIOS_OVERRIDE_UNII4_US_CA = (__force iwl_ucode_tlv_capa_t)96, - - /* - * @IWL_UCODE_TLV_CAPA_PSC_CHAN_SUPPORT: supports PSC channels - */ IWL_UCODE_TLV_CAPA_PSC_CHAN_SUPPORT = (__force iwl_ucode_tlv_capa_t)98, - IWL_UCODE_TLV_CAPA_BIGTK_SUPPORT = (__force iwl_ucode_tlv_capa_t)100, IWL_UCODE_TLV_CAPA_SPP_AMSDU_SUPPORT = (__force iwl_ucode_tlv_capa_t)103, IWL_UCODE_TLV_CAPA_DRAM_FRAG_SUPPORT = (__force iwl_ucode_tlv_capa_t)104, @@ -514,11 +547,8 @@ enum iwl_ucode_tlv_capa { IWL_UCODE_TLV_CAPA_EXT_FSEQ_IMAGE_SUPPORT = (__force iwl_ucode_tlv_capa_t)125, /* set 4 */ - /** - * @IWL_UCODE_TLV_CAPA_RESET_DURING_ASSERT: FW reset handshake is needed - * during assert handling even if the dump isn't split - */ - IWL_UCODE_TLV_CAPA_RESET_DURING_ASSERT = (__force iwl_ucode_tlv_capa_t)(4 * 32 + 0), + + IWL_UCODE_TLV_CAPA_RESET_DURING_ASSERT = (__force iwl_ucode_tlv_capa_t)(4 * 32 + 0), IWL_UCODE_TLV_CAPA_FW_ACCEPTS_RAW_DSM_TABLE = (__force iwl_ucode_tlv_capa_t)(4 * 32 + 1), NUM_IWL_UCODE_TLV_CAPA /* @@ -852,6 +882,8 @@ struct iwl_fw_dbg_trigger_low_rssi { * @start_assoc_denied: number of denied association to start recording * @start_assoc_timeout: number of association timeout to start recording * @start_connection_loss: number of connection loss to start recording + * @reserved: reserved + * @reserved2: reserved */ struct iwl_fw_dbg_trigger_mlme { u8 stop_auth_denied; @@ -885,6 +917,7 @@ struct iwl_fw_dbg_trigger_mlme { * @p2p_device: timeout for the queues of a P2P device in ms * @ibss: timeout for the queues of an IBSS in ms * @tdls: timeout for the queues of a TDLS station in ms + * @reserved: reserved */ struct iwl_fw_dbg_trigger_txq_timer { __le32 command_queue; @@ -900,7 +933,7 @@ struct iwl_fw_dbg_trigger_txq_timer { /** * struct iwl_fw_dbg_trigger_time_event - configures a time event trigger - * time_Events: a list of tuples . The driver will issue a + * @time_events: a list of tuples . The driver will issue a * trigger each time a time event notification that relates to time event * id with one of the actions in the bitmap is received and * BIT(notif->status) is set in status_bitmap. @@ -916,19 +949,19 @@ struct iwl_fw_dbg_trigger_time_event { /** * struct iwl_fw_dbg_trigger_ba - configures BlockAck related trigger - * rx_ba_start: tid bitmap to configure on what tid the trigger should occur + * @rx_ba_start: tid bitmap to configure on what tid the trigger should occur * when an Rx BlockAck session is started. - * rx_ba_stop: tid bitmap to configure on what tid the trigger should occur + * @rx_ba_stop: tid bitmap to configure on what tid the trigger should occur * when an Rx BlockAck session is stopped. - * tx_ba_start: tid bitmap to configure on what tid the trigger should occur + * @tx_ba_start: tid bitmap to configure on what tid the trigger should occur * when a Tx BlockAck session is started. - * tx_ba_stop: tid bitmap to configure on what tid the trigger should occur + * @tx_ba_stop: tid bitmap to configure on what tid the trigger should occur * when a Tx BlockAck session is stopped. - * rx_bar: tid bitmap to configure on what tid the trigger should occur + * @rx_bar: tid bitmap to configure on what tid the trigger should occur * when a BAR is received (for a Tx BlockAck session). - * tx_bar: tid bitmap to configure on what tid the trigger should occur + * @tx_bar: tid bitmap to configure on what tid the trigger should occur * when a BAR is send (for an Rx BlocAck session). - * frame_timeout: tid bitmap to configure on what tid the trigger should occur + * @frame_timeout: tid bitmap to configure on what tid the trigger should occur * when a frame times out in the reordering buffer. */ struct iwl_fw_dbg_trigger_ba { @@ -946,6 +979,7 @@ struct iwl_fw_dbg_trigger_ba { * @action_bitmap: the TDLS action to trigger the collection upon * @peer_mode: trigger on specific peer or all * @peer: the TDLS peer to trigger the collection on + * @reserved: reserved */ struct iwl_fw_dbg_trigger_tdls { u8 action_bitmap; @@ -958,6 +992,7 @@ struct iwl_fw_dbg_trigger_tdls { * struct iwl_fw_dbg_trigger_tx_status - configures trigger for tx response * status. * @statuses: the list of statuses to trigger the collection on + * @reserved: reserved */ struct iwl_fw_dbg_trigger_tx_status { struct tx_status { @@ -971,6 +1006,7 @@ struct iwl_fw_dbg_trigger_tx_status { * struct iwl_fw_dbg_conf_tlv - a TLV that describes a debug configuration. * @id: conf id * @usniffer: should the uSniffer image be used + * @reserved: reserved * @num_of_hcmds: how many HCMDs to send are present here * @hcmd: a variable length host command to be sent to apply the configuration. * If there is more than one HCMD to send, they will appear one after the diff --git a/drivers/net/wireless/intel/iwlwifi/fw/img.h b/drivers/net/wireless/intel/iwlwifi/fw/img.h index 5256f20623e9..045a3e009429 100644 --- a/drivers/net/wireless/intel/iwlwifi/fw/img.h +++ b/drivers/net/wireless/intel/iwlwifi/fw/img.h @@ -14,14 +14,13 @@ #include "error-dump.h" /** - * enum iwl_ucode_type - * - * The type of ucode. + * enum iwl_ucode_type - type of ucode * * @IWL_UCODE_REGULAR: Normal runtime ucode * @IWL_UCODE_INIT: Initial ucode * @IWL_UCODE_WOWLAN: Wake on Wireless enabled ucode * @IWL_UCODE_REGULAR_USNIFFER: Normal runtime ucode when using usniffer image + * @IWL_UCODE_TYPE_MAX: (internal value) */ enum iwl_ucode_type { IWL_UCODE_REGULAR, @@ -122,7 +121,7 @@ struct fw_img { #define FW_ADDR_CACHE_CONTROL 0xC0000000UL /** - * struct iwl_fw_paging + * struct iwl_fw_paging - FW paging descriptor * @fw_paging_phys: page phy pointer * @fw_paging_block: pointer to the allocated block * @fw_paging_size: page size @@ -197,6 +196,11 @@ struct iwl_dump_exclude { * @dump_excl_wowlan: image dump exclusion areas for WoWLAN image * @pnvm_data: PNVM data embedded in the .ucode file, if any * @pnvm_size: size of the embedded PNVM data + * @dbg: debug data, see &struct iwl_fw_dbg + * @default_calib: default calibration data + * @phy_config: PHY configuration flags + * @valid_rx_ant: valid RX antenna bitmap + * @valid_tx_ant: valid TX antenna bitmap */ struct iwl_fw { u32 ucode_ver; diff --git a/drivers/net/wireless/intel/iwlwifi/fw/runtime.h b/drivers/net/wireless/intel/iwlwifi/fw/runtime.h index 806f9bcdf4f5..57570ff15622 100644 --- a/drivers/net/wireless/intel/iwlwifi/fw/runtime.h +++ b/drivers/net/wireless/intel/iwlwifi/fw/runtime.h @@ -45,6 +45,8 @@ struct iwl_fwrt_shared_mem_cfg { * struct iwl_fwrt_dump_data - dump data * @trig: trigger the worker was scheduled upon * @fw_pkt: packet received from FW + * @desc: dump descriptor + * @monitor_only: only dump for monitor * * Note that the decision which part of the union is used * is based on iwl_trans_dbg_ini_valid(): the 'trig' part @@ -68,6 +70,7 @@ struct iwl_fwrt_dump_data { * struct iwl_fwrt_wk_data - dump worker data struct * @idx: index of the worker * @wk: worker + * @dump_data: dump data */ struct iwl_fwrt_wk_data { u8 idx; @@ -91,8 +94,8 @@ struct iwl_txf_iter_data { /** * struct iwl_fw_runtime - runtime data for firmware + * @trans: transport pointer * @fw: firmware image - * @cfg: NIC configuration * @dev: device pointer * @ops: user ops * @ops_ctx: user ops context @@ -117,6 +120,23 @@ struct iwl_txf_iter_data { * zero (default initialization) means it hasn't been read yet, * and BIT(0) is set when it has since function 0 also has this * bitmap and is always supported + * @geo_enabled: WGDS table is present + * @geo_num_profiles: number of geo profiles + * @geo_rev: geo profiles table revision + * @ppag_chains: PPAG table data + * @ppag_flags: PPAG flags + * @reduced_power_flags: reduced power flags + * @sanitize_ctx: context for dump sanitizer + * @sanitize_ops: dump sanitizer ops + * @sar_chain_a_profile: SAR chain A profile + * @sar_chain_b_profile: SAR chain B profile + * @sgom_enabled: SGOM enabled + * @sgom_table: SGOM table + * @timestamp: timestamp marker data + * @timestamp.wk: timestamp marking worker + * @timestamp.seq: timestamp marking sequence + * @timestamp.delay: timestamp marking worker delay + * @tpc_enabled: TPC enabled */ struct iwl_fw_runtime { struct iwl_trans *trans; diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-config.h b/drivers/net/wireless/intel/iwlwifi/iwl-config.h index cfd1629f0016..0b34c9f90b3f 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-config.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-config.h @@ -385,7 +385,7 @@ struct iwl_mac_cfg { #define IWL_NUM_RBDS_EHT (512 * 8) /** - * struct iwl_rf_cfg + * struct iwl_rf_cfg - RF/CRF configuration data * @fw_name_pre: Firmware filename prefix. The api version and extension * (.ucode) will be added to filename before loading from disk. The * filename is constructed as -.ucode. diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.h b/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.h index 7ed6329fd8ca..fe4e46a0edbd 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.h @@ -1,6 +1,6 @@ /* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */ /* - * Copyright (C) 2018-2023 Intel Corporation + * Copyright (C) 2018-2023, 2025 Intel Corporation */ #ifndef __iwl_dbg_tlv_h__ #define __iwl_dbg_tlv_h__ @@ -32,7 +32,7 @@ union iwl_dbg_tlv_tp_data { }; /** - * struct iwl_dbg_tlv_time_point_data + * struct iwl_dbg_tlv_time_point_data - debug time point data * @trig_list: list of triggers * @active_trig_list: list of active triggers * @hcmd_list: list of host commands diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-drv.h b/drivers/net/wireless/intel/iwlwifi/iwl-drv.h index 595300a14639..a0b67e8aba8d 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-drv.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-drv.h @@ -62,7 +62,8 @@ struct iwl_rf_cfg; * starts the driver: fetches the firmware. This should be called by bus * specific system flows implementations. For example, the bus specific probe * function should do bus related operations only, and then call to this - * function. It returns the driver object or %NULL if an error occurred. + * function. + * Return: the driver object or %NULL if an error occurred. */ struct iwl_drv *iwl_drv_start(struct iwl_trans *trans); diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-modparams.h b/drivers/net/wireless/intel/iwlwifi/iwl-modparams.h index 21eabfc3ffc8..0476df7b7f17 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-modparams.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-modparams.h @@ -1,6 +1,6 @@ /* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */ /* - * Copyright (C) 2005-2014, 2018-2022, 2024 Intel Corporation + * Copyright (C) 2005-2014, 2018-2022, 2024-2025 Intel Corporation */ #ifndef __iwl_modparams_h__ #define __iwl_modparams_h__ @@ -42,7 +42,7 @@ enum iwl_uapsd_disable { }; /** - * struct iwl_mod_params + * struct iwl_mod_params - module parameters for iwlwifi * * Holds the module parameters * diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.h b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.h index cbc92abf9f87..12f28bb0e859 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.h @@ -115,11 +115,12 @@ iwl_parse_nvm_data(struct iwl_trans *trans, const struct iwl_rf_cfg *cfg, * iwl_parse_nvm_mcc_info - parse MCC (mobile country code) info coming from FW * * This function parses the regulatory channel data received as a - * MCC_UPDATE_CMD command. It returns a newly allocation regulatory domain, - * to be fed into the regulatory core. In case the geo_info is set handle - * accordingly. An ERR_PTR is returned on error. - * If not given to the regulatory core, the user is responsible for freeing - * the regdomain returned here with kfree. + * MCC_UPDATE_CMD command. + * + * Return: a newly allocation regulatory domain, to be given to the regulatory + * core. In case the geo_info is set handle accordingly. An ERR_PTR is + * returned on error. If not given to the regulatory core, the user is + * responsible for freeing the regdomain returned here with kfree(). * * @trans: the transport * @num_of_ch: the number of channels @@ -140,6 +141,8 @@ iwl_parse_nvm_mcc_info(struct iwl_trans *trans, * This struct holds an NVM section read from the NIC using NVM_ACCESS_CMD, * and saved for later use by the driver. Not all NVM sections are saved * this way, only the needed ones. + * @length: length of the section + * @data: section data */ struct iwl_nvm_section { u16 length; @@ -148,6 +151,10 @@ struct iwl_nvm_section { /** * iwl_read_external_nvm - Reads external NVM from a file into nvm_sections + * @trans: the transport + * @nvm_file_name: the filename to request + * @nvm_sections: sections data to fill + * Return: 0 on success or an error code */ int iwl_read_external_nvm(struct iwl_trans *trans, const char *nvm_file_name, diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-op-mode.h b/drivers/net/wireless/intel/iwlwifi/iwl-op-mode.h index a146d0e399f2..df6341dfc4a1 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-op-mode.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-op-mode.h @@ -185,6 +185,7 @@ void iwl_opmode_deregister(const char *name); /** * struct iwl_op_mode - operational mode * @ops: pointer to its own ops + * @op_mode_specific: per-opmode data * * This holds an implementation of the mac80211 / fw API. */ diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h index a0cc5d7745e8..a552669db6e2 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h @@ -121,7 +121,7 @@ enum CMD_MODE { #define DEF_CMD_PAYLOAD_SIZE 320 /** - * struct iwl_device_cmd + * struct iwl_device_cmd - device command structure * * For allocation of the command and tx queues, this establishes the overall * size of the largest command we send to uCode, except for commands that @@ -516,7 +516,7 @@ enum iwl_trans_state { */ /** - * enum iwl_ini_cfg_state + * enum iwl_ini_cfg_state - debug config state * @IWL_INI_CFG_STATE_NOT_LOADED: no debug cfg was given * @IWL_INI_CFG_STATE_LOADED: debug cfg was found and loaded * @IWL_INI_CFG_STATE_CORRUPTED: debug cfg was found and some of the TLVs @@ -532,7 +532,7 @@ enum iwl_ini_cfg_state { #define IWL_TRANS_NMI_TIMEOUT (HZ / 4) /** - * struct iwl_dram_data + * struct iwl_dram_data - DRAM data descriptor * @physical: page phy pointer * @block: pointer to the allocated block/page * @size: size of the block/page From d852e72d9425ef1401442c0226cd7b936c02a0ca Mon Sep 17 00:00:00 2001 From: Miri Korenblit Date: Mon, 15 Sep 2025 11:34:30 +0300 Subject: [PATCH 0311/1381] wifi: iwlwifi: mvm: cleanup unsupported phy command versions The iwlmvm op mode is used for pre EHT HWs. Those HWs doesn't have wider OFDMA, so phy command versions 5+ (that added support for wider OFDMA) are not supported. Remove support for them. This means that we also don't need to set the IEEE80211_VIF_IGNORE_OFDMA_WIDER_BW, as we don't care about the ap chandef anyway. Reviewed-by: Johannes Berg Signed-off-by: Miri Korenblit Link: https://patch.msgid.link/20250915113137.e7e26fe71132.I8ec95ff86521661118782ecee1be20ef6e8e48e1@changeid --- .../wireless/intel/iwlwifi/mvm/mld-mac80211.c | 3 --- .../net/wireless/intel/iwlwifi/mvm/phy-ctxt.c | 24 +++++-------------- 2 files changed, 6 insertions(+), 21 deletions(-) diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c index b1dca76b7141..380b6f8a53fd 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c @@ -102,9 +102,6 @@ static int iwl_mvm_mld_mac_add_interface(struct ieee80211_hw *hw, mvm->csme_vif = vif; } - if (vif->p2p || iwl_fw_lookup_cmd_ver(mvm->fw, PHY_CONTEXT_CMD, 1) < 5) - vif->driver_flags |= IEEE80211_VIF_IGNORE_OFDMA_WIDER_BW; - return 0; out_free_bf: diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/phy-ctxt.c b/drivers/net/wireless/intel/iwlwifi/mvm/phy-ctxt.c index 5e7e2926be0c..4f4111055ddd 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/phy-ctxt.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/phy-ctxt.c @@ -1,6 +1,6 @@ // SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause /* - * Copyright (C) 2012-2014, 2018-2024 Intel Corporation + * Copyright (C) 2012-2014, 2018-2025 Intel Corporation * Copyright (C) 2013-2014 Intel Mobile Communications GmbH * Copyright (C) 2017 Intel Deutschland GmbH */ @@ -202,17 +202,13 @@ int iwl_mvm_phy_send_rlc(struct iwl_mvm *mvm, struct iwl_mvm_phy_ctxt *ctxt, static int iwl_mvm_phy_ctxt_apply(struct iwl_mvm *mvm, struct iwl_mvm_phy_ctxt *ctxt, const struct cfg80211_chan_def *chandef, - const struct cfg80211_chan_def *ap, u8 chains_static, u8 chains_dynamic, u32 action) { int ret; int ver = iwl_fw_lookup_cmd_ver(mvm->fw, PHY_CONTEXT_CMD, 1); - if (ver < 5 || !ap || !ap->chan) - ap = NULL; - - if (ver >= 3 && ver <= 6) { + if (ver >= 3 && ver <= 4) { struct iwl_phy_context_cmd cmd = {}; /* Set the command header fields */ @@ -223,14 +219,6 @@ static int iwl_mvm_phy_ctxt_apply(struct iwl_mvm *mvm, chains_static, chains_dynamic); - if (ap) { - cmd.sbb_bandwidth = iwl_mvm_get_channel_width(ap); - cmd.sbb_ctrl_channel_loc = iwl_mvm_get_ctrl_pos(ap); - } - - if (ver == 6) - cmd.puncture_mask = cpu_to_le16(chandef->punctured); - ret = iwl_mvm_send_cmd_pdu(mvm, PHY_CONTEXT_CMD, 0, sizeof(cmd), &cmd); } else if (ver < 3) { @@ -284,7 +272,7 @@ int iwl_mvm_phy_ctxt_add(struct iwl_mvm *mvm, struct iwl_mvm_phy_ctxt *ctxt, ctxt->width = chandef->width; ctxt->center_freq1 = chandef->center_freq1; - ret = iwl_mvm_phy_ctxt_apply(mvm, ctxt, chandef, ap, + ret = iwl_mvm_phy_ctxt_apply(mvm, ctxt, chandef, chains_static, chains_dynamic, FW_CTXT_ACTION_ADD); @@ -342,7 +330,7 @@ int iwl_mvm_phy_ctxt_changed(struct iwl_mvm *mvm, struct iwl_mvm_phy_ctxt *ctxt, int ret; /* ... remove it here ...*/ - ret = iwl_mvm_phy_ctxt_apply(mvm, ctxt, chandef, NULL, + ret = iwl_mvm_phy_ctxt_apply(mvm, ctxt, chandef, chains_static, chains_dynamic, FW_CTXT_ACTION_REMOVE); if (ret) @@ -356,7 +344,7 @@ int iwl_mvm_phy_ctxt_changed(struct iwl_mvm *mvm, struct iwl_mvm_phy_ctxt *ctxt, ctxt->width = chandef->width; ctxt->center_freq1 = chandef->center_freq1; - return iwl_mvm_phy_ctxt_apply(mvm, ctxt, chandef, ap, + return iwl_mvm_phy_ctxt_apply(mvm, ctxt, chandef, chains_static, chains_dynamic, action); } @@ -376,7 +364,7 @@ void iwl_mvm_phy_ctxt_unref(struct iwl_mvm *mvm, struct iwl_mvm_phy_ctxt *ctxt) cfg80211_chandef_create(&chandef, ctxt->channel, NL80211_CHAN_NO_HT); - iwl_mvm_phy_ctxt_apply(mvm, ctxt, &chandef, NULL, 1, 1, + iwl_mvm_phy_ctxt_apply(mvm, ctxt, &chandef, 1, 1, FW_CTXT_ACTION_REMOVE); } From 355431679a910eedc3d1a01cfffaec590a02033b Mon Sep 17 00:00:00 2001 From: Emmanuel Grumbach Date: Mon, 15 Sep 2025 11:34:31 +0300 Subject: [PATCH 0312/1381] wifi: iwlwifi: mld: support get/set_antenna This allows to set the antennas from user space. Signed-off-by: Emmanuel Grumbach Signed-off-by: Miri Korenblit Link: https://patch.msgid.link/20250915113137.5a45baf9513c.I5912e6b6d9a9ae6530d0ac45e9517d07f98b8d05@changeid --- .../net/wireless/intel/iwlwifi/mld/mac80211.c | 34 +++++++++++++++++++ drivers/net/wireless/intel/iwlwifi/mld/mld.h | 11 ++++++ 2 files changed, 45 insertions(+) diff --git a/drivers/net/wireless/intel/iwlwifi/mld/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mld/mac80211.c index 5725104a53bf..98d47fed8421 100644 --- a/drivers/net/wireless/intel/iwlwifi/mld/mac80211.c +++ b/drivers/net/wireless/intel/iwlwifi/mld/mac80211.c @@ -23,6 +23,7 @@ #include "roc.h" #include "mlo.h" #include "stats.h" +#include "iwl-nvm-parse.h" #include "ftm-initiator.h" #include "low_latency.h" #include "fw/api/scan.h" @@ -2591,11 +2592,44 @@ iwl_mld_can_neg_ttlm(struct ieee80211_hw *hw, struct ieee80211_vif *vif, return NEG_TTLM_RES_ACCEPT; } +static int iwl_mld_get_antenna(struct ieee80211_hw *hw, int radio_idx, + u32 *tx_ant, u32 *rx_ant) +{ + struct iwl_mld *mld = IWL_MAC80211_GET_MLD(hw); + + *tx_ant = iwl_mld_get_valid_tx_ant(mld); + *rx_ant = iwl_mld_get_valid_rx_ant(mld); + + return 0; +} + +static int iwl_mld_set_antenna(struct ieee80211_hw *hw, int radio_idx, + u32 tx_ant, u32 rx_ant) +{ + struct iwl_mld *mld = IWL_MAC80211_GET_MLD(hw); + + if (WARN_ON(!mld->nvm_data)) + return -EBUSY; + + /* mac80211 ensures the device is not started, + * so the firmware cannot be running + */ + + mld->set_tx_ant = tx_ant; + mld->set_rx_ant = rx_ant; + + iwl_reinit_cab(mld->trans, mld->nvm_data, tx_ant, rx_ant, mld->fw); + + return 0; +} + const struct ieee80211_ops iwl_mld_hw_ops = { .tx = iwl_mld_mac80211_tx, .start = iwl_mld_mac80211_start, .stop = iwl_mld_mac80211_stop, .config = iwl_mld_mac80211_config, + .get_antenna = iwl_mld_get_antenna, + .set_antenna = iwl_mld_set_antenna, .add_interface = iwl_mld_mac80211_add_interface, .remove_interface = iwl_mld_mac80211_remove_interface, .conf_tx = iwl_mld_mac80211_conf_tx, diff --git a/drivers/net/wireless/intel/iwlwifi/mld/mld.h b/drivers/net/wireless/intel/iwlwifi/mld/mld.h index 94dc9da6360d..b1d44fdaa61b 100644 --- a/drivers/net/wireless/intel/iwlwifi/mld/mld.h +++ b/drivers/net/wireless/intel/iwlwifi/mld/mld.h @@ -180,6 +180,8 @@ * @mcast_filter_cmd: pointer to the multicast filter command. * @mgmt_tx_ant: stores the last TX antenna index; used for setting * TX rate_n_flags for non-STA mgmt frames (toggles on every TX failure). + * @set_tx_ant: stores the last TX antenna bitmask set by user space (if any) + * @set_rx_ant: stores the last RX antenna bitmask set by user space (if any) * @fw_rates_ver_3: FW rates are in version 3 * @low_latency: low-latency manager. * @tzone: thermal zone device's data @@ -279,6 +281,9 @@ struct iwl_mld { u8 mgmt_tx_ant; + u8 set_tx_ant; + u8 set_rx_ant; + bool fw_rates_ver_3; struct iwl_mld_low_latency low_latency; @@ -374,6 +379,9 @@ static inline u8 iwl_mld_get_valid_tx_ant(const struct iwl_mld *mld) if (mld->nvm_data && mld->nvm_data->valid_tx_ant) tx_ant &= mld->nvm_data->valid_tx_ant; + if (mld->set_tx_ant) + tx_ant &= mld->set_tx_ant; + return tx_ant; } @@ -384,6 +392,9 @@ static inline u8 iwl_mld_get_valid_rx_ant(const struct iwl_mld *mld) if (mld->nvm_data && mld->nvm_data->valid_rx_ant) rx_ant &= mld->nvm_data->valid_rx_ant; + if (mld->set_rx_ant) + rx_ant &= mld->set_rx_ant; + return rx_ant; } From d676e01357682c966b5e135aa3359fd90be85be9 Mon Sep 17 00:00:00 2001 From: Miri Korenblit Date: Mon, 15 Sep 2025 11:34:32 +0300 Subject: [PATCH 0313/1381] wifi: iwlwifi: mld: set wiphy::iftype_ext_capab dynamically Instead of having a static const array for each possible combination of features, build the extended capabilities dynamically. With this we will also stop setting EHT capabilities when it might actually be disabled. Reviewed-by: Johannes Berg Signed-off-by: Miri Korenblit Link: https://patch.msgid.link/20250915113137.b3c03b56d5a3.I38eaf8ebaf3256e78b4643bef7e3a54aeb4989df@changeid --- .../wireless/intel/iwlwifi/mld/constants.h | 2 + .../net/wireless/intel/iwlwifi/mld/mac80211.c | 63 ++++++++++--------- drivers/net/wireless/intel/iwlwifi/mld/mld.h | 4 ++ 3 files changed, 39 insertions(+), 30 deletions(-) diff --git a/drivers/net/wireless/intel/iwlwifi/mld/constants.h b/drivers/net/wireless/intel/iwlwifi/mld/constants.h index 49accf96f44b..5d23a618ae3c 100644 --- a/drivers/net/wireless/intel/iwlwifi/mld/constants.h +++ b/drivers/net/wireless/intel/iwlwifi/mld/constants.h @@ -75,5 +75,7 @@ #define IWL_MLD_FTM_RESP_LMR_FEEDBACK_SUPPORT true #define IWL_MLD_FTM_NON_TB_MIN_TIME_BETWEEN_MSR 7 #define IWL_MLD_FTM_NON_TB_MAX_TIME_BETWEEN_MSR 1000 +#define IWL_MLD_STA_EXT_CAPA_SIZE 9 +#define IWL_MLD_EXT_CAPA_NUM_IFTYPES 1 #endif /* __iwl_mld_constants_h__ */ diff --git a/drivers/net/wireless/intel/iwlwifi/mld/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mld/mac80211.c index 98d47fed8421..5323c73ac827 100644 --- a/drivers/net/wireless/intel/iwlwifi/mld/mac80211.c +++ b/drivers/net/wireless/intel/iwlwifi/mld/mac80211.c @@ -76,13 +76,12 @@ iwl_mld_iface_combinations[] = { }, }; -static const u8 if_types_ext_capa_sta[] = { - [0] = WLAN_EXT_CAPA1_EXT_CHANNEL_SWITCHING, - [2] = WLAN_EXT_CAPA3_MULTI_BSSID_SUPPORT, - [7] = WLAN_EXT_CAPA8_OPMODE_NOTIF | - WLAN_EXT_CAPA8_MAX_MSDU_IN_AMSDU_LSB, - [8] = WLAN_EXT_CAPA9_MAX_MSDU_IN_AMSDU_MSB, - [9] = WLAN_EXT_CAPA10_TWT_REQUESTER_SUPPORT, +static const u8 ext_capa_base[IWL_MLD_STA_EXT_CAPA_SIZE] = { + [0] = WLAN_EXT_CAPA1_EXT_CHANNEL_SWITCHING, + [2] = WLAN_EXT_CAPA3_MULTI_BSSID_SUPPORT, + [7] = WLAN_EXT_CAPA8_OPMODE_NOTIF | + WLAN_EXT_CAPA8_MAX_MSDU_IN_AMSDU_LSB, + [8] = WLAN_EXT_CAPA9_MAX_MSDU_IN_AMSDU_MSB, }; #define IWL_MLD_EMLSR_CAPA (IEEE80211_EML_CAP_EMLSR_SUPP | \ @@ -95,18 +94,6 @@ static const u8 if_types_ext_capa_sta[] = { IEEE80211_MLD_CAP_OP_TID_TO_LINK_MAP_NEG_SUPP_SAME) | \ IEEE80211_MLD_CAP_OP_LINK_RECONF_SUPPORT) -static const struct wiphy_iftype_ext_capab iftypes_ext_capa[] = { - { - .iftype = NL80211_IFTYPE_STATION, - .extended_capabilities = if_types_ext_capa_sta, - .extended_capabilities_mask = if_types_ext_capa_sta, - .extended_capabilities_len = sizeof(if_types_ext_capa_sta), - /* relevant only if EHT is supported */ - .eml_capabilities = IWL_MLD_EMLSR_CAPA, - .mld_capa_and_ops = IWL_MLD_CAPA_OPS, - }, -}; - static void iwl_mld_hw_set_addresses(struct iwl_mld *mld) { struct wiphy *wiphy = mld->wiphy; @@ -336,21 +323,37 @@ static void iwl_mac_hw_set_wiphy(struct iwl_mld *mld) if (fw_has_capa(ucode_capa, IWL_UCODE_TLV_CAPA_PROTECTED_TWT)) wiphy_ext_feature_set(wiphy, NL80211_EXT_FEATURE_PROTECTED_TWT); - wiphy->iftype_ext_capab = NULL; - wiphy->num_iftype_ext_capab = 0; - - if (!iwlwifi_mod_params.disable_11ax) { - wiphy->iftype_ext_capab = iftypes_ext_capa; - wiphy->num_iftype_ext_capab = ARRAY_SIZE(iftypes_ext_capa); - - ieee80211_hw_set(hw, SUPPORTS_MULTI_BSSID); - ieee80211_hw_set(hw, SUPPORTS_ONLY_HE_MULTI_BSSID); - } - if (iwlmld_mod_params.power_scheme != IWL_POWER_SCHEME_CAM) wiphy->flags |= WIPHY_FLAG_PS_ON_BY_DEFAULT; else wiphy->flags &= ~WIPHY_FLAG_PS_ON_BY_DEFAULT; + + /* We are done for non-HE */ + if (iwlwifi_mod_params.disable_11ax) + return; + + ieee80211_hw_set(hw, SUPPORTS_MULTI_BSSID); + ieee80211_hw_set(hw, SUPPORTS_ONLY_HE_MULTI_BSSID); + + wiphy->iftype_ext_capab = mld->ext_capab; + wiphy->num_iftype_ext_capab = ARRAY_SIZE(mld->ext_capab); + + BUILD_BUG_ON(sizeof(mld->sta_ext_capab) < sizeof(ext_capa_base)); + + memcpy(mld->sta_ext_capab, ext_capa_base, sizeof(ext_capa_base)); + + mld->ext_capab[0].iftype = NL80211_IFTYPE_STATION; + mld->ext_capab[0].extended_capabilities = mld->sta_ext_capab; + mld->ext_capab[0].extended_capabilities_mask = mld->sta_ext_capab; + mld->ext_capab[0].extended_capabilities_len = sizeof(mld->sta_ext_capab); + + if (!mld->nvm_data->sku_cap_11be_enable || + iwlwifi_mod_params.disable_11be) + return; + + mld->ext_capab[0].eml_capabilities = IWL_MLD_EMLSR_CAPA; + mld->ext_capab[0].mld_capa_and_ops = IWL_MLD_CAPA_OPS; + } static void iwl_mac_hw_set_misc(struct iwl_mld *mld) diff --git a/drivers/net/wireless/intel/iwlwifi/mld/mld.h b/drivers/net/wireless/intel/iwlwifi/mld/mld.h index b1d44fdaa61b..ceda12c1672d 100644 --- a/drivers/net/wireless/intel/iwlwifi/mld/mld.h +++ b/drivers/net/wireless/intel/iwlwifi/mld/mld.h @@ -134,6 +134,8 @@ * @fw: a pointer to the fw object * @hw: pointer to the hw object. * @wiphy: a pointer to the wiphy struct, for easier access to it. + * @ext_capab: extended capabilities that will be set to wiphy on registration. + * @sta_ext_capab: extended capabilities for the station interface. * @nvm_data: pointer to the nvm_data that includes all our capabilities * @fwrt: fw runtime data * @debugfs_dir: debugfs directory @@ -227,6 +229,8 @@ struct iwl_mld { const struct iwl_fw *fw; struct ieee80211_hw *hw; struct wiphy *wiphy; + struct wiphy_iftype_ext_capab ext_capab[IWL_MLD_EXT_CAPA_NUM_IFTYPES]; + u8 sta_ext_capab[IWL_MLD_STA_EXT_CAPA_SIZE]; struct iwl_nvm_data *nvm_data; struct iwl_fw_runtime fwrt; struct dentry *debugfs_dir; From 92e87cee465cb5c293aa0a88c4cb5fe5e618583c Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Sun, 19 Oct 2025 11:45:02 +0300 Subject: [PATCH 0314/1381] wifi: iwlwifi: mld: update to new sniffer API This will break current sniffer functionality for firmware versions that don't have the new API, but supporting both would be very complex. Convert the code to use only the new sniffer notification. Signed-off-by: Johannes Berg Signed-off-by: Miri Korenblit Link: https://patch.msgid.link/20251019114304.85b75a084a2f.I4a18b2043703c1f9a8f55c108dcaaeca7891e19c@changeid --- .../wireless/intel/iwlwifi/fw/api/datapath.h | 5 + .../net/wireless/intel/iwlwifi/fw/api/rx.h | 286 +++ .../net/wireless/intel/iwlwifi/mld/mac80211.c | 6 +- drivers/net/wireless/intel/iwlwifi/mld/mld.c | 1 + drivers/net/wireless/intel/iwlwifi/mld/mld.h | 10 +- .../net/wireless/intel/iwlwifi/mld/notif.c | 4 +- drivers/net/wireless/intel/iwlwifi/mld/rx.c | 1663 +++++++++-------- drivers/net/wireless/intel/iwlwifi/mld/rx.h | 5 +- 8 files changed, 1178 insertions(+), 802 deletions(-) diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/datapath.h b/drivers/net/wireless/intel/iwlwifi/fw/api/datapath.h index b1c6ee8ae2df..6a6e11a57dbf 100644 --- a/drivers/net/wireless/intel/iwlwifi/fw/api/datapath.h +++ b/drivers/net/wireless/intel/iwlwifi/fw/api/datapath.h @@ -123,6 +123,11 @@ enum iwl_data_path_subcmd_ids { */ BEACON_FILTER_IN_NOTIF = 0xF8, + /** + * @PHY_AIR_SNIFFER_NOTIF: &struct iwl_rx_phy_air_sniffer_ntfy + */ + PHY_AIR_SNIFFER_NOTIF = 0xF9, + /** * @STA_PM_NOTIF: &struct iwl_mvm_pm_state_notification */ diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/rx.h b/drivers/net/wireless/intel/iwlwifi/fw/api/rx.h index d751789998ac..3ed7e0807b90 100644 --- a/drivers/net/wireless/intel/iwlwifi/fw/api/rx.h +++ b/drivers/net/wireless/intel/iwlwifi/fw/api/rx.h @@ -262,6 +262,7 @@ enum iwl_rx_mpdu_reorder_data { }; enum iwl_rx_mpdu_phy_info { + IWL_RX_MPDU_PHY_EOF_INDICATION = BIT(0), IWL_RX_MPDU_PHY_AMPDU = BIT(5), IWL_RX_MPDU_PHY_AMPDU_TOGGLE = BIT(6), IWL_RX_MPDU_PHY_SHORT_PREAMBLE = BIT(7), @@ -1041,4 +1042,289 @@ struct iwl_beacon_filter_notif { __le32 link_id; } __packed; /* BEACON_FILTER_IN_NTFY_API_S_VER_2 */ +union iwl_legacy_sig { +#define OFDM_RX_LEGACY_LENGTH 0x00000fff +#define OFDM_RX_RATE 0x0000f000 + __le32 ofdm; +#define CCK_CRFR_SHORT_PREAMBLE 0x00000040 + __le32 cck; +}; + +struct iwl_ht_sigs { +#define OFDM_RX_FRAME_HT_MCS 0x0000007f +#define OFDM_RX_FRAME_HT_BANDWIDTH 0x00000080 +#define OFDM_RX_FRAME_HT_LENGTH 0x03ffff00 + __le32 a1; + __le32 a2; +}; + +struct iwl_vht_sigs { +#define OFDM_RX_FRAME_VHT_NUM_OF_DATA_SYM 0x000007ff +#define OFDM_RX_FRAME_VHT_NUM_OF_DATA_SYM_VALID 0x80000000 + __le32 a0; + __le32 a1, a2; +}; + +struct iwl_he_sigs { +#define OFDM_RX_FRAME_HE_BEAM_CHANGE 0x00000001 +#define OFDM_RX_FRAME_HE_UL_FLAG 0x00000002 +#define OFDM_RX_FRAME_HE_MCS 0x0000003c +#define OFDM_RX_FRAME_HE_DCM 0x00000040 +#define OFDM_RX_FRAME_HE_BSS_COLOR 0x00001f80 +#define OFDM_RX_FRAME_HE_SPATIAL_REUSE 0x0001e000 +#define OFDM_RX_FRAME_HE_BANDWIDTH 0x00060000 +#define OFDM_RX_FRAME_HE_SU_EXT_BW10 0x00080000 +#define OFDM_RX_FRAME_HE_GI_LTF_TYPE 0x00700000 +#define OFDM_RX_FRAME_HE_NSTS 0x03800000 +#define OFDM_RX_FRAME_HE_PRMBL_PUNC_TYPE 0x0c000000 + __le32 a1; +#define OFDM_RX_FRAME_HE_TXOP_DURATION 0x0000007f +#define OFDM_RX_FRAME_HE_CODING 0x00000080 +#define OFDM_RX_FRAME_HE_CODING_EXTRA_SYM 0x00000100 +#define OFDM_RX_FRAME_HE_STBC 0x00000200 +#define OFDM_RX_FRAME_HE_BF 0x00000400 +#define OFDM_RX_FRAME_HE_PRE_FEC_PAD_FACTOR 0x00001800 +#define OFDM_RX_FRAME_HE_PE_DISAMBIG 0x00002000 +#define OFDM_RX_FRAME_HE_DOPPLER 0x00004000 +#define OFDM_RX_FRAME_HE_TYPE 0x00038000 +#define OFDM_RX_FRAME_HE_MU_NUM_OF_SIGB_SYM_OR_USER_NUM 0x003c0000 +#define OFDM_RX_FRAME_HE_MU_SIGB_COMP 0x00400000 +#define OFDM_RX_FRAME_HE_MU_NUM_OF_LTF_SYM 0x03800000 + __le32 a2; +#define OFDM_RX_FRAME_HE_NUM_OF_DATA_SYM 0x000007ff +#define OFDM_RX_FRAME_HE_PE_DURATION 0x00003800 +#define OFDM_RX_FRAME_HE_NUM_OF_DATA_SYM_VALID 0x80000000 + __le32 a3; +#define OFDM_RX_FRAME_HE_SIGB_STA_ID_FOUND 0x00000001 +#define OFDM_RX_FRAME_HE_SIGB_STA_ID_INDX 0x0000000e +#define OFDM_RX_FRAME_HE_SIGB_NSTS 0x00000070 +#define OFDM_RX_FRAME_HE_SIGB_BF 0x00000080 +#define OFDM_RX_FRAME_HE_SIGB_MCS 0x00000f00 +#define OFDM_RX_FRAME_HE_SIGB_DCM 0x00001000 +#define OFDM_RX_FRAME_HE_SIGB_CODING 0x00002000 +#define OFDM_RX_FRAME_HE_SIGB_SPATIAL_CONFIG 0x0003c000 +#define OFDM_RX_FRAME_HE_SIGB_STA_RU 0x03fc0000 +#define OFDM_RX_FRAME_HE_SIGB_NUM_OF_SYM 0x3c000000 +#define OFDM_RX_FRAME_HE_SIGB_CRC_OK 0x40000000 + __le32 b; +/* index 0 */ +#define OFDM_RX_FRAME_HE_RU_ALLOC_0_A1 0x000000ff +#define OFDM_RX_FRAME_HE_RU_ALLOC_0_A2 0x0000ff00 +#define OFDM_RX_FRAME_HE_RU_ALLOC_0_B1 0x00ff0000 +#define OFDM_RX_FRAME_HE_RU_ALLOC_0_B2 0xff000000 +/* index 1 */ +#define OFDM_RX_FRAME_HE_RU_ALLOC_1_C1 0x000000ff +#define OFDM_RX_FRAME_HE_RU_ALLOC_1_C2 0x0000ff00 +#define OFDM_RX_FRAME_HE_RU_ALLOC_1_D1 0x00ff0000 +#define OFDM_RX_FRAME_HE_RU_ALLOC_1_D2 0xff000000 +/* index 2 */ +#define OFDM_RX_FRAME_HE_CENTER_RU_CC1 0x00000001 +#define OFDM_RX_FRAME_HE_CENTER_RU_CC2 0x00000002 +#define OFDM_RX_FRAME_HE_COMMON_CC1_CRC_OK 0x00000004 +#define OFDM_RX_FRAME_HE_COMMON_CC2_CRC_OK 0x00000008 + __le32 cmn[3]; +}; + +struct iwl_he_tb_sigs { +#define OFDM_RX_HE_TRIG_FORMAT 0x00000001 +#define OFDM_RX_HE_TRIG_BSS_COLOR 0x0000007e +#define OFDM_RX_HE_TRIG_SPATIAL_REUSE_1 0x00000780 +#define OFDM_RX_HE_TRIG_SPATIAL_REUSE_2 0x00007800 +#define OFDM_RX_HE_TRIG_SPATIAL_REUSE_3 0x00078000 +#define OFDM_RX_HE_TRIG_SPATIAL_REUSE_4 0x00780000 +#define OFDM_RX_HE_TRIG_BANDWIDTH 0x03000000 + __le32 a1; +#define OFDM_RX_HE_TRIG_TXOP_DURATION 0x0000007f +#define OFDM_RX_HE_TRIG_SIG2_RESERVED 0x0000ff80 +#define OFDM_RX_HE_TRIG_FORMAT_ERR 0x08000000 +#define OFDM_RX_HE_TRIG_BW_ERR 0x10000000 +#define OFDM_RX_HE_TRIG_LEGACY_LENGTH_ERR 0x20000000 +#define OFDM_RX_HE_TRIG_CRC_OK 0x40000000 + __le32 a2; +#define OFDM_UCODE_TRIG_BASE_RX_LGCY_LENGTH 0x00000fff +#define OFDM_UCODE_TRIG_BASE_RX_BANDWIDTH 0x00007000 +#define OFDM_UCODE_TRIG_BASE_PS160 0x00008000 +#define OFDM_UCODE_EHT_TRIG_CONTROL_CHANNEL 0x000f0000 + __le32 tb_rx0; +#define OFDM_UCODE_TRIG_BASE_RX_MCS 0x0000000f +#define OFDM_UCODE_TRIG_BASE_RX_DCM 0x00000010 +#define OFDM_UCODE_TRIG_BASE_RX_GI_LTF_TYPE 0x00000060 +#define OFDM_UCODE_TRIG_BASE_RX_NSTS 0x00000380 +#define OFDM_UCODE_TRIG_BASE_RX_CODING 0x00000400 +#define OFDM_UCODE_TRIG_BASE_RX_CODING_EXTRA_SYM 0x00000800 +#define OFDM_UCODE_TRIG_BASE_RX_STBC 0x00001000 +#define OFDM_UCODE_TRIG_BASE_RX_PRE_FEC_PAD_FACTOR 0x00006000 +#define OFDM_UCODE_TRIG_BASE_RX_PE_DISAMBIG 0x00008000 +#define OFDM_UCODE_TRIG_BASE_RX_DOPPLER 0x00010000 +#define OFDM_UCODE_TRIG_BASE_RX_RU 0x01fe0000 +#define OFDM_UCODE_TRIG_BASE_RX_RU_P80 0x00020000 +#define OFDM_UCODE_TRIG_BASE_RX_NUM_OF_LTF_SYM 0x0e000000 +#define OFDM_UCODE_TRIG_BASE_RX_LTF_PILOT_TYPE 0x10000000 +#define OFDM_UCODE_TRIG_BASE_RX_LOWEST_SS_ALLOCATION 0xe0000000 + __le32 tb_rx1; +}; + +struct iwl_eht_sigs { +#define OFDM_RX_FRAME_ENHANCED_WIFI_VER_ID 0x00000007 +#define OFDM_RX_FRAME_ENHANCED_WIFI_BANDWIDTH 0x00000038 +#define OFDM_RX_FRAME_ENHANCED_WIFI_UL_FLAG 0x00000040 +#define OFDM_RX_FRAME_ENHANCED_WIFI_BSS_COLOR 0x00001f80 +#define OFDM_RX_FRAME_ENHANCED_WIFI_TXOP_DURATION 0x000fe000 +#define OFDM_RX_FRAME_EHT_USIG1_DISREGARD 0x01f00000 +#define OFDM_RX_FRAME_EHT_USIG1_VALIDATE 0x02000000 +#define OFDM_RX_FRAME_EHT_BW320_SLOT 0x04000000 +#define OFDM_RX_FRAME_EHT_TYPE 0x18000000 +#define OFDM_RX_FRAME_ENHANCED_ER_NO_STREAMS 0x20000000 + __le32 usig_a1; +#define OFDM_RX_FRAME_EHT_PPDU_TYPE 0x00000003 +#define OFDM_RX_FRAME_EHT_USIG2_VALIDATE_B2 0x00000004 +#define OFDM_RX_FRAME_EHT_PUNC_CHANNEL 0x000000f8 +#define OFDM_RX_FRAME_EHT_USIG2_VALIDATE_B8 0x00000100 +#define OFDM_RX_FRAME_EHT_SIG_MCS 0x00000600 +#define OFDM_RX_FRAME_EHT_SIG_SYM_NUM 0x0000f800 +#define OFDM_RX_FRAME_EHT_TRIG_SPATIAL_REUSE_1 0x000f0000 +#define OFDM_RX_FRAME_EHT_TRIG_SPATIAL_REUSE_2 0x00f00000 +#define OFDM_RX_FRAME_EHT_TRIG_USIG2_DISREGARD 0x1f000000 +#define OFDM_RX_FRAME_EHT_TRIG_NO_STREAMS 0x20000000 +#define OFDM_RX_USIG_CRC_OK 0x40000000 + __le32 usig_a2_eht; +#define OFDM_RX_FRAME_EHT_SPATIAL_REUSE 0x0000000f +#define OFDM_RX_FRAME_EHT_GI_LTF_TYPE 0x00000030 +#define OFDM_RX_FRAME_EHT_NUM_OF_LTF_SYM 0x000001c0 +#define OFDM_RX_FRAME_EHT_CODING_EXTRA_SYM 0x00000200 +#define OFDM_RX_FRAME_EHT_PRE_FEC_PAD_FACTOR 0x00000c00 +#define OFDM_RX_FRAME_EHT_PE_DISAMBIG 0x00001000 +#define OFDM_RX_FRAME_EHT_USIG_OVF_DISREGARD 0x0001e000 +#define OFDM_RX_FRAME_EHT_NUM_OF_USERS 0x000e0000 +#define OFDM_RX_FRAME_EHT_NSTS 0x00f00000 +#define OFDM_RX_FRAME_EHT_BF 0x01000000 +#define OFDM_RX_FRAME_EHT_USIG_OVF_NDP_DISREGARD 0x06000000 +#define OFDM_RX_FRAME_EHTSIG_COMM_CC1_CRC_OK 0x08000000 +#define OFDM_RX_FRAME_EHTSIG_COMM_CC2_CRC_OK 0x10000000 +#define OFDM_RX_FRAME_EHT_NON_VALID_RU_ALLOC 0x20000000 +#define OFDM_RX_FRAME_EHT_NO_STREAMS 0x40000000 + __le32 b1; +#define OFDM_RX_FRAME_EHT_MATCH_ID_FOUND 0x00000001 +#define OFDM_RX_FRAME_EHT_ID_INDX 0x0000000e +#define OFDM_RX_FRAME_EHT_MCS 0x000000f0 +#define OFDM_RX_FRAME_EHT_CODING 0x00000100 +#define OFDM_RX_FRAME_EHT_SPATIAL_CONFIG 0x00007e00 +#define OFDM_RX_FRAME_EHT_STA_RU 0x007f8000 +#define OFDM_RX_FRAME_EHT_STA_RU_P80 0x00008000 +#define OFDM_RX_FRAME_EHT_STA_RU_PS160 0x00800000 +#define OFDM_RX_FRAME_EHT_USER_FIELD_CRC_OK 0x40000000 + __le32 b2; +#define OFDM_RX_FRAME_EHT_NUM_OF_DATA_SYM 0x000007ff +#define OFDM_RX_FRAME_EHT_PE_DURATION 0x00003800 +#define OFDM_RX_FRAME_EHT_NUM_OF_DATA_SYM_VALID 0x80000000 + __le32 sig2; +#define OFDM_RX_FRAME_EHT_RU_ALLOC_0_A1 0x000001ff +#define OFDM_RX_FRAME_EHT_RU_ALLOC_0_A2 0x0003fe00 +#define OFDM_RX_FRAME_EHT_RU_ALLOC_0_A3 0x07fc0000 +#define OFDM_RX_FRAME_EHT_RU_ALLOC_1_B1 0x000001ff +#define OFDM_RX_FRAME_EHT_RU_ALLOC_1_B2 0x0003fe00 +#define OFDM_RX_FRAME_EHT_RU_ALLOC_1_B3 0x07fc0000 +#define OFDM_RX_FRAME_EHT_RU_ALLOC_2_C1 0x000001ff +#define OFDM_RX_FRAME_EHT_RU_ALLOC_2_C2 0x0003fe00 +#define OFDM_RX_FRAME_EHT_RU_ALLOC_2_C3 0x07fc0000 +#define OFDM_RX_FRAME_EHT_RU_ALLOC_3_D1 0x000001ff +#define OFDM_RX_FRAME_EHT_RU_ALLOC_3_D2 0x0003fe00 +#define OFDM_RX_FRAME_EHT_RU_ALLOC_3_D3 0x07fc0000 +#define OFDM_RX_FRAME_EHT_RU_ALLOC_4_A4 0x000001ff +#define OFDM_RX_FRAME_EHT_RU_ALLOC_4_B4 0x0003fe00 +#define OFDM_RX_FRAME_EHT_RU_ALLOC_5_C4 0x000001ff +#define OFDM_RX_FRAME_EHT_RU_ALLOC_5_D4 0x0003fe00 + __le32 cmn[6]; +#define OFDM_RX_FRAME_EHT_USER_FIELD_ID 0x000007ff + __le32 user_id; +}; + +struct iwl_eht_tb_sigs { + /* same as non-TB above */ + __le32 usig_a1, usig_a2_eht; + /* same as HE TB above */ + __le32 tb_rx0, tb_rx1; +}; + +struct iwl_uhr_sigs { + __le32 usig_a1, usig_a1_uhr, usig_a2_uhr, b1, b2; + __le32 sig2; + __le32 cmn[6]; + __le32 user_id; +}; + +struct iwl_uhr_tb_sigs { + __le32 usig_a1, usig_a2_uhr, tb_rx0, tb_rx1; +}; + +struct iwl_uhr_elr_sigs { + __le32 usig_a1, usig_a2_uhr; + __le32 uhr_sig_elr1, uhr_sig_elr2; +}; + +union iwl_sigs { + struct iwl_ht_sigs ht; + struct iwl_vht_sigs vht; + struct iwl_he_sigs he; + struct iwl_he_tb_sigs he_tb; + struct iwl_eht_sigs eht; + struct iwl_eht_tb_sigs eht_tb; + struct iwl_uhr_sigs uhr; + struct iwl_uhr_tb_sigs uhr_tb; + struct iwl_uhr_elr_sigs uhr_elr; +}; + +enum iwl_sniffer_status { + IWL_SNIF_STAT_PLCP_RX_OK = 0, + IWL_SNIF_STAT_AID_NOT_FOR_US = 1, + IWL_SNIF_STAT_PLCP_RX_LSIG_ERR = 2, + IWL_SNIF_STAT_PLCP_RX_SIGA_ERR = 3, + IWL_SNIF_STAT_PLCP_RX_SIGB_ERR = 4, + IWL_SNIF_STAT_UNEXPECTED_TB = 5, + IWL_SNIF_STAT_UNSUPPORTED_RATE = 6, + IWL_SNIF_STAT_UNKNOWN_ERROR = 7, +}; /* AIR_SNIFFER_STATUS_E_VER_1 */ + +enum iwl_sniffer_flags { + IWL_SNIF_FLAG_VALID_TB_RX = BIT(0), + IWL_SNIF_FLAG_VALID_RU = BIT(1), +}; /* AIR_SNIFFER_FLAGS_E_VER_1 */ + +/** + * struct iwl_rx_phy_air_sniffer_ntfy - air sniffer notification + * + * @status: &enum iwl_sniffer_status + * @flags: &enum iwl_sniffer_flags + * @reserved1: reserved + * @rssi_a: energy chain-A in negative dBm, measured at FINA time + * @rssi_b: energy chain-B in negative dBm, measured at FINA time + * @channel: channel number + * @band: band information, PHY_BAND_* + * @on_air_rise_time: GP2 at on air rise + * @frame_time: frame time in us + * @rate: RATE_MCS_* + * @bytecount: byte count for legay and HT, otherwise number of symbols + * @legacy_sig: CCK signal information if %RATE_MCS_MOD_TYPE_MSK in @rate is + * %RATE_MCS_MOD_TYPE_CCK, otherwise OFDM signal information + * @sigs: PHY signal information, depending on %RATE_MCS_MOD_TYPE_MSK in @rate + * @reserved2: reserved + * + * Sent for every frame and before the normal RX command if data is included. + */ +struct iwl_rx_phy_air_sniffer_ntfy { + u8 status; + u8 flags; + u8 reserved1[2]; + u8 rssi_a, rssi_b; + u8 channel, band; + __le32 on_air_rise_time; + __le32 frame_time; + /* note: MCS in rate is not valid for MU-VHT */ + __le32 rate; + __le32 bytecount; + union iwl_legacy_sig legacy_sig; + union iwl_sigs sigs; + __le32 reserved2; +}; /* RX_PHY_AIR_SNIFFER_NTFY_API_S_VER_1 */ + #endif /* __iwl_fw_api_rx_h__ */ diff --git a/drivers/net/wireless/intel/iwlwifi/mld/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mld/mac80211.c index 5323c73ac827..55b484c16280 100644 --- a/drivers/net/wireless/intel/iwlwifi/mld/mac80211.c +++ b/drivers/net/wireless/intel/iwlwifi/mld/mac80211.c @@ -397,11 +397,9 @@ static int iwl_mld_hw_verify_preconditions(struct iwl_mld *mld) TLC_MNG_UPDATE_NOTIF, 0) >= 4) + (iwl_fw_lookup_notif_ver(mld->fw, LEGACY_GROUP, REPLY_RX_MPDU_CMD, 0) >= 6) + - (iwl_fw_lookup_notif_ver(mld->fw, DATA_PATH_GROUP, - RX_NO_DATA_NOTIF, 0) >= 4) + (iwl_fw_lookup_notif_ver(mld->fw, LONG_GROUP, TX_CMD, 0) >= 9); - if (ratecheck != 0 && ratecheck != 5) { + if (ratecheck != 0 && ratecheck != 4) { IWL_ERR(mld, "Firmware has inconsistent rates\n"); return -EINVAL; } @@ -684,6 +682,8 @@ void iwl_mld_mac80211_remove_interface(struct ieee80211_hw *hw, #endif iwl_mld_rm_vif(mld, vif); + + mld->monitor.phy.valid = false; } struct iwl_mld_mc_iter_data { diff --git a/drivers/net/wireless/intel/iwlwifi/mld/mld.c b/drivers/net/wireless/intel/iwlwifi/mld/mld.c index a6962256bdd1..8a4c96385640 100644 --- a/drivers/net/wireless/intel/iwlwifi/mld/mld.c +++ b/drivers/net/wireless/intel/iwlwifi/mld/mld.c @@ -259,6 +259,7 @@ static const struct iwl_hcmd_names iwl_mld_data_path_names[] = { HCMD_NAME(MONITOR_NOTIF), HCMD_NAME(TLC_MNG_UPDATE_NOTIF), HCMD_NAME(BEACON_FILTER_IN_NOTIF), + HCMD_NAME(PHY_AIR_SNIFFER_NOTIF), HCMD_NAME(MU_GROUP_MGMT_NOTIF), }; diff --git a/drivers/net/wireless/intel/iwlwifi/mld/mld.h b/drivers/net/wireless/intel/iwlwifi/mld/mld.h index ceda12c1672d..22efe8e10f53 100644 --- a/drivers/net/wireless/intel/iwlwifi/mld/mld.h +++ b/drivers/net/wireless/intel/iwlwifi/mld/mld.h @@ -118,7 +118,11 @@ * @monitor.cur_bssid: current bssid tracked by the sniffer * @monitor.ptp_time: set the Rx mactime using the device's PTP clock time * @monitor.p80: primary channel position relative to he whole bandwidth, in - * steps of 80 MHz + * steps of 80 MHz + * @monitor.phy: PHY data information + * @monitor.phy.data: PHY data (&struct iwl_rx_phy_air_sniffer_ntfy) received + * @monitor.phy.valid: PHY data is valid (was received) + * @monitor.phy.used: PHY data was used by an RX * @fw_id_to_link_sta: maps a fw id of a sta to the corresponding * ieee80211_link_sta. This is not cleaned up on restart since we want to * preserve the fw sta ids during a restart (for SN/PN restoring). @@ -209,6 +213,10 @@ struct iwl_mld { u32 ampdu_ref; bool ampdu_toggle; u8 p80; + struct { + struct iwl_rx_phy_air_sniffer_ntfy data; + u8 valid:1, used:1; + } phy; #ifdef CONFIG_IWLWIFI_DEBUGFS __le16 cur_aid; u8 cur_bssid[ETH_ALEN]; diff --git a/drivers/net/wireless/intel/iwlwifi/mld/notif.c b/drivers/net/wireless/intel/iwlwifi/mld/notif.c index 884973d0b344..4cf3920b005f 100644 --- a/drivers/net/wireless/intel/iwlwifi/mld/notif.c +++ b/drivers/net/wireless/intel/iwlwifi/mld/notif.c @@ -589,8 +589,8 @@ void iwl_mld_rx(struct iwl_op_mode *op_mode, struct napi_struct *napi, else if (unlikely(cmd_id == WIDE_ID(DATA_PATH_GROUP, RX_QUEUES_NOTIFICATION))) iwl_mld_handle_rx_queues_sync_notif(mld, napi, pkt, 0); - else if (cmd_id == WIDE_ID(DATA_PATH_GROUP, RX_NO_DATA_NOTIF)) - iwl_mld_rx_monitor_no_data(mld, napi, pkt, 0); + else if (cmd_id == WIDE_ID(DATA_PATH_GROUP, PHY_AIR_SNIFFER_NOTIF)) + iwl_mld_handle_phy_air_sniffer_notif(mld, napi, pkt); else iwl_mld_rx_notif(mld, rxb, pkt); } diff --git a/drivers/net/wireless/intel/iwlwifi/mld/rx.c b/drivers/net/wireless/intel/iwlwifi/mld/rx.c index 20d866dd92c2..052a19bb85b4 100644 --- a/drivers/net/wireless/intel/iwlwifi/mld/rx.c +++ b/drivers/net/wireless/intel/iwlwifi/mld/rx.c @@ -18,41 +18,32 @@ /* stores relevant PHY data fields extracted from iwl_rx_mpdu_desc */ struct iwl_mld_rx_phy_data { - enum iwl_rx_phy_info_type info_type; - __le32 data0; - __le32 data1; - __le32 data2; - __le32 data3; - __le32 eht_data4; - __le32 data5; - __le16 data4; + struct iwl_rx_phy_air_sniffer_ntfy *ntfy; bool first_subframe; bool with_data; - __le32 rx_vec[4]; u32 rate_n_flags; u32 gp2_on_air_rise; + /* phy_info is only valid when we have a frame, i.e. with_data=true */ u16 phy_info; u8 energy_a, energy_b; }; static void -iwl_mld_fill_phy_data(struct iwl_mld *mld, - struct iwl_rx_mpdu_desc *desc, - struct iwl_mld_rx_phy_data *phy_data) +iwl_mld_fill_phy_data_from_mpdu(struct iwl_mld *mld, + struct iwl_rx_mpdu_desc *desc, + struct iwl_mld_rx_phy_data *phy_data) { + if (unlikely(mld->monitor.phy.valid)) { + mld->monitor.phy.used = true; + phy_data->ntfy = &mld->monitor.phy.data; + } + phy_data->phy_info = le16_to_cpu(desc->phy_info); phy_data->rate_n_flags = iwl_v3_rate_from_v2_v3(desc->v3.rate_n_flags, mld->fw_rates_ver_3); phy_data->gp2_on_air_rise = le32_to_cpu(desc->v3.gp2_on_air_rise); phy_data->energy_a = desc->v3.energy_a; phy_data->energy_b = desc->v3.energy_b; - phy_data->data0 = desc->v3.phy_data0; - phy_data->data1 = desc->v3.phy_data1; - phy_data->data2 = desc->v3.phy_data2; - phy_data->data3 = desc->v3.phy_data3; - phy_data->data4 = desc->phy_data4; - phy_data->eht_data4 = desc->phy_eht_data4; - phy_data->data5 = desc->v3.phy_data5; phy_data->with_data = true; } @@ -217,26 +208,19 @@ static void iwl_mld_fill_signal(struct iwl_mld *mld, int link_id, } static void -iwl_mld_decode_he_phy_ru_alloc(struct iwl_mld_rx_phy_data *phy_data, - struct ieee80211_radiotap_he *he, - struct ieee80211_radiotap_he_mu *he_mu, - struct ieee80211_rx_status *rx_status) +iwl_mld_he_set_ru_alloc(struct ieee80211_rx_status *rx_status, + struct ieee80211_radiotap_he *he, + u8 ru_with_p80) { - /* Unfortunately, we have to leave the mac80211 data - * incorrect for the case that we receive an HE-MU - * transmission and *don't* have the HE phy data (due - * to the bits being used for TSF). This shouldn't - * happen though as management frames where we need - * the TSF/timers are not be transmitted in HE-MU. - */ - u8 ru = le32_get_bits(phy_data->data1, IWL_RX_PHY_DATA1_HE_RU_ALLOC_MASK); - u32 rate_n_flags = phy_data->rate_n_flags; - u32 he_type = rate_n_flags & RATE_MCS_HE_TYPE_MSK; + u8 ru = ru_with_p80 >> 1; + u8 p80 = ru_with_p80 & 1; u8 offs = 0; rx_status->bw = RATE_INFO_BW_HE_RU; he->data1 |= cpu_to_le16(IEEE80211_RADIOTAP_HE_DATA1_BW_RU_ALLOC_KNOWN); + he->data2 |= cpu_to_le16(IEEE80211_RADIOTAP_HE_DATA2_PRISEC_80_KNOWN | + IEEE80211_RADIOTAP_HE_DATA2_RU_OFFSET_KNOWN); switch (ru) { case 0 ... 36: @@ -266,227 +250,262 @@ iwl_mld_decode_he_phy_ru_alloc(struct iwl_mld_rx_phy_data *phy_data, rx_status->he_ru = NL80211_RATE_INFO_HE_RU_ALLOC_2x996; break; } + he->data2 |= le16_encode_bits(offs, IEEE80211_RADIOTAP_HE_DATA2_RU_OFFSET); - he->data2 |= cpu_to_le16(IEEE80211_RADIOTAP_HE_DATA2_PRISEC_80_KNOWN | - IEEE80211_RADIOTAP_HE_DATA2_RU_OFFSET_KNOWN); - if (phy_data->data1 & cpu_to_le32(IWL_RX_PHY_DATA1_HE_RU_ALLOC_SEC80)) - he->data2 |= - cpu_to_le16(IEEE80211_RADIOTAP_HE_DATA2_PRISEC_80_SEC); -#define CHECK_BW(bw) \ - BUILD_BUG_ON(IEEE80211_RADIOTAP_HE_MU_FLAGS2_BW_FROM_SIG_A_BW_ ## bw ## MHZ != \ - RATE_MCS_CHAN_WIDTH_##bw >> RATE_MCS_CHAN_WIDTH_POS); \ - BUILD_BUG_ON(IEEE80211_RADIOTAP_HE_DATA6_TB_PPDU_BW_ ## bw ## MHZ != \ - RATE_MCS_CHAN_WIDTH_##bw >> RATE_MCS_CHAN_WIDTH_POS) - CHECK_BW(20); - CHECK_BW(40); - CHECK_BW(80); - CHECK_BW(160); - - if (he_mu) - he_mu->flags2 |= - le16_encode_bits(u32_get_bits(rate_n_flags, - RATE_MCS_CHAN_WIDTH_MSK), - IEEE80211_RADIOTAP_HE_MU_FLAGS2_BW_FROM_SIG_A_BW); - else if (he_type == RATE_MCS_HE_TYPE_TRIG) - he->data6 |= - cpu_to_le16(IEEE80211_RADIOTAP_HE_DATA6_TB_PPDU_BW_KNOWN) | - le16_encode_bits(u32_get_bits(rate_n_flags, - RATE_MCS_CHAN_WIDTH_MSK), - IEEE80211_RADIOTAP_HE_DATA6_TB_PPDU_BW); + he->data2 |= le16_encode_bits(p80, IEEE80211_RADIOTAP_HE_DATA2_PRISEC_80_SEC); } +#define RTAP_ENC_HE(src, src_msk, dst_msk) \ + le16_encode_bits(le32_get_bits(src, src_msk), dst_msk) + static void -iwl_mld_decode_he_mu_ext(struct iwl_mld_rx_phy_data *phy_data, - struct ieee80211_radiotap_he_mu *he_mu) +iwl_mld_decode_he_mu(struct iwl_mld_rx_phy_data *phy_data, + struct ieee80211_radiotap_he *he, + struct ieee80211_radiotap_he_mu *he_mu, + struct ieee80211_rx_status *rx_status) { - u32 phy_data2 = le32_to_cpu(phy_data->data2); - u32 phy_data3 = le32_to_cpu(phy_data->data3); - u16 phy_data4 = le16_to_cpu(phy_data->data4); u32 rate_n_flags = phy_data->rate_n_flags; - if (u32_get_bits(phy_data4, IWL_RX_PHY_DATA4_HE_MU_EXT_CH1_CRC_OK)) { + he_mu->flags1 |= RTAP_ENC_HE(phy_data->ntfy->sigs.he.b, + OFDM_RX_FRAME_HE_SIGB_DCM, + IEEE80211_RADIOTAP_HE_MU_FLAGS1_SIG_B_DCM); + he_mu->flags1 |= RTAP_ENC_HE(phy_data->ntfy->sigs.he.b, + OFDM_RX_FRAME_HE_SIGB_MCS, + IEEE80211_RADIOTAP_HE_MU_FLAGS1_SIG_B_MCS); + he_mu->flags2 |= RTAP_ENC_HE(phy_data->ntfy->sigs.he.a1, + OFDM_RX_FRAME_HE_PRMBL_PUNC_TYPE, + IEEE80211_RADIOTAP_HE_MU_FLAGS2_PUNC_FROM_SIG_A_BW); + he_mu->flags2 |= RTAP_ENC_HE(phy_data->ntfy->sigs.he.a2, + OFDM_RX_FRAME_HE_MU_NUM_OF_SIGB_SYM_OR_USER_NUM, + IEEE80211_RADIOTAP_HE_MU_FLAGS2_SIG_B_SYMS_USERS); + he_mu->flags2 |= RTAP_ENC_HE(phy_data->ntfy->sigs.he.b, + OFDM_RX_FRAME_HE_MU_SIGB_COMP, + IEEE80211_RADIOTAP_HE_MU_FLAGS2_SIG_B_COMP); + + if (phy_data->ntfy->flags & IWL_SNIF_FLAG_VALID_RU && + le32_get_bits(phy_data->ntfy->sigs.he.cmn[2], + OFDM_RX_FRAME_HE_COMMON_CC1_CRC_OK)) { he_mu->flags1 |= cpu_to_le16(IEEE80211_RADIOTAP_HE_MU_FLAGS1_CH1_RU_KNOWN | IEEE80211_RADIOTAP_HE_MU_FLAGS1_CH1_CTR_26T_RU_KNOWN); he_mu->flags1 |= - le16_encode_bits(u32_get_bits(phy_data4, - IWL_RX_PHY_DATA4_HE_MU_EXT_CH1_CTR_RU), - IEEE80211_RADIOTAP_HE_MU_FLAGS1_CH1_CTR_26T_RU); + RTAP_ENC_HE(phy_data->ntfy->sigs.he.cmn[2], + OFDM_RX_FRAME_HE_CENTER_RU_CC1, + IEEE80211_RADIOTAP_HE_MU_FLAGS1_CH1_CTR_26T_RU); - he_mu->ru_ch1[0] = u32_get_bits(phy_data2, - IWL_RX_PHY_DATA2_HE_MU_EXT_CH1_RU0); - he_mu->ru_ch1[1] = u32_get_bits(phy_data3, - IWL_RX_PHY_DATA3_HE_MU_EXT_CH1_RU1); - he_mu->ru_ch1[2] = u32_get_bits(phy_data2, - IWL_RX_PHY_DATA2_HE_MU_EXT_CH1_RU2); - he_mu->ru_ch1[3] = u32_get_bits(phy_data3, - IWL_RX_PHY_DATA3_HE_MU_EXT_CH1_RU3); + he_mu->ru_ch1[0] = le32_get_bits(phy_data->ntfy->sigs.he.cmn[0], + OFDM_RX_FRAME_HE_RU_ALLOC_0_A1); + he_mu->ru_ch1[1] = le32_get_bits(phy_data->ntfy->sigs.he.cmn[1], + OFDM_RX_FRAME_HE_RU_ALLOC_1_C1); + he_mu->ru_ch1[2] = le32_get_bits(phy_data->ntfy->sigs.he.cmn[0], + OFDM_RX_FRAME_HE_RU_ALLOC_0_A2); + he_mu->ru_ch1[3] = le32_get_bits(phy_data->ntfy->sigs.he.cmn[1], + OFDM_RX_FRAME_HE_RU_ALLOC_1_C2); } - if (u32_get_bits(phy_data4, IWL_RX_PHY_DATA4_HE_MU_EXT_CH2_CRC_OK) && + if (phy_data->ntfy->flags & IWL_SNIF_FLAG_VALID_RU && + le32_get_bits(phy_data->ntfy->sigs.he.cmn[2], + OFDM_RX_FRAME_HE_COMMON_CC2_CRC_OK) && (rate_n_flags & RATE_MCS_CHAN_WIDTH_MSK) != RATE_MCS_CHAN_WIDTH_20) { he_mu->flags1 |= cpu_to_le16(IEEE80211_RADIOTAP_HE_MU_FLAGS1_CH2_RU_KNOWN | IEEE80211_RADIOTAP_HE_MU_FLAGS1_CH2_CTR_26T_RU_KNOWN); he_mu->flags2 |= - le16_encode_bits(u32_get_bits(phy_data4, - IWL_RX_PHY_DATA4_HE_MU_EXT_CH2_CTR_RU), - IEEE80211_RADIOTAP_HE_MU_FLAGS2_CH2_CTR_26T_RU); + RTAP_ENC_HE(phy_data->ntfy->sigs.he.cmn[2], + OFDM_RX_FRAME_HE_CENTER_RU_CC2, + IEEE80211_RADIOTAP_HE_MU_FLAGS2_CH2_CTR_26T_RU); - he_mu->ru_ch2[0] = u32_get_bits(phy_data2, - IWL_RX_PHY_DATA2_HE_MU_EXT_CH2_RU0); - he_mu->ru_ch2[1] = u32_get_bits(phy_data3, - IWL_RX_PHY_DATA3_HE_MU_EXT_CH2_RU1); - he_mu->ru_ch2[2] = u32_get_bits(phy_data2, - IWL_RX_PHY_DATA2_HE_MU_EXT_CH2_RU2); - he_mu->ru_ch2[3] = u32_get_bits(phy_data3, - IWL_RX_PHY_DATA3_HE_MU_EXT_CH2_RU3); + he_mu->ru_ch2[0] = le32_get_bits(phy_data->ntfy->sigs.he.cmn[0], + OFDM_RX_FRAME_HE_RU_ALLOC_0_B1); + he_mu->ru_ch2[1] = le32_get_bits(phy_data->ntfy->sigs.he.cmn[1], + OFDM_RX_FRAME_HE_RU_ALLOC_1_D1); + he_mu->ru_ch2[2] = le32_get_bits(phy_data->ntfy->sigs.he.cmn[0], + OFDM_RX_FRAME_HE_RU_ALLOC_0_B2); + he_mu->ru_ch2[3] = le32_get_bits(phy_data->ntfy->sigs.he.cmn[1], + OFDM_RX_FRAME_HE_RU_ALLOC_1_D2); } + +#define CHECK_BW(bw) \ + BUILD_BUG_ON(IEEE80211_RADIOTAP_HE_MU_FLAGS2_BW_FROM_SIG_A_BW_ ## bw ## MHZ != \ + RATE_MCS_CHAN_WIDTH_##bw >> RATE_MCS_CHAN_WIDTH_POS) + CHECK_BW(20); + CHECK_BW(40); + CHECK_BW(80); + CHECK_BW(160); +#undef CHECK_BW + + he_mu->flags2 |= + le16_encode_bits(u32_get_bits(rate_n_flags, RATE_MCS_CHAN_WIDTH_MSK), + IEEE80211_RADIOTAP_HE_MU_FLAGS2_BW_FROM_SIG_A_BW); + + iwl_mld_he_set_ru_alloc(rx_status, he, + le32_get_bits(phy_data->ntfy->sigs.he.b, + OFDM_RX_FRAME_HE_SIGB_STA_RU)); +} + +static void +iwl_mld_decode_he_tb_phy_data(struct iwl_mld_rx_phy_data *phy_data, + struct ieee80211_radiotap_he *he, + struct ieee80211_rx_status *rx_status) +{ + u32 rate_n_flags = phy_data->rate_n_flags; + u32 nsts; + + he->data1 |= cpu_to_le16(IEEE80211_RADIOTAP_HE_DATA1_BSS_COLOR_KNOWN | + IEEE80211_RADIOTAP_HE_DATA1_SPTL_REUSE_KNOWN | + IEEE80211_RADIOTAP_HE_DATA1_SPTL_REUSE2_KNOWN | + IEEE80211_RADIOTAP_HE_DATA1_SPTL_REUSE3_KNOWN | + IEEE80211_RADIOTAP_HE_DATA1_SPTL_REUSE4_KNOWN); + + he->data4 |= RTAP_ENC_HE(phy_data->ntfy->sigs.he_tb.a1, + OFDM_RX_HE_TRIG_SPATIAL_REUSE_1, + IEEE80211_RADIOTAP_HE_DATA4_TB_SPTL_REUSE1); + he->data4 |= RTAP_ENC_HE(phy_data->ntfy->sigs.he_tb.a1, + OFDM_RX_HE_TRIG_SPATIAL_REUSE_2, + IEEE80211_RADIOTAP_HE_DATA4_TB_SPTL_REUSE2); + he->data4 |= RTAP_ENC_HE(phy_data->ntfy->sigs.he_tb.a1, + OFDM_RX_HE_TRIG_SPATIAL_REUSE_3, + IEEE80211_RADIOTAP_HE_DATA4_TB_SPTL_REUSE3); + he->data4 |= RTAP_ENC_HE(phy_data->ntfy->sigs.he_tb.a1, + OFDM_RX_HE_TRIG_SPATIAL_REUSE_4, + IEEE80211_RADIOTAP_HE_DATA4_TB_SPTL_REUSE4); + he->data3 |= RTAP_ENC_HE(phy_data->ntfy->sigs.he_tb.a1, + OFDM_RX_HE_TRIG_BSS_COLOR, + IEEE80211_RADIOTAP_HE_DATA3_BSS_COLOR); + +#define CHECK_BW(bw) \ + BUILD_BUG_ON(IEEE80211_RADIOTAP_HE_DATA6_TB_PPDU_BW_ ## bw ## MHZ != \ + RATE_MCS_CHAN_WIDTH_##bw >> RATE_MCS_CHAN_WIDTH_POS) + CHECK_BW(20); + CHECK_BW(40); + CHECK_BW(80); + CHECK_BW(160); +#undef CHECK_BW + + he->data6 |= cpu_to_le16(IEEE80211_RADIOTAP_HE_DATA6_TB_PPDU_BW_KNOWN) | + le16_encode_bits(u32_get_bits(rate_n_flags, RATE_MCS_CHAN_WIDTH_MSK), + IEEE80211_RADIOTAP_HE_DATA6_TB_PPDU_BW); + + if (!(phy_data->ntfy->flags & IWL_SNIF_FLAG_VALID_TB_RX)) + return; + + he->data1 |= cpu_to_le16(IEEE80211_RADIOTAP_HE_DATA1_LDPC_XSYMSEG_KNOWN | + IEEE80211_RADIOTAP_HE_DATA1_DOPPLER_KNOWN); + he->data2 |= cpu_to_le16(IEEE80211_RADIOTAP_HE_DATA2_PRE_FEC_PAD_KNOWN | + IEEE80211_RADIOTAP_HE_DATA2_PE_DISAMBIG_KNOWN | + IEEE80211_RADIOTAP_HE_DATA2_TXOP_KNOWN | + IEEE80211_RADIOTAP_HE_DATA2_NUM_LTF_SYMS_KNOWN); + + he->data3 |= RTAP_ENC_HE(phy_data->ntfy->sigs.he_tb.tb_rx1, + OFDM_UCODE_TRIG_BASE_RX_CODING_EXTRA_SYM, + IEEE80211_RADIOTAP_HE_DATA3_LDPC_XSYMSEG); + he->data6 |= RTAP_ENC_HE(phy_data->ntfy->sigs.he_tb.tb_rx1, + OFDM_UCODE_TRIG_BASE_RX_DOPPLER, + IEEE80211_RADIOTAP_HE_DATA6_DOPPLER); + he->data5 |= RTAP_ENC_HE(phy_data->ntfy->sigs.he_tb.tb_rx1, + OFDM_UCODE_TRIG_BASE_RX_PRE_FEC_PAD_FACTOR, + IEEE80211_RADIOTAP_HE_DATA5_PRE_FEC_PAD); + he->data5 |= RTAP_ENC_HE(phy_data->ntfy->sigs.he_tb.tb_rx1, + OFDM_UCODE_TRIG_BASE_RX_PE_DISAMBIG, + IEEE80211_RADIOTAP_HE_DATA5_PE_DISAMBIG); + he->data5 |= RTAP_ENC_HE(phy_data->ntfy->sigs.he_tb.tb_rx1, + OFDM_UCODE_TRIG_BASE_RX_NUM_OF_LTF_SYM, + IEEE80211_RADIOTAP_HE_DATA5_NUM_LTF_SYMS); + he->data6 |= RTAP_ENC_HE(phy_data->ntfy->sigs.he_tb.a2, + OFDM_RX_HE_TRIG_TXOP_DURATION, + IEEE80211_RADIOTAP_HE_DATA6_TXOP); + + iwl_mld_he_set_ru_alloc(rx_status, he, + le32_get_bits(phy_data->ntfy->sigs.he_tb.tb_rx1, + OFDM_UCODE_TRIG_BASE_RX_RU)); + + nsts = le32_get_bits(phy_data->ntfy->sigs.he_tb.tb_rx1, + OFDM_UCODE_TRIG_BASE_RX_NSTS) + 1; + rx_status->nss = nsts >> !!(rate_n_flags & RATE_MCS_STBC_MSK); } static void iwl_mld_decode_he_phy_data(struct iwl_mld_rx_phy_data *phy_data, struct ieee80211_radiotap_he *he, struct ieee80211_radiotap_he_mu *he_mu, - struct ieee80211_rx_status *rx_status, - int queue) + struct ieee80211_rx_status *rx_status) { - switch (phy_data->info_type) { - case IWL_RX_PHY_INFO_TYPE_NONE: - case IWL_RX_PHY_INFO_TYPE_CCK: - case IWL_RX_PHY_INFO_TYPE_OFDM_LGCY: - case IWL_RX_PHY_INFO_TYPE_HT: - case IWL_RX_PHY_INFO_TYPE_VHT_SU: - case IWL_RX_PHY_INFO_TYPE_VHT_MU: - case IWL_RX_PHY_INFO_TYPE_EHT_MU: - case IWL_RX_PHY_INFO_TYPE_EHT_TB: - case IWL_RX_PHY_INFO_TYPE_EHT_MU_EXT: - case IWL_RX_PHY_INFO_TYPE_EHT_TB_EXT: + u32 rate_n_flags = phy_data->rate_n_flags; + u32 he_type = rate_n_flags & RATE_MCS_HE_TYPE_MSK; + u32 nsts; + + switch (he_type) { + case RATE_MCS_HE_TYPE_TRIG: + iwl_mld_decode_he_tb_phy_data(phy_data, he, rx_status); + /* that's it, below is only for SU/MU */ return; - case IWL_RX_PHY_INFO_TYPE_HE_TB_EXT: - he->data1 |= cpu_to_le16(IEEE80211_RADIOTAP_HE_DATA1_SPTL_REUSE_KNOWN | - IEEE80211_RADIOTAP_HE_DATA1_SPTL_REUSE2_KNOWN | - IEEE80211_RADIOTAP_HE_DATA1_SPTL_REUSE3_KNOWN | - IEEE80211_RADIOTAP_HE_DATA1_SPTL_REUSE4_KNOWN); - he->data4 |= le16_encode_bits(le32_get_bits(phy_data->data2, - IWL_RX_PHY_DATA2_HE_TB_EXT_SPTL_REUSE1), - IEEE80211_RADIOTAP_HE_DATA4_TB_SPTL_REUSE1); - he->data4 |= le16_encode_bits(le32_get_bits(phy_data->data2, - IWL_RX_PHY_DATA2_HE_TB_EXT_SPTL_REUSE2), - IEEE80211_RADIOTAP_HE_DATA4_TB_SPTL_REUSE2); - he->data4 |= le16_encode_bits(le32_get_bits(phy_data->data2, - IWL_RX_PHY_DATA2_HE_TB_EXT_SPTL_REUSE3), - IEEE80211_RADIOTAP_HE_DATA4_TB_SPTL_REUSE3); - he->data4 |= le16_encode_bits(le32_get_bits(phy_data->data2, - IWL_RX_PHY_DATA2_HE_TB_EXT_SPTL_REUSE4), - IEEE80211_RADIOTAP_HE_DATA4_TB_SPTL_REUSE4); - fallthrough; - case IWL_RX_PHY_INFO_TYPE_HE_SU: - case IWL_RX_PHY_INFO_TYPE_HE_MU: - case IWL_RX_PHY_INFO_TYPE_HE_MU_EXT: - case IWL_RX_PHY_INFO_TYPE_HE_TB: - /* HE common */ - he->data1 |= cpu_to_le16(IEEE80211_RADIOTAP_HE_DATA1_LDPC_XSYMSEG_KNOWN | - IEEE80211_RADIOTAP_HE_DATA1_DOPPLER_KNOWN | - IEEE80211_RADIOTAP_HE_DATA1_BSS_COLOR_KNOWN); - he->data2 |= cpu_to_le16(IEEE80211_RADIOTAP_HE_DATA2_PRE_FEC_PAD_KNOWN | - IEEE80211_RADIOTAP_HE_DATA2_PE_DISAMBIG_KNOWN | - IEEE80211_RADIOTAP_HE_DATA2_TXOP_KNOWN | - IEEE80211_RADIOTAP_HE_DATA2_NUM_LTF_SYMS_KNOWN); - he->data3 |= le16_encode_bits(le32_get_bits(phy_data->data0, - IWL_RX_PHY_DATA0_HE_BSS_COLOR_MASK), - IEEE80211_RADIOTAP_HE_DATA3_BSS_COLOR); - if (phy_data->info_type != IWL_RX_PHY_INFO_TYPE_HE_TB && - phy_data->info_type != IWL_RX_PHY_INFO_TYPE_HE_TB_EXT) { - he->data1 |= cpu_to_le16(IEEE80211_RADIOTAP_HE_DATA1_UL_DL_KNOWN); - he->data3 |= le16_encode_bits(le32_get_bits(phy_data->data0, - IWL_RX_PHY_DATA0_HE_UPLINK), - IEEE80211_RADIOTAP_HE_DATA3_UL_DL); - } - he->data3 |= le16_encode_bits(le32_get_bits(phy_data->data0, - IWL_RX_PHY_DATA0_HE_LDPC_EXT_SYM), - IEEE80211_RADIOTAP_HE_DATA3_LDPC_XSYMSEG); - he->data5 |= le16_encode_bits(le32_get_bits(phy_data->data0, - IWL_RX_PHY_DATA0_HE_PRE_FEC_PAD_MASK), - IEEE80211_RADIOTAP_HE_DATA5_PRE_FEC_PAD); - he->data5 |= le16_encode_bits(le32_get_bits(phy_data->data0, - IWL_RX_PHY_DATA0_HE_PE_DISAMBIG), - IEEE80211_RADIOTAP_HE_DATA5_PE_DISAMBIG); - he->data5 |= le16_encode_bits(le32_get_bits(phy_data->data1, - IWL_RX_PHY_DATA1_HE_LTF_NUM_MASK), - IEEE80211_RADIOTAP_HE_DATA5_NUM_LTF_SYMS); - he->data6 |= le16_encode_bits(le32_get_bits(phy_data->data0, - IWL_RX_PHY_DATA0_HE_TXOP_DUR_MASK), - IEEE80211_RADIOTAP_HE_DATA6_TXOP); - he->data6 |= le16_encode_bits(le32_get_bits(phy_data->data0, - IWL_RX_PHY_DATA0_HE_DOPPLER), - IEEE80211_RADIOTAP_HE_DATA6_DOPPLER); - break; - } + case RATE_MCS_HE_TYPE_MU: + iwl_mld_decode_he_mu(phy_data, he, he_mu, rx_status); - switch (phy_data->info_type) { - case IWL_RX_PHY_INFO_TYPE_HE_MU_EXT: - case IWL_RX_PHY_INFO_TYPE_HE_MU: - case IWL_RX_PHY_INFO_TYPE_HE_SU: - he->data1 |= cpu_to_le16(IEEE80211_RADIOTAP_HE_DATA1_SPTL_REUSE_KNOWN); - he->data4 |= le16_encode_bits(le32_get_bits(phy_data->data0, - IWL_RX_PHY_DATA0_HE_SPATIAL_REUSE_MASK), - IEEE80211_RADIOTAP_HE_DATA4_SU_MU_SPTL_REUSE); + nsts = le32_get_bits(phy_data->ntfy->sigs.he.b, + OFDM_RX_FRAME_HE_SIGB_NSTS) + 1; break; - default: - /* nothing here */ - break; - } - - switch (phy_data->info_type) { - case IWL_RX_PHY_INFO_TYPE_HE_MU_EXT: - he_mu->flags1 |= - le16_encode_bits(le16_get_bits(phy_data->data4, - IWL_RX_PHY_DATA4_HE_MU_EXT_SIGB_DCM), - IEEE80211_RADIOTAP_HE_MU_FLAGS1_SIG_B_DCM); - he_mu->flags1 |= - le16_encode_bits(le16_get_bits(phy_data->data4, - IWL_RX_PHY_DATA4_HE_MU_EXT_SIGB_MCS_MASK), - IEEE80211_RADIOTAP_HE_MU_FLAGS1_SIG_B_MCS); - he_mu->flags2 |= - le16_encode_bits(le16_get_bits(phy_data->data4, - IWL_RX_PHY_DATA4_HE_MU_EXT_PREAMBLE_PUNC_TYPE_MASK), - IEEE80211_RADIOTAP_HE_MU_FLAGS2_PUNC_FROM_SIG_A_BW); - iwl_mld_decode_he_mu_ext(phy_data, he_mu); - fallthrough; - case IWL_RX_PHY_INFO_TYPE_HE_MU: - he_mu->flags2 |= - le16_encode_bits(le32_get_bits(phy_data->data1, - IWL_RX_PHY_DATA1_HE_MU_SIBG_SYM_OR_USER_NUM_MASK), - IEEE80211_RADIOTAP_HE_MU_FLAGS2_SIG_B_SYMS_USERS); - he_mu->flags2 |= - le16_encode_bits(le32_get_bits(phy_data->data1, - IWL_RX_PHY_DATA1_HE_MU_SIGB_COMPRESSION), - IEEE80211_RADIOTAP_HE_MU_FLAGS2_SIG_B_COMP); - fallthrough; - case IWL_RX_PHY_INFO_TYPE_HE_TB: - case IWL_RX_PHY_INFO_TYPE_HE_TB_EXT: - iwl_mld_decode_he_phy_ru_alloc(phy_data, he, he_mu, rx_status); - break; - case IWL_RX_PHY_INFO_TYPE_HE_SU: + case RATE_MCS_HE_TYPE_SU: + case RATE_MCS_HE_TYPE_EXT_SU: he->data1 |= cpu_to_le16(IEEE80211_RADIOTAP_HE_DATA1_BEAM_CHANGE_KNOWN); - he->data3 |= le16_encode_bits(le32_get_bits(phy_data->data0, - IWL_RX_PHY_DATA0_HE_BEAM_CHNG), - IEEE80211_RADIOTAP_HE_DATA3_BEAM_CHANGE); - break; - default: - /* nothing */ + he->data3 |= RTAP_ENC_HE(phy_data->ntfy->sigs.he.a1, + OFDM_RX_FRAME_HE_BEAM_CHANGE, + IEEE80211_RADIOTAP_HE_DATA3_BEAM_CHANGE); + + nsts = le32_get_bits(phy_data->ntfy->sigs.he.a1, + OFDM_RX_FRAME_HE_NSTS) + 1; break; } + + rx_status->nss = nsts >> !!(rate_n_flags & RATE_MCS_STBC_MSK); + + he->data1 |= cpu_to_le16(IEEE80211_RADIOTAP_HE_DATA1_LDPC_XSYMSEG_KNOWN | + IEEE80211_RADIOTAP_HE_DATA1_DOPPLER_KNOWN); + he->data2 |= cpu_to_le16(IEEE80211_RADIOTAP_HE_DATA2_PRE_FEC_PAD_KNOWN | + IEEE80211_RADIOTAP_HE_DATA2_PE_DISAMBIG_KNOWN | + IEEE80211_RADIOTAP_HE_DATA2_TXOP_KNOWN | + IEEE80211_RADIOTAP_HE_DATA2_NUM_LTF_SYMS_KNOWN); + + he->data3 |= RTAP_ENC_HE(phy_data->ntfy->sigs.he.a2, + OFDM_RX_FRAME_HE_CODING_EXTRA_SYM, + IEEE80211_RADIOTAP_HE_DATA3_LDPC_XSYMSEG); + he->data5 |= RTAP_ENC_HE(phy_data->ntfy->sigs.he.a2, + OFDM_RX_FRAME_HE_PRE_FEC_PAD_FACTOR, + IEEE80211_RADIOTAP_HE_DATA5_PRE_FEC_PAD); + he->data5 |= RTAP_ENC_HE(phy_data->ntfy->sigs.he.a2, + OFDM_RX_FRAME_HE_PE_DISAMBIG, + IEEE80211_RADIOTAP_HE_DATA5_PE_DISAMBIG); + he->data5 |= RTAP_ENC_HE(phy_data->ntfy->sigs.he.a2, + OFDM_RX_FRAME_HE_MU_NUM_OF_LTF_SYM, + IEEE80211_RADIOTAP_HE_DATA5_NUM_LTF_SYMS); + he->data6 |= RTAP_ENC_HE(phy_data->ntfy->sigs.he.a2, + OFDM_RX_FRAME_HE_TXOP_DURATION, + IEEE80211_RADIOTAP_HE_DATA6_TXOP); + he->data6 |= RTAP_ENC_HE(phy_data->ntfy->sigs.he.a2, + OFDM_RX_FRAME_HE_DOPPLER, + IEEE80211_RADIOTAP_HE_DATA6_DOPPLER); + + he->data1 |= cpu_to_le16(IEEE80211_RADIOTAP_HE_DATA1_UL_DL_KNOWN | + IEEE80211_RADIOTAP_HE_DATA1_BSS_COLOR_KNOWN | + IEEE80211_RADIOTAP_HE_DATA1_SPTL_REUSE_KNOWN); + + he->data3 |= RTAP_ENC_HE(phy_data->ntfy->sigs.he.a1, + OFDM_RX_FRAME_HE_BSS_COLOR, + IEEE80211_RADIOTAP_HE_DATA3_BSS_COLOR); + he->data3 |= RTAP_ENC_HE(phy_data->ntfy->sigs.he.a1, + OFDM_RX_FRAME_HE_UL_FLAG, + IEEE80211_RADIOTAP_HE_DATA3_UL_DL); + he->data4 |= RTAP_ENC_HE(phy_data->ntfy->sigs.he.a1, + OFDM_RX_FRAME_HE_SPATIAL_REUSE, + IEEE80211_RADIOTAP_HE_DATA4_SU_MU_SPTL_REUSE); } -static void iwl_mld_rx_he(struct iwl_mld *mld, struct sk_buff *skb, - struct iwl_mld_rx_phy_data *phy_data, - int queue) +static void iwl_mld_rx_he(struct sk_buff *skb, + struct iwl_mld_rx_phy_data *phy_data) { struct ieee80211_rx_status *rx_status = IEEE80211_SKB_RXCB(skb); struct ieee80211_radiotap_he *he = NULL; @@ -510,48 +529,28 @@ static void iwl_mld_rx_he(struct iwl_mld *mld, struct sk_buff *skb, .flags2 = cpu_to_le16(IEEE80211_RADIOTAP_HE_MU_FLAGS2_PUNC_FROM_SIG_A_BW_KNOWN | IEEE80211_RADIOTAP_HE_MU_FLAGS2_BW_FROM_SIG_A_BW_KNOWN), }; - u16 phy_info = phy_data->phy_info; he = skb_put_data(skb, &known, sizeof(known)); rx_status->flag |= RX_FLAG_RADIOTAP_HE; - if (phy_data->info_type == IWL_RX_PHY_INFO_TYPE_HE_MU || - phy_data->info_type == IWL_RX_PHY_INFO_TYPE_HE_MU_EXT) { - he_mu = skb_put_data(skb, &mu_known, sizeof(mu_known)); - rx_status->flag |= RX_FLAG_RADIOTAP_HE_MU; - } - - /* report the AMPDU-EOF bit on single frames */ - if (!queue && !(phy_info & IWL_RX_MPDU_PHY_AMPDU)) { - rx_status->flag |= RX_FLAG_AMPDU_DETAILS; - rx_status->flag |= RX_FLAG_AMPDU_EOF_BIT_KNOWN; - if (phy_data->data0 & cpu_to_le32(IWL_RX_PHY_DATA0_HE_DELIM_EOF)) - rx_status->flag |= RX_FLAG_AMPDU_EOF_BIT; - } - - if (phy_info & IWL_RX_MPDU_PHY_TSF_OVERLOAD) - iwl_mld_decode_he_phy_data(phy_data, he, he_mu, rx_status, - queue); - - /* update aggregation data for monitor sake on default queue */ - if (!queue && (phy_info & IWL_RX_MPDU_PHY_TSF_OVERLOAD) && - (phy_info & IWL_RX_MPDU_PHY_AMPDU) && phy_data->first_subframe) { - rx_status->flag |= RX_FLAG_AMPDU_EOF_BIT_KNOWN; - if (phy_data->data0 & cpu_to_le32(IWL_RX_PHY_DATA0_EHT_DELIM_EOF)) - rx_status->flag |= RX_FLAG_AMPDU_EOF_BIT; - } - - if (he_type == RATE_MCS_HE_TYPE_EXT_SU && - rate_n_flags & RATE_MCS_HE_106T_MSK) { - rx_status->bw = RATE_INFO_BW_HE_RU; - rx_status->he_ru = NL80211_RATE_INFO_HE_RU_ALLOC_106; - } - - /* actually data is filled in mac80211 */ - if (he_type == RATE_MCS_HE_TYPE_SU || - he_type == RATE_MCS_HE_TYPE_EXT_SU) + switch (he_type) { + case RATE_MCS_HE_TYPE_EXT_SU: + /* + * Except for this special case we won't have + * HE RU allocation info outside of monitor mode + * since we don't get the PHY notif. + */ + if (rate_n_flags & RATE_MCS_HE_106T_MSK) { + rx_status->bw = RATE_INFO_BW_HE_RU; + rx_status->he_ru = NL80211_RATE_INFO_HE_RU_ALLOC_106; + } + fallthrough; + case RATE_MCS_HE_TYPE_SU: + /* actual data is filled in mac80211 */ he->data1 |= cpu_to_le16(IEEE80211_RADIOTAP_HE_DATA1_BW_RU_ALLOC_KNOWN); + break; + } #define CHECK_TYPE(F) \ BUILD_BUG_ON(IEEE80211_RADIOTAP_HE_DATA1_FORMAT_ ## F != \ @@ -567,8 +566,7 @@ static void iwl_mld_rx_he(struct iwl_mld *mld, struct sk_buff *skb, if (rate_n_flags & RATE_MCS_BF_MSK) he->data5 |= cpu_to_le16(IEEE80211_RADIOTAP_HE_DATA5_TXBF); - switch ((rate_n_flags & RATE_MCS_HE_GI_LTF_MSK) >> - RATE_MCS_HE_GI_LTF_POS) { + switch (u32_get_bits(rate_n_flags, RATE_MCS_HE_GI_LTF_MSK)) { case 0: if (he_type == RATE_MCS_HE_TYPE_TRIG) rx_status->he_gi = NL80211_RATE_INFO_HE_GI_1_6; @@ -609,37 +607,52 @@ static void iwl_mld_rx_he(struct iwl_mld *mld, struct sk_buff *skb, he->data5 |= le16_encode_bits(ltf, IEEE80211_RADIOTAP_HE_DATA5_LTF_SIZE); + + if (likely(!phy_data->ntfy)) + return; + + if (he_type == RATE_MCS_HE_TYPE_MU) { + he_mu = skb_put_data(skb, &mu_known, sizeof(mu_known)); + rx_status->flag |= RX_FLAG_RADIOTAP_HE_MU; + } + + iwl_mld_decode_he_phy_data(phy_data, he, he_mu, rx_status); } static void iwl_mld_decode_lsig(struct sk_buff *skb, struct iwl_mld_rx_phy_data *phy_data) { struct ieee80211_rx_status *rx_status = IEEE80211_SKB_RXCB(skb); + u32 format = phy_data->rate_n_flags & RATE_MCS_MOD_TYPE_MSK; struct ieee80211_radiotap_lsig *lsig; + u32 lsig_len, rate; - switch (phy_data->info_type) { - case IWL_RX_PHY_INFO_TYPE_HT: - case IWL_RX_PHY_INFO_TYPE_VHT_SU: - case IWL_RX_PHY_INFO_TYPE_VHT_MU: - case IWL_RX_PHY_INFO_TYPE_HE_TB_EXT: - case IWL_RX_PHY_INFO_TYPE_HE_SU: - case IWL_RX_PHY_INFO_TYPE_HE_MU: - case IWL_RX_PHY_INFO_TYPE_HE_MU_EXT: - case IWL_RX_PHY_INFO_TYPE_HE_TB: - case IWL_RX_PHY_INFO_TYPE_EHT_MU: - case IWL_RX_PHY_INFO_TYPE_EHT_TB: - case IWL_RX_PHY_INFO_TYPE_EHT_MU_EXT: - case IWL_RX_PHY_INFO_TYPE_EHT_TB_EXT: - lsig = skb_put(skb, sizeof(*lsig)); - lsig->data1 = cpu_to_le16(IEEE80211_RADIOTAP_LSIG_DATA1_LENGTH_KNOWN); - lsig->data2 = le16_encode_bits(le32_get_bits(phy_data->data1, - IWL_RX_PHY_DATA1_LSIG_LEN_MASK), - IEEE80211_RADIOTAP_LSIG_DATA2_LENGTH); - rx_status->flag |= RX_FLAG_RADIOTAP_LSIG; - break; - default: - break; - } + if (likely(!phy_data->ntfy)) + return; + + /* + * Technically legacy CCK/OFDM frames don't have an L-SIG + * since that's the compat format for HT (non-greenfield) + * and up. However, it's meant to be compatible with the + * LENGTH and RATE fields in Clause 17 and 18 OFDM frames + * so include the field for any non-CCK frame. For CCK it + * cannot work, since the LENGTH field for them is 16-bit + * and the radiotap field only has 12 bits. + */ + if (format == RATE_MCS_MOD_TYPE_CCK) + return; + + lsig_len = le32_get_bits(phy_data->ntfy->legacy_sig.ofdm, + OFDM_RX_LEGACY_LENGTH); + rate = le32_get_bits(phy_data->ntfy->legacy_sig.ofdm, OFDM_RX_RATE); + + lsig = skb_put(skb, sizeof(*lsig)); + lsig->data1 = cpu_to_le16(IEEE80211_RADIOTAP_LSIG_DATA1_LENGTH_KNOWN) | + cpu_to_le16(IEEE80211_RADIOTAP_LSIG_DATA1_RATE_KNOWN); + lsig->data2 = le16_encode_bits(lsig_len, + IEEE80211_RADIOTAP_LSIG_DATA2_LENGTH) | + le16_encode_bits(rate, IEEE80211_RADIOTAP_LSIG_DATA2_RATE); + rx_status->flag |= RX_FLAG_RADIOTAP_LSIG; } /* Put a TLV on the skb and return data pointer @@ -667,209 +680,144 @@ iwl_mld_radiotap_put_tlv(struct sk_buff *skb, u16 type, u16 len) (_usig)->value |= LE32_DEC_ENC(in_value, dec_bits, _enc_bits); \ } while (0) -#define __IWL_MLD_ENC_EHT_RU(rt_data, rt_ru, fw_data, fw_ru) \ - eht->data[(rt_data)] |= \ - (cpu_to_le32 \ - (IEEE80211_RADIOTAP_EHT_DATA ## rt_data ## _RU_ALLOC_CC_ ## rt_ru ## _KNOWN) | \ - LE32_DEC_ENC(data ## fw_data, \ - IWL_RX_PHY_DATA ## fw_data ## _EHT_MU_EXT_RU_ALLOC_ ## fw_ru, \ - IEEE80211_RADIOTAP_EHT_DATA ## rt_data ## _RU_ALLOC_CC_ ## rt_ru)) - -#define _IWL_MLD_ENC_EHT_RU(rt_data, rt_ru, fw_data, fw_ru) \ - __IWL_MLD_ENC_EHT_RU(rt_data, rt_ru, fw_data, fw_ru) - -#define IEEE80211_RADIOTAP_RU_DATA_1_1_1 1 -#define IEEE80211_RADIOTAP_RU_DATA_2_1_1 2 -#define IEEE80211_RADIOTAP_RU_DATA_1_1_2 2 -#define IEEE80211_RADIOTAP_RU_DATA_2_1_2 2 -#define IEEE80211_RADIOTAP_RU_DATA_1_2_1 3 -#define IEEE80211_RADIOTAP_RU_DATA_2_2_1 3 -#define IEEE80211_RADIOTAP_RU_DATA_1_2_2 3 -#define IEEE80211_RADIOTAP_RU_DATA_2_2_2 4 - -#define IWL_RX_RU_DATA_A1 2 -#define IWL_RX_RU_DATA_A2 2 -#define IWL_RX_RU_DATA_B1 2 -#define IWL_RX_RU_DATA_B2 4 -#define IWL_RX_RU_DATA_C1 3 -#define IWL_RX_RU_DATA_C2 3 -#define IWL_RX_RU_DATA_D1 4 -#define IWL_RX_RU_DATA_D2 4 - -#define IWL_MLD_ENC_EHT_RU(rt_ru, fw_ru) \ - _IWL_MLD_ENC_EHT_RU(IEEE80211_RADIOTAP_RU_DATA_ ## rt_ru, \ - rt_ru, \ - IWL_RX_RU_DATA_ ## fw_ru, \ - fw_ru) - -static void iwl_mld_decode_eht_ext_mu(struct iwl_mld *mld, - struct iwl_mld_rx_phy_data *phy_data, - struct ieee80211_rx_status *rx_status, - struct ieee80211_radiotap_eht *eht, - struct ieee80211_radiotap_eht_usig *usig) +static void iwl_mld_decode_eht_usig_tb(struct iwl_mld_rx_phy_data *phy_data, + struct ieee80211_radiotap_eht_usig *usig) { - if (phy_data->with_data) { - __le32 data1 = phy_data->data1; - __le32 data2 = phy_data->data2; - __le32 data3 = phy_data->data3; - __le32 data4 = phy_data->eht_data4; - __le32 data5 = phy_data->data5; - u32 phy_bw = phy_data->rate_n_flags & RATE_MCS_CHAN_WIDTH_MSK; + __le32 usig_a1 = phy_data->ntfy->sigs.eht_tb.usig_a1; + __le32 usig_a2 = phy_data->ntfy->sigs.eht_tb.usig_a2_eht; - IWL_MLD_ENC_USIG_VALUE_MASK(usig, data5, - IWL_RX_PHY_DATA5_EHT_TYPE_AND_COMP, - IEEE80211_RADIOTAP_EHT_USIG2_MU_B0_B1_PPDU_TYPE); - IWL_MLD_ENC_USIG_VALUE_MASK(usig, data5, - IWL_RX_PHY_DATA5_EHT_MU_PUNC_CH_CODE, - IEEE80211_RADIOTAP_EHT_USIG2_MU_B3_B7_PUNCTURED_INFO); - IWL_MLD_ENC_USIG_VALUE_MASK(usig, data4, - IWL_RX_PHY_DATA4_EHT_MU_EXT_SIGB_MCS, - IEEE80211_RADIOTAP_EHT_USIG2_MU_B9_B10_SIG_MCS); - IWL_MLD_ENC_USIG_VALUE_MASK - (usig, data1, IWL_RX_PHY_DATA1_EHT_MU_NUM_SIG_SYM_USIGA2, - IEEE80211_RADIOTAP_EHT_USIG2_MU_B11_B15_EHT_SIG_SYMBOLS); - - eht->user_info[0] |= - cpu_to_le32(IEEE80211_RADIOTAP_EHT_USER_INFO_STA_ID_KNOWN) | - LE32_DEC_ENC(data5, IWL_RX_PHY_DATA5_EHT_MU_STA_ID_USR, - IEEE80211_RADIOTAP_EHT_USER_INFO_STA_ID); - - eht->known |= cpu_to_le32(IEEE80211_RADIOTAP_EHT_KNOWN_NR_NON_OFDMA_USERS_M); - eht->data[7] |= LE32_DEC_ENC - (data5, IWL_RX_PHY_DATA5_EHT_MU_NUM_USR_NON_OFDMA, - IEEE80211_RADIOTAP_EHT_DATA7_NUM_OF_NON_OFDMA_USERS); - - /* - * Hardware labels the content channels/RU allocation values - * as follows: - * Content Channel 1 Content Channel 2 - * 20 MHz: A1 - * 40 MHz: A1 B1 - * 80 MHz: A1 C1 B1 D1 - * 160 MHz: A1 C1 A2 C2 B1 D1 B2 D2 - * 320 MHz: A1 C1 A2 C2 A3 C3 A4 C4 B1 D1 B2 D2 B3 D3 B4 D4 - * - * However firmware can only give us A1-D2, so the higher - * frequencies are missing. - */ - - switch (phy_bw) { - case RATE_MCS_CHAN_WIDTH_320: - /* additional values are missing in RX metadata */ - fallthrough; - case RATE_MCS_CHAN_WIDTH_160: - /* content channel 1 */ - IWL_MLD_ENC_EHT_RU(1_2_1, A2); - IWL_MLD_ENC_EHT_RU(1_2_2, C2); - /* content channel 2 */ - IWL_MLD_ENC_EHT_RU(2_2_1, B2); - IWL_MLD_ENC_EHT_RU(2_2_2, D2); - fallthrough; - case RATE_MCS_CHAN_WIDTH_80: - /* content channel 1 */ - IWL_MLD_ENC_EHT_RU(1_1_2, C1); - /* content channel 2 */ - IWL_MLD_ENC_EHT_RU(2_1_2, D1); - fallthrough; - case RATE_MCS_CHAN_WIDTH_40: - /* content channel 2 */ - IWL_MLD_ENC_EHT_RU(2_1_1, B1); - fallthrough; - case RATE_MCS_CHAN_WIDTH_20: - IWL_MLD_ENC_EHT_RU(1_1_1, A1); - break; - } - } else { - __le32 usig_a1 = phy_data->rx_vec[0]; - __le32 usig_a2 = phy_data->rx_vec[1]; - - IWL_MLD_ENC_USIG_VALUE_MASK(usig, usig_a1, - IWL_RX_USIG_A1_DISREGARD, - IEEE80211_RADIOTAP_EHT_USIG1_MU_B20_B24_DISREGARD); - IWL_MLD_ENC_USIG_VALUE_MASK(usig, usig_a1, - IWL_RX_USIG_A1_VALIDATE, - IEEE80211_RADIOTAP_EHT_USIG1_MU_B25_VALIDATE); - IWL_MLD_ENC_USIG_VALUE_MASK(usig, usig_a2, - IWL_RX_USIG_A2_EHT_PPDU_TYPE, - IEEE80211_RADIOTAP_EHT_USIG2_MU_B0_B1_PPDU_TYPE); - IWL_MLD_ENC_USIG_VALUE_MASK(usig, usig_a2, - IWL_RX_USIG_A2_EHT_USIG2_VALIDATE_B2, - IEEE80211_RADIOTAP_EHT_USIG2_MU_B2_VALIDATE); - IWL_MLD_ENC_USIG_VALUE_MASK(usig, usig_a2, - IWL_RX_USIG_A2_EHT_PUNC_CHANNEL, - IEEE80211_RADIOTAP_EHT_USIG2_MU_B3_B7_PUNCTURED_INFO); - IWL_MLD_ENC_USIG_VALUE_MASK(usig, usig_a2, - IWL_RX_USIG_A2_EHT_USIG2_VALIDATE_B8, - IEEE80211_RADIOTAP_EHT_USIG2_MU_B8_VALIDATE); - IWL_MLD_ENC_USIG_VALUE_MASK(usig, usig_a2, - IWL_RX_USIG_A2_EHT_SIG_MCS, - IEEE80211_RADIOTAP_EHT_USIG2_MU_B9_B10_SIG_MCS); - IWL_MLD_ENC_USIG_VALUE_MASK - (usig, usig_a2, IWL_RX_USIG_A2_EHT_SIG_SYM_NUM, - IEEE80211_RADIOTAP_EHT_USIG2_MU_B11_B15_EHT_SIG_SYMBOLS); - IWL_MLD_ENC_USIG_VALUE_MASK(usig, usig_a2, - IWL_RX_USIG_A2_EHT_CRC_OK, - IEEE80211_RADIOTAP_EHT_USIG2_MU_B16_B19_CRC); - } + IWL_MLD_ENC_USIG_VALUE_MASK(usig, usig_a1, + OFDM_RX_FRAME_EHT_USIG1_DISREGARD, + IEEE80211_RADIOTAP_EHT_USIG1_TB_B20_B25_DISREGARD); + IWL_MLD_ENC_USIG_VALUE_MASK(usig, usig_a2, + OFDM_RX_FRAME_EHT_PPDU_TYPE, + IEEE80211_RADIOTAP_EHT_USIG2_TB_B0_B1_PPDU_TYPE); + IWL_MLD_ENC_USIG_VALUE_MASK(usig, usig_a2, + OFDM_RX_FRAME_EHT_USIG2_VALIDATE_B2, + IEEE80211_RADIOTAP_EHT_USIG2_TB_B2_VALIDATE); + IWL_MLD_ENC_USIG_VALUE_MASK(usig, usig_a2, + OFDM_RX_FRAME_EHT_TRIG_SPATIAL_REUSE_1, + IEEE80211_RADIOTAP_EHT_USIG2_TB_B3_B6_SPATIAL_REUSE_1); + IWL_MLD_ENC_USIG_VALUE_MASK(usig, usig_a2, + OFDM_RX_FRAME_EHT_TRIG_SPATIAL_REUSE_2, + IEEE80211_RADIOTAP_EHT_USIG2_TB_B7_B10_SPATIAL_REUSE_2); + IWL_MLD_ENC_USIG_VALUE_MASK(usig, usig_a2, + OFDM_RX_FRAME_EHT_TRIG_USIG2_DISREGARD, + IEEE80211_RADIOTAP_EHT_USIG2_TB_B11_B15_DISREGARD); } -static void iwl_mld_decode_eht_ext_tb(struct iwl_mld *mld, - struct iwl_mld_rx_phy_data *phy_data, - struct ieee80211_rx_status *rx_status, - struct ieee80211_radiotap_eht *eht, - struct ieee80211_radiotap_eht_usig *usig) +static void iwl_mld_decode_eht_usig_non_tb(struct iwl_mld_rx_phy_data *phy_data, + struct ieee80211_radiotap_eht_usig *usig) { - if (phy_data->with_data) { - __le32 data5 = phy_data->data5; + __le32 usig_a1 = phy_data->ntfy->sigs.eht.usig_a1; + __le32 usig_a2 = phy_data->ntfy->sigs.eht.usig_a2_eht; - IWL_MLD_ENC_USIG_VALUE_MASK(usig, data5, - IWL_RX_PHY_DATA5_EHT_TYPE_AND_COMP, - IEEE80211_RADIOTAP_EHT_USIG2_TB_B0_B1_PPDU_TYPE); - IWL_MLD_ENC_USIG_VALUE_MASK(usig, data5, - IWL_RX_PHY_DATA5_EHT_TB_SPATIAL_REUSE1, - IEEE80211_RADIOTAP_EHT_USIG2_TB_B3_B6_SPATIAL_REUSE_1); - - IWL_MLD_ENC_USIG_VALUE_MASK(usig, data5, - IWL_RX_PHY_DATA5_EHT_TB_SPATIAL_REUSE2, - IEEE80211_RADIOTAP_EHT_USIG2_TB_B7_B10_SPATIAL_REUSE_2); - } else { - __le32 usig_a1 = phy_data->rx_vec[0]; - __le32 usig_a2 = phy_data->rx_vec[1]; - - IWL_MLD_ENC_USIG_VALUE_MASK(usig, usig_a1, - IWL_RX_USIG_A1_DISREGARD, - IEEE80211_RADIOTAP_EHT_USIG1_TB_B20_B25_DISREGARD); - IWL_MLD_ENC_USIG_VALUE_MASK(usig, usig_a2, - IWL_RX_USIG_A2_EHT_PPDU_TYPE, - IEEE80211_RADIOTAP_EHT_USIG2_TB_B0_B1_PPDU_TYPE); - IWL_MLD_ENC_USIG_VALUE_MASK(usig, usig_a2, - IWL_RX_USIG_A2_EHT_USIG2_VALIDATE_B2, - IEEE80211_RADIOTAP_EHT_USIG2_TB_B2_VALIDATE); - IWL_MLD_ENC_USIG_VALUE_MASK(usig, usig_a2, - IWL_RX_USIG_A2_EHT_TRIG_SPATIAL_REUSE_1, - IEEE80211_RADIOTAP_EHT_USIG2_TB_B3_B6_SPATIAL_REUSE_1); - IWL_MLD_ENC_USIG_VALUE_MASK(usig, usig_a2, - IWL_RX_USIG_A2_EHT_TRIG_SPATIAL_REUSE_2, - IEEE80211_RADIOTAP_EHT_USIG2_TB_B7_B10_SPATIAL_REUSE_2); - IWL_MLD_ENC_USIG_VALUE_MASK(usig, usig_a2, - IWL_RX_USIG_A2_EHT_TRIG_USIG2_DISREGARD, - IEEE80211_RADIOTAP_EHT_USIG2_TB_B11_B15_DISREGARD); - IWL_MLD_ENC_USIG_VALUE_MASK(usig, usig_a2, - IWL_RX_USIG_A2_EHT_CRC_OK, - IEEE80211_RADIOTAP_EHT_USIG2_TB_B16_B19_CRC); - } + IWL_MLD_ENC_USIG_VALUE_MASK(usig, usig_a1, + OFDM_RX_FRAME_EHT_USIG1_DISREGARD, + IEEE80211_RADIOTAP_EHT_USIG1_MU_B20_B24_DISREGARD); + IWL_MLD_ENC_USIG_VALUE_MASK(usig, usig_a1, + OFDM_RX_FRAME_EHT_USIG1_VALIDATE, + IEEE80211_RADIOTAP_EHT_USIG1_MU_B25_VALIDATE); + IWL_MLD_ENC_USIG_VALUE_MASK(usig, usig_a2, + OFDM_RX_FRAME_EHT_PPDU_TYPE, + IEEE80211_RADIOTAP_EHT_USIG2_MU_B0_B1_PPDU_TYPE); + IWL_MLD_ENC_USIG_VALUE_MASK(usig, usig_a2, + OFDM_RX_FRAME_EHT_USIG2_VALIDATE_B2, + IEEE80211_RADIOTAP_EHT_USIG2_MU_B2_VALIDATE); + IWL_MLD_ENC_USIG_VALUE_MASK(usig, usig_a2, + OFDM_RX_FRAME_EHT_PUNC_CHANNEL, + IEEE80211_RADIOTAP_EHT_USIG2_MU_B3_B7_PUNCTURED_INFO); + IWL_MLD_ENC_USIG_VALUE_MASK(usig, usig_a2, + OFDM_RX_FRAME_EHT_USIG2_VALIDATE_B8, + IEEE80211_RADIOTAP_EHT_USIG2_MU_B8_VALIDATE); + IWL_MLD_ENC_USIG_VALUE_MASK(usig, usig_a2, + OFDM_RX_FRAME_EHT_SIG_MCS, + IEEE80211_RADIOTAP_EHT_USIG2_MU_B9_B10_SIG_MCS); + IWL_MLD_ENC_USIG_VALUE_MASK(usig, usig_a2, + OFDM_RX_FRAME_EHT_SIG_SYM_NUM, + IEEE80211_RADIOTAP_EHT_USIG2_MU_B11_B15_EHT_SIG_SYMBOLS); } -static void iwl_mld_decode_eht_ru(struct iwl_mld *mld, - struct ieee80211_rx_status *rx_status, - struct ieee80211_radiotap_eht *eht) +static void iwl_mld_decode_eht_usig(struct iwl_mld_rx_phy_data *phy_data, + struct sk_buff *skb) +{ + u32 he_type = phy_data->rate_n_flags & RATE_MCS_HE_TYPE_MSK; + __le32 usig_a1 = phy_data->ntfy->sigs.eht.usig_a1; + __le32 usig_a2 = phy_data->ntfy->sigs.eht.usig_a2_eht; + struct ieee80211_radiotap_eht_usig *usig; + u32 bw; + + usig = iwl_mld_radiotap_put_tlv(skb, IEEE80211_RADIOTAP_EHT_USIG, + sizeof(*usig)); + + BUILD_BUG_ON(offsetof(union iwl_sigs, eht.usig_a1) != + offsetof(union iwl_sigs, eht_tb.usig_a1)); + BUILD_BUG_ON(offsetof(union iwl_sigs, eht.usig_a2_eht) != + offsetof(union iwl_sigs, eht_tb.usig_a2_eht)); + + usig->common |= cpu_to_le32(IEEE80211_RADIOTAP_EHT_USIG_COMMON_UL_DL_KNOWN | + IEEE80211_RADIOTAP_EHT_USIG_COMMON_BSS_COLOR_KNOWN | + IEEE80211_RADIOTAP_EHT_USIG_COMMON_VALIDATE_BITS_CHECKED | + IEEE80211_RADIOTAP_EHT_USIG_COMMON_BW_KNOWN | + IEEE80211_RADIOTAP_EHT_USIG_COMMON_TXOP_KNOWN); + +#define CHECK_BW(bw) \ + BUILD_BUG_ON(IEEE80211_RADIOTAP_EHT_USIG_COMMON_BW_ ## bw ## MHZ != \ + RATE_MCS_CHAN_WIDTH_ ## bw ## _VAL) + CHECK_BW(20); + CHECK_BW(40); + CHECK_BW(80); + CHECK_BW(160); +#undef CHECK_BW + BUILD_BUG_ON(IEEE80211_RADIOTAP_EHT_USIG_COMMON_BW_320MHZ_1 != + RATE_MCS_CHAN_WIDTH_320_VAL); + bw = u32_get_bits(phy_data->rate_n_flags, RATE_MCS_CHAN_WIDTH_MSK); + /* specific handling for 320MHz-1/320MHz-2 */ + if (bw == RATE_MCS_CHAN_WIDTH_320_VAL) + bw += le32_get_bits(usig_a1, OFDM_RX_FRAME_EHT_BW320_SLOT); + usig->common |= le32_encode_bits(bw, + IEEE80211_RADIOTAP_EHT_USIG_COMMON_BW); + + usig->common |= LE32_DEC_ENC(usig_a1, OFDM_RX_FRAME_ENHANCED_WIFI_UL_FLAG, + IEEE80211_RADIOTAP_EHT_USIG_COMMON_UL_DL); + usig->common |= LE32_DEC_ENC(usig_a1, OFDM_RX_FRAME_ENHANCED_WIFI_BSS_COLOR, + IEEE80211_RADIOTAP_EHT_USIG_COMMON_BSS_COLOR); + + if (le32_get_bits(usig_a1, OFDM_RX_FRAME_EHT_USIG1_VALIDATE) && + le32_get_bits(usig_a2, OFDM_RX_FRAME_EHT_USIG2_VALIDATE_B2) && + le32_get_bits(usig_a2, OFDM_RX_FRAME_EHT_USIG2_VALIDATE_B8)) + usig->common |= cpu_to_le32(IEEE80211_RADIOTAP_EHT_USIG_COMMON_VALIDATE_BITS_OK); + + usig->common |= LE32_DEC_ENC(usig_a1, + OFDM_RX_FRAME_ENHANCED_WIFI_TXOP_DURATION, + IEEE80211_RADIOTAP_EHT_USIG_COMMON_TXOP); + + if (!le32_get_bits(usig_a2, OFDM_RX_USIG_CRC_OK)) + usig->common |= cpu_to_le32(IEEE80211_RADIOTAP_EHT_USIG_COMMON_BAD_USIG_CRC); + + usig->common |= cpu_to_le32(IEEE80211_RADIOTAP_EHT_USIG_COMMON_PHY_VER_KNOWN); + usig->common |= LE32_DEC_ENC(usig_a1, + OFDM_RX_FRAME_ENHANCED_WIFI_VER_ID, + IEEE80211_RADIOTAP_EHT_USIG_COMMON_PHY_VER); + + if (he_type == RATE_MCS_HE_TYPE_TRIG) + iwl_mld_decode_eht_usig_tb(phy_data, usig); + else + iwl_mld_decode_eht_usig_non_tb(phy_data, usig); +} + +static void +iwl_mld_eht_set_ru_alloc(struct ieee80211_rx_status *rx_status, + u32 ru_with_p80) { - u32 ru = le32_get_bits(eht->data[8], - IEEE80211_RADIOTAP_EHT_DATA8_RU_ALLOC_TB_FMT_B7_B1); enum nl80211_eht_ru_alloc nl_ru; + u32 ru = ru_with_p80 >> 1; - /* Using D1.5 Table 9-53a - Encoding of PS160 and RU Allocation subfields - * in an EHT variant User Info field + /* + * HW always uses trigger frame format: + * + * Draft PIEEE802.11be D7.0 Table 9-46l - Encoding of the PS160 and + * RU Allocation subfields in an EHT variant User Info field */ switch (ru) { @@ -929,135 +877,228 @@ static void iwl_mld_decode_eht_ru(struct iwl_mld *mld, rx_status->eht.ru = nl_ru; } -static void iwl_mld_decode_eht_phy_data(struct iwl_mld *mld, - struct iwl_mld_rx_phy_data *phy_data, - struct ieee80211_rx_status *rx_status, - struct ieee80211_radiotap_eht *eht, - struct ieee80211_radiotap_eht_usig *usig) - +static void iwl_mld_decode_eht_tb(struct iwl_mld_rx_phy_data *phy_data, + struct ieee80211_rx_status *rx_status, + struct ieee80211_radiotap_eht *eht) { - __le32 data0 = phy_data->data0; - __le32 data1 = phy_data->data1; - __le32 usig_a1 = phy_data->rx_vec[0]; - u8 info_type = phy_data->info_type; - - /* Not in EHT range */ - if (info_type < IWL_RX_PHY_INFO_TYPE_EHT_MU || - info_type > IWL_RX_PHY_INFO_TYPE_EHT_TB_EXT) + if (!(phy_data->ntfy->flags & IWL_SNIF_FLAG_VALID_TB_RX)) return; - usig->common |= cpu_to_le32 - (IEEE80211_RADIOTAP_EHT_USIG_COMMON_UL_DL_KNOWN | - IEEE80211_RADIOTAP_EHT_USIG_COMMON_BSS_COLOR_KNOWN); - if (phy_data->with_data) { - usig->common |= LE32_DEC_ENC(data0, - IWL_RX_PHY_DATA0_EHT_UPLINK, - IEEE80211_RADIOTAP_EHT_USIG_COMMON_UL_DL); - usig->common |= LE32_DEC_ENC(data0, - IWL_RX_PHY_DATA0_EHT_BSS_COLOR_MASK, - IEEE80211_RADIOTAP_EHT_USIG_COMMON_BSS_COLOR); - } else { - usig->common |= LE32_DEC_ENC(usig_a1, - IWL_RX_USIG_A1_UL_FLAG, - IEEE80211_RADIOTAP_EHT_USIG_COMMON_UL_DL); - usig->common |= LE32_DEC_ENC(usig_a1, - IWL_RX_USIG_A1_BSS_COLOR, - IEEE80211_RADIOTAP_EHT_USIG_COMMON_BSS_COLOR); - } + eht->known |= cpu_to_le32(IEEE80211_RADIOTAP_EHT_KNOWN_RU_ALLOC_TB_FMT | + IEEE80211_RADIOTAP_EHT_KNOWN_LDPC_EXTRA_SYM_OM | + IEEE80211_RADIOTAP_EHT_KNOWN_PRE_PADD_FACOR_OM | + IEEE80211_RADIOTAP_EHT_KNOWN_PE_DISAMBIGUITY_OM | + IEEE80211_RADIOTAP_EHT_KNOWN_EHT_LTF | + IEEE80211_RADIOTAP_EHT_KNOWN_PRIMARY_80); - usig->common |= - cpu_to_le32(IEEE80211_RADIOTAP_EHT_USIG_COMMON_VALIDATE_BITS_CHECKED); - usig->common |= - LE32_DEC_ENC(data0, IWL_RX_PHY_DATA0_EHT_VALIDATE, - IEEE80211_RADIOTAP_EHT_USIG_COMMON_VALIDATE_BITS_OK); - - eht->known |= cpu_to_le32(IEEE80211_RADIOTAP_EHT_KNOWN_SPATIAL_REUSE); - eht->data[0] |= LE32_DEC_ENC(data0, - IWL_RX_PHY_DATA0_ETH_SPATIAL_REUSE_MASK, - IEEE80211_RADIOTAP_EHT_DATA0_SPATIAL_REUSE); - - /* All RU allocating size/index is in TB format */ - eht->known |= cpu_to_le32(IEEE80211_RADIOTAP_EHT_KNOWN_RU_ALLOC_TB_FMT); - eht->data[8] |= LE32_DEC_ENC(data0, IWL_RX_PHY_DATA0_EHT_PS160, + eht->data[8] |= LE32_DEC_ENC(phy_data->ntfy->sigs.eht_tb.tb_rx0, + OFDM_UCODE_TRIG_BASE_PS160, IEEE80211_RADIOTAP_EHT_DATA8_RU_ALLOC_TB_FMT_PS_160); - eht->data[8] |= LE32_DEC_ENC(data1, IWL_RX_PHY_DATA1_EHT_RU_ALLOC_B0, - IEEE80211_RADIOTAP_EHT_DATA8_RU_ALLOC_TB_FMT_B0); - eht->data[8] |= LE32_DEC_ENC(data1, IWL_RX_PHY_DATA1_EHT_RU_ALLOC_B1_B7, + eht->data[8] |= LE32_DEC_ENC(phy_data->ntfy->sigs.eht_tb.tb_rx1, + OFDM_UCODE_TRIG_BASE_RX_RU, + IEEE80211_RADIOTAP_EHT_DATA8_RU_ALLOC_TB_FMT_B0 | IEEE80211_RADIOTAP_EHT_DATA8_RU_ALLOC_TB_FMT_B7_B1); - - iwl_mld_decode_eht_ru(mld, rx_status, eht); - - /* We only get here in case of IWL_RX_MPDU_PHY_TSF_OVERLOAD is set - * which is on only in case of monitor mode so no need to check monitor - * mode - */ - eht->known |= cpu_to_le32(IEEE80211_RADIOTAP_EHT_KNOWN_PRIMARY_80); - eht->data[1] |= - le32_encode_bits(mld->monitor.p80, - IEEE80211_RADIOTAP_EHT_DATA1_PRIMARY_80); - - usig->common |= cpu_to_le32(IEEE80211_RADIOTAP_EHT_USIG_COMMON_TXOP_KNOWN); - if (phy_data->with_data) - usig->common |= LE32_DEC_ENC(data0, IWL_RX_PHY_DATA0_EHT_TXOP_DUR_MASK, - IEEE80211_RADIOTAP_EHT_USIG_COMMON_TXOP); - else - usig->common |= LE32_DEC_ENC(usig_a1, IWL_RX_USIG_A1_TXOP_DURATION, - IEEE80211_RADIOTAP_EHT_USIG_COMMON_TXOP); - - eht->known |= cpu_to_le32(IEEE80211_RADIOTAP_EHT_KNOWN_LDPC_EXTRA_SYM_OM); - eht->data[0] |= LE32_DEC_ENC(data0, IWL_RX_PHY_DATA0_EHT_LDPC_EXT_SYM, + eht->data[0] |= LE32_DEC_ENC(phy_data->ntfy->sigs.eht_tb.tb_rx1, + OFDM_UCODE_TRIG_BASE_RX_CODING_EXTRA_SYM, IEEE80211_RADIOTAP_EHT_DATA0_LDPC_EXTRA_SYM_OM); - - eht->known |= cpu_to_le32(IEEE80211_RADIOTAP_EHT_KNOWN_PRE_PADD_FACOR_OM); - eht->data[0] |= LE32_DEC_ENC(data0, IWL_RX_PHY_DATA0_EHT_PRE_FEC_PAD_MASK, - IEEE80211_RADIOTAP_EHT_DATA0_PRE_PADD_FACOR_OM); - - eht->known |= cpu_to_le32(IEEE80211_RADIOTAP_EHT_KNOWN_PE_DISAMBIGUITY_OM); - eht->data[0] |= LE32_DEC_ENC(data0, IWL_RX_PHY_DATA0_EHT_PE_DISAMBIG, + eht->data[0] |= LE32_DEC_ENC(phy_data->ntfy->sigs.eht_tb.tb_rx1, + OFDM_UCODE_TRIG_BASE_RX_PRE_FEC_PAD_FACTOR, + IEEE80211_RADIOTAP_EHT_DATA0_PRE_PADD_FACOR_OM); + eht->data[0] |= LE32_DEC_ENC(phy_data->ntfy->sigs.eht_tb.tb_rx1, + OFDM_UCODE_TRIG_BASE_RX_PE_DISAMBIG, IEEE80211_RADIOTAP_EHT_DATA0_PE_DISAMBIGUITY_OM); + eht->data[0] |= LE32_DEC_ENC(phy_data->ntfy->sigs.eht_tb.tb_rx1, + OFDM_UCODE_TRIG_BASE_RX_NUM_OF_LTF_SYM, + IEEE80211_RADIOTAP_EHT_DATA0_EHT_LTF); + eht->data[1] |= LE32_DEC_ENC(phy_data->ntfy->sigs.eht_tb.tb_rx0, + OFDM_UCODE_TRIG_BASE_RX_RU_P80, + IEEE80211_RADIOTAP_EHT_DATA1_PRIMARY_80); - /* TODO: what about IWL_RX_PHY_DATA0_EHT_BW320_SLOT */ + iwl_mld_eht_set_ru_alloc(rx_status, + le32_get_bits(phy_data->ntfy->sigs.eht_tb.tb_rx1, + OFDM_UCODE_TRIG_BASE_RX_RU)); +} - if (!le32_get_bits(data0, IWL_RX_PHY_DATA0_EHT_SIGA_CRC_OK)) - usig->common |= cpu_to_le32(IEEE80211_RADIOTAP_EHT_USIG_COMMON_BAD_USIG_CRC); +static void iwl_mld_eht_decode_user_ru(struct iwl_mld_rx_phy_data *phy_data, + struct ieee80211_radiotap_eht *eht) +{ + u32 phy_bw = phy_data->rate_n_flags & RATE_MCS_CHAN_WIDTH_MSK; - usig->common |= cpu_to_le32(IEEE80211_RADIOTAP_EHT_USIG_COMMON_PHY_VER_KNOWN); - usig->common |= LE32_DEC_ENC(data0, IWL_RX_PHY_DATA0_EHT_PHY_VER, - IEEE80211_RADIOTAP_EHT_USIG_COMMON_PHY_VER); + if (!(phy_data->ntfy->flags & IWL_SNIF_FLAG_VALID_RU)) + return; + +#define __IWL_MLD_ENC_EHT_RU(rt_data, rt_ru, fw_data, fw_ru) \ + eht->data[(rt_data)] |= \ + (cpu_to_le32(IEEE80211_RADIOTAP_EHT_DATA ## rt_data ## _RU_ALLOC_CC_ ## rt_ru ## _KNOWN) | \ + LE32_DEC_ENC(phy_data->ntfy->sigs.eht.cmn[fw_data], \ + OFDM_RX_FRAME_EHT_RU_ALLOC_ ## fw_data ## _ ## fw_ru, \ + IEEE80211_RADIOTAP_EHT_DATA ## rt_data ## _RU_ALLOC_CC_ ## rt_ru)) + +#define _IWL_MLD_ENC_EHT_RU(rt_data, rt_ru, fw_data, fw_ru) \ + __IWL_MLD_ENC_EHT_RU(rt_data, rt_ru, fw_data, fw_ru) + +#define IEEE80211_RADIOTAP_RU_DATA_1_1_1 1 +#define IEEE80211_RADIOTAP_RU_DATA_2_1_1 2 +#define IEEE80211_RADIOTAP_RU_DATA_1_1_2 2 +#define IEEE80211_RADIOTAP_RU_DATA_2_1_2 2 +#define IEEE80211_RADIOTAP_RU_DATA_1_2_1 3 +#define IEEE80211_RADIOTAP_RU_DATA_2_2_1 3 +#define IEEE80211_RADIOTAP_RU_DATA_1_2_2 3 +#define IEEE80211_RADIOTAP_RU_DATA_2_2_2 4 +#define IEEE80211_RADIOTAP_RU_DATA_1_2_3 4 +#define IEEE80211_RADIOTAP_RU_DATA_2_2_3 4 +#define IEEE80211_RADIOTAP_RU_DATA_1_2_4 5 +#define IEEE80211_RADIOTAP_RU_DATA_2_2_4 5 +#define IEEE80211_RADIOTAP_RU_DATA_1_2_5 5 +#define IEEE80211_RADIOTAP_RU_DATA_2_2_5 6 +#define IEEE80211_RADIOTAP_RU_DATA_1_2_6 6 +#define IEEE80211_RADIOTAP_RU_DATA_2_2_6 6 + +#define IWL_RX_RU_DATA_A1 0 +#define IWL_RX_RU_DATA_A2 0 +#define IWL_RX_RU_DATA_A3 0 +#define IWL_RX_RU_DATA_A4 4 +#define IWL_RX_RU_DATA_B1 1 +#define IWL_RX_RU_DATA_B2 1 +#define IWL_RX_RU_DATA_B3 1 +#define IWL_RX_RU_DATA_B4 4 +#define IWL_RX_RU_DATA_C1 2 +#define IWL_RX_RU_DATA_C2 2 +#define IWL_RX_RU_DATA_C3 2 +#define IWL_RX_RU_DATA_C4 5 +#define IWL_RX_RU_DATA_D1 3 +#define IWL_RX_RU_DATA_D2 3 +#define IWL_RX_RU_DATA_D3 3 +#define IWL_RX_RU_DATA_D4 5 + +#define IWL_MLD_ENC_EHT_RU(rt_ru, fw_ru) \ + _IWL_MLD_ENC_EHT_RU(IEEE80211_RADIOTAP_RU_DATA_ ## rt_ru, \ + rt_ru, \ + IWL_RX_RU_DATA_ ## fw_ru, \ + fw_ru) /* - * TODO: what about TB - IWL_RX_PHY_DATA1_EHT_TB_PILOT_TYPE, - * IWL_RX_PHY_DATA1_EHT_TB_LOW_SS + * Hardware labels the content channels/RU allocation values + * as follows: + * + * Content Channel 1 Content Channel 2 + * 20 MHz: A1 + * 40 MHz: A1 B1 + * 80 MHz: A1 C1 B1 D1 + * 160 MHz: A1 C1 A2 C2 B1 D1 B2 D2 + * 320 MHz: A1 C1 A2 C2 A3 C3 A4 C4 B1 D1 B2 D2 B3 D3 B4 D4 */ - eht->known |= cpu_to_le32(IEEE80211_RADIOTAP_EHT_KNOWN_EHT_LTF); - eht->data[0] |= LE32_DEC_ENC(data1, IWL_RX_PHY_DATA1_EHT_SIG_LTF_NUM, + switch (phy_bw) { + case RATE_MCS_CHAN_WIDTH_320: + /* content channel 1 */ + IWL_MLD_ENC_EHT_RU(1_2_3, A3); + IWL_MLD_ENC_EHT_RU(1_2_4, C3); + IWL_MLD_ENC_EHT_RU(1_2_5, A4); + IWL_MLD_ENC_EHT_RU(1_2_6, C4); + /* content channel 2 */ + IWL_MLD_ENC_EHT_RU(2_2_3, B3); + IWL_MLD_ENC_EHT_RU(2_2_4, D3); + IWL_MLD_ENC_EHT_RU(2_2_5, B4); + IWL_MLD_ENC_EHT_RU(2_2_6, D4); + fallthrough; + case RATE_MCS_CHAN_WIDTH_160: + /* content channel 1 */ + IWL_MLD_ENC_EHT_RU(1_2_1, A2); + IWL_MLD_ENC_EHT_RU(1_2_2, C2); + /* content channel 2 */ + IWL_MLD_ENC_EHT_RU(2_2_1, B2); + IWL_MLD_ENC_EHT_RU(2_2_2, D2); + fallthrough; + case RATE_MCS_CHAN_WIDTH_80: + /* content channel 1 */ + IWL_MLD_ENC_EHT_RU(1_1_2, C1); + /* content channel 2 */ + IWL_MLD_ENC_EHT_RU(2_1_2, D1); + fallthrough; + case RATE_MCS_CHAN_WIDTH_40: + /* content channel 2 */ + IWL_MLD_ENC_EHT_RU(2_1_1, B1); + fallthrough; + case RATE_MCS_CHAN_WIDTH_20: + /* content channel 1 */ + IWL_MLD_ENC_EHT_RU(1_1_1, A1); + break; + } +} + +static void iwl_mld_decode_eht_non_tb(struct iwl_mld_rx_phy_data *phy_data, + struct ieee80211_rx_status *rx_status, + struct ieee80211_radiotap_eht *eht) +{ + eht->known |= cpu_to_le32(IEEE80211_RADIOTAP_EHT_KNOWN_SPATIAL_REUSE | + /* All RU allocating size/index is in TB format */ + IEEE80211_RADIOTAP_EHT_KNOWN_RU_ALLOC_TB_FMT | + IEEE80211_RADIOTAP_EHT_KNOWN_LDPC_EXTRA_SYM_OM | + IEEE80211_RADIOTAP_EHT_KNOWN_PRE_PADD_FACOR_OM | + IEEE80211_RADIOTAP_EHT_KNOWN_PE_DISAMBIGUITY_OM | + IEEE80211_RADIOTAP_EHT_KNOWN_EHT_LTF | + IEEE80211_RADIOTAP_EHT_KNOWN_PRIMARY_80 | + IEEE80211_RADIOTAP_EHT_KNOWN_NR_NON_OFDMA_USERS_M); + + eht->data[0] |= LE32_DEC_ENC(phy_data->ntfy->sigs.eht.b1, + OFDM_RX_FRAME_EHT_SPATIAL_REUSE, + IEEE80211_RADIOTAP_EHT_DATA0_SPATIAL_REUSE); + eht->data[8] |= LE32_DEC_ENC(phy_data->ntfy->sigs.eht.b2, + OFDM_RX_FRAME_EHT_STA_RU_PS160, + IEEE80211_RADIOTAP_EHT_DATA8_RU_ALLOC_TB_FMT_PS_160); + eht->data[8] |= LE32_DEC_ENC(phy_data->ntfy->sigs.eht.b2, + OFDM_RX_FRAME_EHT_STA_RU, + IEEE80211_RADIOTAP_EHT_DATA8_RU_ALLOC_TB_FMT_B0 | + IEEE80211_RADIOTAP_EHT_DATA8_RU_ALLOC_TB_FMT_B7_B1); + eht->data[0] |= LE32_DEC_ENC(phy_data->ntfy->sigs.eht.b1, + OFDM_RX_FRAME_EHT_CODING_EXTRA_SYM, + IEEE80211_RADIOTAP_EHT_DATA0_LDPC_EXTRA_SYM_OM); + eht->data[0] |= LE32_DEC_ENC(phy_data->ntfy->sigs.eht.b1, + OFDM_RX_FRAME_EHT_PRE_FEC_PAD_FACTOR, + IEEE80211_RADIOTAP_EHT_DATA0_PRE_PADD_FACOR_OM); + eht->data[0] |= LE32_DEC_ENC(phy_data->ntfy->sigs.eht.b1, + OFDM_RX_FRAME_EHT_PE_DISAMBIG, + IEEE80211_RADIOTAP_EHT_DATA0_PE_DISAMBIGUITY_OM); + eht->data[0] |= LE32_DEC_ENC(phy_data->ntfy->sigs.eht.b1, + OFDM_RX_FRAME_EHT_NUM_OF_LTF_SYM, IEEE80211_RADIOTAP_EHT_DATA0_EHT_LTF); + eht->data[1] |= LE32_DEC_ENC(phy_data->ntfy->sigs.eht.b2, + OFDM_RX_FRAME_EHT_STA_RU_P80, + IEEE80211_RADIOTAP_EHT_DATA1_PRIMARY_80); + eht->data[7] |= LE32_DEC_ENC(phy_data->ntfy->sigs.eht.b1, + OFDM_RX_FRAME_EHT_NUM_OF_USERS, + IEEE80211_RADIOTAP_EHT_DATA7_NUM_OF_NON_OFDMA_USERS); - if (info_type == IWL_RX_PHY_INFO_TYPE_EHT_TB_EXT || - info_type == IWL_RX_PHY_INFO_TYPE_EHT_TB) - iwl_mld_decode_eht_ext_tb(mld, phy_data, rx_status, eht, usig); + iwl_mld_eht_decode_user_ru(phy_data, eht); - if (info_type == IWL_RX_PHY_INFO_TYPE_EHT_MU_EXT || - info_type == IWL_RX_PHY_INFO_TYPE_EHT_MU) - iwl_mld_decode_eht_ext_mu(mld, phy_data, rx_status, eht, usig); + iwl_mld_eht_set_ru_alloc(rx_status, + le32_get_bits(phy_data->ntfy->sigs.eht.b2, + OFDM_RX_FRAME_EHT_STA_RU)); +} + +static void iwl_mld_decode_eht_phy_data(struct iwl_mld_rx_phy_data *phy_data, + struct ieee80211_rx_status *rx_status, + struct ieee80211_radiotap_eht *eht) +{ + u32 he_type = phy_data->rate_n_flags & RATE_MCS_HE_TYPE_MSK; + + if (he_type == RATE_MCS_HE_TYPE_TRIG) + iwl_mld_decode_eht_tb(phy_data, rx_status, eht); + else + iwl_mld_decode_eht_non_tb(phy_data, rx_status, eht); } static void iwl_mld_rx_eht(struct iwl_mld *mld, struct sk_buff *skb, - struct iwl_mld_rx_phy_data *phy_data, - int queue) + struct iwl_mld_rx_phy_data *phy_data) { struct ieee80211_rx_status *rx_status = IEEE80211_SKB_RXCB(skb); struct ieee80211_radiotap_eht *eht; - struct ieee80211_radiotap_eht_usig *usig; size_t eht_len = sizeof(*eht); - u32 rate_n_flags = phy_data->rate_n_flags; u32 he_type = rate_n_flags & RATE_MCS_HE_TYPE_MSK; /* EHT and HE have the same values for LTF */ u8 ltf = IEEE80211_RADIOTAP_HE_DATA5_LTF_SIZE_UNKNOWN; - u16 phy_info = phy_data->phy_info; - u32 bw; /* u32 for 1 user_info */ if (phy_data->with_data) @@ -1065,50 +1106,7 @@ static void iwl_mld_rx_eht(struct iwl_mld *mld, struct sk_buff *skb, eht = iwl_mld_radiotap_put_tlv(skb, IEEE80211_RADIOTAP_EHT, eht_len); - usig = iwl_mld_radiotap_put_tlv(skb, IEEE80211_RADIOTAP_EHT_USIG, - sizeof(*usig)); rx_status->flag |= RX_FLAG_RADIOTAP_TLV_AT_END; - usig->common |= - cpu_to_le32(IEEE80211_RADIOTAP_EHT_USIG_COMMON_BW_KNOWN); - - /* specific handling for 320MHz */ - bw = u32_get_bits(rate_n_flags, RATE_MCS_CHAN_WIDTH_MSK); - if (bw == RATE_MCS_CHAN_WIDTH_320_VAL) - bw += le32_get_bits(phy_data->data0, - IWL_RX_PHY_DATA0_EHT_BW320_SLOT); - - usig->common |= cpu_to_le32 - (FIELD_PREP(IEEE80211_RADIOTAP_EHT_USIG_COMMON_BW, bw)); - - /* report the AMPDU-EOF bit on single frames */ - if (!queue && !(phy_info & IWL_RX_MPDU_PHY_AMPDU)) { - rx_status->flag |= RX_FLAG_AMPDU_DETAILS; - rx_status->flag |= RX_FLAG_AMPDU_EOF_BIT_KNOWN; - if (phy_data->data0 & - cpu_to_le32(IWL_RX_PHY_DATA0_EHT_DELIM_EOF)) - rx_status->flag |= RX_FLAG_AMPDU_EOF_BIT; - } - - /* update aggregation data for monitor sake on default queue */ - if (!queue && (phy_info & IWL_RX_MPDU_PHY_TSF_OVERLOAD) && - (phy_info & IWL_RX_MPDU_PHY_AMPDU) && phy_data->first_subframe) { - rx_status->flag |= RX_FLAG_AMPDU_EOF_BIT_KNOWN; - if (phy_data->data0 & - cpu_to_le32(IWL_RX_PHY_DATA0_EHT_DELIM_EOF)) - rx_status->flag |= RX_FLAG_AMPDU_EOF_BIT; - } - - if (phy_info & IWL_RX_MPDU_PHY_TSF_OVERLOAD) - iwl_mld_decode_eht_phy_data(mld, phy_data, rx_status, eht, usig); - -#define CHECK_TYPE(F) \ - BUILD_BUG_ON(IEEE80211_RADIOTAP_HE_DATA1_FORMAT_ ## F != \ - (RATE_MCS_HE_TYPE_ ## F >> RATE_MCS_HE_TYPE_POS)) - - CHECK_TYPE(SU); - CHECK_TYPE(EXT_SU); - CHECK_TYPE(MU); - CHECK_TYPE(TRIG); switch (u32_get_bits(rate_n_flags, RATE_MCS_HE_GI_LTF_MSK)) { case 0: @@ -1144,20 +1142,18 @@ static void iwl_mld_rx_eht(struct iwl_mld *mld, struct sk_buff *skb, if (ltf != IEEE80211_RADIOTAP_HE_DATA5_LTF_SIZE_UNKNOWN) { eht->known |= cpu_to_le32(IEEE80211_RADIOTAP_EHT_KNOWN_GI); - eht->data[0] |= cpu_to_le32 - (FIELD_PREP(IEEE80211_RADIOTAP_EHT_DATA0_LTF, - ltf) | - FIELD_PREP(IEEE80211_RADIOTAP_EHT_DATA0_GI, - rx_status->eht.gi)); + eht->data[0] |= le32_encode_bits(ltf, + IEEE80211_RADIOTAP_EHT_DATA0_LTF) | + le32_encode_bits(rx_status->eht.gi, + IEEE80211_RADIOTAP_EHT_DATA0_GI); } if (!phy_data->with_data) { eht->known |= cpu_to_le32(IEEE80211_RADIOTAP_EHT_KNOWN_NSS_S | IEEE80211_RADIOTAP_EHT_KNOWN_BEAMFORMED_S); - eht->data[7] |= - le32_encode_bits(le32_get_bits(phy_data->rx_vec[2], - RX_NO_DATA_RX_VEC2_EHT_NSTS_MSK), - IEEE80211_RADIOTAP_EHT_DATA7_NSS_S); + eht->data[7] |= LE32_DEC_ENC(phy_data->ntfy->sigs.eht.b1, + OFDM_RX_FRAME_EHT_NSTS, + IEEE80211_RADIOTAP_EHT_DATA7_NSS_S); if (rate_n_flags & RATE_MCS_BF_MSK) eht->data[7] |= cpu_to_le32(IEEE80211_RADIOTAP_EHT_DATA7_BEAMFORMED_S); @@ -1177,14 +1173,28 @@ static void iwl_mld_rx_eht(struct iwl_mld *mld, struct sk_buff *skb, eht->user_info[0] |= cpu_to_le32(IEEE80211_RADIOTAP_EHT_USER_INFO_CODING); - eht->user_info[0] |= cpu_to_le32 - (FIELD_PREP(IEEE80211_RADIOTAP_EHT_USER_INFO_MCS, - u32_get_bits(rate_n_flags, - RATE_VHT_MCS_RATE_CODE_MSK)) | - FIELD_PREP(IEEE80211_RADIOTAP_EHT_USER_INFO_NSS_O, - u32_get_bits(rate_n_flags, - RATE_MCS_NSS_MSK))); + eht->user_info[0] |= + le32_encode_bits(u32_get_bits(rate_n_flags, + RATE_VHT_MCS_RATE_CODE_MSK), + IEEE80211_RADIOTAP_EHT_USER_INFO_MCS) | + le32_encode_bits(u32_get_bits(rate_n_flags, + RATE_MCS_NSS_MSK), + IEEE80211_RADIOTAP_EHT_USER_INFO_NSS_O); } + + if (likely(!phy_data->ntfy)) + return; + + if (phy_data->with_data) { + eht->user_info[0] |= + cpu_to_le32(IEEE80211_RADIOTAP_EHT_USER_INFO_STA_ID_KNOWN) | + LE32_DEC_ENC(phy_data->ntfy->sigs.eht.user_id, + OFDM_RX_FRAME_EHT_USER_FIELD_ID, + IEEE80211_RADIOTAP_EHT_USER_INFO_STA_ID); + } + + iwl_mld_decode_eht_usig(phy_data, skb); + iwl_mld_decode_eht_phy_data(phy_data, rx_status, eht); } #ifdef CONFIG_IWLWIFI_DEBUGFS @@ -1219,34 +1229,30 @@ static void iwl_mld_add_rtap_sniffer_config(struct iwl_mld *mld, } #endif -/* Note: hdr can be NULL */ -static void iwl_mld_rx_fill_status(struct iwl_mld *mld, int link_id, - struct ieee80211_hdr *hdr, - struct sk_buff *skb, - struct iwl_mld_rx_phy_data *phy_data, - int queue) +static void +iwl_mld_set_rx_nonlegacy_rate_info(u32 rate_n_flags, + struct ieee80211_rx_status *rx_status) +{ + u8 stbc = u32_get_bits(rate_n_flags, RATE_MCS_STBC_MSK); + + /* NSS may be overridden by PHY ntfy with full value */ + rx_status->nss = u32_get_bits(rate_n_flags, RATE_MCS_NSS_MSK) + 1; + rx_status->rate_idx = rate_n_flags & RATE_MCS_CODE_MSK; + rx_status->enc_flags |= stbc << RX_ENC_FLAG_STBC_SHIFT; + if (rate_n_flags & RATE_MCS_LDPC_MSK) + rx_status->enc_flags |= RX_ENC_FLAG_LDPC; +} + +static void iwl_mld_set_rx_rate(struct iwl_mld *mld, + struct iwl_mld_rx_phy_data *phy_data, + struct ieee80211_rx_status *rx_status) { - struct ieee80211_rx_status *rx_status = IEEE80211_SKB_RXCB(skb); - u32 format = phy_data->rate_n_flags & RATE_MCS_MOD_TYPE_MSK; u32 rate_n_flags = phy_data->rate_n_flags; u8 stbc = u32_get_bits(rate_n_flags, RATE_MCS_STBC_MSK); + u32 format = rate_n_flags & RATE_MCS_MOD_TYPE_MSK; bool is_sgi = rate_n_flags & RATE_MCS_SGI_MSK; - phy_data->info_type = IWL_RX_PHY_INFO_TYPE_NONE; - - if (phy_data->phy_info & IWL_RX_MPDU_PHY_TSF_OVERLOAD) - phy_data->info_type = - le32_get_bits(phy_data->data1, - IWL_RX_PHY_DATA1_INFO_TYPE_MASK); - - /* set the preamble flag if appropriate */ - if (format == RATE_MCS_MOD_TYPE_CCK && - phy_data->phy_info & IWL_RX_MPDU_PHY_SHORT_PREAMBLE) - rx_status->enc_flags |= RX_ENC_FLAG_SHORTPRE; - - iwl_mld_fill_signal(mld, link_id, hdr, rx_status, phy_data); - - /* This may be overridden by iwl_mld_rx_he() to HE_RU */ + /* bandwidth may be overridden to RU by PHY ntfy */ switch (rate_n_flags & RATE_MCS_CHAN_WIDTH_MSK) { case RATE_MCS_CHAN_WIDTH_20: break; @@ -1264,66 +1270,12 @@ static void iwl_mld_rx_fill_status(struct iwl_mld *mld, int link_id, break; } - /* must be before L-SIG data */ - if (format == RATE_MCS_MOD_TYPE_HE) - iwl_mld_rx_he(mld, skb, phy_data, queue); - - iwl_mld_decode_lsig(skb, phy_data); - - rx_status->device_timestamp = phy_data->gp2_on_air_rise; - - /* using TLV format and must be after all fixed len fields */ - if (format == RATE_MCS_MOD_TYPE_EHT) - iwl_mld_rx_eht(mld, skb, phy_data, queue); - -#ifdef CONFIG_IWLWIFI_DEBUGFS - if (unlikely(mld->monitor.on)) { - iwl_mld_add_rtap_sniffer_config(mld, skb); - - if (mld->monitor.ptp_time) { - u64 adj_time = - iwl_mld_ptp_get_adj_time(mld, - phy_data->gp2_on_air_rise * - NSEC_PER_USEC); - - rx_status->mactime = div64_u64(adj_time, NSEC_PER_USEC); - rx_status->flag |= RX_FLAG_MACTIME_IS_RTAP_TS64; - rx_status->flag &= ~RX_FLAG_MACTIME; - } - } -#endif - - if (format != RATE_MCS_MOD_TYPE_CCK && is_sgi) - rx_status->enc_flags |= RX_ENC_FLAG_SHORT_GI; - - if (rate_n_flags & RATE_MCS_LDPC_MSK) - rx_status->enc_flags |= RX_ENC_FLAG_LDPC; - switch (format) { - case RATE_MCS_MOD_TYPE_HT: - rx_status->encoding = RX_ENC_HT; - rx_status->rate_idx = RATE_HT_MCS_INDEX(rate_n_flags); - rx_status->enc_flags |= stbc << RX_ENC_FLAG_STBC_SHIFT; - break; - case RATE_MCS_MOD_TYPE_VHT: - case RATE_MCS_MOD_TYPE_HE: - case RATE_MCS_MOD_TYPE_EHT: - if (format == RATE_MCS_MOD_TYPE_VHT) { - rx_status->encoding = RX_ENC_VHT; - } else if (format == RATE_MCS_MOD_TYPE_HE) { - rx_status->encoding = RX_ENC_HE; - rx_status->he_dcm = - !!(rate_n_flags & RATE_HE_DUAL_CARRIER_MODE_MSK); - } else if (format == RATE_MCS_MOD_TYPE_EHT) { - rx_status->encoding = RX_ENC_EHT; - } - - rx_status->nss = u32_get_bits(rate_n_flags, - RATE_MCS_NSS_MSK) + 1; - rx_status->rate_idx = rate_n_flags & RATE_MCS_CODE_MSK; - rx_status->enc_flags |= stbc << RX_ENC_FLAG_STBC_SHIFT; - break; - default: { + case RATE_MCS_MOD_TYPE_CCK: + if (phy_data->phy_info & IWL_RX_MPDU_PHY_SHORT_PREAMBLE) + rx_status->enc_flags |= RX_ENC_FLAG_SHORTPRE; + fallthrough; + case RATE_MCS_MOD_TYPE_LEGACY_OFDM: { int rate = iwl_mld_legacy_hw_idx_to_mac80211_idx(rate_n_flags, rx_status->band); @@ -1337,12 +1289,88 @@ static void iwl_mld_rx_fill_status(struct iwl_mld *mld, int link_id, /* invalid rate */ rx_status->rate_idx = 0; - if (net_ratelimit()) + /* + * In monitor mode we can see CCK frames on 5 or 6 GHz, usually + * just the (possibly malformed) PHY header by accident, since + * the decoder doesn't seem to turn off CCK. We cannot correctly + * encode the rate to mac80211 (and therefore not in radiotap) + * since we give the per-band index which doesn't cover those + * rates. + */ + if (!mld->monitor.on && net_ratelimit()) IWL_ERR(mld, "invalid rate_n_flags=0x%x, band=%d\n", rate_n_flags, rx_status->band); break; } + case RATE_MCS_MOD_TYPE_HT: + rx_status->encoding = RX_ENC_HT; + rx_status->rate_idx = RATE_HT_MCS_INDEX(rate_n_flags); + rx_status->enc_flags |= stbc << RX_ENC_FLAG_STBC_SHIFT; + break; + case RATE_MCS_MOD_TYPE_VHT: + rx_status->encoding = RX_ENC_VHT; + iwl_mld_set_rx_nonlegacy_rate_info(rate_n_flags, rx_status); + break; + case RATE_MCS_MOD_TYPE_HE: + rx_status->encoding = RX_ENC_HE; + rx_status->he_dcm = + !!(rate_n_flags & RATE_HE_DUAL_CARRIER_MODE_MSK); + iwl_mld_set_rx_nonlegacy_rate_info(rate_n_flags, rx_status); + break; + case RATE_MCS_MOD_TYPE_EHT: + rx_status->encoding = RX_ENC_EHT; + iwl_mld_set_rx_nonlegacy_rate_info(rate_n_flags, rx_status); + break; + default: + WARN_ON_ONCE(1); } + + if (format != RATE_MCS_MOD_TYPE_CCK && is_sgi) + rx_status->enc_flags |= RX_ENC_FLAG_SHORT_GI; +} + +/* Note: hdr can be NULL */ +static void iwl_mld_rx_fill_status(struct iwl_mld *mld, int link_id, + struct ieee80211_hdr *hdr, + struct sk_buff *skb, + struct iwl_mld_rx_phy_data *phy_data) +{ + struct ieee80211_rx_status *rx_status = IEEE80211_SKB_RXCB(skb); + u32 rate_n_flags = phy_data->rate_n_flags; + u32 format = rate_n_flags & RATE_MCS_MOD_TYPE_MSK; + + iwl_mld_fill_signal(mld, link_id, hdr, rx_status, phy_data); + + rx_status->device_timestamp = phy_data->gp2_on_air_rise; + + iwl_mld_set_rx_rate(mld, phy_data, rx_status); + + /* must be before L-SIG data (radiotap field order) */ + if (format == RATE_MCS_MOD_TYPE_HE) + iwl_mld_rx_he(skb, phy_data); + + iwl_mld_decode_lsig(skb, phy_data); + + /* TLVs - must be after radiotap fixed fields */ + if (format == RATE_MCS_MOD_TYPE_EHT) + iwl_mld_rx_eht(mld, skb, phy_data); + +#ifdef CONFIG_IWLWIFI_DEBUGFS + if (unlikely(mld->monitor.on)) { + iwl_mld_add_rtap_sniffer_config(mld, skb); + + if (mld->monitor.ptp_time) { + u64 adj_time = + iwl_mld_ptp_get_adj_time(mld, + phy_data->gp2_on_air_rise * + NSEC_PER_USEC); + + rx_status->mactime = div64_u64(adj_time, NSEC_PER_USEC); + rx_status->flag |= RX_FLAG_MACTIME_IS_RTAP_TS64; + rx_status->flag &= ~RX_FLAG_MACTIME; + } + } +#endif } /* iwl_mld_create_skb adds the rxb to a new skb */ @@ -1763,13 +1791,36 @@ static int iwl_mld_rx_crypto(struct iwl_mld *mld, return 0; } -static void iwl_mld_rx_update_ampdu_ref(struct iwl_mld *mld, - struct iwl_mld_rx_phy_data *phy_data, - struct ieee80211_rx_status *rx_status) +static void iwl_mld_rx_update_ampdu_data(struct iwl_mld *mld, + struct iwl_mld_rx_phy_data *phy_data, + struct ieee80211_rx_status *rx_status) { + u32 format = phy_data->rate_n_flags & RATE_MCS_MOD_TYPE_MSK; bool toggle_bit = phy_data->phy_info & IWL_RX_MPDU_PHY_AMPDU_TOGGLE; + switch (format) { + case RATE_MCS_MOD_TYPE_CCK: + case RATE_MCS_MOD_TYPE_LEGACY_OFDM: + /* no aggregation possible */ + return; + case RATE_MCS_MOD_TYPE_HT: + case RATE_MCS_MOD_TYPE_VHT: + /* single frames are not A-MPDU format */ + if (!(phy_data->phy_info & IWL_RX_MPDU_PHY_AMPDU)) + return; + break; + default: + /* HE/EHT/UHR have A-MPDU format for single frames */ + if (!(phy_data->phy_info & IWL_RX_MPDU_PHY_AMPDU)) { + rx_status->flag |= RX_FLAG_AMPDU_DETAILS; + rx_status->flag |= RX_FLAG_AMPDU_EOF_BIT_KNOWN; + if (phy_data->phy_info & IWL_RX_MPDU_PHY_EOF_INDICATION) + rx_status->flag |= RX_FLAG_AMPDU_EOF_BIT; + return; + } + } + rx_status->flag |= RX_FLAG_AMPDU_DETAILS; /* Toggle is switched whenever new aggregation starts. Make * sure ampdu_reference is never 0 so we can later use it to @@ -1781,6 +1832,11 @@ static void iwl_mld_rx_update_ampdu_ref(struct iwl_mld *mld, mld->monitor.ampdu_ref++; mld->monitor.ampdu_toggle = toggle_bit; phy_data->first_subframe = true; + + /* report EOF bit on the first subframe */ + rx_status->flag |= RX_FLAG_AMPDU_EOF_BIT_KNOWN; + if (phy_data->phy_info & IWL_RX_MPDU_PHY_EOF_INDICATION) + rx_status->flag |= RX_FLAG_AMPDU_EOF_BIT; } rx_status->ampdu_reference = mld->monitor.ampdu_ref; } @@ -1824,6 +1880,8 @@ void iwl_mld_rx_mpdu(struct iwl_mld *mld, struct napi_struct *napi, "FW lied about packet len (%d)\n", pkt_len)) return; + iwl_mld_fill_phy_data_from_mpdu(mld, mpdu_desc, &phy_data); + /* Don't use dev_alloc_skb(), we'll have enough headroom once * ieee80211_hdr pulled. */ @@ -1835,8 +1893,6 @@ void iwl_mld_rx_mpdu(struct iwl_mld *mld, struct napi_struct *napi, hdr = (void *)(pkt->data + mpdu_desc_size); - iwl_mld_fill_phy_data(mld, mpdu_desc, &phy_data); - if (mpdu_desc->mac_flags2 & IWL_RX_MPDU_MFLG2_PAD) { /* If the device inserted padding it means that (it thought) * the 802.11 header wasn't a multiple of 4 bytes long. In @@ -1861,9 +1917,8 @@ void iwl_mld_rx_mpdu(struct iwl_mld *mld, struct napi_struct *napi, if (drop) goto drop; - /* update aggregation data for monitor sake on default queue */ - if (!queue && (phy_data.phy_info & IWL_RX_MPDU_PHY_AMPDU)) - iwl_mld_rx_update_ampdu_ref(mld, &phy_data, rx_status); + if (unlikely(mld->monitor.on)) + iwl_mld_rx_update_ampdu_data(mld, &phy_data, rx_status); /* Keep packets with CRC errors (and with overrun) for monitor mode * (otherwise the firmware discards them) but mark them as bad. @@ -1897,7 +1952,7 @@ void iwl_mld_rx_mpdu(struct iwl_mld *mld, struct napi_struct *napi, link_id = u8_get_bits(mpdu_desc->mac_phy_band, IWL_RX_MPDU_MAC_PHY_BAND_LINK_MASK); - iwl_mld_rx_fill_status(mld, link_id, hdr, skb, &phy_data, queue); + iwl_mld_rx_fill_status(mld, link_id, hdr, skb, &phy_data); if (iwl_mld_rx_crypto(mld, sta, hdr, rx_status, mpdu_desc, queue, le32_to_cpu(pkt->len_n_flags), &crypto_len)) @@ -2031,87 +2086,64 @@ void iwl_mld_handle_rx_queues_sync_notif(struct iwl_mld *mld, wake_up(&mld->rxq_sync.waitq); } -void iwl_mld_rx_monitor_no_data(struct iwl_mld *mld, struct napi_struct *napi, - struct iwl_rx_packet *pkt, int queue) +static void iwl_mld_no_data_rx(struct iwl_mld *mld, + struct napi_struct *napi, + struct iwl_rx_phy_air_sniffer_ntfy *ntfy) { - struct iwl_rx_no_data_ver_3 *desc; - struct iwl_mld_rx_phy_data phy_data; struct ieee80211_rx_status *rx_status; + struct iwl_mld_rx_phy_data phy_data = { + .ntfy = ntfy, + .phy_info = 0, /* short preamble set below */ + .rate_n_flags = le32_to_cpu(ntfy->rate), + .gp2_on_air_rise = le32_to_cpu(ntfy->on_air_rise_time), + .energy_a = ntfy->rssi_a, + .energy_b = ntfy->rssi_b, + }; + u32 format = phy_data.rate_n_flags & RATE_MCS_MOD_TYPE_MSK; struct sk_buff *skb; - u32 format, rssi; - u8 channel; - if (unlikely(mld->fw_status.in_hw_restart)) - return; - - if (IWL_FW_CHECK(mld, iwl_rx_packet_payload_len(pkt) < sizeof(*desc), - "Bad RX_NO_DATA_NOTIF size (%d)\n", - iwl_rx_packet_payload_len(pkt))) - return; - - desc = (void *)pkt->data; - - rssi = le32_to_cpu(desc->rssi); - channel = u32_get_bits(rssi, RX_NO_DATA_CHANNEL_MSK); - - phy_data.energy_a = u32_get_bits(rssi, RX_NO_DATA_CHAIN_A_MSK); - phy_data.energy_b = u32_get_bits(rssi, RX_NO_DATA_CHAIN_B_MSK); - phy_data.data0 = desc->phy_info[0]; - phy_data.data1 = desc->phy_info[1]; - phy_data.phy_info = IWL_RX_MPDU_PHY_TSF_OVERLOAD; - phy_data.gp2_on_air_rise = le32_to_cpu(desc->on_air_rise_time); - phy_data.rate_n_flags = iwl_v3_rate_from_v2_v3(desc->rate, - mld->fw_rates_ver_3); - phy_data.with_data = false; - - BUILD_BUG_ON(sizeof(phy_data.rx_vec) != sizeof(desc->rx_vec)); - memcpy(phy_data.rx_vec, desc->rx_vec, sizeof(phy_data.rx_vec)); - - format = phy_data.rate_n_flags & RATE_MCS_MOD_TYPE_MSK; - - /* Don't use dev_alloc_skb(), we'll have enough headroom once - * ieee80211_hdr pulled. - */ skb = alloc_skb(128, GFP_ATOMIC); - if (!skb) { - IWL_ERR(mld, "alloc_skb failed\n"); + if (!skb) return; - } rx_status = IEEE80211_SKB_RXCB(skb); /* 0-length PSDU */ rx_status->flag |= RX_FLAG_NO_PSDU; - /* mark as failed PLCP on any errors to skip checks in mac80211 */ - if (le32_get_bits(desc->info, RX_NO_DATA_INFO_ERR_MSK) != - RX_NO_DATA_INFO_ERR_NONE) - rx_status->flag |= RX_FLAG_FAILED_PLCP_CRC; - - switch (le32_get_bits(desc->info, RX_NO_DATA_INFO_TYPE_MSK)) { - case RX_NO_DATA_INFO_TYPE_NDP: + switch (ntfy->status) { + case IWL_SNIF_STAT_PLCP_RX_OK: + /* we only get here with sounding PPDUs */ rx_status->zero_length_psdu_type = IEEE80211_RADIOTAP_ZERO_LEN_PSDU_SOUNDING; break; - case RX_NO_DATA_INFO_TYPE_MU_UNMATCHED: - case RX_NO_DATA_INFO_TYPE_TB_UNMATCHED: + case IWL_SNIF_STAT_AID_NOT_FOR_US: rx_status->zero_length_psdu_type = IEEE80211_RADIOTAP_ZERO_LEN_PSDU_NOT_CAPTURED; break; + case IWL_SNIF_STAT_PLCP_RX_LSIG_ERR: + case IWL_SNIF_STAT_PLCP_RX_SIGA_ERR: + case IWL_SNIF_STAT_PLCP_RX_SIGB_ERR: + case IWL_SNIF_STAT_UNKNOWN_ERROR: default: + rx_status->flag |= RX_FLAG_FAILED_PLCP_CRC; + fallthrough; + case IWL_SNIF_STAT_UNEXPECTED_TB: + case IWL_SNIF_STAT_UNSUPPORTED_RATE: rx_status->zero_length_psdu_type = IEEE80211_RADIOTAP_ZERO_LEN_PSDU_VENDOR; - break; + /* we could include the real reason in a vendor TLV */ } - rx_status->band = channel > 14 ? NL80211_BAND_5GHZ : - NL80211_BAND_2GHZ; + if (format == RATE_MCS_MOD_TYPE_CCK && + ntfy->legacy_sig.cck & cpu_to_le32(CCK_CRFR_SHORT_PREAMBLE)) + phy_data.phy_info |= IWL_RX_MPDU_PHY_SHORT_PREAMBLE; - rx_status->freq = ieee80211_channel_to_frequency(channel, - rx_status->band); + iwl_mld_fill_rx_status_band_freq(IEEE80211_SKB_RXCB(skb), + ntfy->band, ntfy->channel); /* link ID is ignored for NULL header */ - iwl_mld_rx_fill_status(mld, -1, NULL, skb, &phy_data, queue); + iwl_mld_rx_fill_status(mld, -1, NULL, skb, &phy_data); /* No more radiotap info should be added after this point. * Mark it as mac header for upper layers to know where @@ -2119,29 +2151,72 @@ void iwl_mld_rx_monitor_no_data(struct iwl_mld *mld, struct napi_struct *napi, */ skb_set_mac_header(skb, skb->len); - /* Override the nss from the rx_vec since the rate_n_flags has - * only 1 bit for the nss which gives a max of 2 ss but there - * may be up to 8 spatial streams. - */ - switch (format) { - case RATE_MCS_MOD_TYPE_VHT: - rx_status->nss = - le32_get_bits(desc->rx_vec[0], - RX_NO_DATA_RX_VEC0_VHT_NSTS_MSK) + 1; - break; - case RATE_MCS_MOD_TYPE_HE: - rx_status->nss = - le32_get_bits(desc->rx_vec[0], - RX_NO_DATA_RX_VEC0_HE_NSTS_MSK) + 1; - break; - case RATE_MCS_MOD_TYPE_EHT: - rx_status->nss = - le32_get_bits(desc->rx_vec[2], - RX_NO_DATA_RX_VEC2_EHT_NSTS_MSK) + 1; - } - /* pass the packet to mac80211 */ rcu_read_lock(); ieee80211_rx_napi(mld->hw, NULL, skb, napi); rcu_read_unlock(); } + +void iwl_mld_handle_phy_air_sniffer_notif(struct iwl_mld *mld, + struct napi_struct *napi, + struct iwl_rx_packet *pkt) +{ + struct iwl_rx_phy_air_sniffer_ntfy *ntfy = (void *)pkt->data; + bool is_ndp = false; + u32 he_type; + + if (IWL_FW_CHECK(mld, iwl_rx_packet_payload_len(pkt) < sizeof(*ntfy), + "invalid air sniffer notification size\n")) + return; + + /* check if there's an old one to release as errored */ + if (mld->monitor.phy.valid && !mld->monitor.phy.used) { + /* didn't capture data, so override status */ + mld->monitor.phy.data.status = IWL_SNIF_STAT_AID_NOT_FOR_US; + iwl_mld_no_data_rx(mld, napi, &mld->monitor.phy.data); + } + + /* old data is no longer valid now */ + mld->monitor.phy.valid = false; + + he_type = le32_to_cpu(ntfy->rate) & RATE_MCS_HE_TYPE_MSK; + + switch (le32_to_cpu(ntfy->rate) & RATE_MCS_MOD_TYPE_MSK) { + case RATE_MCS_MOD_TYPE_HT: + is_ndp = !le32_get_bits(ntfy->sigs.ht.a1, + OFDM_RX_FRAME_HT_LENGTH); + break; + case RATE_MCS_MOD_TYPE_VHT: + is_ndp = le32_get_bits(ntfy->sigs.vht.a0, + OFDM_RX_FRAME_VHT_NUM_OF_DATA_SYM_VALID) && + !le32_get_bits(ntfy->sigs.vht.a0, + OFDM_RX_FRAME_VHT_NUM_OF_DATA_SYM); + break; + case RATE_MCS_MOD_TYPE_HE: + if (he_type == RATE_MCS_HE_TYPE_TRIG) + break; + is_ndp = le32_get_bits(ntfy->sigs.he.a3, + OFDM_RX_FRAME_HE_NUM_OF_DATA_SYM_VALID) && + !le32_get_bits(ntfy->sigs.he.a3, + OFDM_RX_FRAME_HE_NUM_OF_DATA_SYM); + break; + case RATE_MCS_MOD_TYPE_EHT: + if (he_type == RATE_MCS_HE_TYPE_TRIG) + break; + is_ndp = le32_get_bits(ntfy->sigs.eht.sig2, + OFDM_RX_FRAME_EHT_NUM_OF_DATA_SYM_VALID) && + !le32_get_bits(ntfy->sigs.eht.sig2, + OFDM_RX_FRAME_EHT_NUM_OF_DATA_SYM); + break; + } + + if (ntfy->status != IWL_SNIF_STAT_PLCP_RX_OK || is_ndp) { + iwl_mld_no_data_rx(mld, napi, ntfy); + return; + } + + /* hang on to it for the RX_MPDU data packet(s) */ + mld->monitor.phy.data = *ntfy; + mld->monitor.phy.valid = true; + mld->monitor.phy.used = false; +} diff --git a/drivers/net/wireless/intel/iwlwifi/mld/rx.h b/drivers/net/wireless/intel/iwlwifi/mld/rx.h index 2beabd7e70b1..09dddbd40f55 100644 --- a/drivers/net/wireless/intel/iwlwifi/mld/rx.h +++ b/drivers/net/wireless/intel/iwlwifi/mld/rx.h @@ -66,7 +66,8 @@ void iwl_mld_pass_packet_to_mac80211(struct iwl_mld *mld, struct sk_buff *skb, int queue, struct ieee80211_sta *sta); -void iwl_mld_rx_monitor_no_data(struct iwl_mld *mld, struct napi_struct *napi, - struct iwl_rx_packet *pkt, int queue); +void iwl_mld_handle_phy_air_sniffer_notif(struct iwl_mld *mld, + struct napi_struct *napi, + struct iwl_rx_packet *pkt); #endif /* __iwl_mld_agg_h__ */ From 38f79506195c32f4f1d1fe5808e7294cd1f298ad Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Sun, 19 Oct 2025 11:45:03 +0300 Subject: [PATCH 0315/1381] wifi: iwlwifi: mld: include raw PHY notification in radiotap This is useful for debugging and can also be used to see anything that isn't encoded in radiotap (yet.) Signed-off-by: Johannes Berg Signed-off-by: Miri Korenblit Link: https://patch.msgid.link/20251019114304.5fe26e9061f2.Iefb45e3a6a2a62ff3247db4de3777059d390af95@changeid --- drivers/net/wireless/intel/iwlwifi/mld/rx.c | 45 +++++++++++++++++++-- 1 file changed, 42 insertions(+), 3 deletions(-) diff --git a/drivers/net/wireless/intel/iwlwifi/mld/rx.c b/drivers/net/wireless/intel/iwlwifi/mld/rx.c index 052a19bb85b4..19e78d2fbf9b 100644 --- a/drivers/net/wireless/intel/iwlwifi/mld/rx.c +++ b/drivers/net/wireless/intel/iwlwifi/mld/rx.c @@ -1217,8 +1217,9 @@ static void iwl_mld_add_rtap_sniffer_config(struct iwl_mld *mld, radiotap->oui[0] = 0xf6; radiotap->oui[1] = 0x54; radiotap->oui[2] = 0x25; - /* radiotap sniffer config sub-namespace */ + /* Intel OUI default radiotap subtype */ radiotap->oui_subtype = 1; + /* Sniffer config element type */ radiotap->vendor_type = 0; /* fill the data now */ @@ -1229,6 +1230,34 @@ static void iwl_mld_add_rtap_sniffer_config(struct iwl_mld *mld, } #endif +static void iwl_mld_add_rtap_sniffer_phy_data(struct iwl_mld *mld, + struct sk_buff *skb, + struct iwl_rx_phy_air_sniffer_ntfy *ntfy) +{ + struct ieee80211_rx_status *rx_status = IEEE80211_SKB_RXCB(skb); + struct ieee80211_radiotap_vendor_content *radiotap; + const u16 vendor_data_len = sizeof(*ntfy); + + radiotap = + iwl_mld_radiotap_put_tlv(skb, + IEEE80211_RADIOTAP_VENDOR_NAMESPACE, + sizeof(*radiotap) + vendor_data_len); + + /* Intel OUI */ + radiotap->oui[0] = 0xf6; + radiotap->oui[1] = 0x54; + radiotap->oui[2] = 0x25; + /* Intel OUI default radiotap subtype */ + radiotap->oui_subtype = 1; + /* PHY data element type */ + radiotap->vendor_type = cpu_to_le16(1); + + /* fill the data now */ + memcpy(radiotap->data, ntfy, vendor_data_len); + + rx_status->flag |= RX_FLAG_RADIOTAP_TLV_AT_END; +} + static void iwl_mld_set_rx_nonlegacy_rate_info(u32 rate_n_flags, struct ieee80211_rx_status *rx_status) @@ -1371,6 +1400,9 @@ static void iwl_mld_rx_fill_status(struct iwl_mld *mld, int link_id, } } #endif + + if (phy_data->ntfy) + iwl_mld_add_rtap_sniffer_phy_data(mld, skb, phy_data->ntfy); } /* iwl_mld_create_skb adds the rxb to a new skb */ @@ -1866,6 +1898,7 @@ void iwl_mld_rx_mpdu(struct iwl_mld *mld, struct napi_struct *napi, u32 mpdu_len; enum iwl_mld_reorder_result reorder_res; struct ieee80211_rx_status *rx_status; + unsigned int alloc_size = 128; if (unlikely(mld->fw_status.in_hw_restart)) return; @@ -1884,8 +1917,13 @@ void iwl_mld_rx_mpdu(struct iwl_mld *mld, struct napi_struct *napi, /* Don't use dev_alloc_skb(), we'll have enough headroom once * ieee80211_hdr pulled. + * + * For monitor mode we need more space to include the full PHY + * notification data. */ - skb = alloc_skb(128, GFP_ATOMIC); + if (unlikely(mld->monitor.on) && phy_data.ntfy) + alloc_size += sizeof(struct iwl_rx_phy_air_sniffer_ntfy); + skb = alloc_skb(alloc_size, GFP_ATOMIC); if (!skb) { IWL_ERR(mld, "alloc_skb failed\n"); return; @@ -2102,7 +2140,8 @@ static void iwl_mld_no_data_rx(struct iwl_mld *mld, u32 format = phy_data.rate_n_flags & RATE_MCS_MOD_TYPE_MSK; struct sk_buff *skb; - skb = alloc_skb(128, GFP_ATOMIC); + skb = alloc_skb(128 + sizeof(struct iwl_rx_phy_air_sniffer_ntfy), + GFP_ATOMIC); if (!skb) return; From 7f59fadbcbbc7881e434544531a1d6c04036f034 Mon Sep 17 00:00:00 2001 From: Nidhish A N Date: Sun, 19 Oct 2025 11:45:04 +0300 Subject: [PATCH 0316/1381] wifi: iwlwifi: fw: remove support of several iwl_lari_config_change_cmd versions We only need versions 1, 6, 8 and 12. Remove versions 2, 3, 4, 5, 7, 9, 10, 11. Signed-off-by: Nidhish A N Signed-off-by: Miri Korenblit Link: https://patch.msgid.link/20251019114304.5c3de58594e8.I6e6f8707eb66a2b540fb19303c06393f13e1f68e@changeid --- .../wireless/intel/iwlwifi/fw/api/nvm-reg.h | 134 +----------------- .../wireless/intel/iwlwifi/fw/regulatory.c | 26 +--- 2 files changed, 11 insertions(+), 149 deletions(-) diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/nvm-reg.h b/drivers/net/wireless/intel/iwlwifi/fw/api/nvm-reg.h index e90f3187e55c..4644fc1aa1ec 100644 --- a/drivers/net/wireless/intel/iwlwifi/fw/api/nvm-reg.h +++ b/drivers/net/wireless/intel/iwlwifi/fw/api/nvm-reg.h @@ -18,13 +18,8 @@ enum iwl_regulatory_and_nvm_subcmd_ids { /** * @LARI_CONFIG_CHANGE: &struct iwl_lari_config_change_cmd_v1, - * &struct iwl_lari_config_change_cmd_v2, - * &struct iwl_lari_config_change_cmd_v3, - * &struct iwl_lari_config_change_cmd_v4, - * &struct iwl_lari_config_change_cmd_v5, * &struct iwl_lari_config_change_cmd_v6, - * &struct iwl_lari_config_change_cmd_v7, - * &struct iwl_lari_config_change_cmd_v10 or + * &struct iwl_lari_config_change_cmd_v8, * &struct iwl_lari_config_change_cmd */ LARI_CONFIG_CHANGE = 0x1, @@ -564,74 +559,6 @@ struct iwl_lari_config_change_cmd_v1 { __le32 config_bitmap; } __packed; /* LARI_CHANGE_CONF_CMD_S_VER_1 */ -/** - * struct iwl_lari_config_change_cmd_v2 - change LARI configuration - * @config_bitmap: bit map of the config commands. each bit will trigger a - * different predefined FW config operation - * @oem_uhb_allow_bitmap: bitmap of UHB enabled MCC sets - */ -struct iwl_lari_config_change_cmd_v2 { - __le32 config_bitmap; - __le32 oem_uhb_allow_bitmap; -} __packed; /* LARI_CHANGE_CONF_CMD_S_VER_2 */ - -/** - * struct iwl_lari_config_change_cmd_v3 - change LARI configuration - * @config_bitmap: bit map of the config commands. each bit will trigger a - * different predefined FW config operation - * @oem_uhb_allow_bitmap: bitmap of UHB enabled MCC sets - * @oem_11ax_allow_bitmap: bitmap of 11ax allowed MCCs. - * For each supported country, a pair of regulatory override bit and 11ax mode exist - * in the bit field. - */ -struct iwl_lari_config_change_cmd_v3 { - __le32 config_bitmap; - __le32 oem_uhb_allow_bitmap; - __le32 oem_11ax_allow_bitmap; -} __packed; /* LARI_CHANGE_CONF_CMD_S_VER_3 */ - -/** - * struct iwl_lari_config_change_cmd_v4 - change LARI configuration - * @config_bitmap: Bitmap of the config commands. Each bit will trigger a - * different predefined FW config operation. - * @oem_uhb_allow_bitmap: Bitmap of UHB enabled MCC sets. - * @oem_11ax_allow_bitmap: Bitmap of 11ax allowed MCCs. There are two bits - * per country, one to indicate whether to override and the other to - * indicate the value to use. - * @oem_unii4_allow_bitmap: Bitmap of unii4 allowed MCCs.There are two bits - * per country, one to indicate whether to override and the other to - * indicate allow/disallow unii4 channels. - */ -struct iwl_lari_config_change_cmd_v4 { - __le32 config_bitmap; - __le32 oem_uhb_allow_bitmap; - __le32 oem_11ax_allow_bitmap; - __le32 oem_unii4_allow_bitmap; -} __packed; /* LARI_CHANGE_CONF_CMD_S_VER_4 */ - -/** - * struct iwl_lari_config_change_cmd_v5 - change LARI configuration - * @config_bitmap: Bitmap of the config commands. Each bit will trigger a - * different predefined FW config operation. - * @oem_uhb_allow_bitmap: Bitmap of UHB enabled MCC sets. - * @oem_11ax_allow_bitmap: Bitmap of 11ax allowed MCCs. There are two bits - * per country, one to indicate whether to override and the other to - * indicate the value to use. - * @oem_unii4_allow_bitmap: Bitmap of unii4 allowed MCCs.There are two bits - * per country, one to indicate whether to override and the other to - * indicate allow/disallow unii4 channels. - * @chan_state_active_bitmap: Bitmap for overriding channel state to active. - * Each bit represents a country or region to activate, according to the BIOS - * definitions. - */ -struct iwl_lari_config_change_cmd_v5 { - __le32 config_bitmap; - __le32 oem_uhb_allow_bitmap; - __le32 oem_11ax_allow_bitmap; - __le32 oem_unii4_allow_bitmap; - __le32 chan_state_active_bitmap; -} __packed; /* LARI_CHANGE_CONF_CMD_S_VER_5 */ - /** * struct iwl_lari_config_change_cmd_v6 - change LARI configuration * @config_bitmap: Bitmap of the config commands. Each bit will trigger a @@ -659,8 +586,7 @@ struct iwl_lari_config_change_cmd_v6 { } __packed; /* LARI_CHANGE_CONF_CMD_S_VER_6 */ /** - * struct iwl_lari_config_change_cmd_v7 - change LARI configuration - * This structure is used also for lari cmd version 8 and 9. + * struct iwl_lari_config_change_cmd_v8 - change LARI configuration * @config_bitmap: Bitmap of the config commands. Each bit will trigger a * different predefined FW config operation. * @oem_uhb_allow_bitmap: Bitmap of UHB enabled MCC sets. @@ -670,21 +596,19 @@ struct iwl_lari_config_change_cmd_v6 { * @oem_unii4_allow_bitmap: Bitmap of unii4 allowed MCCs.There are two bits * per country, one to indicate whether to override and the other to * indicate allow/disallow unii4 channels. - * For LARI cmd version 4 to 8 - bits 0:3 are supported. - * For LARI cmd version 9 - bits 0:5 are supported. + * bit 0 - 3: supported. * @chan_state_active_bitmap: Bitmap to enable different bands per country * or region. * Each bit represents a country or region, and a band to activate * according to the BIOS definitions. - * For LARI cmd version 7 - bits 0:3 are supported. - * For LARI cmd version 8 - bits 0:4 are supported. + * bit 0 - 4: supported. * @force_disable_channels_bitmap: Bitmap of disabled bands/channels. * Each bit represents a set of channels in a specific band that should be * disabled * @edt_bitmap: Bitmap of energy detection threshold table. * Disable/enable the EDT optimization method for different band. */ -struct iwl_lari_config_change_cmd_v7 { +struct iwl_lari_config_change_cmd_v8 { __le32 config_bitmap; __le32 oem_uhb_allow_bitmap; __le32 oem_11ax_allow_bitmap; @@ -693,48 +617,8 @@ struct iwl_lari_config_change_cmd_v7 { __le32 force_disable_channels_bitmap; __le32 edt_bitmap; } __packed; -/* LARI_CHANGE_CONF_CMD_S_VER_7 */ /* LARI_CHANGE_CONF_CMD_S_VER_8 */ -/* LARI_CHANGE_CONF_CMD_S_VER_9 */ -/** - * struct iwl_lari_config_change_cmd_v10 - change LARI configuration - * @config_bitmap: Bitmap of the config commands. Each bit will trigger a - * different predefined FW config operation. - * @oem_uhb_allow_bitmap: Bitmap of UHB enabled MCC sets. - * @oem_11ax_allow_bitmap: Bitmap of 11ax allowed MCCs. There are two bits - * per country, one to indicate whether to override and the other to - * indicate the value to use. - * @oem_unii4_allow_bitmap: Bitmap of unii4 allowed MCCs.There are two bits - * per country, one to indicate whether to override and the other to - * indicate allow/disallow unii4 channels. - * For LARI cmd version 10 - bits 0:5 are supported. - * @chan_state_active_bitmap: Bitmap to enable different bands per country - * or region. - * Each bit represents a country or region, and a band to activate - * according to the BIOS definitions. - * For LARI cmd version 10 - bits 0:4 are supported. - * @force_disable_channels_bitmap: Bitmap of disabled bands/channels. - * Each bit represents a set of channels in a specific band that should be - * disabled - * @edt_bitmap: Bitmap of energy detection threshold table. - * Disable/enable the EDT optimization method for different band. - * @oem_320mhz_allow_bitmap: 320Mhz bandwidth enablement bitmap per MCC. - * bit0: enable 320Mhz in Japan. - * bit1: enable 320Mhz in South Korea. - * bit 2 - 31: reserved. - */ -struct iwl_lari_config_change_cmd_v10 { - __le32 config_bitmap; - __le32 oem_uhb_allow_bitmap; - __le32 oem_11ax_allow_bitmap; - __le32 oem_unii4_allow_bitmap; - __le32 chan_state_active_bitmap; - __le32 force_disable_channels_bitmap; - __le32 edt_bitmap; - __le32 oem_320mhz_allow_bitmap; -} __packed; -/* LARI_CHANGE_CONF_CMD_S_VER_10 */ /** * struct iwl_lari_config_change_cmd - change LARI configuration @@ -747,14 +631,11 @@ struct iwl_lari_config_change_cmd_v10 { * @oem_unii4_allow_bitmap: Bitmap of unii4 allowed MCCs.There are two bits * per country, one to indicate whether to override and the other to * indicate allow/disallow unii4 channels. - * For LARI cmd version 11 - bits 0:5 are supported. * @chan_state_active_bitmap: Bitmap to enable different bands per country * or region. * Each bit represents a country or region, and a band to activate * according to the BIOS definitions. - * For LARI cmd version 11 - bits 0:4 are supported. - * For LARI cmd version 12 - bits 0:6 are supported and bits 7:31 are - * reserved. + * bit 0 - 6: supported. * @force_disable_channels_bitmap: Bitmap of disabled bands/channels. * Each bit represents a set of channels in a specific band that should be * disabled @@ -781,12 +662,11 @@ struct iwl_lari_config_change_cmd { __le32 oem_320mhz_allow_bitmap; __le32 oem_11be_allow_bitmap; } __packed; -/* LARI_CHANGE_CONF_CMD_S_VER_11 */ /* LARI_CHANGE_CONF_CMD_S_VER_12 */ /* Activate UNII-1 (5.2GHz) for World Wide */ #define ACTIVATE_5G2_IN_WW_MASK BIT(4) -#define CHAN_STATE_ACTIVE_BITMAP_CMD_V11 0x1F +#define CHAN_STATE_ACTIVE_BITMAP_CMD_V8 0x1F #define CHAN_STATE_ACTIVE_BITMAP_CMD_V12 0x7F /** diff --git a/drivers/net/wireless/intel/iwlwifi/fw/regulatory.c b/drivers/net/wireless/intel/iwlwifi/fw/regulatory.c index e1f28b053253..d2ad169ae880 100644 --- a/drivers/net/wireless/intel/iwlwifi/fw/regulatory.c +++ b/drivers/net/wireless/intel/iwlwifi/fw/regulatory.c @@ -543,32 +543,14 @@ static size_t iwl_get_lari_config_cmd_size(u8 cmd_ver) switch (cmd_ver) { case 12: - case 11: cmd_size = sizeof(struct iwl_lari_config_change_cmd); break; - case 10: - cmd_size = sizeof(struct iwl_lari_config_change_cmd_v10); - break; - case 9: case 8: - case 7: - cmd_size = sizeof(struct iwl_lari_config_change_cmd_v7); + cmd_size = sizeof(struct iwl_lari_config_change_cmd_v8); break; case 6: cmd_size = sizeof(struct iwl_lari_config_change_cmd_v6); break; - case 5: - cmd_size = sizeof(struct iwl_lari_config_change_cmd_v5); - break; - case 4: - cmd_size = sizeof(struct iwl_lari_config_change_cmd_v4); - break; - case 3: - cmd_size = sizeof(struct iwl_lari_config_change_cmd_v3); - break; - case 2: - cmd_size = sizeof(struct iwl_lari_config_change_cmd_v2); - break; default: cmd_size = sizeof(struct iwl_lari_config_change_cmd_v1); break; @@ -609,11 +591,11 @@ int iwl_fill_lari_config(struct iwl_fw_runtime *fwrt, if (!has_raw_dsm_capa) value &= DSM_UNII4_ALLOW_BITMAP; - /* Since version 9, bits 4 and 5 are supported + /* Since version 12, bits 4 and 5 are supported * regardless of this capability, By pass this masking * if firmware has capability of accepting raw DSM table. */ - if (!has_raw_dsm_capa && cmd_ver < 9 && + if (!has_raw_dsm_capa && cmd_ver < 12 && !fw_has_capa(&fwrt->fw->ucode_capa, IWL_UCODE_TLV_CAPA_BIOS_OVERRIDE_5G9_FOR_CA)) value &= ~(DSM_VALUE_UNII4_CANADA_OVERRIDE_MSK | @@ -637,7 +619,7 @@ int iwl_fill_lari_config(struct iwl_fw_runtime *fwrt, if (!has_raw_dsm_capa && cmd_ver < 12 && !fw_has_capa(&fwrt->fw->ucode_capa, IWL_UCODE_TLV_CAPA_BIOS_OVERRIDE_UNII4_US_CA)) - value &= CHAN_STATE_ACTIVE_BITMAP_CMD_V11; + value &= CHAN_STATE_ACTIVE_BITMAP_CMD_V8; cmd->chan_state_active_bitmap = cpu_to_le32(value); } From 8377e92a3a08304b06e4a5cef89d49ab93728dc1 Mon Sep 17 00:00:00 2001 From: Emmanuel Grumbach Date: Sun, 19 Oct 2025 11:45:05 +0300 Subject: [PATCH 0317/1381] wifi: iwlwifi: be more chatty when we fail to find a wifi7 device All wifi7 devices need CONFIG_IWLMLD to be enabled. If we can't support the wifi7 device and the module is not enabled, complain to the user. The check in iwl_req_fw_callback is then no longer required. Signed-off-by: Emmanuel Grumbach Signed-off-by: Miri Korenblit Link: https://patch.msgid.link/20251019114304.071dae9a5de2.I1603085bac5a796442faa75982f8675647becfec@changeid --- drivers/net/wireless/intel/iwlwifi/iwl-drv.c | 9 +-------- drivers/net/wireless/intel/iwlwifi/iwl-drv.h | 6 ++++++ drivers/net/wireless/intel/iwlwifi/pcie/gen1_2/trans.c | 9 +++++++++ 3 files changed, 16 insertions(+), 8 deletions(-) diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-drv.c b/drivers/net/wireless/intel/iwlwifi/iwl-drv.c index 7d58e294618d..94ef422dc56a 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-drv.c +++ b/drivers/net/wireless/intel/iwlwifi/iwl-drv.c @@ -177,7 +177,7 @@ static inline char iwl_drv_get_step(int step) return 'a' + step; } -static bool iwl_drv_is_wifi7_supported(struct iwl_trans *trans) +bool iwl_drv_is_wifi7_supported(struct iwl_trans *trans) { return trans->mac_cfg->device_family >= IWL_DEVICE_FAMILY_BZ && CSR_HW_RFID_TYPE(trans->info.hw_rf_id) >= IWL_CFG_RF_TYPE_FM; @@ -1859,13 +1859,6 @@ static void iwl_req_fw_callback(const struct firmware *ucode_raw, void *context) #if IS_ENABLED(CONFIG_IWLMLD) if (iwl_drv_is_wifi7_supported(drv->trans)) op = &iwlwifi_opmode_table[MLD_OP_MODE]; -#else - if (iwl_drv_is_wifi7_supported(drv->trans)) { - IWL_ERR(drv, - "IWLMLD needs to be compiled to support this firmware\n"); - mutex_unlock(&iwlwifi_opmode_table_mtx); - goto out_unbind; - } #endif IWL_INFO(drv, "loaded firmware version %s op_mode %s\n", diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-drv.h b/drivers/net/wireless/intel/iwlwifi/iwl-drv.h index a0b67e8aba8d..6e60953de2ec 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-drv.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-drv.h @@ -78,6 +78,12 @@ struct iwl_drv *iwl_drv_start(struct iwl_trans *trans); */ void iwl_drv_stop(struct iwl_drv *drv); +/* + * iwl_drv_is_wifi7_supported - returns if wifi7 is supported + * If yes, iwlmld needs to be used to drive the device. + */ +bool iwl_drv_is_wifi7_supported(struct iwl_trans *trans); + /* * exported symbol management * diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/gen1_2/trans.c b/drivers/net/wireless/intel/iwlwifi/pcie/gen1_2/trans.c index 59307b5df441..164d060ec617 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/gen1_2/trans.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/gen1_2/trans.c @@ -4218,6 +4218,15 @@ int iwl_pci_gen1_2_probe(struct pci_dev *pdev, pdev->device, pdev->subsystem_device, info.hw_rev, info.hw_rf_id); +#if !IS_ENABLED(CONFIG_IWLMLD) + if (iwl_drv_is_wifi7_supported(iwl_trans)) { + IWL_ERR(iwl_trans, + "IWLMLD needs to be compiled to support this device\n"); + ret = -EOPNOTSUPP; + goto out_free_trans; + } +#endif + dev_info = iwl_pci_find_dev_info(pdev->device, pdev->subsystem_device, CSR_HW_RFID_TYPE(info.hw_rf_id), CSR_HW_RFID_IS_CDB(info.hw_rf_id), From d24076e0758ff8aa788e181c565683abae8bb397 Mon Sep 17 00:00:00 2001 From: Emmanuel Grumbach Date: Sun, 19 Oct 2025 11:45:06 +0300 Subject: [PATCH 0318/1381] wifi: iwlwifi: stop checking the firmware's error pointer It is not very clear what values we put in min_umac_error_event_table. For Ma (Meteor Lake), this value is wrong and we get the print: Not valid error log pointer ... Just remove the check. Signed-off-by: Emmanuel Grumbach Signed-off-by: Miri Korenblit Link: https://patch.msgid.link/20251019114304.a64d0803150f.Ie2db385f68e17fb0adcdcb16e5bf0125289e177d@changeid --- drivers/net/wireless/intel/iwlwifi/cfg/22000.c | 1 - drivers/net/wireless/intel/iwlwifi/cfg/8000.c | 1 - drivers/net/wireless/intel/iwlwifi/cfg/9000.c | 1 - drivers/net/wireless/intel/iwlwifi/cfg/ax210.c | 1 - drivers/net/wireless/intel/iwlwifi/cfg/bz.c | 1 - drivers/net/wireless/intel/iwlwifi/cfg/dr.c | 1 - drivers/net/wireless/intel/iwlwifi/cfg/sc.c | 1 - drivers/net/wireless/intel/iwlwifi/iwl-config.h | 2 -- drivers/net/wireless/intel/iwlwifi/mld/fw.c | 6 +----- drivers/net/wireless/intel/iwlwifi/mvm/fw.c | 13 ++----------- 10 files changed, 3 insertions(+), 25 deletions(-) diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/22000.c b/drivers/net/wireless/intel/iwlwifi/cfg/22000.c index ca488931a33c..f0453f3f6ba6 100644 --- a/drivers/net/wireless/intel/iwlwifi/cfg/22000.c +++ b/drivers/net/wireless/intel/iwlwifi/cfg/22000.c @@ -38,7 +38,6 @@ static const struct iwl_family_base_params iwl_22000_base = { .features = IWL_TX_CSUM_NETIF_FLAGS | NETIF_F_RXCSUM, .apmg_not_supported = true, .mac_addr_from_csr = 0x380, - .min_umac_error_event_table = 0x400000, .d3_debug_data_base_addr = 0x401000, .d3_debug_data_length = 60 * 1024, .mon_smem_regs = { diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/8000.c b/drivers/net/wireless/intel/iwlwifi/cfg/8000.c index b56574006ee0..3c844cd419e8 100644 --- a/drivers/net/wireless/intel/iwlwifi/cfg/8000.c +++ b/drivers/net/wireless/intel/iwlwifi/cfg/8000.c @@ -50,7 +50,6 @@ static const struct iwl_family_base_params iwl8000_base = { .smem_offset = IWL8260_SMEM_OFFSET, .smem_len = IWL8260_SMEM_LEN, .apmg_not_supported = true, - .min_umac_error_event_table = 0x800000, }; static const struct iwl_tt_params iwl8000_tt_params = { diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/9000.c b/drivers/net/wireless/intel/iwlwifi/cfg/9000.c index ac1fa291cf2f..5872fc9b8caf 100644 --- a/drivers/net/wireless/intel/iwlwifi/cfg/9000.c +++ b/drivers/net/wireless/intel/iwlwifi/cfg/9000.c @@ -41,7 +41,6 @@ static const struct iwl_family_base_params iwl9000_base = { .features = IWL_TX_CSUM_NETIF_FLAGS | NETIF_F_RXCSUM, .apmg_not_supported = true, .mac_addr_from_csr = 0x380, - .min_umac_error_event_table = 0x800000, .d3_debug_data_base_addr = 0x401000, .d3_debug_data_length = 92 * 1024, .nvm_hw_section_num = 10, diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/ax210.c b/drivers/net/wireless/intel/iwlwifi/cfg/ax210.c index ddf3d313da5a..582f61661062 100644 --- a/drivers/net/wireless/intel/iwlwifi/cfg/ax210.c +++ b/drivers/net/wireless/intel/iwlwifi/cfg/ax210.c @@ -33,7 +33,6 @@ static const struct iwl_family_base_params iwl_ax210_base = { .features = IWL_TX_CSUM_NETIF_FLAGS | NETIF_F_RXCSUM, .apmg_not_supported = true, .mac_addr_from_csr = 0x380, - .min_umac_error_event_table = 0x400000, .d3_debug_data_base_addr = 0x401000, .d3_debug_data_length = 60 * 1024, .mon_smem_regs = { diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/bz.c b/drivers/net/wireless/intel/iwlwifi/cfg/bz.c index 18c9244ee8ef..7babb60463be 100644 --- a/drivers/net/wireless/intel/iwlwifi/cfg/bz.c +++ b/drivers/net/wireless/intel/iwlwifi/cfg/bz.c @@ -38,7 +38,6 @@ static const struct iwl_family_base_params iwl_bz_base = { .smem_len = IWL_BZ_SMEM_LEN, .apmg_not_supported = true, .mac_addr_from_csr = 0x30, - .min_umac_error_event_table = 0xD0000, .d3_debug_data_base_addr = 0x401000, .d3_debug_data_length = 60 * 1024, .mon_smem_regs = { diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/dr.c b/drivers/net/wireless/intel/iwlwifi/cfg/dr.c index e53a785686c8..ad65951d5643 100644 --- a/drivers/net/wireless/intel/iwlwifi/cfg/dr.c +++ b/drivers/net/wireless/intel/iwlwifi/cfg/dr.c @@ -33,7 +33,6 @@ static const struct iwl_family_base_params iwl_dr_base = { .smem_len = IWL_DR_SMEM_LEN, .apmg_not_supported = true, .mac_addr_from_csr = 0x30, - .min_umac_error_event_table = 0xD0000, .d3_debug_data_base_addr = 0x401000, .d3_debug_data_length = 60 * 1024, .mon_smem_regs = { diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/sc.c b/drivers/net/wireless/intel/iwlwifi/cfg/sc.c index e9449b59114a..abb6283dfb8f 100644 --- a/drivers/net/wireless/intel/iwlwifi/cfg/sc.c +++ b/drivers/net/wireless/intel/iwlwifi/cfg/sc.c @@ -41,7 +41,6 @@ static const struct iwl_family_base_params iwl_sc_base = { .smem_len = IWL_SC_SMEM_LEN, .apmg_not_supported = true, .mac_addr_from_csr = 0x30, - .min_umac_error_event_table = 0xD0000, .d3_debug_data_base_addr = 0x401000, .d3_debug_data_length = 60 * 1024, .mon_smem_regs = { diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-config.h b/drivers/net/wireless/intel/iwlwifi/iwl-config.h index 0b34c9f90b3f..e44936204068 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-config.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-config.h @@ -170,7 +170,6 @@ struct iwl_fw_mon_regs { * for aggregation * @min_txq_size: minimum number of slots required in a TX queue * @gp2_reg_addr: GP2 (timer) register address - * @min_umac_error_event_table: minimum SMEM location of UMAC error table * @mon_dbgi_regs: monitor DBGI registers * @mon_dram_regs: monitor DRAM registers * @mon_smem_regs: monitor SMEM registers @@ -203,7 +202,6 @@ struct iwl_family_base_params { netdev_features_t features; u32 smem_offset; u32 smem_len; - u32 min_umac_error_event_table; u32 d3_debug_data_base_addr; u32 d3_debug_data_length; u32 min_txq_size; diff --git a/drivers/net/wireless/intel/iwlwifi/mld/fw.c b/drivers/net/wireless/intel/iwlwifi/mld/fw.c index b3abfa1ec810..19da521a4bab 100644 --- a/drivers/net/wireless/intel/iwlwifi/mld/fw.c +++ b/drivers/net/wireless/intel/iwlwifi/mld/fw.c @@ -167,11 +167,7 @@ static bool iwl_alive_fn(struct iwl_notif_wait_data *notif_wait, umac_error_table = le32_to_cpu(umac->dbg_ptrs.error_info_addr) & ~FW_ADDR_CACHE_CONTROL; - if (umac_error_table >= trans->mac_cfg->base->min_umac_error_event_table) - iwl_fw_umac_set_alive_err_table(trans, umac_error_table); - else - IWL_ERR(mld, "Not valid error log pointer 0x%08X\n", - umac_error_table); + iwl_fw_umac_set_alive_err_table(trans, umac_error_table); alive_data->valid = status == IWL_ALIVE_STATUS_OK; diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c index 6b76ce35443d..edae13755ee6 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c @@ -214,17 +214,8 @@ static bool iwl_alive_fn(struct iwl_notif_wait_data *notif_wait, ~FW_ADDR_CACHE_CONTROL; if (umac_error_table) { - if (umac_error_table >= - mvm->trans->mac_cfg->base->min_umac_error_event_table) { - iwl_fw_umac_set_alive_err_table(mvm->trans, - umac_error_table); - } else { - IWL_ERR(mvm, - "Not valid error log pointer 0x%08X for %s uCode\n", - umac_error_table, - (mvm->fwrt.cur_fw_img == IWL_UCODE_INIT) ? - "Init" : "RT"); - } + iwl_fw_umac_set_alive_err_table(mvm->trans, + umac_error_table); } alive_data->valid = status == IWL_ALIVE_STATUS_OK; From 7906c61a8faf6660bc2d5c6e71985be695a71065 Mon Sep 17 00:00:00 2001 From: Miri Korenblit Date: Sun, 19 Oct 2025 11:45:07 +0300 Subject: [PATCH 0319/1381] wifi: iwlwifi: mld: check the validity of noa_len Validate iwl_probe_resp_data_notif::noa_attr::len_low since we are using its value to determine the noa_len, which is later used for the NoA attribute. Reviewed-by: Emmanuel Grumbach Signed-off-by: Miri Korenblit Link: https://patch.msgid.link/20251019114304.b127a2b57e8c.I7ccaf118d236fb39da5da351b95ad9b37b825bc2@changeid --- drivers/net/wireless/intel/iwlwifi/mld/iface.c | 13 +++++++++++++ 1 file changed, 13 insertions(+) diff --git a/drivers/net/wireless/intel/iwlwifi/mld/iface.c b/drivers/net/wireless/intel/iwlwifi/mld/iface.c index ed379825a923..a5ececfc13e4 100644 --- a/drivers/net/wireless/intel/iwlwifi/mld/iface.c +++ b/drivers/net/wireless/intel/iwlwifi/mld/iface.c @@ -528,6 +528,19 @@ void iwl_mld_handle_probe_resp_data_notif(struct iwl_mld *mld, mld_link = &iwl_mld_vif_from_mac80211(vif)->deflink; + /* len_low should be 2 + n*13 (where n is the number of descriptors. + * 13 is the size of a NoA descriptor). We can have either one or two + * descriptors. + */ + if (IWL_FW_CHECK(mld, notif->noa_active && + notif->noa_attr.len_low != 2 + + sizeof(struct ieee80211_p2p_noa_desc) && + notif->noa_attr.len_low != 2 + + sizeof(struct ieee80211_p2p_noa_desc) * 2, + "Invalid noa_attr.len_low (%d)\n", + notif->noa_attr.len_low)) + return; + new_data = kzalloc(sizeof(*new_data), GFP_KERNEL); if (!new_data) return; From f67cf9aaae4681692dce78ce31ba08cdf9cc0705 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Sun, 19 Oct 2025 11:45:08 +0300 Subject: [PATCH 0320/1381] wifi: iwlwifi: fix build when mvm/mld not configured When neither mvm nor mld are configured, we don't have the iwl_bz_mac_cfg symbol and thus cannot check for it. But in that case the relevant device entries aren't and cannot be present, so just ifdef the test code for that. Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-kbuild-all/202509170625.BAJBe7Bi-lkp@intel.com/ Signed-off-by: Johannes Berg Signed-off-by: Miri Korenblit Link: https://patch.msgid.link/20251019114304.615810979e7b.I9a215f955bb3208d99239be8496d19e0f186b4d0@changeid --- drivers/net/wireless/intel/iwlwifi/tests/devinfo.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/wireless/intel/iwlwifi/tests/devinfo.c b/drivers/net/wireless/intel/iwlwifi/tests/devinfo.c index 3054bc0a9c67..6bf2ad18b009 100644 --- a/drivers/net/wireless/intel/iwlwifi/tests/devinfo.c +++ b/drivers/net/wireless/intel/iwlwifi/tests/devinfo.c @@ -274,6 +274,7 @@ static void devinfo_pci_ids_config(struct kunit *test) if (s->device == PCI_ANY_ID || s->subdevice == PCI_ANY_ID) continue; +#if IS_ENABLED(CONFIG_IWLMVM) || IS_ENABLED(CONFIG_IWLMLD) /* * The check below only works for old (pre-CNVI) devices. Most * new have subdevice==ANY, so are already skipped, but for some @@ -281,6 +282,7 @@ static void devinfo_pci_ids_config(struct kunit *test) */ if (s->driver_data == (kernel_ulong_t)&iwl_bz_mac_cfg) continue; +#endif di = iwl_pci_find_dev_info(s->device, s->subdevice, 0, 0, 0, 0, true); From 50d0cafec60ce2c6ece236806df7fbef0cc8a0da Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Sun, 19 Oct 2025 11:45:09 +0300 Subject: [PATCH 0321/1381] wifi: iwlwifi: bump core version for BZ/SC/DR Start supporting Core 100 FW on those devices. In addition, the move to the new Core scheme (instead of API scheme) will start Core 100 and not 99, as planned. Adjust for that as well. Signed-off-by: Johannes Berg Signed-off-by: Miri Korenblit Link: https://patch.msgid.link/20251019114304.00c28b4259da.Idd6281cd647f1b33f2572a8c107c3a3228e03665@changeid --- drivers/net/wireless/intel/iwlwifi/cfg/bz.c | 2 +- drivers/net/wireless/intel/iwlwifi/cfg/dr.c | 2 +- drivers/net/wireless/intel/iwlwifi/cfg/sc.c | 2 +- drivers/net/wireless/intel/iwlwifi/iwl-drv.c | 4 ++-- 4 files changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/bz.c b/drivers/net/wireless/intel/iwlwifi/cfg/bz.c index 7babb60463be..28005f25c38b 100644 --- a/drivers/net/wireless/intel/iwlwifi/cfg/bz.c +++ b/drivers/net/wireless/intel/iwlwifi/cfg/bz.c @@ -10,7 +10,7 @@ #include "fw/api/txq.h" /* Highest firmware core release supported */ -#define IWL_BZ_UCODE_CORE_MAX 99 +#define IWL_BZ_UCODE_CORE_MAX 100 /* Lowest firmware API version supported */ #define IWL_BZ_UCODE_API_MIN 100 diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/dr.c b/drivers/net/wireless/intel/iwlwifi/cfg/dr.c index ad65951d5643..c1c6174534d7 100644 --- a/drivers/net/wireless/intel/iwlwifi/cfg/dr.c +++ b/drivers/net/wireless/intel/iwlwifi/cfg/dr.c @@ -9,7 +9,7 @@ #include "fw/api/txq.h" /* Highest firmware core release supported */ -#define IWL_DR_UCODE_CORE_MAX 99 +#define IWL_DR_UCODE_CORE_MAX 100 /* Lowest firmware API version supported */ #define IWL_DR_UCODE_API_MIN 100 diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/sc.c b/drivers/net/wireless/intel/iwlwifi/cfg/sc.c index abb6283dfb8f..d9943d252ae0 100644 --- a/drivers/net/wireless/intel/iwlwifi/cfg/sc.c +++ b/drivers/net/wireless/intel/iwlwifi/cfg/sc.c @@ -10,7 +10,7 @@ #include "fw/api/txq.h" /* Highest firmware core release supported */ -#define IWL_SC_UCODE_CORE_MAX 99 +#define IWL_SC_UCODE_CORE_MAX 100 /* Lowest firmware API version supported */ #define IWL_SC_UCODE_API_MIN 100 diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-drv.c b/drivers/net/wireless/intel/iwlwifi/iwl-drv.c index 94ef422dc56a..3391f07b01de 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-drv.c +++ b/drivers/net/wireless/intel/iwlwifi/iwl-drv.c @@ -348,8 +348,8 @@ static int iwl_request_firmware(struct iwl_drv *drv, bool first) if (first) drv->fw_index = ucode_api_max; - else if (drv->fw_index == ENCODE_CORE_AS_API(99)) - drv->fw_index = 101; /* last API-scheme number below core 99 */ + else if (drv->fw_index == ENCODE_CORE_AS_API(100)) + drv->fw_index = 102; /* last API-scheme number below core 100 */ else drv->fw_index--; From 3e24ba621ba122ff765c8c9790800a05196fbbba Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Sun, 19 Oct 2025 11:45:10 +0300 Subject: [PATCH 0322/1381] wifi: iwlwifi: mvm/mld: report non-HT frames as 20 MHz Non-HT frames can only be encoded in 20 MHz, however, they could be duplicated on all/some of the subchannels (mostly used for RTS/CTS), in which case the firmware will report and estimate of the overall used bandwidth based on energy detected. This could be confusing so don't report it that way, always use 20 MHz for non-HT/legacy frames instead. Note that currently the value doesn't appear to be used by mac80211, it never checks the bandwidth field for legacy encodings. Signed-off-by: Johannes Berg Reviewed-by: Benjamin Berg Signed-off-by: Miri Korenblit Link: https://patch.msgid.link/20251019114304.23e6695039ca.I3da7c542bde6de4362755f200248dbcc12aa246e@changeid --- drivers/net/wireless/intel/iwlwifi/mld/rx.c | 3 +++ drivers/net/wireless/intel/iwlwifi/mvm/rx.c | 2 ++ 2 files changed, 5 insertions(+) diff --git a/drivers/net/wireless/intel/iwlwifi/mld/rx.c b/drivers/net/wireless/intel/iwlwifi/mld/rx.c index 19e78d2fbf9b..6a76e3fcb581 100644 --- a/drivers/net/wireless/intel/iwlwifi/mld/rx.c +++ b/drivers/net/wireless/intel/iwlwifi/mld/rx.c @@ -1309,6 +1309,9 @@ static void iwl_mld_set_rx_rate(struct iwl_mld *mld, iwl_mld_legacy_hw_idx_to_mac80211_idx(rate_n_flags, rx_status->band); + /* override BW - it could be DUP and indicate the wrong BW */ + rx_status->bw = RATE_INFO_BW_20; + /* valid rate */ if (rate >= 0 && rate <= 0xFF) { rx_status->rate_idx = rate; diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rx.c b/drivers/net/wireless/intel/iwlwifi/mvm/rx.c index 8c1bb3a7ffca..d0c0faae0122 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/rx.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/rx.c @@ -519,6 +519,8 @@ void iwl_mvm_rx_rx_mpdu(struct iwl_mvm *mvm, struct napi_struct *napi, return; } rx_status->rate_idx = rate; + /* override BW - it could be DUP and indicate the wrong BW */ + rx_status->bw = RATE_INFO_BW_20; } #ifdef CONFIG_IWLWIFI_DEBUGFS From bd8a6e46e6b89f758669614d2aa1219e8ea048ef Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Sun, 19 Oct 2025 11:45:11 +0300 Subject: [PATCH 0323/1381] wifi: iwlwifi: mld: use FW_CHECK on bad ROC notification If the firmware sends a ROC notification after the driver cancelled it, we can get into this WARN_ON(). Don't do that, use IWL_FW_CHECK() instead. Signed-off-by: Johannes Berg Tested-by: Emmanuel Grumbach Reviewed-by: Emmanuel Grumbach Signed-off-by: Miri Korenblit Link: https://patch.msgid.link/20251019114304.48aff2c8922e.Ie27b21eb26b67c8010d13ce9590751cad417d1ad@changeid --- drivers/net/wireless/intel/iwlwifi/mld/roc.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/intel/iwlwifi/mld/roc.c b/drivers/net/wireless/intel/iwlwifi/mld/roc.c index 4136c98030d0..4e37a288471e 100644 --- a/drivers/net/wireless/intel/iwlwifi/mld/roc.c +++ b/drivers/net/wireless/intel/iwlwifi/mld/roc.c @@ -231,7 +231,9 @@ void iwl_mld_handle_roc_notif(struct iwl_mld *mld, struct ieee80211_vif *vif; vif = iwl_mld_find_roc_vif(mld, activity); - if (WARN_ON(!vif)) + if (IWL_FW_CHECK(mld, !vif, + "unexpected ROC notif from FW for activity %d\n", + activity)) return; mld_vif = iwl_mld_vif_from_mac80211(vif); From 0d0e8149c6d77fe22dfb10b985ebedfbc192a2cd Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Sun, 19 Oct 2025 11:45:12 +0300 Subject: [PATCH 0324/1381] wifi: iwlwifi: bump core version for BZ/SC/DR Start supporting Core 101 FW on these devices. Signed-off-by: Johannes Berg Signed-off-by: Miri Korenblit Link: https://patch.msgid.link/20251019114304.23300d52cd8b.I6aad50aed469d7734c165086796dfa9cdf9d81bd@changeid --- drivers/net/wireless/intel/iwlwifi/cfg/bz.c | 2 +- drivers/net/wireless/intel/iwlwifi/cfg/dr.c | 2 +- drivers/net/wireless/intel/iwlwifi/cfg/sc.c | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/bz.c b/drivers/net/wireless/intel/iwlwifi/cfg/bz.c index 28005f25c38b..d25445bd1e5c 100644 --- a/drivers/net/wireless/intel/iwlwifi/cfg/bz.c +++ b/drivers/net/wireless/intel/iwlwifi/cfg/bz.c @@ -10,7 +10,7 @@ #include "fw/api/txq.h" /* Highest firmware core release supported */ -#define IWL_BZ_UCODE_CORE_MAX 100 +#define IWL_BZ_UCODE_CORE_MAX 101 /* Lowest firmware API version supported */ #define IWL_BZ_UCODE_API_MIN 100 diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/dr.c b/drivers/net/wireless/intel/iwlwifi/cfg/dr.c index c1c6174534d7..a279dcfd3083 100644 --- a/drivers/net/wireless/intel/iwlwifi/cfg/dr.c +++ b/drivers/net/wireless/intel/iwlwifi/cfg/dr.c @@ -9,7 +9,7 @@ #include "fw/api/txq.h" /* Highest firmware core release supported */ -#define IWL_DR_UCODE_CORE_MAX 100 +#define IWL_DR_UCODE_CORE_MAX 101 /* Lowest firmware API version supported */ #define IWL_DR_UCODE_API_MIN 100 diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/sc.c b/drivers/net/wireless/intel/iwlwifi/cfg/sc.c index d9943d252ae0..ee00b2af7a1d 100644 --- a/drivers/net/wireless/intel/iwlwifi/cfg/sc.c +++ b/drivers/net/wireless/intel/iwlwifi/cfg/sc.c @@ -10,7 +10,7 @@ #include "fw/api/txq.h" /* Highest firmware core release supported */ -#define IWL_SC_UCODE_CORE_MAX 100 +#define IWL_SC_UCODE_CORE_MAX 101 /* Lowest firmware API version supported */ #define IWL_SC_UCODE_API_MIN 100 From 7ed47d42943fba8ced505f62d4358f63963bb968 Mon Sep 17 00:00:00 2001 From: Emmanuel Grumbach Date: Sun, 19 Oct 2025 11:45:13 +0300 Subject: [PATCH 0325/1381] wifi: iwlwifi: disable EHT if the device doesn't allow it We have a few devices that don't allow EHT. Make sure we reflect this towards mac80211 so that we won't try to enable it. Signed-off-by: Emmanuel Grumbach Signed-off-by: Miri Korenblit Link: https://patch.msgid.link/20251019114304.71121f4e5557.I49e2329d4121f9e52d0889156d0c3e8778e27d88@changeid --- .../net/wireless/intel/iwlwifi/cfg/rf-fm.c | 1 + .../net/wireless/intel/iwlwifi/cfg/rf-wh.c | 23 +++++++++++++++++++ .../net/wireless/intel/iwlwifi/iwl-config.h | 5 +++- .../wireless/intel/iwlwifi/iwl-nvm-parse.c | 2 +- drivers/net/wireless/intel/iwlwifi/pcie/drv.c | 6 +++-- 5 files changed, 33 insertions(+), 4 deletions(-) diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/rf-fm.c b/drivers/net/wireless/intel/iwlwifi/cfg/rf-fm.c index 456a666c8dfd..fd82050e33a3 100644 --- a/drivers/net/wireless/intel/iwlwifi/cfg/rf-fm.c +++ b/drivers/net/wireless/intel/iwlwifi/cfg/rf-fm.c @@ -19,6 +19,7 @@ .non_shared_ant = ANT_B, \ .vht_mu_mimo_supported = true, \ .uhb_supported = true, \ + .eht_supported = true, \ .num_rbds = IWL_NUM_RBDS_EHT, \ .nvm_ver = IWL_FM_NVM_VERSION, \ .nvm_type = IWL_NVM_EXT diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/rf-wh.c b/drivers/net/wireless/intel/iwlwifi/cfg/rf-wh.c index b8c6b06e7099..b5803ea1eb78 100644 --- a/drivers/net/wireless/intel/iwlwifi/cfg/rf-wh.c +++ b/drivers/net/wireless/intel/iwlwifi/cfg/rf-wh.c @@ -4,8 +4,31 @@ */ #include "iwl-config.h" +/* NVM versions */ +#define IWL_WH_NVM_VERSION 0x0a1d + +#define IWL_DEVICE_WH \ + .ht_params = { \ + .stbc = true, \ + .ldpc = true, \ + .ht40_bands = BIT(NL80211_BAND_2GHZ) | \ + BIT(NL80211_BAND_5GHZ), \ + }, \ + .led_mode = IWL_LED_RF_STATE, \ + .non_shared_ant = ANT_B, \ + .vht_mu_mimo_supported = true, \ + .uhb_supported = true, \ + .num_rbds = IWL_NUM_RBDS_EHT, \ + .nvm_ver = IWL_WH_NVM_VERSION, \ + .nvm_type = IWL_NVM_EXT + /* currently iwl_rf_wh/iwl_rf_wh_160mhz are just defines for the FM ones */ +const struct iwl_rf_cfg iwl_rf_wh_non_eht = { + IWL_DEVICE_WH, + .eht_supported = false, +}; + const char iwl_killer_be1775s_name[] = "Killer(R) Wi-Fi 7 BE1775s 320MHz Wireless Network Adapter (BE211D2W)"; const char iwl_killer_be1775i_name[] = diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-config.h b/drivers/net/wireless/intel/iwlwifi/iwl-config.h index e44936204068..0a5a683325ef 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-config.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-config.h @@ -416,6 +416,7 @@ struct iwl_mac_cfg { * @vht_mu_mimo_supported: VHT MU-MIMO support * @nvm_type: see &enum iwl_nvm_type * @uhb_supported: ultra high band channels supported + * @eht_supported: EHT supported * @num_rbds: number of receive buffer descriptors to use * (only used for multi-queue capable devices) * @@ -448,7 +449,8 @@ struct iwl_rf_cfg { host_interrupt_operation_mode:1, lp_xtal_workaround:1, vht_mu_mimo_supported:1, - uhb_supported:1; + uhb_supported:1, + eht_supported:1; u8 valid_tx_ant; u8 valid_rx_ant; u8 non_shared_ant; @@ -742,6 +744,7 @@ extern const struct iwl_rf_cfg iwl_rf_fm; extern const struct iwl_rf_cfg iwl_rf_fm_160mhz; #define iwl_rf_wh iwl_rf_fm #define iwl_rf_wh_160mhz iwl_rf_fm_160mhz +extern const struct iwl_rf_cfg iwl_rf_wh_non_eht; #define iwl_rf_pe iwl_rf_fm #endif /* CONFIG_IWLMLD */ diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c index 23465e4c4b39..e021fc57d85d 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c +++ b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c @@ -2080,7 +2080,7 @@ struct iwl_nvm_data *iwl_get_nvm(struct iwl_trans *trans, !!(mac_flags & NVM_MAC_SKU_FLAGS_BAND_5_2_ENABLED); nvm->sku_cap_mimo_disabled = !!(mac_flags & NVM_MAC_SKU_FLAGS_MIMO_DISABLED); - if (CSR_HW_RFID_TYPE(trans->info.hw_rf_id) >= IWL_CFG_RF_TYPE_FM) + if (trans->cfg->eht_supported) nvm->sku_cap_11be_enable = true; /* Initialize PHY sku data */ diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c index de04a84def0d..73001cdce13a 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c @@ -1061,8 +1061,10 @@ VISIBLE_IF_IWLWIFI_KUNIT const struct iwl_dev_info iwl_dev_info_table[] = { /* WH RF */ IWL_DEV_INFO(iwl_rf_wh, iwl_be211_name, RF_TYPE(WH)), - IWL_DEV_INFO(iwl_rf_wh, iwl_ax221_name, RF_TYPE(WH), SUBDEV(0x0514)), - IWL_DEV_INFO(iwl_rf_wh, iwl_ax221_name, RF_TYPE(WH), SUBDEV(0x4514)), + IWL_DEV_INFO(iwl_rf_wh_non_eht, iwl_ax221_name, RF_TYPE(WH), + SUBDEV(0x0514)), + IWL_DEV_INFO(iwl_rf_wh_non_eht, iwl_ax221_name, RF_TYPE(WH), + SUBDEV(0x4514)), IWL_DEV_INFO(iwl_rf_wh_160mhz, iwl_be213_name, RF_TYPE(WH), BW_LIMITED), /* PE RF */ From 9e69bcb527ea66288d49b97a85e33ce429825079 Mon Sep 17 00:00:00 2001 From: Nidhish A N Date: Sun, 19 Oct 2025 11:45:14 +0300 Subject: [PATCH 0326/1381] wifi: iwlwifi: mld: Move EMLSR prints to IWL_DL_EHT Modify EMLSR debug prints to use IWL_DL_EHT instead of IWL_DL_INFO. This will allow better communication with validation as they might enable only IWL_DL_EHT or IWL_DL_INFO as required. Add prints to log attempt to switch links when missed beacons exceed threshold. Print both link ids and missed beacons when in EMLSR mode. Signed-off-by: Nidhish A N Signed-off-by: Miri Korenblit Link: https://patch.msgid.link/20251019114304.3bfc2bc8f410.I405ab2aa81af1ba0ea5eaff343eae1778f2035d9@changeid --- drivers/net/wireless/intel/iwlwifi/mld/link.c | 9 +- drivers/net/wireless/intel/iwlwifi/mld/mlo.c | 97 +++++++++---------- drivers/net/wireless/intel/iwlwifi/mld/sta.c | 2 +- 3 files changed, 52 insertions(+), 56 deletions(-) diff --git a/drivers/net/wireless/intel/iwlwifi/mld/link.c b/drivers/net/wireless/intel/iwlwifi/mld/link.c index 738f80fe0c50..deb1e7227dd8 100644 --- a/drivers/net/wireless/intel/iwlwifi/mld/link.c +++ b/drivers/net/wireless/intel/iwlwifi/mld/link.c @@ -571,8 +571,12 @@ void iwl_mld_handle_missed_beacon_notif(struct iwl_mld *mld, /* Not in EMLSR and we can't hear the link. * Try to switch to a better link. EMLSR case is handled below. */ - if (!iwl_mld_emlsr_active(vif)) + if (!iwl_mld_emlsr_active(vif)) { + IWL_DEBUG_EHT(mld, + "missed beacons exceeds threshold. link_id=%u. Try to switch to a better link.\n", + link_id); iwl_mld_int_mlo_scan(mld, vif); + } } /* no more logic if we're not in EMLSR */ @@ -591,7 +595,8 @@ void iwl_mld_handle_missed_beacon_notif(struct iwl_mld *mld, return; IWL_DEBUG_EHT(mld, - "missed bcn on the other link (link_id=%u): %u\n", + "missed bcn link_id=%u: %u consecutive=%u, other link_id=%u: %u\n", + link_id, missed_bcon, missed_bcon_since_rx, other_link->link_id, scnd_lnk_bcn_lost); /* Exit EMLSR if we lost more than diff --git a/drivers/net/wireless/intel/iwlwifi/mld/mlo.c b/drivers/net/wireless/intel/iwlwifi/mld/mlo.c index fa04fbe06656..c6b151f26921 100644 --- a/drivers/net/wireless/intel/iwlwifi/mld/mlo.c +++ b/drivers/net/wireless/intel/iwlwifi/mld/mlo.c @@ -31,11 +31,9 @@ static void iwl_mld_print_emlsr_blocked(struct iwl_mld *mld, u32 mask) { #define NAME_FMT(x) "%s" #define NAME_PR(x) (mask & IWL_MLD_EMLSR_BLOCKED_##x) ? "[" #x "]" : "", - IWL_DEBUG_INFO(mld, - "EMLSR blocked = " HANDLE_EMLSR_BLOCKED_REASONS(NAME_FMT) - " (0x%x)\n", - HANDLE_EMLSR_BLOCKED_REASONS(NAME_PR) - mask); + IWL_DEBUG_EHT(mld, + "EMLSR blocked = " HANDLE_EMLSR_BLOCKED_REASONS(NAME_FMT) + " (0x%x)\n", HANDLE_EMLSR_BLOCKED_REASONS(NAME_PR) mask); #undef NAME_FMT #undef NAME_PR } @@ -72,11 +70,9 @@ static void iwl_mld_print_emlsr_exit(struct iwl_mld *mld, u32 mask) { #define NAME_FMT(x) "%s" #define NAME_PR(x) (mask & IWL_MLD_EMLSR_EXIT_##x) ? "[" #x "]" : "", - IWL_DEBUG_INFO(mld, - "EMLSR exit = " HANDLE_EMLSR_EXIT_REASONS(NAME_FMT) - " (0x%x)\n", - HANDLE_EMLSR_EXIT_REASONS(NAME_PR) - mask); + IWL_DEBUG_EHT(mld, + "EMLSR exit = " HANDLE_EMLSR_EXIT_REASONS(NAME_FMT) + " (0x%x)\n", HANDLE_EMLSR_EXIT_REASONS(NAME_PR) mask); #undef NAME_FMT #undef NAME_PR } @@ -170,10 +166,10 @@ static void iwl_mld_check_emlsr_prevention(struct iwl_mld *mld, WARN_ON(mld_vif->emlsr.exit_repeat_count > 3); } - IWL_DEBUG_INFO(mld, - "Preventing EMLSR for %ld seconds due to %u exits with the reason = %s (0x%x)\n", - delay / HZ, mld_vif->emlsr.exit_repeat_count, - iwl_mld_get_emlsr_exit_string(reason), reason); + IWL_DEBUG_EHT(mld, + "Preventing EMLSR for %ld seconds due to %u exits with the reason = %s (0x%x)\n", + delay / HZ, mld_vif->emlsr.exit_repeat_count, + iwl_mld_get_emlsr_exit_string(reason), reason); wiphy_delayed_work_queue(mld->wiphy, &mld_vif->emlsr.prevent_done_wk, delay); @@ -217,10 +213,10 @@ static int _iwl_mld_exit_emlsr(struct iwl_mld *mld, struct ieee80211_vif *vif, link_to_keep = __ffs(vif->active_links); new_active_links = BIT(link_to_keep); - IWL_DEBUG_INFO(mld, - "Exiting EMLSR. reason = %s (0x%x). Current active links=0x%x, new active links = 0x%x\n", - iwl_mld_get_emlsr_exit_string(exit), exit, - vif->active_links, new_active_links); + IWL_DEBUG_EHT(mld, + "Exiting EMLSR. reason = %s (0x%x). Current active links=0x%x, new active links = 0x%x\n", + iwl_mld_get_emlsr_exit_string(exit), exit, + vif->active_links, new_active_links); if (sync) ret = ieee80211_set_active_links(vif, new_active_links); @@ -262,9 +258,8 @@ static int _iwl_mld_emlsr_block(struct iwl_mld *mld, struct ieee80211_vif *vif, mld_vif->emlsr.blocked_reasons |= reason; - IWL_DEBUG_INFO(mld, - "Blocking EMLSR mode. reason = %s (0x%x)\n", - iwl_mld_get_emlsr_blocked_string(reason), reason); + IWL_DEBUG_EHT(mld, "Blocking EMLSR mode. reason = %s (0x%x)\n", + iwl_mld_get_emlsr_blocked_string(reason), reason); iwl_mld_print_emlsr_blocked(mld, mld_vif->emlsr.blocked_reasons); if (reason == IWL_MLD_EMLSR_BLOCKED_TPT) @@ -335,9 +330,8 @@ void iwl_mld_unblock_emlsr(struct iwl_mld *mld, struct ieee80211_vif *vif, mld_vif->emlsr.blocked_reasons &= ~reason; - IWL_DEBUG_INFO(mld, - "Unblocking EMLSR mode. reason = %s (0x%x)\n", - iwl_mld_get_emlsr_blocked_string(reason), reason); + IWL_DEBUG_EHT(mld, "Unblocking EMLSR mode. reason = %s (0x%x)\n", + iwl_mld_get_emlsr_blocked_string(reason), reason); iwl_mld_print_emlsr_blocked(mld, mld_vif->emlsr.blocked_reasons); if (reason == IWL_MLD_EMLSR_BLOCKED_TPT) @@ -348,7 +342,7 @@ void iwl_mld_unblock_emlsr(struct iwl_mld *mld, struct ieee80211_vif *vif, if (mld_vif->emlsr.blocked_reasons) return; - IWL_DEBUG_INFO(mld, "EMLSR is unblocked\n"); + IWL_DEBUG_EHT(mld, "EMLSR is unblocked\n"); iwl_mld_int_mlo_scan(mld, vif); } @@ -365,18 +359,17 @@ iwl_mld_vif_iter_emlsr_mode_notif(void *data, u8 *mac, switch (action) { case ESR_RECOMMEND_LEAVE: - IWL_DEBUG_INFO(mld_vif->mld, - "FW recommend leave reason = 0x%x\n", - le32_to_cpu(notif->leave_reason_mask)); + IWL_DEBUG_EHT(mld_vif->mld, + "FW recommend leave reason = 0x%x\n", + le32_to_cpu(notif->leave_reason_mask)); iwl_mld_exit_emlsr(mld_vif->mld, vif, IWL_MLD_EMLSR_EXIT_FW_REQUEST, iwl_mld_get_primary_link(vif)); break; case ESR_FORCE_LEAVE: - IWL_DEBUG_INFO(mld_vif->mld, - "FW force leave reason = 0x%x\n", - le32_to_cpu(notif->leave_reason_mask)); + IWL_DEBUG_EHT(mld_vif->mld, "FW force leave reason = 0x%x\n", + le32_to_cpu(notif->leave_reason_mask)); fallthrough; case ESR_RECOMMEND_ENTER: default: @@ -412,11 +405,12 @@ void iwl_mld_handle_emlsr_trans_fail_notif(struct iwl_mld *mld, struct ieee80211_bss_conf *bss_conf = iwl_mld_fw_id_to_link_conf(mld, fw_link_id); - IWL_DEBUG_INFO(mld, "Failed to %s EMLSR on link %d (FW: %d), reason %d\n", - le32_to_cpu(notif->activation) ? "enter" : "exit", - bss_conf ? bss_conf->link_id : -1, - le32_to_cpu(notif->link_id), - le32_to_cpu(notif->err_code)); + IWL_DEBUG_EHT(mld, + "Failed to %s EMLSR on link %d (FW: %d), reason %d\n", + le32_to_cpu(notif->activation) ? "enter" : "exit", + bss_conf ? bss_conf->link_id : -1, + le32_to_cpu(notif->link_id), + le32_to_cpu(notif->err_code)); if (IWL_FW_CHECK(mld, !bss_conf, "FW reported failure to %sactivate EMLSR on a non-existing link: %d\n", @@ -590,8 +584,8 @@ void iwl_mld_emlsr_check_tpt(struct wiphy *wiphy, struct wiphy_work *wk) spin_unlock_bh(&queue_counter->lock); } - IWL_DEBUG_INFO(mld, "total Tx MPDUs: %ld. total Rx MPDUs: %ld\n", - total_tx, total_rx); + IWL_DEBUG_EHT(mld, "total Tx MPDUs: %ld. total Rx MPDUs: %ld\n", + total_tx, total_rx); /* If we don't have enough MPDUs - exit EMLSR */ if (total_tx < IWL_MLD_ENTER_EMLSR_TPT_THRESH && @@ -605,8 +599,8 @@ void iwl_mld_emlsr_check_tpt(struct wiphy *wiphy, struct wiphy_work *wk) if (sec_link_id == -1) goto schedule; - IWL_DEBUG_INFO(mld, "Secondary Link %d: Tx MPDUs: %ld. Rx MPDUs: %ld\n", - sec_link_id, sec_link_tx, sec_link_rx); + IWL_DEBUG_EHT(mld, "Secondary Link %d: Tx MPDUs: %ld. Rx MPDUs: %ld\n", + sec_link_id, sec_link_tx, sec_link_rx); /* Calculate the percentage of the secondary link TX/RX */ sec_link_tx_perc = total_tx ? sec_link_tx * 100 / total_tx : 0; @@ -703,10 +697,8 @@ iwl_mld_emlsr_disallowed_with_link(struct iwl_mld *mld, ret |= IWL_MLD_EMLSR_EXIT_CSA; if (ret) { - IWL_DEBUG_INFO(mld, - "Link %d is not allowed for EMLSR as %s\n", - link->link_id, - primary ? "primary" : "secondary"); + IWL_DEBUG_EHT(mld, "Link %d is not allowed for EMLSR as %s\n", + link->link_id, primary ? "primary" : "secondary"); iwl_mld_print_emlsr_exit(mld, ret); } @@ -870,13 +862,12 @@ iwl_mld_emlsr_pair_state(struct ieee80211_vif *vif, reason_mask |= IWL_MLD_EMLSR_EXIT_CHAN_LOAD; if (reason_mask) { - IWL_DEBUG_INFO(mld, - "Links %d and %d are not a valid pair for EMLSR\n", - a->link_id, b->link_id); - IWL_DEBUG_INFO(mld, - "Links bandwidth are: %d and %d\n", - nl80211_chan_width_to_mhz(a->chandef->width), - nl80211_chan_width_to_mhz(b->chandef->width)); + IWL_DEBUG_EHT(mld, + "Links %d and %d are not a valid pair for EMLSR\n", + a->link_id, b->link_id); + IWL_DEBUG_EHT(mld, "Links bandwidth are: %d and %d\n", + nl80211_chan_width_to_mhz(a->chandef->width), + nl80211_chan_width_to_mhz(b->chandef->width)); iwl_mld_print_emlsr_exit(mld, reason_mask); } @@ -994,8 +985,8 @@ static void _iwl_mld_select_links(struct iwl_mld *mld, } set_active: - IWL_DEBUG_INFO(mld, "Link selection result: 0x%x. Primary = %d\n", - new_active, new_primary); + IWL_DEBUG_EHT(mld, "Link selection result: 0x%x. Primary = %d\n", + new_active, new_primary); mld_vif->emlsr.selected_primary = new_primary; mld_vif->emlsr.selected_links = new_active; diff --git a/drivers/net/wireless/intel/iwlwifi/mld/sta.c b/drivers/net/wireless/intel/iwlwifi/mld/sta.c index 5cdbfa29a202..61ecc33116cf 100644 --- a/drivers/net/wireless/intel/iwlwifi/mld/sta.c +++ b/drivers/net/wireless/intel/iwlwifi/mld/sta.c @@ -890,7 +890,7 @@ static void iwl_mld_count_mpdu(struct ieee80211_link_sta *link_sta, int queue, sizeof(queue_counter->per_link)); queue_counter->window_start_time = jiffies; - IWL_DEBUG_INFO(mld, "MPDU counters are cleared\n"); + IWL_DEBUG_EHT(mld, "MPDU counters are cleared\n"); } link_counter = &queue_counter->per_link[mld_link->fw_id]; From 5ee10092eb77dee1bed468777ff8e16130f21877 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Sun, 19 Oct 2025 11:45:15 +0300 Subject: [PATCH 0327/1381] wifi: iwlwifi: cfg: fix a few device names There are going to be some devices called BN203 instead of BN201, adjust the names accordingly. Signed-off-by: Johannes Berg Signed-off-by: Miri Korenblit Link: https://patch.msgid.link/20251019114304.2ad0f42ffb57.I2c7864d33f0d7d3dc49381949571c4ce620a9723@changeid --- drivers/net/wireless/intel/iwlwifi/cfg/rf-pe.c | 1 + drivers/net/wireless/intel/iwlwifi/iwl-config.h | 1 + drivers/net/wireless/intel/iwlwifi/pcie/drv.c | 7 ++++--- 3 files changed, 6 insertions(+), 3 deletions(-) diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/rf-pe.c b/drivers/net/wireless/intel/iwlwifi/cfg/rf-pe.c index 483f21659eff..408b9850bd10 100644 --- a/drivers/net/wireless/intel/iwlwifi/cfg/rf-pe.c +++ b/drivers/net/wireless/intel/iwlwifi/cfg/rf-pe.c @@ -12,5 +12,6 @@ const char iwl_killer_bn1850i_name[] = "Killer(R) Wi-Fi 8 BN1850i 320MHz Wireless Network Adapter (BN201.NGW)"; const char iwl_bn201_name[] = "Intel(R) Wi-Fi 8 BN201"; +const char iwl_bn203_name[] = "Intel(R) Wi-Fi 8 BN203"; const char iwl_be221_name[] = "Intel(R) Wi-Fi 7 BE221"; const char iwl_be223_name[] = "Intel(R) Wi-Fi 7 BE223"; diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-config.h b/drivers/net/wireless/intel/iwlwifi/iwl-config.h index 0a5a683325ef..076810ee5d34 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-config.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-config.h @@ -686,6 +686,7 @@ extern const char iwl_be211_name[]; extern const char iwl_killer_bn1850w2_name[]; extern const char iwl_killer_bn1850i_name[]; extern const char iwl_bn201_name[]; +extern const char iwl_bn203_name[]; extern const char iwl_be221_name[]; extern const char iwl_be223_name[]; extern const char iwl_ax221_name[]; diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c index 73001cdce13a..dc99e7ac4726 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c @@ -1069,9 +1069,10 @@ VISIBLE_IF_IWLWIFI_KUNIT const struct iwl_dev_info iwl_dev_info_table[] = { /* PE RF */ IWL_DEV_INFO(iwl_rf_pe, iwl_bn201_name, RF_TYPE(PE)), - IWL_DEV_INFO(iwl_rf_pe, iwl_be223_name, RF_TYPE(PE), SUBDEV(0x0524)), - IWL_DEV_INFO(iwl_rf_pe, iwl_be223_name, RF_TYPE(PE), SUBDEV(0x4524)), - IWL_DEV_INFO(iwl_rf_pe, iwl_be221_name, RF_TYPE(PE), SUBDEV(0x0324)), + IWL_DEV_INFO(iwl_rf_pe, iwl_be223_name, RF_TYPE(PE), + SUBDEV_MASKED(0x0524, 0xFFF)), + IWL_DEV_INFO(iwl_rf_pe, iwl_bn203_name, RF_TYPE(PE), + SUBDEV_MASKED(0x0324, 0xFFF)), /* Killer */ IWL_DEV_INFO(iwl_rf_wh, iwl_killer_be1775s_name, SUBDEV(0x1776)), From 75dd87e3f181bbf7d333c69fa45ef49eb75e4acd Mon Sep 17 00:00:00 2001 From: Emmanuel Grumbach Date: Sun, 19 Oct 2025 11:45:16 +0300 Subject: [PATCH 0328/1381] wifi: iwlwifi: mld: check for NULL pointer after kmalloc Coverity complained that we didn't add a NULL check for the link we allocate. Signed-off-by: Emmanuel Grumbach Reviewed-by: Johannes Berg Signed-off-by: Miri Korenblit Link: https://patch.msgid.link/20251019114304.d1f958160c5a.Icc891c14c633c3b8625372680fdc67ca33c83cc7@changeid --- drivers/net/wireless/intel/iwlwifi/mld/link.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/drivers/net/wireless/intel/iwlwifi/mld/link.c b/drivers/net/wireless/intel/iwlwifi/mld/link.c index deb1e7227dd8..27ae8bd6f407 100644 --- a/drivers/net/wireless/intel/iwlwifi/mld/link.c +++ b/drivers/net/wireless/intel/iwlwifi/mld/link.c @@ -465,10 +465,13 @@ int iwl_mld_add_link(struct iwl_mld *mld, int ret; if (!link) { - if (is_deflink) + if (is_deflink) { link = &mld_vif->deflink; - else + } else { link = kzalloc(sizeof(*link), GFP_KERNEL); + if (!link) + return -ENOMEM; + } } else { WARN_ON(!mld->fw_status.in_hw_restart); } From 3df28496673bd8009f1cd3a85a63650c96e369f4 Mon Sep 17 00:00:00 2001 From: Li Qiang Date: Fri, 17 Oct 2025 12:11:28 +0800 Subject: [PATCH 0329/1381] wifi: iwlwifi: mld: add null check for kzalloc() in iwl_mld_send_proto_offload() Add a missing NULL pointer check after kzalloc() in iwl_mld_send_proto_offload(). Without this check, a failed allocation could lead to a NULL dereference. Fixes: d1e879ec600f9 ("wifi: iwlwifi: add iwlmld sub-driver") Signed-off-by: Li Qiang Link: https://patch.msgid.link/20251017041128.1379715-1-liqiang01@kylinos.cn Signed-off-by: Miri Korenblit --- drivers/net/wireless/intel/iwlwifi/mld/d3.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/net/wireless/intel/iwlwifi/mld/d3.c b/drivers/net/wireless/intel/iwlwifi/mld/d3.c index 1d4282a21f09..dd85be94433c 100644 --- a/drivers/net/wireless/intel/iwlwifi/mld/d3.c +++ b/drivers/net/wireless/intel/iwlwifi/mld/d3.c @@ -1794,6 +1794,10 @@ iwl_mld_send_proto_offload(struct iwl_mld *mld, u32 enabled = 0; cmd = kzalloc(hcmd.len[0], GFP_KERNEL); + if (!cmd) { + IWL_DEBUG_WOWLAN(mld, "Failed to allocate proto offload cmd\n"); + return -ENOMEM; + } #if IS_ENABLED(CONFIG_IPV6) struct iwl_mld_vif *mld_vif = iwl_mld_vif_from_mac80211(vif); From c51aa14be9c4ad9f3d45f9dd2890776cfbccb55a Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Th=C3=A9o=20Lebrun?= Date: Thu, 23 Oct 2025 18:22:51 +0200 Subject: [PATCH 0330/1381] dt-bindings: net: cdns,macb: add Mobileye EyeQ5 ethernet interface MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Add "cdns,eyeq5-gem" as compatible for the integrated GEM block inside Mobileye EyeQ5 SoCs. It is different from other compatibles in two main ways: (1) it requires a generic PHY and (2) it is better to keep TCP Segmentation Offload (TSO) disabled. Reviewed-by: Andrew Lunn Acked-by: Conor Dooley Signed-off-by: Théo Lebrun Link: https://patch.msgid.link/20251023-macb-eyeq5-v3-1-af509422c204@bootlin.com Signed-off-by: Paolo Abeni --- Documentation/devicetree/bindings/net/cdns,macb.yaml | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/Documentation/devicetree/bindings/net/cdns,macb.yaml b/Documentation/devicetree/bindings/net/cdns,macb.yaml index 02f14a0b72f9..ea8337846ab2 100644 --- a/Documentation/devicetree/bindings/net/cdns,macb.yaml +++ b/Documentation/devicetree/bindings/net/cdns,macb.yaml @@ -57,6 +57,7 @@ properties: - cdns,np4-macb # NP4 SoC devices - microchip,sama7g5-emac # Microchip SAMA7G5 ethernet interface - microchip,sama7g5-gem # Microchip SAMA7G5 gigabit ethernet interface + - mobileye,eyeq5-gem # Mobileye EyeQ5 SoCs - raspberrypi,rp1-gem # Raspberry Pi RP1 gigabit ethernet interface - sifive,fu540-c000-gem # SiFive FU540-C000 SoC @@ -183,6 +184,15 @@ allOf: reg: maxItems: 1 + - if: + properties: + compatible: + contains: + const: mobileye,eyeq5-gem + then: + required: + - phys + unevaluatedProperties: false examples: From ae7a9585ea6974bd7a772fee96bb8514e250dbd6 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Th=C3=A9o=20Lebrun?= Date: Thu, 23 Oct 2025 18:22:52 +0200 Subject: [PATCH 0331/1381] net: macb: match skb_reserve(skb, NET_IP_ALIGN) with HW alignment MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit If HW is RSC capable, it cannot add dummy bytes at the start of IP packets. Alignment (ie number of dummy bytes) is configured using the RBOF field inside the NCFGR register. On the software side, the skb_reserve(skb, NET_IP_ALIGN) call must only be done if those dummy bytes are added by the hardware; notice the skb_reserve() is done AFTER writing the address to the device. We cannot do the skb_reserve() call BEFORE writing the address because the address field ignores the low 2/3 bits. Conclusion: in some cases, we risk not being able to respect the NET_IP_ALIGN value (which is picked based on unaligned CPU access performance). Signed-off-by: Théo Lebrun Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20251023-macb-eyeq5-v3-2-af509422c204@bootlin.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/cadence/macb.h | 3 +++ drivers/net/ethernet/cadence/macb_main.c | 23 ++++++++++++++++++++--- 2 files changed, 23 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/cadence/macb.h b/drivers/net/ethernet/cadence/macb.h index 5b7d4cdb204d..93e8dd092313 100644 --- a/drivers/net/ethernet/cadence/macb.h +++ b/drivers/net/ethernet/cadence/macb.h @@ -537,6 +537,8 @@ /* Bitfields in DCFG6. */ #define GEM_PBUF_LSO_OFFSET 27 #define GEM_PBUF_LSO_SIZE 1 +#define GEM_PBUF_RSC_OFFSET 26 +#define GEM_PBUF_RSC_SIZE 1 #define GEM_PBUF_CUTTHRU_OFFSET 25 #define GEM_PBUF_CUTTHRU_SIZE 1 #define GEM_DAW64_OFFSET 23 @@ -775,6 +777,7 @@ #define MACB_CAPS_MACB_IS_GEM BIT(20) #define MACB_CAPS_DMA_64B BIT(21) #define MACB_CAPS_DMA_PTP BIT(22) +#define MACB_CAPS_RSC BIT(23) /* LSO settings */ #define MACB_LSO_UFO_ENABLE 0x01 diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c index 39673f5c3337..be3d0c2313a1 100644 --- a/drivers/net/ethernet/cadence/macb_main.c +++ b/drivers/net/ethernet/cadence/macb_main.c @@ -1300,8 +1300,19 @@ static void gem_rx_refill(struct macb_queue *queue) dma_wmb(); macb_set_addr(bp, desc, paddr); - /* properly align Ethernet header */ - skb_reserve(skb, NET_IP_ALIGN); + /* Properly align Ethernet header. + * + * Hardware can add dummy bytes if asked using the RBOF + * field inside the NCFGR register. That feature isn't + * available if hardware is RSC capable. + * + * We cannot fallback to doing the 2-byte shift before + * DMA mapping because the address field does not allow + * setting the low 2/3 bits. + * It is 3 bits if HW_DMA_CAP_PTP, else 2 bits. + */ + if (!(bp->caps & MACB_CAPS_RSC)) + skb_reserve(skb, NET_IP_ALIGN); } else { desc->ctrl = 0; dma_wmb(); @@ -2773,7 +2784,11 @@ static void macb_init_hw(struct macb *bp) macb_set_hwaddr(bp); config = macb_mdc_clk_div(bp); - config |= MACB_BF(RBOF, NET_IP_ALIGN); /* Make eth data aligned */ + /* Make eth data aligned. + * If RSC capable, that offset is ignored by HW. + */ + if (!(bp->caps & MACB_CAPS_RSC)) + config |= MACB_BF(RBOF, NET_IP_ALIGN); config |= MACB_BIT(DRFCS); /* Discard Rx FCS */ if (bp->caps & MACB_CAPS_JUMBO) config |= MACB_BIT(JFRAME); /* Enable jumbo frames */ @@ -4321,6 +4336,8 @@ static void macb_configure_caps(struct macb *bp, dcfg = gem_readl(bp, DCFG2); if ((dcfg & (GEM_BIT(RX_PKT_BUFF) | GEM_BIT(TX_PKT_BUFF))) == 0) bp->caps |= MACB_CAPS_FIFO_MODE; + if (GEM_BFEXT(PBUF_RSC, gem_readl(bp, DCFG6))) + bp->caps |= MACB_CAPS_RSC; if (gem_has_ptp(bp)) { if (!GEM_BFEXT(TSU, gem_readl(bp, DCFG5))) dev_err(&bp->pdev->dev, From 7a3d209145d17b4e0d24a51864fb2def1f5ecf0b Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Th=C3=A9o=20Lebrun?= Date: Thu, 23 Oct 2025 18:22:53 +0200 Subject: [PATCH 0332/1381] net: macb: add no LSO capability (MACB_CAPS_NO_LSO) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit LSO is runtime-detected using the PBUF_LSO field inside register DCFG6. Allow disabling that feature if it is broken by using bp->caps coming from match data. Reviewed-by: Andrew Lunn Signed-off-by: Théo Lebrun Link: https://patch.msgid.link/20251023-macb-eyeq5-v3-3-af509422c204@bootlin.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/cadence/macb.h | 1 + drivers/net/ethernet/cadence/macb_main.c | 7 +++++-- 2 files changed, 6 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/cadence/macb.h b/drivers/net/ethernet/cadence/macb.h index 93e8dd092313..05bfa9bd4782 100644 --- a/drivers/net/ethernet/cadence/macb.h +++ b/drivers/net/ethernet/cadence/macb.h @@ -778,6 +778,7 @@ #define MACB_CAPS_DMA_64B BIT(21) #define MACB_CAPS_DMA_PTP BIT(22) #define MACB_CAPS_RSC BIT(23) +#define MACB_CAPS_NO_LSO BIT(24) /* LSO settings */ #define MACB_LSO_UFO_ENABLE 0x01 diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c index be3d0c2313a1..8b688a6cb2f9 100644 --- a/drivers/net/ethernet/cadence/macb_main.c +++ b/drivers/net/ethernet/cadence/macb_main.c @@ -4564,8 +4564,11 @@ static int macb_init(struct platform_device *pdev) /* Set features */ dev->hw_features = NETIF_F_SG; - /* Check LSO capability */ - if (GEM_BFEXT(PBUF_LSO, gem_readl(bp, DCFG6))) + /* Check LSO capability; runtime detection can be overridden by a cap + * flag if the hardware is known to be buggy + */ + if (!(bp->caps & MACB_CAPS_NO_LSO) && + GEM_BFEXT(PBUF_LSO, gem_readl(bp, DCFG6))) dev->hw_features |= MACB_NETIF_LSO; /* Checksum offload is only available on gem with packet buffer */ From 3f7e51cd5fbf4d970b14956ee9464515bb40666f Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Th=C3=A9o=20Lebrun?= Date: Thu, 23 Oct 2025 18:22:54 +0200 Subject: [PATCH 0333/1381] net: macb: rename bp->sgmii_phy field to bp->phy MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The bp->sgmii_phy field is initialised at probe by init_reset_optional() if bp->phy_interface == PHY_INTERFACE_MODE_SGMII. It gets used by: - zynqmp_config: "cdns,zynqmp-gem" or "xlnx,zynqmp-gem" compatibles. - mpfs_config: "microchip,mpfs-macb" compatible. - versal_config: "xlnx,versal-gem" compatible. Make name more generic as EyeQ5 requires the PHY in SGMII & RGMII cases. Drop "for ZynqMP SGMII mode" comment that is already a lie, as it gets used on Microchip platforms as well. And soon it won't be SGMII-only. Reviewed-by: Andrew Lunn Reviewed-by: Maxime Chevallier Signed-off-by: Théo Lebrun Link: https://patch.msgid.link/20251023-macb-eyeq5-v3-4-af509422c204@bootlin.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/cadence/macb.h | 2 +- drivers/net/ethernet/cadence/macb_main.c | 26 ++++++++++++------------ 2 files changed, 14 insertions(+), 14 deletions(-) diff --git a/drivers/net/ethernet/cadence/macb.h b/drivers/net/ethernet/cadence/macb.h index 05bfa9bd4782..87414a2ddf6e 100644 --- a/drivers/net/ethernet/cadence/macb.h +++ b/drivers/net/ethernet/cadence/macb.h @@ -1341,7 +1341,7 @@ struct macb { struct macb_ptp_info *ptp_info; /* macb-ptp interface */ - struct phy *sgmii_phy; /* for ZynqMP SGMII mode */ + struct phy *phy; spinlock_t tsu_clk_lock; /* gem tsu clock locking */ unsigned int tsu_rate; diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c index 8b688a6cb2f9..44188e7eee56 100644 --- a/drivers/net/ethernet/cadence/macb_main.c +++ b/drivers/net/ethernet/cadence/macb_main.c @@ -2965,7 +2965,7 @@ static int macb_open(struct net_device *dev) macb_init_hw(bp); - err = phy_power_on(bp->sgmii_phy); + err = phy_power_on(bp->phy); if (err) goto reset_hw; @@ -2981,7 +2981,7 @@ static int macb_open(struct net_device *dev) return 0; phy_off: - phy_power_off(bp->sgmii_phy); + phy_power_off(bp->phy); reset_hw: macb_reset_hw(bp); @@ -3013,7 +3013,7 @@ static int macb_close(struct net_device *dev) phylink_stop(bp->phylink); phylink_disconnect_phy(bp->phylink); - phy_power_off(bp->sgmii_phy); + phy_power_off(bp->phy); spin_lock_irqsave(&bp->lock, flags); macb_reset_hw(bp); @@ -5141,13 +5141,13 @@ static int init_reset_optional(struct platform_device *pdev) if (bp->phy_interface == PHY_INTERFACE_MODE_SGMII) { /* Ensure PHY device used in SGMII mode is ready */ - bp->sgmii_phy = devm_phy_optional_get(&pdev->dev, NULL); + bp->phy = devm_phy_optional_get(&pdev->dev, NULL); - if (IS_ERR(bp->sgmii_phy)) - return dev_err_probe(&pdev->dev, PTR_ERR(bp->sgmii_phy), + if (IS_ERR(bp->phy)) + return dev_err_probe(&pdev->dev, PTR_ERR(bp->phy), "failed to get SGMII PHY\n"); - ret = phy_init(bp->sgmii_phy); + ret = phy_init(bp->phy); if (ret) return dev_err_probe(&pdev->dev, ret, "failed to init SGMII PHY\n"); @@ -5176,7 +5176,7 @@ static int init_reset_optional(struct platform_device *pdev) /* Fully reset controller at hardware level if mapped in device tree */ ret = device_reset_optional(&pdev->dev); if (ret) { - phy_exit(bp->sgmii_phy); + phy_exit(bp->phy); return dev_err_probe(&pdev->dev, ret, "failed to reset controller"); } @@ -5184,7 +5184,7 @@ static int init_reset_optional(struct platform_device *pdev) err_out_phy_exit: if (ret) - phy_exit(bp->sgmii_phy); + phy_exit(bp->phy); return ret; } @@ -5594,7 +5594,7 @@ static int macb_probe(struct platform_device *pdev) mdiobus_free(bp->mii_bus); err_out_phy_exit: - phy_exit(bp->sgmii_phy); + phy_exit(bp->phy); err_out_free_netdev: free_netdev(dev); @@ -5618,7 +5618,7 @@ static void macb_remove(struct platform_device *pdev) if (dev) { bp = netdev_priv(dev); unregister_netdev(dev); - phy_exit(bp->sgmii_phy); + phy_exit(bp->phy); mdiobus_unregister(bp->mii_bus); mdiobus_free(bp->mii_bus); @@ -5645,7 +5645,7 @@ static int __maybe_unused macb_suspend(struct device *dev) u32 tmp; if (!device_may_wakeup(&bp->dev->dev)) - phy_exit(bp->sgmii_phy); + phy_exit(bp->phy); if (!netif_running(netdev)) return 0; @@ -5774,7 +5774,7 @@ static int __maybe_unused macb_resume(struct device *dev) int err; if (!device_may_wakeup(&bp->dev->dev)) - phy_init(bp->sgmii_phy); + phy_init(bp->phy); if (!netif_running(netdev)) return 0; From 48cf0be9b9a66ea64192beb215911d3d7c94a409 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Th=C3=A9o=20Lebrun?= Date: Thu, 23 Oct 2025 18:22:55 +0200 Subject: [PATCH 0334/1381] net: macb: Add "mobileye,eyeq5-gem" compatible MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Add support for the two GEM instances inside Mobileye EyeQ5 SoCs, using compatible "mobileye,eyeq5-gem". With it, add a custom init sequence that must grab a generic PHY and initialise it. We use bp->phy in both RGMII and SGMII cases. Tell our mode by adding a phy_set_mode_ext() during macb_open(), before phy_power_on(). We are the first users of bp->phy that use it in non-SGMII cases. The phy_set_mode_ext() call is made unconditionally. It cannot cause issues on platforms where !bp->phy or !bp->phy->ops->set_mode as, in those cases, the call is a no-op (returning zero). From reading upstream DTS, we can figure out that no platform has a bp->phy and a PHY driver that has a .set_mode() implementation: - cdns,zynqmp-gem: no DTS upstream. - microchip,mpfs-macb: microchip/mpfs.dtsi, &mac0..1, no PHY attached. - xlnx,versal-gem: xilinx/versal-net.dtsi, &gem0..1, no PHY attached. - xlnx,zynqmp-gem: xilinx/zynqmp.dtsi, &gem0..3, PHY attached to drivers/phy/xilinx/phy-zynqmp.c which has no .set_mode(). Reviewed-by: Maxime Chevallier Signed-off-by: Théo Lebrun Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20251023-macb-eyeq5-v3-5-af509422c204@bootlin.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/cadence/macb_main.c | 38 ++++++++++++++++++++++++ 1 file changed, 38 insertions(+) diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c index 44188e7eee56..b1ed98d9c438 100644 --- a/drivers/net/ethernet/cadence/macb_main.c +++ b/drivers/net/ethernet/cadence/macb_main.c @@ -2965,6 +2965,10 @@ static int macb_open(struct net_device *dev) macb_init_hw(bp); + err = phy_set_mode_ext(bp->phy, PHY_MODE_ETHERNET, bp->phy_interface); + if (err) + goto reset_hw; + err = phy_power_on(bp->phy); if (err) goto reset_hw; @@ -5189,6 +5193,28 @@ static int init_reset_optional(struct platform_device *pdev) return ret; } +static int eyeq5_init(struct platform_device *pdev) +{ + struct net_device *netdev = platform_get_drvdata(pdev); + struct macb *bp = netdev_priv(netdev); + struct device *dev = &pdev->dev; + int ret; + + bp->phy = devm_phy_get(dev, NULL); + if (IS_ERR(bp->phy)) + return dev_err_probe(dev, PTR_ERR(bp->phy), + "failed to get PHY\n"); + + ret = phy_init(bp->phy); + if (ret) + return dev_err_probe(dev, ret, "failed to init PHY\n"); + + ret = macb_init(pdev); + if (ret) + phy_exit(bp->phy); + return ret; +} + static const struct macb_usrio_config sama7g5_usrio = { .mii = 0, .rmii = 1, @@ -5343,6 +5369,17 @@ static const struct macb_config versal_config = { .usrio = &macb_default_usrio, }; +static const struct macb_config eyeq5_config = { + .caps = MACB_CAPS_GIGABIT_MODE_AVAILABLE | MACB_CAPS_JUMBO | + MACB_CAPS_GEM_HAS_PTP | MACB_CAPS_QUEUE_DISABLE | + MACB_CAPS_NO_LSO, + .dma_burst_length = 16, + .clk_init = macb_clk_init, + .init = eyeq5_init, + .jumbo_max_len = 10240, + .usrio = &macb_default_usrio, +}; + static const struct macb_config raspberrypi_rp1_config = { .caps = MACB_CAPS_GIGABIT_MODE_AVAILABLE | MACB_CAPS_CLK_HW_CHG | MACB_CAPS_JUMBO | @@ -5374,6 +5411,7 @@ static const struct of_device_id macb_dt_ids[] = { { .compatible = "microchip,mpfs-macb", .data = &mpfs_config }, { .compatible = "microchip,sama7g5-gem", .data = &sama7g5_gem_config }, { .compatible = "microchip,sama7g5-emac", .data = &sama7g5_emac_config }, + { .compatible = "mobileye,eyeq5-gem", .data = &eyeq5_config }, { .compatible = "raspberrypi,rp1-gem", .data = &raspberrypi_rp1_config }, { .compatible = "xlnx,zynqmp-gem", .data = &zynqmp_config}, { .compatible = "xlnx,zynq-gem", .data = &zynq_config }, From 792000fbcd0ca32f358c4e14eaa40a73690be24c Mon Sep 17 00:00:00 2001 From: Maxime Chevallier Date: Fri, 24 Oct 2025 09:07:17 +0200 Subject: [PATCH 0335/1381] net: stmmac: Move subsecond increment configuration in dedicated helper In preparation for fine/coarse support, let's move the subsecond increment and addend configuration in a dedicated helper. Signed-off-by: Maxime Chevallier Reviewed-by: Russell King (Oracle) Link: https://patch.msgid.link/20251024070720.71174-2-maxime.chevallier@bootlin.com Signed-off-by: Paolo Abeni --- .../net/ethernet/stmicro/stmmac/stmmac_main.c | 48 +++++++++++-------- 1 file changed, 28 insertions(+), 20 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index fd5106880192..6706ae7dd89f 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -464,6 +464,33 @@ static void stmmac_get_rx_hwtstamp(struct stmmac_priv *priv, struct dma_desc *p, } } +static void stmmac_update_subsecond_increment(struct stmmac_priv *priv) +{ + bool xmac = dwmac_is_xmac(priv->plat->core_type); + u32 sec_inc = 0; + u64 temp = 0; + + stmmac_config_hw_tstamping(priv, priv->ptpaddr, priv->systime_flags); + + /* program Sub Second Increment reg */ + stmmac_config_sub_second_increment(priv, priv->ptpaddr, + priv->plat->clk_ptp_rate, + xmac, &sec_inc); + temp = div_u64(1000000000ULL, sec_inc); + + /* Store sub second increment for later use */ + priv->sub_second_inc = sec_inc; + + /* calculate default added value: + * formula is : + * addend = (2^32)/freq_div_ratio; + * where, freq_div_ratio = 1e9ns/sec_inc + */ + temp = (u64)(temp << 32); + priv->default_addend = div_u64(temp, priv->plat->clk_ptp_rate); + stmmac_config_addend(priv, priv->ptpaddr, priv->default_addend); +} + /** * stmmac_hwtstamp_set - control hardware timestamping. * @dev: device pointer. @@ -697,10 +724,7 @@ static int stmmac_hwtstamp_get(struct net_device *dev, static int stmmac_init_tstamp_counter(struct stmmac_priv *priv, u32 systime_flags) { - bool xmac = dwmac_is_xmac(priv->plat->core_type); struct timespec64 now; - u32 sec_inc = 0; - u64 temp = 0; if (!priv->plat->clk_ptp_rate) { netdev_err(priv->dev, "Invalid PTP clock rate"); @@ -710,23 +734,7 @@ static int stmmac_init_tstamp_counter(struct stmmac_priv *priv, stmmac_config_hw_tstamping(priv, priv->ptpaddr, systime_flags); priv->systime_flags = systime_flags; - /* program Sub Second Increment reg */ - stmmac_config_sub_second_increment(priv, priv->ptpaddr, - priv->plat->clk_ptp_rate, - xmac, &sec_inc); - temp = div_u64(1000000000ULL, sec_inc); - - /* Store sub second increment for later use */ - priv->sub_second_inc = sec_inc; - - /* calculate default added value: - * formula is : - * addend = (2^32)/freq_div_ratio; - * where, freq_div_ratio = 1e9ns/sec_inc - */ - temp = (u64)(temp << 32); - priv->default_addend = div_u64(temp, priv->plat->clk_ptp_rate); - stmmac_config_addend(priv, priv->ptpaddr, priv->default_addend); + stmmac_update_subsecond_increment(priv); /* initialize system time */ ktime_get_real_ts64(&now); From 6920fa0c764dbdd35d311d4df986226bb48165f6 Mon Sep 17 00:00:00 2001 From: Maxime Chevallier Date: Fri, 24 Oct 2025 09:07:18 +0200 Subject: [PATCH 0336/1381] net: stmmac: Add a devlink attribute to control timestamping mode The DWMAC1000 supports 2 timestamping configurations to configure how frequency adjustments are made to the ptp_clock, as well as the reported timestamp values. There was a previous attempt at upstreaming support for configuring this mode by Olivier Dautricourt and Julien Beraud a few years back [1] In a nutshell, the timestamping can be either set in fine mode or in coarse mode. In fine mode, which is the default, we use the overflow of an accumulator to trigger frequency adjustments, but by doing so we lose precision on the timetamps that are produced by the timestamping unit. The main drawback is that the sub-second increment value, used to generate timestamps, can't be set to lower than (2 / ptp_clock_freq). The "fine" qualification comes from the frequent frequency adjustments we are able to do, which is perfect for a PTP follower usecase. In Coarse mode, we don't do frequency adjustments based on an accumulator overflow. We can therefore have very fine subsecond increment values, allowing for better timestamping precision. However this mode works best when the ptp clock frequency is adjusted based on an external signal, such as a PPS input produced by a GPS clock. This mode is therefore perfect for a Grand-master usecase. Introduce a driver-specific devlink parameter "ts_coarse" to enable or disable coarse mode, keeping the "fine" mode as a default. This can then be changed with: devlink dev param set name ts_coarse value true cmode runtime The associated documentation is also added. [1] : https://lore.kernel.org/netdev/20200514102808.31163-1-olivier.dautricourt@orolia.com/ Signed-off-by: Maxime Chevallier Reviewed-by: Russell King (Oracle) Reviewed-by: Kory Maincent Link: https://patch.msgid.link/20251024070720.71174-3-maxime.chevallier@bootlin.com Signed-off-by: Paolo Abeni --- Documentation/networking/devlink/index.rst | 1 + Documentation/networking/devlink/stmmac.rst | 31 +++++ drivers/net/ethernet/stmicro/stmmac/Kconfig | 1 + drivers/net/ethernet/stmicro/stmmac/stmmac.h | 3 + .../net/ethernet/stmicro/stmmac/stmmac_main.c | 115 +++++++++++++++++- 5 files changed, 148 insertions(+), 3 deletions(-) create mode 100644 Documentation/networking/devlink/stmmac.rst diff --git a/Documentation/networking/devlink/index.rst b/Documentation/networking/devlink/index.rst index 0c58e5c729d9..35b12a2bfeba 100644 --- a/Documentation/networking/devlink/index.rst +++ b/Documentation/networking/devlink/index.rst @@ -99,5 +99,6 @@ parameters, info versions, and other features it supports. prestera qed sfc + stmmac ti-cpsw-switch zl3073x diff --git a/Documentation/networking/devlink/stmmac.rst b/Documentation/networking/devlink/stmmac.rst new file mode 100644 index 000000000000..e8e33d1c7baf --- /dev/null +++ b/Documentation/networking/devlink/stmmac.rst @@ -0,0 +1,31 @@ +.. SPDX-License-Identifier: GPL-2.0 + +======================================= +stmmac (synopsys dwmac) devlink support +======================================= + +This document describes the devlink features implemented by the ``stmmac`` +device driver. + +Parameters +========== + +The ``stmmac`` driver implements the following driver-specific parameters. + +.. list-table:: Driver-specific parameters implemented + :widths: 5 5 5 85 + + * - Name + - Type + - Mode + - Description + * - ``ts_coarse`` + - Boolean + - runtime + - Enable the Coarse timestamping mode. In Coarse mode, the ptp clock is + expected to be updated through an external PPS input, but the subsecond + increment used for timestamping is set to 1/ptp_clock_rate. In Fine mode + (i.e. Coarse mode == false), the ptp clock frequency is adjusted more + frequently, but the subsecond increment is set to 2/ptp_clock_rate. + Coarse mode is suitable for PTP Grand Master operation. If unsure, leave + the parameter to False. diff --git a/drivers/net/ethernet/stmicro/stmmac/Kconfig b/drivers/net/ethernet/stmicro/stmmac/Kconfig index 716daa51df7e..87c5bea6c2a2 100644 --- a/drivers/net/ethernet/stmicro/stmmac/Kconfig +++ b/drivers/net/ethernet/stmicro/stmmac/Kconfig @@ -10,6 +10,7 @@ config STMMAC_ETH select PHYLINK select CRC32 select RESET_CONTROLLER + select NET_DEVLINK help This is the driver for the Ethernet IPs built around a Synopsys IP Core. diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac.h b/drivers/net/ethernet/stmicro/stmmac/stmmac.h index d5af9344dfb0..3ea680cc63d8 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac.h +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac.h @@ -259,6 +259,7 @@ struct stmmac_priv { u32 sarc_type; u32 rx_riwt[MTL_MAX_RX_QUEUES]; int hwts_rx_en; + bool tsfupdt_coarse; void __iomem *ioaddr; struct net_device *dev; @@ -369,6 +370,8 @@ struct stmmac_priv { /* XDP BPF Program */ unsigned long *af_xdp_zc_qps; struct bpf_prog *xdp_prog; + + struct devlink *devlink; }; enum stmmac_state { diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index 6706ae7dd89f..ba4eeba14baa 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -40,6 +40,7 @@ #include #include #include +#include #include #include #include @@ -58,8 +59,7 @@ * with fine resolution and binary rollover. This avoid non-monotonic behavior * (clock jumps) when changing timestamping settings at runtime. */ -#define STMMAC_HWTS_ACTIVE (PTP_TCR_TSENA | PTP_TCR_TSCFUPDT | \ - PTP_TCR_TSCTRLSSR) +#define STMMAC_HWTS_ACTIVE (PTP_TCR_TSENA | PTP_TCR_TSCTRLSSR) #define STMMAC_ALIGN(x) ALIGN(ALIGN(x, SMP_CACHE_BYTES), 16) #define TSO_MAX_BUFF_SIZE (SZ_16K - 1) @@ -148,6 +148,15 @@ static void stmmac_exit_fs(struct net_device *dev); #define STMMAC_COAL_TIMER(x) (ns_to_ktime((x) * NSEC_PER_USEC)) +struct stmmac_devlink_priv { + struct stmmac_priv *stmmac_priv; +}; + +enum stmmac_dl_param_id { + STMMAC_DEVLINK_PARAM_ID_BASE = DEVLINK_PARAM_GENERIC_ID_MAX, + STMMAC_DEVLINK_PARAM_ID_TS_COARSE, +}; + /** * stmmac_set_clk_tx_rate() - set the clock rate for the MAC transmit clock * @bsp_priv: BSP private data structure (unused) @@ -675,6 +684,8 @@ static int stmmac_hwtstamp_set(struct net_device *dev, priv->hwts_tx_en = config->tx_type == HWTSTAMP_TX_ON; priv->systime_flags = STMMAC_HWTS_ACTIVE; + if (!priv->tsfupdt_coarse) + priv->systime_flags |= PTP_TCR_TSCFUPDT; if (priv->hwts_tx_en || priv->hwts_rx_en) { priv->systime_flags |= tstamp_all | ptp_v2 | @@ -765,7 +776,8 @@ static int stmmac_init_timestamping(struct stmmac_priv *priv) return -EOPNOTSUPP; } - ret = stmmac_init_tstamp_counter(priv, STMMAC_HWTS_ACTIVE); + ret = stmmac_init_tstamp_counter(priv, STMMAC_HWTS_ACTIVE | + PTP_TCR_TSCFUPDT); if (ret) { netdev_warn(priv->dev, "PTP init failed\n"); return ret; @@ -7400,6 +7412,95 @@ static const struct xdp_metadata_ops stmmac_xdp_metadata_ops = { .xmo_rx_timestamp = stmmac_xdp_rx_timestamp, }; +static int stmmac_dl_ts_coarse_set(struct devlink *dl, u32 id, + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) +{ + struct stmmac_devlink_priv *dl_priv = devlink_priv(dl); + struct stmmac_priv *priv = dl_priv->stmmac_priv; + + priv->tsfupdt_coarse = ctx->val.vbool; + + if (priv->tsfupdt_coarse) + priv->systime_flags &= ~PTP_TCR_TSCFUPDT; + else + priv->systime_flags |= PTP_TCR_TSCFUPDT; + + /* In Coarse mode, we can use a smaller subsecond increment, let's + * reconfigure the systime, subsecond increment and addend. + */ + stmmac_update_subsecond_increment(priv); + + return 0; +} + +static int stmmac_dl_ts_coarse_get(struct devlink *dl, u32 id, + struct devlink_param_gset_ctx *ctx) +{ + struct stmmac_devlink_priv *dl_priv = devlink_priv(dl); + struct stmmac_priv *priv = dl_priv->stmmac_priv; + + ctx->val.vbool = priv->tsfupdt_coarse; + + return 0; +} + +static const struct devlink_param stmmac_devlink_params[] = { + DEVLINK_PARAM_DRIVER(STMMAC_DEVLINK_PARAM_ID_TS_COARSE, "ts_coarse", + DEVLINK_PARAM_TYPE_BOOL, + BIT(DEVLINK_PARAM_CMODE_RUNTIME), + stmmac_dl_ts_coarse_get, + stmmac_dl_ts_coarse_set, NULL), +}; + +/* None of the generic devlink parameters are implemented */ +static const struct devlink_ops stmmac_devlink_ops = {}; + +static int stmmac_register_devlink(struct stmmac_priv *priv) +{ + struct stmmac_devlink_priv *dl_priv; + int ret; + + /* For now, what is exposed over devlink is only relevant when + * timestamping is available and we have a valid ptp clock rate + */ + if (!(priv->dma_cap.time_stamp || priv->dma_cap.atime_stamp) || + !priv->plat->clk_ptp_rate) + return 0; + + priv->devlink = devlink_alloc(&stmmac_devlink_ops, sizeof(*dl_priv), + priv->device); + if (!priv->devlink) + return -ENOMEM; + + dl_priv = devlink_priv(priv->devlink); + dl_priv->stmmac_priv = priv; + + ret = devlink_params_register(priv->devlink, stmmac_devlink_params, + ARRAY_SIZE(stmmac_devlink_params)); + if (ret) + goto dl_free; + + devlink_register(priv->devlink); + return 0; + +dl_free: + devlink_free(priv->devlink); + + return ret; +} + +static void stmmac_unregister_devlink(struct stmmac_priv *priv) +{ + if (!priv->devlink) + return; + + devlink_unregister(priv->devlink); + devlink_params_unregister(priv->devlink, stmmac_devlink_params, + ARRAY_SIZE(stmmac_devlink_params)); + devlink_free(priv->devlink); +} + /** * stmmac_dvr_probe * @device: device pointer @@ -7673,6 +7774,10 @@ int stmmac_dvr_probe(struct device *device, goto error_phy_setup; } + ret = stmmac_register_devlink(priv); + if (ret) + goto error_devlink_setup; + ret = register_netdev(ndev); if (ret) { dev_err(priv->device, "%s: ERROR %i registering the device\n", @@ -7695,6 +7800,8 @@ int stmmac_dvr_probe(struct device *device, return ret; error_netdev_register: + stmmac_unregister_devlink(priv); +error_devlink_setup: phylink_destroy(priv->phylink); error_phy_setup: stmmac_pcs_clean(ndev); @@ -7731,6 +7838,8 @@ void stmmac_dvr_remove(struct device *dev) #ifdef CONFIG_DEBUG_FS stmmac_exit_fs(ndev); #endif + stmmac_unregister_devlink(priv); + phylink_destroy(priv->phylink); if (priv->plat->stmmac_rst) reset_control_assert(priv->plat->stmmac_rst); From 09e2603513841c48e5422d132f57f8a4c7337721 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 27 Oct 2025 12:29:57 -0700 Subject: [PATCH 0337/1381] tools: ynl: fix indent issues in the main Python lib Class NlError() and operation_do_attributes() are indented by 2 spaces rather than 4 spaces used by the rest of the file. Reviewed-by: Donald Hunter Link: https://patch.msgid.link/20251027192958.2058340-1-kuba@kernel.org Signed-off-by: Jakub Kicinski --- tools/net/ynl/pyynl/lib/ynl.py | 26 +++++++++++++------------- 1 file changed, 13 insertions(+), 13 deletions(-) diff --git a/tools/net/ynl/pyynl/lib/ynl.py b/tools/net/ynl/pyynl/lib/ynl.py index 62383c70ebb9..bdcc4f031d39 100644 --- a/tools/net/ynl/pyynl/lib/ynl.py +++ b/tools/net/ynl/pyynl/lib/ynl.py @@ -100,12 +100,12 @@ class Netlink: 'bitfield32', 'sint', 'uint']) class NlError(Exception): - def __init__(self, nl_msg): - self.nl_msg = nl_msg - self.error = -nl_msg.error + def __init__(self, nl_msg): + self.nl_msg = nl_msg + self.error = -nl_msg.error - def __str__(self): - return f"Netlink error: {os.strerror(self.error)}\n{self.nl_msg}" + def __str__(self): + return f"Netlink error: {os.strerror(self.error)}\n{self.nl_msg}" class ConfigError(Exception): @@ -1039,15 +1039,15 @@ class YnlFamily(SpecFamily): self.check_ntf() def operation_do_attributes(self, name): - """ - For a given operation name, find and return a supported - set of attributes (as a dict). - """ - op = self.find_operation(name) - if not op: - return None + """ + For a given operation name, find and return a supported + set of attributes (as a dict). + """ + op = self.find_operation(name) + if not op: + return None - return op['do']['request']['attributes'].copy() + return op['do']['request']['attributes'].copy() def _encode_message(self, op, vals, flags, req_seq): nl_flags = Netlink.NLM_F_REQUEST | Netlink.NLM_F_ACK From 34164142b5fd6878cd487f531ae074e3227031ac Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 27 Oct 2025 12:29:58 -0700 Subject: [PATCH 0338/1381] tools: ynl: rework the string representation of NlError In early days of YNL development dumping the NlMsg on errors was quite useful, as the library itself could have been buggy. These days increasingly the NlMsg is just taking up screen space and means nothing to a typical user. Try to format the errors more in line with how YNL C formats its errors strings. Before: $ ynl --family ethtool --do channels-set --json '{}' Netlink error: Invalid argument nl_len = 44 (28) nl_flags = 0x300 nl_type = 2 error: -22 extack: {'miss-type': 'header'} $ ynl --family ethtool --do channels-set --json '{..., "tx-count": 999}' Netlink error: Invalid argument nl_len = 88 (72) nl_flags = 0x300 nl_type = 2 error: -22 extack: {'msg': 'requested channel count exceeds maximum', 'bad-attr': '.tx-count'} After: $ ynl --family ethtool --do channels-set --json '{}' Netlink error: Invalid argument {'miss-type': 'header'} $ ynl --family ethtool --do channels-set --json '{..., "tx-count": 999}' Netlink error: requested channel count exceeds maximum: Invalid argument {'bad-attr': '.tx-count'} Reviewed-by: Donald Hunter Link: https://patch.msgid.link/20251027192958.2058340-2-kuba@kernel.org Signed-off-by: Jakub Kicinski --- tools/net/ynl/pyynl/lib/ynl.py | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/tools/net/ynl/pyynl/lib/ynl.py b/tools/net/ynl/pyynl/lib/ynl.py index bdcc4f031d39..225baad3c8f8 100644 --- a/tools/net/ynl/pyynl/lib/ynl.py +++ b/tools/net/ynl/pyynl/lib/ynl.py @@ -105,7 +105,16 @@ class NlError(Exception): self.error = -nl_msg.error def __str__(self): - return f"Netlink error: {os.strerror(self.error)}\n{self.nl_msg}" + msg = "Netlink error: " + + extack = self.nl_msg.extack.copy() if self.nl_msg.extack else {} + if 'msg' in extack: + msg += extack['msg'] + ': ' + del extack['msg'] + msg += os.strerror(self.error) + if extack: + msg += ' ' + str(extack) + return msg class ConfigError(Exception): From a086e9860ce6a751acd71dbec54d24a819dd6baa Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 24 Oct 2025 09:05:17 +0000 Subject: [PATCH 0339/1381] net: optimize enqueue_to_backlog() for the fast path Add likely() and unlikely() clauses for the common cases: Device is running. Queue is not full. Queue is less than half capacity. Add max_backlog parameter to skb_flow_limit() to avoid a second READ_ONCE(net_hotdata.max_backlog). skb_flow_limit() does not need the backlog_lock protection, and can be called before we acquire the lock, for even better resistance to attacks. Signed-off-by: Eric Dumazet Reviewed-by: Willem de Bruijn Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20251024090517.3289181-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/core/dev.c | 18 ++++++++++-------- 1 file changed, 10 insertions(+), 8 deletions(-) diff --git a/net/core/dev.c b/net/core/dev.c index 378c2d010faf..d32f0b0c03bb 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -5249,14 +5249,15 @@ void kick_defer_list_purge(unsigned int cpu) int netdev_flow_limit_table_len __read_mostly = (1 << 12); #endif -static bool skb_flow_limit(struct sk_buff *skb, unsigned int qlen) +static bool skb_flow_limit(struct sk_buff *skb, unsigned int qlen, + int max_backlog) { #ifdef CONFIG_NET_FLOW_LIMIT - struct sd_flow_limit *fl; - struct softnet_data *sd; unsigned int old_flow, new_flow; + const struct softnet_data *sd; + struct sd_flow_limit *fl; - if (qlen < (READ_ONCE(net_hotdata.max_backlog) >> 1)) + if (likely(qlen < (max_backlog >> 1))) return false; sd = this_cpu_ptr(&softnet_data); @@ -5301,19 +5302,19 @@ static int enqueue_to_backlog(struct sk_buff *skb, int cpu, u32 tail; reason = SKB_DROP_REASON_DEV_READY; - if (!netif_running(skb->dev)) + if (unlikely(!netif_running(skb->dev))) goto bad_dev; - reason = SKB_DROP_REASON_CPU_BACKLOG; sd = &per_cpu(softnet_data, cpu); qlen = skb_queue_len_lockless(&sd->input_pkt_queue); max_backlog = READ_ONCE(net_hotdata.max_backlog); - if (unlikely(qlen > max_backlog)) + if (unlikely(qlen > max_backlog) || + skb_flow_limit(skb, qlen, max_backlog)) goto cpu_backlog_drop; backlog_lock_irq_save(sd, &flags); qlen = skb_queue_len(&sd->input_pkt_queue); - if (qlen <= max_backlog && !skb_flow_limit(skb, qlen)) { + if (likely(qlen <= max_backlog)) { if (!qlen) { /* Schedule NAPI for backlog device. We can use * non atomic operation as we own the queue lock. @@ -5334,6 +5335,7 @@ static int enqueue_to_backlog(struct sk_buff *skb, int cpu, backlog_unlock_irq_restore(sd, &flags); cpu_backlog_drop: + reason = SKB_DROP_REASON_CPU_BACKLOG; numa_drop_add(&sd->drop_counters, 1); bad_dev: dev_core_stats_rx_dropped_inc(skb->dev); From c72568c21b97dbc48d02b769f4eec6667ad13d5a Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 24 Oct 2025 09:12:40 +0000 Subject: [PATCH 0340/1381] net: rps: softnet_data reorg to make enqueue_to_backlog() fast enqueue_to_backlog() is showing up in kernel profiles on hosts with many cores, when RFS/RPS is used. The following softnet_data fields need to be updated: - input_queue_tail - input_pkt_queue (next, prev, qlen, lock) - backlog.state (if input_pkt_queue was empty) Unfortunately they are currenly using two cache lines: /* --- cacheline 3 boundary (192 bytes) --- */ call_single_data_t csd __attribute__((__aligned__(64))); /* 0xc0 0x20 */ struct softnet_data * rps_ipi_next; /* 0xe0 0x8 */ unsigned int cpu; /* 0xe8 0x4 */ unsigned int input_queue_tail; /* 0xec 0x4 */ struct sk_buff_head input_pkt_queue; /* 0xf0 0x18 */ /* --- cacheline 4 boundary (256 bytes) was 8 bytes ago --- */ struct napi_struct backlog __attribute__((__aligned__(8))); /* 0x108 0x1f0 */ Add one ____cacheline_aligned_in_smp to make sure they now are using a single cache line. Also, because napi_struct has written fields, make @state its first field. We want to make sure that cpus adding packets to sd->input_pkt_queue are not slowing down cpus processing their backlog because of false sharing. After this patch new layout is: /* --- cacheline 5 boundary (320 bytes) --- */ long int pad[3] __attribute__((__aligned__(64))); /* 0x140 0x18 */ unsigned int input_queue_tail; /* 0x158 0x4 */ /* XXX 4 bytes hole, try to pack */ struct sk_buff_head input_pkt_queue; /* 0x160 0x18 */ struct napi_struct backlog __attribute__((__aligned__(8))); /* 0x178 0x1f0 */ Signed-off-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20251024091240.3292546-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- include/linux/netdevice.h | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 7f5aad5cc9a1..9c1e5042c5e7 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -377,6 +377,8 @@ struct napi_config { * Structure for NAPI scheduling similar to tasklet but with weighting */ struct napi_struct { + /* This field should be first or softnet_data.backlog needs tweaks. */ + unsigned long state; /* The poll_list must only be managed by the entity which * changes the state of the NAPI_STATE_SCHED bit. This means * whoever atomically sets that bit can add this napi_struct @@ -385,7 +387,6 @@ struct napi_struct { */ struct list_head poll_list; - unsigned long state; int weight; u32 defer_hard_irqs_count; int (*poll)(struct napi_struct *, int); @@ -3529,9 +3530,17 @@ struct softnet_data { call_single_data_t csd ____cacheline_aligned_in_smp; struct softnet_data *rps_ipi_next; unsigned int cpu; + + /* We force a cacheline alignment from here, to hold together + * input_queue_tail, input_pkt_queue and backlog.state. + * We add holes so that backlog.state is the last field + * of this cache line. + */ + long pad[3] ____cacheline_aligned_in_smp; unsigned int input_queue_tail; #endif struct sk_buff_head input_pkt_queue; + struct napi_struct backlog; struct numa_drop_counters drop_counters; From 8443c3160858b860bfc2db6a8397c72c9f6b513e Mon Sep 17 00:00:00 2001 From: Bobby Eshleman Date: Fri, 24 Oct 2025 11:02:56 -0700 Subject: [PATCH 0341/1381] net: netmem: remove NET_IOV_MAX from net_iov_type enum Remove the NET_IOV_MAX workaround from the net_iov_type enum. This entry was previously added to force the enum size to unsigned long to satisfy the NET_IOV_ASSERT_OFFSET static assertions. After commit f3d85c9ee510 ("netmem: introduce struct netmem_desc mirroring struct page") this approach became unnecessary by placing the net_iov_type after the netmem_desc. Placing the net_iov_type after netmem_desc results in the net_iov_type size having no effect on the position or layout of the fields that mirror the struct page. The layout before this patch: struct net_iov { union { struct netmem_desc desc; /* 0 48 */ struct { long unsigned int _flags; /* 0 8 */ long unsigned int pp_magic; /* 8 8 */ struct page_pool * pp; /* 16 8 */ long unsigned int _pp_mapping_pad; /* 24 8 */ long unsigned int dma_addr; /* 32 8 */ atomic_long_t pp_ref_count; /* 40 8 */ }; /* 0 48 */ }; /* 0 48 */ struct net_iov_area * owner; /* 48 8 */ enum net_iov_type type; /* 56 8 */ /* size: 64, cachelines: 1, members: 3 */ }; The layout after this patch: struct net_iov { union { struct netmem_desc desc; /* 0 48 */ struct { long unsigned int _flags; /* 0 8 */ long unsigned int pp_magic; /* 8 8 */ struct page_pool * pp; /* 16 8 */ long unsigned int _pp_mapping_pad; /* 24 8 */ long unsigned int dma_addr; /* 32 8 */ atomic_long_t pp_ref_count; /* 40 8 */ }; /* 0 48 */ }; /* 0 48 */ struct net_iov_area * owner; /* 48 8 */ enum net_iov_type type; /* 56 4 */ /* size: 64, cachelines: 1, members: 3 */ /* padding: 4 */ }; Signed-off-by: Bobby Eshleman Reviewed-by: Mina Almasry Link: https://patch.msgid.link/20251024-b4-devmem-remove-niov-max-v1-1-ba72c68bc869@meta.com Signed-off-by: Jakub Kicinski --- include/net/netmem.h | 4 ---- 1 file changed, 4 deletions(-) diff --git a/include/net/netmem.h b/include/net/netmem.h index 651e2c62d1dd..9e10f4ac50c3 100644 --- a/include/net/netmem.h +++ b/include/net/netmem.h @@ -68,10 +68,6 @@ DECLARE_STATIC_KEY_FALSE(page_pool_mem_providers); enum net_iov_type { NET_IOV_DMABUF, NET_IOV_IOURING, - - /* Force size to unsigned long to make the NET_IOV_ASSERTS below pass. - */ - NET_IOV_MAX = ULONG_MAX }; /* A memory descriptor representing abstract networking I/O vectors, From 294bfe0343da3b59db040c3a4dac05b4c91ce013 Mon Sep 17 00:00:00 2001 From: Christophe JAILLET Date: Sat, 25 Oct 2025 09:40:59 +0200 Subject: [PATCH 0342/1381] sctp: Constify struct sctp_sched_ops 'struct sctp_sched_ops' is not modified in these drivers. Constifying this structure moves some data to a read-only section, so increases overall security, especially when the structure holds some function pointers. On a x86_64, with allmodconfig, as an example: Before: ====== text data bss dec hex filename 8019 568 0 8587 218b net/sctp/stream_sched_fc.o After: ===== text data bss dec hex filename 8275 312 0 8587 218b net/sctp/stream_sched_fc.o Signed-off-by: Christophe JAILLET Link: https://patch.msgid.link/dce03527eb7b7cc8a3c26d5cdac12bafe3350135.1761377890.git.christophe.jaillet@wanadoo.fr Signed-off-by: Jakub Kicinski --- include/net/sctp/stream_sched.h | 4 ++-- include/net/sctp/structs.h | 2 +- net/sctp/stream.c | 8 ++++---- net/sctp/stream_sched.c | 16 ++++++++-------- net/sctp/stream_sched_fc.c | 4 ++-- net/sctp/stream_sched_prio.c | 2 +- net/sctp/stream_sched_rr.c | 2 +- 7 files changed, 19 insertions(+), 19 deletions(-) diff --git a/include/net/sctp/stream_sched.h b/include/net/sctp/stream_sched.h index 8034bf5febbe..77806ef1cb70 100644 --- a/include/net/sctp/stream_sched.h +++ b/include/net/sctp/stream_sched.h @@ -52,10 +52,10 @@ void sctp_sched_dequeue_done(struct sctp_outq *q, struct sctp_chunk *ch); void sctp_sched_dequeue_common(struct sctp_outq *q, struct sctp_chunk *ch); int sctp_sched_init_sid(struct sctp_stream *stream, __u16 sid, gfp_t gfp); -struct sctp_sched_ops *sctp_sched_ops_from_stream(struct sctp_stream *stream); +const struct sctp_sched_ops *sctp_sched_ops_from_stream(struct sctp_stream *stream); void sctp_sched_ops_register(enum sctp_sched_type sched, - struct sctp_sched_ops *sched_ops); + const struct sctp_sched_ops *sched_ops); void sctp_sched_ops_prio_init(void); void sctp_sched_ops_rr_init(void); void sctp_sched_ops_fc_init(void); diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 3dd304e411d0..5900196d65fd 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -1073,7 +1073,7 @@ struct sctp_outq { struct list_head out_chunk_list; /* Stream scheduler being used */ - struct sctp_sched_ops *sched; + const struct sctp_sched_ops *sched; unsigned int out_qlen; /* Total length of queued data chunks. */ diff --git a/net/sctp/stream.c b/net/sctp/stream.c index f205556c5b24..0615e4426341 100644 --- a/net/sctp/stream.c +++ b/net/sctp/stream.c @@ -54,7 +54,7 @@ static void sctp_stream_shrink_out(struct sctp_stream *stream, __u16 outcnt) static void sctp_stream_free_ext(struct sctp_stream *stream, __u16 sid) { - struct sctp_sched_ops *sched; + const struct sctp_sched_ops *sched; if (!SCTP_SO(stream, sid)->ext) return; @@ -130,7 +130,7 @@ static int sctp_stream_alloc_in(struct sctp_stream *stream, __u16 incnt, int sctp_stream_init(struct sctp_stream *stream, __u16 outcnt, __u16 incnt, gfp_t gfp) { - struct sctp_sched_ops *sched = sctp_sched_ops_from_stream(stream); + const struct sctp_sched_ops *sched = sctp_sched_ops_from_stream(stream); int i, ret = 0; gfp |= __GFP_NOWARN; @@ -182,7 +182,7 @@ int sctp_stream_init_ext(struct sctp_stream *stream, __u16 sid) void sctp_stream_free(struct sctp_stream *stream) { - struct sctp_sched_ops *sched = sctp_sched_ops_from_stream(stream); + const struct sctp_sched_ops *sched = sctp_sched_ops_from_stream(stream); int i; sched->unsched_all(stream); @@ -207,7 +207,7 @@ void sctp_stream_clear(struct sctp_stream *stream) void sctp_stream_update(struct sctp_stream *stream, struct sctp_stream *new) { - struct sctp_sched_ops *sched = sctp_sched_ops_from_stream(stream); + const struct sctp_sched_ops *sched = sctp_sched_ops_from_stream(stream); sched->unsched_all(stream); sctp_stream_outq_migrate(stream, new, new->outcnt); diff --git a/net/sctp/stream_sched.c b/net/sctp/stream_sched.c index 54afbe4fb087..50f8b5240359 100644 --- a/net/sctp/stream_sched.c +++ b/net/sctp/stream_sched.c @@ -91,7 +91,7 @@ static void sctp_sched_fcfs_unsched_all(struct sctp_stream *stream) { } -static struct sctp_sched_ops sctp_sched_fcfs = { +static const struct sctp_sched_ops sctp_sched_fcfs = { .set = sctp_sched_fcfs_set, .get = sctp_sched_fcfs_get, .init = sctp_sched_fcfs_init, @@ -111,10 +111,10 @@ static void sctp_sched_ops_fcfs_init(void) /* API to other parts of the stack */ -static struct sctp_sched_ops *sctp_sched_ops[SCTP_SS_MAX + 1]; +static const struct sctp_sched_ops *sctp_sched_ops[SCTP_SS_MAX + 1]; void sctp_sched_ops_register(enum sctp_sched_type sched, - struct sctp_sched_ops *sched_ops) + const struct sctp_sched_ops *sched_ops) { sctp_sched_ops[sched] = sched_ops; } @@ -130,7 +130,7 @@ void sctp_sched_ops_init(void) static void sctp_sched_free_sched(struct sctp_stream *stream) { - struct sctp_sched_ops *sched = sctp_sched_ops_from_stream(stream); + const struct sctp_sched_ops *sched = sctp_sched_ops_from_stream(stream); struct sctp_stream_out_ext *soute; int i; @@ -148,9 +148,9 @@ static void sctp_sched_free_sched(struct sctp_stream *stream) int sctp_sched_set_sched(struct sctp_association *asoc, enum sctp_sched_type sched) { - struct sctp_sched_ops *old = asoc->outqueue.sched; + const struct sctp_sched_ops *old = asoc->outqueue.sched; struct sctp_datamsg *msg = NULL; - struct sctp_sched_ops *n; + const struct sctp_sched_ops *n; struct sctp_chunk *ch; int i, ret = 0; @@ -263,14 +263,14 @@ void sctp_sched_dequeue_common(struct sctp_outq *q, struct sctp_chunk *ch) int sctp_sched_init_sid(struct sctp_stream *stream, __u16 sid, gfp_t gfp) { - struct sctp_sched_ops *sched = sctp_sched_ops_from_stream(stream); + const struct sctp_sched_ops *sched = sctp_sched_ops_from_stream(stream); struct sctp_stream_out_ext *ext = SCTP_SO(stream, sid)->ext; INIT_LIST_HEAD(&ext->outq); return sched->init_sid(stream, sid, gfp); } -struct sctp_sched_ops *sctp_sched_ops_from_stream(struct sctp_stream *stream) +const struct sctp_sched_ops *sctp_sched_ops_from_stream(struct sctp_stream *stream) { struct sctp_association *asoc; diff --git a/net/sctp/stream_sched_fc.c b/net/sctp/stream_sched_fc.c index 4bd18a497a6d..776c6de46c22 100644 --- a/net/sctp/stream_sched_fc.c +++ b/net/sctp/stream_sched_fc.c @@ -188,7 +188,7 @@ static void sctp_sched_fc_unsched_all(struct sctp_stream *stream) list_del_init(&soute->fc_list); } -static struct sctp_sched_ops sctp_sched_fc = { +static const struct sctp_sched_ops sctp_sched_fc = { .set = sctp_sched_fc_set, .get = sctp_sched_fc_get, .init = sctp_sched_fc_init, @@ -206,7 +206,7 @@ void sctp_sched_ops_fc_init(void) sctp_sched_ops_register(SCTP_SS_FC, &sctp_sched_fc); } -static struct sctp_sched_ops sctp_sched_wfq = { +static const struct sctp_sched_ops sctp_sched_wfq = { .set = sctp_sched_wfq_set, .get = sctp_sched_wfq_get, .init = sctp_sched_fc_init, diff --git a/net/sctp/stream_sched_prio.c b/net/sctp/stream_sched_prio.c index 4d4d9da331f4..fb6c55e5615d 100644 --- a/net/sctp/stream_sched_prio.c +++ b/net/sctp/stream_sched_prio.c @@ -300,7 +300,7 @@ static void sctp_sched_prio_unsched_all(struct sctp_stream *stream) sctp_sched_prio_unsched(soute); } -static struct sctp_sched_ops sctp_sched_prio = { +static const struct sctp_sched_ops sctp_sched_prio = { .set = sctp_sched_prio_set, .get = sctp_sched_prio_get, .init = sctp_sched_prio_init, diff --git a/net/sctp/stream_sched_rr.c b/net/sctp/stream_sched_rr.c index 1f235e7f643a..9157b653f196 100644 --- a/net/sctp/stream_sched_rr.c +++ b/net/sctp/stream_sched_rr.c @@ -171,7 +171,7 @@ static void sctp_sched_rr_unsched_all(struct sctp_stream *stream) sctp_sched_rr_unsched(stream, soute); } -static struct sctp_sched_ops sctp_sched_rr = { +static const struct sctp_sched_ops sctp_sched_rr = { .set = sctp_sched_rr_set, .get = sctp_sched_rr_get, .init = sctp_sched_rr_init, From 5c00da851c31edd58be06a628d6c57684fd0d6a3 Mon Sep 17 00:00:00 2001 From: Rakuram Eswaran Date: Sat, 25 Oct 2025 17:35:18 +0530 Subject: [PATCH 0343/1381] net: tcp_lp: fix kernel-doc warnings and update outdated reference links Fix kernel-doc warnings in tcp_lp.c by adding missing parameter descriptions for tcp_lp_cong_avoid() and tcp_lp_pkts_acked() when building with W=1. Also replace invalid URLs in the file header comment with the currently valid links to the TCP-LP paper and implementation page. No functional changes. Signed-off-by: Rakuram Eswaran Reviewed-by: Simon Horman Link: https://patch.msgid.link/20251025-net_ipv4_tcp_lp_c-v1-1-058cc221499e@gmail.com Signed-off-by: Jakub Kicinski --- net/ipv4/tcp_lp.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/net/ipv4/tcp_lp.c b/net/ipv4/tcp_lp.c index 52fe17167460..976b56644a8a 100644 --- a/net/ipv4/tcp_lp.c +++ b/net/ipv4/tcp_lp.c @@ -23,9 +23,9 @@ * Original Author: * Aleksandar Kuzmanovic * Available from: - * http://www.ece.rice.edu/~akuzma/Doc/akuzma/TCP-LP.pdf + * https://users.cs.northwestern.edu/~akuzma/doc/TCP-LP-ToN.pdf * Original implementation for 2.4.19: - * http://www-ece.rice.edu/networks/TCP-LP/ + * https://users.cs.northwestern.edu/~akuzma/rice/TCP-LP/linux/tcp-lp-linux.htm * * 2.6.x module Authors: * Wong Hoi Sing, Edison @@ -113,6 +113,8 @@ static void tcp_lp_init(struct sock *sk) /** * tcp_lp_cong_avoid * @sk: socket to avoid congesting + * @ack: current ack sequence number + * @acked: number of ACKed packets * * Implementation of cong_avoid. * Will only call newReno CA when away from inference. @@ -261,6 +263,7 @@ static void tcp_lp_rtt_sample(struct sock *sk, u32 rtt) /** * tcp_lp_pkts_acked * @sk: socket requiring congestion avoidance calculations + * @sample: ACK sample containing timing and rate information * * Implementation of pkts_acked. * Deal with active drop under Early Congestion Indication. From 61958b33ef0bab1c1874c933cd3910f495526782 Mon Sep 17 00:00:00 2001 From: Issam Hamdi Date: Fri, 24 Oct 2025 11:49:00 +0200 Subject: [PATCH 0344/1381] net: phy: realtek: Add RTL8224 cable testing support The RTL8224 can detect open pairs and short types (in same pair or some other pair). The distance to this problem can be estimated. This is done for each of the 4 pairs separately. It is not meant to be run while there is an active link partner because this interferes with the active test pulses. Output with open 50 m cable: Pair A code Open Circuit, source: TDR Pair A, fault length: 51.79m, source: TDR Pair B code Open Circuit, source: TDR Pair B, fault length: 51.28m, source: TDR Pair C code Open Circuit, source: TDR Pair C, fault length: 50.46m, source: TDR Pair D code Open Circuit, source: TDR Pair D, fault length: 51.12m, source: TDR Terminated cable: Pair A code OK, source: TDR Pair B code OK, source: TDR Pair C code OK, source: TDR Pair D code OK, source: TDR Shorted cable (both short types are at roughly the same distance) Pair A code Short to another pair, source: TDR Pair A, fault length: 2.35m, source: TDR Pair B code Short to another pair, source: TDR Pair B, fault length: 2.15m, source: TDR Pair C code OK, source: TDR Pair D code Short within Pair, source: TDR Pair D, fault length: 1.94m, source: TDR Signed-off-by: Issam Hamdi Co-developed-by: Sven Eckelmann Signed-off-by: Sven Eckelmann Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20251024-rtl8224-cable-test-v1-1-e3cda89ac98f@simonwunderlich.de Signed-off-by: Jakub Kicinski --- drivers/net/phy/realtek/realtek_main.c | 187 +++++++++++++++++++++++++ 1 file changed, 187 insertions(+) diff --git a/drivers/net/phy/realtek/realtek_main.c b/drivers/net/phy/realtek/realtek_main.c index 16a347084293..1fd4b6cf5c1e 100644 --- a/drivers/net/phy/realtek/realtek_main.c +++ b/drivers/net/phy/realtek/realtek_main.c @@ -8,6 +8,7 @@ * Copyright (c) 2004 Freescale Semiconductor, Inc. */ #include +#include #include #include #include @@ -127,6 +128,27 @@ */ #define RTL822X_VND2_C22_REG(reg) (0xa400 + 2 * (reg)) +#define RTL8224_MII_RTCT 0x11 +#define RTL8224_MII_RTCT_ENABLE BIT(0) +#define RTL8224_MII_RTCT_PAIR_A BIT(4) +#define RTL8224_MII_RTCT_PAIR_B BIT(5) +#define RTL8224_MII_RTCT_PAIR_C BIT(6) +#define RTL8224_MII_RTCT_PAIR_D BIT(7) +#define RTL8224_MII_RTCT_DONE BIT(15) + +#define RTL8224_MII_SRAM_ADDR 0x1b +#define RTL8224_MII_SRAM_DATA 0x1c + +#define RTL8224_SRAM_RTCT_FAULT(pair) (0x8026 + (pair) * 4) +#define RTL8224_SRAM_RTCT_FAULT_BUSY BIT(0) +#define RTL8224_SRAM_RTCT_FAULT_OPEN BIT(3) +#define RTL8224_SRAM_RTCT_FAULT_SAME_SHORT BIT(4) +#define RTL8224_SRAM_RTCT_FAULT_OK BIT(5) +#define RTL8224_SRAM_RTCT_FAULT_DONE BIT(6) +#define RTL8224_SRAM_RTCT_FAULT_CROSS_SHORT BIT(7) + +#define RTL8224_SRAM_RTCT_LEN(pair) (0x8028 + (pair) * 4) + #define RTL8366RB_POWER_SAVE 0x15 #define RTL8366RB_POWER_SAVE_ON BIT(12) @@ -1453,6 +1475,168 @@ static int rtl822xb_c45_read_status(struct phy_device *phydev) return 0; } +static int rtl8224_cable_test_start(struct phy_device *phydev) +{ + u32 val; + int ret; + + /* disable auto-negotiation and force 1000/Full */ + ret = phy_modify_mmd(phydev, MDIO_MMD_VEND2, + RTL822X_VND2_C22_REG(MII_BMCR), + BMCR_ANENABLE | BMCR_SPEED100 | BMCR_SPEED10, + BMCR_SPEED1000 | BMCR_FULLDPLX); + if (ret) + return ret; + + mdelay(500); + + /* trigger cable test */ + val = RTL8224_MII_RTCT_ENABLE; + val |= RTL8224_MII_RTCT_PAIR_A; + val |= RTL8224_MII_RTCT_PAIR_B; + val |= RTL8224_MII_RTCT_PAIR_C; + val |= RTL8224_MII_RTCT_PAIR_D; + + return phy_modify_mmd(phydev, MDIO_MMD_VEND2, + RTL822X_VND2_C22_REG(RTL8224_MII_RTCT), + RTL8224_MII_RTCT_DONE, val); +} + +static int rtl8224_sram_read(struct phy_device *phydev, u32 reg) +{ + int ret; + + ret = phy_write_mmd(phydev, MDIO_MMD_VEND2, + RTL822X_VND2_C22_REG(RTL8224_MII_SRAM_ADDR), + reg); + if (ret) + return ret; + + return phy_read_mmd(phydev, MDIO_MMD_VEND2, + RTL822X_VND2_C22_REG(RTL8224_MII_SRAM_DATA)); +} + +static int rtl8224_pair_len_get(struct phy_device *phydev, u32 pair) +{ + int cable_len; + u32 reg_len; + int ret; + u32 cm; + + reg_len = RTL8224_SRAM_RTCT_LEN(pair); + + ret = rtl8224_sram_read(phydev, reg_len); + if (ret < 0) + return ret; + + cable_len = ret & 0xff00; + + ret = rtl8224_sram_read(phydev, reg_len + 1); + if (ret < 0) + return ret; + + cable_len |= (ret & 0xff00) >> 8; + + cable_len -= 620; + cable_len = max(cable_len, 0); + + cm = cable_len * 100 / 78; + + return cm; +} + +static int rtl8224_cable_test_result_trans(u32 result) +{ + if (!(result & RTL8224_SRAM_RTCT_FAULT_DONE)) + return -EBUSY; + + if (result & RTL8224_SRAM_RTCT_FAULT_OK) + return ETHTOOL_A_CABLE_RESULT_CODE_OK; + + if (result & RTL8224_SRAM_RTCT_FAULT_OPEN) + return ETHTOOL_A_CABLE_RESULT_CODE_OPEN; + + if (result & RTL8224_SRAM_RTCT_FAULT_SAME_SHORT) + return ETHTOOL_A_CABLE_RESULT_CODE_SAME_SHORT; + + if (result & RTL8224_SRAM_RTCT_FAULT_BUSY) + return ETHTOOL_A_CABLE_RESULT_CODE_UNSPEC; + + if (result & RTL8224_SRAM_RTCT_FAULT_CROSS_SHORT) + return ETHTOOL_A_CABLE_RESULT_CODE_CROSS_SHORT; + + return ETHTOOL_A_CABLE_RESULT_CODE_UNSPEC; +} + +static int rtl8224_cable_test_report_pair(struct phy_device *phydev, unsigned int pair) +{ + int fault_rslt; + int ret; + + ret = rtl8224_sram_read(phydev, RTL8224_SRAM_RTCT_FAULT(pair)); + if (ret < 0) + return ret; + + fault_rslt = rtl8224_cable_test_result_trans(ret); + if (fault_rslt < 0) + return 0; + + ret = ethnl_cable_test_result(phydev, pair, fault_rslt); + if (ret < 0) + return ret; + + switch (fault_rslt) { + case ETHTOOL_A_CABLE_RESULT_CODE_OPEN: + case ETHTOOL_A_CABLE_RESULT_CODE_SAME_SHORT: + case ETHTOOL_A_CABLE_RESULT_CODE_CROSS_SHORT: + ret = rtl8224_pair_len_get(phydev, pair); + if (ret < 0) + return ret; + + return ethnl_cable_test_fault_length(phydev, pair, ret); + default: + return 0; + } +} + +static int rtl8224_cable_test_report(struct phy_device *phydev, bool *finished) +{ + unsigned int pair; + int ret; + + for (pair = ETHTOOL_A_CABLE_PAIR_A; pair <= ETHTOOL_A_CABLE_PAIR_D; pair++) { + ret = rtl8224_cable_test_report_pair(phydev, pair); + if (ret == -EBUSY) { + *finished = false; + return 0; + } + + if (ret < 0) + return ret; + } + + return 0; +} + +static int rtl8224_cable_test_get_status(struct phy_device *phydev, bool *finished) +{ + int ret; + + *finished = false; + + ret = phy_read_mmd(phydev, MDIO_MMD_VEND2, + RTL822X_VND2_C22_REG(RTL8224_MII_RTCT)); + if (ret < 0) + return ret; + + if (!(ret & RTL8224_MII_RTCT_DONE)) + return 0; + + *finished = true; + + return rtl8224_cable_test_report(phydev, finished); +} + static bool rtlgen_supports_2_5gbps(struct phy_device *phydev) { int val; @@ -1930,11 +2114,14 @@ static struct phy_driver realtek_drvs[] = { }, { PHY_ID_MATCH_EXACT(0x001ccad0), .name = "RTL8224 2.5Gbps PHY", + .flags = PHY_POLL_CABLE_TEST, .get_features = rtl822x_c45_get_features, .config_aneg = rtl822x_c45_config_aneg, .read_status = rtl822x_c45_read_status, .suspend = genphy_c45_pma_suspend, .resume = rtlgen_c45_resume, + .cable_test_start = rtl8224_cable_test_start, + .cable_test_get_status = rtl8224_cable_test_get_status, }, { PHY_ID_MATCH_EXACT(0x001cc961), .name = "RTL8366RB Gigabit Ethernet", From a8abe8e210c175b1d5a7e53df069e107b65c13cb Mon Sep 17 00:00:00 2001 From: Tianling Shen Date: Sun, 26 Oct 2025 21:36:52 +0800 Subject: [PATCH 0345/1381] net: phy: motorcomm: Add support for PHY LEDs on YT8531 The LED registers on YT8531 are exactly same as YT8521, so simply reuse yt8521_led_hw_* functions. Tested on OrangePi R1 Plus LTS and Zero3. Signed-off-by: Tianling Shen Reviewed-by: Andrew Lunn Reviewed-by: Jijie Shao Link: https://patch.msgid.link/20251026133652.1288732-1-cnsztl@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/phy/motorcomm.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/net/phy/motorcomm.c b/drivers/net/phy/motorcomm.c index a3593e663059..89b5b19a9bd2 100644 --- a/drivers/net/phy/motorcomm.c +++ b/drivers/net/phy/motorcomm.c @@ -3048,6 +3048,9 @@ static struct phy_driver motorcomm_phy_drvs[] = { .get_wol = ytphy_get_wol, .set_wol = yt8531_set_wol, .link_change_notify = yt8531_link_change_notify, + .led_hw_is_supported = yt8521_led_hw_is_supported, + .led_hw_control_set = yt8521_led_hw_control_set, + .led_hw_control_get = yt8521_led_hw_control_get, }, { PHY_ID_MATCH_EXACT(PHY_ID_YT8531S), From 9e314a3c525c91c1e918546f472585cdbd44fbd1 Mon Sep 17 00:00:00 2001 From: Michal Kubiak Date: Thu, 25 Sep 2025 11:22:51 +0200 Subject: [PATCH 0346/1381] ice: remove legacy Rx and construct SKB The commit 53844673d555 ("iavf: kill 'legacy-rx' for good") removed the legacy Rx path in the iavf driver. This change applies the same rationale to the ice driver. The legacy Rx path relied on manual skb allocation and header copying, which has become increasingly inefficient and difficult to maintain. With the stabilization of build_skb() and the growing adoption of features like XDP, page_pool, and multi-buffer support, the legacy approach is no longer viable. Key drawbacks of the legacy path included: - Higher memory pressure due to direct page allocations and splitting; - Redundant memcpy() operations for packet headers; - CPU overhead from eth_get_headlen() and Flow Dissector usage; - Compatibility issues with XDP, which imposes strict headroom and tailroom requirements. The ice driver, like iavf, does not benefit from the minimal headroom savings that legacy Rx once offered, as it already splits pages into fixed halves. Removing this path simplifies the Rx logic, eliminates unnecessary branches in the hotpath, and prepares the driver for upcoming enhancements. In addition to removing the legacy Rx path, this change also eliminates the custom construct_skb() functions from both the standard and zero-copy (ZC) Rx paths. These are replaced with the build_skb() and standardized xdp_build_skb_from_zc() helpers, aligning the driver with the modern XDP infrastructure and reducing code duplication. This cleanup also reduces code complexity and improves maintainability as we move toward a more unified and modern Rx model across drivers. Co-developed-by: Alexander Lobakin Signed-off-by: Alexander Lobakin Reviewed-by: Alexander Lobakin Reviewed-by: Jacob Keller Signed-off-by: Michal Kubiak Tested-by: Alexander Nowlin Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice.h | 1 - drivers/net/ethernet/intel/ice/ice_base.c | 23 +----- drivers/net/ethernet/intel/ice/ice_ethtool.c | 5 -- drivers/net/ethernet/intel/ice/ice_main.c | 11 +-- drivers/net/ethernet/intel/ice/ice_txrx.c | 86 +------------------- drivers/net/ethernet/intel/ice/ice_txrx.h | 16 ---- drivers/net/ethernet/intel/ice/ice_xsk.c | 72 +--------------- 7 files changed, 6 insertions(+), 208 deletions(-) diff --git a/drivers/net/ethernet/intel/ice/ice.h b/drivers/net/ethernet/intel/ice/ice.h index 9ee596773f34..28de4273c2e8 100644 --- a/drivers/net/ethernet/intel/ice/ice.h +++ b/drivers/net/ethernet/intel/ice/ice.h @@ -509,7 +509,6 @@ enum ice_pf_flags { ICE_FLAG_MOD_POWER_UNSUPPORTED, ICE_FLAG_PHY_FW_LOAD_FAILED, ICE_FLAG_ETHTOOL_CTXT, /* set when ethtool holds RTNL lock */ - ICE_FLAG_LEGACY_RX, ICE_FLAG_VF_TRUE_PROMISC_ENA, ICE_FLAG_MDD_AUTO_RESET_VF, ICE_FLAG_VF_VLAN_PRUNING, diff --git a/drivers/net/ethernet/intel/ice/ice_base.c b/drivers/net/ethernet/intel/ice/ice_base.c index 2d35a278c555..b3eb9f512500 100644 --- a/drivers/net/ethernet/intel/ice/ice_base.c +++ b/drivers/net/ethernet/intel/ice/ice_base.c @@ -461,19 +461,6 @@ u16 ice_calc_ts_ring_count(struct ice_tx_ring *tx_ring) return tx_ring->count + max_fetch_desc; } -/** - * ice_rx_offset - Return expected offset into page to access data - * @rx_ring: Ring we are requesting offset of - * - * Returns the offset value for ring into the data buffer. - */ -static unsigned int ice_rx_offset(struct ice_rx_ring *rx_ring) -{ - if (ice_ring_uses_build_skb(rx_ring)) - return ICE_SKB_PAD; - return 0; -} - /** * ice_setup_rx_ctx - Configure a receive ring context * @ring: The Rx ring to configure @@ -586,13 +573,7 @@ static int ice_setup_rx_ctx(struct ice_rx_ring *ring) if (vsi->type == ICE_VSI_VF) return 0; - /* configure Rx buffer alignment */ - if (!vsi->netdev || test_bit(ICE_FLAG_LEGACY_RX, vsi->back->flags)) - ice_clear_ring_build_skb_ena(ring); - else - ice_set_ring_build_skb_ena(ring); - - ring->rx_offset = ice_rx_offset(ring); + ring->rx_offset = ICE_SKB_PAD; /* init queue specific tail register */ ring->tail = hw->hw_addr + QRX_TAIL(pf_q); @@ -753,7 +734,7 @@ int ice_vsi_cfg_single_rxq(struct ice_vsi *vsi, u16 q_idx) */ static void ice_vsi_cfg_frame_size(struct ice_vsi *vsi, struct ice_rx_ring *ring) { - if (!vsi->netdev || test_bit(ICE_FLAG_LEGACY_RX, vsi->back->flags)) { + if (!vsi->netdev) { ring->max_frame = ICE_MAX_FRAME_LEGACY_RX; ring->rx_buf_len = ICE_RXBUF_1664; #if (PAGE_SIZE < 8192) diff --git a/drivers/net/ethernet/intel/ice/ice_ethtool.c b/drivers/net/ethernet/intel/ice/ice_ethtool.c index cb34d4675a78..240e3f35c10a 100644 --- a/drivers/net/ethernet/intel/ice/ice_ethtool.c +++ b/drivers/net/ethernet/intel/ice/ice_ethtool.c @@ -340,7 +340,6 @@ static const struct ice_priv_flag ice_gstrings_priv_flags[] = { ICE_FLAG_VF_TRUE_PROMISC_ENA), ICE_PRIV_FLAG("mdd-auto-reset-vf", ICE_FLAG_MDD_AUTO_RESET_VF), ICE_PRIV_FLAG("vf-vlan-pruning", ICE_FLAG_VF_VLAN_PRUNING), - ICE_PRIV_FLAG("legacy-rx", ICE_FLAG_LEGACY_RX), }; #define ICE_PRIV_FLAG_ARRAY_SIZE ARRAY_SIZE(ice_gstrings_priv_flags) @@ -1856,10 +1855,6 @@ static int ice_set_priv_flags(struct net_device *netdev, u32 flags) ice_nway_reset(netdev); } } - if (test_bit(ICE_FLAG_LEGACY_RX, change_flags)) { - /* down and up VSI so that changes of Rx cfg are reflected. */ - ice_down_up(vsi); - } /* don't allow modification of this flag when a single VF is in * promiscuous mode because it's not supported */ diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c index 1de3da7b3907..36e6a078c84c 100644 --- a/drivers/net/ethernet/intel/ice/ice_main.c +++ b/drivers/net/ethernet/intel/ice/ice_main.c @@ -2957,10 +2957,7 @@ int ice_vsi_determine_xdp_res(struct ice_vsi *vsi) */ static int ice_max_xdp_frame_size(struct ice_vsi *vsi) { - if (test_bit(ICE_FLAG_LEGACY_RX, vsi->back->flags)) - return ICE_RXBUF_1664; - else - return ICE_RXBUF_3072; + return ICE_RXBUF_3072; } /** @@ -7864,12 +7861,6 @@ int ice_change_mtu(struct net_device *netdev, int new_mtu) frame_size - ICE_ETH_PKT_HDR_PAD); return -EINVAL; } - } else if (test_bit(ICE_FLAG_LEGACY_RX, pf->flags)) { - if (new_mtu + ICE_ETH_PKT_HDR_PAD > ICE_MAX_FRAME_LEGACY_RX) { - netdev_err(netdev, "Too big MTU for legacy-rx; Max is %d\n", - ICE_MAX_FRAME_LEGACY_RX - ICE_ETH_PKT_HDR_PAD); - return -EINVAL; - } } /* if a reset is in progress, wait for some time for it to complete */ diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.c b/drivers/net/ethernet/intel/ice/ice_txrx.c index 73f08d02f9c7..5d59ee45d3da 100644 --- a/drivers/net/ethernet/intel/ice/ice_txrx.c +++ b/drivers/net/ethernet/intel/ice/ice_txrx.c @@ -1169,87 +1169,6 @@ ice_build_skb(struct ice_rx_ring *rx_ring, struct xdp_buff *xdp) return skb; } -/** - * ice_construct_skb - Allocate skb and populate it - * @rx_ring: Rx descriptor ring to transact packets on - * @xdp: xdp_buff pointing to the data - * - * This function allocates an skb. It then populates it with the page - * data from the current receive descriptor, taking care to set up the - * skb correctly. - */ -static struct sk_buff * -ice_construct_skb(struct ice_rx_ring *rx_ring, struct xdp_buff *xdp) -{ - unsigned int size = xdp->data_end - xdp->data; - struct skb_shared_info *sinfo = NULL; - struct ice_rx_buf *rx_buf; - unsigned int nr_frags = 0; - unsigned int headlen; - struct sk_buff *skb; - - /* prefetch first cache line of first page */ - net_prefetch(xdp->data); - - if (unlikely(xdp_buff_has_frags(xdp))) { - sinfo = xdp_get_shared_info_from_buff(xdp); - nr_frags = sinfo->nr_frags; - } - - /* allocate a skb to store the frags */ - skb = napi_alloc_skb(&rx_ring->q_vector->napi, ICE_RX_HDR_SIZE); - if (unlikely(!skb)) - return NULL; - - rx_buf = &rx_ring->rx_buf[rx_ring->first_desc]; - skb_record_rx_queue(skb, rx_ring->q_index); - /* Determine available headroom for copy */ - headlen = size; - if (headlen > ICE_RX_HDR_SIZE) - headlen = eth_get_headlen(skb->dev, xdp->data, ICE_RX_HDR_SIZE); - - /* align pull length to size of long to optimize memcpy performance */ - memcpy(__skb_put(skb, headlen), xdp->data, ALIGN(headlen, - sizeof(long))); - - /* if we exhaust the linear part then add what is left as a frag */ - size -= headlen; - if (size) { - /* besides adding here a partial frag, we are going to add - * frags from xdp_buff, make sure there is enough space for - * them - */ - if (unlikely(nr_frags >= MAX_SKB_FRAGS - 1)) { - dev_kfree_skb(skb); - return NULL; - } - skb_add_rx_frag(skb, 0, rx_buf->page, - rx_buf->page_offset + headlen, size, - xdp->frame_sz); - } else { - /* buffer is unused, restore biased page count in Rx buffer; - * data was copied onto skb's linear part so there's no - * need for adjusting page offset and we can reuse this buffer - * as-is - */ - rx_buf->pagecnt_bias++; - } - - if (unlikely(xdp_buff_has_frags(xdp))) { - struct skb_shared_info *skinfo = skb_shinfo(skb); - - memcpy(&skinfo->frags[skinfo->nr_frags], &sinfo->frags[0], - sizeof(skb_frag_t) * nr_frags); - - xdp_update_skb_frags_info(skb, skinfo->nr_frags + nr_frags, - sinfo->xdp_frags_size, - nr_frags * xdp->frame_sz, - xdp_buff_get_skb_flags(xdp)); - } - - return skb; -} - /** * ice_put_rx_buf - Clean up used buffer and either recycle or free * @rx_ring: Rx descriptor ring to transact packets on @@ -1464,10 +1383,7 @@ static int ice_clean_rx_irq(struct ice_rx_ring *rx_ring, int budget) continue; construct_skb: - if (likely(ice_ring_uses_build_skb(rx_ring))) - skb = ice_build_skb(rx_ring, xdp); - else - skb = ice_construct_skb(rx_ring, xdp); + skb = ice_build_skb(rx_ring, xdp); /* exit if we failed to retrieve a buffer */ if (!skb) { rx_ring->ring_stats->rx_stats.alloc_buf_failed++; diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.h b/drivers/net/ethernet/intel/ice/ice_txrx.h index 841a07bfba54..427f672fe053 100644 --- a/drivers/net/ethernet/intel/ice/ice_txrx.h +++ b/drivers/net/ethernet/intel/ice/ice_txrx.h @@ -373,7 +373,6 @@ struct ice_rx_ring { dma_addr_t dma; /* physical address of ring */ u8 dcb_tc; /* Traffic class of ring */ u8 ptp_rx; -#define ICE_RX_FLAGS_RING_BUILD_SKB BIT(1) #define ICE_RX_FLAGS_CRC_STRIP_DIS BIT(2) #define ICE_RX_FLAGS_MULTIDEV BIT(3) #define ICE_RX_FLAGS_RING_GCS BIT(4) @@ -422,21 +421,6 @@ struct ice_tx_ring { u16 quanta_prof_id; } ____cacheline_internodealigned_in_smp; -static inline bool ice_ring_uses_build_skb(struct ice_rx_ring *ring) -{ - return !!(ring->flags & ICE_RX_FLAGS_RING_BUILD_SKB); -} - -static inline void ice_set_ring_build_skb_ena(struct ice_rx_ring *ring) -{ - ring->flags |= ICE_RX_FLAGS_RING_BUILD_SKB; -} - -static inline void ice_clear_ring_build_skb_ena(struct ice_rx_ring *ring) -{ - ring->flags &= ~ICE_RX_FLAGS_RING_BUILD_SKB; -} - static inline bool ice_ring_ch_enabled(struct ice_tx_ring *ring) { return !!ring->ch; diff --git a/drivers/net/ethernet/intel/ice/ice_xsk.c b/drivers/net/ethernet/intel/ice/ice_xsk.c index 575fd48f485f..b25bc5ba40ab 100644 --- a/drivers/net/ethernet/intel/ice/ice_xsk.c +++ b/drivers/net/ethernet/intel/ice/ice_xsk.c @@ -392,69 +392,6 @@ bool ice_alloc_rx_bufs_zc(struct ice_rx_ring *rx_ring, return __ice_alloc_rx_bufs_zc(rx_ring, xsk_pool, leftover); } -/** - * ice_construct_skb_zc - Create an sk_buff from zero-copy buffer - * @rx_ring: Rx ring - * @xdp: Pointer to XDP buffer - * - * This function allocates a new skb from a zero-copy Rx buffer. - * - * Returns the skb on success, NULL on failure. - */ -static struct sk_buff * -ice_construct_skb_zc(struct ice_rx_ring *rx_ring, struct xdp_buff *xdp) -{ - unsigned int totalsize = xdp->data_end - xdp->data_meta; - unsigned int metasize = xdp->data - xdp->data_meta; - struct skb_shared_info *sinfo = NULL; - struct sk_buff *skb; - u32 nr_frags = 0; - - if (unlikely(xdp_buff_has_frags(xdp))) { - sinfo = xdp_get_shared_info_from_buff(xdp); - nr_frags = sinfo->nr_frags; - } - net_prefetch(xdp->data_meta); - - skb = napi_alloc_skb(&rx_ring->q_vector->napi, totalsize); - if (unlikely(!skb)) - return NULL; - - memcpy(__skb_put(skb, totalsize), xdp->data_meta, - ALIGN(totalsize, sizeof(long))); - - if (metasize) { - skb_metadata_set(skb, metasize); - __skb_pull(skb, metasize); - } - - if (likely(!xdp_buff_has_frags(xdp))) - goto out; - - for (int i = 0; i < nr_frags; i++) { - struct skb_shared_info *skinfo = skb_shinfo(skb); - skb_frag_t *frag = &sinfo->frags[i]; - struct page *page; - void *addr; - - page = dev_alloc_page(); - if (!page) { - dev_kfree_skb(skb); - return NULL; - } - addr = page_to_virt(page); - - memcpy(addr, skb_frag_page(frag), skb_frag_size(frag)); - - __skb_fill_page_desc_noacc(skinfo, skinfo->nr_frags++, - addr, 0, skb_frag_size(frag)); - } - -out: - xsk_buff_free(xdp); - return skb; -} - /** * ice_clean_xdp_irq_zc - produce AF_XDP descriptors to CQ * @xdp_ring: XDP Tx ring @@ -757,20 +694,15 @@ int ice_clean_rx_irq_zc(struct ice_rx_ring *rx_ring, construct_skb: /* XDP_PASS path */ - skb = ice_construct_skb_zc(rx_ring, first); + skb = xdp_build_skb_from_zc(first); if (!skb) { + xsk_buff_free(first); rx_ring->ring_stats->rx_stats.alloc_buf_failed++; break; } first = NULL; rx_ring->first_desc = ntc; - - if (eth_skb_pad(skb)) { - skb = NULL; - continue; - } - total_rx_bytes += skb->len; total_rx_packets++; From 3a4f419f750946181e3d6a339a1ef1942c5b5685 Mon Sep 17 00:00:00 2001 From: Michal Kubiak Date: Thu, 25 Sep 2025 11:22:52 +0200 Subject: [PATCH 0347/1381] ice: drop page splitting and recycling As part of the transition toward Page Pool integration, remove the legacy page splitting and recycling logic from the ice driver. This mirrors the approach taken in commit 920d86f3c552 ("iavf: drop page splitting and recycling"). The previous model attempted to reuse partially consumed pages by splitting them and tracking their usage across descriptors. While this was once a memory optimization, it introduced significant complexity and overhead in the Rx path, including: - Manual refcount management and page reuse heuristics; - Per-descriptor buffer shuffling, which could involve moving dozens of `ice_rx_buf` structures per NAPI cycle; - Increased branching and cache pressure in the hotpath. This change simplifies the Rx logic by always allocating fresh pages and letting the networking stack handle their lifecycle. Although this may temporarily reduce performance (up to ~98% in some XDP cases), it greatly improves maintainability and paves the way for Page Pool, which will restore and exceed previous performance levels. The `ice_rx_buf` array is retained for now to minimize diffstat and ease future replacement with a shared buffer abstraction. Co-developed-by: Alexander Lobakin Signed-off-by: Alexander Lobakin Reviewed-by: Alexander Lobakin Reviewed-by: Jacob Keller Signed-off-by: Michal Kubiak Tested-by: Alexander Nowlin Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice.h | 2 + drivers/net/ethernet/intel/ice/ice_base.c | 29 ++-- drivers/net/ethernet/intel/ice/ice_txrx.c | 139 ++----------------- drivers/net/ethernet/intel/ice/ice_txrx.h | 72 ---------- drivers/net/ethernet/intel/ice/virt/queues.c | 5 +- 5 files changed, 24 insertions(+), 223 deletions(-) diff --git a/drivers/net/ethernet/intel/ice/ice.h b/drivers/net/ethernet/intel/ice/ice.h index 28de4273c2e8..3d4d8b88631b 100644 --- a/drivers/net/ethernet/intel/ice/ice.h +++ b/drivers/net/ethernet/intel/ice/ice.h @@ -374,6 +374,8 @@ struct ice_vsi { spinlock_t arfs_lock; /* protects aRFS hash table and filter state */ atomic_t *arfs_last_fltr_id; + u16 max_frame; + struct ice_aqc_vsi_props info; /* VSI properties */ struct ice_vsi_vlan_info vlan_info; /* vlan config to be restored */ diff --git a/drivers/net/ethernet/intel/ice/ice_base.c b/drivers/net/ethernet/intel/ice/ice_base.c index b3eb9f512500..fee58f879d9e 100644 --- a/drivers/net/ethernet/intel/ice/ice_base.c +++ b/drivers/net/ethernet/intel/ice/ice_base.c @@ -495,7 +495,7 @@ static int ice_setup_rx_ctx(struct ice_rx_ring *ring) /* Receive Packet Data Buffer Size. * The Packet Data Buffer Size is defined in 128 byte units. */ - rlan_ctx.dbuf = DIV_ROUND_UP(ring->rx_buf_len, + rlan_ctx.dbuf = DIV_ROUND_UP(ICE_RXBUF_3072, BIT_ULL(ICE_RLAN_CTX_DBUF_S)); /* use 32 byte descriptors */ @@ -536,8 +536,8 @@ static int ice_setup_rx_ctx(struct ice_rx_ring *ring) /* Max packet size for this queue - must not be set to a larger value * than 5 x DBUF */ - rlan_ctx.rxmax = min_t(u32, ring->max_frame, - ICE_MAX_CHAINED_RX_BUFS * ring->rx_buf_len); + rlan_ctx.rxmax = min_t(u32, vsi->max_frame, + ICE_MAX_CHAINED_RX_BUFS * ICE_RXBUF_3072); /* Rx queue threshold in units of 64 */ rlan_ctx.lrxqthresh = 1; @@ -608,7 +608,7 @@ static unsigned int ice_get_frame_sz(struct ice_rx_ring *rx_ring) #if (PAGE_SIZE >= 8192) frame_sz = rx_ring->rx_buf_len; #else - frame_sz = ice_rx_pg_size(rx_ring) / 2; + frame_sz = PAGE_SIZE; #endif return frame_sz; @@ -624,6 +624,7 @@ static int ice_vsi_cfg_rxq(struct ice_rx_ring *ring) { struct device *dev = ice_pf_to_dev(ring->vsi->back); u32 num_bufs = ICE_RX_DESC_UNUSED(ring); + u32 rx_buf_len; int err; if (ring->vsi->type == ICE_VSI_PF || ring->vsi->type == ICE_VSI_SF) { @@ -631,7 +632,7 @@ static int ice_vsi_cfg_rxq(struct ice_rx_ring *ring) err = __xdp_rxq_info_reg(&ring->xdp_rxq, ring->netdev, ring->q_index, ring->q_vector->napi.napi_id, - ring->rx_buf_len); + ICE_RXBUF_3072); if (err) return err; } @@ -640,12 +641,12 @@ static int ice_vsi_cfg_rxq(struct ice_rx_ring *ring) if (ring->xsk_pool) { xdp_rxq_info_unreg(&ring->xdp_rxq); - ring->rx_buf_len = + rx_buf_len = xsk_pool_get_rx_frame_size(ring->xsk_pool); err = __xdp_rxq_info_reg(&ring->xdp_rxq, ring->netdev, ring->q_index, ring->q_vector->napi.napi_id, - ring->rx_buf_len); + rx_buf_len); if (err) return err; err = xdp_rxq_info_reg_mem_model(&ring->xdp_rxq, @@ -663,7 +664,7 @@ static int ice_vsi_cfg_rxq(struct ice_rx_ring *ring) err = __xdp_rxq_info_reg(&ring->xdp_rxq, ring->netdev, ring->q_index, ring->q_vector->napi.napi_id, - ring->rx_buf_len); + ICE_RXBUF_3072); if (err) return err; } @@ -735,17 +736,9 @@ int ice_vsi_cfg_single_rxq(struct ice_vsi *vsi, u16 q_idx) static void ice_vsi_cfg_frame_size(struct ice_vsi *vsi, struct ice_rx_ring *ring) { if (!vsi->netdev) { - ring->max_frame = ICE_MAX_FRAME_LEGACY_RX; - ring->rx_buf_len = ICE_RXBUF_1664; -#if (PAGE_SIZE < 8192) - } else if (!ICE_2K_TOO_SMALL_WITH_PADDING && - (vsi->netdev->mtu <= ETH_DATA_LEN)) { - ring->max_frame = ICE_RXBUF_1536 - NET_IP_ALIGN; - ring->rx_buf_len = ICE_RXBUF_1536 - NET_IP_ALIGN; -#endif + vsi->max_frame = ICE_MAX_FRAME_LEGACY_RX; } else { - ring->max_frame = ICE_AQ_SET_MAC_FRAME_SIZE_MAX; - ring->rx_buf_len = ICE_RXBUF_3072; + vsi->max_frame = ICE_AQ_SET_MAC_FRAME_SIZE_MAX; } } diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.c b/drivers/net/ethernet/intel/ice/ice_txrx.c index 5d59ee45d3da..2b46e4c8be86 100644 --- a/drivers/net/ethernet/intel/ice/ice_txrx.c +++ b/drivers/net/ethernet/intel/ice/ice_txrx.c @@ -543,13 +543,13 @@ void ice_clean_rx_ring(struct ice_rx_ring *rx_ring) */ dma_sync_single_range_for_cpu(dev, rx_buf->dma, rx_buf->page_offset, - rx_ring->rx_buf_len, + ICE_RXBUF_3072, DMA_FROM_DEVICE); /* free resources associated with mapping */ - dma_unmap_page_attrs(dev, rx_buf->dma, ice_rx_pg_size(rx_ring), + dma_unmap_page_attrs(dev, rx_buf->dma, PAGE_SIZE, DMA_FROM_DEVICE, ICE_RX_DMA_ATTR); - __page_frag_cache_drain(rx_buf->page, rx_buf->pagecnt_bias); + __free_page(rx_buf->page); rx_buf->page = NULL; rx_buf->page_offset = 0; @@ -803,10 +803,6 @@ ice_alloc_mapped_page(struct ice_rx_ring *rx_ring, struct ice_rx_buf *bi) struct page *page = bi->page; dma_addr_t dma; - /* since we are recycling buffers we should seldom need to alloc */ - if (likely(page)) - return true; - /* alloc new page for storage */ page = dev_alloc_pages(ice_rx_pg_order(rx_ring)); if (unlikely(!page)) { @@ -815,7 +811,7 @@ ice_alloc_mapped_page(struct ice_rx_ring *rx_ring, struct ice_rx_buf *bi) } /* map page for use */ - dma = dma_map_page_attrs(rx_ring->dev, page, 0, ice_rx_pg_size(rx_ring), + dma = dma_map_page_attrs(rx_ring->dev, page, 0, PAGE_SIZE, DMA_FROM_DEVICE, ICE_RX_DMA_ATTR); /* if mapping failed free memory back to system since @@ -831,7 +827,6 @@ ice_alloc_mapped_page(struct ice_rx_ring *rx_ring, struct ice_rx_buf *bi) bi->page = page; bi->page_offset = rx_ring->rx_offset; page_ref_add(page, USHRT_MAX - 1); - bi->pagecnt_bias = USHRT_MAX; return true; } @@ -902,7 +897,7 @@ bool ice_alloc_rx_bufs(struct ice_rx_ring *rx_ring, unsigned int cleaned_count) /* sync the buffer for use by the device */ dma_sync_single_range_for_device(rx_ring->dev, bi->dma, bi->page_offset, - rx_ring->rx_buf_len, + ICE_RXBUF_3072, DMA_FROM_DEVICE); /* Refresh the desc even if buffer_addrs didn't change @@ -931,69 +926,6 @@ bool ice_alloc_rx_bufs(struct ice_rx_ring *rx_ring, unsigned int cleaned_count) return !!cleaned_count; } -/** - * ice_rx_buf_adjust_pg_offset - Prepare Rx buffer for reuse - * @rx_buf: Rx buffer to adjust - * @size: Size of adjustment - * - * Update the offset within page so that Rx buf will be ready to be reused. - * For systems with PAGE_SIZE < 8192 this function will flip the page offset - * so the second half of page assigned to Rx buffer will be used, otherwise - * the offset is moved by "size" bytes - */ -static void -ice_rx_buf_adjust_pg_offset(struct ice_rx_buf *rx_buf, unsigned int size) -{ -#if (PAGE_SIZE < 8192) - /* flip page offset to other buffer */ - rx_buf->page_offset ^= size; -#else - /* move offset up to the next cache line */ - rx_buf->page_offset += size; -#endif -} - -/** - * ice_can_reuse_rx_page - Determine if page can be reused for another Rx - * @rx_buf: buffer containing the page - * - * If page is reusable, we have a green light for calling ice_reuse_rx_page, - * which will assign the current buffer to the buffer that next_to_alloc is - * pointing to; otherwise, the DMA mapping needs to be destroyed and - * page freed - */ -static bool -ice_can_reuse_rx_page(struct ice_rx_buf *rx_buf) -{ - unsigned int pagecnt_bias = rx_buf->pagecnt_bias; - struct page *page = rx_buf->page; - - /* avoid re-using remote and pfmemalloc pages */ - if (!dev_page_is_reusable(page)) - return false; - - /* if we are only owner of page we can reuse it */ - if (unlikely(rx_buf->pgcnt - pagecnt_bias > 1)) - return false; -#if (PAGE_SIZE >= 8192) -#define ICE_LAST_OFFSET \ - (SKB_WITH_OVERHEAD(PAGE_SIZE) - ICE_RXBUF_3072) - if (rx_buf->page_offset > ICE_LAST_OFFSET) - return false; -#endif /* PAGE_SIZE >= 8192) */ - - /* If we have drained the page fragment pool we need to update - * the pagecnt_bias and page count so that we fully restock the - * number of references the driver holds. - */ - if (unlikely(pagecnt_bias == 1)) { - page_ref_add(page, USHRT_MAX - 1); - rx_buf->pagecnt_bias = USHRT_MAX; - } - - return true; -} - /** * ice_add_xdp_frag - Add contents of Rx buffer to xdp buf as a frag * @rx_ring: Rx descriptor ring to transact packets on @@ -1032,35 +964,6 @@ ice_add_xdp_frag(struct ice_rx_ring *rx_ring, struct xdp_buff *xdp, return 0; } -/** - * ice_reuse_rx_page - page flip buffer and store it back on the ring - * @rx_ring: Rx descriptor ring to store buffers on - * @old_buf: donor buffer to have page reused - * - * Synchronizes page for reuse by the adapter - */ -static void -ice_reuse_rx_page(struct ice_rx_ring *rx_ring, struct ice_rx_buf *old_buf) -{ - u16 nta = rx_ring->next_to_alloc; - struct ice_rx_buf *new_buf; - - new_buf = &rx_ring->rx_buf[nta]; - - /* update, and store next to alloc */ - nta++; - rx_ring->next_to_alloc = (nta < rx_ring->count) ? nta : 0; - - /* Transfer page from old buffer to new buffer. - * Move each member individually to avoid possible store - * forwarding stalls and unnecessary copy of skb. - */ - new_buf->dma = old_buf->dma; - new_buf->page = old_buf->page; - new_buf->page_offset = old_buf->page_offset; - new_buf->pagecnt_bias = old_buf->pagecnt_bias; -} - /** * ice_get_rx_buf - Fetch Rx buffer and synchronize data for use * @rx_ring: Rx descriptor ring to transact packets on @@ -1086,9 +989,6 @@ ice_get_rx_buf(struct ice_rx_ring *rx_ring, const unsigned int size, rx_buf->page_offset, size, DMA_FROM_DEVICE); - /* We have pulled a buffer for use, so decrement pagecnt_bias */ - rx_buf->pagecnt_bias--; - return rx_buf; } @@ -1183,16 +1083,10 @@ ice_put_rx_buf(struct ice_rx_ring *rx_ring, struct ice_rx_buf *rx_buf) if (!rx_buf) return; - if (ice_can_reuse_rx_page(rx_buf)) { - /* hand second half of page back to the ring */ - ice_reuse_rx_page(rx_ring, rx_buf); - } else { - /* we are not reusing the buffer so unmap it */ - dma_unmap_page_attrs(rx_ring->dev, rx_buf->dma, - ice_rx_pg_size(rx_ring), DMA_FROM_DEVICE, - ICE_RX_DMA_ATTR); - __page_frag_cache_drain(rx_buf->page, rx_buf->pagecnt_bias); - } + /* we are not reusing the buffer so unmap it */ + dma_unmap_page_attrs(rx_ring->dev, rx_buf->dma, + PAGE_SIZE, DMA_FROM_DEVICE, + ICE_RX_DMA_ATTR); /* clear contents of buffer_info */ rx_buf->page = NULL; @@ -1218,27 +1112,12 @@ static void ice_put_rx_mbuf(struct ice_rx_ring *rx_ring, struct xdp_buff *xdp, u32 idx = rx_ring->first_desc; u32 cnt = rx_ring->count; struct ice_rx_buf *buf; - u32 xdp_frags = 0; - int i = 0; - - if (unlikely(xdp_buff_has_frags(xdp))) - xdp_frags = xdp_get_shared_info_from_buff(xdp)->nr_frags; while (idx != ntc) { buf = &rx_ring->rx_buf[idx]; if (++idx == cnt) idx = 0; - /* An XDP program could release fragments from the end of the - * buffer. For these, we need to keep the pagecnt_bias as-is. - * To do this, only adjust pagecnt_bias for fragments up to - * the total remaining after the XDP program has run. - */ - if (verdict != ICE_XDP_CONSUMED) - ice_rx_buf_adjust_pg_offset(buf, xdp->frame_sz); - else if (i++ <= xdp_frags) - buf->pagecnt_bias++; - ice_put_rx_buf(rx_ring, buf); } diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.h b/drivers/net/ethernet/intel/ice/ice_txrx.h index 427f672fe053..3c7830f787de 100644 --- a/drivers/net/ethernet/intel/ice/ice_txrx.h +++ b/drivers/net/ethernet/intel/ice/ice_txrx.h @@ -27,71 +27,7 @@ #define ICE_MAX_TXQ_PER_TXQG 128 -/* Attempt to maximize the headroom available for incoming frames. We use a 2K - * buffer for MTUs <= 1500 and need 1536/1534 to store the data for the frame. - * This leaves us with 512 bytes of room. From that we need to deduct the - * space needed for the shared info and the padding needed to IP align the - * frame. - * - * Note: For cache line sizes 256 or larger this value is going to end - * up negative. In these cases we should fall back to the legacy - * receive path. - */ -#if (PAGE_SIZE < 8192) -#define ICE_2K_TOO_SMALL_WITH_PADDING \ - ((unsigned int)(NET_SKB_PAD + ICE_RXBUF_1536) > \ - SKB_WITH_OVERHEAD(ICE_RXBUF_2048)) - -/** - * ice_compute_pad - compute the padding - * @rx_buf_len: buffer length - * - * Figure out the size of half page based on given buffer length and - * then subtract the skb_shared_info followed by subtraction of the - * actual buffer length; this in turn results in the actual space that - * is left for padding usage - */ -static inline int ice_compute_pad(int rx_buf_len) -{ - int half_page_size; - - half_page_size = ALIGN(rx_buf_len, PAGE_SIZE / 2); - return SKB_WITH_OVERHEAD(half_page_size) - rx_buf_len; -} - -/** - * ice_skb_pad - determine the padding that we can supply - * - * Figure out the right Rx buffer size and based on that calculate the - * padding - */ -static inline int ice_skb_pad(void) -{ - int rx_buf_len; - - /* If a 2K buffer cannot handle a standard Ethernet frame then - * optimize padding for a 3K buffer instead of a 1.5K buffer. - * - * For a 3K buffer we need to add enough padding to allow for - * tailroom due to NET_IP_ALIGN possibly shifting us out of - * cache-line alignment. - */ - if (ICE_2K_TOO_SMALL_WITH_PADDING) - rx_buf_len = ICE_RXBUF_3072 + SKB_DATA_ALIGN(NET_IP_ALIGN); - else - rx_buf_len = ICE_RXBUF_1536; - - /* if needed make room for NET_IP_ALIGN */ - rx_buf_len -= NET_IP_ALIGN; - - return ice_compute_pad(rx_buf_len); -} - -#define ICE_SKB_PAD ice_skb_pad() -#else -#define ICE_2K_TOO_SMALL_WITH_PADDING false #define ICE_SKB_PAD (NET_SKB_PAD + NET_IP_ALIGN) -#endif /* We are assuming that the cache line is always 64 Bytes here for ice. * In order to make sure that is a correct assumption there is a check in probe @@ -202,7 +138,6 @@ struct ice_rx_buf { struct page *page; unsigned int page_offset; unsigned int pgcnt; - unsigned int pagecnt_bias; }; struct ice_q_stats { @@ -368,7 +303,6 @@ struct ice_rx_ring { struct ice_tx_ring *xdp_ring; struct ice_rx_ring *next; /* pointer to next ring in q_vector */ struct xsk_buff_pool *xsk_pool; - u16 max_frame; u16 rx_buf_len; dma_addr_t dma; /* physical address of ring */ u8 dcb_tc; /* Traffic class of ring */ @@ -475,15 +409,9 @@ struct ice_coalesce_stored { static inline unsigned int ice_rx_pg_order(struct ice_rx_ring *ring) { -#if (PAGE_SIZE < 8192) - if (ring->rx_buf_len > (PAGE_SIZE / 2)) - return 1; -#endif return 0; } -#define ice_rx_pg_size(_ring) (PAGE_SIZE << ice_rx_pg_order(_ring)) - union ice_32b_rx_flex_desc; void ice_init_ctrl_rx_descs(struct ice_rx_ring *rx_ring, u32 num_descs); diff --git a/drivers/net/ethernet/intel/ice/virt/queues.c b/drivers/net/ethernet/intel/ice/virt/queues.c index 370f6ec2a374..7928f4e8e788 100644 --- a/drivers/net/ethernet/intel/ice/virt/queues.c +++ b/drivers/net/ethernet/intel/ice/virt/queues.c @@ -842,18 +842,17 @@ int ice_vc_cfg_qs_msg(struct ice_vf *vf, u8 *msg) (qpi->rxq.databuffer_size > ((16 * 1024) - 128) || qpi->rxq.databuffer_size < 1024)) goto error_param; - ring->rx_buf_len = qpi->rxq.databuffer_size; if (qpi->rxq.max_pkt_size > max_frame_size || qpi->rxq.max_pkt_size < 64) goto error_param; - ring->max_frame = qpi->rxq.max_pkt_size; + vsi->max_frame = qpi->rxq.max_pkt_size; /* add space for the port VLAN since the VF driver is * not expected to account for it in the MTU * calculation */ if (ice_vf_is_port_vlan_ena(vf)) - ring->max_frame += VLAN_HLEN; + vsi->max_frame += VLAN_HLEN; if (ice_vsi_cfg_single_rxq(vsi, q_idx)) { dev_warn(ice_pf_to_dev(pf), "VF-%d failed to configure RX queue %d\n", From 93f53db9f9dc4a16b40ecd18e6d338ad57e4b670 Mon Sep 17 00:00:00 2001 From: Michal Kubiak Date: Thu, 25 Sep 2025 11:22:53 +0200 Subject: [PATCH 0348/1381] ice: switch to Page Pool This patch completes the transition of the ice driver to use the Page Pool and libeth APIs, following the same direction as commit 5fa4caff59f2 ("iavf: switch to Page Pool"). With the legacy page splitting and recycling logic already removed, the driver is now in a clean state to adopt the modern memory model. The Page Pool integration simplifies buffer management by offloading DMA mapping and recycling to the core infrastructure. This eliminates the need for driver-specific handling of headroom, buffer sizing, and page order. The libeth helper is used for CPU-side processing, while DMA-for-device is handled by the Page Pool core. Additionally, this patch extends the conversion to cover XDP support. The driver now uses libeth_xdp helpers for Rx buffer processing, and optimizes XDP_TX by skipping per-frame DMA mapping. Instead, all buffers are mapped as bi-directional up front, leveraging Page Pool's lifecycle management. This significantly reduces overhead in virtualized environments. Performance observations: - In typical scenarios (netperf, XDP_PASS, XDP_DROP), performance remains on par with the previous implementation. - In XDP_TX mode: * With IOMMU enabled, performance improves dramatically - over 5x increase - due to reduced DMA mapping overhead and better memory reuse. * With IOMMU disabled, performance remains comparable to the previous implementation, with no significant changes observed. - In XDP_DROP mode: * For small MTUs, (where multiple buffers can be allocated on a single memory page), a performance drop of approximately 20% is observed. According to 'perf top' analysis, the bottleneck is caused by atomic reference counter increments in the Page Pool. * For normal MTUs, (where only one buffer can be allocated within a single memory page), performance remains comparable to baseline levels. This change is also a step toward a more modular and unified XDP implementation across Intel Ethernet drivers, aligning with ongoing efforts to consolidate and streamline feature support. Suggested-by: Maciej Fijalkowski Suggested-by: Alexander Lobakin Reviewed-by: Alexander Lobakin Reviewed-by: Jacob Keller Signed-off-by: Michal Kubiak Tested-by: Alexander Nowlin Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/Kconfig | 1 + drivers/net/ethernet/intel/ice/ice_base.c | 91 ++-- drivers/net/ethernet/intel/ice/ice_ethtool.c | 17 +- drivers/net/ethernet/intel/ice/ice_lib.c | 1 - drivers/net/ethernet/intel/ice/ice_main.c | 10 +- drivers/net/ethernet/intel/ice/ice_txrx.c | 444 ++++-------------- drivers/net/ethernet/intel/ice/ice_txrx.h | 37 +- drivers/net/ethernet/intel/ice/ice_txrx_lib.c | 69 ++- drivers/net/ethernet/intel/ice/ice_txrx_lib.h | 9 - drivers/net/ethernet/intel/ice/ice_xsk.c | 76 +-- drivers/net/ethernet/intel/ice/ice_xsk.h | 6 +- 11 files changed, 206 insertions(+), 555 deletions(-) diff --git a/drivers/net/ethernet/intel/Kconfig b/drivers/net/ethernet/intel/Kconfig index a563a94e2780..7c6ee1e604aa 100644 --- a/drivers/net/ethernet/intel/Kconfig +++ b/drivers/net/ethernet/intel/Kconfig @@ -296,6 +296,7 @@ config ICE depends on GNSS || GNSS = n select AUXILIARY_BUS select DIMLIB + select LIBETH_XDP select LIBIE select LIBIE_ADMINQ select LIBIE_FWLOG diff --git a/drivers/net/ethernet/intel/ice/ice_base.c b/drivers/net/ethernet/intel/ice/ice_base.c index fee58f879d9e..eabab50fab33 100644 --- a/drivers/net/ethernet/intel/ice/ice_base.c +++ b/drivers/net/ethernet/intel/ice/ice_base.c @@ -2,6 +2,7 @@ /* Copyright (c) 2019, Intel Corporation. */ #include +#include #include "ice_base.h" #include "ice_lib.h" #include "ice_dcb_lib.h" @@ -495,7 +496,7 @@ static int ice_setup_rx_ctx(struct ice_rx_ring *ring) /* Receive Packet Data Buffer Size. * The Packet Data Buffer Size is defined in 128 byte units. */ - rlan_ctx.dbuf = DIV_ROUND_UP(ICE_RXBUF_3072, + rlan_ctx.dbuf = DIV_ROUND_UP(ring->rx_buf_len, BIT_ULL(ICE_RLAN_CTX_DBUF_S)); /* use 32 byte descriptors */ @@ -537,7 +538,7 @@ static int ice_setup_rx_ctx(struct ice_rx_ring *ring) * than 5 x DBUF */ rlan_ctx.rxmax = min_t(u32, vsi->max_frame, - ICE_MAX_CHAINED_RX_BUFS * ICE_RXBUF_3072); + ICE_MAX_CHAINED_RX_BUFS * ring->rx_buf_len); /* Rx queue threshold in units of 64 */ rlan_ctx.lrxqthresh = 1; @@ -573,8 +574,6 @@ static int ice_setup_rx_ctx(struct ice_rx_ring *ring) if (vsi->type == ICE_VSI_VF) return 0; - ring->rx_offset = ICE_SKB_PAD; - /* init queue specific tail register */ ring->tail = hw->hw_addr + QRX_TAIL(pf_q); writel(0, ring->tail); @@ -582,38 +581,6 @@ static int ice_setup_rx_ctx(struct ice_rx_ring *ring) return 0; } -static void ice_xsk_pool_fill_cb(struct ice_rx_ring *ring) -{ - void *ctx_ptr = &ring->pkt_ctx; - struct xsk_cb_desc desc = {}; - - XSK_CHECK_PRIV_TYPE(struct ice_xdp_buff); - desc.src = &ctx_ptr; - desc.off = offsetof(struct ice_xdp_buff, pkt_ctx) - - sizeof(struct xdp_buff); - desc.bytes = sizeof(ctx_ptr); - xsk_pool_fill_cb(ring->xsk_pool, &desc); -} - -/** - * ice_get_frame_sz - calculate xdp_buff::frame_sz - * @rx_ring: the ring being configured - * - * Return frame size based on underlying PAGE_SIZE - */ -static unsigned int ice_get_frame_sz(struct ice_rx_ring *rx_ring) -{ - unsigned int frame_sz; - -#if (PAGE_SIZE >= 8192) - frame_sz = rx_ring->rx_buf_len; -#else - frame_sz = PAGE_SIZE; -#endif - - return frame_sz; -} - /** * ice_vsi_cfg_rxq - Configure an Rx queue * @ring: the ring being configured @@ -622,8 +589,14 @@ static unsigned int ice_get_frame_sz(struct ice_rx_ring *rx_ring) */ static int ice_vsi_cfg_rxq(struct ice_rx_ring *ring) { + struct libeth_fq fq = { + .count = ring->count, + .nid = NUMA_NO_NODE, + .xdp = ice_is_xdp_ena_vsi(ring->vsi), + .buf_len = LIBIE_MAX_RX_BUF_LEN, + }; struct device *dev = ice_pf_to_dev(ring->vsi->back); - u32 num_bufs = ICE_RX_DESC_UNUSED(ring); + u32 num_bufs = ICE_DESC_UNUSED(ring); u32 rx_buf_len; int err; @@ -632,12 +605,16 @@ static int ice_vsi_cfg_rxq(struct ice_rx_ring *ring) err = __xdp_rxq_info_reg(&ring->xdp_rxq, ring->netdev, ring->q_index, ring->q_vector->napi.napi_id, - ICE_RXBUF_3072); + ring->rx_buf_len); if (err) return err; } ice_rx_xsk_pool(ring); + err = ice_realloc_rx_xdp_bufs(ring, ring->xsk_pool); + if (err) + return err; + if (ring->xsk_pool) { xdp_rxq_info_unreg(&ring->xdp_rxq); @@ -655,36 +632,38 @@ static int ice_vsi_cfg_rxq(struct ice_rx_ring *ring) if (err) return err; xsk_pool_set_rxq_info(ring->xsk_pool, &ring->xdp_rxq); - ice_xsk_pool_fill_cb(ring); dev_info(dev, "Registered XDP mem model MEM_TYPE_XSK_BUFF_POOL on Rx ring %d\n", ring->q_index); } else { + err = libeth_rx_fq_create(&fq, &ring->q_vector->napi); + if (err) + return err; + + ring->pp = fq.pp; + ring->rx_fqes = fq.fqes; + ring->truesize = fq.truesize; + ring->rx_buf_len = fq.buf_len; + if (!xdp_rxq_info_is_reg(&ring->xdp_rxq)) { err = __xdp_rxq_info_reg(&ring->xdp_rxq, ring->netdev, ring->q_index, ring->q_vector->napi.napi_id, - ICE_RXBUF_3072); + ring->rx_buf_len); if (err) - return err; + goto err_destroy_fq; } - - err = xdp_rxq_info_reg_mem_model(&ring->xdp_rxq, - MEM_TYPE_PAGE_SHARED, - NULL); - if (err) - return err; + xdp_rxq_info_attach_page_pool(&ring->xdp_rxq, + ring->pp); } } - xdp_init_buff(&ring->xdp, ice_get_frame_sz(ring), &ring->xdp_rxq); ring->xdp.data = NULL; - ring->xdp_ext.pkt_ctx = &ring->pkt_ctx; err = ice_setup_rx_ctx(ring); if (err) { dev_err(dev, "ice_setup_rx_ctx failed for RxQ %d, err %d\n", ring->q_index, err); - return err; + goto err_destroy_fq; } if (ring->xsk_pool) { @@ -712,9 +691,19 @@ static int ice_vsi_cfg_rxq(struct ice_rx_ring *ring) if (ring->vsi->type == ICE_VSI_CTRL) ice_init_ctrl_rx_descs(ring, num_bufs); else - ice_alloc_rx_bufs(ring, num_bufs); + err = ice_alloc_rx_bufs(ring, num_bufs); + + if (err) + goto err_destroy_fq; return 0; + +err_destroy_fq: + libeth_rx_fq_destroy(&fq); + ring->rx_fqes = NULL; + ring->pp = NULL; + + return err; } int ice_vsi_cfg_single_rxq(struct ice_vsi *vsi, u16 q_idx) diff --git a/drivers/net/ethernet/intel/ice/ice_ethtool.c b/drivers/net/ethernet/intel/ice/ice_ethtool.c index 240e3f35c10a..36fdac4fddc3 100644 --- a/drivers/net/ethernet/intel/ice/ice_ethtool.c +++ b/drivers/net/ethernet/intel/ice/ice_ethtool.c @@ -10,6 +10,7 @@ #include "ice_lib.h" #include "ice_dcb_lib.h" #include +#include struct ice_stats { char stat_string[ETH_GSTRING_LEN]; @@ -1230,8 +1231,9 @@ static int ice_diag_send(struct ice_tx_ring *tx_ring, u8 *data, u16 size) */ static int ice_lbtest_receive_frames(struct ice_rx_ring *rx_ring) { - struct ice_rx_buf *rx_buf; + struct libeth_fqe *rx_buf; int valid_frames, i; + struct page *page; u8 *received_buf; valid_frames = 0; @@ -1246,8 +1248,10 @@ static int ice_lbtest_receive_frames(struct ice_rx_ring *rx_ring) cpu_to_le16(BIT(ICE_RX_FLEX_DESC_STATUS0_EOF_S))))) continue; - rx_buf = &rx_ring->rx_buf[i]; - received_buf = page_address(rx_buf->page) + rx_buf->page_offset; + rx_buf = &rx_ring->rx_fqes[i]; + page = __netmem_to_page(rx_buf->netmem); + received_buf = page_address(page) + rx_buf->offset + + page->pp->p.offset; if (ice_lbtest_check_frame(received_buf)) valid_frames++; @@ -3303,7 +3307,8 @@ ice_set_ringparam(struct net_device *netdev, struct ethtool_ringparam *ring, rx_rings[i].count = new_rx_cnt; rx_rings[i].cached_phctime = pf->ptp.cached_phc_time; rx_rings[i].desc = NULL; - rx_rings[i].rx_buf = NULL; + rx_rings[i].xdp_buf = NULL; + /* this is to allow wr32 to have something to write to * during early allocation of Rx buffers */ @@ -3312,10 +3317,6 @@ ice_set_ringparam(struct net_device *netdev, struct ethtool_ringparam *ring, err = ice_setup_rx_ring(&rx_rings[i]); if (err) goto rx_unwind; - - /* allocate Rx buffers */ - err = ice_alloc_rx_bufs(&rx_rings[i], - ICE_RX_DESC_UNUSED(&rx_rings[i])); rx_unwind: if (err) { while (i) { diff --git a/drivers/net/ethernet/intel/ice/ice_lib.c b/drivers/net/ethernet/intel/ice/ice_lib.c index 69cb0381c460..15621707fbf8 100644 --- a/drivers/net/ethernet/intel/ice/ice_lib.c +++ b/drivers/net/ethernet/intel/ice/ice_lib.c @@ -1427,7 +1427,6 @@ static int ice_vsi_alloc_rings(struct ice_vsi *vsi) ring->reg_idx = vsi->rxq_map[i]; ring->vsi = vsi; ring->netdev = vsi->netdev; - ring->dev = dev; ring->count = vsi->num_rx_desc; ring->cached_phctime = pf->ptp.cached_phc_time; diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c index 36e6a078c84c..2533876f1a2f 100644 --- a/drivers/net/ethernet/intel/ice/ice_main.c +++ b/drivers/net/ethernet/intel/ice/ice_main.c @@ -37,6 +37,8 @@ static const char ice_copyright[] = "Copyright (c) 2018, Intel Corporation."; #define ICE_DDP_PKG_FILE ICE_DDP_PKG_PATH "ice.pkg" MODULE_DESCRIPTION(DRV_SUMMARY); +MODULE_IMPORT_NS("LIBETH"); +MODULE_IMPORT_NS("LIBETH_XDP"); MODULE_IMPORT_NS("LIBIE"); MODULE_IMPORT_NS("LIBIE_ADMINQ"); MODULE_IMPORT_NS("LIBIE_FWLOG"); @@ -3015,19 +3017,11 @@ ice_xdp_setup_prog(struct ice_vsi *vsi, struct bpf_prog *prog, } } xdp_features_set_redirect_target(vsi->netdev, true); - /* reallocate Rx queues that are used for zero-copy */ - xdp_ring_err = ice_realloc_zc_buf(vsi, true); - if (xdp_ring_err) - NL_SET_ERR_MSG_MOD(extack, "Setting up XDP Rx resources failed"); } else if (ice_is_xdp_ena_vsi(vsi) && !prog) { xdp_features_clear_redirect_target(vsi->netdev); xdp_ring_err = ice_destroy_xdp_rings(vsi, ICE_XDP_CFG_FULL); if (xdp_ring_err) NL_SET_ERR_MSG_MOD(extack, "Freeing XDP Tx resources failed"); - /* reallocate Rx queues that were used for zero-copy */ - xdp_ring_err = ice_realloc_zc_buf(vsi, false); - if (xdp_ring_err) - NL_SET_ERR_MSG_MOD(extack, "Freeing XDP Rx resources failed"); } resume_if: diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.c b/drivers/net/ethernet/intel/ice/ice_txrx.c index 2b46e4c8be86..5a966138eacf 100644 --- a/drivers/net/ethernet/intel/ice/ice_txrx.c +++ b/drivers/net/ethernet/intel/ice/ice_txrx.c @@ -7,6 +7,8 @@ #include #include #include +#include +#include #include #include #include @@ -111,7 +113,7 @@ ice_prgm_fdir_fltr(struct ice_vsi *vsi, struct ice_fltr_desc *fdir_desc, static void ice_unmap_and_free_tx_buf(struct ice_tx_ring *ring, struct ice_tx_buf *tx_buf) { - if (dma_unmap_len(tx_buf, len)) + if (tx_buf->type != ICE_TX_BUF_XDP_TX && dma_unmap_len(tx_buf, len)) dma_unmap_page(ring->dev, dma_unmap_addr(tx_buf, dma), dma_unmap_len(tx_buf, len), @@ -125,7 +127,7 @@ ice_unmap_and_free_tx_buf(struct ice_tx_ring *ring, struct ice_tx_buf *tx_buf) dev_kfree_skb_any(tx_buf->skb); break; case ICE_TX_BUF_XDP_TX: - page_frag_free(tx_buf->raw_buf); + libeth_xdp_return_va(tx_buf->raw_buf, false); break; case ICE_TX_BUF_XDP_XMIT: xdp_return_frame(tx_buf->xdpf); @@ -512,54 +514,44 @@ int ice_setup_tx_ring(struct ice_tx_ring *tx_ring) */ void ice_clean_rx_ring(struct ice_rx_ring *rx_ring) { - struct xdp_buff *xdp = &rx_ring->xdp; - struct device *dev = rx_ring->dev; + struct libeth_fq fq = { + .fqes = rx_ring->rx_fqes, + .pp = rx_ring->pp, + }; u32 size; - u16 i; - - /* ring already cleared, nothing to do */ - if (!rx_ring->rx_buf) - return; if (rx_ring->xsk_pool) { ice_xsk_clean_rx_ring(rx_ring); goto rx_skip_free; } - if (xdp->data) { - xdp_return_buff(xdp); - xdp->data = NULL; - } + /* ring already cleared, nothing to do */ + if (!rx_ring->rx_fqes) + return; + + libeth_xdp_return_stash(&rx_ring->xdp); /* Free all the Rx ring sk_buffs */ - for (i = 0; i < rx_ring->count; i++) { - struct ice_rx_buf *rx_buf = &rx_ring->rx_buf[i]; + for (u32 i = rx_ring->next_to_clean; i != rx_ring->next_to_use; ) { + const struct libeth_fqe *rx_fqes = &rx_ring->rx_fqes[i]; - if (!rx_buf->page) - continue; + libeth_rx_recycle_slow(rx_fqes->netmem); - /* Invalidate cache lines that may have been written to by - * device so that we avoid corrupting memory. - */ - dma_sync_single_range_for_cpu(dev, rx_buf->dma, - rx_buf->page_offset, - ICE_RXBUF_3072, - DMA_FROM_DEVICE); - - /* free resources associated with mapping */ - dma_unmap_page_attrs(dev, rx_buf->dma, PAGE_SIZE, - DMA_FROM_DEVICE, ICE_RX_DMA_ATTR); - __free_page(rx_buf->page); - - rx_buf->page = NULL; - rx_buf->page_offset = 0; + if (unlikely(++i == rx_ring->count)) + i = 0; } + if (rx_ring->vsi->type == ICE_VSI_PF && + xdp_rxq_info_is_reg(&rx_ring->xdp_rxq)) { + xdp_rxq_info_detach_mem_model(&rx_ring->xdp_rxq); + xdp_rxq_info_unreg(&rx_ring->xdp_rxq); + } + + libeth_rx_fq_destroy(&fq); + rx_ring->rx_fqes = NULL; + rx_ring->pp = NULL; + rx_skip_free: - if (rx_ring->xsk_pool) - memset(rx_ring->xdp_buf, 0, array_size(rx_ring->count, sizeof(*rx_ring->xdp_buf))); - else - memset(rx_ring->rx_buf, 0, array_size(rx_ring->count, sizeof(*rx_ring->rx_buf))); /* Zero out the descriptor ring */ size = ALIGN(rx_ring->count * sizeof(union ice_32byte_rx_desc), @@ -568,7 +560,6 @@ void ice_clean_rx_ring(struct ice_rx_ring *rx_ring) rx_ring->next_to_alloc = 0; rx_ring->next_to_clean = 0; - rx_ring->first_desc = 0; rx_ring->next_to_use = 0; } @@ -580,26 +571,20 @@ void ice_clean_rx_ring(struct ice_rx_ring *rx_ring) */ void ice_free_rx_ring(struct ice_rx_ring *rx_ring) { + struct device *dev = ice_pf_to_dev(rx_ring->vsi->back); u32 size; ice_clean_rx_ring(rx_ring); - if (rx_ring->vsi->type == ICE_VSI_PF) - if (xdp_rxq_info_is_reg(&rx_ring->xdp_rxq)) - xdp_rxq_info_unreg(&rx_ring->xdp_rxq); WRITE_ONCE(rx_ring->xdp_prog, NULL); if (rx_ring->xsk_pool) { kfree(rx_ring->xdp_buf); rx_ring->xdp_buf = NULL; - } else { - kfree(rx_ring->rx_buf); - rx_ring->rx_buf = NULL; } if (rx_ring->desc) { size = ALIGN(rx_ring->count * sizeof(union ice_32byte_rx_desc), PAGE_SIZE); - dmam_free_coherent(rx_ring->dev, size, - rx_ring->desc, rx_ring->dma); + dmam_free_coherent(dev, size, rx_ring->desc, rx_ring->dma); rx_ring->desc = NULL; } } @@ -612,19 +597,9 @@ void ice_free_rx_ring(struct ice_rx_ring *rx_ring) */ int ice_setup_rx_ring(struct ice_rx_ring *rx_ring) { - struct device *dev = rx_ring->dev; + struct device *dev = ice_pf_to_dev(rx_ring->vsi->back); u32 size; - if (!dev) - return -ENOMEM; - - /* warn if we are about to overwrite the pointer */ - WARN_ON(rx_ring->rx_buf); - rx_ring->rx_buf = - kcalloc(rx_ring->count, sizeof(*rx_ring->rx_buf), GFP_KERNEL); - if (!rx_ring->rx_buf) - return -ENOMEM; - /* round up to nearest page */ size = ALIGN(rx_ring->count * sizeof(union ice_32byte_rx_desc), PAGE_SIZE); @@ -633,22 +608,16 @@ int ice_setup_rx_ring(struct ice_rx_ring *rx_ring) if (!rx_ring->desc) { dev_err(dev, "Unable to allocate memory for the Rx descriptor ring, size=%d\n", size); - goto err; + return -ENOMEM; } rx_ring->next_to_use = 0; rx_ring->next_to_clean = 0; - rx_ring->first_desc = 0; if (ice_is_xdp_ena_vsi(rx_ring->vsi)) WRITE_ONCE(rx_ring->xdp_prog, rx_ring->vsi->xdp_prog); return 0; - -err: - kfree(rx_ring->rx_buf); - rx_ring->rx_buf = NULL; - return -ENOMEM; } /** @@ -662,7 +631,7 @@ int ice_setup_rx_ring(struct ice_rx_ring *rx_ring) * Returns any of ICE_XDP_{PASS, CONSUMED, TX, REDIR} */ static u32 -ice_run_xdp(struct ice_rx_ring *rx_ring, struct xdp_buff *xdp, +ice_run_xdp(struct ice_rx_ring *rx_ring, struct libeth_xdp_buff *xdp, struct bpf_prog *xdp_prog, struct ice_tx_ring *xdp_ring, union ice_32b_rx_flex_desc *eop_desc) { @@ -672,23 +641,23 @@ ice_run_xdp(struct ice_rx_ring *rx_ring, struct xdp_buff *xdp, if (!xdp_prog) goto exit; - ice_xdp_meta_set_desc(xdp, eop_desc); + xdp->desc = eop_desc; - act = bpf_prog_run_xdp(xdp_prog, xdp); + act = bpf_prog_run_xdp(xdp_prog, &xdp->base); switch (act) { case XDP_PASS: break; case XDP_TX: if (static_branch_unlikely(&ice_xdp_locking_key)) spin_lock(&xdp_ring->tx_lock); - ret = __ice_xmit_xdp_ring(xdp, xdp_ring, false); + ret = __ice_xmit_xdp_ring(&xdp->base, xdp_ring, false); if (static_branch_unlikely(&ice_xdp_locking_key)) spin_unlock(&xdp_ring->tx_lock); if (ret == ICE_XDP_CONSUMED) goto out_failure; break; case XDP_REDIRECT: - if (xdp_do_redirect(rx_ring->netdev, xdp, xdp_prog)) + if (xdp_do_redirect(rx_ring->netdev, &xdp->base, xdp_prog)) goto out_failure; ret = ICE_XDP_REDIR; break; @@ -700,8 +669,10 @@ ice_run_xdp(struct ice_rx_ring *rx_ring, struct xdp_buff *xdp, trace_xdp_exception(rx_ring->netdev, xdp_prog, act); fallthrough; case XDP_DROP: + libeth_xdp_return_buff(xdp); ret = ICE_XDP_CONSUMED; } + exit: return ret; } @@ -789,48 +760,6 @@ ice_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frames, return nxmit; } -/** - * ice_alloc_mapped_page - recycle or make a new page - * @rx_ring: ring to use - * @bi: rx_buf struct to modify - * - * Returns true if the page was successfully allocated or - * reused. - */ -static bool -ice_alloc_mapped_page(struct ice_rx_ring *rx_ring, struct ice_rx_buf *bi) -{ - struct page *page = bi->page; - dma_addr_t dma; - - /* alloc new page for storage */ - page = dev_alloc_pages(ice_rx_pg_order(rx_ring)); - if (unlikely(!page)) { - rx_ring->ring_stats->rx_stats.alloc_page_failed++; - return false; - } - - /* map page for use */ - dma = dma_map_page_attrs(rx_ring->dev, page, 0, PAGE_SIZE, - DMA_FROM_DEVICE, ICE_RX_DMA_ATTR); - - /* if mapping failed free memory back to system since - * there isn't much point in holding memory we can't use - */ - if (dma_mapping_error(rx_ring->dev, dma)) { - __free_pages(page, ice_rx_pg_order(rx_ring)); - rx_ring->ring_stats->rx_stats.alloc_page_failed++; - return false; - } - - bi->dma = dma; - bi->page = page; - bi->page_offset = rx_ring->rx_offset; - page_ref_add(page, USHRT_MAX - 1); - - return true; -} - /** * ice_init_ctrl_rx_descs - Initialize Rx descriptors for control vsi. * @rx_ring: ring to init descriptors on @@ -877,9 +806,14 @@ void ice_init_ctrl_rx_descs(struct ice_rx_ring *rx_ring, u32 count) */ bool ice_alloc_rx_bufs(struct ice_rx_ring *rx_ring, unsigned int cleaned_count) { + const struct libeth_fq_fp fq = { + .pp = rx_ring->pp, + .fqes = rx_ring->rx_fqes, + .truesize = rx_ring->truesize, + .count = rx_ring->count, + }; union ice_32b_rx_flex_desc *rx_desc; u16 ntu = rx_ring->next_to_use; - struct ice_rx_buf *bi; /* do nothing if no valid netdev defined */ if (!rx_ring->netdev || !cleaned_count) @@ -887,30 +821,25 @@ bool ice_alloc_rx_bufs(struct ice_rx_ring *rx_ring, unsigned int cleaned_count) /* get the Rx descriptor and buffer based on next_to_use */ rx_desc = ICE_RX_DESC(rx_ring, ntu); - bi = &rx_ring->rx_buf[ntu]; do { - /* if we fail here, we have work remaining */ - if (!ice_alloc_mapped_page(rx_ring, bi)) - break; + dma_addr_t addr; - /* sync the buffer for use by the device */ - dma_sync_single_range_for_device(rx_ring->dev, bi->dma, - bi->page_offset, - ICE_RXBUF_3072, - DMA_FROM_DEVICE); + addr = libeth_rx_alloc(&fq, ntu); + if (addr == DMA_MAPPING_ERROR) { + rx_ring->ring_stats->rx_stats.alloc_page_failed++; + break; + } /* Refresh the desc even if buffer_addrs didn't change * because each write-back erases this info. */ - rx_desc->read.pkt_addr = cpu_to_le64(bi->dma + bi->page_offset); + rx_desc->read.pkt_addr = cpu_to_le64(addr); rx_desc++; - bi++; ntu++; if (unlikely(ntu == rx_ring->count)) { rx_desc = ICE_RX_DESC(rx_ring, 0); - bi = rx_ring->rx_buf; ntu = 0; } @@ -926,205 +855,6 @@ bool ice_alloc_rx_bufs(struct ice_rx_ring *rx_ring, unsigned int cleaned_count) return !!cleaned_count; } -/** - * ice_add_xdp_frag - Add contents of Rx buffer to xdp buf as a frag - * @rx_ring: Rx descriptor ring to transact packets on - * @xdp: xdp buff to place the data into - * @rx_buf: buffer containing page to add - * @size: packet length from rx_desc - * - * This function will add the data contained in rx_buf->page to the xdp buf. - * It will just attach the page as a frag. - */ -static int -ice_add_xdp_frag(struct ice_rx_ring *rx_ring, struct xdp_buff *xdp, - struct ice_rx_buf *rx_buf, const unsigned int size) -{ - struct skb_shared_info *sinfo = xdp_get_shared_info_from_buff(xdp); - - if (!size) - return 0; - - if (!xdp_buff_has_frags(xdp)) { - sinfo->nr_frags = 0; - sinfo->xdp_frags_size = 0; - xdp_buff_set_frags_flag(xdp); - } - - if (unlikely(sinfo->nr_frags == MAX_SKB_FRAGS)) - return -ENOMEM; - - __skb_fill_page_desc_noacc(sinfo, sinfo->nr_frags++, rx_buf->page, - rx_buf->page_offset, size); - sinfo->xdp_frags_size += size; - - if (page_is_pfmemalloc(rx_buf->page)) - xdp_buff_set_frag_pfmemalloc(xdp); - - return 0; -} - -/** - * ice_get_rx_buf - Fetch Rx buffer and synchronize data for use - * @rx_ring: Rx descriptor ring to transact packets on - * @size: size of buffer to add to skb - * @ntc: index of next to clean element - * - * This function will pull an Rx buffer from the ring and synchronize it - * for use by the CPU. - */ -static struct ice_rx_buf * -ice_get_rx_buf(struct ice_rx_ring *rx_ring, const unsigned int size, - const unsigned int ntc) -{ - struct ice_rx_buf *rx_buf; - - rx_buf = &rx_ring->rx_buf[ntc]; - prefetchw(rx_buf->page); - - if (!size) - return rx_buf; - /* we are reusing so sync this buffer for CPU use */ - dma_sync_single_range_for_cpu(rx_ring->dev, rx_buf->dma, - rx_buf->page_offset, size, - DMA_FROM_DEVICE); - - return rx_buf; -} - -/** - * ice_get_pgcnts - grab page_count() for gathered fragments - * @rx_ring: Rx descriptor ring to store the page counts on - * @ntc: the next to clean element (not included in this frame!) - * - * This function is intended to be called right before running XDP - * program so that the page recycling mechanism will be able to take - * a correct decision regarding underlying pages; this is done in such - * way as XDP program can change the refcount of page - */ -static void ice_get_pgcnts(struct ice_rx_ring *rx_ring, unsigned int ntc) -{ - u32 idx = rx_ring->first_desc; - struct ice_rx_buf *rx_buf; - u32 cnt = rx_ring->count; - - while (idx != ntc) { - rx_buf = &rx_ring->rx_buf[idx]; - rx_buf->pgcnt = page_count(rx_buf->page); - - if (++idx == cnt) - idx = 0; - } -} - -/** - * ice_build_skb - Build skb around an existing buffer - * @rx_ring: Rx descriptor ring to transact packets on - * @xdp: xdp_buff pointing to the data - * - * This function builds an skb around an existing XDP buffer, taking care - * to set up the skb correctly and avoid any memcpy overhead. Driver has - * already combined frags (if any) to skb_shared_info. - */ -static struct sk_buff * -ice_build_skb(struct ice_rx_ring *rx_ring, struct xdp_buff *xdp) -{ - u8 metasize = xdp->data - xdp->data_meta; - struct skb_shared_info *sinfo = NULL; - unsigned int nr_frags; - struct sk_buff *skb; - - if (unlikely(xdp_buff_has_frags(xdp))) { - sinfo = xdp_get_shared_info_from_buff(xdp); - nr_frags = sinfo->nr_frags; - } - - /* Prefetch first cache line of first page. If xdp->data_meta - * is unused, this points exactly as xdp->data, otherwise we - * likely have a consumer accessing first few bytes of meta - * data, and then actual data. - */ - net_prefetch(xdp->data_meta); - /* build an skb around the page buffer */ - skb = napi_build_skb(xdp->data_hard_start, xdp->frame_sz); - if (unlikely(!skb)) - return NULL; - - /* must to record Rx queue, otherwise OS features such as - * symmetric queue won't work - */ - skb_record_rx_queue(skb, rx_ring->q_index); - - /* update pointers within the skb to store the data */ - skb_reserve(skb, xdp->data - xdp->data_hard_start); - __skb_put(skb, xdp->data_end - xdp->data); - if (metasize) - skb_metadata_set(skb, metasize); - - if (unlikely(xdp_buff_has_frags(xdp))) - xdp_update_skb_frags_info(skb, nr_frags, sinfo->xdp_frags_size, - nr_frags * xdp->frame_sz, - xdp_buff_get_skb_flags(xdp)); - - return skb; -} - -/** - * ice_put_rx_buf - Clean up used buffer and either recycle or free - * @rx_ring: Rx descriptor ring to transact packets on - * @rx_buf: Rx buffer to pull data from - * - * This function will clean up the contents of the rx_buf. It will either - * recycle the buffer or unmap it and free the associated resources. - */ -static void -ice_put_rx_buf(struct ice_rx_ring *rx_ring, struct ice_rx_buf *rx_buf) -{ - if (!rx_buf) - return; - - /* we are not reusing the buffer so unmap it */ - dma_unmap_page_attrs(rx_ring->dev, rx_buf->dma, - PAGE_SIZE, DMA_FROM_DEVICE, - ICE_RX_DMA_ATTR); - - /* clear contents of buffer_info */ - rx_buf->page = NULL; -} - -/** - * ice_put_rx_mbuf - ice_put_rx_buf() caller, for all buffers in frame - * @rx_ring: Rx ring with all the auxiliary data - * @xdp: XDP buffer carrying linear + frags part - * @ntc: the next to clean element (not included in this frame!) - * @verdict: return code from XDP program execution - * - * Called after XDP program is completed, or on error with verdict set to - * ICE_XDP_CONSUMED. - * - * Walk through buffers from first_desc to the end of the frame, releasing - * buffers and satisfying internal page recycle mechanism. The action depends - * on verdict from XDP program. - */ -static void ice_put_rx_mbuf(struct ice_rx_ring *rx_ring, struct xdp_buff *xdp, - u32 ntc, u32 verdict) -{ - u32 idx = rx_ring->first_desc; - u32 cnt = rx_ring->count; - struct ice_rx_buf *buf; - - while (idx != ntc) { - buf = &rx_ring->rx_buf[idx]; - if (++idx == cnt) - idx = 0; - - ice_put_rx_buf(rx_ring, buf); - } - - xdp->data = NULL; - rx_ring->first_desc = ntc; -} - /** * ice_clean_ctrl_rx_irq - Clean descriptors from flow director Rx ring * @rx_ring: Rx descriptor ring for ctrl_vsi to transact packets on @@ -1159,9 +889,8 @@ void ice_clean_ctrl_rx_irq(struct ice_rx_ring *rx_ring) total_rx_pkts++; } - rx_ring->first_desc = ntc; rx_ring->next_to_clean = ntc; - ice_init_ctrl_rx_descs(rx_ring, ICE_RX_DESC_UNUSED(rx_ring)); + ice_init_ctrl_rx_descs(rx_ring, ICE_DESC_UNUSED(rx_ring)); } /** @@ -1179,16 +908,17 @@ void ice_clean_ctrl_rx_irq(struct ice_rx_ring *rx_ring) static int ice_clean_rx_irq(struct ice_rx_ring *rx_ring, int budget) { unsigned int total_rx_bytes = 0, total_rx_pkts = 0; - unsigned int offset = rx_ring->rx_offset; - struct xdp_buff *xdp = &rx_ring->xdp; struct ice_tx_ring *xdp_ring = NULL; struct bpf_prog *xdp_prog = NULL; u32 ntc = rx_ring->next_to_clean; + LIBETH_XDP_ONSTACK_BUFF(xdp); u32 cached_ntu, xdp_verdict; u32 cnt = rx_ring->count; u32 xdp_xmit = 0; bool failure; + libeth_xdp_init_buff(xdp, &rx_ring->xdp, &rx_ring->xdp_rxq); + xdp_prog = READ_ONCE(rx_ring->xdp_prog); if (xdp_prog) { xdp_ring = rx_ring->xdp_ring; @@ -1198,7 +928,7 @@ static int ice_clean_rx_irq(struct ice_rx_ring *rx_ring, int budget) /* start the loop to process Rx packets bounded by 'budget' */ while (likely(total_rx_pkts < (unsigned int)budget)) { union ice_32b_rx_flex_desc *rx_desc; - struct ice_rx_buf *rx_buf; + struct libeth_fqe *rx_buf; struct sk_buff *skb; unsigned int size; u16 stat_err_bits; @@ -1227,65 +957,47 @@ static int ice_clean_rx_irq(struct ice_rx_ring *rx_ring, int budget) size = le16_to_cpu(rx_desc->wb.pkt_len) & ICE_RX_FLX_DESC_PKT_LEN_M; - /* retrieve a buffer from the ring */ - rx_buf = ice_get_rx_buf(rx_ring, size, ntc); + stat_err_bits = BIT(ICE_RX_FLEX_DESC_STATUS0_RXE_S); + if (unlikely(ice_test_staterr(rx_desc->wb.status_error0, + stat_err_bits))) + size = 0; + + /* retrieve a buffer from the ring */ + rx_buf = &rx_ring->rx_fqes[ntc]; + libeth_xdp_process_buff(xdp, rx_buf, size); - /* Increment ntc before calls to ice_put_rx_mbuf() */ if (++ntc == cnt) ntc = 0; - if (!xdp->data) { - void *hard_start; - - hard_start = page_address(rx_buf->page) + rx_buf->page_offset - - offset; - xdp_prepare_buff(xdp, hard_start, offset, size, !!offset); - xdp_buff_clear_frags_flag(xdp); - } else if (ice_add_xdp_frag(rx_ring, xdp, rx_buf, size)) { - ice_put_rx_mbuf(rx_ring, xdp, ntc, ICE_XDP_CONSUMED); - break; - } - /* skip if it is NOP desc */ - if (ice_is_non_eop(rx_ring, rx_desc)) + if (ice_is_non_eop(rx_ring, rx_desc) || unlikely(!xdp->data)) continue; - ice_get_pgcnts(rx_ring, ntc); xdp_verdict = ice_run_xdp(rx_ring, xdp, xdp_prog, xdp_ring, rx_desc); if (xdp_verdict == ICE_XDP_PASS) goto construct_skb; - total_rx_bytes += xdp_get_buff_len(xdp); + + if (xdp_verdict & (ICE_XDP_TX | ICE_XDP_REDIR)) + xdp_xmit |= xdp_verdict; + total_rx_bytes += xdp_get_buff_len(&xdp->base); total_rx_pkts++; - ice_put_rx_mbuf(rx_ring, xdp, ntc, xdp_verdict); - xdp_xmit |= xdp_verdict & (ICE_XDP_TX | ICE_XDP_REDIR); - + xdp->data = NULL; continue; + construct_skb: - skb = ice_build_skb(rx_ring, xdp); + skb = xdp_build_skb_from_buff(&xdp->base); + xdp->data = NULL; + /* exit if we failed to retrieve a buffer */ if (!skb) { + libeth_xdp_return_buff_slow(xdp); rx_ring->ring_stats->rx_stats.alloc_buf_failed++; - xdp_verdict = ICE_XDP_CONSUMED; - } - ice_put_rx_mbuf(rx_ring, xdp, ntc, xdp_verdict); - - if (!skb) - break; - - stat_err_bits = BIT(ICE_RX_FLEX_DESC_STATUS0_RXE_S); - if (unlikely(ice_test_staterr(rx_desc->wb.status_error0, - stat_err_bits))) { - dev_kfree_skb_any(skb); continue; } vlan_tci = ice_get_vlan_tci(rx_desc); - /* pad the skb if needed, to make a valid ethernet frame */ - if (eth_skb_pad(skb)) - continue; - /* probably a little skewed due to removing CRC */ total_rx_bytes += skb->len; @@ -1302,11 +1014,13 @@ static int ice_clean_rx_irq(struct ice_rx_ring *rx_ring, int budget) rx_ring->next_to_clean = ntc; /* return up to cleaned_count buffers to hardware */ - failure = ice_alloc_rx_bufs(rx_ring, ICE_RX_DESC_UNUSED(rx_ring)); + failure = ice_alloc_rx_bufs(rx_ring, ICE_DESC_UNUSED(rx_ring)); if (xdp_xmit) ice_finalize_xdp_rx(xdp_ring, xdp_xmit, cached_ntu); + libeth_xdp_save_buff(&rx_ring->xdp, xdp); + if (rx_ring->ring_stats) ice_update_rx_ring_stats(rx_ring, total_rx_pkts, total_rx_bytes); diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.h b/drivers/net/ethernet/intel/ice/ice_txrx.h index 3c7830f787de..e97a38ef3fe7 100644 --- a/drivers/net/ethernet/intel/ice/ice_txrx.h +++ b/drivers/net/ethernet/intel/ice/ice_txrx.h @@ -4,6 +4,8 @@ #ifndef _ICE_TXRX_H_ #define _ICE_TXRX_H_ +#include + #include "ice_type.h" #define ICE_DFLT_IRQ_WORK 256 @@ -27,8 +29,6 @@ #define ICE_MAX_TXQ_PER_TXQG 128 -#define ICE_SKB_PAD (NET_SKB_PAD + NET_IP_ALIGN) - /* We are assuming that the cache line is always 64 Bytes here for ice. * In order to make sure that is a correct assumption there is a check in probe * to print a warning if the read from GLPCI_CNF2 tells us that the cache line @@ -48,10 +48,6 @@ (u16)((((R)->next_to_clean > (R)->next_to_use) ? 0 : (R)->count) + \ (R)->next_to_clean - (R)->next_to_use - 1) -#define ICE_RX_DESC_UNUSED(R) \ - ((((R)->first_desc > (R)->next_to_use) ? 0 : (R)->count) + \ - (R)->first_desc - (R)->next_to_use - 1) - #define ICE_RING_QUARTER(R) ((R)->count >> 2) #define ICE_TX_FLAGS_TSO BIT(0) @@ -133,13 +129,6 @@ struct ice_tx_offload_params { u8 header_len; }; -struct ice_rx_buf { - dma_addr_t dma; - struct page *page; - unsigned int page_offset; - unsigned int pgcnt; -}; - struct ice_q_stats { u64 pkts; u64 bytes; @@ -197,15 +186,6 @@ struct ice_pkt_ctx { __be16 vlan_proto; }; -struct ice_xdp_buff { - struct xdp_buff xdp_buff; - const union ice_32b_rx_flex_desc *eop_desc; - const struct ice_pkt_ctx *pkt_ctx; -}; - -/* Required for compatibility with xdp_buffs from xsk_pool */ -static_assert(offsetof(struct ice_xdp_buff, xdp_buff) == 0); - /* indices into GLINT_ITR registers */ #define ICE_RX_ITR ICE_IDX_ITR0 #define ICE_TX_ITR ICE_IDX_ITR1 @@ -258,7 +238,7 @@ struct ice_tstamp_ring { struct ice_rx_ring { /* CL1 - 1st cacheline starts here */ void *desc; /* Descriptor ring memory */ - struct device *dev; /* Used for DMA mapping */ + struct page_pool *pp; struct net_device *netdev; /* netdev ring maps to */ struct ice_vsi *vsi; /* Backreference to associated VSI */ struct ice_q_vector *q_vector; /* Backreference to associated vector */ @@ -270,14 +250,16 @@ struct ice_rx_ring { u16 next_to_alloc; union { - struct ice_rx_buf *rx_buf; + struct libeth_fqe *rx_fqes; struct xdp_buff **xdp_buf; }; + /* CL2 - 2nd cacheline starts here */ union { - struct ice_xdp_buff xdp_ext; - struct xdp_buff xdp; + struct libeth_xdp_buff_stash xdp; + struct libeth_xdp_buff *xsk; }; + /* CL3 - 3rd cacheline starts here */ union { struct ice_pkt_ctx pkt_ctx; @@ -287,12 +269,11 @@ struct ice_rx_ring { }; }; struct bpf_prog *xdp_prog; - u16 rx_offset; /* used in interrupt processing */ u16 next_to_use; u16 next_to_clean; - u16 first_desc; + u32 truesize; /* stats structs */ struct ice_ring_stats *ring_stats; diff --git a/drivers/net/ethernet/intel/ice/ice_txrx_lib.c b/drivers/net/ethernet/intel/ice/ice_txrx_lib.c index 45cfaabc41cb..956da38d63b0 100644 --- a/drivers/net/ethernet/intel/ice/ice_txrx_lib.c +++ b/drivers/net/ethernet/intel/ice/ice_txrx_lib.c @@ -3,6 +3,7 @@ #include #include +#include #include "ice_txrx_lib.h" #include "ice_eswitch.h" @@ -230,9 +231,12 @@ ice_process_skb_fields(struct ice_rx_ring *rx_ring, if (ice_is_port_repr_netdev(netdev)) ice_repr_inc_rx_stats(netdev, skb->len); + + /* __skb_push() is needed because xdp_build_skb_from_buff() + * calls eth_type_trans() + */ + __skb_push(skb, ETH_HLEN); skb->protocol = eth_type_trans(skb, netdev); - } else { - skb->protocol = eth_type_trans(skb, rx_ring->netdev); } ice_rx_csum(rx_ring, skb, rx_desc, ptype); @@ -270,19 +274,18 @@ static void ice_clean_xdp_tx_buf(struct device *dev, struct ice_tx_buf *tx_buf, struct xdp_frame_bulk *bq) { - dma_unmap_single(dev, dma_unmap_addr(tx_buf, dma), - dma_unmap_len(tx_buf, len), DMA_TO_DEVICE); - dma_unmap_len_set(tx_buf, len, 0); - switch (tx_buf->type) { case ICE_TX_BUF_XDP_TX: - page_frag_free(tx_buf->raw_buf); + libeth_xdp_return_va(tx_buf->raw_buf, true); break; case ICE_TX_BUF_XDP_XMIT: + dma_unmap_single(dev, dma_unmap_addr(tx_buf, dma), + dma_unmap_len(tx_buf, len), DMA_TO_DEVICE); xdp_return_frame_bulk(tx_buf->xdpf, bq); break; } + dma_unmap_len_set(tx_buf, len, 0); tx_buf->type = ICE_TX_BUF_EMPTY; } @@ -377,9 +380,11 @@ int __ice_xmit_xdp_ring(struct xdp_buff *xdp, struct ice_tx_ring *xdp_ring, struct ice_tx_buf *tx_buf; u32 cnt = xdp_ring->count; void *data = xdp->data; + struct page *page; u32 nr_frags = 0; u32 free_space; u32 frag = 0; + u32 offset; free_space = ICE_DESC_UNUSED(xdp_ring); if (free_space < ICE_RING_QUARTER(xdp_ring)) @@ -399,24 +404,28 @@ int __ice_xmit_xdp_ring(struct xdp_buff *xdp, struct ice_tx_ring *xdp_ring, tx_head = &xdp_ring->tx_buf[ntu]; tx_buf = tx_head; + page = virt_to_page(data); + offset = offset_in_page(xdp->data); + for (;;) { dma_addr_t dma; - dma = dma_map_single(dev, data, size, DMA_TO_DEVICE); - if (dma_mapping_error(dev, dma)) - goto dma_unmap; + if (frame) { + dma = dma_map_single(dev, data, size, DMA_TO_DEVICE); + if (dma_mapping_error(dev, dma)) + goto dma_unmap; + tx_buf->type = ICE_TX_BUF_FRAG; + } else { + dma = page_pool_get_dma_addr(page) + offset; + dma_sync_single_for_device(dev, dma, size, DMA_BIDIRECTIONAL); + tx_buf->type = ICE_TX_BUF_XDP_TX; + tx_buf->raw_buf = data; + } /* record length, and DMA address */ dma_unmap_len_set(tx_buf, len, size); dma_unmap_addr_set(tx_buf, dma, dma); - if (frame) { - tx_buf->type = ICE_TX_BUF_FRAG; - } else { - tx_buf->type = ICE_TX_BUF_XDP_TX; - tx_buf->raw_buf = data; - } - tx_desc->buf_addr = cpu_to_le64(dma); tx_desc->cmd_type_offset_bsz = ice_build_ctob(0, 0, size, 0); @@ -430,6 +439,8 @@ int __ice_xmit_xdp_ring(struct xdp_buff *xdp, struct ice_tx_ring *xdp_ring, tx_desc = ICE_TX_DESC(xdp_ring, ntu); tx_buf = &xdp_ring->tx_buf[ntu]; + page = skb_frag_page(&sinfo->frags[frag]); + offset = skb_frag_off(&sinfo->frags[frag]); data = skb_frag_address(&sinfo->frags[frag]); size = skb_frag_size(&sinfo->frags[frag]); frag++; @@ -514,10 +525,13 @@ void ice_finalize_xdp_rx(struct ice_tx_ring *xdp_ring, unsigned int xdp_res, */ static int ice_xdp_rx_hw_ts(const struct xdp_md *ctx, u64 *ts_ns) { - const struct ice_xdp_buff *xdp_ext = (void *)ctx; + const struct libeth_xdp_buff *xdp_ext = (void *)ctx; + struct ice_rx_ring *rx_ring; - *ts_ns = ice_ptp_get_rx_hwts(xdp_ext->eop_desc, - xdp_ext->pkt_ctx); + rx_ring = libeth_xdp_buff_to_rq(xdp_ext, typeof(*rx_ring), xdp_rxq); + + *ts_ns = ice_ptp_get_rx_hwts(xdp_ext->desc, + &rx_ring->pkt_ctx); if (!*ts_ns) return -ENODATA; @@ -545,10 +559,10 @@ ice_xdp_rx_hash_type(const union ice_32b_rx_flex_desc *eop_desc) static int ice_xdp_rx_hash(const struct xdp_md *ctx, u32 *hash, enum xdp_rss_hash_type *rss_type) { - const struct ice_xdp_buff *xdp_ext = (void *)ctx; + const struct libeth_xdp_buff *xdp_ext = (void *)ctx; - *hash = ice_get_rx_hash(xdp_ext->eop_desc); - *rss_type = ice_xdp_rx_hash_type(xdp_ext->eop_desc); + *hash = ice_get_rx_hash(xdp_ext->desc); + *rss_type = ice_xdp_rx_hash_type(xdp_ext->desc); if (!likely(*hash)) return -ENODATA; @@ -567,13 +581,16 @@ static int ice_xdp_rx_hash(const struct xdp_md *ctx, u32 *hash, static int ice_xdp_rx_vlan_tag(const struct xdp_md *ctx, __be16 *vlan_proto, u16 *vlan_tci) { - const struct ice_xdp_buff *xdp_ext = (void *)ctx; + const struct libeth_xdp_buff *xdp_ext = (void *)ctx; + struct ice_rx_ring *rx_ring; - *vlan_proto = xdp_ext->pkt_ctx->vlan_proto; + rx_ring = libeth_xdp_buff_to_rq(xdp_ext, typeof(*rx_ring), xdp_rxq); + + *vlan_proto = rx_ring->pkt_ctx.vlan_proto; if (!*vlan_proto) return -ENODATA; - *vlan_tci = ice_get_vlan_tci(xdp_ext->eop_desc); + *vlan_tci = ice_get_vlan_tci(xdp_ext->desc); if (!*vlan_tci) return -ENODATA; diff --git a/drivers/net/ethernet/intel/ice/ice_txrx_lib.h b/drivers/net/ethernet/intel/ice/ice_txrx_lib.h index 99717730f21a..6a3f10f7a53f 100644 --- a/drivers/net/ethernet/intel/ice/ice_txrx_lib.h +++ b/drivers/net/ethernet/intel/ice/ice_txrx_lib.h @@ -135,13 +135,4 @@ ice_process_skb_fields(struct ice_rx_ring *rx_ring, void ice_receive_skb(struct ice_rx_ring *rx_ring, struct sk_buff *skb, u16 vlan_tci); -static inline void -ice_xdp_meta_set_desc(struct xdp_buff *xdp, - union ice_32b_rx_flex_desc *eop_desc) -{ - struct ice_xdp_buff *xdp_ext = container_of(xdp, struct ice_xdp_buff, - xdp_buff); - - xdp_ext->eop_desc = eop_desc; -} #endif /* !_ICE_TXRX_LIB_H_ */ diff --git a/drivers/net/ethernet/intel/ice/ice_xsk.c b/drivers/net/ethernet/intel/ice/ice_xsk.c index b25bc5ba40ab..989ff1fd9110 100644 --- a/drivers/net/ethernet/intel/ice/ice_xsk.c +++ b/drivers/net/ethernet/intel/ice/ice_xsk.c @@ -3,6 +3,7 @@ #include #include +#include #include #include #include "ice.h" @@ -169,50 +170,18 @@ ice_xsk_pool_enable(struct ice_vsi *vsi, struct xsk_buff_pool *pool, u16 qid) * If allocation was successful, substitute buffer with allocated one. * Returns 0 on success, negative on failure */ -static int +int ice_realloc_rx_xdp_bufs(struct ice_rx_ring *rx_ring, bool pool_present) { - size_t elem_size = pool_present ? sizeof(*rx_ring->xdp_buf) : - sizeof(*rx_ring->rx_buf); - void *sw_ring = kcalloc(rx_ring->count, elem_size, GFP_KERNEL); - - if (!sw_ring) - return -ENOMEM; - if (pool_present) { - kfree(rx_ring->rx_buf); - rx_ring->rx_buf = NULL; - rx_ring->xdp_buf = sw_ring; + rx_ring->xdp_buf = kcalloc(rx_ring->count, + sizeof(*rx_ring->xdp_buf), + GFP_KERNEL); + if (!rx_ring->xdp_buf) + return -ENOMEM; } else { kfree(rx_ring->xdp_buf); rx_ring->xdp_buf = NULL; - rx_ring->rx_buf = sw_ring; - } - - return 0; -} - -/** - * ice_realloc_zc_buf - reallocate XDP ZC queue pairs - * @vsi: Current VSI - * @zc: is zero copy set - * - * Reallocate buffer for rx_rings that might be used by XSK. - * XDP requires more memory, than rx_buf provides. - * Returns 0 on success, negative on failure - */ -int ice_realloc_zc_buf(struct ice_vsi *vsi, bool zc) -{ - struct ice_rx_ring *rx_ring; - uint i; - - ice_for_each_rxq(vsi, i) { - rx_ring = vsi->rx_rings[i]; - if (!rx_ring->xsk_pool) - continue; - - if (ice_realloc_rx_xdp_bufs(rx_ring, zc)) - return -ENOMEM; } return 0; @@ -228,6 +197,7 @@ int ice_realloc_zc_buf(struct ice_vsi *vsi, bool zc) */ int ice_xsk_pool_setup(struct ice_vsi *vsi, struct xsk_buff_pool *pool, u16 qid) { + struct ice_rx_ring *rx_ring = vsi->rx_rings[qid]; bool if_running, pool_present = !!pool; int ret = 0, pool_failure = 0; @@ -241,8 +211,6 @@ int ice_xsk_pool_setup(struct ice_vsi *vsi, struct xsk_buff_pool *pool, u16 qid) ice_is_xdp_ena_vsi(vsi); if (if_running) { - struct ice_rx_ring *rx_ring = vsi->rx_rings[qid]; - ret = ice_qp_dis(vsi, qid); if (ret) { netdev_err(vsi->netdev, "ice_qp_dis error = %d\n", ret); @@ -303,11 +271,6 @@ static u16 ice_fill_rx_descs(struct xsk_buff_pool *pool, struct xdp_buff **xdp, rx_desc->read.pkt_addr = cpu_to_le64(dma); rx_desc->wb.status_error0 = 0; - /* Put private info that changes on a per-packet basis - * into xdp_buff_xsk->cb. - */ - ice_xdp_meta_set_desc(*xdp, rx_desc); - rx_desc++; xdp++; } @@ -606,10 +569,10 @@ int ice_clean_rx_irq_zc(struct ice_rx_ring *rx_ring, struct xsk_buff_pool *xsk_pool, int budget) { + struct xdp_buff *first = (struct xdp_buff *)rx_ring->xsk; unsigned int total_rx_bytes = 0, total_rx_packets = 0; u32 ntc = rx_ring->next_to_clean; u32 ntu = rx_ring->next_to_use; - struct xdp_buff *first = NULL; struct ice_tx_ring *xdp_ring; unsigned int xdp_xmit = 0; struct bpf_prog *xdp_prog; @@ -623,9 +586,6 @@ int ice_clean_rx_irq_zc(struct ice_rx_ring *rx_ring, xdp_prog = READ_ONCE(rx_ring->xdp_prog); xdp_ring = rx_ring->xdp_ring; - if (ntc != rx_ring->first_desc) - first = *ice_xdp_buf(rx_ring, rx_ring->first_desc); - while (likely(total_rx_packets < (unsigned int)budget)) { union ice_32b_rx_flex_desc *rx_desc; unsigned int size, xdp_res = 0; @@ -661,15 +621,17 @@ int ice_clean_rx_irq_zc(struct ice_rx_ring *rx_ring, first = xdp; } else if (likely(size) && !xsk_buff_add_frag(first, xdp)) { xsk_buff_free(first); - break; + first = NULL; } if (++ntc == cnt) ntc = 0; - if (ice_is_non_eop(rx_ring, rx_desc)) + if (ice_is_non_eop(rx_ring, rx_desc) || unlikely(!first)) continue; + ((struct libeth_xdp_buff *)first)->desc = rx_desc; + xdp_res = ice_run_xdp_zc(rx_ring, first, xdp_prog, xdp_ring, xsk_pool); if (likely(xdp_res & (ICE_XDP_TX | ICE_XDP_REDIR))) { @@ -677,7 +639,6 @@ int ice_clean_rx_irq_zc(struct ice_rx_ring *rx_ring, } else if (xdp_res == ICE_XDP_EXIT) { failure = true; first = NULL; - rx_ring->first_desc = ntc; break; } else if (xdp_res == ICE_XDP_CONSUMED) { xsk_buff_free(first); @@ -689,7 +650,6 @@ int ice_clean_rx_irq_zc(struct ice_rx_ring *rx_ring, total_rx_packets++; first = NULL; - rx_ring->first_desc = ntc; continue; construct_skb: @@ -697,12 +657,14 @@ int ice_clean_rx_irq_zc(struct ice_rx_ring *rx_ring, skb = xdp_build_skb_from_zc(first); if (!skb) { xsk_buff_free(first); + first = NULL; + rx_ring->ring_stats->rx_stats.alloc_buf_failed++; - break; + continue; } first = NULL; - rx_ring->first_desc = ntc; + total_rx_bytes += skb->len; total_rx_packets++; @@ -713,7 +675,9 @@ int ice_clean_rx_irq_zc(struct ice_rx_ring *rx_ring, } rx_ring->next_to_clean = ntc; - entries_to_alloc = ICE_RX_DESC_UNUSED(rx_ring); + rx_ring->xsk = (struct libeth_xdp_buff *)first; + + entries_to_alloc = ICE_DESC_UNUSED(rx_ring); if (entries_to_alloc > ICE_RING_QUARTER(rx_ring)) failure |= !ice_alloc_rx_bufs_zc(rx_ring, xsk_pool, entries_to_alloc); diff --git a/drivers/net/ethernet/intel/ice/ice_xsk.h b/drivers/net/ethernet/intel/ice/ice_xsk.h index 600cbeeaa203..5275fcedc9e1 100644 --- a/drivers/net/ethernet/intel/ice/ice_xsk.h +++ b/drivers/net/ethernet/intel/ice/ice_xsk.h @@ -22,7 +22,7 @@ bool ice_xsk_any_rx_ring_ena(struct ice_vsi *vsi); void ice_xsk_clean_rx_ring(struct ice_rx_ring *rx_ring); void ice_xsk_clean_xdp_ring(struct ice_tx_ring *xdp_ring); bool ice_xmit_zc(struct ice_tx_ring *xdp_ring, struct xsk_buff_pool *xsk_pool); -int ice_realloc_zc_buf(struct ice_vsi *vsi, bool zc); +int ice_realloc_rx_xdp_bufs(struct ice_rx_ring *rx_ring, bool pool_present); void ice_qvec_cfg_msix(struct ice_vsi *vsi, struct ice_q_vector *q_vector, u16 qid); void ice_qvec_toggle_napi(struct ice_vsi *vsi, struct ice_q_vector *q_vector, @@ -77,8 +77,8 @@ static inline void ice_xsk_clean_rx_ring(struct ice_rx_ring *rx_ring) { } static inline void ice_xsk_clean_xdp_ring(struct ice_tx_ring *xdp_ring) { } static inline int -ice_realloc_zc_buf(struct ice_vsi __always_unused *vsi, - bool __always_unused zc) +ice_realloc_rx_xdp_bufs(struct ice_rx_ring *rx_ring, + bool __always_unused pool_present) { return 0; } From 8adfcfd6a2eedbe4007ad6732bed829f41ec720f Mon Sep 17 00:00:00 2001 From: Alexander Lobakin Date: Mon, 6 Oct 2025 18:20:53 +0200 Subject: [PATCH 0349/1381] ice: implement configurable header split for regular Rx Add second page_pool for header buffers to each Rx queue and ability to toggle the header split on/off using Ethtool (default to off to match the current behaviour). Unlike idpf, all HW backed up by ice doesn't require any W/As and correctly splits all types of packets as configured: after L4 headers for TCP/UDP/SCTP, after L3 headers for other IPv4/IPv6 frames, after the Ethernet header otherwise (in case of tunneling, same as above, but after innermost headers). This doesn't affect the XSk path as there are no benefits of having it there. Signed-off-by: Alexander Lobakin Tested-by: Alexander Nowlin Reviewed-by: Simon Horman Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice.h | 1 + drivers/net/ethernet/intel/ice/ice_base.c | 89 +++++++++++++++---- drivers/net/ethernet/intel/ice/ice_ethtool.c | 15 +++- .../net/ethernet/intel/ice/ice_lan_tx_rx.h | 3 + drivers/net/ethernet/intel/ice/ice_txrx.c | 89 +++++++++++++++---- drivers/net/ethernet/intel/ice/ice_txrx.h | 7 ++ 6 files changed, 168 insertions(+), 36 deletions(-) diff --git a/drivers/net/ethernet/intel/ice/ice.h b/drivers/net/ethernet/intel/ice/ice.h index 3d4d8b88631b..147aaee192a7 100644 --- a/drivers/net/ethernet/intel/ice/ice.h +++ b/drivers/net/ethernet/intel/ice/ice.h @@ -351,6 +351,7 @@ struct ice_vsi { u16 num_q_vectors; /* tell if only dynamic irq allocation is allowed */ bool irq_dyn_alloc; + bool hsplit:1; u16 vsi_num; /* HW (absolute) index of this VSI */ u16 idx; /* software index in pf->vsi[] */ diff --git a/drivers/net/ethernet/intel/ice/ice_base.c b/drivers/net/ethernet/intel/ice/ice_base.c index eabab50fab33..eadb1e3d12b3 100644 --- a/drivers/net/ethernet/intel/ice/ice_base.c +++ b/drivers/net/ethernet/intel/ice/ice_base.c @@ -524,8 +524,29 @@ static int ice_setup_rx_ctx(struct ice_rx_ring *ring) else rlan_ctx.l2tsel = 1; - rlan_ctx.dtype = ICE_RX_DTYPE_NO_SPLIT; - rlan_ctx.hsplit_0 = ICE_RLAN_RX_HSPLIT_0_NO_SPLIT; + if (ring->hdr_pp) { + rlan_ctx.hbuf = ring->rx_hdr_len >> ICE_RLAN_CTX_HBUF_S; + rlan_ctx.dtype = ICE_RX_DTYPE_HEADER_SPLIT; + + /* + * If the frame is TCP/UDP/SCTP, it will be split by the + * payload. + * If not, but it's an IPv4/IPv6 frame, it will be split by + * the IP header. + * If not IP, it will be split by the Ethernet header. + * + * In any case, the header buffer will never be left empty. + */ + rlan_ctx.hsplit_0 = ICE_RLAN_RX_HSPLIT_0_SPLIT_L2 | + ICE_RLAN_RX_HSPLIT_0_SPLIT_IP | + ICE_RLAN_RX_HSPLIT_0_SPLIT_TCP_UDP | + ICE_RLAN_RX_HSPLIT_0_SPLIT_SCTP; + } else { + rlan_ctx.hbuf = 0; + rlan_ctx.dtype = ICE_RX_DTYPE_NO_SPLIT; + rlan_ctx.hsplit_0 = ICE_RLAN_RX_HSPLIT_0_NO_SPLIT; + } + rlan_ctx.hsplit_1 = ICE_RLAN_RX_HSPLIT_1_NO_SPLIT; /* This controls whether VLAN is stripped from inner headers @@ -581,6 +602,53 @@ static int ice_setup_rx_ctx(struct ice_rx_ring *ring) return 0; } +static int ice_rxq_pp_create(struct ice_rx_ring *rq) +{ + struct libeth_fq fq = { + .count = rq->count, + .nid = NUMA_NO_NODE, + .hsplit = rq->vsi->hsplit, + .xdp = ice_is_xdp_ena_vsi(rq->vsi), + .buf_len = LIBIE_MAX_RX_BUF_LEN, + }; + int err; + + err = libeth_rx_fq_create(&fq, &rq->q_vector->napi); + if (err) + return err; + + rq->pp = fq.pp; + rq->rx_fqes = fq.fqes; + rq->truesize = fq.truesize; + rq->rx_buf_len = fq.buf_len; + + if (!fq.hsplit) + return 0; + + fq = (struct libeth_fq){ + .count = rq->count, + .type = LIBETH_FQE_HDR, + .nid = NUMA_NO_NODE, + .xdp = ice_is_xdp_ena_vsi(rq->vsi), + }; + + err = libeth_rx_fq_create(&fq, &rq->q_vector->napi); + if (err) + goto destroy; + + rq->hdr_pp = fq.pp; + rq->hdr_fqes = fq.fqes; + rq->hdr_truesize = fq.truesize; + rq->rx_hdr_len = fq.buf_len; + + return 0; + +destroy: + ice_rxq_pp_destroy(rq); + + return err; +} + /** * ice_vsi_cfg_rxq - Configure an Rx queue * @ring: the ring being configured @@ -589,12 +657,6 @@ static int ice_setup_rx_ctx(struct ice_rx_ring *ring) */ static int ice_vsi_cfg_rxq(struct ice_rx_ring *ring) { - struct libeth_fq fq = { - .count = ring->count, - .nid = NUMA_NO_NODE, - .xdp = ice_is_xdp_ena_vsi(ring->vsi), - .buf_len = LIBIE_MAX_RX_BUF_LEN, - }; struct device *dev = ice_pf_to_dev(ring->vsi->back); u32 num_bufs = ICE_DESC_UNUSED(ring); u32 rx_buf_len; @@ -636,15 +698,10 @@ static int ice_vsi_cfg_rxq(struct ice_rx_ring *ring) dev_info(dev, "Registered XDP mem model MEM_TYPE_XSK_BUFF_POOL on Rx ring %d\n", ring->q_index); } else { - err = libeth_rx_fq_create(&fq, &ring->q_vector->napi); + err = ice_rxq_pp_create(ring); if (err) return err; - ring->pp = fq.pp; - ring->rx_fqes = fq.fqes; - ring->truesize = fq.truesize; - ring->rx_buf_len = fq.buf_len; - if (!xdp_rxq_info_is_reg(&ring->xdp_rxq)) { err = __xdp_rxq_info_reg(&ring->xdp_rxq, ring->netdev, ring->q_index, @@ -699,9 +756,7 @@ static int ice_vsi_cfg_rxq(struct ice_rx_ring *ring) return 0; err_destroy_fq: - libeth_rx_fq_destroy(&fq); - ring->rx_fqes = NULL; - ring->pp = NULL; + ice_rxq_pp_destroy(ring); return err; } diff --git a/drivers/net/ethernet/intel/ice/ice_ethtool.c b/drivers/net/ethernet/intel/ice/ice_ethtool.c index 36fdac4fddc3..a1d9abee97e5 100644 --- a/drivers/net/ethernet/intel/ice/ice_ethtool.c +++ b/drivers/net/ethernet/intel/ice/ice_ethtool.c @@ -3151,6 +3151,10 @@ ice_get_ringparam(struct net_device *netdev, struct ethtool_ringparam *ring, ring->rx_jumbo_max_pending = 0; ring->rx_mini_pending = 0; ring->rx_jumbo_pending = 0; + + kernel_ring->tcp_data_split = vsi->hsplit ? + ETHTOOL_TCP_DATA_SPLIT_ENABLED : + ETHTOOL_TCP_DATA_SPLIT_DISABLED; } static int @@ -3167,6 +3171,7 @@ ice_set_ringparam(struct net_device *netdev, struct ethtool_ringparam *ring, int i, timeout = 50, err = 0; struct ice_hw *hw = &pf->hw; u16 new_rx_cnt, new_tx_cnt; + bool hsplit; if (ring->tx_pending > ICE_MAX_NUM_DESC_BY_MAC(hw) || ring->tx_pending < ICE_MIN_NUM_DESC || @@ -3192,9 +3197,12 @@ ice_set_ringparam(struct net_device *netdev, struct ethtool_ringparam *ring, netdev_info(netdev, "Requested Rx descriptor count rounded up to %d\n", new_rx_cnt); + hsplit = kernel_ring->tcp_data_split == ETHTOOL_TCP_DATA_SPLIT_ENABLED; + /* if nothing to do return success */ if (new_tx_cnt == vsi->tx_rings[0]->count && - new_rx_cnt == vsi->rx_rings[0]->count) { + new_rx_cnt == vsi->rx_rings[0]->count && + hsplit == vsi->hsplit) { netdev_dbg(netdev, "Nothing to change, descriptor count is same as requested\n"); return 0; } @@ -3224,6 +3232,8 @@ ice_set_ringparam(struct net_device *netdev, struct ethtool_ringparam *ring, vsi->xdp_rings[i]->count = new_tx_cnt; vsi->num_tx_desc = (u16)new_tx_cnt; vsi->num_rx_desc = (u16)new_rx_cnt; + vsi->hsplit = hsplit; + netdev_dbg(netdev, "Link is down, descriptor count change happens when link is brought up\n"); goto done; } @@ -3330,6 +3340,8 @@ ice_set_ringparam(struct net_device *netdev, struct ethtool_ringparam *ring, } process_link: + vsi->hsplit = hsplit; + /* Bring interface down, copy in the new ring info, then restore the * interface. if VSI is up, bring it down and then back up */ @@ -4811,6 +4823,7 @@ static const struct ethtool_ops ice_ethtool_ops = { ETHTOOL_COALESCE_USE_ADAPTIVE | ETHTOOL_COALESCE_RX_USECS_HIGH, .supported_input_xfrm = RXH_XFRM_SYM_XOR, + .supported_ring_params = ETHTOOL_RING_USE_TCP_DATA_SPLIT, .get_link_ksettings = ice_get_link_ksettings, .set_link_ksettings = ice_set_link_ksettings, .get_fec_stats = ice_get_fec_stats, diff --git a/drivers/net/ethernet/intel/ice/ice_lan_tx_rx.h b/drivers/net/ethernet/intel/ice/ice_lan_tx_rx.h index 10c312d49e05..185672c7e17d 100644 --- a/drivers/net/ethernet/intel/ice/ice_lan_tx_rx.h +++ b/drivers/net/ethernet/intel/ice/ice_lan_tx_rx.h @@ -342,6 +342,9 @@ enum ice_flg64_bits { /* for ice_32byte_rx_flex_desc.pkt_length member */ #define ICE_RX_FLX_DESC_PKT_LEN_M (0x3FFF) /* 14-bits */ +/* ice_32byte_rx_flex_desc::hdr_len_sph_flex_flags1 */ +#define ICE_RX_FLEX_DESC_HDR_LEN_M GENMASK(10, 0) + enum ice_rx_flex_desc_status_error_0_bits { /* Note: These are predefined bit offsets */ ICE_RX_FLEX_DESC_STATUS0_DD_S = 0, diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.c b/drivers/net/ethernet/intel/ice/ice_txrx.c index 5a966138eacf..ad76768a4232 100644 --- a/drivers/net/ethernet/intel/ice/ice_txrx.c +++ b/drivers/net/ethernet/intel/ice/ice_txrx.c @@ -508,16 +508,34 @@ int ice_setup_tx_ring(struct ice_tx_ring *tx_ring) return -ENOMEM; } +void ice_rxq_pp_destroy(struct ice_rx_ring *rq) +{ + struct libeth_fq fq = { + .fqes = rq->rx_fqes, + .pp = rq->pp, + }; + + libeth_rx_fq_destroy(&fq); + rq->rx_fqes = NULL; + rq->pp = NULL; + + if (!rq->hdr_pp) + return; + + fq.fqes = rq->hdr_fqes; + fq.pp = rq->hdr_pp; + + libeth_rx_fq_destroy(&fq); + rq->hdr_fqes = NULL; + rq->hdr_pp = NULL; +} + /** * ice_clean_rx_ring - Free Rx buffers * @rx_ring: ring to be cleaned */ void ice_clean_rx_ring(struct ice_rx_ring *rx_ring) { - struct libeth_fq fq = { - .fqes = rx_ring->rx_fqes, - .pp = rx_ring->pp, - }; u32 size; if (rx_ring->xsk_pool) { @@ -533,9 +551,10 @@ void ice_clean_rx_ring(struct ice_rx_ring *rx_ring) /* Free all the Rx ring sk_buffs */ for (u32 i = rx_ring->next_to_clean; i != rx_ring->next_to_use; ) { - const struct libeth_fqe *rx_fqes = &rx_ring->rx_fqes[i]; + libeth_rx_recycle_slow(rx_ring->rx_fqes[i].netmem); - libeth_rx_recycle_slow(rx_fqes->netmem); + if (rx_ring->hdr_pp) + libeth_rx_recycle_slow(rx_ring->hdr_fqes[i].netmem); if (unlikely(++i == rx_ring->count)) i = 0; @@ -547,12 +566,9 @@ void ice_clean_rx_ring(struct ice_rx_ring *rx_ring) xdp_rxq_info_unreg(&rx_ring->xdp_rxq); } - libeth_rx_fq_destroy(&fq); - rx_ring->rx_fqes = NULL; - rx_ring->pp = NULL; + ice_rxq_pp_destroy(rx_ring); rx_skip_free: - /* Zero out the descriptor ring */ size = ALIGN(rx_ring->count * sizeof(union ice_32byte_rx_desc), PAGE_SIZE); @@ -806,6 +822,12 @@ void ice_init_ctrl_rx_descs(struct ice_rx_ring *rx_ring, u32 count) */ bool ice_alloc_rx_bufs(struct ice_rx_ring *rx_ring, unsigned int cleaned_count) { + const struct libeth_fq_fp hdr_fq = { + .pp = rx_ring->hdr_pp, + .fqes = rx_ring->hdr_fqes, + .truesize = rx_ring->hdr_truesize, + .count = rx_ring->count, + }; const struct libeth_fq_fp fq = { .pp = rx_ring->pp, .fqes = rx_ring->rx_fqes, @@ -836,6 +858,20 @@ bool ice_alloc_rx_bufs(struct ice_rx_ring *rx_ring, unsigned int cleaned_count) */ rx_desc->read.pkt_addr = cpu_to_le64(addr); + if (!hdr_fq.pp) + goto next; + + addr = libeth_rx_alloc(&hdr_fq, ntu); + if (addr == DMA_MAPPING_ERROR) { + rx_ring->ring_stats->rx_stats.alloc_page_failed++; + + libeth_rx_recycle_slow(fq.fqes[ntu].netmem); + break; + } + + rx_desc->read.hdr_addr = cpu_to_le64(addr); + +next: rx_desc++; ntu++; if (unlikely(ntu == rx_ring->count)) { @@ -933,14 +969,16 @@ static int ice_clean_rx_irq(struct ice_rx_ring *rx_ring, int budget) unsigned int size; u16 stat_err_bits; u16 vlan_tci; + bool rxe; /* get the Rx desc from Rx ring based on 'next_to_clean' */ rx_desc = ICE_RX_DESC(rx_ring, ntc); - /* status_error_len will always be zero for unused descriptors - * because it's cleared in cleanup, and overlaps with hdr_addr - * which is always zero because packet split isn't used, if the - * hardware wrote DD then it will be non-zero + /* + * The DD bit will always be zero for unused descriptors + * because it's cleared in cleanup or when setting the DMA + * address of the header buffer, which never uses the DD bit. + * If the hardware wrote the descriptor, it will be non-zero. */ stat_err_bits = BIT(ICE_RX_FLEX_DESC_STATUS0_DD_S); if (!ice_test_staterr(rx_desc->wb.status_error0, stat_err_bits)) @@ -954,12 +992,27 @@ static int ice_clean_rx_irq(struct ice_rx_ring *rx_ring, int budget) ice_trace(clean_rx_irq, rx_ring, rx_desc); + stat_err_bits = BIT(ICE_RX_FLEX_DESC_STATUS0_HBO_S) | + BIT(ICE_RX_FLEX_DESC_STATUS0_RXE_S); + rxe = ice_test_staterr(rx_desc->wb.status_error0, + stat_err_bits); + + if (!rx_ring->hdr_pp) + goto payload; + + size = le16_get_bits(rx_desc->wb.hdr_len_sph_flex_flags1, + ICE_RX_FLEX_DESC_HDR_LEN_M); + if (unlikely(rxe)) + size = 0; + + rx_buf = &rx_ring->hdr_fqes[ntc]; + libeth_xdp_process_buff(xdp, rx_buf, size); + rx_buf->netmem = 0; + +payload: size = le16_to_cpu(rx_desc->wb.pkt_len) & ICE_RX_FLX_DESC_PKT_LEN_M; - - stat_err_bits = BIT(ICE_RX_FLEX_DESC_STATUS0_RXE_S); - if (unlikely(ice_test_staterr(rx_desc->wb.status_error0, - stat_err_bits))) + if (unlikely(rxe)) size = 0; /* retrieve a buffer from the ring */ diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.h b/drivers/net/ethernet/intel/ice/ice_txrx.h index e97a38ef3fe7..e440c55d9e9f 100644 --- a/drivers/net/ethernet/intel/ice/ice_txrx.h +++ b/drivers/net/ethernet/intel/ice/ice_txrx.h @@ -255,6 +255,9 @@ struct ice_rx_ring { }; /* CL2 - 2nd cacheline starts here */ + struct libeth_fqe *hdr_fqes; + struct page_pool *hdr_pp; + union { struct libeth_xdp_buff_stash xdp; struct libeth_xdp_buff *xsk; @@ -273,6 +276,8 @@ struct ice_rx_ring { /* used in interrupt processing */ u16 next_to_use; u16 next_to_clean; + + u32 hdr_truesize; u32 truesize; /* stats structs */ @@ -284,6 +289,7 @@ struct ice_rx_ring { struct ice_tx_ring *xdp_ring; struct ice_rx_ring *next; /* pointer to next ring in q_vector */ struct xsk_buff_pool *xsk_pool; + u16 rx_hdr_len; u16 rx_buf_len; dma_addr_t dma; /* physical address of ring */ u8 dcb_tc; /* Traffic class of ring */ @@ -396,6 +402,7 @@ static inline unsigned int ice_rx_pg_order(struct ice_rx_ring *ring) union ice_32b_rx_flex_desc; void ice_init_ctrl_rx_descs(struct ice_rx_ring *rx_ring, u32 num_descs); +void ice_rxq_pp_destroy(struct ice_rx_ring *rq); bool ice_alloc_rx_bufs(struct ice_rx_ring *rxr, unsigned int cleaned_count); netdev_tx_t ice_start_xmit(struct sk_buff *skb, struct net_device *netdev); u16 From ba2807b869a16d36b8bc76efa774fe433b3c45de Mon Sep 17 00:00:00 2001 From: Grzegorz Nitka Date: Mon, 24 Feb 2025 21:59:24 +0100 Subject: [PATCH 0350/1381] ice: Allow 100M speed for E825C SGMII device Add E825C 10GbE SGMII device to the list of devices supporting 100Mbit link mode. Without that change, 100Mbit link mode is ignored in ethtool interface. This change was missed while adding the support for E825C devices family. Testing hints (please note, for previous version, 100baseT/Full entry was missing): [root@localhost]# ethtool eth3 Settings for eth3: Supported ports: [ TP ] Supported link modes: 100baseT/Full 1000baseT/Full 10000baseT/Full Supported pause frame use: Symmetric Supports auto-negotiation: Yes Supported FEC modes: None Advertised link modes: 100baseT/Full 1000baseT/Full 10000baseT/Full ... Signed-off-by: Grzegorz Nitka Reviewed-by: Aleksandr Loktionov Reviewed-by: Paul Menzel Reviewed-by: Simon Horman Tested-by: Rinitha S (A Contingent worker at Intel) Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_common.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/intel/ice/ice_common.c b/drivers/net/ethernet/intel/ice/ice_common.c index b097cc8b175c..83f5217bce9f 100644 --- a/drivers/net/ethernet/intel/ice/ice_common.c +++ b/drivers/net/ethernet/intel/ice/ice_common.c @@ -3392,6 +3392,7 @@ bool ice_is_100m_speed_supported(struct ice_hw *hw) case ICE_DEV_ID_E822L_SGMII: case ICE_DEV_ID_E823L_1GBE: case ICE_DEV_ID_E823C_SGMII: + case ICE_DEV_ID_E825C_SGMII: return true; default: return false; From a7ae783da0b919550e260aebfca1c6ef030b99a4 Mon Sep 17 00:00:00 2001 From: Jay Vosburgh Date: Wed, 22 Oct 2025 18:11:43 +0200 Subject: [PATCH 0351/1381] i40e: avoid redundant VF link state updates Multiple sources can request VF link state changes with identical parameters. For example, OpenStack Neutron may request to set the VF link state to IFLA_VF_LINK_STATE_AUTO during every initialization or user can issue: `ip link set vf 0 state auto` multiple times. Currently, the i40e driver processes each of these requests, even if the requested state is the same as the current one. This leads to unnecessary VF resets and can cause performance degradation or instability in the VF driver, particularly in environment using Data Plane Development Kit (DPDK). With this patch i40e will skip VF link state change requests when the desired link state matches the current configuration. This prevents unnecessary VF resets and reduces PF-VF communication overhead. To reproduce the problem run following command multiple times on the same interface: 'ip link set vf 0 state auto' Every time command is executed, PF driver will trigger VF reset. Co-developed-by: Robert Malz Signed-off-by: Robert Malz Signed-off-by: Jay Vosburgh Reviewed-by: Aleksandr Loktionov Tested-by: Rafal Romanowski Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c index 081a4526a2f0..0fe0d52c796b 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c +++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c @@ -4788,6 +4788,7 @@ int i40e_ndo_set_vf_link_state(struct net_device *netdev, int vf_id, int link) unsigned long q_map; struct i40e_vf *vf; int abs_vf_id; + int old_link; int ret = 0; int tmp; @@ -4806,6 +4807,17 @@ int i40e_ndo_set_vf_link_state(struct net_device *netdev, int vf_id, int link) vf = &pf->vf[vf_id]; abs_vf_id = vf->vf_id + hw->func_caps.vf_base_id; + /* skip VF link state change if requested state is already set */ + if (!vf->link_forced) + old_link = IFLA_VF_LINK_STATE_AUTO; + else if (vf->link_up) + old_link = IFLA_VF_LINK_STATE_ENABLE; + else + old_link = IFLA_VF_LINK_STATE_DISABLE; + + if (link == old_link) + goto error_out; + pfe.event = VIRTCHNL_EVENT_LINK_CHANGE; pfe.severity = PF_EVENT_SEVERITY_INFO; From 5d9b400e6f7e2bfb2bd80709e03a6db42e34076b Mon Sep 17 00:00:00 2001 From: Sreedevi Joshi Date: Fri, 19 Sep 2025 11:59:05 -0500 Subject: [PATCH 0352/1381] idpf: remove duplicate defines in IDPF_CAP_RSS Remove duplicate defines from the OR operation when defining IDPF_CAP_RSS. Duplicate definitions were introduced when IDPF_CAP_RSS was originally defined and were left behind and went unnoticed during a previous commit that renamed them. Review of the original out-of-tree code confirms these duplicates were the result of a typing error. Remove the duplicates to clean up the code and avoid potential confusion. Also verify no other duplicate occurrences of these defines exist elsewhere in the codebase. Reviewed-by: Aleksandr Loktionov Reviewed-by: Przemek Kitszel Signed-off-by: Sreedevi Joshi Tested-by: Samuel Salin Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/idpf/idpf.h | 2 -- 1 file changed, 2 deletions(-) diff --git a/drivers/net/ethernet/intel/idpf/idpf.h b/drivers/net/ethernet/intel/idpf/idpf.h index ca4da0c89979..50fa7be0c00d 100644 --- a/drivers/net/ethernet/intel/idpf/idpf.h +++ b/drivers/net/ethernet/intel/idpf/idpf.h @@ -734,13 +734,11 @@ static inline bool idpf_is_rdma_cap_ena(struct idpf_adapter *adapter) } #define IDPF_CAP_RSS (\ - VIRTCHNL2_FLOW_IPV4_TCP |\ VIRTCHNL2_FLOW_IPV4_TCP |\ VIRTCHNL2_FLOW_IPV4_UDP |\ VIRTCHNL2_FLOW_IPV4_SCTP |\ VIRTCHNL2_FLOW_IPV4_OTHER |\ VIRTCHNL2_FLOW_IPV6_TCP |\ - VIRTCHNL2_FLOW_IPV6_TCP |\ VIRTCHNL2_FLOW_IPV6_UDP |\ VIRTCHNL2_FLOW_IPV6_SCTP |\ VIRTCHNL2_FLOW_IPV6_OTHER) From 6ef670d833a8f3b142cb80739135c56ade8a25b0 Mon Sep 17 00:00:00 2001 From: Alok Tiwari Date: Tue, 21 Oct 2025 12:12:14 -0700 Subject: [PATCH 0353/1381] ixgbe: fix typos in ixgbe driver comments Corrected function reference: - "proc_autoc_read_82599" -> "prot_autoc_read_82599" Fixed spelling of: - "big-enian" -> "big-endian" - "Virtualiztion" -> "Virtualization" Signed-off-by: Alok Tiwari Reviewed-by: Simon Horman Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ixgbe/ixgbe_82599.c | 4 ++-- drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_82599.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_82599.c index d5b1b974b4a3..3069b583fd81 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_82599.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_82599.c @@ -198,7 +198,7 @@ static int prot_autoc_read_82599(struct ixgbe_hw *hw, bool *locked, * @hw: pointer to hardware structure * @autoc: value to write to AUTOC * @locked: bool to indicate whether the SW/FW lock was already taken by - * previous proc_autoc_read_82599. + * previous prot_autoc_read_82599. * * This part (82599) may need to hold a the SW/FW lock around all writes to * AUTOC. Likewise after a write we need to do a pipeline reset. @@ -1622,7 +1622,7 @@ int ixgbe_fdir_set_input_mask_82599(struct ixgbe_hw *hw, break; } - /* store source and destination IP masks (big-enian) */ + /* store source and destination IP masks (big-endian) */ IXGBE_WRITE_REG_BE32(hw, IXGBE_FDIRSIP4M, ~input_mask->formatted.src_ip[0]); IXGBE_WRITE_REG_BE32(hw, IXGBE_FDIRDIP4M, diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c index 170a29d162c6..a1d04914fbbc 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c @@ -318,7 +318,7 @@ static int ixgbe_xdp_queues(struct ixgbe_adapter *adapter) * ixgbe_set_dcb_sriov_queues: Allocate queues for SR-IOV devices w/ DCB * @adapter: board private structure to initialize * - * When SR-IOV (Single Root IO Virtualiztion) is enabled, allocate queues + * When SR-IOV (Single Root IO Virtualization) is enabled, allocate queues * and VM pools where appropriate. Also assign queues based on DCB * priorities and map accordingly.. * @@ -492,7 +492,7 @@ static bool ixgbe_set_dcb_queues(struct ixgbe_adapter *adapter) * ixgbe_set_sriov_queues - Allocate queues for SR-IOV devices * @adapter: board private structure to initialize * - * When SR-IOV (Single Root IO Virtualiztion) is enabled, allocate queues + * When SR-IOV (Single Root IO Virtualization) is enabled, allocate queues * and VM pools where appropriate. If RSS is available, then also try and * enable RSS and map accordingly. * From 9157b8a88c0bd769808caaecce4b8c96bd826304 Mon Sep 17 00:00:00 2001 From: Alok Tiwari Date: Tue, 21 Oct 2025 12:32:01 -0700 Subject: [PATCH 0354/1381] igbvf: fix misplaced newline in VLAN add warning message Corrected the dev_warn format string: - "Vlan id %d\n is not added" -> "Vlan id %d is not added\n" Signed-off-by: Alok Tiwari Reviewed-by: Simon Horman Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/igbvf/netdev.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/intel/igbvf/netdev.c b/drivers/net/ethernet/intel/igbvf/netdev.c index 61dfcd8cb370..ac57212ab02b 100644 --- a/drivers/net/ethernet/intel/igbvf/netdev.c +++ b/drivers/net/ethernet/intel/igbvf/netdev.c @@ -1235,7 +1235,7 @@ static int igbvf_vlan_rx_add_vid(struct net_device *netdev, spin_lock_bh(&hw->mbx_lock); if (hw->mac.ops.set_vfta(hw, vid, true)) { - dev_warn(&adapter->pdev->dev, "Vlan id %d\n is not added", vid); + dev_warn(&adapter->pdev->dev, "Vlan id %d is not added\n", vid); spin_unlock_bh(&hw->mbx_lock); return -EINVAL; } From f58abec23da51f03ffe25f9464f2ce2d1a2a592a Mon Sep 17 00:00:00 2001 From: Wang Liang Date: Sat, 25 Oct 2025 17:26:37 +0800 Subject: [PATCH 0355/1381] net: ipv4: Remove extern udp_v4_early_demux()/tcp_v4_early_demux() in .c files Function udp_v4_early_demux() was already declared in 'include/net/udp.h', no need to keep the extern in 'ip_input.c', which may produce the following checkpatch warning: WARNING: externs should be avoided in .c files #45: FILE: net/ipv4/ip_input.c:322: +enum skb_drop_reason udp_v4_early_demux(struct sk_buff *skb); Replace it by including 'net/udp.h'. Do the same for tcp_v4_early_demux(). Signed-off-by: Wang Liang Link: https://patch.msgid.link/20251025092637.1020960-1-wangliang74@huawei.com Signed-off-by: Jakub Kicinski --- net/ipv4/ip_input.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c index 273578579a6b..19d3141dad1f 100644 --- a/net/ipv4/ip_input.c +++ b/net/ipv4/ip_input.c @@ -141,6 +141,8 @@ #include #include #include +#include +#include /* * Process Router Attention IP option (RFC 2113) @@ -317,8 +319,6 @@ static bool ip_can_use_hint(const struct sk_buff *skb, const struct iphdr *iph, ip_hdr(hint)->tos == iph->tos; } -int tcp_v4_early_demux(struct sk_buff *skb); -enum skb_drop_reason udp_v4_early_demux(struct sk_buff *skb); static int ip_rcv_finish_core(struct net *net, struct sk_buff *skb, struct net_device *dev, const struct sk_buff *hint) From fc18b6e98cce2380e8f31b0e5089b11ceecb541d Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Wed, 29 Oct 2025 00:03:10 +0000 Subject: [PATCH 0356/1381] net: stmmac: move version handling into own function Move the version handling out of stmmac_hwif_init() and into its own function, returning the version information through a structure. Reviewed-by: Andrew Lunn Tested-by: Maxime Chevallier Tested-by: Mohd Ayaan Anwar Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vDteg-0000000CCBr-2m7q@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/hwif.c | 42 +++++++++++++++------- 1 file changed, 29 insertions(+), 13 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/hwif.c b/drivers/net/ethernet/stmicro/stmmac/hwif.c index 41a7e1841227..4924e74997e4 100644 --- a/drivers/net/ethernet/stmicro/stmmac/hwif.c +++ b/drivers/net/ethernet/stmicro/stmmac/hwif.c @@ -13,6 +13,11 @@ #include "dwmac4_descs.h" #include "dwxgmac2.h" +struct stmmac_version { + u8 snpsver; + u8 dev_id; +}; + static u32 stmmac_get_id(struct stmmac_priv *priv, u32 id_reg) { u32 reg = readl(priv->ioaddr + id_reg); @@ -40,6 +45,24 @@ static u32 stmmac_get_dev_id(struct stmmac_priv *priv, u32 id_reg) return (reg & GENMASK(15, 8)) >> 8; } +static void stmmac_get_version(struct stmmac_priv *priv, + struct stmmac_version *ver) +{ + enum dwmac_core_type core_type = priv->plat->core_type; + + ver->dev_id = 0; + + if (core_type == DWMAC_CORE_GMAC) { + ver->snpsver = stmmac_get_id(priv, GMAC_VERSION); + } else if (dwmac_is_xmac(core_type)) { + ver->snpsver = stmmac_get_id(priv, GMAC4_VERSION); + if (core_type == DWMAC_CORE_XGMAC) + ver->dev_id = stmmac_get_dev_id(priv, GMAC4_VERSION); + } else { + ver->snpsver = 0; + } +} + static void stmmac_dwmac_mode_quirk(struct stmmac_priv *priv) { struct mac_device_info *mac = priv->hw; @@ -292,23 +315,15 @@ int stmmac_hwif_init(struct stmmac_priv *priv) { enum dwmac_core_type core_type = priv->plat->core_type; const struct stmmac_hwif_entry *entry; + struct stmmac_version version; struct mac_device_info *mac; bool needs_setup = true; - u32 id, dev_id = 0; int i, ret; - if (core_type == DWMAC_CORE_GMAC) { - id = stmmac_get_id(priv, GMAC_VERSION); - } else if (dwmac_is_xmac(core_type)) { - id = stmmac_get_id(priv, GMAC4_VERSION); - if (core_type == DWMAC_CORE_XGMAC) - dev_id = stmmac_get_dev_id(priv, GMAC4_VERSION); - } else { - id = 0; - } + stmmac_get_version(priv, &version); /* Save ID for later use */ - priv->synopsys_id = id; + priv->synopsys_id = version.snpsver; /* Lets assume some safe values first */ if (core_type == DWMAC_CORE_GMAC4) { @@ -344,7 +359,8 @@ int stmmac_hwif_init(struct stmmac_priv *priv) /* Use synopsys_id var because some setups can override this */ if (priv->synopsys_id < entry->min_id) continue; - if (core_type == DWMAC_CORE_XGMAC && (dev_id ^ entry->dev_id)) + if (core_type == DWMAC_CORE_XGMAC && + (version.dev_id ^ entry->dev_id)) continue; /* Only use generic HW helpers if needed */ @@ -380,7 +396,7 @@ int stmmac_hwif_init(struct stmmac_priv *priv) } dev_err(priv->device, "Failed to find HW IF (id=0x%x, gmac=%d/%d)\n", - id, core_type == DWMAC_CORE_GMAC, + version.snpsver, core_type == DWMAC_CORE_GMAC, core_type == DWMAC_CORE_GMAC4); return -EINVAL; } From f49838f77cf6c50961afa33a5bc1ea95e061d7f0 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Wed, 29 Oct 2025 00:03:15 +0000 Subject: [PATCH 0357/1381] net: stmmac: simplify stmmac_get_version() We can simplify stmmac_get_version() by pre-initialising the version members to zero, detecting the MAC100 core and returning, otherwise determining the version register offset separately from calling stmmac_get_id() and stmmac_get_dev_id(). Do this. Reviewed-by: Andrew Lunn Tested-by: Maxime Chevallier Tested-by: Mohd Ayaan Anwar Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vDtel-0000000CCBx-3Lpf@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/hwif.c | 22 +++++++++++++--------- 1 file changed, 13 insertions(+), 9 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/hwif.c b/drivers/net/ethernet/stmicro/stmmac/hwif.c index 4924e74997e4..f6ada5a905fe 100644 --- a/drivers/net/ethernet/stmicro/stmmac/hwif.c +++ b/drivers/net/ethernet/stmicro/stmmac/hwif.c @@ -49,18 +49,22 @@ static void stmmac_get_version(struct stmmac_priv *priv, struct stmmac_version *ver) { enum dwmac_core_type core_type = priv->plat->core_type; + unsigned int version_offset; + ver->snpsver = 0; ver->dev_id = 0; - if (core_type == DWMAC_CORE_GMAC) { - ver->snpsver = stmmac_get_id(priv, GMAC_VERSION); - } else if (dwmac_is_xmac(core_type)) { - ver->snpsver = stmmac_get_id(priv, GMAC4_VERSION); - if (core_type == DWMAC_CORE_XGMAC) - ver->dev_id = stmmac_get_dev_id(priv, GMAC4_VERSION); - } else { - ver->snpsver = 0; - } + if (core_type == DWMAC_CORE_MAC100) + return; + + if (core_type == DWMAC_CORE_GMAC) + version_offset = GMAC_VERSION; + else + version_offset = GMAC4_VERSION; + + ver->snpsver = stmmac_get_id(priv, version_offset); + if (core_type == DWMAC_CORE_XGMAC) + ver->dev_id = stmmac_get_dev_id(priv, version_offset); } static void stmmac_dwmac_mode_quirk(struct stmmac_priv *priv) From c36b97e4ca773f20ff20db34ed97e35cc36ad97c Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Wed, 29 Oct 2025 00:03:20 +0000 Subject: [PATCH 0358/1381] net: stmmac: consolidate version reading and validation There is no need to read the version register twice, once in stmmac_get_id() and then again in stmmac_get_dev_id(). Consolidate this into stmmac_get_version() and pass each of these this value. As both functions unnecessarily issue the same warning for a zero register value, also move this into stmmac_get_version(). Reviewed-by: Andrew Lunn Tested-by: Maxime Chevallier Tested-by: Mohd Ayaan Anwar Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vDteq-0000000CCC3-3zbJ@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/hwif.c | 29 ++++++++-------------- 1 file changed, 11 insertions(+), 18 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/hwif.c b/drivers/net/ethernet/stmicro/stmmac/hwif.c index f6ada5a905fe..ffdc101ce3ce 100644 --- a/drivers/net/ethernet/stmicro/stmmac/hwif.c +++ b/drivers/net/ethernet/stmicro/stmmac/hwif.c @@ -18,30 +18,16 @@ struct stmmac_version { u8 dev_id; }; -static u32 stmmac_get_id(struct stmmac_priv *priv, u32 id_reg) +static u32 stmmac_get_id(struct stmmac_priv *priv, u32 reg) { - u32 reg = readl(priv->ioaddr + id_reg); - - if (!reg) { - dev_info(priv->device, "Version ID not available\n"); - return 0x0; - } - dev_info(priv->device, "User ID: 0x%x, Synopsys ID: 0x%x\n", (unsigned int)(reg & GENMASK(15, 8)) >> 8, (unsigned int)(reg & GENMASK(7, 0))); return reg & GENMASK(7, 0); } -static u32 stmmac_get_dev_id(struct stmmac_priv *priv, u32 id_reg) +static u32 stmmac_get_dev_id(struct stmmac_priv *priv, u32 reg) { - u32 reg = readl(priv->ioaddr + id_reg); - - if (!reg) { - dev_info(priv->device, "Version ID not available\n"); - return 0x0; - } - return (reg & GENMASK(15, 8)) >> 8; } @@ -50,6 +36,7 @@ static void stmmac_get_version(struct stmmac_priv *priv, { enum dwmac_core_type core_type = priv->plat->core_type; unsigned int version_offset; + u32 version; ver->snpsver = 0; ver->dev_id = 0; @@ -62,9 +49,15 @@ static void stmmac_get_version(struct stmmac_priv *priv, else version_offset = GMAC4_VERSION; - ver->snpsver = stmmac_get_id(priv, version_offset); + version = readl(priv->ioaddr + version_offset); + if (version == 0) { + dev_info(priv->device, "Version ID not available\n"); + return; + } + + ver->snpsver = stmmac_get_id(priv, version); if (core_type == DWMAC_CORE_XGMAC) - ver->dev_id = stmmac_get_dev_id(priv, version_offset); + ver->dev_id = stmmac_get_dev_id(priv, version); } static void stmmac_dwmac_mode_quirk(struct stmmac_priv *priv) From 7b2e41fff76f53b53e7e00222674e3d87cbd4a67 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Wed, 29 Oct 2025 00:03:26 +0000 Subject: [PATCH 0359/1381] net: stmmac: move stmmac_get_*id() into stmmac_get_version() Move the contents of both stmmac_get_id() and stmmac_get_dev_id() into stmmac_get_version() as it no longer makes sense for these to be separate functions. Reviewed-by: Andrew Lunn Tested-by: Maxime Chevallier Tested-by: Mohd Ayaan Anwar Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vDtew-0000000CCC9-0KeM@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/hwif.c | 21 ++++++--------------- 1 file changed, 6 insertions(+), 15 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/hwif.c b/drivers/net/ethernet/stmicro/stmmac/hwif.c index ffdc101ce3ce..a4df51a7aef1 100644 --- a/drivers/net/ethernet/stmicro/stmmac/hwif.c +++ b/drivers/net/ethernet/stmicro/stmmac/hwif.c @@ -18,19 +18,6 @@ struct stmmac_version { u8 dev_id; }; -static u32 stmmac_get_id(struct stmmac_priv *priv, u32 reg) -{ - dev_info(priv->device, "User ID: 0x%x, Synopsys ID: 0x%x\n", - (unsigned int)(reg & GENMASK(15, 8)) >> 8, - (unsigned int)(reg & GENMASK(7, 0))); - return reg & GENMASK(7, 0); -} - -static u32 stmmac_get_dev_id(struct stmmac_priv *priv, u32 reg) -{ - return (reg & GENMASK(15, 8)) >> 8; -} - static void stmmac_get_version(struct stmmac_priv *priv, struct stmmac_version *ver) { @@ -55,9 +42,13 @@ static void stmmac_get_version(struct stmmac_priv *priv, return; } - ver->snpsver = stmmac_get_id(priv, version); + dev_info(priv->device, "User ID: 0x%x, Synopsys ID: 0x%x\n", + (unsigned int)(version & GENMASK(15, 8)) >> 8, + (unsigned int)(version & GENMASK(7, 0))); + + ver->snpsver = version & GENMASK(7, 0); if (core_type == DWMAC_CORE_XGMAC) - ver->dev_id = stmmac_get_dev_id(priv, version); + ver->dev_id = (version & GENMASK(15, 8)) >> 8; } static void stmmac_dwmac_mode_quirk(struct stmmac_priv *priv) From b2fe9e29b5f65aa5ad87e859966871061eb37303 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Wed, 29 Oct 2025 00:03:31 +0000 Subject: [PATCH 0360/1381] net: stmmac: use FIELD_GET() for version register Provide field definitions in common.h, and use these with FIELD_GET() to extract the fields from the version register. Reviewed-by: Andrew Lunn Tested-by: Maxime Chevallier Tested-by: Mohd Ayaan Anwar Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vDtf1-0000000CCCF-0uUV@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/common.h | 3 +++ drivers/net/ethernet/stmicro/stmmac/hwif.c | 8 ++++---- 2 files changed, 7 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/common.h b/drivers/net/ethernet/stmicro/stmmac/common.h index 553a8897b005..27083af54568 100644 --- a/drivers/net/ethernet/stmicro/stmmac/common.h +++ b/drivers/net/ethernet/stmicro/stmmac/common.h @@ -26,6 +26,9 @@ #include "hwif.h" #include "mmc.h" +#define DWMAC_SNPSVER GENMASK_U32(7, 0) +#define DWMAC_USERVER GENMASK_U32(15, 8) + /* Synopsys Core versions */ #define DWMAC_CORE_3_40 0x34 #define DWMAC_CORE_3_50 0x35 diff --git a/drivers/net/ethernet/stmicro/stmmac/hwif.c b/drivers/net/ethernet/stmicro/stmmac/hwif.c index a4df51a7aef1..26cc1bc758bf 100644 --- a/drivers/net/ethernet/stmicro/stmmac/hwif.c +++ b/drivers/net/ethernet/stmicro/stmmac/hwif.c @@ -43,12 +43,12 @@ static void stmmac_get_version(struct stmmac_priv *priv, } dev_info(priv->device, "User ID: 0x%x, Synopsys ID: 0x%x\n", - (unsigned int)(version & GENMASK(15, 8)) >> 8, - (unsigned int)(version & GENMASK(7, 0))); + FIELD_GET(DWMAC_USERVER, version), + FIELD_GET(DWMAC_SNPSVER, version)); - ver->snpsver = version & GENMASK(7, 0); + ver->snpsver = FIELD_GET(DWMAC_SNPSVER, version); if (core_type == DWMAC_CORE_XGMAC) - ver->dev_id = (version & GENMASK(15, 8)) >> 8; + ver->dev_id = FIELD_GET(DWMAC_USERVER, version); } static void stmmac_dwmac_mode_quirk(struct stmmac_priv *priv) From 7b510ea8e58eb0ead6bc41fd0f15ec064312dcbf Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Wed, 29 Oct 2025 00:03:36 +0000 Subject: [PATCH 0361/1381] net: stmmac: provide function to lookup hwif Provide a function to lookup the hwif entry given the core type, Synopsys version, and device ID (used for XGMAC cores). Reviewed-by: Andrew Lunn Tested-by: Maxime Chevallier Tested-by: Mohd Ayaan Anwar Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vDtf6-0000000CCCL-1cQA@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/hwif.c | 40 +++++++++++++++------- 1 file changed, 28 insertions(+), 12 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/hwif.c b/drivers/net/ethernet/stmicro/stmmac/hwif.c index 26cc1bc758bf..892cef79c4d1 100644 --- a/drivers/net/ethernet/stmicro/stmmac/hwif.c +++ b/drivers/net/ethernet/stmicro/stmmac/hwif.c @@ -299,6 +299,30 @@ static const struct stmmac_hwif_entry { }, }; +static const struct stmmac_hwif_entry * +stmmac_hwif_find(enum dwmac_core_type core_type, u8 snpsver, u8 dev_id) +{ + const struct stmmac_hwif_entry *entry; + int i; + + for (i = ARRAY_SIZE(stmmac_hw) - 1; i >= 0; i--) { + entry = &stmmac_hw[i]; + + if (core_type != entry->core_type) + continue; + /* Use synopsys_id var because some setups can override this */ + if (snpsver < entry->min_id) + continue; + if (core_type == DWMAC_CORE_XGMAC && + (dev_id ^ entry->dev_id)) + continue; + + return entry; + } + + return NULL; +} + int stmmac_hwif_init(struct stmmac_priv *priv) { enum dwmac_core_type core_type = priv->plat->core_type; @@ -306,7 +330,7 @@ int stmmac_hwif_init(struct stmmac_priv *priv) struct stmmac_version version; struct mac_device_info *mac; bool needs_setup = true; - int i, ret; + int ret; stmmac_get_version(priv, &version); @@ -339,18 +363,10 @@ int stmmac_hwif_init(struct stmmac_priv *priv) spin_lock_init(&mac->irq_ctrl_lock); /* Fallback to generic HW */ - for (i = ARRAY_SIZE(stmmac_hw) - 1; i >= 0; i--) { - entry = &stmmac_hw[i]; - - if (core_type != entry->core_type) - continue; - /* Use synopsys_id var because some setups can override this */ - if (priv->synopsys_id < entry->min_id) - continue; - if (core_type == DWMAC_CORE_XGMAC && - (version.dev_id ^ entry->dev_id)) - continue; + /* Use synopsys_id var because some setups can override this */ + entry = stmmac_hwif_find(core_type, priv->synopsys_id, version.dev_id); + if (entry) { /* Only use generic HW helpers if needed */ mac->desc = mac->desc ? : entry->desc; mac->dma = mac->dma ? : entry->dma; From f9326b139b4c77128564574ea2d7b83c8a758114 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Wed, 29 Oct 2025 00:03:41 +0000 Subject: [PATCH 0362/1381] net: stmmac: use != rather than ^ for comparing dev_id Use the more usual not-equals rather than exclusive-or operator when comparing the dev_id in stmmac_hwif_find(). Reviewed-by: Andrew Lunn Tested-by: Maxime Chevallier Tested-by: Mohd Ayaan Anwar Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vDtfB-0000000CCCR-25rr@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/hwif.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/hwif.c b/drivers/net/ethernet/stmicro/stmmac/hwif.c index 892cef79c4d1..e1f99b9d9d7f 100644 --- a/drivers/net/ethernet/stmicro/stmmac/hwif.c +++ b/drivers/net/ethernet/stmicro/stmmac/hwif.c @@ -314,7 +314,7 @@ stmmac_hwif_find(enum dwmac_core_type core_type, u8 snpsver, u8 dev_id) if (snpsver < entry->min_id) continue; if (core_type == DWMAC_CORE_XGMAC && - (dev_id ^ entry->dev_id)) + dev_id != entry->dev_id) continue; return entry; From 6436f408eb214bce9c5c308d51372a1611940dff Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Wed, 29 Oct 2025 00:03:46 +0000 Subject: [PATCH 0363/1381] net: stmmac: reorganise stmmac_hwif_init() Reorganise stmmac_hwif_init() to handle the error case of stmmac_hwif_find() in the indented block, which follows normal programming pattern. Reviewed-by: Andrew Lunn Tested-by: Maxime Chevallier Tested-by: Mohd Ayaan Anwar Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vDtfG-0000000CCCX-2YwQ@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/hwif.c | 72 ++++++++++++---------- 1 file changed, 38 insertions(+), 34 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/hwif.c b/drivers/net/ethernet/stmicro/stmmac/hwif.c index e1f99b9d9d7f..8212441f9826 100644 --- a/drivers/net/ethernet/stmicro/stmmac/hwif.c +++ b/drivers/net/ethernet/stmicro/stmmac/hwif.c @@ -366,41 +366,45 @@ int stmmac_hwif_init(struct stmmac_priv *priv) /* Use synopsys_id var because some setups can override this */ entry = stmmac_hwif_find(core_type, priv->synopsys_id, version.dev_id); - if (entry) { - /* Only use generic HW helpers if needed */ - mac->desc = mac->desc ? : entry->desc; - mac->dma = mac->dma ? : entry->dma; - mac->mac = mac->mac ? : entry->mac; - mac->ptp = mac->ptp ? : entry->hwtimestamp; - mac->mode = mac->mode ? : entry->mode; - mac->tc = mac->tc ? : entry->tc; - mac->mmc = mac->mmc ? : entry->mmc; - mac->est = mac->est ? : entry->est; - mac->vlan = mac->vlan ? : entry->vlan; + if (!entry) { + dev_err(priv->device, + "Failed to find HW IF (id=0x%x, gmac=%d/%d)\n", + version.snpsver, core_type == DWMAC_CORE_GMAC, + core_type == DWMAC_CORE_GMAC4); - priv->hw = mac; - priv->fpe_cfg.reg = entry->regs.fpe_reg; - priv->ptpaddr = priv->ioaddr + entry->regs.ptp_off; - priv->mmcaddr = priv->ioaddr + entry->regs.mmc_off; - memcpy(&priv->ptp_clock_ops, entry->ptp, - sizeof(struct ptp_clock_info)); - if (entry->est) - priv->estaddr = priv->ioaddr + entry->regs.est_off; - - /* Entry found */ - if (needs_setup) { - ret = entry->setup(priv); - if (ret) - return ret; - } - - /* Save quirks, if needed for posterior use */ - priv->hwif_quirks = entry->quirks; - return 0; + return -EINVAL; } - dev_err(priv->device, "Failed to find HW IF (id=0x%x, gmac=%d/%d)\n", - version.snpsver, core_type == DWMAC_CORE_GMAC, - core_type == DWMAC_CORE_GMAC4); - return -EINVAL; + /* Only use generic HW helpers if needed */ + mac->desc = mac->desc ? : entry->desc; + mac->dma = mac->dma ? : entry->dma; + mac->mac = mac->mac ? : entry->mac; + mac->ptp = mac->ptp ? : entry->hwtimestamp; + mac->mode = mac->mode ? : entry->mode; + mac->tc = mac->tc ? : entry->tc; + mac->mmc = mac->mmc ? : entry->mmc; + mac->est = mac->est ? : entry->est; + mac->vlan = mac->vlan ? : entry->vlan; + + priv->hw = mac; + priv->fpe_cfg.reg = entry->regs.fpe_reg; + priv->ptpaddr = priv->ioaddr + entry->regs.ptp_off; + priv->mmcaddr = priv->ioaddr + entry->regs.mmc_off; + memcpy(&priv->ptp_clock_ops, entry->ptp, + sizeof(struct ptp_clock_info)); + + if (entry->est) + priv->estaddr = priv->ioaddr + entry->regs.est_off; + + /* Entry found */ + if (needs_setup) { + ret = entry->setup(priv); + if (ret) + return ret; + } + + /* Save quirks, if needed for posterior use */ + priv->hwif_quirks = entry->quirks; + + return 0; } From afb8f6567a5b4bb4e673608048939fef854b8709 Mon Sep 17 00:00:00 2001 From: Ankit Khushwaha Date: Tue, 28 Oct 2025 22:59:47 +0530 Subject: [PATCH 0364/1381] selftest: net: fix socklen_t type mismatch in sctp_collision test Socket APIs like recvfrom(), accept(), and getsockname() expect socklen_t* arg, but tests were using int variables. This causes -Wpointer-sign warnings on platforms where socklen_t is unsigned. Change the variable type from int to socklen_t to resolve the warning and ensure type safety across platforms. warning fixed: sctp_collision.c:62:70: warning: passing 'int *' to parameter of type 'socklen_t *' (aka 'unsigned int *') converts between pointers to integer types with different sign [-Wpointer-sign] 62 | ret = recvfrom(sd, buf, sizeof(buf), 0, (struct sockaddr *)&daddr, &len); | ^~~~ /usr/include/sys/socket.h:165:27: note: passing argument to parameter '__addr_len' here 165 | socklen_t *__restrict __addr_len); | ^ Reviewed-by: Muhammad Usama Anjum Signed-off-by: Ankit Khushwaha Reviewed-by: Simon Horman Link: https://patch.msgid.link/20251028172947.53153-1-ankitkhushwaha.linux@gmail.com Signed-off-by: Jakub Kicinski --- tools/testing/selftests/net/netfilter/sctp_collision.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/tools/testing/selftests/net/netfilter/sctp_collision.c b/tools/testing/selftests/net/netfilter/sctp_collision.c index 21bb1cfd8a85..b282d1785c9b 100644 --- a/tools/testing/selftests/net/netfilter/sctp_collision.c +++ b/tools/testing/selftests/net/netfilter/sctp_collision.c @@ -9,9 +9,10 @@ int main(int argc, char *argv[]) { struct sockaddr_in saddr = {}, daddr = {}; - int sd, ret, len = sizeof(daddr); + socklen_t len = sizeof(daddr); struct timeval tv = {25, 0}; char buf[] = "hello"; + int sd, ret; if (argc != 6 || (strcmp(argv[1], "server") && strcmp(argv[1], "client"))) { printf("%s \n", From b8a7826e4b1aab3fabb29cbf0b73da9993d356de Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Tue, 28 Oct 2025 03:58:23 +0000 Subject: [PATCH 0365/1381] net: sched: Don't use WARN_ON_ONCE() for -ENOMEM in tcf_classify(). As demonstrated by syzbot, WARN_ON_ONCE() in tcf_classify() can be easily triggered by fault injection. [0] We should not use WARN_ON_ONCE() for the simple -ENOMEM case. Also, we provide SKB_DROP_REASON_NOMEM for the same error. Let's remove WARN_ON_ONCE() there. [0]: FAULT_INJECTION: forcing a failure. name failslab, interval 1, probability 0, space 0, times 0 CPU: 0 UID: 0 PID: 31392 Comm: syz.8.7081 Not tainted syzkaller #0 PREEMPT(full) Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 10/02/2025 Call Trace: dump_stack_lvl+0x189/0x250 should_fail_ex+0x414/0x560 should_failslab+0xa8/0x100 kmem_cache_alloc_noprof+0x74/0x6e0 skb_ext_add+0x148/0x8f0 tcf_classify+0xeba/0x1140 multiq_enqueue+0xfd/0x4c0 net/sched/sch_multiq.c:66 ... WARNING: CPU: 0 PID: 31392 at net/sched/cls_api.c:1869 tcf_classify+0xfd7/0x1140 Modules linked in: CPU: 0 UID: 0 PID: 31392 Comm: syz.8.7081 Not tainted syzkaller #0 PREEMPT(full) Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 10/02/2025 RIP: 0010:tcf_classify+0xfd7/0x1140 Code: e8 03 42 0f b6 04 30 84 c0 0f 85 41 01 00 00 66 41 89 1f eb 05 e8 89 26 75 f8 bb ff ff ff ff e9 04 f9 ff ff e8 7a 26 75 f8 90 <0f> 0b 90 49 83 c5 44 4c 89 eb 49 c1 ed 03 43 0f b6 44 35 00 84 c0 RSP: 0018:ffffc9000b7671f0 EFLAGS: 00010293 RAX: ffffffff894addf6 RBX: 0000000000000002 RCX: ffff888025029e40 RDX: 0000000000000000 RSI: ffffffff8bbf05c0 RDI: ffffffff8bbf0580 RBP: 0000000000000000 R08: 00000000ffffffff R09: 1ffffffff1c0bfd6 R10: dffffc0000000000 R11: fffffbfff1c0bfd7 R12: ffff88805a90de5c R13: ffff88805a90ddc0 R14: dffffc0000000000 R15: ffffc9000b7672c0 FS: 00007f20739f66c0(0000) GS:ffff88812613e000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 000000110c2d2a80 CR3: 0000000024e36000 CR4: 00000000003526f0 Call Trace: multiq_classify net/sched/sch_multiq.c:39 [inline] multiq_enqueue+0xfd/0x4c0 net/sched/sch_multiq.c:66 dev_qdisc_enqueue+0x4e/0x260 net/core/dev.c:4118 __dev_xmit_skb net/core/dev.c:4214 [inline] __dev_queue_xmit+0xe83/0x3b50 net/core/dev.c:4729 packet_snd net/packet/af_packet.c:3076 [inline] packet_sendmsg+0x3e33/0x5080 net/packet/af_packet.c:3108 sock_sendmsg_nosec net/socket.c:727 [inline] __sock_sendmsg+0x21c/0x270 net/socket.c:742 ____sys_sendmsg+0x505/0x830 net/socket.c:2630 ___sys_sendmsg+0x21f/0x2a0 net/socket.c:2684 __sys_sendmsg net/socket.c:2716 [inline] __do_sys_sendmsg net/socket.c:2721 [inline] __se_sys_sendmsg net/socket.c:2719 [inline] __x64_sys_sendmsg+0x19b/0x260 net/socket.c:2719 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline] do_syscall_64+0xfa/0xfa0 arch/x86/entry/syscall_64.c:94 entry_SYSCALL_64_after_hwframe+0x77/0x7f RIP: 0033:0x7f207578efc9 Code: ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 40 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 a8 ff ff ff f7 d8 64 89 01 48 RSP: 002b:00007f20739f6038 EFLAGS: 00000246 ORIG_RAX: 000000000000002e RAX: ffffffffffffffda RBX: 00007f20759e5fa0 RCX: 00007f207578efc9 RDX: 0000000000000004 RSI: 00002000000000c0 RDI: 0000000000000008 RBP: 00007f20739f6090 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000001 R13: 00007f20759e6038 R14: 00007f20759e5fa0 R15: 00007f2075b0fa28 Reported-by: syzbot+87e1289a044fcd0c5f62@syzkaller.appspotmail.com Closes: https://lore.kernel.org/netdev/69003e33.050a0220.32483.00e8.GAE@google.com/ Signed-off-by: Kuniyuki Iwashima Reviewed-by: Jamal Hadi Salim Link: https://patch.msgid.link/20251028035859.2067690-1-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/sched/cls_api.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index ecec0a1e1c1a..f751cd5eeac8 100644 --- a/net/sched/cls_api.c +++ b/net/sched/cls_api.c @@ -1866,7 +1866,7 @@ int tcf_classify(struct sk_buff *skb, struct tc_skb_cb *cb = tc_skb_cb(skb); ext = tc_skb_ext_alloc(skb); - if (WARN_ON_ONCE(!ext)) { + if (!ext) { tcf_set_drop_reason(skb, SKB_DROP_REASON_NOMEM); return TC_ACT_SHOT; } From f0e7036fc9cb08bdfb27d64eee7fc003ba0bc2e5 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Mon, 27 Oct 2025 10:22:30 +0200 Subject: [PATCH 0366/1381] ipv4: icmp: Add RFC 5837 support Add the ability to append the incoming IP interface information to ICMPv4 error messages in accordance with RFC 5837 and RFC 4884. This is required for more meaningful traceroute results in unnumbered networks. The feature is disabled by default and controlled via a new sysctl ("net.ipv4.icmp_errors_extension_mask") which accepts a bitmask of ICMP extensions to append to ICMP error messages. Currently, only a single value is supported, but the interface and the implementation should be able to support more extensions, if needed. Clone the skb and copy the relevant data portions before modifying the skb as the caller of __icmp_send() still owns the skb after the function returns. This should be fine since by default ICMP error messages are rate limited to 1000 per second and no more than 1 per second per specific host. Trim or pad the packet to 128 bytes before appending the ICMP extension structure in order to be compatible with legacy applications that assume that the ICMP extension structure always starts at this offset (the minimum length specified by RFC 4884). Reviewed-by: Petr Machata Reviewed-by: David Ahern Reviewed-by: Willem de Bruijn Signed-off-by: Ido Schimmel Reviewed-by: Eric Dumazet Link: https://patch.msgid.link/20251027082232.232571-2-idosch@nvidia.com Signed-off-by: Jakub Kicinski --- Documentation/networking/ip-sysctl.rst | 17 +++ include/linux/icmp.h | 32 +++++ include/net/netns/ipv4.h | 1 + net/ipv4/icmp.c | 191 ++++++++++++++++++++++++- net/ipv4/sysctl_net_ipv4.c | 11 ++ 5 files changed, 251 insertions(+), 1 deletion(-) diff --git a/Documentation/networking/ip-sysctl.rst b/Documentation/networking/ip-sysctl.rst index a06cb99d66dc..ece1187ba0f1 100644 --- a/Documentation/networking/ip-sysctl.rst +++ b/Documentation/networking/ip-sysctl.rst @@ -1796,6 +1796,23 @@ icmp_errors_use_inbound_ifaddr - BOOLEAN Default: 0 (disabled) +icmp_errors_extension_mask - UNSIGNED INTEGER + Bitmask of ICMP extensions to append to ICMPv4 error messages + ("Destination Unreachable", "Time Exceeded" and "Parameter Problem"). + The original datagram is trimmed / padded to 128 bytes in order to be + compatible with applications that do not comply with RFC 4884. + + Possible extensions are: + + ==== ============================================================== + 0x01 Incoming IP interface information according to RFC 5837. + Extension will include the index, IPv4 address (if present), + name and MTU of the IP interface that received the datagram + which elicited the ICMP error. + ==== ============================================================== + + Default: 0x00 (no extensions) + igmp_max_memberships - INTEGER Change the maximum number of multicast groups we can subscribe to. Default: 20 diff --git a/include/linux/icmp.h b/include/linux/icmp.h index 0af4d210ee31..043ec5d9c882 100644 --- a/include/linux/icmp.h +++ b/include/linux/icmp.h @@ -40,4 +40,36 @@ void ip_icmp_error_rfc4884(const struct sk_buff *skb, struct sock_ee_data_rfc4884 *out, int thlen, int off); +/* RFC 4884 */ +#define ICMP_EXT_ORIG_DGRAM_MIN_LEN 128 +#define ICMP_EXT_VERSION_2 2 + +/* ICMP Extension Object Classes */ +#define ICMP_EXT_OBJ_CLASS_IIO 2 /* RFC 5837 */ + +/* Interface Information Object - RFC 5837 */ +enum { + ICMP_EXT_CTYPE_IIO_ROLE_IIF, +}; + +#define ICMP_EXT_CTYPE_IIO_ROLE(ROLE) ((ROLE) << 6) +#define ICMP_EXT_CTYPE_IIO_MTU BIT(0) +#define ICMP_EXT_CTYPE_IIO_NAME BIT(1) +#define ICMP_EXT_CTYPE_IIO_IPADDR BIT(2) +#define ICMP_EXT_CTYPE_IIO_IFINDEX BIT(3) + +struct icmp_ext_iio_name_subobj { + u8 len; + char name[IFNAMSIZ]; +}; + +enum { + /* RFC 5837 - Incoming IP Interface Role */ + ICMP_ERR_EXT_IIO_IIF, + /* Add new constants above. Used by "icmp_errors_extension_mask" + * sysctl. + */ + ICMP_ERR_EXT_COUNT, +}; + #endif /* _LINUX_ICMP_H */ diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index 34eb3aecb3f2..0e96c90e56c6 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -135,6 +135,7 @@ struct netns_ipv4 { u8 sysctl_icmp_echo_ignore_broadcasts; u8 sysctl_icmp_ignore_bogus_error_responses; u8 sysctl_icmp_errors_use_inbound_ifaddr; + u8 sysctl_icmp_errors_extension_mask; int sysctl_icmp_ratelimit; int sysctl_icmp_ratemask; int sysctl_icmp_msgs_per_sec; diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c index 1b7fb5d935ed..4abbec2f47ef 100644 --- a/net/ipv4/icmp.c +++ b/net/ipv4/icmp.c @@ -582,6 +582,185 @@ static struct rtable *icmp_route_lookup(struct net *net, struct flowi4 *fl4, return ERR_PTR(err); } +struct icmp_ext_iio_addr4_subobj { + __be16 afi; + __be16 reserved; + __be32 addr4; +}; + +static unsigned int icmp_ext_iio_len(void) +{ + return sizeof(struct icmp_extobj_hdr) + + /* ifIndex */ + sizeof(__be32) + + /* Interface Address Sub-Object */ + sizeof(struct icmp_ext_iio_addr4_subobj) + + /* Interface Name Sub-Object. Length must be a multiple of 4 + * bytes. + */ + ALIGN(sizeof(struct icmp_ext_iio_name_subobj), 4) + + /* MTU */ + sizeof(__be32); +} + +static unsigned int icmp_ext_max_len(u8 ext_objs) +{ + unsigned int ext_max_len; + + ext_max_len = sizeof(struct icmp_ext_hdr); + + if (ext_objs & BIT(ICMP_ERR_EXT_IIO_IIF)) + ext_max_len += icmp_ext_iio_len(); + + return ext_max_len; +} + +static __be32 icmp_ext_iio_addr4_find(const struct net_device *dev) +{ + struct in_device *in_dev; + struct in_ifaddr *ifa; + + in_dev = __in_dev_get_rcu(dev); + if (!in_dev) + return 0; + + /* It is unclear from RFC 5837 which IP address should be chosen, but + * it makes sense to choose a global unicast address. + */ + in_dev_for_each_ifa_rcu(ifa, in_dev) { + if (READ_ONCE(ifa->ifa_flags) & IFA_F_SECONDARY) + continue; + if (ifa->ifa_scope != RT_SCOPE_UNIVERSE || + ipv4_is_multicast(ifa->ifa_address)) + continue; + return ifa->ifa_address; + } + + return 0; +} + +static void icmp_ext_iio_iif_append(struct net *net, struct sk_buff *skb, + int iif) +{ + struct icmp_ext_iio_name_subobj *name_subobj; + struct icmp_extobj_hdr *objh; + struct net_device *dev; + __be32 data; + + if (!iif) + return; + + /* Add the fields in the order specified by RFC 5837. */ + objh = skb_put(skb, sizeof(*objh)); + objh->class_num = ICMP_EXT_OBJ_CLASS_IIO; + objh->class_type = ICMP_EXT_CTYPE_IIO_ROLE(ICMP_EXT_CTYPE_IIO_ROLE_IIF); + + data = htonl(iif); + skb_put_data(skb, &data, sizeof(__be32)); + objh->class_type |= ICMP_EXT_CTYPE_IIO_IFINDEX; + + rcu_read_lock(); + + dev = dev_get_by_index_rcu(net, iif); + if (!dev) + goto out; + + data = icmp_ext_iio_addr4_find(dev); + if (data) { + struct icmp_ext_iio_addr4_subobj *addr4_subobj; + + addr4_subobj = skb_put_zero(skb, sizeof(*addr4_subobj)); + addr4_subobj->afi = htons(ICMP_AFI_IP); + addr4_subobj->addr4 = data; + objh->class_type |= ICMP_EXT_CTYPE_IIO_IPADDR; + } + + name_subobj = skb_put_zero(skb, ALIGN(sizeof(*name_subobj), 4)); + name_subobj->len = ALIGN(sizeof(*name_subobj), 4); + netdev_copy_name(dev, name_subobj->name); + objh->class_type |= ICMP_EXT_CTYPE_IIO_NAME; + + data = htonl(READ_ONCE(dev->mtu)); + skb_put_data(skb, &data, sizeof(__be32)); + objh->class_type |= ICMP_EXT_CTYPE_IIO_MTU; + +out: + rcu_read_unlock(); + objh->length = htons(skb_tail_pointer(skb) - (unsigned char *)objh); +} + +static void icmp_ext_objs_append(struct net *net, struct sk_buff *skb, + u8 ext_objs, int iif) +{ + if (ext_objs & BIT(ICMP_ERR_EXT_IIO_IIF)) + icmp_ext_iio_iif_append(net, skb, iif); +} + +static struct sk_buff * +icmp_ext_append(struct net *net, struct sk_buff *skb_in, struct icmphdr *icmph, + unsigned int room, int iif) +{ + unsigned int payload_len, ext_max_len, ext_len; + struct icmp_ext_hdr *ext_hdr; + struct sk_buff *skb; + u8 ext_objs; + int nhoff; + + switch (icmph->type) { + case ICMP_DEST_UNREACH: + case ICMP_TIME_EXCEEDED: + case ICMP_PARAMETERPROB: + break; + default: + return NULL; + } + + ext_objs = READ_ONCE(net->ipv4.sysctl_icmp_errors_extension_mask); + if (!ext_objs) + return NULL; + + ext_max_len = icmp_ext_max_len(ext_objs); + if (ICMP_EXT_ORIG_DGRAM_MIN_LEN + ext_max_len > room) + return NULL; + + skb = skb_clone(skb_in, GFP_ATOMIC); + if (!skb) + return NULL; + + nhoff = skb_network_offset(skb); + payload_len = min(skb->len - nhoff, ICMP_EXT_ORIG_DGRAM_MIN_LEN); + + if (!pskb_network_may_pull(skb, payload_len)) + goto free_skb; + + if (pskb_trim(skb, nhoff + ICMP_EXT_ORIG_DGRAM_MIN_LEN) || + __skb_put_padto(skb, nhoff + ICMP_EXT_ORIG_DGRAM_MIN_LEN, false)) + goto free_skb; + + if (pskb_expand_head(skb, 0, ext_max_len, GFP_ATOMIC)) + goto free_skb; + + ext_hdr = skb_put_zero(skb, sizeof(*ext_hdr)); + ext_hdr->version = ICMP_EXT_VERSION_2; + + icmp_ext_objs_append(net, skb, ext_objs, iif); + + /* Do not send an empty extension structure. */ + ext_len = skb_tail_pointer(skb) - (unsigned char *)ext_hdr; + if (ext_len == sizeof(*ext_hdr)) + goto free_skb; + + ext_hdr->checksum = ip_compute_csum(ext_hdr, ext_len); + /* The length of the original datagram in 32-bit words (RFC 4884). */ + icmph->un.reserved[1] = ICMP_EXT_ORIG_DGRAM_MIN_LEN / sizeof(u32); + + return skb; + +free_skb: + consume_skb(skb); + return NULL; +} + /* * Send an ICMP message in response to a situation * @@ -601,6 +780,7 @@ void __icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info, struct icmp_bxm icmp_param; struct rtable *rt = skb_rtable(skb_in); bool apply_ratelimit = false; + struct sk_buff *ext_skb; struct ipcm_cookie ipc; struct flowi4 fl4; __be32 saddr; @@ -770,7 +950,12 @@ void __icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info, if (room <= (int)sizeof(struct iphdr)) goto ende; - icmp_param.data_len = skb_in->len - icmp_param.offset; + ext_skb = icmp_ext_append(net, skb_in, &icmp_param.data.icmph, room, + parm->iif); + if (ext_skb) + icmp_param.skb = ext_skb; + + icmp_param.data_len = icmp_param.skb->len - icmp_param.offset; if (icmp_param.data_len > room) icmp_param.data_len = room; icmp_param.head_len = sizeof(struct icmphdr); @@ -785,6 +970,9 @@ void __icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info, trace_icmp_send(skb_in, type, code); icmp_push_reply(sk, &icmp_param, &fl4, &ipc, &rt); + + if (ext_skb) + consume_skb(ext_skb); ende: ip_rt_put(rt); out_unlock: @@ -1502,6 +1690,7 @@ static int __net_init icmp_sk_init(struct net *net) net->ipv4.sysctl_icmp_ratelimit = 1 * HZ; net->ipv4.sysctl_icmp_ratemask = 0x1818; net->ipv4.sysctl_icmp_errors_use_inbound_ifaddr = 0; + net->ipv4.sysctl_icmp_errors_extension_mask = 0; net->ipv4.sysctl_icmp_msgs_per_sec = 1000; net->ipv4.sysctl_icmp_msgs_burst = 50; diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 24dbc603cc44..0c7c8f9041cb 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -48,6 +48,8 @@ static int tcp_plb_max_rounds = 31; static int tcp_plb_max_cong_thresh = 256; static unsigned int tcp_tw_reuse_delay_max = TCP_PAWS_MSL * MSEC_PER_SEC; static int tcp_ecn_mode_max = 2; +static u32 icmp_errors_extension_mask_all = + GENMASK_U8(ICMP_ERR_EXT_COUNT - 1, 0); /* obsolete */ static int sysctl_tcp_low_latency __read_mostly; @@ -674,6 +676,15 @@ static struct ctl_table ipv4_net_table[] = { .extra1 = SYSCTL_ZERO, .extra2 = SYSCTL_ONE }, + { + .procname = "icmp_errors_extension_mask", + .data = &init_net.ipv4.sysctl_icmp_errors_extension_mask, + .maxlen = sizeof(u8), + .mode = 0644, + .proc_handler = proc_dou8vec_minmax, + .extra1 = SYSCTL_ZERO, + .extra2 = &icmp_errors_extension_mask_all, + }, { .procname = "icmp_ratelimit", .data = &init_net.ipv4.sysctl_icmp_ratelimit, From d12d04d221f8d928a27a66236228e7501cd4cad5 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Mon, 27 Oct 2025 10:22:31 +0200 Subject: [PATCH 0367/1381] ipv6: icmp: Add RFC 5837 support Add the ability to append the incoming IP interface information to ICMPv6 error messages in accordance with RFC 5837 and RFC 4884. This is required for more meaningful traceroute results in unnumbered networks. The feature is disabled by default and controlled via a new sysctl ("net.ipv6.icmp.errors_extension_mask") which accepts a bitmask of ICMP extensions to append to ICMP error messages. Currently, only a single value is supported, but the interface and the implementation should be able to support more extensions, if needed. Clone the skb and copy the relevant data portions before modifying the skb as the caller of icmp6_send() still owns the skb after the function returns. This should be fine since by default ICMP error messages are rate limited to 1000 per second and no more than 1 per second per specific host. Trim or pad the packet to 128 bytes before appending the ICMP extension structure in order to be compatible with legacy applications that assume that the ICMP extension structure always starts at this offset (the minimum length specified by RFC 4884). Since commit 20e1954fe238 ("ipv6: RFC 4884 partial support for SIT/GRE tunnels") it is possible for icmp6_send() to be called with an skb that already contains ICMP extensions. This can happen when we receive an ICMPv4 message with extensions from a tunnel and translate it to an ICMPv6 message towards an IPv6 host in the overlay network. I could not find an RFC that supports this behavior, but it makes sense to not overwrite the original extensions that were appended to the packet. Therefore, avoid appending extensions if the length field in the provided ICMPv6 header is already filled. Export netdev_copy_name() using EXPORT_IPV6_MOD_GPL() to make it available to IPv6 when it is built as a module. Reviewed-by: Petr Machata Reviewed-by: David Ahern Reviewed-by: Willem de Bruijn Signed-off-by: Ido Schimmel Reviewed-by: Eric Dumazet Link: https://patch.msgid.link/20251027082232.232571-3-idosch@nvidia.com Signed-off-by: Jakub Kicinski --- Documentation/networking/ip-sysctl.rst | 17 ++ include/net/netns/ipv6.h | 1 + net/core/dev.c | 1 + net/ipv6/af_inet6.c | 1 + net/ipv6/icmp.c | 214 ++++++++++++++++++++++++- 5 files changed, 232 insertions(+), 2 deletions(-) diff --git a/Documentation/networking/ip-sysctl.rst b/Documentation/networking/ip-sysctl.rst index ece1187ba0f1..7cd35bfd39e6 100644 --- a/Documentation/networking/ip-sysctl.rst +++ b/Documentation/networking/ip-sysctl.rst @@ -3279,6 +3279,23 @@ error_anycast_as_unicast - BOOLEAN Default: 0 (disabled) +errors_extension_mask - UNSIGNED INTEGER + Bitmask of ICMP extensions to append to ICMPv6 error messages + ("Destination Unreachable" and "Time Exceeded"). The original datagram + is trimmed / padded to 128 bytes in order to be compatible with + applications that do not comply with RFC 4884. + + Possible extensions are: + + ==== ============================================================== + 0x01 Incoming IP interface information according to RFC 5837. + Extension will include the index, IPv6 address (if present), + name and MTU of the IP interface that received the datagram + which elicited the ICMP error. + ==== ============================================================== + + Default: 0x00 (no extensions) + xfrm6_gc_thresh - INTEGER (Obsolete since linux-4.14) The threshold at which we will start garbage collecting for IPv6 diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h index 47dc70d8100a..08d2ecc96e2b 100644 --- a/include/net/netns/ipv6.h +++ b/include/net/netns/ipv6.h @@ -56,6 +56,7 @@ struct netns_sysctl_ipv6 { u8 skip_notify_on_dev_down; u8 fib_notify_on_flag_change; u8 icmpv6_error_anycast_as_unicast; + u8 icmpv6_errors_extension_mask; }; struct netns_ipv6 { diff --git a/net/core/dev.c b/net/core/dev.c index d32f0b0c03bb..dccc1176f3c6 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1163,6 +1163,7 @@ void netdev_copy_name(struct net_device *dev, char *name) strscpy(name, dev->name, IFNAMSIZ); } while (read_seqretry(&netdev_rename_lock, seq)); } +EXPORT_IPV6_MOD_GPL(netdev_copy_name); /** * netdev_get_name - get a netdevice name, knowing its ifindex. diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index 1b0314644e0c..44d7de1eec4f 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -960,6 +960,7 @@ static int __net_init inet6_net_init(struct net *net) net->ipv6.sysctl.icmpv6_echo_ignore_multicast = 0; net->ipv6.sysctl.icmpv6_echo_ignore_anycast = 0; net->ipv6.sysctl.icmpv6_error_anycast_as_unicast = 0; + net->ipv6.sysctl.icmpv6_errors_extension_mask = 0; /* By default, rate limit error messages. * Except for pmtu discovery, it would break it. diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index 56c974cf75d1..5d2f90babaa5 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -444,6 +444,193 @@ static int icmp6_iif(const struct sk_buff *skb) return icmp6_dev(skb)->ifindex; } +struct icmp6_ext_iio_addr6_subobj { + __be16 afi; + __be16 reserved; + struct in6_addr addr6; +}; + +static unsigned int icmp6_ext_iio_len(void) +{ + return sizeof(struct icmp_extobj_hdr) + + /* ifIndex */ + sizeof(__be32) + + /* Interface Address Sub-Object */ + sizeof(struct icmp6_ext_iio_addr6_subobj) + + /* Interface Name Sub-Object. Length must be a multiple of 4 + * bytes. + */ + ALIGN(sizeof(struct icmp_ext_iio_name_subobj), 4) + + /* MTU */ + sizeof(__be32); +} + +static unsigned int icmp6_ext_max_len(u8 ext_objs) +{ + unsigned int ext_max_len; + + ext_max_len = sizeof(struct icmp_ext_hdr); + + if (ext_objs & BIT(ICMP_ERR_EXT_IIO_IIF)) + ext_max_len += icmp6_ext_iio_len(); + + return ext_max_len; +} + +static struct in6_addr *icmp6_ext_iio_addr6_find(const struct net_device *dev) +{ + struct inet6_dev *in6_dev; + struct inet6_ifaddr *ifa; + + in6_dev = __in6_dev_get(dev); + if (!in6_dev) + return NULL; + + /* It is unclear from RFC 5837 which IP address should be chosen, but + * it makes sense to choose a global unicast address. + */ + list_for_each_entry_rcu(ifa, &in6_dev->addr_list, if_list) { + if (ifa->flags & (IFA_F_TENTATIVE | IFA_F_DADFAILED)) + continue; + if (ipv6_addr_type(&ifa->addr) != IPV6_ADDR_UNICAST || + ipv6_addr_src_scope(&ifa->addr) != IPV6_ADDR_SCOPE_GLOBAL) + continue; + return &ifa->addr; + } + + return NULL; +} + +static void icmp6_ext_iio_iif_append(struct net *net, struct sk_buff *skb, + int iif) +{ + struct icmp_ext_iio_name_subobj *name_subobj; + struct icmp_extobj_hdr *objh; + struct net_device *dev; + struct in6_addr *addr6; + __be32 data; + + if (!iif) + return; + + /* Add the fields in the order specified by RFC 5837. */ + objh = skb_put(skb, sizeof(*objh)); + objh->class_num = ICMP_EXT_OBJ_CLASS_IIO; + objh->class_type = ICMP_EXT_CTYPE_IIO_ROLE(ICMP_EXT_CTYPE_IIO_ROLE_IIF); + + data = htonl(iif); + skb_put_data(skb, &data, sizeof(__be32)); + objh->class_type |= ICMP_EXT_CTYPE_IIO_IFINDEX; + + rcu_read_lock(); + + dev = dev_get_by_index_rcu(net, iif); + if (!dev) + goto out; + + addr6 = icmp6_ext_iio_addr6_find(dev); + if (addr6) { + struct icmp6_ext_iio_addr6_subobj *addr6_subobj; + + addr6_subobj = skb_put_zero(skb, sizeof(*addr6_subobj)); + addr6_subobj->afi = htons(ICMP_AFI_IP6); + addr6_subobj->addr6 = *addr6; + objh->class_type |= ICMP_EXT_CTYPE_IIO_IPADDR; + } + + name_subobj = skb_put_zero(skb, ALIGN(sizeof(*name_subobj), 4)); + name_subobj->len = ALIGN(sizeof(*name_subobj), 4); + netdev_copy_name(dev, name_subobj->name); + objh->class_type |= ICMP_EXT_CTYPE_IIO_NAME; + + data = htonl(READ_ONCE(dev->mtu)); + skb_put_data(skb, &data, sizeof(__be32)); + objh->class_type |= ICMP_EXT_CTYPE_IIO_MTU; + +out: + rcu_read_unlock(); + objh->length = htons(skb_tail_pointer(skb) - (unsigned char *)objh); +} + +static void icmp6_ext_objs_append(struct net *net, struct sk_buff *skb, + u8 ext_objs, int iif) +{ + if (ext_objs & BIT(ICMP_ERR_EXT_IIO_IIF)) + icmp6_ext_iio_iif_append(net, skb, iif); +} + +static struct sk_buff * +icmp6_ext_append(struct net *net, struct sk_buff *skb_in, + struct icmp6hdr *icmp6h, unsigned int room, int iif) +{ + unsigned int payload_len, ext_max_len, ext_len; + struct icmp_ext_hdr *ext_hdr; + struct sk_buff *skb; + u8 ext_objs; + int nhoff; + + switch (icmp6h->icmp6_type) { + case ICMPV6_DEST_UNREACH: + case ICMPV6_TIME_EXCEED: + break; + default: + return NULL; + } + + /* Do not overwrite existing extensions. This can happen when we + * receive an ICMPv4 message with extensions from a tunnel and + * translate it to an ICMPv6 message towards an IPv6 host in the + * overlay network. + */ + if (icmp6h->icmp6_datagram_len) + return NULL; + + ext_objs = READ_ONCE(net->ipv6.sysctl.icmpv6_errors_extension_mask); + if (!ext_objs) + return NULL; + + ext_max_len = icmp6_ext_max_len(ext_objs); + if (ICMP_EXT_ORIG_DGRAM_MIN_LEN + ext_max_len > room) + return NULL; + + skb = skb_clone(skb_in, GFP_ATOMIC); + if (!skb) + return NULL; + + nhoff = skb_network_offset(skb); + payload_len = min(skb->len - nhoff, ICMP_EXT_ORIG_DGRAM_MIN_LEN); + + if (!pskb_network_may_pull(skb, payload_len)) + goto free_skb; + + if (pskb_trim(skb, nhoff + ICMP_EXT_ORIG_DGRAM_MIN_LEN) || + __skb_put_padto(skb, nhoff + ICMP_EXT_ORIG_DGRAM_MIN_LEN, false)) + goto free_skb; + + if (pskb_expand_head(skb, 0, ext_max_len, GFP_ATOMIC)) + goto free_skb; + + ext_hdr = skb_put_zero(skb, sizeof(*ext_hdr)); + ext_hdr->version = ICMP_EXT_VERSION_2; + + icmp6_ext_objs_append(net, skb, ext_objs, iif); + + /* Do not send an empty extension structure. */ + ext_len = skb_tail_pointer(skb) - (unsigned char *)ext_hdr; + if (ext_len == sizeof(*ext_hdr)) + goto free_skb; + + ext_hdr->checksum = ip_compute_csum(ext_hdr, ext_len); + /* The length of the original datagram in 64-bit words (RFC 4884). */ + icmp6h->icmp6_datagram_len = ICMP_EXT_ORIG_DGRAM_MIN_LEN / sizeof(u64); + + return skb; + +free_skb: + consume_skb(skb); + return NULL; +} + /* * Send an ICMP message in response to a packet in error */ @@ -458,7 +645,9 @@ void icmp6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info, struct ipv6_pinfo *np; const struct in6_addr *saddr = NULL; bool apply_ratelimit = false; + struct sk_buff *ext_skb; struct dst_entry *dst; + unsigned int room; struct icmp6hdr tmp_hdr; struct flowi6 fl6; struct icmpv6_msg msg; @@ -612,8 +801,13 @@ void icmp6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info, msg.offset = skb_network_offset(skb); msg.type = type; - len = skb->len - msg.offset; - len = min_t(unsigned int, len, IPV6_MIN_MTU - sizeof(struct ipv6hdr) - sizeof(struct icmp6hdr)); + room = IPV6_MIN_MTU - sizeof(struct ipv6hdr) - sizeof(struct icmp6hdr); + ext_skb = icmp6_ext_append(net, skb, &tmp_hdr, room, parm->iif); + if (ext_skb) + msg.skb = ext_skb; + + len = msg.skb->len - msg.offset; + len = min_t(unsigned int, len, room); if (len < 0) { net_dbg_ratelimited("icmp: len problem [%pI6c > %pI6c]\n", &hdr->saddr, &hdr->daddr); @@ -635,6 +829,8 @@ void icmp6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info, } out_dst_release: + if (ext_skb) + consume_skb(ext_skb); dst_release(dst); out_unlock: icmpv6_xmit_unlock(sk); @@ -1171,6 +1367,10 @@ int icmpv6_err_convert(u8 type, u8 code, int *err) EXPORT_SYMBOL(icmpv6_err_convert); #ifdef CONFIG_SYSCTL + +static u32 icmpv6_errors_extension_mask_all = + GENMASK_U8(ICMP_ERR_EXT_COUNT - 1, 0); + static struct ctl_table ipv6_icmp_table_template[] = { { .procname = "ratelimit", @@ -1216,6 +1416,15 @@ static struct ctl_table ipv6_icmp_table_template[] = { .extra1 = SYSCTL_ZERO, .extra2 = SYSCTL_ONE, }, + { + .procname = "errors_extension_mask", + .data = &init_net.ipv6.sysctl.icmpv6_errors_extension_mask, + .maxlen = sizeof(u8), + .mode = 0644, + .proc_handler = proc_dou8vec_minmax, + .extra1 = SYSCTL_ZERO, + .extra2 = &icmpv6_errors_extension_mask_all, + }, }; struct ctl_table * __net_init ipv6_icmp_sysctl_init(struct net *net) @@ -1233,6 +1442,7 @@ struct ctl_table * __net_init ipv6_icmp_sysctl_init(struct net *net) table[3].data = &net->ipv6.sysctl.icmpv6_echo_ignore_anycast; table[4].data = &net->ipv6.sysctl.icmpv6_ratemask_ptr; table[5].data = &net->ipv6.sysctl.icmpv6_error_anycast_as_unicast; + table[6].data = &net->ipv6.sysctl.icmpv6_errors_extension_mask; } return table; } From 02da595751833a272ce0e30438a544a77eb7c103 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Mon, 27 Oct 2025 10:22:32 +0200 Subject: [PATCH 0368/1381] selftests: traceroute: Add ICMP extensions tests Test that ICMP extensions are reported correctly when enabled and not reported when disabled. Test both IPv4 and IPv6 and using different packet sizes, to make sure trimming / padding works correctly. Disable ICMP rate limiting (defaults to 1 per-second per-target) so that the kernel will always generate ICMP errors when needed. Signed-off-by: Ido Schimmel Link: https://patch.msgid.link/20251027082232.232571-4-idosch@nvidia.com Signed-off-by: Jakub Kicinski --- tools/testing/selftests/net/traceroute.sh | 313 ++++++++++++++++++++++ 1 file changed, 313 insertions(+) diff --git a/tools/testing/selftests/net/traceroute.sh b/tools/testing/selftests/net/traceroute.sh index dbb34c7e09ce..a7c6ab8a0347 100755 --- a/tools/testing/selftests/net/traceroute.sh +++ b/tools/testing/selftests/net/traceroute.sh @@ -36,6 +36,35 @@ run_cmd() return $rc } +__check_traceroute_version() +{ + local cmd=$1; shift + local req_ver=$1; shift + local ver + + req_ver=$(echo "$req_ver" | sed 's/\.//g') + ver=$($cmd -V 2>&1 | grep -Eo '[0-9]+.[0-9]+.[0-9]+' | sed 's/\.//g') + if [[ $ver -lt $req_ver ]]; then + return 1 + else + return 0 + fi +} + +check_traceroute6_version() +{ + local req_ver=$1; shift + + __check_traceroute_version traceroute6 "$req_ver" +} + +check_traceroute_version() +{ + local req_ver=$1; shift + + __check_traceroute_version traceroute "$req_ver" +} + ################################################################################ # create namespaces and interconnects @@ -59,6 +88,8 @@ create_ns() ip netns exec ${ns} ip -6 ro add unreachable default metric 8192 ip netns exec ${ns} sysctl -qw net.ipv4.ip_forward=1 + ip netns exec ${ns} sysctl -qw net.ipv4.icmp_ratelimit=0 + ip netns exec ${ns} sysctl -qw net.ipv6.icmp.ratelimit=0 ip netns exec ${ns} sysctl -qw net.ipv6.conf.all.keep_addr_on_down=1 ip netns exec ${ns} sysctl -qw net.ipv6.conf.all.forwarding=1 ip netns exec ${ns} sysctl -qw net.ipv6.conf.default.forwarding=1 @@ -297,6 +328,144 @@ run_traceroute6_vrf() cleanup_traceroute6_vrf } +################################################################################ +# traceroute6 with ICMP extensions test +# +# Verify that in this scenario +# +# ---- ---- ---- +# |H1|--------------------------|R1|--------------------------|H2| +# ---- N1 ---- N2 ---- +# +# ICMP extensions are correctly reported. The loopback interfaces on all the +# nodes are assigned global addresses and the interfaces connecting the nodes +# are assigned IPv6 link-local addresses. + +cleanup_traceroute6_ext() +{ + cleanup_all_ns +} + +setup_traceroute6_ext() +{ + # Start clean + cleanup_traceroute6_ext + + setup_ns h1 r1 h2 + create_ns "$h1" + create_ns "$r1" + create_ns "$h2" + + # Setup N1 + connect_ns "$h1" eth1 - fe80::1/64 "$r1" eth1 - fe80::2/64 + # Setup N2 + connect_ns "$r1" eth2 - fe80::3/64 "$h2" eth2 - fe80::4/64 + + # Setup H1 + ip -n "$h1" address add 2001:db8:1::1/128 dev lo + ip -n "$h1" route add ::/0 nexthop via fe80::2 dev eth1 + + # Setup R1 + ip -n "$r1" address add 2001:db8:1::2/128 dev lo + ip -n "$r1" route add 2001:db8:1::1/128 nexthop via fe80::1 dev eth1 + ip -n "$r1" route add 2001:db8:1::3/128 nexthop via fe80::4 dev eth2 + + # Setup H2 + ip -n "$h2" address add 2001:db8:1::3/128 dev lo + ip -n "$h2" route add ::/0 nexthop via fe80::3 dev eth2 + + # Prime the network + ip netns exec "$h1" ping6 -c5 2001:db8:1::3 >/dev/null 2>&1 +} + +traceroute6_ext_iio_iif_test() +{ + local r1_ifindex h2_ifindex + local pkt_len=$1; shift + + # Test that incoming interface info is not appended by default. + run_cmd "$h1" "traceroute6 -e 2001:db8:1::3 $pkt_len | grep INC" + check_fail $? "Incoming interface info appended by default when should not" + + # Test that the extension is appended when enabled. + run_cmd "$r1" "bash -c \"echo 0x01 > /proc/sys/net/ipv6/icmp/errors_extension_mask\"" + check_err $? "Failed to enable incoming interface info extension on R1" + + run_cmd "$h1" "traceroute6 -e 2001:db8:1::3 $pkt_len | grep INC" + check_err $? "Incoming interface info not appended after enable" + + # Test that the extension is not appended when disabled. + run_cmd "$r1" "bash -c \"echo 0x00 > /proc/sys/net/ipv6/icmp/errors_extension_mask\"" + check_err $? "Failed to disable incoming interface info extension on R1" + + run_cmd "$h1" "traceroute6 -e 2001:db8:1::3 $pkt_len | grep INC" + check_fail $? "Incoming interface info appended after disable" + + # Test that the extension is sent correctly from both R1 and H2. + run_cmd "$r1" "sysctl -w net.ipv6.icmp.errors_extension_mask=0x01" + r1_ifindex=$(ip -n "$r1" -j link show dev eth1 | jq '.[]["ifindex"]') + run_cmd "$h1" "traceroute6 -e 2001:db8:1::3 $pkt_len | grep ''" + check_err $? "Wrong incoming interface info reported from R1" + + run_cmd "$h2" "sysctl -w net.ipv6.icmp.errors_extension_mask=0x01" + h2_ifindex=$(ip -n "$h2" -j link show dev eth2 | jq '.[]["ifindex"]') + run_cmd "$h1" "traceroute6 -e 2001:db8:1::3 $pkt_len | grep ''" + check_err $? "Wrong incoming interface info reported from H2" + + # Add a global address on the incoming interface of R1 and check that + # it is reported. + run_cmd "$r1" "ip address add 2001:db8:100::1/64 dev eth1 nodad" + run_cmd "$h1" "traceroute6 -e 2001:db8:1::3 $pkt_len | grep ''" + check_err $? "Wrong incoming interface info reported from R1 after address addition" + run_cmd "$r1" "ip address del 2001:db8:100::1/64 dev eth1" + + # Change name and MTU and make sure the result is still correct. + run_cmd "$r1" "ip link set dev eth1 name eth1tag mtu 1501" + run_cmd "$h1" "traceroute6 -e 2001:db8:1::3 $pkt_len | grep ''" + check_err $? "Wrong incoming interface info reported from R1 after name and MTU change" + run_cmd "$r1" "ip link set dev eth1tag name eth1 mtu 1500" + + run_cmd "$r1" "sysctl -w net.ipv6.icmp.errors_extension_mask=0x00" + run_cmd "$h2" "sysctl -w net.ipv6.icmp.errors_extension_mask=0x00" +} + +run_traceroute6_ext() +{ + # Need at least version 2.1.5 for RFC 5837 support. + if ! check_traceroute6_version 2.1.5; then + log_test_skip "traceroute6 too old, missing ICMP extensions support" + return + fi + + setup_traceroute6_ext + + RET=0 + + ## General ICMP extensions tests + + # Test that ICMP extensions are disabled by default. + run_cmd "$h1" "sysctl net.ipv6.icmp.errors_extension_mask | grep \"= 0$\"" + check_err $? "ICMP extensions are not disabled by default" + + # Test that unsupported values are rejected. Do not use "sysctl" as + # older versions do not return an error code upon failure. + run_cmd "$h1" "bash -c \"echo 0x80 > /proc/sys/net/ipv6/icmp/errors_extension_mask\"" + check_fail $? "Unsupported sysctl value was not rejected" + + ## Extension-specific tests + + # Incoming interface info test. Test with various packet sizes, + # including the default one. + traceroute6_ext_iio_iif_test + traceroute6_ext_iio_iif_test 127 + traceroute6_ext_iio_iif_test 128 + traceroute6_ext_iio_iif_test 129 + + log_test "IPv6 traceroute with ICMP extensions" + + cleanup_traceroute6_ext +} + ################################################################################ # traceroute test # @@ -437,6 +606,147 @@ run_traceroute_vrf() cleanup_traceroute_vrf } +################################################################################ +# traceroute with ICMP extensions test +# +# Verify that in this scenario +# +# ---- ---- ---- +# |H1|--------------------------|R1|--------------------------|H2| +# ---- N1 ---- N2 ---- +# +# ICMP extensions are correctly reported. The loopback interfaces on all the +# nodes are assigned global addresses and the interfaces connecting the nodes +# are assigned IPv6 link-local addresses. + +cleanup_traceroute_ext() +{ + cleanup_all_ns +} + +setup_traceroute_ext() +{ + # Start clean + cleanup_traceroute_ext + + setup_ns h1 r1 h2 + create_ns "$h1" + create_ns "$r1" + create_ns "$h2" + + # Setup N1 + connect_ns "$h1" eth1 - fe80::1/64 "$r1" eth1 - fe80::2/64 + # Setup N2 + connect_ns "$r1" eth2 - fe80::3/64 "$h2" eth2 - fe80::4/64 + + # Setup H1 + ip -n "$h1" address add 192.0.2.1/32 dev lo + ip -n "$h1" route add 0.0.0.0/0 nexthop via inet6 fe80::2 dev eth1 + + # Setup R1 + ip -n "$r1" address add 192.0.2.2/32 dev lo + ip -n "$r1" route add 192.0.2.1/32 nexthop via inet6 fe80::1 dev eth1 + ip -n "$r1" route add 192.0.2.3/32 nexthop via inet6 fe80::4 dev eth2 + + # Setup H2 + ip -n "$h2" address add 192.0.2.3/32 dev lo + ip -n "$h2" route add 0.0.0.0/0 nexthop via inet6 fe80::3 dev eth2 + + # Prime the network + ip netns exec "$h1" ping -c5 192.0.2.3 >/dev/null 2>&1 +} + +traceroute_ext_iio_iif_test() +{ + local r1_ifindex h2_ifindex + local pkt_len=$1; shift + + # Test that incoming interface info is not appended by default. + run_cmd "$h1" "traceroute -e 192.0.2.3 $pkt_len | grep INC" + check_fail $? "Incoming interface info appended by default when should not" + + # Test that the extension is appended when enabled. + run_cmd "$r1" "bash -c \"echo 0x01 > /proc/sys/net/ipv4/icmp_errors_extension_mask\"" + check_err $? "Failed to enable incoming interface info extension on R1" + + run_cmd "$h1" "traceroute -e 192.0.2.3 $pkt_len | grep INC" + check_err $? "Incoming interface info not appended after enable" + + # Test that the extension is not appended when disabled. + run_cmd "$r1" "bash -c \"echo 0x00 > /proc/sys/net/ipv4/icmp_errors_extension_mask\"" + check_err $? "Failed to disable incoming interface info extension on R1" + + run_cmd "$h1" "traceroute -e 192.0.2.3 $pkt_len | grep INC" + check_fail $? "Incoming interface info appended after disable" + + # Test that the extension is sent correctly from both R1 and H2. + run_cmd "$r1" "sysctl -w net.ipv4.icmp_errors_extension_mask=0x01" + r1_ifindex=$(ip -n "$r1" -j link show dev eth1 | jq '.[]["ifindex"]') + run_cmd "$h1" "traceroute -e 192.0.2.3 $pkt_len | grep ''" + check_err $? "Wrong incoming interface info reported from R1" + + run_cmd "$h2" "sysctl -w net.ipv4.icmp_errors_extension_mask=0x01" + h2_ifindex=$(ip -n "$h2" -j link show dev eth2 | jq '.[]["ifindex"]') + run_cmd "$h1" "traceroute -e 192.0.2.3 $pkt_len | grep ''" + check_err $? "Wrong incoming interface info reported from H2" + + # Add a global address on the incoming interface of R1 and check that + # it is reported. + run_cmd "$r1" "ip address add 198.51.100.1/24 dev eth1" + run_cmd "$h1" "traceroute -e 192.0.2.3 $pkt_len | grep ''" + check_err $? "Wrong incoming interface info reported from R1 after address addition" + run_cmd "$r1" "ip address del 198.51.100.1/24 dev eth1" + + # Change name and MTU and make sure the result is still correct. + # Re-add the route towards H1 since it was deleted when we removed the + # last IPv4 address from eth1 on R1. + run_cmd "$r1" "ip route add 192.0.2.1/32 nexthop via inet6 fe80::1 dev eth1" + run_cmd "$r1" "ip link set dev eth1 name eth1tag mtu 1501" + run_cmd "$h1" "traceroute -e 192.0.2.3 $pkt_len | grep ''" + check_err $? "Wrong incoming interface info reported from R1 after name and MTU change" + run_cmd "$r1" "ip link set dev eth1tag name eth1 mtu 1500" + + run_cmd "$r1" "sysctl -w net.ipv4.icmp_errors_extension_mask=0x00" + run_cmd "$h2" "sysctl -w net.ipv4.icmp_errors_extension_mask=0x00" +} + +run_traceroute_ext() +{ + # Need at least version 2.1.5 for RFC 5837 support. + if ! check_traceroute_version 2.1.5; then + log_test_skip "traceroute too old, missing ICMP extensions support" + return + fi + + setup_traceroute_ext + + RET=0 + + ## General ICMP extensions tests + + # Test that ICMP extensions are disabled by default. + run_cmd "$h1" "sysctl net.ipv4.icmp_errors_extension_mask | grep \"= 0$\"" + check_err $? "ICMP extensions are not disabled by default" + + # Test that unsupported values are rejected. Do not use "sysctl" as + # older versions do not return an error code upon failure. + run_cmd "$h1" "bash -c \"echo 0x80 > /proc/sys/net/ipv4/icmp_errors_extension_mask\"" + check_fail $? "Unsupported sysctl value was not rejected" + + ## Extension-specific tests + + # Incoming interface info test. Test with various packet sizes, + # including the default one. + traceroute_ext_iio_iif_test + traceroute_ext_iio_iif_test 127 + traceroute_ext_iio_iif_test 128 + traceroute_ext_iio_iif_test 129 + + log_test "IPv4 traceroute with ICMP extensions" + + cleanup_traceroute_ext +} + ################################################################################ # Run tests @@ -444,8 +754,10 @@ run_tests() { run_traceroute6 run_traceroute6_vrf + run_traceroute6_ext run_traceroute run_traceroute_vrf + run_traceroute_ext } ################################################################################ @@ -462,6 +774,7 @@ done require_command traceroute6 require_command traceroute +require_command jq run_tests From 3a85ec37bc11e77bda8b1cba34501ef3308200b0 Mon Sep 17 00:00:00 2001 From: Wei Fang Date: Wed, 29 Oct 2025 09:38:55 +0800 Subject: [PATCH 0369/1381] dt-bindings: net: netc-blk-ctrl: add compatible string for i.MX94 platforms Add the compatible string "nxp,imx94-netc-blk-ctrl" for i.MX94 platforms. Signed-off-by: Wei Fang Acked-by: Rob Herring (Arm) Link: https://patch.msgid.link/20251029013900.407583-2-wei.fang@nxp.com Signed-off-by: Jakub Kicinski --- Documentation/devicetree/bindings/net/nxp,netc-blk-ctrl.yaml | 1 + 1 file changed, 1 insertion(+) diff --git a/Documentation/devicetree/bindings/net/nxp,netc-blk-ctrl.yaml b/Documentation/devicetree/bindings/net/nxp,netc-blk-ctrl.yaml index 97389fd5dbbf..deea4fd73d76 100644 --- a/Documentation/devicetree/bindings/net/nxp,netc-blk-ctrl.yaml +++ b/Documentation/devicetree/bindings/net/nxp,netc-blk-ctrl.yaml @@ -21,6 +21,7 @@ maintainers: properties: compatible: enum: + - nxp,imx94-netc-blk-ctrl - nxp,imx95-netc-blk-ctrl reg: From c4430f2ac0475e3525d9b2d0550e4421af54d7e9 Mon Sep 17 00:00:00 2001 From: Wei Fang Date: Wed, 29 Oct 2025 09:38:56 +0800 Subject: [PATCH 0370/1381] dt-bindings: net: enetc: add compatible string for ENETC with pseduo MAC The ENETC with pseudo MAC is used to connect to the CPU port of the NETC switch. This ENETC has a different PCI device ID, so add a standard PCI device compatible string to it. Signed-off-by: Wei Fang Acked-by: Rob Herring (Arm) Link: https://patch.msgid.link/20251029013900.407583-3-wei.fang@nxp.com Signed-off-by: Jakub Kicinski --- Documentation/devicetree/bindings/net/fsl,enetc.yaml | 1 + 1 file changed, 1 insertion(+) diff --git a/Documentation/devicetree/bindings/net/fsl,enetc.yaml b/Documentation/devicetree/bindings/net/fsl,enetc.yaml index ca70f0050171..aac20ab72ace 100644 --- a/Documentation/devicetree/bindings/net/fsl,enetc.yaml +++ b/Documentation/devicetree/bindings/net/fsl,enetc.yaml @@ -27,6 +27,7 @@ properties: - const: fsl,enetc - enum: - pci1131,e101 + - pci1131,e110 reg: maxItems: 1 From ba5d7d45ce8eec93fafd171da110386f97ecfed4 Mon Sep 17 00:00:00 2001 From: Wei Fang Date: Wed, 29 Oct 2025 09:38:57 +0800 Subject: [PATCH 0371/1381] net: enetc: add preliminary i.MX94 NETC blocks control support NETC blocks control is used for warm reset and pre-boot initialization. Different versions of NETC blocks control are not exactly the same. We need to add corresponding netc_devinfo data for each version. i.MX94 series are launched after i.MX95, so its NETC version (v4.3) is higher than i.MX95 NETC (v4.1). Currently, the patch adds the following configurations for ENETCs. 1. Set the link's MII protocol. 2. ENETC 0 (MAC 3) and the switch port 2 (MAC 2) share the same parallel interface, but due to SoC constraint, they cannot be used simultaneously. Since the switch is not supported yet, so the interface is assigned to ENETC 0 by default. The switch configuration will be added separately in a subsequent patch. Signed-off-by: Wei Fang Reviewed-by: Frank Li Link: https://patch.msgid.link/20251029013900.407583-4-wei.fang@nxp.com Signed-off-by: Jakub Kicinski --- .../ethernet/freescale/enetc/netc_blk_ctrl.c | 104 ++++++++++++++++++ 1 file changed, 104 insertions(+) diff --git a/drivers/net/ethernet/freescale/enetc/netc_blk_ctrl.c b/drivers/net/ethernet/freescale/enetc/netc_blk_ctrl.c index bcb8eefeb93c..5978ea096e80 100644 --- a/drivers/net/ethernet/freescale/enetc/netc_blk_ctrl.c +++ b/drivers/net/ethernet/freescale/enetc/netc_blk_ctrl.c @@ -47,6 +47,13 @@ #define PCS_PROT_SFI BIT(4) #define PCS_PROT_10G_SXGMII BIT(6) +#define IMX94_EXT_PIN_CONTROL 0x10 +#define MAC2_MAC3_SEL BIT(1) + +#define IMX94_NETC_LINK_CFG(a) (0x4c + (a) * 4) +#define NETC_LINK_CFG_MII_PROT GENMASK(3, 0) +#define NETC_LINK_CFG_IO_VAR GENMASK(19, 16) + /* NETC privileged register block register */ #define PRB_NETCRR 0x100 #define NETCRR_SR BIT(0) @@ -68,6 +75,13 @@ #define IMX95_ENETC1_BUS_DEVFN 0x40 #define IMX95_ENETC2_BUS_DEVFN 0x80 +#define IMX94_ENETC0_BUS_DEVFN 0x100 +#define IMX94_ENETC1_BUS_DEVFN 0x140 +#define IMX94_ENETC2_BUS_DEVFN 0x180 +#define IMX94_ENETC0_LINK 3 +#define IMX94_ENETC1_LINK 4 +#define IMX94_ENETC2_LINK 5 + /* Flags for different platforms */ #define NETC_HAS_NETCMIX BIT(0) @@ -192,6 +206,90 @@ static int imx95_netcmix_init(struct platform_device *pdev) return 0; } +static int imx94_enetc_get_link_id(struct device_node *np) +{ + int bus_devfn = netc_of_pci_get_bus_devfn(np); + + /* Parse ENETC link number */ + switch (bus_devfn) { + case IMX94_ENETC0_BUS_DEVFN: + return IMX94_ENETC0_LINK; + case IMX94_ENETC1_BUS_DEVFN: + return IMX94_ENETC1_LINK; + case IMX94_ENETC2_BUS_DEVFN: + return IMX94_ENETC2_LINK; + default: + return -EINVAL; + } +} + +static int imx94_link_config(struct netc_blk_ctrl *priv, + struct device_node *np, int link_id) +{ + phy_interface_t interface; + int mii_proto; + u32 val; + + /* The node may be disabled and does not have a 'phy-mode' + * or 'phy-connection-type' property. + */ + if (of_get_phy_mode(np, &interface)) + return 0; + + mii_proto = netc_get_link_mii_protocol(interface); + if (mii_proto < 0) + return mii_proto; + + val = mii_proto & NETC_LINK_CFG_MII_PROT; + if (val == MII_PROT_SERIAL) + val = u32_replace_bits(val, IO_VAR_16FF_16G_SERDES, + NETC_LINK_CFG_IO_VAR); + + netc_reg_write(priv->netcmix, IMX94_NETC_LINK_CFG(link_id), val); + + return 0; +} + +static int imx94_enetc_link_config(struct netc_blk_ctrl *priv, + struct device_node *np) +{ + int link_id = imx94_enetc_get_link_id(np); + + if (link_id < 0) + return link_id; + + return imx94_link_config(priv, np, link_id); +} + +static int imx94_netcmix_init(struct platform_device *pdev) +{ + struct netc_blk_ctrl *priv = platform_get_drvdata(pdev); + struct device_node *np = pdev->dev.of_node; + u32 val; + int err; + + for_each_child_of_node_scoped(np, child) { + for_each_child_of_node_scoped(child, gchild) { + if (!of_device_is_compatible(gchild, "pci1131,e101")) + continue; + + err = imx94_enetc_link_config(priv, gchild); + if (err) + return err; + } + } + + /* ENETC 0 and switch port 2 share the same parallel interface. + * Currently, the switch is not supported, so this interface is + * used by ENETC 0 by default. + */ + val = netc_reg_read(priv->netcmix, IMX94_EXT_PIN_CONTROL); + val |= MAC2_MAC3_SEL; + netc_reg_write(priv->netcmix, IMX94_EXT_PIN_CONTROL, val); + + return 0; +} + static bool netc_ierb_is_locked(struct netc_blk_ctrl *priv) { return !!(netc_reg_read(priv->prb, PRB_NETCRR) & NETCRR_LOCK); @@ -340,8 +438,14 @@ static const struct netc_devinfo imx95_devinfo = { .ierb_init = imx95_ierb_init, }; +static const struct netc_devinfo imx94_devinfo = { + .flags = NETC_HAS_NETCMIX, + .netcmix_init = imx94_netcmix_init, +}; + static const struct of_device_id netc_blk_ctrl_match[] = { { .compatible = "nxp,imx95-netc-blk-ctrl", .data = &imx95_devinfo }, + { .compatible = "nxp,imx94-netc-blk-ctrl", .data = &imx94_devinfo }, {}, }; MODULE_DEVICE_TABLE(of, netc_blk_ctrl_match); From 1cd3f21c18c293392a8b4abcb1f0a51d7a9efe8c Mon Sep 17 00:00:00 2001 From: Clark Wang Date: Wed, 29 Oct 2025 09:38:58 +0800 Subject: [PATCH 0372/1381] net: enetc: add ptp timer binding support for i.MX94 The i.MX94 has three PTP timers, and all standalone ENETCs can select one of them to bind to as their PHC. The 'ptp-timer' property is used to represent the PTP device of the Ethernet controller. So users can add 'ptp-timer' to the ENETC node to specify the PTP timer. The driver parses this property to bind the two hardware devices. If the "ptp-timer" property is not present, the first timer of the PCIe bus where the ENETC is located is used as the default bound PTP timer. Signed-off-by: Clark Wang Signed-off-by: Wei Fang Link: https://patch.msgid.link/20251029013900.407583-5-wei.fang@nxp.com Signed-off-by: Jakub Kicinski --- .../ethernet/freescale/enetc/netc_blk_ctrl.c | 100 ++++++++++++++++++ 1 file changed, 100 insertions(+) diff --git a/drivers/net/ethernet/freescale/enetc/netc_blk_ctrl.c b/drivers/net/ethernet/freescale/enetc/netc_blk_ctrl.c index 5978ea096e80..d7aee3c934d3 100644 --- a/drivers/net/ethernet/freescale/enetc/netc_blk_ctrl.c +++ b/drivers/net/ethernet/freescale/enetc/netc_blk_ctrl.c @@ -66,6 +66,7 @@ /* NETC integrated endpoint register block register */ #define IERB_EMDIOFAUXR 0x344 #define IERB_T0FAUXR 0x444 +#define IERB_ETBCR(a) (0x300c + 0x100 * (a)) #define IERB_EFAUXR(a) (0x3044 + 0x100 * (a)) #define IERB_VFAUXR(a) (0x4004 + 0x40 * (a)) #define FAUXR_LDID GENMASK(3, 0) @@ -78,10 +79,16 @@ #define IMX94_ENETC0_BUS_DEVFN 0x100 #define IMX94_ENETC1_BUS_DEVFN 0x140 #define IMX94_ENETC2_BUS_DEVFN 0x180 +#define IMX94_TIMER0_BUS_DEVFN 0x1 +#define IMX94_TIMER1_BUS_DEVFN 0x101 +#define IMX94_TIMER2_BUS_DEVFN 0x181 #define IMX94_ENETC0_LINK 3 #define IMX94_ENETC1_LINK 4 #define IMX94_ENETC2_LINK 5 +#define NETC_ENETC_ID(a) (a) +#define NETC_TIMER_ID(a) (a) + /* Flags for different platforms */ #define NETC_HAS_NETCMIX BIT(0) @@ -345,6 +352,98 @@ static int imx95_ierb_init(struct platform_device *pdev) return 0; } +static int imx94_get_enetc_id(struct device_node *np) +{ + int bus_devfn = netc_of_pci_get_bus_devfn(np); + + /* Parse ENETC offset */ + switch (bus_devfn) { + case IMX94_ENETC0_BUS_DEVFN: + return NETC_ENETC_ID(0); + case IMX94_ENETC1_BUS_DEVFN: + return NETC_ENETC_ID(1); + case IMX94_ENETC2_BUS_DEVFN: + return NETC_ENETC_ID(2); + default: + return -EINVAL; + } +} + +static int imx94_get_timer_id(struct device_node *np) +{ + int bus_devfn = netc_of_pci_get_bus_devfn(np); + + /* Parse NETC PTP timer ID, the timer0 is on bus 0, + * the timer 1 and timer2 is on bus 1. + */ + switch (bus_devfn) { + case IMX94_TIMER0_BUS_DEVFN: + return NETC_TIMER_ID(0); + case IMX94_TIMER1_BUS_DEVFN: + return NETC_TIMER_ID(1); + case IMX94_TIMER2_BUS_DEVFN: + return NETC_TIMER_ID(2); + default: + return -EINVAL; + } +} + +static int imx94_enetc_update_tid(struct netc_blk_ctrl *priv, + struct device_node *np) +{ + struct device *dev = &priv->pdev->dev; + struct device_node *timer_np; + int eid, tid; + + eid = imx94_get_enetc_id(np); + if (eid < 0) { + dev_err(dev, "Failed to get ENETC ID\n"); + return eid; + } + + timer_np = of_parse_phandle(np, "ptp-timer", 0); + if (!timer_np) { + /* If 'ptp-timer' is not present, the timer1 is the default + * timer of all standalone ENETCs, which is on the same PCIe + * bus as these ENETCs. + */ + tid = NETC_TIMER_ID(1); + goto end; + } + + tid = imx94_get_timer_id(timer_np); + of_node_put(timer_np); + if (tid < 0) { + dev_err(dev, "Failed to get NETC Timer ID\n"); + return tid; + } + +end: + netc_reg_write(priv->ierb, IERB_ETBCR(eid), tid); + + return 0; +} + +static int imx94_ierb_init(struct platform_device *pdev) +{ + struct netc_blk_ctrl *priv = platform_get_drvdata(pdev); + struct device_node *np = pdev->dev.of_node; + int err; + + for_each_child_of_node_scoped(np, child) { + for_each_child_of_node_scoped(child, gchild) { + if (!of_device_is_compatible(gchild, "pci1131,e101")) + continue; + + err = imx94_enetc_update_tid(priv, gchild); + if (err) + return err; + } + } + + return 0; +} + static int netc_ierb_init(struct platform_device *pdev) { struct netc_blk_ctrl *priv = platform_get_drvdata(pdev); @@ -441,6 +540,7 @@ static const struct netc_devinfo imx95_devinfo = { static const struct netc_devinfo imx94_devinfo = { .flags = NETC_HAS_NETCMIX, .netcmix_init = imx94_netcmix_init, + .ierb_init = imx94_ierb_init, }; static const struct of_device_id netc_blk_ctrl_match[] = { From 5175c1e4adcad3af2cf9d0e5f7a7d1142c3ed320 Mon Sep 17 00:00:00 2001 From: Wei Fang Date: Wed, 29 Oct 2025 09:38:59 +0800 Subject: [PATCH 0373/1381] net: enetc: add basic support for the ENETC with pseudo MAC for i.MX94 The ENETC with pseudo MAC is an internal port which connects to the CPU port of the switch. The switch CPU/host ENETC is fully integrated with the switch and does not require a back-to-back MAC, instead a light weight "pseudo MAC" provides the delineation between switch and ENETC. This translates to lower power (less logic and memory) and lower delay (as there is no serialization delay across this link). Different from the standalone ENETC which is used as the external port, the internal ENETC has a different PCIe device ID, and it does not have Ethernet MAC port registers, instead, it has a small number of pseudo MAC port registers, so some features are not supported by pseudo MAC, such as loopback, half duplex, one-step timestamping and so on. Therefore, the configuration of this internal ENETC is also somewhat different from that of the standalone ENETC. So add the basic support for ENETC with pseudo MAC. More supports will be added in the future. Signed-off-by: Wei Fang Reviewed-by: Claudiu Manoil Link: https://patch.msgid.link/20251029013900.407583-6-wei.fang@nxp.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/freescale/enetc/enetc.c | 24 +++++++- drivers/net/ethernet/freescale/enetc/enetc.h | 8 +++ .../net/ethernet/freescale/enetc/enetc4_hw.h | 30 +++++++++ .../net/ethernet/freescale/enetc/enetc4_pf.c | 15 +++++ .../ethernet/freescale/enetc/enetc_ethtool.c | 61 +++++++++++++++++++ .../net/ethernet/freescale/enetc/enetc_hw.h | 1 + .../freescale/enetc/enetc_pf_common.c | 5 +- 7 files changed, 142 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/freescale/enetc/enetc.c b/drivers/net/ethernet/freescale/enetc/enetc.c index 0535e92404e3..3ed0e04eb589 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc.c +++ b/drivers/net/ethernet/freescale/enetc/enetc.c @@ -14,12 +14,21 @@ u32 enetc_port_mac_rd(struct enetc_si *si, u32 reg) { + /* ENETC with pseudo MAC does not have Ethernet MAC + * port registers. + */ + if (enetc_is_pseudo_mac(si)) + return 0; + return enetc_port_rd(&si->hw, reg); } EXPORT_SYMBOL_GPL(enetc_port_mac_rd); void enetc_port_mac_wr(struct enetc_si *si, u32 reg, u32 val) { + if (enetc_is_pseudo_mac(si)) + return; + enetc_port_wr(&si->hw, reg, val); if (si->hw_features & ENETC_SI_F_QBU) enetc_port_wr(&si->hw, reg + si->drvdata->pmac_offset, val); @@ -3367,7 +3376,8 @@ int enetc_hwtstamp_set(struct net_device *ndev, new_offloads |= ENETC_F_TX_TSTAMP; break; case HWTSTAMP_TX_ONESTEP_SYNC: - if (!enetc_si_is_pf(priv->si)) + if (!enetc_si_is_pf(priv->si) || + enetc_is_pseudo_mac(priv->si)) return -EOPNOTSUPP; new_offloads &= ~ENETC_F_TX_TSTAMP_MASK; @@ -3708,6 +3718,13 @@ static const struct enetc_drvdata enetc4_pf_data = { .eth_ops = &enetc4_pf_ethtool_ops, }; +static const struct enetc_drvdata enetc4_ppm_data = { + .sysclk_freq = ENETC_CLK_333M, + .tx_csum = true, + .max_frags = ENETC4_MAX_SKB_FRAGS, + .eth_ops = &enetc4_ppm_ethtool_ops, +}; + static const struct enetc_drvdata enetc_vf_data = { .sysclk_freq = ENETC_CLK_400M, .max_frags = ENETC_MAX_SKB_FRAGS, @@ -3727,6 +3744,11 @@ static const struct enetc_platform_info enetc_info[] = { .dev_id = ENETC_DEV_ID_VF, .data = &enetc_vf_data, }, + { + .revision = ENETC_REV_4_3, + .dev_id = NXP_ENETC_PPM_DEV_ID, + .data = &enetc4_ppm_data, + }, }; int enetc_get_driver_data(struct enetc_si *si) diff --git a/drivers/net/ethernet/freescale/enetc/enetc.h b/drivers/net/ethernet/freescale/enetc/enetc.h index f279fa597991..dce27bd67a7d 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc.h +++ b/drivers/net/ethernet/freescale/enetc/enetc.h @@ -273,6 +273,7 @@ enum enetc_errata { #define ENETC_SI_F_QBV BIT(1) #define ENETC_SI_F_QBU BIT(2) #define ENETC_SI_F_LSO BIT(3) +#define ENETC_SI_F_PPM BIT(4) /* pseudo MAC */ struct enetc_drvdata { u32 pmac_offset; /* Only valid for PSI which supports 802.1Qbu */ @@ -362,6 +363,11 @@ static inline int enetc_pf_to_port(struct pci_dev *pf_pdev) } } +static inline bool enetc_is_pseudo_mac(struct enetc_si *si) +{ + return si->hw_features & ENETC_SI_F_PPM; +} + #define ENETC_MAX_NUM_TXQS 8 #define ENETC_INT_NAME_MAX (IFNAMSIZ + 8) @@ -534,6 +540,8 @@ int enetc_hwtstamp_set(struct net_device *ndev, extern const struct ethtool_ops enetc_pf_ethtool_ops; extern const struct ethtool_ops enetc4_pf_ethtool_ops; extern const struct ethtool_ops enetc_vf_ethtool_ops; +extern const struct ethtool_ops enetc4_ppm_ethtool_ops; + void enetc_set_ethtool_ops(struct net_device *ndev); void enetc_mm_link_state_update(struct enetc_ndev_priv *priv, bool link); void enetc_mm_commit_preemptible_tcs(struct enetc_ndev_priv *priv); diff --git a/drivers/net/ethernet/freescale/enetc/enetc4_hw.h b/drivers/net/ethernet/freescale/enetc/enetc4_hw.h index 19bf0e89cdc2..ebea4298791c 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc4_hw.h +++ b/drivers/net/ethernet/freescale/enetc/enetc4_hw.h @@ -11,6 +11,7 @@ #define NXP_ENETC_VENDOR_ID 0x1131 #define NXP_ENETC_PF_DEV_ID 0xe101 +#define NXP_ENETC_PPM_DEV_ID 0xe110 /**********************Station interface registers************************/ /* Station interface LSO segmentation flag mask register 0/1 */ @@ -115,6 +116,10 @@ #define PMCAPR_HD BIT(8) #define PMCAPR_FP GENMASK(10, 9) +/* Port capability register */ +#define ENETC4_PCAPR 0x4000 +#define PCAPR_LINK_TYPE BIT(4) + /* Port configuration register */ #define ENETC4_PCR 0x4010 #define PCR_HDR_FMT BIT(0) @@ -193,4 +198,29 @@ #define SSP_1G 2 #define PM_IF_MODE_ENA BIT(15) +/**********************ENETC Pseudo MAC port registers************************/ +/* Port pseudo MAC receive octets counter (64-bit) */ +#define ENETC4_PPMROCR 0x5080 + +/* Port pseudo MAC receive unicast frame counter register (64-bit) */ +#define ENETC4_PPMRUFCR 0x5088 + +/* Port pseudo MAC receive multicast frame counter register (64-bit) */ +#define ENETC4_PPMRMFCR 0x5090 + +/* Port pseudo MAC receive broadcast frame counter register (64-bit) */ +#define ENETC4_PPMRBFCR 0x5098 + +/* Port pseudo MAC transmit octets counter (64-bit) */ +#define ENETC4_PPMTOCR 0x50c0 + +/* Port pseudo MAC transmit unicast frame counter register (64-bit) */ +#define ENETC4_PPMTUFCR 0x50c8 + +/* Port pseudo MAC transmit multicast frame counter register (64-bit) */ +#define ENETC4_PPMTMFCR 0x50d0 + +/* Port pseudo MAC transmit broadcast frame counter register (64-bit) */ +#define ENETC4_PPMTBFCR 0x50d8 + #endif diff --git a/drivers/net/ethernet/freescale/enetc/enetc4_pf.c b/drivers/net/ethernet/freescale/enetc/enetc4_pf.c index 82c443b28b15..498346dd996a 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc4_pf.c +++ b/drivers/net/ethernet/freescale/enetc/enetc4_pf.c @@ -41,6 +41,16 @@ static void enetc4_get_port_caps(struct enetc_pf *pf) pf->caps.mac_filter_num = val & PSIMAFCAPR_NUM_MAC_AFTE; } +static void enetc4_get_psi_hw_features(struct enetc_si *si) +{ + struct enetc_hw *hw = &si->hw; + u32 val; + + val = enetc_port_rd(hw, ENETC4_PCAPR); + if (val & PCAPR_LINK_TYPE) + si->hw_features |= ENETC_SI_F_PPM; +} + static void enetc4_pf_set_si_primary_mac(struct enetc_hw *hw, int si, const u8 *addr) { @@ -277,6 +287,7 @@ static int enetc4_pf_struct_init(struct enetc_si *si) pf->ops = &enetc4_pf_ops; enetc4_get_port_caps(pf); + enetc4_get_psi_hw_features(si); return 0; } @@ -589,6 +600,9 @@ static void enetc4_mac_config(struct enetc_pf *pf, unsigned int mode, struct enetc_si *si = pf->si; u32 val; + if (enetc_is_pseudo_mac(si)) + return; + val = enetc_port_mac_rd(si, ENETC4_PM_IF_MODE(0)); val &= ~(PM_IF_MODE_IFMODE | PM_IF_MODE_ENA); @@ -1071,6 +1085,7 @@ static void enetc4_pf_remove(struct pci_dev *pdev) static const struct pci_device_id enetc4_pf_id_table[] = { { PCI_DEVICE(NXP_ENETC_VENDOR_ID, NXP_ENETC_PF_DEV_ID) }, + { PCI_DEVICE(NXP_ENETC_VENDOR_ID, NXP_ENETC_PPM_DEV_ID) }, { 0, } /* End of table. */ }; MODULE_DEVICE_TABLE(pci, enetc4_pf_id_table); diff --git a/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c b/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c index 71d052de669a..5ef2c5f3ff8f 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c +++ b/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c @@ -435,6 +435,48 @@ static void enetc_get_eth_mac_stats(struct net_device *ndev, } } +static void enetc_ppm_mac_stats(struct enetc_si *si, + struct ethtool_eth_mac_stats *s) +{ + struct enetc_hw *hw = &si->hw; + u64 rufcr, rmfcr, rbfcr; + u64 tufcr, tmfcr, tbfcr; + + rufcr = enetc_port_rd64(hw, ENETC4_PPMRUFCR); + rmfcr = enetc_port_rd64(hw, ENETC4_PPMRMFCR); + rbfcr = enetc_port_rd64(hw, ENETC4_PPMRBFCR); + + tufcr = enetc_port_rd64(hw, ENETC4_PPMTUFCR); + tmfcr = enetc_port_rd64(hw, ENETC4_PPMTMFCR); + tbfcr = enetc_port_rd64(hw, ENETC4_PPMTBFCR); + + s->FramesTransmittedOK = tufcr + tmfcr + tbfcr; + s->FramesReceivedOK = rufcr + rmfcr + rbfcr; + s->OctetsTransmittedOK = enetc_port_rd64(hw, ENETC4_PPMTOCR); + s->OctetsReceivedOK = enetc_port_rd64(hw, ENETC4_PPMROCR); + s->MulticastFramesXmittedOK = tmfcr; + s->BroadcastFramesXmittedOK = tbfcr; + s->MulticastFramesReceivedOK = rmfcr; + s->BroadcastFramesReceivedOK = rbfcr; +} + +static void enetc_ppm_get_eth_mac_stats(struct net_device *ndev, + struct ethtool_eth_mac_stats *mac_stats) +{ + struct enetc_ndev_priv *priv = netdev_priv(ndev); + + switch (mac_stats->src) { + case ETHTOOL_MAC_STATS_SRC_EMAC: + enetc_ppm_mac_stats(priv->si, mac_stats); + break; + case ETHTOOL_MAC_STATS_SRC_PMAC: + break; + case ETHTOOL_MAC_STATS_SRC_AGGREGATE: + ethtool_aggregate_mac_stats(ndev, mac_stats); + break; + } +} + static void enetc_get_eth_ctrl_stats(struct net_device *ndev, struct ethtool_eth_ctrl_stats *ctrl_stats) { @@ -1313,6 +1355,25 @@ const struct ethtool_ops enetc_pf_ethtool_ops = { .get_mm_stats = enetc_get_mm_stats, }; +const struct ethtool_ops enetc4_ppm_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_MAX_FRAMES | + ETHTOOL_COALESCE_USE_ADAPTIVE_RX, + .get_eth_mac_stats = enetc_ppm_get_eth_mac_stats, + .get_rxnfc = enetc4_get_rxnfc, + .get_rxfh_key_size = enetc_get_rxfh_key_size, + .get_rxfh_indir_size = enetc_get_rxfh_indir_size, + .get_rxfh = enetc_get_rxfh, + .set_rxfh = enetc_set_rxfh, + .get_rxfh_fields = enetc_get_rxfh_fields, + .get_ringparam = enetc_get_ringparam, + .get_coalesce = enetc_get_coalesce, + .set_coalesce = enetc_set_coalesce, + .get_link_ksettings = enetc_get_link_ksettings, + .set_link_ksettings = enetc_set_link_ksettings, + .get_link = ethtool_op_get_link, +}; + const struct ethtool_ops enetc_vf_ethtool_ops = { .supported_coalesce_params = ETHTOOL_COALESCE_USECS | ETHTOOL_COALESCE_MAX_FRAMES | diff --git a/drivers/net/ethernet/freescale/enetc/enetc_hw.h b/drivers/net/ethernet/freescale/enetc/enetc_hw.h index 377c96325814..7b882b8921fe 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc_hw.h +++ b/drivers/net/ethernet/freescale/enetc/enetc_hw.h @@ -378,6 +378,7 @@ enum enetc_bdr_type {TX, RX}; #define EIPBRR0_REVISION GENMASK(15, 0) #define ENETC_REV_1_0 0x0100 #define ENETC_REV_4_1 0X0401 +#define ENETC_REV_4_3 0x0403 #define ENETC_G_EIPBRR1 0x0bfc #define ENETC_G_EPFBLPR(n) (0xd00 + 4 * (n)) diff --git a/drivers/net/ethernet/freescale/enetc/enetc_pf_common.c b/drivers/net/ethernet/freescale/enetc/enetc_pf_common.c index edf14a95cab7..9c634205e2a7 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc_pf_common.c +++ b/drivers/net/ethernet/freescale/enetc/enetc_pf_common.c @@ -109,7 +109,7 @@ void enetc_pf_netdev_setup(struct enetc_si *si, struct net_device *ndev, ndev->hw_features = NETIF_F_SG | NETIF_F_RXCSUM | NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_CTAG_RX | - NETIF_F_HW_VLAN_CTAG_FILTER | NETIF_F_LOOPBACK | + NETIF_F_HW_VLAN_CTAG_FILTER | NETIF_F_HW_CSUM | NETIF_F_TSO | NETIF_F_TSO6 | NETIF_F_GSO_UDP_L4; ndev->features = NETIF_F_HIGHDMA | NETIF_F_SG | NETIF_F_RXCSUM | @@ -133,6 +133,9 @@ void enetc_pf_netdev_setup(struct enetc_si *si, struct net_device *ndev, ndev->features |= NETIF_F_RXHASH; } + if (!enetc_is_pseudo_mac(si)) + ndev->hw_features |= NETIF_F_LOOPBACK; + /* TODO: currently, i.MX95 ENETC driver does not support advanced features */ if (!is_enetc_rev1(si)) goto end; From 2d673b0e2f8d75b66aa21e5ca5d95c415dec5be7 Mon Sep 17 00:00:00 2001 From: Wei Fang Date: Wed, 29 Oct 2025 09:39:00 +0800 Subject: [PATCH 0374/1381] net: enetc: add standalone ENETC support for i.MX94 The revision of i.MX94 ENETC is changed to v4.3, so add this revision to enetc_info to support i.MX94 ENETC. And add PTP suspport for i.MX94. Signed-off-by: Wei Fang Reviewed-by: Frank Li Link: https://patch.msgid.link/20251029013900.407583-7-wei.fang@nxp.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/freescale/enetc/enetc.c | 4 ++++ drivers/net/ethernet/freescale/enetc/enetc_ethtool.c | 3 +++ 2 files changed, 7 insertions(+) diff --git a/drivers/net/ethernet/freescale/enetc/enetc.c b/drivers/net/ethernet/freescale/enetc/enetc.c index 3ed0e04eb589..d5e5800b84ef 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc.c +++ b/drivers/net/ethernet/freescale/enetc/enetc.c @@ -3749,6 +3749,10 @@ static const struct enetc_platform_info enetc_info[] = { .dev_id = NXP_ENETC_PPM_DEV_ID, .data = &enetc4_ppm_data, }, + { .revision = ENETC_REV_4_3, + .dev_id = NXP_ENETC_PF_DEV_ID, + .data = &enetc4_pf_data, + }, }; int enetc_get_driver_data(struct enetc_si *si) diff --git a/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c b/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c index 5ef2c5f3ff8f..3e222321b937 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c +++ b/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c @@ -936,6 +936,9 @@ static int enetc_get_phc_index_by_pdev(struct enetc_si *si) case ENETC_REV_4_1: devfn = PCI_DEVFN(24, 0); break; + case ENETC_REV_4_3: + devfn = PCI_DEVFN(0, 1); + break; default: return -1; } From 9f2674e1c3356ec3895d2da404b5c11af965776c Mon Sep 17 00:00:00 2001 From: Sakari Ailus Date: Mon, 27 Oct 2025 13:50:19 +0200 Subject: [PATCH 0375/1381] net: ethernet: Remove redundant pm_runtime_mark_last_busy() calls MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit pm_runtime_put_autosuspend(), pm_runtime_put_sync_autosuspend(), pm_runtime_autosuspend() and pm_request_autosuspend() now include a call to pm_runtime_mark_last_busy(). Remove the now-reduntant explicit call to pm_runtime_mark_last_busy(). Signed-off-by: Sakari Ailus Reviewed-by: Niklas Söderlund Link: https://patch.msgid.link/20251027115022.390997-1-sakari.ailus@linux.intel.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/cadence/macb_main.c | 5 ----- drivers/net/ethernet/freescale/fec_main.c | 8 -------- drivers/net/ethernet/renesas/ravb_main.c | 7 +------ drivers/net/ethernet/ti/davinci_mdio.c | 7 ------- 4 files changed, 1 insertion(+), 26 deletions(-) diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c index b1ed98d9c438..e461f5072884 100644 --- a/drivers/net/ethernet/cadence/macb_main.c +++ b/drivers/net/ethernet/cadence/macb_main.c @@ -327,7 +327,6 @@ static int macb_mdio_read_c22(struct mii_bus *bus, int mii_id, int regnum) status = MACB_BFEXT(DATA, macb_readl(bp, MAN)); mdio_read_exit: - pm_runtime_mark_last_busy(&bp->pdev->dev); pm_runtime_put_autosuspend(&bp->pdev->dev); mdio_pm_exit: return status; @@ -373,7 +372,6 @@ static int macb_mdio_read_c45(struct mii_bus *bus, int mii_id, int devad, status = MACB_BFEXT(DATA, macb_readl(bp, MAN)); mdio_read_exit: - pm_runtime_mark_last_busy(&bp->pdev->dev); pm_runtime_put_autosuspend(&bp->pdev->dev); mdio_pm_exit: return status; @@ -405,7 +403,6 @@ static int macb_mdio_write_c22(struct mii_bus *bus, int mii_id, int regnum, goto mdio_write_exit; mdio_write_exit: - pm_runtime_mark_last_busy(&bp->pdev->dev); pm_runtime_put_autosuspend(&bp->pdev->dev); mdio_pm_exit: return status; @@ -451,7 +448,6 @@ static int macb_mdio_write_c45(struct mii_bus *bus, int mii_id, goto mdio_write_exit; mdio_write_exit: - pm_runtime_mark_last_busy(&bp->pdev->dev); pm_runtime_put_autosuspend(&bp->pdev->dev); mdio_pm_exit: return status; @@ -5622,7 +5618,6 @@ static int macb_probe(struct platform_device *pdev) macb_is_gem(bp) ? "GEM" : "MACB", macb_readl(bp, MID), dev->base_addr, dev->irq, dev->dev_addr); - pm_runtime_mark_last_busy(&bp->pdev->dev); pm_runtime_put_autosuspend(&bp->pdev->dev); return 0; diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c index 1edcfaee6819..4875101c4f3f 100644 --- a/drivers/net/ethernet/freescale/fec_main.c +++ b/drivers/net/ethernet/freescale/fec_main.c @@ -2231,7 +2231,6 @@ static int fec_enet_mdio_read_c22(struct mii_bus *bus, int mii_id, int regnum) ret = FEC_MMFR_DATA(readl(fep->hwp + FEC_MII_DATA)); out: - pm_runtime_mark_last_busy(dev); pm_runtime_put_autosuspend(dev); return ret; @@ -2280,7 +2279,6 @@ static int fec_enet_mdio_read_c45(struct mii_bus *bus, int mii_id, ret = FEC_MMFR_DATA(readl(fep->hwp + FEC_MII_DATA)); out: - pm_runtime_mark_last_busy(dev); pm_runtime_put_autosuspend(dev); return ret; @@ -2312,7 +2310,6 @@ static int fec_enet_mdio_write_c22(struct mii_bus *bus, int mii_id, int regnum, if (ret) netdev_err(fep->netdev, "MDIO write timeout\n"); - pm_runtime_mark_last_busy(dev); pm_runtime_put_autosuspend(dev); return ret; @@ -2356,7 +2353,6 @@ static int fec_enet_mdio_write_c45(struct mii_bus *bus, int mii_id, netdev_err(fep->netdev, "MDIO write timeout\n"); out: - pm_runtime_mark_last_busy(dev); pm_runtime_put_autosuspend(dev); return ret; @@ -2839,7 +2835,6 @@ static void fec_enet_get_regs(struct net_device *ndev, buf[off] = readl(&theregs[off]); } - pm_runtime_mark_last_busy(dev); pm_runtime_put_autosuspend(dev); } @@ -3616,7 +3611,6 @@ fec_enet_open(struct net_device *ndev) err_enet_alloc: fec_enet_clk_enable(ndev, false); clk_enable: - pm_runtime_mark_last_busy(&fep->pdev->dev); pm_runtime_put_autosuspend(&fep->pdev->dev); pinctrl_pm_select_sleep_state(&fep->pdev->dev); return ret; @@ -3647,7 +3641,6 @@ fec_enet_close(struct net_device *ndev) cpu_latency_qos_remove_request(&fep->pm_qos_req); pinctrl_pm_select_sleep_state(&fep->pdev->dev); - pm_runtime_mark_last_busy(&fep->pdev->dev); pm_runtime_put_autosuspend(&fep->pdev->dev); fec_enet_free_buffers(ndev); @@ -4616,7 +4609,6 @@ fec_probe(struct platform_device *pdev) INIT_WORK(&fep->tx_timeout_work, fec_enet_timeout_work); - pm_runtime_mark_last_busy(&pdev->dev); pm_runtime_put_autosuspend(&pdev->dev); return 0; diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c index edfaa5850527..cc619dbebf9d 100644 --- a/drivers/net/ethernet/renesas/ravb_main.c +++ b/drivers/net/ethernet/renesas/ravb_main.c @@ -1975,7 +1975,6 @@ static int ravb_open(struct net_device *ndev) out_set_reset: ravb_set_opmode(ndev, CCC_OPC_RESET); out_rpm_put: - pm_runtime_mark_last_busy(dev); pm_runtime_put_autosuspend(dev); out_napi_off: if (info->nc_queues) @@ -2404,7 +2403,6 @@ static int ravb_close(struct net_device *ndev) if (error) return error; - pm_runtime_mark_last_busy(dev); pm_runtime_put_autosuspend(dev); return 0; @@ -3093,7 +3091,6 @@ static int ravb_probe(struct platform_device *pdev) netdev_info(ndev, "Base address at %#x, %pM, IRQ %d.\n", (u32)ndev->base_addr, ndev->dev_addr, ndev->irq); - pm_runtime_mark_last_busy(&pdev->dev); pm_runtime_put_autosuspend(&pdev->dev); return 0; @@ -3277,10 +3274,8 @@ static int ravb_resume(struct device *dev) return 0; out_rpm_put: - if (!priv->wol_enabled) { - pm_runtime_mark_last_busy(dev); + if (!priv->wol_enabled) pm_runtime_put_autosuspend(dev); - } return ret; } diff --git a/drivers/net/ethernet/ti/davinci_mdio.c b/drivers/net/ethernet/ti/davinci_mdio.c index 68507126be8e..9f049ebbf107 100644 --- a/drivers/net/ethernet/ti/davinci_mdio.c +++ b/drivers/net/ethernet/ti/davinci_mdio.c @@ -234,7 +234,6 @@ static int davinci_mdiobb_read_c22(struct mii_bus *bus, int phy, int reg) ret = mdiobb_read_c22(bus, phy, reg); - pm_runtime_mark_last_busy(bus->parent); pm_runtime_put_autosuspend(bus->parent); return ret; @@ -251,7 +250,6 @@ static int davinci_mdiobb_write_c22(struct mii_bus *bus, int phy, int reg, ret = mdiobb_write_c22(bus, phy, reg, val); - pm_runtime_mark_last_busy(bus->parent); pm_runtime_put_autosuspend(bus->parent); return ret; @@ -268,7 +266,6 @@ static int davinci_mdiobb_read_c45(struct mii_bus *bus, int phy, int devad, ret = mdiobb_read_c45(bus, phy, devad, reg); - pm_runtime_mark_last_busy(bus->parent); pm_runtime_put_autosuspend(bus->parent); return ret; @@ -285,7 +282,6 @@ static int davinci_mdiobb_write_c45(struct mii_bus *bus, int phy, int devad, ret = mdiobb_write_c45(bus, phy, devad, reg, val); - pm_runtime_mark_last_busy(bus->parent); pm_runtime_put_autosuspend(bus->parent); return ret; @@ -332,7 +328,6 @@ static int davinci_mdio_common_reset(struct davinci_mdio_data *data) data->bus->phy_mask = phy_mask; done: - pm_runtime_mark_last_busy(data->dev); pm_runtime_put_autosuspend(data->dev); return 0; @@ -441,7 +436,6 @@ static int davinci_mdio_read(struct mii_bus *bus, int phy_id, int phy_reg) break; } - pm_runtime_mark_last_busy(data->dev); pm_runtime_put_autosuspend(data->dev); return ret; } @@ -478,7 +472,6 @@ static int davinci_mdio_write(struct mii_bus *bus, int phy_id, break; } - pm_runtime_mark_last_busy(data->dev); pm_runtime_put_autosuspend(data->dev); return ret; From a5d937dd0ead5ff826f94eb926f26106b6d178fa Mon Sep 17 00:00:00 2001 From: Sakari Ailus Date: Mon, 27 Oct 2025 13:50:20 +0200 Subject: [PATCH 0376/1381] net: ipa: Remove redundant pm_runtime_mark_last_busy() calls pm_runtime_put_autosuspend(), pm_runtime_put_sync_autosuspend(), pm_runtime_autosuspend() and pm_request_autosuspend() now include a call to pm_runtime_mark_last_busy(). Remove the now-reduntant explicit call to pm_runtime_mark_last_busy(). Signed-off-by: Sakari Ailus Reviewed-by: Dmitry Baryshkov Link: https://patch.msgid.link/20251027115022.390997-2-sakari.ailus@linux.intel.com Signed-off-by: Jakub Kicinski --- drivers/net/ipa/ipa_interrupt.c | 1 - drivers/net/ipa/ipa_main.c | 1 - drivers/net/ipa/ipa_modem.c | 4 ---- drivers/net/ipa/ipa_smp2p.c | 2 -- drivers/net/ipa/ipa_uc.c | 2 -- 5 files changed, 10 deletions(-) diff --git a/drivers/net/ipa/ipa_interrupt.c b/drivers/net/ipa/ipa_interrupt.c index 245a06997055..8336596b1247 100644 --- a/drivers/net/ipa/ipa_interrupt.c +++ b/drivers/net/ipa/ipa_interrupt.c @@ -149,7 +149,6 @@ static irqreturn_t ipa_isr_thread(int irq, void *dev_id) iowrite32(pending, ipa->reg_virt + reg_offset(reg)); } out_power_put: - pm_runtime_mark_last_busy(dev); (void)pm_runtime_put_autosuspend(dev); return IRQ_HANDLED; diff --git a/drivers/net/ipa/ipa_main.c b/drivers/net/ipa/ipa_main.c index 25500c5a6928..95a61bae3124 100644 --- a/drivers/net/ipa/ipa_main.c +++ b/drivers/net/ipa/ipa_main.c @@ -903,7 +903,6 @@ static int ipa_probe(struct platform_device *pdev) if (ret) goto err_deconfig; done: - pm_runtime_mark_last_busy(dev); (void)pm_runtime_put_autosuspend(dev); return 0; diff --git a/drivers/net/ipa/ipa_modem.c b/drivers/net/ipa/ipa_modem.c index 8fe0d0e1a00f..9b136f6b8b4a 100644 --- a/drivers/net/ipa/ipa_modem.c +++ b/drivers/net/ipa/ipa_modem.c @@ -71,7 +71,6 @@ static int ipa_open(struct net_device *netdev) netif_start_queue(netdev); - pm_runtime_mark_last_busy(dev); (void)pm_runtime_put_autosuspend(dev); return 0; @@ -102,7 +101,6 @@ static int ipa_stop(struct net_device *netdev) ipa_endpoint_disable_one(priv->rx); ipa_endpoint_disable_one(priv->tx); out_power_put: - pm_runtime_mark_last_busy(dev); (void)pm_runtime_put_autosuspend(dev); return 0; @@ -175,7 +173,6 @@ ipa_start_xmit(struct sk_buff *skb, struct net_device *netdev) ret = ipa_endpoint_skb_tx(endpoint, skb); - pm_runtime_mark_last_busy(dev); (void)pm_runtime_put_autosuspend(dev); if (ret) { @@ -432,7 +429,6 @@ static void ipa_modem_crashed(struct ipa *ipa) dev_err(dev, "error %d zeroing modem memory regions\n", ret); out_power_put: - pm_runtime_mark_last_busy(dev); (void)pm_runtime_put_autosuspend(dev); } diff --git a/drivers/net/ipa/ipa_smp2p.c b/drivers/net/ipa/ipa_smp2p.c index fcaadd111a8a..420098796eec 100644 --- a/drivers/net/ipa/ipa_smp2p.c +++ b/drivers/net/ipa/ipa_smp2p.c @@ -171,7 +171,6 @@ static irqreturn_t ipa_smp2p_modem_setup_ready_isr(int irq, void *dev_id) WARN(ret != 0, "error %d from ipa_setup()\n", ret); out_power_put: - pm_runtime_mark_last_busy(dev); (void)pm_runtime_put_autosuspend(dev); return IRQ_HANDLED; @@ -213,7 +212,6 @@ static void ipa_smp2p_power_release(struct ipa *ipa) if (!ipa->smp2p->power_on) return; - pm_runtime_mark_last_busy(dev); (void)pm_runtime_put_autosuspend(dev); ipa->smp2p->power_on = false; } diff --git a/drivers/net/ipa/ipa_uc.c b/drivers/net/ipa/ipa_uc.c index 2963db83ab6b..dc7e92f2a4fb 100644 --- a/drivers/net/ipa/ipa_uc.c +++ b/drivers/net/ipa/ipa_uc.c @@ -158,7 +158,6 @@ static void ipa_uc_response_hdlr(struct ipa *ipa) if (ipa->uc_powered) { ipa->uc_loaded = true; ipa_power_retention(ipa, true); - pm_runtime_mark_last_busy(dev); (void)pm_runtime_put_autosuspend(dev); ipa->uc_powered = false; } else { @@ -203,7 +202,6 @@ void ipa_uc_deconfig(struct ipa *ipa) if (!ipa->uc_powered) return; - pm_runtime_mark_last_busy(dev); (void)pm_runtime_put_autosuspend(dev); } From 10c7b9be47e683634535c360e75ec2a48e73b2ce Mon Sep 17 00:00:00 2001 From: Sakari Ailus Date: Mon, 27 Oct 2025 13:50:22 +0200 Subject: [PATCH 0377/1381] net: wwan: Remove redundant pm_runtime_mark_last_busy() calls pm_runtime_put_autosuspend(), pm_runtime_put_sync_autosuspend(), pm_runtime_autosuspend() and pm_request_autosuspend() now include a call to pm_runtime_mark_last_busy(). Remove the now-reduntant explicit call to pm_runtime_mark_last_busy(). Signed-off-by: Sakari Ailus Link: https://patch.msgid.link/20251027115022.390997-4-sakari.ailus@linux.intel.com Signed-off-by: Jakub Kicinski --- drivers/net/wwan/qcom_bam_dmux.c | 2 -- drivers/net/wwan/t7xx/t7xx_hif_cldma.c | 3 --- drivers/net/wwan/t7xx/t7xx_hif_dpmaif_rx.c | 2 -- drivers/net/wwan/t7xx/t7xx_hif_dpmaif_tx.c | 2 -- 4 files changed, 9 deletions(-) diff --git a/drivers/net/wwan/qcom_bam_dmux.c b/drivers/net/wwan/qcom_bam_dmux.c index 64dab8b57611..6a5b22589af4 100644 --- a/drivers/net/wwan/qcom_bam_dmux.c +++ b/drivers/net/wwan/qcom_bam_dmux.c @@ -162,7 +162,6 @@ static void bam_dmux_tx_done(struct bam_dmux_skb_dma *skb_dma) struct bam_dmux *dmux = skb_dma->dmux; unsigned long flags; - pm_runtime_mark_last_busy(dmux->dev); pm_runtime_put_autosuspend(dmux->dev); if (skb_dma->addr) @@ -397,7 +396,6 @@ static void bam_dmux_tx_wakeup_work(struct work_struct *work) dma_async_issue_pending(dmux->tx); out: - pm_runtime_mark_last_busy(dmux->dev); pm_runtime_put_autosuspend(dmux->dev); } diff --git a/drivers/net/wwan/t7xx/t7xx_hif_cldma.c b/drivers/net/wwan/t7xx/t7xx_hif_cldma.c index 97163e1e5783..689c920ca898 100644 --- a/drivers/net/wwan/t7xx/t7xx_hif_cldma.c +++ b/drivers/net/wwan/t7xx/t7xx_hif_cldma.c @@ -250,7 +250,6 @@ static void t7xx_cldma_rx_done(struct work_struct *work) t7xx_cldma_clear_ip_busy(&md_ctrl->hw_info); t7xx_cldma_hw_irq_en_txrx(&md_ctrl->hw_info, queue->index, MTK_RX); t7xx_cldma_hw_irq_en_eq(&md_ctrl->hw_info, queue->index, MTK_RX); - pm_runtime_mark_last_busy(md_ctrl->dev); pm_runtime_put_autosuspend(md_ctrl->dev); } @@ -362,7 +361,6 @@ static void t7xx_cldma_tx_done(struct work_struct *work) } spin_unlock_irqrestore(&md_ctrl->cldma_lock, flags); - pm_runtime_mark_last_busy(md_ctrl->dev); pm_runtime_put_autosuspend(md_ctrl->dev); } @@ -987,7 +985,6 @@ int t7xx_cldma_send_skb(struct cldma_ctrl *md_ctrl, int qno, struct sk_buff *skb allow_sleep: t7xx_pci_enable_sleep(md_ctrl->t7xx_dev); - pm_runtime_mark_last_busy(md_ctrl->dev); pm_runtime_put_autosuspend(md_ctrl->dev); return ret; } diff --git a/drivers/net/wwan/t7xx/t7xx_hif_dpmaif_rx.c b/drivers/net/wwan/t7xx/t7xx_hif_dpmaif_rx.c index 2310493203d3..b76bea6ab2d7 100644 --- a/drivers/net/wwan/t7xx/t7xx_hif_dpmaif_rx.c +++ b/drivers/net/wwan/t7xx/t7xx_hif_dpmaif_rx.c @@ -877,7 +877,6 @@ int t7xx_dpmaif_napi_rx_poll(struct napi_struct *napi, const int budget) t7xx_dpmaif_clr_ip_busy_sts(&rxq->dpmaif_ctrl->hw_info); t7xx_dpmaif_dlq_unmask_rx_done(&rxq->dpmaif_ctrl->hw_info, rxq->index); t7xx_pci_enable_sleep(rxq->dpmaif_ctrl->t7xx_dev); - pm_runtime_mark_last_busy(rxq->dpmaif_ctrl->dev); pm_runtime_put_autosuspend(rxq->dpmaif_ctrl->dev); atomic_set(&rxq->rx_processing, 0); } else { @@ -1078,7 +1077,6 @@ static void t7xx_dpmaif_bat_release_work(struct work_struct *work) } t7xx_pci_enable_sleep(dpmaif_ctrl->t7xx_dev); - pm_runtime_mark_last_busy(dpmaif_ctrl->dev); pm_runtime_put_autosuspend(dpmaif_ctrl->dev); } diff --git a/drivers/net/wwan/t7xx/t7xx_hif_dpmaif_tx.c b/drivers/net/wwan/t7xx/t7xx_hif_dpmaif_tx.c index 8dab025a088a..236d632cf591 100644 --- a/drivers/net/wwan/t7xx/t7xx_hif_dpmaif_tx.c +++ b/drivers/net/wwan/t7xx/t7xx_hif_dpmaif_tx.c @@ -185,7 +185,6 @@ static void t7xx_dpmaif_tx_done(struct work_struct *work) } t7xx_pci_enable_sleep(dpmaif_ctrl->t7xx_dev); - pm_runtime_mark_last_busy(dpmaif_ctrl->dev); pm_runtime_put_autosuspend(dpmaif_ctrl->dev); } @@ -468,7 +467,6 @@ static int t7xx_dpmaif_tx_hw_push_thread(void *arg) t7xx_pci_disable_sleep(dpmaif_ctrl->t7xx_dev); t7xx_do_tx_hw_push(dpmaif_ctrl); t7xx_pci_enable_sleep(dpmaif_ctrl->t7xx_dev); - pm_runtime_mark_last_busy(dpmaif_ctrl->dev); pm_runtime_put_autosuspend(dpmaif_ctrl->dev); } From cf35f4347ddd42ceca156e57a87ec489788f1bd7 Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Sat, 25 Oct 2025 20:35:47 +0200 Subject: [PATCH 0378/1381] net: stmmac: mdio: fix incorrect phy address check max_addr is the max number of addresses, not the highest possible address, therefore check phydev->mdio.addr > max_addr isn't correct. To fix this change the semantics of max_addr, so that it represents the highest possible address. IMO this is also a little bit more intuitive wrt name max_addr. Fixes: 4a107a0e8361 ("net: stmmac: mdio: use phy_find_first to simplify stmmac_mdio_register") Reported-by: Dan Carpenter Reported-by: Simon Horman Signed-off-by: Heiner Kallweit Reviewed-by: Simon Horman Link: https://patch.msgid.link/e869999b-2d4b-4dc1-9890-c2d3d1e8d0f8@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c | 9 ++------- 1 file changed, 2 insertions(+), 7 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c index 3f8cc3293964..1e82850f2a25 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c @@ -583,9 +583,9 @@ int stmmac_mdio_register(struct net_device *ndev) struct device_node *mdio_node = priv->plat->mdio_node; struct device *dev = ndev->dev.parent; struct fwnode_handle *fixed_node; + int max_addr = PHY_MAX_ADDR - 1; struct fwnode_handle *fwnode; struct phy_device *phydev; - int max_addr; if (!mdio_bus_data) return 0; @@ -609,15 +609,12 @@ int stmmac_mdio_register(struct net_device *ndev) if (priv->synopsys_id < DWXGMAC_CORE_2_20) { /* Right now only C22 phys are supported */ - max_addr = MII_XGMAC_MAX_C22ADDR + 1; + max_addr = MII_XGMAC_MAX_C22ADDR; /* Check if DT specified an unsupported phy addr */ if (priv->plat->phy_addr > MII_XGMAC_MAX_C22ADDR) dev_err(dev, "Unsupported phy_addr (max=%d)\n", MII_XGMAC_MAX_C22ADDR); - } else { - /* XGMAC version 2.20 onwards support 32 phy addr */ - max_addr = PHY_MAX_ADDR; } } else { new_bus->read = &stmmac_mdio_read_c22; @@ -626,8 +623,6 @@ int stmmac_mdio_register(struct net_device *ndev) new_bus->read_c45 = &stmmac_mdio_read_c45; new_bus->write_c45 = &stmmac_mdio_write_c45; } - - max_addr = PHY_MAX_ADDR; } if (mdio_bus_data->needs_reset) From 26888de97b2ffe0267c12dd4e9fcd552545903f1 Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Sat, 25 Oct 2025 20:49:50 +0200 Subject: [PATCH 0379/1381] net: phy: add iterator mdiobus_for_each_phy Add an iterator for all PHY's on a MII bus, and phy_find_next() as a prerequisite. Signed-off-by: Heiner Kallweit Reviewed-by: Wei Fang Link: https://patch.msgid.link/cd112f15-401a-43d9-8525-9ff0965a68cd@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/phy/phy_device.c | 16 +++++++++------- include/linux/phy.h | 11 ++++++++++- 2 files changed, 19 insertions(+), 8 deletions(-) diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c index b7feaf0cb1df..737747cf1906 100644 --- a/drivers/net/phy/phy_device.c +++ b/drivers/net/phy/phy_device.c @@ -1224,22 +1224,24 @@ int phy_get_c45_ids(struct phy_device *phydev) EXPORT_SYMBOL(phy_get_c45_ids); /** - * phy_find_first - finds the first PHY device on the bus + * phy_find_next - finds the next PHY device on the bus * @bus: the target MII bus + * @pos: cursor + * + * Return: next phy_device on the bus, or NULL */ -struct phy_device *phy_find_first(struct mii_bus *bus) +struct phy_device *phy_find_next(struct mii_bus *bus, struct phy_device *pos) { - struct phy_device *phydev; - int addr; + for (int addr = pos ? pos->mdio.addr + 1 : 0; + addr < PHY_MAX_ADDR; addr++) { + struct phy_device *phydev = mdiobus_get_phy(bus, addr); - for (addr = 0; addr < PHY_MAX_ADDR; addr++) { - phydev = mdiobus_get_phy(bus, addr); if (phydev) return phydev; } return NULL; } -EXPORT_SYMBOL(phy_find_first); +EXPORT_SYMBOL_GPL(phy_find_next); /** * phy_prepare_link - prepares the PHY layer to monitor link status diff --git a/include/linux/phy.h b/include/linux/phy.h index 17a2cdc9f1a0..358dd6f0ff96 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -1869,7 +1869,7 @@ int phy_sfp_probe(struct phy_device *phydev, const struct sfp_upstream_ops *ops); struct phy_device *phy_attach(struct net_device *dev, const char *bus_id, phy_interface_t interface); -struct phy_device *phy_find_first(struct mii_bus *bus); +struct phy_device *phy_find_next(struct mii_bus *bus, struct phy_device *pos); int phy_attach_direct(struct net_device *dev, struct phy_device *phydev, u32 flags, phy_interface_t interface); int phy_connect_direct(struct net_device *dev, struct phy_device *phydev, @@ -1896,6 +1896,15 @@ bool phy_check_valid(int speed, int duplex, unsigned long *features); int phy_restart_aneg(struct phy_device *phydev); int phy_reset_after_clk_enable(struct phy_device *phydev); +static inline struct phy_device *phy_find_first(struct mii_bus *bus) +{ + return phy_find_next(bus, NULL); +} + +#define mdiobus_for_each_phy(_bus, _phydev) \ + for (_phydev = phy_find_first(_bus); _phydev; \ + _phydev = phy_find_next(_bus, _phydev)) + #if IS_ENABLED(CONFIG_PHYLIB) int phy_start_cable_test(struct phy_device *phydev, struct netlink_ext_ack *extack); From 0514010d553aa8e820edaee5227cc4f3ae183ae3 Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Sat, 25 Oct 2025 20:50:41 +0200 Subject: [PATCH 0380/1381] net: fec: use new iterator mdiobus_for_each_phy Use new iterator mdiobus_for_each_phy() to simplify the code. Signed-off-by: Heiner Kallweit Reviewed-by: Wei Fang Link: https://patch.msgid.link/65eb9490-5666-4b4a-8d26-3fca738b1315@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/freescale/fec_main.c | 8 ++------ 1 file changed, 2 insertions(+), 6 deletions(-) diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c index 4875101c4f3f..024dd443bfbc 100644 --- a/drivers/net/ethernet/freescale/fec_main.c +++ b/drivers/net/ethernet/freescale/fec_main.c @@ -2548,7 +2548,6 @@ static int fec_enet_mii_init(struct platform_device *pdev) int err = -ENXIO; u32 mii_speed, holdtime; u32 bus_freq; - int addr; /* * The i.MX28 dual fec interfaces are not equal. @@ -2663,11 +2662,8 @@ static int fec_enet_mii_init(struct platform_device *pdev) of_node_put(node); /* find all the PHY devices on the bus and set mac_managed_pm to true */ - for (addr = 0; addr < PHY_MAX_ADDR; addr++) { - phydev = mdiobus_get_phy(fep->mii_bus, addr); - if (phydev) - phydev->mac_managed_pm = true; - } + mdiobus_for_each_phy(fep->mii_bus, phydev) + phydev->mac_managed_pm = true; mii_cnt++; From 4575875065dee7f60452b557230e12f45c4bb012 Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Sat, 25 Oct 2025 20:52:09 +0200 Subject: [PATCH 0381/1381] net: davinci_mdio: use new iterator mdiobus_for_each_phy Use new iterator mdiobus_for_each_phy() to simplify the code. Reviewed-by: Siddharth Vadapalli Signed-off-by: Heiner Kallweit Link: https://patch.msgid.link/326d1337-2c22-42e3-a152-046ac5c43095@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/ti/davinci_mdio.c | 14 +++++--------- 1 file changed, 5 insertions(+), 9 deletions(-) diff --git a/drivers/net/ethernet/ti/davinci_mdio.c b/drivers/net/ethernet/ti/davinci_mdio.c index 9f049ebbf107..48f85a3649b2 100644 --- a/drivers/net/ethernet/ti/davinci_mdio.c +++ b/drivers/net/ethernet/ti/davinci_mdio.c @@ -541,8 +541,8 @@ static int davinci_mdio_probe(struct platform_device *pdev) struct davinci_mdio_data *data; struct resource *res; struct phy_device *phy; - int ret, addr; int autosuspend_delay_ms = -1; + int ret; data = devm_kzalloc(dev, sizeof(*data), GFP_KERNEL); if (!data) @@ -645,14 +645,10 @@ static int davinci_mdio_probe(struct platform_device *pdev) goto bail_out; /* scan and dump the bus */ - for (addr = 0; addr < PHY_MAX_ADDR; addr++) { - phy = mdiobus_get_phy(data->bus, addr); - if (phy) { - dev_info(dev, "phy[%d]: device %s, driver %s\n", - phy->mdio.addr, phydev_name(phy), - phy->drv ? phy->drv->name : "unknown"); - } - } + mdiobus_for_each_phy(data->bus, phy) + dev_info(dev, "phy[%d]: device %s, driver %s\n", + phy->mdio.addr, phydev_name(phy), + phy->drv ? phy->drv->name : "unknown"); return 0; From d4780abb8cce3692f24ea2f1cf162092075e91e8 Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Sat, 25 Oct 2025 20:52:56 +0200 Subject: [PATCH 0382/1381] net: phy: use new iterator mdiobus_for_each_phy in mdiobus_prevent_c45_scan Use new iterator mdiobus_for_each_phy() to simplify the code. Signed-off-by: Heiner Kallweit Link: https://patch.msgid.link/6d792b1e-d23d-4b7e-a94f-89c6617b620f@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/phy/mdio_bus_provider.c | 13 ++++--------- 1 file changed, 4 insertions(+), 9 deletions(-) diff --git a/drivers/net/phy/mdio_bus_provider.c b/drivers/net/phy/mdio_bus_provider.c index a2391d4b7e5c..4b0637405740 100644 --- a/drivers/net/phy/mdio_bus_provider.c +++ b/drivers/net/phy/mdio_bus_provider.c @@ -249,20 +249,15 @@ static int mdiobus_scan_bus_c45(struct mii_bus *bus) */ static bool mdiobus_prevent_c45_scan(struct mii_bus *bus) { - int i; + struct phy_device *phydev; - for (i = 0; i < PHY_MAX_ADDR; i++) { - struct phy_device *phydev; - u32 oui; - - phydev = mdiobus_get_phy(bus, i); - if (!phydev) - continue; - oui = phydev->phy_id >> 10; + mdiobus_for_each_phy(bus, phydev) { + u32 oui = phydev->phy_id >> 10; if (oui == MICREL_OUI) return true; } + return false; } From 0a119c68d0e6b64f460516ad832a7f171be55594 Mon Sep 17 00:00:00 2001 From: Sakari Ailus Date: Mon, 27 Oct 2025 13:50:21 +0200 Subject: [PATCH 0383/1381] net: wireless: Remove redundant pm_runtime_mark_last_busy() calls pm_runtime_put_autosuspend(), pm_runtime_put_sync_autosuspend(), pm_runtime_autosuspend() and pm_request_autosuspend() now include a call to pm_runtime_mark_last_busy(). Remove the now-reduntant explicit call to pm_runtime_mark_last_busy(). Signed-off-by: Sakari Ailus Link: https://patch.msgid.link/20251027115022.390997-3-sakari.ailus@linux.intel.com Signed-off-by: Johannes Berg --- drivers/net/wireless/ath/wil6210/pm.c | 1 - drivers/net/wireless/ti/wl18xx/debugfs.c | 3 -- drivers/net/wireless/ti/wlcore/cmd.c | 1 - drivers/net/wireless/ti/wlcore/debugfs.c | 11 ------- drivers/net/wireless/ti/wlcore/main.c | 36 --------------------- drivers/net/wireless/ti/wlcore/scan.c | 1 - drivers/net/wireless/ti/wlcore/sysfs.c | 1 - drivers/net/wireless/ti/wlcore/testmode.c | 2 -- drivers/net/wireless/ti/wlcore/tx.c | 1 - drivers/net/wireless/ti/wlcore/vendor_cmd.c | 3 -- 10 files changed, 60 deletions(-) diff --git a/drivers/net/wireless/ath/wil6210/pm.c b/drivers/net/wireless/ath/wil6210/pm.c index f521af575e9b..c866cfd144c7 100644 --- a/drivers/net/wireless/ath/wil6210/pm.c +++ b/drivers/net/wireless/ath/wil6210/pm.c @@ -458,6 +458,5 @@ void wil_pm_runtime_put(struct wil6210_priv *wil) { struct device *dev = wil_to_dev(wil); - pm_runtime_mark_last_busy(dev); pm_runtime_put_autosuspend(dev); } diff --git a/drivers/net/wireless/ti/wl18xx/debugfs.c b/drivers/net/wireless/ti/wl18xx/debugfs.c index 80fbf740fe6d..ac756318e8ea 100644 --- a/drivers/net/wireless/ti/wl18xx/debugfs.c +++ b/drivers/net/wireless/ti/wl18xx/debugfs.c @@ -272,7 +272,6 @@ static ssize_t radar_detection_write(struct file *file, if (ret < 0) count = ret; - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: mutex_unlock(&wl->mutex); @@ -312,7 +311,6 @@ static ssize_t dynamic_fw_traces_write(struct file *file, if (ret < 0) count = ret; - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: mutex_unlock(&wl->mutex); @@ -374,7 +372,6 @@ static ssize_t radar_debug_mode_write(struct file *file, wl->radar_debug_mode, 0); } - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: mutex_unlock(&wl->mutex); diff --git a/drivers/net/wireless/ti/wlcore/cmd.c b/drivers/net/wireless/ti/wlcore/cmd.c index fa3a3f71dd15..9d73ba933a16 100644 --- a/drivers/net/wireless/ti/wlcore/cmd.c +++ b/drivers/net/wireless/ti/wlcore/cmd.c @@ -213,7 +213,6 @@ int wlcore_cmd_wait_for_event_or_timeout(struct wl1271 *wl, } while (!event); out: - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); free_vector: kfree(events_vector); diff --git a/drivers/net/wireless/ti/wlcore/debugfs.c b/drivers/net/wireless/ti/wlcore/debugfs.c index eb3d3f0e0b4d..bbfd2725215b 100644 --- a/drivers/net/wireless/ti/wlcore/debugfs.c +++ b/drivers/net/wireless/ti/wlcore/debugfs.c @@ -63,7 +63,6 @@ void wl1271_debugfs_update_stats(struct wl1271 *wl) wl->stats.fw_stats_update = jiffies; } - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: @@ -113,7 +112,6 @@ static void chip_op_handler(struct wl1271 *wl, unsigned long value, chip_op = arg; chip_op(wl); - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); } @@ -287,7 +285,6 @@ static ssize_t dynamic_ps_timeout_write(struct file *file, wl1271_ps_set_mode(wl, wlvif, STATION_AUTO_PS_MODE); } - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: @@ -357,7 +354,6 @@ static ssize_t forced_ps_write(struct file *file, wl1271_ps_set_mode(wl, wlvif, ps_mode); } - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: @@ -830,7 +826,6 @@ static ssize_t rx_streaming_interval_write(struct file *file, wl1271_recalc_rx_streaming(wl, wlvif); } - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: mutex_unlock(&wl->mutex); @@ -886,7 +881,6 @@ static ssize_t rx_streaming_always_write(struct file *file, wl1271_recalc_rx_streaming(wl, wlvif); } - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: mutex_unlock(&wl->mutex); @@ -934,7 +928,6 @@ static ssize_t beacon_filtering_write(struct file *file, ret = wl1271_acx_beacon_filter_opt(wl, wlvif, !!value); } - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: mutex_unlock(&wl->mutex); @@ -1015,7 +1008,6 @@ static ssize_t sleep_auth_write(struct file *file, goto out_sleep; out_sleep: - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: mutex_unlock(&wl->mutex); @@ -1090,7 +1082,6 @@ static ssize_t dev_mem_read(struct file *file, goto part_err; part_err: - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); skip_read: @@ -1172,7 +1163,6 @@ static ssize_t dev_mem_write(struct file *file, const char __user *user_buf, goto part_err; part_err: - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); skip_write: @@ -1247,7 +1237,6 @@ static ssize_t fw_logger_write(struct file *file, ret = wl12xx_cmd_config_fwlog(wl); - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: diff --git a/drivers/net/wireless/ti/wlcore/main.c b/drivers/net/wireless/ti/wlcore/main.c index 6116a8522d96..12f0167d7380 100644 --- a/drivers/net/wireless/ti/wlcore/main.c +++ b/drivers/net/wireless/ti/wlcore/main.c @@ -154,7 +154,6 @@ static void wl1271_rx_streaming_enable_work(struct work_struct *work) jiffies + msecs_to_jiffies(wl->conf.rx_streaming.duration)); out_sleep: - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: mutex_unlock(&wl->mutex); @@ -181,7 +180,6 @@ static void wl1271_rx_streaming_disable_work(struct work_struct *work) goto out_sleep; out_sleep: - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: mutex_unlock(&wl->mutex); @@ -234,7 +232,6 @@ static void wlcore_rc_update_work(struct work_struct *work) } out_sleep: - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: mutex_unlock(&wl->mutex); @@ -711,7 +708,6 @@ static int wlcore_irq_locked(struct wl1271 *wl) } err_ret: - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: @@ -1047,7 +1043,6 @@ static void wl1271_recovery_work(struct work_struct *work) } wlcore_op_stop_locked(wl); - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); ieee80211_restart_hw(wl->hw); @@ -1943,7 +1938,6 @@ static int __maybe_unused wl1271_op_resume(struct ieee80211_hw *hw) goto out_sleep; out_sleep: - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: @@ -2131,7 +2125,6 @@ static void wlcore_channel_switch_work(struct work_struct *work) wl12xx_cmd_stop_channel_switch(wl, wlvif); - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: mutex_unlock(&wl->mutex); @@ -2201,7 +2194,6 @@ static void wlcore_pending_auth_complete_work(struct work_struct *work) /* cancel the ROC if active */ wlcore_update_inconn_sta(wl, wlvif, NULL, false); - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: mutex_unlock(&wl->mutex); @@ -2694,7 +2686,6 @@ static int wl1271_op_add_interface(struct ieee80211_hw *hw, else wl->sta_count++; out: - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out_unlock: mutex_unlock(&wl->mutex); @@ -2774,7 +2765,6 @@ static void __wl1271_op_remove_interface(struct wl1271 *wl, } } - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); } deinit: @@ -3200,7 +3190,6 @@ static int wl1271_op_config(struct ieee80211_hw *hw, int radio_idx, u32 changed) } out_sleep: - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: @@ -3315,7 +3304,6 @@ static void wl1271_op_configure_filter(struct ieee80211_hw *hw, */ out_sleep: - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: @@ -3531,7 +3519,6 @@ static int wlcore_op_set_key(struct ieee80211_hw *hw, enum set_key_cmd cmd, ret = wlcore_hw_set_key(wl, cmd, vif, sta, key_conf); - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out_wake_queues: @@ -3695,7 +3682,6 @@ static void wl1271_op_set_default_key_idx(struct ieee80211_hw *hw, } out_sleep: - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out_unlock: @@ -3724,7 +3710,6 @@ void wlcore_regdomain_config(struct wl1271 *wl) goto out; } - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: mutex_unlock(&wl->mutex); @@ -3772,7 +3757,6 @@ static int wl1271_op_hw_scan(struct ieee80211_hw *hw, ret = wlcore_scan(hw->priv, vif, ssid, len, req); out_sleep: - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: mutex_unlock(&wl->mutex); @@ -3823,7 +3807,6 @@ static void wl1271_op_cancel_hw_scan(struct ieee80211_hw *hw, ieee80211_scan_completed(wl->hw, &info); out_sleep: - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: mutex_unlock(&wl->mutex); @@ -3860,7 +3843,6 @@ static int wl1271_op_sched_scan_start(struct ieee80211_hw *hw, wl->sched_vif = wlvif; out_sleep: - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: mutex_unlock(&wl->mutex); @@ -3887,7 +3869,6 @@ static int wl1271_op_sched_scan_stop(struct ieee80211_hw *hw, wl->ops->sched_scan_stop(wl, wlvif); - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: mutex_unlock(&wl->mutex); @@ -3916,7 +3897,6 @@ static int wl1271_op_set_frag_threshold(struct ieee80211_hw *hw, if (ret < 0) wl1271_warning("wl1271_op_set_frag_threshold failed: %d", ret); - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: @@ -3948,7 +3928,6 @@ static int wl1271_op_set_rts_threshold(struct ieee80211_hw *hw, int radio_idx, if (ret < 0) wl1271_warning("set rts threshold failed: %d", ret); } - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: @@ -4714,7 +4693,6 @@ static void wl1271_op_bss_info_changed(struct ieee80211_hw *hw, else wl1271_bss_info_changed_sta(wl, vif, bss_conf, changed); - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: @@ -4779,7 +4757,6 @@ static void wlcore_op_change_chanctx(struct ieee80211_hw *hw, } } - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: mutex_unlock(&wl->mutex); @@ -4828,7 +4805,6 @@ static int wlcore_op_assign_vif_chanctx(struct ieee80211_hw *hw, wlvif->radar_enabled = true; } - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: mutex_unlock(&wl->mutex); @@ -4871,7 +4847,6 @@ static void wlcore_op_unassign_vif_chanctx(struct ieee80211_hw *hw, wlvif->radar_enabled = false; } - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: mutex_unlock(&wl->mutex); @@ -4941,7 +4916,6 @@ wlcore_op_switch_vif_chanctx(struct ieee80211_hw *hw, goto out_sleep; } out_sleep: - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: mutex_unlock(&wl->mutex); @@ -4995,7 +4969,6 @@ static int wl1271_op_conf_tx(struct ieee80211_hw *hw, 0, 0); out_sleep: - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: @@ -5029,7 +5002,6 @@ static u64 wl1271_op_get_tsf(struct ieee80211_hw *hw, goto out_sleep; out_sleep: - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: @@ -5342,7 +5314,6 @@ static int wl12xx_op_sta_state(struct ieee80211_hw *hw, ret = wl12xx_update_sta_state(wl, wlvif, sta, old_state, new_state); - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: mutex_unlock(&wl->mutex); @@ -5467,7 +5438,6 @@ static int wl1271_op_ampdu_action(struct ieee80211_hw *hw, ret = -EINVAL; } - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: @@ -5511,7 +5481,6 @@ static int wl12xx_set_bitrate_mask(struct ieee80211_hw *hw, wl1271_tx_min_rate_get(wl, wlvif->basic_rate_set); ret = wl1271_acx_sta_rate_policies(wl, wlvif); - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); } out: @@ -5566,7 +5535,6 @@ static void wl12xx_op_channel_switch(struct ieee80211_hw *hw, } out_sleep: - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: @@ -5645,7 +5613,6 @@ static void wlcore_op_channel_switch_beacon(struct ieee80211_hw *hw, set_bit(WLVIF_FLAG_CS_PROGRESS, &wlvif->flags); out_sleep: - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: mutex_unlock(&wl->mutex); @@ -5699,7 +5666,6 @@ static int wlcore_op_remain_on_channel(struct ieee80211_hw *hw, ieee80211_queue_delayed_work(hw, &wl->roc_complete_work, msecs_to_jiffies(duration)); out_sleep: - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: mutex_unlock(&wl->mutex); @@ -5748,7 +5714,6 @@ static int wlcore_roc_completed(struct wl1271 *wl) ret = __wlcore_roc_completed(wl); - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: mutex_unlock(&wl->mutex); @@ -5839,7 +5804,6 @@ static void wlcore_op_sta_statistics(struct ieee80211_hw *hw, sinfo->signal = rssi_dbm; out_sleep: - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: diff --git a/drivers/net/wireless/ti/wlcore/scan.c b/drivers/net/wireless/ti/wlcore/scan.c index b414305acc32..f6dc54c1dbad 100644 --- a/drivers/net/wireless/ti/wlcore/scan.c +++ b/drivers/net/wireless/ti/wlcore/scan.c @@ -69,7 +69,6 @@ void wl1271_scan_complete_work(struct work_struct *work) wlcore_cmd_regdomain_config_locked(wl); - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); ieee80211_scan_completed(wl->hw, &info); diff --git a/drivers/net/wireless/ti/wlcore/sysfs.c b/drivers/net/wireless/ti/wlcore/sysfs.c index 65ca5dc569a0..5ab6c1683675 100644 --- a/drivers/net/wireless/ti/wlcore/sysfs.c +++ b/drivers/net/wireless/ti/wlcore/sysfs.c @@ -58,7 +58,6 @@ static ssize_t bt_coex_state_store(struct device *dev, goto out; wl1271_acx_sg_enable(wl, wl->sg_enabled); - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: diff --git a/drivers/net/wireless/ti/wlcore/testmode.c b/drivers/net/wireless/ti/wlcore/testmode.c index fc8ea58bc165..7c0cb1b7fef0 100644 --- a/drivers/net/wireless/ti/wlcore/testmode.c +++ b/drivers/net/wireless/ti/wlcore/testmode.c @@ -127,7 +127,6 @@ static int wl1271_tm_cmd_test(struct wl1271 *wl, struct nlattr *tb[]) } out_sleep: - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: mutex_unlock(&wl->mutex); @@ -192,7 +191,6 @@ static int wl1271_tm_cmd_interrogate(struct wl1271 *wl, struct nlattr *tb[]) out_free: kfree(cmd); out_sleep: - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: mutex_unlock(&wl->mutex); diff --git a/drivers/net/wireless/ti/wlcore/tx.c b/drivers/net/wireless/ti/wlcore/tx.c index 464587d16ab2..f76087be2f75 100644 --- a/drivers/net/wireless/ti/wlcore/tx.c +++ b/drivers/net/wireless/ti/wlcore/tx.c @@ -863,7 +863,6 @@ void wl1271_tx_work(struct work_struct *work) goto out; } - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: mutex_unlock(&wl->mutex); diff --git a/drivers/net/wireless/ti/wlcore/vendor_cmd.c b/drivers/net/wireless/ti/wlcore/vendor_cmd.c index e4269e2b0098..5bb9eb300f97 100644 --- a/drivers/net/wireless/ti/wlcore/vendor_cmd.c +++ b/drivers/net/wireless/ti/wlcore/vendor_cmd.c @@ -60,7 +60,6 @@ wlcore_vendor_cmd_smart_config_start(struct wiphy *wiphy, ret = wlcore_smart_config_start(wl, nla_get_u32(tb[WLCORE_VENDOR_ATTR_GROUP_ID])); - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: mutex_unlock(&wl->mutex); @@ -92,7 +91,6 @@ wlcore_vendor_cmd_smart_config_stop(struct wiphy *wiphy, ret = wlcore_smart_config_stop(wl); - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: mutex_unlock(&wl->mutex); @@ -140,7 +138,6 @@ wlcore_vendor_cmd_smart_config_set_group_key(struct wiphy *wiphy, nla_len(tb[WLCORE_VENDOR_ATTR_GROUP_KEY]), nla_data(tb[WLCORE_VENDOR_ATTR_GROUP_KEY])); - pm_runtime_mark_last_busy(wl->dev); pm_runtime_put_autosuspend(wl->dev); out: mutex_unlock(&wl->mutex); From db82ddeaf42b93799a52df347284062893ea2ad6 Mon Sep 17 00:00:00 2001 From: Benjamin Berg Date: Mon, 27 Oct 2025 14:22:14 +0200 Subject: [PATCH 0384/1381] wifi: mac80211: add RX flag to report radiotap VHT information mac80211 already reports some basic information in the radiotap header with the known fields declared by the driver. However, drivers may want to report more accurate information and in that case the full VHT radiotap structure needs to be provided. Add a new RX_FLAG_RADIOTAP_VHT which is set when the VHT information should be pulled from the skb. Update the code to fill in the VHT fields to only do so when requested by the driver or if the information has not yet been set. This way the driver can fully control the information if it chooses so. Signed-off-by: Benjamin Berg Reviewed-by: Johannes Berg Signed-off-by: Miri Korenblit Link: https://patch.msgid.link/20251027142118.0bad1c307a21.I2cf285c20a822698039603f2af00ed9c548f2ee0@changeid Signed-off-by: Johannes Berg --- include/net/ieee80211_radiotap.h | 20 +++++- include/net/mac80211.h | 2 + net/mac80211/rx.c | 104 ++++++++++++++++++++----------- 3 files changed, 89 insertions(+), 37 deletions(-) diff --git a/include/net/ieee80211_radiotap.h b/include/net/ieee80211_radiotap.h index 813e163ce27c..c60867e7e43c 100644 --- a/include/net/ieee80211_radiotap.h +++ b/include/net/ieee80211_radiotap.h @@ -1,6 +1,6 @@ /* * Copyright (c) 2017 Intel Deutschland GmbH - * Copyright (c) 2018-2019, 2021-2022 Intel Corporation + * Copyright (c) 2018-2019, 2021-2022, 2025 Intel Corporation * * Permission to use, copy, modify, and/or distribute this software for any * purpose with or without fee is hereby granted, provided that the above @@ -202,6 +202,24 @@ enum ieee80211_radiotap_vht_coding { IEEE80211_RADIOTAP_CODING_LDPC_USER3 = 0x08, }; +enum ieee80211_radiotap_vht_bandwidth { + /* Note: more values are defined but can't really be used */ + IEEE80211_RADIOTAP_VHT_BW_20 = 0, + IEEE80211_RADIOTAP_VHT_BW_40 = 1, + IEEE80211_RADIOTAP_VHT_BW_80 = 4, + IEEE80211_RADIOTAP_VHT_BW_160 = 11, +}; + +struct ieee80211_radiotap_vht { + __le16 known; + u8 flags; + u8 bandwidth; + u8 mcs_nss[4]; + u8 coding; + u8 group_id; + __le16 partial_aid; +} __packed; + /* for IEEE80211_RADIOTAP_TIMESTAMP */ enum ieee80211_radiotap_timestamp_unit_spos { IEEE80211_RADIOTAP_TIMESTAMP_UNIT_MASK = 0x000F, diff --git a/include/net/mac80211.h b/include/net/mac80211.h index a55085cf4ec4..c326243e1f01 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -1529,6 +1529,7 @@ ieee80211_tx_info_clear_status(struct ieee80211_tx_info *info) * known the frame shouldn't be reported. * @RX_FLAG_8023: the frame has an 802.3 header (decap offload performed by * hardware or driver) + * @RX_FLAG_RADIOTAP_VHT: VHT radiotap data is present */ enum mac80211_rx_flags { RX_FLAG_MMIC_ERROR = BIT(0), @@ -1564,6 +1565,7 @@ enum mac80211_rx_flags { RX_FLAG_RADIOTAP_LSIG = BIT(28), RX_FLAG_NO_PSDU = BIT(29), RX_FLAG_8023 = BIT(30), + RX_FLAG_RADIOTAP_VHT = BIT(31), }; /** diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c index 4641a2a80856..b59aeed340b3 100644 --- a/net/mac80211/rx.c +++ b/net/mac80211/rx.c @@ -59,7 +59,8 @@ static struct sk_buff *ieee80211_clean_skb(struct sk_buff *skb, status->flag &= ~(RX_FLAG_RADIOTAP_TLV_AT_END | RX_FLAG_RADIOTAP_LSIG | RX_FLAG_RADIOTAP_HE_MU | - RX_FLAG_RADIOTAP_HE); + RX_FLAG_RADIOTAP_HE | + RX_FLAG_RADIOTAP_VHT); hdr = (void *)skb->data; fc = hdr->frame_control; @@ -151,8 +152,10 @@ ieee80211_rx_radiotap_hdrlen(struct ieee80211_local *local, } if (status->encoding == RX_ENC_VHT) { + /* Included even if RX_FLAG_RADIOTAP_VHT is not set */ len = ALIGN(len, 2); len += 12; + BUILD_BUG_ON(sizeof(struct ieee80211_radiotap_vht) != 12); } if (local->hw.radiotap_timestamp.units_pos >= 0) { @@ -195,6 +198,9 @@ ieee80211_rx_radiotap_hdrlen(struct ieee80211_local *local, * The position to look at depends on the existence (or non- * existence) of other elements, so take that into account... */ + if (status->flag & RX_FLAG_RADIOTAP_VHT) + tlv_offset += + sizeof(struct ieee80211_radiotap_vht); if (status->flag & RX_FLAG_RADIOTAP_HE) tlv_offset += sizeof(struct ieee80211_radiotap_he); @@ -319,10 +325,17 @@ ieee80211_add_rx_radiotap_header(struct ieee80211_local *local, u32 tlvs_len = 0; int mpdulen, chain; unsigned long chains = status->chains; + struct ieee80211_radiotap_vht vht = {}; struct ieee80211_radiotap_he he = {}; struct ieee80211_radiotap_he_mu he_mu = {}; struct ieee80211_radiotap_lsig lsig = {}; + if (status->flag & RX_FLAG_RADIOTAP_VHT) { + vht = *(struct ieee80211_radiotap_vht *)skb->data; + skb_pull(skb, sizeof(vht)); + WARN_ON_ONCE(status->encoding != RX_ENC_VHT); + } + if (status->flag & RX_FLAG_RADIOTAP_HE) { he = *(struct ieee80211_radiotap_he *)skb->data; skb_pull(skb, sizeof(he)); @@ -530,45 +543,61 @@ ieee80211_add_rx_radiotap_header(struct ieee80211_local *local, } if (status->encoding == RX_ENC_VHT) { - u16 known = local->hw.radiotap_vht_details; + u16 fill = local->hw.radiotap_vht_details; - rthdr->it_present |= cpu_to_le32(BIT(IEEE80211_RADIOTAP_VHT)); - put_unaligned_le16(known, pos); - pos += 2; - /* flags */ - if (status->enc_flags & RX_ENC_FLAG_SHORT_GI) - *pos |= IEEE80211_RADIOTAP_VHT_FLAG_SGI; + /* Leave driver filled fields alone */ + fill &= ~le16_to_cpu(vht.known); + vht.known |= cpu_to_le16(fill); + + if (fill & IEEE80211_RADIOTAP_VHT_KNOWN_GI && + status->enc_flags & RX_ENC_FLAG_SHORT_GI) + vht.flags |= IEEE80211_RADIOTAP_VHT_FLAG_SGI; /* in VHT, STBC is binary */ - if (status->enc_flags & RX_ENC_FLAG_STBC_MASK) - *pos |= IEEE80211_RADIOTAP_VHT_FLAG_STBC; - if (status->enc_flags & RX_ENC_FLAG_BF) + if (fill & IEEE80211_RADIOTAP_VHT_KNOWN_STBC && + status->enc_flags & RX_ENC_FLAG_STBC_MASK) + vht.flags |= IEEE80211_RADIOTAP_VHT_FLAG_STBC; + if (fill & IEEE80211_RADIOTAP_VHT_KNOWN_BEAMFORMED && + status->enc_flags & RX_ENC_FLAG_BF) *pos |= IEEE80211_RADIOTAP_VHT_FLAG_BEAMFORMED; - pos++; - /* bandwidth */ - switch (status->bw) { - case RATE_INFO_BW_80: - *pos++ = 4; - break; - case RATE_INFO_BW_160: - *pos++ = 11; - break; - case RATE_INFO_BW_40: - *pos++ = 1; - break; - default: - *pos++ = 0; + + if (fill & IEEE80211_RADIOTAP_VHT_KNOWN_BANDWIDTH) { + switch (status->bw) { + case RATE_INFO_BW_40: + vht.bandwidth = IEEE80211_RADIOTAP_VHT_BW_40; + break; + case RATE_INFO_BW_80: + vht.bandwidth = IEEE80211_RADIOTAP_VHT_BW_80; + break; + case RATE_INFO_BW_160: + vht.bandwidth = IEEE80211_RADIOTAP_VHT_BW_160; + break; + default: + vht.bandwidth = IEEE80211_RADIOTAP_VHT_BW_20; + break; + } } - /* MCS/NSS */ - *pos = (status->rate_idx << 4) | status->nss; - pos += 4; - /* coding field */ - if (status->enc_flags & RX_ENC_FLAG_LDPC) - *pos |= IEEE80211_RADIOTAP_CODING_LDPC_USER0; - pos++; - /* group ID */ - pos++; - /* partial_aid */ - pos += 2; + + /* + * If the driver filled in mcs_nss[0], then do not touch it. + * + * Otherwise, put some information about MCS/NSS into the + * user 0 field. Note that this is not technically correct for + * an MU frame as we might have decoded a different user. + */ + if (!vht.mcs_nss[0]) { + vht.mcs_nss[0] = (status->rate_idx << 4) | status->nss; + + /* coding field */ + if (status->enc_flags & RX_ENC_FLAG_LDPC) + vht.coding |= IEEE80211_RADIOTAP_CODING_LDPC_USER0; + } + + /* ensure 2 byte alignment */ + while ((pos - (u8 *)rthdr) & 1) + pos++; + rthdr->it_present |= cpu_to_le32(BIT(IEEE80211_RADIOTAP_VHT)); + memcpy(pos, &vht, sizeof(vht)); + pos += sizeof(vht); } if (local->hw.radiotap_timestamp.units_pos >= 0) { @@ -834,6 +863,9 @@ ieee80211_rx_monitor(struct ieee80211_local *local, struct sk_buff *origskb, return NULL; } + if (status->flag & RX_FLAG_RADIOTAP_VHT) + rtap_space += sizeof(struct ieee80211_radiotap_vht); + if (status->flag & RX_FLAG_RADIOTAP_HE) rtap_space += sizeof(struct ieee80211_radiotap_he); From ec81b33b23bed73d7f407dc2079867a3f271aba0 Mon Sep 17 00:00:00 2001 From: Rosen Penev Date: Mon, 27 Oct 2025 11:06:38 -0700 Subject: [PATCH 0385/1381] wifi: rt2x00: add nvmem eeprom support Some embedded platforms have eeproms located in flash. Add nvmem support to handle this. Support is added for PCI and SOC backends. Signed-off-by: Rosen Penev Link: https://patch.msgid.link/20251027180639.3797-1-rosenp@gmail.com Signed-off-by: Johannes Berg --- .../net/wireless/ralink/rt2x00/rt2800lib.c | 31 +++++++++++++++++++ .../net/wireless/ralink/rt2x00/rt2800lib.h | 2 ++ .../net/wireless/ralink/rt2x00/rt2800pci.c | 3 ++ .../net/wireless/ralink/rt2x00/rt2800soc.c | 6 +++- 4 files changed, 41 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/ralink/rt2x00/rt2800lib.c b/drivers/net/wireless/ralink/rt2x00/rt2800lib.c index f07152fa3725..65d0f805459c 100644 --- a/drivers/net/wireless/ralink/rt2x00/rt2800lib.c +++ b/drivers/net/wireless/ralink/rt2x00/rt2800lib.c @@ -24,6 +24,7 @@ #include #include #include +#include #include #include "rt2x00.h" @@ -10962,6 +10963,36 @@ int rt2800_read_eeprom_efuse(struct rt2x00_dev *rt2x00dev) } EXPORT_SYMBOL_GPL(rt2800_read_eeprom_efuse); +int rt2800_read_eeprom_nvmem(struct rt2x00_dev *rt2x00dev) +{ + struct device_node *np = rt2x00dev->dev->of_node; + unsigned int len = rt2x00dev->ops->eeprom_size; + struct nvmem_cell *cell; + const void *data; + size_t retlen; + + cell = of_nvmem_cell_get(np, "eeprom"); + if (IS_ERR(cell)) + return PTR_ERR(cell); + + data = nvmem_cell_read(cell, &retlen); + nvmem_cell_put(cell); + + if (IS_ERR(data)) + return PTR_ERR(data); + + if (retlen != len) { + dev_err(rt2x00dev->dev, "invalid eeprom size, required: 0x%04x\n", len); + kfree(data); + return -EINVAL; + } + + memcpy(rt2x00dev->eeprom, data, len); + kfree(data); + return 0; +} +EXPORT_SYMBOL_GPL(rt2800_read_eeprom_nvmem); + static u8 rt2800_get_txmixer_gain_24g(struct rt2x00_dev *rt2x00dev) { u16 word; diff --git a/drivers/net/wireless/ralink/rt2x00/rt2800lib.h b/drivers/net/wireless/ralink/rt2x00/rt2800lib.h index 620a3d9872ce..a3c3a751f57e 100644 --- a/drivers/net/wireless/ralink/rt2x00/rt2800lib.h +++ b/drivers/net/wireless/ralink/rt2x00/rt2800lib.h @@ -248,6 +248,8 @@ void rt2800_disable_radio(struct rt2x00_dev *rt2x00dev); int rt2800_efuse_detect(struct rt2x00_dev *rt2x00dev); int rt2800_read_eeprom_efuse(struct rt2x00_dev *rt2x00dev); +int rt2800_read_eeprom_nvmem(struct rt2x00_dev *rt2x00dev); + int rt2800_probe_hw(struct rt2x00_dev *rt2x00dev); void rt2800_get_key_seq(struct ieee80211_hw *hw, diff --git a/drivers/net/wireless/ralink/rt2x00/rt2800pci.c b/drivers/net/wireless/ralink/rt2x00/rt2800pci.c index 14c45aba836f..4fa14bb573ad 100644 --- a/drivers/net/wireless/ralink/rt2x00/rt2800pci.c +++ b/drivers/net/wireless/ralink/rt2x00/rt2800pci.c @@ -278,6 +278,9 @@ static int rt2800pci_read_eeprom(struct rt2x00_dev *rt2x00dev) { int retval; + if (!rt2800_read_eeprom_nvmem(rt2x00dev)) + return 0; + if (rt2800pci_efuse_detect(rt2x00dev)) retval = rt2800pci_read_eeprom_efuse(rt2x00dev); else diff --git a/drivers/net/wireless/ralink/rt2x00/rt2800soc.c b/drivers/net/wireless/ralink/rt2x00/rt2800soc.c index 8f510a84e7f1..5c29201b34c8 100644 --- a/drivers/net/wireless/ralink/rt2x00/rt2800soc.c +++ b/drivers/net/wireless/ralink/rt2x00/rt2800soc.c @@ -92,8 +92,12 @@ static int rt2800soc_set_device_state(struct rt2x00_dev *rt2x00dev, static int rt2800soc_read_eeprom(struct rt2x00_dev *rt2x00dev) { - void __iomem *base_addr = ioremap(0x1F040000, EEPROM_SIZE); + void __iomem *base_addr; + if (!rt2800_read_eeprom_nvmem(rt2x00dev)) + return 0; + + base_addr = ioremap(0x1F040000, EEPROM_SIZE); if (!base_addr) return -ENOMEM; From 508dfc1f2ccdc480893332aaeb5e8d076769f7c3 Mon Sep 17 00:00:00 2001 From: Thomas Wu Date: Tue, 28 Oct 2025 10:04:42 +0530 Subject: [PATCH 0386/1381] wifi: mac80211: Allow HT Action frame processing on 6 GHz when HE is supported Management frames on 6 GHz do not include HT Capabilities, causing HT Action frames to be dropped in ieee80211_rx_h_action(). The current logic checks only ht_cap.ht_supported, which fails for 6 GHz radios that support only HE and EHT. Update the condition to also allow HT Action frame processing when he_cap.has_he is true. This enables support for HE dynamic SM power save as defined in IEEE Std 802.11ax-2021, section 26.14.4. Signed-off-by: Thomas Wu Signed-off-by: Aaradhana Sahu Link: https://patch.msgid.link/20251028043442.523647-1-aaradhana.sahu@oss.qualcomm.com Signed-off-by: Johannes Berg --- net/mac80211/rx.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c index b59aeed340b3..80067ed1da2f 100644 --- a/net/mac80211/rx.c +++ b/net/mac80211/rx.c @@ -3602,8 +3602,11 @@ ieee80211_rx_h_action(struct ieee80211_rx_data *rx) switch (mgmt->u.action.category) { case WLAN_CATEGORY_HT: - /* reject HT action frames from stations not supporting HT */ - if (!rx->link_sta->pub->ht_cap.ht_supported) + /* reject HT action frames from stations not supporting HT + * or not HE Capable + */ + if (!rx->link_sta->pub->ht_cap.ht_supported && + !rx->link_sta->pub->he_cap.has_he) goto invalid; if (sdata->vif.type != NL80211_IFTYPE_STATION && From 6b3b6e59c4f8584806cbed63b1593fd56d54cb71 Mon Sep 17 00:00:00 2001 From: Steffen Klassert Date: Tue, 28 Oct 2025 07:29:09 +0100 Subject: [PATCH 0387/1381] pfkey: Deprecate pfkey The pfkey user configuration interface was replaced by the netlink user configuration interface more than a decade ago. In between all maintained IKE implementations moved to the netlink interface. So let config NET_KEY default to no in Kconfig. The pfkey code will be removed in a second step. Signed-off-by: Steffen Klassert Reviewed-by: Sabrina Dubroca Acked-by: Antony Antony Acked-by: Tobias Brunner Acked-by: Herbert Xu Acked-by: Tuomo Soini Acked-by: Paul Wouters --- net/key/af_key.c | 2 ++ net/xfrm/Kconfig | 11 +++++++---- 2 files changed, 9 insertions(+), 4 deletions(-) diff --git a/net/key/af_key.c b/net/key/af_key.c index 2ebde0352245..571200433aa9 100644 --- a/net/key/af_key.c +++ b/net/key/af_key.c @@ -3903,6 +3903,8 @@ static int __init ipsec_pfkey_init(void) { int err = proto_register(&key_proto, 0); + pr_warn_once("PFKEY is deprecated and scheduled to be removed in 2027, " + "please contact the netdev mailing list\n"); if (err != 0) goto out; diff --git a/net/xfrm/Kconfig b/net/xfrm/Kconfig index f0157702718f..4a62817a88f8 100644 --- a/net/xfrm/Kconfig +++ b/net/xfrm/Kconfig @@ -110,14 +110,17 @@ config XFRM_IPCOMP select CRYPTO_DEFLATE config NET_KEY - tristate "PF_KEY sockets" + tristate "PF_KEY sockets (deprecated)" select XFRM_ALGO help PF_KEYv2 socket family, compatible to KAME ones. - They are required if you are going to use IPsec tools ported - from KAME. - Say Y unless you know what you are doing. + The PF_KEYv2 socket interface is deprecated and + scheduled for removal. All maintained IKE daemons + no longer need PF_KEY sockets. Please use the netlink + interface (XFRM_USER) to configure IPsec. + + If unsure, say N. config NET_KEY_MIGRATE bool "PF_KEY MIGRATE" From 85708c5d5f5bd2a1e63a1fcfab6a1030cc195bfc Mon Sep 17 00:00:00 2001 From: Subbaraya Sundeep Date: Sat, 25 Oct 2025 16:02:37 +0530 Subject: [PATCH 0388/1381] octeontx2-af: Simplify context writing and reading to hardware Simplify NIX context reading and writing by using hardware maximum context size instead of using individual sizes of each context type. Signed-off-by: Subbaraya Sundeep Link: https://patch.msgid.link/1761388367-16579-2-git-send-email-sbhatta@marvell.com Signed-off-by: Paolo Abeni --- .../ethernet/marvell/octeontx2/af/rvu_nix.c | 44 +++++++++---------- .../marvell/octeontx2/af/rvu_struct.h | 25 ++++++++++- 2 files changed, 46 insertions(+), 23 deletions(-) diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c index 828316211b24..26dc0fbeafa6 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c @@ -1149,36 +1149,36 @@ static int rvu_nix_blk_aq_enq_inst(struct rvu *rvu, struct nix_hw *nix_hw, case NIX_AQ_INSTOP_WRITE: if (req->ctype == NIX_AQ_CTYPE_RQ) memcpy(mask, &req->rq_mask, - sizeof(struct nix_rq_ctx_s)); + NIX_MAX_CTX_SIZE); else if (req->ctype == NIX_AQ_CTYPE_SQ) memcpy(mask, &req->sq_mask, - sizeof(struct nix_sq_ctx_s)); + NIX_MAX_CTX_SIZE); else if (req->ctype == NIX_AQ_CTYPE_CQ) memcpy(mask, &req->cq_mask, - sizeof(struct nix_cq_ctx_s)); + NIX_MAX_CTX_SIZE); else if (req->ctype == NIX_AQ_CTYPE_RSS) memcpy(mask, &req->rss_mask, - sizeof(struct nix_rsse_s)); + NIX_MAX_CTX_SIZE); else if (req->ctype == NIX_AQ_CTYPE_MCE) memcpy(mask, &req->mce_mask, - sizeof(struct nix_rx_mce_s)); + NIX_MAX_CTX_SIZE); else if (req->ctype == NIX_AQ_CTYPE_BANDPROF) memcpy(mask, &req->prof_mask, - sizeof(struct nix_bandprof_s)); + NIX_MAX_CTX_SIZE); fallthrough; case NIX_AQ_INSTOP_INIT: if (req->ctype == NIX_AQ_CTYPE_RQ) - memcpy(ctx, &req->rq, sizeof(struct nix_rq_ctx_s)); + memcpy(ctx, &req->rq, NIX_MAX_CTX_SIZE); else if (req->ctype == NIX_AQ_CTYPE_SQ) - memcpy(ctx, &req->sq, sizeof(struct nix_sq_ctx_s)); + memcpy(ctx, &req->sq, NIX_MAX_CTX_SIZE); else if (req->ctype == NIX_AQ_CTYPE_CQ) - memcpy(ctx, &req->cq, sizeof(struct nix_cq_ctx_s)); + memcpy(ctx, &req->cq, NIX_MAX_CTX_SIZE); else if (req->ctype == NIX_AQ_CTYPE_RSS) - memcpy(ctx, &req->rss, sizeof(struct nix_rsse_s)); + memcpy(ctx, &req->rss, NIX_MAX_CTX_SIZE); else if (req->ctype == NIX_AQ_CTYPE_MCE) - memcpy(ctx, &req->mce, sizeof(struct nix_rx_mce_s)); + memcpy(ctx, &req->mce, NIX_MAX_CTX_SIZE); else if (req->ctype == NIX_AQ_CTYPE_BANDPROF) - memcpy(ctx, &req->prof, sizeof(struct nix_bandprof_s)); + memcpy(ctx, &req->prof, NIX_MAX_CTX_SIZE); break; case NIX_AQ_INSTOP_NOP: case NIX_AQ_INSTOP_READ: @@ -1243,22 +1243,22 @@ static int rvu_nix_blk_aq_enq_inst(struct rvu *rvu, struct nix_hw *nix_hw, if (req->op == NIX_AQ_INSTOP_READ) { if (req->ctype == NIX_AQ_CTYPE_RQ) memcpy(&rsp->rq, ctx, - sizeof(struct nix_rq_ctx_s)); + NIX_MAX_CTX_SIZE); else if (req->ctype == NIX_AQ_CTYPE_SQ) memcpy(&rsp->sq, ctx, - sizeof(struct nix_sq_ctx_s)); + NIX_MAX_CTX_SIZE); else if (req->ctype == NIX_AQ_CTYPE_CQ) memcpy(&rsp->cq, ctx, - sizeof(struct nix_cq_ctx_s)); + NIX_MAX_CTX_SIZE); else if (req->ctype == NIX_AQ_CTYPE_RSS) memcpy(&rsp->rss, ctx, - sizeof(struct nix_rsse_s)); + NIX_MAX_CTX_SIZE); else if (req->ctype == NIX_AQ_CTYPE_MCE) memcpy(&rsp->mce, ctx, - sizeof(struct nix_rx_mce_s)); + NIX_MAX_CTX_SIZE); else if (req->ctype == NIX_AQ_CTYPE_BANDPROF) memcpy(&rsp->prof, ctx, - sizeof(struct nix_bandprof_s)); + NIX_MAX_CTX_SIZE); } } @@ -1289,8 +1289,8 @@ static int rvu_nix_verify_aq_ctx(struct rvu *rvu, struct nix_hw *nix_hw, /* Make copy of original context & mask which are required * for resubmission */ - memcpy(&aq_req.cq_mask, &req->cq_mask, sizeof(struct nix_cq_ctx_s)); - memcpy(&aq_req.cq, &req->cq, sizeof(struct nix_cq_ctx_s)); + memcpy(&aq_req.cq_mask, &req->cq_mask, NIX_MAX_CTX_SIZE); + memcpy(&aq_req.cq, &req->cq, NIX_MAX_CTX_SIZE); /* exclude fields which HW can update */ aq_req.cq_mask.cq_err = 0; @@ -1309,7 +1309,7 @@ static int rvu_nix_verify_aq_ctx(struct rvu *rvu, struct nix_hw *nix_hw, * updated fields are masked out for request and response * comparison */ - for (word = 0; word < sizeof(struct nix_cq_ctx_s) / sizeof(u64); + for (word = 0; word < NIX_MAX_CTX_SIZE / sizeof(u64); word++) { *(u64 *)((u8 *)&aq_rsp.cq + word * 8) &= (*(u64 *)((u8 *)&aq_req.cq_mask + word * 8)); @@ -1317,7 +1317,7 @@ static int rvu_nix_verify_aq_ctx(struct rvu *rvu, struct nix_hw *nix_hw, (*(u64 *)((u8 *)&aq_req.cq_mask + word * 8)); } - if (memcmp(&aq_req.cq, &aq_rsp.cq, sizeof(struct nix_cq_ctx_s))) + if (memcmp(&aq_req.cq, &aq_rsp.cq, NIX_MAX_CTX_SIZE)) return NIX_AF_ERR_AQ_CTX_RETRY_WRITE; return 0; diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_struct.h b/drivers/net/ethernet/marvell/octeontx2/af/rvu_struct.h index 0596a3ac4c12..8d41cb8f85ef 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_struct.h +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_struct.h @@ -13,6 +13,8 @@ #define RVU_MULTI_BLK_VER 0x7ULL +#define NIX_MAX_CTX_SIZE 128 + /* RVU Block Address Enumeration */ enum rvu_block_addr_e { BLKADDR_RVUM = 0x0ULL, @@ -370,8 +372,12 @@ struct nix_cq_ctx_s { u64 qsize : 4; u64 cq_err_int : 8; u64 cq_err_int_ena : 8; + /* Ensure all context sizes are 128 bytes */ + u64 padding[12]; }; +static_assert(sizeof(struct nix_cq_ctx_s) == NIX_MAX_CTX_SIZE); + /* CN10K NIX Receive queue context structure */ struct nix_cn10k_rq_ctx_s { u64 ena : 1; @@ -460,6 +466,8 @@ struct nix_cn10k_rq_ctx_s { u64 rsvd_1023_960; /* W15 */ }; +static_assert(sizeof(struct nix_cn10k_rq_ctx_s) == NIX_MAX_CTX_SIZE); + /* CN10K NIX Send queue context structure */ struct nix_cn10k_sq_ctx_s { u64 ena : 1; @@ -523,6 +531,8 @@ struct nix_cn10k_sq_ctx_s { u64 rsvd_1023_1008 : 16; }; +static_assert(sizeof(struct nix_cn10k_sq_ctx_s) == NIX_MAX_CTX_SIZE); + /* NIX Receive queue context structure */ struct nix_rq_ctx_s { u64 ena : 1; @@ -594,6 +604,8 @@ struct nix_rq_ctx_s { u64 rsvd_1023_960; /* W15 */ }; +static_assert(sizeof(struct nix_rq_ctx_s) == NIX_MAX_CTX_SIZE); + /* NIX sqe sizes */ enum nix_maxsqesz { NIX_MAXSQESZ_W16 = 0x0, @@ -668,13 +680,18 @@ struct nix_sq_ctx_s { u64 rsvd_1023_1008 : 16; }; +static_assert(sizeof(struct nix_sq_ctx_s) == NIX_MAX_CTX_SIZE); + /* NIX Receive side scaling entry structure*/ struct nix_rsse_s { uint32_t rq : 20; uint32_t reserved_20_31 : 12; - + /* Ensure all context sizes are minimum 128 bytes */ + u64 padding[15]; }; +static_assert(sizeof(struct nix_rsse_s) == NIX_MAX_CTX_SIZE); + /* NIX receive multicast/mirror entry structure */ struct nix_rx_mce_s { uint64_t op : 2; @@ -684,8 +701,12 @@ struct nix_rx_mce_s { uint64_t rsvd_31_24 : 8; uint64_t pf_func : 16; uint64_t next : 16; + /* Ensure all context sizes are minimum 128 bytes */ + u64 padding[15]; }; +static_assert(sizeof(struct nix_rx_mce_s) == NIX_MAX_CTX_SIZE); + enum nix_band_prof_layers { BAND_PROF_LEAF_LAYER = 0, BAND_PROF_INVAL_LAYER = 1, @@ -769,6 +790,8 @@ struct nix_bandprof_s { uint64_t reserved_1008_1023 : 16; }; +static_assert(sizeof(struct nix_bandprof_s) == NIX_MAX_CTX_SIZE); + enum nix_lsoalg { NIX_LSOALG_NOP, NIX_LSOALG_ADD_SEGNUM, From b5dcdde074d5560bc513007e2104f5e0a33e1726 Mon Sep 17 00:00:00 2001 From: Subbaraya Sundeep Date: Sat, 25 Oct 2025 16:02:38 +0530 Subject: [PATCH 0389/1381] octeontx2-af: Add cn20k NIX block contexts New CN20K silicon has NIX hardware context structures different from previous silicons. Add NIX send and completion queue context definitions for cn20k. Extend NIX context handling support to cn20k. Signed-off-by: Subbaraya Sundeep Link: https://patch.msgid.link/1761388367-16579-3-git-send-email-sbhatta@marvell.com Signed-off-by: Paolo Abeni --- .../ethernet/marvell/octeontx2/af/Makefile | 2 +- .../ethernet/marvell/octeontx2/af/cn20k/nix.c | 20 ++ .../marvell/octeontx2/af/cn20k/struct.h | 205 ++++++++++++++++++ .../net/ethernet/marvell/octeontx2/af/mbox.h | 38 ++++ .../net/ethernet/marvell/octeontx2/af/rvu.h | 15 +- .../ethernet/marvell/octeontx2/af/rvu_nix.c | 10 +- 6 files changed, 286 insertions(+), 4 deletions(-) create mode 100644 drivers/net/ethernet/marvell/octeontx2/af/cn20k/nix.c diff --git a/drivers/net/ethernet/marvell/octeontx2/af/Makefile b/drivers/net/ethernet/marvell/octeontx2/af/Makefile index 532813d8d028..cb77b978eda5 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/Makefile +++ b/drivers/net/ethernet/marvell/octeontx2/af/Makefile @@ -12,4 +12,4 @@ rvu_af-y := cgx.o rvu.o rvu_cgx.o rvu_npa.o rvu_nix.o \ rvu_reg.o rvu_npc.o rvu_debugfs.o ptp.o rvu_npc_fs.o \ rvu_cpt.o rvu_devlink.o rpm.o rvu_cn10k.o rvu_switch.o \ rvu_sdp.o rvu_npc_hash.o mcs.o mcs_rvu_if.o mcs_cnf10kb.o \ - rvu_rep.o cn20k/mbox_init.o + rvu_rep.o cn20k/mbox_init.o cn20k/nix.o diff --git a/drivers/net/ethernet/marvell/octeontx2/af/cn20k/nix.c b/drivers/net/ethernet/marvell/octeontx2/af/cn20k/nix.c new file mode 100644 index 000000000000..aa2016fd1bba --- /dev/null +++ b/drivers/net/ethernet/marvell/octeontx2/af/cn20k/nix.c @@ -0,0 +1,20 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Marvell RVU Admin Function driver + * + * Copyright (C) 2024 Marvell. + * + */ + +#include +#include + +#include "struct.h" +#include "../rvu.h" + +int rvu_mbox_handler_nix_cn20k_aq_enq(struct rvu *rvu, + struct nix_cn20k_aq_enq_req *req, + struct nix_cn20k_aq_enq_rsp *rsp) +{ + return rvu_nix_aq_enq_inst(rvu, (struct nix_aq_enq_req *)req, + (struct nix_aq_enq_rsp *)rsp); +} diff --git a/drivers/net/ethernet/marvell/octeontx2/af/cn20k/struct.h b/drivers/net/ethernet/marvell/octeontx2/af/cn20k/struct.h index 76ce3ec6da9c..ff8f0d1809c8 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/cn20k/struct.h +++ b/drivers/net/ethernet/marvell/octeontx2/af/cn20k/struct.h @@ -8,6 +8,8 @@ #ifndef STRUCT_H #define STRUCT_H +#define NIX_MAX_CTX_SIZE 128 + /* * CN20k RVU PF MBOX Interrupt Vector Enumeration * @@ -37,4 +39,207 @@ enum rvu_af_cn20k_int_vec_e { RVU_AF_CN20K_INT_VEC_PFAF1_MBOX1 = 0x9, RVU_AF_CN20K_INT_VEC_CNT = 0xa, }; + +struct nix_cn20k_sq_ctx_s { + u64 ena : 1; /* W0 */ + u64 qint_idx : 6; + u64 substream : 20; + u64 sdp_mcast : 1; + u64 cq : 20; + u64 sqe_way_mask : 16; + u64 smq : 11; /* W1 */ + u64 cq_ena : 1; + u64 xoff : 1; + u64 sso_ena : 1; + u64 smq_rr_weight : 14; + u64 default_chan : 12; + u64 sqb_count : 16; + u64 reserved_120_120 : 1; + u64 smq_rr_count_lb : 7; + u64 smq_rr_count_ub : 25; /* W2 */ + u64 sqb_aura : 20; + u64 sq_int : 8; + u64 sq_int_ena : 8; + u64 sqe_stype : 2; + u64 reserved_191_191 : 1; + u64 max_sqe_size : 2; /* W3 */ + u64 cq_limit : 8; + u64 lmt_dis : 1; + u64 mnq_dis : 1; + u64 smq_next_sq : 20; + u64 smq_lso_segnum : 8; + u64 tail_offset : 6; + u64 smenq_offset : 6; + u64 head_offset : 6; + u64 smenq_next_sqb_vld : 1; + u64 smq_pend : 1; + u64 smq_next_sq_vld : 1; + u64 reserved_253_255 : 3; + u64 next_sqb : 64; /* W4 */ + u64 tail_sqb : 64; /* W5 */ + u64 smenq_sqb : 64; /* W6 */ + u64 smenq_next_sqb : 64; /* W7 */ + u64 head_sqb : 64; /* W8 */ + u64 reserved_576_583 : 8; /* W9 */ + u64 vfi_lso_total : 18; + u64 vfi_lso_sizem1 : 3; + u64 vfi_lso_sb : 8; + u64 vfi_lso_mps : 14; + u64 vfi_lso_vlan0_ins_ena : 1; + u64 vfi_lso_vlan1_ins_ena : 1; + u64 vfi_lso_vld : 1; + u64 reserved_630_639 : 10; + u64 scm_lso_rem : 18; /* W10 */ + u64 reserved_658_703 : 46; + u64 octs : 48; /* W11 */ + u64 reserved_752_767 : 16; + u64 pkts : 48; /* W12 */ + u64 reserved_816_831 : 16; + u64 aged_drop_octs : 32; /* W13 */ + u64 aged_drop_pkts : 32; + u64 dropped_octs : 48; /* W14 */ + u64 reserved_944_959 : 16; + u64 dropped_pkts : 48; /* W15 */ + u64 reserved_1008_1023 : 16; +}; + +static_assert(sizeof(struct nix_cn20k_sq_ctx_s) == NIX_MAX_CTX_SIZE); + +struct nix_cn20k_cq_ctx_s { + u64 base : 64; /* W0 */ + u64 lbp_ena : 1; /* W1 */ + u64 lbpid_low : 3; + u64 bp_ena : 1; + u64 lbpid_med : 3; + u64 bpid : 9; + u64 lbpid_high : 3; + u64 qint_idx : 7; + u64 cq_err : 1; + u64 cint_idx : 7; + u64 avg_con : 9; + u64 wrptr : 20; + u64 tail : 20; /* W2 */ + u64 head : 20; + u64 avg_level : 8; + u64 update_time : 16; + u64 bp : 8; /* W3 */ + u64 drop : 8; + u64 drop_ena : 1; + u64 ena : 1; + u64 cpt_drop_err_en : 1; + u64 reserved_211_211 : 1; + u64 msh_dst : 11; + u64 msh_valid : 1; + u64 stash_thresh : 4; + u64 lbp_frac : 4; + u64 caching : 1; + u64 stashing : 1; + u64 reserved_234_235 : 2; + u64 qsize : 4; + u64 cq_err_int : 8; + u64 cq_err_int_ena : 8; + u64 bpid_ext : 2; /* W4 */ + u64 reserved_258_259 : 2; + u64 lbpid_ext : 2; + u64 reserved_262_319 : 58; + u64 reserved_320_383 : 64; /* W5 */ + u64 reserved_384_447 : 64; /* W6 */ + u64 reserved_448_511 : 64; /* W7 */ + u64 padding[8]; +}; + +static_assert(sizeof(struct nix_cn20k_sq_ctx_s) == NIX_MAX_CTX_SIZE); + +struct nix_cn20k_rq_ctx_s { + u64 ena : 1; + u64 sso_ena : 1; + u64 ipsech_ena : 1; + u64 ena_wqwd : 1; + u64 cq : 20; + u64 reserved_24_34 : 11; + u64 port_il4_dis : 1; + u64 port_ol4_dis : 1; + u64 lenerr_dis : 1; + u64 csum_il4_dis : 1; + u64 csum_ol4_dis : 1; + u64 len_il4_dis : 1; + u64 len_il3_dis : 1; + u64 len_ol4_dis : 1; + u64 len_ol3_dis : 1; + u64 wqe_aura : 20; + u64 spb_aura : 20; + u64 lpb_aura : 20; + u64 sso_grp : 10; + u64 sso_tt : 2; + u64 pb_caching : 2; + u64 wqe_caching : 1; + u64 xqe_drop_ena : 1; + u64 spb_drop_ena : 1; + u64 lpb_drop_ena : 1; + u64 pb_stashing : 1; + u64 ipsecd_drop_en : 1; + u64 chi_ena : 1; + u64 reserved_125_127 : 3; + u64 band_prof_id_l : 10; + u64 sso_fc_ena : 1; + u64 policer_ena : 1; + u64 spb_sizem1 : 6; + u64 wqe_skip : 2; + u64 spb_high_sizem1 : 3; + u64 spb_ena : 1; + u64 lpb_sizem1 : 12; + u64 first_skip : 7; + u64 reserved_171_171 : 1; + u64 later_skip : 6; + u64 xqe_imm_size : 6; + u64 band_prof_id_h : 4; + u64 reserved_188_189 : 2; + u64 xqe_imm_copy : 1; + u64 xqe_hdr_split : 1; + u64 xqe_drop : 8; + u64 xqe_pass : 8; + u64 wqe_pool_drop : 8; + u64 wqe_pool_pass : 8; + u64 spb_aura_drop : 8; + u64 spb_aura_pass : 8; + u64 spb_pool_drop : 8; + u64 spb_pool_pass : 8; + u64 lpb_aura_drop : 8; + u64 lpb_aura_pass : 8; + u64 lpb_pool_drop : 8; + u64 lpb_pool_pass : 8; + u64 reserved_288_291 : 4; + u64 rq_int : 8; + u64 rq_int_ena : 8; + u64 qint_idx : 7; + u64 reserved_315_319 : 5; + u64 ltag : 24; + u64 good_utag : 8; + u64 bad_utag : 8; + u64 flow_tagw : 6; + u64 ipsec_vwqe : 1; + u64 vwqe_ena : 1; + u64 vtime_wait : 8; + u64 max_vsize_exp : 4; + u64 vwqe_skip : 2; + u64 reserved_382_383 : 2; + u64 octs : 48; + u64 reserved_432_447 : 16; + u64 pkts : 48; + u64 reserved_496_511 : 16; + u64 drop_octs : 48; + u64 reserved_560_575 : 16; + u64 drop_pkts : 48; + u64 reserved_624_639 : 16; + u64 re_pkts : 48; + u64 reserved_688_703 : 16; + u64 reserved_704_767 : 64; + u64 reserved_768_831 : 64; + u64 reserved_832_895 : 64; + u64 reserved_896_959 : 64; + u64 reserved_960_1023 : 64; +}; + +static_assert(sizeof(struct nix_cn20k_rq_ctx_s) == NIX_MAX_CTX_SIZE); + #endif diff --git a/drivers/net/ethernet/marvell/octeontx2/af/mbox.h b/drivers/net/ethernet/marvell/octeontx2/af/mbox.h index 933073cd2280..01086c52e78d 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/mbox.h +++ b/drivers/net/ethernet/marvell/octeontx2/af/mbox.h @@ -336,6 +336,8 @@ M(NIX_MCAST_GRP_UPDATE, 0x802d, nix_mcast_grp_update, \ nix_mcast_grp_update_req, \ nix_mcast_grp_update_rsp) \ M(NIX_LF_STATS, 0x802e, nix_lf_stats, nix_stats_req, nix_stats_rsp) \ +M(NIX_CN20K_AQ_ENQ, 0x802f, nix_cn20k_aq_enq, nix_cn20k_aq_enq_req, \ + nix_cn20k_aq_enq_rsp) \ /* MCS mbox IDs (range 0xA000 - 0xBFFF) */ \ M(MCS_ALLOC_RESOURCES, 0xa000, mcs_alloc_resources, mcs_alloc_rsrc_req, \ mcs_alloc_rsrc_rsp) \ @@ -940,6 +942,42 @@ struct nix_lf_free_req { u64 flags; }; +/* CN20K NIX AQ enqueue msg */ +struct nix_cn20k_aq_enq_req { + struct mbox_msghdr hdr; + u32 qidx; + u8 ctype; + u8 op; + union { + struct nix_cn20k_rq_ctx_s rq; + struct nix_cn20k_sq_ctx_s sq; + struct nix_cn20k_cq_ctx_s cq; + struct nix_rsse_s rss; + struct nix_rx_mce_s mce; + struct nix_bandprof_s prof; + }; + union { + struct nix_cn20k_rq_ctx_s rq_mask; + struct nix_cn20k_sq_ctx_s sq_mask; + struct nix_cn20k_cq_ctx_s cq_mask; + struct nix_rsse_s rss_mask; + struct nix_rx_mce_s mce_mask; + struct nix_bandprof_s prof_mask; + }; +}; + +struct nix_cn20k_aq_enq_rsp { + struct mbox_msghdr hdr; + union { + struct nix_cn20k_rq_ctx_s rq; + struct nix_cn20k_sq_ctx_s sq; + struct nix_cn20k_cq_ctx_s cq; + struct nix_rsse_s rss; + struct nix_rx_mce_s mce; + struct nix_bandprof_s prof; + }; +}; + /* CN10K NIX AQ enqueue msg */ struct nix_cn10k_aq_enq_req { struct mbox_msghdr hdr; diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu.h b/drivers/net/ethernet/marvell/octeontx2/af/rvu.h index b58283341923..e85dac2c806d 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu.h +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu.h @@ -498,6 +498,14 @@ struct channel_fwdata { u8 reserved[RVU_CHANL_INFO_RESERVED]; }; +struct altaf_intr_notify { + unsigned long flr_pf_bmap[2]; + unsigned long flr_vf_bmap[2]; + unsigned long gint_paddr; + unsigned long gint_iova_addr; + unsigned long reserved[6]; +}; + struct rvu_fwdata { #define RVU_FWDATA_HEADER_MAGIC 0xCFDA /* Custom Firmware Data*/ #define RVU_FWDATA_VERSION 0x0001 @@ -517,7 +525,8 @@ struct rvu_fwdata { u32 ptp_ext_clk_rate; u32 ptp_ext_tstamp; struct channel_fwdata channel_data; -#define FWDATA_RESERVED_MEM 958 + struct altaf_intr_notify altaf_intr_info; +#define FWDATA_RESERVED_MEM 946 u64 reserved[FWDATA_RESERVED_MEM]; #define CGX_MAX 9 #define CGX_LMACS_MAX 4 @@ -648,6 +657,7 @@ struct rvu { struct mutex mbox_lock; /* Serialize mbox up and down msgs */ u16 rep_pcifunc; + bool altaf_ready; int rep_cnt; u16 *rep2pfvf_map; u8 rep_mode; @@ -1032,6 +1042,9 @@ void rvu_nix_flr_free_bpids(struct rvu *rvu, u16 pcifunc); int rvu_alloc_cint_qint_mem(struct rvu *rvu, struct rvu_pfvf *pfvf, int blkaddr, int nixlf); void rvu_block_bcast_xon(struct rvu *rvu, int blkaddr); +int rvu_nix_aq_enq_inst(struct rvu *rvu, struct nix_aq_enq_req *req, + struct nix_aq_enq_rsp *rsp); + /* NPC APIs */ void rvu_npc_freemem(struct rvu *rvu); int rvu_npc_get_pkind(struct rvu *rvu, u16 pf); diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c index 26dc0fbeafa6..d156d124f079 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c @@ -1019,6 +1019,12 @@ static void nix_get_aq_req_smq(struct rvu *rvu, struct nix_aq_enq_req *req, { struct nix_cn10k_aq_enq_req *aq_req; + if (is_cn20k(rvu->pdev)) { + *smq = ((struct nix_cn20k_aq_enq_req *)req)->sq.smq; + *smq_mask = ((struct nix_cn20k_aq_enq_req *)req)->sq_mask.smq; + return; + } + if (!is_rvu_otx2(rvu)) { aq_req = (struct nix_cn10k_aq_enq_req *)req; *smq = aq_req->sq.smq; @@ -1323,8 +1329,8 @@ static int rvu_nix_verify_aq_ctx(struct rvu *rvu, struct nix_hw *nix_hw, return 0; } -static int rvu_nix_aq_enq_inst(struct rvu *rvu, struct nix_aq_enq_req *req, - struct nix_aq_enq_rsp *rsp) +int rvu_nix_aq_enq_inst(struct rvu *rvu, struct nix_aq_enq_req *req, + struct nix_aq_enq_rsp *rsp) { struct nix_hw *nix_hw; int err, retries = 5; From 45229e9a9ab5edf089d4c16efdf690a06d11a9bf Mon Sep 17 00:00:00 2001 From: Subbaraya Sundeep Date: Sat, 25 Oct 2025 16:02:39 +0530 Subject: [PATCH 0390/1381] octeontx2-af: Extend debugfs support for cn20k NIX Extend debugfs to display CN20K NIX send, receive and completion queue contexts. Signed-off-by: Subbaraya Sundeep Link: https://patch.msgid.link/1761388367-16579-4-git-send-email-sbhatta@marvell.com Signed-off-by: Paolo Abeni --- .../ethernet/marvell/octeontx2/af/Makefile | 2 +- .../marvell/octeontx2/af/cn20k/debugfs.c | 134 ++++++++++++++++++ .../marvell/octeontx2/af/cn20k/debugfs.h | 24 ++++ .../marvell/octeontx2/af/rvu_debugfs.c | 17 +++ 4 files changed, 176 insertions(+), 1 deletion(-) create mode 100644 drivers/net/ethernet/marvell/octeontx2/af/cn20k/debugfs.c create mode 100644 drivers/net/ethernet/marvell/octeontx2/af/cn20k/debugfs.h diff --git a/drivers/net/ethernet/marvell/octeontx2/af/Makefile b/drivers/net/ethernet/marvell/octeontx2/af/Makefile index cb77b978eda5..57eeaa116116 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/Makefile +++ b/drivers/net/ethernet/marvell/octeontx2/af/Makefile @@ -12,4 +12,4 @@ rvu_af-y := cgx.o rvu.o rvu_cgx.o rvu_npa.o rvu_nix.o \ rvu_reg.o rvu_npc.o rvu_debugfs.o ptp.o rvu_npc_fs.o \ rvu_cpt.o rvu_devlink.o rpm.o rvu_cn10k.o rvu_switch.o \ rvu_sdp.o rvu_npc_hash.o mcs.o mcs_rvu_if.o mcs_cnf10kb.o \ - rvu_rep.o cn20k/mbox_init.o cn20k/nix.o + rvu_rep.o cn20k/mbox_init.o cn20k/nix.o cn20k/debugfs.o diff --git a/drivers/net/ethernet/marvell/octeontx2/af/cn20k/debugfs.c b/drivers/net/ethernet/marvell/octeontx2/af/cn20k/debugfs.c new file mode 100644 index 000000000000..50b1bd1d2c86 --- /dev/null +++ b/drivers/net/ethernet/marvell/octeontx2/af/cn20k/debugfs.c @@ -0,0 +1,134 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Marvell RVU Admin Function driver + * + * Copyright (C) 2024 Marvell. + * + */ + +#include +#include +#include +#include + +#include "struct.h" +#include "debugfs.h" + +void print_nix_cn20k_sq_ctx(struct seq_file *m, + struct nix_cn20k_sq_ctx_s *sq_ctx) +{ + seq_printf(m, "W0: ena \t\t\t%d\nW0: qint_idx \t\t\t%d\n", + sq_ctx->ena, sq_ctx->qint_idx); + seq_printf(m, "W0: substream \t\t\t0x%03x\nW0: sdp_mcast \t\t\t%d\n", + sq_ctx->substream, sq_ctx->sdp_mcast); + seq_printf(m, "W0: cq \t\t\t\t%d\nW0: sqe_way_mask \t\t%d\n\n", + sq_ctx->cq, sq_ctx->sqe_way_mask); + + seq_printf(m, "W1: smq \t\t\t%d\nW1: cq_ena \t\t\t%d\nW1: xoff\t\t\t%d\n", + sq_ctx->smq, sq_ctx->cq_ena, sq_ctx->xoff); + seq_printf(m, "W1: sso_ena \t\t\t%d\nW1: smq_rr_weight\t\t%d\n", + sq_ctx->sso_ena, sq_ctx->smq_rr_weight); + seq_printf(m, "W1: default_chan\t\t%d\nW1: sqb_count\t\t\t%d\n\n", + sq_ctx->default_chan, sq_ctx->sqb_count); + + seq_printf(m, "W1: smq_rr_count_lb \t\t%d\n", sq_ctx->smq_rr_count_lb); + seq_printf(m, "W2: smq_rr_count_ub \t\t%d\n", sq_ctx->smq_rr_count_ub); + seq_printf(m, "W2: sqb_aura \t\t\t%d\nW2: sq_int \t\t\t%d\n", + sq_ctx->sqb_aura, sq_ctx->sq_int); + seq_printf(m, "W2: sq_int_ena \t\t\t%d\nW2: sqe_stype \t\t\t%d\n", + sq_ctx->sq_int_ena, sq_ctx->sqe_stype); + + seq_printf(m, "W3: max_sqe_size\t\t%d\nW3: cq_limit\t\t\t%d\n", + sq_ctx->max_sqe_size, sq_ctx->cq_limit); + seq_printf(m, "W3: lmt_dis \t\t\t%d\nW3: mnq_dis \t\t\t%d\n", + sq_ctx->lmt_dis, sq_ctx->mnq_dis); + seq_printf(m, "W3: smq_next_sq\t\t\t%d\nW3: smq_lso_segnum\t\t%d\n", + sq_ctx->smq_next_sq, sq_ctx->smq_lso_segnum); + seq_printf(m, "W3: tail_offset \t\t%d\nW3: smenq_offset\t\t%d\n", + sq_ctx->tail_offset, sq_ctx->smenq_offset); + seq_printf(m, "W3: head_offset\t\t\t%d\nW3: smenq_next_sqb_vld\t\t%d\n\n", + sq_ctx->head_offset, sq_ctx->smenq_next_sqb_vld); + + seq_printf(m, "W3: smq_next_sq_vld\t\t%d\nW3: smq_pend\t\t\t%d\n", + sq_ctx->smq_next_sq_vld, sq_ctx->smq_pend); + seq_printf(m, "W4: next_sqb \t\t\t%llx\n\n", sq_ctx->next_sqb); + seq_printf(m, "W5: tail_sqb \t\t\t%llx\n\n", sq_ctx->tail_sqb); + seq_printf(m, "W6: smenq_sqb \t\t\t%llx\n\n", sq_ctx->smenq_sqb); + seq_printf(m, "W7: smenq_next_sqb \t\t%llx\n\n", + sq_ctx->smenq_next_sqb); + + seq_printf(m, "W8: head_sqb\t\t\t%llx\n\n", sq_ctx->head_sqb); + + seq_printf(m, "W9: vfi_lso_total\t\t%d\n", sq_ctx->vfi_lso_total); + seq_printf(m, "W9: vfi_lso_sizem1\t\t%d\nW9: vfi_lso_sb\t\t\t%d\n", + sq_ctx->vfi_lso_sizem1, sq_ctx->vfi_lso_sb); + seq_printf(m, "W9: vfi_lso_mps\t\t\t%d\nW9: vfi_lso_vlan0_ins_ena\t%d\n", + sq_ctx->vfi_lso_mps, sq_ctx->vfi_lso_vlan0_ins_ena); + seq_printf(m, "W9: vfi_lso_vlan1_ins_ena\t%d\nW9: vfi_lso_vld \t\t%d\n\n", + sq_ctx->vfi_lso_vld, sq_ctx->vfi_lso_vlan1_ins_ena); + + seq_printf(m, "W10: scm_lso_rem \t\t%llu\n\n", + (u64)sq_ctx->scm_lso_rem); + seq_printf(m, "W11: octs \t\t\t%llu\n\n", (u64)sq_ctx->octs); + seq_printf(m, "W12: pkts \t\t\t%llu\n\n", (u64)sq_ctx->pkts); + seq_printf(m, "W13: aged_drop_octs \t\t\t%llu\n\n", + (u64)sq_ctx->aged_drop_octs); + seq_printf(m, "W13: aged_drop_pkts \t\t\t%llu\n\n", + (u64)sq_ctx->aged_drop_pkts); + seq_printf(m, "W14: dropped_octs \t\t%llu\n\n", + (u64)sq_ctx->dropped_octs); + seq_printf(m, "W15: dropped_pkts \t\t%llu\n\n", + (u64)sq_ctx->dropped_pkts); +} + +void print_nix_cn20k_cq_ctx(struct seq_file *m, + struct nix_cn20k_aq_enq_rsp *rsp) +{ + struct nix_cn20k_cq_ctx_s *cq_ctx = &rsp->cq; + + seq_printf(m, "W0: base \t\t\t%llx\n\n", cq_ctx->base); + + seq_printf(m, "W1: wrptr \t\t\t%llx\n", (u64)cq_ctx->wrptr); + seq_printf(m, "W1: avg_con \t\t\t%d\nW1: cint_idx \t\t\t%d\n", + cq_ctx->avg_con, cq_ctx->cint_idx); + seq_printf(m, "W1: cq_err \t\t\t%d\nW1: qint_idx \t\t\t%d\n", + cq_ctx->cq_err, cq_ctx->qint_idx); + seq_printf(m, "W1: bpid \t\t\t%d\nW1: bp_ena \t\t\t%d\n\n", + cq_ctx->bpid, cq_ctx->bp_ena); + + seq_printf(m, "W1: lbpid_high \t\t\t0x%03x\n", cq_ctx->lbpid_high); + seq_printf(m, "W1: lbpid_med \t\t\t0x%03x\n", cq_ctx->lbpid_med); + seq_printf(m, "W1: lbpid_low \t\t\t0x%03x\n", cq_ctx->lbpid_low); + seq_printf(m, "(W1: lbpid) \t\t\t0x%03x\n", + cq_ctx->lbpid_high << 6 | cq_ctx->lbpid_med << 3 | + cq_ctx->lbpid_low); + seq_printf(m, "W1: lbp_ena \t\t\t\t%d\n\n", cq_ctx->lbp_ena); + + seq_printf(m, "W2: update_time \t\t%d\nW2:avg_level \t\t\t%d\n", + cq_ctx->update_time, cq_ctx->avg_level); + seq_printf(m, "W2: head \t\t\t%d\nW2:tail \t\t\t%d\n\n", + cq_ctx->head, cq_ctx->tail); + + seq_printf(m, "W3: cq_err_int_ena \t\t%d\nW3:cq_err_int \t\t\t%d\n", + cq_ctx->cq_err_int_ena, cq_ctx->cq_err_int); + seq_printf(m, "W3: qsize \t\t\t%d\nW3:stashing \t\t\t%d\n", + cq_ctx->qsize, cq_ctx->stashing); + + seq_printf(m, "W3: caching \t\t\t%d\n", cq_ctx->caching); + seq_printf(m, "W3: lbp_frac \t\t\t%d\n", cq_ctx->lbp_frac); + seq_printf(m, "W3: stash_thresh \t\t\t%d\n", + cq_ctx->stash_thresh); + + seq_printf(m, "W3: msh_valid \t\t\t%d\nW3:msh_dst \t\t\t%d\n", + cq_ctx->msh_valid, cq_ctx->msh_dst); + + seq_printf(m, "W3: cpt_drop_err_en \t\t\t%d\n", + cq_ctx->cpt_drop_err_en); + seq_printf(m, "W3: ena \t\t\t%d\n", + cq_ctx->ena); + seq_printf(m, "W3: drop_ena \t\t\t%d\nW3: drop \t\t\t%d\n", + cq_ctx->drop_ena, cq_ctx->drop); + seq_printf(m, "W3: bp \t\t\t\t%d\n\n", cq_ctx->bp); + + seq_printf(m, "W4: lbpid_ext \t\t\t\t%d\n\n", cq_ctx->lbpid_ext); + seq_printf(m, "W4: bpid_ext \t\t\t\t%d\n\n", cq_ctx->bpid_ext); +} diff --git a/drivers/net/ethernet/marvell/octeontx2/af/cn20k/debugfs.h b/drivers/net/ethernet/marvell/octeontx2/af/cn20k/debugfs.h new file mode 100644 index 000000000000..9d3a98dc3000 --- /dev/null +++ b/drivers/net/ethernet/marvell/octeontx2/af/cn20k/debugfs.h @@ -0,0 +1,24 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Marvell OcteonTx2 CGX driver + * + * Copyright (C) 2024 Marvell. + * + */ + +#ifndef DEBUFS_H +#define DEBUFS_H + +#include +#include +#include +#include + +#include "struct.h" +#include "../mbox.h" + +void print_nix_cn20k_sq_ctx(struct seq_file *m, + struct nix_cn20k_sq_ctx_s *sq_ctx); +void print_nix_cn20k_cq_ctx(struct seq_file *m, + struct nix_cn20k_aq_enq_rsp *rsp); + +#endif diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c index 8375f18c8e07..eeca8cef7964 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c @@ -21,6 +21,8 @@ #include "rvu_npc_hash.h" #include "mcs.h" +#include "cn20k/debugfs.h" + #define DEBUGFS_DIR_NAME "octeontx2" enum { @@ -2009,10 +2011,16 @@ static void print_nix_sq_ctx(struct seq_file *m, struct nix_aq_enq_rsp *rsp) struct nix_hw *nix_hw = m->private; struct rvu *rvu = nix_hw->rvu; + if (is_cn20k(rvu->pdev)) { + print_nix_cn20k_sq_ctx(m, (struct nix_cn20k_sq_ctx_s *)sq_ctx); + return; + } + if (!is_rvu_otx2(rvu)) { print_nix_cn10k_sq_ctx(m, (struct nix_cn10k_sq_ctx_s *)sq_ctx); return; } + seq_printf(m, "W0: sqe_way_mask \t\t%d\nW0: cq \t\t\t\t%d\n", sq_ctx->sqe_way_mask, sq_ctx->cq); seq_printf(m, "W0: sdp_mcast \t\t\t%d\nW0: substream \t\t\t0x%03x\n", @@ -2225,6 +2233,11 @@ static void print_nix_cq_ctx(struct seq_file *m, struct nix_aq_enq_rsp *rsp) struct nix_hw *nix_hw = m->private; struct rvu *rvu = nix_hw->rvu; + if (is_cn20k(rvu->pdev)) { + print_nix_cn20k_cq_ctx(m, (struct nix_cn20k_aq_enq_rsp *)rsp); + return; + } + seq_printf(m, "W0: base \t\t\t%llx\n\n", cq_ctx->base); seq_printf(m, "W1: wrptr \t\t\t%llx\n", (u64)cq_ctx->wrptr); @@ -2254,6 +2267,7 @@ static void print_nix_cq_ctx(struct seq_file *m, struct nix_aq_enq_rsp *rsp) cq_ctx->cq_err_int_ena, cq_ctx->cq_err_int); seq_printf(m, "W3: qsize \t\t\t%d\nW3:caching \t\t\t%d\n", cq_ctx->qsize, cq_ctx->caching); + seq_printf(m, "W3: substream \t\t\t0x%03x\nW3: ena \t\t\t%d\n", cq_ctx->substream, cq_ctx->ena); if (!is_rvu_otx2(rvu)) { @@ -3950,6 +3964,9 @@ static void rvu_dbg_cpt_init(struct rvu *rvu, int blkaddr) static const char *rvu_get_dbg_dir_name(struct rvu *rvu) { + if (is_cn20k(rvu->pdev)) + return "cn20k"; + if (!is_rvu_otx2(rvu)) return "cn10k"; else From 8a8b1301277405047f73e641eadf0555f502756a Mon Sep 17 00:00:00 2001 From: Linu Cherian Date: Sat, 25 Oct 2025 16:02:40 +0530 Subject: [PATCH 0391/1381] octeontx2-af: Add cn20k NPA block contexts New CN20K silicon has NPA hardware context structures different from previous silicons. Add NPA aura and pool context definitions for cn20k. Extend NPA context handling support to cn20k. Signed-off-by: Linu Cherian Signed-off-by: Subbaraya Sundeep Link: https://patch.msgid.link/1761388367-16579-5-git-send-email-sbhatta@marvell.com Signed-off-by: Paolo Abeni --- .../ethernet/marvell/octeontx2/af/Makefile | 3 +- .../ethernet/marvell/octeontx2/af/cn20k/npa.c | 21 +++ .../marvell/octeontx2/af/cn20k/struct.h | 135 ++++++++++++++++++ .../net/ethernet/marvell/octeontx2/af/mbox.h | 35 +++++ 4 files changed, 193 insertions(+), 1 deletion(-) create mode 100644 drivers/net/ethernet/marvell/octeontx2/af/cn20k/npa.c diff --git a/drivers/net/ethernet/marvell/octeontx2/af/Makefile b/drivers/net/ethernet/marvell/octeontx2/af/Makefile index 57eeaa116116..244de500963e 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/Makefile +++ b/drivers/net/ethernet/marvell/octeontx2/af/Makefile @@ -12,4 +12,5 @@ rvu_af-y := cgx.o rvu.o rvu_cgx.o rvu_npa.o rvu_nix.o \ rvu_reg.o rvu_npc.o rvu_debugfs.o ptp.o rvu_npc_fs.o \ rvu_cpt.o rvu_devlink.o rpm.o rvu_cn10k.o rvu_switch.o \ rvu_sdp.o rvu_npc_hash.o mcs.o mcs_rvu_if.o mcs_cnf10kb.o \ - rvu_rep.o cn20k/mbox_init.o cn20k/nix.o cn20k/debugfs.o + rvu_rep.o cn20k/mbox_init.o cn20k/nix.o cn20k/debugfs.o \ + cn20k/npa.o diff --git a/drivers/net/ethernet/marvell/octeontx2/af/cn20k/npa.c b/drivers/net/ethernet/marvell/octeontx2/af/cn20k/npa.c new file mode 100644 index 000000000000..fe8f926c8b75 --- /dev/null +++ b/drivers/net/ethernet/marvell/octeontx2/af/cn20k/npa.c @@ -0,0 +1,21 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Marvell RVU Admin Function driver + * + * Copyright (C) 2024 Marvell. + * + */ + +#include +#include + +#include "struct.h" +#include "../rvu.h" + +int rvu_mbox_handler_npa_cn20k_aq_enq(struct rvu *rvu, + struct npa_cn20k_aq_enq_req *req, + struct npa_cn20k_aq_enq_rsp *rsp) +{ + return rvu_npa_aq_enq_inst(rvu, (struct npa_aq_enq_req *)req, + (struct npa_aq_enq_rsp *)rsp); +} +EXPORT_SYMBOL(rvu_mbox_handler_npa_cn20k_aq_enq); diff --git a/drivers/net/ethernet/marvell/octeontx2/af/cn20k/struct.h b/drivers/net/ethernet/marvell/octeontx2/af/cn20k/struct.h index ff8f0d1809c8..763f6cabd7c2 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/cn20k/struct.h +++ b/drivers/net/ethernet/marvell/octeontx2/af/cn20k/struct.h @@ -242,4 +242,139 @@ struct nix_cn20k_rq_ctx_s { static_assert(sizeof(struct nix_cn20k_rq_ctx_s) == NIX_MAX_CTX_SIZE); +struct npa_cn20k_aura_s { + u64 pool_addr; /* W0 */ + u64 ena : 1; /* W1 */ + u64 reserved_65 : 2; + u64 pool_caching : 1; + u64 reserved_68 : 16; + u64 avg_con : 9; + u64 reserved_93 : 1; + u64 pool_drop_ena : 1; + u64 aura_drop_ena : 1; + u64 bp_ena : 1; + u64 reserved_97_103 : 7; + u64 aura_drop : 8; + u64 shift : 6; + u64 reserved_118_119 : 2; + u64 avg_level : 8; + u64 count : 36; /* W2 */ + u64 reserved_164_167 : 4; + u64 bpid : 12; + u64 reserved_180_191 : 12; + u64 limit : 36; /* W3 */ + u64 reserved_228_231 : 4; + u64 bp : 7; + u64 reserved_239_243 : 5; + u64 fc_ena : 1; + u64 fc_up_crossing : 1; + u64 fc_stype : 2; + u64 fc_hyst_bits : 4; + u64 reserved_252_255 : 4; + u64 fc_addr; /* W4 */ + u64 pool_drop : 8; /* W5 */ + u64 update_time : 16; + u64 err_int : 8; + u64 err_int_ena : 8; + u64 thresh_int : 1; + u64 thresh_int_ena : 1; + u64 thresh_up : 1; + u64 reserved_363 : 1; + u64 thresh_qint_idx : 7; + u64 reserved_371 : 1; + u64 err_qint_idx : 7; + u64 reserved_379_383 : 5; + u64 thresh : 36; /* W6*/ + u64 rsvd_423_420 : 4; + u64 fc_msh_dst : 11; + u64 reserved_435_438 : 4; + u64 op_dpc_ena : 1; + u64 op_dpc_set : 5; + u64 reserved_445_445 : 1; + u64 stream_ctx : 1; + u64 unified_ctx : 1; + u64 reserved_448_511; /* W7 */ + u64 padding[8]; +}; + +static_assert(sizeof(struct npa_cn20k_aura_s) == NIX_MAX_CTX_SIZE); + +struct npa_cn20k_pool_s { + u64 stack_base; /* W0 */ + u64 ena : 1; + u64 nat_align : 1; + u64 reserved_66_67 : 2; + u64 stack_caching : 1; + u64 reserved_69_87 : 19; + u64 buf_offset : 12; + u64 reserved_100_103 : 4; + u64 buf_size : 12; + u64 reserved_116_119 : 4; + u64 ref_cnt_prof : 3; + u64 reserved_123_127 : 5; + u64 stack_max_pages : 32; + u64 stack_pages : 32; + uint64_t bp_0 : 7; + uint64_t bp_1 : 7; + uint64_t bp_2 : 7; + uint64_t bp_3 : 7; + uint64_t bp_4 : 7; + uint64_t bp_5 : 7; + uint64_t bp_6 : 7; + uint64_t bp_7 : 7; + uint64_t bp_ena_0 : 1; + uint64_t bp_ena_1 : 1; + uint64_t bp_ena_2 : 1; + uint64_t bp_ena_3 : 1; + uint64_t bp_ena_4 : 1; + uint64_t bp_ena_5 : 1; + uint64_t bp_ena_6 : 1; + uint64_t bp_ena_7 : 1; + u64 stack_offset : 4; + u64 reserved_260_263 : 4; + u64 shift : 6; + u64 reserved_270_271 : 2; + u64 avg_level : 8; + u64 avg_con : 9; + u64 fc_ena : 1; + u64 fc_stype : 2; + u64 fc_hyst_bits : 4; + u64 fc_up_crossing : 1; + u64 reserved_297_299 : 3; + u64 update_time : 16; + u64 reserved_316_319 : 4; + u64 fc_addr; /* W5 */ + u64 ptr_start; /* W6 */ + u64 ptr_end; /* W7 */ + u64 bpid_0 : 12; + u64 reserved_524_535 : 12; + u64 err_int : 8; + u64 err_int_ena : 8; + u64 thresh_int : 1; + u64 thresh_int_ena : 1; + u64 thresh_up : 1; + u64 reserved_555 : 1; + u64 thresh_qint_idx : 7; + u64 reserved_563 : 1; + u64 err_qint_idx : 7; + u64 reserved_571_575 : 5; + u64 thresh : 36; + u64 rsvd_612_615 : 4; + u64 fc_msh_dst : 11; + u64 reserved_627_630 : 4; + u64 op_dpc_ena : 1; + u64 op_dpc_set : 5; + u64 reserved_637_637 : 1; + u64 stream_ctx : 1; + u64 reserved_639 : 1; + u64 reserved_640_703; /* W10 */ + u64 reserved_704_767; /* W11 */ + u64 reserved_768_831; /* W12 */ + u64 reserved_832_895; /* W13 */ + u64 reserved_896_959; /* W14 */ + u64 reserved_960_1023; /* W15 */ +}; + +static_assert(sizeof(struct npa_cn20k_pool_s) == NIX_MAX_CTX_SIZE); + #endif diff --git a/drivers/net/ethernet/marvell/octeontx2/af/mbox.h b/drivers/net/ethernet/marvell/octeontx2/af/mbox.h index 01086c52e78d..a3e273126e4e 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/mbox.h +++ b/drivers/net/ethernet/marvell/octeontx2/af/mbox.h @@ -203,6 +203,8 @@ M(NPA_LF_ALLOC, 0x400, npa_lf_alloc, \ M(NPA_LF_FREE, 0x401, npa_lf_free, msg_req, msg_rsp) \ M(NPA_AQ_ENQ, 0x402, npa_aq_enq, npa_aq_enq_req, npa_aq_enq_rsp) \ M(NPA_HWCTX_DISABLE, 0x403, npa_hwctx_disable, hwctx_disable_req, msg_rsp)\ +M(NPA_CN20K_AQ_ENQ, 0x404, npa_cn20k_aq_enq, npa_cn20k_aq_enq_req, \ + npa_cn20k_aq_enq_rsp) \ /* SSO/SSOW mbox IDs (range 0x600 - 0x7FF) */ \ /* TIM mbox IDs (range 0x800 - 0x9FF) */ \ /* CPT mbox IDs (range 0xA00 - 0xBFF) */ \ @@ -834,6 +836,39 @@ struct npa_aq_enq_rsp { }; }; +struct npa_cn20k_aq_enq_req { + struct mbox_msghdr hdr; + u32 aura_id; + u8 ctype; + u8 op; + union { + /* Valid when op == WRITE/INIT and ctype == AURA. + * LF fills the pool_id in aura.pool_addr. AF will translate + * the pool_id to pool context pointer. + */ + struct npa_cn20k_aura_s aura; + /* Valid when op == WRITE/INIT and ctype == POOL */ + struct npa_cn20k_pool_s pool; + }; + /* Mask data when op == WRITE (1=write, 0=don't write) */ + union { + /* Valid when op == WRITE and ctype == AURA */ + struct npa_cn20k_aura_s aura_mask; + /* Valid when op == WRITE and ctype == POOL */ + struct npa_cn20k_pool_s pool_mask; + }; +}; + +struct npa_cn20k_aq_enq_rsp { + struct mbox_msghdr hdr; + union { + /* Valid when op == READ and ctype == AURA */ + struct npa_cn20k_aura_s aura; + /* Valid when op == READ and ctype == POOL */ + struct npa_cn20k_pool_s pool; + }; +}; + /* Disable all contexts of type 'ctype' */ struct hwctx_disable_req { struct mbox_msghdr hdr; From e4a8e78aca5e9e1bc5dc0f69a069c04fc9af943e Mon Sep 17 00:00:00 2001 From: Linu Cherian Date: Sat, 25 Oct 2025 16:02:41 +0530 Subject: [PATCH 0392/1381] octeontx2-af: Extend debugfs support for cn20k NPA Extend debugfs to display CN20K NPA aura and pool contexts. Signed-off-by: Linu Cherian Signed-off-by: Subbaraya Sundeep Link: https://patch.msgid.link/1761388367-16579-6-git-send-email-sbhatta@marvell.com Signed-off-by: Paolo Abeni --- .../marvell/octeontx2/af/cn20k/debugfs.c | 84 +++++++++++++++++++ .../marvell/octeontx2/af/cn20k/debugfs.h | 4 + .../marvell/octeontx2/af/rvu_debugfs.c | 10 +++ 3 files changed, 98 insertions(+) diff --git a/drivers/net/ethernet/marvell/octeontx2/af/cn20k/debugfs.c b/drivers/net/ethernet/marvell/octeontx2/af/cn20k/debugfs.c index 50b1bd1d2c86..498968bf4cf5 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/cn20k/debugfs.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/cn20k/debugfs.c @@ -132,3 +132,87 @@ void print_nix_cn20k_cq_ctx(struct seq_file *m, seq_printf(m, "W4: lbpid_ext \t\t\t\t%d\n\n", cq_ctx->lbpid_ext); seq_printf(m, "W4: bpid_ext \t\t\t\t%d\n\n", cq_ctx->bpid_ext); } + +void print_npa_cn20k_aura_ctx(struct seq_file *m, + struct npa_cn20k_aq_enq_rsp *rsp) +{ + struct npa_cn20k_aura_s *aura = &rsp->aura; + + seq_printf(m, "W0: Pool addr\t\t%llx\n", aura->pool_addr); + + seq_printf(m, "W1: ena\t\t\t%d\nW1: pool caching\t%d\n", + aura->ena, aura->pool_caching); + seq_printf(m, "W1: avg con\t\t%d\n", aura->avg_con); + seq_printf(m, "W1: pool drop ena\t%d\nW1: aura drop ena\t%d\n", + aura->pool_drop_ena, aura->aura_drop_ena); + seq_printf(m, "W1: bp_ena\t\t%d\nW1: aura drop\t\t%d\n", + aura->bp_ena, aura->aura_drop); + seq_printf(m, "W1: aura shift\t\t%d\nW1: avg_level\t\t%d\n", + aura->shift, aura->avg_level); + + seq_printf(m, "W2: count\t\t%llu\nW2: nix_bpid\t\t%d\n", + (u64)aura->count, aura->bpid); + + seq_printf(m, "W3: limit\t\t%llu\nW3: bp\t\t\t%d\nW3: fc_ena\t\t%d\n", + (u64)aura->limit, aura->bp, aura->fc_ena); + + seq_printf(m, "W3: fc_up_crossing\t%d\nW3: fc_stype\t\t%d\n", + aura->fc_up_crossing, aura->fc_stype); + seq_printf(m, "W3: fc_hyst_bits\t%d\n", aura->fc_hyst_bits); + + seq_printf(m, "W4: fc_addr\t\t%llx\n", aura->fc_addr); + + seq_printf(m, "W5: pool_drop\t\t%d\nW5: update_time\t\t%d\n", + aura->pool_drop, aura->update_time); + seq_printf(m, "W5: err_int \t\t%d\nW5: err_int_ena\t\t%d\n", + aura->err_int, aura->err_int_ena); + seq_printf(m, "W5: thresh_int\t\t%d\nW5: thresh_int_ena \t%d\n", + aura->thresh_int, aura->thresh_int_ena); + seq_printf(m, "W5: thresh_up\t\t%d\nW5: thresh_qint_idx\t%d\n", + aura->thresh_up, aura->thresh_qint_idx); + seq_printf(m, "W5: err_qint_idx \t%d\n", aura->err_qint_idx); + + seq_printf(m, "W6: thresh\t\t%llu\n", (u64)aura->thresh); + seq_printf(m, "W6: fc_msh_dst\t\t%d\n", aura->fc_msh_dst); +} + +void print_npa_cn20k_pool_ctx(struct seq_file *m, + struct npa_cn20k_aq_enq_rsp *rsp) +{ + struct npa_cn20k_pool_s *pool = &rsp->pool; + + seq_printf(m, "W0: Stack base\t\t%llx\n", pool->stack_base); + + seq_printf(m, "W1: ena \t\t%d\nW1: nat_align \t\t%d\n", + pool->ena, pool->nat_align); + seq_printf(m, "W1: stack_caching\t%d\n", + pool->stack_caching); + seq_printf(m, "W1: buf_offset\t\t%d\nW1: buf_size\t\t%d\n", + pool->buf_offset, pool->buf_size); + + seq_printf(m, "W2: stack_max_pages \t%d\nW2: stack_pages\t\t%d\n", + pool->stack_max_pages, pool->stack_pages); + + seq_printf(m, "W4: stack_offset\t%d\nW4: shift\t\t%d\nW4: avg_level\t\t%d\n", + pool->stack_offset, pool->shift, pool->avg_level); + seq_printf(m, "W4: avg_con \t\t%d\nW4: fc_ena\t\t%d\nW4: fc_stype\t\t%d\n", + pool->avg_con, pool->fc_ena, pool->fc_stype); + seq_printf(m, "W4: fc_hyst_bits\t%d\nW4: fc_up_crossing\t%d\n", + pool->fc_hyst_bits, pool->fc_up_crossing); + seq_printf(m, "W4: update_time\t\t%d\n", pool->update_time); + + seq_printf(m, "W5: fc_addr\t\t%llx\n", pool->fc_addr); + + seq_printf(m, "W6: ptr_start\t\t%llx\n", pool->ptr_start); + + seq_printf(m, "W7: ptr_end\t\t%llx\n", pool->ptr_end); + + seq_printf(m, "W8: err_int\t\t%d\nW8: err_int_ena\t\t%d\n", + pool->err_int, pool->err_int_ena); + seq_printf(m, "W8: thresh_int\t\t%d\n", pool->thresh_int); + seq_printf(m, "W8: thresh_int_ena\t%d\nW8: thresh_up\t\t%d\n", + pool->thresh_int_ena, pool->thresh_up); + seq_printf(m, "W8: thresh_qint_idx\t%d\nW8: err_qint_idx\t%d\n", + pool->thresh_qint_idx, pool->err_qint_idx); + seq_printf(m, "W8: fc_msh_dst\t\t%d\n", pool->fc_msh_dst); +} diff --git a/drivers/net/ethernet/marvell/octeontx2/af/cn20k/debugfs.h b/drivers/net/ethernet/marvell/octeontx2/af/cn20k/debugfs.h index 9d3a98dc3000..a2e3a2cd6edb 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/cn20k/debugfs.h +++ b/drivers/net/ethernet/marvell/octeontx2/af/cn20k/debugfs.h @@ -20,5 +20,9 @@ void print_nix_cn20k_sq_ctx(struct seq_file *m, struct nix_cn20k_sq_ctx_s *sq_ctx); void print_nix_cn20k_cq_ctx(struct seq_file *m, struct nix_cn20k_aq_enq_rsp *rsp); +void print_npa_cn20k_aura_ctx(struct seq_file *m, + struct npa_cn20k_aq_enq_rsp *rsp); +void print_npa_cn20k_pool_ctx(struct seq_file *m, + struct npa_cn20k_aq_enq_rsp *rsp); #endif diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c index eeca8cef7964..c55a0f15380d 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c @@ -1103,6 +1103,11 @@ static void print_npa_aura_ctx(struct seq_file *m, struct npa_aq_enq_rsp *rsp) struct npa_aura_s *aura = &rsp->aura; struct rvu *rvu = m->private; + if (is_cn20k(rvu->pdev)) { + print_npa_cn20k_aura_ctx(m, (struct npa_cn20k_aq_enq_rsp *)rsp); + return; + } + seq_printf(m, "W0: Pool addr\t\t%llx\n", aura->pool_addr); seq_printf(m, "W1: ena\t\t\t%d\nW1: pool caching\t%d\n", @@ -1151,6 +1156,11 @@ static void print_npa_pool_ctx(struct seq_file *m, struct npa_aq_enq_rsp *rsp) struct npa_pool_s *pool = &rsp->pool; struct rvu *rvu = m->private; + if (is_cn20k(rvu->pdev)) { + print_npa_cn20k_pool_ctx(m, (struct npa_cn20k_aq_enq_rsp *)rsp); + return; + } + seq_printf(m, "W0: Stack base\t\t%llx\n", pool->stack_base); seq_printf(m, "W1: ena \t\t%d\nW1: nat_align \t\t%d\n", From a861e5809f3e7ecbfde09546f6073da34bde1b47 Mon Sep 17 00:00:00 2001 From: Linu Cherian Date: Sat, 25 Oct 2025 16:02:42 +0530 Subject: [PATCH 0393/1381] octeontx2-af: Skip NDC operations for cn20k For cn20k, NPA block doesn't use the general purpose NDC (Near Coprocessor Bus Data cache Unit) for caching, hence skip the NDC related operations. Also refactor NDC configuration code to a helper function. Signed-off-by: Linu Cherian Signed-off-by: Subbaraya Sundeep Link: https://patch.msgid.link/1761388367-16579-7-git-send-email-sbhatta@marvell.com Signed-off-by: Paolo Abeni --- .../marvell/octeontx2/af/rvu_debugfs.c | 3 ++ .../ethernet/marvell/octeontx2/af/rvu_npa.c | 29 ++++++++++++++----- 2 files changed, 24 insertions(+), 8 deletions(-) diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c index c55a0f15380d..8ab82700e826 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c @@ -2808,6 +2808,9 @@ static void rvu_dbg_npa_init(struct rvu *rvu) &rvu_dbg_npa_aura_ctx_fops); debugfs_create_file("pool_ctx", 0600, rvu->rvu_dbg.npa, rvu, &rvu_dbg_npa_pool_ctx_fops); + + if (is_cn20k(rvu->pdev)) /* NDC not appliable for cn20k */ + return; debugfs_create_file("ndc_cache", 0600, rvu->rvu_dbg.npa, rvu, &rvu_dbg_npa_ndc_cache_fops); debugfs_create_file("ndc_hits_miss", 0600, rvu->rvu_dbg.npa, rvu, diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npa.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npa.c index 4f5ca5ab13a4..e2a33e46b48a 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npa.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npa.c @@ -464,6 +464,23 @@ int rvu_mbox_handler_npa_lf_free(struct rvu *rvu, struct msg_req *req, return 0; } +static void npa_aq_ndc_config(struct rvu *rvu, struct rvu_block *block) +{ + u64 cfg; + + if (is_cn20k(rvu->pdev)) /* NDC not applicable to cn20k */ + return; + + /* Do not bypass NDC cache */ + cfg = rvu_read64(rvu, block->addr, NPA_AF_NDC_CFG); + cfg &= ~0x03DULL; +#ifdef CONFIG_NDC_DIS_DYNAMIC_CACHING + /* Disable caching of stack pages */ + cfg |= 0x10ULL; +#endif + rvu_write64(rvu, block->addr, NPA_AF_NDC_CFG, cfg); +} + static int npa_aq_init(struct rvu *rvu, struct rvu_block *block) { u64 cfg; @@ -479,14 +496,7 @@ static int npa_aq_init(struct rvu *rvu, struct rvu_block *block) rvu_write64(rvu, block->addr, NPA_AF_GEN_CFG, cfg); #endif - /* Do not bypass NDC cache */ - cfg = rvu_read64(rvu, block->addr, NPA_AF_NDC_CFG); - cfg &= ~0x03DULL; -#ifdef CONFIG_NDC_DIS_DYNAMIC_CACHING - /* Disable caching of stack pages */ - cfg |= 0x10ULL; -#endif - rvu_write64(rvu, block->addr, NPA_AF_NDC_CFG, cfg); + npa_aq_ndc_config(rvu, block); /* For CN10K NPA BATCH DMA set 35 cache lines */ if (!is_rvu_otx2(rvu)) { @@ -567,6 +577,9 @@ int rvu_ndc_fix_locked_cacheline(struct rvu *rvu, int blkaddr) int bank, max_bank, line, max_line, err; u64 reg, ndc_af_const; + if (is_cn20k(rvu->pdev)) /* NDC not applicable to cn20k */ + return 0; + /* Set the ENABLE bit(63) to '0' */ reg = rvu_read64(rvu, blkaddr, NDC_AF_CAMS_RD_INTERVAL); rvu_write64(rvu, blkaddr, NDC_AF_CAMS_RD_INTERVAL, reg & GENMASK_ULL(62, 0)); From d322fbd1720382a37ec4d2cf3f3eda7e59439d40 Mon Sep 17 00:00:00 2001 From: Linu Cherian Date: Sat, 25 Oct 2025 16:02:43 +0530 Subject: [PATCH 0394/1381] octeontx2-pf: Initialize cn20k specific aura and pool contexts With new CN20K NPA pool and aura contexts supported in AF driver this patch modifies PF driver to use new NPA contexts. Implement new hw_ops for intializing aura and pool contexts for all the silicons. Signed-off-by: Linu Cherian Signed-off-by: Subbaraya Sundeep Link: https://patch.msgid.link/1761388367-16579-8-git-send-email-sbhatta@marvell.com Signed-off-by: Paolo Abeni --- .../ethernet/marvell/octeontx2/nic/cn10k.c | 4 + .../ethernet/marvell/octeontx2/nic/cn20k.c | 186 ++++++++++++++++-- .../marvell/octeontx2/nic/otx2_common.c | 14 ++ .../marvell/octeontx2/nic/otx2_common.h | 10 + 4 files changed, 203 insertions(+), 11 deletions(-) diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/cn10k.c b/drivers/net/ethernet/marvell/octeontx2/nic/cn10k.c index bec7d5b4d7cc..cab157aac251 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/cn10k.c +++ b/drivers/net/ethernet/marvell/octeontx2/nic/cn10k.c @@ -15,6 +15,8 @@ static struct dev_hw_ops otx2_hw_ops = { .aura_freeptr = otx2_aura_freeptr, .refill_pool_ptrs = otx2_refill_pool_ptrs, .pfaf_mbox_intr_handler = otx2_pfaf_mbox_intr_handler, + .aura_aq_init = otx2_aura_aq_init, + .pool_aq_init = otx2_pool_aq_init, }; static struct dev_hw_ops cn10k_hw_ops = { @@ -23,6 +25,8 @@ static struct dev_hw_ops cn10k_hw_ops = { .aura_freeptr = cn10k_aura_freeptr, .refill_pool_ptrs = cn10k_refill_pool_ptrs, .pfaf_mbox_intr_handler = otx2_pfaf_mbox_intr_handler, + .aura_aq_init = otx2_aura_aq_init, + .pool_aq_init = otx2_pool_aq_init, }; void otx2_init_hw_ops(struct otx2_nic *pfvf) diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/cn20k.c b/drivers/net/ethernet/marvell/octeontx2/nic/cn20k.c index ec8cde98076d..6063025824ec 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/cn20k.c +++ b/drivers/net/ethernet/marvell/octeontx2/nic/cn20k.c @@ -10,17 +10,6 @@ #include "otx2_struct.h" #include "cn10k.h" -static struct dev_hw_ops cn20k_hw_ops = { - .pfaf_mbox_intr_handler = cn20k_pfaf_mbox_intr_handler, - .vfaf_mbox_intr_handler = cn20k_vfaf_mbox_intr_handler, - .pfvf_mbox_intr_handler = cn20k_pfvf_mbox_intr_handler, -}; - -void cn20k_init(struct otx2_nic *pfvf) -{ - pfvf->hw_ops = &cn20k_hw_ops; -} -EXPORT_SYMBOL(cn20k_init); /* CN20K mbox AF => PFx irq handler */ irqreturn_t cn20k_pfaf_mbox_intr_handler(int irq, void *pf_irq) { @@ -250,3 +239,178 @@ int cn20k_register_pfvf_mbox_intr(struct otx2_nic *pf, int numvfs) return 0; } + +#define RQ_BP_LVL_AURA (255 - ((85 * 256) / 100)) /* BP when 85% is full */ + +static u8 cn20k_aura_bpid_idx(struct otx2_nic *pfvf, int aura_id) +{ +#ifdef CONFIG_DCB + return pfvf->queue_to_pfc_map[aura_id]; +#else + return 0; +#endif +} + +static int cn20k_aura_aq_init(struct otx2_nic *pfvf, int aura_id, + int pool_id, int numptrs) +{ + struct npa_cn20k_aq_enq_req *aq; + struct otx2_pool *pool; + u8 bpid_idx; + int err; + + pool = &pfvf->qset.pool[pool_id]; + + /* Allocate memory for HW to update Aura count. + * Alloc one cache line, so that it fits all FC_STYPE modes. + */ + if (!pool->fc_addr) { + err = qmem_alloc(pfvf->dev, &pool->fc_addr, 1, OTX2_ALIGN); + if (err) + return err; + } + + /* Initialize this aura's context via AF */ + aq = otx2_mbox_alloc_msg_npa_cn20k_aq_enq(&pfvf->mbox); + if (!aq) { + /* Shared mbox memory buffer is full, flush it and retry */ + err = otx2_sync_mbox_msg(&pfvf->mbox); + if (err) + return err; + aq = otx2_mbox_alloc_msg_npa_cn20k_aq_enq(&pfvf->mbox); + if (!aq) + return -ENOMEM; + } + + aq->aura_id = aura_id; + + /* Will be filled by AF with correct pool context address */ + aq->aura.pool_addr = pool_id; + aq->aura.pool_caching = 1; + aq->aura.shift = ilog2(numptrs) - 8; + aq->aura.count = numptrs; + aq->aura.limit = numptrs; + aq->aura.avg_level = 255; + aq->aura.ena = 1; + aq->aura.fc_ena = 1; + aq->aura.fc_addr = pool->fc_addr->iova; + aq->aura.fc_hyst_bits = 0; /* Store count on all updates */ + + /* Enable backpressure for RQ aura */ + if (aura_id < pfvf->hw.rqpool_cnt && !is_otx2_lbkvf(pfvf->pdev)) { + aq->aura.bp_ena = 0; + /* If NIX1 LF is attached then specify NIX1_RX. + * + * Below NPA_AURA_S[BP_ENA] is set according to the + * NPA_BPINTF_E enumeration given as: + * 0x0 + a*0x1 where 'a' is 0 for NIX0_RX and 1 for NIX1_RX so + * NIX0_RX is 0x0 + 0*0x1 = 0 + * NIX1_RX is 0x0 + 1*0x1 = 1 + * But in HRM it is given that + * "NPA_AURA_S[BP_ENA](w1[33:32]) - Enable aura backpressure to + * NIX-RX based on [BP] level. One bit per NIX-RX; index + * enumerated by NPA_BPINTF_E." + */ + if (pfvf->nix_blkaddr == BLKADDR_NIX1) + aq->aura.bp_ena = 1; + + bpid_idx = cn20k_aura_bpid_idx(pfvf, aura_id); + aq->aura.bpid = pfvf->bpid[bpid_idx]; + + /* Set backpressure level for RQ's Aura */ + aq->aura.bp = RQ_BP_LVL_AURA; + } + + /* Fill AQ info */ + aq->ctype = NPA_AQ_CTYPE_AURA; + aq->op = NPA_AQ_INSTOP_INIT; + + return 0; +} + +static int cn20k_pool_aq_init(struct otx2_nic *pfvf, u16 pool_id, + int stack_pages, int numptrs, int buf_size, + int type) +{ + struct page_pool_params pp_params = { 0 }; + struct npa_cn20k_aq_enq_req *aq; + struct otx2_pool *pool; + int err, sz; + + pool = &pfvf->qset.pool[pool_id]; + /* Alloc memory for stack which is used to store buffer pointers */ + err = qmem_alloc(pfvf->dev, &pool->stack, + stack_pages, pfvf->hw.stack_pg_bytes); + if (err) + return err; + + pool->rbsize = buf_size; + + /* Initialize this pool's context via AF */ + aq = otx2_mbox_alloc_msg_npa_cn20k_aq_enq(&pfvf->mbox); + if (!aq) { + /* Shared mbox memory buffer is full, flush it and retry */ + err = otx2_sync_mbox_msg(&pfvf->mbox); + if (err) { + qmem_free(pfvf->dev, pool->stack); + return err; + } + aq = otx2_mbox_alloc_msg_npa_cn20k_aq_enq(&pfvf->mbox); + if (!aq) { + qmem_free(pfvf->dev, pool->stack); + return -ENOMEM; + } + } + + aq->aura_id = pool_id; + aq->pool.stack_base = pool->stack->iova; + aq->pool.stack_caching = 1; + aq->pool.ena = 1; + aq->pool.buf_size = buf_size / 128; + aq->pool.stack_max_pages = stack_pages; + aq->pool.shift = ilog2(numptrs) - 8; + aq->pool.ptr_start = 0; + aq->pool.ptr_end = ~0ULL; + + /* Fill AQ info */ + aq->ctype = NPA_AQ_CTYPE_POOL; + aq->op = NPA_AQ_INSTOP_INIT; + + if (type != AURA_NIX_RQ) { + pool->page_pool = NULL; + return 0; + } + + sz = ALIGN(ALIGN(SKB_DATA_ALIGN(buf_size), OTX2_ALIGN), PAGE_SIZE); + pp_params.order = get_order(sz); + pp_params.flags = PP_FLAG_DMA_MAP; + pp_params.pool_size = min(OTX2_PAGE_POOL_SZ, numptrs); + pp_params.nid = NUMA_NO_NODE; + pp_params.dev = pfvf->dev; + pp_params.dma_dir = DMA_FROM_DEVICE; + pool->page_pool = page_pool_create(&pp_params); + if (IS_ERR(pool->page_pool)) { + netdev_err(pfvf->netdev, "Creation of page pool failed\n"); + return PTR_ERR(pool->page_pool); + } + + return 0; +} + +static struct dev_hw_ops cn20k_hw_ops = { + .pfaf_mbox_intr_handler = cn20k_pfaf_mbox_intr_handler, + .vfaf_mbox_intr_handler = cn20k_vfaf_mbox_intr_handler, + .pfvf_mbox_intr_handler = cn20k_pfvf_mbox_intr_handler, + .sq_aq_init = cn10k_sq_aq_init, + .sqe_flush = cn10k_sqe_flush, + .aura_freeptr = cn10k_aura_freeptr, + .refill_pool_ptrs = cn10k_refill_pool_ptrs, + .aura_aq_init = cn20k_aura_aq_init, + .pool_aq_init = cn20k_pool_aq_init, +}; + +void cn20k_init(struct otx2_nic *pfvf) +{ + pfvf->hw_ops = &cn20k_hw_ops; +} +EXPORT_SYMBOL(cn20k_init); diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c index aff17c37ddde..3378be87a473 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c @@ -1368,6 +1368,13 @@ void otx2_aura_pool_free(struct otx2_nic *pfvf) int otx2_aura_init(struct otx2_nic *pfvf, int aura_id, int pool_id, int numptrs) +{ + return pfvf->hw_ops->aura_aq_init(pfvf, aura_id, pool_id, + numptrs); +} + +int otx2_aura_aq_init(struct otx2_nic *pfvf, int aura_id, + int pool_id, int numptrs) { struct npa_aq_enq_req *aq; struct otx2_pool *pool; @@ -1445,6 +1452,13 @@ int otx2_aura_init(struct otx2_nic *pfvf, int aura_id, int otx2_pool_init(struct otx2_nic *pfvf, u16 pool_id, int stack_pages, int numptrs, int buf_size, int type) +{ + return pfvf->hw_ops->pool_aq_init(pfvf, pool_id, stack_pages, numptrs, + buf_size, type); +} + +int otx2_pool_aq_init(struct otx2_nic *pfvf, u16 pool_id, + int stack_pages, int numptrs, int buf_size, int type) { struct page_pool_params pp_params = { 0 }; struct xsk_buff_pool *xsk_pool; diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h index ec26d1b6c789..e616a727a3a9 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h @@ -14,6 +14,7 @@ #include #include #include +#include #include #include #include @@ -375,6 +376,11 @@ struct dev_hw_ops { irqreturn_t (*pfaf_mbox_intr_handler)(int irq, void *pf_irq); irqreturn_t (*vfaf_mbox_intr_handler)(int irq, void *pf_irq); irqreturn_t (*pfvf_mbox_intr_handler)(int irq, void *pf_irq); + int (*aura_aq_init)(struct otx2_nic *pfvf, int aura_id, + int pool_id, int numptrs); + int (*pool_aq_init)(struct otx2_nic *pfvf, u16 pool_id, + int stack_pages, int numptrs, int buf_size, + int type); }; #define CN10K_MCS_SA_PER_SC 4 @@ -1059,6 +1065,10 @@ irqreturn_t otx2_cq_intr_handler(int irq, void *cq_irq); int otx2_rq_init(struct otx2_nic *pfvf, u16 qidx, u16 lpb_aura); int otx2_cq_init(struct otx2_nic *pfvf, u16 qidx); int otx2_set_hw_capabilities(struct otx2_nic *pfvf); +int otx2_aura_aq_init(struct otx2_nic *pfvf, int aura_id, + int pool_id, int numptrs); +int otx2_pool_aq_init(struct otx2_nic *pfvf, u16 pool_id, + int stack_pages, int numptrs, int buf_size, int type); /* RSS configuration APIs*/ int otx2_rss_init(struct otx2_nic *pfvf); From 81f12533572d929cd503b908593c7cfd5d13717e Mon Sep 17 00:00:00 2001 From: Subbaraya Sundeep Date: Sat, 25 Oct 2025 16:02:44 +0530 Subject: [PATCH 0395/1381] octeontx2-pf: Initialize new NIX SQ context for cn20k cn20k has different NIX context for send queue hence use the new cn20k mailbox to init SQ context. Signed-off-by: Subbaraya Sundeep Link: https://patch.msgid.link/1761388367-16579-9-git-send-email-sbhatta@marvell.com Signed-off-by: Paolo Abeni --- .../ethernet/marvell/octeontx2/nic/cn20k.c | 36 ++++++++++++++++++- 1 file changed, 35 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/cn20k.c b/drivers/net/ethernet/marvell/octeontx2/nic/cn20k.c index 6063025824ec..a60f8cf53feb 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/cn20k.c +++ b/drivers/net/ethernet/marvell/octeontx2/nic/cn20k.c @@ -397,11 +397,45 @@ static int cn20k_pool_aq_init(struct otx2_nic *pfvf, u16 pool_id, return 0; } +static int cn20k_sq_aq_init(void *dev, u16 qidx, u8 chan_offset, u16 sqb_aura) +{ + struct nix_cn20k_aq_enq_req *aq; + struct otx2_nic *pfvf = dev; + + /* Get memory to put this msg */ + aq = otx2_mbox_alloc_msg_nix_cn20k_aq_enq(&pfvf->mbox); + if (!aq) + return -ENOMEM; + + aq->sq.cq = pfvf->hw.rx_queues + qidx; + aq->sq.max_sqe_size = NIX_MAXSQESZ_W16; /* 128 byte */ + aq->sq.cq_ena = 1; + aq->sq.ena = 1; + aq->sq.smq = otx2_get_smq_idx(pfvf, qidx); + aq->sq.smq_rr_weight = mtu_to_dwrr_weight(pfvf, pfvf->tx_max_pktlen); + aq->sq.default_chan = pfvf->hw.tx_chan_base + chan_offset; + aq->sq.sqe_stype = NIX_STYPE_STF; /* Cache SQB */ + aq->sq.sqb_aura = sqb_aura; + aq->sq.sq_int_ena = NIX_SQINT_BITS; + aq->sq.qint_idx = 0; + /* Due pipelining impact minimum 2000 unused SQ CQE's + * need to maintain to avoid CQ overflow. + */ + aq->sq.cq_limit = (SEND_CQ_SKID * 256) / (pfvf->qset.sqe_cnt); + + /* Fill AQ info */ + aq->qidx = qidx; + aq->ctype = NIX_AQ_CTYPE_SQ; + aq->op = NIX_AQ_INSTOP_INIT; + + return otx2_sync_mbox_msg(&pfvf->mbox); +} + static struct dev_hw_ops cn20k_hw_ops = { .pfaf_mbox_intr_handler = cn20k_pfaf_mbox_intr_handler, .vfaf_mbox_intr_handler = cn20k_vfaf_mbox_intr_handler, .pfvf_mbox_intr_handler = cn20k_pfvf_mbox_intr_handler, - .sq_aq_init = cn10k_sq_aq_init, + .sq_aq_init = cn20k_sq_aq_init, .sqe_flush = cn10k_sqe_flush, .aura_freeptr = cn10k_aura_freeptr, .refill_pool_ptrs = cn10k_refill_pool_ptrs, From f7774633cf251b66d68695efe230fb0003a06d0a Mon Sep 17 00:00:00 2001 From: Subbaraya Sundeep Date: Sat, 25 Oct 2025 16:02:45 +0530 Subject: [PATCH 0396/1381] octeontx2-af: Accommodate more bandwidth profiles for cn20k CN20K has 16k of leaf profiles, 2k of middle profiles and 256 of top profiles. This patch modifies existing receive queue and bandwidth profile context structures to accommodate additional profiles of cn20k. Signed-off-by: Subbaraya Sundeep Link: https://patch.msgid.link/1761388367-16579-10-git-send-email-sbhatta@marvell.com Signed-off-by: Paolo Abeni --- .../ethernet/marvell/octeontx2/af/rvu_nix.c | 22 ++++++++++++++----- .../marvell/octeontx2/af/rvu_struct.h | 6 +++-- 2 files changed, 21 insertions(+), 7 deletions(-) diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c index d156d124f079..2f485a930edd 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c @@ -5818,6 +5818,8 @@ static void nix_ipolicer_freemem(struct rvu *rvu, struct nix_hw *nix_hw) } } +#define NIX_BW_PROF_HI_MASK GENMASK(10, 7) + static int nix_verify_bandprof(struct nix_cn10k_aq_enq_req *req, struct nix_hw *nix_hw, u16 pcifunc) { @@ -5856,7 +5858,8 @@ static int nix_verify_bandprof(struct nix_cn10k_aq_enq_req *req, return -EINVAL; ipolicer = &nix_hw->ipolicer[hi_layer]; - prof_idx = req->prof.band_prof_id; + prof_idx = FIELD_PREP(NIX_BW_PROF_HI_MASK, req->prof.band_prof_id_h); + prof_idx |= req->prof.band_prof_id; if (prof_idx >= ipolicer->band_prof.max || ipolicer->pfvf_map[prof_idx] != pcifunc) return -EINVAL; @@ -6021,8 +6024,10 @@ static int nix_ipolicer_map_leaf_midprofs(struct rvu *rvu, aq_req->op = NIX_AQ_INSTOP_WRITE; aq_req->qidx = leaf_prof; - aq_req->prof.band_prof_id = mid_prof; + aq_req->prof.band_prof_id = mid_prof & 0x7F; aq_req->prof_mask.band_prof_id = GENMASK(6, 0); + aq_req->prof.band_prof_id_h = FIELD_GET(NIX_BW_PROF_HI_MASK, mid_prof); + aq_req->prof_mask.band_prof_id_h = GENMASK(3, 0); aq_req->prof.hl_en = 1; aq_req->prof_mask.hl_en = 1; @@ -6031,6 +6036,8 @@ static int nix_ipolicer_map_leaf_midprofs(struct rvu *rvu, (struct nix_aq_enq_rsp *)aq_rsp); } +#define NIX_RQ_PROF_HI_MASK GENMASK(13, 10) + int rvu_nix_setup_ratelimit_aggr(struct rvu *rvu, u16 pcifunc, u16 rq_idx, u16 match_id) { @@ -6062,7 +6069,8 @@ int rvu_nix_setup_ratelimit_aggr(struct rvu *rvu, u16 pcifunc, return 0; /* Get the bandwidth profile ID mapped to this RQ */ - leaf_prof = aq_rsp.rq.band_prof_id; + leaf_prof = FIELD_PREP(NIX_RQ_PROF_HI_MASK, aq_rsp.rq.band_prof_id_h); + leaf_prof |= aq_rsp.rq.band_prof_id; ipolicer = &nix_hw->ipolicer[BAND_PROF_LEAF_LAYER]; ipolicer->match_id[leaf_prof] = match_id; @@ -6100,7 +6108,10 @@ int rvu_nix_setup_ratelimit_aggr(struct rvu *rvu, u16 pcifunc, * to different RQs and marked with same match_id * are rate limited in a aggregate fashion */ - mid_prof = aq_rsp.prof.band_prof_id; + mid_prof = FIELD_PREP(NIX_BW_PROF_HI_MASK, + aq_rsp.prof.band_prof_id_h); + mid_prof |= aq_rsp.prof.band_prof_id; + rc = nix_ipolicer_map_leaf_midprofs(rvu, nix_hw, &aq_req, &aq_rsp, leaf_prof, mid_prof); @@ -6222,7 +6233,8 @@ static void nix_clear_ratelimit_aggr(struct rvu *rvu, struct nix_hw *nix_hw, if (!aq_rsp.prof.hl_en) return; - mid_prof = aq_rsp.prof.band_prof_id; + mid_prof = FIELD_PREP(NIX_BW_PROF_HI_MASK, aq_rsp.prof.band_prof_id_h); + mid_prof |= aq_rsp.prof.band_prof_id; ipolicer = &nix_hw->ipolicer[BAND_PROF_MID_LAYER]; ipolicer->ref_count[mid_prof]--; /* If ref_count is zero, free mid layer profile */ diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_struct.h b/drivers/net/ethernet/marvell/octeontx2/af/rvu_struct.h index 8d41cb8f85ef..8e868f815de1 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_struct.h +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_struct.h @@ -419,7 +419,8 @@ struct nix_cn10k_rq_ctx_s { u64 rsvd_171 : 1; u64 later_skip : 6; u64 xqe_imm_size : 6; - u64 rsvd_189_184 : 6; + u64 band_prof_id_h : 4; + u64 rsvd_189_188 : 2; u64 xqe_imm_copy : 1; u64 xqe_hdr_split : 1; u64 xqe_drop : 8; /* W3 */ @@ -757,7 +758,8 @@ struct nix_bandprof_s { uint64_t rc_action : 2; uint64_t meter_algo : 2; uint64_t band_prof_id : 7; - uint64_t reserved_111_118 : 8; + uint64_t band_prof_id_h : 4; + uint64_t reserved_115_118 : 4; uint64_t hl_en : 1; uint64_t reserved_120_127 : 8; uint64_t ts : 48; /* W2 */ From 47a1208776d7f57170e9061465380777a7722eb6 Mon Sep 17 00:00:00 2001 From: Subbaraya Sundeep Date: Sat, 25 Oct 2025 16:02:46 +0530 Subject: [PATCH 0397/1381] octeontx2-af: Display new bandwidth profiles too in debugfs Consider the new profiles of cn20k too while displaying bandwidth profile contexts in debugfs. Signed-off-by: Subbaraya Sundeep Link: https://patch.msgid.link/1761388367-16579-11-git-send-email-sbhatta@marvell.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c index 8ab82700e826..7370812ece2a 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c @@ -2121,7 +2121,9 @@ static void print_nix_cn10k_rq_ctx(struct seq_file *m, seq_printf(m, "W1: ipsecd_drop_ena \t\t%d\nW1: chi_ena \t\t\t%d\n\n", rq_ctx->ipsecd_drop_ena, rq_ctx->chi_ena); - seq_printf(m, "W2: band_prof_id \t\t%d\n", rq_ctx->band_prof_id); + seq_printf(m, "W2: band_prof_id \t\t%d\n", + (u16)rq_ctx->band_prof_id_h << 10 | rq_ctx->band_prof_id); + seq_printf(m, "W2: policer_ena \t\t%d\n", rq_ctx->policer_ena); seq_printf(m, "W2: spb_sizem1 \t\t\t%d\n", rq_ctx->spb_sizem1); seq_printf(m, "W2: wqe_skip \t\t\t%d\nW2: sqb_ena \t\t\t%d\n", @@ -2639,7 +2641,10 @@ static void print_band_prof_ctx(struct seq_file *m, (prof->rc_action == 1) ? "DROP" : "RED"; seq_printf(m, "W1: rc_action\t\t%s\n", str); seq_printf(m, "W1: meter_algo\t\t%d\n", prof->meter_algo); - seq_printf(m, "W1: band_prof_id\t%d\n", prof->band_prof_id); + + seq_printf(m, "W1: band_prof_id\t%d\n", + (u16)prof->band_prof_id_h << 7 | prof->band_prof_id); + seq_printf(m, "W1: hl_en\t\t%d\n", prof->hl_en); seq_printf(m, "W2: ts\t\t\t%lld\n", (u64)prof->ts); From 33d8a1f45729b972e5e633a4296507be2e1866c3 Mon Sep 17 00:00:00 2001 From: Subbaraya Sundeep Date: Sat, 25 Oct 2025 16:02:47 +0530 Subject: [PATCH 0398/1381] octeontx2-pf: Use new bandwidth profiles in receive queue Receive queue points to a bandwidth profile for rate limiting. Since cn20k has additional bandwidth profiles use them too while mapping receive queue to bandwidth profile. Signed-off-by: Subbaraya Sundeep Link: https://patch.msgid.link/1761388367-16579-12-git-send-email-sbhatta@marvell.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/marvell/octeontx2/nic/cn10k.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/cn10k.c b/drivers/net/ethernet/marvell/octeontx2/nic/cn10k.c index cab157aac251..3e1bf22cba69 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/cn10k.c +++ b/drivers/net/ethernet/marvell/octeontx2/nic/cn10k.c @@ -341,6 +341,12 @@ int cn10k_map_unmap_rq_policer(struct otx2_nic *pfvf, int rq_idx, aq->rq.band_prof_id = policer; aq->rq_mask.band_prof_id = GENMASK(9, 0); + /* If policer id is greater than 1023 then it implies hardware supports + * more leaf profiles. In that case use band_prof_id_h for 4 MSBs. + */ + aq->rq.band_prof_id_h = policer >> 10; + aq->rq_mask.band_prof_id_h = GENMASK(3, 0); + /* Fill AQ info */ aq->qidx = rq_idx; aq->ctype = NIX_AQ_CTYPE_RQ; From 320d80eeb22219bc1dd14780113c36e11dad7c04 Mon Sep 17 00:00:00 2001 From: Fernando Fernandez Mancera Date: Mon, 13 Oct 2025 10:59:48 +0200 Subject: [PATCH 0399/1381] netfilter: nf_tables: use C99 struct initializer for nft_set_iter Use C99 struct initializer for nft_set_iter, simplifying the code and preventing future errors due to uninitialized fields if new fields are added to the struct. Signed-off-by: Fernando Fernandez Mancera Signed-off-by: Florian Westphal --- net/netfilter/nf_tables_api.c | 34 ++++++++++++++++------------------ net/netfilter/nft_lookup.c | 13 +++++-------- 2 files changed, 21 insertions(+), 26 deletions(-) diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index eed434e0a970..f3de2f9bbebf 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -5770,7 +5770,11 @@ int nf_tables_bind_set(const struct nft_ctx *ctx, struct nft_set *set, struct nft_set_binding *binding) { struct nft_set_binding *i; - struct nft_set_iter iter; + struct nft_set_iter iter = { + .genmask = nft_genmask_next(ctx->net), + .type = NFT_ITER_UPDATE, + .fn = nf_tables_bind_check_setelem, + }; if (!list_empty(&set->bindings) && nft_set_is_anonymous(set)) return -EBUSY; @@ -5785,13 +5789,6 @@ int nf_tables_bind_set(const struct nft_ctx *ctx, struct nft_set *set, goto bind; } - iter.genmask = nft_genmask_next(ctx->net); - iter.type = NFT_ITER_UPDATE; - iter.skip = 0; - iter.count = 0; - iter.err = 0; - iter.fn = nf_tables_bind_check_setelem; - set->ops->walk(ctx, set, &iter); if (!iter.err) iter.err = nft_set_catchall_bind_check(ctx, set); @@ -6195,7 +6192,17 @@ static int nf_tables_dump_set(struct sk_buff *skb, struct netlink_callback *cb) struct nftables_pernet *nft_net; struct nft_table *table; struct nft_set *set; - struct nft_set_dump_args args; + struct nft_set_dump_args args = { + .cb = cb, + .skb = skb, + .reset = dump_ctx->reset, + .iter = { + .genmask = nft_genmask_cur(net), + .type = NFT_ITER_READ, + .skip = cb->args[0], + .fn = nf_tables_dump_setelem, + }, + }; bool set_found = false; struct nlmsghdr *nlh; struct nlattr *nest; @@ -6246,15 +6253,6 @@ static int nf_tables_dump_set(struct sk_buff *skb, struct netlink_callback *cb) if (nest == NULL) goto nla_put_failure; - args.cb = cb; - args.skb = skb; - args.reset = dump_ctx->reset; - args.iter.genmask = nft_genmask_cur(net); - args.iter.type = NFT_ITER_READ; - args.iter.skip = cb->args[0]; - args.iter.count = 0; - args.iter.err = 0; - args.iter.fn = nf_tables_dump_setelem; set->ops->walk(&dump_ctx->ctx, set, &args.iter); if (!args.iter.err && args.iter.count == cb->args[0]) diff --git a/net/netfilter/nft_lookup.c b/net/netfilter/nft_lookup.c index 58c5b14889c4..fc2d7c5d83c8 100644 --- a/net/netfilter/nft_lookup.c +++ b/net/netfilter/nft_lookup.c @@ -246,19 +246,16 @@ static int nft_lookup_validate(const struct nft_ctx *ctx, const struct nft_expr *expr) { const struct nft_lookup *priv = nft_expr_priv(expr); - struct nft_set_iter iter; + struct nft_set_iter iter = { + .genmask = nft_genmask_next(ctx->net), + .type = NFT_ITER_UPDATE, + .fn = nft_setelem_validate, + }; if (!(priv->set->flags & NFT_SET_MAP) || priv->set->dtype != NFT_DATA_VERDICT) return 0; - iter.genmask = nft_genmask_next(ctx->net); - iter.type = NFT_ITER_UPDATE; - iter.skip = 0; - iter.count = 0; - iter.err = 0; - iter.fn = nft_setelem_validate; - priv->set->ops->walk(ctx, priv->set, &iter); if (!iter.err) iter.err = nft_set_catchall_validate(ctx, priv->set); From 2b749f257645c54f8659bddbdb5b2ede999bec00 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Sun, 21 Sep 2025 17:45:30 +0200 Subject: [PATCH 0400/1381] netfilter: conntrack: disable 0 value for conntrack_max setting Undocumented historical artifact inherited from ip_conntrack. If value is 0, then no limit is applied at all, conntrack table can grow to huge value, only limited by size of conntrack hashes and the kernel-internal upper limit on the hash chain lengths. This feature makes no sense; users can just set conntrack_max=2147483647 (INT_MAX). Disallow a 0 value. This will make it slightly easier to allow per-netns constraints for this value in a future patch. Signed-off-by: Florian Westphal --- net/netfilter/nf_conntrack_core.c | 2 +- net/netfilter/nf_conntrack_standalone.c | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index 344f88295976..0b95f226f211 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -1668,7 +1668,7 @@ __nf_conntrack_alloc(struct net *net, /* We don't want any race condition at early drop stage */ ct_count = atomic_inc_return(&cnet->count); - if (nf_conntrack_max && unlikely(ct_count > nf_conntrack_max)) { + if (unlikely(ct_count > nf_conntrack_max)) { if (!early_drop(net, hash)) { if (!conntrack_gc_work.early_drop) conntrack_gc_work.early_drop = true; diff --git a/net/netfilter/nf_conntrack_standalone.c b/net/netfilter/nf_conntrack_standalone.c index 708b79380f04..207b240b14e5 100644 --- a/net/netfilter/nf_conntrack_standalone.c +++ b/net/netfilter/nf_conntrack_standalone.c @@ -648,7 +648,7 @@ static struct ctl_table nf_ct_sysctl_table[] = { .maxlen = sizeof(int), .mode = 0644, .proc_handler = proc_dointvec_minmax, - .extra1 = SYSCTL_ZERO, + .extra1 = SYSCTL_ONE, .extra2 = SYSCTL_INT_MAX, }, [NF_SYSCTL_CT_COUNT] = { @@ -929,7 +929,7 @@ static struct ctl_table nf_ct_netfilter_table[] = { .maxlen = sizeof(int), .mode = 0644, .proc_handler = proc_dointvec_minmax, - .extra1 = SYSCTL_ZERO, + .extra1 = SYSCTL_ONE, .extra2 = SYSCTL_INT_MAX, }, }; From 57347d58a4011551e7d0e030f2f12e4d1a28feb6 Mon Sep 17 00:00:00 2001 From: "caivive (Weibiao Tu)" Date: Thu, 28 Nov 2024 20:52:04 +0800 Subject: [PATCH 0401/1381] netfilter: fix typo in nf_conntrack_l4proto.h comment In the comment for nf_conntrack_l4proto.h, the word "nfnetink" was incorrectly spelled. It has been corrected to "nfnetlink". Fixes a typo to enhance readability and ensure consistency. Signed-off-by: caivive (Weibiao Tu) Signed-off-by: Florian Westphal --- include/net/netfilter/nf_conntrack_l4proto.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/net/netfilter/nf_conntrack_l4proto.h b/include/net/netfilter/nf_conntrack_l4proto.h index 6929f8daf1ed..cd5020835a6d 100644 --- a/include/net/netfilter/nf_conntrack_l4proto.h +++ b/include/net/netfilter/nf_conntrack_l4proto.h @@ -30,7 +30,7 @@ struct nf_conntrack_l4proto { /* called by gc worker if table is full */ bool (*can_early_drop)(const struct nf_conn *ct); - /* convert protoinfo to nfnetink attributes */ + /* convert protoinfo to nfnetlink attributes */ int (*to_nlattr)(struct sk_buff *skb, struct nlattr *nla, struct nf_conn *ct, bool destroy); From aef3cdb47bbbef9fea9512ed6c02d64394449d53 Mon Sep 17 00:00:00 2001 From: Halil Pasic Date: Mon, 27 Oct 2025 23:48:55 +0100 Subject: [PATCH 0402/1381] net/smc: make wr buffer count configurable Think SMC_WR_BUF_CNT_SEND := SMC_WR_BUF_CNT used in send context and SMC_WR_BUF_CNT_RECV := 3 * SMC_WR_BUF_CNT used in recv context. Those get replaced with lgr->max_send_wr and lgr->max_recv_wr respective. Please note that although with the default sysctl values qp_attr.cap.max_send_wr == qp_attr.cap.max_recv_wr is maintained but can not be assumed to be generally true any more. I see no downside to that, but my confidence level is rather modest. Signed-off-by: Halil Pasic Reviewed-by: Sidraya Jayagond Reviewed-by: Dust Li Tested-by: Mahanta Jambigi Link: https://patch.msgid.link/20251027224856.2970019-2-pasic@linux.ibm.com Signed-off-by: Paolo Abeni --- Documentation/networking/smc-sysctl.rst | 36 +++++++++++++++++++++++++ include/net/netns/smc.h | 2 ++ net/smc/smc_core.h | 6 +++++ net/smc/smc_ib.c | 10 +++---- net/smc/smc_llc.c | 2 ++ net/smc/smc_sysctl.c | 22 +++++++++++++++ net/smc/smc_sysctl.h | 2 ++ net/smc/smc_wr.c | 31 ++++++++++----------- net/smc/smc_wr.h | 2 -- 9 files changed, 91 insertions(+), 22 deletions(-) diff --git a/Documentation/networking/smc-sysctl.rst b/Documentation/networking/smc-sysctl.rst index a874d007f2db..337ac2be167e 100644 --- a/Documentation/networking/smc-sysctl.rst +++ b/Documentation/networking/smc-sysctl.rst @@ -71,3 +71,39 @@ smcr_max_conns_per_lgr - INTEGER acceptable value ranges from 16 to 255. Only for SMC-R v2.1 and later. Default: 255 + +smcr_max_send_wr - INTEGER + So-called work request buffers are SMCR link (and RDMA queue pair) level + resources necessary for performing RDMA operations. Since up to 255 + connections can share a link group and thus also a link and the number + of the work request buffers is decided when the link is allocated, + depending on the workload it can be a bottleneck in a sense that threads + have to wait for work request buffers to become available. Before the + introduction of this control the maximal number of work request buffers + available on the send path used to be hard coded to 16. With this control + it becomes configurable. The acceptable range is between 2 and 2048. + + Please be aware that all the buffers need to be allocated as a physically + continuous array in which each element is a single buffer and has the size + of SMC_WR_BUF_SIZE (48) bytes. If the allocation fails we give up much + like before having this control. + + Default: 16 + +smcr_max_recv_wr - INTEGER + So-called work request buffers are SMCR link (and RDMA queue pair) level + resources necessary for performing RDMA operations. Since up to 255 + connections can share a link group and thus also a link and the number + of the work request buffers is decided when the link is allocated, + depending on the workload it can be a bottleneck in a sense that threads + have to wait for work request buffers to become available. Before the + introduction of this control the maximal number of work request buffers + available on the receive path used to be hard coded to 16. With this control + it becomes configurable. The acceptable range is between 2 and 2048. + + Please be aware that all the buffers need to be allocated as a physically + continuous array in which each element is a single buffer and has the size + of SMC_WR_BUF_SIZE (48) bytes. If the allocation fails we give up much + like before having this control. + + Default: 48 diff --git a/include/net/netns/smc.h b/include/net/netns/smc.h index fc752a50f91b..6ceb12baec24 100644 --- a/include/net/netns/smc.h +++ b/include/net/netns/smc.h @@ -24,5 +24,7 @@ struct netns_smc { int sysctl_rmem; int sysctl_max_links_per_lgr; int sysctl_max_conns_per_lgr; + unsigned int sysctl_smcr_max_send_wr; + unsigned int sysctl_smcr_max_recv_wr; }; #endif diff --git a/net/smc/smc_core.h b/net/smc/smc_core.h index a5a78cbff341..8d06c8bb14e9 100644 --- a/net/smc/smc_core.h +++ b/net/smc/smc_core.h @@ -34,6 +34,8 @@ * distributions may modify it to a value between * 16-255 as needed. */ +#define SMCR_MAX_SEND_WR_DEF 16 /* Default number of work requests per send queue */ +#define SMCR_MAX_RECV_WR_DEF 48 /* Default number of work requests per recv queue */ struct smc_lgr_list { /* list of link group definition */ struct list_head list; @@ -366,6 +368,10 @@ struct smc_link_group { /* max conn can be assigned to lgr */ u8 max_links; /* max links can be added in lgr */ + u16 max_send_wr; + /* number of WR buffers on send */ + u16 max_recv_wr; + /* number of WR buffers on recv */ }; struct { /* SMC-D */ struct smcd_gid peer_gid; diff --git a/net/smc/smc_ib.c b/net/smc/smc_ib.c index 0052f02756eb..1154907c5c05 100644 --- a/net/smc/smc_ib.c +++ b/net/smc/smc_ib.c @@ -669,11 +669,6 @@ int smc_ib_create_queue_pair(struct smc_link *lnk) .recv_cq = lnk->smcibdev->roce_cq_recv, .srq = NULL, .cap = { - /* include unsolicited rdma_writes as well, - * there are max. 2 RDMA_WRITE per 1 WR_SEND - */ - .max_send_wr = SMC_WR_BUF_CNT * 3, - .max_recv_wr = SMC_WR_BUF_CNT * 3, .max_send_sge = SMC_IB_MAX_SEND_SGE, .max_recv_sge = lnk->wr_rx_sge_cnt, .max_inline_data = 0, @@ -683,6 +678,11 @@ int smc_ib_create_queue_pair(struct smc_link *lnk) }; int rc; + /* include unsolicited rdma_writes as well, + * there are max. 2 RDMA_WRITE per 1 WR_SEND + */ + qp_attr.cap.max_send_wr = 3 * lnk->lgr->max_send_wr; + qp_attr.cap.max_recv_wr = lnk->lgr->max_recv_wr; lnk->roce_qp = ib_create_qp(lnk->roce_pd, &qp_attr); rc = PTR_ERR_OR_ZERO(lnk->roce_qp); if (IS_ERR(lnk->roce_qp)) diff --git a/net/smc/smc_llc.c b/net/smc/smc_llc.c index f865c58c3aa7..f5d5eb617526 100644 --- a/net/smc/smc_llc.c +++ b/net/smc/smc_llc.c @@ -2157,6 +2157,8 @@ void smc_llc_lgr_init(struct smc_link_group *lgr, struct smc_sock *smc) init_waitqueue_head(&lgr->llc_msg_waiter); init_rwsem(&lgr->llc_conf_mutex); lgr->llc_testlink_time = READ_ONCE(net->smc.sysctl_smcr_testlink_time); + lgr->max_send_wr = (u16)(READ_ONCE(net->smc.sysctl_smcr_max_send_wr)); + lgr->max_recv_wr = (u16)(READ_ONCE(net->smc.sysctl_smcr_max_recv_wr)); } /* called after lgr was removed from lgr_list */ diff --git a/net/smc/smc_sysctl.c b/net/smc/smc_sysctl.c index 2fab6456f765..7b2471904d04 100644 --- a/net/smc/smc_sysctl.c +++ b/net/smc/smc_sysctl.c @@ -29,6 +29,8 @@ static int links_per_lgr_min = SMC_LINKS_ADD_LNK_MIN; static int links_per_lgr_max = SMC_LINKS_ADD_LNK_MAX; static int conns_per_lgr_min = SMC_CONN_PER_LGR_MIN; static int conns_per_lgr_max = SMC_CONN_PER_LGR_MAX; +static unsigned int smcr_max_wr_min = 2; +static unsigned int smcr_max_wr_max = 2048; static struct ctl_table smc_table[] = { { @@ -99,6 +101,24 @@ static struct ctl_table smc_table[] = { .extra1 = SYSCTL_ZERO, .extra2 = SYSCTL_ONE, }, + { + .procname = "smcr_max_send_wr", + .data = &init_net.smc.sysctl_smcr_max_send_wr, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = &smcr_max_wr_min, + .extra2 = &smcr_max_wr_max, + }, + { + .procname = "smcr_max_recv_wr", + .data = &init_net.smc.sysctl_smcr_max_recv_wr, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = &smcr_max_wr_min, + .extra2 = &smcr_max_wr_max, + }, }; int __net_init smc_sysctl_net_init(struct net *net) @@ -130,6 +150,8 @@ int __net_init smc_sysctl_net_init(struct net *net) WRITE_ONCE(net->smc.sysctl_rmem, net_smc_rmem_init); net->smc.sysctl_max_links_per_lgr = SMC_LINKS_PER_LGR_MAX_PREFER; net->smc.sysctl_max_conns_per_lgr = SMC_CONN_PER_LGR_PREFER; + net->smc.sysctl_smcr_max_send_wr = SMCR_MAX_SEND_WR_DEF; + net->smc.sysctl_smcr_max_recv_wr = SMCR_MAX_RECV_WR_DEF; /* disable handshake limitation by default */ net->smc.limit_smc_hs = 0; diff --git a/net/smc/smc_sysctl.h b/net/smc/smc_sysctl.h index eb2465ae1e15..8538915af7af 100644 --- a/net/smc/smc_sysctl.h +++ b/net/smc/smc_sysctl.h @@ -25,6 +25,8 @@ static inline int smc_sysctl_net_init(struct net *net) net->smc.sysctl_autocorking_size = SMC_AUTOCORKING_DEFAULT_SIZE; net->smc.sysctl_max_links_per_lgr = SMC_LINKS_PER_LGR_MAX_PREFER; net->smc.sysctl_max_conns_per_lgr = SMC_CONN_PER_LGR_PREFER; + net->smc.sysctl_smcr_max_send_wr = SMCR_MAX_SEND_WR_DEF; + net->smc.sysctl_smcr_max_recv_wr = SMCR_MAX_RECV_WR_DEF; return 0; } diff --git a/net/smc/smc_wr.c b/net/smc/smc_wr.c index b04a21b8c511..883fb0f1ce43 100644 --- a/net/smc/smc_wr.c +++ b/net/smc/smc_wr.c @@ -547,9 +547,9 @@ void smc_wr_remember_qp_attr(struct smc_link *lnk) IB_QP_DEST_QPN, &init_attr); - lnk->wr_tx_cnt = min_t(size_t, SMC_WR_BUF_CNT, + lnk->wr_tx_cnt = min_t(size_t, lnk->lgr->max_send_wr, lnk->qp_attr.cap.max_send_wr); - lnk->wr_rx_cnt = min_t(size_t, SMC_WR_BUF_CNT * 3, + lnk->wr_rx_cnt = min_t(size_t, lnk->lgr->max_recv_wr, lnk->qp_attr.cap.max_recv_wr); } @@ -741,50 +741,51 @@ int smc_wr_alloc_lgr_mem(struct smc_link_group *lgr) int smc_wr_alloc_link_mem(struct smc_link *link) { /* allocate link related memory */ - link->wr_tx_bufs = kcalloc(SMC_WR_BUF_CNT, SMC_WR_BUF_SIZE, GFP_KERNEL); + link->wr_tx_bufs = kcalloc(link->lgr->max_send_wr, + SMC_WR_BUF_SIZE, GFP_KERNEL); if (!link->wr_tx_bufs) goto no_mem; - link->wr_rx_bufs = kcalloc(SMC_WR_BUF_CNT * 3, link->wr_rx_buflen, + link->wr_rx_bufs = kcalloc(link->lgr->max_recv_wr, link->wr_rx_buflen, GFP_KERNEL); if (!link->wr_rx_bufs) goto no_mem_wr_tx_bufs; - link->wr_tx_ibs = kcalloc(SMC_WR_BUF_CNT, sizeof(link->wr_tx_ibs[0]), - GFP_KERNEL); + link->wr_tx_ibs = kcalloc(link->lgr->max_send_wr, + sizeof(link->wr_tx_ibs[0]), GFP_KERNEL); if (!link->wr_tx_ibs) goto no_mem_wr_rx_bufs; - link->wr_rx_ibs = kcalloc(SMC_WR_BUF_CNT * 3, + link->wr_rx_ibs = kcalloc(link->lgr->max_recv_wr, sizeof(link->wr_rx_ibs[0]), GFP_KERNEL); if (!link->wr_rx_ibs) goto no_mem_wr_tx_ibs; - link->wr_tx_rdmas = kcalloc(SMC_WR_BUF_CNT, + link->wr_tx_rdmas = kcalloc(link->lgr->max_send_wr, sizeof(link->wr_tx_rdmas[0]), GFP_KERNEL); if (!link->wr_tx_rdmas) goto no_mem_wr_rx_ibs; - link->wr_tx_rdma_sges = kcalloc(SMC_WR_BUF_CNT, + link->wr_tx_rdma_sges = kcalloc(link->lgr->max_send_wr, sizeof(link->wr_tx_rdma_sges[0]), GFP_KERNEL); if (!link->wr_tx_rdma_sges) goto no_mem_wr_tx_rdmas; - link->wr_tx_sges = kcalloc(SMC_WR_BUF_CNT, sizeof(link->wr_tx_sges[0]), + link->wr_tx_sges = kcalloc(link->lgr->max_send_wr, sizeof(link->wr_tx_sges[0]), GFP_KERNEL); if (!link->wr_tx_sges) goto no_mem_wr_tx_rdma_sges; - link->wr_rx_sges = kcalloc(SMC_WR_BUF_CNT * 3, + link->wr_rx_sges = kcalloc(link->lgr->max_recv_wr, sizeof(link->wr_rx_sges[0]) * link->wr_rx_sge_cnt, GFP_KERNEL); if (!link->wr_rx_sges) goto no_mem_wr_tx_sges; - link->wr_tx_mask = bitmap_zalloc(SMC_WR_BUF_CNT, GFP_KERNEL); + link->wr_tx_mask = bitmap_zalloc(link->lgr->max_send_wr, GFP_KERNEL); if (!link->wr_tx_mask) goto no_mem_wr_rx_sges; - link->wr_tx_pends = kcalloc(SMC_WR_BUF_CNT, + link->wr_tx_pends = kcalloc(link->lgr->max_send_wr, sizeof(link->wr_tx_pends[0]), GFP_KERNEL); if (!link->wr_tx_pends) goto no_mem_wr_tx_mask; - link->wr_tx_compl = kcalloc(SMC_WR_BUF_CNT, + link->wr_tx_compl = kcalloc(link->lgr->max_send_wr, sizeof(link->wr_tx_compl[0]), GFP_KERNEL); if (!link->wr_tx_compl) @@ -905,7 +906,7 @@ int smc_wr_create_link(struct smc_link *lnk) goto dma_unmap; } smc_wr_init_sge(lnk); - bitmap_zero(lnk->wr_tx_mask, SMC_WR_BUF_CNT); + bitmap_zero(lnk->wr_tx_mask, lnk->lgr->max_send_wr); init_waitqueue_head(&lnk->wr_tx_wait); rc = percpu_ref_init(&lnk->wr_tx_refs, smcr_wr_tx_refs_free, 0, GFP_KERNEL); if (rc) diff --git a/net/smc/smc_wr.h b/net/smc/smc_wr.h index f3008dda222a..aa4533af9122 100644 --- a/net/smc/smc_wr.h +++ b/net/smc/smc_wr.h @@ -19,8 +19,6 @@ #include "smc.h" #include "smc_core.h" -#define SMC_WR_BUF_CNT 16 /* # of ctrl buffers per link */ - #define SMC_WR_TX_WAIT_FREE_SLOT_TIME (10 * HZ) #define SMC_WR_TX_SIZE 44 /* actual size of wr_send data (<=SMC_WR_BUF_SIZE) */ From 8f736087e52f1cd4234b871c137f5a3eb0d2741a Mon Sep 17 00:00:00 2001 From: Halil Pasic Date: Mon, 27 Oct 2025 23:48:56 +0100 Subject: [PATCH 0403/1381] net/smc: handle -ENOMEM from smc_wr_alloc_link_mem gracefully Currently if a -ENOMEM from smc_wr_alloc_link_mem() is handled by giving up and going the way of a TCP fallback. This was reasonable before the sizes of the allocations there were compile time constants and reasonably small. But now those are actually configurable. So instead of giving up, keep retrying with half of the requested size unless we dip below the old static sizes -- then give up! In terms of numbers that means we give up when it is certain that we at best would end up allocating less than 16 send WR buffers or less than 48 recv WR buffers. This is to avoid regressions due to having fewer buffers compared the static values of the past. Please note that SMC-R is supposed to be an optimisation over TCP, and falling back to TCP is superior to establishing an SMC connection that is going to perform worse. If the memory allocation fails (and we propagate -ENOMEM), we fall back to TCP. Preserve (modulo truncation) the ratio of send/recv WR buffer counts. Signed-off-by: Halil Pasic Reviewed-by: Wenjia Zhang Reviewed-by: Mahanta Jambigi Reviewed-by: Sidraya Jayagond Reviewed-by: Dust Li Tested-by: Mahanta Jambigi Link: https://patch.msgid.link/20251027224856.2970019-3-pasic@linux.ibm.com Signed-off-by: Paolo Abeni --- Documentation/networking/smc-sysctl.rst | 8 ++++-- net/smc/smc_core.c | 34 +++++++++++++++++-------- net/smc/smc_core.h | 2 ++ net/smc/smc_wr.c | 28 ++++++++++---------- 4 files changed, 46 insertions(+), 26 deletions(-) diff --git a/Documentation/networking/smc-sysctl.rst b/Documentation/networking/smc-sysctl.rst index 337ac2be167e..904a910f198e 100644 --- a/Documentation/networking/smc-sysctl.rst +++ b/Documentation/networking/smc-sysctl.rst @@ -85,7 +85,9 @@ smcr_max_send_wr - INTEGER Please be aware that all the buffers need to be allocated as a physically continuous array in which each element is a single buffer and has the size - of SMC_WR_BUF_SIZE (48) bytes. If the allocation fails we give up much + of SMC_WR_BUF_SIZE (48) bytes. If the allocation fails, we keep retrying + with half of the buffer count until it is ether successful or (unlikely) + we dip below the old hard coded value which is 16 where we give up much like before having this control. Default: 16 @@ -103,7 +105,9 @@ smcr_max_recv_wr - INTEGER Please be aware that all the buffers need to be allocated as a physically continuous array in which each element is a single buffer and has the size - of SMC_WR_BUF_SIZE (48) bytes. If the allocation fails we give up much + of SMC_WR_BUF_SIZE (48) bytes. If the allocation fails, we keep retrying + with half of the buffer count until it is ether successful or (unlikely) + we dip below the old hard coded value which is 16 where we give up much like before having this control. Default: 48 diff --git a/net/smc/smc_core.c b/net/smc/smc_core.c index be0c2da83d2b..e4eabc83719e 100644 --- a/net/smc/smc_core.c +++ b/net/smc/smc_core.c @@ -810,6 +810,8 @@ int smcr_link_init(struct smc_link_group *lgr, struct smc_link *lnk, lnk->clearing = 0; lnk->path_mtu = lnk->smcibdev->pattr[lnk->ibport - 1].active_mtu; lnk->link_id = smcr_next_link_id(lgr); + lnk->max_send_wr = lgr->max_send_wr; + lnk->max_recv_wr = lgr->max_recv_wr; lnk->lgr = lgr; smc_lgr_hold(lgr); /* lgr_put in smcr_link_clear() */ lnk->link_idx = link_idx; @@ -836,27 +838,39 @@ int smcr_link_init(struct smc_link_group *lgr, struct smc_link *lnk, rc = smc_llc_link_init(lnk); if (rc) goto out; - rc = smc_wr_alloc_link_mem(lnk); - if (rc) - goto clear_llc_lnk; rc = smc_ib_create_protection_domain(lnk); if (rc) - goto free_link_mem; - rc = smc_ib_create_queue_pair(lnk); - if (rc) - goto dealloc_pd; + goto clear_llc_lnk; + do { + rc = smc_ib_create_queue_pair(lnk); + if (rc) + goto dealloc_pd; + rc = smc_wr_alloc_link_mem(lnk); + if (!rc) + break; + else if (rc != -ENOMEM) /* give up */ + goto destroy_qp; + /* retry with smaller ... */ + lnk->max_send_wr /= 2; + lnk->max_recv_wr /= 2; + /* ... unless droping below old SMC_WR_BUF_SIZE */ + if (lnk->max_send_wr < 16 || lnk->max_recv_wr < 48) + goto destroy_qp; + smc_ib_destroy_queue_pair(lnk); + } while (1); + rc = smc_wr_create_link(lnk); if (rc) - goto destroy_qp; + goto free_link_mem; lnk->state = SMC_LNK_ACTIVATING; return 0; +free_link_mem: + smc_wr_free_link_mem(lnk); destroy_qp: smc_ib_destroy_queue_pair(lnk); dealloc_pd: smc_ib_dealloc_protection_domain(lnk); -free_link_mem: - smc_wr_free_link_mem(lnk); clear_llc_lnk: smc_llc_link_clear(lnk, false); out: diff --git a/net/smc/smc_core.h b/net/smc/smc_core.h index 8d06c8bb14e9..5c18f08a4c8a 100644 --- a/net/smc/smc_core.h +++ b/net/smc/smc_core.h @@ -175,6 +175,8 @@ struct smc_link { struct completion llc_testlink_resp; /* wait for rx of testlink */ int llc_testlink_time; /* testlink interval */ atomic_t conn_cnt; /* connections on this link */ + u16 max_send_wr; + u16 max_recv_wr; }; /* For now we just allow one parallel link per link group. The SMC protocol diff --git a/net/smc/smc_wr.c b/net/smc/smc_wr.c index 883fb0f1ce43..5feafa98ab1a 100644 --- a/net/smc/smc_wr.c +++ b/net/smc/smc_wr.c @@ -547,9 +547,9 @@ void smc_wr_remember_qp_attr(struct smc_link *lnk) IB_QP_DEST_QPN, &init_attr); - lnk->wr_tx_cnt = min_t(size_t, lnk->lgr->max_send_wr, + lnk->wr_tx_cnt = min_t(size_t, lnk->max_send_wr, lnk->qp_attr.cap.max_send_wr); - lnk->wr_rx_cnt = min_t(size_t, lnk->lgr->max_recv_wr, + lnk->wr_rx_cnt = min_t(size_t, lnk->max_recv_wr, lnk->qp_attr.cap.max_recv_wr); } @@ -741,51 +741,51 @@ int smc_wr_alloc_lgr_mem(struct smc_link_group *lgr) int smc_wr_alloc_link_mem(struct smc_link *link) { /* allocate link related memory */ - link->wr_tx_bufs = kcalloc(link->lgr->max_send_wr, + link->wr_tx_bufs = kcalloc(link->max_send_wr, SMC_WR_BUF_SIZE, GFP_KERNEL); if (!link->wr_tx_bufs) goto no_mem; - link->wr_rx_bufs = kcalloc(link->lgr->max_recv_wr, link->wr_rx_buflen, + link->wr_rx_bufs = kcalloc(link->max_recv_wr, link->wr_rx_buflen, GFP_KERNEL); if (!link->wr_rx_bufs) goto no_mem_wr_tx_bufs; - link->wr_tx_ibs = kcalloc(link->lgr->max_send_wr, + link->wr_tx_ibs = kcalloc(link->max_send_wr, sizeof(link->wr_tx_ibs[0]), GFP_KERNEL); if (!link->wr_tx_ibs) goto no_mem_wr_rx_bufs; - link->wr_rx_ibs = kcalloc(link->lgr->max_recv_wr, + link->wr_rx_ibs = kcalloc(link->max_recv_wr, sizeof(link->wr_rx_ibs[0]), GFP_KERNEL); if (!link->wr_rx_ibs) goto no_mem_wr_tx_ibs; - link->wr_tx_rdmas = kcalloc(link->lgr->max_send_wr, + link->wr_tx_rdmas = kcalloc(link->max_send_wr, sizeof(link->wr_tx_rdmas[0]), GFP_KERNEL); if (!link->wr_tx_rdmas) goto no_mem_wr_rx_ibs; - link->wr_tx_rdma_sges = kcalloc(link->lgr->max_send_wr, + link->wr_tx_rdma_sges = kcalloc(link->max_send_wr, sizeof(link->wr_tx_rdma_sges[0]), GFP_KERNEL); if (!link->wr_tx_rdma_sges) goto no_mem_wr_tx_rdmas; - link->wr_tx_sges = kcalloc(link->lgr->max_send_wr, sizeof(link->wr_tx_sges[0]), + link->wr_tx_sges = kcalloc(link->max_send_wr, sizeof(link->wr_tx_sges[0]), GFP_KERNEL); if (!link->wr_tx_sges) goto no_mem_wr_tx_rdma_sges; - link->wr_rx_sges = kcalloc(link->lgr->max_recv_wr, + link->wr_rx_sges = kcalloc(link->max_recv_wr, sizeof(link->wr_rx_sges[0]) * link->wr_rx_sge_cnt, GFP_KERNEL); if (!link->wr_rx_sges) goto no_mem_wr_tx_sges; - link->wr_tx_mask = bitmap_zalloc(link->lgr->max_send_wr, GFP_KERNEL); + link->wr_tx_mask = bitmap_zalloc(link->max_send_wr, GFP_KERNEL); if (!link->wr_tx_mask) goto no_mem_wr_rx_sges; - link->wr_tx_pends = kcalloc(link->lgr->max_send_wr, + link->wr_tx_pends = kcalloc(link->max_send_wr, sizeof(link->wr_tx_pends[0]), GFP_KERNEL); if (!link->wr_tx_pends) goto no_mem_wr_tx_mask; - link->wr_tx_compl = kcalloc(link->lgr->max_send_wr, + link->wr_tx_compl = kcalloc(link->max_send_wr, sizeof(link->wr_tx_compl[0]), GFP_KERNEL); if (!link->wr_tx_compl) @@ -906,7 +906,7 @@ int smc_wr_create_link(struct smc_link *lnk) goto dma_unmap; } smc_wr_init_sge(lnk); - bitmap_zero(lnk->wr_tx_mask, lnk->lgr->max_send_wr); + bitmap_zero(lnk->wr_tx_mask, lnk->max_send_wr); init_waitqueue_head(&lnk->wr_tx_wait); rc = percpu_ref_init(&lnk->wr_tx_refs, smcr_wr_tx_refs_free, 0, GFP_KERNEL); if (rc) From be5febd51c478bc8e24ad3480435f2754a403b14 Mon Sep 17 00:00:00 2001 From: Abdun Nihaal Date: Tue, 28 Oct 2025 22:34:55 +0530 Subject: [PATCH 0404/1381] wifi: ath12k: fix potential memory leak in ath12k_wow_arp_ns_offload() When the call to ath12k_wmi_arp_ns_offload() fails, the temporary memory allocation for offload is not freed before returning. Fix that by freeing offload in the error path. Fixes: 1666108c74c4 ("wifi: ath12k: support ARP and NS offload") Signed-off-by: Abdun Nihaal Reviewed-by: Baochen Qiang Link: https://patch.msgid.link/20251028170457.134608-1-nihaal@cse.iitm.ac.in Signed-off-by: Jeff Johnson --- drivers/net/wireless/ath/ath12k/wow.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/wireless/ath/ath12k/wow.c b/drivers/net/wireless/ath/ath12k/wow.c index dce9bd0bcaef..e8481626f194 100644 --- a/drivers/net/wireless/ath/ath12k/wow.c +++ b/drivers/net/wireless/ath/ath12k/wow.c @@ -758,6 +758,7 @@ static int ath12k_wow_arp_ns_offload(struct ath12k *ar, bool enable) if (ret) { ath12k_warn(ar->ab, "failed to set arp ns offload vdev %i: enable %d, ret %d\n", arvif->vdev_id, enable, ret); + kfree(offload); return ret; } } From 00575bb44b2c2aa53d0a768de2b80c9c1af0174d Mon Sep 17 00:00:00 2001 From: Baochen Qiang Date: Wed, 29 Oct 2025 10:07:14 +0800 Subject: [PATCH 0405/1381] wifi: ath12k: fix reusing m3 memory During firmware recovery or suspend/resume, m3 memory could be reused if the size of the new m3 binary is equal to or less than that of the existing memory. There will be issues for the latter case, since m3_mem->size will be updated with a smaller value and this value is eventually used in the free path, where the original total size should be used instead. To fix it, add a new member in m3_mem_region structure to track the original memory size and use it in free path. Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.1.c5-00302-QCAHMTSWPL_V1.0_V2.0_SILICONZ-1.115823.3 Fixes: 05090ae82f44 ("wifi: ath12k: check M3 buffer size as well whey trying to reuse it") Signed-off-by: Baochen Qiang Reviewed-by: Vasanthakumar Thiagarajan Link: https://patch.msgid.link/20251029-ath12k-fix-m3-reuse-v1-1-69225bacfc5d@oss.qualcomm.com Signed-off-by: Jeff Johnson --- drivers/net/wireless/ath/ath12k/qmi.c | 11 +++++++---- drivers/net/wireless/ath/ath12k/qmi.h | 5 ++++- 2 files changed, 11 insertions(+), 5 deletions(-) diff --git a/drivers/net/wireless/ath/ath12k/qmi.c b/drivers/net/wireless/ath/ath12k/qmi.c index 36325e62aa24..8de9aee2498e 100644 --- a/drivers/net/wireless/ath/ath12k/qmi.c +++ b/drivers/net/wireless/ath/ath12k/qmi.c @@ -1,7 +1,7 @@ // SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (c) 2018-2021 The Linux Foundation. All rights reserved. - * Copyright (c) 2021-2025 Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. */ #include @@ -3114,9 +3114,10 @@ static void ath12k_qmi_m3_free(struct ath12k_base *ab) if (!m3_mem->vaddr) return; - dma_free_coherent(ab->dev, m3_mem->size, + dma_free_coherent(ab->dev, m3_mem->total_size, m3_mem->vaddr, m3_mem->paddr); m3_mem->vaddr = NULL; + m3_mem->total_size = 0; m3_mem->size = 0; } @@ -3152,7 +3153,7 @@ static int ath12k_qmi_m3_load(struct ath12k_base *ab) /* In recovery/resume cases, M3 buffer is not freed, try to reuse that */ if (m3_mem->vaddr) { - if (m3_mem->size >= m3_len) + if (m3_mem->total_size >= m3_len) goto skip_m3_alloc; /* Old buffer is too small, free and reallocate */ @@ -3164,11 +3165,13 @@ static int ath12k_qmi_m3_load(struct ath12k_base *ab) GFP_KERNEL); if (!m3_mem->vaddr) { ath12k_err(ab, "failed to allocate memory for M3 with size %zu\n", - fw->size); + m3_len); ret = -ENOMEM; goto out; } + m3_mem->total_size = m3_len; + skip_m3_alloc: memcpy(m3_mem->vaddr, m3_data, m3_len); m3_mem->size = m3_len; diff --git a/drivers/net/wireless/ath/ath12k/qmi.h b/drivers/net/wireless/ath/ath12k/qmi.h index 4767d9a2e309..7a88268aa1e9 100644 --- a/drivers/net/wireless/ath/ath12k/qmi.h +++ b/drivers/net/wireless/ath/ath12k/qmi.h @@ -1,7 +1,7 @@ /* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* * Copyright (c) 2018-2021 The Linux Foundation. All rights reserved. - * Copyright (c) 2021-2025 Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. */ #ifndef ATH12K_QMI_H @@ -120,6 +120,9 @@ struct target_info { }; struct m3_mem_region { + /* total memory allocated */ + u32 total_size; + /* actual memory being used */ u32 size; dma_addr_t paddr; void *vaddr; From 088a099690e4c0d291db505013317ab5dd58b4d5 Mon Sep 17 00:00:00 2001 From: Baochen Qiang Date: Thu, 30 Oct 2025 10:08:43 +0800 Subject: [PATCH 0406/1381] wifi: ath12k: fix error handling in creating hardware group In ath12k_core_init() when ath12k_core_hw_group_create() fails, ath12k_core_hw_group_destroy() is called where for each device below path would get executed ath12k_core_soc_destroy() ath12k_qmi_deinit_service() qmi_handle_release() This results in kernel crash in case one of the device fails at qmi_handle_init() when creating hardware group: ath12k_pci 0000:10:00.0: failed to initialize qmi handle ath12k_pci 0000:10:00.0: failed to initialize qmi :-517 ath12k_pci 0000:10:00.0: failed to create soc core: -517 ath12k_pci 0000:10:00.0: unable to create hw group BUG: unable to handle page fault for address: ffffffffffffffb7 RIP: 0010:qmi_handle_release Call Trace: ath12k_qmi_deinit_service ath12k_core_hw_group_destroy ath12k_core_init ath12k_pci_probe The detailed reason is, when qmi_handle_init() fails for a device ab->qmi.handle is not correctly initialized. Then ath12k_core_hw_group_create() returns failure, since error handing is done for all device, eventually qmi_handle_release() is called for the issue device and finally kernel crashes due to the uninitialized ab->qmi.handle. Fix this by moving error handling to ath12k_core_hw_group_create(), this way the issue device can be skipped. Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.1.c5-00284.1-QCAHMTSWPL_V1.0_V2.0_SILICONZ-3 Fixes: 6f245ea0ec6c ("wifi: ath12k: introduce device group abstraction") Link: https://lore.kernel.org/ath12k/fabc97122016d1a66a53ddedd965d134@posteo.net Reported-by: a-development Closes: https://bugzilla.kernel.org/show_bug.cgi?id=220518 Tested-by: a-development Signed-off-by: Baochen Qiang Reviewed-by: Vasanthakumar Thiagarajan Link: https://patch.msgid.link/20251030-fix-hw-group-create-err-handling-v1-1-0659e4d15fb9@oss.qualcomm.com Signed-off-by: Jeff Johnson --- drivers/net/wireless/ath/ath12k/core.c | 22 +++++++++++++++++----- 1 file changed, 17 insertions(+), 5 deletions(-) diff --git a/drivers/net/wireless/ath/ath12k/core.c b/drivers/net/wireless/ath/ath12k/core.c index 5d494c5cdc0d..a2137b363c2f 100644 --- a/drivers/net/wireless/ath/ath12k/core.c +++ b/drivers/net/wireless/ath/ath12k/core.c @@ -2106,14 +2106,27 @@ static int ath12k_core_hw_group_create(struct ath12k_hw_group *ag) ret = ath12k_core_soc_create(ab); if (ret) { mutex_unlock(&ab->core_lock); - ath12k_err(ab, "failed to create soc core: %d\n", ret); - return ret; + ath12k_err(ab, "failed to create soc %d core: %d\n", i, ret); + goto destroy; } mutex_unlock(&ab->core_lock); } return 0; + +destroy: + for (i--; i >= 0; i--) { + ab = ag->ab[i]; + if (!ab) + continue; + + mutex_lock(&ab->core_lock); + ath12k_core_soc_destroy(ab); + mutex_unlock(&ab->core_lock); + } + + return ret; } void ath12k_core_hw_group_set_mlo_capable(struct ath12k_hw_group *ag) @@ -2188,7 +2201,7 @@ int ath12k_core_init(struct ath12k_base *ab) if (ret) { mutex_unlock(&ag->mutex); ath12k_warn(ab, "unable to create hw group\n"); - goto err_destroy_hw_group; + goto err_unassign_hw_group; } } @@ -2196,8 +2209,7 @@ int ath12k_core_init(struct ath12k_base *ab) return 0; -err_destroy_hw_group: - ath12k_core_hw_group_destroy(ab->ag); +err_unassign_hw_group: ath12k_core_hw_group_unassign(ab); err_unregister_notifier: ath12k_core_panic_notifier_unregister(ab); From 770bff79424beec2edb8e7cc63b0e8d1b1a927a3 Mon Sep 17 00:00:00 2001 From: Muna Sinada Date: Thu, 23 Oct 2025 17:19:23 -0700 Subject: [PATCH 0407/1381] wifi: ath12k: generalize GI and LTF fixed rate functions Currently, functions in mac.c for setting GI and LTF rates are specifically for HE rates. Remove any mention of "HE" in such functions in order to allow for other modes to utilize the functions. The intention is to prepare for the addition of EHT GI and LTF fixed rate settings. Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1 Signed-off-by: Muna Sinada Reviewed-by: Vasanthakumar Thiagarajan Link: https://patch.msgid.link/20251024001928.257356-2-muna.sinada@oss.qualcomm.com Signed-off-by: Jeff Johnson --- drivers/net/wireless/ath/ath12k/mac.c | 77 ++++++++++++++------------- drivers/net/wireless/ath/ath12k/mac.h | 14 ++++- drivers/net/wireless/ath/ath12k/wmi.h | 12 ++--- 3 files changed, 60 insertions(+), 43 deletions(-) diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c index e79d457e3c03..c17500a3e95f 100644 --- a/drivers/net/wireless/ath/ath12k/mac.c +++ b/drivers/net/wireless/ath/ath12k/mac.c @@ -12094,55 +12094,57 @@ ath12k_mac_get_single_legacy_rate(struct ath12k *ar, } static int -ath12k_mac_set_fixed_rate_gi_ltf(struct ath12k_link_vif *arvif, u8 he_gi, u8 he_ltf) +ath12k_mac_set_fixed_rate_gi_ltf(struct ath12k_link_vif *arvif, u8 gi, u8 ltf) { struct ath12k *ar = arvif->ar; - int ret; + int param, ret; lockdep_assert_wiphy(ath12k_ar_to_hw(ar)->wiphy); /* 0.8 = 0, 1.6 = 2 and 3.2 = 3. */ - if (he_gi && he_gi != 0xFF) - he_gi += 1; + if (gi && gi != 0xFF) + gi += 1; ret = ath12k_wmi_vdev_set_param_cmd(ar, arvif->vdev_id, - WMI_VDEV_PARAM_SGI, he_gi); + WMI_VDEV_PARAM_SGI, gi); if (ret) { - ath12k_warn(ar->ab, "failed to set HE GI:%d, error:%d\n", - he_gi, ret); + ath12k_warn(ar->ab, "failed to set GI:%d, error:%d\n", + gi, ret); return ret; } /* start from 1 */ - if (he_ltf != 0xFF) - he_ltf += 1; + if (ltf != 0xFF) + ltf += 1; + + param = WMI_VDEV_PARAM_HE_LTF; ret = ath12k_wmi_vdev_set_param_cmd(ar, arvif->vdev_id, - WMI_VDEV_PARAM_HE_LTF, he_ltf); + param, ltf); if (ret) { - ath12k_warn(ar->ab, "failed to set HE LTF:%d, error:%d\n", - he_ltf, ret); + ath12k_warn(ar->ab, "failed to set LTF:%d, error:%d\n", + ltf, ret); return ret; } return 0; } static int -ath12k_mac_set_auto_rate_gi_ltf(struct ath12k_link_vif *arvif, u16 he_gi, u8 he_ltf) +ath12k_mac_set_auto_rate_gi_ltf(struct ath12k_link_vif *arvif, u16 gi, u8 ltf) { struct ath12k *ar = arvif->ar; int ret; - u32 he_ar_gi_ltf; + u32 ar_gi_ltf; - if (he_gi != 0xFF) { - switch (he_gi) { - case NL80211_RATE_INFO_HE_GI_0_8: - he_gi = WMI_AUTORATE_800NS_GI; + if (gi != 0xFF) { + switch (gi) { + case ATH12K_RATE_INFO_GI_0_8: + gi = WMI_AUTORATE_800NS_GI; break; - case NL80211_RATE_INFO_HE_GI_1_6: - he_gi = WMI_AUTORATE_1600NS_GI; + case ATH12K_RATE_INFO_GI_1_6: + gi = WMI_AUTORATE_1600NS_GI; break; - case NL80211_RATE_INFO_HE_GI_3_2: - he_gi = WMI_AUTORATE_3200NS_GI; + case ATH12K_RATE_INFO_GI_3_2: + gi = WMI_AUTORATE_3200NS_GI; break; default: ath12k_warn(ar->ab, "Invalid GI\n"); @@ -12150,16 +12152,16 @@ ath12k_mac_set_auto_rate_gi_ltf(struct ath12k_link_vif *arvif, u16 he_gi, u8 he_ } } - if (he_ltf != 0xFF) { - switch (he_ltf) { - case NL80211_RATE_INFO_HE_1XLTF: - he_ltf = WMI_HE_AUTORATE_LTF_1X; + if (ltf != 0xFF) { + switch (ltf) { + case ATH12K_RATE_INFO_1XLTF: + ltf = WMI_AUTORATE_LTF_1X; break; - case NL80211_RATE_INFO_HE_2XLTF: - he_ltf = WMI_HE_AUTORATE_LTF_2X; + case ATH12K_RATE_INFO_2XLTF: + ltf = WMI_AUTORATE_LTF_2X; break; - case NL80211_RATE_INFO_HE_4XLTF: - he_ltf = WMI_HE_AUTORATE_LTF_4X; + case ATH12K_RATE_INFO_4XLTF: + ltf = WMI_AUTORATE_LTF_4X; break; default: ath12k_warn(ar->ab, "Invalid LTF\n"); @@ -12167,15 +12169,15 @@ ath12k_mac_set_auto_rate_gi_ltf(struct ath12k_link_vif *arvif, u16 he_gi, u8 he_ } } - he_ar_gi_ltf = he_gi | he_ltf; + ar_gi_ltf = gi | ltf; ret = ath12k_wmi_vdev_set_param_cmd(ar, arvif->vdev_id, WMI_VDEV_PARAM_AUTORATE_MISC_CFG, - he_ar_gi_ltf); + ar_gi_ltf); if (ret) { ath12k_warn(ar->ab, - "failed to set HE autorate GI:%u, LTF:%u params, error:%d\n", - he_gi, he_ltf, ret); + "failed to set autorate GI:%u, LTF:%u params, error:%d\n", + gi, ltf, ret); return ret; } @@ -12200,10 +12202,10 @@ static int ath12k_mac_set_rate_params(struct ath12k_link_vif *arvif, { struct ieee80211_bss_conf *link_conf; struct ath12k *ar = arvif->ar; + bool he_support, gi_ltf_set = false; u32 vdev_param; u32 param_value; int ret; - bool he_support; lockdep_assert_wiphy(ath12k_ar_to_hw(ar)->wiphy); @@ -12257,7 +12259,10 @@ static int ath12k_mac_set_rate_params(struct ath12k_link_vif *arvif, ret = ath12k_mac_set_auto_rate_gi_ltf(arvif, he_gi, he_ltf); if (ret) return ret; - } else { + gi_ltf_set = true; + } + + if (!gi_ltf_set) { vdev_param = WMI_VDEV_PARAM_SGI; param_value = ath12k_mac_nlgi_to_wmigi(sgi); ret = ath12k_wmi_vdev_set_param_cmd(ar, arvif->vdev_id, diff --git a/drivers/net/wireless/ath/ath12k/mac.h b/drivers/net/wireless/ath/ath12k/mac.h index c05af40bd7a2..1f689e367c8a 100644 --- a/drivers/net/wireless/ath/ath12k/mac.h +++ b/drivers/net/wireless/ath/ath12k/mac.h @@ -1,7 +1,7 @@ /* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* * Copyright (c) 2018-2021 The Linux Foundation. All rights reserved. - * Copyright (c) 2021-2025 Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. */ #ifndef ATH12K_MAC_H @@ -84,6 +84,18 @@ enum ath12k_supported_bw { ATH12K_BW_320 = 4, }; +enum ath12k_gi { + ATH12K_RATE_INFO_GI_0_8, + ATH12K_RATE_INFO_GI_1_6, + ATH12K_RATE_INFO_GI_3_2, +}; + +enum ath12k_ltf { + ATH12K_RATE_INFO_1XLTF, + ATH12K_RATE_INFO_2XLTF, + ATH12K_RATE_INFO_4XLTF, +}; + struct ath12k_mac_get_any_chanctx_conf_arg { struct ath12k *ar; struct ieee80211_chanctx_conf *chanctx_conf; diff --git a/drivers/net/wireless/ath/ath12k/wmi.h b/drivers/net/wireless/ath/ath12k/wmi.h index 911ef9d52817..467fc32feee2 100644 --- a/drivers/net/wireless/ath/ath12k/wmi.h +++ b/drivers/net/wireless/ath/ath12k/wmi.h @@ -223,15 +223,15 @@ enum WMI_HOST_WLAN_BAND { }; /* Parameters used for WMI_VDEV_PARAM_AUTORATE_MISC_CFG command. - * Used only for HE auto rate mode. + * Used for HE and EHT auto rate mode. */ enum { - /* HE LTF related configuration */ - WMI_HE_AUTORATE_LTF_1X = BIT(0), - WMI_HE_AUTORATE_LTF_2X = BIT(1), - WMI_HE_AUTORATE_LTF_4X = BIT(2), + /* LTF related configuration */ + WMI_AUTORATE_LTF_1X = BIT(0), + WMI_AUTORATE_LTF_2X = BIT(1), + WMI_AUTORATE_LTF_4X = BIT(2), - /* HE GI related configuration */ + /* GI related configuration */ WMI_AUTORATE_400NS_GI = BIT(8), WMI_AUTORATE_800NS_GI = BIT(9), WMI_AUTORATE_1600NS_GI = BIT(10), From ec1d9b79be5df30f1998f37a49bf14d34ecd6c50 Mon Sep 17 00:00:00 2001 From: Muna Sinada Date: Thu, 23 Oct 2025 17:19:24 -0700 Subject: [PATCH 0408/1381] wifi: ath12k: add EHT rate handling to existing set rate functions Add EHT rate handling to the existing rate functions that validate, prepare and set rates. Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1 Co-developed-by: Aloka Dixit Signed-off-by: Aloka Dixit Signed-off-by: Muna Sinada Reviewed-by: Vasanthakumar Thiagarajan Link: https://patch.msgid.link/20251024001928.257356-3-muna.sinada@oss.qualcomm.com Signed-off-by: Jeff Johnson --- drivers/net/wireless/ath/ath12k/mac.c | 211 +++++++++++++++++++++----- 1 file changed, 171 insertions(+), 40 deletions(-) diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c index c17500a3e95f..0efdf87153e2 100644 --- a/drivers/net/wireless/ath/ath12k/mac.c +++ b/drivers/net/wireless/ath/ath12k/mac.c @@ -533,6 +533,18 @@ ath12k_mac_max_he_nss(const u16 he_mcs_mask[NL80211_HE_NSS_MAX]) return 1; } +static u32 +ath12k_mac_max_eht_nss(const u16 eht_mcs_mask[NL80211_EHT_NSS_MAX]) +{ + int nss; + + for (nss = NL80211_EHT_NSS_MAX - 1; nss >= 0; nss--) + if (eht_mcs_mask[nss]) + return nss + 1; + + return 1; +} + static u8 ath12k_parse_mpdudensity(u8 mpdudensity) { /* From IEEE Std 802.11-2020 defined values for "Minimum MPDU Start Spacing": @@ -3102,37 +3114,50 @@ static void ath12k_peer_assoc_h_phymode(struct ath12k *ar, WARN_ON(phymode == MODE_UNKNOWN); } +#define ATH12K_EHT_MCS_7_ENABLED 0x00FF +#define ATH12K_EHT_MCS_9_ENABLED 0x0300 +#define ATH12K_EHT_MCS_11_ENABLED 0x0C00 +#define ATH12K_EHT_MCS_13_ENABLED 0x3000 + static void ath12k_mac_set_eht_mcs(u8 rx_tx_mcs7, u8 rx_tx_mcs9, u8 rx_tx_mcs11, u8 rx_tx_mcs13, - u32 *rx_mcs, u32 *tx_mcs) + u32 *rx_mcs, u32 *tx_mcs, + const u16 eht_mcs_limit[NL80211_EHT_NSS_MAX]) { - *rx_mcs = 0; - u32p_replace_bits(rx_mcs, - u8_get_bits(rx_tx_mcs7, IEEE80211_EHT_MCS_NSS_RX), - WMI_EHT_MCS_NSS_0_7); - u32p_replace_bits(rx_mcs, - u8_get_bits(rx_tx_mcs9, IEEE80211_EHT_MCS_NSS_RX), - WMI_EHT_MCS_NSS_8_9); - u32p_replace_bits(rx_mcs, - u8_get_bits(rx_tx_mcs11, IEEE80211_EHT_MCS_NSS_RX), - WMI_EHT_MCS_NSS_10_11); - u32p_replace_bits(rx_mcs, - u8_get_bits(rx_tx_mcs13, IEEE80211_EHT_MCS_NSS_RX), - WMI_EHT_MCS_NSS_12_13); + int nss; + u8 mcs_7 = 0, mcs_9 = 0, mcs_11 = 0, mcs_13 = 0; + u8 peer_mcs_7, peer_mcs_9, peer_mcs_11, peer_mcs_13; - *tx_mcs = 0; - u32p_replace_bits(tx_mcs, - u8_get_bits(rx_tx_mcs7, IEEE80211_EHT_MCS_NSS_TX), - WMI_EHT_MCS_NSS_0_7); - u32p_replace_bits(tx_mcs, - u8_get_bits(rx_tx_mcs9, IEEE80211_EHT_MCS_NSS_TX), - WMI_EHT_MCS_NSS_8_9); - u32p_replace_bits(tx_mcs, - u8_get_bits(rx_tx_mcs11, IEEE80211_EHT_MCS_NSS_TX), - WMI_EHT_MCS_NSS_10_11); - u32p_replace_bits(tx_mcs, - u8_get_bits(rx_tx_mcs13, IEEE80211_EHT_MCS_NSS_TX), - WMI_EHT_MCS_NSS_12_13); + for (nss = 0; nss < NL80211_EHT_NSS_MAX; nss++) { + if (eht_mcs_limit[nss] & ATH12K_EHT_MCS_7_ENABLED) + mcs_7++; + if (eht_mcs_limit[nss] & ATH12K_EHT_MCS_9_ENABLED) + mcs_9++; + if (eht_mcs_limit[nss] & ATH12K_EHT_MCS_11_ENABLED) + mcs_11++; + if (eht_mcs_limit[nss] & ATH12K_EHT_MCS_13_ENABLED) + mcs_13++; + } + + peer_mcs_7 = u8_get_bits(rx_tx_mcs7, IEEE80211_EHT_MCS_NSS_RX); + peer_mcs_9 = u8_get_bits(rx_tx_mcs9, IEEE80211_EHT_MCS_NSS_RX); + peer_mcs_11 = u8_get_bits(rx_tx_mcs11, IEEE80211_EHT_MCS_NSS_RX); + peer_mcs_13 = u8_get_bits(rx_tx_mcs13, IEEE80211_EHT_MCS_NSS_RX); + + *rx_mcs = u32_encode_bits(min(peer_mcs_7, mcs_7), WMI_EHT_MCS_NSS_0_7) | + u32_encode_bits(min(peer_mcs_9, mcs_9), WMI_EHT_MCS_NSS_8_9) | + u32_encode_bits(min(peer_mcs_11, mcs_11), WMI_EHT_MCS_NSS_10_11) | + u32_encode_bits(min(peer_mcs_13, mcs_13), WMI_EHT_MCS_NSS_12_13); + + peer_mcs_7 = u8_get_bits(rx_tx_mcs7, IEEE80211_EHT_MCS_NSS_TX); + peer_mcs_9 = u8_get_bits(rx_tx_mcs9, IEEE80211_EHT_MCS_NSS_TX); + peer_mcs_11 = u8_get_bits(rx_tx_mcs11, IEEE80211_EHT_MCS_NSS_TX); + peer_mcs_13 = u8_get_bits(rx_tx_mcs13, IEEE80211_EHT_MCS_NSS_TX); + + *tx_mcs = u32_encode_bits(min(peer_mcs_7, mcs_7), WMI_EHT_MCS_NSS_0_7) | + u32_encode_bits(min(peer_mcs_9, mcs_9), WMI_EHT_MCS_NSS_8_9) | + u32_encode_bits(min(peer_mcs_11, mcs_11), WMI_EHT_MCS_NSS_10_11) | + u32_encode_bits(min(peer_mcs_13, mcs_13), WMI_EHT_MCS_NSS_12_13); } static void ath12k_mac_set_eht_ppe_threshold(const u8 *ppe_thres, @@ -3171,13 +3196,17 @@ static void ath12k_peer_assoc_h_eht(struct ath12k *ar, struct ath12k_wmi_peer_assoc_arg *arg) { struct ieee80211_sta *sta = ath12k_ahsta_to_sta(arsta->ahsta); + struct ieee80211_vif *vif = ath12k_ahvif_to_vif(arvif->ahvif); const struct ieee80211_eht_mcs_nss_supp_20mhz_only *bw_20; const struct ieee80211_eht_mcs_nss_supp_bw *bw; const struct ieee80211_sta_eht_cap *eht_cap; const struct ieee80211_sta_he_cap *he_cap; struct ieee80211_link_sta *link_sta; struct ieee80211_bss_conf *link_conf; + struct cfg80211_chan_def def; + enum nl80211_band band; u32 *rx_mcs, *tx_mcs; + u16 *eht_mcs_mask; lockdep_assert_wiphy(ath12k_ar_to_hw(ar)->wiphy); @@ -3199,6 +3228,12 @@ static void ath12k_peer_assoc_h_eht(struct ath12k *ar, if (!he_cap->has_he || !eht_cap->has_eht) return; + if (WARN_ON(ath12k_mac_vif_link_chan(vif, arvif->link_id, &def))) + return; + + band = def.chan->band; + eht_mcs_mask = arvif->bitrate_mask.control[band].eht_mcs; + arg->eht_flag = true; if ((eht_cap->eht_cap_elem.phy_cap_info[5] & @@ -3223,7 +3258,8 @@ static void ath12k_peer_assoc_h_eht(struct ath12k *ar, bw->rx_tx_mcs11_max_nss, bw->rx_tx_mcs13_max_nss, &rx_mcs[WMI_EHTCAP_TXRX_MCS_NSS_IDX_320], - &tx_mcs[WMI_EHTCAP_TXRX_MCS_NSS_IDX_320]); + &tx_mcs[WMI_EHTCAP_TXRX_MCS_NSS_IDX_320], + eht_mcs_mask); arg->peer_eht_mcs_count++; fallthrough; case IEEE80211_STA_RX_BW_160: @@ -3233,7 +3269,8 @@ static void ath12k_peer_assoc_h_eht(struct ath12k *ar, bw->rx_tx_mcs11_max_nss, bw->rx_tx_mcs13_max_nss, &rx_mcs[WMI_EHTCAP_TXRX_MCS_NSS_IDX_160], - &tx_mcs[WMI_EHTCAP_TXRX_MCS_NSS_IDX_160]); + &tx_mcs[WMI_EHTCAP_TXRX_MCS_NSS_IDX_160], + eht_mcs_mask); arg->peer_eht_mcs_count++; fallthrough; default: @@ -3249,7 +3286,8 @@ static void ath12k_peer_assoc_h_eht(struct ath12k *ar, bw_20->rx_tx_mcs11_max_nss, bw_20->rx_tx_mcs13_max_nss, &rx_mcs[WMI_EHTCAP_TXRX_MCS_NSS_IDX_80], - &tx_mcs[WMI_EHTCAP_TXRX_MCS_NSS_IDX_80]); + &tx_mcs[WMI_EHTCAP_TXRX_MCS_NSS_IDX_80], + eht_mcs_mask); } else { bw = &eht_cap->eht_mcs_nss_supp.bw._80; ath12k_mac_set_eht_mcs(bw->rx_tx_mcs9_max_nss, @@ -3257,7 +3295,8 @@ static void ath12k_peer_assoc_h_eht(struct ath12k *ar, bw->rx_tx_mcs11_max_nss, bw->rx_tx_mcs13_max_nss, &rx_mcs[WMI_EHTCAP_TXRX_MCS_NSS_IDX_80], - &tx_mcs[WMI_EHTCAP_TXRX_MCS_NSS_IDX_80]); + &tx_mcs[WMI_EHTCAP_TXRX_MCS_NSS_IDX_80], + eht_mcs_mask); } arg->peer_eht_mcs_count++; @@ -3908,6 +3947,8 @@ static void ath12k_mac_init_arvif(struct ath12k_vif *ahvif, sizeof(arvif->bitrate_mask.control[i].vht_mcs)); memset(arvif->bitrate_mask.control[i].he_mcs, 0xff, sizeof(arvif->bitrate_mask.control[i].he_mcs)); + memset(arvif->bitrate_mask.control[i].eht_mcs, 0xff, + sizeof(arvif->bitrate_mask.control[i].eht_mcs)); } /* Handle MLO related assignments */ @@ -5835,6 +5876,20 @@ ath12k_mac_bitrate_mask_num_he_rates(struct ath12k *ar, return num_rates; } +static int +ath12k_mac_bitrate_mask_num_eht_rates(struct ath12k *ar, + enum nl80211_band band, + const struct cfg80211_bitrate_mask *mask) +{ + int num_rates = 0; + int i; + + for (i = 0; i < ARRAY_SIZE(mask->control[band].eht_mcs); i++) + num_rates += hweight16(mask->control[band].eht_mcs[i]); + + return num_rates; +} + static int ath12k_mac_set_peer_vht_fixed_rate(struct ath12k_link_vif *arvif, struct ath12k_link_sta *arsta, @@ -11976,6 +12031,9 @@ ath12k_mac_has_single_legacy_rate(struct ath12k *ar, if (ath12k_mac_bitrate_mask_num_he_rates(ar, band, mask)) return false; + if (ath12k_mac_bitrate_mask_num_eht_rates(ar, band, mask)) + return false; + return num_rates == 1; } @@ -11998,11 +12056,15 @@ ath12k_mac_bitrate_mask_get_single_nss(struct ath12k *ar, { struct ieee80211_supported_band *sband = &ar->mac.sbands[band]; u16 vht_mcs_map = le16_to_cpu(sband->vht_cap.vht_mcs.tx_mcs_map); + const struct ieee80211_sband_iftype_data *data; const struct ieee80211_sta_he_cap *he_cap; u16 he_mcs_map = 0; + u16 eht_mcs_map = 0; u8 ht_nss_mask = 0; u8 vht_nss_mask = 0; u8 he_nss_mask = 0; + u8 eht_nss_mask = 0; + u8 mcs_nss_len; int i; /* No need to consider legacy here. Basic rates are always present @@ -12046,7 +12108,60 @@ ath12k_mac_bitrate_mask_get_single_nss(struct ath12k *ar, return false; } - if (ht_nss_mask != vht_nss_mask || ht_nss_mask != he_nss_mask) + data = ieee80211_get_sband_iftype_data(sband, vif->type); + + mcs_nss_len = ieee80211_eht_mcs_nss_size(&data->he_cap.he_cap_elem, + &data->eht_cap.eht_cap_elem, + false); + if (mcs_nss_len == 4) { + /* 20 MHz only STA case */ + const struct ieee80211_eht_mcs_nss_supp_20mhz_only *eht_mcs_nss = + &data->eht_cap.eht_mcs_nss_supp.only_20mhz; + if (eht_mcs_nss->rx_tx_mcs13_max_nss) + eht_mcs_map = 0x1fff; + else if (eht_mcs_nss->rx_tx_mcs11_max_nss) + eht_mcs_map = 0x07ff; + else if (eht_mcs_nss->rx_tx_mcs9_max_nss) + eht_mcs_map = 0x01ff; + else + eht_mcs_map = 0x007f; + } else { + const struct ieee80211_eht_mcs_nss_supp_bw *eht_mcs_nss; + + switch (mcs_nss_len) { + case 9: + eht_mcs_nss = &data->eht_cap.eht_mcs_nss_supp.bw._320; + break; + case 6: + eht_mcs_nss = &data->eht_cap.eht_mcs_nss_supp.bw._160; + break; + case 3: + eht_mcs_nss = &data->eht_cap.eht_mcs_nss_supp.bw._80; + break; + default: + return false; + } + + if (eht_mcs_nss->rx_tx_mcs13_max_nss) + eht_mcs_map = 0x1fff; + else if (eht_mcs_nss->rx_tx_mcs11_max_nss) + eht_mcs_map = 0x7ff; + else + eht_mcs_map = 0x1ff; + } + + for (i = 0; i < ARRAY_SIZE(mask->control[band].eht_mcs); i++) { + if (mask->control[band].eht_mcs[i] == 0) + continue; + + if (mask->control[band].eht_mcs[i] < eht_mcs_map) + eht_nss_mask |= BIT(i); + else + return false; + } + + if (ht_nss_mask != vht_nss_mask || ht_nss_mask != he_nss_mask || + ht_nss_mask != eht_nss_mask) return false; if (ht_nss_mask == 0) @@ -12198,11 +12313,12 @@ static u32 ath12k_mac_nlgi_to_wmigi(enum nl80211_txrate_gi gi) static int ath12k_mac_set_rate_params(struct ath12k_link_vif *arvif, u32 rate, u8 nss, u8 sgi, u8 ldpc, - u8 he_gi, u8 he_ltf, bool he_fixed_rate) + u8 he_gi, u8 he_ltf, bool he_fixed_rate, + bool eht_fixed_rate) { struct ieee80211_bss_conf *link_conf; struct ath12k *ar = arvif->ar; - bool he_support, gi_ltf_set = false; + bool he_support, eht_support, gi_ltf_set = false; u32 vdev_param; u32 param_value; int ret; @@ -12214,6 +12330,7 @@ static int ath12k_mac_set_rate_params(struct ath12k_link_vif *arvif, return -EINVAL; he_support = link_conf->he_support; + eht_support = link_conf->eht_support; ath12k_dbg(ar->ab, ATH12K_DBG_MAC, "mac set rate params vdev %i rate 0x%02x nss 0x%02x sgi 0x%02x ldpc 0x%02x\n", @@ -12223,7 +12340,10 @@ static int ath12k_mac_set_rate_params(struct ath12k_link_vif *arvif, "he_gi 0x%02x he_ltf 0x%02x he_fixed_rate %d\n", he_gi, he_ltf, he_fixed_rate); - if (!he_support) { + ath12k_dbg(ar->ab, ATH12K_DBG_MAC, "eht_fixed_rate %d\n", + eht_fixed_rate); + + if (!he_support && !eht_support) { vdev_param = WMI_VDEV_PARAM_FIXED_RATE; ret = ath12k_wmi_vdev_set_param_cmd(ar, arvif->vdev_id, vdev_param, rate); @@ -12381,15 +12501,16 @@ ath12k_mac_validate_fixed_rate_settings(struct ath12k *ar, enum nl80211_band ban const struct cfg80211_bitrate_mask *mask, unsigned int link_id) { - bool he_fixed_rate = false, vht_fixed_rate = false; - const u16 *vht_mcs_mask, *he_mcs_mask; + bool eht_fixed_rate = false, he_fixed_rate = false, vht_fixed_rate = false; + const u16 *vht_mcs_mask, *he_mcs_mask, *eht_mcs_mask; struct ieee80211_link_sta *link_sta; struct ath12k_peer *peer, *tmp; - u8 vht_nss, he_nss; + u8 vht_nss, he_nss, eht_nss; int ret = true; vht_mcs_mask = mask->control[band].vht_mcs; he_mcs_mask = mask->control[band].he_mcs; + eht_mcs_mask = mask->control[band].eht_mcs; if (ath12k_mac_bitrate_mask_num_vht_rates(ar, band, mask) == 1) vht_fixed_rate = true; @@ -12397,11 +12518,15 @@ ath12k_mac_validate_fixed_rate_settings(struct ath12k *ar, enum nl80211_band ban if (ath12k_mac_bitrate_mask_num_he_rates(ar, band, mask) == 1) he_fixed_rate = true; - if (!vht_fixed_rate && !he_fixed_rate) + if (ath12k_mac_bitrate_mask_num_eht_rates(ar, band, mask) == 1) + eht_fixed_rate = true; + + if (!vht_fixed_rate && !he_fixed_rate && !eht_fixed_rate) return true; vht_nss = ath12k_mac_max_vht_nss(vht_mcs_mask); he_nss = ath12k_mac_max_he_nss(he_mcs_mask); + eht_nss = ath12k_mac_max_eht_nss(eht_mcs_mask); rcu_read_lock(); spin_lock_bh(&ar->ab->base_lock); @@ -12423,6 +12548,11 @@ ath12k_mac_validate_fixed_rate_settings(struct ath12k *ar, enum nl80211_band ban ret = false; goto exit; } + if (eht_fixed_rate && (!link_sta->eht_cap.has_eht || + link_sta->rx_nss < eht_nss)) { + ret = false; + goto exit; + } } } exit: @@ -12454,6 +12584,7 @@ ath12k_mac_op_set_bitrate_mask(struct ieee80211_hw *hw, int ret; int num_rates; bool he_fixed_rate = false; + bool eht_fixed_rate = false; lockdep_assert_wiphy(hw->wiphy); @@ -12578,7 +12709,7 @@ ath12k_mac_op_set_bitrate_mask(struct ieee80211_hw *hw, } ret = ath12k_mac_set_rate_params(arvif, rate, nss, sgi, ldpc, he_gi, - he_ltf, he_fixed_rate); + he_ltf, he_fixed_rate, eht_fixed_rate); if (ret) { ath12k_warn(ar->ab, "failed to set rate params on vdev %i: %d\n", arvif->vdev_id, ret); From 6c95151e2e776462de6b6fd9e577865e8b32203e Mon Sep 17 00:00:00 2001 From: Muna Sinada Date: Thu, 23 Oct 2025 17:19:25 -0700 Subject: [PATCH 0409/1381] wifi: ath12k: Add EHT MCS/NSS rates to Peer Assoc Add EHT MCS/NSS rate functionality to peer association. As part of ath12k_peer_assoc_h_eht() add the calculation of EHT MCS/NSS using intersection of link_sta and phy capability. ath12k_mac_max_eht_mcs_nss() function is utilized when comparing the max NSS of link STA and phy capability since in split phy case, phy supports max NSS of 2 for 5G band. Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1 Co-developed-by: Vishal Kumar Signed-off-by: Vishal Kumar Co-developed-by: Aloka Dixit Signed-off-by: Aloka Dixit Signed-off-by: Muna Sinada Reviewed-by: Vasanthakumar Thiagarajan Link: https://patch.msgid.link/20251024001928.257356-4-muna.sinada@oss.qualcomm.com Signed-off-by: Jeff Johnson --- drivers/net/wireless/ath/ath12k/mac.c | 110 ++++++++++++++++++++++++-- 1 file changed, 103 insertions(+), 7 deletions(-) diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c index 0efdf87153e2..f1b5755bb85e 100644 --- a/drivers/net/wireless/ath/ath12k/mac.c +++ b/drivers/net/wireless/ath/ath12k/mac.c @@ -545,6 +545,18 @@ ath12k_mac_max_eht_nss(const u16 eht_mcs_mask[NL80211_EHT_NSS_MAX]) return 1; } +static u32 +ath12k_mac_max_eht_mcs_nss(const u8 *eht_mcs, int eht_mcs_set_size) +{ + int i; + u8 nss = 0; + + for (i = 0; i < eht_mcs_set_size; i++) + nss = max(nss, u8_get_bits(eht_mcs[i], IEEE80211_EHT_MCS_NSS_RX)); + + return nss; +} + static u8 ath12k_parse_mpdudensity(u8 mpdudensity) { /* From IEEE Std 802.11-2020 defined values for "Minimum MPDU Start Spacing": @@ -3016,6 +3028,18 @@ static enum wmi_phy_mode ath12k_mac_get_phymode_eht(struct ath12k *ar, return MODE_UNKNOWN; } +static bool +ath12k_peer_assoc_h_eht_masked(const u16 eht_mcs_mask[NL80211_EHT_NSS_MAX]) +{ + int nss; + + for (nss = 0; nss < NL80211_EHT_NSS_MAX; nss++) + if (eht_mcs_mask[nss]) + return false; + + return true; +} + static void ath12k_peer_assoc_h_phymode(struct ath12k *ar, struct ath12k_link_vif *arvif, struct ath12k_link_sta *arsta, @@ -3027,6 +3051,7 @@ static void ath12k_peer_assoc_h_phymode(struct ath12k *ar, const u8 *ht_mcs_mask; const u16 *vht_mcs_mask; const u16 *he_mcs_mask; + const u16 *eht_mcs_mask; enum wmi_phy_mode phymode = MODE_UNKNOWN; lockdep_assert_wiphy(ath12k_ar_to_hw(ar)->wiphy); @@ -3041,6 +3066,7 @@ static void ath12k_peer_assoc_h_phymode(struct ath12k *ar, ht_mcs_mask = arvif->bitrate_mask.control[band].ht_mcs; vht_mcs_mask = arvif->bitrate_mask.control[band].vht_mcs; he_mcs_mask = arvif->bitrate_mask.control[band].he_mcs; + eht_mcs_mask = arvif->bitrate_mask.control[band].eht_mcs; link_sta = ath12k_mac_get_link_sta(arsta); if (!link_sta) { @@ -3051,7 +3077,8 @@ static void ath12k_peer_assoc_h_phymode(struct ath12k *ar, switch (band) { case NL80211_BAND_2GHZ: - if (link_sta->eht_cap.has_eht) { + if (link_sta->eht_cap.has_eht && + !ath12k_peer_assoc_h_eht_masked(eht_mcs_mask)) { if (link_sta->bandwidth == IEEE80211_STA_RX_BW_40) phymode = MODE_11BE_EHT40_2G; else @@ -3197,16 +3224,21 @@ static void ath12k_peer_assoc_h_eht(struct ath12k *ar, { struct ieee80211_sta *sta = ath12k_ahsta_to_sta(arsta->ahsta); struct ieee80211_vif *vif = ath12k_ahvif_to_vif(arvif->ahvif); + const struct ieee80211_eht_mcs_nss_supp *own_eht_mcs_nss_supp; const struct ieee80211_eht_mcs_nss_supp_20mhz_only *bw_20; + const struct ieee80211_sta_eht_cap *eht_cap, *own_eht_cap; + const struct ieee80211_sband_iftype_data *iftd; const struct ieee80211_eht_mcs_nss_supp_bw *bw; - const struct ieee80211_sta_eht_cap *eht_cap; const struct ieee80211_sta_he_cap *he_cap; struct ieee80211_link_sta *link_sta; struct ieee80211_bss_conf *link_conf; struct cfg80211_chan_def def; + bool user_rate_valid = true; enum nl80211_band band; + int eht_nss, nss_idx; u32 *rx_mcs, *tx_mcs; u16 *eht_mcs_mask; + u8 max_nss = 0; lockdep_assert_wiphy(ath12k_ar_to_hw(ar)->wiphy); @@ -3234,6 +3266,16 @@ static void ath12k_peer_assoc_h_eht(struct ath12k *ar, band = def.chan->band; eht_mcs_mask = arvif->bitrate_mask.control[band].eht_mcs; + iftd = ieee80211_get_sband_iftype_data(&ar->mac.sbands[band], vif->type); + if (!iftd) { + ath12k_warn(ar->ab, + "unable to access iftype_data in struct ieee80211_supported_band\n"); + return; + } + + own_eht_cap = &iftd->eht_cap; + own_eht_mcs_nss_supp = &own_eht_cap->eht_mcs_nss_supp; + arg->eht_flag = true; if ((eht_cap->eht_cap_elem.phy_cap_info[5] & @@ -3250,6 +3292,28 @@ static void ath12k_peer_assoc_h_eht(struct ath12k *ar, rx_mcs = arg->peer_eht_rx_mcs_set; tx_mcs = arg->peer_eht_tx_mcs_set; + eht_nss = ath12k_mac_max_eht_mcs_nss((void *)own_eht_mcs_nss_supp, + sizeof(*own_eht_mcs_nss_supp)); + if (eht_nss > link_sta->rx_nss) { + user_rate_valid = false; + for (nss_idx = (link_sta->rx_nss - 1); nss_idx >= 0; nss_idx--) { + if (eht_mcs_mask[nss_idx]) { + user_rate_valid = true; + break; + } + } + } + + if (!user_rate_valid) { + ath12k_dbg(ar->ab, ATH12K_DBG_MAC, + "Setting eht range MCS value to peer supported nss %d for peer %pM\n", + link_sta->rx_nss, arsta->addr); + eht_mcs_mask[link_sta->rx_nss - 1] = eht_mcs_mask[eht_nss - 1]; + } + + bw_20 = &eht_cap->eht_mcs_nss_supp.only_20mhz; + bw = &eht_cap->eht_mcs_nss_supp.bw._80; + switch (link_sta->bandwidth) { case IEEE80211_STA_RX_BW_320: bw = &eht_cap->eht_mcs_nss_supp.bw._320; @@ -3274,11 +3338,8 @@ static void ath12k_peer_assoc_h_eht(struct ath12k *ar, arg->peer_eht_mcs_count++; fallthrough; default: - if ((he_cap->he_cap_elem.phy_cap_info[0] & - (IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_40MHZ_IN_2G | - IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_40MHZ_80MHZ_IN_5G | - IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_160MHZ_IN_5G | - IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_80PLUS80_MHZ_IN_5G)) == 0) { + if (!(link_sta->he_cap.he_cap_elem.phy_cap_info[0] & + IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_MASK_ALL)) { bw_20 = &eht_cap->eht_mcs_nss_supp.only_20mhz; ath12k_mac_set_eht_mcs(bw_20->rx_tx_mcs7_max_nss, @@ -3305,6 +3366,41 @@ static void ath12k_peer_assoc_h_eht(struct ath12k *ar, arg->punct_bitmap = ~arvif->punct_bitmap; arg->eht_disable_mcs15 = link_conf->eht_disable_mcs15; + + if (!(link_sta->he_cap.he_cap_elem.phy_cap_info[0] & + IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_MASK_ALL)) { + if (bw_20->rx_tx_mcs13_max_nss) + max_nss = max(max_nss, u8_get_bits(bw_20->rx_tx_mcs13_max_nss, + IEEE80211_EHT_MCS_NSS_RX)); + if (bw_20->rx_tx_mcs11_max_nss) + max_nss = max(max_nss, u8_get_bits(bw_20->rx_tx_mcs11_max_nss, + IEEE80211_EHT_MCS_NSS_RX)); + if (bw_20->rx_tx_mcs9_max_nss) + max_nss = max(max_nss, u8_get_bits(bw_20->rx_tx_mcs9_max_nss, + IEEE80211_EHT_MCS_NSS_RX)); + if (bw_20->rx_tx_mcs7_max_nss) + max_nss = max(max_nss, u8_get_bits(bw_20->rx_tx_mcs7_max_nss, + IEEE80211_EHT_MCS_NSS_RX)); + } else { + if (bw->rx_tx_mcs13_max_nss) + max_nss = max(max_nss, u8_get_bits(bw->rx_tx_mcs13_max_nss, + IEEE80211_EHT_MCS_NSS_RX)); + if (bw->rx_tx_mcs11_max_nss) + max_nss = max(max_nss, u8_get_bits(bw->rx_tx_mcs11_max_nss, + IEEE80211_EHT_MCS_NSS_RX)); + if (bw->rx_tx_mcs9_max_nss) + max_nss = max(max_nss, u8_get_bits(bw->rx_tx_mcs9_max_nss, + IEEE80211_EHT_MCS_NSS_RX)); + } + + max_nss = min(max_nss, (uint8_t)eht_nss); + + arg->peer_nss = min(link_sta->rx_nss, max_nss); + + ath12k_dbg(ar->ab, ATH12K_DBG_MAC, + "mac eht peer %pM nss %d mcs cnt %d ru_punct_bitmap 0x%x\n", + arsta->addr, arg->peer_nss, arg->peer_eht_mcs_count, + arg->punct_bitmap); } static void ath12k_peer_assoc_h_mlo(struct ath12k_link_sta *arsta, From ab31a9b73c95ca8a1b527a0d9fd9192a27acb26f Mon Sep 17 00:00:00 2001 From: Muna Sinada Date: Thu, 23 Oct 2025 17:19:26 -0700 Subject: [PATCH 0410/1381] wifi: ath12k: Add EHT fixed GI/LTF Add EHT functionality to set fixed GI/LTF parameters. Add new wmi vdev parameter id for EHT LTF Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1 Signed-off-by: Muna Sinada Reviewed-by: Vasanthakumar Thiagarajan Link: https://patch.msgid.link/20251024001928.257356-5-muna.sinada@oss.qualcomm.com Signed-off-by: Jeff Johnson --- drivers/net/wireless/ath/ath12k/mac.c | 50 +++++++++++++++++++++------ drivers/net/wireless/ath/ath12k/wmi.h | 1 + 2 files changed, 41 insertions(+), 10 deletions(-) diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c index f1b5755bb85e..ca06807c4e0f 100644 --- a/drivers/net/wireless/ath/ath12k/mac.c +++ b/drivers/net/wireless/ath/ath12k/mac.c @@ -12305,10 +12305,11 @@ ath12k_mac_get_single_legacy_rate(struct ath12k *ar, } static int -ath12k_mac_set_fixed_rate_gi_ltf(struct ath12k_link_vif *arvif, u8 gi, u8 ltf) +ath12k_mac_set_fixed_rate_gi_ltf(struct ath12k_link_vif *arvif, u8 gi, u8 ltf, + u32 param) { struct ath12k *ar = arvif->ar; - int param, ret; + int ret; lockdep_assert_wiphy(ath12k_ar_to_hw(ar)->wiphy); @@ -12323,11 +12324,16 @@ ath12k_mac_set_fixed_rate_gi_ltf(struct ath12k_link_vif *arvif, u8 gi, u8 ltf) gi, ret); return ret; } - /* start from 1 */ - if (ltf != 0xFF) - ltf += 1; - param = WMI_VDEV_PARAM_HE_LTF; + if (param == WMI_VDEV_PARAM_HE_LTF) { + /* HE values start from 1 */ + if (ltf != 0xFF) + ltf += 1; + } else { + /* EHT values start from 5 */ + if (ltf != 0xFF) + ltf += 4; + } ret = ath12k_wmi_vdev_set_param_cmd(ar, arvif->vdev_id, param, ltf); @@ -12410,6 +12416,7 @@ static u32 ath12k_mac_nlgi_to_wmigi(enum nl80211_txrate_gi gi) static int ath12k_mac_set_rate_params(struct ath12k_link_vif *arvif, u32 rate, u8 nss, u8 sgi, u8 ldpc, u8 he_gi, u8 he_ltf, bool he_fixed_rate, + u8 eht_gi, u8 eht_ltf, bool eht_fixed_rate) { struct ieee80211_bss_conf *link_conf; @@ -12436,8 +12443,9 @@ static int ath12k_mac_set_rate_params(struct ath12k_link_vif *arvif, "he_gi 0x%02x he_ltf 0x%02x he_fixed_rate %d\n", he_gi, he_ltf, he_fixed_rate); - ath12k_dbg(ar->ab, ATH12K_DBG_MAC, "eht_fixed_rate %d\n", - eht_fixed_rate); + ath12k_dbg(ar->ab, ATH12K_DBG_MAC, + "eht_gi 0x%02x eht_ltf 0x%02x eht_fixed_rate %d\n", + eht_gi, eht_ltf, eht_fixed_rate); if (!he_support && !eht_support) { vdev_param = WMI_VDEV_PARAM_FIXED_RATE; @@ -12468,9 +12476,26 @@ static int ath12k_mac_set_rate_params(struct ath12k_link_vif *arvif, return ret; } + if (eht_support) { + if (eht_fixed_rate) + ret = ath12k_mac_set_fixed_rate_gi_ltf(arvif, eht_gi, eht_ltf, + WMI_VDEV_PARAM_EHT_LTF); + else + ret = ath12k_mac_set_auto_rate_gi_ltf(arvif, eht_gi, eht_ltf); + + if (ret) { + ath12k_warn(ar->ab, + "failed to set EHT LTF/GI params %d/%d: %d\n", + eht_gi, eht_ltf, ret); + return ret; + } + gi_ltf_set = true; + } + if (he_support) { if (he_fixed_rate) - ret = ath12k_mac_set_fixed_rate_gi_ltf(arvif, he_gi, he_ltf); + ret = ath12k_mac_set_fixed_rate_gi_ltf(arvif, he_gi, he_ltf, + WMI_VDEV_PARAM_HE_LTF); else ret = ath12k_mac_set_auto_rate_gi_ltf(arvif, he_gi, he_ltf); if (ret) @@ -12672,6 +12697,7 @@ ath12k_mac_op_set_bitrate_mask(struct ieee80211_hw *hw, const u16 *he_mcs_mask; u8 he_ltf = 0; u8 he_gi = 0; + u8 eht_ltf = 0, eht_gi = 0; u32 rate; u8 nss, mac_nss; u8 sgi; @@ -12707,6 +12733,9 @@ ath12k_mac_op_set_bitrate_mask(struct ieee80211_hw *hw, he_gi = mask->control[band].he_gi; he_ltf = mask->control[band].he_ltf; + eht_gi = mask->control[band].eht_gi; + eht_ltf = mask->control[band].eht_ltf; + /* mac80211 doesn't support sending a fixed HT/VHT MCS alone, rather it * requires passing at least one of used basic rates along with them. * Fixed rate setting across different preambles(legacy, HT, VHT) is @@ -12805,7 +12834,8 @@ ath12k_mac_op_set_bitrate_mask(struct ieee80211_hw *hw, } ret = ath12k_mac_set_rate_params(arvif, rate, nss, sgi, ldpc, he_gi, - he_ltf, he_fixed_rate, eht_fixed_rate); + he_ltf, he_fixed_rate, eht_gi, eht_ltf, + eht_fixed_rate); if (ret) { ath12k_warn(ar->ab, "failed to set rate params on vdev %i: %d\n", arvif->vdev_id, ret); diff --git a/drivers/net/wireless/ath/ath12k/wmi.h b/drivers/net/wireless/ath/ath12k/wmi.h index 467fc32feee2..f99fced1610e 100644 --- a/drivers/net/wireless/ath/ath12k/wmi.h +++ b/drivers/net/wireless/ath/ath12k/wmi.h @@ -1197,6 +1197,7 @@ enum wmi_tlv_vdev_param { WMI_VDEV_PARAM_SET_HEMU_MODE, WMI_VDEV_PARAM_HEOPS_0_31 = 0x8003, WMI_VDEV_PARAM_SET_EHT_MU_MODE = 0x8005, + WMI_VDEV_PARAM_EHT_LTF, }; enum wmi_tlv_peer_flags { From 5ee9cb2c236b45c4d58d6d464a12d985e453576b Mon Sep 17 00:00:00 2001 From: Muna Sinada Date: Thu, 23 Oct 2025 17:19:27 -0700 Subject: [PATCH 0411/1381] wifi: ath12k: add EHT rates to ath12k_mac_op_set_bitrate_mask() Extend ath12k_mac_op_set_bitrate_mask() to handle EHT rates. Create and pass EHT mask containing MCS and NSS along with EHT GI and LTF when calling ath12k_mac_set_rate_params() Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1 Co-developed-by: Aaradhana Sahu Signed-off-by: Aaradhana Sahu Signed-off-by: Muna Sinada Reviewed-by: Vasanthakumar Thiagarajan Link: https://patch.msgid.link/20251024001928.257356-6-muna.sinada@oss.qualcomm.com Signed-off-by: Jeff Johnson --- drivers/net/wireless/ath/ath12k/mac.c | 55 +++++++++++++++++++++++++-- 1 file changed, 52 insertions(+), 3 deletions(-) diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c index ca06807c4e0f..fa53a588ceff 100644 --- a/drivers/net/wireless/ath/ath12k/mac.c +++ b/drivers/net/wireless/ath/ath12k/mac.c @@ -12568,6 +12568,38 @@ ath12k_mac_he_mcs_range_present(struct ath12k *ar, return true; } +static bool +ath12k_mac_eht_mcs_range_present(struct ath12k *ar, + enum nl80211_band band, + const struct cfg80211_bitrate_mask *mask) +{ + u16 eht_mcs; + int i; + + for (i = 0; i < NL80211_EHT_NSS_MAX; i++) { + eht_mcs = mask->control[band].eht_mcs[i]; + + switch (eht_mcs) { + case 0: + case BIT(8) - 1: + case BIT(10) - 1: + case BIT(12) - 1: + case BIT(14) - 1: + break; + case BIT(15) - 1: + case BIT(16) - 1: + case BIT(16) - BIT(14) - 1: + if (i != 0) + return false; + break; + default: + return false; + } + } + + return true; +} + static void ath12k_mac_set_bitrate_mask_iter(void *data, struct ieee80211_sta *sta) { @@ -12695,6 +12727,7 @@ ath12k_mac_op_set_bitrate_mask(struct ieee80211_hw *hw, const u8 *ht_mcs_mask; const u16 *vht_mcs_mask; const u16 *he_mcs_mask; + const u16 *eht_mcs_mask; u8 he_ltf = 0; u8 he_gi = 0; u8 eht_ltf = 0, eht_gi = 0; @@ -12722,6 +12755,7 @@ ath12k_mac_op_set_bitrate_mask(struct ieee80211_hw *hw, ht_mcs_mask = mask->control[band].ht_mcs; vht_mcs_mask = mask->control[band].vht_mcs; he_mcs_mask = mask->control[band].he_mcs; + eht_mcs_mask = mask->control[band].eht_mcs; ldpc = !!(ar->ht_cap_info & WMI_HT_CAP_LDPC); sgi = mask->control[band].gi; @@ -12773,9 +12807,10 @@ ath12k_mac_op_set_bitrate_mask(struct ieee80211_hw *hw, ath12k_warn(ar->ab, "failed to update fixed rate settings due to mcs/nss incompatibility\n"); - mac_nss = max3(ath12k_mac_max_ht_nss(ht_mcs_mask), - ath12k_mac_max_vht_nss(vht_mcs_mask), - ath12k_mac_max_he_nss(he_mcs_mask)); + mac_nss = max(max3(ath12k_mac_max_ht_nss(ht_mcs_mask), + ath12k_mac_max_vht_nss(vht_mcs_mask), + ath12k_mac_max_he_nss(he_mcs_mask)), + ath12k_mac_max_eht_nss(eht_mcs_mask)); nss = min_t(u32, ar->num_tx_chains, mac_nss); /* If multiple rates across different preambles are given @@ -12823,6 +12858,20 @@ ath12k_mac_op_set_bitrate_mask(struct ieee80211_hw *hw, ret = -EINVAL; goto out; } + + num_rates = ath12k_mac_bitrate_mask_num_eht_rates(ar, band, + mask); + if (num_rates == 1) + eht_fixed_rate = true; + + if (!ath12k_mac_eht_mcs_range_present(ar, band, mask) && + num_rates > 1) { + ath12k_warn(ar->ab, + "Setting more than one EHT MCS Value in bitrate mask not supported\n"); + ret = -EINVAL; + goto out; + } + ieee80211_iterate_stations_mtx(hw, ath12k_mac_disable_peer_fixed_rate, arvif); From 09486128caef9efd88794c8aaa1e9ab16b16f383 Mon Sep 17 00:00:00 2001 From: Muna Sinada Date: Thu, 23 Oct 2025 17:19:28 -0700 Subject: [PATCH 0412/1381] wifi: ath12k: Set EHT fixed rates for associated STAs Fixed rate is set for STAs that are associated. This will be done during association or with ath12k_sta_rc_update_wk(). Add EHT fixed rate setting for STAs by adding call to ath12k_mac_set_peer_eht_fixed_rate() during the times fixed rate is set for STAs. This new function sets EHT fixed rate for a peer, which sends WMI command with the updated MCS/NSS rate using WMI_PEER_PARAM_FIXED_RATE command id. Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1 Co-developed-by: Aaradhana Sahu Signed-off-by: Aaradhana Sahu Signed-off-by: Muna Sinada Reviewed-by: Vasanthakumar Thiagarajan Link: https://patch.msgid.link/20251024001928.257356-7-muna.sinada@oss.qualcomm.com Signed-off-by: Jeff Johnson --- drivers/net/wireless/ath/ath12k/mac.c | 92 +++++++++++++++++++++++++-- 1 file changed, 85 insertions(+), 7 deletions(-) diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c index fa53a588ceff..29a0b022c32e 100644 --- a/drivers/net/wireless/ath/ath12k/mac.c +++ b/drivers/net/wireless/ath/ath12k/mac.c @@ -6086,6 +6086,65 @@ ath12k_mac_set_peer_he_fixed_rate(struct ath12k_link_vif *arvif, return ret; } +static int +ath12k_mac_set_peer_eht_fixed_rate(struct ath12k_link_vif *arvif, + struct ath12k_link_sta *arsta, + const struct cfg80211_bitrate_mask *mask, + enum nl80211_band band) +{ + struct ath12k_sta *ahsta = arsta->ahsta; + struct ath12k *ar = arvif->ar; + struct ieee80211_sta *sta; + struct ieee80211_link_sta *link_sta; + u8 eht_rate, nss = 0; + u32 rate_code; + int ret, i; + + lockdep_assert_wiphy(ath12k_ar_to_hw(ar)->wiphy); + + sta = ath12k_ahsta_to_sta(ahsta); + + for (i = 0; i < ARRAY_SIZE(mask->control[band].eht_mcs); i++) { + if (hweight16(mask->control[band].eht_mcs[i]) == 1) { + nss = i + 1; + eht_rate = ffs(mask->control[band].eht_mcs[i]) - 1; + } + } + + if (!nss) { + ath12k_warn(ar->ab, "No single EHT Fixed rate found to set for %pM\n", + arsta->addr); + return -EINVAL; + } + + /* Avoid updating invalid nss as fixed rate*/ + link_sta = ath12k_mac_get_link_sta(arsta); + if (!link_sta || nss > link_sta->rx_nss) { + ath12k_warn(ar->ab, + "unable to access link sta for sta %pM link %u or fixed nss of %u is not supported by sta\n", + sta->addr, arsta->link_id, nss); + return -EINVAL; + } + + ath12k_dbg(ar->ab, ATH12K_DBG_MAC, + "Setting Fixed EHT Rate for peer %pM. Device will not switch to any other selected rates\n", + arsta->addr); + + rate_code = ATH12K_HW_RATE_CODE(eht_rate, nss - 1, + WMI_RATE_PREAMBLE_EHT); + + ret = ath12k_wmi_set_peer_param(ar, arsta->addr, + arvif->vdev_id, + WMI_PEER_PARAM_FIXED_RATE, + rate_code); + if (ret) + ath12k_warn(ar->ab, + "failed to update STA %pM Fixed Rate %d: %d\n", + arsta->addr, rate_code, ret); + + return ret; +} + static int ath12k_mac_station_assoc(struct ath12k *ar, struct ath12k_link_vif *arvif, struct ath12k_link_sta *arsta, @@ -6098,7 +6157,7 @@ static int ath12k_mac_station_assoc(struct ath12k *ar, struct cfg80211_chan_def def; enum nl80211_band band; struct cfg80211_bitrate_mask *mask; - u8 num_vht_rates, num_he_rates; + u8 num_vht_rates, num_he_rates, num_eht_rates; u8 link_id = arvif->link_id; lockdep_assert_wiphy(ath12k_ar_to_hw(ar)->wiphy); @@ -6141,10 +6200,11 @@ static int ath12k_mac_station_assoc(struct ath12k *ar, num_vht_rates = ath12k_mac_bitrate_mask_num_vht_rates(ar, band, mask); num_he_rates = ath12k_mac_bitrate_mask_num_he_rates(ar, band, mask); + num_eht_rates = ath12k_mac_bitrate_mask_num_eht_rates(ar, band, mask); - /* If single VHT/HE rate is configured (by set_bitrate_mask()), - * peer_assoc will disable VHT/HE. This is now enabled by a peer specific - * fixed param. + /* If single VHT/HE/EHT rate is configured (by set_bitrate_mask()), + * peer_assoc will disable VHT/HE/EHT. This is now enabled by a peer + * specific fixed param. * Note that all other rates and NSS will be disabled for this peer. */ link_sta = ath12k_mac_get_link_sta(arsta); @@ -6164,6 +6224,10 @@ static int ath12k_mac_station_assoc(struct ath12k *ar, ret = ath12k_mac_set_peer_he_fixed_rate(arvif, arsta, mask, band); if (ret) return ret; + } else if (link_sta->eht_cap.has_eht && num_eht_rates == 1) { + ret = ath12k_mac_set_peer_eht_fixed_rate(arvif, arsta, mask, band); + if (ret) + return ret; } /* Re-assoc is run only to update supported rates for given station. It @@ -6226,8 +6290,9 @@ static void ath12k_sta_rc_update_wk(struct wiphy *wiphy, struct wiphy_work *wk) const u8 *ht_mcs_mask; const u16 *vht_mcs_mask; const u16 *he_mcs_mask; + const u16 *eht_mcs_mask; u32 changed, bw, nss, mac_nss, smps, bw_prev; - int err, num_vht_rates, num_he_rates; + int err, num_vht_rates, num_he_rates, num_eht_rates; const struct cfg80211_bitrate_mask *mask; enum wmi_phy_mode peer_phymode; struct ath12k_link_sta *arsta; @@ -6248,6 +6313,7 @@ static void ath12k_sta_rc_update_wk(struct wiphy *wiphy, struct wiphy_work *wk) ht_mcs_mask = arvif->bitrate_mask.control[band].ht_mcs; vht_mcs_mask = arvif->bitrate_mask.control[band].vht_mcs; he_mcs_mask = arvif->bitrate_mask.control[band].he_mcs; + eht_mcs_mask = arvif->bitrate_mask.control[band].eht_mcs; spin_lock_bh(&ar->data_lock); @@ -6265,6 +6331,7 @@ static void ath12k_sta_rc_update_wk(struct wiphy *wiphy, struct wiphy_work *wk) mac_nss = max3(ath12k_mac_max_ht_nss(ht_mcs_mask), ath12k_mac_max_vht_nss(vht_mcs_mask), ath12k_mac_max_he_nss(he_mcs_mask)); + mac_nss = max(mac_nss, ath12k_mac_max_eht_nss(eht_mcs_mask)); nss = min(nss, mac_nss); struct ath12k_wmi_peer_assoc_arg *peer_arg __free(kfree) = @@ -6350,6 +6417,8 @@ static void ath12k_sta_rc_update_wk(struct wiphy *wiphy, struct wiphy_work *wk) mask); num_he_rates = ath12k_mac_bitrate_mask_num_he_rates(ar, band, mask); + num_eht_rates = ath12k_mac_bitrate_mask_num_eht_rates(ar, band, + mask); /* Peer_assoc_prepare will reject vht rates in * bitrate_mask if its not available in range format and @@ -6374,9 +6443,18 @@ static void ath12k_sta_rc_update_wk(struct wiphy *wiphy, struct wiphy_work *wk) band); } else if (link_sta->he_cap.has_he && num_he_rates == 1) { ath12k_mac_set_peer_he_fixed_rate(arvif, arsta, mask, band); + } else if (link_sta->eht_cap.has_eht && num_eht_rates == 1) { + err = ath12k_mac_set_peer_eht_fixed_rate(arvif, arsta, + mask, band); + if (err) { + ath12k_warn(ar->ab, + "failed to set peer EHT fixed rate for STA %pM ret %d\n", + arsta->addr, err); + return; + } } else { - /* If the peer is non-VHT/HE or no fixed VHT/HE rate - * is provided in the new bitrate mask we set the + /* If the peer is non-VHT/HE/EHT or no fixed VHT/HE/EHT + * rate is provided in the new bitrate mask we set the * other rates using peer_assoc command. Also clear * the peer fixed rate settings as it has higher proprity * than peer assoc From 448bf7b51426bcca54b5ac1ddd1045a36c9d1dea Mon Sep 17 00:00:00 2001 From: Rameshkumar Sundaram Date: Sun, 26 Oct 2025 23:52:53 +0530 Subject: [PATCH 0413/1381] wifi: ath12k: enforce vdev limit in ath12k_mac_vdev_create() Currently, vdev limit check is performed only in ath12k_mac_assign_vif_to_vdev(). If the host has already created maximum number of vdevs for the radio (ar) and a scan request arrives for the same radio, ath12k_mac_initiate_hw_scan() attempts to create a vdev without checking the limit, causing firmware asserts. Centralize the vdev limit guard by moving the check into ath12k_mac_vdev_create() so that all callers obey the limit. While doing this, update the condition from `num_created_vdevs > (TARGET_NUM_VDEVS(ab) - 1)` to `num_created_vdevs >= TARGET_NUM_VDEVS(ab)` for clarity and to eliminate unnecessary arithmetic. Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.4.1-00199-QCAHKSWPL_SILICONZ-1 Fixes: 0d6e6736ed9f ("wifi: ath12k: scan statemachine changes for single wiphy") Fixes: 4938ba733ee2 ("wifi: ath12k: modify remain on channel for single wiphy") Signed-off-by: Rameshkumar Sundaram Reviewed-by: Vasanthakumar Thiagarajan Link: https://patch.msgid.link/20251026182254.1399650-2-rameshkumar.sundaram@oss.qualcomm.com Signed-off-by: Jeff Johnson --- drivers/net/wireless/ath/ath12k/mac.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c index 29a0b022c32e..8f0d914d7fc3 100644 --- a/drivers/net/wireless/ath/ath12k/mac.c +++ b/drivers/net/wireless/ath/ath12k/mac.c @@ -10031,6 +10031,12 @@ int ath12k_mac_vdev_create(struct ath12k *ar, struct ath12k_link_vif *arvif) if (vif->type == NL80211_IFTYPE_MONITOR && ar->monitor_vdev_created) return -EINVAL; + if (ar->num_created_vdevs >= TARGET_NUM_VDEVS(ab)) { + ath12k_warn(ab, "failed to create vdev, reached max vdev limit %d\n", + TARGET_NUM_VDEVS(ab)); + return -ENOSPC; + } + link_id = arvif->link_id; if (link_id < IEEE80211_MLD_MAX_NUM_LINKS) { @@ -10390,12 +10396,6 @@ static struct ath12k *ath12k_mac_assign_vif_to_vdev(struct ieee80211_hw *hw, if (arvif->is_created) goto flush; - if (ar->num_created_vdevs > (TARGET_NUM_VDEVS(ab) - 1)) { - ath12k_warn(ab, "failed to create vdev, reached max vdev limit %d\n", - TARGET_NUM_VDEVS(ab)); - goto unlock; - } - ret = ath12k_mac_vdev_create(ar, arvif); if (ret) { ath12k_warn(ab, "failed to create vdev %pM ret %d", vif->addr, ret); From e70515039d44be61b6a73aafb401d141b0034d12 Mon Sep 17 00:00:00 2001 From: Rameshkumar Sundaram Date: Sun, 26 Oct 2025 23:52:54 +0530 Subject: [PATCH 0414/1381] wifi: ath12k: unassign arvif on scan vdev create failure During scan and remain-on-channel requests, a scan link vif (arvif) is assigned and a temporary vdev is created. If vdev creation fails, the assigned arvif is left attached until the virtual interface is removed, leaving a stale link in ahvif. Fix this by freeing the stale arvif and resetting the corresponding link in ahvif by calling ath12k_mac_unassign_link_vif() when vdev creation fails. While at it, propagate the actual error code from ath12k_mac_vdev_create() instead of returning -EINVAL in ath12k_mac_initiate_hw_scan(). Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.4.1-00199-QCAHKSWPL_SILICONZ-1 Fixes: 477cabfdb776 ("wifi: ath12k: modify link arvif creation and removal for MLO") Signed-off-by: Rameshkumar Sundaram Reviewed-by: Vasanthakumar Thiagarajan Link: https://patch.msgid.link/20251026182254.1399650-3-rameshkumar.sundaram@oss.qualcomm.com Signed-off-by: Jeff Johnson --- drivers/net/wireless/ath/ath12k/mac.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c index 8f0d914d7fc3..7639fb2c52fe 100644 --- a/drivers/net/wireless/ath/ath12k/mac.c +++ b/drivers/net/wireless/ath/ath12k/mac.c @@ -5313,7 +5313,8 @@ static int ath12k_mac_initiate_hw_scan(struct ieee80211_hw *hw, ret = ath12k_mac_vdev_create(ar, arvif); if (ret) { ath12k_warn(ar->ab, "unable to create scan vdev %d\n", ret); - return -EINVAL; + ath12k_mac_unassign_link_vif(arvif); + return ret; } } @@ -13411,6 +13412,7 @@ static int ath12k_mac_op_remain_on_channel(struct ieee80211_hw *hw, if (ret) { ath12k_warn(ar->ab, "unable to create scan vdev for roc: %d\n", ret); + ath12k_mac_unassign_link_vif(arvif); return ret; } } From 877f9c22fdf424c657de757bfe8543cf77461324 Mon Sep 17 00:00:00 2001 From: Zhongqiu Han Date: Thu, 30 Oct 2025 21:00:22 +0800 Subject: [PATCH 0415/1381] wifi: ath10k: use = {} to initialize pm_qos_request instead of memset Initialize the pm_qos_request structure using = {} instead of memset() in ath10k_download_fw(). This ensures the structure is properly zeroed before passing it to cpu_latency_qos_add_request(), and improves efficiency by avoiding an explicit runtime memset. Signed-off-by: Zhongqiu Han Reviewed-by: Vasanthakumar Thiagarajan Link: https://patch.msgid.link/20251030130023.1836808-2-zhongqiu.han@oss.qualcomm.com Signed-off-by: Jeff Johnson --- drivers/net/wireless/ath/ath10k/core.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/net/wireless/ath/ath10k/core.c b/drivers/net/wireless/ath/ath10k/core.c index 9ae3595fb698..670c31a52a12 100644 --- a/drivers/net/wireless/ath/ath10k/core.c +++ b/drivers/net/wireless/ath/ath10k/core.c @@ -1186,7 +1186,7 @@ static int ath10k_download_fw(struct ath10k *ar) u32 address, data_len; const void *data; int ret; - struct pm_qos_request latency_qos; + struct pm_qos_request latency_qos = {}; address = ar->hw_params.patch_load_addr; @@ -1220,7 +1220,6 @@ static int ath10k_download_fw(struct ath10k *ar) ret); } - memset(&latency_qos, 0, sizeof(latency_qos)); cpu_latency_qos_add_request(&latency_qos, 0); ret = ath10k_bmi_fast_download(ar, address, data, data_len); From 059ca8fd692b67a77fb89e9d4e8f57cf08e32b08 Mon Sep 17 00:00:00 2001 From: Zhongqiu Han Date: Thu, 30 Oct 2025 21:00:23 +0800 Subject: [PATCH 0416/1381] wifi: ath10k: use = {} to initialize bmi_target_info instead of memset Initialize the bmi_target_info structure using = {} at declaration time instead of calling memset() in each bus-specific code path. This simplifies the code and avoids an explicit memset. Signed-off-by: Zhongqiu Han Reviewed-by: Vasanthakumar Thiagarajan Link: https://patch.msgid.link/20251030130023.1836808-3-zhongqiu.han@oss.qualcomm.com Signed-off-by: Jeff Johnson --- drivers/net/wireless/ath/ath10k/core.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/drivers/net/wireless/ath/ath10k/core.c b/drivers/net/wireless/ath/ath10k/core.c index 670c31a52a12..7c2939cbde5f 100644 --- a/drivers/net/wireless/ath/ath10k/core.c +++ b/drivers/net/wireless/ath/ath10k/core.c @@ -3352,7 +3352,7 @@ EXPORT_SYMBOL(ath10k_core_stop); */ static int ath10k_core_probe_fw(struct ath10k *ar) { - struct bmi_target_info target_info; + struct bmi_target_info target_info = {}; int ret = 0; ret = ath10k_hif_power_up(ar, ATH10K_FIRMWARE_MODE_NORMAL); @@ -3363,7 +3363,6 @@ static int ath10k_core_probe_fw(struct ath10k *ar) switch (ar->hif.bus) { case ATH10K_BUS_SDIO: - memset(&target_info, 0, sizeof(target_info)); ret = ath10k_bmi_get_target_info_sdio(ar, &target_info); if (ret) { ath10k_err(ar, "could not get target info (%d)\n", ret); @@ -3375,7 +3374,6 @@ static int ath10k_core_probe_fw(struct ath10k *ar) case ATH10K_BUS_PCI: case ATH10K_BUS_AHB: case ATH10K_BUS_USB: - memset(&target_info, 0, sizeof(target_info)); ret = ath10k_bmi_get_target_info(ar, &target_info); if (ret) { ath10k_err(ar, "could not get target info (%d)\n", ret); @@ -3385,7 +3383,6 @@ static int ath10k_core_probe_fw(struct ath10k *ar) ar->hw->wiphy->hw_version = target_info.version; break; case ATH10K_BUS_SNOC: - memset(&target_info, 0, sizeof(target_info)); ret = ath10k_hif_get_target_info(ar, &target_info); if (ret) { ath10k_err(ar, "could not get target info (%d)\n", ret); From ecca75ae5ae66d13f701c4a6523644d51946ddec Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 29 Oct 2025 09:49:30 -0700 Subject: [PATCH 0417/1381] selftests: drv-net: replace the nsim ring test with a drv-net one We are trying to move away from netdevsim-only tests and towards tests which can be run both against netdevsim and real drivers. Replace the simple bash script we have for checking ethtool -g/-G on netdevsim with a Python test tweaking those params as well as channel count. The new test is not exactly equivalent to the netdevsim one, but real drivers don't often support random ring sizes, let alone modifying max values via debugfs. Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20251029164930.2923448-1-kuba@kernel.org Signed-off-by: Jakub Kicinski --- tools/testing/selftests/drivers/net/Makefile | 1 + .../selftests/drivers/net/netdevsim/Makefile | 1 - .../drivers/net/netdevsim/ethtool-ring.sh | 85 --------- .../selftests/drivers/net/ring_reconfig.py | 167 ++++++++++++++++++ 4 files changed, 168 insertions(+), 86 deletions(-) delete mode 100755 tools/testing/selftests/drivers/net/netdevsim/ethtool-ring.sh create mode 100755 tools/testing/selftests/drivers/net/ring_reconfig.py diff --git a/tools/testing/selftests/drivers/net/Makefile b/tools/testing/selftests/drivers/net/Makefile index 6e41635bd55a..68e0bb603a9d 100644 --- a/tools/testing/selftests/drivers/net/Makefile +++ b/tools/testing/selftests/drivers/net/Makefile @@ -22,6 +22,7 @@ TEST_PROGS := \ ping.py \ psp.py \ queues.py \ + ring_reconfig.py \ shaper.py \ stats.py \ xdp.py \ diff --git a/tools/testing/selftests/drivers/net/netdevsim/Makefile b/tools/testing/selftests/drivers/net/netdevsim/Makefile index daf51113c827..833abd8e6fdc 100644 --- a/tools/testing/selftests/drivers/net/netdevsim/Makefile +++ b/tools/testing/selftests/drivers/net/netdevsim/Makefile @@ -8,7 +8,6 @@ TEST_PROGS := \ ethtool-features.sh \ ethtool-fec.sh \ ethtool-pause.sh \ - ethtool-ring.sh \ fib.sh \ fib_notifications.sh \ hw_stats_l3.sh \ diff --git a/tools/testing/selftests/drivers/net/netdevsim/ethtool-ring.sh b/tools/testing/selftests/drivers/net/netdevsim/ethtool-ring.sh deleted file mode 100755 index c969559ffa7a..000000000000 --- a/tools/testing/selftests/drivers/net/netdevsim/ethtool-ring.sh +++ /dev/null @@ -1,85 +0,0 @@ -#!/bin/bash -# SPDX-License-Identifier: GPL-2.0-only - -source ethtool-common.sh - -function get_value { - local query="${SETTINGS_MAP[$1]}" - - echo $(ethtool -g $NSIM_NETDEV | \ - tail -n +$CURR_SETT_LINE | \ - awk -F':' -v pattern="$query:" '$0 ~ pattern {gsub(/[\t ]/, "", $2); print $2}') -} - -function update_current_settings { - for key in ${!SETTINGS_MAP[@]}; do - CURRENT_SETTINGS[$key]=$(get_value $key) - done - echo ${CURRENT_SETTINGS[@]} -} - -if ! ethtool -h | grep -q set-ring >/dev/null; then - echo "SKIP: No --set-ring support in ethtool" - exit 4 -fi - -NSIM_NETDEV=$(make_netdev) - -set -o pipefail - -declare -A SETTINGS_MAP=( - ["rx"]="RX" - ["rx-mini"]="RX Mini" - ["rx-jumbo"]="RX Jumbo" - ["tx"]="TX" -) - -declare -A EXPECTED_SETTINGS=( - ["rx"]="" - ["rx-mini"]="" - ["rx-jumbo"]="" - ["tx"]="" -) - -declare -A CURRENT_SETTINGS=( - ["rx"]="" - ["rx-mini"]="" - ["rx-jumbo"]="" - ["tx"]="" -) - -MAX_VALUE=$((RANDOM % $((2**32-1)))) -RING_MAX_LIST=$(ls $NSIM_DEV_DFS/ethtool/ring/) - -for ring_max_entry in $RING_MAX_LIST; do - echo $MAX_VALUE > $NSIM_DEV_DFS/ethtool/ring/$ring_max_entry -done - -CURR_SETT_LINE=$(ethtool -g $NSIM_NETDEV | grep -i -m1 -n 'Current hardware settings' | cut -f1 -d:) - -# populate the expected settings map -for key in ${!SETTINGS_MAP[@]}; do - EXPECTED_SETTINGS[$key]=$(get_value $key) -done - -# test -for key in ${!SETTINGS_MAP[@]}; do - value=$((RANDOM % $MAX_VALUE)) - - ethtool -G $NSIM_NETDEV "$key" "$value" - - EXPECTED_SETTINGS[$key]="$value" - expected=${EXPECTED_SETTINGS[@]} - current=$(update_current_settings) - - check $? "$current" "$expected" - set +x -done - -if [ $num_errors -eq 0 ]; then - echo "PASSED all $((num_passes)) checks" - exit 0 -else - echo "FAILED $num_errors/$((num_errors+num_passes)) checks" - exit 1 -fi diff --git a/tools/testing/selftests/drivers/net/ring_reconfig.py b/tools/testing/selftests/drivers/net/ring_reconfig.py new file mode 100755 index 000000000000..f9530a8b0856 --- /dev/null +++ b/tools/testing/selftests/drivers/net/ring_reconfig.py @@ -0,0 +1,167 @@ +#!/usr/bin/env python3 +# SPDX-License-Identifier: GPL-2.0 + +""" +Test channel and ring size configuration via ethtool (-L / -G). +""" + +from lib.py import ksft_run, ksft_exit, ksft_pr +from lib.py import ksft_eq +from lib.py import NetDrvEpEnv, EthtoolFamily, GenerateTraffic +from lib.py import defer, NlError + + +def channels(cfg) -> None: + """ + Twiddle channel counts in various combinations of parameters. + We're only looking for driver adhering to the requested config + if the config is accepted and crashes. + """ + ehdr = {'header':{'dev-index': cfg.ifindex}} + chans = cfg.eth.channels_get(ehdr) + + all_keys = ["rx", "tx", "combined"] + mixes = [{"combined"}, {"rx", "tx"}, {"rx", "combined"}, {"tx", "combined"}, + {"rx", "tx", "combined"},] + + # Get the set of keys that device actually supports + restore = {} + supported = set() + for key in all_keys: + if key + "-max" in chans: + supported.add(key) + restore |= {key + "-count": chans[key + "-count"]} + + defer(cfg.eth.channels_set, ehdr | restore) + + def test_config(config): + try: + cfg.eth.channels_set(ehdr | config) + get = cfg.eth.channels_get(ehdr) + for k, v in config.items(): + ksft_eq(get.get(k, 0), v) + except NlError as e: + failed.append(mix) + ksft_pr("Can't set", config, e) + else: + ksft_pr("Okay", config) + + failed = [] + for mix in mixes: + if not mix.issubset(supported): + continue + + # Set all the values in the mix to 1, other supported to 0 + config = {} + for key in all_keys: + config[key + "-count"] = 1 if key in mix else 0 + test_config(config) + + for mix in mixes: + if not mix.issubset(supported): + continue + if mix in failed: + continue + + # Set all the values in the mix to max, other supported to 0 + config = {} + for key in all_keys: + config[key + "-count"] = chans[key + '-max'] if key in mix else 0 + test_config(config) + + +def _configure_min_ring_cnt(cfg) -> None: + """ Try to configure a single Rx/Tx ring. """ + ehdr = {'header':{'dev-index': cfg.ifindex}} + chans = cfg.eth.channels_get(ehdr) + + all_keys = ["rx-count", "tx-count", "combined-count"] + restore = {} + config = {} + for key in all_keys: + if key in chans: + restore[key] = chans[key] + config[key] = 0 + + if chans.get('combined-count', 0) > 1: + config['combined-count'] = 1 + elif chans.get('rx-count', 0) > 1 and chans.get('tx-count', 0) > 1: + config['tx-count'] = 1 + config['rx-count'] = 1 + else: + # looks like we're already on 1 channel + return + + cfg.eth.channels_set(ehdr | config) + defer(cfg.eth.channels_set, ehdr | restore) + + +def ringparam(cfg) -> None: + """ + Tweak the ringparam configuration. Try to run some traffic over min + ring size to make sure it actually functions. + """ + ehdr = {'header':{'dev-index': cfg.ifindex}} + rings = cfg.eth.rings_get(ehdr) + + restore = {} + maxes = {} + params = set() + for key in rings.keys(): + if 'max' in key: + param = key[:-4] + maxes[param] = rings[key] + params.add(param) + restore[param] = rings[param] + + defer(cfg.eth.rings_set, ehdr | restore) + + # Speed up the reconfig by configuring just one ring + _configure_min_ring_cnt(cfg) + + # Try to reach min on all settings + for param in params: + val = rings[param] + while True: + try: + cfg.eth.rings_set({'header':{'dev-index': cfg.ifindex}, + param: val // 2}) + if val == 0: + break + val //= 2 + except NlError: + break + + get = cfg.eth.rings_get(ehdr) + ksft_eq(get[param], val) + + ksft_pr(f"Reached min for '{param}' at {val} (max {rings[param]})") + + GenerateTraffic(cfg).wait_pkts_and_stop(10000) + + # Try max across all params, if the driver supports large rings + # this may OOM so we ignore errors + try: + ksft_pr("Applying max settings") + config = {p: maxes[p] for p in params} + cfg.eth.rings_set(ehdr | config) + except NlError as e: + ksft_pr("Can't set max params", config, e) + else: + GenerateTraffic(cfg).wait_pkts_and_stop(10000) + + +def main() -> None: + """ Ksft boiler plate main """ + + with NetDrvEpEnv(__file__) as cfg: + cfg.eth = EthtoolFamily() + + ksft_run([channels, + ringparam], + args=(cfg, )) + ksft_exit() + + +if __name__ == "__main__": + main() From 0d0eb186421d0886ac466008235f6d9eedaf918e Mon Sep 17 00:00:00 2001 From: Harshita V Rajput Date: Tue, 28 Oct 2025 13:22:55 +0530 Subject: [PATCH 0418/1381] cxgb4: flower: add support for fragmentation This patch adds support for matching fragmented packets in tc flower filters. Previously, commit 93a8540aac72 ("cxgb4: flower: validate control flags") added a check using flow_rule_match_has_control_flags() to reject any rules with control flags, as the driver did not support fragmentation at that time. Now, with this patch, support for FLOW_DIS_IS_FRAGMENT is added: - The driver checks for control flags using flow_rule_is_supp_control_flags(), as recommended in commit d11e63119432 ("flow_offload: add control flag checking helpers"). - If the fragmentation flag is present, the driver sets `fs->val.frag` and `fs->mask.frag` accordingly in the filter specification. Since fragmentation is now supported, the earlier check that rejected all control flags (flow_rule_match_has_control_flags()) has been removed. Signed-off-by: Harshita V Rajput Signed-off-by: Potnuri Bharat Teja Reviewed-by: Simon Horman Link: https://patch.msgid.link/20251028075255.1391596-1-harshitha.vr@chelsio.com Signed-off-by: Jakub Kicinski --- .../ethernet/chelsio/cxgb4/cxgb4_tc_flower.c | 40 +++++++++++-------- 1 file changed, 24 insertions(+), 16 deletions(-) diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c index 0765d000eaef..e2b5554531b5 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c @@ -161,20 +161,9 @@ static struct ch_tc_flower_entry *ch_flower_lookup(struct adapter *adap, static void cxgb4_process_flow_match(struct net_device *dev, struct flow_rule *rule, + u16 addr_type, struct ch_filter_specification *fs) { - u16 addr_type = 0; - - if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_CONTROL)) { - struct flow_match_control match; - - flow_rule_match_control(rule, &match); - addr_type = match.key->addr_type; - } else if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_IPV4_ADDRS)) { - addr_type = FLOW_DISSECTOR_KEY_IPV4_ADDRS; - } else if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_IPV6_ADDRS)) { - addr_type = FLOW_DISSECTOR_KEY_IPV6_ADDRS; - } if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_BASIC)) { struct flow_match_basic match; @@ -327,9 +316,6 @@ static int cxgb4_validate_flow_match(struct netlink_ext_ack *extack, return -EOPNOTSUPP; } - if (flow_rule_match_has_control_flags(rule, extack)) - return -EOPNOTSUPP; - if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_BASIC)) { struct flow_match_basic match; @@ -858,6 +844,7 @@ int cxgb4_flow_rule_replace(struct net_device *dev, struct flow_rule *rule, { struct adapter *adap = netdev2adap(dev); struct filter_ctx ctx; + u16 addr_type = 0; u8 inet_family; int fidx, ret; @@ -867,7 +854,28 @@ int cxgb4_flow_rule_replace(struct net_device *dev, struct flow_rule *rule, if (cxgb4_validate_flow_match(extack, rule)) return -EOPNOTSUPP; - cxgb4_process_flow_match(dev, rule, fs); + if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_CONTROL)) { + struct flow_match_control match; + + flow_rule_match_control(rule, &match); + addr_type = match.key->addr_type; + + if (match.mask->flags & FLOW_DIS_IS_FRAGMENT) { + fs->val.frag = match.key->flags & FLOW_DIS_IS_FRAGMENT; + fs->mask.frag = true; + } + + if (!flow_rule_is_supp_control_flags(FLOW_DIS_IS_FRAGMENT, + match.mask->flags, extack)) + return -EOPNOTSUPP; + + } else if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_IPV4_ADDRS)) { + addr_type = FLOW_DISSECTOR_KEY_IPV4_ADDRS; + } else if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_IPV6_ADDRS)) { + addr_type = FLOW_DISSECTOR_KEY_IPV6_ADDRS; + } + + cxgb4_process_flow_match(dev, rule, addr_type, fs); cxgb4_process_flow_actions(dev, &rule->action, fs); fs->hash = is_filter_exact_match(adap, fs); From 4f6b0435c613fdb76d85bb4aae009309a8ce8784 Mon Sep 17 00:00:00 2001 From: Vadim Fedorenko Date: Wed, 29 Oct 2025 23:16:18 +0000 Subject: [PATCH 0419/1381] can: convert generic HW timestamp ioctl to ndo_hwtstamp callbacks Can has generic implementation of ndo_eth_ioctl which implements only HW timestamping commands. Implement generic ndo_hwtstamp callbacks and use it in drivers instead of generic ioctl interface. Signed-off-by: Vadim Fedorenko Reviewed-by: Kory Maincent Reviewed-by: Vincent Mailhol Link: https://patch.msgid.link/20251029231620.1135640-2-vadim.fedorenko@linux.dev Signed-off-by: Marc Kleine-Budde --- drivers/net/can/dev/dev.c | 45 +++++++++---------- drivers/net/can/esd/esd_402_pci-core.c | 3 +- .../can/kvaser_pciefd/kvaser_pciefd_core.c | 3 +- .../net/can/spi/mcp251xfd/mcp251xfd-core.c | 3 +- drivers/net/can/usb/etas_es58x/es58x_core.c | 3 +- drivers/net/can/usb/gs_usb.c | 20 +++++++-- .../net/can/usb/kvaser_usb/kvaser_usb_core.c | 3 +- include/linux/can/dev.h | 6 ++- 8 files changed, 54 insertions(+), 32 deletions(-) diff --git a/drivers/net/can/dev/dev.c b/drivers/net/can/dev/dev.c index 0cc3d008adb3..80e1ab18de87 100644 --- a/drivers/net/can/dev/dev.c +++ b/drivers/net/can/dev/dev.c @@ -379,34 +379,33 @@ int can_set_static_ctrlmode(struct net_device *dev, u32 static_mode) } EXPORT_SYMBOL_GPL(can_set_static_ctrlmode); -/* generic implementation of netdev_ops::ndo_eth_ioctl for CAN devices +/* generic implementation of netdev_ops::ndo_hwtstamp_get for CAN devices * supporting hardware timestamps */ -int can_eth_ioctl_hwts(struct net_device *netdev, struct ifreq *ifr, int cmd) +int can_hwtstamp_get(struct net_device *netdev, + struct kernel_hwtstamp_config *cfg) { - struct hwtstamp_config hwts_cfg = { 0 }; + cfg->tx_type = HWTSTAMP_TX_ON; + cfg->rx_filter = HWTSTAMP_FILTER_ALL; - switch (cmd) { - case SIOCSHWTSTAMP: /* set */ - if (copy_from_user(&hwts_cfg, ifr->ifr_data, sizeof(hwts_cfg))) - return -EFAULT; - if (hwts_cfg.tx_type == HWTSTAMP_TX_ON && - hwts_cfg.rx_filter == HWTSTAMP_FILTER_ALL) - return 0; - return -ERANGE; - - case SIOCGHWTSTAMP: /* get */ - hwts_cfg.tx_type = HWTSTAMP_TX_ON; - hwts_cfg.rx_filter = HWTSTAMP_FILTER_ALL; - if (copy_to_user(ifr->ifr_data, &hwts_cfg, sizeof(hwts_cfg))) - return -EFAULT; - return 0; - - default: - return -EOPNOTSUPP; - } + return 0; } -EXPORT_SYMBOL(can_eth_ioctl_hwts); +EXPORT_SYMBOL(can_hwtstamp_get); + +/* generic implementation of netdev_ops::ndo_hwtstamp_set for CAN devices + * supporting hardware timestamps + */ +int can_hwtstamp_set(struct net_device *netdev, + struct kernel_hwtstamp_config *cfg, + struct netlink_ext_ack *extack) +{ + if (cfg->tx_type == HWTSTAMP_TX_ON && + cfg->rx_filter == HWTSTAMP_FILTER_ALL) + return 0; + NL_SET_ERR_MSG_MOD(extack, "Only TX on and RX all packets filter supported"); + return -ERANGE; +} +EXPORT_SYMBOL(can_hwtstamp_set); /* generic implementation of ethtool_ops::get_ts_info for CAN devices * supporting hardware timestamps diff --git a/drivers/net/can/esd/esd_402_pci-core.c b/drivers/net/can/esd/esd_402_pci-core.c index 05adecae6375..c826f00c551b 100644 --- a/drivers/net/can/esd/esd_402_pci-core.c +++ b/drivers/net/can/esd/esd_402_pci-core.c @@ -86,7 +86,8 @@ static const struct net_device_ops pci402_acc_netdev_ops = { .ndo_open = acc_open, .ndo_stop = acc_close, .ndo_start_xmit = acc_start_xmit, - .ndo_eth_ioctl = can_eth_ioctl_hwts, + .ndo_hwtstamp_get = can_hwtstamp_get, + .ndo_hwtstamp_set = can_hwtstamp_set, }; static const struct ethtool_ops pci402_acc_ethtool_ops = { diff --git a/drivers/net/can/kvaser_pciefd/kvaser_pciefd_core.c b/drivers/net/can/kvaser_pciefd/kvaser_pciefd_core.c index 705f9bb74cd2..d8c9bfb20230 100644 --- a/drivers/net/can/kvaser_pciefd/kvaser_pciefd_core.c +++ b/drivers/net/can/kvaser_pciefd/kvaser_pciefd_core.c @@ -902,8 +902,9 @@ static void kvaser_pciefd_bec_poll_timer(struct timer_list *data) static const struct net_device_ops kvaser_pciefd_netdev_ops = { .ndo_open = kvaser_pciefd_open, .ndo_stop = kvaser_pciefd_stop, - .ndo_eth_ioctl = can_eth_ioctl_hwts, .ndo_start_xmit = kvaser_pciefd_start_xmit, + .ndo_hwtstamp_get = can_hwtstamp_get, + .ndo_hwtstamp_set = can_hwtstamp_set, }; static int kvaser_pciefd_set_phys_id(struct net_device *netdev, diff --git a/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c b/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c index 9402530ba3d4..c0f9d9fed02e 100644 --- a/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c +++ b/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c @@ -1714,7 +1714,8 @@ static const struct net_device_ops mcp251xfd_netdev_ops = { .ndo_open = mcp251xfd_open, .ndo_stop = mcp251xfd_stop, .ndo_start_xmit = mcp251xfd_start_xmit, - .ndo_eth_ioctl = can_eth_ioctl_hwts, + .ndo_hwtstamp_get = can_hwtstamp_get, + .ndo_hwtstamp_set = can_hwtstamp_set, }; static void diff --git a/drivers/net/can/usb/etas_es58x/es58x_core.c b/drivers/net/can/usb/etas_es58x/es58x_core.c index 47d9e03f3044..f799233c2b72 100644 --- a/drivers/net/can/usb/etas_es58x/es58x_core.c +++ b/drivers/net/can/usb/etas_es58x/es58x_core.c @@ -1976,7 +1976,8 @@ static const struct net_device_ops es58x_netdev_ops = { .ndo_open = es58x_open, .ndo_stop = es58x_stop, .ndo_start_xmit = es58x_start_xmit, - .ndo_eth_ioctl = can_eth_ioctl_hwts, + .ndo_hwtstamp_get = can_hwtstamp_get, + .ndo_hwtstamp_set = can_hwtstamp_set, }; static const struct ethtool_ops es58x_ethtool_ops = { diff --git a/drivers/net/can/usb/gs_usb.c b/drivers/net/can/usb/gs_usb.c index 30608901a974..1321eb5e89ae 100644 --- a/drivers/net/can/usb/gs_usb.c +++ b/drivers/net/can/usb/gs_usb.c @@ -1087,12 +1087,25 @@ static int gs_can_close(struct net_device *netdev) return 0; } -static int gs_can_eth_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) +static int gs_can_hwtstamp_get(struct net_device *netdev, + struct kernel_hwtstamp_config *cfg) { const struct gs_can *dev = netdev_priv(netdev); if (dev->feature & GS_CAN_FEATURE_HW_TIMESTAMP) - return can_eth_ioctl_hwts(netdev, ifr, cmd); + return can_hwtstamp_get(netdev, cfg); + + return -EOPNOTSUPP; +} + +static int gs_can_hwtstamp_set(struct net_device *netdev, + struct kernel_hwtstamp_config *cfg, + struct netlink_ext_ack *extack) +{ + const struct gs_can *dev = netdev_priv(netdev); + + if (dev->feature & GS_CAN_FEATURE_HW_TIMESTAMP) + return can_hwtstamp_set(netdev, cfg, extack); return -EOPNOTSUPP; } @@ -1101,7 +1114,8 @@ static const struct net_device_ops gs_usb_netdev_ops = { .ndo_open = gs_can_open, .ndo_stop = gs_can_close, .ndo_start_xmit = gs_can_start_xmit, - .ndo_eth_ioctl = gs_can_eth_ioctl, + .ndo_hwtstamp_get = gs_can_hwtstamp_get, + .ndo_hwtstamp_set = gs_can_hwtstamp_set, }; static int gs_usb_set_identify(struct net_device *netdev, bool do_identify) diff --git a/drivers/net/can/usb/kvaser_usb/kvaser_usb_core.c b/drivers/net/can/usb/kvaser_usb/kvaser_usb_core.c index 89e22b66f919..62701ec34272 100644 --- a/drivers/net/can/usb/kvaser_usb/kvaser_usb_core.c +++ b/drivers/net/can/usb/kvaser_usb/kvaser_usb_core.c @@ -784,8 +784,9 @@ static int kvaser_usb_set_phys_id(struct net_device *netdev, static const struct net_device_ops kvaser_usb_netdev_ops = { .ndo_open = kvaser_usb_open, .ndo_stop = kvaser_usb_close, - .ndo_eth_ioctl = can_eth_ioctl_hwts, .ndo_start_xmit = kvaser_usb_start_xmit, + .ndo_hwtstamp_get = can_hwtstamp_get, + .ndo_hwtstamp_set = can_hwtstamp_set, }; static const struct ethtool_ops kvaser_usb_ethtool_ops = { diff --git a/include/linux/can/dev.h b/include/linux/can/dev.h index 0fe8f80f223e..bd7410b5d8a6 100644 --- a/include/linux/can/dev.h +++ b/include/linux/can/dev.h @@ -129,7 +129,11 @@ void close_candev(struct net_device *dev); void can_set_default_mtu(struct net_device *dev); int __must_check can_set_static_ctrlmode(struct net_device *dev, u32 static_mode); -int can_eth_ioctl_hwts(struct net_device *netdev, struct ifreq *ifr, int cmd); +int can_hwtstamp_get(struct net_device *netdev, + struct kernel_hwtstamp_config *cfg); +int can_hwtstamp_set(struct net_device *netdev, + struct kernel_hwtstamp_config *cfg, + struct netlink_ext_ack *extack); int can_ethtool_op_get_ts_info_hwts(struct net_device *dev, struct kernel_ethtool_ts_info *info); From 336e2232583018c7f386fbc33c8944922a18d7ba Mon Sep 17 00:00:00 2001 From: Vadim Fedorenko Date: Wed, 29 Oct 2025 23:16:19 +0000 Subject: [PATCH 0420/1381] can: peak_canfd: convert to use ndo_hwtstamp callbacks Convert driver to use ndo_hwtstamp_set()/ndo_hwtstamp_get() callbacks. ndo_eth_ioctl handler does nothing after conversion - remove it. Signed-off-by: Vadim Fedorenko Reviewed-by: Kory Maincent Reviewed-by: Vincent Mailhol Link: https://patch.msgid.link/20251029231620.1135640-3-vadim.fedorenko@linux.dev Signed-off-by: Marc Kleine-Budde --- drivers/net/can/peak_canfd/peak_canfd.c | 35 +++++++++++-------------- 1 file changed, 16 insertions(+), 19 deletions(-) diff --git a/drivers/net/can/peak_canfd/peak_canfd.c b/drivers/net/can/peak_canfd/peak_canfd.c index a53c9d347b7b..06cb2629f66a 100644 --- a/drivers/net/can/peak_canfd/peak_canfd.c +++ b/drivers/net/can/peak_canfd/peak_canfd.c @@ -743,36 +743,33 @@ static netdev_tx_t peak_canfd_start_xmit(struct sk_buff *skb, return NETDEV_TX_OK; } -static int peak_eth_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) +static int peak_eth_hwtstamp_get(struct net_device *netdev, + struct kernel_hwtstamp_config *config) { - struct hwtstamp_config hwts_cfg = { 0 }; + config->tx_type = HWTSTAMP_TX_OFF; + config->rx_filter = HWTSTAMP_FILTER_ALL; - switch (cmd) { - case SIOCSHWTSTAMP: /* set */ - if (copy_from_user(&hwts_cfg, ifr->ifr_data, sizeof(hwts_cfg))) - return -EFAULT; - if (hwts_cfg.tx_type == HWTSTAMP_TX_OFF && - hwts_cfg.rx_filter == HWTSTAMP_FILTER_ALL) - return 0; - return -ERANGE; + return 0; +} - case SIOCGHWTSTAMP: /* get */ - hwts_cfg.tx_type = HWTSTAMP_TX_OFF; - hwts_cfg.rx_filter = HWTSTAMP_FILTER_ALL; - if (copy_to_user(ifr->ifr_data, &hwts_cfg, sizeof(hwts_cfg))) - return -EFAULT; +static int peak_eth_hwtstamp_set(struct net_device *netdev, + struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack) +{ + if (config->tx_type == HWTSTAMP_TX_OFF && + config->rx_filter == HWTSTAMP_FILTER_ALL) return 0; - default: - return -EOPNOTSUPP; - } + NL_SET_ERR_MSG_MOD(extack, "Only RX HWTSTAMP_FILTER_ALL is supported"); + return -ERANGE; } static const struct net_device_ops peak_canfd_netdev_ops = { .ndo_open = peak_canfd_open, .ndo_stop = peak_canfd_close, - .ndo_eth_ioctl = peak_eth_ioctl, .ndo_start_xmit = peak_canfd_start_xmit, + .ndo_hwtstamp_get = peak_eth_hwtstamp_get, + .ndo_hwtstamp_set = peak_eth_hwtstamp_set, }; static int peak_get_ts_info(struct net_device *dev, From 243449f99238486f40fcc74d249d011b217ddcef Mon Sep 17 00:00:00 2001 From: Vadim Fedorenko Date: Wed, 29 Oct 2025 23:16:20 +0000 Subject: [PATCH 0421/1381] can: peak_usb: convert to use ndo_hwtstamp callbacks Convert driver to use ndo_hwtstamp_set()/ndo_hwtstamp_get() callbacks. ndo_eth_ioctl handler does nothing after conversion - remove it. Signed-off-by: Vadim Fedorenko Reviewed-by: Kory Maincent Reviewed-by: Vincent Mailhol Link: https://patch.msgid.link/20251029231620.1135640-4-vadim.fedorenko@linux.dev Signed-off-by: Marc Kleine-Budde --- drivers/net/can/usb/peak_usb/pcan_usb_core.c | 35 +++++++++----------- 1 file changed, 16 insertions(+), 19 deletions(-) diff --git a/drivers/net/can/usb/peak_usb/pcan_usb_core.c b/drivers/net/can/usb/peak_usb/pcan_usb_core.c index 94b1d7f15d27..cf48bb26d46d 100644 --- a/drivers/net/can/usb/peak_usb/pcan_usb_core.c +++ b/drivers/net/can/usb/peak_usb/pcan_usb_core.c @@ -784,36 +784,33 @@ static int peak_usb_set_data_bittiming(struct net_device *netdev) return 0; } -static int peak_eth_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) +static int peak_hwtstamp_get(struct net_device *netdev, + struct kernel_hwtstamp_config *config) { - struct hwtstamp_config hwts_cfg = { 0 }; + config->tx_type = HWTSTAMP_TX_OFF; + config->rx_filter = HWTSTAMP_FILTER_ALL; - switch (cmd) { - case SIOCSHWTSTAMP: /* set */ - if (copy_from_user(&hwts_cfg, ifr->ifr_data, sizeof(hwts_cfg))) - return -EFAULT; - if (hwts_cfg.tx_type == HWTSTAMP_TX_OFF && - hwts_cfg.rx_filter == HWTSTAMP_FILTER_ALL) - return 0; - return -ERANGE; + return 0; +} - case SIOCGHWTSTAMP: /* get */ - hwts_cfg.tx_type = HWTSTAMP_TX_OFF; - hwts_cfg.rx_filter = HWTSTAMP_FILTER_ALL; - if (copy_to_user(ifr->ifr_data, &hwts_cfg, sizeof(hwts_cfg))) - return -EFAULT; +static int peak_hwtstamp_set(struct net_device *netdev, + struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack) +{ + if (config->tx_type == HWTSTAMP_TX_OFF && + config->rx_filter == HWTSTAMP_FILTER_ALL) return 0; - default: - return -EOPNOTSUPP; - } + NL_SET_ERR_MSG_MOD(extack, "Only RX HWTSTAMP_FILTER_ALL is supported"); + return -ERANGE; } static const struct net_device_ops peak_usb_netdev_ops = { .ndo_open = peak_usb_ndo_open, .ndo_stop = peak_usb_ndo_stop, - .ndo_eth_ioctl = peak_eth_ioctl, .ndo_start_xmit = peak_usb_ndo_start_xmit, + .ndo_hwtstamp_get = peak_hwtstamp_get, + .ndo_hwtstamp_set = peak_hwtstamp_set, }; /* CAN-USB devices generally handle 32-bit CAN channel IDs. From 5cf236b89f4a8254f28af4badc0f0dbdaba09193 Mon Sep 17 00:00:00 2001 From: Maud Spierings Date: Thu, 30 Oct 2025 08:07:42 +0100 Subject: [PATCH 0422/1381] can: mcp251x: mcp251x_can_probe(): use dev_err_probe() The currently used combination of dev_err() plus return leaves a loud error in dmesg even when the error is a deferred probe which gets resolved later. For example a supply that has not been probed yet. Use dev_err_probe() to improve the handling/display of errors. Signed-off-by: Maud Spierings Link: https://patch.msgid.link/20251030-mcp_err-v1-1-eecf737823b7@gocontroll.com Signed-off-by: Marc Kleine-Budde --- drivers/net/can/spi/mcp251x.c | 31 ++++++++++++++++++++----------- 1 file changed, 20 insertions(+), 11 deletions(-) diff --git a/drivers/net/can/spi/mcp251x.c b/drivers/net/can/spi/mcp251x.c index 1e54e1a22702..fa97adf25b73 100644 --- a/drivers/net/can/spi/mcp251x.c +++ b/drivers/net/can/spi/mcp251x.c @@ -1320,7 +1320,7 @@ static int mcp251x_can_probe(struct spi_device *spi) clk = devm_clk_get_optional(&spi->dev, NULL); if (IS_ERR(clk)) - return PTR_ERR(clk); + return dev_err_probe(&spi->dev, PTR_ERR(clk), "Cannot get clock\n"); freq = clk_get_rate(clk); if (freq == 0) @@ -1328,7 +1328,7 @@ static int mcp251x_can_probe(struct spi_device *spi) /* Sanity check */ if (freq < 1000000 || freq > 25000000) - return -ERANGE; + return dev_err_probe(&spi->dev, -ERANGE, "clock frequency out of range\n"); /* Allocate can/net device */ net = alloc_candev(sizeof(struct mcp251x_priv), TX_ECHO_SKB_MAX); @@ -1336,8 +1336,10 @@ static int mcp251x_can_probe(struct spi_device *spi) return -ENOMEM; ret = clk_prepare_enable(clk); - if (ret) + if (ret) { + dev_err_probe(&spi->dev, ret, "Cannot enable clock\n"); goto out_free; + } net->netdev_ops = &mcp251x_netdev_ops; net->ethtool_ops = &mcp251x_ethtool_ops; @@ -1362,20 +1364,25 @@ static int mcp251x_can_probe(struct spi_device *spi) else spi->max_speed_hz = spi->max_speed_hz ? : 10 * 1000 * 1000; ret = spi_setup(spi); - if (ret) + if (ret) { + dev_err_probe(&spi->dev, ret, "Cannot set up spi\n"); goto out_clk; + } priv->power = devm_regulator_get_optional(&spi->dev, "vdd"); priv->transceiver = devm_regulator_get_optional(&spi->dev, "xceiver"); if ((PTR_ERR(priv->power) == -EPROBE_DEFER) || (PTR_ERR(priv->transceiver) == -EPROBE_DEFER)) { ret = -EPROBE_DEFER; + dev_err_probe(&spi->dev, ret, "supply deferred\n"); goto out_clk; } ret = mcp251x_power_enable(priv->power, 1); - if (ret) + if (ret) { + dev_err_probe(&spi->dev, ret, "Cannot enable power\n"); goto out_clk; + } priv->wq = alloc_workqueue("mcp251x_wq", WQ_FREEZABLE | WQ_MEM_RECLAIM | WQ_PERCPU, @@ -1409,21 +1416,24 @@ static int mcp251x_can_probe(struct spi_device *spi) /* Here is OK to not lock the MCP, no one knows about it yet */ ret = mcp251x_hw_probe(spi); if (ret) { - if (ret == -ENODEV) - dev_err(&spi->dev, "Cannot initialize MCP%x. Wrong wiring?\n", - priv->model); + dev_err_probe(&spi->dev, ret, "Cannot initialize MCP%x. Wrong wiring?\n", + priv->model); goto error_probe; } mcp251x_hw_sleep(spi); ret = register_candev(net); - if (ret) + if (ret) { + dev_err_probe(&spi->dev, ret, "Cannot register CAN device\n"); goto error_probe; + } ret = mcp251x_gpio_setup(priv); - if (ret) + if (ret) { + dev_err_probe(&spi->dev, ret, "Cannot set up gpios\n"); goto out_unregister_candev; + } netdev_info(net, "MCP%x successfully initialized.\n", priv->model); return 0; @@ -1442,7 +1452,6 @@ static int mcp251x_can_probe(struct spi_device *spi) out_free: free_candev(net); - dev_err(&spi->dev, "Probe failed, err=%d\n", -ret); return ret; } From 54133f9b4b53ffa2204eb27cfc9d50072c9a52d2 Mon Sep 17 00:00:00 2001 From: Haiyang Zhang Date: Wed, 29 Oct 2025 13:43:10 -0700 Subject: [PATCH 0423/1381] net: mana: Support HW link state events Handle the NIC hardware link state events received from the HW channel, then set the proper link state accordingly. And, add a feature bit, GDMA_DRV_CAP_FLAG_1_HW_VPORT_LINK_AWARE, to inform the NIC hardware this handler exists. Our MANA NIC only sends out the link state down/up messages when we need to let the VM rerun DHCP client and change IP address. So, add netif_carrier_on() in the probe(), let the NIC show the right initial state in /sys/class/net/ethX/operstate. Signed-off-by: Haiyang Zhang Link: https://patch.msgid.link/1761770601-16920-1-git-send-email-haiyangz@linux.microsoft.com Signed-off-by: Jakub Kicinski --- .../net/ethernet/microsoft/mana/gdma_main.c | 1 + .../net/ethernet/microsoft/mana/hw_channel.c | 12 +++++ drivers/net/ethernet/microsoft/mana/mana_en.c | 54 +++++++++++++++++-- include/net/mana/gdma.h | 4 +- include/net/mana/hw_channel.h | 2 + include/net/mana/mana.h | 4 ++ 6 files changed, 71 insertions(+), 6 deletions(-) diff --git a/drivers/net/ethernet/microsoft/mana/gdma_main.c b/drivers/net/ethernet/microsoft/mana/gdma_main.c index 43f034e180c4..effe0a2f207a 100644 --- a/drivers/net/ethernet/microsoft/mana/gdma_main.c +++ b/drivers/net/ethernet/microsoft/mana/gdma_main.c @@ -528,6 +528,7 @@ static void mana_gd_process_eqe(struct gdma_queue *eq) case GDMA_EQE_HWC_INIT_DONE: case GDMA_EQE_HWC_SOC_SERVICE: case GDMA_EQE_RNIC_QP_FATAL: + case GDMA_EQE_HWC_SOC_RECONFIG_DATA: if (!eq->eq.callback) break; diff --git a/drivers/net/ethernet/microsoft/mana/hw_channel.c b/drivers/net/ethernet/microsoft/mana/hw_channel.c index ada6c78a2bef..aa4e2731e2ba 100644 --- a/drivers/net/ethernet/microsoft/mana/hw_channel.c +++ b/drivers/net/ethernet/microsoft/mana/hw_channel.c @@ -118,6 +118,7 @@ static void mana_hwc_init_event_handler(void *ctx, struct gdma_queue *q_self, struct gdma_dev *gd = hwc->gdma_dev; union hwc_init_type_data type_data; union hwc_init_eq_id_db eq_db; + struct mana_context *ac; u32 type, val; int ret; @@ -196,6 +197,17 @@ static void mana_hwc_init_event_handler(void *ctx, struct gdma_queue *q_self, hwc->hwc_timeout = val; break; + case HWC_DATA_HW_LINK_CONNECT: + case HWC_DATA_HW_LINK_DISCONNECT: + ac = gd->gdma_context->mana.driver_data; + if (!ac) + break; + + WRITE_ONCE(ac->link_event, type); + schedule_work(&ac->link_change_work); + + break; + default: dev_warn(hwc->dev, "Received unknown reconfig type %u\n", type); break; diff --git a/drivers/net/ethernet/microsoft/mana/mana_en.c b/drivers/net/ethernet/microsoft/mana/mana_en.c index 0142fd98392c..739087081dfd 100644 --- a/drivers/net/ethernet/microsoft/mana/mana_en.c +++ b/drivers/net/ethernet/microsoft/mana/mana_en.c @@ -20,6 +20,7 @@ #include #include +#include static DEFINE_IDA(mana_adev_ida); @@ -84,7 +85,6 @@ static int mana_open(struct net_device *ndev) /* Ensure port state updated before txq state */ smp_wmb(); - netif_carrier_on(ndev); netif_tx_wake_all_queues(ndev); netdev_dbg(ndev, "%s successful\n", __func__); return 0; @@ -100,6 +100,46 @@ static int mana_close(struct net_device *ndev) return mana_detach(ndev, true); } +static void mana_link_state_handle(struct work_struct *w) +{ + struct mana_context *ac; + struct net_device *ndev; + u32 link_event; + bool link_up; + int i; + + ac = container_of(w, struct mana_context, link_change_work); + + rtnl_lock(); + + link_event = READ_ONCE(ac->link_event); + + if (link_event == HWC_DATA_HW_LINK_CONNECT) + link_up = true; + else if (link_event == HWC_DATA_HW_LINK_DISCONNECT) + link_up = false; + else + goto out; + + /* Process all ports */ + for (i = 0; i < ac->num_ports; i++) { + ndev = ac->ports[i]; + if (!ndev) + continue; + + if (link_up) { + netif_carrier_on(ndev); + + __netdev_notify_peers(ndev); + } else { + netif_carrier_off(ndev); + } + } + +out: + rtnl_unlock(); +} + static bool mana_can_tx(struct gdma_queue *wq) { return mana_gd_wq_avail_space(wq) >= MAX_TX_WQE_SIZE; @@ -3059,9 +3099,6 @@ int mana_attach(struct net_device *ndev) /* Ensure port state updated before txq state */ smp_wmb(); - if (apc->port_is_up) - netif_carrier_on(ndev); - netif_device_attach(ndev); return 0; @@ -3154,7 +3191,6 @@ int mana_detach(struct net_device *ndev, bool from_close) smp_wmb(); netif_tx_disable(ndev); - netif_carrier_off(ndev); if (apc->port_st_save) { err = mana_dealloc_queues(ndev); @@ -3243,6 +3279,8 @@ static int mana_probe_port(struct mana_context *ac, int port_idx, goto free_indir; } + netif_carrier_on(ndev); + debugfs_create_u32("current_speed", 0400, apc->mana_port_debugfs, &apc->speed); return 0; @@ -3431,6 +3469,8 @@ int mana_probe(struct gdma_dev *gd, bool resuming) if (!resuming) { ac->num_ports = num_ports; + + INIT_WORK(&ac->link_change_work, mana_link_state_handle); } else { if (ac->num_ports != num_ports) { dev_err(dev, "The number of vPorts changed: %d->%d\n", @@ -3438,6 +3478,8 @@ int mana_probe(struct gdma_dev *gd, bool resuming) err = -EPROTO; goto out; } + + enable_work(&ac->link_change_work); } if (ac->num_ports == 0) @@ -3500,6 +3542,8 @@ void mana_remove(struct gdma_dev *gd, bool suspending) int err; int i; + disable_work_sync(&ac->link_change_work); + /* adev currently doesn't support suspending, always remove it */ if (gd->adev) remove_adev(gd); diff --git a/include/net/mana/gdma.h b/include/net/mana/gdma.h index 57df78cfbf82..637f42485dba 100644 --- a/include/net/mana/gdma.h +++ b/include/net/mana/gdma.h @@ -590,6 +590,7 @@ enum { /* Driver can self reset on FPGA Reconfig EQE notification */ #define GDMA_DRV_CAP_FLAG_1_HANDLE_RECONFIG_EQE BIT(17) +#define GDMA_DRV_CAP_FLAG_1_HW_VPORT_LINK_AWARE BIT(6) #define GDMA_DRV_CAP_FLAGS1 \ (GDMA_DRV_CAP_FLAG_1_EQ_SHARING_MULTI_VPORT | \ @@ -599,7 +600,8 @@ enum { GDMA_DRV_CAP_FLAG_1_DEV_LIST_HOLES_SUP | \ GDMA_DRV_CAP_FLAG_1_DYNAMIC_IRQ_ALLOC_SUPPORT | \ GDMA_DRV_CAP_FLAG_1_SELF_RESET_ON_EQE | \ - GDMA_DRV_CAP_FLAG_1_HANDLE_RECONFIG_EQE) + GDMA_DRV_CAP_FLAG_1_HANDLE_RECONFIG_EQE | \ + GDMA_DRV_CAP_FLAG_1_HW_VPORT_LINK_AWARE) #define GDMA_DRV_CAP_FLAGS2 0 diff --git a/include/net/mana/hw_channel.h b/include/net/mana/hw_channel.h index 83cf93338eb3..16feb39616c1 100644 --- a/include/net/mana/hw_channel.h +++ b/include/net/mana/hw_channel.h @@ -24,6 +24,8 @@ #define HWC_INIT_DATA_PF_DEST_CQ_ID 11 #define HWC_DATA_CFG_HWC_TIMEOUT 1 +#define HWC_DATA_HW_LINK_CONNECT 2 +#define HWC_DATA_HW_LINK_DISCONNECT 3 #define HW_CHANNEL_WAIT_RESOURCE_TIMEOUT_MS 30000 diff --git a/include/net/mana/mana.h b/include/net/mana/mana.h index 0921485565c0..8906901535f5 100644 --- a/include/net/mana/mana.h +++ b/include/net/mana/mana.h @@ -477,6 +477,10 @@ struct mana_context { struct dentry *mana_eqs_debugfs; struct net_device *ports[MAX_PORTS_IN_MANA_DEV]; + + /* Link state change work */ + struct work_struct link_change_work; + u32 link_event; }; struct mana_port_context { From 7ea4376b3972d89385599307d1ad4f20eb763a05 Mon Sep 17 00:00:00 2001 From: Carolina Jubran Date: Thu, 30 Oct 2025 12:25:05 +0200 Subject: [PATCH 0424/1381] net/mlx5e: Remove redundant tstamp pointer from channel structures Remove the tstamp pointer field from mlx5e_channel, mlx5e_ptp, and mlx5e_trap structures, since it was only used to reference the tstamp field in the priv structure. Instead, directly use the tstamp field from priv when initializing RQ structures. Also remove the unused hwtstamp_config field from mlx5_clock structure as part of the cleanup. Signed-off-by: Carolina Jubran Reviewed-by: Cosmin Ratiu Signed-off-by: Tariq Toukan Link: https://patch.msgid.link/1761819910-1011051-2-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mellanox/mlx5/core/en.h | 1 - drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c | 1 - drivers/net/ethernet/mellanox/mlx5/core/en/ptp.h | 1 - drivers/net/ethernet/mellanox/mlx5/core/en/trap.c | 1 - drivers/net/ethernet/mellanox/mlx5/core/en/trap.h | 1 - drivers/net/ethernet/mellanox/mlx5/core/en/xsk/setup.c | 2 +- drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 3 +-- drivers/net/ethernet/mellanox/mlx5/core/lib/clock.h | 1 - 8 files changed, 2 insertions(+), 9 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index 14e3207b14e7..5485cf014926 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -784,7 +784,6 @@ struct mlx5e_channel { /* control */ struct mlx5e_priv *priv; struct mlx5_core_dev *mdev; - struct hwtstamp_config *tstamp; DECLARE_BITMAP(state, MLX5E_CHANNEL_NUM_STATES); int ix; int vec_ix; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c b/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c index c93ee969ea64..96a78b6d4904 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c @@ -896,7 +896,6 @@ int mlx5e_ptp_open(struct mlx5e_priv *priv, struct mlx5e_params *params, c->priv = priv; c->mdev = priv->mdev; - c->tstamp = &priv->tstamp; c->pdev = mlx5_core_dma_dev(priv->mdev); c->netdev = priv->netdev; c->mkey_be = cpu_to_be32(priv->mdev->mlx5e_res.hw_objs.mkey); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.h b/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.h index 1b3c9648220b..1c0e0a86a9ac 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.h @@ -64,7 +64,6 @@ struct mlx5e_ptp { /* control */ struct mlx5e_priv *priv; struct mlx5_core_dev *mdev; - struct hwtstamp_config *tstamp; DECLARE_BITMAP(state, MLX5E_PTP_STATE_NUM_STATES); struct mlx5_sq_bfreg *bfreg; }; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/trap.c b/drivers/net/ethernet/mellanox/mlx5/core/en/trap.c index 996fcdb5a29d..db6932b0cedf 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/trap.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/trap.c @@ -144,7 +144,6 @@ static struct mlx5e_trap *mlx5e_open_trap(struct mlx5e_priv *priv) t->priv = priv; t->mdev = priv->mdev; - t->tstamp = &priv->tstamp; t->pdev = mlx5_core_dma_dev(priv->mdev); t->netdev = priv->netdev; t->mkey_be = cpu_to_be32(priv->mdev->mlx5e_res.hw_objs.mkey); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/trap.h b/drivers/net/ethernet/mellanox/mlx5/core/en/trap.h index aa3f17658c6d..394e917ea2b0 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/trap.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/trap.h @@ -22,7 +22,6 @@ struct mlx5e_trap { /* control */ struct mlx5e_priv *priv; struct mlx5_core_dev *mdev; - struct hwtstamp_config *tstamp; DECLARE_BITMAP(state, MLX5E_CHANNEL_NUM_STATES); struct mlx5e_params params; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/setup.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/setup.c index dbd88eb5c082..dc5a4afa4974 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/setup.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/setup.c @@ -71,7 +71,7 @@ static int mlx5e_init_xsk_rq(struct mlx5e_channel *c, rq->pdev = c->pdev; rq->netdev = c->netdev; rq->priv = c->priv; - rq->tstamp = c->tstamp; + rq->tstamp = &c->priv->tstamp; rq->clock = mdev->clock; rq->icosq = &c->icosq; rq->ix = c->ix; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index 9c46511e7b43..20f55542433d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -735,7 +735,7 @@ static int mlx5e_init_rxq_rq(struct mlx5e_channel *c, struct mlx5e_params *param rq->pdev = c->pdev; rq->netdev = c->netdev; rq->priv = c->priv; - rq->tstamp = c->tstamp; + rq->tstamp = &c->priv->tstamp; rq->clock = mdev->clock; rq->icosq = &c->icosq; rq->ix = c->ix; @@ -2803,7 +2803,6 @@ static int mlx5e_open_channel(struct mlx5e_priv *priv, int ix, c->priv = priv; c->mdev = mdev; - c->tstamp = &priv->tstamp; c->ix = ix; c->vec_ix = vec_ix; c->sd_ix = mlx5_sd_ch_ix_get_dev_ix(mdev, ix); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.h b/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.h index c18a652c0faa..aff3aed62c74 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.h @@ -54,7 +54,6 @@ struct mlx5_timer { struct mlx5_clock { seqlock_t lock; - struct hwtstamp_config hwtstamp_config; struct ptp_clock *ptp; struct ptp_clock_info ptp_info; struct mlx5_pps pps_info; From bf791659743b1a8e20f5810b1ac893b7b24f650e Mon Sep 17 00:00:00 2001 From: Carolina Jubran Date: Thu, 30 Oct 2025 12:25:06 +0200 Subject: [PATCH 0425/1381] net/mlx5e: Remove unnecessary tstamp local variable in mlx5i_complete_rx_cqe Remove the tstamp local variable in mlx5i_complete_rx_cqe() and directly pass the tstamp field from priv to mlx5e_rx_hw_stamp(). The local variable was only used once and provided no additional value. Signed-off-by: Carolina Jubran Reviewed-by: Cosmin Ratiu Signed-off-by: Tariq Toukan Link: https://patch.msgid.link/1761819910-1011051-3-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mellanox/mlx5/core/en_rx.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c index 26621a2972ec..2fc1ec76770d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c @@ -2650,7 +2650,6 @@ static inline void mlx5i_complete_rx_cqe(struct mlx5e_rq *rq, u32 cqe_bcnt, struct sk_buff *skb) { - struct hwtstamp_config *tstamp; struct mlx5e_rq_stats *stats; struct net_device *netdev; struct mlx5e_priv *priv; @@ -2674,7 +2673,6 @@ static inline void mlx5i_complete_rx_cqe(struct mlx5e_rq *rq, } priv = mlx5i_epriv(netdev); - tstamp = &priv->tstamp; stats = &priv->channel_stats[rq->ix]->rq; flags_rqpn = be32_to_cpu(cqe->flags_rqpn); @@ -2710,7 +2708,7 @@ static inline void mlx5i_complete_rx_cqe(struct mlx5e_rq *rq, stats->csum_none++; } - if (unlikely(mlx5e_rx_hw_stamp(tstamp))) + if (unlikely(mlx5e_rx_hw_stamp(&priv->tstamp))) skb_hwtstamps(skb)->hwtstamp = mlx5e_cqe_ts_to_ns(rq->ptp_cyc2time, rq->clock, get_cqe_ts(cqe)); skb_record_rx_queue(skb, rq->ix); From fee182371a59414d43633a5ea6f1cda160418a16 Mon Sep 17 00:00:00 2001 From: Carolina Jubran Date: Thu, 30 Oct 2025 12:25:07 +0200 Subject: [PATCH 0426/1381] net/mlx5e: Rename hwstamp functions to hwtstamp Rename mlx5e_hwstamp_set/get() functions to mlx5e_hwtstamp_set/get() to better reflect that these functions handle hardware timestamping, not just hardware stamping. Signed-off-by: Carolina Jubran Reviewed-by: Cosmin Ratiu Signed-off-by: Tariq Toukan Link: https://patch.msgid.link/1761819910-1011051-4-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mellanox/mlx5/core/en.h | 4 ++-- drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 8 ++++---- drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c | 4 ++-- 3 files changed, 8 insertions(+), 8 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index 5485cf014926..ebd7493888d7 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -1026,8 +1026,8 @@ void mlx5e_self_test(struct net_device *ndev, struct ethtool_test *etest, u64 *buf); void mlx5e_set_rx_mode_work(struct work_struct *work); -int mlx5e_hwstamp_set(struct mlx5e_priv *priv, struct ifreq *ifr); -int mlx5e_hwstamp_get(struct mlx5e_priv *priv, struct ifreq *ifr); +int mlx5e_hwtstamp_set(struct mlx5e_priv *priv, struct ifreq *ifr); +int mlx5e_hwtstamp_get(struct mlx5e_priv *priv, struct ifreq *ifr); int mlx5e_modify_rx_cqe_compression_locked(struct mlx5e_priv *priv, bool val, bool rx_filter); int mlx5e_vlan_rx_add_vid(struct net_device *dev, __always_unused __be16 proto, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index 20f55542433d..2ecbd735584e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -4740,7 +4740,7 @@ static int mlx5e_hwstamp_config_ptp_rx(struct mlx5e_priv *priv, bool ptp_rx) &new_params.ptp_rx, true); } -int mlx5e_hwstamp_set(struct mlx5e_priv *priv, struct ifreq *ifr) +int mlx5e_hwtstamp_set(struct mlx5e_priv *priv, struct ifreq *ifr) { struct hwtstamp_config config; bool rx_cqe_compress_def; @@ -4818,7 +4818,7 @@ int mlx5e_hwstamp_set(struct mlx5e_priv *priv, struct ifreq *ifr) return err; } -int mlx5e_hwstamp_get(struct mlx5e_priv *priv, struct ifreq *ifr) +int mlx5e_hwtstamp_get(struct mlx5e_priv *priv, struct ifreq *ifr) { struct hwtstamp_config *cfg = &priv->tstamp; @@ -4834,9 +4834,9 @@ static int mlx5e_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) switch (cmd) { case SIOCSHWTSTAMP: - return mlx5e_hwstamp_set(priv, ifr); + return mlx5e_hwtstamp_set(priv, ifr); case SIOCGHWTSTAMP: - return mlx5e_hwstamp_get(priv, ifr); + return mlx5e_hwtstamp_get(priv, ifr); default: return -EOPNOTSUPP; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c index 79ae3a51a4b3..11d950f58ae3 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c @@ -563,9 +563,9 @@ int mlx5i_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) switch (cmd) { case SIOCSHWTSTAMP: - return mlx5e_hwstamp_set(priv, ifr); + return mlx5e_hwtstamp_set(priv, ifr); case SIOCGHWTSTAMP: - return mlx5e_hwstamp_get(priv, ifr); + return mlx5e_hwtstamp_get(priv, ifr); default: return -EOPNOTSUPP; } From 91baaf96f5d0b764aec462dd50a8433f5c8d621f Mon Sep 17 00:00:00 2001 From: Carolina Jubran Date: Thu, 30 Oct 2025 12:25:08 +0200 Subject: [PATCH 0427/1381] net/mlx5e: Rename timestamp fields to hwtstamp_config Rename hardware timestamp-related fields from 'tstamp' to 'hwtstamp_config' throughout the MLX5 driver. The new name is more descriptive as it clearly indicates these fields contain hardware timestamp configuration. Signed-off-by: Carolina Jubran Reviewed-by: Cosmin Ratiu Signed-off-by: Tariq Toukan Link: https://patch.msgid.link/1761819910-1011051-5-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mellanox/mlx5/core/en.h | 4 ++-- drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c | 2 +- .../net/ethernet/mellanox/mlx5/core/en/reporter_rx.c | 3 ++- drivers/net/ethernet/mellanox/mlx5/core/en/trap.c | 2 +- drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c | 2 +- drivers/net/ethernet/mellanox/mlx5/core/en/xsk/setup.c | 2 +- drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c | 2 +- drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 10 +++++----- drivers/net/ethernet/mellanox/mlx5/core/en_rx.c | 4 ++-- 9 files changed, 16 insertions(+), 15 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index ebd7493888d7..eb3eef1a496e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -696,7 +696,7 @@ struct mlx5e_rq { struct mlx5e_rq_stats *stats; struct mlx5e_cq cq; struct mlx5e_cq_decomp cqd; - struct hwtstamp_config *tstamp; + struct hwtstamp_config *hwtstamp_config; struct mlx5_clock *clock; struct mlx5e_icosq *icosq; struct mlx5e_priv *priv; @@ -917,7 +917,7 @@ struct mlx5e_priv { u8 max_opened_tc; bool tx_ptp_opened; bool rx_ptp_opened; - struct hwtstamp_config tstamp; + struct hwtstamp_config hwtstamp_config; u16 q_counter[MLX5_SD_MAX_GROUP_SZ]; u16 drop_rq_q_counter; struct notifier_block events_nb; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c b/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c index 96a78b6d4904..12e10feb30f0 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c @@ -713,7 +713,7 @@ static int mlx5e_init_ptp_rq(struct mlx5e_ptp *c, struct mlx5e_params *params, rq->netdev = priv->netdev; rq->priv = priv; rq->clock = mdev->clock; - rq->tstamp = &priv->tstamp; + rq->hwtstamp_config = &priv->hwtstamp_config; rq->mdev = mdev; rq->hw_mtu = MLX5E_SW2HW_MTU(params, params->sw_mtu); rq->stats = &c->priv->ptp_stats.rq; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c index b1415992ffa2..0686fbdd5a05 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c @@ -318,7 +318,8 @@ mlx5e_rx_reporter_diagnose_common_ptp_config(struct mlx5e_priv *priv, struct mlx struct devlink_fmsg *fmsg) { mlx5e_health_fmsg_named_obj_nest_start(fmsg, "PTP"); - devlink_fmsg_u32_pair_put(fmsg, "filter_type", priv->tstamp.rx_filter); + devlink_fmsg_u32_pair_put(fmsg, "filter_type", + priv->hwtstamp_config.rx_filter); mlx5e_rx_reporter_diagnose_generic_rq(&ptp_ch->rq, fmsg); mlx5e_health_fmsg_named_obj_nest_end(fmsg); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/trap.c b/drivers/net/ethernet/mellanox/mlx5/core/en/trap.c index db6932b0cedf..da8c44f46edb 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/trap.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/trap.c @@ -47,7 +47,7 @@ static void mlx5e_init_trap_rq(struct mlx5e_trap *t, struct mlx5e_params *params rq->netdev = priv->netdev; rq->priv = priv; rq->clock = mdev->clock; - rq->tstamp = &priv->tstamp; + rq->hwtstamp_config = &priv->hwtstamp_config; rq->mdev = mdev; rq->hw_mtu = MLX5E_SW2HW_MTU(params, params->sw_mtu); rq->stats = &priv->trap_stats.rq; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c index 5d51600935a6..80f9fc10877a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c @@ -179,7 +179,7 @@ static int mlx5e_xdp_rx_timestamp(const struct xdp_md *ctx, u64 *timestamp) { const struct mlx5e_xdp_buff *_ctx = (void *)ctx; - if (unlikely(!mlx5e_rx_hw_stamp(_ctx->rq->tstamp))) + if (unlikely(!mlx5e_rx_hw_stamp(_ctx->rq->hwtstamp_config))) return -ENODATA; *timestamp = mlx5e_cqe_ts_to_ns(_ctx->rq->ptp_cyc2time, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/setup.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/setup.c index dc5a4afa4974..5981c71cae2d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/setup.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/setup.c @@ -71,7 +71,7 @@ static int mlx5e_init_xsk_rq(struct mlx5e_channel *c, rq->pdev = c->pdev; rq->netdev = c->netdev; rq->priv = c->priv; - rq->tstamp = &c->priv->tstamp; + rq->hwtstamp_config = &c->priv->hwtstamp_config; rq->clock = mdev->clock; rq->icosq = &c->icosq; rq->ix = c->ix; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c index 53e5ae252eac..47a3770fb0f7 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -2273,7 +2273,7 @@ static int set_pflag_rx_cqe_compress(struct net_device *netdev, if (!MLX5_CAP_GEN(mdev, cqe_compression)) return -EOPNOTSUPP; - rx_filter = priv->tstamp.rx_filter != HWTSTAMP_FILTER_NONE; + rx_filter = priv->hwtstamp_config.rx_filter != HWTSTAMP_FILTER_NONE; err = mlx5e_modify_rx_cqe_compression_locked(priv, enable, rx_filter); if (err) return err; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index 2ecbd735584e..5b2491e19baa 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -735,7 +735,7 @@ static int mlx5e_init_rxq_rq(struct mlx5e_channel *c, struct mlx5e_params *param rq->pdev = c->pdev; rq->netdev = c->netdev; rq->priv = c->priv; - rq->tstamp = &c->priv->tstamp; + rq->hwtstamp_config = &c->priv->hwtstamp_config; rq->clock = mdev->clock; rq->icosq = &c->icosq; rq->ix = c->ix; @@ -3444,8 +3444,8 @@ int mlx5e_safe_reopen_channels(struct mlx5e_priv *priv) void mlx5e_timestamp_init(struct mlx5e_priv *priv) { - priv->tstamp.tx_type = HWTSTAMP_TX_OFF; - priv->tstamp.rx_filter = HWTSTAMP_FILTER_NONE; + priv->hwtstamp_config.tx_type = HWTSTAMP_TX_OFF; + priv->hwtstamp_config.rx_filter = HWTSTAMP_FILTER_NONE; } static void mlx5e_modify_admin_state(struct mlx5_core_dev *mdev, @@ -4805,7 +4805,7 @@ int mlx5e_hwtstamp_set(struct mlx5e_priv *priv, struct ifreq *ifr) if (err) goto err_unlock; - memcpy(&priv->tstamp, &config, sizeof(config)); + memcpy(&priv->hwtstamp_config, &config, sizeof(config)); mutex_unlock(&priv->state_lock); /* might need to fix some features */ @@ -4820,7 +4820,7 @@ int mlx5e_hwtstamp_set(struct mlx5e_priv *priv, struct ifreq *ifr) int mlx5e_hwtstamp_get(struct mlx5e_priv *priv, struct ifreq *ifr) { - struct hwtstamp_config *cfg = &priv->tstamp; + struct hwtstamp_config *cfg = &priv->hwtstamp_config; if (!MLX5_CAP_GEN(priv->mdev, device_frequency_khz)) return -EOPNOTSUPP; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c index 2fc1ec76770d..6c8ce1d0d233 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c @@ -1602,7 +1602,7 @@ static inline bool mlx5e_build_rx_skb(struct mlx5_cqe64 *cqe, stats->lro_bytes += cqe_bcnt; } - if (unlikely(mlx5e_rx_hw_stamp(rq->tstamp))) + if (unlikely(mlx5e_rx_hw_stamp(rq->hwtstamp_config))) skb_hwtstamps(skb)->hwtstamp = mlx5e_cqe_ts_to_ns(rq->ptp_cyc2time, rq->clock, get_cqe_ts(cqe)); skb_record_rx_queue(skb, rq->ix); @@ -2708,7 +2708,7 @@ static inline void mlx5i_complete_rx_cqe(struct mlx5e_rq *rq, stats->csum_none++; } - if (unlikely(mlx5e_rx_hw_stamp(&priv->tstamp))) + if (unlikely(mlx5e_rx_hw_stamp(&priv->hwtstamp_config))) skb_hwtstamps(skb)->hwtstamp = mlx5e_cqe_ts_to_ns(rq->ptp_cyc2time, rq->clock, get_cqe_ts(cqe)); skb_record_rx_queue(skb, rq->ix); From 250da3c8fe811b14fd5e610760ed7469166fd0f7 Mon Sep 17 00:00:00 2001 From: Carolina Jubran Date: Thu, 30 Oct 2025 12:25:09 +0200 Subject: [PATCH 0428/1381] IB/IPoIB: Add support for hwtstamp get/set ndos Add support for the ndo_hwtstamp_get and ndo_hwtstamp_set operations in IPoIB. This allows lower devices to handle hardware timestamp configuration through the new ndos instead of the legacy ioctls. Signed-off-by: Carolina Jubran Signed-off-by: Tariq Toukan Link: https://patch.msgid.link/1761819910-1011051-6-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- drivers/infiniband/ulp/ipoib/ipoib_main.c | 29 +++++++++++++++++++++++ 1 file changed, 29 insertions(+) diff --git a/drivers/infiniband/ulp/ipoib/ipoib_main.c b/drivers/infiniband/ulp/ipoib/ipoib_main.c index 5b4d76e97437..300afc27c561 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_main.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_main.c @@ -1825,6 +1825,31 @@ static int ipoib_ioctl(struct net_device *dev, struct ifreq *ifr, return priv->rn_ops->ndo_eth_ioctl(dev, ifr, cmd); } +static int ipoib_hwtstamp_get(struct net_device *dev, + struct kernel_hwtstamp_config *config) +{ + struct ipoib_dev_priv *priv = ipoib_priv(dev); + + if (!priv->rn_ops->ndo_hwtstamp_get) + /* legacy */ + return dev_eth_ioctl(dev, config->ifr, SIOCGHWTSTAMP); + + return priv->rn_ops->ndo_hwtstamp_get(dev, config); +} + +static int ipoib_hwtstamp_set(struct net_device *dev, + struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack) +{ + struct ipoib_dev_priv *priv = ipoib_priv(dev); + + if (!priv->rn_ops->ndo_hwtstamp_set) + /* legacy */ + return dev_eth_ioctl(dev, config->ifr, SIOCSHWTSTAMP); + + return priv->rn_ops->ndo_hwtstamp_set(dev, config, extack); +} + static int ipoib_dev_init(struct net_device *dev) { struct ipoib_dev_priv *priv = ipoib_priv(dev); @@ -2149,6 +2174,8 @@ static const struct net_device_ops ipoib_netdev_ops_pf = { .ndo_set_mac_address = ipoib_set_mac, .ndo_get_stats64 = ipoib_get_stats, .ndo_eth_ioctl = ipoib_ioctl, + .ndo_hwtstamp_get = ipoib_hwtstamp_get, + .ndo_hwtstamp_set = ipoib_hwtstamp_set, }; static const struct net_device_ops ipoib_netdev_ops_vf = { @@ -2164,6 +2191,8 @@ static const struct net_device_ops ipoib_netdev_ops_vf = { .ndo_get_iflink = ipoib_get_iflink, .ndo_get_stats64 = ipoib_get_stats, .ndo_eth_ioctl = ipoib_ioctl, + .ndo_hwtstamp_get = ipoib_hwtstamp_get, + .ndo_hwtstamp_set = ipoib_hwtstamp_set, }; static const struct net_device_ops ipoib_netdev_default_pf = { From 1c7fe48a90158486a66665f1401d0b90bd390ef0 Mon Sep 17 00:00:00 2001 From: Carolina Jubran Date: Thu, 30 Oct 2025 12:25:10 +0200 Subject: [PATCH 0429/1381] net/mlx5e: Convert to new hwtstamp_get/set interface Migrate from the legacy ioctl hardware timestamping interface to the ndo_hwtstamp_get/set operations. Signed-off-by: Carolina Jubran Reviewed-by: Cosmin Ratiu Signed-off-by: Tariq Toukan Link: https://patch.msgid.link/1761819910-1011051-7-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mellanox/mlx5/core/en.h | 11 ++-- .../net/ethernet/mellanox/mlx5/core/en/txrx.h | 2 +- .../net/ethernet/mellanox/mlx5/core/en_main.c | 57 ++++++++++--------- .../ethernet/mellanox/mlx5/core/ipoib/ipoib.c | 34 ++++++----- .../ethernet/mellanox/mlx5/core/ipoib/ipoib.h | 6 +- .../mellanox/mlx5/core/ipoib/ipoib_vlan.c | 9 +-- 6 files changed, 65 insertions(+), 54 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index eb3eef1a496e..fd107906bc28 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -696,7 +696,7 @@ struct mlx5e_rq { struct mlx5e_rq_stats *stats; struct mlx5e_cq cq; struct mlx5e_cq_decomp cqd; - struct hwtstamp_config *hwtstamp_config; + struct kernel_hwtstamp_config *hwtstamp_config; struct mlx5_clock *clock; struct mlx5e_icosq *icosq; struct mlx5e_priv *priv; @@ -917,7 +917,7 @@ struct mlx5e_priv { u8 max_opened_tc; bool tx_ptp_opened; bool rx_ptp_opened; - struct hwtstamp_config hwtstamp_config; + struct kernel_hwtstamp_config hwtstamp_config; u16 q_counter[MLX5_SD_MAX_GROUP_SZ]; u16 drop_rq_q_counter; struct notifier_block events_nb; @@ -1026,8 +1026,11 @@ void mlx5e_self_test(struct net_device *ndev, struct ethtool_test *etest, u64 *buf); void mlx5e_set_rx_mode_work(struct work_struct *work); -int mlx5e_hwtstamp_set(struct mlx5e_priv *priv, struct ifreq *ifr); -int mlx5e_hwtstamp_get(struct mlx5e_priv *priv, struct ifreq *ifr); +int mlx5e_hwtstamp_set(struct mlx5e_priv *priv, + struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack); +int mlx5e_hwtstamp_get(struct mlx5e_priv *priv, + struct kernel_hwtstamp_config *config); int mlx5e_modify_rx_cqe_compression_locked(struct mlx5e_priv *priv, bool val, bool rx_filter); int mlx5e_vlan_rx_add_vid(struct net_device *dev, __always_unused __be16 proto, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h index 6760bb0336df..7e191e1569e8 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h @@ -92,7 +92,7 @@ int mlx5e_poll_rx_cq(struct mlx5e_cq *cq, int budget); void mlx5e_free_rx_descs(struct mlx5e_rq *rq); void mlx5e_free_rx_missing_descs(struct mlx5e_rq *rq); -static inline bool mlx5e_rx_hw_stamp(struct hwtstamp_config *config) +static inline bool mlx5e_rx_hw_stamp(struct kernel_hwtstamp_config *config) { return config->rx_filter == HWTSTAMP_FILTER_ALL; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index 5b2491e19baa..6b905848fe86 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -4740,22 +4740,23 @@ static int mlx5e_hwstamp_config_ptp_rx(struct mlx5e_priv *priv, bool ptp_rx) &new_params.ptp_rx, true); } -int mlx5e_hwtstamp_set(struct mlx5e_priv *priv, struct ifreq *ifr) +int mlx5e_hwtstamp_set(struct mlx5e_priv *priv, + struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack) { - struct hwtstamp_config config; bool rx_cqe_compress_def; bool ptp_rx; int err; if (!MLX5_CAP_GEN(priv->mdev, device_frequency_khz) || - (mlx5_clock_get_ptp_index(priv->mdev) == -1)) + (mlx5_clock_get_ptp_index(priv->mdev) == -1)) { + NL_SET_ERR_MSG_MOD(extack, + "Timestamps are not supported on this device"); return -EOPNOTSUPP; - - if (copy_from_user(&config, ifr->ifr_data, sizeof(config))) - return -EFAULT; + } /* TX HW timestamp */ - switch (config.tx_type) { + switch (config->tx_type) { case HWTSTAMP_TX_OFF: case HWTSTAMP_TX_ON: break; @@ -4767,7 +4768,7 @@ int mlx5e_hwtstamp_set(struct mlx5e_priv *priv, struct ifreq *ifr) rx_cqe_compress_def = priv->channels.params.rx_cqe_compress_def; /* RX HW timestamp */ - switch (config.rx_filter) { + switch (config->rx_filter) { case HWTSTAMP_FILTER_NONE: ptp_rx = false; break; @@ -4786,7 +4787,7 @@ int mlx5e_hwtstamp_set(struct mlx5e_priv *priv, struct ifreq *ifr) case HWTSTAMP_FILTER_PTP_V2_SYNC: case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ: case HWTSTAMP_FILTER_NTP_ALL: - config.rx_filter = HWTSTAMP_FILTER_ALL; + config->rx_filter = HWTSTAMP_FILTER_ALL; /* ptp_rx is set if both HW TS is set and CQE * compression is set */ @@ -4799,47 +4800,50 @@ int mlx5e_hwtstamp_set(struct mlx5e_priv *priv, struct ifreq *ifr) if (!mlx5e_profile_feature_cap(priv->profile, PTP_RX)) err = mlx5e_hwstamp_config_no_ptp_rx(priv, - config.rx_filter != HWTSTAMP_FILTER_NONE); + config->rx_filter != HWTSTAMP_FILTER_NONE); else err = mlx5e_hwstamp_config_ptp_rx(priv, ptp_rx); if (err) goto err_unlock; - memcpy(&priv->hwtstamp_config, &config, sizeof(config)); + priv->hwtstamp_config = *config; mutex_unlock(&priv->state_lock); /* might need to fix some features */ netdev_update_features(priv->netdev); - return copy_to_user(ifr->ifr_data, &config, - sizeof(config)) ? -EFAULT : 0; + return 0; err_unlock: mutex_unlock(&priv->state_lock); return err; } -int mlx5e_hwtstamp_get(struct mlx5e_priv *priv, struct ifreq *ifr) +static int mlx5e_hwtstamp_set_ndo(struct net_device *netdev, + struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack) { - struct hwtstamp_config *cfg = &priv->hwtstamp_config; + struct mlx5e_priv *priv = netdev_priv(netdev); + return mlx5e_hwtstamp_set(priv, config, extack); +} + +int mlx5e_hwtstamp_get(struct mlx5e_priv *priv, + struct kernel_hwtstamp_config *config) +{ if (!MLX5_CAP_GEN(priv->mdev, device_frequency_khz)) return -EOPNOTSUPP; - return copy_to_user(ifr->ifr_data, cfg, sizeof(*cfg)) ? -EFAULT : 0; + *config = priv->hwtstamp_config; + + return 0; } -static int mlx5e_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) +static int mlx5e_hwtstamp_get_ndo(struct net_device *dev, + struct kernel_hwtstamp_config *config) { struct mlx5e_priv *priv = netdev_priv(dev); - switch (cmd) { - case SIOCSHWTSTAMP: - return mlx5e_hwtstamp_set(priv, ifr); - case SIOCGHWTSTAMP: - return mlx5e_hwtstamp_get(priv, ifr); - default: - return -EOPNOTSUPP; - } + return mlx5e_hwtstamp_get(priv, config); } #ifdef CONFIG_MLX5_ESWITCH @@ -5280,13 +5284,14 @@ const struct net_device_ops mlx5e_netdev_ops = { .ndo_set_features = mlx5e_set_features, .ndo_fix_features = mlx5e_fix_features, .ndo_change_mtu = mlx5e_change_nic_mtu, - .ndo_eth_ioctl = mlx5e_ioctl, .ndo_set_tx_maxrate = mlx5e_set_tx_maxrate, .ndo_features_check = mlx5e_features_check, .ndo_tx_timeout = mlx5e_tx_timeout, .ndo_bpf = mlx5e_xdp, .ndo_xdp_xmit = mlx5e_xdp_xmit, .ndo_xsk_wakeup = mlx5e_xsk_wakeup, + .ndo_hwtstamp_get = mlx5e_hwtstamp_get_ndo, + .ndo_hwtstamp_set = mlx5e_hwtstamp_set_ndo, #ifdef CONFIG_MLX5_EN_ARFS .ndo_rx_flow_steer = mlx5e_rx_flow_steer, #endif diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c index 11d950f58ae3..906b1fbc27aa 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c @@ -45,6 +45,23 @@ static int mlx5i_open(struct net_device *netdev); static int mlx5i_close(struct net_device *netdev); static int mlx5i_change_mtu(struct net_device *netdev, int new_mtu); +int mlx5i_hwtstamp_set(struct net_device *dev, + struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack) +{ + struct mlx5e_priv *epriv = mlx5i_epriv(dev); + + return mlx5e_hwtstamp_set(epriv, config, extack); +} + +int mlx5i_hwtstamp_get(struct net_device *dev, + struct kernel_hwtstamp_config *config) +{ + struct mlx5e_priv *epriv = mlx5i_epriv(dev); + + return mlx5e_hwtstamp_get(epriv, config); +} + static const struct net_device_ops mlx5i_netdev_ops = { .ndo_open = mlx5i_open, .ndo_stop = mlx5i_close, @@ -52,7 +69,8 @@ static const struct net_device_ops mlx5i_netdev_ops = { .ndo_init = mlx5i_dev_init, .ndo_uninit = mlx5i_dev_cleanup, .ndo_change_mtu = mlx5i_change_mtu, - .ndo_eth_ioctl = mlx5i_ioctl, + .ndo_hwtstamp_get = mlx5i_hwtstamp_get, + .ndo_hwtstamp_set = mlx5i_hwtstamp_set, }; /* IPoIB mlx5 netdev profile */ @@ -557,20 +575,6 @@ int mlx5i_dev_init(struct net_device *dev) return 0; } -int mlx5i_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) -{ - struct mlx5e_priv *priv = mlx5i_epriv(dev); - - switch (cmd) { - case SIOCSHWTSTAMP: - return mlx5e_hwtstamp_set(priv, ifr); - case SIOCGHWTSTAMP: - return mlx5e_hwtstamp_get(priv, ifr); - default: - return -EOPNOTSUPP; - } -} - void mlx5i_dev_cleanup(struct net_device *dev) { struct mlx5e_priv *priv = mlx5i_epriv(dev); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.h b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.h index 2ab6437a1c49..d67d5a72bb41 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.h @@ -88,7 +88,11 @@ struct net_device *mlx5i_pkey_get_netdev(struct net_device *netdev, u32 qpn); /* Shared ndo functions */ int mlx5i_dev_init(struct net_device *dev); void mlx5i_dev_cleanup(struct net_device *dev); -int mlx5i_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd); +int mlx5i_hwtstamp_set(struct net_device *dev, + struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack); +int mlx5i_hwtstamp_get(struct net_device *dev, + struct kernel_hwtstamp_config *config); /* Parent profile functions */ int mlx5i_init(struct mlx5_core_dev *mdev, struct net_device *netdev); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib_vlan.c b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib_vlan.c index 028a76944d82..04444dad3a0d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib_vlan.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib_vlan.c @@ -140,7 +140,6 @@ static int mlx5i_pkey_close(struct net_device *netdev); static int mlx5i_pkey_dev_init(struct net_device *dev); static void mlx5i_pkey_dev_cleanup(struct net_device *netdev); static int mlx5i_pkey_change_mtu(struct net_device *netdev, int new_mtu); -static int mlx5i_pkey_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd); static const struct net_device_ops mlx5i_pkey_netdev_ops = { .ndo_open = mlx5i_pkey_open, @@ -149,7 +148,8 @@ static const struct net_device_ops mlx5i_pkey_netdev_ops = { .ndo_get_stats64 = mlx5i_get_stats, .ndo_uninit = mlx5i_pkey_dev_cleanup, .ndo_change_mtu = mlx5i_pkey_change_mtu, - .ndo_eth_ioctl = mlx5i_pkey_ioctl, + .ndo_hwtstamp_get = mlx5i_hwtstamp_get, + .ndo_hwtstamp_set = mlx5i_hwtstamp_set, }; /* Child NDOs */ @@ -184,11 +184,6 @@ static int mlx5i_pkey_dev_init(struct net_device *dev) return mlx5i_dev_init(dev); } -static int mlx5i_pkey_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) -{ - return mlx5i_ioctl(dev, ifr, cmd); -} - static void mlx5i_pkey_dev_cleanup(struct net_device *netdev) { mlx5i_parent_put(netdev); From 9b443e58a896fce4d377d83da8dfd083664d8739 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Thu, 30 Oct 2025 10:20:32 +0000 Subject: [PATCH 0430/1381] net: stmmac: qcom-ethqos: remove MAC_CTRL_REG modification When operating in "SGMII" mode (Cisco SGMII or 2500BASE-X), qcom-ethqos modifies the MAC control register in its ethqos_configure_sgmii() function, which is only called from one path: stmmac_mac_link_up() +- reads MAC_CTRL_REG +- masks out priv->hw->link.speed_mask +- sets bits according to speed (2500, 1000, 100, 10) from priv->hw.link.speed* +- ethqos_fix_mac_speed() | +- qcom_ethqos_set_sgmii_loopback(false) | +- ethqos_update_link_clk(speed) | `- ethqos_configure(speed) | `- ethqos_configure_sgmii(speed) | +- reads MAC_CTRL_REG, | +- configures PS/FES bits according to speed | `- writes MAC_CTRL_REG as the last operation +- sets duplex bit(s) +- stmmac_mac_flow_ctrl() +- writes MAC_CTRL_REG if changed from original read ... As can be seen, the modification of the control register that stmmac_mac_link_up() overwrites the changes that ethqos_fix_mac_speed() does to the register. This makes ethqos_configure_sgmii()'s modification questionable at best. Analysing the values written, GMAC4 sets the speed bits as: speed_mask = GMAC_CONFIG_FES | GMAC_CONFIG_PS speed2500 = GMAC_CONFIG_FES B14=1 B15=0 speed1000 = 0 B14=0 B15=0 speed100 = GMAC_CONFIG_FES | GMAC_CONFIG_PS B14=1 B15=1 speed10 = GMAC_CONFIG_PS B14=0 B15=1 Whereas ethqos_configure_sgmii(): 2500: clears ETHQOS_MAC_CTRL_PORT_SEL B14=X B15=0 1000: clears ETHQOS_MAC_CTRL_PORT_SEL B14=X B15=0 100: sets ETHQOS_MAC_CTRL_PORT_SEL | B14=1 B15=1 ETHQOS_MAC_CTRL_SPEED_MODE 10: sets ETHQOS_MAC_CTRL_PORT_SEL B14=0 B15=1 clears ETHQOS_MAC_CTRL_SPEED_MODE Thus, they appear to be doing very similar, with the exception of the FES bit (bit 14) for 1G and 2.5G speeds. Given that stmmac_mac_link_up() will write the MAC_CTRL_REG after ethqos_configure_sgmii(), remove the unnecessary update in the glue driver's ethqos_configure_sgmii() method, simplifying the code. Konrad states: Without any additional knowledge, the register description says: 2500: B14=1 B15=0 1000: B14=0 B15=0 100: B14=1 B15=1 10: B14=0 B15=1 Tested-by: Mohd Ayaan Anwar Reviewed-by: Konrad Dybcio Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vEPlg-0000000CFHY-282A@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c | 16 +--------------- 1 file changed, 1 insertion(+), 15 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c index d1e48b524d7a..1a616a71c36a 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c @@ -76,10 +76,6 @@ #define RGMII_CONFIG2_DATA_DIVIDE_CLK_SEL BIT(6) #define RGMII_CONFIG2_TX_CLK_PHASE_SHIFT_EN BIT(5) -/* MAC_CTRL_REG bits */ -#define ETHQOS_MAC_CTRL_SPEED_MODE BIT(14) -#define ETHQOS_MAC_CTRL_PORT_SEL BIT(15) - /* EMAC_WRAPPER_SGMII_PHY_CNTRL1 bits */ #define SGMII_PHY_CNTRL1_SGMII_TX_TO_RX_LOOPBACK_EN BIT(3) @@ -632,13 +628,9 @@ static int ethqos_configure_sgmii(struct qcom_ethqos *ethqos, int speed) { struct net_device *dev = platform_get_drvdata(ethqos->pdev); struct stmmac_priv *priv = netdev_priv(dev); - int val; - - val = readl(ethqos->mac_base + MAC_CTRL_REG); switch (speed) { case SPEED_2500: - val &= ~ETHQOS_MAC_CTRL_PORT_SEL; rgmii_updatel(ethqos, RGMII_CONFIG2_RGMII_CLK_SEL_CFG, RGMII_CONFIG2_RGMII_CLK_SEL_CFG, RGMII_IO_MACRO_CONFIG2); @@ -646,7 +638,6 @@ static int ethqos_configure_sgmii(struct qcom_ethqos *ethqos, int speed) ethqos_pcs_set_inband(priv, false); break; case SPEED_1000: - val &= ~ETHQOS_MAC_CTRL_PORT_SEL; rgmii_updatel(ethqos, RGMII_CONFIG2_RGMII_CLK_SEL_CFG, RGMII_CONFIG2_RGMII_CLK_SEL_CFG, RGMII_IO_MACRO_CONFIG2); @@ -654,13 +645,10 @@ static int ethqos_configure_sgmii(struct qcom_ethqos *ethqos, int speed) ethqos_pcs_set_inband(priv, true); break; case SPEED_100: - val |= ETHQOS_MAC_CTRL_PORT_SEL | ETHQOS_MAC_CTRL_SPEED_MODE; ethqos_set_serdes_speed(ethqos, SPEED_1000); ethqos_pcs_set_inband(priv, true); break; case SPEED_10: - val |= ETHQOS_MAC_CTRL_PORT_SEL; - val &= ~ETHQOS_MAC_CTRL_SPEED_MODE; rgmii_updatel(ethqos, RGMII_CONFIG_SGMII_CLK_DVDR, FIELD_PREP(RGMII_CONFIG_SGMII_CLK_DVDR, SGMII_10M_RX_CLK_DVDR), @@ -670,9 +658,7 @@ static int ethqos_configure_sgmii(struct qcom_ethqos *ethqos, int speed) break; } - writel(val, ethqos->mac_base + MAC_CTRL_REG); - - return val; + return 0; } static int ethqos_configure(struct qcom_ethqos *ethqos, int speed) From e7e756779afa102fc8f9d648ccdc2ecbc41ce2f8 Mon Sep 17 00:00:00 2001 From: Parthiban Veerasooran Date: Thu, 30 Oct 2025 15:52:57 +0530 Subject: [PATCH 0431/1381] net: phy: microchip_t1s: add support for Microchip LAN867X Rev.D0 PHY MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Add support for the LAN8670/1/2 Rev.D0 10BASE-T1S PHYs from Microchip. The new Rev.D0 silicon requires a specific set of initialization settings to be configured for optimal performance and compliance with OPEN Alliance specifications, as described in Microchip Application Note AN1699 (Revision G, DS60001699G – October 2025). https://www.microchip.com/en-us/application-notes/an1699 Signed-off-by: Parthiban Veerasooran Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20251030102258.180061-2-parthiban.veerasooran@microchip.com Signed-off-by: Jakub Kicinski --- drivers/net/phy/Kconfig | 2 +- drivers/net/phy/microchip_t1s.c | 47 ++++++++++++++++++++++++++++++++- 2 files changed, 47 insertions(+), 2 deletions(-) diff --git a/drivers/net/phy/Kconfig b/drivers/net/phy/Kconfig index 98700d069191..a7ade7b95a2e 100644 --- a/drivers/net/phy/Kconfig +++ b/drivers/net/phy/Kconfig @@ -308,7 +308,7 @@ config MICREL_PHY config MICROCHIP_T1S_PHY tristate "Microchip 10BASE-T1S Ethernet PHYs" help - Currently supports the LAN8670/1/2 Rev.B1/C1/C2 and + Currently supports the LAN8670/1/2 Rev.B1/C1/C2/D0 and LAN8650/1 Rev.B0/B1 Internal PHYs. config MICROCHIP_PHY diff --git a/drivers/net/phy/microchip_t1s.c b/drivers/net/phy/microchip_t1s.c index e50a0c102a86..03e3bacb02bd 100644 --- a/drivers/net/phy/microchip_t1s.c +++ b/drivers/net/phy/microchip_t1s.c @@ -3,7 +3,7 @@ * Driver for Microchip 10BASE-T1S PHYs * * Support: Microchip Phys: - * lan8670/1/2 Rev.B1/C1/C2 + * lan8670/1/2 Rev.B1/C1/C2/D0 * lan8650/1 Rev.B0/B1 Internal PHYs */ @@ -14,6 +14,7 @@ #define PHY_ID_LAN867X_REVB1 0x0007C162 #define PHY_ID_LAN867X_REVC1 0x0007C164 #define PHY_ID_LAN867X_REVC2 0x0007C165 +#define PHY_ID_LAN867X_REVD0 0x0007C166 /* Both Rev.B0 and B1 clause 22 PHYID's are same due to B1 chip limitation */ #define PHY_ID_LAN865X_REVB 0x0007C1B3 @@ -109,6 +110,21 @@ static const u16 lan865x_revb_sqi_fixup_cfg_regs[3] = { 0x00AD, 0x00AE, 0x00AF, }; +/* LAN867x Rev.D0 configuration parameters from AN1699 + * As per the Configuration Application Note AN1699 published in the below link, + * https://www.microchip.com/en-us/application-notes/an1699 + * Revision G (DS60001699G - October 2025) + */ +static const u16 lan867x_revd0_fixup_regs[8] = { + 0x0037, 0x008A, 0x0118, 0x00D6, + 0x0082, 0x00FD, 0x00FD, 0x0091, +}; + +static const u16 lan867x_revd0_fixup_values[8] = { + 0x0800, 0xBFC0, 0x029C, 0x1001, + 0x001C, 0x0C0B, 0x8C07, 0x9660, +}; + /* Pulled from AN1760 describing 'indirect read' * * write_register(0x4, 0x00D8, addr) @@ -407,6 +423,25 @@ static int lan86xx_plca_set_cfg(struct phy_device *phydev, COL_DET_CTRL0_ENABLE_BIT_MASK, COL_DET_ENABLE); } +static int lan867x_revd0_config_init(struct phy_device *phydev) +{ + int ret; + + ret = lan867x_check_reset_complete(phydev); + if (ret) + return ret; + + for (int i = 0; i < ARRAY_SIZE(lan867x_revd0_fixup_regs); i++) { + ret = phy_write_mmd(phydev, MDIO_MMD_VEND2, + lan867x_revd0_fixup_regs[i], + lan867x_revd0_fixup_values[i]); + if (ret) + return ret; + } + + return 0; +} + static int lan86xx_read_status(struct phy_device *phydev) { /* The phy has some limitations, namely: @@ -481,6 +516,15 @@ static struct phy_driver microchip_t1s_driver[] = { .set_plca_cfg = lan86xx_plca_set_cfg, .get_plca_status = genphy_c45_plca_get_status, }, + { + PHY_ID_MATCH_EXACT(PHY_ID_LAN867X_REVD0), + .name = "LAN867X Rev.D0", + .features = PHY_BASIC_T1S_P2MP_FEATURES, + .config_init = lan867x_revd0_config_init, + .get_plca_cfg = genphy_c45_plca_get_cfg, + .set_plca_cfg = lan86xx_plca_set_cfg, + .get_plca_status = genphy_c45_plca_get_status, + }, { PHY_ID_MATCH_EXACT(PHY_ID_LAN865X_REVB), .name = "LAN865X Rev.B0/B1 Internal Phy", @@ -501,6 +545,7 @@ static const struct mdio_device_id __maybe_unused tbl[] = { { PHY_ID_MATCH_EXACT(PHY_ID_LAN867X_REVB1) }, { PHY_ID_MATCH_EXACT(PHY_ID_LAN867X_REVC1) }, { PHY_ID_MATCH_EXACT(PHY_ID_LAN867X_REVC2) }, + { PHY_ID_MATCH_EXACT(PHY_ID_LAN867X_REVD0) }, { PHY_ID_MATCH_EXACT(PHY_ID_LAN865X_REVB) }, { } }; From 07f5765f26c3c7381a59d37c73d3ec51b4fd5cf0 Mon Sep 17 00:00:00 2001 From: Parthiban Veerasooran Date: Thu, 30 Oct 2025 15:52:58 +0530 Subject: [PATCH 0432/1381] net: phy: microchip_t1s: configure link status control for LAN867x Rev.D0 Configure the link status in the Link Status Control register for LAN8670/1/2 Rev.D0 PHYs, depending on whether PLCA or CSMA/CD mode is enabled. When PLCA is enabled, the link status reflects the PLCA status. When PLCA is disabled (CSMA/CD mode), the PHY does not support autonegotiation, so the link status is forced active by setting the LINK_STATUS_SEMAPHORE bit. The link status control is configured: - During PHY initialization, for default CSMA/CD mode. - Whenever PLCA configuration is updated. This ensures correct link reporting and consistent behavior for LAN867x Rev.D0 devices. Signed-off-by: Parthiban Veerasooran Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20251030102258.180061-3-parthiban.veerasooran@microchip.com Signed-off-by: Jakub Kicinski --- drivers/net/phy/microchip_t1s.c | 51 ++++++++++++++++++++++++++++++++- 1 file changed, 50 insertions(+), 1 deletion(-) diff --git a/drivers/net/phy/microchip_t1s.c b/drivers/net/phy/microchip_t1s.c index 03e3bacb02bd..bce5cf087b19 100644 --- a/drivers/net/phy/microchip_t1s.c +++ b/drivers/net/phy/microchip_t1s.c @@ -33,6 +33,17 @@ #define COL_DET_ENABLE BIT(15) #define COL_DET_DISABLE 0x0000 +/* LAN8670/1/2 Rev.D0 Link Status Selection Register */ +#define LAN867X_REG_LINK_STATUS_CTRL 0x0012 +#define LINK_STATUS_CONFIGURATION GENMASK(12, 11) +#define LINK_STATUS_SEMAPHORE BIT(0) + +/* Link Status Configuration */ +#define LINK_STATUS_CONFIG_PLCA_STATUS 0x1 +#define LINK_STATUS_CONFIG_SEMAPHORE 0x2 + +#define LINK_STATUS_SEMAPHORE_SET 0x1 + #define LAN865X_CFGPARAM_READ_ENABLE BIT(1) /* The arrays below are pulled from the following table from AN1699 @@ -393,6 +404,32 @@ static int lan867x_revb1_config_init(struct phy_device *phydev) return 0; } +static int lan867x_revd0_link_active_selection(struct phy_device *phydev, + bool plca_enabled) +{ + u16 value; + + if (plca_enabled) { + /* 0x1 - When PLCA is enabled: link status reflects plca_status. + */ + value = FIELD_PREP(LINK_STATUS_CONFIGURATION, + LINK_STATUS_CONFIG_PLCA_STATUS); + } else { + /* 0x2 - Link status is controlled by the value written into the + * LINK_STATUS_SEMAPHORE bit written. Here the link semaphore + * bit is written with 0x1 to set the link always active in + * CSMA/CD mode as it doesn't support autoneg. + */ + value = FIELD_PREP(LINK_STATUS_CONFIGURATION, + LINK_STATUS_CONFIG_SEMAPHORE) | + FIELD_PREP(LINK_STATUS_SEMAPHORE, + LINK_STATUS_SEMAPHORE_SET); + } + + return phy_write_mmd(phydev, MDIO_MMD_VEND2, + LAN867X_REG_LINK_STATUS_CTRL, value); +} + /* As per LAN8650/1 Rev.B0/B1 AN1760 (Revision F (DS60001760G - June 2024)) and * LAN8670/1/2 Rev.C1/C2 AN1699 (Revision E (DS60001699F - June 2024)), under * normal operation, the device should be operated in PLCA mode. Disabling @@ -409,6 +446,14 @@ static int lan86xx_plca_set_cfg(struct phy_device *phydev, { int ret; + /* Link status selection must be configured for LAN8670/1/2 Rev.D0 */ + if (phydev->phy_id == PHY_ID_LAN867X_REVD0) { + ret = lan867x_revd0_link_active_selection(phydev, + plca_cfg->enabled); + if (ret) + return ret; + } + ret = genphy_c45_plca_set_cfg(phydev, plca_cfg); if (ret) return ret; @@ -439,7 +484,11 @@ static int lan867x_revd0_config_init(struct phy_device *phydev) return ret; } - return 0; + /* Initially the PHY will be in CSMA/CD mode by default. So it is + * required to set the link always active as it doesn't support + * autoneg. + */ + return lan867x_revd0_link_active_selection(phydev, false); } static int lan86xx_read_status(struct phy_device *phydev) From a7aca10c0091d511030ec7907667e1448869b71c Mon Sep 17 00:00:00 2001 From: Bagas Sanjaya Date: Thu, 30 Oct 2025 14:50:13 +0700 Subject: [PATCH 0433/1381] Documentation: netconsole: Separate literal code blocks for full and short netcat command name versions Both full and short (abbreviated) command name versions of netcat example are combined in single literal code block due to 'or::' paragraph being indented one more space than the preceding paragraph (before the short version example). Unindent it to separate the versions. Signed-off-by: Bagas Sanjaya Reviewed-by: Randy Dunlap Tested-by: Randy Dunlap Link: https://patch.msgid.link/20251030075013.40418-1-bagasdotme@gmail.com Signed-off-by: Jakub Kicinski --- Documentation/networking/netconsole.rst | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Documentation/networking/netconsole.rst b/Documentation/networking/netconsole.rst index 2555e75e5cc1..4ab5d7b05cf1 100644 --- a/Documentation/networking/netconsole.rst +++ b/Documentation/networking/netconsole.rst @@ -88,7 +88,7 @@ for example: nc -u -l -p ' / 'nc -u -l - or:: + or:: netcat -u -l -p ' / 'netcat -u -l From 4d07797faaa19aa8e80e10a04ca1a72c643ef5cf Mon Sep 17 00:00:00 2001 From: Thomas Wismer Date: Wed, 29 Oct 2025 22:23:09 +0100 Subject: [PATCH 0434/1381] net: pse-pd: tps23881: Add support for TPS23881B The TPS23881B uses different firmware than the TPS23881. Trying to load the TPS23881 firmware on a TPS23881B device fails and must be omitted. The TPS23881B ships with a more recent ROM firmware. Moreover, no updated firmware has been released yet and so the firmware loading step must be skipped. As of today, the TPS23881B is intended to use its ROM firmware. Signed-off-by: Thomas Wismer Reviewed-by: Kory Maincent Acked-by: Oleksij Rempel Link: https://patch.msgid.link/20251029212312.108749-2-thomas@wismer.xyz Signed-off-by: Jakub Kicinski --- drivers/net/pse-pd/tps23881.c | 69 +++++++++++++++++++++++++++-------- 1 file changed, 54 insertions(+), 15 deletions(-) diff --git a/drivers/net/pse-pd/tps23881.c b/drivers/net/pse-pd/tps23881.c index b724b222ab44..76ec1555d60d 100644 --- a/drivers/net/pse-pd/tps23881.c +++ b/drivers/net/pse-pd/tps23881.c @@ -55,8 +55,6 @@ #define TPS23881_REG_TPON BIT(0) #define TPS23881_REG_FWREV 0x41 #define TPS23881_REG_DEVID 0x43 -#define TPS23881_REG_DEVID_MASK 0xF0 -#define TPS23881_DEVICE_ID 0x02 #define TPS23881_REG_CHAN1_CLASS 0x4c #define TPS23881_REG_SRAM_CTRL 0x60 #define TPS23881_REG_SRAM_DATA 0x61 @@ -1012,8 +1010,28 @@ static const struct pse_controller_ops tps23881_ops = { .pi_get_pw_req = tps23881_pi_get_pw_req, }; -static const char fw_parity_name[] = "ti/tps23881/tps23881-parity-14.bin"; -static const char fw_sram_name[] = "ti/tps23881/tps23881-sram-14.bin"; +struct tps23881_info { + u8 dev_id; /* device ID and silicon revision */ + const char *fw_parity_name; /* parity code firmware file name */ + const char *fw_sram_name; /* SRAM code firmware file name */ +}; + +enum tps23881_model { + TPS23881, + TPS23881B, +}; + +static const struct tps23881_info tps23881_info[] = { + [TPS23881] = { + .dev_id = 0x22, + .fw_parity_name = "ti/tps23881/tps23881-parity-14.bin", + .fw_sram_name = "ti/tps23881/tps23881-sram-14.bin", + }, + [TPS23881B] = { + .dev_id = 0x24, + /* skip SRAM load, ROM provides Clause 145 hardware-level support */ + }, +}; struct tps23881_fw_conf { u8 reg; @@ -1085,16 +1103,17 @@ static int tps23881_flash_sram_fw_part(struct i2c_client *client, return ret; } -static int tps23881_flash_sram_fw(struct i2c_client *client) +static int tps23881_flash_sram_fw(struct i2c_client *client, + const struct tps23881_info *info) { int ret; - ret = tps23881_flash_sram_fw_part(client, fw_parity_name, + ret = tps23881_flash_sram_fw_part(client, info->fw_parity_name, tps23881_fw_parity_conf); if (ret) return ret; - ret = tps23881_flash_sram_fw_part(client, fw_sram_name, + ret = tps23881_flash_sram_fw_part(client, info->fw_sram_name, tps23881_fw_sram_conf); if (ret) return ret; @@ -1412,6 +1431,7 @@ static int tps23881_setup_irq(struct tps23881_priv *priv, int irq) static int tps23881_i2c_probe(struct i2c_client *client) { struct device *dev = &client->dev; + const struct tps23881_info *info; struct tps23881_priv *priv; struct gpio_desc *reset; int ret; @@ -1422,6 +1442,10 @@ static int tps23881_i2c_probe(struct i2c_client *client) return -ENXIO; } + info = i2c_get_match_data(client); + if (!info) + return -EINVAL; + priv = devm_kzalloc(dev, sizeof(*priv), GFP_KERNEL); if (!priv) return -ENOMEM; @@ -1440,7 +1464,7 @@ static int tps23881_i2c_probe(struct i2c_client *client) * to Load TPS2388x SRAM and Parity Code over I2C" (Rev E)) * indicates we should delay that programming by at least 50ms. So * we'll wait the entire 50ms here to ensure we're safe to go to the - * SRAM loading proceedure. + * SRAM loading procedure. */ msleep(50); } @@ -1449,20 +1473,27 @@ static int tps23881_i2c_probe(struct i2c_client *client) if (ret < 0) return ret; - if (FIELD_GET(TPS23881_REG_DEVID_MASK, ret) != TPS23881_DEVICE_ID) { + if (ret != info->dev_id) { dev_err(dev, "Wrong device ID\n"); return -ENXIO; } - ret = tps23881_flash_sram_fw(client); - if (ret < 0) - return ret; + if (info->fw_sram_name) { + ret = tps23881_flash_sram_fw(client, info); + if (ret < 0) + return ret; + } ret = i2c_smbus_read_byte_data(client, TPS23881_REG_FWREV); if (ret < 0) return ret; - dev_info(&client->dev, "Firmware revision 0x%x\n", ret); + if (ret == 0xFF) { + dev_err(&client->dev, "Device entered safe mode\n"); + return -ENXIO; + } + dev_info(&client->dev, "Firmware revision 0x%x%s\n", ret, + ret == 0x00 ? " (ROM firmware)" : ""); /* Set configuration B, 16 bit access on a single device address */ ret = i2c_smbus_read_byte_data(client, TPS23881_REG_GEN_MASK); @@ -1498,13 +1529,21 @@ static int tps23881_i2c_probe(struct i2c_client *client) } static const struct i2c_device_id tps23881_id[] = { - { "tps23881" }, + { "tps23881", .driver_data = (kernel_ulong_t)&tps23881_info[TPS23881] }, + { "tps23881b", .driver_data = (kernel_ulong_t)&tps23881_info[TPS23881B] }, { } }; MODULE_DEVICE_TABLE(i2c, tps23881_id); static const struct of_device_id tps23881_of_match[] = { - { .compatible = "ti,tps23881", }, + { + .compatible = "ti,tps23881", + .data = &tps23881_info[TPS23881] + }, + { + .compatible = "ti,tps23881b", + .data = &tps23881_info[TPS23881B] + }, { }, }; MODULE_DEVICE_TABLE(of, tps23881_of_match); From 32032eb166a6d05d6bb4803c9b9e39659990b18a Mon Sep 17 00:00:00 2001 From: Thomas Wismer Date: Wed, 29 Oct 2025 22:23:10 +0100 Subject: [PATCH 0435/1381] dt-bindings: pse-pd: ti,tps23881: Add TPS23881B Add the TPS23881B I2C power sourcing equipment controller to the list of supported devices. Falling back to the TPS23881 predecessor device is not suitable as firmware loading needs to handled differently by the driver. The TPS23881 and TPS23881B devices require different firmware. Trying to load the TPS23881 firmware on a TPS23881B device fails and must therefore be omitted. Signed-off-by: Thomas Wismer Acked-by: Conor Dooley Reviewed-by: Kory Maincent Link: https://patch.msgid.link/20251029212312.108749-3-thomas@wismer.xyz Signed-off-by: Jakub Kicinski --- Documentation/devicetree/bindings/net/pse-pd/ti,tps23881.yaml | 1 + 1 file changed, 1 insertion(+) diff --git a/Documentation/devicetree/bindings/net/pse-pd/ti,tps23881.yaml b/Documentation/devicetree/bindings/net/pse-pd/ti,tps23881.yaml index bb1ee3398655..0b3803f647b7 100644 --- a/Documentation/devicetree/bindings/net/pse-pd/ti,tps23881.yaml +++ b/Documentation/devicetree/bindings/net/pse-pd/ti,tps23881.yaml @@ -16,6 +16,7 @@ properties: compatible: enum: - ti,tps23881 + - ti,tps23881b reg: maxItems: 1 From 30176bf7c871681df506f3165ffe76ec462db991 Mon Sep 17 00:00:00 2001 From: Ivan Vecera Date: Wed, 29 Oct 2025 16:32:06 +0100 Subject: [PATCH 0436/1381] dpll: add phase-adjust-gran pin attribute Phase-adjust values are currently limited by a min-max range. Some hardware requires, for certain pin types, that values be multiples of a specific granularity, as in the zl3073x driver. Add a `phase-adjust-gran` pin attribute and an appropriate field in dpll_pin_properties. If set by the driver, use its value to validate user-provided phase-adjust values. Reviewed-by: Michal Schmidt Reviewed-by: Petr Oros Tested-by: Prathosh Satish Signed-off-by: Ivan Vecera Reviewed-by: Jiri Pirko Reviewed-by: Arkadiusz Kubalewski Link: https://patch.msgid.link/20251029153207.178448-2-ivecera@redhat.com Signed-off-by: Jakub Kicinski --- Documentation/driver-api/dpll.rst | 36 +++++++++++++++------------ Documentation/netlink/specs/dpll.yaml | 7 ++++++ drivers/dpll/dpll_netlink.c | 12 ++++++++- include/linux/dpll.h | 1 + include/uapi/linux/dpll.h | 1 + 5 files changed, 40 insertions(+), 17 deletions(-) diff --git a/Documentation/driver-api/dpll.rst b/Documentation/driver-api/dpll.rst index be1fc643b645..83118c728ed9 100644 --- a/Documentation/driver-api/dpll.rst +++ b/Documentation/driver-api/dpll.rst @@ -198,26 +198,28 @@ be requested with the same attribute with ``DPLL_CMD_DEVICE_SET`` command. ================================== ====================================== Device may also provide ability to adjust a signal phase on a pin. -If pin phase adjustment is supported, minimal and maximal values that pin -handle shall be provide to the user on ``DPLL_CMD_PIN_GET`` respond -with ``DPLL_A_PIN_PHASE_ADJUST_MIN`` and ``DPLL_A_PIN_PHASE_ADJUST_MAX`` +If pin phase adjustment is supported, minimal and maximal values and +granularity that pin handle shall be provided to the user on +``DPLL_CMD_PIN_GET`` respond with ``DPLL_A_PIN_PHASE_ADJUST_MIN``, +``DPLL_A_PIN_PHASE_ADJUST_MAX`` and ``DPLL_A_PIN_PHASE_ADJUST_GRAN`` attributes. Configured phase adjust value is provided with ``DPLL_A_PIN_PHASE_ADJUST`` attribute of a pin, and value change can be requested with the same attribute with ``DPLL_CMD_PIN_SET`` command. - =============================== ====================================== - ``DPLL_A_PIN_ID`` configured pin id - ``DPLL_A_PIN_PHASE_ADJUST_MIN`` attr minimum value of phase adjustment - ``DPLL_A_PIN_PHASE_ADJUST_MAX`` attr maximum value of phase adjustment - ``DPLL_A_PIN_PHASE_ADJUST`` attr configured value of phase - adjustment on parent dpll device - ``DPLL_A_PIN_PARENT_DEVICE`` nested attribute for requesting - configuration on given parent dpll - device - ``DPLL_A_PIN_PARENT_ID`` parent dpll device id - ``DPLL_A_PIN_PHASE_OFFSET`` attr measured phase difference - between a pin and parent dpll device - =============================== ====================================== + ================================ ========================================== + ``DPLL_A_PIN_ID`` configured pin id + ``DPLL_A_PIN_PHASE_ADJUST_GRAN`` attr granularity of phase adjustment value + ``DPLL_A_PIN_PHASE_ADJUST_MIN`` attr minimum value of phase adjustment + ``DPLL_A_PIN_PHASE_ADJUST_MAX`` attr maximum value of phase adjustment + ``DPLL_A_PIN_PHASE_ADJUST`` attr configured value of phase + adjustment on parent dpll device + ``DPLL_A_PIN_PARENT_DEVICE`` nested attribute for requesting + configuration on given parent dpll + device + ``DPLL_A_PIN_PARENT_ID`` parent dpll device id + ``DPLL_A_PIN_PHASE_OFFSET`` attr measured phase difference + between a pin and parent dpll device + ================================ ========================================== All phase related values are provided in pico seconds, which represents time difference between signals phase. The negative value means that @@ -384,6 +386,8 @@ according to attribute purpose. frequencies ``DPLL_A_PIN_ANY_FREQUENCY_MIN`` attr minimum value of frequency ``DPLL_A_PIN_ANY_FREQUENCY_MAX`` attr maximum value of frequency + ``DPLL_A_PIN_PHASE_ADJUST_GRAN`` attr granularity of phase + adjustment value ``DPLL_A_PIN_PHASE_ADJUST_MIN`` attr minimum value of phase adjustment ``DPLL_A_PIN_PHASE_ADJUST_MAX`` attr maximum value of phase diff --git a/Documentation/netlink/specs/dpll.yaml b/Documentation/netlink/specs/dpll.yaml index 80728f6f9bc8..78d0724d7e12 100644 --- a/Documentation/netlink/specs/dpll.yaml +++ b/Documentation/netlink/specs/dpll.yaml @@ -440,6 +440,12 @@ attribute-sets: doc: | Capable pin provides list of pins that can be bound to create a reference-sync pin pair. + - + name: phase-adjust-gran + type: u32 + doc: | + Granularity of phase adjustment, in picoseconds. The value of + phase adjustment must be a multiple of this granularity. - name: pin-parent-device @@ -616,6 +622,7 @@ operations: - capabilities - parent-device - parent-pin + - phase-adjust-gran - phase-adjust-min - phase-adjust-max - phase-adjust diff --git a/drivers/dpll/dpll_netlink.c b/drivers/dpll/dpll_netlink.c index a4153bcb6dcf..64944f601ee5 100644 --- a/drivers/dpll/dpll_netlink.c +++ b/drivers/dpll/dpll_netlink.c @@ -637,6 +637,10 @@ dpll_cmd_pin_get_one(struct sk_buff *msg, struct dpll_pin *pin, ret = dpll_msg_add_pin_freq(msg, pin, ref, extack); if (ret) return ret; + if (prop->phase_gran && + nla_put_u32(msg, DPLL_A_PIN_PHASE_ADJUST_GRAN, + prop->phase_gran)) + return -EMSGSIZE; if (nla_put_s32(msg, DPLL_A_PIN_PHASE_ADJUST_MIN, prop->phase_range.min)) return -EMSGSIZE; @@ -1261,7 +1265,13 @@ dpll_pin_phase_adj_set(struct dpll_pin *pin, struct nlattr *phase_adj_attr, if (phase_adj > pin->prop.phase_range.max || phase_adj < pin->prop.phase_range.min) { NL_SET_ERR_MSG_ATTR(extack, phase_adj_attr, - "phase adjust value not supported"); + "phase adjust value of out range"); + return -EINVAL; + } + if (pin->prop.phase_gran && phase_adj % (s32)pin->prop.phase_gran) { + NL_SET_ERR_MSG_ATTR_FMT(extack, phase_adj_attr, + "phase adjust value not multiple of %u", + pin->prop.phase_gran); return -EINVAL; } diff --git a/include/linux/dpll.h b/include/linux/dpll.h index 25be745bf41f..562f520b23c2 100644 --- a/include/linux/dpll.h +++ b/include/linux/dpll.h @@ -163,6 +163,7 @@ struct dpll_pin_properties { u32 freq_supported_num; struct dpll_pin_frequency *freq_supported; struct dpll_pin_phase_adjust_range phase_range; + u32 phase_gran; }; #if IS_ENABLED(CONFIG_DPLL) diff --git a/include/uapi/linux/dpll.h b/include/uapi/linux/dpll.h index ab1725a954d7..69d35570ac4f 100644 --- a/include/uapi/linux/dpll.h +++ b/include/uapi/linux/dpll.h @@ -251,6 +251,7 @@ enum dpll_a_pin { DPLL_A_PIN_ESYNC_FREQUENCY_SUPPORTED, DPLL_A_PIN_ESYNC_PULSE, DPLL_A_PIN_REFERENCE_SYNC, + DPLL_A_PIN_PHASE_ADJUST_GRAN, __DPLL_A_PIN_MAX, DPLL_A_PIN_MAX = (__DPLL_A_PIN_MAX - 1) From 055a01b29fd643e33b9b1e88e24bbe1afe6fc6d9 Mon Sep 17 00:00:00 2001 From: Ivan Vecera Date: Wed, 29 Oct 2025 16:32:07 +0100 Subject: [PATCH 0437/1381] dpll: zl3073x: Specify phase adjustment granularity for pins Output pins phase adjustment values in the device are expressed in half synth clock cycles. Use this number of cycles as output pins' phase adjust granularity and simplify both get/set callbacks. Reviewed-by: Michal Schmidt Reviewed-by: Petr Oros Tested-by: Prathosh Satish Signed-off-by: Ivan Vecera Reviewed-by: Arkadiusz Kubalewski Link: https://patch.msgid.link/20251029153207.178448-3-ivecera@redhat.com Signed-off-by: Jakub Kicinski --- drivers/dpll/zl3073x/dpll.c | 58 +++++++++---------------------------- drivers/dpll/zl3073x/prop.c | 11 +++++++ 2 files changed, 25 insertions(+), 44 deletions(-) diff --git a/drivers/dpll/zl3073x/dpll.c b/drivers/dpll/zl3073x/dpll.c index f93f9a458324..d90150671d37 100644 --- a/drivers/dpll/zl3073x/dpll.c +++ b/drivers/dpll/zl3073x/dpll.c @@ -35,6 +35,7 @@ * @prio: pin priority <0, 14> * @selectable: pin is selectable in automatic mode * @esync_control: embedded sync is controllable + * @phase_gran: phase adjustment granularity * @pin_state: last saved pin state * @phase_offset: last saved pin phase offset * @freq_offset: last saved fractional frequency offset @@ -49,6 +50,7 @@ struct zl3073x_dpll_pin { u8 prio; bool selectable; bool esync_control; + s32 phase_gran; enum dpll_pin_state pin_state; s64 phase_offset; s64 freq_offset; @@ -1388,25 +1390,14 @@ zl3073x_dpll_output_pin_phase_adjust_get(const struct dpll_pin *dpll_pin, struct zl3073x_dpll *zldpll = dpll_priv; struct zl3073x_dev *zldev = zldpll->dev; struct zl3073x_dpll_pin *pin = pin_priv; - u32 synth_freq; s32 phase_comp; - u8 out, synth; + u8 out; int rc; - out = zl3073x_output_pin_out_get(pin->id); - synth = zl3073x_out_synth_get(zldev, out); - synth_freq = zl3073x_synth_freq_get(zldev, synth); - - /* Check synth freq for zero */ - if (!synth_freq) { - dev_err(zldev->dev, "Got zero synth frequency for output %u\n", - out); - return -EINVAL; - } - guard(mutex)(&zldev->multiop_lock); /* Read output configuration */ + out = zl3073x_output_pin_out_get(pin->id); rc = zl3073x_mb_op(zldev, ZL_REG_OUTPUT_MB_SEM, ZL_OUTPUT_MB_SEM_RD, ZL_REG_OUTPUT_MB_MASK, BIT(out)); if (rc) @@ -1417,11 +1408,10 @@ zl3073x_dpll_output_pin_phase_adjust_get(const struct dpll_pin *dpll_pin, if (rc) return rc; - /* Value in register is expressed in half synth clock cycles */ - phase_comp *= (int)div_u64(PSEC_PER_SEC, 2 * synth_freq); - - /* Reverse two's complement negation applied during 'set' */ - *phase_adjust = -phase_comp; + /* Convert value to ps and reverse two's complement negation applied + * during 'set' + */ + *phase_adjust = -phase_comp * pin->phase_gran; return rc; } @@ -1437,39 +1427,18 @@ zl3073x_dpll_output_pin_phase_adjust_set(const struct dpll_pin *dpll_pin, struct zl3073x_dpll *zldpll = dpll_priv; struct zl3073x_dev *zldev = zldpll->dev; struct zl3073x_dpll_pin *pin = pin_priv; - int half_synth_cycle; - u32 synth_freq; - u8 out, synth; + u8 out; int rc; - /* Get attached synth */ - out = zl3073x_output_pin_out_get(pin->id); - synth = zl3073x_out_synth_get(zldev, out); - - /* Get synth's frequency */ - synth_freq = zl3073x_synth_freq_get(zldev, synth); - - /* Value in register is expressed in half synth clock cycles so - * the given phase adjustment a multiple of half synth clock. - */ - half_synth_cycle = (int)div_u64(PSEC_PER_SEC, 2 * synth_freq); - - if ((phase_adjust % half_synth_cycle) != 0) { - NL_SET_ERR_MSG_FMT(extack, - "Phase adjustment value has to be multiple of %d", - half_synth_cycle); - return -EINVAL; - } - phase_adjust /= half_synth_cycle; - /* The value in the register is stored as two's complement negation - * of requested value. + * of requested value and expressed in half synth clock cycles. */ - phase_adjust = -phase_adjust; + phase_adjust = -phase_adjust / pin->phase_gran; guard(mutex)(&zldev->multiop_lock); /* Read output configuration */ + out = zl3073x_output_pin_out_get(pin->id); rc = zl3073x_mb_op(zldev, ZL_REG_OUTPUT_MB_SEM, ZL_OUTPUT_MB_SEM_RD, ZL_REG_OUTPUT_MB_MASK, BIT(out)); if (rc) @@ -1758,9 +1727,10 @@ zl3073x_dpll_pin_register(struct zl3073x_dpll_pin *pin, u32 index) if (IS_ERR(props)) return PTR_ERR(props); - /* Save package label & esync capability */ + /* Save package label, esync capability and phase adjust granularity */ strscpy(pin->label, props->package_label); pin->esync_control = props->esync_control; + pin->phase_gran = props->dpll_props.phase_gran; if (zl3073x_dpll_is_input_pin(pin)) { rc = zl3073x_dpll_ref_prio_get(pin, &pin->prio); diff --git a/drivers/dpll/zl3073x/prop.c b/drivers/dpll/zl3073x/prop.c index 4cf7e8aefcb3..9e1fca5cdaf1 100644 --- a/drivers/dpll/zl3073x/prop.c +++ b/drivers/dpll/zl3073x/prop.c @@ -208,7 +208,18 @@ struct zl3073x_pin_props *zl3073x_pin_props_get(struct zl3073x_dev *zldev, DPLL_PIN_CAPABILITIES_PRIORITY_CAN_CHANGE | DPLL_PIN_CAPABILITIES_STATE_CAN_CHANGE; } else { + u8 out, synth; + u32 f; + props->dpll_props.type = DPLL_PIN_TYPE_GNSS; + + /* The output pin phase adjustment granularity equals half of + * the synth frequency count. + */ + out = zl3073x_output_pin_out_get(index); + synth = zl3073x_out_synth_get(zldev, out); + f = 2 * zl3073x_synth_freq_get(zldev, synth); + props->dpll_props.phase_gran = f ? div_u64(PSEC_PER_SEC, f) : 1; } props->dpll_props.phase_range.min = S32_MIN; From 01cc760632b875c4ad0d8fec0b0c01896b8a36d4 Mon Sep 17 00:00:00 2001 From: Bagas Sanjaya Date: Tue, 28 Oct 2025 08:44:52 +0700 Subject: [PATCH 0438/1381] Documentation: ARCnet: Update obsolete contact info ARCnet docs states that inquiries on the subsystem should be emailed to Avery Pennarun , for whom has been in CREDITS since the beginning of kernel git history and her email address is unreachable (bounce). The subsystem is now maintained by Michael Grzeschik since c38f6ac74c9980 ("MAINTAINERS: add arcnet and take maintainership"). In addition, there used to be a dedicated ARCnet mailing list but its archive at epistolary.org has been shut down. ARCnet discussion nowadays take place in netdev list. The arcnet.com domain mentioned has become AIoT (Artificial Intelligence of Things) related Typeform page and ARCnet info now resides on arcnet.cc (ARCnet Resource Center) instead. Update contact information. Signed-off-by: Bagas Sanjaya Reviewed-by: Randy Dunlap Tested-by: Randy Dunlap Link: https://patch.msgid.link/20251028014451.10521-2-bagasdotme@gmail.com Signed-off-by: Jakub Kicinski --- Documentation/networking/arcnet-hardware.rst | 22 ++++----- Documentation/networking/arcnet.rst | 48 +++++--------------- 2 files changed, 21 insertions(+), 49 deletions(-) diff --git a/Documentation/networking/arcnet-hardware.rst b/Documentation/networking/arcnet-hardware.rst index 3bf7f99cd7bb..20e5075d0d0e 100644 --- a/Documentation/networking/arcnet-hardware.rst +++ b/Documentation/networking/arcnet-hardware.rst @@ -4,18 +4,20 @@ ARCnet Hardware =============== +:Author: Avery Pennarun + .. note:: - 1) This file is a supplement to arcnet.txt. Please read that for general + 1) This file is a supplement to arcnet.rst. Please read that for general driver configuration help. 2) This file is no longer Linux-specific. It should probably be moved out of the kernel sources. Ideas? Because so many people (myself included) seem to have obtained ARCnet cards without manuals, this file contains a quick introduction to ARCnet hardware, -some cabling tips, and a listing of all jumper settings I can find. Please -e-mail apenwarr@worldvisions.ca with any settings for your particular card, -or any other information you have! +some cabling tips, and a listing of all jumper settings I can find. If you +have any settings for your particular card, and/or any other information you +have, do not hesitate to :ref:`email to netdev `. Introduction to ARCnet @@ -72,11 +74,10 @@ level of encapsulation is defined by RFC1201, which I call "packet splitting," that allows "virtual packets" to grow as large as 64K each, although they are generally kept down to the Ethernet-style 1500 bytes. -For more information on the advantages and disadvantages (mostly the -advantages) of ARCnet networks, you might try the "ARCnet Trade Association" -WWW page: +For more information on ARCnet networks, visit the "ARCNET Resource Center" +WWW page at: - http://www.arcnet.com + https://www.arcnet.cc Cabling ARCnet Networks @@ -3226,9 +3227,6 @@ Settings for IRQ Selection (Lower Jumper Line) Other Cards =========== -I have no information on other models of ARCnet cards at the moment. Please -send any and all info to: - - apenwarr@worldvisions.ca +I have no information on other models of ARCnet cards at the moment. Thanks. diff --git a/Documentation/networking/arcnet.rst b/Documentation/networking/arcnet.rst index 82fce606c0f0..cd43a18ad149 100644 --- a/Documentation/networking/arcnet.rst +++ b/Documentation/networking/arcnet.rst @@ -4,6 +4,8 @@ ARCnet ====== +:Author: Avery Pennarun + .. note:: See also arcnet-hardware.txt in this directory for jumper-setting @@ -30,18 +32,7 @@ Come on, be a sport! Send me a success report! (hey, that was even better than my original poem... this is getting bad!) - -.. warning:: - - If you don't e-mail me about your success/failure soon, I may be forced to - start SINGING. And we don't want that, do we? - - (You know, it might be argued that I'm pushing this point a little too much. - If you think so, why not flame me in a quick little e-mail? Please also - include the type of card(s) you're using, software, size of network, and - whether it's working or not.) - - My e-mail address is: apenwarr@worldvisions.ca +---- These are the ARCnet drivers for Linux. @@ -59,23 +50,14 @@ ARCnet 2.10 ALPHA, Tomasz's all-new-and-improved RFC1051 support has been included and seems to be working fine! +.. _arcnet-netdev: + Where do I discuss these drivers? --------------------------------- -Tomasz has been so kind as to set up a new and improved mailing list. -Subscribe by sending a message with the BODY "subscribe linux-arcnet YOUR -REAL NAME" to listserv@tichy.ch.uj.edu.pl. Then, to submit messages to the -list, mail to linux-arcnet@tichy.ch.uj.edu.pl. - -There are archives of the mailing list at: - - http://epistolary.org/mailman/listinfo.cgi/arcnet - -The people on linux-net@vger.kernel.org (now defunct, replaced by -netdev@vger.kernel.org) have also been known to be very helpful, especially -when we're talking about ALPHA Linux kernels that may or may not work right -in the first place. - +ARCnet discussions take place on netdev. Simply send your email to +netdev@vger.kernel.org and make sure to Cc: maintainer listed in +"ARCNET NETWORK LAYER" heading of Documentation/process/maintainers.rst. Other Drivers and Info ---------------------- @@ -523,17 +505,9 @@ can set up your network then: It works: what now? ------------------- -Send mail describing your setup, preferably including driver version, kernel -version, ARCnet card model, CPU type, number of systems on your network, and -list of software in use to me at the following address: - - apenwarr@worldvisions.ca - -I do send (sometimes automated) replies to all messages I receive. My email -can be weird (and also usually gets forwarded all over the place along the -way to me), so if you don't get a reply within a reasonable time, please -resend. - +Send mail following :ref:`arcnet-netdev`. Describe your setup, preferably +including driver version, kernel version, ARCnet card model, CPU type, number +of systems on your network, and list of software in use. It doesn't work: what now? -------------------------- From acbf1d0a9aeb1035e74c3750a6b31db0b7b69ed4 Mon Sep 17 00:00:00 2001 From: Alok Tiwari Date: Fri, 31 Oct 2025 04:26:44 -0700 Subject: [PATCH 0439/1381] hinic3: fix misleading error message in hinic3_open_channel() The error message printed when hinic3_configure() fails incorrectly reports "Failed to init txrxq irq", which does not match the actual operation performed. The hinic3_configure() function sets up various device resources such as MTU and RSS parameters , not IRQ initialization. Update the log to "Failed to configure device resources" to make the message accurate and clearer for debugging. Signed-off-by: Alok Tiwari Reviewed-by: Fan Gong Link: https://patch.msgid.link/20251031112654.46187-1-alok.a.tiwari@oracle.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/huawei/hinic3/hinic3_netdev_ops.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/huawei/hinic3/hinic3_netdev_ops.c b/drivers/net/ethernet/huawei/hinic3/hinic3_netdev_ops.c index 0fa3c7900225..bbf22811a029 100644 --- a/drivers/net/ethernet/huawei/hinic3/hinic3_netdev_ops.c +++ b/drivers/net/ethernet/huawei/hinic3/hinic3_netdev_ops.c @@ -304,7 +304,7 @@ static int hinic3_open_channel(struct net_device *netdev) err = hinic3_configure(netdev); if (err) { - netdev_err(netdev, "Failed to init txrxq irq\n"); + netdev_err(netdev, "Failed to configure device resources\n"); goto err_uninit_qps_irq; } From 18aa36238a4d835c1644dcccd63d32c7fdd4b310 Mon Sep 17 00:00:00 2001 From: Jianhui Zhao Date: Sun, 2 Nov 2025 16:26:37 +0100 Subject: [PATCH 0440/1381] net: phy: realtek: add interrupt support for RTL8221B This commit introduces interrupt support for RTL8221B (C45 mode). Interrupts are mapped on the VEND2 page. VEND2 registers are only accessible via C45 reads and cannot be accessed by C45 over C22. Signed-off-by: Jianhui Zhao [Enable only link state change interrupts] Signed-off-by: Aleksander Jan Bajkowski Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20251102152644.1676482-1-olek2@wp.pl Signed-off-by: Jakub Kicinski --- drivers/net/phy/realtek/realtek_main.c | 56 ++++++++++++++++++++++++++ 1 file changed, 56 insertions(+) diff --git a/drivers/net/phy/realtek/realtek_main.c b/drivers/net/phy/realtek/realtek_main.c index 1fd4b6cf5c1e..417f9a88aab6 100644 --- a/drivers/net/phy/realtek/realtek_main.c +++ b/drivers/net/phy/realtek/realtek_main.c @@ -128,6 +128,11 @@ */ #define RTL822X_VND2_C22_REG(reg) (0xa400 + 2 * (reg)) +#define RTL8221B_VND2_INER 0xa4d2 +#define RTL8221B_VND2_INER_LINK_STATUS BIT(4) + +#define RTL8221B_VND2_INSR 0xa4d4 + #define RTL8224_MII_RTCT 0x11 #define RTL8224_MII_RTCT_ENABLE BIT(0) #define RTL8224_MII_RTCT_PAIR_A BIT(4) @@ -1880,6 +1885,53 @@ static irqreturn_t rtl9000a_handle_interrupt(struct phy_device *phydev) return IRQ_HANDLED; } +static int rtl8221b_ack_interrupt(struct phy_device *phydev) +{ + int err; + + err = phy_read_mmd(phydev, MDIO_MMD_VEND2, RTL8221B_VND2_INSR); + + return (err < 0) ? err : 0; +} + +static int rtl8221b_config_intr(struct phy_device *phydev) +{ + int err; + + if (phydev->interrupts == PHY_INTERRUPT_ENABLED) { + err = rtl8221b_ack_interrupt(phydev); + if (err) + return err; + + err = phy_write_mmd(phydev, MDIO_MMD_VEND2, RTL8221B_VND2_INER, + RTL8221B_VND2_INER_LINK_STATUS); + } else { + err = phy_write_mmd(phydev, MDIO_MMD_VEND2, + RTL8221B_VND2_INER, 0); + if (err) + return err; + + err = rtl8221b_ack_interrupt(phydev); + } + + return err; +} + +static irqreturn_t rtl8221b_handle_interrupt(struct phy_device *phydev) +{ + int err; + + err = rtl8221b_ack_interrupt(phydev); + if (err) { + phy_error(phydev); + return IRQ_NONE; + } + + phy_trigger_machine(phydev); + + return IRQ_HANDLED; +} + static struct phy_driver realtek_drvs[] = { { PHY_ID_MATCH_EXACT(0x00008201), @@ -2054,6 +2106,8 @@ static struct phy_driver realtek_drvs[] = { }, { .match_phy_device = rtl8221b_vb_cg_c45_match_phy_device, .name = "RTL8221B-VB-CG 2.5Gbps PHY (C45)", + .config_intr = rtl8221b_config_intr, + .handle_interrupt = rtl8221b_handle_interrupt, .probe = rtl822x_probe, .config_init = rtl822xb_config_init, .get_rate_matching = rtl822xb_get_rate_matching, @@ -2078,6 +2132,8 @@ static struct phy_driver realtek_drvs[] = { }, { .match_phy_device = rtl8221b_vm_cg_c45_match_phy_device, .name = "RTL8221B-VM-CG 2.5Gbps PHY (C45)", + .config_intr = rtl8221b_config_intr, + .handle_interrupt = rtl8221b_handle_interrupt, .probe = rtl822x_probe, .config_init = rtl822xb_config_init, .get_rate_matching = rtl822xb_get_rate_matching, From 22795871edea35dfaf63011782ea77a3f440a655 Mon Sep 17 00:00:00 2001 From: Colin Ian King Date: Sat, 1 Nov 2025 18:34:46 +0000 Subject: [PATCH 0441/1381] net: dsa: yt921x: Fix spelling mistake "stucked" -> "stuck" There is a spelling mistake in a dev_err message. Fix it. Signed-off-by: Colin Ian King Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20251101183446.32134-1-colin.i.king@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/yt921x.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/dsa/yt921x.c b/drivers/net/dsa/yt921x.c index ab762ffc4661..944988e29127 100644 --- a/drivers/net/dsa/yt921x.c +++ b/drivers/net/dsa/yt921x.c @@ -1131,7 +1131,7 @@ static int yt921x_fdb_wait(struct yt921x_priv *priv, u32 *valp) res = yt921x_reg_wait(priv, YT921X_FDB_RESULT, YT921X_FDB_RESULT_DONE, &val); if (res) { - dev_err(dev, "FDB probably stucked\n"); + dev_err(dev, "FDB probably stuck\n"); return res; } From 209ff7af79bf495e6c3d300bf3dea6aeea973bc7 Mon Sep 17 00:00:00 2001 From: Maxime Chevallier Date: Thu, 30 Oct 2025 19:24:53 +0100 Subject: [PATCH 0442/1381] net: stmmac: rename devlink parameter ts_coarse into phc_coarse_adj The devlink param "ts_coarse" doesn't indicate that we get coarse timestamps, but rather that the PHC clock adjusments are coarse as the frequency won't be continuously adjusted. Adjust the devlink parameter name to reflect that. The Coarse terminlogy comes from the dwmac register naming, update the documentation to better explain what the parameter is about. With this change, the parameter can now be adjusted using: devlink dev param set name phc_coarse_adj value true cmode runtime Signed-off-by: Maxime Chevallier Link: https://patch.msgid.link/20251030182454.182406-1-maxime.chevallier@bootlin.com Signed-off-by: Jakub Kicinski --- Documentation/networking/devlink/stmmac.rst | 21 +++++++++++++------ .../net/ethernet/stmicro/stmmac/stmmac_main.c | 2 +- 2 files changed, 16 insertions(+), 7 deletions(-) diff --git a/Documentation/networking/devlink/stmmac.rst b/Documentation/networking/devlink/stmmac.rst index e8e33d1c7baf..47e3ff10bc08 100644 --- a/Documentation/networking/devlink/stmmac.rst +++ b/Documentation/networking/devlink/stmmac.rst @@ -19,13 +19,22 @@ The ``stmmac`` driver implements the following driver-specific parameters. - Type - Mode - Description - * - ``ts_coarse`` + * - ``phc_coarse_adj`` - Boolean - runtime - - Enable the Coarse timestamping mode. In Coarse mode, the ptp clock is - expected to be updated through an external PPS input, but the subsecond - increment used for timestamping is set to 1/ptp_clock_rate. In Fine mode - (i.e. Coarse mode == false), the ptp clock frequency is adjusted more - frequently, but the subsecond increment is set to 2/ptp_clock_rate. + - Enable the Coarse timestamping mode, as defined in the DWMAC TRM. + A detailed explanation of this timestamping mode can be found in the + Socfpga Functionnal Description [1]. + + In Coarse mode, the ptp clock is expected to be fed by a high-precision + clock that is externally adjusted, and the subsecond increment used for + timestamping is set to 1/ptp_clock_rate. + + In Fine mode (i.e. Coarse mode == false), the ptp clock frequency is + continuously adjusted, but the subsecond increment is set to + 2/ptp_clock_rate. + Coarse mode is suitable for PTP Grand Master operation. If unsure, leave the parameter to False. + + [1] https://www.intel.com/content/www/us/en/docs/programmable/683126/21-2/functional-description-of-the-emac.html diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index 1e69c1a7dc6c..c2a783c8022d 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -7445,7 +7445,7 @@ static int stmmac_dl_ts_coarse_get(struct devlink *dl, u32 id, } static const struct devlink_param stmmac_devlink_params[] = { - DEVLINK_PARAM_DRIVER(STMMAC_DEVLINK_PARAM_ID_TS_COARSE, "ts_coarse", + DEVLINK_PARAM_DRIVER(STMMAC_DEVLINK_PARAM_ID_TS_COARSE, "phc_coarse_adj", DEVLINK_PARAM_TYPE_BOOL, BIT(DEVLINK_PARAM_CMODE_RUNTIME), stmmac_dl_ts_coarse_get, From 030b8d5878e87ba9b91c48727135dba6e0f0fc11 Mon Sep 17 00:00:00 2001 From: Ping-Ke Shih Date: Sat, 1 Nov 2025 21:19:55 +0200 Subject: [PATCH 0443/1381] wifi: rtw89: 8832cu: turn off TX partial mode The TX partial mode in USB devices will cause timeout to wait for payload, causing SER 0x999 and disconnection. Turn off this mode according to design suggestion. rtw89_8852cu 2-4:1.0: FW status = 0xee001108 rtw89_8852cu 2-4:1.0: FW BADADDR = 0x18605fc8 rtw89_8852cu 2-4:1.0: FW EPC/RA = 0x0 rtw89_8852cu 2-4:1.0: FW MISC = 0x1010000 rtw89_8852cu 2-4:1.0: R_AX_HALT_C2H = 0x999 rtw89_8852cu 2-4:1.0: R_AX_SER_DBG_INFO = 0x71020010 rtw89_8852cu 2-4:1.0: [ERR]fw PC = 0x2013f55e rtw89_8852cu 2-4:1.0: [ERR]fw PC = 0x2013f55a rtw89_8852cu 2-4:1.0: [ERR]fw PC = 0x2013f55a rtw89_8852cu 2-4:1.0: [ERR]fw PC = 0x2013f55e rtw89_8852cu 2-4:1.0: [ERR]fw PC = 0x2013f55e rtw89_8852cu 2-4:1.0: [ERR]fw PC = 0x2013f55e rtw89_8852cu 2-4:1.0: [ERR]fw PC = 0x2013f55a rtw89_8852cu 2-4:1.0: [ERR]fw PC = 0x2013f55e rtw89_8852cu 2-4:1.0: [ERR]fw PC = 0x2013f55a rtw89_8852cu 2-4:1.0: [ERR]fw PC = 0x2013f55e rtw89_8852cu 2-4:1.0: [ERR]fw PC = 0x2013f554 rtw89_8852cu 2-4:1.0: [ERR]fw PC = 0x2013f556 rtw89_8852cu 2-4:1.0: [ERR]fw PC = 0x2013f55a rtw89_8852cu 2-4:1.0: [ERR]fw PC = 0x2013f55a rtw89_8852cu 2-4:1.0: [ERR]fw PC = 0x2013f55e rtw89_8852cu 2-4:1.0: ---> err=0x999 rtw89_8852cu 2-4:1.0: R_AX_SER_DBG_INFO =0x71020010 rtw89_8852cu 2-4:1.0: R_AX_SER_DBG_INFO =0x71020010 rtw89_8852cu 2-4:1.0: DBG Counter 1 (R_AX_DRV_FW_HSK_4)=0x00000000 rtw89_8852cu 2-4:1.0: DBG Counter 2 (R_AX_DRV_FW_HSK_5)=0x00000000 rtw89_8852cu 2-4:1.0: R_AX_DMAC_ERR_ISR=0x00000000 rtw89_8852cu 2-4:1.0: R_AX_DMAC_ERR_IMR=0x00000000 rtw89_8852cu 2-4:1.0: R_AX_CMAC_ERR_ISR [0]=0x00000000 rtw89_8852cu 2-4:1.0: R_AX_CMAC_FUNC_EN [0]=0xf000803f rtw89_8852cu 2-4:1.0: R_AX_CK_EN [0]=0xffffffff rtw89_8852cu 2-4:1.0: R_AX_CMAC_ERR_IMR [0]=0x00000000 rtw89_8852cu 2-4:1.0: [CMAC] : CMAC1 not enabled Signed-off-by: Ping-Ke Shih Signed-off-by: Bitterblue Smith Link: https://patch.msgid.link/3904dd9f-2178-41e5-95c2-7a9f6268e935@gmail.com --- drivers/net/wireless/realtek/rtw89/mac.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/realtek/rtw89/mac.c b/drivers/net/wireless/realtek/rtw89/mac.c index d837513f4e92..df040b4e8855 100644 --- a/drivers/net/wireless/realtek/rtw89/mac.c +++ b/drivers/net/wireless/realtek/rtw89/mac.c @@ -2327,7 +2327,8 @@ static int sec_eng_init_ax(struct rtw89_dev *rtwdev) if (chip->chip_id == RTL8852C) val |= B_AX_UC_MGNT_DEC; if (chip->chip_id == RTL8852A || chip->chip_id == RTL8852B || - chip->chip_id == RTL8851B) + chip->chip_id == RTL8851B || + (chip->chip_id == RTL8852C && rtwdev->hci.type == RTW89_HCI_TYPE_USB)) val &= ~B_AX_TX_PARTIAL_MODE; rtw89_write32(rtwdev, R_AX_SEC_ENG_CTRL, val); From 89acd6c49322db4603a122e20780904021dc0491 Mon Sep 17 00:00:00 2001 From: Bitterblue Smith Date: Sat, 1 Nov 2025 21:20:38 +0200 Subject: [PATCH 0444/1381] wifi: rtw89: Add rtw89_core_get_ch_dma_v2() RTL8852CU, RTL8852AU, and RTL8922AU need a different mapping of TX queue to DMA channel compared to their PCI versions, so make get_ch_dma in struct rtw89_chip_ops an array and add rtw89_core_get_ch_dma_v2(). Signed-off-by: Bitterblue Smith Acked-by: Ping-Ke Shih Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/6c7b2f01-2c78-47c8-a4c4-98cd3060d7f3@gmail.com --- drivers/net/wireless/realtek/rtw89/core.c | 19 +++++++++++++++++++ drivers/net/wireless/realtek/rtw89/core.h | 5 +++-- drivers/net/wireless/realtek/rtw89/rtw8851b.c | 4 +++- drivers/net/wireless/realtek/rtw89/rtw8852a.c | 4 +++- drivers/net/wireless/realtek/rtw89/rtw8852b.c | 4 +++- .../net/wireless/realtek/rtw89/rtw8852bt.c | 4 +++- drivers/net/wireless/realtek/rtw89/rtw8852c.c | 4 +++- drivers/net/wireless/realtek/rtw89/rtw8922a.c | 4 +++- 8 files changed, 40 insertions(+), 8 deletions(-) diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c index 1b5a40e9821c..1f92be300ea7 100644 --- a/drivers/net/wireless/realtek/rtw89/core.c +++ b/drivers/net/wireless/realtek/rtw89/core.c @@ -759,6 +759,25 @@ u8 rtw89_core_get_ch_dma_v1(struct rtw89_dev *rtwdev, u8 qsel) } EXPORT_SYMBOL(rtw89_core_get_ch_dma_v1); +u8 rtw89_core_get_ch_dma_v2(struct rtw89_dev *rtwdev, u8 qsel) +{ + switch (qsel) { + default: + rtw89_warn(rtwdev, "Cannot map qsel to dma v2: %d\n", qsel); + fallthrough; + case RTW89_TX_QSEL_BE_0: + case RTW89_TX_QSEL_VO_0: + return RTW89_TXCH_ACH0; + case RTW89_TX_QSEL_BK_0: + case RTW89_TX_QSEL_VI_0: + return RTW89_TXCH_ACH2; + case RTW89_TX_QSEL_B0_MGMT: + case RTW89_TX_QSEL_B0_HI: + return RTW89_TXCH_CH8; + } +} +EXPORT_SYMBOL(rtw89_core_get_ch_dma_v2); + static void rtw89_core_tx_update_mgmt_info(struct rtw89_dev *rtwdev, struct rtw89_core_tx_request *tx_req) diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h index f8b443894db9..c7b48bd000c7 100644 --- a/drivers/net/wireless/realtek/rtw89/core.h +++ b/drivers/net/wireless/realtek/rtw89/core.h @@ -3764,7 +3764,7 @@ struct rtw89_chip_ops { void (*fill_txdesc_fwcmd)(struct rtw89_dev *rtwdev, struct rtw89_tx_desc_info *desc_info, void *txdesc); - u8 (*get_ch_dma)(struct rtw89_dev *rtwdev, u8 qsel); + u8 (*get_ch_dma[RTW89_HCI_TYPE_NUM])(struct rtw89_dev *rtwdev, u8 qsel); int (*cfg_ctrl_path)(struct rtw89_dev *rtwdev, bool wl); int (*mac_cfg_gnt)(struct rtw89_dev *rtwdev, const struct rtw89_mac_ax_coex_gnt *gnt_cfg); @@ -7253,7 +7253,7 @@ u8 rtw89_chip_get_ch_dma(struct rtw89_dev *rtwdev, u8 qsel) { const struct rtw89_chip_info *chip = rtwdev->chip; - return chip->ops->get_ch_dma(rtwdev, qsel); + return chip->ops->get_ch_dma[rtwdev->hci.type](rtwdev, qsel); } static inline @@ -7507,6 +7507,7 @@ void rtw89_core_fill_txdesc_fwcmd_v2(struct rtw89_dev *rtwdev, void *txdesc); u8 rtw89_core_get_ch_dma(struct rtw89_dev *rtwdev, u8 qsel); u8 rtw89_core_get_ch_dma_v1(struct rtw89_dev *rtwdev, u8 qsel); +u8 rtw89_core_get_ch_dma_v2(struct rtw89_dev *rtwdev, u8 qsel); void rtw89_core_rx(struct rtw89_dev *rtwdev, struct rtw89_rx_desc_info *desc_info, struct sk_buff *skb); diff --git a/drivers/net/wireless/realtek/rtw89/rtw8851b.c b/drivers/net/wireless/realtek/rtw89/rtw8851b.c index edcbda124916..2019f6022cbb 100644 --- a/drivers/net/wireless/realtek/rtw89/rtw8851b.c +++ b/drivers/net/wireless/realtek/rtw89/rtw8851b.c @@ -2537,7 +2537,9 @@ static const struct rtw89_chip_ops rtw8851b_chip_ops = { .query_rxdesc = rtw89_core_query_rxdesc, .fill_txdesc = rtw89_core_fill_txdesc, .fill_txdesc_fwcmd = rtw89_core_fill_txdesc, - .get_ch_dma = rtw89_core_get_ch_dma, + .get_ch_dma = {rtw89_core_get_ch_dma, + rtw89_core_get_ch_dma, + NULL,}, .cfg_ctrl_path = rtw89_mac_cfg_ctrl_path, .mac_cfg_gnt = rtw89_mac_cfg_gnt, .stop_sch_tx = rtw89_mac_stop_sch_tx, diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852a.c b/drivers/net/wireless/realtek/rtw89/rtw8852a.c index 232f4c1bee1b..fa347dbebf9a 100644 --- a/drivers/net/wireless/realtek/rtw89/rtw8852a.c +++ b/drivers/net/wireless/realtek/rtw89/rtw8852a.c @@ -2178,7 +2178,9 @@ static const struct rtw89_chip_ops rtw8852a_chip_ops = { .query_rxdesc = rtw89_core_query_rxdesc, .fill_txdesc = rtw89_core_fill_txdesc, .fill_txdesc_fwcmd = rtw89_core_fill_txdesc, - .get_ch_dma = rtw89_core_get_ch_dma, + .get_ch_dma = {rtw89_core_get_ch_dma, + rtw89_core_get_ch_dma_v2, + NULL,}, .cfg_ctrl_path = rtw89_mac_cfg_ctrl_path, .mac_cfg_gnt = rtw89_mac_cfg_gnt, .stop_sch_tx = rtw89_mac_stop_sch_tx, diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852b.c b/drivers/net/wireless/realtek/rtw89/rtw8852b.c index 0777e336aaa1..38cd151f8c3f 100644 --- a/drivers/net/wireless/realtek/rtw89/rtw8852b.c +++ b/drivers/net/wireless/realtek/rtw89/rtw8852b.c @@ -842,7 +842,9 @@ static const struct rtw89_chip_ops rtw8852b_chip_ops = { .query_rxdesc = rtw89_core_query_rxdesc, .fill_txdesc = rtw89_core_fill_txdesc, .fill_txdesc_fwcmd = rtw89_core_fill_txdesc, - .get_ch_dma = rtw89_core_get_ch_dma, + .get_ch_dma = {rtw89_core_get_ch_dma, + rtw89_core_get_ch_dma, + NULL,}, .cfg_ctrl_path = rtw89_mac_cfg_ctrl_path, .mac_cfg_gnt = rtw89_mac_cfg_gnt, .stop_sch_tx = rtw89_mac_stop_sch_tx, diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852bt.c b/drivers/net/wireless/realtek/rtw89/rtw8852bt.c index b3a79ebc7e75..15ba780492d6 100644 --- a/drivers/net/wireless/realtek/rtw89/rtw8852bt.c +++ b/drivers/net/wireless/realtek/rtw89/rtw8852bt.c @@ -708,7 +708,9 @@ static const struct rtw89_chip_ops rtw8852bt_chip_ops = { .query_rxdesc = rtw89_core_query_rxdesc, .fill_txdesc = rtw89_core_fill_txdesc, .fill_txdesc_fwcmd = rtw89_core_fill_txdesc, - .get_ch_dma = rtw89_core_get_ch_dma, + .get_ch_dma = {rtw89_core_get_ch_dma, + NULL, + NULL,}, .cfg_ctrl_path = rtw89_mac_cfg_ctrl_path, .mac_cfg_gnt = rtw89_mac_cfg_gnt, .stop_sch_tx = rtw89_mac_stop_sch_tx, diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852c.c b/drivers/net/wireless/realtek/rtw89/rtw8852c.c index 440801d63343..ee1915307376 100644 --- a/drivers/net/wireless/realtek/rtw89/rtw8852c.c +++ b/drivers/net/wireless/realtek/rtw89/rtw8852c.c @@ -2962,7 +2962,9 @@ static const struct rtw89_chip_ops rtw8852c_chip_ops = { .query_rxdesc = rtw89_core_query_rxdesc, .fill_txdesc = rtw89_core_fill_txdesc_v1, .fill_txdesc_fwcmd = rtw89_core_fill_txdesc_fwcmd_v1, - .get_ch_dma = rtw89_core_get_ch_dma, + .get_ch_dma = {rtw89_core_get_ch_dma, + rtw89_core_get_ch_dma_v2, + NULL,}, .cfg_ctrl_path = rtw89_mac_cfg_ctrl_path_v1, .mac_cfg_gnt = rtw89_mac_cfg_gnt_v1, .stop_sch_tx = rtw89_mac_stop_sch_tx_v1, diff --git a/drivers/net/wireless/realtek/rtw89/rtw8922a.c b/drivers/net/wireless/realtek/rtw89/rtw8922a.c index 6aa19ad259ac..61f3a0a3f440 100644 --- a/drivers/net/wireless/realtek/rtw89/rtw8922a.c +++ b/drivers/net/wireless/realtek/rtw89/rtw8922a.c @@ -2821,7 +2821,9 @@ static const struct rtw89_chip_ops rtw8922a_chip_ops = { .query_rxdesc = rtw89_core_query_rxdesc_v2, .fill_txdesc = rtw89_core_fill_txdesc_v2, .fill_txdesc_fwcmd = rtw89_core_fill_txdesc_fwcmd_v2, - .get_ch_dma = rtw89_core_get_ch_dma, + .get_ch_dma = {rtw89_core_get_ch_dma, + rtw89_core_get_ch_dma_v2, + NULL,}, .cfg_ctrl_path = rtw89_mac_cfg_ctrl_path_v2, .mac_cfg_gnt = rtw89_mac_cfg_gnt_v2, .stop_sch_tx = rtw89_mac_stop_sch_tx_v2, From 994944aa58f84a4798664408a95606bd9f8a40f2 Mon Sep 17 00:00:00 2001 From: Bitterblue Smith Date: Sat, 1 Nov 2025 21:22:19 +0200 Subject: [PATCH 0445/1381] wifi: rtw89: usb: Move bulk out map to new struct rtw89_usb_info RTL8852AU, RTL8852CU, and RTL8922AU will need a different TX channel to bulk out endpoint mapping, so create a new struct rtw89_usb_info and move the mapping there. Initialise it in each chip's driver. Struct rtw89_usb_info will also hold some registers which are located at different offsets in RTL8852CU compared to the other wifi 6 chips. Signed-off-by: Bitterblue Smith Acked-by: Ping-Ke Shih Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/de11cfae-1dc0-4514-95b2-02b1bbfa92be@gmail.com --- drivers/net/wireless/realtek/rtw89/core.h | 2 ++ .../net/wireless/realtek/rtw89/rtw8851bu.c | 15 +++++++++++ .../net/wireless/realtek/rtw89/rtw8852bu.c | 15 +++++++++++ drivers/net/wireless/realtek/rtw89/usb.c | 25 +++---------------- drivers/net/wireless/realtek/rtw89/usb.h | 5 ++++ 5 files changed, 40 insertions(+), 22 deletions(-) diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h index c7b48bd000c7..97cb0e5f0898 100644 --- a/drivers/net/wireless/realtek/rtw89/core.h +++ b/drivers/net/wireless/realtek/rtw89/core.h @@ -15,6 +15,7 @@ struct rtw89_dev; struct rtw89_pci_info; +struct rtw89_usb_info; struct rtw89_mac_gen_def; struct rtw89_phy_gen_def; struct rtw89_fw_blacklist; @@ -4514,6 +4515,7 @@ struct rtw89_chip_variant { union rtw89_bus_info { const struct rtw89_pci_info *pci; + const struct rtw89_usb_info *usb; }; struct rtw89_driver_info { diff --git a/drivers/net/wireless/realtek/rtw89/rtw8851bu.c b/drivers/net/wireless/realtek/rtw89/rtw8851bu.c index 04e1ab13b753..c39acbec12a9 100644 --- a/drivers/net/wireless/realtek/rtw89/rtw8851bu.c +++ b/drivers/net/wireless/realtek/rtw89/rtw8851bu.c @@ -7,10 +7,25 @@ #include "rtw8851b.h" #include "usb.h" +static const struct rtw89_usb_info rtw8851b_usb_info = { + .bulkout_id = { + [RTW89_DMA_ACH0] = 3, + [RTW89_DMA_ACH1] = 4, + [RTW89_DMA_ACH2] = 5, + [RTW89_DMA_ACH3] = 6, + [RTW89_DMA_B0MG] = 0, + [RTW89_DMA_B0HI] = 1, + [RTW89_DMA_H2C] = 2, + }, +}; + static const struct rtw89_driver_info rtw89_8851bu_info = { .chip = &rtw8851b_chip_info, .variant = NULL, .quirks = NULL, + .bus = { + .usb = &rtw8851b_usb_info, + } }; static const struct usb_device_id rtw_8851bu_id_table[] = { diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852bu.c b/drivers/net/wireless/realtek/rtw89/rtw8852bu.c index 0694272f7ffa..28e69473918e 100644 --- a/drivers/net/wireless/realtek/rtw89/rtw8852bu.c +++ b/drivers/net/wireless/realtek/rtw89/rtw8852bu.c @@ -7,10 +7,25 @@ #include "rtw8852b.h" #include "usb.h" +static const struct rtw89_usb_info rtw8852b_usb_info = { + .bulkout_id = { + [RTW89_DMA_ACH0] = 3, + [RTW89_DMA_ACH1] = 4, + [RTW89_DMA_ACH2] = 5, + [RTW89_DMA_ACH3] = 6, + [RTW89_DMA_B0MG] = 0, + [RTW89_DMA_B0HI] = 1, + [RTW89_DMA_H2C] = 2, + }, +}; + static const struct rtw89_driver_info rtw89_8852bu_info = { .chip = &rtw8852b_chip_info, .variant = NULL, .quirks = NULL, + .bus = { + .usb = &rtw8852b_usb_info, + } }; static const struct usb_device_id rtw_8852bu_id_table[] = { diff --git a/drivers/net/wireless/realtek/rtw89/usb.c b/drivers/net/wireless/realtek/rtw89/usb.c index afbd1fb79021..8c0129d54b6c 100644 --- a/drivers/net/wireless/realtek/rtw89/usb.c +++ b/drivers/net/wireless/realtek/rtw89/usb.c @@ -167,27 +167,6 @@ rtw89_usb_ops_check_and_reclaim_tx_resource(struct rtw89_dev *rtwdev, return 42; /* TODO some kind of calculation? */ } -static u8 rtw89_usb_get_bulkout_id(u8 ch_dma) -{ - switch (ch_dma) { - case RTW89_DMA_ACH0: - return 3; - case RTW89_DMA_ACH1: - return 4; - case RTW89_DMA_ACH2: - return 5; - case RTW89_DMA_ACH3: - return 6; - default: - case RTW89_DMA_B0MG: - return 0; - case RTW89_DMA_B0HI: - return 1; - case RTW89_DMA_H2C: - return 2; - } -} - static void rtw89_usb_write_port_complete(struct urb *urb) { struct rtw89_usb_tx_ctrl_block *txcb = urb->context; @@ -249,9 +228,10 @@ static int rtw89_usb_write_port(struct rtw89_dev *rtwdev, u8 ch_dma, void *data, int len, void *context) { struct rtw89_usb *rtwusb = rtw89_usb_priv(rtwdev); + const struct rtw89_usb_info *info = rtwusb->info; struct usb_device *usbd = rtwusb->udev; struct urb *urb; - u8 bulkout_id = rtw89_usb_get_bulkout_id(ch_dma); + u8 bulkout_id = info->bulkout_id[ch_dma]; unsigned int pipe; int ret; @@ -949,6 +929,7 @@ int rtw89_usb_probe(struct usb_interface *intf, rtwusb = rtw89_usb_priv(rtwdev); rtwusb->rtwdev = rtwdev; + rtwusb->info = info->bus.usb; rtwdev->hci.ops = &rtw89_usb_ops; rtwdev->hci.type = RTW89_HCI_TYPE_USB; diff --git a/drivers/net/wireless/realtek/rtw89/usb.h b/drivers/net/wireless/realtek/rtw89/usb.h index c1b4bfa20979..a27c698a424c 100644 --- a/drivers/net/wireless/realtek/rtw89/usb.h +++ b/drivers/net/wireless/realtek/rtw89/usb.h @@ -20,6 +20,10 @@ #define RTW89_MAX_ENDPOINT_NUM 9 #define RTW89_MAX_BULKOUT_NUM 7 +struct rtw89_usb_info { + u8 bulkout_id[RTW89_DMA_CH_NUM]; +}; + struct rtw89_usb_rx_ctrl_block { struct rtw89_dev *rtwdev; struct urb *rx_urb; @@ -35,6 +39,7 @@ struct rtw89_usb_tx_ctrl_block { struct rtw89_usb { struct rtw89_dev *rtwdev; struct usb_device *udev; + const struct rtw89_usb_info *info; __le32 *vendor_req_buf; From 7697701d6c0356cecda03be516023cb527ddb007 Mon Sep 17 00:00:00 2001 From: Bitterblue Smith Date: Sat, 1 Nov 2025 21:22:56 +0200 Subject: [PATCH 0446/1381] wifi: rtw89: usb: Prepare rtw89_usb_ops_mac_pre_init() for RTL8852CU The registers used in rtw89_usb_ops_mac_pre_init() are located at different offsets in RTL8852CU, so move them to struct rtw89_usb_info, which is filled in each chip's driver. Signed-off-by: Bitterblue Smith Acked-by: Ping-Ke Shih Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/36b823c9-2cb5-4191-8492-2a291f9dc1db@gmail.com --- drivers/net/wireless/realtek/rtw89/rtw8851bu.c | 4 ++++ drivers/net/wireless/realtek/rtw89/rtw8852bu.c | 4 ++++ drivers/net/wireless/realtek/rtw89/usb.c | 13 ++++++++----- drivers/net/wireless/realtek/rtw89/usb.h | 3 +++ 4 files changed, 19 insertions(+), 5 deletions(-) diff --git a/drivers/net/wireless/realtek/rtw89/rtw8851bu.c b/drivers/net/wireless/realtek/rtw89/rtw8851bu.c index c39acbec12a9..9cc682872d24 100644 --- a/drivers/net/wireless/realtek/rtw89/rtw8851bu.c +++ b/drivers/net/wireless/realtek/rtw89/rtw8851bu.c @@ -5,9 +5,13 @@ #include #include #include "rtw8851b.h" +#include "reg.h" #include "usb.h" static const struct rtw89_usb_info rtw8851b_usb_info = { + .usb_host_request_2 = R_AX_USB_HOST_REQUEST_2, + .usb_wlan0_1 = R_AX_USB_WLAN0_1, + .hci_func_en = R_AX_HCI_FUNC_EN, .bulkout_id = { [RTW89_DMA_ACH0] = 3, [RTW89_DMA_ACH1] = 4, diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852bu.c b/drivers/net/wireless/realtek/rtw89/rtw8852bu.c index 28e69473918e..b913cae95e3b 100644 --- a/drivers/net/wireless/realtek/rtw89/rtw8852bu.c +++ b/drivers/net/wireless/realtek/rtw89/rtw8852bu.c @@ -5,9 +5,13 @@ #include #include #include "rtw8852b.h" +#include "reg.h" #include "usb.h" static const struct rtw89_usb_info rtw8852b_usb_info = { + .usb_host_request_2 = R_AX_USB_HOST_REQUEST_2, + .usb_wlan0_1 = R_AX_USB_WLAN0_1, + .hci_func_en = R_AX_HCI_FUNC_EN, .bulkout_id = { [RTW89_DMA_ACH0] = 3, [RTW89_DMA_ACH1] = 4, diff --git a/drivers/net/wireless/realtek/rtw89/usb.c b/drivers/net/wireless/realtek/rtw89/usb.c index 8c0129d54b6c..ea1d8726ec3f 100644 --- a/drivers/net/wireless/realtek/rtw89/usb.c +++ b/drivers/net/wireless/realtek/rtw89/usb.c @@ -678,20 +678,23 @@ static int rtw89_usb_ops_deinit(struct rtw89_dev *rtwdev) static int rtw89_usb_ops_mac_pre_init(struct rtw89_dev *rtwdev) { + struct rtw89_usb *rtwusb = rtw89_usb_priv(rtwdev); + const struct rtw89_usb_info *info = rtwusb->info; u32 val32; - rtw89_write32_set(rtwdev, R_AX_USB_HOST_REQUEST_2, B_AX_R_USBIO_MODE); + rtw89_write32_set(rtwdev, info->usb_host_request_2, + B_AX_R_USBIO_MODE); /* fix USB IO hang suggest by chihhanli@realtek.com */ - rtw89_write32_clr(rtwdev, R_AX_USB_WLAN0_1, + rtw89_write32_clr(rtwdev, info->usb_wlan0_1, B_AX_USBRX_RST | B_AX_USBTX_RST); - val32 = rtw89_read32(rtwdev, R_AX_HCI_FUNC_EN); + val32 = rtw89_read32(rtwdev, info->hci_func_en); val32 &= ~(B_AX_HCI_RXDMA_EN | B_AX_HCI_TXDMA_EN); - rtw89_write32(rtwdev, R_AX_HCI_FUNC_EN, val32); + rtw89_write32(rtwdev, info->hci_func_en, val32); val32 |= B_AX_HCI_RXDMA_EN | B_AX_HCI_TXDMA_EN; - rtw89_write32(rtwdev, R_AX_HCI_FUNC_EN, val32); + rtw89_write32(rtwdev, info->hci_func_en, val32); /* fix USB TRX hang suggest by chihhanli@realtek.com */ return 0; diff --git a/drivers/net/wireless/realtek/rtw89/usb.h b/drivers/net/wireless/realtek/rtw89/usb.h index a27c698a424c..be3a9fbe42c0 100644 --- a/drivers/net/wireless/realtek/rtw89/usb.h +++ b/drivers/net/wireless/realtek/rtw89/usb.h @@ -21,6 +21,9 @@ #define RTW89_MAX_BULKOUT_NUM 7 struct rtw89_usb_info { + u32 usb_host_request_2; + u32 usb_wlan0_1; + u32 hci_func_en; u8 bulkout_id[RTW89_DMA_CH_NUM]; }; From 199afd3af12c81f663734d8fe290197288618ada Mon Sep 17 00:00:00 2001 From: Bitterblue Smith Date: Sat, 1 Nov 2025 21:23:33 +0200 Subject: [PATCH 0447/1381] wifi: rtw89: usb: Prepare rtw89_usb_ops_mac_post_init() for RTL8852CU The registers used in rtw89_usb_ops_mac_post_init() are located at different offsets in RTL8852CU, so move them to struct rtw89_usb_info, which is filled in each chip's driver. Signed-off-by: Bitterblue Smith Acked-by: Ping-Ke Shih Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/c12b621a-037c-4870-ac12-9795ddde6b53@gmail.com --- drivers/net/wireless/realtek/rtw89/rtw8851bu.c | 3 +++ drivers/net/wireless/realtek/rtw89/rtw8852bu.c | 3 +++ drivers/net/wireless/realtek/rtw89/usb.c | 7 ++++--- drivers/net/wireless/realtek/rtw89/usb.h | 3 +++ 4 files changed, 13 insertions(+), 3 deletions(-) diff --git a/drivers/net/wireless/realtek/rtw89/rtw8851bu.c b/drivers/net/wireless/realtek/rtw89/rtw8851bu.c index 9cc682872d24..b8799a37add3 100644 --- a/drivers/net/wireless/realtek/rtw89/rtw8851bu.c +++ b/drivers/net/wireless/realtek/rtw89/rtw8851bu.c @@ -12,6 +12,9 @@ static const struct rtw89_usb_info rtw8851b_usb_info = { .usb_host_request_2 = R_AX_USB_HOST_REQUEST_2, .usb_wlan0_1 = R_AX_USB_WLAN0_1, .hci_func_en = R_AX_HCI_FUNC_EN, + .usb3_mac_npi_config_intf_0 = R_AX_USB3_MAC_NPI_CONFIG_INTF_0, + .usb_endpoint_0 = R_AX_USB_ENDPOINT_0, + .usb_endpoint_2 = R_AX_USB_ENDPOINT_2, .bulkout_id = { [RTW89_DMA_ACH0] = 3, [RTW89_DMA_ACH1] = 4, diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852bu.c b/drivers/net/wireless/realtek/rtw89/rtw8852bu.c index b913cae95e3b..c8a644403341 100644 --- a/drivers/net/wireless/realtek/rtw89/rtw8852bu.c +++ b/drivers/net/wireless/realtek/rtw89/rtw8852bu.c @@ -12,6 +12,9 @@ static const struct rtw89_usb_info rtw8852b_usb_info = { .usb_host_request_2 = R_AX_USB_HOST_REQUEST_2, .usb_wlan0_1 = R_AX_USB_WLAN0_1, .hci_func_en = R_AX_HCI_FUNC_EN, + .usb3_mac_npi_config_intf_0 = R_AX_USB3_MAC_NPI_CONFIG_INTF_0, + .usb_endpoint_0 = R_AX_USB_ENDPOINT_0, + .usb_endpoint_2 = R_AX_USB_ENDPOINT_2, .bulkout_id = { [RTW89_DMA_ACH0] = 3, [RTW89_DMA_ACH1] = 4, diff --git a/drivers/net/wireless/realtek/rtw89/usb.c b/drivers/net/wireless/realtek/rtw89/usb.c index ea1d8726ec3f..4c47071b30fd 100644 --- a/drivers/net/wireless/realtek/rtw89/usb.c +++ b/drivers/net/wireless/realtek/rtw89/usb.c @@ -708,10 +708,11 @@ static int rtw89_usb_ops_mac_pre_deinit(struct rtw89_dev *rtwdev) static int rtw89_usb_ops_mac_post_init(struct rtw89_dev *rtwdev) { struct rtw89_usb *rtwusb = rtw89_usb_priv(rtwdev); + const struct rtw89_usb_info *info = rtwusb->info; enum usb_device_speed speed; u32 ep; - rtw89_write32_clr(rtwdev, R_AX_USB3_MAC_NPI_CONFIG_INTF_0, + rtw89_write32_clr(rtwdev, info->usb3_mac_npi_config_intf_0, B_AX_SSPHY_LFPS_FILTER); speed = rtwusb->udev->speed; @@ -727,9 +728,9 @@ static int rtw89_usb_ops_mac_post_init(struct rtw89_dev *rtwdev) if (ep == 8) continue; - rtw89_write8_mask(rtwdev, R_AX_USB_ENDPOINT_0, + rtw89_write8_mask(rtwdev, info->usb_endpoint_0, B_AX_EP_IDX, ep); - rtw89_write8(rtwdev, R_AX_USB_ENDPOINT_2 + 1, NUMP); + rtw89_write8(rtwdev, info->usb_endpoint_2 + 1, NUMP); } return 0; diff --git a/drivers/net/wireless/realtek/rtw89/usb.h b/drivers/net/wireless/realtek/rtw89/usb.h index be3a9fbe42c0..32cdbf5128e4 100644 --- a/drivers/net/wireless/realtek/rtw89/usb.h +++ b/drivers/net/wireless/realtek/rtw89/usb.h @@ -24,6 +24,9 @@ struct rtw89_usb_info { u32 usb_host_request_2; u32 usb_wlan0_1; u32 hci_func_en; + u32 usb3_mac_npi_config_intf_0; + u32 usb_endpoint_0; + u32 usb_endpoint_2; u8 bulkout_id[RTW89_DMA_CH_NUM]; }; From 32e0381e86031d0289b98e82a7a872ed1698676d Mon Sep 17 00:00:00 2001 From: Bitterblue Smith Date: Sat, 1 Nov 2025 21:24:05 +0200 Subject: [PATCH 0448/1381] wifi: rtw89: Fix rtw89_mac_dmac_func_pre_en_ax() for USB/SDIO Set the DMA mode according to the interface type. Signed-off-by: Bitterblue Smith Acked-by: Ping-Ke Shih Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/d2250e72-0aa1-422c-8f7f-9aeb283ca376@gmail.com --- drivers/net/wireless/realtek/rtw89/mac.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/drivers/net/wireless/realtek/rtw89/mac.c b/drivers/net/wireless/realtek/rtw89/mac.c index df040b4e8855..b5309ae3fd57 100644 --- a/drivers/net/wireless/realtek/rtw89/mac.c +++ b/drivers/net/wireless/realtek/rtw89/mac.c @@ -3984,8 +3984,15 @@ static void rtw89_mac_dmac_func_pre_en_ax(struct rtw89_dev *rtwdev) val = rtw89_read32(rtwdev, R_AX_HAXI_INIT_CFG1); val &= ~(B_AX_DMA_MODE_MASK | B_AX_STOP_AXI_MST); - val |= FIELD_PREP(B_AX_DMA_MODE_MASK, DMA_MOD_PCIE_1B) | - B_AX_TXHCI_EN_V1 | B_AX_RXHCI_EN_V1; + val |= B_AX_TXHCI_EN_V1 | B_AX_RXHCI_EN_V1; + + if (rtwdev->hci.type == RTW89_HCI_TYPE_PCIE) + val |= FIELD_PREP(B_AX_DMA_MODE_MASK, DMA_MOD_PCIE_1B); + else if (rtwdev->hci.type == RTW89_HCI_TYPE_USB) + val |= FIELD_PREP(B_AX_DMA_MODE_MASK, DMA_MOD_USB); + else + val |= FIELD_PREP(B_AX_DMA_MODE_MASK, DMA_MOD_SDIO); + rtw89_write32(rtwdev, R_AX_HAXI_INIT_CFG1, val); rtw89_write32_clr(rtwdev, R_AX_HAXI_DMA_STOP1, From 97259766b18e2a22f2d5258d4ae151260e9165b0 Mon Sep 17 00:00:00 2001 From: Bitterblue Smith Date: Sat, 1 Nov 2025 21:24:39 +0200 Subject: [PATCH 0449/1381] wifi: rtw89: 8852c: Fix rtw8852c_pwr_{on,off}_func() for USB There are a few differences in the power on/off functions between PCIE and USB. The changes in the power off function in particular are needed for the RTL8832CU to be able to power on again after it's powered off. While the RTL8832CU appears to work without the changes in the power on function, it's probably best to implement them, in case they are needed in some situations. Signed-off-by: Bitterblue Smith Acked-by: Ping-Ke Shih Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/598dec66-b5cc-435a-bcf6-fa66577f8cfc@gmail.com --- drivers/net/wireless/realtek/rtw89/rtw8852c.c | 31 ++++++++++++++----- 1 file changed, 24 insertions(+), 7 deletions(-) diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852c.c b/drivers/net/wireless/realtek/rtw89/rtw8852c.c index ee1915307376..6aedbd03e3d4 100644 --- a/drivers/net/wireless/realtek/rtw89/rtw8852c.c +++ b/drivers/net/wireless/realtek/rtw89/rtw8852c.c @@ -214,7 +214,8 @@ static int rtw8852c_pwr_on_func(struct rtw89_dev *rtwdev) int ret; val32 = rtw89_read32_mask(rtwdev, R_AX_SYS_STATUS1, B_AX_PAD_HCI_SEL_V2_MASK); - if (val32 == MAC_AX_HCI_SEL_PCIE_USB) + if (val32 == MAC_AX_HCI_SEL_PCIE_USB || + rtwdev->hci.type == RTW89_HCI_TYPE_USB) rtw89_write32_set(rtwdev, R_AX_LDO_AON_CTRL0, B_AX_PD_REGU_L); rtw89_write32_clr(rtwdev, R_AX_SYS_PW_CTRL, B_AX_AFSM_WLSUS_EN | @@ -246,7 +247,9 @@ static int rtw8852c_pwr_on_func(struct rtw89_dev *rtwdev) rtw89_write8_clr(rtwdev, R_AX_PLATFORM_ENABLE, B_AX_PLATFORM_EN); rtw89_write8_set(rtwdev, R_AX_PLATFORM_ENABLE, B_AX_PLATFORM_EN); - rtw89_write32_clr(rtwdev, R_AX_SYS_SDIO_CTRL, B_AX_PCIE_CALIB_EN_V1); + + if (rtwdev->hci.type == RTW89_HCI_TYPE_PCIE) + rtw89_write32_clr(rtwdev, R_AX_SYS_SDIO_CTRL, B_AX_PCIE_CALIB_EN_V1); rtw89_write32_clr(rtwdev, R_AX_SYS_ISO_CTRL_EXTEND, B_AX_CMAC1_FEN); rtw89_write32_set(rtwdev, R_AX_SYS_ISO_CTRL_EXTEND, B_AX_R_SYM_ISO_CMAC12PP); @@ -305,9 +308,11 @@ static int rtw8852c_pwr_on_func(struct rtw89_dev *rtwdev) rtw89_write32_clr(rtwdev, R_AX_SYS_ISO_CTRL, B_AX_PWC_EV2EF_B14); rtw89_write32_clr(rtwdev, R_AX_PMC_DBG_CTRL2, B_AX_SYSON_DIS_PMCR_AX_WRMSK); - rtw89_write32_set(rtwdev, R_AX_GPIO0_15_EECS_EESK_LED1_PULL_LOW_EN, - B_AX_EECS_PULL_LOW_EN | B_AX_EESK_PULL_LOW_EN | - B_AX_LED1_PULL_LOW_EN); + + if (rtwdev->hci.type == RTW89_HCI_TYPE_PCIE) + rtw89_write32_set(rtwdev, R_AX_GPIO0_15_EECS_EESK_LED1_PULL_LOW_EN, + B_AX_EECS_PULL_LOW_EN | B_AX_EESK_PULL_LOW_EN | + B_AX_LED1_PULL_LOW_EN); rtw89_write32_set(rtwdev, R_AX_DMAC_FUNC_EN, B_AX_MAC_FUNC_EN | B_AX_DMAC_FUNC_EN | B_AX_MPDU_PROC_EN | @@ -385,12 +390,24 @@ static int rtw8852c_pwr_off_func(struct rtw89_dev *rtwdev) if (ret) return ret; - rtw89_write32(rtwdev, R_AX_WLLPS_CTRL, SW_LPS_OPTION); + if (rtwdev->hci.type == RTW89_HCI_TYPE_PCIE) + rtw89_write32(rtwdev, R_AX_WLLPS_CTRL, SW_LPS_OPTION); + else if (rtwdev->hci.type == RTW89_HCI_TYPE_USB) + rtw89_write32_clr(rtwdev, R_AX_SYS_PW_CTRL, B_AX_SOP_EDSWR); + rtw89_write32_set(rtwdev, R_AX_SYS_PW_CTRL, B_AX_XTAL_OFF_A_DIE); rtw89_write32_set(rtwdev, R_AX_SYS_SWR_CTRL1, B_AX_SYM_CTRL_SPS_PWMFREQ); rtw89_write32_mask(rtwdev, R_AX_SPS_DIG_ON_CTRL0, B_AX_REG_ZCDC_H_MASK, 0x3); - rtw89_write32_set(rtwdev, R_AX_SYS_PW_CTRL, B_AX_APFM_SWLPS); + + if (rtwdev->hci.type == RTW89_HCI_TYPE_PCIE) { + rtw89_write32_set(rtwdev, R_AX_SYS_PW_CTRL, B_AX_APFM_SWLPS); + } else if (rtwdev->hci.type == RTW89_HCI_TYPE_USB) { + val32 = rtw89_read32(rtwdev, R_AX_SYS_PW_CTRL); + val32 &= ~B_AX_AFSM_PCIE_SUS_EN; + val32 |= B_AX_AFSM_WLSUS_EN; + rtw89_write32(rtwdev, R_AX_SYS_PW_CTRL, val32); + } return 0; } From a8658990846058fc9a76a38b20edc5ddf05fd39d Mon Sep 17 00:00:00 2001 From: Bitterblue Smith Date: Sat, 1 Nov 2025 21:25:41 +0200 Subject: [PATCH 0450/1381] wifi: rtw89: Add rtw8852c_dle_mem_usb{2,3} Add rtw8852c_dle_mem_usb2 and rtw8852c_dle_mem_usb3 and their various quotas and sizes in struct rtw89_mac_size_set. "dle" could be "Data Link Engine" or "Double Link Engine". These are some parameters needed for RTL8852CU. Signed-off-by: Bitterblue Smith Acked-by: Ping-Ke Shih Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/40a58644-13ce-48a4-85e2-ba4f3cbb975b@gmail.com --- drivers/net/wireless/realtek/rtw89/mac.c | 20 ++++++++++++ drivers/net/wireless/realtek/rtw89/mac.h | 10 ++++++ drivers/net/wireless/realtek/rtw89/rtw8852c.c | 31 ++++++++++++++++++- 3 files changed, 60 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/realtek/rtw89/mac.c b/drivers/net/wireless/realtek/rtw89/mac.c index b5309ae3fd57..bc39b4f7bc4f 100644 --- a/drivers/net/wireless/realtek/rtw89/mac.c +++ b/drivers/net/wireless/realtek/rtw89/mac.c @@ -1663,6 +1663,8 @@ const struct rtw89_mac_size_set rtw89_mac_size = { .wde_size7 = {RTW89_WDE_PG_64, 510, 2,}, /* DLFW */ .wde_size9 = {RTW89_WDE_PG_64, 0, 1024,}, + /* 8852C USB3.0 */ + .wde_size17 = {RTW89_WDE_PG_64, 354, 30,}, /* 8852C DLFW */ .wde_size18 = {RTW89_WDE_PG_64, 0, 2048,}, /* 8852C PCIE SCC */ @@ -1670,6 +1672,8 @@ const struct rtw89_mac_size_set rtw89_mac_size = { .wde_size23 = {RTW89_WDE_PG_64, 1022, 2,}, /* 8852B USB2.0/USB3.0 SCC */ .wde_size25 = {RTW89_WDE_PG_64, 162, 94,}, + /* 8852C USB2.0 */ + .wde_size31 = {RTW89_WDE_PG_64, 384, 0,}, /* PCIE */ .ple_size0 = {RTW89_PLE_PG_128, 1520, 16,}, .ple_size0_v1 = {RTW89_PLE_PG_128, 2688, 240, 212992,}, @@ -1681,6 +1685,8 @@ const struct rtw89_mac_size_set rtw89_mac_size = { /* DLFW */ .ple_size8 = {RTW89_PLE_PG_128, 64, 960,}, .ple_size9 = {RTW89_PLE_PG_128, 2288, 16,}, + /* 8852C USB */ + .ple_size17 = {RTW89_PLE_PG_128, 3368, 24,}, /* 8852C DLFW */ .ple_size18 = {RTW89_PLE_PG_128, 2544, 16,}, /* 8852C PCIE SCC */ @@ -1689,6 +1695,8 @@ const struct rtw89_mac_size_set rtw89_mac_size = { .ple_size32 = {RTW89_PLE_PG_128, 620, 20,}, /* 8852B USB3.0 SCC */ .ple_size33 = {RTW89_PLE_PG_128, 632, 8,}, + /* 8852C USB2.0 */ + .ple_size34 = {RTW89_PLE_PG_128, 3374, 18,}, /* PCIE 64 */ .wde_qt0 = {3792, 196, 0, 107,}, .wde_qt0_v1 = {3302, 6, 0, 20,}, @@ -1698,6 +1706,8 @@ const struct rtw89_mac_size_set rtw89_mac_size = { .wde_qt6 = {448, 48, 0, 16,}, /* 8852B PCIE SCC */ .wde_qt7 = {446, 48, 0, 16,}, + /* 8852C USB3.0 */ + .wde_qt16 = {344, 2, 0, 8,}, /* 8852C DLFW */ .wde_qt17 = {0, 0, 0, 0,}, /* 8852C PCIE SCC */ @@ -1705,6 +1715,8 @@ const struct rtw89_mac_size_set rtw89_mac_size = { .wde_qt23 = {958, 48, 0, 16,}, /* 8852B USB2.0/USB3.0 SCC */ .wde_qt25 = {152, 2, 0, 8,}, + /* 8852C USB2.0 */ + .wde_qt31 = {338, 6, 0, 40,}, .ple_qt0 = {320, 320, 32, 16, 13, 13, 292, 292, 64, 18, 1, 4, 0,}, .ple_qt1 = {320, 320, 32, 16, 1316, 1316, 1595, 1595, 1367, 1321, 1, 1307, 0,}, /* PCIE SCC */ @@ -1716,6 +1728,10 @@ const struct rtw89_mac_size_set rtw89_mac_size = { .ple_qt13 = {0, 0, 16, 48, 0, 0, 0, 0, 0, 0, 0,}, /* PCIE 64 */ .ple_qt18 = {147, 0, 16, 20, 17, 13, 89, 0, 32, 14, 8, 0,}, + /* USB 52C USB3.0 */ + .ple_qt42 = {1068, 0, 16, 48, 4, 13, 178, 0, 16, 1, 8, 16, 0,}, + /* USB 52C USB3.0 */ + .ple_qt43 = {3068, 0, 32, 48, 4, 13, 178, 0, 16, 1, 8, 16, 0,}, /* DLFW 52C */ .ple_qt44 = {0, 0, 16, 256, 0, 0, 0, 0, 0, 0, 0, 0,}, /* DLFW 52C */ @@ -1735,6 +1751,10 @@ const struct rtw89_mac_size_set rtw89_mac_size = { /* USB3.0 52B 92K */ .ple_qt74 = {286, 0, 16, 48, 4, 13, 178, 0, 32, 14, 8, 0, 0,}, .ple_qt75 = {286, 0, 32, 48, 37, 13, 211, 0, 65, 14, 24, 0, 0,}, + /* USB2.0 52C */ + .ple_qt78 = {1560, 0, 16, 48, 13, 13, 390, 0, 32, 38, 8, 16, 0,}, + /* USB2.0 52C */ + .ple_qt79 = {1560, 0, 32, 48, 1253, 13, 1630, 0, 1272, 38, 120, 1256, 0,}, /* 8852A PCIE WOW */ .ple_qt_52a_wow = {264, 0, 32, 20, 64, 13, 1005, 0, 64, 128, 120,}, /* 8852B PCIE WOW */ diff --git a/drivers/net/wireless/realtek/rtw89/mac.h b/drivers/net/wireless/realtek/rtw89/mac.h index 25fe5e5c8a97..46302f4aa3d9 100644 --- a/drivers/net/wireless/realtek/rtw89/mac.h +++ b/drivers/net/wireless/realtek/rtw89/mac.h @@ -923,10 +923,12 @@ struct rtw89_mac_size_set { const struct rtw89_dle_size wde_size6; const struct rtw89_dle_size wde_size7; const struct rtw89_dle_size wde_size9; + const struct rtw89_dle_size wde_size17; const struct rtw89_dle_size wde_size18; const struct rtw89_dle_size wde_size19; const struct rtw89_dle_size wde_size23; const struct rtw89_dle_size wde_size25; + const struct rtw89_dle_size wde_size31; const struct rtw89_dle_size ple_size0; const struct rtw89_dle_size ple_size0_v1; const struct rtw89_dle_size ple_size3_v1; @@ -934,19 +936,23 @@ struct rtw89_mac_size_set { const struct rtw89_dle_size ple_size6; const struct rtw89_dle_size ple_size8; const struct rtw89_dle_size ple_size9; + const struct rtw89_dle_size ple_size17; const struct rtw89_dle_size ple_size18; const struct rtw89_dle_size ple_size19; const struct rtw89_dle_size ple_size32; const struct rtw89_dle_size ple_size33; + const struct rtw89_dle_size ple_size34; const struct rtw89_wde_quota wde_qt0; const struct rtw89_wde_quota wde_qt0_v1; const struct rtw89_wde_quota wde_qt4; const struct rtw89_wde_quota wde_qt6; const struct rtw89_wde_quota wde_qt7; + const struct rtw89_wde_quota wde_qt16; const struct rtw89_wde_quota wde_qt17; const struct rtw89_wde_quota wde_qt18; const struct rtw89_wde_quota wde_qt23; const struct rtw89_wde_quota wde_qt25; + const struct rtw89_wde_quota wde_qt31; const struct rtw89_ple_quota ple_qt0; const struct rtw89_ple_quota ple_qt1; const struct rtw89_ple_quota ple_qt4; @@ -954,6 +960,8 @@ struct rtw89_mac_size_set { const struct rtw89_ple_quota ple_qt9; const struct rtw89_ple_quota ple_qt13; const struct rtw89_ple_quota ple_qt18; + const struct rtw89_ple_quota ple_qt42; + const struct rtw89_ple_quota ple_qt43; const struct rtw89_ple_quota ple_qt44; const struct rtw89_ple_quota ple_qt45; const struct rtw89_ple_quota ple_qt46; @@ -965,6 +973,8 @@ struct rtw89_mac_size_set { const struct rtw89_ple_quota ple_qt73; const struct rtw89_ple_quota ple_qt74; const struct rtw89_ple_quota ple_qt75; + const struct rtw89_ple_quota ple_qt78; + const struct rtw89_ple_quota ple_qt79; const struct rtw89_ple_quota ple_qt_52a_wow; const struct rtw89_ple_quota ple_qt_52b_wow; const struct rtw89_ple_quota ple_qt_52bt_wow; diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852c.c b/drivers/net/wireless/realtek/rtw89/rtw8852c.c index 6aedbd03e3d4..651ef96d143a 100644 --- a/drivers/net/wireless/realtek/rtw89/rtw8852c.c +++ b/drivers/net/wireless/realtek/rtw89/rtw8852c.c @@ -64,6 +64,32 @@ static const struct rtw89_dle_mem rtw8852c_dle_mem_pcie[] = { NULL}, }; +static const struct rtw89_dle_mem rtw8852c_dle_mem_usb2[] = { + [RTW89_QTA_SCC] = {RTW89_QTA_SCC, &rtw89_mac_size.wde_size31, + &rtw89_mac_size.ple_size34, &rtw89_mac_size.wde_qt31, + &rtw89_mac_size.wde_qt31, &rtw89_mac_size.ple_qt78, + &rtw89_mac_size.ple_qt79}, + [RTW89_QTA_DLFW] = {RTW89_QTA_DLFW, &rtw89_mac_size.wde_size18, + &rtw89_mac_size.ple_size18, &rtw89_mac_size.wde_qt17, + &rtw89_mac_size.wde_qt17, &rtw89_mac_size.ple_qt44, + &rtw89_mac_size.ple_qt45}, + [RTW89_QTA_INVALID] = {RTW89_QTA_INVALID, NULL, NULL, NULL, NULL, NULL, + NULL}, +}; + +static const struct rtw89_dle_mem rtw8852c_dle_mem_usb3[] = { + [RTW89_QTA_SCC] = {RTW89_QTA_SCC, &rtw89_mac_size.wde_size17, + &rtw89_mac_size.ple_size17, &rtw89_mac_size.wde_qt16, + &rtw89_mac_size.wde_qt16, &rtw89_mac_size.ple_qt42, + &rtw89_mac_size.ple_qt43}, + [RTW89_QTA_DLFW] = {RTW89_QTA_DLFW, &rtw89_mac_size.wde_size18, + &rtw89_mac_size.ple_size18, &rtw89_mac_size.wde_qt17, + &rtw89_mac_size.wde_qt17, &rtw89_mac_size.ple_qt44, + &rtw89_mac_size.ple_qt45}, + [RTW89_QTA_INVALID] = {RTW89_QTA_INVALID, NULL, NULL, NULL, NULL, NULL, + NULL}, +}; + static const u32 rtw8852c_h2c_regs[RTW89_H2CREG_MAX] = { R_AX_H2CREG_DATA0_V1, R_AX_H2CREG_DATA1_V1, R_AX_H2CREG_DATA2_V1, R_AX_H2CREG_DATA3_V1 @@ -3026,7 +3052,10 @@ const struct rtw89_chip_info rtw8852c_chip_info = { .dis_2g_40m_ul_ofdma = false, .rsvd_ple_ofst = 0x6f800, .hfc_param_ini = {rtw8852c_hfc_param_ini_pcie, NULL, NULL}, - .dle_mem = {rtw8852c_dle_mem_pcie, NULL, NULL, NULL}, + .dle_mem = {rtw8852c_dle_mem_pcie, + rtw8852c_dle_mem_usb2, + rtw8852c_dle_mem_usb3, + NULL}, .wde_qempty_acq_grpnum = 16, .wde_qempty_mgq_grpsel = 16, .rf_base_addr = {0xe000, 0xf000}, From 19faad8c729df195f189763151dceb7ff5851fde Mon Sep 17 00:00:00 2001 From: Bitterblue Smith Date: Sat, 1 Nov 2025 21:26:41 +0200 Subject: [PATCH 0451/1381] wifi: rtw89: Add rtw8852c_hfc_param_ini_usb "hfc" means "hci fc" which is "Host Control Interface Flow Control". These are some parameters needed for RTL8852CU. Signed-off-by: Bitterblue Smith Acked-by: Ping-Ke Shih Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/5b90204c-60ad-4579-b241-b7ac2e1fee91@gmail.com --- drivers/net/wireless/realtek/rtw89/rtw8852c.c | 46 ++++++++++++++++++- 1 file changed, 45 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852c.c b/drivers/net/wireless/realtek/rtw89/rtw8852c.c index 651ef96d143a..10e3963b692c 100644 --- a/drivers/net/wireless/realtek/rtw89/rtw8852c.c +++ b/drivers/net/wireless/realtek/rtw89/rtw8852c.c @@ -51,6 +51,48 @@ static const struct rtw89_hfc_param_ini rtw8852c_hfc_param_ini_pcie[] = { [RTW89_QTA_INVALID] = {NULL}, }; +static const struct rtw89_hfc_ch_cfg rtw8852c_hfc_chcfg_usb[] = { + {18, 344, grp_0}, /* ACH 0 */ + {0, 0, grp_0}, /* ACH 1 */ + {18, 344, grp_0}, /* ACH 2 */ + {0, 0, grp_0}, /* ACH 3 */ + {18, 344, grp_0}, /* ACH 4 */ + {0, 0, grp_0}, /* ACH 5 */ + {18, 344, grp_0}, /* ACH 6 */ + {0, 0, grp_0}, /* ACH 7 */ + {18, 344, grp_0}, /* B0MGQ */ + {0, 0, grp_0}, /* B0HIQ */ + {18, 344, grp_0}, /* B1MGQ */ + {0, 0, grp_0}, /* B1HIQ */ + {0, 0, 0} /* FWCMDQ */ +}; + +static const struct rtw89_hfc_pub_cfg rtw8852c_hfc_pubcfg_usb = { + 344, /* Group 0 */ + 0, /* Group 1 */ + 344, /* Public Max */ + 0 /* WP threshold */ +}; + +static const struct rtw89_hfc_prec_cfg rtw8852c_hfc_preccfg_usb = { + 9, /* CH 0-11 pre-cost */ + 32, /* H2C pre-cost */ + 146, /* WP CH 0-7 pre-cost */ + 146, /* WP CH 8-11 pre-cost */ + 1, /* CH 0-11 full condition */ + 1, /* H2C full condition */ + 1, /* WP CH 0-7 full condition */ + 1, /* WP CH 8-11 full condition */ +}; + +static const struct rtw89_hfc_param_ini rtw8852c_hfc_param_ini_usb[] = { + [RTW89_QTA_SCC] = {rtw8852c_hfc_chcfg_usb, &rtw8852c_hfc_pubcfg_usb, + &rtw8852c_hfc_preccfg_usb, RTW89_HCIFC_STF}, + [RTW89_QTA_DLFW] = {NULL, NULL, + &rtw8852c_hfc_preccfg_usb, RTW89_HCIFC_STF}, + [RTW89_QTA_INVALID] = {NULL}, +}; + static const struct rtw89_dle_mem rtw8852c_dle_mem_pcie[] = { [RTW89_QTA_SCC] = {RTW89_QTA_SCC, &rtw89_mac_size.wde_size19, &rtw89_mac_size.ple_size19, &rtw89_mac_size.wde_qt18, @@ -3051,7 +3093,9 @@ const struct rtw89_chip_info rtw8852c_chip_info = { .max_amsdu_limit = 8000, .dis_2g_40m_ul_ofdma = false, .rsvd_ple_ofst = 0x6f800, - .hfc_param_ini = {rtw8852c_hfc_param_ini_pcie, NULL, NULL}, + .hfc_param_ini = {rtw8852c_hfc_param_ini_pcie, + rtw8852c_hfc_param_ini_usb, + NULL}, .dle_mem = {rtw8852c_dle_mem_pcie, rtw8852c_dle_mem_usb2, rtw8852c_dle_mem_usb3, From 6bc2711085a52bd4d8da0173b22b8d4d532c5d21 Mon Sep 17 00:00:00 2001 From: Bitterblue Smith Date: Sat, 1 Nov 2025 21:27:09 +0200 Subject: [PATCH 0452/1381] wifi: rtw89: 8852c: Accept USB devices and load their MAC address Make rtw8852c_read_efuse() accept USB devices and load the MAC address from the correct offset. Also fix the offset of the MAC address because it was wrong. Signed-off-by: Bitterblue Smith Acked-by: Ping-Ke Shih Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/78138507-99ea-4a58-a02b-f4d11dbfba3b@gmail.com --- drivers/net/wireless/realtek/rtw89/rtw8852c.c | 16 +++++++--------- drivers/net/wireless/realtek/rtw89/rtw8852c.h | 2 +- 2 files changed, 8 insertions(+), 10 deletions(-) diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852c.c b/drivers/net/wireless/realtek/rtw89/rtw8852c.c index 10e3963b692c..8550ae3bcb64 100644 --- a/drivers/net/wireless/realtek/rtw89/rtw8852c.c +++ b/drivers/net/wireless/realtek/rtw89/rtw8852c.c @@ -480,14 +480,6 @@ static int rtw8852c_pwr_off_func(struct rtw89_dev *rtwdev) return 0; } -static void rtw8852c_e_efuse_parsing(struct rtw89_efuse *efuse, - struct rtw8852c_efuse *map) -{ - ether_addr_copy(efuse->addr, map->e.mac_addr); - efuse->rfe_type = map->rfe_type; - efuse->xtal_cap = map->xtal_k; -} - static void rtw8852c_efuse_parsing_tssi(struct rtw89_dev *rtwdev, struct rtw8852c_efuse *map) { @@ -596,12 +588,18 @@ static int rtw8852c_read_efuse(struct rtw89_dev *rtwdev, u8 *log_map, switch (rtwdev->hci.type) { case RTW89_HCI_TYPE_PCIE: - rtw8852c_e_efuse_parsing(efuse, map); + ether_addr_copy(efuse->addr, map->e.mac_addr); + break; + case RTW89_HCI_TYPE_USB: + ether_addr_copy(efuse->addr, map->u.mac_addr); break; default: return -ENOTSUPP; } + efuse->rfe_type = map->rfe_type; + efuse->xtal_cap = map->xtal_k; + rtw89_info(rtwdev, "chip rfe_type is %d\n", efuse->rfe_type); return 0; diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852c.h b/drivers/net/wireless/realtek/rtw89/rtw8852c.h index 77b05daedd10..8585921ac6c4 100644 --- a/drivers/net/wireless/realtek/rtw89/rtw8852c.h +++ b/drivers/net/wireless/realtek/rtw89/rtw8852c.h @@ -11,7 +11,7 @@ #define BB_PATH_NUM_8852C 2 struct rtw8852c_u_efuse { - u8 rsvd[0x38]; + u8 rsvd[0x88]; u8 mac_addr[ETH_ALEN]; }; From 406849000df417a0618dca96f57f4bd06cf2b7d9 Mon Sep 17 00:00:00 2001 From: Bitterblue Smith Date: Sat, 1 Nov 2025 21:27:38 +0200 Subject: [PATCH 0453/1381] wifi: rtw89: Add rtw8852cu.c This is the entry point for the new rtw89_8852cu module. Signed-off-by: Bitterblue Smith Acked-by: Ping-Ke Shih Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/a63a71f0-8003-4390-8bee-5cfd298867a0@gmail.com --- .../net/wireless/realtek/rtw89/rtw8852cu.c | 69 +++++++++++++++++++ 1 file changed, 69 insertions(+) create mode 100644 drivers/net/wireless/realtek/rtw89/rtw8852cu.c diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852cu.c b/drivers/net/wireless/realtek/rtw89/rtw8852cu.c new file mode 100644 index 000000000000..2708b523ca14 --- /dev/null +++ b/drivers/net/wireless/realtek/rtw89/rtw8852cu.c @@ -0,0 +1,69 @@ +// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause +/* Copyright(c) 2025 Realtek Corporation + */ + +#include +#include +#include "rtw8852c.h" +#include "reg.h" +#include "usb.h" + +static const struct rtw89_usb_info rtw8852c_usb_info = { + .usb_host_request_2 = R_AX_USB_HOST_REQUEST_2_V1, + .usb_wlan0_1 = R_AX_USB_WLAN0_1_V1, + .hci_func_en = R_AX_HCI_FUNC_EN_V1, + .usb3_mac_npi_config_intf_0 = R_AX_USB3_MAC_NPI_CONFIG_INTF_0_V1, + .usb_endpoint_0 = R_AX_USB_ENDPOINT_0_V1, + .usb_endpoint_2 = R_AX_USB_ENDPOINT_2_V1, + .bulkout_id = { + [RTW89_DMA_ACH0] = 3, + [RTW89_DMA_ACH2] = 5, + [RTW89_DMA_ACH4] = 4, + [RTW89_DMA_ACH6] = 6, + [RTW89_DMA_B0MG] = 0, + [RTW89_DMA_B0HI] = 0, + [RTW89_DMA_B1MG] = 1, + [RTW89_DMA_B1HI] = 1, + [RTW89_DMA_H2C] = 2, + }, +}; + +static const struct rtw89_driver_info rtw89_8852cu_info = { + .chip = &rtw8852c_chip_info, + .variant = NULL, + .quirks = NULL, + .bus = { + .usb = &rtw8852c_usb_info, + }, +}; + +static const struct usb_device_id rtw_8852cu_id_table[] = { + { USB_DEVICE_AND_INTERFACE_INFO(0x0bda, 0xc832, 0xff, 0xff, 0xff), + .driver_info = (kernel_ulong_t)&rtw89_8852cu_info }, + { USB_DEVICE_AND_INTERFACE_INFO(0x0bda, 0xc85a, 0xff, 0xff, 0xff), + .driver_info = (kernel_ulong_t)&rtw89_8852cu_info }, + { USB_DEVICE_AND_INTERFACE_INFO(0x0bda, 0xc85d, 0xff, 0xff, 0xff), + .driver_info = (kernel_ulong_t)&rtw89_8852cu_info }, + { USB_DEVICE_AND_INTERFACE_INFO(0x0db0, 0x991d, 0xff, 0xff, 0xff), + .driver_info = (kernel_ulong_t)&rtw89_8852cu_info }, + { USB_DEVICE_AND_INTERFACE_INFO(0x35b2, 0x0502, 0xff, 0xff, 0xff), + .driver_info = (kernel_ulong_t)&rtw89_8852cu_info }, + { USB_DEVICE_AND_INTERFACE_INFO(0x35bc, 0x0101, 0xff, 0xff, 0xff), + .driver_info = (kernel_ulong_t)&rtw89_8852cu_info }, + { USB_DEVICE_AND_INTERFACE_INFO(0x35bc, 0x0102, 0xff, 0xff, 0xff), + .driver_info = (kernel_ulong_t)&rtw89_8852cu_info }, + {}, +}; +MODULE_DEVICE_TABLE(usb, rtw_8852cu_id_table); + +static struct usb_driver rtw_8852cu_driver = { + .name = KBUILD_MODNAME, + .id_table = rtw_8852cu_id_table, + .probe = rtw89_usb_probe, + .disconnect = rtw89_usb_disconnect, +}; +module_usb_driver(rtw_8852cu_driver); + +MODULE_AUTHOR("Bitterblue Smith "); +MODULE_DESCRIPTION("Realtek 802.11ax wireless 8852CU driver"); +MODULE_LICENSE("Dual BSD/GPL"); From 0d971ffdae584ea4a5b413318b4945b3d273a75a Mon Sep 17 00:00:00 2001 From: Bitterblue Smith Date: Sat, 1 Nov 2025 21:28:17 +0200 Subject: [PATCH 0454/1381] wifi: rtw89: Enable the new rtw89_8852cu module Tested mostly in station mode, and a little bit in AP mode. Signed-off-by: Bitterblue Smith Acked-by: Ping-Ke Shih Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/859a45ce-2730-4eeb-99d2-37d7ff277bd7@gmail.com --- drivers/net/wireless/realtek/rtw89/Kconfig | 11 +++++++++++ drivers/net/wireless/realtek/rtw89/Makefile | 3 +++ 2 files changed, 14 insertions(+) diff --git a/drivers/net/wireless/realtek/rtw89/Kconfig b/drivers/net/wireless/realtek/rtw89/Kconfig index 4288c30b400a..e89f7481475a 100644 --- a/drivers/net/wireless/realtek/rtw89/Kconfig +++ b/drivers/net/wireless/realtek/rtw89/Kconfig @@ -121,6 +121,17 @@ config RTW89_8852CE 802.11ax PCIe wireless network (Wi-Fi 6E) adapter +config RTW89_8852CU + tristate "Realtek 8852CU USB wireless network (Wi-Fi 6E) adapter" + depends on USB + select RTW89_CORE + select RTW89_USB + select RTW89_8852C + help + Select this option will enable support for 8852CU chipset + + 802.11ax USB wireless network (Wi-Fi 6E) adapter + config RTW89_8922AE tristate "Realtek 8922AE/8922AE-VS PCI wireless network (Wi-Fi 7) adapter" depends on PCI diff --git a/drivers/net/wireless/realtek/rtw89/Makefile b/drivers/net/wireless/realtek/rtw89/Makefile index 23e43c444f69..e0d21972e57c 100644 --- a/drivers/net/wireless/realtek/rtw89/Makefile +++ b/drivers/net/wireless/realtek/rtw89/Makefile @@ -75,6 +75,9 @@ rtw89_8852c-objs := rtw8852c.o \ obj-$(CONFIG_RTW89_8852CE) += rtw89_8852ce.o rtw89_8852ce-objs := rtw8852ce.o +obj-$(CONFIG_RTW89_8852CU) += rtw89_8852cu.o +rtw89_8852cu-objs := rtw8852cu.o + obj-$(CONFIG_RTW89_8922A) += rtw89_8922a.o rtw89_8922a-objs := rtw8922a.o \ rtw8922a_rfk.o From 2214ca1ff6df1c1faab4fb95f0296b6d9bf6e1ee Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Wed, 29 Oct 2025 17:32:53 +0000 Subject: [PATCH 0455/1381] mpls: Return early in mpls_label_ok(). When mpls_label_ok() returns false, it does not need to update *index. Let's remove is_ok and return early. Signed-off-by: Kuniyuki Iwashima Reviewed-by: Guillaume Nault Link: https://patch.msgid.link/20251029173344.2934622-2-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/mpls/af_mpls.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) diff --git a/net/mpls/af_mpls.c b/net/mpls/af_mpls.c index 25c88cba5c48..e3533d85d372 100644 --- a/net/mpls/af_mpls.c +++ b/net/mpls/af_mpls.c @@ -940,24 +940,23 @@ static int mpls_nh_build_multi(struct mpls_route_config *cfg, static bool mpls_label_ok(struct net *net, unsigned int *index, struct netlink_ext_ack *extack) { - bool is_ok = true; - /* Reserved labels may not be set */ if (*index < MPLS_LABEL_FIRST_UNRESERVED) { NL_SET_ERR_MSG(extack, "Invalid label - must be MPLS_LABEL_FIRST_UNRESERVED or higher"); - is_ok = false; + return false; } /* The full 20 bit range may not be supported. */ - if (is_ok && *index >= net->mpls.platform_labels) { + if (*index >= net->mpls.platform_labels) { NL_SET_ERR_MSG(extack, "Label >= configured maximum in platform_labels"); - is_ok = false; + return false; } *index = array_index_nospec(*index, net->mpls.platform_labels); - return is_ok; + + return true; } static int mpls_route_add(struct mpls_route_config *cfg, From f0914b8436c589b7ab32c614d8d7868eb4ebd5bf Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Wed, 29 Oct 2025 17:32:54 +0000 Subject: [PATCH 0456/1381] mpls: Hold dev refcnt for mpls_nh. MPLS uses RTNL 1) to guarantee the lifetime of struct mpls_nh.nh_dev 2) to protect net->mpls.platform_label , but neither actually requires RTNL. If we do not call dev_put() in find_outdev() and call it just before freeing struct mpls_route, we can drop RTNL for 1). Let's hold the refcnt of mpls_nh.nh_dev and track it with netdevice_tracker. Two notable changes: If mpls_nh_build_multi() fails to set up a neighbour, we need to call netdev_put() for successfully created neighbours in mpls_rt_free_rcu(), so the number of neighbours (rt->rt_nhn) is now updated in each iteration. When a dev is unregistered, mpls_ifdown() clones mpls_route and replaces it with the clone, so the clone requires extra netdev_hold(). Signed-off-by: Kuniyuki Iwashima Reviewed-by: Guillaume Nault Link: https://patch.msgid.link/20251029173344.2934622-3-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/mpls/af_mpls.c | 63 +++++++++++++++++++++++++++++++-------------- net/mpls/internal.h | 1 + 2 files changed, 45 insertions(+), 19 deletions(-) diff --git a/net/mpls/af_mpls.c b/net/mpls/af_mpls.c index e3533d85d372..e7be87466809 100644 --- a/net/mpls/af_mpls.c +++ b/net/mpls/af_mpls.c @@ -530,10 +530,23 @@ static struct mpls_route *mpls_rt_alloc(u8 num_nh, u8 max_alen, u8 max_labels) return rt; } +static void mpls_rt_free_rcu(struct rcu_head *head) +{ + struct mpls_route *rt; + + rt = container_of(head, struct mpls_route, rt_rcu); + + change_nexthops(rt) { + netdev_put(nh->nh_dev, &nh->nh_dev_tracker); + } endfor_nexthops(rt); + + kfree(rt); +} + static void mpls_rt_free(struct mpls_route *rt) { if (rt) - kfree_rcu(rt, rt_rcu); + call_rcu(&rt->rt_rcu, mpls_rt_free_rcu); } static void mpls_notify_route(struct net *net, unsigned index, @@ -587,6 +600,7 @@ static unsigned find_free_label(struct net *net) #if IS_ENABLED(CONFIG_INET) static struct net_device *inet_fib_lookup_dev(struct net *net, + struct mpls_nh *nh, const void *addr) { struct net_device *dev; @@ -599,14 +613,14 @@ static struct net_device *inet_fib_lookup_dev(struct net *net, return ERR_CAST(rt); dev = rt->dst.dev; - dev_hold(dev); - + netdev_hold(dev, &nh->nh_dev_tracker, GFP_KERNEL); ip_rt_put(rt); return dev; } #else static struct net_device *inet_fib_lookup_dev(struct net *net, + struct mpls_nh *nh, const void *addr) { return ERR_PTR(-EAFNOSUPPORT); @@ -615,6 +629,7 @@ static struct net_device *inet_fib_lookup_dev(struct net *net, #if IS_ENABLED(CONFIG_IPV6) static struct net_device *inet6_fib_lookup_dev(struct net *net, + struct mpls_nh *nh, const void *addr) { struct net_device *dev; @@ -631,13 +646,14 @@ static struct net_device *inet6_fib_lookup_dev(struct net *net, return ERR_CAST(dst); dev = dst->dev; - dev_hold(dev); + netdev_hold(dev, &nh->nh_dev_tracker, GFP_KERNEL); dst_release(dst); return dev; } #else static struct net_device *inet6_fib_lookup_dev(struct net *net, + struct mpls_nh *nh, const void *addr) { return ERR_PTR(-EAFNOSUPPORT); @@ -653,16 +669,17 @@ static struct net_device *find_outdev(struct net *net, if (!oif) { switch (nh->nh_via_table) { case NEIGH_ARP_TABLE: - dev = inet_fib_lookup_dev(net, mpls_nh_via(rt, nh)); + dev = inet_fib_lookup_dev(net, nh, mpls_nh_via(rt, nh)); break; case NEIGH_ND_TABLE: - dev = inet6_fib_lookup_dev(net, mpls_nh_via(rt, nh)); + dev = inet6_fib_lookup_dev(net, nh, mpls_nh_via(rt, nh)); break; case NEIGH_LINK_TABLE: break; } } else { - dev = dev_get_by_index(net, oif); + dev = netdev_get_by_index(net, oif, + &nh->nh_dev_tracker, GFP_KERNEL); } if (!dev) @@ -671,8 +688,7 @@ static struct net_device *find_outdev(struct net *net, if (IS_ERR(dev)) return dev; - /* The caller is holding rtnl anyways, so release the dev reference */ - dev_put(dev); + nh->nh_dev = dev; return dev; } @@ -686,20 +702,17 @@ static int mpls_nh_assign_dev(struct net *net, struct mpls_route *rt, dev = find_outdev(net, rt, nh, oif); if (IS_ERR(dev)) { err = PTR_ERR(dev); - dev = NULL; goto errout; } /* Ensure this is a supported device */ err = -EINVAL; if (!mpls_dev_get(dev)) - goto errout; + goto errout_put; if ((nh->nh_via_table == NEIGH_LINK_TABLE) && (dev->addr_len != nh->nh_via_alen)) - goto errout; - - nh->nh_dev = dev; + goto errout_put; if (!(dev->flags & IFF_UP)) { nh->nh_flags |= RTNH_F_DEAD; @@ -713,6 +726,9 @@ static int mpls_nh_assign_dev(struct net *net, struct mpls_route *rt, return 0; +errout_put: + netdev_put(nh->nh_dev, &nh->nh_dev_tracker); + nh->nh_dev = NULL; errout: return err; } @@ -890,7 +906,8 @@ static int mpls_nh_build_multi(struct mpls_route_config *cfg, struct nlattr *nla_via, *nla_newdst; int remaining = cfg->rc_mp_len; int err = 0; - u8 nhs = 0; + + rt->rt_nhn = 0; change_nexthops(rt) { int attrlen; @@ -926,11 +943,9 @@ static int mpls_nh_build_multi(struct mpls_route_config *cfg, rt->rt_nhn_alive--; rtnh = rtnh_next(rtnh, &remaining); - nhs++; + rt->rt_nhn++; } endfor_nexthops(rt); - rt->rt_nhn = nhs; - return 0; errout: @@ -1523,8 +1538,12 @@ static int mpls_ifdown(struct net_device *dev, int event) change_nexthops(rt) { unsigned int nh_flags = nh->nh_flags; - if (nh->nh_dev != dev) + if (nh->nh_dev != dev) { + if (nh_del) + netdev_hold(nh->nh_dev, &nh->nh_dev_tracker, + GFP_KERNEL); goto next; + } switch (event) { case NETDEV_DOWN: @@ -2518,10 +2537,13 @@ static int resize_platform_label_table(struct net *net, size_t limit) /* In case the predefined labels need to be populated */ if (limit > MPLS_LABEL_IPV4NULL) { struct net_device *lo = net->loopback_dev; + rt0 = mpls_rt_alloc(1, lo->addr_len, 0); if (IS_ERR(rt0)) goto nort0; + rt0->rt_nh->nh_dev = lo; + netdev_hold(lo, &rt0->rt_nh->nh_dev_tracker, GFP_KERNEL); rt0->rt_protocol = RTPROT_KERNEL; rt0->rt_payload_type = MPT_IPV4; rt0->rt_ttl_propagate = MPLS_TTL_PROP_DEFAULT; @@ -2532,10 +2554,13 @@ static int resize_platform_label_table(struct net *net, size_t limit) } if (limit > MPLS_LABEL_IPV6NULL) { struct net_device *lo = net->loopback_dev; + rt2 = mpls_rt_alloc(1, lo->addr_len, 0); if (IS_ERR(rt2)) goto nort2; + rt2->rt_nh->nh_dev = lo; + netdev_hold(lo, &rt2->rt_nh->nh_dev_tracker, GFP_KERNEL); rt2->rt_protocol = RTPROT_KERNEL; rt2->rt_payload_type = MPT_IPV6; rt2->rt_ttl_propagate = MPLS_TTL_PROP_DEFAULT; diff --git a/net/mpls/internal.h b/net/mpls/internal.h index 83c629529b57..3a5feca27d6a 100644 --- a/net/mpls/internal.h +++ b/net/mpls/internal.h @@ -88,6 +88,7 @@ enum mpls_payload_type { struct mpls_nh { /* next hop label forwarding entry */ struct net_device *nh_dev; + netdevice_tracker nh_dev_tracker; /* nh_flags is accessed under RCU in the packet path; it is * modified handling netdev events with rtnl lock held From 451c538ec067e84c1bf1c2b99ebc2b1ca0a09090 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Wed, 29 Oct 2025 17:32:55 +0000 Subject: [PATCH 0457/1381] mpls: Unify return paths in mpls_dev_notify(). We will protect net->mpls.platform_label by a dedicated mutex. Then, we need to wrap functions called from mpls_dev_notify() with the mutex. As a prep, let's unify the return paths. Signed-off-by: Kuniyuki Iwashima Reviewed-by: Guillaume Nault Link: https://patch.msgid.link/20251029173344.2934622-4-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/mpls/af_mpls.c | 24 ++++++++++++++++-------- 1 file changed, 16 insertions(+), 8 deletions(-) diff --git a/net/mpls/af_mpls.c b/net/mpls/af_mpls.c index e7be87466809..c5bbf712f8be 100644 --- a/net/mpls/af_mpls.c +++ b/net/mpls/af_mpls.c @@ -1616,22 +1616,24 @@ static int mpls_dev_notify(struct notifier_block *this, unsigned long event, if (event == NETDEV_REGISTER) { mdev = mpls_add_dev(dev); - if (IS_ERR(mdev)) - return notifier_from_errno(PTR_ERR(mdev)); + if (IS_ERR(mdev)) { + err = PTR_ERR(mdev); + goto err; + } - return NOTIFY_OK; + goto out; } mdev = mpls_dev_get(dev); if (!mdev) - return NOTIFY_OK; + goto out; switch (event) { case NETDEV_DOWN: err = mpls_ifdown(dev, event); if (err) - return notifier_from_errno(err); + goto err; break; case NETDEV_UP: flags = netif_get_flags(dev); @@ -1647,13 +1649,14 @@ static int mpls_dev_notify(struct notifier_block *this, unsigned long event, } else { err = mpls_ifdown(dev, event); if (err) - return notifier_from_errno(err); + goto err; } break; case NETDEV_UNREGISTER: err = mpls_ifdown(dev, event); if (err) - return notifier_from_errno(err); + goto err; + mdev = mpls_dev_get(dev); if (mdev) { mpls_dev_sysctl_unregister(dev, mdev); @@ -1667,11 +1670,16 @@ static int mpls_dev_notify(struct notifier_block *this, unsigned long event, mpls_dev_sysctl_unregister(dev, mdev); err = mpls_dev_sysctl_register(dev, mdev); if (err) - return notifier_from_errno(err); + goto err; } break; } + +out: return NOTIFY_OK; + +err: + return notifier_from_errno(err); } static struct notifier_block mpls_dev_notifier = { From d8f9581e1b7f1fe2e1ac985f4ea508d044c90733 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Wed, 29 Oct 2025 17:32:56 +0000 Subject: [PATCH 0458/1381] ipv6: Add in6_dev_rcu(). rcu_dereference_rtnl() does not clearly tell whether the caller is under RCU or RTNL. Let's add in6_dev_rcu() to make it easy to remove __in6_dev_get() in the future. Signed-off-by: Kuniyuki Iwashima Reviewed-by: Guillaume Nault Link: https://patch.msgid.link/20251029173344.2934622-5-kuniyu@google.com Signed-off-by: Jakub Kicinski --- include/net/addrconf.h | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/include/net/addrconf.h b/include/net/addrconf.h index 9e5e95988b9e..78e8b877fb25 100644 --- a/include/net/addrconf.h +++ b/include/net/addrconf.h @@ -347,6 +347,11 @@ static inline struct inet6_dev *__in6_dev_get(const struct net_device *dev) return rcu_dereference_rtnl(dev->ip6_ptr); } +static inline struct inet6_dev *in6_dev_rcu(const struct net_device *dev) +{ + return rcu_dereference(dev->ip6_ptr); +} + static inline struct inet6_dev *__in6_dev_get_rtnl_net(const struct net_device *dev) { return rtnl_net_dereference(dev_net(dev), dev->ip6_ptr); From bc7ebc569e8cc768342dfb01af1a26c7fbef513e Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Wed, 29 Oct 2025 17:32:57 +0000 Subject: [PATCH 0459/1381] mpls: Use in6_dev_rcu() and dev_net_rcu() in mpls_forward() and mpls_xmit(). mpls_forward() and mpls_xmit() are called under RCU. Let's use in6_dev_rcu() and dev_net_rcu() there to annotate as such. Now we pass net to mpls_stats_inc_outucastpkts() not to read dev_net_rcu() twice. Signed-off-by: Kuniyuki Iwashima Reviewed-by: Guillaume Nault Link: https://patch.msgid.link/20251029173344.2934622-6-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/mpls/af_mpls.c | 15 ++++++++------- net/mpls/internal.h | 3 ++- net/mpls/mpls_iptunnel.c | 4 ++-- 3 files changed, 12 insertions(+), 10 deletions(-) diff --git a/net/mpls/af_mpls.c b/net/mpls/af_mpls.c index c5bbf712f8be..efc6c7da5766 100644 --- a/net/mpls/af_mpls.c +++ b/net/mpls/af_mpls.c @@ -129,7 +129,8 @@ bool mpls_pkt_too_big(const struct sk_buff *skb, unsigned int mtu) } EXPORT_SYMBOL_GPL(mpls_pkt_too_big); -void mpls_stats_inc_outucastpkts(struct net_device *dev, +void mpls_stats_inc_outucastpkts(struct net *net, + struct net_device *dev, const struct sk_buff *skb) { struct mpls_dev *mdev; @@ -141,13 +142,13 @@ void mpls_stats_inc_outucastpkts(struct net_device *dev, tx_packets, tx_bytes); } else if (skb->protocol == htons(ETH_P_IP)) { - IP_UPD_PO_STATS(dev_net(dev), IPSTATS_MIB_OUT, skb->len); + IP_UPD_PO_STATS(net, IPSTATS_MIB_OUT, skb->len); #if IS_ENABLED(CONFIG_IPV6) } else if (skb->protocol == htons(ETH_P_IPV6)) { - struct inet6_dev *in6dev = __in6_dev_get(dev); + struct inet6_dev *in6dev = in6_dev_rcu(dev); if (in6dev) - IP6_UPD_PO_STATS(dev_net(dev), in6dev, + IP6_UPD_PO_STATS(net, in6dev, IPSTATS_MIB_OUT, skb->len); #endif } @@ -342,7 +343,7 @@ static bool mpls_egress(struct net *net, struct mpls_route *rt, static int mpls_forward(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt, struct net_device *orig_dev) { - struct net *net = dev_net(dev); + struct net *net = dev_net_rcu(dev); struct mpls_shim_hdr *hdr; const struct mpls_nh *nh; struct mpls_route *rt; @@ -434,7 +435,7 @@ static int mpls_forward(struct sk_buff *skb, struct net_device *dev, dec.ttl -= 1; if (unlikely(!new_header_size && dec.bos)) { /* Penultimate hop popping */ - if (!mpls_egress(dev_net(out_dev), rt, skb, dec)) + if (!mpls_egress(net, rt, skb, dec)) goto err; } else { bool bos; @@ -451,7 +452,7 @@ static int mpls_forward(struct sk_buff *skb, struct net_device *dev, } } - mpls_stats_inc_outucastpkts(out_dev, skb); + mpls_stats_inc_outucastpkts(net, out_dev, skb); /* If via wasn't specified then send out using device address */ if (nh->nh_via_table == MPLS_NEIGH_TABLE_UNSPEC) diff --git a/net/mpls/internal.h b/net/mpls/internal.h index 3a5feca27d6a..e491427ea08a 100644 --- a/net/mpls/internal.h +++ b/net/mpls/internal.h @@ -197,7 +197,8 @@ int nla_get_labels(const struct nlattr *nla, u8 max_labels, u8 *labels, bool mpls_output_possible(const struct net_device *dev); unsigned int mpls_dev_mtu(const struct net_device *dev); bool mpls_pkt_too_big(const struct sk_buff *skb, unsigned int mtu); -void mpls_stats_inc_outucastpkts(struct net_device *dev, +void mpls_stats_inc_outucastpkts(struct net *net, + struct net_device *dev, const struct sk_buff *skb); #endif /* MPLS_INTERNAL_H */ diff --git a/net/mpls/mpls_iptunnel.c b/net/mpls/mpls_iptunnel.c index 6e73da94af7f..cfbab7b2fec7 100644 --- a/net/mpls/mpls_iptunnel.c +++ b/net/mpls/mpls_iptunnel.c @@ -53,7 +53,7 @@ static int mpls_xmit(struct sk_buff *skb) /* Find the output device */ out_dev = dst->dev; - net = dev_net(out_dev); + net = dev_net_rcu(out_dev); if (!mpls_output_possible(out_dev) || !dst->lwtstate || skb_warn_if_lro(skb)) @@ -128,7 +128,7 @@ static int mpls_xmit(struct sk_buff *skb) bos = false; } - mpls_stats_inc_outucastpkts(out_dev, skb); + mpls_stats_inc_outucastpkts(net, out_dev, skb); if (rt) { if (rt->rt_gw_family == AF_INET6) From ab061f3347923b6e3aa7731056dc58cbe5044c9f Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Wed, 29 Oct 2025 17:32:58 +0000 Subject: [PATCH 0460/1381] mpls: Add mpls_dev_rcu(). mpls_dev_get() uses rcu_dereference_rtnl() to fetch dev->mpls_ptr. We will replace RTNL with a dedicated mutex to protect the field. Then, we will use rcu_dereference_protected() for clarity. Let's add mpls_dev_rcu() for the RCU reader. Signed-off-by: Kuniyuki Iwashima Reviewed-by: Guillaume Nault Link: https://patch.msgid.link/20251029173344.2934622-7-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/mpls/af_mpls.c | 12 ++++++------ net/mpls/internal.h | 5 +++++ net/mpls/mpls_iptunnel.c | 2 +- 3 files changed, 12 insertions(+), 7 deletions(-) diff --git a/net/mpls/af_mpls.c b/net/mpls/af_mpls.c index efc6c7da5766..10130b90c439 100644 --- a/net/mpls/af_mpls.c +++ b/net/mpls/af_mpls.c @@ -136,7 +136,7 @@ void mpls_stats_inc_outucastpkts(struct net *net, struct mpls_dev *mdev; if (skb->protocol == htons(ETH_P_MPLS_UC)) { - mdev = mpls_dev_get(dev); + mdev = mpls_dev_rcu(dev); if (mdev) MPLS_INC_STATS_LEN(mdev, skb->len, tx_packets, @@ -358,7 +358,7 @@ static int mpls_forward(struct sk_buff *skb, struct net_device *dev, /* Careful this entire function runs inside of an rcu critical section */ - mdev = mpls_dev_get(dev); + mdev = mpls_dev_rcu(dev); if (!mdev) goto drop; @@ -467,7 +467,7 @@ static int mpls_forward(struct sk_buff *skb, struct net_device *dev, return 0; tx_err: - out_mdev = out_dev ? mpls_dev_get(out_dev) : NULL; + out_mdev = out_dev ? mpls_dev_rcu(out_dev) : NULL; if (out_mdev) MPLS_INC_STATS(out_mdev, tx_errors); goto drop; @@ -1118,7 +1118,7 @@ static int mpls_fill_stats_af(struct sk_buff *skb, struct mpls_dev *mdev; struct nlattr *nla; - mdev = mpls_dev_get(dev); + mdev = mpls_dev_rcu(dev); if (!mdev) return -ENODATA; @@ -1138,7 +1138,7 @@ static size_t mpls_get_stats_af_size(const struct net_device *dev) { struct mpls_dev *mdev; - mdev = mpls_dev_get(dev); + mdev = mpls_dev_rcu(dev); if (!mdev) return 0; @@ -1341,7 +1341,7 @@ static int mpls_netconf_dump_devconf(struct sk_buff *skb, rcu_read_lock(); for_each_netdev_dump(net, dev, ctx->ifindex) { - mdev = mpls_dev_get(dev); + mdev = mpls_dev_rcu(dev); if (!mdev) continue; err = mpls_netconf_fill_devconf(skb, mdev, diff --git a/net/mpls/internal.h b/net/mpls/internal.h index e491427ea08a..080e82010022 100644 --- a/net/mpls/internal.h +++ b/net/mpls/internal.h @@ -185,6 +185,11 @@ static inline struct mpls_entry_decoded mpls_entry_decode(struct mpls_shim_hdr * return result; } +static inline struct mpls_dev *mpls_dev_rcu(const struct net_device *dev) +{ + return rcu_dereference(dev->mpls_ptr); +} + static inline struct mpls_dev *mpls_dev_get(const struct net_device *dev) { return rcu_dereference_rtnl(dev->mpls_ptr); diff --git a/net/mpls/mpls_iptunnel.c b/net/mpls/mpls_iptunnel.c index cfbab7b2fec7..1a1a0eb5b787 100644 --- a/net/mpls/mpls_iptunnel.c +++ b/net/mpls/mpls_iptunnel.c @@ -153,7 +153,7 @@ static int mpls_xmit(struct sk_buff *skb) return LWTUNNEL_XMIT_DONE; drop: - out_mdev = out_dev ? mpls_dev_get(out_dev) : NULL; + out_mdev = out_dev ? mpls_dev_rcu(out_dev) : NULL; if (out_mdev) MPLS_INC_STATS(out_mdev, tx_errors); kfree_skb(skb); From 1fb462de9329731c17267c7ccf19619f22790a0a Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Wed, 29 Oct 2025 17:32:59 +0000 Subject: [PATCH 0461/1381] mpls: Pass net to mpls_dev_get(). We will replace RTNL with a per-netns mutex to protect dev->mpls_ptr. Then, we will use rcu_dereference_protected() with the lockdep_is_held() annotation, which requires net to access the per-netns mutex. However, dev_net(dev) is not safe without RTNL. Let's pass net to mpls_dev_get(). Signed-off-by: Kuniyuki Iwashima Reviewed-by: Guillaume Nault Link: https://patch.msgid.link/20251029173344.2934622-8-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/mpls/af_mpls.c | 11 ++++++----- net/mpls/internal.h | 3 ++- 2 files changed, 8 insertions(+), 6 deletions(-) diff --git a/net/mpls/af_mpls.c b/net/mpls/af_mpls.c index 10130b90c439..a715b12860e9 100644 --- a/net/mpls/af_mpls.c +++ b/net/mpls/af_mpls.c @@ -708,7 +708,7 @@ static int mpls_nh_assign_dev(struct net *net, struct mpls_route *rt, /* Ensure this is a supported device */ err = -EINVAL; - if (!mpls_dev_get(dev)) + if (!mpls_dev_get(net, dev)) goto errout_put; if ((nh->nh_via_table == NEIGH_LINK_TABLE) && @@ -1288,7 +1288,7 @@ static int mpls_netconf_get_devconf(struct sk_buff *in_skb, if (!dev) goto errout; - mdev = mpls_dev_get(dev); + mdev = mpls_dev_get(net, dev); if (!mdev) goto errout; @@ -1611,6 +1611,7 @@ static int mpls_dev_notify(struct notifier_block *this, unsigned long event, void *ptr) { struct net_device *dev = netdev_notifier_info_to_dev(ptr); + struct net *net = dev_net(dev); struct mpls_dev *mdev; unsigned int flags; int err; @@ -1625,7 +1626,7 @@ static int mpls_dev_notify(struct notifier_block *this, unsigned long event, goto out; } - mdev = mpls_dev_get(dev); + mdev = mpls_dev_get(net, dev); if (!mdev) goto out; @@ -1658,7 +1659,7 @@ static int mpls_dev_notify(struct notifier_block *this, unsigned long event, if (err) goto err; - mdev = mpls_dev_get(dev); + mdev = mpls_dev_get(net, dev); if (mdev) { mpls_dev_sysctl_unregister(dev, mdev); RCU_INIT_POINTER(dev->mpls_ptr, NULL); @@ -1666,7 +1667,7 @@ static int mpls_dev_notify(struct notifier_block *this, unsigned long event, } break; case NETDEV_CHANGENAME: - mdev = mpls_dev_get(dev); + mdev = mpls_dev_get(net, dev); if (mdev) { mpls_dev_sysctl_unregister(dev, mdev); err = mpls_dev_sysctl_register(dev, mdev); diff --git a/net/mpls/internal.h b/net/mpls/internal.h index 080e82010022..0df01a5395ee 100644 --- a/net/mpls/internal.h +++ b/net/mpls/internal.h @@ -190,7 +190,8 @@ static inline struct mpls_dev *mpls_dev_rcu(const struct net_device *dev) return rcu_dereference(dev->mpls_ptr); } -static inline struct mpls_dev *mpls_dev_get(const struct net_device *dev) +static inline struct mpls_dev *mpls_dev_get(const struct net *net, + const struct net_device *dev) { return rcu_dereference_rtnl(dev->mpls_ptr); } From 73e40539399101667945ed8b8299d0fa67a4fca2 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Wed, 29 Oct 2025 17:33:00 +0000 Subject: [PATCH 0462/1381] mpls: Add mpls_route_input(). mpls_route_input_rcu() is called from mpls_forward() and mpls_getroute(). The former is under RCU, and the latter is under RTNL, so mpls_route_input_rcu() uses rcu_dereference_rtnl(). Let's use rcu_dereference() in mpls_route_input_rcu() and add an RTNL variant for mpls_getroute(). Later, we will remove rtnl_dereference() there. Signed-off-by: Kuniyuki Iwashima Reviewed-by: Guillaume Nault Link: https://patch.msgid.link/20251029173344.2934622-9-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/mpls/af_mpls.c | 28 ++++++++++++++++++---------- 1 file changed, 18 insertions(+), 10 deletions(-) diff --git a/net/mpls/af_mpls.c b/net/mpls/af_mpls.c index a715b12860e9..530f7e6f7b3c 100644 --- a/net/mpls/af_mpls.c +++ b/net/mpls/af_mpls.c @@ -75,16 +75,23 @@ static void rtmsg_lfib(int event, u32 label, struct mpls_route *rt, struct nlmsghdr *nlh, struct net *net, u32 portid, unsigned int nlm_flags); -static struct mpls_route *mpls_route_input_rcu(struct net *net, unsigned index) +static struct mpls_route *mpls_route_input(struct net *net, unsigned int index) { - struct mpls_route *rt = NULL; + struct mpls_route __rcu **platform_label; - if (index < net->mpls.platform_labels) { - struct mpls_route __rcu **platform_label = - rcu_dereference_rtnl(net->mpls.platform_label); - rt = rcu_dereference_rtnl(platform_label[index]); - } - return rt; + platform_label = rtnl_dereference(net->mpls.platform_label); + return rtnl_dereference(platform_label[index]); +} + +static struct mpls_route *mpls_route_input_rcu(struct net *net, unsigned int index) +{ + struct mpls_route __rcu **platform_label; + + if (index >= net->mpls.platform_labels) + return NULL; + + platform_label = rcu_dereference(net->mpls.platform_label); + return rcu_dereference(platform_label[index]); } bool mpls_output_possible(const struct net_device *dev) @@ -2373,12 +2380,12 @@ static int mpls_getroute(struct sk_buff *in_skb, struct nlmsghdr *in_nlh, u32 portid = NETLINK_CB(in_skb).portid; u32 in_label = LABEL_NOT_SPECIFIED; struct nlattr *tb[RTA_MAX + 1]; + struct mpls_route *rt = NULL; u32 labels[MAX_NEW_LABELS]; struct mpls_shim_hdr *hdr; unsigned int hdr_size = 0; const struct mpls_nh *nh; struct net_device *dev; - struct mpls_route *rt; struct rtmsg *rtm, *r; struct nlmsghdr *nlh; struct sk_buff *skb; @@ -2406,7 +2413,8 @@ static int mpls_getroute(struct sk_buff *in_skb, struct nlmsghdr *in_nlh, } } - rt = mpls_route_input_rcu(net, in_label); + if (in_label < net->mpls.platform_labels) + rt = mpls_route_input(net, in_label); if (!rt) { err = -ENETUNREACH; goto errout; From 3a49629335a523341d3fae895435d5217341a022 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Wed, 29 Oct 2025 17:33:01 +0000 Subject: [PATCH 0463/1381] mpls: Use mpls_route_input() where appropriate. In many places, we uses rtnl_dereference() twice for net->mpls.platform_label and net->mpls.platform_label[index]. Let's replace the code with mpls_route_input(). We do not use mpls_route_input() in mpls_dump_routes() since we will rely on RCU there. Signed-off-by: Kuniyuki Iwashima Reviewed-by: Guillaume Nault Link: https://patch.msgid.link/20251029173344.2934622-10-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/mpls/af_mpls.c | 32 +++++++++++++------------------- 1 file changed, 13 insertions(+), 19 deletions(-) diff --git a/net/mpls/af_mpls.c b/net/mpls/af_mpls.c index 530f7e6f7b3c..35ae3dbd7bdc 100644 --- a/net/mpls/af_mpls.c +++ b/net/mpls/af_mpls.c @@ -590,19 +590,17 @@ static void mpls_route_update(struct net *net, unsigned index, mpls_rt_free(rt); } -static unsigned find_free_label(struct net *net) +static unsigned int find_free_label(struct net *net) { - struct mpls_route __rcu **platform_label; - size_t platform_labels; - unsigned index; + unsigned int index; - platform_label = rtnl_dereference(net->mpls.platform_label); - platform_labels = net->mpls.platform_labels; - for (index = MPLS_LABEL_FIRST_UNRESERVED; index < platform_labels; + for (index = MPLS_LABEL_FIRST_UNRESERVED; + index < net->mpls.platform_labels; index++) { - if (!rtnl_dereference(platform_label[index])) + if (!mpls_route_input(net, index)) return index; } + return LABEL_NOT_SPECIFIED; } @@ -985,7 +983,6 @@ static bool mpls_label_ok(struct net *net, unsigned int *index, static int mpls_route_add(struct mpls_route_config *cfg, struct netlink_ext_ack *extack) { - struct mpls_route __rcu **platform_label; struct net *net = cfg->rc_nlinfo.nl_net; struct mpls_route *rt, *old; int err = -EINVAL; @@ -1013,8 +1010,7 @@ static int mpls_route_add(struct mpls_route_config *cfg, } err = -EEXIST; - platform_label = rtnl_dereference(net->mpls.platform_label); - old = rtnl_dereference(platform_label[index]); + old = mpls_route_input(net, index); if ((cfg->rc_nlflags & NLM_F_EXCL) && old) goto errout; @@ -1503,16 +1499,15 @@ static void mpls_dev_destroy_rcu(struct rcu_head *head) static int mpls_ifdown(struct net_device *dev, int event) { - struct mpls_route __rcu **platform_label; struct net *net = dev_net(dev); - unsigned index; + unsigned int index; - platform_label = rtnl_dereference(net->mpls.platform_label); for (index = 0; index < net->mpls.platform_labels; index++) { - struct mpls_route *rt = rtnl_dereference(platform_label[index]); + struct mpls_route *rt; bool nh_del = false; u8 alive = 0; + rt = mpls_route_input(net, index); if (!rt) continue; @@ -1583,15 +1578,14 @@ static int mpls_ifdown(struct net_device *dev, int event) static void mpls_ifup(struct net_device *dev, unsigned int flags) { - struct mpls_route __rcu **platform_label; struct net *net = dev_net(dev); - unsigned index; + unsigned int index; u8 alive; - platform_label = rtnl_dereference(net->mpls.platform_label); for (index = 0; index < net->mpls.platform_labels; index++) { - struct mpls_route *rt = rtnl_dereference(platform_label[index]); + struct mpls_route *rt; + rt = mpls_route_input(net, index); if (!rt) continue; From dde1b38e873cff70f0af36e884bbeb1b14a536ed Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Wed, 29 Oct 2025 17:33:02 +0000 Subject: [PATCH 0464/1381] mpls: Convert mpls_dump_routes() to RCU. mpls_dump_routes() sets fib_dump_filter.rtnl_held to true and calls __dev_get_by_index() in mpls_valid_fib_dump_req(). This is the only RTNL dependant in mpls_dump_routes(). Also, synchronize_rcu() in resize_platform_label_table() guarantees that net->mpls.platform_label is alive under RCU. Let's convert mpls_dump_routes() to RCU and use dev_get_by_index_rcu(). Signed-off-by: Kuniyuki Iwashima Reviewed-by: Guillaume Nault Link: https://patch.msgid.link/20251029173344.2934622-11-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/mpls/af_mpls.c | 26 ++++++++++++++++---------- 1 file changed, 16 insertions(+), 10 deletions(-) diff --git a/net/mpls/af_mpls.c b/net/mpls/af_mpls.c index 35ae3dbd7bdc..f00f75c137dc 100644 --- a/net/mpls/af_mpls.c +++ b/net/mpls/af_mpls.c @@ -2153,7 +2153,7 @@ static int mpls_valid_fib_dump_req(struct net *net, const struct nlmsghdr *nlh, if (i == RTA_OIF) { ifindex = nla_get_u32(tb[i]); - filter->dev = __dev_get_by_index(net, ifindex); + filter->dev = dev_get_by_index_rcu(net, ifindex); if (!filter->dev) return -ENODEV; filter->filter_set = 1; @@ -2191,20 +2191,19 @@ static int mpls_dump_routes(struct sk_buff *skb, struct netlink_callback *cb) struct net *net = sock_net(skb->sk); struct mpls_route __rcu **platform_label; struct fib_dump_filter filter = { - .rtnl_held = true, + .rtnl_held = false, }; unsigned int flags = NLM_F_MULTI; size_t platform_labels; unsigned int index; + int err; - ASSERT_RTNL(); + rcu_read_lock(); if (cb->strict_check) { - int err; - err = mpls_valid_fib_dump_req(net, nlh, &filter, cb); if (err < 0) - return err; + goto err; /* for MPLS, there is only 1 table with fixed type and flags. * If either are set in the filter then return nothing. @@ -2212,14 +2211,14 @@ static int mpls_dump_routes(struct sk_buff *skb, struct netlink_callback *cb) if ((filter.table_id && filter.table_id != RT_TABLE_MAIN) || (filter.rt_type && filter.rt_type != RTN_UNICAST) || filter.flags) - return skb->len; + goto unlock; } index = cb->args[0]; if (index < MPLS_LABEL_FIRST_UNRESERVED) index = MPLS_LABEL_FIRST_UNRESERVED; - platform_label = rtnl_dereference(net->mpls.platform_label); + platform_label = rcu_dereference(net->mpls.platform_label); platform_labels = net->mpls.platform_labels; if (filter.filter_set) @@ -2228,7 +2227,7 @@ static int mpls_dump_routes(struct sk_buff *skb, struct netlink_callback *cb) for (; index < platform_labels; index++) { struct mpls_route *rt; - rt = rtnl_dereference(platform_label[index]); + rt = rcu_dereference(platform_label[index]); if (!rt) continue; @@ -2243,7 +2242,13 @@ static int mpls_dump_routes(struct sk_buff *skb, struct netlink_callback *cb) } cb->args[0] = index; +unlock: + rcu_read_unlock(); return skb->len; + +err: + rcu_read_unlock(); + return err; } static inline size_t lfib_nlmsg_size(struct mpls_route *rt) @@ -2767,7 +2772,8 @@ static struct rtnl_af_ops mpls_af_ops __read_mostly = { static const struct rtnl_msg_handler mpls_rtnl_msg_handlers[] __initdata_or_module = { {THIS_MODULE, PF_MPLS, RTM_NEWROUTE, mpls_rtm_newroute, NULL, 0}, {THIS_MODULE, PF_MPLS, RTM_DELROUTE, mpls_rtm_delroute, NULL, 0}, - {THIS_MODULE, PF_MPLS, RTM_GETROUTE, mpls_getroute, mpls_dump_routes, 0}, + {THIS_MODULE, PF_MPLS, RTM_GETROUTE, mpls_getroute, mpls_dump_routes, + RTNL_FLAG_DUMP_UNLOCKED}, {THIS_MODULE, PF_MPLS, RTM_GETNETCONF, mpls_netconf_get_devconf, mpls_netconf_dump_devconf, RTNL_FLAG_DUMP_UNLOCKED}, From fb2b77b9b1dbd90cf7db5580b15df40a20b42bd8 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Wed, 29 Oct 2025 17:33:03 +0000 Subject: [PATCH 0465/1381] mpls: Convert RTM_GETNETCONF to RCU. mpls_netconf_get_devconf() calls __dev_get_by_index(), and this only depends on RTNL. Let's convert mpls_netconf_get_devconf() to RCU and use dev_get_by_index_rcu(). Note that nlmsg_new() is moved ahead to use GFP_KERNEL. Signed-off-by: Kuniyuki Iwashima Reviewed-by: Guillaume Nault Link: https://patch.msgid.link/20251029173344.2934622-12-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/mpls/af_mpls.c | 44 ++++++++++++++++++++++++++++++-------------- 1 file changed, 30 insertions(+), 14 deletions(-) diff --git a/net/mpls/af_mpls.c b/net/mpls/af_mpls.c index f00f75c137dc..49fd15232dbe 100644 --- a/net/mpls/af_mpls.c +++ b/net/mpls/af_mpls.c @@ -1282,23 +1282,32 @@ static int mpls_netconf_get_devconf(struct sk_buff *in_skb, if (err < 0) goto errout; - err = -EINVAL; - if (!tb[NETCONFA_IFINDEX]) + if (!tb[NETCONFA_IFINDEX]) { + err = -EINVAL; goto errout; + } ifindex = nla_get_s32(tb[NETCONFA_IFINDEX]); - dev = __dev_get_by_index(net, ifindex); - if (!dev) - goto errout; - mdev = mpls_dev_get(net, dev); - if (!mdev) - goto errout; - - err = -ENOBUFS; skb = nlmsg_new(mpls_netconf_msgsize_devconf(NETCONFA_ALL), GFP_KERNEL); - if (!skb) + if (!skb) { + err = -ENOBUFS; goto errout; + } + + rcu_read_lock(); + + dev = dev_get_by_index_rcu(net, ifindex); + if (!dev) { + err = -EINVAL; + goto errout_unlock; + } + + mdev = mpls_dev_rcu(dev); + if (!mdev) { + err = -EINVAL; + goto errout_unlock; + } err = mpls_netconf_fill_devconf(skb, mdev, NETLINK_CB(in_skb).portid, @@ -1307,12 +1316,19 @@ static int mpls_netconf_get_devconf(struct sk_buff *in_skb, if (err < 0) { /* -EMSGSIZE implies BUG in mpls_netconf_msgsize_devconf() */ WARN_ON(err == -EMSGSIZE); - kfree_skb(skb); - goto errout; + goto errout_unlock; } + err = rtnl_unicast(skb, net, NETLINK_CB(in_skb).portid); + + rcu_read_unlock(); errout: return err; + +errout_unlock: + rcu_read_unlock(); + kfree_skb(skb); + goto errout; } static int mpls_netconf_dump_devconf(struct sk_buff *skb, @@ -2776,7 +2792,7 @@ static const struct rtnl_msg_handler mpls_rtnl_msg_handlers[] __initdata_or_modu RTNL_FLAG_DUMP_UNLOCKED}, {THIS_MODULE, PF_MPLS, RTM_GETNETCONF, mpls_netconf_get_devconf, mpls_netconf_dump_devconf, - RTNL_FLAG_DUMP_UNLOCKED}, + RTNL_FLAG_DOIT_UNLOCKED | RTNL_FLAG_DUMP_UNLOCKED}, }; static int __init mpls_init(void) From e833eb25161aae6cd0caf14782f405d0ed5765ed Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Wed, 29 Oct 2025 17:33:04 +0000 Subject: [PATCH 0466/1381] mpls: Protect net->mpls.platform_label with a per-netns mutex. MPLS (re)uses RTNL to protect net->mpls.platform_label, but the lock does not need to be RTNL at all. Let's protect net->mpls.platform_label with a dedicated per-netns mutex. Signed-off-by: Kuniyuki Iwashima Reviewed-by: Guillaume Nault Link: https://patch.msgid.link/20251029173344.2934622-13-kuniyu@google.com Signed-off-by: Jakub Kicinski --- include/net/netns/mpls.h | 1 + net/mpls/af_mpls.c | 55 ++++++++++++++++++++++++++-------------- net/mpls/internal.h | 7 ++++- 3 files changed, 43 insertions(+), 20 deletions(-) diff --git a/include/net/netns/mpls.h b/include/net/netns/mpls.h index 19ad2574b267..6682e51513ef 100644 --- a/include/net/netns/mpls.h +++ b/include/net/netns/mpls.h @@ -16,6 +16,7 @@ struct netns_mpls { int default_ttl; size_t platform_labels; struct mpls_route __rcu * __rcu *platform_label; + struct mutex platform_mutex; struct ctl_table_header *ctl; }; diff --git a/net/mpls/af_mpls.c b/net/mpls/af_mpls.c index 49fd15232dbe..d0d047dd2245 100644 --- a/net/mpls/af_mpls.c +++ b/net/mpls/af_mpls.c @@ -79,8 +79,8 @@ static struct mpls_route *mpls_route_input(struct net *net, unsigned int index) { struct mpls_route __rcu **platform_label; - platform_label = rtnl_dereference(net->mpls.platform_label); - return rtnl_dereference(platform_label[index]); + platform_label = mpls_dereference(net, net->mpls.platform_label); + return mpls_dereference(net, platform_label[index]); } static struct mpls_route *mpls_route_input_rcu(struct net *net, unsigned int index) @@ -578,10 +578,8 @@ static void mpls_route_update(struct net *net, unsigned index, struct mpls_route __rcu **platform_label; struct mpls_route *rt; - ASSERT_RTNL(); - - platform_label = rtnl_dereference(net->mpls.platform_label); - rt = rtnl_dereference(platform_label[index]); + platform_label = mpls_dereference(net, net->mpls.platform_label); + rt = mpls_dereference(net, platform_label[index]); rcu_assign_pointer(platform_label[index], new); mpls_notify_route(net, index, rt, new, info); @@ -1472,8 +1470,6 @@ static struct mpls_dev *mpls_add_dev(struct net_device *dev) int err = -ENOMEM; int i; - ASSERT_RTNL(); - mdev = kzalloc(sizeof(*mdev), GFP_KERNEL); if (!mdev) return ERR_PTR(err); @@ -1633,6 +1629,8 @@ static int mpls_dev_notify(struct notifier_block *this, unsigned long event, unsigned int flags; int err; + mutex_lock(&net->mpls.platform_mutex); + if (event == NETDEV_REGISTER) { mdev = mpls_add_dev(dev); if (IS_ERR(mdev)) { @@ -1695,9 +1693,11 @@ static int mpls_dev_notify(struct notifier_block *this, unsigned long event, } out: + mutex_unlock(&net->mpls.platform_mutex); return NOTIFY_OK; err: + mutex_unlock(&net->mpls.platform_mutex); return notifier_from_errno(err); } @@ -1973,6 +1973,7 @@ static int rtm_to_route_config(struct sk_buff *skb, static int mpls_rtm_delroute(struct sk_buff *skb, struct nlmsghdr *nlh, struct netlink_ext_ack *extack) { + struct net *net = sock_net(skb->sk); struct mpls_route_config *cfg; int err; @@ -1984,7 +1985,9 @@ static int mpls_rtm_delroute(struct sk_buff *skb, struct nlmsghdr *nlh, if (err < 0) goto out; + mutex_lock(&net->mpls.platform_mutex); err = mpls_route_del(cfg, extack); + mutex_unlock(&net->mpls.platform_mutex); out: kfree(cfg); @@ -1995,6 +1998,7 @@ static int mpls_rtm_delroute(struct sk_buff *skb, struct nlmsghdr *nlh, static int mpls_rtm_newroute(struct sk_buff *skb, struct nlmsghdr *nlh, struct netlink_ext_ack *extack) { + struct net *net = sock_net(skb->sk); struct mpls_route_config *cfg; int err; @@ -2006,7 +2010,9 @@ static int mpls_rtm_newroute(struct sk_buff *skb, struct nlmsghdr *nlh, if (err < 0) goto out; + mutex_lock(&net->mpls.platform_mutex); err = mpls_route_add(cfg, extack); + mutex_unlock(&net->mpls.platform_mutex); out: kfree(cfg); @@ -2407,6 +2413,8 @@ static int mpls_getroute(struct sk_buff *in_skb, struct nlmsghdr *in_nlh, u8 n_labels; int err; + mutex_lock(&net->mpls.platform_mutex); + err = mpls_valid_getroute_req(in_skb, in_nlh, tb, extack); if (err < 0) goto errout; @@ -2450,7 +2458,8 @@ static int mpls_getroute(struct sk_buff *in_skb, struct nlmsghdr *in_nlh, goto errout_free; } - return rtnl_unicast(skb, net, portid); + err = rtnl_unicast(skb, net, portid); + goto errout; } if (tb[RTA_NEWDST]) { @@ -2542,12 +2551,14 @@ static int mpls_getroute(struct sk_buff *in_skb, struct nlmsghdr *in_nlh, err = rtnl_unicast(skb, net, portid); errout: + mutex_unlock(&net->mpls.platform_mutex); return err; nla_put_failure: nlmsg_cancel(skb, nlh); err = -EMSGSIZE; errout_free: + mutex_unlock(&net->mpls.platform_mutex); kfree_skb(skb); return err; } @@ -2603,9 +2614,10 @@ static int resize_platform_label_table(struct net *net, size_t limit) lo->addr_len); } - rtnl_lock(); + mutex_lock(&net->mpls.platform_mutex); + /* Remember the original table */ - old = rtnl_dereference(net->mpls.platform_label); + old = mpls_dereference(net, net->mpls.platform_label); old_limit = net->mpls.platform_labels; /* Free any labels beyond the new table */ @@ -2636,7 +2648,7 @@ static int resize_platform_label_table(struct net *net, size_t limit) net->mpls.platform_labels = limit; rcu_assign_pointer(net->mpls.platform_label, labels); - rtnl_unlock(); + mutex_unlock(&net->mpls.platform_mutex); mpls_rt_free(rt2); mpls_rt_free(rt0); @@ -2709,12 +2721,13 @@ static const struct ctl_table mpls_table[] = { }, }; -static int mpls_net_init(struct net *net) +static __net_init int mpls_net_init(struct net *net) { size_t table_size = ARRAY_SIZE(mpls_table); struct ctl_table *table; int i; + mutex_init(&net->mpls.platform_mutex); net->mpls.platform_labels = 0; net->mpls.platform_label = NULL; net->mpls.ip_ttl_propagate = 1; @@ -2740,7 +2753,7 @@ static int mpls_net_init(struct net *net) return 0; } -static void mpls_net_exit(struct net *net) +static __net_exit void mpls_net_exit(struct net *net) { struct mpls_route __rcu **platform_label; size_t platform_labels; @@ -2760,16 +2773,20 @@ static void mpls_net_exit(struct net *net) * As such no additional rcu synchronization is necessary when * freeing the platform_label table. */ - rtnl_lock(); - platform_label = rtnl_dereference(net->mpls.platform_label); + mutex_lock(&net->mpls.platform_mutex); + + platform_label = mpls_dereference(net, net->mpls.platform_label); platform_labels = net->mpls.platform_labels; + for (index = 0; index < platform_labels; index++) { - struct mpls_route *rt = rtnl_dereference(platform_label[index]); - RCU_INIT_POINTER(platform_label[index], NULL); + struct mpls_route *rt; + + rt = mpls_dereference(net, platform_label[index]); mpls_notify_route(net, index, rt, NULL, NULL); mpls_rt_free(rt); } - rtnl_unlock(); + + mutex_unlock(&net->mpls.platform_mutex); kvfree(platform_label); } diff --git a/net/mpls/internal.h b/net/mpls/internal.h index 0df01a5395ee..80cb5bbcd946 100644 --- a/net/mpls/internal.h +++ b/net/mpls/internal.h @@ -185,6 +185,11 @@ static inline struct mpls_entry_decoded mpls_entry_decode(struct mpls_shim_hdr * return result; } +#define mpls_dereference(net, p) \ + rcu_dereference_protected( \ + (p), \ + lockdep_is_held(&(net)->mpls.platform_mutex)) + static inline struct mpls_dev *mpls_dev_rcu(const struct net_device *dev) { return rcu_dereference(dev->mpls_ptr); @@ -193,7 +198,7 @@ static inline struct mpls_dev *mpls_dev_rcu(const struct net_device *dev) static inline struct mpls_dev *mpls_dev_get(const struct net *net, const struct net_device *dev) { - return rcu_dereference_rtnl(dev->mpls_ptr); + return mpls_dereference(net, dev->mpls_ptr); } int nla_put_labels(struct sk_buff *skb, int attrtype, u8 labels, From 7d99a7c6c6a3d0d6456520baa85d58095bf262ee Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Wed, 29 Oct 2025 17:33:05 +0000 Subject: [PATCH 0467/1381] mpls: Drop RTNL for RTM_NEWROUTE, RTM_DELROUTE, and RTM_GETROUTE. RTM_NEWROUTE looks up dev under RCU (ip_route_output(), ipv6_stub->ipv6_dst_lookup_flow(), netdev_get_by_index()), and each neighbour holds the refcnt of its dev. Also, net->mpls.platform_label is protected by a dedicated per-netns mutex. Now, no MPLS code depends on RTNL. Let's drop RTNL for RTM_NEWROUTE, RTM_DELROUTE, and RTM_GETROUTE. Signed-off-by: Kuniyuki Iwashima Reviewed-by: Guillaume Nault Link: https://patch.msgid.link/20251029173344.2934622-14-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/mpls/af_mpls.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/net/mpls/af_mpls.c b/net/mpls/af_mpls.c index d0d047dd2245..580aac112dd2 100644 --- a/net/mpls/af_mpls.c +++ b/net/mpls/af_mpls.c @@ -2803,10 +2803,12 @@ static struct rtnl_af_ops mpls_af_ops __read_mostly = { }; static const struct rtnl_msg_handler mpls_rtnl_msg_handlers[] __initdata_or_module = { - {THIS_MODULE, PF_MPLS, RTM_NEWROUTE, mpls_rtm_newroute, NULL, 0}, - {THIS_MODULE, PF_MPLS, RTM_DELROUTE, mpls_rtm_delroute, NULL, 0}, + {THIS_MODULE, PF_MPLS, RTM_NEWROUTE, mpls_rtm_newroute, NULL, + RTNL_FLAG_DOIT_UNLOCKED}, + {THIS_MODULE, PF_MPLS, RTM_DELROUTE, mpls_rtm_delroute, NULL, + RTNL_FLAG_DOIT_UNLOCKED}, {THIS_MODULE, PF_MPLS, RTM_GETROUTE, mpls_getroute, mpls_dump_routes, - RTNL_FLAG_DUMP_UNLOCKED}, + RTNL_FLAG_DOIT_UNLOCKED | RTNL_FLAG_DUMP_UNLOCKED}, {THIS_MODULE, PF_MPLS, RTM_GETNETCONF, mpls_netconf_get_devconf, mpls_netconf_dump_devconf, RTNL_FLAG_DOIT_UNLOCKED | RTNL_FLAG_DUMP_UNLOCKED}, From 71ee0d565985cbcf9f9c5c6b046e6d4ad52d0b3f Mon Sep 17 00:00:00 2001 From: Chin-Yen Lee Date: Mon, 27 Oct 2025 15:02:59 +0800 Subject: [PATCH 0468/1381] wifi: rtw88: 8822c: use fixed rate and bandwidth to reply CSI packets AP could trigger beamform and send NDPA packet with 6Mbps rate on bandwidth 80MHz, but RTL8822C can't reply CSI packet with the same setting. Therefore, force to use OFDM rate and bandwidth 20MHz instead. Signed-off-by: Chin-Yen Lee Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251027070259.18931-1-pkshih@realtek.com --- drivers/net/wireless/realtek/rtw88/bf.c | 8 +++++++- drivers/net/wireless/realtek/rtw88/bf.h | 7 +++++++ 2 files changed, 14 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/realtek/rtw88/bf.c b/drivers/net/wireless/realtek/rtw88/bf.c index c827c4a2814b..0d0ccbc7d00c 100644 --- a/drivers/net/wireless/realtek/rtw88/bf.c +++ b/drivers/net/wireless/realtek/rtw88/bf.c @@ -124,8 +124,11 @@ void rtw_bf_init_bfer_entry_mu(struct rtw_dev *rtwdev, void rtw_bf_cfg_sounding(struct rtw_dev *rtwdev, struct rtw_vif *vif, enum rtw_trx_desc_rate rate) { + u8 csi_rsc = CSI_RSC_FOLLOW_RX_PACKET_BW; u32 psf_ctl = 0; - u8 csi_rsc = 0x1; + + if (rtwdev->chip->id == RTW_CHIP_TYPE_8822C) + csi_rsc = CSI_RSC_PRIMARY_20M_BW; psf_ctl = rtw_read32(rtwdev, REG_BBPSF_CTRL) | BIT_WMAC_USE_NDPARATE | @@ -387,6 +390,9 @@ void rtw_bf_cfg_csi_rate(struct rtw_dev *rtwdev, u8 rssi, u8 cur_rate, csi_cfg = rtw_read32(rtwdev, REG_BBPSF_CTRL) & ~BIT_MASK_CSI_RATE; cur_rrsr = rtw_read16(rtwdev, REG_RRSR); + if (rtwdev->chip->id == RTW_CHIP_TYPE_8822C) + csi_cfg |= BIT_CSI_FORCE_RATE; + if (rssi >= 40) { if (cur_rate != DESC_RATE54M) { cur_rrsr |= BIT(DESC_RATE54M); diff --git a/drivers/net/wireless/realtek/rtw88/bf.h b/drivers/net/wireless/realtek/rtw88/bf.h index 7b40c2c03856..a5d3010e6be6 100644 --- a/drivers/net/wireless/realtek/rtw88/bf.h +++ b/drivers/net/wireless/realtek/rtw88/bf.h @@ -33,6 +33,7 @@ #define BIT_SHIFT_R_MU_RL 12 #define BIT_SHIFT_WMAC_TXMU_ACKPOLICY 4 #define BIT_SHIFT_CSI_RATE 24 +#define BIT_CSI_FORCE_RATE BIT(15) #define BIT_MASK_R_MU_RL (R_MU_RL << BIT_SHIFT_R_MU_RL) #define BIT_MASK_R_MU_TABLE_VALID 0x3f @@ -48,6 +49,12 @@ #define RTW_SND_CTRL_REMOVE 0x98 #define RTW_SND_CTRL_SOUNDING 0x9B +enum csi_rsc { + CSI_RSC_PRIMARY_20M_BW = 0, + CSI_RSC_FOLLOW_RX_PACKET_BW = 1, + CSI_RSC_DUPLICATE_MODE = 2, +}; + enum csi_seg_len { HAL_CSI_SEG_4K = 0, HAL_CSI_SEG_8K = 1, From 8adb609f645b42148c14989066a971bbb5f3587e Mon Sep 17 00:00:00 2001 From: Hiroaki Yamamoto Date: Fri, 31 Oct 2025 20:16:34 +0900 Subject: [PATCH 0469/1381] wifi: rtw88: Add BUFFALO WI-U3-866DHP to the USB ID list BUFFALO WI-U3-866DHP (0411:03d0) is based on rtl8812bu. I locally tested this patch with a retail sample and it worked fine. Signed-off-by: Hiroaki Yamamoto Acked-by: Ping-Ke Shih Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251031111641.33653-1-hrak1529@gmail.com --- drivers/net/wireless/realtek/rtw88/rtw8822bu.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/wireless/realtek/rtw88/rtw8822bu.c b/drivers/net/wireless/realtek/rtw88/rtw8822bu.c index 44e28e583964..2769b86ce1b2 100644 --- a/drivers/net/wireless/realtek/rtw88/rtw8822bu.c +++ b/drivers/net/wireless/realtek/rtw88/rtw8822bu.c @@ -79,6 +79,8 @@ static const struct usb_device_id rtw_8822bu_id_table[] = { .driver_info = (kernel_ulong_t)&(rtw8822b_hw_spec) }, /* D-Link DWA-T185 rev. A1 */ { USB_DEVICE_AND_INTERFACE_INFO(0x0411, 0x03d1, 0xff, 0xff, 0xff), .driver_info = (kernel_ulong_t)&(rtw8822b_hw_spec) }, /* BUFFALO WI-U2-866DM */ + { USB_DEVICE_AND_INTERFACE_INFO(0x0411, 0x03d0, 0xff, 0xff, 0xff), + .driver_info = (kernel_ulong_t)&(rtw8822b_hw_spec) }, /* BUFFALO WI-U3-866DHP */ {}, }; MODULE_DEVICE_TABLE(usb, rtw_8822bu_id_table); From c18d4b190a46651726c9a952667c74d2deb33c28 Mon Sep 17 00:00:00 2001 From: Samiullah Khawaja Date: Tue, 28 Oct 2025 20:30:05 +0000 Subject: [PATCH 0470/1381] net: Extend NAPI threaded polling to allow kthread based busy polling Add a new state NAPI_STATE_THREADED_BUSY_POLL to the NAPI state enum to enable and disable threaded busy polling. When threaded busy polling is enabled for a NAPI, enable NAPI_STATE_THREADED also. When the threaded NAPI is scheduled, set NAPI_STATE_IN_BUSY_POLL to signal napi_complete_done not to rearm interrupts. Whenever NAPI_STATE_THREADED_BUSY_POLL is unset, the NAPI_STATE_IN_BUSY_POLL will be unset, napi_complete_done unsets the NAPI_STATE_SCHED_THREADED bit also, which in turn will make the kthread go to sleep. Signed-off-by: Samiullah Khawaja Reviewed-by: Willem de Bruijn Acked-by: Martin Karsten Tested-by: Martin Karsten Link: https://patch.msgid.link/20251028203007.575686-2-skhawaja@google.com Signed-off-by: Jakub Kicinski --- Documentation/netlink/specs/netdev.yaml | 5 ++- Documentation/networking/napi.rst | 50 ++++++++++++++++++++- include/linux/netdevice.h | 4 +- include/uapi/linux/netdev.h | 1 + net/core/dev.c | 58 ++++++++++++++++++++----- net/core/dev.h | 3 ++ net/core/netdev-genl-gen.c | 2 +- tools/include/uapi/linux/netdev.h | 1 + 8 files changed, 109 insertions(+), 15 deletions(-) diff --git a/Documentation/netlink/specs/netdev.yaml b/Documentation/netlink/specs/netdev.yaml index e00d3fa1c152..10c412b7433f 100644 --- a/Documentation/netlink/specs/netdev.yaml +++ b/Documentation/netlink/specs/netdev.yaml @@ -88,7 +88,7 @@ definitions: - name: napi-threaded type: enum - entries: [disabled, enabled] + entries: [disabled, enabled, busy-poll] attribute-sets: - @@ -291,7 +291,8 @@ attribute-sets: name: threaded doc: Whether the NAPI is configured to operate in threaded polling mode. If this is set to enabled then the NAPI context operates - in threaded polling mode. + in threaded polling mode. If this is set to busy-poll, then the + threaded polling mode also busy polls. type: u32 enum: napi-threaded - diff --git a/Documentation/networking/napi.rst b/Documentation/networking/napi.rst index 7dd60366f4ff..4e008efebb35 100644 --- a/Documentation/networking/napi.rst +++ b/Documentation/networking/napi.rst @@ -263,7 +263,9 @@ are not well known). Busy polling is enabled by either setting ``SO_BUSY_POLL`` on selected sockets or using the global ``net.core.busy_poll`` and ``net.core.busy_read`` sysctls. An io_uring API for NAPI busy polling -also exists. +also exists. Threaded polling of NAPI also has a mode to busy poll for +packets (:ref:`threaded busy polling`) using the NAPI +processing kthread. epoll-based busy polling ------------------------ @@ -426,6 +428,52 @@ Therefore, setting ``gro_flush_timeout`` and ``napi_defer_hard_irqs`` is the recommended usage, because otherwise setting ``irq-suspend-timeout`` might not have any discernible effect. +.. _threaded_busy_poll: + +Threaded NAPI busy polling +-------------------------- + +Threaded NAPI busy polling extends threaded NAPI and adds support to do +continuous busy polling of the NAPI. This can be useful for forwarding or +AF_XDP applications. + +Threaded NAPI busy polling can be enabled on per NIC queue basis using Netlink. + +For example, using the following script: + +.. code-block:: bash + + $ ynl --family netdev --do napi-set \ + --json='{"id": 66, "threaded": "busy-poll"}' + +The kernel will create a kthread that busy polls on this NAPI. + +The user may elect to set the CPU affinity of this kthread to an unused CPU +core to improve how often the NAPI is polled at the expense of wasted CPU +cycles. Note that this will keep the CPU core busy with 100% usage. + +Once threaded busy polling is enabled for a NAPI, PID of the kthread can be +retrieved using Netlink so the affinity of the kthread can be set up. + +For example, the following script can be used to fetch the PID: + +.. code-block:: bash + + $ ynl --family netdev --do napi-get --json='{"id": 66}' + +This will output something like following, the pid `258` is the PID of the +kthread that is polling this NAPI. + +.. code-block:: bash + + $ {'defer-hard-irqs': 0, + 'gro-flush-timeout': 0, + 'id': 66, + 'ifindex': 2, + 'irq-suspend-timeout': 0, + 'pid': 258, + 'threaded': 'busy-poll'} + .. _threaded: Threaded NAPI diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 9c1e5042c5e7..e808071dbb7d 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -423,11 +423,12 @@ enum { NAPI_STATE_NPSVC, /* Netpoll - don't dequeue from poll_list */ NAPI_STATE_LISTED, /* NAPI added to system lists */ NAPI_STATE_NO_BUSY_POLL, /* Do not add in napi_hash, no busy polling */ - NAPI_STATE_IN_BUSY_POLL, /* sk_busy_loop() owns this NAPI */ + NAPI_STATE_IN_BUSY_POLL, /* Do not rearm NAPI interrupt */ NAPI_STATE_PREFER_BUSY_POLL, /* prefer busy-polling over softirq processing*/ NAPI_STATE_THREADED, /* The poll is performed inside its own thread*/ NAPI_STATE_SCHED_THREADED, /* Napi is currently scheduled in threaded mode */ NAPI_STATE_HAS_NOTIFIER, /* Napi has an IRQ notifier */ + NAPI_STATE_THREADED_BUSY_POLL, /* The threaded NAPI poller will busy poll */ }; enum { @@ -442,6 +443,7 @@ enum { NAPIF_STATE_THREADED = BIT(NAPI_STATE_THREADED), NAPIF_STATE_SCHED_THREADED = BIT(NAPI_STATE_SCHED_THREADED), NAPIF_STATE_HAS_NOTIFIER = BIT(NAPI_STATE_HAS_NOTIFIER), + NAPIF_STATE_THREADED_BUSY_POLL = BIT(NAPI_STATE_THREADED_BUSY_POLL), }; enum gro_result { diff --git a/include/uapi/linux/netdev.h b/include/uapi/linux/netdev.h index 48eb49aa03d4..048c8de1a130 100644 --- a/include/uapi/linux/netdev.h +++ b/include/uapi/linux/netdev.h @@ -80,6 +80,7 @@ enum netdev_qstats_scope { enum netdev_napi_threaded { NETDEV_NAPI_THREADED_DISABLED, NETDEV_NAPI_THREADED_ENABLED, + NETDEV_NAPI_THREADED_BUSY_POLL, }; enum { diff --git a/net/core/dev.c b/net/core/dev.c index dccc1176f3c6..2c1de5fb97d9 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -7089,7 +7089,8 @@ static void napi_stop_kthread(struct napi_struct *napi) */ if ((val & NAPIF_STATE_SCHED_THREADED) || !(val & NAPIF_STATE_SCHED)) { - new = val & (~NAPIF_STATE_THREADED); + new = val & (~(NAPIF_STATE_THREADED | + NAPIF_STATE_THREADED_BUSY_POLL)); } else { msleep(20); continue; @@ -7113,6 +7114,16 @@ static void napi_stop_kthread(struct napi_struct *napi) napi->thread = NULL; } +static void napi_set_threaded_state(struct napi_struct *napi, + enum netdev_napi_threaded threaded_mode) +{ + bool threaded = threaded_mode != NETDEV_NAPI_THREADED_DISABLED; + bool busy_poll = threaded_mode == NETDEV_NAPI_THREADED_BUSY_POLL; + + assign_bit(NAPI_STATE_THREADED, &napi->state, threaded); + assign_bit(NAPI_STATE_THREADED_BUSY_POLL, &napi->state, busy_poll); +} + int napi_set_threaded(struct napi_struct *napi, enum netdev_napi_threaded threaded) { @@ -7139,7 +7150,7 @@ int napi_set_threaded(struct napi_struct *napi, } else { /* Make sure kthread is created before THREADED bit is set. */ smp_mb__before_atomic(); - assign_bit(NAPI_STATE_THREADED, &napi->state, threaded); + napi_set_threaded_state(napi, threaded); } return 0; @@ -7531,7 +7542,9 @@ void napi_disable_locked(struct napi_struct *n) } new = val | NAPIF_STATE_SCHED | NAPIF_STATE_NPSVC; - new &= ~(NAPIF_STATE_THREADED | NAPIF_STATE_PREFER_BUSY_POLL); + new &= ~(NAPIF_STATE_THREADED | + NAPIF_STATE_THREADED_BUSY_POLL | + NAPIF_STATE_PREFER_BUSY_POLL); } while (!try_cmpxchg(&n->state, &val, new)); hrtimer_cancel(&n->timer); @@ -7743,7 +7756,7 @@ static int napi_thread_wait(struct napi_struct *napi) return -1; } -static void napi_threaded_poll_loop(struct napi_struct *napi) +static void napi_threaded_poll_loop(struct napi_struct *napi, bool busy_poll) { struct bpf_net_context __bpf_net_ctx, *bpf_net_ctx; struct softnet_data *sd; @@ -7772,22 +7785,47 @@ static void napi_threaded_poll_loop(struct napi_struct *napi) } skb_defer_free_flush(); bpf_net_ctx_clear(bpf_net_ctx); + + /* When busy poll is enabled, the old packets are not flushed in + * napi_complete_done. So flush them here. + */ + if (busy_poll) + gro_flush_normal(&napi->gro, HZ >= 1000); local_bh_enable(); + /* Call cond_resched here to avoid watchdog warnings. */ + if (repoll || busy_poll) { + rcu_softirq_qs_periodic(last_qs); + cond_resched(); + } + if (!repoll) break; - - rcu_softirq_qs_periodic(last_qs); - cond_resched(); } } static int napi_threaded_poll(void *data) { struct napi_struct *napi = data; + bool want_busy_poll; + bool in_busy_poll; + unsigned long val; - while (!napi_thread_wait(napi)) - napi_threaded_poll_loop(napi); + while (!napi_thread_wait(napi)) { + val = READ_ONCE(napi->state); + + want_busy_poll = val & NAPIF_STATE_THREADED_BUSY_POLL; + in_busy_poll = val & NAPIF_STATE_IN_BUSY_POLL; + + if (unlikely(val & NAPIF_STATE_DISABLE)) + want_busy_poll = false; + + if (want_busy_poll != in_busy_poll) + assign_bit(NAPI_STATE_IN_BUSY_POLL, &napi->state, + want_busy_poll); + + napi_threaded_poll_loop(napi, want_busy_poll); + } return 0; } @@ -13097,7 +13135,7 @@ static void run_backlog_napi(unsigned int cpu) { struct softnet_data *sd = per_cpu_ptr(&softnet_data, cpu); - napi_threaded_poll_loop(&sd->backlog); + napi_threaded_poll_loop(&sd->backlog, false); } static void backlog_napi_setup(unsigned int cpu) diff --git a/net/core/dev.h b/net/core/dev.h index 900880e8b5b4..4d872a79bafb 100644 --- a/net/core/dev.h +++ b/net/core/dev.h @@ -317,6 +317,9 @@ static inline void napi_set_irq_suspend_timeout(struct napi_struct *n, static inline enum netdev_napi_threaded napi_get_threaded(struct napi_struct *n) { + if (test_bit(NAPI_STATE_THREADED_BUSY_POLL, &n->state)) + return NETDEV_NAPI_THREADED_BUSY_POLL; + if (test_bit(NAPI_STATE_THREADED, &n->state)) return NETDEV_NAPI_THREADED_ENABLED; diff --git a/net/core/netdev-genl-gen.c b/net/core/netdev-genl-gen.c index e9a2a6f26cb7..ff20435c45d2 100644 --- a/net/core/netdev-genl-gen.c +++ b/net/core/netdev-genl-gen.c @@ -97,7 +97,7 @@ static const struct nla_policy netdev_napi_set_nl_policy[NETDEV_A_NAPI_THREADED [NETDEV_A_NAPI_DEFER_HARD_IRQS] = NLA_POLICY_FULL_RANGE(NLA_U32, &netdev_a_napi_defer_hard_irqs_range), [NETDEV_A_NAPI_GRO_FLUSH_TIMEOUT] = { .type = NLA_UINT, }, [NETDEV_A_NAPI_IRQ_SUSPEND_TIMEOUT] = { .type = NLA_UINT, }, - [NETDEV_A_NAPI_THREADED] = NLA_POLICY_MAX(NLA_U32, 1), + [NETDEV_A_NAPI_THREADED] = NLA_POLICY_MAX(NLA_U32, 2), }; /* NETDEV_CMD_BIND_TX - do */ diff --git a/tools/include/uapi/linux/netdev.h b/tools/include/uapi/linux/netdev.h index 48eb49aa03d4..048c8de1a130 100644 --- a/tools/include/uapi/linux/netdev.h +++ b/tools/include/uapi/linux/netdev.h @@ -80,6 +80,7 @@ enum netdev_qstats_scope { enum netdev_napi_threaded { NETDEV_NAPI_THREADED_DISABLED, NETDEV_NAPI_THREADED_ENABLED, + NETDEV_NAPI_THREADED_BUSY_POLL, }; enum { From add3c1324a8912b1abbf7afeb976fb719563f454 Mon Sep 17 00:00:00 2001 From: Samiullah Khawaja Date: Tue, 28 Oct 2025 20:30:06 +0000 Subject: [PATCH 0471/1381] selftests: Add napi threaded busy poll test in `busy_poller` Add testcase to run busy poll test with threaded napi busy poll enabled. Signed-off-by: Samiullah Khawaja Reviewed-by: Willem de Bruijn Acked-by: Martin Karsten Tested-by: Martin Karsten Link: https://patch.msgid.link/20251028203007.575686-3-skhawaja@google.com Signed-off-by: Jakub Kicinski --- tools/testing/selftests/net/busy_poll_test.sh | 24 ++++++++++++++++++- tools/testing/selftests/net/busy_poller.c | 16 ++++++++++--- 2 files changed, 36 insertions(+), 4 deletions(-) diff --git a/tools/testing/selftests/net/busy_poll_test.sh b/tools/testing/selftests/net/busy_poll_test.sh index 7d2d40812074..5ec1c85c1623 100755 --- a/tools/testing/selftests/net/busy_poll_test.sh +++ b/tools/testing/selftests/net/busy_poll_test.sh @@ -27,6 +27,8 @@ NAPI_DEFER_HARD_IRQS=100 GRO_FLUSH_TIMEOUT=50000 SUSPEND_TIMEOUT=20000000 +NAPI_THREADED_MODE_BUSY_POLL=2 + setup_ns() { set -e @@ -62,6 +64,9 @@ cleanup_ns() test_busypoll() { suspend_value=${1:-0} + napi_threaded_value=${2:-0} + prefer_busy_poll_value=${3:-$PREFER_BUSY_POLL} + tmp_file=$(mktemp) out_file=$(mktemp) @@ -73,10 +78,11 @@ test_busypoll() -b${SERVER_IP} \ -m${MAX_EVENTS} \ -u${BUSY_POLL_USECS} \ - -P${PREFER_BUSY_POLL} \ + -P${prefer_busy_poll_value} \ -g${BUSY_POLL_BUDGET} \ -i${NSIM_SV_IFIDX} \ -s${suspend_value} \ + -t${napi_threaded_value} \ -o${out_file}& wait_local_port_listen nssv ${SERVER_PORT} tcp @@ -109,6 +115,15 @@ test_busypoll_with_suspend() return $? } +test_busypoll_with_napi_threaded() +{ + # Only enable napi threaded poll. Set suspend timeout and prefer busy + # poll to 0. + test_busypoll 0 ${NAPI_THREADED_MODE_BUSY_POLL} 0 + + return $? +} + ### ### Code start ### @@ -154,6 +169,13 @@ if [ $? -ne 0 ]; then exit 1 fi +test_busypoll_with_napi_threaded +if [ $? -ne 0 ]; then + echo "test_busypoll_with_napi_threaded failed" + cleanup_ns + exit 1 +fi + echo "$NSIM_SV_FD:$NSIM_SV_IFIDX" > $NSIM_DEV_SYS_UNLINK echo $NSIM_CL_ID > $NSIM_DEV_SYS_DEL diff --git a/tools/testing/selftests/net/busy_poller.c b/tools/testing/selftests/net/busy_poller.c index 04c7ff577bb8..3a81f9c94795 100644 --- a/tools/testing/selftests/net/busy_poller.c +++ b/tools/testing/selftests/net/busy_poller.c @@ -65,15 +65,16 @@ static uint32_t cfg_busy_poll_usecs; static uint16_t cfg_busy_poll_budget; static uint8_t cfg_prefer_busy_poll; -/* IRQ params */ +/* NAPI params */ static uint32_t cfg_defer_hard_irqs; static uint64_t cfg_gro_flush_timeout; static uint64_t cfg_irq_suspend_timeout; +static enum netdev_napi_threaded cfg_napi_threaded_poll = NETDEV_NAPI_THREADED_DISABLED; static void usage(const char *filepath) { error(1, 0, - "Usage: %s -p -b -m -u -P -g -o -d -r -s -i", + "Usage: %s -p -b -m -u -P -g -o -d -r -s -t -i", filepath); } @@ -86,7 +87,7 @@ static void parse_opts(int argc, char **argv) if (argc <= 1) usage(argv[0]); - while ((c = getopt(argc, argv, "p:m:b:u:P:g:o:d:r:s:i:")) != -1) { + while ((c = getopt(argc, argv, "p:m:b:u:P:g:o:d:r:s:i:t:")) != -1) { /* most options take integer values, except o and b, so reduce * code duplication a bit for the common case by calling * strtoull here and leave bounds checking and casting per @@ -168,6 +169,12 @@ static void parse_opts(int argc, char **argv) cfg_ifindex = (int)tmp; break; + case 't': + if (tmp > 2) + error(1, ERANGE, "napi threaded poll value must be 0-2"); + + cfg_napi_threaded_poll = (enum netdev_napi_threaded)tmp; + break; } } @@ -247,6 +254,9 @@ static void setup_queue(void) netdev_napi_set_req_set_irq_suspend_timeout(set_req, cfg_irq_suspend_timeout); + if (cfg_napi_threaded_poll) + netdev_napi_set_req_set_threaded(set_req, cfg_napi_threaded_poll); + if (netdev_napi_set(ys, set_req)) error(1, 0, "can't set NAPI params: %s\n", yerr.msg); From abcf6eef90c6e47efed62a7c233ffc1a6a90797e Mon Sep 17 00:00:00 2001 From: Oleksij Rempel Date: Mon, 27 Oct 2025 13:27:58 +0100 Subject: [PATCH 0472/1381] net: phy: introduce internal API for PHY MSE diagnostics Add the base infrastructure for Mean Square Error (MSE) diagnostics, as proposed by the OPEN Alliance "Advanced diagnostic features for 100BASE-T1 automotive Ethernet PHYs" [1] specification. The OPEN Alliance spec defines only average MSE and average peak MSE over a fixed number of symbols. However, other PHYs, such as the KSZ9131, additionally expose a worst-peak MSE value latched since the last channel capture. This API accounts for such vendor extensions by adding a distinct capability bit and snapshot field. Channel-to-pair mapping is normally straightforward, but in some cases (e.g. 100BASE-TX with MDI-X resolution unknown) the mapping is ambiguous. If hardware does not expose MDI-X status, the exact pair cannot be determined. To avoid returning misleading per-channel data in this case, a LINK selector is defined for aggregate MSE measurements. All investigated devices differ in MSE capabilities, such as sample rate, number of analyzed symbols, and scaling factors. For example, the KSZ9131 uses different scaling for MSE and pMSE. To make this visible to callers, scale limits and timing information are returned via get_mse_capability(). Some PHYs sample very few symbols at high frequency (e.g. 2 us update rate). To cover such cases and allow for future high-speed PHYs with even shorter intervals, the refresh rate is reported as u64 in picoseconds. This patch introduces the internal PHY API for Mean Square Error diagnostics. It defines new kernel-side data types and driver hooks: - struct phy_mse_capability: describes supported metrics, scale limits, update interval, and sampling length. - struct phy_mse_snapshot: holds one correlated measurement set. - New phy_driver ops: `get_mse_capability()` and `get_mse_snapshot()`. These definitions form the core kernel API. No user-visible interfaces are added in this commit. Standardization notes: OPEN Alliance defines presence and interpretation of some metrics but does not fix numeric scales or sampling internals: - SQI (3-bit, 0..7) is mandatory; correlation to SNR/BER is informative (OA 100BASE-T1 TC1 v1.0 6.1.2; OA 1000BASE-T1 TC12 v2.2 6.1.2). - MSE is optional; OA recommends 2^16 symbols and scaling to 0..511, with a worst-case latch since last read (OA 100BASE-T1 TC1 v1.0 6.1.1; OA 1000BASE-T1 TC12 v2.2 6.1.1). Refresh is recommended (~0.8-2.0 ms for 100BASE-T1; ~80-200 us for 1000BASE-T1). Exact scaling/time windows are vendor-specific. - Peak MSE (pMSE) is defined only for 100BASE-T1 as optional, e.g. 128-symbol sliding window with 8-bit range and worst-case latch (OA 100BASE-T1 TC1 v1.0 6.1.3). Therefore this API exposes which measures and selectors a PHY supports, and documents where behavior is standard-referenced vs vendor-specific. [1] Signed-off-by: Oleksij Rempel Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/20251027122801.982364-2-o.rempel@pengutronix.de Signed-off-by: Jakub Kicinski --- include/linux/phy.h | 206 ++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 206 insertions(+) diff --git a/include/linux/phy.h b/include/linux/phy.h index 358dd6f0ff96..e3474f03cbc1 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -903,6 +903,165 @@ struct phy_led { #define to_phy_led(d) container_of(d, struct phy_led, led_cdev) +/* + * PHY_MSE_CAP_* - Bitmask flags for Mean Square Error (MSE) capabilities + * + * These flags describe which MSE metrics and selectors are implemented + * by the PHY for the current link mode. They are used in + * struct phy_mse_capability.supported_caps. + * + * Standardization: + * The OPEN Alliance (OA) defines the presence of MSE/SQI/pMSE but not their + * numeric scaling, update intervals, or aggregation windows. See: + * OA 100BASE-T1 TC1 v1.0, sections 6.1.1-6.1.3 + * OA 1000BASE-T1 TC12 v2.2, sections 6.1.1-6.1.2 + * + * Description of flags: + * + * PHY_MSE_CAP_CHANNEL_A + * Per-pair diagnostics for Channel A are supported. Mapping to the + * physical wire pair may depend on MDI/MDI-X polarity. + * + * PHY_MSE_CAP_CHANNEL_B, _C, _D + * Same as above for channels B-D. + * + * PHY_MSE_CAP_WORST_CHANNEL + * The PHY or driver can identify and report the single worst-performing + * channel without querying each one individually. + * + * PHY_MSE_CAP_LINK + * The PHY provides only a link-wide aggregate measurement or cannot map + * results to a specific pair (for example 100BASE-TX with unknown + * MDI/MDI-X). + * + * PHY_MSE_CAP_AVG + * Average MSE (mean DCQ metric) is supported. For 100/1000BASE-T1 the OA + * recommends 2^16 symbols, scaled 0..511, but the exact scaling is + * vendor-specific. + * + * PHY_MSE_CAP_PEAK + * Peak MSE (current peak within the measurement window) is supported. + * Defined as pMSE for 100BASE-T1; vendor-specific for others. + * + * PHY_MSE_CAP_WORST_PEAK + * Latched worst-case peak MSE since the last read (read-to-clear if + * implemented). Optional in OA 100BASE-T1 TC1 6.1.3. + */ +#define PHY_MSE_CAP_CHANNEL_A BIT(0) +#define PHY_MSE_CAP_CHANNEL_B BIT(1) +#define PHY_MSE_CAP_CHANNEL_C BIT(2) +#define PHY_MSE_CAP_CHANNEL_D BIT(3) +#define PHY_MSE_CAP_WORST_CHANNEL BIT(4) +#define PHY_MSE_CAP_LINK BIT(5) +#define PHY_MSE_CAP_AVG BIT(6) +#define PHY_MSE_CAP_PEAK BIT(7) +#define PHY_MSE_CAP_WORST_PEAK BIT(8) + +/* + * enum phy_mse_channel - Identifiers for selecting MSE measurement channels + * + * PHY_MSE_CHANNEL_A - PHY_MSE_CHANNEL_D + * Select per-pair measurement for the corresponding channel. + * + * PHY_MSE_CHANNEL_WORST + * Select the single worst-performing channel reported by hardware. + * + * PHY_MSE_CHANNEL_LINK + * Select link-wide aggregate data (used when per-pair results are + * unavailable). + */ +enum phy_mse_channel { + PHY_MSE_CHANNEL_A, + PHY_MSE_CHANNEL_B, + PHY_MSE_CHANNEL_C, + PHY_MSE_CHANNEL_D, + PHY_MSE_CHANNEL_WORST, + PHY_MSE_CHANNEL_LINK, +}; + +/** + * struct phy_mse_capability - Capabilities of Mean Square Error (MSE) + * measurement interface + * + * Standardization notes: + * + * - Presence of MSE/SQI/pMSE is defined by OPEN Alliance specs, but numeric + * scaling, refresh/update rate and aggregation windows are not fixed and + * are vendor-/product-specific. (OA 100BASE-T1 TC1 v1.0 6.1.*; + * OA 1000BASE-T1 TC12 v2.2 6.1.*) + * + * - Typical recommendations: 2^16 symbols and 0..511 scaling for MSE; pMSE only + * defined for 100BASE-T1 (sliding window example), others are vendor + * extensions. Drivers must report actual scale/limits here. + * + * Describes the MSE measurement capabilities for the current link mode. These + * properties are dynamic and may change when link settings are modified. + * Callers should re-query this capability after any link state change to + * ensure they have the most up-to-date information. + * + * Callers should only request measurements for channels and types that are + * indicated as supported by the @supported_caps bitmask. If @supported_caps + * is 0, the device provides no MSE diagnostics, and driver operations should + * typically return -EOPNOTSUPP. + * + * Snapshot values for average and peak MSE can be normalized to a 0..1 ratio + * by dividing the raw snapshot by the corresponding @max_average_mse or + * @max_peak_mse value. + * + * @max_average_mse: The maximum value for an average MSE snapshot. This + * defines the scale for the measurement. If the PHY_MSE_CAP_AVG capability is + * supported, this value MUST be greater than 0. (vendor-specific units). + * @max_peak_mse: The maximum value for a peak MSE snapshot. If either + * PHY_MSE_CAP_PEAK or PHY_MSE_CAP_WORST_PEAK is supported, this value MUST + * be greater than 0. (vendor-specific units). + * @refresh_rate_ps: The typical interval, in picoseconds, between hardware + * updates of the MSE values. This is an estimate, and callers should not + * assume synchronous sampling. (vendor-specific units). + * @num_symbols: The number of symbols aggregated per hardware sample to + * calculate the MSE. (vendor-specific units). + * @supported_caps: A bitmask of PHY_MSE_CAP_* values indicating which + * measurement types (e.g., average, peak) and channels + * (e.g., per-pair or link-wide) are supported. + */ +struct phy_mse_capability { + u64 max_average_mse; + u64 max_peak_mse; + u64 refresh_rate_ps; + u64 num_symbols; + u32 supported_caps; +}; + +/** + * struct phy_mse_snapshot - A snapshot of Mean Square Error (MSE) diagnostics + * + * Holds a set of MSE diagnostic values that were all captured from a single + * measurement window. + * + * Values are raw, device-scaled and not normalized. Use struct + * phy_mse_capability to interpret the scale and sampling window. + * + * @average_mse: The average MSE value over the measurement window. + * OPEN Alliance references MSE as a DCQ metric; recommends 2^16 symbols and + * 0..511 scaling. Exact scale and refresh are vendor-specific. + * (100BASE-T1 TC1 v1.0 6.1.1; 1000BASE-T1 TC12 v2.2 6.1.1). + * + * @peak_mse: The peak MSE value observed within the measurement window. + * For 100BASE-T1, "pMSE" is optional and may be implemented via a sliding + * 128-symbol window with periodic capture; not standardized for 1000BASE-T1. + * (100BASE-T1 TC1 v1.0 6.1.3, Table "DCQ.peakMSE"). + * + * @worst_peak_mse: A latched high-water mark of the peak MSE since last read + * (read-to-clear if implemented). OPEN Alliance shows a latched "worst case + * peak MSE" for 100BASE-T1 pMSE; availability/semantics outside that are + * vendor-specific. (100BASE-T1 TC1 v1.0 6.1.3, DCQ.peakMSE high byte; + * 1000BASE-T1 TC12 v2.2 treats DCQ details as vendor-specific.) + */ +struct phy_mse_snapshot { + u64 average_mse; + u64 peak_mse; + u64 worst_peak_mse; +}; + /** * struct phy_driver - Driver structure for a particular PHY type * @@ -1184,6 +1343,53 @@ struct phy_driver { /** @get_sqi_max: Get the maximum signal quality indication */ int (*get_sqi_max)(struct phy_device *dev); + /** + * @get_mse_capability: Get capabilities and scale of MSE measurement + * @dev: PHY device + * @cap: Output (filled on success) + * + * Fill @cap with the PHY's MSE capability for the current + * link mode: scale limits (max_average_mse, max_peak_mse), update + * interval (refresh_rate_ps), sample length (num_symbols) and the + * capability bitmask (supported_caps). + * + * Implementations may defer capability report until hardware has + * converged; in that case they should return -EAGAIN and allow the + * caller to retry later. + * + * Return: 0 on success. On failure, returns a negative errno code, such + * as -EOPNOTSUPP if MSE measurement is not supported by the PHY or in + * the current link mode, or -EAGAIN if the capability information is + * not yet available. + */ + int (*get_mse_capability)(struct phy_device *dev, + struct phy_mse_capability *cap); + + /** + * @get_mse_snapshot: Retrieve a snapshot of MSE diagnostic values + * @dev: PHY device + * @channel: Channel identifier (PHY_MSE_CHANNEL_*) + * @snapshot: Output (filled on success) + * + * Fill @snapshot with a correlated set of MSE values from the most + * recent measurement window. + * + * Callers must validate @channel against supported_caps returned by + * get_mse_capability(). Drivers must not coerce @channel; if the + * requested selector is not implemented by the device or current link + * mode, the operation must fail. + * + * worst_peak_mse is latched and must be treated as read-to-clear. + * + * Return: 0 on success. On failure, returns a negative errno code, such + * as -EOPNOTSUPP if MSE measurement is not supported by the PHY or in + * the current link mode, or -EAGAIN if measurements are not yet + * available. + */ + int (*get_mse_snapshot)(struct phy_device *dev, + enum phy_mse_channel channel, + struct phy_mse_snapshot *snapshot); + /* PLCA RS interface */ /** @get_plca_cfg: Return the current PLCA configuration */ int (*get_plca_cfg)(struct phy_device *dev, From e6e93fb01302e9b7a15d17f3b8a00eff8a601654 Mon Sep 17 00:00:00 2001 From: Oleksij Rempel Date: Mon, 27 Oct 2025 13:27:59 +0100 Subject: [PATCH 0473/1381] ethtool: netlink: add ETHTOOL_MSG_MSE_GET and wire up PHY MSE access Introduce the userspace entry point for PHY MSE diagnostics via ethtool netlink. This exposes the core API added previously and returns both capability information and one or more snapshots. Userspace sends ETHTOOL_MSG_MSE_GET. The reply carries: - ETHTOOL_A_MSE_CAPABILITIES: scale limits and timing information - ETHTOOL_A_MSE_CHANNEL_* nests: one or more snapshots (per-channel if available, otherwise WORST, otherwise LINK) Link down returns -ENETDOWN. Changes: - YAML: add attribute sets (mse, mse-capabilities, mse-snapshot) and the mse-get operation - UAPI (generated): add ETHTOOL_A_MSE_* enums and message IDs, ETHTOOL_MSG_MSE_GET/REPLY - ethtool core: add net/ethtool/mse.c implementing the request, register genl op, and hook into ethnl dispatch - docs: document MSE_GET in ethtool-netlink.rst The include/uapi/linux/ethtool_netlink_generated.h is generated from Documentation/netlink/specs/ethtool.yaml. Signed-off-by: Oleksij Rempel Link: https://patch.msgid.link/20251027122801.982364-3-o.rempel@pengutronix.de Signed-off-by: Jakub Kicinski --- Documentation/netlink/specs/ethtool.yaml | 86 +++++ Documentation/networking/ethtool-netlink.rst | 64 ++++ .../uapi/linux/ethtool_netlink_generated.h | 35 ++ net/ethtool/Makefile | 2 +- net/ethtool/mse.c | 329 ++++++++++++++++++ net/ethtool/netlink.c | 10 + net/ethtool/netlink.h | 2 + 7 files changed, 527 insertions(+), 1 deletion(-) create mode 100644 net/ethtool/mse.c diff --git a/Documentation/netlink/specs/ethtool.yaml b/Documentation/netlink/specs/ethtool.yaml index 6a0fb1974513..05d2b6508b59 100644 --- a/Documentation/netlink/specs/ethtool.yaml +++ b/Documentation/netlink/specs/ethtool.yaml @@ -1823,6 +1823,73 @@ attribute-sets: type: uint enum: pse-event doc: List of events reported by the PSE controller + - + name: mse-capabilities + doc: MSE capabilities attribute set + attr-cnt-name: --ethtool-a-mse-capabilities-cnt + attributes: + - + name: max-average-mse + type: uint + - + name: max-peak-mse + type: uint + - + name: refresh-rate-ps + type: uint + - + name: num-symbols + type: uint + - + name: mse-snapshot + doc: MSE snapshot attribute set + attr-cnt-name: --ethtool-a-mse-snapshot-cnt + attributes: + - + name: average-mse + type: uint + - + name: peak-mse + type: uint + - + name: worst-peak-mse + type: uint + - + name: mse + attr-cnt-name: --ethtool-a-mse-cnt + attributes: + - + name: header + type: nest + nested-attributes: header + - + name: capabilities + type: nest + nested-attributes: mse-capabilities + - + name: channel-a + type: nest + nested-attributes: mse-snapshot + - + name: channel-b + type: nest + nested-attributes: mse-snapshot + - + name: channel-c + type: nest + nested-attributes: mse-snapshot + - + name: channel-d + type: nest + nested-attributes: mse-snapshot + - + name: worst-channel + type: nest + nested-attributes: mse-snapshot + - + name: link + type: nest + nested-attributes: mse-snapshot operations: enum-model: directional @@ -2756,6 +2823,25 @@ operations: attributes: - header - context + - + name: mse-get + doc: Get PHY MSE measurement data and capabilities. + attribute-set: mse + do: &mse-get-op + request: + attributes: + - header + reply: + attributes: + - header + - capabilities + - channel-a + - channel-b + - channel-c + - channel-d + - worst-channel + - link + dump: *mse-get-op mcast-groups: list: diff --git a/Documentation/networking/ethtool-netlink.rst b/Documentation/networking/ethtool-netlink.rst index b270886c5f5d..af56c304cef4 100644 --- a/Documentation/networking/ethtool-netlink.rst +++ b/Documentation/networking/ethtool-netlink.rst @@ -242,6 +242,7 @@ Userspace to kernel: ``ETHTOOL_MSG_RSS_SET`` set RSS settings ``ETHTOOL_MSG_RSS_CREATE_ACT`` create an additional RSS context ``ETHTOOL_MSG_RSS_DELETE_ACT`` delete an additional RSS context + ``ETHTOOL_MSG_MSE_GET`` get MSE diagnostic data ===================================== ================================= Kernel to userspace: @@ -299,6 +300,7 @@ Kernel to userspace: ``ETHTOOL_MSG_RSS_CREATE_ACT_REPLY`` create an additional RSS context ``ETHTOOL_MSG_RSS_CREATE_NTF`` additional RSS context created ``ETHTOOL_MSG_RSS_DELETE_NTF`` additional RSS context deleted + ``ETHTOOL_MSG_MSE_GET_REPLY`` MSE diagnostic data ======================================== ================================= ``GET`` requests are sent by userspace applications to retrieve device @@ -2458,6 +2460,68 @@ Kernel response contents: For a description of each attribute, see ``TSCONFIG_GET``. +MSE_GET +======= + +Retrieves detailed Mean Square Error (MSE) diagnostic information from the PHY. + +Request Contents: + + ==================================== ====== ============================ + ``ETHTOOL_A_MSE_HEADER`` nested request header + ==================================== ====== ============================ + +Kernel Response Contents: + + ==================================== ====== ================================ + ``ETHTOOL_A_MSE_HEADER`` nested reply header + ``ETHTOOL_A_MSE_CAPABILITIES`` nested capability/scale info for MSE + measurements + ``ETHTOOL_A_MSE_CHANNEL_A`` nested snapshot for Channel A + ``ETHTOOL_A_MSE_CHANNEL_B`` nested snapshot for Channel B + ``ETHTOOL_A_MSE_CHANNEL_C`` nested snapshot for Channel C + ``ETHTOOL_A_MSE_CHANNEL_D`` nested snapshot for Channel D + ``ETHTOOL_A_MSE_WORST_CHANNEL`` nested snapshot for worst channel + ``ETHTOOL_A_MSE_LINK`` nested snapshot for link-wide aggregate + ==================================== ====== ================================ + +MSE Capabilities +---------------- + +This nested attribute reports the capability / scaling properties used to +interpret snapshot values. + + ============================================== ====== ========================= + ``ETHTOOL_A_MSE_CAPABILITIES_MAX_AVERAGE_MSE`` uint max avg_mse scale + ``ETHTOOL_A_MSE_CAPABILITIES_MAX_PEAK_MSE`` uint max peak_mse scale + ``ETHTOOL_A_MSE_CAPABILITIES_REFRESH_RATE_PS`` uint sample rate (picoseconds) + ``ETHTOOL_A_MSE_CAPABILITIES_NUM_SYMBOLS`` uint symbols per HW sample + ============================================== ====== ========================= + +The max-average/peak fields are included only if the corresponding metric +is supported by the PHY. Their absence indicates that the metric is not +available. + +See ``struct phy_mse_capability`` kernel documentation in +``include/linux/phy.h``. + +MSE Snapshot +------------ + +Each per-channel nest contains an atomic snapshot of MSE values for that +selector (channel A/B/C/D, worst channel, or link). + + ========================================== ====== =================== + ``ETHTOOL_A_MSE_SNAPSHOT_AVERAGE_MSE`` uint average MSE value + ``ETHTOOL_A_MSE_SNAPSHOT_PEAK_MSE`` uint current peak MSE + ``ETHTOOL_A_MSE_SNAPSHOT_WORST_PEAK_MSE`` uint worst-case peak MSE + ========================================== ====== =================== + +Within each channel nest, only the metrics supported by the PHY will be present. + +See ``struct phy_mse_snapshot`` kernel documentation in +``include/linux/phy.h``. + Request translation =================== diff --git a/include/uapi/linux/ethtool_netlink_generated.h b/include/uapi/linux/ethtool_netlink_generated.h index 0e8ac0d974e2..b71b175df46d 100644 --- a/include/uapi/linux/ethtool_netlink_generated.h +++ b/include/uapi/linux/ethtool_netlink_generated.h @@ -803,6 +803,39 @@ enum { ETHTOOL_A_PSE_NTF_MAX = (__ETHTOOL_A_PSE_NTF_CNT - 1) }; +enum { + ETHTOOL_A_MSE_CAPABILITIES_MAX_AVERAGE_MSE = 1, + ETHTOOL_A_MSE_CAPABILITIES_MAX_PEAK_MSE, + ETHTOOL_A_MSE_CAPABILITIES_REFRESH_RATE_PS, + ETHTOOL_A_MSE_CAPABILITIES_NUM_SYMBOLS, + + __ETHTOOL_A_MSE_CAPABILITIES_CNT, + ETHTOOL_A_MSE_CAPABILITIES_MAX = (__ETHTOOL_A_MSE_CAPABILITIES_CNT - 1) +}; + +enum { + ETHTOOL_A_MSE_SNAPSHOT_AVERAGE_MSE = 1, + ETHTOOL_A_MSE_SNAPSHOT_PEAK_MSE, + ETHTOOL_A_MSE_SNAPSHOT_WORST_PEAK_MSE, + + __ETHTOOL_A_MSE_SNAPSHOT_CNT, + ETHTOOL_A_MSE_SNAPSHOT_MAX = (__ETHTOOL_A_MSE_SNAPSHOT_CNT - 1) +}; + +enum { + ETHTOOL_A_MSE_HEADER = 1, + ETHTOOL_A_MSE_CAPABILITIES, + ETHTOOL_A_MSE_CHANNEL_A, + ETHTOOL_A_MSE_CHANNEL_B, + ETHTOOL_A_MSE_CHANNEL_C, + ETHTOOL_A_MSE_CHANNEL_D, + ETHTOOL_A_MSE_WORST_CHANNEL, + ETHTOOL_A_MSE_LINK, + + __ETHTOOL_A_MSE_CNT, + ETHTOOL_A_MSE_MAX = (__ETHTOOL_A_MSE_CNT - 1) +}; + enum { ETHTOOL_MSG_USER_NONE = 0, ETHTOOL_MSG_STRSET_GET = 1, @@ -855,6 +888,7 @@ enum { ETHTOOL_MSG_RSS_SET, ETHTOOL_MSG_RSS_CREATE_ACT, ETHTOOL_MSG_RSS_DELETE_ACT, + ETHTOOL_MSG_MSE_GET, __ETHTOOL_MSG_USER_CNT, ETHTOOL_MSG_USER_MAX = (__ETHTOOL_MSG_USER_CNT - 1) @@ -915,6 +949,7 @@ enum { ETHTOOL_MSG_RSS_CREATE_ACT_REPLY, ETHTOOL_MSG_RSS_CREATE_NTF, ETHTOOL_MSG_RSS_DELETE_NTF, + ETHTOOL_MSG_MSE_GET_REPLY, __ETHTOOL_MSG_KERNEL_CNT, ETHTOOL_MSG_KERNEL_MAX = (__ETHTOOL_MSG_KERNEL_CNT - 1) diff --git a/net/ethtool/Makefile b/net/ethtool/Makefile index 1e493553b977..629c10916670 100644 --- a/net/ethtool/Makefile +++ b/net/ethtool/Makefile @@ -9,4 +9,4 @@ ethtool_nl-y := netlink.o bitset.o strset.o linkinfo.o linkmodes.o rss.o \ channels.o coalesce.o pause.o eee.o tsinfo.o cabletest.o \ tunnels.o fec.o eeprom.o stats.o phc_vclocks.o mm.o \ module.o cmis_fw_update.o cmis_cdb.o pse-pd.o plca.o \ - phy.o tsconfig.o + phy.o tsconfig.o mse.o diff --git a/net/ethtool/mse.c b/net/ethtool/mse.c new file mode 100644 index 000000000000..6aac004c3ffc --- /dev/null +++ b/net/ethtool/mse.c @@ -0,0 +1,329 @@ +// SPDX-License-Identifier: GPL-2.0-only + +#include +#include +#include + +#include "netlink.h" +#include "common.h" + +/* Channels A-D only; WORST and LINK are exclusive alternatives */ +#define PHY_MSE_CHANNEL_COUNT 4 + +struct mse_req_info { + struct ethnl_req_info base; +}; + +struct mse_snapshot_entry { + struct phy_mse_snapshot snapshot; + int channel; +}; + +struct mse_reply_data { + struct ethnl_reply_data base; + struct phy_mse_capability capability; + struct mse_snapshot_entry *snapshots; + unsigned int num_snapshots; +}; + +static struct mse_reply_data * +mse_repdata(const struct ethnl_reply_data *reply_base) +{ + return container_of(reply_base, struct mse_reply_data, base); +} + +const struct nla_policy ethnl_mse_get_policy[] = { + [ETHTOOL_A_MSE_HEADER] = NLA_POLICY_NESTED(ethnl_header_policy_phy), +}; + +static int get_snapshot_if_supported(struct phy_device *phydev, + struct mse_reply_data *data, + unsigned int *idx, u32 cap_bit, + enum phy_mse_channel channel) +{ + int ret; + + if (data->capability.supported_caps & cap_bit) { + ret = phydev->drv->get_mse_snapshot(phydev, channel, + &data->snapshots[*idx].snapshot); + if (ret) + return ret; + data->snapshots[*idx].channel = channel; + (*idx)++; + } + + return 0; +} + +static int mse_get_channels(struct phy_device *phydev, + struct mse_reply_data *data) +{ + unsigned int i = 0; + int ret; + + if (!data->capability.supported_caps) + return 0; + + data->snapshots = kcalloc(PHY_MSE_CHANNEL_COUNT, + sizeof(*data->snapshots), GFP_KERNEL); + if (!data->snapshots) + return -ENOMEM; + + /* Priority 1: Individual channels */ + ret = get_snapshot_if_supported(phydev, data, &i, PHY_MSE_CAP_CHANNEL_A, + PHY_MSE_CHANNEL_A); + if (ret) + return ret; + ret = get_snapshot_if_supported(phydev, data, &i, PHY_MSE_CAP_CHANNEL_B, + PHY_MSE_CHANNEL_B); + if (ret) + return ret; + ret = get_snapshot_if_supported(phydev, data, &i, PHY_MSE_CAP_CHANNEL_C, + PHY_MSE_CHANNEL_C); + if (ret) + return ret; + ret = get_snapshot_if_supported(phydev, data, &i, PHY_MSE_CAP_CHANNEL_D, + PHY_MSE_CHANNEL_D); + if (ret) + return ret; + + /* If any individual channels were found, we are done. */ + if (i > 0) { + data->num_snapshots = i; + return 0; + } + + /* Priority 2: Worst channel, if no individual channels supported. */ + ret = get_snapshot_if_supported(phydev, data, &i, + PHY_MSE_CAP_WORST_CHANNEL, + PHY_MSE_CHANNEL_WORST); + if (ret) + return ret; + + /* If worst channel was found, we are done. */ + if (i > 0) { + data->num_snapshots = i; + return 0; + } + + /* Priority 3: Link-wide, if nothing else is supported. */ + ret = get_snapshot_if_supported(phydev, data, &i, PHY_MSE_CAP_LINK, + PHY_MSE_CHANNEL_LINK); + if (ret) + return ret; + + data->num_snapshots = i; + return 0; +} + +static int mse_prepare_data(const struct ethnl_req_info *req_base, + struct ethnl_reply_data *reply_base, + const struct genl_info *info) +{ + struct mse_reply_data *data = mse_repdata(reply_base); + struct net_device *dev = reply_base->dev; + struct phy_device *phydev; + int ret; + + phydev = ethnl_req_get_phydev(req_base, info->attrs, + ETHTOOL_A_MSE_HEADER, info->extack); + if (IS_ERR(phydev)) + return PTR_ERR(phydev); + if (!phydev) + return -EOPNOTSUPP; + + ret = ethnl_ops_begin(dev); + if (ret) + return ret; + + mutex_lock(&phydev->lock); + + if (!phydev->drv || !phydev->drv->get_mse_capability || + !phydev->drv->get_mse_snapshot) { + ret = -EOPNOTSUPP; + goto out_unlock; + } + if (!phydev->link) { + ret = -ENETDOWN; + goto out_unlock; + } + + ret = phydev->drv->get_mse_capability(phydev, &data->capability); + if (ret) + goto out_unlock; + + ret = mse_get_channels(phydev, data); + +out_unlock: + mutex_unlock(&phydev->lock); + ethnl_ops_complete(dev); + if (ret) + kfree(data->snapshots); + return ret; +} + +static void mse_cleanup_data(struct ethnl_reply_data *reply_base) +{ + struct mse_reply_data *data = mse_repdata(reply_base); + + kfree(data->snapshots); +} + +static int mse_reply_size(const struct ethnl_req_info *req_base, + const struct ethnl_reply_data *reply_base) +{ + const struct mse_reply_data *data = mse_repdata(reply_base); + size_t len = 0; + unsigned int i; + + /* ETHTOOL_A_MSE_CAPABILITIES */ + len += nla_total_size(0); + if (data->capability.supported_caps & PHY_MSE_CAP_AVG) + /* ETHTOOL_A_MSE_CAPABILITIES_MAX_AVERAGE_MSE */ + len += nla_total_size(sizeof(u64)); + if (data->capability.supported_caps & (PHY_MSE_CAP_PEAK | + PHY_MSE_CAP_WORST_PEAK)) + /* ETHTOOL_A_MSE_CAPABILITIES_MAX_PEAK_MSE */ + len += nla_total_size(sizeof(u64)); + /* ETHTOOL_A_MSE_CAPABILITIES_REFRESH_RATE_PS */ + len += nla_total_size(sizeof(u64)); + /* ETHTOOL_A_MSE_CAPABILITIES_NUM_SYMBOLS */ + len += nla_total_size(sizeof(u64)); + + for (i = 0; i < data->num_snapshots; i++) { + size_t snapshot_len = 0; + + /* Per-channel nest (e.g., ETHTOOL_A_MSE_CHANNEL_A / _B / _C / + * _D / _WORST_CHANNEL / _LINK) + */ + snapshot_len += nla_total_size(0); + + if (data->capability.supported_caps & PHY_MSE_CAP_AVG) + snapshot_len += nla_total_size(sizeof(u64)); + if (data->capability.supported_caps & PHY_MSE_CAP_PEAK) + snapshot_len += nla_total_size(sizeof(u64)); + if (data->capability.supported_caps & PHY_MSE_CAP_WORST_PEAK) + snapshot_len += nla_total_size(sizeof(u64)); + + len += snapshot_len; + } + + return len; +} + +static int mse_channel_to_attr(int ch) +{ + switch (ch) { + case PHY_MSE_CHANNEL_A: + return ETHTOOL_A_MSE_CHANNEL_A; + case PHY_MSE_CHANNEL_B: + return ETHTOOL_A_MSE_CHANNEL_B; + case PHY_MSE_CHANNEL_C: + return ETHTOOL_A_MSE_CHANNEL_C; + case PHY_MSE_CHANNEL_D: + return ETHTOOL_A_MSE_CHANNEL_D; + case PHY_MSE_CHANNEL_WORST: + return ETHTOOL_A_MSE_WORST_CHANNEL; + case PHY_MSE_CHANNEL_LINK: + return ETHTOOL_A_MSE_LINK; + default: + return -EINVAL; + } +} + +static int mse_fill_reply(struct sk_buff *skb, + const struct ethnl_req_info *req_base, + const struct ethnl_reply_data *reply_base) +{ + const struct mse_reply_data *data = mse_repdata(reply_base); + struct nlattr *nest; + unsigned int i; + int ret; + + nest = nla_nest_start(skb, ETHTOOL_A_MSE_CAPABILITIES); + if (!nest) + return -EMSGSIZE; + + if (data->capability.supported_caps & PHY_MSE_CAP_AVG) { + ret = nla_put_uint(skb, + ETHTOOL_A_MSE_CAPABILITIES_MAX_AVERAGE_MSE, + data->capability.max_average_mse); + if (ret < 0) + goto nla_put_nest_failure; + } + + if (data->capability.supported_caps & (PHY_MSE_CAP_PEAK | + PHY_MSE_CAP_WORST_PEAK)) { + ret = nla_put_uint(skb, ETHTOOL_A_MSE_CAPABILITIES_MAX_PEAK_MSE, + data->capability.max_peak_mse); + if (ret < 0) + goto nla_put_nest_failure; + } + + ret = nla_put_uint(skb, ETHTOOL_A_MSE_CAPABILITIES_REFRESH_RATE_PS, + data->capability.refresh_rate_ps); + if (ret < 0) + goto nla_put_nest_failure; + + ret = nla_put_uint(skb, ETHTOOL_A_MSE_CAPABILITIES_NUM_SYMBOLS, + data->capability.num_symbols); + if (ret < 0) + goto nla_put_nest_failure; + + nla_nest_end(skb, nest); + + for (i = 0; i < data->num_snapshots; i++) { + const struct mse_snapshot_entry *s = &data->snapshots[i]; + int chan_attr; + + chan_attr = mse_channel_to_attr(s->channel); + if (chan_attr < 0) + return chan_attr; + + nest = nla_nest_start(skb, chan_attr); + if (!nest) + return -EMSGSIZE; + + if (data->capability.supported_caps & PHY_MSE_CAP_AVG) { + ret = nla_put_uint(skb, + ETHTOOL_A_MSE_SNAPSHOT_AVERAGE_MSE, + s->snapshot.average_mse); + if (ret) + goto nla_put_nest_failure; + } + if (data->capability.supported_caps & PHY_MSE_CAP_PEAK) { + ret = nla_put_uint(skb, ETHTOOL_A_MSE_SNAPSHOT_PEAK_MSE, + s->snapshot.peak_mse); + if (ret) + goto nla_put_nest_failure; + } + if (data->capability.supported_caps & PHY_MSE_CAP_WORST_PEAK) { + ret = nla_put_uint(skb, + ETHTOOL_A_MSE_SNAPSHOT_WORST_PEAK_MSE, + s->snapshot.worst_peak_mse); + if (ret) + goto nla_put_nest_failure; + } + + nla_nest_end(skb, nest); + } + + return 0; + +nla_put_nest_failure: + nla_nest_cancel(skb, nest); + return ret; +} + +const struct ethnl_request_ops ethnl_mse_request_ops = { + .request_cmd = ETHTOOL_MSG_MSE_GET, + .reply_cmd = ETHTOOL_MSG_MSE_GET_REPLY, + .hdr_attr = ETHTOOL_A_MSE_HEADER, + .req_info_size = sizeof(struct mse_req_info), + .reply_data_size = sizeof(struct mse_reply_data), + + .prepare_data = mse_prepare_data, + .cleanup_data = mse_cleanup_data, + .reply_size = mse_reply_size, + .fill_reply = mse_fill_reply, +}; diff --git a/net/ethtool/netlink.c b/net/ethtool/netlink.c index 2f813f25f07e..6e5f0f4f815a 100644 --- a/net/ethtool/netlink.c +++ b/net/ethtool/netlink.c @@ -420,6 +420,7 @@ ethnl_default_requests[__ETHTOOL_MSG_USER_CNT] = { [ETHTOOL_MSG_TSCONFIG_GET] = ðnl_tsconfig_request_ops, [ETHTOOL_MSG_TSCONFIG_SET] = ðnl_tsconfig_request_ops, [ETHTOOL_MSG_PHY_GET] = ðnl_phy_request_ops, + [ETHTOOL_MSG_MSE_GET] = ðnl_mse_request_ops, }; static struct ethnl_dump_ctx *ethnl_dump_context(struct netlink_callback *cb) @@ -1534,6 +1535,15 @@ static const struct genl_ops ethtool_genl_ops[] = { .policy = ethnl_rss_delete_policy, .maxattr = ARRAY_SIZE(ethnl_rss_delete_policy) - 1, }, + { + .cmd = ETHTOOL_MSG_MSE_GET, + .doit = ethnl_default_doit, + .start = ethnl_perphy_start, + .dumpit = ethnl_perphy_dumpit, + .done = ethnl_perphy_done, + .policy = ethnl_mse_get_policy, + .maxattr = ARRAY_SIZE(ethnl_mse_get_policy) - 1, + }, }; static const struct genl_multicast_group ethtool_nl_mcgrps[] = { diff --git a/net/ethtool/netlink.h b/net/ethtool/netlink.h index 1d4f9ecb3d26..89010eaa67df 100644 --- a/net/ethtool/netlink.h +++ b/net/ethtool/netlink.h @@ -442,6 +442,7 @@ extern const struct ethnl_request_ops ethnl_plca_status_request_ops; extern const struct ethnl_request_ops ethnl_mm_request_ops; extern const struct ethnl_request_ops ethnl_phy_request_ops; extern const struct ethnl_request_ops ethnl_tsconfig_request_ops; +extern const struct ethnl_request_ops ethnl_mse_request_ops; extern const struct nla_policy ethnl_header_policy[ETHTOOL_A_HEADER_FLAGS + 1]; extern const struct nla_policy ethnl_header_policy_stats[ETHTOOL_A_HEADER_FLAGS + 1]; @@ -497,6 +498,7 @@ extern const struct nla_policy ethnl_module_fw_flash_act_policy[ETHTOOL_A_MODULE extern const struct nla_policy ethnl_phy_get_policy[ETHTOOL_A_PHY_HEADER + 1]; extern const struct nla_policy ethnl_tsconfig_get_policy[ETHTOOL_A_TSCONFIG_HEADER + 1]; extern const struct nla_policy ethnl_tsconfig_set_policy[ETHTOOL_A_TSCONFIG_MAX + 1]; +extern const struct nla_policy ethnl_mse_get_policy[ETHTOOL_A_MSE_HEADER + 1]; int ethnl_set_features(struct sk_buff *skb, struct genl_info *info); int ethnl_act_cable_test(struct sk_buff *skb, struct genl_info *info); From 335a9660e141349d7751b3b880d7531ea401a8db Mon Sep 17 00:00:00 2001 From: Oleksij Rempel Date: Mon, 27 Oct 2025 13:28:00 +0100 Subject: [PATCH 0474/1381] net: phy: micrel: add MSE interface support for KSZ9477 family MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Implement the get_mse_capability() and get_mse_snapshot() PHY driver ops for KSZ9477-series integrated PHYs to demonstrate the new PHY MSE UAPI. These PHYs do not expose a documented direct MSE register, but the Signal Quality Indicator (SQI) registers are derived from the internal MSE computation. This hook maps SQI readings into the MSE interface so that tooling can retrieve the raw value together with metadata for correct interpretation in userspace. Behaviour: - For 1000BASE-T, report per-channel (A–D) values and support a WORST channel selector. - For 100BASE-TX, only LINK-wide measurements are available. - Report average MSE only, with a max scale based on KSZ9477_MMD_SQI_MASK and a fixed refresh rate of 2 µs. This mapping differs from the OPEN Alliance SQI definition, which assigns thresholds such as pre-fail indices; the MSE interface instead provides the raw measurement, leaving interpretation to userspace. Signed-off-by: Oleksij Rempel Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/20251027122801.982364-4-o.rempel@pengutronix.de Signed-off-by: Jakub Kicinski --- drivers/net/phy/micrel.c | 102 +++++++++++++++++++++++++++++++++++++++ 1 file changed, 102 insertions(+) diff --git a/drivers/net/phy/micrel.c b/drivers/net/phy/micrel.c index edca0024b7c7..06080b1c753d 100644 --- a/drivers/net/phy/micrel.c +++ b/drivers/net/phy/micrel.c @@ -2325,6 +2325,106 @@ static int kszphy_get_sqi_max(struct phy_device *phydev) return KSZ9477_SQI_MAX; } +static int kszphy_get_mse_capability(struct phy_device *phydev, + struct phy_mse_capability *cap) +{ + /* Capabilities depend on link mode: + * - 1000BASE-T: per-pair SQI registers exist => expose A..D + * and a WORST selector. + * - 100BASE-TX: HW provides a single MSE/SQI reading in the "channel A" + * register, but with auto MDI-X there is no MDI-X resolution bit, + * so we cannot map that register to a specific wire pair reliably. + * To avoid misleading per-channel data, advertise only LINK. + * Other speeds: no MSE exposure via this driver. + * + * Note: WORST is *not* a hardware selector on this family. + * We expose it because the driver computes it in software + * by scanning per-channel readouts (A..D) and picking the + * maximum average MSE. + */ + if (phydev->speed == SPEED_1000) + cap->supported_caps = PHY_MSE_CAP_CHANNEL_A | + PHY_MSE_CAP_CHANNEL_B | + PHY_MSE_CAP_CHANNEL_C | + PHY_MSE_CAP_CHANNEL_D | + PHY_MSE_CAP_WORST_CHANNEL; + else if (phydev->speed == SPEED_100) + cap->supported_caps = PHY_MSE_CAP_LINK; + else + return -EOPNOTSUPP; + + cap->max_average_mse = FIELD_MAX(KSZ9477_MMD_SQI_MASK); + cap->refresh_rate_ps = 2000000; /* 2 us */ + /* Estimated from link modulation (125 MBd per channel) and documented + * refresh rate of 2 us + */ + cap->num_symbols = 250; + + cap->supported_caps |= PHY_MSE_CAP_AVG; + + return 0; +} + +static int kszphy_get_mse_snapshot(struct phy_device *phydev, + enum phy_mse_channel channel, + struct phy_mse_snapshot *snapshot) +{ + u8 num_channels; + int ret; + + if (phydev->speed == SPEED_1000) + num_channels = 4; + else if (phydev->speed == SPEED_100) + num_channels = 1; + else + return -EOPNOTSUPP; + + if (channel == PHY_MSE_CHANNEL_WORST) { + u32 worst_val = 0; + int i; + + /* WORST is implemented in software: select the maximum + * average MSE across the available per-channel registers. + * Only defined when multiple channels exist (1000BASE-T). + */ + if (num_channels < 2) + return -EOPNOTSUPP; + + for (i = 0; i < num_channels; i++) { + ret = phy_read_mmd(phydev, MDIO_MMD_PMAPMD, + KSZ9477_MMD_SIGNAL_QUALITY_CHAN_A + i); + if (ret < 0) + return ret; + + ret = FIELD_GET(KSZ9477_MMD_SQI_MASK, ret); + if (ret > worst_val) + worst_val = ret; + } + snapshot->average_mse = worst_val; + } else if (channel == PHY_MSE_CHANNEL_LINK && num_channels == 1) { + ret = phy_read_mmd(phydev, MDIO_MMD_PMAPMD, + KSZ9477_MMD_SIGNAL_QUALITY_CHAN_A); + if (ret < 0) + return ret; + snapshot->average_mse = FIELD_GET(KSZ9477_MMD_SQI_MASK, ret); + } else if (channel >= PHY_MSE_CHANNEL_A && + channel <= PHY_MSE_CHANNEL_D) { + /* Per-channel readouts are valid only for 1000BASE-T. */ + if (phydev->speed != SPEED_1000) + return -EOPNOTSUPP; + + ret = phy_read_mmd(phydev, MDIO_MMD_PMAPMD, + KSZ9477_MMD_SIGNAL_QUALITY_CHAN_A + channel); + if (ret < 0) + return ret; + snapshot->average_mse = FIELD_GET(KSZ9477_MMD_SQI_MASK, ret); + } else { + return -EOPNOTSUPP; + } + + return 0; +} + static void kszphy_enable_clk(struct phy_device *phydev) { struct kszphy_priv *priv = phydev->priv; @@ -6497,6 +6597,8 @@ static struct phy_driver ksphy_driver[] = { .cable_test_get_status = ksz9x31_cable_test_get_status, .get_sqi = kszphy_get_sqi, .get_sqi_max = kszphy_get_sqi_max, + .get_mse_capability = kszphy_get_mse_capability, + .get_mse_snapshot = kszphy_get_mse_snapshot, } }; module_phy_driver(ksphy_driver); From fd93ed77efe4735cd2b9a3fbccd5e199ced19bba Mon Sep 17 00:00:00 2001 From: Oleksij Rempel Date: Mon, 27 Oct 2025 13:28:01 +0100 Subject: [PATCH 0475/1381] net: phy: dp83td510: add MSE interface support for 10BASE-T1L Implement get_mse_capability() and get_mse_snapshot() for the DP83TD510E to expose its Mean Square Error (MSE) register via the new PHY MSE UAPI. The DP83TD510E does not document any peak MSE values; it only exposes a single average MSE register used internally to derive SQI. This implementation therefore advertises only PHY_MSE_CAP_AVG, along with LINK and channel-A selectors. Scaling is fixed to 0xFFFF, and the refresh interval/number of symbols are estimated from 10BASE-T1L symbol rate (7.5 MBd) and typical diagnostic intervals (~1 ms). For 10BASE-T1L deployments, SQI is a reliable indicator of link modulation quality once the link is established, but it does not indicate whether autonegotiation pulses will be correctly received in marginal conditions. MSE provides a direct measurement of slicer error rate that can be used to evaluate if autonegotiation is likely to succeed under a given cable length and condition. In practice, testing such scenarios often requires forcing a fixed-link setup to isolate MSE behaviour from the autonegotiation process. Signed-off-by: Oleksij Rempel Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/20251027122801.982364-5-o.rempel@pengutronix.de Signed-off-by: Jakub Kicinski --- drivers/net/phy/dp83td510.c | 62 +++++++++++++++++++++++++++++++++++++ 1 file changed, 62 insertions(+) diff --git a/drivers/net/phy/dp83td510.c b/drivers/net/phy/dp83td510.c index 23af1ac194fa..d75dae6071ad 100644 --- a/drivers/net/phy/dp83td510.c +++ b/drivers/net/phy/dp83td510.c @@ -61,6 +61,7 @@ #define DP83TD510E_MASTER_SLAVE_RESOL_FAIL BIT(15) #define DP83TD510E_MSE_DETECT 0xa85 +#define DP83TD510E_MSE_MAX U16_MAX #define DP83TD510_SQI_MAX 7 @@ -249,6 +250,64 @@ struct dp83td510_priv { #define DP83TD510E_ALCD_COMPLETE BIT(15) #define DP83TD510E_ALCD_CABLE_LENGTH GENMASK(10, 0) +static int dp83td510_get_mse_capability(struct phy_device *phydev, + struct phy_mse_capability *cap) +{ + /* DP83TD510E documents only a single (average) MSE register + * (used to derive SQI); no peak or worst-peak counters are + * described. Advertise only PHY_MSE_CAP_AVG. + */ + cap->supported_caps = PHY_MSE_CAP_AVG; + /* 10BASE-T1L is a single-pair medium, so there are no B/C/D channels. + * We still advertise PHY_MSE_CAP_CHANNEL_A to indicate that the PHY + * can attribute the measurement to a specific pair (the only one), + * rather than exposing it only as a link-aggregate. + * + * Rationale: + * - Keeps the ethtool MSE_GET selection logic consistent: per-channel + * (A/B/C/D) is preferred over WORST/LINK, so userspace receives a + * CHANNEL_A nest instead of LINK. + * - Signals to tools that "per-pair" data is available (even if there's + * just one pair), avoiding the impression that only aggregate values + * are supported. + * - Remains compatible with multi-pair PHYs and uniform UI handling. + * + * Note: WORST and other channels are not advertised on 10BASE-T1L. + */ + cap->supported_caps |= PHY_MSE_CHANNEL_A | PHY_MSE_CAP_LINK; + cap->max_average_mse = DP83TD510E_MSE_MAX; + + /* The datasheet does not specify the refresh rate or symbol count, + * but based on similar PHYs and standards, we can assume a common + * value. For 10BASE-T1L, the symbol rate is 7.5 MBd. A common + * diagnostic interval is around 1ms. + * 7.5e6 symbols/sec * 0.001 sec = 7500 symbols. + */ + cap->refresh_rate_ps = 1000000000; /* 1 ms */ + cap->num_symbols = 7500; + + return 0; +} + +static int dp83td510_get_mse_snapshot(struct phy_device *phydev, + enum phy_mse_channel channel, + struct phy_mse_snapshot *snapshot) +{ + int ret; + + if (channel != PHY_MSE_CHANNEL_LINK && + channel != PHY_MSE_CHANNEL_A) + return -EOPNOTSUPP; + + ret = phy_read_mmd(phydev, MDIO_MMD_VEND2, DP83TD510E_MSE_DETECT); + if (ret < 0) + return ret; + + snapshot->average_mse = ret; + + return 0; +} + static int dp83td510_led_brightness_set(struct phy_device *phydev, u8 index, enum led_brightness brightness) { @@ -893,6 +952,9 @@ static struct phy_driver dp83td510_driver[] = { .get_phy_stats = dp83td510_get_phy_stats, .update_stats = dp83td510_update_stats, + .get_mse_capability = dp83td510_get_mse_capability, + .get_mse_snapshot = dp83td510_get_mse_snapshot, + .led_brightness_set = dp83td510_led_brightness_set, .led_hw_is_supported = dp83td510_led_hw_is_supported, .led_hw_control_set = dp83td510_led_hw_control_set, From 27cb3de7f43ac0263474d87a2c84d96f904d73e2 Mon Sep 17 00:00:00 2001 From: Tonghao Zhang Date: Tue, 28 Oct 2025 12:32:44 +0800 Subject: [PATCH 0476/1381] net: add net cookie for net device trace events In a multi-network card or container environment, this is needed in order to differentiate between trace events relating to net devices that exist in different network namespaces and share the same name. for xmit_timeout trace events: [002] ..s1. 1838.311662: net_dev_xmit_timeout: dev=eth0 driver=virtio_net queue=10 net_cookie=3 [007] ..s1. 1839.335650: net_dev_xmit_timeout: dev=eth0 driver=virtio_net queue=10 net_cookie=4100 [007] ..s1. 1844.455659: net_dev_xmit_timeout: dev=eth0 driver=virtio_net queue=10 net_cookie=3 [002] ..s1. 1850.087647: net_dev_xmit_timeout: dev=eth0 driver=virtio_net queue=10 net_cookie=3 Cc: Eran Ben Elisha Cc: Jiri Pirko Cc: Cong Wang Cc: Jakub Kicinski Cc: Eric Dumazet Cc: Simon Horman Cc: Paolo Abeni Suggested-by: Ido Schimmel Signed-off-by: Tonghao Zhang Reviewed-by: Ido Schimmel Link: https://patch.msgid.link/20251028043244.82288-1-tonghao@bamaicloud.com Signed-off-by: Paolo Abeni --- include/trace/events/net.h | 37 +++++++++++++++++++++++++++---------- 1 file changed, 27 insertions(+), 10 deletions(-) diff --git a/include/trace/events/net.h b/include/trace/events/net.h index d55162c12f90..fdd9ad474ce3 100644 --- a/include/trace/events/net.h +++ b/include/trace/events/net.h @@ -35,6 +35,7 @@ TRACE_EVENT(net_dev_start_xmit, __field( u16, gso_size ) __field( u16, gso_segs ) __field( u16, gso_type ) + __field( u64, net_cookie ) ), TP_fast_assign( @@ -57,16 +58,18 @@ TRACE_EVENT(net_dev_start_xmit, __entry->gso_size = skb_shinfo(skb)->gso_size; __entry->gso_segs = skb_shinfo(skb)->gso_segs; __entry->gso_type = skb_shinfo(skb)->gso_type; + __entry->net_cookie = dev_net(dev)->net_cookie; ), - TP_printk("dev=%s queue_mapping=%u skbaddr=%p vlan_tagged=%d vlan_proto=0x%04x vlan_tci=0x%04x protocol=0x%04x ip_summed=%d len=%u data_len=%u network_offset=%d transport_offset_valid=%d transport_offset=%d tx_flags=%d gso_size=%d gso_segs=%d gso_type=%#x", + TP_printk("dev=%s queue_mapping=%u skbaddr=%p vlan_tagged=%d vlan_proto=0x%04x vlan_tci=0x%04x protocol=0x%04x ip_summed=%d len=%u data_len=%u network_offset=%d transport_offset_valid=%d transport_offset=%d tx_flags=%d gso_size=%d gso_segs=%d gso_type=%#x net_cookie=%llu", __get_str(name), __entry->queue_mapping, __entry->skbaddr, __entry->vlan_tagged, __entry->vlan_proto, __entry->vlan_tci, __entry->protocol, __entry->ip_summed, __entry->len, __entry->data_len, __entry->network_offset, __entry->transport_offset_valid, __entry->transport_offset, __entry->tx_flags, - __entry->gso_size, __entry->gso_segs, __entry->gso_type) + __entry->gso_size, __entry->gso_segs, + __entry->gso_type, __entry->net_cookie) ); TRACE_EVENT(net_dev_xmit, @@ -83,17 +86,21 @@ TRACE_EVENT(net_dev_xmit, __field( unsigned int, len ) __field( int, rc ) __string( name, dev->name ) + __field( u64, net_cookie ) ), TP_fast_assign( __entry->skbaddr = skb; __entry->len = skb_len; __entry->rc = rc; + __entry->net_cookie = dev_net(dev)->net_cookie; __assign_str(name); ), - TP_printk("dev=%s skbaddr=%p len=%u rc=%d", - __get_str(name), __entry->skbaddr, __entry->len, __entry->rc) + TP_printk("dev=%s skbaddr=%p len=%u rc=%d net_cookie=%llu", + __get_str(name), __entry->skbaddr, + __entry->len, __entry->rc, + __entry->net_cookie) ); TRACE_EVENT(net_dev_xmit_timeout, @@ -107,16 +114,19 @@ TRACE_EVENT(net_dev_xmit_timeout, __string( name, dev->name ) __string( driver, netdev_drivername(dev)) __field( int, queue_index ) + __field( u64, net_cookie ) ), TP_fast_assign( __assign_str(name); __assign_str(driver); __entry->queue_index = queue_index; + __entry->net_cookie = dev_net(dev)->net_cookie; ), - TP_printk("dev=%s driver=%s queue=%d", - __get_str(name), __get_str(driver), __entry->queue_index) + TP_printk("dev=%s driver=%s queue=%d net_cookie=%llu", + __get_str(name), __get_str(driver), + __entry->queue_index, __entry->net_cookie) ); DECLARE_EVENT_CLASS(net_dev_template, @@ -129,16 +139,20 @@ DECLARE_EVENT_CLASS(net_dev_template, __field( void *, skbaddr ) __field( unsigned int, len ) __string( name, skb->dev->name ) + __field( u64, net_cookie ) ), TP_fast_assign( __entry->skbaddr = skb; __entry->len = skb->len; + __entry->net_cookie = dev_net(skb->dev)->net_cookie; __assign_str(name); ), - TP_printk("dev=%s skbaddr=%p len=%u", - __get_str(name), __entry->skbaddr, __entry->len) + TP_printk("dev=%s skbaddr=%p len=%u net_cookie=%llu", + __get_str(name), __entry->skbaddr, + __entry->len, + __entry->net_cookie) ) DEFINE_EVENT(net_dev_template, net_dev_queue, @@ -188,6 +202,7 @@ DECLARE_EVENT_CLASS(net_dev_rx_verbose_template, __field( unsigned char, nr_frags ) __field( u16, gso_size ) __field( u16, gso_type ) + __field( u64, net_cookie ) ), TP_fast_assign( @@ -214,16 +229,18 @@ DECLARE_EVENT_CLASS(net_dev_rx_verbose_template, __entry->nr_frags = skb_shinfo(skb)->nr_frags; __entry->gso_size = skb_shinfo(skb)->gso_size; __entry->gso_type = skb_shinfo(skb)->gso_type; + __entry->net_cookie = dev_net(skb->dev)->net_cookie; ), - TP_printk("dev=%s napi_id=%#x queue_mapping=%u skbaddr=%p vlan_tagged=%d vlan_proto=0x%04x vlan_tci=0x%04x protocol=0x%04x ip_summed=%d hash=0x%08x l4_hash=%d len=%u data_len=%u truesize=%u mac_header_valid=%d mac_header=%d nr_frags=%d gso_size=%d gso_type=%#x", + TP_printk("dev=%s napi_id=%#x queue_mapping=%u skbaddr=%p vlan_tagged=%d vlan_proto=0x%04x vlan_tci=0x%04x protocol=0x%04x ip_summed=%d hash=0x%08x l4_hash=%d len=%u data_len=%u truesize=%u mac_header_valid=%d mac_header=%d nr_frags=%d gso_size=%d gso_type=%#x net_cookie=%llu", __get_str(name), __entry->napi_id, __entry->queue_mapping, __entry->skbaddr, __entry->vlan_tagged, __entry->vlan_proto, __entry->vlan_tci, __entry->protocol, __entry->ip_summed, __entry->hash, __entry->l4_hash, __entry->len, __entry->data_len, __entry->truesize, __entry->mac_header_valid, __entry->mac_header, - __entry->nr_frags, __entry->gso_size, __entry->gso_type) + __entry->nr_frags, __entry->gso_size, + __entry->gso_type, __entry->net_cookie) ); DEFINE_EVENT(net_dev_rx_verbose_template, napi_gro_frags_entry, From 462280043466b2bc74483c56a5d5316ff6b16380 Mon Sep 17 00:00:00 2001 From: Jason Xing Date: Thu, 30 Oct 2025 08:06:45 +0800 Subject: [PATCH 0477/1381] xsk: do not enable/disable irq when grabbing/releasing xsk_tx_list_lock The commit ac98d8aab61b ("xsk: wire upp Tx zero-copy functions") originally introducing this lock put the deletion process in the sk_destruct which can run in irq context obviously, so the xxx_irqsave()/xxx_irqrestore() pair was used. But later another commit 541d7fdd7694 ("xsk: proper AF_XDP socket teardown ordering") moved the deletion into xsk_release() that only happens in process context. It means that since this commit, it doesn't necessarily need that pair. Now, there are two places that use this xsk_tx_list_lock and only run in the process context. So avoid manipulating the irq then. Signed-off-by: Jason Xing Acked-by: Maciej Fijalkowski Link: https://patch.msgid.link/20251030000646.18859-2-kerneljasonxing@gmail.com Signed-off-by: Paolo Abeni --- net/xdp/xsk_buff_pool.c | 12 ++++-------- 1 file changed, 4 insertions(+), 8 deletions(-) diff --git a/net/xdp/xsk_buff_pool.c b/net/xdp/xsk_buff_pool.c index aa9788f20d0d..309075050b2a 100644 --- a/net/xdp/xsk_buff_pool.c +++ b/net/xdp/xsk_buff_pool.c @@ -12,26 +12,22 @@ void xp_add_xsk(struct xsk_buff_pool *pool, struct xdp_sock *xs) { - unsigned long flags; - if (!xs->tx) return; - spin_lock_irqsave(&pool->xsk_tx_list_lock, flags); + spin_lock(&pool->xsk_tx_list_lock); list_add_rcu(&xs->tx_list, &pool->xsk_tx_list); - spin_unlock_irqrestore(&pool->xsk_tx_list_lock, flags); + spin_unlock(&pool->xsk_tx_list_lock); } void xp_del_xsk(struct xsk_buff_pool *pool, struct xdp_sock *xs) { - unsigned long flags; - if (!xs->tx) return; - spin_lock_irqsave(&pool->xsk_tx_list_lock, flags); + spin_lock(&pool->xsk_tx_list_lock); list_del_rcu(&xs->tx_list); - spin_unlock_irqrestore(&pool->xsk_tx_list_lock, flags); + spin_unlock(&pool->xsk_tx_list_lock); } void xp_destroy(struct xsk_buff_pool *pool) From 30ed05adca4a05c50594384cff18910858dd1d35 Mon Sep 17 00:00:00 2001 From: Jason Xing Date: Thu, 30 Oct 2025 08:06:46 +0800 Subject: [PATCH 0478/1381] xsk: use a smaller new lock for shared pool case - Split cq_lock into two smaller locks: cq_prod_lock and cq_cached_prod_lock - Avoid disabling/enabling interrupts in the hot xmit path In either xsk_cq_cancel_locked() or xsk_cq_reserve_locked() function, the race condition is only between multiple xsks sharing the same pool. They are all in the process context rather than interrupt context, so now the small lock named cq_cached_prod_lock can be used without handling interrupts. While cq_cached_prod_lock ensures the exclusive modification of @cached_prod, cq_prod_lock in xsk_cq_submit_addr_locked() only cares about @producer and corresponding @desc. Both of them don't necessarily be consistent with @cached_prod protected by cq_cached_prod_lock. That's the reason why the previous big lock can be split into two smaller ones. Please note that SPSC rule is all about the global state of producer and consumer that can affect both layers instead of local or cached ones. Frequently disabling and enabling interrupt are very time consuming in some cases, especially in a per-descriptor granularity, which now can be avoided after this optimization, even when the pool is shared by multiple xsks. With this patch, the performance number[1] could go from 1,872,565 pps to 1,961,009 pps. It's a minor rise of around 5%. [1]: taskset -c 1 ./xdpsock -i enp2s0f1 -q 0 -t -S -s 64 Signed-off-by: Jason Xing Acked-by: Maciej Fijalkowski Link: https://patch.msgid.link/20251030000646.18859-3-kerneljasonxing@gmail.com Signed-off-by: Paolo Abeni --- include/net/xsk_buff_pool.h | 13 +++++++++---- net/xdp/xsk.c | 15 ++++++--------- net/xdp/xsk_buff_pool.c | 3 ++- 3 files changed, 17 insertions(+), 14 deletions(-) diff --git a/include/net/xsk_buff_pool.h b/include/net/xsk_buff_pool.h index cac56e6b0869..92a2358c6ce3 100644 --- a/include/net/xsk_buff_pool.h +++ b/include/net/xsk_buff_pool.h @@ -85,11 +85,16 @@ struct xsk_buff_pool { bool unaligned; bool tx_sw_csum; void *addrs; - /* Mutual exclusion of the completion ring in the SKB mode. Two cases to protect: - * NAPI TX thread and sendmsg error paths in the SKB destructor callback and when - * sockets share a single cq when the same netdev and queue id is shared. + /* Mutual exclusion of the completion ring in the SKB mode. + * Protect: NAPI TX thread and sendmsg error paths in the SKB + * destructor callback. */ - spinlock_t cq_lock; + spinlock_t cq_prod_lock; + /* Mutual exclusion of the completion ring in the SKB mode. + * Protect: when sockets share a single cq when the same netdev + * and queue id is shared. + */ + spinlock_t cq_cached_prod_lock; struct xdp_buff_xsk *free_heads[]; }; diff --git a/net/xdp/xsk.c b/net/xdp/xsk.c index 7b0c68a70888..2f26c918d448 100644 --- a/net/xdp/xsk.c +++ b/net/xdp/xsk.c @@ -548,12 +548,11 @@ static int xsk_wakeup(struct xdp_sock *xs, u8 flags) static int xsk_cq_reserve_locked(struct xsk_buff_pool *pool) { - unsigned long flags; int ret; - spin_lock_irqsave(&pool->cq_lock, flags); + spin_lock(&pool->cq_cached_prod_lock); ret = xskq_prod_reserve(pool->cq); - spin_unlock_irqrestore(&pool->cq_lock, flags); + spin_unlock(&pool->cq_cached_prod_lock); return ret; } @@ -566,7 +565,7 @@ static void xsk_cq_submit_addr_locked(struct xsk_buff_pool *pool, unsigned long flags; u32 idx; - spin_lock_irqsave(&pool->cq_lock, flags); + spin_lock_irqsave(&pool->cq_prod_lock, flags); idx = xskq_get_prod(pool->cq); xskq_prod_write_addr(pool->cq, idx, @@ -583,16 +582,14 @@ static void xsk_cq_submit_addr_locked(struct xsk_buff_pool *pool, } } xskq_prod_submit_n(pool->cq, descs_processed); - spin_unlock_irqrestore(&pool->cq_lock, flags); + spin_unlock_irqrestore(&pool->cq_prod_lock, flags); } static void xsk_cq_cancel_locked(struct xsk_buff_pool *pool, u32 n) { - unsigned long flags; - - spin_lock_irqsave(&pool->cq_lock, flags); + spin_lock(&pool->cq_cached_prod_lock); xskq_prod_cancel_n(pool->cq, n); - spin_unlock_irqrestore(&pool->cq_lock, flags); + spin_unlock(&pool->cq_cached_prod_lock); } static void xsk_inc_num_desc(struct sk_buff *skb) diff --git a/net/xdp/xsk_buff_pool.c b/net/xdp/xsk_buff_pool.c index 309075050b2a..00a4eddaa0cd 100644 --- a/net/xdp/xsk_buff_pool.c +++ b/net/xdp/xsk_buff_pool.c @@ -90,7 +90,8 @@ struct xsk_buff_pool *xp_create_and_assign_umem(struct xdp_sock *xs, INIT_LIST_HEAD(&pool->xskb_list); INIT_LIST_HEAD(&pool->xsk_tx_list); spin_lock_init(&pool->xsk_tx_list_lock); - spin_lock_init(&pool->cq_lock); + spin_lock_init(&pool->cq_prod_lock); + spin_lock_init(&pool->cq_cached_prod_lock); refcount_set(&pool->users, 1); pool->fq = xs->fq_tmp; From 105bae321862b3bac300c73748192ff61a5129cd Mon Sep 17 00:00:00 2001 From: Adrian Moreno Date: Mon, 3 Nov 2025 16:40:04 +0100 Subject: [PATCH 0479/1381] rtnetlink: honor RTEXT_FILTER_SKIP_STATS in IFLA_STATS MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Gathering interface statistics can be a relatively expensive operation on certain systems as it requires iterating over all the cpus. RTEXT_FILTER_SKIP_STATS was first introduced [1] to skip AF_INET6 statistics from interface dumps and it was then extended [2] to also exclude IFLA_VF_INFO. The semantics of the flag does not seem to be limited to AF_INET or VF statistics and having a way to query the interface status (e.g: carrier, address) without retrieving its statistics seems reasonable. So this patch extends the use RTEXT_FILTER_SKIP_STATS to also affect IFLA_STATS. [1] https://lore.kernel.org/all/20150911204848.GC9687@oracle.com/ [2] https://lore.kernel.org/all/20230611105108.122586-1-gal@nvidia.com/ Signed-off-by: Adrian Moreno Reviewed-by: Toke Høiland-Jørgensen Reviewed-by: Eric Dumazet Reviewed-by: Nicolas Dichtel Link: https://patch.msgid.link/20251103154006.1189707-1-amorenoz@redhat.com Signed-off-by: Jakub Kicinski --- net/core/rtnetlink.c | 15 +++++++++++---- 1 file changed, 11 insertions(+), 4 deletions(-) diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 576d5ec3bb36..b1ed55141d8a 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -1270,13 +1270,13 @@ static size_t rtnl_dpll_pin_size(const struct net_device *dev) static noinline size_t if_nlmsg_size(const struct net_device *dev, u32 ext_filter_mask) { - return NLMSG_ALIGN(sizeof(struct ifinfomsg)) + size_t size; + + size = NLMSG_ALIGN(sizeof(struct ifinfomsg)) + nla_total_size(IFNAMSIZ) /* IFLA_IFNAME */ + nla_total_size(IFALIASZ) /* IFLA_IFALIAS */ + nla_total_size(IFNAMSIZ) /* IFLA_QDISC */ + nla_total_size_64bit(sizeof(struct rtnl_link_ifmap)) - + nla_total_size(sizeof(struct rtnl_link_stats)) - + nla_total_size_64bit(sizeof(struct rtnl_link_stats64)) + nla_total_size(MAX_ADDR_LEN) /* IFLA_ADDRESS */ + nla_total_size(MAX_ADDR_LEN) /* IFLA_BROADCAST */ + nla_total_size(4) /* IFLA_TXQLEN */ @@ -1329,6 +1329,12 @@ static noinline size_t if_nlmsg_size(const struct net_device *dev, + nla_total_size(2) /* IFLA_HEADROOM */ + nla_total_size(2) /* IFLA_TAILROOM */ + 0; + + if (!(ext_filter_mask & RTEXT_FILTER_SKIP_STATS)) + size += nla_total_size(sizeof(struct rtnl_link_stats)) + + nla_total_size_64bit(sizeof(struct rtnl_link_stats64)); + + return size; } static int rtnl_vf_ports_fill(struct sk_buff *skb, struct net_device *dev) @@ -2123,7 +2129,8 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, if (rtnl_phys_switch_id_fill(skb, dev)) goto nla_put_failure; - if (rtnl_fill_stats(skb, dev)) + if (!(ext_filter_mask & RTEXT_FILTER_SKIP_STATS) && + rtnl_fill_stats(skb, dev)) goto nla_put_failure; if (rtnl_fill_vf(skb, dev, ext_filter_mask)) From 46173144e03d87beddf02ee785cbdf818087687a Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 3 Nov 2025 16:52:56 +0000 Subject: [PATCH 0480/1381] net: mark deliver_skb() as unlikely and not inlined deliver_skb() should not be inlined as is it not called in the fast path. Add unlikely() clauses giving hints to the compiler about this fact. Before this patch: size net/core/dev.o text data bss dec hex filename 121794 13330 176 135300 21084 net/core/dev.o __netif_receive_skb_core() size on x86_64 : 4080 bytes. After: size net/core/dev.o text data bss dec hex filenamee 120330 13338 176 133844 20ad4 net/core/dev.o __netif_receive_skb_core() size on x86_64 : 2781 bytes. Signed-off-by: Eric Dumazet Link: https://patch.msgid.link/20251103165256.1712169-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/core/dev.c | 22 +++++++++++----------- 1 file changed, 11 insertions(+), 11 deletions(-) diff --git a/net/core/dev.c b/net/core/dev.c index 2c1de5fb97d9..ba39146bbd25 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2463,9 +2463,9 @@ int dev_forward_skb_nomtu(struct net_device *dev, struct sk_buff *skb) return __dev_forward_skb2(dev, skb, false) ?: netif_rx_internal(skb); } -static inline int deliver_skb(struct sk_buff *skb, - struct packet_type *pt_prev, - struct net_device *orig_dev) +static int deliver_skb(struct sk_buff *skb, + struct packet_type *pt_prev, + struct net_device *orig_dev) { if (unlikely(skb_orphan_frags_rx(skb, GFP_ATOMIC))) return -ENOMEM; @@ -2484,7 +2484,7 @@ static inline void deliver_ptype_list_skb(struct sk_buff *skb, list_for_each_entry_rcu(ptype, ptype_list, list) { if (ptype->type != type) continue; - if (pt_prev) + if (unlikely(pt_prev)) deliver_skb(skb, pt_prev, orig_dev); pt_prev = ptype; } @@ -2545,7 +2545,7 @@ void dev_queue_xmit_nit(struct sk_buff *skb, struct net_device *dev) if (skb_loop_sk(ptype, skb)) continue; - if (pt_prev) { + if (unlikely(pt_prev)) { deliver_skb(skb2, pt_prev, skb->dev); pt_prev = ptype; continue; @@ -4421,7 +4421,7 @@ sch_handle_ingress(struct sk_buff *skb, struct packet_type **pt_prev, int *ret, return skb; bpf_net_ctx = bpf_net_ctx_set(&__bpf_net_ctx); - if (*pt_prev) { + if (unlikely(*pt_prev)) { *ret = deliver_skb(skb, *pt_prev, orig_dev); *pt_prev = NULL; } @@ -5883,7 +5883,7 @@ static inline int nf_ingress(struct sk_buff *skb, struct packet_type **pt_prev, if (nf_hook_ingress_active(skb)) { int ingress_retval; - if (*pt_prev) { + if (unlikely(*pt_prev)) { *ret = deliver_skb(skb, *pt_prev, orig_dev); *pt_prev = NULL; } @@ -5960,13 +5960,13 @@ static int __netif_receive_skb_core(struct sk_buff **pskb, bool pfmemalloc, list_for_each_entry_rcu(ptype, &dev_net_rcu(skb->dev)->ptype_all, list) { - if (pt_prev) + if (unlikely(pt_prev)) ret = deliver_skb(skb, pt_prev, orig_dev); pt_prev = ptype; } list_for_each_entry_rcu(ptype, &skb->dev->ptype_all, list) { - if (pt_prev) + if (unlikely(pt_prev)) ret = deliver_skb(skb, pt_prev, orig_dev); pt_prev = ptype; } @@ -5997,7 +5997,7 @@ static int __netif_receive_skb_core(struct sk_buff **pskb, bool pfmemalloc, } if (skb_vlan_tag_present(skb)) { - if (pt_prev) { + if (unlikely(pt_prev)) { ret = deliver_skb(skb, pt_prev, orig_dev); pt_prev = NULL; } @@ -6009,7 +6009,7 @@ static int __netif_receive_skb_core(struct sk_buff **pskb, bool pfmemalloc, rx_handler = rcu_dereference(skb->dev->rx_handler); if (rx_handler) { - if (pt_prev) { + if (unlikely(pt_prev)) { ret = deliver_skb(skb, pt_prev, orig_dev); pt_prev = NULL; } From dec568a36f9b16f0334aed8e95ec4225606830cc Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Mon, 3 Nov 2025 11:49:55 +0000 Subject: [PATCH 0481/1381] net: stmmac: imx: use phylink's interface mode for set_clk_tx_rate() imx_dwmac_set_clk_tx_rate() is passed the interface mode from phylink which will be the same as plat_dat->phy_interface. Use the passed-in interface mode rather than plat_dat->phy_interface. Reviewed-by: Maxime Chevallier Tested-by: Maxime Chevallier Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vFt4N-0000000ChoM-1llp@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c index 4268b9987237..147fa08d5b6e 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c @@ -195,9 +195,6 @@ static void imx_dwmac_exit(struct platform_device *pdev, void *priv) static int imx_dwmac_set_clk_tx_rate(void *bsp_priv, struct clk *clk_tx_i, phy_interface_t interface, int speed) { - struct imx_priv_data *dwmac = bsp_priv; - - interface = dwmac->plat_dat->phy_interface; if (interface == PHY_INTERFACE_MODE_RMII || interface == PHY_INTERFACE_MODE_MII) return 0; From 553f23d1953527eb277efa902cd498131b2527e1 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Mon, 3 Nov 2025 11:50:00 +0000 Subject: [PATCH 0482/1381] net: stmmac: s32: move PHY_INTF_SEL_x definitions out of the way S32's PHY_INTF_SEL_x definitions conflict with those for the dwmac cores as they use a different bitmapping. Add a S32 prefix so that they are unique. Signed-off-by: Russell King (Oracle) Reviewed-by: Maxime Chevallier Reviewed-by: Jan Petrous (OSS) Link: https://patch.msgid.link/E1vFt4S-0000000ChoS-2Ahi@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/dwmac-s32.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-s32.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-s32.c index ee095ac13203..2b7ad64bfdf7 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-s32.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-s32.c @@ -24,10 +24,10 @@ #define GMAC_INTF_RATE_125M 125000000 /* 125MHz */ /* SoC PHY interface control register */ -#define PHY_INTF_SEL_MII 0x00 -#define PHY_INTF_SEL_SGMII 0x01 -#define PHY_INTF_SEL_RGMII 0x02 -#define PHY_INTF_SEL_RMII 0x08 +#define S32_PHY_INTF_SEL_MII 0x00 +#define S32_PHY_INTF_SEL_SGMII 0x01 +#define S32_PHY_INTF_SEL_RGMII 0x02 +#define S32_PHY_INTF_SEL_RMII 0x08 struct s32_priv_data { void __iomem *ioaddr; @@ -40,7 +40,7 @@ struct s32_priv_data { static int s32_gmac_write_phy_intf_select(struct s32_priv_data *gmac) { - writel(PHY_INTF_SEL_RGMII, gmac->ctrl_sts); + writel(S32_PHY_INTF_SEL_RGMII, gmac->ctrl_sts); dev_dbg(gmac->dev, "PHY mode set to %s\n", phy_modes(*gmac->intf_mode)); From 4a4692e9091867dd413764c7d81f09e8109a233a Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Mon, 3 Nov 2025 11:50:05 +0000 Subject: [PATCH 0483/1381] net: stmmac: add phy_intf_sel and ACTPHYIF definitions Add definitions for the active PHY interface found in DMA hardware feature register 0, and also used to configure the core in multi- interface designs via phy_intf_sel. Signed-off-by: Russell King (Oracle) Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/E1vFt4X-0000000ChoY-30p9@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/common.h | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/drivers/net/ethernet/stmicro/stmmac/common.h b/drivers/net/ethernet/stmicro/stmmac/common.h index 27083af54568..7395bbb94aea 100644 --- a/drivers/net/ethernet/stmicro/stmmac/common.h +++ b/drivers/net/ethernet/stmicro/stmmac/common.h @@ -313,6 +313,16 @@ struct stmmac_safety_stats { #define DMA_HW_FEAT_ACTPHYIF 0x70000000 /* Active/selected PHY iface */ #define DEFAULT_DMA_PBL 8 +/* phy_intf_sel_i and ACTPHYIF encodings */ +#define PHY_INTF_SEL_GMII_MII 0 +#define PHY_INTF_SEL_RGMII 1 +#define PHY_INTF_SEL_SGMII 2 +#define PHY_INTF_SEL_TBI 3 +#define PHY_INTF_SEL_RMII 4 +#define PHY_INTF_SEL_RTBI 5 +#define PHY_INTF_SEL_SMII 6 +#define PHY_INTF_SEL_REVMII 7 + /* MSI defines */ #define STMMAC_MSI_VEC_MAX 32 From b459790d3fd6d7ead31182ae0cd8632fe79deed6 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Mon, 3 Nov 2025 11:50:10 +0000 Subject: [PATCH 0484/1381] net: stmmac: add stmmac_get_phy_intf_sel() Provide a function to translate the PHY interface mode to the phy_intf_sel pin configuration for dwmac1000 and dwmac4 cores that support multiple interfaces. We currently handle MII, GMII, RGMII, SGMII, RMII and REVMII, but not TBI, RTBI nor SMII as drivers do not appear to use these three and the driver doesn't currently support these. Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vFt4c-0000000Choe-3SII@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/stmmac.h | 1 + .../net/ethernet/stmicro/stmmac/stmmac_main.c | 20 +++++++++++++++++++ 2 files changed, 21 insertions(+) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac.h b/drivers/net/ethernet/stmicro/stmmac/stmmac.h index 3ea680cc63d8..0ea74c88a779 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac.h +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac.h @@ -396,6 +396,7 @@ void stmmac_ptp_register(struct stmmac_priv *priv); void stmmac_ptp_unregister(struct stmmac_priv *priv); int stmmac_xdp_open(struct net_device *dev); void stmmac_xdp_release(struct net_device *dev); +int stmmac_get_phy_intf_sel(phy_interface_t interface); int stmmac_resume(struct device *dev); int stmmac_suspend(struct device *dev); void stmmac_dvr_remove(struct device *dev); diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index c2a783c8022d..6d4323d04573 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -3082,6 +3082,26 @@ static void stmmac_check_ether_addr(struct stmmac_priv *priv) } } +int stmmac_get_phy_intf_sel(phy_interface_t interface) +{ + int phy_intf_sel = -EINVAL; + + if (interface == PHY_INTERFACE_MODE_MII || + interface == PHY_INTERFACE_MODE_GMII) + phy_intf_sel = PHY_INTF_SEL_GMII_MII; + else if (phy_interface_mode_is_rgmii(interface)) + phy_intf_sel = PHY_INTF_SEL_RGMII; + else if (interface == PHY_INTERFACE_MODE_SGMII) + phy_intf_sel = PHY_INTF_SEL_SGMII; + else if (interface == PHY_INTERFACE_MODE_RMII) + phy_intf_sel = PHY_INTF_SEL_RMII; + else if (interface == PHY_INTERFACE_MODE_REVMII) + phy_intf_sel = PHY_INTF_SEL_REVMII; + + return phy_intf_sel; +} +EXPORT_SYMBOL_GPL(stmmac_get_phy_intf_sel); + /** * stmmac_init_dma_engine - DMA init. * @priv: driver private structure From 1b6aa81c85621d6b55099906585ff09a477203b8 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Mon, 3 Nov 2025 11:50:15 +0000 Subject: [PATCH 0485/1381] net: stmmac: add support for configuring the phy_intf_sel inputs When dwmac is synthesised with support for multiple PHY interfaces, the core provides phy_intf_sel inputs, sampled on reset, to configure the PHY facing interface. Use stmmac_get_phy_intf_sel() in core code to determine the dwmac phy_intf_sel input value, and provide a new platform method called with this value just before we issue a soft reset to the dwmac core. Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vFt4h-0000000Chos-3wxX@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../net/ethernet/stmicro/stmmac/stmmac_main.c | 34 +++++++++++++++++++ include/linux/stmmac.h | 1 + 2 files changed, 35 insertions(+) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index 6d4323d04573..ccf383b355e7 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -3102,6 +3102,36 @@ int stmmac_get_phy_intf_sel(phy_interface_t interface) } EXPORT_SYMBOL_GPL(stmmac_get_phy_intf_sel); +static int stmmac_prereset_configure(struct stmmac_priv *priv) +{ + struct plat_stmmacenet_data *plat_dat = priv->plat; + phy_interface_t interface; + int phy_intf_sel, ret; + + if (!plat_dat->set_phy_intf_sel) + return 0; + + interface = plat_dat->phy_interface; + phy_intf_sel = stmmac_get_phy_intf_sel(interface); + if (phy_intf_sel < 0) { + netdev_err(priv->dev, + "failed to get phy_intf_sel for %s: %pe\n", + phy_modes(interface), ERR_PTR(phy_intf_sel)); + return phy_intf_sel; + } + + ret = plat_dat->set_phy_intf_sel(plat_dat->bsp_priv, phy_intf_sel); + if (ret == -EINVAL) + netdev_err(priv->dev, "platform does not support %s\n", + phy_modes(interface)); + else if (ret < 0) + netdev_err(priv->dev, + "platform failed to set interface %s: %pe\n", + phy_modes(interface), ERR_PTR(ret)); + + return ret; +} + /** * stmmac_init_dma_engine - DMA init. * @priv: driver private structure @@ -3128,6 +3158,10 @@ static int stmmac_init_dma_engine(struct stmmac_priv *priv) if (priv->extend_desc && (priv->mode == STMMAC_RING_MODE)) priv->plat->dma_cfg->atds = 1; + ret = stmmac_prereset_configure(priv); + if (ret) + return ret; + ret = stmmac_reset(priv, priv->ioaddr); if (ret) { netdev_err(priv->dev, "Failed to reset the dma\n"); diff --git a/include/linux/stmmac.h b/include/linux/stmmac.h index 151c81c560c8..48e9f1d4e17e 100644 --- a/include/linux/stmmac.h +++ b/include/linux/stmmac.h @@ -250,6 +250,7 @@ struct plat_stmmacenet_data { struct stmmac_txq_cfg tx_queues_cfg[MTL_MAX_TX_QUEUES]; void (*get_interfaces)(struct stmmac_priv *priv, void *bsp_priv, unsigned long *interfaces); + int (*set_phy_intf_sel)(void *priv, u8 phy_intf_sel); int (*set_clk_tx_rate)(void *priv, struct clk *clk_tx_i, phy_interface_t interface, int speed); void (*fix_mac_speed)(void *priv, int speed, unsigned int mode); From 8233cc439779eac1d2682d334c1aa6bb6d95120c Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Mon, 3 Nov 2025 11:50:21 +0000 Subject: [PATCH 0486/1381] net: stmmac: imx: convert to PHY_INTF_SEL_xxx Convert dwmac-imx to use the PHY_INTF_SEL_xxx definitions rather than constants via: - ensuring that the prefix for the MASK and value definitions is the same. - using FIELD_PREP() to shift the PHY_INTF_SEL_xxx definition to the appropriate bitfield. Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vFt4n-0000000Choy-0IeG@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../net/ethernet/stmicro/stmmac/dwmac-imx.c | 23 ++++++++++++------- 1 file changed, 15 insertions(+), 8 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c index 147fa08d5b6e..4fbee59e7337 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c @@ -23,18 +23,25 @@ #include "stmmac_platform.h" #define GPR_ENET_QOS_INTF_MODE_MASK GENMASK(21, 16) -#define GPR_ENET_QOS_INTF_SEL_MII (0x0 << 16) -#define GPR_ENET_QOS_INTF_SEL_RMII (0x4 << 16) -#define GPR_ENET_QOS_INTF_SEL_RGMII (0x1 << 16) +#define GPR_ENET_QOS_INTF_SEL_MASK GENMASK(20, 16) +#define GPR_ENET_QOS_INTF_SEL_MII FIELD_PREP(GPR_ENET_QOS_INTF_SEL_MASK, \ + PHY_INTF_SEL_GMII_MII) +#define GPR_ENET_QOS_INTF_SEL_RMII FIELD_PREP(GPR_ENET_QOS_INTF_SEL_MASK, \ + PHY_INTF_SEL_RMII) +#define GPR_ENET_QOS_INTF_SEL_RGMII FIELD_PREP(GPR_ENET_QOS_INTF_SEL_MASK, \ + PHY_INTF_SEL_RGMII) #define GPR_ENET_QOS_CLK_GEN_EN (0x1 << 19) #define GPR_ENET_QOS_CLK_TX_CLK_SEL (0x1 << 20) #define GPR_ENET_QOS_RGMII_EN (0x1 << 21) #define MX93_GPR_ENET_QOS_INTF_MODE_MASK GENMASK(3, 0) -#define MX93_GPR_ENET_QOS_INTF_MASK GENMASK(3, 1) -#define MX93_GPR_ENET_QOS_INTF_SEL_MII (0x0 << 1) -#define MX93_GPR_ENET_QOS_INTF_SEL_RMII (0x4 << 1) -#define MX93_GPR_ENET_QOS_INTF_SEL_RGMII (0x1 << 1) +#define MX93_GPR_ENET_QOS_INTF_SEL_MASK GENMASK(3, 1) +#define MX93_GPR_ENET_QOS_INTF_SEL_MII FIELD_PREP(MX93_GPR_ENET_QOS_INTF_SEL_MASK, \ + PHY_INTF_SEL_GMII_MII) +#define MX93_GPR_ENET_QOS_INTF_SEL_RMII FIELD_PREP(MX93_GPR_ENET_QOS_INTF_SEL_MASK, \ + PHY_INTF_SEL_RMII) +#define MX93_GPR_ENET_QOS_INTF_SEL_RGMII FIELD_PREP(MX93_GPR_ENET_QOS_INTF_SEL_MASK, \ + PHY_INTF_SEL_RGMII) #define MX93_GPR_ENET_QOS_CLK_GEN_EN (0x1 << 0) #define MX93_GPR_ENET_QOS_CLK_SEL_MASK BIT_MASK(0) #define MX93_GPR_CLK_SEL_OFFSET (4) @@ -241,7 +248,7 @@ static void imx93_dwmac_fix_speed(void *priv, int speed, unsigned int mode) if (regmap_read(dwmac->intf_regmap, dwmac->intf_reg_off, &iface)) return; - iface &= MX93_GPR_ENET_QOS_INTF_MASK; + iface &= MX93_GPR_ENET_QOS_INTF_SEL_MASK; if (iface != MX93_GPR_ENET_QOS_INTF_SEL_RGMII) return; From d73c1dccfb9909f0e2d517af887fe414ab421cea Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Mon, 3 Nov 2025 11:50:26 +0000 Subject: [PATCH 0487/1381] net: stmmac: imx: use FIELD_PREP()/FIELD_GET() for PHY_INTF_SEL_x Use FIELD_PREP()/FIELD_GET() in the functions to construct the PHY interface selection bitfield or to extract its value. Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vFt4s-0000000Chp4-0kwf@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../net/ethernet/stmicro/stmmac/dwmac-imx.c | 44 +++++++++---------- 1 file changed, 20 insertions(+), 24 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c index 4fbee59e7337..f1cfccd4269c 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c @@ -24,24 +24,12 @@ #define GPR_ENET_QOS_INTF_MODE_MASK GENMASK(21, 16) #define GPR_ENET_QOS_INTF_SEL_MASK GENMASK(20, 16) -#define GPR_ENET_QOS_INTF_SEL_MII FIELD_PREP(GPR_ENET_QOS_INTF_SEL_MASK, \ - PHY_INTF_SEL_GMII_MII) -#define GPR_ENET_QOS_INTF_SEL_RMII FIELD_PREP(GPR_ENET_QOS_INTF_SEL_MASK, \ - PHY_INTF_SEL_RMII) -#define GPR_ENET_QOS_INTF_SEL_RGMII FIELD_PREP(GPR_ENET_QOS_INTF_SEL_MASK, \ - PHY_INTF_SEL_RGMII) #define GPR_ENET_QOS_CLK_GEN_EN (0x1 << 19) #define GPR_ENET_QOS_CLK_TX_CLK_SEL (0x1 << 20) #define GPR_ENET_QOS_RGMII_EN (0x1 << 21) #define MX93_GPR_ENET_QOS_INTF_MODE_MASK GENMASK(3, 0) #define MX93_GPR_ENET_QOS_INTF_SEL_MASK GENMASK(3, 1) -#define MX93_GPR_ENET_QOS_INTF_SEL_MII FIELD_PREP(MX93_GPR_ENET_QOS_INTF_SEL_MASK, \ - PHY_INTF_SEL_GMII_MII) -#define MX93_GPR_ENET_QOS_INTF_SEL_RMII FIELD_PREP(MX93_GPR_ENET_QOS_INTF_SEL_MASK, \ - PHY_INTF_SEL_RMII) -#define MX93_GPR_ENET_QOS_INTF_SEL_RGMII FIELD_PREP(MX93_GPR_ENET_QOS_INTF_SEL_MASK, \ - PHY_INTF_SEL_RGMII) #define MX93_GPR_ENET_QOS_CLK_GEN_EN (0x1 << 0) #define MX93_GPR_ENET_QOS_CLK_SEL_MASK BIT_MASK(0) #define MX93_GPR_CLK_SEL_OFFSET (4) @@ -77,22 +65,24 @@ struct imx_priv_data { static int imx8mp_set_intf_mode(struct plat_stmmacenet_data *plat_dat) { struct imx_priv_data *dwmac = plat_dat->bsp_priv; + u8 phy_intf_sel; int val; switch (plat_dat->phy_interface) { case PHY_INTERFACE_MODE_MII: - val = GPR_ENET_QOS_INTF_SEL_MII; + phy_intf_sel = PHY_INTF_SEL_GMII_MII; + val = 0; break; case PHY_INTERFACE_MODE_RMII: - val = GPR_ENET_QOS_INTF_SEL_RMII; - val |= (dwmac->rmii_refclk_ext ? 0 : GPR_ENET_QOS_CLK_TX_CLK_SEL); + phy_intf_sel = PHY_INTF_SEL_RMII; + val = dwmac->rmii_refclk_ext ? 0 : GPR_ENET_QOS_CLK_TX_CLK_SEL; break; case PHY_INTERFACE_MODE_RGMII: case PHY_INTERFACE_MODE_RGMII_ID: case PHY_INTERFACE_MODE_RGMII_RXID: case PHY_INTERFACE_MODE_RGMII_TXID: - val = GPR_ENET_QOS_INTF_SEL_RGMII | - GPR_ENET_QOS_RGMII_EN; + phy_intf_sel = PHY_INTF_SEL_RGMII; + val = GPR_ENET_QOS_RGMII_EN; break; default: pr_debug("imx dwmac doesn't support %s interface\n", @@ -100,7 +90,9 @@ static int imx8mp_set_intf_mode(struct plat_stmmacenet_data *plat_dat) return -EINVAL; } - val |= GPR_ENET_QOS_CLK_GEN_EN; + val |= FIELD_PREP(GPR_ENET_QOS_INTF_SEL_MASK, phy_intf_sel) | + GPR_ENET_QOS_CLK_GEN_EN; + return regmap_update_bits(dwmac->intf_regmap, dwmac->intf_reg_off, GPR_ENET_QOS_INTF_MODE_MASK, val); }; @@ -117,11 +109,12 @@ imx8dxl_set_intf_mode(struct plat_stmmacenet_data *plat_dat) static int imx93_set_intf_mode(struct plat_stmmacenet_data *plat_dat) { struct imx_priv_data *dwmac = plat_dat->bsp_priv; + u8 phy_intf_sel; int val, ret; switch (plat_dat->phy_interface) { case PHY_INTERFACE_MODE_MII: - val = MX93_GPR_ENET_QOS_INTF_SEL_MII; + phy_intf_sel = PHY_INTF_SEL_GMII_MII; break; case PHY_INTERFACE_MODE_RMII: if (dwmac->rmii_refclk_ext) { @@ -132,13 +125,13 @@ static int imx93_set_intf_mode(struct plat_stmmacenet_data *plat_dat) if (ret) return ret; } - val = MX93_GPR_ENET_QOS_INTF_SEL_RMII; + phy_intf_sel = PHY_INTF_SEL_RMII; break; case PHY_INTERFACE_MODE_RGMII: case PHY_INTERFACE_MODE_RGMII_ID: case PHY_INTERFACE_MODE_RGMII_RXID: case PHY_INTERFACE_MODE_RGMII_TXID: - val = MX93_GPR_ENET_QOS_INTF_SEL_RGMII; + phy_intf_sel = PHY_INTF_SEL_RGMII; break; default: dev_dbg(dwmac->dev, "imx dwmac doesn't support %s interface\n", @@ -146,7 +139,9 @@ static int imx93_set_intf_mode(struct plat_stmmacenet_data *plat_dat) return -EINVAL; } - val |= MX93_GPR_ENET_QOS_CLK_GEN_EN; + val = FIELD_PREP(MX93_GPR_ENET_QOS_INTF_SEL_MASK, phy_intf_sel) | + MX93_GPR_ENET_QOS_CLK_GEN_EN; + return regmap_update_bits(dwmac->intf_regmap, dwmac->intf_reg_off, MX93_GPR_ENET_QOS_INTF_MODE_MASK, val); }; @@ -248,8 +243,8 @@ static void imx93_dwmac_fix_speed(void *priv, int speed, unsigned int mode) if (regmap_read(dwmac->intf_regmap, dwmac->intf_reg_off, &iface)) return; - iface &= MX93_GPR_ENET_QOS_INTF_SEL_MASK; - if (iface != MX93_GPR_ENET_QOS_INTF_SEL_RGMII) + if (FIELD_GET(MX93_GPR_ENET_QOS_INTF_SEL_MASK, iface) != + PHY_INTF_SEL_RGMII) return; old_ctrl = readl(dwmac->base_addr + MAC_CTRL_REG); @@ -262,6 +257,7 @@ static void imx93_dwmac_fix_speed(void *priv, int speed, unsigned int mode) readl(dwmac->base_addr + MAC_CTRL_REG); usleep_range(10, 20); + iface &= MX93_GPR_ENET_QOS_INTF_SEL_MASK; iface |= MX93_GPR_ENET_QOS_CLK_GEN_EN; regmap_update_bits(dwmac->intf_regmap, dwmac->intf_reg_off, MX93_GPR_ENET_QOS_INTF_MODE_MASK, iface); From c012710c14a70dfa21691e2542d18dd4b621c518 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Mon, 3 Nov 2025 11:50:31 +0000 Subject: [PATCH 0488/1381] net: stmmac: imx: use stmmac_get_phy_intf_sel() i.MX implementations other than IMX8DXL involve setting the dwmac core phy_intf_sel input. Use stmmac_get_phy_intf_sel() to decode the PHY interface mode to the phy_intf_sel value, validating the result, and passing it into the implementation specific .set_intf_mode() method rather than each .set_intf_mode() method doing this. Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vFt4x-0000000ChpA-1Edr@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../net/ethernet/stmicro/stmmac/dwmac-imx.c | 43 +++++++++++-------- 1 file changed, 25 insertions(+), 18 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c index f1cfccd4269c..dc28486a7af0 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c @@ -45,7 +45,8 @@ struct imx_dwmac_ops { bool mac_rgmii_txclk_auto_adj; int (*fix_soc_reset)(struct stmmac_priv *priv, void __iomem *ioaddr); - int (*set_intf_mode)(struct plat_stmmacenet_data *plat_dat); + int (*set_intf_mode)(struct plat_stmmacenet_data *plat_dat, + u8 phy_intf_sel); void (*fix_mac_speed)(void *priv, int speed, unsigned int mode); }; @@ -62,26 +63,23 @@ struct imx_priv_data { struct plat_stmmacenet_data *plat_dat; }; -static int imx8mp_set_intf_mode(struct plat_stmmacenet_data *plat_dat) +static int imx8mp_set_intf_mode(struct plat_stmmacenet_data *plat_dat, + u8 phy_intf_sel) { struct imx_priv_data *dwmac = plat_dat->bsp_priv; - u8 phy_intf_sel; int val; switch (plat_dat->phy_interface) { case PHY_INTERFACE_MODE_MII: - phy_intf_sel = PHY_INTF_SEL_GMII_MII; val = 0; break; case PHY_INTERFACE_MODE_RMII: - phy_intf_sel = PHY_INTF_SEL_RMII; val = dwmac->rmii_refclk_ext ? 0 : GPR_ENET_QOS_CLK_TX_CLK_SEL; break; case PHY_INTERFACE_MODE_RGMII: case PHY_INTERFACE_MODE_RGMII_ID: case PHY_INTERFACE_MODE_RGMII_RXID: case PHY_INTERFACE_MODE_RGMII_TXID: - phy_intf_sel = PHY_INTF_SEL_RGMII; val = GPR_ENET_QOS_RGMII_EN; break; default: @@ -98,7 +96,8 @@ static int imx8mp_set_intf_mode(struct plat_stmmacenet_data *plat_dat) }; static int -imx8dxl_set_intf_mode(struct plat_stmmacenet_data *plat_dat) +imx8dxl_set_intf_mode(struct plat_stmmacenet_data *plat_dat, + u8 phy_intf_sel) { int ret = 0; @@ -106,16 +105,13 @@ imx8dxl_set_intf_mode(struct plat_stmmacenet_data *plat_dat) return ret; } -static int imx93_set_intf_mode(struct plat_stmmacenet_data *plat_dat) +static int imx93_set_intf_mode(struct plat_stmmacenet_data *plat_dat, + u8 phy_intf_sel) { struct imx_priv_data *dwmac = plat_dat->bsp_priv; - u8 phy_intf_sel; int val, ret; switch (plat_dat->phy_interface) { - case PHY_INTERFACE_MODE_MII: - phy_intf_sel = PHY_INTF_SEL_GMII_MII; - break; case PHY_INTERFACE_MODE_RMII: if (dwmac->rmii_refclk_ext) { ret = regmap_clear_bits(dwmac->intf_regmap, @@ -125,13 +121,12 @@ static int imx93_set_intf_mode(struct plat_stmmacenet_data *plat_dat) if (ret) return ret; } - phy_intf_sel = PHY_INTF_SEL_RMII; break; + case PHY_INTERFACE_MODE_MII: case PHY_INTERFACE_MODE_RGMII: case PHY_INTERFACE_MODE_RGMII_ID: case PHY_INTERFACE_MODE_RGMII_RXID: case PHY_INTERFACE_MODE_RGMII_TXID: - phy_intf_sel = PHY_INTF_SEL_RGMII; break; default: dev_dbg(dwmac->dev, "imx dwmac doesn't support %s interface\n", @@ -176,12 +171,24 @@ static int imx_dwmac_init(struct platform_device *pdev, void *priv) { struct plat_stmmacenet_data *plat_dat; struct imx_priv_data *dwmac = priv; - int ret; - - plat_dat = dwmac->plat_dat; + phy_interface_t interface; + int phy_intf_sel, ret; if (dwmac->ops->set_intf_mode) { - ret = dwmac->ops->set_intf_mode(plat_dat); + plat_dat = dwmac->plat_dat; + interface = plat_dat->phy_interface; + + phy_intf_sel = stmmac_get_phy_intf_sel(interface); + if (phy_intf_sel != PHY_INTF_SEL_GMII_MII && + phy_intf_sel != PHY_INTF_SEL_RGMII && + phy_intf_sel != PHY_INTF_SEL_RMII) { + dev_dbg(dwmac->dev, + "imx dwmac doesn't support %s interface\n", + phy_modes(interface)); + return phy_intf_sel < 0 ? phy_intf_sel : -EINVAL; + } + + ret = dwmac->ops->set_intf_mode(plat_dat, phy_intf_sel); if (ret) return ret; } From 35103babce3036058cd9ed8674c98e9ab397d715 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Mon, 3 Nov 2025 11:50:36 +0000 Subject: [PATCH 0489/1381] net: stmmac: imx: simplify set_intf_mode() implementations Simplify the set_intf_mode() implementations, testing the phy_intf_sel value rather than the PHY interface mode. Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vFt52-0000000ChpG-1bsd@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../net/ethernet/stmicro/stmmac/dwmac-imx.c | 63 +++++-------------- 1 file changed, 17 insertions(+), 46 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c index dc28486a7af0..d69be9de4468 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c @@ -67,29 +67,15 @@ static int imx8mp_set_intf_mode(struct plat_stmmacenet_data *plat_dat, u8 phy_intf_sel) { struct imx_priv_data *dwmac = plat_dat->bsp_priv; - int val; + unsigned int val; - switch (plat_dat->phy_interface) { - case PHY_INTERFACE_MODE_MII: - val = 0; - break; - case PHY_INTERFACE_MODE_RMII: - val = dwmac->rmii_refclk_ext ? 0 : GPR_ENET_QOS_CLK_TX_CLK_SEL; - break; - case PHY_INTERFACE_MODE_RGMII: - case PHY_INTERFACE_MODE_RGMII_ID: - case PHY_INTERFACE_MODE_RGMII_RXID: - case PHY_INTERFACE_MODE_RGMII_TXID: - val = GPR_ENET_QOS_RGMII_EN; - break; - default: - pr_debug("imx dwmac doesn't support %s interface\n", - phy_modes(plat_dat->phy_interface)); - return -EINVAL; - } + val = FIELD_PREP(GPR_ENET_QOS_INTF_SEL_MASK, phy_intf_sel) | + GPR_ENET_QOS_CLK_GEN_EN; - val |= FIELD_PREP(GPR_ENET_QOS_INTF_SEL_MASK, phy_intf_sel) | - GPR_ENET_QOS_CLK_GEN_EN; + if (phy_intf_sel == PHY_INTF_SEL_RMII && !dwmac->rmii_refclk_ext) + val |= GPR_ENET_QOS_CLK_TX_CLK_SEL; + else if (phy_intf_sel == PHY_INTF_SEL_RGMII) + val |= GPR_ENET_QOS_RGMII_EN; return regmap_update_bits(dwmac->intf_regmap, dwmac->intf_reg_off, GPR_ENET_QOS_INTF_MODE_MASK, val); @@ -99,39 +85,24 @@ static int imx8dxl_set_intf_mode(struct plat_stmmacenet_data *plat_dat, u8 phy_intf_sel) { - int ret = 0; - /* TBD: depends on imx8dxl scu interfaces to be upstreamed */ - return ret; + return 0; } static int imx93_set_intf_mode(struct plat_stmmacenet_data *plat_dat, u8 phy_intf_sel) { struct imx_priv_data *dwmac = plat_dat->bsp_priv; - int val, ret; + unsigned int val; + int ret; - switch (plat_dat->phy_interface) { - case PHY_INTERFACE_MODE_RMII: - if (dwmac->rmii_refclk_ext) { - ret = regmap_clear_bits(dwmac->intf_regmap, - dwmac->intf_reg_off + - MX93_GPR_CLK_SEL_OFFSET, - MX93_GPR_ENET_QOS_CLK_SEL_MASK); - if (ret) - return ret; - } - break; - case PHY_INTERFACE_MODE_MII: - case PHY_INTERFACE_MODE_RGMII: - case PHY_INTERFACE_MODE_RGMII_ID: - case PHY_INTERFACE_MODE_RGMII_RXID: - case PHY_INTERFACE_MODE_RGMII_TXID: - break; - default: - dev_dbg(dwmac->dev, "imx dwmac doesn't support %s interface\n", - phy_modes(plat_dat->phy_interface)); - return -EINVAL; + if (phy_intf_sel == PHY_INTF_SEL_RMII && dwmac->rmii_refclk_ext) { + ret = regmap_clear_bits(dwmac->intf_regmap, + dwmac->intf_reg_off + + MX93_GPR_CLK_SEL_OFFSET, + MX93_GPR_ENET_QOS_CLK_SEL_MASK); + if (ret) + return ret; } val = FIELD_PREP(MX93_GPR_ENET_QOS_INTF_SEL_MASK, phy_intf_sel) | From 38cd4e84b369c11680966fdea129e11dbb28a6ec Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Mon, 3 Nov 2025 11:50:41 +0000 Subject: [PATCH 0490/1381] net: stmmac: imx: cleanup arguments for set_intf_mode() method Pass the imx_priv_data instead of the plat_stmmacenet_data into the set_intf_mode() SoC specific methods. Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vFt57-0000000ChpL-25kS@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../net/ethernet/stmicro/stmmac/dwmac-imx.c | 22 +++++++------------ 1 file changed, 8 insertions(+), 14 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c index d69be9de4468..ae1b73e1bcb2 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c @@ -39,14 +39,15 @@ #define RMII_RESET_SPEED (0x3 << 14) #define CTRL_SPEED_MASK GENMASK(15, 14) +struct imx_priv_data; + struct imx_dwmac_ops { u32 addr_width; u32 flags; bool mac_rgmii_txclk_auto_adj; int (*fix_soc_reset)(struct stmmac_priv *priv, void __iomem *ioaddr); - int (*set_intf_mode)(struct plat_stmmacenet_data *plat_dat, - u8 phy_intf_sel); + int (*set_intf_mode)(struct imx_priv_data *dwmac, u8 phy_intf_sel); void (*fix_mac_speed)(void *priv, int speed, unsigned int mode); }; @@ -63,10 +64,8 @@ struct imx_priv_data { struct plat_stmmacenet_data *plat_dat; }; -static int imx8mp_set_intf_mode(struct plat_stmmacenet_data *plat_dat, - u8 phy_intf_sel) +static int imx8mp_set_intf_mode(struct imx_priv_data *dwmac, u8 phy_intf_sel) { - struct imx_priv_data *dwmac = plat_dat->bsp_priv; unsigned int val; val = FIELD_PREP(GPR_ENET_QOS_INTF_SEL_MASK, phy_intf_sel) | @@ -82,17 +81,14 @@ static int imx8mp_set_intf_mode(struct plat_stmmacenet_data *plat_dat, }; static int -imx8dxl_set_intf_mode(struct plat_stmmacenet_data *plat_dat, - u8 phy_intf_sel) +imx8dxl_set_intf_mode(struct imx_priv_data *dwmac, u8 phy_intf_sel) { /* TBD: depends on imx8dxl scu interfaces to be upstreamed */ return 0; } -static int imx93_set_intf_mode(struct plat_stmmacenet_data *plat_dat, - u8 phy_intf_sel) +static int imx93_set_intf_mode(struct imx_priv_data *dwmac, u8 phy_intf_sel) { - struct imx_priv_data *dwmac = plat_dat->bsp_priv; unsigned int val; int ret; @@ -140,14 +136,12 @@ static int imx_dwmac_clks_config(void *priv, bool enabled) static int imx_dwmac_init(struct platform_device *pdev, void *priv) { - struct plat_stmmacenet_data *plat_dat; struct imx_priv_data *dwmac = priv; phy_interface_t interface; int phy_intf_sel, ret; if (dwmac->ops->set_intf_mode) { - plat_dat = dwmac->plat_dat; - interface = plat_dat->phy_interface; + interface = dwmac->plat_dat->phy_interface; phy_intf_sel = stmmac_get_phy_intf_sel(interface); if (phy_intf_sel != PHY_INTF_SEL_GMII_MII && @@ -159,7 +153,7 @@ static int imx_dwmac_init(struct platform_device *pdev, void *priv) return phy_intf_sel < 0 ? phy_intf_sel : -EINVAL; } - ret = dwmac->ops->set_intf_mode(plat_dat, phy_intf_sel); + ret = dwmac->ops->set_intf_mode(dwmac, phy_intf_sel); if (ret) return ret; } From eaca1a4dc51e5e4979e45a4ad72a1c2a88a80a72 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Mon, 3 Nov 2025 11:50:46 +0000 Subject: [PATCH 0491/1381] net: stmmac: imx: use ->set_phy_intf_sel() Rather than placing the phy_intf_sel() setup in the ->init() method, move it to the new ->set_phy_intf_sel() method. Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vFt5C-0000000ChpR-2kAB@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../net/ethernet/stmicro/stmmac/dwmac-imx.c | 38 +++++-------------- 1 file changed, 10 insertions(+), 28 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c index ae1b73e1bcb2..db288fbd5a4d 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c @@ -134,36 +134,19 @@ static int imx_dwmac_clks_config(void *priv, bool enabled) return ret; } -static int imx_dwmac_init(struct platform_device *pdev, void *priv) +static int imx_set_phy_intf_sel(void *bsp_priv, u8 phy_intf_sel) { - struct imx_priv_data *dwmac = priv; - phy_interface_t interface; - int phy_intf_sel, ret; + struct imx_priv_data *dwmac = bsp_priv; - if (dwmac->ops->set_intf_mode) { - interface = dwmac->plat_dat->phy_interface; + if (!dwmac->ops->set_intf_mode) + return 0; - phy_intf_sel = stmmac_get_phy_intf_sel(interface); - if (phy_intf_sel != PHY_INTF_SEL_GMII_MII && - phy_intf_sel != PHY_INTF_SEL_RGMII && - phy_intf_sel != PHY_INTF_SEL_RMII) { - dev_dbg(dwmac->dev, - "imx dwmac doesn't support %s interface\n", - phy_modes(interface)); - return phy_intf_sel < 0 ? phy_intf_sel : -EINVAL; - } + if (phy_intf_sel != PHY_INTF_SEL_GMII_MII && + phy_intf_sel != PHY_INTF_SEL_RGMII && + phy_intf_sel != PHY_INTF_SEL_RMII) + return -EINVAL; - ret = dwmac->ops->set_intf_mode(dwmac, phy_intf_sel); - if (ret) - return ret; - } - - return 0; -} - -static void imx_dwmac_exit(struct platform_device *pdev, void *priv) -{ - /* nothing to do now */ + return dwmac->ops->set_intf_mode(dwmac, phy_intf_sel); } static int imx_dwmac_set_clk_tx_rate(void *bsp_priv, struct clk *clk_tx_i, @@ -342,8 +325,7 @@ static int imx_dwmac_probe(struct platform_device *pdev) plat_dat->tx_queues_cfg[i].tbs_en = 1; plat_dat->host_dma_width = dwmac->ops->addr_width; - plat_dat->init = imx_dwmac_init; - plat_dat->exit = imx_dwmac_exit; + plat_dat->set_phy_intf_sel = imx_set_phy_intf_sel; plat_dat->clks_config = imx_dwmac_clks_config; plat_dat->bsp_priv = dwmac; dwmac->plat_dat = plat_dat; From f4b2786fb14bef2b16c66be076e41863da1e511b Mon Sep 17 00:00:00 2001 From: Chu Guangqing Date: Mon, 3 Nov 2025 15:43:05 +0800 Subject: [PATCH 0492/1381] virtio_net: Fix a typo error in virtio_net Fix the spelling error of "separate". Signed-off-by: Chu Guangqing Reviewed-by: Jacob Keller Acked-by: Michael S. Tsirkin Link: https://patch.msgid.link/20251103074305.4727-1-chuguangqing@inspur.com Signed-off-by: Jakub Kicinski --- drivers/net/virtio_net.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index 8e8a179aaa49..1e6f5e650f11 100644 --- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -3760,7 +3760,7 @@ static int virtnet_set_queues(struct virtnet_info *vi, u16 queue_pairs) * (2) no user configuration. * * During rss command processing, device updates queue_pairs using rss.max_tx_vq. That is, - * the device updates queue_pairs together with rss, so we can skip the sperate queue_pairs + * the device updates queue_pairs together with rss, so we can skip the separate queue_pairs * update (VIRTIO_NET_CTRL_MQ_VQ_PAIRS_SET below) and return directly. */ if (vi->has_rss && !netif_is_rxfh_configured(dev)) { From 2428803d5eef1fd8451a6c4ba41d17cd199f3715 Mon Sep 17 00:00:00 2001 From: Chu Guangqing Date: Mon, 3 Nov 2025 14:05:04 +0800 Subject: [PATCH 0493/1381] gtp: Fix a typo error for size Fix the spelling error of "size". Signed-off-by: Chu Guangqing Reviewed-by: Jacob Keller Link: https://patch.msgid.link/20251103060504.3524-1-chuguangqing@inspur.com Signed-off-by: Jakub Kicinski --- drivers/net/gtp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/gtp.c b/drivers/net/gtp.c index 5cb59d72bc82..4213c3b2d532 100644 --- a/drivers/net/gtp.c +++ b/drivers/net/gtp.c @@ -633,7 +633,7 @@ static void gtp1u_build_echo_msg(struct gtp1_header_long *hdr, __u8 msg_type) hdr->tid = 0; /* seq, npdu and next should be counted to the length of the GTP packet - * that's why szie of gtp1_header should be subtracted, + * that's why size of gtp1_header should be subtracted, * not size of gtp1_header_long. */ From 9781642e58903e52f3c607e957e5220e95e792b6 Mon Sep 17 00:00:00 2001 From: Chu Guangqing Date: Mon, 3 Nov 2025 13:53:51 +0800 Subject: [PATCH 0494/1381] veth: Fix a typo error in veth Fix a spellling error for resources Signed-off-by: Chu Guangqing Reviewed-by: Jacob Keller Link: https://patch.msgid.link/20251103055351.3150-1-chuguangqing@inspur.com Signed-off-by: Jakub Kicinski --- drivers/net/veth.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/veth.c b/drivers/net/veth.c index a3046142cb8e..87a63c4bee77 100644 --- a/drivers/net/veth.c +++ b/drivers/net/veth.c @@ -1323,7 +1323,7 @@ static int veth_set_channels(struct net_device *dev, if (peer) netif_carrier_off(peer); - /* try to allocate new resurces, as needed*/ + /* try to allocate new resources, as needed*/ err = veth_enable_range_safe(dev, old_rx_count, new_rx_count); if (err) goto out; From 96c68954cd3b65f321d40b9078cdb49a26bdaf9a Mon Sep 17 00:00:00 2001 From: Chu Guangqing Date: Mon, 3 Nov 2025 13:44:43 +0800 Subject: [PATCH 0495/1381] net: sungem_phy: Fix a typo error in sungem_phy Fix a spelling mistakes for regularly Signed-off-by: Chu Guangqing Reviewed-by: Simon Horman Link: https://patch.msgid.link/20251103054443.2878-1-chuguangqing@inspur.com Signed-off-by: Jakub Kicinski --- drivers/net/sungem_phy.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/sungem_phy.c b/drivers/net/sungem_phy.c index 55aa8d0c8e1f..c10198d44576 100644 --- a/drivers/net/sungem_phy.c +++ b/drivers/net/sungem_phy.c @@ -1165,7 +1165,7 @@ int sungem_phy_probe(struct mii_phy *phy, int mii_id) int i; /* We do not reset the mii_phy structure as the driver - * may re-probe the PHY regulary + * may re-probe the PHY regularly */ phy->mii_id = mii_id; From 52665fcc2241f8f9a17543d9a6531b1a1b029bde Mon Sep 17 00:00:00 2001 From: Chu Guangqing Date: Mon, 3 Nov 2025 11:22:12 +0800 Subject: [PATCH 0496/1381] xen/netfront: Comment Correction: Fix Spelling Error and Description of Queue Quantity Rules The original comments contained spelling errors and incomplete logical descriptions, which could easily lead to misunderstandings of the code logic. The specific modifications are as follows: Correct the spelling error by changing "inut max" to "but not exceed the maximum limit"; Add the note "If the user has not specified a value, the default maximum limit is 8" to clarify the default value logic; Improve the coherence of the statement to make the queue quantity rules clearer. After the modification, the comments can accurately reflect the code behavior of "taking the smaller value between the number of CPUs and the default maximum limit of 8 for the number of queues", enhancing code maintainability. Signed-off-by: Chu Guangqing Reviewed-by: Juergen Gross Link: https://patch.msgid.link/20251103032212.2462-1-chuguangqing@inspur.com Signed-off-by: Jakub Kicinski --- drivers/net/xen-netfront.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/net/xen-netfront.c b/drivers/net/xen-netfront.c index a11a0e949400..7c2220366623 100644 --- a/drivers/net/xen-netfront.c +++ b/drivers/net/xen-netfront.c @@ -2696,8 +2696,9 @@ static int __init netif_init(void) pr_info("Initialising Xen virtual ethernet driver\n"); - /* Allow as many queues as there are CPUs inut max. 8 if user has not - * specified a value. + /* Allow the number of queues to match the number of CPUs, but not exceed + * the maximum limit. If the user has not specified a value, the default + * maximum limit is 8. */ if (xennet_max_queues == 0) xennet_max_queues = min_t(unsigned int, MAX_QUEUES_DEFAULT, From 091400a5d411ee7398095ba832361eb12b345f3d Mon Sep 17 00:00:00 2001 From: Tariq Toukan Date: Thu, 30 Oct 2025 15:32:33 +0200 Subject: [PATCH 0497/1381] net/mlx5e: Enhance function structures for self loopback prevention application The re-application of self loopback prevention attributes in TIRs is necessary in old firmwares (where tis_tir_td_order cap is cleared) after recreation of SQs. However, this is not needed in new firmware with tis_tir_td_order=1. As a preparation patch, enhance the function structures to differentiate between an explicit loopback prevention configuration apply, and the re-apply operation required by old firmware. Loopback selftests should now call mlx5e_modify_tirs_lb() directly, as their use case is not related to the firmware limitation. Signed-off-by: Tariq Toukan Reviewed-by: Carolina Jubran Reviewed-by: Simon Horman Link: https://patch.msgid.link/1761831159-1013140-2-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mellanox/mlx5/core/en.h | 4 +++- .../net/ethernet/mellanox/mlx5/core/en_common.c | 16 ++++++++++++---- .../net/ethernet/mellanox/mlx5/core/en_main.c | 2 +- .../ethernet/mellanox/mlx5/core/en_selftest.c | 4 ++-- .../ethernet/mellanox/mlx5/core/ipoib/ipoib.c | 2 +- 5 files changed, 19 insertions(+), 9 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index fd107906bc28..4a29333285c0 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -1156,7 +1156,9 @@ extern const struct ethtool_ops mlx5e_ethtool_ops; int mlx5e_create_mkey(struct mlx5_core_dev *mdev, u32 pdn, u32 *mkey); int mlx5e_create_mdev_resources(struct mlx5_core_dev *mdev, bool create_tises); void mlx5e_destroy_mdev_resources(struct mlx5_core_dev *mdev); -int mlx5e_refresh_tirs(struct mlx5e_priv *priv, bool enable_uc_lb, +int mlx5e_modify_tirs_lb(struct mlx5_core_dev *mdev, bool enable_uc_lb, + bool enable_mc_lb); +int mlx5e_refresh_tirs(struct mlx5_core_dev *mdev, bool enable_uc_lb, bool enable_mc_lb); void mlx5e_mkey_set_relaxed_ordering(struct mlx5_core_dev *mdev, void *mkc); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_common.c b/drivers/net/ethernet/mellanox/mlx5/core/en_common.c index 30424ccad584..376a018b2db1 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_common.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_common.c @@ -247,10 +247,9 @@ void mlx5e_destroy_mdev_resources(struct mlx5_core_dev *mdev) memset(res, 0, sizeof(*res)); } -int mlx5e_refresh_tirs(struct mlx5e_priv *priv, bool enable_uc_lb, - bool enable_mc_lb) +int mlx5e_modify_tirs_lb(struct mlx5_core_dev *mdev, bool enable_uc_lb, + bool enable_mc_lb) { - struct mlx5_core_dev *mdev = priv->mdev; struct mlx5e_tir *tir; u8 lb_flags = 0; int err = 0; @@ -285,7 +284,16 @@ int mlx5e_refresh_tirs(struct mlx5e_priv *priv, bool enable_uc_lb, kvfree(in); if (err) - netdev_err(priv->netdev, "refresh tir(0x%x) failed, %d\n", tirn, err); + mlx5_core_err(mdev, + "modify tir(0x%x) enable_lb uc(%d) mc(%d) failed, %d\n", + tirn, + enable_uc_lb, enable_mc_lb, err); return err; } + +int mlx5e_refresh_tirs(struct mlx5_core_dev *mdev, bool enable_uc_lb, + bool enable_mc_lb) +{ + return mlx5e_modify_tirs_lb(mdev, enable_uc_lb, enable_mc_lb); +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index 6b905848fe86..f0d7a61e014b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -6136,7 +6136,7 @@ static void mlx5e_nic_disable(struct mlx5e_priv *priv) static int mlx5e_update_nic_rx(struct mlx5e_priv *priv) { - return mlx5e_refresh_tirs(priv, false, false); + return mlx5e_refresh_tirs(priv->mdev, false, false); } static const struct mlx5e_profile mlx5e_nic_profile = { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_selftest.c b/drivers/net/ethernet/mellanox/mlx5/core/en_selftest.c index 2f7a543feca6..fcad464bc4d5 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_selftest.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_selftest.c @@ -214,7 +214,7 @@ static int mlx5e_test_loopback_setup(struct mlx5e_priv *priv, return err; } - err = mlx5e_refresh_tirs(priv, true, false); + err = mlx5e_modify_tirs_lb(priv->mdev, true, false); if (err) goto out; @@ -243,7 +243,7 @@ static void mlx5e_test_loopback_cleanup(struct mlx5e_priv *priv, mlx5_nic_vport_update_local_lb(priv->mdev, false); dev_remove_pack(&lbtp->pt); - mlx5e_refresh_tirs(priv, false, false); + mlx5e_modify_tirs_lb(priv->mdev, false, false); } static int mlx5e_cond_loopback(struct mlx5e_priv *priv) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c index 906b1fbc27aa..976347ac1faf 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c @@ -334,7 +334,7 @@ void mlx5i_destroy_underlay_qp(struct mlx5_core_dev *mdev, u32 qpn) int mlx5i_update_nic_rx(struct mlx5e_priv *priv) { - return mlx5e_refresh_tirs(priv, true, true); + return mlx5e_refresh_tirs(priv->mdev, true, true); } int mlx5i_create_tis(struct mlx5_core_dev *mdev, u32 underlay_qpn, u32 *tisn) From 5c51a86122b20326229c6c9dff4a92c186cbb6bf Mon Sep 17 00:00:00 2001 From: Tariq Toukan Date: Thu, 30 Oct 2025 15:32:34 +0200 Subject: [PATCH 0498/1381] net/mlx5e: Use TIR API in mlx5e_modify_tirs_lb() Extend the TIR API and use it in mlx5e_modify_tirs_lb() instead of the explicit modify_tir code. Signed-off-by: Tariq Toukan Reviewed-by: Carolina Jubran Reviewed-by: Dragos Tatulea Reviewed-by: Simon Horman Link: https://patch.msgid.link/1761831159-1013140-3-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- .../net/ethernet/mellanox/mlx5/core/en/tir.c | 29 +++++++++++-- .../net/ethernet/mellanox/mlx5/core/en/tir.h | 3 ++ .../ethernet/mellanox/mlx5/core/en_common.c | 41 +++++++------------ 3 files changed, 43 insertions(+), 30 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tir.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tir.c index 19499072f67f..0b55e77f19c8 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tir.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tir.c @@ -146,6 +146,31 @@ void mlx5e_tir_builder_build_direct(struct mlx5e_tir_builder *builder) MLX5_SET(tirc, tirc, rx_hash_fn, MLX5_RX_HASH_FN_INVERTED_XOR8); } +static void mlx5e_tir_context_self_lb_block(void *tirc, bool enable_uc_lb, + bool enable_mc_lb) +{ + u8 lb_flags = 0; + + if (enable_uc_lb) + lb_flags = MLX5_TIRC_SELF_LB_BLOCK_BLOCK_UNICAST; + if (enable_mc_lb) + lb_flags |= MLX5_TIRC_SELF_LB_BLOCK_BLOCK_MULTICAST; + + MLX5_SET(tirc, tirc, self_lb_block, lb_flags); +} + +void mlx5e_tir_builder_build_self_lb_block(struct mlx5e_tir_builder *builder, + bool enable_uc_lb, + bool enable_mc_lb) +{ + void *tirc = mlx5e_tir_builder_get_tirc(builder); + + if (builder->modify) + MLX5_SET(modify_tir_in, builder->in, bitmask.self_lb_en, 1); + + mlx5e_tir_context_self_lb_block(tirc, enable_uc_lb, enable_mc_lb); +} + void mlx5e_tir_builder_build_tls(struct mlx5e_tir_builder *builder) { void *tirc = mlx5e_tir_builder_get_tirc(builder); @@ -153,9 +178,7 @@ void mlx5e_tir_builder_build_tls(struct mlx5e_tir_builder *builder) WARN_ON(builder->modify); MLX5_SET(tirc, tirc, tls_en, 1); - MLX5_SET(tirc, tirc, self_lb_block, - MLX5_TIRC_SELF_LB_BLOCK_BLOCK_UNICAST | - MLX5_TIRC_SELF_LB_BLOCK_BLOCK_MULTICAST); + mlx5e_tir_context_self_lb_block(tirc, true, true); } int mlx5e_tir_init(struct mlx5e_tir *tir, struct mlx5e_tir_builder *builder, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tir.h b/drivers/net/ethernet/mellanox/mlx5/core/en/tir.h index e8df3aaf6562..958eeb959a19 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tir.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tir.h @@ -35,6 +35,9 @@ void mlx5e_tir_builder_build_rss(struct mlx5e_tir_builder *builder, const struct mlx5e_rss_params_traffic_type *rss_tt, bool inner); void mlx5e_tir_builder_build_direct(struct mlx5e_tir_builder *builder); +void mlx5e_tir_builder_build_self_lb_block(struct mlx5e_tir_builder *builder, + bool enable_uc_lb, + bool enable_mc_lb); void mlx5e_tir_builder_build_tls(struct mlx5e_tir_builder *builder); struct mlx5_core_dev; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_common.c b/drivers/net/ethernet/mellanox/mlx5/core/en_common.c index 376a018b2db1..022a0cf7063c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_common.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_common.c @@ -250,44 +250,31 @@ void mlx5e_destroy_mdev_resources(struct mlx5_core_dev *mdev) int mlx5e_modify_tirs_lb(struct mlx5_core_dev *mdev, bool enable_uc_lb, bool enable_mc_lb) { + struct mlx5e_tir_builder *builder; struct mlx5e_tir *tir; - u8 lb_flags = 0; - int err = 0; - u32 tirn = 0; - int inlen; - void *in; + int err = 0; - inlen = MLX5_ST_SZ_BYTES(modify_tir_in); - in = kvzalloc(inlen, GFP_KERNEL); - if (!in) + builder = mlx5e_tir_builder_alloc(true); + if (!builder) return -ENOMEM; - if (enable_uc_lb) - lb_flags = MLX5_TIRC_SELF_LB_BLOCK_BLOCK_UNICAST; - - if (enable_mc_lb) - lb_flags |= MLX5_TIRC_SELF_LB_BLOCK_BLOCK_MULTICAST; - - if (lb_flags) - MLX5_SET(modify_tir_in, in, ctx.self_lb_block, lb_flags); - - MLX5_SET(modify_tir_in, in, bitmask.self_lb_en, 1); + mlx5e_tir_builder_build_self_lb_block(builder, enable_uc_lb, + enable_mc_lb); mutex_lock(&mdev->mlx5e_res.hw_objs.td.list_lock); list_for_each_entry(tir, &mdev->mlx5e_res.hw_objs.td.tirs_list, list) { - tirn = tir->tirn; - err = mlx5_core_modify_tir(mdev, tirn, in); - if (err) + err = mlx5e_tir_modify(tir, builder); + if (err) { + mlx5_core_err(mdev, + "modify tir(0x%x) enable_lb uc(%d) mc(%d) failed, %d\n", + mlx5e_tir_get_tirn(tir), + enable_uc_lb, enable_mc_lb, err); break; + } } mutex_unlock(&mdev->mlx5e_res.hw_objs.td.list_lock); - kvfree(in); - if (err) - mlx5_core_err(mdev, - "modify tir(0x%x) enable_lb uc(%d) mc(%d) failed, %d\n", - tirn, - enable_uc_lb, enable_mc_lb, err); + mlx5e_tir_builder_free(builder); return err; } From 99b002018f6a3dc08c789e2962070d6de7cb3bac Mon Sep 17 00:00:00 2001 From: Tariq Toukan Date: Thu, 30 Oct 2025 15:32:35 +0200 Subject: [PATCH 0499/1381] net/mlx5e: Allow setting self loopback prevention bits on TIR init Until now, IPoIB was creating TIRs without setting self loopback prevention, then modifying them in activation stage. This is a preparation patch, that will be used by IPoIB to init TIRs properly without the need for following calls of modify_tir. Signed-off-by: Tariq Toukan Reviewed-by: Carolina Jubran Reviewed-by: Simon Horman Link: https://patch.msgid.link/1761831159-1013140-4-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mellanox/mlx5/core/en/rss.c | 2 ++ drivers/net/ethernet/mellanox/mlx5/core/en/rss.h | 1 + drivers/net/ethernet/mellanox/mlx5/core/en/rx_res.c | 7 +++++++ drivers/net/ethernet/mellanox/mlx5/core/en/rx_res.h | 1 + 4 files changed, 11 insertions(+) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/rss.c b/drivers/net/ethernet/mellanox/mlx5/core/en/rss.c index c96cbc4b0dbf..88b0e1050d1a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/rss.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/rss.c @@ -231,6 +231,8 @@ mlx5e_rss_create_tir(struct mlx5e_rss *rss, enum mlx5_traffic_types tt, rqtn, rss_inner); mlx5e_tir_builder_build_packet_merge(builder, pkt_merge_param); rss_tt = mlx5e_rss_get_tt_config(rss, tt); + mlx5e_tir_builder_build_self_lb_block(builder, rss->params.self_lb_blk, + rss->params.self_lb_blk); mlx5e_tir_builder_build_rss(builder, &rss->hash, &rss_tt, inner); err = mlx5e_tir_init(tir, builder, rss->mdev, true); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/rss.h b/drivers/net/ethernet/mellanox/mlx5/core/en/rss.h index 5fb03cd0a411..17664757a561 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/rss.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/rss.h @@ -23,6 +23,7 @@ struct mlx5e_rss_init_params { struct mlx5e_rss_params { bool inner_ft_support; u32 drop_rqn; + bool self_lb_blk; }; struct mlx5e_rss_params_traffic_type diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/rx_res.c b/drivers/net/ethernet/mellanox/mlx5/core/en/rx_res.c index ac26a32845d0..55c117b7d8c4 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/rx_res.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/rx_res.c @@ -71,6 +71,8 @@ static int mlx5e_rx_res_rss_init_def(struct mlx5e_rx_res *res, rss_params = (struct mlx5e_rss_params) { .inner_ft_support = inner_ft_support, .drop_rqn = res->drop_rqn, + .self_lb_blk = + res->features & MLX5E_RX_RES_FEATURE_SELF_LB_BLOCK, }; rss = mlx5e_rss_init(res->mdev, &rss_params, &init_params); @@ -104,6 +106,8 @@ int mlx5e_rx_res_rss_init(struct mlx5e_rx_res *res, u32 rss_idx, unsigned int in rss_params = (struct mlx5e_rss_params) { .inner_ft_support = inner_ft_support, .drop_rqn = res->drop_rqn, + .self_lb_blk = + res->features & MLX5E_RX_RES_FEATURE_SELF_LB_BLOCK, }; rss = mlx5e_rss_init(res->mdev, &rss_params, &init_params); @@ -346,6 +350,7 @@ static struct mlx5e_rx_res *mlx5e_rx_res_alloc(struct mlx5_core_dev *mdev, unsig static int mlx5e_rx_res_channels_init(struct mlx5e_rx_res *res) { bool inner_ft_support = res->features & MLX5E_RX_RES_FEATURE_INNER_FT; + bool self_lb_blk = res->features & MLX5E_RX_RES_FEATURE_SELF_LB_BLOCK; struct mlx5e_tir_builder *builder; int err = 0; int ix; @@ -376,6 +381,8 @@ static int mlx5e_rx_res_channels_init(struct mlx5e_rx_res *res) mlx5e_rqt_get_rqtn(&res->channels[ix].direct_rqt), inner_ft_support); mlx5e_tir_builder_build_packet_merge(builder, &res->pkt_merge_param); + mlx5e_tir_builder_build_self_lb_block(builder, self_lb_blk, + self_lb_blk); mlx5e_tir_builder_build_direct(builder); err = mlx5e_tir_init(&res->channels[ix].direct_tir, builder, res->mdev, true); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/rx_res.h b/drivers/net/ethernet/mellanox/mlx5/core/en/rx_res.h index 65a857c215e1..675780120a20 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/rx_res.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/rx_res.h @@ -21,6 +21,7 @@ enum mlx5e_rx_res_features { MLX5E_RX_RES_FEATURE_INNER_FT = BIT(0), MLX5E_RX_RES_FEATURE_PTP = BIT(1), MLX5E_RX_RES_FEATURE_MULTI_VHCA = BIT(2), + MLX5E_RX_RES_FEATURE_SELF_LB_BLOCK = BIT(3), }; /* Setup */ From a4c81e72f132b93a3b920196621a7b78c71fb7fc Mon Sep 17 00:00:00 2001 From: Tariq Toukan Date: Thu, 30 Oct 2025 15:32:36 +0200 Subject: [PATCH 0500/1381] net/mlx5: IPoIB, set self loopback prevention in TIR init In IPoIB, the self loopback prevention configuration apply in activation stage has two roles: fulfill a firmware requirement for old firmware (tis_tir_td_order=0), and update the proper configuration as it was not set in init. Here we set the proper configuration in init, to allow skipping the modify_tirs commands on new firmware in a downstream patch. Signed-off-by: Tariq Toukan Reviewed-by: Carolina Jubran Reviewed-by: Simon Horman Link: https://patch.msgid.link/1761831159-1013140-5-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c index 976347ac1faf..0a6003fe60e9 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c @@ -427,6 +427,7 @@ static void mlx5i_destroy_flow_steering(struct mlx5e_priv *priv) static int mlx5i_init_rx(struct mlx5e_priv *priv) { struct mlx5_core_dev *mdev = priv->mdev; + enum mlx5e_rx_res_features features; int err; priv->fs = mlx5e_fs_init(priv->profile, mdev, @@ -445,7 +446,9 @@ static int mlx5i_init_rx(struct mlx5e_priv *priv) goto err_destroy_q_counters; } - priv->rx_res = mlx5e_rx_res_create(priv->mdev, 0, priv->max_nch, priv->drop_rq.rqn, + features = MLX5E_RX_RES_FEATURE_SELF_LB_BLOCK; + priv->rx_res = mlx5e_rx_res_create(priv->mdev, features, priv->max_nch, + priv->drop_rq.rqn, &priv->channels.params.packet_merge, priv->channels.params.num_channels); if (IS_ERR(priv->rx_res)) { From 477c352adda4ba0bd80c945ab13165161802239e Mon Sep 17 00:00:00 2001 From: Tariq Toukan Date: Thu, 30 Oct 2025 15:32:37 +0200 Subject: [PATCH 0501/1381] net/mlx5e: Do not re-apply TIR loopback configuration if not necessary On old firmware, (tis_tir_td_order=0), TIR of a transport domain should either be created after all SQs of the same domain, or TIR.self_lb_en should be reapplied using MODIFY_TIR, for self loopback filtering to function correctly. This is not necessary anymnore on new FW (tis_tir_td_order=1), thus there's no need for calling modify_tir operations after creating a new set of SQs to maintain the self loopback prevention functional. Skip these operations. This saves O(max_num_channels) MODIFY_TIR firmware commands in operations like interface up or channels configuration change. Signed-off-by: Tariq Toukan Reviewed-by: Carolina Jubran Reviewed-by: Simon Horman Link: https://patch.msgid.link/1761831159-1013140-6-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mellanox/mlx5/core/en_common.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_common.c b/drivers/net/ethernet/mellanox/mlx5/core/en_common.c index 022a0cf7063c..5a2ac7b6f260 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_common.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_common.c @@ -282,5 +282,8 @@ int mlx5e_modify_tirs_lb(struct mlx5_core_dev *mdev, bool enable_uc_lb, int mlx5e_refresh_tirs(struct mlx5_core_dev *mdev, bool enable_uc_lb, bool enable_mc_lb) { + if (MLX5_CAP_GEN(mdev, tis_tir_td_order)) + return 0; /* refresh not needed */ + return mlx5e_modify_tirs_lb(mdev, enable_uc_lb, enable_mc_lb); } From 911e3a37b024163d8329e3560d6fd5f0f0da2558 Mon Sep 17 00:00:00 2001 From: Tariq Toukan Date: Thu, 30 Oct 2025 15:32:38 +0200 Subject: [PATCH 0502/1381] net/mlx5e: Pass old channels as argument to mlx5e_switch_priv_channels Let the caller function mlx5e_safe_switch_params() maintain a copy of the old channels, and pass it to mlx5e_switch_priv_channels(). This is in preparation for the next patch. Signed-off-by: Tariq Toukan Reviewed-by: Carolina Jubran Reviewed-by: Simon Horman Link: https://patch.msgid.link/1761831159-1013140-7-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- .../net/ethernet/mellanox/mlx5/core/en_main.c | 25 +++++++++++++------ 1 file changed, 17 insertions(+), 8 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index f0d7a61e014b..4adc5a429745 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -3356,12 +3356,12 @@ static int mlx5e_switch_priv_params(struct mlx5e_priv *priv, } static int mlx5e_switch_priv_channels(struct mlx5e_priv *priv, + struct mlx5e_channels *old_chs, struct mlx5e_channels *new_chs, mlx5e_fp_preactivate preactivate, void *context) { struct net_device *netdev = priv->netdev; - struct mlx5e_channels old_chs; int carrier_ok; int err = 0; @@ -3370,7 +3370,6 @@ static int mlx5e_switch_priv_channels(struct mlx5e_priv *priv, mlx5e_deactivate_priv_channels(priv); - old_chs = priv->channels; priv->channels = *new_chs; /* New channels are ready to roll, call the preactivate hook if needed @@ -3379,12 +3378,12 @@ static int mlx5e_switch_priv_channels(struct mlx5e_priv *priv, if (preactivate) { err = preactivate(priv, context); if (err) { - priv->channels = old_chs; + priv->channels = *old_chs; goto out; } } - mlx5e_close_channels(&old_chs); + mlx5e_close_channels(old_chs); priv->profile->update_rx(priv); mlx5e_selq_apply(&priv->selq); @@ -3403,16 +3402,20 @@ int mlx5e_safe_switch_params(struct mlx5e_priv *priv, mlx5e_fp_preactivate preactivate, void *context, bool reset) { - struct mlx5e_channels *new_chs; + struct mlx5e_channels *old_chs, *new_chs; int err; reset &= test_bit(MLX5E_STATE_OPENED, &priv->state); if (!reset) return mlx5e_switch_priv_params(priv, params, preactivate, context); + old_chs = kzalloc(sizeof(*old_chs), GFP_KERNEL); new_chs = kzalloc(sizeof(*new_chs), GFP_KERNEL); - if (!new_chs) - return -ENOMEM; + if (!old_chs || !new_chs) { + err = -ENOMEM; + goto err_free_chs; + } + new_chs->params = *params; mlx5e_selq_prepare_params(&priv->selq, &new_chs->params); @@ -3421,11 +3424,15 @@ int mlx5e_safe_switch_params(struct mlx5e_priv *priv, if (err) goto err_cancel_selq; - err = mlx5e_switch_priv_channels(priv, new_chs, preactivate, context); + *old_chs = priv->channels; + + err = mlx5e_switch_priv_channels(priv, old_chs, new_chs, + preactivate, context); if (err) goto err_close; kfree(new_chs); + kfree(old_chs); return 0; err_close: @@ -3433,7 +3440,9 @@ int mlx5e_safe_switch_params(struct mlx5e_priv *priv, err_cancel_selq: mlx5e_selq_cancel(&priv->selq); +err_free_chs: kfree(new_chs); + kfree(old_chs); return err; } From 3b88a535a8e10d83335f04c60aafbdfd37146a01 Mon Sep 17 00:00:00 2001 From: Tariq Toukan Date: Thu, 30 Oct 2025 15:32:39 +0200 Subject: [PATCH 0503/1381] net/mlx5e: Defer channels closure to reduce interface down time Cap bit tis_tir_td_order=1 indicates that an old firmware requirement / limitation no longer exists. When unset, the latency of several firmware commands significantly increases with the presence of high number of co-existing channels (both old and new sets). Hence, we used to close unneeded old channels before invoking those firmware commands. Today, on capable devices, this is no longer the case. Minimize the interface down time by deferring the old channels closure, after the activation of the new ones. Perf numbers: Measured the number of dropped packets in a simple ping flood test, during a configuration change operation, that switches the number of channels from 247 to 248. Before: 71 packets lost After: 15 packets lost, ~80% saving. Signed-off-by: Tariq Toukan Reviewed-by: Carolina Jubran Reviewed-by: Simon Horman Link: https://patch.msgid.link/1761831159-1013140-8-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index 4adc5a429745..7b7a0060979d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -3383,7 +3383,8 @@ static int mlx5e_switch_priv_channels(struct mlx5e_priv *priv, } } - mlx5e_close_channels(old_chs); + if (!MLX5_CAP_GEN(priv->mdev, tis_tir_td_order)) + mlx5e_close_channels(old_chs); priv->profile->update_rx(priv); mlx5e_selq_apply(&priv->selq); @@ -3431,6 +3432,9 @@ int mlx5e_safe_switch_params(struct mlx5e_priv *priv, if (err) goto err_close; + if (MLX5_CAP_GEN(priv->mdev, tis_tir_td_order)) + mlx5e_close_channels(old_chs); + kfree(new_chs); kfree(old_chs); return 0; From f88191c7f3618405f1fc5c331a94ebfe601c5b08 Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" Date: Sat, 1 Nov 2025 18:56:51 +0100 Subject: [PATCH 0504/1381] mptcp: pm: in-kernel: record fullmesh endp nb Instead of iterating over all endpoints, under RCU read lock, just to check if one of them as the fullmesh flag, we can keep a counter of fullmesh endpoint, similar to what is done with the other flags. This counter is now checked, before iterating over all endpoints. Similar to the other counters, this new one is also exposed. A userspace app can then know when it is being used in a fullmesh mode, with potentially (too) many subflows. Reviewed-by: Geliang Tang Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20251101-net-next-mptcp-fm-endp-nb-bind-v1-1-b4166772d6bb@kernel.org Signed-off-by: Jakub Kicinski --- include/uapi/linux/mptcp.h | 3 ++- net/mptcp/pm_kernel.c | 38 +++++++++++++++++++++++++++++++++++--- net/mptcp/protocol.h | 1 + net/mptcp/sockopt.c | 2 ++ 4 files changed, 40 insertions(+), 4 deletions(-) diff --git a/include/uapi/linux/mptcp.h b/include/uapi/linux/mptcp.h index 87cfab874e24..04eea6d1d0a9 100644 --- a/include/uapi/linux/mptcp.h +++ b/include/uapi/linux/mptcp.h @@ -70,7 +70,8 @@ struct mptcp_info { __u64 mptcpi_bytes_acked; __u8 mptcpi_subflows_total; __u8 mptcpi_endp_laminar_max; - __u8 reserved[2]; + __u8 mptcpi_endp_fullmesh_max; + __u8 reserved; __u32 mptcpi_last_data_sent; __u32 mptcpi_last_data_recv; __u32 mptcpi_last_ack_recv; diff --git a/net/mptcp/pm_kernel.c b/net/mptcp/pm_kernel.c index 2ae95476dba3..e2918c68ff02 100644 --- a/net/mptcp/pm_kernel.c +++ b/net/mptcp/pm_kernel.c @@ -22,6 +22,7 @@ struct pm_nl_pernet { u8 endp_signal_max; u8 endp_subflow_max; u8 endp_laminar_max; + u8 endp_fullmesh_max; u8 limit_add_addr_accepted; u8 limit_extra_subflows; u8 next_id; @@ -70,6 +71,14 @@ u8 mptcp_pm_get_endp_laminar_max(const struct mptcp_sock *msk) } EXPORT_SYMBOL_GPL(mptcp_pm_get_endp_laminar_max); +u8 mptcp_pm_get_endp_fullmesh_max(const struct mptcp_sock *msk) +{ + struct pm_nl_pernet *pernet = pm_nl_get_pernet_from_msk(msk); + + return READ_ONCE(pernet->endp_fullmesh_max); +} +EXPORT_SYMBOL_GPL(mptcp_pm_get_endp_fullmesh_max); + u8 mptcp_pm_get_limit_add_addr_accepted(const struct mptcp_sock *msk) { struct pm_nl_pernet *pernet = pm_nl_get_pernet_from_msk(msk); @@ -603,9 +612,12 @@ fill_local_addresses_vec(struct mptcp_sock *msk, struct mptcp_addr_info *remote, int i; /* If there is at least one MPTCP endpoint with a fullmesh flag */ - i = fill_local_addresses_vec_fullmesh(msk, remote, locals, c_flag_case); - if (i) - return i; + if (mptcp_pm_get_endp_fullmesh_max(msk)) { + i = fill_local_addresses_vec_fullmesh(msk, remote, locals, + c_flag_case); + if (i) + return i; + } /* If there is at least one MPTCP endpoint with a laminar flag */ if (mptcp_pm_get_endp_laminar_max(msk)) @@ -790,6 +802,10 @@ static int mptcp_pm_nl_append_new_local_addr(struct pm_nl_pernet *pernet, addr_max = pernet->endp_laminar_max; WRITE_ONCE(pernet->endp_laminar_max, addr_max + 1); } + if (entry->flags & MPTCP_PM_ADDR_FLAG_FULLMESH) { + addr_max = pernet->endp_fullmesh_max; + WRITE_ONCE(pernet->endp_fullmesh_max, addr_max + 1); + } pernet->endpoints++; if (!entry->addr.port) @@ -1187,6 +1203,10 @@ int mptcp_pm_nl_del_addr_doit(struct sk_buff *skb, struct genl_info *info) addr_max = pernet->endp_laminar_max; WRITE_ONCE(pernet->endp_laminar_max, addr_max - 1); } + if (entry->flags & MPTCP_PM_ADDR_FLAG_FULLMESH) { + addr_max = pernet->endp_fullmesh_max; + WRITE_ONCE(pernet->endp_fullmesh_max, addr_max - 1); + } pernet->endpoints--; list_del_rcu(&entry->list); @@ -1502,6 +1522,18 @@ int mptcp_pm_nl_set_flags(struct mptcp_pm_addr_entry *local, changed = (local->flags ^ entry->flags) & mask; entry->flags = (entry->flags & ~mask) | (local->flags & mask); *local = *entry; + + if (changed & MPTCP_PM_ADDR_FLAG_FULLMESH) { + u8 addr_max = pernet->endp_fullmesh_max; + + if (entry->flags & MPTCP_PM_ADDR_FLAG_FULLMESH) + addr_max++; + else + addr_max--; + + WRITE_ONCE(pernet->endp_fullmesh_max, addr_max); + } + spin_unlock_bh(&pernet->lock); mptcp_pm_nl_set_flags_all(net, local, changed); diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h index 379a88e14e8d..9a3429175758 100644 --- a/net/mptcp/protocol.h +++ b/net/mptcp/protocol.h @@ -1183,6 +1183,7 @@ void __mptcp_pm_kernel_worker(struct mptcp_sock *msk); u8 mptcp_pm_get_endp_signal_max(const struct mptcp_sock *msk); u8 mptcp_pm_get_endp_subflow_max(const struct mptcp_sock *msk); u8 mptcp_pm_get_endp_laminar_max(const struct mptcp_sock *msk); +u8 mptcp_pm_get_endp_fullmesh_max(const struct mptcp_sock *msk); u8 mptcp_pm_get_limit_add_addr_accepted(const struct mptcp_sock *msk); u8 mptcp_pm_get_limit_extra_subflows(const struct mptcp_sock *msk); diff --git a/net/mptcp/sockopt.c b/net/mptcp/sockopt.c index a28a48385885..de90a2897d2d 100644 --- a/net/mptcp/sockopt.c +++ b/net/mptcp/sockopt.c @@ -982,6 +982,8 @@ void mptcp_diag_fill_info(struct mptcp_sock *msk, struct mptcp_info *info) mptcp_pm_get_endp_subflow_max(msk); info->mptcpi_endp_laminar_max = mptcp_pm_get_endp_laminar_max(msk); + info->mptcpi_endp_fullmesh_max = + mptcp_pm_get_endp_fullmesh_max(msk); } if (__mptcp_check_fallback(msk)) From e461e8a799a2984e9b55f40c65d123a114496dff Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" Date: Sat, 1 Nov 2025 18:56:52 +0100 Subject: [PATCH 0505/1381] mptcp: pm: in kernel: only use fullmesh endp if any Our documentation is saying that the in-kernel PM is only using fullmesh endpoints to establish subflows to announced addresses when at least one endpoint has a fullmesh flag. But this was not totally correct: only fullmesh endpoints were used if at least one endpoint *from the same address family as the received ADD_ADDR* has the fullmesh flag. This is confusing, and it seems clearer not to have differences depending on the address family. So, now, when at least one MPTCP endpoint has a fullmesh flag, the local addresses are picked from all fullmesh endpoints, which might be 0 if there are no endpoints for the correct address family. One selftest needs to be adapted for this behaviour change. Reviewed-by: Geliang Tang Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20251101-net-next-mptcp-fm-endp-nb-bind-v1-2-b4166772d6bb@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/pm_kernel.c | 10 +++------- tools/testing/selftests/net/mptcp/mptcp_join.sh | 6 +++++- 2 files changed, 8 insertions(+), 8 deletions(-) diff --git a/net/mptcp/pm_kernel.c b/net/mptcp/pm_kernel.c index e2918c68ff02..e50721c670d0 100644 --- a/net/mptcp/pm_kernel.c +++ b/net/mptcp/pm_kernel.c @@ -609,15 +609,11 @@ fill_local_addresses_vec(struct mptcp_sock *msk, struct mptcp_addr_info *remote, struct mptcp_pm_local *locals) { bool c_flag_case = remote->id && mptcp_pm_add_addr_c_flag_case(msk); - int i; /* If there is at least one MPTCP endpoint with a fullmesh flag */ - if (mptcp_pm_get_endp_fullmesh_max(msk)) { - i = fill_local_addresses_vec_fullmesh(msk, remote, locals, - c_flag_case); - if (i) - return i; - } + if (mptcp_pm_get_endp_fullmesh_max(msk)) + return fill_local_addresses_vec_fullmesh(msk, remote, locals, + c_flag_case); /* If there is at least one MPTCP endpoint with a laminar flag */ if (mptcp_pm_get_endp_laminar_max(msk)) diff --git a/tools/testing/selftests/net/mptcp/mptcp_join.sh b/tools/testing/selftests/net/mptcp/mptcp_join.sh index 78a1aa4ecff2..e7a498dd5a46 100755 --- a/tools/testing/selftests/net/mptcp/mptcp_join.sh +++ b/tools/testing/selftests/net/mptcp/mptcp_join.sh @@ -2952,7 +2952,11 @@ mixed_tests() pm_nl_add_endpoint $ns1 10.0.1.1 flags signal speed=slow \ run_tests $ns1 $ns2 dead:beef:2::1 - chk_join_nr 1 1 1 + if mptcp_lib_kallsyms_has "mptcp_pm_get_endp_fullmesh_max$"; then + chk_join_nr 0 0 0 + else + chk_join_nr 1 1 1 + fi fi # fullmesh still tries to create all the possibly subflows with From 4a6220a453c8afd46a01f0bbe56bc6734adb77b7 Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" Date: Sat, 1 Nov 2025 18:56:53 +0100 Subject: [PATCH 0506/1381] selftests: mptcp: join: do_transfer: reduce code dup The same extra long commands are present twice, with small differences: the variable for the stdin file is different. Use new dedicated variables in one command to avoid this code duplication. Reviewed-by: Geliang Tang Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20251101-net-next-mptcp-fm-endp-nb-bind-v1-3-b4166772d6bb@kernel.org Signed-off-by: Jakub Kicinski --- .../testing/selftests/net/mptcp/mptcp_join.sh | 30 +++++++------------ 1 file changed, 11 insertions(+), 19 deletions(-) diff --git a/tools/testing/selftests/net/mptcp/mptcp_join.sh b/tools/testing/selftests/net/mptcp/mptcp_join.sh index e7a498dd5a46..4c9ee094381e 100755 --- a/tools/testing/selftests/net/mptcp/mptcp_join.sh +++ b/tools/testing/selftests/net/mptcp/mptcp_join.sh @@ -951,6 +951,8 @@ do_transfer() local FAILING_LINKS=${FAILING_LINKS:-""} local fastclose=${fastclose:-""} local speed=${speed:-"fast"} + local listener_in="${sin}" + local connector_in="${cin}" port=$(get_port) :> "$cout" @@ -999,16 +1001,12 @@ do_transfer() extra_srv_args="$extra_args $extra_srv_args" if [ "$test_linkfail" -gt 1 ];then - timeout ${timeout_test} \ - ip netns exec ${listener_ns} \ - ./mptcp_connect -t ${timeout_poll} -l -p $port -s ${srv_proto} \ - $extra_srv_args "::" < "$sinfail" > "$sout" & - else - timeout ${timeout_test} \ - ip netns exec ${listener_ns} \ - ./mptcp_connect -t ${timeout_poll} -l -p $port -s ${srv_proto} \ - $extra_srv_args "::" < "$sin" > "$sout" & + listener_in="${sinfail}" fi + timeout ${timeout_test} \ + ip netns exec ${listener_ns} \ + ./mptcp_connect -t ${timeout_poll} -l -p ${port} -s ${srv_proto} \ + ${extra_srv_args} "::" < "${listener_in}" > "${sout}" & local spid=$! mptcp_lib_wait_local_port_listen "${listener_ns}" "${port}" @@ -1020,6 +1018,7 @@ do_transfer() ./mptcp_connect -t ${timeout_poll} -p $port -s ${cl_proto} \ $extra_cl_args $connect_addr < "$cin" > "$cout" & elif [ "$test_linkfail" -eq 1 ] || [ "$test_linkfail" -eq 2 ];then + connector_in="${cinsent}" ( cat "$cinfail" ; sleep 2; link_failure $listener_ns ; cat "$cinfail" ) | \ tee "$cinsent" | \ timeout ${timeout_test} \ @@ -1027,6 +1026,7 @@ do_transfer() ./mptcp_connect -t ${timeout_poll} -p $port -s ${cl_proto} \ $extra_cl_args $connect_addr > "$cout" & else + connector_in="${cinsent}" tee "$cinsent" < "$cinfail" | \ timeout ${timeout_test} \ ip netns exec ${connector_ns} \ @@ -1057,17 +1057,9 @@ do_transfer() return 1 fi - if [ "$test_linkfail" -gt 1 ];then - check_transfer $sinfail $cout "file received by client" $trunc_size - else - check_transfer $sin $cout "file received by client" $trunc_size - fi + check_transfer $listener_in $cout "file received by client" $trunc_size retc=$? - if [ "$test_linkfail" -eq 0 ];then - check_transfer $cin $sout "file received by server" $trunc_size - else - check_transfer $cinsent $sout "file received by server" $trunc_size - fi + check_transfer $connector_in $sout "file received by server" $trunc_size rets=$? [ $retc -eq 0 ] && [ $rets -eq 0 ] From 5c59df126bae1f2bb6fce6c11eef0e9776b32598 Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" Date: Sat, 1 Nov 2025 18:56:54 +0100 Subject: [PATCH 0507/1381] selftests: mptcp: join: validate extra bind cases By design, an MPTCP connection will not accept extra subflows where no MPTCP listening sockets can accept such requests. In other words, it means that if the 'server' listens on a specific address / device, it cannot accept MP_JOIN sent to a different address / device. Except if there is another MPTCP listening socket accepting them. This is what the new tests are validating: - Forcing a bind on the main v4/v6 address, and checking that MP_JOIN to announced addresses are not accepted. - Also forcing a bind on the main v4/v6 address, but before, another listening socket is created to accept additional subflows. Note that 'mptcpize run nc -l' -- or something else only doing: socket(MPTCP), bind(), listen(0) -- would be enough, but here mptcp_connect is reused not to depend on another tool just for that. - Same as the previous one, but using v6 link-local addresses: this is a bit particular because it is required to specify the outgoing network interface when connecting to a link-local address announced by the other peer. When using the routing rules, this doesn't work (the outgoing interface is not known) ; but it does work with a 'laminar' endpoint having a specified interface. Note that extra small modifications are needed for these tests to work: - mptcp_connect's check_getpeername_connect() check should strip the specified interface when comparing addresses. - With IPv6 link-local addresses, it is required to wait for them to be ready (no longer in 'tentative' mode) before using them, otherwise the bind() will not be allowed. Link: https://github.com/multipath-tcp/mptcp_net-next/issues/591 Reviewed-by: Geliang Tang Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20251101-net-next-mptcp-fm-endp-nb-bind-v1-4-b4166772d6bb@kernel.org Signed-off-by: Jakub Kicinski --- .../selftests/net/mptcp/mptcp_connect.c | 10 +- .../testing/selftests/net/mptcp/mptcp_join.sh | 153 +++++++++++++++++- 2 files changed, 161 insertions(+), 2 deletions(-) diff --git a/tools/testing/selftests/net/mptcp/mptcp_connect.c b/tools/testing/selftests/net/mptcp/mptcp_connect.c index b148cadb96d0..c030b08a7195 100644 --- a/tools/testing/selftests/net/mptcp/mptcp_connect.c +++ b/tools/testing/selftests/net/mptcp/mptcp_connect.c @@ -1064,6 +1064,8 @@ static void check_getpeername_connect(int fd) socklen_t salen = sizeof(ss); char a[INET6_ADDRSTRLEN]; char b[INET6_ADDRSTRLEN]; + const char *iface; + size_t len; if (getpeername(fd, (struct sockaddr *)&ss, &salen) < 0) { perror("getpeername"); @@ -1073,7 +1075,13 @@ static void check_getpeername_connect(int fd) xgetnameinfo((struct sockaddr *)&ss, salen, a, sizeof(a), b, sizeof(b)); - if (strcmp(cfg_host, a) || strcmp(cfg_port, b)) + iface = strchr(cfg_host, '%'); + if (iface) + len = iface - cfg_host; + else + len = strlen(cfg_host) + 1; + + if (strncmp(cfg_host, a, len) || strcmp(cfg_port, b)) fprintf(stderr, "%s: %s vs %s, %s vs %s\n", __func__, cfg_host, a, cfg_port, b); } diff --git a/tools/testing/selftests/net/mptcp/mptcp_join.sh b/tools/testing/selftests/net/mptcp/mptcp_join.sh index 4c9ee094381e..4faf58fecc94 100755 --- a/tools/testing/selftests/net/mptcp/mptcp_join.sh +++ b/tools/testing/selftests/net/mptcp/mptcp_join.sh @@ -62,6 +62,7 @@ unset sflags unset fastclose unset fullmesh unset speed +unset bind_addr unset join_syn_rej unset join_csum_ns1 unset join_csum_ns2 @@ -645,6 +646,27 @@ wait_mpj() done } +wait_ll_ready() +{ + local ns="${1}" + + local i + for i in $(seq 50); do + ip -n "${ns}" -6 addr show scope link | grep "inet6 fe80" | + grep -qw "tentative" || break + sleep 0.1 + done +} + +get_ll_addr() +{ + local ns="${1}" + local iface="${2}" + + ip -n "${ns}" -6 addr show dev "${iface}" scope link | + grep "inet6 fe80" | sed 's#.*\(fe80::.*\)/.*#\1#' +} + kill_events_pids() { mptcp_lib_kill_wait $evts_ns1_pid @@ -951,6 +973,7 @@ do_transfer() local FAILING_LINKS=${FAILING_LINKS:-""} local fastclose=${fastclose:-""} local speed=${speed:-"fast"} + local bind_addr=${bind_addr:-"::"} local listener_in="${sin}" local connector_in="${cin}" port=$(get_port) @@ -1006,7 +1029,7 @@ do_transfer() timeout ${timeout_test} \ ip netns exec ${listener_ns} \ ./mptcp_connect -t ${timeout_poll} -l -p ${port} -s ${srv_proto} \ - ${extra_srv_args} "::" < "${listener_in}" > "${sout}" & + ${extra_srv_args} "${bind_addr}" < "${listener_in}" > "${sout}" & local spid=$! mptcp_lib_wait_local_port_listen "${listener_ns}" "${port}" @@ -3229,6 +3252,133 @@ add_addr_ports_tests() fi } +bind_tests() +{ + # bind to one address should not allow extra subflows to other addresses + if reset "bind main address v4, no join v4"; then + pm_nl_set_limits $ns1 0 2 + pm_nl_set_limits $ns2 2 2 + pm_nl_add_endpoint $ns1 10.0.2.1 flags signal + bind_addr="10.0.1.1" \ + run_tests $ns1 $ns2 10.0.1.1 + join_syn_tx=1 \ + chk_join_nr 0 0 0 + chk_add_nr 1 1 + fi + + # bind to one address should not allow extra subflows to other addresses + if reset "bind main address v6, no join v6"; then + pm_nl_set_limits $ns1 0 2 + pm_nl_set_limits $ns2 2 2 + pm_nl_add_endpoint $ns1 dead:beef:2::1 flags signal + bind_addr="dead:beef:1::1" \ + run_tests $ns1 $ns2 dead:beef:1::1 + join_syn_tx=1 \ + chk_join_nr 0 0 0 + chk_add_nr 1 1 + fi + + # multiple binds to allow extra subflows to other addresses + if reset "multiple bind to allow joins v4"; then + local extra_bind + + pm_nl_set_limits $ns1 0 2 + pm_nl_set_limits $ns2 2 2 + pm_nl_add_endpoint $ns1 10.0.2.1 flags signal + + # Launching another app listening on a different address + # Note: it could be a totally different app, e.g. nc, socat, ... + ip netns exec ${ns1} ./mptcp_connect -l -t -1 -p "$(get_port)" \ + -s MPTCP 10.0.2.1 & + extra_bind=$! + + bind_addr="10.0.1.1" \ + run_tests $ns1 $ns2 10.0.1.1 + chk_join_nr 1 1 1 + chk_add_nr 1 1 + + kill ${extra_bind} + fi + + # multiple binds to allow extra subflows to other addresses + if reset "multiple bind to allow joins v6"; then + local extra_bind + + pm_nl_set_limits $ns1 0 2 + pm_nl_set_limits $ns2 2 2 + pm_nl_add_endpoint $ns1 dead:beef:2::1 flags signal + + # Launching another app listening on a different address + # Note: it could be a totally different app, e.g. nc, socat, ... + ip netns exec ${ns1} ./mptcp_connect -l -t -1 -p "$(get_port)" \ + -s MPTCP dead:beef:2::1 & + extra_bind=$! + + bind_addr="dead:beef:1::1" \ + run_tests $ns1 $ns2 dead:beef:1::1 + chk_join_nr 1 1 1 + chk_add_nr 1 1 + + kill ${extra_bind} + fi + + # multiple binds to allow extra subflows to other addresses: v6 LL case + if reset "multiple bind to allow joins v6 link-local routing"; then + local extra_bind ns1ll1 ns1ll2 + + ns1ll1="$(get_ll_addr $ns1 ns1eth1)" + ns1ll2="$(get_ll_addr $ns1 ns1eth2)" + + pm_nl_set_limits $ns1 0 2 + pm_nl_set_limits $ns2 2 2 + pm_nl_add_endpoint $ns1 "${ns1ll2}" flags signal + + wait_ll_ready $ns1 # to be able to bind + wait_ll_ready $ns2 # also needed to bind on the client side + ip netns exec ${ns1} ./mptcp_connect -l -t -1 -p "$(get_port)" \ + -s MPTCP "${ns1ll2}%ns1eth2" & + extra_bind=$! + + bind_addr="${ns1ll1}%ns1eth1" \ + run_tests $ns1 $ns2 "${ns1ll1}%ns2eth1" + # it is not possible to connect to the announced LL addr without + # specifying the outgoing interface. + join_connect_err=1 \ + chk_join_nr 0 0 0 + chk_add_nr 1 1 + + kill ${extra_bind} + fi + + # multiple binds to allow extra subflows to v6 LL addresses: laminar + if reset "multiple bind to allow joins v6 link-local laminar" && + continue_if mptcp_lib_kallsyms_has "mptcp_pm_get_endp_laminar_max$"; then + local extra_bind ns1ll1 ns1ll2 ns2ll2 + + ns1ll1="$(get_ll_addr $ns1 ns1eth1)" + ns1ll2="$(get_ll_addr $ns1 ns1eth2)" + ns2ll2="$(get_ll_addr $ns2 ns2eth2)" + + pm_nl_set_limits $ns1 0 2 + pm_nl_set_limits $ns2 2 2 + pm_nl_add_endpoint $ns1 "${ns1ll2}" flags signal + pm_nl_add_endpoint $ns2 "${ns2ll2}" flags laminar dev ns2eth2 + + wait_ll_ready $ns1 # to be able to bind + wait_ll_ready $ns2 # also needed to bind on the client side + ip netns exec ${ns1} ./mptcp_connect -l -t -1 -p "$(get_port)" \ + -s MPTCP "${ns1ll2}%ns1eth2" & + extra_bind=$! + + bind_addr="${ns1ll1}%ns1eth1" \ + run_tests $ns1 $ns2 "${ns1ll1}%ns2eth1" + chk_join_nr 1 1 1 + chk_add_nr 1 1 + + kill ${extra_bind} + fi +} + syncookies_tests() { # single subflow, syncookies @@ -4183,6 +4333,7 @@ all_tests_sorted=( M@mixed_tests b@backup_tests p@add_addr_ports_tests + B@bind_tests k@syncookies_tests S@checksum_tests d@deny_join_id0_tests From f2143e283c6b993f4ad8b85a45aa16ac899f1abc Mon Sep 17 00:00:00 2001 From: Yue Haibing Date: Mon, 3 Nov 2025 15:20:46 +0800 Subject: [PATCH 0508/1381] net: devmem: Remove unused declaration net_devmem_bind_tx_release() Commit bd61848900bf ("net: devmem: Implement TX path") declared this but never implemented it. Signed-off-by: Yue Haibing Acked-by: Stanislav Fomichev Reviewed-by: Mina Almasry Link: https://patch.msgid.link/20251103072046.1670574-1-yuehaibing@huawei.com Signed-off-by: Jakub Kicinski --- net/core/devmem.h | 1 - 1 file changed, 1 deletion(-) diff --git a/net/core/devmem.h b/net/core/devmem.h index 101150d761af..0b43a648cd2e 100644 --- a/net/core/devmem.h +++ b/net/core/devmem.h @@ -94,7 +94,6 @@ void net_devmem_unbind_dmabuf(struct net_devmem_dmabuf_binding *binding); int net_devmem_bind_dmabuf_to_queue(struct net_device *dev, u32 rxq_idx, struct net_devmem_dmabuf_binding *binding, struct netlink_ext_ack *extack); -void net_devmem_bind_tx_release(struct sock *sk); static inline struct dmabuf_genpool_chunk_owner * net_devmem_iov_to_chunk_owner(const struct net_iov *niov) From e0c78fcad2bb04651af2ad40ed20714501931d7a Mon Sep 17 00:00:00 2001 From: Buday Csaba Date: Mon, 3 Nov 2025 09:13:42 +0100 Subject: [PATCH 0509/1381] dt-bindings: net: ethernet-phy: clarify when compatible must specify PHY ID Change PHY ID description in ethernet-phy.yaml to clarify that a PHY ID is required (may -> must) when the PHY requires special initialization sequence. Link: https://lore.kernel.org/netdev/20251026212026.GA2959311-robh@kernel.org/ Link: https://lore.kernel.org/netdev/aQIZvDt5gooZSTcp@debianbuilder/ Signed-off-by: Buday Csaba Acked-by: Conor Dooley Reviewed-by: Andrew Lunn Reviewed-by: Florian Fainelli Link: https://patch.msgid.link/64c52d1a726944a68a308355433e8ef0f82c4240.1762157515.git.buday.csaba@prolan.hu Signed-off-by: Jakub Kicinski --- .../devicetree/bindings/net/ethernet-phy.yaml | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/Documentation/devicetree/bindings/net/ethernet-phy.yaml b/Documentation/devicetree/bindings/net/ethernet-phy.yaml index 2ec2d9fda7e3..bb4c49fc5fd8 100644 --- a/Documentation/devicetree/bindings/net/ethernet-phy.yaml +++ b/Documentation/devicetree/bindings/net/ethernet-phy.yaml @@ -35,9 +35,13 @@ properties: description: PHYs that implement IEEE802.3 clause 45 - pattern: "^ethernet-phy-id[a-f0-9]{4}\\.[a-f0-9]{4}$" description: - If the PHY reports an incorrect ID (or none at all) then the - compatible list may contain an entry with the correct PHY ID - in the above form. + PHYs contain identification registers. These will be read to + identify the PHY. If the PHY reports an incorrect ID, or the + PHY requires a specific initialization sequence (like a + particular order of clocks, resets, power supplies), in + order to be able to read the ID registers, then the + compatible list must contain an entry with the correct PHY + ID in the above form. The first group of digits is the 16 bit Phy Identifier 1 register, this is the chip vendor OUI bits 3:18. The second group of digits is the Phy Identifier 2 register, From 2b38447548813ccdb8ad79385094a35d77c01e3b Mon Sep 17 00:00:00 2001 From: Vadim Fedorenko Date: Mon, 3 Nov 2025 15:09:47 +0000 Subject: [PATCH 0510/1381] net: liquidio: convert to use ndo_hwtstamp callbacks The driver implemented SIOCSHWTSTAMP ioctl command only, but there is a way to get configured status. Implement both ndo_hwtstamp_set and ndo_hwtstamp_get callbacks. Reviewed-by: Kory Maincent Signed-off-by: Vadim Fedorenko Link: https://patch.msgid.link/20251103150952.3538205-3-vadim.fedorenko@linux.dev Signed-off-by: Jakub Kicinski --- .../net/ethernet/cavium/liquidio/lio_main.c | 50 ++++++++----------- 1 file changed, 20 insertions(+), 30 deletions(-) diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c index 8e2fcec26ea1..0732440eeacd 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_main.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c @@ -2107,20 +2107,16 @@ liquidio_get_stats64(struct net_device *netdev, lstats->tx_fifo_errors; } -/** - * hwtstamp_ioctl - Handler for SIOCSHWTSTAMP ioctl - * @netdev: network device - * @ifr: interface request - */ -static int hwtstamp_ioctl(struct net_device *netdev, struct ifreq *ifr) +static int liquidio_hwtstamp_set(struct net_device *netdev, + struct kernel_hwtstamp_config *conf, + struct netlink_ext_ack *extack) { - struct hwtstamp_config conf; struct lio *lio = GET_LIO(netdev); - if (copy_from_user(&conf, ifr->ifr_data, sizeof(conf))) - return -EFAULT; + if (!lio->oct_dev->ptp_enable) + return -EOPNOTSUPP; - switch (conf.tx_type) { + switch (conf->tx_type) { case HWTSTAMP_TX_ON: case HWTSTAMP_TX_OFF: break; @@ -2128,7 +2124,7 @@ static int hwtstamp_ioctl(struct net_device *netdev, struct ifreq *ifr) return -ERANGE; } - switch (conf.rx_filter) { + switch (conf->rx_filter) { case HWTSTAMP_FILTER_NONE: break; case HWTSTAMP_FILTER_ALL: @@ -2146,39 +2142,32 @@ static int hwtstamp_ioctl(struct net_device *netdev, struct ifreq *ifr) case HWTSTAMP_FILTER_PTP_V2_SYNC: case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ: case HWTSTAMP_FILTER_NTP_ALL: - conf.rx_filter = HWTSTAMP_FILTER_ALL; + conf->rx_filter = HWTSTAMP_FILTER_ALL; break; default: return -ERANGE; } - if (conf.rx_filter == HWTSTAMP_FILTER_ALL) + if (conf->rx_filter == HWTSTAMP_FILTER_ALL) ifstate_set(lio, LIO_IFSTATE_RX_TIMESTAMP_ENABLED); else ifstate_reset(lio, LIO_IFSTATE_RX_TIMESTAMP_ENABLED); - return copy_to_user(ifr->ifr_data, &conf, sizeof(conf)) ? -EFAULT : 0; + return 0; } -/** - * liquidio_ioctl - ioctl handler - * @netdev: network device - * @ifr: interface request - * @cmd: command - */ -static int liquidio_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) +static int liquidio_hwtstamp_get(struct net_device *netdev, + struct kernel_hwtstamp_config *conf) { struct lio *lio = GET_LIO(netdev); - switch (cmd) { - case SIOCSHWTSTAMP: - if (lio->oct_dev->ptp_enable) - return hwtstamp_ioctl(netdev, ifr); - fallthrough; - default: - return -EOPNOTSUPP; - } + /* TX timestamping is technically always on */ + conf->tx_type = HWTSTAMP_TX_ON; + conf->rx_filter = ifstate_check(lio, LIO_IFSTATE_RX_TIMESTAMP_ENABLED) ? + HWTSTAMP_FILTER_ALL : HWTSTAMP_FILTER_NONE; + + return 0; } /** @@ -3227,7 +3216,6 @@ static const struct net_device_ops lionetdevops = { .ndo_vlan_rx_add_vid = liquidio_vlan_rx_add_vid, .ndo_vlan_rx_kill_vid = liquidio_vlan_rx_kill_vid, .ndo_change_mtu = liquidio_change_mtu, - .ndo_eth_ioctl = liquidio_ioctl, .ndo_fix_features = liquidio_fix_features, .ndo_set_features = liquidio_set_features, .ndo_set_vf_mac = liquidio_set_vf_mac, @@ -3238,6 +3226,8 @@ static const struct net_device_ops lionetdevops = { .ndo_set_vf_link_state = liquidio_set_vf_link_state, .ndo_get_vf_stats = liquidio_get_vf_stats, .ndo_get_port_parent_id = liquidio_get_port_parent_id, + .ndo_hwtstamp_get = liquidio_hwtstamp_get, + .ndo_hwtstamp_set = liquidio_hwtstamp_set, }; /** From 94037a0e18e3340912a039fa8435a69bee50bfd1 Mon Sep 17 00:00:00 2001 From: Vadim Fedorenko Date: Mon, 3 Nov 2025 15:09:48 +0000 Subject: [PATCH 0511/1381] net: liquidio_vf: convert to use ndo_hwtstamp callbacks The driver implemented SIOCSHWTSTAMP ioctl command only, but there is a way to get configuration back. Implement both ndo_hwtstamp_set and ndo_hwtstamp_set callbacks. Reviewed-by: Kory Maincent Signed-off-by: Vadim Fedorenko Link: https://patch.msgid.link/20251103150952.3538205-4-vadim.fedorenko@linux.dev Signed-off-by: Jakub Kicinski --- .../ethernet/cavium/liquidio/lio_vf_main.c | 48 ++++++++----------- 1 file changed, 19 insertions(+), 29 deletions(-) diff --git a/drivers/net/ethernet/cavium/liquidio/lio_vf_main.c b/drivers/net/ethernet/cavium/liquidio/lio_vf_main.c index 3230dff5ba05..e02942dbbcce 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_vf_main.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_vf_main.c @@ -1236,20 +1236,13 @@ liquidio_get_stats64(struct net_device *netdev, lstats->tx_carrier_errors; } -/** - * hwtstamp_ioctl - Handler for SIOCSHWTSTAMP ioctl - * @netdev: network device - * @ifr: interface request - */ -static int hwtstamp_ioctl(struct net_device *netdev, struct ifreq *ifr) +static int liquidio_hwtstamp_set(struct net_device *netdev, + struct kernel_hwtstamp_config *conf, + struct netlink_ext_ack *extack) { struct lio *lio = GET_LIO(netdev); - struct hwtstamp_config conf; - if (copy_from_user(&conf, ifr->ifr_data, sizeof(conf))) - return -EFAULT; - - switch (conf.tx_type) { + switch (conf->tx_type) { case HWTSTAMP_TX_ON: case HWTSTAMP_TX_OFF: break; @@ -1257,7 +1250,7 @@ static int hwtstamp_ioctl(struct net_device *netdev, struct ifreq *ifr) return -ERANGE; } - switch (conf.rx_filter) { + switch (conf->rx_filter) { case HWTSTAMP_FILTER_NONE: break; case HWTSTAMP_FILTER_ALL: @@ -1275,35 +1268,31 @@ static int hwtstamp_ioctl(struct net_device *netdev, struct ifreq *ifr) case HWTSTAMP_FILTER_PTP_V2_SYNC: case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ: case HWTSTAMP_FILTER_NTP_ALL: - conf.rx_filter = HWTSTAMP_FILTER_ALL; + conf->rx_filter = HWTSTAMP_FILTER_ALL; break; default: return -ERANGE; } - if (conf.rx_filter == HWTSTAMP_FILTER_ALL) + if (conf->rx_filter == HWTSTAMP_FILTER_ALL) ifstate_set(lio, LIO_IFSTATE_RX_TIMESTAMP_ENABLED); else ifstate_reset(lio, LIO_IFSTATE_RX_TIMESTAMP_ENABLED); - return copy_to_user(ifr->ifr_data, &conf, sizeof(conf)) ? -EFAULT : 0; + return 0; } -/** - * liquidio_ioctl - ioctl handler - * @netdev: network device - * @ifr: interface request - * @cmd: command - */ -static int liquidio_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) +static int liquidio_hwtstamp_get(struct net_device *netdev, + struct kernel_hwtstamp_config *conf) { - switch (cmd) { - case SIOCSHWTSTAMP: - return hwtstamp_ioctl(netdev, ifr); - default: - return -EOPNOTSUPP; - } + struct lio *lio = GET_LIO(netdev); + + /* TX timestamping is techically always on */ + conf->tx_type = HWTSTAMP_TX_ON; + conf->rx_filter = ifstate_check(lio, LIO_IFSTATE_RX_TIMESTAMP_ENABLED) ? + HWTSTAMP_FILTER_ALL : HWTSTAMP_FILTER_NONE; + return 0; } static void handle_timestamp(struct octeon_device *oct, u32 status, void *buf) @@ -1881,9 +1870,10 @@ static const struct net_device_ops lionetdevops = { .ndo_vlan_rx_add_vid = liquidio_vlan_rx_add_vid, .ndo_vlan_rx_kill_vid = liquidio_vlan_rx_kill_vid, .ndo_change_mtu = liquidio_change_mtu, - .ndo_eth_ioctl = liquidio_ioctl, .ndo_fix_features = liquidio_fix_features, .ndo_set_features = liquidio_set_features, + .ndo_hwtstamp_get = liquidio_hwtstamp_get, + .ndo_hwtstamp_set = liquidio_hwtstamp_set, }; static int lio_nic_info(struct octeon_recv_info *recv_info, void *buf) From 72c35e3a9589005e6ba55d63902eeb8b9fdb3a4e Mon Sep 17 00:00:00 2001 From: Vadim Fedorenko Date: Mon, 3 Nov 2025 15:09:49 +0000 Subject: [PATCH 0512/1381] net: octeon: mgmt: convert to use ndo_hwtstamp callbacks The driver implemented SIOCSHWTSTAMP ioctl command only. But it stores timestamping configuration, so it is possible to report it to users. Implement both ndo_hwtstamp_set and ndo_hwtstamp_get callbacks. After this the ndo_eth_ioctl effectively becomes phy_do_ioctl - adjust callback accordingly. Reviewed-by: Kory Maincent Signed-off-by: Vadim Fedorenko Link: https://patch.msgid.link/20251103150952.3538205-5-vadim.fedorenko@linux.dev Signed-off-by: Jakub Kicinski --- .../net/ethernet/cavium/octeon/octeon_mgmt.c | 62 ++++++++++--------- 1 file changed, 32 insertions(+), 30 deletions(-) diff --git a/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c b/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c index 393b9951490a..c190fc6538d4 100644 --- a/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c +++ b/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c @@ -690,19 +690,16 @@ static irqreturn_t octeon_mgmt_interrupt(int cpl, void *dev_id) return IRQ_HANDLED; } -static int octeon_mgmt_ioctl_hwtstamp(struct net_device *netdev, - struct ifreq *rq, int cmd) +static int octeon_mgmt_hwtstamp_set(struct net_device *netdev, + struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack) { struct octeon_mgmt *p = netdev_priv(netdev); - struct hwtstamp_config config; - union cvmx_mio_ptp_clock_cfg ptp; union cvmx_agl_gmx_rxx_frm_ctl rxx_frm_ctl; + union cvmx_mio_ptp_clock_cfg ptp; bool have_hw_timestamps = false; - if (copy_from_user(&config, rq->ifr_data, sizeof(config))) - return -EFAULT; - - /* Check the status of hardware for tiemstamps */ + /* Check the status of hardware for timestamps */ if (OCTEON_IS_MODEL(OCTEON_CN6XXX)) { /* Get the current state of the PTP clock */ ptp.u64 = cvmx_read_csr(CVMX_MIO_PTP_CLOCK_CFG); @@ -733,10 +730,12 @@ static int octeon_mgmt_ioctl_hwtstamp(struct net_device *netdev, have_hw_timestamps = true; } - if (!have_hw_timestamps) + if (!have_hw_timestamps) { + NL_SET_ERR_MSG_MOD(extack, "HW doesn't support timestamping"); return -EINVAL; + } - switch (config.tx_type) { + switch (config->tx_type) { case HWTSTAMP_TX_OFF: case HWTSTAMP_TX_ON: break; @@ -744,7 +743,7 @@ static int octeon_mgmt_ioctl_hwtstamp(struct net_device *netdev, return -ERANGE; } - switch (config.rx_filter) { + switch (config->rx_filter) { case HWTSTAMP_FILTER_NONE: p->has_rx_tstamp = false; rxx_frm_ctl.u64 = cvmx_read_csr(p->agl + AGL_GMX_RX_FRM_CTL); @@ -766,33 +765,34 @@ static int octeon_mgmt_ioctl_hwtstamp(struct net_device *netdev, case HWTSTAMP_FILTER_PTP_V2_SYNC: case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ: case HWTSTAMP_FILTER_NTP_ALL: - p->has_rx_tstamp = have_hw_timestamps; - config.rx_filter = HWTSTAMP_FILTER_ALL; - if (p->has_rx_tstamp) { - rxx_frm_ctl.u64 = cvmx_read_csr(p->agl + AGL_GMX_RX_FRM_CTL); - rxx_frm_ctl.s.ptp_mode = 1; - cvmx_write_csr(p->agl + AGL_GMX_RX_FRM_CTL, rxx_frm_ctl.u64); - } + p->has_rx_tstamp = true; + config->rx_filter = HWTSTAMP_FILTER_ALL; + rxx_frm_ctl.u64 = cvmx_read_csr(p->agl + AGL_GMX_RX_FRM_CTL); + rxx_frm_ctl.s.ptp_mode = 1; + cvmx_write_csr(p->agl + AGL_GMX_RX_FRM_CTL, rxx_frm_ctl.u64); break; default: return -ERANGE; } - if (copy_to_user(rq->ifr_data, &config, sizeof(config))) - return -EFAULT; - return 0; } -static int octeon_mgmt_ioctl(struct net_device *netdev, - struct ifreq *rq, int cmd) +static int octeon_mgmt_hwtstamp_get(struct net_device *netdev, + struct kernel_hwtstamp_config *config) { - switch (cmd) { - case SIOCSHWTSTAMP: - return octeon_mgmt_ioctl_hwtstamp(netdev, rq, cmd); - default: - return phy_do_ioctl(netdev, rq, cmd); - } + struct octeon_mgmt *p = netdev_priv(netdev); + + /* Check the status of hardware for timestamps */ + if (!OCTEON_IS_MODEL(OCTEON_CN6XXX)) + return -EINVAL; + + config->tx_type = HWTSTAMP_TX_ON; + config->rx_filter = p->has_rx_tstamp ? + HWTSTAMP_FILTER_ALL : + HWTSTAMP_FILTER_NONE; + + return 0; } static void octeon_mgmt_disable_link(struct octeon_mgmt *p) @@ -1370,11 +1370,13 @@ static const struct net_device_ops octeon_mgmt_ops = { .ndo_start_xmit = octeon_mgmt_xmit, .ndo_set_rx_mode = octeon_mgmt_set_rx_filtering, .ndo_set_mac_address = octeon_mgmt_set_mac_address, - .ndo_eth_ioctl = octeon_mgmt_ioctl, + .ndo_eth_ioctl = phy_do_ioctl, .ndo_change_mtu = octeon_mgmt_change_mtu, #ifdef CONFIG_NET_POLL_CONTROLLER .ndo_poll_controller = octeon_mgmt_poll_controller, #endif + .ndo_hwtstamp_get = octeon_mgmt_hwtstamp_get, + .ndo_hwtstamp_set = octeon_mgmt_hwtstamp_set, }; static int octeon_mgmt_probe(struct platform_device *pdev) From a23d0486d05a7b6d816d55099a7574c3c4ca4869 Mon Sep 17 00:00:00 2001 From: Vadim Fedorenko Date: Mon, 3 Nov 2025 15:09:50 +0000 Subject: [PATCH 0513/1381] net: thunderx: convert to use ndo_hwtstamp callbacks The driver implemented SIOCSHWTSTAMP ioctl command only, but it also stores configuration in private data, so it's possible to report it back to users. Implement both ndo_hwtstamp_set and ndo_hwtstamp_get callbacks. Reviewed-by: Kory Maincent Signed-off-by: Vadim Fedorenko Link: https://patch.msgid.link/20251103150952.3538205-6-vadim.fedorenko@linux.dev Signed-off-by: Jakub Kicinski --- .../net/ethernet/cavium/thunder/nicvf_main.c | 45 ++++++++++--------- 1 file changed, 25 insertions(+), 20 deletions(-) diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_main.c b/drivers/net/ethernet/cavium/thunder/nicvf_main.c index 1be2dc40a1a6..0b6e30a8feb0 100644 --- a/drivers/net/ethernet/cavium/thunder/nicvf_main.c +++ b/drivers/net/ethernet/cavium/thunder/nicvf_main.c @@ -1899,18 +1899,18 @@ static int nicvf_xdp(struct net_device *netdev, struct netdev_bpf *xdp) } } -static int nicvf_config_hwtstamp(struct net_device *netdev, struct ifreq *ifr) +static int nicvf_hwtstamp_set(struct net_device *netdev, + struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack) { - struct hwtstamp_config config; struct nicvf *nic = netdev_priv(netdev); - if (!nic->ptp_clock) + if (!nic->ptp_clock) { + NL_SET_ERR_MSG_MOD(extack, "HW timestamping is not supported"); return -ENODEV; + } - if (copy_from_user(&config, ifr->ifr_data, sizeof(config))) - return -EFAULT; - - switch (config.tx_type) { + switch (config->tx_type) { case HWTSTAMP_TX_OFF: case HWTSTAMP_TX_ON: break; @@ -1918,7 +1918,7 @@ static int nicvf_config_hwtstamp(struct net_device *netdev, struct ifreq *ifr) return -ERANGE; } - switch (config.rx_filter) { + switch (config->rx_filter) { case HWTSTAMP_FILTER_NONE: nic->hw_rx_tstamp = false; break; @@ -1937,7 +1937,7 @@ static int nicvf_config_hwtstamp(struct net_device *netdev, struct ifreq *ifr) case HWTSTAMP_FILTER_PTP_V2_SYNC: case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ: nic->hw_rx_tstamp = true; - config.rx_filter = HWTSTAMP_FILTER_ALL; + config->rx_filter = HWTSTAMP_FILTER_ALL; break; default: return -ERANGE; @@ -1946,20 +1946,24 @@ static int nicvf_config_hwtstamp(struct net_device *netdev, struct ifreq *ifr) if (netif_running(netdev)) nicvf_config_hw_rx_tstamp(nic, nic->hw_rx_tstamp); - if (copy_to_user(ifr->ifr_data, &config, sizeof(config))) - return -EFAULT; - return 0; } -static int nicvf_ioctl(struct net_device *netdev, struct ifreq *req, int cmd) +static int nicvf_hwtstamp_get(struct net_device *netdev, + struct kernel_hwtstamp_config *config) { - switch (cmd) { - case SIOCSHWTSTAMP: - return nicvf_config_hwtstamp(netdev, req); - default: - return -EOPNOTSUPP; - } + struct nicvf *nic = netdev_priv(netdev); + + if (!nic->ptp_clock) + return -ENODEV; + + /* TX timestamping is technically always on */ + config->tx_type = HWTSTAMP_TX_ON; + config->rx_filter = nic->hw_rx_tstamp ? + HWTSTAMP_FILTER_ALL : + HWTSTAMP_FILTER_NONE; + + return 0; } static void __nicvf_set_rx_mode_task(u8 mode, struct xcast_addr_list *mc_addrs, @@ -2081,8 +2085,9 @@ static const struct net_device_ops nicvf_netdev_ops = { .ndo_fix_features = nicvf_fix_features, .ndo_set_features = nicvf_set_features, .ndo_bpf = nicvf_xdp, - .ndo_eth_ioctl = nicvf_ioctl, .ndo_set_rx_mode = nicvf_set_rx_mode, + .ndo_hwtstamp_get = nicvf_hwtstamp_get, + .ndo_hwtstamp_set = nicvf_hwtstamp_set, }; static int nicvf_probe(struct pci_dev *pdev, const struct pci_device_id *ent) From d8fdc7069474a298ec63ee338b5fcc879a33c754 Mon Sep 17 00:00:00 2001 From: Vadim Fedorenko Date: Mon, 3 Nov 2025 15:09:51 +0000 Subject: [PATCH 0514/1381] net: pch_gbe: convert to use ndo_hwtstamp callbacks The driver implemented SIOCSHWTSTAMP ioctl command only, but it stores configuration in the private data, so it is possible to report it back to users. Implement both ndo_hwtstamp_set and ndo_hwtstamp_get callbacks. To properly report RX filter type, store it in hwts_rx_en instead of using this field as a simple flag. The logic didn't change because receive path used this field as boolean flag. Signed-off-by: Vadim Fedorenko Reviewed-by: Kory Maincent Link: https://patch.msgid.link/20251103150952.3538205-7-vadim.fedorenko@linux.dev Signed-off-by: Jakub Kicinski --- .../ethernet/oki-semi/pch_gbe/pch_gbe_main.c | 38 +++++++++++-------- 1 file changed, 23 insertions(+), 15 deletions(-) diff --git a/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c b/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c index e5a6f59af0b6..62f05f4569b1 100644 --- a/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c +++ b/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c @@ -198,23 +198,21 @@ pch_tx_timestamp(struct pch_gbe_adapter *adapter, struct sk_buff *skb) pch_ch_event_write(pdev, TX_SNAPSHOT_LOCKED); } -static int hwtstamp_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) +static int pch_gbe_hwtstamp_set(struct net_device *netdev, + struct kernel_hwtstamp_config *cfg, + struct netlink_ext_ack *extack) { - struct hwtstamp_config cfg; struct pch_gbe_adapter *adapter = netdev_priv(netdev); struct pci_dev *pdev; u8 station[20]; - if (copy_from_user(&cfg, ifr->ifr_data, sizeof(cfg))) - return -EFAULT; - /* Get ieee1588's dev information */ pdev = adapter->ptp_pdev; - if (cfg.tx_type != HWTSTAMP_TX_OFF && cfg.tx_type != HWTSTAMP_TX_ON) + if (cfg->tx_type != HWTSTAMP_TX_OFF && cfg->tx_type != HWTSTAMP_TX_ON) return -ERANGE; - switch (cfg.rx_filter) { + switch (cfg->rx_filter) { case HWTSTAMP_FILTER_NONE: adapter->hwts_rx_en = 0; break; @@ -223,17 +221,17 @@ static int hwtstamp_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) pch_ch_control_write(pdev, SLAVE_MODE | CAP_MODE0); break; case HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ: - adapter->hwts_rx_en = 1; + adapter->hwts_rx_en = cfg->rx_filter; pch_ch_control_write(pdev, MASTER_MODE | CAP_MODE0); break; case HWTSTAMP_FILTER_PTP_V2_L4_EVENT: - adapter->hwts_rx_en = 1; + adapter->hwts_rx_en = cfg->rx_filter; pch_ch_control_write(pdev, V2_MODE | CAP_MODE2); strcpy(station, PTP_L4_MULTICAST_SA); pch_set_station_address(station, pdev); break; case HWTSTAMP_FILTER_PTP_V2_L2_EVENT: - adapter->hwts_rx_en = 1; + adapter->hwts_rx_en = cfg->rx_filter; pch_ch_control_write(pdev, V2_MODE | CAP_MODE2); strcpy(station, PTP_L2_MULTICAST_SA); pch_set_station_address(station, pdev); @@ -242,12 +240,23 @@ static int hwtstamp_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) return -ERANGE; } - adapter->hwts_tx_en = cfg.tx_type == HWTSTAMP_TX_ON; + adapter->hwts_tx_en = cfg->tx_type == HWTSTAMP_TX_ON; /* Clear out any old time stamps. */ pch_ch_event_write(pdev, TX_SNAPSHOT_LOCKED | RX_SNAPSHOT_LOCKED); - return copy_to_user(ifr->ifr_data, &cfg, sizeof(cfg)) ? -EFAULT : 0; + return 0; +} + +static int pch_gbe_hwtstamp_get(struct net_device *netdev, + struct kernel_hwtstamp_config *cfg) +{ + struct pch_gbe_adapter *adapter = netdev_priv(netdev); + + cfg->tx_type = adapter->hwts_tx_en ? HWTSTAMP_TX_ON : HWTSTAMP_TX_OFF; + cfg->rx_filter = adapter->hwts_rx_en; + + return 0; } static inline void pch_gbe_mac_load_mac_addr(struct pch_gbe_hw *hw) @@ -2234,9 +2243,6 @@ static int pch_gbe_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) netdev_dbg(netdev, "cmd : 0x%04x\n", cmd); - if (cmd == SIOCSHWTSTAMP) - return hwtstamp_ioctl(netdev, ifr, cmd); - return generic_mii_ioctl(&adapter->mii, if_mii(ifr), cmd, NULL); } @@ -2328,6 +2334,8 @@ static const struct net_device_ops pch_gbe_netdev_ops = { #ifdef CONFIG_NET_POLL_CONTROLLER .ndo_poll_controller = pch_gbe_netpoll, #endif + .ndo_hwtstamp_get = pch_gbe_hwtstamp_get, + .ndo_hwtstamp_set = pch_gbe_hwtstamp_set, }; static pci_ers_result_t pch_gbe_io_error_detected(struct pci_dev *pdev, From 3f02b82725576a85a1219547e28a2ab30b53666f Mon Sep 17 00:00:00 2001 From: Vadim Fedorenko Date: Mon, 3 Nov 2025 17:29:02 +0000 Subject: [PATCH 0515/1381] ti: netcp: convert to ndo_hwtstamp callbacks Convert TI NetCP driver to use ndo_hwtstamp_get()/ndo_hwtstamp_set() callbacks. The logic is slightly changed, because I believe the original logic was not really correct. Config reading part is using the very first module to get the configuration instead of iterating over all of them and keep the last one as the configuration is supposed to be identical for all modules. HW timestamp config set path is now trying to configure all modules, but in case of error from one module it adds extack message. This way the configuration will be as synchronized as possible. There are only 2 modules using netcp core infrastructure, and both use the very same function to configure HW timestamping, so no actual difference in behavior is expected. Signed-off-by: Vadim Fedorenko Reviewed-by: Kory Maincent Link: https://patch.msgid.link/20251103172902.3538392-1-vadim.fedorenko@linux.dev Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/ti/netcp.h | 5 ++ drivers/net/ethernet/ti/netcp_core.c | 58 +++++++++++++++++++++ drivers/net/ethernet/ti/netcp_ethss.c | 72 +++++++++++++++------------ 3 files changed, 103 insertions(+), 32 deletions(-) diff --git a/drivers/net/ethernet/ti/netcp.h b/drivers/net/ethernet/ti/netcp.h index 7007eb8bed36..b9cbd3b4a8a2 100644 --- a/drivers/net/ethernet/ti/netcp.h +++ b/drivers/net/ethernet/ti/netcp.h @@ -207,6 +207,11 @@ struct netcp_module { int (*del_vid)(void *intf_priv, int vid); int (*ioctl)(void *intf_priv, struct ifreq *req, int cmd); int (*set_rx_mode)(void *intf_priv, bool promisc); + int (*hwtstamp_get)(void *intf_priv, + struct kernel_hwtstamp_config *cfg); + int (*hwtstamp_set)(void *intf_priv, + struct kernel_hwtstamp_config *cfg, + struct netlink_ext_ack *extack); /* used internally */ struct list_head module_list; diff --git a/drivers/net/ethernet/ti/netcp_core.c b/drivers/net/ethernet/ti/netcp_core.c index 857820657bac..cee2686a4893 100644 --- a/drivers/net/ethernet/ti/netcp_core.c +++ b/drivers/net/ethernet/ti/netcp_core.c @@ -1781,6 +1781,62 @@ static int netcp_ndo_stop(struct net_device *ndev) return 0; } +static int netcp_ndo_hwtstamp_get(struct net_device *ndev, + struct kernel_hwtstamp_config *config) +{ + struct netcp_intf *netcp = netdev_priv(ndev); + struct netcp_intf_modpriv *intf_modpriv; + struct netcp_module *module; + int err = -EOPNOTSUPP; + + if (!netif_running(ndev)) + return -EINVAL; + + for_each_module(netcp, intf_modpriv) { + module = intf_modpriv->netcp_module; + if (!module->hwtstamp_get) + continue; + + err = module->hwtstamp_get(intf_modpriv->module_priv, config); + break; + } + + return err; +} + +static int netcp_ndo_hwtstamp_set(struct net_device *ndev, + struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack) +{ + struct netcp_intf *netcp = netdev_priv(ndev); + struct netcp_intf_modpriv *intf_modpriv; + struct netcp_module *module; + int ret = -1, err = -EOPNOTSUPP; + + if (!netif_running(ndev)) + return -EINVAL; + + for_each_module(netcp, intf_modpriv) { + module = intf_modpriv->netcp_module; + if (!module->hwtstamp_set) + continue; + + err = module->hwtstamp_set(intf_modpriv->module_priv, config, + extack); + if ((err < 0) && (err != -EOPNOTSUPP)) { + NL_SET_ERR_MSG_WEAK_MOD(extack, + "At least one module failed to setup HW timestamps"); + ret = err; + goto out; + } + if (err == 0) + ret = err; + } + +out: + return (ret == 0) ? 0 : err; +} + static int netcp_ndo_ioctl(struct net_device *ndev, struct ifreq *req, int cmd) { @@ -1952,6 +2008,8 @@ static const struct net_device_ops netcp_netdev_ops = { .ndo_tx_timeout = netcp_ndo_tx_timeout, .ndo_select_queue = dev_pick_tx_zero, .ndo_setup_tc = netcp_setup_tc, + .ndo_hwtstamp_get = netcp_ndo_hwtstamp_get, + .ndo_hwtstamp_set = netcp_ndo_hwtstamp_set, }; static int netcp_create_interface(struct netcp_device *netcp_device, diff --git a/drivers/net/ethernet/ti/netcp_ethss.c b/drivers/net/ethernet/ti/netcp_ethss.c index 55a1a96cd834..0ae44112812c 100644 --- a/drivers/net/ethernet/ti/netcp_ethss.c +++ b/drivers/net/ethernet/ti/netcp_ethss.c @@ -2591,20 +2591,26 @@ static int gbe_rxtstamp(struct gbe_intf *gbe_intf, struct netcp_packet *p_info) return 0; } -static int gbe_hwtstamp_get(struct gbe_intf *gbe_intf, struct ifreq *ifr) +static int gbe_hwtstamp_get(void *intf_priv, struct kernel_hwtstamp_config *cfg) { - struct gbe_priv *gbe_dev = gbe_intf->gbe_dev; - struct cpts *cpts = gbe_dev->cpts; - struct hwtstamp_config cfg; + struct gbe_intf *gbe_intf = intf_priv; + struct gbe_priv *gbe_dev; + struct phy_device *phy; - if (!cpts) + gbe_dev = gbe_intf->gbe_dev; + + if (!gbe_dev->cpts) return -EOPNOTSUPP; - cfg.flags = 0; - cfg.tx_type = gbe_dev->tx_ts_enabled ? HWTSTAMP_TX_ON : HWTSTAMP_TX_OFF; - cfg.rx_filter = gbe_dev->rx_ts_enabled; + phy = gbe_intf->slave->phy; + if (phy_has_hwtstamp(phy)) + return -EOPNOTSUPP; - return copy_to_user(ifr->ifr_data, &cfg, sizeof(cfg)) ? -EFAULT : 0; + cfg->flags = 0; + cfg->tx_type = gbe_dev->tx_ts_enabled ? HWTSTAMP_TX_ON : HWTSTAMP_TX_OFF; + cfg->rx_filter = gbe_dev->rx_ts_enabled; + + return 0; } static void gbe_hwtstamp(struct gbe_intf *gbe_intf) @@ -2637,19 +2643,23 @@ static void gbe_hwtstamp(struct gbe_intf *gbe_intf) writel(ctl, GBE_REG_ADDR(slave, port_regs, ts_ctl_ltype2)); } -static int gbe_hwtstamp_set(struct gbe_intf *gbe_intf, struct ifreq *ifr) +static int gbe_hwtstamp_set(void *intf_priv, struct kernel_hwtstamp_config *cfg, + struct netlink_ext_ack *extack) { - struct gbe_priv *gbe_dev = gbe_intf->gbe_dev; - struct cpts *cpts = gbe_dev->cpts; - struct hwtstamp_config cfg; + struct gbe_intf *gbe_intf = intf_priv; + struct gbe_priv *gbe_dev; + struct phy_device *phy; - if (!cpts) + gbe_dev = gbe_intf->gbe_dev; + + if (!gbe_dev->cpts) return -EOPNOTSUPP; - if (copy_from_user(&cfg, ifr->ifr_data, sizeof(cfg))) - return -EFAULT; + phy = gbe_intf->slave->phy; + if (phy_has_hwtstamp(phy)) + return phy->mii_ts->hwtstamp(phy->mii_ts, cfg, extack); - switch (cfg.tx_type) { + switch (cfg->tx_type) { case HWTSTAMP_TX_OFF: gbe_dev->tx_ts_enabled = 0; break; @@ -2660,7 +2670,7 @@ static int gbe_hwtstamp_set(struct gbe_intf *gbe_intf, struct ifreq *ifr) return -ERANGE; } - switch (cfg.rx_filter) { + switch (cfg->rx_filter) { case HWTSTAMP_FILTER_NONE: gbe_dev->rx_ts_enabled = HWTSTAMP_FILTER_NONE; break; @@ -2668,7 +2678,7 @@ static int gbe_hwtstamp_set(struct gbe_intf *gbe_intf, struct ifreq *ifr) case HWTSTAMP_FILTER_PTP_V1_L4_SYNC: case HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ: gbe_dev->rx_ts_enabled = HWTSTAMP_FILTER_PTP_V1_L4_EVENT; - cfg.rx_filter = HWTSTAMP_FILTER_PTP_V1_L4_EVENT; + cfg->rx_filter = HWTSTAMP_FILTER_PTP_V1_L4_EVENT; break; case HWTSTAMP_FILTER_PTP_V2_L4_EVENT: case HWTSTAMP_FILTER_PTP_V2_L4_SYNC: @@ -2680,7 +2690,7 @@ static int gbe_hwtstamp_set(struct gbe_intf *gbe_intf, struct ifreq *ifr) case HWTSTAMP_FILTER_PTP_V2_SYNC: case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ: gbe_dev->rx_ts_enabled = HWTSTAMP_FILTER_PTP_V2_EVENT; - cfg.rx_filter = HWTSTAMP_FILTER_PTP_V2_EVENT; + cfg->rx_filter = HWTSTAMP_FILTER_PTP_V2_EVENT; break; default: return -ERANGE; @@ -2688,7 +2698,7 @@ static int gbe_hwtstamp_set(struct gbe_intf *gbe_intf, struct ifreq *ifr) gbe_hwtstamp(gbe_intf); - return copy_to_user(ifr->ifr_data, &cfg, sizeof(cfg)) ? -EFAULT : 0; + return 0; } static void gbe_register_cpts(struct gbe_priv *gbe_dev) @@ -2745,12 +2755,15 @@ static inline void gbe_unregister_cpts(struct gbe_priv *gbe_dev) { } -static inline int gbe_hwtstamp_get(struct gbe_intf *gbe_intf, struct ifreq *req) +static inline int gbe_hwtstamp_get(struct gbe_intf *gbe_intf, + struct kernel_hwtstamp_config *cfg) { return -EOPNOTSUPP; } -static inline int gbe_hwtstamp_set(struct gbe_intf *gbe_intf, struct ifreq *req) +static inline int gbe_hwtstamp_set(struct gbe_intf *gbe_intf, + struct kernel_hwtstamp_config *cfg, + struct netlink_ext_ack *extack) { return -EOPNOTSUPP; } @@ -2816,15 +2829,6 @@ static int gbe_ioctl(void *intf_priv, struct ifreq *req, int cmd) struct gbe_intf *gbe_intf = intf_priv; struct phy_device *phy = gbe_intf->slave->phy; - if (!phy_has_hwtstamp(phy)) { - switch (cmd) { - case SIOCGHWTSTAMP: - return gbe_hwtstamp_get(gbe_intf, req); - case SIOCSHWTSTAMP: - return gbe_hwtstamp_set(gbe_intf, req); - } - } - if (phy) return phy_mii_ioctl(phy, req, cmd); @@ -3824,6 +3828,8 @@ static struct netcp_module gbe_module = { .add_vid = gbe_add_vid, .del_vid = gbe_del_vid, .ioctl = gbe_ioctl, + .hwtstamp_get = gbe_hwtstamp_get, + .hwtstamp_set = gbe_hwtstamp_set, }; static struct netcp_module xgbe_module = { @@ -3841,6 +3847,8 @@ static struct netcp_module xgbe_module = { .add_vid = gbe_add_vid, .del_vid = gbe_del_vid, .ioctl = gbe_ioctl, + .hwtstamp_get = gbe_hwtstamp_get, + .hwtstamp_set = gbe_hwtstamp_set, }; static int __init keystone_gbe_init(void) From ee61c10cd4820e8844dba4315f2d1e522f1f3b98 Mon Sep 17 00:00:00 2001 From: Dong Yibo Date: Sat, 1 Nov 2025 09:38:45 +0800 Subject: [PATCH 0516/1381] net: rnpgbe: Add build support for rnpgbe Add build options and doc for mucse. Initialize pci device access for MUCSE devices. Signed-off-by: Dong Yibo Reviewed-by: Andrew Lunn Reviewed-by: Vadim Fedorenko Reviewed-by: MD Danish Anwar Link: https://patch.msgid.link/20251101013849.120565-2-dong100@mucse.com Signed-off-by: Jakub Kicinski --- .../device_drivers/ethernet/index.rst | 1 + .../device_drivers/ethernet/mucse/rnpgbe.rst | 17 +++ MAINTAINERS | 8 ++ drivers/net/ethernet/Kconfig | 1 + drivers/net/ethernet/Makefile | 1 + drivers/net/ethernet/mucse/Kconfig | 33 +++++ drivers/net/ethernet/mucse/Makefile | 7 + drivers/net/ethernet/mucse/rnpgbe/Makefile | 8 ++ drivers/net/ethernet/mucse/rnpgbe/rnpgbe.h | 18 +++ .../net/ethernet/mucse/rnpgbe/rnpgbe_main.c | 120 ++++++++++++++++++ 10 files changed, 214 insertions(+) create mode 100644 Documentation/networking/device_drivers/ethernet/mucse/rnpgbe.rst create mode 100644 drivers/net/ethernet/mucse/Kconfig create mode 100644 drivers/net/ethernet/mucse/Makefile create mode 100644 drivers/net/ethernet/mucse/rnpgbe/Makefile create mode 100644 drivers/net/ethernet/mucse/rnpgbe/rnpgbe.h create mode 100644 drivers/net/ethernet/mucse/rnpgbe/rnpgbe_main.c diff --git a/Documentation/networking/device_drivers/ethernet/index.rst b/Documentation/networking/device_drivers/ethernet/index.rst index 7cfcd183054f..bcc02355f828 100644 --- a/Documentation/networking/device_drivers/ethernet/index.rst +++ b/Documentation/networking/device_drivers/ethernet/index.rst @@ -47,6 +47,7 @@ Contents: mellanox/mlx5/index meta/fbnic microsoft/netvsc + mucse/rnpgbe neterion/s2io netronome/nfp pensando/ionic diff --git a/Documentation/networking/device_drivers/ethernet/mucse/rnpgbe.rst b/Documentation/networking/device_drivers/ethernet/mucse/rnpgbe.rst new file mode 100644 index 000000000000..d35cf8a46b6c --- /dev/null +++ b/Documentation/networking/device_drivers/ethernet/mucse/rnpgbe.rst @@ -0,0 +1,17 @@ +.. SPDX-License-Identifier: GPL-2.0 + +=========================================================== +Linux Base Driver for MUCSE(R) Gigabit PCI Express Adapters +=========================================================== + +Contents +======== + +- Identifying Your Adapter + +Identifying Your Adapter +======================== +The driver is compatible with devices based on the following: + + * MUCSE(R) Ethernet Controller N210 series + * MUCSE(R) Ethernet Controller N500 series diff --git a/MAINTAINERS b/MAINTAINERS index 1ab7e8746299..bdf0a3a0dd36 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -17610,6 +17610,14 @@ T: git git://linuxtv.org/media.git F: Documentation/devicetree/bindings/media/i2c/aptina,mt9v111.yaml F: drivers/media/i2c/mt9v111.c +MUCSE ETHERNET DRIVER +M: Yibo Dong +L: netdev@vger.kernel.org +S: Maintained +W: https://www.mucse.com/en/ +F: Documentation/networking/device_drivers/ethernet/mucse/ +F: drivers/net/ethernet/mucse/ + MULTIFUNCTION DEVICES (MFD) M: Lee Jones S: Maintained diff --git a/drivers/net/ethernet/Kconfig b/drivers/net/ethernet/Kconfig index aead145dd91d..4a1b368ca7e6 100644 --- a/drivers/net/ethernet/Kconfig +++ b/drivers/net/ethernet/Kconfig @@ -129,6 +129,7 @@ source "drivers/net/ethernet/microchip/Kconfig" source "drivers/net/ethernet/mscc/Kconfig" source "drivers/net/ethernet/microsoft/Kconfig" source "drivers/net/ethernet/moxa/Kconfig" +source "drivers/net/ethernet/mucse/Kconfig" source "drivers/net/ethernet/myricom/Kconfig" config FEALNX diff --git a/drivers/net/ethernet/Makefile b/drivers/net/ethernet/Makefile index 998dd628b202..2e18df8ca8ec 100644 --- a/drivers/net/ethernet/Makefile +++ b/drivers/net/ethernet/Makefile @@ -65,6 +65,7 @@ obj-$(CONFIG_NET_VENDOR_MICREL) += micrel/ obj-$(CONFIG_NET_VENDOR_MICROCHIP) += microchip/ obj-$(CONFIG_NET_VENDOR_MICROSEMI) += mscc/ obj-$(CONFIG_NET_VENDOR_MOXART) += moxa/ +obj-$(CONFIG_NET_VENDOR_MUCSE) += mucse/ obj-$(CONFIG_NET_VENDOR_MYRI) += myricom/ obj-$(CONFIG_FEALNX) += fealnx.o obj-$(CONFIG_NET_VENDOR_NATSEMI) += natsemi/ diff --git a/drivers/net/ethernet/mucse/Kconfig b/drivers/net/ethernet/mucse/Kconfig new file mode 100644 index 000000000000..0b3e853d625f --- /dev/null +++ b/drivers/net/ethernet/mucse/Kconfig @@ -0,0 +1,33 @@ +# SPDX-License-Identifier: GPL-2.0-only +# +# Mucse network device configuration +# + +config NET_VENDOR_MUCSE + bool "Mucse devices" + default y + help + If you have a network (Ethernet) card from Mucse(R), say Y. + + Note that the answer to this question doesn't directly affect the + kernel: saying N will just cause the configurator to skip all + the questions about Mucse(R) cards. If you say Y, you will + be asked for your specific card in the following questions. + +if NET_VENDOR_MUCSE + +config MGBE + tristate "Mucse(R) 1GbE PCI Express adapters support" + depends on PCI + help + This driver supports Mucse(R) 1GbE PCI Express family of + adapters. + + More specific information on configuring the driver is in + . + + To compile this driver as a module, choose M here. The module + will be called rnpgbe. + +endif # NET_VENDOR_MUCSE + diff --git a/drivers/net/ethernet/mucse/Makefile b/drivers/net/ethernet/mucse/Makefile new file mode 100644 index 000000000000..675173fa05f7 --- /dev/null +++ b/drivers/net/ethernet/mucse/Makefile @@ -0,0 +1,7 @@ +# SPDX-License-Identifier: GPL-2.0 +# Copyright(c) 2020 - 2025 MUCSE Corporation. +# +# Makefile for the MUCSE(R) network device drivers +# + +obj-$(CONFIG_MGBE) += rnpgbe/ diff --git a/drivers/net/ethernet/mucse/rnpgbe/Makefile b/drivers/net/ethernet/mucse/rnpgbe/Makefile new file mode 100644 index 000000000000..9df536f0d04c --- /dev/null +++ b/drivers/net/ethernet/mucse/rnpgbe/Makefile @@ -0,0 +1,8 @@ +# SPDX-License-Identifier: GPL-2.0 +# Copyright(c) 2020 - 2025 MUCSE Corporation. +# +# Makefile for the MUCSE(R) 1GbE PCI Express ethernet driver +# + +obj-$(CONFIG_MGBE) += rnpgbe.o +rnpgbe-objs := rnpgbe_main.o diff --git a/drivers/net/ethernet/mucse/rnpgbe/rnpgbe.h b/drivers/net/ethernet/mucse/rnpgbe/rnpgbe.h new file mode 100644 index 000000000000..d3439d28c654 --- /dev/null +++ b/drivers/net/ethernet/mucse/rnpgbe/rnpgbe.h @@ -0,0 +1,18 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright(c) 2020 - 2025 Mucse Corporation. */ + +#ifndef _RNPGBE_H +#define _RNPGBE_H + +enum rnpgbe_boards { + board_n500, + board_n210 +}; + +/* Device IDs */ +#define PCI_VENDOR_ID_MUCSE 0x8848 +#define RNPGBE_DEVICE_ID_N500_QUAD_PORT 0x8308 +#define RNPGBE_DEVICE_ID_N500_DUAL_PORT 0x8318 +#define RNPGBE_DEVICE_ID_N210 0x8208 +#define RNPGBE_DEVICE_ID_N210L 0x820a +#endif /* _RNPGBE_H */ diff --git a/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_main.c b/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_main.c new file mode 100644 index 000000000000..019e819fb497 --- /dev/null +++ b/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_main.c @@ -0,0 +1,120 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright(c) 2020 - 2025 Mucse Corporation. */ + +#include + +#include "rnpgbe.h" + +static const char rnpgbe_driver_name[] = "rnpgbe"; + +/* rnpgbe_pci_tbl - PCI Device ID Table + * + * { PCI_VDEVICE(Vendor ID, Device ID), + * private_data (used for different hw chip) } + */ +static struct pci_device_id rnpgbe_pci_tbl[] = { + { PCI_VDEVICE(MUCSE, RNPGBE_DEVICE_ID_N210), board_n210 }, + { PCI_VDEVICE(MUCSE, RNPGBE_DEVICE_ID_N210L), board_n210 }, + { PCI_VDEVICE(MUCSE, RNPGBE_DEVICE_ID_N500_DUAL_PORT), board_n500 }, + { PCI_VDEVICE(MUCSE, RNPGBE_DEVICE_ID_N500_QUAD_PORT), board_n500 }, + /* required last entry */ + {0, }, +}; + +/** + * rnpgbe_probe - Device initialization routine + * @pdev: PCI device information struct + * @id: entry in rnpgbe_pci_tbl + * + * rnpgbe_probe initializes a PF adapter identified by a pci_dev + * structure. + * + * Return: 0 on success, negative errno on failure + **/ +static int rnpgbe_probe(struct pci_dev *pdev, const struct pci_device_id *id) +{ + int err; + + err = pci_enable_device_mem(pdev); + if (err) + return err; + + err = dma_set_coherent_mask(&pdev->dev, DMA_BIT_MASK(56)); + if (err) { + dev_err(&pdev->dev, + "No usable DMA configuration, aborting %d\n", err); + goto err_disable_dev; + } + + err = pci_request_mem_regions(pdev, rnpgbe_driver_name); + if (err) { + dev_err(&pdev->dev, + "pci_request_selected_regions failed %d\n", err); + goto err_disable_dev; + } + + pci_set_master(pdev); + err = pci_save_state(pdev); + if (err) { + dev_err(&pdev->dev, "pci_save_state failed %d\n", err); + goto err_free_regions; + } + + return 0; +err_free_regions: + pci_release_mem_regions(pdev); +err_disable_dev: + pci_disable_device(pdev); + return err; +} + +/** + * rnpgbe_remove - Device removal routine + * @pdev: PCI device information struct + * + * rnpgbe_remove is called by the PCI subsystem to alert the driver + * that it should release a PCI device. This could be caused by a + * Hot-Plug event, or because the driver is going to be removed from + * memory. + **/ +static void rnpgbe_remove(struct pci_dev *pdev) +{ + pci_release_mem_regions(pdev); + pci_disable_device(pdev); +} + +/** + * rnpgbe_dev_shutdown - Device shutdown routine + * @pdev: PCI device information struct + **/ +static void rnpgbe_dev_shutdown(struct pci_dev *pdev) +{ + pci_disable_device(pdev); +} + +/** + * rnpgbe_shutdown - Device shutdown routine + * @pdev: PCI device information struct + * + * rnpgbe_shutdown is called by the PCI subsystem to alert the driver + * that os shutdown. Device should setup wakeup state here. + **/ +static void rnpgbe_shutdown(struct pci_dev *pdev) +{ + rnpgbe_dev_shutdown(pdev); +} + +static struct pci_driver rnpgbe_driver = { + .name = rnpgbe_driver_name, + .id_table = rnpgbe_pci_tbl, + .probe = rnpgbe_probe, + .remove = rnpgbe_remove, + .shutdown = rnpgbe_shutdown, +}; + +module_pci_driver(rnpgbe_driver); + +MODULE_DEVICE_TABLE(pci, rnpgbe_pci_tbl); +MODULE_AUTHOR("Yibo Dong, "); +MODULE_DESCRIPTION("Mucse(R) 1 Gigabit PCI Express Network Driver"); +MODULE_LICENSE("GPL"); From 1b7f85f733fd243d7c9073b9ff3d93e6ba5d1055 Mon Sep 17 00:00:00 2001 From: Dong Yibo Date: Sat, 1 Nov 2025 09:38:46 +0800 Subject: [PATCH 0517/1381] net: rnpgbe: Add n500/n210 chip support with BAR2 mapping Add hardware initialization foundation for MUCSE 1Gbe controller, including: 1. Map PCI BAR2 as hardware register base; 2. Bind PCI device to driver private data (struct mucse) and initialize hardware context (struct mucse_hw); 3. Reserve board-specific init framework via rnpgbe_init_hw. Signed-off-by: Dong Yibo Reviewed-by: Vadim Fedorenko Reviewed-by: MD Danish Anwar Link: https://patch.msgid.link/20251101013849.120565-3-dong100@mucse.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mucse/rnpgbe/rnpgbe.h | 10 +++ drivers/net/ethernet/mucse/rnpgbe/rnpgbe_hw.h | 8 ++ .../net/ethernet/mucse/rnpgbe/rnpgbe_main.c | 80 +++++++++++++++++++ 3 files changed, 98 insertions(+) create mode 100644 drivers/net/ethernet/mucse/rnpgbe/rnpgbe_hw.h diff --git a/drivers/net/ethernet/mucse/rnpgbe/rnpgbe.h b/drivers/net/ethernet/mucse/rnpgbe/rnpgbe.h index d3439d28c654..a121ce4872a6 100644 --- a/drivers/net/ethernet/mucse/rnpgbe/rnpgbe.h +++ b/drivers/net/ethernet/mucse/rnpgbe/rnpgbe.h @@ -9,6 +9,16 @@ enum rnpgbe_boards { board_n210 }; +struct mucse_hw { + void __iomem *hw_addr; +}; + +struct mucse { + struct net_device *netdev; + struct pci_dev *pdev; + struct mucse_hw hw; +}; + /* Device IDs */ #define PCI_VENDOR_ID_MUCSE 0x8848 #define RNPGBE_DEVICE_ID_N500_QUAD_PORT 0x8308 diff --git a/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_hw.h b/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_hw.h new file mode 100644 index 000000000000..3a779806e8be --- /dev/null +++ b/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_hw.h @@ -0,0 +1,8 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright(c) 2020 - 2025 Mucse Corporation. */ + +#ifndef _RNPGBE_HW_H +#define _RNPGBE_HW_H + +#define RNPGBE_MAX_QUEUES 8 +#endif /* _RNPGBE_HW_H */ diff --git a/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_main.c b/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_main.c index 019e819fb497..305657d73e25 100644 --- a/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_main.c +++ b/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_main.c @@ -2,8 +2,11 @@ /* Copyright(c) 2020 - 2025 Mucse Corporation. */ #include +#include +#include #include "rnpgbe.h" +#include "rnpgbe_hw.h" static const char rnpgbe_driver_name[] = "rnpgbe"; @@ -21,6 +24,54 @@ static struct pci_device_id rnpgbe_pci_tbl[] = { {0, }, }; +/** + * rnpgbe_add_adapter - Add netdev for this pci_dev + * @pdev: PCI device information structure + * @board_type: board type + * + * rnpgbe_add_adapter initializes a netdev for this pci_dev + * structure. Initializes Bar map, private structure, and a + * hardware reset occur. + * + * Return: 0 on success, negative errno on failure + **/ +static int rnpgbe_add_adapter(struct pci_dev *pdev, + int board_type) +{ + struct net_device *netdev; + void __iomem *hw_addr; + struct mucse *mucse; + struct mucse_hw *hw; + int err; + + netdev = alloc_etherdev_mq(sizeof(struct mucse), RNPGBE_MAX_QUEUES); + if (!netdev) + return -ENOMEM; + + SET_NETDEV_DEV(netdev, &pdev->dev); + mucse = netdev_priv(netdev); + mucse->netdev = netdev; + mucse->pdev = pdev; + pci_set_drvdata(pdev, mucse); + + hw = &mucse->hw; + hw_addr = devm_ioremap(&pdev->dev, + pci_resource_start(pdev, 2), + pci_resource_len(pdev, 2)); + if (!hw_addr) { + err = -EIO; + goto err_free_net; + } + + hw->hw_addr = hw_addr; + + return 0; + +err_free_net: + free_netdev(netdev); + return err; +} + /** * rnpgbe_probe - Device initialization routine * @pdev: PCI device information struct @@ -33,6 +84,7 @@ static struct pci_device_id rnpgbe_pci_tbl[] = { **/ static int rnpgbe_probe(struct pci_dev *pdev, const struct pci_device_id *id) { + int board_type = id->driver_data; int err; err = pci_enable_device_mem(pdev); @@ -60,6 +112,10 @@ static int rnpgbe_probe(struct pci_dev *pdev, const struct pci_device_id *id) goto err_free_regions; } + err = rnpgbe_add_adapter(pdev, board_type); + if (err) + goto err_free_regions; + return 0; err_free_regions: pci_release_mem_regions(pdev); @@ -68,6 +124,23 @@ static int rnpgbe_probe(struct pci_dev *pdev, const struct pci_device_id *id) return err; } +/** + * rnpgbe_rm_adapter - Remove netdev for this mucse structure + * @pdev: PCI device information struct + * + * rnpgbe_rm_adapter remove a netdev for this mucse structure + **/ +static void rnpgbe_rm_adapter(struct pci_dev *pdev) +{ + struct mucse *mucse = pci_get_drvdata(pdev); + struct net_device *netdev; + + if (!mucse) + return; + netdev = mucse->netdev; + free_netdev(netdev); +} + /** * rnpgbe_remove - Device removal routine * @pdev: PCI device information struct @@ -79,6 +152,7 @@ static int rnpgbe_probe(struct pci_dev *pdev, const struct pci_device_id *id) **/ static void rnpgbe_remove(struct pci_dev *pdev) { + rnpgbe_rm_adapter(pdev); pci_release_mem_regions(pdev); pci_disable_device(pdev); } @@ -89,6 +163,12 @@ static void rnpgbe_remove(struct pci_dev *pdev) **/ static void rnpgbe_dev_shutdown(struct pci_dev *pdev) { + struct mucse *mucse = pci_get_drvdata(pdev); + struct net_device *netdev = mucse->netdev; + + rtnl_lock(); + netif_device_detach(netdev); + rtnl_unlock(); pci_disable_device(pdev); } From 4543534c3ef5115189ee2981e0911d1626af57ed Mon Sep 17 00:00:00 2001 From: Dong Yibo Date: Sat, 1 Nov 2025 09:38:47 +0800 Subject: [PATCH 0518/1381] net: rnpgbe: Add basic mbx ops support Add fundamental mailbox (MBX) communication operations between PF (Physical Function) and firmware for n500/n210 chips Signed-off-by: Dong Yibo Reviewed-by: Vadim Fedorenko Link: https://patch.msgid.link/20251101013849.120565-4-dong100@mucse.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mucse/rnpgbe/Makefile | 4 +- drivers/net/ethernet/mucse/rnpgbe/rnpgbe.h | 17 + .../net/ethernet/mucse/rnpgbe/rnpgbe_chip.c | 70 +++ drivers/net/ethernet/mucse/rnpgbe/rnpgbe_hw.h | 7 + .../net/ethernet/mucse/rnpgbe/rnpgbe_main.c | 5 + .../net/ethernet/mucse/rnpgbe/rnpgbe_mbx.c | 405 ++++++++++++++++++ .../net/ethernet/mucse/rnpgbe/rnpgbe_mbx.h | 20 + 7 files changed, 527 insertions(+), 1 deletion(-) create mode 100644 drivers/net/ethernet/mucse/rnpgbe/rnpgbe_chip.c create mode 100644 drivers/net/ethernet/mucse/rnpgbe/rnpgbe_mbx.c create mode 100644 drivers/net/ethernet/mucse/rnpgbe/rnpgbe_mbx.h diff --git a/drivers/net/ethernet/mucse/rnpgbe/Makefile b/drivers/net/ethernet/mucse/rnpgbe/Makefile index 9df536f0d04c..5fc878ada4b1 100644 --- a/drivers/net/ethernet/mucse/rnpgbe/Makefile +++ b/drivers/net/ethernet/mucse/rnpgbe/Makefile @@ -5,4 +5,6 @@ # obj-$(CONFIG_MGBE) += rnpgbe.o -rnpgbe-objs := rnpgbe_main.o +rnpgbe-objs := rnpgbe_main.o\ + rnpgbe_chip.o\ + rnpgbe_mbx.o diff --git a/drivers/net/ethernet/mucse/rnpgbe/rnpgbe.h b/drivers/net/ethernet/mucse/rnpgbe/rnpgbe.h index a121ce4872a6..4c70b0cedd1f 100644 --- a/drivers/net/ethernet/mucse/rnpgbe/rnpgbe.h +++ b/drivers/net/ethernet/mucse/rnpgbe/rnpgbe.h @@ -4,13 +4,28 @@ #ifndef _RNPGBE_H #define _RNPGBE_H +#include + enum rnpgbe_boards { board_n500, board_n210 }; +struct mucse_mbx_info { + u32 timeout_us; + u32 delay_us; + u16 fw_req; + u16 fw_ack; + /* fw <--> pf mbx */ + u32 fwpf_shm_base; + u32 pf2fw_mbx_ctrl; + u32 fwpf_mbx_mask; + u32 fwpf_ctrl_base; +}; + struct mucse_hw { void __iomem *hw_addr; + struct mucse_mbx_info mbx; }; struct mucse { @@ -19,6 +34,8 @@ struct mucse { struct mucse_hw hw; }; +int rnpgbe_init_hw(struct mucse_hw *hw, int board_type); + /* Device IDs */ #define PCI_VENDOR_ID_MUCSE 0x8848 #define RNPGBE_DEVICE_ID_N500_QUAD_PORT 0x8308 diff --git a/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_chip.c b/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_chip.c new file mode 100644 index 000000000000..5739db98f12a --- /dev/null +++ b/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_chip.c @@ -0,0 +1,70 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright(c) 2020 - 2025 Mucse Corporation. */ + +#include + +#include "rnpgbe.h" +#include "rnpgbe_hw.h" +#include "rnpgbe_mbx.h" + +/** + * rnpgbe_init_n500 - Setup n500 hw info + * @hw: hw information structure + * + * rnpgbe_init_n500 initializes all private + * structure for n500 + **/ +static void rnpgbe_init_n500(struct mucse_hw *hw) +{ + struct mucse_mbx_info *mbx = &hw->mbx; + + mbx->fwpf_ctrl_base = MUCSE_N500_FWPF_CTRL_BASE; + mbx->fwpf_shm_base = MUCSE_N500_FWPF_SHM_BASE; +} + +/** + * rnpgbe_init_n210 - Setup n210 hw info + * @hw: hw information structure + * + * rnpgbe_init_n210 initializes all private + * structure for n210 + **/ +static void rnpgbe_init_n210(struct mucse_hw *hw) +{ + struct mucse_mbx_info *mbx = &hw->mbx; + + mbx->fwpf_ctrl_base = MUCSE_N210_FWPF_CTRL_BASE; + mbx->fwpf_shm_base = MUCSE_N210_FWPF_SHM_BASE; +} + +/** + * rnpgbe_init_hw - Setup hw info according to board_type + * @hw: hw information structure + * @board_type: board type + * + * rnpgbe_init_hw initializes all hw data + * + * Return: 0 on success, -EINVAL on failure + **/ +int rnpgbe_init_hw(struct mucse_hw *hw, int board_type) +{ + struct mucse_mbx_info *mbx = &hw->mbx; + + mbx->pf2fw_mbx_ctrl = MUCSE_GBE_PFFW_MBX_CTRL_OFFSET; + mbx->fwpf_mbx_mask = MUCSE_GBE_FWPF_MBX_MASK_OFFSET; + + switch (board_type) { + case board_n500: + rnpgbe_init_n500(hw); + break; + case board_n210: + rnpgbe_init_n210(hw); + break; + default: + return -EINVAL; + } + /* init_params with mbx base */ + mucse_init_mbx_params_pf(hw); + + return 0; +} diff --git a/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_hw.h b/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_hw.h index 3a779806e8be..268f572936aa 100644 --- a/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_hw.h +++ b/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_hw.h @@ -4,5 +4,12 @@ #ifndef _RNPGBE_HW_H #define _RNPGBE_HW_H +#define MUCSE_N500_FWPF_CTRL_BASE 0x28b00 +#define MUCSE_N500_FWPF_SHM_BASE 0x2d000 +#define MUCSE_GBE_PFFW_MBX_CTRL_OFFSET 0x5500 +#define MUCSE_GBE_FWPF_MBX_MASK_OFFSET 0x5700 +#define MUCSE_N210_FWPF_CTRL_BASE 0x29400 +#define MUCSE_N210_FWPF_SHM_BASE 0x2d900 + #define RNPGBE_MAX_QUEUES 8 #endif /* _RNPGBE_HW_H */ diff --git a/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_main.c b/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_main.c index 305657d73e25..d8aaac79ff4b 100644 --- a/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_main.c +++ b/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_main.c @@ -64,6 +64,11 @@ static int rnpgbe_add_adapter(struct pci_dev *pdev, } hw->hw_addr = hw_addr; + err = rnpgbe_init_hw(hw, board_type); + if (err) { + dev_err(&pdev->dev, "Init hw err %d\n", err); + goto err_free_net; + } return 0; diff --git a/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_mbx.c b/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_mbx.c new file mode 100644 index 000000000000..5de4b104455e --- /dev/null +++ b/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_mbx.c @@ -0,0 +1,405 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright(c) 2022 - 2025 Mucse Corporation. */ + +#include +#include +#include + +#include "rnpgbe_mbx.h" + +/** + * mbx_data_rd32 - Reads reg with base mbx->fwpf_shm_base + * @mbx: pointer to the MBX structure + * @reg: register offset + * + * Return: register value + **/ +static u32 mbx_data_rd32(struct mucse_mbx_info *mbx, u32 reg) +{ + struct mucse_hw *hw = container_of(mbx, struct mucse_hw, mbx); + + return readl(hw->hw_addr + mbx->fwpf_shm_base + reg); +} + +/** + * mbx_data_wr32 - Writes value to reg with base mbx->fwpf_shm_base + * @mbx: pointer to the MBX structure + * @reg: register offset + * @value: value to be written + * + **/ +static void mbx_data_wr32(struct mucse_mbx_info *mbx, u32 reg, u32 value) +{ + struct mucse_hw *hw = container_of(mbx, struct mucse_hw, mbx); + + writel(value, hw->hw_addr + mbx->fwpf_shm_base + reg); +} + +/** + * mbx_ctrl_rd32 - Reads reg with base mbx->fwpf_ctrl_base + * @mbx: pointer to the MBX structure + * @reg: register offset + * + * Return: register value + **/ +static u32 mbx_ctrl_rd32(struct mucse_mbx_info *mbx, u32 reg) +{ + struct mucse_hw *hw = container_of(mbx, struct mucse_hw, mbx); + + return readl(hw->hw_addr + mbx->fwpf_ctrl_base + reg); +} + +/** + * mbx_ctrl_wr32 - Writes value to reg with base mbx->fwpf_ctrl_base + * @mbx: pointer to the MBX structure + * @reg: register offset + * @value: value to be written + * + **/ +static void mbx_ctrl_wr32(struct mucse_mbx_info *mbx, u32 reg, u32 value) +{ + struct mucse_hw *hw = container_of(mbx, struct mucse_hw, mbx); + + writel(value, hw->hw_addr + mbx->fwpf_ctrl_base + reg); +} + +/** + * mucse_mbx_get_lock_pf - Write ctrl and read back lock status + * @hw: pointer to the HW structure + * + * Return: register value after write + **/ +static u32 mucse_mbx_get_lock_pf(struct mucse_hw *hw) +{ + struct mucse_mbx_info *mbx = &hw->mbx; + u32 reg = MUCSE_MBX_PF2FW_CTRL(mbx); + + mbx_ctrl_wr32(mbx, reg, MUCSE_MBX_PFU); + + return mbx_ctrl_rd32(mbx, reg); +} + +/** + * mucse_obtain_mbx_lock_pf - Obtain mailbox lock + * @hw: pointer to the HW structure + * + * Pair with mucse_release_mbx_lock_pf() + * This function maybe used in an irq handler. + * + * Return: 0 on success, negative errno on failure + **/ +static int mucse_obtain_mbx_lock_pf(struct mucse_hw *hw) +{ + struct mucse_mbx_info *mbx = &hw->mbx; + u32 val; + + return read_poll_timeout_atomic(mucse_mbx_get_lock_pf, + val, val & MUCSE_MBX_PFU, + mbx->delay_us, + mbx->timeout_us, + false, hw); +} + +/** + * mucse_release_mbx_lock_pf - Release mailbox lock + * @hw: pointer to the HW structure + * @req: send a request or not + * + * Pair with mucse_obtain_mbx_lock_pf(): + * - Releases the mailbox lock by clearing MUCSE_MBX_PFU bit + * - Simultaneously sends the request by setting MUCSE_MBX_REQ bit + * if req is true + * (Both bits are in the same mailbox control register, + * so operations are combined) + **/ +static void mucse_release_mbx_lock_pf(struct mucse_hw *hw, bool req) +{ + struct mucse_mbx_info *mbx = &hw->mbx; + u32 reg = MUCSE_MBX_PF2FW_CTRL(mbx); + + mbx_ctrl_wr32(mbx, reg, req ? MUCSE_MBX_REQ : 0); +} + +/** + * mucse_mbx_get_fwreq - Read fw req from reg + * @mbx: pointer to the mbx structure + * + * Return: the fwreq value + **/ +static u16 mucse_mbx_get_fwreq(struct mucse_mbx_info *mbx) +{ + u32 val = mbx_data_rd32(mbx, MUCSE_MBX_FW2PF_CNT); + + return FIELD_GET(GENMASK_U32(15, 0), val); +} + +/** + * mucse_mbx_inc_pf_ack - Increase ack + * @hw: pointer to the HW structure + * + * mucse_mbx_inc_pf_ack reads pf_ack from hw, then writes + * new value back after increase + **/ +static void mucse_mbx_inc_pf_ack(struct mucse_hw *hw) +{ + struct mucse_mbx_info *mbx = &hw->mbx; + u16 ack; + u32 val; + + val = mbx_data_rd32(mbx, MUCSE_MBX_PF2FW_CNT); + ack = FIELD_GET(GENMASK_U32(31, 16), val); + ack++; + val &= ~GENMASK_U32(31, 16); + val |= FIELD_PREP(GENMASK_U32(31, 16), ack); + mbx_data_wr32(mbx, MUCSE_MBX_PF2FW_CNT, val); +} + +/** + * mucse_read_mbx_pf - Read a message from the mailbox + * @hw: pointer to the HW structure + * @msg: the message buffer + * @size: length of buffer + * + * mucse_read_mbx_pf copies a message from the mbx buffer to the caller's + * memory buffer. The presumption is that the caller knows that there was + * a message due to a fw request so no polling for message is needed. + * + * Return: 0 on success, negative errno on failure + **/ +static int mucse_read_mbx_pf(struct mucse_hw *hw, u32 *msg, u16 size) +{ + const int size_in_words = size / sizeof(u32); + struct mucse_mbx_info *mbx = &hw->mbx; + int err; + + err = mucse_obtain_mbx_lock_pf(hw); + if (err) + return err; + + for (int i = 0; i < size_in_words; i++) + msg[i] = mbx_data_rd32(mbx, MUCSE_MBX_FWPF_SHM + 4 * i); + /* Hw needs write data_reg at last */ + mbx_data_wr32(mbx, MUCSE_MBX_FWPF_SHM, 0); + /* flush reqs as we have read this request data */ + hw->mbx.fw_req = mucse_mbx_get_fwreq(mbx); + mucse_mbx_inc_pf_ack(hw); + mucse_release_mbx_lock_pf(hw, false); + + return 0; +} + +/** + * mucse_check_for_msg_pf - Check to see if the fw has sent mail + * @hw: pointer to the HW structure + * + * Return: 0 if the fw has set the Status bit or else -EIO + **/ +static int mucse_check_for_msg_pf(struct mucse_hw *hw) +{ + struct mucse_mbx_info *mbx = &hw->mbx; + u16 fw_req; + + fw_req = mucse_mbx_get_fwreq(mbx); + /* chip's register is reset to 0 when rc send reset + * mbx command. Return -EIO if in this state, others + * fw == hw->mbx.fw_req means no new msg. + **/ + if (fw_req == 0 || fw_req == hw->mbx.fw_req) + return -EIO; + + return 0; +} + +/** + * mucse_poll_for_msg - Wait for message notification + * @hw: pointer to the HW structure + * + * Return: 0 on success, negative errno on failure + **/ +static int mucse_poll_for_msg(struct mucse_hw *hw) +{ + struct mucse_mbx_info *mbx = &hw->mbx; + int val; + + return read_poll_timeout(mucse_check_for_msg_pf, + val, !val, mbx->delay_us, + mbx->timeout_us, + false, hw); +} + +/** + * mucse_poll_and_read_mbx - Wait for message notification and receive message + * @hw: pointer to the HW structure + * @msg: the message buffer + * @size: length of buffer + * + * Return: 0 if it successfully received a message notification and + * copied it into the receive buffer, negative errno on failure + **/ +int mucse_poll_and_read_mbx(struct mucse_hw *hw, u32 *msg, u16 size) +{ + int err; + + err = mucse_poll_for_msg(hw); + if (err) + return err; + + return mucse_read_mbx_pf(hw, msg, size); +} + +/** + * mucse_mbx_get_fwack - Read fw ack from reg + * @mbx: pointer to the MBX structure + * + * Return: the fwack value + **/ +static u16 mucse_mbx_get_fwack(struct mucse_mbx_info *mbx) +{ + u32 val = mbx_data_rd32(mbx, MUCSE_MBX_FW2PF_CNT); + + return FIELD_GET(GENMASK_U32(31, 16), val); +} + +/** + * mucse_mbx_inc_pf_req - Increase req + * @hw: pointer to the HW structure + * + * mucse_mbx_inc_pf_req reads pf_req from hw, then writes + * new value back after increase + **/ +static void mucse_mbx_inc_pf_req(struct mucse_hw *hw) +{ + struct mucse_mbx_info *mbx = &hw->mbx; + u16 req; + u32 val; + + val = mbx_data_rd32(mbx, MUCSE_MBX_PF2FW_CNT); + req = FIELD_GET(GENMASK_U32(15, 0), val); + req++; + val &= ~GENMASK_U32(15, 0); + val |= FIELD_PREP(GENMASK_U32(15, 0), req); + mbx_data_wr32(mbx, MUCSE_MBX_PF2FW_CNT, val); +} + +/** + * mucse_write_mbx_pf - Place a message in the mailbox + * @hw: pointer to the HW structure + * @msg: the message buffer + * @size: length of buffer + * + * Return: 0 if it successfully copied message into the buffer, + * negative errno on failure + **/ +static int mucse_write_mbx_pf(struct mucse_hw *hw, u32 *msg, u16 size) +{ + const int size_in_words = size / sizeof(u32); + struct mucse_mbx_info *mbx = &hw->mbx; + int err; + + err = mucse_obtain_mbx_lock_pf(hw); + if (err) + return err; + + for (int i = 0; i < size_in_words; i++) + mbx_data_wr32(mbx, MUCSE_MBX_FWPF_SHM + i * 4, msg[i]); + + /* flush acks as we are overwriting the message buffer */ + hw->mbx.fw_ack = mucse_mbx_get_fwack(mbx); + mucse_mbx_inc_pf_req(hw); + mucse_release_mbx_lock_pf(hw, true); + + return 0; +} + +/** + * mucse_check_for_ack_pf - Check to see if the fw has ACKed + * @hw: pointer to the HW structure + * + * Return: 0 if the fw has set the Status bit or else -EIO + **/ +static int mucse_check_for_ack_pf(struct mucse_hw *hw) +{ + struct mucse_mbx_info *mbx = &hw->mbx; + u16 fw_ack; + + fw_ack = mucse_mbx_get_fwack(mbx); + /* chip's register is reset to 0 when rc send reset + * mbx command. Return -EIO if in this state, others + * fw_ack == hw->mbx.fw_ack means no new ack. + **/ + if (fw_ack == 0 || fw_ack == hw->mbx.fw_ack) + return -EIO; + + return 0; +} + +/** + * mucse_poll_for_ack - Wait for message acknowledgment + * @hw: pointer to the HW structure + * + * Return: 0 if it successfully received a message acknowledgment, + * else negative errno + **/ +static int mucse_poll_for_ack(struct mucse_hw *hw) +{ + struct mucse_mbx_info *mbx = &hw->mbx; + int val; + + return read_poll_timeout(mucse_check_for_ack_pf, + val, !val, mbx->delay_us, + mbx->timeout_us, + false, hw); +} + +/** + * mucse_write_and_wait_ack_mbx - Write a message to the mailbox, wait for ack + * @hw: pointer to the HW structure + * @msg: the message buffer + * @size: length of buffer + * + * Return: 0 if it successfully copied message into the buffer and + * received an ack to that message within delay * timeout_cnt period + **/ +int mucse_write_and_wait_ack_mbx(struct mucse_hw *hw, u32 *msg, u16 size) +{ + int err; + + err = mucse_write_mbx_pf(hw, msg, size); + if (err) + return err; + + return mucse_poll_for_ack(hw); +} + +/** + * mucse_mbx_reset - Reset mbx info, sync info from regs + * @hw: pointer to the HW structure + * + * mucse_mbx_reset resets all mbx variables to default. + **/ +static void mucse_mbx_reset(struct mucse_hw *hw) +{ + struct mucse_mbx_info *mbx = &hw->mbx; + u32 val; + + val = mbx_data_rd32(mbx, MUCSE_MBX_FW2PF_CNT); + hw->mbx.fw_req = FIELD_GET(GENMASK_U32(15, 0), val); + hw->mbx.fw_ack = FIELD_GET(GENMASK_U32(31, 16), val); + mbx_ctrl_wr32(mbx, MUCSE_MBX_PF2FW_CTRL(mbx), 0); + mbx_ctrl_wr32(mbx, MUCSE_MBX_FWPF_MASK(mbx), GENMASK_U32(31, 16)); +} + +/** + * mucse_init_mbx_params_pf - Set initial values for pf mailbox + * @hw: pointer to the HW structure + * + * Initializes the hw->mbx struct to correct values for pf mailbox + */ +void mucse_init_mbx_params_pf(struct mucse_hw *hw) +{ + struct mucse_mbx_info *mbx = &hw->mbx; + + mbx->delay_us = 100; + mbx->timeout_us = 4 * USEC_PER_SEC; + mucse_mbx_reset(hw); +} diff --git a/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_mbx.h b/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_mbx.h new file mode 100644 index 000000000000..e6fcc8d1d3ca --- /dev/null +++ b/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_mbx.h @@ -0,0 +1,20 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright(c) 2020 - 2025 Mucse Corporation. */ + +#ifndef _RNPGBE_MBX_H +#define _RNPGBE_MBX_H + +#include "rnpgbe.h" + +#define MUCSE_MBX_FW2PF_CNT 0 +#define MUCSE_MBX_PF2FW_CNT 4 +#define MUCSE_MBX_FWPF_SHM 8 +#define MUCSE_MBX_PF2FW_CTRL(mbx) ((mbx)->pf2fw_mbx_ctrl) +#define MUCSE_MBX_FWPF_MASK(mbx) ((mbx)->fwpf_mbx_mask) +#define MUCSE_MBX_REQ BIT(0) /* Request a req to mailbox */ +#define MUCSE_MBX_PFU BIT(3) /* PF owns the mailbox buffer */ + +int mucse_write_and_wait_ack_mbx(struct mucse_hw *hw, u32 *msg, u16 size); +void mucse_init_mbx_params_pf(struct mucse_hw *hw); +int mucse_poll_and_read_mbx(struct mucse_hw *hw, u32 *msg, u16 size); +#endif /* _RNPGBE_MBX_H */ From c6d3f0198eaa4efba506933d2dbcd417e987630a Mon Sep 17 00:00:00 2001 From: Dong Yibo Date: Sat, 1 Nov 2025 09:38:48 +0800 Subject: [PATCH 0519/1381] net: rnpgbe: Add basic mbx_fw support Add fundamental firmware (FW) communication operations via PF-FW mailbox, including: - FW sync (via HW info query with retries) - HW reset (post FW command to reset hardware) - MAC address retrieval (request FW for port-specific MAC) - Power management (powerup/powerdown notification to FW) Signed-off-by: Dong Yibo Reviewed-by: Vadim Fedorenko Link: https://patch.msgid.link/20251101013849.120565-5-dong100@mucse.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mucse/rnpgbe/Makefile | 3 +- drivers/net/ethernet/mucse/rnpgbe/rnpgbe.h | 4 + .../net/ethernet/mucse/rnpgbe/rnpgbe_mbx.c | 1 + .../net/ethernet/mucse/rnpgbe/rnpgbe_mbx_fw.c | 191 ++++++++++++++++++ .../net/ethernet/mucse/rnpgbe/rnpgbe_mbx_fw.h | 88 ++++++++ 5 files changed, 286 insertions(+), 1 deletion(-) create mode 100644 drivers/net/ethernet/mucse/rnpgbe/rnpgbe_mbx_fw.c create mode 100644 drivers/net/ethernet/mucse/rnpgbe/rnpgbe_mbx_fw.h diff --git a/drivers/net/ethernet/mucse/rnpgbe/Makefile b/drivers/net/ethernet/mucse/rnpgbe/Makefile index 5fc878ada4b1..de8bcb7772ab 100644 --- a/drivers/net/ethernet/mucse/rnpgbe/Makefile +++ b/drivers/net/ethernet/mucse/rnpgbe/Makefile @@ -7,4 +7,5 @@ obj-$(CONFIG_MGBE) += rnpgbe.o rnpgbe-objs := rnpgbe_main.o\ rnpgbe_chip.o\ - rnpgbe_mbx.o + rnpgbe_mbx.o\ + rnpgbe_mbx_fw.o diff --git a/drivers/net/ethernet/mucse/rnpgbe/rnpgbe.h b/drivers/net/ethernet/mucse/rnpgbe/rnpgbe.h index 4c70b0cedd1f..37bd9278beaa 100644 --- a/drivers/net/ethernet/mucse/rnpgbe/rnpgbe.h +++ b/drivers/net/ethernet/mucse/rnpgbe/rnpgbe.h @@ -5,6 +5,7 @@ #define _RNPGBE_H #include +#include enum rnpgbe_boards { board_n500, @@ -16,6 +17,8 @@ struct mucse_mbx_info { u32 delay_us; u16 fw_req; u16 fw_ack; + /* lock for only one use mbx */ + struct mutex lock; /* fw <--> pf mbx */ u32 fwpf_shm_base; u32 pf2fw_mbx_ctrl; @@ -26,6 +29,7 @@ struct mucse_mbx_info { struct mucse_hw { void __iomem *hw_addr; struct mucse_mbx_info mbx; + u8 pfvfnum; }; struct mucse { diff --git a/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_mbx.c b/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_mbx.c index 5de4b104455e..de5e29230b3c 100644 --- a/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_mbx.c +++ b/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_mbx.c @@ -401,5 +401,6 @@ void mucse_init_mbx_params_pf(struct mucse_hw *hw) mbx->delay_us = 100; mbx->timeout_us = 4 * USEC_PER_SEC; + mutex_init(&mbx->lock); mucse_mbx_reset(hw); } diff --git a/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_mbx_fw.c b/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_mbx_fw.c new file mode 100644 index 000000000000..8c8bd5e8e1db --- /dev/null +++ b/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_mbx_fw.c @@ -0,0 +1,191 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright(c) 2020 - 2025 Mucse Corporation. */ + +#include +#include + +#include "rnpgbe.h" +#include "rnpgbe_mbx.h" +#include "rnpgbe_mbx_fw.h" + +/** + * mucse_fw_send_cmd_wait_resp - Send cmd req and wait for response + * @hw: pointer to the HW structure + * @req: pointer to the cmd req structure + * @reply: pointer to the fw reply structure + * + * mucse_fw_send_cmd_wait_resp sends req to pf-fw mailbox and wait + * reply from fw. + * + * Return: 0 on success, negative errno on failure + **/ +static int mucse_fw_send_cmd_wait_resp(struct mucse_hw *hw, + struct mbx_fw_cmd_req *req, + struct mbx_fw_cmd_reply *reply) +{ + int len = le16_to_cpu(req->datalen); + int retry_cnt = 3; + int err; + + mutex_lock(&hw->mbx.lock); + err = mucse_write_and_wait_ack_mbx(hw, (u32 *)req, len); + if (err) + goto out; + do { + err = mucse_poll_and_read_mbx(hw, (u32 *)reply, + sizeof(*reply)); + if (err) + goto out; + /* mucse_write_and_wait_ack_mbx return 0 means fw has + * received request, wait for the expect opcode + * reply with 'retry_cnt' times. + */ + } while (--retry_cnt >= 0 && reply->opcode != req->opcode); +out: + mutex_unlock(&hw->mbx.lock); + if (!err && retry_cnt < 0) + return -ETIMEDOUT; + if (!err && reply->error_code) + return -EIO; + + return err; +} + +/** + * mucse_mbx_get_info - Get hw info from fw + * @hw: pointer to the HW structure + * + * mucse_mbx_get_info tries to get hw info from hw. + * + * Return: 0 on success, negative errno on failure + **/ +static int mucse_mbx_get_info(struct mucse_hw *hw) +{ + struct mbx_fw_cmd_req req = { + .datalen = cpu_to_le16(MUCSE_MBX_REQ_HDR_LEN), + .opcode = cpu_to_le16(GET_HW_INFO), + }; + struct mbx_fw_cmd_reply reply = {}; + int err; + + err = mucse_fw_send_cmd_wait_resp(hw, &req, &reply); + if (!err) + hw->pfvfnum = FIELD_GET(GENMASK_U16(7, 0), + le16_to_cpu(reply.hw_info.pfnum)); + + return err; +} + +/** + * mucse_mbx_sync_fw - Try to sync with fw + * @hw: pointer to the HW structure + * + * mucse_mbx_sync_fw tries to sync with fw. It is only called in + * probe. Nothing (register network) todo if failed. + * Try more times to do sync. + * + * Return: 0 on success, negative errno on failure + **/ +int mucse_mbx_sync_fw(struct mucse_hw *hw) +{ + int try_cnt = 3; + int err; + + do { + err = mucse_mbx_get_info(hw); + } while (err == -ETIMEDOUT && try_cnt--); + + return err; +} + +/** + * mucse_mbx_powerup - Echo fw to powerup + * @hw: pointer to the HW structure + * @is_powerup: true for powerup, false for powerdown + * + * mucse_mbx_powerup echo fw to change working frequency + * to normal after received true, and reduce working frequency + * if false. + * + * Return: 0 on success, negative errno on failure + **/ +int mucse_mbx_powerup(struct mucse_hw *hw, bool is_powerup) +{ + struct mbx_fw_cmd_req req = { + .datalen = cpu_to_le16(sizeof(req.powerup) + + MUCSE_MBX_REQ_HDR_LEN), + .opcode = cpu_to_le16(POWER_UP), + .powerup = { + /* fw needs this to reply correct cmd */ + .version = cpu_to_le32(GENMASK_U32(31, 0)), + .status = cpu_to_le32(is_powerup ? 1 : 0), + }, + }; + int len, err; + + len = le16_to_cpu(req.datalen); + mutex_lock(&hw->mbx.lock); + err = mucse_write_and_wait_ack_mbx(hw, (u32 *)&req, len); + mutex_unlock(&hw->mbx.lock); + + return err; +} + +/** + * mucse_mbx_reset_hw - Posts a mbx req to reset hw + * @hw: pointer to the HW structure + * + * mucse_mbx_reset_hw posts a mbx req to firmware to reset hw. + * We use mucse_fw_send_cmd_wait_resp to wait hw reset ok. + * + * Return: 0 on success, negative errno on failure + **/ +int mucse_mbx_reset_hw(struct mucse_hw *hw) +{ + struct mbx_fw_cmd_req req = { + .datalen = cpu_to_le16(MUCSE_MBX_REQ_HDR_LEN), + .opcode = cpu_to_le16(RESET_HW), + }; + struct mbx_fw_cmd_reply reply = {}; + + return mucse_fw_send_cmd_wait_resp(hw, &req, &reply); +} + +/** + * mucse_mbx_get_macaddr - Posts a mbx req to request macaddr + * @hw: pointer to the HW structure + * @pfvfnum: index of pf/vf num + * @mac_addr: pointer to store mac_addr + * @port: port index + * + * mucse_mbx_get_macaddr posts a mbx req to firmware to get mac_addr. + * + * Return: 0 on success, negative errno on failure + **/ +int mucse_mbx_get_macaddr(struct mucse_hw *hw, int pfvfnum, + u8 *mac_addr, + int port) +{ + struct mbx_fw_cmd_req req = { + .datalen = cpu_to_le16(sizeof(req.get_mac_addr) + + MUCSE_MBX_REQ_HDR_LEN), + .opcode = cpu_to_le16(GET_MAC_ADDRESS), + .get_mac_addr = { + .port_mask = cpu_to_le32(BIT(port)), + .pfvf_num = cpu_to_le32(pfvfnum), + }, + }; + struct mbx_fw_cmd_reply reply = {}; + int err; + + err = mucse_fw_send_cmd_wait_resp(hw, &req, &reply); + if (err) + return err; + + if (le32_to_cpu(reply.mac_addr.ports) & BIT(port)) + memcpy(mac_addr, reply.mac_addr.addrs[port].mac, ETH_ALEN); + else + return -ENODATA; + + return 0; +} diff --git a/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_mbx_fw.h b/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_mbx_fw.h new file mode 100644 index 000000000000..fb24fc12b613 --- /dev/null +++ b/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_mbx_fw.h @@ -0,0 +1,88 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright(c) 2020 - 2025 Mucse Corporation. */ + +#ifndef _RNPGBE_MBX_FW_H +#define _RNPGBE_MBX_FW_H + +#include + +#include "rnpgbe.h" + +#define MUCSE_MBX_REQ_HDR_LEN 24 + +enum MUCSE_FW_CMD { + GET_HW_INFO = 0x0601, + GET_MAC_ADDRESS = 0x0602, + RESET_HW = 0x0603, + POWER_UP = 0x0803, +}; + +struct mucse_hw_info { + u8 link_stat; + u8 port_mask; + __le32 speed; + __le16 phy_type; + __le16 nic_mode; + __le16 pfnum; + __le32 fw_version; + __le32 axi_mhz; + union { + u8 port_id[4]; + __le32 port_ids; + }; + __le32 bd_uid; + __le32 phy_id; + __le32 wol_status; + __le32 ext_info; +} __packed; + +struct mbx_fw_cmd_req { + __le16 flags; + __le16 opcode; + __le16 datalen; + __le16 ret_value; + __le32 cookie_lo; + __le32 cookie_hi; + __le32 reply_lo; + __le32 reply_hi; + union { + u8 data[32]; + struct { + __le32 version; + __le32 status; + } powerup; + struct { + __le32 port_mask; + __le32 pfvf_num; + } get_mac_addr; + }; +} __packed; + +struct mbx_fw_cmd_reply { + __le16 flags; + __le16 opcode; + __le16 error_code; + __le16 datalen; + __le32 cookie_lo; + __le32 cookie_hi; + union { + u8 data[40]; + struct mac_addr { + __le32 ports; + struct _addr { + /* for macaddr:01:02:03:04:05:06 + * mac-hi=0x01020304 mac-lo=0x05060000 + */ + u8 mac[8]; + } addrs[4]; + } mac_addr; + struct mucse_hw_info hw_info; + }; +} __packed; + +int mucse_mbx_sync_fw(struct mucse_hw *hw); +int mucse_mbx_powerup(struct mucse_hw *hw, bool is_powerup); +int mucse_mbx_reset_hw(struct mucse_hw *hw); +int mucse_mbx_get_macaddr(struct mucse_hw *hw, int pfvfnum, + u8 *mac_addr, int port); +#endif /* _RNPGBE_MBX_FW_H */ From 2ee95ec17e97c58b65e978a08b75fa8cb6424e4e Mon Sep 17 00:00:00 2001 From: Dong Yibo Date: Sat, 1 Nov 2025 09:38:49 +0800 Subject: [PATCH 0520/1381] net: rnpgbe: Add register_netdev Complete the network device (netdev) registration flow for Mucse Gbe Ethernet chips, including: 1. Hardware state initialization: - Send powerup notification to firmware (via echo_fw_status) - Sync with firmware - Reset hardware 2. MAC address handling: - Retrieve permanent MAC from firmware (via mucse_mbx_get_macaddr) - Fallback to random valid MAC (eth_random_addr) if not valid mac from Fw Signed-off-by: Dong Yibo Reviewed-by: Vadim Fedorenko Link: https://patch.msgid.link/20251101013849.120565-6-dong100@mucse.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mucse/rnpgbe/rnpgbe.h | 22 ++++ .../net/ethernet/mucse/rnpgbe/rnpgbe_chip.c | 73 +++++++++++ drivers/net/ethernet/mucse/rnpgbe/rnpgbe_hw.h | 2 + .../net/ethernet/mucse/rnpgbe/rnpgbe_main.c | 119 +++++++++++++++++- 4 files changed, 214 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/mucse/rnpgbe/rnpgbe.h b/drivers/net/ethernet/mucse/rnpgbe/rnpgbe.h index 37bd9278beaa..5b024f9f7e17 100644 --- a/drivers/net/ethernet/mucse/rnpgbe/rnpgbe.h +++ b/drivers/net/ethernet/mucse/rnpgbe/rnpgbe.h @@ -26,18 +26,37 @@ struct mucse_mbx_info { u32 fwpf_ctrl_base; }; +/* Enum for firmware notification modes, + * more modes (e.g., portup, link_report) will be added in future + **/ +enum { + mucse_fw_powerup, +}; + struct mucse_hw { void __iomem *hw_addr; + struct pci_dev *pdev; struct mucse_mbx_info mbx; + int port; u8 pfvfnum; }; +struct mucse_stats { + u64 tx_dropped; +}; + struct mucse { struct net_device *netdev; struct pci_dev *pdev; struct mucse_hw hw; + struct mucse_stats stats; }; +int rnpgbe_get_permanent_mac(struct mucse_hw *hw, u8 *perm_addr); +int rnpgbe_reset_hw(struct mucse_hw *hw); +int rnpgbe_send_notify(struct mucse_hw *hw, + bool enable, + int mode); int rnpgbe_init_hw(struct mucse_hw *hw, int board_type); /* Device IDs */ @@ -46,4 +65,7 @@ int rnpgbe_init_hw(struct mucse_hw *hw, int board_type); #define RNPGBE_DEVICE_ID_N500_DUAL_PORT 0x8318 #define RNPGBE_DEVICE_ID_N210 0x8208 #define RNPGBE_DEVICE_ID_N210L 0x820a + +#define mucse_hw_wr32(hw, reg, val) \ + writel((val), (hw)->hw_addr + (reg)) #endif /* _RNPGBE_H */ diff --git a/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_chip.c b/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_chip.c index 5739db98f12a..ebc7b3750157 100644 --- a/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_chip.c +++ b/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_chip.c @@ -1,11 +1,82 @@ // SPDX-License-Identifier: GPL-2.0 /* Copyright(c) 2020 - 2025 Mucse Corporation. */ +#include #include +#include #include "rnpgbe.h" #include "rnpgbe_hw.h" #include "rnpgbe_mbx.h" +#include "rnpgbe_mbx_fw.h" + +/** + * rnpgbe_get_permanent_mac - Get permanent mac + * @hw: hw information structure + * @perm_addr: pointer to store perm_addr + * + * rnpgbe_get_permanent_mac tries to get mac from hw + * + * Return: 0 on success, negative errno on failure + **/ +int rnpgbe_get_permanent_mac(struct mucse_hw *hw, u8 *perm_addr) +{ + struct device *dev = &hw->pdev->dev; + int err; + + err = mucse_mbx_get_macaddr(hw, hw->pfvfnum, perm_addr, hw->port); + if (err) { + dev_err(dev, "Failed to get MAC from FW %d\n", err); + return err; + } + + if (!is_valid_ether_addr(perm_addr)) { + dev_err(dev, "Failed to get valid MAC from FW\n"); + return -EINVAL; + } + + return 0; +} + +/** + * rnpgbe_reset_hw - Do a hardware reset + * @hw: hw information structure + * + * rnpgbe_reset_hw calls fw to do a hardware + * reset, and cleans some regs to default. + * + * Return: 0 on success, negative errno on failure + **/ +int rnpgbe_reset_hw(struct mucse_hw *hw) +{ + mucse_hw_wr32(hw, RNPGBE_DMA_AXI_EN, 0); + return mucse_mbx_reset_hw(hw); +} + +/** + * rnpgbe_send_notify - Echo fw status + * @hw: hw information structure + * @enable: true or false status + * @mode: status mode + * + * Return: 0 on success, negative errno on failure + **/ +int rnpgbe_send_notify(struct mucse_hw *hw, + bool enable, + int mode) +{ + int err; + /* Keep switch struct to support more modes in the future */ + switch (mode) { + case mucse_fw_powerup: + err = mucse_mbx_powerup(hw, enable); + break; + default: + err = -EINVAL; + } + + return err; +} /** * rnpgbe_init_n500 - Setup n500 hw info @@ -50,6 +121,8 @@ int rnpgbe_init_hw(struct mucse_hw *hw, int board_type) { struct mucse_mbx_info *mbx = &hw->mbx; + hw->port = 0; + mbx->pf2fw_mbx_ctrl = MUCSE_GBE_PFFW_MBX_CTRL_OFFSET; mbx->fwpf_mbx_mask = MUCSE_GBE_FWPF_MBX_MASK_OFFSET; diff --git a/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_hw.h b/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_hw.h index 268f572936aa..e77e6bc3d3e3 100644 --- a/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_hw.h +++ b/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_hw.h @@ -11,5 +11,7 @@ #define MUCSE_N210_FWPF_CTRL_BASE 0x29400 #define MUCSE_N210_FWPF_SHM_BASE 0x2d900 +#define RNPGBE_DMA_AXI_EN 0x0010 + #define RNPGBE_MAX_QUEUES 8 #endif /* _RNPGBE_HW_H */ diff --git a/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_main.c b/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_main.c index d8aaac79ff4b..316f941629d4 100644 --- a/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_main.c +++ b/drivers/net/ethernet/mucse/rnpgbe/rnpgbe_main.c @@ -7,6 +7,7 @@ #include "rnpgbe.h" #include "rnpgbe_hw.h" +#include "rnpgbe_mbx_fw.h" static const char rnpgbe_driver_name[] = "rnpgbe"; @@ -24,6 +25,58 @@ static struct pci_device_id rnpgbe_pci_tbl[] = { {0, }, }; +/** + * rnpgbe_open - Called when a network interface is made active + * @netdev: network interface device structure + * + * The open entry point is called when a network interface is made + * active by the system (IFF_UP). + * + * Return: 0 + **/ +static int rnpgbe_open(struct net_device *netdev) +{ + return 0; +} + +/** + * rnpgbe_close - Disables a network interface + * @netdev: network interface device structure + * + * The close entry point is called when an interface is de-activated + * by the OS. + * + * Return: 0, this is not allowed to fail + **/ +static int rnpgbe_close(struct net_device *netdev) +{ + return 0; +} + +/** + * rnpgbe_xmit_frame - Send a skb to driver + * @skb: skb structure to be sent + * @netdev: network interface device structure + * + * Return: NETDEV_TX_OK + **/ +static netdev_tx_t rnpgbe_xmit_frame(struct sk_buff *skb, + struct net_device *netdev) +{ + struct mucse *mucse = netdev_priv(netdev); + + dev_kfree_skb_any(skb); + mucse->stats.tx_dropped++; + + return NETDEV_TX_OK; +} + +static const struct net_device_ops rnpgbe_netdev_ops = { + .ndo_open = rnpgbe_open, + .ndo_stop = rnpgbe_close, + .ndo_start_xmit = rnpgbe_xmit_frame, +}; + /** * rnpgbe_add_adapter - Add netdev for this pci_dev * @pdev: PCI device information structure @@ -39,10 +92,11 @@ static int rnpgbe_add_adapter(struct pci_dev *pdev, int board_type) { struct net_device *netdev; + u8 perm_addr[ETH_ALEN]; void __iomem *hw_addr; struct mucse *mucse; struct mucse_hw *hw; - int err; + int err, err_notify; netdev = alloc_etherdev_mq(sizeof(struct mucse), RNPGBE_MAX_QUEUES); if (!netdev) @@ -64,14 +118,67 @@ static int rnpgbe_add_adapter(struct pci_dev *pdev, } hw->hw_addr = hw_addr; + hw->pdev = pdev; + err = rnpgbe_init_hw(hw, board_type); if (err) { dev_err(&pdev->dev, "Init hw err %d\n", err); goto err_free_net; } + /* Step 1: Send power-up notification to firmware (no response expected) + * This informs firmware to initialize hardware power state, but + * firmware only acknowledges receipt without returning data. Must be + * done before synchronization as firmware may be in low-power idle + * state initially. + */ + err_notify = rnpgbe_send_notify(hw, true, mucse_fw_powerup); + if (err_notify) { + dev_warn(&pdev->dev, "Send powerup to hw failed %d\n", + err_notify); + dev_warn(&pdev->dev, "Maybe low performance\n"); + } + /* Step 2: Synchronize mailbox communication with firmware (requires + * response) After power-up, confirm firmware is ready to process + * requests with responses. This ensures subsequent request/response + * interactions work reliably. + */ + err = mucse_mbx_sync_fw(hw); + if (err) { + dev_err(&pdev->dev, "Sync fw failed! %d\n", err); + goto err_powerdown; + } + + netdev->netdev_ops = &rnpgbe_netdev_ops; + err = rnpgbe_reset_hw(hw); + if (err) { + dev_err(&pdev->dev, "Hw reset failed %d\n", err); + goto err_powerdown; + } + + err = rnpgbe_get_permanent_mac(hw, perm_addr); + if (!err) { + eth_hw_addr_set(netdev, perm_addr); + } else if (err == -EINVAL) { + dev_warn(&pdev->dev, "Using random MAC\n"); + eth_hw_addr_random(netdev); + } else if (err) { + dev_err(&pdev->dev, "get perm_addr failed %d\n", err); + goto err_powerdown; + } + + err = register_netdev(netdev); + if (err) + goto err_powerdown; return 0; - +err_powerdown: + /* notify powerdown only powerup ok */ + if (!err_notify) { + err_notify = rnpgbe_send_notify(hw, false, mucse_fw_powerup); + if (err_notify) + dev_warn(&pdev->dev, "Send powerdown to hw failed %d\n", + err_notify); + } err_free_net: free_netdev(netdev); return err; @@ -138,11 +245,17 @@ static int rnpgbe_probe(struct pci_dev *pdev, const struct pci_device_id *id) static void rnpgbe_rm_adapter(struct pci_dev *pdev) { struct mucse *mucse = pci_get_drvdata(pdev); + struct mucse_hw *hw = &mucse->hw; struct net_device *netdev; + int err; if (!mucse) return; netdev = mucse->netdev; + unregister_netdev(netdev); + err = rnpgbe_send_notify(hw, false, mucse_fw_powerup); + if (err) + dev_warn(&pdev->dev, "Send powerdown to hw failed %d\n", err); free_netdev(netdev); } @@ -173,6 +286,8 @@ static void rnpgbe_dev_shutdown(struct pci_dev *pdev) rtnl_lock(); netif_device_detach(netdev); + if (netif_running(netdev)) + rnpgbe_close(netdev); rtnl_unlock(); pci_disable_device(pdev); } From 617a0dd24ef2b4e6240df48b1fbac1c3ebfa9282 Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Mon, 3 Nov 2025 23:26:49 +0100 Subject: [PATCH 0521/1381] net: phy: make phy_device members pause and asym_pause bitfield bits We can reduce the size of struct phy_device a little by switching the type of members pause and asym_pause from int to a single bit. As C99 is supported now, we can use type bool for the bitfield members, what provides us with the benefit of the usual implicit bool conversions. Signed-off-by: Heiner Kallweit Link: https://patch.msgid.link/764e9a31-b40b-4dc9-b808-118192a16d87@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/phy/phy-c45.c | 20 ++++++++++---------- drivers/net/phy/phy_device.c | 16 ++++++++-------- include/linux/phy.h | 4 ++-- 3 files changed, 20 insertions(+), 20 deletions(-) diff --git a/drivers/net/phy/phy-c45.c b/drivers/net/phy/phy-c45.c index 61670be0f095..1a7b32be4625 100644 --- a/drivers/net/phy/phy-c45.c +++ b/drivers/net/phy/phy-c45.c @@ -485,8 +485,8 @@ static int genphy_c45_baset1_read_lpa(struct phy_device *phydev) mii_t1_adv_l_mod_linkmode_t(phydev->lp_advertising, 0); mii_t1_adv_m_mod_linkmode_t(phydev->lp_advertising, 0); - phydev->pause = 0; - phydev->asym_pause = 0; + phydev->pause = false; + phydev->asym_pause = false; return 0; } @@ -498,8 +498,8 @@ static int genphy_c45_baset1_read_lpa(struct phy_device *phydev) return val; mii_t1_adv_l_mod_linkmode_t(phydev->lp_advertising, val); - phydev->pause = val & MDIO_AN_T1_ADV_L_PAUSE_CAP ? 1 : 0; - phydev->asym_pause = val & MDIO_AN_T1_ADV_L_PAUSE_ASYM ? 1 : 0; + phydev->pause = val & MDIO_AN_T1_ADV_L_PAUSE_CAP; + phydev->asym_pause = val & MDIO_AN_T1_ADV_L_PAUSE_ASYM; val = phy_read_mmd(phydev, MDIO_MMD_AN, MDIO_AN_T1_LP_M); if (val < 0) @@ -536,8 +536,8 @@ int genphy_c45_read_lpa(struct phy_device *phydev) phydev->lp_advertising); mii_10gbt_stat_mod_linkmode_lpa_t(phydev->lp_advertising, 0); mii_adv_mod_linkmode_adv_t(phydev->lp_advertising, 0); - phydev->pause = 0; - phydev->asym_pause = 0; + phydev->pause = false; + phydev->asym_pause = false; return 0; } @@ -551,8 +551,8 @@ int genphy_c45_read_lpa(struct phy_device *phydev) return val; mii_adv_mod_linkmode_adv_t(phydev->lp_advertising, val); - phydev->pause = val & LPA_PAUSE_CAP ? 1 : 0; - phydev->asym_pause = val & LPA_PAUSE_ASYM ? 1 : 0; + phydev->pause = val & LPA_PAUSE_CAP; + phydev->asym_pause = val & LPA_PAUSE_ASYM; /* Read the link partner's 10G advertisement */ val = phy_read_mmd(phydev, MDIO_MMD_AN, MDIO_AN_10GBT_STAT); @@ -1171,8 +1171,8 @@ int genphy_c45_read_status(struct phy_device *phydev) phydev->speed = SPEED_UNKNOWN; phydev->duplex = DUPLEX_UNKNOWN; - phydev->pause = 0; - phydev->asym_pause = 0; + phydev->pause = false; + phydev->asym_pause = false; if (phydev->autoneg == AUTONEG_ENABLE) { ret = genphy_c45_read_lpa(phydev); diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c index 737747cf1906..81984d4ebb7c 100644 --- a/drivers/net/phy/phy_device.c +++ b/drivers/net/phy/phy_device.c @@ -825,8 +825,8 @@ struct phy_device *phy_device_create(struct mii_bus *bus, int addr, u32 phy_id, dev->speed = SPEED_UNKNOWN; dev->duplex = DUPLEX_UNKNOWN; - dev->pause = 0; - dev->asym_pause = 0; + dev->pause = false; + dev->asym_pause = false; dev->link = 0; dev->port = PORT_TP; dev->interface = PHY_INTERFACE_MODE_GMII; @@ -2092,8 +2092,8 @@ int genphy_setup_forced(struct phy_device *phydev) { u16 ctl; - phydev->pause = 0; - phydev->asym_pause = 0; + phydev->pause = false; + phydev->asym_pause = false; ctl = mii_bmcr_encode_fixed(phydev->speed, phydev->duplex); @@ -2500,8 +2500,8 @@ int genphy_read_status(struct phy_device *phydev) phydev->master_slave_state = MASTER_SLAVE_STATE_UNSUPPORTED; phydev->speed = SPEED_UNKNOWN; phydev->duplex = DUPLEX_UNKNOWN; - phydev->pause = 0; - phydev->asym_pause = 0; + phydev->pause = false; + phydev->asym_pause = false; if (phydev->is_gigabit_capable) { err = genphy_read_master_slave(phydev); @@ -2554,8 +2554,8 @@ int genphy_c37_read_status(struct phy_device *phydev, bool *changed) /* Signal link has changed */ *changed = true; phydev->duplex = DUPLEX_UNKNOWN; - phydev->pause = 0; - phydev->asym_pause = 0; + phydev->pause = false; + phydev->asym_pause = false; if (phydev->autoneg == AUTONEG_ENABLE && phydev->autoneg_complete) { lpa = phy_read(phydev, MII_LPA); diff --git a/include/linux/phy.h b/include/linux/phy.h index e3474f03cbc1..d145a200ea21 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -666,6 +666,8 @@ struct phy_device { /* The most recently read link state */ unsigned link:1; unsigned autoneg_complete:1; + bool pause:1; + bool asym_pause:1; /* Interrupts are enabled */ unsigned interrupts:1; @@ -690,8 +692,6 @@ struct phy_device { int speed; int duplex; int port; - int pause; - int asym_pause; u8 master_slave_get; u8 master_slave_set; u8 master_slave_state; From 6874520518868a660bda182da060b69372265a16 Mon Sep 17 00:00:00 2001 From: Maxime Chevallier Date: Mon, 3 Nov 2025 11:49:24 +0100 Subject: [PATCH 0522/1381] net: altera-tse: Set platform drvdata before registering netdev We don't have to wait until netdev is registered before setting it as the pdev's drvdata. Move it at netdev alloc time. Reviewed-by: Andrew Lunn Signed-off-by: Maxime Chevallier Link: https://patch.msgid.link/20251103104928.58461-2-maxime.chevallier@bootlin.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/altera/altera_tse_main.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/net/ethernet/altera/altera_tse_main.c b/drivers/net/ethernet/altera/altera_tse_main.c index 3f6204de9e6b..6ba1249f027d 100644 --- a/drivers/net/ethernet/altera/altera_tse_main.c +++ b/drivers/net/ethernet/altera/altera_tse_main.c @@ -1150,6 +1150,7 @@ static int altera_tse_probe(struct platform_device *pdev) } SET_NETDEV_DEV(ndev, &pdev->dev); + platform_set_drvdata(pdev, ndev); priv = netdev_priv(ndev); priv->device = &pdev->dev; @@ -1394,8 +1395,6 @@ static int altera_tse_probe(struct platform_device *pdev) goto err_register_netdev; } - platform_set_drvdata(pdev, ndev); - priv->revision = ioread32(&priv->mac_dev->megacore_revision); if (netif_msg_probe(priv)) From dd2619d38d7e97d17995e1156293be1ec08b058d Mon Sep 17 00:00:00 2001 From: Maxime Chevallier Date: Mon, 3 Nov 2025 11:49:25 +0100 Subject: [PATCH 0523/1381] net: altera-tse: Warn on bad revision at probe time Instead of reading the core revision at probe time, and print a warning for an unexecpected version at .ndo_open() time, let's print that warning directly in .probe(). This allows getting rid of the "revision" private field, and also prevent a potential race between reading the revision in .probe() after netdev registration, and accessing that revision in .ndo_open(). By printing the warning after register_netdev(), we are sure that we have a netdev name, and that we try to print the revision after having read it from the internal registers. Suggested-by: Andrew Lunn Signed-off-by: Maxime Chevallier Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20251103104928.58461-3-maxime.chevallier@bootlin.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/altera/altera_tse.h | 3 --- drivers/net/ethernet/altera/altera_tse_main.c | 12 ++++++------ 2 files changed, 6 insertions(+), 9 deletions(-) diff --git a/drivers/net/ethernet/altera/altera_tse.h b/drivers/net/ethernet/altera/altera_tse.h index 82f2363a45cd..e5a56bb989da 100644 --- a/drivers/net/ethernet/altera/altera_tse.h +++ b/drivers/net/ethernet/altera/altera_tse.h @@ -401,9 +401,6 @@ struct altera_tse_private { /* MAC address space */ struct altera_tse_mac __iomem *mac_dev; - /* TSE Revision */ - u32 revision; - /* mSGDMA Rx Dispatcher address space */ void __iomem *rx_dma_csr; void __iomem *rx_dma_desc; diff --git a/drivers/net/ethernet/altera/altera_tse_main.c b/drivers/net/ethernet/altera/altera_tse_main.c index 6ba1249f027d..343c78a493a1 100644 --- a/drivers/net/ethernet/altera/altera_tse_main.c +++ b/drivers/net/ethernet/altera/altera_tse_main.c @@ -892,9 +892,6 @@ static int tse_open(struct net_device *dev) netdev_warn(dev, "device MAC address %pM\n", dev->dev_addr); - if ((priv->revision < 0xd00) || (priv->revision > 0xe00)) - netdev_warn(dev, "TSE revision %x\n", priv->revision); - spin_lock(&priv->mac_cfg_lock); ret = reset_mac(priv); @@ -1142,6 +1139,7 @@ static int altera_tse_probe(struct platform_device *pdev) struct net_device *ndev; void __iomem *descmap; int ret = -ENODEV; + u32 revision; ndev = alloc_etherdev(sizeof(struct altera_tse_private)); if (!ndev) { @@ -1395,12 +1393,14 @@ static int altera_tse_probe(struct platform_device *pdev) goto err_register_netdev; } - priv->revision = ioread32(&priv->mac_dev->megacore_revision); + revision = ioread32(&priv->mac_dev->megacore_revision); + + if (revision < 0xd00 || revision > 0xe00) + netdev_warn(ndev, "TSE revision %x\n", revision); if (netif_msg_probe(priv)) dev_info(&pdev->dev, "Altera TSE MAC version %d.%d at 0x%08lx irq %d/%d\n", - (priv->revision >> 8) & 0xff, - priv->revision & 0xff, + (revision >> 8) & 0xff, revision & 0xff, (unsigned long) control_port->start, priv->rx_irq, priv->tx_irq); From 9350ea63fec6f0cd713b6e90bd60cc2ee433b14f Mon Sep 17 00:00:00 2001 From: Maxime Chevallier Date: Mon, 3 Nov 2025 11:49:26 +0100 Subject: [PATCH 0524/1381] net: altera-tse: Don't use netdev name for the PCS mdio bus The PCS mdio bus must be created before registering the net_device. To do that, we musn't depend on the netdev name to create the mdio bus name. Let's use the device's name instead. Note that this changes the bus name in /sys/bus/mdiobus Reviewed-by: Andrew Lunn Signed-off-by: Maxime Chevallier Link: https://patch.msgid.link/20251103104928.58461-4-maxime.chevallier@bootlin.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/altera/altera_tse_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/altera/altera_tse_main.c b/drivers/net/ethernet/altera/altera_tse_main.c index 343c78a493a1..003df8970998 100644 --- a/drivers/net/ethernet/altera/altera_tse_main.c +++ b/drivers/net/ethernet/altera/altera_tse_main.c @@ -1404,7 +1404,7 @@ static int altera_tse_probe(struct platform_device *pdev) (unsigned long) control_port->start, priv->rx_irq, priv->tx_irq); - snprintf(mrc.name, MII_BUS_ID_SIZE, "%s-pcs-mii", ndev->name); + snprintf(mrc.name, MII_BUS_ID_SIZE, "%s-pcs-mii", dev_name(&pdev->dev)); pcs_bus = devm_mdio_regmap_register(&pdev->dev, &mrc); if (IS_ERR(pcs_bus)) { ret = PTR_ERR(pcs_bus); From 055e554b8fff7bb47e026bb8d199b213756e4321 Mon Sep 17 00:00:00 2001 From: Maxime Chevallier Date: Mon, 3 Nov 2025 11:49:27 +0100 Subject: [PATCH 0525/1381] net: altera-tse: Init PCS and phylink before registering netdev register_netdev() must be done only once all resources are ready, as they may be used in .ndo_open() immediately upon registration. Move the lynx PCS and phylink initialisation before registerng the netdevice. We also remove the call to netif_carrier_off(), as phylink takes care of that. Reviewed-by: Andrew Lunn Signed-off-by: Maxime Chevallier Link: https://patch.msgid.link/20251103104928.58461-5-maxime.chevallier@bootlin.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/altera/altera_tse_main.c | 40 +++++++++---------- 1 file changed, 20 insertions(+), 20 deletions(-) diff --git a/drivers/net/ethernet/altera/altera_tse_main.c b/drivers/net/ethernet/altera/altera_tse_main.c index 003df8970998..ca55c5fd11df 100644 --- a/drivers/net/ethernet/altera/altera_tse_main.c +++ b/drivers/net/ethernet/altera/altera_tse_main.c @@ -1386,24 +1386,6 @@ static int altera_tse_probe(struct platform_device *pdev) spin_lock_init(&priv->tx_lock); spin_lock_init(&priv->rxdma_irq_lock); - netif_carrier_off(ndev); - ret = register_netdev(ndev); - if (ret) { - dev_err(&pdev->dev, "failed to register TSE net device\n"); - goto err_register_netdev; - } - - revision = ioread32(&priv->mac_dev->megacore_revision); - - if (revision < 0xd00 || revision > 0xe00) - netdev_warn(ndev, "TSE revision %x\n", revision); - - if (netif_msg_probe(priv)) - dev_info(&pdev->dev, "Altera TSE MAC version %d.%d at 0x%08lx irq %d/%d\n", - (revision >> 8) & 0xff, revision & 0xff, - (unsigned long) control_port->start, priv->rx_irq, - priv->tx_irq); - snprintf(mrc.name, MII_BUS_ID_SIZE, "%s-pcs-mii", dev_name(&pdev->dev)); pcs_bus = devm_mdio_regmap_register(&pdev->dev, &mrc); if (IS_ERR(pcs_bus)) { @@ -1441,12 +1423,30 @@ static int altera_tse_probe(struct platform_device *pdev) goto err_init_phylink; } + ret = register_netdev(ndev); + if (ret) { + dev_err(&pdev->dev, "failed to register TSE net device\n"); + goto err_register_netdev; + } + + revision = ioread32(&priv->mac_dev->megacore_revision); + + if (revision < 0xd00 || revision > 0xe00) + netdev_warn(ndev, "TSE revision %x\n", revision); + + if (netif_msg_probe(priv)) + dev_info(&pdev->dev, "Altera TSE MAC version %d.%d at 0x%08lx irq %d/%d\n", + (revision >> 8) & 0xff, revision & 0xff, + (unsigned long)control_port->start, priv->rx_irq, + priv->tx_irq); + return 0; + +err_register_netdev: + phylink_destroy(priv->phylink); err_init_phylink: lynx_pcs_destroy(priv->pcs); err_init_pcs: - unregister_netdev(ndev); -err_register_netdev: netif_napi_del(&priv->napi); altera_tse_mdio_destroy(ndev); err_free_netdev: From c9445e3c087656e01d0160a48f90389856baf368 Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Thu, 30 Oct 2025 22:41:19 +0100 Subject: [PATCH 0526/1381] net: phy: fixed_phy: add helper fixed_phy_register_100fd In few places a 100FD fixed PHY is used. Create a helper so that users don't have to define the struct fixed_phy_status. Signed-off-by: Heiner Kallweit Link: https://patch.msgid.link/bf564b19-e9bc-4896-aeae-9f721cc4fecd@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/phy/fixed_phy.c | 12 ++++++++++++ include/linux/phy_fixed.h | 6 ++++++ 2 files changed, 18 insertions(+) diff --git a/drivers/net/phy/fixed_phy.c b/drivers/net/phy/fixed_phy.c index 0e1b28f06f18..bdc3a4bffede 100644 --- a/drivers/net/phy/fixed_phy.c +++ b/drivers/net/phy/fixed_phy.c @@ -227,6 +227,18 @@ struct phy_device *fixed_phy_register(const struct fixed_phy_status *status, } EXPORT_SYMBOL_GPL(fixed_phy_register); +struct phy_device *fixed_phy_register_100fd(void) +{ + static const struct fixed_phy_status status = { + .link = 1, + .speed = SPEED_100, + .duplex = DUPLEX_FULL, + }; + + return fixed_phy_register(&status, NULL); +} +EXPORT_SYMBOL_GPL(fixed_phy_register_100fd); + void fixed_phy_unregister(struct phy_device *phy) { phy_device_remove(phy); diff --git a/include/linux/phy_fixed.h b/include/linux/phy_fixed.h index d17ff750c708..08275ef64147 100644 --- a/include/linux/phy_fixed.h +++ b/include/linux/phy_fixed.h @@ -20,6 +20,7 @@ extern int fixed_phy_change_carrier(struct net_device *dev, bool new_carrier); void fixed_phy_add(const struct fixed_phy_status *status); struct phy_device *fixed_phy_register(const struct fixed_phy_status *status, struct device_node *np); +struct phy_device *fixed_phy_register_100fd(void); extern void fixed_phy_unregister(struct phy_device *phydev); extern int fixed_phy_set_link_update(struct phy_device *phydev, @@ -34,6 +35,11 @@ fixed_phy_register(const struct fixed_phy_status *status, return ERR_PTR(-ENODEV); } +static inline struct phy_device *fixed_phy_register_100fd(void) +{ + return ERR_PTR(-ENODEV); +} + static inline void fixed_phy_unregister(struct phy_device *phydev) { } From dc86b621e1b4129cc9ceea09ee449ae97fcf106f Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Thu, 30 Oct 2025 22:42:30 +0100 Subject: [PATCH 0527/1381] net: fec: register a fixed phy using fixed_phy_register_100fd if needed In case of coldfire/5272 a fixed phy is used, which so far is created by platform code, using fixed_phy_add(). This function has a number of problems, therefore create a potentially needed fixed phy here, using fixed_phy_register_100fd. Note 1: This includes a small functional change, as coldfire/5272 created a fixed phy in half-duplex mode. Likely this was by mistake, because the fec MAC is 100FD-capable, and connection is to a switch. Note 2: Usage of phy_find_next() makes use of the fact that dev_id can only be 0 or 1. Due to lack of hardware, this is compile-tested only. Signed-off-by: Heiner Kallweit Link: https://patch.msgid.link/adf4dc5c-5fa3-4ae6-a75c-a73954dede73@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/freescale/Kconfig | 1 + drivers/net/ethernet/freescale/fec_main.c | 52 +++++++++++------------ 2 files changed, 27 insertions(+), 26 deletions(-) diff --git a/drivers/net/ethernet/freescale/Kconfig b/drivers/net/ethernet/freescale/Kconfig index bbef47c3480c..e2a591cf9601 100644 --- a/drivers/net/ethernet/freescale/Kconfig +++ b/drivers/net/ethernet/freescale/Kconfig @@ -28,6 +28,7 @@ config FEC depends on PTP_1588_CLOCK_OPTIONAL select CRC32 select PHYLIB + select FIXED_PHY if M5272 select PAGE_POOL imply PAGE_POOL_STATS imply NET_SELFTESTS diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c index 024dd443bfbc..742f3e81cc7c 100644 --- a/drivers/net/ethernet/freescale/fec_main.c +++ b/drivers/net/ethernet/freescale/fec_main.c @@ -52,6 +52,7 @@ #include #include #include +#include #include #include #include @@ -2472,11 +2473,8 @@ static int fec_enet_parse_rgmii_delay(struct fec_enet_private *fep, static int fec_enet_mii_probe(struct net_device *ndev) { struct fec_enet_private *fep = netdev_priv(ndev); - struct phy_device *phy_dev = NULL; - char mdio_bus_id[MII_BUS_ID_SIZE]; - char phy_name[MII_BUS_ID_SIZE + 3]; - int phy_id; - int dev_id = fep->dev_id; + struct phy_device *phy_dev; + int ret; if (fep->phy_node) { phy_dev = of_phy_connect(ndev, fep->phy_node, @@ -2488,30 +2486,28 @@ static int fec_enet_mii_probe(struct net_device *ndev) } } else { /* check for attached phy */ - for (phy_id = 0; (phy_id < PHY_MAX_ADDR); phy_id++) { - if (!mdiobus_is_registered_device(fep->mii_bus, phy_id)) - continue; - if (dev_id--) - continue; - strscpy(mdio_bus_id, fep->mii_bus->id, MII_BUS_ID_SIZE); - break; - } + phy_dev = phy_find_first(fep->mii_bus); + if (fep->dev_id && phy_dev) + phy_dev = phy_find_next(fep->mii_bus, phy_dev); - if (phy_id >= PHY_MAX_ADDR) { + if (!phy_dev) { netdev_info(ndev, "no PHY, assuming direct connection to switch\n"); - strscpy(mdio_bus_id, "fixed-0", MII_BUS_ID_SIZE); - phy_id = 0; + phy_dev = fixed_phy_register_100fd(); + if (IS_ERR(phy_dev)) { + netdev_err(ndev, "could not register fixed PHY\n"); + return PTR_ERR(phy_dev); + } } - snprintf(phy_name, sizeof(phy_name), - PHY_ID_FMT, mdio_bus_id, phy_id); - phy_dev = phy_connect(ndev, phy_name, &fec_enet_adjust_link, - fep->phy_interface); - } + ret = phy_connect_direct(ndev, phy_dev, &fec_enet_adjust_link, + fep->phy_interface); + if (ret) { + if (phy_is_pseudo_fixed_link(phy_dev)) + fixed_phy_unregister(phy_dev); + netdev_err(ndev, "could not attach to PHY\n"); + return ret; + } - if (IS_ERR(phy_dev)) { - netdev_err(ndev, "could not attach to PHY\n"); - return PTR_ERR(phy_dev); } /* mask with MAC supported features */ @@ -3616,8 +3612,9 @@ static int fec_enet_close(struct net_device *ndev) { struct fec_enet_private *fep = netdev_priv(ndev); + struct phy_device *phy_dev = ndev->phydev; - phy_stop(ndev->phydev); + phy_stop(phy_dev); if (netif_device_present(ndev)) { napi_disable(&fep->napi); @@ -3625,7 +3622,10 @@ fec_enet_close(struct net_device *ndev) fec_stop(ndev); } - phy_disconnect(ndev->phydev); + phy_disconnect(phy_dev); + + if (!fep->phy_node && phy_is_pseudo_fixed_link(phy_dev)) + fixed_phy_unregister(phy_dev); if (fep->quirks & FEC_QUIRK_ERR006687) imx6q_cpuidle_fec_irqs_unused(); From 0ee21f39c5d844e0b30ea323542b39ce73b3dd86 Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Thu, 30 Oct 2025 22:43:36 +0100 Subject: [PATCH 0528/1381] m68k: coldfire: remove creating a fixed phy Now that the fec ethernet driver creates a fixed phy if needed, we can remove this here. Signed-off-by: Heiner Kallweit Link: https://patch.msgid.link/212e0cb5-a2f5-460f-8e03-3c3369d0acf1@gmail.com Signed-off-by: Jakub Kicinski --- arch/m68k/coldfire/m5272.c | 15 --------------- 1 file changed, 15 deletions(-) diff --git a/arch/m68k/coldfire/m5272.c b/arch/m68k/coldfire/m5272.c index 918e2a3236c5..28b3ffa25ba0 100644 --- a/arch/m68k/coldfire/m5272.c +++ b/arch/m68k/coldfire/m5272.c @@ -16,7 +16,6 @@ #include #include #include -#include #include #include #include @@ -103,23 +102,9 @@ void __init config_BSP(char *commandp, int size) /***************************************************************************/ -/* - * Some 5272 based boards have the FEC ethernet directly connected to - * an ethernet switch. In this case we need to use the fixed phy type, - * and we need to declare it early in boot. - */ -static const struct fixed_phy_status nettel_fixed_phy_status __initconst = { - .link = 1, - .speed = 100, - .duplex = 0, -}; - -/***************************************************************************/ - static int __init init_BSP(void) { m5272_uarts_init(); - fixed_phy_add(&nettel_fixed_phy_status); clkdev_add_table(m5272_clk_lookup, ARRAY_SIZE(m5272_clk_lookup)); return 0; } From 10d2f15afba2391471576846fd22d49631e34b21 Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Thu, 30 Oct 2025 22:44:35 +0100 Subject: [PATCH 0529/1381] net: b44: register a fixed phy using fixed_phy_register_100fd if needed In case of bcm47xx a fixed phy is used, which so far is created by platform code, using fixed_phy_add(). This function has a number of problems, therefore create a potentially needed fixed phy here, using fixed_phy_register_100fd. Due to lack of hardware, this is compile-tested only. Signed-off-by: Heiner Kallweit Link: https://patch.msgid.link/53e4e74d-a49e-4f37-b970-5543a35041db@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/broadcom/Kconfig | 1 + drivers/net/ethernet/broadcom/b44.c | 37 +++++++++++++++------------ 2 files changed, 21 insertions(+), 17 deletions(-) diff --git a/drivers/net/ethernet/broadcom/Kconfig b/drivers/net/ethernet/broadcom/Kconfig index 9fdef874f5ca..666522d64775 100644 --- a/drivers/net/ethernet/broadcom/Kconfig +++ b/drivers/net/ethernet/broadcom/Kconfig @@ -25,6 +25,7 @@ config B44 select SSB select MII select PHYLIB + select FIXED_PHY if BCM47XX help If you have a network (Ethernet) controller of this type, say Y or M here. diff --git a/drivers/net/ethernet/broadcom/b44.c b/drivers/net/ethernet/broadcom/b44.c index 0353359c3fe9..888f28f11406 100644 --- a/drivers/net/ethernet/broadcom/b44.c +++ b/drivers/net/ethernet/broadcom/b44.c @@ -31,6 +31,7 @@ #include #include #include +#include #include #include @@ -2233,7 +2234,6 @@ static int b44_register_phy_one(struct b44 *bp) struct mii_bus *mii_bus; struct ssb_device *sdev = bp->sdev; struct phy_device *phydev; - char bus_id[MII_BUS_ID_SIZE + 3]; struct ssb_sprom *sprom = &sdev->bus->sprom; int err; @@ -2260,27 +2260,26 @@ static int b44_register_phy_one(struct b44 *bp) goto err_out_mdiobus; } - if (!mdiobus_is_registered_device(bp->mii_bus, bp->phy_addr) && - (sprom->boardflags_lo & (B44_BOARDFLAG_ROBO | B44_BOARDFLAG_ADM))) { - + phydev = mdiobus_get_phy(bp->mii_bus, bp->phy_addr); + if (!phydev && + sprom->boardflags_lo & (B44_BOARDFLAG_ROBO | B44_BOARDFLAG_ADM)) { dev_info(sdev->dev, "could not find PHY at %i, use fixed one\n", bp->phy_addr); - bp->phy_addr = 0; - snprintf(bus_id, sizeof(bus_id), PHY_ID_FMT, "fixed-0", - bp->phy_addr); - } else { - snprintf(bus_id, sizeof(bus_id), PHY_ID_FMT, mii_bus->id, - bp->phy_addr); + phydev = fixed_phy_register_100fd(); + if (!IS_ERR(phydev)) + bp->phy_addr = phydev->mdio.addr; } - phydev = phy_connect(bp->dev, bus_id, &b44_adjust_link, - PHY_INTERFACE_MODE_MII); - if (IS_ERR(phydev)) { + if (IS_ERR_OR_NULL(phydev)) + err = -ENODEV; + else + err = phy_connect_direct(bp->dev, phydev, &b44_adjust_link, + PHY_INTERFACE_MODE_MII); + if (err) { dev_err(sdev->dev, "could not attach PHY at %i\n", bp->phy_addr); - err = PTR_ERR(phydev); goto err_out_mdiobus_unregister; } @@ -2293,7 +2292,6 @@ static int b44_register_phy_one(struct b44 *bp) linkmode_copy(phydev->advertising, phydev->supported); bp->old_link = 0; - bp->phy_addr = phydev->mdio.addr; phy_attached_info(phydev); @@ -2311,10 +2309,15 @@ static int b44_register_phy_one(struct b44 *bp) static void b44_unregister_phy_one(struct b44 *bp) { - struct net_device *dev = bp->dev; struct mii_bus *mii_bus = bp->mii_bus; + struct net_device *dev = bp->dev; + struct phy_device *phydev; - phy_disconnect(dev->phydev); + phydev = dev->phydev; + + phy_disconnect(phydev); + if (phy_is_pseudo_fixed_link(phydev)) + fixed_phy_unregister(phydev); mdiobus_unregister(mii_bus); mdiobus_free(mii_bus); } From 458639c42b7e6927a746bbbf0954ce3dd738c468 Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Thu, 30 Oct 2025 22:45:30 +0100 Subject: [PATCH 0530/1381] MIPS: BCM47XX: remove creating a fixed phy Now that b44 ethernet driver creates a fixed phy if needed, we can remove this here. Signed-off-by: Heiner Kallweit Link: https://patch.msgid.link/8983b705-6bca-4728-9283-7aa60f49340f@gmail.com Signed-off-by: Jakub Kicinski --- arch/mips/bcm47xx/setup.c | 7 ------- 1 file changed, 7 deletions(-) diff --git a/arch/mips/bcm47xx/setup.c b/arch/mips/bcm47xx/setup.c index a93a4266dc1e..38ed61b4bd96 100644 --- a/arch/mips/bcm47xx/setup.c +++ b/arch/mips/bcm47xx/setup.c @@ -256,12 +256,6 @@ static int __init bcm47xx_cpu_fixes(void) } arch_initcall(bcm47xx_cpu_fixes); -static const struct fixed_phy_status bcm47xx_fixed_phy_status __initconst = { - .link = 1, - .speed = SPEED_100, - .duplex = DUPLEX_FULL, -}; - static int __init bcm47xx_register_bus_complete(void) { switch (bcm47xx_bus_type) { @@ -282,7 +276,6 @@ static int __init bcm47xx_register_bus_complete(void) bcm47xx_leds_register(); bcm47xx_workarounds(); - fixed_phy_add(&bcm47xx_fixed_phy_status); return 0; } device_initcall(bcm47xx_register_bus_complete); From 5de9ea1c50f061892625388880e83fdc50a4ef66 Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Thu, 30 Oct 2025 22:46:32 +0100 Subject: [PATCH 0531/1381] net: phy: fixed_phy: remove fixed_phy_add fixed_phy_add() has a number of problems/disadvantages: - It uses phy address 0 w/o checking whether a fixed phy with this address exists already. - A subsequent call to fixed_phy_register() would also use phy address 0, because fixed_phy_add() doesn't mark it as used. - fixed_phy_add() is used from platform code, therefore requires that fixed_phy code is built-in. Now that for the only two users (coldfire/5272 and bcm47xx) fixed_phy creation has been moved to the respective ethernet driver (fec, b44), we can remove fixed_phy_add(). Signed-off-by: Heiner Kallweit Link: https://patch.msgid.link/bee046a1-1e77-4057-8b04-fdb2a1bbbd08@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/phy/fixed_phy.c | 6 ------ include/linux/phy_fixed.h | 2 -- 2 files changed, 8 deletions(-) diff --git a/drivers/net/phy/fixed_phy.c b/drivers/net/phy/fixed_phy.c index bdc3a4bffede..d498d8a9bba6 100644 --- a/drivers/net/phy/fixed_phy.c +++ b/drivers/net/phy/fixed_phy.c @@ -131,12 +131,6 @@ static int __fixed_phy_add(int phy_addr, return 0; } -void fixed_phy_add(const struct fixed_phy_status *status) -{ - __fixed_phy_add(0, status); -} -EXPORT_SYMBOL_GPL(fixed_phy_add); - static DEFINE_IDA(phy_fixed_ida); static void fixed_phy_del(int phy_addr) diff --git a/include/linux/phy_fixed.h b/include/linux/phy_fixed.h index 08275ef64147..8bade999831c 100644 --- a/include/linux/phy_fixed.h +++ b/include/linux/phy_fixed.h @@ -17,7 +17,6 @@ struct net_device; #if IS_ENABLED(CONFIG_FIXED_PHY) extern int fixed_phy_change_carrier(struct net_device *dev, bool new_carrier); -void fixed_phy_add(const struct fixed_phy_status *status); struct phy_device *fixed_phy_register(const struct fixed_phy_status *status, struct device_node *np); struct phy_device *fixed_phy_register_100fd(void); @@ -27,7 +26,6 @@ extern int fixed_phy_set_link_update(struct phy_device *phydev, int (*link_update)(struct net_device *, struct fixed_phy_status *)); #else -static inline void fixed_phy_add(const struct fixed_phy_status *status) {} static inline struct phy_device * fixed_phy_register(const struct fixed_phy_status *status, struct device_node *np) From bf33247a90d3e85d53a9b55bb276b725456ff0bf Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Mon, 3 Nov 2025 16:26:09 -0800 Subject: [PATCH 0532/1381] net: Add struct sockaddr_unsized for sockaddr of unknown length Add flexible sockaddr structure to support addresses longer than the traditional 14-byte struct sockaddr::sa_data limitation without requiring the full 128-byte sa_data of struct sockaddr_storage. This allows the network APIs to pass around a pointer to an object that isn't lying to the compiler about how big it is, but must be accompanied by its actual size as an additional parameter. It's possible we may way to migrate to including the size with the struct in the future, e.g.: struct sockaddr_unsized { u16 sa_data_len; u16 sa_family; u8 sa_data[] __counted_by(sa_data_len); }; Signed-off-by: Kees Cook Link: https://patch.msgid.link/20251104002617.2752303-1-kees@kernel.org Signed-off-by: Jakub Kicinski --- include/linux/socket.h | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) diff --git a/include/linux/socket.h b/include/linux/socket.h index 3b262487ec06..7b1a01be29da 100644 --- a/include/linux/socket.h +++ b/include/linux/socket.h @@ -40,6 +40,23 @@ struct sockaddr { }; }; +/** + * struct sockaddr_unsized - Unspecified size sockaddr for callbacks + * @sa_family: Address family (AF_UNIX, AF_INET, AF_INET6, etc.) + * @sa_data: Flexible array for address data + * + * This structure is designed for callback interfaces where the + * total size is known via the sockaddr_len parameter. Unlike struct + * sockaddr which has a fixed 14-byte sa_data limit or struct + * sockaddr_storage which has a fixed 128-byte sa_data limit, this + * structure can accommodate addresses of any size, but must be used + * carefully. + */ +struct sockaddr_unsized { + __kernel_sa_family_t sa_family; /* address family, AF_xxx */ + char sa_data[]; /* flexible address data */ +}; + struct linger { int l_onoff; /* Linger active */ int l_linger; /* How long to linger for */ From 0e50474fa514822e9d990874e554bf8043a201d7 Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Mon, 3 Nov 2025 16:26:10 -0800 Subject: [PATCH 0533/1381] net: Convert proto_ops bind() callbacks to use sockaddr_unsized Update all struct proto_ops bind() callback function prototypes from "struct sockaddr *" to "struct sockaddr_unsized *" to avoid lying to the compiler about object sizes. Calls into struct proto handlers gain casts that will be removed in the struct proto conversion patch. No binary changes expected. Signed-off-by: Kees Cook Link: https://patch.msgid.link/20251104002617.2752303-2-kees@kernel.org Signed-off-by: Jakub Kicinski --- crypto/af_alg.c | 2 +- drivers/block/drbd/drbd_receiver.c | 4 ++-- drivers/infiniband/hw/erdma/erdma_cm.c | 4 ++-- drivers/infiniband/sw/siw/siw_cm.c | 6 +++--- drivers/isdn/mISDN/l1oip_core.c | 2 +- drivers/isdn/mISDN/socket.c | 4 ++-- drivers/net/ppp/pptp.c | 4 ++-- drivers/nvme/host/tcp.c | 2 +- drivers/nvme/target/tcp.c | 2 +- drivers/target/iscsi/iscsi_target_login.c | 2 +- drivers/xen/pvcalls-back.c | 2 +- fs/afs/rxrpc.c | 6 +++--- fs/dlm/lowcomms.c | 6 +++--- fs/ocfs2/cluster/tcp.c | 4 ++-- fs/smb/client/connect.c | 2 +- fs/smb/server/transport_tcp.c | 4 ++-- include/linux/net.h | 4 ++-- include/net/inet_common.h | 2 +- include/net/ipv6.h | 2 +- include/net/sock.h | 2 +- net/9p/trans_fd.c | 2 +- net/appletalk/ddp.c | 2 +- net/atm/pvc.c | 4 ++-- net/atm/svc.c | 2 +- net/ax25/af_ax25.c | 2 +- net/bluetooth/hci_sock.c | 2 +- net/bluetooth/iso.c | 4 ++-- net/bluetooth/l2cap_sock.c | 2 +- net/bluetooth/rfcomm/core.c | 4 ++-- net/bluetooth/rfcomm/sock.c | 2 +- net/bluetooth/sco.c | 2 +- net/can/isotp.c | 2 +- net/can/j1939/socket.c | 2 +- net/can/raw.c | 2 +- net/core/sock.c | 2 +- net/ieee802154/socket.c | 4 ++-- net/ipv4/af_inet.c | 4 ++-- net/ipv4/udp_tunnel_core.c | 2 +- net/ipv6/af_inet6.c | 4 ++-- net/ipv6/ip6_udp_tunnel.c | 2 +- net/iucv/af_iucv.c | 2 +- net/l2tp/l2tp_core.c | 4 ++-- net/llc/af_llc.c | 2 +- net/mctp/af_mctp.c | 2 +- net/mctp/test/route-test.c | 2 +- net/mptcp/protocol.c | 6 +++--- net/mptcp/subflow.c | 2 +- net/netfilter/ipvs/ip_vs_sync.c | 4 ++-- net/netlink/af_netlink.c | 2 +- net/netrom/af_netrom.c | 2 +- net/nfc/llcp_sock.c | 4 ++-- net/packet/af_packet.c | 11 ++++++----- net/phonet/socket.c | 8 ++++---- net/qrtr/af_qrtr.c | 2 +- net/qrtr/ns.c | 2 +- net/rds/bind.c | 2 +- net/rds/rds.h | 2 +- net/rds/tcp_connect.c | 2 +- net/rds/tcp_listen.c | 2 +- net/rose/af_rose.c | 2 +- net/rxrpc/af_rxrpc.c | 2 +- net/rxrpc/rxperf.c | 2 +- net/smc/af_smc.c | 2 +- net/smc/smc.h | 2 +- net/socket.c | 6 +++--- net/sunrpc/clnt.c | 4 ++-- net/sunrpc/svcsock.c | 2 +- net/sunrpc/xprtsock.c | 4 ++-- net/tipc/socket.c | 4 ++-- net/unix/af_unix.c | 4 ++-- net/vmw_vsock/af_vsock.c | 4 ++-- net/x25/af_x25.c | 2 +- net/xdp/xsk.c | 2 +- tools/testing/selftests/bpf/test_kmods/bpf_testmod.c | 2 +- 74 files changed, 113 insertions(+), 112 deletions(-) diff --git a/crypto/af_alg.c b/crypto/af_alg.c index ca6fdcc6c54a..5e760ab62618 100644 --- a/crypto/af_alg.c +++ b/crypto/af_alg.c @@ -145,7 +145,7 @@ void af_alg_release_parent(struct sock *sk) } EXPORT_SYMBOL_GPL(af_alg_release_parent); -static int alg_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) +static int alg_bind(struct socket *sock, struct sockaddr_unsized *uaddr, int addr_len) { const u32 allowed = CRYPTO_ALG_KERN_DRIVER_ONLY; struct sock *sk = sock->sk; diff --git a/drivers/block/drbd/drbd_receiver.c b/drivers/block/drbd/drbd_receiver.c index caaf2781136d..d9296f74f902 100644 --- a/drivers/block/drbd/drbd_receiver.c +++ b/drivers/block/drbd/drbd_receiver.c @@ -450,7 +450,7 @@ static struct socket *drbd_try_connect(struct drbd_connection *connection) * a free one dynamically. */ what = "bind before connect"; - err = sock->ops->bind(sock, (struct sockaddr *) &src_in6, my_addr_len); + err = sock->ops->bind(sock, (struct sockaddr_unsized *) &src_in6, my_addr_len); if (err < 0) goto out; @@ -537,7 +537,7 @@ static int prepare_listen_socket(struct drbd_connection *connection, struct acce drbd_setbufsize(s_listen, sndbuf_size, rcvbuf_size); what = "bind before listen"; - err = s_listen->ops->bind(s_listen, (struct sockaddr *)&my_addr, my_addr_len); + err = s_listen->ops->bind(s_listen, (struct sockaddr_unsized *)&my_addr, my_addr_len); if (err < 0) goto out; diff --git a/drivers/infiniband/hw/erdma/erdma_cm.c b/drivers/infiniband/hw/erdma/erdma_cm.c index e0acc185e719..ef66a6359eb9 100644 --- a/drivers/infiniband/hw/erdma/erdma_cm.c +++ b/drivers/infiniband/hw/erdma/erdma_cm.c @@ -993,7 +993,7 @@ static int kernel_bindconnect(struct socket *s, struct sockaddr *laddr, int ret; sock_set_reuseaddr(s->sk); - ret = s->ops->bind(s, laddr, laddrlen); + ret = s->ops->bind(s, (struct sockaddr_unsized *)laddr, laddrlen); if (ret) return ret; ret = s->ops->connect(s, raddr, raddrlen, flags); @@ -1315,7 +1315,7 @@ int erdma_create_listen(struct iw_cm_id *id, int backlog) if (ipv4_is_zeronet(laddr->sin_addr.s_addr)) s->sk->sk_bound_dev_if = dev->netdev->ifindex; - ret = s->ops->bind(s, (struct sockaddr *)laddr, + ret = s->ops->bind(s, (struct sockaddr_unsized *)laddr, sizeof(struct sockaddr_in)); if (ret) goto error; diff --git a/drivers/infiniband/sw/siw/siw_cm.c b/drivers/infiniband/sw/siw/siw_cm.c index 708b13993fdf..7fe118cacb3f 100644 --- a/drivers/infiniband/sw/siw/siw_cm.c +++ b/drivers/infiniband/sw/siw/siw_cm.c @@ -1340,7 +1340,7 @@ static int kernel_bindconnect(struct socket *s, struct sockaddr *laddr, return rv; } - rv = s->ops->bind(s, laddr, size); + rv = s->ops->bind(s, (struct sockaddr_unsized *)laddr, size); if (rv < 0) return rv; @@ -1789,7 +1789,7 @@ int siw_create_listen(struct iw_cm_id *id, int backlog) goto error; } } - rv = s->ops->bind(s, (struct sockaddr *)laddr, + rv = s->ops->bind(s, (struct sockaddr_unsized *)laddr, sizeof(struct sockaddr_in)); } else { struct sockaddr_in6 *laddr = &to_sockaddr_in6(id->local_addr); @@ -1813,7 +1813,7 @@ int siw_create_listen(struct iw_cm_id *id, int backlog) goto error; } } - rv = s->ops->bind(s, (struct sockaddr *)laddr, + rv = s->ops->bind(s, (struct sockaddr_unsized *)laddr, sizeof(struct sockaddr_in6)); } if (rv) { diff --git a/drivers/isdn/mISDN/l1oip_core.c b/drivers/isdn/mISDN/l1oip_core.c index f732f6614d37..6ab036e4a35f 100644 --- a/drivers/isdn/mISDN/l1oip_core.c +++ b/drivers/isdn/mISDN/l1oip_core.c @@ -676,7 +676,7 @@ l1oip_socket_thread(void *data) hc->sin_remote.sin_port = htons((unsigned short)hc->remoteport); /* bind to incoming port */ - if (socket->ops->bind(socket, (struct sockaddr *)&hc->sin_local, + if (socket->ops->bind(socket, (struct sockaddr_unsized *)&hc->sin_local, sizeof(hc->sin_local))) { printk(KERN_ERR "%s: Failed to bind socket to port %d.\n", __func__, hc->localport); diff --git a/drivers/isdn/mISDN/socket.c b/drivers/isdn/mISDN/socket.c index b215b28cad7b..77b900db1cac 100644 --- a/drivers/isdn/mISDN/socket.c +++ b/drivers/isdn/mISDN/socket.c @@ -462,7 +462,7 @@ static int data_sock_getsockopt(struct socket *sock, int level, int optname, } static int -data_sock_bind(struct socket *sock, struct sockaddr *addr, int addr_len) +data_sock_bind(struct socket *sock, struct sockaddr_unsized *addr, int addr_len) { struct sockaddr_mISDN *maddr = (struct sockaddr_mISDN *) addr; struct sock *sk = sock->sk; @@ -696,7 +696,7 @@ base_sock_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg) } static int -base_sock_bind(struct socket *sock, struct sockaddr *addr, int addr_len) +base_sock_bind(struct socket *sock, struct sockaddr_unsized *addr, int addr_len) { struct sockaddr_mISDN *maddr = (struct sockaddr_mISDN *) addr; struct sock *sk = sock->sk; diff --git a/drivers/net/ppp/pptp.c b/drivers/net/ppp/pptp.c index 90737cb71892..d07e87a0974c 100644 --- a/drivers/net/ppp/pptp.c +++ b/drivers/net/ppp/pptp.c @@ -382,8 +382,8 @@ static int pptp_rcv(struct sk_buff *skb) return NET_RX_DROP; } -static int pptp_bind(struct socket *sock, struct sockaddr *uservaddr, - int sockaddr_len) +static int pptp_bind(struct socket *sock, struct sockaddr_unsized *uservaddr, + int sockaddr_len) { struct sock *sk = sock->sk; struct sockaddr_pppox *sp = (struct sockaddr_pppox *) uservaddr; diff --git a/drivers/nvme/host/tcp.c b/drivers/nvme/host/tcp.c index 9a96df1a511c..35d0bd91f6fd 100644 --- a/drivers/nvme/host/tcp.c +++ b/drivers/nvme/host/tcp.c @@ -1834,7 +1834,7 @@ static int nvme_tcp_alloc_queue(struct nvme_ctrl *nctrl, int qid, sk_set_memalloc(queue->sock->sk); if (nctrl->opts->mask & NVMF_OPT_HOST_TRADDR) { - ret = kernel_bind(queue->sock, (struct sockaddr *)&ctrl->src_addr, + ret = kernel_bind(queue->sock, (struct sockaddr_unsized *)&ctrl->src_addr, sizeof(ctrl->src_addr)); if (ret) { dev_err(nctrl->device, diff --git a/drivers/nvme/target/tcp.c b/drivers/nvme/target/tcp.c index 470bf37e5a63..d543da09ef8e 100644 --- a/drivers/nvme/target/tcp.c +++ b/drivers/nvme/target/tcp.c @@ -2055,7 +2055,7 @@ static int nvmet_tcp_add_port(struct nvmet_port *nport) if (so_priority > 0) sock_set_priority(port->sock->sk, so_priority); - ret = kernel_bind(port->sock, (struct sockaddr *)&port->addr, + ret = kernel_bind(port->sock, (struct sockaddr_unsized *)&port->addr, sizeof(port->addr)); if (ret) { pr_err("failed to bind port socket %d\n", ret); diff --git a/drivers/target/iscsi/iscsi_target_login.c b/drivers/target/iscsi/iscsi_target_login.c index c2ac9a99ebbb..53aca059dc16 100644 --- a/drivers/target/iscsi/iscsi_target_login.c +++ b/drivers/target/iscsi/iscsi_target_login.c @@ -822,7 +822,7 @@ int iscsit_setup_np( sock_set_reuseaddr(sock->sk); ip_sock_set_freebind(sock->sk); - ret = kernel_bind(sock, (struct sockaddr *)&np->np_sockaddr, len); + ret = kernel_bind(sock, (struct sockaddr_unsized *)&np->np_sockaddr, len); if (ret < 0) { pr_err("kernel_bind() failed: %d\n", ret); goto fail; diff --git a/drivers/xen/pvcalls-back.c b/drivers/xen/pvcalls-back.c index fd7ed65e0197..da1b516b9cfd 100644 --- a/drivers/xen/pvcalls-back.c +++ b/drivers/xen/pvcalls-back.c @@ -650,7 +650,7 @@ static int pvcalls_back_bind(struct xenbus_device *dev, if (ret < 0) goto out; - ret = inet_bind(map->sock, (struct sockaddr *)&req->u.bind.addr, + ret = inet_bind(map->sock, (struct sockaddr_unsized *)&req->u.bind.addr, req->u.bind.len); if (ret < 0) goto out; diff --git a/fs/afs/rxrpc.c b/fs/afs/rxrpc.c index c1cadf8fb346..bf0e4ea0aafd 100644 --- a/fs/afs/rxrpc.c +++ b/fs/afs/rxrpc.c @@ -82,16 +82,16 @@ int afs_open_socket(struct afs_net *net) if (ret < 0) pr_err("Couldn't create RxGK CM key: %d\n", ret); - ret = kernel_bind(socket, (struct sockaddr *) &srx, sizeof(srx)); + ret = kernel_bind(socket, (struct sockaddr_unsized *) &srx, sizeof(srx)); if (ret == -EADDRINUSE) { srx.transport.sin6.sin6_port = 0; - ret = kernel_bind(socket, (struct sockaddr *) &srx, sizeof(srx)); + ret = kernel_bind(socket, (struct sockaddr_unsized *) &srx, sizeof(srx)); } if (ret < 0) goto error_2; srx.srx_service = YFS_CM_SERVICE; - ret = kernel_bind(socket, (struct sockaddr *) &srx, sizeof(srx)); + ret = kernel_bind(socket, (struct sockaddr_unsized *) &srx, sizeof(srx)); if (ret < 0) goto error_2; diff --git a/fs/dlm/lowcomms.c b/fs/dlm/lowcomms.c index 9a0b6c2b6b01..0500421b6e3b 100644 --- a/fs/dlm/lowcomms.c +++ b/fs/dlm/lowcomms.c @@ -1134,7 +1134,7 @@ static int sctp_bind_addrs(struct socket *sock, __be16 port) make_sockaddr(&localaddr, port, &addr_len); if (!i) - result = kernel_bind(sock, addr, addr_len); + result = kernel_bind(sock, (struct sockaddr_unsized *)addr, addr_len); else result = sock_bind_add(sock->sk, addr, addr_len); @@ -1813,7 +1813,7 @@ static int dlm_tcp_bind(struct socket *sock) memcpy(&src_addr, &dlm_local_addr[0], sizeof(src_addr)); make_sockaddr(&src_addr, 0, &addr_len); - result = kernel_bind(sock, (struct sockaddr *)&src_addr, + result = kernel_bind(sock, (struct sockaddr_unsized *)&src_addr, addr_len); if (result < 0) { /* This *may* not indicate a critical error */ @@ -1852,7 +1852,7 @@ static int dlm_tcp_listen_bind(struct socket *sock) /* Bind to our port */ make_sockaddr(&dlm_local_addr[0], dlm_config.ci_tcp_port, &addr_len); - return kernel_bind(sock, (struct sockaddr *)&dlm_local_addr[0], + return kernel_bind(sock, (struct sockaddr_unsized *)&dlm_local_addr[0], addr_len); } diff --git a/fs/ocfs2/cluster/tcp.c b/fs/ocfs2/cluster/tcp.c index b05d4e9d13b2..c7734193d8d7 100644 --- a/fs/ocfs2/cluster/tcp.c +++ b/fs/ocfs2/cluster/tcp.c @@ -1615,7 +1615,7 @@ static void o2net_start_connect(struct work_struct *work) myaddr.sin_addr.s_addr = mynode->nd_ipv4_address; myaddr.sin_port = htons(0); /* any port */ - ret = sock->ops->bind(sock, (struct sockaddr *)&myaddr, + ret = sock->ops->bind(sock, (struct sockaddr_unsized *)&myaddr, sizeof(myaddr)); if (ret) { mlog(ML_ERROR, "bind failed with %d at address %pI4\n", @@ -2002,7 +2002,7 @@ static int o2net_open_listening_sock(__be32 addr, __be16 port) INIT_WORK(&o2net_listen_work, o2net_accept_many); sock->sk->sk_reuse = SK_CAN_REUSE; - ret = sock->ops->bind(sock, (struct sockaddr *)&sin, sizeof(sin)); + ret = sock->ops->bind(sock, (struct sockaddr_unsized *)&sin, sizeof(sin)); if (ret < 0) { printk(KERN_ERR "o2net: Error %d while binding socket at " "%pI4:%u\n", ret, &addr, ntohs(port)); diff --git a/fs/smb/client/connect.c b/fs/smb/client/connect.c index dd12f3eb61dc..96d972263020 100644 --- a/fs/smb/client/connect.c +++ b/fs/smb/client/connect.c @@ -3112,7 +3112,7 @@ bind_socket(struct TCP_Server_Info *server) struct socket *socket = server->ssocket; rc = kernel_bind(socket, - (struct sockaddr *) &server->srcaddr, + (struct sockaddr_unsized *) &server->srcaddr, sizeof(server->srcaddr)); if (rc < 0) { struct sockaddr_in *saddr4; diff --git a/fs/smb/server/transport_tcp.c b/fs/smb/server/transport_tcp.c index 7a1e3dcc2cde..bf694bc78c65 100644 --- a/fs/smb/server/transport_tcp.c +++ b/fs/smb/server/transport_tcp.c @@ -519,10 +519,10 @@ static int create_socket(struct interface *iface) } if (ipv4) - ret = kernel_bind(ksmbd_socket, (struct sockaddr *)&sin, + ret = kernel_bind(ksmbd_socket, (struct sockaddr_unsized *)&sin, sizeof(sin)); else - ret = kernel_bind(ksmbd_socket, (struct sockaddr *)&sin6, + ret = kernel_bind(ksmbd_socket, (struct sockaddr_unsized *)&sin6, sizeof(sin6)); if (ret) { pr_err("Failed to bind socket: %d\n", ret); diff --git a/include/linux/net.h b/include/linux/net.h index ec09620f40f7..0e316f063113 100644 --- a/include/linux/net.h +++ b/include/linux/net.h @@ -163,7 +163,7 @@ struct proto_ops { struct module *owner; int (*release) (struct socket *sock); int (*bind) (struct socket *sock, - struct sockaddr *myaddr, + struct sockaddr_unsized *myaddr, int sockaddr_len); int (*connect) (struct socket *sock, struct sockaddr *vaddr, @@ -345,7 +345,7 @@ int kernel_sendmsg(struct socket *sock, struct msghdr *msg, struct kvec *vec, int kernel_recvmsg(struct socket *sock, struct msghdr *msg, struct kvec *vec, size_t num, size_t len, int flags); -int kernel_bind(struct socket *sock, struct sockaddr *addr, int addrlen); +int kernel_bind(struct socket *sock, struct sockaddr_unsized *addr, int addrlen); int kernel_listen(struct socket *sock, int backlog); int kernel_accept(struct socket *sock, struct socket **newsock, int flags); int kernel_connect(struct socket *sock, struct sockaddr *addr, int addrlen, diff --git a/include/net/inet_common.h b/include/net/inet_common.h index c17a6585d0b0..1666cf6f539e 100644 --- a/include/net/inet_common.h +++ b/include/net/inet_common.h @@ -42,7 +42,7 @@ int inet_shutdown(struct socket *sock, int how); int inet_listen(struct socket *sock, int backlog); int __inet_listen_sk(struct sock *sk, int backlog); void inet_sock_destruct(struct sock *sk); -int inet_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len); +int inet_bind(struct socket *sock, struct sockaddr_unsized *uaddr, int addr_len); int inet_bind_sk(struct sock *sk, struct sockaddr *uaddr, int addr_len); /* Don't allocate port at this moment, defer to connect. */ #define BIND_FORCE_ADDRESS_NO_PORT (1 << 0) diff --git a/include/net/ipv6.h b/include/net/ipv6.h index 2ccdf85f34f1..2188bad9a687 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -1208,7 +1208,7 @@ void ipv6_local_rxpmtu(struct sock *sk, struct flowi6 *fl6, u32 mtu); void inet6_cleanup_sock(struct sock *sk); void inet6_sock_destruct(struct sock *sk); int inet6_release(struct socket *sock); -int inet6_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len); +int inet6_bind(struct socket *sock, struct sockaddr_unsized *uaddr, int addr_len); int inet6_bind_sk(struct sock *sk, struct sockaddr *uaddr, int addr_len); int inet6_getname(struct socket *sock, struct sockaddr *uaddr, int peer); diff --git a/include/net/sock.h b/include/net/sock.h index c7e58b8e8a90..acbb78c96d69 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -1920,7 +1920,7 @@ int sock_cmsg_send(struct sock *sk, struct msghdr *msg, * Functions to fill in entries in struct proto_ops when a protocol * does not implement a particular function. */ -int sock_no_bind(struct socket *, struct sockaddr *, int); +int sock_no_bind(struct socket *sock, struct sockaddr_unsized *saddr, int len); int sock_no_connect(struct socket *, struct sockaddr *, int, int); int sock_no_socketpair(struct socket *, struct socket *); int sock_no_accept(struct socket *, struct socket *, struct proto_accept_arg *); diff --git a/net/9p/trans_fd.c b/net/9p/trans_fd.c index a516745f732f..ef517bb307e2 100644 --- a/net/9p/trans_fd.c +++ b/net/9p/trans_fd.c @@ -966,7 +966,7 @@ static int p9_bind_privport(struct socket *sock) ((struct sockaddr_in *)&stor)->sin_port = htons((ushort)port); else ((struct sockaddr_in6 *)&stor)->sin6_port = htons((ushort)port); - err = kernel_bind(sock, (struct sockaddr *)&stor, sizeof(stor)); + err = kernel_bind(sock, (struct sockaddr_unsized *)&stor, sizeof(stor)); if (err != -EADDRINUSE) break; } diff --git a/net/appletalk/ddp.c b/net/appletalk/ddp.c index 30242fe10341..45db43cde67f 100644 --- a/net/appletalk/ddp.c +++ b/net/appletalk/ddp.c @@ -1149,7 +1149,7 @@ static int atalk_autobind(struct sock *sk) } /* Set the address 'our end' of the connection */ -static int atalk_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) +static int atalk_bind(struct socket *sock, struct sockaddr_unsized *uaddr, int addr_len) { struct sockaddr_at *addr = (struct sockaddr_at *)uaddr; struct sock *sk = sock->sk; diff --git a/net/atm/pvc.c b/net/atm/pvc.c index 66d9a9bd5896..62fdf07c53de 100644 --- a/net/atm/pvc.c +++ b/net/atm/pvc.c @@ -24,7 +24,7 @@ static int pvc_shutdown(struct socket *sock, int how) return 0; } -static int pvc_bind(struct socket *sock, struct sockaddr *sockaddr, +static int pvc_bind(struct socket *sock, struct sockaddr_unsized *sockaddr, int sockaddr_len) { struct sock *sk = sock->sk; @@ -59,7 +59,7 @@ static int pvc_bind(struct socket *sock, struct sockaddr *sockaddr, static int pvc_connect(struct socket *sock, struct sockaddr *sockaddr, int sockaddr_len, int flags) { - return pvc_bind(sock, sockaddr, sockaddr_len); + return pvc_bind(sock, (struct sockaddr_unsized *)sockaddr, sockaddr_len); } static int pvc_setsockopt(struct socket *sock, int level, int optname, diff --git a/net/atm/svc.c b/net/atm/svc.c index f8137ae693b0..1906a493c8aa 100644 --- a/net/atm/svc.c +++ b/net/atm/svc.c @@ -97,7 +97,7 @@ static int svc_release(struct socket *sock) return 0; } -static int svc_bind(struct socket *sock, struct sockaddr *sockaddr, +static int svc_bind(struct socket *sock, struct sockaddr_unsized *sockaddr, int sockaddr_len) { DEFINE_WAIT(wait); diff --git a/net/ax25/af_ax25.c b/net/ax25/af_ax25.c index 6ef8b2a57a9b..23c558ff9682 100644 --- a/net/ax25/af_ax25.c +++ b/net/ax25/af_ax25.c @@ -1094,7 +1094,7 @@ static int ax25_release(struct socket *sock) * that we've implemented support for SO_BINDTODEVICE. It is however small * and trivially backward compatible. */ -static int ax25_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) +static int ax25_bind(struct socket *sock, struct sockaddr_unsized *uaddr, int addr_len) { struct sock *sk = sock->sk; struct full_sockaddr_ax25 *addr = (struct full_sockaddr_ax25 *)uaddr; diff --git a/net/bluetooth/hci_sock.c b/net/bluetooth/hci_sock.c index fc866759910d..ba9f48771e11 100644 --- a/net/bluetooth/hci_sock.c +++ b/net/bluetooth/hci_sock.c @@ -1185,7 +1185,7 @@ static int hci_sock_compat_ioctl(struct socket *sock, unsigned int cmd, } #endif -static int hci_sock_bind(struct socket *sock, struct sockaddr *addr, +static int hci_sock_bind(struct socket *sock, struct sockaddr_unsized *addr, int addr_len) { struct sockaddr_hci haddr; diff --git a/net/bluetooth/iso.c b/net/bluetooth/iso.c index 3d98cb6291da..6a7e1b4a8701 100644 --- a/net/bluetooth/iso.c +++ b/net/bluetooth/iso.c @@ -944,7 +944,7 @@ static int iso_sock_create(struct net *net, struct socket *sock, int protocol, return 0; } -static int iso_sock_bind_bc(struct socket *sock, struct sockaddr *addr, +static int iso_sock_bind_bc(struct socket *sock, struct sockaddr_unsized *addr, int addr_len) { struct sockaddr_iso *sa = (struct sockaddr_iso *)addr; @@ -1022,7 +1022,7 @@ static int iso_sock_bind_pa_sk(struct sock *sk, struct sockaddr_iso *sa, return err; } -static int iso_sock_bind(struct socket *sock, struct sockaddr *addr, +static int iso_sock_bind(struct socket *sock, struct sockaddr_unsized *addr, int addr_len) { struct sockaddr_iso *sa = (struct sockaddr_iso *)addr; diff --git a/net/bluetooth/l2cap_sock.c b/net/bluetooth/l2cap_sock.c index 814fb8610ac4..ca7394d8fa4e 100644 --- a/net/bluetooth/l2cap_sock.c +++ b/net/bluetooth/l2cap_sock.c @@ -80,7 +80,7 @@ static int l2cap_validate_le_psm(u16 psm) return 0; } -static int l2cap_sock_bind(struct socket *sock, struct sockaddr *addr, int alen) +static int l2cap_sock_bind(struct socket *sock, struct sockaddr_unsized *addr, int alen) { struct sock *sk = sock->sk; struct l2cap_chan *chan = l2cap_pi(sk)->chan; diff --git a/net/bluetooth/rfcomm/core.c b/net/bluetooth/rfcomm/core.c index 96250807b32b..d62fd6c57617 100644 --- a/net/bluetooth/rfcomm/core.c +++ b/net/bluetooth/rfcomm/core.c @@ -781,7 +781,7 @@ static struct rfcomm_session *rfcomm_session_create(bdaddr_t *src, addr.l2_psm = 0; addr.l2_cid = 0; addr.l2_bdaddr_type = BDADDR_BREDR; - *err = kernel_bind(sock, (struct sockaddr *) &addr, sizeof(addr)); + *err = kernel_bind(sock, (struct sockaddr_unsized *)&addr, sizeof(addr)); if (*err < 0) goto failed; @@ -2068,7 +2068,7 @@ static int rfcomm_add_listener(bdaddr_t *ba) addr.l2_psm = cpu_to_le16(L2CAP_PSM_RFCOMM); addr.l2_cid = 0; addr.l2_bdaddr_type = BDADDR_BREDR; - err = kernel_bind(sock, (struct sockaddr *) &addr, sizeof(addr)); + err = kernel_bind(sock, (struct sockaddr_unsized *)&addr, sizeof(addr)); if (err < 0) { BT_ERR("Bind failed %d", err); goto failed; diff --git a/net/bluetooth/rfcomm/sock.c b/net/bluetooth/rfcomm/sock.c index 913402806fa0..8c8762bbc6de 100644 --- a/net/bluetooth/rfcomm/sock.c +++ b/net/bluetooth/rfcomm/sock.c @@ -324,7 +324,7 @@ static int rfcomm_sock_create(struct net *net, struct socket *sock, return 0; } -static int rfcomm_sock_bind(struct socket *sock, struct sockaddr *addr, int addr_len) +static int rfcomm_sock_bind(struct socket *sock, struct sockaddr_unsized *addr, int addr_len) { struct sockaddr_rc sa; struct sock *sk = sock->sk; diff --git a/net/bluetooth/sco.c b/net/bluetooth/sco.c index ab0cf442d57b..01d878205e58 100644 --- a/net/bluetooth/sco.c +++ b/net/bluetooth/sco.c @@ -605,7 +605,7 @@ static int sco_sock_create(struct net *net, struct socket *sock, int protocol, return 0; } -static int sco_sock_bind(struct socket *sock, struct sockaddr *addr, +static int sco_sock_bind(struct socket *sock, struct sockaddr_unsized *addr, int addr_len) { struct sockaddr_sco *sa = (struct sockaddr_sco *) addr; diff --git a/net/can/isotp.c b/net/can/isotp.c index 74ee1e52249b..ce588b85665a 100644 --- a/net/can/isotp.c +++ b/net/can/isotp.c @@ -1246,7 +1246,7 @@ static int isotp_release(struct socket *sock) return 0; } -static int isotp_bind(struct socket *sock, struct sockaddr *uaddr, int len) +static int isotp_bind(struct socket *sock, struct sockaddr_unsized *uaddr, int len) { struct sockaddr_can *addr = (struct sockaddr_can *)uaddr; struct sock *sk = sock->sk; diff --git a/net/can/j1939/socket.c b/net/can/j1939/socket.c index 88e7160d4248..a2abedc757d0 100644 --- a/net/can/j1939/socket.c +++ b/net/can/j1939/socket.c @@ -440,7 +440,7 @@ static int j1939_sk_sanity_check(struct sockaddr_can *addr, int len) return 0; } -static int j1939_sk_bind(struct socket *sock, struct sockaddr *uaddr, int len) +static int j1939_sk_bind(struct socket *sock, struct sockaddr_unsized *uaddr, int len) { struct sockaddr_can *addr = (struct sockaddr_can *)uaddr; struct j1939_sock *jsk = j1939_sk(sock->sk); diff --git a/net/can/raw.c b/net/can/raw.c index a53853f5e9af..f36a83d3447c 100644 --- a/net/can/raw.c +++ b/net/can/raw.c @@ -449,7 +449,7 @@ static int raw_release(struct socket *sock) return 0; } -static int raw_bind(struct socket *sock, struct sockaddr *uaddr, int len) +static int raw_bind(struct socket *sock, struct sockaddr_unsized *uaddr, int len) { struct sockaddr_can *addr = (struct sockaddr_can *)uaddr; struct sock *sk = sock->sk; diff --git a/net/core/sock.c b/net/core/sock.c index 7a9bbc2afcf0..1e1ce18bba16 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -3462,7 +3462,7 @@ EXPORT_SYMBOL_GPL(sk_set_peek_off); * function, some default processing is provided. */ -int sock_no_bind(struct socket *sock, struct sockaddr *saddr, int len) +int sock_no_bind(struct socket *sock, struct sockaddr_unsized *saddr, int len) { return -EOPNOTSUPP; } diff --git a/net/ieee802154/socket.c b/net/ieee802154/socket.c index 18d267921bb5..99ddfad9bb88 100644 --- a/net/ieee802154/socket.c +++ b/net/ieee802154/socket.c @@ -96,13 +96,13 @@ static int ieee802154_sock_sendmsg(struct socket *sock, struct msghdr *msg, return sk->sk_prot->sendmsg(sk, msg, len); } -static int ieee802154_sock_bind(struct socket *sock, struct sockaddr *uaddr, +static int ieee802154_sock_bind(struct socket *sock, struct sockaddr_unsized *uaddr, int addr_len) { struct sock *sk = sock->sk; if (sk->sk_prot->bind) - return sk->sk_prot->bind(sk, uaddr, addr_len); + return sk->sk_prot->bind(sk, (struct sockaddr *)uaddr, addr_len); return sock_no_bind(sock, uaddr, addr_len); } diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index 0784e2a873a1..aa43d16e48ff 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -464,9 +464,9 @@ int inet_bind_sk(struct sock *sk, struct sockaddr *uaddr, int addr_len) return __inet_bind(sk, uaddr, addr_len, flags); } -int inet_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) +int inet_bind(struct socket *sock, struct sockaddr_unsized *uaddr, int addr_len) { - return inet_bind_sk(sock->sk, uaddr, addr_len); + return inet_bind_sk(sock->sk, (struct sockaddr *)uaddr, addr_len); } EXPORT_SYMBOL(inet_bind); diff --git a/net/ipv4/udp_tunnel_core.c b/net/ipv4/udp_tunnel_core.c index 54386e06a813..11e5a88c923d 100644 --- a/net/ipv4/udp_tunnel_core.c +++ b/net/ipv4/udp_tunnel_core.c @@ -29,7 +29,7 @@ int udp_sock_create4(struct net *net, struct udp_port_cfg *cfg, udp_addr.sin_family = AF_INET; udp_addr.sin_addr = cfg->local_ip; udp_addr.sin_port = cfg->local_udp_port; - err = kernel_bind(sock, (struct sockaddr *)&udp_addr, + err = kernel_bind(sock, (struct sockaddr_unsized *)&udp_addr, sizeof(udp_addr)); if (err < 0) goto error; diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index 44d7de1eec4f..c92d27e35fbc 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -465,9 +465,9 @@ int inet6_bind_sk(struct sock *sk, struct sockaddr *uaddr, int addr_len) } /* bind for INET6 API */ -int inet6_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) +int inet6_bind(struct socket *sock, struct sockaddr_unsized *uaddr, int addr_len) { - return inet6_bind_sk(sock->sk, uaddr, addr_len); + return inet6_bind_sk(sock->sk, (struct sockaddr *)uaddr, addr_len); } EXPORT_SYMBOL(inet6_bind); diff --git a/net/ipv6/ip6_udp_tunnel.c b/net/ipv6/ip6_udp_tunnel.c index 0ff547a4bff7..b0d9286b33c8 100644 --- a/net/ipv6/ip6_udp_tunnel.c +++ b/net/ipv6/ip6_udp_tunnel.c @@ -40,7 +40,7 @@ int udp_sock_create6(struct net *net, struct udp_port_cfg *cfg, memcpy(&udp6_addr.sin6_addr, &cfg->local_ip6, sizeof(udp6_addr.sin6_addr)); udp6_addr.sin6_port = cfg->local_udp_port; - err = kernel_bind(sock, (struct sockaddr *)&udp6_addr, + err = kernel_bind(sock, (struct sockaddr_unsized *)&udp6_addr, sizeof(udp6_addr)); if (err < 0) goto error; diff --git a/net/iucv/af_iucv.c b/net/iucv/af_iucv.c index 4ddfc633d30c..3941e32cda69 100644 --- a/net/iucv/af_iucv.c +++ b/net/iucv/af_iucv.c @@ -563,7 +563,7 @@ static void __iucv_auto_name(struct iucv_sock *iucv) } /* Bind an unbound socket */ -static int iucv_sock_bind(struct socket *sock, struct sockaddr *addr, +static int iucv_sock_bind(struct socket *sock, struct sockaddr_unsized *addr, int addr_len) { DECLARE_SOCKADDR(struct sockaddr_iucv *, sa, addr); diff --git a/net/l2tp/l2tp_core.c b/net/l2tp/l2tp_core.c index 369a2f2e459c..4b5e372a5cd4 100644 --- a/net/l2tp/l2tp_core.c +++ b/net/l2tp/l2tp_core.c @@ -1503,7 +1503,7 @@ static int l2tp_tunnel_sock_create(struct net *net, memcpy(&ip6_addr.l2tp_addr, cfg->local_ip6, sizeof(ip6_addr.l2tp_addr)); ip6_addr.l2tp_conn_id = tunnel_id; - err = kernel_bind(sock, (struct sockaddr *)&ip6_addr, + err = kernel_bind(sock, (struct sockaddr_unsized *)&ip6_addr, sizeof(ip6_addr)); if (err < 0) goto out; @@ -1530,7 +1530,7 @@ static int l2tp_tunnel_sock_create(struct net *net, ip_addr.l2tp_family = AF_INET; ip_addr.l2tp_addr = cfg->local_ip; ip_addr.l2tp_conn_id = tunnel_id; - err = kernel_bind(sock, (struct sockaddr *)&ip_addr, + err = kernel_bind(sock, (struct sockaddr_unsized *)&ip_addr, sizeof(ip_addr)); if (err < 0) goto out; diff --git a/net/llc/af_llc.c b/net/llc/af_llc.c index 5958a80fe14c..e5bb0c0d708c 100644 --- a/net/llc/af_llc.c +++ b/net/llc/af_llc.c @@ -337,7 +337,7 @@ static int llc_ui_autobind(struct socket *sock, struct sockaddr_llc *addr) * otherwise all hell will break loose. * Returns: 0 upon success, negative otherwise. */ -static int llc_ui_bind(struct socket *sock, struct sockaddr *uaddr, int addrlen) +static int llc_ui_bind(struct socket *sock, struct sockaddr_unsized *uaddr, int addrlen) { struct sockaddr_llc *addr = (struct sockaddr_llc *)uaddr; struct sock *sk = sock->sk; diff --git a/net/mctp/af_mctp.c b/net/mctp/af_mctp.c index b99ba14f39d2..5b1ef50637b7 100644 --- a/net/mctp/af_mctp.c +++ b/net/mctp/af_mctp.c @@ -49,7 +49,7 @@ static bool mctp_sockaddr_ext_is_ok(const struct sockaddr_mctp_ext *addr) !addr->__smctp_pad0[2]; } -static int mctp_bind(struct socket *sock, struct sockaddr *addr, int addrlen) +static int mctp_bind(struct socket *sock, struct sockaddr_unsized *addr, int addrlen) { struct sock *sk = sock->sk; struct mctp_sock *msk = container_of(sk, struct mctp_sock, sk); diff --git a/net/mctp/test/route-test.c b/net/mctp/test/route-test.c index 69a3ccfc6310..be9149ac79dd 100644 --- a/net/mctp/test/route-test.c +++ b/net/mctp/test/route-test.c @@ -205,7 +205,7 @@ static void __mctp_route_test_init(struct kunit *test, addr.smctp_network = netid; addr.smctp_addr.s_addr = 8; addr.smctp_type = 0; - rc = kernel_bind(sock, (struct sockaddr *)&addr, sizeof(addr)); + rc = kernel_bind(sock, (struct sockaddr_unsized *)&addr, sizeof(addr)); KUNIT_ASSERT_EQ(test, rc, 0); *devp = dev; diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index d568575cdcb5..53e2b095dfb1 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -3856,7 +3856,7 @@ static struct proto mptcp_prot = { .no_autobind = true, }; -static int mptcp_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) +static int mptcp_bind(struct socket *sock, struct sockaddr_unsized *uaddr, int addr_len) { struct mptcp_sock *msk = mptcp_sk(sock->sk); struct sock *ssk, *sk = sock->sk; @@ -3870,10 +3870,10 @@ static int mptcp_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) } if (sk->sk_family == AF_INET) - err = inet_bind_sk(ssk, uaddr, addr_len); + err = inet_bind_sk(ssk, (struct sockaddr *)uaddr, addr_len); #if IS_ENABLED(CONFIG_MPTCP_IPV6) else if (sk->sk_family == AF_INET6) - err = inet6_bind_sk(ssk, uaddr, addr_len); + err = inet6_bind_sk(ssk, (struct sockaddr *)uaddr, addr_len); #endif if (!err) mptcp_copy_inaddrs(sk, ssk); diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c index e8325890a322..d90237bf433c 100644 --- a/net/mptcp/subflow.c +++ b/net/mptcp/subflow.c @@ -1660,7 +1660,7 @@ int __mptcp_subflow_connect(struct sock *sk, const struct mptcp_pm_local *local, addrlen = sizeof(struct sockaddr_in6); #endif ssk->sk_bound_dev_if = local->ifindex; - err = kernel_bind(sf, (struct sockaddr *)&addr, addrlen); + err = kernel_bind(sf, (struct sockaddr_unsized *)&addr, addrlen); if (err) { MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_JOINSYNTXBINDERR); pr_debug("msk=%p local=%d remote=%d bind error: %d\n", diff --git a/net/netfilter/ipvs/ip_vs_sync.c b/net/netfilter/ipvs/ip_vs_sync.c index 3402675bf521..d8c089ef387c 100644 --- a/net/netfilter/ipvs/ip_vs_sync.c +++ b/net/netfilter/ipvs/ip_vs_sync.c @@ -1435,7 +1435,7 @@ static int bind_mcastif_addr(struct socket *sock, struct net_device *dev) sin.sin_addr.s_addr = addr; sin.sin_port = 0; - return kernel_bind(sock, (struct sockaddr *)&sin, sizeof(sin)); + return kernel_bind(sock, (struct sockaddr_unsized *)&sin, sizeof(sin)); } static void get_mcast_sockaddr(union ipvs_sockaddr *sa, int *salen, @@ -1542,7 +1542,7 @@ static int make_receive_sock(struct netns_ipvs *ipvs, int id, get_mcast_sockaddr(&mcast_addr, &salen, &ipvs->bcfg, id); sock->sk->sk_bound_dev_if = dev->ifindex; - result = kernel_bind(sock, (struct sockaddr *)&mcast_addr, salen); + result = kernel_bind(sock, (struct sockaddr_unsized *)&mcast_addr, salen); if (result < 0) { pr_err("Error binding to the multicast addr\n"); goto error; diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index 687a84c48882..18490a56edd0 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -966,7 +966,7 @@ static void netlink_undo_bind(int group, long unsigned int groups, nlk->netlink_unbind(sock_net(sk), undo + 1); } -static int netlink_bind(struct socket *sock, struct sockaddr *addr, +static int netlink_bind(struct socket *sock, struct sockaddr_unsized *addr, int addr_len) { struct sock *sk = sock->sk; diff --git a/net/netrom/af_netrom.c b/net/netrom/af_netrom.c index 3331669d8e33..33468124d53d 100644 --- a/net/netrom/af_netrom.c +++ b/net/netrom/af_netrom.c @@ -561,7 +561,7 @@ static int nr_release(struct socket *sock) return 0; } -static int nr_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) +static int nr_bind(struct socket *sock, struct sockaddr_unsized *uaddr, int addr_len) { struct sock *sk = sock->sk; struct nr_sock *nr = nr_sk(sk); diff --git a/net/nfc/llcp_sock.c b/net/nfc/llcp_sock.c index 57a2f97004e1..26e6ceb48a82 100644 --- a/net/nfc/llcp_sock.c +++ b/net/nfc/llcp_sock.c @@ -56,7 +56,7 @@ static struct proto llcp_sock_proto = { .obj_size = sizeof(struct nfc_llcp_sock), }; -static int llcp_sock_bind(struct socket *sock, struct sockaddr *addr, int alen) +static int llcp_sock_bind(struct socket *sock, struct sockaddr_unsized *addr, int alen) { struct sock *sk = sock->sk; struct nfc_llcp_sock *llcp_sock = nfc_llcp_sock(sk); @@ -146,7 +146,7 @@ static int llcp_sock_bind(struct socket *sock, struct sockaddr *addr, int alen) return ret; } -static int llcp_raw_sock_bind(struct socket *sock, struct sockaddr *addr, +static int llcp_raw_sock_bind(struct socket *sock, struct sockaddr_unsized *addr, int alen) { struct sock *sk = sock->sk; diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index 173e6edda08f..fccad2a529cc 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -3279,11 +3279,12 @@ static int packet_do_bind(struct sock *sk, const char *name, int ifindex, * Bind a packet socket to a device */ -static int packet_bind_spkt(struct socket *sock, struct sockaddr *uaddr, +static int packet_bind_spkt(struct socket *sock, struct sockaddr_unsized *uaddr, int addr_len) { struct sock *sk = sock->sk; - char name[sizeof(uaddr->sa_data_min) + 1]; + struct sockaddr *sa = (struct sockaddr *)uaddr; + char name[sizeof(sa->sa_data_min) + 1]; /* * Check legality @@ -3294,13 +3295,13 @@ static int packet_bind_spkt(struct socket *sock, struct sockaddr *uaddr, /* uaddr->sa_data comes from the userspace, it's not guaranteed to be * zero-terminated. */ - memcpy(name, uaddr->sa_data, sizeof(uaddr->sa_data_min)); - name[sizeof(uaddr->sa_data_min)] = 0; + memcpy(name, sa->sa_data, sizeof(sa->sa_data_min)); + name[sizeof(sa->sa_data_min)] = 0; return packet_do_bind(sk, name, 0, 0); } -static int packet_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) +static int packet_bind(struct socket *sock, struct sockaddr_unsized *uaddr, int addr_len) { struct sockaddr_ll *sll = (struct sockaddr_ll *)uaddr; struct sock *sk = sock->sk; diff --git a/net/phonet/socket.c b/net/phonet/socket.c index db2d552e9b32..478b02647733 100644 --- a/net/phonet/socket.c +++ b/net/phonet/socket.c @@ -153,7 +153,7 @@ EXPORT_SYMBOL(pn_sock_unhash); static DEFINE_MUTEX(port_mutex); -static int pn_socket_bind(struct socket *sock, struct sockaddr *addr, int len) +static int pn_socket_bind(struct socket *sock, struct sockaddr_unsized *addr, int len) { struct sock *sk = sock->sk; struct pn_sock *pn = pn_sk(sk); @@ -163,7 +163,7 @@ static int pn_socket_bind(struct socket *sock, struct sockaddr *addr, int len) u8 saddr; if (sk->sk_prot->bind) - return sk->sk_prot->bind(sk, addr, len); + return sk->sk_prot->bind(sk, (struct sockaddr *)addr, len); if (len < sizeof(struct sockaddr_pn)) return -EINVAL; @@ -206,8 +206,8 @@ static int pn_socket_autobind(struct socket *sock) memset(&sa, 0, sizeof(sa)); sa.spn_family = AF_PHONET; - err = pn_socket_bind(sock, (struct sockaddr *)&sa, - sizeof(struct sockaddr_pn)); + err = pn_socket_bind(sock, (struct sockaddr_unsized *)&sa, + sizeof(struct sockaddr_pn)); if (err != -EINVAL) return err; BUG_ON(!pn_port(pn_sk(sock->sk)->sobject)); diff --git a/net/qrtr/af_qrtr.c b/net/qrtr/af_qrtr.c index 00c51cf693f3..00bd3dd9f0f9 100644 --- a/net/qrtr/af_qrtr.c +++ b/net/qrtr/af_qrtr.c @@ -824,7 +824,7 @@ static int qrtr_autobind(struct socket *sock) } /* Bind socket to specified sockaddr. */ -static int qrtr_bind(struct socket *sock, struct sockaddr *saddr, int len) +static int qrtr_bind(struct socket *sock, struct sockaddr_unsized *saddr, int len) { DECLARE_SOCKADDR(struct sockaddr_qrtr *, addr, saddr); struct qrtr_sock *ipc = qrtr_sk(sock->sk); diff --git a/net/qrtr/ns.c b/net/qrtr/ns.c index 3de9350cbf30..bfcc1a453f23 100644 --- a/net/qrtr/ns.c +++ b/net/qrtr/ns.c @@ -714,7 +714,7 @@ int qrtr_ns_init(void) sq.sq_port = QRTR_PORT_CTRL; qrtr_ns.local_node = sq.sq_node; - ret = kernel_bind(qrtr_ns.sock, (struct sockaddr *)&sq, sizeof(sq)); + ret = kernel_bind(qrtr_ns.sock, (struct sockaddr_unsized *)&sq, sizeof(sq)); if (ret < 0) { pr_err("failed to bind to socket\n"); goto err_wq; diff --git a/net/rds/bind.c b/net/rds/bind.c index 97a29172a8ee..f800d920d969 100644 --- a/net/rds/bind.c +++ b/net/rds/bind.c @@ -160,7 +160,7 @@ void rds_remove_bound(struct rds_sock *rs) rs->rs_bound_addr = in6addr_any; } -int rds_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) +int rds_bind(struct socket *sock, struct sockaddr_unsized *uaddr, int addr_len) { struct sock *sk = sock->sk; struct rds_sock *rs = rds_sk_to_rs(sk); diff --git a/net/rds/rds.h b/net/rds/rds.h index 5b1c072e2e7f..a029e5fcdea7 100644 --- a/net/rds/rds.h +++ b/net/rds/rds.h @@ -735,7 +735,7 @@ extern wait_queue_head_t rds_poll_waitq; /* bind.c */ -int rds_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len); +int rds_bind(struct socket *sock, struct sockaddr_unsized *uaddr, int addr_len); void rds_remove_bound(struct rds_sock *rs); struct rds_sock *rds_find_bound(const struct in6_addr *addr, __be16 port, __u32 scope_id); diff --git a/net/rds/tcp_connect.c b/net/rds/tcp_connect.c index a0046e99d6df..1eff3b03ab77 100644 --- a/net/rds/tcp_connect.c +++ b/net/rds/tcp_connect.c @@ -145,7 +145,7 @@ int rds_tcp_conn_path_connect(struct rds_conn_path *cp) addrlen = sizeof(sin); } - ret = kernel_bind(sock, addr, addrlen); + ret = kernel_bind(sock, (struct sockaddr_unsized *)addr, addrlen); if (ret) { rdsdebug("bind failed with %d at address %pI6c\n", ret, &conn->c_laddr); diff --git a/net/rds/tcp_listen.c b/net/rds/tcp_listen.c index 91e34af3fe5d..820d3e20de19 100644 --- a/net/rds/tcp_listen.c +++ b/net/rds/tcp_listen.c @@ -290,7 +290,7 @@ struct socket *rds_tcp_listen_init(struct net *net, bool isv6) addr_len = sizeof(*sin); } - ret = kernel_bind(sock, (struct sockaddr *)&ss, addr_len); + ret = kernel_bind(sock, (struct sockaddr_unsized *)&ss, addr_len); if (ret < 0) { rdsdebug("could not bind %s listener socket: %d\n", isv6 ? "IPv6" : "IPv4", ret); diff --git a/net/rose/af_rose.c b/net/rose/af_rose.c index 543f9e8ebb69..47369eab5aec 100644 --- a/net/rose/af_rose.c +++ b/net/rose/af_rose.c @@ -693,7 +693,7 @@ static int rose_release(struct socket *sock) return 0; } -static int rose_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) +static int rose_bind(struct socket *sock, struct sockaddr_unsized *uaddr, int addr_len) { struct sock *sk = sock->sk; struct rose_sock *rose = rose_sk(sk); diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c index 36df0274d7b7..245f37a74394 100644 --- a/net/rxrpc/af_rxrpc.c +++ b/net/rxrpc/af_rxrpc.c @@ -127,7 +127,7 @@ static int rxrpc_validate_address(struct rxrpc_sock *rx, /* * bind a local address to an RxRPC socket */ -static int rxrpc_bind(struct socket *sock, struct sockaddr *saddr, int len) +static int rxrpc_bind(struct socket *sock, struct sockaddr_unsized *saddr, int len) { struct sockaddr_rxrpc *srx = (struct sockaddr_rxrpc *)saddr; struct rxrpc_local *local; diff --git a/net/rxrpc/rxperf.c b/net/rxrpc/rxperf.c index 2ea71e3831f7..98ea76fae70f 100644 --- a/net/rxrpc/rxperf.c +++ b/net/rxrpc/rxperf.c @@ -211,7 +211,7 @@ static int rxperf_open_socket(void) ret = rxrpc_sock_set_security_keyring(socket->sk, rxperf_sec_keyring); - ret = kernel_bind(socket, (struct sockaddr *)&srx, sizeof(srx)); + ret = kernel_bind(socket, (struct sockaddr_unsized *)&srx, sizeof(srx)); if (ret < 0) goto error_2; diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c index e9d0e62e0b1b..be18ab08f15d 100644 --- a/net/smc/af_smc.c +++ b/net/smc/af_smc.c @@ -421,7 +421,7 @@ static struct sock *smc_sock_alloc(struct net *net, struct socket *sock, return sk; } -int smc_bind(struct socket *sock, struct sockaddr *uaddr, +int smc_bind(struct socket *sock, struct sockaddr_unsized *uaddr, int addr_len) { struct sockaddr_in *addr = (struct sockaddr_in *)uaddr; diff --git a/net/smc/smc.h b/net/smc/smc.h index 2c9084963739..a008dbe6d6f6 100644 --- a/net/smc/smc.h +++ b/net/smc/smc.h @@ -42,7 +42,7 @@ void smc_unhash_sk(struct sock *sk); void smc_release_cb(struct sock *sk); int smc_release(struct socket *sock); -int smc_bind(struct socket *sock, struct sockaddr *uaddr, +int smc_bind(struct socket *sock, struct sockaddr_unsized *uaddr, int addr_len); int smc_connect(struct socket *sock, struct sockaddr *addr, int alen, int flags); diff --git a/net/socket.c b/net/socket.c index e8892b218708..aaefb2e519a7 100644 --- a/net/socket.c +++ b/net/socket.c @@ -1872,7 +1872,7 @@ int __sys_bind_socket(struct socket *sock, struct sockaddr_storage *address, addrlen); if (!err) err = READ_ONCE(sock->ops)->bind(sock, - (struct sockaddr *)address, + (struct sockaddr_unsized *)address, addrlen); return err; } @@ -3583,13 +3583,13 @@ static long compat_sock_ioctl(struct file *file, unsigned int cmd, * Returns 0 or an error. */ -int kernel_bind(struct socket *sock, struct sockaddr *addr, int addrlen) +int kernel_bind(struct socket *sock, struct sockaddr_unsized *addr, int addrlen) { struct sockaddr_storage address; memcpy(&address, addr, addrlen); - return READ_ONCE(sock->ops)->bind(sock, (struct sockaddr *)&address, + return READ_ONCE(sock->ops)->bind(sock, (struct sockaddr_unsized *)&address, addrlen); } EXPORT_SYMBOL(kernel_bind); diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c index 8ca354ecfd02..318ee24ad900 100644 --- a/net/sunrpc/clnt.c +++ b/net/sunrpc/clnt.c @@ -1457,12 +1457,12 @@ static int rpc_sockname(struct net *net, struct sockaddr *sap, size_t salen, switch (sap->sa_family) { case AF_INET: err = kernel_bind(sock, - (struct sockaddr *)&rpc_inaddr_loopback, + (struct sockaddr_unsized *)&rpc_inaddr_loopback, sizeof(rpc_inaddr_loopback)); break; case AF_INET6: err = kernel_bind(sock, - (struct sockaddr *)&rpc_in6addr_loopback, + (struct sockaddr_unsized *)&rpc_in6addr_loopback, sizeof(rpc_in6addr_loopback)); break; default: diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c index 7b90abc5cf0e..16ff6c100821 100644 --- a/net/sunrpc/svcsock.c +++ b/net/sunrpc/svcsock.c @@ -1557,7 +1557,7 @@ static struct svc_xprt *svc_create_socket(struct svc_serv *serv, ip6_sock_set_v6only(sock->sk); if (type == SOCK_STREAM) sock->sk->sk_reuse = SK_CAN_REUSE; /* allow address reuse */ - error = kernel_bind(sock, sin, len); + error = kernel_bind(sock, (struct sockaddr_unsized *)sin, len); if (error < 0) goto bummer; diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index 3aa987e7f072..95732a45b059 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -1845,8 +1845,8 @@ static int xs_bind(struct sock_xprt *transport, struct socket *sock) memcpy(&myaddr, &transport->srcaddr, transport->xprt.addrlen); do { rpc_set_port((struct sockaddr *)&myaddr, port); - err = kernel_bind(sock, (struct sockaddr *)&myaddr, - transport->xprt.addrlen); + err = kernel_bind(sock, (struct sockaddr_unsized *)&myaddr, + transport->xprt.addrlen); if (err == 0) { if (transport->xprt.reuseport) transport->srcport = port; diff --git a/net/tipc/socket.c b/net/tipc/socket.c index bc614a1f019c..3903a97ada7d 100644 --- a/net/tipc/socket.c +++ b/net/tipc/socket.c @@ -710,7 +710,7 @@ int tipc_sk_bind(struct socket *sock, struct sockaddr *skaddr, int alen) return res; } -static int tipc_bind(struct socket *sock, struct sockaddr *skaddr, int alen) +static int tipc_bind(struct socket *sock, struct sockaddr_unsized *skaddr, int alen) { struct tipc_uaddr *ua = (struct tipc_uaddr *)skaddr; u32 atype = ua->addrtype; @@ -726,7 +726,7 @@ static int tipc_bind(struct socket *sock, struct sockaddr *skaddr, int alen) return -EACCES; } } - return tipc_sk_bind(sock, skaddr, alen); + return tipc_sk_bind(sock, (struct sockaddr *)skaddr, alen); } /** diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index 54177caa9c12..788775f0eea7 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -843,7 +843,7 @@ static int unix_listen(struct socket *sock, int backlog) } static int unix_release(struct socket *); -static int unix_bind(struct socket *, struct sockaddr *, int); +static int unix_bind(struct socket *, struct sockaddr_unsized *, int); static int unix_stream_connect(struct socket *, struct sockaddr *, int addr_len, int flags); static int unix_socketpair(struct socket *, struct socket *); @@ -1466,7 +1466,7 @@ static int unix_bind_abstract(struct sock *sk, struct sockaddr_un *sunaddr, return err; } -static int unix_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) +static int unix_bind(struct socket *sock, struct sockaddr_unsized *uaddr, int addr_len) { struct sockaddr_un *sunaddr = (struct sockaddr_un *)uaddr; struct sock *sk = sock->sk; diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c index 76763247a377..0e5609e7284b 100644 --- a/net/vmw_vsock/af_vsock.c +++ b/net/vmw_vsock/af_vsock.c @@ -987,7 +987,7 @@ static int vsock_release(struct socket *sock) } static int -vsock_bind(struct socket *sock, struct sockaddr *addr, int addr_len) +vsock_bind(struct socket *sock, struct sockaddr_unsized *addr, int addr_len) { int err; struct sock *sk; @@ -995,7 +995,7 @@ vsock_bind(struct socket *sock, struct sockaddr *addr, int addr_len) sk = sock->sk; - if (vsock_addr_cast(addr, addr_len, &vm_addr) != 0) + if (vsock_addr_cast((struct sockaddr *)addr, addr_len, &vm_addr) != 0) return -EINVAL; lock_sock(sk); diff --git a/net/x25/af_x25.c b/net/x25/af_x25.c index 655d1e0ae25f..ca8006d8f792 100644 --- a/net/x25/af_x25.c +++ b/net/x25/af_x25.c @@ -670,7 +670,7 @@ static int x25_release(struct socket *sock) return 0; } -static int x25_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) +static int x25_bind(struct socket *sock, struct sockaddr_unsized *uaddr, int addr_len) { struct sock *sk = sock->sk; struct sockaddr_x25 *addr = (struct sockaddr_x25 *)uaddr; diff --git a/net/xdp/xsk.c b/net/xdp/xsk.c index 2f26c918d448..ed8b612ec29d 100644 --- a/net/xdp/xsk.c +++ b/net/xdp/xsk.c @@ -1238,7 +1238,7 @@ static bool xsk_validate_queues(struct xdp_sock *xs) return xs->fq_tmp && xs->cq_tmp; } -static int xsk_bind(struct socket *sock, struct sockaddr *addr, int addr_len) +static int xsk_bind(struct socket *sock, struct sockaddr_unsized *addr, int addr_len) { struct sockaddr_xdp *sxdp = (struct sockaddr_xdp *)addr; struct sock *sk = sock->sk; diff --git a/tools/testing/selftests/bpf/test_kmods/bpf_testmod.c b/tools/testing/selftests/bpf/test_kmods/bpf_testmod.c index 8074bc5f6f20..0497b5dea25c 100644 --- a/tools/testing/selftests/bpf/test_kmods/bpf_testmod.c +++ b/tools/testing/selftests/bpf/test_kmods/bpf_testmod.c @@ -923,7 +923,7 @@ __bpf_kfunc int bpf_kfunc_call_kernel_bind(struct addr_args *args) goto out; } - err = kernel_bind(sock, (struct sockaddr *)&args->addr, args->addrlen); + err = kernel_bind(sock, (struct sockaddr_unsized *)&args->addr, args->addrlen); out: mutex_unlock(&sock_lock); From 85cb0757d7e1f9370a8b52a8b8144c37941cba0a Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Mon, 3 Nov 2025 16:26:11 -0800 Subject: [PATCH 0534/1381] net: Convert proto_ops connect() callbacks to use sockaddr_unsized Update all struct proto_ops connect() callback function prototypes from "struct sockaddr *" to "struct sockaddr_unsized *" to avoid lying to the compiler about object sizes. Calls into struct proto handlers gain casts that will be removed in the struct proto conversion patch. No binary changes expected. Signed-off-by: Kees Cook Link: https://patch.msgid.link/20251104002617.2752303-3-kees@kernel.org Signed-off-by: Jakub Kicinski --- drivers/block/drbd/drbd_receiver.c | 2 +- drivers/infiniband/hw/erdma/erdma_cm.c | 2 +- drivers/infiniband/sw/siw/siw_cm.c | 2 +- drivers/net/ppp/pppoe.c | 4 ++-- drivers/net/ppp/pptp.c | 4 ++-- drivers/net/wireless/ath/ath10k/qmi.c | 2 +- drivers/net/wireless/ath/ath11k/qmi.c | 2 +- drivers/net/wireless/ath/ath12k/qmi.c | 2 +- drivers/nvme/host/tcp.c | 2 +- drivers/slimbus/qcom-ngd-ctrl.c | 2 +- drivers/xen/pvcalls-back.c | 2 +- fs/coredump.c | 2 +- fs/dlm/lowcomms.c | 2 +- fs/ocfs2/cluster/tcp.c | 2 +- fs/smb/client/connect.c | 2 +- include/linux/bpf-cgroup.h | 6 +++--- include/linux/net.h | 4 ++-- include/net/inet_common.h | 6 +++--- include/net/sctp/sctp.h | 2 +- include/net/sock.h | 2 +- include/net/vsock_addr.h | 2 +- net/9p/trans_fd.c | 6 +++--- net/appletalk/ddp.c | 2 +- net/atm/pvc.c | 4 ++-- net/atm/svc.c | 2 +- net/ax25/af_ax25.c | 2 +- net/bluetooth/iso.c | 2 +- net/bluetooth/l2cap_sock.c | 2 +- net/bluetooth/rfcomm/core.c | 2 +- net/bluetooth/rfcomm/sock.c | 3 ++- net/bluetooth/sco.c | 2 +- net/caif/caif_socket.c | 2 +- net/can/bcm.c | 2 +- net/can/j1939/socket.c | 2 +- net/ceph/messenger.c | 2 +- net/core/sock.c | 2 +- net/ieee802154/socket.c | 4 ++-- net/ipv4/af_inet.c | 14 +++++++------- net/ipv4/tcp.c | 2 +- net/ipv4/udp_tunnel_core.c | 2 +- net/ipv6/ip6_udp_tunnel.c | 2 +- net/iucv/af_iucv.c | 4 ++-- net/l2tp/l2tp_core.c | 4 ++-- net/l2tp/l2tp_ppp.c | 2 +- net/llc/af_llc.c | 2 +- net/mctp/af_mctp.c | 2 +- net/mctp/test/utils.c | 5 +++-- net/mptcp/subflow.c | 2 +- net/netfilter/ipvs/ip_vs_sync.c | 2 +- net/netlink/af_netlink.c | 2 +- net/netrom/af_netrom.c | 4 ++-- net/nfc/llcp_sock.c | 2 +- net/nfc/rawsock.c | 2 +- net/phonet/socket.c | 6 +++--- net/qrtr/af_qrtr.c | 2 +- net/rds/af_rds.c | 2 +- net/rds/tcp_connect.c | 2 +- net/rose/af_rose.c | 3 ++- net/rxrpc/af_rxrpc.c | 2 +- net/sctp/socket.c | 4 ++-- net/smc/af_smc.c | 4 ++-- net/smc/smc.h | 2 +- net/socket.c | 8 ++++---- net/sunrpc/clnt.c | 2 +- net/sunrpc/xprtsock.c | 5 +++-- net/tipc/socket.c | 2 +- net/unix/af_unix.c | 8 ++++---- net/vmw_vsock/af_vsock.c | 6 +++--- net/vmw_vsock/vsock_addr.c | 2 +- net/x25/af_x25.c | 2 +- samples/qmi/qmi_sample_client.c | 2 +- .../testing/selftests/bpf/test_kmods/bpf_testmod.c | 2 +- 72 files changed, 110 insertions(+), 106 deletions(-) diff --git a/drivers/block/drbd/drbd_receiver.c b/drivers/block/drbd/drbd_receiver.c index d9296f74f902..33bc91665fe8 100644 --- a/drivers/block/drbd/drbd_receiver.c +++ b/drivers/block/drbd/drbd_receiver.c @@ -458,7 +458,7 @@ static struct socket *drbd_try_connect(struct drbd_connection *connection) * stay C_WF_CONNECTION, don't go Disconnecting! */ disconnect_on_error = 0; what = "connect"; - err = sock->ops->connect(sock, (struct sockaddr *) &peer_in6, peer_addr_len, 0); + err = sock->ops->connect(sock, (struct sockaddr_unsized *) &peer_in6, peer_addr_len, 0); out: if (err < 0) { diff --git a/drivers/infiniband/hw/erdma/erdma_cm.c b/drivers/infiniband/hw/erdma/erdma_cm.c index ef66a6359eb9..ed21ba0037a4 100644 --- a/drivers/infiniband/hw/erdma/erdma_cm.c +++ b/drivers/infiniband/hw/erdma/erdma_cm.c @@ -996,7 +996,7 @@ static int kernel_bindconnect(struct socket *s, struct sockaddr *laddr, ret = s->ops->bind(s, (struct sockaddr_unsized *)laddr, laddrlen); if (ret) return ret; - ret = s->ops->connect(s, raddr, raddrlen, flags); + ret = s->ops->connect(s, (struct sockaddr_unsized *)raddr, raddrlen, flags); return ret < 0 ? ret : 0; } diff --git a/drivers/infiniband/sw/siw/siw_cm.c b/drivers/infiniband/sw/siw/siw_cm.c index 7fe118cacb3f..eb0bd4f79a85 100644 --- a/drivers/infiniband/sw/siw/siw_cm.c +++ b/drivers/infiniband/sw/siw/siw_cm.c @@ -1344,7 +1344,7 @@ static int kernel_bindconnect(struct socket *s, struct sockaddr *laddr, if (rv < 0) return rv; - rv = s->ops->connect(s, raddr, size, flags); + rv = s->ops->connect(s, (struct sockaddr_unsized *)raddr, size, flags); return rv < 0 ? rv : 0; } diff --git a/drivers/net/ppp/pppoe.c b/drivers/net/ppp/pppoe.c index 4ac6afce267b..4275b393a454 100644 --- a/drivers/net/ppp/pppoe.c +++ b/drivers/net/ppp/pppoe.c @@ -608,8 +608,8 @@ static int pppoe_release(struct socket *sock) return 0; } -static int pppoe_connect(struct socket *sock, struct sockaddr *uservaddr, - int sockaddr_len, int flags) +static int pppoe_connect(struct socket *sock, struct sockaddr_unsized *uservaddr, + int sockaddr_len, int flags) { struct sock *sk = sock->sk; struct sockaddr_pppox *sp = (struct sockaddr_pppox *)uservaddr; diff --git a/drivers/net/ppp/pptp.c b/drivers/net/ppp/pptp.c index d07e87a0974c..b18acd810561 100644 --- a/drivers/net/ppp/pptp.c +++ b/drivers/net/ppp/pptp.c @@ -415,8 +415,8 @@ static int pptp_bind(struct socket *sock, struct sockaddr_unsized *uservaddr, return error; } -static int pptp_connect(struct socket *sock, struct sockaddr *uservaddr, - int sockaddr_len, int flags) +static int pptp_connect(struct socket *sock, struct sockaddr_unsized *uservaddr, + int sockaddr_len, int flags) { struct sock *sk = sock->sk; struct sockaddr_pppox *sp = (struct sockaddr_pppox *) uservaddr; diff --git a/drivers/net/wireless/ath/ath10k/qmi.c b/drivers/net/wireless/ath/ath10k/qmi.c index f1f33af0170a..8275345631a0 100644 --- a/drivers/net/wireless/ath/ath10k/qmi.c +++ b/drivers/net/wireless/ath/ath10k/qmi.c @@ -986,7 +986,7 @@ static int ath10k_qmi_new_server(struct qmi_handle *qmi_hdl, ath10k_dbg(ar, ATH10K_DBG_QMI, "wifi fw qmi service found\n"); - ret = kernel_connect(qmi_hdl->sock, (struct sockaddr *)&qmi->sq, + ret = kernel_connect(qmi_hdl->sock, (struct sockaddr_unsized *)&qmi->sq, sizeof(qmi->sq), 0); if (ret) { ath10k_err(ar, "failed to connect to a remote QMI service port\n"); diff --git a/drivers/net/wireless/ath/ath11k/qmi.c b/drivers/net/wireless/ath/ath11k/qmi.c index aea56c38bf8f..ff6a97e328b8 100644 --- a/drivers/net/wireless/ath/ath11k/qmi.c +++ b/drivers/net/wireless/ath/ath11k/qmi.c @@ -3177,7 +3177,7 @@ static int ath11k_qmi_ops_new_server(struct qmi_handle *qmi_hdl, sq->sq_node = service->node; sq->sq_port = service->port; - ret = kernel_connect(qmi_hdl->sock, (struct sockaddr *)sq, + ret = kernel_connect(qmi_hdl->sock, (struct sockaddr_unsized *)sq, sizeof(*sq), 0); if (ret) { ath11k_warn(ab, "failed to connect to qmi remote service: %d\n", ret); diff --git a/drivers/net/wireless/ath/ath12k/qmi.c b/drivers/net/wireless/ath/ath12k/qmi.c index 36325e62aa24..cf9c25df3ffd 100644 --- a/drivers/net/wireless/ath/ath12k/qmi.c +++ b/drivers/net/wireless/ath/ath12k/qmi.c @@ -3740,7 +3740,7 @@ static int ath12k_qmi_ops_new_server(struct qmi_handle *qmi_hdl, sq->sq_node = service->node; sq->sq_port = service->port; - ret = kernel_connect(qmi_hdl->sock, (struct sockaddr *)sq, + ret = kernel_connect(qmi_hdl->sock, (struct sockaddr_unsized *)sq, sizeof(*sq), 0); if (ret) { ath12k_warn(ab, "qmi failed to connect to remote service %d\n", ret); diff --git a/drivers/nvme/host/tcp.c b/drivers/nvme/host/tcp.c index 35d0bd91f6fd..6795b8286c35 100644 --- a/drivers/nvme/host/tcp.c +++ b/drivers/nvme/host/tcp.c @@ -1872,7 +1872,7 @@ static int nvme_tcp_alloc_queue(struct nvme_ctrl *nctrl, int qid, dev_dbg(nctrl->device, "connecting queue %d\n", nvme_tcp_queue_id(queue)); - ret = kernel_connect(queue->sock, (struct sockaddr *)&ctrl->addr, + ret = kernel_connect(queue->sock, (struct sockaddr_unsized *)&ctrl->addr, sizeof(ctrl->addr), 0); if (ret) { dev_err(nctrl->device, diff --git a/drivers/slimbus/qcom-ngd-ctrl.c b/drivers/slimbus/qcom-ngd-ctrl.c index 4fb66986cc22..fdb94dc4a730 100644 --- a/drivers/slimbus/qcom-ngd-ctrl.c +++ b/drivers/slimbus/qcom-ngd-ctrl.c @@ -463,7 +463,7 @@ static int qcom_slim_qmi_init(struct qcom_slim_ngd_ctrl *ctrl, } rc = kernel_connect(handle->sock, - (struct sockaddr *)&ctrl->qmi.svc_info, + (struct sockaddr_unsized *)&ctrl->qmi.svc_info, sizeof(ctrl->qmi.svc_info), 0); if (rc < 0) { dev_err(ctrl->dev, "Remote Service connect failed: %d\n", rc); diff --git a/drivers/xen/pvcalls-back.c b/drivers/xen/pvcalls-back.c index da1b516b9cfd..c5b6f6fa11eb 100644 --- a/drivers/xen/pvcalls-back.c +++ b/drivers/xen/pvcalls-back.c @@ -409,7 +409,7 @@ static int pvcalls_back_connect(struct xenbus_device *dev, ret = sock_create(AF_INET, SOCK_STREAM, 0, &sock); if (ret < 0) goto out; - ret = inet_stream_connect(sock, sa, req->u.connect.len, 0); + ret = inet_stream_connect(sock, (struct sockaddr_unsized *)sa, req->u.connect.len, 0); if (ret < 0) { sock_release(sock); goto out; diff --git a/fs/coredump.c b/fs/coredump.c index 5c1c381ee380..14837d9e2abb 100644 --- a/fs/coredump.c +++ b/fs/coredump.c @@ -708,7 +708,7 @@ static bool coredump_sock_connect(struct core_name *cn, struct coredump_params * */ pidfs_coredump(cprm); - retval = kernel_connect(socket, (struct sockaddr *)(&addr), addr_len, + retval = kernel_connect(socket, (struct sockaddr_unsized *)(&addr), addr_len, O_NONBLOCK | SOCK_COREDUMP); if (retval) { diff --git a/fs/dlm/lowcomms.c b/fs/dlm/lowcomms.c index 0500421b6e3b..f832dafdaca8 100644 --- a/fs/dlm/lowcomms.c +++ b/fs/dlm/lowcomms.c @@ -1599,7 +1599,7 @@ static int dlm_connect(struct connection *con) log_print_ratelimited("connecting to %d", con->nodeid); make_sockaddr(&addr, dlm_config.ci_tcp_port, &addr_len); - result = kernel_connect(sock, (struct sockaddr *)&addr, addr_len, 0); + result = kernel_connect(sock, (struct sockaddr_unsized *)&addr, addr_len, 0); switch (result) { case -EINPROGRESS: /* not an error */ diff --git a/fs/ocfs2/cluster/tcp.c b/fs/ocfs2/cluster/tcp.c index c7734193d8d7..79b281e32f4c 100644 --- a/fs/ocfs2/cluster/tcp.c +++ b/fs/ocfs2/cluster/tcp.c @@ -1638,7 +1638,7 @@ static void o2net_start_connect(struct work_struct *work) remoteaddr.sin_port = node->nd_ipv4_port; ret = sc->sc_sock->ops->connect(sc->sc_sock, - (struct sockaddr *)&remoteaddr, + (struct sockaddr_unsized *)&remoteaddr, sizeof(remoteaddr), O_NONBLOCK); if (ret == -EINPROGRESS) diff --git a/fs/smb/client/connect.c b/fs/smb/client/connect.c index 96d972263020..73120988661a 100644 --- a/fs/smb/client/connect.c +++ b/fs/smb/client/connect.c @@ -3411,7 +3411,7 @@ generic_ip_connect(struct TCP_Server_Info *server) socket->sk->sk_sndbuf, socket->sk->sk_rcvbuf, socket->sk->sk_rcvtimeo); - rc = kernel_connect(socket, saddr, slen, + rc = kernel_connect(socket, (struct sockaddr_unsized *)saddr, slen, server->noblockcnt ? O_NONBLOCK : 0); /* * When mounting SMB root file systems, we do not want to block in diff --git a/include/linux/bpf-cgroup.h b/include/linux/bpf-cgroup.h index aedf573bdb42..a7fb4f46974f 100644 --- a/include/linux/bpf-cgroup.h +++ b/include/linux/bpf-cgroup.h @@ -238,7 +238,7 @@ static inline bool cgroup_bpf_sock_enabled(struct sock *sk, ({ \ int __ret = 0; \ if (cgroup_bpf_enabled(atype)) \ - __ret = __cgroup_bpf_run_filter_sock_addr(sk, uaddr, uaddrlen, \ + __ret = __cgroup_bpf_run_filter_sock_addr(sk, (struct sockaddr *)uaddr, uaddrlen, \ atype, NULL, NULL); \ __ret; \ }) @@ -248,7 +248,7 @@ static inline bool cgroup_bpf_sock_enabled(struct sock *sk, int __ret = 0; \ if (cgroup_bpf_enabled(atype)) { \ lock_sock(sk); \ - __ret = __cgroup_bpf_run_filter_sock_addr(sk, uaddr, uaddrlen, \ + __ret = __cgroup_bpf_run_filter_sock_addr(sk, (struct sockaddr *)uaddr, uaddrlen, \ atype, t_ctx, NULL); \ release_sock(sk); \ } \ @@ -266,7 +266,7 @@ static inline bool cgroup_bpf_sock_enabled(struct sock *sk, int __ret = 0; \ if (cgroup_bpf_enabled(atype)) { \ lock_sock(sk); \ - __ret = __cgroup_bpf_run_filter_sock_addr(sk, uaddr, uaddrlen, \ + __ret = __cgroup_bpf_run_filter_sock_addr(sk, (struct sockaddr *)uaddr, uaddrlen, \ atype, NULL, &__flags); \ release_sock(sk); \ if (__flags & BPF_RET_BIND_NO_CAP_NET_BIND_SERVICE) \ diff --git a/include/linux/net.h b/include/linux/net.h index 0e316f063113..db6bc997ca5b 100644 --- a/include/linux/net.h +++ b/include/linux/net.h @@ -166,7 +166,7 @@ struct proto_ops { struct sockaddr_unsized *myaddr, int sockaddr_len); int (*connect) (struct socket *sock, - struct sockaddr *vaddr, + struct sockaddr_unsized *vaddr, int sockaddr_len, int flags); int (*socketpair)(struct socket *sock1, struct socket *sock2); @@ -348,7 +348,7 @@ int kernel_recvmsg(struct socket *sock, struct msghdr *msg, struct kvec *vec, int kernel_bind(struct socket *sock, struct sockaddr_unsized *addr, int addrlen); int kernel_listen(struct socket *sock, int backlog); int kernel_accept(struct socket *sock, struct socket **newsock, int flags); -int kernel_connect(struct socket *sock, struct sockaddr *addr, int addrlen, +int kernel_connect(struct socket *sock, struct sockaddr_unsized *addr, int addrlen, int flags); int kernel_getsockname(struct socket *sock, struct sockaddr *addr); int kernel_getpeername(struct socket *sock, struct sockaddr *addr); diff --git a/include/net/inet_common.h b/include/net/inet_common.h index 1666cf6f539e..ebafd96912bb 100644 --- a/include/net/inet_common.h +++ b/include/net/inet_common.h @@ -23,11 +23,11 @@ struct sockaddr; struct socket; int inet_release(struct socket *sock); -int inet_stream_connect(struct socket *sock, struct sockaddr *uaddr, +int inet_stream_connect(struct socket *sock, struct sockaddr_unsized *uaddr, int addr_len, int flags); -int __inet_stream_connect(struct socket *sock, struct sockaddr *uaddr, +int __inet_stream_connect(struct socket *sock, struct sockaddr_unsized *uaddr, int addr_len, int flags, int is_sendmsg); -int inet_dgram_connect(struct socket *sock, struct sockaddr *uaddr, +int inet_dgram_connect(struct socket *sock, struct sockaddr_unsized *uaddr, int addr_len, int flags); int inet_accept(struct socket *sock, struct socket *newsock, struct proto_accept_arg *arg); diff --git a/include/net/sctp/sctp.h b/include/net/sctp/sctp.h index bb4b80c12541..58242b37b47a 100644 --- a/include/net/sctp/sctp.h +++ b/include/net/sctp/sctp.h @@ -85,7 +85,7 @@ void sctp_udp_sock_stop(struct net *net); /* * sctp/socket.c */ -int sctp_inet_connect(struct socket *sock, struct sockaddr *uaddr, +int sctp_inet_connect(struct socket *sock, struct sockaddr_unsized *uaddr, int addr_len, int flags); int sctp_backlog_rcv(struct sock *sk, struct sk_buff *skb); int sctp_inet_listen(struct socket *sock, int backlog); diff --git a/include/net/sock.h b/include/net/sock.h index acbb78c96d69..589fbce77217 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -1921,7 +1921,7 @@ int sock_cmsg_send(struct sock *sk, struct msghdr *msg, * does not implement a particular function. */ int sock_no_bind(struct socket *sock, struct sockaddr_unsized *saddr, int len); -int sock_no_connect(struct socket *, struct sockaddr *, int, int); +int sock_no_connect(struct socket *sock, struct sockaddr_unsized *saddr, int len, int flags); int sock_no_socketpair(struct socket *, struct socket *); int sock_no_accept(struct socket *, struct socket *, struct proto_accept_arg *); int sock_no_getname(struct socket *, struct sockaddr *, int); diff --git a/include/net/vsock_addr.h b/include/net/vsock_addr.h index cf8cc140d68d..c3f4cc206198 100644 --- a/include/net/vsock_addr.h +++ b/include/net/vsock_addr.h @@ -16,7 +16,7 @@ bool vsock_addr_bound(const struct sockaddr_vm *addr); void vsock_addr_unbind(struct sockaddr_vm *addr); bool vsock_addr_equals_addr(const struct sockaddr_vm *addr, const struct sockaddr_vm *other); -int vsock_addr_cast(const struct sockaddr *addr, size_t len, +int vsock_addr_cast(const struct sockaddr_unsized *addr, size_t len, struct sockaddr_vm **out_addr); #endif diff --git a/net/9p/trans_fd.c b/net/9p/trans_fd.c index ef517bb307e2..49d674f5e73a 100644 --- a/net/9p/trans_fd.c +++ b/net/9p/trans_fd.c @@ -1018,7 +1018,7 @@ p9_fd_create_tcp(struct p9_client *client, const char *addr, char *args) } err = READ_ONCE(csocket->ops)->connect(csocket, - (struct sockaddr *)&stor, + (struct sockaddr_unsized *)&stor, sizeof(stor), 0); if (err < 0) { pr_err("%s (%d): problem connecting socket to %s\n", @@ -1058,8 +1058,8 @@ p9_fd_create_unix(struct p9_client *client, const char *addr, char *args) return err; } - err = READ_ONCE(csocket->ops)->connect(csocket, (struct sockaddr *)&sun_server, - sizeof(struct sockaddr_un) - 1, 0); + err = READ_ONCE(csocket->ops)->connect(csocket, (struct sockaddr_unsized *)&sun_server, + sizeof(struct sockaddr_un) - 1, 0); if (err < 0) { pr_err("%s (%d): problem connecting socket: %s: %d\n", __func__, task_pid_nr(current), addr, err); diff --git a/net/appletalk/ddp.c b/net/appletalk/ddp.c index 45db43cde67f..2a01fff46c9d 100644 --- a/net/appletalk/ddp.c +++ b/net/appletalk/ddp.c @@ -1204,7 +1204,7 @@ static int atalk_bind(struct socket *sock, struct sockaddr_unsized *uaddr, int a } /* Set the address we talk to */ -static int atalk_connect(struct socket *sock, struct sockaddr *uaddr, +static int atalk_connect(struct socket *sock, struct sockaddr_unsized *uaddr, int addr_len, int flags) { struct sock *sk = sock->sk; diff --git a/net/atm/pvc.c b/net/atm/pvc.c index 62fdf07c53de..8f5e76f5dd9e 100644 --- a/net/atm/pvc.c +++ b/net/atm/pvc.c @@ -56,10 +56,10 @@ static int pvc_bind(struct socket *sock, struct sockaddr_unsized *sockaddr, return error; } -static int pvc_connect(struct socket *sock, struct sockaddr *sockaddr, +static int pvc_connect(struct socket *sock, struct sockaddr_unsized *sockaddr, int sockaddr_len, int flags) { - return pvc_bind(sock, (struct sockaddr_unsized *)sockaddr, sockaddr_len); + return pvc_bind(sock, sockaddr, sockaddr_len); } static int pvc_setsockopt(struct socket *sock, int level, int optname, diff --git a/net/atm/svc.c b/net/atm/svc.c index 1906a493c8aa..005964250ecd 100644 --- a/net/atm/svc.c +++ b/net/atm/svc.c @@ -153,7 +153,7 @@ static int svc_bind(struct socket *sock, struct sockaddr_unsized *sockaddr, return error; } -static int svc_connect(struct socket *sock, struct sockaddr *sockaddr, +static int svc_connect(struct socket *sock, struct sockaddr_unsized *sockaddr, int sockaddr_len, int flags) { DEFINE_WAIT(wait); diff --git a/net/ax25/af_ax25.c b/net/ax25/af_ax25.c index 23c558ff9682..7ebbff2f0020 100644 --- a/net/ax25/af_ax25.c +++ b/net/ax25/af_ax25.c @@ -1175,7 +1175,7 @@ static int ax25_bind(struct socket *sock, struct sockaddr_unsized *uaddr, int ad * FIXME: nonblock behaviour looks like it may have a bug. */ static int __must_check ax25_connect(struct socket *sock, - struct sockaddr *uaddr, int addr_len, int flags) + struct sockaddr_unsized *uaddr, int addr_len, int flags) { struct sock *sk = sock->sk; ax25_cb *ax25 = sk_to_ax25(sk), *ax25t; diff --git a/net/bluetooth/iso.c b/net/bluetooth/iso.c index 6a7e1b4a8701..243505b89733 100644 --- a/net/bluetooth/iso.c +++ b/net/bluetooth/iso.c @@ -1080,7 +1080,7 @@ static int iso_sock_bind(struct socket *sock, struct sockaddr_unsized *addr, return err; } -static int iso_sock_connect(struct socket *sock, struct sockaddr *addr, +static int iso_sock_connect(struct socket *sock, struct sockaddr_unsized *addr, int alen, int flags) { struct sockaddr_iso *sa = (struct sockaddr_iso *)addr; diff --git a/net/bluetooth/l2cap_sock.c b/net/bluetooth/l2cap_sock.c index ca7394d8fa4e..9ee189c815d4 100644 --- a/net/bluetooth/l2cap_sock.c +++ b/net/bluetooth/l2cap_sock.c @@ -178,7 +178,7 @@ static int l2cap_sock_bind(struct socket *sock, struct sockaddr_unsized *addr, i return err; } -static int l2cap_sock_connect(struct socket *sock, struct sockaddr *addr, +static int l2cap_sock_connect(struct socket *sock, struct sockaddr_unsized *addr, int alen, int flags) { struct sock *sk = sock->sk; diff --git a/net/bluetooth/rfcomm/core.c b/net/bluetooth/rfcomm/core.c index d62fd6c57617..57b1dca8141f 100644 --- a/net/bluetooth/rfcomm/core.c +++ b/net/bluetooth/rfcomm/core.c @@ -808,7 +808,7 @@ static struct rfcomm_session *rfcomm_session_create(bdaddr_t *src, addr.l2_psm = cpu_to_le16(L2CAP_PSM_RFCOMM); addr.l2_cid = 0; addr.l2_bdaddr_type = BDADDR_BREDR; - *err = kernel_connect(sock, (struct sockaddr *) &addr, sizeof(addr), O_NONBLOCK); + *err = kernel_connect(sock, (struct sockaddr_unsized *)&addr, sizeof(addr), O_NONBLOCK); if (*err == 0 || *err == -EINPROGRESS) return s; diff --git a/net/bluetooth/rfcomm/sock.c b/net/bluetooth/rfcomm/sock.c index 8c8762bbc6de..be6639cd6f59 100644 --- a/net/bluetooth/rfcomm/sock.c +++ b/net/bluetooth/rfcomm/sock.c @@ -371,7 +371,8 @@ static int rfcomm_sock_bind(struct socket *sock, struct sockaddr_unsized *addr, return err; } -static int rfcomm_sock_connect(struct socket *sock, struct sockaddr *addr, int alen, int flags) +static int rfcomm_sock_connect(struct socket *sock, struct sockaddr_unsized *addr, + int alen, int flags) { struct sockaddr_rc *sa = (struct sockaddr_rc *) addr; struct sock *sk = sock->sk; diff --git a/net/bluetooth/sco.c b/net/bluetooth/sco.c index 01d878205e58..7afe65e7ff37 100644 --- a/net/bluetooth/sco.c +++ b/net/bluetooth/sco.c @@ -639,7 +639,7 @@ static int sco_sock_bind(struct socket *sock, struct sockaddr_unsized *addr, return err; } -static int sco_sock_connect(struct socket *sock, struct sockaddr *addr, int alen, int flags) +static int sco_sock_connect(struct socket *sock, struct sockaddr_unsized *addr, int alen, int flags) { struct sockaddr_sco *sa = (struct sockaddr_sco *) addr; struct sock *sk = sock->sk; diff --git a/net/caif/caif_socket.c b/net/caif/caif_socket.c index 039dfbd367c9..af218742af5a 100644 --- a/net/caif/caif_socket.c +++ b/net/caif/caif_socket.c @@ -734,7 +734,7 @@ static int setsockopt(struct socket *sock, int lvl, int opt, sockptr_t ov, * o sock->state: holds the SS_* socket state and is updated by connect and * disconnect. */ -static int caif_connect(struct socket *sock, struct sockaddr *uaddr, +static int caif_connect(struct socket *sock, struct sockaddr_unsized *uaddr, int addr_len, int flags) { struct sock *sk = sock->sk; diff --git a/net/can/bcm.c b/net/can/bcm.c index 5e690a2377e4..7eba8ae01a5b 100644 --- a/net/can/bcm.c +++ b/net/can/bcm.c @@ -1657,7 +1657,7 @@ static int bcm_release(struct socket *sock) return 0; } -static int bcm_connect(struct socket *sock, struct sockaddr *uaddr, int len, +static int bcm_connect(struct socket *sock, struct sockaddr_unsized *uaddr, int len, int flags) { struct sockaddr_can *addr = (struct sockaddr_can *)uaddr; diff --git a/net/can/j1939/socket.c b/net/can/j1939/socket.c index a2abedc757d0..6272326dd614 100644 --- a/net/can/j1939/socket.c +++ b/net/can/j1939/socket.c @@ -535,7 +535,7 @@ static int j1939_sk_bind(struct socket *sock, struct sockaddr_unsized *uaddr, in return ret; } -static int j1939_sk_connect(struct socket *sock, struct sockaddr *uaddr, +static int j1939_sk_connect(struct socket *sock, struct sockaddr_unsized *uaddr, int len, int flags) { struct sockaddr_can *addr = (struct sockaddr_can *)uaddr; diff --git a/net/ceph/messenger.c b/net/ceph/messenger.c index f8181acaf870..70b25f4ecba6 100644 --- a/net/ceph/messenger.c +++ b/net/ceph/messenger.c @@ -460,7 +460,7 @@ int ceph_tcp_connect(struct ceph_connection *con) set_sock_callbacks(sock, con); con_sock_state_connecting(con); - ret = kernel_connect(sock, (struct sockaddr *)&ss, sizeof(ss), + ret = kernel_connect(sock, (struct sockaddr_unsized *)&ss, sizeof(ss), O_NONBLOCK); if (ret == -EINPROGRESS) { dout("connect %s EINPROGRESS sk_state = %u\n", diff --git a/net/core/sock.c b/net/core/sock.c index 1e1ce18bba16..f97a0e958991 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -3468,7 +3468,7 @@ int sock_no_bind(struct socket *sock, struct sockaddr_unsized *saddr, int len) } EXPORT_SYMBOL(sock_no_bind); -int sock_no_connect(struct socket *sock, struct sockaddr *saddr, +int sock_no_connect(struct socket *sock, struct sockaddr_unsized *saddr, int len, int flags) { return -EOPNOTSUPP; diff --git a/net/ieee802154/socket.c b/net/ieee802154/socket.c index 99ddfad9bb88..b93fd85f248a 100644 --- a/net/ieee802154/socket.c +++ b/net/ieee802154/socket.c @@ -107,7 +107,7 @@ static int ieee802154_sock_bind(struct socket *sock, struct sockaddr_unsized *ua return sock_no_bind(sock, uaddr, addr_len); } -static int ieee802154_sock_connect(struct socket *sock, struct sockaddr *uaddr, +static int ieee802154_sock_connect(struct socket *sock, struct sockaddr_unsized *uaddr, int addr_len, int flags) { struct sock *sk = sock->sk; @@ -118,7 +118,7 @@ static int ieee802154_sock_connect(struct socket *sock, struct sockaddr *uaddr, if (uaddr->sa_family == AF_UNSPEC) return sk->sk_prot->disconnect(sk, flags); - return sk->sk_prot->connect(sk, uaddr, addr_len); + return sk->sk_prot->connect(sk, (struct sockaddr *)uaddr, addr_len); } static int ieee802154_dev_ioctl(struct sock *sk, struct ifreq __user *arg, diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index aa43d16e48ff..0844de9ac6a4 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -567,7 +567,7 @@ int __inet_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len, return err; } -int inet_dgram_connect(struct socket *sock, struct sockaddr *uaddr, +int inet_dgram_connect(struct socket *sock, struct sockaddr_unsized *uaddr, int addr_len, int flags) { struct sock *sk = sock->sk; @@ -584,14 +584,14 @@ int inet_dgram_connect(struct socket *sock, struct sockaddr *uaddr, return prot->disconnect(sk, flags); if (BPF_CGROUP_PRE_CONNECT_ENABLED(sk)) { - err = prot->pre_connect(sk, uaddr, addr_len); + err = prot->pre_connect(sk, (struct sockaddr *)uaddr, addr_len); if (err) return err; } if (data_race(!inet_sk(sk)->inet_num) && inet_autobind(sk)) return -EAGAIN; - return prot->connect(sk, uaddr, addr_len); + return prot->connect(sk, (struct sockaddr *)uaddr, addr_len); } EXPORT_SYMBOL(inet_dgram_connect); @@ -623,7 +623,7 @@ static long inet_wait_for_connect(struct sock *sk, long timeo, int writebias) * Connect to a remote host. There is regrettably still a little * TCP 'magic' in here. */ -int __inet_stream_connect(struct socket *sock, struct sockaddr *uaddr, +int __inet_stream_connect(struct socket *sock, struct sockaddr_unsized *uaddr, int addr_len, int flags, int is_sendmsg) { struct sock *sk = sock->sk; @@ -671,12 +671,12 @@ int __inet_stream_connect(struct socket *sock, struct sockaddr *uaddr, goto out; if (BPF_CGROUP_PRE_CONNECT_ENABLED(sk)) { - err = sk->sk_prot->pre_connect(sk, uaddr, addr_len); + err = sk->sk_prot->pre_connect(sk, (struct sockaddr *)uaddr, addr_len); if (err) goto out; } - err = sk->sk_prot->connect(sk, uaddr, addr_len); + err = sk->sk_prot->connect(sk, (struct sockaddr *)uaddr, addr_len); if (err < 0) goto out; @@ -741,7 +741,7 @@ int __inet_stream_connect(struct socket *sock, struct sockaddr *uaddr, } EXPORT_SYMBOL(__inet_stream_connect); -int inet_stream_connect(struct socket *sock, struct sockaddr *uaddr, +int inet_stream_connect(struct socket *sock, struct sockaddr_unsized *uaddr, int addr_len, int flags) { int err; diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index a9345aa5a2e5..dee578aad690 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -1061,7 +1061,7 @@ int tcp_sendmsg_fastopen(struct sock *sk, struct msghdr *msg, int *copied, } } flags = (msg->msg_flags & MSG_DONTWAIT) ? O_NONBLOCK : 0; - err = __inet_stream_connect(sk->sk_socket, uaddr, + err = __inet_stream_connect(sk->sk_socket, (struct sockaddr_unsized *)uaddr, msg->msg_namelen, flags, 1); /* fastopen_req could already be freed in __inet_stream_connect * if the connection times out or gets rst diff --git a/net/ipv4/udp_tunnel_core.c b/net/ipv4/udp_tunnel_core.c index 11e5a88c923d..b1f667c52cb2 100644 --- a/net/ipv4/udp_tunnel_core.c +++ b/net/ipv4/udp_tunnel_core.c @@ -38,7 +38,7 @@ int udp_sock_create4(struct net *net, struct udp_port_cfg *cfg, udp_addr.sin_family = AF_INET; udp_addr.sin_addr = cfg->peer_ip; udp_addr.sin_port = cfg->peer_udp_port; - err = kernel_connect(sock, (struct sockaddr *)&udp_addr, + err = kernel_connect(sock, (struct sockaddr_unsized *)&udp_addr, sizeof(udp_addr), 0); if (err < 0) goto error; diff --git a/net/ipv6/ip6_udp_tunnel.c b/net/ipv6/ip6_udp_tunnel.c index b0d9286b33c8..cef3e0210744 100644 --- a/net/ipv6/ip6_udp_tunnel.c +++ b/net/ipv6/ip6_udp_tunnel.c @@ -52,7 +52,7 @@ int udp_sock_create6(struct net *net, struct udp_port_cfg *cfg, sizeof(udp6_addr.sin6_addr)); udp6_addr.sin6_port = cfg->peer_udp_port; err = kernel_connect(sock, - (struct sockaddr *)&udp6_addr, + (struct sockaddr_unsized *)&udp6_addr, sizeof(udp6_addr), 0); } if (err < 0) diff --git a/net/iucv/af_iucv.c b/net/iucv/af_iucv.c index 3941e32cda69..a4f1df92417d 100644 --- a/net/iucv/af_iucv.c +++ b/net/iucv/af_iucv.c @@ -668,7 +668,7 @@ static int iucv_sock_autobind(struct sock *sk) return err; } -static int afiucv_path_connect(struct socket *sock, struct sockaddr *addr) +static int afiucv_path_connect(struct socket *sock, struct sockaddr_unsized *addr) { DECLARE_SOCKADDR(struct sockaddr_iucv *, sa, addr); struct sock *sk = sock->sk; @@ -714,7 +714,7 @@ static int afiucv_path_connect(struct socket *sock, struct sockaddr *addr) } /* Connect an unconnected socket */ -static int iucv_sock_connect(struct socket *sock, struct sockaddr *addr, +static int iucv_sock_connect(struct socket *sock, struct sockaddr_unsized *addr, int alen, int flags) { DECLARE_SOCKADDR(struct sockaddr_iucv *, sa, addr); diff --git a/net/l2tp/l2tp_core.c b/net/l2tp/l2tp_core.c index 4b5e372a5cd4..c4f4a57cd67c 100644 --- a/net/l2tp/l2tp_core.c +++ b/net/l2tp/l2tp_core.c @@ -1513,7 +1513,7 @@ static int l2tp_tunnel_sock_create(struct net *net, sizeof(ip6_addr.l2tp_addr)); ip6_addr.l2tp_conn_id = peer_tunnel_id; err = kernel_connect(sock, - (struct sockaddr *)&ip6_addr, + (struct sockaddr_unsized *)&ip6_addr, sizeof(ip6_addr), 0); if (err < 0) goto out; @@ -1538,7 +1538,7 @@ static int l2tp_tunnel_sock_create(struct net *net, ip_addr.l2tp_family = AF_INET; ip_addr.l2tp_addr = cfg->peer_ip; ip_addr.l2tp_conn_id = peer_tunnel_id; - err = kernel_connect(sock, (struct sockaddr *)&ip_addr, + err = kernel_connect(sock, (struct sockaddr_unsized *)&ip_addr, sizeof(ip_addr), 0); if (err < 0) goto out; diff --git a/net/l2tp/l2tp_ppp.c b/net/l2tp/l2tp_ppp.c index 5e12e7ce17d8..ae4543d5597b 100644 --- a/net/l2tp/l2tp_ppp.c +++ b/net/l2tp/l2tp_ppp.c @@ -684,7 +684,7 @@ static struct l2tp_tunnel *pppol2tp_tunnel_get(struct net *net, /* connect() handler. Attach a PPPoX socket to a tunnel UDP socket */ -static int pppol2tp_connect(struct socket *sock, struct sockaddr *uservaddr, +static int pppol2tp_connect(struct socket *sock, struct sockaddr_unsized *uservaddr, int sockaddr_len, int flags) { struct sock *sk = sock->sk; diff --git a/net/llc/af_llc.c b/net/llc/af_llc.c index e5bb0c0d708c..59d593bb5d18 100644 --- a/net/llc/af_llc.c +++ b/net/llc/af_llc.c @@ -477,7 +477,7 @@ static int llc_ui_shutdown(struct socket *sock, int how) * This function will autobind if user did not previously call bind. * Returns: 0 upon success, negative otherwise. */ -static int llc_ui_connect(struct socket *sock, struct sockaddr *uaddr, +static int llc_ui_connect(struct socket *sock, struct sockaddr_unsized *uaddr, int addrlen, int flags) { struct sock *sk = sock->sk; diff --git a/net/mctp/af_mctp.c b/net/mctp/af_mctp.c index 5b1ef50637b7..209a963112e3 100644 --- a/net/mctp/af_mctp.c +++ b/net/mctp/af_mctp.c @@ -128,7 +128,7 @@ static int mctp_bind(struct socket *sock, struct sockaddr_unsized *addr, int add /* Used to set a specific peer prior to bind. Not used for outbound * connections (Tag Owner set) since MCTP is a datagram protocol. */ -static int mctp_connect(struct socket *sock, struct sockaddr *addr, +static int mctp_connect(struct socket *sock, struct sockaddr_unsized *addr, int addrlen, int flags) { struct sock *sk = sock->sk; diff --git a/net/mctp/test/utils.c b/net/mctp/test/utils.c index 953d41902771..35f6be814567 100644 --- a/net/mctp/test/utils.c +++ b/net/mctp/test/utils.c @@ -279,7 +279,7 @@ void mctp_test_bind_run(struct kunit *test, addr.smctp_addr.s_addr = setup->peer_addr; /* connect() type must match bind() type */ addr.smctp_type = setup->bind_type; - rc = kernel_connect(*sock, (struct sockaddr *)&addr, + rc = kernel_connect(*sock, (struct sockaddr_unsized *)&addr, sizeof(addr), 0); KUNIT_EXPECT_EQ(test, rc, 0); } @@ -292,5 +292,6 @@ void mctp_test_bind_run(struct kunit *test, addr.smctp_type = setup->bind_type; *ret_bind_errno = - kernel_bind(*sock, (struct sockaddr *)&addr, sizeof(addr)); + kernel_bind(*sock, (struct sockaddr_unsized *)&addr, + sizeof(addr)); } diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c index d90237bf433c..30961b3d1702 100644 --- a/net/mptcp/subflow.c +++ b/net/mptcp/subflow.c @@ -1680,7 +1680,7 @@ int __mptcp_subflow_connect(struct sock *sk, const struct mptcp_pm_local *local, sock_hold(ssk); list_add_tail(&subflow->node, &msk->conn_list); - err = kernel_connect(sf, (struct sockaddr *)&addr, addrlen, O_NONBLOCK); + err = kernel_connect(sf, (struct sockaddr_unsized *)&addr, addrlen, O_NONBLOCK); if (err && err != -EINPROGRESS) { MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_JOINSYNTXCONNECTERR); pr_debug("msk=%p local=%d remote=%d connect error: %d\n", diff --git a/net/netfilter/ipvs/ip_vs_sync.c b/net/netfilter/ipvs/ip_vs_sync.c index d8c089ef387c..5a0c6f42bd8f 100644 --- a/net/netfilter/ipvs/ip_vs_sync.c +++ b/net/netfilter/ipvs/ip_vs_sync.c @@ -1501,7 +1501,7 @@ static int make_send_sock(struct netns_ipvs *ipvs, int id, } get_mcast_sockaddr(&mcast_addr, &salen, &ipvs->mcfg, id); - result = kernel_connect(sock, (struct sockaddr *)&mcast_addr, + result = kernel_connect(sock, (struct sockaddr_unsized *)&mcast_addr, salen, 0); if (result < 0) { pr_err("Error connecting to the multicast addr\n"); diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index 18490a56edd0..8e5151f0c6e4 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -1054,7 +1054,7 @@ static int netlink_bind(struct socket *sock, struct sockaddr_unsized *addr, return err; } -static int netlink_connect(struct socket *sock, struct sockaddr *addr, +static int netlink_connect(struct socket *sock, struct sockaddr_unsized *addr, int alen, int flags) { int err = 0; diff --git a/net/netrom/af_netrom.c b/net/netrom/af_netrom.c index 33468124d53d..5ed1a71ceec1 100644 --- a/net/netrom/af_netrom.c +++ b/net/netrom/af_netrom.c @@ -632,8 +632,8 @@ static int nr_bind(struct socket *sock, struct sockaddr_unsized *uaddr, int addr return 0; } -static int nr_connect(struct socket *sock, struct sockaddr *uaddr, - int addr_len, int flags) +static int nr_connect(struct socket *sock, struct sockaddr_unsized *uaddr, + int addr_len, int flags) { struct sock *sk = sock->sk; struct nr_sock *nr = nr_sk(sk); diff --git a/net/nfc/llcp_sock.c b/net/nfc/llcp_sock.c index 26e6ceb48a82..f1be1e84f665 100644 --- a/net/nfc/llcp_sock.c +++ b/net/nfc/llcp_sock.c @@ -648,7 +648,7 @@ static int llcp_sock_release(struct socket *sock) return err; } -static int llcp_sock_connect(struct socket *sock, struct sockaddr *_addr, +static int llcp_sock_connect(struct socket *sock, struct sockaddr_unsized *_addr, int len, int flags) { struct sock *sk = sock->sk; diff --git a/net/nfc/rawsock.c b/net/nfc/rawsock.c index 5125392bb68e..b049022399ae 100644 --- a/net/nfc/rawsock.c +++ b/net/nfc/rawsock.c @@ -73,7 +73,7 @@ static int rawsock_release(struct socket *sock) return 0; } -static int rawsock_connect(struct socket *sock, struct sockaddr *_addr, +static int rawsock_connect(struct socket *sock, struct sockaddr_unsized *_addr, int len, int flags) { struct sock *sk = sock->sk; diff --git a/net/phonet/socket.c b/net/phonet/socket.c index 478b02647733..9391378083a4 100644 --- a/net/phonet/socket.c +++ b/net/phonet/socket.c @@ -214,8 +214,8 @@ static int pn_socket_autobind(struct socket *sock) return 0; /* socket was already bound */ } -static int pn_socket_connect(struct socket *sock, struct sockaddr *addr, - int len, int flags) +static int pn_socket_connect(struct socket *sock, struct sockaddr_unsized *addr, + int len, int flags) { struct sock *sk = sock->sk; struct pn_sock *pn = pn_sk(sk); @@ -252,7 +252,7 @@ static int pn_socket_connect(struct socket *sock, struct sockaddr *addr, pn->resource = pn_sockaddr_get_resource(spn); sock->state = SS_CONNECTING; - err = sk->sk_prot->connect(sk, addr, len); + err = sk->sk_prot->connect(sk, (struct sockaddr *)addr, len); if (err) { sock->state = SS_UNCONNECTED; pn->dobject = 0; diff --git a/net/qrtr/af_qrtr.c b/net/qrtr/af_qrtr.c index 00bd3dd9f0f9..dab839f61ee9 100644 --- a/net/qrtr/af_qrtr.c +++ b/net/qrtr/af_qrtr.c @@ -1084,7 +1084,7 @@ static int qrtr_recvmsg(struct socket *sock, struct msghdr *msg, return rc; } -static int qrtr_connect(struct socket *sock, struct sockaddr *saddr, +static int qrtr_connect(struct socket *sock, struct sockaddr_unsized *saddr, int len, int flags) { DECLARE_SOCKADDR(struct sockaddr_qrtr *, addr, saddr); diff --git a/net/rds/af_rds.c b/net/rds/af_rds.c index 4a7217fbeab6..b396c673dfaf 100644 --- a/net/rds/af_rds.c +++ b/net/rds/af_rds.c @@ -533,7 +533,7 @@ static int rds_getsockopt(struct socket *sock, int level, int optname, } -static int rds_connect(struct socket *sock, struct sockaddr *uaddr, +static int rds_connect(struct socket *sock, struct sockaddr_unsized *uaddr, int addr_len, int flags) { struct sock *sk = sock->sk; diff --git a/net/rds/tcp_connect.c b/net/rds/tcp_connect.c index 1eff3b03ab77..92891b0d224d 100644 --- a/net/rds/tcp_connect.c +++ b/net/rds/tcp_connect.c @@ -173,7 +173,7 @@ int rds_tcp_conn_path_connect(struct rds_conn_path *cp) * own the socket */ rds_tcp_set_callbacks(sock, cp); - ret = kernel_connect(sock, addr, addrlen, O_NONBLOCK); + ret = kernel_connect(sock, (struct sockaddr_unsized *)addr, addrlen, O_NONBLOCK); rdsdebug("connect to address %pI6c returned %d\n", &conn->c_faddr, ret); if (ret == -EINPROGRESS) diff --git a/net/rose/af_rose.c b/net/rose/af_rose.c index 47369eab5aec..fd67494f2815 100644 --- a/net/rose/af_rose.c +++ b/net/rose/af_rose.c @@ -765,7 +765,8 @@ static int rose_bind(struct socket *sock, struct sockaddr_unsized *uaddr, int ad return err; } -static int rose_connect(struct socket *sock, struct sockaddr *uaddr, int addr_len, int flags) +static int rose_connect(struct socket *sock, struct sockaddr_unsized *uaddr, int addr_len, + int flags) { struct sock *sk = sock->sk; struct rose_sock *rose = rose_sk(sk); diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c index 245f37a74394..0c2c68c4b07e 100644 --- a/net/rxrpc/af_rxrpc.c +++ b/net/rxrpc/af_rxrpc.c @@ -481,7 +481,7 @@ EXPORT_SYMBOL(rxrpc_kernel_set_notifications); * - this just targets it at a specific destination; no actual connection * negotiation takes place */ -static int rxrpc_connect(struct socket *sock, struct sockaddr *addr, +static int rxrpc_connect(struct socket *sock, struct sockaddr_unsized *addr, int addr_len, int flags) { struct sockaddr_rxrpc *srx = (struct sockaddr_rxrpc *)addr; diff --git a/net/sctp/socket.c b/net/sctp/socket.c index ac737e60829b..940abbced191 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -4820,7 +4820,7 @@ static int sctp_connect(struct sock *sk, struct sockaddr *addr, return err; } -int sctp_inet_connect(struct socket *sock, struct sockaddr *uaddr, +int sctp_inet_connect(struct socket *sock, struct sockaddr_unsized *uaddr, int addr_len, int flags) { if (addr_len < sizeof(uaddr->sa_family)) @@ -4829,7 +4829,7 @@ int sctp_inet_connect(struct socket *sock, struct sockaddr *uaddr, if (uaddr->sa_family == AF_UNSPEC) return -EOPNOTSUPP; - return sctp_connect(sock->sk, uaddr, addr_len, flags); + return sctp_connect(sock->sk, (struct sockaddr *)uaddr, addr_len, flags); } /* Only called when shutdown a listening SCTP socket. */ diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c index be18ab08f15d..0ef3e16a8517 100644 --- a/net/smc/af_smc.c +++ b/net/smc/af_smc.c @@ -1642,7 +1642,7 @@ static void smc_connect_work(struct work_struct *work) release_sock(&smc->sk); } -int smc_connect(struct socket *sock, struct sockaddr *addr, +int smc_connect(struct socket *sock, struct sockaddr_unsized *addr, int alen, int flags) { struct sock *sk = sock->sk; @@ -1694,7 +1694,7 @@ int smc_connect(struct socket *sock, struct sockaddr *addr, rc = -EALREADY; goto out; } - rc = kernel_connect(smc->clcsock, addr, alen, flags); + rc = kernel_connect(smc->clcsock, (struct sockaddr_unsized *)addr, alen, flags); if (rc && rc != -EINPROGRESS) goto out; diff --git a/net/smc/smc.h b/net/smc/smc.h index a008dbe6d6f6..9e6af72784ba 100644 --- a/net/smc/smc.h +++ b/net/smc/smc.h @@ -44,7 +44,7 @@ void smc_release_cb(struct sock *sk); int smc_release(struct socket *sock); int smc_bind(struct socket *sock, struct sockaddr_unsized *uaddr, int addr_len); -int smc_connect(struct socket *sock, struct sockaddr *addr, +int smc_connect(struct socket *sock, struct sockaddr_unsized *addr, int alen, int flags); int smc_accept(struct socket *sock, struct socket *new_sock, struct proto_accept_arg *arg); diff --git a/net/socket.c b/net/socket.c index aaefb2e519a7..101a7ed574e7 100644 --- a/net/socket.c +++ b/net/socket.c @@ -2099,8 +2099,8 @@ int __sys_connect_file(struct file *file, struct sockaddr_storage *address, if (err) goto out; - err = READ_ONCE(sock->ops)->connect(sock, (struct sockaddr *)address, - addrlen, sock->file->f_flags | file_flags); + err = READ_ONCE(sock->ops)->connect(sock, (struct sockaddr_unsized *)address, + addrlen, sock->file->f_flags | file_flags); out: return err; } @@ -3662,14 +3662,14 @@ EXPORT_SYMBOL(kernel_accept); * Returns 0 or an error code. */ -int kernel_connect(struct socket *sock, struct sockaddr *addr, int addrlen, +int kernel_connect(struct socket *sock, struct sockaddr_unsized *addr, int addrlen, int flags) { struct sockaddr_storage address; memcpy(&address, addr, addrlen); - return READ_ONCE(sock->ops)->connect(sock, (struct sockaddr *)&address, + return READ_ONCE(sock->ops)->connect(sock, (struct sockaddr_unsized *)&address, addrlen, flags); } EXPORT_SYMBOL(kernel_connect); diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c index 318ee24ad900..58442ae1c2da 100644 --- a/net/sunrpc/clnt.c +++ b/net/sunrpc/clnt.c @@ -1474,7 +1474,7 @@ static int rpc_sockname(struct net *net, struct sockaddr *sap, size_t salen, goto out_release; } - err = kernel_connect(sock, sap, salen, 0); + err = kernel_connect(sock, (struct sockaddr_unsized *)sap, salen, 0); if (err < 0) { dprintk("RPC: can't connect UDP socket (%d)\n", err); goto out_release; diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index 95732a45b059..2e1fe6013361 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -2005,7 +2005,7 @@ static int xs_local_finish_connecting(struct rpc_xprt *xprt, xs_stream_start_connect(transport); - return kernel_connect(sock, xs_addr(xprt), xprt->addrlen, 0); + return kernel_connect(sock, (struct sockaddr_unsized *)xs_addr(xprt), xprt->addrlen, 0); } /** @@ -2405,7 +2405,8 @@ static int xs_tcp_finish_connecting(struct rpc_xprt *xprt, struct socket *sock) /* Tell the socket layer to start connecting... */ set_bit(XPRT_SOCK_CONNECTING, &transport->sock_state); - return kernel_connect(sock, xs_addr(xprt), xprt->addrlen, O_NONBLOCK); + return kernel_connect(sock, (struct sockaddr_unsized *)xs_addr(xprt), + xprt->addrlen, O_NONBLOCK); } /** diff --git a/net/tipc/socket.c b/net/tipc/socket.c index 3903a97ada7d..817b07d95a91 100644 --- a/net/tipc/socket.c +++ b/net/tipc/socket.c @@ -2565,7 +2565,7 @@ static bool tipc_sockaddr_is_sane(struct sockaddr_tipc *addr) * * Return: 0 on success, errno otherwise */ -static int tipc_connect(struct socket *sock, struct sockaddr *dest, +static int tipc_connect(struct socket *sock, struct sockaddr_unsized *dest, int destlen, int flags) { struct sock *sk = sock->sk; diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index 788775f0eea7..3b44cadaed96 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -844,7 +844,7 @@ static int unix_listen(struct socket *sock, int backlog) static int unix_release(struct socket *); static int unix_bind(struct socket *, struct sockaddr_unsized *, int); -static int unix_stream_connect(struct socket *, struct sockaddr *, +static int unix_stream_connect(struct socket *, struct sockaddr_unsized *, int addr_len, int flags); static int unix_socketpair(struct socket *, struct socket *); static int unix_accept(struct socket *, struct socket *, struct proto_accept_arg *arg); @@ -866,7 +866,7 @@ static int unix_dgram_sendmsg(struct socket *, struct msghdr *, size_t); static int unix_dgram_recvmsg(struct socket *, struct msghdr *, size_t, int); static int unix_read_skb(struct sock *sk, skb_read_actor_t recv_actor); static int unix_stream_read_skb(struct sock *sk, skb_read_actor_t recv_actor); -static int unix_dgram_connect(struct socket *, struct sockaddr *, +static int unix_dgram_connect(struct socket *, struct sockaddr_unsized *, int, int); static int unix_seqpacket_sendmsg(struct socket *, struct msghdr *, size_t); static int unix_seqpacket_recvmsg(struct socket *, struct msghdr *, size_t, @@ -1512,7 +1512,7 @@ static void unix_state_double_unlock(struct sock *sk1, struct sock *sk2) unix_state_unlock(sk2); } -static int unix_dgram_connect(struct socket *sock, struct sockaddr *addr, +static int unix_dgram_connect(struct socket *sock, struct sockaddr_unsized *addr, int alen, int flags) { struct sockaddr_un *sunaddr = (struct sockaddr_un *)addr; @@ -1631,7 +1631,7 @@ static long unix_wait_for_peer(struct sock *other, long timeo) return timeo; } -static int unix_stream_connect(struct socket *sock, struct sockaddr *uaddr, +static int unix_stream_connect(struct socket *sock, struct sockaddr_unsized *uaddr, int addr_len, int flags) { struct sockaddr_un *sunaddr = (struct sockaddr_un *)uaddr; diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c index 0e5609e7284b..72bb6b7ed386 100644 --- a/net/vmw_vsock/af_vsock.c +++ b/net/vmw_vsock/af_vsock.c @@ -995,7 +995,7 @@ vsock_bind(struct socket *sock, struct sockaddr_unsized *addr, int addr_len) sk = sock->sk; - if (vsock_addr_cast((struct sockaddr *)addr, addr_len, &vm_addr) != 0) + if (vsock_addr_cast(addr, addr_len, &vm_addr) != 0) return -EINVAL; lock_sock(sk); @@ -1328,7 +1328,7 @@ static int vsock_dgram_sendmsg(struct socket *sock, struct msghdr *msg, } static int vsock_dgram_connect(struct socket *sock, - struct sockaddr *addr, int addr_len, int flags) + struct sockaddr_unsized *addr, int addr_len, int flags) { int err; struct sock *sk; @@ -1528,7 +1528,7 @@ static void vsock_connect_timeout(struct work_struct *work) sock_put(sk); } -static int vsock_connect(struct socket *sock, struct sockaddr *addr, +static int vsock_connect(struct socket *sock, struct sockaddr_unsized *addr, int addr_len, int flags) { int err; diff --git a/net/vmw_vsock/vsock_addr.c b/net/vmw_vsock/vsock_addr.c index 223b9660a759..a986aa6fff9b 100644 --- a/net/vmw_vsock/vsock_addr.c +++ b/net/vmw_vsock/vsock_addr.c @@ -57,7 +57,7 @@ bool vsock_addr_equals_addr(const struct sockaddr_vm *addr, } EXPORT_SYMBOL_GPL(vsock_addr_equals_addr); -int vsock_addr_cast(const struct sockaddr *addr, +int vsock_addr_cast(const struct sockaddr_unsized *addr, size_t len, struct sockaddr_vm **out_addr) { if (len < sizeof(**out_addr)) diff --git a/net/x25/af_x25.c b/net/x25/af_x25.c index ca8006d8f792..af8762b24039 100644 --- a/net/x25/af_x25.c +++ b/net/x25/af_x25.c @@ -743,7 +743,7 @@ static int x25_wait_for_connection_establishment(struct sock *sk) return rc; } -static int x25_connect(struct socket *sock, struct sockaddr *uaddr, +static int x25_connect(struct socket *sock, struct sockaddr_unsized *uaddr, int addr_len, int flags) { struct sock *sk = sock->sk; diff --git a/samples/qmi/qmi_sample_client.c b/samples/qmi/qmi_sample_client.c index b27d861f354f..d1814582319b 100644 --- a/samples/qmi/qmi_sample_client.c +++ b/samples/qmi/qmi_sample_client.c @@ -468,7 +468,7 @@ static int qmi_sample_probe(struct platform_device *pdev) return ret; sq = dev_get_platdata(&pdev->dev); - ret = kernel_connect(sample->qmi.sock, (struct sockaddr *)sq, + ret = kernel_connect(sample->qmi.sock, (struct sockaddr_unsized *)sq, sizeof(*sq), 0); if (ret < 0) { pr_err("failed to connect to remote service port\n"); diff --git a/tools/testing/selftests/bpf/test_kmods/bpf_testmod.c b/tools/testing/selftests/bpf/test_kmods/bpf_testmod.c index 0497b5dea25c..8eeebaa951f0 100644 --- a/tools/testing/selftests/bpf/test_kmods/bpf_testmod.c +++ b/tools/testing/selftests/bpf/test_kmods/bpf_testmod.c @@ -900,7 +900,7 @@ __bpf_kfunc int bpf_kfunc_call_kernel_connect(struct addr_args *args) goto out; } - err = kernel_connect(sock, (struct sockaddr *)&args->addr, + err = kernel_connect(sock, (struct sockaddr_unsized *)&args->addr, args->addrlen, 0); out: mutex_unlock(&sock_lock); From 3d39d34146f2b38127eadf36a0513e130eaa7eec Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Mon, 3 Nov 2025 16:26:12 -0800 Subject: [PATCH 0535/1381] net: Remove struct sockaddr from net.h Now that struct sockaddr is no longer used by net.h, remove it. Signed-off-by: Kees Cook Link: https://patch.msgid.link/20251104002617.2752303-4-kees@kernel.org Signed-off-by: Jakub Kicinski --- include/linux/net.h | 1 - 1 file changed, 1 deletion(-) diff --git a/include/linux/net.h b/include/linux/net.h index db6bc997ca5b..f58b38ab37f8 100644 --- a/include/linux/net.h +++ b/include/linux/net.h @@ -148,7 +148,6 @@ typedef struct { struct vm_area_struct; struct page; -struct sockaddr; struct msghdr; struct module; struct sk_buff; From 449f68f8fffa2c41fc265730bd05a3c4947916c1 Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Mon, 3 Nov 2025 16:26:13 -0800 Subject: [PATCH 0536/1381] net: Convert proto callbacks from sockaddr to sockaddr_unsized Convert struct proto pre_connect(), connect(), bind(), and bind_add() callback function prototypes from struct sockaddr to struct sockaddr_unsized. This does not change per-implementation use of sockaddr for passing around an arbitrarily sized sockaddr struct. Those will be addressed in future patches. Additionally removes the no longer referenced struct sockaddr from include/net/inet_common.h. No binary changes expected. Signed-off-by: Kees Cook Link: https://patch.msgid.link/20251104002617.2752303-5-kees@kernel.org Signed-off-by: Jakub Kicinski --- fs/dlm/lowcomms.c | 4 ++-- include/net/inet_common.h | 5 ++--- include/net/ip.h | 4 ++-- include/net/ipv6.h | 8 ++++---- include/net/ipv6_stubs.h | 2 +- include/net/ping.h | 2 +- include/net/sock.h | 10 +++++----- include/net/tcp.h | 2 +- include/net/udp.h | 2 +- net/core/filter.c | 5 +++-- net/core/sock.c | 2 +- net/ieee802154/socket.c | 12 ++++++------ net/ipv4/af_inet.c | 14 +++++++------- net/ipv4/datagram.c | 4 ++-- net/ipv4/ping.c | 8 ++++---- net/ipv4/raw.c | 3 ++- net/ipv4/tcp_ipv4.c | 4 ++-- net/ipv4/udp.c | 6 ++++-- net/ipv6/af_inet6.c | 6 +++--- net/ipv6/datagram.c | 8 ++++---- net/ipv6/ping.c | 2 +- net/ipv6/raw.c | 3 ++- net/ipv6/tcp_ipv6.c | 6 +++--- net/ipv6/udp.c | 5 +++-- net/l2tp/l2tp_ip.c | 6 ++++-- net/l2tp/l2tp_ip6.c | 5 +++-- net/mptcp/pm_kernel.c | 4 ++-- net/mptcp/protocol.c | 7 ++++--- net/phonet/pep.c | 3 ++- net/phonet/socket.c | 4 ++-- net/sctp/socket.c | 9 +++++---- 31 files changed, 88 insertions(+), 77 deletions(-) diff --git a/fs/dlm/lowcomms.c b/fs/dlm/lowcomms.c index f832dafdaca8..b3958008ba3f 100644 --- a/fs/dlm/lowcomms.c +++ b/fs/dlm/lowcomms.c @@ -1126,7 +1126,7 @@ static void writequeue_entry_complete(struct writequeue_entry *e, int completed) static int sctp_bind_addrs(struct socket *sock, __be16 port) { struct sockaddr_storage localaddr; - struct sockaddr *addr = (struct sockaddr *)&localaddr; + struct sockaddr_unsized *addr = (struct sockaddr_unsized *)&localaddr; int i, addr_len, result = 0; for (i = 0; i < dlm_local_count; i++) { @@ -1134,7 +1134,7 @@ static int sctp_bind_addrs(struct socket *sock, __be16 port) make_sockaddr(&localaddr, port, &addr_len); if (!i) - result = kernel_bind(sock, (struct sockaddr_unsized *)addr, addr_len); + result = kernel_bind(sock, addr, addr_len); else result = sock_bind_add(sock->sk, addr, addr_len); diff --git a/include/net/inet_common.h b/include/net/inet_common.h index ebafd96912bb..5dd2bf24449e 100644 --- a/include/net/inet_common.h +++ b/include/net/inet_common.h @@ -19,7 +19,6 @@ struct msghdr; struct net; struct page; struct sock; -struct sockaddr; struct socket; int inet_release(struct socket *sock); @@ -43,7 +42,7 @@ int inet_listen(struct socket *sock, int backlog); int __inet_listen_sk(struct sock *sk, int backlog); void inet_sock_destruct(struct sock *sk); int inet_bind(struct socket *sock, struct sockaddr_unsized *uaddr, int addr_len); -int inet_bind_sk(struct sock *sk, struct sockaddr *uaddr, int addr_len); +int inet_bind_sk(struct sock *sk, struct sockaddr_unsized *uaddr, int addr_len); /* Don't allocate port at this moment, defer to connect. */ #define BIND_FORCE_ADDRESS_NO_PORT (1 << 0) /* Grab and release socket lock. */ @@ -52,7 +51,7 @@ int inet_bind_sk(struct sock *sk, struct sockaddr *uaddr, int addr_len); #define BIND_FROM_BPF (1 << 2) /* Skip CAP_NET_BIND_SERVICE check. */ #define BIND_NO_CAP_NET_BIND_SERVICE (1 << 3) -int __inet_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len, +int __inet_bind(struct sock *sk, struct sockaddr_unsized *uaddr, int addr_len, u32 flags); int inet_getname(struct socket *sock, struct sockaddr *uaddr, int peer); diff --git a/include/net/ip.h b/include/net/ip.h index 380afb691c41..69d5cef46004 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -261,8 +261,8 @@ static inline u8 ip_sendmsg_scope(const struct inet_sock *inet, } /* datagram.c */ -int __ip4_datagram_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len); -int ip4_datagram_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len); +int __ip4_datagram_connect(struct sock *sk, struct sockaddr_unsized *uaddr, int addr_len); +int ip4_datagram_connect(struct sock *sk, struct sockaddr_unsized *uaddr, int addr_len); void ip4_datagram_release_cb(struct sock *sk); diff --git a/include/net/ipv6.h b/include/net/ipv6.h index 2188bad9a687..74fbf1ad8065 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -1188,10 +1188,10 @@ int do_ipv6_getsockopt(struct sock *sk, int level, int optname, int ipv6_getsockopt(struct sock *sk, int level, int optname, char __user *optval, int __user *optlen); -int __ip6_datagram_connect(struct sock *sk, struct sockaddr *addr, +int __ip6_datagram_connect(struct sock *sk, struct sockaddr_unsized *addr, int addr_len); -int ip6_datagram_connect(struct sock *sk, struct sockaddr *addr, int addr_len); -int ip6_datagram_connect_v6_only(struct sock *sk, struct sockaddr *addr, +int ip6_datagram_connect(struct sock *sk, struct sockaddr_unsized *addr, int addr_len); +int ip6_datagram_connect_v6_only(struct sock *sk, struct sockaddr_unsized *addr, int addr_len); int ip6_datagram_dst_update(struct sock *sk, bool fix_sk_saddr); void ip6_datagram_release_cb(struct sock *sk); @@ -1209,7 +1209,7 @@ void inet6_cleanup_sock(struct sock *sk); void inet6_sock_destruct(struct sock *sk); int inet6_release(struct socket *sock); int inet6_bind(struct socket *sock, struct sockaddr_unsized *uaddr, int addr_len); -int inet6_bind_sk(struct sock *sk, struct sockaddr *uaddr, int addr_len); +int inet6_bind_sk(struct sock *sk, struct sockaddr_unsized *uaddr, int addr_len); int inet6_getname(struct socket *sock, struct sockaddr *uaddr, int peer); int inet6_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg); diff --git a/include/net/ipv6_stubs.h b/include/net/ipv6_stubs.h index 8a3465c8c2c5..d3013e721b14 100644 --- a/include/net/ipv6_stubs.h +++ b/include/net/ipv6_stubs.h @@ -80,7 +80,7 @@ extern const struct ipv6_stub *ipv6_stub __read_mostly; /* A stub used by bpf helpers. Similarly ugly as ipv6_stub */ struct ipv6_bpf_stub { - int (*inet6_bind)(struct sock *sk, struct sockaddr *uaddr, int addr_len, + int (*inet6_bind)(struct sock *sk, struct sockaddr_unsized *uaddr, int addr_len, u32 flags); struct sock *(*udp6_lib_lookup)(const struct net *net, const struct in6_addr *saddr, __be16 sport, diff --git a/include/net/ping.h b/include/net/ping.h index 9634b8800814..05bfd594a64c 100644 --- a/include/net/ping.h +++ b/include/net/ping.h @@ -58,7 +58,7 @@ void ping_unhash(struct sock *sk); int ping_init_sock(struct sock *sk); void ping_close(struct sock *sk, long timeout); -int ping_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len); +int ping_bind(struct sock *sk, struct sockaddr_unsized *uaddr, int addr_len); void ping_err(struct sk_buff *skb, int offset, u32 info); int ping_getfrag(void *from, char *to, int offset, int fraglen, int odd, struct sk_buff *); diff --git a/include/net/sock.h b/include/net/sock.h index 589fbce77217..a5f36ea9d46f 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -1274,10 +1274,10 @@ struct proto { void (*close)(struct sock *sk, long timeout); int (*pre_connect)(struct sock *sk, - struct sockaddr *uaddr, + struct sockaddr_unsized *uaddr, int addr_len); int (*connect)(struct sock *sk, - struct sockaddr *uaddr, + struct sockaddr_unsized *uaddr, int addr_len); int (*disconnect)(struct sock *sk, int flags); @@ -1306,9 +1306,9 @@ struct proto { size_t len, int flags, int *addr_len); void (*splice_eof)(struct socket *sock); int (*bind)(struct sock *sk, - struct sockaddr *addr, int addr_len); + struct sockaddr_unsized *addr, int addr_len); int (*bind_add)(struct sock *sk, - struct sockaddr *addr, int addr_len); + struct sockaddr_unsized *addr, int addr_len); int (*backlog_rcv) (struct sock *sk, struct sk_buff *skb); @@ -3105,7 +3105,7 @@ void sock_set_reuseaddr(struct sock *sk); void sock_set_reuseport(struct sock *sk); void sock_set_sndtimeo(struct sock *sk, s64 secs); -int sock_bind_add(struct sock *sk, struct sockaddr *addr, int addr_len); +int sock_bind_add(struct sock *sk, struct sockaddr_unsized *addr, int addr_len); int sock_get_timeout(long timeo, void *optval, bool old_timeval); int sock_copy_user_timeval(struct __kernel_sock_timeval *tv, diff --git a/include/net/tcp.h b/include/net/tcp.h index 4fd6d8d1230d..0aa1f07d036a 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -535,7 +535,7 @@ struct sock *tcp_v4_syn_recv_sock(const struct sock *sk, struct sk_buff *skb, struct request_sock *req_unhash, bool *own_req); int tcp_v4_do_rcv(struct sock *sk, struct sk_buff *skb); -int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len); +int tcp_v4_connect(struct sock *sk, struct sockaddr_unsized *uaddr, int addr_len); int tcp_connect(struct sock *sk); enum tcp_synack_type { TCP_SYNACK_NORMAL, diff --git a/include/net/udp.h b/include/net/udp.h index cffedb3e40f2..a061d1b22ddc 100644 --- a/include/net/udp.h +++ b/include/net/udp.h @@ -424,7 +424,7 @@ void udp4_hwcsum(struct sk_buff *skb, __be32 src, __be32 dst); int udp_rcv(struct sk_buff *skb); int udp_ioctl(struct sock *sk, int cmd, int *karg); int udp_init_sock(struct sock *sk); -int udp_pre_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len); +int udp_pre_connect(struct sock *sk, struct sockaddr_unsized *uaddr, int addr_len); int __udp_disconnect(struct sock *sk, int flags); int udp_disconnect(struct sock *sk, int flags); __poll_t udp_poll(struct file *file, struct socket *sock, poll_table *wait); diff --git a/net/core/filter.c b/net/core/filter.c index 16105f52927d..90273da74807 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -5978,7 +5978,7 @@ BPF_CALL_3(bpf_bind, struct bpf_sock_addr_kern *, ctx, struct sockaddr *, addr, return err; if (((struct sockaddr_in *)addr)->sin_port == htons(0)) flags |= BIND_FORCE_ADDRESS_NO_PORT; - return __inet_bind(sk, addr, addr_len, flags); + return __inet_bind(sk, (struct sockaddr_unsized *)addr, addr_len, flags); #if IS_ENABLED(CONFIG_IPV6) } else if (addr->sa_family == AF_INET6) { if (addr_len < SIN6_LEN_RFC2133) @@ -5988,7 +5988,8 @@ BPF_CALL_3(bpf_bind, struct bpf_sock_addr_kern *, ctx, struct sockaddr *, addr, /* ipv6_bpf_stub cannot be NULL, since it's called from * bpf_cgroup_inet6_connect hook and ipv6 is already loaded */ - return ipv6_bpf_stub->inet6_bind(sk, addr, addr_len, flags); + return ipv6_bpf_stub->inet6_bind(sk, (struct sockaddr_unsized *)addr, + addr_len, flags); #endif /* CONFIG_IPV6 */ } #endif /* CONFIG_INET */ diff --git a/net/core/sock.c b/net/core/sock.c index f97a0e958991..3b74fc71f51c 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -4395,7 +4395,7 @@ bool sk_busy_loop_end(void *p, unsigned long start_time) EXPORT_SYMBOL(sk_busy_loop_end); #endif /* CONFIG_NET_RX_BUSY_POLL */ -int sock_bind_add(struct sock *sk, struct sockaddr *addr, int addr_len) +int sock_bind_add(struct sock *sk, struct sockaddr_unsized *addr, int addr_len) { if (!sk->sk_prot->bind_add) return -EOPNOTSUPP; diff --git a/net/ieee802154/socket.c b/net/ieee802154/socket.c index b93fd85f248a..e542fbe113e7 100644 --- a/net/ieee802154/socket.c +++ b/net/ieee802154/socket.c @@ -102,7 +102,7 @@ static int ieee802154_sock_bind(struct socket *sock, struct sockaddr_unsized *ua struct sock *sk = sock->sk; if (sk->sk_prot->bind) - return sk->sk_prot->bind(sk, (struct sockaddr *)uaddr, addr_len); + return sk->sk_prot->bind(sk, uaddr, addr_len); return sock_no_bind(sock, uaddr, addr_len); } @@ -118,7 +118,7 @@ static int ieee802154_sock_connect(struct socket *sock, struct sockaddr_unsized if (uaddr->sa_family == AF_UNSPEC) return sk->sk_prot->disconnect(sk, flags); - return sk->sk_prot->connect(sk, (struct sockaddr *)uaddr, addr_len); + return sk->sk_prot->connect(sk, uaddr, addr_len); } static int ieee802154_dev_ioctl(struct sock *sk, struct ifreq __user *arg, @@ -193,7 +193,7 @@ static void raw_close(struct sock *sk, long timeout) sk_common_release(sk); } -static int raw_bind(struct sock *sk, struct sockaddr *_uaddr, int len) +static int raw_bind(struct sock *sk, struct sockaddr_unsized *_uaddr, int len) { struct ieee802154_addr addr; struct sockaddr_ieee802154 *uaddr = (struct sockaddr_ieee802154 *)_uaddr; @@ -227,7 +227,7 @@ static int raw_bind(struct sock *sk, struct sockaddr *_uaddr, int len) return err; } -static int raw_connect(struct sock *sk, struct sockaddr *uaddr, +static int raw_connect(struct sock *sk, struct sockaddr_unsized *uaddr, int addr_len) { return -ENOTSUPP; @@ -485,7 +485,7 @@ static void dgram_close(struct sock *sk, long timeout) sk_common_release(sk); } -static int dgram_bind(struct sock *sk, struct sockaddr *uaddr, int len) +static int dgram_bind(struct sock *sk, struct sockaddr_unsized *uaddr, int len) { struct sockaddr_ieee802154 *addr = (struct sockaddr_ieee802154 *)uaddr; struct ieee802154_addr haddr; @@ -563,7 +563,7 @@ static int dgram_ioctl(struct sock *sk, int cmd, int *karg) } /* FIXME: autobind */ -static int dgram_connect(struct sock *sk, struct sockaddr *uaddr, +static int dgram_connect(struct sock *sk, struct sockaddr_unsized *uaddr, int len) { struct sockaddr_ieee802154 *addr = (struct sockaddr_ieee802154 *)uaddr; diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index 0844de9ac6a4..d5ac089356eb 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -441,7 +441,7 @@ int inet_release(struct socket *sock) } EXPORT_SYMBOL(inet_release); -int inet_bind_sk(struct sock *sk, struct sockaddr *uaddr, int addr_len) +int inet_bind_sk(struct sock *sk, struct sockaddr_unsized *uaddr, int addr_len) { u32 flags = BIND_WITH_LOCK; int err; @@ -466,11 +466,11 @@ int inet_bind_sk(struct sock *sk, struct sockaddr *uaddr, int addr_len) int inet_bind(struct socket *sock, struct sockaddr_unsized *uaddr, int addr_len) { - return inet_bind_sk(sock->sk, (struct sockaddr *)uaddr, addr_len); + return inet_bind_sk(sock->sk, uaddr, addr_len); } EXPORT_SYMBOL(inet_bind); -int __inet_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len, +int __inet_bind(struct sock *sk, struct sockaddr_unsized *uaddr, int addr_len, u32 flags) { struct sockaddr_in *addr = (struct sockaddr_in *)uaddr; @@ -584,14 +584,14 @@ int inet_dgram_connect(struct socket *sock, struct sockaddr_unsized *uaddr, return prot->disconnect(sk, flags); if (BPF_CGROUP_PRE_CONNECT_ENABLED(sk)) { - err = prot->pre_connect(sk, (struct sockaddr *)uaddr, addr_len); + err = prot->pre_connect(sk, uaddr, addr_len); if (err) return err; } if (data_race(!inet_sk(sk)->inet_num) && inet_autobind(sk)) return -EAGAIN; - return prot->connect(sk, (struct sockaddr *)uaddr, addr_len); + return prot->connect(sk, uaddr, addr_len); } EXPORT_SYMBOL(inet_dgram_connect); @@ -671,12 +671,12 @@ int __inet_stream_connect(struct socket *sock, struct sockaddr_unsized *uaddr, goto out; if (BPF_CGROUP_PRE_CONNECT_ENABLED(sk)) { - err = sk->sk_prot->pre_connect(sk, (struct sockaddr *)uaddr, addr_len); + err = sk->sk_prot->pre_connect(sk, uaddr, addr_len); if (err) goto out; } - err = sk->sk_prot->connect(sk, (struct sockaddr *)uaddr, addr_len); + err = sk->sk_prot->connect(sk, uaddr, addr_len); if (err < 0) goto out; diff --git a/net/ipv4/datagram.c b/net/ipv4/datagram.c index c2b2cda1a7e5..1614593b6d72 100644 --- a/net/ipv4/datagram.c +++ b/net/ipv4/datagram.c @@ -16,7 +16,7 @@ #include #include -int __ip4_datagram_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) +int __ip4_datagram_connect(struct sock *sk, struct sockaddr_unsized *uaddr, int addr_len) { struct inet_sock *inet = inet_sk(sk); struct sockaddr_in *usin = (struct sockaddr_in *) uaddr; @@ -84,7 +84,7 @@ int __ip4_datagram_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len } EXPORT_SYMBOL(__ip4_datagram_connect); -int ip4_datagram_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) +int ip4_datagram_connect(struct sock *sk, struct sockaddr_unsized *uaddr, int addr_len) { int res; diff --git a/net/ipv4/ping.c b/net/ipv4/ping.c index 5321c5801c64..ad56588107cc 100644 --- a/net/ipv4/ping.c +++ b/net/ipv4/ping.c @@ -286,7 +286,7 @@ void ping_close(struct sock *sk, long timeout) } EXPORT_IPV6_MOD_GPL(ping_close); -static int ping_pre_connect(struct sock *sk, struct sockaddr *uaddr, +static int ping_pre_connect(struct sock *sk, struct sockaddr_unsized *uaddr, int addr_len) { /* This check is replicated from __ip4_datagram_connect() and @@ -301,7 +301,7 @@ static int ping_pre_connect(struct sock *sk, struct sockaddr *uaddr, /* Checks the bind address and possibly modifies sk->sk_bound_dev_if. */ static int ping_check_bind_addr(struct sock *sk, struct inet_sock *isk, - struct sockaddr *uaddr, int addr_len) + struct sockaddr_unsized *uaddr, int addr_len) { struct net *net = sock_net(sk); if (sk->sk_family == AF_INET) { @@ -387,7 +387,7 @@ static int ping_check_bind_addr(struct sock *sk, struct inet_sock *isk, return 0; } -static void ping_set_saddr(struct sock *sk, struct sockaddr *saddr) +static void ping_set_saddr(struct sock *sk, struct sockaddr_unsized *saddr) { if (saddr->sa_family == AF_INET) { struct inet_sock *isk = inet_sk(sk); @@ -407,7 +407,7 @@ static void ping_set_saddr(struct sock *sk, struct sockaddr *saddr) * Moreover, we don't allow binding to multi- and broadcast addresses. */ -int ping_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len) +int ping_bind(struct sock *sk, struct sockaddr_unsized *uaddr, int addr_len) { struct inet_sock *isk = inet_sk(sk); unsigned short snum; diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c index d54ebb7df966..5998c4cc6f47 100644 --- a/net/ipv4/raw.c +++ b/net/ipv4/raw.c @@ -697,7 +697,8 @@ static void raw_destroy(struct sock *sk) } /* This gets rid of all the nasties in af_inet. -DaveM */ -static int raw_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len) +static int raw_bind(struct sock *sk, struct sockaddr_unsized *uaddr, + int addr_len) { struct inet_sock *inet = inet_sk(sk); struct sockaddr_in *addr = (struct sockaddr_in *) uaddr; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 40a76da5364a..b7526a7888cb 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -205,7 +205,7 @@ int tcp_twsk_unique(struct sock *sk, struct sock *sktw, void *twp) } EXPORT_IPV6_MOD_GPL(tcp_twsk_unique); -static int tcp_v4_pre_connect(struct sock *sk, struct sockaddr *uaddr, +static int tcp_v4_pre_connect(struct sock *sk, struct sockaddr_unsized *uaddr, int addr_len) { /* This check is replicated from tcp_v4_connect() and intended to @@ -221,7 +221,7 @@ static int tcp_v4_pre_connect(struct sock *sk, struct sockaddr *uaddr, } /* This will initiate an outgoing connection. */ -int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) +int tcp_v4_connect(struct sock *sk, struct sockaddr_unsized *uaddr, int addr_len) { struct sockaddr_in *usin = (struct sockaddr_in *)uaddr; struct inet_timewait_death_row *tcp_death_row; diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 30dfbf73729d..ffe074cb5865 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -2159,7 +2159,8 @@ int udp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int flags, goto try_again; } -int udp_pre_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) +int udp_pre_connect(struct sock *sk, struct sockaddr_unsized *uaddr, + int addr_len) { /* This check is replicated from __ip4_datagram_connect() and * intended to prevent BPF program called below from accessing bytes @@ -2172,7 +2173,8 @@ int udp_pre_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) } EXPORT_IPV6_MOD(udp_pre_connect); -static int udp_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) +static int udp_connect(struct sock *sk, struct sockaddr_unsized *uaddr, + int addr_len) { int res; diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index c92d27e35fbc..b705751eb73c 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -277,7 +277,7 @@ static int inet6_create(struct net *net, struct socket *sock, int protocol, goto out; } -static int __inet6_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len, +static int __inet6_bind(struct sock *sk, struct sockaddr_unsized *uaddr, int addr_len, u32 flags) { struct sockaddr_in6 *addr = (struct sockaddr_in6 *)uaddr; @@ -438,7 +438,7 @@ static int __inet6_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len, goto out; } -int inet6_bind_sk(struct sock *sk, struct sockaddr *uaddr, int addr_len) +int inet6_bind_sk(struct sock *sk, struct sockaddr_unsized *uaddr, int addr_len) { u32 flags = BIND_WITH_LOCK; const struct proto *prot; @@ -467,7 +467,7 @@ int inet6_bind_sk(struct sock *sk, struct sockaddr *uaddr, int addr_len) /* bind for INET6 API */ int inet6_bind(struct socket *sock, struct sockaddr_unsized *uaddr, int addr_len) { - return inet6_bind_sk(sock->sk, (struct sockaddr *)uaddr, addr_len); + return inet6_bind_sk(sock->sk, uaddr, addr_len); } EXPORT_SYMBOL(inet6_bind); diff --git a/net/ipv6/datagram.c b/net/ipv6/datagram.c index 33ebe93d80e3..83e03176819c 100644 --- a/net/ipv6/datagram.c +++ b/net/ipv6/datagram.c @@ -138,7 +138,7 @@ void ip6_datagram_release_cb(struct sock *sk) } EXPORT_SYMBOL_GPL(ip6_datagram_release_cb); -int __ip6_datagram_connect(struct sock *sk, struct sockaddr *uaddr, +int __ip6_datagram_connect(struct sock *sk, struct sockaddr_unsized *uaddr, int addr_len) { struct sockaddr_in6 *usin = (struct sockaddr_in6 *) uaddr; @@ -194,7 +194,7 @@ int __ip6_datagram_connect(struct sock *sk, struct sockaddr *uaddr, sin.sin_port = usin->sin6_port; err = __ip4_datagram_connect(sk, - (struct sockaddr *) &sin, + (struct sockaddr_unsized *)&sin, sizeof(sin)); ipv4_connected: @@ -271,7 +271,7 @@ int __ip6_datagram_connect(struct sock *sk, struct sockaddr *uaddr, } EXPORT_SYMBOL_GPL(__ip6_datagram_connect); -int ip6_datagram_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) +int ip6_datagram_connect(struct sock *sk, struct sockaddr_unsized *uaddr, int addr_len) { int res; @@ -282,7 +282,7 @@ int ip6_datagram_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) } EXPORT_SYMBOL_GPL(ip6_datagram_connect); -int ip6_datagram_connect_v6_only(struct sock *sk, struct sockaddr *uaddr, +int ip6_datagram_connect_v6_only(struct sock *sk, struct sockaddr_unsized *uaddr, int addr_len) { DECLARE_SOCKADDR(struct sockaddr_in6 *, sin6, uaddr); diff --git a/net/ipv6/ping.c b/net/ipv6/ping.c index d7a2cdaa2631..e4afc651731a 100644 --- a/net/ipv6/ping.c +++ b/net/ipv6/ping.c @@ -45,7 +45,7 @@ static int dummy_ipv6_chk_addr(struct net *net, const struct in6_addr *addr, return 0; } -static int ping_v6_pre_connect(struct sock *sk, struct sockaddr *uaddr, +static int ping_v6_pre_connect(struct sock *sk, struct sockaddr_unsized *uaddr, int addr_len) { /* This check is replicated from __ip6_datagram_connect() and diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index e369f54844dd..b4cd05dba9b6 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -214,7 +214,8 @@ bool raw6_local_deliver(struct sk_buff *skb, int nexthdr) } /* This cleans up af_inet6 a bit. -DaveM */ -static int rawv6_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len) +static int rawv6_bind(struct sock *sk, struct sockaddr_unsized *uaddr, + int addr_len) { struct inet_sock *inet = inet_sk(sk); struct ipv6_pinfo *np = inet6_sk(sk); diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 06eb90e4078e..7df21c1cba21 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -118,7 +118,7 @@ static u32 tcp_v6_init_ts_off(const struct net *net, const struct sk_buff *skb) ipv6_hdr(skb)->saddr.s6_addr32); } -static int tcp_v6_pre_connect(struct sock *sk, struct sockaddr *uaddr, +static int tcp_v6_pre_connect(struct sock *sk, struct sockaddr_unsized *uaddr, int addr_len) { /* This check is replicated from tcp_v6_connect() and intended to @@ -133,7 +133,7 @@ static int tcp_v6_pre_connect(struct sock *sk, struct sockaddr *uaddr, return BPF_CGROUP_RUN_PROG_INET6_CONNECT(sk, uaddr, &addr_len); } -static int tcp_v6_connect(struct sock *sk, struct sockaddr *uaddr, +static int tcp_v6_connect(struct sock *sk, struct sockaddr_unsized *uaddr, int addr_len) { struct sockaddr_in6 *usin = (struct sockaddr_in6 *) uaddr; @@ -238,7 +238,7 @@ static int tcp_v6_connect(struct sock *sk, struct sockaddr *uaddr, tp->af_specific = &tcp_sock_ipv6_mapped_specific; #endif - err = tcp_v4_connect(sk, (struct sockaddr *)&sin, sizeof(sin)); + err = tcp_v4_connect(sk, (struct sockaddr_unsized *)&sin, sizeof(sin)); if (err) { icsk->icsk_ext_hdr_len = exthdrlen; diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index 813a2ba75824..794c13674e8a 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -1282,7 +1282,7 @@ static void udp_v6_flush_pending_frames(struct sock *sk) } } -static int udpv6_pre_connect(struct sock *sk, struct sockaddr *uaddr, +static int udpv6_pre_connect(struct sock *sk, struct sockaddr_unsized *uaddr, int addr_len) { if (addr_len < offsetofend(struct sockaddr, sa_family)) @@ -1303,7 +1303,8 @@ static int udpv6_pre_connect(struct sock *sk, struct sockaddr *uaddr, return BPF_CGROUP_RUN_PROG_INET6_CONNECT_LOCK(sk, uaddr, &addr_len); } -static int udpv6_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) +static int udpv6_connect(struct sock *sk, struct sockaddr_unsized *uaddr, + int addr_len) { int res; diff --git a/net/l2tp/l2tp_ip.c b/net/l2tp/l2tp_ip.c index 29795d2839e8..cac1ff59cb83 100644 --- a/net/l2tp/l2tp_ip.c +++ b/net/l2tp/l2tp_ip.c @@ -267,7 +267,8 @@ static void l2tp_ip_destroy_sock(struct sock *sk) } } -static int l2tp_ip_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len) +static int l2tp_ip_bind(struct sock *sk, struct sockaddr_unsized *uaddr, + int addr_len) { struct inet_sock *inet = inet_sk(sk); struct sockaddr_l2tpip *addr = (struct sockaddr_l2tpip *)uaddr; @@ -328,7 +329,8 @@ static int l2tp_ip_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len) return ret; } -static int l2tp_ip_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) +static int l2tp_ip_connect(struct sock *sk, struct sockaddr_unsized *uaddr, + int addr_len) { struct sockaddr_l2tpip *lsa = (struct sockaddr_l2tpip *)uaddr; struct l2tp_ip_net *pn = l2tp_ip_pernet(sock_net(sk)); diff --git a/net/l2tp/l2tp_ip6.c b/net/l2tp/l2tp_ip6.c index ea232f338dcb..05a396ba6a3e 100644 --- a/net/l2tp/l2tp_ip6.c +++ b/net/l2tp/l2tp_ip6.c @@ -280,7 +280,8 @@ static void l2tp_ip6_destroy_sock(struct sock *sk) } } -static int l2tp_ip6_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len) +static int l2tp_ip6_bind(struct sock *sk, struct sockaddr_unsized *uaddr, + int addr_len) { struct inet_sock *inet = inet_sk(sk); struct ipv6_pinfo *np = inet6_sk(sk); @@ -383,7 +384,7 @@ static int l2tp_ip6_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len) return err; } -static int l2tp_ip6_connect(struct sock *sk, struct sockaddr *uaddr, +static int l2tp_ip6_connect(struct sock *sk, struct sockaddr_unsized *uaddr, int addr_len) { struct sockaddr_l2tpip6 *lsa = (struct sockaddr_l2tpip6 *)uaddr; diff --git a/net/mptcp/pm_kernel.c b/net/mptcp/pm_kernel.c index e50721c670d0..598f01a573c1 100644 --- a/net/mptcp/pm_kernel.c +++ b/net/mptcp/pm_kernel.c @@ -867,10 +867,10 @@ static int mptcp_pm_nl_create_listen_socket(struct sock *sk, addrlen = sizeof(struct sockaddr_in6); #endif if (ssk->sk_family == AF_INET) - err = inet_bind_sk(ssk, (struct sockaddr *)&addr, addrlen); + err = inet_bind_sk(ssk, (struct sockaddr_unsized *)&addr, addrlen); #if IS_ENABLED(CONFIG_MPTCP_IPV6) else if (ssk->sk_family == AF_INET6) - err = inet6_bind_sk(ssk, (struct sockaddr *)&addr, addrlen); + err = inet6_bind_sk(ssk, (struct sockaddr_unsized *)&addr, addrlen); #endif if (err) return err; diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index 53e2b095dfb1..4cd5df01446e 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -3746,7 +3746,8 @@ static int mptcp_ioctl(struct sock *sk, int cmd, int *karg) return 0; } -static int mptcp_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) +static int mptcp_connect(struct sock *sk, struct sockaddr_unsized *uaddr, + int addr_len) { struct mptcp_subflow_context *subflow; struct mptcp_sock *msk = mptcp_sk(sk); @@ -3870,10 +3871,10 @@ static int mptcp_bind(struct socket *sock, struct sockaddr_unsized *uaddr, int a } if (sk->sk_family == AF_INET) - err = inet_bind_sk(ssk, (struct sockaddr *)uaddr, addr_len); + err = inet_bind_sk(ssk, uaddr, addr_len); #if IS_ENABLED(CONFIG_MPTCP_IPV6) else if (sk->sk_family == AF_INET6) - err = inet6_bind_sk(ssk, (struct sockaddr *)uaddr, addr_len); + err = inet6_bind_sk(ssk, uaddr, addr_len); #endif if (!err) mptcp_copy_inaddrs(sk, ssk); diff --git a/net/phonet/pep.c b/net/phonet/pep.c index 4db564d9d522..120e711ea78c 100644 --- a/net/phonet/pep.c +++ b/net/phonet/pep.c @@ -882,7 +882,8 @@ static struct sock *pep_sock_accept(struct sock *sk, return newsk; } -static int pep_sock_connect(struct sock *sk, struct sockaddr *addr, int len) +static int pep_sock_connect(struct sock *sk, struct sockaddr_unsized *addr, + int len) { struct pep_sock *pn = pep_sk(sk); int err; diff --git a/net/phonet/socket.c b/net/phonet/socket.c index 9391378083a4..4423d483c630 100644 --- a/net/phonet/socket.c +++ b/net/phonet/socket.c @@ -163,7 +163,7 @@ static int pn_socket_bind(struct socket *sock, struct sockaddr_unsized *addr, in u8 saddr; if (sk->sk_prot->bind) - return sk->sk_prot->bind(sk, (struct sockaddr *)addr, len); + return sk->sk_prot->bind(sk, addr, len); if (len < sizeof(struct sockaddr_pn)) return -EINVAL; @@ -252,7 +252,7 @@ static int pn_socket_connect(struct socket *sock, struct sockaddr_unsized *addr, pn->resource = pn_sockaddr_get_resource(spn); sock->state = SS_CONNECTING; - err = sk->sk_prot->connect(sk, (struct sockaddr *)addr, len); + err = sk->sk_prot->connect(sk, addr, len); if (err) { sock->state = SS_UNCONNECTED; pn->dobject = 0; diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 940abbced191..38d2932acebf 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -306,7 +306,8 @@ static struct sctp_transport *sctp_addr_id2transport(struct sock *sk, * sockaddr_in6 [RFC 2553]), * addr_len - the size of the address structure. */ -static int sctp_bind(struct sock *sk, struct sockaddr *addr, int addr_len) +static int sctp_bind(struct sock *sk, struct sockaddr_unsized *addr, + int addr_len) { int retval = 0; @@ -1053,13 +1054,13 @@ static int sctp_setsockopt_bindx(struct sock *sk, struct sockaddr *addrs, } } -static int sctp_bind_add(struct sock *sk, struct sockaddr *addrs, - int addrlen) +static int sctp_bind_add(struct sock *sk, struct sockaddr_unsized *addrs, + int addrlen) { int err; lock_sock(sk); - err = sctp_setsockopt_bindx(sk, addrs, addrlen, SCTP_BINDX_ADD_ADDR); + err = sctp_setsockopt_bindx(sk, (struct sockaddr *)addrs, addrlen, SCTP_BINDX_ADD_ADDR); release_sock(sk); return err; } From 8116d803e7f8f20bf00ce23ff8bd0baab41e1635 Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Mon, 3 Nov 2025 16:26:14 -0800 Subject: [PATCH 0537/1381] bpf: Convert cgroup sockaddr filters to use sockaddr_unsized consistently Update BPF cgroup sockaddr filtering infrastructure to use sockaddr_unsized consistently throughout the call chain, removing redundant explicit casts from callers. No binary changes expected. Signed-off-by: Kees Cook Link: https://patch.msgid.link/20251104002617.2752303-6-kees@kernel.org Signed-off-by: Jakub Kicinski --- include/linux/bpf-cgroup.h | 17 ++++++++++------- kernel/bpf/cgroup.c | 4 ++-- net/ipv4/af_inet.c | 4 ++-- 3 files changed, 14 insertions(+), 11 deletions(-) diff --git a/include/linux/bpf-cgroup.h b/include/linux/bpf-cgroup.h index a7fb4f46974f..d1eb5c7729cb 100644 --- a/include/linux/bpf-cgroup.h +++ b/include/linux/bpf-cgroup.h @@ -120,7 +120,7 @@ int __cgroup_bpf_run_filter_sk(struct sock *sk, enum cgroup_bpf_attach_type atype); int __cgroup_bpf_run_filter_sock_addr(struct sock *sk, - struct sockaddr *uaddr, + struct sockaddr_unsized *uaddr, int *uaddrlen, enum cgroup_bpf_attach_type atype, void *t_ctx, @@ -238,8 +238,9 @@ static inline bool cgroup_bpf_sock_enabled(struct sock *sk, ({ \ int __ret = 0; \ if (cgroup_bpf_enabled(atype)) \ - __ret = __cgroup_bpf_run_filter_sock_addr(sk, (struct sockaddr *)uaddr, uaddrlen, \ - atype, NULL, NULL); \ + __ret = __cgroup_bpf_run_filter_sock_addr(sk, \ + (struct sockaddr_unsized *)uaddr, uaddrlen, \ + atype, NULL, NULL); \ __ret; \ }) @@ -248,8 +249,9 @@ static inline bool cgroup_bpf_sock_enabled(struct sock *sk, int __ret = 0; \ if (cgroup_bpf_enabled(atype)) { \ lock_sock(sk); \ - __ret = __cgroup_bpf_run_filter_sock_addr(sk, (struct sockaddr *)uaddr, uaddrlen, \ - atype, t_ctx, NULL); \ + __ret = __cgroup_bpf_run_filter_sock_addr(sk, \ + (struct sockaddr_unsized *)uaddr, uaddrlen, \ + atype, t_ctx, NULL); \ release_sock(sk); \ } \ __ret; \ @@ -266,8 +268,9 @@ static inline bool cgroup_bpf_sock_enabled(struct sock *sk, int __ret = 0; \ if (cgroup_bpf_enabled(atype)) { \ lock_sock(sk); \ - __ret = __cgroup_bpf_run_filter_sock_addr(sk, (struct sockaddr *)uaddr, uaddrlen, \ - atype, NULL, &__flags); \ + __ret = __cgroup_bpf_run_filter_sock_addr(sk, \ + (struct sockaddr_unsized *)uaddr, uaddrlen, \ + atype, NULL, &__flags); \ release_sock(sk); \ if (__flags & BPF_RET_BIND_NO_CAP_NET_BIND_SERVICE) \ *bind_flags |= BIND_NO_CAP_NET_BIND_SERVICE; \ diff --git a/kernel/bpf/cgroup.c b/kernel/bpf/cgroup.c index 248f517d66d0..497aedc9afa1 100644 --- a/kernel/bpf/cgroup.c +++ b/kernel/bpf/cgroup.c @@ -1665,7 +1665,7 @@ EXPORT_SYMBOL(__cgroup_bpf_run_filter_sk); * returned value != 1 during execution. In all other cases, 0 is returned. */ int __cgroup_bpf_run_filter_sock_addr(struct sock *sk, - struct sockaddr *uaddr, + struct sockaddr_unsized *uaddr, int *uaddrlen, enum cgroup_bpf_attach_type atype, void *t_ctx, @@ -1673,7 +1673,7 @@ int __cgroup_bpf_run_filter_sock_addr(struct sock *sk, { struct bpf_sock_addr_kern ctx = { .sk = sk, - .uaddr = uaddr, + .uaddr = (struct sockaddr *)uaddr, .t_ctx = t_ctx, }; struct sockaddr_storage unspec; diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index d5ac089356eb..a31b94ce8968 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -834,7 +834,7 @@ int inet_getname(struct socket *sock, struct sockaddr *uaddr, } sin->sin_port = inet->inet_dport; sin->sin_addr.s_addr = inet->inet_daddr; - BPF_CGROUP_RUN_SA_PROG(sk, (struct sockaddr *)sin, &sin_addr_len, + BPF_CGROUP_RUN_SA_PROG(sk, sin, &sin_addr_len, CGROUP_INET4_GETPEERNAME); } else { __be32 addr = inet->inet_rcv_saddr; @@ -842,7 +842,7 @@ int inet_getname(struct socket *sock, struct sockaddr *uaddr, addr = inet->inet_saddr; sin->sin_port = inet->inet_sport; sin->sin_addr.s_addr = addr; - BPF_CGROUP_RUN_SA_PROG(sk, (struct sockaddr *)sin, &sin_addr_len, + BPF_CGROUP_RUN_SA_PROG(sk, sin, &sin_addr_len, CGROUP_INET4_GETSOCKNAME); } release_sock(sk); From c1a799eef62b8c3298a4d82753fe0f2a448e5e4f Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Mon, 3 Nov 2025 16:26:15 -0800 Subject: [PATCH 0538/1381] bpf: Convert bpf_sock_addr_kern "uaddr" to sockaddr_unsized Change struct bpf_sock_addr_kern to use sockaddr_unsized for the "uaddr" field instead of sockaddr. This improves type safety in the BPF cgroup socket address filtering code. The casting in __cgroup_bpf_run_filter_sock_addr() is updated to match the new type, removing an unnecessary cast in the initialization and updating the conditional assignment to use the appropriate sockaddr_unsized cast. Additionally rename the "unspec" variable to "storage" to better align with its usage. No binary changes expected. Signed-off-by: Kees Cook Link: https://patch.msgid.link/20251104002617.2752303-7-kees@kernel.org Signed-off-by: Jakub Kicinski --- include/linux/filter.h | 2 +- kernel/bpf/cgroup.c | 8 ++++---- 2 files changed, 5 insertions(+), 5 deletions(-) diff --git a/include/linux/filter.h b/include/linux/filter.h index f5c859b8131a..e116de7edc58 100644 --- a/include/linux/filter.h +++ b/include/linux/filter.h @@ -1515,7 +1515,7 @@ static inline int bpf_tell_extensions(void) struct bpf_sock_addr_kern { struct sock *sk; - struct sockaddr *uaddr; + struct sockaddr_unsized *uaddr; /* Temporary "register" to make indirect stores to nested structures * defined above. We need three registers to make such a store, but * only two (src and dst) are available at convert_ctx_access time diff --git a/kernel/bpf/cgroup.c b/kernel/bpf/cgroup.c index 497aedc9afa1..69988af44b37 100644 --- a/kernel/bpf/cgroup.c +++ b/kernel/bpf/cgroup.c @@ -1673,10 +1673,10 @@ int __cgroup_bpf_run_filter_sock_addr(struct sock *sk, { struct bpf_sock_addr_kern ctx = { .sk = sk, - .uaddr = (struct sockaddr *)uaddr, + .uaddr = uaddr, .t_ctx = t_ctx, }; - struct sockaddr_storage unspec; + struct sockaddr_storage storage; struct cgroup *cgrp; int ret; @@ -1688,8 +1688,8 @@ int __cgroup_bpf_run_filter_sock_addr(struct sock *sk, return 0; if (!ctx.uaddr) { - memset(&unspec, 0, sizeof(unspec)); - ctx.uaddr = (struct sockaddr *)&unspec; + memset(&storage, 0, sizeof(storage)); + ctx.uaddr = (struct sockaddr_unsized *)&storage; ctx.uaddrlen = 0; } else { ctx.uaddrlen = *uaddrlen; From 2b5e9f9b7e414c5eeb20dd7a7b80816ff55cf57b Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Mon, 3 Nov 2025 16:26:16 -0800 Subject: [PATCH 0539/1381] net: Convert struct sockaddr to fixed-size "sa_data[14]" Revert struct sockaddr from flexible array to fixed 14-byte "sa_data", to solve over 36,000 -Wflex-array-member-not-at-end warnings, since struct sockaddr is embedded within many network structs. With socket/proto sockaddr-based internal APIs switched to use struct sockaddr_unsized, there should be no more uses of struct sockaddr that depend on reading beyond the end of struct sockaddr::sa_data that might trigger bounds checking. Comparing an x86_64 "allyesconfig" vmlinux build before and after this patch showed no new "ud1" instructions from CONFIG_UBSAN_BOUNDS nor any new "field-spanning" memcpy CONFIG_FORTIFY_SOURCE instrumentations. Cc: Gustavo A. R. Silva Signed-off-by: Kees Cook Link: https://patch.msgid.link/20251104002617.2752303-8-kees@kernel.org Signed-off-by: Jakub Kicinski --- include/linux/socket.h | 6 ++---- net/core/dev.c | 2 +- net/core/dev_ioctl.c | 2 +- net/ipv4/arp.c | 2 +- net/packet/af_packet.c | 10 +++++----- tools/perf/trace/beauty/include/linux/socket.h | 5 +---- 6 files changed, 11 insertions(+), 16 deletions(-) diff --git a/include/linux/socket.h b/include/linux/socket.h index 7b1a01be29da..944027f9765e 100644 --- a/include/linux/socket.h +++ b/include/linux/socket.h @@ -32,12 +32,10 @@ typedef __kernel_sa_family_t sa_family_t; * 1003.1g requires sa_family_t and that sa_data is char. */ +/* Deprecated for in-kernel use. Use struct sockaddr_unsized instead. */ struct sockaddr { sa_family_t sa_family; /* address family, AF_xxx */ - union { - char sa_data_min[14]; /* Minimum 14 bytes of protocol address */ - DECLARE_FLEX_ARRAY(char, sa_data); - }; + char sa_data[14]; /* 14 bytes of protocol address */ }; /** diff --git a/net/core/dev.c b/net/core/dev.c index ba39146bbd25..537aa43edff0 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -9973,7 +9973,7 @@ DECLARE_RWSEM(dev_addr_sem); /* "sa" is a true struct sockaddr with limited "sa_data" member. */ int netif_get_mac_address(struct sockaddr *sa, struct net *net, char *dev_name) { - size_t size = sizeof(sa->sa_data_min); + size_t size = sizeof(sa->sa_data); struct net_device *dev; int ret = 0; diff --git a/net/core/dev_ioctl.c b/net/core/dev_ioctl.c index ad54b12d4b4c..b3ce0fb24a69 100644 --- a/net/core/dev_ioctl.c +++ b/net/core/dev_ioctl.c @@ -596,7 +596,7 @@ static int dev_ifsioc(struct net *net, struct ifreq *ifr, void __user *data, if (ifr->ifr_hwaddr.sa_family != dev->type) return -EINVAL; memcpy(dev->broadcast, ifr->ifr_hwaddr.sa_data, - min(sizeof(ifr->ifr_hwaddr.sa_data_min), + min(sizeof(ifr->ifr_hwaddr.sa_data), (size_t)dev->addr_len)); netdev_lock_ops(dev); call_netdevice_notifiers(NETDEV_CHANGEADDR, dev); diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c index f3bfecf8a234..7f3863daaa40 100644 --- a/net/ipv4/arp.c +++ b/net/ipv4/arp.c @@ -1189,7 +1189,7 @@ static int arp_req_get(struct net *net, struct arpreq *r) read_lock_bh(&neigh->lock); memcpy(r->arp_ha.sa_data, neigh->ha, - min(dev->addr_len, sizeof(r->arp_ha.sa_data_min))); + min(dev->addr_len, sizeof(r->arp_ha.sa_data))); r->arp_flags = arp_state_to_flags(neigh); read_unlock_bh(&neigh->lock); diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index fccad2a529cc..494d628d10a5 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -3284,7 +3284,7 @@ static int packet_bind_spkt(struct socket *sock, struct sockaddr_unsized *uaddr, { struct sock *sk = sock->sk; struct sockaddr *sa = (struct sockaddr *)uaddr; - char name[sizeof(sa->sa_data_min) + 1]; + char name[sizeof(sa->sa_data) + 1]; /* * Check legality @@ -3295,8 +3295,8 @@ static int packet_bind_spkt(struct socket *sock, struct sockaddr_unsized *uaddr, /* uaddr->sa_data comes from the userspace, it's not guaranteed to be * zero-terminated. */ - memcpy(name, sa->sa_data, sizeof(sa->sa_data_min)); - name[sizeof(sa->sa_data_min)] = 0; + memcpy(name, sa->sa_data, sizeof(sa->sa_data)); + name[sizeof(sa->sa_data)] = 0; return packet_do_bind(sk, name, 0, 0); } @@ -3581,11 +3581,11 @@ static int packet_getname_spkt(struct socket *sock, struct sockaddr *uaddr, return -EOPNOTSUPP; uaddr->sa_family = AF_PACKET; - memset(uaddr->sa_data, 0, sizeof(uaddr->sa_data_min)); + memset(uaddr->sa_data, 0, sizeof(uaddr->sa_data)); rcu_read_lock(); dev = dev_get_by_index_rcu(sock_net(sk), READ_ONCE(pkt_sk(sk)->ifindex)); if (dev) - strscpy(uaddr->sa_data, dev->name, sizeof(uaddr->sa_data_min)); + strscpy(uaddr->sa_data, dev->name, sizeof(uaddr->sa_data)); rcu_read_unlock(); return sizeof(*uaddr); diff --git a/tools/perf/trace/beauty/include/linux/socket.h b/tools/perf/trace/beauty/include/linux/socket.h index 3b262487ec06..77d7c59f5d8b 100644 --- a/tools/perf/trace/beauty/include/linux/socket.h +++ b/tools/perf/trace/beauty/include/linux/socket.h @@ -34,10 +34,7 @@ typedef __kernel_sa_family_t sa_family_t; struct sockaddr { sa_family_t sa_family; /* address family, AF_xxx */ - union { - char sa_data_min[14]; /* Minimum 14 bytes of protocol address */ - DECLARE_FLEX_ARRAY(char, sa_data); - }; + char sa_data[14]; /* 14 bytes of protocol address */ }; struct linger { From dfb073d32cac28cdb597d16d1e0589b25c841661 Mon Sep 17 00:00:00 2001 From: Tim Hostetler Date: Tue, 4 Nov 2025 14:59:15 -0800 Subject: [PATCH 0540/1381] ptp: Return -EINVAL on ptp_clock_register if required ops are NULL ptp_clock should never be registered unless it stubs one of gettimex64() or gettime64() and settime64(). WARN_ON_ONCE and error out if either set of function pointers is null. For consistency, n_alarm validation is also folded into the WARN_ON_ONCE. Suggested-by: Kuniyuki Iwashima Reviewed-by: Kuniyuki Iwashima Reviewed-by: Harshitha Ramamurthy Reviewed-by: Vadim Fedorenko Signed-off-by: Tim Hostetler Acked-by: Richard Cochran Link: https://patch.msgid.link/20251104225915.2040080-1-thostet@google.com Signed-off-by: Jakub Kicinski --- drivers/ptp/ptp_clock.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/ptp/ptp_clock.c b/drivers/ptp/ptp_clock.c index ef020599b771..b0e167c0b3eb 100644 --- a/drivers/ptp/ptp_clock.c +++ b/drivers/ptp/ptp_clock.c @@ -322,7 +322,9 @@ struct ptp_clock *ptp_clock_register(struct ptp_clock_info *info, char debugfsname[16]; size_t size; - if (info->n_alarm > PTP_MAX_ALARMS) + if (WARN_ON_ONCE(info->n_alarm > PTP_MAX_ALARMS || + (!info->gettimex64 && !info->gettime64) || + !info->settime64)) return ERR_PTR(-EINVAL); /* Initialize a clock structure. */ From c79a022524577e486220bc9627ccebc706148c1f Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Fri, 31 Oct 2025 16:05:01 +0300 Subject: [PATCH 0541/1381] net: dsa: microchip: Fix a link check in ksz9477_pcs_read() The BMSR_LSTATUS define is 0x4 but the "p->phydev.link" variable is a 1 bit bitfield in a u32. Since 4 doesn't fit in 0-1 range it means that ".link" is always set to false. Add a !! to fix this. [Jakub: According to Maxime the phydev struct isn't really used and we should consider removing it completely. So not treating this as a fix.] Signed-off-by: Dan Carpenter Link: https://patch.msgid.link/aQSz_euUg0Ja8ZaH@stanley.mountain Signed-off-by: Jakub Kicinski --- drivers/net/dsa/microchip/ksz9477.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/dsa/microchip/ksz9477.c b/drivers/net/dsa/microchip/ksz9477.c index d747ea1c41a7..cf67d6377719 100644 --- a/drivers/net/dsa/microchip/ksz9477.c +++ b/drivers/net/dsa/microchip/ksz9477.c @@ -244,7 +244,7 @@ static int ksz9477_pcs_read(struct mii_bus *bus, int phy, int mmd, int reg) p->phydev.link = 0; } } else if (reg == MII_BMSR) { - p->phydev.link = (val & BMSR_LSTATUS); + p->phydev.link = !!(val & BMSR_LSTATUS); } } From 93d46ea3e984323fae0e5d2919cf5817e1297d41 Mon Sep 17 00:00:00 2001 From: Rohan G Thomas Date: Sat, 1 Nov 2025 01:27:07 +0800 Subject: [PATCH 0542/1381] net: stmmac: socfpga: Agilex5 EMAC platform configuration Agilex5 HPS EMAC uses the dwxgmac-3.10a IP, unlike previous socfpga platforms which use dwmac1000 IP. Due to differences in platform configuration, Agilex5 requires a distinct setup. Introduce a setup_plat_dat() callback in socfpga_dwmac_ops to handle platform-specific setup. This callback is invoked before stmmac_dvr_probe() to ensure the platform data is correctly configured. Also, implemented separate setup_plat_dat() callback for current socfpga platforms and Agilex5. Signed-off-by: Rohan G Thomas Reviewed-by: Maxime Chevallier Tested-by: Maxime Chevallier Link: https://patch.msgid.link/20251101-agilex5_ext-v2-1-a6b51b4dca4d@altera.com Signed-off-by: Jakub Kicinski --- .../ethernet/stmicro/stmmac/dwmac-socfpga.c | 30 +++++++++++++++++-- 1 file changed, 27 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c index 2ff5db6d41ca..5666b0172364 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c @@ -44,6 +44,7 @@ struct socfpga_dwmac; struct socfpga_dwmac_ops { int (*set_phy_mode)(struct socfpga_dwmac *dwmac_priv); + void (*setup_plat_dat)(struct socfpga_dwmac *dwmac_priv); }; struct socfpga_dwmac { @@ -441,6 +442,23 @@ static int socfpga_dwmac_init(struct platform_device *pdev, void *bsp_priv) return dwmac->ops->set_phy_mode(dwmac); } +static void socfpga_gen5_setup_plat_dat(struct socfpga_dwmac *dwmac) +{ + struct plat_stmmacenet_data *plat_dat = dwmac->plat_dat; + + plat_dat->core_type = DWMAC_CORE_GMAC; + + /* Rx watchdog timer in dwmac is buggy in this hw */ + plat_dat->riwt_off = 1; +} + +static void socfpga_agilex5_setup_plat_dat(struct socfpga_dwmac *dwmac) +{ + struct plat_stmmacenet_data *plat_dat = dwmac->plat_dat; + + plat_dat->core_type = DWMAC_CORE_XGMAC; +} + static int socfpga_dwmac_probe(struct platform_device *pdev) { struct plat_stmmacenet_data *plat_dat; @@ -497,25 +515,31 @@ static int socfpga_dwmac_probe(struct platform_device *pdev) plat_dat->pcs_init = socfpga_dwmac_pcs_init; plat_dat->pcs_exit = socfpga_dwmac_pcs_exit; plat_dat->select_pcs = socfpga_dwmac_select_pcs; - plat_dat->core_type = DWMAC_CORE_GMAC; - plat_dat->riwt_off = 1; + ops->setup_plat_dat(dwmac); return devm_stmmac_pltfr_probe(pdev, plat_dat, &stmmac_res); } static const struct socfpga_dwmac_ops socfpga_gen5_ops = { .set_phy_mode = socfpga_gen5_set_phy_mode, + .setup_plat_dat = socfpga_gen5_setup_plat_dat, }; static const struct socfpga_dwmac_ops socfpga_gen10_ops = { .set_phy_mode = socfpga_gen10_set_phy_mode, + .setup_plat_dat = socfpga_gen5_setup_plat_dat, +}; + +static const struct socfpga_dwmac_ops socfpga_agilex5_ops = { + .set_phy_mode = socfpga_gen10_set_phy_mode, + .setup_plat_dat = socfpga_agilex5_setup_plat_dat, }; static const struct of_device_id socfpga_dwmac_match[] = { { .compatible = "altr,socfpga-stmmac", .data = &socfpga_gen5_ops }, { .compatible = "altr,socfpga-stmmac-a10-s10", .data = &socfpga_gen10_ops }, - { .compatible = "altr,socfpga-stmmac-agilex5", .data = &socfpga_gen10_ops }, + { .compatible = "altr,socfpga-stmmac-agilex5", .data = &socfpga_agilex5_ops }, { } }; MODULE_DEVICE_TABLE(of, socfpga_dwmac_match); From 4c00476d44804db3c16838299b87a11741cd0dbd Mon Sep 17 00:00:00 2001 From: Rohan G Thomas Date: Sat, 1 Nov 2025 01:27:08 +0800 Subject: [PATCH 0543/1381] net: stmmac: socfpga: Enable TBS support for Agilex5 Agilex5 supports Time-Based Scheduling(TBS) for Tx queue 6 and Tx queue 7. This commit enables TBS support for these queues. Signed-off-by: Rohan G Thomas Link: https://patch.msgid.link/20251101-agilex5_ext-v2-2-a6b51b4dca4d@altera.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c | 13 +++++++++++++ 1 file changed, 13 insertions(+) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c index 5666b0172364..4f256f0ae05c 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c @@ -457,6 +457,19 @@ static void socfpga_agilex5_setup_plat_dat(struct socfpga_dwmac *dwmac) struct plat_stmmacenet_data *plat_dat = dwmac->plat_dat; plat_dat->core_type = DWMAC_CORE_XGMAC; + + /* Enable TBS */ + switch (plat_dat->tx_queues_to_use) { + case 8: + plat_dat->tx_queues_cfg[7].tbs_en = true; + fallthrough; + case 7: + plat_dat->tx_queues_cfg[6].tbs_en = true; + break; + default: + /* Tx Queues 0 - 5 doesn't support TBS on Agilex5 */ + break; + } } static int socfpga_dwmac_probe(struct platform_device *pdev) From e28988aef70f8f993d93a62161a202e930cfce55 Mon Sep 17 00:00:00 2001 From: Rohan G Thomas Date: Sat, 1 Nov 2025 01:27:09 +0800 Subject: [PATCH 0544/1381] net: stmmac: socfpga: Enable TSO for Agilex5 platform Agilex5 supports TCP Segmentation Offload(TSO). This commit enables TSO for Agilex5 socfpga platforms. Signed-off-by: Rohan G Thomas Link: https://patch.msgid.link/20251101-agilex5_ext-v2-3-a6b51b4dca4d@altera.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c index 4f256f0ae05c..1837346ca2d4 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c @@ -458,6 +458,9 @@ static void socfpga_agilex5_setup_plat_dat(struct socfpga_dwmac *dwmac) plat_dat->core_type = DWMAC_CORE_XGMAC; + /* Enable TSO */ + plat_dat->flags |= STMMAC_FLAG_TSO_EN; + /* Enable TBS */ switch (plat_dat->tx_queues_to_use) { case 8: From fd8c4f6454963aa7ea895657472aa57f33779d57 Mon Sep 17 00:00:00 2001 From: Rohan G Thomas Date: Sat, 1 Nov 2025 01:27:10 +0800 Subject: [PATCH 0545/1381] net: stmmac: socfpga: Add hardware supported cross-timestamp Cross timestamping is supported on Agilex5 platform with Synchronized Multidrop Timestamp Gathering(SMTG) IP. The hardware cross-timestamp result is made available the applications through the ioctl call PTP_SYS_OFFSET_PRECISE, which inturn calls stmmac_getcrosststamp(). Device time is stored in the MAC Auxiliary register. The 64-bit System time (ARM_ARCH_COUNTER) is stored in SMTG IP. SMTG IP is an MDIO device with 0xC - 0xF MDIO register space holds 64-bit system time. This commit is similar to following commit for Intel platforms: Commit 341f67e424e5 ("net: stmmac: Add hardware supported cross-timestamp") Signed-off-by: Rohan G Thomas Link: https://patch.msgid.link/20251101-agilex5_ext-v2-4-a6b51b4dca4d@altera.com Signed-off-by: Jakub Kicinski --- .../ethernet/stmicro/stmmac/dwmac-socfpga.c | 120 ++++++++++++++++++ .../net/ethernet/stmicro/stmmac/dwxgmac2.h | 5 + 2 files changed, 125 insertions(+) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c index 1837346ca2d4..49d651948e2b 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c @@ -5,6 +5,7 @@ */ #include +#include #include #include #include @@ -15,8 +16,10 @@ #include #include +#include "dwxgmac2.h" #include "stmmac.h" #include "stmmac_platform.h" +#include "stmmac_ptp.h" #define SYSMGR_EMACGRP_CTRL_PHYSEL_ENUM_GMII_MII 0x0 #define SYSMGR_EMACGRP_CTRL_PHYSEL_ENUM_RGMII 0x1 @@ -41,6 +44,13 @@ #define SGMII_ADAPTER_ENABLE 0x0000 #define SGMII_ADAPTER_DISABLE 0x0001 +#define SMTG_MDIO_ADDR 0x15 +#define SMTG_TSC_WORD0 0xC +#define SMTG_TSC_WORD1 0xD +#define SMTG_TSC_WORD2 0xE +#define SMTG_TSC_WORD3 0xF +#define SMTG_TSC_SHIFT 16 + struct socfpga_dwmac; struct socfpga_dwmac_ops { int (*set_phy_mode)(struct socfpga_dwmac *dwmac_priv); @@ -269,6 +279,112 @@ static int socfpga_set_phy_mode_common(int phymode, u32 *val) return 0; } +static void get_smtgtime(struct mii_bus *mii, int smtg_addr, u64 *smtg_time) +{ + u64 ns; + + ns = mdiobus_read(mii, smtg_addr, SMTG_TSC_WORD3); + ns <<= SMTG_TSC_SHIFT; + ns |= mdiobus_read(mii, smtg_addr, SMTG_TSC_WORD2); + ns <<= SMTG_TSC_SHIFT; + ns |= mdiobus_read(mii, smtg_addr, SMTG_TSC_WORD1); + ns <<= SMTG_TSC_SHIFT; + ns |= mdiobus_read(mii, smtg_addr, SMTG_TSC_WORD0); + + *smtg_time = ns; +} + +static int smtg_crosststamp(ktime_t *device, struct system_counterval_t *system, + void *ctx) +{ + struct stmmac_priv *priv = (struct stmmac_priv *)ctx; + u32 num_snapshot, gpio_value, acr_value; + void __iomem *ptpaddr = priv->ptpaddr; + void __iomem *ioaddr = priv->hw->pcsr; + unsigned long flags; + u64 smtg_time = 0; + u64 ptp_time = 0; + int i, ret; + u32 v; + + /* Both internal crosstimestamping and external triggered event + * timestamping cannot be run concurrently. + */ + if (priv->plat->flags & STMMAC_FLAG_EXT_SNAPSHOT_EN) + return -EBUSY; + + mutex_lock(&priv->aux_ts_lock); + /* Enable Internal snapshot trigger */ + acr_value = readl(ptpaddr + PTP_ACR); + acr_value &= ~PTP_ACR_MASK; + switch (priv->plat->int_snapshot_num) { + case AUX_SNAPSHOT0: + acr_value |= PTP_ACR_ATSEN0; + break; + case AUX_SNAPSHOT1: + acr_value |= PTP_ACR_ATSEN1; + break; + case AUX_SNAPSHOT2: + acr_value |= PTP_ACR_ATSEN2; + break; + case AUX_SNAPSHOT3: + acr_value |= PTP_ACR_ATSEN3; + break; + default: + mutex_unlock(&priv->aux_ts_lock); + return -EINVAL; + } + writel(acr_value, ptpaddr + PTP_ACR); + + /* Clear FIFO */ + acr_value = readl(ptpaddr + PTP_ACR); + acr_value |= PTP_ACR_ATSFC; + writel(acr_value, ptpaddr + PTP_ACR); + /* Release the mutex */ + mutex_unlock(&priv->aux_ts_lock); + + /* Trigger Internal snapshot signal. Create a rising edge by just toggle + * the GPO0 to low and back to high. + */ + gpio_value = readl(ioaddr + XGMAC_GPIO_STATUS); + gpio_value &= ~XGMAC_GPIO_GPO0; + writel(gpio_value, ioaddr + XGMAC_GPIO_STATUS); + gpio_value |= XGMAC_GPIO_GPO0; + writel(gpio_value, ioaddr + XGMAC_GPIO_STATUS); + + /* Poll for time sync operation done */ + ret = readl_poll_timeout(priv->ioaddr + XGMAC_INT_STATUS, v, + (v & XGMAC_INT_TSIS), 100, 10000); + if (ret) { + netdev_err(priv->dev, "%s: Wait for time sync operation timeout\n", + __func__); + return ret; + } + + *system = (struct system_counterval_t) { + .cycles = 0, + .cs_id = CSID_ARM_ARCH_COUNTER, + .use_nsecs = false, + }; + + num_snapshot = (readl(ioaddr + XGMAC_TIMESTAMP_STATUS) & + XGMAC_TIMESTAMP_ATSNS_MASK) >> + XGMAC_TIMESTAMP_ATSNS_SHIFT; + + /* Repeat until the timestamps are from the FIFO last segment */ + for (i = 0; i < num_snapshot; i++) { + read_lock_irqsave(&priv->ptp_lock, flags); + stmmac_get_ptptime(priv, ptpaddr, &ptp_time); + *device = ns_to_ktime(ptp_time); + read_unlock_irqrestore(&priv->ptp_lock, flags); + } + + get_smtgtime(priv->mii, SMTG_MDIO_ADDR, &smtg_time); + system->cycles = smtg_time; + + return 0; +} + static int socfpga_gen5_set_phy_mode(struct socfpga_dwmac *dwmac) { struct regmap *sys_mgr_base_addr = dwmac->sys_mgr_base_addr; @@ -473,6 +589,10 @@ static void socfpga_agilex5_setup_plat_dat(struct socfpga_dwmac *dwmac) /* Tx Queues 0 - 5 doesn't support TBS on Agilex5 */ break; } + + /* Hw supported cross-timestamp */ + plat_dat->int_snapshot_num = AUX_SNAPSHOT0; + plat_dat->crosststamp = smtg_crosststamp; } static int socfpga_dwmac_probe(struct platform_device *pdev) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h index 0d408ee17f33..e48cfa05000c 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h +++ b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h @@ -79,6 +79,7 @@ #define XGMAC_PSRQ(x) GENMASK((x) * 8 + 7, (x) * 8) #define XGMAC_PSRQ_SHIFT(x) ((x) * 8) #define XGMAC_INT_STATUS 0x000000b0 +#define XGMAC_INT_TSIS BIT(12) #define XGMAC_LPIIS BIT(5) #define XGMAC_PMTIS BIT(4) #define XGMAC_INT_EN 0x000000b4 @@ -173,6 +174,8 @@ #define XGMAC_MDIO_ADDR 0x00000200 #define XGMAC_MDIO_DATA 0x00000204 #define XGMAC_MDIO_C22P 0x00000220 +#define XGMAC_GPIO_STATUS 0x0000027c +#define XGMAC_GPIO_GPO0 BIT(16) #define XGMAC_ADDRx_HIGH(x) (0x00000300 + (x) * 0x8) #define XGMAC_ADDR_MAX 32 #define XGMAC_AE BIT(31) @@ -220,6 +223,8 @@ #define XGMAC_OB BIT(0) #define XGMAC_RSS_DATA 0x00000c8c #define XGMAC_TIMESTAMP_STATUS 0x00000d20 +#define XGMAC_TIMESTAMP_ATSNS_MASK GENMASK(29, 25) +#define XGMAC_TIMESTAMP_ATSNS_SHIFT 25 #define XGMAC_TXTSC BIT(15) #define XGMAC_TXTIMESTAMP_NSEC 0x00000d30 #define XGMAC_TXTSSTSLO GENMASK(30, 0) From 0567c84d683d1f38dc41928eec786ec5c02bf7b4 Mon Sep 17 00:00:00 2001 From: Shangjuan Wei Date: Tue, 4 Nov 2025 15:33:05 +0800 Subject: [PATCH 0546/1381] dt-bindings: ethernet: eswin: fix yaml schema issues eswin,hsp-sp-csr attribute is one phandle with multiple arguments, so the syntax should be in the form of: items: - items: - description: ... - description: ... - description: ... - description: ... To align with the description of the 'eswin-sp-csr' attribute in the mmc,usb modules, the description of the 'eswin,hsp-sp-csr' attribute has been modified. Fixes: 888bd0eca93c ("dt-bindings: ethernet: eswin: Document for EIC7700 SoC") Reported-by: Rob Herring (Arm) Closes: https://lore.kernel.org/all/176096011380.22917.1988679321096076522.robh@kernel.org/ Signed-off-by: Shangjuan Wei Reviewed-by: Krzysztof Kozlowski Link: https://patch.msgid.link/20251104073305.299-1-weishangjuan@eswincomputing.com Signed-off-by: Jakub Kicinski --- .../bindings/net/eswin,eic7700-eth.yaml | 20 ++++++++++--------- 1 file changed, 11 insertions(+), 9 deletions(-) diff --git a/Documentation/devicetree/bindings/net/eswin,eic7700-eth.yaml b/Documentation/devicetree/bindings/net/eswin,eic7700-eth.yaml index 9ddbfe219ae2..91e8cd1db67b 100644 --- a/Documentation/devicetree/bindings/net/eswin,eic7700-eth.yaml +++ b/Documentation/devicetree/bindings/net/eswin,eic7700-eth.yaml @@ -69,17 +69,19 @@ properties: enum: [0, 200, 600, 1200, 1600, 1800, 2000, 2200, 2400] eswin,hsp-sp-csr: + description: + HSP CSR is to control and get status of different high-speed peripherals + (such as Ethernet, USB, SATA, etc.) via register, which can tune + board-level's parameters of PHY, etc. $ref: /schemas/types.yaml#/definitions/phandle-array items: - - description: Phandle to HSP(High-Speed Peripheral) device - - description: Offset of phy control register for internal - or external clock selection - - description: Offset of AXI clock controller Low-Power request - register - - description: Offset of register controlling TX/RX clock delay - description: | - High-Speed Peripheral device needed to configure clock selection, - clock low-power mode and clock delay. + - items: + - description: Phandle to HSP(High-Speed Peripheral) device + - description: Offset of phy control register for internal + or external clock selection + - description: Offset of AXI clock controller Low-Power request + register + - description: Offset of register controlling TX/RX clock delay required: - compatible From 28a45575289f3292aa9cb7bacae18ba3ee7a6adf Mon Sep 17 00:00:00 2001 From: Fedor Pchelkin Date: Tue, 4 Nov 2025 16:57:08 +0300 Subject: [PATCH 0547/1381] wifi: rtw89: usb: use common error path for skbs in rtw89_usb_rx_handler() Allow adding rx_skb to rx_free_queue for later reuse on the common error handling path, otherwise free it. Found by Linux Verification Center (linuxtesting.org). Fixes: 2135c28be6a8 ("wifi: rtw89: Add usb.{c,h}") Acked-by: Ping-Ke Shih Signed-off-by: Fedor Pchelkin Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251104135720.321110-2-pchelkin@ispras.ru --- drivers/net/wireless/realtek/rtw89/usb.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/net/wireless/realtek/rtw89/usb.c b/drivers/net/wireless/realtek/rtw89/usb.c index 4c47071b30fd..450631c77840 100644 --- a/drivers/net/wireless/realtek/rtw89/usb.c +++ b/drivers/net/wireless/realtek/rtw89/usb.c @@ -390,8 +390,7 @@ static void rtw89_usb_rx_handler(struct work_struct *work) if (skb_queue_len(&rtwusb->rx_queue) >= RTW89_USB_MAX_RXQ_LEN) { rtw89_warn(rtwdev, "rx_queue overflow\n"); - dev_kfree_skb_any(rx_skb); - continue; + goto free_or_reuse; } memset(&desc_info, 0, sizeof(desc_info)); @@ -402,7 +401,7 @@ static void rtw89_usb_rx_handler(struct work_struct *work) rtw89_debug(rtwdev, RTW89_DBG_HCI, "failed to allocate RX skb of size %u\n", desc_info.pkt_size); - continue; + goto free_or_reuse; } pkt_offset = desc_info.offset + desc_info.rxd_len; @@ -412,6 +411,7 @@ static void rtw89_usb_rx_handler(struct work_struct *work) rtw89_core_rx(rtwdev, &desc_info, skb); +free_or_reuse: if (skb_queue_len(&rtwusb->rx_free_queue) >= RTW89_USB_RX_SKB_NUM) dev_kfree_skb_any(rx_skb); else From 7543818e97d5d54b3b2f75f1c4dedee298d7d914 Mon Sep 17 00:00:00 2001 From: Fedor Pchelkin Date: Tue, 4 Nov 2025 16:57:09 +0300 Subject: [PATCH 0548/1381] wifi: rtw89: usb: fix leak in rtw89_usb_write_port() When there is an attempt to write data and RTW89_FLAG_UNPLUGGED is set, this means device is disconnected and no urb is submitted. Return appropriate error code to the caller to properly free the allocated resources. Found by Linux Verification Center (linuxtesting.org). Fixes: 2135c28be6a8 ("wifi: rtw89: Add usb.{c,h}") Acked-by: Ping-Ke Shih Signed-off-by: Fedor Pchelkin Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251104135720.321110-3-pchelkin@ispras.ru --- drivers/net/wireless/realtek/rtw89/usb.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/net/wireless/realtek/rtw89/usb.c b/drivers/net/wireless/realtek/rtw89/usb.c index 450631c77840..639b84ebe942 100644 --- a/drivers/net/wireless/realtek/rtw89/usb.c +++ b/drivers/net/wireless/realtek/rtw89/usb.c @@ -236,7 +236,7 @@ static int rtw89_usb_write_port(struct rtw89_dev *rtwdev, u8 ch_dma, int ret; if (test_bit(RTW89_FLAG_UNPLUGGED, rtwdev->flags)) - return 0; + return -ENODEV; urb = usb_alloc_urb(0, GFP_ATOMIC); if (!urb) @@ -285,8 +285,9 @@ static void rtw89_usb_ops_tx_kick_off(struct rtw89_dev *rtwdev, u8 txch) ret = rtw89_usb_write_port(rtwdev, txch, skb->data, skb->len, txcb); if (ret) { - rtw89_err(rtwdev, "write port txch %d failed: %d\n", - txch, ret); + if (ret != -ENODEV) + rtw89_err(rtwdev, "write port txch %d failed: %d\n", + txch, ret); skb_dequeue(&txcb->tx_ack_queue); kfree(txcb); From 45a6a88f011fc57ba8518f4d032cc2b4f6543c12 Mon Sep 17 00:00:00 2001 From: Fedor Pchelkin Date: Tue, 4 Nov 2025 16:57:10 +0300 Subject: [PATCH 0549/1381] wifi: rtw89: usb: use ieee80211_free_txskb() where appropriate rtw89_usb_ops_tx_kick_off() may need to release skb if a failure occurs. It operates mainly on skbs coming from the core wireless stack and the ones containing firmware commands. Use ieee80211_free_txskb() for the former case. Suggested-by: Ping-Ke Shih Acked-by: Ping-Ke Shih Signed-off-by: Fedor Pchelkin Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251104135720.321110-4-pchelkin@ispras.ru --- drivers/net/wireless/realtek/rtw89/usb.c | 13 +++++++++++-- 1 file changed, 11 insertions(+), 2 deletions(-) diff --git a/drivers/net/wireless/realtek/rtw89/usb.c b/drivers/net/wireless/realtek/rtw89/usb.c index 639b84ebe942..59cdee54feef 100644 --- a/drivers/net/wireless/realtek/rtw89/usb.c +++ b/drivers/net/wireless/realtek/rtw89/usb.c @@ -258,6 +258,15 @@ static int rtw89_usb_write_port(struct rtw89_dev *rtwdev, u8 ch_dma, return ret; } +static void rtw89_usb_tx_free_skb(struct rtw89_dev *rtwdev, u8 txch, + struct sk_buff *skb) +{ + if (txch == RTW89_TXCH_CH12) + dev_kfree_skb_any(skb); + else + ieee80211_free_txskb(rtwdev->hw, skb); +} + static void rtw89_usb_ops_tx_kick_off(struct rtw89_dev *rtwdev, u8 txch) { struct rtw89_usb *rtwusb = rtw89_usb_priv(rtwdev); @@ -272,7 +281,7 @@ static void rtw89_usb_ops_tx_kick_off(struct rtw89_dev *rtwdev, u8 txch) txcb = kmalloc(sizeof(*txcb), GFP_ATOMIC); if (!txcb) { - dev_kfree_skb_any(skb); + rtw89_usb_tx_free_skb(rtwdev, txch, skb); continue; } @@ -291,7 +300,7 @@ static void rtw89_usb_ops_tx_kick_off(struct rtw89_dev *rtwdev, u8 txch) skb_dequeue(&txcb->tx_ack_queue); kfree(txcb); - dev_kfree_skb_any(skb); + rtw89_usb_tx_free_skb(rtwdev, txch, skb); } } } From 8986bafb0945821b7fc160780074d39eae5f7247 Mon Sep 17 00:00:00 2001 From: Fedor Pchelkin Date: Tue, 4 Nov 2025 16:57:11 +0300 Subject: [PATCH 0550/1381] wifi: rtw89: refine rtw89_core_tx_wait_complete() Pass TX status value directly into rtw89_core_tx_wait_complete(). This will make it a bit in sync with further patches and will give flexibility in future work. Also use scope based RCU locking which simplifies the code of the function. Found by Linux Verification Center (linuxtesting.org). Acked-by: Ping-Ke Shih Signed-off-by: Fedor Pchelkin Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251104135720.321110-5-pchelkin@ispras.ru --- drivers/net/wireless/realtek/rtw89/core.h | 20 ++++++++++---------- drivers/net/wireless/realtek/rtw89/pci.c | 2 +- drivers/net/wireless/realtek/rtw89/pci.h | 4 ---- 3 files changed, 11 insertions(+), 15 deletions(-) diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h index 97cb0e5f0898..d23ec4d4095a 100644 --- a/drivers/net/wireless/realtek/rtw89/core.h +++ b/drivers/net/wireless/realtek/rtw89/core.h @@ -3509,6 +3509,11 @@ struct rtw89_phy_rate_pattern { bool enable; }; +#define RTW89_TX_DONE 0x0 +#define RTW89_TX_RETRY_LIMIT 0x1 +#define RTW89_TX_LIFE_TIME 0x2 +#define RTW89_TX_MACID_DROP 0x3 + #define RTW89_TX_WAIT_WORK_TIMEOUT msecs_to_jiffies(500) struct rtw89_tx_wait_info { struct rcu_head rcu_head; @@ -7388,25 +7393,20 @@ static inline struct sk_buff *rtw89_alloc_skb_for_rx(struct rtw89_dev *rtwdev, static inline bool rtw89_core_tx_wait_complete(struct rtw89_dev *rtwdev, struct rtw89_tx_skb_data *skb_data, - bool tx_done) + u8 tx_status) { struct rtw89_tx_wait_info *wait; - bool ret = false; - rcu_read_lock(); + guard(rcu)(); wait = rcu_dereference(skb_data->wait); if (!wait) - goto out; + return false; - ret = true; - wait->tx_done = tx_done; + wait->tx_done = tx_status == RTW89_TX_DONE; /* Don't access skb anymore after completion */ complete_all(&wait->completion); - -out: - rcu_read_unlock(); - return ret; + return true; } static inline bool rtw89_is_mlo_1_1(struct rtw89_dev *rtwdev) diff --git a/drivers/net/wireless/realtek/rtw89/pci.c b/drivers/net/wireless/realtek/rtw89/pci.c index 0ee5f8579447..b1985193a18f 100644 --- a/drivers/net/wireless/realtek/rtw89/pci.c +++ b/drivers/net/wireless/realtek/rtw89/pci.c @@ -464,7 +464,7 @@ static void rtw89_pci_tx_status(struct rtw89_dev *rtwdev, struct rtw89_tx_skb_data *skb_data = RTW89_TX_SKB_CB(skb); struct ieee80211_tx_info *info; - if (rtw89_core_tx_wait_complete(rtwdev, skb_data, tx_status == RTW89_TX_DONE)) + if (rtw89_core_tx_wait_complete(rtwdev, skb_data, tx_status)) return; info = IEEE80211_SKB_CB(skb); diff --git a/drivers/net/wireless/realtek/rtw89/pci.h b/drivers/net/wireless/realtek/rtw89/pci.h index cb05c83dfd56..16dfb0e79d77 100644 --- a/drivers/net/wireless/realtek/rtw89/pci.h +++ b/drivers/net/wireless/realtek/rtw89/pci.h @@ -1487,10 +1487,6 @@ struct rtw89_pci_tx_addr_info_32_v1 { #define RTW89_PCI_RPP_POLLUTED BIT(31) #define RTW89_PCI_RPP_SEQ GENMASK(30, 16) #define RTW89_PCI_RPP_TX_STATUS GENMASK(15, 13) -#define RTW89_TX_DONE 0x0 -#define RTW89_TX_RETRY_LIMIT 0x1 -#define RTW89_TX_LIFE_TIME 0x2 -#define RTW89_TX_MACID_DROP 0x3 #define RTW89_PCI_RPP_QSEL GENMASK(12, 8) #define RTW89_PCI_RPP_MACID GENMASK(7, 0) From c33c6a1b6f72ae48570530a271c68131c17abdbb Mon Sep 17 00:00:00 2001 From: Fedor Pchelkin Date: Tue, 4 Nov 2025 16:57:12 +0300 Subject: [PATCH 0551/1381] wifi: rtw89: implement C2H TX report handler rtw89 has several ways of handling TX status report events. The first one is based on RPP feature which is used by PCIe HCI. The other one depends on firmware sending a corresponding C2H message, quite similar to what rtw88 has. Toggle a bit in the TX descriptor to indicate to the firmware that TX report for the frame is expected. This will allow handling TX wait skbs and the ones flagged with IEEE80211_TX_CTL_REQ_TX_STATUS correctly. Do the bulk of the patch according to the vendor driver for RTL8851BU. However, there are slight differences in C2H message format between different types of chips. RTL885xB ones follow format V0. RTL8852C has format V1, and RTL8922AU has format V2. Found by Linux Verification Center (linuxtesting.org). Suggested-by: Bitterblue Smith Acked-by: Ping-Ke Shih Signed-off-by: Fedor Pchelkin Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251104135720.321110-6-pchelkin@ispras.ru --- drivers/net/wireless/realtek/rtw89/core.c | 28 +++++++++++--- drivers/net/wireless/realtek/rtw89/core.h | 4 ++ drivers/net/wireless/realtek/rtw89/fw.h | 41 ++++++++++++++++++++ drivers/net/wireless/realtek/rtw89/mac.c | 46 +++++++++++++++++++++++ drivers/net/wireless/realtek/rtw89/mac.h | 7 ++++ drivers/net/wireless/realtek/rtw89/txrx.h | 6 ++- 6 files changed, 126 insertions(+), 6 deletions(-) diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c index 1f92be300ea7..84f94c796a48 100644 --- a/drivers/net/wireless/realtek/rtw89/core.c +++ b/drivers/net/wireless/realtek/rtw89/core.c @@ -1415,7 +1415,10 @@ static __le32 rtw89_build_txwd_info1(struct rtw89_tx_desc_info *desc_info) u32 dword = FIELD_PREP(RTW89_TXWD_INFO1_MAX_AGGNUM, desc_info->ampdu_num) | FIELD_PREP(RTW89_TXWD_INFO1_A_CTRL_BSR, desc_info->a_ctrl_bsr) | FIELD_PREP(RTW89_TXWD_INFO1_DATA_RTY_LOWEST_RATE, - desc_info->data_retry_lowest_rate); + desc_info->data_retry_lowest_rate) | + FIELD_PREP(RTW89_TXWD_INFO1_DATA_TXCNT_LMT_SEL, + desc_info->tx_cnt_lmt_en) | + FIELD_PREP(RTW89_TXWD_INFO1_DATA_TXCNT_LMT, desc_info->tx_cnt_lmt); return cpu_to_le32(dword); } @@ -1439,11 +1442,19 @@ static __le32 rtw89_build_txwd_info2_v1(struct rtw89_tx_desc_info *desc_info) return cpu_to_le32(dword); } +static __le32 rtw89_build_txwd_info3(struct rtw89_tx_desc_info *desc_info) +{ + u32 dword = FIELD_PREP(RTW89_TXWD_INFO3_SPE_RPT, desc_info->report); + + return cpu_to_le32(dword); +} + static __le32 rtw89_build_txwd_info4(struct rtw89_tx_desc_info *desc_info) { bool rts_en = !desc_info->is_bmc; u32 dword = FIELD_PREP(RTW89_TXWD_INFO4_RTS_EN, rts_en) | - FIELD_PREP(RTW89_TXWD_INFO4_HW_RTS_EN, 1); + FIELD_PREP(RTW89_TXWD_INFO4_HW_RTS_EN, 1) | + FIELD_PREP(RTW89_TXWD_INFO4_SW_DEFINE, desc_info->sn); return cpu_to_le32(dword); } @@ -1466,6 +1477,7 @@ void rtw89_core_fill_txdesc(struct rtw89_dev *rtwdev, txwd_info->dword0 = rtw89_build_txwd_info0(desc_info); txwd_info->dword1 = rtw89_build_txwd_info1(desc_info); txwd_info->dword2 = rtw89_build_txwd_info2(desc_info); + txwd_info->dword3 = rtw89_build_txwd_info3(desc_info); txwd_info->dword4 = rtw89_build_txwd_info4(desc_info); } @@ -1495,6 +1507,7 @@ void rtw89_core_fill_txdesc_v1(struct rtw89_dev *rtwdev, txwd_info->dword0 = rtw89_build_txwd_info0_v1(desc_info); txwd_info->dword1 = rtw89_build_txwd_info1(desc_info); txwd_info->dword2 = rtw89_build_txwd_info2_v1(desc_info); + txwd_info->dword3 = rtw89_build_txwd_info3(desc_info); txwd_info->dword4 = rtw89_build_txwd_info4(desc_info); } EXPORT_SYMBOL(rtw89_core_fill_txdesc_v1); @@ -1580,7 +1593,10 @@ static __le32 rtw89_build_txwd_info0_v2(struct rtw89_tx_desc_info *desc_info) u32 dword = FIELD_PREP(BE_TXD_INFO0_DATA_STBC, desc_info->stbc) | FIELD_PREP(BE_TXD_INFO0_DATA_LDPC, desc_info->ldpc) | FIELD_PREP(BE_TXD_INFO0_DISDATAFB, desc_info->dis_data_fb) | - FIELD_PREP(BE_TXD_INFO0_MULTIPORT_ID, desc_info->port); + FIELD_PREP(BE_TXD_INFO0_MULTIPORT_ID, desc_info->port) | + FIELD_PREP(BE_TXD_INFO0_DATA_TXCNT_LMT_SEL, + desc_info->tx_cnt_lmt_en) | + FIELD_PREP(BE_TXD_INFO0_DATA_TXCNT_LMT, desc_info->tx_cnt_lmt); return cpu_to_le32(dword); } @@ -1590,7 +1606,8 @@ static __le32 rtw89_build_txwd_info1_v2(struct rtw89_tx_desc_info *desc_info) u32 dword = FIELD_PREP(BE_TXD_INFO1_MAX_AGG_NUM, desc_info->ampdu_num) | FIELD_PREP(BE_TXD_INFO1_A_CTRL_BSR, desc_info->a_ctrl_bsr) | FIELD_PREP(BE_TXD_INFO1_DATA_RTY_LOWEST_RATE, - desc_info->data_retry_lowest_rate); + desc_info->data_retry_lowest_rate) | + FIELD_PREP(BE_TXD_INFO1_SW_DEFINE, desc_info->sn); return cpu_to_le32(dword); } @@ -1599,7 +1616,8 @@ static __le32 rtw89_build_txwd_info2_v2(struct rtw89_tx_desc_info *desc_info) { u32 dword = FIELD_PREP(BE_TXD_INFO2_AMPDU_DENSITY, desc_info->ampdu_density) | FIELD_PREP(BE_TXD_INFO2_FORCE_KEY_EN, desc_info->sec_en) | - FIELD_PREP(BE_TXD_INFO2_SEC_CAM_IDX, desc_info->sec_cam_idx); + FIELD_PREP(BE_TXD_INFO2_SEC_CAM_IDX, desc_info->sec_cam_idx) | + FIELD_PREP(BE_TXD_INFO2_SPE_RPT_V1, desc_info->report); return cpu_to_le32(dword); } diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h index d23ec4d4095a..2914cc4e97d5 100644 --- a/drivers/net/wireless/realtek/rtw89/core.h +++ b/drivers/net/wireless/realtek/rtw89/core.h @@ -1169,6 +1169,10 @@ struct rtw89_tx_desc_info { u8 ampdu_density; u8 ampdu_num; bool sec_en; + bool report; + bool tx_cnt_lmt_en; + u8 sn: 4; + u8 tx_cnt_lmt: 6; u8 addr_info_nr; u8 sec_keyid; u8 sec_type; diff --git a/drivers/net/wireless/realtek/rtw89/fw.h b/drivers/net/wireless/realtek/rtw89/fw.h index ddebf7972068..310580206368 100644 --- a/drivers/net/wireless/realtek/rtw89/fw.h +++ b/drivers/net/wireless/realtek/rtw89/fw.h @@ -3747,6 +3747,47 @@ struct rtw89_c2h_scanofld { #define RTW89_GET_MAC_C2H_MCC_REQ_ACK_H2C_FUNC(c2h) \ le32_get_bits(*((const __le32 *)(c2h) + 2), GENMASK(15, 8)) +struct rtw89_c2h_mac_tx_rpt { + struct rtw89_c2h_hdr hdr; + __le32 w2; + __le32 w3; + __le32 w4; + __le32 w5; + __le32 w6; + __le32 w7; +} __packed; + +#define RTW89_C2H_MAC_TX_RPT_W2_TX_STATE GENMASK(7, 6) +#define RTW89_C2H_MAC_TX_RPT_W2_SW_DEFINE GENMASK(11, 8) +#define RTW89_C2H_MAC_TX_RPT_W5_DATA_TX_CNT GENMASK(13, 8) +#define RTW89_C2H_MAC_TX_RPT_W5_DATA_TX_CNT_V1 GENMASK(15, 10) + +struct rtw89_c2h_mac_tx_rpt_v2 { + struct rtw89_c2h_hdr hdr; + __le32 w2; + __le32 w3; + __le32 w4; + __le32 w5; + __le32 w6; + __le32 w7; + __le32 w8; + __le32 w9; + __le32 w10; + __le32 w11; + __le32 w12; + __le32 w13; + __le32 w14; + __le32 w15; + __le32 w16; + __le32 w17; + __le32 w18; + __le32 w19; +} __packed; + +#define RTW89_C2H_MAC_TX_RPT_W12_TX_STATE_V2 GENMASK(9, 8) +#define RTW89_C2H_MAC_TX_RPT_W12_SW_DEFINE_V2 GENMASK(15, 12) +#define RTW89_C2H_MAC_TX_RPT_W14_DATA_TX_CNT_V2 GENMASK(15, 10) + struct rtw89_mac_mcc_tsf_rpt { u32 macid_x; u32 macid_y; diff --git a/drivers/net/wireless/realtek/rtw89/mac.c b/drivers/net/wireless/realtek/rtw89/mac.c index bc39b4f7bc4f..47048d125c01 100644 --- a/drivers/net/wireless/realtek/rtw89/mac.c +++ b/drivers/net/wireless/realtek/rtw89/mac.c @@ -5488,6 +5488,40 @@ rtw89_mac_c2h_mcc_status_rpt(struct rtw89_dev *rtwdev, struct sk_buff *c2h, u32 rtw89_complete_cond(&rtwdev->mcc.wait, cond, &data); } +static void +rtw89_mac_c2h_tx_rpt(struct rtw89_dev *rtwdev, struct sk_buff *c2h, u32 len) +{ + u8 sw_define, tx_status, txcnt; + + if (rtwdev->chip->chip_id == RTL8922A) { + const struct rtw89_c2h_mac_tx_rpt_v2 *rpt_v2; + + rpt_v2 = (const struct rtw89_c2h_mac_tx_rpt_v2 *)c2h->data; + sw_define = le32_get_bits(rpt_v2->w12, + RTW89_C2H_MAC_TX_RPT_W12_SW_DEFINE_V2); + tx_status = le32_get_bits(rpt_v2->w12, + RTW89_C2H_MAC_TX_RPT_W12_TX_STATE_V2); + txcnt = le32_get_bits(rpt_v2->w14, + RTW89_C2H_MAC_TX_RPT_W14_DATA_TX_CNT_V2); + } else { + const struct rtw89_c2h_mac_tx_rpt *rpt; + + rpt = (const struct rtw89_c2h_mac_tx_rpt *)c2h->data; + sw_define = le32_get_bits(rpt->w2, RTW89_C2H_MAC_TX_RPT_W2_SW_DEFINE); + tx_status = le32_get_bits(rpt->w2, RTW89_C2H_MAC_TX_RPT_W2_TX_STATE); + if (rtwdev->chip->chip_id == RTL8852C) + txcnt = le32_get_bits(rpt->w5, + RTW89_C2H_MAC_TX_RPT_W5_DATA_TX_CNT_V1); + else + txcnt = le32_get_bits(rpt->w5, + RTW89_C2H_MAC_TX_RPT_W5_DATA_TX_CNT); + } + + rtw89_debug(rtwdev, RTW89_DBG_TXRX, + "C2H TX RPT: sn %d, tx_status %d, txcnt %d\n", + sw_define, tx_status, txcnt); +} + static void rtw89_mac_c2h_mrc_tsf_rpt(struct rtw89_dev *rtwdev, struct sk_buff *c2h, u32 len) { @@ -5722,6 +5756,12 @@ void (* const rtw89_mac_c2h_mcc_handler[])(struct rtw89_dev *rtwdev, [RTW89_MAC_C2H_FUNC_MCC_STATUS_RPT] = rtw89_mac_c2h_mcc_status_rpt, }; +static +void (* const rtw89_mac_c2h_misc_handler[])(struct rtw89_dev *rtwdev, + struct sk_buff *c2h, u32 len) = { + [RTW89_MAC_C2H_FUNC_TX_REPORT] = rtw89_mac_c2h_tx_rpt, +}; + static void (* const rtw89_mac_c2h_mlo_handler[])(struct rtw89_dev *rtwdev, struct sk_buff *c2h, u32 len) = { @@ -5808,6 +5848,8 @@ bool rtw89_mac_c2h_chk_atomic(struct rtw89_dev *rtwdev, struct sk_buff *c2h, } case RTW89_MAC_C2H_CLASS_MCC: return true; + case RTW89_MAC_C2H_CLASS_MISC: + return true; case RTW89_MAC_C2H_CLASS_MLO: return true; case RTW89_MAC_C2H_CLASS_MRC: @@ -5843,6 +5885,10 @@ void rtw89_mac_c2h_handle(struct rtw89_dev *rtwdev, struct sk_buff *skb, if (func < NUM_OF_RTW89_MAC_C2H_FUNC_MCC) handler = rtw89_mac_c2h_mcc_handler[func]; break; + case RTW89_MAC_C2H_CLASS_MISC: + if (func < NUM_OF_RTW89_MAC_C2H_FUNC_MISC) + handler = rtw89_mac_c2h_misc_handler[func]; + break; case RTW89_MAC_C2H_CLASS_MLO: if (func < NUM_OF_RTW89_MAC_C2H_FUNC_MLO) handler = rtw89_mac_c2h_mlo_handler[func]; diff --git a/drivers/net/wireless/realtek/rtw89/mac.h b/drivers/net/wireless/realtek/rtw89/mac.h index 46302f4aa3d9..dfa85ade38ce 100644 --- a/drivers/net/wireless/realtek/rtw89/mac.h +++ b/drivers/net/wireless/realtek/rtw89/mac.h @@ -432,6 +432,12 @@ enum rtw89_mac_c2h_mcc_func { NUM_OF_RTW89_MAC_C2H_FUNC_MCC, }; +enum rtw89_mac_c2h_misc_func { + RTW89_MAC_C2H_FUNC_TX_REPORT = 1, + + NUM_OF_RTW89_MAC_C2H_FUNC_MISC, +}; + enum rtw89_mac_c2h_mlo_func { RTW89_MAC_C2H_FUNC_MLO_GET_TBL = 0x0, RTW89_MAC_C2H_FUNC_MLO_EMLSR_TRANS_DONE = 0x1, @@ -470,6 +476,7 @@ enum rtw89_mac_c2h_class { RTW89_MAC_C2H_CLASS_WOW = 0x3, RTW89_MAC_C2H_CLASS_MCC = 0x4, RTW89_MAC_C2H_CLASS_FWDBG = 0x5, + RTW89_MAC_C2H_CLASS_MISC = 0x9, RTW89_MAC_C2H_CLASS_MLO = 0xc, RTW89_MAC_C2H_CLASS_MRC = 0xe, RTW89_MAC_C2H_CLASS_AP = 0x18, diff --git a/drivers/net/wireless/realtek/rtw89/txrx.h b/drivers/net/wireless/realtek/rtw89/txrx.h index 984c9fdbb018..b37dbac7b790 100644 --- a/drivers/net/wireless/realtek/rtw89/txrx.h +++ b/drivers/net/wireless/realtek/rtw89/txrx.h @@ -127,6 +127,8 @@ static inline u8 rtw89_get_data_nss(struct rtw89_dev *rtwdev, u16 hw_rate) #define RTW89_TXWD_INFO0_MULTIPORT_ID GENMASK(6, 4) /* TX WD INFO DWORD 1 */ +#define RTW89_TXWD_INFO1_DATA_TXCNT_LMT_SEL BIT(31) +#define RTW89_TXWD_INFO1_DATA_TXCNT_LMT GENMASK(30, 25) #define RTW89_TXWD_INFO1_DATA_RTY_LOWEST_RATE GENMASK(24, 16) #define RTW89_TXWD_INFO1_A_CTRL_BSR BIT(14) #define RTW89_TXWD_INFO1_MAX_AGGNUM GENMASK(7, 0) @@ -139,10 +141,12 @@ static inline u8 rtw89_get_data_nss(struct rtw89_dev *rtwdev, u16 hw_rate) #define RTW89_TXWD_INFO2_SEC_CAM_IDX GENMASK(7, 0) /* TX WD INFO DWORD 3 */ +#define RTW89_TXWD_INFO3_SPE_RPT BIT(10) /* TX WD INFO DWORD 4 */ -#define RTW89_TXWD_INFO4_RTS_EN BIT(27) #define RTW89_TXWD_INFO4_HW_RTS_EN BIT(31) +#define RTW89_TXWD_INFO4_RTS_EN BIT(27) +#define RTW89_TXWD_INFO4_SW_DEFINE GENMASK(3, 0) /* TX WD INFO DWORD 5 */ From 21b946104087cd48e16ea51903c421cc3962225b Mon Sep 17 00:00:00 2001 From: Ping-Ke Shih Date: Tue, 4 Nov 2025 16:57:13 +0300 Subject: [PATCH 0552/1381] wifi: rtw89: fill TX descriptor of FWCMD in shortcut TX type FWCMD is used to download firmware and send H2C commands, and it only fill few fields of TX description, such as desc_info->pkt_size. Therefore, early return the TX type FWCMD. Signed-off-by: Ping-Ke Shih Signed-off-by: Fedor Pchelkin Link: https://patch.msgid.link/20251104135720.321110-7-pchelkin@ispras.ru --- drivers/net/wireless/realtek/rtw89/core.c | 26 ++++++++++++----------- 1 file changed, 14 insertions(+), 12 deletions(-) diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c index 84f94c796a48..fd6624d12efe 100644 --- a/drivers/net/wireless/realtek/rtw89/core.c +++ b/drivers/net/wireless/realtek/rtw89/core.c @@ -1106,32 +1106,35 @@ rtw89_core_tx_update_desc_info(struct rtw89_dev *rtwdev, struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb); struct ieee80211_hdr *hdr = (void *)skb->data; struct rtw89_addr_cam_entry *addr_cam; - enum rtw89_core_tx_type tx_type; enum btc_pkt_type pkt_type; bool upd_wlan_hdr = false; bool is_bmc; u16 seq; + desc_info->pkt_size = skb->len; + + if (unlikely(tx_req->tx_type == RTW89_CORE_TX_TYPE_FWCMD)) { + rtw89_core_tx_update_h2c_info(rtwdev, tx_req); + return; + } + + tx_req->tx_type = rtw89_core_get_tx_type(rtwdev, skb); + if (tx_req->sta) desc_info->mlo = tx_req->sta->mlo; else if (tx_req->vif) desc_info->mlo = ieee80211_vif_is_mld(tx_req->vif); seq = (le16_to_cpu(hdr->seq_ctrl) & IEEE80211_SCTL_SEQ) >> 4; - if (tx_req->tx_type != RTW89_CORE_TX_TYPE_FWCMD) { - tx_type = rtw89_core_get_tx_type(rtwdev, skb); - tx_req->tx_type = tx_type; + addr_cam = rtw89_get_addr_cam_of(tx_req->rtwvif_link, + tx_req->rtwsta_link); + if (addr_cam->valid && desc_info->mlo) + upd_wlan_hdr = true; - addr_cam = rtw89_get_addr_cam_of(tx_req->rtwvif_link, - tx_req->rtwsta_link); - if (addr_cam->valid && desc_info->mlo) - upd_wlan_hdr = true; - } is_bmc = (is_broadcast_ether_addr(hdr->addr1) || is_multicast_ether_addr(hdr->addr1)); desc_info->seq = seq; - desc_info->pkt_size = skb->len; desc_info->is_bmc = is_bmc; desc_info->wd_page = true; desc_info->hiq = info->flags & IEEE80211_TX_CTL_SEND_AFTER_DTIM; @@ -1148,8 +1151,7 @@ rtw89_core_tx_update_desc_info(struct rtw89_dev *rtwdev, rtw89_core_tx_update_ampdu_info(rtwdev, tx_req, pkt_type); rtw89_core_tx_update_llc_hdr(rtwdev, desc_info, skb); break; - case RTW89_CORE_TX_TYPE_FWCMD: - rtw89_core_tx_update_h2c_info(rtwdev, tx_req); + default: break; } } From 26a42d804aa89f6e62be68126114a4b62a8fe713 Mon Sep 17 00:00:00 2001 From: Fedor Pchelkin Date: Tue, 4 Nov 2025 16:57:14 +0300 Subject: [PATCH 0553/1381] wifi: rtw89: usb: anchor TX URBs During HCI reset all pending TX URBs should be canceled. Use anchor to keep track of them and have an ability to cancel them synchronously. Note however that canceling RX URBs can't be done here in rtw89_usb_ops_reset() as it breaks driver initialization. Found by Linux Verification Center (linuxtesting.org). Acked-by: Ping-Ke Shih Signed-off-by: Fedor Pchelkin Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251104135720.321110-8-pchelkin@ispras.ru --- drivers/net/wireless/realtek/rtw89/usb.c | 24 ++++++++++++++++++++---- drivers/net/wireless/realtek/rtw89/usb.h | 1 + 2 files changed, 21 insertions(+), 4 deletions(-) diff --git a/drivers/net/wireless/realtek/rtw89/usb.c b/drivers/net/wireless/realtek/rtw89/usb.c index 59cdee54feef..b7b981bac7bf 100644 --- a/drivers/net/wireless/realtek/rtw89/usb.c +++ b/drivers/net/wireless/realtek/rtw89/usb.c @@ -221,7 +221,6 @@ static void rtw89_usb_write_port_complete(struct urb *urb) } kfree(txcb); - usb_free_urb(urb); } static int rtw89_usb_write_port(struct rtw89_dev *rtwdev, u8 ch_dma, @@ -247,10 +246,17 @@ static int rtw89_usb_write_port(struct rtw89_dev *rtwdev, u8 ch_dma, usb_fill_bulk_urb(urb, usbd, pipe, data, len, rtw89_usb_write_port_complete, context); urb->transfer_flags |= URB_ZERO_PACKET; - ret = usb_submit_urb(urb, GFP_ATOMIC); + usb_anchor_urb(urb, &rtwusb->tx_submitted); + ret = usb_submit_urb(urb, GFP_ATOMIC); if (ret) - usb_free_urb(urb); + usb_unanchor_urb(urb); + + /* release our reference to this URB, USB core will eventually free it + * on its own after the completion callback finishes (or URB is + * immediately freed here if its submission has failed) + */ + usb_free_urb(urb); if (ret == -ENODEV) set_bit(RTW89_FLAG_UNPLUGGED, rtwdev->flags); @@ -557,6 +563,11 @@ static void rtw89_usb_cancel_rx_bufs(struct rtw89_usb *rtwusb) } } +static void rtw89_usb_cancel_tx_bufs(struct rtw89_usb *rtwusb) +{ + usb_kill_anchored_urbs(&rtwusb->tx_submitted); +} + static void rtw89_usb_free_rx_bufs(struct rtw89_usb *rtwusb) { struct rtw89_usb_rx_ctrl_block *rxcb; @@ -658,7 +669,9 @@ static void rtw89_usb_deinit_tx(struct rtw89_dev *rtwdev) static void rtw89_usb_ops_reset(struct rtw89_dev *rtwdev) { - /* TODO: anything to do here? */ + struct rtw89_usb *rtwusb = rtw89_usb_priv(rtwdev); + + rtw89_usb_cancel_tx_bufs(rtwusb); } static int rtw89_usb_ops_start(struct rtw89_dev *rtwdev) @@ -895,6 +908,8 @@ static int rtw89_usb_intf_init(struct rtw89_dev *rtwdev, struct rtw89_usb *rtwusb = rtw89_usb_priv(rtwdev); int ret; + init_usb_anchor(&rtwusb->tx_submitted); + ret = rtw89_usb_parse(rtwdev, intf); if (ret) return ret; @@ -1021,6 +1036,7 @@ void rtw89_usb_disconnect(struct usb_interface *intf) rtwusb = rtw89_usb_priv(rtwdev); rtw89_usb_cancel_rx_bufs(rtwusb); + rtw89_usb_cancel_tx_bufs(rtwusb); rtw89_core_unregister(rtwdev); rtw89_core_deinit(rtwdev); diff --git a/drivers/net/wireless/realtek/rtw89/usb.h b/drivers/net/wireless/realtek/rtw89/usb.h index 32cdbf5128e4..203ec8e993e9 100644 --- a/drivers/net/wireless/realtek/rtw89/usb.h +++ b/drivers/net/wireless/realtek/rtw89/usb.h @@ -60,6 +60,7 @@ struct rtw89_usb { struct sk_buff_head rx_free_queue; struct work_struct rx_work; struct work_struct rx_urb_work; + struct usb_anchor tx_submitted; struct sk_buff_head tx_queue[RTW89_TXCH_NUM]; }; From 816e849ef83aaf989035545d4a927b50a3d03525 Mon Sep 17 00:00:00 2001 From: Fedor Pchelkin Date: Tue, 4 Nov 2025 16:57:15 +0300 Subject: [PATCH 0554/1381] wifi: rtw89: handle IEEE80211_TX_CTL_REQ_TX_STATUS frames for USB Frames flagged with IEEE80211_TX_CTL_REQ_TX_STATUS mean the driver has to report to mac80211 stack whether AP sent ACK for the null frame/probe request or not. It's not implemented in USB part of the driver yet. PCIe HCI has its own way of getting TX status incorporated into RPP feature, and it's always enabled there. Other HCIs need a different scheme based on processing C2H messages. Thus define a .tx_rpt_enabled flag indicating which HCIs need to enable a TX report feature. Currently it is USB only. Toggle a bit in the TX descriptor and place flagged skbs in a fix-sized queue to wait for a message from the firmware. Firmware maintains a 4-bit sequence number for required frames hence the queue can contain just 16 elements simultaneously. That's enough for normal driver / firmware communication. If the firmware crashes for any reason and doesn't provide TX reports in time, driver will handle this and report the obsolete frames as dropped. rtw89 also has a new feature providing a TX report for each transmission attempt. Ignore a failed TX status reported by the firmware until retry limit is reached or successful status appears. When there is no success and the retry limit is reached, report the frame up to the wireless stack as failed eventually. HCI reset should stop all pending TX activity so forcefully flush the queue there. Found by Linux Verification Center (linuxtesting.org). Signed-off-by: Fedor Pchelkin Acked-by: Ping-Ke Shih Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251104135720.321110-9-pchelkin@ispras.ru --- drivers/net/wireless/realtek/rtw89/core.c | 4 ++ drivers/net/wireless/realtek/rtw89/core.h | 14 ++++ drivers/net/wireless/realtek/rtw89/mac.c | 32 +++++++++ drivers/net/wireless/realtek/rtw89/mac.h | 81 +++++++++++++++++++++++ drivers/net/wireless/realtek/rtw89/usb.c | 18 +++++ 5 files changed, 149 insertions(+) diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c index fd6624d12efe..6059e30a0590 100644 --- a/drivers/net/wireless/realtek/rtw89/core.c +++ b/drivers/net/wireless/realtek/rtw89/core.c @@ -1131,6 +1131,9 @@ rtw89_core_tx_update_desc_info(struct rtw89_dev *rtwdev, if (addr_cam->valid && desc_info->mlo) upd_wlan_hdr = true; + if (rtw89_is_tx_rpt_skb(tx_req->skb)) + rtw89_tx_rpt_init(rtwdev, tx_req); + is_bmc = (is_broadcast_ether_addr(hdr->addr1) || is_multicast_ether_addr(hdr->addr1)); @@ -5865,6 +5868,7 @@ int rtw89_core_init(struct rtw89_dev *rtwdev) wiphy_work_init(&rtwdev->cancel_6ghz_probe_work, rtw89_cancel_6ghz_probe_work); INIT_WORK(&rtwdev->load_firmware_work, rtw89_load_firmware_work); + spin_lock_init(&rtwdev->tx_rpt.skb_lock); skb_queue_head_init(&rtwdev->c2h_queue); rtw89_core_ppdu_sts_init(rtwdev); rtw89_traffic_stats_init(rtwdev, &rtwdev->stats); diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h index 2914cc4e97d5..cb71bd87b193 100644 --- a/drivers/net/wireless/realtek/rtw89/core.h +++ b/drivers/net/wireless/realtek/rtw89/core.h @@ -3518,6 +3518,15 @@ struct rtw89_phy_rate_pattern { #define RTW89_TX_LIFE_TIME 0x2 #define RTW89_TX_MACID_DROP 0x3 +#define RTW89_MAX_TX_RPTS 16 +#define RTW89_MAX_TX_RPTS_MASK (RTW89_MAX_TX_RPTS - 1) +struct rtw89_tx_rpt { + struct sk_buff *skbs[RTW89_MAX_TX_RPTS]; + /* protect skbs array access/modification */ + spinlock_t skb_lock; + atomic_t sn; +}; + #define RTW89_TX_WAIT_WORK_TIMEOUT msecs_to_jiffies(500) struct rtw89_tx_wait_info { struct rcu_head rcu_head; @@ -3529,6 +3538,8 @@ struct rtw89_tx_wait_info { struct rtw89_tx_skb_data { struct rtw89_tx_wait_info __rcu *wait; + u8 tx_rpt_sn; + u8 tx_pkt_cnt_lmt; u8 hci_priv[]; }; @@ -3698,6 +3709,7 @@ struct rtw89_hci_info { u32 rpwm_addr; u32 cpwm_addr; bool paused; + bool tx_rpt_enabled; }; struct rtw89_chip_ops { @@ -6021,6 +6033,8 @@ struct rtw89_dev { struct list_head tx_waits; struct wiphy_delayed_work tx_wait_work; + struct rtw89_tx_rpt tx_rpt; + struct rtw89_cam_info cam_info; struct sk_buff_head c2h_queue; diff --git a/drivers/net/wireless/realtek/rtw89/mac.c b/drivers/net/wireless/realtek/rtw89/mac.c index 47048d125c01..0a8474002cb7 100644 --- a/drivers/net/wireless/realtek/rtw89/mac.c +++ b/drivers/net/wireless/realtek/rtw89/mac.c @@ -5491,7 +5491,10 @@ rtw89_mac_c2h_mcc_status_rpt(struct rtw89_dev *rtwdev, struct sk_buff *c2h, u32 static void rtw89_mac_c2h_tx_rpt(struct rtw89_dev *rtwdev, struct sk_buff *c2h, u32 len) { + struct rtw89_tx_rpt *tx_rpt = &rtwdev->tx_rpt; + struct rtw89_tx_skb_data *skb_data; u8 sw_define, tx_status, txcnt; + struct sk_buff *skb; if (rtwdev->chip->chip_id == RTL8922A) { const struct rtw89_c2h_mac_tx_rpt_v2 *rpt_v2; @@ -5520,6 +5523,35 @@ rtw89_mac_c2h_tx_rpt(struct rtw89_dev *rtwdev, struct sk_buff *c2h, u32 len) rtw89_debug(rtwdev, RTW89_DBG_TXRX, "C2H TX RPT: sn %d, tx_status %d, txcnt %d\n", sw_define, tx_status, txcnt); + + /* claim sw_define is not over size of tx_rpt->skbs[] */ + static_assert(hweight32(RTW89_MAX_TX_RPTS_MASK) == + hweight32(RTW89_C2H_MAC_TX_RPT_W12_SW_DEFINE_V2) && + hweight32(RTW89_MAX_TX_RPTS_MASK) == + hweight32(RTW89_C2H_MAC_TX_RPT_W2_SW_DEFINE)); + + scoped_guard(spinlock_irqsave, &tx_rpt->skb_lock) { + skb = tx_rpt->skbs[sw_define]; + + /* skip if no skb (normally shouldn't happen) */ + if (!skb) { + rtw89_debug(rtwdev, RTW89_DBG_TXRX, + "C2H TX RPT: no skb found in queue\n"); + return; + } + + skb_data = RTW89_TX_SKB_CB(skb); + + /* skip if TX attempt has failed and retry limit has not been + * reached yet + */ + if (tx_status != RTW89_TX_DONE && + txcnt != skb_data->tx_pkt_cnt_lmt) + return; + + tx_rpt->skbs[sw_define] = NULL; + rtw89_tx_rpt_tx_status(rtwdev, skb, tx_status); + } } static void diff --git a/drivers/net/wireless/realtek/rtw89/mac.h b/drivers/net/wireless/realtek/rtw89/mac.h index dfa85ade38ce..56751dd6e99b 100644 --- a/drivers/net/wireless/realtek/rtw89/mac.h +++ b/drivers/net/wireless/realtek/rtw89/mac.h @@ -1626,4 +1626,85 @@ int rtw89_mac_scan_offload(struct rtw89_dev *rtwdev, return ret; } + +static inline +void rtw89_tx_rpt_init(struct rtw89_dev *rtwdev, + struct rtw89_core_tx_request *tx_req) +{ + struct rtw89_tx_rpt *tx_rpt = &rtwdev->tx_rpt; + + if (!rtwdev->hci.tx_rpt_enabled) + return; + + tx_req->desc_info.report = true; + /* firmware maintains a 4-bit sequence number */ + tx_req->desc_info.sn = atomic_inc_return(&tx_rpt->sn) & + RTW89_MAX_TX_RPTS_MASK; + tx_req->desc_info.tx_cnt_lmt_en = true; + tx_req->desc_info.tx_cnt_lmt = 8; +} + +static inline +bool rtw89_is_tx_rpt_skb(struct sk_buff *skb) +{ + struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb); + + return info->flags & IEEE80211_TX_CTL_REQ_TX_STATUS; +} + +static inline +void rtw89_tx_rpt_tx_status(struct rtw89_dev *rtwdev, struct sk_buff *skb, + u8 tx_status) +{ + struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb); + + ieee80211_tx_info_clear_status(info); + + if (tx_status == RTW89_TX_DONE) + info->flags |= IEEE80211_TX_STAT_ACK; + else + info->flags &= ~IEEE80211_TX_STAT_ACK; + + ieee80211_tx_status_irqsafe(rtwdev->hw, skb); +} + +static inline +void rtw89_tx_rpt_skb_add(struct rtw89_dev *rtwdev, struct sk_buff *skb) +{ + struct rtw89_tx_rpt *tx_rpt = &rtwdev->tx_rpt; + struct rtw89_tx_skb_data *skb_data; + u8 idx; + + skb_data = RTW89_TX_SKB_CB(skb); + idx = skb_data->tx_rpt_sn; + + scoped_guard(spinlock_irqsave, &tx_rpt->skb_lock) { + /* if skb having the similar seq number is still in the queue, + * this means the queue is overflowed - it isn't normal and + * should indicate firmware doesn't provide TX reports in time; + * report the old skb as dropped, we can't do much more here + */ + if (tx_rpt->skbs[idx]) + rtw89_tx_rpt_tx_status(rtwdev, tx_rpt->skbs[idx], + RTW89_TX_MACID_DROP); + tx_rpt->skbs[idx] = skb; + } +} + +static inline +void rtw89_tx_rpt_skbs_purge(struct rtw89_dev *rtwdev) +{ + struct rtw89_tx_rpt *tx_rpt = &rtwdev->tx_rpt; + struct sk_buff *skbs[RTW89_MAX_TX_RPTS]; + + scoped_guard(spinlock_irqsave, &tx_rpt->skb_lock) { + memcpy(skbs, tx_rpt->skbs, sizeof(tx_rpt->skbs)); + memset(tx_rpt->skbs, 0, sizeof(tx_rpt->skbs)); + } + + for (int i = 0; i < ARRAY_SIZE(skbs); i++) + if (skbs[i]) + rtw89_tx_rpt_tx_status(rtwdev, skbs[i], + RTW89_TX_MACID_DROP); +} #endif diff --git a/drivers/net/wireless/realtek/rtw89/usb.c b/drivers/net/wireless/realtek/rtw89/usb.c index b7b981bac7bf..f54e00c3033e 100644 --- a/drivers/net/wireless/realtek/rtw89/usb.c +++ b/drivers/net/wireless/realtek/rtw89/usb.c @@ -194,6 +194,15 @@ static void rtw89_usb_write_port_complete(struct urb *urb) skb_pull(skb, txdesc_size); + if (rtw89_is_tx_rpt_skb(skb)) { + if (urb->status == 0) + rtw89_tx_rpt_skb_add(rtwdev, skb); + else + rtw89_tx_rpt_tx_status(rtwdev, skb, + RTW89_TX_MACID_DROP); + continue; + } + info = IEEE80211_SKB_CB(skb); ieee80211_tx_info_clear_status(info); @@ -358,6 +367,7 @@ static int rtw89_usb_ops_tx_write(struct rtw89_dev *rtwdev, { struct rtw89_tx_desc_info *desc_info = &tx_req->desc_info; struct rtw89_usb *rtwusb = rtw89_usb_priv(rtwdev); + struct rtw89_tx_skb_data *skb_data; struct sk_buff *skb = tx_req->skb; struct rtw89_txwd_body *txdesc; u32 txdesc_size; @@ -384,6 +394,12 @@ static int rtw89_usb_ops_tx_write(struct rtw89_dev *rtwdev, le32p_replace_bits(&txdesc->dword0, 1, RTW89_TXWD_BODY0_STF_MODE); + skb_data = RTW89_TX_SKB_CB(skb); + if (tx_req->desc_info.sn) + skb_data->tx_rpt_sn = tx_req->desc_info.sn; + if (tx_req->desc_info.tx_cnt_lmt) + skb_data->tx_pkt_cnt_lmt = tx_req->desc_info.tx_cnt_lmt; + skb_queue_tail(&rtwusb->tx_queue[desc_info->ch_dma], skb); return 0; @@ -672,6 +688,7 @@ static void rtw89_usb_ops_reset(struct rtw89_dev *rtwdev) struct rtw89_usb *rtwusb = rtw89_usb_priv(rtwdev); rtw89_usb_cancel_tx_bufs(rtwusb); + rtw89_tx_rpt_skbs_purge(rtwdev); } static int rtw89_usb_ops_start(struct rtw89_dev *rtwdev) @@ -962,6 +979,7 @@ int rtw89_usb_probe(struct usb_interface *intf, rtwdev->hci.ops = &rtw89_usb_ops; rtwdev->hci.type = RTW89_HCI_TYPE_USB; + rtwdev->hci.tx_rpt_enabled = true; ret = rtw89_usb_intf_init(rtwdev, intf); if (ret) { From cc7070e41786a51953f71e39a1e6e803a1ed2d50 Mon Sep 17 00:00:00 2001 From: Fedor Pchelkin Date: Tue, 4 Nov 2025 16:57:16 +0300 Subject: [PATCH 0555/1381] wifi: rtw89: provide TX reports for management frames In order to provide TX reports for the management queue rtw89 should configure the firmware. Do this with SET_CMC_TBL_MGQ_RPT_EN() for the WiFi6 chips and with CCTLINFO_G7_W0_MGQ_RPT_EN flag for the WiFi7 ones. Suggested-by: Bitterblue Smith Acked-by: Ping-Ke Shih Signed-off-by: Fedor Pchelkin Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251104135720.321110-10-pchelkin@ispras.ru --- drivers/net/wireless/realtek/rtw89/fw.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/realtek/rtw89/fw.c b/drivers/net/wireless/realtek/rtw89/fw.c index cb431c8a65ac..96f0463e66d6 100644 --- a/drivers/net/wireless/realtek/rtw89/fw.c +++ b/drivers/net/wireless/realtek/rtw89/fw.c @@ -3165,6 +3165,7 @@ int rtw89_fw_h2c_default_cmac_tbl(struct rtw89_dev *rtwdev, SET_CMC_TBL_ANTSEL_C(skb->data, 0); SET_CMC_TBL_ANTSEL_D(skb->data, 0); } + SET_CMC_TBL_MGQ_RPT_EN(skb->data, rtwdev->hci.tx_rpt_enabled); SET_CMC_TBL_DOPPLER_CTRL(skb->data, 0); SET_CMC_TBL_TXPWR_TOLERENCE(skb->data, 0); if (rtwvif_link->net_type == RTW89_NET_TYPE_AP_MODE) @@ -3210,7 +3211,8 @@ int rtw89_fw_h2c_default_cmac_tbl_g7(struct rtw89_dev *rtwdev, h2c->c0 = le32_encode_bits(mac_id, CCTLINFO_G7_C0_MACID) | le32_encode_bits(1, CCTLINFO_G7_C0_OP); - h2c->w0 = le32_encode_bits(4, CCTLINFO_G7_W0_DATARATE); + h2c->w0 = le32_encode_bits(4, CCTLINFO_G7_W0_DATARATE) | + le32_encode_bits(rtwdev->hci.tx_rpt_enabled, CCTLINFO_G7_W0_MGQ_RPT_EN); h2c->m0 = cpu_to_le32(CCTLINFO_G7_W0_ALL); h2c->w1 = le32_encode_bits(4, CCTLINFO_G7_W1_DATA_RTY_LOWEST_RATE) | From d5da3d9fb05f6e2b2685c58b37067749b342b518 Mon Sep 17 00:00:00 2001 From: Fedor Pchelkin Date: Tue, 4 Nov 2025 16:57:17 +0300 Subject: [PATCH 0556/1381] wifi: rtw89: process TX wait skbs for USB via C2H handler TX wait skbs need to be completed when they are done. PCIe part does this inside rtw89_pci_tx_status() during RPP processing. Other HCIs use a mechanism based on C2H firmware messages. Store TX wait skbs inside TX report queue so that it'll be possible to identify completed items inside the C2H handler. Try to do this as similar to PCIe path as possible. When the corresponding TX wait skb is found inside TX report queue, unlink it from there and call rtw89_core_tx_wait_complete() to mark the completion. If the callee waiting for the completion has already timed out, the TX wait skb is placed into TX wait list (like PCIe part does). It's important that during HCI reset all pending TX wait frames should be completed inside hci.ops->reset method before calling rtw89_tx_wait_list_clear(). Found by Linux Verification Center (linuxtesting.org). Acked-by: Ping-Ke Shih Signed-off-by: Fedor Pchelkin Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251104135720.321110-11-pchelkin@ispras.ru --- drivers/net/wireless/realtek/rtw89/core.c | 5 ++--- drivers/net/wireless/realtek/rtw89/core.h | 7 +++++++ drivers/net/wireless/realtek/rtw89/mac.h | 13 ++++++++++--- drivers/net/wireless/realtek/rtw89/usb.c | 2 +- 4 files changed, 20 insertions(+), 7 deletions(-) diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c index 6059e30a0590..1d30ff27bce1 100644 --- a/drivers/net/wireless/realtek/rtw89/core.c +++ b/drivers/net/wireless/realtek/rtw89/core.c @@ -1131,7 +1131,7 @@ rtw89_core_tx_update_desc_info(struct rtw89_dev *rtwdev, if (addr_cam->valid && desc_info->mlo) upd_wlan_hdr = true; - if (rtw89_is_tx_rpt_skb(tx_req->skb)) + if (rtw89_is_tx_rpt_skb(rtwdev, tx_req->skb)) rtw89_tx_rpt_init(rtwdev, tx_req); is_bmc = (is_broadcast_ether_addr(hdr->addr1) || @@ -1263,14 +1263,13 @@ static int rtw89_core_tx_write_link(struct rtw89_dev *rtwdev, tx_req.rtwvif_link = rtwvif_link; tx_req.rtwsta_link = rtwsta_link; tx_req.desc_info.sw_mld = sw_mld; + rcu_assign_pointer(skb_data->wait, wait); rtw89_traffic_stats_accu(rtwdev, rtwvif, skb, true, true); rtw89_wow_parse_akm(rtwdev, skb); rtw89_core_tx_update_desc_info(rtwdev, &tx_req); rtw89_core_tx_wake(rtwdev, &tx_req); - rcu_assign_pointer(skb_data->wait, wait); - ret = rtw89_hci_tx_write(rtwdev, &tx_req); if (ret) { rtw89_err(rtwdev, "failed to transmit skb to HCI\n"); diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h index cb71bd87b193..71eb7078a0dc 100644 --- a/drivers/net/wireless/realtek/rtw89/core.h +++ b/drivers/net/wireless/realtek/rtw89/core.h @@ -6328,6 +6328,7 @@ static inline int rtw89_hci_tx_write(struct rtw89_dev *rtwdev, static inline void rtw89_hci_reset(struct rtw89_dev *rtwdev) { rtwdev->hci.ops->reset(rtwdev); + /* hci.ops->reset must complete all pending TX wait SKBs */ rtw89_tx_wait_list_clear(rtwdev); } @@ -7409,6 +7410,12 @@ static inline struct sk_buff *rtw89_alloc_skb_for_rx(struct rtw89_dev *rtwdev, return dev_alloc_skb(length); } +static inline bool rtw89_core_is_tx_wait(struct rtw89_dev *rtwdev, + struct rtw89_tx_skb_data *skb_data) +{ + return rcu_access_pointer(skb_data->wait); +} + static inline bool rtw89_core_tx_wait_complete(struct rtw89_dev *rtwdev, struct rtw89_tx_skb_data *skb_data, u8 tx_status) diff --git a/drivers/net/wireless/realtek/rtw89/mac.h b/drivers/net/wireless/realtek/rtw89/mac.h index 56751dd6e99b..ce3474a7ba03 100644 --- a/drivers/net/wireless/realtek/rtw89/mac.h +++ b/drivers/net/wireless/realtek/rtw89/mac.h @@ -1645,19 +1645,26 @@ void rtw89_tx_rpt_init(struct rtw89_dev *rtwdev, } static inline -bool rtw89_is_tx_rpt_skb(struct sk_buff *skb) +bool rtw89_is_tx_rpt_skb(struct rtw89_dev *rtwdev, struct sk_buff *skb) { + struct rtw89_tx_skb_data *skb_data = RTW89_TX_SKB_CB(skb); struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb); - return info->flags & IEEE80211_TX_CTL_REQ_TX_STATUS; + return rtw89_core_is_tx_wait(rtwdev, skb_data) || + (info->flags & IEEE80211_TX_CTL_REQ_TX_STATUS); } static inline void rtw89_tx_rpt_tx_status(struct rtw89_dev *rtwdev, struct sk_buff *skb, u8 tx_status) { - struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb); + struct rtw89_tx_skb_data *skb_data = RTW89_TX_SKB_CB(skb); + struct ieee80211_tx_info *info; + if (rtw89_core_tx_wait_complete(rtwdev, skb_data, tx_status)) + return; + + info = IEEE80211_SKB_CB(skb); ieee80211_tx_info_clear_status(info); if (tx_status == RTW89_TX_DONE) diff --git a/drivers/net/wireless/realtek/rtw89/usb.c b/drivers/net/wireless/realtek/rtw89/usb.c index f54e00c3033e..d7d968207a39 100644 --- a/drivers/net/wireless/realtek/rtw89/usb.c +++ b/drivers/net/wireless/realtek/rtw89/usb.c @@ -194,7 +194,7 @@ static void rtw89_usb_write_port_complete(struct urb *urb) skb_pull(skb, txdesc_size); - if (rtw89_is_tx_rpt_skb(skb)) { + if (rtw89_is_tx_rpt_skb(rtwdev, skb)) { if (urb->status == 0) rtw89_tx_rpt_skb_add(rtwdev, skb); else From 6b47af35a6dded074ff583361f6d6668dd7a401d Mon Sep 17 00:00:00 2001 From: Raju Rangoju Date: Fri, 31 Oct 2025 16:48:11 +0530 Subject: [PATCH 0557/1381] net: selftests: export packet creation helpers for driver use Export the network selftest packet creation infrastructure to allow network drivers to reuse the existing selftest framework instead of duplicating packet creation code. Signed-off-by: Raju Rangoju Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/20251031111811.775434-1-Raju.Rangoju@amd.com Signed-off-by: Paolo Abeni --- include/net/selftests.h | 45 ++++++++++++++++++++++++++++++++++++++ net/core/selftests.c | 48 ++++++----------------------------------- 2 files changed, 52 insertions(+), 41 deletions(-) diff --git a/include/net/selftests.h b/include/net/selftests.h index e65e8d230d33..c36e07406ad4 100644 --- a/include/net/selftests.h +++ b/include/net/selftests.h @@ -3,9 +3,48 @@ #define _NET_SELFTESTS #include +#include + +struct net_packet_attrs { + const unsigned char *src; + const unsigned char *dst; + u32 ip_src; + u32 ip_dst; + bool tcp; + u16 sport; + u16 dport; + int timeout; + int size; + int max_size; + u8 id; + u16 queue_mapping; + bool bad_csum; +}; + +struct net_test_priv { + struct net_packet_attrs *packet; + struct packet_type pt; + struct completion comp; + int double_vlan; + int vlan_id; + int ok; +}; + +struct netsfhdr { + __be32 version; + __be64 magic; + u8 id; +} __packed; + +#define NET_TEST_PKT_SIZE (sizeof(struct ethhdr) + sizeof(struct iphdr) + \ + sizeof(struct netsfhdr)) +#define NET_TEST_PKT_MAGIC 0xdeadcafecafedeadULL +#define NET_LB_TIMEOUT msecs_to_jiffies(200) #if IS_ENABLED(CONFIG_NET_SELFTESTS) +struct sk_buff *net_test_get_skb(struct net_device *ndev, u8 id, + struct net_packet_attrs *attr); void net_selftest(struct net_device *ndev, struct ethtool_test *etest, u64 *buf); int net_selftest_get_count(void); @@ -13,6 +52,12 @@ void net_selftest_get_strings(u8 *data); #else +static inline struct sk_buff *net_test_get_skb(struct net_device *ndev, u8 id, + struct net_packet_attrs *attr) +{ + return NULL; +} + static inline void net_selftest(struct net_device *ndev, struct ethtool_test *etest, u64 *buf) { diff --git a/net/core/selftests.c b/net/core/selftests.c index 3d79133a91a6..8b81feb82c4a 100644 --- a/net/core/selftests.c +++ b/net/core/selftests.c @@ -14,46 +14,10 @@ #include #include -struct net_packet_attrs { - const unsigned char *src; - const unsigned char *dst; - u32 ip_src; - u32 ip_dst; - bool tcp; - u16 sport; - u16 dport; - int timeout; - int size; - int max_size; - u8 id; - u16 queue_mapping; - bool bad_csum; -}; - -struct net_test_priv { - struct net_packet_attrs *packet; - struct packet_type pt; - struct completion comp; - int double_vlan; - int vlan_id; - int ok; -}; - -struct netsfhdr { - __be32 version; - __be64 magic; - u8 id; -} __packed; - static u8 net_test_next_id; -#define NET_TEST_PKT_SIZE (sizeof(struct ethhdr) + sizeof(struct iphdr) + \ - sizeof(struct netsfhdr)) -#define NET_TEST_PKT_MAGIC 0xdeadcafecafedeadULL -#define NET_LB_TIMEOUT msecs_to_jiffies(200) - -static struct sk_buff *net_test_get_skb(struct net_device *ndev, - struct net_packet_attrs *attr) +struct sk_buff *net_test_get_skb(struct net_device *ndev, u8 id, + struct net_packet_attrs *attr) { struct sk_buff *skb = NULL; struct udphdr *uhdr = NULL; @@ -142,8 +106,8 @@ static struct sk_buff *net_test_get_skb(struct net_device *ndev, shdr = skb_put(skb, sizeof(*shdr)); shdr->version = 0; shdr->magic = cpu_to_be64(NET_TEST_PKT_MAGIC); - attr->id = net_test_next_id; - shdr->id = net_test_next_id++; + attr->id = id; + shdr->id = id; if (attr->size) { void *payload = skb_put(skb, attr->size); @@ -190,6 +154,7 @@ static struct sk_buff *net_test_get_skb(struct net_device *ndev, return skb; } +EXPORT_SYMBOL_GPL(net_test_get_skb); static int net_test_loopback_validate(struct sk_buff *skb, struct net_device *ndev, @@ -286,12 +251,13 @@ static int __net_test_loopback(struct net_device *ndev, tpriv->packet = attr; dev_add_pack(&tpriv->pt); - skb = net_test_get_skb(ndev, attr); + skb = net_test_get_skb(ndev, net_test_next_id, attr); if (!skb) { ret = -ENOMEM; goto cleanup; } + net_test_next_id++; ret = dev_direct_xmit(skb, attr->queue_mapping); if (ret < 0) { goto cleanup; From 862a64c83faf7708e7e79498193ff5270543a68d Mon Sep 17 00:00:00 2001 From: Raju Rangoju Date: Fri, 31 Oct 2025 16:45:53 +0530 Subject: [PATCH 0558/1381] amd-xgbe: introduce support ethtool selftest Add support for ethtool selftest for MAC loopback. This includes the sanity check and helps in finding the misconfiguration of HW. Uses the existing selftest infrastructure to create test packets. Signed-off-by: Raju Rangoju Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/20251031111555.774425-2-Raju.Rangoju@amd.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/amd/Kconfig | 1 + drivers/net/ethernet/amd/xgbe/Makefile | 2 +- drivers/net/ethernet/amd/xgbe/xgbe-dev.c | 17 ++ drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c | 7 + drivers/net/ethernet/amd/xgbe/xgbe-selftest.c | 237 ++++++++++++++++++ drivers/net/ethernet/amd/xgbe/xgbe.h | 10 + 6 files changed, 273 insertions(+), 1 deletion(-) create mode 100644 drivers/net/ethernet/amd/xgbe/xgbe-selftest.c diff --git a/drivers/net/ethernet/amd/Kconfig b/drivers/net/ethernet/amd/Kconfig index b39c6f3e1eda..d54dca3074eb 100644 --- a/drivers/net/ethernet/amd/Kconfig +++ b/drivers/net/ethernet/amd/Kconfig @@ -165,6 +165,7 @@ config AMD_XGBE select CRC32 select PHYLIB select AMD_XGBE_HAVE_ECC if X86 + select NET_SELFTESTS help This driver supports the AMD 10GbE Ethernet device found on an AMD SoC. diff --git a/drivers/net/ethernet/amd/xgbe/Makefile b/drivers/net/ethernet/amd/xgbe/Makefile index 980e27652237..5992f7fd4d9b 100644 --- a/drivers/net/ethernet/amd/xgbe/Makefile +++ b/drivers/net/ethernet/amd/xgbe/Makefile @@ -5,7 +5,7 @@ amd-xgbe-objs := xgbe-main.o xgbe-drv.o xgbe-dev.o \ xgbe-desc.o xgbe-ethtool.o xgbe-mdio.o \ xgbe-hwtstamp.o xgbe-ptp.o xgbe-pps.o \ xgbe-i2c.o xgbe-phy-v1.o xgbe-phy-v2.o \ - xgbe-platform.o + xgbe-platform.o xgbe-selftest.o amd-xgbe-$(CONFIG_PCI) += xgbe-pci.o amd-xgbe-$(CONFIG_AMD_XGBE_DCB) += xgbe-dcb.o diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-dev.c b/drivers/net/ethernet/amd/xgbe/xgbe-dev.c index e5391a2eca51..ffc7d83522c7 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-dev.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-dev.c @@ -3578,3 +3578,20 @@ void xgbe_init_function_ptrs_dev(struct xgbe_hw_if *hw_if) DBGPR("<--xgbe_init_function_ptrs\n"); } + +int xgbe_enable_mac_loopback(struct xgbe_prv_data *pdata) +{ + /* Enable MAC loopback mode */ + XGMAC_IOWRITE_BITS(pdata, MAC_RCR, LM, 1); + + /* Wait for loopback to stabilize */ + usleep_range(10, 15); + + return 0; +} + +void xgbe_disable_mac_loopback(struct xgbe_prv_data *pdata) +{ + /* Disable MAC loopback mode */ + XGMAC_IOWRITE_BITS(pdata, MAC_RCR, LM, 0); +} diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c b/drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c index b6e1b67a2d0e..0d19b09497a0 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c @@ -85,6 +85,9 @@ static void xgbe_get_strings(struct net_device *netdev, u32 stringset, u8 *data) int i; switch (stringset) { + case ETH_SS_TEST: + xgbe_selftest_get_strings(pdata, data); + break; case ETH_SS_STATS: for (i = 0; i < XGBE_STATS_COUNT; i++) ethtool_puts(&data, xgbe_gstring_stats[i].stat_string); @@ -131,6 +134,9 @@ static int xgbe_get_sset_count(struct net_device *netdev, int stringset) int ret; switch (stringset) { + case ETH_SS_TEST: + ret = xgbe_selftest_get_count(pdata); + break; case ETH_SS_STATS: ret = XGBE_STATS_COUNT + (pdata->tx_ring_count * 2) + @@ -760,6 +766,7 @@ static const struct ethtool_ops xgbe_ethtool_ops = { .set_ringparam = xgbe_set_ringparam, .get_channels = xgbe_get_channels, .set_channels = xgbe_set_channels, + .self_test = xgbe_selftest_run, }; const struct ethtool_ops *xgbe_get_ethtool_ops(void) diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-selftest.c b/drivers/net/ethernet/amd/xgbe/xgbe-selftest.c new file mode 100644 index 000000000000..8a3a6279584d --- /dev/null +++ b/drivers/net/ethernet/amd/xgbe/xgbe-selftest.c @@ -0,0 +1,237 @@ +// SPDX-License-Identifier: (GPL-2.0-or-later OR BSD-3-Clause) +/* + * Copyright (c) 2014-2025, Advanced Micro Devices, Inc. + * Copyright (c) 2014, Synopsys, Inc. + * All rights reserved + * + * Author: Raju Rangoju + */ +#include +#include +#include +#include +#include +#include +#include + +#include "xgbe.h" +#include "xgbe-common.h" + +#define XGBE_LOOPBACK_NONE 0 +#define XGBE_LOOPBACK_MAC 1 + +struct xgbe_test { + char name[ETH_GSTRING_LEN]; + int lb; + int (*fn)(struct xgbe_prv_data *pdata); +}; + +static u8 xgbe_test_id; + +static int xgbe_test_loopback_validate(struct sk_buff *skb, + struct net_device *ndev, + struct packet_type *pt, + struct net_device *orig_ndev) +{ + struct net_test_priv *tdata = pt->af_packet_priv; + const unsigned char *dst = tdata->packet->dst; + const unsigned char *src = tdata->packet->src; + struct netsfhdr *hdr; + struct ethhdr *eh; + struct tcphdr *th; + struct udphdr *uh; + struct iphdr *ih; + + skb = skb_unshare(skb, GFP_ATOMIC); + if (!skb) + goto out; + + if (skb_linearize(skb)) + goto out; + + if (skb_headlen(skb) < (NET_TEST_PKT_SIZE - ETH_HLEN)) + goto out; + + eh = (struct ethhdr *)skb_mac_header(skb); + if (dst) { + if (!ether_addr_equal_unaligned(eh->h_dest, dst)) + goto out; + } + if (src) { + if (!ether_addr_equal_unaligned(eh->h_source, src)) + goto out; + } + + ih = ip_hdr(skb); + + if (tdata->packet->tcp) { + if (ih->protocol != IPPROTO_TCP) + goto out; + + th = (struct tcphdr *)((u8 *)ih + 4 * ih->ihl); + if (th->dest != htons(tdata->packet->dport)) + goto out; + + hdr = (struct netsfhdr *)((u8 *)th + sizeof(*th)); + } else { + if (ih->protocol != IPPROTO_UDP) + goto out; + + uh = (struct udphdr *)((u8 *)ih + 4 * ih->ihl); + if (uh->dest != htons(tdata->packet->dport)) + goto out; + + hdr = (struct netsfhdr *)((u8 *)uh + sizeof(*uh)); + } + + if (hdr->magic != cpu_to_be64(NET_TEST_PKT_MAGIC)) + goto out; + if (tdata->packet->id != hdr->id) + goto out; + + tdata->ok = true; + complete(&tdata->comp); +out: + kfree_skb(skb); + return 0; +} + +static int __xgbe_test_loopback(struct xgbe_prv_data *pdata, + struct net_packet_attrs *attr) +{ + struct net_test_priv *tdata; + struct sk_buff *skb = NULL; + int ret = 0; + + tdata = kzalloc(sizeof(*tdata), GFP_KERNEL); + if (!tdata) + return -ENOMEM; + + tdata->ok = false; + init_completion(&tdata->comp); + + tdata->pt.type = htons(ETH_P_IP); + tdata->pt.func = xgbe_test_loopback_validate; + tdata->pt.dev = pdata->netdev; + tdata->pt.af_packet_priv = tdata; + tdata->packet = attr; + + dev_add_pack(&tdata->pt); + + skb = net_test_get_skb(pdata->netdev, xgbe_test_id, attr); + if (!skb) { + ret = -ENOMEM; + goto cleanup; + } + + xgbe_test_id++; + ret = dev_direct_xmit(skb, attr->queue_mapping); + if (ret) + goto cleanup; + + if (!attr->timeout) + attr->timeout = NET_LB_TIMEOUT; + + wait_for_completion_timeout(&tdata->comp, attr->timeout); + ret = tdata->ok ? 0 : -ETIMEDOUT; + + if (ret) + netdev_err(pdata->netdev, "Response timedout: ret %d\n", ret); +cleanup: + dev_remove_pack(&tdata->pt); + kfree(tdata); + return ret; +} + +static int xgbe_test_mac_loopback(struct xgbe_prv_data *pdata) +{ + struct net_packet_attrs attr = {}; + + attr.dst = pdata->netdev->dev_addr; + return __xgbe_test_loopback(pdata, &attr); +} + +static const struct xgbe_test xgbe_selftests[] = { + { + .name = "MAC Loopback ", + .lb = XGBE_LOOPBACK_MAC, + .fn = xgbe_test_mac_loopback, + }, +}; + +void xgbe_selftest_run(struct net_device *dev, + struct ethtool_test *etest, u64 *buf) +{ + struct xgbe_prv_data *pdata = netdev_priv(dev); + int count = xgbe_selftest_get_count(pdata); + int i, ret; + + memset(buf, 0, sizeof(*buf) * count); + xgbe_test_id = 0; + + if (etest->flags != ETH_TEST_FL_OFFLINE) { + netdev_err(pdata->netdev, "Only offline tests are supported\n"); + etest->flags |= ETH_TEST_FL_FAILED; + return; + } else if (!netif_carrier_ok(dev)) { + netdev_err(pdata->netdev, + "Invalid link, cannot execute tests\n"); + etest->flags |= ETH_TEST_FL_FAILED; + return; + } + + /* Wait for queues drain */ + msleep(200); + + for (i = 0; i < count; i++) { + ret = 0; + + switch (xgbe_selftests[i].lb) { + case XGBE_LOOPBACK_MAC: + ret = xgbe_enable_mac_loopback(pdata); + break; + case XGBE_LOOPBACK_NONE: + break; + default: + ret = -EOPNOTSUPP; + break; + } + + /* + * First tests will always be MAC / PHY loopback. + * If any of them is not supported we abort earlier. + */ + if (ret) { + netdev_err(pdata->netdev, "Loopback not supported\n"); + etest->flags |= ETH_TEST_FL_FAILED; + break; + } + + ret = xgbe_selftests[i].fn(pdata); + if (ret && (ret != -EOPNOTSUPP)) + etest->flags |= ETH_TEST_FL_FAILED; + buf[i] = ret; + + switch (xgbe_selftests[i].lb) { + case XGBE_LOOPBACK_MAC: + xgbe_disable_mac_loopback(pdata); + break; + default: + break; + } + } +} + +void xgbe_selftest_get_strings(struct xgbe_prv_data *pdata, u8 *data) +{ + u8 *p = data; + int i; + + for (i = 0; i < xgbe_selftest_get_count(pdata); i++) + ethtool_puts(&p, xgbe_selftests[i].name); +} + +int xgbe_selftest_get_count(struct xgbe_prv_data *pdata) +{ + return ARRAY_SIZE(xgbe_selftests); +} diff --git a/drivers/net/ethernet/amd/xgbe/xgbe.h b/drivers/net/ethernet/amd/xgbe/xgbe.h index 381f72a33d1a..dc03082c59aa 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe.h +++ b/drivers/net/ethernet/amd/xgbe/xgbe.h @@ -1322,6 +1322,16 @@ void xgbe_update_tstamp_time(struct xgbe_prv_data *pdata, unsigned int sec, int xgbe_pps_config(struct xgbe_prv_data *pdata, struct xgbe_pps_config *cfg, int index, bool on); +/* Selftest functions */ +void xgbe_selftest_run(struct net_device *dev, + struct ethtool_test *etest, u64 *buf); +void xgbe_selftest_get_strings(struct xgbe_prv_data *pdata, u8 *data); +int xgbe_selftest_get_count(struct xgbe_prv_data *pdata); + +/* Loopback control */ +int xgbe_enable_mac_loopback(struct xgbe_prv_data *pdata); +void xgbe_disable_mac_loopback(struct xgbe_prv_data *pdata); + #ifdef CONFIG_DEBUG_FS void xgbe_debugfs_init(struct xgbe_prv_data *); void xgbe_debugfs_exit(struct xgbe_prv_data *); From 42b06fcc878d08785a0c44d2af42c8db453487e2 Mon Sep 17 00:00:00 2001 From: Raju Rangoju Date: Fri, 31 Oct 2025 16:45:55 +0530 Subject: [PATCH 0559/1381] amd-xgbe: add ethtool phy loopback selftest Add support for PHY loopback testing via ethtool self-test. The test uses phy_loopback() which enables PHY-level loopback through the PHY driver's set_loopback callback if provided, else uses the genphy_loopback(). Signed-off-by: Raju Rangoju Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/20251031111555.774425-3-Raju.Rangoju@amd.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/amd/xgbe/xgbe-selftest.c | 40 +++++++++++++++++++ 1 file changed, 40 insertions(+) diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-selftest.c b/drivers/net/ethernet/amd/xgbe/xgbe-selftest.c index 8a3a6279584d..23b9d568a861 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-selftest.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-selftest.c @@ -19,6 +19,7 @@ #define XGBE_LOOPBACK_NONE 0 #define XGBE_LOOPBACK_MAC 1 +#define XGBE_LOOPBACK_PHY 2 struct xgbe_test { char name[ETH_GSTRING_LEN]; @@ -151,11 +152,36 @@ static int xgbe_test_mac_loopback(struct xgbe_prv_data *pdata) return __xgbe_test_loopback(pdata, &attr); } +static int xgbe_test_phy_loopback(struct xgbe_prv_data *pdata) +{ + struct net_packet_attrs attr = {}; + int ret; + + if (!pdata->netdev->phydev) { + netdev_err(pdata->netdev, "phydev not found: cannot start PHY loopback test\n"); + return -EOPNOTSUPP; + } + + ret = phy_loopback(pdata->netdev->phydev, true, 0); + if (ret) + return ret; + + attr.dst = pdata->netdev->dev_addr; + ret = __xgbe_test_loopback(pdata, &attr); + + phy_loopback(pdata->netdev->phydev, false, 0); + return ret; +} + static const struct xgbe_test xgbe_selftests[] = { { .name = "MAC Loopback ", .lb = XGBE_LOOPBACK_MAC, .fn = xgbe_test_mac_loopback, + }, { + .name = "PHY Loopback ", + .lb = XGBE_LOOPBACK_NONE, + .fn = xgbe_test_phy_loopback, }, }; @@ -187,6 +213,13 @@ void xgbe_selftest_run(struct net_device *dev, ret = 0; switch (xgbe_selftests[i].lb) { + case XGBE_LOOPBACK_PHY: + ret = -EOPNOTSUPP; + if (dev->phydev) + ret = phy_loopback(dev->phydev, true, 0); + if (!ret) + break; + fallthrough; case XGBE_LOOPBACK_MAC: ret = xgbe_enable_mac_loopback(pdata); break; @@ -213,6 +246,13 @@ void xgbe_selftest_run(struct net_device *dev, buf[i] = ret; switch (xgbe_selftests[i].lb) { + case XGBE_LOOPBACK_PHY: + ret = -EOPNOTSUPP; + if (dev->phydev) + ret = phy_loopback(dev->phydev, false, 0); + if (!ret) + break; + fallthrough; case XGBE_LOOPBACK_MAC: xgbe_disable_mac_loopback(pdata); break; From d7735c6bb2310f7ca8235af7f946e6c8716cdb5e Mon Sep 17 00:00:00 2001 From: Raju Rangoju Date: Fri, 31 Oct 2025 16:45:57 +0530 Subject: [PATCH 0560/1381] amd-xgbe: add ethtool split header selftest Adds support for ethtool split header selftest. Performs UDP and TCP check to ensure split header selft test works for both packet types. Signed-off-by: Raju Rangoju Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/20251031111555.774425-4-Raju.Rangoju@amd.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/amd/xgbe/xgbe-dev.c | 2 + drivers/net/ethernet/amd/xgbe/xgbe-selftest.c | 46 +++++++++++++++++++ drivers/net/ethernet/amd/xgbe/xgbe.h | 1 + 3 files changed, 49 insertions(+) diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-dev.c b/drivers/net/ethernet/amd/xgbe/xgbe-dev.c index ffc7d83522c7..b646ae575e6a 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-dev.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-dev.c @@ -211,6 +211,7 @@ static void xgbe_config_sph_mode(struct xgbe_prv_data *pdata) } XGMAC_IOWRITE_BITS(pdata, MAC_RCR, HDSMS, XGBE_SPH_HDSMS_SIZE); + pdata->sph = true; } static void xgbe_disable_sph_mode(struct xgbe_prv_data *pdata) @@ -223,6 +224,7 @@ static void xgbe_disable_sph_mode(struct xgbe_prv_data *pdata) XGMAC_DMA_IOWRITE_BITS(pdata->channel[i], DMA_CH_CR, SPH, 0); } + pdata->sph = false; } static int xgbe_write_rss_reg(struct xgbe_prv_data *pdata, unsigned int type, diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-selftest.c b/drivers/net/ethernet/amd/xgbe/xgbe-selftest.c index 23b9d568a861..15c51e96bcdf 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-selftest.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-selftest.c @@ -173,6 +173,48 @@ static int xgbe_test_phy_loopback(struct xgbe_prv_data *pdata) return ret; } +static int xgbe_test_sph(struct xgbe_prv_data *pdata) +{ + struct net_packet_attrs attr = {}; + unsigned long cnt_end, cnt_start; + int ret; + + cnt_start = pdata->ext_stats.rx_split_header_packets; + + if (!pdata->sph) { + netdev_err(pdata->netdev, "Split Header not enabled\n"); + return -EOPNOTSUPP; + } + + /* UDP test */ + attr.dst = pdata->netdev->dev_addr; + attr.tcp = false; + + ret = __xgbe_test_loopback(pdata, &attr); + if (ret) + return ret; + + cnt_end = pdata->ext_stats.rx_split_header_packets; + if (cnt_end <= cnt_start) + return -EINVAL; + + /* TCP test */ + cnt_start = cnt_end; + + attr.dst = pdata->netdev->dev_addr; + attr.tcp = true; + + ret = __xgbe_test_loopback(pdata, &attr); + if (ret) + return ret; + + cnt_end = pdata->ext_stats.rx_split_header_packets; + if (cnt_end <= cnt_start) + return -EINVAL; + + return 0; +} + static const struct xgbe_test xgbe_selftests[] = { { .name = "MAC Loopback ", @@ -182,6 +224,10 @@ static const struct xgbe_test xgbe_selftests[] = { .name = "PHY Loopback ", .lb = XGBE_LOOPBACK_NONE, .fn = xgbe_test_phy_loopback, + }, { + .name = "Split Header ", + .lb = XGBE_LOOPBACK_PHY, + .fn = xgbe_test_sph, }, }; diff --git a/drivers/net/ethernet/amd/xgbe/xgbe.h b/drivers/net/ethernet/amd/xgbe/xgbe.h index dc03082c59aa..03ef0f548483 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe.h +++ b/drivers/net/ethernet/amd/xgbe/xgbe.h @@ -1246,6 +1246,7 @@ struct xgbe_prv_data { int rx_adapt_retries; bool rx_adapt_done; bool mode_set; + bool sph; }; /* Function prototypes*/ From 9c11b6b1abcd328136fc0cbc381734d6815d1c16 Mon Sep 17 00:00:00 2001 From: Raju Rangoju Date: Fri, 31 Oct 2025 16:45:59 +0530 Subject: [PATCH 0561/1381] amd-xgbe: add ethtool jumbo frame selftest Adds support for jumbo frame selftest. Works only for mtu size greater than 1500. Signed-off-by: Raju Rangoju Link: https://patch.msgid.link/20251031111555.774425-5-Raju.Rangoju@amd.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/amd/xgbe/xgbe-selftest.c | 23 +++++++++++++++++++ 1 file changed, 23 insertions(+) diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-selftest.c b/drivers/net/ethernet/amd/xgbe/xgbe-selftest.c index 15c51e96bcdf..55e5e467facd 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-selftest.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-selftest.c @@ -42,11 +42,19 @@ static int xgbe_test_loopback_validate(struct sk_buff *skb, struct tcphdr *th; struct udphdr *uh; struct iphdr *ih; + int eat; skb = skb_unshare(skb, GFP_ATOMIC); if (!skb) goto out; + eat = (skb->tail + skb->data_len) - skb->end; + if (eat > 0 && skb_shared(skb)) { + skb = skb_share_check(skb, GFP_ATOMIC); + if (!skb) + goto out; + } + if (skb_linearize(skb)) goto out; @@ -215,6 +223,17 @@ static int xgbe_test_sph(struct xgbe_prv_data *pdata) return 0; } +static int xgbe_test_jumbo(struct xgbe_prv_data *pdata) +{ + struct net_packet_attrs attr = {}; + int size = pdata->rx_buf_size; + + attr.dst = pdata->netdev->dev_addr; + attr.max_size = size - ETH_FCS_LEN; + + return __xgbe_test_loopback(pdata, &attr); +} + static const struct xgbe_test xgbe_selftests[] = { { .name = "MAC Loopback ", @@ -228,6 +247,10 @@ static const struct xgbe_test xgbe_selftests[] = { .name = "Split Header ", .lb = XGBE_LOOPBACK_PHY, .fn = xgbe_test_sph, + }, { + .name = "Jumbo Frame ", + .lb = XGBE_LOOPBACK_PHY, + .fn = xgbe_test_jumbo, }, }; From 0cc4b846159184892f4210c440daeb97cbe9583a Mon Sep 17 00:00:00 2001 From: Aswin Karuvally Date: Mon, 3 Nov 2025 11:16:52 +0100 Subject: [PATCH 0562/1381] s390/ctcm: Use info level for handshake UC_RCRESET CTC adapter throws CTC_EVENT_UC_RCRESET (Unit check remote reset event) during initial handshake, if the peer is not ready yet. This causes the ctcm driver to re-attempt the handshake. As it is normal for the event to occur during initialization, use info instead of warn level in kernel log and NOTICE instead of ERROR level in s390 debug feature. Also reword the log message for clarity. Reviewed-by: Alexandra Winter Signed-off-by: Aswin Karuvally Link: https://patch.msgid.link/20251103101652.2349855-1-aswin@linux.ibm.com Signed-off-by: Paolo Abeni --- drivers/s390/net/ctcm_fsms.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+) diff --git a/drivers/s390/net/ctcm_fsms.c b/drivers/s390/net/ctcm_fsms.c index 9678c6a2cda7..1a48258b63b2 100644 --- a/drivers/s390/net/ctcm_fsms.c +++ b/drivers/s390/net/ctcm_fsms.c @@ -882,6 +882,13 @@ static void ctcm_chx_rxiniterr(fsm_instance *fi, int event, void *arg) fsm_newstate(fi, CTC_STATE_RXERR); fsm_event(priv->fsm, DEV_EVENT_RXDOWN, dev); } + } else if (event == CTC_EVENT_UC_RCRESET) { + CTCM_DBF_TEXT_(TRACE, CTC_DBF_NOTICE, + "%s(%s): %s in %s", CTCM_FUNTAIL, ch->id, + ctc_ch_event_names[event], fsm_getstate_str(fi)); + + dev_info(&dev->dev, + "Init handshake not received, peer not ready yet\n"); } else { CTCM_DBF_TEXT_(ERROR, CTC_DBF_ERROR, "%s(%s): %s in %s", CTCM_FUNTAIL, ch->id, @@ -967,6 +974,13 @@ static void ctcm_chx_txiniterr(fsm_instance *fi, int event, void *arg) fsm_newstate(fi, CTC_STATE_TXERR); fsm_event(priv->fsm, DEV_EVENT_TXDOWN, dev); } + } else if (event == CTC_EVENT_UC_RCRESET) { + CTCM_DBF_TEXT_(TRACE, CTC_DBF_NOTICE, + "%s(%s): %s in %s", CTCM_FUNTAIL, ch->id, + ctc_ch_event_names[event], fsm_getstate_str(fi)); + + dev_info(&dev->dev, + "Init handshake not sent, peer not ready yet\n"); } else { CTCM_DBF_TEXT_(ERROR, CTC_DBF_ERROR, "%s(%s): %s in %s", CTCM_FUNTAIL, ch->id, From 13068e9d57264d0a86b8195817a01155ba33d230 Mon Sep 17 00:00:00 2001 From: Pavan Kumar Linga Date: Mon, 3 Nov 2025 14:46:30 -0800 Subject: [PATCH 0563/1381] idpf: add support for IDPF PCI programming interface At present IDPF supports only 0x1452 and 0x145C as PF and VF device IDs on our current generation hardware. Future hardware exposes a new set of device IDs for each generation. To avoid adding a new device ID for each generation and to make the driver forward and backward compatible, make use of the IDPF PCI programming interface to load the driver. Write and read the VF_ARQBAL mailbox register to find if the current device is a PF or a VF. PCI SIG allocated a new programming interface for the IDPF compliant ethernet network controller devices. It can be found at: https://members.pcisig.com/wg/PCI-SIG/document/20113 with the document titled as 'PCI Code and ID Assignment Revision 1.16' or any latest revisions. Tested this patch by doing a simple driver load/unload on Intel IPU E2000 hardware which supports 0x1452 and 0x145C device IDs and new hardware which supports the IDPF PCI programming interface. Reviewed-by: Sridhar Samudrala Reviewed-by: Simon Horman Signed-off-by: Pavan Kumar Linga Signed-off-by: Madhu Chittim Reviewed-by: Aleksandr Loktionov Tested-by: Marek Landowski Signed-off-by: Tony Nguyen Link: https://patch.msgid.link/20251103224631.595527-1-anthony.l.nguyen@intel.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/intel/idpf/idpf_main.c | 105 ++++++++++++++++---- 1 file changed, 88 insertions(+), 17 deletions(-) diff --git a/drivers/net/ethernet/intel/idpf/idpf_main.c b/drivers/net/ethernet/intel/idpf/idpf_main.c index 8c46481d2e1f..7a06eaf46a08 100644 --- a/drivers/net/ethernet/intel/idpf/idpf_main.c +++ b/drivers/net/ethernet/intel/idpf/idpf_main.c @@ -3,15 +3,93 @@ #include "idpf.h" #include "idpf_devids.h" +#include "idpf_lan_vf_regs.h" #include "idpf_virtchnl.h" #define DRV_SUMMARY "Intel(R) Infrastructure Data Path Function Linux Driver" +#define IDPF_NETWORK_ETHERNET_PROGIF 0x01 +#define IDPF_CLASS_NETWORK_ETHERNET_PROGIF \ + (PCI_CLASS_NETWORK_ETHERNET << 8 | IDPF_NETWORK_ETHERNET_PROGIF) +#define IDPF_VF_TEST_VAL 0xfeed0000u + MODULE_DESCRIPTION(DRV_SUMMARY); MODULE_IMPORT_NS("LIBETH"); MODULE_IMPORT_NS("LIBETH_XDP"); MODULE_LICENSE("GPL"); +/** + * idpf_get_device_type - Helper to find if it is a VF or PF device + * @pdev: PCI device information struct + * + * Return: PF/VF device ID or -%errno on failure. + */ +static int idpf_get_device_type(struct pci_dev *pdev) +{ + void __iomem *addr; + int ret; + + addr = ioremap(pci_resource_start(pdev, 0) + VF_ARQBAL, 4); + if (!addr) { + pci_err(pdev, "Failed to allocate BAR0 mbx region\n"); + return -EIO; + } + + writel(IDPF_VF_TEST_VAL, addr); + if (readl(addr) == IDPF_VF_TEST_VAL) + ret = IDPF_DEV_ID_VF; + else + ret = IDPF_DEV_ID_PF; + + iounmap(addr); + + return ret; +} + +/** + * idpf_dev_init - Initialize device specific parameters + * @adapter: adapter to initialize + * @ent: entry in idpf_pci_tbl + * + * Return: %0 on success, -%errno on failure. + */ +static int idpf_dev_init(struct idpf_adapter *adapter, + const struct pci_device_id *ent) +{ + int ret; + + if (ent->class == IDPF_CLASS_NETWORK_ETHERNET_PROGIF) { + ret = idpf_get_device_type(adapter->pdev); + switch (ret) { + case IDPF_DEV_ID_VF: + idpf_vf_dev_ops_init(adapter); + adapter->crc_enable = true; + break; + case IDPF_DEV_ID_PF: + idpf_dev_ops_init(adapter); + break; + default: + return ret; + } + + return 0; + } + + switch (ent->device) { + case IDPF_DEV_ID_PF: + idpf_dev_ops_init(adapter); + break; + case IDPF_DEV_ID_VF: + idpf_vf_dev_ops_init(adapter); + adapter->crc_enable = true; + break; + default: + return -ENODEV; + } + + return 0; +} + /** * idpf_remove - Device removal routine * @pdev: PCI device information struct @@ -165,21 +243,6 @@ static int idpf_probe(struct pci_dev *pdev, const struct pci_device_id *ent) adapter->req_tx_splitq = true; adapter->req_rx_splitq = true; - switch (ent->device) { - case IDPF_DEV_ID_PF: - idpf_dev_ops_init(adapter); - break; - case IDPF_DEV_ID_VF: - idpf_vf_dev_ops_init(adapter); - adapter->crc_enable = true; - break; - default: - err = -ENODEV; - dev_err(&pdev->dev, "Unexpected dev ID 0x%x in idpf probe\n", - ent->device); - goto err_free; - } - adapter->pdev = pdev; err = pcim_enable_device(pdev); if (err) @@ -259,11 +322,18 @@ static int idpf_probe(struct pci_dev *pdev, const struct pci_device_id *ent) /* setup msglvl */ adapter->msg_enable = netif_msg_init(-1, IDPF_AVAIL_NETIF_M); + err = idpf_dev_init(adapter, ent); + if (err) { + dev_err(&pdev->dev, "Unexpected dev ID 0x%x in idpf probe\n", + ent->device); + goto destroy_vc_event_wq; + } + err = idpf_cfg_hw(adapter); if (err) { dev_err(dev, "Failed to configure HW structure for adapter: %d\n", err); - goto err_cfg_hw; + goto destroy_vc_event_wq; } mutex_init(&adapter->vport_ctrl_lock); @@ -284,7 +354,7 @@ static int idpf_probe(struct pci_dev *pdev, const struct pci_device_id *ent) return 0; -err_cfg_hw: +destroy_vc_event_wq: destroy_workqueue(adapter->vc_event_wq); err_vc_event_wq_alloc: destroy_workqueue(adapter->stats_wq); @@ -304,6 +374,7 @@ static int idpf_probe(struct pci_dev *pdev, const struct pci_device_id *ent) static const struct pci_device_id idpf_pci_tbl[] = { { PCI_VDEVICE(INTEL, IDPF_DEV_ID_PF)}, { PCI_VDEVICE(INTEL, IDPF_DEV_ID_VF)}, + { PCI_DEVICE_CLASS(IDPF_CLASS_NETWORK_ETHERNET_PROGIF, ~0)}, { /* Sentinel */ } }; MODULE_DEVICE_TABLE(pci, idpf_pci_tbl); From bd6ec8111e6546d01f16bb113aee1fcada6e86a9 Mon Sep 17 00:00:00 2001 From: Manish Dharanenthiran Date: Fri, 31 Oct 2025 08:37:45 +0530 Subject: [PATCH 0564/1381] wifi: ath12k: Make firmware stats reset caller-driven Currently, ath12k_fw_stats_reset() is called in ath12k_mac_get_fw_stats() before fetching the required stats from the firmware. However, ath12k_open_bcn_stats() requests firmware stats for each enabled BSS individually. Since the firmware stats are reset before fetching, only the last BSS's data is displayed. Also, in ath12k_mac_op_get_txpower(), ath12k_mac_op_sta_statistics(), and ath12k_mac_op_link_sta_statistics(), after getting the stats from the firmware, the reset function is not called until the next firmware stats are requested or while unloading the module. Hence, the stats buffer will not be freed until one of the above sequences is executed. However, in ath12k_open_vdev_stats(), ath12k_open_bcn_stats() and ath12k_open_pdev_stats(), firmware stats are reset after copying the necessary data in ath12k_wmi_fw_stats_dump(). This leads to inconsistent usage of ath12k_fw_stats_reset() for freeing the firmware stats. Avoid these discrepancies by making it the caller's responsibility to free the stats buffer, thereby removing the need to free the stats buffer in ath12k_mac_get_fw_stats() and ath12k_wmi_fw_stats_dump(). Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.4.1-00199-QCAHKSWPL_SILICONZ-1 Signed-off-by: Manish Dharanenthiran Reviewed-by: Vasanthakumar Thiagarajan Reviewed-by: Baochen Qiang Link: https://patch.msgid.link/20251031-beacon_stats-v1-1-f52fce7b03ac@qti.qualcomm.com Signed-off-by: Jeff Johnson --- drivers/net/wireless/ath/ath12k/debugfs.c | 9 +++------ drivers/net/wireless/ath/ath12k/mac.c | 15 ++++++++++----- drivers/net/wireless/ath/ath12k/wmi.c | 2 -- 3 files changed, 13 insertions(+), 13 deletions(-) diff --git a/drivers/net/wireless/ath/ath12k/debugfs.c b/drivers/net/wireless/ath/ath12k/debugfs.c index 15219429d4ed..d6a86f075d73 100644 --- a/drivers/net/wireless/ath/ath12k/debugfs.c +++ b/drivers/net/wireless/ath/ath12k/debugfs.c @@ -1286,6 +1286,7 @@ static int ath12k_open_vdev_stats(struct inode *inode, struct file *file) ath12k_wmi_fw_stats_dump(ar, &ar->fw_stats, param.stats_id, buf); + ath12k_fw_stats_reset(ar); file->private_data = no_free_ptr(buf); @@ -1352,12 +1353,7 @@ static int ath12k_open_bcn_stats(struct inode *inode, struct file *file) ath12k_wmi_fw_stats_dump(ar, &ar->fw_stats, param.stats_id, buf); - /* since beacon stats request is looped for all active VDEVs, saved fw - * stats is not freed for each request until done for all active VDEVs - */ - spin_lock_bh(&ar->data_lock); - ath12k_fw_stats_bcn_free(&ar->fw_stats.bcn); - spin_unlock_bh(&ar->data_lock); + ath12k_fw_stats_reset(ar); file->private_data = no_free_ptr(buf); @@ -1418,6 +1414,7 @@ static int ath12k_open_pdev_stats(struct inode *inode, struct file *file) ath12k_wmi_fw_stats_dump(ar, &ar->fw_stats, param.stats_id, buf); + ath12k_fw_stats_reset(ar); file->private_data = no_free_ptr(buf); diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c index 0348601ecc50..65e0ac19ad11 100644 --- a/drivers/net/wireless/ath/ath12k/mac.c +++ b/drivers/net/wireless/ath/ath12k/mac.c @@ -5091,8 +5091,6 @@ int ath12k_mac_get_fw_stats(struct ath12k *ar, if (ah->state != ATH12K_HW_STATE_ON) return -ENETDOWN; - ath12k_fw_stats_reset(ar); - reinit_completion(&ar->fw_stats_complete); reinit_completion(&ar->fw_stats_done); @@ -5190,6 +5188,7 @@ static int ath12k_mac_op_get_txpower(struct ieee80211_hw *hw, ar->chan_tx_pwr = pdev->chan_tx_power / 2; spin_unlock_bh(&ar->data_lock); ar->last_tx_power_update = jiffies; + ath12k_fw_stats_reset(ar); send_tx_power: *dbm = ar->chan_tx_pwr; @@ -13220,14 +13219,18 @@ static void ath12k_mac_op_sta_statistics(struct ieee80211_hw *hw, if (!signal && ahsta->ahvif->vdev_type == WMI_VDEV_TYPE_STA && - !(ath12k_mac_get_fw_stats(ar, ¶ms))) + !(ath12k_mac_get_fw_stats(ar, ¶ms))) { signal = arsta->rssi_beacon; + ath12k_fw_stats_reset(ar); + } params.stats_id = WMI_REQUEST_RSSI_PER_CHAIN_STAT; if (!(sinfo->filled & BIT_ULL(NL80211_STA_INFO_CHAIN_SIGNAL)) && ahsta->ahvif->vdev_type == WMI_VDEV_TYPE_STA && - !(ath12k_mac_get_fw_stats(ar, ¶ms))) + !(ath12k_mac_get_fw_stats(ar, ¶ms))) { ath12k_mac_put_chain_rssi(sinfo, arsta); + ath12k_fw_stats_reset(ar); + } spin_lock_bh(&ar->data_lock); noise_floor = ath12k_pdev_get_noise_floor(ar); @@ -13311,8 +13314,10 @@ static void ath12k_mac_op_link_sta_statistics(struct ieee80211_hw *hw, if (!signal && ahsta->ahvif->vdev_type == WMI_VDEV_TYPE_STA && - !(ath12k_mac_get_fw_stats(ar, ¶ms))) + !(ath12k_mac_get_fw_stats(ar, ¶ms))) { signal = arsta->rssi_beacon; + ath12k_fw_stats_reset(ar); + } if (signal) { link_sinfo->signal = diff --git a/drivers/net/wireless/ath/ath12k/wmi.c b/drivers/net/wireless/ath/ath12k/wmi.c index 5075d86df36f..f812da24b21e 100644 --- a/drivers/net/wireless/ath/ath12k/wmi.c +++ b/drivers/net/wireless/ath/ath12k/wmi.c @@ -8089,8 +8089,6 @@ void ath12k_wmi_fw_stats_dump(struct ath12k *ar, buf[len - 1] = 0; else buf[len] = 0; - - ath12k_fw_stats_reset(ar); } static void From 2977567b244f056d86658160659f06cd6c78ba3d Mon Sep 17 00:00:00 2001 From: Manish Dharanenthiran Date: Fri, 31 Oct 2025 08:37:46 +0530 Subject: [PATCH 0565/1381] wifi: ath12k: Fix timeout error during beacon stats retrieval Currently, for beacon_stats, ath12k_mac_get_fw_stats() is called for each started BSS on the specified hardware. ath12k_mac_get_fw_stats() will wait for the fw_stats_done completion after fetching the requested data from firmware. For the beacon_stats, fw_stats_done completion will be set only when stats are received for all BSSes. However, for other stats like vdev_stats or pdev_stats, there is one request to the firmware for all enabled BSSes. Since beacon_stats is fetched individually for all BSSes enabled in that pdev, waiting for the completion event results in a timeout error when multiple BSSes are enabled. Avoid this by completing the fw_stats_done immediately after updating the requested BSS's beacon stats in the list. Subsequently, this list will be used to display the beacon stats for all enabled BSSes in the requested pdev. Additionally, remove 'num_bcn_recvd' from the ath12k_fw_stats struct as it is no longer needed. Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.4.1-00199-QCAHKSWPL_SILICONZ-1 Fixes: 9fe4669ae919 ("wifi: ath12k: Request beacon stats from firmware") Signed-off-by: Manish Dharanenthiran Reviewed-by: Vasanthakumar Thiagarajan Reviewed-by: Baochen Qiang Link: https://patch.msgid.link/20251031-beacon_stats-v1-2-f52fce7b03ac@qti.qualcomm.com Signed-off-by: Jeff Johnson --- drivers/net/wireless/ath/ath12k/core.c | 2 -- drivers/net/wireless/ath/ath12k/core.h | 1 - drivers/net/wireless/ath/ath12k/wmi.c | 10 +--------- 3 files changed, 1 insertion(+), 12 deletions(-) diff --git a/drivers/net/wireless/ath/ath12k/core.c b/drivers/net/wireless/ath/ath12k/core.c index a2137b363c2f..cc352eef1939 100644 --- a/drivers/net/wireless/ath/ath12k/core.c +++ b/drivers/net/wireless/ath/ath12k/core.c @@ -1,7 +1,6 @@ // SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (c) 2018-2021 The Linux Foundation. All rights reserved. - * Copyright (c) 2021-2025 Qualcomm Innovation Center, Inc. All rights reserved. * Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. */ @@ -1250,7 +1249,6 @@ void ath12k_fw_stats_reset(struct ath12k *ar) spin_lock_bh(&ar->data_lock); ath12k_fw_stats_free(&ar->fw_stats); ar->fw_stats.num_vdev_recvd = 0; - ar->fw_stats.num_bcn_recvd = 0; spin_unlock_bh(&ar->data_lock); } diff --git a/drivers/net/wireless/ath/ath12k/core.h b/drivers/net/wireless/ath/ath12k/core.h index 41da0efaa854..3c1e0069be1e 100644 --- a/drivers/net/wireless/ath/ath12k/core.h +++ b/drivers/net/wireless/ath/ath12k/core.h @@ -646,7 +646,6 @@ struct ath12k_fw_stats { struct list_head vdevs; struct list_head bcn; u32 num_vdev_recvd; - u32 num_bcn_recvd; }; struct ath12k_dbg_htt_stats { diff --git a/drivers/net/wireless/ath/ath12k/wmi.c b/drivers/net/wireless/ath/ath12k/wmi.c index f812da24b21e..be8b2943094f 100644 --- a/drivers/net/wireless/ath/ath12k/wmi.c +++ b/drivers/net/wireless/ath/ath12k/wmi.c @@ -8485,18 +8485,10 @@ static void ath12k_wmi_fw_stats_process(struct ath12k *ar, ath12k_warn(ab, "empty beacon stats"); return; } - /* Mark end until we reached the count of all started VDEVs - * within the PDEV - */ - if (ar->num_started_vdevs) - is_end = ((++ar->fw_stats.num_bcn_recvd) == - ar->num_started_vdevs); list_splice_tail_init(&stats->bcn, &ar->fw_stats.bcn); - - if (is_end) - complete(&ar->fw_stats_done); + complete(&ar->fw_stats_done); } } From 9352d40c8bcd2ef29366d2c38b163c0b115039ed Mon Sep 17 00:00:00 2001 From: Mohammad Heib Date: Sat, 25 Oct 2025 16:08:58 +0300 Subject: [PATCH 0566/1381] devlink: Add new "max_mac_per_vf" generic device param Add a new device generic parameter to controls the maximum number of MAC filters allowed per VF. For example, to limit a VF to 3 MAC addresses: $ devlink dev param set pci/0000:3b:00.0 name max_mac_per_vf \ value 3 \ cmode runtime Signed-off-by: Mohammad Heib Reviewed-by: Simon Horman Signed-off-by: Jacob Keller Signed-off-by: Tony Nguyen --- Documentation/networking/devlink/devlink-params.rst | 4 ++++ include/net/devlink.h | 4 ++++ net/devlink/param.c | 5 +++++ 3 files changed, 13 insertions(+) diff --git a/Documentation/networking/devlink/devlink-params.rst b/Documentation/networking/devlink/devlink-params.rst index 0a9c20d70122..c0597d456641 100644 --- a/Documentation/networking/devlink/devlink-params.rst +++ b/Documentation/networking/devlink/devlink-params.rst @@ -151,3 +151,7 @@ own name. * - ``num_doorbells`` - u32 - Controls the number of doorbells used by the device. + * - ``max_mac_per_vf`` + - u32 + - Controls the maximum number of MAC address filters that can be assigned + to a Virtual Function (VF). diff --git a/include/net/devlink.h b/include/net/devlink.h index 9e824f61e40f..d01046ef0577 100644 --- a/include/net/devlink.h +++ b/include/net/devlink.h @@ -532,6 +532,7 @@ enum devlink_param_generic_id { DEVLINK_PARAM_GENERIC_ID_CLOCK_ID, DEVLINK_PARAM_GENERIC_ID_TOTAL_VFS, DEVLINK_PARAM_GENERIC_ID_NUM_DOORBELLS, + DEVLINK_PARAM_GENERIC_ID_MAX_MAC_PER_VF, /* add new param generic ids above here*/ __DEVLINK_PARAM_GENERIC_ID_MAX, @@ -602,6 +603,9 @@ enum devlink_param_generic_id { #define DEVLINK_PARAM_GENERIC_NUM_DOORBELLS_NAME "num_doorbells" #define DEVLINK_PARAM_GENERIC_NUM_DOORBELLS_TYPE DEVLINK_PARAM_TYPE_U32 +#define DEVLINK_PARAM_GENERIC_MAX_MAC_PER_VF_NAME "max_mac_per_vf" +#define DEVLINK_PARAM_GENERIC_MAX_MAC_PER_VF_TYPE DEVLINK_PARAM_TYPE_U32 + #define DEVLINK_PARAM_GENERIC(_id, _cmodes, _get, _set, _validate) \ { \ .id = DEVLINK_PARAM_GENERIC_ID_##_id, \ diff --git a/net/devlink/param.c b/net/devlink/param.c index 70e69523412c..6b233b13b69a 100644 --- a/net/devlink/param.c +++ b/net/devlink/param.c @@ -112,6 +112,11 @@ static const struct devlink_param devlink_param_generic[] = { .name = DEVLINK_PARAM_GENERIC_NUM_DOORBELLS_NAME, .type = DEVLINK_PARAM_GENERIC_NUM_DOORBELLS_TYPE, }, + { + .id = DEVLINK_PARAM_GENERIC_ID_MAX_MAC_PER_VF, + .name = DEVLINK_PARAM_GENERIC_MAX_MAC_PER_VF_NAME, + .type = DEVLINK_PARAM_GENERIC_MAX_MAC_PER_VF_TYPE, + }, }; static int devlink_param_generic_verify(const struct devlink_param *param) From 2c031d4c772f3a9191d04d57a3403ad6a56375c7 Mon Sep 17 00:00:00 2001 From: Mohammad Heib Date: Sat, 25 Oct 2025 16:08:59 +0300 Subject: [PATCH 0567/1381] i40e: support generic devlink param "max_mac_per_vf" Currently the i40e driver enforces its own internally calculated per-VF MAC filter limit, derived from the number of allocated VFs and available hardware resources. This limit is not configurable by the administrator, which makes it difficult to control how many MAC addresses each VF may use. This patch adds support for the new generic devlink runtime parameter "max_mac_per_vf" which provides administrators with a way to cap the number of MAC addresses a VF can use: - When the parameter is set to 0 (default), the driver continues to use its internally calculated limit. - When set to a non-zero value, the driver applies this value as a strict cap for VFs, overriding the internal calculation. Important notes: - The configured value is a theoretical maximum. Hardware limits may still prevent additional MAC addresses from being added, even if the parameter allows it. - Since MAC filters are a shared hardware resource across all VFs, setting a high value may cause resource contention and starve other VFs. - This change gives administrators predictable and flexible control over VF resource allocation, while still respecting hardware limitations. - Previous discussion about this change: https://lore.kernel.org/netdev/20250805134042.2604897-2-dhill@redhat.com https://lore.kernel.org/netdev/20250823094952.182181-1-mheib@redhat.com Signed-off-by: Mohammad Heib Reviewed-by: Jacob Keller Reviewed-by: Aleksandr Loktionov Reviewed-by: Simon Horman Tested-by: Rafal Romanowski Signed-off-by: Jacob Keller Signed-off-by: Tony Nguyen --- Documentation/networking/devlink/i40e.rst | 34 ++++++++++++ drivers/net/ethernet/intel/i40e/i40e.h | 4 ++ .../net/ethernet/intel/i40e/i40e_devlink.c | 54 ++++++++++++++++++- .../ethernet/intel/i40e/i40e_virtchnl_pf.c | 31 ++++++++--- 4 files changed, 113 insertions(+), 10 deletions(-) diff --git a/Documentation/networking/devlink/i40e.rst b/Documentation/networking/devlink/i40e.rst index d3cb5bb5197e..51c887f0dc83 100644 --- a/Documentation/networking/devlink/i40e.rst +++ b/Documentation/networking/devlink/i40e.rst @@ -7,6 +7,40 @@ i40e devlink support This document describes the devlink features implemented by the ``i40e`` device driver. +Parameters +========== + +.. list-table:: Generic parameters implemented + :widths: 5 5 90 + + * - Name + - Mode + - Notes + * - ``max_mac_per_vf`` + - runtime + - Controls the maximum number of MAC addresses a VF can use + on i40e devices. + + By default (``0``), the driver enforces its internally calculated per-VF + MAC filter limit, which is based on the number of allocated VFS. + + If set to a non-zero value, this parameter acts as a strict cap: + the driver will use the user-provided value instead of its internal + calculation. + + **Important notes:** + + - This value **must be set before enabling SR-IOV**. + Attempting to change it while SR-IOV is enabled will return an error. + - MAC filters are a **shared hardware resource** across all VFs. + Setting a high value may cause other VFs to be starved of filters. + - This value is a **Administrative policy**. The hardware may return + errors when its absolute limit is reached, regardless of the value + set here. + + The default value is ``0`` (internal calculation is used). + + Info versions ============= diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h index 801a57a925da..d2d03db2acec 100644 --- a/drivers/net/ethernet/intel/i40e/i40e.h +++ b/drivers/net/ethernet/intel/i40e/i40e.h @@ -574,6 +574,10 @@ struct i40e_pf { struct i40e_vf *vf; int num_alloc_vfs; /* actual number of VFs allocated */ u32 vf_aq_requests; + /* If set to non-zero, the device uses this value + * as maximum number of MAC filters per VF. + */ + u32 max_mac_per_vf; u32 arq_overflows; /* Not fatal, possibly indicative of problems */ struct ratelimit_state mdd_message_rate_limit; /* DCBx/DCBNL capability for PF that indicates diff --git a/drivers/net/ethernet/intel/i40e/i40e_devlink.c b/drivers/net/ethernet/intel/i40e/i40e_devlink.c index cc4e9e2addb7..bc205e3077c7 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_devlink.c +++ b/drivers/net/ethernet/intel/i40e/i40e_devlink.c @@ -5,6 +5,41 @@ #include "i40e.h" #include "i40e_devlink.h" +static int i40e_max_mac_per_vf_set(struct devlink *devlink, + u32 id, + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) +{ + struct i40e_pf *pf = devlink_priv(devlink); + + if (pf->num_alloc_vfs > 0) { + NL_SET_ERR_MSG_MOD(extack, + "Cannot change max_mac_per_vf while SR-IOV is enabled"); + return -EBUSY; + } + + pf->max_mac_per_vf = ctx->val.vu32; + return 0; +} + +static int i40e_max_mac_per_vf_get(struct devlink *devlink, + u32 id, + struct devlink_param_gset_ctx *ctx) +{ + struct i40e_pf *pf = devlink_priv(devlink); + + ctx->val.vu32 = pf->max_mac_per_vf; + return 0; +} + +static const struct devlink_param i40e_dl_params[] = { + DEVLINK_PARAM_GENERIC(MAX_MAC_PER_VF, + BIT(DEVLINK_PARAM_CMODE_RUNTIME), + i40e_max_mac_per_vf_get, + i40e_max_mac_per_vf_set, + NULL), +}; + static void i40e_info_get_dsn(struct i40e_pf *pf, char *buf, size_t len) { u8 dsn[8]; @@ -165,7 +200,18 @@ void i40e_free_pf(struct i40e_pf *pf) **/ void i40e_devlink_register(struct i40e_pf *pf) { - devlink_register(priv_to_devlink(pf)); + struct devlink *dl = priv_to_devlink(pf); + struct device *dev = &pf->pdev->dev; + int err; + + err = devlink_params_register(dl, i40e_dl_params, + ARRAY_SIZE(i40e_dl_params)); + if (err) + dev_err(dev, + "devlink params register failed with error %d", err); + + devlink_register(dl); + } /** @@ -176,7 +222,11 @@ void i40e_devlink_register(struct i40e_pf *pf) **/ void i40e_devlink_unregister(struct i40e_pf *pf) { - devlink_unregister(priv_to_devlink(pf)); + struct devlink *dl = priv_to_devlink(pf); + + devlink_unregister(dl); + devlink_params_unregister(dl, i40e_dl_params, + ARRAY_SIZE(i40e_dl_params)); } /** diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c index 0fe0d52c796b..9d91a382612d 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c +++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c @@ -2935,33 +2935,48 @@ static inline int i40e_check_vf_permission(struct i40e_vf *vf, if (!f) ++mac_add_cnt; } - - /* If this VF is not privileged, then we can't add more than a limited - * number of addresses. + /* Determine the maximum number of MAC addresses this VF may use. * - * If this VF is trusted, it can use more resources than untrusted. - * However to ensure that every trusted VF has appropriate number of - * resources, divide whole pool of resources per port and then across - * all VFs. + * - For untrusted VFs: use a fixed small limit. + * + * - For trusted VFs: limit is calculated by dividing total MAC + * filter pool across all VFs/ports. + * + * - User can override this by devlink param "max_mac_per_vf". + * If set its value is used as a strict cap for both trusted and + * untrusted VFs. + * Note: + * even when overridden, this is a theoretical maximum; hardware + * may reject additional MACs if the absolute HW limit is reached. */ if (!vf_trusted) mac_add_max = I40E_VC_MAX_MAC_ADDR_PER_VF; else mac_add_max = I40E_VC_MAX_MACVLAN_PER_TRUSTED_VF(pf->num_alloc_vfs, hw->num_ports); + if (pf->max_mac_per_vf > 0) + mac_add_max = pf->max_mac_per_vf; + /* VF can replace all its filters in one step, in this case mac_add_max * will be added as active and another mac_add_max will be in * a to-be-removed state. Account for that. */ if ((i40e_count_active_filters(vsi) + mac_add_cnt) > mac_add_max || (i40e_count_all_filters(vsi) + mac_add_cnt) > 2 * mac_add_max) { + if (pf->max_mac_per_vf == mac_add_max && mac_add_max > 0) { + dev_err(&pf->pdev->dev, + "Cannot add more MAC addresses: VF reached its maximum allowed limit (%d)\n", + mac_add_max); + return -EPERM; + } if (!vf_trusted) { dev_err(&pf->pdev->dev, "Cannot add more MAC addresses, VF is not trusted, switch the VF to trusted to add more functionality\n"); return -EPERM; } else { dev_err(&pf->pdev->dev, - "Cannot add more MAC addresses, trusted VF exhausted it's resources\n"); + "Cannot add more MAC addresses: trusted VF reached its maximum allowed limit (%d)\n", + mac_add_max); return -EPERM; } } From 322a1e6f3d687d3b116837036299abeb4d79ba76 Mon Sep 17 00:00:00 2001 From: Daniel Golle Date: Mon, 3 Nov 2025 12:18:41 +0000 Subject: [PATCH 0568/1381] net: dsa: lantiq_gswip: split into common and MMIO parts Move all parts specific for the MMIO/SoC driver into a module of its own to prepare for supporting MDIO-connected switch ICs. Modify gswip_probe() functions by splitting it into a common function gswip_probe_common() which covers allocating, initializing and registering the DSA switch, while keeping transport-specific regmap initialization as well as PHY firmware loading in the new MMIO/SoC-specific gswip_probe() function. Signed-off-by: Daniel Golle Tested-by: Alexander Sverdlin Reviewed-by: Vladimir Oltean Link: https://patch.msgid.link/dc7da5b65ec220ba8e9bc4bd04fe1ed7de046656.1762170107.git.daniel@makrotopia.org Signed-off-by: Jakub Kicinski --- drivers/net/dsa/lantiq/Kconfig | 6 +- drivers/net/dsa/lantiq/Makefile | 1 + drivers/net/dsa/lantiq/lantiq_gswip.c | 1591 +---------------- drivers/net/dsa/lantiq/lantiq_gswip.h | 4 + drivers/net/dsa/lantiq/lantiq_gswip_common.c | 1622 ++++++++++++++++++ 5 files changed, 1645 insertions(+), 1579 deletions(-) create mode 100644 drivers/net/dsa/lantiq/lantiq_gswip_common.c diff --git a/drivers/net/dsa/lantiq/Kconfig b/drivers/net/dsa/lantiq/Kconfig index 3cfa16840cf5..78db82a47d09 100644 --- a/drivers/net/dsa/lantiq/Kconfig +++ b/drivers/net/dsa/lantiq/Kconfig @@ -1,8 +1,12 @@ +config NET_DSA_LANTIQ_COMMON + tristate + select REGMAP + config NET_DSA_LANTIQ_GSWIP tristate "Lantiq / Intel GSWIP" depends on HAS_IOMEM select NET_DSA_TAG_GSWIP - select REGMAP + select NET_DSA_LANTIQ_COMMON help This enables support for the Lantiq / Intel GSWIP 2.1 found in the xrx200 / VR9 SoC. diff --git a/drivers/net/dsa/lantiq/Makefile b/drivers/net/dsa/lantiq/Makefile index 849f85ebebd6..65ffa7bb71aa 100644 --- a/drivers/net/dsa/lantiq/Makefile +++ b/drivers/net/dsa/lantiq/Makefile @@ -1 +1,2 @@ obj-$(CONFIG_NET_DSA_LANTIQ_GSWIP) += lantiq_gswip.o +obj-$(CONFIG_NET_DSA_LANTIQ_COMMON) += lantiq_gswip_common.o diff --git a/drivers/net/dsa/lantiq/lantiq_gswip.c b/drivers/net/dsa/lantiq/lantiq_gswip.c index 38f7f6352e8d..57dd063c0740 100644 --- a/drivers/net/dsa/lantiq/lantiq_gswip.c +++ b/drivers/net/dsa/lantiq/lantiq_gswip.c @@ -2,1227 +2,33 @@ /* * Lantiq / Intel GSWIP switch driver for VRX200, xRX300 and xRX330 SoCs * - * Copyright (C) 2010 Lantiq Deutschland - * Copyright (C) 2012 John Crispin + * Copyright (C) 2025 Daniel Golle * Copyright (C) 2017 - 2019 Hauke Mehrtens - * - * The VLAN and bridge model the GSWIP hardware uses does not directly - * matches the model DSA uses. - * - * The hardware has 64 possible table entries for bridges with one VLAN - * ID, one flow id and a list of ports for each bridge. All entries which - * match the same flow ID are combined in the mac learning table, they - * act as one global bridge. - * The hardware does not support VLAN filter on the port, but on the - * bridge, this driver converts the DSA model to the hardware. - * - * The CPU gets all the exception frames which do not match any forwarding - * rule and the CPU port is also added to all bridges. This makes it possible - * to handle all the special cases easily in software. - * At the initialization the driver allocates one bridge table entry for - * each switch port which is used when the port is used without an - * explicit bridge. This prevents the frames from being forwarded - * between all LAN ports by default. + * Copyright (C) 2012 John Crispin + * Copyright (C) 2010 Lantiq Deutschland */ #include "lantiq_gswip.h" #include "lantiq_pce.h" +#include #include -#include #include -#include -#include -#include #include #include -#include -#include #include -#include -#include +#include +#include +#include #include +#include + struct xway_gphy_match_data { char *fe_firmware_name; char *ge_firmware_name; }; -struct gswip_pce_table_entry { - u16 index; // PCE_TBL_ADDR.ADDR = pData->table_index - u16 table; // PCE_TBL_CTRL.ADDR = pData->table - u16 key[8]; - u16 val[5]; - u16 mask; - u8 gmap; - bool type; - bool valid; - bool key_mode; -}; - -struct gswip_rmon_cnt_desc { - unsigned int size; - unsigned int offset; - const char *name; -}; - -#define MIB_DESC(_size, _offset, _name) {.size = _size, .offset = _offset, .name = _name} - -static const struct gswip_rmon_cnt_desc gswip_rmon_cnt[] = { - /** Receive Packet Count (only packets that are accepted and not discarded). */ - MIB_DESC(1, 0x1F, "RxGoodPkts"), - MIB_DESC(1, 0x23, "RxUnicastPkts"), - MIB_DESC(1, 0x22, "RxMulticastPkts"), - MIB_DESC(1, 0x21, "RxFCSErrorPkts"), - MIB_DESC(1, 0x1D, "RxUnderSizeGoodPkts"), - MIB_DESC(1, 0x1E, "RxUnderSizeErrorPkts"), - MIB_DESC(1, 0x1B, "RxOversizeGoodPkts"), - MIB_DESC(1, 0x1C, "RxOversizeErrorPkts"), - MIB_DESC(1, 0x20, "RxGoodPausePkts"), - MIB_DESC(1, 0x1A, "RxAlignErrorPkts"), - MIB_DESC(1, 0x12, "Rx64BytePkts"), - MIB_DESC(1, 0x13, "Rx127BytePkts"), - MIB_DESC(1, 0x14, "Rx255BytePkts"), - MIB_DESC(1, 0x15, "Rx511BytePkts"), - MIB_DESC(1, 0x16, "Rx1023BytePkts"), - /** Receive Size 1024-1522 (or more, if configured) Packet Count. */ - MIB_DESC(1, 0x17, "RxMaxBytePkts"), - MIB_DESC(1, 0x18, "RxDroppedPkts"), - MIB_DESC(1, 0x19, "RxFilteredPkts"), - MIB_DESC(2, 0x24, "RxGoodBytes"), - MIB_DESC(2, 0x26, "RxBadBytes"), - MIB_DESC(1, 0x11, "TxAcmDroppedPkts"), - MIB_DESC(1, 0x0C, "TxGoodPkts"), - MIB_DESC(1, 0x06, "TxUnicastPkts"), - MIB_DESC(1, 0x07, "TxMulticastPkts"), - MIB_DESC(1, 0x00, "Tx64BytePkts"), - MIB_DESC(1, 0x01, "Tx127BytePkts"), - MIB_DESC(1, 0x02, "Tx255BytePkts"), - MIB_DESC(1, 0x03, "Tx511BytePkts"), - MIB_DESC(1, 0x04, "Tx1023BytePkts"), - /** Transmit Size 1024-1522 (or more, if configured) Packet Count. */ - MIB_DESC(1, 0x05, "TxMaxBytePkts"), - MIB_DESC(1, 0x08, "TxSingleCollCount"), - MIB_DESC(1, 0x09, "TxMultCollCount"), - MIB_DESC(1, 0x0A, "TxLateCollCount"), - MIB_DESC(1, 0x0B, "TxExcessCollCount"), - MIB_DESC(1, 0x0D, "TxPauseCount"), - MIB_DESC(1, 0x10, "TxDroppedPkts"), - MIB_DESC(2, 0x0E, "TxGoodBytes"), -}; - -static u32 gswip_switch_r_timeout(struct gswip_priv *priv, u32 offset, - u32 cleared) -{ - u32 val; - - return regmap_read_poll_timeout(priv->gswip, offset, val, - !(val & cleared), 20, 50000); -} - -static void gswip_mii_mask_cfg(struct gswip_priv *priv, u32 mask, u32 set, - int port) -{ - int reg_port; - - /* MII_CFG register only exists for MII ports */ - if (!(priv->hw_info->mii_ports & BIT(port))) - return; - - reg_port = port + priv->hw_info->mii_port_reg_offset; - - regmap_write_bits(priv->mii, GSWIP_MII_CFGp(reg_port), mask, - set); -} - -static void gswip_mii_mask_pcdu(struct gswip_priv *priv, u32 mask, u32 set, - int port) -{ - int reg_port; - - /* MII_PCDU register only exists for MII ports */ - if (!(priv->hw_info->mii_ports & BIT(port))) - return; - - reg_port = port + priv->hw_info->mii_port_reg_offset; - - switch (reg_port) { - case 0: - regmap_write_bits(priv->mii, GSWIP_MII_PCDU0, mask, set); - break; - case 1: - regmap_write_bits(priv->mii, GSWIP_MII_PCDU1, mask, set); - break; - case 5: - regmap_write_bits(priv->mii, GSWIP_MII_PCDU5, mask, set); - break; - } -} - -static int gswip_mdio_poll(struct gswip_priv *priv) -{ - u32 ctrl; - - return regmap_read_poll_timeout(priv->mdio, GSWIP_MDIO_CTRL, ctrl, - !(ctrl & GSWIP_MDIO_CTRL_BUSY), 40, 4000); -} - -static int gswip_mdio_wr(struct mii_bus *bus, int addr, int reg, u16 val) -{ - struct gswip_priv *priv = bus->priv; - int err; - - err = gswip_mdio_poll(priv); - if (err) { - dev_err(&bus->dev, "waiting for MDIO bus busy timed out\n"); - return err; - } - - regmap_write(priv->mdio, GSWIP_MDIO_WRITE, val); - regmap_write(priv->mdio, GSWIP_MDIO_CTRL, - GSWIP_MDIO_CTRL_BUSY | GSWIP_MDIO_CTRL_WR | - ((addr & GSWIP_MDIO_CTRL_PHYAD_MASK) << GSWIP_MDIO_CTRL_PHYAD_SHIFT) | - (reg & GSWIP_MDIO_CTRL_REGAD_MASK)); - - return 0; -} - -static int gswip_mdio_rd(struct mii_bus *bus, int addr, int reg) -{ - struct gswip_priv *priv = bus->priv; - u32 val; - int err; - - err = gswip_mdio_poll(priv); - if (err) { - dev_err(&bus->dev, "waiting for MDIO bus busy timed out\n"); - return err; - } - - regmap_write(priv->mdio, GSWIP_MDIO_CTRL, - GSWIP_MDIO_CTRL_BUSY | GSWIP_MDIO_CTRL_RD | - ((addr & GSWIP_MDIO_CTRL_PHYAD_MASK) << GSWIP_MDIO_CTRL_PHYAD_SHIFT) | - (reg & GSWIP_MDIO_CTRL_REGAD_MASK)); - - err = gswip_mdio_poll(priv); - if (err) { - dev_err(&bus->dev, "waiting for MDIO bus busy timed out\n"); - return err; - } - - err = regmap_read(priv->mdio, GSWIP_MDIO_READ, &val); - if (err) - return err; - - return val; -} - -static int gswip_mdio(struct gswip_priv *priv) -{ - struct device_node *mdio_np, *switch_np = priv->dev->of_node; - struct device *dev = priv->dev; - struct mii_bus *bus; - int err = 0; - - mdio_np = of_get_compatible_child(switch_np, "lantiq,xrx200-mdio"); - if (!mdio_np) - mdio_np = of_get_child_by_name(switch_np, "mdio"); - - if (!of_device_is_available(mdio_np)) - goto out_put_node; - - bus = devm_mdiobus_alloc(dev); - if (!bus) { - err = -ENOMEM; - goto out_put_node; - } - - bus->priv = priv; - bus->read = gswip_mdio_rd; - bus->write = gswip_mdio_wr; - bus->name = "lantiq,xrx200-mdio"; - snprintf(bus->id, MII_BUS_ID_SIZE, "%s-mii", dev_name(priv->dev)); - bus->parent = priv->dev; - - err = devm_of_mdiobus_register(dev, bus, mdio_np); - -out_put_node: - of_node_put(mdio_np); - - return err; -} - -static int gswip_pce_table_entry_read(struct gswip_priv *priv, - struct gswip_pce_table_entry *tbl) -{ - int i; - int err; - u32 crtl; - u32 tmp; - u16 addr_mode = tbl->key_mode ? GSWIP_PCE_TBL_CTRL_OPMOD_KSRD : - GSWIP_PCE_TBL_CTRL_OPMOD_ADRD; - - mutex_lock(&priv->pce_table_lock); - - err = gswip_switch_r_timeout(priv, GSWIP_PCE_TBL_CTRL, - GSWIP_PCE_TBL_CTRL_BAS); - if (err) - goto out_unlock; - - regmap_write(priv->gswip, GSWIP_PCE_TBL_ADDR, tbl->index); - regmap_write_bits(priv->gswip, GSWIP_PCE_TBL_CTRL, - GSWIP_PCE_TBL_CTRL_ADDR_MASK | - GSWIP_PCE_TBL_CTRL_OPMOD_MASK | - GSWIP_PCE_TBL_CTRL_BAS, - tbl->table | addr_mode | GSWIP_PCE_TBL_CTRL_BAS); - - err = gswip_switch_r_timeout(priv, GSWIP_PCE_TBL_CTRL, - GSWIP_PCE_TBL_CTRL_BAS); - if (err) - goto out_unlock; - - for (i = 0; i < ARRAY_SIZE(tbl->key); i++) { - err = regmap_read(priv->gswip, GSWIP_PCE_TBL_KEY(i), &tmp); - if (err) - goto out_unlock; - tbl->key[i] = tmp; - } - for (i = 0; i < ARRAY_SIZE(tbl->val); i++) { - err = regmap_read(priv->gswip, GSWIP_PCE_TBL_VAL(i), &tmp); - if (err) - goto out_unlock; - tbl->val[i] = tmp; - } - - err = regmap_read(priv->gswip, GSWIP_PCE_TBL_MASK, &tmp); - if (err) - goto out_unlock; - - tbl->mask = tmp; - err = regmap_read(priv->gswip, GSWIP_PCE_TBL_CTRL, &crtl); - if (err) - goto out_unlock; - - tbl->type = !!(crtl & GSWIP_PCE_TBL_CTRL_TYPE); - tbl->valid = !!(crtl & GSWIP_PCE_TBL_CTRL_VLD); - tbl->gmap = (crtl & GSWIP_PCE_TBL_CTRL_GMAP_MASK) >> 7; - -out_unlock: - mutex_unlock(&priv->pce_table_lock); - - return err; -} - -static int gswip_pce_table_entry_write(struct gswip_priv *priv, - struct gswip_pce_table_entry *tbl) -{ - int i; - int err; - u32 crtl; - u16 addr_mode = tbl->key_mode ? GSWIP_PCE_TBL_CTRL_OPMOD_KSWR : - GSWIP_PCE_TBL_CTRL_OPMOD_ADWR; - - mutex_lock(&priv->pce_table_lock); - - err = gswip_switch_r_timeout(priv, GSWIP_PCE_TBL_CTRL, - GSWIP_PCE_TBL_CTRL_BAS); - if (err) { - mutex_unlock(&priv->pce_table_lock); - return err; - } - - regmap_write(priv->gswip, GSWIP_PCE_TBL_ADDR, tbl->index); - regmap_write_bits(priv->gswip, GSWIP_PCE_TBL_CTRL, - GSWIP_PCE_TBL_CTRL_ADDR_MASK | - GSWIP_PCE_TBL_CTRL_OPMOD_MASK, - tbl->table | addr_mode); - - for (i = 0; i < ARRAY_SIZE(tbl->key); i++) - regmap_write(priv->gswip, GSWIP_PCE_TBL_KEY(i), tbl->key[i]); - - for (i = 0; i < ARRAY_SIZE(tbl->val); i++) - regmap_write(priv->gswip, GSWIP_PCE_TBL_VAL(i), tbl->val[i]); - - regmap_write_bits(priv->gswip, GSWIP_PCE_TBL_CTRL, - GSWIP_PCE_TBL_CTRL_ADDR_MASK | - GSWIP_PCE_TBL_CTRL_OPMOD_MASK, - tbl->table | addr_mode); - - regmap_write(priv->gswip, GSWIP_PCE_TBL_MASK, tbl->mask); - - regmap_read(priv->gswip, GSWIP_PCE_TBL_CTRL, &crtl); - crtl &= ~(GSWIP_PCE_TBL_CTRL_TYPE | GSWIP_PCE_TBL_CTRL_VLD | - GSWIP_PCE_TBL_CTRL_GMAP_MASK); - if (tbl->type) - crtl |= GSWIP_PCE_TBL_CTRL_TYPE; - if (tbl->valid) - crtl |= GSWIP_PCE_TBL_CTRL_VLD; - crtl |= (tbl->gmap << 7) & GSWIP_PCE_TBL_CTRL_GMAP_MASK; - crtl |= GSWIP_PCE_TBL_CTRL_BAS; - regmap_write(priv->gswip, GSWIP_PCE_TBL_CTRL, crtl); - - err = gswip_switch_r_timeout(priv, GSWIP_PCE_TBL_CTRL, - GSWIP_PCE_TBL_CTRL_BAS); - - mutex_unlock(&priv->pce_table_lock); - - return err; -} - -/* Add the LAN port into a bridge with the CPU port by - * default. This prevents automatic forwarding of - * packages between the LAN ports when no explicit - * bridge is configured. - */ -static int gswip_add_single_port_br(struct gswip_priv *priv, int port, bool add) -{ - struct gswip_pce_table_entry vlan_active = {0,}; - struct gswip_pce_table_entry vlan_mapping = {0,}; - int err; - - vlan_active.index = port + 1; - vlan_active.table = GSWIP_TABLE_ACTIVE_VLAN; - vlan_active.key[0] = GSWIP_VLAN_UNAWARE_PVID; - vlan_active.val[0] = port + 1 /* fid */; - vlan_active.valid = add; - err = gswip_pce_table_entry_write(priv, &vlan_active); - if (err) { - dev_err(priv->dev, "failed to write active VLAN: %d\n", err); - return err; - } - - if (!add) - return 0; - - vlan_mapping.index = port + 1; - vlan_mapping.table = GSWIP_TABLE_VLAN_MAPPING; - vlan_mapping.val[0] = GSWIP_VLAN_UNAWARE_PVID; - vlan_mapping.val[1] = BIT(port) | dsa_cpu_ports(priv->ds); - vlan_mapping.val[2] = 0; - err = gswip_pce_table_entry_write(priv, &vlan_mapping); - if (err) { - dev_err(priv->dev, "failed to write VLAN mapping: %d\n", err); - return err; - } - - return 0; -} - -static int gswip_port_setup(struct dsa_switch *ds, int port) -{ - struct gswip_priv *priv = ds->priv; - int err; - - if (!dsa_is_cpu_port(ds, port)) { - err = gswip_add_single_port_br(priv, port, true); - if (err) - return err; - } - - return 0; -} - -static int gswip_port_enable(struct dsa_switch *ds, int port, - struct phy_device *phydev) -{ - struct gswip_priv *priv = ds->priv; - - if (!dsa_is_cpu_port(ds, port)) { - u32 mdio_phy = 0; - - if (phydev) - mdio_phy = phydev->mdio.addr & GSWIP_MDIO_PHY_ADDR_MASK; - - regmap_write_bits(priv->mdio, GSWIP_MDIO_PHYp(port), - GSWIP_MDIO_PHY_ADDR_MASK, - mdio_phy); - } - - /* RMON Counter Enable for port */ - regmap_write(priv->gswip, GSWIP_BM_PCFGp(port), GSWIP_BM_PCFG_CNTEN); - - /* enable port fetch/store dma & VLAN Modification */ - regmap_set_bits(priv->gswip, GSWIP_FDMA_PCTRLp(port), - GSWIP_FDMA_PCTRL_EN | GSWIP_FDMA_PCTRL_VLANMOD_BOTH); - regmap_set_bits(priv->gswip, GSWIP_SDMA_PCTRLp(port), - GSWIP_SDMA_PCTRL_EN); - - return 0; -} - -static void gswip_port_disable(struct dsa_switch *ds, int port) -{ - struct gswip_priv *priv = ds->priv; - - regmap_clear_bits(priv->gswip, GSWIP_FDMA_PCTRLp(port), - GSWIP_FDMA_PCTRL_EN); - regmap_clear_bits(priv->gswip, GSWIP_SDMA_PCTRLp(port), - GSWIP_SDMA_PCTRL_EN); -} - -static int gswip_pce_load_microcode(struct gswip_priv *priv) -{ - int i; - int err; - - regmap_write_bits(priv->gswip, GSWIP_PCE_TBL_CTRL, - GSWIP_PCE_TBL_CTRL_ADDR_MASK | - GSWIP_PCE_TBL_CTRL_OPMOD_MASK | - GSWIP_PCE_TBL_CTRL_OPMOD_ADWR, - GSWIP_PCE_TBL_CTRL_OPMOD_ADWR); - regmap_write(priv->gswip, GSWIP_PCE_TBL_MASK, 0); - - for (i = 0; i < priv->hw_info->pce_microcode_size; i++) { - regmap_write(priv->gswip, GSWIP_PCE_TBL_ADDR, i); - regmap_write(priv->gswip, GSWIP_PCE_TBL_VAL(0), - (*priv->hw_info->pce_microcode)[i].val_0); - regmap_write(priv->gswip, GSWIP_PCE_TBL_VAL(1), - (*priv->hw_info->pce_microcode)[i].val_1); - regmap_write(priv->gswip, GSWIP_PCE_TBL_VAL(2), - (*priv->hw_info->pce_microcode)[i].val_2); - regmap_write(priv->gswip, GSWIP_PCE_TBL_VAL(3), - (*priv->hw_info->pce_microcode)[i].val_3); - - /* start the table access: */ - regmap_set_bits(priv->gswip, GSWIP_PCE_TBL_CTRL, - GSWIP_PCE_TBL_CTRL_BAS); - err = gswip_switch_r_timeout(priv, GSWIP_PCE_TBL_CTRL, - GSWIP_PCE_TBL_CTRL_BAS); - if (err) - return err; - } - - /* tell the switch that the microcode is loaded */ - regmap_set_bits(priv->gswip, GSWIP_PCE_GCTRL_0, - GSWIP_PCE_GCTRL_0_MC_VALID); - - return 0; -} - -static void gswip_port_commit_pvid(struct gswip_priv *priv, int port) -{ - struct dsa_port *dp = dsa_to_port(priv->ds, port); - struct net_device *br = dsa_port_bridge_dev_get(dp); - u32 vinr; - int idx; - - if (!dsa_port_is_user(dp)) - return; - - if (br) { - u16 pvid = GSWIP_VLAN_UNAWARE_PVID; - - if (br_vlan_enabled(br)) - br_vlan_get_pvid(br, &pvid); - - /* VLAN-aware bridge ports with no PVID will use Active VLAN - * index 0. The expectation is that this drops all untagged and - * VID-0 tagged ingress traffic. - */ - idx = 0; - for (int i = priv->hw_info->max_ports; - i < ARRAY_SIZE(priv->vlans); i++) { - if (priv->vlans[i].bridge == br && - priv->vlans[i].vid == pvid) { - idx = i; - break; - } - } - } else { - /* The Active VLAN table index as configured by - * gswip_add_single_port_br() - */ - idx = port + 1; - } - - vinr = idx ? GSWIP_PCE_VCTRL_VINR_ALL : GSWIP_PCE_VCTRL_VINR_TAGGED; - regmap_write_bits(priv->gswip, GSWIP_PCE_VCTRL(port), - GSWIP_PCE_VCTRL_VINR, - FIELD_PREP(GSWIP_PCE_VCTRL_VINR, vinr)); - - /* Note that in GSWIP 2.2 VLAN mode the VID needs to be programmed - * directly instead of referencing the index in the Active VLAN Tablet. - * However, without the VLANMD bit (9) in PCE_GCTRL_1 (0x457) even - * GSWIP 2.2 and newer hardware maintain the GSWIP 2.1 behavior. - */ - regmap_write(priv->gswip, GSWIP_PCE_DEFPVID(port), idx); -} - -static int gswip_port_vlan_filtering(struct dsa_switch *ds, int port, - bool vlan_filtering, - struct netlink_ext_ack *extack) -{ - struct gswip_priv *priv = ds->priv; - - if (vlan_filtering) { - /* Use tag based VLAN */ - regmap_write_bits(priv->gswip, GSWIP_PCE_VCTRL(port), - GSWIP_PCE_VCTRL_VSR | - GSWIP_PCE_VCTRL_UVR | - GSWIP_PCE_VCTRL_VIMR | - GSWIP_PCE_VCTRL_VEMR | - GSWIP_PCE_VCTRL_VID0, - GSWIP_PCE_VCTRL_UVR | - GSWIP_PCE_VCTRL_VIMR | - GSWIP_PCE_VCTRL_VEMR | - GSWIP_PCE_VCTRL_VID0); - regmap_clear_bits(priv->gswip, GSWIP_PCE_PCTRL_0p(port), - GSWIP_PCE_PCTRL_0_TVM); - } else { - /* Use port based VLAN */ - regmap_write_bits(priv->gswip, GSWIP_PCE_VCTRL(port), - GSWIP_PCE_VCTRL_UVR | - GSWIP_PCE_VCTRL_VIMR | - GSWIP_PCE_VCTRL_VEMR | - GSWIP_PCE_VCTRL_VID0 | - GSWIP_PCE_VCTRL_VSR, - GSWIP_PCE_VCTRL_VSR); - regmap_set_bits(priv->gswip, GSWIP_PCE_PCTRL_0p(port), - GSWIP_PCE_PCTRL_0_TVM); - } - - gswip_port_commit_pvid(priv, port); - - return 0; -} - -static int gswip_setup(struct dsa_switch *ds) -{ - unsigned int cpu_ports = dsa_cpu_ports(ds); - struct gswip_priv *priv = ds->priv; - struct dsa_port *cpu_dp; - int err, i; - - regmap_write(priv->gswip, GSWIP_SWRES, GSWIP_SWRES_R0); - usleep_range(5000, 10000); - regmap_write(priv->gswip, GSWIP_SWRES, 0); - - /* disable port fetch/store dma on all ports */ - for (i = 0; i < priv->hw_info->max_ports; i++) { - gswip_port_disable(ds, i); - gswip_port_vlan_filtering(ds, i, false, NULL); - } - - /* enable Switch */ - regmap_set_bits(priv->mdio, GSWIP_MDIO_GLOB, GSWIP_MDIO_GLOB_ENABLE); - - err = gswip_pce_load_microcode(priv); - if (err) { - dev_err(priv->dev, "writing PCE microcode failed, %i\n", err); - return err; - } - - /* Default unknown Broadcast/Multicast/Unicast port maps */ - regmap_write(priv->gswip, GSWIP_PCE_PMAP1, cpu_ports); - regmap_write(priv->gswip, GSWIP_PCE_PMAP2, cpu_ports); - regmap_write(priv->gswip, GSWIP_PCE_PMAP3, cpu_ports); - - /* Deactivate MDIO PHY auto polling. Some PHYs as the AR8030 have an - * interoperability problem with this auto polling mechanism because - * their status registers think that the link is in a different state - * than it actually is. For the AR8030 it has the BMSR_ESTATEN bit set - * as well as ESTATUS_1000_TFULL and ESTATUS_1000_XFULL. This makes the - * auto polling state machine consider the link being negotiated with - * 1Gbit/s. Since the PHY itself is a Fast Ethernet RMII PHY this leads - * to the switch port being completely dead (RX and TX are both not - * working). - * Also with various other PHY / port combinations (PHY11G GPHY, PHY22F - * GPHY, external RGMII PEF7071/7072) any traffic would stop. Sometimes - * it would work fine for a few minutes to hours and then stop, on - * other device it would no traffic could be sent or received at all. - * Testing shows that when PHY auto polling is disabled these problems - * go away. - */ - regmap_write(priv->mdio, GSWIP_MDIO_MDC_CFG0, 0x0); - - /* Configure the MDIO Clock 2.5 MHz */ - regmap_write_bits(priv->mdio, GSWIP_MDIO_MDC_CFG1, 0xff, 0x09); - - /* bring up the mdio bus */ - err = gswip_mdio(priv); - if (err) { - dev_err(priv->dev, "mdio bus setup failed\n"); - return err; - } - - /* Disable the xMII interface and clear it's isolation bit */ - for (i = 0; i < priv->hw_info->max_ports; i++) - gswip_mii_mask_cfg(priv, - GSWIP_MII_CFG_EN | GSWIP_MII_CFG_ISOLATE, - 0, i); - - dsa_switch_for_each_cpu_port(cpu_dp, ds) { - /* enable special tag insertion on cpu port */ - regmap_set_bits(priv->gswip, GSWIP_FDMA_PCTRLp(cpu_dp->index), - GSWIP_FDMA_PCTRL_STEN); - - /* accept special tag in ingress direction */ - regmap_set_bits(priv->gswip, - GSWIP_PCE_PCTRL_0p(cpu_dp->index), - GSWIP_PCE_PCTRL_0_INGRESS); - } - - regmap_set_bits(priv->gswip, GSWIP_BM_QUEUE_GCTRL, - GSWIP_BM_QUEUE_GCTRL_GL_MOD); - - /* VLAN aware Switching */ - regmap_set_bits(priv->gswip, GSWIP_PCE_GCTRL_0, - GSWIP_PCE_GCTRL_0_VLAN); - - /* Flush MAC Table */ - regmap_set_bits(priv->gswip, GSWIP_PCE_GCTRL_0, - GSWIP_PCE_GCTRL_0_MTFL); - - err = gswip_switch_r_timeout(priv, GSWIP_PCE_GCTRL_0, - GSWIP_PCE_GCTRL_0_MTFL); - if (err) { - dev_err(priv->dev, "MAC flushing didn't finish\n"); - return err; - } - - ds->mtu_enforcement_ingress = true; - - return 0; -} - -static enum dsa_tag_protocol gswip_get_tag_protocol(struct dsa_switch *ds, - int port, - enum dsa_tag_protocol mp) -{ - struct gswip_priv *priv = ds->priv; - - return priv->hw_info->tag_protocol; -} - -static int gswip_vlan_active_create(struct gswip_priv *priv, - struct net_device *bridge, - int fid, u16 vid) -{ - struct gswip_pce_table_entry vlan_active = {0,}; - unsigned int max_ports = priv->hw_info->max_ports; - int idx = -1; - int err; - int i; - - /* Look for a free slot */ - for (i = max_ports; i < ARRAY_SIZE(priv->vlans); i++) { - if (!priv->vlans[i].bridge) { - idx = i; - break; - } - } - - if (idx == -1) - return -ENOSPC; - - if (fid == -1) - fid = idx; - - vlan_active.index = idx; - vlan_active.table = GSWIP_TABLE_ACTIVE_VLAN; - vlan_active.key[0] = vid; - vlan_active.val[0] = fid; - vlan_active.valid = true; - - err = gswip_pce_table_entry_write(priv, &vlan_active); - if (err) { - dev_err(priv->dev, "failed to write active VLAN: %d\n", err); - return err; - } - - priv->vlans[idx].bridge = bridge; - priv->vlans[idx].vid = vid; - priv->vlans[idx].fid = fid; - - return idx; -} - -static int gswip_vlan_active_remove(struct gswip_priv *priv, int idx) -{ - struct gswip_pce_table_entry vlan_active = {0,}; - int err; - - vlan_active.index = idx; - vlan_active.table = GSWIP_TABLE_ACTIVE_VLAN; - vlan_active.valid = false; - err = gswip_pce_table_entry_write(priv, &vlan_active); - if (err) - dev_err(priv->dev, "failed to delete active VLAN: %d\n", err); - priv->vlans[idx].bridge = NULL; - - return err; -} - -static int gswip_vlan_add(struct gswip_priv *priv, struct net_device *bridge, - int port, u16 vid, bool untagged, bool pvid, - bool vlan_aware) -{ - struct gswip_pce_table_entry vlan_mapping = {0,}; - unsigned int max_ports = priv->hw_info->max_ports; - unsigned int cpu_ports = dsa_cpu_ports(priv->ds); - bool active_vlan_created = false; - int fid = -1, idx = -1; - int i, err; - - /* Check if there is already a page for this bridge */ - for (i = max_ports; i < ARRAY_SIZE(priv->vlans); i++) { - if (priv->vlans[i].bridge == bridge) { - if (vlan_aware) { - if (fid != -1 && fid != priv->vlans[i].fid) - dev_err(priv->dev, "one bridge with multiple flow ids\n"); - fid = priv->vlans[i].fid; - } - if (priv->vlans[i].vid == vid) { - idx = i; - break; - } - } - } - - /* If this bridge is not programmed yet, add a Active VLAN table - * entry in a free slot and prepare the VLAN mapping table entry. - */ - if (idx == -1) { - idx = gswip_vlan_active_create(priv, bridge, fid, vid); - if (idx < 0) - return idx; - active_vlan_created = true; - - vlan_mapping.index = idx; - vlan_mapping.table = GSWIP_TABLE_VLAN_MAPPING; - } else { - /* Read the existing VLAN mapping entry from the switch */ - vlan_mapping.index = idx; - vlan_mapping.table = GSWIP_TABLE_VLAN_MAPPING; - err = gswip_pce_table_entry_read(priv, &vlan_mapping); - if (err) { - dev_err(priv->dev, "failed to read VLAN mapping: %d\n", - err); - return err; - } - } - - /* VLAN ID byte, maps to the VLAN ID of vlan active table */ - vlan_mapping.val[0] = vid; - /* Update the VLAN mapping entry and write it to the switch */ - vlan_mapping.val[1] |= cpu_ports; - vlan_mapping.val[1] |= BIT(port); - if (vlan_aware) - vlan_mapping.val[2] |= cpu_ports; - if (untagged) - vlan_mapping.val[2] &= ~BIT(port); - else - vlan_mapping.val[2] |= BIT(port); - err = gswip_pce_table_entry_write(priv, &vlan_mapping); - if (err) { - dev_err(priv->dev, "failed to write VLAN mapping: %d\n", err); - /* In case an Active VLAN was creaetd delete it again */ - if (active_vlan_created) - gswip_vlan_active_remove(priv, idx); - return err; - } - - gswip_port_commit_pvid(priv, port); - - return 0; -} - -static int gswip_vlan_remove(struct gswip_priv *priv, - struct net_device *bridge, int port, - u16 vid) -{ - struct gswip_pce_table_entry vlan_mapping = {0,}; - unsigned int max_ports = priv->hw_info->max_ports; - int idx = -1; - int i; - int err; - - /* Check if there is already a page for this bridge */ - for (i = max_ports; i < ARRAY_SIZE(priv->vlans); i++) { - if (priv->vlans[i].bridge == bridge && - priv->vlans[i].vid == vid) { - idx = i; - break; - } - } - - if (idx == -1) { - dev_err(priv->dev, "Port %d cannot find VID %u of bridge %s\n", - port, vid, bridge ? bridge->name : "(null)"); - return -ENOENT; - } - - vlan_mapping.index = idx; - vlan_mapping.table = GSWIP_TABLE_VLAN_MAPPING; - err = gswip_pce_table_entry_read(priv, &vlan_mapping); - if (err) { - dev_err(priv->dev, "failed to read VLAN mapping: %d\n", err); - return err; - } - - vlan_mapping.val[1] &= ~BIT(port); - vlan_mapping.val[2] &= ~BIT(port); - err = gswip_pce_table_entry_write(priv, &vlan_mapping); - if (err) { - dev_err(priv->dev, "failed to write VLAN mapping: %d\n", err); - return err; - } - - /* In case all ports are removed from the bridge, remove the VLAN */ - if (!(vlan_mapping.val[1] & ~dsa_cpu_ports(priv->ds))) { - err = gswip_vlan_active_remove(priv, idx); - if (err) { - dev_err(priv->dev, "failed to write active VLAN: %d\n", - err); - return err; - } - } - - gswip_port_commit_pvid(priv, port); - - return 0; -} - -static int gswip_port_bridge_join(struct dsa_switch *ds, int port, - struct dsa_bridge bridge, - bool *tx_fwd_offload, - struct netlink_ext_ack *extack) -{ - struct net_device *br = bridge.dev; - struct gswip_priv *priv = ds->priv; - int err; - - /* Set up the VLAN for VLAN-unaware bridging for this port, and remove - * it from the "single-port bridge" through which it was operating as - * standalone. - */ - err = gswip_vlan_add(priv, br, port, GSWIP_VLAN_UNAWARE_PVID, - true, true, false); - if (err) - return err; - - return gswip_add_single_port_br(priv, port, false); -} - -static void gswip_port_bridge_leave(struct dsa_switch *ds, int port, - struct dsa_bridge bridge) -{ - struct net_device *br = bridge.dev; - struct gswip_priv *priv = ds->priv; - - /* Add the port back to the "single-port bridge", and remove it from - * the VLAN-unaware PVID created for this bridge. - */ - gswip_add_single_port_br(priv, port, true); - gswip_vlan_remove(priv, br, port, GSWIP_VLAN_UNAWARE_PVID); -} - -static int gswip_port_vlan_prepare(struct dsa_switch *ds, int port, - const struct switchdev_obj_port_vlan *vlan, - struct netlink_ext_ack *extack) -{ - struct net_device *bridge = dsa_port_bridge_dev_get(dsa_to_port(ds, port)); - struct gswip_priv *priv = ds->priv; - unsigned int max_ports = priv->hw_info->max_ports; - int pos = max_ports; - int i, idx = -1; - - /* We only support VLAN filtering on bridges */ - if (!dsa_is_cpu_port(ds, port) && !bridge) - return -EOPNOTSUPP; - - /* Check if there is already a page for this VLAN */ - for (i = max_ports; i < ARRAY_SIZE(priv->vlans); i++) { - if (priv->vlans[i].bridge == bridge && - priv->vlans[i].vid == vlan->vid) { - idx = i; - break; - } - } - - /* If this VLAN is not programmed yet, we have to reserve - * one entry in the VLAN table. Make sure we start at the - * next position round. - */ - if (idx == -1) { - /* Look for a free slot */ - for (; pos < ARRAY_SIZE(priv->vlans); pos++) { - if (!priv->vlans[pos].bridge) { - idx = pos; - pos++; - break; - } - } - - if (idx == -1) { - NL_SET_ERR_MSG_MOD(extack, "No slot in VLAN table"); - return -ENOSPC; - } - } - - return 0; -} - -static int gswip_port_vlan_add(struct dsa_switch *ds, int port, - const struct switchdev_obj_port_vlan *vlan, - struct netlink_ext_ack *extack) -{ - struct net_device *bridge = dsa_port_bridge_dev_get(dsa_to_port(ds, port)); - struct gswip_priv *priv = ds->priv; - bool untagged = vlan->flags & BRIDGE_VLAN_INFO_UNTAGGED; - bool pvid = vlan->flags & BRIDGE_VLAN_INFO_PVID; - int err; - - if (vlan->vid == GSWIP_VLAN_UNAWARE_PVID) - return 0; - - err = gswip_port_vlan_prepare(ds, port, vlan, extack); - if (err) - return err; - - /* We have to receive all packets on the CPU port and should not - * do any VLAN filtering here. This is also called with bridge - * NULL and then we do not know for which bridge to configure - * this. - */ - if (dsa_is_cpu_port(ds, port)) - return 0; - - return gswip_vlan_add(priv, bridge, port, vlan->vid, untagged, pvid, - true); -} - -static int gswip_port_vlan_del(struct dsa_switch *ds, int port, - const struct switchdev_obj_port_vlan *vlan) -{ - struct net_device *bridge = dsa_port_bridge_dev_get(dsa_to_port(ds, port)); - struct gswip_priv *priv = ds->priv; - - if (vlan->vid == GSWIP_VLAN_UNAWARE_PVID) - return 0; - - /* We have to receive all packets on the CPU port and should not - * do any VLAN filtering here. This is also called with bridge - * NULL and then we do not know for which bridge to configure - * this. - */ - if (dsa_is_cpu_port(ds, port)) - return 0; - - return gswip_vlan_remove(priv, bridge, port, vlan->vid); -} - -static void gswip_port_fast_age(struct dsa_switch *ds, int port) -{ - struct gswip_priv *priv = ds->priv; - struct gswip_pce_table_entry mac_bridge = {0,}; - int i; - int err; - - for (i = 0; i < 2048; i++) { - mac_bridge.table = GSWIP_TABLE_MAC_BRIDGE; - mac_bridge.index = i; - - err = gswip_pce_table_entry_read(priv, &mac_bridge); - if (err) { - dev_err(priv->dev, "failed to read mac bridge: %d\n", - err); - return; - } - - if (!mac_bridge.valid) - continue; - - if (mac_bridge.val[1] & GSWIP_TABLE_MAC_BRIDGE_VAL1_STATIC) - continue; - - if (port != FIELD_GET(GSWIP_TABLE_MAC_BRIDGE_VAL0_PORT, - mac_bridge.val[0])) - continue; - - mac_bridge.valid = false; - err = gswip_pce_table_entry_write(priv, &mac_bridge); - if (err) { - dev_err(priv->dev, "failed to write mac bridge: %d\n", - err); - return; - } - } -} - -static void gswip_port_stp_state_set(struct dsa_switch *ds, int port, u8 state) -{ - struct gswip_priv *priv = ds->priv; - u32 stp_state; - - switch (state) { - case BR_STATE_DISABLED: - regmap_clear_bits(priv->gswip, GSWIP_SDMA_PCTRLp(port), - GSWIP_SDMA_PCTRL_EN); - return; - case BR_STATE_BLOCKING: - case BR_STATE_LISTENING: - stp_state = GSWIP_PCE_PCTRL_0_PSTATE_LISTEN; - break; - case BR_STATE_LEARNING: - stp_state = GSWIP_PCE_PCTRL_0_PSTATE_LEARNING; - break; - case BR_STATE_FORWARDING: - stp_state = GSWIP_PCE_PCTRL_0_PSTATE_FORWARDING; - break; - default: - dev_err(priv->dev, "invalid STP state: %d\n", state); - return; - } - - regmap_set_bits(priv->gswip, GSWIP_SDMA_PCTRLp(port), - GSWIP_SDMA_PCTRL_EN); - regmap_write_bits(priv->gswip, GSWIP_PCE_PCTRL_0p(port), - GSWIP_PCE_PCTRL_0_PSTATE_MASK, - stp_state); -} - -static int gswip_port_fdb(struct dsa_switch *ds, int port, - struct net_device *bridge, const unsigned char *addr, - u16 vid, bool add) -{ - struct gswip_priv *priv = ds->priv; - struct gswip_pce_table_entry mac_bridge = {0,}; - unsigned int max_ports = priv->hw_info->max_ports; - int fid = -1; - int i; - int err; - - for (i = max_ports; i < ARRAY_SIZE(priv->vlans); i++) { - if (priv->vlans[i].bridge == bridge) { - fid = priv->vlans[i].fid; - break; - } - } - - if (fid == -1) { - dev_err(priv->dev, "no FID found for bridge %s\n", - bridge->name); - return -EINVAL; - } - - mac_bridge.table = GSWIP_TABLE_MAC_BRIDGE; - mac_bridge.key_mode = true; - mac_bridge.key[0] = addr[5] | (addr[4] << 8); - mac_bridge.key[1] = addr[3] | (addr[2] << 8); - mac_bridge.key[2] = addr[1] | (addr[0] << 8); - mac_bridge.key[3] = FIELD_PREP(GSWIP_TABLE_MAC_BRIDGE_KEY3_FID, fid); - mac_bridge.val[0] = add ? BIT(port) : 0; /* port map */ - mac_bridge.val[1] = GSWIP_TABLE_MAC_BRIDGE_VAL1_STATIC; - mac_bridge.valid = add; - - err = gswip_pce_table_entry_write(priv, &mac_bridge); - if (err) - dev_err(priv->dev, "failed to write mac bridge: %d\n", err); - - return err; -} - -static int gswip_port_fdb_add(struct dsa_switch *ds, int port, - const unsigned char *addr, u16 vid, - struct dsa_db db) -{ - if (db.type != DSA_DB_BRIDGE) - return -EOPNOTSUPP; - - return gswip_port_fdb(ds, port, db.bridge.dev, addr, vid, true); -} - -static int gswip_port_fdb_del(struct dsa_switch *ds, int port, - const unsigned char *addr, u16 vid, - struct dsa_db db) -{ - if (db.type != DSA_DB_BRIDGE) - return -EOPNOTSUPP; - - return gswip_port_fdb(ds, port, db.bridge.dev, addr, vid, false); -} - -static int gswip_port_fdb_dump(struct dsa_switch *ds, int port, - dsa_fdb_dump_cb_t *cb, void *data) -{ - struct gswip_priv *priv = ds->priv; - struct gswip_pce_table_entry mac_bridge = {0,}; - unsigned char addr[ETH_ALEN]; - int i; - int err; - - for (i = 0; i < 2048; i++) { - mac_bridge.table = GSWIP_TABLE_MAC_BRIDGE; - mac_bridge.index = i; - - err = gswip_pce_table_entry_read(priv, &mac_bridge); - if (err) { - dev_err(priv->dev, - "failed to read mac bridge entry %d: %d\n", - i, err); - return err; - } - - if (!mac_bridge.valid) - continue; - - addr[5] = mac_bridge.key[0] & 0xff; - addr[4] = (mac_bridge.key[0] >> 8) & 0xff; - addr[3] = mac_bridge.key[1] & 0xff; - addr[2] = (mac_bridge.key[1] >> 8) & 0xff; - addr[1] = mac_bridge.key[2] & 0xff; - addr[0] = (mac_bridge.key[2] >> 8) & 0xff; - if (mac_bridge.val[1] & GSWIP_TABLE_MAC_BRIDGE_VAL1_STATIC) { - if (mac_bridge.val[0] & BIT(port)) { - err = cb(addr, 0, true, data); - if (err) - return err; - } - } else { - if (port == FIELD_GET(GSWIP_TABLE_MAC_BRIDGE_VAL0_PORT, - mac_bridge.val[0])) { - err = cb(addr, 0, false, data); - if (err) - return err; - } - } - } - return 0; -} - -static int gswip_port_max_mtu(struct dsa_switch *ds, int port) -{ - /* Includes 8 bytes for special header. */ - return GSWIP_MAX_PACKET_LENGTH - VLAN_ETH_HLEN - ETH_FCS_LEN; -} - -static int gswip_port_change_mtu(struct dsa_switch *ds, int port, int new_mtu) -{ - struct gswip_priv *priv = ds->priv; - - /* CPU port always has maximum mtu of user ports, so use it to set - * switch frame size, including 8 byte special header. - */ - if (dsa_is_cpu_port(ds, port)) { - new_mtu += 8; - regmap_write(priv->gswip, GSWIP_MAC_FLEN, - VLAN_ETH_HLEN + new_mtu + ETH_FCS_LEN); - } - - /* Enable MLEN for ports with non-standard MTUs, including the special - * header on the CPU port added above. - */ - if (new_mtu != ETH_DATA_LEN) - regmap_set_bits(priv->gswip, GSWIP_MAC_CTRL_2p(port), - GSWIP_MAC_CTRL_2_MLEN); - else - regmap_clear_bits(priv->gswip, GSWIP_MAC_CTRL_2p(port), - GSWIP_MAC_CTRL_2_MLEN); - - return 0; -} - static void gswip_xrx200_phylink_get_caps(struct dsa_switch *ds, int port, struct phylink_config *config) { @@ -1291,327 +97,6 @@ static void gswip_xrx300_phylink_get_caps(struct dsa_switch *ds, int port, MAC_10 | MAC_100 | MAC_1000; } -static void gswip_phylink_get_caps(struct dsa_switch *ds, int port, - struct phylink_config *config) -{ - struct gswip_priv *priv = ds->priv; - - priv->hw_info->phylink_get_caps(ds, port, config); -} - -static void gswip_port_set_link(struct gswip_priv *priv, int port, bool link) -{ - u32 mdio_phy; - - if (link) - mdio_phy = GSWIP_MDIO_PHY_LINK_UP; - else - mdio_phy = GSWIP_MDIO_PHY_LINK_DOWN; - - regmap_write_bits(priv->mdio, GSWIP_MDIO_PHYp(port), - GSWIP_MDIO_PHY_LINK_MASK, mdio_phy); -} - -static void gswip_port_set_speed(struct gswip_priv *priv, int port, int speed, - phy_interface_t interface) -{ - u32 mdio_phy = 0, mii_cfg = 0, mac_ctrl_0 = 0; - - switch (speed) { - case SPEED_10: - mdio_phy = GSWIP_MDIO_PHY_SPEED_M10; - - if (interface == PHY_INTERFACE_MODE_RMII) - mii_cfg = GSWIP_MII_CFG_RATE_M50; - else - mii_cfg = GSWIP_MII_CFG_RATE_M2P5; - - mac_ctrl_0 = GSWIP_MAC_CTRL_0_GMII_MII; - break; - - case SPEED_100: - mdio_phy = GSWIP_MDIO_PHY_SPEED_M100; - - if (interface == PHY_INTERFACE_MODE_RMII) - mii_cfg = GSWIP_MII_CFG_RATE_M50; - else - mii_cfg = GSWIP_MII_CFG_RATE_M25; - - mac_ctrl_0 = GSWIP_MAC_CTRL_0_GMII_MII; - break; - - case SPEED_1000: - mdio_phy = GSWIP_MDIO_PHY_SPEED_G1; - - mii_cfg = GSWIP_MII_CFG_RATE_M125; - - mac_ctrl_0 = GSWIP_MAC_CTRL_0_GMII_RGMII; - break; - } - - regmap_write_bits(priv->mdio, GSWIP_MDIO_PHYp(port), - GSWIP_MDIO_PHY_SPEED_MASK, mdio_phy); - gswip_mii_mask_cfg(priv, GSWIP_MII_CFG_RATE_MASK, mii_cfg, port); - regmap_write_bits(priv->gswip, GSWIP_MAC_CTRL_0p(port), - GSWIP_MAC_CTRL_0_GMII_MASK, mac_ctrl_0); -} - -static void gswip_port_set_duplex(struct gswip_priv *priv, int port, int duplex) -{ - u32 mac_ctrl_0, mdio_phy; - - if (duplex == DUPLEX_FULL) { - mac_ctrl_0 = GSWIP_MAC_CTRL_0_FDUP_EN; - mdio_phy = GSWIP_MDIO_PHY_FDUP_EN; - } else { - mac_ctrl_0 = GSWIP_MAC_CTRL_0_FDUP_DIS; - mdio_phy = GSWIP_MDIO_PHY_FDUP_DIS; - } - - regmap_write_bits(priv->gswip, GSWIP_MAC_CTRL_0p(port), - GSWIP_MAC_CTRL_0_FDUP_MASK, mac_ctrl_0); - regmap_write_bits(priv->mdio, GSWIP_MDIO_PHYp(port), - GSWIP_MDIO_PHY_FDUP_MASK, mdio_phy); -} - -static void gswip_port_set_pause(struct gswip_priv *priv, int port, - bool tx_pause, bool rx_pause) -{ - u32 mac_ctrl_0, mdio_phy; - - if (tx_pause && rx_pause) { - mac_ctrl_0 = GSWIP_MAC_CTRL_0_FCON_RXTX; - mdio_phy = GSWIP_MDIO_PHY_FCONTX_EN | - GSWIP_MDIO_PHY_FCONRX_EN; - } else if (tx_pause) { - mac_ctrl_0 = GSWIP_MAC_CTRL_0_FCON_TX; - mdio_phy = GSWIP_MDIO_PHY_FCONTX_EN | - GSWIP_MDIO_PHY_FCONRX_DIS; - } else if (rx_pause) { - mac_ctrl_0 = GSWIP_MAC_CTRL_0_FCON_RX; - mdio_phy = GSWIP_MDIO_PHY_FCONTX_DIS | - GSWIP_MDIO_PHY_FCONRX_EN; - } else { - mac_ctrl_0 = GSWIP_MAC_CTRL_0_FCON_NONE; - mdio_phy = GSWIP_MDIO_PHY_FCONTX_DIS | - GSWIP_MDIO_PHY_FCONRX_DIS; - } - - regmap_write_bits(priv->gswip, GSWIP_MAC_CTRL_0p(port), - GSWIP_MAC_CTRL_0_FCON_MASK, mac_ctrl_0); - regmap_write_bits(priv->mdio, GSWIP_MDIO_PHYp(port), - GSWIP_MDIO_PHY_FCONTX_MASK | GSWIP_MDIO_PHY_FCONRX_MASK, - mdio_phy); -} - -static void gswip_phylink_mac_config(struct phylink_config *config, - unsigned int mode, - const struct phylink_link_state *state) -{ - struct dsa_port *dp = dsa_phylink_to_port(config); - struct gswip_priv *priv = dp->ds->priv; - int port = dp->index; - u32 miicfg = 0; - - miicfg |= GSWIP_MII_CFG_LDCLKDIS; - - switch (state->interface) { - case PHY_INTERFACE_MODE_SGMII: - case PHY_INTERFACE_MODE_1000BASEX: - case PHY_INTERFACE_MODE_2500BASEX: - return; - case PHY_INTERFACE_MODE_MII: - case PHY_INTERFACE_MODE_INTERNAL: - miicfg |= GSWIP_MII_CFG_MODE_MIIM; - break; - case PHY_INTERFACE_MODE_REVMII: - miicfg |= GSWIP_MII_CFG_MODE_MIIP; - break; - case PHY_INTERFACE_MODE_RMII: - miicfg |= GSWIP_MII_CFG_MODE_RMIIM; - break; - case PHY_INTERFACE_MODE_RGMII: - case PHY_INTERFACE_MODE_RGMII_ID: - case PHY_INTERFACE_MODE_RGMII_RXID: - case PHY_INTERFACE_MODE_RGMII_TXID: - miicfg |= GSWIP_MII_CFG_MODE_RGMII; - break; - case PHY_INTERFACE_MODE_GMII: - miicfg |= GSWIP_MII_CFG_MODE_GMII; - break; - default: - dev_err(dp->ds->dev, - "Unsupported interface: %d\n", state->interface); - return; - } - - gswip_mii_mask_cfg(priv, - GSWIP_MII_CFG_MODE_MASK | GSWIP_MII_CFG_RMII_CLK | - GSWIP_MII_CFG_RGMII_IBS | GSWIP_MII_CFG_LDCLKDIS, - miicfg, port); - - switch (state->interface) { - case PHY_INTERFACE_MODE_RGMII_ID: - gswip_mii_mask_pcdu(priv, GSWIP_MII_PCDU_TXDLY_MASK | - GSWIP_MII_PCDU_RXDLY_MASK, 0, port); - break; - case PHY_INTERFACE_MODE_RGMII_RXID: - gswip_mii_mask_pcdu(priv, GSWIP_MII_PCDU_RXDLY_MASK, 0, port); - break; - case PHY_INTERFACE_MODE_RGMII_TXID: - gswip_mii_mask_pcdu(priv, GSWIP_MII_PCDU_TXDLY_MASK, 0, port); - break; - default: - break; - } -} - -static void gswip_phylink_mac_link_down(struct phylink_config *config, - unsigned int mode, - phy_interface_t interface) -{ - struct dsa_port *dp = dsa_phylink_to_port(config); - struct gswip_priv *priv = dp->ds->priv; - int port = dp->index; - - gswip_mii_mask_cfg(priv, GSWIP_MII_CFG_EN, 0, port); - - if (!dsa_port_is_cpu(dp)) - gswip_port_set_link(priv, port, false); -} - -static void gswip_phylink_mac_link_up(struct phylink_config *config, - struct phy_device *phydev, - unsigned int mode, - phy_interface_t interface, - int speed, int duplex, - bool tx_pause, bool rx_pause) -{ - struct dsa_port *dp = dsa_phylink_to_port(config); - struct gswip_priv *priv = dp->ds->priv; - int port = dp->index; - - if (!dsa_port_is_cpu(dp)) { - gswip_port_set_link(priv, port, true); - gswip_port_set_speed(priv, port, speed, interface); - gswip_port_set_duplex(priv, port, duplex); - gswip_port_set_pause(priv, port, tx_pause, rx_pause); - } - - gswip_mii_mask_cfg(priv, GSWIP_MII_CFG_EN, GSWIP_MII_CFG_EN, port); -} - -static void gswip_get_strings(struct dsa_switch *ds, int port, u32 stringset, - uint8_t *data) -{ - int i; - - if (stringset != ETH_SS_STATS) - return; - - for (i = 0; i < ARRAY_SIZE(gswip_rmon_cnt); i++) - ethtool_puts(&data, gswip_rmon_cnt[i].name); -} - -static u32 gswip_bcm_ram_entry_read(struct gswip_priv *priv, u32 table, - u32 index) -{ - u32 result, val; - int err; - - regmap_write(priv->gswip, GSWIP_BM_RAM_ADDR, index); - regmap_write_bits(priv->gswip, GSWIP_BM_RAM_CTRL, - GSWIP_BM_RAM_CTRL_ADDR_MASK | GSWIP_BM_RAM_CTRL_OPMOD | - GSWIP_BM_RAM_CTRL_BAS, - table | GSWIP_BM_RAM_CTRL_BAS); - - err = gswip_switch_r_timeout(priv, GSWIP_BM_RAM_CTRL, - GSWIP_BM_RAM_CTRL_BAS); - if (err) { - dev_err(priv->dev, "timeout while reading table: %u, index: %u\n", - table, index); - return 0; - } - - regmap_read(priv->gswip, GSWIP_BM_RAM_VAL(0), &result); - regmap_read(priv->gswip, GSWIP_BM_RAM_VAL(1), &val); - result |= val << 16; - - return result; -} - -static void gswip_get_ethtool_stats(struct dsa_switch *ds, int port, - uint64_t *data) -{ - struct gswip_priv *priv = ds->priv; - const struct gswip_rmon_cnt_desc *rmon_cnt; - int i; - u64 high; - - for (i = 0; i < ARRAY_SIZE(gswip_rmon_cnt); i++) { - rmon_cnt = &gswip_rmon_cnt[i]; - - data[i] = gswip_bcm_ram_entry_read(priv, port, - rmon_cnt->offset); - if (rmon_cnt->size == 2) { - high = gswip_bcm_ram_entry_read(priv, port, - rmon_cnt->offset + 1); - data[i] |= high << 32; - } - } -} - -static int gswip_get_sset_count(struct dsa_switch *ds, int port, int sset) -{ - if (sset != ETH_SS_STATS) - return 0; - - return ARRAY_SIZE(gswip_rmon_cnt); -} - -static struct phylink_pcs *gswip_phylink_mac_select_pcs(struct phylink_config *config, - phy_interface_t interface) -{ - struct dsa_port *dp = dsa_phylink_to_port(config); - struct gswip_priv *priv = dp->ds->priv; - - if (priv->hw_info->mac_select_pcs) - return priv->hw_info->mac_select_pcs(config, interface); - - return NULL; -} - -static const struct phylink_mac_ops gswip_phylink_mac_ops = { - .mac_config = gswip_phylink_mac_config, - .mac_link_down = gswip_phylink_mac_link_down, - .mac_link_up = gswip_phylink_mac_link_up, - .mac_select_pcs = gswip_phylink_mac_select_pcs, -}; - -static const struct dsa_switch_ops gswip_switch_ops = { - .get_tag_protocol = gswip_get_tag_protocol, - .setup = gswip_setup, - .port_setup = gswip_port_setup, - .port_enable = gswip_port_enable, - .port_disable = gswip_port_disable, - .port_bridge_join = gswip_port_bridge_join, - .port_bridge_leave = gswip_port_bridge_leave, - .port_fast_age = gswip_port_fast_age, - .port_vlan_filtering = gswip_port_vlan_filtering, - .port_vlan_add = gswip_port_vlan_add, - .port_vlan_del = gswip_port_vlan_del, - .port_stp_state_set = gswip_port_stp_state_set, - .port_fdb_add = gswip_port_fdb_add, - .port_fdb_del = gswip_port_fdb_del, - .port_fdb_dump = gswip_port_fdb_dump, - .port_change_mtu = gswip_port_change_mtu, - .port_max_mtu = gswip_port_max_mtu, - .phylink_get_caps = gswip_phylink_get_caps, - .get_strings = gswip_get_strings, - .get_ethtool_stats = gswip_get_ethtool_stats, - .get_sset_count = gswip_get_sset_count, -}; - static const struct xway_gphy_match_data xrx200a1x_gphy_data = { .fe_firmware_name = "lantiq/xrx200_phy22f_a14.bin", .ge_firmware_name = "lantiq/xrx200_phy11g_a14.bin", @@ -1832,30 +317,6 @@ static int gswip_gphy_fw_list(struct gswip_priv *priv, return err; } -static int gswip_validate_cpu_port(struct dsa_switch *ds) -{ - struct gswip_priv *priv = ds->priv; - struct dsa_port *cpu_dp; - int cpu_port = -1; - - dsa_switch_for_each_cpu_port(cpu_dp, ds) { - if (cpu_port != -1) - return dev_err_probe(ds->dev, -EINVAL, - "only a single CPU port is supported\n"); - - cpu_port = cpu_dp->index; - } - - if (cpu_port == -1) - return dev_err_probe(ds->dev, -EINVAL, "no CPU port defined\n"); - - if (BIT(cpu_port) & ~priv->hw_info->allowed_cpu_ports) - return dev_err_probe(ds->dev, -EINVAL, - "unsupported CPU port defined\n"); - - return 0; -} - static const struct regmap_config sw_regmap_config = { .name = "switch", .reg_bits = 32, @@ -1929,24 +390,9 @@ static int gswip_probe(struct platform_device *pdev) if (!priv->ds) return -ENOMEM; - priv->ds->dev = dev; - priv->ds->num_ports = priv->hw_info->max_ports; - priv->ds->priv = priv; - priv->ds->ops = &gswip_switch_ops; - priv->ds->phylink_mac_ops = &gswip_phylink_mac_ops; priv->dev = dev; - mutex_init(&priv->pce_table_lock); - regmap_read(priv->gswip, GSWIP_VERSION, &version); - /* The hardware has the 'major/minor' version bytes in the wrong order - * preventing numerical comparisons. Construct a 16-bit unsigned integer - * having the REV field as most significant byte and the MOD field as - * least significant byte. This is effectively swapping the two bytes of - * the version variable, but other than using swab16 it doesn't affect - * the source variable. - */ - priv->version = GSWIP_VERSION_REV(version) << 8 | - GSWIP_VERSION_MOD(version); + regmap_read(priv->gswip, GSWIP_VERSION, &version); np = dev->of_node; switch (version) { @@ -1976,25 +422,14 @@ static int gswip_probe(struct platform_device *pdev) "gphy fw probe failed\n"); } - err = dsa_register_switch(priv->ds); - if (err) { - dev_err_probe(dev, err, "dsa switch registration failed\n"); - goto gphy_fw_remove; - } - - err = gswip_validate_cpu_port(priv->ds); + err = gswip_probe_common(priv, version); if (err) - goto disable_switch; + goto gphy_fw_remove; platform_set_drvdata(pdev, priv); - dev_info(dev, "probed GSWIP version %lx mod %lx\n", - GSWIP_VERSION_REV(version), GSWIP_VERSION_MOD(version)); return 0; -disable_switch: - regmap_clear_bits(priv->mdio, GSWIP_MDIO_GLOB, GSWIP_MDIO_GLOB_ENABLE); - dsa_unregister_switch(priv->ds); gphy_fw_remove: for (i = 0; i < priv->num_gphy_fw; i++) gswip_gphy_fw_remove(priv, &priv->gphy_fw[i]); @@ -2010,7 +445,7 @@ static void gswip_remove(struct platform_device *pdev) return; /* disable the switch */ - regmap_clear_bits(priv->mdio, GSWIP_MDIO_GLOB, GSWIP_MDIO_GLOB_ENABLE); + gswip_disable_switch(priv); dsa_unregister_switch(priv->ds); diff --git a/drivers/net/dsa/lantiq/lantiq_gswip.h b/drivers/net/dsa/lantiq/lantiq_gswip.h index 24d759e06e15..d86290db19b4 100644 --- a/drivers/net/dsa/lantiq/lantiq_gswip.h +++ b/drivers/net/dsa/lantiq/lantiq_gswip.h @@ -278,4 +278,8 @@ struct gswip_priv { u16 version; }; +void gswip_disable_switch(struct gswip_priv *priv); + +int gswip_probe_common(struct gswip_priv *priv, u32 version); + #endif /* __LANTIQ_GSWIP_H */ diff --git a/drivers/net/dsa/lantiq/lantiq_gswip_common.c b/drivers/net/dsa/lantiq/lantiq_gswip_common.c new file mode 100644 index 000000000000..a0e361622acb --- /dev/null +++ b/drivers/net/dsa/lantiq/lantiq_gswip_common.c @@ -0,0 +1,1622 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Lantiq / Intel / MaxLinear GSWIP common function library + * + * Copyright (C) 2025 Daniel Golle + * Copyright (C) 2023 - 2024 MaxLinear Inc. + * Copyright (C) 2022 Snap One, LLC. All rights reserved. + * Copyright (C) 2017 - 2019 Hauke Mehrtens + * Copyright (C) 2012 John Crispin + * Copyright (C) 2010 Lantiq Deutschland + * + * The VLAN and bridge model the GSWIP hardware uses does not directly + * matches the model DSA uses. + * + * The hardware has 64 possible table entries for bridges with one VLAN + * ID, one flow id and a list of ports for each bridge. All entries which + * match the same flow ID are combined in the mac learning table, they + * act as one global bridge. + * The hardware does not support VLAN filter on the port, but on the + * bridge, this driver converts the DSA model to the hardware. + * + * The CPU gets all the exception frames which do not match any forwarding + * rule and the CPU port is also added to all bridges. This makes it possible + * to handle all the special cases easily in software. + * At the initialization the driver allocates one bridge table entry for + * each switch port which is used when the port is used without an + * explicit bridge. This prevents the frames from being forwarded + * between all LAN ports by default. + */ + +#include "lantiq_gswip.h" + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +struct gswip_pce_table_entry { + u16 index; // PCE_TBL_ADDR.ADDR = pData->table_index + u16 table; // PCE_TBL_CTRL.ADDR = pData->table + u16 key[8]; + u16 val[5]; + u16 mask; + u8 gmap; + bool type; + bool valid; + bool key_mode; +}; + +struct gswip_rmon_cnt_desc { + unsigned int size; + unsigned int offset; + const char *name; +}; + +#define MIB_DESC(_size, _offset, _name) {.size = _size, .offset = _offset, .name = _name} + +static const struct gswip_rmon_cnt_desc gswip_rmon_cnt[] = { + /** Receive Packet Count (only packets that are accepted and not discarded). */ + MIB_DESC(1, 0x1F, "RxGoodPkts"), + MIB_DESC(1, 0x23, "RxUnicastPkts"), + MIB_DESC(1, 0x22, "RxMulticastPkts"), + MIB_DESC(1, 0x21, "RxFCSErrorPkts"), + MIB_DESC(1, 0x1D, "RxUnderSizeGoodPkts"), + MIB_DESC(1, 0x1E, "RxUnderSizeErrorPkts"), + MIB_DESC(1, 0x1B, "RxOversizeGoodPkts"), + MIB_DESC(1, 0x1C, "RxOversizeErrorPkts"), + MIB_DESC(1, 0x20, "RxGoodPausePkts"), + MIB_DESC(1, 0x1A, "RxAlignErrorPkts"), + MIB_DESC(1, 0x12, "Rx64BytePkts"), + MIB_DESC(1, 0x13, "Rx127BytePkts"), + MIB_DESC(1, 0x14, "Rx255BytePkts"), + MIB_DESC(1, 0x15, "Rx511BytePkts"), + MIB_DESC(1, 0x16, "Rx1023BytePkts"), + /** Receive Size 1024-1522 (or more, if configured) Packet Count. */ + MIB_DESC(1, 0x17, "RxMaxBytePkts"), + MIB_DESC(1, 0x18, "RxDroppedPkts"), + MIB_DESC(1, 0x19, "RxFilteredPkts"), + MIB_DESC(2, 0x24, "RxGoodBytes"), + MIB_DESC(2, 0x26, "RxBadBytes"), + MIB_DESC(1, 0x11, "TxAcmDroppedPkts"), + MIB_DESC(1, 0x0C, "TxGoodPkts"), + MIB_DESC(1, 0x06, "TxUnicastPkts"), + MIB_DESC(1, 0x07, "TxMulticastPkts"), + MIB_DESC(1, 0x00, "Tx64BytePkts"), + MIB_DESC(1, 0x01, "Tx127BytePkts"), + MIB_DESC(1, 0x02, "Tx255BytePkts"), + MIB_DESC(1, 0x03, "Tx511BytePkts"), + MIB_DESC(1, 0x04, "Tx1023BytePkts"), + /** Transmit Size 1024-1522 (or more, if configured) Packet Count. */ + MIB_DESC(1, 0x05, "TxMaxBytePkts"), + MIB_DESC(1, 0x08, "TxSingleCollCount"), + MIB_DESC(1, 0x09, "TxMultCollCount"), + MIB_DESC(1, 0x0A, "TxLateCollCount"), + MIB_DESC(1, 0x0B, "TxExcessCollCount"), + MIB_DESC(1, 0x0D, "TxPauseCount"), + MIB_DESC(1, 0x10, "TxDroppedPkts"), + MIB_DESC(2, 0x0E, "TxGoodBytes"), +}; + +static u32 gswip_switch_r_timeout(struct gswip_priv *priv, u32 offset, + u32 cleared) +{ + u32 val; + + return regmap_read_poll_timeout(priv->gswip, offset, val, + !(val & cleared), 20, 50000); +} + +static void gswip_mii_mask_cfg(struct gswip_priv *priv, u32 mask, u32 set, + int port) +{ + int reg_port; + + /* MII_CFG register only exists for MII ports */ + if (!(priv->hw_info->mii_ports & BIT(port))) + return; + + reg_port = port + priv->hw_info->mii_port_reg_offset; + + regmap_write_bits(priv->mii, GSWIP_MII_CFGp(reg_port), mask, + set); +} + +static void gswip_mii_mask_pcdu(struct gswip_priv *priv, u32 mask, u32 set, + int port) +{ + int reg_port; + + /* MII_PCDU register only exists for MII ports */ + if (!(priv->hw_info->mii_ports & BIT(port))) + return; + + reg_port = port + priv->hw_info->mii_port_reg_offset; + + switch (reg_port) { + case 0: + regmap_write_bits(priv->mii, GSWIP_MII_PCDU0, mask, set); + break; + case 1: + regmap_write_bits(priv->mii, GSWIP_MII_PCDU1, mask, set); + break; + case 5: + regmap_write_bits(priv->mii, GSWIP_MII_PCDU5, mask, set); + break; + } +} + +static int gswip_mdio_poll(struct gswip_priv *priv) +{ + u32 ctrl; + + return regmap_read_poll_timeout(priv->mdio, GSWIP_MDIO_CTRL, ctrl, + !(ctrl & GSWIP_MDIO_CTRL_BUSY), 40, 4000); +} + +static int gswip_mdio_wr(struct mii_bus *bus, int addr, int reg, u16 val) +{ + struct gswip_priv *priv = bus->priv; + int err; + + err = gswip_mdio_poll(priv); + if (err) { + dev_err(&bus->dev, "waiting for MDIO bus busy timed out\n"); + return err; + } + + regmap_write(priv->mdio, GSWIP_MDIO_WRITE, val); + regmap_write(priv->mdio, GSWIP_MDIO_CTRL, + GSWIP_MDIO_CTRL_BUSY | GSWIP_MDIO_CTRL_WR | + ((addr & GSWIP_MDIO_CTRL_PHYAD_MASK) << GSWIP_MDIO_CTRL_PHYAD_SHIFT) | + (reg & GSWIP_MDIO_CTRL_REGAD_MASK)); + + return 0; +} + +static int gswip_mdio_rd(struct mii_bus *bus, int addr, int reg) +{ + struct gswip_priv *priv = bus->priv; + u32 val; + int err; + + err = gswip_mdio_poll(priv); + if (err) { + dev_err(&bus->dev, "waiting for MDIO bus busy timed out\n"); + return err; + } + + regmap_write(priv->mdio, GSWIP_MDIO_CTRL, + GSWIP_MDIO_CTRL_BUSY | GSWIP_MDIO_CTRL_RD | + ((addr & GSWIP_MDIO_CTRL_PHYAD_MASK) << GSWIP_MDIO_CTRL_PHYAD_SHIFT) | + (reg & GSWIP_MDIO_CTRL_REGAD_MASK)); + + err = gswip_mdio_poll(priv); + if (err) { + dev_err(&bus->dev, "waiting for MDIO bus busy timed out\n"); + return err; + } + + err = regmap_read(priv->mdio, GSWIP_MDIO_READ, &val); + if (err) + return err; + + return val; +} + +static int gswip_mdio(struct gswip_priv *priv) +{ + struct device_node *mdio_np, *switch_np = priv->dev->of_node; + struct device *dev = priv->dev; + struct mii_bus *bus; + int err = 0; + + mdio_np = of_get_compatible_child(switch_np, "lantiq,xrx200-mdio"); + if (!mdio_np) + mdio_np = of_get_child_by_name(switch_np, "mdio"); + + if (!of_device_is_available(mdio_np)) + goto out_put_node; + + bus = devm_mdiobus_alloc(dev); + if (!bus) { + err = -ENOMEM; + goto out_put_node; + } + + bus->priv = priv; + bus->read = gswip_mdio_rd; + bus->write = gswip_mdio_wr; + bus->name = "lantiq,xrx200-mdio"; + snprintf(bus->id, MII_BUS_ID_SIZE, "%s-mii", dev_name(priv->dev)); + bus->parent = priv->dev; + + err = devm_of_mdiobus_register(dev, bus, mdio_np); + +out_put_node: + of_node_put(mdio_np); + + return err; +} + +static int gswip_pce_table_entry_read(struct gswip_priv *priv, + struct gswip_pce_table_entry *tbl) +{ + int i; + int err; + u32 crtl; + u32 tmp; + u16 addr_mode = tbl->key_mode ? GSWIP_PCE_TBL_CTRL_OPMOD_KSRD : + GSWIP_PCE_TBL_CTRL_OPMOD_ADRD; + + mutex_lock(&priv->pce_table_lock); + + err = gswip_switch_r_timeout(priv, GSWIP_PCE_TBL_CTRL, + GSWIP_PCE_TBL_CTRL_BAS); + if (err) + goto out_unlock; + + regmap_write(priv->gswip, GSWIP_PCE_TBL_ADDR, tbl->index); + regmap_write_bits(priv->gswip, GSWIP_PCE_TBL_CTRL, + GSWIP_PCE_TBL_CTRL_ADDR_MASK | + GSWIP_PCE_TBL_CTRL_OPMOD_MASK | + GSWIP_PCE_TBL_CTRL_BAS, + tbl->table | addr_mode | GSWIP_PCE_TBL_CTRL_BAS); + + err = gswip_switch_r_timeout(priv, GSWIP_PCE_TBL_CTRL, + GSWIP_PCE_TBL_CTRL_BAS); + if (err) + goto out_unlock; + + for (i = 0; i < ARRAY_SIZE(tbl->key); i++) { + err = regmap_read(priv->gswip, GSWIP_PCE_TBL_KEY(i), &tmp); + if (err) + goto out_unlock; + tbl->key[i] = tmp; + } + for (i = 0; i < ARRAY_SIZE(tbl->val); i++) { + err = regmap_read(priv->gswip, GSWIP_PCE_TBL_VAL(i), &tmp); + if (err) + goto out_unlock; + tbl->val[i] = tmp; + } + + err = regmap_read(priv->gswip, GSWIP_PCE_TBL_MASK, &tmp); + if (err) + goto out_unlock; + + tbl->mask = tmp; + err = regmap_read(priv->gswip, GSWIP_PCE_TBL_CTRL, &crtl); + if (err) + goto out_unlock; + + tbl->type = !!(crtl & GSWIP_PCE_TBL_CTRL_TYPE); + tbl->valid = !!(crtl & GSWIP_PCE_TBL_CTRL_VLD); + tbl->gmap = (crtl & GSWIP_PCE_TBL_CTRL_GMAP_MASK) >> 7; + +out_unlock: + mutex_unlock(&priv->pce_table_lock); + + return err; +} + +static int gswip_pce_table_entry_write(struct gswip_priv *priv, + struct gswip_pce_table_entry *tbl) +{ + int i; + int err; + u32 crtl; + u16 addr_mode = tbl->key_mode ? GSWIP_PCE_TBL_CTRL_OPMOD_KSWR : + GSWIP_PCE_TBL_CTRL_OPMOD_ADWR; + + mutex_lock(&priv->pce_table_lock); + + err = gswip_switch_r_timeout(priv, GSWIP_PCE_TBL_CTRL, + GSWIP_PCE_TBL_CTRL_BAS); + if (err) { + mutex_unlock(&priv->pce_table_lock); + return err; + } + + regmap_write(priv->gswip, GSWIP_PCE_TBL_ADDR, tbl->index); + regmap_write_bits(priv->gswip, GSWIP_PCE_TBL_CTRL, + GSWIP_PCE_TBL_CTRL_ADDR_MASK | + GSWIP_PCE_TBL_CTRL_OPMOD_MASK, + tbl->table | addr_mode); + + for (i = 0; i < ARRAY_SIZE(tbl->key); i++) + regmap_write(priv->gswip, GSWIP_PCE_TBL_KEY(i), tbl->key[i]); + + for (i = 0; i < ARRAY_SIZE(tbl->val); i++) + regmap_write(priv->gswip, GSWIP_PCE_TBL_VAL(i), tbl->val[i]); + + regmap_write_bits(priv->gswip, GSWIP_PCE_TBL_CTRL, + GSWIP_PCE_TBL_CTRL_ADDR_MASK | + GSWIP_PCE_TBL_CTRL_OPMOD_MASK, + tbl->table | addr_mode); + + regmap_write(priv->gswip, GSWIP_PCE_TBL_MASK, tbl->mask); + + regmap_read(priv->gswip, GSWIP_PCE_TBL_CTRL, &crtl); + crtl &= ~(GSWIP_PCE_TBL_CTRL_TYPE | GSWIP_PCE_TBL_CTRL_VLD | + GSWIP_PCE_TBL_CTRL_GMAP_MASK); + if (tbl->type) + crtl |= GSWIP_PCE_TBL_CTRL_TYPE; + if (tbl->valid) + crtl |= GSWIP_PCE_TBL_CTRL_VLD; + crtl |= (tbl->gmap << 7) & GSWIP_PCE_TBL_CTRL_GMAP_MASK; + crtl |= GSWIP_PCE_TBL_CTRL_BAS; + regmap_write(priv->gswip, GSWIP_PCE_TBL_CTRL, crtl); + + err = gswip_switch_r_timeout(priv, GSWIP_PCE_TBL_CTRL, + GSWIP_PCE_TBL_CTRL_BAS); + + mutex_unlock(&priv->pce_table_lock); + + return err; +} + +/* Add the LAN port into a bridge with the CPU port by + * default. This prevents automatic forwarding of + * packages between the LAN ports when no explicit + * bridge is configured. + */ +static int gswip_add_single_port_br(struct gswip_priv *priv, int port, bool add) +{ + struct gswip_pce_table_entry vlan_active = {0,}; + struct gswip_pce_table_entry vlan_mapping = {0,}; + int err; + + vlan_active.index = port + 1; + vlan_active.table = GSWIP_TABLE_ACTIVE_VLAN; + vlan_active.key[0] = GSWIP_VLAN_UNAWARE_PVID; + vlan_active.val[0] = port + 1 /* fid */; + vlan_active.valid = add; + err = gswip_pce_table_entry_write(priv, &vlan_active); + if (err) { + dev_err(priv->dev, "failed to write active VLAN: %d\n", err); + return err; + } + + if (!add) + return 0; + + vlan_mapping.index = port + 1; + vlan_mapping.table = GSWIP_TABLE_VLAN_MAPPING; + vlan_mapping.val[0] = GSWIP_VLAN_UNAWARE_PVID; + vlan_mapping.val[1] = BIT(port) | dsa_cpu_ports(priv->ds); + vlan_mapping.val[2] = 0; + err = gswip_pce_table_entry_write(priv, &vlan_mapping); + if (err) { + dev_err(priv->dev, "failed to write VLAN mapping: %d\n", err); + return err; + } + + return 0; +} + +static int gswip_port_setup(struct dsa_switch *ds, int port) +{ + struct gswip_priv *priv = ds->priv; + int err; + + if (!dsa_is_cpu_port(ds, port)) { + err = gswip_add_single_port_br(priv, port, true); + if (err) + return err; + } + + return 0; +} + +static int gswip_port_enable(struct dsa_switch *ds, int port, + struct phy_device *phydev) +{ + struct gswip_priv *priv = ds->priv; + + if (!dsa_is_cpu_port(ds, port)) { + u32 mdio_phy = 0; + + if (phydev) + mdio_phy = phydev->mdio.addr & GSWIP_MDIO_PHY_ADDR_MASK; + + regmap_write_bits(priv->mdio, GSWIP_MDIO_PHYp(port), + GSWIP_MDIO_PHY_ADDR_MASK, + mdio_phy); + } + + /* RMON Counter Enable for port */ + regmap_write(priv->gswip, GSWIP_BM_PCFGp(port), GSWIP_BM_PCFG_CNTEN); + + /* enable port fetch/store dma & VLAN Modification */ + regmap_set_bits(priv->gswip, GSWIP_FDMA_PCTRLp(port), + GSWIP_FDMA_PCTRL_EN | GSWIP_FDMA_PCTRL_VLANMOD_BOTH); + regmap_set_bits(priv->gswip, GSWIP_SDMA_PCTRLp(port), + GSWIP_SDMA_PCTRL_EN); + + return 0; +} + +static void gswip_port_disable(struct dsa_switch *ds, int port) +{ + struct gswip_priv *priv = ds->priv; + + regmap_clear_bits(priv->gswip, GSWIP_FDMA_PCTRLp(port), + GSWIP_FDMA_PCTRL_EN); + regmap_clear_bits(priv->gswip, GSWIP_SDMA_PCTRLp(port), + GSWIP_SDMA_PCTRL_EN); +} + +static int gswip_pce_load_microcode(struct gswip_priv *priv) +{ + int i; + int err; + + regmap_write_bits(priv->gswip, GSWIP_PCE_TBL_CTRL, + GSWIP_PCE_TBL_CTRL_ADDR_MASK | + GSWIP_PCE_TBL_CTRL_OPMOD_MASK | + GSWIP_PCE_TBL_CTRL_OPMOD_ADWR, + GSWIP_PCE_TBL_CTRL_OPMOD_ADWR); + regmap_write(priv->gswip, GSWIP_PCE_TBL_MASK, 0); + + for (i = 0; i < priv->hw_info->pce_microcode_size; i++) { + regmap_write(priv->gswip, GSWIP_PCE_TBL_ADDR, i); + regmap_write(priv->gswip, GSWIP_PCE_TBL_VAL(0), + (*priv->hw_info->pce_microcode)[i].val_0); + regmap_write(priv->gswip, GSWIP_PCE_TBL_VAL(1), + (*priv->hw_info->pce_microcode)[i].val_1); + regmap_write(priv->gswip, GSWIP_PCE_TBL_VAL(2), + (*priv->hw_info->pce_microcode)[i].val_2); + regmap_write(priv->gswip, GSWIP_PCE_TBL_VAL(3), + (*priv->hw_info->pce_microcode)[i].val_3); + + /* start the table access: */ + regmap_set_bits(priv->gswip, GSWIP_PCE_TBL_CTRL, + GSWIP_PCE_TBL_CTRL_BAS); + err = gswip_switch_r_timeout(priv, GSWIP_PCE_TBL_CTRL, + GSWIP_PCE_TBL_CTRL_BAS); + if (err) + return err; + } + + /* tell the switch that the microcode is loaded */ + regmap_set_bits(priv->gswip, GSWIP_PCE_GCTRL_0, + GSWIP_PCE_GCTRL_0_MC_VALID); + + return 0; +} + +static void gswip_port_commit_pvid(struct gswip_priv *priv, int port) +{ + struct dsa_port *dp = dsa_to_port(priv->ds, port); + struct net_device *br = dsa_port_bridge_dev_get(dp); + u32 vinr; + int idx; + + if (!dsa_port_is_user(dp)) + return; + + if (br) { + u16 pvid = GSWIP_VLAN_UNAWARE_PVID; + + if (br_vlan_enabled(br)) + br_vlan_get_pvid(br, &pvid); + + /* VLAN-aware bridge ports with no PVID will use Active VLAN + * index 0. The expectation is that this drops all untagged and + * VID-0 tagged ingress traffic. + */ + idx = 0; + for (int i = priv->hw_info->max_ports; + i < ARRAY_SIZE(priv->vlans); i++) { + if (priv->vlans[i].bridge == br && + priv->vlans[i].vid == pvid) { + idx = i; + break; + } + } + } else { + /* The Active VLAN table index as configured by + * gswip_add_single_port_br() + */ + idx = port + 1; + } + + vinr = idx ? GSWIP_PCE_VCTRL_VINR_ALL : GSWIP_PCE_VCTRL_VINR_TAGGED; + regmap_write_bits(priv->gswip, GSWIP_PCE_VCTRL(port), + GSWIP_PCE_VCTRL_VINR, + FIELD_PREP(GSWIP_PCE_VCTRL_VINR, vinr)); + + /* Note that in GSWIP 2.2 VLAN mode the VID needs to be programmed + * directly instead of referencing the index in the Active VLAN Tablet. + * However, without the VLANMD bit (9) in PCE_GCTRL_1 (0x457) even + * GSWIP 2.2 and newer hardware maintain the GSWIP 2.1 behavior. + */ + regmap_write(priv->gswip, GSWIP_PCE_DEFPVID(port), idx); +} + +static int gswip_port_vlan_filtering(struct dsa_switch *ds, int port, + bool vlan_filtering, + struct netlink_ext_ack *extack) +{ + struct gswip_priv *priv = ds->priv; + + if (vlan_filtering) { + /* Use tag based VLAN */ + regmap_write_bits(priv->gswip, GSWIP_PCE_VCTRL(port), + GSWIP_PCE_VCTRL_VSR | + GSWIP_PCE_VCTRL_UVR | + GSWIP_PCE_VCTRL_VIMR | + GSWIP_PCE_VCTRL_VEMR | + GSWIP_PCE_VCTRL_VID0, + GSWIP_PCE_VCTRL_UVR | + GSWIP_PCE_VCTRL_VIMR | + GSWIP_PCE_VCTRL_VEMR | + GSWIP_PCE_VCTRL_VID0); + regmap_clear_bits(priv->gswip, GSWIP_PCE_PCTRL_0p(port), + GSWIP_PCE_PCTRL_0_TVM); + } else { + /* Use port based VLAN */ + regmap_write_bits(priv->gswip, GSWIP_PCE_VCTRL(port), + GSWIP_PCE_VCTRL_UVR | + GSWIP_PCE_VCTRL_VIMR | + GSWIP_PCE_VCTRL_VEMR | + GSWIP_PCE_VCTRL_VID0 | + GSWIP_PCE_VCTRL_VSR, + GSWIP_PCE_VCTRL_VSR); + regmap_set_bits(priv->gswip, GSWIP_PCE_PCTRL_0p(port), + GSWIP_PCE_PCTRL_0_TVM); + } + + gswip_port_commit_pvid(priv, port); + + return 0; +} + +static int gswip_setup(struct dsa_switch *ds) +{ + unsigned int cpu_ports = dsa_cpu_ports(ds); + struct gswip_priv *priv = ds->priv; + struct dsa_port *cpu_dp; + int err, i; + + regmap_write(priv->gswip, GSWIP_SWRES, GSWIP_SWRES_R0); + usleep_range(5000, 10000); + regmap_write(priv->gswip, GSWIP_SWRES, 0); + + /* disable port fetch/store dma on all ports */ + for (i = 0; i < priv->hw_info->max_ports; i++) { + gswip_port_disable(ds, i); + gswip_port_vlan_filtering(ds, i, false, NULL); + } + + /* enable Switch */ + regmap_set_bits(priv->mdio, GSWIP_MDIO_GLOB, GSWIP_MDIO_GLOB_ENABLE); + + err = gswip_pce_load_microcode(priv); + if (err) { + dev_err(priv->dev, "writing PCE microcode failed, %i\n", err); + return err; + } + + /* Default unknown Broadcast/Multicast/Unicast port maps */ + regmap_write(priv->gswip, GSWIP_PCE_PMAP1, cpu_ports); + regmap_write(priv->gswip, GSWIP_PCE_PMAP2, cpu_ports); + regmap_write(priv->gswip, GSWIP_PCE_PMAP3, cpu_ports); + + /* Deactivate MDIO PHY auto polling. Some PHYs as the AR8030 have an + * interoperability problem with this auto polling mechanism because + * their status registers think that the link is in a different state + * than it actually is. For the AR8030 it has the BMSR_ESTATEN bit set + * as well as ESTATUS_1000_TFULL and ESTATUS_1000_XFULL. This makes the + * auto polling state machine consider the link being negotiated with + * 1Gbit/s. Since the PHY itself is a Fast Ethernet RMII PHY this leads + * to the switch port being completely dead (RX and TX are both not + * working). + * Also with various other PHY / port combinations (PHY11G GPHY, PHY22F + * GPHY, external RGMII PEF7071/7072) any traffic would stop. Sometimes + * it would work fine for a few minutes to hours and then stop, on + * other device it would no traffic could be sent or received at all. + * Testing shows that when PHY auto polling is disabled these problems + * go away. + */ + regmap_write(priv->mdio, GSWIP_MDIO_MDC_CFG0, 0x0); + + /* Configure the MDIO Clock 2.5 MHz */ + regmap_write_bits(priv->mdio, GSWIP_MDIO_MDC_CFG1, 0xff, 0x09); + + /* bring up the mdio bus */ + err = gswip_mdio(priv); + if (err) { + dev_err(priv->dev, "mdio bus setup failed\n"); + return err; + } + + /* Disable the xMII interface and clear it's isolation bit */ + for (i = 0; i < priv->hw_info->max_ports; i++) + gswip_mii_mask_cfg(priv, + GSWIP_MII_CFG_EN | GSWIP_MII_CFG_ISOLATE, + 0, i); + + dsa_switch_for_each_cpu_port(cpu_dp, ds) { + /* enable special tag insertion on cpu port */ + regmap_set_bits(priv->gswip, GSWIP_FDMA_PCTRLp(cpu_dp->index), + GSWIP_FDMA_PCTRL_STEN); + + /* accept special tag in ingress direction */ + regmap_set_bits(priv->gswip, + GSWIP_PCE_PCTRL_0p(cpu_dp->index), + GSWIP_PCE_PCTRL_0_INGRESS); + } + + regmap_set_bits(priv->gswip, GSWIP_BM_QUEUE_GCTRL, + GSWIP_BM_QUEUE_GCTRL_GL_MOD); + + /* VLAN aware Switching */ + regmap_set_bits(priv->gswip, GSWIP_PCE_GCTRL_0, + GSWIP_PCE_GCTRL_0_VLAN); + + /* Flush MAC Table */ + regmap_set_bits(priv->gswip, GSWIP_PCE_GCTRL_0, + GSWIP_PCE_GCTRL_0_MTFL); + + err = gswip_switch_r_timeout(priv, GSWIP_PCE_GCTRL_0, + GSWIP_PCE_GCTRL_0_MTFL); + if (err) { + dev_err(priv->dev, "MAC flushing didn't finish\n"); + return err; + } + + ds->mtu_enforcement_ingress = true; + + return 0; +} + +static enum dsa_tag_protocol gswip_get_tag_protocol(struct dsa_switch *ds, + int port, + enum dsa_tag_protocol mp) +{ + struct gswip_priv *priv = ds->priv; + + return priv->hw_info->tag_protocol; +} + +static int gswip_vlan_active_create(struct gswip_priv *priv, + struct net_device *bridge, + int fid, u16 vid) +{ + struct gswip_pce_table_entry vlan_active = {0,}; + unsigned int max_ports = priv->hw_info->max_ports; + int idx = -1; + int err; + int i; + + /* Look for a free slot */ + for (i = max_ports; i < ARRAY_SIZE(priv->vlans); i++) { + if (!priv->vlans[i].bridge) { + idx = i; + break; + } + } + + if (idx == -1) + return -ENOSPC; + + if (fid == -1) + fid = idx; + + vlan_active.index = idx; + vlan_active.table = GSWIP_TABLE_ACTIVE_VLAN; + vlan_active.key[0] = vid; + vlan_active.val[0] = fid; + vlan_active.valid = true; + + err = gswip_pce_table_entry_write(priv, &vlan_active); + if (err) { + dev_err(priv->dev, "failed to write active VLAN: %d\n", err); + return err; + } + + priv->vlans[idx].bridge = bridge; + priv->vlans[idx].vid = vid; + priv->vlans[idx].fid = fid; + + return idx; +} + +static int gswip_vlan_active_remove(struct gswip_priv *priv, int idx) +{ + struct gswip_pce_table_entry vlan_active = {0,}; + int err; + + vlan_active.index = idx; + vlan_active.table = GSWIP_TABLE_ACTIVE_VLAN; + vlan_active.valid = false; + err = gswip_pce_table_entry_write(priv, &vlan_active); + if (err) + dev_err(priv->dev, "failed to delete active VLAN: %d\n", err); + priv->vlans[idx].bridge = NULL; + + return err; +} + +static int gswip_vlan_add(struct gswip_priv *priv, struct net_device *bridge, + int port, u16 vid, bool untagged, bool pvid, + bool vlan_aware) +{ + struct gswip_pce_table_entry vlan_mapping = {0,}; + unsigned int max_ports = priv->hw_info->max_ports; + unsigned int cpu_ports = dsa_cpu_ports(priv->ds); + bool active_vlan_created = false; + int fid = -1, idx = -1; + int i, err; + + /* Check if there is already a page for this bridge */ + for (i = max_ports; i < ARRAY_SIZE(priv->vlans); i++) { + if (priv->vlans[i].bridge == bridge) { + if (vlan_aware) { + if (fid != -1 && fid != priv->vlans[i].fid) + dev_err(priv->dev, "one bridge with multiple flow ids\n"); + fid = priv->vlans[i].fid; + } + if (priv->vlans[i].vid == vid) { + idx = i; + break; + } + } + } + + /* If this bridge is not programmed yet, add a Active VLAN table + * entry in a free slot and prepare the VLAN mapping table entry. + */ + if (idx == -1) { + idx = gswip_vlan_active_create(priv, bridge, fid, vid); + if (idx < 0) + return idx; + active_vlan_created = true; + + vlan_mapping.index = idx; + vlan_mapping.table = GSWIP_TABLE_VLAN_MAPPING; + } else { + /* Read the existing VLAN mapping entry from the switch */ + vlan_mapping.index = idx; + vlan_mapping.table = GSWIP_TABLE_VLAN_MAPPING; + err = gswip_pce_table_entry_read(priv, &vlan_mapping); + if (err) { + dev_err(priv->dev, "failed to read VLAN mapping: %d\n", + err); + return err; + } + } + + /* VLAN ID byte, maps to the VLAN ID of vlan active table */ + vlan_mapping.val[0] = vid; + /* Update the VLAN mapping entry and write it to the switch */ + vlan_mapping.val[1] |= cpu_ports; + vlan_mapping.val[1] |= BIT(port); + if (vlan_aware) + vlan_mapping.val[2] |= cpu_ports; + if (untagged) + vlan_mapping.val[2] &= ~BIT(port); + else + vlan_mapping.val[2] |= BIT(port); + err = gswip_pce_table_entry_write(priv, &vlan_mapping); + if (err) { + dev_err(priv->dev, "failed to write VLAN mapping: %d\n", err); + /* In case an Active VLAN was creaetd delete it again */ + if (active_vlan_created) + gswip_vlan_active_remove(priv, idx); + return err; + } + + gswip_port_commit_pvid(priv, port); + + return 0; +} + +static int gswip_vlan_remove(struct gswip_priv *priv, + struct net_device *bridge, int port, + u16 vid) +{ + struct gswip_pce_table_entry vlan_mapping = {0,}; + unsigned int max_ports = priv->hw_info->max_ports; + int idx = -1; + int i; + int err; + + /* Check if there is already a page for this bridge */ + for (i = max_ports; i < ARRAY_SIZE(priv->vlans); i++) { + if (priv->vlans[i].bridge == bridge && + priv->vlans[i].vid == vid) { + idx = i; + break; + } + } + + if (idx == -1) { + dev_err(priv->dev, "Port %d cannot find VID %u of bridge %s\n", + port, vid, bridge ? bridge->name : "(null)"); + return -ENOENT; + } + + vlan_mapping.index = idx; + vlan_mapping.table = GSWIP_TABLE_VLAN_MAPPING; + err = gswip_pce_table_entry_read(priv, &vlan_mapping); + if (err) { + dev_err(priv->dev, "failed to read VLAN mapping: %d\n", err); + return err; + } + + vlan_mapping.val[1] &= ~BIT(port); + vlan_mapping.val[2] &= ~BIT(port); + err = gswip_pce_table_entry_write(priv, &vlan_mapping); + if (err) { + dev_err(priv->dev, "failed to write VLAN mapping: %d\n", err); + return err; + } + + /* In case all ports are removed from the bridge, remove the VLAN */ + if (!(vlan_mapping.val[1] & ~dsa_cpu_ports(priv->ds))) { + err = gswip_vlan_active_remove(priv, idx); + if (err) { + dev_err(priv->dev, "failed to write active VLAN: %d\n", + err); + return err; + } + } + + gswip_port_commit_pvid(priv, port); + + return 0; +} + +static int gswip_port_bridge_join(struct dsa_switch *ds, int port, + struct dsa_bridge bridge, + bool *tx_fwd_offload, + struct netlink_ext_ack *extack) +{ + struct net_device *br = bridge.dev; + struct gswip_priv *priv = ds->priv; + int err; + + /* Set up the VLAN for VLAN-unaware bridging for this port, and remove + * it from the "single-port bridge" through which it was operating as + * standalone. + */ + err = gswip_vlan_add(priv, br, port, GSWIP_VLAN_UNAWARE_PVID, + true, true, false); + if (err) + return err; + + return gswip_add_single_port_br(priv, port, false); +} + +static void gswip_port_bridge_leave(struct dsa_switch *ds, int port, + struct dsa_bridge bridge) +{ + struct net_device *br = bridge.dev; + struct gswip_priv *priv = ds->priv; + + /* Add the port back to the "single-port bridge", and remove it from + * the VLAN-unaware PVID created for this bridge. + */ + gswip_add_single_port_br(priv, port, true); + gswip_vlan_remove(priv, br, port, GSWIP_VLAN_UNAWARE_PVID); +} + +static int gswip_port_vlan_prepare(struct dsa_switch *ds, int port, + const struct switchdev_obj_port_vlan *vlan, + struct netlink_ext_ack *extack) +{ + struct net_device *bridge = dsa_port_bridge_dev_get(dsa_to_port(ds, port)); + struct gswip_priv *priv = ds->priv; + unsigned int max_ports = priv->hw_info->max_ports; + int pos = max_ports; + int i, idx = -1; + + /* We only support VLAN filtering on bridges */ + if (!dsa_is_cpu_port(ds, port) && !bridge) + return -EOPNOTSUPP; + + /* Check if there is already a page for this VLAN */ + for (i = max_ports; i < ARRAY_SIZE(priv->vlans); i++) { + if (priv->vlans[i].bridge == bridge && + priv->vlans[i].vid == vlan->vid) { + idx = i; + break; + } + } + + /* If this VLAN is not programmed yet, we have to reserve + * one entry in the VLAN table. Make sure we start at the + * next position round. + */ + if (idx == -1) { + /* Look for a free slot */ + for (; pos < ARRAY_SIZE(priv->vlans); pos++) { + if (!priv->vlans[pos].bridge) { + idx = pos; + pos++; + break; + } + } + + if (idx == -1) { + NL_SET_ERR_MSG_MOD(extack, "No slot in VLAN table"); + return -ENOSPC; + } + } + + return 0; +} + +static int gswip_port_vlan_add(struct dsa_switch *ds, int port, + const struct switchdev_obj_port_vlan *vlan, + struct netlink_ext_ack *extack) +{ + struct net_device *bridge = dsa_port_bridge_dev_get(dsa_to_port(ds, port)); + struct gswip_priv *priv = ds->priv; + bool untagged = vlan->flags & BRIDGE_VLAN_INFO_UNTAGGED; + bool pvid = vlan->flags & BRIDGE_VLAN_INFO_PVID; + int err; + + if (vlan->vid == GSWIP_VLAN_UNAWARE_PVID) + return 0; + + err = gswip_port_vlan_prepare(ds, port, vlan, extack); + if (err) + return err; + + /* We have to receive all packets on the CPU port and should not + * do any VLAN filtering here. This is also called with bridge + * NULL and then we do not know for which bridge to configure + * this. + */ + if (dsa_is_cpu_port(ds, port)) + return 0; + + return gswip_vlan_add(priv, bridge, port, vlan->vid, untagged, pvid, + true); +} + +static int gswip_port_vlan_del(struct dsa_switch *ds, int port, + const struct switchdev_obj_port_vlan *vlan) +{ + struct net_device *bridge = dsa_port_bridge_dev_get(dsa_to_port(ds, port)); + struct gswip_priv *priv = ds->priv; + + if (vlan->vid == GSWIP_VLAN_UNAWARE_PVID) + return 0; + + /* We have to receive all packets on the CPU port and should not + * do any VLAN filtering here. This is also called with bridge + * NULL and then we do not know for which bridge to configure + * this. + */ + if (dsa_is_cpu_port(ds, port)) + return 0; + + return gswip_vlan_remove(priv, bridge, port, vlan->vid); +} + +static void gswip_port_fast_age(struct dsa_switch *ds, int port) +{ + struct gswip_priv *priv = ds->priv; + struct gswip_pce_table_entry mac_bridge = {0,}; + int i; + int err; + + for (i = 0; i < 2048; i++) { + mac_bridge.table = GSWIP_TABLE_MAC_BRIDGE; + mac_bridge.index = i; + + err = gswip_pce_table_entry_read(priv, &mac_bridge); + if (err) { + dev_err(priv->dev, "failed to read mac bridge: %d\n", + err); + return; + } + + if (!mac_bridge.valid) + continue; + + if (mac_bridge.val[1] & GSWIP_TABLE_MAC_BRIDGE_VAL1_STATIC) + continue; + + if (port != FIELD_GET(GSWIP_TABLE_MAC_BRIDGE_VAL0_PORT, + mac_bridge.val[0])) + continue; + + mac_bridge.valid = false; + err = gswip_pce_table_entry_write(priv, &mac_bridge); + if (err) { + dev_err(priv->dev, "failed to write mac bridge: %d\n", + err); + return; + } + } +} + +static void gswip_port_stp_state_set(struct dsa_switch *ds, int port, u8 state) +{ + struct gswip_priv *priv = ds->priv; + u32 stp_state; + + switch (state) { + case BR_STATE_DISABLED: + regmap_clear_bits(priv->gswip, GSWIP_SDMA_PCTRLp(port), + GSWIP_SDMA_PCTRL_EN); + return; + case BR_STATE_BLOCKING: + case BR_STATE_LISTENING: + stp_state = GSWIP_PCE_PCTRL_0_PSTATE_LISTEN; + break; + case BR_STATE_LEARNING: + stp_state = GSWIP_PCE_PCTRL_0_PSTATE_LEARNING; + break; + case BR_STATE_FORWARDING: + stp_state = GSWIP_PCE_PCTRL_0_PSTATE_FORWARDING; + break; + default: + dev_err(priv->dev, "invalid STP state: %d\n", state); + return; + } + + regmap_set_bits(priv->gswip, GSWIP_SDMA_PCTRLp(port), + GSWIP_SDMA_PCTRL_EN); + regmap_write_bits(priv->gswip, GSWIP_PCE_PCTRL_0p(port), + GSWIP_PCE_PCTRL_0_PSTATE_MASK, + stp_state); +} + +static int gswip_port_fdb(struct dsa_switch *ds, int port, + struct net_device *bridge, const unsigned char *addr, + u16 vid, bool add) +{ + struct gswip_priv *priv = ds->priv; + struct gswip_pce_table_entry mac_bridge = {0,}; + unsigned int max_ports = priv->hw_info->max_ports; + int fid = -1; + int i; + int err; + + for (i = max_ports; i < ARRAY_SIZE(priv->vlans); i++) { + if (priv->vlans[i].bridge == bridge) { + fid = priv->vlans[i].fid; + break; + } + } + + if (fid == -1) { + dev_err(priv->dev, "no FID found for bridge %s\n", + bridge->name); + return -EINVAL; + } + + mac_bridge.table = GSWIP_TABLE_MAC_BRIDGE; + mac_bridge.key_mode = true; + mac_bridge.key[0] = addr[5] | (addr[4] << 8); + mac_bridge.key[1] = addr[3] | (addr[2] << 8); + mac_bridge.key[2] = addr[1] | (addr[0] << 8); + mac_bridge.key[3] = FIELD_PREP(GSWIP_TABLE_MAC_BRIDGE_KEY3_FID, fid); + mac_bridge.val[0] = add ? BIT(port) : 0; /* port map */ + mac_bridge.val[1] = GSWIP_TABLE_MAC_BRIDGE_VAL1_STATIC; + mac_bridge.valid = add; + + err = gswip_pce_table_entry_write(priv, &mac_bridge); + if (err) + dev_err(priv->dev, "failed to write mac bridge: %d\n", err); + + return err; +} + +static int gswip_port_fdb_add(struct dsa_switch *ds, int port, + const unsigned char *addr, u16 vid, + struct dsa_db db) +{ + if (db.type != DSA_DB_BRIDGE) + return -EOPNOTSUPP; + + return gswip_port_fdb(ds, port, db.bridge.dev, addr, vid, true); +} + +static int gswip_port_fdb_del(struct dsa_switch *ds, int port, + const unsigned char *addr, u16 vid, + struct dsa_db db) +{ + if (db.type != DSA_DB_BRIDGE) + return -EOPNOTSUPP; + + return gswip_port_fdb(ds, port, db.bridge.dev, addr, vid, false); +} + +static int gswip_port_fdb_dump(struct dsa_switch *ds, int port, + dsa_fdb_dump_cb_t *cb, void *data) +{ + struct gswip_priv *priv = ds->priv; + struct gswip_pce_table_entry mac_bridge = {0,}; + unsigned char addr[ETH_ALEN]; + int i; + int err; + + for (i = 0; i < 2048; i++) { + mac_bridge.table = GSWIP_TABLE_MAC_BRIDGE; + mac_bridge.index = i; + + err = gswip_pce_table_entry_read(priv, &mac_bridge); + if (err) { + dev_err(priv->dev, + "failed to read mac bridge entry %d: %d\n", + i, err); + return err; + } + + if (!mac_bridge.valid) + continue; + + addr[5] = mac_bridge.key[0] & 0xff; + addr[4] = (mac_bridge.key[0] >> 8) & 0xff; + addr[3] = mac_bridge.key[1] & 0xff; + addr[2] = (mac_bridge.key[1] >> 8) & 0xff; + addr[1] = mac_bridge.key[2] & 0xff; + addr[0] = (mac_bridge.key[2] >> 8) & 0xff; + if (mac_bridge.val[1] & GSWIP_TABLE_MAC_BRIDGE_VAL1_STATIC) { + if (mac_bridge.val[0] & BIT(port)) { + err = cb(addr, 0, true, data); + if (err) + return err; + } + } else { + if (port == FIELD_GET(GSWIP_TABLE_MAC_BRIDGE_VAL0_PORT, + mac_bridge.val[0])) { + err = cb(addr, 0, false, data); + if (err) + return err; + } + } + } + return 0; +} + +static int gswip_port_max_mtu(struct dsa_switch *ds, int port) +{ + /* Includes 8 bytes for special header. */ + return GSWIP_MAX_PACKET_LENGTH - VLAN_ETH_HLEN - ETH_FCS_LEN; +} + +static int gswip_port_change_mtu(struct dsa_switch *ds, int port, int new_mtu) +{ + struct gswip_priv *priv = ds->priv; + + /* CPU port always has maximum mtu of user ports, so use it to set + * switch frame size, including 8 byte special header. + */ + if (dsa_is_cpu_port(ds, port)) { + new_mtu += 8; + regmap_write(priv->gswip, GSWIP_MAC_FLEN, + VLAN_ETH_HLEN + new_mtu + ETH_FCS_LEN); + } + + /* Enable MLEN for ports with non-standard MTUs, including the special + * header on the CPU port added above. + */ + if (new_mtu != ETH_DATA_LEN) + regmap_set_bits(priv->gswip, GSWIP_MAC_CTRL_2p(port), + GSWIP_MAC_CTRL_2_MLEN); + else + regmap_clear_bits(priv->gswip, GSWIP_MAC_CTRL_2p(port), + GSWIP_MAC_CTRL_2_MLEN); + + return 0; +} + +static void gswip_phylink_get_caps(struct dsa_switch *ds, int port, + struct phylink_config *config) +{ + struct gswip_priv *priv = ds->priv; + + priv->hw_info->phylink_get_caps(ds, port, config); +} + +static void gswip_port_set_link(struct gswip_priv *priv, int port, bool link) +{ + u32 mdio_phy; + + if (link) + mdio_phy = GSWIP_MDIO_PHY_LINK_UP; + else + mdio_phy = GSWIP_MDIO_PHY_LINK_DOWN; + + regmap_write_bits(priv->mdio, GSWIP_MDIO_PHYp(port), + GSWIP_MDIO_PHY_LINK_MASK, mdio_phy); +} + +static void gswip_port_set_speed(struct gswip_priv *priv, int port, int speed, + phy_interface_t interface) +{ + u32 mdio_phy = 0, mii_cfg = 0, mac_ctrl_0 = 0; + + switch (speed) { + case SPEED_10: + mdio_phy = GSWIP_MDIO_PHY_SPEED_M10; + + if (interface == PHY_INTERFACE_MODE_RMII) + mii_cfg = GSWIP_MII_CFG_RATE_M50; + else + mii_cfg = GSWIP_MII_CFG_RATE_M2P5; + + mac_ctrl_0 = GSWIP_MAC_CTRL_0_GMII_MII; + break; + + case SPEED_100: + mdio_phy = GSWIP_MDIO_PHY_SPEED_M100; + + if (interface == PHY_INTERFACE_MODE_RMII) + mii_cfg = GSWIP_MII_CFG_RATE_M50; + else + mii_cfg = GSWIP_MII_CFG_RATE_M25; + + mac_ctrl_0 = GSWIP_MAC_CTRL_0_GMII_MII; + break; + + case SPEED_1000: + mdio_phy = GSWIP_MDIO_PHY_SPEED_G1; + + mii_cfg = GSWIP_MII_CFG_RATE_M125; + + mac_ctrl_0 = GSWIP_MAC_CTRL_0_GMII_RGMII; + break; + } + + regmap_write_bits(priv->mdio, GSWIP_MDIO_PHYp(port), + GSWIP_MDIO_PHY_SPEED_MASK, mdio_phy); + gswip_mii_mask_cfg(priv, GSWIP_MII_CFG_RATE_MASK, mii_cfg, port); + regmap_write_bits(priv->gswip, GSWIP_MAC_CTRL_0p(port), + GSWIP_MAC_CTRL_0_GMII_MASK, mac_ctrl_0); +} + +static void gswip_port_set_duplex(struct gswip_priv *priv, int port, int duplex) +{ + u32 mac_ctrl_0, mdio_phy; + + if (duplex == DUPLEX_FULL) { + mac_ctrl_0 = GSWIP_MAC_CTRL_0_FDUP_EN; + mdio_phy = GSWIP_MDIO_PHY_FDUP_EN; + } else { + mac_ctrl_0 = GSWIP_MAC_CTRL_0_FDUP_DIS; + mdio_phy = GSWIP_MDIO_PHY_FDUP_DIS; + } + + regmap_write_bits(priv->gswip, GSWIP_MAC_CTRL_0p(port), + GSWIP_MAC_CTRL_0_FDUP_MASK, mac_ctrl_0); + regmap_write_bits(priv->mdio, GSWIP_MDIO_PHYp(port), + GSWIP_MDIO_PHY_FDUP_MASK, mdio_phy); +} + +static void gswip_port_set_pause(struct gswip_priv *priv, int port, + bool tx_pause, bool rx_pause) +{ + u32 mac_ctrl_0, mdio_phy; + + if (tx_pause && rx_pause) { + mac_ctrl_0 = GSWIP_MAC_CTRL_0_FCON_RXTX; + mdio_phy = GSWIP_MDIO_PHY_FCONTX_EN | + GSWIP_MDIO_PHY_FCONRX_EN; + } else if (tx_pause) { + mac_ctrl_0 = GSWIP_MAC_CTRL_0_FCON_TX; + mdio_phy = GSWIP_MDIO_PHY_FCONTX_EN | + GSWIP_MDIO_PHY_FCONRX_DIS; + } else if (rx_pause) { + mac_ctrl_0 = GSWIP_MAC_CTRL_0_FCON_RX; + mdio_phy = GSWIP_MDIO_PHY_FCONTX_DIS | + GSWIP_MDIO_PHY_FCONRX_EN; + } else { + mac_ctrl_0 = GSWIP_MAC_CTRL_0_FCON_NONE; + mdio_phy = GSWIP_MDIO_PHY_FCONTX_DIS | + GSWIP_MDIO_PHY_FCONRX_DIS; + } + + regmap_write_bits(priv->gswip, GSWIP_MAC_CTRL_0p(port), + GSWIP_MAC_CTRL_0_FCON_MASK, mac_ctrl_0); + regmap_write_bits(priv->mdio, GSWIP_MDIO_PHYp(port), + GSWIP_MDIO_PHY_FCONTX_MASK | GSWIP_MDIO_PHY_FCONRX_MASK, + mdio_phy); +} + +static void gswip_phylink_mac_config(struct phylink_config *config, + unsigned int mode, + const struct phylink_link_state *state) +{ + struct dsa_port *dp = dsa_phylink_to_port(config); + struct gswip_priv *priv = dp->ds->priv; + int port = dp->index; + u32 miicfg = 0; + + miicfg |= GSWIP_MII_CFG_LDCLKDIS; + + switch (state->interface) { + case PHY_INTERFACE_MODE_SGMII: + case PHY_INTERFACE_MODE_1000BASEX: + case PHY_INTERFACE_MODE_2500BASEX: + return; + case PHY_INTERFACE_MODE_MII: + case PHY_INTERFACE_MODE_INTERNAL: + miicfg |= GSWIP_MII_CFG_MODE_MIIM; + break; + case PHY_INTERFACE_MODE_REVMII: + miicfg |= GSWIP_MII_CFG_MODE_MIIP; + break; + case PHY_INTERFACE_MODE_RMII: + miicfg |= GSWIP_MII_CFG_MODE_RMIIM; + break; + case PHY_INTERFACE_MODE_RGMII: + case PHY_INTERFACE_MODE_RGMII_ID: + case PHY_INTERFACE_MODE_RGMII_RXID: + case PHY_INTERFACE_MODE_RGMII_TXID: + miicfg |= GSWIP_MII_CFG_MODE_RGMII; + break; + case PHY_INTERFACE_MODE_GMII: + miicfg |= GSWIP_MII_CFG_MODE_GMII; + break; + default: + dev_err(dp->ds->dev, + "Unsupported interface: %d\n", state->interface); + return; + } + + gswip_mii_mask_cfg(priv, + GSWIP_MII_CFG_MODE_MASK | GSWIP_MII_CFG_RMII_CLK | + GSWIP_MII_CFG_RGMII_IBS | GSWIP_MII_CFG_LDCLKDIS, + miicfg, port); + + switch (state->interface) { + case PHY_INTERFACE_MODE_RGMII_ID: + gswip_mii_mask_pcdu(priv, GSWIP_MII_PCDU_TXDLY_MASK | + GSWIP_MII_PCDU_RXDLY_MASK, 0, port); + break; + case PHY_INTERFACE_MODE_RGMII_RXID: + gswip_mii_mask_pcdu(priv, GSWIP_MII_PCDU_RXDLY_MASK, 0, port); + break; + case PHY_INTERFACE_MODE_RGMII_TXID: + gswip_mii_mask_pcdu(priv, GSWIP_MII_PCDU_TXDLY_MASK, 0, port); + break; + default: + break; + } +} + +static void gswip_phylink_mac_link_down(struct phylink_config *config, + unsigned int mode, + phy_interface_t interface) +{ + struct dsa_port *dp = dsa_phylink_to_port(config); + struct gswip_priv *priv = dp->ds->priv; + int port = dp->index; + + gswip_mii_mask_cfg(priv, GSWIP_MII_CFG_EN, 0, port); + + if (!dsa_port_is_cpu(dp)) + gswip_port_set_link(priv, port, false); +} + +static void gswip_phylink_mac_link_up(struct phylink_config *config, + struct phy_device *phydev, + unsigned int mode, + phy_interface_t interface, + int speed, int duplex, + bool tx_pause, bool rx_pause) +{ + struct dsa_port *dp = dsa_phylink_to_port(config); + struct gswip_priv *priv = dp->ds->priv; + int port = dp->index; + + if (!dsa_port_is_cpu(dp)) { + gswip_port_set_link(priv, port, true); + gswip_port_set_speed(priv, port, speed, interface); + gswip_port_set_duplex(priv, port, duplex); + gswip_port_set_pause(priv, port, tx_pause, rx_pause); + } + + gswip_mii_mask_cfg(priv, GSWIP_MII_CFG_EN, GSWIP_MII_CFG_EN, port); +} + +static void gswip_get_strings(struct dsa_switch *ds, int port, u32 stringset, + uint8_t *data) +{ + int i; + + if (stringset != ETH_SS_STATS) + return; + + for (i = 0; i < ARRAY_SIZE(gswip_rmon_cnt); i++) + ethtool_puts(&data, gswip_rmon_cnt[i].name); +} + +static u32 gswip_bcm_ram_entry_read(struct gswip_priv *priv, u32 table, + u32 index) +{ + u32 result, val; + int err; + + regmap_write(priv->gswip, GSWIP_BM_RAM_ADDR, index); + regmap_write_bits(priv->gswip, GSWIP_BM_RAM_CTRL, + GSWIP_BM_RAM_CTRL_ADDR_MASK | GSWIP_BM_RAM_CTRL_OPMOD | + GSWIP_BM_RAM_CTRL_BAS, + table | GSWIP_BM_RAM_CTRL_BAS); + + err = gswip_switch_r_timeout(priv, GSWIP_BM_RAM_CTRL, + GSWIP_BM_RAM_CTRL_BAS); + if (err) { + dev_err(priv->dev, "timeout while reading table: %u, index: %u\n", + table, index); + return 0; + } + + regmap_read(priv->gswip, GSWIP_BM_RAM_VAL(0), &result); + regmap_read(priv->gswip, GSWIP_BM_RAM_VAL(1), &val); + result |= val << 16; + + return result; +} + +static void gswip_get_ethtool_stats(struct dsa_switch *ds, int port, + uint64_t *data) +{ + struct gswip_priv *priv = ds->priv; + const struct gswip_rmon_cnt_desc *rmon_cnt; + int i; + u64 high; + + for (i = 0; i < ARRAY_SIZE(gswip_rmon_cnt); i++) { + rmon_cnt = &gswip_rmon_cnt[i]; + + data[i] = gswip_bcm_ram_entry_read(priv, port, + rmon_cnt->offset); + if (rmon_cnt->size == 2) { + high = gswip_bcm_ram_entry_read(priv, port, + rmon_cnt->offset + 1); + data[i] |= high << 32; + } + } +} + +static int gswip_get_sset_count(struct dsa_switch *ds, int port, int sset) +{ + if (sset != ETH_SS_STATS) + return 0; + + return ARRAY_SIZE(gswip_rmon_cnt); +} + +static struct phylink_pcs *gswip_phylink_mac_select_pcs(struct phylink_config *config, + phy_interface_t interface) +{ + struct dsa_port *dp = dsa_phylink_to_port(config); + struct gswip_priv *priv = dp->ds->priv; + + if (priv->hw_info->mac_select_pcs) + return priv->hw_info->mac_select_pcs(config, interface); + + return NULL; +} + +static const struct phylink_mac_ops gswip_phylink_mac_ops = { + .mac_config = gswip_phylink_mac_config, + .mac_link_down = gswip_phylink_mac_link_down, + .mac_link_up = gswip_phylink_mac_link_up, + .mac_select_pcs = gswip_phylink_mac_select_pcs, +}; + +static const struct dsa_switch_ops gswip_switch_ops = { + .get_tag_protocol = gswip_get_tag_protocol, + .setup = gswip_setup, + .port_setup = gswip_port_setup, + .port_enable = gswip_port_enable, + .port_disable = gswip_port_disable, + .port_bridge_join = gswip_port_bridge_join, + .port_bridge_leave = gswip_port_bridge_leave, + .port_fast_age = gswip_port_fast_age, + .port_vlan_filtering = gswip_port_vlan_filtering, + .port_vlan_add = gswip_port_vlan_add, + .port_vlan_del = gswip_port_vlan_del, + .port_stp_state_set = gswip_port_stp_state_set, + .port_fdb_add = gswip_port_fdb_add, + .port_fdb_del = gswip_port_fdb_del, + .port_fdb_dump = gswip_port_fdb_dump, + .port_change_mtu = gswip_port_change_mtu, + .port_max_mtu = gswip_port_max_mtu, + .phylink_get_caps = gswip_phylink_get_caps, + .get_strings = gswip_get_strings, + .get_ethtool_stats = gswip_get_ethtool_stats, + .get_sset_count = gswip_get_sset_count, +}; + +void gswip_disable_switch(struct gswip_priv *priv) +{ + regmap_clear_bits(priv->mdio, GSWIP_MDIO_GLOB, GSWIP_MDIO_GLOB_ENABLE); +} +EXPORT_SYMBOL_GPL(gswip_disable_switch); + +static int gswip_validate_cpu_port(struct dsa_switch *ds) +{ + struct gswip_priv *priv = ds->priv; + struct dsa_port *cpu_dp; + int cpu_port = -1; + + dsa_switch_for_each_cpu_port(cpu_dp, ds) { + if (cpu_port != -1) + return dev_err_probe(ds->dev, -EINVAL, + "only a single CPU port is supported\n"); + + cpu_port = cpu_dp->index; + } + + if (cpu_port == -1) + return dev_err_probe(ds->dev, -EINVAL, "no CPU port defined\n"); + + if (BIT(cpu_port) & ~priv->hw_info->allowed_cpu_ports) + return dev_err_probe(ds->dev, -EINVAL, + "unsupported CPU port defined\n"); + + return 0; +} + +int gswip_probe_common(struct gswip_priv *priv, u32 version) +{ + int err; + + mutex_init(&priv->pce_table_lock); + + priv->ds = devm_kzalloc(priv->dev, sizeof(*priv->ds), GFP_KERNEL); + if (!priv->ds) + return -ENOMEM; + + priv->ds->dev = priv->dev; + priv->ds->num_ports = priv->hw_info->max_ports; + priv->ds->ops = &gswip_switch_ops; + priv->ds->phylink_mac_ops = &gswip_phylink_mac_ops; + priv->ds->priv = priv; + + /* The hardware has the 'major/minor' version bytes in the wrong order + * preventing numerical comparisons. Construct a 16-bit unsigned integer + * having the REV field as most significant byte and the MOD field as + * least significant byte. This is effectively swapping the two bytes of + * the version variable, but other than using swab16 it doesn't affect + * the source variable. + */ + priv->version = GSWIP_VERSION_REV(version) << 8 | + GSWIP_VERSION_MOD(version); + + err = dsa_register_switch(priv->ds); + if (err) + return dev_err_probe(priv->dev, err, "dsa switch registration failed\n"); + + err = gswip_validate_cpu_port(priv->ds); + if (err) + goto disable_switch; + + dev_info(priv->dev, "probed GSWIP version %lx mod %lx\n", + GSWIP_VERSION_REV(version), GSWIP_VERSION_MOD(version)); + + return 0; + +disable_switch: + gswip_disable_switch(priv); + dsa_unregister_switch(priv->ds); + + return err; +} +EXPORT_SYMBOL_GPL(gswip_probe_common); + +MODULE_AUTHOR("Hauke Mehrtens "); +MODULE_AUTHOR("Daniel Golle "); +MODULE_DESCRIPTION("Lantiq / Intel / MaxLinear GSWIP common functions"); +MODULE_LICENSE("GPL"); From a7d4b05f9d748fef80de89d2dd650d5d2ae2f590 Mon Sep 17 00:00:00 2001 From: Daniel Golle Date: Mon, 3 Nov 2025 12:18:49 +0000 Subject: [PATCH 0569/1381] net: dsa: lantiq_gswip: support enable/disable learning Switch API 2.2 or later supports enabling or disabling learning on each port. Implement support for BR_LEARNING bridge flag and announce support for BR_LEARNING on GSWIP 2.2 or later. Signed-off-by: Daniel Golle Reviewed-by: Vladimir Oltean Link: https://patch.msgid.link/0aa4621e01c998378ad5812464bc17d23aa3bf62.1762170107.git.daniel@makrotopia.org Signed-off-by: Jakub Kicinski --- drivers/net/dsa/lantiq/lantiq_gswip.h | 3 ++ drivers/net/dsa/lantiq/lantiq_gswip_common.c | 43 ++++++++++++++++++++ 2 files changed, 46 insertions(+) diff --git a/drivers/net/dsa/lantiq/lantiq_gswip.h b/drivers/net/dsa/lantiq/lantiq_gswip.h index d86290db19b4..fb7d2c02bde9 100644 --- a/drivers/net/dsa/lantiq/lantiq_gswip.h +++ b/drivers/net/dsa/lantiq/lantiq_gswip.h @@ -157,6 +157,9 @@ #define GSWIP_PCE_PCTRL_0_PSTATE_LEARNING 0x3 #define GSWIP_PCE_PCTRL_0_PSTATE_FORWARDING 0x7 #define GSWIP_PCE_PCTRL_0_PSTATE_MASK GENMASK(2, 0) +/* Ethernet Switch PCE Port Control Register 3 */ +#define GSWIP_PCE_PCTRL_3p(p) (0x483 + ((p) * 0xA)) +#define GSWIP_PCE_PCTRL_3_LNDIS BIT(15) /* Learning Disable */ #define GSWIP_PCE_VCTRL(p) (0x485 + ((p) * 0xA)) #define GSWIP_PCE_VCTRL_UVR BIT(0) /* Unknown VLAN Rule */ #define GSWIP_PCE_VCTRL_VINR GENMASK(2, 1) /* VLAN Ingress Tag Rule */ diff --git a/drivers/net/dsa/lantiq/lantiq_gswip_common.c b/drivers/net/dsa/lantiq/lantiq_gswip_common.c index a0e361622acb..f130bf6642a7 100644 --- a/drivers/net/dsa/lantiq/lantiq_gswip_common.c +++ b/drivers/net/dsa/lantiq/lantiq_gswip_common.c @@ -403,6 +403,47 @@ static int gswip_add_single_port_br(struct gswip_priv *priv, int port, bool add) return 0; } +static int gswip_port_set_learning(struct gswip_priv *priv, int port, + bool enable) +{ + if (!GSWIP_VERSION_GE(priv, GSWIP_VERSION_2_2)) + return -EOPNOTSUPP; + + /* learning disable bit */ + return regmap_update_bits(priv->gswip, GSWIP_PCE_PCTRL_3p(port), + GSWIP_PCE_PCTRL_3_LNDIS, + enable ? 0 : GSWIP_PCE_PCTRL_3_LNDIS); +} + +static int gswip_port_pre_bridge_flags(struct dsa_switch *ds, int port, + struct switchdev_brport_flags flags, + struct netlink_ext_ack *extack) +{ + struct gswip_priv *priv = ds->priv; + unsigned long supported = 0; + + if (GSWIP_VERSION_GE(priv, GSWIP_VERSION_2_2)) + supported |= BR_LEARNING; + + if (flags.mask & ~supported) + return -EINVAL; + + return 0; +} + +static int gswip_port_bridge_flags(struct dsa_switch *ds, int port, + struct switchdev_brport_flags flags, + struct netlink_ext_ack *extack) +{ + struct gswip_priv *priv = ds->priv; + + if (flags.mask & BR_LEARNING) + return gswip_port_set_learning(priv, port, + !!(flags.val & BR_LEARNING)); + + return 0; +} + static int gswip_port_setup(struct dsa_switch *ds, int port) { struct gswip_priv *priv = ds->priv; @@ -1521,6 +1562,8 @@ static const struct dsa_switch_ops gswip_switch_ops = { .port_setup = gswip_port_setup, .port_enable = gswip_port_enable, .port_disable = gswip_port_disable, + .port_pre_bridge_flags = gswip_port_pre_bridge_flags, + .port_bridge_flags = gswip_port_bridge_flags, .port_bridge_join = gswip_port_bridge_join, .port_bridge_leave = gswip_port_bridge_leave, .port_fast_age = gswip_port_fast_age, From 9ec1fc0bf2b0ee89b3195d1c1db2cdb471a8c85d Mon Sep 17 00:00:00 2001 From: Daniel Golle Date: Mon, 3 Nov 2025 12:19:00 +0000 Subject: [PATCH 0570/1381] net: dsa: lantiq_gswip: support Energy Efficient Ethernet Introduce support for Energy Efficient Ethernet (EEE) on hardware version 2.2 or later. Signed-off-by: Daniel Golle Reviewed-by: Russell King (Oracle) Link: https://patch.msgid.link/753e45acb25e185689ca1afd8a9bd0c199d1c15b.1762170107.git.daniel@makrotopia.org Signed-off-by: Jakub Kicinski --- drivers/net/dsa/lantiq/lantiq_gswip.h | 7 +++ drivers/net/dsa/lantiq/lantiq_gswip_common.c | 47 ++++++++++++++++++++ 2 files changed, 54 insertions(+) diff --git a/drivers/net/dsa/lantiq/lantiq_gswip.h b/drivers/net/dsa/lantiq/lantiq_gswip.h index fb7d2c02bde9..56de869fc472 100644 --- a/drivers/net/dsa/lantiq/lantiq_gswip.h +++ b/drivers/net/dsa/lantiq/lantiq_gswip.h @@ -2,6 +2,7 @@ #ifndef __LANTIQ_GSWIP_H #define __LANTIQ_GSWIP_H +#include #include #include #include @@ -193,6 +194,12 @@ #define GSWIP_MAC_CTRL_2p(p) (0x905 + ((p) * 0xC)) #define GSWIP_MAC_CTRL_2_LCHKL BIT(2) /* Frame Length Check Long Enable */ #define GSWIP_MAC_CTRL_2_MLEN BIT(3) /* Maximum Untagged Frame Lnegth */ +#define GSWIP_MAC_CTRL_4p(p) (0x907 + ((p) * 0xC)) +#define GSWIP_MAC_CTRL_4_LPIEN BIT(7) /* LPI Mode Enable */ +#define GSWIP_MAC_CTRL_4_GWAIT_MASK GENMASK(14, 8) /* LPI Wait Time 1G */ +#define GSWIP_MAC_CTRL_4_GWAIT(t) u16_encode_bits((t), GSWIP_MAC_CTRL_4_GWAIT_MASK) +#define GSWIP_MAC_CTRL_4_WAIT_MASK GENMASK(6, 0) /* LPI Wait Time 100M */ +#define GSWIP_MAC_CTRL_4_WAIT(t) u16_encode_bits((t), GSWIP_MAC_CTRL_4_WAIT_MASK) /* Ethernet Switch Fetch DMA Port Control Register */ #define GSWIP_FDMA_PCTRLp(p) (0xA80 + ((p) * 0x6)) diff --git a/drivers/net/dsa/lantiq/lantiq_gswip_common.c b/drivers/net/dsa/lantiq/lantiq_gswip_common.c index f130bf6642a7..092187603dea 100644 --- a/drivers/net/dsa/lantiq/lantiq_gswip_common.c +++ b/drivers/net/dsa/lantiq/lantiq_gswip_common.c @@ -1537,6 +1537,49 @@ static int gswip_get_sset_count(struct dsa_switch *ds, int port, int sset) return ARRAY_SIZE(gswip_rmon_cnt); } +static int gswip_set_mac_eee(struct dsa_switch *ds, int port, + struct ethtool_keee *e) +{ + if (e->tx_lpi_timer > 0x7f) + return -EINVAL; + + return 0; +} + +static void gswip_phylink_mac_disable_tx_lpi(struct phylink_config *config) +{ + struct dsa_port *dp = dsa_phylink_to_port(config); + struct gswip_priv *priv = dp->ds->priv; + + regmap_clear_bits(priv->gswip, GSWIP_MAC_CTRL_4p(dp->index), + GSWIP_MAC_CTRL_4_LPIEN); +} + +static int gswip_phylink_mac_enable_tx_lpi(struct phylink_config *config, + u32 timer, bool tx_clock_stop) +{ + struct dsa_port *dp = dsa_phylink_to_port(config); + struct gswip_priv *priv = dp->ds->priv; + + return regmap_update_bits(priv->gswip, GSWIP_MAC_CTRL_4p(dp->index), + GSWIP_MAC_CTRL_4_LPIEN | + GSWIP_MAC_CTRL_4_GWAIT_MASK | + GSWIP_MAC_CTRL_4_WAIT_MASK, + GSWIP_MAC_CTRL_4_LPIEN | + GSWIP_MAC_CTRL_4_GWAIT(timer) | + GSWIP_MAC_CTRL_4_WAIT(timer)); +} + +static bool gswip_support_eee(struct dsa_switch *ds, int port) +{ + struct gswip_priv *priv = ds->priv; + + if (GSWIP_VERSION_GE(priv, GSWIP_VERSION_2_2)) + return true; + + return false; +} + static struct phylink_pcs *gswip_phylink_mac_select_pcs(struct phylink_config *config, phy_interface_t interface) { @@ -1553,6 +1596,8 @@ static const struct phylink_mac_ops gswip_phylink_mac_ops = { .mac_config = gswip_phylink_mac_config, .mac_link_down = gswip_phylink_mac_link_down, .mac_link_up = gswip_phylink_mac_link_up, + .mac_disable_tx_lpi = gswip_phylink_mac_disable_tx_lpi, + .mac_enable_tx_lpi = gswip_phylink_mac_enable_tx_lpi, .mac_select_pcs = gswip_phylink_mac_select_pcs, }; @@ -1580,6 +1625,8 @@ static const struct dsa_switch_ops gswip_switch_ops = { .get_strings = gswip_get_strings, .get_ethtool_stats = gswip_get_ethtool_stats, .get_sset_count = gswip_get_sset_count, + .set_mac_eee = gswip_set_mac_eee, + .support_eee = gswip_support_eee, }; void gswip_disable_switch(struct gswip_priv *priv) From 3e5ef3b1709afecd86886bc8eb2425e39b09196d Mon Sep 17 00:00:00 2001 From: Daniel Golle Date: Mon, 3 Nov 2025 12:19:10 +0000 Subject: [PATCH 0571/1381] net: dsa: lantiq_gswip: set link parameters also for CPU port On standalone switch ICs the link parameters of the CPU port need to be setup just like user ports. The destinction in the driver to not carry out link parameter setup for the CPU port does make sense for in-SoC switches on which the CPU port is internally connected to the SoC's Ethernet MAC. Set link parameters also for the CPU port unless it is an internal interface. Note that the internal TP PHYs anyway cannot be used as CPU ports, hence it doesn't matter that they are now also covered by that condition. Signed-off-by: Daniel Golle Reviewed-by: Alexander Sverdlin Tested-by: Alexander Sverdlin Reviewed-by: Vladimir Oltean Link: https://patch.msgid.link/07c6b8d3a12296123be5e5938b454fc620f819e6.1762170107.git.daniel@makrotopia.org Signed-off-by: Jakub Kicinski --- drivers/net/dsa/lantiq/lantiq_gswip_common.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/dsa/lantiq/lantiq_gswip_common.c b/drivers/net/dsa/lantiq/lantiq_gswip_common.c index 092187603dea..0ac87eb23bb5 100644 --- a/drivers/net/dsa/lantiq/lantiq_gswip_common.c +++ b/drivers/net/dsa/lantiq/lantiq_gswip_common.c @@ -1459,7 +1459,7 @@ static void gswip_phylink_mac_link_up(struct phylink_config *config, struct gswip_priv *priv = dp->ds->priv; int port = dp->index; - if (!dsa_port_is_cpu(dp)) { + if (!dsa_port_is_cpu(dp) || interface != PHY_INTERFACE_MODE_INTERNAL) { gswip_port_set_link(priv, port, true); gswip_port_set_speed(priv, port, speed, interface); gswip_port_set_duplex(priv, port, duplex); From 0c56a98560c16d118d35e58a25541a3ac2717b78 Mon Sep 17 00:00:00 2001 From: Daniel Golle Date: Mon, 3 Nov 2025 12:19:25 +0000 Subject: [PATCH 0572/1381] net: dsa: lantiq_gswip: define and use GSWIP_TABLE_MAC_BRIDGE_VAL1_VALID When adding FDB entries to the MAC bridge table on GSWIP 2.2 or later it is needed to set an (undocumented) bit to mark the entry as valid. If this bit isn't set for entries in the MAC bridge table, then those entries won't be considered as valid MAC addresses. Signed-off-by: Daniel Golle Link: https://patch.msgid.link/e02fe0d946c98920bc55b5f389a8f56382aae7df.1762170107.git.daniel@makrotopia.org Signed-off-by: Jakub Kicinski --- drivers/net/dsa/lantiq/lantiq_gswip.h | 1 + drivers/net/dsa/lantiq/lantiq_gswip_common.c | 7 ++++++- 2 files changed, 7 insertions(+), 1 deletion(-) diff --git a/drivers/net/dsa/lantiq/lantiq_gswip.h b/drivers/net/dsa/lantiq/lantiq_gswip.h index 56de869fc472..42000954d842 100644 --- a/drivers/net/dsa/lantiq/lantiq_gswip.h +++ b/drivers/net/dsa/lantiq/lantiq_gswip.h @@ -224,6 +224,7 @@ #define GSWIP_TABLE_MAC_BRIDGE_KEY3_FID GENMASK(5, 0) /* Filtering identifier */ #define GSWIP_TABLE_MAC_BRIDGE_VAL0_PORT GENMASK(7, 4) /* Port on learned entries */ #define GSWIP_TABLE_MAC_BRIDGE_VAL1_STATIC BIT(0) /* Static, non-aging entry */ +#define GSWIP_TABLE_MAC_BRIDGE_VAL1_VALID BIT(1) /* Valid bit */ #define XRX200_GPHY_FW_ALIGN (16 * 1024) diff --git a/drivers/net/dsa/lantiq/lantiq_gswip_common.c b/drivers/net/dsa/lantiq/lantiq_gswip_common.c index 0ac87eb23bb5..fdfc265b4c73 100644 --- a/drivers/net/dsa/lantiq/lantiq_gswip_common.c +++ b/drivers/net/dsa/lantiq/lantiq_gswip_common.c @@ -1149,7 +1149,12 @@ static int gswip_port_fdb(struct dsa_switch *ds, int port, mac_bridge.key[2] = addr[1] | (addr[0] << 8); mac_bridge.key[3] = FIELD_PREP(GSWIP_TABLE_MAC_BRIDGE_KEY3_FID, fid); mac_bridge.val[0] = add ? BIT(port) : 0; /* port map */ - mac_bridge.val[1] = GSWIP_TABLE_MAC_BRIDGE_VAL1_STATIC; + if (GSWIP_VERSION_GE(priv, GSWIP_VERSION_2_2_ETC)) + mac_bridge.val[1] = add ? (GSWIP_TABLE_MAC_BRIDGE_VAL1_STATIC | + GSWIP_TABLE_MAC_BRIDGE_VAL1_VALID) : 0; + else + mac_bridge.val[1] = GSWIP_TABLE_MAC_BRIDGE_VAL1_STATIC; + mac_bridge.valid = add; err = gswip_pce_table_entry_write(priv, &mac_bridge); From e836824116b5644eb681777cd58cba915f4cbe75 Mon Sep 17 00:00:00 2001 From: Daniel Golle Date: Mon, 3 Nov 2025 12:19:34 +0000 Subject: [PATCH 0573/1381] dt-bindings: net: dsa: lantiq,gswip: add MaxLinear RMII refclk output property Add support for the maxlinear,rmii-refclk-out boolean property on port nodes to configure the RMII reference clock to be an output rather than an input. This property is only applicable for ports in RMII mode and allows the switch to provide the reference clock for RMII-connected PHYs instead of requiring an external clock source. This corresponds to the driver changes that read this Device Tree property to configure the RMII clock direction. Signed-off-by: Daniel Golle Reviewed-by: Alexander Sverdlin Reviewed-by: Krzysztof Kozlowski Link: https://patch.msgid.link/9813bb916ecce9bae366e6c50c081014fe5371ea.1762170107.git.daniel@makrotopia.org Signed-off-by: Jakub Kicinski --- .../bindings/net/dsa/lantiq,gswip.yaml | 18 ++++++++++++++++-- 1 file changed, 16 insertions(+), 2 deletions(-) diff --git a/Documentation/devicetree/bindings/net/dsa/lantiq,gswip.yaml b/Documentation/devicetree/bindings/net/dsa/lantiq,gswip.yaml index f3154b19af78..809d0e9d0a15 100644 --- a/Documentation/devicetree/bindings/net/dsa/lantiq,gswip.yaml +++ b/Documentation/devicetree/bindings/net/dsa/lantiq,gswip.yaml @@ -6,8 +6,7 @@ $schema: http://devicetree.org/meta-schemas/core.yaml# title: Lantiq GSWIP Ethernet switches -allOf: - - $ref: dsa.yaml#/$defs/ethernet-ports +$ref: dsa.yaml# maintainers: - Hauke Mehrtens @@ -91,6 +90,21 @@ properties: additionalProperties: false +patternProperties: + "^(ethernet-)?ports$": + type: object + patternProperties: + "^(ethernet-)?port@[0-6]$": + $ref: dsa-port.yaml# + unevaluatedProperties: false + + properties: + maxlinear,rmii-refclk-out: + type: boolean + description: + Configure the RMII reference clock to be a clock output + rather than an input. Only applicable for RMII mode. + required: - compatible - reg From 319fd7e9d446bb90469a82f876e78785f6da0bc5 Mon Sep 17 00:00:00 2001 From: Daniel Golle Date: Mon, 3 Nov 2025 12:19:47 +0000 Subject: [PATCH 0574/1381] net: dsa: lantiq_gswip: add vendor property to setup MII refclk output Read boolean Device Tree property "maxlinear,rmii-refclk-out" and switch the RMII reference clock to be a clock output rather than an input if it is set. Signed-off-by: Daniel Golle Reviewed-by: Alexander Sverdlin Tested-by: Alexander Sverdlin Link: https://patch.msgid.link/947d14970f74f760e4a60c777aabee64e7e4f356.1762170107.git.daniel@makrotopia.org Signed-off-by: Jakub Kicinski --- drivers/net/dsa/lantiq/lantiq_gswip_common.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/dsa/lantiq/lantiq_gswip_common.c b/drivers/net/dsa/lantiq/lantiq_gswip_common.c index fdfc265b4c73..7b3debd45b91 100644 --- a/drivers/net/dsa/lantiq/lantiq_gswip_common.c +++ b/drivers/net/dsa/lantiq/lantiq_gswip_common.c @@ -1402,6 +1402,8 @@ static void gswip_phylink_mac_config(struct phylink_config *config, break; case PHY_INTERFACE_MODE_RMII: miicfg |= GSWIP_MII_CFG_MODE_RMIIM; + if (of_property_read_bool(dp->dn, "maxlinear,rmii-refclk-out")) + miicfg |= GSWIP_MII_CFG_RMII_CLK; break; case PHY_INTERFACE_MODE_RGMII: case PHY_INTERFACE_MODE_RGMII_ID: From bea0c17786116141aaf3980dc73758e3cc0d2748 Mon Sep 17 00:00:00 2001 From: Daniel Golle Date: Mon, 3 Nov 2025 12:19:58 +0000 Subject: [PATCH 0575/1381] dt-bindings: net: dsa: lantiq,gswip: add support for MII delay properties Add support for standard tx-internal-delay-ps and rx-internal-delay-ps properties on port nodes to allow fine-tuning of RGMII clock delays. The GSWIP switch hardware supports delay values in 500 picosecond increments from 0 to 3500 picoseconds, with a post-reset default of 2000 picoseconds for both TX and RX delays. The driver currently sets the delay to 0 in case the PHY is setup to carry out the delay by the corresponding interface modes ("rgmii-id", "rgmii-rxid", "rgmii-txid"). This corresponds to the driver changes that allow adjusting MII delays using Device Tree properties instead of relying solely on the PHY interface mode. Signed-off-by: Daniel Golle Reviewed-by: Krzysztof Kozlowski Link: https://patch.msgid.link/9e007d4f85c2c6d69e0b91f3663d99e0f6fc8eac.1762170107.git.daniel@makrotopia.org Signed-off-by: Jakub Kicinski --- .../bindings/net/dsa/lantiq,gswip.yaml | 18 +++++++++++++++++- 1 file changed, 17 insertions(+), 1 deletion(-) diff --git a/Documentation/devicetree/bindings/net/dsa/lantiq,gswip.yaml b/Documentation/devicetree/bindings/net/dsa/lantiq,gswip.yaml index 809d0e9d0a15..929f6f8e4534 100644 --- a/Documentation/devicetree/bindings/net/dsa/lantiq,gswip.yaml +++ b/Documentation/devicetree/bindings/net/dsa/lantiq,gswip.yaml @@ -104,6 +104,20 @@ patternProperties: description: Configure the RMII reference clock to be a clock output rather than an input. Only applicable for RMII mode. + tx-internal-delay-ps: + enum: [0, 500, 1000, 1500, 2000, 2500, 3000, 3500] + description: + RGMII TX Clock Delay defined in pico seconds. + The delay lines adjust the MII clock vs. data timing. + If this property is not present the delay is determined by + the interface mode. + rx-internal-delay-ps: + enum: [0, 500, 1000, 1500, 2000, 2500, 3000, 3500] + description: + RGMII RX Clock Delay defined in pico seconds. + The delay lines adjust the MII clock vs. data timing. + If this property is not present the delay is determined by + the interface mode. required: - compatible @@ -127,8 +141,10 @@ examples: port@0 { reg = <0>; label = "lan3"; - phy-mode = "rgmii"; + phy-mode = "rgmii-id"; phy-handle = <&phy0>; + tx-internal-delay-ps = <2000>; + rx-internal-delay-ps = <2000>; }; port@1 { From cdef8e47b638bcc35b0e05f48269a2667ec665da Mon Sep 17 00:00:00 2001 From: Daniel Golle Date: Mon, 3 Nov 2025 12:20:05 +0000 Subject: [PATCH 0576/1381] net: dsa: lantiq_gswip: allow adjusting MII delays Currently the MII clk vs. data delay is configured based on the PHY interface mode. In addition to that add support for setting up MII delays using the standard Device Tree properties 'tx-internal-delay-ps' and 'rx-internal-delay-ps', using the values determined by the PHY interface mode as default to maintain backward compatibility with legacy device trees. Signed-off-by: Daniel Golle Link: https://patch.msgid.link/37203e831cff87dc46e5ef9e8cbd68fb8689773d.1762170107.git.daniel@makrotopia.org Signed-off-by: Jakub Kicinski --- drivers/net/dsa/lantiq/lantiq_gswip.h | 4 + drivers/net/dsa/lantiq/lantiq_gswip_common.c | 94 ++++++++++++-------- 2 files changed, 60 insertions(+), 38 deletions(-) diff --git a/drivers/net/dsa/lantiq/lantiq_gswip.h b/drivers/net/dsa/lantiq/lantiq_gswip.h index 42000954d842..0c32ec85e127 100644 --- a/drivers/net/dsa/lantiq/lantiq_gswip.h +++ b/drivers/net/dsa/lantiq/lantiq_gswip.h @@ -82,6 +82,10 @@ #define GSWIP_MII_PCDU5 0x05 #define GSWIP_MII_PCDU_TXDLY_MASK GENMASK(2, 0) #define GSWIP_MII_PCDU_RXDLY_MASK GENMASK(9, 7) +#define GSWIP_MII_PCDU_TXDLY(x) u16_encode_bits(((x) / 500), GSWIP_MII_PCDU_TXDLY_MASK) +#define GSWIP_MII_PCDU_RXDLY(x) u16_encode_bits(((x) / 500), GSWIP_MII_PCDU_RXDLY_MASK) +#define GSWIP_MII_PCDU_RXDLY_DEFAULT 2000 /* picoseconds */ +#define GSWIP_MII_PCDU_TXDLY_DEFAULT 2000 /* picoseconds */ /* GSWIP Core Registers */ #define GSWIP_SWRES 0x000 diff --git a/drivers/net/dsa/lantiq/lantiq_gswip_common.c b/drivers/net/dsa/lantiq/lantiq_gswip_common.c index 7b3debd45b91..122ccea4057b 100644 --- a/drivers/net/dsa/lantiq/lantiq_gswip_common.c +++ b/drivers/net/dsa/lantiq/lantiq_gswip_common.c @@ -130,30 +130,6 @@ static void gswip_mii_mask_cfg(struct gswip_priv *priv, u32 mask, u32 set, set); } -static void gswip_mii_mask_pcdu(struct gswip_priv *priv, u32 mask, u32 set, - int port) -{ - int reg_port; - - /* MII_PCDU register only exists for MII ports */ - if (!(priv->hw_info->mii_ports & BIT(port))) - return; - - reg_port = port + priv->hw_info->mii_port_reg_offset; - - switch (reg_port) { - case 0: - regmap_write_bits(priv->mii, GSWIP_MII_PCDU0, mask, set); - break; - case 1: - regmap_write_bits(priv->mii, GSWIP_MII_PCDU1, mask, set); - break; - case 5: - regmap_write_bits(priv->mii, GSWIP_MII_PCDU5, mask, set); - break; - } -} - static int gswip_mdio_poll(struct gswip_priv *priv) { u32 ctrl; @@ -622,6 +598,61 @@ static int gswip_port_vlan_filtering(struct dsa_switch *ds, int port, return 0; } +static void gswip_mii_delay_setup(struct gswip_priv *priv, struct dsa_port *dp, + phy_interface_t interface) +{ + u32 tx_delay = GSWIP_MII_PCDU_TXDLY_DEFAULT; + u32 rx_delay = GSWIP_MII_PCDU_RXDLY_DEFAULT; + struct device_node *port_dn = dp->dn; + u16 mii_pcdu_reg; + + /* As MII_PCDU registers only exist for MII ports, silently return + * unless the port is an MII port + */ + if (!(priv->hw_info->mii_ports & BIT(dp->index))) + return; + + switch (dp->index + priv->hw_info->mii_port_reg_offset) { + case 0: + mii_pcdu_reg = GSWIP_MII_PCDU0; + break; + case 1: + mii_pcdu_reg = GSWIP_MII_PCDU1; + break; + case 5: + mii_pcdu_reg = GSWIP_MII_PCDU5; + break; + default: + return; + } + + /* legacy code to set default delays according to the interface mode */ + switch (interface) { + case PHY_INTERFACE_MODE_RGMII_ID: + tx_delay = 0; + rx_delay = 0; + break; + case PHY_INTERFACE_MODE_RGMII_RXID: + rx_delay = 0; + break; + case PHY_INTERFACE_MODE_RGMII_TXID: + tx_delay = 0; + break; + default: + break; + } + + /* allow settings delays using device tree properties */ + of_property_read_u32(port_dn, "rx-internal-delay-ps", &rx_delay); + of_property_read_u32(port_dn, "tx-internal-delay-ps", &tx_delay); + + regmap_write_bits(priv->mii, mii_pcdu_reg, + GSWIP_MII_PCDU_TXDLY_MASK | + GSWIP_MII_PCDU_RXDLY_MASK, + GSWIP_MII_PCDU_TXDLY(tx_delay) | + GSWIP_MII_PCDU_RXDLY(rx_delay)); +} + static int gswip_setup(struct dsa_switch *ds) { unsigned int cpu_ports = dsa_cpu_ports(ds); @@ -1425,20 +1456,7 @@ static void gswip_phylink_mac_config(struct phylink_config *config, GSWIP_MII_CFG_RGMII_IBS | GSWIP_MII_CFG_LDCLKDIS, miicfg, port); - switch (state->interface) { - case PHY_INTERFACE_MODE_RGMII_ID: - gswip_mii_mask_pcdu(priv, GSWIP_MII_PCDU_TXDLY_MASK | - GSWIP_MII_PCDU_RXDLY_MASK, 0, port); - break; - case PHY_INTERFACE_MODE_RGMII_RXID: - gswip_mii_mask_pcdu(priv, GSWIP_MII_PCDU_RXDLY_MASK, 0, port); - break; - case PHY_INTERFACE_MODE_RGMII_TXID: - gswip_mii_mask_pcdu(priv, GSWIP_MII_PCDU_TXDLY_MASK, 0, port); - break; - default: - break; - } + gswip_mii_delay_setup(priv, dp, state->interface); } static void gswip_phylink_mac_link_down(struct phylink_config *config, From e1bb4b36a7ae0915f16abb5fd7073d2547235fa7 Mon Sep 17 00:00:00 2001 From: Daniel Golle Date: Mon, 3 Nov 2025 12:20:12 +0000 Subject: [PATCH 0577/1381] dt-bindings: net: dsa: lantiq,gswip: add support for MaxLinear GSW1xx switches Extend the Lantiq GSWIP device tree binding to also cover MaxLinear GSW1xx switches which are based on the same hardware IP but connected via MDIO instead of being memory-mapped. Add compatible strings for MaxLinear GSW120, GSW125, GSW140, GSW141, and GSW145 switches and adjust the schema to handle the different connection methods with conditional properties. Add MaxLinear GSW125 example showing MDIO-connected configuration. Signed-off-by: Daniel Golle Reviewed-by: Krzysztof Kozlowski Link: https://patch.msgid.link/fc96f1dedb2b418a63e69960356dde7f6eb86424.1762170107.git.daniel@makrotopia.org Signed-off-by: Jakub Kicinski --- .../bindings/net/dsa/lantiq,gswip.yaml | 128 +++++++++++++++++- 1 file changed, 123 insertions(+), 5 deletions(-) diff --git a/Documentation/devicetree/bindings/net/dsa/lantiq,gswip.yaml b/Documentation/devicetree/bindings/net/dsa/lantiq,gswip.yaml index 929f6f8e4534..205b683849a5 100644 --- a/Documentation/devicetree/bindings/net/dsa/lantiq,gswip.yaml +++ b/Documentation/devicetree/bindings/net/dsa/lantiq,gswip.yaml @@ -4,7 +4,12 @@ $id: http://devicetree.org/schemas/net/dsa/lantiq,gswip.yaml# $schema: http://devicetree.org/meta-schemas/core.yaml# -title: Lantiq GSWIP Ethernet switches +title: Lantiq GSWIP and MaxLinear GSW1xx Ethernet switches + +description: + Lantiq GSWIP and MaxLinear GSW1xx switches share the same hardware IP. + Lantiq switches are embedded in SoCs and accessed via memory-mapped I/O, + while MaxLinear switches are standalone ICs connected via MDIO. $ref: dsa.yaml# @@ -17,9 +22,14 @@ properties: - lantiq,xrx200-gswip - lantiq,xrx300-gswip - lantiq,xrx330-gswip + - maxlinear,gsw120 + - maxlinear,gsw125 + - maxlinear,gsw140 + - maxlinear,gsw141 + - maxlinear,gsw145 reg: - minItems: 3 + minItems: 1 maxItems: 3 reg-names: @@ -36,9 +46,6 @@ properties: compatible: const: lantiq,xrx200-mdio - required: - - compatible - gphy-fw: type: object properties: @@ -123,6 +130,30 @@ required: - compatible - reg +allOf: + - if: + properties: + compatible: + contains: + enum: + - lantiq,xrx200-gswip + - lantiq,xrx300-gswip + - lantiq,xrx330-gswip + then: + properties: + reg: + minItems: 3 + maxItems: 3 + mdio: + required: + - compatible + else: + properties: + reg: + maxItems: 1 + reg-names: false + gphy-fw: false + unevaluatedProperties: false examples: @@ -230,3 +261,90 @@ examples: }; }; }; + + - | + #include + + mdio { + #address-cells = <1>; + #size-cells = <0>; + + switch@1f { + compatible = "maxlinear,gsw125"; + reg = <0x1f>; + + ports { + #address-cells = <1>; + #size-cells = <0>; + + port@0 { + reg = <0>; + label = "lan0"; + phy-handle = <&switchphy0>; + phy-mode = "internal"; + }; + + port@1 { + reg = <1>; + label = "lan1"; + phy-handle = <&switchphy1>; + phy-mode = "internal"; + }; + + port@4 { + reg = <4>; + label = "wan"; + phy-mode = "1000base-x"; + managed = "in-band-status"; + }; + + port@5 { + reg = <5>; + phy-mode = "rgmii-id"; + tx-internal-delay-ps = <2000>; + rx-internal-delay-ps = <2000>; + ethernet = <ð0>; + + fixed-link { + speed = <1000>; + full-duplex; + }; + }; + }; + + mdio { + #address-cells = <1>; + #size-cells = <0>; + + switchphy0: switchphy@0 { + reg = <0>; + + leds { + #address-cells = <1>; + #size-cells = <0>; + + led@0 { + reg = <0>; + color = ; + function = LED_FUNCTION_LAN; + }; + }; + }; + + switchphy1: switchphy@1 { + reg = <1>; + + leds { + #address-cells = <1>; + #size-cells = <0>; + + led@0 { + reg = <0>; + color = ; + function = LED_FUNCTION_LAN; + }; + }; + }; + }; + }; + }; From c6230446b1a6f3c91effafd99f604de455da52e5 Mon Sep 17 00:00:00 2001 From: Daniel Golle Date: Mon, 3 Nov 2025 12:20:20 +0000 Subject: [PATCH 0578/1381] net: dsa: add tagging driver for MaxLinear GSW1xx switch family Add support for a new DSA tagging protocol driver for the MaxLinear GSW1xx switch family. The GSW1xx switches use a proprietary 8-byte special tag inserted between the source MAC address and the EtherType field to indicate the source and destination ports for frames traversing the CPU port. Implement the tag handling logic to insert the special tag on transmit and parse it on receive. Signed-off-by: Daniel Golle Reviewed-by: Alexander Sverdlin Tested-by: Alexander Sverdlin Link: https://patch.msgid.link/0e973ebfd9433c30c96f50670da9e9449a0d98f2.1762170107.git.daniel@makrotopia.org Signed-off-by: Jakub Kicinski --- MAINTAINERS | 3 +- include/net/dsa.h | 2 + include/uapi/linux/if_ether.h | 1 + net/dsa/Kconfig | 8 +++ net/dsa/Makefile | 1 + net/dsa/tag_mxl-gsw1xx.c | 116 ++++++++++++++++++++++++++++++++++ 6 files changed, 130 insertions(+), 1 deletion(-) create mode 100644 net/dsa/tag_mxl-gsw1xx.c diff --git a/MAINTAINERS b/MAINTAINERS index 12cd8a5ab274..0dc4aa37d903 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -14053,7 +14053,7 @@ F: tools/testing/selftests/landlock/ K: landlock K: LANDLOCK -LANTIQ / INTEL Ethernet drivers +LANTIQ / MAXLINEAR / INTEL Ethernet DSA drivers M: Hauke Mehrtens L: netdev@vger.kernel.org S: Maintained @@ -14061,6 +14061,7 @@ F: Documentation/devicetree/bindings/net/dsa/lantiq,gswip.yaml F: drivers/net/dsa/lantiq/* F: drivers/net/ethernet/lantiq_xrx200.c F: net/dsa/tag_gswip.c +F: net/dsa/tag_mxl-gsw1xx.c LANTIQ MIPS ARCHITECTURE M: John Crispin diff --git a/include/net/dsa.h b/include/net/dsa.h index 67762fdaf3c7..2df2e2ead9a8 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -56,6 +56,7 @@ struct tc_action; #define DSA_TAG_PROTO_VSC73XX_8021Q_VALUE 28 #define DSA_TAG_PROTO_BRCM_LEGACY_FCS_VALUE 29 #define DSA_TAG_PROTO_YT921X_VALUE 30 +#define DSA_TAG_PROTO_MXL_GSW1XX_VALUE 31 enum dsa_tag_protocol { DSA_TAG_PROTO_NONE = DSA_TAG_PROTO_NONE_VALUE, @@ -89,6 +90,7 @@ enum dsa_tag_protocol { DSA_TAG_PROTO_LAN937X = DSA_TAG_PROTO_LAN937X_VALUE, DSA_TAG_PROTO_VSC73XX_8021Q = DSA_TAG_PROTO_VSC73XX_8021Q_VALUE, DSA_TAG_PROTO_YT921X = DSA_TAG_PROTO_YT921X_VALUE, + DSA_TAG_PROTO_MXL_GSW1XX = DSA_TAG_PROTO_MXL_GSW1XX_VALUE, }; struct dsa_switch; diff --git a/include/uapi/linux/if_ether.h b/include/uapi/linux/if_ether.h index cfd200c87e5e..2c93b7b731c8 100644 --- a/include/uapi/linux/if_ether.h +++ b/include/uapi/linux/if_ether.h @@ -92,6 +92,7 @@ #define ETH_P_ETHERCAT 0x88A4 /* EtherCAT */ #define ETH_P_8021AD 0x88A8 /* 802.1ad Service VLAN */ #define ETH_P_802_EX1 0x88B5 /* 802.1 Local Experimental 1. */ +#define ETH_P_MXLGSW 0x88C3 /* MaxLinear GSW DSA [ NOT AN OFFICIALLY REGISTERED ID ] */ #define ETH_P_PREAUTH 0x88C7 /* 802.11 Preauthentication */ #define ETH_P_TIPC 0x88CA /* TIPC */ #define ETH_P_LLDP 0x88CC /* Link Layer Discovery Protocol */ diff --git a/net/dsa/Kconfig b/net/dsa/Kconfig index 6b94028b1fcc..f86b30742122 100644 --- a/net/dsa/Kconfig +++ b/net/dsa/Kconfig @@ -104,6 +104,14 @@ config NET_DSA_TAG_MTK Say Y or M if you want to enable support for tagging frames for Mediatek switches. +config NET_DSA_TAG_MXL_GSW1XX + tristate "Tag driver for MaxLinear GSW1xx switches" + help + The GSW1xx family of switches supports an 8-byte special tag which + can be used on the CPU port of the switch. + Say Y or M if you want to enable support for tagging frames for + MaxLinear GSW1xx switches. + config NET_DSA_TAG_KSZ tristate "Tag driver for Microchip 8795/937x/9477/9893 families of switches" help diff --git a/net/dsa/Makefile b/net/dsa/Makefile index 4b011a1d5c87..42d173f5a701 100644 --- a/net/dsa/Makefile +++ b/net/dsa/Makefile @@ -28,6 +28,7 @@ obj-$(CONFIG_NET_DSA_TAG_HELLCREEK) += tag_hellcreek.o obj-$(CONFIG_NET_DSA_TAG_KSZ) += tag_ksz.o obj-$(CONFIG_NET_DSA_TAG_LAN9303) += tag_lan9303.o obj-$(CONFIG_NET_DSA_TAG_MTK) += tag_mtk.o +obj-$(CONFIG_NET_DSA_TAG_MXL_GSW1XX) += tag_mxl-gsw1xx.o obj-$(CONFIG_NET_DSA_TAG_NONE) += tag_none.o obj-$(CONFIG_NET_DSA_TAG_OCELOT) += tag_ocelot.o obj-$(CONFIG_NET_DSA_TAG_OCELOT_8021Q) += tag_ocelot_8021q.o diff --git a/net/dsa/tag_mxl-gsw1xx.c b/net/dsa/tag_mxl-gsw1xx.c new file mode 100644 index 000000000000..701a079955f2 --- /dev/null +++ b/net/dsa/tag_mxl-gsw1xx.c @@ -0,0 +1,116 @@ +// SPDX-License-Identifier: GPL-2.0+ +/* + * DSA driver Special Tag support for MaxLinear GSW1xx switch chips + * + * Copyright (C) 2025 Daniel Golle + * Copyright (C) 2023 - 2024 MaxLinear Inc. + */ + +#include +#include +#include +#include + +#include "tag.h" + +/* To define the outgoing port and to discover the incoming port a special + * tag is used by the GSW1xx. + * + * Dest MAC Src MAC special TAG EtherType + * ...| 1 2 3 4 5 6 | 1 2 3 4 5 6 | 1 2 3 4 5 6 7 8 | 1 2 |... + * |<--------------->| + */ + +#define GSW1XX_TAG_NAME "gsw1xx" + +/* special tag header length (RX and TX) */ +#define GSW1XX_HEADER_LEN 8 + +/* Word 0 = Ethertype -> 0x88C3 */ + +/* Word 1 */ +#define GSW1XX_TX_PORT_MAP GENMASK(7, 0) +#define GSW1XX_TX_PORT_MAP_EN BIT(15) +#define GSW1XX_TX_CLASS_EN BIT(14) +#define GSW1XX_TX_TIME_STAMP_EN BIT(13) +#define GSW1XX_TX_LRN_DIS BIT(12) +#define GSW1XX_TX_CLASS GENMASK(11, 8) + +/* special tag in RX path header */ +/* Word 2 */ +#define GSW1XX_RX_PORT_MAP GENMASK(15, 8) + +static struct sk_buff *gsw1xx_tag_xmit(struct sk_buff *skb, + struct net_device *dev) +{ + struct dsa_port *dp = dsa_user_to_port(dev); + __be16 *gsw1xx_tag; + + /* provide additional space 'GSW1XX_HEADER_LEN' bytes */ + skb_push(skb, GSW1XX_HEADER_LEN); + + /* add space between MAC address and Ethertype */ + dsa_alloc_etype_header(skb, GSW1XX_HEADER_LEN); + + /* special tag ingress */ + gsw1xx_tag = dsa_etype_header_pos_tx(skb); + gsw1xx_tag[0] = htons(ETH_P_MXLGSW); + gsw1xx_tag[1] = htons(GSW1XX_TX_PORT_MAP_EN | GSW1XX_TX_LRN_DIS | + FIELD_PREP(GSW1XX_TX_PORT_MAP, BIT(dp->index))); + + gsw1xx_tag[2] = 0; + gsw1xx_tag[3] = 0; + + return skb; +} + +static struct sk_buff *gsw1xx_tag_rcv(struct sk_buff *skb, + struct net_device *dev) +{ + int port; + __be16 *gsw1xx_tag; + + if (unlikely(!pskb_may_pull(skb, GSW1XX_HEADER_LEN))) { + dev_warn_ratelimited(&dev->dev, "Dropping packet, cannot pull SKB\n"); + return NULL; + } + + gsw1xx_tag = dsa_etype_header_pos_rx(skb); + + if (unlikely(ntohs(gsw1xx_tag[0]) != ETH_P_MXLGSW)) { + dev_warn_ratelimited(&dev->dev, "Dropping packet due to invalid special tag\n"); + dev_warn_ratelimited(&dev->dev, "Tag: %8ph\n", gsw1xx_tag); + return NULL; + } + + /* Get source port information */ + port = FIELD_GET(GSW1XX_RX_PORT_MAP, ntohs(gsw1xx_tag[1])); + skb->dev = dsa_conduit_find_user(dev, 0, port); + if (!skb->dev) { + dev_warn_ratelimited(&dev->dev, "Dropping packet due to invalid source port\n"); + dev_warn_ratelimited(&dev->dev, "Tag: %8ph\n", gsw1xx_tag); + return NULL; + } + + /* remove the GSW1xx special tag between MAC addresses and the current + * ethertype field. + */ + skb_pull_rcsum(skb, GSW1XX_HEADER_LEN); + dsa_strip_etype_header(skb, GSW1XX_HEADER_LEN); + + return skb; +} + +static const struct dsa_device_ops gsw1xx_netdev_ops = { + .name = GSW1XX_TAG_NAME, + .proto = DSA_TAG_PROTO_MXL_GSW1XX, + .xmit = gsw1xx_tag_xmit, + .rcv = gsw1xx_tag_rcv, + .needed_headroom = GSW1XX_HEADER_LEN, +}; + +MODULE_DESCRIPTION("DSA tag driver for MaxLinear GSW1xx 8 byte protocol"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS_DSA_TAG_DRIVER(DSA_TAG_PROTO_MXL_GSW1XX, GSW1XX_TAG_NAME); + +module_dsa_tag_driver(gsw1xx_netdev_ops); From 22335939ec907cca26be41f10f6cc01f0df8b0e9 Mon Sep 17 00:00:00 2001 From: Daniel Golle Date: Mon, 3 Nov 2025 12:20:28 +0000 Subject: [PATCH 0579/1381] net: dsa: add driver for MaxLinear GSW1xx switch family Add driver for the MaxLinear GSW1xx family of Ethernet switch ICs which are based on the same IP as the Lantiq/Intel GSWIP found in the Lantiq VR9 and Intel GRX MIPS router SoCs. The main difference is that instead of using memory-mapped I/O to communicate with the host CPU these ICs are connected via MDIO (or SPI, which isn't supported by this driver). Implement the regmap API to access the switch registers over MDIO to allow reusing lantiq_gswip_common for all core functionality. The GSW1xx also comes with a SerDes port capable of 1000Base-X, SGMII and 2500Base-X, which can either be used to connect an external PHY or SFP cage, or as the CPU port. Support for the SerDes interface is implemented in this driver using the phylink_pcs interface. Signed-off-by: Daniel Golle Tested-by: Alexander Sverdlin Link: https://patch.msgid.link/b567ec1b4beb08fd37abf18b280c56d5d8253c26.1762170107.git.daniel@makrotopia.org Signed-off-by: Jakub Kicinski --- drivers/net/dsa/lantiq/Kconfig | 12 + drivers/net/dsa/lantiq/Makefile | 1 + drivers/net/dsa/lantiq/lantiq_gswip.h | 1 + drivers/net/dsa/lantiq/mxl-gsw1xx.c | 733 ++++++++++++++++++++++++ drivers/net/dsa/lantiq/mxl-gsw1xx.h | 126 ++++ drivers/net/dsa/lantiq/mxl-gsw1xx_pce.h | 154 +++++ 6 files changed, 1027 insertions(+) create mode 100644 drivers/net/dsa/lantiq/mxl-gsw1xx.c create mode 100644 drivers/net/dsa/lantiq/mxl-gsw1xx.h create mode 100644 drivers/net/dsa/lantiq/mxl-gsw1xx_pce.h diff --git a/drivers/net/dsa/lantiq/Kconfig b/drivers/net/dsa/lantiq/Kconfig index 78db82a47d09..4a9771be5d58 100644 --- a/drivers/net/dsa/lantiq/Kconfig +++ b/drivers/net/dsa/lantiq/Kconfig @@ -10,3 +10,15 @@ config NET_DSA_LANTIQ_GSWIP help This enables support for the Lantiq / Intel GSWIP 2.1 found in the xrx200 / VR9 SoC. + +config NET_DSA_MXL_GSW1XX + tristate "MaxLinear GSW1xx Ethernet switch support" + select NET_DSA_TAG_MXL_GSW1XX + select NET_DSA_LANTIQ_COMMON + help + This enables support for the MaxLinear GSW1xx family of 1GE switches + GSW120 4 port, 2 PHYs, RGMII & SGMII/2500Base-X + GSW125 4 port, 2 PHYs, RGMII & SGMII/2500Base-X, industrial temperature + GSW140 6 port, 4 PHYs, RGMII & SGMII/2500Base-X + GSW141 6 port, 4 PHYs, RGMII & SGMII + GSW145 6 port, 4 PHYs, RGMII & SGMII/2500Base-X, industrial temperature diff --git a/drivers/net/dsa/lantiq/Makefile b/drivers/net/dsa/lantiq/Makefile index 65ffa7bb71aa..85fce605310b 100644 --- a/drivers/net/dsa/lantiq/Makefile +++ b/drivers/net/dsa/lantiq/Makefile @@ -1,2 +1,3 @@ obj-$(CONFIG_NET_DSA_LANTIQ_GSWIP) += lantiq_gswip.o obj-$(CONFIG_NET_DSA_LANTIQ_COMMON) += lantiq_gswip_common.o +obj-$(CONFIG_NET_DSA_MXL_GSW1XX) += mxl-gsw1xx.o diff --git a/drivers/net/dsa/lantiq/lantiq_gswip.h b/drivers/net/dsa/lantiq/lantiq_gswip.h index 0c32ec85e127..9c38e51a75e8 100644 --- a/drivers/net/dsa/lantiq/lantiq_gswip.h +++ b/drivers/net/dsa/lantiq/lantiq_gswip.h @@ -255,6 +255,7 @@ struct gswip_hw_info { unsigned int allowed_cpu_ports; unsigned int mii_ports; int mii_port_reg_offset; + bool supports_2500m; const struct gswip_pce_microcode (*pce_microcode)[]; size_t pce_microcode_size; enum dsa_tag_protocol tag_protocol; diff --git a/drivers/net/dsa/lantiq/mxl-gsw1xx.c b/drivers/net/dsa/lantiq/mxl-gsw1xx.c new file mode 100644 index 000000000000..0816c61a47f1 --- /dev/null +++ b/drivers/net/dsa/lantiq/mxl-gsw1xx.c @@ -0,0 +1,733 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* DSA Driver for MaxLinear GSW1xx switch devices + * + * Copyright (C) 2025 Daniel Golle + * Copyright (C) 2023 - 2024 MaxLinear Inc. + * Copyright (C) 2022 Snap One, LLC. All rights reserved. + * Copyright (C) 2017 - 2019 Hauke Mehrtens + * Copyright (C) 2012 John Crispin + * Copyright (C) 2010 Lantiq Deutschland + */ + +#include +#include +#include +#include +#include +#include +#include + +#include "lantiq_gswip.h" +#include "mxl-gsw1xx.h" +#include "mxl-gsw1xx_pce.h" + +struct gsw1xx_priv { + struct mdio_device *mdio_dev; + int smdio_badr; + struct regmap *sgmii; + struct regmap *gpio; + struct regmap *clk; + struct regmap *shell; + struct phylink_pcs pcs; + phy_interface_t tbi_interface; + struct gswip_priv gswip; +}; + +static int gsw1xx_config_smdio_badr(struct gsw1xx_priv *priv, + unsigned int reg) +{ + struct mii_bus *bus = priv->mdio_dev->bus; + int sw_addr = priv->mdio_dev->addr; + int smdio_badr = priv->smdio_badr; + int res; + + if (smdio_badr == GSW1XX_SMDIO_BADR_UNKNOWN || + reg - smdio_badr >= GSW1XX_SMDIO_BADR || + smdio_badr > reg) { + /* Configure the Switch Base Address */ + smdio_badr = reg & ~GENMASK(3, 0); + res = __mdiobus_write(bus, sw_addr, GSW1XX_SMDIO_BADR, smdio_badr); + if (res < 0) { + dev_err(&priv->mdio_dev->dev, + "%s: Error %d, configuring switch base\n", + __func__, res); + return res; + } + priv->smdio_badr = smdio_badr; + } + + return smdio_badr; +} + +static int gsw1xx_regmap_read(void *context, unsigned int reg, + unsigned int *val) +{ + struct gsw1xx_priv *priv = context; + struct mii_bus *bus = priv->mdio_dev->bus; + int sw_addr = priv->mdio_dev->addr; + int smdio_badr; + int res; + + smdio_badr = gsw1xx_config_smdio_badr(priv, reg); + if (smdio_badr < 0) + return smdio_badr; + + res = __mdiobus_read(bus, sw_addr, reg - smdio_badr); + if (res < 0) { + dev_err(&priv->mdio_dev->dev, "%s: Error %d reading 0x%x\n", + __func__, res, reg); + return res; + } + + *val = res; + + return 0; +} + +static int gsw1xx_regmap_write(void *context, unsigned int reg, + unsigned int val) +{ + struct gsw1xx_priv *priv = context; + struct mii_bus *bus = priv->mdio_dev->bus; + int sw_addr = priv->mdio_dev->addr; + int smdio_badr; + int res; + + smdio_badr = gsw1xx_config_smdio_badr(priv, reg); + if (smdio_badr < 0) + return smdio_badr; + + res = __mdiobus_write(bus, sw_addr, reg - smdio_badr, val); + if (res < 0) + dev_err(&priv->mdio_dev->dev, + "%s: Error %d, writing 0x%x:0x%x\n", __func__, res, reg, + val); + + return res; +} + +static const struct regmap_bus gsw1xx_regmap_bus = { + .reg_write = gsw1xx_regmap_write, + .reg_read = gsw1xx_regmap_read, +}; + +static void gsw1xx_mdio_regmap_lock(void *mdio_lock) +{ + mutex_lock_nested(mdio_lock, MDIO_MUTEX_NESTED); +} + +static void gsw1xx_mdio_regmap_unlock(void *mdio_lock) +{ + mutex_unlock(mdio_lock); +} + +static unsigned int gsw1xx_pcs_inband_caps(struct phylink_pcs *pcs, + phy_interface_t interface) +{ + return LINK_INBAND_DISABLE | LINK_INBAND_ENABLE; +} + +static struct gsw1xx_priv *pcs_to_gsw1xx(struct phylink_pcs *pcs) +{ + return container_of(pcs, struct gsw1xx_priv, pcs); +} + +static int gsw1xx_pcs_enable(struct phylink_pcs *pcs) +{ + struct gsw1xx_priv *priv = pcs_to_gsw1xx(pcs); + + /* Deassert SGMII shell reset */ + return regmap_clear_bits(priv->shell, GSW1XX_SHELL_RST_REQ, + GSW1XX_RST_REQ_SGMII_SHELL); +} + +static void gsw1xx_pcs_disable(struct phylink_pcs *pcs) +{ + struct gsw1xx_priv *priv = pcs_to_gsw1xx(pcs); + + /* Assert SGMII shell reset */ + regmap_set_bits(priv->shell, GSW1XX_SHELL_RST_REQ, + GSW1XX_RST_REQ_SGMII_SHELL); + + priv->tbi_interface = PHY_INTERFACE_MODE_NA; +} + +static void gsw1xx_pcs_get_state(struct phylink_pcs *pcs, + unsigned int neg_mode, + struct phylink_link_state *state) +{ + struct gsw1xx_priv *priv = pcs_to_gsw1xx(pcs); + int ret; + u32 val; + + ret = regmap_read(priv->sgmii, GSW1XX_SGMII_TBI_TBISTAT, &val); + if (ret < 0) + return; + + state->link = !!(val & GSW1XX_SGMII_TBI_TBISTAT_LINK); + state->an_complete = !!(val & GSW1XX_SGMII_TBI_TBISTAT_AN_COMPLETE); + + ret = regmap_read(priv->sgmii, GSW1XX_SGMII_TBI_LPSTAT, &val); + if (ret < 0) + return; + + state->duplex = (val & GSW1XX_SGMII_TBI_LPSTAT_DUPLEX) ? + DUPLEX_FULL : DUPLEX_HALF; + if (val & GSW1XX_SGMII_TBI_LPSTAT_PAUSE_RX) + state->pause |= MLO_PAUSE_RX; + + if (val & GSW1XX_SGMII_TBI_LPSTAT_PAUSE_TX) + state->pause |= MLO_PAUSE_TX; + + switch (FIELD_GET(GSW1XX_SGMII_TBI_LPSTAT_SPEED, val)) { + case GSW1XX_SGMII_TBI_LPSTAT_SPEED_10: + state->speed = SPEED_10; + break; + case GSW1XX_SGMII_TBI_LPSTAT_SPEED_100: + state->speed = SPEED_100; + break; + case GSW1XX_SGMII_TBI_LPSTAT_SPEED_1000: + state->speed = SPEED_1000; + break; + case GSW1XX_SGMII_TBI_LPSTAT_SPEED_NOSGMII: + if (state->interface == PHY_INTERFACE_MODE_1000BASEX) + state->speed = SPEED_1000; + else if (state->interface == PHY_INTERFACE_MODE_2500BASEX) + state->speed = SPEED_2500; + else + state->speed = SPEED_UNKNOWN; + break; + } +} + +static int gsw1xx_pcs_phy_xaui_write(struct gsw1xx_priv *priv, u16 addr, + u16 data) +{ + int ret, val; + + ret = regmap_write(priv->sgmii, GSW1XX_SGMII_PHY_D, data); + if (ret < 0) + return ret; + + ret = regmap_write(priv->sgmii, GSW1XX_SGMII_PHY_A, addr); + if (ret < 0) + return ret; + + ret = regmap_write(priv->sgmii, GSW1XX_SGMII_PHY_C, + GSW1XX_SGMII_PHY_WRITE | + GSW1XX_SGMII_PHY_RESET_N); + if (ret < 0) + return ret; + + return regmap_read_poll_timeout(priv->sgmii, GSW1XX_SGMII_PHY_C, + val, val & GSW1XX_SGMII_PHY_STATUS, + 1000, 100000); +} + +static int gsw1xx_pcs_reset(struct gsw1xx_priv *priv) +{ + int ret; + u16 val; + + /* Assert and deassert SGMII shell reset */ + ret = regmap_set_bits(priv->shell, GSW1XX_SHELL_RST_REQ, + GSW1XX_RST_REQ_SGMII_SHELL); + if (ret < 0) + return ret; + + ret = regmap_clear_bits(priv->shell, GSW1XX_SHELL_RST_REQ, + GSW1XX_RST_REQ_SGMII_SHELL); + if (ret < 0) + return ret; + + /* Hardware Bringup FSM Enable */ + ret = regmap_write(priv->sgmii, GSW1XX_SGMII_PHY_HWBU_CTRL, + GSW1XX_SGMII_PHY_HWBU_CTRL_EN_HWBU_FSM | + GSW1XX_SGMII_PHY_HWBU_CTRL_HW_FSM_EN); + if (ret < 0) + return ret; + + /* Configure SGMII PHY Receiver */ + val = FIELD_PREP(GSW1XX_SGMII_PHY_RX0_CFG2_EQ, + GSW1XX_SGMII_PHY_RX0_CFG2_EQ_DEF) | + GSW1XX_SGMII_PHY_RX0_CFG2_LOS_EN | + GSW1XX_SGMII_PHY_RX0_CFG2_TERM_EN | + FIELD_PREP(GSW1XX_SGMII_PHY_RX0_CFG2_FILT_CNT, + GSW1XX_SGMII_PHY_RX0_CFG2_FILT_CNT_DEF); + + /* TODO: Take care of inverted RX pair once generic property is + * available + */ + + ret = regmap_write(priv->sgmii, GSW1XX_SGMII_PHY_RX0_CFG2, val); + if (ret < 0) + return ret; + + val = FIELD_PREP(GSW1XX_SGMII_PHY_TX0_CFG3_VBOOST_LEVEL, + GSW1XX_SGMII_PHY_TX0_CFG3_VBOOST_LEVEL_DEF); + + /* TODO: Take care of inverted TX pair once generic property is + * available + */ + + ret = regmap_write(priv->sgmii, GSW1XX_SGMII_PHY_TX0_CFG3, val); + if (ret < 0) + return ret; + + /* Reset and Release TBI */ + val = GSW1XX_SGMII_TBI_TBICTL_INITTBI | GSW1XX_SGMII_TBI_TBICTL_ENTBI | + GSW1XX_SGMII_TBI_TBICTL_CRSTRR | GSW1XX_SGMII_TBI_TBICTL_CRSOFF; + ret = regmap_write(priv->sgmii, GSW1XX_SGMII_TBI_TBICTL, val); + if (ret < 0) + return ret; + val &= ~GSW1XX_SGMII_TBI_TBICTL_INITTBI; + ret = regmap_write(priv->sgmii, GSW1XX_SGMII_TBI_TBICTL, val); + if (ret < 0) + return ret; + + /* Release Tx Data Buffers */ + ret = regmap_set_bits(priv->sgmii, GSW1XX_SGMII_PCS_TXB_CTL, + GSW1XX_SGMII_PCS_TXB_CTL_INIT_TX_TXB); + if (ret < 0) + return ret; + ret = regmap_clear_bits(priv->sgmii, GSW1XX_SGMII_PCS_TXB_CTL, + GSW1XX_SGMII_PCS_TXB_CTL_INIT_TX_TXB); + if (ret < 0) + return ret; + + /* Release Rx Data Buffers */ + ret = regmap_set_bits(priv->sgmii, GSW1XX_SGMII_PCS_RXB_CTL, + GSW1XX_SGMII_PCS_RXB_CTL_INIT_RX_RXB); + if (ret < 0) + return ret; + return regmap_clear_bits(priv->sgmii, GSW1XX_SGMII_PCS_RXB_CTL, + GSW1XX_SGMII_PCS_RXB_CTL_INIT_RX_RXB); +} + +static int gsw1xx_pcs_config(struct phylink_pcs *pcs, unsigned int neg_mode, + phy_interface_t interface, + const unsigned long *advertising, + bool permit_pause_to_mac) +{ + struct gsw1xx_priv *priv = pcs_to_gsw1xx(pcs); + u16 txaneg, anegctl, nco_ctrl; + bool reconf = false; + int ret = 0; + + /* do not unnecessarily disrupt link and skip resetting the hardware in + * case the PCS has previously been successfully configured for this + * interface mode + */ + if (priv->tbi_interface == interface) + reconf = true; + + /* mark PCS configuration as incomplete */ + priv->tbi_interface = PHY_INTERFACE_MODE_NA; + + if (!reconf) + ret = gsw1xx_pcs_reset(priv); + + if (ret) + return ret; + + /* override bootstrap pin settings + * OVRANEG sets ANEG Mode, Enable ANEG and restart ANEG to be + * taken from bits ANMODE, ANEGEN, RANEG of the ANEGCTL register. + * OVERABL sets ability bits in tx_config_reg to be taken from + * the TXANEGH and TXANEGL registers. + */ + anegctl = GSW1XX_SGMII_TBI_ANEGCTL_OVRANEG | + GSW1XX_SGMII_TBI_ANEGCTL_OVRABL; + + switch (phylink_get_link_timer_ns(interface)) { + case 10000: + anegctl |= FIELD_PREP(GSW1XX_SGMII_TBI_ANEGCTL_LT, + GSW1XX_SGMII_TBI_ANEGCTL_LT_10US); + break; + case 1600000: + anegctl |= FIELD_PREP(GSW1XX_SGMII_TBI_ANEGCTL_LT, + GSW1XX_SGMII_TBI_ANEGCTL_LT_1_6MS); + break; + case 5000000: + anegctl |= FIELD_PREP(GSW1XX_SGMII_TBI_ANEGCTL_LT, + GSW1XX_SGMII_TBI_ANEGCTL_LT_5MS); + break; + case 10000000: + anegctl |= FIELD_PREP(GSW1XX_SGMII_TBI_ANEGCTL_LT, + GSW1XX_SGMII_TBI_ANEGCTL_LT_10MS); + break; + default: + return -EINVAL; + } + + if (neg_mode & PHYLINK_PCS_NEG_INBAND) + anegctl |= GSW1XX_SGMII_TBI_ANEGCTL_ANEGEN; + + txaneg = phylink_mii_c22_pcs_encode_advertisement(interface, advertising); + + if (interface == PHY_INTERFACE_MODE_SGMII) { + /* lacking a defined reverse-SGMII interface mode this + * driver only supports SGMII (MAC side) for now + */ + anegctl |= FIELD_PREP(GSW1XX_SGMII_TBI_ANEGCTL_ANMODE, + GSW1XX_SGMII_TBI_ANEGCTL_ANMODE_SGMII_MAC); + txaneg |= ADVERTISE_LPACK; + } else if (interface == PHY_INTERFACE_MODE_1000BASEX || + interface == PHY_INTERFACE_MODE_2500BASEX) { + anegctl |= FIELD_PREP(GSW1XX_SGMII_TBI_ANEGCTL_ANMODE, + GSW1XX_SGMII_TBI_ANEGCTL_ANMODE_1000BASEX); + } else { + dev_err(priv->gswip.dev, "%s: wrong interface mode %s\n", + __func__, phy_modes(interface)); + return -EINVAL; + } + + ret = regmap_write(priv->sgmii, GSW1XX_SGMII_TBI_TXANEGH, + FIELD_GET(GENMASK(15, 8), txaneg)); + if (ret < 0) + return ret; + ret = regmap_write(priv->sgmii, GSW1XX_SGMII_TBI_TXANEGL, + FIELD_GET(GENMASK(7, 0), txaneg)); + if (ret < 0) + return ret; + ret = regmap_write(priv->sgmii, GSW1XX_SGMII_TBI_ANEGCTL, anegctl); + if (ret < 0) + return ret; + + if (!reconf) { + /* setup SerDes clock speed */ + if (interface == PHY_INTERFACE_MODE_2500BASEX) + nco_ctrl = GSW1XX_SGMII_2G5 | GSW1XX_SGMII_2G5_NCO2; + else + nco_ctrl = GSW1XX_SGMII_1G | GSW1XX_SGMII_1G_NCO1; + + ret = regmap_update_bits(priv->clk, GSW1XX_CLK_NCO_CTRL, + GSW1XX_SGMII_HSP_MASK | + GSW1XX_SGMII_SEL, + nco_ctrl); + if (ret) + return ret; + + ret = gsw1xx_pcs_phy_xaui_write(priv, 0x30, 0x80); + if (ret) + return ret; + } + + /* PCS configuration has now been completed, store mode to prevent + * disrupting the link in case of future calls of this function for the + * same interface mode. + */ + priv->tbi_interface = interface; + + return 0; +} + +static void gsw1xx_pcs_an_restart(struct phylink_pcs *pcs) +{ + struct gsw1xx_priv *priv = pcs_to_gsw1xx(pcs); + + regmap_set_bits(priv->sgmii, GSW1XX_SGMII_TBI_ANEGCTL, + GSW1XX_SGMII_TBI_ANEGCTL_RANEG); +} + +static void gsw1xx_pcs_link_up(struct phylink_pcs *pcs, + unsigned int neg_mode, + phy_interface_t interface, int speed, + int duplex) +{ + struct gsw1xx_priv *priv = pcs_to_gsw1xx(pcs); + u16 lpstat; + + /* When in-band AN is enabled hardware will set lpstat */ + if (neg_mode == PHYLINK_PCS_NEG_INBAND_ENABLED) + return; + + /* Force speed and duplex settings */ + if (interface == PHY_INTERFACE_MODE_SGMII) { + if (speed == SPEED_10) + lpstat = FIELD_PREP(GSW1XX_SGMII_TBI_LPSTAT_SPEED, + GSW1XX_SGMII_TBI_LPSTAT_SPEED_10); + else if (speed == SPEED_100) + lpstat = FIELD_PREP(GSW1XX_SGMII_TBI_LPSTAT_SPEED, + GSW1XX_SGMII_TBI_LPSTAT_SPEED_100); + else + lpstat = FIELD_PREP(GSW1XX_SGMII_TBI_LPSTAT_SPEED, + GSW1XX_SGMII_TBI_LPSTAT_SPEED_1000); + } else { + lpstat = FIELD_PREP(GSW1XX_SGMII_TBI_LPSTAT_SPEED, + GSW1XX_SGMII_TBI_LPSTAT_SPEED_NOSGMII); + } + + if (duplex == DUPLEX_FULL) + lpstat |= GSW1XX_SGMII_TBI_LPSTAT_DUPLEX; + + regmap_write(priv->sgmii, GSW1XX_SGMII_TBI_LPSTAT, lpstat); +} + +static const struct phylink_pcs_ops gsw1xx_pcs_ops = { + .pcs_inband_caps = gsw1xx_pcs_inband_caps, + .pcs_enable = gsw1xx_pcs_enable, + .pcs_disable = gsw1xx_pcs_disable, + .pcs_get_state = gsw1xx_pcs_get_state, + .pcs_config = gsw1xx_pcs_config, + .pcs_an_restart = gsw1xx_pcs_an_restart, + .pcs_link_up = gsw1xx_pcs_link_up, +}; + +static void gsw1xx_phylink_get_caps(struct dsa_switch *ds, int port, + struct phylink_config *config) +{ + struct gswip_priv *priv = ds->priv; + + config->mac_capabilities = MAC_ASYM_PAUSE | MAC_SYM_PAUSE | + MAC_10 | MAC_100 | MAC_1000; + + switch (port) { + case 0: + case 1: + case 2: + case 3: + __set_bit(PHY_INTERFACE_MODE_INTERNAL, + config->supported_interfaces); + break; + case 4: /* port 4: SGMII */ + __set_bit(PHY_INTERFACE_MODE_SGMII, + config->supported_interfaces); + __set_bit(PHY_INTERFACE_MODE_1000BASEX, + config->supported_interfaces); + if (priv->hw_info->supports_2500m) { + __set_bit(PHY_INTERFACE_MODE_2500BASEX, + config->supported_interfaces); + config->mac_capabilities |= MAC_2500FD; + } + return; /* no support for EEE on SGMII port */ + case 5: /* port 5: RGMII or RMII */ + __set_bit(PHY_INTERFACE_MODE_RMII, + config->supported_interfaces); + phy_interface_set_rgmii(config->supported_interfaces); + break; + } + + config->lpi_capabilities = MAC_100FD | MAC_1000FD; + config->lpi_timer_default = 20; + memcpy(config->lpi_interfaces, config->supported_interfaces, + sizeof(config->lpi_interfaces)); +} + +static struct phylink_pcs *gsw1xx_phylink_mac_select_pcs(struct phylink_config *config, + phy_interface_t interface) +{ + struct dsa_port *dp = dsa_phylink_to_port(config); + struct gswip_priv *gswip_priv = dp->ds->priv; + struct gsw1xx_priv *gsw1xx_priv = container_of(gswip_priv, + struct gsw1xx_priv, + gswip); + + switch (dp->index) { + case GSW1XX_SGMII_PORT: + return &gsw1xx_priv->pcs; + default: + return NULL; + } +} + +static struct regmap *gsw1xx_regmap_init(struct gsw1xx_priv *priv, + const char *name, + unsigned int reg_base, + unsigned int max_register) +{ + const struct regmap_config config = { + .name = name, + .reg_bits = 16, + .val_bits = 16, + .reg_base = reg_base, + .max_register = max_register, + .lock = gsw1xx_mdio_regmap_lock, + .unlock = gsw1xx_mdio_regmap_unlock, + .lock_arg = &priv->mdio_dev->bus->mdio_lock, + }; + + return devm_regmap_init(&priv->mdio_dev->dev, &gsw1xx_regmap_bus, + priv, &config); +} + +static int gsw1xx_probe(struct mdio_device *mdiodev) +{ + struct device *dev = &mdiodev->dev; + struct gsw1xx_priv *priv; + u32 version; + int ret; + + priv = devm_kzalloc(dev, sizeof(*priv), GFP_KERNEL); + if (!priv) + return -ENOMEM; + + priv->mdio_dev = mdiodev; + priv->smdio_badr = GSW1XX_SMDIO_BADR_UNKNOWN; + + priv->gswip.dev = dev; + priv->gswip.hw_info = of_device_get_match_data(dev); + if (!priv->gswip.hw_info) + return -EINVAL; + + priv->gswip.gswip = gsw1xx_regmap_init(priv, "switch", + GSW1XX_SWITCH_BASE, 0xfff); + if (IS_ERR(priv->gswip.gswip)) + return PTR_ERR(priv->gswip.gswip); + + priv->gswip.mdio = gsw1xx_regmap_init(priv, "mdio", GSW1XX_MMDIO_BASE, + 0xff); + if (IS_ERR(priv->gswip.mdio)) + return PTR_ERR(priv->gswip.mdio); + + priv->gswip.mii = gsw1xx_regmap_init(priv, "mii", GSW1XX_RGMII_BASE, + 0xff); + if (IS_ERR(priv->gswip.mii)) + return PTR_ERR(priv->gswip.mii); + + priv->sgmii = gsw1xx_regmap_init(priv, "sgmii", GSW1XX_SGMII_BASE, + 0xfff); + if (IS_ERR(priv->sgmii)) + return PTR_ERR(priv->sgmii); + + priv->gpio = gsw1xx_regmap_init(priv, "gpio", GSW1XX_GPIO_BASE, 0xff); + if (IS_ERR(priv->gpio)) + return PTR_ERR(priv->gpio); + + priv->clk = gsw1xx_regmap_init(priv, "clk", GSW1XX_CLK_BASE, 0xff); + if (IS_ERR(priv->clk)) + return PTR_ERR(priv->clk); + + priv->shell = gsw1xx_regmap_init(priv, "shell", GSW1XX_SHELL_BASE, + 0xff); + if (IS_ERR(priv->shell)) + return PTR_ERR(priv->shell); + + priv->pcs.ops = &gsw1xx_pcs_ops; + priv->pcs.poll = true; + __set_bit(PHY_INTERFACE_MODE_SGMII, + priv->pcs.supported_interfaces); + __set_bit(PHY_INTERFACE_MODE_1000BASEX, + priv->pcs.supported_interfaces); + if (priv->gswip.hw_info->supports_2500m) + __set_bit(PHY_INTERFACE_MODE_2500BASEX, + priv->pcs.supported_interfaces); + priv->tbi_interface = PHY_INTERFACE_MODE_NA; + + /* assert SGMII reset to power down SGMII unit */ + ret = regmap_set_bits(priv->shell, GSW1XX_SHELL_RST_REQ, + GSW1XX_RST_REQ_SGMII_SHELL); + if (ret < 0) + return ret; + + /* configure GPIO pin-mux for MMDIO in case of external PHY connected to + * SGMII or RGMII as slave interface + */ + regmap_set_bits(priv->gpio, GPIO_ALTSEL0, 3); + regmap_set_bits(priv->gpio, GPIO_ALTSEL1, 3); + + ret = regmap_read(priv->gswip.gswip, GSWIP_VERSION, &version); + if (ret) + return ret; + + ret = gswip_probe_common(&priv->gswip, version); + if (ret) + return ret; + + dev_set_drvdata(dev, &priv->gswip); + + return 0; +} + +static void gsw1xx_remove(struct mdio_device *mdiodev) +{ + struct gswip_priv *priv = dev_get_drvdata(&mdiodev->dev); + + if (!priv) + return; + + gswip_disable_switch(priv); + + dsa_unregister_switch(priv->ds); +} + +static void gsw1xx_shutdown(struct mdio_device *mdiodev) +{ + struct gswip_priv *priv = dev_get_drvdata(&mdiodev->dev); + + if (!priv) + return; + + dev_set_drvdata(&mdiodev->dev, NULL); + + gswip_disable_switch(priv); +} + +static const struct gswip_hw_info gsw12x_data = { + .max_ports = GSW1XX_PORTS, + .allowed_cpu_ports = BIT(GSW1XX_MII_PORT) | BIT(GSW1XX_SGMII_PORT), + .mii_ports = BIT(GSW1XX_MII_PORT), + .mii_port_reg_offset = -GSW1XX_MII_PORT, + .mac_select_pcs = gsw1xx_phylink_mac_select_pcs, + .phylink_get_caps = &gsw1xx_phylink_get_caps, + .supports_2500m = true, + .pce_microcode = &gsw1xx_pce_microcode, + .pce_microcode_size = ARRAY_SIZE(gsw1xx_pce_microcode), + .tag_protocol = DSA_TAG_PROTO_MXL_GSW1XX, +}; + +static const struct gswip_hw_info gsw140_data = { + .max_ports = GSW1XX_PORTS, + .allowed_cpu_ports = BIT(GSW1XX_MII_PORT) | BIT(GSW1XX_SGMII_PORT), + .mii_ports = BIT(GSW1XX_MII_PORT), + .mii_port_reg_offset = -GSW1XX_MII_PORT, + .mac_select_pcs = gsw1xx_phylink_mac_select_pcs, + .phylink_get_caps = &gsw1xx_phylink_get_caps, + .supports_2500m = true, + .pce_microcode = &gsw1xx_pce_microcode, + .pce_microcode_size = ARRAY_SIZE(gsw1xx_pce_microcode), + .tag_protocol = DSA_TAG_PROTO_MXL_GSW1XX, +}; + +static const struct gswip_hw_info gsw141_data = { + .max_ports = GSW1XX_PORTS, + .allowed_cpu_ports = BIT(GSW1XX_MII_PORT) | BIT(GSW1XX_SGMII_PORT), + .mii_ports = BIT(GSW1XX_MII_PORT), + .mii_port_reg_offset = -GSW1XX_MII_PORT, + .mac_select_pcs = gsw1xx_phylink_mac_select_pcs, + .phylink_get_caps = gsw1xx_phylink_get_caps, + .pce_microcode = &gsw1xx_pce_microcode, + .pce_microcode_size = ARRAY_SIZE(gsw1xx_pce_microcode), + .tag_protocol = DSA_TAG_PROTO_MXL_GSW1XX, +}; + +/* + * GSW125 is the industrial temperature version of GSW120. + * GSW145 is the industrial temperature version of GSW140. + */ +static const struct of_device_id gsw1xx_of_match[] = { + { .compatible = "maxlinear,gsw120", .data = &gsw12x_data }, + { .compatible = "maxlinear,gsw125", .data = &gsw12x_data }, + { .compatible = "maxlinear,gsw140", .data = &gsw140_data }, + { .compatible = "maxlinear,gsw141", .data = &gsw141_data }, + { .compatible = "maxlinear,gsw145", .data = &gsw140_data }, + { /* sentinel */ }, +}; + +MODULE_DEVICE_TABLE(of, gsw1xx_of_match); + +static struct mdio_driver gsw1xx_driver = { + .probe = gsw1xx_probe, + .remove = gsw1xx_remove, + .shutdown = gsw1xx_shutdown, + .mdiodrv.driver = { + .name = "mxl-gsw1xx", + .of_match_table = gsw1xx_of_match, + }, +}; + +mdio_module_driver(gsw1xx_driver); + +MODULE_AUTHOR("Daniel Golle "); +MODULE_DESCRIPTION("Driver for MaxLinear GSW1xx ethernet switch"); +MODULE_LICENSE("GPL"); diff --git a/drivers/net/dsa/lantiq/mxl-gsw1xx.h b/drivers/net/dsa/lantiq/mxl-gsw1xx.h new file mode 100644 index 000000000000..38e03c048a26 --- /dev/null +++ b/drivers/net/dsa/lantiq/mxl-gsw1xx.h @@ -0,0 +1,126 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Register definitions for MaxLinear GSW1xx series switches + * + * Copyright (C) 2025 Daniel Golle + * Copyright (C) 2023 - 2024 MaxLinear Inc. + */ +#ifndef __MXL_GSW1XX_H +#define __MXL_GSW1XX_H + +#include + +#define GSW1XX_PORTS 6 +/* Port used for RGMII or optional RMII */ +#define GSW1XX_MII_PORT 5 +/* Port used for SGMII */ +#define GSW1XX_SGMII_PORT 4 + +#define GSW1XX_SYS_CLK_FREQ 340000000 + +/* SMDIO switch register base address */ +#define GSW1XX_SMDIO_BADR 0x1f +#define GSW1XX_SMDIO_BADR_UNKNOWN -1 + +/* GSW1XX SGMII PCS */ +#define GSW1XX_SGMII_BASE 0xd000 +#define GSW1XX_SGMII_PHY_HWBU_CTRL 0x009 +#define GSW1XX_SGMII_PHY_HWBU_CTRL_EN_HWBU_FSM BIT(0) +#define GSW1XX_SGMII_PHY_HWBU_CTRL_HW_FSM_EN BIT(3) +#define GSW1XX_SGMII_TBI_TXANEGH 0x300 +#define GSW1XX_SGMII_TBI_TXANEGL 0x301 +#define GSW1XX_SGMII_TBI_ANEGCTL 0x304 +#define GSW1XX_SGMII_TBI_ANEGCTL_LT GENMASK(1, 0) +#define GSW1XX_SGMII_TBI_ANEGCTL_LT_10US 0 +#define GSW1XX_SGMII_TBI_ANEGCTL_LT_1_6MS 1 +#define GSW1XX_SGMII_TBI_ANEGCTL_LT_5MS 2 +#define GSW1XX_SGMII_TBI_ANEGCTL_LT_10MS 3 +#define GSW1XX_SGMII_TBI_ANEGCTL_ANEGEN BIT(2) +#define GSW1XX_SGMII_TBI_ANEGCTL_RANEG BIT(3) +#define GSW1XX_SGMII_TBI_ANEGCTL_OVRABL BIT(4) +#define GSW1XX_SGMII_TBI_ANEGCTL_OVRANEG BIT(5) +#define GSW1XX_SGMII_TBI_ANEGCTL_ANMODE GENMASK(7, 6) +#define GSW1XX_SGMII_TBI_ANEGCTL_ANMODE_1000BASEX 1 +#define GSW1XX_SGMII_TBI_ANEGCTL_ANMODE_SGMII_PHY 2 +#define GSW1XX_SGMII_TBI_ANEGCTL_ANMODE_SGMII_MAC 3 +#define GSW1XX_SGMII_TBI_ANEGCTL_BCOMP BIT(15) + +#define GSW1XX_SGMII_TBI_TBICTL 0x305 +#define GSW1XX_SGMII_TBI_TBICTL_INITTBI BIT(0) +#define GSW1XX_SGMII_TBI_TBICTL_ENTBI BIT(1) +#define GSW1XX_SGMII_TBI_TBICTL_CRSTRR BIT(4) +#define GSW1XX_SGMII_TBI_TBICTL_CRSOFF BIT(5) +#define GSW1XX_SGMII_TBI_TBISTAT 0x309 +#define GSW1XX_SGMII_TBI_TBISTAT_LINK BIT(0) +#define GSW1XX_SGMII_TBI_TBISTAT_AN_COMPLETE BIT(1) +#define GSW1XX_SGMII_TBI_LPSTAT 0x30a +#define GSW1XX_SGMII_TBI_LPSTAT_DUPLEX BIT(0) +#define GSW1XX_SGMII_TBI_LPSTAT_PAUSE_RX BIT(1) +#define GSW1XX_SGMII_TBI_LPSTAT_PAUSE_TX BIT(2) +#define GSW1XX_SGMII_TBI_LPSTAT_SPEED GENMASK(6, 5) +#define GSW1XX_SGMII_TBI_LPSTAT_SPEED_10 0 +#define GSW1XX_SGMII_TBI_LPSTAT_SPEED_100 1 +#define GSW1XX_SGMII_TBI_LPSTAT_SPEED_1000 2 +#define GSW1XX_SGMII_TBI_LPSTAT_SPEED_NOSGMII 3 +#define GSW1XX_SGMII_PHY_D 0x100 +#define GSW1XX_SGMII_PHY_A 0x101 +#define GSW1XX_SGMII_PHY_C 0x102 +#define GSW1XX_SGMII_PHY_STATUS BIT(0) +#define GSW1XX_SGMII_PHY_READ BIT(4) +#define GSW1XX_SGMII_PHY_WRITE BIT(8) +#define GSW1XX_SGMII_PHY_RESET_N BIT(12) +#define GSW1XX_SGMII_PCS_RXB_CTL 0x401 +#define GSW1XX_SGMII_PCS_RXB_CTL_INIT_RX_RXB BIT(1) +#define GSW1XX_SGMII_PCS_TXB_CTL 0x404 +#define GSW1XX_SGMII_PCS_TXB_CTL_INIT_TX_TXB BIT(1) + +#define GSW1XX_SGMII_PHY_RX0_CFG2 0x004 +#define GSW1XX_SGMII_PHY_RX0_CFG2_EQ GENMASK(2, 0) +#define GSW1XX_SGMII_PHY_RX0_CFG2_EQ_DEF 2 +#define GSW1XX_SGMII_PHY_RX0_CFG2_INVERT BIT(3) +#define GSW1XX_SGMII_PHY_RX0_CFG2_LOS_EN BIT(4) +#define GSW1XX_SGMII_PHY_RX0_CFG2_TERM_EN BIT(5) +#define GSW1XX_SGMII_PHY_RX0_CFG2_FILT_CNT GENMASK(12, 6) +#define GSW1XX_SGMII_PHY_RX0_CFG2_FILT_CNT_DEF 20 + +#define GSW1XX_SGMII_PHY_TX0_CFG3 0x007 +#define GSW1XX_SGMII_PHY_TX0_CFG3_VBOOST_EN BIT(12) +#define GSW1XX_SGMII_PHY_TX0_CFG3_VBOOST_LEVEL GENMASK(11, 9) +#define GSW1XX_SGMII_PHY_TX0_CFG3_VBOOST_LEVEL_DEF 4 +#define GSW1XX_SGMII_PHY_TX0_CFG3_INVERT BIT(8) + +/* GSW1XX PDI Registers */ +#define GSW1XX_SWITCH_BASE 0xe000 + +/* GSW1XX MII Registers */ +#define GSW1XX_RGMII_BASE 0xf100 + +/* GSW1XX GPIO Registers */ +#define GSW1XX_GPIO_BASE 0xf300 +#define GPIO_ALTSEL0 0x83 +#define GPIO_ALTSEL0_EXTPHY_MUX_VAL 0x03c3 +#define GPIO_ALTSEL1 0x84 +#define GPIO_ALTSEL1_EXTPHY_MUX_VAL 0x003f + +/* MDIO bus controller */ +#define GSW1XX_MMDIO_BASE 0xf400 + +/* generic IC registers */ +#define GSW1XX_SHELL_BASE 0xfa00 +#define GSW1XX_SHELL_RST_REQ 0x01 +#define GSW1XX_RST_REQ_SGMII_SHELL BIT(5) +/* RGMII PAD Slew Control Register */ +#define GSW1XX_SHELL_RGMII_SLEW_CFG 0x78 +#define RGMII_SLEW_CFG_RX_2_5_V BIT(4) +#define RGMII_SLEW_CFG_TX_2_5_V BIT(5) + +/* SGMII clock related settings */ +#define GSW1XX_CLK_BASE 0xf900 +#define GSW1XX_CLK_NCO_CTRL 0x68 +#define GSW1XX_SGMII_HSP_MASK GENMASK(3, 2) +#define GSW1XX_SGMII_SEL BIT(1) +#define GSW1XX_SGMII_1G 0x0 +#define GSW1XX_SGMII_2G5 0xc +#define GSW1XX_SGMII_1G_NCO1 0x0 +#define GSW1XX_SGMII_2G5_NCO2 0x2 + +#endif /* __MXL_GSW1XX_H */ diff --git a/drivers/net/dsa/lantiq/mxl-gsw1xx_pce.h b/drivers/net/dsa/lantiq/mxl-gsw1xx_pce.h new file mode 100644 index 000000000000..eefcd411a340 --- /dev/null +++ b/drivers/net/dsa/lantiq/mxl-gsw1xx_pce.h @@ -0,0 +1,154 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * PCE microcode code update for driver for MaxLinear GSW1xx switch chips + * + * Copyright (C) 2023 - 2024 MaxLinear Inc. + * Copyright (C) 2022 Snap One, LLC. All rights reserved. + * Copyright (C) 2017 - 2019 Hauke Mehrtens + * Copyright (C) 2012 John Crispin + * Copyright (C) 2010 Lantiq Deutschland + */ + +#include "lantiq_gswip.h" + +#define INSTR 0 +#define IPV6 1 +#define LENACCU 2 + +/* GSWIP_2.X */ +enum { + OUT_MAC0 = 0, + OUT_MAC1, + OUT_MAC2, + OUT_MAC3, + OUT_MAC4, + OUT_MAC5, + OUT_ETHTYP, + OUT_VTAG0, + OUT_VTAG1, + OUT_ITAG0, + OUT_ITAG1, /* 10 */ + OUT_ITAG2, + OUT_ITAG3, + OUT_IP0, + OUT_IP1, + OUT_IP2, + OUT_IP3, + OUT_SIP0, + OUT_SIP1, + OUT_SIP2, + OUT_SIP3, /* 20 */ + OUT_SIP4, + OUT_SIP5, + OUT_SIP6, + OUT_SIP7, + OUT_DIP0, + OUT_DIP1, + OUT_DIP2, + OUT_DIP3, + OUT_DIP4, + OUT_DIP5, /* 30 */ + OUT_DIP6, + OUT_DIP7, + OUT_SESID, + OUT_PROT, + OUT_APP0, + OUT_APP1, + OUT_IGMP0, + OUT_IGMP1, + OUT_STAG0 = 61, + OUT_STAG1 = 62, + OUT_NONE = 63, +}; + +/* parser's microcode flag type */ +enum { + FLAG_ITAG = 0, + FLAG_VLAN, + FLAG_SNAP, + FLAG_PPPOE, + FLAG_IPV6, + FLAG_IPV6FL, + FLAG_IPV4, + FLAG_IGMP, + FLAG_TU, + FLAG_HOP, + FLAG_NN1, /* 10 */ + FLAG_NN2, + FLAG_END, + FLAG_NO, /* 13 */ + FLAG_SVLAN, /* 14 */ +}; + +#define PCE_MC_M(val, msk, ns, out, len, type, flags, ipv4_len) \ + { (val), (msk), ((ns) << 10 | (out) << 4 | (len) >> 1),\ + ((len) & 1) << 15 | (type) << 13 | (flags) << 9 | (ipv4_len) << 8 } + +/* V22_2X (IPv6 issue fixed) */ +static const struct gswip_pce_microcode gsw1xx_pce_microcode[] = { + /* value mask ns fields L type flags ipv4_len */ + PCE_MC_M(0x88c3, 0xFFFF, 1, OUT_ITAG0, 4, INSTR, FLAG_ITAG, 0), + PCE_MC_M(0x8100, 0xFFFF, 4, OUT_STAG0, 2, INSTR, FLAG_SVLAN, 0), + PCE_MC_M(0x88A8, 0xFFFF, 4, OUT_STAG0, 2, INSTR, FLAG_SVLAN, 0), + PCE_MC_M(0x9100, 0xFFFF, 4, OUT_STAG0, 2, INSTR, FLAG_SVLAN, 0), + PCE_MC_M(0x8100, 0xFFFF, 5, OUT_VTAG0, 2, INSTR, FLAG_VLAN, 0), + PCE_MC_M(0x88A8, 0xFFFF, 6, OUT_VTAG0, 2, INSTR, FLAG_VLAN, 0), + PCE_MC_M(0x9100, 0xFFFF, 4, OUT_VTAG0, 2, INSTR, FLAG_VLAN, 0), + PCE_MC_M(0x8864, 0xFFFF, 20, OUT_ETHTYP, 1, INSTR, FLAG_NO, 0), + PCE_MC_M(0x0800, 0xFFFF, 24, OUT_ETHTYP, 1, INSTR, FLAG_NO, 0), + PCE_MC_M(0x86DD, 0xFFFF, 25, OUT_ETHTYP, 1, INSTR, FLAG_NO, 0), + PCE_MC_M(0x8863, 0xFFFF, 19, OUT_ETHTYP, 1, INSTR, FLAG_NO, 0), + PCE_MC_M(0x0000, 0xF800, 13, OUT_NONE, 0, INSTR, FLAG_NO, 0), + PCE_MC_M(0x0000, 0x0000, 44, OUT_ETHTYP, 1, INSTR, FLAG_NO, 0), + PCE_MC_M(0x0600, 0x0600, 44, OUT_ETHTYP, 1, INSTR, FLAG_NO, 0), + PCE_MC_M(0x0000, 0x0000, 15, OUT_NONE, 1, INSTR, FLAG_NO, 0), + PCE_MC_M(0xAAAA, 0xFFFF, 17, OUT_NONE, 1, INSTR, FLAG_NO, 0), + PCE_MC_M(0x0000, 0x0000, 45, OUT_NONE, 0, INSTR, FLAG_NO, 0), + PCE_MC_M(0x0300, 0xFF00, 45, OUT_NONE, 0, INSTR, FLAG_SNAP, 0), + PCE_MC_M(0x0000, 0x0000, 45, OUT_NONE, 0, INSTR, FLAG_NO, 0), + PCE_MC_M(0x0000, 0x0000, 45, OUT_DIP7, 3, INSTR, FLAG_NO, 0), + PCE_MC_M(0x0000, 0x0000, 21, OUT_DIP7, 3, INSTR, FLAG_PPPOE, 0), + PCE_MC_M(0x0021, 0xFFFF, 24, OUT_NONE, 1, INSTR, FLAG_NO, 0), + PCE_MC_M(0x0057, 0xFFFF, 25, OUT_NONE, 1, INSTR, FLAG_NO, 0), + PCE_MC_M(0x0000, 0x0000, 44, OUT_NONE, 0, INSTR, FLAG_NO, 0), + PCE_MC_M(0x4000, 0xF000, 27, OUT_IP0, 4, INSTR, FLAG_IPV4, 1), + PCE_MC_M(0x6000, 0xF000, 30, OUT_IP0, 3, INSTR, FLAG_IPV6, 0), + PCE_MC_M(0x0000, 0x0000, 45, OUT_NONE, 0, INSTR, FLAG_NO, 0), + PCE_MC_M(0x0000, 0x0000, 28, OUT_IP3, 2, INSTR, FLAG_NO, 0), + PCE_MC_M(0x0000, 0x0000, 29, OUT_SIP0, 4, INSTR, FLAG_NO, 0), + PCE_MC_M(0x0000, 0x0000, 44, OUT_NONE, 0, LENACCU, FLAG_NO, 0), + PCE_MC_M(0x1100, 0xFF00, 43, OUT_PROT, 1, INSTR, FLAG_NO, 0), + PCE_MC_M(0x0600, 0xFF00, 43, OUT_PROT, 1, INSTR, FLAG_NO, 0), + PCE_MC_M(0x0000, 0xFF00, 36, OUT_IP3, 17, INSTR, FLAG_HOP, 0), + PCE_MC_M(0x2B00, 0xFF00, 36, OUT_IP3, 17, INSTR, FLAG_NN1, 0), + PCE_MC_M(0x3C00, 0xFF00, 36, OUT_IP3, 17, INSTR, FLAG_NN2, 0), + PCE_MC_M(0x0000, 0x0000, 43, OUT_PROT, 1, INSTR, FLAG_NO, 0), + PCE_MC_M(0x0000, 0x00F0, 38, OUT_NONE, 0, INSTR, FLAG_NO, 0), + PCE_MC_M(0x0000, 0x0000, 44, OUT_NONE, 0, INSTR, FLAG_NO, 0), + PCE_MC_M(0x0000, 0xFF00, 36, OUT_NONE, 0, IPV6, FLAG_HOP, 0), + PCE_MC_M(0x2B00, 0xFF00, 36, OUT_NONE, 0, IPV6, FLAG_NN1, 0), + PCE_MC_M(0x3C00, 0xFF00, 36, OUT_NONE, 0, IPV6, FLAG_NN2, 0), + PCE_MC_M(0x0000, 0x00FC, 44, OUT_PROT, 0, IPV6, FLAG_NO, 0), + PCE_MC_M(0x0000, 0x0000, 44, OUT_NONE, 0, IPV6, FLAG_NO, 0), + PCE_MC_M(0x0000, 0x0000, 44, OUT_SIP0, 16, INSTR, FLAG_NO, 0), + PCE_MC_M(0x0000, 0x0000, 45, OUT_APP0, 4, INSTR, FLAG_IGMP, 0), + PCE_MC_M(0x0000, 0x0000, 45, OUT_NONE, 0, INSTR, FLAG_END, 0), + PCE_MC_M(0x0000, 0x0000, 45, OUT_NONE, 0, INSTR, FLAG_END, 0), + PCE_MC_M(0x0000, 0x0000, 45, OUT_NONE, 0, INSTR, FLAG_END, 0), + PCE_MC_M(0x0000, 0x0000, 45, OUT_NONE, 0, INSTR, FLAG_END, 0), + PCE_MC_M(0x0000, 0x0000, 45, OUT_NONE, 0, INSTR, FLAG_END, 0), + PCE_MC_M(0x0000, 0x0000, 45, OUT_NONE, 0, INSTR, FLAG_END, 0), + PCE_MC_M(0x0000, 0x0000, 45, OUT_NONE, 0, INSTR, FLAG_END, 0), + PCE_MC_M(0x0000, 0x0000, 45, OUT_NONE, 0, INSTR, FLAG_END, 0), + PCE_MC_M(0x0000, 0x0000, 45, OUT_NONE, 0, INSTR, FLAG_END, 0), + PCE_MC_M(0x0000, 0x0000, 45, OUT_NONE, 0, INSTR, FLAG_END, 0), + PCE_MC_M(0x0000, 0x0000, 45, OUT_NONE, 0, INSTR, FLAG_END, 0), + PCE_MC_M(0x0000, 0x0000, 45, OUT_NONE, 0, INSTR, FLAG_END, 0), + PCE_MC_M(0x0000, 0x0000, 45, OUT_NONE, 0, INSTR, FLAG_END, 0), + PCE_MC_M(0x0000, 0x0000, 45, OUT_NONE, 0, INSTR, FLAG_END, 0), + PCE_MC_M(0x0000, 0x0000, 45, OUT_NONE, 0, INSTR, FLAG_END, 0), + PCE_MC_M(0x0000, 0x0000, 45, OUT_NONE, 0, INSTR, FLAG_END, 0), + PCE_MC_M(0x0000, 0x0000, 45, OUT_NONE, 0, INSTR, FLAG_END, 0), + PCE_MC_M(0x0000, 0x0000, 45, OUT_NONE, 0, INSTR, FLAG_END, 0), + PCE_MC_M(0x0000, 0x0000, 45, OUT_NONE, 0, INSTR, FLAG_END, 0), +}; From 12ed3e5a03a8a5150977c2fa7e4dd739320592bc Mon Sep 17 00:00:00 2001 From: Aleksandr Loktionov Date: Thu, 30 Oct 2025 14:59:45 +0100 Subject: [PATCH 0580/1381] ice: add flow parsing for GTP and new protocol field support Introduce new protocol header types and field sizes to support GTPU, GTPC tunneling protocols. - Add field size macros for GTP TEID, QFI, and other headers - Extend ice_flow_field_info and enum definitions - Update hash macros for new protocols - Add support for IPv6 prefix matching and fragment headers This patch lays the groundwork for enhanced RSS and flow classification capabilities. Co-developed-by: Dan Nowlin Signed-off-by: Dan Nowlin Co-developed-by: Junfeng Guo Signed-off-by: Junfeng Guo Co-developed-by: Ting Xu Signed-off-by: Ting Xu Signed-off-by: Przemek Kitszel Reviewed-by: Simon Horman Signed-off-by: Aleksandr Loktionov Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_flow.c | 217 +++++++++++++++++- drivers/net/ethernet/intel/ice/ice_flow.h | 94 +++++++- .../ethernet/intel/ice/ice_protocol_type.h | 20 ++ 3 files changed, 322 insertions(+), 9 deletions(-) diff --git a/drivers/net/ethernet/intel/ice/ice_flow.c b/drivers/net/ethernet/intel/ice/ice_flow.c index 6d5c939dc8a5..a2f2a612428d 100644 --- a/drivers/net/ethernet/intel/ice/ice_flow.c +++ b/drivers/net/ethernet/intel/ice/ice_flow.c @@ -5,6 +5,38 @@ #include "ice_flow.h" #include +/* Size of known protocol header fields */ +#define ICE_FLOW_FLD_SZ_ETH_TYPE 2 +#define ICE_FLOW_FLD_SZ_VLAN 2 +#define ICE_FLOW_FLD_SZ_IPV4_ADDR 4 +#define ICE_FLOW_FLD_SZ_IPV6_ADDR 16 +#define ICE_FLOW_FLD_SZ_IPV6_PRE32_ADDR 4 +#define ICE_FLOW_FLD_SZ_IPV6_PRE48_ADDR 6 +#define ICE_FLOW_FLD_SZ_IPV6_PRE64_ADDR 8 +#define ICE_FLOW_FLD_SZ_IPV4_ID 2 +#define ICE_FLOW_FLD_SZ_IPV6_ID 4 +#define ICE_FLOW_FLD_SZ_IP_CHKSUM 2 +#define ICE_FLOW_FLD_SZ_TCP_CHKSUM 2 +#define ICE_FLOW_FLD_SZ_UDP_CHKSUM 2 +#define ICE_FLOW_FLD_SZ_SCTP_CHKSUM 4 +#define ICE_FLOW_FLD_SZ_IP_DSCP 1 +#define ICE_FLOW_FLD_SZ_IP_TTL 1 +#define ICE_FLOW_FLD_SZ_IP_PROT 1 +#define ICE_FLOW_FLD_SZ_PORT 2 +#define ICE_FLOW_FLD_SZ_TCP_FLAGS 1 +#define ICE_FLOW_FLD_SZ_ICMP_TYPE 1 +#define ICE_FLOW_FLD_SZ_ICMP_CODE 1 +#define ICE_FLOW_FLD_SZ_ARP_OPER 2 +#define ICE_FLOW_FLD_SZ_GRE_KEYID 4 +#define ICE_FLOW_FLD_SZ_GTP_TEID 4 +#define ICE_FLOW_FLD_SZ_GTP_QFI 2 +#define ICE_FLOW_FLD_SZ_PFCP_SEID 8 +#define ICE_FLOW_FLD_SZ_ESP_SPI 4 +#define ICE_FLOW_FLD_SZ_AH_SPI 4 +#define ICE_FLOW_FLD_SZ_NAT_T_ESP_SPI 4 +#define ICE_FLOW_FLD_SZ_L2TPV2_SESS_ID 2 +#define ICE_FLOW_FLD_SZ_L2TPV2_LEN_SESS_ID 2 + /* Describe properties of a protocol header field */ struct ice_flow_field_info { enum ice_flow_seg_hdr hdr; @@ -20,6 +52,7 @@ struct ice_flow_field_info { .mask = 0, \ } +/* QFI: 6-bit field in GTP-U PDU Session Container (3GPP TS 38.415) */ #define ICE_FLOW_FLD_INFO_MSK(_hdr, _offset_bytes, _size_bytes, _mask) { \ .hdr = _hdr, \ .off = (_offset_bytes) * BITS_PER_BYTE, \ @@ -61,7 +94,33 @@ struct ice_flow_field_info ice_flds_info[ICE_FLOW_FIELD_IDX_MAX] = { /* ICE_FLOW_FIELD_IDX_IPV6_SA */ ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_IPV6, 8, sizeof(struct in6_addr)), /* ICE_FLOW_FIELD_IDX_IPV6_DA */ - ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_IPV6, 24, sizeof(struct in6_addr)), + ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_IPV6, 24, ICE_FLOW_FLD_SZ_IPV6_ADDR), + /* ICE_FLOW_FIELD_IDX_IPV4_CHKSUM */ + ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_IPV4, 10, ICE_FLOW_FLD_SZ_IP_CHKSUM), + /* ICE_FLOW_FIELD_IDX_IPV4_FRAG */ + ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_IPV_FRAG, 4, + ICE_FLOW_FLD_SZ_IPV4_ID), + /* ICE_FLOW_FIELD_IDX_IPV6_FRAG */ + ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_IPV_FRAG, 4, + ICE_FLOW_FLD_SZ_IPV6_ID), + /* ICE_FLOW_FIELD_IDX_IPV6_PRE32_SA */ + ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_IPV6, 8, + ICE_FLOW_FLD_SZ_IPV6_PRE32_ADDR), + /* ICE_FLOW_FIELD_IDX_IPV6_PRE32_DA */ + ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_IPV6, 24, + ICE_FLOW_FLD_SZ_IPV6_PRE32_ADDR), + /* ICE_FLOW_FIELD_IDX_IPV6_PRE48_SA */ + ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_IPV6, 8, + ICE_FLOW_FLD_SZ_IPV6_PRE48_ADDR), + /* ICE_FLOW_FIELD_IDX_IPV6_PRE48_DA */ + ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_IPV6, 24, + ICE_FLOW_FLD_SZ_IPV6_PRE48_ADDR), + /* ICE_FLOW_FIELD_IDX_IPV6_PRE64_SA */ + ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_IPV6, 8, + ICE_FLOW_FLD_SZ_IPV6_PRE64_ADDR), + /* ICE_FLOW_FIELD_IDX_IPV6_PRE64_DA */ + ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_IPV6, 24, + ICE_FLOW_FLD_SZ_IPV6_PRE64_ADDR), /* Transport */ /* ICE_FLOW_FIELD_IDX_TCP_SRC_PORT */ ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_TCP, 0, sizeof(__be16)), @@ -76,7 +135,14 @@ struct ice_flow_field_info ice_flds_info[ICE_FLOW_FIELD_IDX_MAX] = { /* ICE_FLOW_FIELD_IDX_SCTP_DST_PORT */ ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_SCTP, 2, sizeof(__be16)), /* ICE_FLOW_FIELD_IDX_TCP_FLAGS */ - ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_TCP, 13, 1), + ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_TCP, 13, ICE_FLOW_FLD_SZ_TCP_FLAGS), + /* ICE_FLOW_FIELD_IDX_TCP_CHKSUM */ + ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_TCP, 16, ICE_FLOW_FLD_SZ_TCP_CHKSUM), + /* ICE_FLOW_FIELD_IDX_UDP_CHKSUM */ + ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_UDP, 6, ICE_FLOW_FLD_SZ_UDP_CHKSUM), + /* ICE_FLOW_FIELD_IDX_SCTP_CHKSUM */ + ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_SCTP, 8, + ICE_FLOW_FLD_SZ_SCTP_CHKSUM), /* ARP */ /* ICE_FLOW_FIELD_IDX_ARP_SIP */ ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_ARP, 14, sizeof(struct in_addr)), @@ -108,9 +174,17 @@ struct ice_flow_field_info ice_flds_info[ICE_FLOW_FIELD_IDX_MAX] = { ICE_FLOW_FLD_INFO_MSK(ICE_FLOW_SEG_HDR_GTPU_EH, 22, sizeof(__be16), 0x3f00), /* ICE_FLOW_FIELD_IDX_GTPU_UP_TEID */ - ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_GTPU_UP, 12, sizeof(__be32)), + ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_GTPU_UP, 12, + ICE_FLOW_FLD_SZ_GTP_TEID), + /* ICE_FLOW_FIELD_IDX_GTPU_UP_QFI */ + ICE_FLOW_FLD_INFO_MSK(ICE_FLOW_SEG_HDR_GTPU_UP, 22, + ICE_FLOW_FLD_SZ_GTP_QFI, 0x3f00), /* ICE_FLOW_FIELD_IDX_GTPU_DWN_TEID */ - ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_GTPU_DWN, 12, sizeof(__be32)), + ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_GTPU_DWN, 12, + ICE_FLOW_FLD_SZ_GTP_TEID), + /* ICE_FLOW_FIELD_IDX_GTPU_DWN_QFI */ + ICE_FLOW_FLD_INFO_MSK(ICE_FLOW_SEG_HDR_GTPU_DWN, 22, + ICE_FLOW_FLD_SZ_GTP_QFI, 0x3f00), /* PPPoE */ /* ICE_FLOW_FIELD_IDX_PPPOE_SESS_ID */ ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_PPPOE, 2, sizeof(__be16)), @@ -128,7 +202,16 @@ struct ice_flow_field_info ice_flds_info[ICE_FLOW_FIELD_IDX_MAX] = { ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_AH, 4, sizeof(__be32)), /* NAT_T_ESP */ /* ICE_FLOW_FIELD_IDX_NAT_T_ESP_SPI */ - ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_NAT_T_ESP, 8, sizeof(__be32)), + ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_NAT_T_ESP, 8, + ICE_FLOW_FLD_SZ_NAT_T_ESP_SPI), + /* L2TPV2 */ + /* ICE_FLOW_FIELD_IDX_L2TPV2_SESS_ID */ + ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_L2TPV2, 12, + ICE_FLOW_FLD_SZ_L2TPV2_SESS_ID), + /* L2TPV2_LEN */ + /* ICE_FLOW_FIELD_IDX_L2TPV2_LEN_SESS_ID */ + ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_L2TPV2, 14, + ICE_FLOW_FLD_SZ_L2TPV2_LEN_SESS_ID), }; /* Bitmaps indicating relevant packet types for a particular protocol header @@ -2324,6 +2407,130 @@ static void ice_rss_set_symm(struct ice_hw *hw, struct ice_flow_prof *prof) } } +/** + * ice_rss_cfg_raw_symm - Configure symmetric RSS for a raw parser profile + * @hw: device HW + * @prof: parser profile describing extracted FV (field vector) entries + * @prof_id: RSS profile identifier used to program symmetry registers + * + * The routine scans the parser profile's FV entries and looks for + * direction-sensitive pairs (L3 src/dst, L4 src/dst). When a pair is found, + * it programs XOR-based symmetry so that flows hash identically regardless + * of packet direction. This preserves CPU affinity for the same 5-tuple. + * + * Notes: + * - The size of each logical field (IPv4/IPv6 address, L4 port) is expressed + * in units of ICE_FLOW_FV_EXTRACT_SZ so we can step across fv[] correctly. + * - We guard against out-of-bounds access before looking at fv[i + len]. + */ +static void ice_rss_cfg_raw_symm(struct ice_hw *hw, + const struct ice_parser_profile *prof, + u64 prof_id) +{ + for (size_t i = 0; i < prof->fv_num; i++) { + u8 proto_id = prof->fv[i].proto_id; + u16 src_off = 0, dst_off = 0; + size_t src_idx, dst_idx; + bool is_matched = false; + unsigned int len = 0; + + switch (proto_id) { + /* IPv4 address pairs (outer/inner variants) */ + case ICE_PROT_IPV4_OF_OR_S: + case ICE_PROT_IPV4_IL: + case ICE_PROT_IPV4_IL_IL: + len = ICE_FLOW_FLD_SZ_IPV4_ADDR / + ICE_FLOW_FV_EXTRACT_SZ; + src_off = ICE_FLOW_FIELD_IPV4_SRC_OFFSET; + dst_off = ICE_FLOW_FIELD_IPV4_DST_OFFSET; + break; + + /* IPv6 address pairs (outer/inner variants) */ + case ICE_PROT_IPV6_OF_OR_S: + case ICE_PROT_IPV6_IL: + case ICE_PROT_IPV6_IL_IL: + len = ICE_FLOW_FLD_SZ_IPV6_ADDR / + ICE_FLOW_FV_EXTRACT_SZ; + src_off = ICE_FLOW_FIELD_IPV6_SRC_OFFSET; + dst_off = ICE_FLOW_FIELD_IPV6_DST_OFFSET; + break; + + /* L4 port pairs (TCP/UDP/SCTP) */ + case ICE_PROT_TCP_IL: + case ICE_PROT_UDP_IL_OR_S: + case ICE_PROT_SCTP_IL: + len = ICE_FLOW_FLD_SZ_PORT / ICE_FLOW_FV_EXTRACT_SZ; + src_off = ICE_FLOW_FIELD_SRC_PORT_OFFSET; + dst_off = ICE_FLOW_FIELD_DST_PORT_OFFSET; + break; + + default: + continue; + } + + /* Bounds check before accessing fv[i + len]. */ + if (i + len >= prof->fv_num) + continue; + + /* Verify src/dst pairing for this protocol id. */ + is_matched = prof->fv[i].offset == src_off && + prof->fv[i + len].proto_id == proto_id && + prof->fv[i + len].offset == dst_off; + if (!is_matched) + continue; + + /* Program XOR symmetry for this field pair. */ + src_idx = i; + dst_idx = i + len; + + ice_rss_config_xor(hw, prof_id, src_idx, dst_idx, len); + + /* Skip over the pair we just handled; the loop's ++i advances + * one more element, hence the --i after the jump. + */ + i += (2 * len); + /* not strictly needed; keeps static analyzers happy */ + if (i == 0) + break; + --i; + } +} + +/* Max registers index per packet profile */ +#define ICE_SYMM_REG_INDEX_MAX 6 + +/** + * ice_rss_update_raw_symm - update symmetric hash configuration + * for raw pattern + * @hw: pointer to the hardware structure + * @cfg: configure parameters for raw pattern + * @id: profile tracking ID + * + * Update symmetric hash configuration for raw pattern if required. + * Otherwise only clear to default. + */ +void +ice_rss_update_raw_symm(struct ice_hw *hw, + struct ice_rss_raw_cfg *cfg, u64 id) +{ + struct ice_prof_map *map; + u8 prof_id, m; + + mutex_lock(&hw->blk[ICE_BLK_RSS].es.prof_map_lock); + map = ice_search_prof_id(hw, ICE_BLK_RSS, id); + if (map) + prof_id = map->prof_id; + mutex_unlock(&hw->blk[ICE_BLK_RSS].es.prof_map_lock); + if (!map) + return; + /* clear to default */ + for (m = 0; m < ICE_SYMM_REG_INDEX_MAX; m++) + wr32(hw, GLQF_HSYMM(prof_id, m), 0); + + if (cfg->symm) + ice_rss_cfg_raw_symm(hw, &cfg->prof, prof_id); +} + /** * ice_add_rss_cfg_sync - add an RSS configuration * @hw: pointer to the hardware structure diff --git a/drivers/net/ethernet/intel/ice/ice_flow.h b/drivers/net/ethernet/intel/ice/ice_flow.h index 52f906d89eca..6c6cdc8addb1 100644 --- a/drivers/net/ethernet/intel/ice/ice_flow.h +++ b/drivers/net/ethernet/intel/ice/ice_flow.h @@ -22,6 +22,15 @@ #define ICE_FLOW_HASH_IPV6 \ (BIT_ULL(ICE_FLOW_FIELD_IDX_IPV6_SA) | \ BIT_ULL(ICE_FLOW_FIELD_IDX_IPV6_DA)) +#define ICE_FLOW_HASH_IPV6_PRE32 \ + (BIT_ULL(ICE_FLOW_FIELD_IDX_IPV6_PRE32_SA) | \ + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV6_PRE32_DA)) +#define ICE_FLOW_HASH_IPV6_PRE48 \ + (BIT_ULL(ICE_FLOW_FIELD_IDX_IPV6_PRE48_SA) | \ + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV6_PRE48_DA)) +#define ICE_FLOW_HASH_IPV6_PRE64 \ + (BIT_ULL(ICE_FLOW_FIELD_IDX_IPV6_PRE64_SA) | \ + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV6_PRE64_DA)) #define ICE_FLOW_HASH_TCP_PORT \ (BIT_ULL(ICE_FLOW_FIELD_IDX_TCP_SRC_PORT) | \ BIT_ULL(ICE_FLOW_FIELD_IDX_TCP_DST_PORT)) @@ -40,6 +49,33 @@ #define ICE_HASH_SCTP_IPV4 (ICE_FLOW_HASH_IPV4 | ICE_FLOW_HASH_SCTP_PORT) #define ICE_HASH_SCTP_IPV6 (ICE_FLOW_HASH_IPV6 | ICE_FLOW_HASH_SCTP_PORT) +#define ICE_HASH_TCP_IPV6_PRE32 \ + (ICE_FLOW_HASH_IPV6_PRE32 | ICE_FLOW_HASH_TCP_PORT) +#define ICE_HASH_UDP_IPV6_PRE32 \ + (ICE_FLOW_HASH_IPV6_PRE32 | ICE_FLOW_HASH_UDP_PORT) +#define ICE_HASH_SCTP_IPV6_PRE32 \ + (ICE_FLOW_HASH_IPV6_PRE32 | ICE_FLOW_HASH_SCTP_PORT) +#define ICE_HASH_TCP_IPV6_PRE48 \ + (ICE_FLOW_HASH_IPV6_PRE48 | ICE_FLOW_HASH_TCP_PORT) +#define ICE_HASH_UDP_IPV6_PRE48 \ + (ICE_FLOW_HASH_IPV6_PRE48 | ICE_FLOW_HASH_UDP_PORT) +#define ICE_HASH_SCTP_IPV6_PRE48 \ + (ICE_FLOW_HASH_IPV6_PRE48 | ICE_FLOW_HASH_SCTP_PORT) +#define ICE_HASH_TCP_IPV6_PRE64 \ + (ICE_FLOW_HASH_IPV6_PRE64 | ICE_FLOW_HASH_TCP_PORT) +#define ICE_HASH_UDP_IPV6_PRE64 \ + (ICE_FLOW_HASH_IPV6_PRE64 | ICE_FLOW_HASH_UDP_PORT) +#define ICE_HASH_SCTP_IPV6_PRE64 \ + (ICE_FLOW_HASH_IPV6_PRE64 | ICE_FLOW_HASH_SCTP_PORT) + +#define ICE_FLOW_HASH_GTP_TEID \ + (BIT_ULL(ICE_FLOW_FIELD_IDX_GTPC_TEID)) + +#define ICE_FLOW_HASH_GTP_IPV4_TEID \ + (ICE_FLOW_HASH_IPV4 | ICE_FLOW_HASH_GTP_TEID) +#define ICE_FLOW_HASH_GTP_IPV6_TEID \ + (ICE_FLOW_HASH_IPV6 | ICE_FLOW_HASH_GTP_TEID) + #define ICE_FLOW_HASH_GTP_C_TEID \ (BIT_ULL(ICE_FLOW_FIELD_IDX_GTPC_TEID)) @@ -128,6 +164,23 @@ #define ICE_FLOW_HASH_NAT_T_ESP_IPV6_SPI \ (ICE_FLOW_HASH_IPV6 | ICE_FLOW_HASH_NAT_T_ESP_SPI) +#define ICE_FLOW_HASH_L2TPV2_SESS_ID \ + (BIT_ULL(ICE_FLOW_FIELD_IDX_L2TPV2_SESS_ID)) +#define ICE_FLOW_HASH_L2TPV2_SESS_ID_ETH \ + (ICE_FLOW_HASH_ETH | ICE_FLOW_HASH_L2TPV2_SESS_ID) + +#define ICE_FLOW_HASH_L2TPV2_LEN_SESS_ID \ + (BIT_ULL(ICE_FLOW_FIELD_IDX_L2TPV2_LEN_SESS_ID)) +#define ICE_FLOW_HASH_L2TPV2_LEN_SESS_ID_ETH \ + (ICE_FLOW_HASH_ETH | ICE_FLOW_HASH_L2TPV2_LEN_SESS_ID) + +#define ICE_FLOW_FIELD_IPV4_SRC_OFFSET 12 +#define ICE_FLOW_FIELD_IPV4_DST_OFFSET 16 +#define ICE_FLOW_FIELD_IPV6_SRC_OFFSET 8 +#define ICE_FLOW_FIELD_IPV6_DST_OFFSET 24 +#define ICE_FLOW_FIELD_SRC_PORT_OFFSET 0 +#define ICE_FLOW_FIELD_DST_PORT_OFFSET 2 + /* Protocol header fields within a packet segment. A segment consists of one or * more protocol headers that make up a logical group of protocol headers. Each * logical group of protocol headers encapsulates or is encapsulated using/by @@ -160,10 +213,13 @@ enum ice_flow_seg_hdr { ICE_FLOW_SEG_HDR_AH = 0x00200000, ICE_FLOW_SEG_HDR_NAT_T_ESP = 0x00400000, ICE_FLOW_SEG_HDR_ETH_NON_IP = 0x00800000, + ICE_FLOW_SEG_HDR_GTPU_NON_IP = 0x01000000, + ICE_FLOW_SEG_HDR_L2TPV2 = 0x10000000, /* The following is an additive bit for ICE_FLOW_SEG_HDR_IPV4 and - * ICE_FLOW_SEG_HDR_IPV6 which include the IPV4 other PTYPEs + * ICE_FLOW_SEG_HDR_IPV6. */ - ICE_FLOW_SEG_HDR_IPV_OTHER = 0x20000000, + ICE_FLOW_SEG_HDR_IPV_FRAG = 0x40000000, + ICE_FLOW_SEG_HDR_IPV_OTHER = 0x80000000, }; /* These segments all have the same PTYPES, but are otherwise distinguished by @@ -200,6 +256,15 @@ enum ice_flow_field { ICE_FLOW_FIELD_IDX_IPV4_DA, ICE_FLOW_FIELD_IDX_IPV6_SA, ICE_FLOW_FIELD_IDX_IPV6_DA, + ICE_FLOW_FIELD_IDX_IPV4_CHKSUM, + ICE_FLOW_FIELD_IDX_IPV4_ID, + ICE_FLOW_FIELD_IDX_IPV6_ID, + ICE_FLOW_FIELD_IDX_IPV6_PRE32_SA, + ICE_FLOW_FIELD_IDX_IPV6_PRE32_DA, + ICE_FLOW_FIELD_IDX_IPV6_PRE48_SA, + ICE_FLOW_FIELD_IDX_IPV6_PRE48_DA, + ICE_FLOW_FIELD_IDX_IPV6_PRE64_SA, + ICE_FLOW_FIELD_IDX_IPV6_PRE64_DA, /* L4 */ ICE_FLOW_FIELD_IDX_TCP_SRC_PORT, ICE_FLOW_FIELD_IDX_TCP_DST_PORT, @@ -208,6 +273,9 @@ enum ice_flow_field { ICE_FLOW_FIELD_IDX_SCTP_SRC_PORT, ICE_FLOW_FIELD_IDX_SCTP_DST_PORT, ICE_FLOW_FIELD_IDX_TCP_FLAGS, + ICE_FLOW_FIELD_IDX_TCP_CHKSUM, + ICE_FLOW_FIELD_IDX_UDP_CHKSUM, + ICE_FLOW_FIELD_IDX_SCTP_CHKSUM, /* ARP */ ICE_FLOW_FIELD_IDX_ARP_SIP, ICE_FLOW_FIELD_IDX_ARP_DIP, @@ -228,13 +296,13 @@ enum ice_flow_field { ICE_FLOW_FIELD_IDX_GTPU_EH_QFI, /* GTPU_UP */ ICE_FLOW_FIELD_IDX_GTPU_UP_TEID, + ICE_FLOW_FIELD_IDX_GTPU_UP_QFI, /* GTPU_DWN */ ICE_FLOW_FIELD_IDX_GTPU_DWN_TEID, - /* PPPoE */ + ICE_FLOW_FIELD_IDX_GTPU_DWN_QFI, ICE_FLOW_FIELD_IDX_PPPOE_SESS_ID, /* PFCP */ ICE_FLOW_FIELD_IDX_PFCP_SEID, - /* L2TPv3 */ ICE_FLOW_FIELD_IDX_L2TPV3_SESS_ID, /* ESP */ ICE_FLOW_FIELD_IDX_ESP_SPI, @@ -242,10 +310,16 @@ enum ice_flow_field { ICE_FLOW_FIELD_IDX_AH_SPI, /* NAT_T ESP */ ICE_FLOW_FIELD_IDX_NAT_T_ESP_SPI, + /* L2TPV2 SESSION ID*/ + ICE_FLOW_FIELD_IDX_L2TPV2_SESS_ID, + /* L2TPV2_LEN SESSION ID */ + ICE_FLOW_FIELD_IDX_L2TPV2_LEN_SESS_ID, /* The total number of enums must not exceed 64 */ ICE_FLOW_FIELD_IDX_MAX }; +static_assert(ICE_FLOW_FIELD_IDX_MAX <= 64, "The total number of enums must not exceed 64"); + #define ICE_FLOW_HASH_FLD_IPV4_SA BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_SA) #define ICE_FLOW_HASH_FLD_IPV6_SA BIT_ULL(ICE_FLOW_FIELD_IDX_IPV6_SA) #define ICE_FLOW_HASH_FLD_IPV4_DA BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_DA) @@ -296,6 +370,10 @@ enum ice_rss_cfg_hdr_type { /* take inner headers as inputset for packet with outer ipv6. */ ICE_RSS_INNER_HEADERS_W_OUTER_IPV6, /* take outer headers first then inner headers as inputset */ + /* take inner as inputset for GTPoGRE with outer IPv4 + GRE. */ + ICE_RSS_INNER_HEADERS_W_OUTER_IPV4_GRE, + /* take inner as inputset for GTPoGRE with outer IPv6 + GRE. */ + ICE_RSS_INNER_HEADERS_W_OUTER_IPV6_GRE, ICE_RSS_ANY_HEADERS }; @@ -406,6 +484,12 @@ struct ice_flow_prof { bool symm; /* Symmetric Hash for RSS */ }; +struct ice_rss_raw_cfg { + struct ice_parser_profile prof; + bool raw_ena; + bool symm; +}; + struct ice_rss_cfg { struct list_head l_entry; /* bitmap of VSIs added to the RSS entry */ @@ -444,4 +528,6 @@ int ice_add_rss_cfg(struct ice_hw *hw, struct ice_vsi *vsi, int ice_rem_rss_cfg(struct ice_hw *hw, u16 vsi_handle, const struct ice_rss_hash_cfg *cfg); u64 ice_get_rss_cfg(struct ice_hw *hw, u16 vsi_handle, u32 hdrs, bool *symm); +void ice_rss_update_raw_symm(struct ice_hw *hw, + struct ice_rss_raw_cfg *cfg, u64 id); #endif /* _ICE_FLOW_H_ */ diff --git a/drivers/net/ethernet/intel/ice/ice_protocol_type.h b/drivers/net/ethernet/intel/ice/ice_protocol_type.h index 7c09ea0f03ba..725167d557a8 100644 --- a/drivers/net/ethernet/intel/ice/ice_protocol_type.h +++ b/drivers/net/ethernet/intel/ice/ice_protocol_type.h @@ -82,26 +82,46 @@ enum ice_sw_tunnel_type { enum ice_prot_id { ICE_PROT_ID_INVAL = 0, ICE_PROT_MAC_OF_OR_S = 1, + ICE_PROT_MAC_O2 = 2, ICE_PROT_MAC_IL = 4, + ICE_PROT_MAC_IN_MAC = 7, ICE_PROT_ETYPE_OL = 9, ICE_PROT_ETYPE_IL = 10, + ICE_PROT_PAY = 15, + ICE_PROT_EVLAN_O = 16, + ICE_PROT_VLAN_O = 17, + ICE_PROT_VLAN_IF = 18, + ICE_PROT_MPLS_OL_MINUS_1 = 27, + ICE_PROT_MPLS_OL_OR_OS = 28, + ICE_PROT_MPLS_IL = 29, ICE_PROT_IPV4_OF_OR_S = 32, ICE_PROT_IPV4_IL = 33, + ICE_PROT_IPV4_IL_IL = 34, ICE_PROT_IPV6_OF_OR_S = 40, ICE_PROT_IPV6_IL = 41, + ICE_PROT_IPV6_IL_IL = 42, + ICE_PROT_IPV6_NEXT_PROTO = 43, + ICE_PROT_IPV6_FRAG = 47, ICE_PROT_TCP_IL = 49, ICE_PROT_UDP_OF = 52, ICE_PROT_UDP_IL_OR_S = 53, ICE_PROT_GRE_OF = 64, + ICE_PROT_NSH_F = 84, ICE_PROT_ESP_F = 88, ICE_PROT_ESP_2 = 89, ICE_PROT_SCTP_IL = 96, ICE_PROT_ICMP_IL = 98, ICE_PROT_ICMPV6_IL = 100, + ICE_PROT_VRRP_F = 101, + ICE_PROT_OSPF = 102, ICE_PROT_PPPOE = 103, ICE_PROT_L2TPV3 = 104, + ICE_PROT_ATAOE_OF = 114, + ICE_PROT_CTRL_OF = 116, + ICE_PROT_LLDP_OF = 117, ICE_PROT_ARP_OF = 118, ICE_PROT_META_ID = 255, /* when offset == metadata */ + ICE_PROT_EAPOL_OF = 120, ICE_PROT_INVALID = 255 /* when offset == ICE_FV_OFFSET_INVAL */ }; From 38724a474c0fc37b6604e8b20c75d87446fc2fd1 Mon Sep 17 00:00:00 2001 From: Aleksandr Loktionov Date: Thu, 30 Oct 2025 14:59:46 +0100 Subject: [PATCH 0581/1381] ice: add virtchnl definitions and static data for GTP RSS Add virtchnl protocol header and field definitions for advanced RSS configuration including GTPC, GTPU, L2TPv2, ECPRI, PPP, GRE, and IP fragment headers. - Define new virtchnl protocol header types - Add RSS field selectors for tunnel protocols - Extend static mapping arrays for protocol field matching - Add L2TPv2 session ID and length+session ID field support This provides the foundational definitions needed for VF RSS configuration of tunnel protocols. Co-developed-by: Dan Nowlin Signed-off-by: Dan Nowlin Co-developed-by: Jie Wang Signed-off-by: Jie Wang Co-developed-by: Junfeng Guo Signed-off-by: Junfeng Guo Co-developed-by: Qi Zhang Signed-off-by: Qi Zhang Co-developed-by: Ting Xu Signed-off-by: Ting Xu Signed-off-by: Przemek Kitszel Signed-off-by: Aleksandr Loktionov Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_vf_lib.h | 48 +++++ drivers/net/ethernet/intel/ice/virt/rss.c | 219 +++++++++++++++++++- include/linux/avf/virtchnl.h | 50 +++++ 3 files changed, 316 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/intel/ice/ice_vf_lib.h b/drivers/net/ethernet/intel/ice/ice_vf_lib.h index b00708907176..7a9c75d1d07c 100644 --- a/drivers/net/ethernet/intel/ice/ice_vf_lib.h +++ b/drivers/net/ethernet/intel/ice/ice_vf_lib.h @@ -53,6 +53,46 @@ struct ice_mdd_vf_events { u16 last_printed; }; +enum ice_hash_ip_ctx_type { + ICE_HASH_IP_CTX_IP = 0, + ICE_HASH_IP_CTX_IP_ESP, + ICE_HASH_IP_CTX_IP_UDP_ESP, + ICE_HASH_IP_CTX_IP_AH, + ICE_HASH_IP_CTX_IP_PFCP, + ICE_HASH_IP_CTX_IP_UDP, + ICE_HASH_IP_CTX_IP_TCP, + ICE_HASH_IP_CTX_IP_SCTP, + ICE_HASH_IP_CTX_MAX, +}; + +struct ice_vf_hash_ip_ctx { + struct ice_rss_hash_cfg ctx[ICE_HASH_IP_CTX_MAX]; +}; + +enum ice_hash_gtpu_ctx_type { + ICE_HASH_GTPU_CTX_EH_IP = 0, + ICE_HASH_GTPU_CTX_EH_IP_UDP, + ICE_HASH_GTPU_CTX_EH_IP_TCP, + ICE_HASH_GTPU_CTX_UP_IP, + ICE_HASH_GTPU_CTX_UP_IP_UDP, + ICE_HASH_GTPU_CTX_UP_IP_TCP, + ICE_HASH_GTPU_CTX_DW_IP, + ICE_HASH_GTPU_CTX_DW_IP_UDP, + ICE_HASH_GTPU_CTX_DW_IP_TCP, + ICE_HASH_GTPU_CTX_MAX, +}; + +struct ice_vf_hash_gtpu_ctx { + struct ice_rss_hash_cfg ctx[ICE_HASH_GTPU_CTX_MAX]; +}; + +struct ice_vf_hash_ctx { + struct ice_vf_hash_ip_ctx v4; + struct ice_vf_hash_ip_ctx v6; + struct ice_vf_hash_gtpu_ctx ipv4; + struct ice_vf_hash_gtpu_ctx ipv6; +}; + /* Structure to store fdir fv entry */ struct ice_fdir_prof_info { struct ice_parser_profile prof; @@ -66,6 +106,12 @@ struct ice_vf_qs_bw { u8 tc; }; +/* Structure to store RSS field vector entry */ +struct ice_rss_prof_info { + struct ice_parser_profile prof; + bool symm; +}; + /* VF operations */ struct ice_vf_ops { enum ice_disq_rst_src reset_type; @@ -106,6 +152,8 @@ struct ice_vf { u16 ctrl_vsi_idx; struct ice_vf_fdir fdir; struct ice_fdir_prof_info fdir_prof_info[ICE_MAX_PTGS]; + struct ice_rss_prof_info rss_prof_info[ICE_MAX_PTGS]; + struct ice_vf_hash_ctx hash_ctx; u64 rss_hashcfg; /* RSS hash configuration */ struct ice_sw *vf_sw_id; /* switch ID the VF VSIs connect to */ struct virtchnl_version_info vf_ver; diff --git a/drivers/net/ethernet/intel/ice/virt/rss.c b/drivers/net/ethernet/intel/ice/virt/rss.c index cbdbb32d512b..ee0d1ec32d56 100644 --- a/drivers/net/ethernet/intel/ice/virt/rss.c +++ b/drivers/net/ethernet/intel/ice/virt/rss.c @@ -36,6 +36,11 @@ static const struct ice_vc_hdr_match_type ice_vc_hdr_list[] = { {VIRTCHNL_PROTO_HDR_ESP, ICE_FLOW_SEG_HDR_ESP}, {VIRTCHNL_PROTO_HDR_AH, ICE_FLOW_SEG_HDR_AH}, {VIRTCHNL_PROTO_HDR_PFCP, ICE_FLOW_SEG_HDR_PFCP_SESSION}, + {VIRTCHNL_PROTO_HDR_GTPC, ICE_FLOW_SEG_HDR_GTPC}, + {VIRTCHNL_PROTO_HDR_L2TPV2, ICE_FLOW_SEG_HDR_L2TPV2}, + {VIRTCHNL_PROTO_HDR_IPV4_FRAG, ICE_FLOW_SEG_HDR_IPV_FRAG}, + {VIRTCHNL_PROTO_HDR_IPV6_EH_FRAG, ICE_FLOW_SEG_HDR_IPV_FRAG}, + {VIRTCHNL_PROTO_HDR_GRE, ICE_FLOW_SEG_HDR_GRE}, }; struct ice_vc_hash_field_match_type { @@ -87,8 +92,125 @@ ice_vc_hash_field_match_type ice_vc_hash_field_list[] = { FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_DST) | FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_PROT), ICE_FLOW_HASH_IPV4 | BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_PROT)}, - {VIRTCHNL_PROTO_HDR_IPV4, FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_PROT), + {VIRTCHNL_PROTO_HDR_IPV4, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_FRAG_PKID), + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_ID)}, + {VIRTCHNL_PROTO_HDR_IPV4, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_CHKSUM), + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_CHKSUM)}, + {VIRTCHNL_PROTO_HDR_IPV4, FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_SRC) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_CHKSUM), + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_SA) | + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_CHKSUM)}, + {VIRTCHNL_PROTO_HDR_IPV4, FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_DST) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_CHKSUM), + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_DA) | + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_CHKSUM)}, + {VIRTCHNL_PROTO_HDR_IPV4, FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_SRC) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_DST) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_CHKSUM), + ICE_FLOW_HASH_IPV4 | BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_CHKSUM)}, + {VIRTCHNL_PROTO_HDR_IPV4, FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_SRC) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_PROT) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_CHKSUM), + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_SA) | + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_PROT) | + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_CHKSUM)}, + {VIRTCHNL_PROTO_HDR_IPV4, FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_DST) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_PROT) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_CHKSUM), + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_DA) | + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_PROT) | + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_CHKSUM)}, + {VIRTCHNL_PROTO_HDR_IPV4, FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_SRC) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_DST) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_PROT) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_CHKSUM), + ICE_FLOW_HASH_IPV4 | BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_PROT) | + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_CHKSUM)}, + {VIRTCHNL_PROTO_HDR_IPV4, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_PROT) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_CHKSUM), + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_PROT) | + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_CHKSUM)}, + {VIRTCHNL_PROTO_HDR_IPV4_FRAG, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_PROT), BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_PROT)}, + {VIRTCHNL_PROTO_HDR_IPV4_FRAG, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_SRC), + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_SA)}, + {VIRTCHNL_PROTO_HDR_IPV4_FRAG, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_DST), + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_DA)}, + {VIRTCHNL_PROTO_HDR_IPV4_FRAG, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_SRC) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_DST), + ICE_FLOW_HASH_IPV4}, + {VIRTCHNL_PROTO_HDR_IPV4_FRAG, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_SRC) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_PROT), + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_SA) | + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_PROT)}, + {VIRTCHNL_PROTO_HDR_IPV4_FRAG, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_DST) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_PROT), + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_DA) | + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_PROT)}, + {VIRTCHNL_PROTO_HDR_IPV4_FRAG, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_SRC) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_DST) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_PROT), + ICE_FLOW_HASH_IPV4 | BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_PROT)}, + {VIRTCHNL_PROTO_HDR_IPV4_FRAG, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_PROT), + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_PROT)}, + {VIRTCHNL_PROTO_HDR_IPV4_FRAG, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_FRAG_PKID), + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_ID)}, + {VIRTCHNL_PROTO_HDR_IPV4_FRAG, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_CHKSUM), + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_CHKSUM)}, + {VIRTCHNL_PROTO_HDR_IPV4_FRAG, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_SRC) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_CHKSUM), + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_SA) | + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_CHKSUM)}, + {VIRTCHNL_PROTO_HDR_IPV4_FRAG, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_DST) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_CHKSUM), + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_DA) | + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_CHKSUM)}, + {VIRTCHNL_PROTO_HDR_IPV4_FRAG, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_SRC) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_DST) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_CHKSUM), + ICE_FLOW_HASH_IPV4 | BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_CHKSUM)}, + {VIRTCHNL_PROTO_HDR_IPV4_FRAG, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_SRC) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_PROT) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_CHKSUM), + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_SA) | + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_PROT) | + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_CHKSUM)}, + {VIRTCHNL_PROTO_HDR_IPV4_FRAG, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_DST) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_PROT) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_CHKSUM), + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_DA) | + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_PROT) | + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_CHKSUM)}, + {VIRTCHNL_PROTO_HDR_IPV4_FRAG, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_SRC) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_DST) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_PROT) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_CHKSUM), + ICE_FLOW_HASH_IPV4 | BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_PROT) | + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_CHKSUM)}, + {VIRTCHNL_PROTO_HDR_IPV4_FRAG, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_PROT) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV4_CHKSUM), + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_PROT) | + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_CHKSUM)}, {VIRTCHNL_PROTO_HDR_IPV6, FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV6_SRC), BIT_ULL(ICE_FLOW_FIELD_IDX_IPV6_SA)}, {VIRTCHNL_PROTO_HDR_IPV6, FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV6_DST), @@ -110,6 +232,35 @@ ice_vc_hash_field_match_type ice_vc_hash_field_list[] = { ICE_FLOW_HASH_IPV6 | BIT_ULL(ICE_FLOW_FIELD_IDX_IPV6_PROT)}, {VIRTCHNL_PROTO_HDR_IPV6, FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV6_PROT), BIT_ULL(ICE_FLOW_FIELD_IDX_IPV6_PROT)}, + {VIRTCHNL_PROTO_HDR_IPV6_EH_FRAG, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV6_EH_FRAG_PKID), + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV6_ID)}, + {VIRTCHNL_PROTO_HDR_IPV6, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV6_PREFIX64_SRC) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV6_PREFIX64_DST), + ICE_FLOW_HASH_IPV6_PRE64}, + {VIRTCHNL_PROTO_HDR_IPV6, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV6_PREFIX64_SRC), + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV6_PRE64_SA)}, + {VIRTCHNL_PROTO_HDR_IPV6, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV6_PREFIX64_DST), + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV6_PRE64_DA)}, + {VIRTCHNL_PROTO_HDR_IPV6, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV6_PREFIX64_SRC) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV6_PREFIX64_DST) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV6_PROT), + ICE_FLOW_HASH_IPV6_PRE64 | + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV6_PROT)}, + {VIRTCHNL_PROTO_HDR_IPV6, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV6_PREFIX64_SRC) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV6_PROT), + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV6_PRE64_SA) | + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV6_PROT)}, + {VIRTCHNL_PROTO_HDR_IPV6, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV6_PREFIX64_DST) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_IPV6_PROT), + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV6_PRE64_DA) | + BIT_ULL(ICE_FLOW_FIELD_IDX_IPV6_PROT)}, {VIRTCHNL_PROTO_HDR_TCP, FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_TCP_SRC_PORT), BIT_ULL(ICE_FLOW_FIELD_IDX_TCP_SRC_PORT)}, @@ -120,6 +271,25 @@ ice_vc_hash_field_match_type ice_vc_hash_field_list[] = { FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_TCP_SRC_PORT) | FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_TCP_DST_PORT), ICE_FLOW_HASH_TCP_PORT}, + {VIRTCHNL_PROTO_HDR_TCP, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_TCP_CHKSUM), + BIT_ULL(ICE_FLOW_FIELD_IDX_TCP_CHKSUM)}, + {VIRTCHNL_PROTO_HDR_TCP, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_TCP_SRC_PORT) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_TCP_CHKSUM), + BIT_ULL(ICE_FLOW_FIELD_IDX_TCP_SRC_PORT) | + BIT_ULL(ICE_FLOW_FIELD_IDX_TCP_CHKSUM)}, + {VIRTCHNL_PROTO_HDR_TCP, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_TCP_DST_PORT) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_TCP_CHKSUM), + BIT_ULL(ICE_FLOW_FIELD_IDX_TCP_DST_PORT) | + BIT_ULL(ICE_FLOW_FIELD_IDX_TCP_CHKSUM)}, + {VIRTCHNL_PROTO_HDR_TCP, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_TCP_SRC_PORT) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_TCP_DST_PORT) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_TCP_CHKSUM), + ICE_FLOW_HASH_TCP_PORT | + BIT_ULL(ICE_FLOW_FIELD_IDX_TCP_CHKSUM)}, {VIRTCHNL_PROTO_HDR_UDP, FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_UDP_SRC_PORT), BIT_ULL(ICE_FLOW_FIELD_IDX_UDP_SRC_PORT)}, @@ -130,6 +300,25 @@ ice_vc_hash_field_match_type ice_vc_hash_field_list[] = { FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_UDP_SRC_PORT) | FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_UDP_DST_PORT), ICE_FLOW_HASH_UDP_PORT}, + {VIRTCHNL_PROTO_HDR_UDP, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_UDP_CHKSUM), + BIT_ULL(ICE_FLOW_FIELD_IDX_UDP_CHKSUM)}, + {VIRTCHNL_PROTO_HDR_UDP, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_UDP_SRC_PORT) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_UDP_CHKSUM), + BIT_ULL(ICE_FLOW_FIELD_IDX_UDP_SRC_PORT) | + BIT_ULL(ICE_FLOW_FIELD_IDX_UDP_CHKSUM)}, + {VIRTCHNL_PROTO_HDR_UDP, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_UDP_DST_PORT) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_UDP_CHKSUM), + BIT_ULL(ICE_FLOW_FIELD_IDX_UDP_DST_PORT) | + BIT_ULL(ICE_FLOW_FIELD_IDX_UDP_CHKSUM)}, + {VIRTCHNL_PROTO_HDR_UDP, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_UDP_SRC_PORT) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_UDP_DST_PORT) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_UDP_CHKSUM), + ICE_FLOW_HASH_UDP_PORT | + BIT_ULL(ICE_FLOW_FIELD_IDX_UDP_CHKSUM)}, {VIRTCHNL_PROTO_HDR_SCTP, FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_SCTP_SRC_PORT), BIT_ULL(ICE_FLOW_FIELD_IDX_SCTP_SRC_PORT)}, @@ -140,6 +329,25 @@ ice_vc_hash_field_match_type ice_vc_hash_field_list[] = { FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_SCTP_SRC_PORT) | FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_SCTP_DST_PORT), ICE_FLOW_HASH_SCTP_PORT}, + {VIRTCHNL_PROTO_HDR_SCTP, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_SCTP_CHKSUM), + BIT_ULL(ICE_FLOW_FIELD_IDX_SCTP_CHKSUM)}, + {VIRTCHNL_PROTO_HDR_SCTP, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_SCTP_SRC_PORT) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_SCTP_CHKSUM), + BIT_ULL(ICE_FLOW_FIELD_IDX_SCTP_SRC_PORT) | + BIT_ULL(ICE_FLOW_FIELD_IDX_SCTP_CHKSUM)}, + {VIRTCHNL_PROTO_HDR_SCTP, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_SCTP_DST_PORT) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_SCTP_CHKSUM), + BIT_ULL(ICE_FLOW_FIELD_IDX_SCTP_DST_PORT) | + BIT_ULL(ICE_FLOW_FIELD_IDX_SCTP_CHKSUM)}, + {VIRTCHNL_PROTO_HDR_SCTP, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_SCTP_SRC_PORT) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_SCTP_DST_PORT) | + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_SCTP_CHKSUM), + ICE_FLOW_HASH_SCTP_PORT | + BIT_ULL(ICE_FLOW_FIELD_IDX_SCTP_CHKSUM)}, {VIRTCHNL_PROTO_HDR_PPPOE, FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_PPPOE_SESS_ID), BIT_ULL(ICE_FLOW_FIELD_IDX_PPPOE_SESS_ID)}, @@ -155,6 +363,15 @@ ice_vc_hash_field_match_type ice_vc_hash_field_list[] = { BIT_ULL(ICE_FLOW_FIELD_IDX_AH_SPI)}, {VIRTCHNL_PROTO_HDR_PFCP, FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_PFCP_SEID), BIT_ULL(ICE_FLOW_FIELD_IDX_PFCP_SEID)}, + {VIRTCHNL_PROTO_HDR_GTPC, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_GTPC_TEID), + BIT_ULL(ICE_FLOW_FIELD_IDX_GTPC_TEID)}, + {VIRTCHNL_PROTO_HDR_L2TPV2, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_L2TPV2_SESS_ID), + BIT_ULL(ICE_FLOW_FIELD_IDX_L2TPV2_SESS_ID)}, + {VIRTCHNL_PROTO_HDR_L2TPV2, + FIELD_SELECTOR(VIRTCHNL_PROTO_HDR_L2TPV2_LEN_SESS_ID), + BIT_ULL(ICE_FLOW_FIELD_IDX_L2TPV2_LEN_SESS_ID)}, }; /** diff --git a/include/linux/avf/virtchnl.h b/include/linux/avf/virtchnl.h index 5be1881abbb6..11bdab5522fd 100644 --- a/include/linux/avf/virtchnl.h +++ b/include/linux/avf/virtchnl.h @@ -1253,6 +1253,17 @@ enum virtchnl_proto_hdr_type { VIRTCHNL_PROTO_HDR_ESP, VIRTCHNL_PROTO_HDR_AH, VIRTCHNL_PROTO_HDR_PFCP, + VIRTCHNL_PROTO_HDR_GTPC, + VIRTCHNL_PROTO_HDR_ECPRI, + VIRTCHNL_PROTO_HDR_L2TPV2, + VIRTCHNL_PROTO_HDR_PPP, + /* IPv4 and IPv6 Fragment header types are only associated to + * VIRTCHNL_PROTO_HDR_IPV4 and VIRTCHNL_PROTO_HDR_IPV6 respectively, + * cannot be used independently. + */ + VIRTCHNL_PROTO_HDR_IPV4_FRAG, + VIRTCHNL_PROTO_HDR_IPV6_EH_FRAG, + VIRTCHNL_PROTO_HDR_GRE, }; /* Protocol header field within a protocol header. */ @@ -1275,6 +1286,7 @@ enum virtchnl_proto_hdr_field { VIRTCHNL_PROTO_HDR_IPV4_DSCP, VIRTCHNL_PROTO_HDR_IPV4_TTL, VIRTCHNL_PROTO_HDR_IPV4_PROT, + VIRTCHNL_PROTO_HDR_IPV4_CHKSUM, /* IPV6 */ VIRTCHNL_PROTO_HDR_IPV6_SRC = PROTO_HDR_FIELD_START(VIRTCHNL_PROTO_HDR_IPV6), @@ -1282,18 +1294,34 @@ enum virtchnl_proto_hdr_field { VIRTCHNL_PROTO_HDR_IPV6_TC, VIRTCHNL_PROTO_HDR_IPV6_HOP_LIMIT, VIRTCHNL_PROTO_HDR_IPV6_PROT, + /* IPV6 Prefix */ + VIRTCHNL_PROTO_HDR_IPV6_PREFIX32_SRC, + VIRTCHNL_PROTO_HDR_IPV6_PREFIX32_DST, + VIRTCHNL_PROTO_HDR_IPV6_PREFIX40_SRC, + VIRTCHNL_PROTO_HDR_IPV6_PREFIX40_DST, + VIRTCHNL_PROTO_HDR_IPV6_PREFIX48_SRC, + VIRTCHNL_PROTO_HDR_IPV6_PREFIX48_DST, + VIRTCHNL_PROTO_HDR_IPV6_PREFIX56_SRC, + VIRTCHNL_PROTO_HDR_IPV6_PREFIX56_DST, + VIRTCHNL_PROTO_HDR_IPV6_PREFIX64_SRC, + VIRTCHNL_PROTO_HDR_IPV6_PREFIX64_DST, + VIRTCHNL_PROTO_HDR_IPV6_PREFIX96_SRC, + VIRTCHNL_PROTO_HDR_IPV6_PREFIX96_DST, /* TCP */ VIRTCHNL_PROTO_HDR_TCP_SRC_PORT = PROTO_HDR_FIELD_START(VIRTCHNL_PROTO_HDR_TCP), VIRTCHNL_PROTO_HDR_TCP_DST_PORT, + VIRTCHNL_PROTO_HDR_TCP_CHKSUM, /* UDP */ VIRTCHNL_PROTO_HDR_UDP_SRC_PORT = PROTO_HDR_FIELD_START(VIRTCHNL_PROTO_HDR_UDP), VIRTCHNL_PROTO_HDR_UDP_DST_PORT, + VIRTCHNL_PROTO_HDR_UDP_CHKSUM, /* SCTP */ VIRTCHNL_PROTO_HDR_SCTP_SRC_PORT = PROTO_HDR_FIELD_START(VIRTCHNL_PROTO_HDR_SCTP), VIRTCHNL_PROTO_HDR_SCTP_DST_PORT, + VIRTCHNL_PROTO_HDR_SCTP_CHKSUM, /* GTPU_IP */ VIRTCHNL_PROTO_HDR_GTPU_IP_TEID = PROTO_HDR_FIELD_START(VIRTCHNL_PROTO_HDR_GTPU_IP), @@ -1317,6 +1345,28 @@ enum virtchnl_proto_hdr_field { VIRTCHNL_PROTO_HDR_PFCP_S_FIELD = PROTO_HDR_FIELD_START(VIRTCHNL_PROTO_HDR_PFCP), VIRTCHNL_PROTO_HDR_PFCP_SEID, + /* GTPC */ + VIRTCHNL_PROTO_HDR_GTPC_TEID = + PROTO_HDR_FIELD_START(VIRTCHNL_PROTO_HDR_GTPC), + /* ECPRI */ + VIRTCHNL_PROTO_HDR_ECPRI_MSG_TYPE = + PROTO_HDR_FIELD_START(VIRTCHNL_PROTO_HDR_ECPRI), + VIRTCHNL_PROTO_HDR_ECPRI_PC_RTC_ID, + /* IPv4 Dummy Fragment */ + VIRTCHNL_PROTO_HDR_IPV4_FRAG_PKID = + PROTO_HDR_FIELD_START(VIRTCHNL_PROTO_HDR_IPV4_FRAG), + /* IPv6 Extension Fragment */ + VIRTCHNL_PROTO_HDR_IPV6_EH_FRAG_PKID = + PROTO_HDR_FIELD_START(VIRTCHNL_PROTO_HDR_IPV6_EH_FRAG), + /* GTPU_DWN/UP */ + VIRTCHNL_PROTO_HDR_GTPU_DWN_QFI = + PROTO_HDR_FIELD_START(VIRTCHNL_PROTO_HDR_GTPU_EH_PDU_DWN), + VIRTCHNL_PROTO_HDR_GTPU_UP_QFI = + PROTO_HDR_FIELD_START(VIRTCHNL_PROTO_HDR_GTPU_EH_PDU_UP), + /* L2TPv2 */ + VIRTCHNL_PROTO_HDR_L2TPV2_SESS_ID = + PROTO_HDR_FIELD_START(VIRTCHNL_PROTO_HDR_L2TPV2), + VIRTCHNL_PROTO_HDR_L2TPV2_LEN_SESS_ID, }; struct virtchnl_proto_hdr { From 3a6d87e2eaaced60411b32c06191a5971c1303c2 Mon Sep 17 00:00:00 2001 From: Aleksandr Loktionov Date: Thu, 30 Oct 2025 14:59:47 +0100 Subject: [PATCH 0582/1381] ice: implement GTP RSS context tracking and configuration This commit implements the core RSS context management and configuration logic for GTP (GTPU) protocol support in VF RSS operations. Key implementation features: - GTPU hash context management with pre/post processing functions - Context index calculation and mapping for different GTPU scenarios - Integration with main RSS configuration flow via wrapper functions - Support for IPv4/IPv6 GTPU RSS configurations - Rollback mechanism for handling RSS rule conflicts - Hash context reset and cleanup functionality The implementation provides comprehensive GTPU RSS support by: 1. Adding ice_add_rss_cfg_pre_gtpu() for preprocessing GTPU contexts 2. Adding ice_add_rss_cfg_post_gtpu() for postprocessing configurations 3. Adding ice_calc_gtpu_ctx_idx() for context index calculation 4. Integrating GTPU logic into ice_add_rss_cfg_wrap() and ice_rem_rss_cfg_wrap() 5. Supporting context tracking in VF hash_ctx structures This completes the GTP RSS infrastructure enabling VFs to configure RSS hashing on GTP-encapsulated traffic. Co-developed-by: Dan Nowlin Signed-off-by: Dan Nowlin Co-developed-by: Jie Wang Signed-off-by: Jie Wang Co-developed-by: Junfeng Guo Signed-off-by: Junfeng Guo Co-developed-by: Qi Zhang Signed-off-by: Qi Zhang Co-developed-by: Ting Xu Signed-off-by: Ting Xu Signed-off-by: Przemek Kitszel Signed-off-by: Aleksandr Loktionov Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/virt/rss.c | 1094 ++++++++++++++++++++- 1 file changed, 1040 insertions(+), 54 deletions(-) diff --git a/drivers/net/ethernet/intel/ice/virt/rss.c b/drivers/net/ethernet/intel/ice/virt/rss.c index ee0d1ec32d56..085e69ec0cfc 100644 --- a/drivers/net/ethernet/intel/ice/virt/rss.c +++ b/drivers/net/ethernet/intel/ice/virt/rss.c @@ -374,6 +374,43 @@ ice_vc_hash_field_match_type ice_vc_hash_field_list[] = { BIT_ULL(ICE_FLOW_FIELD_IDX_L2TPV2_LEN_SESS_ID)}, }; +static int +ice_vc_rss_hash_update(struct ice_hw *hw, struct ice_vsi *vsi, u8 hash_type) +{ + struct ice_vsi_ctx *ctx; + int ret; + + ctx = kzalloc(sizeof(*ctx), GFP_KERNEL); + if (!ctx) + return -ENOMEM; + + /* clear previous hash_type */ + ctx->info.q_opt_rss = vsi->info.q_opt_rss & + ~ICE_AQ_VSI_Q_OPT_RSS_HASH_M; + /* hash_type is passed in as ICE_AQ_VSI_Q_OPT_RSS_info.q_opt_rss |= FIELD_PREP(ICE_AQ_VSI_Q_OPT_RSS_HASH_M, + hash_type); + + /* Preserve existing queueing option setting */ + ctx->info.q_opt_tc = vsi->info.q_opt_tc; + ctx->info.q_opt_flags = vsi->info.q_opt_flags; + + ctx->info.valid_sections = + cpu_to_le16(ICE_AQ_VSI_PROP_Q_OPT_VALID); + + ret = ice_update_vsi(hw, vsi->idx, ctx, NULL); + if (ret) { + dev_err(ice_hw_to_dev(hw), "update VSI for RSS failed, err %d aq_err %s\n", + ret, libie_aq_str(hw->adminq.sq_last_status)); + } else { + vsi->info.q_opt_rss = ctx->info.q_opt_rss; + } + + kfree(ctx); + + return ret; +} + /** * ice_vc_validate_pattern * @vf: pointer to the VF info @@ -488,6 +525,11 @@ static bool ice_vc_parse_rss_cfg(struct ice_hw *hw, const struct ice_vc_hash_field_match_type *hf_list; const struct ice_vc_hdr_match_type *hdr_list; int i, hf_list_len, hdr_list_len; + bool outer_ipv4 = false; + bool outer_ipv6 = false; + bool inner_hdr = false; + bool has_gre = false; + u32 *addl_hdrs = &hash_cfg->addl_hdrs; u64 *hash_flds = &hash_cfg->hash_flds; @@ -507,17 +549,17 @@ static bool ice_vc_parse_rss_cfg(struct ice_hw *hw, for (i = 0; i < rss_cfg->proto_hdrs.count; i++) { struct virtchnl_proto_hdr *proto_hdr = &rss_cfg->proto_hdrs.proto_hdr[i]; - bool hdr_found = false; + u32 hdr_found = 0; int j; - /* Find matched ice headers according to virtchnl headers. */ + /* Find matched ice headers according to virtchnl headers. + * Also figure out the outer type of GTPU headers. + */ for (j = 0; j < hdr_list_len; j++) { struct ice_vc_hdr_match_type hdr_map = hdr_list[j]; - if (proto_hdr->type == hdr_map.vc_hdr) { - *addl_hdrs |= hdr_map.ice_hdr; - hdr_found = true; - } + if (proto_hdr->type == hdr_map.vc_hdr) + hdr_found = hdr_map.ice_hdr; } if (!hdr_found) @@ -535,8 +577,98 @@ static bool ice_vc_parse_rss_cfg(struct ice_hw *hw, break; } } + + if (proto_hdr->type == VIRTCHNL_PROTO_HDR_IPV4 && !inner_hdr) + outer_ipv4 = true; + else if (proto_hdr->type == VIRTCHNL_PROTO_HDR_IPV6 && + !inner_hdr) + outer_ipv6 = true; + /* for GRE and L2TPv2, take inner header as input set if no + * any field is selected from outer headers. + * for GTPU, take inner header and GTPU teid as input set. + */ + else if ((proto_hdr->type == VIRTCHNL_PROTO_HDR_GTPU_IP || + proto_hdr->type == VIRTCHNL_PROTO_HDR_GTPU_EH || + proto_hdr->type == VIRTCHNL_PROTO_HDR_GTPU_EH_PDU_DWN || + proto_hdr->type == + VIRTCHNL_PROTO_HDR_GTPU_EH_PDU_UP) || + ((proto_hdr->type == VIRTCHNL_PROTO_HDR_L2TPV2 || + proto_hdr->type == VIRTCHNL_PROTO_HDR_GRE) && + *hash_flds == 0)) { + /* set inner_hdr flag, and clean up outer header */ + inner_hdr = true; + + /* clear outer headers */ + *addl_hdrs = 0; + + if (outer_ipv4 && outer_ipv6) + return false; + + if (outer_ipv4) + hash_cfg->hdr_type = ICE_RSS_INNER_HEADERS_W_OUTER_IPV4; + else if (outer_ipv6) + hash_cfg->hdr_type = ICE_RSS_INNER_HEADERS_W_OUTER_IPV6; + else + hash_cfg->hdr_type = ICE_RSS_INNER_HEADERS; + + if (has_gre && outer_ipv4) + hash_cfg->hdr_type = + ICE_RSS_INNER_HEADERS_W_OUTER_IPV4_GRE; + if (has_gre && outer_ipv6) + hash_cfg->hdr_type = + ICE_RSS_INNER_HEADERS_W_OUTER_IPV6_GRE; + + if (proto_hdr->type == VIRTCHNL_PROTO_HDR_GRE) + has_gre = true; + } + + *addl_hdrs |= hdr_found; + + /* refine hash hdrs and fields for IP fragment */ + if (VIRTCHNL_TEST_PROTO_HDR_FIELD(proto_hdr, + VIRTCHNL_PROTO_HDR_IPV4_FRAG_PKID) && + proto_hdr->type == VIRTCHNL_PROTO_HDR_IPV4_FRAG) { + *addl_hdrs |= ICE_FLOW_SEG_HDR_IPV_FRAG; + *addl_hdrs &= ~(ICE_FLOW_SEG_HDR_IPV_OTHER); + *hash_flds |= BIT_ULL(ICE_FLOW_FIELD_IDX_IPV4_ID); + VIRTCHNL_DEL_PROTO_HDR_FIELD(proto_hdr, + VIRTCHNL_PROTO_HDR_IPV4_FRAG_PKID); + } + if (VIRTCHNL_TEST_PROTO_HDR_FIELD(proto_hdr, + VIRTCHNL_PROTO_HDR_IPV6_EH_FRAG_PKID) && + proto_hdr->type == VIRTCHNL_PROTO_HDR_IPV6_EH_FRAG) { + *addl_hdrs |= ICE_FLOW_SEG_HDR_IPV_FRAG; + *addl_hdrs &= ~(ICE_FLOW_SEG_HDR_IPV_OTHER); + *hash_flds |= BIT_ULL(ICE_FLOW_FIELD_IDX_IPV6_ID); + VIRTCHNL_DEL_PROTO_HDR_FIELD(proto_hdr, + VIRTCHNL_PROTO_HDR_IPV6_EH_FRAG_PKID); + } } + /* refine gtpu header if we take outer as input set for a no inner + * ip gtpu flow. + */ + if (hash_cfg->hdr_type == ICE_RSS_OUTER_HEADERS && + *addl_hdrs & ICE_FLOW_SEG_HDR_GTPU_IP) { + *addl_hdrs &= ~(ICE_FLOW_SEG_HDR_GTPU_IP); + *addl_hdrs |= ICE_FLOW_SEG_HDR_GTPU_NON_IP; + } + + /* refine hash field for esp and nat-t-esp. */ + if ((*addl_hdrs & ICE_FLOW_SEG_HDR_UDP) && + (*addl_hdrs & ICE_FLOW_SEG_HDR_ESP)) { + *addl_hdrs &= ~(ICE_FLOW_SEG_HDR_ESP | ICE_FLOW_SEG_HDR_UDP); + *addl_hdrs |= ICE_FLOW_SEG_HDR_NAT_T_ESP; + *hash_flds &= ~(BIT_ULL(ICE_FLOW_FIELD_IDX_ESP_SPI)); + *hash_flds |= BIT_ULL(ICE_FLOW_FIELD_IDX_NAT_T_ESP_SPI); + } + + /* refine hash hdrs for L4 udp/tcp/sctp. */ + if (*addl_hdrs & (ICE_FLOW_SEG_HDR_TCP | ICE_FLOW_SEG_HDR_UDP | + ICE_FLOW_SEG_HDR_SCTP) && + *addl_hdrs & ICE_FLOW_SEG_HDR_IPV_OTHER) + *addl_hdrs &= ~ICE_FLOW_SEG_HDR_IPV_OTHER; + return true; } @@ -553,6 +685,874 @@ static bool ice_vf_adv_rss_offload_ena(u32 caps) return !!(caps & VIRTCHNL_VF_OFFLOAD_ADV_RSS_PF); } +/** + * ice_is_hash_cfg_valid - Check whether an RSS hash context is valid + * @cfg: RSS hash configuration to test + * + * Return: true if both @cfg->hash_flds and @cfg->addl_hdrs are non-zero; false otherwise. + */ +static bool ice_is_hash_cfg_valid(struct ice_rss_hash_cfg *cfg) +{ + return cfg->hash_flds && cfg->addl_hdrs; +} + +/** + * ice_hash_cfg_reset - Reset an RSS hash context + * @cfg: RSS hash configuration to reset + * + * Reset fields of @cfg that store the active rule information. + */ +static void ice_hash_cfg_reset(struct ice_rss_hash_cfg *cfg) +{ + cfg->hash_flds = 0; + cfg->addl_hdrs = 0; + cfg->hdr_type = ICE_RSS_OUTER_HEADERS; + cfg->symm = 0; +} + +/** + * ice_hash_cfg_record - Record an RSS hash context + * @ctx: destination (global) RSS hash configuration + * @cfg: source RSS hash configuration to record + * + * Copy the active rule information from @cfg into @ctx. + */ +static void ice_hash_cfg_record(struct ice_rss_hash_cfg *ctx, + struct ice_rss_hash_cfg *cfg) +{ + ctx->hash_flds = cfg->hash_flds; + ctx->addl_hdrs = cfg->addl_hdrs; + ctx->hdr_type = cfg->hdr_type; + ctx->symm = cfg->symm; +} + +/** + * ice_hash_moveout - Delete an RSS configuration (keep context) + * @vf: VF pointer + * @cfg: RSS hash configuration + * + * Return: 0 on success (including when already absent); -ENOENT if @cfg is + * invalid or VSI is missing; -EBUSY on hardware removal failure. + */ +static int +ice_hash_moveout(struct ice_vf *vf, struct ice_rss_hash_cfg *cfg) +{ + struct device *dev = ice_pf_to_dev(vf->pf); + struct ice_vsi *vsi = ice_get_vf_vsi(vf); + struct ice_hw *hw = &vf->pf->hw; + int ret; + + if (!ice_is_hash_cfg_valid(cfg) || !vsi) + return -ENOENT; + + ret = ice_rem_rss_cfg(hw, vsi->idx, cfg); + if (ret && ret != -ENOENT) { + dev_err(dev, "ice_rem_rss_cfg failed for VF %d, VSI %d, error:%d\n", + vf->vf_id, vf->lan_vsi_idx, ret); + return -EBUSY; + } + + return 0; +} + +/** + * ice_hash_moveback - Add an RSS hash configuration for a VF + * @vf: VF pointer + * @cfg: RSS hash configuration to apply + * + * Add @cfg to @vf if the context is valid and VSI exists; programs HW. + * + * Return: + * * 0 on success + * * -ENOENT if @cfg is invalid or VSI is missing + * * -EBUSY if hardware programming fails + */ +static int +ice_hash_moveback(struct ice_vf *vf, struct ice_rss_hash_cfg *cfg) +{ + struct device *dev = ice_pf_to_dev(vf->pf); + struct ice_vsi *vsi = ice_get_vf_vsi(vf); + struct ice_hw *hw = &vf->pf->hw; + int ret; + + if (!ice_is_hash_cfg_valid(cfg) || !vsi) + return -ENOENT; + + ret = ice_add_rss_cfg(hw, vsi, cfg); + if (ret) { + dev_err(dev, "ice_add_rss_cfg failed for VF %d, VSI %d, error:%d\n", + vf->vf_id, vf->lan_vsi_idx, ret); + return -EBUSY; + } + + return 0; +} + +/** + * ice_hash_remove - remove a RSS configuration + * @vf: pointer to the VF info + * @cfg: pointer to the RSS hash configuration + * + * This function will delete a RSS hash configuration and also delete the + * hash context which stores the rule info. + * + * Return: 0 on success, or a negative error code on failure. + */ +static int +ice_hash_remove(struct ice_vf *vf, struct ice_rss_hash_cfg *cfg) +{ + int ret; + + ret = ice_hash_moveout(vf, cfg); + if (ret && ret != -ENOENT) + return ret; + + ice_hash_cfg_reset(cfg); + + return 0; +} + +struct ice_gtpu_ctx_action { + u32 ctx_idx; + const u32 *remove_list; + int remove_count; + const u32 *moveout_list; + int moveout_count; +}; + +/** + * ice_add_rss_cfg_pre_gtpu - Pre-process the GTPU RSS configuration + * @vf: pointer to the VF info + * @ctx: pointer to the context of the GTPU hash + * @ctx_idx: index of the hash context + * + * Pre-processes the GTPU hash configuration before adding a new + * hash context. It removes or reorders existing hash configurations that may + * conflict with the new one. For example, if a GTPU_UP or GTPU_DWN rule is + * configured after a GTPU_EH rule, the GTPU_EH hash will be matched first due + * to TCAM write and match order (top-down). In such cases, the GTPU_EH rule + * must be moved after the GTPU_UP/DWN rule. Conversely, if a GTPU_EH rule is + * configured after a GTPU_UP/DWN rule, the UP/DWN rules should be removed to + * avoid conflict. + * + * Return: 0 on success or a negative error code on failure + */ +static int ice_add_rss_cfg_pre_gtpu(struct ice_vf *vf, + struct ice_vf_hash_gtpu_ctx *ctx, + u32 ctx_idx) +{ + int ret, i; + + static const u32 remove_eh_ip[] = { + ICE_HASH_GTPU_CTX_EH_IP_UDP, ICE_HASH_GTPU_CTX_EH_IP_TCP, + ICE_HASH_GTPU_CTX_UP_IP, ICE_HASH_GTPU_CTX_UP_IP_UDP, + ICE_HASH_GTPU_CTX_UP_IP_TCP, ICE_HASH_GTPU_CTX_DW_IP, + ICE_HASH_GTPU_CTX_DW_IP_UDP, ICE_HASH_GTPU_CTX_DW_IP_TCP, + }; + + static const u32 remove_eh_ip_udp[] = { + ICE_HASH_GTPU_CTX_UP_IP_UDP, + ICE_HASH_GTPU_CTX_DW_IP_UDP, + }; + static const u32 moveout_eh_ip_udp[] = { + ICE_HASH_GTPU_CTX_UP_IP, + ICE_HASH_GTPU_CTX_UP_IP_TCP, + ICE_HASH_GTPU_CTX_DW_IP, + ICE_HASH_GTPU_CTX_DW_IP_TCP, + }; + + static const u32 remove_eh_ip_tcp[] = { + ICE_HASH_GTPU_CTX_UP_IP_TCP, + ICE_HASH_GTPU_CTX_DW_IP_TCP, + }; + static const u32 moveout_eh_ip_tcp[] = { + ICE_HASH_GTPU_CTX_UP_IP, + ICE_HASH_GTPU_CTX_UP_IP_UDP, + ICE_HASH_GTPU_CTX_DW_IP, + ICE_HASH_GTPU_CTX_DW_IP_UDP, + }; + + static const u32 remove_up_ip[] = { + ICE_HASH_GTPU_CTX_UP_IP_UDP, + ICE_HASH_GTPU_CTX_UP_IP_TCP, + }; + static const u32 moveout_up_ip[] = { + ICE_HASH_GTPU_CTX_EH_IP, + ICE_HASH_GTPU_CTX_EH_IP_UDP, + ICE_HASH_GTPU_CTX_EH_IP_TCP, + }; + + static const u32 moveout_up_ip_udp_tcp[] = { + ICE_HASH_GTPU_CTX_EH_IP, + ICE_HASH_GTPU_CTX_EH_IP_UDP, + ICE_HASH_GTPU_CTX_EH_IP_TCP, + }; + + static const u32 remove_dw_ip[] = { + ICE_HASH_GTPU_CTX_DW_IP_UDP, + ICE_HASH_GTPU_CTX_DW_IP_TCP, + }; + static const u32 moveout_dw_ip[] = { + ICE_HASH_GTPU_CTX_EH_IP, + ICE_HASH_GTPU_CTX_EH_IP_UDP, + ICE_HASH_GTPU_CTX_EH_IP_TCP, + }; + + static const struct ice_gtpu_ctx_action actions[] = { + { ICE_HASH_GTPU_CTX_EH_IP, remove_eh_ip, + ARRAY_SIZE(remove_eh_ip), NULL, 0 }, + { ICE_HASH_GTPU_CTX_EH_IP_UDP, remove_eh_ip_udp, + ARRAY_SIZE(remove_eh_ip_udp), moveout_eh_ip_udp, + ARRAY_SIZE(moveout_eh_ip_udp) }, + { ICE_HASH_GTPU_CTX_EH_IP_TCP, remove_eh_ip_tcp, + ARRAY_SIZE(remove_eh_ip_tcp), moveout_eh_ip_tcp, + ARRAY_SIZE(moveout_eh_ip_tcp) }, + { ICE_HASH_GTPU_CTX_UP_IP, remove_up_ip, + ARRAY_SIZE(remove_up_ip), moveout_up_ip, + ARRAY_SIZE(moveout_up_ip) }, + { ICE_HASH_GTPU_CTX_UP_IP_UDP, NULL, 0, moveout_up_ip_udp_tcp, + ARRAY_SIZE(moveout_up_ip_udp_tcp) }, + { ICE_HASH_GTPU_CTX_UP_IP_TCP, NULL, 0, moveout_up_ip_udp_tcp, + ARRAY_SIZE(moveout_up_ip_udp_tcp) }, + { ICE_HASH_GTPU_CTX_DW_IP, remove_dw_ip, + ARRAY_SIZE(remove_dw_ip), moveout_dw_ip, + ARRAY_SIZE(moveout_dw_ip) }, + { ICE_HASH_GTPU_CTX_DW_IP_UDP, NULL, 0, moveout_dw_ip, + ARRAY_SIZE(moveout_dw_ip) }, + { ICE_HASH_GTPU_CTX_DW_IP_TCP, NULL, 0, moveout_dw_ip, + ARRAY_SIZE(moveout_dw_ip) }, + }; + + for (i = 0; i < ARRAY_SIZE(actions); i++) { + if (actions[i].ctx_idx != ctx_idx) + continue; + + if (actions[i].remove_list) { + for (int j = 0; j < actions[i].remove_count; j++) { + u16 rm = actions[i].remove_list[j]; + + ret = ice_hash_remove(vf, &ctx->ctx[rm]); + if (ret && ret != -ENOENT) + return ret; + } + } + + if (actions[i].moveout_list) { + for (int j = 0; j < actions[i].moveout_count; j++) { + u16 mv = actions[i].moveout_list[j]; + + ret = ice_hash_moveout(vf, &ctx->ctx[mv]); + if (ret && ret != -ENOENT) + return ret; + } + } + break; + } + + return 0; +} + +/** + * ice_add_rss_cfg_pre_ip - Pre-process IP-layer RSS configuration + * @vf: VF pointer + * @ctx: IP L4 hash context (ESP/UDP-ESP/AH/PFCP and UDP/TCP/SCTP) + * + * Remove covered/recorded IP RSS configurations prior to adding a new one. + * + * Return: 0 on success; negative error code on failure. + */ +static int +ice_add_rss_cfg_pre_ip(struct ice_vf *vf, struct ice_vf_hash_ip_ctx *ctx) +{ + int i, ret; + + for (i = 1; i < ICE_HASH_IP_CTX_MAX; i++) + if (ice_is_hash_cfg_valid(&ctx->ctx[i])) { + ret = ice_hash_remove(vf, &ctx->ctx[i]); + if (ret) + return ret; + } + + return 0; +} + +/** + * ice_calc_gtpu_ctx_idx - Calculate GTPU hash context index + * @hdrs: Bitmask of protocol headers prefixed with ICE_FLOW_SEG_HDR_* + * + * Determine the GTPU hash context index based on the combination of + * encapsulation headers (GTPU_EH, GTPU_UP, GTPU_DWN) and transport + * protocols (UDP, TCP) within IPv4 or IPv6 flows. + * + * Return: A valid context index (0-8) if the header combination is supported, + * or ICE_HASH_GTPU_CTX_MAX if the combination is invalid. + */ +static enum ice_hash_gtpu_ctx_type ice_calc_gtpu_ctx_idx(u32 hdrs) +{ + u32 eh_idx, ip_idx; + + if (hdrs & ICE_FLOW_SEG_HDR_GTPU_EH) + eh_idx = 0; + else if (hdrs & ICE_FLOW_SEG_HDR_GTPU_UP) + eh_idx = 1; + else if (hdrs & ICE_FLOW_SEG_HDR_GTPU_DWN) + eh_idx = 2; + else + return ICE_HASH_GTPU_CTX_MAX; + + ip_idx = 0; + if (hdrs & ICE_FLOW_SEG_HDR_UDP) + ip_idx = 1; + else if (hdrs & ICE_FLOW_SEG_HDR_TCP) + ip_idx = 2; + + if (hdrs & (ICE_FLOW_SEG_HDR_IPV4 | ICE_FLOW_SEG_HDR_IPV6)) + return eh_idx * 3 + ip_idx; + else + return ICE_HASH_GTPU_CTX_MAX; +} + +/** + * ice_map_ip_ctx_idx - map the index of the IP L4 hash context + * @hdrs: protocol headers prefix with ICE_FLOW_SEG_HDR_XXX. + * + * The IP L4 hash context use the index to classify for IPv4/IPv6 with + * ESP/UDP_ESP/AH/PFCP and non-tunnel UDP/TCP/SCTP + * this function map the index based on the protocol headers. + * + * Return: The mapped IP context index on success, or ICE_HASH_IP_CTX_MAX + * if no matching context is found. + */ +static u8 ice_map_ip_ctx_idx(u32 hdrs) +{ + u8 i; + + static struct { + u32 hdrs; + u8 ctx_idx; + } ip_ctx_idx_map[] = { + { ICE_FLOW_SEG_HDR_IPV4 | ICE_FLOW_SEG_HDR_IPV_OTHER | + ICE_FLOW_SEG_HDR_ESP, + ICE_HASH_IP_CTX_IP_ESP }, + { ICE_FLOW_SEG_HDR_IPV4 | ICE_FLOW_SEG_HDR_IPV_OTHER | + ICE_FLOW_SEG_HDR_NAT_T_ESP, + ICE_HASH_IP_CTX_IP_UDP_ESP }, + { ICE_FLOW_SEG_HDR_IPV4 | ICE_FLOW_SEG_HDR_IPV_OTHER | + ICE_FLOW_SEG_HDR_AH, + ICE_HASH_IP_CTX_IP_AH }, + { ICE_FLOW_SEG_HDR_IPV4 | ICE_FLOW_SEG_HDR_IPV_OTHER | + ICE_FLOW_SEG_HDR_PFCP_SESSION, + ICE_HASH_IP_CTX_IP_PFCP }, + { ICE_FLOW_SEG_HDR_ETH | ICE_FLOW_SEG_HDR_VLAN | + ICE_FLOW_SEG_HDR_IPV4 | ICE_FLOW_SEG_HDR_UDP, + ICE_HASH_IP_CTX_IP_UDP }, + { ICE_FLOW_SEG_HDR_ETH | ICE_FLOW_SEG_HDR_VLAN | + ICE_FLOW_SEG_HDR_IPV4 | ICE_FLOW_SEG_HDR_TCP, + ICE_HASH_IP_CTX_IP_TCP }, + { ICE_FLOW_SEG_HDR_ETH | ICE_FLOW_SEG_HDR_VLAN | + ICE_FLOW_SEG_HDR_IPV4 | ICE_FLOW_SEG_HDR_SCTP, + ICE_HASH_IP_CTX_IP_SCTP }, + { ICE_FLOW_SEG_HDR_ETH | ICE_FLOW_SEG_HDR_VLAN | + ICE_FLOW_SEG_HDR_IPV4 | ICE_FLOW_SEG_HDR_IPV_OTHER, + ICE_HASH_IP_CTX_IP }, + { ICE_FLOW_SEG_HDR_IPV6 | ICE_FLOW_SEG_HDR_IPV_OTHER | + ICE_FLOW_SEG_HDR_ESP, + ICE_HASH_IP_CTX_IP_ESP }, + { ICE_FLOW_SEG_HDR_IPV6 | ICE_FLOW_SEG_HDR_IPV_OTHER | + ICE_FLOW_SEG_HDR_NAT_T_ESP, + ICE_HASH_IP_CTX_IP_UDP_ESP }, + { ICE_FLOW_SEG_HDR_IPV6 | ICE_FLOW_SEG_HDR_IPV_OTHER | + ICE_FLOW_SEG_HDR_AH, + ICE_HASH_IP_CTX_IP_AH }, + { ICE_FLOW_SEG_HDR_IPV6 | ICE_FLOW_SEG_HDR_IPV_OTHER | + ICE_FLOW_SEG_HDR_PFCP_SESSION, + ICE_HASH_IP_CTX_IP_PFCP }, + { ICE_FLOW_SEG_HDR_ETH | ICE_FLOW_SEG_HDR_VLAN | + ICE_FLOW_SEG_HDR_IPV6 | ICE_FLOW_SEG_HDR_UDP, + ICE_HASH_IP_CTX_IP_UDP }, + { ICE_FLOW_SEG_HDR_ETH | ICE_FLOW_SEG_HDR_VLAN | + ICE_FLOW_SEG_HDR_IPV6 | ICE_FLOW_SEG_HDR_TCP, + ICE_HASH_IP_CTX_IP_TCP }, + { ICE_FLOW_SEG_HDR_ETH | ICE_FLOW_SEG_HDR_VLAN | + ICE_FLOW_SEG_HDR_IPV6 | ICE_FLOW_SEG_HDR_SCTP, + ICE_HASH_IP_CTX_IP_SCTP }, + { ICE_FLOW_SEG_HDR_ETH | ICE_FLOW_SEG_HDR_VLAN | + ICE_FLOW_SEG_HDR_IPV6 | ICE_FLOW_SEG_HDR_IPV_OTHER, + ICE_HASH_IP_CTX_IP }, + /* the remaining mappings are used for default RSS */ + { ICE_FLOW_SEG_HDR_IPV4 | ICE_FLOW_SEG_HDR_UDP, + ICE_HASH_IP_CTX_IP_UDP }, + { ICE_FLOW_SEG_HDR_IPV4 | ICE_FLOW_SEG_HDR_TCP, + ICE_HASH_IP_CTX_IP_TCP }, + { ICE_FLOW_SEG_HDR_IPV4 | ICE_FLOW_SEG_HDR_SCTP, + ICE_HASH_IP_CTX_IP_SCTP }, + { ICE_FLOW_SEG_HDR_IPV4 | ICE_FLOW_SEG_HDR_IPV_OTHER, + ICE_HASH_IP_CTX_IP }, + { ICE_FLOW_SEG_HDR_IPV6 | ICE_FLOW_SEG_HDR_UDP, + ICE_HASH_IP_CTX_IP_UDP }, + { ICE_FLOW_SEG_HDR_IPV6 | ICE_FLOW_SEG_HDR_TCP, + ICE_HASH_IP_CTX_IP_TCP }, + { ICE_FLOW_SEG_HDR_IPV6 | ICE_FLOW_SEG_HDR_SCTP, + ICE_HASH_IP_CTX_IP_SCTP }, + { ICE_FLOW_SEG_HDR_IPV6 | ICE_FLOW_SEG_HDR_IPV_OTHER, + ICE_HASH_IP_CTX_IP }, + }; + + for (i = 0; i < ARRAY_SIZE(ip_ctx_idx_map); i++) { + if (hdrs == ip_ctx_idx_map[i].hdrs) + return ip_ctx_idx_map[i].ctx_idx; + } + + return ICE_HASH_IP_CTX_MAX; +} + +/** + * ice_add_rss_cfg_pre - Prepare RSS configuration context for a VF + * @vf: pointer to the VF structure + * @cfg: pointer to the RSS hash configuration + * + * Prepare the RSS hash context for a given VF based on the additional + * protocol headers specified in @cfg. This includes pre-configuration + * for IP and GTPU-based flows. + * + * If the configuration matches a known IP context, the function sets up + * the appropriate IP hash context. If the configuration includes GTPU + * headers, it prepares the GTPU-specific context accordingly. + * + * Return: 0 on success, or a negative error code on failure. + */ +static int +ice_add_rss_cfg_pre(struct ice_vf *vf, struct ice_rss_hash_cfg *cfg) +{ + u32 ice_gtpu_ctx_idx = ice_calc_gtpu_ctx_idx(cfg->addl_hdrs); + u8 ip_ctx_idx = ice_map_ip_ctx_idx(cfg->addl_hdrs); + + if (ip_ctx_idx == ICE_HASH_IP_CTX_IP) { + int ret = 0; + + if (cfg->addl_hdrs & ICE_FLOW_SEG_HDR_IPV4) + ret = ice_add_rss_cfg_pre_ip(vf, &vf->hash_ctx.v4); + else if (cfg->addl_hdrs & ICE_FLOW_SEG_HDR_IPV6) + ret = ice_add_rss_cfg_pre_ip(vf, &vf->hash_ctx.v6); + + if (ret) + return ret; + } + + if (cfg->addl_hdrs & ICE_FLOW_SEG_HDR_IPV4) { + return ice_add_rss_cfg_pre_gtpu(vf, &vf->hash_ctx.ipv4, + ice_gtpu_ctx_idx); + } else if (cfg->addl_hdrs & ICE_FLOW_SEG_HDR_IPV6) { + return ice_add_rss_cfg_pre_gtpu(vf, &vf->hash_ctx.ipv6, + ice_gtpu_ctx_idx); + } + + return 0; +} + +/** + * ice_add_rss_cfg_post_gtpu - Post-process GTPU RSS configuration + * @vf: pointer to the VF info + * @ctx: pointer to the context of the GTPU hash + * @cfg: pointer to the RSS hash configuration + * @ctx_idx: index of the hash context + * + * Post-processes the GTPU hash configuration after a new hash + * context has been successfully added. It updates the context with the new + * configuration and restores any previously removed hash contexts that need + * to be re-applied. This ensures proper TCAM rule ordering and avoids + * conflicts between overlapping GTPU rules. + * + * Return: 0 on success or a negative error code on failure + */ +static int ice_add_rss_cfg_post_gtpu(struct ice_vf *vf, + struct ice_vf_hash_gtpu_ctx *ctx, + struct ice_rss_hash_cfg *cfg, u32 ctx_idx) +{ + /* GTPU hash moveback lookup table indexed by context ID. + * Each entry is a bitmap indicating which contexts need moveback + * operations when the corresponding context index is processed. + */ + static const unsigned long + ice_gtpu_moveback_tbl[ICE_HASH_GTPU_CTX_MAX] = { + [ICE_HASH_GTPU_CTX_EH_IP] = 0, + [ICE_HASH_GTPU_CTX_EH_IP_UDP] = + BIT(ICE_HASH_GTPU_CTX_UP_IP) | + BIT(ICE_HASH_GTPU_CTX_UP_IP_TCP) | + BIT(ICE_HASH_GTPU_CTX_DW_IP) | + BIT(ICE_HASH_GTPU_CTX_DW_IP_TCP), + [ICE_HASH_GTPU_CTX_EH_IP_TCP] = + BIT(ICE_HASH_GTPU_CTX_UP_IP) | + BIT(ICE_HASH_GTPU_CTX_UP_IP_UDP) | + BIT(ICE_HASH_GTPU_CTX_DW_IP) | + BIT(ICE_HASH_GTPU_CTX_DW_IP_UDP), + [ICE_HASH_GTPU_CTX_UP_IP] = + BIT(ICE_HASH_GTPU_CTX_EH_IP) | + BIT(ICE_HASH_GTPU_CTX_EH_IP_UDP) | + BIT(ICE_HASH_GTPU_CTX_EH_IP_TCP), + [ICE_HASH_GTPU_CTX_UP_IP_UDP] = + BIT(ICE_HASH_GTPU_CTX_EH_IP) | + BIT(ICE_HASH_GTPU_CTX_EH_IP_UDP) | + BIT(ICE_HASH_GTPU_CTX_EH_IP_TCP), + [ICE_HASH_GTPU_CTX_UP_IP_TCP] = + BIT(ICE_HASH_GTPU_CTX_EH_IP) | + BIT(ICE_HASH_GTPU_CTX_EH_IP_UDP) | + BIT(ICE_HASH_GTPU_CTX_EH_IP_TCP), + [ICE_HASH_GTPU_CTX_DW_IP] = + BIT(ICE_HASH_GTPU_CTX_EH_IP) | + BIT(ICE_HASH_GTPU_CTX_EH_IP_UDP) | + BIT(ICE_HASH_GTPU_CTX_EH_IP_TCP), + [ICE_HASH_GTPU_CTX_DW_IP_UDP] = + BIT(ICE_HASH_GTPU_CTX_EH_IP) | + BIT(ICE_HASH_GTPU_CTX_EH_IP_UDP) | + BIT(ICE_HASH_GTPU_CTX_EH_IP_TCP), + [ICE_HASH_GTPU_CTX_DW_IP_TCP] = + BIT(ICE_HASH_GTPU_CTX_EH_IP) | + BIT(ICE_HASH_GTPU_CTX_EH_IP_UDP) | + BIT(ICE_HASH_GTPU_CTX_EH_IP_TCP), + }; + unsigned long moveback_mask; + int ret; + int i; + + if (unlikely(ctx_idx >= ICE_HASH_GTPU_CTX_MAX)) + return 0; + + ctx->ctx[ctx_idx].addl_hdrs = cfg->addl_hdrs; + ctx->ctx[ctx_idx].hash_flds = cfg->hash_flds; + ctx->ctx[ctx_idx].hdr_type = cfg->hdr_type; + ctx->ctx[ctx_idx].symm = cfg->symm; + + moveback_mask = ice_gtpu_moveback_tbl[ctx_idx]; + for_each_set_bit(i, &moveback_mask, ICE_HASH_GTPU_CTX_MAX) { + ret = ice_hash_moveback(vf, &ctx->ctx[i]); + if (ret && ret != -ENOENT) + return ret; + } + + return 0; +} + +static int +ice_add_rss_cfg_post(struct ice_vf *vf, struct ice_rss_hash_cfg *cfg) +{ + u32 ice_gtpu_ctx_idx = ice_calc_gtpu_ctx_idx(cfg->addl_hdrs); + u8 ip_ctx_idx = ice_map_ip_ctx_idx(cfg->addl_hdrs); + + if (ip_ctx_idx && ip_ctx_idx < ICE_HASH_IP_CTX_MAX) { + if (cfg->addl_hdrs & ICE_FLOW_SEG_HDR_IPV4) + ice_hash_cfg_record(&vf->hash_ctx.v4.ctx[ip_ctx_idx], cfg); + else if (cfg->addl_hdrs & ICE_FLOW_SEG_HDR_IPV6) + ice_hash_cfg_record(&vf->hash_ctx.v6.ctx[ip_ctx_idx], cfg); + } + + if (cfg->addl_hdrs & ICE_FLOW_SEG_HDR_IPV4) { + return ice_add_rss_cfg_post_gtpu(vf, &vf->hash_ctx.ipv4, + cfg, ice_gtpu_ctx_idx); + } else if (cfg->addl_hdrs & ICE_FLOW_SEG_HDR_IPV6) { + return ice_add_rss_cfg_post_gtpu(vf, &vf->hash_ctx.ipv6, + cfg, ice_gtpu_ctx_idx); + } + + return 0; +} + +/** + * ice_rem_rss_cfg_post - post-process the RSS configuration + * @vf: pointer to the VF info + * @cfg: pointer to the RSS hash configuration + * + * Post process the RSS hash configuration after deleting a hash + * config. Such as, it will reset the hash context for the GTPU hash. + */ +static void +ice_rem_rss_cfg_post(struct ice_vf *vf, struct ice_rss_hash_cfg *cfg) +{ + u32 ice_gtpu_ctx_idx = ice_calc_gtpu_ctx_idx(cfg->addl_hdrs); + u8 ip_ctx_idx = ice_map_ip_ctx_idx(cfg->addl_hdrs); + + if (ip_ctx_idx && ip_ctx_idx < ICE_HASH_IP_CTX_MAX) { + if (cfg->addl_hdrs & ICE_FLOW_SEG_HDR_IPV4) + ice_hash_cfg_reset(&vf->hash_ctx.v4.ctx[ip_ctx_idx]); + else if (cfg->addl_hdrs & ICE_FLOW_SEG_HDR_IPV6) + ice_hash_cfg_reset(&vf->hash_ctx.v6.ctx[ip_ctx_idx]); + } + + if (ice_gtpu_ctx_idx >= ICE_HASH_GTPU_CTX_MAX) + return; + + if (cfg->addl_hdrs & ICE_FLOW_SEG_HDR_IPV4) + ice_hash_cfg_reset(&vf->hash_ctx.ipv4.ctx[ice_gtpu_ctx_idx]); + else if (cfg->addl_hdrs & ICE_FLOW_SEG_HDR_IPV6) + ice_hash_cfg_reset(&vf->hash_ctx.ipv6.ctx[ice_gtpu_ctx_idx]); +} + +/** + * ice_rem_rss_cfg_wrap - Wrapper for deleting an RSS configuration + * @vf: pointer to the VF info + * @cfg: pointer to the RSS hash configuration + * + * Wrapper function to delete a flow profile base on an RSS configuration, + * and also post process the hash context base on the rollback mechanism + * which handle some rules conflict by ice_add_rss_cfg_wrap. + * + * Return: 0 on success; negative error code on failure. + */ +static int +ice_rem_rss_cfg_wrap(struct ice_vf *vf, struct ice_rss_hash_cfg *cfg) +{ + struct device *dev = ice_pf_to_dev(vf->pf); + struct ice_vsi *vsi = ice_get_vf_vsi(vf); + struct ice_hw *hw = &vf->pf->hw; + int ret; + + ret = ice_rem_rss_cfg(hw, vsi->idx, cfg); + /* We just ignore -ENOENT, because if two configurations share the same + * profile remove one of them actually removes both, since the + * profile is deleted. + */ + if (ret && ret != -ENOENT) { + dev_err(dev, "ice_rem_rss_cfg failed for VF %d, VSI %d, error:%d\n", + vf->vf_id, vf->lan_vsi_idx, ret); + return ret; + } + + ice_rem_rss_cfg_post(vf, cfg); + + return 0; +} + +/** + * ice_add_rss_cfg_wrap - Wrapper for adding an RSS configuration + * @vf: pointer to the VF info + * @cfg: pointer to the RSS hash configuration + * + * Add a flow profile based on an RSS configuration. Use a rollback + * mechanism to handle rule conflicts due to TCAM + * write sequence from top to down. + * + * Return: 0 on success; negative error code on failure. + */ +static int +ice_add_rss_cfg_wrap(struct ice_vf *vf, struct ice_rss_hash_cfg *cfg) +{ + struct device *dev = ice_pf_to_dev(vf->pf); + struct ice_vsi *vsi = ice_get_vf_vsi(vf); + struct ice_hw *hw = &vf->pf->hw; + int ret; + + if (ice_add_rss_cfg_pre(vf, cfg)) + return -EINVAL; + + ret = ice_add_rss_cfg(hw, vsi, cfg); + if (ret) { + dev_err(dev, "ice_add_rss_cfg failed for VF %d, VSI %d, error:%d\n", + vf->vf_id, vf->lan_vsi_idx, ret); + return ret; + } + + if (ice_add_rss_cfg_post(vf, cfg)) + ret = -EINVAL; + + return ret; +} + +/** + * ice_parse_raw_rss_pattern - Parse raw pattern spec and mask for RSS + * @vf: pointer to the VF info + * @proto: pointer to the virtchnl protocol header + * @raw_cfg: pointer to the RSS raw pattern configuration + * + * Parser function to get spec and mask from virtchnl message, and parse + * them to get the corresponding profile and offset. The profile is used + * to add RSS configuration. + * + * Return: 0 on success; negative error code on failure. + */ +static int +ice_parse_raw_rss_pattern(struct ice_vf *vf, struct virtchnl_proto_hdrs *proto, + struct ice_rss_raw_cfg *raw_cfg) +{ + struct ice_parser_result pkt_parsed; + struct ice_hw *hw = &vf->pf->hw; + struct ice_parser_profile prof; + struct ice_parser *psr; + u8 *pkt_buf, *msk_buf; + u16 pkt_len; + int ret = 0; + + pkt_len = proto->raw.pkt_len; + if (!pkt_len) + return -EINVAL; + if (pkt_len > VIRTCHNL_MAX_SIZE_RAW_PACKET) + pkt_len = VIRTCHNL_MAX_SIZE_RAW_PACKET; + + pkt_buf = kzalloc(pkt_len, GFP_KERNEL); + msk_buf = kzalloc(pkt_len, GFP_KERNEL); + if (!pkt_buf || !msk_buf) { + ret = -ENOMEM; + goto free_alloc; + } + + memcpy(pkt_buf, proto->raw.spec, pkt_len); + memcpy(msk_buf, proto->raw.mask, pkt_len); + + psr = ice_parser_create(hw); + if (IS_ERR(psr)) { + ret = PTR_ERR(psr); + goto free_alloc; + } + + ret = ice_parser_run(psr, pkt_buf, pkt_len, &pkt_parsed); + if (ret) + goto parser_destroy; + + ret = ice_parser_profile_init(&pkt_parsed, pkt_buf, msk_buf, + pkt_len, ICE_BLK_RSS, &prof); + if (ret) + goto parser_destroy; + + memcpy(&raw_cfg->prof, &prof, sizeof(prof)); + +parser_destroy: + ice_parser_destroy(psr); +free_alloc: + kfree(pkt_buf); + kfree(msk_buf); + return ret; +} + +/** + * ice_add_raw_rss_cfg - add RSS configuration for raw pattern + * @vf: pointer to the VF info + * @cfg: pointer to the RSS raw pattern configuration + * + * This function adds the RSS configuration for raw pattern. + * Check if current profile is matched. If not, remove the old + * one and add the new profile to HW directly. Update the symmetric + * hash configuration as well. + * + * Return: 0 on success; negative error code on failure. + */ +static int +ice_add_raw_rss_cfg(struct ice_vf *vf, struct ice_rss_raw_cfg *cfg) +{ + struct ice_parser_profile *prof = &cfg->prof; + struct device *dev = ice_pf_to_dev(vf->pf); + struct ice_rss_prof_info *rss_prof; + struct ice_hw *hw = &vf->pf->hw; + int i, ptg, ret = 0; + u16 vsi_handle; + u64 id; + + vsi_handle = vf->lan_vsi_idx; + id = find_first_bit(prof->ptypes, ICE_FLOW_PTYPE_MAX); + + ptg = hw->blk[ICE_BLK_RSS].xlt1.t[id]; + rss_prof = &vf->rss_prof_info[ptg]; + + /* check if ptg already has a profile */ + if (rss_prof->prof.fv_num) { + for (i = 0; i < ICE_MAX_FV_WORDS; i++) { + if (rss_prof->prof.fv[i].proto_id != + prof->fv[i].proto_id || + rss_prof->prof.fv[i].offset != + prof->fv[i].offset) + break; + } + + /* current profile is matched, check symmetric hash */ + if (i == ICE_MAX_FV_WORDS) { + if (rss_prof->symm != cfg->symm) + goto update_symm; + return ret; + } + + /* current profile is not matched, remove it */ + ret = + ice_rem_prof_id_flow(hw, ICE_BLK_RSS, + ice_get_hw_vsi_num(hw, vsi_handle), + id); + if (ret) { + dev_err(dev, "remove RSS flow failed\n"); + return ret; + } + + ret = ice_rem_prof(hw, ICE_BLK_RSS, id); + if (ret) { + dev_err(dev, "remove RSS profile failed\n"); + return ret; + } + } + + /* add new profile */ + ret = ice_flow_set_parser_prof(hw, vsi_handle, 0, prof, ICE_BLK_RSS); + if (ret) { + dev_err(dev, "HW profile add failed\n"); + return ret; + } + + memcpy(&rss_prof->prof, prof, sizeof(struct ice_parser_profile)); + +update_symm: + rss_prof->symm = cfg->symm; + ice_rss_update_raw_symm(hw, cfg, id); + return ret; +} + +/** + * ice_rem_raw_rss_cfg - remove RSS configuration for raw pattern + * @vf: pointer to the VF info + * @cfg: pointer to the RSS raw pattern configuration + * + * This function removes the RSS configuration for raw pattern. + * Check if vsi group is already removed first. If not, remove the + * profile. + * + * Return: 0 on success; negative error code on failure. + */ +static int +ice_rem_raw_rss_cfg(struct ice_vf *vf, struct ice_rss_raw_cfg *cfg) +{ + struct ice_parser_profile *prof = &cfg->prof; + struct device *dev = ice_pf_to_dev(vf->pf); + struct ice_hw *hw = &vf->pf->hw; + int ptg, ret = 0; + u16 vsig, vsi; + u64 id; + + id = find_first_bit(prof->ptypes, ICE_FLOW_PTYPE_MAX); + + ptg = hw->blk[ICE_BLK_RSS].xlt1.t[id]; + + memset(&vf->rss_prof_info[ptg], 0, + sizeof(struct ice_rss_prof_info)); + + /* check if vsig is already removed */ + vsi = ice_get_hw_vsi_num(hw, vf->lan_vsi_idx); + if (vsi >= ICE_MAX_VSI) { + ret = -EINVAL; + goto err; + } + + vsig = hw->blk[ICE_BLK_RSS].xlt2.vsis[vsi].vsig; + if (vsig) { + ret = ice_rem_prof_id_flow(hw, ICE_BLK_RSS, vsi, id); + if (ret) + goto err; + + ret = ice_rem_prof(hw, ICE_BLK_RSS, id); + if (ret) + goto err; + } + + return ret; + +err: + dev_err(dev, "HW profile remove failed\n"); + return ret; +} + /** * ice_vc_handle_rss_cfg * @vf: pointer to the VF info @@ -569,6 +1569,9 @@ int ice_vc_handle_rss_cfg(struct ice_vf *vf, u8 *msg, bool add) struct device *dev = ice_pf_to_dev(vf->pf); struct ice_hw *hw = &vf->pf->hw; struct ice_vsi *vsi; + u8 hash_type; + bool symm; + int ret; if (!test_bit(ICE_FLAG_RSS_ENA, vf->pf->flags)) { dev_dbg(dev, "VF %d attempting to configure RSS, but RSS is not supported by the PF\n", @@ -604,49 +1607,44 @@ int ice_vc_handle_rss_cfg(struct ice_vf *vf, u8 *msg, bool add) goto error_param; } - if (!ice_vc_validate_pattern(vf, &rss_cfg->proto_hdrs)) { - v_ret = VIRTCHNL_STATUS_ERR_PARAM; + if (rss_cfg->rss_algorithm == VIRTCHNL_RSS_ALG_R_ASYMMETRIC) { + hash_type = add ? ICE_AQ_VSI_Q_OPT_RSS_HASH_XOR : + ICE_AQ_VSI_Q_OPT_RSS_HASH_TPLZ; + + ret = ice_vc_rss_hash_update(hw, vsi, hash_type); + if (ret) + v_ret = ice_err_to_virt_err(ret); goto error_param; } - if (rss_cfg->rss_algorithm == VIRTCHNL_RSS_ALG_R_ASYMMETRIC) { - struct ice_vsi_ctx *ctx; - u8 lut_type, hash_type; - int status; + hash_type = add ? ICE_AQ_VSI_Q_OPT_RSS_HASH_SYM_TPLZ : + ICE_AQ_VSI_Q_OPT_RSS_HASH_TPLZ; + ret = ice_vc_rss_hash_update(hw, vsi, hash_type); + if (ret) { + v_ret = ice_err_to_virt_err(ret); + goto error_param; + } - lut_type = ICE_AQ_VSI_Q_OPT_RSS_LUT_VSI; - hash_type = add ? ICE_AQ_VSI_Q_OPT_RSS_HASH_XOR : - ICE_AQ_VSI_Q_OPT_RSS_HASH_TPLZ; + symm = rss_cfg->rss_algorithm == VIRTCHNL_RSS_ALG_TOEPLITZ_SYMMETRIC; + /* Configure RSS hash for raw pattern */ + if (rss_cfg->proto_hdrs.tunnel_level == 0 && + rss_cfg->proto_hdrs.count == 0) { + struct ice_rss_raw_cfg raw_cfg; - ctx = kzalloc(sizeof(*ctx), GFP_KERNEL); - if (!ctx) { - v_ret = VIRTCHNL_STATUS_ERR_NO_MEMORY; + if (ice_parse_raw_rss_pattern(vf, &rss_cfg->proto_hdrs, + &raw_cfg)) { + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } - ctx->info.q_opt_rss = - FIELD_PREP(ICE_AQ_VSI_Q_OPT_RSS_LUT_M, lut_type) | - FIELD_PREP(ICE_AQ_VSI_Q_OPT_RSS_HASH_M, hash_type); - - /* Preserve existing queueing option setting */ - ctx->info.q_opt_rss |= (vsi->info.q_opt_rss & - ICE_AQ_VSI_Q_OPT_RSS_GBL_LUT_M); - ctx->info.q_opt_tc = vsi->info.q_opt_tc; - ctx->info.q_opt_flags = vsi->info.q_opt_rss; - - ctx->info.valid_sections = - cpu_to_le16(ICE_AQ_VSI_PROP_Q_OPT_VALID); - - status = ice_update_vsi(hw, vsi->idx, ctx, NULL); - if (status) { - dev_err(dev, "update VSI for RSS failed, err %d aq_err %s\n", - status, libie_aq_str(hw->adminq.sq_last_status)); - v_ret = VIRTCHNL_STATUS_ERR_PARAM; + if (add) { + raw_cfg.symm = symm; + if (ice_add_raw_rss_cfg(vf, &raw_cfg)) + v_ret = VIRTCHNL_STATUS_ERR_PARAM; } else { - vsi->info.q_opt_rss = ctx->info.q_opt_rss; + if (ice_rem_raw_rss_cfg(vf, &raw_cfg)) + v_ret = VIRTCHNL_STATUS_ERR_PARAM; } - - kfree(ctx); } else { struct ice_rss_hash_cfg cfg; @@ -665,24 +1663,12 @@ int ice_vc_handle_rss_cfg(struct ice_vf *vf, u8 *msg, bool add) } if (add) { - if (ice_add_rss_cfg(hw, vsi, &cfg)) { + cfg.symm = symm; + if (ice_add_rss_cfg_wrap(vf, &cfg)) v_ret = VIRTCHNL_STATUS_ERR_PARAM; - dev_err(dev, "ice_add_rss_cfg failed for vsi = %d, v_ret = %d\n", - vsi->vsi_num, v_ret); - } } else { - int status; - - status = ice_rem_rss_cfg(hw, vsi->idx, &cfg); - /* We just ignore -ENOENT, because if two configurations - * share the same profile remove one of them actually - * removes both, since the profile is deleted. - */ - if (status && status != -ENOENT) { + if (ice_rem_rss_cfg_wrap(vf, &cfg)) v_ret = VIRTCHNL_STATUS_ERR_PARAM; - dev_err(dev, "ice_rem_rss_cfg failed for VF ID:%d, error:%d\n", - vf->vf_id, status); - } } } From f89e4e1512333fcb92474e0e8487929b352c50de Mon Sep 17 00:00:00 2001 From: Aleksandr Loktionov Date: Thu, 30 Oct 2025 14:59:48 +0100 Subject: [PATCH 0583/1381] ice: improve TCAM priority handling for RSS profiles Enhance TCAM priority logic to avoid conflicts between RSS profiles with overlapping PTGs and attributes. Track used PTG and attribute combinations. Ensure higher-priority profiles override lower ones. Add helper for setting TCAM flags and masks. Ensure RSS rule consistency and prevent unintended matches. Co-developed-by: Dan Nowlin Signed-off-by: Dan Nowlin Signed-off-by: Przemek Kitszel Reviewed-by: Simon Horman Signed-off-by: Aleksandr Loktionov Signed-off-by: Tony Nguyen --- .../net/ethernet/intel/ice/ice_flex_pipe.c | 91 ++++++++++++++++--- .../net/ethernet/intel/ice/ice_flex_type.h | 1 + 2 files changed, 78 insertions(+), 14 deletions(-) diff --git a/drivers/net/ethernet/intel/ice/ice_flex_pipe.c b/drivers/net/ethernet/intel/ice/ice_flex_pipe.c index c8cb492fddf4..c0dbec369366 100644 --- a/drivers/net/ethernet/intel/ice/ice_flex_pipe.c +++ b/drivers/net/ethernet/intel/ice/ice_flex_pipe.c @@ -3577,6 +3577,19 @@ ice_move_vsi(struct ice_hw *hw, enum ice_block blk, u16 vsi, u16 vsig, return 0; } +/** + * ice_set_tcam_flags - set TCAM flag don't care mask + * @mask: mask for flags + * @dc_mask: pointer to the don't care mask + */ +static void ice_set_tcam_flags(u16 mask, u8 dc_mask[ICE_TCAM_KEY_VAL_SZ]) +{ + u16 inverted_mask = ~mask; + + /* flags are lowest u16 */ + put_unaligned_le16(inverted_mask, dc_mask); +} + /** * ice_rem_chg_tcam_ent - remove a specific TCAM entry from change list * @hw: pointer to the HW struct @@ -3647,6 +3660,9 @@ ice_prof_tcam_ena_dis(struct ice_hw *hw, enum ice_block blk, bool enable, if (!p) return -ENOMEM; + /* set don't care masks for TCAM flags */ + ice_set_tcam_flags(tcam->attr.mask, dc_msk); + status = ice_tcam_write_entry(hw, blk, tcam->tcam_idx, tcam->prof_id, tcam->ptg, vsig, 0, tcam->attr.flags, vl_msk, dc_msk, nm_msk); @@ -3672,6 +3688,34 @@ ice_prof_tcam_ena_dis(struct ice_hw *hw, enum ice_block blk, bool enable, return status; } +/** + * ice_ptg_attr_in_use - determine if PTG and attribute pair is in use + * @ptg_attr: pointer to the PTG and attribute pair to check + * @ptgs_used: bitmap that denotes which PTGs are in use + * @attr_used: array of PTG and attributes pairs already used + * @attr_cnt: count of entries in the attr_used array + * + * Return: true if the PTG and attribute pair is in use, false otherwise. + */ +static bool +ice_ptg_attr_in_use(struct ice_tcam_inf *ptg_attr, unsigned long *ptgs_used, + struct ice_tcam_inf *attr_used[], u16 attr_cnt) +{ + u16 i; + + if (!test_bit(ptg_attr->ptg, ptgs_used)) + return false; + + /* the PTG is used, so now look for correct attributes */ + for (i = 0; i < attr_cnt; i++) + if (attr_used[i]->ptg == ptg_attr->ptg && + attr_used[i]->attr.flags == ptg_attr->attr.flags && + attr_used[i]->attr.mask == ptg_attr->attr.mask) + return true; + + return false; +} + /** * ice_adj_prof_priorities - adjust profile based on priorities * @hw: pointer to the HW struct @@ -3684,10 +3728,16 @@ ice_adj_prof_priorities(struct ice_hw *hw, enum ice_block blk, u16 vsig, struct list_head *chg) { DECLARE_BITMAP(ptgs_used, ICE_XLT1_CNT); + struct ice_tcam_inf **attr_used; struct ice_vsig_prof *t; - int status; + u16 attr_used_cnt = 0; + int status = 0; u16 idx; + attr_used = kcalloc(ICE_MAX_PTG_ATTRS, sizeof(*attr_used), GFP_KERNEL); + if (!attr_used) + return -ENOMEM; + bitmap_zero(ptgs_used, ICE_XLT1_CNT); idx = vsig & ICE_VSIG_IDX_M; @@ -3705,11 +3755,15 @@ ice_adj_prof_priorities(struct ice_hw *hw, enum ice_block blk, u16 vsig, u16 i; for (i = 0; i < t->tcam_count; i++) { + bool used; + /* Scan the priorities from newest to oldest. * Make sure that the newest profiles take priority. */ - if (test_bit(t->tcam[i].ptg, ptgs_used) && - t->tcam[i].in_use) { + used = ice_ptg_attr_in_use(&t->tcam[i], ptgs_used, + attr_used, attr_used_cnt); + + if (used && t->tcam[i].in_use) { /* need to mark this PTG as never match, as it * was already in use and therefore duplicate * (and lower priority) @@ -3719,9 +3773,8 @@ ice_adj_prof_priorities(struct ice_hw *hw, enum ice_block blk, u16 vsig, &t->tcam[i], chg); if (status) - return status; - } else if (!test_bit(t->tcam[i].ptg, ptgs_used) && - !t->tcam[i].in_use) { + goto free_attr_used; + } else if (!used && !t->tcam[i].in_use) { /* need to enable this PTG, as it in not in use * and not enabled (highest priority) */ @@ -3730,15 +3783,21 @@ ice_adj_prof_priorities(struct ice_hw *hw, enum ice_block blk, u16 vsig, &t->tcam[i], chg); if (status) - return status; + goto free_attr_used; } /* keep track of used ptgs */ - __set_bit(t->tcam[i].ptg, ptgs_used); + set_bit(t->tcam[i].ptg, ptgs_used); + if (attr_used_cnt < ICE_MAX_PTG_ATTRS) + attr_used[attr_used_cnt++] = &t->tcam[i]; + else + ice_debug(hw, ICE_DBG_INIT, "Warn: ICE_MAX_PTG_ATTRS exceeded\n"); } } - return 0; +free_attr_used: + kfree(attr_used); + return status; } /** @@ -3821,11 +3880,15 @@ ice_add_prof_id_vsig(struct ice_hw *hw, enum ice_block blk, u16 vsig, u64 hdl, p->vsig = vsig; p->tcam_idx = t->tcam[i].tcam_idx; + /* set don't care masks for TCAM flags */ + ice_set_tcam_flags(t->tcam[i].attr.mask, dc_msk); + /* write the TCAM entry */ status = ice_tcam_write_entry(hw, blk, t->tcam[i].tcam_idx, t->tcam[i].prof_id, - t->tcam[i].ptg, vsig, 0, 0, - vl_msk, dc_msk, nm_msk); + t->tcam[i].ptg, vsig, 0, + t->tcam[i].attr.flags, vl_msk, + dc_msk, nm_msk); if (status) { devm_kfree(ice_hw_to_dev(hw), p); goto err_ice_add_prof_id_vsig; @@ -4139,9 +4202,6 @@ ice_flow_assoc_fdir_prof(struct ice_hw *hw, enum ice_block blk, u16 vsi_num; int status; - if (blk != ICE_BLK_FD) - return -EINVAL; - vsi_num = ice_get_hw_vsi_num(hw, dest_vsi); status = ice_add_prof_id_flow(hw, blk, vsi_num, hdl); if (status) { @@ -4150,6 +4210,9 @@ ice_flow_assoc_fdir_prof(struct ice_hw *hw, enum ice_block blk, return status; } + if (blk != ICE_BLK_FD) + return 0; + vsi_num = ice_get_hw_vsi_num(hw, fdir_vsi); status = ice_add_prof_id_flow(hw, blk, vsi_num, hdl); if (status) { diff --git a/drivers/net/ethernet/intel/ice/ice_flex_type.h b/drivers/net/ethernet/intel/ice/ice_flex_type.h index 817beca591e0..80c9e7c749c2 100644 --- a/drivers/net/ethernet/intel/ice/ice_flex_type.h +++ b/drivers/net/ethernet/intel/ice/ice_flex_type.h @@ -187,6 +187,7 @@ struct ice_prof_map { }; #define ICE_INVALID_TCAM 0xFFFF +#define ICE_MAX_PTG_ATTRS 1024 struct ice_tcam_inf { u16 tcam_idx; From 41e880eb8482499f0acb6371b30fe7d8dd44eb8b Mon Sep 17 00:00:00 2001 From: Przemek Kitszel Date: Thu, 30 Oct 2025 14:59:49 +0100 Subject: [PATCH 0584/1381] ice: Extend PTYPE bitmap coverage for GTP encapsulated flows Consolidate updates to the Protocol Type (PTYPE) bitmap definitions across multiple flow types in the Intel ICE driver to support GTP (GPRS Tunneling Protocol) encapsulated traffic. Enable improved Receive Side Scaling (RSS) configuration for both user and control plane GTP flows. Cover a wide range of protocol and encapsulation scenarios, including: - MAC OFOS and IL - IPv4 and IPv6 (OFOS, IL, ALL, no-L4) - TCP, SCTP, ICMP - GRE OF - GTPC (control plane) Expand the PTYPE bitmap entries to improve classification and distribution of GTP traffic across multiple queues, enhancing performance and scalability in mobile network environments. Co-developed-by: Dan Nowlin Signed-off-by: Dan Nowlin Co-developed-by: Qi Zhang Signed-off-by: Qi Zhang Co-developed-by: Jie Wang Signed-off-by: Jie Wang Co-developed-by: Junfeng Guo Signed-off-by: Junfeng Guo Signed-off-by: Przemek Kitszel Reviewed-by: Jedrzej Jagielski Reviewed-by: Simon Horman Signed-off-by: Aleksandr Loktionov Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_flow.c | 52 +++++++++++------------ 1 file changed, 26 insertions(+), 26 deletions(-) diff --git a/drivers/net/ethernet/intel/ice/ice_flow.c b/drivers/net/ethernet/intel/ice/ice_flow.c index a2f2a612428d..c9b6d0a84bd1 100644 --- a/drivers/net/ethernet/intel/ice/ice_flow.c +++ b/drivers/net/ethernet/intel/ice/ice_flow.c @@ -220,9 +220,9 @@ struct ice_flow_field_info ice_flds_info[ICE_FLOW_FIELD_IDX_MAX] = { */ static const u32 ice_ptypes_mac_ofos[] = { 0xFDC00846, 0xBFBF7F7E, 0xF70001DF, 0xFEFDFDFB, - 0x0000077E, 0x00000000, 0x00000000, 0x00000000, - 0x00400000, 0x03FFF000, 0x7FFFFFE0, 0x00000000, - 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x0000077E, 0x000003FF, 0x00000000, 0x00000000, + 0x00400000, 0x03FFF000, 0xFFFFFFE0, 0x00000707, + 0xFFFFF000, 0x000003FF, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, @@ -245,10 +245,10 @@ static const u32 ice_ptypes_macvlan_il[] = { * include IPv4 other PTYPEs */ static const u32 ice_ptypes_ipv4_ofos[] = { - 0x1DC00000, 0x04000800, 0x00000000, 0x00000000, + 0x1D800000, 0xBFBF7800, 0x000001DF, 0x00000000, 0x00000000, 0x00000155, 0x00000000, 0x00000000, - 0x00000000, 0x000FC000, 0x00000000, 0x00000000, - 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x000FC000, 0x000002A0, 0x00000000, + 0x00015000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, @@ -259,10 +259,10 @@ static const u32 ice_ptypes_ipv4_ofos[] = { * IPv4 other PTYPEs */ static const u32 ice_ptypes_ipv4_ofos_all[] = { - 0x1DC00000, 0x04000800, 0x00000000, 0x00000000, + 0x1D800000, 0x27BF7800, 0x00000000, 0x00000000, 0x00000000, 0x00000155, 0x00000000, 0x00000000, - 0x00000000, 0x000FC000, 0x83E0F800, 0x00000101, - 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x000FC000, 0x83E0FAA0, 0x00000101, + 0x3FFD5000, 0x00000000, 0x02FBEFBC, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, @@ -274,7 +274,7 @@ static const u32 ice_ptypes_ipv4_il[] = { 0xE0000000, 0xB807700E, 0x80000003, 0xE01DC03B, 0x0000000E, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x001FF800, 0x00000000, - 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0xC0FC0000, 0x0000000F, 0xBC0BC0BC, 0x00000BC0, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, @@ -285,10 +285,10 @@ static const u32 ice_ptypes_ipv4_il[] = { * include IPv6 other PTYPEs */ static const u32 ice_ptypes_ipv6_ofos[] = { - 0x00000000, 0x00000000, 0x77000000, 0x10002000, + 0x00000000, 0x00000000, 0x76000000, 0x10002000, 0x00000000, 0x000002AA, 0x00000000, 0x00000000, - 0x00000000, 0x03F00000, 0x00000000, 0x00000000, - 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x03F00000, 0x00000540, 0x00000000, + 0x0002A000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, @@ -299,10 +299,10 @@ static const u32 ice_ptypes_ipv6_ofos[] = { * IPv6 other PTYPEs */ static const u32 ice_ptypes_ipv6_ofos_all[] = { - 0x00000000, 0x00000000, 0x77000000, 0x10002000, - 0x00000000, 0x000002AA, 0x00000000, 0x00000000, - 0x00080F00, 0x03F00000, 0x7C1F0000, 0x00000206, - 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x76000000, 0xFEFDE000, + 0x0000077E, 0x000002AA, 0x00000000, 0x00000000, + 0x00000000, 0x03F00000, 0x7C1F0540, 0x00000206, + 0xC002A000, 0x000003FF, 0xBC000000, 0x0002FBEF, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, @@ -314,7 +314,7 @@ static const u32 ice_ptypes_ipv6_il[] = { 0x00000000, 0x03B80770, 0x000001DC, 0x0EE00000, 0x00000770, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x7FE00000, 0x00000000, - 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x3F000000, 0x000003F0, 0x02F02F00, 0x0002F02F, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, @@ -387,8 +387,8 @@ static const u32 ice_ptypes_ipv6_il_no_l4[] = { static const u32 ice_ptypes_udp_il[] = { 0x81000000, 0x20204040, 0x04000010, 0x80810102, 0x00000040, 0x00000000, 0x00000000, 0x00000000, - 0x00000000, 0x00410000, 0x90842000, 0x00000007, - 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00410000, 0x908427E0, 0x00000007, + 0x0413F000, 0x00000041, 0x10410410, 0x00004104, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, @@ -400,7 +400,7 @@ static const u32 ice_ptypes_tcp_il[] = { 0x04000000, 0x80810102, 0x10000040, 0x02040408, 0x00000102, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00820000, 0x21084000, 0x00000000, - 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x08200000, 0x00000082, 0x20820820, 0x00008208, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, @@ -412,7 +412,7 @@ static const u32 ice_ptypes_sctp_il[] = { 0x08000000, 0x01020204, 0x20000081, 0x04080810, 0x00000204, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x01040000, 0x00000000, 0x00000000, - 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x10400000, 0x00000104, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, @@ -436,7 +436,7 @@ static const u32 ice_ptypes_icmp_il[] = { 0x00000000, 0x02040408, 0x40000102, 0x08101020, 0x00000408, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x42108000, 0x00000000, - 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x20800000, 0x00000208, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, @@ -448,7 +448,7 @@ static const u32 ice_ptypes_gre_of[] = { 0x00000000, 0xBFBF7800, 0x000001DF, 0xFEFDE000, 0x0000017E, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, - 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0xBEFBEFBC, 0x0002FBEF, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, @@ -457,7 +457,7 @@ static const u32 ice_ptypes_gre_of[] = { /* Packet types for packets with an Innermost/Last MAC header */ static const u32 ice_ptypes_mac_il[] = { - 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x20000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, @@ -471,7 +471,7 @@ static const u32 ice_ptypes_mac_il[] = { static const u32 ice_ptypes_gtpc[] = { 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, - 0x00000000, 0x00000000, 0x00000180, 0x00000000, + 0x00000000, 0x00000000, 0x000001E0, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, From 3da28eb277c1c175df1aa048e404cc765ae04327 Mon Sep 17 00:00:00 2001 From: Aleksandr Loktionov Date: Thu, 30 Oct 2025 14:59:50 +0100 Subject: [PATCH 0585/1381] iavf: add RSS support for GTP protocol via ethtool Extend the iavf driver to support Receive Side Scaling (RSS) configuration for GTP (GPRS Tunneling Protocol) flows using ethtool. The implementation introduces new RSS flow segment headers and hash field definitions for various GTP encapsulations, including: - GTPC - GTPU (IP, Extension Header, Uplink, Downlink) - TEID-based hashing The ethtool interface is updated to parse and apply these new flow types and hash fields, enabling fine-grained traffic distribution for GTP-based mobile workloads. This enhancement improves performance and scalability for virtualized network functions (VNFs) and user plane functions (UPFs) in 5G and LTE deployments. Reviewed-by: Jedrzej Jagielski Reviewed-by: Simon Horman Signed-off-by: Aleksandr Loktionov Signed-off-by: Tony Nguyen --- .../net/ethernet/intel/iavf/iavf_adv_rss.c | 119 ++++++++++++++---- .../net/ethernet/intel/iavf/iavf_adv_rss.h | 31 +++++ .../net/ethernet/intel/iavf/iavf_ethtool.c | 89 +++++++++++++ 3 files changed, 216 insertions(+), 23 deletions(-) diff --git a/drivers/net/ethernet/intel/iavf/iavf_adv_rss.c b/drivers/net/ethernet/intel/iavf/iavf_adv_rss.c index a9e1da35e248..4d12dfe1b481 100644 --- a/drivers/net/ethernet/intel/iavf/iavf_adv_rss.c +++ b/drivers/net/ethernet/intel/iavf/iavf_adv_rss.c @@ -90,6 +90,55 @@ iavf_fill_adv_rss_sctp_hdr(struct virtchnl_proto_hdr *hdr, u64 hash_flds) VIRTCHNL_ADD_PROTO_HDR_FIELD_BIT(hdr, SCTP, DST_PORT); } +/** + * iavf_fill_adv_rss_gtp_hdr - Fill GTP-related RSS protocol headers + * @proto_hdrs: pointer to the virtchnl protocol headers structure to populate + * @packet_hdrs: bitmask of packet header types to configure + * @hash_flds: RSS hash field configuration + * + * This function populates the virtchnl protocol header structure with + * appropriate GTP-related header types based on the specified packet_hdrs. + * It supports GTPC, GTPU with extension headers, and uplink/downlink PDU + * types. For certain GTPU types, it also appends an IPv4 header to enable + * hashing on the destination IP address. + * + * Return: 0 on success or -EOPNOTSUPP if the packet_hdrs value is unsupported. + */ +static int +iavf_fill_adv_rss_gtp_hdr(struct virtchnl_proto_hdrs *proto_hdrs, + u32 packet_hdrs, u64 hash_flds) +{ + struct virtchnl_proto_hdr *hdr; + + hdr = &proto_hdrs->proto_hdr[proto_hdrs->count - 1]; + + switch (packet_hdrs & IAVF_ADV_RSS_FLOW_SEG_HDR_GTP) { + case IAVF_ADV_RSS_FLOW_SEG_HDR_GTPC_TEID: + case IAVF_ADV_RSS_FLOW_SEG_HDR_GTPC: + VIRTCHNL_SET_PROTO_HDR_TYPE(hdr, GTPC); + break; + case IAVF_ADV_RSS_FLOW_SEG_HDR_GTPU_EH: + VIRTCHNL_SET_PROTO_HDR_TYPE(hdr, GTPU_EH); + break; + case IAVF_ADV_RSS_FLOW_SEG_HDR_GTPU_UP: + VIRTCHNL_SET_PROTO_HDR_TYPE(hdr, GTPU_EH_PDU_UP); + hdr = &proto_hdrs->proto_hdr[proto_hdrs->count++]; + iavf_fill_adv_rss_ip4_hdr(hdr, IAVF_ADV_RSS_HASH_FLD_IPV4_DA); + break; + case IAVF_ADV_RSS_FLOW_SEG_HDR_GTPU_DWN: + VIRTCHNL_SET_PROTO_HDR_TYPE(hdr, GTPU_EH_PDU_DWN); + fallthrough; + case IAVF_ADV_RSS_FLOW_SEG_HDR_GTPU_IP: + hdr = &proto_hdrs->proto_hdr[proto_hdrs->count++]; + iavf_fill_adv_rss_ip4_hdr(hdr, IAVF_ADV_RSS_HASH_FLD_IPV4_DA); + break; + default: + return -EOPNOTSUPP; + } + + return 0; +} + /** * iavf_fill_adv_rss_cfg_msg - fill the RSS configuration into virtchnl message * @rss_cfg: the virtchnl message to be filled with RSS configuration setting @@ -103,6 +152,8 @@ int iavf_fill_adv_rss_cfg_msg(struct virtchnl_rss_cfg *rss_cfg, u32 packet_hdrs, u64 hash_flds, bool symm) { + const u32 packet_l3_hdrs = packet_hdrs & IAVF_ADV_RSS_FLOW_SEG_HDR_L3; + const u32 packet_l4_hdrs = packet_hdrs & IAVF_ADV_RSS_FLOW_SEG_HDR_L4; struct virtchnl_proto_hdrs *proto_hdrs = &rss_cfg->proto_hdrs; struct virtchnl_proto_hdr *hdr; @@ -113,31 +164,41 @@ iavf_fill_adv_rss_cfg_msg(struct virtchnl_rss_cfg *rss_cfg, proto_hdrs->tunnel_level = 0; /* always outer layer */ - hdr = &proto_hdrs->proto_hdr[proto_hdrs->count++]; - switch (packet_hdrs & IAVF_ADV_RSS_FLOW_SEG_HDR_L3) { - case IAVF_ADV_RSS_FLOW_SEG_HDR_IPV4: - iavf_fill_adv_rss_ip4_hdr(hdr, hash_flds); - break; - case IAVF_ADV_RSS_FLOW_SEG_HDR_IPV6: - iavf_fill_adv_rss_ip6_hdr(hdr, hash_flds); - break; - default: - return -EINVAL; + if (packet_l3_hdrs) { + hdr = &proto_hdrs->proto_hdr[proto_hdrs->count++]; + switch (packet_l3_hdrs) { + case IAVF_ADV_RSS_FLOW_SEG_HDR_IPV4: + iavf_fill_adv_rss_ip4_hdr(hdr, hash_flds); + break; + case IAVF_ADV_RSS_FLOW_SEG_HDR_IPV6: + iavf_fill_adv_rss_ip6_hdr(hdr, hash_flds); + break; + default: + return -EINVAL; + } } - hdr = &proto_hdrs->proto_hdr[proto_hdrs->count++]; - switch (packet_hdrs & IAVF_ADV_RSS_FLOW_SEG_HDR_L4) { - case IAVF_ADV_RSS_FLOW_SEG_HDR_TCP: - iavf_fill_adv_rss_tcp_hdr(hdr, hash_flds); - break; - case IAVF_ADV_RSS_FLOW_SEG_HDR_UDP: - iavf_fill_adv_rss_udp_hdr(hdr, hash_flds); - break; - case IAVF_ADV_RSS_FLOW_SEG_HDR_SCTP: - iavf_fill_adv_rss_sctp_hdr(hdr, hash_flds); - break; - default: - return -EINVAL; + if (packet_l4_hdrs) { + hdr = &proto_hdrs->proto_hdr[proto_hdrs->count++]; + switch (packet_l4_hdrs) { + case IAVF_ADV_RSS_FLOW_SEG_HDR_TCP: + iavf_fill_adv_rss_tcp_hdr(hdr, hash_flds); + break; + case IAVF_ADV_RSS_FLOW_SEG_HDR_UDP: + iavf_fill_adv_rss_udp_hdr(hdr, hash_flds); + break; + case IAVF_ADV_RSS_FLOW_SEG_HDR_SCTP: + iavf_fill_adv_rss_sctp_hdr(hdr, hash_flds); + break; + default: + return -EINVAL; + } + } + + if (packet_hdrs & IAVF_ADV_RSS_FLOW_SEG_HDR_GTP) { + hdr = &proto_hdrs->proto_hdr[proto_hdrs->count++]; + if (iavf_fill_adv_rss_gtp_hdr(proto_hdrs, packet_hdrs, hash_flds)) + return -EINVAL; } return 0; @@ -186,6 +247,8 @@ iavf_print_adv_rss_cfg(struct iavf_adapter *adapter, struct iavf_adv_rss *rss, proto = "UDP"; else if (packet_hdrs & IAVF_ADV_RSS_FLOW_SEG_HDR_SCTP) proto = "SCTP"; + else if (packet_hdrs & IAVF_ADV_RSS_FLOW_SEG_HDR_GTP) + proto = "GTP"; else return; @@ -211,6 +274,16 @@ iavf_print_adv_rss_cfg(struct iavf_adapter *adapter, struct iavf_adv_rss *rss, IAVF_ADV_RSS_HASH_FLD_UDP_DST_PORT | IAVF_ADV_RSS_HASH_FLD_SCTP_DST_PORT)) strcat(hash_opt, "dst port,"); + if (hash_flds & IAVF_ADV_RSS_HASH_FLD_GTPC_TEID) + strcat(hash_opt, "gtp-c,"); + if (hash_flds & IAVF_ADV_RSS_HASH_FLD_GTPU_IP_TEID) + strcat(hash_opt, "gtp-u ip,"); + if (hash_flds & IAVF_ADV_RSS_HASH_FLD_GTPU_EH_TEID) + strcat(hash_opt, "gtp-u ext,"); + if (hash_flds & IAVF_ADV_RSS_HASH_FLD_GTPU_UP_TEID) + strcat(hash_opt, "gtp-u ul,"); + if (hash_flds & IAVF_ADV_RSS_HASH_FLD_GTPU_DWN_TEID) + strcat(hash_opt, "gtp-u dl,"); if (!action) action = ""; diff --git a/drivers/net/ethernet/intel/iavf/iavf_adv_rss.h b/drivers/net/ethernet/intel/iavf/iavf_adv_rss.h index e31eb2afebea..74cc9e0d528c 100644 --- a/drivers/net/ethernet/intel/iavf/iavf_adv_rss.h +++ b/drivers/net/ethernet/intel/iavf/iavf_adv_rss.h @@ -22,6 +22,12 @@ enum iavf_adv_rss_flow_seg_hdr { IAVF_ADV_RSS_FLOW_SEG_HDR_TCP = 0x00000004, IAVF_ADV_RSS_FLOW_SEG_HDR_UDP = 0x00000008, IAVF_ADV_RSS_FLOW_SEG_HDR_SCTP = 0x00000010, + IAVF_ADV_RSS_FLOW_SEG_HDR_GTPC = 0x00000400, + IAVF_ADV_RSS_FLOW_SEG_HDR_GTPC_TEID = 0x00000800, + IAVF_ADV_RSS_FLOW_SEG_HDR_GTPU_IP = 0x00001000, + IAVF_ADV_RSS_FLOW_SEG_HDR_GTPU_EH = 0x00002000, + IAVF_ADV_RSS_FLOW_SEG_HDR_GTPU_DWN = 0x00004000, + IAVF_ADV_RSS_FLOW_SEG_HDR_GTPU_UP = 0x00008000, }; #define IAVF_ADV_RSS_FLOW_SEG_HDR_L3 \ @@ -33,6 +39,14 @@ enum iavf_adv_rss_flow_seg_hdr { IAVF_ADV_RSS_FLOW_SEG_HDR_UDP | \ IAVF_ADV_RSS_FLOW_SEG_HDR_SCTP) +#define IAVF_ADV_RSS_FLOW_SEG_HDR_GTP \ + (IAVF_ADV_RSS_FLOW_SEG_HDR_GTPC | \ + IAVF_ADV_RSS_FLOW_SEG_HDR_GTPC_TEID | \ + IAVF_ADV_RSS_FLOW_SEG_HDR_GTPU_IP | \ + IAVF_ADV_RSS_FLOW_SEG_HDR_GTPU_EH | \ + IAVF_ADV_RSS_FLOW_SEG_HDR_GTPU_DWN | \ + IAVF_ADV_RSS_FLOW_SEG_HDR_GTPU_UP) + enum iavf_adv_rss_flow_field { /* L3 */ IAVF_ADV_RSS_FLOW_FIELD_IDX_IPV4_SA, @@ -46,6 +60,17 @@ enum iavf_adv_rss_flow_field { IAVF_ADV_RSS_FLOW_FIELD_IDX_UDP_DST_PORT, IAVF_ADV_RSS_FLOW_FIELD_IDX_SCTP_SRC_PORT, IAVF_ADV_RSS_FLOW_FIELD_IDX_SCTP_DST_PORT, + /* GTPC_TEID */ + IAVF_ADV_RSS_FLOW_FIELD_IDX_GTPC_TEID, + /* GTPU_IP */ + IAVF_ADV_RSS_FLOW_FIELD_IDX_GTPU_IP_TEID, + /* GTPU_EH */ + IAVF_ADV_RSS_FLOW_FIELD_IDX_GTPU_EH_TEID, + IAVF_ADV_RSS_FLOW_FIELD_IDX_GTPU_EH_QFI, + /* GTPU_UP */ + IAVF_ADV_RSS_FLOW_FIELD_IDX_GTPU_UP_TEID, + /* GTPU_DWN */ + IAVF_ADV_RSS_FLOW_FIELD_IDX_GTPU_DWN_TEID, /* The total number of enums must not exceed 64 */ IAVF_ADV_RSS_FLOW_FIELD_IDX_MAX @@ -72,6 +97,12 @@ enum iavf_adv_rss_flow_field { BIT_ULL(IAVF_ADV_RSS_FLOW_FIELD_IDX_SCTP_SRC_PORT) #define IAVF_ADV_RSS_HASH_FLD_SCTP_DST_PORT \ BIT_ULL(IAVF_ADV_RSS_FLOW_FIELD_IDX_SCTP_DST_PORT) +#define IAVF_ADV_RSS_HASH_FLD_GTPC_TEID BIT_ULL(IAVF_ADV_RSS_FLOW_FIELD_IDX_GTPC_TEID) +#define IAVF_ADV_RSS_HASH_FLD_GTPU_IP_TEID BIT_ULL(IAVF_ADV_RSS_FLOW_FIELD_IDX_GTPU_IP_TEID) +#define IAVF_ADV_RSS_HASH_FLD_GTPU_EH_TEID BIT_ULL(IAVF_ADV_RSS_FLOW_FIELD_IDX_GTPU_EH_TEID) +#define IAVF_ADV_RSS_HASH_FLD_GTPU_UP_TEID BIT_ULL(IAVF_ADV_RSS_FLOW_FIELD_IDX_GTPU_UP_TEID) +#define IAVF_ADV_RSS_HASH_FLD_GTPU_DWN_TEID \ + BIT_ULL(IAVF_ADV_RSS_FLOW_FIELD_IDX_GTPU_DWN_TEID) /* bookkeeping of advanced RSS configuration */ struct iavf_adv_rss { diff --git a/drivers/net/ethernet/intel/iavf/iavf_ethtool.c b/drivers/net/ethernet/intel/iavf/iavf_ethtool.c index 05d72be3fe80..a3f8ced23266 100644 --- a/drivers/net/ethernet/intel/iavf/iavf_ethtool.c +++ b/drivers/net/ethernet/intel/iavf/iavf_ethtool.c @@ -1336,6 +1336,56 @@ static u32 iavf_adv_rss_parse_hdrs(const struct ethtool_rxfh_fields *cmd) hdrs |= IAVF_ADV_RSS_FLOW_SEG_HDR_SCTP | IAVF_ADV_RSS_FLOW_SEG_HDR_IPV6; break; + case GTPU_V4_FLOW: + hdrs |= IAVF_ADV_RSS_FLOW_SEG_HDR_GTPU_IP | + IAVF_ADV_RSS_FLOW_SEG_HDR_IPV4; + break; + case GTPC_V4_FLOW: + hdrs |= IAVF_ADV_RSS_FLOW_SEG_HDR_GTPC | + IAVF_ADV_RSS_FLOW_SEG_HDR_UDP | + IAVF_ADV_RSS_FLOW_SEG_HDR_IPV4; + break; + case GTPC_TEID_V4_FLOW: + hdrs |= IAVF_ADV_RSS_FLOW_SEG_HDR_GTPC_TEID | + IAVF_ADV_RSS_FLOW_SEG_HDR_UDP | + IAVF_ADV_RSS_FLOW_SEG_HDR_IPV4; + break; + case GTPU_EH_V4_FLOW: + hdrs |= IAVF_ADV_RSS_FLOW_SEG_HDR_GTPU_EH | + IAVF_ADV_RSS_FLOW_SEG_HDR_IPV4; + break; + case GTPU_UL_V4_FLOW: + hdrs |= IAVF_ADV_RSS_FLOW_SEG_HDR_GTPU_UP | + IAVF_ADV_RSS_FLOW_SEG_HDR_IPV4; + break; + case GTPU_DL_V4_FLOW: + hdrs |= IAVF_ADV_RSS_FLOW_SEG_HDR_GTPU_DWN | + IAVF_ADV_RSS_FLOW_SEG_HDR_IPV4; + break; + case GTPU_V6_FLOW: + hdrs |= IAVF_ADV_RSS_FLOW_SEG_HDR_GTPU_IP | + IAVF_ADV_RSS_FLOW_SEG_HDR_IPV6; + break; + case GTPC_V6_FLOW: + hdrs |= IAVF_ADV_RSS_FLOW_SEG_HDR_GTPC | + IAVF_ADV_RSS_FLOW_SEG_HDR_IPV6; + break; + case GTPC_TEID_V6_FLOW: + hdrs |= IAVF_ADV_RSS_FLOW_SEG_HDR_GTPC_TEID | + IAVF_ADV_RSS_FLOW_SEG_HDR_IPV6; + break; + case GTPU_EH_V6_FLOW: + hdrs |= IAVF_ADV_RSS_FLOW_SEG_HDR_GTPU_EH | + IAVF_ADV_RSS_FLOW_SEG_HDR_IPV6; + break; + case GTPU_UL_V6_FLOW: + hdrs |= IAVF_ADV_RSS_FLOW_SEG_HDR_GTPU_UP | + IAVF_ADV_RSS_FLOW_SEG_HDR_IPV6; + break; + case GTPU_DL_V6_FLOW: + hdrs |= IAVF_ADV_RSS_FLOW_SEG_HDR_GTPU_DWN | + IAVF_ADV_RSS_FLOW_SEG_HDR_IPV6; + break; default: break; } @@ -1353,6 +1403,12 @@ iavf_adv_rss_parse_hash_flds(const struct ethtool_rxfh_fields *cmd, bool symm) case TCP_V4_FLOW: case UDP_V4_FLOW: case SCTP_V4_FLOW: + case GTPU_V4_FLOW: + case GTPC_V4_FLOW: + case GTPC_TEID_V4_FLOW: + case GTPU_EH_V4_FLOW: + case GTPU_UL_V4_FLOW: + case GTPU_DL_V4_FLOW: if (cmd->data & RXH_IP_SRC) hfld |= IAVF_ADV_RSS_HASH_FLD_IPV4_SA; if (cmd->data & RXH_IP_DST) @@ -1361,6 +1417,12 @@ iavf_adv_rss_parse_hash_flds(const struct ethtool_rxfh_fields *cmd, bool symm) case TCP_V6_FLOW: case UDP_V6_FLOW: case SCTP_V6_FLOW: + case GTPU_V6_FLOW: + case GTPC_V6_FLOW: + case GTPC_TEID_V6_FLOW: + case GTPU_EH_V6_FLOW: + case GTPU_UL_V6_FLOW: + case GTPU_DL_V6_FLOW: if (cmd->data & RXH_IP_SRC) hfld |= IAVF_ADV_RSS_HASH_FLD_IPV6_SA; if (cmd->data & RXH_IP_DST) @@ -1382,6 +1444,7 @@ iavf_adv_rss_parse_hash_flds(const struct ethtool_rxfh_fields *cmd, bool symm) break; case UDP_V4_FLOW: case UDP_V6_FLOW: + case GTPC_V4_FLOW: if (cmd->data & RXH_L4_B_0_1) hfld |= IAVF_ADV_RSS_HASH_FLD_UDP_SRC_PORT; if (cmd->data & RXH_L4_B_2_3) @@ -1398,6 +1461,32 @@ iavf_adv_rss_parse_hash_flds(const struct ethtool_rxfh_fields *cmd, bool symm) break; } } + if (cmd->data & RXH_GTP_TEID) { + switch (cmd->flow_type) { + case GTPC_TEID_V4_FLOW: + case GTPC_TEID_V6_FLOW: + hfld |= IAVF_ADV_RSS_HASH_FLD_GTPC_TEID; + break; + case GTPU_V4_FLOW: + case GTPU_V6_FLOW: + hfld |= IAVF_ADV_RSS_HASH_FLD_GTPU_IP_TEID; + break; + case GTPU_EH_V4_FLOW: + case GTPU_EH_V6_FLOW: + hfld |= IAVF_ADV_RSS_HASH_FLD_GTPU_EH_TEID; + break; + case GTPU_UL_V4_FLOW: + case GTPU_UL_V6_FLOW: + hfld |= IAVF_ADV_RSS_HASH_FLD_GTPU_UP_TEID; + break; + case GTPU_DL_V4_FLOW: + case GTPU_DL_V6_FLOW: + hfld |= IAVF_ADV_RSS_HASH_FLD_GTPU_DWN_TEID; + break; + default: + break; + } + } return hfld; } From c04956cccb78cc233a20cc18f663689671f03c65 Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Wed, 5 Nov 2025 10:01:12 -0800 Subject: [PATCH 0586/1381] tg3: extract GRXRINGS from .get_rxnfc Commit 84eaf4359c36 ("net: ethtool: add get_rx_ring_count callback to optimize RX ring queries") added specific support for GRXRINGS callback, simplifying .get_rxnfc. Remove the handling of GRXRINGS in .get_rxnfc() by moving it to the new .get_rx_ring_count(). Given that tg3_get_rxnfc() only handles ETHTOOL_GRXRINGS, then this function becomes useless now, and it is removed. This also fixes the behavior for devices without MSIX support. Previously, the function would return -EOPNOTSUPP, but now it correctly returns 1. The functionality remains the same: return the current queue count if the device is running, otherwise return the minimum of online CPUs and TG3_RSS_MAX_NUM_QS. Signed-off-by: Breno Leitao Reviewed-by: Michael Chan Link: https://patch.msgid.link/20251105-grxrings_v1-v1-1-54c2caafa1fd@debian.org Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/broadcom/tg3.c | 24 ++++++------------------ 1 file changed, 6 insertions(+), 18 deletions(-) diff --git a/drivers/net/ethernet/broadcom/tg3.c b/drivers/net/ethernet/broadcom/tg3.c index d78cafdb2094..fa58c3ffceb0 100644 --- a/drivers/net/ethernet/broadcom/tg3.c +++ b/drivers/net/ethernet/broadcom/tg3.c @@ -12719,29 +12719,17 @@ static int tg3_get_sset_count(struct net_device *dev, int sset) } } -static int tg3_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *info, - u32 *rules __always_unused) +static u32 tg3_get_rx_ring_count(struct net_device *dev) { struct tg3 *tp = netdev_priv(dev); if (!tg3_flag(tp, SUPPORT_MSIX)) - return -EOPNOTSUPP; + return 1; - switch (info->cmd) { - case ETHTOOL_GRXRINGS: - if (netif_running(tp->dev)) - info->data = tp->rxq_cnt; - else { - info->data = num_online_cpus(); - if (info->data > TG3_RSS_MAX_NUM_QS) - info->data = TG3_RSS_MAX_NUM_QS; - } + if (netif_running(tp->dev)) + return tp->rxq_cnt; - return 0; - - default: - return -EOPNOTSUPP; - } + return min(num_online_cpus(), TG3_RSS_MAX_NUM_QS); } static u32 tg3_get_rxfh_indir_size(struct net_device *dev) @@ -14268,7 +14256,7 @@ static const struct ethtool_ops tg3_ethtool_ops = { .get_coalesce = tg3_get_coalesce, .set_coalesce = tg3_set_coalesce, .get_sset_count = tg3_get_sset_count, - .get_rxnfc = tg3_get_rxnfc, + .get_rx_ring_count = tg3_get_rx_ring_count, .get_rxfh_indir_size = tg3_get_rxfh_indir_size, .get_rxfh = tg3_get_rxfh, .set_rxfh = tg3_set_rxfh, From 8a25a2e34157d882032112e4194ccdfb29c499e8 Mon Sep 17 00:00:00 2001 From: FUJITA Tomonori Date: Wed, 5 Nov 2025 22:31:26 +0900 Subject: [PATCH 0587/1381] net: phy: qt2025: Wait until PHY becomes ready Wait until a PHY becomes ready in the probe callback by using read_poll_timeout function. Reviewed-by: Andrew Lunn Reviewed-by: Alice Ryhl Reviewed-by: Gary Guo Signed-off-by: FUJITA Tomonori Link: https://patch.msgid.link/20251105133126.3221948-1-fujita.tomonori@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/phy/qt2025.rs | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/drivers/net/phy/qt2025.rs b/drivers/net/phy/qt2025.rs index 0b9400dcb4c1..aaaead6512a0 100644 --- a/drivers/net/phy/qt2025.rs +++ b/drivers/net/phy/qt2025.rs @@ -12,6 +12,7 @@ use kernel::c_str; use kernel::error::code; use kernel::firmware::Firmware; +use kernel::io::poll::read_poll_timeout; use kernel::net::phy::{ self, reg::{Mmd, C45}, @@ -19,6 +20,7 @@ }; use kernel::prelude::*; use kernel::sizes::{SZ_16K, SZ_8K}; +use kernel::time::Delta; kernel::module_phy_driver! { drivers: [PhyQT2025], @@ -93,7 +95,13 @@ fn probe(dev: &mut phy::Device) -> Result<()> { // The micro-controller will start running from SRAM. dev.write(C45::new(Mmd::PCS, 0xe854), 0x0040)?; - // TODO: sleep here until the hw becomes ready. + read_poll_timeout( + || dev.read(C45::new(Mmd::PCS, 0xd7fd)), + |val| *val != 0x00 && *val != 0x10, + Delta::from_millis(50), + Delta::from_secs(3), + )?; + Ok(()) } From 24ab8efb9aea77764dd99d2bad41fd8991223013 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Fri, 31 Oct 2025 22:20:55 +0100 Subject: [PATCH 0588/1381] xsk: Move NETDEV_XDP_ACT_ZC into generic header Move NETDEV_XDP_ACT_ZC into xdp_sock_drv.h header such that external code can reuse it, and rename it into more generic NETDEV_XDP_ACT_XSK. Signed-off-by: Daniel Borkmann Co-developed-by: David Wei Signed-off-by: David Wei Reviewed-by: Maciej Fijalkowski Reviewed-by: Nikolay Aleksandrov Link: https://patch.msgid.link/20251031212103.310683-7-daniel@iogearbox.net Signed-off-by: Jakub Kicinski --- include/net/xdp_sock_drv.h | 4 ++++ net/xdp/xsk_buff_pool.c | 6 +----- 2 files changed, 5 insertions(+), 5 deletions(-) diff --git a/include/net/xdp_sock_drv.h b/include/net/xdp_sock_drv.h index 4f2d3268a676..242e34f771cc 100644 --- a/include/net/xdp_sock_drv.h +++ b/include/net/xdp_sock_drv.h @@ -12,6 +12,10 @@ #define XDP_UMEM_MIN_CHUNK_SHIFT 11 #define XDP_UMEM_MIN_CHUNK_SIZE (1 << XDP_UMEM_MIN_CHUNK_SHIFT) +#define NETDEV_XDP_ACT_XSK (NETDEV_XDP_ACT_BASIC | \ + NETDEV_XDP_ACT_REDIRECT | \ + NETDEV_XDP_ACT_XSK_ZEROCOPY) + struct xsk_cb_desc { void *src; u8 off; diff --git a/net/xdp/xsk_buff_pool.c b/net/xdp/xsk_buff_pool.c index 00a4eddaa0cd..51526034c42a 100644 --- a/net/xdp/xsk_buff_pool.c +++ b/net/xdp/xsk_buff_pool.c @@ -155,10 +155,6 @@ static void xp_disable_drv_zc(struct xsk_buff_pool *pool) } } -#define NETDEV_XDP_ACT_ZC (NETDEV_XDP_ACT_BASIC | \ - NETDEV_XDP_ACT_REDIRECT | \ - NETDEV_XDP_ACT_XSK_ZEROCOPY) - int xp_assign_dev(struct xsk_buff_pool *pool, struct net_device *netdev, u16 queue_id, u16 flags) { @@ -200,7 +196,7 @@ int xp_assign_dev(struct xsk_buff_pool *pool, /* For copy-mode, we are done. */ return 0; - if ((netdev->xdp_features & NETDEV_XDP_ACT_ZC) != NETDEV_XDP_ACT_ZC) { + if ((netdev->xdp_features & NETDEV_XDP_ACT_XSK) != NETDEV_XDP_ACT_XSK) { err = -EOPNOTSUPP; goto err_unreg_pool; } From 25e63e559c41b5caa2ad8f076eba90f9897c0a07 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Fri, 31 Oct 2025 22:20:59 +0100 Subject: [PATCH 0589/1381] netkit: Document fast vs slowpath members via macros Instead of a comment, just use two cachline groups to document the intent for members often accessed in fast or slow path. Signed-off-by: Daniel Borkmann Co-developed-by: David Wei Signed-off-by: David Wei Reviewed-by: Nikolay Aleksandrov Link: https://patch.msgid.link/20251031212103.310683-11-daniel@iogearbox.net Signed-off-by: Jakub Kicinski --- drivers/net/netkit.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/net/netkit.c b/drivers/net/netkit.c index 492be60f2e70..0a2fef7caccb 100644 --- a/drivers/net/netkit.c +++ b/drivers/net/netkit.c @@ -16,17 +16,19 @@ #define DRV_NAME "netkit" struct netkit { - /* Needed in fast-path */ + __cacheline_group_begin(netkit_fastpath); struct net_device __rcu *peer; struct bpf_mprog_entry __rcu *active; enum netkit_action policy; enum netkit_scrub scrub; struct bpf_mprog_bundle bundle; + __cacheline_group_end(netkit_fastpath); - /* Needed in slow-path */ + __cacheline_group_begin(netkit_slowpath); enum netkit_mode mode; bool primary; u32 headroom; + __cacheline_group_end(netkit_slowpath); }; struct netkit_link { From e98d8792929df31ec51710773509f6fd2964ea81 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Niklas=20S=C3=B6derlund?= Date: Tue, 4 Nov 2025 23:24:14 +0100 Subject: [PATCH 0590/1381] net: rswitch: Move definition of S4 gPTP offset MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The files rcar_gen4_ptp.{c,h} implements an abstraction of the gPTP support implemented together with different other IP blocks. The first device added which supported this was RSWITCH on R-Car S4. While doing so the RSWITCH R-Car S4 specific offset was added to the generic Gen4 gPTP header file. Move it to the RSWITCH driver to make it clear it only applies to this driver. Signed-off-by: Niklas Söderlund Link: https://patch.msgid.link/20251104222420.882731-2-niklas.soderlund+renesas@ragnatech.se Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/renesas/rcar_gen4_ptp.h | 2 -- drivers/net/ethernet/renesas/rswitch_main.c | 4 +++- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/renesas/rcar_gen4_ptp.h b/drivers/net/ethernet/renesas/rcar_gen4_ptp.h index f77e79e47357..536badd798cc 100644 --- a/drivers/net/ethernet/renesas/rcar_gen4_ptp.h +++ b/drivers/net/ethernet/renesas/rcar_gen4_ptp.h @@ -9,8 +9,6 @@ #include -#define RCAR_GEN4_GPTP_OFFSET_S4 0x00018000 - /* driver's definitions */ #define RCAR_GEN4_RXTSTAMP_ENABLED BIT(0) #define RCAR_GEN4_RXTSTAMP_TYPE_V2_L2_EVENT BIT(1) diff --git a/drivers/net/ethernet/renesas/rswitch_main.c b/drivers/net/ethernet/renesas/rswitch_main.c index f21a814aa9d1..24ed33ac4bcd 100644 --- a/drivers/net/ethernet/renesas/rswitch_main.c +++ b/drivers/net/ethernet/renesas/rswitch_main.c @@ -30,6 +30,8 @@ #include "rswitch.h" #include "rswitch_l2.h" +#define RSWITCH_GPTP_OFFSET_S4 0x00018000 + static int rswitch_reg_wait(void __iomem *addr, u32 offs, u32 mask, u32 expected) { u32 val; @@ -2175,7 +2177,7 @@ static int renesas_eth_sw_probe(struct platform_device *pdev) if (IS_ERR(priv->addr)) return PTR_ERR(priv->addr); - priv->ptp_priv->addr = priv->addr + RCAR_GEN4_GPTP_OFFSET_S4; + priv->ptp_priv->addr = priv->addr + RSWITCH_GPTP_OFFSET_S4; ret = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(40)); if (ret < 0) { From 50ab1c6becdec42ce7cc9b52f2ea67c6a81218f2 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Niklas=20S=C3=B6derlund?= Date: Tue, 4 Nov 2025 23:24:15 +0100 Subject: [PATCH 0591/1381] net: rcar_gen4_ptp: Move control fields to users MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The struct rcar_gen4_ptp_private provides two fields for convenience of its users, tstamp_tx_ctrl and tstamp_rx_ctrl. These fields are not used by the rcar_gen4_ptp driver itself but only by the drivers using it. Upcoming work will enable the RAVB driver currently only supporting gPTP on pre-Gen4 SoCs to use the Gen4 implementation as well. To facilitate this the convenience of having these fields in struct rcar_gen4_ptp_private becomes a problem as the RAVB driver already have it's own driver specific fields for the same thing. Move the fields from struct rcar_gen4_ptp_private to each driver using the Gen4 gPTP clocks own private data structures. There is no functional change. Signed-off-by: Niklas Söderlund Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20251104222420.882731-3-niklas.soderlund+renesas@ragnatech.se Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/renesas/rcar_gen4_ptp.h | 2 -- drivers/net/ethernet/renesas/rswitch.h | 3 +++ drivers/net/ethernet/renesas/rswitch_main.c | 17 ++++++++--------- drivers/net/ethernet/renesas/rtsn.c | 17 ++++++++--------- 4 files changed, 19 insertions(+), 20 deletions(-) diff --git a/drivers/net/ethernet/renesas/rcar_gen4_ptp.h b/drivers/net/ethernet/renesas/rcar_gen4_ptp.h index 536badd798cc..1a1e43add129 100644 --- a/drivers/net/ethernet/renesas/rcar_gen4_ptp.h +++ b/drivers/net/ethernet/renesas/rcar_gen4_ptp.h @@ -23,8 +23,6 @@ struct rcar_gen4_ptp_private { struct ptp_clock *clock; struct ptp_clock_info info; spinlock_t lock; /* For multiple registers access */ - u32 tstamp_tx_ctrl; - u32 tstamp_rx_ctrl; s64 default_addend; bool initialized; }; diff --git a/drivers/net/ethernet/renesas/rswitch.h b/drivers/net/ethernet/renesas/rswitch.h index a1d4a877e5bd..3b348ebf6742 100644 --- a/drivers/net/ethernet/renesas/rswitch.h +++ b/drivers/net/ethernet/renesas/rswitch.h @@ -1063,6 +1063,9 @@ struct rswitch_private { bool etha_no_runtime_change; bool gwca_halt; struct net_device *offload_brdev; + + u32 tstamp_tx_ctrl; + u32 tstamp_rx_ctrl; }; bool is_rdev(const struct net_device *ndev); diff --git a/drivers/net/ethernet/renesas/rswitch_main.c b/drivers/net/ethernet/renesas/rswitch_main.c index 24ed33ac4bcd..31aabc6fc462 100644 --- a/drivers/net/ethernet/renesas/rswitch_main.c +++ b/drivers/net/ethernet/renesas/rswitch_main.c @@ -845,7 +845,7 @@ static bool rswitch_rx(struct net_device *ndev, int *quota) if (!skb) goto out; - get_ts = rdev->priv->ptp_priv->tstamp_rx_ctrl & RCAR_GEN4_RXTSTAMP_TYPE_V2_L2_EVENT; + get_ts = rdev->priv->tstamp_rx_ctrl & RCAR_GEN4_RXTSTAMP_TYPE_V2_L2_EVENT; if (get_ts) { struct skb_shared_hwtstamps *shhwtstamps; struct timespec64 ts; @@ -1799,14 +1799,13 @@ static int rswitch_hwstamp_get(struct net_device *ndev, struct kernel_hwtstamp_config *config) { struct rswitch_device *rdev = netdev_priv(ndev); - struct rcar_gen4_ptp_private *ptp_priv; - - ptp_priv = rdev->priv->ptp_priv; + struct rswitch_private *priv = rdev->priv; config->flags = 0; - config->tx_type = ptp_priv->tstamp_tx_ctrl ? HWTSTAMP_TX_ON : - HWTSTAMP_TX_OFF; - switch (ptp_priv->tstamp_rx_ctrl & RCAR_GEN4_RXTSTAMP_TYPE) { + config->tx_type = + priv->tstamp_tx_ctrl ? HWTSTAMP_TX_ON : HWTSTAMP_TX_OFF; + + switch (priv->tstamp_rx_ctrl & RCAR_GEN4_RXTSTAMP_TYPE) { case RCAR_GEN4_RXTSTAMP_TYPE_V2_L2_EVENT: config->rx_filter = HWTSTAMP_FILTER_PTP_V2_L2_EVENT; break; @@ -1856,8 +1855,8 @@ static int rswitch_hwstamp_set(struct net_device *ndev, break; } - rdev->priv->ptp_priv->tstamp_tx_ctrl = tstamp_tx_ctrl; - rdev->priv->ptp_priv->tstamp_rx_ctrl = tstamp_rx_ctrl; + rdev->priv->tstamp_tx_ctrl = tstamp_tx_ctrl; + rdev->priv->tstamp_rx_ctrl = tstamp_rx_ctrl; return 0; } diff --git a/drivers/net/ethernet/renesas/rtsn.c b/drivers/net/ethernet/renesas/rtsn.c index 15a043e85431..958c19808472 100644 --- a/drivers/net/ethernet/renesas/rtsn.c +++ b/drivers/net/ethernet/renesas/rtsn.c @@ -62,6 +62,9 @@ struct rtsn_private { int tx_data_irq; int rx_data_irq; + + u32 tstamp_tx_ctrl; + u32 tstamp_rx_ctrl; }; static u32 rtsn_read(struct rtsn_private *priv, enum rtsn_reg reg) @@ -162,7 +165,7 @@ static int rtsn_rx(struct net_device *ndev, int budget) unsigned int i; bool get_ts; - get_ts = priv->ptp_priv->tstamp_rx_ctrl & + get_ts = priv->tstamp_rx_ctrl & RCAR_GEN4_RXTSTAMP_TYPE_V2_L2_EVENT; ndescriptors = priv->dirty_rx + priv->num_rx_ring - priv->cur_rx; @@ -1122,21 +1125,19 @@ static int rtsn_do_ioctl(struct net_device *ndev, struct ifreq *ifr, int cmd) static int rtsn_hwtstamp_get(struct net_device *ndev, struct kernel_hwtstamp_config *config) { - struct rcar_gen4_ptp_private *ptp_priv; struct rtsn_private *priv; if (!netif_running(ndev)) return -ENODEV; priv = netdev_priv(ndev); - ptp_priv = priv->ptp_priv; config->flags = 0; config->tx_type = - ptp_priv->tstamp_tx_ctrl ? HWTSTAMP_TX_ON : HWTSTAMP_TX_OFF; + priv->tstamp_tx_ctrl ? HWTSTAMP_TX_ON : HWTSTAMP_TX_OFF; - switch (ptp_priv->tstamp_rx_ctrl & RCAR_GEN4_RXTSTAMP_TYPE) { + switch (priv->tstamp_rx_ctrl & RCAR_GEN4_RXTSTAMP_TYPE) { case RCAR_GEN4_RXTSTAMP_TYPE_V2_L2_EVENT: config->rx_filter = HWTSTAMP_FILTER_PTP_V2_L2_EVENT; break; @@ -1155,7 +1156,6 @@ static int rtsn_hwtstamp_set(struct net_device *ndev, struct kernel_hwtstamp_config *config, struct netlink_ext_ack *extack) { - struct rcar_gen4_ptp_private *ptp_priv; struct rtsn_private *priv; u32 tstamp_rx_ctrl; u32 tstamp_tx_ctrl; @@ -1164,7 +1164,6 @@ static int rtsn_hwtstamp_set(struct net_device *ndev, return -ENODEV; priv = netdev_priv(ndev); - ptp_priv = priv->ptp_priv; if (config->flags) return -EINVAL; @@ -1195,8 +1194,8 @@ static int rtsn_hwtstamp_set(struct net_device *ndev, break; } - ptp_priv->tstamp_tx_ctrl = tstamp_tx_ctrl; - ptp_priv->tstamp_rx_ctrl = tstamp_rx_ctrl; + priv->tstamp_tx_ctrl = tstamp_tx_ctrl; + priv->tstamp_rx_ctrl = tstamp_rx_ctrl; return 0; } From b314e4f7a9d9ba7821cfac57e1bf3648bb27a289 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Niklas=20S=C3=B6derlund?= Date: Tue, 4 Nov 2025 23:24:16 +0100 Subject: [PATCH 0592/1381] net: rswitch: Use common defines for time stamping control MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Instead of translating to/from driver specific flags for packet time stamp control use the common flags directly. This simplifies the driver as the translating code can be removed while at the same time making it clear the flags are not flags written to hardware registers. One thing to note is that the bit-wise and check in rswitch_rx() of RCAR_GEN4_RXTSTAMP_TYPE_V2_L2_EVENT is replaced with a not set check of HWTSTAMP_FILTER_NONE. This is okay as the bit of device specific event replaced was set for all modes except HWTSTAMP_FILTER_NONE. Signed-off-by: Niklas Söderlund Link: https://patch.msgid.link/20251104222420.882731-4-niklas.soderlund+renesas@ragnatech.se Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/renesas/rswitch.h | 4 +-- drivers/net/ethernet/renesas/rswitch_main.c | 32 ++++++--------------- 2 files changed, 10 insertions(+), 26 deletions(-) diff --git a/drivers/net/ethernet/renesas/rswitch.h b/drivers/net/ethernet/renesas/rswitch.h index 3b348ebf6742..aa605304fed0 100644 --- a/drivers/net/ethernet/renesas/rswitch.h +++ b/drivers/net/ethernet/renesas/rswitch.h @@ -1064,8 +1064,8 @@ struct rswitch_private { bool gwca_halt; struct net_device *offload_brdev; - u32 tstamp_tx_ctrl; - u32 tstamp_rx_ctrl; + enum hwtstamp_tx_types tstamp_tx_ctrl; + enum hwtstamp_rx_filters tstamp_rx_ctrl; }; bool is_rdev(const struct net_device *ndev); diff --git a/drivers/net/ethernet/renesas/rswitch_main.c b/drivers/net/ethernet/renesas/rswitch_main.c index 31aabc6fc462..e14b21148f27 100644 --- a/drivers/net/ethernet/renesas/rswitch_main.c +++ b/drivers/net/ethernet/renesas/rswitch_main.c @@ -845,7 +845,7 @@ static bool rswitch_rx(struct net_device *ndev, int *quota) if (!skb) goto out; - get_ts = rdev->priv->tstamp_rx_ctrl & RCAR_GEN4_RXTSTAMP_TYPE_V2_L2_EVENT; + get_ts = rdev->priv->tstamp_rx_ctrl != HWTSTAMP_FILTER_NONE; if (get_ts) { struct skb_shared_hwtstamps *shhwtstamps; struct timespec64 ts; @@ -1802,20 +1802,8 @@ static int rswitch_hwstamp_get(struct net_device *ndev, struct rswitch_private *priv = rdev->priv; config->flags = 0; - config->tx_type = - priv->tstamp_tx_ctrl ? HWTSTAMP_TX_ON : HWTSTAMP_TX_OFF; - - switch (priv->tstamp_rx_ctrl & RCAR_GEN4_RXTSTAMP_TYPE) { - case RCAR_GEN4_RXTSTAMP_TYPE_V2_L2_EVENT: - config->rx_filter = HWTSTAMP_FILTER_PTP_V2_L2_EVENT; - break; - case RCAR_GEN4_RXTSTAMP_TYPE_ALL: - config->rx_filter = HWTSTAMP_FILTER_ALL; - break; - default: - config->rx_filter = HWTSTAMP_FILTER_NONE; - break; - } + config->tx_type = priv->tstamp_tx_ctrl; + config->rx_filter = priv->tstamp_rx_ctrl; return 0; } @@ -1825,18 +1813,16 @@ static int rswitch_hwstamp_set(struct net_device *ndev, struct netlink_ext_ack *extack) { struct rswitch_device *rdev = netdev_priv(ndev); - u32 tstamp_rx_ctrl = RCAR_GEN4_RXTSTAMP_ENABLED; - u32 tstamp_tx_ctrl; + enum hwtstamp_rx_filters tstamp_rx_ctrl; + enum hwtstamp_tx_types tstamp_tx_ctrl; if (config->flags) return -EINVAL; switch (config->tx_type) { case HWTSTAMP_TX_OFF: - tstamp_tx_ctrl = 0; - break; case HWTSTAMP_TX_ON: - tstamp_tx_ctrl = RCAR_GEN4_TXTSTAMP_ENABLED; + tstamp_tx_ctrl = config->tx_type; break; default: return -ERANGE; @@ -1844,14 +1830,12 @@ static int rswitch_hwstamp_set(struct net_device *ndev, switch (config->rx_filter) { case HWTSTAMP_FILTER_NONE: - tstamp_rx_ctrl = 0; - break; case HWTSTAMP_FILTER_PTP_V2_L2_EVENT: - tstamp_rx_ctrl |= RCAR_GEN4_RXTSTAMP_TYPE_V2_L2_EVENT; + tstamp_rx_ctrl = config->rx_filter; break; default: config->rx_filter = HWTSTAMP_FILTER_ALL; - tstamp_rx_ctrl |= RCAR_GEN4_RXTSTAMP_TYPE_ALL; + tstamp_rx_ctrl = HWTSTAMP_FILTER_ALL; break; } From e43791f40b814d00b8e2be41f9d2bc2a97246d6d Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Niklas=20S=C3=B6derlund?= Date: Tue, 4 Nov 2025 23:24:17 +0100 Subject: [PATCH 0593/1381] net: rtsn: Use common defines for time stamping control MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Instead of translating to/from driver specific flags for packet time stamp control use the common flags directly. This simplifies the driver as the translating code can be removed while at the same time making it clear the flags are not flags written to hardware registers. One thing to note is that the bit-wise and check in rtsn_rx() of RCAR_GEN4_RXTSTAMP_TYPE_V2_L2_EVENT is replaced with a not set check of HWTSTAMP_FILTER_NONE. This is okay as the bit of device specific event replaced was set for all modes except HWTSTAMP_FILTER_NONE. Signed-off-by: Niklas Söderlund Link: https://patch.msgid.link/20251104222420.882731-5-niklas.soderlund+renesas@ragnatech.se Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/renesas/rtsn.c | 36 +++++++---------------------- 1 file changed, 8 insertions(+), 28 deletions(-) diff --git a/drivers/net/ethernet/renesas/rtsn.c b/drivers/net/ethernet/renesas/rtsn.c index 958c19808472..fdb1e7b7fb06 100644 --- a/drivers/net/ethernet/renesas/rtsn.c +++ b/drivers/net/ethernet/renesas/rtsn.c @@ -165,8 +165,7 @@ static int rtsn_rx(struct net_device *ndev, int budget) unsigned int i; bool get_ts; - get_ts = priv->tstamp_rx_ctrl & - RCAR_GEN4_RXTSTAMP_TYPE_V2_L2_EVENT; + get_ts = priv->tstamp_rx_ctrl != HWTSTAMP_FILTER_NONE; ndescriptors = priv->dirty_rx + priv->num_rx_ring - priv->cur_rx; rx_packets = 0; @@ -1133,21 +1132,8 @@ static int rtsn_hwtstamp_get(struct net_device *ndev, priv = netdev_priv(ndev); config->flags = 0; - - config->tx_type = - priv->tstamp_tx_ctrl ? HWTSTAMP_TX_ON : HWTSTAMP_TX_OFF; - - switch (priv->tstamp_rx_ctrl & RCAR_GEN4_RXTSTAMP_TYPE) { - case RCAR_GEN4_RXTSTAMP_TYPE_V2_L2_EVENT: - config->rx_filter = HWTSTAMP_FILTER_PTP_V2_L2_EVENT; - break; - case RCAR_GEN4_RXTSTAMP_TYPE_ALL: - config->rx_filter = HWTSTAMP_FILTER_ALL; - break; - default: - config->rx_filter = HWTSTAMP_FILTER_NONE; - break; - } + config->tx_type = priv->tstamp_tx_ctrl; + config->rx_filter = priv->tstamp_rx_ctrl; return 0; } @@ -1156,9 +1142,9 @@ static int rtsn_hwtstamp_set(struct net_device *ndev, struct kernel_hwtstamp_config *config, struct netlink_ext_ack *extack) { + enum hwtstamp_rx_filters tstamp_rx_ctrl; + enum hwtstamp_tx_types tstamp_tx_ctrl; struct rtsn_private *priv; - u32 tstamp_rx_ctrl; - u32 tstamp_tx_ctrl; if (!netif_running(ndev)) return -ENODEV; @@ -1170,10 +1156,8 @@ static int rtsn_hwtstamp_set(struct net_device *ndev, switch (config->tx_type) { case HWTSTAMP_TX_OFF: - tstamp_tx_ctrl = 0; - break; case HWTSTAMP_TX_ON: - tstamp_tx_ctrl = RCAR_GEN4_TXTSTAMP_ENABLED; + tstamp_tx_ctrl = config->tx_type; break; default: return -ERANGE; @@ -1181,16 +1165,12 @@ static int rtsn_hwtstamp_set(struct net_device *ndev, switch (config->rx_filter) { case HWTSTAMP_FILTER_NONE: - tstamp_rx_ctrl = 0; - break; case HWTSTAMP_FILTER_PTP_V2_L2_EVENT: - tstamp_rx_ctrl = RCAR_GEN4_RXTSTAMP_ENABLED | - RCAR_GEN4_RXTSTAMP_TYPE_V2_L2_EVENT; + tstamp_rx_ctrl = config->rx_filter; break; default: config->rx_filter = HWTSTAMP_FILTER_ALL; - tstamp_rx_ctrl = RCAR_GEN4_RXTSTAMP_ENABLED | - RCAR_GEN4_RXTSTAMP_TYPE_ALL; + tstamp_rx_ctrl = HWTSTAMP_FILTER_ALL; break; } From 3614d249d1da83b4806e015740117ed1069e3ca1 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Niklas=20S=C3=B6derlund?= Date: Tue, 4 Nov 2025 23:24:18 +0100 Subject: [PATCH 0594/1381] net: rcar_gen4_ptp: Remove unused defines MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The driver specific flags to control packet time stamps have all been replaced by values from enum hwtstamp_tx_types and enum hwtstamp_rx_filters. Remove the driver specific flags as there are no more users. Signed-off-by: Niklas Söderlund Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20251104222420.882731-6-niklas.soderlund+renesas@ragnatech.se Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/renesas/rcar_gen4_ptp.h | 9 --------- 1 file changed, 9 deletions(-) diff --git a/drivers/net/ethernet/renesas/rcar_gen4_ptp.h b/drivers/net/ethernet/renesas/rcar_gen4_ptp.h index 1a1e43add129..9a9c232c854e 100644 --- a/drivers/net/ethernet/renesas/rcar_gen4_ptp.h +++ b/drivers/net/ethernet/renesas/rcar_gen4_ptp.h @@ -9,15 +9,6 @@ #include -/* driver's definitions */ -#define RCAR_GEN4_RXTSTAMP_ENABLED BIT(0) -#define RCAR_GEN4_RXTSTAMP_TYPE_V2_L2_EVENT BIT(1) -#define RCAR_GEN4_RXTSTAMP_TYPE_ALL (RCAR_GEN4_RXTSTAMP_TYPE_V2_L2_EVENT | BIT(2)) -#define RCAR_GEN4_RXTSTAMP_TYPE RCAR_GEN4_RXTSTAMP_TYPE_ALL - -#define RCAR_GEN4_TXTSTAMP_ENABLED BIT(0) - - struct rcar_gen4_ptp_private { void __iomem *addr; struct ptp_clock *clock; From 5ce97b8d6132a0f252c68d58fed35f73c31e0bee Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Niklas=20S=C3=B6derlund?= Date: Tue, 4 Nov 2025 23:24:19 +0100 Subject: [PATCH 0595/1381] net: ravb: Break out Rx hardware timestamping MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Prepare for moving away from device specific bit-fields to track how to do hardware Rx timestamping to using net common enums by breaking out the timestamping to a helper function. This is done to create cleaner code and prepare for easier changes improving the hardware timestapming. There is no functional change. Signed-off-by: Niklas Söderlund Link: https://patch.msgid.link/20251104222420.882731-7-niklas.soderlund+renesas@ragnatech.se Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/renesas/ravb_main.c | 38 +++++++++++++++--------- 1 file changed, 24 insertions(+), 14 deletions(-) diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c index cc619dbebf9d..5477bb5c69ae 100644 --- a/drivers/net/ethernet/renesas/ravb_main.c +++ b/drivers/net/ethernet/renesas/ravb_main.c @@ -946,6 +946,29 @@ static int ravb_rx_gbeth(struct net_device *ndev, int budget, int q) return rx_packets; } +static void ravb_rx_rcar_hwstamp(struct ravb_private *priv, int q, + struct ravb_ex_rx_desc *desc, + struct sk_buff *skb) +{ + u32 get_ts = priv->tstamp_rx_ctrl & RAVB_RXTSTAMP_TYPE; + struct skb_shared_hwtstamps *shhwtstamps; + struct timespec64 ts; + + get_ts &= (q == RAVB_NC) ? + RAVB_RXTSTAMP_TYPE_V2_L2_EVENT : + ~RAVB_RXTSTAMP_TYPE_V2_L2_EVENT; + + if (!get_ts) + return; + + shhwtstamps = skb_hwtstamps(skb); + memset(shhwtstamps, 0, sizeof(*shhwtstamps)); + ts.tv_sec = ((u64)le16_to_cpu(desc->ts_sh) << 32) + | le32_to_cpu(desc->ts_sl); + ts.tv_nsec = le32_to_cpu(desc->ts_n); + shhwtstamps->hwtstamp = timespec64_to_ktime(ts); +} + /* Packet receive function for Ethernet AVB */ static int ravb_rx_rcar(struct net_device *ndev, int budget, int q) { @@ -955,7 +978,6 @@ static int ravb_rx_rcar(struct net_device *ndev, int budget, int q) struct ravb_ex_rx_desc *desc; unsigned int limit, i; struct sk_buff *skb; - struct timespec64 ts; int rx_packets = 0; u8 desc_status; u16 pkt_len; @@ -992,7 +1014,6 @@ static int ravb_rx_rcar(struct net_device *ndev, int budget, int q) if (desc_status & MSC_CEEF) stats->rx_missed_errors++; } else { - u32 get_ts = priv->tstamp_rx_ctrl & RAVB_RXTSTAMP_TYPE; struct ravb_rx_buffer *rx_buff; void *rx_addr; @@ -1010,19 +1031,8 @@ static int ravb_rx_rcar(struct net_device *ndev, int budget, int q) break; } skb_mark_for_recycle(skb); - get_ts &= (q == RAVB_NC) ? - RAVB_RXTSTAMP_TYPE_V2_L2_EVENT : - ~RAVB_RXTSTAMP_TYPE_V2_L2_EVENT; - if (get_ts) { - struct skb_shared_hwtstamps *shhwtstamps; - shhwtstamps = skb_hwtstamps(skb); - memset(shhwtstamps, 0, sizeof(*shhwtstamps)); - ts.tv_sec = ((u64) le16_to_cpu(desc->ts_sh) << - 32) | le32_to_cpu(desc->ts_sl); - ts.tv_nsec = le32_to_cpu(desc->ts_n); - shhwtstamps->hwtstamp = timespec64_to_ktime(ts); - } + ravb_rx_rcar_hwstamp(priv, q, desc, skb); skb_put(skb, pkt_len); skb->protocol = eth_type_trans(skb, ndev); From 16e2e6cf75e62182b9e28d0e48ab72d93c464bda Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Niklas=20S=C3=B6derlund?= Date: Tue, 4 Nov 2025 23:24:20 +0100 Subject: [PATCH 0596/1381] net: ravb: Use common defines for time stamping control MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Instead of translating to/from driver specific flags for packet time stamp control use the common flags directly. This simplifies the driver as the translating code can be removed while at the same time making it clear the flags are not flags written to hardware registers. The change from a device specific bit-field track variable to the common enum datatypes forces us to touch the ravb_rx_rcar_hwstamp() in a non trivial way. To make this cleaner and easier to understand expand the nested conditions. Signed-off-by: Niklas Söderlund Link: https://patch.msgid.link/20251104222420.882731-8-niklas.soderlund+renesas@ragnatech.se Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/renesas/ravb.h | 14 ++------- drivers/net/ethernet/renesas/ravb_main.c | 37 ++++++++---------------- 2 files changed, 14 insertions(+), 37 deletions(-) diff --git a/drivers/net/ethernet/renesas/ravb.h b/drivers/net/ethernet/renesas/ravb.h index d65cd83ddd16..5e56ec9b1013 100644 --- a/drivers/net/ethernet/renesas/ravb.h +++ b/drivers/net/ethernet/renesas/ravb.h @@ -35,16 +35,6 @@ /* Driver's parameters */ #define RAVB_ALIGN 128 -/* Hardware time stamp */ -#define RAVB_TXTSTAMP_VALID 0x00000001 /* TX timestamp valid */ -#define RAVB_TXTSTAMP_ENABLED 0x00000010 /* Enable TX timestamping */ - -#define RAVB_RXTSTAMP_VALID 0x00000001 /* RX timestamp valid */ -#define RAVB_RXTSTAMP_TYPE 0x00000006 /* RX type mask */ -#define RAVB_RXTSTAMP_TYPE_V2_L2_EVENT 0x00000002 -#define RAVB_RXTSTAMP_TYPE_ALL 0x00000006 -#define RAVB_RXTSTAMP_ENABLED 0x00000010 /* Enable RX timestamping */ - enum ravb_reg { /* AVB-DMAC registers */ CCC = 0x0000, @@ -1114,8 +1104,8 @@ struct ravb_private { u32 rx_over_errors; u32 rx_fifo_errors; struct net_device_stats stats[NUM_RX_QUEUE]; - u32 tstamp_tx_ctrl; - u32 tstamp_rx_ctrl; + enum hwtstamp_tx_types tstamp_tx_ctrl; + enum hwtstamp_rx_filters tstamp_rx_ctrl; struct list_head ts_skb_list; u32 ts_skb_tag; struct ravb_ptp ptp; diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c index 5477bb5c69ae..1680e94b9242 100644 --- a/drivers/net/ethernet/renesas/ravb_main.c +++ b/drivers/net/ethernet/renesas/ravb_main.c @@ -950,13 +950,14 @@ static void ravb_rx_rcar_hwstamp(struct ravb_private *priv, int q, struct ravb_ex_rx_desc *desc, struct sk_buff *skb) { - u32 get_ts = priv->tstamp_rx_ctrl & RAVB_RXTSTAMP_TYPE; struct skb_shared_hwtstamps *shhwtstamps; struct timespec64 ts; + bool get_ts; - get_ts &= (q == RAVB_NC) ? - RAVB_RXTSTAMP_TYPE_V2_L2_EVENT : - ~RAVB_RXTSTAMP_TYPE_V2_L2_EVENT; + if (q == RAVB_NC) + get_ts = priv->tstamp_rx_ctrl == HWTSTAMP_FILTER_PTP_V2_L2_EVENT; + else + get_ts = priv->tstamp_rx_ctrl != HWTSTAMP_FILTER_PTP_V2_L2_EVENT; if (!get_ts) return; @@ -2424,18 +2425,8 @@ static int ravb_hwtstamp_get(struct net_device *ndev, struct ravb_private *priv = netdev_priv(ndev); config->flags = 0; - config->tx_type = priv->tstamp_tx_ctrl ? HWTSTAMP_TX_ON : - HWTSTAMP_TX_OFF; - switch (priv->tstamp_rx_ctrl & RAVB_RXTSTAMP_TYPE) { - case RAVB_RXTSTAMP_TYPE_V2_L2_EVENT: - config->rx_filter = HWTSTAMP_FILTER_PTP_V2_L2_EVENT; - break; - case RAVB_RXTSTAMP_TYPE_ALL: - config->rx_filter = HWTSTAMP_FILTER_ALL; - break; - default: - config->rx_filter = HWTSTAMP_FILTER_NONE; - } + config->tx_type = priv->tstamp_tx_ctrl; + config->rx_filter = priv->tstamp_rx_ctrl; return 0; } @@ -2446,15 +2437,13 @@ static int ravb_hwtstamp_set(struct net_device *ndev, struct netlink_ext_ack *extack) { struct ravb_private *priv = netdev_priv(ndev); - u32 tstamp_rx_ctrl = RAVB_RXTSTAMP_ENABLED; - u32 tstamp_tx_ctrl; + enum hwtstamp_rx_filters tstamp_rx_ctrl; + enum hwtstamp_tx_types tstamp_tx_ctrl; switch (config->tx_type) { case HWTSTAMP_TX_OFF: - tstamp_tx_ctrl = 0; - break; case HWTSTAMP_TX_ON: - tstamp_tx_ctrl = RAVB_TXTSTAMP_ENABLED; + tstamp_tx_ctrl = config->tx_type; break; default: return -ERANGE; @@ -2462,14 +2451,12 @@ static int ravb_hwtstamp_set(struct net_device *ndev, switch (config->rx_filter) { case HWTSTAMP_FILTER_NONE: - tstamp_rx_ctrl = 0; - break; case HWTSTAMP_FILTER_PTP_V2_L2_EVENT: - tstamp_rx_ctrl |= RAVB_RXTSTAMP_TYPE_V2_L2_EVENT; + tstamp_rx_ctrl = config->rx_filter; break; default: config->rx_filter = HWTSTAMP_FILTER_ALL; - tstamp_rx_ctrl |= RAVB_RXTSTAMP_TYPE_ALL; + tstamp_rx_ctrl = HWTSTAMP_FILTER_ALL; } priv->tstamp_tx_ctrl = tstamp_tx_ctrl; From c6934c4e049c8a7f2b7fab620c04bc5dfbd947c1 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 4 Nov 2025 15:23:44 -0800 Subject: [PATCH 0597/1381] netlink: specs: netdev add missing stats to qstat-get Add missing entries in C attribute list. Link: https://patch.msgid.link/20251104232348.1954349-2-kuba@kernel.org Signed-off-by: Jakub Kicinski --- Documentation/netlink/specs/netdev.yaml | 23 +++++++++++++++++++++++ 1 file changed, 23 insertions(+) diff --git a/Documentation/netlink/specs/netdev.yaml b/Documentation/netlink/specs/netdev.yaml index 10c412b7433f..82bf5cb2617d 100644 --- a/Documentation/netlink/specs/netdev.yaml +++ b/Documentation/netlink/specs/netdev.yaml @@ -733,6 +733,29 @@ operations: - rx-bytes - tx-packets - tx-bytes + - rx-alloc-fail + - rx-hw-drops + - rx-hw-drop-overruns + - rx-csum-complete + - rx-csum-unnecessary + - rx-csum-none + - rx-csum-bad + - rx-hw-gro-packets + - rx-hw-gro-bytes + - rx-hw-gro-wire-packets + - rx-hw-gro-wire-bytes + - rx-hw-drop-ratelimits + - tx-hw-drops + - tx-hw-drop-errors + - tx-csum-none + - tx-needs-csum + - tx-hw-gso-packets + - tx-hw-gso-bytes + - tx-hw-gso-wire-packets + - tx-hw-gso-wire-bytes + - tx-hw-drop-ratelimits + - tx-stop + - tx-wake - name: bind-rx doc: Bind dmabuf to netdev From be88c549e9d78828a2e06126ed7e17fc2e030f1f Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Thu, 6 Nov 2025 00:32:40 +0000 Subject: [PATCH 0598/1381] tcp: Call tcp_syn_ack_timeout() directly. Since DCCP has been removed, we do not need to use request_sock_ops.syn_ack_timeout(). Let's call tcp_syn_ack_timeout() directly. Now other function pointers of request_sock_ops are protocol-dependent. Signed-off-by: Kuniyuki Iwashima Reviewed-by: Eric Dumazet Link: https://patch.msgid.link/20251106003357.273403-2-kuniyu@google.com Signed-off-by: Jakub Kicinski --- include/net/request_sock.h | 1 - net/ipv4/inet_connection_sock.c | 4 +++- net/ipv4/tcp_ipv4.c | 1 - net/ipv4/tcp_timer.c | 3 +-- net/ipv6/tcp_ipv6.c | 1 - 5 files changed, 4 insertions(+), 6 deletions(-) diff --git a/include/net/request_sock.h b/include/net/request_sock.h index cd4d4cf71d0d..9b9e04f6bb89 100644 --- a/include/net/request_sock.h +++ b/include/net/request_sock.h @@ -36,7 +36,6 @@ struct request_sock_ops { struct sk_buff *skb, enum sk_rst_reason reason); void (*destructor)(struct request_sock *req); - void (*syn_ack_timeout)(const struct request_sock *req); }; struct saved_syn { diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index 3b83b66b2284..6a86c1ac3011 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -1096,9 +1096,11 @@ static void reqsk_timer_handler(struct timer_list *t) young <<= 1; } } + syn_ack_recalc(req, max_syn_ack_retries, READ_ONCE(queue->rskq_defer_accept), &expire, &resend); - req->rsk_ops->syn_ack_timeout(req); + tcp_syn_ack_timeout(req); + if (!expire && (!resend || !tcp_rtx_synack(sk_listener, req) || diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index b7526a7888cb..0cfebac33a91 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -1660,7 +1660,6 @@ struct request_sock_ops tcp_request_sock_ops __read_mostly = { .send_ack = tcp_v4_reqsk_send_ack, .destructor = tcp_v4_reqsk_destructor, .send_reset = tcp_v4_send_reset, - .syn_ack_timeout = tcp_syn_ack_timeout, }; const struct tcp_request_sock_ops tcp_request_sock_ipv4_ops = { diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index 2dd73a4e8e51..0672c3d8f4f1 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -458,7 +458,7 @@ static void tcp_fastopen_synack_timer(struct sock *sk, struct request_sock *req) struct tcp_sock *tp = tcp_sk(sk); int max_retries; - req->rsk_ops->syn_ack_timeout(req); + tcp_syn_ack_timeout(req); /* Add one more retry for fastopen. * Paired with WRITE_ONCE() in tcp_sock_set_syncnt() @@ -752,7 +752,6 @@ void tcp_syn_ack_timeout(const struct request_sock *req) __NET_INC_STATS(net, LINUX_MIB_TCPTIMEOUTS); } -EXPORT_IPV6_MOD(tcp_syn_ack_timeout); void tcp_reset_keepalive_timer(struct sock *sk, unsigned long len) { diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 7df21c1cba21..08113f430124 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -796,7 +796,6 @@ struct request_sock_ops tcp6_request_sock_ops __read_mostly = { .send_ack = tcp_v6_reqsk_send_ack, .destructor = tcp_v6_reqsk_destructor, .send_reset = tcp_v6_send_reset, - .syn_ack_timeout = tcp_syn_ack_timeout, }; const struct tcp_request_sock_ops tcp_request_sock_ipv6_ops = { From 3ce5dd8161ecdf12ffe0af99ff8980f1432f64a5 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Thu, 6 Nov 2025 00:32:41 +0000 Subject: [PATCH 0599/1381] tcp: Remove timeout arg from reqsk_queue_hash_req(). inet_csk_reqsk_queue_hash_add() is no longer shared by DCCP. We do not need to pass req->timeout down to reqsk_queue_hash_req(). Let's move tcp_timeout_init() from tcp_conn_request() to reqsk_queue_hash_req(). Signed-off-by: Kuniyuki Iwashima Reviewed-by: Eric Dumazet Link: https://patch.msgid.link/20251106003357.273403-3-kuniyu@google.com Signed-off-by: Jakub Kicinski --- include/net/inet_connection_sock.h | 3 +-- net/ipv4/inet_connection_sock.c | 11 +++++------ net/ipv4/tcp_input.c | 14 +++++--------- 3 files changed, 11 insertions(+), 17 deletions(-) diff --git a/include/net/inet_connection_sock.h b/include/net/inet_connection_sock.h index b4b886647607..90a99a2fc804 100644 --- a/include/net/inet_connection_sock.h +++ b/include/net/inet_connection_sock.h @@ -267,8 +267,7 @@ struct dst_entry *inet_csk_route_child_sock(const struct sock *sk, struct sock *inet_csk_reqsk_queue_add(struct sock *sk, struct request_sock *req, struct sock *child); -bool inet_csk_reqsk_queue_hash_add(struct sock *sk, struct request_sock *req, - unsigned long timeout); +bool inet_csk_reqsk_queue_hash_add(struct sock *sk, struct request_sock *req); struct sock *inet_csk_complete_hashdance(struct sock *sk, struct sock *child, struct request_sock *req, bool own_req); diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index 6a86c1ac3011..d9c674403eb0 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -1144,8 +1144,7 @@ static void reqsk_timer_handler(struct timer_list *t) reqsk_put(oreq); } -static bool reqsk_queue_hash_req(struct request_sock *req, - unsigned long timeout) +static bool reqsk_queue_hash_req(struct request_sock *req) { bool found_dup_sk = false; @@ -1153,8 +1152,9 @@ static bool reqsk_queue_hash_req(struct request_sock *req, return false; /* The timer needs to be setup after a successful insertion. */ + req->timeout = tcp_timeout_init((struct sock *)req); timer_setup(&req->rsk_timer, reqsk_timer_handler, TIMER_PINNED); - mod_timer(&req->rsk_timer, jiffies + timeout); + mod_timer(&req->rsk_timer, jiffies + req->timeout); /* before letting lookups find us, make sure all req fields * are committed to memory and refcnt initialized. @@ -1164,10 +1164,9 @@ static bool reqsk_queue_hash_req(struct request_sock *req, return true; } -bool inet_csk_reqsk_queue_hash_add(struct sock *sk, struct request_sock *req, - unsigned long timeout) +bool inet_csk_reqsk_queue_hash_add(struct sock *sk, struct request_sock *req) { - if (!reqsk_queue_hash_req(req, timeout)) + if (!reqsk_queue_hash_req(req)) return false; inet_csk_reqsk_queue_added(sk); diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 6db1d4c36a88..804ec56bdd24 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -7531,15 +7531,11 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops, sock_put(fastopen_sk); } else { tcp_rsk(req)->tfo_listener = false; - if (!want_cookie) { - req->timeout = tcp_timeout_init((struct sock *)req); - if (unlikely(!inet_csk_reqsk_queue_hash_add(sk, req, - req->timeout))) { - reqsk_free(req); - dst_release(dst); - return 0; - } - + if (!want_cookie && + unlikely(!inet_csk_reqsk_queue_hash_add(sk, req))) { + reqsk_free(req); + dst_release(dst); + return 0; } af_ops->send_synack(sk, dst, &fl, req, &foc, !want_cookie ? TCP_SYNACK_NORMAL : From 6fbf648d5cc48ebf250f32a8793a95a2596c0d50 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Thu, 6 Nov 2025 00:32:42 +0000 Subject: [PATCH 0600/1381] tcp: Remove redundant init for req->num_timeout. Commit 5903123f662e ("tcp: Use BPF timeout setting for SYN ACK RTO") introduced req->timeout and initialised it in 3 places: 1. reqsk_alloc() sets 0 2. inet_reqsk_alloc() sets TCP_TIMEOUT_INIT 3. tcp_conn_request() sets tcp_timeout_init() 1. has been always redundant as 2. overwrites it immediately. 2. was necessary for TFO SYN+ACK but is no longer needed after commit 8ea731d4c2ce ("tcp: Make SYN ACK RTO tunable by BPF programs with TFO"). 3. was moved to reqsk_queue_hash_req() in the previous patch. Now, we always initialise req->timeout just before scheduling the SYN+ACK timer: * For non-TFO SYN+ACK : reqsk_queue_hash_req() * For TFO SYN+ACK : tcp_fastopen_create_child() Let's remove the redundant initialisation of req->timeout in reqsk_alloc() and inet_reqsk_alloc(). Signed-off-by: Kuniyuki Iwashima Reviewed-by: Eric Dumazet Link: https://patch.msgid.link/20251106003357.273403-4-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/inet_connection_sock.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index d9c674403eb0..2bfe7af51bbb 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -885,7 +885,6 @@ reqsk_alloc_noprof(const struct request_sock_ops *ops, struct sock *sk_listener, sk_tx_queue_clear(req_to_sk(req)); req->saved_syn = NULL; req->syncookie = 0; - req->timeout = 0; req->num_timeout = 0; req->num_retrans = 0; req->sk = NULL; @@ -913,7 +912,6 @@ struct request_sock *inet_reqsk_alloc(const struct request_sock_ops *ops, ireq->ireq_state = TCP_NEW_SYN_RECV; write_pnet(&ireq->ireq_net, sock_net(sk_listener)); ireq->ireq_family = sk_listener->sk_family; - req->timeout = TCP_TIMEOUT_INIT; } return req; From 207ce0f6bc131812c96cf4f6db328af5396cebac Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Thu, 6 Nov 2025 00:32:43 +0000 Subject: [PATCH 0601/1381] tcp: Remove timeout arg from reqsk_timeout(). reqsk_timeout() is always called with @timeout being TCP_RTO_MAX. Let's remove the arg. As a prep for the next patch, reqsk_timeout() is moved to tcp.h and renamed to tcp_reqsk_timeout(). Signed-off-by: Kuniyuki Iwashima Reviewed-by: Eric Dumazet Link: https://patch.msgid.link/20251106003357.273403-5-kuniyu@google.com Signed-off-by: Jakub Kicinski --- include/net/inet_connection_sock.h | 8 -------- include/net/tcp.h | 7 +++++++ net/ipv4/inet_connection_sock.c | 2 +- net/ipv4/tcp_minisocks.c | 5 +++-- 4 files changed, 11 insertions(+), 11 deletions(-) diff --git a/include/net/inet_connection_sock.h b/include/net/inet_connection_sock.h index 90a99a2fc804..fd40af2221b9 100644 --- a/include/net/inet_connection_sock.h +++ b/include/net/inet_connection_sock.h @@ -290,14 +290,6 @@ static inline int inet_csk_reqsk_queue_is_full(const struct sock *sk) bool inet_csk_reqsk_queue_drop(struct sock *sk, struct request_sock *req); void inet_csk_reqsk_queue_drop_and_put(struct sock *sk, struct request_sock *req); -static inline unsigned long -reqsk_timeout(struct request_sock *req, unsigned long max_timeout) -{ - u64 timeout = (u64)req->timeout << req->num_timeout; - - return (unsigned long)min_t(u64, timeout, max_timeout); -} - void inet_csk_destroy_sock(struct sock *sk); void inet_csk_prepare_for_destroy_sock(struct sock *sk); void inet_csk_prepare_forced_close(struct sock *sk); diff --git a/include/net/tcp.h b/include/net/tcp.h index 0aa1f07d036a..0c7274ac7ed5 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -841,6 +841,13 @@ static inline u32 __tcp_set_rto(const struct tcp_sock *tp) return usecs_to_jiffies((tp->srtt_us >> 3) + tp->rttvar_us); } +static inline unsigned long tcp_reqsk_timeout(struct request_sock *req) +{ + u64 timeout = (u64)req->timeout << req->num_timeout; + + return (unsigned long)min_t(u64, timeout, TCP_RTO_MAX); +} + u32 tcp_delack_max(const struct sock *sk); /* Compute the actual rto_min value */ diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index 2bfe7af51bbb..b4eae731c9ba 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -1105,7 +1105,7 @@ static void reqsk_timer_handler(struct timer_list *t) inet_rsk(req)->acked)) { if (req->num_timeout++ == 0) atomic_dec(&queue->young); - mod_timer(&req->rsk_timer, jiffies + reqsk_timeout(req, TCP_RTO_MAX)); + mod_timer(&req->rsk_timer, jiffies + tcp_reqsk_timeout(req)); if (!nreq) return; diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index ded2cf1f6006..d8f4d813e8dd 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -714,7 +714,8 @@ struct sock *tcp_check_req(struct sock *sk, struct sk_buff *skb, * it can be estimated (approximately) * from another data. */ - tmp_opt.ts_recent_stamp = ktime_get_seconds() - reqsk_timeout(req, TCP_RTO_MAX) / HZ; + tmp_opt.ts_recent_stamp = ktime_get_seconds() - + tcp_reqsk_timeout(req) / HZ; paws_reject = tcp_paws_reject(&tmp_opt, th->rst); } } @@ -753,7 +754,7 @@ struct sock *tcp_check_req(struct sock *sk, struct sk_buff *skb, !tcp_rtx_synack(sk, req)) { unsigned long expires = jiffies; - expires += reqsk_timeout(req, TCP_RTO_MAX); + expires += tcp_reqsk_timeout(req); if (!fastopen) mod_timer_pending(&req->rsk_timer, expires); else From 1e9d3005e02cba82047d49f859982fc73b9a100b Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Thu, 6 Nov 2025 00:32:44 +0000 Subject: [PATCH 0602/1381] tcp: Apply max RTO to non-TFO SYN+ACK. Since commit 54a378f43425 ("tcp: add the ability to control max RTO"), TFO SYN+ACK RTO is capped by the TFO full sk's inet_csk(sk)->icsk_rto_max. The value is inherited from the parent listener. Let's apply the same cap to non-TFO SYN+ACK. Note that req->rsk_listener is always non-NULL when we call tcp_reqsk_timeout() in reqsk_timer_handler() or tcp_check_req(). It could be NULL for SYN cookie req, but we do not use req->timeout then. Signed-off-by: Kuniyuki Iwashima Reviewed-by: Eric Dumazet Link: https://patch.msgid.link/20251106003357.273403-6-kuniyu@google.com Signed-off-by: Jakub Kicinski --- include/net/tcp.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/include/net/tcp.h b/include/net/tcp.h index 0c7274ac7ed5..4833ec7903ec 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -845,7 +845,8 @@ static inline unsigned long tcp_reqsk_timeout(struct request_sock *req) { u64 timeout = (u64)req->timeout << req->num_timeout; - return (unsigned long)min_t(u64, timeout, TCP_RTO_MAX); + return (unsigned long)min_t(u64, timeout, + tcp_rto_max(req->rsk_listener)); } u32 tcp_delack_max(const struct sock *sk); From ffc56c90819e86d3a8c4eff6f831317d1c1476b6 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Thu, 6 Nov 2025 00:32:45 +0000 Subject: [PATCH 0603/1381] selftest: packetdrill: Add max RTO test for SYN+ACK. This script sets net.ipv4.tcp_rto_max_ms to 1000 and checks if SYN+ACK RTO is capped at 1s for TFO and non-TFO. Without the previous patch, the max RTO is applied to TFO SYN+ACK only, and non-TFO SYN+ACK RTO increases exponentially. # selftests: net/packetdrill: tcp_rto_synack_rto_max.pkt # TAP version 13 # 1..2 # tcp_rto_synack_rto_max.pkt:46: error handling packet: timing error: expected outbound packet at 5.091936 sec but happened at 6.107826 sec; tolerance 0.127974 sec # script packet: 5.091936 S. 0:0(0) ack 1 # actual packet: 6.107826 S. 0:0(0) ack 1 win 65535 # not ok 1 ipv4 # tcp_rto_synack_rto_max.pkt:46: error handling packet: timing error: expected outbound packet at 5.075901 sec but happened at 6.091841 sec; tolerance 0.127976 sec # script packet: 5.075901 S. 0:0(0) ack 1 # actual packet: 6.091841 S. 0:0(0) ack 1 win 65535 # not ok 2 ipv6 # # Totals: pass:0 fail:2 xfail:0 xpass:0 skip:0 error:0 not ok 49 selftests: net/packetdrill: tcp_rto_synack_rto_max.pkt # exit=1 With the previous patch, all SYN+ACKs are retransmitted after 1s. # selftests: net/packetdrill: tcp_rto_synack_rto_max.pkt # TAP version 13 # 1..2 # ok 1 ipv4 # ok 2 ipv6 # # Totals: pass:2 fail:0 xfail:0 xpass:0 skip:0 error:0 ok 49 selftests: net/packetdrill: tcp_rto_synack_rto_max.pkt Signed-off-by: Kuniyuki Iwashima Reviewed-by: Eric Dumazet Link: https://patch.msgid.link/20251106003357.273403-7-kuniyu@google.com Signed-off-by: Jakub Kicinski --- .../packetdrill/tcp_rto_synack_rto_max.pkt | 54 +++++++++++++++++++ 1 file changed, 54 insertions(+) create mode 100644 tools/testing/selftests/net/packetdrill/tcp_rto_synack_rto_max.pkt diff --git a/tools/testing/selftests/net/packetdrill/tcp_rto_synack_rto_max.pkt b/tools/testing/selftests/net/packetdrill/tcp_rto_synack_rto_max.pkt new file mode 100644 index 000000000000..47550df124ce --- /dev/null +++ b/tools/testing/selftests/net/packetdrill/tcp_rto_synack_rto_max.pkt @@ -0,0 +1,54 @@ +// SPDX-License-Identifier: GPL-2.0 +// +// Test SYN+ACK RTX with 1s RTO. +// +`./defaults.sh + ./set_sysctls.py /proc/sys/net/ipv4/tcp_rto_max_ms=1000` + +// +// Test 1: TFO SYN+ACK +// + 0 socket(..., SOCK_STREAM|SOCK_NONBLOCK, IPPROTO_TCP) = 3 + +0 setsockopt(3, SOL_SOCKET, SO_REUSEADDR, [1], 4) = 0 + +0 bind(3, ..., ...) = 0 + +0 listen(3, 1) = 0 + +0 setsockopt(3, SOL_TCP, TCP_FASTOPEN, [1], 4) = 0 + + +0 < S 0:10(10) win 1000 + +0 > S. 0:0(0) ack 11 + +// RTO must be capped to 1s + +1 > S. 0:0(0) ack 11 + +1 > S. 0:0(0) ack 11 + +1 > S. 0:0(0) ack 11 + + +0 < . 11:11(0) ack 1 win 1000 + +0 accept(3, ..., ...) = 4 + +0 %{ assert (tcpi_options & TCPI_OPT_SYN_DATA) != 0, tcpi_options }% + + +0 close(4) = 0 + +0 close(3) = 0 + + +// +// Test 2: non-TFO SYN+ACK +// + +0 socket(..., SOCK_STREAM|SOCK_NONBLOCK, IPPROTO_TCP) = 3 + +0 setsockopt(3, SOL_SOCKET, SO_REUSEADDR, [1], 4) = 0 + +0 bind(3, ..., ...) = 0 + +0 listen(3, 1) = 0 + + +0 < S 0:0(0) win 1000 + +0 > S. 0:0(0) ack 1 + +// RTO must be capped to 1s + +1 > S. 0:0(0) ack 1 + +1 > S. 0:0(0) ack 1 + +1 > S. 0:0(0) ack 1 + + +0 < . 1:1(0) ack 1 win 1000 + +0 accept(3, ..., ...) = 4 + +0 %{ assert (tcpi_options & TCPI_OPT_SYN_DATA) == 0, tcpi_options }% + + +0 close(4) = 0 + +0 close(3) = 0 From 416dd649f3aa3774907c668167a29c668dbc634b Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 6 Nov 2025 11:52:36 +0000 Subject: [PATCH 0604/1381] tcp: add net.ipv4.tcp_comp_sack_rtt_percent TCP SACK compression has been added in 2018 in commit 5d9f4262b7ea ("tcp: add SACK compression"). It is working great for WAN flows (with large RTT). Wifi in particular gets a significant boost _when_ ACK are suppressed. Add a new sysctl so that we can tune the very conservative 5 % value that has been used so far in this formula, so that small RTT flows can benefit from this feature. delay = min ( 5 % of RTT, 1 ms) This patch adds new tcp_comp_sack_rtt_percent sysctl to ease experiments and tuning. Given that we cap the delay to 1ms (tcp_comp_sack_delay_ns sysctl), set the default value to 33 %. Quoting Neal Cardwell ( https://lore.kernel.org/netdev/CADVnQymZ1tFnEA1Q=vtECs0=Db7zHQ8=+WCQtnhHFVbEOzjVnQ@mail.gmail.com/ ) The rationale for 33% is basically to try to facilitate pipelining, where there are always at least 3 ACKs and 3 GSO/TSO skbs per SRTT, so that the path can maintain a budget for 3 full-sized GSO/TSO skbs "in flight" at all times: + 1 skb in the qdisc waiting to be sent by the NIC next + 1 skb being sent by the NIC (being serialized by the NIC out onto the wire) + 1 skb being received and aggregated by the receiver machine's aggregation mechanism (some combination of LRO, GRO, and sack compression) Note that this is basically the same magic number (3) and the same rationales as: (a) tcp_tso_should_defer() ensuring that we defer sending data for no longer than cwnd/tcp_tso_win_divisor (where tcp_tso_win_divisor = 3), and (b) bbr_quantization_budget() ensuring that cwnd is at least 3 GSO/TSO skbs to maintain pipelining and full throughput at low RTTs Signed-off-by: Eric Dumazet Reviewed-by: Neal Cardwell Link: https://patch.msgid.link/20251106115236.3450026-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- Documentation/networking/ip-sysctl.rst | 13 +++++++++++-- include/net/netns/ipv4.h | 1 + net/ipv4/sysctl_net_ipv4.c | 9 +++++++++ net/ipv4/tcp_input.c | 26 ++++++++++++++++++-------- net/ipv4/tcp_ipv4.c | 1 + 5 files changed, 40 insertions(+), 10 deletions(-) diff --git a/Documentation/networking/ip-sysctl.rst b/Documentation/networking/ip-sysctl.rst index 7cd35bfd39e6..2bae61be1859 100644 --- a/Documentation/networking/ip-sysctl.rst +++ b/Documentation/networking/ip-sysctl.rst @@ -854,9 +854,18 @@ tcp_sack - BOOLEAN Default: 1 (enabled) +tcp_comp_sack_rtt_percent - INTEGER + Percentage of SRTT used for the compressed SACK feature. + See tcp_comp_sack_nr, tcp_comp_sack_delay_ns, tcp_comp_sack_slack_ns. + + Possible values : 1 - 1000 + + Default : 33 % + tcp_comp_sack_delay_ns - LONG INTEGER - TCP tries to reduce number of SACK sent, using a timer - based on 5% of SRTT, capped by this sysctl, in nano seconds. + TCP tries to reduce number of SACK sent, using a timer based + on tcp_comp_sack_rtt_percent of SRTT, capped by this sysctl + in nano seconds. The default is 1ms, based on TSO autosizing period. Default : 1,000,000 ns (1 ms) diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index 0e96c90e56c6..de9d36acc8e2 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -221,6 +221,7 @@ struct netns_ipv4 { int sysctl_tcp_pacing_ss_ratio; int sysctl_tcp_pacing_ca_ratio; unsigned int sysctl_tcp_child_ehash_entries; + int sysctl_tcp_comp_sack_rtt_percent; unsigned long sysctl_tcp_comp_sack_delay_ns; unsigned long sysctl_tcp_comp_sack_slack_ns; int sysctl_max_syn_backlog; diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 0c7c8f9041cb..35367f8e2da3 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -1451,6 +1451,15 @@ static struct ctl_table ipv4_net_table[] = { .mode = 0644, .proc_handler = proc_doulongvec_minmax, }, + { + .procname = "tcp_comp_sack_rtt_percent", + .data = &init_net.ipv4.sysctl_tcp_comp_sack_rtt_percent, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = SYSCTL_ONE, + .extra2 = SYSCTL_ONE_THOUSAND, + }, { .procname = "tcp_comp_sack_slack_ns", .data = &init_net.ipv4.sysctl_tcp_comp_sack_slack_ns, diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 804ec56bdd24..9df5d7515605 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -5893,7 +5893,9 @@ static inline void tcp_data_snd_check(struct sock *sk) static void __tcp_ack_snd_check(struct sock *sk, int ofo_possible) { struct tcp_sock *tp = tcp_sk(sk); - unsigned long rtt, delay; + struct net *net = sock_net(sk); + unsigned long rtt; + u64 delay; /* More than one full frame received... */ if (((tp->rcv_nxt - tp->rcv_wup) > inet_csk(sk)->icsk_ack.rcv_mss && @@ -5912,7 +5914,7 @@ static void __tcp_ack_snd_check(struct sock *sk, int ofo_possible) * Defer the ack until tcp_release_cb(). */ if (sock_owned_by_user_nocheck(sk) && - READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_backlog_ack_defer)) { + READ_ONCE(net->ipv4.sysctl_tcp_backlog_ack_defer)) { set_bit(TCP_ACK_DEFERRED, &sk->sk_tsq_flags); return; } @@ -5927,7 +5929,7 @@ static void __tcp_ack_snd_check(struct sock *sk, int ofo_possible) } if (!tcp_is_sack(tp) || - tp->compressed_ack >= READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_comp_sack_nr)) + tp->compressed_ack >= READ_ONCE(net->ipv4.sysctl_tcp_comp_sack_nr)) goto send_now; if (tp->compressed_ack_rcv_nxt != tp->rcv_nxt) { @@ -5942,18 +5944,26 @@ static void __tcp_ack_snd_check(struct sock *sk, int ofo_possible) if (hrtimer_is_queued(&tp->compressed_ack_timer)) return; - /* compress ack timer : 5 % of rtt, but no more than tcp_comp_sack_delay_ns */ + /* compress ack timer : comp_sack_rtt_percent of rtt, + * but no more than tcp_comp_sack_delay_ns. + */ rtt = tp->rcv_rtt_est.rtt_us; if (tp->srtt_us && tp->srtt_us < rtt) rtt = tp->srtt_us; - delay = min_t(unsigned long, - READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_comp_sack_delay_ns), - rtt * (NSEC_PER_USEC >> 3)/20); + /* delay = (rtt >> 3) * NSEC_PER_USEC * comp_sack_rtt_percent / 100 + * -> + * delay = rtt * 1.25 * comp_sack_rtt_percent + */ + delay = (u64)(rtt + (rtt >> 2)) * + READ_ONCE(net->ipv4.sysctl_tcp_comp_sack_rtt_percent); + + delay = min(delay, READ_ONCE(net->ipv4.sysctl_tcp_comp_sack_delay_ns)); + sock_hold(sk); hrtimer_start_range_ns(&tp->compressed_ack_timer, ns_to_ktime(delay), - READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_comp_sack_slack_ns), + READ_ONCE(net->ipv4.sysctl_tcp_comp_sack_slack_ns), HRTIMER_MODE_REL_PINNED_SOFT); } diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 0cfebac33a91..a7d9fec2950b 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -3595,6 +3595,7 @@ static int __net_init tcp_sk_init(struct net *net) net->ipv4.sysctl_tcp_comp_sack_delay_ns = NSEC_PER_MSEC; net->ipv4.sysctl_tcp_comp_sack_slack_ns = 100 * NSEC_PER_USEC; net->ipv4.sysctl_tcp_comp_sack_nr = 44; + net->ipv4.sysctl_tcp_comp_sack_rtt_percent = 33; net->ipv4.sysctl_tcp_backlog_ack_defer = 1; net->ipv4.sysctl_tcp_fastopen = TFO_CLIENT_ENABLE; net->ipv4.sysctl_tcp_fastopen_blackhole_timeout = 0; From 3f47e67dff1f7266e112c50313d63824f6f17102 Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Thu, 6 Nov 2025 13:53:23 +0100 Subject: [PATCH 0605/1381] net: airoha: Add the capability to consume out-of-order DMA tx descriptors EN7581 and AN7583 SoCs are capable of DMA mapping non-linear tx skbs on non-consecutive DMA descriptors. This feature is useful when multiple flows are queued on the same hw tx queue since it allows to fully utilize the available tx DMA descriptors and to avoid the starvation of high-priority flow we have in the current codebase due to head-of-line blocking introduced by low-priority flows. Tested-by: Xuegang Lu Reviewed-by: Jacob Keller Signed-off-by: Lorenzo Bianconi Link: https://patch.msgid.link/20251106-airoha-tx-linked-list-v2-1-0706d4a322bd@kernel.org Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/airoha/airoha_eth.c | 85 +++++++++++------------- drivers/net/ethernet/airoha/airoha_eth.h | 7 +- 2 files changed, 45 insertions(+), 47 deletions(-) diff --git a/drivers/net/ethernet/airoha/airoha_eth.c b/drivers/net/ethernet/airoha/airoha_eth.c index 688faf999e4c..75893c90a0a1 100644 --- a/drivers/net/ethernet/airoha/airoha_eth.c +++ b/drivers/net/ethernet/airoha/airoha_eth.c @@ -892,19 +892,13 @@ static int airoha_qdma_tx_napi_poll(struct napi_struct *napi, int budget) dma_unmap_single(eth->dev, e->dma_addr, e->dma_len, DMA_TO_DEVICE); - memset(e, 0, sizeof(*e)); + e->dma_addr = 0; + list_add_tail(&e->list, &q->tx_list); + WRITE_ONCE(desc->msg0, 0); WRITE_ONCE(desc->msg1, 0); q->queued--; - /* completion ring can report out-of-order indexes if hw QoS - * is enabled and packets with different priority are queued - * to same DMA ring. Take into account possible out-of-order - * reports incrementing DMA ring tail pointer - */ - while (q->tail != q->head && !q->entry[q->tail].dma_addr) - q->tail = (q->tail + 1) % q->ndesc; - if (skb) { u16 queue = skb_get_queue_mapping(skb); struct netdev_queue *txq; @@ -949,6 +943,7 @@ static int airoha_qdma_init_tx_queue(struct airoha_queue *q, q->ndesc = size; q->qdma = qdma; q->free_thr = 1 + MAX_SKB_FRAGS; + INIT_LIST_HEAD(&q->tx_list); q->entry = devm_kzalloc(eth->dev, q->ndesc * sizeof(*q->entry), GFP_KERNEL); @@ -961,9 +956,9 @@ static int airoha_qdma_init_tx_queue(struct airoha_queue *q, return -ENOMEM; for (i = 0; i < q->ndesc; i++) { - u32 val; + u32 val = FIELD_PREP(QDMA_DESC_DONE_MASK, 1); - val = FIELD_PREP(QDMA_DESC_DONE_MASK, 1); + list_add_tail(&q->entry[i].list, &q->tx_list); WRITE_ONCE(q->desc[i].ctrl, cpu_to_le32(val)); } @@ -973,9 +968,9 @@ static int airoha_qdma_init_tx_queue(struct airoha_queue *q, airoha_qdma_wr(qdma, REG_TX_RING_BASE(qid), dma_addr); airoha_qdma_rmw(qdma, REG_TX_CPU_IDX(qid), TX_RING_CPU_IDX_MASK, - FIELD_PREP(TX_RING_CPU_IDX_MASK, q->head)); + FIELD_PREP(TX_RING_CPU_IDX_MASK, 0)); airoha_qdma_rmw(qdma, REG_TX_DMA_IDX(qid), TX_RING_DMA_IDX_MASK, - FIELD_PREP(TX_RING_DMA_IDX_MASK, q->head)); + FIELD_PREP(TX_RING_DMA_IDX_MASK, 0)); return 0; } @@ -1031,17 +1026,21 @@ static int airoha_qdma_init_tx(struct airoha_qdma *qdma) static void airoha_qdma_cleanup_tx_queue(struct airoha_queue *q) { struct airoha_eth *eth = q->qdma->eth; + int i; spin_lock_bh(&q->lock); - while (q->queued) { - struct airoha_queue_entry *e = &q->entry[q->tail]; + for (i = 0; i < q->ndesc; i++) { + struct airoha_queue_entry *e = &q->entry[i]; + + if (!e->dma_addr) + continue; dma_unmap_single(eth->dev, e->dma_addr, e->dma_len, DMA_TO_DEVICE); dev_kfree_skb_any(e->skb); + e->dma_addr = 0; e->skb = NULL; - - q->tail = (q->tail + 1) % q->ndesc; + list_add_tail(&e->list, &q->tx_list); q->queued--; } spin_unlock_bh(&q->lock); @@ -1883,20 +1882,6 @@ static u32 airoha_get_dsa_tag(struct sk_buff *skb, struct net_device *dev) #endif } -static bool airoha_dev_tx_queue_busy(struct airoha_queue *q, u32 nr_frags) -{ - u32 tail = q->tail <= q->head ? q->tail + q->ndesc : q->tail; - u32 index = q->head + nr_frags; - - /* completion napi can free out-of-order tx descriptors if hw QoS is - * enabled and packets with different priorities are queued to the same - * DMA ring. Take into account possible out-of-order reports checking - * if the tx queue is full using circular buffer head/tail pointers - * instead of the number of queued packets. - */ - return index >= tail; -} - static int airoha_get_fe_port(struct airoha_gdm_port *port) { struct airoha_qdma *qdma = port->qdma; @@ -1919,8 +1904,10 @@ static netdev_tx_t airoha_dev_xmit(struct sk_buff *skb, struct airoha_gdm_port *port = netdev_priv(dev); struct airoha_qdma *qdma = port->qdma; u32 nr_frags, tag, msg0, msg1, len; + struct airoha_queue_entry *e; struct netdev_queue *txq; struct airoha_queue *q; + LIST_HEAD(tx_list); void *data; int i, qid; u16 index; @@ -1966,7 +1953,7 @@ static netdev_tx_t airoha_dev_xmit(struct sk_buff *skb, txq = netdev_get_tx_queue(dev, qid); nr_frags = 1 + skb_shinfo(skb)->nr_frags; - if (airoha_dev_tx_queue_busy(q, nr_frags)) { + if (q->queued + nr_frags >= q->ndesc) { /* not enough space in the queue */ netif_tx_stop_queue(txq); spin_unlock_bh(&q->lock); @@ -1975,11 +1962,13 @@ static netdev_tx_t airoha_dev_xmit(struct sk_buff *skb, len = skb_headlen(skb); data = skb->data; - index = q->head; + + e = list_first_entry(&q->tx_list, struct airoha_queue_entry, + list); + index = e - q->entry; for (i = 0; i < nr_frags; i++) { struct airoha_qdma_desc *desc = &q->desc[index]; - struct airoha_queue_entry *e = &q->entry[index]; skb_frag_t *frag = &skb_shinfo(skb)->frags[i]; dma_addr_t addr; u32 val; @@ -1989,7 +1978,14 @@ static netdev_tx_t airoha_dev_xmit(struct sk_buff *skb, if (unlikely(dma_mapping_error(dev->dev.parent, addr))) goto error_unmap; - index = (index + 1) % q->ndesc; + list_move_tail(&e->list, &tx_list); + e->skb = i ? NULL : skb; + e->dma_addr = addr; + e->dma_len = len; + + e = list_first_entry(&q->tx_list, struct airoha_queue_entry, + list); + index = e - q->entry; val = FIELD_PREP(QDMA_DESC_LEN_MASK, len); if (i < nr_frags - 1) @@ -2002,15 +1998,9 @@ static netdev_tx_t airoha_dev_xmit(struct sk_buff *skb, WRITE_ONCE(desc->msg1, cpu_to_le32(msg1)); WRITE_ONCE(desc->msg2, cpu_to_le32(0xffff)); - e->skb = i ? NULL : skb; - e->dma_addr = addr; - e->dma_len = len; - data = skb_frag_address(frag); len = skb_frag_size(frag); } - - q->head = index; q->queued += i; skb_tx_timestamp(skb); @@ -2019,7 +2009,7 @@ static netdev_tx_t airoha_dev_xmit(struct sk_buff *skb, if (netif_xmit_stopped(txq) || !netdev_xmit_more()) airoha_qdma_rmw(qdma, REG_TX_CPU_IDX(qid), TX_RING_CPU_IDX_MASK, - FIELD_PREP(TX_RING_CPU_IDX_MASK, q->head)); + FIELD_PREP(TX_RING_CPU_IDX_MASK, index)); if (q->ndesc - q->queued < q->free_thr) netif_tx_stop_queue(txq); @@ -2029,10 +2019,13 @@ static netdev_tx_t airoha_dev_xmit(struct sk_buff *skb, return NETDEV_TX_OK; error_unmap: - for (i--; i >= 0; i--) { - index = (q->head + i) % q->ndesc; - dma_unmap_single(dev->dev.parent, q->entry[index].dma_addr, - q->entry[index].dma_len, DMA_TO_DEVICE); + while (!list_empty(&tx_list)) { + e = list_first_entry(&tx_list, struct airoha_queue_entry, + list); + dma_unmap_single(dev->dev.parent, e->dma_addr, e->dma_len, + DMA_TO_DEVICE); + e->dma_addr = 0; + list_move_tail(&e->list, &q->tx_list); } spin_unlock_bh(&q->lock); diff --git a/drivers/net/ethernet/airoha/airoha_eth.h b/drivers/net/ethernet/airoha/airoha_eth.h index eb27a4ff5198..fbbc58133364 100644 --- a/drivers/net/ethernet/airoha/airoha_eth.h +++ b/drivers/net/ethernet/airoha/airoha_eth.h @@ -169,7 +169,10 @@ enum trtcm_param { struct airoha_queue_entry { union { void *buf; - struct sk_buff *skb; + struct { + struct list_head list; + struct sk_buff *skb; + }; }; dma_addr_t dma_addr; u16 dma_len; @@ -193,6 +196,8 @@ struct airoha_queue { struct napi_struct napi; struct page_pool *page_pool; struct sk_buff *skb; + + struct list_head tx_list; }; struct airoha_tx_irq_queue { From 140039580efa96d18576a6c9fa6b1158be8a1d0f Mon Sep 17 00:00:00 2001 From: Erni Sri Satya Vennela Date: Wed, 5 Nov 2025 11:04:28 -0800 Subject: [PATCH 0606/1381] net: mana: Fix incorrect speed reported by debugfs Once the netshaper is created for MANA, the current bandwidth is reported in debugfs like this: $ sudo ./tools/net/ynl/pyynl/cli.py \ --spec Documentation/netlink/specs/net_shaper.yaml \ --do set \ --json '{"ifindex":'3', "handle":{ "scope": "netdev", "id":'1' }, "bw-max": 200000000 }' None $ sudo cat /sys/kernel/debug/mana/1/vport0/current_speed 200 After the shaper is deleted, it is expected to report the maximum speed supported by the SKU. But currently it is reporting 0, which is incorrect. Fix this inconsistency, by resetting apc->speed to apc->max_speed during deletion of the shaper object. This will improve readability and debuggability. Signed-off-by: Erni Sri Satya Vennela Reviewed-by: Jacob Keller Link: https://patch.msgid.link/1762369468-32570-1-git-send-email-ernis@linux.microsoft.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/microsoft/mana/mana_en.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/microsoft/mana/mana_en.c b/drivers/net/ethernet/microsoft/mana/mana_en.c index 739087081dfd..cccd5b63cee6 100644 --- a/drivers/net/ethernet/microsoft/mana/mana_en.c +++ b/drivers/net/ethernet/microsoft/mana/mana_en.c @@ -854,7 +854,7 @@ static int mana_shaper_del(struct net_shaper_binding *binding, /* Reset mana port context parameters */ apc->handle.id = 0; apc->handle.scope = NET_SHAPER_SCOPE_UNSPEC; - apc->speed = 0; + apc->speed = apc->max_speed; } return err; From b87ee13e34931779ac1dcd3264beba50b54966fd Mon Sep 17 00:00:00 2001 From: Parthiban Veerasooran Date: Wed, 5 Nov 2025 10:42:12 +0530 Subject: [PATCH 0607/1381] net: phy: phy-c45: add OATC14 10BASE-T1S PHY cable diagnostic support Add support for Open Alliance TC14 (OATC14) 10BASE-T1S PHYs cable diagnostic feature. This patch implements: - genphy_c45_oatc14_cable_test_start() to initiate a cable test - genphy_c45_oatc14_cable_test_get_status() to retrieve test results - Helper function to map PHY cable test status to ethtool result codes - Function declarations and exports for use by PHY drivers This enables ethtool to report ok, open, short, and undetectable cable conditions on OATC14 10Base-T1S PHYs. Open Alliance TC14 10BASE-T1S Advanced Diagnostic PHY Features Specification ref: https://opensig.org/wp-content/uploads/2025/06/OPEN_Alliance_10BASE-T1S_Advanced_PHY_features_for-automotive_Ethernet_V2.1b.pdf Signed-off-by: Parthiban Veerasooran Link: https://patch.msgid.link/20251105051213.50443-2-parthiban.veerasooran@microchip.com Signed-off-by: Jakub Kicinski --- drivers/net/phy/mdio-open-alliance.h | 36 ++++++++ drivers/net/phy/phy-c45.c | 122 +++++++++++++++++++++++++++ include/linux/phy.h | 3 + 3 files changed, 161 insertions(+) diff --git a/drivers/net/phy/mdio-open-alliance.h b/drivers/net/phy/mdio-open-alliance.h index 931e14660d75..6850a3f0b31e 100644 --- a/drivers/net/phy/mdio-open-alliance.h +++ b/drivers/net/phy/mdio-open-alliance.h @@ -43,4 +43,40 @@ /* Version Identifiers */ #define OATC14_IDM 0x0a00 +/* + * Open Alliance TC14 (10BASE-T1S) - Advanced Diagnostic Features Registers + * + * Refer to the OPEN Alliance documentation: + * https://opensig.org/automotive-ethernet-specifications/ + * + * Specification: + * "10BASE-T1S Advanced Diagnostic PHY Features" + * https://opensig.org/wp-content/uploads/2025/06/OPEN_Alliance_10BASE-T1S_Advanced_PHY_features_for-automotive_Ethernet_V2.1b.pdf + */ +/* Advanced Diagnostic Features Capability Register*/ +#define MDIO_OATC14_ADFCAP 0xcc00 +#define OATC14_ADFCAP_HDD_CAPABILITY GENMASK(10, 8) + +/* Harness Defect Detection Register */ +#define MDIO_OATC14_HDD 0xcc01 +#define OATC14_HDD_CONTROL BIT(15) +#define OATC14_HDD_READY BIT(14) +#define OATC14_HDD_START_CONTROL BIT(13) +#define OATC14_HDD_VALID BIT(2) +#define OATC14_HDD_SHORT_OPEN_STATUS GENMASK(1, 0) + +/* Bus Short/Open Status: + * 0 0 - no fault; everything is ok. (Default) + * 0 1 - detected as an open or missing termination(s) + * 1 0 - detected as a short or extra termination(s) + * 1 1 - fault but fault type not detectable. More details can be available by + * vender specific register if supported. + */ +enum oatc14_hdd_status { + OATC14_HDD_STATUS_CABLE_OK = 0, + OATC14_HDD_STATUS_OPEN, + OATC14_HDD_STATUS_SHORT, + OATC14_HDD_STATUS_NOT_DETECTABLE, +}; + #endif /* __MDIO_OPEN_ALLIANCE__ */ diff --git a/drivers/net/phy/phy-c45.c b/drivers/net/phy/phy-c45.c index 1a7b32be4625..e8e5be4684ab 100644 --- a/drivers/net/phy/phy-c45.c +++ b/drivers/net/phy/phy-c45.c @@ -7,6 +7,7 @@ #include #include #include +#include #include "mdio-open-alliance.h" #include "phylib-internal.h" @@ -1573,3 +1574,124 @@ int genphy_c45_ethtool_set_eee(struct phy_device *phydev, return ret; } EXPORT_SYMBOL(genphy_c45_ethtool_set_eee); + +/** + * oatc14_cable_test_get_result_code - Convert hardware cable test status to + * ethtool result code. + * @status: The hardware-reported cable test status + * + * This helper function maps the OATC14 HDD cable test status to the + * corresponding ethtool cable test result code. It provides a translation + * between the device-specific status values and the standardized ethtool + * result codes. + * + * Return: + * * ETHTOOL_A_CABLE_RESULT_CODE_OK - Cable is OK + * * ETHTOOL_A_CABLE_RESULT_CODE_OPEN - Open circuit detected + * * ETHTOOL_A_CABLE_RESULT_CODE_SAME_SHORT - Short circuit detected + * * ETHTOOL_A_CABLE_RESULT_CODE_UNSPEC - Status not detectable or invalid + */ +static int oatc14_cable_test_get_result_code(enum oatc14_hdd_status status) +{ + switch (status) { + case OATC14_HDD_STATUS_CABLE_OK: + return ETHTOOL_A_CABLE_RESULT_CODE_OK; + case OATC14_HDD_STATUS_OPEN: + return ETHTOOL_A_CABLE_RESULT_CODE_OPEN; + case OATC14_HDD_STATUS_SHORT: + return ETHTOOL_A_CABLE_RESULT_CODE_SAME_SHORT; + case OATC14_HDD_STATUS_NOT_DETECTABLE: + default: + return ETHTOOL_A_CABLE_RESULT_CODE_UNSPEC; + } +} + +/** + * genphy_c45_oatc14_cable_test_get_status - Get status of OATC14 10Base-T1S + * PHY cable test. + * @phydev: pointer to the PHY device structure + * @finished: pointer to a boolean set true if the test is complete + * + * Retrieves the current status of the OATC14 10Base-T1S PHY cable test. + * This function reads the OATC14 HDD register to determine whether the test + * results are valid and whether the test has finished. + * + * If the test is complete, the function reports the cable test result via + * the ethtool cable test interface using ethnl_cable_test_result(), and then + * clears the test control bit in the PHY register to reset the test state. + * + * Return: 0 on success, or a negative error code on failure (e.g. register + * read/write error). + */ +int genphy_c45_oatc14_cable_test_get_status(struct phy_device *phydev, + bool *finished) +{ + int ret; + u8 sts; + + *finished = false; + + ret = phy_read_mmd(phydev, MDIO_MMD_VEND2, MDIO_OATC14_HDD); + if (ret < 0) + return ret; + + if (!(ret & OATC14_HDD_VALID)) + return 0; + + *finished = true; + + sts = FIELD_GET(OATC14_HDD_SHORT_OPEN_STATUS, ret); + + ret = ethnl_cable_test_result(phydev, ETHTOOL_A_CABLE_PAIR_A, + oatc14_cable_test_get_result_code(sts)); + if (ret) + return ret; + + return phy_clear_bits_mmd(phydev, MDIO_MMD_VEND2, + MDIO_OATC14_HDD, OATC14_HDD_CONTROL); +} +EXPORT_SYMBOL(genphy_c45_oatc14_cable_test_get_status); + +/** + * genphy_c45_oatc14_cable_test_start - Start a cable test on an OATC14 + * 10Base-T1S PHY. + * @phydev: Pointer to the PHY device structure + * + * This function initiates a cable diagnostic test on a Clause 45 OATC14 + * 10Base-T1S capable PHY device. It first reads the PHY’s advanced diagnostic + * capability register to check if High Definition Diagnostics (HDD) mode is + * supported. If the PHY does not report HDD capability, cable testing is not + * supported and the function returns -EOPNOTSUPP. + * + * For PHYs that support HDD, the function sets the appropriate control bits in + * the OATC14_HDD register to enable and start the cable diagnostic test. + * + * Return: + * * 0 on success + * * -EOPNOTSUPP if the PHY does not support HDD capability + * * A negative error code on I/O or register access failures + */ +int genphy_c45_oatc14_cable_test_start(struct phy_device *phydev) +{ + int ret; + + ret = phy_read_mmd(phydev, MDIO_MMD_VEND2, MDIO_OATC14_ADFCAP); + if (ret < 0) + return ret; + + if (!(ret & OATC14_ADFCAP_HDD_CAPABILITY)) + return -EOPNOTSUPP; + + ret = phy_set_bits_mmd(phydev, MDIO_MMD_VEND2, MDIO_OATC14_HDD, + OATC14_HDD_CONTROL); + if (ret) + return ret; + + ret = phy_read_mmd(phydev, MDIO_MMD_VEND2, MDIO_OATC14_HDD); + if (ret < 0) + return ret; + + return phy_set_bits_mmd(phydev, MDIO_MMD_VEND2, MDIO_OATC14_HDD, + OATC14_HDD_START_CONTROL); +} +EXPORT_SYMBOL(genphy_c45_oatc14_cable_test_start); diff --git a/include/linux/phy.h b/include/linux/phy.h index d145a200ea21..bf5457341ca8 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -2251,6 +2251,9 @@ int genphy_c45_ethtool_get_eee(struct phy_device *phydev, int genphy_c45_ethtool_set_eee(struct phy_device *phydev, struct ethtool_keee *data); int genphy_c45_an_config_eee_aneg(struct phy_device *phydev); +int genphy_c45_oatc14_cable_test_start(struct phy_device *phydev); +int genphy_c45_oatc14_cable_test_get_status(struct phy_device *phydev, + bool *finished); /* The gen10g_* functions are the old Clause 45 stub */ int gen10g_config_aneg(struct phy_device *phydev); From f424409483d26b3d54439cda832f273a54745cf6 Mon Sep 17 00:00:00 2001 From: Parthiban Veerasooran Date: Wed, 5 Nov 2025 10:42:13 +0530 Subject: [PATCH 0608/1381] net: phy: microchip_t1s:: add cable diagnostic support for LAN867x Rev.D0 Enable Open Alliance TC14 (OATC14) 10Base-T1S cable diagnostic feature for Microchip LAN867x Rev.D0 PHY by implementing `cable_test_start` and `cable_test_get_status` using the generic C45 functions. This allows the `ethtool` utility to perform cable diagnostic tests directly on the PHY, improving network troubleshooting and maintenance. Signed-off-by: Parthiban Veerasooran Link: https://patch.msgid.link/20251105051213.50443-3-parthiban.veerasooran@microchip.com Signed-off-by: Jakub Kicinski --- drivers/net/phy/microchip_t1s.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/phy/microchip_t1s.c b/drivers/net/phy/microchip_t1s.c index bce5cf087b19..5a0a66778977 100644 --- a/drivers/net/phy/microchip_t1s.c +++ b/drivers/net/phy/microchip_t1s.c @@ -573,6 +573,8 @@ static struct phy_driver microchip_t1s_driver[] = { .get_plca_cfg = genphy_c45_plca_get_cfg, .set_plca_cfg = lan86xx_plca_set_cfg, .get_plca_status = genphy_c45_plca_get_status, + .cable_test_start = genphy_c45_oatc14_cable_test_start, + .cable_test_get_status = genphy_c45_oatc14_cable_test_get_status, }, { PHY_ID_MATCH_EXACT(PHY_ID_LAN865X_REVB), From f73e0f46bbfab29b111ff52d047f15aa13623972 Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Wed, 5 Nov 2025 23:09:17 +0100 Subject: [PATCH 0609/1381] net: phy: fixed_phy: shrink size of struct fixed_phy_status All three members are effectively of type bool, so make this explicit and shrink size of struct fixed_phy_status. Signed-off-by: Heiner Kallweit Link: https://patch.msgid.link/9eca3d7e-fa64-4724-8fdc-f2c1a8f2ae8f@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/phy/fixed_phy.c | 2 +- include/linux/phy_fixed.h | 6 +++--- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/net/phy/fixed_phy.c b/drivers/net/phy/fixed_phy.c index d498d8a9bba6..9bd6937411e4 100644 --- a/drivers/net/phy/fixed_phy.c +++ b/drivers/net/phy/fixed_phy.c @@ -224,7 +224,7 @@ EXPORT_SYMBOL_GPL(fixed_phy_register); struct phy_device *fixed_phy_register_100fd(void) { static const struct fixed_phy_status status = { - .link = 1, + .link = true, .speed = SPEED_100, .duplex = DUPLEX_FULL, }; diff --git a/include/linux/phy_fixed.h b/include/linux/phy_fixed.h index 8bade999831c..436bff20f324 100644 --- a/include/linux/phy_fixed.h +++ b/include/linux/phy_fixed.h @@ -5,11 +5,11 @@ #include struct fixed_phy_status { - int link; int speed; int duplex; - int pause; - int asym_pause; + bool link:1; + bool pause:1; + bool asym_pause:1; }; struct device_node; From dae4a92399fa8d68aa917db6bb3245f83021e762 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 5 Nov 2025 16:26:02 -0800 Subject: [PATCH 0610/1381] psp: report basic stats from the core Track and report stats common to all psp devices from the core. A 'stale-event' is when the core marks the rx state of an active psp_assoc as incapable of authenticating psp encapsulated data. Signed-off-by: Daniel Zahka Link: https://patch.msgid.link/20251106002608.1578518-2-daniel.zahka@gmail.com Signed-off-by: Jakub Kicinski --- Documentation/netlink/specs/psp.yaml | 40 +++++++++++++++ include/net/psp/types.h | 9 ++++ include/uapi/linux/psp.h | 10 ++++ net/psp/psp-nl-gen.c | 19 +++++++ net/psp/psp-nl-gen.h | 2 + net/psp/psp_nl.c | 74 ++++++++++++++++++++++++++++ net/psp/psp_sock.c | 4 +- 7 files changed, 157 insertions(+), 1 deletion(-) diff --git a/Documentation/netlink/specs/psp.yaml b/Documentation/netlink/specs/psp.yaml index 944429e5c9a8..914148221384 100644 --- a/Documentation/netlink/specs/psp.yaml +++ b/Documentation/netlink/specs/psp.yaml @@ -76,6 +76,28 @@ attribute-sets: name: spi doc: Security Parameters Index (SPI) of the association. type: u32 + - + name: stats + attributes: + - + name: dev-id + doc: PSP device ID. + type: u32 + checks: + min: 1 + - + name: key-rotations + type: uint + doc: | + Number of key rotations during the lifetime of the device. + Kernel statistic. + - + name: stale-events + type: uint + doc: | + Number of times a socket's Rx got shut down due to using + a key which went stale (fully rotated out). + Kernel statistic. operations: list: @@ -177,6 +199,24 @@ operations: pre: psp-assoc-device-get-locked post: psp-device-unlock + - + name: get-stats + doc: Get device statistics. + attribute-set: stats + do: + request: + attributes: + - dev-id + reply: &stats-all + attributes: + - dev-id + - key-rotations + - stale-events + pre: psp-device-get-locked + post: psp-device-unlock + dump: + reply: *stats-all + mcast-groups: list: - diff --git a/include/net/psp/types.h b/include/net/psp/types.h index 31cee64b7c86..5b0ccaac3882 100644 --- a/include/net/psp/types.h +++ b/include/net/psp/types.h @@ -59,6 +59,10 @@ struct psp_dev_config { * device key * @stale_assocs: associations which use a rotated out key * + * @stats: statistics maintained by the core + * @stats.rotations: See stats attr key-rotations + * @stats.stales: See stats attr stale-events + * * @rcu: RCU head for freeing the structure */ struct psp_dev { @@ -81,6 +85,11 @@ struct psp_dev { struct list_head prev_assocs; struct list_head stale_assocs; + struct { + unsigned long rotations; + unsigned long stales; + } stats; + struct rcu_head rcu; }; diff --git a/include/uapi/linux/psp.h b/include/uapi/linux/psp.h index 607c42c39ba5..31592760ad79 100644 --- a/include/uapi/linux/psp.h +++ b/include/uapi/linux/psp.h @@ -45,6 +45,15 @@ enum { PSP_A_KEYS_MAX = (__PSP_A_KEYS_MAX - 1) }; +enum { + PSP_A_STATS_DEV_ID = 1, + PSP_A_STATS_KEY_ROTATIONS, + PSP_A_STATS_STALE_EVENTS, + + __PSP_A_STATS_MAX, + PSP_A_STATS_MAX = (__PSP_A_STATS_MAX - 1) +}; + enum { PSP_CMD_DEV_GET = 1, PSP_CMD_DEV_ADD_NTF, @@ -55,6 +64,7 @@ enum { PSP_CMD_KEY_ROTATE_NTF, PSP_CMD_RX_ASSOC, PSP_CMD_TX_ASSOC, + PSP_CMD_GET_STATS, __PSP_CMD_MAX, PSP_CMD_MAX = (__PSP_CMD_MAX - 1) diff --git a/net/psp/psp-nl-gen.c b/net/psp/psp-nl-gen.c index 9fdd6f831803..73f8b06d66f0 100644 --- a/net/psp/psp-nl-gen.c +++ b/net/psp/psp-nl-gen.c @@ -47,6 +47,11 @@ static const struct nla_policy psp_tx_assoc_nl_policy[PSP_A_ASSOC_SOCK_FD + 1] = [PSP_A_ASSOC_SOCK_FD] = { .type = NLA_U32, }, }; +/* PSP_CMD_GET_STATS - do */ +static const struct nla_policy psp_get_stats_nl_policy[PSP_A_STATS_DEV_ID + 1] = { + [PSP_A_STATS_DEV_ID] = NLA_POLICY_MIN(NLA_U32, 1), +}; + /* Ops table for psp */ static const struct genl_split_ops psp_nl_ops[] = { { @@ -99,6 +104,20 @@ static const struct genl_split_ops psp_nl_ops[] = { .maxattr = PSP_A_ASSOC_SOCK_FD, .flags = GENL_CMD_CAP_DO, }, + { + .cmd = PSP_CMD_GET_STATS, + .pre_doit = psp_device_get_locked, + .doit = psp_nl_get_stats_doit, + .post_doit = psp_device_unlock, + .policy = psp_get_stats_nl_policy, + .maxattr = PSP_A_STATS_DEV_ID, + .flags = GENL_CMD_CAP_DO, + }, + { + .cmd = PSP_CMD_GET_STATS, + .dumpit = psp_nl_get_stats_dumpit, + .flags = GENL_CMD_CAP_DUMP, + }, }; static const struct genl_multicast_group psp_nl_mcgrps[] = { diff --git a/net/psp/psp-nl-gen.h b/net/psp/psp-nl-gen.h index 25268ed11fb5..5bc3b5d5a53e 100644 --- a/net/psp/psp-nl-gen.h +++ b/net/psp/psp-nl-gen.h @@ -28,6 +28,8 @@ int psp_nl_dev_set_doit(struct sk_buff *skb, struct genl_info *info); int psp_nl_key_rotate_doit(struct sk_buff *skb, struct genl_info *info); int psp_nl_rx_assoc_doit(struct sk_buff *skb, struct genl_info *info); int psp_nl_tx_assoc_doit(struct sk_buff *skb, struct genl_info *info); +int psp_nl_get_stats_doit(struct sk_buff *skb, struct genl_info *info); +int psp_nl_get_stats_dumpit(struct sk_buff *skb, struct netlink_callback *cb); enum { PSP_NLGRP_MGMT, diff --git a/net/psp/psp_nl.c b/net/psp/psp_nl.c index 8aaca62744c3..f990cccbe99c 100644 --- a/net/psp/psp_nl.c +++ b/net/psp/psp_nl.c @@ -262,6 +262,7 @@ int psp_nl_key_rotate_doit(struct sk_buff *skb, struct genl_info *info) psd->generation & ~PSP_GEN_VALID_MASK); psp_assocs_key_rotated(psd); + psd->stats.rotations++; nlmsg_end(ntf, (struct nlmsghdr *)ntf->data); genlmsg_multicast_netns(&psp_nl_family, dev_net(psd->main_netdev), ntf, @@ -503,3 +504,76 @@ int psp_nl_tx_assoc_doit(struct sk_buff *skb, struct genl_info *info) nlmsg_free(rsp); return err; } + +static int +psp_nl_stats_fill(struct psp_dev *psd, struct sk_buff *rsp, + const struct genl_info *info) +{ + void *hdr; + + hdr = genlmsg_iput(rsp, info); + if (!hdr) + return -EMSGSIZE; + + if (nla_put_u32(rsp, PSP_A_STATS_DEV_ID, psd->id) || + nla_put_uint(rsp, PSP_A_STATS_KEY_ROTATIONS, + psd->stats.rotations) || + nla_put_uint(rsp, PSP_A_STATS_STALE_EVENTS, psd->stats.stales)) + goto err_cancel_msg; + + genlmsg_end(rsp, hdr); + return 0; + +err_cancel_msg: + genlmsg_cancel(rsp, hdr); + return -EMSGSIZE; +} + +int psp_nl_get_stats_doit(struct sk_buff *skb, struct genl_info *info) +{ + struct psp_dev *psd = info->user_ptr[0]; + struct sk_buff *rsp; + int err; + + rsp = genlmsg_new(GENLMSG_DEFAULT_SIZE, GFP_KERNEL); + if (!rsp) + return -ENOMEM; + + err = psp_nl_stats_fill(psd, rsp, info); + if (err) + goto err_free_msg; + + return genlmsg_reply(rsp, info); + +err_free_msg: + nlmsg_free(rsp); + return err; +} + +static int +psp_nl_stats_get_dumpit_one(struct sk_buff *rsp, struct netlink_callback *cb, + struct psp_dev *psd) +{ + if (psp_dev_check_access(psd, sock_net(rsp->sk))) + return 0; + + return psp_nl_stats_fill(psd, rsp, genl_info_dump(cb)); +} + +int psp_nl_get_stats_dumpit(struct sk_buff *rsp, struct netlink_callback *cb) +{ + struct psp_dev *psd; + int err = 0; + + mutex_lock(&psp_devs_lock); + xa_for_each_start(&psp_devs, cb->args[0], psd, cb->args[0]) { + mutex_lock(&psd->lock); + err = psp_nl_stats_get_dumpit_one(rsp, cb, psd); + mutex_unlock(&psd->lock); + if (err) + break; + } + mutex_unlock(&psp_devs_lock); + + return err; +} diff --git a/net/psp/psp_sock.c b/net/psp/psp_sock.c index a931d825d1cc..f785672b7df6 100644 --- a/net/psp/psp_sock.c +++ b/net/psp/psp_sock.c @@ -253,8 +253,10 @@ void psp_assocs_key_rotated(struct psp_dev *psd) /* Mark the stale associations as invalid, they will no longer * be able to Rx any traffic. */ - list_for_each_entry_safe(pas, next, &psd->prev_assocs, assocs_list) + list_for_each_entry_safe(pas, next, &psd->prev_assocs, assocs_list) { pas->generation |= ~PSP_GEN_VALID_MASK; + psd->stats.stales++; + } list_splice_init(&psd->prev_assocs, &psd->stale_assocs); list_splice_init(&psd->active_assocs, &psd->prev_assocs); From 2098cec32865422b3ddc6a1bd8beb055a3850451 Mon Sep 17 00:00:00 2001 From: Daniel Zahka Date: Wed, 5 Nov 2025 16:26:03 -0800 Subject: [PATCH 0611/1381] selftests: drv-net: psp: add assertions on core-tracked psp dev stats Add assertions to existing test cases to cover key rotations and 'stale-events'. Signed-off-by: Daniel Zahka Link: https://patch.msgid.link/20251106002608.1578518-3-daniel.zahka@gmail.com Signed-off-by: Jakub Kicinski --- tools/testing/selftests/drivers/net/psp.py | 13 +++++++++++++ 1 file changed, 13 insertions(+) diff --git a/tools/testing/selftests/drivers/net/psp.py b/tools/testing/selftests/drivers/net/psp.py index 4ae7a785ff10..06559ef49b9a 100755 --- a/tools/testing/selftests/drivers/net/psp.py +++ b/tools/testing/selftests/drivers/net/psp.py @@ -109,6 +109,10 @@ def _check_data_outq(s, exp_len, force_wait=False): time.sleep(0.01) ksft_eq(outq, exp_len) + +def _get_stat(cfg, key): + return cfg.pspnl.get_stats({'dev-id': cfg.psp_dev_id})[key] + # # Test case boiler plate # @@ -171,11 +175,16 @@ def dev_rotate(cfg): """ Test key rotation """ _init_psp_dev(cfg) + prev_rotations = _get_stat(cfg, 'key-rotations') + rot = cfg.pspnl.key_rotate({"id": cfg.psp_dev_id}) ksft_eq(rot['id'], cfg.psp_dev_id) rot = cfg.pspnl.key_rotate({"id": cfg.psp_dev_id}) ksft_eq(rot['id'], cfg.psp_dev_id) + cur_rotations = _get_stat(cfg, 'key-rotations') + ksft_eq(cur_rotations, prev_rotations + 2) + def dev_rotate_spi(cfg): """ Test key rotation and SPI check """ @@ -475,6 +484,7 @@ def data_stale_key(cfg): """ Test send on a double-rotated key """ _init_psp_dev(cfg) + prev_stale = _get_stat(cfg, 'stale-events') s = _make_psp_conn(cfg) try: rx_assoc = cfg.pspnl.rx_assoc({"version": 0, @@ -495,6 +505,9 @@ def data_stale_key(cfg): cfg.pspnl.key_rotate({"id": cfg.psp_dev_id}) cfg.pspnl.key_rotate({"id": cfg.psp_dev_id}) + cur_stale = _get_stat(cfg, 'stale-events') + ksft_gt(cur_stale, prev_stale) + s.send(b'0123456789' * 200) _check_data_outq(s, 2000, force_wait=True) finally: From f05d26198cf2c71f25f6bbe62ca4481c15543922 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 5 Nov 2025 16:26:04 -0800 Subject: [PATCH 0612/1381] psp: add stats from psp spec to driver facing api Provide a driver api for reporting device statistics required by the "Implementation Requirements" section of the PSP Architecture Specification. Use a warning to ensure drivers report stats required by the spec. Signed-off-by: Daniel Zahka Link: https://patch.msgid.link/20251106002608.1578518-4-daniel.zahka@gmail.com Signed-off-by: Jakub Kicinski --- Documentation/netlink/specs/psp.yaml | 55 ++++++++++++++++++++++++++++ include/net/psp/types.h | 23 ++++++++++++ include/uapi/linux/psp.h | 8 ++++ net/psp/psp_main.c | 3 +- net/psp/psp_nl.c | 21 ++++++++++- 5 files changed, 108 insertions(+), 2 deletions(-) diff --git a/Documentation/netlink/specs/psp.yaml b/Documentation/netlink/specs/psp.yaml index 914148221384..f3a57782d2cf 100644 --- a/Documentation/netlink/specs/psp.yaml +++ b/Documentation/netlink/specs/psp.yaml @@ -98,6 +98,61 @@ attribute-sets: Number of times a socket's Rx got shut down due to using a key which went stale (fully rotated out). Kernel statistic. + - + name: rx-packets + type: uint + doc: | + Number of successfully processed and authenticated PSP packets. + Device statistic (from the PSP spec). + - + name: rx-bytes + type: uint + doc: | + Number of successfully authenticated PSP bytes received, counting from + the first byte after the IV through the last byte of payload. + The fixed initial portion of the PSP header (16 bytes) + and the PSP trailer/ICV (16 bytes) are not included in this count. + Device statistic (from the PSP spec). + - + name: rx-auth-fail + type: uint + doc: | + Number of received PSP packets with unsuccessful authentication. + Device statistic (from the PSP spec). + - + name: rx-error + type: uint + doc: | + Number of received PSP packets with length/framing errors. + Device statistic (from the PSP spec). + - + name: rx-bad + type: uint + doc: | + Number of received PSP packets with miscellaneous errors + (invalid master key indicated by SPI, unsupported version, etc.) + Device statistic (from the PSP spec). + - + name: tx-packets + type: uint + doc: | + Number of successfully processed PSP packets for transmission. + Device statistic (from the PSP spec). + - + name: tx-bytes + type: uint + doc: | + Number of successfully processed PSP bytes for transmit, counting from + the first byte after the IV through the last byte of payload. + The fixed initial portion of the PSP header (16 bytes) + and the PSP trailer/ICV (16 bytes) are not included in this count. + Device statistic (from the PSP spec). + - + name: tx-error + type: uint + doc: | + Number of PSP packets for transmission with errors. + Device statistic (from the PSP spec). operations: list: diff --git a/include/net/psp/types.h b/include/net/psp/types.h index 5b0ccaac3882..25a9096d4e7d 100644 --- a/include/net/psp/types.h +++ b/include/net/psp/types.h @@ -150,6 +150,22 @@ struct psp_assoc { u8 drv_data[] __aligned(8); }; +struct psp_dev_stats { + union { + struct { + u64 rx_packets; + u64 rx_bytes; + u64 rx_auth_fail; + u64 rx_error; + u64 rx_bad; + u64 tx_packets; + u64 tx_bytes; + u64 tx_error; + }; + DECLARE_FLEX_ARRAY(u64, required); + }; +}; + /** * struct psp_dev_ops - netdev driver facing PSP callbacks */ @@ -188,6 +204,13 @@ struct psp_dev_ops { * Remove an association from the device. */ void (*tx_key_del)(struct psp_dev *psd, struct psp_assoc *pas); + + /** + * @get_stats: get statistics from the device + * Stats required by the spec must be maintained and filled in. + * Stats must be filled in member-by-member, never memset the struct. + */ + void (*get_stats)(struct psp_dev *psd, struct psp_dev_stats *stats); }; #endif /* __NET_PSP_H */ diff --git a/include/uapi/linux/psp.h b/include/uapi/linux/psp.h index 31592760ad79..d8449c043ba1 100644 --- a/include/uapi/linux/psp.h +++ b/include/uapi/linux/psp.h @@ -49,6 +49,14 @@ enum { PSP_A_STATS_DEV_ID = 1, PSP_A_STATS_KEY_ROTATIONS, PSP_A_STATS_STALE_EVENTS, + PSP_A_STATS_RX_PACKETS, + PSP_A_STATS_RX_BYTES, + PSP_A_STATS_RX_AUTH_FAIL, + PSP_A_STATS_RX_ERROR, + PSP_A_STATS_RX_BAD, + PSP_A_STATS_TX_PACKETS, + PSP_A_STATS_TX_BYTES, + PSP_A_STATS_TX_ERROR, __PSP_A_STATS_MAX, PSP_A_STATS_MAX = (__PSP_A_STATS_MAX - 1) diff --git a/net/psp/psp_main.c b/net/psp/psp_main.c index 481aaf0fc9fc..a8534124f626 100644 --- a/net/psp/psp_main.c +++ b/net/psp/psp_main.c @@ -60,7 +60,8 @@ psp_dev_create(struct net_device *netdev, !psd_ops->key_rotate || !psd_ops->rx_spi_alloc || !psd_ops->tx_key_add || - !psd_ops->tx_key_del)) + !psd_ops->tx_key_del || + !psd_ops->get_stats)) return ERR_PTR(-EINVAL); psd = kzalloc(sizeof(*psd), GFP_KERNEL); diff --git a/net/psp/psp_nl.c b/net/psp/psp_nl.c index f990cccbe99c..6afd7707ec12 100644 --- a/net/psp/psp_nl.c +++ b/net/psp/psp_nl.c @@ -1,5 +1,6 @@ // SPDX-License-Identifier: GPL-2.0-only +#include #include #include #include @@ -509,7 +510,17 @@ static int psp_nl_stats_fill(struct psp_dev *psd, struct sk_buff *rsp, const struct genl_info *info) { + unsigned int required_cnt = sizeof(struct psp_dev_stats) / sizeof(u64); + struct psp_dev_stats stats; void *hdr; + int i; + + memset(&stats, 0xff, sizeof(stats)); + psd->ops->get_stats(psd, &stats); + + for (i = 0; i < required_cnt; i++) + if (WARN_ON_ONCE(stats.required[i] == ETHTOOL_STAT_NOT_SET)) + return -EOPNOTSUPP; hdr = genlmsg_iput(rsp, info); if (!hdr) @@ -518,7 +529,15 @@ psp_nl_stats_fill(struct psp_dev *psd, struct sk_buff *rsp, if (nla_put_u32(rsp, PSP_A_STATS_DEV_ID, psd->id) || nla_put_uint(rsp, PSP_A_STATS_KEY_ROTATIONS, psd->stats.rotations) || - nla_put_uint(rsp, PSP_A_STATS_STALE_EVENTS, psd->stats.stales)) + nla_put_uint(rsp, PSP_A_STATS_STALE_EVENTS, psd->stats.stales) || + nla_put_uint(rsp, PSP_A_STATS_RX_PACKETS, stats.rx_packets) || + nla_put_uint(rsp, PSP_A_STATS_RX_BYTES, stats.rx_bytes) || + nla_put_uint(rsp, PSP_A_STATS_RX_AUTH_FAIL, stats.rx_auth_fail) || + nla_put_uint(rsp, PSP_A_STATS_RX_ERROR, stats.rx_error) || + nla_put_uint(rsp, PSP_A_STATS_RX_BAD, stats.rx_bad) || + nla_put_uint(rsp, PSP_A_STATS_TX_PACKETS, stats.tx_packets) || + nla_put_uint(rsp, PSP_A_STATS_TX_BYTES, stats.tx_bytes) || + nla_put_uint(rsp, PSP_A_STATS_TX_ERROR, stats.tx_error)) goto err_cancel_msg; genlmsg_end(rsp, hdr); From b1346219e5350990ab36cdf8d1b302c1b53fc21a Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 5 Nov 2025 16:26:05 -0800 Subject: [PATCH 0613/1381] net/mlx5e: Add PSP stats support for Rx/Tx flows Add all statistics described under the "Implementation Requirements" section of the PSP Architecture Specification: Rx successfully decrypted PSP packets: psp_rx_pkts : Number of packets decrypted successfully psp_rx_bytes : Number of bytes decrypted successfully Rx PSP authentication failure statistics: psp_rx_pkts_auth_fail : Number of PSP packets that failed authentication psp_rx_bytes_auth_fail : Number of PSP bytes that failed authentication Rx PSP bad frame error statistics: psp_rx_pkts_frame_err; psp_rx_bytes_frame_err; Rx PSP drop statistics: psp_rx_pkts_drop : Number of PSP packets dropped psp_rx_bytes_drop : Number of PSP bytes dropped Tx successfully encrypted PSP packets: psp_tx_pkts : Number of packets encrypted successfully psp_tx_bytes : Number of bytes encrypted successfully Tx drops: tx_drop : Number of misc psp related drops The above can be seen using the ynl cli: ./pyynl/cli.py --spec netlink/specs/psp.yaml --dump get-stats Signed-off-by: Raed Salem Signed-off-by: Rahul Rameshbabu Signed-off-by: Daniel Zahka Link: https://patch.msgid.link/20251106002608.1578518-5-daniel.zahka@gmail.com Signed-off-by: Jakub Kicinski --- .../mellanox/mlx5/core/en_accel/psp.c | 235 ++++++++++++++++-- .../mellanox/mlx5/core/en_accel/psp.h | 16 ++ .../mellanox/mlx5/core/en_accel/psp_rxtx.c | 1 + .../net/ethernet/mellanox/mlx5/core/en_main.c | 5 + 4 files changed, 241 insertions(+), 16 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/psp.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/psp.c index 8565cfe8d7dc..38e7c77cc851 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/psp.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/psp.c @@ -28,12 +28,15 @@ struct mlx5e_psp_tx { struct mlx5_flow_handle *rule; struct mutex mutex; /* Protect PSP TX steering */ u32 refcnt; + struct mlx5_fc *tx_counter; }; struct mlx5e_psp_rx_err { struct mlx5_flow_table *ft; struct mlx5_flow_handle *rule; - struct mlx5_flow_handle *drop_rule; + struct mlx5_flow_handle *auth_fail_rule; + struct mlx5_flow_handle *err_rule; + struct mlx5_flow_handle *bad_rule; struct mlx5_modify_hdr *copy_modify_hdr; }; @@ -50,6 +53,10 @@ struct mlx5e_accel_fs_psp_prot { struct mlx5e_accel_fs_psp { struct mlx5e_accel_fs_psp_prot fs_prot[ACCEL_FS_PSP_NUM_TYPES]; + struct mlx5_fc *rx_counter; + struct mlx5_fc *rx_auth_fail_counter; + struct mlx5_fc *rx_err_counter; + struct mlx5_fc *rx_bad_counter; }; struct mlx5e_psp_fs { @@ -72,9 +79,19 @@ static enum mlx5_traffic_types fs_psp2tt(enum accel_fs_psp_type i) static void accel_psp_fs_rx_err_del_rules(struct mlx5e_psp_fs *fs, struct mlx5e_psp_rx_err *rx_err) { - if (rx_err->drop_rule) { - mlx5_del_flow_rules(rx_err->drop_rule); - rx_err->drop_rule = NULL; + if (rx_err->bad_rule) { + mlx5_del_flow_rules(rx_err->bad_rule); + rx_err->bad_rule = NULL; + } + + if (rx_err->err_rule) { + mlx5_del_flow_rules(rx_err->err_rule); + rx_err->err_rule = NULL; + } + + if (rx_err->auth_fail_rule) { + mlx5_del_flow_rules(rx_err->auth_fail_rule); + rx_err->auth_fail_rule = NULL; } if (rx_err->rule) { @@ -117,6 +134,7 @@ static int accel_psp_fs_rx_err_add_rule(struct mlx5e_psp_fs *fs, { u8 action[MLX5_UN_SZ_BYTES(set_add_copy_action_in_auto)] = {}; struct mlx5_core_dev *mdev = fs->mdev; + struct mlx5_flow_destination dest[2]; struct mlx5_flow_act flow_act = {}; struct mlx5_modify_hdr *modify_hdr; struct mlx5_flow_handle *fte; @@ -147,10 +165,14 @@ static int accel_psp_fs_rx_err_add_rule(struct mlx5e_psp_fs *fs, accel_psp_setup_syndrome_match(spec, PSP_OK); /* create fte */ flow_act.action = MLX5_FLOW_CONTEXT_ACTION_MOD_HDR | - MLX5_FLOW_CONTEXT_ACTION_FWD_DEST; + MLX5_FLOW_CONTEXT_ACTION_FWD_DEST | + MLX5_FLOW_CONTEXT_ACTION_COUNT; flow_act.modify_hdr = modify_hdr; - fte = mlx5_add_flow_rules(rx_err->ft, spec, &flow_act, - &fs_prot->default_dest, 1); + dest[0].type = fs_prot->default_dest.type; + dest[0].ft = fs_prot->default_dest.ft; + dest[1].type = MLX5_FLOW_DESTINATION_TYPE_COUNTER; + dest[1].counter = fs->rx_fs->rx_counter; + fte = mlx5_add_flow_rules(rx_err->ft, spec, &flow_act, dest, 2); if (IS_ERR(fte)) { err = PTR_ERR(fte); mlx5_core_err(mdev, "fail to add psp rx err copy rule err=%d\n", err); @@ -158,22 +180,69 @@ static int accel_psp_fs_rx_err_add_rule(struct mlx5e_psp_fs *fs, } rx_err->rule = fte; - /* add default drop rule */ + /* add auth fail drop rule */ + memset(spec, 0, sizeof(*spec)); + memset(&flow_act, 0, sizeof(flow_act)); + accel_psp_setup_syndrome_match(spec, PSP_ICV_FAIL); + /* create fte */ + flow_act.action = MLX5_FLOW_CONTEXT_ACTION_DROP | + MLX5_FLOW_CONTEXT_ACTION_COUNT; + dest[0].type = MLX5_FLOW_DESTINATION_TYPE_COUNTER; + dest[0].counter = fs->rx_fs->rx_auth_fail_counter; + fte = mlx5_add_flow_rules(rx_err->ft, spec, &flow_act, dest, 1); + if (IS_ERR(fte)) { + err = PTR_ERR(fte); + mlx5_core_err(mdev, "fail to add psp rx auth fail drop rule err=%d\n", + err); + goto out_drop_rule; + } + rx_err->auth_fail_rule = fte; + + /* add framing drop rule */ + memset(spec, 0, sizeof(*spec)); + memset(&flow_act, 0, sizeof(flow_act)); + accel_psp_setup_syndrome_match(spec, PSP_BAD_TRAILER); + /* create fte */ + flow_act.action = MLX5_FLOW_CONTEXT_ACTION_DROP | + MLX5_FLOW_CONTEXT_ACTION_COUNT; + dest[0].type = MLX5_FLOW_DESTINATION_TYPE_COUNTER; + dest[0].counter = fs->rx_fs->rx_err_counter; + fte = mlx5_add_flow_rules(rx_err->ft, spec, &flow_act, dest, 1); + if (IS_ERR(fte)) { + err = PTR_ERR(fte); + mlx5_core_err(mdev, "fail to add psp rx framing err drop rule err=%d\n", + err); + goto out_drop_auth_fail_rule; + } + rx_err->err_rule = fte; + + /* add misc. errors drop rule */ memset(spec, 0, sizeof(*spec)); memset(&flow_act, 0, sizeof(flow_act)); /* create fte */ - flow_act.action = MLX5_FLOW_CONTEXT_ACTION_DROP; - fte = mlx5_add_flow_rules(rx_err->ft, spec, &flow_act, NULL, 0); + flow_act.action = MLX5_FLOW_CONTEXT_ACTION_DROP | + MLX5_FLOW_CONTEXT_ACTION_COUNT; + dest[0].type = MLX5_FLOW_DESTINATION_TYPE_COUNTER; + dest[0].counter = fs->rx_fs->rx_bad_counter; + fte = mlx5_add_flow_rules(rx_err->ft, spec, &flow_act, dest, 1); if (IS_ERR(fte)) { err = PTR_ERR(fte); - mlx5_core_err(mdev, "fail to add psp rx err drop rule err=%d\n", err); - goto out_drop_rule; + mlx5_core_err(mdev, "fail to add psp rx misc. err drop rule err=%d\n", + err); + goto out_drop_error_rule; } - rx_err->drop_rule = fte; + rx_err->bad_rule = fte; + rx_err->copy_modify_hdr = modify_hdr; goto out_spec; +out_drop_error_rule: + mlx5_del_flow_rules(rx_err->err_rule); + rx_err->err_rule = NULL; +out_drop_auth_fail_rule: + mlx5_del_flow_rules(rx_err->auth_fail_rule); + rx_err->auth_fail_rule = NULL; out_drop_rule: mlx5_del_flow_rules(rx_err->rule); rx_err->rule = NULL; @@ -461,6 +530,10 @@ static void accel_psp_fs_cleanup_rx(struct mlx5e_psp_fs *fs) return; accel_psp = fs->rx_fs; + mlx5_fc_destroy(fs->mdev, accel_psp->rx_bad_counter); + mlx5_fc_destroy(fs->mdev, accel_psp->rx_err_counter); + mlx5_fc_destroy(fs->mdev, accel_psp->rx_auth_fail_counter); + mlx5_fc_destroy(fs->mdev, accel_psp->rx_counter); for (i = 0; i < ACCEL_FS_PSP_NUM_TYPES; i++) { fs_prot = &accel_psp->fs_prot[i]; mutex_destroy(&fs_prot->prot_mutex); @@ -474,7 +547,10 @@ static int accel_psp_fs_init_rx(struct mlx5e_psp_fs *fs) { struct mlx5e_accel_fs_psp_prot *fs_prot; struct mlx5e_accel_fs_psp *accel_psp; + struct mlx5_core_dev *mdev = fs->mdev; + struct mlx5_fc *flow_counter; enum accel_fs_psp_type i; + int err; accel_psp = kzalloc(sizeof(*accel_psp), GFP_KERNEL); if (!accel_psp) @@ -485,9 +561,68 @@ static int accel_psp_fs_init_rx(struct mlx5e_psp_fs *fs) mutex_init(&fs_prot->prot_mutex); } + flow_counter = mlx5_fc_create(mdev, false); + if (IS_ERR(flow_counter)) { + mlx5_core_warn(mdev, + "fail to create psp rx flow counter err=%pe\n", + flow_counter); + err = PTR_ERR(flow_counter); + goto out_err; + } + accel_psp->rx_counter = flow_counter; + + flow_counter = mlx5_fc_create(mdev, false); + if (IS_ERR(flow_counter)) { + mlx5_core_warn(mdev, + "fail to create psp rx auth fail flow counter err=%pe\n", + flow_counter); + err = PTR_ERR(flow_counter); + goto out_counter_err; + } + accel_psp->rx_auth_fail_counter = flow_counter; + + flow_counter = mlx5_fc_create(mdev, false); + if (IS_ERR(flow_counter)) { + mlx5_core_warn(mdev, + "fail to create psp rx error flow counter err=%pe\n", + flow_counter); + err = PTR_ERR(flow_counter); + goto out_auth_fail_counter_err; + } + accel_psp->rx_err_counter = flow_counter; + + flow_counter = mlx5_fc_create(mdev, false); + if (IS_ERR(flow_counter)) { + mlx5_core_warn(mdev, + "fail to create psp rx bad flow counter err=%pe\n", + flow_counter); + err = PTR_ERR(flow_counter); + goto out_err_counter_err; + } + accel_psp->rx_bad_counter = flow_counter; + fs->rx_fs = accel_psp; return 0; + +out_err_counter_err: + mlx5_fc_destroy(mdev, accel_psp->rx_err_counter); + accel_psp->rx_err_counter = NULL; +out_auth_fail_counter_err: + mlx5_fc_destroy(mdev, accel_psp->rx_auth_fail_counter); + accel_psp->rx_auth_fail_counter = NULL; +out_counter_err: + mlx5_fc_destroy(mdev, accel_psp->rx_counter); + accel_psp->rx_counter = NULL; +out_err: + for (i = 0; i < ACCEL_FS_PSP_NUM_TYPES; i++) { + fs_prot = &accel_psp->fs_prot[i]; + mutex_destroy(&fs_prot->prot_mutex); + } + kfree(accel_psp); + fs->rx_fs = NULL; + + return err; } void mlx5_accel_psp_fs_cleanup_rx_tables(struct mlx5e_priv *priv) @@ -532,6 +667,7 @@ static int accel_psp_fs_tx_create_ft_table(struct mlx5e_psp_fs *fs) { int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in); struct mlx5_flow_table_attr ft_attr = {}; + struct mlx5_flow_destination dest = {}; struct mlx5_core_dev *mdev = fs->mdev; struct mlx5_flow_act flow_act = {}; u32 *in, *mc, *outer_headers_c; @@ -580,8 +716,11 @@ static int accel_psp_fs_tx_create_ft_table(struct mlx5e_psp_fs *fs) flow_act.crypto.type = MLX5_FLOW_CONTEXT_ENCRYPT_DECRYPT_TYPE_PSP; flow_act.flags |= FLOW_ACT_NO_APPEND; flow_act.action = MLX5_FLOW_CONTEXT_ACTION_ALLOW | - MLX5_FLOW_CONTEXT_ACTION_CRYPTO_ENCRYPT; - rule = mlx5_add_flow_rules(ft, spec, &flow_act, NULL, 0); + MLX5_FLOW_CONTEXT_ACTION_CRYPTO_ENCRYPT | + MLX5_FLOW_CONTEXT_ACTION_COUNT; + dest.type = MLX5_FLOW_DESTINATION_TYPE_COUNTER; + dest.counter = tx_fs->tx_counter; + rule = mlx5_add_flow_rules(ft, spec, &flow_act, &dest, 1); if (IS_ERR(rule)) { err = PTR_ERR(rule); mlx5_core_err(mdev, "PSP: fail to add psp tx flow rule, err = %d\n", err); @@ -650,6 +789,7 @@ static void accel_psp_fs_cleanup_tx(struct mlx5e_psp_fs *fs) if (!tx_fs) return; + mlx5_fc_destroy(fs->mdev, tx_fs->tx_counter); mutex_destroy(&tx_fs->mutex); WARN_ON(tx_fs->refcnt); kfree(tx_fs); @@ -658,10 +798,12 @@ static void accel_psp_fs_cleanup_tx(struct mlx5e_psp_fs *fs) static int accel_psp_fs_init_tx(struct mlx5e_psp_fs *fs) { + struct mlx5_core_dev *mdev = fs->mdev; struct mlx5_flow_namespace *ns; + struct mlx5_fc *flow_counter; struct mlx5e_psp_tx *tx_fs; - ns = mlx5_get_flow_namespace(fs->mdev, MLX5_FLOW_NAMESPACE_EGRESS_IPSEC); + ns = mlx5_get_flow_namespace(mdev, MLX5_FLOW_NAMESPACE_EGRESS_IPSEC); if (!ns) return -EOPNOTSUPP; @@ -669,12 +811,55 @@ static int accel_psp_fs_init_tx(struct mlx5e_psp_fs *fs) if (!tx_fs) return -ENOMEM; + flow_counter = mlx5_fc_create(mdev, false); + if (IS_ERR(flow_counter)) { + mlx5_core_warn(mdev, + "fail to create psp tx flow counter err=%pe\n", + flow_counter); + kfree(tx_fs); + return PTR_ERR(flow_counter); + } + tx_fs->tx_counter = flow_counter; mutex_init(&tx_fs->mutex); tx_fs->ns = ns; fs->tx_fs = tx_fs; return 0; } +static void +mlx5e_accel_psp_fs_get_stats_fill(struct mlx5e_priv *priv, + struct mlx5e_psp_stats *stats) +{ + struct mlx5e_psp_tx *tx_fs = priv->psp->fs->tx_fs; + struct mlx5_core_dev *mdev = priv->mdev; + struct mlx5e_accel_fs_psp *accel_psp; + + accel_psp = (struct mlx5e_accel_fs_psp *)priv->psp->fs->rx_fs; + + if (tx_fs->tx_counter) + mlx5_fc_query(mdev, tx_fs->tx_counter, &stats->psp_tx_pkts, + &stats->psp_tx_bytes); + + if (accel_psp->rx_counter) + mlx5_fc_query(mdev, accel_psp->rx_counter, &stats->psp_rx_pkts, + &stats->psp_rx_bytes); + + if (accel_psp->rx_auth_fail_counter) + mlx5_fc_query(mdev, accel_psp->rx_auth_fail_counter, + &stats->psp_rx_pkts_auth_fail, + &stats->psp_rx_bytes_auth_fail); + + if (accel_psp->rx_err_counter) + mlx5_fc_query(mdev, accel_psp->rx_err_counter, + &stats->psp_rx_pkts_frame_err, + &stats->psp_rx_bytes_frame_err); + + if (accel_psp->rx_bad_counter) + mlx5_fc_query(mdev, accel_psp->rx_bad_counter, + &stats->psp_rx_pkts_drop, + &stats->psp_rx_bytes_drop); +} + void mlx5_accel_psp_fs_cleanup_tx_tables(struct mlx5e_priv *priv) { if (!priv->psp) @@ -849,12 +1034,30 @@ mlx5e_psp_key_rotate(struct psp_dev *psd, struct netlink_ext_ack *exack) return mlx5e_psp_rotate_key(priv->mdev); } +static void +mlx5e_psp_get_stats(struct psp_dev *psd, struct psp_dev_stats *stats) +{ + struct mlx5e_priv *priv = netdev_priv(psd->main_netdev); + struct mlx5e_psp_stats nstats; + + mlx5e_accel_psp_fs_get_stats_fill(priv, &nstats); + stats->rx_packets = nstats.psp_rx_pkts; + stats->rx_bytes = nstats.psp_rx_bytes; + stats->rx_auth_fail = nstats.psp_rx_pkts_auth_fail; + stats->rx_error = nstats.psp_rx_pkts_frame_err; + stats->rx_bad = nstats.psp_rx_pkts_drop; + stats->tx_packets = nstats.psp_tx_pkts; + stats->tx_bytes = nstats.psp_tx_bytes; + stats->tx_error = atomic_read(&priv->psp->tx_drop); +} + static struct psp_dev_ops mlx5_psp_ops = { .set_config = mlx5e_psp_set_config, .rx_spi_alloc = mlx5e_psp_rx_spi_alloc, .tx_key_add = mlx5e_psp_assoc_add, .tx_key_del = mlx5e_psp_assoc_del, .key_rotate = mlx5e_psp_key_rotate, + .get_stats = mlx5e_psp_get_stats, }; void mlx5e_psp_unregister(struct mlx5e_priv *priv) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/psp.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/psp.h index 42bb671fb2cb..6b62fef0d9a7 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/psp.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/psp.h @@ -7,11 +7,27 @@ #include #include "en.h" +struct mlx5e_psp_stats { + u64 psp_rx_pkts; + u64 psp_rx_bytes; + u64 psp_rx_pkts_auth_fail; + u64 psp_rx_bytes_auth_fail; + u64 psp_rx_pkts_frame_err; + u64 psp_rx_bytes_frame_err; + u64 psp_rx_pkts_drop; + u64 psp_rx_bytes_drop; + u64 psp_tx_pkts; + u64 psp_tx_bytes; + u64 psp_tx_pkts_drop; + u64 psp_tx_bytes_drop; +}; + struct mlx5e_psp { struct psp_dev *psp; struct psp_dev_caps caps; struct mlx5e_psp_fs *fs; atomic_t tx_key_cnt; + atomic_t tx_drop; }; static inline bool mlx5_is_psp_device(struct mlx5_core_dev *mdev) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/psp_rxtx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/psp_rxtx.c index 828bff1137af..c17ea0fcd8ef 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/psp_rxtx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/psp_rxtx.c @@ -186,6 +186,7 @@ bool mlx5e_psp_handle_tx_skb(struct net_device *netdev, /* psp_encap of the packet */ if (!psp_dev_encapsulate(net, skb, psp_st->spi, psp_st->ver, 0)) { kfree_skb_reason(skb, SKB_DROP_REASON_PSP_OUTPUT); + atomic_inc(&priv->psp->tx_drop); return false; } if (skb_is_gso(skb)) { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index 8e394b16c92b..5ec0f5ca45b4 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -4025,6 +4025,11 @@ void mlx5e_fold_sw_stats64(struct mlx5e_priv *priv, struct rtnl_link_stats64 *s) s->rx_bytes += rq_stats->bytes; s->multicast += rq_stats->mcast_packets; } + +#ifdef CONFIG_MLX5_EN_PSP + if (priv->psp) + s->tx_dropped += atomic_read(&priv->psp->tx_drop); +#endif } void From 178f0763c5f3cf710062ddf2e4d659127fee66b2 Mon Sep 17 00:00:00 2001 From: Daniel Zahka Date: Wed, 5 Nov 2025 16:26:06 -0800 Subject: [PATCH 0614/1381] netdevsim: implement psp device stats For now only tx/rx packets/bytes are reported. This is not compliant with the PSP Architecture Specification. Signed-off-by: Daniel Zahka Link: https://patch.msgid.link/20251106002608.1578518-6-daniel.zahka@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/netdevsim/netdevsim.h | 5 +++++ drivers/net/netdevsim/psp.c | 27 +++++++++++++++++++++++++++ 2 files changed, 32 insertions(+) diff --git a/drivers/net/netdevsim/netdevsim.h b/drivers/net/netdevsim/netdevsim.h index 02c1c97b7008..af6fcfcda8ba 100644 --- a/drivers/net/netdevsim/netdevsim.h +++ b/drivers/net/netdevsim/netdevsim.h @@ -109,6 +109,11 @@ struct netdevsim { int rq_reset_mode; struct { + u64 rx_packets; + u64 rx_bytes; + u64 tx_packets; + u64 tx_bytes; + struct u64_stats_sync syncp; struct psp_dev *dev; u32 spi; u32 assoc_cnt; diff --git a/drivers/net/netdevsim/psp.c b/drivers/net/netdevsim/psp.c index 332b5b744f01..727da06101ca 100644 --- a/drivers/net/netdevsim/psp.c +++ b/drivers/net/netdevsim/psp.c @@ -70,6 +70,13 @@ nsim_do_psp(struct sk_buff *skb, struct netdevsim *ns, *psp_ext = skb->extensions; refcount_inc(&(*psp_ext)->refcnt); skb->decrypted = 1; + + u64_stats_update_begin(&ns->psp.syncp); + ns->psp.tx_packets++; + ns->psp.rx_packets++; + ns->psp.tx_bytes += skb->len - skb_inner_transport_offset(skb); + ns->psp.rx_bytes += skb->len - skb_inner_transport_offset(skb); + u64_stats_update_end(&ns->psp.syncp); } else { struct ipv6hdr *ip6h __maybe_unused; struct iphdr *iph; @@ -164,12 +171,32 @@ static void nsim_assoc_del(struct psp_dev *psd, struct psp_assoc *pas) ns->psp.assoc_cnt--; } +static void nsim_get_stats(struct psp_dev *psd, struct psp_dev_stats *stats) +{ + struct netdevsim *ns = psd->drv_priv; + unsigned int start; + + /* WARNING: do *not* blindly zero stats in real drivers! + * All required stats must be reported by the device! + */ + memset(stats, 0, sizeof(struct psp_dev_stats)); + + do { + start = u64_stats_fetch_begin(&ns->psp.syncp); + stats->rx_bytes = ns->psp.rx_bytes; + stats->rx_packets = ns->psp.rx_packets; + stats->tx_bytes = ns->psp.tx_bytes; + stats->tx_packets = ns->psp.tx_packets; + } while (u64_stats_fetch_retry(&ns->psp.syncp, start)); +} + static struct psp_dev_ops nsim_psp_ops = { .set_config = nsim_psp_set_config, .rx_spi_alloc = nsim_rx_spi_alloc, .tx_key_add = nsim_assoc_add, .tx_key_del = nsim_assoc_del, .key_rotate = nsim_key_rotate, + .get_stats = nsim_get_stats, }; static struct psp_dev_caps nsim_psp_caps = { From fd9557c3606bb683c01a6c7627e915b539b9a8df Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 6 Nov 2025 08:55:00 +0000 Subject: [PATCH 0615/1381] net: add prefetch() in skb_defer_free_flush() skb_defer_free_flush() is becoming more important these days. Add a prefetch operation to reduce latency a bit on some platforms like AMD EPYC 7B12. On more recent cpus, a stall happens when reading skb_shinfo(). Avoiding it will require a more elaborate strategy. Signed-off-by: Eric Dumazet Acked-by: Paolo Abeni Link: https://patch.msgid.link/20251106085500.2438951-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/core/dev.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/core/dev.c b/net/core/dev.c index 537aa43edff0..69515edd17bc 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -6782,6 +6782,7 @@ static void skb_defer_free_flush(void) free_list = llist_del_all(&sdn->defer_list); llist_for_each_entry_safe(skb, next, free_list, ll_node) { + prefetch(next); napi_consume_skb(skb, 1); } } From 1fcf572211da626982223ab4e4641bcd92db0701 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 6 Nov 2025 20:29:33 +0000 Subject: [PATCH 0616/1381] net: allow skb_release_head_state() to be called multiple times MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Currently, only skb dst is cleared (thanks to skb_dst_drop()) Make sure skb->destructor, conntrack and extensions are cleared. Signed-off-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Reviewed-by: Toke Høiland-Jørgensen Link: https://patch.msgid.link/20251106202935.1776179-2-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/core/skbuff.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 5b4bc8b1c7d5..eeddb9e737ff 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -1149,11 +1149,10 @@ void skb_release_head_state(struct sk_buff *skb) skb); #endif + skb->destructor = NULL; } -#if IS_ENABLED(CONFIG_NF_CONNTRACK) - nf_conntrack_put(skb_nfct(skb)); -#endif - skb_ext_put(skb); + nf_reset_ct(skb); + skb_ext_reset(skb); } /* Free everything but the sk_buff shell. */ From e20dfbad8aab2b7c72571ae3c3e2e646d6b04cb7 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 6 Nov 2025 20:29:34 +0000 Subject: [PATCH 0617/1381] net: fix napi_consume_skb() with alien skbs MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit There is a lack of NUMA awareness and more generally lack of slab caches affinity on TX completion path. Modern drivers are using napi_consume_skb(), hoping to cache sk_buff in per-cpu caches so that they can be recycled in RX path. Only use this if the skb was allocated on the same cpu, otherwise use skb_attempt_defer_free() so that the skb is freed on the original cpu. This removes contention on SLUB spinlocks and data structures. After this patch, I get ~50% improvement for an UDP tx workload on an AMD EPYC 9B45 (IDPF 200Gbit NIC with 32 TX queues). 80 Mpps -> 120 Mpps. Profiling one of the 32 cpus servicing NIC interrupts : Before: mpstat -P 511 1 1 Average: CPU %usr %nice %sys %iowait %irq %soft %steal %guest %gnice %idle Average: 511 0.00 0.00 0.00 0.00 0.00 98.00 0.00 0.00 0.00 2.00 31.01% ksoftirqd/511 [kernel.kallsyms] [k] queued_spin_lock_slowpath 12.45% swapper [kernel.kallsyms] [k] queued_spin_lock_slowpath 5.60% ksoftirqd/511 [kernel.kallsyms] [k] __slab_free 3.31% ksoftirqd/511 [kernel.kallsyms] [k] idpf_tx_clean_buf_ring 3.27% ksoftirqd/511 [kernel.kallsyms] [k] idpf_tx_splitq_clean_all 2.95% ksoftirqd/511 [kernel.kallsyms] [k] idpf_tx_splitq_start 2.52% ksoftirqd/511 [kernel.kallsyms] [k] fq_dequeue 2.32% ksoftirqd/511 [kernel.kallsyms] [k] read_tsc 2.25% ksoftirqd/511 [kernel.kallsyms] [k] build_detached_freelist 2.15% ksoftirqd/511 [kernel.kallsyms] [k] kmem_cache_free 2.11% swapper [kernel.kallsyms] [k] __slab_free 2.06% ksoftirqd/511 [kernel.kallsyms] [k] idpf_features_check 2.01% ksoftirqd/511 [kernel.kallsyms] [k] idpf_tx_splitq_clean_hdr 1.97% ksoftirqd/511 [kernel.kallsyms] [k] skb_release_data 1.52% ksoftirqd/511 [kernel.kallsyms] [k] sock_wfree 1.34% swapper [kernel.kallsyms] [k] idpf_tx_clean_buf_ring 1.23% swapper [kernel.kallsyms] [k] idpf_tx_splitq_clean_all 1.15% ksoftirqd/511 [kernel.kallsyms] [k] dma_unmap_page_attrs 1.11% swapper [kernel.kallsyms] [k] idpf_tx_splitq_start 1.03% swapper [kernel.kallsyms] [k] fq_dequeue 0.94% swapper [kernel.kallsyms] [k] kmem_cache_free 0.93% swapper [kernel.kallsyms] [k] read_tsc 0.81% ksoftirqd/511 [kernel.kallsyms] [k] napi_consume_skb 0.79% swapper [kernel.kallsyms] [k] idpf_tx_splitq_clean_hdr 0.77% ksoftirqd/511 [kernel.kallsyms] [k] skb_free_head 0.76% swapper [kernel.kallsyms] [k] idpf_features_check 0.72% swapper [kernel.kallsyms] [k] skb_release_data 0.69% swapper [kernel.kallsyms] [k] build_detached_freelist 0.58% ksoftirqd/511 [kernel.kallsyms] [k] skb_release_head_state 0.56% ksoftirqd/511 [kernel.kallsyms] [k] __put_partials 0.55% ksoftirqd/511 [kernel.kallsyms] [k] kmem_cache_free_bulk 0.48% swapper [kernel.kallsyms] [k] sock_wfree After: mpstat -P 511 1 1 Average: CPU %usr %nice %sys %iowait %irq %soft %steal %guest %gnice %idle Average: 511 0.00 0.00 0.00 0.00 0.00 51.49 0.00 0.00 0.00 48.51 19.10% swapper [kernel.kallsyms] [k] idpf_tx_splitq_clean_hdr 13.86% swapper [kernel.kallsyms] [k] idpf_tx_clean_buf_ring 10.80% swapper [kernel.kallsyms] [k] skb_attempt_defer_free 10.57% swapper [kernel.kallsyms] [k] idpf_tx_splitq_clean_all 7.18% swapper [kernel.kallsyms] [k] queued_spin_lock_slowpath 6.69% swapper [kernel.kallsyms] [k] sock_wfree 5.55% swapper [kernel.kallsyms] [k] dma_unmap_page_attrs 3.10% swapper [kernel.kallsyms] [k] fq_dequeue 3.00% swapper [kernel.kallsyms] [k] skb_release_head_state 2.73% swapper [kernel.kallsyms] [k] read_tsc 2.48% swapper [kernel.kallsyms] [k] idpf_tx_splitq_start 1.20% swapper [kernel.kallsyms] [k] idpf_features_check 1.13% swapper [kernel.kallsyms] [k] napi_consume_skb 0.93% swapper [kernel.kallsyms] [k] idpf_vport_splitq_napi_poll 0.64% swapper [kernel.kallsyms] [k] native_send_call_func_single_ipi 0.60% swapper [kernel.kallsyms] [k] acpi_processor_ffh_cstate_enter 0.53% swapper [kernel.kallsyms] [k] io_idle 0.43% swapper [kernel.kallsyms] [k] netif_skb_features 0.41% swapper [kernel.kallsyms] [k] __direct_call_cpuidle_state_enter2 0.40% swapper [kernel.kallsyms] [k] native_irq_return_iret 0.40% swapper [kernel.kallsyms] [k] idpf_tx_buf_hw_update 0.36% swapper [kernel.kallsyms] [k] sched_clock_noinstr 0.34% swapper [kernel.kallsyms] [k] handle_softirqs 0.32% swapper [kernel.kallsyms] [k] net_rx_action 0.32% swapper [kernel.kallsyms] [k] dql_completed 0.32% swapper [kernel.kallsyms] [k] validate_xmit_skb 0.31% swapper [kernel.kallsyms] [k] skb_network_protocol 0.29% swapper [kernel.kallsyms] [k] skb_csum_hwoffload_help 0.29% swapper [kernel.kallsyms] [k] x2apic_send_IPI 0.28% swapper [kernel.kallsyms] [k] ktime_get 0.24% swapper [kernel.kallsyms] [k] __qdisc_run Signed-off-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Reviewed-by: Toke Høiland-Jørgensen Reviewed-by: Jason Xing Link: https://patch.msgid.link/20251106202935.1776179-3-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/core/skbuff.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/net/core/skbuff.c b/net/core/skbuff.c index eeddb9e737ff..7ac5f8aa1235 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -1476,6 +1476,11 @@ void napi_consume_skb(struct sk_buff *skb, int budget) DEBUG_NET_WARN_ON_ONCE(!in_softirq()); + if (skb->alloc_cpu != smp_processor_id() && !skb_shared(skb)) { + skb_release_head_state(skb); + return skb_attempt_defer_free(skb); + } + if (!skb_unref(skb)) return; From b61785852ed0a0e7dc16b606157e4a0228cd76cf Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 6 Nov 2025 20:29:35 +0000 Subject: [PATCH 0618/1381] net: increase skb_defer_max default to 128 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit skb_defer_max value is very conservative, and can be increased to avoid too many calls to kick_defer_list_purge(). Signed-off-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Reviewed-by: Toke Høiland-Jørgensen Reviewed-by: Jason Xing Link: https://patch.msgid.link/20251106202935.1776179-4-edumazet@google.com Signed-off-by: Jakub Kicinski --- Documentation/admin-guide/sysctl/net.rst | 4 ++-- net/core/hotdata.c | 2 +- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/Documentation/admin-guide/sysctl/net.rst b/Documentation/admin-guide/sysctl/net.rst index 991773dcb9cf..369a738a6819 100644 --- a/Documentation/admin-guide/sysctl/net.rst +++ b/Documentation/admin-guide/sysctl/net.rst @@ -355,9 +355,9 @@ skb_defer_max ------------- Max size (in skbs) of the per-cpu list of skbs being freed -by the cpu which allocated them. Used by TCP stack so far. +by the cpu which allocated them. -Default: 64 +Default: 128 optmem_max ---------- diff --git a/net/core/hotdata.c b/net/core/hotdata.c index 95d0a4df1006..dddd5c287cf0 100644 --- a/net/core/hotdata.c +++ b/net/core/hotdata.c @@ -20,7 +20,7 @@ struct net_hotdata net_hotdata __cacheline_aligned = { .dev_tx_weight = 64, .dev_rx_weight = 64, .sysctl_max_skb_frags = MAX_SKB_FRAGS, - .sysctl_skb_defer_max = 64, + .sysctl_skb_defer_max = 128, .sysctl_mem_pcpu_rsv = SK_MEMORY_PCPU_RESERVE }; EXPORT_SYMBOL(net_hotdata); From 5636fcdb02114f492dd922cef67d1b1b5f09893b Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Thu, 6 Nov 2025 11:23:11 +0000 Subject: [PATCH 0619/1381] net: stmmac: lpc18xx: convert to PHY_INTF_SEL_x Use the common dwmac definitions for the PHY interface selection field. Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vGy59-0000000DhQ1-393H@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/dwmac-lpc18xx.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-lpc18xx.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-lpc18xx.c index 6fffc9dfbae5..66c309a7afb3 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-lpc18xx.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-lpc18xx.c @@ -22,8 +22,8 @@ /* Register defines for CREG syscon */ #define LPC18XX_CREG_CREG6 0x12c # define LPC18XX_CREG_CREG6_ETHMODE_MASK 0x7 -# define LPC18XX_CREG_CREG6_ETHMODE_MII 0x0 -# define LPC18XX_CREG_CREG6_ETHMODE_RMII 0x4 +# define LPC18XX_CREG_CREG6_ETHMODE_MII PHY_INTF_SEL_GMII_MII +# define LPC18XX_CREG_CREG6_ETHMODE_RMII PHY_INTF_SEL_RMII static int lpc18xx_dwmac_probe(struct platform_device *pdev) { From eb0533c7e63be03c1220f24bd00aeb2a4f0e3f78 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Thu, 6 Nov 2025 11:23:16 +0000 Subject: [PATCH 0620/1381] net: stmmac: lpc18xx: use PHY_INTF_SEL_x directly Use the PHY_INTF_SEL_x values directly rather than the driver private LPC18XX_CREG_CREG6_ETHMODE_x definitions, and convert LPC18XX_CREG_CREG6_ETHMODE_MASK to use GENMASK(). Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vGy5E-0000000DhQ7-3cuy@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/dwmac-lpc18xx.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-lpc18xx.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-lpc18xx.c index 66c309a7afb3..895d16dc0a4b 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-lpc18xx.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-lpc18xx.c @@ -21,9 +21,7 @@ /* Register defines for CREG syscon */ #define LPC18XX_CREG_CREG6 0x12c -# define LPC18XX_CREG_CREG6_ETHMODE_MASK 0x7 -# define LPC18XX_CREG_CREG6_ETHMODE_MII PHY_INTF_SEL_GMII_MII -# define LPC18XX_CREG_CREG6_ETHMODE_RMII PHY_INTF_SEL_RMII +# define LPC18XX_CREG_CREG6_ETHMODE_MASK GENMASK(2, 0) static int lpc18xx_dwmac_probe(struct platform_device *pdev) { @@ -50,9 +48,9 @@ static int lpc18xx_dwmac_probe(struct platform_device *pdev) } if (plat_dat->phy_interface == PHY_INTERFACE_MODE_MII) { - ethmode = LPC18XX_CREG_CREG6_ETHMODE_MII; + ethmode = PHY_INTF_SEL_GMII_MII; } else if (plat_dat->phy_interface == PHY_INTERFACE_MODE_RMII) { - ethmode = LPC18XX_CREG_CREG6_ETHMODE_RMII; + ethmode = PHY_INTF_SEL_RMII; } else { dev_err(&pdev->dev, "Only MII and RMII mode supported\n"); return -EINVAL; From 9882f12194082a0178a025946911e677cfdcd8d8 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Thu, 6 Nov 2025 11:23:21 +0000 Subject: [PATCH 0621/1381] net: stmmac: lpc18xx: use stmmac_get_phy_intf_sel() Use stmmac_get_phy_intf_sel() to decode the PHY interface mode to the phy_intf_sel value, and use the result to program the ethernet mode. Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vGy5J-0000000DhQD-46Ob@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/dwmac-lpc18xx.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-lpc18xx.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-lpc18xx.c index 895d16dc0a4b..0f6be2a17e65 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-lpc18xx.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-lpc18xx.c @@ -47,15 +47,14 @@ static int lpc18xx_dwmac_probe(struct platform_device *pdev) return PTR_ERR(reg); } - if (plat_dat->phy_interface == PHY_INTERFACE_MODE_MII) { - ethmode = PHY_INTF_SEL_GMII_MII; - } else if (plat_dat->phy_interface == PHY_INTERFACE_MODE_RMII) { - ethmode = PHY_INTF_SEL_RMII; - } else { + if (plat_dat->phy_interface != PHY_INTERFACE_MODE_MII && + plat_dat->phy_interface != PHY_INTERFACE_MODE_RMII) { dev_err(&pdev->dev, "Only MII and RMII mode supported\n"); return -EINVAL; } + ethmode = stmmac_get_phy_intf_sel(plat_dat->phy_interface); + regmap_update_bits(reg, LPC18XX_CREG_CREG6, LPC18XX_CREG_CREG6_ETHMODE_MASK, ethmode); From 4bad4219249f12dabb5c246f5cf9378f3693e226 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Thu, 6 Nov 2025 11:23:27 +0000 Subject: [PATCH 0622/1381] net: stmmac: lpc18xx: validate phy_intf_sel Validate the phy_intf_sel value rather than the PHY interface mode. This will allow us to transition to the ->set_phy_intf_sel() method. Note that this will allow GMII as well as MII as the phy_intf_sel value is the same for both. Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vGy5P-0000000DhQJ-0Oi3@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/dwmac-lpc18xx.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-lpc18xx.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-lpc18xx.c index 0f6be2a17e65..ec60968113b8 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-lpc18xx.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-lpc18xx.c @@ -47,14 +47,13 @@ static int lpc18xx_dwmac_probe(struct platform_device *pdev) return PTR_ERR(reg); } - if (plat_dat->phy_interface != PHY_INTERFACE_MODE_MII && - plat_dat->phy_interface != PHY_INTERFACE_MODE_RMII) { + ethmode = stmmac_get_phy_intf_sel(plat_dat->phy_interface); + if (ethmode != PHY_INTF_SEL_GMII_MII && + ethmode != PHY_INTF_SEL_RMII) { dev_err(&pdev->dev, "Only MII and RMII mode supported\n"); return -EINVAL; } - ethmode = stmmac_get_phy_intf_sel(plat_dat->phy_interface); - regmap_update_bits(reg, LPC18XX_CREG_CREG6, LPC18XX_CREG_CREG6_ETHMODE_MASK, ethmode); From 7fe0e06a7364ff7ffb3b44fe1c9340824b45c1a8 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Thu, 6 Nov 2025 11:23:32 +0000 Subject: [PATCH 0623/1381] net: stmmac: lpc18xx: use ->set_phy_intf_sel() Move the configuration of the dwmac PHY interface selection to the new ->set_phy_intf_sel() method. Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vGy5U-0000000DhQP-19Hd@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../ethernet/stmicro/stmmac/dwmac-lpc18xx.c | 36 +++++++++++-------- 1 file changed, 22 insertions(+), 14 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-lpc18xx.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-lpc18xx.c index ec60968113b8..c68d7de1f8ac 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-lpc18xx.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-lpc18xx.c @@ -23,12 +23,27 @@ #define LPC18XX_CREG_CREG6 0x12c # define LPC18XX_CREG_CREG6_ETHMODE_MASK GENMASK(2, 0) +static int lpc18xx_set_phy_intf_sel(void *bsp_priv, u8 phy_intf_sel) +{ + struct regmap *reg = bsp_priv; + + if (phy_intf_sel != PHY_INTF_SEL_GMII_MII && + phy_intf_sel != PHY_INTF_SEL_RMII) + return -EINVAL; + + regmap_update_bits(reg, LPC18XX_CREG_CREG6, + LPC18XX_CREG_CREG6_ETHMODE_MASK, + FIELD_PREP(LPC18XX_CREG_CREG6_ETHMODE_MASK, + phy_intf_sel)); + + return 0; +} + static int lpc18xx_dwmac_probe(struct platform_device *pdev) { struct plat_stmmacenet_data *plat_dat; struct stmmac_resources stmmac_res; - struct regmap *reg; - u8 ethmode; + struct regmap *regmap; int ret; ret = stmmac_get_platform_resources(pdev, &stmmac_res); @@ -41,21 +56,14 @@ static int lpc18xx_dwmac_probe(struct platform_device *pdev) plat_dat->core_type = DWMAC_CORE_GMAC; - reg = syscon_regmap_lookup_by_compatible("nxp,lpc1850-creg"); - if (IS_ERR(reg)) { + regmap = syscon_regmap_lookup_by_compatible("nxp,lpc1850-creg"); + if (IS_ERR(regmap)) { dev_err(&pdev->dev, "syscon lookup failed\n"); - return PTR_ERR(reg); + return PTR_ERR(regmap); } - ethmode = stmmac_get_phy_intf_sel(plat_dat->phy_interface); - if (ethmode != PHY_INTF_SEL_GMII_MII && - ethmode != PHY_INTF_SEL_RMII) { - dev_err(&pdev->dev, "Only MII and RMII mode supported\n"); - return -EINVAL; - } - - regmap_update_bits(reg, LPC18XX_CREG_CREG6, - LPC18XX_CREG_CREG6_ETHMODE_MASK, ethmode); + plat_dat->bsp_priv = regmap; + plat_dat->set_phy_intf_sel = lpc18xx_set_phy_intf_sel; return stmmac_dvr_probe(&pdev->dev, plat_dat, &stmmac_res); } From 9cd23c02ac57a4727095e1d1d62974e0bea445fb Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Thu, 6 Nov 2025 11:23:37 +0000 Subject: [PATCH 0624/1381] net: stmmac: sti: use PHY_INTF_SEL_x to select PHY interface Use the common dwmac definitions for the PHY interface selection field, adding MII_PHY_SEL_VAL() temporarily to avoid line wrapping. Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vGy5Z-0000000DhQV-1e2l@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/dwmac-sti.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-sti.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-sti.c index 53d5ce1f6dc6..1e8769a81d77 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-sti.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-sti.c @@ -77,13 +77,15 @@ * 001-RGMII * 010-SGMII * 100-RMII + * These are the DW MAC phy_intf_sel values */ #define MII_PHY_SEL_MASK GENMASK(4, 2) -#define ETH_PHY_SEL_RMII BIT(4) -#define ETH_PHY_SEL_SGMII BIT(3) -#define ETH_PHY_SEL_RGMII BIT(2) -#define ETH_PHY_SEL_GMII 0x0 -#define ETH_PHY_SEL_MII 0x0 +#define MII_PHY_SEL_VAL(val) FIELD_PREP_CONST(MII_PHY_SEL_MASK, val) +#define ETH_PHY_SEL_RMII MII_PHY_SEL_VAL(PHY_INTF_SEL_RMII) +#define ETH_PHY_SEL_SGMII MII_PHY_SEL_VAL(PHY_INTF_SEL_SGMII) +#define ETH_PHY_SEL_RGMII MII_PHY_SEL_VAL(PHY_INTF_SEL_RGMII) +#define ETH_PHY_SEL_GMII MII_PHY_SEL_VAL(PHY_INTF_SEL_GMII_MII) +#define ETH_PHY_SEL_MII MII_PHY_SEL_VAL(PHY_INTF_SEL_GMII_MII) struct sti_dwmac { phy_interface_t interface; /* MII interface */ From bd5a68159259efc910a5b199b3cd1a27d47205fb Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Thu, 6 Nov 2025 11:23:42 +0000 Subject: [PATCH 0625/1381] net: stmmac: sti: use PHY_INTF_SEL_x directly Use the PHY_INTF_SEL_x values directly rather than the driver private ETH_PHY_SEL_x values. Move the FIELD_PREP() into sti_dwmac_set_mode(). Use dwmac->interface directly. Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vGy5e-0000000DhQb-2B7I@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../net/ethernet/stmicro/stmmac/dwmac-sti.c | 27 +++++++++---------- 1 file changed, 12 insertions(+), 15 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-sti.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-sti.c index 1e8769a81d77..c97535824be0 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-sti.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-sti.c @@ -81,11 +81,6 @@ */ #define MII_PHY_SEL_MASK GENMASK(4, 2) #define MII_PHY_SEL_VAL(val) FIELD_PREP_CONST(MII_PHY_SEL_MASK, val) -#define ETH_PHY_SEL_RMII MII_PHY_SEL_VAL(PHY_INTF_SEL_RMII) -#define ETH_PHY_SEL_SGMII MII_PHY_SEL_VAL(PHY_INTF_SEL_SGMII) -#define ETH_PHY_SEL_RGMII MII_PHY_SEL_VAL(PHY_INTF_SEL_RGMII) -#define ETH_PHY_SEL_GMII MII_PHY_SEL_VAL(PHY_INTF_SEL_GMII_MII) -#define ETH_PHY_SEL_MII MII_PHY_SEL_VAL(PHY_INTF_SEL_GMII_MII) struct sti_dwmac { phy_interface_t interface; /* MII interface */ @@ -104,13 +99,13 @@ struct sti_dwmac_of_data { void (*fix_retime_src)(void *priv, int speed, unsigned int mode); }; -static u32 phy_intf_sels[] = { - [PHY_INTERFACE_MODE_MII] = ETH_PHY_SEL_MII, - [PHY_INTERFACE_MODE_GMII] = ETH_PHY_SEL_GMII, - [PHY_INTERFACE_MODE_RGMII] = ETH_PHY_SEL_RGMII, - [PHY_INTERFACE_MODE_RGMII_ID] = ETH_PHY_SEL_RGMII, - [PHY_INTERFACE_MODE_SGMII] = ETH_PHY_SEL_SGMII, - [PHY_INTERFACE_MODE_RMII] = ETH_PHY_SEL_RMII, +static u8 phy_intf_sels[] = { + [PHY_INTERFACE_MODE_MII] = PHY_INTF_SEL_GMII_MII, + [PHY_INTERFACE_MODE_GMII] = PHY_INTF_SEL_GMII_MII, + [PHY_INTERFACE_MODE_RGMII] = PHY_INTF_SEL_RGMII, + [PHY_INTERFACE_MODE_RGMII_ID] = PHY_INTF_SEL_RGMII, + [PHY_INTERFACE_MODE_SGMII] = PHY_INTF_SEL_SGMII, + [PHY_INTERFACE_MODE_RMII] = PHY_INTF_SEL_RMII, }; enum { @@ -164,16 +159,18 @@ static void stih4xx_fix_retime_src(void *priv, int spd, unsigned int mode) static int sti_dwmac_set_mode(struct sti_dwmac *dwmac) { struct regmap *regmap = dwmac->regmap; - int iface = dwmac->interface; u32 reg = dwmac->ctrl_reg; + u8 phy_intf_sel; u32 val; if (dwmac->gmac_en) regmap_update_bits(regmap, reg, EN_MASK, EN); - regmap_update_bits(regmap, reg, MII_PHY_SEL_MASK, phy_intf_sels[iface]); + phy_intf_sel = phy_intf_sels[dwmac->interface]; + regmap_update_bits(regmap, reg, MII_PHY_SEL_MASK, + FIELD_PREP(MII_PHY_SEL_MASK, phy_intf_sel)); - val = (iface == PHY_INTERFACE_MODE_REVMII) ? 0 : ENMII; + val = (dwmac->interface == PHY_INTERFACE_MODE_REVMII) ? 0 : ENMII; regmap_update_bits(regmap, reg, ENMII_MASK, val); dwmac->fix_retime_src(dwmac, dwmac->speed, 0); From ef5e870be90ffc390ae9afc0b50d411d3538126c Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Thu, 6 Nov 2025 11:23:47 +0000 Subject: [PATCH 0626/1381] net: stmmac: sti: use stmmac_get_phy_intf_sel() Use stmmac_get_phy_intf_sel() to decode the PHY interface mode to the phy_intf_sel value, validate the result and use that to set the control register to select the operating mode for the DWMAC core. Note that when an unsupported interface mode is used, the array would decode this to PHY_INTF_SEL_GMII_MII, so preserve this behaviour. Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vGy5j-0000000DhQh-2e0x@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../net/ethernet/stmicro/stmmac/dwmac-sti.c | 22 ++++++++----------- 1 file changed, 9 insertions(+), 13 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-sti.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-sti.c index c97535824be0..593e154b5957 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-sti.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-sti.c @@ -77,10 +77,9 @@ * 001-RGMII * 010-SGMII * 100-RMII - * These are the DW MAC phy_intf_sel values + * These are the DW MAC phy_intf_sel values. */ #define MII_PHY_SEL_MASK GENMASK(4, 2) -#define MII_PHY_SEL_VAL(val) FIELD_PREP_CONST(MII_PHY_SEL_MASK, val) struct sti_dwmac { phy_interface_t interface; /* MII interface */ @@ -99,15 +98,6 @@ struct sti_dwmac_of_data { void (*fix_retime_src)(void *priv, int speed, unsigned int mode); }; -static u8 phy_intf_sels[] = { - [PHY_INTERFACE_MODE_MII] = PHY_INTF_SEL_GMII_MII, - [PHY_INTERFACE_MODE_GMII] = PHY_INTF_SEL_GMII_MII, - [PHY_INTERFACE_MODE_RGMII] = PHY_INTF_SEL_RGMII, - [PHY_INTERFACE_MODE_RGMII_ID] = PHY_INTF_SEL_RGMII, - [PHY_INTERFACE_MODE_SGMII] = PHY_INTF_SEL_SGMII, - [PHY_INTERFACE_MODE_RMII] = PHY_INTF_SEL_RMII, -}; - enum { TX_RETIME_SRC_NA = 0, TX_RETIME_SRC_TXCLK = 1, @@ -160,13 +150,19 @@ static int sti_dwmac_set_mode(struct sti_dwmac *dwmac) { struct regmap *regmap = dwmac->regmap; u32 reg = dwmac->ctrl_reg; - u8 phy_intf_sel; + int phy_intf_sel; u32 val; if (dwmac->gmac_en) regmap_update_bits(regmap, reg, EN_MASK, EN); - phy_intf_sel = phy_intf_sels[dwmac->interface]; + phy_intf_sel = stmmac_get_phy_intf_sel(dwmac->interface); + if (phy_intf_sel != PHY_INTF_SEL_GMII_MII && + phy_intf_sel != PHY_INTF_SEL_RGMII && + phy_intf_sel != PHY_INTF_SEL_SGMII && + phy_intf_sel != PHY_INTF_SEL_RMII) + phy_intf_sel = PHY_INTF_SEL_GMII_MII; + regmap_update_bits(regmap, reg, MII_PHY_SEL_MASK, FIELD_PREP(MII_PHY_SEL_MASK, phy_intf_sel)); From e3c8f25cf2aa3f2547da26e7a46424c016550d31 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Thu, 6 Nov 2025 11:23:52 +0000 Subject: [PATCH 0627/1381] net: stmmac: sti: use ->set_phy_intf_sel() Rather than placing the phy_intf_sel() setup in the ->init() method, move it to the new ->set_phy_intf_sel() method. Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vGy5o-0000000DhQn-34JE@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../net/ethernet/stmicro/stmmac/dwmac-sti.c | 25 +++++++------------ 1 file changed, 9 insertions(+), 16 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-sti.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-sti.c index 593e154b5957..b0509ab6b31c 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-sti.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-sti.c @@ -146,17 +146,18 @@ static void stih4xx_fix_retime_src(void *priv, int spd, unsigned int mode) stih4xx_tx_retime_val[src]); } -static int sti_dwmac_set_mode(struct sti_dwmac *dwmac) +static int sti_set_phy_intf_sel(void *bsp_priv, u8 phy_intf_sel) { - struct regmap *regmap = dwmac->regmap; - u32 reg = dwmac->ctrl_reg; - int phy_intf_sel; - u32 val; + struct sti_dwmac *dwmac = bsp_priv; + struct regmap *regmap; + u32 reg, val; + + regmap = dwmac->regmap; + reg = dwmac->ctrl_reg; if (dwmac->gmac_en) regmap_update_bits(regmap, reg, EN_MASK, EN); - phy_intf_sel = stmmac_get_phy_intf_sel(dwmac->interface); if (phy_intf_sel != PHY_INTF_SEL_GMII_MII && phy_intf_sel != PHY_INTF_SEL_RGMII && phy_intf_sel != PHY_INTF_SEL_SGMII && @@ -231,17 +232,8 @@ static int sti_dwmac_parse_data(struct sti_dwmac *dwmac, static int sti_dwmac_init(struct platform_device *pdev, void *bsp_priv) { struct sti_dwmac *dwmac = bsp_priv; - int ret; - ret = clk_prepare_enable(dwmac->clk); - if (ret) - return ret; - - ret = sti_dwmac_set_mode(dwmac); - if (ret) - clk_disable_unprepare(dwmac->clk); - - return ret; + return clk_prepare_enable(dwmac->clk); } static void sti_dwmac_exit(struct platform_device *pdev, void *bsp_priv) @@ -286,6 +278,7 @@ static int sti_dwmac_probe(struct platform_device *pdev) dwmac->fix_retime_src = data->fix_retime_src; plat_dat->bsp_priv = dwmac; + plat_dat->set_phy_intf_sel = sti_set_phy_intf_sel; plat_dat->fix_mac_speed = data->fix_retime_src; plat_dat->init = sti_dwmac_init; plat_dat->exit = sti_dwmac_exit; From 7b8a8ec20cfce2298f6737089f5d17407ea346b4 Mon Sep 17 00:00:00 2001 From: Yishai Hadas Date: Mon, 27 Oct 2025 11:34:01 +0200 Subject: [PATCH 0628/1381] PCI/TPH: Expose pcie_tph_get_st_table_loc() Expose pcie_tph_get_st_table_loc() to be used by drivers as will be done in the next patch from the series. Signed-off-by: Yishai Hadas Signed-off-by: Edward Srouji Link: https://patch.msgid.link/20251027-st-direct-mode-v1-1-e0ad953866b6@nvidia.com Acked-by: Bjorn Helgaas Signed-off-by: Leon Romanovsky --- drivers/pci/tph.c | 16 +++++++++++++--- include/linux/pci-tph.h | 1 + 2 files changed, 14 insertions(+), 3 deletions(-) diff --git a/drivers/pci/tph.c b/drivers/pci/tph.c index cc64f93709a4..ca4f97be7538 100644 --- a/drivers/pci/tph.c +++ b/drivers/pci/tph.c @@ -155,7 +155,16 @@ static u8 get_st_modes(struct pci_dev *pdev) return reg; } -static u32 get_st_table_loc(struct pci_dev *pdev) +/** + * pcie_tph_get_st_table_loc - Return the device's ST table location + * @pdev: PCI device to query + * + * Return: + * PCI_TPH_LOC_NONE - Not present + * PCI_TPH_LOC_CAP - Located in the TPH Requester Extended Capability + * PCI_TPH_LOC_MSIX - Located in the MSI-X Table + */ +u32 pcie_tph_get_st_table_loc(struct pci_dev *pdev) { u32 reg; @@ -163,6 +172,7 @@ static u32 get_st_table_loc(struct pci_dev *pdev) return FIELD_GET(PCI_TPH_CAP_LOC_MASK, reg); } +EXPORT_SYMBOL(pcie_tph_get_st_table_loc); /* * Return the size of ST table. If ST table is not in TPH Requester Extended @@ -174,7 +184,7 @@ u16 pcie_tph_get_st_table_size(struct pci_dev *pdev) u32 loc; /* Check ST table location first */ - loc = get_st_table_loc(pdev); + loc = pcie_tph_get_st_table_loc(pdev); /* Convert loc to match with PCI_TPH_LOC_* defined in pci_regs.h */ loc = FIELD_PREP(PCI_TPH_CAP_LOC_MASK, loc); @@ -299,7 +309,7 @@ int pcie_tph_set_st_entry(struct pci_dev *pdev, unsigned int index, u16 tag) */ set_ctrl_reg_req_en(pdev, PCI_TPH_REQ_DISABLE); - loc = get_st_table_loc(pdev); + loc = pcie_tph_get_st_table_loc(pdev); /* Convert loc to match with PCI_TPH_LOC_* */ loc = FIELD_PREP(PCI_TPH_CAP_LOC_MASK, loc); diff --git a/include/linux/pci-tph.h b/include/linux/pci-tph.h index 9e4e331b1603..ba28140ce670 100644 --- a/include/linux/pci-tph.h +++ b/include/linux/pci-tph.h @@ -29,6 +29,7 @@ int pcie_tph_get_cpu_st(struct pci_dev *dev, void pcie_disable_tph(struct pci_dev *pdev); int pcie_enable_tph(struct pci_dev *pdev, int mode); u16 pcie_tph_get_st_table_size(struct pci_dev *pdev); +u32 pcie_tph_get_st_table_loc(struct pci_dev *pdev); #else static inline int pcie_tph_set_st_entry(struct pci_dev *pdev, unsigned int index, u16 tag) From 2d838c11e10e9169cae4f7778345c11b5447ef05 Mon Sep 17 00:00:00 2001 From: Yishai Hadas Date: Mon, 27 Oct 2025 11:34:02 +0200 Subject: [PATCH 0629/1381] net/mlx5: Add direct ST mode support for RDMA Add support for direct ST mode where ST Table Location equals PCI_TPH_LOC_NONE. In that case, no steering table exists, the steering tag itself will be used directly by the SW, FW, HW from the mkey. This enables RDMA users to use the current exposed APIs to work in direct mode. Signed-off-by: Yishai Hadas Signed-off-by: Edward Srouji Link: https://patch.msgid.link/20251027-st-direct-mode-v1-2-e0ad953866b6@nvidia.com Signed-off-by: Leon Romanovsky --- .../net/ethernet/mellanox/mlx5/core/lib/st.c | 29 ++++++++++++++++--- 1 file changed, 25 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/st.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/st.c index 47fe215f66bf..ef06fe6cbb51 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lib/st.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/st.c @@ -19,13 +19,16 @@ struct mlx5_st { struct mutex lock; struct xa_limit index_limit; struct xarray idx_xa; /* key == index, value == struct mlx5_st_idx_data */ + u8 direct_mode : 1; }; struct mlx5_st *mlx5_st_create(struct mlx5_core_dev *dev) { struct pci_dev *pdev = dev->pdev; struct mlx5_st *st; + u8 direct_mode = 0; u16 num_entries; + u32 tbl_loc; int ret; if (!MLX5_CAP_GEN(dev, mkey_pcie_tph)) @@ -40,10 +43,16 @@ struct mlx5_st *mlx5_st_create(struct mlx5_core_dev *dev) if (!pdev->tph_cap) return NULL; - num_entries = pcie_tph_get_st_table_size(pdev); - /* We need a reserved entry for non TPH cases */ - if (num_entries < 2) - return NULL; + tbl_loc = pcie_tph_get_st_table_loc(pdev); + if (tbl_loc == PCI_TPH_LOC_NONE) + direct_mode = 1; + + if (!direct_mode) { + num_entries = pcie_tph_get_st_table_size(pdev); + /* We need a reserved entry for non TPH cases */ + if (num_entries < 2) + return NULL; + } /* The OS doesn't support ST */ ret = pcie_enable_tph(pdev, PCI_TPH_ST_DS_MODE); @@ -56,6 +65,10 @@ struct mlx5_st *mlx5_st_create(struct mlx5_core_dev *dev) mutex_init(&st->lock); xa_init_flags(&st->idx_xa, XA_FLAGS_ALLOC); + st->direct_mode = direct_mode; + if (st->direct_mode) + return st; + /* entry 0 is reserved for non TPH cases */ st->index_limit.min = MLX5_MKC_PCIE_TPH_NO_STEERING_TAG_INDEX + 1; st->index_limit.max = num_entries - 1; @@ -96,6 +109,11 @@ int mlx5_st_alloc_index(struct mlx5_core_dev *dev, enum tph_mem_type mem_type, if (ret) return ret; + if (st->direct_mode) { + *st_index = tag; + return 0; + } + mutex_lock(&st->lock); xa_for_each(&st->idx_xa, index, idx_data) { @@ -145,6 +163,9 @@ int mlx5_st_dealloc_index(struct mlx5_core_dev *dev, u16 st_index) if (!st) return -EOPNOTSUPP; + if (st->direct_mode) + return 0; + mutex_lock(&st->lock); idx_data = xa_load(&st->idx_xa, st_index); if (WARN_ON_ONCE(!idx_data)) { From 6948417b3f1fafbeab85c051f8dba5e305a8f9c4 Mon Sep 17 00:00:00 2001 From: Patrisious Haddad Date: Wed, 29 Oct 2025 17:42:53 +0200 Subject: [PATCH 0630/1381] net/mlx5: Add OTHER_ESWITCH HW capabilities Add OTHER_ESWITCH capabilities which includes other_eswitch and eswitch_owner_vhca_id to all steering objects. Signed-off-by: Patrisious Haddad Signed-off-by: Edward Srouji Link: https://patch.msgid.link/20251029-support-other-eswitch-v1-1-98bb707b5d57@nvidia.com Signed-off-by: Leon Romanovsky --- include/linux/mlx5/mlx5_ifc.h | 47 +++++++++++++++++++++++------------ 1 file changed, 31 insertions(+), 16 deletions(-) diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h index 07614cd95bed..9b8f88987d2f 100644 --- a/include/linux/mlx5/mlx5_ifc.h +++ b/include/linux/mlx5/mlx5_ifc.h @@ -5251,13 +5251,15 @@ struct mlx5_ifc_set_fte_in_bits { u8 op_mod[0x10]; u8 other_vport[0x1]; - u8 reserved_at_41[0xf]; + u8 other_eswitch[0x1]; + u8 reserved_at_42[0xe]; u8 vport_number[0x10]; u8 reserved_at_60[0x20]; u8 table_type[0x8]; - u8 reserved_at_88[0x18]; + u8 reserved_at_88[0x8]; + u8 eswitch_owner_vhca_id[0x10]; u8 reserved_at_a0[0x8]; u8 table_id[0x18]; @@ -8809,13 +8811,15 @@ struct mlx5_ifc_destroy_flow_table_in_bits { u8 op_mod[0x10]; u8 other_vport[0x1]; - u8 reserved_at_41[0xf]; + u8 other_eswitch[0x1]; + u8 reserved_at_42[0xe]; u8 vport_number[0x10]; u8 reserved_at_60[0x20]; u8 table_type[0x8]; - u8 reserved_at_88[0x18]; + u8 reserved_at_88[0x8]; + u8 eswitch_owner_vhca_id[0x10]; u8 reserved_at_a0[0x8]; u8 table_id[0x18]; @@ -8840,13 +8844,15 @@ struct mlx5_ifc_destroy_flow_group_in_bits { u8 op_mod[0x10]; u8 other_vport[0x1]; - u8 reserved_at_41[0xf]; + u8 other_eswitch[0x1]; + u8 reserved_at_42[0xe]; u8 vport_number[0x10]; u8 reserved_at_60[0x20]; u8 table_type[0x8]; - u8 reserved_at_88[0x18]; + u8 reserved_at_88[0x8]; + u8 eswitch_owner_vhca_id[0x10]; u8 reserved_at_a0[0x8]; u8 table_id[0x18]; @@ -8985,13 +8991,15 @@ struct mlx5_ifc_delete_fte_in_bits { u8 op_mod[0x10]; u8 other_vport[0x1]; - u8 reserved_at_41[0xf]; + u8 other_eswitch[0x1]; + u8 reserved_at_42[0xe]; u8 vport_number[0x10]; u8 reserved_at_60[0x20]; u8 table_type[0x8]; - u8 reserved_at_88[0x18]; + u8 reserved_at_88[0x8]; + u8 eswitch_owner_vhca_id[0x10]; u8 reserved_at_a0[0x8]; u8 table_id[0x18]; @@ -9535,13 +9543,15 @@ struct mlx5_ifc_create_flow_table_in_bits { u8 op_mod[0x10]; u8 other_vport[0x1]; - u8 reserved_at_41[0xf]; + u8 other_eswitch[0x1]; + u8 reserved_at_42[0xe]; u8 vport_number[0x10]; u8 reserved_at_60[0x20]; u8 table_type[0x8]; - u8 reserved_at_88[0x18]; + u8 reserved_at_88[0x8]; + u8 eswitch_owner_vhca_id[0x10]; u8 reserved_at_a0[0x20]; @@ -9580,7 +9590,8 @@ struct mlx5_ifc_create_flow_group_in_bits { u8 op_mod[0x10]; u8 other_vport[0x1]; - u8 reserved_at_41[0xf]; + u8 other_eswitch[0x1]; + u8 reserved_at_42[0xe]; u8 vport_number[0x10]; u8 reserved_at_60[0x20]; @@ -9588,7 +9599,7 @@ struct mlx5_ifc_create_flow_group_in_bits { u8 table_type[0x8]; u8 reserved_at_88[0x4]; u8 group_type[0x4]; - u8 reserved_at_90[0x10]; + u8 eswitch_owner_vhca_id[0x10]; u8 reserved_at_a0[0x8]; u8 table_id[0x18]; @@ -11876,10 +11887,12 @@ struct mlx5_ifc_set_flow_table_root_in_bits { u8 op_mod[0x10]; u8 other_vport[0x1]; - u8 reserved_at_41[0xf]; + u8 other_eswitch[0x1]; + u8 reserved_at_42[0xe]; u8 vport_number[0x10]; - u8 reserved_at_60[0x20]; + u8 reserved_at_60[0x10]; + u8 eswitch_owner_vhca_id[0x10]; u8 table_type[0x8]; u8 reserved_at_88[0x7]; @@ -11919,14 +11932,16 @@ struct mlx5_ifc_modify_flow_table_in_bits { u8 op_mod[0x10]; u8 other_vport[0x1]; - u8 reserved_at_41[0xf]; + u8 other_eswitch[0x1]; + u8 reserved_at_42[0xe]; u8 vport_number[0x10]; u8 reserved_at_60[0x10]; u8 modify_field_select[0x10]; u8 table_type[0x8]; - u8 reserved_at_88[0x18]; + u8 reserved_at_88[0x8]; + u8 eswitch_owner_vhca_id[0x10]; u8 reserved_at_a0[0x8]; u8 table_id[0x18]; From 3b848dec7e821bace785b9e405bf1884c077635a Mon Sep 17 00:00:00 2001 From: Patrisious Haddad Date: Wed, 29 Oct 2025 17:42:54 +0200 Subject: [PATCH 0631/1381] net/mlx5: fs, Add other_eswitch support for steering tables Add other_eswitch support which allows flow tables creation above vports that reside on different esw managers. The new flag MLX5_FLOW_TABLE_OTHER_ESWITCH indicates if the esw_owner_vhca_id attribute is supported. Note that this is only supported if the Advanced-RDMA cap- rdma_transport_manager_other_eswitch is set. And it is the caller responsibility to check that. Signed-off-by: Patrisious Haddad Signed-off-by: Edward Srouji Link: https://patch.msgid.link/20251029-support-other-eswitch-v1-2-98bb707b5d57@nvidia.com Signed-off-by: Leon Romanovsky --- .../net/ethernet/mellanox/mlx5/core/fs_cmd.c | 31 +++++++++++++++++++ .../net/ethernet/mellanox/mlx5/core/fs_core.c | 18 +++++------ .../net/ethernet/mellanox/mlx5/core/fs_core.h | 1 + include/linux/mlx5/fs.h | 2 ++ 4 files changed, 42 insertions(+), 10 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c index 1af76da8b132..ced747bef641 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c @@ -239,6 +239,10 @@ static int mlx5_cmd_update_root_ft(struct mlx5_flow_root_namespace *ns, MLX5_SET(set_flow_table_root_in, in, vport_number, ft->vport); MLX5_SET(set_flow_table_root_in, in, other_vport, !!(ft->flags & MLX5_FLOW_TABLE_OTHER_VPORT)); + MLX5_SET(set_flow_table_root_in, in, eswitch_owner_vhca_id, + ft->esw_owner_vhca_id); + MLX5_SET(set_flow_table_root_in, in, other_eswitch, + !!(ft->flags & MLX5_FLOW_TABLE_OTHER_ESWITCH)); err = mlx5_cmd_exec_in(dev, set_flow_table_root, in); if (!err && @@ -302,6 +306,10 @@ static int mlx5_cmd_create_flow_table(struct mlx5_flow_root_namespace *ns, MLX5_SET(create_flow_table_in, in, vport_number, ft->vport); MLX5_SET(create_flow_table_in, in, other_vport, !!(ft->flags & MLX5_FLOW_TABLE_OTHER_VPORT)); + MLX5_SET(create_flow_table_in, in, eswitch_owner_vhca_id, + ft->esw_owner_vhca_id); + MLX5_SET(create_flow_table_in, in, other_eswitch, + !!(ft->flags & MLX5_FLOW_TABLE_OTHER_ESWITCH)); MLX5_SET(create_flow_table_in, in, flow_table_context.decap_en, en_decap); @@ -360,6 +368,10 @@ static int mlx5_cmd_destroy_flow_table(struct mlx5_flow_root_namespace *ns, MLX5_SET(destroy_flow_table_in, in, vport_number, ft->vport); MLX5_SET(destroy_flow_table_in, in, other_vport, !!(ft->flags & MLX5_FLOW_TABLE_OTHER_VPORT)); + MLX5_SET(destroy_flow_table_in, in, eswitch_owner_vhca_id, + ft->esw_owner_vhca_id); + MLX5_SET(destroy_flow_table_in, in, other_eswitch, + !!(ft->flags & MLX5_FLOW_TABLE_OTHER_ESWITCH)); err = mlx5_cmd_exec_in(dev, destroy_flow_table, in); if (!err) @@ -394,6 +406,10 @@ static int mlx5_cmd_modify_flow_table(struct mlx5_flow_root_namespace *ns, MLX5_SET(modify_flow_table_in, in, vport_number, ft->vport); MLX5_SET(modify_flow_table_in, in, other_vport, !!(ft->flags & MLX5_FLOW_TABLE_OTHER_VPORT)); + MLX5_SET(modify_flow_table_in, in, eswitch_owner_vhca_id, + ft->esw_owner_vhca_id); + MLX5_SET(modify_flow_table_in, in, other_eswitch, + !!(ft->flags & MLX5_FLOW_TABLE_OTHER_ESWITCH)); MLX5_SET(modify_flow_table_in, in, modify_field_select, MLX5_MODIFY_FLOW_TABLE_MISS_TABLE_ID); if (next_ft) { @@ -429,6 +445,10 @@ static int mlx5_cmd_create_flow_group(struct mlx5_flow_root_namespace *ns, MLX5_SET(create_flow_group_in, in, vport_number, ft->vport); MLX5_SET(create_flow_group_in, in, other_vport, !!(ft->flags & MLX5_FLOW_TABLE_OTHER_VPORT)); + MLX5_SET(create_flow_group_in, in, eswitch_owner_vhca_id, + ft->esw_owner_vhca_id); + MLX5_SET(create_flow_group_in, in, other_eswitch, + !!(ft->flags & MLX5_FLOW_TABLE_OTHER_ESWITCH)); err = mlx5_cmd_exec_inout(dev, create_flow_group, in, out); if (!err) fg->id = MLX5_GET(create_flow_group_out, out, @@ -451,6 +471,10 @@ static int mlx5_cmd_destroy_flow_group(struct mlx5_flow_root_namespace *ns, MLX5_SET(destroy_flow_group_in, in, vport_number, ft->vport); MLX5_SET(destroy_flow_group_in, in, other_vport, !!(ft->flags & MLX5_FLOW_TABLE_OTHER_VPORT)); + MLX5_SET(destroy_flow_group_in, in, eswitch_owner_vhca_id, + ft->esw_owner_vhca_id); + MLX5_SET(destroy_flow_group_in, in, other_eswitch, + !!(ft->flags & MLX5_FLOW_TABLE_OTHER_ESWITCH)); return mlx5_cmd_exec_in(dev, destroy_flow_group, in); } @@ -559,6 +583,9 @@ static int mlx5_cmd_set_fte(struct mlx5_core_dev *dev, MLX5_SET(set_fte_in, in, vport_number, ft->vport); MLX5_SET(set_fte_in, in, other_vport, !!(ft->flags & MLX5_FLOW_TABLE_OTHER_VPORT)); + MLX5_SET(set_fte_in, in, eswitch_owner_vhca_id, ft->esw_owner_vhca_id); + MLX5_SET(set_fte_in, in, other_eswitch, + !!(ft->flags & MLX5_FLOW_TABLE_OTHER_ESWITCH)); in_flow_context = MLX5_ADDR_OF(set_fte_in, in, flow_context); MLX5_SET(flow_context, in_flow_context, group_id, group_id); @@ -788,6 +815,10 @@ static int mlx5_cmd_delete_fte(struct mlx5_flow_root_namespace *ns, MLX5_SET(delete_fte_in, in, vport_number, ft->vport); MLX5_SET(delete_fte_in, in, other_vport, !!(ft->flags & MLX5_FLOW_TABLE_OTHER_VPORT)); + MLX5_SET(delete_fte_in, in, eswitch_owner_vhca_id, + ft->esw_owner_vhca_id); + MLX5_SET(delete_fte_in, in, other_eswitch, + !!(ft->flags & MLX5_FLOW_TABLE_OTHER_ESWITCH)); return mlx5_cmd_exec_in(dev, delete_fte, in); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c index 2db3ffb0a2b2..87e381c82ed3 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c @@ -939,10 +939,10 @@ static struct mlx5_flow_group *alloc_insert_flow_group(struct mlx5_flow_table *f return fg; } -static struct mlx5_flow_table *alloc_flow_table(int level, u16 vport, - enum fs_flow_table_type table_type, - enum fs_flow_table_op_mod op_mod, - u32 flags) +static struct mlx5_flow_table * +alloc_flow_table(struct mlx5_flow_table_attr *ft_attr, u16 vport, + enum fs_flow_table_type table_type, + enum fs_flow_table_op_mod op_mod) { struct mlx5_flow_table *ft; int ret; @@ -957,12 +957,13 @@ static struct mlx5_flow_table *alloc_flow_table(int level, u16 vport, return ERR_PTR(ret); } - ft->level = level; + ft->level = ft_attr->level; ft->node.type = FS_TYPE_FLOW_TABLE; ft->op_mod = op_mod; ft->type = table_type; ft->vport = vport; - ft->flags = flags; + ft->esw_owner_vhca_id = ft_attr->esw_owner_vhca_id; + ft->flags = ft_attr->flags; INIT_LIST_HEAD(&ft->fwd_rules); mutex_init(&ft->lock); @@ -1370,10 +1371,7 @@ static struct mlx5_flow_table *__mlx5_create_flow_table(struct mlx5_flow_namespa /* The level is related to the * priority level range. */ - ft = alloc_flow_table(ft_attr->level, - vport, - root->table_type, - op_mod, ft_attr->flags); + ft = alloc_flow_table(ft_attr, vport, root->table_type, op_mod); if (IS_ERR(ft)) { err = PTR_ERR(ft); goto unlock_root; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h index 8458ce203dac..0a9a5ef34c21 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h @@ -205,6 +205,7 @@ struct mlx5_flow_table { }; u32 id; u16 vport; + u16 esw_owner_vhca_id; unsigned int max_fte; unsigned int level; enum fs_flow_table_type type; diff --git a/include/linux/mlx5/fs.h b/include/linux/mlx5/fs.h index 6ac76a0c3827..6325a7fa0df2 100644 --- a/include/linux/mlx5/fs.h +++ b/include/linux/mlx5/fs.h @@ -71,6 +71,7 @@ enum { MLX5_FLOW_TABLE_UNMANAGED = BIT(3), MLX5_FLOW_TABLE_OTHER_VPORT = BIT(4), MLX5_FLOW_TABLE_UPLINK_VPORT = BIT(5), + MLX5_FLOW_TABLE_OTHER_ESWITCH = BIT(6), }; #define LEFTOVERS_RULE_NUM 2 @@ -208,6 +209,7 @@ struct mlx5_flow_table_attr { u32 flags; u16 uid; u16 vport; + u16 esw_owner_vhca_id; struct mlx5_flow_table *next_ft; struct { From 583b4fe1c19d978bb787e0adf9ce469cb7f68455 Mon Sep 17 00:00:00 2001 From: Patrisious Haddad Date: Wed, 29 Oct 2025 17:42:55 +0200 Subject: [PATCH 0632/1381] net/mlx5: fs, set non default device per namespace Add mlx5_fs_set_root_dev() function which swaps the root namespace core device with another one for a given table_type. It is intended for usage only by RDMA_TRANSPORT tables in case of LAG configuration, to allow the creation of tables during LAG always through the LAG master device, which is valid since during LAG the master is allowed to manage the RDMA_TRANSPORT tables of its slaves. In addition move the table_type enum to global include to allow its use in a downstream patch in the RDMA driver. Signed-off-by: Patrisious Haddad Signed-off-by: Edward Srouji Link: https://patch.msgid.link/20251029-support-other-eswitch-v1-3-98bb707b5d57@nvidia.com Signed-off-by: Leon Romanovsky --- .../net/ethernet/mellanox/mlx5/core/fs_core.c | 56 +++++++++++++++++++ .../net/ethernet/mellanox/mlx5/core/fs_core.h | 18 ------ include/linux/mlx5/fs.h | 22 ++++++++ 3 files changed, 78 insertions(+), 18 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c index 87e381c82ed3..5b210c54a592 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c @@ -3308,6 +3308,62 @@ init_rdma_transport_tx_root_ns_one(struct mlx5_flow_steering *steering, return ret; } +static bool mlx5_fs_ns_is_empty(struct mlx5_flow_namespace *ns) +{ + struct fs_prio *iter_prio; + + fs_for_each_prio(iter_prio, ns) { + if (iter_prio->num_ft) + return false; + } + + return true; +} + +int mlx5_fs_set_root_dev(struct mlx5_core_dev *dev, + struct mlx5_core_dev *new_dev, + enum fs_flow_table_type table_type) +{ + struct mlx5_flow_root_namespace **root; + int total_vports; + int i; + + switch (table_type) { + case FS_FT_RDMA_TRANSPORT_TX: + root = dev->priv.steering->rdma_transport_tx_root_ns; + total_vports = dev->priv.steering->rdma_transport_tx_vports; + break; + case FS_FT_RDMA_TRANSPORT_RX: + root = dev->priv.steering->rdma_transport_rx_root_ns; + total_vports = dev->priv.steering->rdma_transport_rx_vports; + break; + default: + WARN_ON_ONCE(true); + return -EINVAL; + } + + for (i = 0; i < total_vports; i++) { + mutex_lock(&root[i]->chain_lock); + if (!mlx5_fs_ns_is_empty(&root[i]->ns)) { + mutex_unlock(&root[i]->chain_lock); + goto err; + } + root[i]->dev = new_dev; + mutex_unlock(&root[i]->chain_lock); + } + return 0; +err: + while (i--) { + mutex_lock(&root[i]->chain_lock); + root[i]->dev = dev; + mutex_unlock(&root[i]->chain_lock); + } + /* If you hit this error try destroying all flow tables and try again */ + mlx5_core_err(dev, "Failed to set root device for RDMA TRANSPORT\n"); + return -EINVAL; +} +EXPORT_SYMBOL(mlx5_fs_set_root_dev); + static int init_rdma_transport_rx_root_ns(struct mlx5_flow_steering *steering) { struct mlx5_core_dev *dev = steering->dev; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h index 0a9a5ef34c21..1c6591425260 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h @@ -103,24 +103,6 @@ enum fs_node_type { FS_TYPE_FLOW_DEST }; -enum fs_flow_table_type { - FS_FT_NIC_RX = 0x0, - FS_FT_NIC_TX = 0x1, - FS_FT_ESW_EGRESS_ACL = 0x2, - FS_FT_ESW_INGRESS_ACL = 0x3, - FS_FT_FDB = 0X4, - FS_FT_SNIFFER_RX = 0X5, - FS_FT_SNIFFER_TX = 0X6, - FS_FT_RDMA_RX = 0X7, - FS_FT_RDMA_TX = 0X8, - FS_FT_PORT_SEL = 0X9, - FS_FT_FDB_RX = 0xa, - FS_FT_FDB_TX = 0xb, - FS_FT_RDMA_TRANSPORT_RX = 0xd, - FS_FT_RDMA_TRANSPORT_TX = 0xe, - FS_FT_MAX_TYPE = FS_FT_RDMA_TRANSPORT_TX, -}; - enum fs_flow_table_op_mod { FS_FT_OP_MOD_NORMAL, FS_FT_OP_MOD_LAG_DEMUX, diff --git a/include/linux/mlx5/fs.h b/include/linux/mlx5/fs.h index 6325a7fa0df2..fe721557bd1d 100644 --- a/include/linux/mlx5/fs.h +++ b/include/linux/mlx5/fs.h @@ -128,6 +128,24 @@ enum { FDB_PER_VPORT, }; +enum fs_flow_table_type { + FS_FT_NIC_RX = 0x0, + FS_FT_NIC_TX = 0x1, + FS_FT_ESW_EGRESS_ACL = 0x2, + FS_FT_ESW_INGRESS_ACL = 0x3, + FS_FT_FDB = 0X4, + FS_FT_SNIFFER_RX = 0X5, + FS_FT_SNIFFER_TX = 0X6, + FS_FT_RDMA_RX = 0X7, + FS_FT_RDMA_TX = 0X8, + FS_FT_PORT_SEL = 0X9, + FS_FT_FDB_RX = 0xa, + FS_FT_FDB_TX = 0xb, + FS_FT_RDMA_TRANSPORT_RX = 0xd, + FS_FT_RDMA_TRANSPORT_TX = 0xe, + FS_FT_MAX_TYPE = FS_FT_RDMA_TRANSPORT_TX, +}; + struct mlx5_pkt_reformat; struct mlx5_modify_hdr; struct mlx5_flow_definer; @@ -355,4 +373,8 @@ u32 mlx5_flow_table_id(struct mlx5_flow_table *ft); struct mlx5_flow_root_namespace * mlx5_get_root_namespace(struct mlx5_core_dev *dev, enum mlx5_flow_namespace_type ns_type); + +int mlx5_fs_set_root_dev(struct mlx5_core_dev *dev, + struct mlx5_core_dev *new_dev, + enum fs_flow_table_type table_type); #endif From 69674282fc97fffd98a85ab5b4837edbc5898145 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 5 Nov 2025 15:36:49 +0100 Subject: [PATCH 0633/1381] wifi: ieee80211: split mesh definitions out The ieee80211.h file has gotten very long, start splitting it by putting mesh definitions into a separate file. Link: https://patch.msgid.link/20251105153843.489713ca8b34.I3befb4bf6ace0315758a1794224ddd18c4652e32@changeid Signed-off-by: Johannes Berg --- include/linux/ieee80211-mesh.h | 230 +++++++++++++++++++++++++++++++++ include/linux/ieee80211.h | 211 +----------------------------- 2 files changed, 232 insertions(+), 209 deletions(-) create mode 100644 include/linux/ieee80211-mesh.h diff --git a/include/linux/ieee80211-mesh.h b/include/linux/ieee80211-mesh.h new file mode 100644 index 000000000000..4b829bcb38b6 --- /dev/null +++ b/include/linux/ieee80211-mesh.h @@ -0,0 +1,230 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * IEEE 802.11 mesh definitions + * + * Copyright (c) 2001-2002, SSH Communications Security Corp and Jouni Malinen + * + * Copyright (c) 2002-2003, Jouni Malinen + * Copyright (c) 2005, Devicescape Software, Inc. + * Copyright (c) 2006, Michael Wu + * Copyright (c) 2013 - 2014 Intel Mobile Communications GmbH + * Copyright (c) 2016 - 2017 Intel Deutschland GmbH + * Copyright (c) 2018 - 2025 Intel Corporation + */ + +#ifndef LINUX_IEEE80211_MESH_H +#define LINUX_IEEE80211_MESH_H + +#include +#include + +#define IEEE80211_MAX_MESH_ID_LEN 32 + +struct ieee80211s_hdr { + u8 flags; + u8 ttl; + __le32 seqnum; + u8 eaddr1[ETH_ALEN]; + u8 eaddr2[ETH_ALEN]; +} __packed __aligned(2); + +/* Mesh flags */ +#define MESH_FLAGS_AE_A4 0x1 +#define MESH_FLAGS_AE_A5_A6 0x2 +#define MESH_FLAGS_AE 0x3 +#define MESH_FLAGS_PS_DEEP 0x4 + +/** + * enum ieee80211_preq_flags - mesh PREQ element flags + * + * @IEEE80211_PREQ_PROACTIVE_PREP_FLAG: proactive PREP subfield + */ +enum ieee80211_preq_flags { + IEEE80211_PREQ_PROACTIVE_PREP_FLAG = 1<<2, +}; + +/** + * enum ieee80211_preq_target_flags - mesh PREQ element per target flags + * + * @IEEE80211_PREQ_TO_FLAG: target only subfield + * @IEEE80211_PREQ_USN_FLAG: unknown target HWMP sequence number subfield + */ +enum ieee80211_preq_target_flags { + IEEE80211_PREQ_TO_FLAG = 1<<0, + IEEE80211_PREQ_USN_FLAG = 1<<2, +}; + +/** + * struct ieee80211_mesh_chansw_params_ie - mesh channel switch parameters IE + * @mesh_ttl: Time To Live + * @mesh_flags: Flags + * @mesh_reason: Reason Code + * @mesh_pre_value: Precedence Value + * + * This structure represents the payload of the "Mesh Channel Switch + * Parameters element" as described in IEEE Std 802.11-2020 section + * 9.4.2.102. + */ +struct ieee80211_mesh_chansw_params_ie { + u8 mesh_ttl; + u8 mesh_flags; + __le16 mesh_reason; + __le16 mesh_pre_value; +} __packed; + +/** + * struct ieee80211_meshconf_ie - Mesh Configuration element + * @meshconf_psel: Active Path Selection Protocol Identifier + * @meshconf_pmetric: Active Path Selection Metric Identifier + * @meshconf_congest: Congestion Control Mode Identifier + * @meshconf_synch: Synchronization Method Identifier + * @meshconf_auth: Authentication Protocol Identifier + * @meshconf_form: Mesh Formation Info + * @meshconf_cap: Mesh Capability (see &enum mesh_config_capab_flags) + * + * This structure represents the payload of the "Mesh Configuration + * element" as described in IEEE Std 802.11-2020 section 9.4.2.97. + */ +struct ieee80211_meshconf_ie { + u8 meshconf_psel; + u8 meshconf_pmetric; + u8 meshconf_congest; + u8 meshconf_synch; + u8 meshconf_auth; + u8 meshconf_form; + u8 meshconf_cap; +} __packed; + +/** + * enum mesh_config_capab_flags - Mesh Configuration IE capability field flags + * + * @IEEE80211_MESHCONF_CAPAB_ACCEPT_PLINKS: STA is willing to establish + * additional mesh peerings with other mesh STAs + * @IEEE80211_MESHCONF_CAPAB_FORWARDING: the STA forwards MSDUs + * @IEEE80211_MESHCONF_CAPAB_TBTT_ADJUSTING: TBTT adjustment procedure + * is ongoing + * @IEEE80211_MESHCONF_CAPAB_POWER_SAVE_LEVEL: STA is in deep sleep mode or has + * neighbors in deep sleep mode + * + * Enumerates the "Mesh Capability" as described in IEEE Std + * 802.11-2020 section 9.4.2.97.7. + */ +enum mesh_config_capab_flags { + IEEE80211_MESHCONF_CAPAB_ACCEPT_PLINKS = 0x01, + IEEE80211_MESHCONF_CAPAB_FORWARDING = 0x08, + IEEE80211_MESHCONF_CAPAB_TBTT_ADJUSTING = 0x20, + IEEE80211_MESHCONF_CAPAB_POWER_SAVE_LEVEL = 0x40, +}; + +#define IEEE80211_MESHCONF_FORM_CONNECTED_TO_GATE 0x1 + +/* + * mesh channel switch parameters element's flag indicator + * + */ +#define WLAN_EID_CHAN_SWITCH_PARAM_TX_RESTRICT BIT(0) +#define WLAN_EID_CHAN_SWITCH_PARAM_INITIATOR BIT(1) +#define WLAN_EID_CHAN_SWITCH_PARAM_REASON BIT(2) + +/** + * struct ieee80211_rann_ie - RANN (root announcement) element + * @rann_flags: Flags + * @rann_hopcount: Hop Count + * @rann_ttl: Element TTL + * @rann_addr: Root Mesh STA Address + * @rann_seq: HWMP Sequence Number + * @rann_interval: Interval + * @rann_metric: Metric + * + * This structure represents the payload of the "RANN element" as + * described in IEEE Std 802.11-2020 section 9.4.2.111. + */ +struct ieee80211_rann_ie { + u8 rann_flags; + u8 rann_hopcount; + u8 rann_ttl; + u8 rann_addr[ETH_ALEN]; + __le32 rann_seq; + __le32 rann_interval; + __le32 rann_metric; +} __packed; + +enum ieee80211_rann_flags { + RANN_FLAG_IS_GATE = 1 << 0, +}; + +/* Mesh action codes */ +enum ieee80211_mesh_actioncode { + WLAN_MESH_ACTION_LINK_METRIC_REPORT, + WLAN_MESH_ACTION_HWMP_PATH_SELECTION, + WLAN_MESH_ACTION_GATE_ANNOUNCEMENT, + WLAN_MESH_ACTION_CONGESTION_CONTROL_NOTIFICATION, + WLAN_MESH_ACTION_MCCA_SETUP_REQUEST, + WLAN_MESH_ACTION_MCCA_SETUP_REPLY, + WLAN_MESH_ACTION_MCCA_ADVERTISEMENT_REQUEST, + WLAN_MESH_ACTION_MCCA_ADVERTISEMENT, + WLAN_MESH_ACTION_MCCA_TEARDOWN, + WLAN_MESH_ACTION_TBTT_ADJUSTMENT_REQUEST, + WLAN_MESH_ACTION_TBTT_ADJUSTMENT_RESPONSE, +}; + +/** + * enum ieee80211_mesh_sync_method - mesh synchronization method identifier + * + * @IEEE80211_SYNC_METHOD_NEIGHBOR_OFFSET: the default synchronization method + * @IEEE80211_SYNC_METHOD_VENDOR: a vendor specific synchronization method + * that will be specified in a vendor specific information element + */ +enum ieee80211_mesh_sync_method { + IEEE80211_SYNC_METHOD_NEIGHBOR_OFFSET = 1, + IEEE80211_SYNC_METHOD_VENDOR = 255, +}; + +/** + * enum ieee80211_mesh_path_protocol - mesh path selection protocol identifier + * + * @IEEE80211_PATH_PROTOCOL_HWMP: the default path selection protocol + * @IEEE80211_PATH_PROTOCOL_VENDOR: a vendor specific protocol that will + * be specified in a vendor specific information element + */ +enum ieee80211_mesh_path_protocol { + IEEE80211_PATH_PROTOCOL_HWMP = 1, + IEEE80211_PATH_PROTOCOL_VENDOR = 255, +}; + +/** + * enum ieee80211_mesh_path_metric - mesh path selection metric identifier + * + * @IEEE80211_PATH_METRIC_AIRTIME: the default path selection metric + * @IEEE80211_PATH_METRIC_VENDOR: a vendor specific metric that will be + * specified in a vendor specific information element + */ +enum ieee80211_mesh_path_metric { + IEEE80211_PATH_METRIC_AIRTIME = 1, + IEEE80211_PATH_METRIC_VENDOR = 255, +}; + +/** + * enum ieee80211_root_mode_identifier - root mesh STA mode identifier + * + * These attribute are used by dot11MeshHWMPRootMode to set root mesh STA mode + * + * @IEEE80211_ROOTMODE_NO_ROOT: the mesh STA is not a root mesh STA (default) + * @IEEE80211_ROOTMODE_ROOT: the mesh STA is a root mesh STA if greater than + * this value + * @IEEE80211_PROACTIVE_PREQ_NO_PREP: the mesh STA is a root mesh STA supports + * the proactive PREQ with proactive PREP subfield set to 0 + * @IEEE80211_PROACTIVE_PREQ_WITH_PREP: the mesh STA is a root mesh STA + * supports the proactive PREQ with proactive PREP subfield set to 1 + * @IEEE80211_PROACTIVE_RANN: the mesh STA is a root mesh STA supports + * the proactive RANN + */ +enum ieee80211_root_mode_identifier { + IEEE80211_ROOTMODE_NO_ROOT = 0, + IEEE80211_ROOTMODE_ROOT = 1, + IEEE80211_PROACTIVE_PREQ_NO_PREP = 2, + IEEE80211_PROACTIVE_PREQ_WITH_PREP = 3, + IEEE80211_PROACTIVE_RANN = 4, +}; + +#endif /* LINUX_IEEE80211_MESH_H */ diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h index ddff9102f633..fe78b150ab45 100644 --- a/include/linux/ieee80211.h +++ b/include/linux/ieee80211.h @@ -252,8 +252,6 @@ static inline u16 ieee80211_sn_sub(u16 sn1, u16 sn2) #define IEEE80211_MAX_SSID_LEN 32 -#define IEEE80211_MAX_MESH_ID_LEN 32 - #define IEEE80211_FIRST_TSPEC_TSID 8 #define IEEE80211_NUM_TIDS 16 @@ -881,40 +879,6 @@ static inline u16 ieee80211_get_sn(struct ieee80211_hdr *hdr) return le16_get_bits(hdr->seq_ctrl, IEEE80211_SCTL_SEQ); } -struct ieee80211s_hdr { - u8 flags; - u8 ttl; - __le32 seqnum; - u8 eaddr1[ETH_ALEN]; - u8 eaddr2[ETH_ALEN]; -} __packed __aligned(2); - -/* Mesh flags */ -#define MESH_FLAGS_AE_A4 0x1 -#define MESH_FLAGS_AE_A5_A6 0x2 -#define MESH_FLAGS_AE 0x3 -#define MESH_FLAGS_PS_DEEP 0x4 - -/** - * enum ieee80211_preq_flags - mesh PREQ element flags - * - * @IEEE80211_PREQ_PROACTIVE_PREP_FLAG: proactive PREP subfield - */ -enum ieee80211_preq_flags { - IEEE80211_PREQ_PROACTIVE_PREP_FLAG = 1<<2, -}; - -/** - * enum ieee80211_preq_target_flags - mesh PREQ element per target flags - * - * @IEEE80211_PREQ_TO_FLAG: target only subfield - * @IEEE80211_PREQ_USN_FLAG: unknown target HWMP sequence number subfield - */ -enum ieee80211_preq_target_flags { - IEEE80211_PREQ_TO_FLAG = 1<<0, - IEEE80211_PREQ_USN_FLAG = 1<<2, -}; - /** * struct ieee80211_quiet_ie - Quiet element * @count: Quiet Count @@ -993,24 +957,6 @@ struct ieee80211_sec_chan_offs_ie { u8 sec_chan_offs; } __packed; -/** - * struct ieee80211_mesh_chansw_params_ie - mesh channel switch parameters IE - * @mesh_ttl: Time To Live - * @mesh_flags: Flags - * @mesh_reason: Reason Code - * @mesh_pre_value: Precedence Value - * - * This structure represents the payload of the "Mesh Channel Switch - * Parameters element" as described in IEEE Std 802.11-2020 section - * 9.4.2.102. - */ -struct ieee80211_mesh_chansw_params_ie { - u8 mesh_ttl; - u8 mesh_flags; - __le16 mesh_reason; - __le16 mesh_pre_value; -} __packed; - /** * struct ieee80211_wide_bw_chansw_ie - wide bandwidth channel switch IE * @new_channel_width: New Channel Width @@ -1051,87 +997,6 @@ struct ieee80211_tim_ie { }; } __packed; -/** - * struct ieee80211_meshconf_ie - Mesh Configuration element - * @meshconf_psel: Active Path Selection Protocol Identifier - * @meshconf_pmetric: Active Path Selection Metric Identifier - * @meshconf_congest: Congestion Control Mode Identifier - * @meshconf_synch: Synchronization Method Identifier - * @meshconf_auth: Authentication Protocol Identifier - * @meshconf_form: Mesh Formation Info - * @meshconf_cap: Mesh Capability (see &enum mesh_config_capab_flags) - * - * This structure represents the payload of the "Mesh Configuration - * element" as described in IEEE Std 802.11-2020 section 9.4.2.97. - */ -struct ieee80211_meshconf_ie { - u8 meshconf_psel; - u8 meshconf_pmetric; - u8 meshconf_congest; - u8 meshconf_synch; - u8 meshconf_auth; - u8 meshconf_form; - u8 meshconf_cap; -} __packed; - -/** - * enum mesh_config_capab_flags - Mesh Configuration IE capability field flags - * - * @IEEE80211_MESHCONF_CAPAB_ACCEPT_PLINKS: STA is willing to establish - * additional mesh peerings with other mesh STAs - * @IEEE80211_MESHCONF_CAPAB_FORWARDING: the STA forwards MSDUs - * @IEEE80211_MESHCONF_CAPAB_TBTT_ADJUSTING: TBTT adjustment procedure - * is ongoing - * @IEEE80211_MESHCONF_CAPAB_POWER_SAVE_LEVEL: STA is in deep sleep mode or has - * neighbors in deep sleep mode - * - * Enumerates the "Mesh Capability" as described in IEEE Std - * 802.11-2020 section 9.4.2.97.7. - */ -enum mesh_config_capab_flags { - IEEE80211_MESHCONF_CAPAB_ACCEPT_PLINKS = 0x01, - IEEE80211_MESHCONF_CAPAB_FORWARDING = 0x08, - IEEE80211_MESHCONF_CAPAB_TBTT_ADJUSTING = 0x20, - IEEE80211_MESHCONF_CAPAB_POWER_SAVE_LEVEL = 0x40, -}; - -#define IEEE80211_MESHCONF_FORM_CONNECTED_TO_GATE 0x1 - -/* - * mesh channel switch parameters element's flag indicator - * - */ -#define WLAN_EID_CHAN_SWITCH_PARAM_TX_RESTRICT BIT(0) -#define WLAN_EID_CHAN_SWITCH_PARAM_INITIATOR BIT(1) -#define WLAN_EID_CHAN_SWITCH_PARAM_REASON BIT(2) - -/** - * struct ieee80211_rann_ie - RANN (root announcement) element - * @rann_flags: Flags - * @rann_hopcount: Hop Count - * @rann_ttl: Element TTL - * @rann_addr: Root Mesh STA Address - * @rann_seq: HWMP Sequence Number - * @rann_interval: Interval - * @rann_metric: Metric - * - * This structure represents the payload of the "RANN element" as - * described in IEEE Std 802.11-2020 section 9.4.2.111. - */ -struct ieee80211_rann_ie { - u8 rann_flags; - u8 rann_hopcount; - u8 rann_ttl; - u8 rann_addr[ETH_ALEN]; - __le32 rann_seq; - __le32 rann_interval; - __le32 rann_metric; -} __packed; - -enum ieee80211_rann_flags { - RANN_FLAG_IS_GATE = 1 << 0, -}; - enum ieee80211_ht_chanwidth_values { IEEE80211_HT_CHANWIDTH_20MHZ = 0, IEEE80211_HT_CHANWIDTH_ANY = 1, @@ -3971,21 +3836,6 @@ enum ieee80211_self_protected_actioncode { WLAN_SP_MGK_ACK = 5, }; -/* Mesh action codes */ -enum ieee80211_mesh_actioncode { - WLAN_MESH_ACTION_LINK_METRIC_REPORT, - WLAN_MESH_ACTION_HWMP_PATH_SELECTION, - WLAN_MESH_ACTION_GATE_ANNOUNCEMENT, - WLAN_MESH_ACTION_CONGESTION_CONTROL_NOTIFICATION, - WLAN_MESH_ACTION_MCCA_SETUP_REQUEST, - WLAN_MESH_ACTION_MCCA_SETUP_REPLY, - WLAN_MESH_ACTION_MCCA_ADVERTISEMENT_REQUEST, - WLAN_MESH_ACTION_MCCA_ADVERTISEMENT, - WLAN_MESH_ACTION_MCCA_TEARDOWN, - WLAN_MESH_ACTION_TBTT_ADJUSTMENT_REQUEST, - WLAN_MESH_ACTION_TBTT_ADJUSTMENT_RESPONSE, -}; - /* Unprotected WNM action codes */ enum ieee80211_unprotected_wnm_actioncode { WLAN_UNPROTECTED_WNM_ACTION_TIM = 0, @@ -4198,65 +4048,6 @@ enum ieee80211_tdls_actioncode { /* BSS Coex IE information field bits */ #define WLAN_BSS_COEX_INFORMATION_REQUEST BIT(0) -/** - * enum ieee80211_mesh_sync_method - mesh synchronization method identifier - * - * @IEEE80211_SYNC_METHOD_NEIGHBOR_OFFSET: the default synchronization method - * @IEEE80211_SYNC_METHOD_VENDOR: a vendor specific synchronization method - * that will be specified in a vendor specific information element - */ -enum ieee80211_mesh_sync_method { - IEEE80211_SYNC_METHOD_NEIGHBOR_OFFSET = 1, - IEEE80211_SYNC_METHOD_VENDOR = 255, -}; - -/** - * enum ieee80211_mesh_path_protocol - mesh path selection protocol identifier - * - * @IEEE80211_PATH_PROTOCOL_HWMP: the default path selection protocol - * @IEEE80211_PATH_PROTOCOL_VENDOR: a vendor specific protocol that will - * be specified in a vendor specific information element - */ -enum ieee80211_mesh_path_protocol { - IEEE80211_PATH_PROTOCOL_HWMP = 1, - IEEE80211_PATH_PROTOCOL_VENDOR = 255, -}; - -/** - * enum ieee80211_mesh_path_metric - mesh path selection metric identifier - * - * @IEEE80211_PATH_METRIC_AIRTIME: the default path selection metric - * @IEEE80211_PATH_METRIC_VENDOR: a vendor specific metric that will be - * specified in a vendor specific information element - */ -enum ieee80211_mesh_path_metric { - IEEE80211_PATH_METRIC_AIRTIME = 1, - IEEE80211_PATH_METRIC_VENDOR = 255, -}; - -/** - * enum ieee80211_root_mode_identifier - root mesh STA mode identifier - * - * These attribute are used by dot11MeshHWMPRootMode to set root mesh STA mode - * - * @IEEE80211_ROOTMODE_NO_ROOT: the mesh STA is not a root mesh STA (default) - * @IEEE80211_ROOTMODE_ROOT: the mesh STA is a root mesh STA if greater than - * this value - * @IEEE80211_PROACTIVE_PREQ_NO_PREP: the mesh STA is a root mesh STA supports - * the proactive PREQ with proactive PREP subfield set to 0 - * @IEEE80211_PROACTIVE_PREQ_WITH_PREP: the mesh STA is a root mesh STA - * supports the proactive PREQ with proactive PREP subfield set to 1 - * @IEEE80211_PROACTIVE_RANN: the mesh STA is a root mesh STA supports - * the proactive RANN - */ -enum ieee80211_root_mode_identifier { - IEEE80211_ROOTMODE_NO_ROOT = 0, - IEEE80211_ROOTMODE_ROOT = 1, - IEEE80211_PROACTIVE_PREQ_NO_PREP = 2, - IEEE80211_PROACTIVE_PREQ_WITH_PREP = 3, - IEEE80211_PROACTIVE_RANN = 4, -}; - /* * IEEE 802.11-2007 7.3.2.9 Country information element * @@ -6098,4 +5889,6 @@ static inline u32 ieee80211_eml_trans_timeout_in_us(u16 eml_cap) #define NAN_DEV_CAPA_NDPE_SUPPORTED 0x08 #define NAN_DEV_CAPA_S3_SUPPORTED 0x10 +#include "ieee80211-mesh.h" + #endif /* LINUX_IEEE80211_H */ From fdc1c141f3ef4dc94e3880e973061681843f62c0 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 5 Nov 2025 15:36:50 +0100 Subject: [PATCH 0634/1381] wifi: ieee80211: split HT definitions out The ieee80211.h file has gotten very long, continue splitting it by putting HT definitions into a separate file. Link: https://patch.msgid.link/20251105153843.7532471178d0.Id956a5433ad8658e4e5c0272dbcbb59587206142@changeid Signed-off-by: Johannes Berg --- include/linux/ieee80211-ht.h | 292 +++++++++++++++++++++++++++++++++++ include/linux/ieee80211.h | 272 +------------------------------- 2 files changed, 293 insertions(+), 271 deletions(-) create mode 100644 include/linux/ieee80211-ht.h diff --git a/include/linux/ieee80211-ht.h b/include/linux/ieee80211-ht.h new file mode 100644 index 000000000000..21bbf470540f --- /dev/null +++ b/include/linux/ieee80211-ht.h @@ -0,0 +1,292 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * IEEE 802.11 HT definitions + * + * Copyright (c) 2001-2002, SSH Communications Security Corp and Jouni Malinen + * + * Copyright (c) 2002-2003, Jouni Malinen + * Copyright (c) 2005, Devicescape Software, Inc. + * Copyright (c) 2006, Michael Wu + * Copyright (c) 2013 - 2014 Intel Mobile Communications GmbH + * Copyright (c) 2016 - 2017 Intel Deutschland GmbH + * Copyright (c) 2018 - 2025 Intel Corporation + */ + +#ifndef LINUX_IEEE80211_HT_H +#define LINUX_IEEE80211_HT_H + +#include +#include + +/* Maximal size of an A-MSDU that can be transported in a HT BA session */ +#define IEEE80211_MAX_MPDU_LEN_HT_BA 4095 + +/* Maximal size of an A-MSDU */ +#define IEEE80211_MAX_MPDU_LEN_HT_3839 3839 +#define IEEE80211_MAX_MPDU_LEN_HT_7935 7935 + +#define IEEE80211_HT_CTL_LEN 4 + +enum ieee80211_ht_chanwidth_values { + IEEE80211_HT_CHANWIDTH_20MHZ = 0, + IEEE80211_HT_CHANWIDTH_ANY = 1, +}; + +/** + * struct ieee80211_bar - Block Ack Request frame format + * @frame_control: Frame Control + * @duration: Duration + * @ra: RA + * @ta: TA + * @control: BAR Control + * @start_seq_num: Starting Sequence Number (see Figure 9-37) + * + * This structure represents the "BlockAckReq frame format" + * as described in IEEE Std 802.11-2020 section 9.3.1.7. +*/ +struct ieee80211_bar { + __le16 frame_control; + __le16 duration; + __u8 ra[ETH_ALEN]; + __u8 ta[ETH_ALEN]; + __le16 control; + __le16 start_seq_num; +} __packed; + +/* 802.11 BAR control masks */ +#define IEEE80211_BAR_CTRL_ACK_POLICY_NORMAL 0x0000 +#define IEEE80211_BAR_CTRL_MULTI_TID 0x0002 +#define IEEE80211_BAR_CTRL_CBMTID_COMPRESSED_BA 0x0004 +#define IEEE80211_BAR_CTRL_TID_INFO_MASK 0xf000 +#define IEEE80211_BAR_CTRL_TID_INFO_SHIFT 12 + +#define IEEE80211_HT_MCS_MASK_LEN 10 + +/** + * struct ieee80211_mcs_info - Supported MCS Set field + * @rx_mask: RX mask + * @rx_highest: highest supported RX rate. If set represents + * the highest supported RX data rate in units of 1 Mbps. + * If this field is 0 this value should not be used to + * consider the highest RX data rate supported. + * @tx_params: TX parameters + * @reserved: Reserved bits + * + * This structure represents the "Supported MCS Set field" as + * described in IEEE Std 802.11-2020 section 9.4.2.55.4. + */ +struct ieee80211_mcs_info { + u8 rx_mask[IEEE80211_HT_MCS_MASK_LEN]; + __le16 rx_highest; + u8 tx_params; + u8 reserved[3]; +} __packed; + +/* 802.11n HT capability MSC set */ +#define IEEE80211_HT_MCS_RX_HIGHEST_MASK 0x3ff +#define IEEE80211_HT_MCS_TX_DEFINED 0x01 +#define IEEE80211_HT_MCS_TX_RX_DIFF 0x02 +/* value 0 == 1 stream etc */ +#define IEEE80211_HT_MCS_TX_MAX_STREAMS_MASK 0x0C +#define IEEE80211_HT_MCS_TX_MAX_STREAMS_SHIFT 2 +#define IEEE80211_HT_MCS_TX_MAX_STREAMS 4 +#define IEEE80211_HT_MCS_TX_UNEQUAL_MODULATION 0x10 + +#define IEEE80211_HT_MCS_CHAINS(mcs) ((mcs) == 32 ? 1 : (1 + ((mcs) >> 3))) + +/* + * 802.11n D5.0 20.3.5 / 20.6 says: + * - indices 0 to 7 and 32 are single spatial stream + * - 8 to 31 are multiple spatial streams using equal modulation + * [8..15 for two streams, 16..23 for three and 24..31 for four] + * - remainder are multiple spatial streams using unequal modulation + */ +#define IEEE80211_HT_MCS_UNEQUAL_MODULATION_START 33 +#define IEEE80211_HT_MCS_UNEQUAL_MODULATION_START_BYTE \ + (IEEE80211_HT_MCS_UNEQUAL_MODULATION_START / 8) + +/** + * struct ieee80211_ht_cap - HT capabilities element + * @cap_info: HT Capability Information + * @ampdu_params_info: A-MPDU Parameters + * @mcs: Supported MCS Set + * @extended_ht_cap_info: HT Extended Capabilities + * @tx_BF_cap_info: Transmit Beamforming Capabilities + * @antenna_selection_info: ASEL Capability + * + * This structure represents the payload of the "HT Capabilities + * element" as described in IEEE Std 802.11-2020 section 9.4.2.55. + */ +struct ieee80211_ht_cap { + __le16 cap_info; + u8 ampdu_params_info; + + /* 16 bytes MCS information */ + struct ieee80211_mcs_info mcs; + + __le16 extended_ht_cap_info; + __le32 tx_BF_cap_info; + u8 antenna_selection_info; +} __packed; + +/* 802.11n HT capabilities masks (for cap_info) */ +#define IEEE80211_HT_CAP_LDPC_CODING 0x0001 +#define IEEE80211_HT_CAP_SUP_WIDTH_20_40 0x0002 +#define IEEE80211_HT_CAP_SM_PS 0x000C +#define IEEE80211_HT_CAP_SM_PS_SHIFT 2 +#define IEEE80211_HT_CAP_GRN_FLD 0x0010 +#define IEEE80211_HT_CAP_SGI_20 0x0020 +#define IEEE80211_HT_CAP_SGI_40 0x0040 +#define IEEE80211_HT_CAP_TX_STBC 0x0080 +#define IEEE80211_HT_CAP_RX_STBC 0x0300 +#define IEEE80211_HT_CAP_RX_STBC_SHIFT 8 +#define IEEE80211_HT_CAP_DELAY_BA 0x0400 +#define IEEE80211_HT_CAP_MAX_AMSDU 0x0800 +#define IEEE80211_HT_CAP_DSSSCCK40 0x1000 +#define IEEE80211_HT_CAP_RESERVED 0x2000 +#define IEEE80211_HT_CAP_40MHZ_INTOLERANT 0x4000 +#define IEEE80211_HT_CAP_LSIG_TXOP_PROT 0x8000 + +/* 802.11n HT extended capabilities masks (for extended_ht_cap_info) */ +#define IEEE80211_HT_EXT_CAP_PCO 0x0001 +#define IEEE80211_HT_EXT_CAP_PCO_TIME 0x0006 +#define IEEE80211_HT_EXT_CAP_PCO_TIME_SHIFT 1 +#define IEEE80211_HT_EXT_CAP_MCS_FB 0x0300 +#define IEEE80211_HT_EXT_CAP_MCS_FB_SHIFT 8 +#define IEEE80211_HT_EXT_CAP_HTC_SUP 0x0400 +#define IEEE80211_HT_EXT_CAP_RD_RESPONDER 0x0800 + +/* 802.11n HT capability AMPDU settings (for ampdu_params_info) */ +#define IEEE80211_HT_AMPDU_PARM_FACTOR 0x03 +#define IEEE80211_HT_AMPDU_PARM_DENSITY 0x1C +#define IEEE80211_HT_AMPDU_PARM_DENSITY_SHIFT 2 + +/* + * Maximum length of AMPDU that the STA can receive in high-throughput (HT). + * Length = 2 ^ (13 + max_ampdu_length_exp) - 1 (octets) + */ +enum ieee80211_max_ampdu_length_exp { + IEEE80211_HT_MAX_AMPDU_8K = 0, + IEEE80211_HT_MAX_AMPDU_16K = 1, + IEEE80211_HT_MAX_AMPDU_32K = 2, + IEEE80211_HT_MAX_AMPDU_64K = 3 +}; + +#define IEEE80211_HT_MAX_AMPDU_FACTOR 13 + +/* Minimum MPDU start spacing */ +enum ieee80211_min_mpdu_spacing { + IEEE80211_HT_MPDU_DENSITY_NONE = 0, /* No restriction */ + IEEE80211_HT_MPDU_DENSITY_0_25 = 1, /* 1/4 usec */ + IEEE80211_HT_MPDU_DENSITY_0_5 = 2, /* 1/2 usec */ + IEEE80211_HT_MPDU_DENSITY_1 = 3, /* 1 usec */ + IEEE80211_HT_MPDU_DENSITY_2 = 4, /* 2 usec */ + IEEE80211_HT_MPDU_DENSITY_4 = 5, /* 4 usec */ + IEEE80211_HT_MPDU_DENSITY_8 = 6, /* 8 usec */ + IEEE80211_HT_MPDU_DENSITY_16 = 7 /* 16 usec */ +}; + +/** + * struct ieee80211_ht_operation - HT operation IE + * @primary_chan: Primary Channel + * @ht_param: HT Operation Information parameters + * @operation_mode: HT Operation Information operation mode + * @stbc_param: HT Operation Information STBC params + * @basic_set: Basic HT-MCS Set + * + * This structure represents the payload of the "HT Operation + * element" as described in IEEE Std 802.11-2020 section 9.4.2.56. + */ +struct ieee80211_ht_operation { + u8 primary_chan; + u8 ht_param; + __le16 operation_mode; + __le16 stbc_param; + u8 basic_set[16]; +} __packed; + +/* for ht_param */ +#define IEEE80211_HT_PARAM_CHA_SEC_OFFSET 0x03 +#define IEEE80211_HT_PARAM_CHA_SEC_NONE 0x00 +#define IEEE80211_HT_PARAM_CHA_SEC_ABOVE 0x01 +#define IEEE80211_HT_PARAM_CHA_SEC_BELOW 0x03 +#define IEEE80211_HT_PARAM_CHAN_WIDTH_ANY 0x04 +#define IEEE80211_HT_PARAM_RIFS_MODE 0x08 + +/* for operation_mode */ +#define IEEE80211_HT_OP_MODE_PROTECTION 0x0003 +#define IEEE80211_HT_OP_MODE_PROTECTION_NONE 0 +#define IEEE80211_HT_OP_MODE_PROTECTION_NONMEMBER 1 +#define IEEE80211_HT_OP_MODE_PROTECTION_20MHZ 2 +#define IEEE80211_HT_OP_MODE_PROTECTION_NONHT_MIXED 3 +#define IEEE80211_HT_OP_MODE_NON_GF_STA_PRSNT 0x0004 +#define IEEE80211_HT_OP_MODE_NON_HT_STA_PRSNT 0x0010 +#define IEEE80211_HT_OP_MODE_CCFS2_SHIFT 5 +#define IEEE80211_HT_OP_MODE_CCFS2_MASK 0x1fe0 + +/* for stbc_param */ +#define IEEE80211_HT_STBC_PARAM_DUAL_BEACON 0x0040 +#define IEEE80211_HT_STBC_PARAM_DUAL_CTS_PROT 0x0080 +#define IEEE80211_HT_STBC_PARAM_STBC_BEACON 0x0100 +#define IEEE80211_HT_STBC_PARAM_LSIG_TXOP_FULLPROT 0x0200 +#define IEEE80211_HT_STBC_PARAM_PCO_ACTIVE 0x0400 +#define IEEE80211_HT_STBC_PARAM_PCO_PHASE 0x0800 + + +/* block-ack parameters */ +#define IEEE80211_ADDBA_PARAM_AMSDU_MASK 0x0001 +#define IEEE80211_ADDBA_PARAM_POLICY_MASK 0x0002 +#define IEEE80211_ADDBA_PARAM_TID_MASK 0x003C +#define IEEE80211_ADDBA_PARAM_BUF_SIZE_MASK 0xFFC0 +#define IEEE80211_DELBA_PARAM_TID_MASK 0xF000 +#define IEEE80211_DELBA_PARAM_INITIATOR_MASK 0x0800 + +/* + * A-MPDU buffer sizes + * According to HT size varies from 8 to 64 frames + * HE adds the ability to have up to 256 frames. + * EHT adds the ability to have up to 1K frames. + */ +#define IEEE80211_MIN_AMPDU_BUF 0x8 +#define IEEE80211_MAX_AMPDU_BUF_HT 0x40 +#define IEEE80211_MAX_AMPDU_BUF_HE 0x100 +#define IEEE80211_MAX_AMPDU_BUF_EHT 0x400 + + +/* Spatial Multiplexing Power Save Modes (for capability) */ +#define WLAN_HT_CAP_SM_PS_STATIC 0 +#define WLAN_HT_CAP_SM_PS_DYNAMIC 1 +#define WLAN_HT_CAP_SM_PS_INVALID 2 +#define WLAN_HT_CAP_SM_PS_DISABLED 3 + +/* for SM power control field lower two bits */ +#define WLAN_HT_SMPS_CONTROL_DISABLED 0 +#define WLAN_HT_SMPS_CONTROL_STATIC 1 +#define WLAN_HT_SMPS_CONTROL_DYNAMIC 3 + +/* HT action codes */ +enum ieee80211_ht_actioncode { + WLAN_HT_ACTION_NOTIFY_CHANWIDTH = 0, + WLAN_HT_ACTION_SMPS = 1, + WLAN_HT_ACTION_PSMP = 2, + WLAN_HT_ACTION_PCO_PHASE = 3, + WLAN_HT_ACTION_CSI = 4, + WLAN_HT_ACTION_NONCOMPRESSED_BF = 5, + WLAN_HT_ACTION_COMPRESSED_BF = 6, + WLAN_HT_ACTION_ASEL_IDX_FEEDBACK = 7, +}; + +/* BACK action code */ +enum ieee80211_back_actioncode { + WLAN_ACTION_ADDBA_REQ = 0, + WLAN_ACTION_ADDBA_RESP = 1, + WLAN_ACTION_DELBA = 2, +}; + +/* BACK (block-ack) parties */ +enum ieee80211_back_parties { + WLAN_BACK_RECIPIENT = 0, + WLAN_BACK_INITIATOR = 1, +}; + +#endif /* LINUX_IEEE80211_HT_H */ diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h index fe78b150ab45..0a9b4a8025cd 100644 --- a/include/linux/ieee80211.h +++ b/include/linux/ieee80211.h @@ -239,13 +239,6 @@ static inline u16 ieee80211_sn_sub(u16 sn1, u16 sn2) /* 30 byte 4 addr hdr, 2 byte QoS, 2304 byte MSDU, 12 byte crypt, 4 byte FCS */ #define IEEE80211_MAX_FRAME_LEN 2352 -/* Maximal size of an A-MSDU that can be transported in a HT BA session */ -#define IEEE80211_MAX_MPDU_LEN_HT_BA 4095 - -/* Maximal size of an A-MSDU */ -#define IEEE80211_MAX_MPDU_LEN_HT_3839 3839 -#define IEEE80211_MAX_MPDU_LEN_HT_7935 7935 - #define IEEE80211_MAX_MPDU_LEN_VHT_3895 3895 #define IEEE80211_MAX_MPDU_LEN_VHT_7991 7991 #define IEEE80211_MAX_MPDU_LEN_VHT_11454 11454 @@ -302,8 +295,6 @@ static inline u16 ieee80211_sn_sub(u16 sn1, u16 sn2) #define IEEE80211_WMM_IE_STA_QOSINFO_SP_MASK 0x03 #define IEEE80211_WMM_IE_STA_QOSINFO_SP_SHIFT 5 -#define IEEE80211_HT_CTL_LEN 4 - /* trigger type within common_info of trigger frame */ #define IEEE80211_TRIGGER_TYPE_MASK 0xf #define IEEE80211_TRIGGER_TYPE_BASIC 0x0 @@ -997,11 +988,6 @@ struct ieee80211_tim_ie { }; } __packed; -enum ieee80211_ht_chanwidth_values { - IEEE80211_HT_CHANWIDTH_20MHZ = 0, - IEEE80211_HT_CHANWIDTH_ANY = 1, -}; - /** * enum ieee80211_vht_opmode_bits - VHT operating mode field bits * @IEEE80211_OPMODE_NOTIF_CHANWIDTH_MASK: channel width mask @@ -1677,146 +1663,6 @@ struct ieee80211_p2p_noa_attr { #define IEEE80211_P2P_OPPPS_ENABLE_BIT BIT(7) #define IEEE80211_P2P_OPPPS_CTWINDOW_MASK 0x7F -/** - * struct ieee80211_bar - Block Ack Request frame format - * @frame_control: Frame Control - * @duration: Duration - * @ra: RA - * @ta: TA - * @control: BAR Control - * @start_seq_num: Starting Sequence Number (see Figure 9-37) - * - * This structure represents the "BlockAckReq frame format" - * as described in IEEE Std 802.11-2020 section 9.3.1.7. -*/ -struct ieee80211_bar { - __le16 frame_control; - __le16 duration; - __u8 ra[ETH_ALEN]; - __u8 ta[ETH_ALEN]; - __le16 control; - __le16 start_seq_num; -} __packed; - -/* 802.11 BAR control masks */ -#define IEEE80211_BAR_CTRL_ACK_POLICY_NORMAL 0x0000 -#define IEEE80211_BAR_CTRL_MULTI_TID 0x0002 -#define IEEE80211_BAR_CTRL_CBMTID_COMPRESSED_BA 0x0004 -#define IEEE80211_BAR_CTRL_TID_INFO_MASK 0xf000 -#define IEEE80211_BAR_CTRL_TID_INFO_SHIFT 12 - -#define IEEE80211_HT_MCS_MASK_LEN 10 - -/** - * struct ieee80211_mcs_info - Supported MCS Set field - * @rx_mask: RX mask - * @rx_highest: highest supported RX rate. If set represents - * the highest supported RX data rate in units of 1 Mbps. - * If this field is 0 this value should not be used to - * consider the highest RX data rate supported. - * @tx_params: TX parameters - * @reserved: Reserved bits - * - * This structure represents the "Supported MCS Set field" as - * described in IEEE Std 802.11-2020 section 9.4.2.55.4. - */ -struct ieee80211_mcs_info { - u8 rx_mask[IEEE80211_HT_MCS_MASK_LEN]; - __le16 rx_highest; - u8 tx_params; - u8 reserved[3]; -} __packed; - -/* 802.11n HT capability MSC set */ -#define IEEE80211_HT_MCS_RX_HIGHEST_MASK 0x3ff -#define IEEE80211_HT_MCS_TX_DEFINED 0x01 -#define IEEE80211_HT_MCS_TX_RX_DIFF 0x02 -/* value 0 == 1 stream etc */ -#define IEEE80211_HT_MCS_TX_MAX_STREAMS_MASK 0x0C -#define IEEE80211_HT_MCS_TX_MAX_STREAMS_SHIFT 2 -#define IEEE80211_HT_MCS_TX_MAX_STREAMS 4 -#define IEEE80211_HT_MCS_TX_UNEQUAL_MODULATION 0x10 - -#define IEEE80211_HT_MCS_CHAINS(mcs) ((mcs) == 32 ? 1 : (1 + ((mcs) >> 3))) - -/* - * 802.11n D5.0 20.3.5 / 20.6 says: - * - indices 0 to 7 and 32 are single spatial stream - * - 8 to 31 are multiple spatial streams using equal modulation - * [8..15 for two streams, 16..23 for three and 24..31 for four] - * - remainder are multiple spatial streams using unequal modulation - */ -#define IEEE80211_HT_MCS_UNEQUAL_MODULATION_START 33 -#define IEEE80211_HT_MCS_UNEQUAL_MODULATION_START_BYTE \ - (IEEE80211_HT_MCS_UNEQUAL_MODULATION_START / 8) - -/** - * struct ieee80211_ht_cap - HT capabilities element - * @cap_info: HT Capability Information - * @ampdu_params_info: A-MPDU Parameters - * @mcs: Supported MCS Set - * @extended_ht_cap_info: HT Extended Capabilities - * @tx_BF_cap_info: Transmit Beamforming Capabilities - * @antenna_selection_info: ASEL Capability - * - * This structure represents the payload of the "HT Capabilities - * element" as described in IEEE Std 802.11-2020 section 9.4.2.55. - */ -struct ieee80211_ht_cap { - __le16 cap_info; - u8 ampdu_params_info; - - /* 16 bytes MCS information */ - struct ieee80211_mcs_info mcs; - - __le16 extended_ht_cap_info; - __le32 tx_BF_cap_info; - u8 antenna_selection_info; -} __packed; - -/* 802.11n HT capabilities masks (for cap_info) */ -#define IEEE80211_HT_CAP_LDPC_CODING 0x0001 -#define IEEE80211_HT_CAP_SUP_WIDTH_20_40 0x0002 -#define IEEE80211_HT_CAP_SM_PS 0x000C -#define IEEE80211_HT_CAP_SM_PS_SHIFT 2 -#define IEEE80211_HT_CAP_GRN_FLD 0x0010 -#define IEEE80211_HT_CAP_SGI_20 0x0020 -#define IEEE80211_HT_CAP_SGI_40 0x0040 -#define IEEE80211_HT_CAP_TX_STBC 0x0080 -#define IEEE80211_HT_CAP_RX_STBC 0x0300 -#define IEEE80211_HT_CAP_RX_STBC_SHIFT 8 -#define IEEE80211_HT_CAP_DELAY_BA 0x0400 -#define IEEE80211_HT_CAP_MAX_AMSDU 0x0800 -#define IEEE80211_HT_CAP_DSSSCCK40 0x1000 -#define IEEE80211_HT_CAP_RESERVED 0x2000 -#define IEEE80211_HT_CAP_40MHZ_INTOLERANT 0x4000 -#define IEEE80211_HT_CAP_LSIG_TXOP_PROT 0x8000 - -/* 802.11n HT extended capabilities masks (for extended_ht_cap_info) */ -#define IEEE80211_HT_EXT_CAP_PCO 0x0001 -#define IEEE80211_HT_EXT_CAP_PCO_TIME 0x0006 -#define IEEE80211_HT_EXT_CAP_PCO_TIME_SHIFT 1 -#define IEEE80211_HT_EXT_CAP_MCS_FB 0x0300 -#define IEEE80211_HT_EXT_CAP_MCS_FB_SHIFT 8 -#define IEEE80211_HT_EXT_CAP_HTC_SUP 0x0400 -#define IEEE80211_HT_EXT_CAP_RD_RESPONDER 0x0800 - -/* 802.11n HT capability AMPDU settings (for ampdu_params_info) */ -#define IEEE80211_HT_AMPDU_PARM_FACTOR 0x03 -#define IEEE80211_HT_AMPDU_PARM_DENSITY 0x1C -#define IEEE80211_HT_AMPDU_PARM_DENSITY_SHIFT 2 - -/* - * Maximum length of AMPDU that the STA can receive in high-throughput (HT). - * Length = 2 ^ (13 + max_ampdu_length_exp) - 1 (octets) - */ -enum ieee80211_max_ampdu_length_exp { - IEEE80211_HT_MAX_AMPDU_8K = 0, - IEEE80211_HT_MAX_AMPDU_16K = 1, - IEEE80211_HT_MAX_AMPDU_32K = 2, - IEEE80211_HT_MAX_AMPDU_64K = 3 -}; - /* * Maximum length of AMPDU that the STA can receive in VHT. * Length = 2 ^ (13 + max_ampdu_length_exp) - 1 (octets) @@ -1832,98 +1678,6 @@ enum ieee80211_vht_max_ampdu_length_exp { IEEE80211_VHT_MAX_AMPDU_1024K = 7 }; -#define IEEE80211_HT_MAX_AMPDU_FACTOR 13 - -/* Minimum MPDU start spacing */ -enum ieee80211_min_mpdu_spacing { - IEEE80211_HT_MPDU_DENSITY_NONE = 0, /* No restriction */ - IEEE80211_HT_MPDU_DENSITY_0_25 = 1, /* 1/4 usec */ - IEEE80211_HT_MPDU_DENSITY_0_5 = 2, /* 1/2 usec */ - IEEE80211_HT_MPDU_DENSITY_1 = 3, /* 1 usec */ - IEEE80211_HT_MPDU_DENSITY_2 = 4, /* 2 usec */ - IEEE80211_HT_MPDU_DENSITY_4 = 5, /* 4 usec */ - IEEE80211_HT_MPDU_DENSITY_8 = 6, /* 8 usec */ - IEEE80211_HT_MPDU_DENSITY_16 = 7 /* 16 usec */ -}; - -/** - * struct ieee80211_ht_operation - HT operation IE - * @primary_chan: Primary Channel - * @ht_param: HT Operation Information parameters - * @operation_mode: HT Operation Information operation mode - * @stbc_param: HT Operation Information STBC params - * @basic_set: Basic HT-MCS Set - * - * This structure represents the payload of the "HT Operation - * element" as described in IEEE Std 802.11-2020 section 9.4.2.56. - */ -struct ieee80211_ht_operation { - u8 primary_chan; - u8 ht_param; - __le16 operation_mode; - __le16 stbc_param; - u8 basic_set[16]; -} __packed; - -/* for ht_param */ -#define IEEE80211_HT_PARAM_CHA_SEC_OFFSET 0x03 -#define IEEE80211_HT_PARAM_CHA_SEC_NONE 0x00 -#define IEEE80211_HT_PARAM_CHA_SEC_ABOVE 0x01 -#define IEEE80211_HT_PARAM_CHA_SEC_BELOW 0x03 -#define IEEE80211_HT_PARAM_CHAN_WIDTH_ANY 0x04 -#define IEEE80211_HT_PARAM_RIFS_MODE 0x08 - -/* for operation_mode */ -#define IEEE80211_HT_OP_MODE_PROTECTION 0x0003 -#define IEEE80211_HT_OP_MODE_PROTECTION_NONE 0 -#define IEEE80211_HT_OP_MODE_PROTECTION_NONMEMBER 1 -#define IEEE80211_HT_OP_MODE_PROTECTION_20MHZ 2 -#define IEEE80211_HT_OP_MODE_PROTECTION_NONHT_MIXED 3 -#define IEEE80211_HT_OP_MODE_NON_GF_STA_PRSNT 0x0004 -#define IEEE80211_HT_OP_MODE_NON_HT_STA_PRSNT 0x0010 -#define IEEE80211_HT_OP_MODE_CCFS2_SHIFT 5 -#define IEEE80211_HT_OP_MODE_CCFS2_MASK 0x1fe0 - -/* for stbc_param */ -#define IEEE80211_HT_STBC_PARAM_DUAL_BEACON 0x0040 -#define IEEE80211_HT_STBC_PARAM_DUAL_CTS_PROT 0x0080 -#define IEEE80211_HT_STBC_PARAM_STBC_BEACON 0x0100 -#define IEEE80211_HT_STBC_PARAM_LSIG_TXOP_FULLPROT 0x0200 -#define IEEE80211_HT_STBC_PARAM_PCO_ACTIVE 0x0400 -#define IEEE80211_HT_STBC_PARAM_PCO_PHASE 0x0800 - - -/* block-ack parameters */ -#define IEEE80211_ADDBA_PARAM_AMSDU_MASK 0x0001 -#define IEEE80211_ADDBA_PARAM_POLICY_MASK 0x0002 -#define IEEE80211_ADDBA_PARAM_TID_MASK 0x003C -#define IEEE80211_ADDBA_PARAM_BUF_SIZE_MASK 0xFFC0 -#define IEEE80211_DELBA_PARAM_TID_MASK 0xF000 -#define IEEE80211_DELBA_PARAM_INITIATOR_MASK 0x0800 - -/* - * A-MPDU buffer sizes - * According to HT size varies from 8 to 64 frames - * HE adds the ability to have up to 256 frames. - * EHT adds the ability to have up to 1K frames. - */ -#define IEEE80211_MIN_AMPDU_BUF 0x8 -#define IEEE80211_MAX_AMPDU_BUF_HT 0x40 -#define IEEE80211_MAX_AMPDU_BUF_HE 0x100 -#define IEEE80211_MAX_AMPDU_BUF_EHT 0x400 - - -/* Spatial Multiplexing Power Save Modes (for capability) */ -#define WLAN_HT_CAP_SM_PS_STATIC 0 -#define WLAN_HT_CAP_SM_PS_DYNAMIC 1 -#define WLAN_HT_CAP_SM_PS_INVALID 2 -#define WLAN_HT_CAP_SM_PS_DISABLED 3 - -/* for SM power control field lower two bits */ -#define WLAN_HT_SMPS_CONTROL_DISABLED 0 -#define WLAN_HT_SMPS_CONTROL_STATIC 1 -#define WLAN_HT_SMPS_CONTROL_DYNAMIC 3 - /** * struct ieee80211_vht_mcs_info - VHT MCS information * @rx_mcs_map: RX MCS map 2 bits for each stream, total 8 streams @@ -3807,18 +3561,6 @@ enum ieee80211_spectrum_mgmt_actioncode { WLAN_ACTION_SPCT_CHL_SWITCH = 4, }; -/* HT action codes */ -enum ieee80211_ht_actioncode { - WLAN_HT_ACTION_NOTIFY_CHANWIDTH = 0, - WLAN_HT_ACTION_SMPS = 1, - WLAN_HT_ACTION_PSMP = 2, - WLAN_HT_ACTION_PCO_PHASE = 3, - WLAN_HT_ACTION_CSI = 4, - WLAN_HT_ACTION_NONCOMPRESSED_BF = 5, - WLAN_HT_ACTION_COMPRESSED_BF = 6, - WLAN_HT_ACTION_ASEL_IDX_FEEDBACK = 7, -}; - /* VHT action codes */ enum ieee80211_vht_actioncode { WLAN_VHT_ACTION_COMPRESSED_BF = 0, @@ -4155,19 +3897,6 @@ struct ieee80211_bss_max_idle_period_ie { u8 idle_options; } __packed; -/* BACK action code */ -enum ieee80211_back_actioncode { - WLAN_ACTION_ADDBA_REQ = 0, - WLAN_ACTION_ADDBA_RESP = 1, - WLAN_ACTION_DELBA = 2, -}; - -/* BACK (block-ack) parties */ -enum ieee80211_back_parties { - WLAN_BACK_RECIPIENT = 0, - WLAN_BACK_INITIATOR = 1, -}; - /* SA Query action */ enum ieee80211_sa_query_action { WLAN_ACTION_SA_QUERY_REQUEST = 0, @@ -5889,6 +5618,7 @@ static inline u32 ieee80211_eml_trans_timeout_in_us(u16 eml_cap) #define NAN_DEV_CAPA_NDPE_SUPPORTED 0x08 #define NAN_DEV_CAPA_S3_SUPPORTED 0x10 +#include "ieee80211-ht.h" #include "ieee80211-mesh.h" #endif /* LINUX_IEEE80211_H */ From 7cb14da1d7bbfa4a6417ed7f1bc07dd77bcd9c83 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 5 Nov 2025 15:36:51 +0100 Subject: [PATCH 0635/1381] wifi: ieee80211: split VHT definitions out The ieee80211.h file has gotten very long, continue splitting it by putting VHT definitions into a separate file. Link: https://patch.msgid.link/20251105153843.c31cb771a250.I787a13064db7d80440101de3445be17881daf1b6@changeid Signed-off-by: Johannes Berg --- include/linux/ieee80211-vht.h | 236 ++++++++++++++++++++++++++++++++++ include/linux/ieee80211.h | 216 +------------------------------ 2 files changed, 237 insertions(+), 215 deletions(-) create mode 100644 include/linux/ieee80211-vht.h diff --git a/include/linux/ieee80211-vht.h b/include/linux/ieee80211-vht.h new file mode 100644 index 000000000000..898dfb561fef --- /dev/null +++ b/include/linux/ieee80211-vht.h @@ -0,0 +1,236 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * IEEE 802.11 VHT definitions + * + * Copyright (c) 2001-2002, SSH Communications Security Corp and Jouni Malinen + * + * Copyright (c) 2002-2003, Jouni Malinen + * Copyright (c) 2005, Devicescape Software, Inc. + * Copyright (c) 2006, Michael Wu + * Copyright (c) 2013 - 2014 Intel Mobile Communications GmbH + * Copyright (c) 2016 - 2017 Intel Deutschland GmbH + * Copyright (c) 2018 - 2025 Intel Corporation + */ + +#ifndef LINUX_IEEE80211_VHT_H +#define LINUX_IEEE80211_VHT_H + +#include +#include + +#define IEEE80211_MAX_MPDU_LEN_VHT_3895 3895 +#define IEEE80211_MAX_MPDU_LEN_VHT_7991 7991 +#define IEEE80211_MAX_MPDU_LEN_VHT_11454 11454 + +/** + * enum ieee80211_vht_opmode_bits - VHT operating mode field bits + * @IEEE80211_OPMODE_NOTIF_CHANWIDTH_MASK: channel width mask + * @IEEE80211_OPMODE_NOTIF_CHANWIDTH_20MHZ: 20 MHz channel width + * @IEEE80211_OPMODE_NOTIF_CHANWIDTH_40MHZ: 40 MHz channel width + * @IEEE80211_OPMODE_NOTIF_CHANWIDTH_80MHZ: 80 MHz channel width + * @IEEE80211_OPMODE_NOTIF_CHANWIDTH_160MHZ: 160 MHz or 80+80 MHz channel width + * @IEEE80211_OPMODE_NOTIF_BW_160_80P80: 160 / 80+80 MHz indicator flag + * @IEEE80211_OPMODE_NOTIF_RX_NSS_MASK: number of spatial streams mask + * (the NSS value is the value of this field + 1) + * @IEEE80211_OPMODE_NOTIF_RX_NSS_SHIFT: number of spatial streams shift + * @IEEE80211_OPMODE_NOTIF_RX_NSS_TYPE_BF: indicates streams in SU-MIMO PPDU + * using a beamforming steering matrix + */ +enum ieee80211_vht_opmode_bits { + IEEE80211_OPMODE_NOTIF_CHANWIDTH_MASK = 0x03, + IEEE80211_OPMODE_NOTIF_CHANWIDTH_20MHZ = 0, + IEEE80211_OPMODE_NOTIF_CHANWIDTH_40MHZ = 1, + IEEE80211_OPMODE_NOTIF_CHANWIDTH_80MHZ = 2, + IEEE80211_OPMODE_NOTIF_CHANWIDTH_160MHZ = 3, + IEEE80211_OPMODE_NOTIF_BW_160_80P80 = 0x04, + IEEE80211_OPMODE_NOTIF_RX_NSS_MASK = 0x70, + IEEE80211_OPMODE_NOTIF_RX_NSS_SHIFT = 4, + IEEE80211_OPMODE_NOTIF_RX_NSS_TYPE_BF = 0x80, +}; + +/* + * Maximum length of AMPDU that the STA can receive in VHT. + * Length = 2 ^ (13 + max_ampdu_length_exp) - 1 (octets) + */ +enum ieee80211_vht_max_ampdu_length_exp { + IEEE80211_VHT_MAX_AMPDU_8K = 0, + IEEE80211_VHT_MAX_AMPDU_16K = 1, + IEEE80211_VHT_MAX_AMPDU_32K = 2, + IEEE80211_VHT_MAX_AMPDU_64K = 3, + IEEE80211_VHT_MAX_AMPDU_128K = 4, + IEEE80211_VHT_MAX_AMPDU_256K = 5, + IEEE80211_VHT_MAX_AMPDU_512K = 6, + IEEE80211_VHT_MAX_AMPDU_1024K = 7 +}; + +/** + * struct ieee80211_vht_mcs_info - VHT MCS information + * @rx_mcs_map: RX MCS map 2 bits for each stream, total 8 streams + * @rx_highest: Indicates highest long GI VHT PPDU data rate + * STA can receive. Rate expressed in units of 1 Mbps. + * If this field is 0 this value should not be used to + * consider the highest RX data rate supported. + * The top 3 bits of this field indicate the Maximum NSTS,total + * (a beamformee capability.) + * @tx_mcs_map: TX MCS map 2 bits for each stream, total 8 streams + * @tx_highest: Indicates highest long GI VHT PPDU data rate + * STA can transmit. Rate expressed in units of 1 Mbps. + * If this field is 0 this value should not be used to + * consider the highest TX data rate supported. + * The top 2 bits of this field are reserved, the + * 3rd bit from the top indiciates VHT Extended NSS BW + * Capability. + */ +struct ieee80211_vht_mcs_info { + __le16 rx_mcs_map; + __le16 rx_highest; + __le16 tx_mcs_map; + __le16 tx_highest; +} __packed; + +/* for rx_highest */ +#define IEEE80211_VHT_MAX_NSTS_TOTAL_SHIFT 13 +#define IEEE80211_VHT_MAX_NSTS_TOTAL_MASK (7 << IEEE80211_VHT_MAX_NSTS_TOTAL_SHIFT) + +/* for tx_highest */ +#define IEEE80211_VHT_EXT_NSS_BW_CAPABLE (1 << 13) + +/** + * enum ieee80211_vht_mcs_support - VHT MCS support definitions + * @IEEE80211_VHT_MCS_SUPPORT_0_7: MCSes 0-7 are supported for the + * number of streams + * @IEEE80211_VHT_MCS_SUPPORT_0_8: MCSes 0-8 are supported + * @IEEE80211_VHT_MCS_SUPPORT_0_9: MCSes 0-9 are supported + * @IEEE80211_VHT_MCS_NOT_SUPPORTED: This number of streams isn't supported + * + * These definitions are used in each 2-bit subfield of the @rx_mcs_map + * and @tx_mcs_map fields of &struct ieee80211_vht_mcs_info, which are + * both split into 8 subfields by number of streams. These values indicate + * which MCSes are supported for the number of streams the value appears + * for. + */ +enum ieee80211_vht_mcs_support { + IEEE80211_VHT_MCS_SUPPORT_0_7 = 0, + IEEE80211_VHT_MCS_SUPPORT_0_8 = 1, + IEEE80211_VHT_MCS_SUPPORT_0_9 = 2, + IEEE80211_VHT_MCS_NOT_SUPPORTED = 3, +}; + +/** + * struct ieee80211_vht_cap - VHT capabilities + * + * This structure is the "VHT capabilities element" as + * described in 802.11ac D3.0 8.4.2.160 + * @vht_cap_info: VHT capability info + * @supp_mcs: VHT MCS supported rates + */ +struct ieee80211_vht_cap { + __le32 vht_cap_info; + struct ieee80211_vht_mcs_info supp_mcs; +} __packed; + +/** + * enum ieee80211_vht_chanwidth - VHT channel width + * @IEEE80211_VHT_CHANWIDTH_USE_HT: use the HT operation IE to + * determine the channel width (20 or 40 MHz) + * @IEEE80211_VHT_CHANWIDTH_80MHZ: 80 MHz bandwidth + * @IEEE80211_VHT_CHANWIDTH_160MHZ: 160 MHz bandwidth + * @IEEE80211_VHT_CHANWIDTH_80P80MHZ: 80+80 MHz bandwidth + */ +enum ieee80211_vht_chanwidth { + IEEE80211_VHT_CHANWIDTH_USE_HT = 0, + IEEE80211_VHT_CHANWIDTH_80MHZ = 1, + IEEE80211_VHT_CHANWIDTH_160MHZ = 2, + IEEE80211_VHT_CHANWIDTH_80P80MHZ = 3, +}; + +/** + * struct ieee80211_vht_operation - VHT operation IE + * + * This structure is the "VHT operation element" as + * described in 802.11ac D3.0 8.4.2.161 + * @chan_width: Operating channel width + * @center_freq_seg0_idx: center freq segment 0 index + * @center_freq_seg1_idx: center freq segment 1 index + * @basic_mcs_set: VHT Basic MCS rate set + */ +struct ieee80211_vht_operation { + u8 chan_width; + u8 center_freq_seg0_idx; + u8 center_freq_seg1_idx; + __le16 basic_mcs_set; +} __packed; + +/* 802.11ac VHT Capabilities */ +#define IEEE80211_VHT_CAP_MAX_MPDU_LENGTH_3895 0x00000000 +#define IEEE80211_VHT_CAP_MAX_MPDU_LENGTH_7991 0x00000001 +#define IEEE80211_VHT_CAP_MAX_MPDU_LENGTH_11454 0x00000002 +#define IEEE80211_VHT_CAP_MAX_MPDU_MASK 0x00000003 +#define IEEE80211_VHT_CAP_SUPP_CHAN_WIDTH_160MHZ 0x00000004 +#define IEEE80211_VHT_CAP_SUPP_CHAN_WIDTH_160_80PLUS80MHZ 0x00000008 +#define IEEE80211_VHT_CAP_SUPP_CHAN_WIDTH_MASK 0x0000000C +#define IEEE80211_VHT_CAP_SUPP_CHAN_WIDTH_SHIFT 2 +#define IEEE80211_VHT_CAP_RXLDPC 0x00000010 +#define IEEE80211_VHT_CAP_SHORT_GI_80 0x00000020 +#define IEEE80211_VHT_CAP_SHORT_GI_160 0x00000040 +#define IEEE80211_VHT_CAP_TXSTBC 0x00000080 +#define IEEE80211_VHT_CAP_RXSTBC_1 0x00000100 +#define IEEE80211_VHT_CAP_RXSTBC_2 0x00000200 +#define IEEE80211_VHT_CAP_RXSTBC_3 0x00000300 +#define IEEE80211_VHT_CAP_RXSTBC_4 0x00000400 +#define IEEE80211_VHT_CAP_RXSTBC_MASK 0x00000700 +#define IEEE80211_VHT_CAP_RXSTBC_SHIFT 8 +#define IEEE80211_VHT_CAP_SU_BEAMFORMER_CAPABLE 0x00000800 +#define IEEE80211_VHT_CAP_SU_BEAMFORMEE_CAPABLE 0x00001000 +#define IEEE80211_VHT_CAP_BEAMFORMEE_STS_SHIFT 13 +#define IEEE80211_VHT_CAP_BEAMFORMEE_STS_MASK \ + (7 << IEEE80211_VHT_CAP_BEAMFORMEE_STS_SHIFT) +#define IEEE80211_VHT_CAP_SOUNDING_DIMENSIONS_SHIFT 16 +#define IEEE80211_VHT_CAP_SOUNDING_DIMENSIONS_MASK \ + (7 << IEEE80211_VHT_CAP_SOUNDING_DIMENSIONS_SHIFT) +#define IEEE80211_VHT_CAP_MU_BEAMFORMER_CAPABLE 0x00080000 +#define IEEE80211_VHT_CAP_MU_BEAMFORMEE_CAPABLE 0x00100000 +#define IEEE80211_VHT_CAP_VHT_TXOP_PS 0x00200000 +#define IEEE80211_VHT_CAP_HTC_VHT 0x00400000 +#define IEEE80211_VHT_CAP_MAX_A_MPDU_LENGTH_EXPONENT_SHIFT 23 +#define IEEE80211_VHT_CAP_MAX_A_MPDU_LENGTH_EXPONENT_MASK \ + (7 << IEEE80211_VHT_CAP_MAX_A_MPDU_LENGTH_EXPONENT_SHIFT) +#define IEEE80211_VHT_CAP_VHT_LINK_ADAPTATION_VHT_UNSOL_MFB 0x08000000 +#define IEEE80211_VHT_CAP_VHT_LINK_ADAPTATION_VHT_MRQ_MFB 0x0c000000 +#define IEEE80211_VHT_CAP_RX_ANTENNA_PATTERN 0x10000000 +#define IEEE80211_VHT_CAP_TX_ANTENNA_PATTERN 0x20000000 +#define IEEE80211_VHT_CAP_EXT_NSS_BW_SHIFT 30 +#define IEEE80211_VHT_CAP_EXT_NSS_BW_MASK 0xc0000000 + +/** + * ieee80211_get_vht_max_nss - return max NSS for a given bandwidth/MCS + * @cap: VHT capabilities of the peer + * @bw: bandwidth to use + * @mcs: MCS index to use + * @ext_nss_bw_capable: indicates whether or not the local transmitter + * (rate scaling algorithm) can deal with the new logic + * (dot11VHTExtendedNSSBWCapable) + * @max_vht_nss: current maximum NSS as advertised by the STA in + * operating mode notification, can be 0 in which case the + * capability data will be used to derive this (from MCS support) + * Return: The maximum NSS that can be used for the given bandwidth/MCS + * combination + * + * Due to the VHT Extended NSS Bandwidth Support, the maximum NSS can + * vary for a given BW/MCS. This function parses the data. + * + * Note: This function is exported by cfg80211. + */ +int ieee80211_get_vht_max_nss(struct ieee80211_vht_cap *cap, + enum ieee80211_vht_chanwidth bw, + int mcs, bool ext_nss_bw_capable, + unsigned int max_vht_nss); + +/* VHT action codes */ +enum ieee80211_vht_actioncode { + WLAN_VHT_ACTION_COMPRESSED_BF = 0, + WLAN_VHT_ACTION_GROUPID_MGMT = 1, + WLAN_VHT_ACTION_OPMODE_NOTIF = 2, +}; + +#endif /* LINUX_IEEE80211_VHT_H */ diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h index 0a9b4a8025cd..0b247b28c661 100644 --- a/include/linux/ieee80211.h +++ b/include/linux/ieee80211.h @@ -239,10 +239,6 @@ static inline u16 ieee80211_sn_sub(u16 sn1, u16 sn2) /* 30 byte 4 addr hdr, 2 byte QoS, 2304 byte MSDU, 12 byte crypt, 4 byte FCS */ #define IEEE80211_MAX_FRAME_LEN 2352 -#define IEEE80211_MAX_MPDU_LEN_VHT_3895 3895 -#define IEEE80211_MAX_MPDU_LEN_VHT_7991 7991 -#define IEEE80211_MAX_MPDU_LEN_VHT_11454 11454 - #define IEEE80211_MAX_SSID_LEN 32 #define IEEE80211_FIRST_TSPEC_TSID 8 @@ -988,32 +984,6 @@ struct ieee80211_tim_ie { }; } __packed; -/** - * enum ieee80211_vht_opmode_bits - VHT operating mode field bits - * @IEEE80211_OPMODE_NOTIF_CHANWIDTH_MASK: channel width mask - * @IEEE80211_OPMODE_NOTIF_CHANWIDTH_20MHZ: 20 MHz channel width - * @IEEE80211_OPMODE_NOTIF_CHANWIDTH_40MHZ: 40 MHz channel width - * @IEEE80211_OPMODE_NOTIF_CHANWIDTH_80MHZ: 80 MHz channel width - * @IEEE80211_OPMODE_NOTIF_CHANWIDTH_160MHZ: 160 MHz or 80+80 MHz channel width - * @IEEE80211_OPMODE_NOTIF_BW_160_80P80: 160 / 80+80 MHz indicator flag - * @IEEE80211_OPMODE_NOTIF_RX_NSS_MASK: number of spatial streams mask - * (the NSS value is the value of this field + 1) - * @IEEE80211_OPMODE_NOTIF_RX_NSS_SHIFT: number of spatial streams shift - * @IEEE80211_OPMODE_NOTIF_RX_NSS_TYPE_BF: indicates streams in SU-MIMO PPDU - * using a beamforming steering matrix - */ -enum ieee80211_vht_opmode_bits { - IEEE80211_OPMODE_NOTIF_CHANWIDTH_MASK = 0x03, - IEEE80211_OPMODE_NOTIF_CHANWIDTH_20MHZ = 0, - IEEE80211_OPMODE_NOTIF_CHANWIDTH_40MHZ = 1, - IEEE80211_OPMODE_NOTIF_CHANWIDTH_80MHZ = 2, - IEEE80211_OPMODE_NOTIF_CHANWIDTH_160MHZ = 3, - IEEE80211_OPMODE_NOTIF_BW_160_80P80 = 0x04, - IEEE80211_OPMODE_NOTIF_RX_NSS_MASK = 0x70, - IEEE80211_OPMODE_NOTIF_RX_NSS_SHIFT = 4, - IEEE80211_OPMODE_NOTIF_RX_NSS_TYPE_BF = 0x80, -}; - /** * enum ieee80211_s1g_chanwidth - S1G channel widths * These are defined in IEEE802.11-2016ah Table 10-20 @@ -1663,119 +1633,6 @@ struct ieee80211_p2p_noa_attr { #define IEEE80211_P2P_OPPPS_ENABLE_BIT BIT(7) #define IEEE80211_P2P_OPPPS_CTWINDOW_MASK 0x7F -/* - * Maximum length of AMPDU that the STA can receive in VHT. - * Length = 2 ^ (13 + max_ampdu_length_exp) - 1 (octets) - */ -enum ieee80211_vht_max_ampdu_length_exp { - IEEE80211_VHT_MAX_AMPDU_8K = 0, - IEEE80211_VHT_MAX_AMPDU_16K = 1, - IEEE80211_VHT_MAX_AMPDU_32K = 2, - IEEE80211_VHT_MAX_AMPDU_64K = 3, - IEEE80211_VHT_MAX_AMPDU_128K = 4, - IEEE80211_VHT_MAX_AMPDU_256K = 5, - IEEE80211_VHT_MAX_AMPDU_512K = 6, - IEEE80211_VHT_MAX_AMPDU_1024K = 7 -}; - -/** - * struct ieee80211_vht_mcs_info - VHT MCS information - * @rx_mcs_map: RX MCS map 2 bits for each stream, total 8 streams - * @rx_highest: Indicates highest long GI VHT PPDU data rate - * STA can receive. Rate expressed in units of 1 Mbps. - * If this field is 0 this value should not be used to - * consider the highest RX data rate supported. - * The top 3 bits of this field indicate the Maximum NSTS,total - * (a beamformee capability.) - * @tx_mcs_map: TX MCS map 2 bits for each stream, total 8 streams - * @tx_highest: Indicates highest long GI VHT PPDU data rate - * STA can transmit. Rate expressed in units of 1 Mbps. - * If this field is 0 this value should not be used to - * consider the highest TX data rate supported. - * The top 2 bits of this field are reserved, the - * 3rd bit from the top indiciates VHT Extended NSS BW - * Capability. - */ -struct ieee80211_vht_mcs_info { - __le16 rx_mcs_map; - __le16 rx_highest; - __le16 tx_mcs_map; - __le16 tx_highest; -} __packed; - -/* for rx_highest */ -#define IEEE80211_VHT_MAX_NSTS_TOTAL_SHIFT 13 -#define IEEE80211_VHT_MAX_NSTS_TOTAL_MASK (7 << IEEE80211_VHT_MAX_NSTS_TOTAL_SHIFT) - -/* for tx_highest */ -#define IEEE80211_VHT_EXT_NSS_BW_CAPABLE (1 << 13) - -/** - * enum ieee80211_vht_mcs_support - VHT MCS support definitions - * @IEEE80211_VHT_MCS_SUPPORT_0_7: MCSes 0-7 are supported for the - * number of streams - * @IEEE80211_VHT_MCS_SUPPORT_0_8: MCSes 0-8 are supported - * @IEEE80211_VHT_MCS_SUPPORT_0_9: MCSes 0-9 are supported - * @IEEE80211_VHT_MCS_NOT_SUPPORTED: This number of streams isn't supported - * - * These definitions are used in each 2-bit subfield of the @rx_mcs_map - * and @tx_mcs_map fields of &struct ieee80211_vht_mcs_info, which are - * both split into 8 subfields by number of streams. These values indicate - * which MCSes are supported for the number of streams the value appears - * for. - */ -enum ieee80211_vht_mcs_support { - IEEE80211_VHT_MCS_SUPPORT_0_7 = 0, - IEEE80211_VHT_MCS_SUPPORT_0_8 = 1, - IEEE80211_VHT_MCS_SUPPORT_0_9 = 2, - IEEE80211_VHT_MCS_NOT_SUPPORTED = 3, -}; - -/** - * struct ieee80211_vht_cap - VHT capabilities - * - * This structure is the "VHT capabilities element" as - * described in 802.11ac D3.0 8.4.2.160 - * @vht_cap_info: VHT capability info - * @supp_mcs: VHT MCS supported rates - */ -struct ieee80211_vht_cap { - __le32 vht_cap_info; - struct ieee80211_vht_mcs_info supp_mcs; -} __packed; - -/** - * enum ieee80211_vht_chanwidth - VHT channel width - * @IEEE80211_VHT_CHANWIDTH_USE_HT: use the HT operation IE to - * determine the channel width (20 or 40 MHz) - * @IEEE80211_VHT_CHANWIDTH_80MHZ: 80 MHz bandwidth - * @IEEE80211_VHT_CHANWIDTH_160MHZ: 160 MHz bandwidth - * @IEEE80211_VHT_CHANWIDTH_80P80MHZ: 80+80 MHz bandwidth - */ -enum ieee80211_vht_chanwidth { - IEEE80211_VHT_CHANWIDTH_USE_HT = 0, - IEEE80211_VHT_CHANWIDTH_80MHZ = 1, - IEEE80211_VHT_CHANWIDTH_160MHZ = 2, - IEEE80211_VHT_CHANWIDTH_80P80MHZ = 3, -}; - -/** - * struct ieee80211_vht_operation - VHT operation IE - * - * This structure is the "VHT operation element" as - * described in 802.11ac D3.0 8.4.2.161 - * @chan_width: Operating channel width - * @center_freq_seg0_idx: center freq segment 0 index - * @center_freq_seg1_idx: center freq segment 1 index - * @basic_mcs_set: VHT Basic MCS rate set - */ -struct ieee80211_vht_operation { - u8 chan_width; - u8 center_freq_seg0_idx; - u8 center_freq_seg1_idx; - __le16 basic_mcs_set; -} __packed; - /** * struct ieee80211_he_cap_elem - HE capabilities element * @mac_cap_info: HE MAC Capabilities Information @@ -2045,71 +1902,6 @@ struct ieee80211_eht_operation_info { u8 optional[]; } __packed; -/* 802.11ac VHT Capabilities */ -#define IEEE80211_VHT_CAP_MAX_MPDU_LENGTH_3895 0x00000000 -#define IEEE80211_VHT_CAP_MAX_MPDU_LENGTH_7991 0x00000001 -#define IEEE80211_VHT_CAP_MAX_MPDU_LENGTH_11454 0x00000002 -#define IEEE80211_VHT_CAP_MAX_MPDU_MASK 0x00000003 -#define IEEE80211_VHT_CAP_SUPP_CHAN_WIDTH_160MHZ 0x00000004 -#define IEEE80211_VHT_CAP_SUPP_CHAN_WIDTH_160_80PLUS80MHZ 0x00000008 -#define IEEE80211_VHT_CAP_SUPP_CHAN_WIDTH_MASK 0x0000000C -#define IEEE80211_VHT_CAP_SUPP_CHAN_WIDTH_SHIFT 2 -#define IEEE80211_VHT_CAP_RXLDPC 0x00000010 -#define IEEE80211_VHT_CAP_SHORT_GI_80 0x00000020 -#define IEEE80211_VHT_CAP_SHORT_GI_160 0x00000040 -#define IEEE80211_VHT_CAP_TXSTBC 0x00000080 -#define IEEE80211_VHT_CAP_RXSTBC_1 0x00000100 -#define IEEE80211_VHT_CAP_RXSTBC_2 0x00000200 -#define IEEE80211_VHT_CAP_RXSTBC_3 0x00000300 -#define IEEE80211_VHT_CAP_RXSTBC_4 0x00000400 -#define IEEE80211_VHT_CAP_RXSTBC_MASK 0x00000700 -#define IEEE80211_VHT_CAP_RXSTBC_SHIFT 8 -#define IEEE80211_VHT_CAP_SU_BEAMFORMER_CAPABLE 0x00000800 -#define IEEE80211_VHT_CAP_SU_BEAMFORMEE_CAPABLE 0x00001000 -#define IEEE80211_VHT_CAP_BEAMFORMEE_STS_SHIFT 13 -#define IEEE80211_VHT_CAP_BEAMFORMEE_STS_MASK \ - (7 << IEEE80211_VHT_CAP_BEAMFORMEE_STS_SHIFT) -#define IEEE80211_VHT_CAP_SOUNDING_DIMENSIONS_SHIFT 16 -#define IEEE80211_VHT_CAP_SOUNDING_DIMENSIONS_MASK \ - (7 << IEEE80211_VHT_CAP_SOUNDING_DIMENSIONS_SHIFT) -#define IEEE80211_VHT_CAP_MU_BEAMFORMER_CAPABLE 0x00080000 -#define IEEE80211_VHT_CAP_MU_BEAMFORMEE_CAPABLE 0x00100000 -#define IEEE80211_VHT_CAP_VHT_TXOP_PS 0x00200000 -#define IEEE80211_VHT_CAP_HTC_VHT 0x00400000 -#define IEEE80211_VHT_CAP_MAX_A_MPDU_LENGTH_EXPONENT_SHIFT 23 -#define IEEE80211_VHT_CAP_MAX_A_MPDU_LENGTH_EXPONENT_MASK \ - (7 << IEEE80211_VHT_CAP_MAX_A_MPDU_LENGTH_EXPONENT_SHIFT) -#define IEEE80211_VHT_CAP_VHT_LINK_ADAPTATION_VHT_UNSOL_MFB 0x08000000 -#define IEEE80211_VHT_CAP_VHT_LINK_ADAPTATION_VHT_MRQ_MFB 0x0c000000 -#define IEEE80211_VHT_CAP_RX_ANTENNA_PATTERN 0x10000000 -#define IEEE80211_VHT_CAP_TX_ANTENNA_PATTERN 0x20000000 -#define IEEE80211_VHT_CAP_EXT_NSS_BW_SHIFT 30 -#define IEEE80211_VHT_CAP_EXT_NSS_BW_MASK 0xc0000000 - -/** - * ieee80211_get_vht_max_nss - return max NSS for a given bandwidth/MCS - * @cap: VHT capabilities of the peer - * @bw: bandwidth to use - * @mcs: MCS index to use - * @ext_nss_bw_capable: indicates whether or not the local transmitter - * (rate scaling algorithm) can deal with the new logic - * (dot11VHTExtendedNSSBWCapable) - * @max_vht_nss: current maximum NSS as advertised by the STA in - * operating mode notification, can be 0 in which case the - * capability data will be used to derive this (from MCS support) - * Return: The maximum NSS that can be used for the given bandwidth/MCS - * combination - * - * Due to the VHT Extended NSS Bandwidth Support, the maximum NSS can - * vary for a given BW/MCS. This function parses the data. - * - * Note: This function is exported by cfg80211. - */ -int ieee80211_get_vht_max_nss(struct ieee80211_vht_cap *cap, - enum ieee80211_vht_chanwidth bw, - int mcs, bool ext_nss_bw_capable, - unsigned int max_vht_nss); - /* 802.11ax HE MAC capabilities */ #define IEEE80211_HE_MAC_CAP0_HTC_HE 0x01 #define IEEE80211_HE_MAC_CAP0_TWT_REQ 0x02 @@ -3561,13 +3353,6 @@ enum ieee80211_spectrum_mgmt_actioncode { WLAN_ACTION_SPCT_CHL_SWITCH = 4, }; -/* VHT action codes */ -enum ieee80211_vht_actioncode { - WLAN_VHT_ACTION_COMPRESSED_BF = 0, - WLAN_VHT_ACTION_GROUPID_MGMT = 1, - WLAN_VHT_ACTION_OPMODE_NOTIF = 2, -}; - /* Self Protected Action codes */ enum ieee80211_self_protected_actioncode { WLAN_SP_RESERVED = 0, @@ -5619,6 +5404,7 @@ static inline u32 ieee80211_eml_trans_timeout_in_us(u16 eml_cap) #define NAN_DEV_CAPA_S3_SUPPORTED 0x10 #include "ieee80211-ht.h" +#include "ieee80211-vht.h" #include "ieee80211-mesh.h" #endif /* LINUX_IEEE80211_H */ From 02a2cf302557eb59794bba0b05d6755f44928d78 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 5 Nov 2025 15:36:52 +0100 Subject: [PATCH 0636/1381] wifi: ieee80211: split HE definitions out The ieee80211.h file has gotten very long, continue splitting it by putting HE definitions into a separate file. Link: https://patch.msgid.link/20251105153843.6998c0802104.I3dd7cfea6abbd118b999ecdedd48437d39cb0533@changeid Signed-off-by: Johannes Berg --- include/linux/ieee80211-he.h | 824 +++++++++++++++++++++++++++++++++++ include/linux/ieee80211.h | 806 +--------------------------------- 2 files changed, 827 insertions(+), 803 deletions(-) create mode 100644 include/linux/ieee80211-he.h diff --git a/include/linux/ieee80211-he.h b/include/linux/ieee80211-he.h new file mode 100644 index 000000000000..904d50db5bb8 --- /dev/null +++ b/include/linux/ieee80211-he.h @@ -0,0 +1,824 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * IEEE 802.11 HE definitions + * + * Copyright (c) 2001-2002, SSH Communications Security Corp and Jouni Malinen + * + * Copyright (c) 2002-2003, Jouni Malinen + * Copyright (c) 2005, Devicescape Software, Inc. + * Copyright (c) 2006, Michael Wu + * Copyright (c) 2013 - 2014 Intel Mobile Communications GmbH + * Copyright (c) 2016 - 2017 Intel Deutschland GmbH + * Copyright (c) 2018 - 2025 Intel Corporation + */ + +#ifndef LINUX_IEEE80211_HE_H +#define LINUX_IEEE80211_HE_H + +#include +#include + +#define IEEE80211_TWT_CONTROL_NDP BIT(0) +#define IEEE80211_TWT_CONTROL_RESP_MODE BIT(1) +#define IEEE80211_TWT_CONTROL_NEG_TYPE_BROADCAST BIT(3) +#define IEEE80211_TWT_CONTROL_RX_DISABLED BIT(4) +#define IEEE80211_TWT_CONTROL_WAKE_DUR_UNIT BIT(5) + +#define IEEE80211_TWT_REQTYPE_REQUEST BIT(0) +#define IEEE80211_TWT_REQTYPE_SETUP_CMD GENMASK(3, 1) +#define IEEE80211_TWT_REQTYPE_TRIGGER BIT(4) +#define IEEE80211_TWT_REQTYPE_IMPLICIT BIT(5) +#define IEEE80211_TWT_REQTYPE_FLOWTYPE BIT(6) +#define IEEE80211_TWT_REQTYPE_FLOWID GENMASK(9, 7) +#define IEEE80211_TWT_REQTYPE_WAKE_INT_EXP GENMASK(14, 10) +#define IEEE80211_TWT_REQTYPE_PROTECTION BIT(15) + +enum ieee80211_twt_setup_cmd { + TWT_SETUP_CMD_REQUEST, + TWT_SETUP_CMD_SUGGEST, + TWT_SETUP_CMD_DEMAND, + TWT_SETUP_CMD_GROUPING, + TWT_SETUP_CMD_ACCEPT, + TWT_SETUP_CMD_ALTERNATE, + TWT_SETUP_CMD_DICTATE, + TWT_SETUP_CMD_REJECT, +}; + +struct ieee80211_twt_params { + __le16 req_type; + __le64 twt; + u8 min_twt_dur; + __le16 mantissa; + u8 channel; +} __packed; + +struct ieee80211_twt_setup { + u8 dialog_token; + u8 element_id; + u8 length; + u8 control; + u8 params[]; +} __packed; + +/** + * struct ieee80211_he_cap_elem - HE capabilities element + * @mac_cap_info: HE MAC Capabilities Information + * @phy_cap_info: HE PHY Capabilities Information + * + * This structure represents the fixed fields of the payload of the + * "HE capabilities element" as described in IEEE Std 802.11ax-2021 + * sections 9.4.2.248.2 and 9.4.2.248.3. + */ +struct ieee80211_he_cap_elem { + u8 mac_cap_info[6]; + u8 phy_cap_info[11]; +} __packed; + +#define IEEE80211_TX_RX_MCS_NSS_DESC_MAX_LEN 5 + +/** + * enum ieee80211_he_mcs_support - HE MCS support definitions + * @IEEE80211_HE_MCS_SUPPORT_0_7: MCSes 0-7 are supported for the + * number of streams + * @IEEE80211_HE_MCS_SUPPORT_0_9: MCSes 0-9 are supported + * @IEEE80211_HE_MCS_SUPPORT_0_11: MCSes 0-11 are supported + * @IEEE80211_HE_MCS_NOT_SUPPORTED: This number of streams isn't supported + * + * These definitions are used in each 2-bit subfield of the rx_mcs_* + * and tx_mcs_* fields of &struct ieee80211_he_mcs_nss_supp, which are + * both split into 8 subfields by number of streams. These values indicate + * which MCSes are supported for the number of streams the value appears + * for. + */ +enum ieee80211_he_mcs_support { + IEEE80211_HE_MCS_SUPPORT_0_7 = 0, + IEEE80211_HE_MCS_SUPPORT_0_9 = 1, + IEEE80211_HE_MCS_SUPPORT_0_11 = 2, + IEEE80211_HE_MCS_NOT_SUPPORTED = 3, +}; + +/** + * struct ieee80211_he_mcs_nss_supp - HE Tx/Rx HE MCS NSS Support Field + * + * This structure holds the data required for the Tx/Rx HE MCS NSS Support Field + * described in P802.11ax_D2.0 section 9.4.2.237.4 + * + * @rx_mcs_80: Rx MCS map 2 bits for each stream, total 8 streams, for channel + * widths less than 80MHz. + * @tx_mcs_80: Tx MCS map 2 bits for each stream, total 8 streams, for channel + * widths less than 80MHz. + * @rx_mcs_160: Rx MCS map 2 bits for each stream, total 8 streams, for channel + * width 160MHz. + * @tx_mcs_160: Tx MCS map 2 bits for each stream, total 8 streams, for channel + * width 160MHz. + * @rx_mcs_80p80: Rx MCS map 2 bits for each stream, total 8 streams, for + * channel width 80p80MHz. + * @tx_mcs_80p80: Tx MCS map 2 bits for each stream, total 8 streams, for + * channel width 80p80MHz. + */ +struct ieee80211_he_mcs_nss_supp { + __le16 rx_mcs_80; + __le16 tx_mcs_80; + __le16 rx_mcs_160; + __le16 tx_mcs_160; + __le16 rx_mcs_80p80; + __le16 tx_mcs_80p80; +} __packed; + +/** + * struct ieee80211_he_operation - HE Operation element + * @he_oper_params: HE Operation Parameters + BSS Color Information + * @he_mcs_nss_set: Basic HE-MCS And NSS Set + * @optional: Optional fields VHT Operation Information, Max Co-Hosted + * BSSID Indicator, and 6 GHz Operation Information + * + * This structure represents the payload of the "HE Operation + * element" as described in IEEE Std 802.11ax-2021 section 9.4.2.249. + */ +struct ieee80211_he_operation { + __le32 he_oper_params; + __le16 he_mcs_nss_set; + u8 optional[]; +} __packed; + +/** + * struct ieee80211_he_spr - Spatial Reuse Parameter Set element + * @he_sr_control: SR Control + * @optional: Optional fields Non-SRG OBSS PD Max Offset, SRG OBSS PD + * Min Offset, SRG OBSS PD Max Offset, SRG BSS Color + * Bitmap, and SRG Partial BSSID Bitmap + * + * This structure represents the payload of the "Spatial Reuse + * Parameter Set element" as described in IEEE Std 802.11ax-2021 + * section 9.4.2.252. + */ +struct ieee80211_he_spr { + u8 he_sr_control; + u8 optional[]; +} __packed; + +/** + * struct ieee80211_he_mu_edca_param_ac_rec - MU AC Parameter Record field + * @aifsn: ACI/AIFSN + * @ecw_min_max: ECWmin/ECWmax + * @mu_edca_timer: MU EDCA Timer + * + * This structure represents the "MU AC Parameter Record" as described + * in IEEE Std 802.11ax-2021 section 9.4.2.251, Figure 9-788p. + */ +struct ieee80211_he_mu_edca_param_ac_rec { + u8 aifsn; + u8 ecw_min_max; + u8 mu_edca_timer; +} __packed; + +/** + * struct ieee80211_mu_edca_param_set - MU EDCA Parameter Set element + * @mu_qos_info: QoS Info + * @ac_be: MU AC_BE Parameter Record + * @ac_bk: MU AC_BK Parameter Record + * @ac_vi: MU AC_VI Parameter Record + * @ac_vo: MU AC_VO Parameter Record + * + * This structure represents the payload of the "MU EDCA Parameter Set + * element" as described in IEEE Std 802.11ax-2021 section 9.4.2.251. + */ +struct ieee80211_mu_edca_param_set { + u8 mu_qos_info; + struct ieee80211_he_mu_edca_param_ac_rec ac_be; + struct ieee80211_he_mu_edca_param_ac_rec ac_bk; + struct ieee80211_he_mu_edca_param_ac_rec ac_vi; + struct ieee80211_he_mu_edca_param_ac_rec ac_vo; +} __packed; + +/* 802.11ax HE MAC capabilities */ +#define IEEE80211_HE_MAC_CAP0_HTC_HE 0x01 +#define IEEE80211_HE_MAC_CAP0_TWT_REQ 0x02 +#define IEEE80211_HE_MAC_CAP0_TWT_RES 0x04 +#define IEEE80211_HE_MAC_CAP0_DYNAMIC_FRAG_NOT_SUPP 0x00 +#define IEEE80211_HE_MAC_CAP0_DYNAMIC_FRAG_LEVEL_1 0x08 +#define IEEE80211_HE_MAC_CAP0_DYNAMIC_FRAG_LEVEL_2 0x10 +#define IEEE80211_HE_MAC_CAP0_DYNAMIC_FRAG_LEVEL_3 0x18 +#define IEEE80211_HE_MAC_CAP0_DYNAMIC_FRAG_MASK 0x18 +#define IEEE80211_HE_MAC_CAP0_MAX_NUM_FRAG_MSDU_1 0x00 +#define IEEE80211_HE_MAC_CAP0_MAX_NUM_FRAG_MSDU_2 0x20 +#define IEEE80211_HE_MAC_CAP0_MAX_NUM_FRAG_MSDU_4 0x40 +#define IEEE80211_HE_MAC_CAP0_MAX_NUM_FRAG_MSDU_8 0x60 +#define IEEE80211_HE_MAC_CAP0_MAX_NUM_FRAG_MSDU_16 0x80 +#define IEEE80211_HE_MAC_CAP0_MAX_NUM_FRAG_MSDU_32 0xa0 +#define IEEE80211_HE_MAC_CAP0_MAX_NUM_FRAG_MSDU_64 0xc0 +#define IEEE80211_HE_MAC_CAP0_MAX_NUM_FRAG_MSDU_UNLIMITED 0xe0 +#define IEEE80211_HE_MAC_CAP0_MAX_NUM_FRAG_MSDU_MASK 0xe0 + +#define IEEE80211_HE_MAC_CAP1_MIN_FRAG_SIZE_UNLIMITED 0x00 +#define IEEE80211_HE_MAC_CAP1_MIN_FRAG_SIZE_128 0x01 +#define IEEE80211_HE_MAC_CAP1_MIN_FRAG_SIZE_256 0x02 +#define IEEE80211_HE_MAC_CAP1_MIN_FRAG_SIZE_512 0x03 +#define IEEE80211_HE_MAC_CAP1_MIN_FRAG_SIZE_MASK 0x03 +#define IEEE80211_HE_MAC_CAP1_TF_MAC_PAD_DUR_0US 0x00 +#define IEEE80211_HE_MAC_CAP1_TF_MAC_PAD_DUR_8US 0x04 +#define IEEE80211_HE_MAC_CAP1_TF_MAC_PAD_DUR_16US 0x08 +#define IEEE80211_HE_MAC_CAP1_TF_MAC_PAD_DUR_MASK 0x0c +#define IEEE80211_HE_MAC_CAP1_MULTI_TID_AGG_RX_QOS_1 0x00 +#define IEEE80211_HE_MAC_CAP1_MULTI_TID_AGG_RX_QOS_2 0x10 +#define IEEE80211_HE_MAC_CAP1_MULTI_TID_AGG_RX_QOS_3 0x20 +#define IEEE80211_HE_MAC_CAP1_MULTI_TID_AGG_RX_QOS_4 0x30 +#define IEEE80211_HE_MAC_CAP1_MULTI_TID_AGG_RX_QOS_5 0x40 +#define IEEE80211_HE_MAC_CAP1_MULTI_TID_AGG_RX_QOS_6 0x50 +#define IEEE80211_HE_MAC_CAP1_MULTI_TID_AGG_RX_QOS_7 0x60 +#define IEEE80211_HE_MAC_CAP1_MULTI_TID_AGG_RX_QOS_8 0x70 +#define IEEE80211_HE_MAC_CAP1_MULTI_TID_AGG_RX_QOS_MASK 0x70 + +/* Link adaptation is split between byte HE_MAC_CAP1 and + * HE_MAC_CAP2. It should be set only if IEEE80211_HE_MAC_CAP0_HTC_HE + * in which case the following values apply: + * 0 = No feedback. + * 1 = reserved. + * 2 = Unsolicited feedback. + * 3 = both + */ +#define IEEE80211_HE_MAC_CAP1_LINK_ADAPTATION 0x80 + +#define IEEE80211_HE_MAC_CAP2_LINK_ADAPTATION 0x01 +#define IEEE80211_HE_MAC_CAP2_ALL_ACK 0x02 +#define IEEE80211_HE_MAC_CAP2_TRS 0x04 +#define IEEE80211_HE_MAC_CAP2_BSR 0x08 +#define IEEE80211_HE_MAC_CAP2_BCAST_TWT 0x10 +#define IEEE80211_HE_MAC_CAP2_32BIT_BA_BITMAP 0x20 +#define IEEE80211_HE_MAC_CAP2_MU_CASCADING 0x40 +#define IEEE80211_HE_MAC_CAP2_ACK_EN 0x80 + +#define IEEE80211_HE_MAC_CAP3_OMI_CONTROL 0x02 +#define IEEE80211_HE_MAC_CAP3_OFDMA_RA 0x04 + +/* The maximum length of an A-MDPU is defined by the combination of the Maximum + * A-MDPU Length Exponent field in the HT capabilities, VHT capabilities and the + * same field in the HE capabilities. + */ +#define IEEE80211_HE_MAC_CAP3_MAX_AMPDU_LEN_EXP_EXT_0 0x00 +#define IEEE80211_HE_MAC_CAP3_MAX_AMPDU_LEN_EXP_EXT_1 0x08 +#define IEEE80211_HE_MAC_CAP3_MAX_AMPDU_LEN_EXP_EXT_2 0x10 +#define IEEE80211_HE_MAC_CAP3_MAX_AMPDU_LEN_EXP_EXT_3 0x18 +#define IEEE80211_HE_MAC_CAP3_MAX_AMPDU_LEN_EXP_MASK 0x18 +#define IEEE80211_HE_MAC_CAP3_AMSDU_FRAG 0x20 +#define IEEE80211_HE_MAC_CAP3_FLEX_TWT_SCHED 0x40 +#define IEEE80211_HE_MAC_CAP3_RX_CTRL_FRAME_TO_MULTIBSS 0x80 + +#define IEEE80211_HE_MAC_CAP4_BSRP_BQRP_A_MPDU_AGG 0x01 +#define IEEE80211_HE_MAC_CAP4_QTP 0x02 +#define IEEE80211_HE_MAC_CAP4_BQR 0x04 +#define IEEE80211_HE_MAC_CAP4_PSR_RESP 0x08 +#define IEEE80211_HE_MAC_CAP4_NDP_FB_REP 0x10 +#define IEEE80211_HE_MAC_CAP4_OPS 0x20 +#define IEEE80211_HE_MAC_CAP4_AMSDU_IN_AMPDU 0x40 +/* Multi TID agg TX is split between byte #4 and #5 + * The value is a combination of B39,B40,B41 + */ +#define IEEE80211_HE_MAC_CAP4_MULTI_TID_AGG_TX_QOS_B39 0x80 + +#define IEEE80211_HE_MAC_CAP5_MULTI_TID_AGG_TX_QOS_B40 0x01 +#define IEEE80211_HE_MAC_CAP5_MULTI_TID_AGG_TX_QOS_B41 0x02 +#define IEEE80211_HE_MAC_CAP5_SUBCHAN_SELECTIVE_TRANSMISSION 0x04 +#define IEEE80211_HE_MAC_CAP5_UL_2x996_TONE_RU 0x08 +#define IEEE80211_HE_MAC_CAP5_OM_CTRL_UL_MU_DATA_DIS_RX 0x10 +#define IEEE80211_HE_MAC_CAP5_HE_DYNAMIC_SM_PS 0x20 +#define IEEE80211_HE_MAC_CAP5_PUNCTURED_SOUNDING 0x40 +#define IEEE80211_HE_MAC_CAP5_HT_VHT_TRIG_FRAME_RX 0x80 + +#define IEEE80211_HE_VHT_MAX_AMPDU_FACTOR 20 +#define IEEE80211_HE_HT_MAX_AMPDU_FACTOR 16 +#define IEEE80211_HE_6GHZ_MAX_AMPDU_FACTOR 13 + +/* 802.11ax HE PHY capabilities */ +#define IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_40MHZ_IN_2G 0x02 +#define IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_40MHZ_80MHZ_IN_5G 0x04 +#define IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_160MHZ_IN_5G 0x08 +#define IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_80PLUS80_MHZ_IN_5G 0x10 +#define IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_MASK_ALL 0x1e + +#define IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_RU_MAPPING_IN_2G 0x20 +#define IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_RU_MAPPING_IN_5G 0x40 +#define IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_MASK 0xfe + +#define IEEE80211_HE_PHY_CAP1_PREAMBLE_PUNC_RX_80MHZ_ONLY_SECOND_20MHZ 0x01 +#define IEEE80211_HE_PHY_CAP1_PREAMBLE_PUNC_RX_80MHZ_ONLY_SECOND_40MHZ 0x02 +#define IEEE80211_HE_PHY_CAP1_PREAMBLE_PUNC_RX_160MHZ_ONLY_SECOND_20MHZ 0x04 +#define IEEE80211_HE_PHY_CAP1_PREAMBLE_PUNC_RX_160MHZ_ONLY_SECOND_40MHZ 0x08 +#define IEEE80211_HE_PHY_CAP1_PREAMBLE_PUNC_RX_MASK 0x0f +#define IEEE80211_HE_PHY_CAP1_DEVICE_CLASS_A 0x10 +#define IEEE80211_HE_PHY_CAP1_LDPC_CODING_IN_PAYLOAD 0x20 +#define IEEE80211_HE_PHY_CAP1_HE_LTF_AND_GI_FOR_HE_PPDUS_0_8US 0x40 +/* Midamble RX/TX Max NSTS is split between byte #2 and byte #3 */ +#define IEEE80211_HE_PHY_CAP1_MIDAMBLE_RX_TX_MAX_NSTS 0x80 + +#define IEEE80211_HE_PHY_CAP2_MIDAMBLE_RX_TX_MAX_NSTS 0x01 +#define IEEE80211_HE_PHY_CAP2_NDP_4x_LTF_AND_3_2US 0x02 +#define IEEE80211_HE_PHY_CAP2_STBC_TX_UNDER_80MHZ 0x04 +#define IEEE80211_HE_PHY_CAP2_STBC_RX_UNDER_80MHZ 0x08 +#define IEEE80211_HE_PHY_CAP2_DOPPLER_TX 0x10 +#define IEEE80211_HE_PHY_CAP2_DOPPLER_RX 0x20 + +/* Note that the meaning of UL MU below is different between an AP and a non-AP + * sta, where in the AP case it indicates support for Rx and in the non-AP sta + * case it indicates support for Tx. + */ +#define IEEE80211_HE_PHY_CAP2_UL_MU_FULL_MU_MIMO 0x40 +#define IEEE80211_HE_PHY_CAP2_UL_MU_PARTIAL_MU_MIMO 0x80 + +#define IEEE80211_HE_PHY_CAP3_DCM_MAX_CONST_TX_NO_DCM 0x00 +#define IEEE80211_HE_PHY_CAP3_DCM_MAX_CONST_TX_BPSK 0x01 +#define IEEE80211_HE_PHY_CAP3_DCM_MAX_CONST_TX_QPSK 0x02 +#define IEEE80211_HE_PHY_CAP3_DCM_MAX_CONST_TX_16_QAM 0x03 +#define IEEE80211_HE_PHY_CAP3_DCM_MAX_CONST_TX_MASK 0x03 +#define IEEE80211_HE_PHY_CAP3_DCM_MAX_TX_NSS_1 0x00 +#define IEEE80211_HE_PHY_CAP3_DCM_MAX_TX_NSS_2 0x04 +#define IEEE80211_HE_PHY_CAP3_DCM_MAX_CONST_RX_NO_DCM 0x00 +#define IEEE80211_HE_PHY_CAP3_DCM_MAX_CONST_RX_BPSK 0x08 +#define IEEE80211_HE_PHY_CAP3_DCM_MAX_CONST_RX_QPSK 0x10 +#define IEEE80211_HE_PHY_CAP3_DCM_MAX_CONST_RX_16_QAM 0x18 +#define IEEE80211_HE_PHY_CAP3_DCM_MAX_CONST_RX_MASK 0x18 +#define IEEE80211_HE_PHY_CAP3_DCM_MAX_RX_NSS_1 0x00 +#define IEEE80211_HE_PHY_CAP3_DCM_MAX_RX_NSS_2 0x20 +#define IEEE80211_HE_PHY_CAP3_RX_PARTIAL_BW_SU_IN_20MHZ_MU 0x40 +#define IEEE80211_HE_PHY_CAP3_SU_BEAMFORMER 0x80 + +#define IEEE80211_HE_PHY_CAP4_SU_BEAMFORMEE 0x01 +#define IEEE80211_HE_PHY_CAP4_MU_BEAMFORMER 0x02 + +/* Minimal allowed value of Max STS under 80MHz is 3 */ +#define IEEE80211_HE_PHY_CAP4_BEAMFORMEE_MAX_STS_UNDER_80MHZ_4 0x0c +#define IEEE80211_HE_PHY_CAP4_BEAMFORMEE_MAX_STS_UNDER_80MHZ_5 0x10 +#define IEEE80211_HE_PHY_CAP4_BEAMFORMEE_MAX_STS_UNDER_80MHZ_6 0x14 +#define IEEE80211_HE_PHY_CAP4_BEAMFORMEE_MAX_STS_UNDER_80MHZ_7 0x18 +#define IEEE80211_HE_PHY_CAP4_BEAMFORMEE_MAX_STS_UNDER_80MHZ_8 0x1c +#define IEEE80211_HE_PHY_CAP4_BEAMFORMEE_MAX_STS_UNDER_80MHZ_MASK 0x1c + +/* Minimal allowed value of Max STS above 80MHz is 3 */ +#define IEEE80211_HE_PHY_CAP4_BEAMFORMEE_MAX_STS_ABOVE_80MHZ_4 0x60 +#define IEEE80211_HE_PHY_CAP4_BEAMFORMEE_MAX_STS_ABOVE_80MHZ_5 0x80 +#define IEEE80211_HE_PHY_CAP4_BEAMFORMEE_MAX_STS_ABOVE_80MHZ_6 0xa0 +#define IEEE80211_HE_PHY_CAP4_BEAMFORMEE_MAX_STS_ABOVE_80MHZ_7 0xc0 +#define IEEE80211_HE_PHY_CAP4_BEAMFORMEE_MAX_STS_ABOVE_80MHZ_8 0xe0 +#define IEEE80211_HE_PHY_CAP4_BEAMFORMEE_MAX_STS_ABOVE_80MHZ_MASK 0xe0 + +#define IEEE80211_HE_PHY_CAP5_BEAMFORMEE_NUM_SND_DIM_UNDER_80MHZ_1 0x00 +#define IEEE80211_HE_PHY_CAP5_BEAMFORMEE_NUM_SND_DIM_UNDER_80MHZ_2 0x01 +#define IEEE80211_HE_PHY_CAP5_BEAMFORMEE_NUM_SND_DIM_UNDER_80MHZ_3 0x02 +#define IEEE80211_HE_PHY_CAP5_BEAMFORMEE_NUM_SND_DIM_UNDER_80MHZ_4 0x03 +#define IEEE80211_HE_PHY_CAP5_BEAMFORMEE_NUM_SND_DIM_UNDER_80MHZ_5 0x04 +#define IEEE80211_HE_PHY_CAP5_BEAMFORMEE_NUM_SND_DIM_UNDER_80MHZ_6 0x05 +#define IEEE80211_HE_PHY_CAP5_BEAMFORMEE_NUM_SND_DIM_UNDER_80MHZ_7 0x06 +#define IEEE80211_HE_PHY_CAP5_BEAMFORMEE_NUM_SND_DIM_UNDER_80MHZ_8 0x07 +#define IEEE80211_HE_PHY_CAP5_BEAMFORMEE_NUM_SND_DIM_UNDER_80MHZ_MASK 0x07 + +#define IEEE80211_HE_PHY_CAP5_BEAMFORMEE_NUM_SND_DIM_ABOVE_80MHZ_1 0x00 +#define IEEE80211_HE_PHY_CAP5_BEAMFORMEE_NUM_SND_DIM_ABOVE_80MHZ_2 0x08 +#define IEEE80211_HE_PHY_CAP5_BEAMFORMEE_NUM_SND_DIM_ABOVE_80MHZ_3 0x10 +#define IEEE80211_HE_PHY_CAP5_BEAMFORMEE_NUM_SND_DIM_ABOVE_80MHZ_4 0x18 +#define IEEE80211_HE_PHY_CAP5_BEAMFORMEE_NUM_SND_DIM_ABOVE_80MHZ_5 0x20 +#define IEEE80211_HE_PHY_CAP5_BEAMFORMEE_NUM_SND_DIM_ABOVE_80MHZ_6 0x28 +#define IEEE80211_HE_PHY_CAP5_BEAMFORMEE_NUM_SND_DIM_ABOVE_80MHZ_7 0x30 +#define IEEE80211_HE_PHY_CAP5_BEAMFORMEE_NUM_SND_DIM_ABOVE_80MHZ_8 0x38 +#define IEEE80211_HE_PHY_CAP5_BEAMFORMEE_NUM_SND_DIM_ABOVE_80MHZ_MASK 0x38 + +#define IEEE80211_HE_PHY_CAP5_NG16_SU_FEEDBACK 0x40 +#define IEEE80211_HE_PHY_CAP5_NG16_MU_FEEDBACK 0x80 + +#define IEEE80211_HE_PHY_CAP6_CODEBOOK_SIZE_42_SU 0x01 +#define IEEE80211_HE_PHY_CAP6_CODEBOOK_SIZE_75_MU 0x02 +#define IEEE80211_HE_PHY_CAP6_TRIG_SU_BEAMFORMING_FB 0x04 +#define IEEE80211_HE_PHY_CAP6_TRIG_MU_BEAMFORMING_PARTIAL_BW_FB 0x08 +#define IEEE80211_HE_PHY_CAP6_TRIG_CQI_FB 0x10 +#define IEEE80211_HE_PHY_CAP6_PARTIAL_BW_EXT_RANGE 0x20 +#define IEEE80211_HE_PHY_CAP6_PARTIAL_BANDWIDTH_DL_MUMIMO 0x40 +#define IEEE80211_HE_PHY_CAP6_PPE_THRESHOLD_PRESENT 0x80 + +#define IEEE80211_HE_PHY_CAP7_PSR_BASED_SR 0x01 +#define IEEE80211_HE_PHY_CAP7_POWER_BOOST_FACTOR_SUPP 0x02 +#define IEEE80211_HE_PHY_CAP7_HE_SU_MU_PPDU_4XLTF_AND_08_US_GI 0x04 +#define IEEE80211_HE_PHY_CAP7_MAX_NC_1 0x08 +#define IEEE80211_HE_PHY_CAP7_MAX_NC_2 0x10 +#define IEEE80211_HE_PHY_CAP7_MAX_NC_3 0x18 +#define IEEE80211_HE_PHY_CAP7_MAX_NC_4 0x20 +#define IEEE80211_HE_PHY_CAP7_MAX_NC_5 0x28 +#define IEEE80211_HE_PHY_CAP7_MAX_NC_6 0x30 +#define IEEE80211_HE_PHY_CAP7_MAX_NC_7 0x38 +#define IEEE80211_HE_PHY_CAP7_MAX_NC_MASK 0x38 +#define IEEE80211_HE_PHY_CAP7_STBC_TX_ABOVE_80MHZ 0x40 +#define IEEE80211_HE_PHY_CAP7_STBC_RX_ABOVE_80MHZ 0x80 + +#define IEEE80211_HE_PHY_CAP8_HE_ER_SU_PPDU_4XLTF_AND_08_US_GI 0x01 +#define IEEE80211_HE_PHY_CAP8_20MHZ_IN_40MHZ_HE_PPDU_IN_2G 0x02 +#define IEEE80211_HE_PHY_CAP8_20MHZ_IN_160MHZ_HE_PPDU 0x04 +#define IEEE80211_HE_PHY_CAP8_80MHZ_IN_160MHZ_HE_PPDU 0x08 +#define IEEE80211_HE_PHY_CAP8_HE_ER_SU_1XLTF_AND_08_US_GI 0x10 +#define IEEE80211_HE_PHY_CAP8_MIDAMBLE_RX_TX_2X_AND_1XLTF 0x20 +#define IEEE80211_HE_PHY_CAP8_DCM_MAX_RU_242 0x00 +#define IEEE80211_HE_PHY_CAP8_DCM_MAX_RU_484 0x40 +#define IEEE80211_HE_PHY_CAP8_DCM_MAX_RU_996 0x80 +#define IEEE80211_HE_PHY_CAP8_DCM_MAX_RU_2x996 0xc0 +#define IEEE80211_HE_PHY_CAP8_DCM_MAX_RU_MASK 0xc0 + +#define IEEE80211_HE_PHY_CAP9_LONGER_THAN_16_SIGB_OFDM_SYM 0x01 +#define IEEE80211_HE_PHY_CAP9_NON_TRIGGERED_CQI_FEEDBACK 0x02 +#define IEEE80211_HE_PHY_CAP9_TX_1024_QAM_LESS_THAN_242_TONE_RU 0x04 +#define IEEE80211_HE_PHY_CAP9_RX_1024_QAM_LESS_THAN_242_TONE_RU 0x08 +#define IEEE80211_HE_PHY_CAP9_RX_FULL_BW_SU_USING_MU_WITH_COMP_SIGB 0x10 +#define IEEE80211_HE_PHY_CAP9_RX_FULL_BW_SU_USING_MU_WITH_NON_COMP_SIGB 0x20 +#define IEEE80211_HE_PHY_CAP9_NOMINAL_PKT_PADDING_0US 0x0 +#define IEEE80211_HE_PHY_CAP9_NOMINAL_PKT_PADDING_8US 0x1 +#define IEEE80211_HE_PHY_CAP9_NOMINAL_PKT_PADDING_16US 0x2 +#define IEEE80211_HE_PHY_CAP9_NOMINAL_PKT_PADDING_RESERVED 0x3 +#define IEEE80211_HE_PHY_CAP9_NOMINAL_PKT_PADDING_POS 6 +#define IEEE80211_HE_PHY_CAP9_NOMINAL_PKT_PADDING_MASK 0xc0 + +#define IEEE80211_HE_PHY_CAP10_HE_MU_M1RU_MAX_LTF 0x01 + +/* 802.11ax HE TX/RX MCS NSS Support */ +#define IEEE80211_TX_RX_MCS_NSS_SUPP_HIGHEST_MCS_POS (3) +#define IEEE80211_TX_RX_MCS_NSS_SUPP_TX_BITMAP_POS (6) +#define IEEE80211_TX_RX_MCS_NSS_SUPP_RX_BITMAP_POS (11) +#define IEEE80211_TX_RX_MCS_NSS_SUPP_TX_BITMAP_MASK 0x07c0 +#define IEEE80211_TX_RX_MCS_NSS_SUPP_RX_BITMAP_MASK 0xf800 + +/* TX/RX HE MCS Support field Highest MCS subfield encoding */ +enum ieee80211_he_highest_mcs_supported_subfield_enc { + HIGHEST_MCS_SUPPORTED_MCS7 = 0, + HIGHEST_MCS_SUPPORTED_MCS8, + HIGHEST_MCS_SUPPORTED_MCS9, + HIGHEST_MCS_SUPPORTED_MCS10, + HIGHEST_MCS_SUPPORTED_MCS11, +}; + +/* Calculate 802.11ax HE capabilities IE Tx/Rx HE MCS NSS Support Field size */ +static inline u8 +ieee80211_he_mcs_nss_size(const struct ieee80211_he_cap_elem *he_cap) +{ + u8 count = 4; + + if (he_cap->phy_cap_info[0] & + IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_160MHZ_IN_5G) + count += 4; + + if (he_cap->phy_cap_info[0] & + IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_80PLUS80_MHZ_IN_5G) + count += 4; + + return count; +} + +/* 802.11ax HE PPE Thresholds */ +#define IEEE80211_PPE_THRES_NSS_SUPPORT_2NSS (1) +#define IEEE80211_PPE_THRES_NSS_POS (0) +#define IEEE80211_PPE_THRES_NSS_MASK (7) +#define IEEE80211_PPE_THRES_RU_INDEX_BITMASK_2x966_AND_966_RU \ + (BIT(5) | BIT(6)) +#define IEEE80211_PPE_THRES_RU_INDEX_BITMASK_MASK 0x78 +#define IEEE80211_PPE_THRES_RU_INDEX_BITMASK_POS (3) +#define IEEE80211_PPE_THRES_INFO_PPET_SIZE (3) +#define IEEE80211_HE_PPE_THRES_INFO_HEADER_SIZE (7) + +/* + * Calculate 802.11ax HE capabilities IE PPE field size + * Input: Header byte of ppe_thres (first byte), and HE capa IE's PHY cap u8* + */ +static inline u8 +ieee80211_he_ppe_size(u8 ppe_thres_hdr, const u8 *phy_cap_info) +{ + u8 n; + + if ((phy_cap_info[6] & + IEEE80211_HE_PHY_CAP6_PPE_THRESHOLD_PRESENT) == 0) + return 0; + + n = hweight8(ppe_thres_hdr & + IEEE80211_PPE_THRES_RU_INDEX_BITMASK_MASK); + n *= (1 + ((ppe_thres_hdr & IEEE80211_PPE_THRES_NSS_MASK) >> + IEEE80211_PPE_THRES_NSS_POS)); + + /* + * Each pair is 6 bits, and we need to add the 7 "header" bits to the + * total size. + */ + n = (n * IEEE80211_PPE_THRES_INFO_PPET_SIZE * 2) + 7; + n = DIV_ROUND_UP(n, 8); + + return n; +} + +static inline bool ieee80211_he_capa_size_ok(const u8 *data, u8 len) +{ + const struct ieee80211_he_cap_elem *he_cap_ie_elem = (const void *)data; + u8 needed = sizeof(*he_cap_ie_elem); + + if (len < needed) + return false; + + needed += ieee80211_he_mcs_nss_size(he_cap_ie_elem); + if (len < needed) + return false; + + if (he_cap_ie_elem->phy_cap_info[6] & + IEEE80211_HE_PHY_CAP6_PPE_THRESHOLD_PRESENT) { + if (len < needed + 1) + return false; + needed += ieee80211_he_ppe_size(data[needed], + he_cap_ie_elem->phy_cap_info); + } + + return len >= needed; +} + +/* HE Operation defines */ +#define IEEE80211_HE_OPERATION_DFLT_PE_DURATION_MASK 0x00000007 +#define IEEE80211_HE_OPERATION_TWT_REQUIRED 0x00000008 +#define IEEE80211_HE_OPERATION_RTS_THRESHOLD_MASK 0x00003ff0 +#define IEEE80211_HE_OPERATION_RTS_THRESHOLD_OFFSET 4 +#define IEEE80211_HE_OPERATION_VHT_OPER_INFO 0x00004000 +#define IEEE80211_HE_OPERATION_CO_HOSTED_BSS 0x00008000 +#define IEEE80211_HE_OPERATION_ER_SU_DISABLE 0x00010000 +#define IEEE80211_HE_OPERATION_6GHZ_OP_INFO 0x00020000 +#define IEEE80211_HE_OPERATION_BSS_COLOR_MASK 0x3f000000 +#define IEEE80211_HE_OPERATION_BSS_COLOR_OFFSET 24 +#define IEEE80211_HE_OPERATION_PARTIAL_BSS_COLOR 0x40000000 +#define IEEE80211_HE_OPERATION_BSS_COLOR_DISABLED 0x80000000 + +#define IEEE80211_6GHZ_CTRL_REG_LPI_AP 0 +#define IEEE80211_6GHZ_CTRL_REG_SP_AP 1 +#define IEEE80211_6GHZ_CTRL_REG_VLP_AP 2 +#define IEEE80211_6GHZ_CTRL_REG_INDOOR_LPI_AP 3 +#define IEEE80211_6GHZ_CTRL_REG_INDOOR_SP_AP_OLD 4 +#define IEEE80211_6GHZ_CTRL_REG_INDOOR_SP_AP 8 + +/** + * struct ieee80211_he_6ghz_oper - HE 6 GHz operation Information field + * @primary: primary channel + * @control: control flags + * @ccfs0: channel center frequency segment 0 + * @ccfs1: channel center frequency segment 1 + * @minrate: minimum rate (in 1 Mbps units) + */ +struct ieee80211_he_6ghz_oper { + u8 primary; +#define IEEE80211_HE_6GHZ_OPER_CTRL_CHANWIDTH 0x3 +#define IEEE80211_HE_6GHZ_OPER_CTRL_CHANWIDTH_20MHZ 0 +#define IEEE80211_HE_6GHZ_OPER_CTRL_CHANWIDTH_40MHZ 1 +#define IEEE80211_HE_6GHZ_OPER_CTRL_CHANWIDTH_80MHZ 2 +#define IEEE80211_HE_6GHZ_OPER_CTRL_CHANWIDTH_160MHZ 3 +#define IEEE80211_HE_6GHZ_OPER_CTRL_DUP_BEACON 0x4 +#define IEEE80211_HE_6GHZ_OPER_CTRL_REG_INFO 0x78 + u8 control; + u8 ccfs0; + u8 ccfs1; + u8 minrate; +} __packed; + +/** + * enum ieee80211_reg_conn_bits - represents Regulatory connectivity field bits. + * + * This enumeration defines bit flags used to represent regulatory connectivity + * field bits. + * + * @IEEE80211_REG_CONN_LPI_VALID: Indicates whether the LPI bit is valid. + * @IEEE80211_REG_CONN_LPI_VALUE: Represents the value of the LPI bit. + * @IEEE80211_REG_CONN_SP_VALID: Indicates whether the SP bit is valid. + * @IEEE80211_REG_CONN_SP_VALUE: Represents the value of the SP bit. + */ +enum ieee80211_reg_conn_bits { + IEEE80211_REG_CONN_LPI_VALID = BIT(0), + IEEE80211_REG_CONN_LPI_VALUE = BIT(1), + IEEE80211_REG_CONN_SP_VALID = BIT(2), + IEEE80211_REG_CONN_SP_VALUE = BIT(3), +}; + +/* transmit power interpretation type of transmit power envelope element */ +enum ieee80211_tx_power_intrpt_type { + IEEE80211_TPE_LOCAL_EIRP, + IEEE80211_TPE_LOCAL_EIRP_PSD, + IEEE80211_TPE_REG_CLIENT_EIRP, + IEEE80211_TPE_REG_CLIENT_EIRP_PSD, +}; + +/* category type of transmit power envelope element */ +enum ieee80211_tx_power_category_6ghz { + IEEE80211_TPE_CAT_6GHZ_DEFAULT = 0, + IEEE80211_TPE_CAT_6GHZ_SUBORDINATE = 1, +}; + +/* + * For IEEE80211_TPE_LOCAL_EIRP / IEEE80211_TPE_REG_CLIENT_EIRP, + * setting to 63.5 dBm means no constraint. + */ +#define IEEE80211_TPE_MAX_TX_PWR_NO_CONSTRAINT 127 + +/* + * For IEEE80211_TPE_LOCAL_EIRP_PSD / IEEE80211_TPE_REG_CLIENT_EIRP_PSD, + * setting to 127 indicates no PSD limit for the 20 MHz channel. + */ +#define IEEE80211_TPE_PSD_NO_LIMIT 127 + +/** + * struct ieee80211_tx_pwr_env - Transmit Power Envelope + * @info: Transmit Power Information field + * @variable: Maximum Transmit Power field + * + * This structure represents the payload of the "Transmit Power + * Envelope element" as described in IEEE Std 802.11ax-2021 section + * 9.4.2.161 + */ +struct ieee80211_tx_pwr_env { + u8 info; + u8 variable[]; +} __packed; + +#define IEEE80211_TX_PWR_ENV_INFO_COUNT 0x7 +#define IEEE80211_TX_PWR_ENV_INFO_INTERPRET 0x38 +#define IEEE80211_TX_PWR_ENV_INFO_CATEGORY 0xC0 + +#define IEEE80211_TX_PWR_ENV_EXT_COUNT 0xF + +static inline bool ieee80211_valid_tpe_element(const u8 *data, u8 len) +{ + const struct ieee80211_tx_pwr_env *env = (const void *)data; + u8 count, interpret, category; + u8 needed = sizeof(*env); + u8 N; /* also called N in the spec */ + + if (len < needed) + return false; + + count = u8_get_bits(env->info, IEEE80211_TX_PWR_ENV_INFO_COUNT); + interpret = u8_get_bits(env->info, IEEE80211_TX_PWR_ENV_INFO_INTERPRET); + category = u8_get_bits(env->info, IEEE80211_TX_PWR_ENV_INFO_CATEGORY); + + switch (category) { + case IEEE80211_TPE_CAT_6GHZ_DEFAULT: + case IEEE80211_TPE_CAT_6GHZ_SUBORDINATE: + break; + default: + return false; + } + + switch (interpret) { + case IEEE80211_TPE_LOCAL_EIRP: + case IEEE80211_TPE_REG_CLIENT_EIRP: + if (count > 3) + return false; + + /* count == 0 encodes 1 value for 20 MHz, etc. */ + needed += count + 1; + + if (len < needed) + return false; + + /* there can be extension fields not accounted for in 'count' */ + + return true; + case IEEE80211_TPE_LOCAL_EIRP_PSD: + case IEEE80211_TPE_REG_CLIENT_EIRP_PSD: + if (count > 4) + return false; + + N = count ? 1 << (count - 1) : 1; + needed += N; + + if (len < needed) + return false; + + if (len > needed) { + u8 K = u8_get_bits(env->variable[N], + IEEE80211_TX_PWR_ENV_EXT_COUNT); + + needed += 1 + K; + if (len < needed) + return false; + } + + return true; + } + + return false; +} + +/* + * ieee80211_he_oper_size - calculate 802.11ax HE Operations IE size + * @he_oper_ie: byte data of the He Operations IE, stating from the byte + * after the ext ID byte. It is assumed that he_oper_ie has at least + * sizeof(struct ieee80211_he_operation) bytes, the caller must have + * validated this. + * @return the actual size of the IE data (not including header), or 0 on error + */ +static inline u8 +ieee80211_he_oper_size(const u8 *he_oper_ie) +{ + const struct ieee80211_he_operation *he_oper = (const void *)he_oper_ie; + u8 oper_len = sizeof(struct ieee80211_he_operation); + u32 he_oper_params; + + /* Make sure the input is not NULL */ + if (!he_oper_ie) + return 0; + + /* Calc required length */ + he_oper_params = le32_to_cpu(he_oper->he_oper_params); + if (he_oper_params & IEEE80211_HE_OPERATION_VHT_OPER_INFO) + oper_len += 3; + if (he_oper_params & IEEE80211_HE_OPERATION_CO_HOSTED_BSS) + oper_len++; + if (he_oper_params & IEEE80211_HE_OPERATION_6GHZ_OP_INFO) + oper_len += sizeof(struct ieee80211_he_6ghz_oper); + + /* Add the first byte (extension ID) to the total length */ + oper_len++; + + return oper_len; +} + +/** + * ieee80211_he_6ghz_oper - obtain 6 GHz operation field + * @he_oper: HE operation element (must be pre-validated for size) + * but may be %NULL + * + * Return: a pointer to the 6 GHz operation field, or %NULL + */ +static inline const struct ieee80211_he_6ghz_oper * +ieee80211_he_6ghz_oper(const struct ieee80211_he_operation *he_oper) +{ + const u8 *ret; + u32 he_oper_params; + + if (!he_oper) + return NULL; + + ret = (const void *)&he_oper->optional; + + he_oper_params = le32_to_cpu(he_oper->he_oper_params); + + if (!(he_oper_params & IEEE80211_HE_OPERATION_6GHZ_OP_INFO)) + return NULL; + if (he_oper_params & IEEE80211_HE_OPERATION_VHT_OPER_INFO) + ret += 3; + if (he_oper_params & IEEE80211_HE_OPERATION_CO_HOSTED_BSS) + ret++; + + return (const void *)ret; +} + +/* HE Spatial Reuse defines */ +#define IEEE80211_HE_SPR_PSR_DISALLOWED BIT(0) +#define IEEE80211_HE_SPR_NON_SRG_OBSS_PD_SR_DISALLOWED BIT(1) +#define IEEE80211_HE_SPR_NON_SRG_OFFSET_PRESENT BIT(2) +#define IEEE80211_HE_SPR_SRG_INFORMATION_PRESENT BIT(3) +#define IEEE80211_HE_SPR_HESIGA_SR_VAL15_ALLOWED BIT(4) + +/* + * ieee80211_he_spr_size - calculate 802.11ax HE Spatial Reuse IE size + * @he_spr_ie: byte data of the He Spatial Reuse IE, stating from the byte + * after the ext ID byte. It is assumed that he_spr_ie has at least + * sizeof(struct ieee80211_he_spr) bytes, the caller must have validated + * this + * @return the actual size of the IE data (not including header), or 0 on error + */ +static inline u8 +ieee80211_he_spr_size(const u8 *he_spr_ie) +{ + const struct ieee80211_he_spr *he_spr = (const void *)he_spr_ie; + u8 spr_len = sizeof(struct ieee80211_he_spr); + u8 he_spr_params; + + /* Make sure the input is not NULL */ + if (!he_spr_ie) + return 0; + + /* Calc required length */ + he_spr_params = he_spr->he_sr_control; + if (he_spr_params & IEEE80211_HE_SPR_NON_SRG_OFFSET_PRESENT) + spr_len++; + if (he_spr_params & IEEE80211_HE_SPR_SRG_INFORMATION_PRESENT) + spr_len += 18; + + /* Add the first byte (extension ID) to the total length */ + spr_len++; + + return spr_len; +} + +struct ieee80211_he_6ghz_capa { + /* uses IEEE80211_HE_6GHZ_CAP_* below */ + __le16 capa; +} __packed; + +/* HE 6 GHz band capabilities */ +/* uses enum ieee80211_min_mpdu_spacing values */ +#define IEEE80211_HE_6GHZ_CAP_MIN_MPDU_START 0x0007 +/* uses enum ieee80211_vht_max_ampdu_length_exp values */ +#define IEEE80211_HE_6GHZ_CAP_MAX_AMPDU_LEN_EXP 0x0038 +/* uses IEEE80211_VHT_CAP_MAX_MPDU_LENGTH_* values */ +#define IEEE80211_HE_6GHZ_CAP_MAX_MPDU_LEN 0x00c0 +/* WLAN_HT_CAP_SM_PS_* values */ +#define IEEE80211_HE_6GHZ_CAP_SM_PS 0x0600 +#define IEEE80211_HE_6GHZ_CAP_RD_RESPONDER 0x0800 +#define IEEE80211_HE_6GHZ_CAP_RX_ANTPAT_CONS 0x1000 +#define IEEE80211_HE_6GHZ_CAP_TX_ANTPAT_CONS 0x2000 + +#endif /* LINUX_IEEE80211_HE_H */ diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h index 0b247b28c661..a3dbbcee00ee 100644 --- a/include/linux/ieee80211.h +++ b/include/linux/ieee80211.h @@ -1141,48 +1141,6 @@ ieee80211_s1g_optional_len(__le16 fc) return len; } -#define IEEE80211_TWT_CONTROL_NDP BIT(0) -#define IEEE80211_TWT_CONTROL_RESP_MODE BIT(1) -#define IEEE80211_TWT_CONTROL_NEG_TYPE_BROADCAST BIT(3) -#define IEEE80211_TWT_CONTROL_RX_DISABLED BIT(4) -#define IEEE80211_TWT_CONTROL_WAKE_DUR_UNIT BIT(5) - -#define IEEE80211_TWT_REQTYPE_REQUEST BIT(0) -#define IEEE80211_TWT_REQTYPE_SETUP_CMD GENMASK(3, 1) -#define IEEE80211_TWT_REQTYPE_TRIGGER BIT(4) -#define IEEE80211_TWT_REQTYPE_IMPLICIT BIT(5) -#define IEEE80211_TWT_REQTYPE_FLOWTYPE BIT(6) -#define IEEE80211_TWT_REQTYPE_FLOWID GENMASK(9, 7) -#define IEEE80211_TWT_REQTYPE_WAKE_INT_EXP GENMASK(14, 10) -#define IEEE80211_TWT_REQTYPE_PROTECTION BIT(15) - -enum ieee80211_twt_setup_cmd { - TWT_SETUP_CMD_REQUEST, - TWT_SETUP_CMD_SUGGEST, - TWT_SETUP_CMD_DEMAND, - TWT_SETUP_CMD_GROUPING, - TWT_SETUP_CMD_ACCEPT, - TWT_SETUP_CMD_ALTERNATE, - TWT_SETUP_CMD_DICTATE, - TWT_SETUP_CMD_REJECT, -}; - -struct ieee80211_twt_params { - __le16 req_type; - __le64 twt; - u8 min_twt_dur; - __le16 mantissa; - u8 channel; -} __packed; - -struct ieee80211_twt_setup { - u8 dialog_token; - u8 element_id; - u8 length; - u8 control; - u8 params[]; -} __packed; - #define IEEE80211_TTLM_MAX_CNT 2 #define IEEE80211_TTLM_CONTROL_DIRECTION 0x03 #define IEEE80211_TTLM_CONTROL_DEF_LINK_MAP 0x04 @@ -1633,137 +1591,6 @@ struct ieee80211_p2p_noa_attr { #define IEEE80211_P2P_OPPPS_ENABLE_BIT BIT(7) #define IEEE80211_P2P_OPPPS_CTWINDOW_MASK 0x7F -/** - * struct ieee80211_he_cap_elem - HE capabilities element - * @mac_cap_info: HE MAC Capabilities Information - * @phy_cap_info: HE PHY Capabilities Information - * - * This structure represents the fixed fields of the payload of the - * "HE capabilities element" as described in IEEE Std 802.11ax-2021 - * sections 9.4.2.248.2 and 9.4.2.248.3. - */ -struct ieee80211_he_cap_elem { - u8 mac_cap_info[6]; - u8 phy_cap_info[11]; -} __packed; - -#define IEEE80211_TX_RX_MCS_NSS_DESC_MAX_LEN 5 - -/** - * enum ieee80211_he_mcs_support - HE MCS support definitions - * @IEEE80211_HE_MCS_SUPPORT_0_7: MCSes 0-7 are supported for the - * number of streams - * @IEEE80211_HE_MCS_SUPPORT_0_9: MCSes 0-9 are supported - * @IEEE80211_HE_MCS_SUPPORT_0_11: MCSes 0-11 are supported - * @IEEE80211_HE_MCS_NOT_SUPPORTED: This number of streams isn't supported - * - * These definitions are used in each 2-bit subfield of the rx_mcs_* - * and tx_mcs_* fields of &struct ieee80211_he_mcs_nss_supp, which are - * both split into 8 subfields by number of streams. These values indicate - * which MCSes are supported for the number of streams the value appears - * for. - */ -enum ieee80211_he_mcs_support { - IEEE80211_HE_MCS_SUPPORT_0_7 = 0, - IEEE80211_HE_MCS_SUPPORT_0_9 = 1, - IEEE80211_HE_MCS_SUPPORT_0_11 = 2, - IEEE80211_HE_MCS_NOT_SUPPORTED = 3, -}; - -/** - * struct ieee80211_he_mcs_nss_supp - HE Tx/Rx HE MCS NSS Support Field - * - * This structure holds the data required for the Tx/Rx HE MCS NSS Support Field - * described in P802.11ax_D2.0 section 9.4.2.237.4 - * - * @rx_mcs_80: Rx MCS map 2 bits for each stream, total 8 streams, for channel - * widths less than 80MHz. - * @tx_mcs_80: Tx MCS map 2 bits for each stream, total 8 streams, for channel - * widths less than 80MHz. - * @rx_mcs_160: Rx MCS map 2 bits for each stream, total 8 streams, for channel - * width 160MHz. - * @tx_mcs_160: Tx MCS map 2 bits for each stream, total 8 streams, for channel - * width 160MHz. - * @rx_mcs_80p80: Rx MCS map 2 bits for each stream, total 8 streams, for - * channel width 80p80MHz. - * @tx_mcs_80p80: Tx MCS map 2 bits for each stream, total 8 streams, for - * channel width 80p80MHz. - */ -struct ieee80211_he_mcs_nss_supp { - __le16 rx_mcs_80; - __le16 tx_mcs_80; - __le16 rx_mcs_160; - __le16 tx_mcs_160; - __le16 rx_mcs_80p80; - __le16 tx_mcs_80p80; -} __packed; - -/** - * struct ieee80211_he_operation - HE Operation element - * @he_oper_params: HE Operation Parameters + BSS Color Information - * @he_mcs_nss_set: Basic HE-MCS And NSS Set - * @optional: Optional fields VHT Operation Information, Max Co-Hosted - * BSSID Indicator, and 6 GHz Operation Information - * - * This structure represents the payload of the "HE Operation - * element" as described in IEEE Std 802.11ax-2021 section 9.4.2.249. - */ -struct ieee80211_he_operation { - __le32 he_oper_params; - __le16 he_mcs_nss_set; - u8 optional[]; -} __packed; - -/** - * struct ieee80211_he_spr - Spatial Reuse Parameter Set element - * @he_sr_control: SR Control - * @optional: Optional fields Non-SRG OBSS PD Max Offset, SRG OBSS PD - * Min Offset, SRG OBSS PD Max Offset, SRG BSS Color - * Bitmap, and SRG Partial BSSID Bitmap - * - * This structure represents the payload of the "Spatial Reuse - * Parameter Set element" as described in IEEE Std 802.11ax-2021 - * section 9.4.2.252. - */ -struct ieee80211_he_spr { - u8 he_sr_control; - u8 optional[]; -} __packed; - -/** - * struct ieee80211_he_mu_edca_param_ac_rec - MU AC Parameter Record field - * @aifsn: ACI/AIFSN - * @ecw_min_max: ECWmin/ECWmax - * @mu_edca_timer: MU EDCA Timer - * - * This structure represents the "MU AC Parameter Record" as described - * in IEEE Std 802.11ax-2021 section 9.4.2.251, Figure 9-788p. - */ -struct ieee80211_he_mu_edca_param_ac_rec { - u8 aifsn; - u8 ecw_min_max; - u8 mu_edca_timer; -} __packed; - -/** - * struct ieee80211_mu_edca_param_set - MU EDCA Parameter Set element - * @mu_qos_info: QoS Info - * @ac_be: MU AC_BE Parameter Record - * @ac_bk: MU AC_BK Parameter Record - * @ac_vi: MU AC_VI Parameter Record - * @ac_vo: MU AC_VO Parameter Record - * - * This structure represents the payload of the "MU EDCA Parameter Set - * element" as described in IEEE Std 802.11ax-2021 section 9.4.2.251. - */ -struct ieee80211_mu_edca_param_set { - u8 mu_qos_info; - struct ieee80211_he_mu_edca_param_ac_rec ac_be; - struct ieee80211_he_mu_edca_param_ac_rec ac_bk; - struct ieee80211_he_mu_edca_param_ac_rec ac_vi; - struct ieee80211_he_mu_edca_param_ac_rec ac_vo; -} __packed; - #define IEEE80211_EHT_MCS_NSS_RX 0x0f #define IEEE80211_EHT_MCS_NSS_TX 0xf0 @@ -1902,618 +1729,6 @@ struct ieee80211_eht_operation_info { u8 optional[]; } __packed; -/* 802.11ax HE MAC capabilities */ -#define IEEE80211_HE_MAC_CAP0_HTC_HE 0x01 -#define IEEE80211_HE_MAC_CAP0_TWT_REQ 0x02 -#define IEEE80211_HE_MAC_CAP0_TWT_RES 0x04 -#define IEEE80211_HE_MAC_CAP0_DYNAMIC_FRAG_NOT_SUPP 0x00 -#define IEEE80211_HE_MAC_CAP0_DYNAMIC_FRAG_LEVEL_1 0x08 -#define IEEE80211_HE_MAC_CAP0_DYNAMIC_FRAG_LEVEL_2 0x10 -#define IEEE80211_HE_MAC_CAP0_DYNAMIC_FRAG_LEVEL_3 0x18 -#define IEEE80211_HE_MAC_CAP0_DYNAMIC_FRAG_MASK 0x18 -#define IEEE80211_HE_MAC_CAP0_MAX_NUM_FRAG_MSDU_1 0x00 -#define IEEE80211_HE_MAC_CAP0_MAX_NUM_FRAG_MSDU_2 0x20 -#define IEEE80211_HE_MAC_CAP0_MAX_NUM_FRAG_MSDU_4 0x40 -#define IEEE80211_HE_MAC_CAP0_MAX_NUM_FRAG_MSDU_8 0x60 -#define IEEE80211_HE_MAC_CAP0_MAX_NUM_FRAG_MSDU_16 0x80 -#define IEEE80211_HE_MAC_CAP0_MAX_NUM_FRAG_MSDU_32 0xa0 -#define IEEE80211_HE_MAC_CAP0_MAX_NUM_FRAG_MSDU_64 0xc0 -#define IEEE80211_HE_MAC_CAP0_MAX_NUM_FRAG_MSDU_UNLIMITED 0xe0 -#define IEEE80211_HE_MAC_CAP0_MAX_NUM_FRAG_MSDU_MASK 0xe0 - -#define IEEE80211_HE_MAC_CAP1_MIN_FRAG_SIZE_UNLIMITED 0x00 -#define IEEE80211_HE_MAC_CAP1_MIN_FRAG_SIZE_128 0x01 -#define IEEE80211_HE_MAC_CAP1_MIN_FRAG_SIZE_256 0x02 -#define IEEE80211_HE_MAC_CAP1_MIN_FRAG_SIZE_512 0x03 -#define IEEE80211_HE_MAC_CAP1_MIN_FRAG_SIZE_MASK 0x03 -#define IEEE80211_HE_MAC_CAP1_TF_MAC_PAD_DUR_0US 0x00 -#define IEEE80211_HE_MAC_CAP1_TF_MAC_PAD_DUR_8US 0x04 -#define IEEE80211_HE_MAC_CAP1_TF_MAC_PAD_DUR_16US 0x08 -#define IEEE80211_HE_MAC_CAP1_TF_MAC_PAD_DUR_MASK 0x0c -#define IEEE80211_HE_MAC_CAP1_MULTI_TID_AGG_RX_QOS_1 0x00 -#define IEEE80211_HE_MAC_CAP1_MULTI_TID_AGG_RX_QOS_2 0x10 -#define IEEE80211_HE_MAC_CAP1_MULTI_TID_AGG_RX_QOS_3 0x20 -#define IEEE80211_HE_MAC_CAP1_MULTI_TID_AGG_RX_QOS_4 0x30 -#define IEEE80211_HE_MAC_CAP1_MULTI_TID_AGG_RX_QOS_5 0x40 -#define IEEE80211_HE_MAC_CAP1_MULTI_TID_AGG_RX_QOS_6 0x50 -#define IEEE80211_HE_MAC_CAP1_MULTI_TID_AGG_RX_QOS_7 0x60 -#define IEEE80211_HE_MAC_CAP1_MULTI_TID_AGG_RX_QOS_8 0x70 -#define IEEE80211_HE_MAC_CAP1_MULTI_TID_AGG_RX_QOS_MASK 0x70 - -/* Link adaptation is split between byte HE_MAC_CAP1 and - * HE_MAC_CAP2. It should be set only if IEEE80211_HE_MAC_CAP0_HTC_HE - * in which case the following values apply: - * 0 = No feedback. - * 1 = reserved. - * 2 = Unsolicited feedback. - * 3 = both - */ -#define IEEE80211_HE_MAC_CAP1_LINK_ADAPTATION 0x80 - -#define IEEE80211_HE_MAC_CAP2_LINK_ADAPTATION 0x01 -#define IEEE80211_HE_MAC_CAP2_ALL_ACK 0x02 -#define IEEE80211_HE_MAC_CAP2_TRS 0x04 -#define IEEE80211_HE_MAC_CAP2_BSR 0x08 -#define IEEE80211_HE_MAC_CAP2_BCAST_TWT 0x10 -#define IEEE80211_HE_MAC_CAP2_32BIT_BA_BITMAP 0x20 -#define IEEE80211_HE_MAC_CAP2_MU_CASCADING 0x40 -#define IEEE80211_HE_MAC_CAP2_ACK_EN 0x80 - -#define IEEE80211_HE_MAC_CAP3_OMI_CONTROL 0x02 -#define IEEE80211_HE_MAC_CAP3_OFDMA_RA 0x04 - -/* The maximum length of an A-MDPU is defined by the combination of the Maximum - * A-MDPU Length Exponent field in the HT capabilities, VHT capabilities and the - * same field in the HE capabilities. - */ -#define IEEE80211_HE_MAC_CAP3_MAX_AMPDU_LEN_EXP_EXT_0 0x00 -#define IEEE80211_HE_MAC_CAP3_MAX_AMPDU_LEN_EXP_EXT_1 0x08 -#define IEEE80211_HE_MAC_CAP3_MAX_AMPDU_LEN_EXP_EXT_2 0x10 -#define IEEE80211_HE_MAC_CAP3_MAX_AMPDU_LEN_EXP_EXT_3 0x18 -#define IEEE80211_HE_MAC_CAP3_MAX_AMPDU_LEN_EXP_MASK 0x18 -#define IEEE80211_HE_MAC_CAP3_AMSDU_FRAG 0x20 -#define IEEE80211_HE_MAC_CAP3_FLEX_TWT_SCHED 0x40 -#define IEEE80211_HE_MAC_CAP3_RX_CTRL_FRAME_TO_MULTIBSS 0x80 - -#define IEEE80211_HE_MAC_CAP4_BSRP_BQRP_A_MPDU_AGG 0x01 -#define IEEE80211_HE_MAC_CAP4_QTP 0x02 -#define IEEE80211_HE_MAC_CAP4_BQR 0x04 -#define IEEE80211_HE_MAC_CAP4_PSR_RESP 0x08 -#define IEEE80211_HE_MAC_CAP4_NDP_FB_REP 0x10 -#define IEEE80211_HE_MAC_CAP4_OPS 0x20 -#define IEEE80211_HE_MAC_CAP4_AMSDU_IN_AMPDU 0x40 -/* Multi TID agg TX is split between byte #4 and #5 - * The value is a combination of B39,B40,B41 - */ -#define IEEE80211_HE_MAC_CAP4_MULTI_TID_AGG_TX_QOS_B39 0x80 - -#define IEEE80211_HE_MAC_CAP5_MULTI_TID_AGG_TX_QOS_B40 0x01 -#define IEEE80211_HE_MAC_CAP5_MULTI_TID_AGG_TX_QOS_B41 0x02 -#define IEEE80211_HE_MAC_CAP5_SUBCHAN_SELECTIVE_TRANSMISSION 0x04 -#define IEEE80211_HE_MAC_CAP5_UL_2x996_TONE_RU 0x08 -#define IEEE80211_HE_MAC_CAP5_OM_CTRL_UL_MU_DATA_DIS_RX 0x10 -#define IEEE80211_HE_MAC_CAP5_HE_DYNAMIC_SM_PS 0x20 -#define IEEE80211_HE_MAC_CAP5_PUNCTURED_SOUNDING 0x40 -#define IEEE80211_HE_MAC_CAP5_HT_VHT_TRIG_FRAME_RX 0x80 - -#define IEEE80211_HE_VHT_MAX_AMPDU_FACTOR 20 -#define IEEE80211_HE_HT_MAX_AMPDU_FACTOR 16 -#define IEEE80211_HE_6GHZ_MAX_AMPDU_FACTOR 13 - -/* 802.11ax HE PHY capabilities */ -#define IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_40MHZ_IN_2G 0x02 -#define IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_40MHZ_80MHZ_IN_5G 0x04 -#define IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_160MHZ_IN_5G 0x08 -#define IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_80PLUS80_MHZ_IN_5G 0x10 -#define IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_MASK_ALL 0x1e - -#define IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_RU_MAPPING_IN_2G 0x20 -#define IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_RU_MAPPING_IN_5G 0x40 -#define IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_MASK 0xfe - -#define IEEE80211_HE_PHY_CAP1_PREAMBLE_PUNC_RX_80MHZ_ONLY_SECOND_20MHZ 0x01 -#define IEEE80211_HE_PHY_CAP1_PREAMBLE_PUNC_RX_80MHZ_ONLY_SECOND_40MHZ 0x02 -#define IEEE80211_HE_PHY_CAP1_PREAMBLE_PUNC_RX_160MHZ_ONLY_SECOND_20MHZ 0x04 -#define IEEE80211_HE_PHY_CAP1_PREAMBLE_PUNC_RX_160MHZ_ONLY_SECOND_40MHZ 0x08 -#define IEEE80211_HE_PHY_CAP1_PREAMBLE_PUNC_RX_MASK 0x0f -#define IEEE80211_HE_PHY_CAP1_DEVICE_CLASS_A 0x10 -#define IEEE80211_HE_PHY_CAP1_LDPC_CODING_IN_PAYLOAD 0x20 -#define IEEE80211_HE_PHY_CAP1_HE_LTF_AND_GI_FOR_HE_PPDUS_0_8US 0x40 -/* Midamble RX/TX Max NSTS is split between byte #2 and byte #3 */ -#define IEEE80211_HE_PHY_CAP1_MIDAMBLE_RX_TX_MAX_NSTS 0x80 - -#define IEEE80211_HE_PHY_CAP2_MIDAMBLE_RX_TX_MAX_NSTS 0x01 -#define IEEE80211_HE_PHY_CAP2_NDP_4x_LTF_AND_3_2US 0x02 -#define IEEE80211_HE_PHY_CAP2_STBC_TX_UNDER_80MHZ 0x04 -#define IEEE80211_HE_PHY_CAP2_STBC_RX_UNDER_80MHZ 0x08 -#define IEEE80211_HE_PHY_CAP2_DOPPLER_TX 0x10 -#define IEEE80211_HE_PHY_CAP2_DOPPLER_RX 0x20 - -/* Note that the meaning of UL MU below is different between an AP and a non-AP - * sta, where in the AP case it indicates support for Rx and in the non-AP sta - * case it indicates support for Tx. - */ -#define IEEE80211_HE_PHY_CAP2_UL_MU_FULL_MU_MIMO 0x40 -#define IEEE80211_HE_PHY_CAP2_UL_MU_PARTIAL_MU_MIMO 0x80 - -#define IEEE80211_HE_PHY_CAP3_DCM_MAX_CONST_TX_NO_DCM 0x00 -#define IEEE80211_HE_PHY_CAP3_DCM_MAX_CONST_TX_BPSK 0x01 -#define IEEE80211_HE_PHY_CAP3_DCM_MAX_CONST_TX_QPSK 0x02 -#define IEEE80211_HE_PHY_CAP3_DCM_MAX_CONST_TX_16_QAM 0x03 -#define IEEE80211_HE_PHY_CAP3_DCM_MAX_CONST_TX_MASK 0x03 -#define IEEE80211_HE_PHY_CAP3_DCM_MAX_TX_NSS_1 0x00 -#define IEEE80211_HE_PHY_CAP3_DCM_MAX_TX_NSS_2 0x04 -#define IEEE80211_HE_PHY_CAP3_DCM_MAX_CONST_RX_NO_DCM 0x00 -#define IEEE80211_HE_PHY_CAP3_DCM_MAX_CONST_RX_BPSK 0x08 -#define IEEE80211_HE_PHY_CAP3_DCM_MAX_CONST_RX_QPSK 0x10 -#define IEEE80211_HE_PHY_CAP3_DCM_MAX_CONST_RX_16_QAM 0x18 -#define IEEE80211_HE_PHY_CAP3_DCM_MAX_CONST_RX_MASK 0x18 -#define IEEE80211_HE_PHY_CAP3_DCM_MAX_RX_NSS_1 0x00 -#define IEEE80211_HE_PHY_CAP3_DCM_MAX_RX_NSS_2 0x20 -#define IEEE80211_HE_PHY_CAP3_RX_PARTIAL_BW_SU_IN_20MHZ_MU 0x40 -#define IEEE80211_HE_PHY_CAP3_SU_BEAMFORMER 0x80 - -#define IEEE80211_HE_PHY_CAP4_SU_BEAMFORMEE 0x01 -#define IEEE80211_HE_PHY_CAP4_MU_BEAMFORMER 0x02 - -/* Minimal allowed value of Max STS under 80MHz is 3 */ -#define IEEE80211_HE_PHY_CAP4_BEAMFORMEE_MAX_STS_UNDER_80MHZ_4 0x0c -#define IEEE80211_HE_PHY_CAP4_BEAMFORMEE_MAX_STS_UNDER_80MHZ_5 0x10 -#define IEEE80211_HE_PHY_CAP4_BEAMFORMEE_MAX_STS_UNDER_80MHZ_6 0x14 -#define IEEE80211_HE_PHY_CAP4_BEAMFORMEE_MAX_STS_UNDER_80MHZ_7 0x18 -#define IEEE80211_HE_PHY_CAP4_BEAMFORMEE_MAX_STS_UNDER_80MHZ_8 0x1c -#define IEEE80211_HE_PHY_CAP4_BEAMFORMEE_MAX_STS_UNDER_80MHZ_MASK 0x1c - -/* Minimal allowed value of Max STS above 80MHz is 3 */ -#define IEEE80211_HE_PHY_CAP4_BEAMFORMEE_MAX_STS_ABOVE_80MHZ_4 0x60 -#define IEEE80211_HE_PHY_CAP4_BEAMFORMEE_MAX_STS_ABOVE_80MHZ_5 0x80 -#define IEEE80211_HE_PHY_CAP4_BEAMFORMEE_MAX_STS_ABOVE_80MHZ_6 0xa0 -#define IEEE80211_HE_PHY_CAP4_BEAMFORMEE_MAX_STS_ABOVE_80MHZ_7 0xc0 -#define IEEE80211_HE_PHY_CAP4_BEAMFORMEE_MAX_STS_ABOVE_80MHZ_8 0xe0 -#define IEEE80211_HE_PHY_CAP4_BEAMFORMEE_MAX_STS_ABOVE_80MHZ_MASK 0xe0 - -#define IEEE80211_HE_PHY_CAP5_BEAMFORMEE_NUM_SND_DIM_UNDER_80MHZ_1 0x00 -#define IEEE80211_HE_PHY_CAP5_BEAMFORMEE_NUM_SND_DIM_UNDER_80MHZ_2 0x01 -#define IEEE80211_HE_PHY_CAP5_BEAMFORMEE_NUM_SND_DIM_UNDER_80MHZ_3 0x02 -#define IEEE80211_HE_PHY_CAP5_BEAMFORMEE_NUM_SND_DIM_UNDER_80MHZ_4 0x03 -#define IEEE80211_HE_PHY_CAP5_BEAMFORMEE_NUM_SND_DIM_UNDER_80MHZ_5 0x04 -#define IEEE80211_HE_PHY_CAP5_BEAMFORMEE_NUM_SND_DIM_UNDER_80MHZ_6 0x05 -#define IEEE80211_HE_PHY_CAP5_BEAMFORMEE_NUM_SND_DIM_UNDER_80MHZ_7 0x06 -#define IEEE80211_HE_PHY_CAP5_BEAMFORMEE_NUM_SND_DIM_UNDER_80MHZ_8 0x07 -#define IEEE80211_HE_PHY_CAP5_BEAMFORMEE_NUM_SND_DIM_UNDER_80MHZ_MASK 0x07 - -#define IEEE80211_HE_PHY_CAP5_BEAMFORMEE_NUM_SND_DIM_ABOVE_80MHZ_1 0x00 -#define IEEE80211_HE_PHY_CAP5_BEAMFORMEE_NUM_SND_DIM_ABOVE_80MHZ_2 0x08 -#define IEEE80211_HE_PHY_CAP5_BEAMFORMEE_NUM_SND_DIM_ABOVE_80MHZ_3 0x10 -#define IEEE80211_HE_PHY_CAP5_BEAMFORMEE_NUM_SND_DIM_ABOVE_80MHZ_4 0x18 -#define IEEE80211_HE_PHY_CAP5_BEAMFORMEE_NUM_SND_DIM_ABOVE_80MHZ_5 0x20 -#define IEEE80211_HE_PHY_CAP5_BEAMFORMEE_NUM_SND_DIM_ABOVE_80MHZ_6 0x28 -#define IEEE80211_HE_PHY_CAP5_BEAMFORMEE_NUM_SND_DIM_ABOVE_80MHZ_7 0x30 -#define IEEE80211_HE_PHY_CAP5_BEAMFORMEE_NUM_SND_DIM_ABOVE_80MHZ_8 0x38 -#define IEEE80211_HE_PHY_CAP5_BEAMFORMEE_NUM_SND_DIM_ABOVE_80MHZ_MASK 0x38 - -#define IEEE80211_HE_PHY_CAP5_NG16_SU_FEEDBACK 0x40 -#define IEEE80211_HE_PHY_CAP5_NG16_MU_FEEDBACK 0x80 - -#define IEEE80211_HE_PHY_CAP6_CODEBOOK_SIZE_42_SU 0x01 -#define IEEE80211_HE_PHY_CAP6_CODEBOOK_SIZE_75_MU 0x02 -#define IEEE80211_HE_PHY_CAP6_TRIG_SU_BEAMFORMING_FB 0x04 -#define IEEE80211_HE_PHY_CAP6_TRIG_MU_BEAMFORMING_PARTIAL_BW_FB 0x08 -#define IEEE80211_HE_PHY_CAP6_TRIG_CQI_FB 0x10 -#define IEEE80211_HE_PHY_CAP6_PARTIAL_BW_EXT_RANGE 0x20 -#define IEEE80211_HE_PHY_CAP6_PARTIAL_BANDWIDTH_DL_MUMIMO 0x40 -#define IEEE80211_HE_PHY_CAP6_PPE_THRESHOLD_PRESENT 0x80 - -#define IEEE80211_HE_PHY_CAP7_PSR_BASED_SR 0x01 -#define IEEE80211_HE_PHY_CAP7_POWER_BOOST_FACTOR_SUPP 0x02 -#define IEEE80211_HE_PHY_CAP7_HE_SU_MU_PPDU_4XLTF_AND_08_US_GI 0x04 -#define IEEE80211_HE_PHY_CAP7_MAX_NC_1 0x08 -#define IEEE80211_HE_PHY_CAP7_MAX_NC_2 0x10 -#define IEEE80211_HE_PHY_CAP7_MAX_NC_3 0x18 -#define IEEE80211_HE_PHY_CAP7_MAX_NC_4 0x20 -#define IEEE80211_HE_PHY_CAP7_MAX_NC_5 0x28 -#define IEEE80211_HE_PHY_CAP7_MAX_NC_6 0x30 -#define IEEE80211_HE_PHY_CAP7_MAX_NC_7 0x38 -#define IEEE80211_HE_PHY_CAP7_MAX_NC_MASK 0x38 -#define IEEE80211_HE_PHY_CAP7_STBC_TX_ABOVE_80MHZ 0x40 -#define IEEE80211_HE_PHY_CAP7_STBC_RX_ABOVE_80MHZ 0x80 - -#define IEEE80211_HE_PHY_CAP8_HE_ER_SU_PPDU_4XLTF_AND_08_US_GI 0x01 -#define IEEE80211_HE_PHY_CAP8_20MHZ_IN_40MHZ_HE_PPDU_IN_2G 0x02 -#define IEEE80211_HE_PHY_CAP8_20MHZ_IN_160MHZ_HE_PPDU 0x04 -#define IEEE80211_HE_PHY_CAP8_80MHZ_IN_160MHZ_HE_PPDU 0x08 -#define IEEE80211_HE_PHY_CAP8_HE_ER_SU_1XLTF_AND_08_US_GI 0x10 -#define IEEE80211_HE_PHY_CAP8_MIDAMBLE_RX_TX_2X_AND_1XLTF 0x20 -#define IEEE80211_HE_PHY_CAP8_DCM_MAX_RU_242 0x00 -#define IEEE80211_HE_PHY_CAP8_DCM_MAX_RU_484 0x40 -#define IEEE80211_HE_PHY_CAP8_DCM_MAX_RU_996 0x80 -#define IEEE80211_HE_PHY_CAP8_DCM_MAX_RU_2x996 0xc0 -#define IEEE80211_HE_PHY_CAP8_DCM_MAX_RU_MASK 0xc0 - -#define IEEE80211_HE_PHY_CAP9_LONGER_THAN_16_SIGB_OFDM_SYM 0x01 -#define IEEE80211_HE_PHY_CAP9_NON_TRIGGERED_CQI_FEEDBACK 0x02 -#define IEEE80211_HE_PHY_CAP9_TX_1024_QAM_LESS_THAN_242_TONE_RU 0x04 -#define IEEE80211_HE_PHY_CAP9_RX_1024_QAM_LESS_THAN_242_TONE_RU 0x08 -#define IEEE80211_HE_PHY_CAP9_RX_FULL_BW_SU_USING_MU_WITH_COMP_SIGB 0x10 -#define IEEE80211_HE_PHY_CAP9_RX_FULL_BW_SU_USING_MU_WITH_NON_COMP_SIGB 0x20 -#define IEEE80211_HE_PHY_CAP9_NOMINAL_PKT_PADDING_0US 0x0 -#define IEEE80211_HE_PHY_CAP9_NOMINAL_PKT_PADDING_8US 0x1 -#define IEEE80211_HE_PHY_CAP9_NOMINAL_PKT_PADDING_16US 0x2 -#define IEEE80211_HE_PHY_CAP9_NOMINAL_PKT_PADDING_RESERVED 0x3 -#define IEEE80211_HE_PHY_CAP9_NOMINAL_PKT_PADDING_POS 6 -#define IEEE80211_HE_PHY_CAP9_NOMINAL_PKT_PADDING_MASK 0xc0 - -#define IEEE80211_HE_PHY_CAP10_HE_MU_M1RU_MAX_LTF 0x01 - -/* 802.11ax HE TX/RX MCS NSS Support */ -#define IEEE80211_TX_RX_MCS_NSS_SUPP_HIGHEST_MCS_POS (3) -#define IEEE80211_TX_RX_MCS_NSS_SUPP_TX_BITMAP_POS (6) -#define IEEE80211_TX_RX_MCS_NSS_SUPP_RX_BITMAP_POS (11) -#define IEEE80211_TX_RX_MCS_NSS_SUPP_TX_BITMAP_MASK 0x07c0 -#define IEEE80211_TX_RX_MCS_NSS_SUPP_RX_BITMAP_MASK 0xf800 - -/* TX/RX HE MCS Support field Highest MCS subfield encoding */ -enum ieee80211_he_highest_mcs_supported_subfield_enc { - HIGHEST_MCS_SUPPORTED_MCS7 = 0, - HIGHEST_MCS_SUPPORTED_MCS8, - HIGHEST_MCS_SUPPORTED_MCS9, - HIGHEST_MCS_SUPPORTED_MCS10, - HIGHEST_MCS_SUPPORTED_MCS11, -}; - -/* Calculate 802.11ax HE capabilities IE Tx/Rx HE MCS NSS Support Field size */ -static inline u8 -ieee80211_he_mcs_nss_size(const struct ieee80211_he_cap_elem *he_cap) -{ - u8 count = 4; - - if (he_cap->phy_cap_info[0] & - IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_160MHZ_IN_5G) - count += 4; - - if (he_cap->phy_cap_info[0] & - IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_80PLUS80_MHZ_IN_5G) - count += 4; - - return count; -} - -/* 802.11ax HE PPE Thresholds */ -#define IEEE80211_PPE_THRES_NSS_SUPPORT_2NSS (1) -#define IEEE80211_PPE_THRES_NSS_POS (0) -#define IEEE80211_PPE_THRES_NSS_MASK (7) -#define IEEE80211_PPE_THRES_RU_INDEX_BITMASK_2x966_AND_966_RU \ - (BIT(5) | BIT(6)) -#define IEEE80211_PPE_THRES_RU_INDEX_BITMASK_MASK 0x78 -#define IEEE80211_PPE_THRES_RU_INDEX_BITMASK_POS (3) -#define IEEE80211_PPE_THRES_INFO_PPET_SIZE (3) -#define IEEE80211_HE_PPE_THRES_INFO_HEADER_SIZE (7) - -/* - * Calculate 802.11ax HE capabilities IE PPE field size - * Input: Header byte of ppe_thres (first byte), and HE capa IE's PHY cap u8* - */ -static inline u8 -ieee80211_he_ppe_size(u8 ppe_thres_hdr, const u8 *phy_cap_info) -{ - u8 n; - - if ((phy_cap_info[6] & - IEEE80211_HE_PHY_CAP6_PPE_THRESHOLD_PRESENT) == 0) - return 0; - - n = hweight8(ppe_thres_hdr & - IEEE80211_PPE_THRES_RU_INDEX_BITMASK_MASK); - n *= (1 + ((ppe_thres_hdr & IEEE80211_PPE_THRES_NSS_MASK) >> - IEEE80211_PPE_THRES_NSS_POS)); - - /* - * Each pair is 6 bits, and we need to add the 7 "header" bits to the - * total size. - */ - n = (n * IEEE80211_PPE_THRES_INFO_PPET_SIZE * 2) + 7; - n = DIV_ROUND_UP(n, 8); - - return n; -} - -static inline bool ieee80211_he_capa_size_ok(const u8 *data, u8 len) -{ - const struct ieee80211_he_cap_elem *he_cap_ie_elem = (const void *)data; - u8 needed = sizeof(*he_cap_ie_elem); - - if (len < needed) - return false; - - needed += ieee80211_he_mcs_nss_size(he_cap_ie_elem); - if (len < needed) - return false; - - if (he_cap_ie_elem->phy_cap_info[6] & - IEEE80211_HE_PHY_CAP6_PPE_THRESHOLD_PRESENT) { - if (len < needed + 1) - return false; - needed += ieee80211_he_ppe_size(data[needed], - he_cap_ie_elem->phy_cap_info); - } - - return len >= needed; -} - -/* HE Operation defines */ -#define IEEE80211_HE_OPERATION_DFLT_PE_DURATION_MASK 0x00000007 -#define IEEE80211_HE_OPERATION_TWT_REQUIRED 0x00000008 -#define IEEE80211_HE_OPERATION_RTS_THRESHOLD_MASK 0x00003ff0 -#define IEEE80211_HE_OPERATION_RTS_THRESHOLD_OFFSET 4 -#define IEEE80211_HE_OPERATION_VHT_OPER_INFO 0x00004000 -#define IEEE80211_HE_OPERATION_CO_HOSTED_BSS 0x00008000 -#define IEEE80211_HE_OPERATION_ER_SU_DISABLE 0x00010000 -#define IEEE80211_HE_OPERATION_6GHZ_OP_INFO 0x00020000 -#define IEEE80211_HE_OPERATION_BSS_COLOR_MASK 0x3f000000 -#define IEEE80211_HE_OPERATION_BSS_COLOR_OFFSET 24 -#define IEEE80211_HE_OPERATION_PARTIAL_BSS_COLOR 0x40000000 -#define IEEE80211_HE_OPERATION_BSS_COLOR_DISABLED 0x80000000 - -#define IEEE80211_6GHZ_CTRL_REG_LPI_AP 0 -#define IEEE80211_6GHZ_CTRL_REG_SP_AP 1 -#define IEEE80211_6GHZ_CTRL_REG_VLP_AP 2 -#define IEEE80211_6GHZ_CTRL_REG_INDOOR_LPI_AP 3 -#define IEEE80211_6GHZ_CTRL_REG_INDOOR_SP_AP_OLD 4 -#define IEEE80211_6GHZ_CTRL_REG_INDOOR_SP_AP 8 - -/** - * struct ieee80211_he_6ghz_oper - HE 6 GHz operation Information field - * @primary: primary channel - * @control: control flags - * @ccfs0: channel center frequency segment 0 - * @ccfs1: channel center frequency segment 1 - * @minrate: minimum rate (in 1 Mbps units) - */ -struct ieee80211_he_6ghz_oper { - u8 primary; -#define IEEE80211_HE_6GHZ_OPER_CTRL_CHANWIDTH 0x3 -#define IEEE80211_HE_6GHZ_OPER_CTRL_CHANWIDTH_20MHZ 0 -#define IEEE80211_HE_6GHZ_OPER_CTRL_CHANWIDTH_40MHZ 1 -#define IEEE80211_HE_6GHZ_OPER_CTRL_CHANWIDTH_80MHZ 2 -#define IEEE80211_HE_6GHZ_OPER_CTRL_CHANWIDTH_160MHZ 3 -#define IEEE80211_HE_6GHZ_OPER_CTRL_DUP_BEACON 0x4 -#define IEEE80211_HE_6GHZ_OPER_CTRL_REG_INFO 0x78 - u8 control; - u8 ccfs0; - u8 ccfs1; - u8 minrate; -} __packed; - -/** - * enum ieee80211_reg_conn_bits - represents Regulatory connectivity field bits. - * - * This enumeration defines bit flags used to represent regulatory connectivity - * field bits. - * - * @IEEE80211_REG_CONN_LPI_VALID: Indicates whether the LPI bit is valid. - * @IEEE80211_REG_CONN_LPI_VALUE: Represents the value of the LPI bit. - * @IEEE80211_REG_CONN_SP_VALID: Indicates whether the SP bit is valid. - * @IEEE80211_REG_CONN_SP_VALUE: Represents the value of the SP bit. - */ -enum ieee80211_reg_conn_bits { - IEEE80211_REG_CONN_LPI_VALID = BIT(0), - IEEE80211_REG_CONN_LPI_VALUE = BIT(1), - IEEE80211_REG_CONN_SP_VALID = BIT(2), - IEEE80211_REG_CONN_SP_VALUE = BIT(3), -}; - -/* transmit power interpretation type of transmit power envelope element */ -enum ieee80211_tx_power_intrpt_type { - IEEE80211_TPE_LOCAL_EIRP, - IEEE80211_TPE_LOCAL_EIRP_PSD, - IEEE80211_TPE_REG_CLIENT_EIRP, - IEEE80211_TPE_REG_CLIENT_EIRP_PSD, -}; - -/* category type of transmit power envelope element */ -enum ieee80211_tx_power_category_6ghz { - IEEE80211_TPE_CAT_6GHZ_DEFAULT = 0, - IEEE80211_TPE_CAT_6GHZ_SUBORDINATE = 1, -}; - -/* - * For IEEE80211_TPE_LOCAL_EIRP / IEEE80211_TPE_REG_CLIENT_EIRP, - * setting to 63.5 dBm means no constraint. - */ -#define IEEE80211_TPE_MAX_TX_PWR_NO_CONSTRAINT 127 - -/* - * For IEEE80211_TPE_LOCAL_EIRP_PSD / IEEE80211_TPE_REG_CLIENT_EIRP_PSD, - * setting to 127 indicates no PSD limit for the 20 MHz channel. - */ -#define IEEE80211_TPE_PSD_NO_LIMIT 127 - -/** - * struct ieee80211_tx_pwr_env - Transmit Power Envelope - * @info: Transmit Power Information field - * @variable: Maximum Transmit Power field - * - * This structure represents the payload of the "Transmit Power - * Envelope element" as described in IEEE Std 802.11ax-2021 section - * 9.4.2.161 - */ -struct ieee80211_tx_pwr_env { - u8 info; - u8 variable[]; -} __packed; - -#define IEEE80211_TX_PWR_ENV_INFO_COUNT 0x7 -#define IEEE80211_TX_PWR_ENV_INFO_INTERPRET 0x38 -#define IEEE80211_TX_PWR_ENV_INFO_CATEGORY 0xC0 - -#define IEEE80211_TX_PWR_ENV_EXT_COUNT 0xF - -static inline bool ieee80211_valid_tpe_element(const u8 *data, u8 len) -{ - const struct ieee80211_tx_pwr_env *env = (const void *)data; - u8 count, interpret, category; - u8 needed = sizeof(*env); - u8 N; /* also called N in the spec */ - - if (len < needed) - return false; - - count = u8_get_bits(env->info, IEEE80211_TX_PWR_ENV_INFO_COUNT); - interpret = u8_get_bits(env->info, IEEE80211_TX_PWR_ENV_INFO_INTERPRET); - category = u8_get_bits(env->info, IEEE80211_TX_PWR_ENV_INFO_CATEGORY); - - switch (category) { - case IEEE80211_TPE_CAT_6GHZ_DEFAULT: - case IEEE80211_TPE_CAT_6GHZ_SUBORDINATE: - break; - default: - return false; - } - - switch (interpret) { - case IEEE80211_TPE_LOCAL_EIRP: - case IEEE80211_TPE_REG_CLIENT_EIRP: - if (count > 3) - return false; - - /* count == 0 encodes 1 value for 20 MHz, etc. */ - needed += count + 1; - - if (len < needed) - return false; - - /* there can be extension fields not accounted for in 'count' */ - - return true; - case IEEE80211_TPE_LOCAL_EIRP_PSD: - case IEEE80211_TPE_REG_CLIENT_EIRP_PSD: - if (count > 4) - return false; - - N = count ? 1 << (count - 1) : 1; - needed += N; - - if (len < needed) - return false; - - if (len > needed) { - u8 K = u8_get_bits(env->variable[N], - IEEE80211_TX_PWR_ENV_EXT_COUNT); - - needed += 1 + K; - if (len < needed) - return false; - } - - return true; - } - - return false; -} - -/* - * ieee80211_he_oper_size - calculate 802.11ax HE Operations IE size - * @he_oper_ie: byte data of the He Operations IE, stating from the byte - * after the ext ID byte. It is assumed that he_oper_ie has at least - * sizeof(struct ieee80211_he_operation) bytes, the caller must have - * validated this. - * @return the actual size of the IE data (not including header), or 0 on error - */ -static inline u8 -ieee80211_he_oper_size(const u8 *he_oper_ie) -{ - const struct ieee80211_he_operation *he_oper = (const void *)he_oper_ie; - u8 oper_len = sizeof(struct ieee80211_he_operation); - u32 he_oper_params; - - /* Make sure the input is not NULL */ - if (!he_oper_ie) - return 0; - - /* Calc required length */ - he_oper_params = le32_to_cpu(he_oper->he_oper_params); - if (he_oper_params & IEEE80211_HE_OPERATION_VHT_OPER_INFO) - oper_len += 3; - if (he_oper_params & IEEE80211_HE_OPERATION_CO_HOSTED_BSS) - oper_len++; - if (he_oper_params & IEEE80211_HE_OPERATION_6GHZ_OP_INFO) - oper_len += sizeof(struct ieee80211_he_6ghz_oper); - - /* Add the first byte (extension ID) to the total length */ - oper_len++; - - return oper_len; -} - -/** - * ieee80211_he_6ghz_oper - obtain 6 GHz operation field - * @he_oper: HE operation element (must be pre-validated for size) - * but may be %NULL - * - * Return: a pointer to the 6 GHz operation field, or %NULL - */ -static inline const struct ieee80211_he_6ghz_oper * -ieee80211_he_6ghz_oper(const struct ieee80211_he_operation *he_oper) -{ - const u8 *ret; - u32 he_oper_params; - - if (!he_oper) - return NULL; - - ret = (const void *)&he_oper->optional; - - he_oper_params = le32_to_cpu(he_oper->he_oper_params); - - if (!(he_oper_params & IEEE80211_HE_OPERATION_6GHZ_OP_INFO)) - return NULL; - if (he_oper_params & IEEE80211_HE_OPERATION_VHT_OPER_INFO) - ret += 3; - if (he_oper_params & IEEE80211_HE_OPERATION_CO_HOSTED_BSS) - ret++; - - return (const void *)ret; -} - -/* HE Spatial Reuse defines */ -#define IEEE80211_HE_SPR_PSR_DISALLOWED BIT(0) -#define IEEE80211_HE_SPR_NON_SRG_OBSS_PD_SR_DISALLOWED BIT(1) -#define IEEE80211_HE_SPR_NON_SRG_OFFSET_PRESENT BIT(2) -#define IEEE80211_HE_SPR_SRG_INFORMATION_PRESENT BIT(3) -#define IEEE80211_HE_SPR_HESIGA_SR_VAL15_ALLOWED BIT(4) - -/* - * ieee80211_he_spr_size - calculate 802.11ax HE Spatial Reuse IE size - * @he_spr_ie: byte data of the He Spatial Reuse IE, stating from the byte - * after the ext ID byte. It is assumed that he_spr_ie has at least - * sizeof(struct ieee80211_he_spr) bytes, the caller must have validated - * this - * @return the actual size of the IE data (not including header), or 0 on error - */ -static inline u8 -ieee80211_he_spr_size(const u8 *he_spr_ie) -{ - const struct ieee80211_he_spr *he_spr = (const void *)he_spr_ie; - u8 spr_len = sizeof(struct ieee80211_he_spr); - u8 he_spr_params; - - /* Make sure the input is not NULL */ - if (!he_spr_ie) - return 0; - - /* Calc required length */ - he_spr_params = he_spr->he_sr_control; - if (he_spr_params & IEEE80211_HE_SPR_NON_SRG_OFFSET_PRESENT) - spr_len++; - if (he_spr_params & IEEE80211_HE_SPR_SRG_INFORMATION_PRESENT) - spr_len += 18; - - /* Add the first byte (extension ID) to the total length */ - spr_len++; - - return spr_len; -} - /* S1G Capabilities Information field */ #define IEEE80211_S1G_CAPABILITY_LEN 15 @@ -2697,6 +1912,9 @@ ieee80211_he_spr_size(const u8 *he_spr_ie) #define IEEE80211_EHT_OPER_CHAN_WIDTH_160MHZ 3 #define IEEE80211_EHT_OPER_CHAN_WIDTH_320MHZ 4 +/* need HE definitions for EHT functions */ +#include "ieee80211-he.h" + /* Calculate 802.11be EHT capabilities IE Tx/Rx EHT MCS NSS Support Field size */ static inline u8 ieee80211_eht_mcs_nss_size(const struct ieee80211_he_cap_elem *he_cap, @@ -3815,24 +3033,6 @@ struct ieee80211_tspec_ie { __le16 medium_time; } __packed; -struct ieee80211_he_6ghz_capa { - /* uses IEEE80211_HE_6GHZ_CAP_* below */ - __le16 capa; -} __packed; - -/* HE 6 GHz band capabilities */ -/* uses enum ieee80211_min_mpdu_spacing values */ -#define IEEE80211_HE_6GHZ_CAP_MIN_MPDU_START 0x0007 -/* uses enum ieee80211_vht_max_ampdu_length_exp values */ -#define IEEE80211_HE_6GHZ_CAP_MAX_AMPDU_LEN_EXP 0x0038 -/* uses IEEE80211_VHT_CAP_MAX_MPDU_LENGTH_* values */ -#define IEEE80211_HE_6GHZ_CAP_MAX_MPDU_LEN 0x00c0 -/* WLAN_HT_CAP_SM_PS_* values */ -#define IEEE80211_HE_6GHZ_CAP_SM_PS 0x0600 -#define IEEE80211_HE_6GHZ_CAP_RD_RESPONDER 0x0800 -#define IEEE80211_HE_6GHZ_CAP_RX_ANTPAT_CONS 0x1000 -#define IEEE80211_HE_6GHZ_CAP_TX_ANTPAT_CONS 0x2000 - /** * ieee80211_get_qos_ctl - get pointer to qos control bytes * @hdr: the frame From 86bc0c662322b4749cd666678d2fdce7015bcae3 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 5 Nov 2025 15:36:53 +0100 Subject: [PATCH 0637/1381] wifi: ieee80211: split EHT definitions out The ieee80211.h file has gotten very long, continue splitting it by putting EHT definitions into a separate file. Link: https://patch.msgid.link/20251105153843.bf77fe169140.I691267e0edd914c604a5bfd447d33be00044c9b4@changeid Signed-off-by: Johannes Berg --- include/linux/ieee80211-eht.h | 1182 +++++++++++++++++++++++++++++++++ include/linux/ieee80211.h | 1164 +------------------------------- 2 files changed, 1184 insertions(+), 1162 deletions(-) create mode 100644 include/linux/ieee80211-eht.h diff --git a/include/linux/ieee80211-eht.h b/include/linux/ieee80211-eht.h new file mode 100644 index 000000000000..f9782e46c5e5 --- /dev/null +++ b/include/linux/ieee80211-eht.h @@ -0,0 +1,1182 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * IEEE 802.11 EHT definitions + * + * Copyright (c) 2001-2002, SSH Communications Security Corp and Jouni Malinen + * + * Copyright (c) 2002-2003, Jouni Malinen + * Copyright (c) 2005, Devicescape Software, Inc. + * Copyright (c) 2006, Michael Wu + * Copyright (c) 2013 - 2014 Intel Mobile Communications GmbH + * Copyright (c) 2016 - 2017 Intel Deutschland GmbH + * Copyright (c) 2018 - 2025 Intel Corporation + */ + +#ifndef LINUX_IEEE80211_EHT_H +#define LINUX_IEEE80211_EHT_H + +#include +#include +/* need HE definitions for the inlines here */ +#include + +#define IEEE80211_TTLM_MAX_CNT 2 +#define IEEE80211_TTLM_CONTROL_DIRECTION 0x03 +#define IEEE80211_TTLM_CONTROL_DEF_LINK_MAP 0x04 +#define IEEE80211_TTLM_CONTROL_SWITCH_TIME_PRESENT 0x08 +#define IEEE80211_TTLM_CONTROL_EXPECTED_DUR_PRESENT 0x10 +#define IEEE80211_TTLM_CONTROL_LINK_MAP_SIZE 0x20 + +#define IEEE80211_TTLM_DIRECTION_DOWN 0 +#define IEEE80211_TTLM_DIRECTION_UP 1 +#define IEEE80211_TTLM_DIRECTION_BOTH 2 + +/** + * struct ieee80211_ttlm_elem - TID-To-Link Mapping element + * + * Defined in section 9.4.2.314 in P802.11be_D4 + * + * @control: the first part of control field + * @optional: the second part of control field + */ +struct ieee80211_ttlm_elem { + u8 control; + u8 optional[]; +} __packed; + +#define IEEE80211_EHT_MCS_NSS_RX 0x0f +#define IEEE80211_EHT_MCS_NSS_TX 0xf0 + +/** + * struct ieee80211_eht_mcs_nss_supp_20mhz_only - EHT 20MHz only station max + * supported NSS for per MCS. + * + * For each field below, bits 0 - 3 indicate the maximal number of spatial + * streams for Rx, and bits 4 - 7 indicate the maximal number of spatial streams + * for Tx. + * + * @rx_tx_mcs7_max_nss: indicates the maximum number of spatial streams + * supported for reception and the maximum number of spatial streams + * supported for transmission for MCS 0 - 7. + * @rx_tx_mcs9_max_nss: indicates the maximum number of spatial streams + * supported for reception and the maximum number of spatial streams + * supported for transmission for MCS 8 - 9. + * @rx_tx_mcs11_max_nss: indicates the maximum number of spatial streams + * supported for reception and the maximum number of spatial streams + * supported for transmission for MCS 10 - 11. + * @rx_tx_mcs13_max_nss: indicates the maximum number of spatial streams + * supported for reception and the maximum number of spatial streams + * supported for transmission for MCS 12 - 13. + * @rx_tx_max_nss: array of the previous fields for easier loop access + */ +struct ieee80211_eht_mcs_nss_supp_20mhz_only { + union { + struct { + u8 rx_tx_mcs7_max_nss; + u8 rx_tx_mcs9_max_nss; + u8 rx_tx_mcs11_max_nss; + u8 rx_tx_mcs13_max_nss; + }; + u8 rx_tx_max_nss[4]; + }; +}; + +/** + * struct ieee80211_eht_mcs_nss_supp_bw - EHT max supported NSS per MCS (except + * 20MHz only stations). + * + * For each field below, bits 0 - 3 indicate the maximal number of spatial + * streams for Rx, and bits 4 - 7 indicate the maximal number of spatial streams + * for Tx. + * + * @rx_tx_mcs9_max_nss: indicates the maximum number of spatial streams + * supported for reception and the maximum number of spatial streams + * supported for transmission for MCS 0 - 9. + * @rx_tx_mcs11_max_nss: indicates the maximum number of spatial streams + * supported for reception and the maximum number of spatial streams + * supported for transmission for MCS 10 - 11. + * @rx_tx_mcs13_max_nss: indicates the maximum number of spatial streams + * supported for reception and the maximum number of spatial streams + * supported for transmission for MCS 12 - 13. + * @rx_tx_max_nss: array of the previous fields for easier loop access + */ +struct ieee80211_eht_mcs_nss_supp_bw { + union { + struct { + u8 rx_tx_mcs9_max_nss; + u8 rx_tx_mcs11_max_nss; + u8 rx_tx_mcs13_max_nss; + }; + u8 rx_tx_max_nss[3]; + }; +}; + +/** + * struct ieee80211_eht_cap_elem_fixed - EHT capabilities fixed data + * + * This structure is the "EHT Capabilities element" fixed fields as + * described in P802.11be_D2.0 section 9.4.2.313. + * + * @mac_cap_info: MAC capabilities, see IEEE80211_EHT_MAC_CAP* + * @phy_cap_info: PHY capabilities, see IEEE80211_EHT_PHY_CAP* + */ +struct ieee80211_eht_cap_elem_fixed { + u8 mac_cap_info[2]; + u8 phy_cap_info[9]; +} __packed; + +/** + * struct ieee80211_eht_cap_elem - EHT capabilities element + * @fixed: fixed parts, see &ieee80211_eht_cap_elem_fixed + * @optional: optional parts + */ +struct ieee80211_eht_cap_elem { + struct ieee80211_eht_cap_elem_fixed fixed; + + /* + * Followed by: + * Supported EHT-MCS And NSS Set field: 4, 3, 6 or 9 octets. + * EHT PPE Thresholds field: variable length. + */ + u8 optional[]; +} __packed; + +#define IEEE80211_EHT_OPER_INFO_PRESENT 0x01 +#define IEEE80211_EHT_OPER_DISABLED_SUBCHANNEL_BITMAP_PRESENT 0x02 +#define IEEE80211_EHT_OPER_EHT_DEF_PE_DURATION 0x04 +#define IEEE80211_EHT_OPER_GROUP_ADDRESSED_BU_IND_LIMIT 0x08 +#define IEEE80211_EHT_OPER_GROUP_ADDRESSED_BU_IND_EXP_MASK 0x30 +#define IEEE80211_EHT_OPER_MCS15_DISABLE 0x40 + +/** + * struct ieee80211_eht_operation - eht operation element + * + * This structure is the "EHT Operation Element" fields as + * described in P802.11be_D2.0 section 9.4.2.311 + * + * @params: EHT operation element parameters. See &IEEE80211_EHT_OPER_* + * @basic_mcs_nss: indicates the EHT-MCSs for each number of spatial streams in + * EHT PPDUs that are supported by all EHT STAs in the BSS in transmit and + * receive. + * @optional: optional parts + */ +struct ieee80211_eht_operation { + u8 params; + struct ieee80211_eht_mcs_nss_supp_20mhz_only basic_mcs_nss; + u8 optional[]; +} __packed; + +/** + * struct ieee80211_eht_operation_info - eht operation information + * + * @control: EHT operation information control. + * @ccfs0: defines a channel center frequency for a 20, 40, 80, 160, or 320 MHz + * EHT BSS. + * @ccfs1: defines a channel center frequency for a 160 or 320 MHz EHT BSS. + * @optional: optional parts + */ +struct ieee80211_eht_operation_info { + u8 control; + u8 ccfs0; + u8 ccfs1; + u8 optional[]; +} __packed; + +/* EHT MAC capabilities as defined in P802.11be_D2.0 section 9.4.2.313.2 */ +#define IEEE80211_EHT_MAC_CAP0_EPCS_PRIO_ACCESS 0x01 +#define IEEE80211_EHT_MAC_CAP0_OM_CONTROL 0x02 +#define IEEE80211_EHT_MAC_CAP0_TRIG_TXOP_SHARING_MODE1 0x04 +#define IEEE80211_EHT_MAC_CAP0_TRIG_TXOP_SHARING_MODE2 0x08 +#define IEEE80211_EHT_MAC_CAP0_RESTRICTED_TWT 0x10 +#define IEEE80211_EHT_MAC_CAP0_SCS_TRAFFIC_DESC 0x20 +#define IEEE80211_EHT_MAC_CAP0_MAX_MPDU_LEN_MASK 0xc0 +#define IEEE80211_EHT_MAC_CAP0_MAX_MPDU_LEN_3895 0 +#define IEEE80211_EHT_MAC_CAP0_MAX_MPDU_LEN_7991 1 +#define IEEE80211_EHT_MAC_CAP0_MAX_MPDU_LEN_11454 2 + +#define IEEE80211_EHT_MAC_CAP1_MAX_AMPDU_LEN_MASK 0x01 +#define IEEE80211_EHT_MAC_CAP1_EHT_TRS 0x02 +#define IEEE80211_EHT_MAC_CAP1_TXOP_RET 0x04 +#define IEEE80211_EHT_MAC_CAP1_TWO_BQRS 0x08 +#define IEEE80211_EHT_MAC_CAP1_EHT_LINK_ADAPT_MASK 0x30 +#define IEEE80211_EHT_MAC_CAP1_UNSOL_EPCS_PRIO_ACCESS 0x40 + +/* EHT PHY capabilities as defined in P802.11be_D2.0 section 9.4.2.313.3 */ +#define IEEE80211_EHT_PHY_CAP0_320MHZ_IN_6GHZ 0x02 +#define IEEE80211_EHT_PHY_CAP0_242_TONE_RU_GT20MHZ 0x04 +#define IEEE80211_EHT_PHY_CAP0_NDP_4_EHT_LFT_32_GI 0x08 +#define IEEE80211_EHT_PHY_CAP0_PARTIAL_BW_UL_MU_MIMO 0x10 +#define IEEE80211_EHT_PHY_CAP0_SU_BEAMFORMER 0x20 +#define IEEE80211_EHT_PHY_CAP0_SU_BEAMFORMEE 0x40 + +/* EHT beamformee number of spatial streams <= 80MHz is split */ +#define IEEE80211_EHT_PHY_CAP0_BEAMFORMEE_SS_80MHZ_MASK 0x80 +#define IEEE80211_EHT_PHY_CAP1_BEAMFORMEE_SS_80MHZ_MASK 0x03 + +#define IEEE80211_EHT_PHY_CAP1_BEAMFORMEE_SS_160MHZ_MASK 0x1c +#define IEEE80211_EHT_PHY_CAP1_BEAMFORMEE_SS_320MHZ_MASK 0xe0 + +#define IEEE80211_EHT_PHY_CAP2_SOUNDING_DIM_80MHZ_MASK 0x07 +#define IEEE80211_EHT_PHY_CAP2_SOUNDING_DIM_160MHZ_MASK 0x38 + +/* EHT number of sounding dimensions for 320MHz is split */ +#define IEEE80211_EHT_PHY_CAP2_SOUNDING_DIM_320MHZ_MASK 0xc0 +#define IEEE80211_EHT_PHY_CAP3_SOUNDING_DIM_320MHZ_MASK 0x01 +#define IEEE80211_EHT_PHY_CAP3_NG_16_SU_FEEDBACK 0x02 +#define IEEE80211_EHT_PHY_CAP3_NG_16_MU_FEEDBACK 0x04 +#define IEEE80211_EHT_PHY_CAP3_CODEBOOK_4_2_SU_FDBK 0x08 +#define IEEE80211_EHT_PHY_CAP3_CODEBOOK_7_5_MU_FDBK 0x10 +#define IEEE80211_EHT_PHY_CAP3_TRIG_SU_BF_FDBK 0x20 +#define IEEE80211_EHT_PHY_CAP3_TRIG_MU_BF_PART_BW_FDBK 0x40 +#define IEEE80211_EHT_PHY_CAP3_TRIG_CQI_FDBK 0x80 + +#define IEEE80211_EHT_PHY_CAP4_PART_BW_DL_MU_MIMO 0x01 +#define IEEE80211_EHT_PHY_CAP4_PSR_SR_SUPP 0x02 +#define IEEE80211_EHT_PHY_CAP4_POWER_BOOST_FACT_SUPP 0x04 +#define IEEE80211_EHT_PHY_CAP4_EHT_MU_PPDU_4_EHT_LTF_08_GI 0x08 +#define IEEE80211_EHT_PHY_CAP4_MAX_NC_MASK 0xf0 + +#define IEEE80211_EHT_PHY_CAP5_NON_TRIG_CQI_FEEDBACK 0x01 +#define IEEE80211_EHT_PHY_CAP5_TX_LESS_242_TONE_RU_SUPP 0x02 +#define IEEE80211_EHT_PHY_CAP5_RX_LESS_242_TONE_RU_SUPP 0x04 +#define IEEE80211_EHT_PHY_CAP5_PPE_THRESHOLD_PRESENT 0x08 +#define IEEE80211_EHT_PHY_CAP5_COMMON_NOMINAL_PKT_PAD_MASK 0x30 +#define IEEE80211_EHT_PHY_CAP5_COMMON_NOMINAL_PKT_PAD_0US 0 +#define IEEE80211_EHT_PHY_CAP5_COMMON_NOMINAL_PKT_PAD_8US 1 +#define IEEE80211_EHT_PHY_CAP5_COMMON_NOMINAL_PKT_PAD_16US 2 +#define IEEE80211_EHT_PHY_CAP5_COMMON_NOMINAL_PKT_PAD_20US 3 + +/* Maximum number of supported EHT LTF is split */ +#define IEEE80211_EHT_PHY_CAP5_MAX_NUM_SUPP_EHT_LTF_MASK 0xc0 +#define IEEE80211_EHT_PHY_CAP5_SUPP_EXTRA_EHT_LTF 0x40 +#define IEEE80211_EHT_PHY_CAP6_MAX_NUM_SUPP_EHT_LTF_MASK 0x07 + +#define IEEE80211_EHT_PHY_CAP6_MCS15_SUPP_80MHZ 0x08 +#define IEEE80211_EHT_PHY_CAP6_MCS15_SUPP_160MHZ 0x30 +#define IEEE80211_EHT_PHY_CAP6_MCS15_SUPP_320MHZ 0x40 +#define IEEE80211_EHT_PHY_CAP6_MCS15_SUPP_MASK 0x78 +#define IEEE80211_EHT_PHY_CAP6_EHT_DUP_6GHZ_SUPP 0x80 + +#define IEEE80211_EHT_PHY_CAP7_20MHZ_STA_RX_NDP_WIDER_BW 0x01 +#define IEEE80211_EHT_PHY_CAP7_NON_OFDMA_UL_MU_MIMO_80MHZ 0x02 +#define IEEE80211_EHT_PHY_CAP7_NON_OFDMA_UL_MU_MIMO_160MHZ 0x04 +#define IEEE80211_EHT_PHY_CAP7_NON_OFDMA_UL_MU_MIMO_320MHZ 0x08 +#define IEEE80211_EHT_PHY_CAP7_MU_BEAMFORMER_80MHZ 0x10 +#define IEEE80211_EHT_PHY_CAP7_MU_BEAMFORMER_160MHZ 0x20 +#define IEEE80211_EHT_PHY_CAP7_MU_BEAMFORMER_320MHZ 0x40 +#define IEEE80211_EHT_PHY_CAP7_TB_SOUNDING_FDBK_RATE_LIMIT 0x80 + +#define IEEE80211_EHT_PHY_CAP8_RX_1024QAM_WIDER_BW_DL_OFDMA 0x01 +#define IEEE80211_EHT_PHY_CAP8_RX_4096QAM_WIDER_BW_DL_OFDMA 0x02 + +/* + * EHT operation channel width as defined in P802.11be_D2.0 section 9.4.2.311 + */ +#define IEEE80211_EHT_OPER_CHAN_WIDTH 0x7 +#define IEEE80211_EHT_OPER_CHAN_WIDTH_20MHZ 0 +#define IEEE80211_EHT_OPER_CHAN_WIDTH_40MHZ 1 +#define IEEE80211_EHT_OPER_CHAN_WIDTH_80MHZ 2 +#define IEEE80211_EHT_OPER_CHAN_WIDTH_160MHZ 3 +#define IEEE80211_EHT_OPER_CHAN_WIDTH_320MHZ 4 + +/* Calculate 802.11be EHT capabilities IE Tx/Rx EHT MCS NSS Support Field size */ +static inline u8 +ieee80211_eht_mcs_nss_size(const struct ieee80211_he_cap_elem *he_cap, + const struct ieee80211_eht_cap_elem_fixed *eht_cap, + bool from_ap) +{ + u8 count = 0; + + /* on 2.4 GHz, if it supports 40 MHz, the result is 3 */ + if (he_cap->phy_cap_info[0] & + IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_40MHZ_IN_2G) + return 3; + + /* on 2.4 GHz, these three bits are reserved, so should be 0 */ + if (he_cap->phy_cap_info[0] & + IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_40MHZ_80MHZ_IN_5G) + count += 3; + + if (he_cap->phy_cap_info[0] & + IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_160MHZ_IN_5G) + count += 3; + + if (eht_cap->phy_cap_info[0] & IEEE80211_EHT_PHY_CAP0_320MHZ_IN_6GHZ) + count += 3; + + if (count) + return count; + + return from_ap ? 3 : 4; +} + +/* 802.11be EHT PPE Thresholds */ +#define IEEE80211_EHT_PPE_THRES_NSS_POS 0 +#define IEEE80211_EHT_PPE_THRES_NSS_MASK 0xf +#define IEEE80211_EHT_PPE_THRES_RU_INDEX_BITMASK_MASK 0x1f0 +#define IEEE80211_EHT_PPE_THRES_INFO_PPET_SIZE 3 +#define IEEE80211_EHT_PPE_THRES_INFO_HEADER_SIZE 9 + +/* + * Calculate 802.11be EHT capabilities IE EHT field size + */ +static inline u8 +ieee80211_eht_ppe_size(u16 ppe_thres_hdr, const u8 *phy_cap_info) +{ + u32 n; + + if (!(phy_cap_info[5] & + IEEE80211_EHT_PHY_CAP5_PPE_THRESHOLD_PRESENT)) + return 0; + + n = hweight16(ppe_thres_hdr & + IEEE80211_EHT_PPE_THRES_RU_INDEX_BITMASK_MASK); + n *= 1 + u16_get_bits(ppe_thres_hdr, IEEE80211_EHT_PPE_THRES_NSS_MASK); + + /* + * Each pair is 6 bits, and we need to add the 9 "header" bits to the + * total size. + */ + n = n * IEEE80211_EHT_PPE_THRES_INFO_PPET_SIZE * 2 + + IEEE80211_EHT_PPE_THRES_INFO_HEADER_SIZE; + return DIV_ROUND_UP(n, 8); +} + +static inline bool +ieee80211_eht_capa_size_ok(const u8 *he_capa, const u8 *data, u8 len, + bool from_ap) +{ + const struct ieee80211_eht_cap_elem_fixed *elem = (const void *)data; + u8 needed = sizeof(struct ieee80211_eht_cap_elem_fixed); + + if (len < needed || !he_capa) + return false; + + needed += ieee80211_eht_mcs_nss_size((const void *)he_capa, + (const void *)data, + from_ap); + if (len < needed) + return false; + + if (elem->phy_cap_info[5] & + IEEE80211_EHT_PHY_CAP5_PPE_THRESHOLD_PRESENT) { + u16 ppe_thres_hdr; + + if (len < needed + sizeof(ppe_thres_hdr)) + return false; + + ppe_thres_hdr = get_unaligned_le16(data + needed); + needed += ieee80211_eht_ppe_size(ppe_thres_hdr, + elem->phy_cap_info); + } + + return len >= needed; +} + +static inline bool +ieee80211_eht_oper_size_ok(const u8 *data, u8 len) +{ + const struct ieee80211_eht_operation *elem = (const void *)data; + u8 needed = sizeof(*elem); + + if (len < needed) + return false; + + if (elem->params & IEEE80211_EHT_OPER_INFO_PRESENT) { + needed += 3; + + if (elem->params & + IEEE80211_EHT_OPER_DISABLED_SUBCHANNEL_BITMAP_PRESENT) + needed += 2; + } + + return len >= needed; +} + +/* must validate ieee80211_eht_oper_size_ok() first */ +static inline u16 +ieee80211_eht_oper_dis_subchan_bitmap(const struct ieee80211_eht_operation *eht_oper) +{ + const struct ieee80211_eht_operation_info *info = + (const void *)eht_oper->optional; + + if (!(eht_oper->params & IEEE80211_EHT_OPER_INFO_PRESENT)) + return 0; + + if (!(eht_oper->params & IEEE80211_EHT_OPER_DISABLED_SUBCHANNEL_BITMAP_PRESENT)) + return 0; + + return get_unaligned_le16(info->optional); +} + +#define IEEE80211_BW_IND_DIS_SUBCH_PRESENT BIT(1) + +struct ieee80211_bandwidth_indication { + u8 params; + struct ieee80211_eht_operation_info info; +} __packed; + +static inline bool +ieee80211_bandwidth_indication_size_ok(const u8 *data, u8 len) +{ + const struct ieee80211_bandwidth_indication *bwi = (const void *)data; + + if (len < sizeof(*bwi)) + return false; + + if (bwi->params & IEEE80211_BW_IND_DIS_SUBCH_PRESENT && + len < sizeof(*bwi) + 2) + return false; + + return true; +} + +/* Protected EHT action codes */ +enum ieee80211_protected_eht_actioncode { + WLAN_PROTECTED_EHT_ACTION_TTLM_REQ = 0, + WLAN_PROTECTED_EHT_ACTION_TTLM_RES = 1, + WLAN_PROTECTED_EHT_ACTION_TTLM_TEARDOWN = 2, + WLAN_PROTECTED_EHT_ACTION_EPCS_ENABLE_REQ = 3, + WLAN_PROTECTED_EHT_ACTION_EPCS_ENABLE_RESP = 4, + WLAN_PROTECTED_EHT_ACTION_EPCS_ENABLE_TEARDOWN = 5, + WLAN_PROTECTED_EHT_ACTION_EML_OP_MODE_NOTIF = 6, + WLAN_PROTECTED_EHT_ACTION_LINK_RECOMMEND = 7, + WLAN_PROTECTED_EHT_ACTION_ML_OP_UPDATE_REQ = 8, + WLAN_PROTECTED_EHT_ACTION_ML_OP_UPDATE_RESP = 9, + WLAN_PROTECTED_EHT_ACTION_LINK_RECONFIG_NOTIF = 10, + WLAN_PROTECTED_EHT_ACTION_LINK_RECONFIG_REQ = 11, + WLAN_PROTECTED_EHT_ACTION_LINK_RECONFIG_RESP = 12, +}; + +/* multi-link device */ +#define IEEE80211_MLD_MAX_NUM_LINKS 15 + +#define IEEE80211_ML_CONTROL_TYPE 0x0007 +#define IEEE80211_ML_CONTROL_TYPE_BASIC 0 +#define IEEE80211_ML_CONTROL_TYPE_PREQ 1 +#define IEEE80211_ML_CONTROL_TYPE_RECONF 2 +#define IEEE80211_ML_CONTROL_TYPE_TDLS 3 +#define IEEE80211_ML_CONTROL_TYPE_PRIO_ACCESS 4 +#define IEEE80211_ML_CONTROL_PRESENCE_MASK 0xfff0 + +struct ieee80211_multi_link_elem { + __le16 control; + u8 variable[]; +} __packed; + +#define IEEE80211_MLC_BASIC_PRES_LINK_ID 0x0010 +#define IEEE80211_MLC_BASIC_PRES_BSS_PARAM_CH_CNT 0x0020 +#define IEEE80211_MLC_BASIC_PRES_MED_SYNC_DELAY 0x0040 +#define IEEE80211_MLC_BASIC_PRES_EML_CAPA 0x0080 +#define IEEE80211_MLC_BASIC_PRES_MLD_CAPA_OP 0x0100 +#define IEEE80211_MLC_BASIC_PRES_MLD_ID 0x0200 +#define IEEE80211_MLC_BASIC_PRES_EXT_MLD_CAPA_OP 0x0400 + +#define IEEE80211_MED_SYNC_DELAY_DURATION 0x00ff +#define IEEE80211_MED_SYNC_DELAY_SYNC_OFDM_ED_THRESH 0x0f00 +#define IEEE80211_MED_SYNC_DELAY_SYNC_MAX_NUM_TXOPS 0xf000 + +/* + * Described in P802.11be_D3.0 + * dot11MSDTimerDuration should default to 5484 (i.e. 171.375) + * dot11MSDOFDMEDthreshold defaults to -72 (i.e. 0) + * dot11MSDTXOPMAX defaults to 1 + */ +#define IEEE80211_MED_SYNC_DELAY_DEFAULT 0x10ac + +#define IEEE80211_EML_CAP_EMLSR_SUPP 0x0001 +#define IEEE80211_EML_CAP_EMLSR_PADDING_DELAY 0x000e +#define IEEE80211_EML_CAP_EMLSR_PADDING_DELAY_0US 0 +#define IEEE80211_EML_CAP_EMLSR_PADDING_DELAY_32US 1 +#define IEEE80211_EML_CAP_EMLSR_PADDING_DELAY_64US 2 +#define IEEE80211_EML_CAP_EMLSR_PADDING_DELAY_128US 3 +#define IEEE80211_EML_CAP_EMLSR_PADDING_DELAY_256US 4 +#define IEEE80211_EML_CAP_EMLSR_TRANSITION_DELAY 0x0070 +#define IEEE80211_EML_CAP_EMLSR_TRANSITION_DELAY_0US 0 +#define IEEE80211_EML_CAP_EMLSR_TRANSITION_DELAY_16US 1 +#define IEEE80211_EML_CAP_EMLSR_TRANSITION_DELAY_32US 2 +#define IEEE80211_EML_CAP_EMLSR_TRANSITION_DELAY_64US 3 +#define IEEE80211_EML_CAP_EMLSR_TRANSITION_DELAY_128US 4 +#define IEEE80211_EML_CAP_EMLSR_TRANSITION_DELAY_256US 5 +#define IEEE80211_EML_CAP_EMLMR_SUPPORT 0x0080 +#define IEEE80211_EML_CAP_EMLMR_DELAY 0x0700 +#define IEEE80211_EML_CAP_EMLMR_DELAY_0US 0 +#define IEEE80211_EML_CAP_EMLMR_DELAY_32US 1 +#define IEEE80211_EML_CAP_EMLMR_DELAY_64US 2 +#define IEEE80211_EML_CAP_EMLMR_DELAY_128US 3 +#define IEEE80211_EML_CAP_EMLMR_DELAY_256US 4 +#define IEEE80211_EML_CAP_TRANSITION_TIMEOUT 0x7800 +#define IEEE80211_EML_CAP_TRANSITION_TIMEOUT_0 0 +#define IEEE80211_EML_CAP_TRANSITION_TIMEOUT_128US 1 +#define IEEE80211_EML_CAP_TRANSITION_TIMEOUT_256US 2 +#define IEEE80211_EML_CAP_TRANSITION_TIMEOUT_512US 3 +#define IEEE80211_EML_CAP_TRANSITION_TIMEOUT_1TU 4 +#define IEEE80211_EML_CAP_TRANSITION_TIMEOUT_2TU 5 +#define IEEE80211_EML_CAP_TRANSITION_TIMEOUT_4TU 6 +#define IEEE80211_EML_CAP_TRANSITION_TIMEOUT_8TU 7 +#define IEEE80211_EML_CAP_TRANSITION_TIMEOUT_16TU 8 +#define IEEE80211_EML_CAP_TRANSITION_TIMEOUT_32TU 9 +#define IEEE80211_EML_CAP_TRANSITION_TIMEOUT_64TU 10 +#define IEEE80211_EML_CAP_TRANSITION_TIMEOUT_128TU 11 + +#define IEEE80211_MLD_CAP_OP_MAX_SIMUL_LINKS 0x000f +#define IEEE80211_MLD_CAP_OP_SRS_SUPPORT 0x0010 +#define IEEE80211_MLD_CAP_OP_TID_TO_LINK_MAP_NEG_SUPP 0x0060 +#define IEEE80211_MLD_CAP_OP_TID_TO_LINK_MAP_NEG_NO_SUPP 0 +#define IEEE80211_MLD_CAP_OP_TID_TO_LINK_MAP_NEG_SUPP_SAME 1 +#define IEEE80211_MLD_CAP_OP_TID_TO_LINK_MAP_NEG_RESERVED 2 +#define IEEE80211_MLD_CAP_OP_TID_TO_LINK_MAP_NEG_SUPP_DIFF 3 +#define IEEE80211_MLD_CAP_OP_FREQ_SEP_TYPE_IND 0x0f80 +#define IEEE80211_MLD_CAP_OP_AAR_SUPPORT 0x1000 +#define IEEE80211_MLD_CAP_OP_LINK_RECONF_SUPPORT 0x2000 +#define IEEE80211_MLD_CAP_OP_ALIGNED_TWT_SUPPORT 0x4000 + +struct ieee80211_mle_basic_common_info { + u8 len; + u8 mld_mac_addr[ETH_ALEN]; + u8 variable[]; +} __packed; + +#define IEEE80211_MLC_PREQ_PRES_MLD_ID 0x0010 + +struct ieee80211_mle_preq_common_info { + u8 len; + u8 variable[]; +} __packed; + +#define IEEE80211_MLC_RECONF_PRES_MLD_MAC_ADDR 0x0010 +#define IEEE80211_MLC_RECONF_PRES_EML_CAPA 0x0020 +#define IEEE80211_MLC_RECONF_PRES_MLD_CAPA_OP 0x0040 +#define IEEE80211_MLC_RECONF_PRES_EXT_MLD_CAPA_OP 0x0080 + +/* no fixed fields in RECONF */ + +struct ieee80211_mle_tdls_common_info { + u8 len; + u8 ap_mld_mac_addr[ETH_ALEN]; +} __packed; + +#define IEEE80211_MLC_PRIO_ACCESS_PRES_AP_MLD_MAC_ADDR 0x0010 + +/* no fixed fields in PRIO_ACCESS */ + +/** + * ieee80211_mle_common_size - check multi-link element common size + * @data: multi-link element, must already be checked for size using + * ieee80211_mle_size_ok() + * Return: the size of the multi-link element's "common" subfield + */ +static inline u8 ieee80211_mle_common_size(const u8 *data) +{ + const struct ieee80211_multi_link_elem *mle = (const void *)data; + u16 control = le16_to_cpu(mle->control); + + switch (u16_get_bits(control, IEEE80211_ML_CONTROL_TYPE)) { + case IEEE80211_ML_CONTROL_TYPE_BASIC: + case IEEE80211_ML_CONTROL_TYPE_PREQ: + case IEEE80211_ML_CONTROL_TYPE_TDLS: + case IEEE80211_ML_CONTROL_TYPE_RECONF: + case IEEE80211_ML_CONTROL_TYPE_PRIO_ACCESS: + /* + * The length is the first octet pointed by mle->variable so no + * need to add anything + */ + break; + default: + WARN_ON(1); + return 0; + } + + return sizeof(*mle) + mle->variable[0]; +} + +/** + * ieee80211_mle_get_link_id - returns the link ID + * @data: the basic multi link element + * Return: the link ID, or -1 if not present + * + * The element is assumed to be of the correct type (BASIC) and big enough, + * this must be checked using ieee80211_mle_type_ok(). + */ +static inline int ieee80211_mle_get_link_id(const u8 *data) +{ + const struct ieee80211_multi_link_elem *mle = (const void *)data; + u16 control = le16_to_cpu(mle->control); + const u8 *common = mle->variable; + + /* common points now at the beginning of ieee80211_mle_basic_common_info */ + common += sizeof(struct ieee80211_mle_basic_common_info); + + if (!(control & IEEE80211_MLC_BASIC_PRES_LINK_ID)) + return -1; + + return *common; +} + +/** + * ieee80211_mle_get_bss_param_ch_cnt - returns the BSS parameter change count + * @data: pointer to the basic multi link element + * Return: the BSS Parameter Change Count field value, or -1 if not present + * + * The element is assumed to be of the correct type (BASIC) and big enough, + * this must be checked using ieee80211_mle_type_ok(). + */ +static inline int +ieee80211_mle_get_bss_param_ch_cnt(const u8 *data) +{ + const struct ieee80211_multi_link_elem *mle = (const void *)data; + u16 control = le16_to_cpu(mle->control); + const u8 *common = mle->variable; + + /* common points now at the beginning of ieee80211_mle_basic_common_info */ + common += sizeof(struct ieee80211_mle_basic_common_info); + + if (!(control & IEEE80211_MLC_BASIC_PRES_BSS_PARAM_CH_CNT)) + return -1; + + if (control & IEEE80211_MLC_BASIC_PRES_LINK_ID) + common += 1; + + return *common; +} + +/** + * ieee80211_mle_get_eml_med_sync_delay - returns the medium sync delay + * @data: pointer to the multi-link element + * Return: the medium synchronization delay field value from the multi-link + * element, or the default value (%IEEE80211_MED_SYNC_DELAY_DEFAULT) + * if not present + * + * The element is assumed to be of the correct type (BASIC) and big enough, + * this must be checked using ieee80211_mle_type_ok(). + */ +static inline u16 ieee80211_mle_get_eml_med_sync_delay(const u8 *data) +{ + const struct ieee80211_multi_link_elem *mle = (const void *)data; + u16 control = le16_to_cpu(mle->control); + const u8 *common = mle->variable; + + /* common points now at the beginning of ieee80211_mle_basic_common_info */ + common += sizeof(struct ieee80211_mle_basic_common_info); + + if (!(control & IEEE80211_MLC_BASIC_PRES_MED_SYNC_DELAY)) + return IEEE80211_MED_SYNC_DELAY_DEFAULT; + + if (control & IEEE80211_MLC_BASIC_PRES_LINK_ID) + common += 1; + if (control & IEEE80211_MLC_BASIC_PRES_BSS_PARAM_CH_CNT) + common += 1; + + return get_unaligned_le16(common); +} + +/** + * ieee80211_mle_get_eml_cap - returns the EML capability + * @data: pointer to the multi-link element + * Return: the EML capability field value from the multi-link element, + * or 0 if not present + * + * The element is assumed to be of the correct type (BASIC) and big enough, + * this must be checked using ieee80211_mle_type_ok(). + */ +static inline u16 ieee80211_mle_get_eml_cap(const u8 *data) +{ + const struct ieee80211_multi_link_elem *mle = (const void *)data; + u16 control = le16_to_cpu(mle->control); + const u8 *common = mle->variable; + + /* common points now at the beginning of ieee80211_mle_basic_common_info */ + common += sizeof(struct ieee80211_mle_basic_common_info); + + if (!(control & IEEE80211_MLC_BASIC_PRES_EML_CAPA)) + return 0; + + if (control & IEEE80211_MLC_BASIC_PRES_LINK_ID) + common += 1; + if (control & IEEE80211_MLC_BASIC_PRES_BSS_PARAM_CH_CNT) + common += 1; + if (control & IEEE80211_MLC_BASIC_PRES_MED_SYNC_DELAY) + common += 2; + + return get_unaligned_le16(common); +} + +/** + * ieee80211_mle_get_mld_capa_op - returns the MLD capabilities and operations. + * @data: pointer to the multi-link element + * Return: the MLD capabilities and operations field value from the multi-link + * element, or 0 if not present + * + * The element is assumed to be of the correct type (BASIC) and big enough, + * this must be checked using ieee80211_mle_type_ok(). + */ +static inline u16 ieee80211_mle_get_mld_capa_op(const u8 *data) +{ + const struct ieee80211_multi_link_elem *mle = (const void *)data; + u16 control = le16_to_cpu(mle->control); + const u8 *common = mle->variable; + + /* + * common points now at the beginning of + * ieee80211_mle_basic_common_info + */ + common += sizeof(struct ieee80211_mle_basic_common_info); + + if (!(control & IEEE80211_MLC_BASIC_PRES_MLD_CAPA_OP)) + return 0; + + if (control & IEEE80211_MLC_BASIC_PRES_LINK_ID) + common += 1; + if (control & IEEE80211_MLC_BASIC_PRES_BSS_PARAM_CH_CNT) + common += 1; + if (control & IEEE80211_MLC_BASIC_PRES_MED_SYNC_DELAY) + common += 2; + if (control & IEEE80211_MLC_BASIC_PRES_EML_CAPA) + common += 2; + + return get_unaligned_le16(common); +} + +/* Defined in Figure 9-1074t in P802.11be_D7.0 */ +#define IEEE80211_EHT_ML_EXT_MLD_CAPA_OP_PARAM_UPDATE 0x0001 +#define IEEE80211_EHT_ML_EXT_MLD_CAPA_OP_RECO_MAX_LINKS_MASK 0x001e +#define IEEE80211_EHT_ML_EXT_MLD_CAPA_NSTR_UPDATE 0x0020 +#define IEEE80211_EHT_ML_EXT_MLD_CAPA_EMLSR_ENA_ON_ONE_LINK 0x0040 +#define IEEE80211_EHT_ML_EXT_MLD_CAPA_BTM_MLD_RECO_MULTI_AP 0x0080 + +/** + * ieee80211_mle_get_ext_mld_capa_op - returns the extended MLD capabilities + * and operations. + * @data: pointer to the multi-link element + * Return: the extended MLD capabilities and operations field value from + * the multi-link element, or 0 if not present + * + * The element is assumed to be of the correct type (BASIC) and big enough, + * this must be checked using ieee80211_mle_type_ok(). + */ +static inline u16 ieee80211_mle_get_ext_mld_capa_op(const u8 *data) +{ + const struct ieee80211_multi_link_elem *mle = (const void *)data; + u16 control = le16_to_cpu(mle->control); + const u8 *common = mle->variable; + + /* + * common points now at the beginning of + * ieee80211_mle_basic_common_info + */ + common += sizeof(struct ieee80211_mle_basic_common_info); + + if (!(control & IEEE80211_MLC_BASIC_PRES_EXT_MLD_CAPA_OP)) + return 0; + + if (control & IEEE80211_MLC_BASIC_PRES_LINK_ID) + common += 1; + if (control & IEEE80211_MLC_BASIC_PRES_BSS_PARAM_CH_CNT) + common += 1; + if (control & IEEE80211_MLC_BASIC_PRES_MED_SYNC_DELAY) + common += 2; + if (control & IEEE80211_MLC_BASIC_PRES_EML_CAPA) + common += 2; + if (control & IEEE80211_MLC_BASIC_PRES_MLD_CAPA_OP) + common += 2; + if (control & IEEE80211_MLC_BASIC_PRES_MLD_ID) + common += 1; + + return get_unaligned_le16(common); +} + +/** + * ieee80211_mle_get_mld_id - returns the MLD ID + * @data: pointer to the multi-link element + * Return: The MLD ID in the given multi-link element, or 0 if not present + * + * The element is assumed to be of the correct type (BASIC) and big enough, + * this must be checked using ieee80211_mle_type_ok(). + */ +static inline u8 ieee80211_mle_get_mld_id(const u8 *data) +{ + const struct ieee80211_multi_link_elem *mle = (const void *)data; + u16 control = le16_to_cpu(mle->control); + const u8 *common = mle->variable; + + /* + * common points now at the beginning of + * ieee80211_mle_basic_common_info + */ + common += sizeof(struct ieee80211_mle_basic_common_info); + + if (!(control & IEEE80211_MLC_BASIC_PRES_MLD_ID)) + return 0; + + if (control & IEEE80211_MLC_BASIC_PRES_LINK_ID) + common += 1; + if (control & IEEE80211_MLC_BASIC_PRES_BSS_PARAM_CH_CNT) + common += 1; + if (control & IEEE80211_MLC_BASIC_PRES_MED_SYNC_DELAY) + common += 2; + if (control & IEEE80211_MLC_BASIC_PRES_EML_CAPA) + common += 2; + if (control & IEEE80211_MLC_BASIC_PRES_MLD_CAPA_OP) + common += 2; + + return *common; +} + +/** + * ieee80211_mle_size_ok - validate multi-link element size + * @data: pointer to the element data + * @len: length of the containing element + * Return: whether or not the multi-link element size is OK + */ +static inline bool ieee80211_mle_size_ok(const u8 *data, size_t len) +{ + const struct ieee80211_multi_link_elem *mle = (const void *)data; + u8 fixed = sizeof(*mle); + u8 common = 0; + bool check_common_len = false; + u16 control; + + if (!data || len < fixed) + return false; + + control = le16_to_cpu(mle->control); + + switch (u16_get_bits(control, IEEE80211_ML_CONTROL_TYPE)) { + case IEEE80211_ML_CONTROL_TYPE_BASIC: + common += sizeof(struct ieee80211_mle_basic_common_info); + check_common_len = true; + if (control & IEEE80211_MLC_BASIC_PRES_LINK_ID) + common += 1; + if (control & IEEE80211_MLC_BASIC_PRES_BSS_PARAM_CH_CNT) + common += 1; + if (control & IEEE80211_MLC_BASIC_PRES_MED_SYNC_DELAY) + common += 2; + if (control & IEEE80211_MLC_BASIC_PRES_EML_CAPA) + common += 2; + if (control & IEEE80211_MLC_BASIC_PRES_MLD_CAPA_OP) + common += 2; + if (control & IEEE80211_MLC_BASIC_PRES_MLD_ID) + common += 1; + if (control & IEEE80211_MLC_BASIC_PRES_EXT_MLD_CAPA_OP) + common += 2; + break; + case IEEE80211_ML_CONTROL_TYPE_PREQ: + common += sizeof(struct ieee80211_mle_preq_common_info); + if (control & IEEE80211_MLC_PREQ_PRES_MLD_ID) + common += 1; + check_common_len = true; + break; + case IEEE80211_ML_CONTROL_TYPE_RECONF: + if (control & IEEE80211_MLC_RECONF_PRES_MLD_MAC_ADDR) + common += ETH_ALEN; + if (control & IEEE80211_MLC_RECONF_PRES_EML_CAPA) + common += 2; + if (control & IEEE80211_MLC_RECONF_PRES_MLD_CAPA_OP) + common += 2; + if (control & IEEE80211_MLC_RECONF_PRES_EXT_MLD_CAPA_OP) + common += 2; + break; + case IEEE80211_ML_CONTROL_TYPE_TDLS: + common += sizeof(struct ieee80211_mle_tdls_common_info); + check_common_len = true; + break; + case IEEE80211_ML_CONTROL_TYPE_PRIO_ACCESS: + common = ETH_ALEN + 1; + break; + default: + /* we don't know this type */ + return true; + } + + if (len < fixed + common) + return false; + + if (!check_common_len) + return true; + + /* if present, common length is the first octet there */ + return mle->variable[0] >= common; +} + +/** + * ieee80211_mle_type_ok - validate multi-link element type and size + * @data: pointer to the element data + * @type: expected type of the element + * @len: length of the containing element + * Return: whether or not the multi-link element type matches and size is OK + */ +static inline bool ieee80211_mle_type_ok(const u8 *data, u8 type, size_t len) +{ + const struct ieee80211_multi_link_elem *mle = (const void *)data; + u16 control; + + if (!ieee80211_mle_size_ok(data, len)) + return false; + + control = le16_to_cpu(mle->control); + + if (u16_get_bits(control, IEEE80211_ML_CONTROL_TYPE) == type) + return true; + + return false; +} + +enum ieee80211_mle_subelems { + IEEE80211_MLE_SUBELEM_PER_STA_PROFILE = 0, + IEEE80211_MLE_SUBELEM_FRAGMENT = 254, +}; + +#define IEEE80211_MLE_STA_CONTROL_LINK_ID 0x000f +#define IEEE80211_MLE_STA_CONTROL_COMPLETE_PROFILE 0x0010 +#define IEEE80211_MLE_STA_CONTROL_STA_MAC_ADDR_PRESENT 0x0020 +#define IEEE80211_MLE_STA_CONTROL_BEACON_INT_PRESENT 0x0040 +#define IEEE80211_MLE_STA_CONTROL_TSF_OFFS_PRESENT 0x0080 +#define IEEE80211_MLE_STA_CONTROL_DTIM_INFO_PRESENT 0x0100 +#define IEEE80211_MLE_STA_CONTROL_NSTR_LINK_PAIR_PRESENT 0x0200 +#define IEEE80211_MLE_STA_CONTROL_NSTR_BITMAP_SIZE 0x0400 +#define IEEE80211_MLE_STA_CONTROL_BSS_PARAM_CHANGE_CNT_PRESENT 0x0800 + +struct ieee80211_mle_per_sta_profile { + __le16 control; + u8 sta_info_len; + u8 variable[]; +} __packed; + +/** + * ieee80211_mle_basic_sta_prof_size_ok - validate basic multi-link element sta + * profile size + * @data: pointer to the sub element data + * @len: length of the containing sub element + * Return: %true if the STA profile is large enough, %false otherwise + */ +static inline bool ieee80211_mle_basic_sta_prof_size_ok(const u8 *data, + size_t len) +{ + const struct ieee80211_mle_per_sta_profile *prof = (const void *)data; + u16 control; + u8 fixed = sizeof(*prof); + u8 info_len = 1; + + if (len < fixed) + return false; + + control = le16_to_cpu(prof->control); + + if (control & IEEE80211_MLE_STA_CONTROL_STA_MAC_ADDR_PRESENT) + info_len += 6; + if (control & IEEE80211_MLE_STA_CONTROL_BEACON_INT_PRESENT) + info_len += 2; + if (control & IEEE80211_MLE_STA_CONTROL_TSF_OFFS_PRESENT) + info_len += 8; + if (control & IEEE80211_MLE_STA_CONTROL_DTIM_INFO_PRESENT) + info_len += 2; + if (control & IEEE80211_MLE_STA_CONTROL_COMPLETE_PROFILE && + control & IEEE80211_MLE_STA_CONTROL_NSTR_LINK_PAIR_PRESENT) { + if (control & IEEE80211_MLE_STA_CONTROL_NSTR_BITMAP_SIZE) + info_len += 2; + else + info_len += 1; + } + if (control & IEEE80211_MLE_STA_CONTROL_BSS_PARAM_CHANGE_CNT_PRESENT) + info_len += 1; + + return prof->sta_info_len >= info_len && + fixed + prof->sta_info_len - 1 <= len; +} + +/** + * ieee80211_mle_basic_sta_prof_bss_param_ch_cnt - get per-STA profile BSS + * parameter change count + * @prof: the per-STA profile, having been checked with + * ieee80211_mle_basic_sta_prof_size_ok() for the correct length + * + * Return: The BSS parameter change count value if present, 0 otherwise. + */ +static inline u8 +ieee80211_mle_basic_sta_prof_bss_param_ch_cnt(const struct ieee80211_mle_per_sta_profile *prof) +{ + u16 control = le16_to_cpu(prof->control); + const u8 *pos = prof->variable; + + if (!(control & IEEE80211_MLE_STA_CONTROL_BSS_PARAM_CHANGE_CNT_PRESENT)) + return 0; + + if (control & IEEE80211_MLE_STA_CONTROL_STA_MAC_ADDR_PRESENT) + pos += 6; + if (control & IEEE80211_MLE_STA_CONTROL_BEACON_INT_PRESENT) + pos += 2; + if (control & IEEE80211_MLE_STA_CONTROL_TSF_OFFS_PRESENT) + pos += 8; + if (control & IEEE80211_MLE_STA_CONTROL_DTIM_INFO_PRESENT) + pos += 2; + if (control & IEEE80211_MLE_STA_CONTROL_COMPLETE_PROFILE && + control & IEEE80211_MLE_STA_CONTROL_NSTR_LINK_PAIR_PRESENT) { + if (control & IEEE80211_MLE_STA_CONTROL_NSTR_BITMAP_SIZE) + pos += 2; + else + pos += 1; + } + + return *pos; +} + +#define IEEE80211_MLE_STA_RECONF_CONTROL_LINK_ID 0x000f +#define IEEE80211_MLE_STA_RECONF_CONTROL_COMPLETE_PROFILE 0x0010 +#define IEEE80211_MLE_STA_RECONF_CONTROL_STA_MAC_ADDR_PRESENT 0x0020 +#define IEEE80211_MLE_STA_RECONF_CONTROL_AP_REM_TIMER_PRESENT 0x0040 +#define IEEE80211_MLE_STA_RECONF_CONTROL_OPERATION_TYPE 0x0780 +#define IEEE80211_MLE_STA_RECONF_CONTROL_OPERATION_TYPE_AP_REM 0 +#define IEEE80211_MLE_STA_RECONF_CONTROL_OPERATION_TYPE_OP_PARAM_UPDATE 1 +#define IEEE80211_MLE_STA_RECONF_CONTROL_OPERATION_TYPE_ADD_LINK 2 +#define IEEE80211_MLE_STA_RECONF_CONTROL_OPERATION_TYPE_DEL_LINK 3 +#define IEEE80211_MLE_STA_RECONF_CONTROL_OPERATION_TYPE_NSTR_STATUS 4 +#define IEEE80211_MLE_STA_RECONF_CONTROL_OPERATION_PARAMS_PRESENT 0x0800 + +/** + * ieee80211_mle_reconf_sta_prof_size_ok - validate reconfiguration multi-link + * element sta profile size. + * @data: pointer to the sub element data + * @len: length of the containing sub element + * Return: %true if the STA profile is large enough, %false otherwise + */ +static inline bool ieee80211_mle_reconf_sta_prof_size_ok(const u8 *data, + size_t len) +{ + const struct ieee80211_mle_per_sta_profile *prof = (const void *)data; + u16 control; + u8 fixed = sizeof(*prof); + u8 info_len = 1; + + if (len < fixed) + return false; + + control = le16_to_cpu(prof->control); + + if (control & IEEE80211_MLE_STA_RECONF_CONTROL_STA_MAC_ADDR_PRESENT) + info_len += ETH_ALEN; + if (control & IEEE80211_MLE_STA_RECONF_CONTROL_AP_REM_TIMER_PRESENT) + info_len += 2; + if (control & IEEE80211_MLE_STA_RECONF_CONTROL_OPERATION_PARAMS_PRESENT) + info_len += 2; + + return prof->sta_info_len >= info_len && + fixed + prof->sta_info_len - 1 <= len; +} + +#define IEEE80211_MLE_STA_EPCS_CONTROL_LINK_ID 0x000f +#define IEEE80211_EPCS_ENA_RESP_BODY_LEN 3 + +static inline bool ieee80211_tid_to_link_map_size_ok(const u8 *data, size_t len) +{ + const struct ieee80211_ttlm_elem *t2l = (const void *)data; + u8 control, fixed = sizeof(*t2l), elem_len = 0; + + if (len < fixed) + return false; + + control = t2l->control; + + if (control & IEEE80211_TTLM_CONTROL_SWITCH_TIME_PRESENT) + elem_len += 2; + if (control & IEEE80211_TTLM_CONTROL_EXPECTED_DUR_PRESENT) + elem_len += 3; + + if (!(control & IEEE80211_TTLM_CONTROL_DEF_LINK_MAP)) { + u8 bm_size; + + elem_len += 1; + if (len < fixed + elem_len) + return false; + + if (control & IEEE80211_TTLM_CONTROL_LINK_MAP_SIZE) + bm_size = 1; + else + bm_size = 2; + + elem_len += hweight8(t2l->optional[0]) * bm_size; + } + + return len >= fixed + elem_len; +} + +/** + * ieee80211_emlsr_pad_delay_in_us - Fetch the EMLSR Padding delay + * in microseconds + * @eml_cap: EML capabilities field value from common info field of + * the Multi-link element + * Return: the EMLSR Padding delay (in microseconds) encoded in the + * EML Capabilities field + */ + +static inline u32 ieee80211_emlsr_pad_delay_in_us(u16 eml_cap) +{ + /* IEEE Std 802.11be-2024 Table 9-417i—Encoding of the EMLSR + * Padding Delay subfield. + */ + u32 pad_delay = u16_get_bits(eml_cap, + IEEE80211_EML_CAP_EMLSR_PADDING_DELAY); + + if (!pad_delay || + pad_delay > IEEE80211_EML_CAP_EMLSR_PADDING_DELAY_256US) + return 0; + + return 32 * (1 << (pad_delay - 1)); +} + +/** + * ieee80211_emlsr_trans_delay_in_us - Fetch the EMLSR Transition + * delay in microseconds + * @eml_cap: EML capabilities field value from common info field of + * the Multi-link element + * Return: the EMLSR Transition delay (in microseconds) encoded in the + * EML Capabilities field + */ + +static inline u32 ieee80211_emlsr_trans_delay_in_us(u16 eml_cap) +{ + /* IEEE Std 802.11be-2024 Table 9-417j—Encoding of the EMLSR + * Transition Delay subfield. + */ + u32 trans_delay = + u16_get_bits(eml_cap, + IEEE80211_EML_CAP_EMLSR_TRANSITION_DELAY); + + /* invalid values also just use 0 */ + if (!trans_delay || + trans_delay > IEEE80211_EML_CAP_EMLSR_TRANSITION_DELAY_256US) + return 0; + + return 16 * (1 << (trans_delay - 1)); +} + +/** + * ieee80211_eml_trans_timeout_in_us - Fetch the EMLSR Transition + * timeout value in microseconds + * @eml_cap: EML capabilities field value from common info field of + * the Multi-link element + * Return: the EMLSR Transition timeout (in microseconds) encoded in + * the EML Capabilities field + */ + +static inline u32 ieee80211_eml_trans_timeout_in_us(u16 eml_cap) +{ + /* IEEE Std 802.11be-2024 Table 9-417m—Encoding of the + * Transition Timeout subfield. + */ + u8 timeout = u16_get_bits(eml_cap, + IEEE80211_EML_CAP_TRANSITION_TIMEOUT); + + /* invalid values also just use 0 */ + if (!timeout || timeout > IEEE80211_EML_CAP_TRANSITION_TIMEOUT_128TU) + return 0; + + return 128 * (1 << (timeout - 1)); +} + +#define for_each_mle_subelement(_elem, _data, _len) \ + if (ieee80211_mle_size_ok(_data, _len)) \ + for_each_element(_elem, \ + _data + ieee80211_mle_common_size(_data),\ + _len - ieee80211_mle_common_size(_data)) + +#endif /* LINUX_IEEE80211_H */ diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h index a3dbbcee00ee..63a9775b059d 100644 --- a/include/linux/ieee80211.h +++ b/include/linux/ieee80211.h @@ -1141,30 +1141,6 @@ ieee80211_s1g_optional_len(__le16 fc) return len; } -#define IEEE80211_TTLM_MAX_CNT 2 -#define IEEE80211_TTLM_CONTROL_DIRECTION 0x03 -#define IEEE80211_TTLM_CONTROL_DEF_LINK_MAP 0x04 -#define IEEE80211_TTLM_CONTROL_SWITCH_TIME_PRESENT 0x08 -#define IEEE80211_TTLM_CONTROL_EXPECTED_DUR_PRESENT 0x10 -#define IEEE80211_TTLM_CONTROL_LINK_MAP_SIZE 0x20 - -#define IEEE80211_TTLM_DIRECTION_DOWN 0 -#define IEEE80211_TTLM_DIRECTION_UP 1 -#define IEEE80211_TTLM_DIRECTION_BOTH 2 - -/** - * struct ieee80211_ttlm_elem - TID-To-Link Mapping element - * - * Defined in section 9.4.2.314 in P802.11be_D4 - * - * @control: the first part of control field - * @optional: the second part of control field - */ -struct ieee80211_ttlm_elem { - u8 control; - u8 optional[]; -} __packed; - /** * struct ieee80211_bss_load_elem - BSS Load elemen * @@ -1591,144 +1567,6 @@ struct ieee80211_p2p_noa_attr { #define IEEE80211_P2P_OPPPS_ENABLE_BIT BIT(7) #define IEEE80211_P2P_OPPPS_CTWINDOW_MASK 0x7F -#define IEEE80211_EHT_MCS_NSS_RX 0x0f -#define IEEE80211_EHT_MCS_NSS_TX 0xf0 - -/** - * struct ieee80211_eht_mcs_nss_supp_20mhz_only - EHT 20MHz only station max - * supported NSS for per MCS. - * - * For each field below, bits 0 - 3 indicate the maximal number of spatial - * streams for Rx, and bits 4 - 7 indicate the maximal number of spatial streams - * for Tx. - * - * @rx_tx_mcs7_max_nss: indicates the maximum number of spatial streams - * supported for reception and the maximum number of spatial streams - * supported for transmission for MCS 0 - 7. - * @rx_tx_mcs9_max_nss: indicates the maximum number of spatial streams - * supported for reception and the maximum number of spatial streams - * supported for transmission for MCS 8 - 9. - * @rx_tx_mcs11_max_nss: indicates the maximum number of spatial streams - * supported for reception and the maximum number of spatial streams - * supported for transmission for MCS 10 - 11. - * @rx_tx_mcs13_max_nss: indicates the maximum number of spatial streams - * supported for reception and the maximum number of spatial streams - * supported for transmission for MCS 12 - 13. - * @rx_tx_max_nss: array of the previous fields for easier loop access - */ -struct ieee80211_eht_mcs_nss_supp_20mhz_only { - union { - struct { - u8 rx_tx_mcs7_max_nss; - u8 rx_tx_mcs9_max_nss; - u8 rx_tx_mcs11_max_nss; - u8 rx_tx_mcs13_max_nss; - }; - u8 rx_tx_max_nss[4]; - }; -}; - -/** - * struct ieee80211_eht_mcs_nss_supp_bw - EHT max supported NSS per MCS (except - * 20MHz only stations). - * - * For each field below, bits 0 - 3 indicate the maximal number of spatial - * streams for Rx, and bits 4 - 7 indicate the maximal number of spatial streams - * for Tx. - * - * @rx_tx_mcs9_max_nss: indicates the maximum number of spatial streams - * supported for reception and the maximum number of spatial streams - * supported for transmission for MCS 0 - 9. - * @rx_tx_mcs11_max_nss: indicates the maximum number of spatial streams - * supported for reception and the maximum number of spatial streams - * supported for transmission for MCS 10 - 11. - * @rx_tx_mcs13_max_nss: indicates the maximum number of spatial streams - * supported for reception and the maximum number of spatial streams - * supported for transmission for MCS 12 - 13. - * @rx_tx_max_nss: array of the previous fields for easier loop access - */ -struct ieee80211_eht_mcs_nss_supp_bw { - union { - struct { - u8 rx_tx_mcs9_max_nss; - u8 rx_tx_mcs11_max_nss; - u8 rx_tx_mcs13_max_nss; - }; - u8 rx_tx_max_nss[3]; - }; -}; - -/** - * struct ieee80211_eht_cap_elem_fixed - EHT capabilities fixed data - * - * This structure is the "EHT Capabilities element" fixed fields as - * described in P802.11be_D2.0 section 9.4.2.313. - * - * @mac_cap_info: MAC capabilities, see IEEE80211_EHT_MAC_CAP* - * @phy_cap_info: PHY capabilities, see IEEE80211_EHT_PHY_CAP* - */ -struct ieee80211_eht_cap_elem_fixed { - u8 mac_cap_info[2]; - u8 phy_cap_info[9]; -} __packed; - -/** - * struct ieee80211_eht_cap_elem - EHT capabilities element - * @fixed: fixed parts, see &ieee80211_eht_cap_elem_fixed - * @optional: optional parts - */ -struct ieee80211_eht_cap_elem { - struct ieee80211_eht_cap_elem_fixed fixed; - - /* - * Followed by: - * Supported EHT-MCS And NSS Set field: 4, 3, 6 or 9 octets. - * EHT PPE Thresholds field: variable length. - */ - u8 optional[]; -} __packed; - -#define IEEE80211_EHT_OPER_INFO_PRESENT 0x01 -#define IEEE80211_EHT_OPER_DISABLED_SUBCHANNEL_BITMAP_PRESENT 0x02 -#define IEEE80211_EHT_OPER_EHT_DEF_PE_DURATION 0x04 -#define IEEE80211_EHT_OPER_GROUP_ADDRESSED_BU_IND_LIMIT 0x08 -#define IEEE80211_EHT_OPER_GROUP_ADDRESSED_BU_IND_EXP_MASK 0x30 -#define IEEE80211_EHT_OPER_MCS15_DISABLE 0x40 - -/** - * struct ieee80211_eht_operation - eht operation element - * - * This structure is the "EHT Operation Element" fields as - * described in P802.11be_D2.0 section 9.4.2.311 - * - * @params: EHT operation element parameters. See &IEEE80211_EHT_OPER_* - * @basic_mcs_nss: indicates the EHT-MCSs for each number of spatial streams in - * EHT PPDUs that are supported by all EHT STAs in the BSS in transmit and - * receive. - * @optional: optional parts - */ -struct ieee80211_eht_operation { - u8 params; - struct ieee80211_eht_mcs_nss_supp_20mhz_only basic_mcs_nss; - u8 optional[]; -} __packed; - -/** - * struct ieee80211_eht_operation_info - eht operation information - * - * @control: EHT operation information control. - * @ccfs0: defines a channel center frequency for a 20, 40, 80, 160, or 320 MHz - * EHT BSS. - * @ccfs1: defines a channel center frequency for a 160 or 320 MHz EHT BSS. - * @optional: optional parts - */ -struct ieee80211_eht_operation_info { - u8 control; - u8 ccfs0; - u8 ccfs1; - u8 optional[]; -} __packed; - /* S1G Capabilities Information field */ #define IEEE80211_S1G_CAPABILITY_LEN 15 @@ -1815,258 +1653,6 @@ struct ieee80211_eht_operation_info { #define S1G_2M_PRIMARY_LOCATION_LOWER 0 #define S1G_2M_PRIMARY_LOCATION_UPPER 1 -/* EHT MAC capabilities as defined in P802.11be_D2.0 section 9.4.2.313.2 */ -#define IEEE80211_EHT_MAC_CAP0_EPCS_PRIO_ACCESS 0x01 -#define IEEE80211_EHT_MAC_CAP0_OM_CONTROL 0x02 -#define IEEE80211_EHT_MAC_CAP0_TRIG_TXOP_SHARING_MODE1 0x04 -#define IEEE80211_EHT_MAC_CAP0_TRIG_TXOP_SHARING_MODE2 0x08 -#define IEEE80211_EHT_MAC_CAP0_RESTRICTED_TWT 0x10 -#define IEEE80211_EHT_MAC_CAP0_SCS_TRAFFIC_DESC 0x20 -#define IEEE80211_EHT_MAC_CAP0_MAX_MPDU_LEN_MASK 0xc0 -#define IEEE80211_EHT_MAC_CAP0_MAX_MPDU_LEN_3895 0 -#define IEEE80211_EHT_MAC_CAP0_MAX_MPDU_LEN_7991 1 -#define IEEE80211_EHT_MAC_CAP0_MAX_MPDU_LEN_11454 2 - -#define IEEE80211_EHT_MAC_CAP1_MAX_AMPDU_LEN_MASK 0x01 -#define IEEE80211_EHT_MAC_CAP1_EHT_TRS 0x02 -#define IEEE80211_EHT_MAC_CAP1_TXOP_RET 0x04 -#define IEEE80211_EHT_MAC_CAP1_TWO_BQRS 0x08 -#define IEEE80211_EHT_MAC_CAP1_EHT_LINK_ADAPT_MASK 0x30 -#define IEEE80211_EHT_MAC_CAP1_UNSOL_EPCS_PRIO_ACCESS 0x40 - -/* EHT PHY capabilities as defined in P802.11be_D2.0 section 9.4.2.313.3 */ -#define IEEE80211_EHT_PHY_CAP0_320MHZ_IN_6GHZ 0x02 -#define IEEE80211_EHT_PHY_CAP0_242_TONE_RU_GT20MHZ 0x04 -#define IEEE80211_EHT_PHY_CAP0_NDP_4_EHT_LFT_32_GI 0x08 -#define IEEE80211_EHT_PHY_CAP0_PARTIAL_BW_UL_MU_MIMO 0x10 -#define IEEE80211_EHT_PHY_CAP0_SU_BEAMFORMER 0x20 -#define IEEE80211_EHT_PHY_CAP0_SU_BEAMFORMEE 0x40 - -/* EHT beamformee number of spatial streams <= 80MHz is split */ -#define IEEE80211_EHT_PHY_CAP0_BEAMFORMEE_SS_80MHZ_MASK 0x80 -#define IEEE80211_EHT_PHY_CAP1_BEAMFORMEE_SS_80MHZ_MASK 0x03 - -#define IEEE80211_EHT_PHY_CAP1_BEAMFORMEE_SS_160MHZ_MASK 0x1c -#define IEEE80211_EHT_PHY_CAP1_BEAMFORMEE_SS_320MHZ_MASK 0xe0 - -#define IEEE80211_EHT_PHY_CAP2_SOUNDING_DIM_80MHZ_MASK 0x07 -#define IEEE80211_EHT_PHY_CAP2_SOUNDING_DIM_160MHZ_MASK 0x38 - -/* EHT number of sounding dimensions for 320MHz is split */ -#define IEEE80211_EHT_PHY_CAP2_SOUNDING_DIM_320MHZ_MASK 0xc0 -#define IEEE80211_EHT_PHY_CAP3_SOUNDING_DIM_320MHZ_MASK 0x01 -#define IEEE80211_EHT_PHY_CAP3_NG_16_SU_FEEDBACK 0x02 -#define IEEE80211_EHT_PHY_CAP3_NG_16_MU_FEEDBACK 0x04 -#define IEEE80211_EHT_PHY_CAP3_CODEBOOK_4_2_SU_FDBK 0x08 -#define IEEE80211_EHT_PHY_CAP3_CODEBOOK_7_5_MU_FDBK 0x10 -#define IEEE80211_EHT_PHY_CAP3_TRIG_SU_BF_FDBK 0x20 -#define IEEE80211_EHT_PHY_CAP3_TRIG_MU_BF_PART_BW_FDBK 0x40 -#define IEEE80211_EHT_PHY_CAP3_TRIG_CQI_FDBK 0x80 - -#define IEEE80211_EHT_PHY_CAP4_PART_BW_DL_MU_MIMO 0x01 -#define IEEE80211_EHT_PHY_CAP4_PSR_SR_SUPP 0x02 -#define IEEE80211_EHT_PHY_CAP4_POWER_BOOST_FACT_SUPP 0x04 -#define IEEE80211_EHT_PHY_CAP4_EHT_MU_PPDU_4_EHT_LTF_08_GI 0x08 -#define IEEE80211_EHT_PHY_CAP4_MAX_NC_MASK 0xf0 - -#define IEEE80211_EHT_PHY_CAP5_NON_TRIG_CQI_FEEDBACK 0x01 -#define IEEE80211_EHT_PHY_CAP5_TX_LESS_242_TONE_RU_SUPP 0x02 -#define IEEE80211_EHT_PHY_CAP5_RX_LESS_242_TONE_RU_SUPP 0x04 -#define IEEE80211_EHT_PHY_CAP5_PPE_THRESHOLD_PRESENT 0x08 -#define IEEE80211_EHT_PHY_CAP5_COMMON_NOMINAL_PKT_PAD_MASK 0x30 -#define IEEE80211_EHT_PHY_CAP5_COMMON_NOMINAL_PKT_PAD_0US 0 -#define IEEE80211_EHT_PHY_CAP5_COMMON_NOMINAL_PKT_PAD_8US 1 -#define IEEE80211_EHT_PHY_CAP5_COMMON_NOMINAL_PKT_PAD_16US 2 -#define IEEE80211_EHT_PHY_CAP5_COMMON_NOMINAL_PKT_PAD_20US 3 - -/* Maximum number of supported EHT LTF is split */ -#define IEEE80211_EHT_PHY_CAP5_MAX_NUM_SUPP_EHT_LTF_MASK 0xc0 -#define IEEE80211_EHT_PHY_CAP5_SUPP_EXTRA_EHT_LTF 0x40 -#define IEEE80211_EHT_PHY_CAP6_MAX_NUM_SUPP_EHT_LTF_MASK 0x07 - -#define IEEE80211_EHT_PHY_CAP6_MCS15_SUPP_80MHZ 0x08 -#define IEEE80211_EHT_PHY_CAP6_MCS15_SUPP_160MHZ 0x30 -#define IEEE80211_EHT_PHY_CAP6_MCS15_SUPP_320MHZ 0x40 -#define IEEE80211_EHT_PHY_CAP6_MCS15_SUPP_MASK 0x78 -#define IEEE80211_EHT_PHY_CAP6_EHT_DUP_6GHZ_SUPP 0x80 - -#define IEEE80211_EHT_PHY_CAP7_20MHZ_STA_RX_NDP_WIDER_BW 0x01 -#define IEEE80211_EHT_PHY_CAP7_NON_OFDMA_UL_MU_MIMO_80MHZ 0x02 -#define IEEE80211_EHT_PHY_CAP7_NON_OFDMA_UL_MU_MIMO_160MHZ 0x04 -#define IEEE80211_EHT_PHY_CAP7_NON_OFDMA_UL_MU_MIMO_320MHZ 0x08 -#define IEEE80211_EHT_PHY_CAP7_MU_BEAMFORMER_80MHZ 0x10 -#define IEEE80211_EHT_PHY_CAP7_MU_BEAMFORMER_160MHZ 0x20 -#define IEEE80211_EHT_PHY_CAP7_MU_BEAMFORMER_320MHZ 0x40 -#define IEEE80211_EHT_PHY_CAP7_TB_SOUNDING_FDBK_RATE_LIMIT 0x80 - -#define IEEE80211_EHT_PHY_CAP8_RX_1024QAM_WIDER_BW_DL_OFDMA 0x01 -#define IEEE80211_EHT_PHY_CAP8_RX_4096QAM_WIDER_BW_DL_OFDMA 0x02 - -/* - * EHT operation channel width as defined in P802.11be_D2.0 section 9.4.2.311 - */ -#define IEEE80211_EHT_OPER_CHAN_WIDTH 0x7 -#define IEEE80211_EHT_OPER_CHAN_WIDTH_20MHZ 0 -#define IEEE80211_EHT_OPER_CHAN_WIDTH_40MHZ 1 -#define IEEE80211_EHT_OPER_CHAN_WIDTH_80MHZ 2 -#define IEEE80211_EHT_OPER_CHAN_WIDTH_160MHZ 3 -#define IEEE80211_EHT_OPER_CHAN_WIDTH_320MHZ 4 - -/* need HE definitions for EHT functions */ -#include "ieee80211-he.h" - -/* Calculate 802.11be EHT capabilities IE Tx/Rx EHT MCS NSS Support Field size */ -static inline u8 -ieee80211_eht_mcs_nss_size(const struct ieee80211_he_cap_elem *he_cap, - const struct ieee80211_eht_cap_elem_fixed *eht_cap, - bool from_ap) -{ - u8 count = 0; - - /* on 2.4 GHz, if it supports 40 MHz, the result is 3 */ - if (he_cap->phy_cap_info[0] & - IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_40MHZ_IN_2G) - return 3; - - /* on 2.4 GHz, these three bits are reserved, so should be 0 */ - if (he_cap->phy_cap_info[0] & - IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_40MHZ_80MHZ_IN_5G) - count += 3; - - if (he_cap->phy_cap_info[0] & - IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_160MHZ_IN_5G) - count += 3; - - if (eht_cap->phy_cap_info[0] & IEEE80211_EHT_PHY_CAP0_320MHZ_IN_6GHZ) - count += 3; - - if (count) - return count; - - return from_ap ? 3 : 4; -} - -/* 802.11be EHT PPE Thresholds */ -#define IEEE80211_EHT_PPE_THRES_NSS_POS 0 -#define IEEE80211_EHT_PPE_THRES_NSS_MASK 0xf -#define IEEE80211_EHT_PPE_THRES_RU_INDEX_BITMASK_MASK 0x1f0 -#define IEEE80211_EHT_PPE_THRES_INFO_PPET_SIZE 3 -#define IEEE80211_EHT_PPE_THRES_INFO_HEADER_SIZE 9 - -/* - * Calculate 802.11be EHT capabilities IE EHT field size - */ -static inline u8 -ieee80211_eht_ppe_size(u16 ppe_thres_hdr, const u8 *phy_cap_info) -{ - u32 n; - - if (!(phy_cap_info[5] & - IEEE80211_EHT_PHY_CAP5_PPE_THRESHOLD_PRESENT)) - return 0; - - n = hweight16(ppe_thres_hdr & - IEEE80211_EHT_PPE_THRES_RU_INDEX_BITMASK_MASK); - n *= 1 + u16_get_bits(ppe_thres_hdr, IEEE80211_EHT_PPE_THRES_NSS_MASK); - - /* - * Each pair is 6 bits, and we need to add the 9 "header" bits to the - * total size. - */ - n = n * IEEE80211_EHT_PPE_THRES_INFO_PPET_SIZE * 2 + - IEEE80211_EHT_PPE_THRES_INFO_HEADER_SIZE; - return DIV_ROUND_UP(n, 8); -} - -static inline bool -ieee80211_eht_capa_size_ok(const u8 *he_capa, const u8 *data, u8 len, - bool from_ap) -{ - const struct ieee80211_eht_cap_elem_fixed *elem = (const void *)data; - u8 needed = sizeof(struct ieee80211_eht_cap_elem_fixed); - - if (len < needed || !he_capa) - return false; - - needed += ieee80211_eht_mcs_nss_size((const void *)he_capa, - (const void *)data, - from_ap); - if (len < needed) - return false; - - if (elem->phy_cap_info[5] & - IEEE80211_EHT_PHY_CAP5_PPE_THRESHOLD_PRESENT) { - u16 ppe_thres_hdr; - - if (len < needed + sizeof(ppe_thres_hdr)) - return false; - - ppe_thres_hdr = get_unaligned_le16(data + needed); - needed += ieee80211_eht_ppe_size(ppe_thres_hdr, - elem->phy_cap_info); - } - - return len >= needed; -} - -static inline bool -ieee80211_eht_oper_size_ok(const u8 *data, u8 len) -{ - const struct ieee80211_eht_operation *elem = (const void *)data; - u8 needed = sizeof(*elem); - - if (len < needed) - return false; - - if (elem->params & IEEE80211_EHT_OPER_INFO_PRESENT) { - needed += 3; - - if (elem->params & - IEEE80211_EHT_OPER_DISABLED_SUBCHANNEL_BITMAP_PRESENT) - needed += 2; - } - - return len >= needed; -} - -/* must validate ieee80211_eht_oper_size_ok() first */ -static inline u16 -ieee80211_eht_oper_dis_subchan_bitmap(const struct ieee80211_eht_operation *eht_oper) -{ - const struct ieee80211_eht_operation_info *info = - (const void *)eht_oper->optional; - - if (!(eht_oper->params & IEEE80211_EHT_OPER_INFO_PRESENT)) - return 0; - - if (!(eht_oper->params & IEEE80211_EHT_OPER_DISABLED_SUBCHANNEL_BITMAP_PRESENT)) - return 0; - - return get_unaligned_le16(info->optional); -} - -#define IEEE80211_BW_IND_DIS_SUBCH_PRESENT BIT(1) - -struct ieee80211_bandwidth_indication { - u8 params; - struct ieee80211_eht_operation_info info; -} __packed; - -static inline bool -ieee80211_bandwidth_indication_size_ok(const u8 *data, u8 len) -{ - const struct ieee80211_bandwidth_indication *bwi = (const void *)data; - - if (len < sizeof(*bwi)) - return false; - - if (bwi->params & IEEE80211_BW_IND_DIS_SUBCH_PRESENT && - len < sizeof(*bwi) + 2) - return false; - - return true; -} - #define LISTEN_INT_USF GENMASK(15, 14) #define LISTEN_INT_UI GENMASK(13, 0) @@ -2587,23 +2173,6 @@ enum ieee80211_unprotected_wnm_actioncode { WLAN_UNPROTECTED_WNM_ACTION_TIMING_MEASUREMENT_RESPONSE = 1, }; -/* Protected EHT action codes */ -enum ieee80211_protected_eht_actioncode { - WLAN_PROTECTED_EHT_ACTION_TTLM_REQ = 0, - WLAN_PROTECTED_EHT_ACTION_TTLM_RES = 1, - WLAN_PROTECTED_EHT_ACTION_TTLM_TEARDOWN = 2, - WLAN_PROTECTED_EHT_ACTION_EPCS_ENABLE_REQ = 3, - WLAN_PROTECTED_EHT_ACTION_EPCS_ENABLE_RESP = 4, - WLAN_PROTECTED_EHT_ACTION_EPCS_ENABLE_TEARDOWN = 5, - WLAN_PROTECTED_EHT_ACTION_EML_OP_MODE_NOTIF = 6, - WLAN_PROTECTED_EHT_ACTION_LINK_RECOMMEND = 7, - WLAN_PROTECTED_EHT_ACTION_ML_OP_UPDATE_REQ = 8, - WLAN_PROTECTED_EHT_ACTION_ML_OP_UPDATE_RESP = 9, - WLAN_PROTECTED_EHT_ACTION_LINK_RECONFIG_NOTIF = 10, - WLAN_PROTECTED_EHT_ACTION_LINK_RECONFIG_REQ = 11, - WLAN_PROTECTED_EHT_ACTION_LINK_RECONFIG_RESP = 12, -}; - /* Security key length */ enum ieee80211_key_len { WLAN_KEY_LEN_WEP40 = 5, @@ -3855,737 +3424,6 @@ struct ieee80211_tbtt_info_ge_11 { struct ieee80211_rnr_mld_params mld_params; } __packed; -/* multi-link device */ -#define IEEE80211_MLD_MAX_NUM_LINKS 15 - -#define IEEE80211_ML_CONTROL_TYPE 0x0007 -#define IEEE80211_ML_CONTROL_TYPE_BASIC 0 -#define IEEE80211_ML_CONTROL_TYPE_PREQ 1 -#define IEEE80211_ML_CONTROL_TYPE_RECONF 2 -#define IEEE80211_ML_CONTROL_TYPE_TDLS 3 -#define IEEE80211_ML_CONTROL_TYPE_PRIO_ACCESS 4 -#define IEEE80211_ML_CONTROL_PRESENCE_MASK 0xfff0 - -struct ieee80211_multi_link_elem { - __le16 control; - u8 variable[]; -} __packed; - -#define IEEE80211_MLC_BASIC_PRES_LINK_ID 0x0010 -#define IEEE80211_MLC_BASIC_PRES_BSS_PARAM_CH_CNT 0x0020 -#define IEEE80211_MLC_BASIC_PRES_MED_SYNC_DELAY 0x0040 -#define IEEE80211_MLC_BASIC_PRES_EML_CAPA 0x0080 -#define IEEE80211_MLC_BASIC_PRES_MLD_CAPA_OP 0x0100 -#define IEEE80211_MLC_BASIC_PRES_MLD_ID 0x0200 -#define IEEE80211_MLC_BASIC_PRES_EXT_MLD_CAPA_OP 0x0400 - -#define IEEE80211_MED_SYNC_DELAY_DURATION 0x00ff -#define IEEE80211_MED_SYNC_DELAY_SYNC_OFDM_ED_THRESH 0x0f00 -#define IEEE80211_MED_SYNC_DELAY_SYNC_MAX_NUM_TXOPS 0xf000 - -/* - * Described in P802.11be_D3.0 - * dot11MSDTimerDuration should default to 5484 (i.e. 171.375) - * dot11MSDOFDMEDthreshold defaults to -72 (i.e. 0) - * dot11MSDTXOPMAX defaults to 1 - */ -#define IEEE80211_MED_SYNC_DELAY_DEFAULT 0x10ac - -#define IEEE80211_EML_CAP_EMLSR_SUPP 0x0001 -#define IEEE80211_EML_CAP_EMLSR_PADDING_DELAY 0x000e -#define IEEE80211_EML_CAP_EMLSR_PADDING_DELAY_0US 0 -#define IEEE80211_EML_CAP_EMLSR_PADDING_DELAY_32US 1 -#define IEEE80211_EML_CAP_EMLSR_PADDING_DELAY_64US 2 -#define IEEE80211_EML_CAP_EMLSR_PADDING_DELAY_128US 3 -#define IEEE80211_EML_CAP_EMLSR_PADDING_DELAY_256US 4 -#define IEEE80211_EML_CAP_EMLSR_TRANSITION_DELAY 0x0070 -#define IEEE80211_EML_CAP_EMLSR_TRANSITION_DELAY_0US 0 -#define IEEE80211_EML_CAP_EMLSR_TRANSITION_DELAY_16US 1 -#define IEEE80211_EML_CAP_EMLSR_TRANSITION_DELAY_32US 2 -#define IEEE80211_EML_CAP_EMLSR_TRANSITION_DELAY_64US 3 -#define IEEE80211_EML_CAP_EMLSR_TRANSITION_DELAY_128US 4 -#define IEEE80211_EML_CAP_EMLSR_TRANSITION_DELAY_256US 5 -#define IEEE80211_EML_CAP_EMLMR_SUPPORT 0x0080 -#define IEEE80211_EML_CAP_EMLMR_DELAY 0x0700 -#define IEEE80211_EML_CAP_EMLMR_DELAY_0US 0 -#define IEEE80211_EML_CAP_EMLMR_DELAY_32US 1 -#define IEEE80211_EML_CAP_EMLMR_DELAY_64US 2 -#define IEEE80211_EML_CAP_EMLMR_DELAY_128US 3 -#define IEEE80211_EML_CAP_EMLMR_DELAY_256US 4 -#define IEEE80211_EML_CAP_TRANSITION_TIMEOUT 0x7800 -#define IEEE80211_EML_CAP_TRANSITION_TIMEOUT_0 0 -#define IEEE80211_EML_CAP_TRANSITION_TIMEOUT_128US 1 -#define IEEE80211_EML_CAP_TRANSITION_TIMEOUT_256US 2 -#define IEEE80211_EML_CAP_TRANSITION_TIMEOUT_512US 3 -#define IEEE80211_EML_CAP_TRANSITION_TIMEOUT_1TU 4 -#define IEEE80211_EML_CAP_TRANSITION_TIMEOUT_2TU 5 -#define IEEE80211_EML_CAP_TRANSITION_TIMEOUT_4TU 6 -#define IEEE80211_EML_CAP_TRANSITION_TIMEOUT_8TU 7 -#define IEEE80211_EML_CAP_TRANSITION_TIMEOUT_16TU 8 -#define IEEE80211_EML_CAP_TRANSITION_TIMEOUT_32TU 9 -#define IEEE80211_EML_CAP_TRANSITION_TIMEOUT_64TU 10 -#define IEEE80211_EML_CAP_TRANSITION_TIMEOUT_128TU 11 - -#define IEEE80211_MLD_CAP_OP_MAX_SIMUL_LINKS 0x000f -#define IEEE80211_MLD_CAP_OP_SRS_SUPPORT 0x0010 -#define IEEE80211_MLD_CAP_OP_TID_TO_LINK_MAP_NEG_SUPP 0x0060 -#define IEEE80211_MLD_CAP_OP_TID_TO_LINK_MAP_NEG_NO_SUPP 0 -#define IEEE80211_MLD_CAP_OP_TID_TO_LINK_MAP_NEG_SUPP_SAME 1 -#define IEEE80211_MLD_CAP_OP_TID_TO_LINK_MAP_NEG_RESERVED 2 -#define IEEE80211_MLD_CAP_OP_TID_TO_LINK_MAP_NEG_SUPP_DIFF 3 -#define IEEE80211_MLD_CAP_OP_FREQ_SEP_TYPE_IND 0x0f80 -#define IEEE80211_MLD_CAP_OP_AAR_SUPPORT 0x1000 -#define IEEE80211_MLD_CAP_OP_LINK_RECONF_SUPPORT 0x2000 -#define IEEE80211_MLD_CAP_OP_ALIGNED_TWT_SUPPORT 0x4000 - -struct ieee80211_mle_basic_common_info { - u8 len; - u8 mld_mac_addr[ETH_ALEN]; - u8 variable[]; -} __packed; - -#define IEEE80211_MLC_PREQ_PRES_MLD_ID 0x0010 - -struct ieee80211_mle_preq_common_info { - u8 len; - u8 variable[]; -} __packed; - -#define IEEE80211_MLC_RECONF_PRES_MLD_MAC_ADDR 0x0010 -#define IEEE80211_MLC_RECONF_PRES_EML_CAPA 0x0020 -#define IEEE80211_MLC_RECONF_PRES_MLD_CAPA_OP 0x0040 -#define IEEE80211_MLC_RECONF_PRES_EXT_MLD_CAPA_OP 0x0080 - -/* no fixed fields in RECONF */ - -struct ieee80211_mle_tdls_common_info { - u8 len; - u8 ap_mld_mac_addr[ETH_ALEN]; -} __packed; - -#define IEEE80211_MLC_PRIO_ACCESS_PRES_AP_MLD_MAC_ADDR 0x0010 - -/* no fixed fields in PRIO_ACCESS */ - -/** - * ieee80211_mle_common_size - check multi-link element common size - * @data: multi-link element, must already be checked for size using - * ieee80211_mle_size_ok() - * Return: the size of the multi-link element's "common" subfield - */ -static inline u8 ieee80211_mle_common_size(const u8 *data) -{ - const struct ieee80211_multi_link_elem *mle = (const void *)data; - u16 control = le16_to_cpu(mle->control); - - switch (u16_get_bits(control, IEEE80211_ML_CONTROL_TYPE)) { - case IEEE80211_ML_CONTROL_TYPE_BASIC: - case IEEE80211_ML_CONTROL_TYPE_PREQ: - case IEEE80211_ML_CONTROL_TYPE_TDLS: - case IEEE80211_ML_CONTROL_TYPE_RECONF: - case IEEE80211_ML_CONTROL_TYPE_PRIO_ACCESS: - /* - * The length is the first octet pointed by mle->variable so no - * need to add anything - */ - break; - default: - WARN_ON(1); - return 0; - } - - return sizeof(*mle) + mle->variable[0]; -} - -/** - * ieee80211_mle_get_link_id - returns the link ID - * @data: the basic multi link element - * Return: the link ID, or -1 if not present - * - * The element is assumed to be of the correct type (BASIC) and big enough, - * this must be checked using ieee80211_mle_type_ok(). - */ -static inline int ieee80211_mle_get_link_id(const u8 *data) -{ - const struct ieee80211_multi_link_elem *mle = (const void *)data; - u16 control = le16_to_cpu(mle->control); - const u8 *common = mle->variable; - - /* common points now at the beginning of ieee80211_mle_basic_common_info */ - common += sizeof(struct ieee80211_mle_basic_common_info); - - if (!(control & IEEE80211_MLC_BASIC_PRES_LINK_ID)) - return -1; - - return *common; -} - -/** - * ieee80211_mle_get_bss_param_ch_cnt - returns the BSS parameter change count - * @data: pointer to the basic multi link element - * Return: the BSS Parameter Change Count field value, or -1 if not present - * - * The element is assumed to be of the correct type (BASIC) and big enough, - * this must be checked using ieee80211_mle_type_ok(). - */ -static inline int -ieee80211_mle_get_bss_param_ch_cnt(const u8 *data) -{ - const struct ieee80211_multi_link_elem *mle = (const void *)data; - u16 control = le16_to_cpu(mle->control); - const u8 *common = mle->variable; - - /* common points now at the beginning of ieee80211_mle_basic_common_info */ - common += sizeof(struct ieee80211_mle_basic_common_info); - - if (!(control & IEEE80211_MLC_BASIC_PRES_BSS_PARAM_CH_CNT)) - return -1; - - if (control & IEEE80211_MLC_BASIC_PRES_LINK_ID) - common += 1; - - return *common; -} - -/** - * ieee80211_mle_get_eml_med_sync_delay - returns the medium sync delay - * @data: pointer to the multi-link element - * Return: the medium synchronization delay field value from the multi-link - * element, or the default value (%IEEE80211_MED_SYNC_DELAY_DEFAULT) - * if not present - * - * The element is assumed to be of the correct type (BASIC) and big enough, - * this must be checked using ieee80211_mle_type_ok(). - */ -static inline u16 ieee80211_mle_get_eml_med_sync_delay(const u8 *data) -{ - const struct ieee80211_multi_link_elem *mle = (const void *)data; - u16 control = le16_to_cpu(mle->control); - const u8 *common = mle->variable; - - /* common points now at the beginning of ieee80211_mle_basic_common_info */ - common += sizeof(struct ieee80211_mle_basic_common_info); - - if (!(control & IEEE80211_MLC_BASIC_PRES_MED_SYNC_DELAY)) - return IEEE80211_MED_SYNC_DELAY_DEFAULT; - - if (control & IEEE80211_MLC_BASIC_PRES_LINK_ID) - common += 1; - if (control & IEEE80211_MLC_BASIC_PRES_BSS_PARAM_CH_CNT) - common += 1; - - return get_unaligned_le16(common); -} - -/** - * ieee80211_mle_get_eml_cap - returns the EML capability - * @data: pointer to the multi-link element - * Return: the EML capability field value from the multi-link element, - * or 0 if not present - * - * The element is assumed to be of the correct type (BASIC) and big enough, - * this must be checked using ieee80211_mle_type_ok(). - */ -static inline u16 ieee80211_mle_get_eml_cap(const u8 *data) -{ - const struct ieee80211_multi_link_elem *mle = (const void *)data; - u16 control = le16_to_cpu(mle->control); - const u8 *common = mle->variable; - - /* common points now at the beginning of ieee80211_mle_basic_common_info */ - common += sizeof(struct ieee80211_mle_basic_common_info); - - if (!(control & IEEE80211_MLC_BASIC_PRES_EML_CAPA)) - return 0; - - if (control & IEEE80211_MLC_BASIC_PRES_LINK_ID) - common += 1; - if (control & IEEE80211_MLC_BASIC_PRES_BSS_PARAM_CH_CNT) - common += 1; - if (control & IEEE80211_MLC_BASIC_PRES_MED_SYNC_DELAY) - common += 2; - - return get_unaligned_le16(common); -} - -/** - * ieee80211_mle_get_mld_capa_op - returns the MLD capabilities and operations. - * @data: pointer to the multi-link element - * Return: the MLD capabilities and operations field value from the multi-link - * element, or 0 if not present - * - * The element is assumed to be of the correct type (BASIC) and big enough, - * this must be checked using ieee80211_mle_type_ok(). - */ -static inline u16 ieee80211_mle_get_mld_capa_op(const u8 *data) -{ - const struct ieee80211_multi_link_elem *mle = (const void *)data; - u16 control = le16_to_cpu(mle->control); - const u8 *common = mle->variable; - - /* - * common points now at the beginning of - * ieee80211_mle_basic_common_info - */ - common += sizeof(struct ieee80211_mle_basic_common_info); - - if (!(control & IEEE80211_MLC_BASIC_PRES_MLD_CAPA_OP)) - return 0; - - if (control & IEEE80211_MLC_BASIC_PRES_LINK_ID) - common += 1; - if (control & IEEE80211_MLC_BASIC_PRES_BSS_PARAM_CH_CNT) - common += 1; - if (control & IEEE80211_MLC_BASIC_PRES_MED_SYNC_DELAY) - common += 2; - if (control & IEEE80211_MLC_BASIC_PRES_EML_CAPA) - common += 2; - - return get_unaligned_le16(common); -} - -/* Defined in Figure 9-1074t in P802.11be_D7.0 */ -#define IEEE80211_EHT_ML_EXT_MLD_CAPA_OP_PARAM_UPDATE 0x0001 -#define IEEE80211_EHT_ML_EXT_MLD_CAPA_OP_RECO_MAX_LINKS_MASK 0x001e -#define IEEE80211_EHT_ML_EXT_MLD_CAPA_NSTR_UPDATE 0x0020 -#define IEEE80211_EHT_ML_EXT_MLD_CAPA_EMLSR_ENA_ON_ONE_LINK 0x0040 -#define IEEE80211_EHT_ML_EXT_MLD_CAPA_BTM_MLD_RECO_MULTI_AP 0x0080 - -/** - * ieee80211_mle_get_ext_mld_capa_op - returns the extended MLD capabilities - * and operations. - * @data: pointer to the multi-link element - * Return: the extended MLD capabilities and operations field value from - * the multi-link element, or 0 if not present - * - * The element is assumed to be of the correct type (BASIC) and big enough, - * this must be checked using ieee80211_mle_type_ok(). - */ -static inline u16 ieee80211_mle_get_ext_mld_capa_op(const u8 *data) -{ - const struct ieee80211_multi_link_elem *mle = (const void *)data; - u16 control = le16_to_cpu(mle->control); - const u8 *common = mle->variable; - - /* - * common points now at the beginning of - * ieee80211_mle_basic_common_info - */ - common += sizeof(struct ieee80211_mle_basic_common_info); - - if (!(control & IEEE80211_MLC_BASIC_PRES_EXT_MLD_CAPA_OP)) - return 0; - - if (control & IEEE80211_MLC_BASIC_PRES_LINK_ID) - common += 1; - if (control & IEEE80211_MLC_BASIC_PRES_BSS_PARAM_CH_CNT) - common += 1; - if (control & IEEE80211_MLC_BASIC_PRES_MED_SYNC_DELAY) - common += 2; - if (control & IEEE80211_MLC_BASIC_PRES_EML_CAPA) - common += 2; - if (control & IEEE80211_MLC_BASIC_PRES_MLD_CAPA_OP) - common += 2; - if (control & IEEE80211_MLC_BASIC_PRES_MLD_ID) - common += 1; - - return get_unaligned_le16(common); -} - -/** - * ieee80211_mle_get_mld_id - returns the MLD ID - * @data: pointer to the multi-link element - * Return: The MLD ID in the given multi-link element, or 0 if not present - * - * The element is assumed to be of the correct type (BASIC) and big enough, - * this must be checked using ieee80211_mle_type_ok(). - */ -static inline u8 ieee80211_mle_get_mld_id(const u8 *data) -{ - const struct ieee80211_multi_link_elem *mle = (const void *)data; - u16 control = le16_to_cpu(mle->control); - const u8 *common = mle->variable; - - /* - * common points now at the beginning of - * ieee80211_mle_basic_common_info - */ - common += sizeof(struct ieee80211_mle_basic_common_info); - - if (!(control & IEEE80211_MLC_BASIC_PRES_MLD_ID)) - return 0; - - if (control & IEEE80211_MLC_BASIC_PRES_LINK_ID) - common += 1; - if (control & IEEE80211_MLC_BASIC_PRES_BSS_PARAM_CH_CNT) - common += 1; - if (control & IEEE80211_MLC_BASIC_PRES_MED_SYNC_DELAY) - common += 2; - if (control & IEEE80211_MLC_BASIC_PRES_EML_CAPA) - common += 2; - if (control & IEEE80211_MLC_BASIC_PRES_MLD_CAPA_OP) - common += 2; - - return *common; -} - -/** - * ieee80211_mle_size_ok - validate multi-link element size - * @data: pointer to the element data - * @len: length of the containing element - * Return: whether or not the multi-link element size is OK - */ -static inline bool ieee80211_mle_size_ok(const u8 *data, size_t len) -{ - const struct ieee80211_multi_link_elem *mle = (const void *)data; - u8 fixed = sizeof(*mle); - u8 common = 0; - bool check_common_len = false; - u16 control; - - if (!data || len < fixed) - return false; - - control = le16_to_cpu(mle->control); - - switch (u16_get_bits(control, IEEE80211_ML_CONTROL_TYPE)) { - case IEEE80211_ML_CONTROL_TYPE_BASIC: - common += sizeof(struct ieee80211_mle_basic_common_info); - check_common_len = true; - if (control & IEEE80211_MLC_BASIC_PRES_LINK_ID) - common += 1; - if (control & IEEE80211_MLC_BASIC_PRES_BSS_PARAM_CH_CNT) - common += 1; - if (control & IEEE80211_MLC_BASIC_PRES_MED_SYNC_DELAY) - common += 2; - if (control & IEEE80211_MLC_BASIC_PRES_EML_CAPA) - common += 2; - if (control & IEEE80211_MLC_BASIC_PRES_MLD_CAPA_OP) - common += 2; - if (control & IEEE80211_MLC_BASIC_PRES_MLD_ID) - common += 1; - if (control & IEEE80211_MLC_BASIC_PRES_EXT_MLD_CAPA_OP) - common += 2; - break; - case IEEE80211_ML_CONTROL_TYPE_PREQ: - common += sizeof(struct ieee80211_mle_preq_common_info); - if (control & IEEE80211_MLC_PREQ_PRES_MLD_ID) - common += 1; - check_common_len = true; - break; - case IEEE80211_ML_CONTROL_TYPE_RECONF: - if (control & IEEE80211_MLC_RECONF_PRES_MLD_MAC_ADDR) - common += ETH_ALEN; - if (control & IEEE80211_MLC_RECONF_PRES_EML_CAPA) - common += 2; - if (control & IEEE80211_MLC_RECONF_PRES_MLD_CAPA_OP) - common += 2; - if (control & IEEE80211_MLC_RECONF_PRES_EXT_MLD_CAPA_OP) - common += 2; - break; - case IEEE80211_ML_CONTROL_TYPE_TDLS: - common += sizeof(struct ieee80211_mle_tdls_common_info); - check_common_len = true; - break; - case IEEE80211_ML_CONTROL_TYPE_PRIO_ACCESS: - common = ETH_ALEN + 1; - break; - default: - /* we don't know this type */ - return true; - } - - if (len < fixed + common) - return false; - - if (!check_common_len) - return true; - - /* if present, common length is the first octet there */ - return mle->variable[0] >= common; -} - -/** - * ieee80211_mle_type_ok - validate multi-link element type and size - * @data: pointer to the element data - * @type: expected type of the element - * @len: length of the containing element - * Return: whether or not the multi-link element type matches and size is OK - */ -static inline bool ieee80211_mle_type_ok(const u8 *data, u8 type, size_t len) -{ - const struct ieee80211_multi_link_elem *mle = (const void *)data; - u16 control; - - if (!ieee80211_mle_size_ok(data, len)) - return false; - - control = le16_to_cpu(mle->control); - - if (u16_get_bits(control, IEEE80211_ML_CONTROL_TYPE) == type) - return true; - - return false; -} - -enum ieee80211_mle_subelems { - IEEE80211_MLE_SUBELEM_PER_STA_PROFILE = 0, - IEEE80211_MLE_SUBELEM_FRAGMENT = 254, -}; - -#define IEEE80211_MLE_STA_CONTROL_LINK_ID 0x000f -#define IEEE80211_MLE_STA_CONTROL_COMPLETE_PROFILE 0x0010 -#define IEEE80211_MLE_STA_CONTROL_STA_MAC_ADDR_PRESENT 0x0020 -#define IEEE80211_MLE_STA_CONTROL_BEACON_INT_PRESENT 0x0040 -#define IEEE80211_MLE_STA_CONTROL_TSF_OFFS_PRESENT 0x0080 -#define IEEE80211_MLE_STA_CONTROL_DTIM_INFO_PRESENT 0x0100 -#define IEEE80211_MLE_STA_CONTROL_NSTR_LINK_PAIR_PRESENT 0x0200 -#define IEEE80211_MLE_STA_CONTROL_NSTR_BITMAP_SIZE 0x0400 -#define IEEE80211_MLE_STA_CONTROL_BSS_PARAM_CHANGE_CNT_PRESENT 0x0800 - -struct ieee80211_mle_per_sta_profile { - __le16 control; - u8 sta_info_len; - u8 variable[]; -} __packed; - -/** - * ieee80211_mle_basic_sta_prof_size_ok - validate basic multi-link element sta - * profile size - * @data: pointer to the sub element data - * @len: length of the containing sub element - * Return: %true if the STA profile is large enough, %false otherwise - */ -static inline bool ieee80211_mle_basic_sta_prof_size_ok(const u8 *data, - size_t len) -{ - const struct ieee80211_mle_per_sta_profile *prof = (const void *)data; - u16 control; - u8 fixed = sizeof(*prof); - u8 info_len = 1; - - if (len < fixed) - return false; - - control = le16_to_cpu(prof->control); - - if (control & IEEE80211_MLE_STA_CONTROL_STA_MAC_ADDR_PRESENT) - info_len += 6; - if (control & IEEE80211_MLE_STA_CONTROL_BEACON_INT_PRESENT) - info_len += 2; - if (control & IEEE80211_MLE_STA_CONTROL_TSF_OFFS_PRESENT) - info_len += 8; - if (control & IEEE80211_MLE_STA_CONTROL_DTIM_INFO_PRESENT) - info_len += 2; - if (control & IEEE80211_MLE_STA_CONTROL_COMPLETE_PROFILE && - control & IEEE80211_MLE_STA_CONTROL_NSTR_LINK_PAIR_PRESENT) { - if (control & IEEE80211_MLE_STA_CONTROL_NSTR_BITMAP_SIZE) - info_len += 2; - else - info_len += 1; - } - if (control & IEEE80211_MLE_STA_CONTROL_BSS_PARAM_CHANGE_CNT_PRESENT) - info_len += 1; - - return prof->sta_info_len >= info_len && - fixed + prof->sta_info_len - 1 <= len; -} - -/** - * ieee80211_mle_basic_sta_prof_bss_param_ch_cnt - get per-STA profile BSS - * parameter change count - * @prof: the per-STA profile, having been checked with - * ieee80211_mle_basic_sta_prof_size_ok() for the correct length - * - * Return: The BSS parameter change count value if present, 0 otherwise. - */ -static inline u8 -ieee80211_mle_basic_sta_prof_bss_param_ch_cnt(const struct ieee80211_mle_per_sta_profile *prof) -{ - u16 control = le16_to_cpu(prof->control); - const u8 *pos = prof->variable; - - if (!(control & IEEE80211_MLE_STA_CONTROL_BSS_PARAM_CHANGE_CNT_PRESENT)) - return 0; - - if (control & IEEE80211_MLE_STA_CONTROL_STA_MAC_ADDR_PRESENT) - pos += 6; - if (control & IEEE80211_MLE_STA_CONTROL_BEACON_INT_PRESENT) - pos += 2; - if (control & IEEE80211_MLE_STA_CONTROL_TSF_OFFS_PRESENT) - pos += 8; - if (control & IEEE80211_MLE_STA_CONTROL_DTIM_INFO_PRESENT) - pos += 2; - if (control & IEEE80211_MLE_STA_CONTROL_COMPLETE_PROFILE && - control & IEEE80211_MLE_STA_CONTROL_NSTR_LINK_PAIR_PRESENT) { - if (control & IEEE80211_MLE_STA_CONTROL_NSTR_BITMAP_SIZE) - pos += 2; - else - pos += 1; - } - - return *pos; -} - -#define IEEE80211_MLE_STA_RECONF_CONTROL_LINK_ID 0x000f -#define IEEE80211_MLE_STA_RECONF_CONTROL_COMPLETE_PROFILE 0x0010 -#define IEEE80211_MLE_STA_RECONF_CONTROL_STA_MAC_ADDR_PRESENT 0x0020 -#define IEEE80211_MLE_STA_RECONF_CONTROL_AP_REM_TIMER_PRESENT 0x0040 -#define IEEE80211_MLE_STA_RECONF_CONTROL_OPERATION_TYPE 0x0780 -#define IEEE80211_MLE_STA_RECONF_CONTROL_OPERATION_TYPE_AP_REM 0 -#define IEEE80211_MLE_STA_RECONF_CONTROL_OPERATION_TYPE_OP_PARAM_UPDATE 1 -#define IEEE80211_MLE_STA_RECONF_CONTROL_OPERATION_TYPE_ADD_LINK 2 -#define IEEE80211_MLE_STA_RECONF_CONTROL_OPERATION_TYPE_DEL_LINK 3 -#define IEEE80211_MLE_STA_RECONF_CONTROL_OPERATION_TYPE_NSTR_STATUS 4 -#define IEEE80211_MLE_STA_RECONF_CONTROL_OPERATION_PARAMS_PRESENT 0x0800 - -/** - * ieee80211_mle_reconf_sta_prof_size_ok - validate reconfiguration multi-link - * element sta profile size. - * @data: pointer to the sub element data - * @len: length of the containing sub element - * Return: %true if the STA profile is large enough, %false otherwise - */ -static inline bool ieee80211_mle_reconf_sta_prof_size_ok(const u8 *data, - size_t len) -{ - const struct ieee80211_mle_per_sta_profile *prof = (const void *)data; - u16 control; - u8 fixed = sizeof(*prof); - u8 info_len = 1; - - if (len < fixed) - return false; - - control = le16_to_cpu(prof->control); - - if (control & IEEE80211_MLE_STA_RECONF_CONTROL_STA_MAC_ADDR_PRESENT) - info_len += ETH_ALEN; - if (control & IEEE80211_MLE_STA_RECONF_CONTROL_AP_REM_TIMER_PRESENT) - info_len += 2; - if (control & IEEE80211_MLE_STA_RECONF_CONTROL_OPERATION_PARAMS_PRESENT) - info_len += 2; - - return prof->sta_info_len >= info_len && - fixed + prof->sta_info_len - 1 <= len; -} - -#define IEEE80211_MLE_STA_EPCS_CONTROL_LINK_ID 0x000f -#define IEEE80211_EPCS_ENA_RESP_BODY_LEN 3 - -static inline bool ieee80211_tid_to_link_map_size_ok(const u8 *data, size_t len) -{ - const struct ieee80211_ttlm_elem *t2l = (const void *)data; - u8 control, fixed = sizeof(*t2l), elem_len = 0; - - if (len < fixed) - return false; - - control = t2l->control; - - if (control & IEEE80211_TTLM_CONTROL_SWITCH_TIME_PRESENT) - elem_len += 2; - if (control & IEEE80211_TTLM_CONTROL_EXPECTED_DUR_PRESENT) - elem_len += 3; - - if (!(control & IEEE80211_TTLM_CONTROL_DEF_LINK_MAP)) { - u8 bm_size; - - elem_len += 1; - if (len < fixed + elem_len) - return false; - - if (control & IEEE80211_TTLM_CONTROL_LINK_MAP_SIZE) - bm_size = 1; - else - bm_size = 2; - - elem_len += hweight8(t2l->optional[0]) * bm_size; - } - - return len >= fixed + elem_len; -} - -/** - * ieee80211_emlsr_pad_delay_in_us - Fetch the EMLSR Padding delay - * in microseconds - * @eml_cap: EML capabilities field value from common info field of - * the Multi-link element - * Return: the EMLSR Padding delay (in microseconds) encoded in the - * EML Capabilities field - */ - -static inline u32 ieee80211_emlsr_pad_delay_in_us(u16 eml_cap) -{ - /* IEEE Std 802.11be-2024 Table 9-417i—Encoding of the EMLSR - * Padding Delay subfield. - */ - u32 pad_delay = u16_get_bits(eml_cap, - IEEE80211_EML_CAP_EMLSR_PADDING_DELAY); - - if (!pad_delay || - pad_delay > IEEE80211_EML_CAP_EMLSR_PADDING_DELAY_256US) - return 0; - - return 32 * (1 << (pad_delay - 1)); -} - -/** - * ieee80211_emlsr_trans_delay_in_us - Fetch the EMLSR Transition - * delay in microseconds - * @eml_cap: EML capabilities field value from common info field of - * the Multi-link element - * Return: the EMLSR Transition delay (in microseconds) encoded in the - * EML Capabilities field - */ - -static inline u32 ieee80211_emlsr_trans_delay_in_us(u16 eml_cap) -{ - /* IEEE Std 802.11be-2024 Table 9-417j—Encoding of the EMLSR - * Transition Delay subfield. - */ - u32 trans_delay = - u16_get_bits(eml_cap, - IEEE80211_EML_CAP_EMLSR_TRANSITION_DELAY); - - /* invalid values also just use 0 */ - if (!trans_delay || - trans_delay > IEEE80211_EML_CAP_EMLSR_TRANSITION_DELAY_256US) - return 0; - - return 16 * (1 << (trans_delay - 1)); -} - -/** - * ieee80211_eml_trans_timeout_in_us - Fetch the EMLSR Transition - * timeout value in microseconds - * @eml_cap: EML capabilities field value from common info field of - * the Multi-link element - * Return: the EMLSR Transition timeout (in microseconds) encoded in - * the EML Capabilities field - */ - -static inline u32 ieee80211_eml_trans_timeout_in_us(u16 eml_cap) -{ - /* IEEE Std 802.11be-2024 Table 9-417m—Encoding of the - * Transition Timeout subfield. - */ - u8 timeout = u16_get_bits(eml_cap, - IEEE80211_EML_CAP_TRANSITION_TIMEOUT); - - /* invalid values also just use 0 */ - if (!timeout || timeout > IEEE80211_EML_CAP_TRANSITION_TIMEOUT_128TU) - return 0; - - return 128 * (1 << (timeout - 1)); -} - -#define for_each_mle_subelement(_elem, _data, _len) \ - if (ieee80211_mle_size_ok(_data, _len)) \ - for_each_element(_elem, \ - _data + ieee80211_mle_common_size(_data),\ - _len - ieee80211_mle_common_size(_data)) - /* NAN operation mode, as defined in Wi-Fi Aware (TM) specification Table 81 */ #define NAN_OP_MODE_PHY_MODE_VHT 0x01 #define NAN_OP_MODE_PHY_MODE_HE 0x10 @@ -4605,6 +3443,8 @@ static inline u32 ieee80211_eml_trans_timeout_in_us(u16 eml_cap) #include "ieee80211-ht.h" #include "ieee80211-vht.h" +#include "ieee80211-he.h" +#include "ieee80211-eht.h" #include "ieee80211-mesh.h" #endif /* LINUX_IEEE80211_H */ From 00105d7600bfb171037783da5f26e2565c7d2106 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 5 Nov 2025 15:36:54 +0100 Subject: [PATCH 0638/1381] wifi: ieee80211: split S1G definitions out The ieee80211.h file has gotten very long, continue splitting it by putting S1G definitions into a separate file. Link: https://patch.msgid.link/20251105153843.82c0bddee6e3.Ic6646615286dad240b42e31e9d428c7e4ea40ce0@changeid Signed-off-by: Johannes Berg --- include/linux/ieee80211-s1g.h | 575 +++++++++++++++++++++++++++++++++ include/linux/ieee80211.h | 585 +--------------------------------- 2 files changed, 591 insertions(+), 569 deletions(-) create mode 100644 include/linux/ieee80211-s1g.h diff --git a/include/linux/ieee80211-s1g.h b/include/linux/ieee80211-s1g.h new file mode 100644 index 000000000000..5b9ed2dcc00e --- /dev/null +++ b/include/linux/ieee80211-s1g.h @@ -0,0 +1,575 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * IEEE 802.11 S1G definitions + * + * Copyright (c) 2001-2002, SSH Communications Security Corp and Jouni Malinen + * + * Copyright (c) 2002-2003, Jouni Malinen + * Copyright (c) 2005, Devicescape Software, Inc. + * Copyright (c) 2006, Michael Wu + * Copyright (c) 2013 - 2014 Intel Mobile Communications GmbH + * Copyright (c) 2016 - 2017 Intel Deutschland GmbH + * Copyright (c) 2018 - 2025 Intel Corporation + */ + +#ifndef LINUX_IEEE80211_S1G_H +#define LINUX_IEEE80211_S1G_H + +#include +#include + +/* bits unique to S1G beacon frame control */ +#define IEEE80211_S1G_BCN_NEXT_TBTT 0x100 +#define IEEE80211_S1G_BCN_CSSID 0x200 +#define IEEE80211_S1G_BCN_ANO 0x400 + +/* see 802.11ah-2016 9.9 NDP CMAC frames */ +#define IEEE80211_S1G_1MHZ_NDP_BITS 25 +#define IEEE80211_S1G_1MHZ_NDP_BYTES 4 +#define IEEE80211_S1G_2MHZ_NDP_BITS 37 +#define IEEE80211_S1G_2MHZ_NDP_BYTES 5 + +/** + * ieee80211_is_s1g_beacon - check if IEEE80211_FTYPE_EXT && + * IEEE80211_STYPE_S1G_BEACON + * @fc: frame control bytes in little-endian byteorder + * Return: whether or not the frame is an S1G beacon + */ +static inline bool ieee80211_is_s1g_beacon(__le16 fc) +{ + return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | + IEEE80211_FCTL_STYPE)) == + cpu_to_le16(IEEE80211_FTYPE_EXT | IEEE80211_STYPE_S1G_BEACON); +} + +/** + * ieee80211_s1g_has_next_tbtt - check if IEEE80211_S1G_BCN_NEXT_TBTT + * @fc: frame control bytes in little-endian byteorder + * Return: whether or not the frame contains the variable-length + * next TBTT field + */ +static inline bool ieee80211_s1g_has_next_tbtt(__le16 fc) +{ + return ieee80211_is_s1g_beacon(fc) && + (fc & cpu_to_le16(IEEE80211_S1G_BCN_NEXT_TBTT)); +} + +/** + * ieee80211_s1g_has_ano - check if IEEE80211_S1G_BCN_ANO + * @fc: frame control bytes in little-endian byteorder + * Return: whether or not the frame contains the variable-length + * ANO field + */ +static inline bool ieee80211_s1g_has_ano(__le16 fc) +{ + return ieee80211_is_s1g_beacon(fc) && + (fc & cpu_to_le16(IEEE80211_S1G_BCN_ANO)); +} + +/** + * ieee80211_s1g_has_cssid - check if IEEE80211_S1G_BCN_CSSID + * @fc: frame control bytes in little-endian byteorder + * Return: whether or not the frame contains the variable-length + * compressed SSID field + */ +static inline bool ieee80211_s1g_has_cssid(__le16 fc) +{ + return ieee80211_is_s1g_beacon(fc) && + (fc & cpu_to_le16(IEEE80211_S1G_BCN_CSSID)); +} + +/** + * enum ieee80211_s1g_chanwidth - S1G channel widths + * These are defined in IEEE802.11-2016ah Table 10-20 + * as BSS Channel Width + * + * @IEEE80211_S1G_CHANWIDTH_1MHZ: 1MHz operating channel + * @IEEE80211_S1G_CHANWIDTH_2MHZ: 2MHz operating channel + * @IEEE80211_S1G_CHANWIDTH_4MHZ: 4MHz operating channel + * @IEEE80211_S1G_CHANWIDTH_8MHZ: 8MHz operating channel + * @IEEE80211_S1G_CHANWIDTH_16MHZ: 16MHz operating channel + */ +enum ieee80211_s1g_chanwidth { + IEEE80211_S1G_CHANWIDTH_1MHZ = 0, + IEEE80211_S1G_CHANWIDTH_2MHZ = 1, + IEEE80211_S1G_CHANWIDTH_4MHZ = 3, + IEEE80211_S1G_CHANWIDTH_8MHZ = 7, + IEEE80211_S1G_CHANWIDTH_16MHZ = 15, +}; + +/** + * enum ieee80211_s1g_pri_chanwidth - S1G primary channel widths + * described in IEEE80211-2024 Table 10-39. + * + * @IEEE80211_S1G_PRI_CHANWIDTH_2MHZ: 2MHz primary channel + * @IEEE80211_S1G_PRI_CHANWIDTH_1MHZ: 1MHz primary channel + */ +enum ieee80211_s1g_pri_chanwidth { + IEEE80211_S1G_PRI_CHANWIDTH_2MHZ = 0, + IEEE80211_S1G_PRI_CHANWIDTH_1MHZ = 1, +}; + +/** + * struct ieee80211_s1g_bcn_compat_ie - S1G Beacon Compatibility element + * @compat_info: Compatibility Information + * @beacon_int: Beacon Interval + * @tsf_completion: TSF Completion + * + * This structure represents the payload of the "S1G Beacon + * Compatibility element" as described in IEEE Std 802.11-2020 section + * 9.4.2.196. + */ +struct ieee80211_s1g_bcn_compat_ie { + __le16 compat_info; + __le16 beacon_int; + __le32 tsf_completion; +} __packed; + +/** + * struct ieee80211_s1g_oper_ie - S1G Operation element + * @ch_width: S1G Operation Information Channel Width + * @oper_class: S1G Operation Information Operating Class + * @primary_ch: S1G Operation Information Primary Channel Number + * @oper_ch: S1G Operation Information Channel Center Frequency + * @basic_mcs_nss: Basic S1G-MCS and NSS Set + * + * This structure represents the payload of the "S1G Operation + * element" as described in IEEE Std 802.11-2020 section 9.4.2.212. + */ +struct ieee80211_s1g_oper_ie { + u8 ch_width; + u8 oper_class; + u8 primary_ch; + u8 oper_ch; + __le16 basic_mcs_nss; +} __packed; + +/** + * struct ieee80211_aid_response_ie - AID Response element + * @aid: AID/Group AID + * @switch_count: AID Switch Count + * @response_int: AID Response Interval + * + * This structure represents the payload of the "AID Response element" + * as described in IEEE Std 802.11-2020 section 9.4.2.194. + */ +struct ieee80211_aid_response_ie { + __le16 aid; + u8 switch_count; + __le16 response_int; +} __packed; + +struct ieee80211_s1g_cap { + u8 capab_info[10]; + u8 supp_mcs_nss[5]; +} __packed; + +/** + * ieee80211_s1g_optional_len - determine length of optional S1G beacon fields + * @fc: frame control bytes in little-endian byteorder + * Return: total length in bytes of the optional fixed-length fields + * + * S1G beacons may contain up to three optional fixed-length fields that + * precede the variable-length elements. Whether these fields are present + * is indicated by flags in the frame control field. + * + * From IEEE 802.11-2024 section 9.3.4.3: + * - Next TBTT field may be 0 or 3 bytes + * - Short SSID field may be 0 or 4 bytes + * - Access Network Options (ANO) field may be 0 or 1 byte + */ +static inline size_t +ieee80211_s1g_optional_len(__le16 fc) +{ + size_t len = 0; + + if (ieee80211_s1g_has_next_tbtt(fc)) + len += 3; + + if (ieee80211_s1g_has_cssid(fc)) + len += 4; + + if (ieee80211_s1g_has_ano(fc)) + len += 1; + + return len; +} + +/* S1G Capabilities Information field */ +#define IEEE80211_S1G_CAPABILITY_LEN 15 + +#define S1G_CAP0_S1G_LONG BIT(0) +#define S1G_CAP0_SGI_1MHZ BIT(1) +#define S1G_CAP0_SGI_2MHZ BIT(2) +#define S1G_CAP0_SGI_4MHZ BIT(3) +#define S1G_CAP0_SGI_8MHZ BIT(4) +#define S1G_CAP0_SGI_16MHZ BIT(5) +#define S1G_CAP0_SUPP_CH_WIDTH GENMASK(7, 6) + +#define S1G_SUPP_CH_WIDTH_2 0 +#define S1G_SUPP_CH_WIDTH_4 1 +#define S1G_SUPP_CH_WIDTH_8 2 +#define S1G_SUPP_CH_WIDTH_16 3 +#define S1G_SUPP_CH_WIDTH_MAX(cap) ((1 << FIELD_GET(S1G_CAP0_SUPP_CH_WIDTH, \ + cap[0])) << 1) + +#define S1G_CAP1_RX_LDPC BIT(0) +#define S1G_CAP1_TX_STBC BIT(1) +#define S1G_CAP1_RX_STBC BIT(2) +#define S1G_CAP1_SU_BFER BIT(3) +#define S1G_CAP1_SU_BFEE BIT(4) +#define S1G_CAP1_BFEE_STS GENMASK(7, 5) + +#define S1G_CAP2_SOUNDING_DIMENSIONS GENMASK(2, 0) +#define S1G_CAP2_MU_BFER BIT(3) +#define S1G_CAP2_MU_BFEE BIT(4) +#define S1G_CAP2_PLUS_HTC_VHT BIT(5) +#define S1G_CAP2_TRAVELING_PILOT GENMASK(7, 6) + +#define S1G_CAP3_RD_RESPONDER BIT(0) +#define S1G_CAP3_HT_DELAYED_BA BIT(1) +#define S1G_CAP3_MAX_MPDU_LEN BIT(2) +#define S1G_CAP3_MAX_AMPDU_LEN_EXP GENMASK(4, 3) +#define S1G_CAP3_MIN_MPDU_START GENMASK(7, 5) + +#define S1G_CAP4_UPLINK_SYNC BIT(0) +#define S1G_CAP4_DYNAMIC_AID BIT(1) +#define S1G_CAP4_BAT BIT(2) +#define S1G_CAP4_TIME_ADE BIT(3) +#define S1G_CAP4_NON_TIM BIT(4) +#define S1G_CAP4_GROUP_AID BIT(5) +#define S1G_CAP4_STA_TYPE GENMASK(7, 6) + +#define S1G_CAP5_CENT_AUTH_CONTROL BIT(0) +#define S1G_CAP5_DIST_AUTH_CONTROL BIT(1) +#define S1G_CAP5_AMSDU BIT(2) +#define S1G_CAP5_AMPDU BIT(3) +#define S1G_CAP5_ASYMMETRIC_BA BIT(4) +#define S1G_CAP5_FLOW_CONTROL BIT(5) +#define S1G_CAP5_SECTORIZED_BEAM GENMASK(7, 6) + +#define S1G_CAP6_OBSS_MITIGATION BIT(0) +#define S1G_CAP6_FRAGMENT_BA BIT(1) +#define S1G_CAP6_NDP_PS_POLL BIT(2) +#define S1G_CAP6_RAW_OPERATION BIT(3) +#define S1G_CAP6_PAGE_SLICING BIT(4) +#define S1G_CAP6_TXOP_SHARING_IMP_ACK BIT(5) +#define S1G_CAP6_VHT_LINK_ADAPT GENMASK(7, 6) + +#define S1G_CAP7_TACK_AS_PS_POLL BIT(0) +#define S1G_CAP7_DUP_1MHZ BIT(1) +#define S1G_CAP7_MCS_NEGOTIATION BIT(2) +#define S1G_CAP7_1MHZ_CTL_RESPONSE_PREAMBLE BIT(3) +#define S1G_CAP7_NDP_BFING_REPORT_POLL BIT(4) +#define S1G_CAP7_UNSOLICITED_DYN_AID BIT(5) +#define S1G_CAP7_SECTOR_TRAINING_OPERATION BIT(6) +#define S1G_CAP7_TEMP_PS_MODE_SWITCH BIT(7) + +#define S1G_CAP8_TWT_GROUPING BIT(0) +#define S1G_CAP8_BDT BIT(1) +#define S1G_CAP8_COLOR GENMASK(4, 2) +#define S1G_CAP8_TWT_REQUEST BIT(5) +#define S1G_CAP8_TWT_RESPOND BIT(6) +#define S1G_CAP8_PV1_FRAME BIT(7) + +#define S1G_CAP9_LINK_ADAPT_PER_CONTROL_RESPONSE BIT(0) + +#define S1G_OPER_CH_WIDTH_PRIMARY BIT(0) +#define S1G_OPER_CH_WIDTH_OPER GENMASK(4, 1) +#define S1G_OPER_CH_PRIMARY_LOCATION BIT(5) + +#define S1G_2M_PRIMARY_LOCATION_LOWER 0 +#define S1G_2M_PRIMARY_LOCATION_UPPER 1 + +#define LISTEN_INT_USF GENMASK(15, 14) +#define LISTEN_INT_UI GENMASK(13, 0) + +#define IEEE80211_MAX_USF FIELD_MAX(LISTEN_INT_USF) +#define IEEE80211_MAX_UI FIELD_MAX(LISTEN_INT_UI) + +/* S1G encoding types */ +#define IEEE80211_S1G_TIM_ENC_MODE_BLOCK 0 +#define IEEE80211_S1G_TIM_ENC_MODE_SINGLE 1 +#define IEEE80211_S1G_TIM_ENC_MODE_OLB 2 + +enum ieee80211_s1g_actioncode { + WLAN_S1G_AID_SWITCH_REQUEST, + WLAN_S1G_AID_SWITCH_RESPONSE, + WLAN_S1G_SYNC_CONTROL, + WLAN_S1G_STA_INFO_ANNOUNCE, + WLAN_S1G_EDCA_PARAM_SET, + WLAN_S1G_EL_OPERATION, + WLAN_S1G_TWT_SETUP, + WLAN_S1G_TWT_TEARDOWN, + WLAN_S1G_SECT_GROUP_ID_LIST, + WLAN_S1G_SECT_ID_FEEDBACK, + WLAN_S1G_TWT_INFORMATION = 11, +}; + +/** + * ieee80211_is_s1g_short_beacon - check if frame is an S1G short beacon + * @fc: frame control bytes in little-endian byteorder + * @variable: pointer to the beacon frame elements + * @variable_len: length of the frame elements + * Return: whether or not the frame is an S1G short beacon. As per + * IEEE80211-2024 11.1.3.10.1, The S1G beacon compatibility element shall + * always be present as the first element in beacon frames generated at a + * TBTT (Target Beacon Transmission Time), so any frame not containing + * this element must have been generated at a TSBTT (Target Short Beacon + * Transmission Time) that is not a TBTT. Additionally, short beacons are + * prohibited from containing the S1G beacon compatibility element as per + * IEEE80211-2024 9.3.4.3 Table 9-76, so if we have an S1G beacon with + * either no elements or the first element is not the beacon compatibility + * element, we have a short beacon. + */ +static inline bool ieee80211_is_s1g_short_beacon(__le16 fc, const u8 *variable, + size_t variable_len) +{ + if (!ieee80211_is_s1g_beacon(fc)) + return false; + + /* + * If the frame does not contain at least 1 element (this is perfectly + * valid in a short beacon) and is an S1G beacon, we have a short + * beacon. + */ + if (variable_len < 2) + return true; + + return variable[0] != WLAN_EID_S1G_BCN_COMPAT; +} + +struct s1g_tim_aid { + u16 aid; + u8 target_blk; /* Target block index */ + u8 target_subblk; /* Target subblock index */ + u8 target_subblk_bit; /* Target subblock bit */ +}; + +struct s1g_tim_enc_block { + u8 enc_mode; + bool inverse; + const u8 *ptr; + u8 len; + + /* + * For an OLB encoded block that spans multiple blocks, this + * is the offset into the span described by that encoded block. + */ + u8 olb_blk_offset; +}; + +/* + * Helper routines to quickly extract the length of an encoded block. Validation + * is also performed to ensure the length extracted lies within the TIM. + */ + +static inline int ieee80211_s1g_len_bitmap(const u8 *ptr, const u8 *end) +{ + u8 blkmap; + u8 n_subblks; + + if (ptr >= end) + return -EINVAL; + + blkmap = *ptr; + n_subblks = hweight8(blkmap); + + if (ptr + 1 + n_subblks > end) + return -EINVAL; + + return 1 + n_subblks; +} + +static inline int ieee80211_s1g_len_single(const u8 *ptr, const u8 *end) +{ + return (ptr + 1 > end) ? -EINVAL : 1; +} + +static inline int ieee80211_s1g_len_olb(const u8 *ptr, const u8 *end) +{ + if (ptr >= end) + return -EINVAL; + + return (ptr + 1 + *ptr > end) ? -EINVAL : 1 + *ptr; +} + +/* + * Enumerate all encoded blocks until we find the encoded block that describes + * our target AID. OLB is a special case as a single encoded block can describe + * multiple blocks as a single encoded block. + */ +static inline int ieee80211_s1g_find_target_block(struct s1g_tim_enc_block *enc, + const struct s1g_tim_aid *aid, + const u8 *ptr, const u8 *end) +{ + /* need at least block-control octet */ + while (ptr + 1 <= end) { + u8 ctrl = *ptr++; + u8 mode = ctrl & 0x03; + bool contains, inverse = ctrl & BIT(2); + u8 span, blk_off = ctrl >> 3; + int len; + + switch (mode) { + case IEEE80211_S1G_TIM_ENC_MODE_BLOCK: + len = ieee80211_s1g_len_bitmap(ptr, end); + contains = blk_off == aid->target_blk; + break; + case IEEE80211_S1G_TIM_ENC_MODE_SINGLE: + len = ieee80211_s1g_len_single(ptr, end); + contains = blk_off == aid->target_blk; + break; + case IEEE80211_S1G_TIM_ENC_MODE_OLB: + len = ieee80211_s1g_len_olb(ptr, end); + /* + * An OLB encoded block can describe more then one + * block, meaning an encoded OLB block can span more + * then a single block. + */ + if (len > 0) { + /* Minus one for the length octet */ + span = DIV_ROUND_UP(len - 1, 8); + /* + * Check if our target block lies within the + * block span described by this encoded block. + */ + contains = (aid->target_blk >= blk_off) && + (aid->target_blk < blk_off + span); + } + break; + default: + return -EOPNOTSUPP; + } + + if (len < 0) + return len; + + if (contains) { + enc->enc_mode = mode; + enc->inverse = inverse; + enc->ptr = ptr; + enc->len = (u8)len; + enc->olb_blk_offset = blk_off; + return 0; + } + + ptr += len; + } + + return -ENOENT; +} + +static inline bool ieee80211_s1g_parse_bitmap(struct s1g_tim_enc_block *enc, + struct s1g_tim_aid *aid) +{ + const u8 *ptr = enc->ptr; + u8 blkmap = *ptr++; + + /* + * If our block bitmap does not contain a set bit that corresponds + * to our AID, it could mean a variety of things depending on if + * the encoding mode is inverted or not. + * + * 1. If inverted, it means the entire subblock is present and hence + * our AID has been set. + * 2. If not inverted, it means our subblock is not present and hence + * it is all zero meaning our AID is not set. + */ + if (!(blkmap & BIT(aid->target_subblk))) + return enc->inverse; + + /* + * Increment ptr by the number of set subblocks that appear before our + * target subblock. If our target subblock is 0, do nothing as ptr + * already points to our target subblock. + */ + if (aid->target_subblk) + ptr += hweight8(blkmap & GENMASK(aid->target_subblk - 1, 0)); + + return !!(*ptr & BIT(aid->target_subblk_bit)) ^ enc->inverse; +} + +static inline bool ieee80211_s1g_parse_single(struct s1g_tim_enc_block *enc, + struct s1g_tim_aid *aid) +{ + /* + * Single AID mode describes, as the name suggests, a single AID + * within the block described by the encoded block. The octet + * contains the 6 LSBs of the AID described in the block. The other + * 2 bits are reserved. When inversed, every single AID described + * by the current block have buffered traffic except for the AID + * described in the single AID octet. + */ + return ((*enc->ptr & 0x3f) == (aid->aid & 0x3f)) ^ enc->inverse; +} + +static inline bool ieee80211_s1g_parse_olb(struct s1g_tim_enc_block *enc, + struct s1g_tim_aid *aid) +{ + const u8 *ptr = enc->ptr; + u8 blk_len = *ptr++; + /* + * Given an OLB encoded block that describes multiple blocks, + * calculate the offset into the span. Then calculate the + * subblock location normally. + */ + u16 span_offset = aid->target_blk - enc->olb_blk_offset; + u16 subblk_idx = span_offset * 8 + aid->target_subblk; + + if (subblk_idx >= blk_len) + return enc->inverse; + + return !!(ptr[subblk_idx] & BIT(aid->target_subblk_bit)) ^ enc->inverse; +} + +/* + * An S1G PVB has 3 non optional encoding types, each that can be inverted. + * An S1G PVB is constructed with zero or more encoded block subfields. Each + * encoded block represents a single "block" of AIDs (64), and each encoded + * block can contain one of the 3 encoding types alongside a single bit for + * whether the bits should be inverted. + * + * As the standard makes no guarantee about the ordering of encoded blocks, + * we must parse every encoded block in the worst case scenario given an + * AID that lies within the last block. + */ +static inline bool ieee80211_s1g_check_tim(const struct ieee80211_tim_ie *tim, + u8 tim_len, u16 aid) +{ + int err; + struct s1g_tim_aid target_aid; + struct s1g_tim_enc_block enc_blk; + + if (tim_len < 3) + return false; + + target_aid.aid = aid; + target_aid.target_blk = (aid >> 6) & 0x1f; + target_aid.target_subblk = (aid >> 3) & 0x7; + target_aid.target_subblk_bit = aid & 0x7; + + /* + * Find our AIDs target encoded block and fill &enc_blk with the + * encoded blocks information. If no entry is found or an error + * occurs return false. + */ + err = ieee80211_s1g_find_target_block(&enc_blk, &target_aid, + tim->virtual_map, + (const u8 *)tim + tim_len + 2); + if (err) + return false; + + switch (enc_blk.enc_mode) { + case IEEE80211_S1G_TIM_ENC_MODE_BLOCK: + return ieee80211_s1g_parse_bitmap(&enc_blk, &target_aid); + case IEEE80211_S1G_TIM_ENC_MODE_SINGLE: + return ieee80211_s1g_parse_single(&enc_blk, &target_aid); + case IEEE80211_S1G_TIM_ENC_MODE_OLB: + return ieee80211_s1g_parse_olb(&enc_blk, &target_aid); + default: + return false; + } +} + +#endif /* LINUX_IEEE80211_H */ diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h index 63a9775b059d..1b27bbac145b 100644 --- a/include/linux/ieee80211.h +++ b/include/linux/ieee80211.h @@ -109,17 +109,6 @@ #define IEEE80211_STYPE_DMG_BEACON 0x0000 #define IEEE80211_STYPE_S1G_BEACON 0x0010 -/* bits unique to S1G beacon */ -#define IEEE80211_S1G_BCN_NEXT_TBTT 0x100 -#define IEEE80211_S1G_BCN_CSSID 0x200 -#define IEEE80211_S1G_BCN_ANO 0x400 - -/* see 802.11ah-2016 9.9 NDP CMAC frames */ -#define IEEE80211_S1G_1MHZ_NDP_BITS 25 -#define IEEE80211_S1G_1MHZ_NDP_BYTES 4 -#define IEEE80211_S1G_2MHZ_NDP_BITS 37 -#define IEEE80211_S1G_2MHZ_NDP_BYTES 5 - #define IEEE80211_NDP_FTYPE_CTS 0 #define IEEE80211_NDP_FTYPE_CF_END 0 #define IEEE80211_NDP_FTYPE_PS_POLL 1 @@ -221,11 +210,6 @@ static inline u16 ieee80211_sn_sub(u16 sn1, u16 sn2) #define IEEE80211_MAX_TIM_LEN 251 #define IEEE80211_MAX_MESH_PEERINGS 63 -/* S1G encoding types */ -#define IEEE80211_S1G_TIM_ENC_MODE_BLOCK 0 -#define IEEE80211_S1G_TIM_ENC_MODE_SINGLE 1 -#define IEEE80211_S1G_TIM_ENC_MODE_OLB 2 - /* Maximum size for the MA-UNITDATA primitive, 802.11 standard section 6.2.1.1.2. @@ -604,55 +588,6 @@ static inline bool ieee80211_is_beacon(__le16 fc) cpu_to_le16(IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_BEACON); } -/** - * ieee80211_is_s1g_beacon - check if IEEE80211_FTYPE_EXT && - * IEEE80211_STYPE_S1G_BEACON - * @fc: frame control bytes in little-endian byteorder - * Return: whether or not the frame is an S1G beacon - */ -static inline bool ieee80211_is_s1g_beacon(__le16 fc) -{ - return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | - IEEE80211_FCTL_STYPE)) == - cpu_to_le16(IEEE80211_FTYPE_EXT | IEEE80211_STYPE_S1G_BEACON); -} - -/** - * ieee80211_s1g_has_next_tbtt - check if IEEE80211_S1G_BCN_NEXT_TBTT - * @fc: frame control bytes in little-endian byteorder - * Return: whether or not the frame contains the variable-length - * next TBTT field - */ -static inline bool ieee80211_s1g_has_next_tbtt(__le16 fc) -{ - return ieee80211_is_s1g_beacon(fc) && - (fc & cpu_to_le16(IEEE80211_S1G_BCN_NEXT_TBTT)); -} - -/** - * ieee80211_s1g_has_ano - check if IEEE80211_S1G_BCN_ANO - * @fc: frame control bytes in little-endian byteorder - * Return: whether or not the frame contains the variable-length - * ANO field - */ -static inline bool ieee80211_s1g_has_ano(__le16 fc) -{ - return ieee80211_is_s1g_beacon(fc) && - (fc & cpu_to_le16(IEEE80211_S1G_BCN_ANO)); -} - -/** - * ieee80211_s1g_has_cssid - check if IEEE80211_S1G_BCN_CSSID - * @fc: frame control bytes in little-endian byteorder - * Return: whether or not the frame contains the variable-length - * compressed SSID field - */ -static inline bool ieee80211_s1g_has_cssid(__le16 fc) -{ - return ieee80211_is_s1g_beacon(fc) && - (fc & cpu_to_le16(IEEE80211_S1G_BCN_CSSID)); -} - /** * ieee80211_is_atim - check if IEEE80211_FTYPE_MGMT && IEEE80211_STYPE_ATIM * @fc: frame control bytes in little-endian byteorder @@ -984,37 +919,6 @@ struct ieee80211_tim_ie { }; } __packed; -/** - * enum ieee80211_s1g_chanwidth - S1G channel widths - * These are defined in IEEE802.11-2016ah Table 10-20 - * as BSS Channel Width - * - * @IEEE80211_S1G_CHANWIDTH_1MHZ: 1MHz operating channel - * @IEEE80211_S1G_CHANWIDTH_2MHZ: 2MHz operating channel - * @IEEE80211_S1G_CHANWIDTH_4MHZ: 4MHz operating channel - * @IEEE80211_S1G_CHANWIDTH_8MHZ: 8MHz operating channel - * @IEEE80211_S1G_CHANWIDTH_16MHZ: 16MHz operating channel - */ -enum ieee80211_s1g_chanwidth { - IEEE80211_S1G_CHANWIDTH_1MHZ = 0, - IEEE80211_S1G_CHANWIDTH_2MHZ = 1, - IEEE80211_S1G_CHANWIDTH_4MHZ = 3, - IEEE80211_S1G_CHANWIDTH_8MHZ = 7, - IEEE80211_S1G_CHANWIDTH_16MHZ = 15, -}; - -/** - * enum ieee80211_s1g_pri_chanwidth - S1G primary channel widths - * described in IEEE80211-2024 Table 10-39. - * - * @IEEE80211_S1G_PRI_CHANWIDTH_2MHZ: 2MHz primary channel - * @IEEE80211_S1G_PRI_CHANWIDTH_1MHZ: 1MHz primary channel - */ -enum ieee80211_s1g_pri_chanwidth { - IEEE80211_S1G_PRI_CHANWIDTH_2MHZ = 0, - IEEE80211_S1G_PRI_CHANWIDTH_1MHZ = 1, -}; - #define WLAN_SA_QUERY_TR_ID_LEN 2 #define WLAN_MEMBERSHIP_LEN 8 #define WLAN_USER_POSITION_LEN 16 @@ -1042,61 +946,6 @@ struct ieee80211_addba_ext_ie { u8 data; } __packed; -/** - * struct ieee80211_s1g_bcn_compat_ie - S1G Beacon Compatibility element - * @compat_info: Compatibility Information - * @beacon_int: Beacon Interval - * @tsf_completion: TSF Completion - * - * This structure represents the payload of the "S1G Beacon - * Compatibility element" as described in IEEE Std 802.11-2020 section - * 9.4.2.196. - */ -struct ieee80211_s1g_bcn_compat_ie { - __le16 compat_info; - __le16 beacon_int; - __le32 tsf_completion; -} __packed; - -/** - * struct ieee80211_s1g_oper_ie - S1G Operation element - * @ch_width: S1G Operation Information Channel Width - * @oper_class: S1G Operation Information Operating Class - * @primary_ch: S1G Operation Information Primary Channel Number - * @oper_ch: S1G Operation Information Channel Center Frequency - * @basic_mcs_nss: Basic S1G-MCS and NSS Set - * - * This structure represents the payload of the "S1G Operation - * element" as described in IEEE Std 802.11-2020 section 9.4.2.212. - */ -struct ieee80211_s1g_oper_ie { - u8 ch_width; - u8 oper_class; - u8 primary_ch; - u8 oper_ch; - __le16 basic_mcs_nss; -} __packed; - -/** - * struct ieee80211_aid_response_ie - AID Response element - * @aid: AID/Group AID - * @switch_count: AID Switch Count - * @response_int: AID Response Interval - * - * This structure represents the payload of the "AID Response element" - * as described in IEEE Std 802.11-2020 section 9.4.2.194. - */ -struct ieee80211_aid_response_ie { - __le16 aid; - u8 switch_count; - __le16 response_int; -} __packed; - -struct ieee80211_s1g_cap { - u8 capab_info[10]; - u8 supp_mcs_nss[5]; -} __packed; - struct ieee80211_ext { __le16 frame_control; __le16 duration; @@ -1110,37 +959,6 @@ struct ieee80211_ext { } u; } __packed __aligned(2); -/** - * ieee80211_s1g_optional_len - determine length of optional S1G beacon fields - * @fc: frame control bytes in little-endian byteorder - * Return: total length in bytes of the optional fixed-length fields - * - * S1G beacons may contain up to three optional fixed-length fields that - * precede the variable-length elements. Whether these fields are present - * is indicated by flags in the frame control field. - * - * From IEEE 802.11-2024 section 9.3.4.3: - * - Next TBTT field may be 0 or 3 bytes - * - Short SSID field may be 0 or 4 bytes - * - Access Network Options (ANO) field may be 0 or 1 byte - */ -static inline size_t -ieee80211_s1g_optional_len(__le16 fc) -{ - size_t len = 0; - - if (ieee80211_s1g_has_next_tbtt(fc)) - len += 3; - - if (ieee80211_s1g_has_cssid(fc)) - len += 4; - - if (ieee80211_s1g_has_ano(fc)) - len += 1; - - return len; -} - /** * struct ieee80211_bss_load_elem - BSS Load elemen * @@ -1567,98 +1385,6 @@ struct ieee80211_p2p_noa_attr { #define IEEE80211_P2P_OPPPS_ENABLE_BIT BIT(7) #define IEEE80211_P2P_OPPPS_CTWINDOW_MASK 0x7F -/* S1G Capabilities Information field */ -#define IEEE80211_S1G_CAPABILITY_LEN 15 - -#define S1G_CAP0_S1G_LONG BIT(0) -#define S1G_CAP0_SGI_1MHZ BIT(1) -#define S1G_CAP0_SGI_2MHZ BIT(2) -#define S1G_CAP0_SGI_4MHZ BIT(3) -#define S1G_CAP0_SGI_8MHZ BIT(4) -#define S1G_CAP0_SGI_16MHZ BIT(5) -#define S1G_CAP0_SUPP_CH_WIDTH GENMASK(7, 6) - -#define S1G_SUPP_CH_WIDTH_2 0 -#define S1G_SUPP_CH_WIDTH_4 1 -#define S1G_SUPP_CH_WIDTH_8 2 -#define S1G_SUPP_CH_WIDTH_16 3 -#define S1G_SUPP_CH_WIDTH_MAX(cap) ((1 << FIELD_GET(S1G_CAP0_SUPP_CH_WIDTH, \ - cap[0])) << 1) - -#define S1G_CAP1_RX_LDPC BIT(0) -#define S1G_CAP1_TX_STBC BIT(1) -#define S1G_CAP1_RX_STBC BIT(2) -#define S1G_CAP1_SU_BFER BIT(3) -#define S1G_CAP1_SU_BFEE BIT(4) -#define S1G_CAP1_BFEE_STS GENMASK(7, 5) - -#define S1G_CAP2_SOUNDING_DIMENSIONS GENMASK(2, 0) -#define S1G_CAP2_MU_BFER BIT(3) -#define S1G_CAP2_MU_BFEE BIT(4) -#define S1G_CAP2_PLUS_HTC_VHT BIT(5) -#define S1G_CAP2_TRAVELING_PILOT GENMASK(7, 6) - -#define S1G_CAP3_RD_RESPONDER BIT(0) -#define S1G_CAP3_HT_DELAYED_BA BIT(1) -#define S1G_CAP3_MAX_MPDU_LEN BIT(2) -#define S1G_CAP3_MAX_AMPDU_LEN_EXP GENMASK(4, 3) -#define S1G_CAP3_MIN_MPDU_START GENMASK(7, 5) - -#define S1G_CAP4_UPLINK_SYNC BIT(0) -#define S1G_CAP4_DYNAMIC_AID BIT(1) -#define S1G_CAP4_BAT BIT(2) -#define S1G_CAP4_TIME_ADE BIT(3) -#define S1G_CAP4_NON_TIM BIT(4) -#define S1G_CAP4_GROUP_AID BIT(5) -#define S1G_CAP4_STA_TYPE GENMASK(7, 6) - -#define S1G_CAP5_CENT_AUTH_CONTROL BIT(0) -#define S1G_CAP5_DIST_AUTH_CONTROL BIT(1) -#define S1G_CAP5_AMSDU BIT(2) -#define S1G_CAP5_AMPDU BIT(3) -#define S1G_CAP5_ASYMMETRIC_BA BIT(4) -#define S1G_CAP5_FLOW_CONTROL BIT(5) -#define S1G_CAP5_SECTORIZED_BEAM GENMASK(7, 6) - -#define S1G_CAP6_OBSS_MITIGATION BIT(0) -#define S1G_CAP6_FRAGMENT_BA BIT(1) -#define S1G_CAP6_NDP_PS_POLL BIT(2) -#define S1G_CAP6_RAW_OPERATION BIT(3) -#define S1G_CAP6_PAGE_SLICING BIT(4) -#define S1G_CAP6_TXOP_SHARING_IMP_ACK BIT(5) -#define S1G_CAP6_VHT_LINK_ADAPT GENMASK(7, 6) - -#define S1G_CAP7_TACK_AS_PS_POLL BIT(0) -#define S1G_CAP7_DUP_1MHZ BIT(1) -#define S1G_CAP7_MCS_NEGOTIATION BIT(2) -#define S1G_CAP7_1MHZ_CTL_RESPONSE_PREAMBLE BIT(3) -#define S1G_CAP7_NDP_BFING_REPORT_POLL BIT(4) -#define S1G_CAP7_UNSOLICITED_DYN_AID BIT(5) -#define S1G_CAP7_SECTOR_TRAINING_OPERATION BIT(6) -#define S1G_CAP7_TEMP_PS_MODE_SWITCH BIT(7) - -#define S1G_CAP8_TWT_GROUPING BIT(0) -#define S1G_CAP8_BDT BIT(1) -#define S1G_CAP8_COLOR GENMASK(4, 2) -#define S1G_CAP8_TWT_REQUEST BIT(5) -#define S1G_CAP8_TWT_RESPOND BIT(6) -#define S1G_CAP8_PV1_FRAME BIT(7) - -#define S1G_CAP9_LINK_ADAPT_PER_CONTROL_RESPONSE BIT(0) - -#define S1G_OPER_CH_WIDTH_PRIMARY BIT(0) -#define S1G_OPER_CH_WIDTH_OPER GENMASK(4, 1) -#define S1G_OPER_CH_PRIMARY_LOCATION BIT(5) - -#define S1G_2M_PRIMARY_LOCATION_LOWER 0 -#define S1G_2M_PRIMARY_LOCATION_UPPER 1 - -#define LISTEN_INT_USF GENMASK(15, 14) -#define LISTEN_INT_UI GENMASK(13, 0) - -#define IEEE80211_MAX_USF FIELD_MAX(LISTEN_INT_USF) -#define IEEE80211_MAX_UI FIELD_MAX(LISTEN_INT_UI) - /* Authentication algorithms */ #define WLAN_AUTH_OPEN 0 #define WLAN_AUTH_SHARED_KEY 1 @@ -2189,20 +1915,6 @@ enum ieee80211_key_len { WLAN_KEY_LEN_BIP_GMAC_256 = 32, }; -enum ieee80211_s1g_actioncode { - WLAN_S1G_AID_SWITCH_REQUEST, - WLAN_S1G_AID_SWITCH_RESPONSE, - WLAN_S1G_SYNC_CONTROL, - WLAN_S1G_STA_INFO_ANNOUNCE, - WLAN_S1G_EDCA_PARAM_SET, - WLAN_S1G_EL_OPERATION, - WLAN_S1G_TWT_SETUP, - WLAN_S1G_TWT_TEARDOWN, - WLAN_S1G_SECT_GROUP_ID_LIST, - WLAN_S1G_SECT_ID_FEEDBACK, - WLAN_S1G_TWT_INFORMATION = 11, -}; - /* Radio measurement action codes as defined in IEEE 802.11-2024 - Table 9-470 */ enum ieee80211_radio_measurement_actioncode { WLAN_RM_ACTION_RADIO_MEASUREMENT_REQUEST = 0, @@ -2877,254 +2589,6 @@ static inline bool __ieee80211_check_tim(const struct ieee80211_tim_ie *tim, return !!(tim->virtual_map[index] & mask); } -struct s1g_tim_aid { - u16 aid; - u8 target_blk; /* Target block index */ - u8 target_subblk; /* Target subblock index */ - u8 target_subblk_bit; /* Target subblock bit */ -}; - -struct s1g_tim_enc_block { - u8 enc_mode; - bool inverse; - const u8 *ptr; - u8 len; - - /* - * For an OLB encoded block that spans multiple blocks, this - * is the offset into the span described by that encoded block. - */ - u8 olb_blk_offset; -}; - -/* - * Helper routines to quickly extract the length of an encoded block. Validation - * is also performed to ensure the length extracted lies within the TIM. - */ - -static inline int ieee80211_s1g_len_bitmap(const u8 *ptr, const u8 *end) -{ - u8 blkmap; - u8 n_subblks; - - if (ptr >= end) - return -EINVAL; - - blkmap = *ptr; - n_subblks = hweight8(blkmap); - - if (ptr + 1 + n_subblks > end) - return -EINVAL; - - return 1 + n_subblks; -} - -static inline int ieee80211_s1g_len_single(const u8 *ptr, const u8 *end) -{ - return (ptr + 1 > end) ? -EINVAL : 1; -} - -static inline int ieee80211_s1g_len_olb(const u8 *ptr, const u8 *end) -{ - if (ptr >= end) - return -EINVAL; - - return (ptr + 1 + *ptr > end) ? -EINVAL : 1 + *ptr; -} - -/* - * Enumerate all encoded blocks until we find the encoded block that describes - * our target AID. OLB is a special case as a single encoded block can describe - * multiple blocks as a single encoded block. - */ -static inline int ieee80211_s1g_find_target_block(struct s1g_tim_enc_block *enc, - const struct s1g_tim_aid *aid, - const u8 *ptr, const u8 *end) -{ - /* need at least block-control octet */ - while (ptr + 1 <= end) { - u8 ctrl = *ptr++; - u8 mode = ctrl & 0x03; - bool contains, inverse = ctrl & BIT(2); - u8 span, blk_off = ctrl >> 3; - int len; - - switch (mode) { - case IEEE80211_S1G_TIM_ENC_MODE_BLOCK: - len = ieee80211_s1g_len_bitmap(ptr, end); - contains = blk_off == aid->target_blk; - break; - case IEEE80211_S1G_TIM_ENC_MODE_SINGLE: - len = ieee80211_s1g_len_single(ptr, end); - contains = blk_off == aid->target_blk; - break; - case IEEE80211_S1G_TIM_ENC_MODE_OLB: - len = ieee80211_s1g_len_olb(ptr, end); - /* - * An OLB encoded block can describe more then one - * block, meaning an encoded OLB block can span more - * then a single block. - */ - if (len > 0) { - /* Minus one for the length octet */ - span = DIV_ROUND_UP(len - 1, 8); - /* - * Check if our target block lies within the - * block span described by this encoded block. - */ - contains = (aid->target_blk >= blk_off) && - (aid->target_blk < blk_off + span); - } - break; - default: - return -EOPNOTSUPP; - } - - if (len < 0) - return len; - - if (contains) { - enc->enc_mode = mode; - enc->inverse = inverse; - enc->ptr = ptr; - enc->len = (u8)len; - enc->olb_blk_offset = blk_off; - return 0; - } - - ptr += len; - } - - return -ENOENT; -} - -static inline bool ieee80211_s1g_parse_bitmap(struct s1g_tim_enc_block *enc, - struct s1g_tim_aid *aid) -{ - const u8 *ptr = enc->ptr; - u8 blkmap = *ptr++; - - /* - * If our block bitmap does not contain a set bit that corresponds - * to our AID, it could mean a variety of things depending on if - * the encoding mode is inverted or not. - * - * 1. If inverted, it means the entire subblock is present and hence - * our AID has been set. - * 2. If not inverted, it means our subblock is not present and hence - * it is all zero meaning our AID is not set. - */ - if (!(blkmap & BIT(aid->target_subblk))) - return enc->inverse; - - /* - * Increment ptr by the number of set subblocks that appear before our - * target subblock. If our target subblock is 0, do nothing as ptr - * already points to our target subblock. - */ - if (aid->target_subblk) - ptr += hweight8(blkmap & GENMASK(aid->target_subblk - 1, 0)); - - return !!(*ptr & BIT(aid->target_subblk_bit)) ^ enc->inverse; -} - -static inline bool ieee80211_s1g_parse_single(struct s1g_tim_enc_block *enc, - struct s1g_tim_aid *aid) -{ - /* - * Single AID mode describes, as the name suggests, a single AID - * within the block described by the encoded block. The octet - * contains the 6 LSBs of the AID described in the block. The other - * 2 bits are reserved. When inversed, every single AID described - * by the current block have buffered traffic except for the AID - * described in the single AID octet. - */ - return ((*enc->ptr & 0x3f) == (aid->aid & 0x3f)) ^ enc->inverse; -} - -static inline bool ieee80211_s1g_parse_olb(struct s1g_tim_enc_block *enc, - struct s1g_tim_aid *aid) -{ - const u8 *ptr = enc->ptr; - u8 blk_len = *ptr++; - /* - * Given an OLB encoded block that describes multiple blocks, - * calculate the offset into the span. Then calculate the - * subblock location normally. - */ - u16 span_offset = aid->target_blk - enc->olb_blk_offset; - u16 subblk_idx = span_offset * 8 + aid->target_subblk; - - if (subblk_idx >= blk_len) - return enc->inverse; - - return !!(ptr[subblk_idx] & BIT(aid->target_subblk_bit)) ^ enc->inverse; -} - -/* - * An S1G PVB has 3 non optional encoding types, each that can be inverted. - * An S1G PVB is constructed with zero or more encoded block subfields. Each - * encoded block represents a single "block" of AIDs (64), and each encoded - * block can contain one of the 3 encoding types alongside a single bit for - * whether the bits should be inverted. - * - * As the standard makes no guarantee about the ordering of encoded blocks, - * we must parse every encoded block in the worst case scenario given an - * AID that lies within the last block. - */ -static inline bool ieee80211_s1g_check_tim(const struct ieee80211_tim_ie *tim, - u8 tim_len, u16 aid) -{ - int err; - struct s1g_tim_aid target_aid; - struct s1g_tim_enc_block enc_blk; - - if (tim_len < 3) - return false; - - target_aid.aid = aid; - target_aid.target_blk = (aid >> 6) & 0x1f; - target_aid.target_subblk = (aid >> 3) & 0x7; - target_aid.target_subblk_bit = aid & 0x7; - - /* - * Find our AIDs target encoded block and fill &enc_blk with the - * encoded blocks information. If no entry is found or an error - * occurs return false. - */ - err = ieee80211_s1g_find_target_block(&enc_blk, &target_aid, - tim->virtual_map, - (const u8 *)tim + tim_len + 2); - if (err) - return false; - - switch (enc_blk.enc_mode) { - case IEEE80211_S1G_TIM_ENC_MODE_BLOCK: - return ieee80211_s1g_parse_bitmap(&enc_blk, &target_aid); - case IEEE80211_S1G_TIM_ENC_MODE_SINGLE: - return ieee80211_s1g_parse_single(&enc_blk, &target_aid); - case IEEE80211_S1G_TIM_ENC_MODE_OLB: - return ieee80211_s1g_parse_olb(&enc_blk, &target_aid); - default: - return false; - } -} - -/** - * ieee80211_check_tim - check if AID bit is set in TIM - * @tim: the TIM IE - * @tim_len: length of the TIM IE - * @aid: the AID to look for - * @s1g: whether the TIM is from an S1G PPDU - * Return: whether or not traffic is indicated in the TIM for the given AID - */ -static inline bool ieee80211_check_tim(const struct ieee80211_tim_ie *tim, - u8 tim_len, u16 aid, bool s1g) -{ - return s1g ? ieee80211_s1g_check_tim(tim, tim_len, aid) : - __ieee80211_check_tim(tim, tim_len, aid); -} - /** * ieee80211_get_tdls_action - get TDLS action code * @skb: the skb containing the frame, length will not be checked @@ -3258,39 +2722,6 @@ static inline bool ieee80211_is_ftm(struct sk_buff *skb) return false; } -/** - * ieee80211_is_s1g_short_beacon - check if frame is an S1G short beacon - * @fc: frame control bytes in little-endian byteorder - * @variable: pointer to the beacon frame elements - * @variable_len: length of the frame elements - * Return: whether or not the frame is an S1G short beacon. As per - * IEEE80211-2024 11.1.3.10.1, The S1G beacon compatibility element shall - * always be present as the first element in beacon frames generated at a - * TBTT (Target Beacon Transmission Time), so any frame not containing - * this element must have been generated at a TSBTT (Target Short Beacon - * Transmission Time) that is not a TBTT. Additionally, short beacons are - * prohibited from containing the S1G beacon compatibility element as per - * IEEE80211-2024 9.3.4.3 Table 9-76, so if we have an S1G beacon with - * either no elements or the first element is not the beacon compatibility - * element, we have a short beacon. - */ -static inline bool ieee80211_is_s1g_short_beacon(__le16 fc, const u8 *variable, - size_t variable_len) -{ - if (!ieee80211_is_s1g_beacon(fc)) - return false; - - /* - * If the frame does not contain at least 1 element (this is perfectly - * valid in a short beacon) and is an S1G beacon, we have a short - * beacon. - */ - if (variable_len < 2) - return true; - - return variable[0] != WLAN_EID_S1G_BCN_COMPAT; -} - struct element { u8 id; u8 datalen; @@ -3446,5 +2877,21 @@ struct ieee80211_tbtt_info_ge_11 { #include "ieee80211-he.h" #include "ieee80211-eht.h" #include "ieee80211-mesh.h" +#include "ieee80211-s1g.h" + +/** + * ieee80211_check_tim - check if AID bit is set in TIM + * @tim: the TIM IE + * @tim_len: length of the TIM IE + * @aid: the AID to look for + * @s1g: whether the TIM is from an S1G PPDU + * Return: whether or not traffic is indicated in the TIM for the given AID + */ +static inline bool ieee80211_check_tim(const struct ieee80211_tim_ie *tim, + u8 tim_len, u16 aid, bool s1g) +{ + return s1g ? ieee80211_s1g_check_tim(tim, tim_len, aid) : + __ieee80211_check_tim(tim, tim_len, aid); +} #endif /* LINUX_IEEE80211_H */ From fcd42b909ba06737dfcda47f3a0a9718bd3ebf03 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 5 Nov 2025 15:36:55 +0100 Subject: [PATCH 0639/1381] wifi: ieee80211: split P2P definitions out The ieee80211.h file has gotten very long, continue splitting it by putting P2P definitions into a separate file. Note that P2P isn't really even IEEE 802.11 but WFA. Link: https://patch.msgid.link/20251105153843.e47b2614e9d2.Id242f61da720e365f6b5d7a4a545fbbc2f1e92b4@changeid Signed-off-by: Johannes Berg --- include/linux/ieee80211-p2p.h | 71 +++++++++++++++++++++++++++++++++++ include/linux/ieee80211.h | 53 +------------------------- 2 files changed, 72 insertions(+), 52 deletions(-) create mode 100644 include/linux/ieee80211-p2p.h diff --git a/include/linux/ieee80211-p2p.h b/include/linux/ieee80211-p2p.h new file mode 100644 index 000000000000..180891c11f08 --- /dev/null +++ b/include/linux/ieee80211-p2p.h @@ -0,0 +1,71 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * WFA P2P definitions + * + * Copyright (c) 2001-2002, SSH Communications Security Corp and Jouni Malinen + * + * Copyright (c) 2002-2003, Jouni Malinen + * Copyright (c) 2005, Devicescape Software, Inc. + * Copyright (c) 2006, Michael Wu + * Copyright (c) 2013 - 2014 Intel Mobile Communications GmbH + * Copyright (c) 2016 - 2017 Intel Deutschland GmbH + * Copyright (c) 2018 - 2025 Intel Corporation + */ + +#ifndef LINUX_IEEE80211_P2P_H +#define LINUX_IEEE80211_P2P_H + +#include +/* + * Peer-to-Peer IE attribute related definitions. + */ +/* + * enum ieee80211_p2p_attr_id - identifies type of peer-to-peer attribute. + */ +enum ieee80211_p2p_attr_id { + IEEE80211_P2P_ATTR_STATUS = 0, + IEEE80211_P2P_ATTR_MINOR_REASON, + IEEE80211_P2P_ATTR_CAPABILITY, + IEEE80211_P2P_ATTR_DEVICE_ID, + IEEE80211_P2P_ATTR_GO_INTENT, + IEEE80211_P2P_ATTR_GO_CONFIG_TIMEOUT, + IEEE80211_P2P_ATTR_LISTEN_CHANNEL, + IEEE80211_P2P_ATTR_GROUP_BSSID, + IEEE80211_P2P_ATTR_EXT_LISTEN_TIMING, + IEEE80211_P2P_ATTR_INTENDED_IFACE_ADDR, + IEEE80211_P2P_ATTR_MANAGABILITY, + IEEE80211_P2P_ATTR_CHANNEL_LIST, + IEEE80211_P2P_ATTR_ABSENCE_NOTICE, + IEEE80211_P2P_ATTR_DEVICE_INFO, + IEEE80211_P2P_ATTR_GROUP_INFO, + IEEE80211_P2P_ATTR_GROUP_ID, + IEEE80211_P2P_ATTR_INTERFACE, + IEEE80211_P2P_ATTR_OPER_CHANNEL, + IEEE80211_P2P_ATTR_INVITE_FLAGS, + /* 19 - 220: Reserved */ + IEEE80211_P2P_ATTR_VENDOR_SPECIFIC = 221, + + IEEE80211_P2P_ATTR_MAX +}; + +/* Notice of Absence attribute - described in P2P spec 4.1.14 */ +/* Typical max value used here */ +#define IEEE80211_P2P_NOA_DESC_MAX 4 + +struct ieee80211_p2p_noa_desc { + u8 count; + __le32 duration; + __le32 interval; + __le32 start_time; +} __packed; + +struct ieee80211_p2p_noa_attr { + u8 index; + u8 oppps_ctwindow; + struct ieee80211_p2p_noa_desc desc[IEEE80211_P2P_NOA_DESC_MAX]; +} __packed; + +#define IEEE80211_P2P_OPPPS_ENABLE_BIT BIT(7) +#define IEEE80211_P2P_OPPPS_CTWINDOW_MASK 0x7F + +#endif /* LINUX_IEEE80211_P2P_H */ diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h index 1b27bbac145b..fa0f7f917ce7 100644 --- a/include/linux/ieee80211.h +++ b/include/linux/ieee80211.h @@ -1333,58 +1333,6 @@ struct ieee80211_tdls_data { } u; } __packed; -/* - * Peer-to-Peer IE attribute related definitions. - */ -/* - * enum ieee80211_p2p_attr_id - identifies type of peer-to-peer attribute. - */ -enum ieee80211_p2p_attr_id { - IEEE80211_P2P_ATTR_STATUS = 0, - IEEE80211_P2P_ATTR_MINOR_REASON, - IEEE80211_P2P_ATTR_CAPABILITY, - IEEE80211_P2P_ATTR_DEVICE_ID, - IEEE80211_P2P_ATTR_GO_INTENT, - IEEE80211_P2P_ATTR_GO_CONFIG_TIMEOUT, - IEEE80211_P2P_ATTR_LISTEN_CHANNEL, - IEEE80211_P2P_ATTR_GROUP_BSSID, - IEEE80211_P2P_ATTR_EXT_LISTEN_TIMING, - IEEE80211_P2P_ATTR_INTENDED_IFACE_ADDR, - IEEE80211_P2P_ATTR_MANAGABILITY, - IEEE80211_P2P_ATTR_CHANNEL_LIST, - IEEE80211_P2P_ATTR_ABSENCE_NOTICE, - IEEE80211_P2P_ATTR_DEVICE_INFO, - IEEE80211_P2P_ATTR_GROUP_INFO, - IEEE80211_P2P_ATTR_GROUP_ID, - IEEE80211_P2P_ATTR_INTERFACE, - IEEE80211_P2P_ATTR_OPER_CHANNEL, - IEEE80211_P2P_ATTR_INVITE_FLAGS, - /* 19 - 220: Reserved */ - IEEE80211_P2P_ATTR_VENDOR_SPECIFIC = 221, - - IEEE80211_P2P_ATTR_MAX -}; - -/* Notice of Absence attribute - described in P2P spec 4.1.14 */ -/* Typical max value used here */ -#define IEEE80211_P2P_NOA_DESC_MAX 4 - -struct ieee80211_p2p_noa_desc { - u8 count; - __le32 duration; - __le32 interval; - __le32 start_time; -} __packed; - -struct ieee80211_p2p_noa_attr { - u8 index; - u8 oppps_ctwindow; - struct ieee80211_p2p_noa_desc desc[IEEE80211_P2P_NOA_DESC_MAX]; -} __packed; - -#define IEEE80211_P2P_OPPPS_ENABLE_BIT BIT(7) -#define IEEE80211_P2P_OPPPS_CTWINDOW_MASK 0x7F - /* Authentication algorithms */ #define WLAN_AUTH_OPEN 0 #define WLAN_AUTH_SHARED_KEY 1 @@ -2878,6 +2826,7 @@ struct ieee80211_tbtt_info_ge_11 { #include "ieee80211-eht.h" #include "ieee80211-mesh.h" #include "ieee80211-s1g.h" +#include "ieee80211-p2p.h" /** * ieee80211_check_tim - check if AID bit is set in TIM From 60a3734192fa6909c48e33b0d212990ebaff54c4 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 5 Nov 2025 15:36:56 +0100 Subject: [PATCH 0640/1381] wifi: ieee80211: split NAN definitions out The ieee80211.h file has gotten very long, continue splitting it by putting NAN definitions into a separate file. Note that NAN isn't really even IEEE 802.11 but WFA. Link: https://patch.msgid.link/20251105153843.8da0e796dda2.I7b2ce11220b70e8794019501eabbf8afbaf431a6@changeid Signed-off-by: Johannes Berg --- include/linux/ieee80211-nan.h | 35 +++++++++++++++++++++++++++++++++++ include/linux/ieee80211.h | 18 +----------------- 2 files changed, 36 insertions(+), 17 deletions(-) create mode 100644 include/linux/ieee80211-nan.h diff --git a/include/linux/ieee80211-nan.h b/include/linux/ieee80211-nan.h new file mode 100644 index 000000000000..d07959bf8a90 --- /dev/null +++ b/include/linux/ieee80211-nan.h @@ -0,0 +1,35 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * WFA NAN definitions + * + * Copyright (c) 2001-2002, SSH Communications Security Corp and Jouni Malinen + * + * Copyright (c) 2002-2003, Jouni Malinen + * Copyright (c) 2005, Devicescape Software, Inc. + * Copyright (c) 2006, Michael Wu + * Copyright (c) 2013 - 2014 Intel Mobile Communications GmbH + * Copyright (c) 2016 - 2017 Intel Deutschland GmbH + * Copyright (c) 2018 - 2025 Intel Corporation + */ + +#ifndef LINUX_IEEE80211_NAN_H +#define LINUX_IEEE80211_NAN_H + +/* NAN operation mode, as defined in Wi-Fi Aware (TM) specification Table 81 */ +#define NAN_OP_MODE_PHY_MODE_VHT 0x01 +#define NAN_OP_MODE_PHY_MODE_HE 0x10 +#define NAN_OP_MODE_PHY_MODE_MASK 0x11 +#define NAN_OP_MODE_80P80MHZ 0x02 +#define NAN_OP_MODE_160MHZ 0x04 +#define NAN_OP_MODE_PNDL_SUPPRTED 0x08 + +/* NAN Device capabilities, as defined in Wi-Fi Aware (TM) specification + * Table 79 + */ +#define NAN_DEV_CAPA_DFS_OWNER 0x01 +#define NAN_DEV_CAPA_EXT_KEY_ID_SUPPORTED 0x02 +#define NAN_DEV_CAPA_SIM_NDP_RX_SUPPORTED 0x04 +#define NAN_DEV_CAPA_NDPE_SUPPORTED 0x08 +#define NAN_DEV_CAPA_S3_SUPPORTED 0x10 + +#endif /* LINUX_IEEE80211_NAN_H */ diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h index fa0f7f917ce7..48ce05e1d203 100644 --- a/include/linux/ieee80211.h +++ b/include/linux/ieee80211.h @@ -2803,23 +2803,6 @@ struct ieee80211_tbtt_info_ge_11 { struct ieee80211_rnr_mld_params mld_params; } __packed; -/* NAN operation mode, as defined in Wi-Fi Aware (TM) specification Table 81 */ -#define NAN_OP_MODE_PHY_MODE_VHT 0x01 -#define NAN_OP_MODE_PHY_MODE_HE 0x10 -#define NAN_OP_MODE_PHY_MODE_MASK 0x11 -#define NAN_OP_MODE_80P80MHZ 0x02 -#define NAN_OP_MODE_160MHZ 0x04 -#define NAN_OP_MODE_PNDL_SUPPRTED 0x08 - -/* NAN Device capabilities, as defined in Wi-Fi Aware (TM) specification - * Table 79 - */ -#define NAN_DEV_CAPA_DFS_OWNER 0x01 -#define NAN_DEV_CAPA_EXT_KEY_ID_SUPPORTED 0x02 -#define NAN_DEV_CAPA_SIM_NDP_RX_SUPPORTED 0x04 -#define NAN_DEV_CAPA_NDPE_SUPPORTED 0x08 -#define NAN_DEV_CAPA_S3_SUPPORTED 0x10 - #include "ieee80211-ht.h" #include "ieee80211-vht.h" #include "ieee80211-he.h" @@ -2827,6 +2810,7 @@ struct ieee80211_tbtt_info_ge_11 { #include "ieee80211-mesh.h" #include "ieee80211-s1g.h" #include "ieee80211-p2p.h" +#include "ieee80211-nan.h" /** * ieee80211_check_tim - check if AID bit is set in TIM From 30b6089aad35500e683025dddc029ac28705385d Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 5 Nov 2025 15:39:57 +0100 Subject: [PATCH 0641/1381] wifi: cfg80211: fix EHT typo This is clearly EHT, not ETH, fix the typo. Link: https://patch.msgid.link/20251105153958.e9d4af3b768e.I5f3378326837e3f62928a2f1fd3403f29cea069b@changeid Signed-off-by: Johannes Berg --- include/net/cfg80211.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index f2e8963cfaac..84be0cdd1da0 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -685,7 +685,7 @@ ieee80211_get_he_6ghz_capa(const struct ieee80211_supported_band *sband, } /** - * ieee80211_get_eht_iftype_cap - return ETH capabilities for an sband's iftype + * ieee80211_get_eht_iftype_cap - return EHT capabilities for an sband's iftype * @sband: the sband to search for the iftype on * @iftype: enum nl80211_iftype * From 1a1cad924e8a60252132446fbba1284035010b4f Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 5 Nov 2025 15:39:58 +0100 Subject: [PATCH 0642/1381] wifi: mac80211: fix EHT typo This is clearly EHT, not ETH, fix the typo. Link: https://patch.msgid.link/20251105153958.12a04517f7ec.Idcf800817fa30605b1002c3d2287cad016e7aea7@changeid Signed-off-by: Johannes Berg --- include/net/mac80211.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/net/mac80211.h b/include/net/mac80211.h index c326243e1f01..c2e49542626c 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -7223,7 +7223,7 @@ ieee80211_get_he_6ghz_capa_vif(const struct ieee80211_supported_band *sband, } /** - * ieee80211_get_eht_iftype_cap_vif - return ETH capabilities for sband/vif + * ieee80211_get_eht_iftype_cap_vif - return EHT capabilities for sband/vif * @sband: the sband to search for the iftype on * @vif: the vif to get the iftype from * From 29cc798e7061e603843091af2f17b5b25009e7b8 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 5 Nov 2025 16:03:42 +0100 Subject: [PATCH 0643/1381] wifi: mac80211: make link iteration safe for 'break' The current link iteration macros for_each_sdata_link() and for_each_sdata_link_rcu() are various nested for loops, but because they iterate all sdata instances and then all links inside, using 'break' inside the iteration doesn't actually break out of the whole iteration. Make it work by tracking whether or not the inner iteration (over links) actually completed, if it broke out then given list_for_each_entry() it still iterates all sdata instances but won't go into them. Link: https://patch.msgid.link/20251105160431.c21956654fc0.I8d4739af061c44c57d172f19a15303a44ad1e596@changeid Signed-off-by: Johannes Berg --- net/mac80211/ieee80211_i.h | 16 +++++++++++----- 1 file changed, 11 insertions(+), 5 deletions(-) diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index 878c3b14aeb8..72e4f326946f 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -1239,9 +1239,12 @@ struct ieee80211_sub_if_data *vif_to_sdata(struct ieee80211_vif *p) for (struct ieee80211_sub_if_data *___sdata = NULL; \ !___sdata; \ ___sdata = (void *)~0 /* always stop */) \ + for (int ___link_id = ARRAY_SIZE(___sdata->link); \ + ___link_id; ___link_id = 0 /* always stop */) \ list_for_each_entry(___sdata, &(_local)->interfaces, list) \ - if (ieee80211_sdata_running(___sdata)) \ - for (int ___link_id = 0; \ + if (___link_id == ARRAY_SIZE(___sdata->link) && \ + ieee80211_sdata_running(___sdata)) \ + for (___link_id = 0; \ ___link_id < ARRAY_SIZE(___sdata->link); \ ___link_id++) \ if ((_link = wiphy_dereference((_local)->hw.wiphy, \ @@ -1255,9 +1258,12 @@ struct ieee80211_sub_if_data *vif_to_sdata(struct ieee80211_vif *p) for (struct ieee80211_sub_if_data *___sdata = NULL; \ !___sdata; \ ___sdata = (void *)~0 /* always stop */) \ - list_for_each_entry_rcu(___sdata, &(_local)->interfaces, list) \ - if (ieee80211_sdata_running(___sdata)) \ - for (int ___link_id = 0; \ + for (int ___link_id = ARRAY_SIZE(___sdata->link); \ + ___link_id; ___link_id = 0 /* always stop */) \ + list_for_each_entry(___sdata, &(_local)->interfaces, list) \ + if (___link_id == ARRAY_SIZE(___sdata->link) && \ + ieee80211_sdata_running(___sdata)) \ + for (___link_id = 0; \ ___link_id < ARRAY_SIZE((___sdata)->link); \ ___link_id++) \ if ((_link = rcu_dereference((___sdata)->link[___link_id]))) From a1dc648aa76d61d8e75692cecea043b1bfdfeda6 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 5 Nov 2025 16:03:43 +0100 Subject: [PATCH 0644/1381] wifi: mac80211: remove chanctx to link back-references Each link can currently use a chanctx and have another one reserved, and both of these are also tracked backwards in the assigned_links and reserved_links lists. If we consider that there aren't *that* many links, this duplicate book- keeping isn't necessary. (I think it used to be necessary before the wiphy locking changes, when chanctx_mtx existed, because we couldn't do any interface iterations while holding only chanctx_mtx.) Additionally, for NAN, we're going to want to track which chanctxs are in use by the (group of) NAN interfaces. For those, links don't really make sense as such, so chanctxs need to be assigned to a different data structure. Thus, as a first step, remove those back-lists of users (right now only links) of each channel context. This is a very basic conversion, ieee80211_vif_use_reserved_switch() should made to iterate smarter. Link: https://patch.msgid.link/20251105160431.dbeea1c42e76.I8d273c407274e1c05a4778aa20b56a9f326e87a7@changeid Signed-off-by: Johannes Berg --- net/mac80211/chan.c | 127 +++++++++++++++++++++---------------- net/mac80211/ieee80211_i.h | 6 -- net/mac80211/link.c | 2 - net/mac80211/util.c | 32 +++++----- 4 files changed, 88 insertions(+), 79 deletions(-) diff --git a/net/mac80211/chan.c b/net/mac80211/chan.c index 7f8799fd673e..944e1c18501a 100644 --- a/net/mac80211/chan.c +++ b/net/mac80211/chan.c @@ -20,8 +20,10 @@ static int ieee80211_chanctx_num_assigned(struct ieee80211_local *local, lockdep_assert_wiphy(local->hw.wiphy); - list_for_each_entry(link, &ctx->assigned_links, assigned_chanctx_list) - num++; + for_each_sdata_link(local, link) { + if (rcu_access_pointer(link->conf->chanctx_conf) == &ctx->conf) + num++; + } return num; } @@ -34,8 +36,10 @@ static int ieee80211_chanctx_num_reserved(struct ieee80211_local *local, lockdep_assert_wiphy(local->hw.wiphy); - list_for_each_entry(link, &ctx->reserved_links, reserved_chanctx_list) - num++; + for_each_sdata_link(local, link) { + if (link->reserved_chanctx == ctx) + num++; + } return num; } @@ -43,8 +47,19 @@ static int ieee80211_chanctx_num_reserved(struct ieee80211_local *local, int ieee80211_chanctx_refcount(struct ieee80211_local *local, struct ieee80211_chanctx *ctx) { - return ieee80211_chanctx_num_assigned(local, ctx) + - ieee80211_chanctx_num_reserved(local, ctx); + struct ieee80211_link_data *link; + int num = 0; + + lockdep_assert_wiphy(local->hw.wiphy); + + for_each_sdata_link(local, link) { + if (rcu_access_pointer(link->conf->chanctx_conf) == &ctx->conf) + num++; + if (link->reserved_chanctx == ctx) + num++; + } + + return num; } static int ieee80211_num_chanctx(struct ieee80211_local *local, int radio_idx) @@ -150,7 +165,10 @@ ieee80211_chanctx_reserved_chanreq(struct ieee80211_local *local, if (WARN_ON(!req)) return NULL; - list_for_each_entry(link, &ctx->reserved_links, reserved_chanctx_list) { + for_each_sdata_link(local, link) { + if (link->reserved_chanctx != ctx) + continue; + req = ieee80211_chanreq_compatible(&link->reserved, req, tmp); if (!req) break; @@ -170,9 +188,12 @@ ieee80211_chanctx_non_reserved_chandef(struct ieee80211_local *local, lockdep_assert_wiphy(local->hw.wiphy); - list_for_each_entry(link, &ctx->assigned_links, assigned_chanctx_list) { + for_each_sdata_link(local, link) { struct ieee80211_bss_conf *link_conf = link->conf; + if (rcu_access_pointer(link_conf->chanctx_conf) != &ctx->conf) + continue; + if (link->reserved_chanctx) continue; @@ -200,7 +221,7 @@ ieee80211_chanctx_can_reserve(struct ieee80211_local *local, if (!ieee80211_chanctx_non_reserved_chandef(local, ctx, req, &tmp)) return false; - if (!list_empty(&ctx->reserved_links) && + if (ieee80211_chanctx_num_reserved(local, ctx) != 0 && ieee80211_chanctx_reserved_chanreq(local, ctx, req, &tmp)) return true; @@ -633,8 +654,6 @@ ieee80211_find_chanctx(struct ieee80211_local *local, * context to actually be removed. */ link->reserved_chanctx = ctx; - list_add(&link->reserved_chanctx_list, - &ctx->reserved_links); ieee80211_change_chanctx(local, ctx, ctx, compat); @@ -705,8 +724,6 @@ ieee80211_alloc_chanctx(struct ieee80211_local *local, if (!ctx) return NULL; - INIT_LIST_HEAD(&ctx->assigned_links); - INIT_LIST_HEAD(&ctx->reserved_links); ctx->conf.def = chanreq->oper; ctx->conf.ap = chanreq->ap; ctx->conf.rx_chains_static = 1; @@ -904,7 +921,6 @@ static int ieee80211_assign_link_chanctx(struct ieee80211_link_data *link, drv_unassign_vif_chanctx(local, sdata, link->conf, curr_ctx); conf = NULL; - list_del(&link->assigned_chanctx_list); } if (new_ctx) { @@ -919,9 +935,6 @@ static int ieee80211_assign_link_chanctx(struct ieee80211_link_data *link, /* succeeded, so commit it to the data structures */ conf = &new_ctx->conf; - if (!local->in_reconfig) - list_add(&link->assigned_chanctx_list, - &new_ctx->assigned_links); } } else { ret = 0; @@ -1108,7 +1121,6 @@ void ieee80211_link_unreserve_chanctx(struct ieee80211_link_data *link) if (WARN_ON(!ctx)) return; - list_del(&link->reserved_chanctx_list); link->reserved_chanctx = NULL; if (ieee80211_chanctx_refcount(sdata->local, ctx) == 0) { @@ -1142,9 +1154,9 @@ ieee80211_replace_chanctx(struct ieee80211_local *local, struct wiphy *wiphy = local->hw.wiphy; const struct wiphy_radio *radio; - if (!curr_ctx || (curr_ctx->replace_state == - IEEE80211_CHANCTX_WILL_BE_REPLACED) || - !list_empty(&curr_ctx->reserved_links)) { + if (!curr_ctx || + curr_ctx->replace_state == IEEE80211_CHANCTX_WILL_BE_REPLACED || + ieee80211_chanctx_num_reserved(local, curr_ctx) != 0) { /* * Another link already requested this context for a * reservation. Find another one hoping all links assigned @@ -1167,7 +1179,7 @@ ieee80211_replace_chanctx(struct ieee80211_local *local, IEEE80211_CHANCTX_REPLACE_NONE) continue; - if (!list_empty(&ctx->reserved_links)) + if (ieee80211_chanctx_num_reserved(local, ctx) != 0) continue; if (ctx->conf.radio_idx >= 0) { @@ -1185,9 +1197,9 @@ ieee80211_replace_chanctx(struct ieee80211_local *local, * If that's true then all available contexts already have reservations * and cannot be used. */ - if (!curr_ctx || (curr_ctx->replace_state == - IEEE80211_CHANCTX_WILL_BE_REPLACED) || - !list_empty(&curr_ctx->reserved_links)) + if (!curr_ctx || + curr_ctx->replace_state == IEEE80211_CHANCTX_WILL_BE_REPLACED || + ieee80211_chanctx_num_reserved(local, curr_ctx) != 0) return ERR_PTR(-EBUSY); new_ctx = ieee80211_alloc_chanctx(local, chanreq, mode, -1); @@ -1267,7 +1279,6 @@ int ieee80211_link_reserve_chanctx(struct ieee80211_link_data *link, return PTR_ERR(new_ctx); } - list_add(&link->reserved_chanctx_list, &new_ctx->reserved_links); link->reserved_chanctx = new_ctx; link->reserved = *chanreq; link->reserved_radar_required = radar_required; @@ -1381,7 +1392,6 @@ ieee80211_link_use_reserved_reassign(struct ieee80211_link_data *link) vif_chsw[0].new_ctx = &new_ctx->conf; vif_chsw[0].link_conf = link->conf; - list_del(&link->reserved_chanctx_list); link->reserved_chanctx = NULL; err = drv_switch_vif_chanctx(local, vif_chsw, 1, @@ -1394,7 +1404,6 @@ ieee80211_link_use_reserved_reassign(struct ieee80211_link_data *link) } link->radar_required = link->reserved_radar_required; - list_move(&link->assigned_chanctx_list, &new_ctx->assigned_links); rcu_assign_pointer(link_conf->chanctx_conf, &new_ctx->conf); if (sdata->vif.type == NL80211_IFTYPE_AP) @@ -1451,7 +1460,6 @@ ieee80211_link_use_reserved_assign(struct ieee80211_link_data *link) ieee80211_change_chanctx(local, new_ctx, new_ctx, chanreq); - list_del(&link->reserved_chanctx_list); link->reserved_chanctx = NULL; err = ieee80211_assign_link_chanctx(link, new_ctx, false); @@ -1517,8 +1525,10 @@ static int ieee80211_chsw_switch_vifs(struct ieee80211_local *local, goto out; } - list_for_each_entry(link, &ctx->reserved_links, - reserved_chanctx_list) { + for_each_sdata_link(local, link) { + if (link->reserved_chanctx != ctx) + continue; + if (!ieee80211_link_has_in_place_reservation(link)) continue; @@ -1551,7 +1561,7 @@ static int ieee80211_chsw_switch_ctxs(struct ieee80211_local *local) if (ctx->replace_state != IEEE80211_CHANCTX_REPLACES_OTHER) continue; - if (!list_empty(&ctx->replace_ctx->assigned_links)) + if (ieee80211_chanctx_num_assigned(local, ctx) != 0) continue; ieee80211_del_chanctx(local, ctx->replace_ctx, false); @@ -1568,7 +1578,7 @@ static int ieee80211_chsw_switch_ctxs(struct ieee80211_local *local) if (ctx->replace_state != IEEE80211_CHANCTX_REPLACES_OTHER) continue; - if (!list_empty(&ctx->replace_ctx->assigned_links)) + if (ieee80211_chanctx_num_assigned(local, ctx) != 0) continue; ieee80211_del_chanctx(local, ctx, false); @@ -1619,8 +1629,10 @@ static int ieee80211_vif_use_reserved_switch(struct ieee80211_local *local) n_reserved = 0; n_ready = 0; - list_for_each_entry(link, &ctx->replace_ctx->assigned_links, - assigned_chanctx_list) { + for_each_sdata_link(local, link) { + if (rcu_access_pointer(link->conf->chanctx_conf) != &ctx->conf) + continue; + n_assigned++; if (link->reserved_chanctx) { n_reserved++; @@ -1641,8 +1653,10 @@ static int ieee80211_vif_use_reserved_switch(struct ieee80211_local *local) } ctx->conf.radar_enabled = false; - list_for_each_entry(link, &ctx->reserved_links, - reserved_chanctx_list) { + for_each_sdata_link(local, link) { + if (link->reserved_chanctx != ctx) + continue; + if (ieee80211_link_has_in_place_reservation(link) && !link->reserved_ready) return -EAGAIN; @@ -1683,8 +1697,10 @@ static int ieee80211_vif_use_reserved_switch(struct ieee80211_local *local) goto err; } - list_for_each_entry(link, &ctx->reserved_links, - reserved_chanctx_list) { + for_each_sdata_link(local, link) { + if (link->reserved_chanctx != ctx) + continue; + if (!ieee80211_link_has_in_place_reservation(link)) continue; @@ -1718,7 +1734,7 @@ static int ieee80211_vif_use_reserved_switch(struct ieee80211_local *local) * context(s). */ list_for_each_entry(ctx, &local->chanctx_list, list) { - struct ieee80211_link_data *link, *link_tmp; + struct ieee80211_link_data *link; if (ctx->replace_state != IEEE80211_CHANCTX_REPLACES_OTHER) continue; @@ -1728,12 +1744,14 @@ static int ieee80211_vif_use_reserved_switch(struct ieee80211_local *local) goto err; } - list_for_each_entry(link, &ctx->reserved_links, - reserved_chanctx_list) { + for_each_sdata_link(local, link) { struct ieee80211_sub_if_data *sdata = link->sdata; struct ieee80211_bss_conf *link_conf = link->conf; u64 changed = 0; + if (link->reserved_chanctx != ctx) + continue; + if (!ieee80211_link_has_in_place_reservation(link)) continue; @@ -1765,14 +1783,13 @@ static int ieee80211_vif_use_reserved_switch(struct ieee80211_local *local) ieee80211_recalc_radar_chanctx(local, ctx); ieee80211_recalc_chanctx_min_def(local, ctx, NULL, false); - list_for_each_entry_safe(link, link_tmp, &ctx->reserved_links, - reserved_chanctx_list) { + for_each_sdata_link(local, link) { + if (link->reserved_chanctx != ctx) + continue; + if (ieee80211_link_get_chanctx(link) != ctx) continue; - list_del(&link->reserved_chanctx_list); - list_move(&link->assigned_chanctx_list, - &ctx->assigned_links); link->reserved_chanctx = NULL; ieee80211_link_chanctx_reservation_complete(link); @@ -1786,12 +1803,11 @@ static int ieee80211_vif_use_reserved_switch(struct ieee80211_local *local) * reservation for originally requested interface has already * succeeded at this point. */ - list_for_each_entry_safe(link, link_tmp, &ctx->reserved_links, - reserved_chanctx_list) { - if (WARN_ON(ieee80211_link_has_in_place_reservation(link))) + for_each_sdata_link(local, link) { + if (link->reserved_chanctx != ctx) continue; - if (WARN_ON(link->reserved_chanctx != ctx)) + if (WARN_ON(ieee80211_link_has_in_place_reservation(link))) continue; if (!link->reserved_ready) @@ -1834,13 +1850,15 @@ static int ieee80211_vif_use_reserved_switch(struct ieee80211_local *local) err: list_for_each_entry(ctx, &local->chanctx_list, list) { - struct ieee80211_link_data *link, *link_tmp; + struct ieee80211_link_data *link; if (ctx->replace_state != IEEE80211_CHANCTX_REPLACES_OTHER) continue; - list_for_each_entry_safe(link, link_tmp, &ctx->reserved_links, - reserved_chanctx_list) { + for_each_sdata_link(local, link) { + if (link->reserved_chanctx != ctx) + continue; + ieee80211_link_unreserve_chanctx(link); ieee80211_link_chanctx_reservation_complete(link); } @@ -1949,7 +1967,6 @@ int _ieee80211_link_use_channel(struct ieee80211_link_data *link, /* remove reservation */ WARN_ON(link->reserved_chanctx != ctx); link->reserved_chanctx = NULL; - list_del(&link->reserved_chanctx_list); } if (ret) { diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index 72e4f326946f..c94156f20995 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -916,9 +916,6 @@ struct ieee80211_chanctx { struct list_head list; struct rcu_head rcu_head; - struct list_head assigned_links; - struct list_head reserved_links; - enum ieee80211_chanctx_replace_state replace_state; struct ieee80211_chanctx *replace_ctx; @@ -1071,9 +1068,6 @@ struct ieee80211_link_data { struct ieee80211_sub_if_data *sdata; unsigned int link_id; - struct list_head assigned_chanctx_list; /* protected by wiphy mutex */ - struct list_head reserved_chanctx_list; /* protected by wiphy mutex */ - /* multicast keys only */ struct ieee80211_key __rcu *gtk[NUM_DEFAULT_KEYS + NUM_DEFAULT_MGMT_KEYS + diff --git a/net/mac80211/link.c b/net/mac80211/link.c index 4a19b765ccb6..85ed3cbb2e36 100644 --- a/net/mac80211/link.c +++ b/net/mac80211/link.c @@ -119,8 +119,6 @@ void ieee80211_link_init(struct ieee80211_sub_if_data *sdata, ieee80211_color_change_finalize_work); wiphy_delayed_work_init(&link->color_collision_detect_work, ieee80211_color_collision_detection_work); - INIT_LIST_HEAD(&link->assigned_chanctx_list); - INIT_LIST_HEAD(&link->reserved_chanctx_list); wiphy_delayed_work_init(&link->dfs_cac_timer_work, ieee80211_dfs_cac_timer_work); diff --git a/net/mac80211/util.c b/net/mac80211/util.c index c9931537f9d2..324d84d11cab 100644 --- a/net/mac80211/util.c +++ b/net/mac80211/util.c @@ -4016,23 +4016,23 @@ static u8 ieee80211_chanctx_radar_detect(struct ieee80211_local *local, if (WARN_ON(ctx->replace_state == IEEE80211_CHANCTX_WILL_BE_REPLACED)) return 0; - list_for_each_entry(link, &ctx->reserved_links, reserved_chanctx_list) - if (link->reserved_radar_required) + for_each_sdata_link(local, link) { + if (rcu_access_pointer(link->conf->chanctx_conf) == &ctx->conf) { + /* + * An in-place reservation context should not have any + * assigned links until it replaces the other context. + */ + WARN_ON(ctx->replace_state == + IEEE80211_CHANCTX_REPLACES_OTHER); + + if (link->radar_required) + radar_detect |= + BIT(link->conf->chanreq.oper.width); + } + + if (link->reserved_chanctx == ctx && + link->reserved_radar_required) radar_detect |= BIT(link->reserved.oper.width); - - /* - * An in-place reservation context should not have any assigned vifs - * until it replaces the other context. - */ - WARN_ON(ctx->replace_state == IEEE80211_CHANCTX_REPLACES_OTHER && - !list_empty(&ctx->assigned_links)); - - list_for_each_entry(link, &ctx->assigned_links, assigned_chanctx_list) { - if (!link->radar_required) - continue; - - radar_detect |= - BIT(link->conf->chanreq.oper.width); } return radar_detect; From 52363af3a9c989e8d8d6f8a2343a7f669e53a57f Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 5 Nov 2025 16:03:44 +0100 Subject: [PATCH 0645/1381] wifi: mac80211: simplify ieee80211_recalc_chanctx_min_def() API ieee80211_recalc_chanctx_min_def() is used outside the chandef code, but then should/is always used with NULL/false for the two last arguments. Remove them, and create another level of indirection for use inside the file. Link: https://patch.msgid.link/20251105160431.33408844c392.I4f52298861780c17a27cd229609e8a3e29c8d740@changeid Signed-off-by: Johannes Berg --- net/mac80211/chan.c | 46 ++++++++++++++++++++++---------------- net/mac80211/he.c | 6 ++--- net/mac80211/ieee80211_i.h | 4 +--- net/mac80211/util.c | 2 +- 4 files changed, 32 insertions(+), 26 deletions(-) diff --git a/net/mac80211/chan.c b/net/mac80211/chan.c index 944e1c18501a..dda11b519252 100644 --- a/net/mac80211/chan.c +++ b/net/mac80211/chan.c @@ -410,10 +410,10 @@ ieee80211_get_chanctx_max_required_bw(struct ieee80211_local *local, * channel context. */ static u32 -_ieee80211_recalc_chanctx_min_def(struct ieee80211_local *local, - struct ieee80211_chanctx *ctx, - struct ieee80211_link_data *rsvd_for, - bool check_reserved) +__ieee80211_recalc_chanctx_min_def(struct ieee80211_local *local, + struct ieee80211_chanctx *ctx, + struct ieee80211_link_data *rsvd_for, + bool check_reserved) { enum nl80211_chan_width max_bw; struct cfg80211_chan_def min_def; @@ -518,13 +518,14 @@ static void ieee80211_chan_bw_change(struct ieee80211_local *local, * the max of min required widths of all the interfaces bound to this * channel context. */ -void ieee80211_recalc_chanctx_min_def(struct ieee80211_local *local, - struct ieee80211_chanctx *ctx, - struct ieee80211_link_data *rsvd_for, - bool check_reserved) +static void +_ieee80211_recalc_chanctx_min_def(struct ieee80211_local *local, + struct ieee80211_chanctx *ctx, + struct ieee80211_link_data *rsvd_for, + bool check_reserved) { - u32 changed = _ieee80211_recalc_chanctx_min_def(local, ctx, rsvd_for, - check_reserved); + u32 changed = __ieee80211_recalc_chanctx_min_def(local, ctx, rsvd_for, + check_reserved); if (!changed) return; @@ -538,6 +539,12 @@ void ieee80211_recalc_chanctx_min_def(struct ieee80211_local *local, ieee80211_chan_bw_change(local, ctx, false, false); } +void ieee80211_recalc_chanctx_min_def(struct ieee80211_local *local, + struct ieee80211_chanctx *ctx) +{ + _ieee80211_recalc_chanctx_min_def(local, ctx, NULL, false); +} + static void _ieee80211_change_chanctx(struct ieee80211_local *local, struct ieee80211_chanctx *ctx, struct ieee80211_chanctx *old_ctx, @@ -572,7 +579,7 @@ static void _ieee80211_change_chanctx(struct ieee80211_local *local, ieee80211_chan_bw_change(local, old_ctx, false, true); if (ieee80211_chanreq_identical(&ctx_req, chanreq)) { - ieee80211_recalc_chanctx_min_def(local, ctx, rsvd_for, false); + _ieee80211_recalc_chanctx_min_def(local, ctx, rsvd_for, false); return; } @@ -593,7 +600,8 @@ static void _ieee80211_change_chanctx(struct ieee80211_local *local, ctx->conf.ap = chanreq->ap; /* check if min chanctx also changed */ - changed |= _ieee80211_recalc_chanctx_min_def(local, ctx, rsvd_for, false); + changed |= __ieee80211_recalc_chanctx_min_def(local, ctx, rsvd_for, + false); ieee80211_add_wbrf(local, &ctx->conf.def); @@ -732,7 +740,7 @@ ieee80211_alloc_chanctx(struct ieee80211_local *local, ctx->conf.radar_enabled = false; ctx->conf.radio_idx = radio_idx; ctx->radar_detected = false; - _ieee80211_recalc_chanctx_min_def(local, ctx, NULL, false); + __ieee80211_recalc_chanctx_min_def(local, ctx, NULL, false); return ctx; } @@ -925,7 +933,7 @@ static int ieee80211_assign_link_chanctx(struct ieee80211_link_data *link, if (new_ctx) { /* recalc considering the link we'll use it for now */ - ieee80211_recalc_chanctx_min_def(local, new_ctx, link, false); + _ieee80211_recalc_chanctx_min_def(local, new_ctx, link, false); ret = drv_assign_vif_chanctx(local, sdata, link->conf, new_ctx); if (assign_on_failure || !ret) { @@ -946,12 +954,12 @@ static int ieee80211_assign_link_chanctx(struct ieee80211_link_data *link, ieee80211_recalc_chanctx_chantype(local, curr_ctx); ieee80211_recalc_smps_chanctx(local, curr_ctx); ieee80211_recalc_radar_chanctx(local, curr_ctx); - ieee80211_recalc_chanctx_min_def(local, curr_ctx, NULL, false); + ieee80211_recalc_chanctx_min_def(local, curr_ctx); } if (new_ctx && ieee80211_chanctx_num_assigned(local, new_ctx) > 0) { ieee80211_recalc_txpower(link, false); - ieee80211_recalc_chanctx_min_def(local, new_ctx, NULL, false); + ieee80211_recalc_chanctx_min_def(local, new_ctx); } if (conf) { @@ -1414,7 +1422,7 @@ ieee80211_link_use_reserved_reassign(struct ieee80211_link_data *link) if (ieee80211_chanctx_refcount(local, old_ctx) == 0) ieee80211_free_chanctx(local, old_ctx, false); - ieee80211_recalc_chanctx_min_def(local, new_ctx, NULL, false); + ieee80211_recalc_chanctx_min_def(local, new_ctx); ieee80211_recalc_smps_chanctx(local, new_ctx); ieee80211_recalc_radar_chanctx(local, new_ctx); @@ -1709,7 +1717,7 @@ static int ieee80211_vif_use_reserved_switch(struct ieee80211_local *local) true, true); } - ieee80211_recalc_chanctx_min_def(local, ctx, NULL, true); + _ieee80211_recalc_chanctx_min_def(local, ctx, NULL, true); } /* @@ -1781,7 +1789,7 @@ static int ieee80211_vif_use_reserved_switch(struct ieee80211_local *local) ieee80211_recalc_chanctx_chantype(local, ctx); ieee80211_recalc_smps_chanctx(local, ctx); ieee80211_recalc_radar_chanctx(local, ctx); - ieee80211_recalc_chanctx_min_def(local, ctx, NULL, false); + ieee80211_recalc_chanctx_min_def(local, ctx); for_each_sdata_link(local, link) { if (link->reserved_chanctx != ctx) diff --git a/net/mac80211/he.c b/net/mac80211/he.c index 5792ef77e986..f7b05e59374c 100644 --- a/net/mac80211/he.c +++ b/net/mac80211/he.c @@ -3,7 +3,7 @@ * HE handling * * Copyright(c) 2017 Intel Deutschland GmbH - * Copyright(c) 2019 - 2024 Intel Corporation + * Copyright(c) 2019-2025 Intel Corporation */ #include "ieee80211_i.h" @@ -313,7 +313,7 @@ bool ieee80211_prepare_rx_omi_bw(struct ieee80211_link_sta *pub_link_sta, ieee80211_link_sta_rc_update_omi(link, link_sta); } else { link_sta->rx_omi_bw_rx = bw; - ieee80211_recalc_chanctx_min_def(local, chanctx, NULL, false); + ieee80211_recalc_chanctx_min_def(local, chanctx); } link_sta->rx_omi_bw_staging = bw; @@ -359,7 +359,7 @@ void ieee80211_finalize_rx_omi_bw(struct ieee80211_link_sta *pub_link_sta) /* channel context in finalize only when narrowing bandwidth */ WARN_ON(link_sta->rx_omi_bw_rx < link_sta->rx_omi_bw_staging); link_sta->rx_omi_bw_rx = link_sta->rx_omi_bw_staging; - ieee80211_recalc_chanctx_min_def(local, chanctx, NULL, false); + ieee80211_recalc_chanctx_min_def(local, chanctx); } trace_api_return_void(local); diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index c94156f20995..898ccbc4ec64 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -2768,9 +2768,7 @@ int ieee80211_chanctx_refcount(struct ieee80211_local *local, void ieee80211_recalc_smps_chanctx(struct ieee80211_local *local, struct ieee80211_chanctx *chanctx); void ieee80211_recalc_chanctx_min_def(struct ieee80211_local *local, - struct ieee80211_chanctx *ctx, - struct ieee80211_link_data *rsvd_for, - bool check_reserved); + struct ieee80211_chanctx *ctx); bool ieee80211_is_radar_required(struct ieee80211_local *local, struct cfg80211_scan_request *req); bool ieee80211_is_radio_idx_in_scan_req(struct wiphy *wiphy, diff --git a/net/mac80211/util.c b/net/mac80211/util.c index 324d84d11cab..1a128f7aae82 100644 --- a/net/mac80211/util.c +++ b/net/mac80211/util.c @@ -2347,7 +2347,7 @@ void ieee80211_recalc_min_chandef(struct ieee80211_sub_if_data *sdata, chanctx = container_of(chanctx_conf, struct ieee80211_chanctx, conf); - ieee80211_recalc_chanctx_min_def(local, chanctx, NULL, false); + ieee80211_recalc_chanctx_min_def(local, chanctx); } } From 1ce954c98b8968e66a304c8586122d706bae97c3 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 5 Nov 2025 16:03:45 +0100 Subject: [PATCH 0646/1381] wifi: mac80211: add and use chanctx usage iteration In preparation for NAN interfaces using multiple channel contexts, add an iterator macro that iterates all users of a given channel context. The logic during reserved assign/reassign handling the bandwidth in ieee80211_get_chanctx_max_required_bw() is a bit more complicated and should be cleaned up, so it isn't yet converted. Link: https://patch.msgid.link/20251105160431.5aaccc2f127d.I2b7fd0858a263916f43abab49c6d3cc0b5aa16ec@changeid Signed-off-by: Johannes Berg --- net/mac80211/chan.c | 342 +++++++++++++++++++++++++------------------- 1 file changed, 195 insertions(+), 147 deletions(-) diff --git a/net/mac80211/chan.c b/net/mac80211/chan.c index dda11b519252..6aa305839f53 100644 --- a/net/mac80211/chan.c +++ b/net/mac80211/chan.c @@ -12,18 +12,132 @@ #include "driver-ops.h" #include "rate.h" +struct ieee80211_chanctx_user_iter { + struct ieee80211_chan_req *chanreq; + struct ieee80211_sub_if_data *sdata; + struct ieee80211_link_data *link; + enum nl80211_iftype iftype; + bool reserved, radar_required, done; + enum { + CHANCTX_ITER_POS_ASSIGNED, + CHANCTX_ITER_POS_RESERVED, + CHANCTX_ITER_POS_DONE, + } per_link; +}; + +enum ieee80211_chanctx_iter_type { + CHANCTX_ITER_ALL, + CHANCTX_ITER_RESERVED, + CHANCTX_ITER_ASSIGNED, +}; + +static void ieee80211_chanctx_user_iter_next(struct ieee80211_local *local, + struct ieee80211_chanctx *ctx, + struct ieee80211_chanctx_user_iter *iter, + enum ieee80211_chanctx_iter_type type, + bool start) +{ + lockdep_assert_wiphy(local->hw.wiphy); + + if (start) { + memset(iter, 0, sizeof(*iter)); + goto next_interface; + } + +next_link: + for (int link_id = iter->link ? iter->link->link_id : 0; + link_id < ARRAY_SIZE(iter->sdata->link); + link_id++) { + struct ieee80211_link_data *link; + + link = sdata_dereference(iter->sdata->link[link_id], + iter->sdata); + if (!link) + continue; + + switch (iter->per_link) { + case CHANCTX_ITER_POS_ASSIGNED: + iter->per_link = CHANCTX_ITER_POS_RESERVED; + if (type != CHANCTX_ITER_RESERVED && + rcu_access_pointer(link->conf->chanctx_conf) == &ctx->conf) { + iter->link = link; + iter->reserved = false; + iter->radar_required = link->radar_required; + iter->chanreq = &link->conf->chanreq; + return; + } + fallthrough; + case CHANCTX_ITER_POS_RESERVED: + iter->per_link = CHANCTX_ITER_POS_DONE; + if (type != CHANCTX_ITER_ASSIGNED && + link->reserved_chanctx == ctx) { + iter->link = link; + iter->reserved = true; + iter->radar_required = + link->reserved_radar_required; + + iter->chanreq = &link->reserved; + return; + } + fallthrough; + case CHANCTX_ITER_POS_DONE: + iter->per_link = CHANCTX_ITER_POS_ASSIGNED; + continue; + } + } + +next_interface: + /* next (or first) interface */ + iter->sdata = list_prepare_entry(iter->sdata, &local->interfaces, list); + list_for_each_entry_continue(iter->sdata, &local->interfaces, list) { + /* AP_VLAN has a chanctx pointer but follows AP */ + if (iter->sdata->vif.type == NL80211_IFTYPE_AP_VLAN) + continue; + + iter->link = NULL; + iter->per_link = CHANCTX_ITER_POS_ASSIGNED; + iter->iftype = iter->sdata->vif.type; + goto next_link; + } + + iter->done = true; +} + +#define for_each_chanctx_user_assigned(local, ctx, iter) \ + for (ieee80211_chanctx_user_iter_next(local, ctx, iter, \ + CHANCTX_ITER_ASSIGNED, \ + true); \ + !((iter)->done); \ + ieee80211_chanctx_user_iter_next(local, ctx, iter, \ + CHANCTX_ITER_ASSIGNED, \ + false)) + +#define for_each_chanctx_user_reserved(local, ctx, iter) \ + for (ieee80211_chanctx_user_iter_next(local, ctx, iter, \ + CHANCTX_ITER_RESERVED, \ + true); \ + !((iter)->done); \ + ieee80211_chanctx_user_iter_next(local, ctx, iter, \ + CHANCTX_ITER_RESERVED, \ + false)) + +#define for_each_chanctx_user_all(local, ctx, iter) \ + for (ieee80211_chanctx_user_iter_next(local, ctx, iter, \ + CHANCTX_ITER_ALL, \ + true); \ + !((iter)->done); \ + ieee80211_chanctx_user_iter_next(local, ctx, iter, \ + CHANCTX_ITER_ALL, \ + false)) + static int ieee80211_chanctx_num_assigned(struct ieee80211_local *local, struct ieee80211_chanctx *ctx) { - struct ieee80211_link_data *link; + struct ieee80211_chanctx_user_iter iter; int num = 0; - lockdep_assert_wiphy(local->hw.wiphy); - - for_each_sdata_link(local, link) { - if (rcu_access_pointer(link->conf->chanctx_conf) == &ctx->conf) - num++; - } + for_each_chanctx_user_assigned(local, ctx, &iter) + num++; return num; } @@ -31,15 +145,11 @@ static int ieee80211_chanctx_num_assigned(struct ieee80211_local *local, static int ieee80211_chanctx_num_reserved(struct ieee80211_local *local, struct ieee80211_chanctx *ctx) { - struct ieee80211_link_data *link; + struct ieee80211_chanctx_user_iter iter; int num = 0; - lockdep_assert_wiphy(local->hw.wiphy); - - for_each_sdata_link(local, link) { - if (link->reserved_chanctx == ctx) - num++; - } + for_each_chanctx_user_reserved(local, ctx, &iter) + num++; return num; } @@ -47,17 +157,11 @@ static int ieee80211_chanctx_num_reserved(struct ieee80211_local *local, int ieee80211_chanctx_refcount(struct ieee80211_local *local, struct ieee80211_chanctx *ctx) { - struct ieee80211_link_data *link; + struct ieee80211_chanctx_user_iter iter; int num = 0; - lockdep_assert_wiphy(local->hw.wiphy); - - for_each_sdata_link(local, link) { - if (rcu_access_pointer(link->conf->chanctx_conf) == &ctx->conf) - num++; - if (link->reserved_chanctx == ctx) - num++; - } + for_each_chanctx_user_all(local, ctx, &iter) + num++; return num; } @@ -158,18 +262,15 @@ ieee80211_chanctx_reserved_chanreq(struct ieee80211_local *local, const struct ieee80211_chan_req *req, struct ieee80211_chan_req *tmp) { - struct ieee80211_link_data *link; + struct ieee80211_chanctx_user_iter iter; lockdep_assert_wiphy(local->hw.wiphy); if (WARN_ON(!req)) return NULL; - for_each_sdata_link(local, link) { - if (link->reserved_chanctx != ctx) - continue; - - req = ieee80211_chanreq_compatible(&link->reserved, req, tmp); + for_each_chanctx_user_reserved(local, ctx, &iter) { + req = ieee80211_chanreq_compatible(iter.chanreq, req, tmp); if (!req) break; } @@ -183,21 +284,16 @@ ieee80211_chanctx_non_reserved_chandef(struct ieee80211_local *local, const struct ieee80211_chan_req *compat, struct ieee80211_chan_req *tmp) { - struct ieee80211_link_data *link; const struct ieee80211_chan_req *comp_def = compat; + struct ieee80211_chanctx_user_iter iter; lockdep_assert_wiphy(local->hw.wiphy); - for_each_sdata_link(local, link) { - struct ieee80211_bss_conf *link_conf = link->conf; - - if (rcu_access_pointer(link_conf->chanctx_conf) != &ctx->conf) + for_each_chanctx_user_assigned(local, ctx, &iter) { + if (iter.link->reserved_chanctx) continue; - if (link->reserved_chanctx) - continue; - - comp_def = ieee80211_chanreq_compatible(&link_conf->chanreq, + comp_def = ieee80211_chanreq_compatible(iter.chanreq, comp_def, tmp); if (!comp_def) break; @@ -702,17 +798,13 @@ static bool ieee80211_chanctx_radar_required(struct ieee80211_local *local, struct ieee80211_chanctx *ctx) { - struct ieee80211_chanctx_conf *conf = &ctx->conf; - struct ieee80211_link_data *link; + struct ieee80211_chanctx_user_iter iter; lockdep_assert_wiphy(local->hw.wiphy); - for_each_sdata_link(local, link) { - if (rcu_access_pointer(link->conf->chanctx_conf) != conf) - continue; - if (!link->radar_required) - continue; - return true; + for_each_chanctx_user_assigned(local, ctx, &iter) { + if (iter.radar_required) + return true; } return false; @@ -829,27 +921,17 @@ void ieee80211_recalc_chanctx_chantype(struct ieee80211_local *local, { struct ieee80211_chanctx_conf *conf = &ctx->conf; const struct ieee80211_chan_req *compat = NULL; - struct ieee80211_link_data *link; + struct ieee80211_chanctx_user_iter iter; struct ieee80211_chan_req tmp; struct sta_info *sta; lockdep_assert_wiphy(local->hw.wiphy); - for_each_sdata_link(local, link) { - struct ieee80211_bss_conf *link_conf; - - if (link->sdata->vif.type == NL80211_IFTYPE_AP_VLAN) - continue; - - link_conf = link->conf; - - if (rcu_access_pointer(link_conf->chanctx_conf) != conf) - continue; - + for_each_chanctx_user_assigned(local, ctx, &iter) { if (!compat) - compat = &link_conf->chanreq; + compat = iter.chanreq; - compat = ieee80211_chanreq_compatible(&link_conf->chanreq, + compat = ieee80211_chanreq_compatible(iter.chanreq, compat, &tmp); if (WARN_ON_ONCE(!compat)) return; @@ -862,6 +944,7 @@ void ieee80211_recalc_chanctx_chantype(struct ieee80211_local *local, list_for_each_entry(sta, &local->sta_list, list) { struct ieee80211_sub_if_data *sdata = sta->sdata; struct ieee80211_chan_req tdls_chanreq = {}; + struct ieee80211_link_data *link; int tdls_link_id; if (!sta->uploaded || @@ -992,21 +1075,21 @@ static int ieee80211_assign_link_chanctx(struct ieee80211_link_data *link, void ieee80211_recalc_smps_chanctx(struct ieee80211_local *local, struct ieee80211_chanctx *chanctx) { + struct ieee80211_chanctx_user_iter iter; struct ieee80211_sub_if_data *sdata; u8 rx_chains_static, rx_chains_dynamic; - struct ieee80211_link_data *link; lockdep_assert_wiphy(local->hw.wiphy); rx_chains_static = 1; rx_chains_dynamic = 1; - for_each_sdata_link(local, link) { + for_each_chanctx_user_assigned(local, chanctx, &iter) { u8 needed_static, needed_dynamic; - switch (link->sdata->vif.type) { + switch (iter.iftype) { case NL80211_IFTYPE_STATION: - if (!link->sdata->u.mgd.associated) + if (!iter.sdata->u.mgd.associated) continue; break; case NL80211_IFTYPE_MONITOR: @@ -1022,26 +1105,23 @@ void ieee80211_recalc_smps_chanctx(struct ieee80211_local *local, continue; } - if (rcu_access_pointer(link->conf->chanctx_conf) != &chanctx->conf) - continue; - - if (link->sdata->vif.type == NL80211_IFTYPE_MONITOR) { + if (iter.iftype == NL80211_IFTYPE_MONITOR) { rx_chains_dynamic = rx_chains_static = local->rx_chains; break; } - switch (link->smps_mode) { + switch (iter.link->smps_mode) { default: WARN_ONCE(1, "Invalid SMPS mode %d\n", - link->smps_mode); + iter.link->smps_mode); fallthrough; case IEEE80211_SMPS_OFF: - needed_static = link->needed_rx_chains; - needed_dynamic = link->needed_rx_chains; + needed_static = iter.link->needed_rx_chains; + needed_dynamic = iter.link->needed_rx_chains; break; case IEEE80211_SMPS_DYNAMIC: needed_static = 1; - needed_dynamic = link->needed_rx_chains; + needed_dynamic = iter.link->needed_rx_chains; break; case IEEE80211_SMPS_STATIC: needed_static = 1; @@ -1513,7 +1593,6 @@ static int ieee80211_chsw_switch_vifs(struct ieee80211_local *local, int n_vifs) { struct ieee80211_vif_chanctx_switch *vif_chsw; - struct ieee80211_link_data *link; struct ieee80211_chanctx *ctx, *old_ctx; int i, err; @@ -1525,6 +1604,8 @@ static int ieee80211_chsw_switch_vifs(struct ieee80211_local *local, i = 0; list_for_each_entry(ctx, &local->chanctx_list, list) { + struct ieee80211_chanctx_user_iter iter; + if (ctx->replace_state != IEEE80211_CHANCTX_REPLACES_OTHER) continue; @@ -1533,18 +1614,15 @@ static int ieee80211_chsw_switch_vifs(struct ieee80211_local *local, goto out; } - for_each_sdata_link(local, link) { - if (link->reserved_chanctx != ctx) + for_each_chanctx_user_reserved(local, ctx, &iter) { + if (!ieee80211_link_has_in_place_reservation(iter.link)) continue; - if (!ieee80211_link_has_in_place_reservation(link)) - continue; - - old_ctx = ieee80211_link_get_chanctx(link); - vif_chsw[i].vif = &link->sdata->vif; + old_ctx = ieee80211_link_get_chanctx(iter.link); + vif_chsw[i].vif = &iter.sdata->vif; vif_chsw[i].old_ctx = &old_ctx->conf; vif_chsw[i].new_ctx = &ctx->conf; - vif_chsw[i].link_conf = link->conf; + vif_chsw[i].link_conf = iter.link->conf; i++; } @@ -1621,7 +1699,7 @@ static int ieee80211_vif_use_reserved_switch(struct ieee80211_local *local) */ list_for_each_entry(ctx, &local->chanctx_list, list) { - struct ieee80211_link_data *link; + struct ieee80211_chanctx_user_iter iter; if (ctx->replace_state != IEEE80211_CHANCTX_REPLACES_OTHER) continue; @@ -1637,14 +1715,11 @@ static int ieee80211_vif_use_reserved_switch(struct ieee80211_local *local) n_reserved = 0; n_ready = 0; - for_each_sdata_link(local, link) { - if (rcu_access_pointer(link->conf->chanctx_conf) != &ctx->conf) - continue; - + for_each_chanctx_user_assigned(local, ctx, &iter) { n_assigned++; - if (link->reserved_chanctx) { + if (iter.link->reserved_chanctx) { n_reserved++; - if (link->reserved_ready) + if (iter.link->reserved_ready) n_ready++; } } @@ -1661,15 +1736,12 @@ static int ieee80211_vif_use_reserved_switch(struct ieee80211_local *local) } ctx->conf.radar_enabled = false; - for_each_sdata_link(local, link) { - if (link->reserved_chanctx != ctx) - continue; - - if (ieee80211_link_has_in_place_reservation(link) && - !link->reserved_ready) + for_each_chanctx_user_reserved(local, ctx, &iter) { + if (ieee80211_link_has_in_place_reservation(iter.link) && + !iter.link->reserved_ready) return -EAGAIN; - old_ctx = ieee80211_link_get_chanctx(link); + old_ctx = ieee80211_link_get_chanctx(iter.link); if (old_ctx) { if (old_ctx->replace_state == IEEE80211_CHANCTX_WILL_BE_REPLACED) @@ -1680,7 +1752,7 @@ static int ieee80211_vif_use_reserved_switch(struct ieee80211_local *local) n_vifs_ctxless++; } - if (link->reserved_radar_required) + if (iter.radar_required) ctx->conf.radar_enabled = true; } } @@ -1695,7 +1767,7 @@ static int ieee80211_vif_use_reserved_switch(struct ieee80211_local *local) /* update station rate control and min width before switch */ list_for_each_entry(ctx, &local->chanctx_list, list) { - struct ieee80211_link_data *link; + struct ieee80211_chanctx_user_iter iter; if (ctx->replace_state != IEEE80211_CHANCTX_REPLACES_OTHER) continue; @@ -1705,15 +1777,12 @@ static int ieee80211_vif_use_reserved_switch(struct ieee80211_local *local) goto err; } - for_each_sdata_link(local, link) { - if (link->reserved_chanctx != ctx) - continue; - - if (!ieee80211_link_has_in_place_reservation(link)) + for_each_chanctx_user_reserved(local, ctx, &iter) { + if (!ieee80211_link_has_in_place_reservation(iter.link)) continue; ieee80211_chan_bw_change(local, - ieee80211_link_get_chanctx(link), + ieee80211_link_get_chanctx(iter.link), true, true); } @@ -1742,7 +1811,7 @@ static int ieee80211_vif_use_reserved_switch(struct ieee80211_local *local) * context(s). */ list_for_each_entry(ctx, &local->chanctx_list, list) { - struct ieee80211_link_data *link; + struct ieee80211_chanctx_user_iter iter; if (ctx->replace_state != IEEE80211_CHANCTX_REPLACES_OTHER) continue; @@ -1752,14 +1821,12 @@ static int ieee80211_vif_use_reserved_switch(struct ieee80211_local *local) goto err; } - for_each_sdata_link(local, link) { - struct ieee80211_sub_if_data *sdata = link->sdata; + for_each_chanctx_user_reserved(local, ctx, &iter) { + struct ieee80211_link_data *link = iter.link; + struct ieee80211_sub_if_data *sdata = iter.sdata; struct ieee80211_bss_conf *link_conf = link->conf; u64 changed = 0; - if (link->reserved_chanctx != ctx) - continue; - if (!ieee80211_link_has_in_place_reservation(link)) continue; @@ -1772,9 +1839,9 @@ static int ieee80211_vif_use_reserved_switch(struct ieee80211_local *local) ieee80211_check_fast_xmit_iface(sdata); - link->radar_required = link->reserved_radar_required; + link->radar_required = iter.radar_required; - if (link_conf->chanreq.oper.width != link->reserved.oper.width) + if (link_conf->chanreq.oper.width != iter.chanreq->oper.width) changed = BSS_CHANGED_BANDWIDTH; ieee80211_link_update_chanreq(link, &link->reserved); @@ -1791,16 +1858,13 @@ static int ieee80211_vif_use_reserved_switch(struct ieee80211_local *local) ieee80211_recalc_radar_chanctx(local, ctx); ieee80211_recalc_chanctx_min_def(local, ctx); - for_each_sdata_link(local, link) { - if (link->reserved_chanctx != ctx) + for_each_chanctx_user_reserved(local, ctx, &iter) { + if (ieee80211_link_get_chanctx(iter.link) != ctx) continue; - if (ieee80211_link_get_chanctx(link) != ctx) - continue; + iter.link->reserved_chanctx = NULL; - link->reserved_chanctx = NULL; - - ieee80211_link_chanctx_reservation_complete(link); + ieee80211_link_chanctx_reservation_complete(iter.link); ieee80211_chan_bw_change(local, ctx, false, false); } @@ -1811,9 +1875,8 @@ static int ieee80211_vif_use_reserved_switch(struct ieee80211_local *local) * reservation for originally requested interface has already * succeeded at this point. */ - for_each_sdata_link(local, link) { - if (link->reserved_chanctx != ctx) - continue; + for_each_chanctx_user_reserved(local, ctx, &iter) { + struct ieee80211_link_data *link = iter.link; if (WARN_ON(ieee80211_link_has_in_place_reservation(link))) continue; @@ -1858,17 +1921,14 @@ static int ieee80211_vif_use_reserved_switch(struct ieee80211_local *local) err: list_for_each_entry(ctx, &local->chanctx_list, list) { - struct ieee80211_link_data *link; + struct ieee80211_chanctx_user_iter iter; if (ctx->replace_state != IEEE80211_CHANCTX_REPLACES_OTHER) continue; - for_each_sdata_link(local, link) { - if (link->reserved_chanctx != ctx) - continue; - - ieee80211_link_unreserve_chanctx(link); - ieee80211_link_chanctx_reservation_complete(link); + for_each_chanctx_user_reserved(local, ctx, &iter) { + ieee80211_link_unreserve_chanctx(iter.link); + ieee80211_link_chanctx_reservation_complete(iter.link); } } @@ -2071,29 +2131,17 @@ ieee80211_chanctx_recheck(struct ieee80211_local *local, struct ieee80211_chan_req *tmp) { const struct ieee80211_chan_req *ret = req; - struct ieee80211_link_data *link; + struct ieee80211_chanctx_user_iter iter; lockdep_assert_wiphy(local->hw.wiphy); - for_each_sdata_link(local, link) { - if (link == skip_link) + for_each_chanctx_user_all(local, ctx, &iter) { + if (iter.link == skip_link) continue; - if (rcu_access_pointer(link->conf->chanctx_conf) == &ctx->conf) { - ret = ieee80211_chanreq_compatible(ret, - &link->conf->chanreq, - tmp); - if (!ret) - return NULL; - } - - if (link->reserved_chanctx == ctx) { - ret = ieee80211_chanreq_compatible(ret, - &link->reserved, - tmp); - if (!ret) - return NULL; - } + ret = ieee80211_chanreq_compatible(ret, iter.chanreq, tmp); + if (!ret) + return NULL; } *tmp = *ret; From 1fba15768585ef75dbf86b1c75533dd5b81e247a Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 5 Nov 2025 16:05:38 +0100 Subject: [PATCH 0647/1381] wifi: mac80211: remove "disabling VHT" message This message always occurs with any AP that is simply HT/11n, remove it. Link: https://patch.msgid.link/20251105160538.f8b286cfba08.Ib91b5b09a3fff753341b0f7cb0df7f6913a4abe6@changeid Signed-off-by: Johannes Berg --- net/mac80211/mlme.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c index 767804e41a34..2ee9eae89d05 100644 --- a/net/mac80211/mlme.c +++ b/net/mac80211/mlme.c @@ -276,11 +276,8 @@ ieee80211_determine_ap_chan(struct ieee80211_sub_if_data *sdata, return IEEE80211_CONN_MODE_VHT; } } else if (!vht_oper || !elems->vht_cap_elem) { - if (sband->band == NL80211_BAND_5GHZ) { - sdata_info(sdata, - "VHT information is missing, disabling VHT\n"); + if (sband->band == NL80211_BAND_5GHZ) return IEEE80211_CONN_MODE_HT; - } no_vht = true; } else if (sband->band == NL80211_BAND_2GHZ) { no_vht = true; From 68eb1b791ac8da7c3d03967143f1417e2978bf5e Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 5 Nov 2025 16:08:10 +0100 Subject: [PATCH 0648/1381] wifi: mac80211: pass frame type to element parsing This will be needed for UHR operation parsing, and we already pass whether or not the frame is an action frame, replace that by the full type. Note this fixes a few cases where 'false' was erroneously passed (mesh and TDLS) and removes ieee802_11_parse_elems_crc() as it's unused. Link: https://patch.msgid.link/20251105160810.a476d20a6e01.Ie659535f9357f2f9a3c73f8c059ccfc96bf93b54@changeid Signed-off-by: Johannes Berg --- include/linux/ieee80211.h | 1 + net/mac80211/agg-rx.c | 7 ++++-- net/mac80211/ibss.c | 14 +++++++----- net/mac80211/ieee80211_i.h | 21 +++++------------- net/mac80211/mesh.c | 26 +++++++++++++--------- net/mac80211/mesh_hwmp.c | 7 ++++-- net/mac80211/mesh_plink.c | 7 ++++-- net/mac80211/mlme.c | 45 ++++++++++++++++++++++++++++---------- net/mac80211/parse.c | 30 ++++++++++++++++++++----- net/mac80211/scan.c | 6 ++++- net/mac80211/tdls.c | 12 +++++++--- net/mac80211/tests/elems.c | 4 +++- 12 files changed, 123 insertions(+), 57 deletions(-) diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h index 48ce05e1d203..6d4bc80caf96 100644 --- a/include/linux/ieee80211.h +++ b/include/linux/ieee80211.h @@ -43,6 +43,7 @@ #define IEEE80211_FCTL_VERS 0x0003 #define IEEE80211_FCTL_FTYPE 0x000c #define IEEE80211_FCTL_STYPE 0x00f0 +#define IEEE80211_FCTL_TYPE (IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE) #define IEEE80211_FCTL_TODS 0x0100 #define IEEE80211_FCTL_FROMDS 0x0200 #define IEEE80211_FCTL_MOREFRAGS 0x0400 diff --git a/net/mac80211/agg-rx.c b/net/mac80211/agg-rx.c index e38f46ffebfa..7da909d78c68 100644 --- a/net/mac80211/agg-rx.c +++ b/net/mac80211/agg-rx.c @@ -9,7 +9,7 @@ * Copyright 2007, Michael Wu * Copyright 2007-2010, Intel Corporation * Copyright(c) 2015-2017 Intel Deutschland GmbH - * Copyright (C) 2018-2024 Intel Corporation + * Copyright (C) 2018-2025 Intel Corporation */ /** @@ -206,7 +206,10 @@ u8 ieee80211_retrieve_addba_ext_data(struct sta_info *sta, if (elem_len <= 0) return 0; - elems = ieee802_11_parse_elems(elem_data, elem_len, true, NULL); + elems = ieee802_11_parse_elems(elem_data, elem_len, + IEEE80211_FTYPE_MGMT | + IEEE80211_STYPE_ACTION, + NULL); if (!elems || elems->parse_error || !elems->addba_ext_ie) goto free; diff --git a/net/mac80211/ibss.c b/net/mac80211/ibss.c index 6e36b09fe97f..168f84a1353b 100644 --- a/net/mac80211/ibss.c +++ b/net/mac80211/ibss.c @@ -9,7 +9,7 @@ * Copyright 2009, Johannes Berg * Copyright 2013-2014 Intel Mobile Communications GmbH * Copyright(c) 2016 Intel Deutschland GmbH - * Copyright(c) 2018-2024 Intel Corporation + * Copyright(c) 2018-2025 Intel Corporation */ #include @@ -1554,6 +1554,7 @@ void ieee80211_rx_mgmt_probe_beacon(struct ieee80211_sub_if_data *sdata, { size_t baselen; struct ieee802_11_elems *elems; + u16 type; BUILD_BUG_ON(offsetof(typeof(mgmt->u.probe_resp), variable) != offsetof(typeof(mgmt->u.beacon), variable)); @@ -1566,8 +1567,9 @@ void ieee80211_rx_mgmt_probe_beacon(struct ieee80211_sub_if_data *sdata, if (baselen > len) return; + type = le16_to_cpu(mgmt->frame_control) & IEEE80211_FCTL_TYPE; elems = ieee802_11_parse_elems(mgmt->u.probe_resp.variable, - len - baselen, false, NULL); + len - baselen, type, NULL); if (elems) { ieee80211_rx_bss_info(sdata, mgmt, len, rx_status, elems); @@ -1616,9 +1618,11 @@ void ieee80211_ibss_rx_queued_mgmt(struct ieee80211_sub_if_data *sdata, if (ies_len < 0) break; - elems = ieee802_11_parse_elems( - mgmt->u.action.u.chan_switch.variable, - ies_len, true, NULL); + elems = ieee802_11_parse_elems(mgmt->u.action.u.chan_switch.variable, + ies_len, + IEEE80211_FTYPE_MGMT | + IEEE80211_STYPE_ACTION, + NULL); if (elems && !elems->parse_error) ieee80211_rx_mgmt_spectrum_mgmt(sdata, mgmt, diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index 898ccbc4ec64..7f1ce9fc01c7 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -2422,7 +2422,8 @@ static inline void ieee80211_tx_skb(struct ieee80211_sub_if_data *sdata, * @mode: connection mode for parsing * @start: pointer to the elements * @len: length of the elements - * @action: %true if the elements came from an action frame + * @type: type of the frame the elements came from + * (action, probe response, beacon, etc.) * @filter: bitmap of element IDs to filter out while calculating * the element CRC * @crc: CRC starting value @@ -2440,7 +2441,7 @@ struct ieee80211_elems_parse_params { enum ieee80211_conn_mode mode; const u8 *start; size_t len; - bool action; + u8 type; u64 filter; u32 crc; struct cfg80211_bss *bss; @@ -2452,17 +2453,14 @@ struct ieee802_11_elems * ieee802_11_parse_elems_full(struct ieee80211_elems_parse_params *params); static inline struct ieee802_11_elems * -ieee802_11_parse_elems_crc(const u8 *start, size_t len, bool action, - u64 filter, u32 crc, - struct cfg80211_bss *bss) +ieee802_11_parse_elems(const u8 *start, size_t len, u8 type, + struct cfg80211_bss *bss) { struct ieee80211_elems_parse_params params = { .mode = IEEE80211_CONN_MODE_HIGHEST, .start = start, .len = len, - .action = action, - .filter = filter, - .crc = crc, + .type = type, .bss = bss, .link_id = -1, }; @@ -2470,13 +2468,6 @@ ieee802_11_parse_elems_crc(const u8 *start, size_t len, bool action, return ieee802_11_parse_elems_full(¶ms); } -static inline struct ieee802_11_elems * -ieee802_11_parse_elems(const u8 *start, size_t len, bool action, - struct cfg80211_bss *bss) -{ - return ieee802_11_parse_elems_crc(start, len, action, 0, 0, bss); -} - extern const int ieee802_1d_to_ac[8]; static inline int ieee80211_ac_from_tid(int tid) diff --git a/net/mac80211/mesh.c b/net/mac80211/mesh.c index f37068a533f4..68901f1def0d 100644 --- a/net/mac80211/mesh.c +++ b/net/mac80211/mesh.c @@ -1,7 +1,7 @@ // SPDX-License-Identifier: GPL-2.0-only /* * Copyright (c) 2008, 2009 open80211s Ltd. - * Copyright (C) 2018 - 2024 Intel Corporation + * Copyright (C) 2018 - 2025 Intel Corporation * Authors: Luis Carlos Cobo * Javier Cardona */ @@ -1410,7 +1410,10 @@ ieee80211_mesh_rx_probe_req(struct ieee80211_sub_if_data *sdata, if (baselen > len) return; - elems = ieee802_11_parse_elems(pos, len - baselen, false, NULL); + elems = ieee802_11_parse_elems(pos, len - baselen, + IEEE80211_FTYPE_MGMT | + IEEE80211_STYPE_PROBE_REQ, + NULL); if (!elems) return; @@ -1455,11 +1458,11 @@ ieee80211_mesh_rx_probe_req(struct ieee80211_sub_if_data *sdata, } static void ieee80211_mesh_rx_bcn_presp(struct ieee80211_sub_if_data *sdata, - u16 stype, struct ieee80211_mgmt *mgmt, size_t len, struct ieee80211_rx_status *rx_status) { + u16 type = le16_to_cpu(mgmt->frame_control) & IEEE80211_FCTL_TYPE; struct ieee80211_local *local = sdata->local; struct ieee80211_if_mesh *ifmsh = &sdata->u.mesh; struct ieee802_11_elems *elems; @@ -1469,7 +1472,7 @@ static void ieee80211_mesh_rx_bcn_presp(struct ieee80211_sub_if_data *sdata, enum nl80211_band band = rx_status->band; /* ignore ProbeResp to foreign address */ - if (stype == IEEE80211_STYPE_PROBE_RESP && + if (type == (IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_PROBE_RESP) && !ether_addr_equal(mgmt->da, sdata->vif.addr)) return; @@ -1478,8 +1481,7 @@ static void ieee80211_mesh_rx_bcn_presp(struct ieee80211_sub_if_data *sdata, return; elems = ieee802_11_parse_elems(mgmt->u.probe_resp.variable, - len - baselen, - false, NULL); + len - baselen, type, NULL); if (!elems) return; @@ -1514,7 +1516,9 @@ static void ieee80211_mesh_rx_bcn_presp(struct ieee80211_sub_if_data *sdata, } if (ifmsh->sync_ops) - ifmsh->sync_ops->rx_bcn_presp(sdata, stype, mgmt, len, + ifmsh->sync_ops->rx_bcn_presp(sdata, + type & IEEE80211_FCTL_STYPE, + mgmt, len, elems->mesh_config, rx_status); free: kfree(elems); @@ -1622,7 +1626,10 @@ static void mesh_rx_csa_frame(struct ieee80211_sub_if_data *sdata, pos = mgmt->u.action.u.chan_switch.variable; baselen = offsetof(struct ieee80211_mgmt, u.action.u.chan_switch.variable); - elems = ieee802_11_parse_elems(pos, len - baselen, true, NULL); + elems = ieee802_11_parse_elems(pos, len - baselen, + IEEE80211_FTYPE_MGMT | + IEEE80211_STYPE_ACTION, + NULL); if (!elems) return; @@ -1699,8 +1706,7 @@ void ieee80211_mesh_rx_queued_mgmt(struct ieee80211_sub_if_data *sdata, switch (stype) { case IEEE80211_STYPE_PROBE_RESP: case IEEE80211_STYPE_BEACON: - ieee80211_mesh_rx_bcn_presp(sdata, stype, mgmt, skb->len, - rx_status); + ieee80211_mesh_rx_bcn_presp(sdata, mgmt, skb->len, rx_status); break; case IEEE80211_STYPE_PROBE_REQ: ieee80211_mesh_rx_probe_req(sdata, mgmt, skb->len); diff --git a/net/mac80211/mesh_hwmp.c b/net/mac80211/mesh_hwmp.c index 9101858525dd..a41b57bd11ff 100644 --- a/net/mac80211/mesh_hwmp.c +++ b/net/mac80211/mesh_hwmp.c @@ -1,7 +1,7 @@ // SPDX-License-Identifier: GPL-2.0-only /* * Copyright (c) 2008, 2009 open80211s Ltd. - * Copyright (C) 2019, 2021-2023 Intel Corporation + * Copyright (C) 2019, 2021-2023, 2025 Intel Corporation * Author: Luis Carlos Cobo */ @@ -951,7 +951,10 @@ void mesh_rx_path_sel_frame(struct ieee80211_sub_if_data *sdata, baselen = (u8 *) mgmt->u.action.u.mesh_action.variable - (u8 *) mgmt; elems = ieee802_11_parse_elems(mgmt->u.action.u.mesh_action.variable, - len - baselen, false, NULL); + len - baselen, + IEEE80211_FTYPE_MGMT | + IEEE80211_STYPE_ACTION, + NULL); if (!elems) return; diff --git a/net/mac80211/mesh_plink.c b/net/mac80211/mesh_plink.c index cb45a5d2009d..04c931cd2063 100644 --- a/net/mac80211/mesh_plink.c +++ b/net/mac80211/mesh_plink.c @@ -1,7 +1,7 @@ // SPDX-License-Identifier: GPL-2.0-only /* * Copyright (c) 2008, 2009 open80211s Ltd. - * Copyright (C) 2019, 2021-2024 Intel Corporation + * Copyright (C) 2019, 2021-2025 Intel Corporation * Author: Luis Carlos Cobo */ #include @@ -1248,7 +1248,10 @@ void mesh_rx_plink_frame(struct ieee80211_sub_if_data *sdata, if (baselen > len) return; } - elems = ieee802_11_parse_elems(baseaddr, len - baselen, true, NULL); + elems = ieee802_11_parse_elems(baseaddr, len - baselen, + IEEE80211_FTYPE_MGMT | + IEEE80211_STYPE_ACTION, + NULL); if (elems) { mesh_process_plink_frame(sdata, mgmt, elems, rx_status); kfree(elems); diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c index 2ee9eae89d05..804c3a95b7c6 100644 --- a/net/mac80211/mlme.c +++ b/net/mac80211/mlme.c @@ -999,6 +999,9 @@ ieee80211_determine_chan_mode(struct ieee80211_sub_if_data *sdata, .from_ap = true, .start = ies->data, .len = ies->len, + .type = ies->from_beacon ? + IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_BEACON : + IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_PROBE_RESP, }; struct ieee802_11_elems *elems; struct ieee80211_supported_band *sband; @@ -5177,7 +5180,9 @@ static void ieee80211_epcs_teardown(struct ieee80211_sub_if_data *sdata) continue; } - elems = ieee802_11_parse_elems(ies->data, ies->len, false, + elems = ieee802_11_parse_elems(ies->data, ies->len, + IEEE80211_FTYPE_MGMT | + IEEE80211_STYPE_BEACON, NULL); if (!elems) { rcu_read_unlock(); @@ -5223,6 +5228,7 @@ static bool ieee80211_assoc_config_link(struct ieee80211_link_data *link, .len = elem_len, .link_id = link_id == assoc_data->assoc_link_id ? -1 : link_id, .from_ap = true, + .type = le16_to_cpu(mgmt->frame_control) & IEEE80211_FCTL_TYPE, }; bool is_5ghz = cbss->channel->band == NL80211_BAND_5GHZ; bool is_6ghz = cbss->channel->band == NL80211_BAND_6GHZ; @@ -6356,6 +6362,7 @@ static void ieee80211_rx_mgmt_assoc_resp(struct ieee80211_sub_if_data *sdata, .bss = NULL, .link_id = -1, .from_ap = true, + .type = le16_to_cpu(mgmt->frame_control) & IEEE80211_FCTL_TYPE, }; struct ieee802_11_elems *elems; int ac; @@ -7264,7 +7271,9 @@ ieee80211_mgd_check_cross_link_csa(struct ieee80211_sub_if_data *sdata, (prof->sta_info_len - 1), len - (prof->sta_info_len - 1), - false, NULL); + IEEE80211_FTYPE_MGMT | + IEEE80211_STYPE_BEACON, + NULL); /* memory allocation failed - let's hope that's transient */ if (!prof_elems) @@ -7368,6 +7377,7 @@ static void ieee80211_rx_mgmt_beacon(struct ieee80211_link_data *link, .mode = link->u.mgd.conn.mode, .link_id = -1, .from_ap = true, + .type = le16_to_cpu(mgmt->frame_control) & IEEE80211_FCTL_TYPE, }; lockdep_assert_wiphy(local->hw.wiphy); @@ -7970,7 +7980,10 @@ void ieee80211_process_neg_ttlm_req(struct ieee80211_sub_if_data *sdata, ies_len = len - offsetof(struct ieee80211_mgmt, u.action.u.ttlm_req.variable); elems = ieee802_11_parse_elems(mgmt->u.action.u.ttlm_req.variable, - ies_len, true, NULL); + ies_len, + IEEE80211_FTYPE_MGMT | + IEEE80211_STYPE_ACTION, + NULL); if (!elems) { ttlm_res = NEG_TTLM_RES_REJECT; goto out; @@ -8176,9 +8189,11 @@ void ieee80211_sta_rx_queued_mgmt(struct ieee80211_sub_if_data *sdata, break; /* CSA IE cannot be overridden, no need for BSSID */ - elems = ieee802_11_parse_elems( - mgmt->u.action.u.chan_switch.variable, - ies_len, true, NULL); + elems = ieee802_11_parse_elems(mgmt->u.action.u.chan_switch.variable, + ies_len, + IEEE80211_FTYPE_MGMT | + IEEE80211_STYPE_ACTION, + NULL); if (elems && !elems->parse_error) { enum ieee80211_csa_source src = @@ -8205,9 +8220,11 @@ void ieee80211_sta_rx_queued_mgmt(struct ieee80211_sub_if_data *sdata, * extended CSA IE can't be overridden, no need for * BSSID */ - elems = ieee802_11_parse_elems( - mgmt->u.action.u.ext_chan_switch.variable, - ies_len, true, NULL); + elems = ieee802_11_parse_elems(mgmt->u.action.u.ext_chan_switch.variable, + ies_len, + IEEE80211_FTYPE_MGMT | + IEEE80211_STYPE_ACTION, + NULL); if (elems && !elems->parse_error) { enum ieee80211_csa_source src; @@ -10985,7 +11002,10 @@ static void ieee80211_ml_epcs(struct ieee80211_sub_if_data *sdata, pos = scratch + sizeof(control); len -= sizeof(control); - link_elems = ieee802_11_parse_elems(pos, len, false, NULL); + link_elems = ieee802_11_parse_elems(pos, len, + IEEE80211_FTYPE_MGMT | + IEEE80211_STYPE_ACTION, + NULL); if (!link_elems) continue; @@ -11036,7 +11056,10 @@ void ieee80211_process_epcs_ena_resp(struct ieee80211_sub_if_data *sdata, u.action.u.epcs.variable) - IEEE80211_EPCS_ENA_RESP_BODY_LEN; - elems = ieee802_11_parse_elems(pos, ies_len, true, NULL); + elems = ieee802_11_parse_elems(pos, ies_len, + IEEE80211_FTYPE_MGMT | + IEEE80211_STYPE_ACTION, + NULL); if (!elems) return; diff --git a/net/mac80211/parse.c b/net/mac80211/parse.c index c5e0f7f46004..bfc4ecb7a048 100644 --- a/net/mac80211/parse.c +++ b/net/mac80211/parse.c @@ -6,7 +6,7 @@ * Copyright 2007 Johannes Berg * Copyright 2013-2014 Intel Mobile Communications GmbH * Copyright (C) 2015-2017 Intel Deutschland GmbH - * Copyright (C) 2018-2024 Intel Corporation + * Copyright (C) 2018-2025 Intel Corporation * * element parsing for mac80211 */ @@ -286,6 +286,24 @@ _ieee802_11_parse_elems_full(struct ieee80211_elems_parse_params *params, bitmap_zero(seen_elems, 256); + switch (params->type) { + /* we don't need to parse assoc request, luckily (it's value 0) */ + case IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_ASSOC_REQ: + case IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_REASSOC_REQ: + default: + WARN(1, "invalid frame type 0x%x for element parsing\n", + params->type); + break; + case IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_ASSOC_RESP: + case IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_REASSOC_RESP: + case IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_PROBE_REQ: + case IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_PROBE_RESP: + case IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_BEACON: + case IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_ACTION: + case IEEE80211_FTYPE_EXT | IEEE80211_STYPE_S1G_BEACON: + break; + } + for_each_element(elem, params->start, params->len) { const struct element *subelem; u8 elem_parse_failed; @@ -566,7 +584,8 @@ _ieee802_11_parse_elems_full(struct ieee80211_elems_parse_params *params, if (params->mode < IEEE80211_CONN_MODE_VHT) break; - if (!params->action) { + if (params->type != (IEEE80211_FTYPE_MGMT | + IEEE80211_STYPE_ACTION)) { elem_parse_failed = IEEE80211_PARSE_ERR_UNEXPECTED_ELEM; break; @@ -582,7 +601,8 @@ _ieee802_11_parse_elems_full(struct ieee80211_elems_parse_params *params, case WLAN_EID_CHANNEL_SWITCH_WRAPPER: if (params->mode < IEEE80211_CONN_MODE_VHT) break; - if (params->action) { + if (params->type == (IEEE80211_FTYPE_MGMT | + IEEE80211_STYPE_ACTION)) { elem_parse_failed = IEEE80211_PARSE_ERR_UNEXPECTED_ELEM; break; @@ -942,7 +962,7 @@ ieee80211_prep_mle_link_parse(struct ieee80211_elems_parse *elems_parse, sub->len = end - sub->start; sub->mode = params->mode; - sub->action = params->action; + sub->type = params->type; sub->from_ap = params->from_ap; sub->link_id = -1; @@ -1041,7 +1061,7 @@ ieee802_11_parse_elems_full(struct ieee80211_elems_parse_params *params) sub.start = elems_parse->scratch_pos; sub.mode = params->mode; sub.len = nontx_len; - sub.action = params->action; + sub.type = params->type; sub.link_id = params->link_id; /* consume the space used for non-transmitted profile */ diff --git a/net/mac80211/scan.c b/net/mac80211/scan.c index bb9563f50e7b..5ef315ed3b0f 100644 --- a/net/mac80211/scan.c +++ b/net/mac80211/scan.c @@ -76,7 +76,11 @@ void ieee80211_inform_bss(struct wiphy *wiphy, if (!update_data) return; - elems = ieee802_11_parse_elems(ies->data, ies->len, false, NULL); + elems = ieee802_11_parse_elems(ies->data, ies->len, + update_data->beacon ? + IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_BEACON : + IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_PROBE_RESP, + NULL); if (!elems) return; diff --git a/net/mac80211/tdls.c b/net/mac80211/tdls.c index ba5fbacbeeda..dbbfe2d6842f 100644 --- a/net/mac80211/tdls.c +++ b/net/mac80211/tdls.c @@ -6,7 +6,7 @@ * Copyright 2014, Intel Corporation * Copyright 2014 Intel Mobile Communications GmbH * Copyright 2015 - 2016 Intel Deutschland GmbH - * Copyright (C) 2019, 2021-2024 Intel Corporation + * Copyright (C) 2019, 2021-2025 Intel Corporation */ #include @@ -1783,7 +1783,10 @@ ieee80211_process_tdls_channel_switch_resp(struct ieee80211_sub_if_data *sdata, } elems = ieee802_11_parse_elems(tf->u.chan_switch_resp.variable, - skb->len - baselen, false, NULL); + skb->len - baselen, + IEEE80211_FTYPE_MGMT | + IEEE80211_STYPE_ACTION, + NULL); if (!elems) { ret = -ENOMEM; goto out; @@ -1902,7 +1905,10 @@ ieee80211_process_tdls_channel_switch_req(struct ieee80211_sub_if_data *sdata, } elems = ieee802_11_parse_elems(tf->u.chan_switch_req.variable, - skb->len - baselen, false, NULL); + skb->len - baselen, + IEEE80211_FTYPE_MGMT | + IEEE80211_STYPE_ACTION, + NULL); if (!elems) return -ENOMEM; diff --git a/net/mac80211/tests/elems.c b/net/mac80211/tests/elems.c index a53c55a879a8..1039794a0183 100644 --- a/net/mac80211/tests/elems.c +++ b/net/mac80211/tests/elems.c @@ -2,7 +2,7 @@ /* * KUnit tests for element parsing * - * Copyright (C) 2023-2024 Intel Corporation + * Copyright (C) 2023-2025 Intel Corporation */ #include #include "../ieee80211_i.h" @@ -15,6 +15,8 @@ static void mle_defrag(struct kunit *test) .link_id = 12, .from_ap = true, .mode = IEEE80211_CONN_MODE_EHT, + /* type is not really relevant here */ + .type = IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_BEACON, }; struct ieee802_11_elems *parsed; struct sk_buff *skb; From 243d30fbb679f3442d2e697bd344dee975982583 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 5 Nov 2025 16:10:33 +0100 Subject: [PATCH 0649/1381] wifi: mac80211: remove unnecessary vlan NULL check In a vlan iteration loop the vlan pointer cannot be NULL. Remove the unnecessary check. Link: https://patch.msgid.link/20251105161033.670b5a06296c.I24cb1a5338736ab0a8a24d6a492c259f894d09fb@changeid Signed-off-by: Johannes Berg --- net/mac80211/link.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/net/mac80211/link.c b/net/mac80211/link.c index 85ed3cbb2e36..1e05845872af 100644 --- a/net/mac80211/link.c +++ b/net/mac80211/link.c @@ -23,9 +23,6 @@ static void ieee80211_update_apvlan_links(struct ieee80211_sub_if_data *sdata) list_for_each_entry(vlan, &sdata->u.ap.vlans, u.vlan.list) { int link_id; - if (!vlan) - continue; - /* No support for 4addr with MLO yet */ if (vlan->wdev.use_4addr) return; From 473235677af46ecb167917887586646e9d70d9ff Mon Sep 17 00:00:00 2001 From: Chien Wong Date: Fri, 7 Nov 2025 22:23:32 +0800 Subject: [PATCH 0650/1381] wifi: cfg80211: fix doc of struct key_params The seq in struct key_params is for many ciphers, including CCMP, GCMP, CMAC, GMAC. In addition to get_key(), it is also used when setting keys. Signed-off-by: Chien Wong Link: https://patch.msgid.link/20251107142332.181308-1-m@xv97.com Signed-off-by: Johannes Berg --- include/net/cfg80211.h | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index 84be0cdd1da0..d87c18e1b133 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -786,8 +786,7 @@ struct vif_params { * @key: key material * @key_len: length of key material * @cipher: cipher suite selector - * @seq: sequence counter (IV/PN) for TKIP and CCMP keys, only used - * with the get_key() callback, must be in little endian, + * @seq: sequence counter (IV/PN), must be in little endian, * length given by @seq_len. * @seq_len: length of @seq. * @vlan_id: vlan_id for VLAN group key (if nonzero) From 8989d328dfe7c7a3b9f4b9f0ef60006d277f81cc Mon Sep 17 00:00:00 2001 From: Jakub Sitnicki Date: Wed, 5 Nov 2025 21:19:38 +0100 Subject: [PATCH 0651/1381] net: Helper to move packet data and metadata after skb_push/pull Lay groundwork for fixing BPF helpers available to TC(X) programs. When skb_push() or skb_pull() is called in a TC(X) ingress BPF program, the skb metadata must be kept in front of the MAC header. Otherwise, BPF programs using the __sk_buff->data_meta pseudo-pointer lose access to it. Introduce a helper that moves both metadata and a specified number of packet data bytes together, suitable as a drop-in replacement for memmove(). Signed-off-by: Jakub Sitnicki Signed-off-by: Martin KaFai Lau Link: https://patch.msgid.link/20251105-skb-meta-rx-path-v4-1-5ceb08a9b37b@cloudflare.com --- include/linux/skbuff.h | 75 ++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 75 insertions(+) diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index a7cc3d1f4fd1..ff90281ddf90 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -4564,6 +4564,81 @@ static inline void skb_metadata_clear(struct sk_buff *skb) skb_metadata_set(skb, 0); } +/** + * skb_data_move - Move packet data and metadata after skb_push() or skb_pull(). + * @skb: packet to operate on + * @len: number of bytes pushed or pulled from &sk_buff->data + * @n: number of bytes to memmove() from pre-push/pull &sk_buff->data + * + * Moves @n bytes of packet data, can be zero, and all bytes of skb metadata. + * + * Assumes metadata is located immediately before &sk_buff->data prior to the + * push/pull, and that sufficient headroom exists to hold it after an + * skb_push(). Otherwise, metadata is cleared and a one-time warning is issued. + * + * Prefer skb_postpull_data_move() or skb_postpush_data_move() to calling this + * helper directly. + */ +static inline void skb_data_move(struct sk_buff *skb, const int len, + const unsigned int n) +{ + const u8 meta_len = skb_metadata_len(skb); + u8 *meta, *meta_end; + + if (!len || (!n && !meta_len)) + return; + + if (!meta_len) + goto no_metadata; + + meta_end = skb_metadata_end(skb); + meta = meta_end - meta_len; + + if (WARN_ON_ONCE(meta_end + len != skb->data || + meta_len > skb_headroom(skb))) { + skb_metadata_clear(skb); + goto no_metadata; + } + + memmove(meta + len, meta, meta_len + n); + return; + +no_metadata: + memmove(skb->data, skb->data - len, n); +} + +/** + * skb_postpull_data_move - Move packet data and metadata after skb_pull(). + * @skb: packet to operate on + * @len: number of bytes pulled from &sk_buff->data + * @n: number of bytes to memmove() from pre-pull &sk_buff->data + * + * See skb_data_move() for details. + */ +static inline void skb_postpull_data_move(struct sk_buff *skb, + const unsigned int len, + const unsigned int n) +{ + DEBUG_NET_WARN_ON_ONCE(len > INT_MAX); + skb_data_move(skb, len, n); +} + +/** + * skb_postpush_data_move - Move packet data and metadata after skb_push(). + * @skb: packet to operate on + * @len: number of bytes pushed onto &sk_buff->data + * @n: number of bytes to memmove() from pre-push &sk_buff->data + * + * See skb_data_move() for details. + */ +static inline void skb_postpush_data_move(struct sk_buff *skb, + const unsigned int len, + const unsigned int n) +{ + DEBUG_NET_WARN_ON_ONCE(len > INT_MAX); + skb_data_move(skb, -len, n); +} + struct sk_buff *skb_clone_sk(struct sk_buff *skb); #ifdef CONFIG_NETWORK_PHY_TIMESTAMPING From 290fc0be09e2f6e08754782616bb7e67f8191fa1 Mon Sep 17 00:00:00 2001 From: Jakub Sitnicki Date: Wed, 5 Nov 2025 21:19:39 +0100 Subject: [PATCH 0652/1381] net: Preserve metadata on pskb_expand_head pskb_expand_head() copies headroom, including skb metadata, into the newly allocated head, but then clears the metadata. As a result, metadata is lost when BPF helpers trigger an skb head reallocation. Let the skb metadata remain in the newly created copy of head. Signed-off-by: Jakub Sitnicki Signed-off-by: Martin KaFai Lau Link: https://patch.msgid.link/20251105-skb-meta-rx-path-v4-2-5ceb08a9b37b@cloudflare.com --- net/core/skbuff.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 7ac5f8aa1235..d95658b738d1 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -2234,6 +2234,10 @@ EXPORT_SYMBOL(__pskb_copy_fclone); * * All the pointers pointing into skb header may change and must be * reloaded after call to this function. + * + * Note: If you skb_push() the start of the buffer after reallocating the + * header, call skb_postpush_data_move() first to move the metadata out of + * the way before writing to &sk_buff->data. */ int pskb_expand_head(struct sk_buff *skb, int nhead, int ntail, @@ -2305,8 +2309,6 @@ int pskb_expand_head(struct sk_buff *skb, int nhead, int ntail, skb->nohdr = 0; atomic_set(&skb_shinfo(skb)->dataref, 1); - skb_metadata_clear(skb); - /* It is not generally safe to change skb->truesize. * For the moment, we really care of rx path, or * when skb is orphaned (not attached to a socket). From f38499ff45f567c932d0911e6a30b8ca022b9b52 Mon Sep 17 00:00:00 2001 From: Jakub Sitnicki Date: Wed, 5 Nov 2025 21:19:40 +0100 Subject: [PATCH 0653/1381] bpf: Unclone skb head on bpf_dynptr_write to skb metadata Currently bpf_dynptr_from_skb_meta() marks the dynptr as read-only when the skb is cloned, preventing writes to metadata. Remove this restriction and unclone the skb head on bpf_dynptr_write() to metadata, now that the metadata is preserved during uncloning. This makes metadata dynptr consistent with skb dynptr, allowing writes regardless of whether the skb is cloned. Signed-off-by: Jakub Sitnicki Signed-off-by: Martin KaFai Lau Link: https://patch.msgid.link/20251105-skb-meta-rx-path-v4-3-5ceb08a9b37b@cloudflare.com --- include/linux/filter.h | 9 +++++++++ kernel/bpf/helpers.c | 6 ++---- net/core/filter.c | 18 ++++++++++++------ 3 files changed, 23 insertions(+), 10 deletions(-) diff --git a/include/linux/filter.h b/include/linux/filter.h index e116de7edc58..a104b3994230 100644 --- a/include/linux/filter.h +++ b/include/linux/filter.h @@ -1781,6 +1781,8 @@ int __bpf_xdp_store_bytes(struct xdp_buff *xdp, u32 offset, void *buf, u32 len); void *bpf_xdp_pointer(struct xdp_buff *xdp, u32 offset, u32 len); void bpf_xdp_copy_buf(struct xdp_buff *xdp, unsigned long off, void *buf, unsigned long len, bool flush); +int __bpf_skb_meta_store_bytes(struct sk_buff *skb, u32 offset, + const void *from, u32 len, u64 flags); void *bpf_skb_meta_pointer(struct sk_buff *skb, u32 offset); #else /* CONFIG_NET */ static inline int __bpf_skb_load_bytes(const struct sk_buff *skb, u32 offset, @@ -1817,6 +1819,13 @@ static inline void bpf_xdp_copy_buf(struct xdp_buff *xdp, unsigned long off, voi { } +static inline int __bpf_skb_meta_store_bytes(struct sk_buff *skb, u32 offset, + const void *from, u32 len, + u64 flags) +{ + return -EOPNOTSUPP; +} + static inline void *bpf_skb_meta_pointer(struct sk_buff *skb, u32 offset) { return ERR_PTR(-EOPNOTSUPP); diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c index eb25e70e0bdc..3e830fd31f5f 100644 --- a/kernel/bpf/helpers.c +++ b/kernel/bpf/helpers.c @@ -1842,10 +1842,8 @@ int __bpf_dynptr_write(const struct bpf_dynptr_kern *dst, u32 offset, void *src, return -EINVAL; return __bpf_xdp_store_bytes(dst->data, dst->offset + offset, src, len); case BPF_DYNPTR_TYPE_SKB_META: - if (flags) - return -EINVAL; - memmove(bpf_skb_meta_pointer(dst->data, dst->offset + offset), src, len); - return 0; + return __bpf_skb_meta_store_bytes(dst->data, dst->offset + offset, src, + len, flags); default: WARN_ONCE(true, "bpf_dynptr_write: unknown dynptr type %d\n", type); return -EFAULT; diff --git a/net/core/filter.c b/net/core/filter.c index 52721efba332..673299fd3d58 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -12102,6 +12102,18 @@ void *bpf_skb_meta_pointer(struct sk_buff *skb, u32 offset) return skb_metadata_end(skb) - skb_metadata_len(skb) + offset; } +int __bpf_skb_meta_store_bytes(struct sk_buff *skb, u32 offset, + const void *from, u32 len, u64 flags) +{ + if (unlikely(flags)) + return -EINVAL; + if (unlikely(bpf_try_make_writable(skb, 0))) + return -EFAULT; + + memmove(bpf_skb_meta_pointer(skb, offset), from, len); + return 0; +} + __bpf_kfunc_start_defs(); __bpf_kfunc int bpf_dynptr_from_skb(struct __sk_buff *s, u64 flags, struct bpf_dynptr *ptr__uninit) @@ -12129,9 +12141,6 @@ __bpf_kfunc int bpf_dynptr_from_skb(struct __sk_buff *s, u64 flags, * XDP context with bpf_xdp_adjust_meta(). Serves as an alternative to * &__sk_buff->data_meta. * - * If passed @skb_ is a clone which shares the data with the original, the - * dynptr will be read-only. This limitation may be lifted in the future. - * * Return: * * %0 - dynptr ready to use * * %-EINVAL - invalid flags, dynptr set to null @@ -12149,9 +12158,6 @@ __bpf_kfunc int bpf_dynptr_from_skb_meta(struct __sk_buff *skb_, u64 flags, bpf_dynptr_init(ptr, skb, BPF_DYNPTR_TYPE_SKB_META, 0, skb_metadata_len(skb)); - if (skb_cloned(skb)) - bpf_dynptr_set_rdonly(ptr); - return 0; } From b85be58e2f7cff47f7477ae61022644a198ee592 Mon Sep 17 00:00:00 2001 From: Jakub Sitnicki Date: Wed, 5 Nov 2025 21:19:41 +0100 Subject: [PATCH 0654/1381] vlan: Make vlan_remove_tag return nothing All callers ignore the return value. Prepare to reorder memmove() after skb_pull() which is a common pattern. Signed-off-by: Jakub Sitnicki Signed-off-by: Martin KaFai Lau Link: https://patch.msgid.link/20251105-skb-meta-rx-path-v4-4-5ceb08a9b37b@cloudflare.com --- include/linux/if_vlan.h | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h index 15e01935d3fa..afa5cc61a0fa 100644 --- a/include/linux/if_vlan.h +++ b/include/linux/if_vlan.h @@ -731,10 +731,8 @@ static inline void vlan_set_encap_proto(struct sk_buff *skb, * * Expects the skb to contain a VLAN tag in the payload, and to have skb->data * pointing at the MAC header. - * - * Returns: a new pointer to skb->data, or NULL on failure to pull. */ -static inline void *vlan_remove_tag(struct sk_buff *skb, u16 *vlan_tci) +static inline void vlan_remove_tag(struct sk_buff *skb, u16 *vlan_tci) { struct vlan_hdr *vhdr = (struct vlan_hdr *)(skb->data + ETH_HLEN); @@ -742,7 +740,7 @@ static inline void *vlan_remove_tag(struct sk_buff *skb, u16 *vlan_tci) memmove(skb->data + VLAN_HLEN, skb->data, 2 * ETH_ALEN); vlan_set_encap_proto(skb, vhdr); - return __skb_pull(skb, VLAN_HLEN); + __skb_pull(skb, VLAN_HLEN); } /** From efd35c26239bed39193201e958d65e695231ccda Mon Sep 17 00:00:00 2001 From: Jakub Sitnicki Date: Wed, 5 Nov 2025 21:19:42 +0100 Subject: [PATCH 0655/1381] bpf: Make bpf_skb_vlan_pop helper metadata-safe Use the metadata-aware helper to move packet bytes after skb_pull(), ensuring metadata remains valid after calling the BPF helper. Signed-off-by: Jakub Sitnicki Signed-off-by: Martin KaFai Lau Link: https://patch.msgid.link/20251105-skb-meta-rx-path-v4-5-5ceb08a9b37b@cloudflare.com --- include/linux/if_vlan.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h index afa5cc61a0fa..4ecc2509b0d4 100644 --- a/include/linux/if_vlan.h +++ b/include/linux/if_vlan.h @@ -738,9 +738,9 @@ static inline void vlan_remove_tag(struct sk_buff *skb, u16 *vlan_tci) *vlan_tci = ntohs(vhdr->h_vlan_TCI); - memmove(skb->data + VLAN_HLEN, skb->data, 2 * ETH_ALEN); vlan_set_encap_proto(skb, vhdr); __skb_pull(skb, VLAN_HLEN); + skb_postpull_data_move(skb, VLAN_HLEN, 2 * ETH_ALEN); } /** From 55ffc98b44d28e13a218306666d16f2c7236d0ae Mon Sep 17 00:00:00 2001 From: Jakub Sitnicki Date: Wed, 5 Nov 2025 21:19:43 +0100 Subject: [PATCH 0656/1381] bpf: Make bpf_skb_vlan_push helper metadata-safe Use the metadata-aware helper to move packet bytes after skb_push(), ensuring metadata remains valid after calling the BPF helper. Also, take care to reserve sufficient headroom for metadata to fit. Signed-off-by: Jakub Sitnicki Signed-off-by: Martin KaFai Lau Link: https://patch.msgid.link/20251105-skb-meta-rx-path-v4-6-5ceb08a9b37b@cloudflare.com --- include/linux/if_vlan.h | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h index 4ecc2509b0d4..f7f34eb15e06 100644 --- a/include/linux/if_vlan.h +++ b/include/linux/if_vlan.h @@ -355,16 +355,17 @@ static inline int __vlan_insert_inner_tag(struct sk_buff *skb, __be16 vlan_proto, u16 vlan_tci, unsigned int mac_len) { + const u8 meta_len = mac_len > ETH_TLEN ? skb_metadata_len(skb) : 0; struct vlan_ethhdr *veth; - if (skb_cow_head(skb, VLAN_HLEN) < 0) + if (skb_cow_head(skb, meta_len + VLAN_HLEN) < 0) return -ENOMEM; skb_push(skb, VLAN_HLEN); /* Move the mac header sans proto to the beginning of the new header. */ if (likely(mac_len > ETH_TLEN)) - memmove(skb->data, skb->data + VLAN_HLEN, mac_len - ETH_TLEN); + skb_postpush_data_move(skb, VLAN_HLEN, mac_len - ETH_TLEN); if (skb_mac_header_was_set(skb)) skb->mac_header -= VLAN_HLEN; From be83105d38ab99737ac0488600efd59ddd57b74c Mon Sep 17 00:00:00 2001 From: Jakub Sitnicki Date: Wed, 5 Nov 2025 21:19:44 +0100 Subject: [PATCH 0657/1381] bpf: Make bpf_skb_adjust_room metadata-safe bpf_skb_adjust_room() may push or pull bytes from skb->data. In both cases, skb metadata must be moved accordingly to stay accessible. Replace existing memmove() calls, which only move payload, with a helper that also handles metadata. Reserve enough space for metadata to fit after skb_push. Signed-off-by: Jakub Sitnicki Signed-off-by: Martin KaFai Lau Link: https://patch.msgid.link/20251105-skb-meta-rx-path-v4-7-5ceb08a9b37b@cloudflare.com --- net/core/filter.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/net/core/filter.c b/net/core/filter.c index 673299fd3d58..297f0ae9dc1f 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -3253,11 +3253,11 @@ static void bpf_skb_change_protocol(struct sk_buff *skb, u16 proto) static int bpf_skb_generic_push(struct sk_buff *skb, u32 off, u32 len) { - /* Caller already did skb_cow() with len as headroom, + /* Caller already did skb_cow() with meta_len+len as headroom, * so no need to do it here. */ skb_push(skb, len); - memmove(skb->data, skb->data + len, off); + skb_postpush_data_move(skb, len, off); memset(skb->data + off, 0, len); /* No skb_postpush_rcsum(skb, skb->data + off, len) @@ -3281,7 +3281,7 @@ static int bpf_skb_generic_pop(struct sk_buff *skb, u32 off, u32 len) old_data = skb->data; __skb_pull(skb, len); skb_postpull_rcsum(skb, old_data + off, len); - memmove(skb->data, old_data, off); + skb_postpull_data_move(skb, len, off); return 0; } @@ -3489,6 +3489,7 @@ static int bpf_skb_net_grow(struct sk_buff *skb, u32 off, u32 len_diff, u8 inner_mac_len = flags >> BPF_ADJ_ROOM_ENCAP_L2_SHIFT; bool encap = flags & BPF_F_ADJ_ROOM_ENCAP_L3_MASK; u16 mac_len = 0, inner_net = 0, inner_trans = 0; + const u8 meta_len = skb_metadata_len(skb); unsigned int gso_type = SKB_GSO_DODGY; int ret; @@ -3499,7 +3500,7 @@ static int bpf_skb_net_grow(struct sk_buff *skb, u32 off, u32 len_diff, return -ENOTSUPP; } - ret = skb_cow_head(skb, len_diff); + ret = skb_cow_head(skb, meta_len + len_diff); if (unlikely(ret < 0)) return ret; From 8cfc172ce28e6559d4d2d1a96df77f0f2d6179d6 Mon Sep 17 00:00:00 2001 From: Jakub Sitnicki Date: Wed, 5 Nov 2025 21:19:45 +0100 Subject: [PATCH 0658/1381] bpf: Make bpf_skb_change_proto helper metadata-safe bpf_skb_change_proto reuses the same headroom operations as bpf_skb_adjust_room, already updated to handle metadata safely. The remaining step is to ensure that there is sufficient headroom to accommodate metadata on skb_push(). Signed-off-by: Jakub Sitnicki Signed-off-by: Martin KaFai Lau Link: https://patch.msgid.link/20251105-skb-meta-rx-path-v4-8-5ceb08a9b37b@cloudflare.com --- net/core/filter.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/core/filter.c b/net/core/filter.c index 297f0ae9dc1f..50775c01c456 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -3326,10 +3326,11 @@ static int bpf_skb_net_hdr_pop(struct sk_buff *skb, u32 off, u32 len) static int bpf_skb_proto_4_to_6(struct sk_buff *skb) { const u32 len_diff = sizeof(struct ipv6hdr) - sizeof(struct iphdr); + const u8 meta_len = skb_metadata_len(skb); u32 off = skb_mac_header_len(skb); int ret; - ret = skb_cow(skb, len_diff); + ret = skb_cow(skb, meta_len + len_diff); if (unlikely(ret < 0)) return ret; From fb206fc3129bc9d4749905d4870ba05dc89126d2 Mon Sep 17 00:00:00 2001 From: Jakub Sitnicki Date: Wed, 5 Nov 2025 21:19:46 +0100 Subject: [PATCH 0659/1381] bpf: Make bpf_skb_change_head helper metadata-safe Although bpf_skb_change_head() doesn't move packet data after skb_push(), skb metadata still needs to be relocated. Use the dedicated helper to handle it. Signed-off-by: Jakub Sitnicki Signed-off-by: Martin KaFai Lau Link: https://patch.msgid.link/20251105-skb-meta-rx-path-v4-9-5ceb08a9b37b@cloudflare.com --- net/core/filter.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/net/core/filter.c b/net/core/filter.c index 50775c01c456..4124becf8604 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -3875,6 +3875,7 @@ static const struct bpf_func_proto sk_skb_change_tail_proto = { static inline int __bpf_skb_change_head(struct sk_buff *skb, u32 head_room, u64 flags) { + const u8 meta_len = skb_metadata_len(skb); u32 max_len = BPF_SKB_MAX_LEN; u32 new_len = skb->len + head_room; int ret; @@ -3884,7 +3885,7 @@ static inline int __bpf_skb_change_head(struct sk_buff *skb, u32 head_room, new_len < skb->len)) return -EINVAL; - ret = skb_cow(skb, head_room); + ret = skb_cow(skb, meta_len + head_room); if (likely(!ret)) { /* Idea for this helper is that we currently only * allow to expand on mac header. This means that @@ -3896,6 +3897,7 @@ static inline int __bpf_skb_change_head(struct sk_buff *skb, u32 head_room, * for redirection into L2 device. */ __skb_push(skb, head_room); + skb_postpush_data_move(skb, head_room, 0); memset(skb->data, 0, head_room); skb_reset_mac_header(skb); skb_reset_mac_len(skb); From 967534e57c4439ba43b31f4af4cb85e84c86e6b7 Mon Sep 17 00:00:00 2001 From: Jakub Sitnicki Date: Wed, 5 Nov 2025 21:19:47 +0100 Subject: [PATCH 0660/1381] selftests/bpf: Verify skb metadata in BPF instead of userspace Move metadata verification into the BPF TC programs. Previously, userspace read metadata from a map and verified it once at test end. Now TC programs compare metadata directly using __builtin_memcmp() and set a test_pass flag. This enables verification at multiple points during test execution rather than a single final check. Signed-off-by: Jakub Sitnicki Signed-off-by: Martin KaFai Lau Link: https://patch.msgid.link/20251105-skb-meta-rx-path-v4-10-5ceb08a9b37b@cloudflare.com --- .../bpf/prog_tests/xdp_context_test_run.c | 52 +++-------- .../selftests/bpf/progs/test_xdp_meta.c | 88 +++++++++---------- 2 files changed, 57 insertions(+), 83 deletions(-) diff --git a/tools/testing/selftests/bpf/prog_tests/xdp_context_test_run.c b/tools/testing/selftests/bpf/prog_tests/xdp_context_test_run.c index 178292d1251a..93a1fbe6a4fd 100644 --- a/tools/testing/selftests/bpf/prog_tests/xdp_context_test_run.c +++ b/tools/testing/selftests/bpf/prog_tests/xdp_context_test_run.c @@ -171,33 +171,6 @@ static int write_test_packet(int tap_fd) return 0; } -static void assert_test_result(const struct bpf_map *result_map) -{ - int err; - __u32 map_key = 0; - __u8 map_value[TEST_PAYLOAD_LEN]; - - err = bpf_map__lookup_elem(result_map, &map_key, sizeof(map_key), - &map_value, TEST_PAYLOAD_LEN, BPF_ANY); - if (!ASSERT_OK(err, "lookup test_result")) - return; - - ASSERT_MEMEQ(&map_value, &test_payload, TEST_PAYLOAD_LEN, - "test_result map contains test payload"); -} - -static bool clear_test_result(struct bpf_map *result_map) -{ - const __u8 v[sizeof(test_payload)] = {}; - const __u32 k = 0; - int err; - - err = bpf_map__update_elem(result_map, &k, sizeof(k), v, sizeof(v), BPF_ANY); - ASSERT_OK(err, "update test_result"); - - return err == 0; -} - void test_xdp_context_veth(void) { LIBBPF_OPTS(bpf_tc_hook, tc_hook, .attach_point = BPF_TC_INGRESS); @@ -270,11 +243,13 @@ void test_xdp_context_veth(void) if (!ASSERT_GE(tx_ifindex, 0, "if_nametoindex tx")) goto close; + skel->bss->test_pass = false; + ret = send_test_packet(tx_ifindex); if (!ASSERT_OK(ret, "send_test_packet")) goto close; - assert_test_result(skel->maps.test_result); + ASSERT_TRUE(skel->bss->test_pass, "test_pass"); close: close_netns(nstoken); @@ -286,7 +261,7 @@ void test_xdp_context_veth(void) static void test_tuntap(struct bpf_program *xdp_prog, struct bpf_program *tc_prio_1_prog, struct bpf_program *tc_prio_2_prog, - struct bpf_map *result_map) + bool *test_pass) { LIBBPF_OPTS(bpf_tc_hook, tc_hook, .attach_point = BPF_TC_INGRESS); LIBBPF_OPTS(bpf_tc_opts, tc_opts, .handle = 1, .priority = 1); @@ -295,8 +270,7 @@ static void test_tuntap(struct bpf_program *xdp_prog, int tap_ifindex; int ret; - if (!clear_test_result(result_map)) - return; + *test_pass = false; ns = netns_new(TAP_NETNS, true); if (!ASSERT_OK_PTR(ns, "create and open ns")) @@ -340,7 +314,7 @@ static void test_tuntap(struct bpf_program *xdp_prog, if (!ASSERT_OK(ret, "write_test_packet")) goto close; - assert_test_result(result_map); + ASSERT_TRUE(*test_pass, "test_pass"); close: if (tap_fd >= 0) @@ -431,37 +405,37 @@ void test_xdp_context_tuntap(void) test_tuntap(skel->progs.ing_xdp, skel->progs.ing_cls, NULL, /* tc prio 2 */ - skel->maps.test_result); + &skel->bss->test_pass); if (test__start_subtest("dynptr_read")) test_tuntap(skel->progs.ing_xdp, skel->progs.ing_cls_dynptr_read, NULL, /* tc prio 2 */ - skel->maps.test_result); + &skel->bss->test_pass); if (test__start_subtest("dynptr_slice")) test_tuntap(skel->progs.ing_xdp, skel->progs.ing_cls_dynptr_slice, NULL, /* tc prio 2 */ - skel->maps.test_result); + &skel->bss->test_pass); if (test__start_subtest("dynptr_write")) test_tuntap(skel->progs.ing_xdp_zalloc_meta, skel->progs.ing_cls_dynptr_write, skel->progs.ing_cls_dynptr_read, - skel->maps.test_result); + &skel->bss->test_pass); if (test__start_subtest("dynptr_slice_rdwr")) test_tuntap(skel->progs.ing_xdp_zalloc_meta, skel->progs.ing_cls_dynptr_slice_rdwr, skel->progs.ing_cls_dynptr_slice, - skel->maps.test_result); + &skel->bss->test_pass); if (test__start_subtest("dynptr_offset")) test_tuntap(skel->progs.ing_xdp_zalloc_meta, skel->progs.ing_cls_dynptr_offset_wr, skel->progs.ing_cls_dynptr_offset_rd, - skel->maps.test_result); + &skel->bss->test_pass); if (test__start_subtest("dynptr_offset_oob")) test_tuntap(skel->progs.ing_xdp, skel->progs.ing_cls_dynptr_offset_oob, skel->progs.ing_cls, - skel->maps.test_result); + &skel->bss->test_pass); if (test__start_subtest("clone_data_meta_empty_on_data_write")) test_tuntap_mirred(skel->progs.ing_xdp, skel->progs.clone_data_meta_empty_on_data_write, diff --git a/tools/testing/selftests/bpf/progs/test_xdp_meta.c b/tools/testing/selftests/bpf/progs/test_xdp_meta.c index d79cb74b571e..11288b20f56c 100644 --- a/tools/testing/selftests/bpf/progs/test_xdp_meta.c +++ b/tools/testing/selftests/bpf/progs/test_xdp_meta.c @@ -11,37 +11,36 @@ #define ctx_ptr(ctx, mem) (void *)(unsigned long)ctx->mem -/* Demonstrates how metadata can be passed from an XDP program to a TC program - * using bpf_xdp_adjust_meta. - * For the sake of testing the metadata support in drivers, the XDP program uses - * a fixed-size payload after the Ethernet header as metadata. The TC program - * copies the metadata it receives into a map so it can be checked from - * userspace. +/* Demonstrate passing metadata from XDP to TC using bpf_xdp_adjust_meta. + * + * The XDP program extracts a fixed-size payload following the Ethernet header + * and stores it as packet metadata to test the driver's metadata support. The + * TC program then verifies if the passed metadata is correct. */ -struct { - __uint(type, BPF_MAP_TYPE_ARRAY); - __uint(max_entries, 1); - __type(key, __u32); - __uint(value_size, META_SIZE); -} test_result SEC(".maps"); - bool test_pass; +static const __u8 meta_want[META_SIZE] = { + 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, 0x08, + 0x11, 0x12, 0x13, 0x14, 0x15, 0x16, 0x17, 0x18, + 0x21, 0x22, 0x23, 0x24, 0x25, 0x26, 0x27, 0x28, + 0x31, 0x32, 0x33, 0x34, 0x35, 0x36, 0x37, 0x38, +}; + SEC("tc") int ing_cls(struct __sk_buff *ctx) { - __u8 *data, *data_meta; - __u32 key = 0; + __u8 *meta_have = ctx_ptr(ctx, data_meta); + __u8 *data = ctx_ptr(ctx, data); - data_meta = ctx_ptr(ctx, data_meta); - data = ctx_ptr(ctx, data); + if (meta_have + META_SIZE > data) + goto out; - if (data_meta + META_SIZE > data) - return TC_ACT_SHOT; - - bpf_map_update_elem(&test_result, &key, data_meta, BPF_ANY); + if (__builtin_memcmp(meta_want, meta_have, META_SIZE)) + goto out; + test_pass = true; +out: return TC_ACT_SHOT; } @@ -49,17 +48,17 @@ int ing_cls(struct __sk_buff *ctx) SEC("tc") int ing_cls_dynptr_read(struct __sk_buff *ctx) { + __u8 meta_have[META_SIZE]; struct bpf_dynptr meta; - const __u32 zero = 0; - __u8 *dst; - - dst = bpf_map_lookup_elem(&test_result, &zero); - if (!dst) - return TC_ACT_SHOT; bpf_dynptr_from_skb_meta(ctx, 0, &meta); - bpf_dynptr_read(dst, META_SIZE, &meta, 0, 0); + bpf_dynptr_read(meta_have, META_SIZE, &meta, 0, 0); + if (__builtin_memcmp(meta_want, meta_have, META_SIZE)) + goto out; + + test_pass = true; +out: return TC_ACT_SHOT; } @@ -86,20 +85,18 @@ SEC("tc") int ing_cls_dynptr_slice(struct __sk_buff *ctx) { struct bpf_dynptr meta; - const __u32 zero = 0; - __u8 *dst, *src; - - dst = bpf_map_lookup_elem(&test_result, &zero); - if (!dst) - return TC_ACT_SHOT; + __u8 *meta_have; bpf_dynptr_from_skb_meta(ctx, 0, &meta); - src = bpf_dynptr_slice(&meta, 0, NULL, META_SIZE); - if (!src) - return TC_ACT_SHOT; + meta_have = bpf_dynptr_slice(&meta, 0, NULL, META_SIZE); + if (!meta_have) + goto out; - __builtin_memcpy(dst, src, META_SIZE); + if (__builtin_memcmp(meta_want, meta_have, META_SIZE)) + goto out; + test_pass = true; +out: return TC_ACT_SHOT; } @@ -129,14 +126,12 @@ int ing_cls_dynptr_slice_rdwr(struct __sk_buff *ctx) SEC("tc") int ing_cls_dynptr_offset_rd(struct __sk_buff *ctx) { - struct bpf_dynptr meta; const __u32 chunk_len = META_SIZE / 4; - const __u32 zero = 0; + __u8 meta_have[META_SIZE]; + struct bpf_dynptr meta; __u8 *dst, *src; - dst = bpf_map_lookup_elem(&test_result, &zero); - if (!dst) - return TC_ACT_SHOT; + dst = meta_have; /* 1. Regular read */ bpf_dynptr_from_skb_meta(ctx, 0, &meta); @@ -155,9 +150,14 @@ int ing_cls_dynptr_offset_rd(struct __sk_buff *ctx) /* 4. Read from a slice starting at an offset */ src = bpf_dynptr_slice(&meta, 2 * chunk_len, NULL, chunk_len); if (!src) - return TC_ACT_SHOT; + goto out; __builtin_memcpy(dst, src, chunk_len); + if (__builtin_memcmp(meta_want, meta_have, META_SIZE)) + goto out; + + test_pass = true; +out: return TC_ACT_SHOT; } From 9ef9ac15a527739135548b87053f4646099e4bd6 Mon Sep 17 00:00:00 2001 From: Jakub Sitnicki Date: Wed, 5 Nov 2025 21:19:48 +0100 Subject: [PATCH 0661/1381] selftests/bpf: Dump skb metadata on verification failure Add diagnostic output when metadata verification fails to help with troubleshooting test failures. Introduce a check_metadata() helper that prints both expected and received metadata to the BPF program's stderr stream on mismatch. The userspace test reads and dumps this stream on failure. Signed-off-by: Jakub Sitnicki Signed-off-by: Martin KaFai Lau Link: https://patch.msgid.link/20251105-skb-meta-rx-path-v4-11-5ceb08a9b37b@cloudflare.com --- .../bpf/prog_tests/xdp_context_test_run.c | 24 +++++++++++++++--- .../selftests/bpf/progs/test_xdp_meta.c | 25 ++++++++++++++++--- 2 files changed, 42 insertions(+), 7 deletions(-) diff --git a/tools/testing/selftests/bpf/prog_tests/xdp_context_test_run.c b/tools/testing/selftests/bpf/prog_tests/xdp_context_test_run.c index 93a1fbe6a4fd..db3027564261 100644 --- a/tools/testing/selftests/bpf/prog_tests/xdp_context_test_run.c +++ b/tools/testing/selftests/bpf/prog_tests/xdp_context_test_run.c @@ -171,6 +171,21 @@ static int write_test_packet(int tap_fd) return 0; } +static void dump_err_stream(const struct bpf_program *prog) +{ + char buf[512]; + int ret; + + ret = 0; + do { + ret = bpf_prog_stream_read(bpf_program__fd(prog), + BPF_STREAM_STDERR, buf, sizeof(buf), + NULL); + if (ret > 0) + fwrite(buf, sizeof(buf[0]), ret, stderr); + } while (ret > 0); +} + void test_xdp_context_veth(void) { LIBBPF_OPTS(bpf_tc_hook, tc_hook, .attach_point = BPF_TC_INGRESS); @@ -249,7 +264,8 @@ void test_xdp_context_veth(void) if (!ASSERT_OK(ret, "send_test_packet")) goto close; - ASSERT_TRUE(skel->bss->test_pass, "test_pass"); + if (!ASSERT_TRUE(skel->bss->test_pass, "test_pass")) + dump_err_stream(tc_prog); close: close_netns(nstoken); @@ -314,7 +330,8 @@ static void test_tuntap(struct bpf_program *xdp_prog, if (!ASSERT_OK(ret, "write_test_packet")) goto close; - ASSERT_TRUE(*test_pass, "test_pass"); + if (!ASSERT_TRUE(*test_pass, "test_pass")) + dump_err_stream(tc_prio_2_prog ? : tc_prio_1_prog); close: if (tap_fd >= 0) @@ -385,7 +402,8 @@ static void test_tuntap_mirred(struct bpf_program *xdp_prog, if (!ASSERT_OK(ret, "write_test_packet")) goto close; - ASSERT_TRUE(*test_pass, "test_pass"); + if (!ASSERT_TRUE(*test_pass, "test_pass")) + dump_err_stream(tc_prog); close: if (tap_fd >= 0) diff --git a/tools/testing/selftests/bpf/progs/test_xdp_meta.c b/tools/testing/selftests/bpf/progs/test_xdp_meta.c index 11288b20f56c..3b137c4eed6c 100644 --- a/tools/testing/selftests/bpf/progs/test_xdp_meta.c +++ b/tools/testing/selftests/bpf/progs/test_xdp_meta.c @@ -27,6 +27,23 @@ static const __u8 meta_want[META_SIZE] = { 0x31, 0x32, 0x33, 0x34, 0x35, 0x36, 0x37, 0x38, }; +static bool check_metadata(const char *file, int line, __u8 *meta_have) +{ + if (!__builtin_memcmp(meta_have, meta_want, META_SIZE)) + return true; + + bpf_stream_printk(BPF_STREAM_STDERR, + "FAIL:%s:%d: metadata mismatch\n" + " have:\n %pI6\n %pI6\n" + " want:\n %pI6\n %pI6\n", + file, line, + &meta_have[0x00], &meta_have[0x10], + &meta_want[0x00], &meta_want[0x10]); + return false; +} + +#define check_metadata(meta_have) check_metadata(__FILE__, __LINE__, meta_have) + SEC("tc") int ing_cls(struct __sk_buff *ctx) { @@ -36,7 +53,7 @@ int ing_cls(struct __sk_buff *ctx) if (meta_have + META_SIZE > data) goto out; - if (__builtin_memcmp(meta_want, meta_have, META_SIZE)) + if (!check_metadata(meta_have)) goto out; test_pass = true; @@ -54,7 +71,7 @@ int ing_cls_dynptr_read(struct __sk_buff *ctx) bpf_dynptr_from_skb_meta(ctx, 0, &meta); bpf_dynptr_read(meta_have, META_SIZE, &meta, 0, 0); - if (__builtin_memcmp(meta_want, meta_have, META_SIZE)) + if (!check_metadata(meta_have)) goto out; test_pass = true; @@ -92,7 +109,7 @@ int ing_cls_dynptr_slice(struct __sk_buff *ctx) if (!meta_have) goto out; - if (__builtin_memcmp(meta_want, meta_have, META_SIZE)) + if (!check_metadata(meta_have)) goto out; test_pass = true; @@ -153,7 +170,7 @@ int ing_cls_dynptr_offset_rd(struct __sk_buff *ctx) goto out; __builtin_memcpy(dst, src, chunk_len); - if (__builtin_memcmp(meta_want, meta_have, META_SIZE)) + if (!check_metadata(meta_have)) goto out; test_pass = true; From 1e1357fde808a35c6069759298660134e5dab053 Mon Sep 17 00:00:00 2001 From: Jakub Sitnicki Date: Wed, 5 Nov 2025 21:19:49 +0100 Subject: [PATCH 0662/1381] selftests/bpf: Expect unclone to preserve skb metadata Since pskb_expand_head() no longer clears metadata on unclone, update tests for cloned packets to expect metadata to remain intact. Also simplify the clone_dynptr_kept_on_{data,meta}_slice_write tests. Creating an r/w dynptr slice is sufficient to trigger an unclone in the prologue, so remove the extraneous writes to the data/meta slice. Signed-off-by: Jakub Sitnicki Signed-off-by: Martin KaFai Lau Link: https://patch.msgid.link/20251105-skb-meta-rx-path-v4-12-5ceb08a9b37b@cloudflare.com --- .../bpf/prog_tests/xdp_context_test_run.c | 24 ++-- .../selftests/bpf/progs/test_xdp_meta.c | 118 ++++++++++-------- 2 files changed, 79 insertions(+), 63 deletions(-) diff --git a/tools/testing/selftests/bpf/prog_tests/xdp_context_test_run.c b/tools/testing/selftests/bpf/prog_tests/xdp_context_test_run.c index db3027564261..a129c3057202 100644 --- a/tools/testing/selftests/bpf/prog_tests/xdp_context_test_run.c +++ b/tools/testing/selftests/bpf/prog_tests/xdp_context_test_run.c @@ -454,29 +454,29 @@ void test_xdp_context_tuntap(void) skel->progs.ing_cls_dynptr_offset_oob, skel->progs.ing_cls, &skel->bss->test_pass); - if (test__start_subtest("clone_data_meta_empty_on_data_write")) + if (test__start_subtest("clone_data_meta_survives_data_write")) test_tuntap_mirred(skel->progs.ing_xdp, - skel->progs.clone_data_meta_empty_on_data_write, + skel->progs.clone_data_meta_survives_data_write, &skel->bss->test_pass); - if (test__start_subtest("clone_data_meta_empty_on_meta_write")) + if (test__start_subtest("clone_data_meta_survives_meta_write")) test_tuntap_mirred(skel->progs.ing_xdp, - skel->progs.clone_data_meta_empty_on_meta_write, + skel->progs.clone_data_meta_survives_meta_write, &skel->bss->test_pass); - if (test__start_subtest("clone_dynptr_empty_on_data_slice_write")) + if (test__start_subtest("clone_meta_dynptr_survives_data_slice_write")) test_tuntap_mirred(skel->progs.ing_xdp, - skel->progs.clone_dynptr_empty_on_data_slice_write, + skel->progs.clone_meta_dynptr_survives_data_slice_write, &skel->bss->test_pass); - if (test__start_subtest("clone_dynptr_empty_on_meta_slice_write")) + if (test__start_subtest("clone_meta_dynptr_survives_meta_slice_write")) test_tuntap_mirred(skel->progs.ing_xdp, - skel->progs.clone_dynptr_empty_on_meta_slice_write, + skel->progs.clone_meta_dynptr_survives_meta_slice_write, &skel->bss->test_pass); - if (test__start_subtest("clone_dynptr_rdonly_before_data_dynptr_write")) + if (test__start_subtest("clone_meta_dynptr_rw_before_data_dynptr_write")) test_tuntap_mirred(skel->progs.ing_xdp, - skel->progs.clone_dynptr_rdonly_before_data_dynptr_write, + skel->progs.clone_meta_dynptr_rw_before_data_dynptr_write, &skel->bss->test_pass); - if (test__start_subtest("clone_dynptr_rdonly_before_meta_dynptr_write")) + if (test__start_subtest("clone_meta_dynptr_rw_before_meta_dynptr_write")) test_tuntap_mirred(skel->progs.ing_xdp, - skel->progs.clone_dynptr_rdonly_before_meta_dynptr_write, + skel->progs.clone_meta_dynptr_rw_before_meta_dynptr_write, &skel->bss->test_pass); test_xdp_meta__destroy(skel); diff --git a/tools/testing/selftests/bpf/progs/test_xdp_meta.c b/tools/testing/selftests/bpf/progs/test_xdp_meta.c index 3b137c4eed6c..a70de55c6997 100644 --- a/tools/testing/selftests/bpf/progs/test_xdp_meta.c +++ b/tools/testing/selftests/bpf/progs/test_xdp_meta.c @@ -321,12 +321,13 @@ int ing_xdp(struct xdp_md *ctx) } /* - * Check that skb->data_meta..skb->data is empty if prog writes to packet - * _payload_ using packet pointers. Applies only to cloned skbs. + * Check that, when operating on a cloned packet, skb->data_meta..skb->data is + * kept intact if prog writes to packet _payload_ using packet pointers. */ SEC("tc") -int clone_data_meta_empty_on_data_write(struct __sk_buff *ctx) +int clone_data_meta_survives_data_write(struct __sk_buff *ctx) { + __u8 *meta_have = ctx_ptr(ctx, data_meta); struct ethhdr *eth = ctx_ptr(ctx, data); if (eth + 1 > ctx_ptr(ctx, data_end)) @@ -335,8 +336,10 @@ int clone_data_meta_empty_on_data_write(struct __sk_buff *ctx) if (eth->h_proto != 0) goto out; - /* Expect no metadata */ - if (ctx->data_meta != ctx->data) + if (meta_have + META_SIZE > eth) + goto out; + + if (!check_metadata(meta_have)) goto out; /* Packet write to trigger unclone in prologue */ @@ -348,14 +351,14 @@ int clone_data_meta_empty_on_data_write(struct __sk_buff *ctx) } /* - * Check that skb->data_meta..skb->data is empty if prog writes to packet - * _metadata_ using packet pointers. Applies only to cloned skbs. + * Check that, when operating on a cloned packet, skb->data_meta..skb->data is + * kept intact if prog writes to packet _metadata_ using packet pointers. */ SEC("tc") -int clone_data_meta_empty_on_meta_write(struct __sk_buff *ctx) +int clone_data_meta_survives_meta_write(struct __sk_buff *ctx) { + __u8 *meta_have = ctx_ptr(ctx, data_meta); struct ethhdr *eth = ctx_ptr(ctx, data); - __u8 *md = ctx_ptr(ctx, data_meta); if (eth + 1 > ctx_ptr(ctx, data_end)) goto out; @@ -363,25 +366,29 @@ int clone_data_meta_empty_on_meta_write(struct __sk_buff *ctx) if (eth->h_proto != 0) goto out; - if (md + 1 > ctx_ptr(ctx, data)) { - /* Expect no metadata */ - test_pass = true; - } else { - /* Metadata write to trigger unclone in prologue */ - *md = 42; - } + if (meta_have + META_SIZE > eth) + goto out; + + if (!check_metadata(meta_have)) + goto out; + + /* Metadata write to trigger unclone in prologue */ + *meta_have = 42; + + test_pass = true; out: return TC_ACT_SHOT; } /* - * Check that skb_meta dynptr is writable but empty if prog writes to packet - * _payload_ using a dynptr slice. Applies only to cloned skbs. + * Check that, when operating on a cloned packet, metadata remains intact if + * prog creates a r/w slice to packet _payload_. */ SEC("tc") -int clone_dynptr_empty_on_data_slice_write(struct __sk_buff *ctx) +int clone_meta_dynptr_survives_data_slice_write(struct __sk_buff *ctx) { struct bpf_dynptr data, meta; + __u8 meta_have[META_SIZE]; struct ethhdr *eth; bpf_dynptr_from_skb(ctx, 0, &data); @@ -392,29 +399,26 @@ int clone_dynptr_empty_on_data_slice_write(struct __sk_buff *ctx) if (eth->h_proto != 0) goto out; - /* Expect no metadata */ bpf_dynptr_from_skb_meta(ctx, 0, &meta); - if (bpf_dynptr_is_rdonly(&meta) || bpf_dynptr_size(&meta) > 0) + bpf_dynptr_read(meta_have, META_SIZE, &meta, 0, 0); + if (!check_metadata(meta_have)) goto out; - /* Packet write to trigger unclone in prologue */ - eth->h_proto = 42; - test_pass = true; out: return TC_ACT_SHOT; } /* - * Check that skb_meta dynptr is writable but empty if prog writes to packet - * _metadata_ using a dynptr slice. Applies only to cloned skbs. + * Check that, when operating on a cloned packet, metadata remains intact if + * prog creates an r/w slice to packet _metadata_. */ SEC("tc") -int clone_dynptr_empty_on_meta_slice_write(struct __sk_buff *ctx) +int clone_meta_dynptr_survives_meta_slice_write(struct __sk_buff *ctx) { struct bpf_dynptr data, meta; const struct ethhdr *eth; - __u8 *md; + __u8 *meta_have; bpf_dynptr_from_skb(ctx, 0, &data); eth = bpf_dynptr_slice(&data, 0, NULL, sizeof(*eth)); @@ -424,16 +428,13 @@ int clone_dynptr_empty_on_meta_slice_write(struct __sk_buff *ctx) if (eth->h_proto != 0) goto out; - /* Expect no metadata */ bpf_dynptr_from_skb_meta(ctx, 0, &meta); - if (bpf_dynptr_is_rdonly(&meta) || bpf_dynptr_size(&meta) > 0) + meta_have = bpf_dynptr_slice_rdwr(&meta, 0, NULL, META_SIZE); + if (!meta_have) goto out; - /* Metadata write to trigger unclone in prologue */ - bpf_dynptr_from_skb_meta(ctx, 0, &meta); - md = bpf_dynptr_slice_rdwr(&meta, 0, NULL, sizeof(*md)); - if (md) - *md = 42; + if (!check_metadata(meta_have)) + goto out; test_pass = true; out: @@ -441,14 +442,17 @@ int clone_dynptr_empty_on_meta_slice_write(struct __sk_buff *ctx) } /* - * Check that skb_meta dynptr is read-only before prog writes to packet payload - * using dynptr_write helper. Applies only to cloned skbs. + * Check that, when operating on a cloned packet, skb_meta dynptr is read-write + * before prog writes to packet _payload_ using dynptr_write helper and metadata + * remains intact before and after the write. */ SEC("tc") -int clone_dynptr_rdonly_before_data_dynptr_write(struct __sk_buff *ctx) +int clone_meta_dynptr_rw_before_data_dynptr_write(struct __sk_buff *ctx) { struct bpf_dynptr data, meta; + __u8 meta_have[META_SIZE]; const struct ethhdr *eth; + int err; bpf_dynptr_from_skb(ctx, 0, &data); eth = bpf_dynptr_slice(&data, 0, NULL, sizeof(*eth)); @@ -458,17 +462,20 @@ int clone_dynptr_rdonly_before_data_dynptr_write(struct __sk_buff *ctx) if (eth->h_proto != 0) goto out; - /* Expect read-only metadata before unclone */ + /* Expect read-write metadata before unclone */ bpf_dynptr_from_skb_meta(ctx, 0, &meta); - if (!bpf_dynptr_is_rdonly(&meta) || bpf_dynptr_size(&meta) != META_SIZE) + if (bpf_dynptr_is_rdonly(&meta)) + goto out; + + err = bpf_dynptr_read(meta_have, META_SIZE, &meta, 0, 0); + if (err || !check_metadata(meta_have)) goto out; /* Helper write to payload will unclone the packet */ bpf_dynptr_write(&data, offsetof(struct ethhdr, h_proto), "x", 1, 0); - /* Expect no metadata after unclone */ - bpf_dynptr_from_skb_meta(ctx, 0, &meta); - if (bpf_dynptr_is_rdonly(&meta) || bpf_dynptr_size(&meta) != 0) + err = bpf_dynptr_read(meta_have, META_SIZE, &meta, 0, 0); + if (err || !check_metadata(meta_have)) goto out; test_pass = true; @@ -477,14 +484,17 @@ int clone_dynptr_rdonly_before_data_dynptr_write(struct __sk_buff *ctx) } /* - * Check that skb_meta dynptr is read-only if prog writes to packet - * metadata using dynptr_write helper. Applies only to cloned skbs. + * Check that, when operating on a cloned packet, skb_meta dynptr is read-write + * before prog writes to packet _metadata_ using dynptr_write helper and + * metadata remains intact before and after the write. */ SEC("tc") -int clone_dynptr_rdonly_before_meta_dynptr_write(struct __sk_buff *ctx) +int clone_meta_dynptr_rw_before_meta_dynptr_write(struct __sk_buff *ctx) { struct bpf_dynptr data, meta; + __u8 meta_have[META_SIZE]; const struct ethhdr *eth; + int err; bpf_dynptr_from_skb(ctx, 0, &data); eth = bpf_dynptr_slice(&data, 0, NULL, sizeof(*eth)); @@ -494,14 +504,20 @@ int clone_dynptr_rdonly_before_meta_dynptr_write(struct __sk_buff *ctx) if (eth->h_proto != 0) goto out; - /* Expect read-only metadata */ + /* Expect read-write metadata before unclone */ bpf_dynptr_from_skb_meta(ctx, 0, &meta); - if (!bpf_dynptr_is_rdonly(&meta) || bpf_dynptr_size(&meta) != META_SIZE) + if (bpf_dynptr_is_rdonly(&meta)) goto out; - /* Metadata write. Expect failure. */ - bpf_dynptr_from_skb_meta(ctx, 0, &meta); - if (bpf_dynptr_write(&meta, 0, "x", 1, 0) != -EINVAL) + err = bpf_dynptr_read(meta_have, META_SIZE, &meta, 0, 0); + if (err || !check_metadata(meta_have)) + goto out; + + /* Helper write to metadata will unclone the packet */ + bpf_dynptr_write(&meta, 0, &meta_have[0], 1, 0); + + err = bpf_dynptr_read(meta_have, META_SIZE, &meta, 0, 0); + if (err || !check_metadata(meta_have)) goto out; test_pass = true; From 354d020c29f72513dce4f3902890158d99b67b67 Mon Sep 17 00:00:00 2001 From: Jakub Sitnicki Date: Wed, 5 Nov 2025 21:19:50 +0100 Subject: [PATCH 0663/1381] selftests/bpf: Cover skb metadata access after vlan push/pop helper Add a test to verify that skb metadata remains accessible after calling bpf_skb_vlan_push() and bpf_skb_vlan_pop(), which modify the packet headroom. Signed-off-by: Jakub Sitnicki Signed-off-by: Martin KaFai Lau Link: https://patch.msgid.link/20251105-skb-meta-rx-path-v4-13-5ceb08a9b37b@cloudflare.com --- .../bpf/prog_tests/xdp_context_test_run.c | 6 +++ .../selftests/bpf/progs/test_xdp_meta.c | 43 +++++++++++++++++++ 2 files changed, 49 insertions(+) diff --git a/tools/testing/selftests/bpf/prog_tests/xdp_context_test_run.c b/tools/testing/selftests/bpf/prog_tests/xdp_context_test_run.c index a129c3057202..97c8f876f673 100644 --- a/tools/testing/selftests/bpf/prog_tests/xdp_context_test_run.c +++ b/tools/testing/selftests/bpf/prog_tests/xdp_context_test_run.c @@ -478,6 +478,12 @@ void test_xdp_context_tuntap(void) test_tuntap_mirred(skel->progs.ing_xdp, skel->progs.clone_meta_dynptr_rw_before_meta_dynptr_write, &skel->bss->test_pass); + /* Tests for BPF helpers which touch headroom */ + if (test__start_subtest("helper_skb_vlan_push_pop")) + test_tuntap(skel->progs.ing_xdp, + skel->progs.helper_skb_vlan_push_pop, + NULL, /* tc prio 2 */ + &skel->bss->test_pass); test_xdp_meta__destroy(skel); } diff --git a/tools/testing/selftests/bpf/progs/test_xdp_meta.c b/tools/testing/selftests/bpf/progs/test_xdp_meta.c index a70de55c6997..04c7487bb350 100644 --- a/tools/testing/selftests/bpf/progs/test_xdp_meta.c +++ b/tools/testing/selftests/bpf/progs/test_xdp_meta.c @@ -44,6 +44,16 @@ static bool check_metadata(const char *file, int line, __u8 *meta_have) #define check_metadata(meta_have) check_metadata(__FILE__, __LINE__, meta_have) +static bool check_skb_metadata(const char *file, int line, struct __sk_buff *skb) +{ + __u8 *data_meta = ctx_ptr(skb, data_meta); + __u8 *data = ctx_ptr(skb, data); + + return data_meta + META_SIZE <= data && (check_metadata)(file, line, data_meta); +} + +#define check_skb_metadata(skb) check_skb_metadata(__FILE__, __LINE__, skb) + SEC("tc") int ing_cls(struct __sk_buff *ctx) { @@ -525,4 +535,37 @@ int clone_meta_dynptr_rw_before_meta_dynptr_write(struct __sk_buff *ctx) return TC_ACT_SHOT; } +SEC("tc") +int helper_skb_vlan_push_pop(struct __sk_buff *ctx) +{ + int err; + + /* bpf_skb_vlan_push assumes HW offload for primary VLAN tag. Only + * secondary tag push triggers an actual MAC header modification. + */ + err = bpf_skb_vlan_push(ctx, 0, 42); + if (err) + goto out; + err = bpf_skb_vlan_push(ctx, 0, 207); + if (err) + goto out; + + if (!check_skb_metadata(ctx)) + goto out; + + err = bpf_skb_vlan_pop(ctx); + if (err) + goto out; + err = bpf_skb_vlan_pop(ctx); + if (err) + goto out; + + if (!check_skb_metadata(ctx)) + goto out; + + test_pass = true; +out: + return TC_ACT_SHOT; +} + char _license[] SEC("license") = "GPL"; From 29960e635b01b148e6db5a84957f99423fd85464 Mon Sep 17 00:00:00 2001 From: Jakub Sitnicki Date: Wed, 5 Nov 2025 21:19:51 +0100 Subject: [PATCH 0664/1381] selftests/bpf: Cover skb metadata access after bpf_skb_adjust_room Add a test to verify that skb metadata remains accessible after calling bpf_skb_adjust_room(), which modifies the packet headroom and can trigger head reallocation. The helper expects an Ethernet frame carrying an IP packet so switch test packet identification by source MAC address since we can no longer rely on Ethernet proto being set to zero. Signed-off-by: Jakub Sitnicki Signed-off-by: Martin KaFai Lau Link: https://patch.msgid.link/20251105-skb-meta-rx-path-v4-14-5ceb08a9b37b@cloudflare.com --- .../bpf/prog_tests/xdp_context_test_run.c | 25 ++++++-- .../selftests/bpf/progs/test_xdp_meta.c | 61 ++++++++++++++++--- 2 files changed, 71 insertions(+), 15 deletions(-) diff --git a/tools/testing/selftests/bpf/prog_tests/xdp_context_test_run.c b/tools/testing/selftests/bpf/prog_tests/xdp_context_test_run.c index 97c8f876f673..a3b82cf2f9e9 100644 --- a/tools/testing/selftests/bpf/prog_tests/xdp_context_test_run.c +++ b/tools/testing/selftests/bpf/prog_tests/xdp_context_test_run.c @@ -124,10 +124,10 @@ static int send_test_packet(int ifindex) int n, sock = -1; __u8 packet[sizeof(struct ethhdr) + TEST_PAYLOAD_LEN]; - /* The ethernet header is not relevant for this test and doesn't need to - * be meaningful. - */ - struct ethhdr eth = { 0 }; + /* We use the Ethernet header only to identify the test packet */ + struct ethhdr eth = { + .h_source = { 0x12, 0x34, 0xDE, 0xAD, 0xBE, 0xEF }, + }; memcpy(packet, ð, sizeof(eth)); memcpy(packet + sizeof(eth), test_payload, TEST_PAYLOAD_LEN); @@ -160,8 +160,16 @@ static int write_test_packet(int tap_fd) __u8 packet[sizeof(struct ethhdr) + TEST_PAYLOAD_LEN]; int n; - /* The ethernet header doesn't need to be valid for this test */ - memset(packet, 0, sizeof(struct ethhdr)); + /* The Ethernet header is mostly not relevant. We use it to identify the + * test packet and some BPF helpers we exercise expect to operate on + * Ethernet frames carrying IP packets. Pretend that's the case. + */ + struct ethhdr eth = { + .h_source = { 0x12, 0x34, 0xDE, 0xAD, 0xBE, 0xEF }, + .h_proto = htons(ETH_P_IP), + }; + + memcpy(packet, ð, sizeof(eth)); memcpy(packet + sizeof(struct ethhdr), test_payload, TEST_PAYLOAD_LEN); n = write(tap_fd, packet, sizeof(packet)); @@ -484,6 +492,11 @@ void test_xdp_context_tuntap(void) skel->progs.helper_skb_vlan_push_pop, NULL, /* tc prio 2 */ &skel->bss->test_pass); + if (test__start_subtest("helper_skb_adjust_room")) + test_tuntap(skel->progs.ing_xdp, + skel->progs.helper_skb_adjust_room, + NULL, /* tc prio 2 */ + &skel->bss->test_pass); test_xdp_meta__destroy(skel); } diff --git a/tools/testing/selftests/bpf/progs/test_xdp_meta.c b/tools/testing/selftests/bpf/progs/test_xdp_meta.c index 04c7487bb350..6edc84d8dc52 100644 --- a/tools/testing/selftests/bpf/progs/test_xdp_meta.c +++ b/tools/testing/selftests/bpf/progs/test_xdp_meta.c @@ -20,6 +20,10 @@ bool test_pass; +static const __u8 smac_want[ETH_ALEN] = { + 0x12, 0x34, 0xDE, 0xAD, 0xBE, 0xEF, +}; + static const __u8 meta_want[META_SIZE] = { 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, 0x08, 0x11, 0x12, 0x13, 0x14, 0x15, 0x16, 0x17, 0x18, @@ -27,6 +31,11 @@ static const __u8 meta_want[META_SIZE] = { 0x31, 0x32, 0x33, 0x34, 0x35, 0x36, 0x37, 0x38, }; +static bool check_smac(const struct ethhdr *eth) +{ + return !__builtin_memcmp(eth->h_source, smac_want, ETH_ALEN); +} + static bool check_metadata(const char *file, int line, __u8 *meta_have) { if (!__builtin_memcmp(meta_have, meta_want, META_SIZE)) @@ -281,7 +290,7 @@ int ing_xdp_zalloc_meta(struct xdp_md *ctx) /* Drop any non-test packets */ if (eth + 1 > ctx_ptr(ctx, data_end)) return XDP_DROP; - if (eth->h_proto != 0) + if (!check_smac(eth)) return XDP_DROP; ret = bpf_xdp_adjust_meta(ctx, -META_SIZE); @@ -321,9 +330,9 @@ int ing_xdp(struct xdp_md *ctx) /* The Linux networking stack may send other packets on the test * interface that interfere with the test. Just drop them. - * The test packets can be recognized by their ethertype of zero. + * The test packets can be recognized by their source MAC address. */ - if (eth->h_proto != 0) + if (!check_smac(eth)) return XDP_DROP; __builtin_memcpy(data_meta, payload, META_SIZE); @@ -343,7 +352,7 @@ int clone_data_meta_survives_data_write(struct __sk_buff *ctx) if (eth + 1 > ctx_ptr(ctx, data_end)) goto out; /* Ignore non-test packets */ - if (eth->h_proto != 0) + if (!check_smac(eth)) goto out; if (meta_have + META_SIZE > eth) @@ -373,7 +382,7 @@ int clone_data_meta_survives_meta_write(struct __sk_buff *ctx) if (eth + 1 > ctx_ptr(ctx, data_end)) goto out; /* Ignore non-test packets */ - if (eth->h_proto != 0) + if (!check_smac(eth)) goto out; if (meta_have + META_SIZE > eth) @@ -406,7 +415,7 @@ int clone_meta_dynptr_survives_data_slice_write(struct __sk_buff *ctx) if (!eth) goto out; /* Ignore non-test packets */ - if (eth->h_proto != 0) + if (!check_smac(eth)) goto out; bpf_dynptr_from_skb_meta(ctx, 0, &meta); @@ -435,7 +444,7 @@ int clone_meta_dynptr_survives_meta_slice_write(struct __sk_buff *ctx) if (!eth) goto out; /* Ignore non-test packets */ - if (eth->h_proto != 0) + if (!check_smac(eth)) goto out; bpf_dynptr_from_skb_meta(ctx, 0, &meta); @@ -469,7 +478,7 @@ int clone_meta_dynptr_rw_before_data_dynptr_write(struct __sk_buff *ctx) if (!eth) goto out; /* Ignore non-test packets */ - if (eth->h_proto != 0) + if (!check_smac(eth)) goto out; /* Expect read-write metadata before unclone */ @@ -511,7 +520,7 @@ int clone_meta_dynptr_rw_before_meta_dynptr_write(struct __sk_buff *ctx) if (!eth) goto out; /* Ignore non-test packets */ - if (eth->h_proto != 0) + if (!check_smac(eth)) goto out; /* Expect read-write metadata before unclone */ @@ -568,4 +577,38 @@ int helper_skb_vlan_push_pop(struct __sk_buff *ctx) return TC_ACT_SHOT; } +SEC("tc") +int helper_skb_adjust_room(struct __sk_buff *ctx) +{ + int err; + + /* Grow a 1 byte hole after the MAC header */ + err = bpf_skb_adjust_room(ctx, 1, BPF_ADJ_ROOM_MAC, 0); + if (err) + goto out; + + if (!check_skb_metadata(ctx)) + goto out; + + /* Shrink a 1 byte hole after the MAC header */ + err = bpf_skb_adjust_room(ctx, -1, BPF_ADJ_ROOM_MAC, 0); + if (err) + goto out; + + if (!check_skb_metadata(ctx)) + goto out; + + /* Grow a 256 byte hole to trigger head reallocation */ + err = bpf_skb_adjust_room(ctx, 256, BPF_ADJ_ROOM_MAC, 0); + if (err) + goto out; + + if (!check_skb_metadata(ctx)) + goto out; + + test_pass = true; +out: + return TC_ACT_SHOT; +} + char _license[] SEC("license") = "GPL"; From 85d454afef612f880c69cb39a7d74772a24411d3 Mon Sep 17 00:00:00 2001 From: Jakub Sitnicki Date: Wed, 5 Nov 2025 21:19:52 +0100 Subject: [PATCH 0665/1381] selftests/bpf: Cover skb metadata access after change_head/tail helper Add a test to verify that skb metadata remains accessible after calling bpf_skb_change_head() and bpf_skb_change_tail(), which modify packet headroom/tailroom and can trigger head reallocation. Signed-off-by: Jakub Sitnicki Signed-off-by: Martin KaFai Lau Link: https://patch.msgid.link/20251105-skb-meta-rx-path-v4-15-5ceb08a9b37b@cloudflare.com --- .../bpf/prog_tests/xdp_context_test_run.c | 5 +++ .../selftests/bpf/progs/test_xdp_meta.c | 34 +++++++++++++++++++ 2 files changed, 39 insertions(+) diff --git a/tools/testing/selftests/bpf/prog_tests/xdp_context_test_run.c b/tools/testing/selftests/bpf/prog_tests/xdp_context_test_run.c index a3b82cf2f9e9..65735a134abb 100644 --- a/tools/testing/selftests/bpf/prog_tests/xdp_context_test_run.c +++ b/tools/testing/selftests/bpf/prog_tests/xdp_context_test_run.c @@ -497,6 +497,11 @@ void test_xdp_context_tuntap(void) skel->progs.helper_skb_adjust_room, NULL, /* tc prio 2 */ &skel->bss->test_pass); + if (test__start_subtest("helper_skb_change_head_tail")) + test_tuntap(skel->progs.ing_xdp, + skel->progs.helper_skb_change_head_tail, + NULL, /* tc prio 2 */ + &skel->bss->test_pass); test_xdp_meta__destroy(skel); } diff --git a/tools/testing/selftests/bpf/progs/test_xdp_meta.c b/tools/testing/selftests/bpf/progs/test_xdp_meta.c index 6edc84d8dc52..e0b2e8ed0cc5 100644 --- a/tools/testing/selftests/bpf/progs/test_xdp_meta.c +++ b/tools/testing/selftests/bpf/progs/test_xdp_meta.c @@ -611,4 +611,38 @@ int helper_skb_adjust_room(struct __sk_buff *ctx) return TC_ACT_SHOT; } +SEC("tc") +int helper_skb_change_head_tail(struct __sk_buff *ctx) +{ + int err; + + /* Reserve 1 extra in the front for packet data */ + err = bpf_skb_change_head(ctx, 1, 0); + if (err) + goto out; + + if (!check_skb_metadata(ctx)) + goto out; + + /* Reserve 256 extra bytes in the front to trigger head reallocation */ + err = bpf_skb_change_head(ctx, 256, 0); + if (err) + goto out; + + if (!check_skb_metadata(ctx)) + goto out; + + /* Reserve 4k extra bytes in the back to trigger head reallocation */ + err = bpf_skb_change_tail(ctx, ctx->len + 4096, 0); + if (err) + goto out; + + if (!check_skb_metadata(ctx)) + goto out; + + test_pass = true; +out: + return TC_ACT_SHOT; +} + char _license[] SEC("license") = "GPL"; From d2c5cca3fb58f96732e1dfd19e36b43e4e54d214 Mon Sep 17 00:00:00 2001 From: Jakub Sitnicki Date: Wed, 5 Nov 2025 21:19:53 +0100 Subject: [PATCH 0666/1381] selftests/bpf: Cover skb metadata access after bpf_skb_change_proto Add a test to verify that skb metadata remains accessible after calling bpf_skb_change_proto(), which modifies packet headroom to accommodate different IP header sizes. Signed-off-by: Jakub Sitnicki Signed-off-by: Martin KaFai Lau Link: https://patch.msgid.link/20251105-skb-meta-rx-path-v4-16-5ceb08a9b37b@cloudflare.com --- .../bpf/prog_tests/xdp_context_test_run.c | 5 ++++ .../selftests/bpf/progs/test_xdp_meta.c | 25 +++++++++++++++++++ 2 files changed, 30 insertions(+) diff --git a/tools/testing/selftests/bpf/prog_tests/xdp_context_test_run.c b/tools/testing/selftests/bpf/prog_tests/xdp_context_test_run.c index 65735a134abb..ee94c281888a 100644 --- a/tools/testing/selftests/bpf/prog_tests/xdp_context_test_run.c +++ b/tools/testing/selftests/bpf/prog_tests/xdp_context_test_run.c @@ -502,6 +502,11 @@ void test_xdp_context_tuntap(void) skel->progs.helper_skb_change_head_tail, NULL, /* tc prio 2 */ &skel->bss->test_pass); + if (test__start_subtest("helper_skb_change_proto")) + test_tuntap(skel->progs.ing_xdp, + skel->progs.helper_skb_change_proto, + NULL, /* tc prio 2 */ + &skel->bss->test_pass); test_xdp_meta__destroy(skel); } diff --git a/tools/testing/selftests/bpf/progs/test_xdp_meta.c b/tools/testing/selftests/bpf/progs/test_xdp_meta.c index e0b2e8ed0cc5..0a0f371a2dec 100644 --- a/tools/testing/selftests/bpf/progs/test_xdp_meta.c +++ b/tools/testing/selftests/bpf/progs/test_xdp_meta.c @@ -4,6 +4,7 @@ #include #include +#include #include #include "bpf_kfuncs.h" @@ -645,4 +646,28 @@ int helper_skb_change_head_tail(struct __sk_buff *ctx) return TC_ACT_SHOT; } +SEC("tc") +int helper_skb_change_proto(struct __sk_buff *ctx) +{ + int err; + + err = bpf_skb_change_proto(ctx, bpf_htons(ETH_P_IPV6), 0); + if (err) + goto out; + + if (!check_skb_metadata(ctx)) + goto out; + + err = bpf_skb_change_proto(ctx, bpf_htons(ETH_P_IP), 0); + if (err) + goto out; + + if (!check_skb_metadata(ctx)) + goto out; + + test_pass = true; +out: + return TC_ACT_SHOT; +} + char _license[] SEC("license") = "GPL"; From 07c428ece3222832bdbfcc4ffa8b8d3991c5eb39 Mon Sep 17 00:00:00 2001 From: "D. Wythe" Date: Fri, 7 Nov 2025 11:56:30 +0800 Subject: [PATCH 0667/1381] bpf: Export necessary symbols for modules with struct_ops Exports three necessary symbols for implementing struct_ops with tristate subsystem. To hold or release refcnt of struct_ops refcnt by inline funcs bpf_try_module_get and bpf_module_put which use bpf_struct_ops_get(put) conditionally. And to copy obj name from one to the other with effective checks by bpf_obj_name_cpy. Signed-off-by: D. Wythe Signed-off-by: Martin KaFai Lau Link: https://patch.msgid.link/20251107035632.115950-2-alibuda@linux.alibaba.com --- kernel/bpf/bpf_struct_ops.c | 2 ++ kernel/bpf/syscall.c | 1 + 2 files changed, 3 insertions(+) diff --git a/kernel/bpf/bpf_struct_ops.c b/kernel/bpf/bpf_struct_ops.c index a41e6730edcf..278490683d28 100644 --- a/kernel/bpf/bpf_struct_ops.c +++ b/kernel/bpf/bpf_struct_ops.c @@ -1162,6 +1162,7 @@ bool bpf_struct_ops_get(const void *kdata) map = __bpf_map_inc_not_zero(&st_map->map, false); return !IS_ERR(map); } +EXPORT_SYMBOL_GPL(bpf_struct_ops_get); void bpf_struct_ops_put(const void *kdata) { @@ -1173,6 +1174,7 @@ void bpf_struct_ops_put(const void *kdata) bpf_map_put(&st_map->map); } +EXPORT_SYMBOL_GPL(bpf_struct_ops_put); u32 bpf_struct_ops_id(const void *kdata) { diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c index 8a129746bd6c..80b86e9d3c39 100644 --- a/kernel/bpf/syscall.c +++ b/kernel/bpf/syscall.c @@ -1234,6 +1234,7 @@ int bpf_obj_name_cpy(char *dst, const char *src, unsigned int size) return src - orig_src; } +EXPORT_SYMBOL_GPL(bpf_obj_name_cpy); int map_check_no_btf(const struct bpf_map *map, const struct btf *btf, From 15f295f55656658e65bdbc9b901d6b2e49d68d72 Mon Sep 17 00:00:00 2001 From: "D. Wythe" Date: Fri, 7 Nov 2025 11:56:31 +0800 Subject: [PATCH 0668/1381] net/smc: bpf: Introduce generic hook for handshake flow The introduction of IPPROTO_SMC enables eBPF programs to determine whether to use SMC based on the context of socket creation, such as network namespaces, PID and comm name, etc. As a subsequent enhancement, to introduce a new generic hook that allows decisions on whether to use SMC or not at runtime, including but not limited to local/remote IP address or ports. User can write their own implememtion via bpf_struct_ops now to choose whether to use SMC or not before TCP 3rd handshake to be comleted. Signed-off-by: D. Wythe Signed-off-by: Martin KaFai Lau Reviewed-by: Dust Li Link: https://patch.msgid.link/20251107035632.115950-3-alibuda@linux.alibaba.com --- include/net/netns/smc.h | 3 + include/net/smc.h | 53 +++++++++++++++ net/ipv4/tcp_output.c | 31 +++++---- net/smc/Kconfig | 10 +++ net/smc/Makefile | 1 + net/smc/af_smc.c | 9 +++ net/smc/smc_hs_bpf.c | 140 ++++++++++++++++++++++++++++++++++++++++ net/smc/smc_hs_bpf.h | 31 +++++++++ net/smc/smc_sysctl.c | 91 ++++++++++++++++++++++++++ 9 files changed, 355 insertions(+), 14 deletions(-) create mode 100644 net/smc/smc_hs_bpf.c create mode 100644 net/smc/smc_hs_bpf.h diff --git a/include/net/netns/smc.h b/include/net/netns/smc.h index 6ceb12baec24..ed24c9f638ee 100644 --- a/include/net/netns/smc.h +++ b/include/net/netns/smc.h @@ -17,6 +17,9 @@ struct netns_smc { #ifdef CONFIG_SYSCTL struct ctl_table_header *smc_hdr; #endif +#if IS_ENABLED(CONFIG_SMC_HS_CTRL_BPF) + struct smc_hs_ctrl __rcu *hs_ctrl; +#endif /* CONFIG_SMC_HS_CTRL_BPF */ unsigned int sysctl_autocorking_size; unsigned int sysctl_smcr_buf_type; int sysctl_smcr_testlink_time; diff --git a/include/net/smc.h b/include/net/smc.h index 08bee529ed8d..bfdc4c41f019 100644 --- a/include/net/smc.h +++ b/include/net/smc.h @@ -17,6 +17,8 @@ #include #include +struct tcp_sock; +struct inet_request_sock; struct sock; #define SMC_MAX_PNETID_LEN 16 /* Max. length of PNET id */ @@ -50,4 +52,55 @@ struct smcd_dev { u8 going_away : 1; }; +#define SMC_HS_CTRL_NAME_MAX 16 + +enum { + /* ops can be inherit from init_net */ + SMC_HS_CTRL_FLAG_INHERITABLE = 0x1, + + SMC_HS_CTRL_ALL_FLAGS = SMC_HS_CTRL_FLAG_INHERITABLE, +}; + +struct smc_hs_ctrl { + /* private */ + + struct list_head list; + struct module *owner; + + /* public */ + + /* unique name */ + char name[SMC_HS_CTRL_NAME_MAX]; + int flags; + + /* Invoked before computing SMC option for SYN packets. + * We can control whether to set SMC options by returning various value. + * Return 0 to disable SMC, or return any other value to enable it. + */ + int (*syn_option)(struct tcp_sock *tp); + + /* Invoked before Set up SMC options for SYN-ACK packets + * We can control whether to respond SMC options by returning various + * value. Return 0 to disable SMC, or return any other value to enable + * it. + */ + int (*synack_option)(const struct tcp_sock *tp, + struct inet_request_sock *ireq); +}; + +#if IS_ENABLED(CONFIG_SMC_HS_CTRL_BPF) +#define smc_call_hsbpf(init_val, tp, func, ...) ({ \ + typeof(init_val) __ret = (init_val); \ + struct smc_hs_ctrl *ctrl; \ + rcu_read_lock(); \ + ctrl = rcu_dereference(sock_net((struct sock *)(tp))->smc.hs_ctrl); \ + if (ctrl && ctrl->func) \ + __ret = ctrl->func(tp, ##__VA_ARGS__); \ + rcu_read_unlock(); \ + __ret; \ +}) +#else +#define smc_call_hsbpf(init_val, tp, ...) ({ (void)(tp); (init_val); }) +#endif /* CONFIG_SMC_HS_CTRL_BPF */ + #endif /* _SMC_H */ diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 7f5df7a71f62..479afb714bdf 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -40,6 +40,7 @@ #include #include #include +#include #include #include @@ -802,34 +803,36 @@ static void tcp_options_write(struct tcphdr *th, struct tcp_sock *tp, mptcp_options_write(th, ptr, tp, opts); } -static void smc_set_option(const struct tcp_sock *tp, +static void smc_set_option(struct tcp_sock *tp, struct tcp_out_options *opts, unsigned int *remaining) { #if IS_ENABLED(CONFIG_SMC) - if (static_branch_unlikely(&tcp_have_smc)) { - if (tp->syn_smc) { - if (*remaining >= TCPOLEN_EXP_SMC_BASE_ALIGNED) { - opts->options |= OPTION_SMC; - *remaining -= TCPOLEN_EXP_SMC_BASE_ALIGNED; - } + if (static_branch_unlikely(&tcp_have_smc) && tp->syn_smc) { + tp->syn_smc = !!smc_call_hsbpf(1, tp, syn_option); + /* re-check syn_smc */ + if (tp->syn_smc && + *remaining >= TCPOLEN_EXP_SMC_BASE_ALIGNED) { + opts->options |= OPTION_SMC; + *remaining -= TCPOLEN_EXP_SMC_BASE_ALIGNED; } } #endif } static void smc_set_option_cond(const struct tcp_sock *tp, - const struct inet_request_sock *ireq, + struct inet_request_sock *ireq, struct tcp_out_options *opts, unsigned int *remaining) { #if IS_ENABLED(CONFIG_SMC) - if (static_branch_unlikely(&tcp_have_smc)) { - if (tp->syn_smc && ireq->smc_ok) { - if (*remaining >= TCPOLEN_EXP_SMC_BASE_ALIGNED) { - opts->options |= OPTION_SMC; - *remaining -= TCPOLEN_EXP_SMC_BASE_ALIGNED; - } + if (static_branch_unlikely(&tcp_have_smc) && tp->syn_smc && ireq->smc_ok) { + ireq->smc_ok = !!smc_call_hsbpf(1, tp, synack_option, ireq); + /* re-check smc_ok */ + if (ireq->smc_ok && + *remaining >= TCPOLEN_EXP_SMC_BASE_ALIGNED) { + opts->options |= OPTION_SMC; + *remaining -= TCPOLEN_EXP_SMC_BASE_ALIGNED; } } #endif diff --git a/net/smc/Kconfig b/net/smc/Kconfig index 99ecd59d1f4b..325addf83cc6 100644 --- a/net/smc/Kconfig +++ b/net/smc/Kconfig @@ -19,3 +19,13 @@ config SMC_DIAG smcss. if unsure, say Y. + +config SMC_HS_CTRL_BPF + bool "Generic eBPF hook for SMC handshake flow" + depends on SMC && BPF_SYSCALL + default y + help + SMC_HS_CTRL_BPF enables support to register generic eBPF hook for SMC + handshake flow, which offer much greater flexibility in modifying the behavior + of the SMC protocol stack compared to a complete kernel-based approach. Select + this option if you want filtring the handshake process via eBPF programs. \ No newline at end of file diff --git a/net/smc/Makefile b/net/smc/Makefile index 0e754cbc38f9..5368634c5dd6 100644 --- a/net/smc/Makefile +++ b/net/smc/Makefile @@ -6,3 +6,4 @@ smc-y := af_smc.o smc_pnet.o smc_ib.o smc_clc.o smc_core.o smc_wr.o smc_llc.o smc-y += smc_cdc.o smc_tx.o smc_rx.o smc_close.o smc_ism.o smc_netlink.o smc_stats.o smc-y += smc_tracepoint.o smc_inet.o smc-$(CONFIG_SYSCTL) += smc_sysctl.o +smc-$(CONFIG_SMC_HS_CTRL_BPF) += smc_hs_bpf.o diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c index 0ef3e16a8517..e388de8dca09 100644 --- a/net/smc/af_smc.c +++ b/net/smc/af_smc.c @@ -58,6 +58,7 @@ #include "smc_tracepoint.h" #include "smc_sysctl.h" #include "smc_inet.h" +#include "smc_hs_bpf.h" static DEFINE_MUTEX(smc_server_lgr_pending); /* serialize link group * creation on server @@ -3600,8 +3601,16 @@ static int __init smc_init(void) pr_err("%s: smc_inet_init fails with %d\n", __func__, rc); goto out_ulp; } + rc = bpf_smc_hs_ctrl_init(); + if (rc) { + pr_err("%s: bpf_smc_hs_ctrl_init fails with %d\n", __func__, + rc); + goto out_inet; + } static_branch_enable(&tcp_have_smc); return 0; +out_inet: + smc_inet_exit(); out_ulp: tcp_unregister_ulp(&smc_ulp_ops); out_ib: diff --git a/net/smc/smc_hs_bpf.c b/net/smc/smc_hs_bpf.c new file mode 100644 index 000000000000..063d23d85850 --- /dev/null +++ b/net/smc/smc_hs_bpf.c @@ -0,0 +1,140 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Shared Memory Communications over RDMA (SMC-R) and RoCE + * + * Generic hook for SMC handshake flow. + * + * Copyright IBM Corp. 2016 + * Copyright (c) 2025, Alibaba Inc. + * + * Author: D. Wythe + */ + +#include +#include +#include +#include + +#include "smc_hs_bpf.h" + +static DEFINE_SPINLOCK(smc_hs_ctrl_list_lock); +static LIST_HEAD(smc_hs_ctrl_list); + +static int smc_hs_ctrl_reg(struct smc_hs_ctrl *ctrl) +{ + int ret = 0; + + spin_lock(&smc_hs_ctrl_list_lock); + /* already exist or duplicate name */ + if (smc_hs_ctrl_find_by_name(ctrl->name)) + ret = -EEXIST; + else + list_add_tail_rcu(&ctrl->list, &smc_hs_ctrl_list); + spin_unlock(&smc_hs_ctrl_list_lock); + return ret; +} + +static void smc_hs_ctrl_unreg(struct smc_hs_ctrl *ctrl) +{ + spin_lock(&smc_hs_ctrl_list_lock); + list_del_rcu(&ctrl->list); + spin_unlock(&smc_hs_ctrl_list_lock); + + /* Ensure that all readers to complete */ + synchronize_rcu(); +} + +struct smc_hs_ctrl *smc_hs_ctrl_find_by_name(const char *name) +{ + struct smc_hs_ctrl *ctrl; + + list_for_each_entry_rcu(ctrl, &smc_hs_ctrl_list, list) { + if (strcmp(ctrl->name, name) == 0) + return ctrl; + } + return NULL; +} + +static int __smc_bpf_stub_set_tcp_option(struct tcp_sock *tp) { return 1; } +static int __smc_bpf_stub_set_tcp_option_cond(const struct tcp_sock *tp, + struct inet_request_sock *ireq) +{ + return 1; +} + +static struct smc_hs_ctrl __smc_bpf_hs_ctrl = { + .syn_option = __smc_bpf_stub_set_tcp_option, + .synack_option = __smc_bpf_stub_set_tcp_option_cond, +}; + +static int smc_bpf_hs_ctrl_init(struct btf *btf) { return 0; } + +static int smc_bpf_hs_ctrl_reg(void *kdata, struct bpf_link *link) +{ + if (link) + return -EOPNOTSUPP; + + return smc_hs_ctrl_reg(kdata); +} + +static void smc_bpf_hs_ctrl_unreg(void *kdata, struct bpf_link *link) +{ + smc_hs_ctrl_unreg(kdata); +} + +static int smc_bpf_hs_ctrl_init_member(const struct btf_type *t, + const struct btf_member *member, + void *kdata, const void *udata) +{ + const struct smc_hs_ctrl *u_ctrl; + struct smc_hs_ctrl *k_ctrl; + u32 moff; + + u_ctrl = (const struct smc_hs_ctrl *)udata; + k_ctrl = (struct smc_hs_ctrl *)kdata; + + moff = __btf_member_bit_offset(t, member) / 8; + switch (moff) { + case offsetof(struct smc_hs_ctrl, name): + if (bpf_obj_name_cpy(k_ctrl->name, u_ctrl->name, + sizeof(u_ctrl->name)) <= 0) + return -EINVAL; + return 1; + case offsetof(struct smc_hs_ctrl, flags): + if (u_ctrl->flags & ~SMC_HS_CTRL_ALL_FLAGS) + return -EINVAL; + k_ctrl->flags = u_ctrl->flags; + return 1; + default: + break; + } + + return 0; +} + +static const struct bpf_func_proto * +bpf_smc_hs_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog) +{ + return bpf_base_func_proto(func_id, prog); +} + +static const struct bpf_verifier_ops smc_bpf_verifier_ops = { + .get_func_proto = bpf_smc_hs_func_proto, + .is_valid_access = bpf_tracing_btf_ctx_access, +}; + +static struct bpf_struct_ops bpf_smc_hs_ctrl_ops = { + .name = "smc_hs_ctrl", + .init = smc_bpf_hs_ctrl_init, + .reg = smc_bpf_hs_ctrl_reg, + .unreg = smc_bpf_hs_ctrl_unreg, + .cfi_stubs = &__smc_bpf_hs_ctrl, + .verifier_ops = &smc_bpf_verifier_ops, + .init_member = smc_bpf_hs_ctrl_init_member, + .owner = THIS_MODULE, +}; + +int bpf_smc_hs_ctrl_init(void) +{ + return register_bpf_struct_ops(&bpf_smc_hs_ctrl_ops, smc_hs_ctrl); +} diff --git a/net/smc/smc_hs_bpf.h b/net/smc/smc_hs_bpf.h new file mode 100644 index 000000000000..f5f1807c079e --- /dev/null +++ b/net/smc/smc_hs_bpf.h @@ -0,0 +1,31 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Shared Memory Communications over RDMA (SMC-R) and RoCE + * + * Generic hook for SMC handshake flow. + * + * Copyright IBM Corp. 2016 + * Copyright (c) 2025, Alibaba Inc. + * + * Author: D. Wythe + */ + +#ifndef __SMC_HS_CTRL +#define __SMC_HS_CTRL + +#include + +/* Find hs_ctrl by the target name, which required to be a c-string. + * Return NULL if no such ctrl was found,otherwise, return a valid ctrl. + * + * Note: Caller MUST ensure it's was invoked under rcu_read_lock. + */ +struct smc_hs_ctrl *smc_hs_ctrl_find_by_name(const char *name); + +#if IS_ENABLED(CONFIG_SMC_HS_CTRL_BPF) +int bpf_smc_hs_ctrl_init(void); +#else +static inline int bpf_smc_hs_ctrl_init(void) { return 0; } +#endif /* CONFIG_SMC_HS_CTRL_BPF */ + +#endif /* __SMC_HS_CTRL */ diff --git a/net/smc/smc_sysctl.c b/net/smc/smc_sysctl.c index 7b2471904d04..b1efed546243 100644 --- a/net/smc/smc_sysctl.c +++ b/net/smc/smc_sysctl.c @@ -12,12 +12,14 @@ #include #include +#include #include #include "smc.h" #include "smc_core.h" #include "smc_llc.h" #include "smc_sysctl.h" +#include "smc_hs_bpf.h" static int min_sndbuf = SMC_BUF_MIN_SIZE; static int min_rcvbuf = SMC_BUF_MIN_SIZE; @@ -32,6 +34,69 @@ static int conns_per_lgr_max = SMC_CONN_PER_LGR_MAX; static unsigned int smcr_max_wr_min = 2; static unsigned int smcr_max_wr_max = 2048; +#if IS_ENABLED(CONFIG_SMC_HS_CTRL_BPF) +static int smc_net_replace_smc_hs_ctrl(struct net *net, const char *name) +{ + struct smc_hs_ctrl *ctrl = NULL; + + rcu_read_lock(); + /* null or empty name ask to clear current ctrl */ + if (name && name[0]) { + ctrl = smc_hs_ctrl_find_by_name(name); + if (!ctrl) { + rcu_read_unlock(); + return -EINVAL; + } + /* no change, just return */ + if (ctrl == rcu_dereference(net->smc.hs_ctrl)) { + rcu_read_unlock(); + return 0; + } + if (!bpf_try_module_get(ctrl, ctrl->owner)) { + rcu_read_unlock(); + return -EBUSY; + } + } + /* xhcg old ctrl with the new one atomically */ + ctrl = unrcu_pointer(xchg(&net->smc.hs_ctrl, RCU_INITIALIZER(ctrl))); + /* release old ctrl */ + if (ctrl) + bpf_module_put(ctrl, ctrl->owner); + + rcu_read_unlock(); + return 0; +} + +static int proc_smc_hs_ctrl(const struct ctl_table *ctl, int write, + void *buffer, size_t *lenp, loff_t *ppos) +{ + struct net *net = container_of(ctl->data, struct net, smc.hs_ctrl); + char val[SMC_HS_CTRL_NAME_MAX]; + const struct ctl_table tbl = { + .data = val, + .maxlen = SMC_HS_CTRL_NAME_MAX, + }; + struct smc_hs_ctrl *ctrl; + int ret; + + rcu_read_lock(); + ctrl = rcu_dereference(net->smc.hs_ctrl); + if (ctrl) + memcpy(val, ctrl->name, sizeof(ctrl->name)); + else + val[0] = '\0'; + rcu_read_unlock(); + + ret = proc_dostring(&tbl, write, buffer, lenp, ppos); + if (ret) + return ret; + + if (write) + ret = smc_net_replace_smc_hs_ctrl(net, val); + return ret; +} +#endif /* CONFIG_SMC_HS_CTRL_BPF */ + static struct ctl_table smc_table[] = { { .procname = "autocorking_size", @@ -119,6 +184,15 @@ static struct ctl_table smc_table[] = { .extra1 = &smcr_max_wr_min, .extra2 = &smcr_max_wr_max, }, +#if IS_ENABLED(CONFIG_SMC_HS_CTRL_BPF) + { + .procname = "hs_ctrl", + .data = &init_net.smc.hs_ctrl, + .mode = 0644, + .maxlen = SMC_HS_CTRL_NAME_MAX, + .proc_handler = proc_smc_hs_ctrl, + }, +#endif /* CONFIG_SMC_HS_CTRL_BPF */ }; int __net_init smc_sysctl_net_init(struct net *net) @@ -129,6 +203,16 @@ int __net_init smc_sysctl_net_init(struct net *net) table = smc_table; if (!net_eq(net, &init_net)) { int i; +#if IS_ENABLED(CONFIG_SMC_HS_CTRL_BPF) + struct smc_hs_ctrl *ctrl; + + rcu_read_lock(); + ctrl = rcu_dereference(init_net.smc.hs_ctrl); + if (ctrl && ctrl->flags & SMC_HS_CTRL_FLAG_INHERITABLE && + bpf_try_module_get(ctrl, ctrl->owner)) + rcu_assign_pointer(net->smc.hs_ctrl, ctrl); + rcu_read_unlock(); +#endif /* CONFIG_SMC_HS_CTRL_BPF */ table = kmemdup(table, sizeof(smc_table), GFP_KERNEL); if (!table) @@ -161,6 +245,9 @@ int __net_init smc_sysctl_net_init(struct net *net) if (!net_eq(net, &init_net)) kfree(table); err_alloc: +#if IS_ENABLED(CONFIG_SMC_HS_CTRL_BPF) + smc_net_replace_smc_hs_ctrl(net, NULL); +#endif /* CONFIG_SMC_HS_CTRL_BPF */ return -ENOMEM; } @@ -170,6 +257,10 @@ void __net_exit smc_sysctl_net_exit(struct net *net) table = net->smc.smc_hdr->ctl_table_arg; unregister_net_sysctl_table(net->smc.smc_hdr); +#if IS_ENABLED(CONFIG_SMC_HS_CTRL_BPF) + smc_net_replace_smc_hs_ctrl(net, NULL); +#endif /* CONFIG_SMC_HS_CTRL_BPF */ + if (!net_eq(net, &init_net)) kfree(table); } From beb3c67297d92f9428484410cf79135d38d0aff3 Mon Sep 17 00:00:00 2001 From: "D. Wythe" Date: Fri, 7 Nov 2025 11:56:32 +0800 Subject: [PATCH 0669/1381] bpf/selftests: Add selftest for bpf_smc_hs_ctrl This tests introduces a tiny smc_hs_ctrl for filtering SMC connections based on IP pairs, and also adds a realistic topology model to verify it. Also, we can only use SMC loopback under CI test, so an additional configuration needs to be enabled. Follow the steps below to run this test. make -C tools/testing/selftests/bpf cd tools/testing/selftests/bpf sudo ./test_progs -t smc Results shows: Summary: 1/1 PASSED, 0 SKIPPED, 0 FAILED Signed-off-by: D. Wythe Signed-off-by: Martin KaFai Lau Tested-by: Saket Kumar Bhaskar Reviewed-by: Zhu Yanjun Link: https://patch.msgid.link/20251107035632.115950-4-alibuda@linux.alibaba.com --- tools/testing/selftests/bpf/config | 5 + .../selftests/bpf/prog_tests/test_bpf_smc.c | 390 ++++++++++++++++++ tools/testing/selftests/bpf/progs/bpf_smc.c | 117 ++++++ 3 files changed, 512 insertions(+) create mode 100644 tools/testing/selftests/bpf/prog_tests/test_bpf_smc.c create mode 100644 tools/testing/selftests/bpf/progs/bpf_smc.c diff --git a/tools/testing/selftests/bpf/config b/tools/testing/selftests/bpf/config index 70b28c1e653e..fcd2f9bf78c9 100644 --- a/tools/testing/selftests/bpf/config +++ b/tools/testing/selftests/bpf/config @@ -123,3 +123,8 @@ CONFIG_XDP_SOCKETS=y CONFIG_XFRM_INTERFACE=y CONFIG_TCP_CONG_DCTCP=y CONFIG_TCP_CONG_BBR=y +CONFIG_INFINIBAND=y +CONFIG_SMC=y +CONFIG_SMC_HS_CTRL_BPF=y +CONFIG_DIBS=y +CONFIG_DIBS_LO=y \ No newline at end of file diff --git a/tools/testing/selftests/bpf/prog_tests/test_bpf_smc.c b/tools/testing/selftests/bpf/prog_tests/test_bpf_smc.c new file mode 100644 index 000000000000..de22734abc4d --- /dev/null +++ b/tools/testing/selftests/bpf/prog_tests/test_bpf_smc.c @@ -0,0 +1,390 @@ +// SPDX-License-Identifier: GPL-2.0 +#include +#include +#include "network_helpers.h" +#include "bpf_smc.skel.h" + +#ifndef IPPROTO_SMC +#define IPPROTO_SMC 256 +#endif + +#define CLIENT_IP "127.0.0.1" +#define SERVER_IP "127.0.1.0" +#define SERVER_IP_VIA_RISK_PATH "127.0.2.0" + +#define SERVICE_1 80 +#define SERVICE_2 443 +#define SERVICE_3 8443 + +#define TEST_NS "bpf_smc_netns" + +static struct netns_obj *test_netns; + +struct smc_policy_ip_key { + __u32 sip; + __u32 dip; +}; + +struct smc_policy_ip_value { + __u8 mode; +}; + +#if defined(__s390x__) +/* s390x has default seid */ +static bool setup_ueid(void) { return true; } +static void cleanup_ueid(void) {} +#else +enum { + SMC_NETLINK_ADD_UEID = 10, + SMC_NETLINK_REMOVE_UEID +}; + +enum { + SMC_NLA_EID_TABLE_UNSPEC, + SMC_NLA_EID_TABLE_ENTRY, /* string */ +}; + +struct msgtemplate { + struct nlmsghdr n; + struct genlmsghdr g; + char buf[1024]; +}; + +#define GENLMSG_DATA(glh) ((void *)(NLMSG_DATA(glh) + GENL_HDRLEN)) +#define GENLMSG_PAYLOAD(glh) (NLMSG_PAYLOAD(glh, 0) - GENL_HDRLEN) +#define NLA_DATA(na) ((void *)((char *)(na) + NLA_HDRLEN)) +#define NLA_PAYLOAD(len) ((len) - NLA_HDRLEN) + +#define SMC_GENL_FAMILY_NAME "SMC_GEN_NETLINK" +#define SMC_BPFTEST_UEID "SMC-BPFTEST-UEID" + +static uint16_t smc_nl_family_id = -1; + +static int send_cmd(int fd, __u16 nlmsg_type, __u32 nlmsg_pid, + __u16 nlmsg_flags, __u8 genl_cmd, __u16 nla_type, + void *nla_data, int nla_len) +{ + struct nlattr *na; + struct sockaddr_nl nladdr; + int r, buflen; + char *buf; + + struct msgtemplate msg = {0}; + + msg.n.nlmsg_len = NLMSG_LENGTH(GENL_HDRLEN); + msg.n.nlmsg_type = nlmsg_type; + msg.n.nlmsg_flags = nlmsg_flags; + msg.n.nlmsg_seq = 0; + msg.n.nlmsg_pid = nlmsg_pid; + msg.g.cmd = genl_cmd; + msg.g.version = 1; + na = (struct nlattr *)GENLMSG_DATA(&msg); + na->nla_type = nla_type; + na->nla_len = nla_len + 1 + NLA_HDRLEN; + memcpy(NLA_DATA(na), nla_data, nla_len); + msg.n.nlmsg_len += NLMSG_ALIGN(na->nla_len); + + buf = (char *)&msg; + buflen = msg.n.nlmsg_len; + memset(&nladdr, 0, sizeof(nladdr)); + nladdr.nl_family = AF_NETLINK; + + while ((r = sendto(fd, buf, buflen, 0, (struct sockaddr *)&nladdr, + sizeof(nladdr))) < buflen) { + if (r > 0) { + buf += r; + buflen -= r; + } else if (errno != EAGAIN) { + return -1; + } + } + return 0; +} + +static bool get_smc_nl_family_id(void) +{ + struct sockaddr_nl nl_src; + struct msgtemplate msg; + struct nlattr *nl; + int fd, ret; + pid_t pid; + + fd = socket(AF_NETLINK, SOCK_RAW, NETLINK_GENERIC); + if (!ASSERT_OK_FD(fd, "nl_family socket")) + return false; + + pid = getpid(); + + memset(&nl_src, 0, sizeof(nl_src)); + nl_src.nl_family = AF_NETLINK; + nl_src.nl_pid = pid; + + ret = bind(fd, (struct sockaddr *)&nl_src, sizeof(nl_src)); + if (!ASSERT_OK(ret, "nl_family bind")) + goto fail; + + ret = send_cmd(fd, GENL_ID_CTRL, pid, + NLM_F_REQUEST, CTRL_CMD_GETFAMILY, + CTRL_ATTR_FAMILY_NAME, (void *)SMC_GENL_FAMILY_NAME, + strlen(SMC_GENL_FAMILY_NAME)); + if (!ASSERT_OK(ret, "nl_family query")) + goto fail; + + ret = recv(fd, &msg, sizeof(msg), 0); + if (!ASSERT_FALSE(msg.n.nlmsg_type == NLMSG_ERROR || ret < 0 || + !NLMSG_OK(&msg.n, ret), "nl_family response")) + goto fail; + + nl = (struct nlattr *)GENLMSG_DATA(&msg); + nl = (struct nlattr *)((char *)nl + NLA_ALIGN(nl->nla_len)); + if (!ASSERT_EQ(nl->nla_type, CTRL_ATTR_FAMILY_ID, "nl_family nla type")) + goto fail; + + smc_nl_family_id = *(uint16_t *)NLA_DATA(nl); + close(fd); + return true; +fail: + close(fd); + return false; +} + +static bool smc_ueid(int op) +{ + struct sockaddr_nl nl_src; + struct msgtemplate msg; + struct nlmsgerr *err; + char test_ueid[32]; + int fd, ret; + pid_t pid; + + /* UEID required */ + memset(test_ueid, '\x20', sizeof(test_ueid)); + memcpy(test_ueid, SMC_BPFTEST_UEID, strlen(SMC_BPFTEST_UEID)); + fd = socket(AF_NETLINK, SOCK_RAW, NETLINK_GENERIC); + if (!ASSERT_OK_FD(fd, "ueid socket")) + return false; + + pid = getpid(); + memset(&nl_src, 0, sizeof(nl_src)); + nl_src.nl_family = AF_NETLINK; + nl_src.nl_pid = pid; + + ret = bind(fd, (struct sockaddr *)&nl_src, sizeof(nl_src)); + if (!ASSERT_OK(ret, "ueid bind")) + goto fail; + + ret = send_cmd(fd, smc_nl_family_id, pid, + NLM_F_REQUEST | NLM_F_ACK, op, SMC_NLA_EID_TABLE_ENTRY, + (void *)test_ueid, sizeof(test_ueid)); + if (!ASSERT_OK(ret, "ueid cmd")) + goto fail; + + ret = recv(fd, &msg, sizeof(msg), 0); + if (!ASSERT_FALSE(ret < 0 || + !NLMSG_OK(&msg.n, ret), "ueid response")) + goto fail; + + if (msg.n.nlmsg_type == NLMSG_ERROR) { + err = NLMSG_DATA(&msg); + switch (op) { + case SMC_NETLINK_REMOVE_UEID: + if (!ASSERT_FALSE((err->error && err->error != -ENOENT), + "ueid remove")) + goto fail; + break; + case SMC_NETLINK_ADD_UEID: + if (!ASSERT_OK(err->error, "ueid add")) + goto fail; + break; + default: + break; + } + } + close(fd); + return true; +fail: + close(fd); + return false; +} + +static bool setup_ueid(void) +{ + /* get smc nl id */ + if (!get_smc_nl_family_id()) + return false; + /* clear old ueid for bpftest */ + smc_ueid(SMC_NETLINK_REMOVE_UEID); + /* smc-loopback required ueid */ + return smc_ueid(SMC_NETLINK_ADD_UEID); +} + +static void cleanup_ueid(void) +{ + smc_ueid(SMC_NETLINK_REMOVE_UEID); +} +#endif /* __s390x__ */ + +static bool setup_netns(void) +{ + test_netns = netns_new(TEST_NS, true); + if (!ASSERT_OK_PTR(test_netns, "open net namespace")) + goto fail_netns; + + SYS(fail_ip, "ip addr add 127.0.1.0/8 dev lo"); + SYS(fail_ip, "ip addr add 127.0.2.0/8 dev lo"); + + return true; +fail_ip: + netns_free(test_netns); +fail_netns: + return false; +} + +static void cleanup_netns(void) +{ + netns_free(test_netns); +} + +static bool setup_smc(void) +{ + if (!setup_ueid()) + return false; + + if (!setup_netns()) + goto fail_netns; + + return true; +fail_netns: + cleanup_ueid(); + return false; +} + +static int set_client_addr_cb(int fd, void *opts) +{ + const char *src = (const char *)opts; + struct sockaddr_in localaddr; + + localaddr.sin_family = AF_INET; + localaddr.sin_port = htons(0); + localaddr.sin_addr.s_addr = inet_addr(src); + return !ASSERT_OK(bind(fd, &localaddr, sizeof(localaddr)), "client bind"); +} + +static void run_link(const char *src, const char *dst, int port) +{ + struct network_helper_opts opts = {0}; + int server, client; + + server = start_server_str(AF_INET, SOCK_STREAM, dst, port, NULL); + if (!ASSERT_OK_FD(server, "start service_1")) + return; + + opts.proto = IPPROTO_TCP; + opts.post_socket_cb = set_client_addr_cb; + opts.cb_opts = (void *)src; + + client = connect_to_fd_opts(server, &opts); + if (!ASSERT_OK_FD(client, "start connect")) + goto fail_client; + + close(client); +fail_client: + close(server); +} + +static void block_link(int map_fd, const char *src, const char *dst) +{ + struct smc_policy_ip_value val = { .mode = /* block */ 0 }; + struct smc_policy_ip_key key = { + .sip = inet_addr(src), + .dip = inet_addr(dst), + }; + + bpf_map_update_elem(map_fd, &key, &val, BPF_ANY); +} + +/* + * This test describes a real-life service topology as follows: + * + * +-------------> service_1 + * link 1 | | + * +--------------------> server | link 2 + * | | V + * | +-------------> service_2 + * | link 3 + * client -------------------> server_via_unsafe_path -> service_3 + * + * Among them, + * 1. link-1 is very suitable for using SMC. + * 2. link-2 is not suitable for using SMC, because the mode of this link is + * kind of short-link services. + * 3. link-3 is also not suitable for using SMC, because the RDMA link is + * unavailable and needs to go through a long timeout before it can fallback + * to TCP. + * To achieve this goal, we use a customized SMC ip strategy via smc_hs_ctrl. + */ +static void test_topo(void) +{ + struct bpf_smc *skel; + int rc, map_fd; + + skel = bpf_smc__open_and_load(); + if (!ASSERT_OK_PTR(skel, "bpf_smc__open_and_load")) + return; + + rc = bpf_smc__attach(skel); + if (!ASSERT_OK(rc, "bpf_smc__attach")) + goto fail; + + map_fd = bpf_map__fd(skel->maps.smc_policy_ip); + if (!ASSERT_OK_FD(map_fd, "bpf_map__fd")) + goto fail; + + /* Mock the process of transparent replacement, since we will modify + * protocol to ipproto_smc accropding to it via + * fmod_ret/update_socket_protocol. + */ + write_sysctl("/proc/sys/net/smc/hs_ctrl", "linkcheck"); + + /* Configure ip strat */ + block_link(map_fd, CLIENT_IP, SERVER_IP_VIA_RISK_PATH); + block_link(map_fd, SERVER_IP, SERVER_IP); + + /* should go with smc */ + run_link(CLIENT_IP, SERVER_IP, SERVICE_1); + /* should go with smc fallback */ + run_link(SERVER_IP, SERVER_IP, SERVICE_2); + + ASSERT_EQ(skel->bss->smc_cnt, 2, "smc count"); + ASSERT_EQ(skel->bss->fallback_cnt, 1, "fallback count"); + + /* should go with smc */ + run_link(CLIENT_IP, SERVER_IP, SERVICE_2); + + ASSERT_EQ(skel->bss->smc_cnt, 3, "smc count"); + ASSERT_EQ(skel->bss->fallback_cnt, 1, "fallback count"); + + /* should go with smc fallback */ + run_link(CLIENT_IP, SERVER_IP_VIA_RISK_PATH, SERVICE_3); + + ASSERT_EQ(skel->bss->smc_cnt, 4, "smc count"); + ASSERT_EQ(skel->bss->fallback_cnt, 2, "fallback count"); + +fail: + bpf_smc__destroy(skel); +} + +void test_bpf_smc(void) +{ + if (!setup_smc()) { + printf("setup for smc test failed, test SKIP:\n"); + test__skip(); + return; + } + + if (test__start_subtest("topo")) + test_topo(); + + cleanup_ueid(); + cleanup_netns(); +} diff --git a/tools/testing/selftests/bpf/progs/bpf_smc.c b/tools/testing/selftests/bpf/progs/bpf_smc.c new file mode 100644 index 000000000000..70d8b08f5914 --- /dev/null +++ b/tools/testing/selftests/bpf/progs/bpf_smc.c @@ -0,0 +1,117 @@ +// SPDX-License-Identifier: GPL-2.0 + +#include "vmlinux.h" + +#include +#include +#include "bpf_tracing_net.h" + +char _license[] SEC("license") = "GPL"; + +enum { + BPF_SMC_LISTEN = 10, +}; + +struct smc_sock___local { + struct sock sk; + struct smc_sock *listen_smc; + bool use_fallback; +} __attribute__((preserve_access_index)); + +int smc_cnt = 0; +int fallback_cnt = 0; + +SEC("fentry/smc_release") +int BPF_PROG(bpf_smc_release, struct socket *sock) +{ + /* only count from one side (client) */ + if (sock->sk->__sk_common.skc_state == BPF_SMC_LISTEN) + return 0; + smc_cnt++; + return 0; +} + +SEC("fentry/smc_switch_to_fallback") +int BPF_PROG(bpf_smc_switch_to_fallback, struct smc_sock___local *smc) +{ + /* only count from one side (client) */ + if (smc && !smc->listen_smc) + fallback_cnt++; + return 0; +} + +/* go with default value if no strat was found */ +bool default_ip_strat_value = true; + +struct smc_policy_ip_key { + __u32 sip; + __u32 dip; +}; + +struct smc_policy_ip_value { + __u8 mode; +}; + +struct { + __uint(type, BPF_MAP_TYPE_HASH); + __uint(key_size, sizeof(struct smc_policy_ip_key)); + __uint(value_size, sizeof(struct smc_policy_ip_value)); + __uint(max_entries, 128); + __uint(map_flags, BPF_F_NO_PREALLOC); +} smc_policy_ip SEC(".maps"); + +static bool smc_check(__u32 src, __u32 dst) +{ + struct smc_policy_ip_value *value; + struct smc_policy_ip_key key = { + .sip = src, + .dip = dst, + }; + + value = bpf_map_lookup_elem(&smc_policy_ip, &key); + return value ? value->mode : default_ip_strat_value; +} + +SEC("fmod_ret/update_socket_protocol") +int BPF_PROG(smc_run, int family, int type, int protocol) +{ + struct task_struct *task; + + if (family != AF_INET && family != AF_INET6) + return protocol; + + if ((type & 0xf) != SOCK_STREAM) + return protocol; + + if (protocol != 0 && protocol != IPPROTO_TCP) + return protocol; + + task = bpf_get_current_task_btf(); + /* Prevent from affecting other tests */ + if (!task || !task->nsproxy->net_ns->smc.hs_ctrl) + return protocol; + + return IPPROTO_SMC; +} + +SEC("struct_ops") +int BPF_PROG(bpf_smc_set_tcp_option_cond, const struct tcp_sock *tp, + struct inet_request_sock *ireq) +{ + return smc_check(ireq->req.__req_common.skc_daddr, + ireq->req.__req_common.skc_rcv_saddr); +} + +SEC("struct_ops") +int BPF_PROG(bpf_smc_set_tcp_option, struct tcp_sock *tp) +{ + return smc_check(tp->inet_conn.icsk_inet.sk.__sk_common.skc_rcv_saddr, + tp->inet_conn.icsk_inet.sk.__sk_common.skc_daddr); +} + +SEC(".struct_ops") +struct smc_hs_ctrl linkcheck = { + .name = "linkcheck", + .syn_option = (void *)bpf_smc_set_tcp_option, + .synack_option = (void *)bpf_smc_set_tcp_option_cond, +}; From 73edb26b06ea0eb5bd8c6cae5b2df212ae3c7ab5 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Thu, 6 Nov 2025 22:34:06 +0000 Subject: [PATCH 0670/1381] sctp: Don't inherit do_auto_asconf in sctp_clone_sock(). syzbot reported list_del(&sp->auto_asconf_list) corruption in sctp_destroy_sock(). The repro calls setsockopt(SCTP_AUTO_ASCONF, 1) to a SCTP listener, calls accept(), and close()s the child socket. setsockopt(SCTP_AUTO_ASCONF, 1) sets sp->do_auto_asconf to 1 and links sp->auto_asconf_list to a per-netns list. Both fields are placed after sp->pd_lobby in struct sctp_sock, and sctp_copy_descendant() did not copy the fields before the cited commit. Also, sctp_clone_sock() did not set them explicitly. In addition, sctp_auto_asconf_init() is called from sctp_sock_migrate(), but it initialises the fields only conditionally. The two fields relied on __GFP_ZERO added in sk_alloc(), but sk_clone() does not use it. Let's clear newsp->do_auto_asconf in sctp_clone_sock(). [0]: list_del corruption. prev->next should be ffff8880799e9148, but was ffff8880799e8808. (prev=ffff88803347d9f8) kernel BUG at lib/list_debug.c:64! Oops: invalid opcode: 0000 [#1] SMP KASAN PTI CPU: 0 UID: 0 PID: 6008 Comm: syz.0.17 Not tainted syzkaller #0 PREEMPT(full) Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 10/02/2025 RIP: 0010:__list_del_entry_valid_or_report+0x15a/0x190 lib/list_debug.c:62 Code: e8 7b 26 71 fd 43 80 3c 2c 00 74 08 4c 89 ff e8 7c ee 92 fd 49 8b 17 48 c7 c7 80 0a bf 8b 48 89 de 4c 89 f9 e8 07 c6 94 fc 90 <0f> 0b 4c 89 f7 e8 4c 26 71 fd 43 80 3c 2c 00 74 08 4c 89 ff e8 4d RSP: 0018:ffffc90003067ad8 EFLAGS: 00010246 RAX: 000000000000006d RBX: ffff8880799e9148 RCX: b056988859ee6e00 RDX: 0000000000000000 RSI: 0000000000000202 RDI: 0000000000000000 RBP: dffffc0000000000 R08: ffffc90003067807 R09: 1ffff9200060cf00 R10: dffffc0000000000 R11: fffff5200060cf01 R12: 1ffff1100668fb3f R13: dffffc0000000000 R14: ffff88803347d9f8 R15: ffff88803347d9f8 FS: 00005555823e5500(0000) GS:ffff88812613e000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000200000000480 CR3: 00000000741ce000 CR4: 00000000003526f0 Call Trace: __list_del_entry_valid include/linux/list.h:132 [inline] __list_del_entry include/linux/list.h:223 [inline] list_del include/linux/list.h:237 [inline] sctp_destroy_sock+0xb4/0x370 net/sctp/socket.c:5163 sk_common_release+0x75/0x310 net/core/sock.c:3961 sctp_close+0x77e/0x900 net/sctp/socket.c:1550 inet_release+0x144/0x190 net/ipv4/af_inet.c:437 __sock_release net/socket.c:662 [inline] sock_close+0xc3/0x240 net/socket.c:1455 __fput+0x44c/0xa70 fs/file_table.c:468 task_work_run+0x1d4/0x260 kernel/task_work.c:227 resume_user_mode_work include/linux/resume_user_mode.h:50 [inline] exit_to_user_mode_loop+0xe9/0x130 kernel/entry/common.c:43 exit_to_user_mode_prepare include/linux/irq-entry-common.h:225 [inline] syscall_exit_to_user_mode_work include/linux/entry-common.h:175 [inline] syscall_exit_to_user_mode include/linux/entry-common.h:210 [inline] do_syscall_64+0x2bd/0xfa0 arch/x86/entry/syscall_64.c:100 entry_SYSCALL_64_after_hwframe+0x77/0x7f Fixes: 16942cf4d3e3 ("sctp: Use sk_clone() in sctp_accept().") Reported-by: syzbot+ba535cb417f106327741@syzkaller.appspotmail.com Closes: https://lore.kernel.org/netdev/690d2185.a70a0220.22f260.000e.GAE@google.com/ Signed-off-by: Kuniyuki Iwashima Acked-by: Xin Long Link: https://patch.msgid.link/20251106223418.1455510-1-kuniyu@google.com Signed-off-by: Jakub Kicinski --- include/net/sctp/structs.h | 4 ---- net/sctp/socket.c | 1 + 2 files changed, 1 insertion(+), 4 deletions(-) diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 5900196d65fd..affee44bd38e 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -228,10 +228,6 @@ struct sctp_sock { atomic_t pd_mode; - /* Fields after this point will be skipped on copies, like on accept - * and peeloff operations - */ - /* Receive to here while partial delivery is in effect. */ struct sk_buff_head pd_lobby; diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 38d2932acebf..d808096f5ab1 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -4885,6 +4885,7 @@ static struct sock *sctp_clone_sock(struct sock *sk, } #endif + newsp->do_auto_asconf = 0; skb_queue_head_init(&newsp->pd_lobby); newsp->ep = sctp_endpoint_new(newsk, GFP_KERNEL); From 5b9192c2c075edf8a35f8c6c2b5ef36cdc8ce9f9 Mon Sep 17 00:00:00 2001 From: Zhongqiu Han Date: Fri, 7 Nov 2025 15:45:33 +0800 Subject: [PATCH 0671/1381] ptp: ocp: Document sysfs output format for backward compatibility Add a comment to ptp_ocp_tty_show() explaining that the sysfs output intentionally does not include a trailing newline. This is required for backward compatibility with existing userspace software that reads the sysfs attribute and uses the value directly as a device path. A previous attempt to add a newline to align with common kernel conventions broke userspace applications that were opening device paths like "/dev/ttyS4\n" instead of "/dev/ttyS4", resulting in ENOENT errors. This comment prevents future attempts to "fix" this behavior, which would break existing userspace applications. Link: https://lore.kernel.org/netdev/20251030124519.1828058-1-zhongqiu.han@oss.qualcomm.com/ Link: https://lore.kernel.org/netdev/aef3b850-5f38-4c28-a018-3b0006dc2f08@linux.dev/ Suggested-by: Jakub Kicinski Signed-off-by: Zhongqiu Han Reviewed-by: Vadim Fedorenko Link: https://patch.msgid.link/20251107074533.416048-1-zhongqiu.han@oss.qualcomm.com Signed-off-by: Jakub Kicinski --- drivers/ptp/ptp_ocp.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/ptp/ptp_ocp.c b/drivers/ptp/ptp_ocp.c index a5c363252986..eeebe4d149f7 100644 --- a/drivers/ptp/ptp_ocp.c +++ b/drivers/ptp/ptp_ocp.c @@ -3430,6 +3430,12 @@ ptp_ocp_tty_show(struct device *dev, struct device_attribute *attr, char *buf) struct dev_ext_attribute *ea = to_ext_attr(attr); struct ptp_ocp *bp = dev_get_drvdata(dev); + /* + * NOTE: This output does not include a trailing newline for backward + * compatibility. Existing userspace software uses this value directly + * as a device path (e.g., "/dev/ttyS4"), and adding a newline would + * break those applications. Do not add a newline to this output. + */ return sysfs_emit(buf, "ttyS%d", bp->port[(uintptr_t)ea->var].line); } From 38f073a71e85c726b09f935e7886de72bc57b15b Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Niklas=20S=C3=B6derlund?= Date: Fri, 7 Nov 2025 21:01:00 +0100 Subject: [PATCH 0672/1381] net: ravb: Correct bad check of timestamp control flags MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When converting the Renesas network drivers to use flags from enum hwtstamp_rx_filters to control when to timestamp packages instead of a driver specific schema with bit-wise flags an error was made. The bit-wise driver specific flags correct logic to set get_ts was: q: RAVB_BE + tstamp_rx_ctrl: 0 => 0 q: RAVB_NC + tstamp_rx_ctrl: 0 => 0 q: RAVB_BE + tstamp_rx_ctrl: RAVB_RXTSTAMP_TYPE_V2_L2_EVENT => 0 q: RAVB_NC + tstamp_rx_ctrl: RAVB_RXTSTAMP_TYPE_V2_L2_EVENT => 1 q: RAVB_BE + tstamp_rx_ctrl: RAVB_RXTSTAMP_TYPE_ALL => 1 q: RAVB_NC + tstamp_rx_ctrl: RAVB_RXTSTAMP_TYPE_ALL => 1 The converted logic to use enum flags mapped tstamp_rx_ctrl as 0 to HWTSTAMP_FILTER_NONE RAVB_RXTSTAMP_TYPE_V2_L2_EVENT to HWTSTAMP_FILTER_PTP_V2_L2_EVENT RAVB_RXTSTAMP_TYPE_ALL to HWTSTAMP_FILTER_ALL But the logic was incorrectly changed to: q: RAVB_BE + tstamp_rx_ctrl: HWTSTAMP_FILTER_NONE => 1 (error) q: RAVB_NC + tstamp_rx_ctrl: HWTSTAMP_FILTER_NONE => 0 q: RAVB_BE + tstamp_rx_ctrl: HWTSTAMP_FILTER_PTP_V2_L2_EVENT => 0 q: RAVB_NC + tstamp_rx_ctrl: HWTSTAMP_FILTER_PTP_V2_L2_EVENT => 1 q: RAVB_BE + tstamp_rx_ctrl: HWTSTAMP_FILTER_ALL => 1 q: RAVB_NC + tstamp_rx_ctrl: HWTSTAMP_FILTER_ALL => 0 (error) This change restores the converted flag check to the correct logic of the bit-wise driver specific flags. Reported-by: Simon Horman Closes: https://lore.kernel.org/linux-renesas-soc/aQ4xSv9629XF-Bt3@horms.kernel.org/ Fixes: 16e2e6cf75e6 ("net: ravb: Use common defines for time stamping control") Signed-off-by: Niklas Söderlund Link: https://patch.msgid.link/20251107200100.3637869-1-niklas.soderlund+renesas@ragnatech.se Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/renesas/ravb_main.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c index 1680e94b9242..57b0db314fb5 100644 --- a/drivers/net/ethernet/renesas/ravb_main.c +++ b/drivers/net/ethernet/renesas/ravb_main.c @@ -955,9 +955,9 @@ static void ravb_rx_rcar_hwstamp(struct ravb_private *priv, int q, bool get_ts; if (q == RAVB_NC) - get_ts = priv->tstamp_rx_ctrl == HWTSTAMP_FILTER_PTP_V2_L2_EVENT; + get_ts = priv->tstamp_rx_ctrl != HWTSTAMP_FILTER_NONE; else - get_ts = priv->tstamp_rx_ctrl != HWTSTAMP_FILTER_PTP_V2_L2_EVENT; + get_ts = priv->tstamp_rx_ctrl == HWTSTAMP_FILTER_ALL; if (!get_ts) return; From a6e4fd38bf2f2e2363b61c27f4e6c49b14e4bb07 Mon Sep 17 00:00:00 2001 From: Jonas Gorski Date: Fri, 7 Nov 2025 09:07:42 +0100 Subject: [PATCH 0673/1381] net: dsa: b53: b53_arl_read{,25}(): use the entry for comparision Align the b53_arl_read{,25}() functions by consistently using the parsed arl entry instead of parsing the raw registers again. Signed-off-by: Jonas Gorski Reviewed-by: Florian Fainelli Link: https://patch.msgid.link/20251107080749.26936-2-jonas.gorski@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/b53/b53_common.c | 22 ++++++++++------------ 1 file changed, 10 insertions(+), 12 deletions(-) diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c index c911d7ea601f..1b94cf7b06e8 100644 --- a/drivers/net/dsa/b53/b53_common.c +++ b/drivers/net/dsa/b53/b53_common.c @@ -1850,7 +1850,7 @@ static int b53_arl_rw_op(struct b53_device *dev, unsigned int op) return b53_arl_op_wait(dev); } -static int b53_arl_read(struct b53_device *dev, u64 mac, +static int b53_arl_read(struct b53_device *dev, const u8 *mac, u16 vid, struct b53_arl_entry *ent, u8 *idx) { DECLARE_BITMAP(free_bins, B53_ARLTBL_MAX_BIN_ENTRIES); @@ -1874,14 +1874,13 @@ static int b53_arl_read(struct b53_device *dev, u64 mac, B53_ARLTBL_DATA_ENTRY(i), &fwd_entry); b53_arl_to_entry(ent, mac_vid, fwd_entry); - if (!(fwd_entry & ARLTBL_VALID)) { + if (!ent->is_valid) { set_bit(i, free_bins); continue; } - if ((mac_vid & ARLTBL_MAC_MASK) != mac) + if (!ether_addr_equal(ent->mac, mac)) continue; - if (dev->vlan_enabled && - ((mac_vid >> ARLTBL_VID_S) & ARLTBL_VID_MASK) != vid) + if (dev->vlan_enabled && ent->vid != vid) continue; *idx = i; return 0; @@ -1891,7 +1890,7 @@ static int b53_arl_read(struct b53_device *dev, u64 mac, return *idx >= dev->num_arl_bins ? -ENOSPC : -ENOENT; } -static int b53_arl_read_25(struct b53_device *dev, u64 mac, +static int b53_arl_read_25(struct b53_device *dev, const u8 *mac, u16 vid, struct b53_arl_entry *ent, u8 *idx) { DECLARE_BITMAP(free_bins, B53_ARLTBL_MAX_BIN_ENTRIES); @@ -1913,14 +1912,13 @@ static int b53_arl_read_25(struct b53_device *dev, u64 mac, b53_arl_to_entry_25(ent, mac_vid); - if (!(mac_vid & ARLTBL_VALID_25)) { + if (!ent->is_valid) { set_bit(i, free_bins); continue; } - if ((mac_vid & ARLTBL_MAC_MASK) != mac) + if (!ether_addr_equal(ent->mac, mac)) continue; - if (dev->vlan_enabled && - ((mac_vid >> ARLTBL_VID_S_65) & ARLTBL_VID_MASK_25) != vid) + if (dev->vlan_enabled && ent->vid != vid) continue; *idx = i; return 0; @@ -1953,9 +1951,9 @@ static int b53_arl_op(struct b53_device *dev, int op, int port, return ret; if (is5325(dev) || is5365(dev)) - ret = b53_arl_read_25(dev, mac, vid, &ent, &idx); + ret = b53_arl_read_25(dev, addr, vid, &ent, &idx); else - ret = b53_arl_read(dev, mac, vid, &ent, &idx); + ret = b53_arl_read(dev, addr, vid, &ent, &idx); /* If this is a read, just finish now */ if (op) From 4a291fe7226736a465ddb3fa93c21fcef7162ec7 Mon Sep 17 00:00:00 2001 From: Jonas Gorski Date: Fri, 7 Nov 2025 09:07:43 +0100 Subject: [PATCH 0674/1381] net: dsa: b53: move reading ARL entries into their own function Instead of duplicating the whole code iterating over all bins for BCM5325, factor out reading and parsing the entry into its own functions, and name it the modern one after the first chip with that ARL format, (BCM53)95. Signed-off-by: Jonas Gorski Reviewed-by: Florian Fainelli Link: https://patch.msgid.link/20251107080749.26936-3-jonas.gorski@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/b53/b53_common.c | 77 +++++++++++--------------------- 1 file changed, 27 insertions(+), 50 deletions(-) diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c index 1b94cf7b06e8..d99e15a7a6bb 100644 --- a/drivers/net/dsa/b53/b53_common.c +++ b/drivers/net/dsa/b53/b53_common.c @@ -1850,6 +1850,28 @@ static int b53_arl_rw_op(struct b53_device *dev, unsigned int op) return b53_arl_op_wait(dev); } +static void b53_arl_read_entry_25(struct b53_device *dev, + struct b53_arl_entry *ent, u8 idx) +{ + u64 mac_vid; + + b53_read64(dev, B53_ARLIO_PAGE, B53_ARLTBL_MAC_VID_ENTRY(idx), + &mac_vid); + b53_arl_to_entry_25(ent, mac_vid); +} + +static void b53_arl_read_entry_95(struct b53_device *dev, + struct b53_arl_entry *ent, u8 idx) +{ + u32 fwd_entry; + u64 mac_vid; + + b53_read64(dev, B53_ARLIO_PAGE, B53_ARLTBL_MAC_VID_ENTRY(idx), + &mac_vid); + b53_read32(dev, B53_ARLIO_PAGE, B53_ARLTBL_DATA_ENTRY(idx), &fwd_entry); + b53_arl_to_entry(ent, mac_vid, fwd_entry); +} + static int b53_arl_read(struct b53_device *dev, const u8 *mac, u16 vid, struct b53_arl_entry *ent, u8 *idx) { @@ -1865,52 +1887,10 @@ static int b53_arl_read(struct b53_device *dev, const u8 *mac, /* Read the bins */ for (i = 0; i < dev->num_arl_bins; i++) { - u64 mac_vid; - u32 fwd_entry; - - b53_read64(dev, B53_ARLIO_PAGE, - B53_ARLTBL_MAC_VID_ENTRY(i), &mac_vid); - b53_read32(dev, B53_ARLIO_PAGE, - B53_ARLTBL_DATA_ENTRY(i), &fwd_entry); - b53_arl_to_entry(ent, mac_vid, fwd_entry); - - if (!ent->is_valid) { - set_bit(i, free_bins); - continue; - } - if (!ether_addr_equal(ent->mac, mac)) - continue; - if (dev->vlan_enabled && ent->vid != vid) - continue; - *idx = i; - return 0; - } - - *idx = find_first_bit(free_bins, dev->num_arl_bins); - return *idx >= dev->num_arl_bins ? -ENOSPC : -ENOENT; -} - -static int b53_arl_read_25(struct b53_device *dev, const u8 *mac, - u16 vid, struct b53_arl_entry *ent, u8 *idx) -{ - DECLARE_BITMAP(free_bins, B53_ARLTBL_MAX_BIN_ENTRIES); - unsigned int i; - int ret; - - ret = b53_arl_op_wait(dev); - if (ret) - return ret; - - bitmap_zero(free_bins, dev->num_arl_bins); - - /* Read the bins */ - for (i = 0; i < dev->num_arl_bins; i++) { - u64 mac_vid; - - b53_read64(dev, B53_ARLIO_PAGE, - B53_ARLTBL_MAC_VID_ENTRY(i), &mac_vid); - - b53_arl_to_entry_25(ent, mac_vid); + if (is5325(dev) || is5365(dev)) + b53_arl_read_entry_25(dev, ent, i); + else + b53_arl_read_entry_95(dev, ent, i); if (!ent->is_valid) { set_bit(i, free_bins); @@ -1950,10 +1930,7 @@ static int b53_arl_op(struct b53_device *dev, int op, int port, if (ret) return ret; - if (is5325(dev) || is5365(dev)) - ret = b53_arl_read_25(dev, addr, vid, &ent, &idx); - else - ret = b53_arl_read(dev, addr, vid, &ent, &idx); + ret = b53_arl_read(dev, addr, vid, &ent, &idx); /* If this is a read, just finish now */ if (op) From bf6e9d2ae1dbafee53ec4ccd126595172e1e5278 Mon Sep 17 00:00:00 2001 From: Jonas Gorski Date: Fri, 7 Nov 2025 09:07:44 +0100 Subject: [PATCH 0675/1381] net: dsa: b53: move writing ARL entries into their own functions Move writing ARL entries into individual functions for each format. Signed-off-by: Jonas Gorski Reviewed-by: Florian Fainelli Link: https://patch.msgid.link/20251107080749.26936-4-jonas.gorski@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/b53/b53_common.c | 38 ++++++++++++++++++++++---------- 1 file changed, 26 insertions(+), 12 deletions(-) diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c index d99e15a7a6bb..9eb7ca878e30 100644 --- a/drivers/net/dsa/b53/b53_common.c +++ b/drivers/net/dsa/b53/b53_common.c @@ -1860,6 +1860,16 @@ static void b53_arl_read_entry_25(struct b53_device *dev, b53_arl_to_entry_25(ent, mac_vid); } +static void b53_arl_write_entry_25(struct b53_device *dev, + const struct b53_arl_entry *ent, u8 idx) +{ + u64 mac_vid; + + b53_arl_from_entry_25(&mac_vid, ent); + b53_write64(dev, B53_ARLIO_PAGE, B53_ARLTBL_MAC_VID_ENTRY(idx), + mac_vid); +} + static void b53_arl_read_entry_95(struct b53_device *dev, struct b53_arl_entry *ent, u8 idx) { @@ -1872,6 +1882,19 @@ static void b53_arl_read_entry_95(struct b53_device *dev, b53_arl_to_entry(ent, mac_vid, fwd_entry); } +static void b53_arl_write_entry_95(struct b53_device *dev, + const struct b53_arl_entry *ent, u8 idx) +{ + u32 fwd_entry; + u64 mac_vid; + + b53_arl_from_entry(&mac_vid, &fwd_entry, ent); + b53_write64(dev, B53_ARLIO_PAGE, B53_ARLTBL_MAC_VID_ENTRY(idx), + mac_vid); + b53_write32(dev, B53_ARLIO_PAGE, B53_ARLTBL_DATA_ENTRY(idx), + fwd_entry); +} + static int b53_arl_read(struct b53_device *dev, const u8 *mac, u16 vid, struct b53_arl_entry *ent, u8 *idx) { @@ -1912,9 +1935,8 @@ static int b53_arl_op(struct b53_device *dev, int op, int port, const unsigned char *addr, u16 vid, bool is_valid) { struct b53_arl_entry ent; - u32 fwd_entry; - u64 mac, mac_vid = 0; u8 idx = 0; + u64 mac; int ret; /* Convert the array into a 64-bit MAC */ @@ -1947,7 +1969,6 @@ static int b53_arl_op(struct b53_device *dev, int op, int port, /* We could not find a matching MAC, so reset to a new entry */ dev_dbg(dev->dev, "{%pM,%.4d} not found, using idx: %d\n", addr, vid, idx); - fwd_entry = 0; break; default: dev_dbg(dev->dev, "{%pM,%.4d} found, using idx: %d\n", @@ -1975,16 +1996,9 @@ static int b53_arl_op(struct b53_device *dev, int op, int port, ent.is_age = false; memcpy(ent.mac, addr, ETH_ALEN); if (is5325(dev) || is5365(dev)) - b53_arl_from_entry_25(&mac_vid, &ent); + b53_arl_write_entry_25(dev, &ent, idx); else - b53_arl_from_entry(&mac_vid, &fwd_entry, &ent); - - b53_write64(dev, B53_ARLIO_PAGE, - B53_ARLTBL_MAC_VID_ENTRY(idx), mac_vid); - - if (!is5325(dev) && !is5365(dev)) - b53_write32(dev, B53_ARLIO_PAGE, - B53_ARLTBL_DATA_ENTRY(idx), fwd_entry); + b53_arl_write_entry_95(dev, &ent, idx); return b53_arl_rw_op(dev, 0); } From 1716be6db04af53bac9b869f01156a460595cf41 Mon Sep 17 00:00:00 2001 From: Jonas Gorski Date: Fri, 7 Nov 2025 09:07:45 +0100 Subject: [PATCH 0676/1381] net: dsa: b53: provide accessors for accessing ARL_SRCH_CTL In order to more easily support more formats, move accessing ARL_SRCH_CTL into helper functions to contain the differences. Signed-off-by: Jonas Gorski Reviewed-by: Florian Fainelli Link: https://patch.msgid.link/20251107080749.26936-5-jonas.gorski@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/b53/b53_common.c | 37 +++++++++++++++++++++----------- 1 file changed, 24 insertions(+), 13 deletions(-) diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c index 9eb7ca878e30..b13437ea21a0 100644 --- a/drivers/net/dsa/b53/b53_common.c +++ b/drivers/net/dsa/b53/b53_common.c @@ -2033,18 +2033,37 @@ int b53_fdb_del(struct dsa_switch *ds, int port, } EXPORT_SYMBOL(b53_fdb_del); -static int b53_arl_search_wait(struct b53_device *dev) +static void b53_read_arl_srch_ctl(struct b53_device *dev, u8 *val) { - unsigned int timeout = 1000; - u8 reg, offset; + u8 offset; if (is5325(dev) || is5365(dev)) offset = B53_ARL_SRCH_CTL_25; else offset = B53_ARL_SRCH_CTL; + b53_read8(dev, B53_ARLIO_PAGE, offset, val); +} + +static void b53_write_arl_srch_ctl(struct b53_device *dev, u8 val) +{ + u8 offset; + + if (is5325(dev) || is5365(dev)) + offset = B53_ARL_SRCH_CTL_25; + else + offset = B53_ARL_SRCH_CTL; + + b53_write8(dev, B53_ARLIO_PAGE, offset, val); +} + +static int b53_arl_search_wait(struct b53_device *dev) +{ + unsigned int timeout = 1000; + u8 reg; + do { - b53_read8(dev, B53_ARLIO_PAGE, offset, ®); + b53_read_arl_srch_ctl(dev, ®); if (!(reg & ARL_SRCH_STDN)) return -ENOENT; @@ -2099,23 +2118,15 @@ int b53_fdb_dump(struct dsa_switch *ds, int port, unsigned int count = 0, results_per_hit = 1; struct b53_device *priv = ds->priv; struct b53_arl_entry results[2]; - u8 offset; int ret; - u8 reg; if (priv->num_arl_bins > 2) results_per_hit = 2; mutex_lock(&priv->arl_mutex); - if (is5325(priv) || is5365(priv)) - offset = B53_ARL_SRCH_CTL_25; - else - offset = B53_ARL_SRCH_CTL; - /* Start search operation */ - reg = ARL_SRCH_STDN; - b53_write8(priv, B53_ARLIO_PAGE, offset, reg); + b53_write_arl_srch_ctl(priv, ARL_SRCH_STDN); do { ret = b53_arl_search_wait(priv); From e0c476f325a8c9b961a3d446c24d3c8ecae7d186 Mon Sep 17 00:00:00 2001 From: Jonas Gorski Date: Fri, 7 Nov 2025 09:07:46 +0100 Subject: [PATCH 0677/1381] net: dsa: b53: split reading search entry into their own functions Split reading search entries into a function for each format. Signed-off-by: Jonas Gorski Reviewed-by: Florian Fainelli Link: https://patch.msgid.link/20251107080749.26936-6-jonas.gorski@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/b53/b53_common.c | 56 ++++++++++++++++++++++---------- 1 file changed, 38 insertions(+), 18 deletions(-) diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c index b13437ea21a0..fa4cf6ceddb8 100644 --- a/drivers/net/dsa/b53/b53_common.c +++ b/drivers/net/dsa/b53/b53_common.c @@ -2076,28 +2076,48 @@ static int b53_arl_search_wait(struct b53_device *dev) return -ETIMEDOUT; } -static void b53_arl_search_rd(struct b53_device *dev, u8 idx, - struct b53_arl_entry *ent) +static void b53_arl_search_read_25(struct b53_device *dev, u8 idx, + struct b53_arl_entry *ent) { u64 mac_vid; - if (is5325(dev)) { - b53_read64(dev, B53_ARLIO_PAGE, B53_ARL_SRCH_RSTL_0_MACVID_25, - &mac_vid); - b53_arl_to_entry_25(ent, mac_vid); - } else if (is5365(dev)) { - b53_read64(dev, B53_ARLIO_PAGE, B53_ARL_SRCH_RSTL_0_MACVID_65, - &mac_vid); - b53_arl_to_entry_25(ent, mac_vid); - } else { - u32 fwd_entry; + b53_read64(dev, B53_ARLIO_PAGE, B53_ARL_SRCH_RSTL_0_MACVID_25, + &mac_vid); + b53_arl_to_entry_25(ent, mac_vid); +} - b53_read64(dev, B53_ARLIO_PAGE, B53_ARL_SRCH_RSTL_MACVID(idx), - &mac_vid); - b53_read32(dev, B53_ARLIO_PAGE, B53_ARL_SRCH_RSTL(idx), - &fwd_entry); - b53_arl_to_entry(ent, mac_vid, fwd_entry); - } +static void b53_arl_search_read_65(struct b53_device *dev, u8 idx, + struct b53_arl_entry *ent) +{ + u64 mac_vid; + + b53_read64(dev, B53_ARLIO_PAGE, B53_ARL_SRCH_RSTL_0_MACVID_65, + &mac_vid); + b53_arl_to_entry_25(ent, mac_vid); +} + +static void b53_arl_search_read_95(struct b53_device *dev, u8 idx, + struct b53_arl_entry *ent) +{ + u32 fwd_entry; + u64 mac_vid; + + b53_read64(dev, B53_ARLIO_PAGE, B53_ARL_SRCH_RSTL_MACVID(idx), + &mac_vid); + b53_read32(dev, B53_ARLIO_PAGE, B53_ARL_SRCH_RSTL(idx), + &fwd_entry); + b53_arl_to_entry(ent, mac_vid, fwd_entry); +} + +static void b53_arl_search_rd(struct b53_device *dev, u8 idx, + struct b53_arl_entry *ent) +{ + if (is5325(dev)) + b53_arl_search_read_25(dev, idx, ent); + else if (is5365(dev)) + b53_arl_search_read_65(dev, idx, ent); + else + b53_arl_search_read_95(dev, idx, ent); } static int b53_fdb_copy(int port, const struct b53_arl_entry *ent, From a7e73339ad46ade76d29fb6cc7d7854222608c26 Mon Sep 17 00:00:00 2001 From: Jonas Gorski Date: Fri, 7 Nov 2025 09:07:47 +0100 Subject: [PATCH 0678/1381] net: dsa: b53: move ARL entry functions into ops struct Now that the differences in ARL entry formats are neatly contained into functions per chip family, wrap them into an ops struct and add wrapper functions to access them. Signed-off-by: Jonas Gorski Reviewed-by: Florian Fainelli Link: https://patch.msgid.link/20251107080749.26936-7-jonas.gorski@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/b53/b53_common.c | 67 ++++++++++++++++++++++---------- drivers/net/dsa/b53/b53_priv.h | 30 ++++++++++++++ 2 files changed, 76 insertions(+), 21 deletions(-) diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c index fa4cf6ceddb8..c69022cc85bf 100644 --- a/drivers/net/dsa/b53/b53_common.c +++ b/drivers/net/dsa/b53/b53_common.c @@ -1910,10 +1910,7 @@ static int b53_arl_read(struct b53_device *dev, const u8 *mac, /* Read the bins */ for (i = 0; i < dev->num_arl_bins; i++) { - if (is5325(dev) || is5365(dev)) - b53_arl_read_entry_25(dev, ent, i); - else - b53_arl_read_entry_95(dev, ent, i); + b53_arl_read_entry(dev, ent, i); if (!ent->is_valid) { set_bit(i, free_bins); @@ -1995,10 +1992,7 @@ static int b53_arl_op(struct b53_device *dev, int op, int port, ent.is_static = true; ent.is_age = false; memcpy(ent.mac, addr, ETH_ALEN); - if (is5325(dev) || is5365(dev)) - b53_arl_write_entry_25(dev, &ent, idx); - else - b53_arl_write_entry_95(dev, &ent, idx); + b53_arl_write_entry(dev, &ent, idx); return b53_arl_rw_op(dev, 0); } @@ -2109,17 +2103,6 @@ static void b53_arl_search_read_95(struct b53_device *dev, u8 idx, b53_arl_to_entry(ent, mac_vid, fwd_entry); } -static void b53_arl_search_rd(struct b53_device *dev, u8 idx, - struct b53_arl_entry *ent) -{ - if (is5325(dev)) - b53_arl_search_read_25(dev, idx, ent); - else if (is5365(dev)) - b53_arl_search_read_65(dev, idx, ent); - else - b53_arl_search_read_95(dev, idx, ent); -} - static int b53_fdb_copy(int port, const struct b53_arl_entry *ent, dsa_fdb_dump_cb_t *cb, void *data) { @@ -2153,13 +2136,13 @@ int b53_fdb_dump(struct dsa_switch *ds, int port, if (ret) break; - b53_arl_search_rd(priv, 0, &results[0]); + b53_arl_search_read(priv, 0, &results[0]); ret = b53_fdb_copy(port, &results[0], cb, data); if (ret) break; if (results_per_hit == 2) { - b53_arl_search_rd(priv, 1, &results[1]); + b53_arl_search_read(priv, 1, &results[1]); ret = b53_fdb_copy(port, &results[1], cb, data); if (ret) break; @@ -2688,6 +2671,24 @@ static const struct dsa_switch_ops b53_switch_ops = { .port_change_mtu = b53_change_mtu, }; +static const struct b53_arl_ops b53_arl_ops_25 = { + .arl_read_entry = b53_arl_read_entry_25, + .arl_write_entry = b53_arl_write_entry_25, + .arl_search_read = b53_arl_search_read_25, +}; + +static const struct b53_arl_ops b53_arl_ops_65 = { + .arl_read_entry = b53_arl_read_entry_25, + .arl_write_entry = b53_arl_write_entry_25, + .arl_search_read = b53_arl_search_read_65, +}; + +static const struct b53_arl_ops b53_arl_ops_95 = { + .arl_read_entry = b53_arl_read_entry_95, + .arl_write_entry = b53_arl_write_entry_95, + .arl_search_read = b53_arl_search_read_95, +}; + struct b53_chip_data { u32 chip_id; const char *dev_name; @@ -2701,6 +2702,7 @@ struct b53_chip_data { u8 duplex_reg; u8 jumbo_pm_reg; u8 jumbo_size_reg; + const struct b53_arl_ops *arl_ops; }; #define B53_VTA_REGS \ @@ -2720,6 +2722,7 @@ static const struct b53_chip_data b53_switch_chips[] = { .arl_buckets = 1024, .imp_port = 5, .duplex_reg = B53_DUPLEX_STAT_FE, + .arl_ops = &b53_arl_ops_25, }, { .chip_id = BCM5365_DEVICE_ID, @@ -2730,6 +2733,7 @@ static const struct b53_chip_data b53_switch_chips[] = { .arl_buckets = 1024, .imp_port = 5, .duplex_reg = B53_DUPLEX_STAT_FE, + .arl_ops = &b53_arl_ops_65, }, { .chip_id = BCM5389_DEVICE_ID, @@ -2743,6 +2747,7 @@ static const struct b53_chip_data b53_switch_chips[] = { .duplex_reg = B53_DUPLEX_STAT_GE, .jumbo_pm_reg = B53_JUMBO_PORT_MASK, .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + .arl_ops = &b53_arl_ops_95, }, { .chip_id = BCM5395_DEVICE_ID, @@ -2756,6 +2761,7 @@ static const struct b53_chip_data b53_switch_chips[] = { .duplex_reg = B53_DUPLEX_STAT_GE, .jumbo_pm_reg = B53_JUMBO_PORT_MASK, .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + .arl_ops = &b53_arl_ops_95, }, { .chip_id = BCM5397_DEVICE_ID, @@ -2769,6 +2775,7 @@ static const struct b53_chip_data b53_switch_chips[] = { .duplex_reg = B53_DUPLEX_STAT_GE, .jumbo_pm_reg = B53_JUMBO_PORT_MASK, .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + .arl_ops = &b53_arl_ops_95, }, { .chip_id = BCM5398_DEVICE_ID, @@ -2782,6 +2789,7 @@ static const struct b53_chip_data b53_switch_chips[] = { .duplex_reg = B53_DUPLEX_STAT_GE, .jumbo_pm_reg = B53_JUMBO_PORT_MASK, .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + .arl_ops = &b53_arl_ops_95, }, { .chip_id = BCM53101_DEVICE_ID, @@ -2795,6 +2803,7 @@ static const struct b53_chip_data b53_switch_chips[] = { .duplex_reg = B53_DUPLEX_STAT_GE, .jumbo_pm_reg = B53_JUMBO_PORT_MASK, .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + .arl_ops = &b53_arl_ops_95, }, { .chip_id = BCM53115_DEVICE_ID, @@ -2808,6 +2817,7 @@ static const struct b53_chip_data b53_switch_chips[] = { .duplex_reg = B53_DUPLEX_STAT_GE, .jumbo_pm_reg = B53_JUMBO_PORT_MASK, .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + .arl_ops = &b53_arl_ops_95, }, { .chip_id = BCM53125_DEVICE_ID, @@ -2821,6 +2831,7 @@ static const struct b53_chip_data b53_switch_chips[] = { .duplex_reg = B53_DUPLEX_STAT_GE, .jumbo_pm_reg = B53_JUMBO_PORT_MASK, .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + .arl_ops = &b53_arl_ops_95, }, { .chip_id = BCM53128_DEVICE_ID, @@ -2834,6 +2845,7 @@ static const struct b53_chip_data b53_switch_chips[] = { .duplex_reg = B53_DUPLEX_STAT_GE, .jumbo_pm_reg = B53_JUMBO_PORT_MASK, .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + .arl_ops = &b53_arl_ops_95, }, { .chip_id = BCM63XX_DEVICE_ID, @@ -2847,6 +2859,7 @@ static const struct b53_chip_data b53_switch_chips[] = { .duplex_reg = B53_DUPLEX_STAT_63XX, .jumbo_pm_reg = B53_JUMBO_PORT_MASK_63XX, .jumbo_size_reg = B53_JUMBO_MAX_SIZE_63XX, + .arl_ops = &b53_arl_ops_95, }, { .chip_id = BCM53010_DEVICE_ID, @@ -2860,6 +2873,7 @@ static const struct b53_chip_data b53_switch_chips[] = { .duplex_reg = B53_DUPLEX_STAT_GE, .jumbo_pm_reg = B53_JUMBO_PORT_MASK, .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + .arl_ops = &b53_arl_ops_95, }, { .chip_id = BCM53011_DEVICE_ID, @@ -2873,6 +2887,7 @@ static const struct b53_chip_data b53_switch_chips[] = { .duplex_reg = B53_DUPLEX_STAT_GE, .jumbo_pm_reg = B53_JUMBO_PORT_MASK, .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + .arl_ops = &b53_arl_ops_95, }, { .chip_id = BCM53012_DEVICE_ID, @@ -2886,6 +2901,7 @@ static const struct b53_chip_data b53_switch_chips[] = { .duplex_reg = B53_DUPLEX_STAT_GE, .jumbo_pm_reg = B53_JUMBO_PORT_MASK, .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + .arl_ops = &b53_arl_ops_95, }, { .chip_id = BCM53018_DEVICE_ID, @@ -2899,6 +2915,7 @@ static const struct b53_chip_data b53_switch_chips[] = { .duplex_reg = B53_DUPLEX_STAT_GE, .jumbo_pm_reg = B53_JUMBO_PORT_MASK, .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + .arl_ops = &b53_arl_ops_95, }, { .chip_id = BCM53019_DEVICE_ID, @@ -2912,6 +2929,7 @@ static const struct b53_chip_data b53_switch_chips[] = { .duplex_reg = B53_DUPLEX_STAT_GE, .jumbo_pm_reg = B53_JUMBO_PORT_MASK, .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + .arl_ops = &b53_arl_ops_95, }, { .chip_id = BCM58XX_DEVICE_ID, @@ -2925,6 +2943,7 @@ static const struct b53_chip_data b53_switch_chips[] = { .duplex_reg = B53_DUPLEX_STAT_GE, .jumbo_pm_reg = B53_JUMBO_PORT_MASK, .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + .arl_ops = &b53_arl_ops_95, }, { .chip_id = BCM583XX_DEVICE_ID, @@ -2938,6 +2957,7 @@ static const struct b53_chip_data b53_switch_chips[] = { .duplex_reg = B53_DUPLEX_STAT_GE, .jumbo_pm_reg = B53_JUMBO_PORT_MASK, .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + .arl_ops = &b53_arl_ops_95, }, /* Starfighter 2 */ { @@ -2952,6 +2972,7 @@ static const struct b53_chip_data b53_switch_chips[] = { .duplex_reg = B53_DUPLEX_STAT_GE, .jumbo_pm_reg = B53_JUMBO_PORT_MASK, .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + .arl_ops = &b53_arl_ops_95, }, { .chip_id = BCM7445_DEVICE_ID, @@ -2965,6 +2986,7 @@ static const struct b53_chip_data b53_switch_chips[] = { .duplex_reg = B53_DUPLEX_STAT_GE, .jumbo_pm_reg = B53_JUMBO_PORT_MASK, .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + .arl_ops = &b53_arl_ops_95, }, { .chip_id = BCM7278_DEVICE_ID, @@ -2978,6 +3000,7 @@ static const struct b53_chip_data b53_switch_chips[] = { .duplex_reg = B53_DUPLEX_STAT_GE, .jumbo_pm_reg = B53_JUMBO_PORT_MASK, .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + .arl_ops = &b53_arl_ops_95, }, { .chip_id = BCM53134_DEVICE_ID, @@ -2992,6 +3015,7 @@ static const struct b53_chip_data b53_switch_chips[] = { .duplex_reg = B53_DUPLEX_STAT_GE, .jumbo_pm_reg = B53_JUMBO_PORT_MASK, .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + .arl_ops = &b53_arl_ops_95, }, }; @@ -3020,6 +3044,7 @@ static int b53_switch_init(struct b53_device *dev) dev->num_vlans = chip->vlans; dev->num_arl_bins = chip->arl_bins; dev->num_arl_buckets = chip->arl_buckets; + dev->arl_ops = chip->arl_ops; break; } } diff --git a/drivers/net/dsa/b53/b53_priv.h b/drivers/net/dsa/b53/b53_priv.h index 458775f95164..ef2413509b5d 100644 --- a/drivers/net/dsa/b53/b53_priv.h +++ b/drivers/net/dsa/b53/b53_priv.h @@ -58,6 +58,17 @@ struct b53_io_ops { bool link_up); }; +struct b53_arl_entry; + +struct b53_arl_ops { + void (*arl_read_entry)(struct b53_device *dev, + struct b53_arl_entry *ent, u8 idx); + void (*arl_write_entry)(struct b53_device *dev, + const struct b53_arl_entry *ent, u8 idx); + void (*arl_search_read)(struct b53_device *dev, u8 idx, + struct b53_arl_entry *ent); +}; + #define B53_INVALID_LANE 0xff enum { @@ -127,6 +138,7 @@ struct b53_device { struct mutex stats_mutex; struct mutex arl_mutex; const struct b53_io_ops *ops; + const struct b53_arl_ops *arl_ops; /* chip specific data */ u32 chip_id; @@ -371,6 +383,24 @@ static inline void b53_arl_from_entry_25(u64 *mac_vid, *mac_vid |= ARLTBL_AGE_25; } +static inline void b53_arl_read_entry(struct b53_device *dev, + struct b53_arl_entry *ent, u8 idx) +{ + dev->arl_ops->arl_read_entry(dev, ent, idx); +} + +static inline void b53_arl_write_entry(struct b53_device *dev, + const struct b53_arl_entry *ent, u8 idx) +{ + dev->arl_ops->arl_write_entry(dev, ent, idx); +} + +static inline void b53_arl_search_read(struct b53_device *dev, u8 idx, + struct b53_arl_entry *ent) +{ + dev->arl_ops->arl_search_read(dev, idx, ent); +} + #ifdef CONFIG_BCM47XX #include From 300f78e8b6b7be17c2c78afeded75be68acb1aa7 Mon Sep 17 00:00:00 2001 From: Jonas Gorski Date: Fri, 7 Nov 2025 09:07:48 +0100 Subject: [PATCH 0679/1381] net: dsa: b53: add support for 5389/5397/5398 ARL entry format BCM5389, BCM5397 and BCM5398 use a different ARL entry format with just a 16 bit fwdentry register, as well as different search control and data offsets. So add appropriate ops for them and switch those chips to use them. Signed-off-by: Jonas Gorski Reviewed-by: Florian Fainelli Link: https://patch.msgid.link/20251107080749.26936-8-jonas.gorski@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/b53/b53_common.c | 53 ++++++++++++++++++++++++++++++-- drivers/net/dsa/b53/b53_priv.h | 26 ++++++++++++++++ drivers/net/dsa/b53/b53_regs.h | 13 ++++++++ 3 files changed, 89 insertions(+), 3 deletions(-) diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c index c69022cc85bf..73ea9adb95b7 100644 --- a/drivers/net/dsa/b53/b53_common.c +++ b/drivers/net/dsa/b53/b53_common.c @@ -1870,6 +1870,31 @@ static void b53_arl_write_entry_25(struct b53_device *dev, mac_vid); } +static void b53_arl_read_entry_89(struct b53_device *dev, + struct b53_arl_entry *ent, u8 idx) +{ + u64 mac_vid; + u16 fwd_entry; + + b53_read64(dev, B53_ARLIO_PAGE, B53_ARLTBL_MAC_VID_ENTRY(idx), + &mac_vid); + b53_read16(dev, B53_ARLIO_PAGE, B53_ARLTBL_DATA_ENTRY(idx), &fwd_entry); + b53_arl_to_entry_89(ent, mac_vid, fwd_entry); +} + +static void b53_arl_write_entry_89(struct b53_device *dev, + const struct b53_arl_entry *ent, u8 idx) +{ + u32 fwd_entry; + u64 mac_vid; + + b53_arl_from_entry_89(&mac_vid, &fwd_entry, ent); + b53_write64(dev, B53_ARLIO_PAGE, + B53_ARLTBL_MAC_VID_ENTRY(idx), mac_vid); + b53_write16(dev, B53_ARLIO_PAGE, + B53_ARLTBL_DATA_ENTRY(idx), fwd_entry); +} + static void b53_arl_read_entry_95(struct b53_device *dev, struct b53_arl_entry *ent, u8 idx) { @@ -2033,6 +2058,8 @@ static void b53_read_arl_srch_ctl(struct b53_device *dev, u8 *val) if (is5325(dev) || is5365(dev)) offset = B53_ARL_SRCH_CTL_25; + else if (dev->chip_id == BCM5389_DEVICE_ID || is5397_98(dev)) + offset = B53_ARL_SRCH_CTL_89; else offset = B53_ARL_SRCH_CTL; @@ -2045,6 +2072,8 @@ static void b53_write_arl_srch_ctl(struct b53_device *dev, u8 val) if (is5325(dev) || is5365(dev)) offset = B53_ARL_SRCH_CTL_25; + else if (dev->chip_id == BCM5389_DEVICE_ID || is5397_98(dev)) + offset = B53_ARL_SRCH_CTL_89; else offset = B53_ARL_SRCH_CTL; @@ -2090,6 +2119,18 @@ static void b53_arl_search_read_65(struct b53_device *dev, u8 idx, b53_arl_to_entry_25(ent, mac_vid); } +static void b53_arl_search_read_89(struct b53_device *dev, u8 idx, + struct b53_arl_entry *ent) +{ + u16 fwd_entry; + u64 mac_vid; + + b53_read64(dev, B53_ARLIO_PAGE, B53_ARL_SRCH_RSLT_MACVID_89, + &mac_vid); + b53_read16(dev, B53_ARLIO_PAGE, B53_ARL_SRCH_RSLT_89, &fwd_entry); + b53_arl_to_entry_89(ent, mac_vid, fwd_entry); +} + static void b53_arl_search_read_95(struct b53_device *dev, u8 idx, struct b53_arl_entry *ent) { @@ -2683,6 +2724,12 @@ static const struct b53_arl_ops b53_arl_ops_65 = { .arl_search_read = b53_arl_search_read_65, }; +static const struct b53_arl_ops b53_arl_ops_89 = { + .arl_read_entry = b53_arl_read_entry_89, + .arl_write_entry = b53_arl_write_entry_89, + .arl_search_read = b53_arl_search_read_89, +}; + static const struct b53_arl_ops b53_arl_ops_95 = { .arl_read_entry = b53_arl_read_entry_95, .arl_write_entry = b53_arl_write_entry_95, @@ -2747,7 +2794,7 @@ static const struct b53_chip_data b53_switch_chips[] = { .duplex_reg = B53_DUPLEX_STAT_GE, .jumbo_pm_reg = B53_JUMBO_PORT_MASK, .jumbo_size_reg = B53_JUMBO_MAX_SIZE, - .arl_ops = &b53_arl_ops_95, + .arl_ops = &b53_arl_ops_89, }, { .chip_id = BCM5395_DEVICE_ID, @@ -2775,7 +2822,7 @@ static const struct b53_chip_data b53_switch_chips[] = { .duplex_reg = B53_DUPLEX_STAT_GE, .jumbo_pm_reg = B53_JUMBO_PORT_MASK, .jumbo_size_reg = B53_JUMBO_MAX_SIZE, - .arl_ops = &b53_arl_ops_95, + .arl_ops = &b53_arl_ops_89, }, { .chip_id = BCM5398_DEVICE_ID, @@ -2789,7 +2836,7 @@ static const struct b53_chip_data b53_switch_chips[] = { .duplex_reg = B53_DUPLEX_STAT_GE, .jumbo_pm_reg = B53_JUMBO_PORT_MASK, .jumbo_size_reg = B53_JUMBO_MAX_SIZE, - .arl_ops = &b53_arl_ops_95, + .arl_ops = &b53_arl_ops_89, }, { .chip_id = BCM53101_DEVICE_ID, diff --git a/drivers/net/dsa/b53/b53_priv.h b/drivers/net/dsa/b53/b53_priv.h index ef2413509b5d..d6d25bb3945b 100644 --- a/drivers/net/dsa/b53/b53_priv.h +++ b/drivers/net/dsa/b53/b53_priv.h @@ -353,6 +353,18 @@ static inline void b53_arl_to_entry_25(struct b53_arl_entry *ent, ent->vid = mac_vid >> ARLTBL_VID_S_65; } +static inline void b53_arl_to_entry_89(struct b53_arl_entry *ent, + u64 mac_vid, u16 fwd_entry) +{ + memset(ent, 0, sizeof(*ent)); + ent->port = fwd_entry & ARLTBL_DATA_PORT_ID_MASK_89; + ent->is_valid = !!(fwd_entry & ARLTBL_VALID_89); + ent->is_age = !!(fwd_entry & ARLTBL_AGE_89); + ent->is_static = !!(fwd_entry & ARLTBL_STATIC_89); + u64_to_ether_addr(mac_vid, ent->mac); + ent->vid = mac_vid >> ARLTBL_VID_S; +} + static inline void b53_arl_from_entry(u64 *mac_vid, u32 *fwd_entry, const struct b53_arl_entry *ent) { @@ -383,6 +395,20 @@ static inline void b53_arl_from_entry_25(u64 *mac_vid, *mac_vid |= ARLTBL_AGE_25; } +static inline void b53_arl_from_entry_89(u64 *mac_vid, u32 *fwd_entry, + const struct b53_arl_entry *ent) +{ + *mac_vid = ether_addr_to_u64(ent->mac); + *mac_vid |= (u64)(ent->vid & ARLTBL_VID_MASK) << ARLTBL_VID_S; + *fwd_entry = ent->port & ARLTBL_DATA_PORT_ID_MASK_89; + if (ent->is_valid) + *fwd_entry |= ARLTBL_VALID_89; + if (ent->is_static) + *fwd_entry |= ARLTBL_STATIC_89; + if (ent->is_age) + *fwd_entry |= ARLTBL_AGE_89; +} + static inline void b53_arl_read_entry(struct b53_device *dev, struct b53_arl_entry *ent, u8 idx) { diff --git a/drivers/net/dsa/b53/b53_regs.h b/drivers/net/dsa/b53/b53_regs.h index c36a3dfb2ee8..c303507d3034 100644 --- a/drivers/net/dsa/b53/b53_regs.h +++ b/drivers/net/dsa/b53/b53_regs.h @@ -346,12 +346,20 @@ #define ARLTBL_STATIC BIT(15) #define ARLTBL_VALID BIT(16) +/* BCM5389 ARL Table Data Entry N Register format (16 bit) */ +#define ARLTBL_DATA_PORT_ID_MASK_89 GENMASK(8, 0) +#define ARLTBL_TC_MASK_89 GENMASK(12, 10) +#define ARLTBL_AGE_89 BIT(13) +#define ARLTBL_STATIC_89 BIT(14) +#define ARLTBL_VALID_89 BIT(15) + /* Maximum number of bin entries in the ARL for all switches */ #define B53_ARLTBL_MAX_BIN_ENTRIES 4 /* ARL Search Control Register (8 bit) */ #define B53_ARL_SRCH_CTL 0x50 #define B53_ARL_SRCH_CTL_25 0x20 +#define B53_ARL_SRCH_CTL_89 0x30 #define ARL_SRCH_VLID BIT(0) #define ARL_SRCH_STDN BIT(7) @@ -359,10 +367,12 @@ #define B53_ARL_SRCH_ADDR 0x51 #define B53_ARL_SRCH_ADDR_25 0x22 #define B53_ARL_SRCH_ADDR_65 0x24 +#define B53_ARL_SRCH_ADDR_89 0x31 #define ARL_ADDR_MASK GENMASK(14, 0) /* ARL Search MAC/VID Result (64 bit) */ #define B53_ARL_SRCH_RSTL_0_MACVID 0x60 +#define B53_ARL_SRCH_RSLT_MACVID_89 0x33 /* Single register search result on 5325 */ #define B53_ARL_SRCH_RSTL_0_MACVID_25 0x24 @@ -372,6 +382,9 @@ /* ARL Search Data Result (32 bit) */ #define B53_ARL_SRCH_RSTL_0 0x68 +/* BCM5389 ARL Search Data Result (16 bit) */ +#define B53_ARL_SRCH_RSLT_89 0x3b + #define B53_ARL_SRCH_RSTL_MACVID(x) (B53_ARL_SRCH_RSTL_0_MACVID + ((x) * 0x10)) #define B53_ARL_SRCH_RSTL(x) (B53_ARL_SRCH_RSTL_0 + ((x) * 0x10)) From 2b3013ac03028a2364d8779719bb6bfbc0212435 Mon Sep 17 00:00:00 2001 From: Jonas Gorski Date: Fri, 7 Nov 2025 09:07:49 +0100 Subject: [PATCH 0680/1381] net: dsa: b53: add support for bcm63xx ARL entry format The ARL registers of BCM63XX embedded switches are somewhat unique. The normal ARL table access registers have the same format as BCM5389, but the ARL search registers differ: * SRCH_CTL is at the same offset of BCM5389, but 16 bits wide. It does not have more fields, just needs to be accessed by a 16 bit read. * SRCH_RSLT_MACVID and SRCH_RSLT are aligned to 32 bit, and have shifted offsets. * SRCH_RSLT has a different format than the normal ARL data entry register. * There is only one set of ENTRY_N registers, implying a 1 bin layout. So add appropriate ops for bcm63xx and let it use it. Signed-off-by: Jonas Gorski Reviewed-by: Florian Fainelli Link: https://patch.msgid.link/20251107080749.26936-9-jonas.gorski@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/b53/b53_common.c | 44 +++++++++++++++++++++++++++----- drivers/net/dsa/b53/b53_priv.h | 15 +++++++++++ drivers/net/dsa/b53/b53_regs.h | 9 +++++++ 3 files changed, 61 insertions(+), 7 deletions(-) diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c index 73ea9adb95b7..72c85cd34a4e 100644 --- a/drivers/net/dsa/b53/b53_common.c +++ b/drivers/net/dsa/b53/b53_common.c @@ -2058,12 +2058,20 @@ static void b53_read_arl_srch_ctl(struct b53_device *dev, u8 *val) if (is5325(dev) || is5365(dev)) offset = B53_ARL_SRCH_CTL_25; - else if (dev->chip_id == BCM5389_DEVICE_ID || is5397_98(dev)) + else if (dev->chip_id == BCM5389_DEVICE_ID || is5397_98(dev) || + is63xx(dev)) offset = B53_ARL_SRCH_CTL_89; else offset = B53_ARL_SRCH_CTL; - b53_read8(dev, B53_ARLIO_PAGE, offset, val); + if (is63xx(dev)) { + u16 val16; + + b53_read16(dev, B53_ARLIO_PAGE, offset, &val16); + *val = val16 & 0xff; + } else { + b53_read8(dev, B53_ARLIO_PAGE, offset, val); + } } static void b53_write_arl_srch_ctl(struct b53_device *dev, u8 val) @@ -2072,12 +2080,16 @@ static void b53_write_arl_srch_ctl(struct b53_device *dev, u8 val) if (is5325(dev) || is5365(dev)) offset = B53_ARL_SRCH_CTL_25; - else if (dev->chip_id == BCM5389_DEVICE_ID || is5397_98(dev)) + else if (dev->chip_id == BCM5389_DEVICE_ID || is5397_98(dev) || + is63xx(dev)) offset = B53_ARL_SRCH_CTL_89; else offset = B53_ARL_SRCH_CTL; - b53_write8(dev, B53_ARLIO_PAGE, offset, val); + if (is63xx(dev)) + b53_write16(dev, B53_ARLIO_PAGE, offset, val); + else + b53_write8(dev, B53_ARLIO_PAGE, offset, val); } static int b53_arl_search_wait(struct b53_device *dev) @@ -2131,6 +2143,18 @@ static void b53_arl_search_read_89(struct b53_device *dev, u8 idx, b53_arl_to_entry_89(ent, mac_vid, fwd_entry); } +static void b53_arl_search_read_63xx(struct b53_device *dev, u8 idx, + struct b53_arl_entry *ent) +{ + u16 fwd_entry; + u64 mac_vid; + + b53_read64(dev, B53_ARLIO_PAGE, B53_ARL_SRCH_RSLT_MACVID_63XX, + &mac_vid); + b53_read16(dev, B53_ARLIO_PAGE, B53_ARL_SRCH_RSLT_63XX, &fwd_entry); + b53_arl_search_to_entry_63xx(ent, mac_vid, fwd_entry); +} + static void b53_arl_search_read_95(struct b53_device *dev, u8 idx, struct b53_arl_entry *ent) { @@ -2730,6 +2754,12 @@ static const struct b53_arl_ops b53_arl_ops_89 = { .arl_search_read = b53_arl_search_read_89, }; +static const struct b53_arl_ops b53_arl_ops_63xx = { + .arl_read_entry = b53_arl_read_entry_89, + .arl_write_entry = b53_arl_write_entry_89, + .arl_search_read = b53_arl_search_read_63xx, +}; + static const struct b53_arl_ops b53_arl_ops_95 = { .arl_read_entry = b53_arl_read_entry_95, .arl_write_entry = b53_arl_write_entry_95, @@ -2899,14 +2929,14 @@ static const struct b53_chip_data b53_switch_chips[] = { .dev_name = "BCM63xx", .vlans = 4096, .enabled_ports = 0, /* pdata must provide them */ - .arl_bins = 4, - .arl_buckets = 1024, + .arl_bins = 1, + .arl_buckets = 4096, .imp_port = 8, .vta_regs = B53_VTA_REGS_63XX, .duplex_reg = B53_DUPLEX_STAT_63XX, .jumbo_pm_reg = B53_JUMBO_PORT_MASK_63XX, .jumbo_size_reg = B53_JUMBO_MAX_SIZE_63XX, - .arl_ops = &b53_arl_ops_95, + .arl_ops = &b53_arl_ops_63xx, }, { .chip_id = BCM53010_DEVICE_ID, diff --git a/drivers/net/dsa/b53/b53_priv.h b/drivers/net/dsa/b53/b53_priv.h index d6d25bb3945b..2bfd0e7c95c9 100644 --- a/drivers/net/dsa/b53/b53_priv.h +++ b/drivers/net/dsa/b53/b53_priv.h @@ -409,6 +409,21 @@ static inline void b53_arl_from_entry_89(u64 *mac_vid, u32 *fwd_entry, *fwd_entry |= ARLTBL_AGE_89; } +static inline void b53_arl_search_to_entry_63xx(struct b53_arl_entry *ent, + u64 mac_vid, u16 fwd_entry) +{ + memset(ent, 0, sizeof(*ent)); + u64_to_ether_addr(mac_vid, ent->mac); + ent->vid = mac_vid >> ARLTBL_VID_S; + + ent->port = fwd_entry & ARL_SRST_PORT_ID_MASK_63XX; + ent->port >>= 1; + + ent->is_age = !!(fwd_entry & ARL_SRST_AGE_63XX); + ent->is_static = !!(fwd_entry & ARL_SRST_STATIC_63XX); + ent->is_valid = 1; +} + static inline void b53_arl_read_entry(struct b53_device *dev, struct b53_arl_entry *ent, u8 idx) { diff --git a/drivers/net/dsa/b53/b53_regs.h b/drivers/net/dsa/b53/b53_regs.h index c303507d3034..69ebbec932f6 100644 --- a/drivers/net/dsa/b53/b53_regs.h +++ b/drivers/net/dsa/b53/b53_regs.h @@ -368,11 +368,13 @@ #define B53_ARL_SRCH_ADDR_25 0x22 #define B53_ARL_SRCH_ADDR_65 0x24 #define B53_ARL_SRCH_ADDR_89 0x31 +#define B53_ARL_SRCH_ADDR_63XX 0x32 #define ARL_ADDR_MASK GENMASK(14, 0) /* ARL Search MAC/VID Result (64 bit) */ #define B53_ARL_SRCH_RSTL_0_MACVID 0x60 #define B53_ARL_SRCH_RSLT_MACVID_89 0x33 +#define B53_ARL_SRCH_RSLT_MACVID_63XX 0x34 /* Single register search result on 5325 */ #define B53_ARL_SRCH_RSTL_0_MACVID_25 0x24 @@ -388,6 +390,13 @@ #define B53_ARL_SRCH_RSTL_MACVID(x) (B53_ARL_SRCH_RSTL_0_MACVID + ((x) * 0x10)) #define B53_ARL_SRCH_RSTL(x) (B53_ARL_SRCH_RSTL_0 + ((x) * 0x10)) +/* 63XX ARL Search Data Result (16 bit) */ +#define B53_ARL_SRCH_RSLT_63XX 0x3c +#define ARL_SRST_PORT_ID_MASK_63XX GENMASK(9, 1) +#define ARL_SRST_TC_MASK_63XX GENMASK(13, 11) +#define ARL_SRST_AGE_63XX BIT(14) +#define ARL_SRST_STATIC_63XX BIT(15) + /************************************************************************* * IEEE 802.1X Registers *************************************************************************/ From 23c52b58cc38e398292f125e68b717001344444c Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Fri, 7 Nov 2025 02:36:59 -0800 Subject: [PATCH 0681/1381] tg3: Fix num of RX queues being reported by ethtool Using num_online_cpus() to report number of queues is actually not correct, as reported by Michael[1]. netif_get_num_default_rss_queues() was used to replace num_online_cpus() in the past, but tg3 ethtool callbacks didn't get converted. Doing it now. Link: https://lore.kernel.org/all/CACKFLim7ruspmqvjr6bNRq5Z_XXVk3vVaLZOons7kMCzsEG23A@mail.gmail.com/#t [1] Signed-off-by: Breno Leitao Suggested-by: Michael Chan Reviewed-by: Michael Chan Link: https://patch.msgid.link/20251107-tg3_counts-v1-1-337fe5c8ccb7@debian.org Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/broadcom/tg3.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/broadcom/tg3.c b/drivers/net/ethernet/broadcom/tg3.c index fa58c3ffceb0..e21f7c6a6de7 100644 --- a/drivers/net/ethernet/broadcom/tg3.c +++ b/drivers/net/ethernet/broadcom/tg3.c @@ -12729,7 +12729,7 @@ static u32 tg3_get_rx_ring_count(struct net_device *dev) if (netif_running(tp->dev)) return tp->rxq_cnt; - return min(num_online_cpus(), TG3_RSS_MAX_NUM_QS); + return min_t(u32, netif_get_num_default_rss_queues(), tp->rxq_max); } static u32 tg3_get_rxfh_indir_size(struct net_device *dev) From 7ff14c52049eafecdd72cd0a12cae6905876566a Mon Sep 17 00:00:00 2001 From: Simon Schippers Date: Thu, 6 Nov 2025 18:56:15 +0100 Subject: [PATCH 0682/1381] usbnet: Add support for Byte Queue Limits (BQL) In the current implementation, usbnet uses a fixed tx_qlen of: USB2: 60 * 1518 bytes = 91.08 KB USB3: 60 * 5 * 1518 bytes = 454.80 KB Such large transmit queues can be problematic, especially for cellular modems. For example, with a typical celluar link speed of 10 Mbit/s, a fully occupied USB3 transmit queue results in: 454.80 KB / (10 Mbit/s / 8 bit/byte) = 363.84 ms of additional latency. This patch adds support for Byte Queue Limits (BQL) [1] to dynamically manage the transmit queue size and reduce latency without sacrificing throughput. Testing was performed on various devices using the usbnet driver for packet transmission: - DELOCK 66045: USB3 to 2.5 GbE adapter (ax88179_178a) - DELOCK 61969: USB2 to 1 GbE adapter (asix) - Quectel RM520: 5G modem (qmi_wwan) - USB2 Android tethering (cdc_ncm) No performance degradation was observed for iperf3 TCP or UDP traffic, while latency for a prioritized ping application was significantly reduced. For example, using the USB3 to 2.5 GbE adapter, which was fully utilized by iperf3 UDP traffic, the prioritized ping was improved from 1.6 ms to 0.6 ms. With the same setup but with a 100 Mbit/s Ethernet connection, the prioritized ping was improved from 35 ms to 5 ms. [1] https://lwn.net/Articles/469652/ Signed-off-by: Simon Schippers Reviewed-by: Eric Dumazet Link: https://patch.msgid.link/20251106175615.26948-1-simon.schippers@tu-dortmund.de Signed-off-by: Jakub Kicinski --- drivers/net/usb/usbnet.c | 11 +++++++++++ include/linux/usb/usbnet.h | 2 ++ 2 files changed, 13 insertions(+) diff --git a/drivers/net/usb/usbnet.c b/drivers/net/usb/usbnet.c index f3087fb62f4f..3d10cf791c51 100644 --- a/drivers/net/usb/usbnet.c +++ b/drivers/net/usb/usbnet.c @@ -831,6 +831,7 @@ int usbnet_stop(struct net_device *net) clear_bit(EVENT_DEV_OPEN, &dev->flags); netif_stop_queue (net); + netdev_reset_queue(net); netif_info(dev, ifdown, dev->net, "stop stats: rx/tx %lu/%lu, errs %lu/%lu\n", @@ -939,6 +940,7 @@ int usbnet_open(struct net_device *net) } set_bit(EVENT_DEV_OPEN, &dev->flags); + netdev_reset_queue(net); netif_start_queue (net); netif_info(dev, ifup, dev->net, "open: enable queueing (rx %d, tx %d) mtu %d %s framing\n", @@ -1500,6 +1502,7 @@ netdev_tx_t usbnet_start_xmit(struct sk_buff *skb, struct net_device *net) case 0: netif_trans_update(net); __usbnet_queue_skb(&dev->txq, skb, tx_start); + netdev_sent_queue(net, skb->len); if (dev->txq.qlen >= TX_QLEN (dev)) netif_stop_queue (net); } @@ -1563,6 +1566,7 @@ static inline void usb_free_skb(struct sk_buff *skb) static void usbnet_bh(struct timer_list *t) { struct usbnet *dev = timer_container_of(dev, t, delay); + unsigned int bytes_compl = 0, pkts_compl = 0; struct sk_buff *skb; struct skb_data *entry; @@ -1574,6 +1578,8 @@ static void usbnet_bh(struct timer_list *t) usb_free_skb(skb); continue; case tx_done: + bytes_compl += skb->len; + pkts_compl++; kfree(entry->urb->sg); fallthrough; case rx_cleanup: @@ -1584,6 +1590,10 @@ static void usbnet_bh(struct timer_list *t) } } + spin_lock_bh(&dev->bql_spinlock); + netdev_completed_queue(dev->net, pkts_compl, bytes_compl); + spin_unlock_bh(&dev->bql_spinlock); + /* restart RX again after disabling due to high error rate */ clear_bit(EVENT_RX_KILL, &dev->flags); @@ -1755,6 +1765,7 @@ usbnet_probe(struct usb_interface *udev, const struct usb_device_id *prod) skb_queue_head_init (&dev->txq); skb_queue_head_init (&dev->done); skb_queue_head_init(&dev->rxq_pause); + spin_lock_init(&dev->bql_spinlock); INIT_WORK(&dev->bh_work, usbnet_bh_work); INIT_WORK (&dev->kevent, usbnet_deferred_kevent); init_usb_anchor(&dev->deferred); diff --git a/include/linux/usb/usbnet.h b/include/linux/usb/usbnet.h index a2d54122823d..2945923a8a95 100644 --- a/include/linux/usb/usbnet.h +++ b/include/linux/usb/usbnet.h @@ -14,6 +14,7 @@ #include #include #include +#include /* interface from usbnet core to each USB networking link we handle */ struct usbnet { @@ -59,6 +60,7 @@ struct usbnet { struct mutex interrupt_mutex; struct usb_anchor deferred; struct work_struct bh_work; + spinlock_t bql_spinlock; struct work_struct kevent; unsigned long flags; From 2dd63c36453408583c2c37a742705a0232612c57 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Fri, 7 Nov 2025 08:28:34 +0000 Subject: [PATCH 0683/1381] net: stmmac: ingenic: move ingenic_mac_init() Move ingenic_mac_init() to between variant specific set_mode() implementations and ingenic_mac_probe(). No code changes. Reviewed-by: Maxime Chevallier Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vHHpi-0000000Djqp-4910@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../ethernet/stmicro/stmmac/dwmac-ingenic.c | 28 +++++++++---------- 1 file changed, 14 insertions(+), 14 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c index c1670f6bae14..8d0627055799 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c @@ -71,20 +71,6 @@ struct ingenic_soc_info { int (*set_mode)(struct plat_stmmacenet_data *plat_dat); }; -static int ingenic_mac_init(struct platform_device *pdev, void *bsp_priv) -{ - struct ingenic_mac *mac = bsp_priv; - int ret; - - if (mac->soc_info->set_mode) { - ret = mac->soc_info->set_mode(mac->plat_dat); - if (ret) - return ret; - } - - return 0; -} - static int jz4775_mac_set_mode(struct plat_stmmacenet_data *plat_dat) { struct ingenic_mac *mac = plat_dat->bsp_priv; @@ -234,6 +220,20 @@ static int x2000_mac_set_mode(struct plat_stmmacenet_data *plat_dat) return regmap_update_bits(mac->regmap, 0, mac->soc_info->mask, val); } +static int ingenic_mac_init(struct platform_device *pdev, void *bsp_priv) +{ + struct ingenic_mac *mac = bsp_priv; + int ret; + + if (mac->soc_info->set_mode) { + ret = mac->soc_info->set_mode(mac->plat_dat); + if (ret) + return ret; + } + + return 0; +} + static int ingenic_mac_probe(struct platform_device *pdev) { struct plat_stmmacenet_data *plat_dat; From 307a575775fd06b1a63ec017a63a82485bedf62f Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Fri, 7 Nov 2025 08:28:40 +0000 Subject: [PATCH 0684/1381] net: stmmac: ingenic: simplify jz4775 mac_set_mode() All paths configure the transmit clock as an input. Move this out of the switch() statement to simplify the code. Reviewed-by: Maxime Chevallier Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vHHpo-0000000Djqv-0RD4@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../net/ethernet/stmicro/stmmac/dwmac-ingenic.c | 14 ++++++-------- 1 file changed, 6 insertions(+), 8 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c index 8d0627055799..c6c82f277f62 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c @@ -78,20 +78,17 @@ static int jz4775_mac_set_mode(struct plat_stmmacenet_data *plat_dat) switch (plat_dat->phy_interface) { case PHY_INTERFACE_MODE_MII: - val = FIELD_PREP(MACPHYC_TXCLK_SEL_MASK, MACPHYC_TXCLK_SEL_INPUT) | - FIELD_PREP(MACPHYC_PHY_INFT_MASK, MACPHYC_PHY_INFT_MII); + val = FIELD_PREP(MACPHYC_PHY_INFT_MASK, MACPHYC_PHY_INFT_MII); dev_dbg(mac->dev, "MAC PHY Control Register: PHY_INTERFACE_MODE_MII\n"); break; case PHY_INTERFACE_MODE_GMII: - val = FIELD_PREP(MACPHYC_TXCLK_SEL_MASK, MACPHYC_TXCLK_SEL_INPUT) | - FIELD_PREP(MACPHYC_PHY_INFT_MASK, MACPHYC_PHY_INFT_GMII); + val = FIELD_PREP(MACPHYC_PHY_INFT_MASK, MACPHYC_PHY_INFT_GMII); dev_dbg(mac->dev, "MAC PHY Control Register: PHY_INTERFACE_MODE_GMII\n"); break; case PHY_INTERFACE_MODE_RMII: - val = FIELD_PREP(MACPHYC_TXCLK_SEL_MASK, MACPHYC_TXCLK_SEL_INPUT) | - FIELD_PREP(MACPHYC_PHY_INFT_MASK, MACPHYC_PHY_INFT_RMII); + val = FIELD_PREP(MACPHYC_PHY_INFT_MASK, MACPHYC_PHY_INFT_RMII); dev_dbg(mac->dev, "MAC PHY Control Register: PHY_INTERFACE_MODE_RMII\n"); break; @@ -99,8 +96,7 @@ static int jz4775_mac_set_mode(struct plat_stmmacenet_data *plat_dat) case PHY_INTERFACE_MODE_RGMII_ID: case PHY_INTERFACE_MODE_RGMII_TXID: case PHY_INTERFACE_MODE_RGMII_RXID: - val = FIELD_PREP(MACPHYC_TXCLK_SEL_MASK, MACPHYC_TXCLK_SEL_INPUT) | - FIELD_PREP(MACPHYC_PHY_INFT_MASK, MACPHYC_PHY_INFT_RGMII); + val = FIELD_PREP(MACPHYC_PHY_INFT_MASK, MACPHYC_PHY_INFT_RGMII); dev_dbg(mac->dev, "MAC PHY Control Register: PHY_INTERFACE_MODE_RGMII\n"); break; @@ -110,6 +106,8 @@ static int jz4775_mac_set_mode(struct plat_stmmacenet_data *plat_dat) return -EINVAL; } + val |= FIELD_PREP(MACPHYC_TXCLK_SEL_MASK, MACPHYC_TXCLK_SEL_INPUT); + /* Update MAC PHY control register */ return regmap_update_bits(mac->regmap, 0, mac->soc_info->mask, val); } From da6e9fd1046f1b936d8dd1db738ffb06f21851c0 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Fri, 7 Nov 2025 08:28:45 +0000 Subject: [PATCH 0685/1381] net: stmmac: ingenic: use PHY_INTF_SEL_x to select PHY interface Use the common dwmac definitions for the PHY interface selection field. Reviewed-by: Maxime Chevallier Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vHHpt-0000000Djr1-0wwr@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c index c6c82f277f62..5de2bd984d34 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c @@ -35,10 +35,10 @@ #define MACPHYC_RX_DELAY_MASK GENMASK(10, 4) #define MACPHYC_SOFT_RST_MASK GENMASK(3, 3) #define MACPHYC_PHY_INFT_MASK GENMASK(2, 0) -#define MACPHYC_PHY_INFT_RMII 0x4 -#define MACPHYC_PHY_INFT_RGMII 0x1 -#define MACPHYC_PHY_INFT_GMII 0x0 -#define MACPHYC_PHY_INFT_MII 0x0 +#define MACPHYC_PHY_INFT_RMII PHY_INTF_SEL_RMII +#define MACPHYC_PHY_INFT_RGMII PHY_INTF_SEL_RGMII +#define MACPHYC_PHY_INFT_GMII PHY_INTF_SEL_GMII_MII +#define MACPHYC_PHY_INFT_MII PHY_INTF_SEL_GMII_MII #define MACPHYC_TX_DELAY_PS_MAX 2496 #define MACPHYC_TX_DELAY_PS_MIN 20 From dbf99dc7d166944df4af1f52351a355512e1c573 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Fri, 7 Nov 2025 08:28:50 +0000 Subject: [PATCH 0686/1381] net: stmmac: ingenic: use PHY_INTF_SEL_x directly Use the PHY_INTF_SEL_x values directly in each of the mac_set_mode methods rather than the driver private MACPHYC_PHY_INFT_x definitions. Remove the MACPHYC_PHY_INFT_x definitions. Reviewed-by: Maxime Chevallier Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vHHpy-0000000Djr7-1R1m@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../ethernet/stmicro/stmmac/dwmac-ingenic.c | 20 ++++++++----------- 1 file changed, 8 insertions(+), 12 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c index 5de2bd984d34..b56d7ada1939 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c @@ -35,10 +35,6 @@ #define MACPHYC_RX_DELAY_MASK GENMASK(10, 4) #define MACPHYC_SOFT_RST_MASK GENMASK(3, 3) #define MACPHYC_PHY_INFT_MASK GENMASK(2, 0) -#define MACPHYC_PHY_INFT_RMII PHY_INTF_SEL_RMII -#define MACPHYC_PHY_INFT_RGMII PHY_INTF_SEL_RGMII -#define MACPHYC_PHY_INFT_GMII PHY_INTF_SEL_GMII_MII -#define MACPHYC_PHY_INFT_MII PHY_INTF_SEL_GMII_MII #define MACPHYC_TX_DELAY_PS_MAX 2496 #define MACPHYC_TX_DELAY_PS_MIN 20 @@ -78,17 +74,17 @@ static int jz4775_mac_set_mode(struct plat_stmmacenet_data *plat_dat) switch (plat_dat->phy_interface) { case PHY_INTERFACE_MODE_MII: - val = FIELD_PREP(MACPHYC_PHY_INFT_MASK, MACPHYC_PHY_INFT_MII); + val = FIELD_PREP(MACPHYC_PHY_INFT_MASK, PHY_INTF_SEL_GMII_MII); dev_dbg(mac->dev, "MAC PHY Control Register: PHY_INTERFACE_MODE_MII\n"); break; case PHY_INTERFACE_MODE_GMII: - val = FIELD_PREP(MACPHYC_PHY_INFT_MASK, MACPHYC_PHY_INFT_GMII); + val = FIELD_PREP(MACPHYC_PHY_INFT_MASK, PHY_INTF_SEL_GMII_MII); dev_dbg(mac->dev, "MAC PHY Control Register: PHY_INTERFACE_MODE_GMII\n"); break; case PHY_INTERFACE_MODE_RMII: - val = FIELD_PREP(MACPHYC_PHY_INFT_MASK, MACPHYC_PHY_INFT_RMII); + val = FIELD_PREP(MACPHYC_PHY_INFT_MASK, PHY_INTF_SEL_RMII); dev_dbg(mac->dev, "MAC PHY Control Register: PHY_INTERFACE_MODE_RMII\n"); break; @@ -96,7 +92,7 @@ static int jz4775_mac_set_mode(struct plat_stmmacenet_data *plat_dat) case PHY_INTERFACE_MODE_RGMII_ID: case PHY_INTERFACE_MODE_RGMII_TXID: case PHY_INTERFACE_MODE_RGMII_RXID: - val = FIELD_PREP(MACPHYC_PHY_INFT_MASK, MACPHYC_PHY_INFT_RGMII); + val = FIELD_PREP(MACPHYC_PHY_INFT_MASK, PHY_INTF_SEL_RGMII); dev_dbg(mac->dev, "MAC PHY Control Register: PHY_INTERFACE_MODE_RGMII\n"); break; @@ -138,7 +134,7 @@ static int x1600_mac_set_mode(struct plat_stmmacenet_data *plat_dat) switch (plat_dat->phy_interface) { case PHY_INTERFACE_MODE_RMII: - val = FIELD_PREP(MACPHYC_PHY_INFT_MASK, MACPHYC_PHY_INFT_RMII); + val = FIELD_PREP(MACPHYC_PHY_INFT_MASK, PHY_INTF_SEL_RMII); dev_dbg(mac->dev, "MAC PHY Control Register: PHY_INTERFACE_MODE_RMII\n"); break; @@ -160,7 +156,7 @@ static int x1830_mac_set_mode(struct plat_stmmacenet_data *plat_dat) switch (plat_dat->phy_interface) { case PHY_INTERFACE_MODE_RMII: val = FIELD_PREP(MACPHYC_MODE_SEL_MASK, MACPHYC_MODE_SEL_RMII) | - FIELD_PREP(MACPHYC_PHY_INFT_MASK, MACPHYC_PHY_INFT_RMII); + FIELD_PREP(MACPHYC_PHY_INFT_MASK, PHY_INTF_SEL_RMII); dev_dbg(mac->dev, "MAC PHY Control Register: PHY_INTERFACE_MODE_RMII\n"); break; @@ -183,7 +179,7 @@ static int x2000_mac_set_mode(struct plat_stmmacenet_data *plat_dat) case PHY_INTERFACE_MODE_RMII: val = FIELD_PREP(MACPHYC_TX_SEL_MASK, MACPHYC_TX_SEL_ORIGIN) | FIELD_PREP(MACPHYC_RX_SEL_MASK, MACPHYC_RX_SEL_ORIGIN) | - FIELD_PREP(MACPHYC_PHY_INFT_MASK, MACPHYC_PHY_INFT_RMII); + FIELD_PREP(MACPHYC_PHY_INFT_MASK, PHY_INTF_SEL_RMII); dev_dbg(mac->dev, "MAC PHY Control Register: PHY_INTERFACE_MODE_RMII\n"); break; @@ -191,7 +187,7 @@ static int x2000_mac_set_mode(struct plat_stmmacenet_data *plat_dat) case PHY_INTERFACE_MODE_RGMII_ID: case PHY_INTERFACE_MODE_RGMII_TXID: case PHY_INTERFACE_MODE_RGMII_RXID: - val = FIELD_PREP(MACPHYC_PHY_INFT_MASK, MACPHYC_PHY_INFT_RGMII); + val = FIELD_PREP(MACPHYC_PHY_INFT_MASK, PHY_INTF_SEL_RGMII); if (mac->tx_delay == 0) val |= FIELD_PREP(MACPHYC_TX_SEL_MASK, MACPHYC_TX_SEL_ORIGIN); From 14497aaa5eb63f7d77cf72c281afe330baea970b Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Fri, 7 Nov 2025 08:28:55 +0000 Subject: [PATCH 0687/1381] net: stmmac: ingenic: prep PHY_INTF_SEL_x field after switch() Move the preparation of the PHY_INTF_SEL_x bitfield out of the switch() statement such that it only appears once. Reviewed-by: Maxime Chevallier Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vHHq3-0000000DjrD-1u8O@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../ethernet/stmicro/stmmac/dwmac-ingenic.c | 34 +++++++++++++------ 1 file changed, 23 insertions(+), 11 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c index b56d7ada1939..6680f7d3a469 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c @@ -71,20 +71,21 @@ static int jz4775_mac_set_mode(struct plat_stmmacenet_data *plat_dat) { struct ingenic_mac *mac = plat_dat->bsp_priv; unsigned int val; + u8 phy_intf_sel; switch (plat_dat->phy_interface) { case PHY_INTERFACE_MODE_MII: - val = FIELD_PREP(MACPHYC_PHY_INFT_MASK, PHY_INTF_SEL_GMII_MII); + phy_intf_sel = PHY_INTF_SEL_GMII_MII; dev_dbg(mac->dev, "MAC PHY Control Register: PHY_INTERFACE_MODE_MII\n"); break; case PHY_INTERFACE_MODE_GMII: - val = FIELD_PREP(MACPHYC_PHY_INFT_MASK, PHY_INTF_SEL_GMII_MII); + phy_intf_sel = PHY_INTF_SEL_GMII_MII; dev_dbg(mac->dev, "MAC PHY Control Register: PHY_INTERFACE_MODE_GMII\n"); break; case PHY_INTERFACE_MODE_RMII: - val = FIELD_PREP(MACPHYC_PHY_INFT_MASK, PHY_INTF_SEL_RMII); + phy_intf_sel = PHY_INTF_SEL_RMII; dev_dbg(mac->dev, "MAC PHY Control Register: PHY_INTERFACE_MODE_RMII\n"); break; @@ -92,7 +93,7 @@ static int jz4775_mac_set_mode(struct plat_stmmacenet_data *plat_dat) case PHY_INTERFACE_MODE_RGMII_ID: case PHY_INTERFACE_MODE_RGMII_TXID: case PHY_INTERFACE_MODE_RGMII_RXID: - val = FIELD_PREP(MACPHYC_PHY_INFT_MASK, PHY_INTF_SEL_RGMII); + phy_intf_sel = PHY_INTF_SEL_RGMII; dev_dbg(mac->dev, "MAC PHY Control Register: PHY_INTERFACE_MODE_RGMII\n"); break; @@ -102,7 +103,8 @@ static int jz4775_mac_set_mode(struct plat_stmmacenet_data *plat_dat) return -EINVAL; } - val |= FIELD_PREP(MACPHYC_TXCLK_SEL_MASK, MACPHYC_TXCLK_SEL_INPUT); + val = FIELD_PREP(MACPHYC_PHY_INFT_MASK, phy_intf_sel) | + FIELD_PREP(MACPHYC_TXCLK_SEL_MASK, MACPHYC_TXCLK_SEL_INPUT); /* Update MAC PHY control register */ return regmap_update_bits(mac->regmap, 0, mac->soc_info->mask, val); @@ -131,10 +133,11 @@ static int x1600_mac_set_mode(struct plat_stmmacenet_data *plat_dat) { struct ingenic_mac *mac = plat_dat->bsp_priv; unsigned int val; + u8 phy_intf_sel; switch (plat_dat->phy_interface) { case PHY_INTERFACE_MODE_RMII: - val = FIELD_PREP(MACPHYC_PHY_INFT_MASK, PHY_INTF_SEL_RMII); + phy_intf_sel = PHY_INTF_SEL_RMII; dev_dbg(mac->dev, "MAC PHY Control Register: PHY_INTERFACE_MODE_RMII\n"); break; @@ -144,6 +147,8 @@ static int x1600_mac_set_mode(struct plat_stmmacenet_data *plat_dat) return -EINVAL; } + val = FIELD_PREP(MACPHYC_PHY_INFT_MASK, phy_intf_sel); + /* Update MAC PHY control register */ return regmap_update_bits(mac->regmap, 0, mac->soc_info->mask, val); } @@ -152,11 +157,12 @@ static int x1830_mac_set_mode(struct plat_stmmacenet_data *plat_dat) { struct ingenic_mac *mac = plat_dat->bsp_priv; unsigned int val; + u8 phy_intf_sel; switch (plat_dat->phy_interface) { case PHY_INTERFACE_MODE_RMII: - val = FIELD_PREP(MACPHYC_MODE_SEL_MASK, MACPHYC_MODE_SEL_RMII) | - FIELD_PREP(MACPHYC_PHY_INFT_MASK, PHY_INTF_SEL_RMII); + val = FIELD_PREP(MACPHYC_MODE_SEL_MASK, MACPHYC_MODE_SEL_RMII); + phy_intf_sel = PHY_INTF_SEL_RMII; dev_dbg(mac->dev, "MAC PHY Control Register: PHY_INTERFACE_MODE_RMII\n"); break; @@ -166,6 +172,8 @@ static int x1830_mac_set_mode(struct plat_stmmacenet_data *plat_dat) return -EINVAL; } + val |= FIELD_PREP(MACPHYC_PHY_INFT_MASK, phy_intf_sel); + /* Update MAC PHY control register */ return regmap_update_bits(mac->regmap, 0, mac->soc_info->mask, val); } @@ -174,12 +182,13 @@ static int x2000_mac_set_mode(struct plat_stmmacenet_data *plat_dat) { struct ingenic_mac *mac = plat_dat->bsp_priv; unsigned int val; + u8 phy_intf_sel; switch (plat_dat->phy_interface) { case PHY_INTERFACE_MODE_RMII: val = FIELD_PREP(MACPHYC_TX_SEL_MASK, MACPHYC_TX_SEL_ORIGIN) | - FIELD_PREP(MACPHYC_RX_SEL_MASK, MACPHYC_RX_SEL_ORIGIN) | - FIELD_PREP(MACPHYC_PHY_INFT_MASK, PHY_INTF_SEL_RMII); + FIELD_PREP(MACPHYC_RX_SEL_MASK, MACPHYC_RX_SEL_ORIGIN); + phy_intf_sel = PHY_INTF_SEL_RMII; dev_dbg(mac->dev, "MAC PHY Control Register: PHY_INTERFACE_MODE_RMII\n"); break; @@ -187,7 +196,8 @@ static int x2000_mac_set_mode(struct plat_stmmacenet_data *plat_dat) case PHY_INTERFACE_MODE_RGMII_ID: case PHY_INTERFACE_MODE_RGMII_TXID: case PHY_INTERFACE_MODE_RGMII_RXID: - val = FIELD_PREP(MACPHYC_PHY_INFT_MASK, PHY_INTF_SEL_RGMII); + val = 0; + phy_intf_sel = PHY_INTF_SEL_RGMII; if (mac->tx_delay == 0) val |= FIELD_PREP(MACPHYC_TX_SEL_MASK, MACPHYC_TX_SEL_ORIGIN); @@ -210,6 +220,8 @@ static int x2000_mac_set_mode(struct plat_stmmacenet_data *plat_dat) return -EINVAL; } + val |= FIELD_PREP(MACPHYC_PHY_INFT_MASK, phy_intf_sel); + /* Update MAC PHY control register */ return regmap_update_bits(mac->regmap, 0, mac->soc_info->mask, val); } From 0e2fa91c55c05f441417489957a1f909ff6c9aa1 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Fri, 7 Nov 2025 08:29:00 +0000 Subject: [PATCH 0688/1381] net: stmmac: ingenic: use stmmac_get_phy_intf_sel() Use stmmac_get_phy_intf_sel() to decode the PHY interface mode to the phy_intf_sel value, validate the result against the SoC specific supported phy_intf_sel values, and pass into the SoC specific set_mode() methods, replacing the local phy_intf_sel variable. This provides the value for the MACPHYC_PHY_INFT_MASK field. Reviewed-by: Maxime Chevallier Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vHHq8-0000000DjrJ-2NRK@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../ethernet/stmicro/stmmac/dwmac-ingenic.c | 55 ++++++++++++------- 1 file changed, 34 insertions(+), 21 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c index 6680f7d3a469..79735a476e86 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c @@ -64,28 +64,27 @@ struct ingenic_soc_info { enum ingenic_mac_version version; u32 mask; - int (*set_mode)(struct plat_stmmacenet_data *plat_dat); + int (*set_mode)(struct plat_stmmacenet_data *plat_dat, u8 phy_intf_sel); + + u8 valid_phy_intf_sel; }; -static int jz4775_mac_set_mode(struct plat_stmmacenet_data *plat_dat) +static int jz4775_mac_set_mode(struct plat_stmmacenet_data *plat_dat, + u8 phy_intf_sel) { struct ingenic_mac *mac = plat_dat->bsp_priv; unsigned int val; - u8 phy_intf_sel; switch (plat_dat->phy_interface) { case PHY_INTERFACE_MODE_MII: - phy_intf_sel = PHY_INTF_SEL_GMII_MII; dev_dbg(mac->dev, "MAC PHY Control Register: PHY_INTERFACE_MODE_MII\n"); break; case PHY_INTERFACE_MODE_GMII: - phy_intf_sel = PHY_INTF_SEL_GMII_MII; dev_dbg(mac->dev, "MAC PHY Control Register: PHY_INTERFACE_MODE_GMII\n"); break; case PHY_INTERFACE_MODE_RMII: - phy_intf_sel = PHY_INTF_SEL_RMII; dev_dbg(mac->dev, "MAC PHY Control Register: PHY_INTERFACE_MODE_RMII\n"); break; @@ -93,7 +92,6 @@ static int jz4775_mac_set_mode(struct plat_stmmacenet_data *plat_dat) case PHY_INTERFACE_MODE_RGMII_ID: case PHY_INTERFACE_MODE_RGMII_TXID: case PHY_INTERFACE_MODE_RGMII_RXID: - phy_intf_sel = PHY_INTF_SEL_RGMII; dev_dbg(mac->dev, "MAC PHY Control Register: PHY_INTERFACE_MODE_RGMII\n"); break; @@ -110,7 +108,8 @@ static int jz4775_mac_set_mode(struct plat_stmmacenet_data *plat_dat) return regmap_update_bits(mac->regmap, 0, mac->soc_info->mask, val); } -static int x1000_mac_set_mode(struct plat_stmmacenet_data *plat_dat) +static int x1000_mac_set_mode(struct plat_stmmacenet_data *plat_dat, + u8 phy_intf_sel) { struct ingenic_mac *mac = plat_dat->bsp_priv; @@ -129,15 +128,14 @@ static int x1000_mac_set_mode(struct plat_stmmacenet_data *plat_dat) return regmap_update_bits(mac->regmap, 0, mac->soc_info->mask, 0); } -static int x1600_mac_set_mode(struct plat_stmmacenet_data *plat_dat) +static int x1600_mac_set_mode(struct plat_stmmacenet_data *plat_dat, + u8 phy_intf_sel) { struct ingenic_mac *mac = plat_dat->bsp_priv; unsigned int val; - u8 phy_intf_sel; switch (plat_dat->phy_interface) { case PHY_INTERFACE_MODE_RMII: - phy_intf_sel = PHY_INTF_SEL_RMII; dev_dbg(mac->dev, "MAC PHY Control Register: PHY_INTERFACE_MODE_RMII\n"); break; @@ -153,16 +151,15 @@ static int x1600_mac_set_mode(struct plat_stmmacenet_data *plat_dat) return regmap_update_bits(mac->regmap, 0, mac->soc_info->mask, val); } -static int x1830_mac_set_mode(struct plat_stmmacenet_data *plat_dat) +static int x1830_mac_set_mode(struct plat_stmmacenet_data *plat_dat, + u8 phy_intf_sel) { struct ingenic_mac *mac = plat_dat->bsp_priv; unsigned int val; - u8 phy_intf_sel; switch (plat_dat->phy_interface) { case PHY_INTERFACE_MODE_RMII: val = FIELD_PREP(MACPHYC_MODE_SEL_MASK, MACPHYC_MODE_SEL_RMII); - phy_intf_sel = PHY_INTF_SEL_RMII; dev_dbg(mac->dev, "MAC PHY Control Register: PHY_INTERFACE_MODE_RMII\n"); break; @@ -178,17 +175,16 @@ static int x1830_mac_set_mode(struct plat_stmmacenet_data *plat_dat) return regmap_update_bits(mac->regmap, 0, mac->soc_info->mask, val); } -static int x2000_mac_set_mode(struct plat_stmmacenet_data *plat_dat) +static int x2000_mac_set_mode(struct plat_stmmacenet_data *plat_dat, + u8 phy_intf_sel) { struct ingenic_mac *mac = plat_dat->bsp_priv; unsigned int val; - u8 phy_intf_sel; switch (plat_dat->phy_interface) { case PHY_INTERFACE_MODE_RMII: val = FIELD_PREP(MACPHYC_TX_SEL_MASK, MACPHYC_TX_SEL_ORIGIN) | FIELD_PREP(MACPHYC_RX_SEL_MASK, MACPHYC_RX_SEL_ORIGIN); - phy_intf_sel = PHY_INTF_SEL_RMII; dev_dbg(mac->dev, "MAC PHY Control Register: PHY_INTERFACE_MODE_RMII\n"); break; @@ -197,8 +193,6 @@ static int x2000_mac_set_mode(struct plat_stmmacenet_data *plat_dat) case PHY_INTERFACE_MODE_RGMII_TXID: case PHY_INTERFACE_MODE_RGMII_RXID: val = 0; - phy_intf_sel = PHY_INTF_SEL_RGMII; - if (mac->tx_delay == 0) val |= FIELD_PREP(MACPHYC_TX_SEL_MASK, MACPHYC_TX_SEL_ORIGIN); else @@ -229,10 +223,21 @@ static int x2000_mac_set_mode(struct plat_stmmacenet_data *plat_dat) static int ingenic_mac_init(struct platform_device *pdev, void *bsp_priv) { struct ingenic_mac *mac = bsp_priv; - int ret; + phy_interface_t interface; + int phy_intf_sel, ret; if (mac->soc_info->set_mode) { - ret = mac->soc_info->set_mode(mac->plat_dat); + interface = mac->plat_dat->phy_interface; + + phy_intf_sel = stmmac_get_phy_intf_sel(interface); + if (phy_intf_sel < 0 || phy_intf_sel >= BITS_PER_BYTE || + ~mac->soc_info->valid_phy_intf_sel & BIT(phy_intf_sel)) { + dev_err(mac->dev, "unsupported interface %s\n", + phy_modes(interface)); + return phy_intf_sel < 0 ? phy_intf_sel : -EINVAL; + } + + ret = mac->soc_info->set_mode(mac->plat_dat, phy_intf_sel); if (ret) return ret; } @@ -309,6 +314,9 @@ static struct ingenic_soc_info jz4775_soc_info = { .mask = MACPHYC_TXCLK_SEL_MASK | MACPHYC_SOFT_RST_MASK | MACPHYC_PHY_INFT_MASK, .set_mode = jz4775_mac_set_mode, + .valid_phy_intf_sel = BIT(PHY_INTF_SEL_GMII_MII) | + BIT(PHY_INTF_SEL_RGMII) | + BIT(PHY_INTF_SEL_RMII), }; static struct ingenic_soc_info x1000_soc_info = { @@ -316,6 +324,7 @@ static struct ingenic_soc_info x1000_soc_info = { .mask = MACPHYC_SOFT_RST_MASK, .set_mode = x1000_mac_set_mode, + .valid_phy_intf_sel = BIT(PHY_INTF_SEL_RMII), }; static struct ingenic_soc_info x1600_soc_info = { @@ -323,6 +332,7 @@ static struct ingenic_soc_info x1600_soc_info = { .mask = MACPHYC_SOFT_RST_MASK | MACPHYC_PHY_INFT_MASK, .set_mode = x1600_mac_set_mode, + .valid_phy_intf_sel = BIT(PHY_INTF_SEL_RMII), }; static struct ingenic_soc_info x1830_soc_info = { @@ -330,6 +340,7 @@ static struct ingenic_soc_info x1830_soc_info = { .mask = MACPHYC_MODE_SEL_MASK | MACPHYC_SOFT_RST_MASK | MACPHYC_PHY_INFT_MASK, .set_mode = x1830_mac_set_mode, + .valid_phy_intf_sel = BIT(PHY_INTF_SEL_RMII), }; static struct ingenic_soc_info x2000_soc_info = { @@ -338,6 +349,8 @@ static struct ingenic_soc_info x2000_soc_info = { MACPHYC_RX_DELAY_MASK | MACPHYC_SOFT_RST_MASK | MACPHYC_PHY_INFT_MASK, .set_mode = x2000_mac_set_mode, + .valid_phy_intf_sel = BIT(PHY_INTF_SEL_RGMII) | + BIT(PHY_INTF_SEL_RMII), }; static const struct of_device_id ingenic_mac_of_matches[] = { From 35147b5c9e414c1ced1903c0447609643d03e2f3 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Fri, 7 Nov 2025 08:29:05 +0000 Subject: [PATCH 0689/1381] net: stmmac: ingenic: move "MAC PHY control register" debug Move the printing of the MAC PHY control register interface mode setting into ingenic_set_phy_intf_sel(), and use phy_modes() to print the string rather than using the enum name. Reviewed-by: Maxime Chevallier Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vHHqD-0000000DjrP-3aaU@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../ethernet/stmicro/stmmac/dwmac-ingenic.c | 18 +++--------------- 1 file changed, 3 insertions(+), 15 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c index 79735a476e86..539513890db1 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c @@ -77,22 +77,12 @@ static int jz4775_mac_set_mode(struct plat_stmmacenet_data *plat_dat, switch (plat_dat->phy_interface) { case PHY_INTERFACE_MODE_MII: - dev_dbg(mac->dev, "MAC PHY Control Register: PHY_INTERFACE_MODE_MII\n"); - break; - case PHY_INTERFACE_MODE_GMII: - dev_dbg(mac->dev, "MAC PHY Control Register: PHY_INTERFACE_MODE_GMII\n"); - break; - case PHY_INTERFACE_MODE_RMII: - dev_dbg(mac->dev, "MAC PHY Control Register: PHY_INTERFACE_MODE_RMII\n"); - break; - case PHY_INTERFACE_MODE_RGMII: case PHY_INTERFACE_MODE_RGMII_ID: case PHY_INTERFACE_MODE_RGMII_TXID: case PHY_INTERFACE_MODE_RGMII_RXID: - dev_dbg(mac->dev, "MAC PHY Control Register: PHY_INTERFACE_MODE_RGMII\n"); break; default: @@ -115,7 +105,6 @@ static int x1000_mac_set_mode(struct plat_stmmacenet_data *plat_dat, switch (plat_dat->phy_interface) { case PHY_INTERFACE_MODE_RMII: - dev_dbg(mac->dev, "MAC PHY Control Register: PHY_INTERFACE_MODE_RMII\n"); break; default: @@ -136,7 +125,6 @@ static int x1600_mac_set_mode(struct plat_stmmacenet_data *plat_dat, switch (plat_dat->phy_interface) { case PHY_INTERFACE_MODE_RMII: - dev_dbg(mac->dev, "MAC PHY Control Register: PHY_INTERFACE_MODE_RMII\n"); break; default: @@ -160,7 +148,6 @@ static int x1830_mac_set_mode(struct plat_stmmacenet_data *plat_dat, switch (plat_dat->phy_interface) { case PHY_INTERFACE_MODE_RMII: val = FIELD_PREP(MACPHYC_MODE_SEL_MASK, MACPHYC_MODE_SEL_RMII); - dev_dbg(mac->dev, "MAC PHY Control Register: PHY_INTERFACE_MODE_RMII\n"); break; default: @@ -185,7 +172,6 @@ static int x2000_mac_set_mode(struct plat_stmmacenet_data *plat_dat, case PHY_INTERFACE_MODE_RMII: val = FIELD_PREP(MACPHYC_TX_SEL_MASK, MACPHYC_TX_SEL_ORIGIN) | FIELD_PREP(MACPHYC_RX_SEL_MASK, MACPHYC_RX_SEL_ORIGIN); - dev_dbg(mac->dev, "MAC PHY Control Register: PHY_INTERFACE_MODE_RMII\n"); break; case PHY_INTERFACE_MODE_RGMII: @@ -205,7 +191,6 @@ static int x2000_mac_set_mode(struct plat_stmmacenet_data *plat_dat, val |= FIELD_PREP(MACPHYC_RX_SEL_MASK, MACPHYC_RX_SEL_DELAY) | FIELD_PREP(MACPHYC_RX_DELAY_MASK, (mac->rx_delay + 9750) / 19500 - 1); - dev_dbg(mac->dev, "MAC PHY Control Register: PHY_INTERFACE_MODE_RGMII\n"); break; default: @@ -237,6 +222,9 @@ static int ingenic_mac_init(struct platform_device *pdev, void *bsp_priv) return phy_intf_sel < 0 ? phy_intf_sel : -EINVAL; } + dev_dbg(mac->dev, "MAC PHY control register: interface %s\n", + phy_modes(interface)); + ret = mac->soc_info->set_mode(mac->plat_dat, phy_intf_sel); if (ret) return ret; From 608975d4d791644d15b286b2b7bfcd25359c979f Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Fri, 7 Nov 2025 08:29:10 +0000 Subject: [PATCH 0690/1381] net: stmmac: ingenic: simplify mac_set_mode() methods x1000, x1600 and x1830 only accept RMII mode. PHY_INTF_SEL_RMII is only selected with PHY_INTERFACE_MODE_RMII, and PHY_INTF_SEL_RMII has been validated by the SoC's .valid_phy_intf_sel bitmask. Thus, checking the interface mode in these functions becomes unnecessary. Remove these. jz4775 is similar, except for a greater set of PHY_INTF_SEL_x valies. Also remove the switch statement here. Reviewed-by: Maxime Chevallier Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vHHqI-0000000DjrV-3ygL@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../ethernet/stmicro/stmmac/dwmac-ingenic.c | 50 +------------------ 1 file changed, 2 insertions(+), 48 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c index 539513890db1..7b2576fbb1e1 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c @@ -75,22 +75,6 @@ static int jz4775_mac_set_mode(struct plat_stmmacenet_data *plat_dat, struct ingenic_mac *mac = plat_dat->bsp_priv; unsigned int val; - switch (plat_dat->phy_interface) { - case PHY_INTERFACE_MODE_MII: - case PHY_INTERFACE_MODE_GMII: - case PHY_INTERFACE_MODE_RMII: - case PHY_INTERFACE_MODE_RGMII: - case PHY_INTERFACE_MODE_RGMII_ID: - case PHY_INTERFACE_MODE_RGMII_TXID: - case PHY_INTERFACE_MODE_RGMII_RXID: - break; - - default: - dev_err(mac->dev, "Unsupported interface %s\n", - phy_modes(plat_dat->phy_interface)); - return -EINVAL; - } - val = FIELD_PREP(MACPHYC_PHY_INFT_MASK, phy_intf_sel) | FIELD_PREP(MACPHYC_TXCLK_SEL_MASK, MACPHYC_TXCLK_SEL_INPUT); @@ -103,16 +87,6 @@ static int x1000_mac_set_mode(struct plat_stmmacenet_data *plat_dat, { struct ingenic_mac *mac = plat_dat->bsp_priv; - switch (plat_dat->phy_interface) { - case PHY_INTERFACE_MODE_RMII: - break; - - default: - dev_err(mac->dev, "Unsupported interface %s\n", - phy_modes(plat_dat->phy_interface)); - return -EINVAL; - } - /* Update MAC PHY control register */ return regmap_update_bits(mac->regmap, 0, mac->soc_info->mask, 0); } @@ -123,16 +97,6 @@ static int x1600_mac_set_mode(struct plat_stmmacenet_data *plat_dat, struct ingenic_mac *mac = plat_dat->bsp_priv; unsigned int val; - switch (plat_dat->phy_interface) { - case PHY_INTERFACE_MODE_RMII: - break; - - default: - dev_err(mac->dev, "Unsupported interface %s\n", - phy_modes(plat_dat->phy_interface)); - return -EINVAL; - } - val = FIELD_PREP(MACPHYC_PHY_INFT_MASK, phy_intf_sel); /* Update MAC PHY control register */ @@ -145,18 +109,8 @@ static int x1830_mac_set_mode(struct plat_stmmacenet_data *plat_dat, struct ingenic_mac *mac = plat_dat->bsp_priv; unsigned int val; - switch (plat_dat->phy_interface) { - case PHY_INTERFACE_MODE_RMII: - val = FIELD_PREP(MACPHYC_MODE_SEL_MASK, MACPHYC_MODE_SEL_RMII); - break; - - default: - dev_err(mac->dev, "Unsupported interface %s\n", - phy_modes(plat_dat->phy_interface)); - return -EINVAL; - } - - val |= FIELD_PREP(MACPHYC_PHY_INFT_MASK, phy_intf_sel); + val = FIELD_PREP(MACPHYC_MODE_SEL_MASK, MACPHYC_MODE_SEL_RMII) | + FIELD_PREP(MACPHYC_PHY_INFT_MASK, phy_intf_sel); /* Update MAC PHY control register */ return regmap_update_bits(mac->regmap, 0, mac->soc_info->mask, val); From 2284cca0bced3cd8c1cb643d1ae6923a75e47265 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Fri, 7 Nov 2025 08:29:16 +0000 Subject: [PATCH 0691/1381] net: stmmac: ingenic: simplify x2000 mac_set_mode() As per the previous commit, we have validated that the phy_intf_sel value is one that is permissible for this SoC, so there is no need to handle invalid PHY interface modes. We can also apply the other configuration based upon the phy_intf_sel value rather than the PHY interface mode. Reviewed-by: Maxime Chevallier Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vHHqO-0000000Djrb-0DPN@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../ethernet/stmicro/stmmac/dwmac-ingenic.c | 26 +++++-------------- 1 file changed, 6 insertions(+), 20 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c index 7b2576fbb1e1..eb5744e0b9ea 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c @@ -122,39 +122,25 @@ static int x2000_mac_set_mode(struct plat_stmmacenet_data *plat_dat, struct ingenic_mac *mac = plat_dat->bsp_priv; unsigned int val; - switch (plat_dat->phy_interface) { - case PHY_INTERFACE_MODE_RMII: - val = FIELD_PREP(MACPHYC_TX_SEL_MASK, MACPHYC_TX_SEL_ORIGIN) | - FIELD_PREP(MACPHYC_RX_SEL_MASK, MACPHYC_RX_SEL_ORIGIN); - break; + val = FIELD_PREP(MACPHYC_PHY_INFT_MASK, phy_intf_sel); - case PHY_INTERFACE_MODE_RGMII: - case PHY_INTERFACE_MODE_RGMII_ID: - case PHY_INTERFACE_MODE_RGMII_TXID: - case PHY_INTERFACE_MODE_RGMII_RXID: - val = 0; + if (phy_intf_sel == PHY_INTF_SEL_RMII) { + val |= FIELD_PREP(MACPHYC_TX_SEL_MASK, MACPHYC_TX_SEL_ORIGIN) | + FIELD_PREP(MACPHYC_RX_SEL_MASK, MACPHYC_RX_SEL_ORIGIN); + } else if (phy_intf_sel == PHY_INTF_SEL_RGMII) { if (mac->tx_delay == 0) val |= FIELD_PREP(MACPHYC_TX_SEL_MASK, MACPHYC_TX_SEL_ORIGIN); else val |= FIELD_PREP(MACPHYC_TX_SEL_MASK, MACPHYC_TX_SEL_DELAY) | - FIELD_PREP(MACPHYC_TX_DELAY_MASK, (mac->tx_delay + 9750) / 19500 - 1); + FIELD_PREP(MACPHYC_TX_DELAY_MASK, (mac->tx_delay + 9750) / 19500 - 1); if (mac->rx_delay == 0) val |= FIELD_PREP(MACPHYC_RX_SEL_MASK, MACPHYC_RX_SEL_ORIGIN); else val |= FIELD_PREP(MACPHYC_RX_SEL_MASK, MACPHYC_RX_SEL_DELAY) | FIELD_PREP(MACPHYC_RX_DELAY_MASK, (mac->rx_delay + 9750) / 19500 - 1); - - break; - - default: - dev_err(mac->dev, "Unsupported interface %s\n", - phy_modes(plat_dat->phy_interface)); - return -EINVAL; } - val |= FIELD_PREP(MACPHYC_PHY_INFT_MASK, phy_intf_sel); - /* Update MAC PHY control register */ return regmap_update_bits(mac->regmap, 0, mac->soc_info->mask, val); } From 9352f74fd13de34468be95291bdb0782aac8ca3d Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Fri, 7 Nov 2025 08:29:21 +0000 Subject: [PATCH 0692/1381] net: stmmac: ingenic: pass ingenic_mac struct rather than plat_dat It no longer makes sense to pass a pointer to struct plat_stmmacenet_data when calling the set_mode() methods to only use it to get a pointer to the ingenic_mac structure that we already had in the caller. Simplify this by passing the struct ingenic_mac pointer. Reviewed-by: Maxime Chevallier Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vHHqT-0000000Djrh-0ka3@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../ethernet/stmicro/stmmac/dwmac-ingenic.c | 25 ++++++------------- 1 file changed, 7 insertions(+), 18 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c index eb5744e0b9ea..41a2071262bc 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c @@ -64,15 +64,13 @@ struct ingenic_soc_info { enum ingenic_mac_version version; u32 mask; - int (*set_mode)(struct plat_stmmacenet_data *plat_dat, u8 phy_intf_sel); + int (*set_mode)(struct ingenic_mac *mac, u8 phy_intf_sel); u8 valid_phy_intf_sel; }; -static int jz4775_mac_set_mode(struct plat_stmmacenet_data *plat_dat, - u8 phy_intf_sel) +static int jz4775_mac_set_mode(struct ingenic_mac *mac, u8 phy_intf_sel) { - struct ingenic_mac *mac = plat_dat->bsp_priv; unsigned int val; val = FIELD_PREP(MACPHYC_PHY_INFT_MASK, phy_intf_sel) | @@ -82,19 +80,14 @@ static int jz4775_mac_set_mode(struct plat_stmmacenet_data *plat_dat, return regmap_update_bits(mac->regmap, 0, mac->soc_info->mask, val); } -static int x1000_mac_set_mode(struct plat_stmmacenet_data *plat_dat, - u8 phy_intf_sel) +static int x1000_mac_set_mode(struct ingenic_mac *mac, u8 phy_intf_sel) { - struct ingenic_mac *mac = plat_dat->bsp_priv; - /* Update MAC PHY control register */ return regmap_update_bits(mac->regmap, 0, mac->soc_info->mask, 0); } -static int x1600_mac_set_mode(struct plat_stmmacenet_data *plat_dat, - u8 phy_intf_sel) +static int x1600_mac_set_mode(struct ingenic_mac *mac, u8 phy_intf_sel) { - struct ingenic_mac *mac = plat_dat->bsp_priv; unsigned int val; val = FIELD_PREP(MACPHYC_PHY_INFT_MASK, phy_intf_sel); @@ -103,10 +96,8 @@ static int x1600_mac_set_mode(struct plat_stmmacenet_data *plat_dat, return regmap_update_bits(mac->regmap, 0, mac->soc_info->mask, val); } -static int x1830_mac_set_mode(struct plat_stmmacenet_data *plat_dat, - u8 phy_intf_sel) +static int x1830_mac_set_mode(struct ingenic_mac *mac, u8 phy_intf_sel) { - struct ingenic_mac *mac = plat_dat->bsp_priv; unsigned int val; val = FIELD_PREP(MACPHYC_MODE_SEL_MASK, MACPHYC_MODE_SEL_RMII) | @@ -116,10 +107,8 @@ static int x1830_mac_set_mode(struct plat_stmmacenet_data *plat_dat, return regmap_update_bits(mac->regmap, 0, mac->soc_info->mask, val); } -static int x2000_mac_set_mode(struct plat_stmmacenet_data *plat_dat, - u8 phy_intf_sel) +static int x2000_mac_set_mode(struct ingenic_mac *mac, u8 phy_intf_sel) { - struct ingenic_mac *mac = plat_dat->bsp_priv; unsigned int val; val = FIELD_PREP(MACPHYC_PHY_INFT_MASK, phy_intf_sel); @@ -165,7 +154,7 @@ static int ingenic_mac_init(struct platform_device *pdev, void *bsp_priv) dev_dbg(mac->dev, "MAC PHY control register: interface %s\n", phy_modes(interface)); - ret = mac->soc_info->set_mode(mac->plat_dat, phy_intf_sel); + ret = mac->soc_info->set_mode(mac, phy_intf_sel); if (ret) return ret; } From 34bf68a691227faac490a313b7925ed10cf8de15 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Fri, 7 Nov 2025 08:29:26 +0000 Subject: [PATCH 0693/1381] net: stmmac: ingenic: use ->set_phy_intf_sel() Rather than placing the phy_intf_sel() setup in the ->init() method, move it to the new ->set_phy_intf_sel() method. Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vHHqY-0000000Djrn-1D6H@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../ethernet/stmicro/stmmac/dwmac-ingenic.c | 31 ++++++------------- 1 file changed, 10 insertions(+), 21 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c index 41a2071262bc..8e4a30c11db0 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-ingenic.c @@ -134,32 +134,21 @@ static int x2000_mac_set_mode(struct ingenic_mac *mac, u8 phy_intf_sel) return regmap_update_bits(mac->regmap, 0, mac->soc_info->mask, val); } -static int ingenic_mac_init(struct platform_device *pdev, void *bsp_priv) +static int ingenic_set_phy_intf_sel(void *bsp_priv, u8 phy_intf_sel) { struct ingenic_mac *mac = bsp_priv; - phy_interface_t interface; - int phy_intf_sel, ret; - if (mac->soc_info->set_mode) { - interface = mac->plat_dat->phy_interface; + if (!mac->soc_info->set_mode) + return 0; - phy_intf_sel = stmmac_get_phy_intf_sel(interface); - if (phy_intf_sel < 0 || phy_intf_sel >= BITS_PER_BYTE || - ~mac->soc_info->valid_phy_intf_sel & BIT(phy_intf_sel)) { - dev_err(mac->dev, "unsupported interface %s\n", - phy_modes(interface)); - return phy_intf_sel < 0 ? phy_intf_sel : -EINVAL; - } + if (phy_intf_sel >= BITS_PER_BYTE || + ~mac->soc_info->valid_phy_intf_sel & BIT(phy_intf_sel)) + return -EINVAL; - dev_dbg(mac->dev, "MAC PHY control register: interface %s\n", - phy_modes(interface)); + dev_dbg(mac->dev, "MAC PHY control register: interface %s\n", + phy_modes(mac->plat_dat->phy_interface)); - ret = mac->soc_info->set_mode(mac, phy_intf_sel); - if (ret) - return ret; - } - - return 0; + return mac->soc_info->set_mode(mac, phy_intf_sel); } static int ingenic_mac_probe(struct platform_device *pdev) @@ -221,7 +210,7 @@ static int ingenic_mac_probe(struct platform_device *pdev) mac->plat_dat = plat_dat; plat_dat->bsp_priv = mac; - plat_dat->init = ingenic_mac_init; + plat_dat->set_phy_intf_sel = ingenic_set_phy_intf_sel; return devm_stmmac_pltfr_probe(pdev, plat_dat, &stmmac_res); } From 40fef85ceb9cce2021f7ba5ca6e3212baa193cb3 Mon Sep 17 00:00:00 2001 From: Ankit Garg Date: Thu, 6 Nov 2025 11:27:43 -0800 Subject: [PATCH 0694/1381] gve: Decouple header split from RX buffer length Previously, enabling header split via `gve_set_hsplit_config` also implicitly changed the RX buffer length to 4K (if supported by the device). This coupled two settings that should be orthogonal; this patch removes that side effect. After this change, `gve_set_hsplit_config` only toggles the header split configuration. The RX buffer length is no longer affected and must be configured independently. Signed-off-by: Ankit Garg Reviewed-by: Harshitha Ramamurthy Reviewed-by: Jordan Rhee Reviewed-by: Willem de Bruijn Signed-off-by: Joshua Washington Link: https://patch.msgid.link/20251106192746.243525-2-joshwash@google.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/google/gve/gve.h | 3 --- drivers/net/ethernet/google/gve/gve_ethtool.c | 2 -- drivers/net/ethernet/google/gve/gve_main.c | 10 ---------- 3 files changed, 15 deletions(-) diff --git a/drivers/net/ethernet/google/gve/gve.h b/drivers/net/ethernet/google/gve/gve.h index cf95ec25b11a..c237d00c5ab3 100644 --- a/drivers/net/ethernet/google/gve/gve.h +++ b/drivers/net/ethernet/google/gve/gve.h @@ -59,8 +59,6 @@ #define GVE_DEFAULT_RX_BUFFER_SIZE 2048 -#define GVE_MAX_RX_BUFFER_SIZE 4096 - #define GVE_XDP_RX_BUFFER_SIZE_DQO 4096 #define GVE_DEFAULT_RX_BUFFER_OFFSET 2048 @@ -1249,7 +1247,6 @@ void gve_rx_free_rings_gqi(struct gve_priv *priv, struct gve_rx_alloc_rings_cfg *cfg); void gve_rx_start_ring_gqi(struct gve_priv *priv, int idx); void gve_rx_stop_ring_gqi(struct gve_priv *priv, int idx); -u16 gve_get_pkt_buf_size(const struct gve_priv *priv, bool enable_hplit); bool gve_header_split_supported(const struct gve_priv *priv); int gve_set_hsplit_config(struct gve_priv *priv, u8 tcp_data_split, struct gve_rx_alloc_rings_cfg *rx_alloc_cfg); diff --git a/drivers/net/ethernet/google/gve/gve_ethtool.c b/drivers/net/ethernet/google/gve/gve_ethtool.c index b030a84b678c..db6fc855a511 100644 --- a/drivers/net/ethernet/google/gve/gve_ethtool.c +++ b/drivers/net/ethernet/google/gve/gve_ethtool.c @@ -606,8 +606,6 @@ static int gve_set_ringparam(struct net_device *netdev, } else { /* Set ring params for the next up */ priv->header_split_enabled = rx_alloc_cfg.enable_header_split; - priv->rx_cfg.packet_buffer_size = - rx_alloc_cfg.packet_buffer_size; priv->tx_desc_cnt = tx_alloc_cfg.ring_size; priv->rx_desc_cnt = rx_alloc_cfg.ring_size; } diff --git a/drivers/net/ethernet/google/gve/gve_main.c b/drivers/net/ethernet/google/gve/gve_main.c index 29845e8f3c0d..8d825218965a 100644 --- a/drivers/net/ethernet/google/gve/gve_main.c +++ b/drivers/net/ethernet/google/gve/gve_main.c @@ -2041,14 +2041,6 @@ static void gve_tx_timeout(struct net_device *dev, unsigned int txqueue) priv->tx_timeo_cnt++; } -u16 gve_get_pkt_buf_size(const struct gve_priv *priv, bool enable_hsplit) -{ - if (enable_hsplit && priv->max_rx_buffer_size >= GVE_MAX_RX_BUFFER_SIZE) - return GVE_MAX_RX_BUFFER_SIZE; - else - return GVE_DEFAULT_RX_BUFFER_SIZE; -} - /* Header split is only supported on DQ RDA queue format. If XDP is enabled, * header split is not allowed. */ @@ -2080,8 +2072,6 @@ int gve_set_hsplit_config(struct gve_priv *priv, u8 tcp_data_split, return 0; rx_alloc_cfg->enable_header_split = enable_hdr_split; - rx_alloc_cfg->packet_buffer_size = - gve_get_pkt_buf_size(priv, enable_hdr_split); return 0; } From 091a3b6ff2b98354270cb9278faad6d17d5aa27d Mon Sep 17 00:00:00 2001 From: Ankit Garg Date: Thu, 6 Nov 2025 11:27:44 -0800 Subject: [PATCH 0695/1381] gve: Use extack to log xdp config verification errors Plumb extack as it allows us to send more detailed error messages back and append 'gve' suffix to method name per convention. NL_SET_ERR_MSG_FMT_MOD doesn't support format string longer than 80 chars so keeping netdev warning with actual queue count details. Signed-off-by: Ankit Garg Reviewed-by: Harshitha Ramamurthy Reviewed-by: Willem de Bruijn Signed-off-by: Joshua Washington Link: https://patch.msgid.link/20251106192746.243525-3-joshwash@google.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/google/gve/gve_main.c | 24 ++++++++++++++-------- 1 file changed, 15 insertions(+), 9 deletions(-) diff --git a/drivers/net/ethernet/google/gve/gve_main.c b/drivers/net/ethernet/google/gve/gve_main.c index 8d825218965a..616182f4fb81 100644 --- a/drivers/net/ethernet/google/gve/gve_main.c +++ b/drivers/net/ethernet/google/gve/gve_main.c @@ -1707,18 +1707,21 @@ static int gve_xsk_wakeup(struct net_device *dev, u32 queue_id, u32 flags) return 0; } -static int verify_xdp_configuration(struct net_device *dev) +static int gve_verify_xdp_configuration(struct net_device *dev, + struct netlink_ext_ack *extack) { struct gve_priv *priv = netdev_priv(dev); u16 max_xdp_mtu; if (dev->features & NETIF_F_LRO) { - netdev_warn(dev, "XDP is not supported when LRO is on.\n"); + NL_SET_ERR_MSG_MOD(extack, + "XDP is not supported when LRO is on."); return -EOPNOTSUPP; } if (priv->header_split_enabled) { - netdev_warn(dev, "XDP is not supported when header-data split is enabled.\n"); + NL_SET_ERR_MSG_MOD(extack, + "XDP is not supported when header-data split is enabled."); return -EOPNOTSUPP; } @@ -1727,17 +1730,20 @@ static int verify_xdp_configuration(struct net_device *dev) max_xdp_mtu -= GVE_RX_PAD; if (dev->mtu > max_xdp_mtu) { - netdev_warn(dev, "XDP is not supported for mtu %d.\n", - dev->mtu); + NL_SET_ERR_MSG_FMT_MOD(extack, + "XDP is not supported for mtu %d.", + dev->mtu); return -EOPNOTSUPP; } if (priv->rx_cfg.num_queues != priv->tx_cfg.num_queues || (2 * priv->tx_cfg.num_queues > priv->tx_cfg.max_queues)) { - netdev_warn(dev, "XDP load failed: The number of configured RX queues %d should be equal to the number of configured TX queues %d and the number of configured RX/TX queues should be less than or equal to half the maximum number of RX/TX queues %d", - priv->rx_cfg.num_queues, - priv->tx_cfg.num_queues, + netdev_warn(dev, + "XDP load failed: The number of configured RX queues %d should be equal to the number of configured TX queues %d and the number of configured RX/TX queues should be less than or equal to half the maximum number of RX/TX queues %d.", + priv->rx_cfg.num_queues, priv->tx_cfg.num_queues, priv->tx_cfg.max_queues); + NL_SET_ERR_MSG_MOD(extack, + "XDP load failed: The number of configured RX queues should be equal to the number of configured TX queues and the number of configured RX/TX queues should be less than or equal to half the maximum number of RX/TX queues"); return -EINVAL; } return 0; @@ -1748,7 +1754,7 @@ static int gve_xdp(struct net_device *dev, struct netdev_bpf *xdp) struct gve_priv *priv = netdev_priv(dev); int err; - err = verify_xdp_configuration(dev); + err = gve_verify_xdp_configuration(dev, xdp->extack); if (err) return err; switch (xdp->command) { From d235bb213f411ace8317bcca3740a1008628ea9c Mon Sep 17 00:00:00 2001 From: Ankit Garg Date: Thu, 6 Nov 2025 11:27:45 -0800 Subject: [PATCH 0696/1381] gve: Allow ethtool to configure rx_buf_len Add support for getting and setting the RX buffer length via the ethtool ring parameters (`ethtool -g`/`-G`). The driver restricts the allowed buffer length to 2048 (SZ_2K) by default and allows 4096 (SZ_4K) based on device options. As XDP is only supported when the `rx_buf_len` is 2048, the driver now enforces this in two places: 1. In `gve_xdp_set`, rejecting XDP programs if the current buffer length is not 2048. 2. In `gve_set_rx_buf_len_config`, rejecting buffer length changes if XDP is loaded and the new length is not 2048. Signed-off-by: Ankit Garg Reviewed-by: Harshitha Ramamurthy Reviewed-by: Jordan Rhee Reviewed-by: Willem de Bruijn Signed-off-by: Joshua Washington Link: https://patch.msgid.link/20251106192746.243525-4-joshwash@google.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/google/gve/gve.h | 9 +++++ drivers/net/ethernet/google/gve/gve_ethtool.c | 13 ++++++- drivers/net/ethernet/google/gve/gve_main.c | 39 +++++++++++++++++++ 3 files changed, 60 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/google/gve/gve.h b/drivers/net/ethernet/google/gve/gve.h index c237d00c5ab3..a33b44c1eb86 100644 --- a/drivers/net/ethernet/google/gve/gve.h +++ b/drivers/net/ethernet/google/gve/gve.h @@ -1167,6 +1167,12 @@ static inline bool gve_is_gqi(struct gve_priv *priv) priv->queue_format == GVE_GQI_QPL_FORMAT; } +static inline bool gve_is_dqo(struct gve_priv *priv) +{ + return priv->queue_format == GVE_DQO_RDA_FORMAT || + priv->queue_format == GVE_DQO_QPL_FORMAT; +} + static inline u32 gve_num_tx_queues(struct gve_priv *priv) { return priv->tx_cfg.num_queues + priv->tx_cfg.num_xdp_queues; @@ -1248,6 +1254,9 @@ void gve_rx_free_rings_gqi(struct gve_priv *priv, void gve_rx_start_ring_gqi(struct gve_priv *priv, int idx); void gve_rx_stop_ring_gqi(struct gve_priv *priv, int idx); bool gve_header_split_supported(const struct gve_priv *priv); +int gve_set_rx_buf_len_config(struct gve_priv *priv, u32 rx_buf_len, + struct netlink_ext_ack *extack, + struct gve_rx_alloc_rings_cfg *rx_alloc_cfg); int gve_set_hsplit_config(struct gve_priv *priv, u8 tcp_data_split, struct gve_rx_alloc_rings_cfg *rx_alloc_cfg); /* rx buffer handling */ diff --git a/drivers/net/ethernet/google/gve/gve_ethtool.c b/drivers/net/ethernet/google/gve/gve_ethtool.c index db6fc855a511..52500ae8348e 100644 --- a/drivers/net/ethernet/google/gve/gve_ethtool.c +++ b/drivers/net/ethernet/google/gve/gve_ethtool.c @@ -529,6 +529,8 @@ static void gve_get_ringparam(struct net_device *netdev, cmd->rx_pending = priv->rx_desc_cnt; cmd->tx_pending = priv->tx_desc_cnt; + kernel_cmd->rx_buf_len = priv->rx_cfg.packet_buffer_size; + if (!gve_header_split_supported(priv)) kernel_cmd->tcp_data_split = ETHTOOL_TCP_DATA_SPLIT_UNKNOWN; else if (priv->header_split_enabled) @@ -589,6 +591,12 @@ static int gve_set_ringparam(struct net_device *netdev, int err; gve_get_curr_alloc_cfgs(priv, &tx_alloc_cfg, &rx_alloc_cfg); + + err = gve_set_rx_buf_len_config(priv, kernel_cmd->rx_buf_len, extack, + &rx_alloc_cfg); + if (err) + return err; + err = gve_set_hsplit_config(priv, kernel_cmd->tcp_data_split, &rx_alloc_cfg); if (err) @@ -605,6 +613,8 @@ static int gve_set_ringparam(struct net_device *netdev, return err; } else { /* Set ring params for the next up */ + priv->rx_cfg.packet_buffer_size = + rx_alloc_cfg.packet_buffer_size; priv->header_split_enabled = rx_alloc_cfg.enable_header_split; priv->tx_desc_cnt = tx_alloc_cfg.ring_size; priv->rx_desc_cnt = rx_alloc_cfg.ring_size; @@ -944,7 +954,8 @@ static int gve_get_ts_info(struct net_device *netdev, const struct ethtool_ops gve_ethtool_ops = { .supported_coalesce_params = ETHTOOL_COALESCE_USECS, - .supported_ring_params = ETHTOOL_RING_USE_TCP_DATA_SPLIT, + .supported_ring_params = ETHTOOL_RING_USE_TCP_DATA_SPLIT | + ETHTOOL_RING_USE_RX_BUF_LEN, .get_drvinfo = gve_get_drvinfo, .get_strings = gve_get_strings, .get_sset_count = gve_get_sset_count, diff --git a/drivers/net/ethernet/google/gve/gve_main.c b/drivers/net/ethernet/google/gve/gve_main.c index 616182f4fb81..6fb8fbb38a7d 100644 --- a/drivers/net/ethernet/google/gve/gve_main.c +++ b/drivers/net/ethernet/google/gve/gve_main.c @@ -1725,6 +1725,13 @@ static int gve_verify_xdp_configuration(struct net_device *dev, return -EOPNOTSUPP; } + if (priv->rx_cfg.packet_buffer_size != SZ_2K) { + NL_SET_ERR_MSG_FMT_MOD(extack, + "XDP is not supported for Rx buf len %d, only %d supported.", + priv->rx_cfg.packet_buffer_size, SZ_2K); + return -EOPNOTSUPP; + } + max_xdp_mtu = priv->rx_cfg.packet_buffer_size - sizeof(struct ethhdr); if (priv->queue_format == GVE_GQI_QPL_FORMAT) max_xdp_mtu -= GVE_RX_PAD; @@ -2056,6 +2063,38 @@ bool gve_header_split_supported(const struct gve_priv *priv) priv->queue_format == GVE_DQO_RDA_FORMAT && !priv->xdp_prog; } +int gve_set_rx_buf_len_config(struct gve_priv *priv, u32 rx_buf_len, + struct netlink_ext_ack *extack, + struct gve_rx_alloc_rings_cfg *rx_alloc_cfg) +{ + u32 old_rx_buf_len = rx_alloc_cfg->packet_buffer_size; + + if (rx_buf_len == old_rx_buf_len) + return 0; + + /* device options may not always contain support for 4K buffers */ + if (!gve_is_dqo(priv) || priv->max_rx_buffer_size < SZ_4K) { + NL_SET_ERR_MSG_MOD(extack, + "Modifying Rx buf len is not supported"); + return -EOPNOTSUPP; + } + + if (priv->xdp_prog && rx_buf_len != SZ_2K) { + NL_SET_ERR_MSG_MOD(extack, + "Rx buf len can only be 2048 when XDP is on"); + return -EINVAL; + } + + if (rx_buf_len != SZ_2K && rx_buf_len != SZ_4K) { + NL_SET_ERR_MSG_MOD(extack, + "Rx buf len can only be 2048 or 4096"); + return -EINVAL; + } + rx_alloc_cfg->packet_buffer_size = rx_buf_len; + + return 0; +} + int gve_set_hsplit_config(struct gve_priv *priv, u8 tcp_data_split, struct gve_rx_alloc_rings_cfg *rx_alloc_cfg) { From 09a81a0f4fb7346f80a7340f725aedccff8ef8e4 Mon Sep 17 00:00:00 2001 From: Ankit Garg Date: Thu, 6 Nov 2025 11:27:46 -0800 Subject: [PATCH 0697/1381] gve: Default to max_rx_buffer_size for DQO if device supported Change the driver's default behavior to prefer the largest available RX buffer length supported by the device for DQO format, rather than always using the hardcoded 2K default. Previously, the driver would initialize with `GVE_DEFAULT_RX_BUFFER_SIZE` (2K), even if the device advertised support for a larger length (e.g., 4K). Performance observations: - With LRO disabled, we observed >10% improvement in RX single stream throughput when MTU >=2048. - With LRO enabled, we observed >10% improvement in RX single stream throughput when MTU >=1460. - No regressions were observed. Signed-off-by: Ankit Garg Reviewed-by: Harshitha Ramamurthy Reviewed-by: Jordan Rhee Reviewed-by: Willem de Bruijn Signed-off-by: Joshua Washington Link: https://patch.msgid.link/20251106192746.243525-5-joshwash@google.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/google/gve/gve_adminq.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/net/ethernet/google/gve/gve_adminq.c b/drivers/net/ethernet/google/gve/gve_adminq.c index 4f33d094a2ef..b72cc0fa2ba2 100644 --- a/drivers/net/ethernet/google/gve/gve_adminq.c +++ b/drivers/net/ethernet/google/gve/gve_adminq.c @@ -987,6 +987,10 @@ static void gve_enable_supported_features(struct gve_priv *priv, dev_info(&priv->pdev->dev, "BUFFER SIZES device option enabled with max_rx_buffer_size of %u, header_buf_size of %u.\n", priv->max_rx_buffer_size, priv->header_buf_size); + if (gve_is_dqo(priv) && + priv->max_rx_buffer_size > GVE_DEFAULT_RX_BUFFER_SIZE) + priv->rx_cfg.packet_buffer_size = + priv->max_rx_buffer_size; } /* Read and store ring size ranges given by device */ From 233542f5b4a8ea1fe3535f8021d37e7e92ffc4e5 Mon Sep 17 00:00:00 2001 From: Bitterblue Smith Date: Fri, 7 Nov 2025 18:06:27 +0200 Subject: [PATCH 0698/1381] wifi: rtw89: Use the correct power sequences for USB/SDIO Make rtw89_mac_pwr_seq() select the right parts of the power sequences based on the interface type. This is only relevant for RTL8852A. The other chips don't use power sequences. Signed-off-by: Bitterblue Smith Acked-by: Ping-Ke Shih Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/dec13310-06eb-429e-acb8-4c5b62656836@gmail.com --- drivers/net/wireless/realtek/rtw89/mac.c | 17 ++++++++++++++++- 1 file changed, 16 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/realtek/rtw89/mac.c b/drivers/net/wireless/realtek/rtw89/mac.c index 0a8474002cb7..e22b5d8c8171 100644 --- a/drivers/net/wireless/realtek/rtw89/mac.c +++ b/drivers/net/wireless/realtek/rtw89/mac.c @@ -1295,11 +1295,26 @@ static int rtw89_mac_sub_pwr_seq(struct rtw89_dev *rtwdev, u8 cv_msk, static int rtw89_mac_pwr_seq(struct rtw89_dev *rtwdev, const struct rtw89_pwr_cfg * const *cfg_seq) { + u8 intf_msk; int ret; + switch (rtwdev->hci.type) { + case RTW89_HCI_TYPE_PCIE: + intf_msk = PWR_INTF_MSK_PCIE; + break; + case RTW89_HCI_TYPE_USB: + intf_msk = PWR_INTF_MSK_USB; + break; + case RTW89_HCI_TYPE_SDIO: + intf_msk = PWR_INTF_MSK_SDIO; + break; + default: + return -EOPNOTSUPP; + } + for (; *cfg_seq; cfg_seq++) { ret = rtw89_mac_sub_pwr_seq(rtwdev, BIT(rtwdev->hal.cv), - PWR_INTF_MSK_PCIE, *cfg_seq); + intf_msk, *cfg_seq); if (ret) return -EBUSY; } From 0eea5e0f03db84d3d1a1a8926fca12295bda1f04 Mon Sep 17 00:00:00 2001 From: Bitterblue Smith Date: Fri, 7 Nov 2025 18:07:06 +0200 Subject: [PATCH 0699/1381] wifi: rtw89: Add rtw8852a_dle_mem_usb Add rtw8852a_dle_mem_usb and its various quotas and sizes in struct rtw89_mac_size_set. "dle" could be "Data Link Engine" or "Double Link Engine". These are some parameters needed for RTL8852AU. Signed-off-by: Bitterblue Smith Acked-by: Ping-Ke Shih Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/d0a09039-97a8-4501-b023-510c126d8c61@gmail.com --- drivers/net/wireless/realtek/rtw89/mac.c | 9 +++++++++ drivers/net/wireless/realtek/rtw89/mac.h | 5 +++++ drivers/net/wireless/realtek/rtw89/rtw8852a.c | 18 +++++++++++++++++- 3 files changed, 31 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/realtek/rtw89/mac.c b/drivers/net/wireless/realtek/rtw89/mac.c index e22b5d8c8171..dab0a76556c9 100644 --- a/drivers/net/wireless/realtek/rtw89/mac.c +++ b/drivers/net/wireless/realtek/rtw89/mac.c @@ -1669,6 +1669,8 @@ const struct rtw89_mac_size_set rtw89_mac_size = { /* PCIE 64 */ .wde_size0 = {RTW89_WDE_PG_64, 4095, 1,}, .wde_size0_v1 = {RTW89_WDE_PG_64, 3328, 0, 0,}, + /* 8852A USB */ + .wde_size1 = {RTW89_WDE_PG_64, 768, 0,}, /* DLFW */ .wde_size4 = {RTW89_WDE_PG_64, 0, 4096,}, .wde_size4_v1 = {RTW89_WDE_PG_64, 0, 3328, 0,}, @@ -1692,6 +1694,8 @@ const struct rtw89_mac_size_set rtw89_mac_size = { /* PCIE */ .ple_size0 = {RTW89_PLE_PG_128, 1520, 16,}, .ple_size0_v1 = {RTW89_PLE_PG_128, 2688, 240, 212992,}, + /* 8852A USB */ + .ple_size1 = {RTW89_PLE_PG_128, 3184, 16,}, .ple_size3_v1 = {RTW89_PLE_PG_128, 2928, 0, 212992,}, /* DLFW */ .ple_size4 = {RTW89_PLE_PG_128, 64, 1472,}, @@ -1715,6 +1719,8 @@ const struct rtw89_mac_size_set rtw89_mac_size = { /* PCIE 64 */ .wde_qt0 = {3792, 196, 0, 107,}, .wde_qt0_v1 = {3302, 6, 0, 20,}, + /* 8852A USB */ + .wde_qt1 = {512, 196, 0, 60,}, /* DLFW */ .wde_qt4 = {0, 0, 0, 0,}, /* PCIE 64 */ @@ -1743,6 +1749,9 @@ const struct rtw89_mac_size_set rtw89_mac_size = { .ple_qt13 = {0, 0, 16, 48, 0, 0, 0, 0, 0, 0, 0,}, /* PCIE 64 */ .ple_qt18 = {147, 0, 16, 20, 17, 13, 89, 0, 32, 14, 8, 0,}, + /* 8852A USB SCC */ + .ple_qt25 = {1536, 0, 16, 48, 13, 13, 360, 0, 32, 40, 8, 0,}, + .ple_qt26 = {2654, 0, 1134, 48, 64, 13, 1478, 0, 64, 128, 120, 0,}, /* USB 52C USB3.0 */ .ple_qt42 = {1068, 0, 16, 48, 4, 13, 178, 0, 16, 1, 8, 16, 0,}, /* USB 52C USB3.0 */ diff --git a/drivers/net/wireless/realtek/rtw89/mac.h b/drivers/net/wireless/realtek/rtw89/mac.h index ce3474a7ba03..868413335f6c 100644 --- a/drivers/net/wireless/realtek/rtw89/mac.h +++ b/drivers/net/wireless/realtek/rtw89/mac.h @@ -924,6 +924,7 @@ struct rtw89_mac_size_set { const struct rtw89_hfc_prec_cfg hfc_prec_cfg_c0; const struct rtw89_hfc_prec_cfg hfc_prec_cfg_c2; const struct rtw89_dle_size wde_size0; + const struct rtw89_dle_size wde_size1; const struct rtw89_dle_size wde_size0_v1; const struct rtw89_dle_size wde_size4; const struct rtw89_dle_size wde_size4_v1; @@ -937,6 +938,7 @@ struct rtw89_mac_size_set { const struct rtw89_dle_size wde_size25; const struct rtw89_dle_size wde_size31; const struct rtw89_dle_size ple_size0; + const struct rtw89_dle_size ple_size1; const struct rtw89_dle_size ple_size0_v1; const struct rtw89_dle_size ple_size3_v1; const struct rtw89_dle_size ple_size4; @@ -950,6 +952,7 @@ struct rtw89_mac_size_set { const struct rtw89_dle_size ple_size33; const struct rtw89_dle_size ple_size34; const struct rtw89_wde_quota wde_qt0; + const struct rtw89_wde_quota wde_qt1; const struct rtw89_wde_quota wde_qt0_v1; const struct rtw89_wde_quota wde_qt4; const struct rtw89_wde_quota wde_qt6; @@ -967,6 +970,8 @@ struct rtw89_mac_size_set { const struct rtw89_ple_quota ple_qt9; const struct rtw89_ple_quota ple_qt13; const struct rtw89_ple_quota ple_qt18; + const struct rtw89_ple_quota ple_qt25; + const struct rtw89_ple_quota ple_qt26; const struct rtw89_ple_quota ple_qt42; const struct rtw89_ple_quota ple_qt43; const struct rtw89_ple_quota ple_qt44; diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852a.c b/drivers/net/wireless/realtek/rtw89/rtw8852a.c index fa347dbebf9a..7855fd55b43e 100644 --- a/drivers/net/wireless/realtek/rtw89/rtw8852a.c +++ b/drivers/net/wireless/realtek/rtw89/rtw8852a.c @@ -65,6 +65,19 @@ static const struct rtw89_dle_mem rtw8852a_dle_mem_pcie[] = { NULL}, }; +static const struct rtw89_dle_mem rtw8852a_dle_mem_usb[] = { + [RTW89_QTA_SCC] = {RTW89_QTA_SCC, &rtw89_mac_size.wde_size1, + &rtw89_mac_size.ple_size1, &rtw89_mac_size.wde_qt1, + &rtw89_mac_size.wde_qt1, &rtw89_mac_size.ple_qt25, + &rtw89_mac_size.ple_qt26}, + [RTW89_QTA_DLFW] = {RTW89_QTA_DLFW, &rtw89_mac_size.wde_size4, + &rtw89_mac_size.ple_size4, &rtw89_mac_size.wde_qt4, + &rtw89_mac_size.wde_qt4, &rtw89_mac_size.ple_qt13, + &rtw89_mac_size.ple_qt13}, + [RTW89_QTA_INVALID] = {RTW89_QTA_INVALID, NULL, NULL, NULL, NULL, NULL, + NULL}, +}; + static const struct rtw89_reg2_def rtw8852a_pmac_ht20_mcs7_tbl[] = { {0x44AC, 0x00000000}, {0x44B0, 0x00000000}, @@ -2225,7 +2238,10 @@ const struct rtw89_chip_info rtw8852a_chip_info = { .dis_2g_40m_ul_ofdma = true, .rsvd_ple_ofst = 0x6f800, .hfc_param_ini = {rtw8852a_hfc_param_ini_pcie, NULL, NULL}, - .dle_mem = {rtw8852a_dle_mem_pcie, NULL, NULL, NULL}, + .dle_mem = {rtw8852a_dle_mem_pcie, + rtw8852a_dle_mem_usb, + rtw8852a_dle_mem_usb, + NULL}, .wde_qempty_acq_grpnum = 16, .wde_qempty_mgq_grpsel = 16, .rf_base_addr = {0xc000, 0xd000}, From c19b106609f398b3b1b14a89dc68ed0244420d08 Mon Sep 17 00:00:00 2001 From: Bitterblue Smith Date: Fri, 7 Nov 2025 18:08:14 +0200 Subject: [PATCH 0700/1381] wifi: rtw89: Add rtw8852a_hfc_param_ini_usb "hfc" means "hci fc" which is "Host Control Interface Flow Control". These are some parameters needed for RTL8852AU. Signed-off-by: Bitterblue Smith Acked-by: Ping-Ke Shih Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/d1032888-2a53-4c52-a8b9-6e00cd6758dc@gmail.com --- drivers/net/wireless/realtek/rtw89/rtw8852a.c | 46 ++++++++++++++++++- 1 file changed, 45 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852a.c b/drivers/net/wireless/realtek/rtw89/rtw8852a.c index 7855fd55b43e..76b64fd5a639 100644 --- a/drivers/net/wireless/realtek/rtw89/rtw8852a.c +++ b/drivers/net/wireless/realtek/rtw89/rtw8852a.c @@ -48,6 +48,48 @@ static const struct rtw89_hfc_param_ini rtw8852a_hfc_param_ini_pcie[] = { [RTW89_QTA_INVALID] = {NULL}, }; +static const struct rtw89_hfc_ch_cfg rtw8852a_hfc_chcfg_usb[] = { + {22, 402, grp_0}, /* ACH 0 */ + {0, 0, grp_0}, /* ACH 1 */ + {22, 402, grp_0}, /* ACH 2 */ + {0, 0, grp_0}, /* ACH 3 */ + {22, 402, grp_0}, /* ACH 4 */ + {0, 0, grp_0}, /* ACH 5 */ + {22, 402, grp_0}, /* ACH 6 */ + {0, 0, grp_0}, /* ACH 7 */ + {22, 402, grp_0}, /* B0MGQ */ + {0, 0, grp_0}, /* B0HIQ */ + {22, 402, grp_0}, /* B1MGQ */ + {0, 0, grp_0}, /* B1HIQ */ + {0, 0, 0} /* FWCMDQ */ +}; + +static const struct rtw89_hfc_pub_cfg rtw8852a_hfc_pubcfg_usb = { + 512, /* Group 0 */ + 0, /* Group 1 */ + 512, /* Public Max */ + 104 /* WP threshold */ +}; + +static const struct rtw89_hfc_prec_cfg rtw8852a_hfc_preccfg_usb = { + 11, /* CH 0-11 pre-cost */ + 32, /* H2C pre-cost */ + 76, /* WP CH 0-7 pre-cost */ + 25, /* WP CH 8-11 pre-cost */ + 1, /* CH 0-11 full condition */ + 1, /* H2C full condition */ + 1, /* WP CH 0-7 full condition */ + 1, /* WP CH 8-11 full condition */ +}; + +static const struct rtw89_hfc_param_ini rtw8852a_hfc_param_ini_usb[] = { + [RTW89_QTA_SCC] = {rtw8852a_hfc_chcfg_usb, &rtw8852a_hfc_pubcfg_usb, + &rtw8852a_hfc_preccfg_usb, RTW89_HCIFC_STF}, + [RTW89_QTA_DLFW] = {NULL, NULL, + &rtw8852a_hfc_preccfg_usb, RTW89_HCIFC_STF}, + [RTW89_QTA_INVALID] = {NULL}, +}; + static const struct rtw89_dle_mem rtw8852a_dle_mem_pcie[] = { [RTW89_QTA_SCC] = {RTW89_QTA_SCC, &rtw89_mac_size.wde_size0, &rtw89_mac_size.ple_size0, &rtw89_mac_size.wde_qt0, @@ -2237,7 +2279,9 @@ const struct rtw89_chip_info rtw8852a_chip_info = { .max_amsdu_limit = 3500, .dis_2g_40m_ul_ofdma = true, .rsvd_ple_ofst = 0x6f800, - .hfc_param_ini = {rtw8852a_hfc_param_ini_pcie, NULL, NULL}, + .hfc_param_ini = {rtw8852a_hfc_param_ini_pcie, + rtw8852a_hfc_param_ini_usb, + NULL}, .dle_mem = {rtw8852a_dle_mem_pcie, rtw8852a_dle_mem_usb, rtw8852a_dle_mem_usb, From 1dfd11e70022a0117bdf3fd02d2b541c9c732845 Mon Sep 17 00:00:00 2001 From: Bitterblue Smith Date: Fri, 7 Nov 2025 18:08:46 +0200 Subject: [PATCH 0701/1381] wifi: rtw89: 8852a: Accept USB devices and load their MAC address Make rtw8852a_read_efuse() accept USB devices and load the MAC address from the correct offset. Signed-off-by: Bitterblue Smith Acked-by: Ping-Ke Shih Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/97412e2c-ac98-403c-9056-9d9fe1ed9f28@gmail.com --- drivers/net/wireless/realtek/rtw89/rtw8852a.c | 16 +++++++--------- 1 file changed, 7 insertions(+), 9 deletions(-) diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852a.c b/drivers/net/wireless/realtek/rtw89/rtw8852a.c index 76b64fd5a639..48205aa4a980 100644 --- a/drivers/net/wireless/realtek/rtw89/rtw8852a.c +++ b/drivers/net/wireless/realtek/rtw89/rtw8852a.c @@ -621,14 +621,6 @@ static const struct rtw89_edcca_regs rtw8852a_edcca_regs = { .tx_collision_t2r_st_mask = B_TX_COLLISION_T2R_ST_M, }; -static void rtw8852ae_efuse_parsing(struct rtw89_efuse *efuse, - struct rtw8852a_efuse *map) -{ - ether_addr_copy(efuse->addr, map->e.mac_addr); - efuse->rfe_type = map->rfe_type; - efuse->xtal_cap = map->xtal_k; -} - static void rtw8852a_efuse_parsing_tssi(struct rtw89_dev *rtwdev, struct rtw8852a_efuse *map) { @@ -674,12 +666,18 @@ static int rtw8852a_read_efuse(struct rtw89_dev *rtwdev, u8 *log_map, switch (rtwdev->hci.type) { case RTW89_HCI_TYPE_PCIE: - rtw8852ae_efuse_parsing(efuse, map); + ether_addr_copy(efuse->addr, map->e.mac_addr); + break; + case RTW89_HCI_TYPE_USB: + ether_addr_copy(efuse->addr, map->u.mac_addr); break; default: return -ENOTSUPP; } + efuse->rfe_type = map->rfe_type; + efuse->xtal_cap = map->xtal_k; + rtw89_info(rtwdev, "chip rfe_type is %d\n", efuse->rfe_type); return 0; From 0029ccab53accee4919971547f4e38036ef62b2f Mon Sep 17 00:00:00 2001 From: Bitterblue Smith Date: Fri, 7 Nov 2025 18:09:22 +0200 Subject: [PATCH 0702/1381] wifi: rtw89: Add rtw8852au.c This is the entry point of the new rtw89_8852au module. Signed-off-by: Bitterblue Smith Acked-by: Ping-Ke Shih Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/9580e5be-2bd1-45f9-ab75-616b86e25694@gmail.com --- .../net/wireless/realtek/rtw89/rtw8852au.c | 79 +++++++++++++++++++ 1 file changed, 79 insertions(+) create mode 100644 drivers/net/wireless/realtek/rtw89/rtw8852au.c diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852au.c b/drivers/net/wireless/realtek/rtw89/rtw8852au.c new file mode 100644 index 000000000000..ca782469c455 --- /dev/null +++ b/drivers/net/wireless/realtek/rtw89/rtw8852au.c @@ -0,0 +1,79 @@ +// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause +/* Copyright(c) 2025 Realtek Corporation + */ + +#include +#include +#include "rtw8852a.h" +#include "reg.h" +#include "usb.h" + +static const struct rtw89_usb_info rtw8852a_usb_info = { + .usb_host_request_2 = R_AX_USB_HOST_REQUEST_2, + .usb_wlan0_1 = R_AX_USB_WLAN0_1, + .hci_func_en = R_AX_HCI_FUNC_EN, + .usb3_mac_npi_config_intf_0 = R_AX_USB3_MAC_NPI_CONFIG_INTF_0, + .usb_endpoint_0 = R_AX_USB_ENDPOINT_0, + .usb_endpoint_2 = R_AX_USB_ENDPOINT_2, + .bulkout_id = { + [RTW89_DMA_ACH0] = 3, + [RTW89_DMA_ACH2] = 5, + [RTW89_DMA_ACH4] = 4, + [RTW89_DMA_ACH6] = 6, + [RTW89_DMA_B0MG] = 0, + [RTW89_DMA_B0HI] = 0, + [RTW89_DMA_B1MG] = 1, + [RTW89_DMA_B1HI] = 1, + [RTW89_DMA_H2C] = 2, + }, +}; + +static const struct rtw89_driver_info rtw89_8852au_info = { + .chip = &rtw8852a_chip_info, + .variant = NULL, + .quirks = NULL, + .bus = { + .usb = &rtw8852a_usb_info, + } +}; + +static const struct usb_device_id rtw_8852au_id_table[] = { + { USB_DEVICE_AND_INTERFACE_INFO(0x0411, 0x0312, 0xff, 0xff, 0xff), + .driver_info = (kernel_ulong_t)&rtw89_8852au_info }, + { USB_DEVICE_AND_INTERFACE_INFO(0x056e, 0x4020, 0xff, 0xff, 0xff), + .driver_info = (kernel_ulong_t)&rtw89_8852au_info }, + { USB_DEVICE_AND_INTERFACE_INFO(0x0b05, 0x1997, 0xff, 0xff, 0xff), + .driver_info = (kernel_ulong_t)&rtw89_8852au_info }, + { USB_DEVICE_AND_INTERFACE_INFO(0x0bda, 0x8832, 0xff, 0xff, 0xff), + .driver_info = (kernel_ulong_t)&rtw89_8852au_info }, + { USB_DEVICE_AND_INTERFACE_INFO(0x0bda, 0x885a, 0xff, 0xff, 0xff), + .driver_info = (kernel_ulong_t)&rtw89_8852au_info }, + { USB_DEVICE_AND_INTERFACE_INFO(0x0bda, 0x885c, 0xff, 0xff, 0xff), + .driver_info = (kernel_ulong_t)&rtw89_8852au_info }, + { USB_DEVICE_AND_INTERFACE_INFO(0x2001, 0x3321, 0xff, 0xff, 0xff), + .driver_info = (kernel_ulong_t)&rtw89_8852au_info }, + { USB_DEVICE_AND_INTERFACE_INFO(0x2001, 0x332c, 0xff, 0xff, 0xff), + .driver_info = (kernel_ulong_t)&rtw89_8852au_info }, + { USB_DEVICE_AND_INTERFACE_INFO(0x2357, 0x013f, 0xff, 0xff, 0xff), + .driver_info = (kernel_ulong_t)&rtw89_8852au_info }, + { USB_DEVICE_AND_INTERFACE_INFO(0x2357, 0x0140, 0xff, 0xff, 0xff), + .driver_info = (kernel_ulong_t)&rtw89_8852au_info }, + { USB_DEVICE_AND_INTERFACE_INFO(0x2357, 0x0141, 0xff, 0xff, 0xff), + .driver_info = (kernel_ulong_t)&rtw89_8852au_info }, + { USB_DEVICE_AND_INTERFACE_INFO(0x3625, 0x010f, 0xff, 0xff, 0xff), + .driver_info = (kernel_ulong_t)&rtw89_8852au_info }, + {}, +}; +MODULE_DEVICE_TABLE(usb, rtw_8852au_id_table); + +static struct usb_driver rtw_8852au_driver = { + .name = KBUILD_MODNAME, + .id_table = rtw_8852au_id_table, + .probe = rtw89_usb_probe, + .disconnect = rtw89_usb_disconnect, +}; +module_usb_driver(rtw_8852au_driver); + +MODULE_AUTHOR("Bitterblue Smith "); +MODULE_DESCRIPTION("Realtek 802.11ax wireless 8852AU driver"); +MODULE_LICENSE("Dual BSD/GPL"); From 623c177323ec16c2c72b4ea0ad68bda223928488 Mon Sep 17 00:00:00 2001 From: Bitterblue Smith Date: Fri, 7 Nov 2025 18:10:19 +0200 Subject: [PATCH 0703/1381] wifi: rtw89: Enable the new rtw89_8852au module Tested in station mode and a little in AP mode. Signed-off-by: Bitterblue Smith Acked-by: Ping-Ke Shih Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/180a5e77-9297-4ffc-80d5-191dfef47661@gmail.com --- drivers/net/wireless/realtek/rtw89/Kconfig | 11 +++++++++++ drivers/net/wireless/realtek/rtw89/Makefile | 3 +++ 2 files changed, 14 insertions(+) diff --git a/drivers/net/wireless/realtek/rtw89/Kconfig b/drivers/net/wireless/realtek/rtw89/Kconfig index e89f7481475a..44d8a7f32bf2 100644 --- a/drivers/net/wireless/realtek/rtw89/Kconfig +++ b/drivers/net/wireless/realtek/rtw89/Kconfig @@ -74,6 +74,17 @@ config RTW89_8852AE 802.11ax PCIe wireless network (Wi-Fi 6) adapter +config RTW89_8852AU + tristate "Realtek 8852AU USB wireless network (Wi-Fi 6) adapter" + depends on USB + select RTW89_CORE + select RTW89_USB + select RTW89_8852A + help + Select this option will enable support for 8852AU chipset + + 802.11ax USB wireless network (Wi-Fi 6) adapter + config RTW89_8852BE tristate "Realtek 8852BE PCI wireless network (Wi-Fi 6) adapter" depends on PCI diff --git a/drivers/net/wireless/realtek/rtw89/Makefile b/drivers/net/wireless/realtek/rtw89/Makefile index e0d21972e57c..1be81f254fca 100644 --- a/drivers/net/wireless/realtek/rtw89/Makefile +++ b/drivers/net/wireless/realtek/rtw89/Makefile @@ -43,6 +43,9 @@ rtw89_8852a-objs := rtw8852a.o \ obj-$(CONFIG_RTW89_8852AE) += rtw89_8852ae.o rtw89_8852ae-objs := rtw8852ae.o +obj-$(CONFIG_RTW89_8852AU) += rtw89_8852au.o +rtw89_8852au-objs := rtw8852au.o + obj-$(CONFIG_RTW89_8852B_COMMON) += rtw89_8852b_common.o rtw89_8852b_common-objs := rtw8852b_common.o From 12e6bea1aac85cc98a55031ebf10db969eeafa34 Mon Sep 17 00:00:00 2001 From: Ping-Ke Shih Date: Thu, 6 Nov 2025 08:49:32 +0800 Subject: [PATCH 0704/1381] wifi: rtlwifi: rtl8188ee: correct allstasleep in P2P PS H2C command The "->allstasleep" variable is a 1 bit bitfield. It can only be 0 or 1. This "= -1" assignement should be "= 0" as other chips and vendor driver. Reported-by: Dan Carpenter Closes: https://lore.kernel.org/linux-wireless/5c992936-4e7b-4c0a-abfc-0ec0fb9ef9fd@suswa.mountain/T/#t Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/1762390172-21091-1-git-send-email-pkshih@realtek.com --- drivers/net/wireless/realtek/rtlwifi/rtl8188ee/fw.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/fw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/fw.c index 7252bc621211..9a9f9e14f472 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/fw.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/fw.c @@ -694,7 +694,7 @@ void rtl88e_set_p2p_ps_offload_cmd(struct ieee80211_hw *hw, u8 p2p_ps_state) if (P2P_ROLE_GO == rtlpriv->mac80211.p2p) { p2p_ps_offload->role = 1; - p2p_ps_offload->allstasleep = -1; + p2p_ps_offload->allstasleep = 0; } else { p2p_ps_offload->role = 0; } From e483a615a609d558d7ca8c161f6aedfb39226e7b Mon Sep 17 00:00:00 2001 From: Marco Crivellari Date: Fri, 7 Nov 2025 14:44:52 +0100 Subject: [PATCH 0705/1381] isdn: kcapi: add WQ_PERCPU to alloc_workqueue users MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Currently if a user enqueues a work item using schedule_delayed_work() the used wq is "system_wq" (per-cpu wq) while queue_delayed_work() use WORK_CPU_UNBOUND (used when a cpu is not specified). The same applies to schedule_work() that is using system_wq and queue_work(), that makes use again of WORK_CPU_UNBOUND. This lack of consistency cannot be addressed without refactoring the API. alloc_workqueue() treats all queues as per-CPU by default, while unbound workqueues must opt-in via WQ_UNBOUND. This default is suboptimal: most workloads benefit from unbound queues, allowing the scheduler to place worker threads where they’re needed and reducing noise when CPUs are isolated. This continues the effort to refactor workqueue APIs, which began with the introduction of new workqueues and a new alloc_workqueue flag in: commit 128ea9f6ccfb ("workqueue: Add system_percpu_wq and system_dfl_wq") commit 930c2ea566af ("workqueue: Add new WQ_PERCPU flag") This change adds a new WQ_PERCPU flag to explicitly request alloc_workqueue() to be per-cpu when WQ_UNBOUND has not been specified. With the introduction of the WQ_PERCPU flag (equivalent to !WQ_UNBOUND), any alloc_workqueue() caller that doesn’t explicitly specify WQ_UNBOUND must now use WQ_PERCPU. Once migration is complete, WQ_UNBOUND can be removed and unbound will become the implicit default. Suggested-by: Tejun Heo Signed-off-by: Marco Crivellari Link: https://patch.msgid.link/20251107134452.198378-1-marco.crivellari@suse.com Signed-off-by: Jakub Kicinski --- drivers/isdn/capi/kcapi.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/isdn/capi/kcapi.c b/drivers/isdn/capi/kcapi.c index c5d13bdc239b..e8f7e52354bc 100644 --- a/drivers/isdn/capi/kcapi.c +++ b/drivers/isdn/capi/kcapi.c @@ -907,7 +907,7 @@ int __init kcapi_init(void) { int err; - kcapi_wq = alloc_workqueue("kcapi", 0, 0); + kcapi_wq = alloc_workqueue("kcapi", WQ_PERCPU, 0); if (!kcapi_wq) return -ENOMEM; From b981e100c19dcd91ce8cca8562c3cdabd4fcf28c Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Sat, 8 Nov 2025 22:59:51 +0100 Subject: [PATCH 0706/1381] net: dsa: loop: use new helper fixed_phy_register_100fd to simplify the code Use new helper fixed_phy_register_100fd to simplify the code. Signed-off-by: Heiner Kallweit Reviewed-by: Vladimir Oltean Link: https://patch.msgid.link/922f1b45-1748-4dd2-87eb-9d018df44731@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/dsa_loop.c | 7 +------ 1 file changed, 1 insertion(+), 6 deletions(-) diff --git a/drivers/net/dsa/dsa_loop.c b/drivers/net/dsa/dsa_loop.c index 650d93226d9f..4a416f2717ba 100644 --- a/drivers/net/dsa/dsa_loop.c +++ b/drivers/net/dsa/dsa_loop.c @@ -441,11 +441,6 @@ static int __init dsa_loop_create_switch_mdiodev(void) static int __init dsa_loop_init(void) { - struct fixed_phy_status status = { - .link = 1, - .speed = SPEED_100, - .duplex = DUPLEX_FULL, - }; unsigned int i; int ret; @@ -454,7 +449,7 @@ static int __init dsa_loop_init(void) return ret; for (i = 0; i < NUM_FIXED_PHYS; i++) - phydevs[i] = fixed_phy_register(&status, NULL); + phydevs[i] = fixed_phy_register_100fd(); ret = mdio_driver_register(&dsa_loop_drv); if (ret) { From 8da7bea7db692e786165b71729fb68b7ff65ee56 Mon Sep 17 00:00:00 2001 From: Jason Xing Date: Fri, 31 Oct 2025 18:33:28 +0800 Subject: [PATCH 0707/1381] xsk: add indirect call for xsk_destruct_skb Since Eric proposed an idea about adding indirect call wrappers for UDP and managed to see a huge improvement[1], the same situation can also be applied in xsk scenario. This patch adds an indirect call for xsk and helps current copy mode improve the performance by around 1% stably which was observed with IXGBE at 10Gb/sec loaded. If the throughput grows, the positive effect will be magnified. I applied this patch on top of batch xmit series[2], and was able to see <5% improvement from our internal application which is a little bit unstable though. Use INDIRECT wrappers to keep xsk_destruct_skb static as it used to be when the mitigation config is off. Be aware of the freeing path that can be very hot since the frequency can reach around 2,000,000 times per second with the xdpsock test. [1]: https://lore.kernel.org/netdev/20251006193103.2684156-2-edumazet@google.com/ [2]: https://lore.kernel.org/all/20251021131209.41491-1-kerneljasonxing@gmail.com/ Suggested-by: Alexander Lobakin Signed-off-by: Jason Xing Reviewed-by: Alexander Lobakin Link: https://patch.msgid.link/20251031103328.95468-1-kerneljasonxing@gmail.com Signed-off-by: Paolo Abeni --- include/net/xdp_sock.h | 7 +++++++ net/core/skbuff.c | 8 +++++--- net/xdp/xsk.c | 3 ++- 3 files changed, 14 insertions(+), 4 deletions(-) diff --git a/include/net/xdp_sock.h b/include/net/xdp_sock.h index ce587a225661..23e8861e8b25 100644 --- a/include/net/xdp_sock.h +++ b/include/net/xdp_sock.h @@ -125,6 +125,7 @@ struct xsk_tx_metadata_ops { int xsk_generic_rcv(struct xdp_sock *xs, struct xdp_buff *xdp); int __xsk_map_redirect(struct xdp_sock *xs, struct xdp_buff *xdp); void __xsk_map_flush(struct list_head *flush_list); +INDIRECT_CALLABLE_DECLARE(void xsk_destruct_skb(struct sk_buff *)); /** * xsk_tx_metadata_to_compl - Save enough relevant metadata information @@ -218,6 +219,12 @@ static inline void __xsk_map_flush(struct list_head *flush_list) { } +#ifdef CONFIG_MITIGATION_RETPOLINE +static inline void xsk_destruct_skb(struct sk_buff *skb) +{ +} +#endif + static inline void xsk_tx_metadata_to_compl(struct xsk_tx_metadata *meta, struct xsk_tx_metadata_compl *compl) { diff --git a/net/core/skbuff.c b/net/core/skbuff.c index d95658b738d1..4f4d7ab7057f 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -81,6 +81,7 @@ #include #include #include +#include #include #include @@ -1140,12 +1141,13 @@ void skb_release_head_state(struct sk_buff *skb) if (skb->destructor) { DEBUG_NET_WARN_ON_ONCE(in_hardirq()); #ifdef CONFIG_INET - INDIRECT_CALL_3(skb->destructor, + INDIRECT_CALL_4(skb->destructor, tcp_wfree, __sock_wfree, sock_wfree, + xsk_destruct_skb, skb); #else - INDIRECT_CALL_1(skb->destructor, - sock_wfree, + INDIRECT_CALL_2(skb->destructor, + sock_wfree, xsk_destruct_skb, skb); #endif diff --git a/net/xdp/xsk.c b/net/xdp/xsk.c index ed8b612ec29d..bcfd400e9cf8 100644 --- a/net/xdp/xsk.c +++ b/net/xdp/xsk.c @@ -602,7 +602,8 @@ static u32 xsk_get_num_desc(struct sk_buff *skb) return XSKCB(skb)->num_descs; } -static void xsk_destruct_skb(struct sk_buff *skb) +INDIRECT_CALLABLE_SCOPE +void xsk_destruct_skb(struct sk_buff *skb) { struct xsk_tx_metadata_compl *compl = &skb_shinfo(skb)->xsk_meta; From 706edca67984faa3d83e397334fa5ec4190f63ba Mon Sep 17 00:00:00 2001 From: Emmanuel Grumbach Date: Mon, 10 Nov 2025 14:03:15 +0200 Subject: [PATCH 0708/1381] wifi: cfg80211: use a C99 initializer in wiphy_register struct regulatory request was not fully initialized. While this is not really a big deal because nl80211_send_reg_change_event doesn't look at the other fields, it still makes sense to zero all the other fields as Coverity suggests. Signed-off-by: Emmanuel Grumbach Reviewed-by: Johannes Berg Signed-off-by: Miri Korenblit Link: https://patch.msgid.link/20251110140230.f8d4fcb1328b.I87170b1caef04356809838e684c9499f5806e624@changeid Signed-off-by: Johannes Berg --- net/wireless/core.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/net/wireless/core.c b/net/wireless/core.c index 9f858a83e912..aa78dbaaf093 100644 --- a/net/wireless/core.c +++ b/net/wireless/core.c @@ -1066,12 +1066,12 @@ int wiphy_register(struct wiphy *wiphy) wiphy_regulatory_register(wiphy); if (wiphy->regulatory_flags & REGULATORY_CUSTOM_REG) { - struct regulatory_request request; - - request.wiphy_idx = get_wiphy_idx(wiphy); - request.initiator = NL80211_REGDOM_SET_BY_DRIVER; - request.alpha2[0] = '9'; - request.alpha2[1] = '9'; + struct regulatory_request request = { + .wiphy_idx = get_wiphy_idx(wiphy), + .initiator = NL80211_REGDOM_SET_BY_DRIVER, + .alpha2[0] = '9', + .alpha2[1] = '9', + }; nl80211_send_reg_change_event(&request); } From e18efacc9c2f17b12c6e019cabad70a2989bd3a9 Mon Sep 17 00:00:00 2001 From: Pagadala Yesu Anjaneyulu Date: Mon, 10 Nov 2025 14:10:29 +0200 Subject: [PATCH 0709/1381] wifi: cfg80211/mac80211: clean up duplicate ap_power handling Move duplicated ap_power type handling code to an inline function in cfg80211. Signed-off-by: Pagadala Yesu Anjaneyulu Reviewed-by: Johannes Berg Signed-off-by: Miri Korenblit Link: https://patch.msgid.link/20251110140806.959948da1cb5.I893b5168329fb3232f249c182a35c99804112da6@changeid Signed-off-by: Johannes Berg --- include/net/cfg80211.h | 24 ++++++++++++++++++++++++ net/mac80211/mlme.c | 20 +------------------- net/wireless/scan.c | 15 +-------------- 3 files changed, 26 insertions(+), 33 deletions(-) diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index d87c18e1b133..1b257eaf8de5 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -10134,6 +10134,30 @@ static inline int cfg80211_color_change_notify(struct net_device *dev, 0, 0, link_id); } +/** + * cfg80211_6ghz_power_type - determine AP regulatory power type + * @control: control flags + * + * Return: regulatory power type from &enum ieee80211_ap_reg_power + */ +static inline enum ieee80211_ap_reg_power +cfg80211_6ghz_power_type(u8 control) +{ + switch (u8_get_bits(control, IEEE80211_HE_6GHZ_OPER_CTRL_REG_INFO)) { + case IEEE80211_6GHZ_CTRL_REG_LPI_AP: + case IEEE80211_6GHZ_CTRL_REG_INDOOR_LPI_AP: + return IEEE80211_REG_LPI_AP; + case IEEE80211_6GHZ_CTRL_REG_SP_AP: + case IEEE80211_6GHZ_CTRL_REG_INDOOR_SP_AP: + case IEEE80211_6GHZ_CTRL_REG_INDOOR_SP_AP_OLD: + return IEEE80211_REG_SP_AP; + case IEEE80211_6GHZ_CTRL_REG_VLP_AP: + return IEEE80211_REG_VLP_AP; + default: + return IEEE80211_REG_UNSET_AP; + } +} + /** * cfg80211_links_removed - Notify about removed STA MLD setup links. * @dev: network device. diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c index 804c3a95b7c6..64230696f3e4 100644 --- a/net/mac80211/mlme.c +++ b/net/mac80211/mlme.c @@ -6034,24 +6034,6 @@ ieee80211_determine_our_sta_mode_assoc(struct ieee80211_sub_if_data *sdata, conn->bw_limit, tmp.bw_limit); } -static enum ieee80211_ap_reg_power -ieee80211_ap_power_type(u8 control) -{ - switch (u8_get_bits(control, IEEE80211_HE_6GHZ_OPER_CTRL_REG_INFO)) { - case IEEE80211_6GHZ_CTRL_REG_LPI_AP: - case IEEE80211_6GHZ_CTRL_REG_INDOOR_LPI_AP: - return IEEE80211_REG_LPI_AP; - case IEEE80211_6GHZ_CTRL_REG_SP_AP: - case IEEE80211_6GHZ_CTRL_REG_INDOOR_SP_AP: - case IEEE80211_6GHZ_CTRL_REG_INDOOR_SP_AP_OLD: - return IEEE80211_REG_SP_AP; - case IEEE80211_6GHZ_CTRL_REG_VLP_AP: - return IEEE80211_REG_VLP_AP; - default: - return IEEE80211_REG_UNSET_AP; - } -} - static int ieee80211_prep_channel(struct ieee80211_sub_if_data *sdata, struct ieee80211_link_data *link, int link_id, @@ -6094,7 +6076,7 @@ static int ieee80211_prep_channel(struct ieee80211_sub_if_data *sdata, he_6ghz_oper = ieee80211_he_6ghz_oper(elems->he_operation); if (he_6ghz_oper) link->conf->power_type = - ieee80211_ap_power_type(he_6ghz_oper->control); + cfg80211_6ghz_power_type(he_6ghz_oper->control); else link_info(link, "HE 6 GHz operation missing (on %d MHz), expect issues\n", diff --git a/net/wireless/scan.c b/net/wireless/scan.c index 90a9187a6b13..68c4130d602f 100644 --- a/net/wireless/scan.c +++ b/net/wireless/scan.c @@ -2230,20 +2230,7 @@ cfg80211_get_6ghz_power_type(const u8 *elems, size_t elems_len) if (!he_6ghz_oper) return IEEE80211_REG_UNSET_AP; - switch (u8_get_bits(he_6ghz_oper->control, - IEEE80211_HE_6GHZ_OPER_CTRL_REG_INFO)) { - case IEEE80211_6GHZ_CTRL_REG_LPI_AP: - case IEEE80211_6GHZ_CTRL_REG_INDOOR_LPI_AP: - return IEEE80211_REG_LPI_AP; - case IEEE80211_6GHZ_CTRL_REG_SP_AP: - case IEEE80211_6GHZ_CTRL_REG_INDOOR_SP_AP: - case IEEE80211_6GHZ_CTRL_REG_INDOOR_SP_AP_OLD: - return IEEE80211_REG_SP_AP; - case IEEE80211_6GHZ_CTRL_REG_VLP_AP: - return IEEE80211_REG_VLP_AP; - default: - return IEEE80211_REG_UNSET_AP; - } + return cfg80211_6ghz_power_type(he_6ghz_oper->control); } static bool cfg80211_6ghz_power_type_valid(const u8 *elems, size_t elems_len, From b54cf0f4495a8f3fa94245cdda7716792400299e Mon Sep 17 00:00:00 2001 From: Pagadala Yesu Anjaneyulu Date: Mon, 10 Nov 2025 14:10:30 +0200 Subject: [PATCH 0710/1381] wifi: cfg80211/mac80211: Add fallback mechanism for INDOOR_SP connection Implement fallback to LPI mode when SP mode is not permitted by regulatory constraints for INDOOR_SP connections. Limit fallback mechanism to client mode. Signed-off-by: Pagadala Yesu Anjaneyulu Reviewed-by: Johannes Berg Signed-off-by: Miri Korenblit Link: https://patch.msgid.link/20251110140806.8b43201a34ae.I37fc7bb5892eb9d044d619802e8f2095fde6b296@changeid Signed-off-by: Johannes Berg --- include/net/cfg80211.h | 9 +++++++-- net/mac80211/mlme.c | 3 ++- net/wireless/core.h | 3 ++- net/wireless/nl80211.c | 4 ++-- net/wireless/scan.c | 7 ++++--- 5 files changed, 17 insertions(+), 9 deletions(-) diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index 1b257eaf8de5..625cb2c78361 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -10137,22 +10137,27 @@ static inline int cfg80211_color_change_notify(struct net_device *dev, /** * cfg80211_6ghz_power_type - determine AP regulatory power type * @control: control flags + * @client_flags: &enum ieee80211_channel_flags for station mode to enable + * SP to LPI fallback, zero otherwise. * * Return: regulatory power type from &enum ieee80211_ap_reg_power */ static inline enum ieee80211_ap_reg_power -cfg80211_6ghz_power_type(u8 control) +cfg80211_6ghz_power_type(u8 control, u32 client_flags) { switch (u8_get_bits(control, IEEE80211_HE_6GHZ_OPER_CTRL_REG_INFO)) { case IEEE80211_6GHZ_CTRL_REG_LPI_AP: case IEEE80211_6GHZ_CTRL_REG_INDOOR_LPI_AP: return IEEE80211_REG_LPI_AP; case IEEE80211_6GHZ_CTRL_REG_SP_AP: - case IEEE80211_6GHZ_CTRL_REG_INDOOR_SP_AP: case IEEE80211_6GHZ_CTRL_REG_INDOOR_SP_AP_OLD: return IEEE80211_REG_SP_AP; case IEEE80211_6GHZ_CTRL_REG_VLP_AP: return IEEE80211_REG_VLP_AP; + case IEEE80211_6GHZ_CTRL_REG_INDOOR_SP_AP: + if (client_flags & IEEE80211_CHAN_NO_6GHZ_AFC_CLIENT) + return IEEE80211_REG_LPI_AP; + return IEEE80211_REG_SP_AP; default: return IEEE80211_REG_UNSET_AP; } diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c index 64230696f3e4..c705d3f45aff 100644 --- a/net/mac80211/mlme.c +++ b/net/mac80211/mlme.c @@ -6076,7 +6076,8 @@ static int ieee80211_prep_channel(struct ieee80211_sub_if_data *sdata, he_6ghz_oper = ieee80211_he_6ghz_oper(elems->he_operation); if (he_6ghz_oper) link->conf->power_type = - cfg80211_6ghz_power_type(he_6ghz_oper->control); + cfg80211_6ghz_power_type(he_6ghz_oper->control, + cbss->channel->flags); else link_info(link, "HE 6 GHz operation missing (on %d MHz), expect issues\n", diff --git a/net/wireless/core.h b/net/wireless/core.h index b6bd7f4d6385..82f343663e8f 100644 --- a/net/wireless/core.h +++ b/net/wireless/core.h @@ -550,7 +550,8 @@ cfg80211_bss_update(struct cfg80211_registered_device *rdev, bool signal_valid, unsigned long ts); enum ieee80211_ap_reg_power -cfg80211_get_6ghz_power_type(const u8 *elems, size_t elems_len); +cfg80211_get_6ghz_power_type(const u8 *elems, size_t elems_len, + u32 client_flags); #ifdef CONFIG_CFG80211_DEVELOPER_WARNINGS #define CFG80211_DEV_WARN_ON(cond) WARN_ON(cond) diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index 2187e148389d..29c92bc8291b 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -6748,7 +6748,7 @@ static int nl80211_start_ap(struct sk_buff *skb, struct genl_info *info) beacon_check.relax = true; beacon_check.reg_power = cfg80211_get_6ghz_power_type(params->beacon.tail, - params->beacon.tail_len); + params->beacon.tail_len, 0); if (!cfg80211_reg_check_beaconing(&rdev->wiphy, ¶ms->chandef, &beacon_check)) { err = -EINVAL; @@ -6927,7 +6927,7 @@ static int nl80211_set_beacon(struct sk_buff *skb, struct genl_info *info) beacon_check.relax = true; beacon_check.reg_power = cfg80211_get_6ghz_power_type(params->beacon.tail, - params->beacon.tail_len); + params->beacon.tail_len, 0); if (!cfg80211_reg_check_beaconing(&rdev->wiphy, &wdev->links[link_id].ap.chandef, &beacon_check)) { diff --git a/net/wireless/scan.c b/net/wireless/scan.c index 68c4130d602f..7546647752fd 100644 --- a/net/wireless/scan.c +++ b/net/wireless/scan.c @@ -2212,7 +2212,8 @@ struct cfg80211_inform_single_bss_data { }; enum ieee80211_ap_reg_power -cfg80211_get_6ghz_power_type(const u8 *elems, size_t elems_len) +cfg80211_get_6ghz_power_type(const u8 *elems, size_t elems_len, + u32 client_flags) { const struct ieee80211_he_6ghz_oper *he_6ghz_oper; struct ieee80211_he_operation *he_oper; @@ -2230,13 +2231,13 @@ cfg80211_get_6ghz_power_type(const u8 *elems, size_t elems_len) if (!he_6ghz_oper) return IEEE80211_REG_UNSET_AP; - return cfg80211_6ghz_power_type(he_6ghz_oper->control); + return cfg80211_6ghz_power_type(he_6ghz_oper->control, client_flags); } static bool cfg80211_6ghz_power_type_valid(const u8 *elems, size_t elems_len, const u32 flags) { - switch (cfg80211_get_6ghz_power_type(elems, elems_len)) { + switch (cfg80211_get_6ghz_power_type(elems, elems_len, flags)) { case IEEE80211_REG_LPI_AP: return true; case IEEE80211_REG_SP_AP: From a5aa46f1ac4f53e03b9b75cbf55634131f2f8cac Mon Sep 17 00:00:00 2001 From: Benjamin Berg Date: Mon, 10 Nov 2025 14:18:20 +0200 Subject: [PATCH 0711/1381] wifi: mac80211: track MU-MIMO configuration on disabled interfaces For monitoring, userspace will try to configure the VIF sdata, while the driver may see the monitor_sdata that is created when only monitor interfaces are up. This causes the odd situation that it may not be possible to store the MU-MIMO configuration on monitor_sdata. Fix this by storing that information on the VIF sdata and updating the monitor_sdata when available and the interface is up. Also, adjust the code that adds monitor_sdata so that it will configure MU-MIMO based on the newly added interface or one of the existing ones. This should give a mostly consistent behaviour when configuring MU-MIMO on sniffer interfaces. Should the user configure MU-MIMO on multiple sniffer interfaces, then mac80211 will simply select one of the configurations. This behaviour should be good enough and avoids breaking user expectations in the common scenarios. Signed-off-by: Benjamin Berg Reviewed-by: Johannes Berg Signed-off-by: Miri Korenblit Link: https://patch.msgid.link/20251110141514.677915f8f6bb.If4e04a57052f9ca763562a67248b06fd80d0c2c1@changeid Signed-off-by: Johannes Berg --- net/mac80211/cfg.c | 47 ++++++++++++++++++++++++++------------ net/mac80211/ieee80211_i.h | 3 ++- net/mac80211/iface.c | 46 ++++++++++++++++++++++++++++++++----- net/mac80211/util.c | 3 ++- 4 files changed, 76 insertions(+), 23 deletions(-) diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c index c52b0456039d..b51c2c8584ae 100644 --- a/net/mac80211/cfg.c +++ b/net/mac80211/cfg.c @@ -63,12 +63,14 @@ static void ieee80211_set_mu_mimo_follow(struct ieee80211_sub_if_data *sdata, memcpy(sdata->vif.bss_conf.mu_group.position, params->vht_mumimo_groups + WLAN_MEMBERSHIP_LEN, WLAN_USER_POSITION_LEN); - ieee80211_link_info_change_notify(sdata, &sdata->deflink, - BSS_CHANGED_MU_GROUPS); + /* don't care about endianness - just check for 0 */ memcpy(&membership, params->vht_mumimo_groups, WLAN_MEMBERSHIP_LEN); mu_mimo_groups = membership != 0; + + /* Unset following if configured explicitly */ + eth_broadcast_addr(sdata->u.mntr.mu_follow_addr); } if (params->vht_mumimo_follow_addr) { @@ -76,16 +78,26 @@ static void ieee80211_set_mu_mimo_follow(struct ieee80211_sub_if_data *sdata, is_valid_ether_addr(params->vht_mumimo_follow_addr); ether_addr_copy(sdata->u.mntr.mu_follow_addr, params->vht_mumimo_follow_addr); + + /* Unset current membership until a management frame is RXed */ + memset(sdata->vif.bss_conf.mu_group.membership, 0, + WLAN_MEMBERSHIP_LEN); } sdata->vif.bss_conf.mu_mimo_owner = mu_mimo_groups || mu_mimo_follow; + + /* Notify only after setting mu_mimo_owner */ + if (sdata->vif.bss_conf.mu_mimo_owner && + sdata->flags & IEEE80211_SDATA_IN_DRIVER) + ieee80211_link_info_change_notify(sdata, &sdata->deflink, + BSS_CHANGED_MU_GROUPS); } static int ieee80211_set_mon_options(struct ieee80211_sub_if_data *sdata, struct vif_params *params) { struct ieee80211_local *local = sdata->local; - struct ieee80211_sub_if_data *monitor_sdata; + struct ieee80211_sub_if_data *monitor_sdata = NULL; /* check flags first */ if (params->flags && ieee80211_sdata_running(sdata)) { @@ -103,23 +115,28 @@ static int ieee80211_set_mon_options(struct ieee80211_sub_if_data *sdata, return -EBUSY; } - /* also validate MU-MIMO change */ - if (ieee80211_hw_check(&local->hw, NO_VIRTUAL_MONITOR)) - monitor_sdata = sdata; - else - monitor_sdata = wiphy_dereference(local->hw.wiphy, - local->monitor_sdata); - - if (!monitor_sdata && + /* validate whether MU-MIMO can be configured */ + if (!ieee80211_hw_check(&local->hw, WANT_MONITOR_VIF) && + !ieee80211_hw_check(&local->hw, NO_VIRTUAL_MONITOR) && (params->vht_mumimo_groups || params->vht_mumimo_follow_addr)) return -EOPNOTSUPP; + /* Also update dependent monitor_sdata if required */ + if (test_bit(SDATA_STATE_RUNNING, &sdata->state) && + !ieee80211_hw_check(&local->hw, NO_VIRTUAL_MONITOR)) + monitor_sdata = wiphy_dereference(local->hw.wiphy, + local->monitor_sdata); + /* apply all changes now - no failures allowed */ - if (monitor_sdata && - (ieee80211_hw_check(&local->hw, WANT_MONITOR_VIF) || - ieee80211_hw_check(&local->hw, NO_VIRTUAL_MONITOR))) - ieee80211_set_mu_mimo_follow(monitor_sdata, params); + if (ieee80211_hw_check(&local->hw, WANT_MONITOR_VIF) || + ieee80211_hw_check(&local->hw, NO_VIRTUAL_MONITOR)) { + /* This is copied in when the VIF is activated */ + ieee80211_set_mu_mimo_follow(sdata, params); + + if (monitor_sdata) + ieee80211_set_mu_mimo_follow(monitor_sdata, params); + } if (params->flags) { if (ieee80211_sdata_running(sdata)) { diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index 7f1ce9fc01c7..9d9313eee59f 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -2107,7 +2107,8 @@ void ieee80211_adjust_monitor_flags(struct ieee80211_sub_if_data *sdata, const int offset); int ieee80211_do_open(struct wireless_dev *wdev, bool coming_up); void ieee80211_sdata_stop(struct ieee80211_sub_if_data *sdata); -int ieee80211_add_virtual_monitor(struct ieee80211_local *local); +int ieee80211_add_virtual_monitor(struct ieee80211_local *local, + struct ieee80211_sub_if_data *creator_sdata); void ieee80211_del_virtual_monitor(struct ieee80211_local *local); bool __ieee80211_recalc_txpower(struct ieee80211_link_data *link); diff --git a/net/mac80211/iface.c b/net/mac80211/iface.c index a7873832d4fa..d7a4f203cde4 100644 --- a/net/mac80211/iface.c +++ b/net/mac80211/iface.c @@ -733,8 +733,9 @@ static void ieee80211_do_stop(struct ieee80211_sub_if_data *sdata, bool going_do ieee80211_configure_filter(local); ieee80211_hw_config(local, -1, hw_reconf_flags); + /* Passing NULL means an interface is picked for configuration */ if (local->virt_monitors == local->open_count) - ieee80211_add_virtual_monitor(local); + ieee80211_add_virtual_monitor(local, NULL); } void ieee80211_stop_mbssid(struct ieee80211_sub_if_data *sdata) @@ -1168,7 +1169,8 @@ static void ieee80211_sdata_init(struct ieee80211_local *local, ieee80211_link_init(sdata, -1, &sdata->deflink, &sdata->vif.bss_conf); } -int ieee80211_add_virtual_monitor(struct ieee80211_local *local) +int ieee80211_add_virtual_monitor(struct ieee80211_local *local, + struct ieee80211_sub_if_data *creator_sdata) { struct ieee80211_sub_if_data *sdata; int ret; @@ -1176,10 +1178,14 @@ int ieee80211_add_virtual_monitor(struct ieee80211_local *local) ASSERT_RTNL(); lockdep_assert_wiphy(local->hw.wiphy); - if (local->monitor_sdata || - ieee80211_hw_check(&local->hw, NO_VIRTUAL_MONITOR)) + if (ieee80211_hw_check(&local->hw, NO_VIRTUAL_MONITOR)) return 0; + /* Already have a monitor set up, configure it */ + sdata = wiphy_dereference(local->hw.wiphy, local->monitor_sdata); + if (sdata) + goto configure_monitor; + sdata = kzalloc(sizeof(*sdata) + local->hw.vif_data_size, GFP_KERNEL); if (!sdata) return -ENOMEM; @@ -1232,6 +1238,32 @@ int ieee80211_add_virtual_monitor(struct ieee80211_local *local) skb_queue_head_init(&sdata->status_queue); wiphy_work_init(&sdata->work, ieee80211_iface_work); +configure_monitor: + /* Copy in the MU-MIMO configuration if set */ + if (!creator_sdata) { + struct ieee80211_sub_if_data *other; + + list_for_each_entry(other, &local->mon_list, list) { + if (!other->vif.bss_conf.mu_mimo_owner) + continue; + + creator_sdata = other; + break; + } + } + + if (creator_sdata && creator_sdata->vif.bss_conf.mu_mimo_owner) { + sdata->vif.bss_conf.mu_mimo_owner = true; + memcpy(&sdata->vif.bss_conf.mu_group, + &creator_sdata->vif.bss_conf.mu_group, + sizeof(sdata->vif.bss_conf.mu_group)); + memcpy(&sdata->u.mntr.mu_follow_addr, + creator_sdata->u.mntr.mu_follow_addr, ETH_ALEN); + + ieee80211_link_info_change_notify(sdata, &sdata->deflink, + BSS_CHANGED_MU_GROUPS); + } + return 0; } @@ -1388,11 +1420,13 @@ int ieee80211_do_open(struct wireless_dev *wdev, bool coming_up) if (res) goto err_stop; } else { - if (local->virt_monitors == 0 && local->open_count == 0) { - res = ieee80211_add_virtual_monitor(local); + /* add/configure if there is no non-monitor interface */ + if (local->virt_monitors == local->open_count) { + res = ieee80211_add_virtual_monitor(local, sdata); if (res) goto err_stop; } + local->virt_monitors++; /* must be before the call to ieee80211_configure_filter */ diff --git a/net/mac80211/util.c b/net/mac80211/util.c index 1a128f7aae82..0c46009a3d63 100644 --- a/net/mac80211/util.c +++ b/net/mac80211/util.c @@ -2206,9 +2206,10 @@ int ieee80211_reconfig(struct ieee80211_local *local) } } + /* Passing NULL means an interface is picked for configuration */ if (local->virt_monitors > 0 && local->virt_monitors == local->open_count) - ieee80211_add_virtual_monitor(local); + ieee80211_add_virtual_monitor(local, NULL); if (!suspended) return 0; From 9e23063a79c4e7e56199d1a0642f63830dffa9ea Mon Sep 17 00:00:00 2001 From: Benjamin Berg Date: Mon, 10 Nov 2025 14:20:20 +0200 Subject: [PATCH 0712/1381] wifi: mac80211: make monitor link info check more specific Verify that only one of the permitted change flags is set when changing the link of a monitor interface. Before the WARN_ON_ONCE would accept anything if mu_mimo_owner was set. Also, split out the mu_mimo_owner flag and enable it for all interface types. The option is set during association when VHT is available and it is not expected that any configuration of the MU groups is done without it being set. Signed-off-by: Benjamin Berg Reviewed-by: Johannes Berg Signed-off-by: Miri Korenblit Link: https://patch.msgid.link/20251110141948.6696dba8678d.Icafac3be4724825dd6140e4407bae3a2adb593a5@changeid Signed-off-by: Johannes Berg --- net/mac80211/driver-ops.c | 8 ++++++-- net/mac80211/main.c | 3 +-- 2 files changed, 7 insertions(+), 4 deletions(-) diff --git a/net/mac80211/driver-ops.c b/net/mac80211/driver-ops.c index ba9fba165926..49753b73aba2 100644 --- a/net/mac80211/driver-ops.c +++ b/net/mac80211/driver-ops.c @@ -476,8 +476,12 @@ void drv_link_info_changed(struct ieee80211_local *local, if (WARN_ON_ONCE(sdata->vif.type == NL80211_IFTYPE_P2P_DEVICE || sdata->vif.type == NL80211_IFTYPE_NAN || (sdata->vif.type == NL80211_IFTYPE_MONITOR && - !sdata->vif.bss_conf.mu_mimo_owner && - !(changed & BSS_CHANGED_TXPOWER)))) + changed & ~(BSS_CHANGED_TXPOWER | + BSS_CHANGED_MU_GROUPS)))) + return; + + if (WARN_ON_ONCE(changed & BSS_CHANGED_MU_GROUPS && + !sdata->vif.bss_conf.mu_mimo_owner)) return; if (!check_sdata_in_driver(sdata)) diff --git a/net/mac80211/main.c b/net/mac80211/main.c index eefa6f7e899b..b05e313c7f17 100644 --- a/net/mac80211/main.c +++ b/net/mac80211/main.c @@ -356,8 +356,7 @@ void ieee80211_bss_info_change_notify(struct ieee80211_sub_if_data *sdata, if (WARN_ON_ONCE(sdata->vif.type == NL80211_IFTYPE_P2P_DEVICE || sdata->vif.type == NL80211_IFTYPE_NAN || (sdata->vif.type == NL80211_IFTYPE_MONITOR && - !sdata->vif.bss_conf.mu_mimo_owner && - !(changed & BSS_CHANGED_TXPOWER)))) + changed & ~BSS_CHANGED_TXPOWER))) return; if (!check_sdata_in_driver(sdata)) From 5e88e864118c20e63a1571d0ff0a152e5d684959 Mon Sep 17 00:00:00 2001 From: Abdun Nihaal Date: Mon, 10 Nov 2025 23:23:15 +0530 Subject: [PATCH 0713/1381] wifi: cw1200: Fix potential memory leak in cw1200_bh_rx_helper() In one of the error paths, the memory allocated for skb_rx is not freed. Fix that by freeing it before returning. Fixes: a910e4a94f69 ("cw1200: add driver for the ST-E CW1100 & CW1200 WLAN chipsets") Signed-off-by: Abdun Nihaal Link: https://patch.msgid.link/20251110175316.106591-1-nihaal@cse.iitm.ac.in Signed-off-by: Johannes Berg --- drivers/net/wireless/st/cw1200/bh.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/net/wireless/st/cw1200/bh.c b/drivers/net/wireless/st/cw1200/bh.c index 3b4ded2ac801..37232ee22037 100644 --- a/drivers/net/wireless/st/cw1200/bh.c +++ b/drivers/net/wireless/st/cw1200/bh.c @@ -317,10 +317,12 @@ static int cw1200_bh_rx_helper(struct cw1200_common *priv, if (wsm_id & 0x0400) { int rc = wsm_release_tx_buffer(priv, 1); - if (WARN_ON(rc < 0)) + if (WARN_ON(rc < 0)) { + dev_kfree_skb(skb_rx); return rc; - else if (rc > 0) + } else if (rc > 0) { *tx = 1; + } } /* cw1200_wsm_rx takes care on SKB livetime */ From b02d229013aad864d89c2cbe8c713ff7da04c253 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 7 Nov 2025 08:22:24 -0800 Subject: [PATCH 0714/1381] tools: ynltool: create skeleton for the C command Based on past discussions it seems like integration of YNL into iproute2 is unlikely. YNL itself is not great as a C library, since it has no backward compat (we routinely change types). Most of the operations can be performed with the generic Python CLI directly. There is, however, a handful of operations where summarization of kernel output is very useful (mostly related to stats: page-pool, qstat). Create a command (inspired by bpftool, I think it stood the test of time reasonably well) to be able to plug the subcommands into. Link: https://lore.kernel.org/1754895902-8790-1-git-send-email-ernis@linux.microsoft.com Signed-off-by: Jakub Kicinski Link: https://patch.msgid.link/20251107162227.980672-2-kuba@kernel.org Acked-by: Stanislav Fomichev Signed-off-by: Paolo Abeni --- tools/net/ynl/Makefile | 3 +- tools/net/ynl/ynltool/.gitignore | 1 + tools/net/ynl/ynltool/Makefile | 52 +++++ tools/net/ynl/ynltool/json_writer.c | 288 ++++++++++++++++++++++++++++ tools/net/ynl/ynltool/json_writer.h | 75 ++++++++ tools/net/ynl/ynltool/main.c | 240 +++++++++++++++++++++++ tools/net/ynl/ynltool/main.h | 62 ++++++ 7 files changed, 720 insertions(+), 1 deletion(-) create mode 100644 tools/net/ynl/ynltool/.gitignore create mode 100644 tools/net/ynl/ynltool/Makefile create mode 100644 tools/net/ynl/ynltool/json_writer.c create mode 100644 tools/net/ynl/ynltool/json_writer.h create mode 100644 tools/net/ynl/ynltool/main.c create mode 100644 tools/net/ynl/ynltool/main.h diff --git a/tools/net/ynl/Makefile b/tools/net/ynl/Makefile index 211df5a93ad9..31ed20c0f3f8 100644 --- a/tools/net/ynl/Makefile +++ b/tools/net/ynl/Makefile @@ -12,10 +12,11 @@ endif libdir ?= $(prefix)/$(libdir_relative) includedir ?= $(prefix)/include -SUBDIRS = lib generated samples +SUBDIRS = lib generated samples ynltool all: $(SUBDIRS) libynl.a +ynltool: | lib generated libynl.a samples: | lib generated libynl.a: | lib generated @echo -e "\tAR $@" diff --git a/tools/net/ynl/ynltool/.gitignore b/tools/net/ynl/ynltool/.gitignore new file mode 100644 index 000000000000..f38848dbb0d3 --- /dev/null +++ b/tools/net/ynl/ynltool/.gitignore @@ -0,0 +1 @@ +ynltool diff --git a/tools/net/ynl/ynltool/Makefile b/tools/net/ynl/ynltool/Makefile new file mode 100644 index 000000000000..cfabab3a20da --- /dev/null +++ b/tools/net/ynl/ynltool/Makefile @@ -0,0 +1,52 @@ +# SPDX-License-Identifier: GPL-2.0-only + +include ../Makefile.deps + +INSTALL ?= install +prefix ?= /usr + +CC := gcc +CFLAGS := -Wall -Wextra -Werror -O2 +ifeq ("$(DEBUG)","1") + CFLAGS += -g -fsanitize=address -fsanitize=leak -static-libasan +endif +CFLAGS += -I../lib + +SRC_VERSION := \ + $(shell make --no-print-directory -sC ../../../.. kernelversion || \ + echo "unknown") + +CFLAGS += -DSRC_VERSION='"$(SRC_VERSION)"' + +SRCS := $(wildcard *.c) +OBJS := $(patsubst %.c,$(OUTPUT)%.o,$(SRCS)) + +YNLTOOL := $(OUTPUT)ynltool + +include $(wildcard *.d) + +all: $(YNLTOOL) + +Q = @ + +$(YNLTOOL): $(OBJS) + $(Q)echo -e "\tLINK $@" + $(Q)$(CC) $(CFLAGS) -o $@ $(OBJS) + +%.o: %.c main.h json_writer.h + $(Q)echo -e "\tCC $@" + $(Q)$(COMPILE.c) -MMD -c -o $@ $< + +clean: + rm -f *.o *.d *~ + +distclean: clean + rm -f $(YNLTOOL) + +bindir ?= /usr/bin + +install: $(YNLTOOL) + install -m 0755 $(YNLTOOL) $(DESTDIR)$(bindir)/$(YNLTOOL) + +.PHONY: all clean distclean +.DEFAULT_GOAL=all diff --git a/tools/net/ynl/ynltool/json_writer.c b/tools/net/ynl/ynltool/json_writer.c new file mode 100644 index 000000000000..c8685e592cd3 --- /dev/null +++ b/tools/net/ynl/ynltool/json_writer.c @@ -0,0 +1,288 @@ +// SPDX-License-Identifier: (GPL-2.0-or-later OR BSD-2-Clause) +/* + * Simple streaming JSON writer + * + * This takes care of the annoying bits of JSON syntax like the commas + * after elements + * + * Authors: Stephen Hemminger + */ + +#include +#include +#include +#include +#include +#include +#include + +#include "json_writer.h" + +struct json_writer { + FILE *out; + unsigned depth; + bool pretty; + char sep; +}; + +static void jsonw_indent(json_writer_t *self) +{ + unsigned i; + for (i = 0; i < self->depth; ++i) + fputs(" ", self->out); +} + +static void jsonw_eol(json_writer_t *self) +{ + if (!self->pretty) + return; + + putc('\n', self->out); + jsonw_indent(self); +} + +static void jsonw_eor(json_writer_t *self) +{ + if (self->sep != '\0') + putc(self->sep, self->out); + self->sep = ','; +} + +static void jsonw_puts(json_writer_t *self, const char *str) +{ + putc('"', self->out); + for (; *str; ++str) + switch (*str) { + case '\t': + fputs("\\t", self->out); + break; + case '\n': + fputs("\\n", self->out); + break; + case '\r': + fputs("\\r", self->out); + break; + case '\f': + fputs("\\f", self->out); + break; + case '\b': + fputs("\\b", self->out); + break; + case '\\': + fputs("\\\\", self->out); + break; + case '"': + fputs("\\\"", self->out); + break; + default: + putc(*str, self->out); + } + putc('"', self->out); +} + +json_writer_t *jsonw_new(FILE *f) +{ + json_writer_t *self = malloc(sizeof(*self)); + if (self) { + self->out = f; + self->depth = 0; + self->pretty = false; + self->sep = '\0'; + } + return self; +} + +void jsonw_destroy(json_writer_t **self_p) +{ + json_writer_t *self = *self_p; + + assert(self->depth == 0); + fputs("\n", self->out); + fflush(self->out); + free(self); + *self_p = NULL; +} + +void jsonw_pretty(json_writer_t *self, bool on) +{ + self->pretty = on; +} + +void jsonw_reset(json_writer_t *self) +{ + assert(self->depth == 0); + self->sep = '\0'; +} + +static void jsonw_begin(json_writer_t *self, int c) +{ + jsonw_eor(self); + putc(c, self->out); + ++self->depth; + self->sep = '\0'; +} + +static void jsonw_end(json_writer_t *self, int c) +{ + assert(self->depth > 0); + + --self->depth; + if (self->sep != '\0') + jsonw_eol(self); + putc(c, self->out); + self->sep = ','; +} + +void jsonw_name(json_writer_t *self, const char *name) +{ + jsonw_eor(self); + jsonw_eol(self); + self->sep = '\0'; + jsonw_puts(self, name); + putc(':', self->out); + if (self->pretty) + putc(' ', self->out); +} + +void jsonw_vprintf_enquote(json_writer_t *self, const char *fmt, va_list ap) +{ + jsonw_eor(self); + putc('"', self->out); + vfprintf(self->out, fmt, ap); + putc('"', self->out); +} + +void jsonw_printf(json_writer_t *self, const char *fmt, ...) +{ + va_list ap; + + va_start(ap, fmt); + jsonw_eor(self); + vfprintf(self->out, fmt, ap); + va_end(ap); +} + +void jsonw_start_object(json_writer_t *self) +{ + jsonw_begin(self, '{'); +} + +void jsonw_end_object(json_writer_t *self) +{ + jsonw_end(self, '}'); +} + +void jsonw_start_array(json_writer_t *self) +{ + jsonw_begin(self, '['); +} + +void jsonw_end_array(json_writer_t *self) +{ + jsonw_end(self, ']'); +} + +void jsonw_string(json_writer_t *self, const char *value) +{ + jsonw_eor(self); + jsonw_puts(self, value); +} + +void jsonw_bool(json_writer_t *self, bool val) +{ + jsonw_printf(self, "%s", val ? "true" : "false"); +} + +void jsonw_null(json_writer_t *self) +{ + jsonw_printf(self, "null"); +} + +void jsonw_float_fmt(json_writer_t *self, const char *fmt, double num) +{ + jsonw_printf(self, fmt, num); +} + +void jsonw_float(json_writer_t *self, double num) +{ + jsonw_printf(self, "%g", num); +} + +void jsonw_hu(json_writer_t *self, unsigned short num) +{ + jsonw_printf(self, "%hu", num); +} + +void jsonw_uint(json_writer_t *self, uint64_t num) +{ + jsonw_printf(self, "%"PRIu64, num); +} + +void jsonw_lluint(json_writer_t *self, unsigned long long int num) +{ + jsonw_printf(self, "%llu", num); +} + +void jsonw_int(json_writer_t *self, int64_t num) +{ + jsonw_printf(self, "%"PRId64, num); +} + +void jsonw_string_field(json_writer_t *self, const char *prop, const char *val) +{ + jsonw_name(self, prop); + jsonw_string(self, val); +} + +void jsonw_bool_field(json_writer_t *self, const char *prop, bool val) +{ + jsonw_name(self, prop); + jsonw_bool(self, val); +} + +void jsonw_float_field(json_writer_t *self, const char *prop, double val) +{ + jsonw_name(self, prop); + jsonw_float(self, val); +} + +void jsonw_float_field_fmt(json_writer_t *self, + const char *prop, + const char *fmt, + double val) +{ + jsonw_name(self, prop); + jsonw_float_fmt(self, fmt, val); +} + +void jsonw_uint_field(json_writer_t *self, const char *prop, uint64_t num) +{ + jsonw_name(self, prop); + jsonw_uint(self, num); +} + +void jsonw_hu_field(json_writer_t *self, const char *prop, unsigned short num) +{ + jsonw_name(self, prop); + jsonw_hu(self, num); +} + +void jsonw_lluint_field(json_writer_t *self, + const char *prop, + unsigned long long int num) +{ + jsonw_name(self, prop); + jsonw_lluint(self, num); +} + +void jsonw_int_field(json_writer_t *self, const char *prop, int64_t num) +{ + jsonw_name(self, prop); + jsonw_int(self, num); +} + +void jsonw_null_field(json_writer_t *self, const char *prop) +{ + jsonw_name(self, prop); + jsonw_null(self); +} diff --git a/tools/net/ynl/ynltool/json_writer.h b/tools/net/ynl/ynltool/json_writer.h new file mode 100644 index 000000000000..0f1e63c88f6a --- /dev/null +++ b/tools/net/ynl/ynltool/json_writer.h @@ -0,0 +1,75 @@ +/* SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) */ +/* + * Simple streaming JSON writer + * + * This takes care of the annoying bits of JSON syntax like the commas + * after elements + * + * Authors: Stephen Hemminger + */ + +#ifndef _JSON_WRITER_H_ +#define _JSON_WRITER_H_ + +#include +#include +#include +#include + +/* Opaque class structure */ +typedef struct json_writer json_writer_t; + +/* Create a new JSON stream */ +json_writer_t *jsonw_new(FILE *f); +/* End output to JSON stream */ +void jsonw_destroy(json_writer_t **self_p); + +/* Cause output to have pretty whitespace */ +void jsonw_pretty(json_writer_t *self, bool on); + +/* Reset separator to create new JSON */ +void jsonw_reset(json_writer_t *self); + +/* Add property name */ +void jsonw_name(json_writer_t *self, const char *name); + +/* Add value */ +void __attribute__((format(printf, 2, 0))) jsonw_vprintf_enquote(json_writer_t *self, + const char *fmt, + va_list ap); +void __attribute__((format(printf, 2, 3))) jsonw_printf(json_writer_t *self, + const char *fmt, ...); +void jsonw_string(json_writer_t *self, const char *value); +void jsonw_bool(json_writer_t *self, bool value); +void jsonw_float(json_writer_t *self, double number); +void jsonw_float_fmt(json_writer_t *self, const char *fmt, double num); +void jsonw_uint(json_writer_t *self, uint64_t number); +void jsonw_hu(json_writer_t *self, unsigned short number); +void jsonw_int(json_writer_t *self, int64_t number); +void jsonw_null(json_writer_t *self); +void jsonw_lluint(json_writer_t *self, unsigned long long int num); + +/* Useful Combinations of name and value */ +void jsonw_string_field(json_writer_t *self, const char *prop, const char *val); +void jsonw_bool_field(json_writer_t *self, const char *prop, bool value); +void jsonw_float_field(json_writer_t *self, const char *prop, double num); +void jsonw_uint_field(json_writer_t *self, const char *prop, uint64_t num); +void jsonw_hu_field(json_writer_t *self, const char *prop, unsigned short num); +void jsonw_int_field(json_writer_t *self, const char *prop, int64_t num); +void jsonw_null_field(json_writer_t *self, const char *prop); +void jsonw_lluint_field(json_writer_t *self, const char *prop, + unsigned long long int num); +void jsonw_float_field_fmt(json_writer_t *self, const char *prop, + const char *fmt, double val); + +/* Collections */ +void jsonw_start_object(json_writer_t *self); +void jsonw_end_object(json_writer_t *self); + +void jsonw_start_array(json_writer_t *self); +void jsonw_end_array(json_writer_t *self); + +/* Override default exception handling */ +typedef void (jsonw_err_handler_fn)(const char *); + +#endif /* _JSON_WRITER_H_ */ diff --git a/tools/net/ynl/ynltool/main.c b/tools/net/ynl/ynltool/main.c new file mode 100644 index 000000000000..8e15e4ee543f --- /dev/null +++ b/tools/net/ynl/ynltool/main.c @@ -0,0 +1,240 @@ +// SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) +/* Copyright (C) 2017-2018 Netronome Systems, Inc. */ +/* Copyright Meta Platforms, Inc. and affiliates */ + +#include +#include +#include +#include +#include +#include +#include + +#include "main.h" + +const char *bin_name; +static int last_argc; +static char **last_argv; +static int (*last_do_help)(int argc, char **argv); +json_writer_t *json_wtr; +bool pretty_output; +bool json_output; + +static void __attribute__((noreturn)) clean_and_exit(int i) +{ + if (json_output) + jsonw_destroy(&json_wtr); + + exit(i); +} + +void usage(void) +{ + last_do_help(last_argc - 1, last_argv + 1); + + clean_and_exit(-1); +} + +static int do_help(int argc __attribute__((unused)), + char **argv __attribute__((unused))) +{ + if (json_output) { + jsonw_null(json_wtr); + return 0; + } + + fprintf(stderr, + "Usage: %s [OPTIONS] OBJECT { COMMAND | help }\n" + " %s version\n" + "\n" + " OBJECT := { }\n" + " " HELP_SPEC_OPTIONS "\n" + "", + bin_name, bin_name); + + return 0; +} + +static int do_version(int argc __attribute__((unused)), + char **argv __attribute__((unused))) +{ + if (json_output) { + jsonw_start_object(json_wtr); + jsonw_name(json_wtr, "version"); + jsonw_printf(json_wtr, SRC_VERSION); + jsonw_end_object(json_wtr); + } else { + printf("%s " SRC_VERSION "\n", bin_name); + } + return 0; +} + +static const struct cmd commands[] = { + { "help", do_help }, + { "version", do_version }, + { 0 } +}; + +int cmd_select(const struct cmd *cmds, int argc, char **argv, + int (*help)(int argc, char **argv)) +{ + unsigned int i; + + last_argc = argc; + last_argv = argv; + last_do_help = help; + + if (argc < 1 && cmds[0].func) + return cmds[0].func(argc, argv); + + for (i = 0; cmds[i].cmd; i++) { + if (is_prefix(*argv, cmds[i].cmd)) { + if (!cmds[i].func) { + p_err("command '%s' is not available", cmds[i].cmd); + return -1; + } + return cmds[i].func(argc - 1, argv + 1); + } + } + + help(argc - 1, argv + 1); + + return -1; +} + +bool is_prefix(const char *pfx, const char *str) +{ + if (!pfx) + return false; + if (strlen(str) < strlen(pfx)) + return false; + + return !memcmp(str, pfx, strlen(pfx)); +} + +/* Last argument MUST be NULL pointer */ +int detect_common_prefix(const char *arg, ...) +{ + unsigned int count = 0; + const char *ref; + char msg[256]; + va_list ap; + + snprintf(msg, sizeof(msg), "ambiguous prefix: '%s' could be '", arg); + va_start(ap, arg); + while ((ref = va_arg(ap, const char *))) { + if (!is_prefix(arg, ref)) + continue; + count++; + if (count > 1) + strncat(msg, "' or '", sizeof(msg) - strlen(msg) - 1); + strncat(msg, ref, sizeof(msg) - strlen(msg) - 1); + } + va_end(ap); + strncat(msg, "'", sizeof(msg) - strlen(msg) - 1); + + if (count >= 2) { + p_err("%s", msg); + return -1; + } + + return 0; +} + +void p_err(const char *fmt, ...) +{ + va_list ap; + + va_start(ap, fmt); + if (json_output) { + jsonw_start_object(json_wtr); + jsonw_name(json_wtr, "error"); + jsonw_vprintf_enquote(json_wtr, fmt, ap); + jsonw_end_object(json_wtr); + } else { + fprintf(stderr, "Error: "); + vfprintf(stderr, fmt, ap); + fprintf(stderr, "\n"); + } + va_end(ap); +} + +void p_info(const char *fmt, ...) +{ + va_list ap; + + if (json_output) + return; + + va_start(ap, fmt); + vfprintf(stderr, fmt, ap); + fprintf(stderr, "\n"); + va_end(ap); +} + +int main(int argc, char **argv) +{ + static const struct option options[] = { + { "json", no_argument, NULL, 'j' }, + { "help", no_argument, NULL, 'h' }, + { "pretty", no_argument, NULL, 'p' }, + { "version", no_argument, NULL, 'V' }, + { 0 } + }; + bool version_requested = false; + int opt, ret; + + setlinebuf(stdout); + + last_do_help = do_help; + pretty_output = false; + json_output = false; + bin_name = "ynltool"; + + opterr = 0; + while ((opt = getopt_long(argc, argv, "Vhjp", + options, NULL)) >= 0) { + switch (opt) { + case 'V': + version_requested = true; + break; + case 'h': + return do_help(argc, argv); + case 'p': + pretty_output = true; + /* fall through */ + case 'j': + if (!json_output) { + json_wtr = jsonw_new(stdout); + if (!json_wtr) { + p_err("failed to create JSON writer"); + return -1; + } + json_output = true; + } + jsonw_pretty(json_wtr, pretty_output); + break; + default: + p_err("unrecognized option '%s'", argv[optind - 1]); + if (json_output) + clean_and_exit(-1); + else + usage(); + } + } + + argc -= optind; + argv += optind; + if (argc < 0) + usage(); + + if (version_requested) + ret = do_version(argc, argv); + else + ret = cmd_select(commands, argc, argv, do_help); + + if (json_output) + jsonw_destroy(&json_wtr); + + return ret; +} diff --git a/tools/net/ynl/ynltool/main.h b/tools/net/ynl/ynltool/main.h new file mode 100644 index 000000000000..f4a70acf2085 --- /dev/null +++ b/tools/net/ynl/ynltool/main.h @@ -0,0 +1,62 @@ +/* SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) */ +/* Copyright (C) 2017-2018 Netronome Systems, Inc. */ +/* Copyright Meta Platforms, Inc. and affiliates */ + +#ifndef __YNLTOOL_H +#define __YNLTOOL_H + +#ifndef _GNU_SOURCE +#define _GNU_SOURCE +#endif +#include +#include +#include +#include +#include + +#include "json_writer.h" + +#define NEXT_ARG() ({ argc--; argv++; if (argc < 0) usage(); }) +#define NEXT_ARGP() ({ (*argc)--; (*argv)++; if (*argc < 0) usage(); }) +#define BAD_ARG() ({ p_err("what is '%s'?", *argv); -1; }) +#define GET_ARG() ({ argc--; *argv++; }) +#define REQ_ARGS(cnt) \ + ({ \ + int _cnt = (cnt); \ + bool _res; \ + \ + if (argc < _cnt) { \ + p_err("'%s' needs at least %d arguments, %d found", \ + argv[-1], _cnt, argc); \ + _res = false; \ + } else { \ + _res = true; \ + } \ + _res; \ + }) + +#define HELP_SPEC_OPTIONS \ + "OPTIONS := { {-j|--json} [{-p|--pretty}] }" + +extern const char *bin_name; + +extern json_writer_t *json_wtr; +extern bool json_output; +extern bool pretty_output; + +void __attribute__((format(printf, 1, 2))) p_err(const char *fmt, ...); +void __attribute__((format(printf, 1, 2))) p_info(const char *fmt, ...); + +bool is_prefix(const char *pfx, const char *str); +int detect_common_prefix(const char *arg, ...); +void usage(void) __attribute__((noreturn)); + +struct cmd { + const char *cmd; + int (*func)(int argc, char **argv); +}; + +int cmd_select(const struct cmd *cmds, int argc, char **argv, + int (*help)(int argc, char **argv)); + +#endif /* __YNLTOOL_H */ From 124dac9b421ca8e69ed11ad7a0fc1794d03c2519 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 7 Nov 2025 08:22:25 -0800 Subject: [PATCH 0715/1381] tools: ynltool: add page-pool stats Replace the page-pool sample with page pool support in ynltool. # ynltool page-pool stats eth0[2] page pools: 18 (zombies: 0) refs: 171456 bytes: 702283776 (refs: 0 bytes: 0) recycling: 97.3% (alloc: 2679:6134966 recycle: 1250981:4719386) # ynltool -j page-pool stats | jq [ { "ifname": "eth0", "ifindex": 2, "page_pools": 18, "zombies": 0, "live": { "refs": 171456, "bytes": 702283776 }, "zombie": { "refs": 0, "bytes": 0 }, "recycling_pct": 97.2746, "alloc": { "slow": 2679, "fast": 6135029 }, "recycle": { "ring": 1250997, "cache": 4719432 } } ] # ynltool page-pool stats group-by pp pool id: 108 dev: eth0[2] napi: 530 inflight: 9472 pages 38797312 bytes recycling: 95.5% (alloc: 148:208379 recycle: 45386:153842) pool id: 107 dev: eth0[2] napi: 529 inflight: 9408 pages 38535168 bytes recycling: 94.9% (alloc: 147:180178 recycle: 42251:128808) Signed-off-by: Jakub Kicinski Link: https://patch.msgid.link/20251107162227.980672-3-kuba@kernel.org Acked-by: Stanislav Fomichev Signed-off-by: Paolo Abeni --- tools/net/ynl/samples/page-pool.c | 149 ---------- tools/net/ynl/ynltool/Makefile | 11 +- tools/net/ynl/ynltool/main.c | 3 +- tools/net/ynl/ynltool/main.h | 3 + tools/net/ynl/ynltool/page-pool.c | 461 ++++++++++++++++++++++++++++++ 5 files changed, 473 insertions(+), 154 deletions(-) delete mode 100644 tools/net/ynl/samples/page-pool.c create mode 100644 tools/net/ynl/ynltool/page-pool.c diff --git a/tools/net/ynl/samples/page-pool.c b/tools/net/ynl/samples/page-pool.c deleted file mode 100644 index e5d521320fbf..000000000000 --- a/tools/net/ynl/samples/page-pool.c +++ /dev/null @@ -1,149 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0 -#define _GNU_SOURCE - -#include -#include - -#include - -#include - -#include "netdev-user.h" - -struct stat { - unsigned int ifc; - - struct { - unsigned int cnt; - size_t refs, bytes; - } live[2]; - - size_t alloc_slow, alloc_fast, recycle_ring, recycle_cache; -}; - -struct stats_array { - unsigned int i, max; - struct stat *s; -}; - -static struct stat *find_ifc(struct stats_array *a, unsigned int ifindex) -{ - unsigned int i; - - for (i = 0; i < a->i; i++) { - if (a->s[i].ifc == ifindex) - return &a->s[i]; - } - - a->i++; - if (a->i == a->max) { - a->max *= 2; - a->s = reallocarray(a->s, a->max, sizeof(*a->s)); - } - a->s[i].ifc = ifindex; - return &a->s[i]; -} - -static void count(struct stat *s, unsigned int l, - struct netdev_page_pool_get_rsp *pp) -{ - s->live[l].cnt++; - if (pp->_present.inflight) - s->live[l].refs += pp->inflight; - if (pp->_present.inflight_mem) - s->live[l].bytes += pp->inflight_mem; -} - -int main(int argc, char **argv) -{ - struct netdev_page_pool_stats_get_list *pp_stats; - struct netdev_page_pool_get_list *pools; - struct stats_array a = {}; - struct ynl_error yerr; - struct ynl_sock *ys; - - ys = ynl_sock_create(&ynl_netdev_family, &yerr); - if (!ys) { - fprintf(stderr, "YNL: %s\n", yerr.msg); - return 1; - } - - a.max = 128; - a.s = calloc(a.max, sizeof(*a.s)); - if (!a.s) - goto err_close; - - pools = netdev_page_pool_get_dump(ys); - if (!pools) - goto err_free; - - ynl_dump_foreach(pools, pp) { - struct stat *s = find_ifc(&a, pp->ifindex); - - count(s, 1, pp); - if (pp->_present.detach_time) - count(s, 0, pp); - } - netdev_page_pool_get_list_free(pools); - - pp_stats = netdev_page_pool_stats_get_dump(ys); - if (!pp_stats) - goto err_free; - - ynl_dump_foreach(pp_stats, pp) { - struct stat *s = find_ifc(&a, pp->info.ifindex); - - if (pp->_present.alloc_fast) - s->alloc_fast += pp->alloc_fast; - if (pp->_present.alloc_refill) - s->alloc_fast += pp->alloc_refill; - if (pp->_present.alloc_slow) - s->alloc_slow += pp->alloc_slow; - if (pp->_present.recycle_ring) - s->recycle_ring += pp->recycle_ring; - if (pp->_present.recycle_cached) - s->recycle_cache += pp->recycle_cached; - } - netdev_page_pool_stats_get_list_free(pp_stats); - - for (unsigned int i = 0; i < a.i; i++) { - char ifname[IF_NAMESIZE]; - struct stat *s = &a.s[i]; - const char *name; - double recycle; - - if (!s->ifc) { - name = "\t"; - } else { - name = if_indextoname(s->ifc, ifname); - if (name) - printf("%8s", name); - printf("[%u]\t", s->ifc); - } - - printf("page pools: %u (zombies: %u)\n", - s->live[1].cnt, s->live[0].cnt); - printf("\t\trefs: %zu bytes: %zu (refs: %zu bytes: %zu)\n", - s->live[1].refs, s->live[1].bytes, - s->live[0].refs, s->live[0].bytes); - - /* We don't know how many pages are sitting in cache and ring - * so we will under-count the recycling rate a bit. - */ - recycle = (double)(s->recycle_ring + s->recycle_cache) / - (s->alloc_fast + s->alloc_slow) * 100; - printf("\t\trecycling: %.1lf%% (alloc: %zu:%zu recycle: %zu:%zu)\n", - recycle, s->alloc_slow, s->alloc_fast, - s->recycle_ring, s->recycle_cache); - } - - ynl_sock_destroy(ys); - return 0; - -err_free: - free(a.s); -err_close: - fprintf(stderr, "YNL: %s\n", ys->err.msg); - ynl_sock_destroy(ys); - return 2; -} diff --git a/tools/net/ynl/ynltool/Makefile b/tools/net/ynl/ynltool/Makefile index cfabab3a20da..11240740ed81 100644 --- a/tools/net/ynl/ynltool/Makefile +++ b/tools/net/ynl/ynltool/Makefile @@ -10,7 +10,7 @@ CFLAGS := -Wall -Wextra -Werror -O2 ifeq ("$(DEBUG)","1") CFLAGS += -g -fsanitize=address -fsanitize=leak -static-libasan endif -CFLAGS += -I../lib +CFLAGS += -I../lib -I../generated -I../../../include/uapi/ SRC_VERSION := \ $(shell make --no-print-directory -sC ../../../.. kernelversion || \ @@ -29,14 +29,17 @@ all: $(YNLTOOL) Q = @ -$(YNLTOOL): $(OBJS) +$(YNLTOOL): ../libynl.a $(OBJS) $(Q)echo -e "\tLINK $@" - $(Q)$(CC) $(CFLAGS) -o $@ $(OBJS) + $(Q)$(CC) $(CFLAGS) -o $@ $(OBJS) ../libynl.a -lmnl -%.o: %.c main.h json_writer.h +%.o: %.c ../libynl.a $(Q)echo -e "\tCC $@" $(Q)$(COMPILE.c) -MMD -c -o $@ $< +../libynl.a: + $(Q)$(MAKE) -C ../ + clean: rm -f *.o *.d *~ diff --git a/tools/net/ynl/ynltool/main.c b/tools/net/ynl/ynltool/main.c index 8e15e4ee543f..f83c6f3245c8 100644 --- a/tools/net/ynl/ynltool/main.c +++ b/tools/net/ynl/ynltool/main.c @@ -47,7 +47,7 @@ static int do_help(int argc __attribute__((unused)), "Usage: %s [OPTIONS] OBJECT { COMMAND | help }\n" " %s version\n" "\n" - " OBJECT := { }\n" + " OBJECT := { page-pool }\n" " " HELP_SPEC_OPTIONS "\n" "", bin_name, bin_name); @@ -71,6 +71,7 @@ static int do_version(int argc __attribute__((unused)), static const struct cmd commands[] = { { "help", do_help }, + { "page-pool", do_page_pool }, { "version", do_version }, { 0 } }; diff --git a/tools/net/ynl/ynltool/main.h b/tools/net/ynl/ynltool/main.h index f4a70acf2085..fd05d21451a2 100644 --- a/tools/net/ynl/ynltool/main.h +++ b/tools/net/ynl/ynltool/main.h @@ -59,4 +59,7 @@ struct cmd { int cmd_select(const struct cmd *cmds, int argc, char **argv, int (*help)(int argc, char **argv)); +/* subcommands */ +int do_page_pool(int argc, char **argv); + #endif /* __YNLTOOL_H */ diff --git a/tools/net/ynl/ynltool/page-pool.c b/tools/net/ynl/ynltool/page-pool.c new file mode 100644 index 000000000000..4b24492abab7 --- /dev/null +++ b/tools/net/ynl/ynltool/page-pool.c @@ -0,0 +1,461 @@ +// SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) + +#include +#include +#include +#include +#include + +#include +#include "netdev-user.h" + +#include "main.h" + +struct pp_stat { + unsigned int ifc; + + struct { + unsigned int cnt; + size_t refs, bytes; + } live[2]; + + size_t alloc_slow, alloc_fast, recycle_ring, recycle_cache; +}; + +struct pp_stats_array { + unsigned int i, max; + struct pp_stat *s; +}; + +static struct pp_stat *find_ifc(struct pp_stats_array *a, unsigned int ifindex) +{ + unsigned int i; + + for (i = 0; i < a->i; i++) { + if (a->s[i].ifc == ifindex) + return &a->s[i]; + } + + a->i++; + if (a->i == a->max) { + a->max *= 2; + a->s = reallocarray(a->s, a->max, sizeof(*a->s)); + } + a->s[i].ifc = ifindex; + return &a->s[i]; +} + +static void count_pool(struct pp_stat *s, unsigned int l, + struct netdev_page_pool_get_rsp *pp) +{ + s->live[l].cnt++; + if (pp->_present.inflight) + s->live[l].refs += pp->inflight; + if (pp->_present.inflight_mem) + s->live[l].bytes += pp->inflight_mem; +} + +/* We don't know how many pages are sitting in cache and ring + * so we will under-count the recycling rate a bit. + */ +static void print_json_recycling_stats(struct pp_stat *s) +{ + double recycle; + + if (s->alloc_fast + s->alloc_slow) { + recycle = (double)(s->recycle_ring + s->recycle_cache) / + (s->alloc_fast + s->alloc_slow) * 100; + jsonw_float_field(json_wtr, "recycling_pct", recycle); + } + + jsonw_name(json_wtr, "alloc"); + jsonw_start_object(json_wtr); + jsonw_uint_field(json_wtr, "slow", s->alloc_slow); + jsonw_uint_field(json_wtr, "fast", s->alloc_fast); + jsonw_end_object(json_wtr); + + jsonw_name(json_wtr, "recycle"); + jsonw_start_object(json_wtr); + jsonw_uint_field(json_wtr, "ring", s->recycle_ring); + jsonw_uint_field(json_wtr, "cache", s->recycle_cache); + jsonw_end_object(json_wtr); +} + +static void print_plain_recycling_stats(struct pp_stat *s) +{ + double recycle; + + if (s->alloc_fast + s->alloc_slow) { + recycle = (double)(s->recycle_ring + s->recycle_cache) / + (s->alloc_fast + s->alloc_slow) * 100; + printf("recycling: %.1lf%% (alloc: %zu:%zu recycle: %zu:%zu)", + recycle, s->alloc_slow, s->alloc_fast, + s->recycle_ring, s->recycle_cache); + } +} + +static void print_json_stats(struct pp_stats_array *a) +{ + jsonw_start_array(json_wtr); + + for (unsigned int i = 0; i < a->i; i++) { + char ifname[IF_NAMESIZE]; + struct pp_stat *s = &a->s[i]; + const char *name; + + jsonw_start_object(json_wtr); + + if (!s->ifc) { + jsonw_string_field(json_wtr, "ifname", ""); + jsonw_uint_field(json_wtr, "ifindex", 0); + } else { + name = if_indextoname(s->ifc, ifname); + if (name) + jsonw_string_field(json_wtr, "ifname", name); + jsonw_uint_field(json_wtr, "ifindex", s->ifc); + } + + jsonw_uint_field(json_wtr, "page_pools", s->live[1].cnt); + jsonw_uint_field(json_wtr, "zombies", s->live[0].cnt); + + jsonw_name(json_wtr, "live"); + jsonw_start_object(json_wtr); + jsonw_uint_field(json_wtr, "refs", s->live[1].refs); + jsonw_uint_field(json_wtr, "bytes", s->live[1].bytes); + jsonw_end_object(json_wtr); + + jsonw_name(json_wtr, "zombie"); + jsonw_start_object(json_wtr); + jsonw_uint_field(json_wtr, "refs", s->live[0].refs); + jsonw_uint_field(json_wtr, "bytes", s->live[0].bytes); + jsonw_end_object(json_wtr); + + if (s->alloc_fast || s->alloc_slow) + print_json_recycling_stats(s); + + jsonw_end_object(json_wtr); + } + + jsonw_end_array(json_wtr); +} + +static void print_plain_stats(struct pp_stats_array *a) +{ + for (unsigned int i = 0; i < a->i; i++) { + char ifname[IF_NAMESIZE]; + struct pp_stat *s = &a->s[i]; + const char *name; + + if (!s->ifc) { + printf("\t"); + } else { + name = if_indextoname(s->ifc, ifname); + if (name) + printf("%8s", name); + printf("[%u]\t", s->ifc); + } + + printf("page pools: %u (zombies: %u)\n", + s->live[1].cnt, s->live[0].cnt); + printf("\t\trefs: %zu bytes: %zu (refs: %zu bytes: %zu)\n", + s->live[1].refs, s->live[1].bytes, + s->live[0].refs, s->live[0].bytes); + + if (s->alloc_fast || s->alloc_slow) { + printf("\t\t"); + print_plain_recycling_stats(s); + printf("\n"); + } + } +} + +static bool +find_pool_stat_in_list(struct netdev_page_pool_stats_get_list *pp_stats, + __u64 pool_id, struct pp_stat *pstat) +{ + ynl_dump_foreach(pp_stats, pp) { + if (!pp->_present.info || !pp->info._present.id) + continue; + if (pp->info.id != pool_id) + continue; + + memset(pstat, 0, sizeof(*pstat)); + if (pp->_present.alloc_fast) + pstat->alloc_fast = pp->alloc_fast; + if (pp->_present.alloc_refill) + pstat->alloc_fast += pp->alloc_refill; + if (pp->_present.alloc_slow) + pstat->alloc_slow = pp->alloc_slow; + if (pp->_present.recycle_ring) + pstat->recycle_ring = pp->recycle_ring; + if (pp->_present.recycle_cached) + pstat->recycle_cache = pp->recycle_cached; + return true; + } + return false; +} + +static void +print_json_pool_list(struct netdev_page_pool_get_list *pools, + struct netdev_page_pool_stats_get_list *pp_stats, + bool zombies_only) +{ + jsonw_start_array(json_wtr); + + ynl_dump_foreach(pools, pp) { + char ifname[IF_NAMESIZE]; + struct pp_stat pstat; + const char *name; + + if (zombies_only && !pp->_present.detach_time) + continue; + + jsonw_start_object(json_wtr); + + jsonw_uint_field(json_wtr, "id", pp->id); + + if (pp->_present.ifindex) { + name = if_indextoname(pp->ifindex, ifname); + if (name) + jsonw_string_field(json_wtr, "ifname", name); + jsonw_uint_field(json_wtr, "ifindex", pp->ifindex); + } + + if (pp->_present.napi_id) + jsonw_uint_field(json_wtr, "napi_id", pp->napi_id); + + if (pp->_present.inflight) + jsonw_uint_field(json_wtr, "refs", pp->inflight); + + if (pp->_present.inflight_mem) + jsonw_uint_field(json_wtr, "bytes", pp->inflight_mem); + + if (pp->_present.detach_time) + jsonw_uint_field(json_wtr, "detach_time", pp->detach_time); + + if (pp->_present.dmabuf) + jsonw_uint_field(json_wtr, "dmabuf", pp->dmabuf); + + if (find_pool_stat_in_list(pp_stats, pp->id, &pstat) && + (pstat.alloc_fast || pstat.alloc_slow)) + print_json_recycling_stats(&pstat); + + jsonw_end_object(json_wtr); + } + + jsonw_end_array(json_wtr); +} + +static void +print_plain_pool_list(struct netdev_page_pool_get_list *pools, + struct netdev_page_pool_stats_get_list *pp_stats, + bool zombies_only) +{ + ynl_dump_foreach(pools, pp) { + char ifname[IF_NAMESIZE]; + struct pp_stat pstat; + const char *name; + + if (zombies_only && !pp->_present.detach_time) + continue; + + printf("pool id: %llu", pp->id); + + if (pp->_present.ifindex) { + name = if_indextoname(pp->ifindex, ifname); + if (name) + printf(" dev: %s", name); + printf("[%u]", pp->ifindex); + } + + if (pp->_present.napi_id) + printf(" napi: %llu", pp->napi_id); + + printf("\n"); + + if (pp->_present.inflight || pp->_present.inflight_mem) { + printf(" inflight:"); + if (pp->_present.inflight) + printf(" %llu pages", pp->inflight); + if (pp->_present.inflight_mem) + printf(" %llu bytes", pp->inflight_mem); + printf("\n"); + } + + if (pp->_present.detach_time) + printf(" detached: %llu\n", pp->detach_time); + + if (pp->_present.dmabuf) + printf(" dmabuf: %u\n", pp->dmabuf); + + if (find_pool_stat_in_list(pp_stats, pp->id, &pstat) && + (pstat.alloc_fast || pstat.alloc_slow)) { + printf(" "); + print_plain_recycling_stats(&pstat); + printf("\n"); + } + } +} + +static void aggregate_device_stats(struct pp_stats_array *a, + struct netdev_page_pool_get_list *pools, + struct netdev_page_pool_stats_get_list *pp_stats) +{ + ynl_dump_foreach(pools, pp) { + struct pp_stat *s = find_ifc(a, pp->ifindex); + + count_pool(s, 1, pp); + if (pp->_present.detach_time) + count_pool(s, 0, pp); + } + + ynl_dump_foreach(pp_stats, pp) { + struct pp_stat *s = find_ifc(a, pp->info.ifindex); + + if (pp->_present.alloc_fast) + s->alloc_fast += pp->alloc_fast; + if (pp->_present.alloc_refill) + s->alloc_fast += pp->alloc_refill; + if (pp->_present.alloc_slow) + s->alloc_slow += pp->alloc_slow; + if (pp->_present.recycle_ring) + s->recycle_ring += pp->recycle_ring; + if (pp->_present.recycle_cached) + s->recycle_cache += pp->recycle_cached; + } +} + +static int do_stats(int argc, char **argv) +{ + struct netdev_page_pool_stats_get_list *pp_stats; + struct netdev_page_pool_get_list *pools; + enum { + GROUP_BY_DEVICE, + GROUP_BY_POOL, + } group_by = GROUP_BY_DEVICE; + bool zombies_only = false; + struct pp_stats_array a = {}; + struct ynl_error yerr; + struct ynl_sock *ys; + int ret = 0; + + /* Parse options */ + while (argc > 0) { + if (is_prefix(*argv, "group-by")) { + NEXT_ARG(); + + if (!REQ_ARGS(1)) + return -1; + + if (is_prefix(*argv, "device")) { + group_by = GROUP_BY_DEVICE; + } else if (is_prefix(*argv, "pp") || + is_prefix(*argv, "page-pool") || + is_prefix(*argv, "none")) { + group_by = GROUP_BY_POOL; + } else { + p_err("invalid group-by value '%s'", *argv); + return -1; + } + NEXT_ARG(); + } else if (is_prefix(*argv, "zombies")) { + zombies_only = true; + group_by = GROUP_BY_POOL; + NEXT_ARG(); + } else { + p_err("unknown option '%s'", *argv); + return -1; + } + } + + ys = ynl_sock_create(&ynl_netdev_family, &yerr); + if (!ys) { + p_err("YNL: %s", yerr.msg); + return -1; + } + + pools = netdev_page_pool_get_dump(ys); + if (!pools) { + p_err("failed to get page pools: %s", ys->err.msg); + ret = -1; + goto exit_close; + } + + pp_stats = netdev_page_pool_stats_get_dump(ys); + if (!pp_stats) { + p_err("failed to get page pool stats: %s", ys->err.msg); + ret = -1; + goto exit_free_pp_list; + } + + /* If grouping by pool, print individual pools */ + if (group_by == GROUP_BY_POOL) { + if (json_output) + print_json_pool_list(pools, pp_stats, zombies_only); + else + print_plain_pool_list(pools, pp_stats, zombies_only); + } else { + /* Aggregated stats mode (group-by device) */ + a.max = 64; + a.s = calloc(a.max, sizeof(*a.s)); + if (!a.s) { + p_err("failed to allocate stats array"); + ret = -1; + goto exit_free_stats_list; + } + + aggregate_device_stats(&a, pools, pp_stats); + + if (json_output) + print_json_stats(&a); + else + print_plain_stats(&a); + + free(a.s); + } + +exit_free_stats_list: + netdev_page_pool_stats_get_list_free(pp_stats); +exit_free_pp_list: + netdev_page_pool_get_list_free(pools); +exit_close: + ynl_sock_destroy(ys); + return ret; +} + +static int do_help(int argc __attribute__((unused)), + char **argv __attribute__((unused))) +{ + if (json_output) { + jsonw_null(json_wtr); + return 0; + } + + fprintf(stderr, + "Usage: %s page-pool { COMMAND | help }\n" + " %s page-pool stats [ OPTIONS ]\n" + "\n" + " OPTIONS := { group-by { device | page-pool | none } | zombies }\n" + "\n" + " stats - Display page pool statistics\n" + " stats group-by device - Group statistics by network device (default)\n" + " stats group-by page-pool | pp | none\n" + " - Show individual page pool details (no grouping)\n" + " stats zombies - Show only zombie page pools (detached but with\n" + " pages in flight). Implies group-by page-pool.\n" + "", + bin_name, bin_name); + + return 0; +} + +static const struct cmd page_pool_cmds[] = { + { "help", do_help }, + { "stats", do_stats }, + { 0 } +}; + +int do_page_pool(int argc, char **argv) +{ + return cmd_select(page_pool_cmds, argc, argv, do_help); +} From 3f0a638d45fcf98f7a53ece0aadf928dfc328a06 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 7 Nov 2025 08:22:26 -0800 Subject: [PATCH 0716/1381] tools: ynltool: add qstats support $ ynltool qstat eth0 rx-packets: 493192163 rx-bytes: 1442544543997 tx-packets: 745999838 tx-bytes: 4574215826482 tx-stop: 7033 tx-wake: 7033 $ ynltool qstat show group-by queue eth0 rx-0 packets: 70196880 bytes: 178633973750 eth0 rx-1 packets: 63623419 bytes: 197274745250 ... eth0 tx-1 packets: 98645810 bytes: 631247647938 stop: 1048 wake: 1048 eth0 tx-2 packets: 86775824 bytes: 563930471952 stop: 1126 wake: 1126 ... $ ynltool -j qstat | jq [ { "ifname": "eth0", "ifindex": 2, "rx": { "packets": 493396439, "bytes": 1443608198921 }, "tx": { "packets": 746239978, "bytes": 4574333772645, "stop": 7072, "wake": 7072 } } ] Signed-off-by: Jakub Kicinski Link: https://patch.msgid.link/20251107162227.980672-4-kuba@kernel.org Acked-by: Stanislav Fomichev Signed-off-by: Paolo Abeni --- tools/net/ynl/ynltool/main.c | 3 +- tools/net/ynl/ynltool/main.h | 1 + tools/net/ynl/ynltool/qstats.c | 330 +++++++++++++++++++++++++++++++++ 3 files changed, 333 insertions(+), 1 deletion(-) create mode 100644 tools/net/ynl/ynltool/qstats.c diff --git a/tools/net/ynl/ynltool/main.c b/tools/net/ynl/ynltool/main.c index f83c6f3245c8..5d0f428eed0a 100644 --- a/tools/net/ynl/ynltool/main.c +++ b/tools/net/ynl/ynltool/main.c @@ -47,7 +47,7 @@ static int do_help(int argc __attribute__((unused)), "Usage: %s [OPTIONS] OBJECT { COMMAND | help }\n" " %s version\n" "\n" - " OBJECT := { page-pool }\n" + " OBJECT := { page-pool | qstats }\n" " " HELP_SPEC_OPTIONS "\n" "", bin_name, bin_name); @@ -72,6 +72,7 @@ static int do_version(int argc __attribute__((unused)), static const struct cmd commands[] = { { "help", do_help }, { "page-pool", do_page_pool }, + { "qstats", do_qstats }, { "version", do_version }, { 0 } }; diff --git a/tools/net/ynl/ynltool/main.h b/tools/net/ynl/ynltool/main.h index fd05d21451a2..c7039f9ac55a 100644 --- a/tools/net/ynl/ynltool/main.h +++ b/tools/net/ynl/ynltool/main.h @@ -61,5 +61,6 @@ int cmd_select(const struct cmd *cmds, int argc, char **argv, /* subcommands */ int do_page_pool(int argc, char **argv); +int do_qstats(int argc, char **argv); #endif /* __YNLTOOL_H */ diff --git a/tools/net/ynl/ynltool/qstats.c b/tools/net/ynl/ynltool/qstats.c new file mode 100644 index 000000000000..fcdbb6d9a852 --- /dev/null +++ b/tools/net/ynl/ynltool/qstats.c @@ -0,0 +1,330 @@ +// SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) + +#include +#include +#include +#include +#include + +#include +#include "netdev-user.h" + +#include "main.h" + +static enum netdev_qstats_scope scope; /* default - device */ + +static void print_json_qstats(struct netdev_qstats_get_list *qstats) +{ + jsonw_start_array(json_wtr); + + ynl_dump_foreach(qstats, qs) { + char ifname[IF_NAMESIZE]; + const char *name; + + jsonw_start_object(json_wtr); + + name = if_indextoname(qs->ifindex, ifname); + if (name) + jsonw_string_field(json_wtr, "ifname", name); + jsonw_uint_field(json_wtr, "ifindex", qs->ifindex); + + if (qs->_present.queue_type) + jsonw_string_field(json_wtr, "queue-type", + netdev_queue_type_str(qs->queue_type)); + if (qs->_present.queue_id) + jsonw_uint_field(json_wtr, "queue-id", qs->queue_id); + + if (qs->_present.rx_packets || qs->_present.rx_bytes || + qs->_present.rx_alloc_fail || qs->_present.rx_hw_drops || + qs->_present.rx_csum_complete || qs->_present.rx_hw_gro_packets) { + jsonw_name(json_wtr, "rx"); + jsonw_start_object(json_wtr); + if (qs->_present.rx_packets) + jsonw_uint_field(json_wtr, "packets", qs->rx_packets); + if (qs->_present.rx_bytes) + jsonw_uint_field(json_wtr, "bytes", qs->rx_bytes); + if (qs->_present.rx_alloc_fail) + jsonw_uint_field(json_wtr, "alloc-fail", qs->rx_alloc_fail); + if (qs->_present.rx_hw_drops) + jsonw_uint_field(json_wtr, "hw-drops", qs->rx_hw_drops); + if (qs->_present.rx_hw_drop_overruns) + jsonw_uint_field(json_wtr, "hw-drop-overruns", qs->rx_hw_drop_overruns); + if (qs->_present.rx_hw_drop_ratelimits) + jsonw_uint_field(json_wtr, "hw-drop-ratelimits", qs->rx_hw_drop_ratelimits); + if (qs->_present.rx_csum_complete) + jsonw_uint_field(json_wtr, "csum-complete", qs->rx_csum_complete); + if (qs->_present.rx_csum_unnecessary) + jsonw_uint_field(json_wtr, "csum-unnecessary", qs->rx_csum_unnecessary); + if (qs->_present.rx_csum_none) + jsonw_uint_field(json_wtr, "csum-none", qs->rx_csum_none); + if (qs->_present.rx_csum_bad) + jsonw_uint_field(json_wtr, "csum-bad", qs->rx_csum_bad); + if (qs->_present.rx_hw_gro_packets) + jsonw_uint_field(json_wtr, "hw-gro-packets", qs->rx_hw_gro_packets); + if (qs->_present.rx_hw_gro_bytes) + jsonw_uint_field(json_wtr, "hw-gro-bytes", qs->rx_hw_gro_bytes); + if (qs->_present.rx_hw_gro_wire_packets) + jsonw_uint_field(json_wtr, "hw-gro-wire-packets", qs->rx_hw_gro_wire_packets); + if (qs->_present.rx_hw_gro_wire_bytes) + jsonw_uint_field(json_wtr, "hw-gro-wire-bytes", qs->rx_hw_gro_wire_bytes); + jsonw_end_object(json_wtr); + } + + if (qs->_present.tx_packets || qs->_present.tx_bytes || + qs->_present.tx_hw_drops || qs->_present.tx_csum_none || + qs->_present.tx_hw_gso_packets) { + jsonw_name(json_wtr, "tx"); + jsonw_start_object(json_wtr); + if (qs->_present.tx_packets) + jsonw_uint_field(json_wtr, "packets", qs->tx_packets); + if (qs->_present.tx_bytes) + jsonw_uint_field(json_wtr, "bytes", qs->tx_bytes); + if (qs->_present.tx_hw_drops) + jsonw_uint_field(json_wtr, "hw-drops", qs->tx_hw_drops); + if (qs->_present.tx_hw_drop_errors) + jsonw_uint_field(json_wtr, "hw-drop-errors", qs->tx_hw_drop_errors); + if (qs->_present.tx_hw_drop_ratelimits) + jsonw_uint_field(json_wtr, "hw-drop-ratelimits", qs->tx_hw_drop_ratelimits); + if (qs->_present.tx_csum_none) + jsonw_uint_field(json_wtr, "csum-none", qs->tx_csum_none); + if (qs->_present.tx_needs_csum) + jsonw_uint_field(json_wtr, "needs-csum", qs->tx_needs_csum); + if (qs->_present.tx_hw_gso_packets) + jsonw_uint_field(json_wtr, "hw-gso-packets", qs->tx_hw_gso_packets); + if (qs->_present.tx_hw_gso_bytes) + jsonw_uint_field(json_wtr, "hw-gso-bytes", qs->tx_hw_gso_bytes); + if (qs->_present.tx_hw_gso_wire_packets) + jsonw_uint_field(json_wtr, "hw-gso-wire-packets", qs->tx_hw_gso_wire_packets); + if (qs->_present.tx_hw_gso_wire_bytes) + jsonw_uint_field(json_wtr, "hw-gso-wire-bytes", qs->tx_hw_gso_wire_bytes); + if (qs->_present.tx_stop) + jsonw_uint_field(json_wtr, "stop", qs->tx_stop); + if (qs->_present.tx_wake) + jsonw_uint_field(json_wtr, "wake", qs->tx_wake); + jsonw_end_object(json_wtr); + } + + jsonw_end_object(json_wtr); + } + + jsonw_end_array(json_wtr); +} + +static void print_one(bool present, const char *name, unsigned long long val, + int *line) +{ + if (!present) + return; + + if (!*line) { + printf(" "); + ++(*line); + } + + /* Don't waste space on tx- and rx- prefix, its implied by queue type */ + if (scope == NETDEV_QSTATS_SCOPE_QUEUE && + (name[0] == 'r' || name[0] == 't') && + name[1] == 'x' && name[2] == '-') + name += 3; + + printf(" %15s: %15llu", name, val); + + if (++(*line) == 3) { + printf("\n"); + *line = 0; + } +} + +static void print_plain_qstats(struct netdev_qstats_get_list *qstats) +{ + ynl_dump_foreach(qstats, qs) { + char ifname[IF_NAMESIZE]; + const char *name; + int n; + + name = if_indextoname(qs->ifindex, ifname); + if (name) + printf("%s", name); + else + printf("ifindex:%u", qs->ifindex); + + if (qs->_present.queue_type && qs->_present.queue_id) + printf("\t%s-%-3u", + netdev_queue_type_str(qs->queue_type), + qs->queue_id); + else + printf("\t "); + + n = 1; + + /* Basic counters */ + print_one(qs->_present.rx_packets, "rx-packets", qs->rx_packets, &n); + print_one(qs->_present.rx_bytes, "rx-bytes", qs->rx_bytes, &n); + print_one(qs->_present.tx_packets, "tx-packets", qs->tx_packets, &n); + print_one(qs->_present.tx_bytes, "tx-bytes", qs->tx_bytes, &n); + + /* RX error/drop counters */ + print_one(qs->_present.rx_alloc_fail, "rx-alloc-fail", + qs->rx_alloc_fail, &n); + print_one(qs->_present.rx_hw_drops, "rx-hw-drops", + qs->rx_hw_drops, &n); + print_one(qs->_present.rx_hw_drop_overruns, "rx-hw-drop-overruns", + qs->rx_hw_drop_overruns, &n); + print_one(qs->_present.rx_hw_drop_ratelimits, "rx-hw-drop-ratelimits", + qs->rx_hw_drop_ratelimits, &n); + + /* RX checksum counters */ + print_one(qs->_present.rx_csum_complete, "rx-csum-complete", + qs->rx_csum_complete, &n); + print_one(qs->_present.rx_csum_unnecessary, "rx-csum-unnecessary", + qs->rx_csum_unnecessary, &n); + print_one(qs->_present.rx_csum_none, "rx-csum-none", + qs->rx_csum_none, &n); + print_one(qs->_present.rx_csum_bad, "rx-csum-bad", + qs->rx_csum_bad, &n); + + /* RX GRO counters */ + print_one(qs->_present.rx_hw_gro_packets, "rx-hw-gro-packets", + qs->rx_hw_gro_packets, &n); + print_one(qs->_present.rx_hw_gro_bytes, "rx-hw-gro-bytes", + qs->rx_hw_gro_bytes, &n); + print_one(qs->_present.rx_hw_gro_wire_packets, "rx-hw-gro-wire-packets", + qs->rx_hw_gro_wire_packets, &n); + print_one(qs->_present.rx_hw_gro_wire_bytes, "rx-hw-gro-wire-bytes", + qs->rx_hw_gro_wire_bytes, &n); + + /* TX error/drop counters */ + print_one(qs->_present.tx_hw_drops, "tx-hw-drops", + qs->tx_hw_drops, &n); + print_one(qs->_present.tx_hw_drop_errors, "tx-hw-drop-errors", + qs->tx_hw_drop_errors, &n); + print_one(qs->_present.tx_hw_drop_ratelimits, "tx-hw-drop-ratelimits", + qs->tx_hw_drop_ratelimits, &n); + + /* TX checksum counters */ + print_one(qs->_present.tx_csum_none, "tx-csum-none", + qs->tx_csum_none, &n); + print_one(qs->_present.tx_needs_csum, "tx-needs-csum", + qs->tx_needs_csum, &n); + + /* TX GSO counters */ + print_one(qs->_present.tx_hw_gso_packets, "tx-hw-gso-packets", + qs->tx_hw_gso_packets, &n); + print_one(qs->_present.tx_hw_gso_bytes, "tx-hw-gso-bytes", + qs->tx_hw_gso_bytes, &n); + print_one(qs->_present.tx_hw_gso_wire_packets, "tx-hw-gso-wire-packets", + qs->tx_hw_gso_wire_packets, &n); + print_one(qs->_present.tx_hw_gso_wire_bytes, "tx-hw-gso-wire-bytes", + qs->tx_hw_gso_wire_bytes, &n); + + /* TX queue control */ + print_one(qs->_present.tx_stop, "tx-stop", qs->tx_stop, &n); + print_one(qs->_present.tx_wake, "tx-wake", qs->tx_wake, &n); + + if (n) + printf("\n"); + } +} + +static int do_show(int argc, char **argv) +{ + struct netdev_qstats_get_list *qstats; + struct netdev_qstats_get_req *req; + struct ynl_error yerr; + struct ynl_sock *ys; + int ret = 0; + + /* Parse options */ + while (argc > 0) { + if (is_prefix(*argv, "scope") || is_prefix(*argv, "group-by")) { + NEXT_ARG(); + + if (!REQ_ARGS(1)) + return -1; + + if (is_prefix(*argv, "queue")) { + scope = NETDEV_QSTATS_SCOPE_QUEUE; + } else if (is_prefix(*argv, "device")) { + scope = 0; + } else { + p_err("invalid scope value '%s'", *argv); + return -1; + } + NEXT_ARG(); + } else { + p_err("unknown option '%s'", *argv); + return -1; + } + } + + ys = ynl_sock_create(&ynl_netdev_family, &yerr); + if (!ys) { + p_err("YNL: %s", yerr.msg); + return -1; + } + + req = netdev_qstats_get_req_alloc(); + if (!req) { + p_err("failed to allocate qstats request"); + ret = -1; + goto exit_close; + } + + if (scope) + netdev_qstats_get_req_set_scope(req, scope); + + qstats = netdev_qstats_get_dump(ys, req); + netdev_qstats_get_req_free(req); + if (!qstats) { + p_err("failed to get queue stats: %s", ys->err.msg); + ret = -1; + goto exit_close; + } + + /* Print the stats as returned by the kernel */ + if (json_output) + print_json_qstats(qstats); + else + print_plain_qstats(qstats); + + netdev_qstats_get_list_free(qstats); +exit_close: + ynl_sock_destroy(ys); + return ret; +} + +static int do_help(int argc __attribute__((unused)), + char **argv __attribute__((unused))) +{ + if (json_output) { + jsonw_null(json_wtr); + return 0; + } + + fprintf(stderr, + "Usage: %s qstats { COMMAND | help }\n" + " %s qstats [ show ] [ OPTIONS ]\n" + "\n" + " OPTIONS := { scope queue | group-by { device | queue } }\n" + "\n" + " show - Display queue statistics (default)\n" + " Statistics are aggregated for the entire device.\n" + " show scope queue - Display per-queue statistics\n" + " show group-by device - Display device-aggregated statistics (default)\n" + " show group-by queue - Display per-queue statistics\n" + "", + bin_name, bin_name); + + return 0; +} + +static const struct cmd qstats_cmds[] = { + { "show", do_show }, + { "help", do_help }, + { 0 } +}; + +int do_qstats(int argc, char **argv) +{ + return cmd_select(qstats_cmds, argc, argv, do_help); +} From 9eef97a9dea3d059cb719a26a055ff09933e964e Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 7 Nov 2025 08:22:27 -0800 Subject: [PATCH 0717/1381] tools: ynltool: add traffic distribution balance The main if not only use case for per-queue stats today is checking for traffic imbalance. Add simple traffic balance analysis to qstats. $ ynltool qstat balance eth0 rx 44 queues: rx-packets : cv=6.9% ns=24.2% stddev=512006493 min=6278921110 max=8011570575 mean=7437054644 rx-bytes : cv=6.9% ns=24.1% stddev=759670503060 min=9326315769440 max=11884393670786 mean=11035439201354 ... $ ynltool -j qstat balance | jq [ { "ifname": "eth0", "ifindex": 2, "queue-type": "rx", "rx-packets": { "queue-count": 44, "min": 6278301665, "max": 8010780185, "mean": 7.43635E+9, "stddev": 5.12012E+8, "coefficient-of-variation": 6.88525, "normalized-spread": 24.249 }, ... Signed-off-by: Jakub Kicinski Link: https://patch.msgid.link/20251107162227.980672-5-kuba@kernel.org Acked-by: Stanislav Fomichev Signed-off-by: Paolo Abeni --- tools/net/ynl/ynltool/Makefile | 2 +- tools/net/ynl/ynltool/qstats.c | 293 ++++++++++++++++++++++++++++++++- 2 files changed, 293 insertions(+), 2 deletions(-) diff --git a/tools/net/ynl/ynltool/Makefile b/tools/net/ynl/ynltool/Makefile index 11240740ed81..86c30b7420cf 100644 --- a/tools/net/ynl/ynltool/Makefile +++ b/tools/net/ynl/ynltool/Makefile @@ -31,7 +31,7 @@ Q = @ $(YNLTOOL): ../libynl.a $(OBJS) $(Q)echo -e "\tLINK $@" - $(Q)$(CC) $(CFLAGS) -o $@ $(OBJS) ../libynl.a -lmnl + $(Q)$(CC) $(CFLAGS) -o $@ $(OBJS) ../libynl.a -lmnl -lm %.o: %.c ../libynl.a $(Q)echo -e "\tCC $@" diff --git a/tools/net/ynl/ynltool/qstats.c b/tools/net/ynl/ynltool/qstats.c index fcdbb6d9a852..31fb45709ffa 100644 --- a/tools/net/ynl/ynltool/qstats.c +++ b/tools/net/ynl/ynltool/qstats.c @@ -5,6 +5,7 @@ #include #include #include +#include #include #include "netdev-user.h" @@ -13,6 +14,16 @@ static enum netdev_qstats_scope scope; /* default - device */ +struct queue_balance { + unsigned int ifindex; + enum netdev_queue_type type; + unsigned int queue_count; + __u64 *rx_packets; + __u64 *rx_bytes; + __u64 *tx_packets; + __u64 *tx_bytes; +}; + static void print_json_qstats(struct netdev_qstats_get_list *qstats) { jsonw_start_array(json_wtr); @@ -293,6 +304,283 @@ static int do_show(int argc, char **argv) return ret; } +static void compute_stats(__u64 *values, unsigned int count, + double *mean, double *stddev, __u64 *min, __u64 *max) +{ + double sum = 0.0, variance = 0.0; + unsigned int i; + + *min = ~0ULL; + *max = 0; + + if (count == 0) { + *mean = 0; + *stddev = 0; + *min = 0; + return; + } + + for (i = 0; i < count; i++) { + sum += values[i]; + if (values[i] < *min) + *min = values[i]; + if (values[i] > *max) + *max = values[i]; + } + + *mean = sum / count; + + if (count > 1) { + for (i = 0; i < count; i++) { + double diff = values[i] - *mean; + + variance += diff * diff; + } + *stddev = sqrt(variance / (count - 1)); + } else { + *stddev = 0; + } +} + +static void print_balance_stats(const char *name, enum netdev_queue_type type, + __u64 *values, unsigned int count) +{ + double mean, stddev, cv, ns; + __u64 min, max; + + if ((name[0] == 'r' && type != NETDEV_QUEUE_TYPE_RX) || + (name[0] == 't' && type != NETDEV_QUEUE_TYPE_TX)) + return; + + compute_stats(values, count, &mean, &stddev, &min, &max); + + cv = mean > 0 ? (stddev / mean) * 100.0 : 0.0; + ns = min + max > 0 ? (double)2 * (max - min) / (max + min) * 100 : 0.0; + + printf(" %-12s: cv=%.1f%% ns=%.1f%% stddev=%.0f\n", + name, cv, ns, stddev); + printf(" %-12s min=%llu max=%llu mean=%.0f\n", + "", min, max, mean); +} + +static void +print_balance_stats_json(const char *name, enum netdev_queue_type type, + __u64 *values, unsigned int count) +{ + double mean, stddev, cv, ns; + __u64 min, max; + + if ((name[0] == 'r' && type != NETDEV_QUEUE_TYPE_RX) || + (name[0] == 't' && type != NETDEV_QUEUE_TYPE_TX)) + return; + + compute_stats(values, count, &mean, &stddev, &min, &max); + + cv = mean > 0 ? (stddev / mean) * 100.0 : 0.0; + ns = min + max > 0 ? (double)2 * (max - min) / (max + min) * 100 : 0.0; + + jsonw_name(json_wtr, name); + jsonw_start_object(json_wtr); + jsonw_uint_field(json_wtr, "queue-count", count); + jsonw_uint_field(json_wtr, "min", min); + jsonw_uint_field(json_wtr, "max", max); + jsonw_float_field(json_wtr, "mean", mean); + jsonw_float_field(json_wtr, "stddev", stddev); + jsonw_float_field(json_wtr, "coefficient-of-variation", cv); + jsonw_float_field(json_wtr, "normalized-spread", ns); + jsonw_end_object(json_wtr); +} + +static int cmp_ifindex_type(const void *a, const void *b) +{ + const struct netdev_qstats_get_rsp *qa = a; + const struct netdev_qstats_get_rsp *qb = b; + + if (qa->ifindex != qb->ifindex) + return qa->ifindex - qb->ifindex; + if (qa->queue_type != qb->queue_type) + return qa->queue_type - qb->queue_type; + return qa->queue_id - qb->queue_id; +} + +static int do_balance(int argc, char **argv __attribute__((unused))) +{ + struct netdev_qstats_get_list *qstats; + struct netdev_qstats_get_req *req; + struct netdev_qstats_get_rsp **sorted; + struct ynl_error yerr; + struct ynl_sock *ys; + unsigned int count = 0; + unsigned int i, j; + int ret = 0; + + if (argc > 0) { + p_err("balance command takes no arguments"); + return -1; + } + + ys = ynl_sock_create(&ynl_netdev_family, &yerr); + if (!ys) { + p_err("YNL: %s", yerr.msg); + return -1; + } + + req = netdev_qstats_get_req_alloc(); + if (!req) { + p_err("failed to allocate qstats request"); + ret = -1; + goto exit_close; + } + + /* Always use queue scope for balance analysis */ + netdev_qstats_get_req_set_scope(req, NETDEV_QSTATS_SCOPE_QUEUE); + + qstats = netdev_qstats_get_dump(ys, req); + netdev_qstats_get_req_free(req); + if (!qstats) { + p_err("failed to get queue stats: %s", ys->err.msg); + ret = -1; + goto exit_close; + } + + /* Count and sort queues */ + ynl_dump_foreach(qstats, qs) + count++; + + if (count == 0) { + if (json_output) + jsonw_start_array(json_wtr); + else + printf("No queue statistics available\n"); + goto exit_free_qstats; + } + + sorted = calloc(count, sizeof(*sorted)); + if (!sorted) { + p_err("failed to allocate sorted array"); + ret = -1; + goto exit_free_qstats; + } + + i = 0; + ynl_dump_foreach(qstats, qs) + sorted[i++] = qs; + + qsort(sorted, count, sizeof(*sorted), cmp_ifindex_type); + + if (json_output) + jsonw_start_array(json_wtr); + + /* Process each device/queue-type combination */ + i = 0; + while (i < count) { + __u64 *rx_packets, *rx_bytes, *tx_packets, *tx_bytes; + enum netdev_queue_type type = sorted[i]->queue_type; + unsigned int ifindex = sorted[i]->ifindex; + unsigned int queue_count = 0; + char ifname[IF_NAMESIZE]; + const char *name; + + /* Count queues for this device/type */ + for (j = i; j < count && sorted[j]->ifindex == ifindex && + sorted[j]->queue_type == type; j++) + queue_count++; + + /* Skip if no packets/bytes (inactive queues) */ + if (!sorted[i]->_present.rx_packets && + !sorted[i]->_present.rx_bytes && + !sorted[i]->_present.tx_packets && + !sorted[i]->_present.tx_bytes) + goto next_ifc; + + /* Allocate arrays for statistics */ + rx_packets = calloc(queue_count, sizeof(*rx_packets)); + rx_bytes = calloc(queue_count, sizeof(*rx_bytes)); + tx_packets = calloc(queue_count, sizeof(*tx_packets)); + tx_bytes = calloc(queue_count, sizeof(*tx_bytes)); + + if (!rx_packets || !rx_bytes || !tx_packets || !tx_bytes) { + p_err("failed to allocate statistics arrays"); + free(rx_packets); + free(rx_bytes); + free(tx_packets); + free(tx_bytes); + ret = -1; + goto exit_free_sorted; + } + + /* Collect statistics */ + for (j = 0; j < queue_count; j++) { + rx_packets[j] = sorted[i + j]->_present.rx_packets ? + sorted[i + j]->rx_packets : 0; + rx_bytes[j] = sorted[i + j]->_present.rx_bytes ? + sorted[i + j]->rx_bytes : 0; + tx_packets[j] = sorted[i + j]->_present.tx_packets ? + sorted[i + j]->tx_packets : 0; + tx_bytes[j] = sorted[i + j]->_present.tx_bytes ? + sorted[i + j]->tx_bytes : 0; + } + + name = if_indextoname(ifindex, ifname); + + if (json_output) { + jsonw_start_object(json_wtr); + if (name) + jsonw_string_field(json_wtr, "ifname", name); + jsonw_uint_field(json_wtr, "ifindex", ifindex); + jsonw_string_field(json_wtr, "queue-type", + netdev_queue_type_str(type)); + + print_balance_stats_json("rx-packets", type, + rx_packets, queue_count); + print_balance_stats_json("rx-bytes", type, + rx_bytes, queue_count); + print_balance_stats_json("tx-packets", type, + tx_packets, queue_count); + print_balance_stats_json("tx-bytes", type, + tx_bytes, queue_count); + + jsonw_end_object(json_wtr); + } else { + if (name) + printf("%s", name); + else + printf("ifindex:%u", ifindex); + printf(" %s %d queues:\n", + netdev_queue_type_str(type), queue_count); + + print_balance_stats("rx-packets", type, + rx_packets, queue_count); + print_balance_stats("rx-bytes", type, + rx_bytes, queue_count); + print_balance_stats("tx-packets", type, + tx_packets, queue_count); + print_balance_stats("tx-bytes", type, + tx_bytes, queue_count); + printf("\n"); + } + + free(rx_packets); + free(rx_bytes); + free(tx_packets); + free(tx_bytes); + +next_ifc: + i += queue_count; + } + + if (json_output) + jsonw_end_array(json_wtr); + +exit_free_sorted: + free(sorted); +exit_free_qstats: + netdev_qstats_get_list_free(qstats); +exit_close: + ynl_sock_destroy(ys); + return ret; +} + static int do_help(int argc __attribute__((unused)), char **argv __attribute__((unused))) { @@ -304,6 +592,7 @@ static int do_help(int argc __attribute__((unused)), fprintf(stderr, "Usage: %s qstats { COMMAND | help }\n" " %s qstats [ show ] [ OPTIONS ]\n" + " %s qstats balance\n" "\n" " OPTIONS := { scope queue | group-by { device | queue } }\n" "\n" @@ -312,14 +601,16 @@ static int do_help(int argc __attribute__((unused)), " show scope queue - Display per-queue statistics\n" " show group-by device - Display device-aggregated statistics (default)\n" " show group-by queue - Display per-queue statistics\n" + " balance - Analyze traffic distribution balance.\n" "", - bin_name, bin_name); + bin_name, bin_name, bin_name); return 0; } static const struct cmd qstats_cmds[] = { { "show", do_show }, + { "balance", do_balance }, { "help", do_help }, { 0 } }; From 0e535824d0bcf7c9bb0532d902283c31c78cd6f3 Mon Sep 17 00:00:00 2001 From: Saeed Mahameed Date: Fri, 7 Nov 2025 23:04:02 -0800 Subject: [PATCH 0718/1381] devlink: Introduce switchdev_inactive eswitch mode Adds DEVLINK_ESWITCH_MODE_SWITCHDEV_INACTIVE attribute to UAPI and documentation. Before having traffic flow through an eswitch, a user may want to have the ability to block traffic towards the FDB until FDB is fully programmed and the user is ready to send traffic to it. For example: when two eswitches are present for vports in a multi-PF setup, one eswitch may take over the traffic from the other when the user chooses. Before this take over, a user may want to first program the inactive eswitch and then once ready redirect traffic to this new eswitch. switchdev modes transition semantics: legacy->switchdev_inactive: Create switchdev mode normally, traffic not allowed to flow yet. switchdev_inactive->switchdev: Enable traffic to flow. switchdev->switchdev_inactive: Block traffic on the FDB, FDB and representros state and content is preserved. When eswitch is configured to this mode, traffic is ignored/dropped on this eswitch FDB, while current configuration is kept, e.g FDB rules and netdev representros are kept available, FDB programming is allowed. Example: # start inactive switchdev devlink dev eswitch set pci/0000:08:00.1 mode switchdev_inactive # setup TC rules, representors etc .. # activate devlink dev eswitch set pci/0000:08:00.1 mode switchdev Signed-off-by: Saeed Mahameed Reviewed-by: Jiri Pirko Link: https://patch.msgid.link/20251108070404.1551708-2-saeed@kernel.org Signed-off-by: Paolo Abeni --- Documentation/netlink/specs/devlink.yaml | 2 ++ .../networking/devlink/devlink-eswitch-attr.rst | 13 +++++++++++++ include/uapi/linux/devlink.h | 1 + net/devlink/netlink_gen.c | 2 +- 4 files changed, 17 insertions(+), 1 deletion(-) diff --git a/Documentation/netlink/specs/devlink.yaml b/Documentation/netlink/specs/devlink.yaml index 3db59c965869..426d5aa7d955 100644 --- a/Documentation/netlink/specs/devlink.yaml +++ b/Documentation/netlink/specs/devlink.yaml @@ -99,6 +99,8 @@ definitions: name: legacy - name: switchdev + - + name: switchdev-inactive - type: enum name: eswitch-inline-mode diff --git a/Documentation/networking/devlink/devlink-eswitch-attr.rst b/Documentation/networking/devlink/devlink-eswitch-attr.rst index 08bb39ab1528..eafe09abc40c 100644 --- a/Documentation/networking/devlink/devlink-eswitch-attr.rst +++ b/Documentation/networking/devlink/devlink-eswitch-attr.rst @@ -39,6 +39,10 @@ The following is a list of E-Switch attributes. rules. * ``switchdev`` allows for more advanced offloading capabilities of the E-Switch to hardware. + * ``switchdev_inactive`` switchdev mode but starts inactive, doesn't allow traffic + until explicitly activated. This mode is useful for orchestrators that + want to prepare the device in switchdev mode but only activate it when + all configurations are done. * - ``inline-mode`` - enum - Some HWs need the VF driver to put part of the packet @@ -74,3 +78,12 @@ Example Usage # enable encap-mode with legacy mode $ devlink dev eswitch set pci/0000:08:00.0 mode legacy inline-mode none encap-mode basic + + # start switchdev mode in inactive state + $ devlink dev eswitch set pci/0000:08:00.0 mode switchdev_inactive + + # setup switchdev configurations, representors, FDB entries, etc.. + ... + + # activate switchdev mode to allow traffic + $ devlink dev eswitch set pci/0000:08:00.0 mode switchdev diff --git a/include/uapi/linux/devlink.h b/include/uapi/linux/devlink.h index bcad11a787a5..157f11d3fb72 100644 --- a/include/uapi/linux/devlink.h +++ b/include/uapi/linux/devlink.h @@ -181,6 +181,7 @@ enum devlink_sb_threshold_type { enum devlink_eswitch_mode { DEVLINK_ESWITCH_MODE_LEGACY, DEVLINK_ESWITCH_MODE_SWITCHDEV, + DEVLINK_ESWITCH_MODE_SWITCHDEV_INACTIVE, }; enum devlink_eswitch_inline_mode { diff --git a/net/devlink/netlink_gen.c b/net/devlink/netlink_gen.c index 9fd00977d59e..5ad435aee29d 100644 --- a/net/devlink/netlink_gen.c +++ b/net/devlink/netlink_gen.c @@ -229,7 +229,7 @@ static const struct nla_policy devlink_eswitch_get_nl_policy[DEVLINK_ATTR_DEV_NA static const struct nla_policy devlink_eswitch_set_nl_policy[DEVLINK_ATTR_ESWITCH_ENCAP_MODE + 1] = { [DEVLINK_ATTR_BUS_NAME] = { .type = NLA_NUL_STRING, }, [DEVLINK_ATTR_DEV_NAME] = { .type = NLA_NUL_STRING, }, - [DEVLINK_ATTR_ESWITCH_MODE] = NLA_POLICY_MAX(NLA_U16, 1), + [DEVLINK_ATTR_ESWITCH_MODE] = NLA_POLICY_MAX(NLA_U16, 2), [DEVLINK_ATTR_ESWITCH_INLINE_MODE] = NLA_POLICY_MAX(NLA_U8, 3), [DEVLINK_ATTR_ESWITCH_ENCAP_MODE] = NLA_POLICY_MAX(NLA_U8, 1), }; From 9902b6381d76ccd2e08e2703390e8c8a3bcda482 Mon Sep 17 00:00:00 2001 From: Saeed Mahameed Date: Fri, 7 Nov 2025 23:04:03 -0800 Subject: [PATCH 0719/1381] net/mlx5: MPFS, add support for dynamic enable/disable MPFS (Multi PF Switch) is enabled by default in Multi-Host environments, the driver keeps a list of desired unicast mac addresses of all vports (vfs/Sfs) and applied to HW via L2_table FW command. Add API to dynamically apply the list of MACs to HW when needed for next patches, to utilize this new API in devlink eswitch active/in-active uAPI. Signed-off-by: Saeed Mahameed Signed-off-by: Adithya Jayachandran Reviewed-by: Jiri Pirko Link: https://patch.msgid.link/20251108070404.1551708-3-saeed@kernel.org Signed-off-by: Paolo Abeni --- .../ethernet/mellanox/mlx5/core/lib/mpfs.c | 116 +++++++++++++++--- .../ethernet/mellanox/mlx5/core/lib/mpfs.h | 9 ++ 2 files changed, 108 insertions(+), 17 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/mpfs.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/mpfs.c index 4450091e181a..99fb7a53add0 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lib/mpfs.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/mpfs.c @@ -65,13 +65,14 @@ static int del_l2table_entry_cmd(struct mlx5_core_dev *dev, u32 index) /* UC L2 table hash node */ struct l2table_node { struct l2addr_node node; - u32 index; /* index in HW l2 table */ + int index; /* index in HW l2 table */ int ref_count; }; struct mlx5_mpfs { struct hlist_head hash[MLX5_L2_ADDR_HASH_SIZE]; struct mutex lock; /* Synchronize l2 table access */ + bool enabled; u32 size; unsigned long *bitmap; }; @@ -114,6 +115,8 @@ int mlx5_mpfs_init(struct mlx5_core_dev *dev) return -ENOMEM; } + mpfs->enabled = true; + dev->priv.mpfs = mpfs; return 0; } @@ -135,7 +138,7 @@ int mlx5_mpfs_add_mac(struct mlx5_core_dev *dev, u8 *mac) struct mlx5_mpfs *mpfs = dev->priv.mpfs; struct l2table_node *l2addr; int err = 0; - u32 index; + int index; if (!mpfs) return 0; @@ -148,30 +151,34 @@ int mlx5_mpfs_add_mac(struct mlx5_core_dev *dev, u8 *mac) goto out; } - err = alloc_l2table_index(mpfs, &index); - if (err) - goto out; - l2addr = l2addr_hash_add(mpfs->hash, mac, struct l2table_node, GFP_KERNEL); if (!l2addr) { err = -ENOMEM; - goto hash_add_err; + goto out; } - err = set_l2table_entry_cmd(dev, index, mac); - if (err) - goto set_table_entry_err; + index = -1; + + if (mpfs->enabled) { + err = alloc_l2table_index(mpfs, &index); + if (err) + goto hash_del; + err = set_l2table_entry_cmd(dev, index, mac); + if (err) + goto free_l2table_index; + mlx5_core_dbg(dev, "MPFS entry %pM, set @index (%d)\n", + l2addr->node.addr, l2addr->index); + } l2addr->index = index; l2addr->ref_count = 1; mlx5_core_dbg(dev, "MPFS mac added %pM, index (%d)\n", mac, index); goto out; - -set_table_entry_err: - l2addr_hash_del(l2addr); -hash_add_err: +free_l2table_index: free_l2table_index(mpfs, index); +hash_del: + l2addr_hash_del(l2addr); out: mutex_unlock(&mpfs->lock); return err; @@ -183,7 +190,7 @@ int mlx5_mpfs_del_mac(struct mlx5_core_dev *dev, u8 *mac) struct mlx5_mpfs *mpfs = dev->priv.mpfs; struct l2table_node *l2addr; int err = 0; - u32 index; + int index; if (!mpfs) return 0; @@ -200,12 +207,87 @@ int mlx5_mpfs_del_mac(struct mlx5_core_dev *dev, u8 *mac) goto unlock; index = l2addr->index; - del_l2table_entry_cmd(dev, index); + if (index >= 0) { + del_l2table_entry_cmd(dev, index); + free_l2table_index(mpfs, index); + mlx5_core_dbg(dev, "MPFS entry %pM, deleted @index (%d)\n", + mac, index); + } l2addr_hash_del(l2addr); - free_l2table_index(mpfs, index); mlx5_core_dbg(dev, "MPFS mac deleted %pM, index (%d)\n", mac, index); unlock: mutex_unlock(&mpfs->lock); return err; } EXPORT_SYMBOL(mlx5_mpfs_del_mac); + +int mlx5_mpfs_enable(struct mlx5_core_dev *dev) +{ + struct mlx5_mpfs *mpfs = dev->priv.mpfs; + struct l2table_node *l2addr; + struct hlist_node *n; + int err = 0, i; + + if (!mpfs) + return -ENODEV; + + mutex_lock(&mpfs->lock); + if (mpfs->enabled) + goto out; + mpfs->enabled = true; + mlx5_core_dbg(dev, "MPFS enabling mpfs\n"); + + mlx5_mpfs_foreach(l2addr, n, mpfs, i) { + u32 index; + + err = alloc_l2table_index(mpfs, &index); + if (err) { + mlx5_core_err(dev, "Failed to allocated MPFS index for %pM, err(%d)\n", + l2addr->node.addr, err); + goto out; + } + + err = set_l2table_entry_cmd(dev, index, l2addr->node.addr); + if (err) { + mlx5_core_err(dev, "Failed to set MPFS l2table entry for %pM index=%d, err(%d)\n", + l2addr->node.addr, index, err); + free_l2table_index(mpfs, index); + goto out; + } + + l2addr->index = index; + mlx5_core_dbg(dev, "MPFS entry %pM, set @index (%d)\n", + l2addr->node.addr, l2addr->index); + } +out: + mutex_unlock(&mpfs->lock); + return err; +} + +void mlx5_mpfs_disable(struct mlx5_core_dev *dev) +{ + struct mlx5_mpfs *mpfs = dev->priv.mpfs; + struct l2table_node *l2addr; + struct hlist_node *n; + int i; + + if (!mpfs) + return; + + mutex_lock(&mpfs->lock); + if (!mpfs->enabled) + goto unlock; + mlx5_mpfs_foreach(l2addr, n, mpfs, i) { + if (l2addr->index < 0) + continue; + del_l2table_entry_cmd(dev, l2addr->index); + free_l2table_index(mpfs, l2addr->index); + mlx5_core_dbg(dev, "MPFS entry %pM, deleted @index (%d)\n", + l2addr->node.addr, l2addr->index); + l2addr->index = -1; + } + mpfs->enabled = false; + mlx5_core_dbg(dev, "MPFS disabled\n"); +unlock: + mutex_unlock(&mpfs->lock); +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/mpfs.h b/drivers/net/ethernet/mellanox/mlx5/core/lib/mpfs.h index 4a293542a7aa..9c63838ce1f3 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lib/mpfs.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/mpfs.h @@ -45,6 +45,10 @@ struct l2addr_node { u8 addr[ETH_ALEN]; }; +#define mlx5_mpfs_foreach(hs, tmp, mpfs, i) \ + for (i = 0; i < MLX5_L2_ADDR_HASH_SIZE; i++) \ + hlist_for_each_entry_safe(hs, tmp, &(mpfs)->hash[i], node.hlist) + #define for_each_l2hash_node(hn, tmp, hash, i) \ for (i = 0; i < MLX5_L2_ADDR_HASH_SIZE; i++) \ hlist_for_each_entry_safe(hn, tmp, &(hash)[i], hlist) @@ -82,11 +86,16 @@ struct l2addr_node { }) #ifdef CONFIG_MLX5_MPFS +struct mlx5_core_dev; int mlx5_mpfs_init(struct mlx5_core_dev *dev); void mlx5_mpfs_cleanup(struct mlx5_core_dev *dev); +int mlx5_mpfs_enable(struct mlx5_core_dev *dev); +void mlx5_mpfs_disable(struct mlx5_core_dev *dev); #else /* #ifndef CONFIG_MLX5_MPFS */ static inline int mlx5_mpfs_init(struct mlx5_core_dev *dev) { return 0; } static inline void mlx5_mpfs_cleanup(struct mlx5_core_dev *dev) {} +static inline int mlx5_mpfs_enable(struct mlx5_core_dev *dev) { return 0; } +static inline void mlx5_mpfs_disable(struct mlx5_core_dev *dev) {} #endif #endif From 9da611df15aa8d519f9947b88a5c733267cba888 Mon Sep 17 00:00:00 2001 From: Saeed Mahameed Date: Fri, 7 Nov 2025 23:04:04 -0800 Subject: [PATCH 0720/1381] net/mlx5: E-Switch, support eswitch inactive mode Add support for eswitch switchdev inactive mode Inactive mode: Drop all traffic going to FDB, Remove mpfs l2 rules and disconnect adjacent vports. Active mode: Traffic flows through FDB, mpfs table populated, and adjacent vports are connected. Signed-off-by: Saeed Mahameed Signed-off-by: Adithya Jayachandran Reviewed-by: Jiri Pirko Link: https://patch.msgid.link/20251108070404.1551708-4-saeed@kernel.org Signed-off-by: Paolo Abeni --- .../mellanox/mlx5/core/esw/adj_vport.c | 15 +- .../net/ethernet/mellanox/mlx5/core/eswitch.h | 6 + .../mellanox/mlx5/core/eswitch_offloads.c | 207 +++++++++++++++++- .../net/ethernet/mellanox/mlx5/core/fs_core.c | 5 + .../ethernet/mellanox/mlx5/core/lib/mpfs.c | 2 +- include/linux/mlx5/fs.h | 1 + 6 files changed, 214 insertions(+), 22 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/adj_vport.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/adj_vport.c index 0091ba697bae..250af09b5af2 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/esw/adj_vport.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/adj_vport.c @@ -4,13 +4,8 @@ #include "fs_core.h" #include "eswitch.h" -enum { - MLX5_ADJ_VPORT_DISCONNECT = 0x0, - MLX5_ADJ_VPORT_CONNECT = 0x1, -}; - -static int mlx5_esw_adj_vport_modify(struct mlx5_core_dev *dev, - u16 vport, bool connect) +int mlx5_esw_adj_vport_modify(struct mlx5_core_dev *dev, u16 vport, + bool connect) { u32 in[MLX5_ST_SZ_DW(modify_vport_state_in)] = {}; @@ -24,7 +19,7 @@ static int mlx5_esw_adj_vport_modify(struct mlx5_core_dev *dev, MLX5_SET(modify_vport_state_in, in, egress_connect_valid, 1); MLX5_SET(modify_vport_state_in, in, ingress_connect, connect); MLX5_SET(modify_vport_state_in, in, egress_connect, connect); - + MLX5_SET(modify_vport_state_in, in, admin_state, connect); return mlx5_cmd_exec_in(dev, modify_vport_state, in); } @@ -96,7 +91,6 @@ static int mlx5_esw_adj_vport_create(struct mlx5_eswitch *esw, u16 vhca_id, if (err) goto acl_ns_remove; - mlx5_esw_adj_vport_modify(esw->dev, vport_num, MLX5_ADJ_VPORT_CONNECT); return 0; acl_ns_remove: @@ -117,8 +111,7 @@ static void mlx5_esw_adj_vport_destroy(struct mlx5_eswitch *esw, esw_debug(esw->dev, "Destroying adjacent vport %d for vhca_id 0x%x\n", vport_num, vport->vhca_id); - mlx5_esw_adj_vport_modify(esw->dev, vport_num, - MLX5_ADJ_VPORT_DISCONNECT); + mlx5_esw_offloads_rep_remove(esw, vport); mlx5_fs_vport_egress_acl_ns_remove(esw->dev->priv.steering, vport->index); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h index 16eb99aba2a7..beaec450a734 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h @@ -264,6 +264,9 @@ struct mlx5_eswitch_fdb { struct offloads_fdb { struct mlx5_flow_namespace *ns; + struct mlx5_flow_table *drop_root; + struct mlx5_flow_handle *drop_root_rule; + struct mlx5_fc *drop_root_fc; struct mlx5_flow_table *tc_miss_table; struct mlx5_flow_table *slow_fdb; struct mlx5_flow_group *send_to_vport_grp; @@ -392,6 +395,7 @@ struct mlx5_eswitch { struct mlx5_esw_offload offloads; u32 last_vport_idx; int mode; + bool offloads_inactive; u16 manager_vport; u16 first_host_vport; u8 num_peers; @@ -634,6 +638,8 @@ const u32 *mlx5_esw_query_functions(struct mlx5_core_dev *dev); void mlx5_esw_adjacent_vhcas_setup(struct mlx5_eswitch *esw); void mlx5_esw_adjacent_vhcas_cleanup(struct mlx5_eswitch *esw); +int mlx5_esw_adj_vport_modify(struct mlx5_core_dev *dev, u16 vport, + bool connect); #define MLX5_DEBUG_ESWITCH_MASK BIT(3) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c index 4092ea29c630..0b1a180ef238 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -1577,6 +1577,7 @@ esw_chains_create(struct mlx5_eswitch *esw, struct mlx5_flow_table *miss_fdb) attr.max_grp_num = esw->params.large_group_num; attr.default_ft = miss_fdb; attr.mapping = esw->offloads.reg_c0_obj_pool; + attr.fs_base_prio = FDB_BYPASS_PATH; chains = mlx5_chains_create(dev, &attr); if (IS_ERR(chains)) { @@ -2355,6 +2356,131 @@ static void esw_mode_change(struct mlx5_eswitch *esw, u16 mode) mlx5_devcom_comp_unlock(esw->dev->priv.hca_devcom_comp); } +static void mlx5_esw_fdb_drop_destroy(struct mlx5_eswitch *esw) +{ + if (!esw->fdb_table.offloads.drop_root) + return; + + esw_debug(esw->dev, "Destroying FDB drop root table %#x fc %#x\n", + esw->fdb_table.offloads.drop_root->id, + esw->fdb_table.offloads.drop_root_fc->id); + mlx5_del_flow_rules(esw->fdb_table.offloads.drop_root_rule); + /* Don't free flow counter here, can be reused on a later activation */ + mlx5_destroy_flow_table(esw->fdb_table.offloads.drop_root); + esw->fdb_table.offloads.drop_root_rule = NULL; + esw->fdb_table.offloads.drop_root = NULL; +} + +static int mlx5_esw_fdb_drop_create(struct mlx5_eswitch *esw) +{ + struct mlx5_flow_destination drop_fc_dst = {}; + struct mlx5_flow_table_attr ft_attr = {}; + struct mlx5_flow_destination *dst = NULL; + struct mlx5_core_dev *dev = esw->dev; + struct mlx5_flow_namespace *root_ns; + struct mlx5_flow_act flow_act = {}; + struct mlx5_flow_handle *flow_rule; + struct mlx5_flow_table *table; + int err = 0, dst_num = 0; + + if (esw->fdb_table.offloads.drop_root) + return 0; + + root_ns = esw->fdb_table.offloads.ns; + + ft_attr.prio = FDB_DROP_ROOT; + ft_attr.max_fte = 1; + ft_attr.autogroup.max_num_groups = 1; + table = mlx5_create_auto_grouped_flow_table(root_ns, &ft_attr); + if (IS_ERR(table)) { + esw_warn(dev, "Failed to create fdb drop root table, err %pe\n", + table); + return PTR_ERR(table); + } + + /* Drop FC reusable, create once on first deactivation of FDB */ + if (!esw->fdb_table.offloads.drop_root_fc) { + struct mlx5_fc *counter = mlx5_fc_create(dev, 0); + + err = PTR_ERR_OR_ZERO(counter); + if (err) + esw_warn(esw->dev, "create fdb drop fc err %d\n", err); + else + esw->fdb_table.offloads.drop_root_fc = counter; + } + + flow_act.action = MLX5_FLOW_CONTEXT_ACTION_DROP; + + if (esw->fdb_table.offloads.drop_root_fc) { + flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_COUNT; + drop_fc_dst.type = MLX5_FLOW_DESTINATION_TYPE_COUNTER; + drop_fc_dst.counter = esw->fdb_table.offloads.drop_root_fc; + dst = &drop_fc_dst; + dst_num++; + } + + flow_rule = mlx5_add_flow_rules(table, NULL, &flow_act, dst, dst_num); + err = PTR_ERR_OR_ZERO(flow_rule); + if (err) { + esw_warn(esw->dev, + "fs offloads: Failed to add vport rx drop rule err %d\n", + err); + goto err_flow_rule; + } + + esw->fdb_table.offloads.drop_root = table; + esw->fdb_table.offloads.drop_root_rule = flow_rule; + esw_debug(esw->dev, "Created FDB drop root table %#x fc %#x\n", + table->id, dst ? dst->counter->id : 0); + return 0; + +err_flow_rule: + /* no need to free drop fc, esw_offloads_steering_cleanup will do it */ + mlx5_destroy_flow_table(table); + return err; +} + +static void mlx5_esw_fdb_active(struct mlx5_eswitch *esw) +{ + struct mlx5_vport *vport; + unsigned long i; + + mlx5_esw_fdb_drop_destroy(esw); + mlx5_mpfs_enable(esw->dev); + + mlx5_esw_for_each_vf_vport(esw, i, vport, U16_MAX) { + if (!vport->adjacent) + continue; + esw_debug(esw->dev, "Connecting vport %d to eswitch\n", + vport->vport); + mlx5_esw_adj_vport_modify(esw->dev, vport->vport, true); + } + + esw->offloads_inactive = false; + esw_warn(esw->dev, "MPFS/FDB active\n"); +} + +static void mlx5_esw_fdb_inactive(struct mlx5_eswitch *esw) +{ + struct mlx5_vport *vport; + unsigned long i; + + mlx5_mpfs_disable(esw->dev); + mlx5_esw_fdb_drop_create(esw); + + mlx5_esw_for_each_vf_vport(esw, i, vport, U16_MAX) { + if (!vport->adjacent) + continue; + esw_debug(esw->dev, "Disconnecting vport %u from eswitch\n", + vport->vport); + + mlx5_esw_adj_vport_modify(esw->dev, vport->vport, false); + } + + esw->offloads_inactive = true; + esw_warn(esw->dev, "MPFS/FDB inactive\n"); +} + static int esw_offloads_start(struct mlx5_eswitch *esw, struct netlink_ext_ack *extack) { @@ -3438,6 +3564,10 @@ static int esw_offloads_steering_init(struct mlx5_eswitch *esw) static void esw_offloads_steering_cleanup(struct mlx5_eswitch *esw) { + mlx5_esw_fdb_drop_destroy(esw); + if (esw->fdb_table.offloads.drop_root_fc) + mlx5_fc_destroy(esw->dev, esw->fdb_table.offloads.drop_root_fc); + esw->fdb_table.offloads.drop_root_fc = NULL; esw_destroy_vport_rx_drop_rule(esw); esw_destroy_vport_rx_drop_group(esw); esw_destroy_vport_rx_group(esw); @@ -3600,6 +3730,11 @@ int esw_offloads_enable(struct mlx5_eswitch *esw) if (err) goto err_steering_init; + if (esw->offloads_inactive) + mlx5_esw_fdb_inactive(esw); + else + mlx5_esw_fdb_active(esw); + /* Representor will control the vport link state */ mlx5_esw_for_each_vf_vport(esw, i, vport, esw->esw_funcs.num_vfs) vport->info.link_state = MLX5_VPORT_ADMIN_STATE_DOWN; @@ -3666,6 +3801,9 @@ void esw_offloads_disable(struct mlx5_eswitch *esw) esw_offloads_metadata_uninit(esw); mlx5_rdma_disable_roce(esw->dev); mlx5_esw_adjacent_vhcas_cleanup(esw); + /* must be done after vhcas cleanup to avoid adjacent vports connect */ + if (esw->offloads_inactive) + mlx5_esw_fdb_active(esw); /* legacy mode always active */ mutex_destroy(&esw->offloads.termtbl_mutex); } @@ -3676,6 +3814,7 @@ static int esw_mode_from_devlink(u16 mode, u16 *mlx5_mode) *mlx5_mode = MLX5_ESWITCH_LEGACY; break; case DEVLINK_ESWITCH_MODE_SWITCHDEV: + case DEVLINK_ESWITCH_MODE_SWITCHDEV_INACTIVE: *mlx5_mode = MLX5_ESWITCH_OFFLOADS; break; default: @@ -3685,14 +3824,17 @@ static int esw_mode_from_devlink(u16 mode, u16 *mlx5_mode) return 0; } -static int esw_mode_to_devlink(u16 mlx5_mode, u16 *mode) +static int esw_mode_to_devlink(struct mlx5_eswitch *esw, u16 *mode) { - switch (mlx5_mode) { + switch (esw->mode) { case MLX5_ESWITCH_LEGACY: *mode = DEVLINK_ESWITCH_MODE_LEGACY; break; case MLX5_ESWITCH_OFFLOADS: - *mode = DEVLINK_ESWITCH_MODE_SWITCHDEV; + if (esw->offloads_inactive) + *mode = DEVLINK_ESWITCH_MODE_SWITCHDEV_INACTIVE; + else + *mode = DEVLINK_ESWITCH_MODE_SWITCHDEV; break; default: return -EINVAL; @@ -3798,6 +3940,45 @@ static bool mlx5_devlink_netdev_netns_immutable_set(struct devlink *devlink, return ret; } +/* Returns true when only changing between active and inactive switchdev mode */ +static bool mlx5_devlink_switchdev_active_mode_change(struct mlx5_eswitch *esw, + u16 devlink_mode) +{ + /* current mode is not switchdev */ + if (esw->mode != MLX5_ESWITCH_OFFLOADS) + return false; + + /* new mode is not switchdev */ + if (devlink_mode != DEVLINK_ESWITCH_MODE_SWITCHDEV && + devlink_mode != DEVLINK_ESWITCH_MODE_SWITCHDEV_INACTIVE) + return false; + + /* already inactive: no change in current state */ + if (devlink_mode == DEVLINK_ESWITCH_MODE_SWITCHDEV_INACTIVE && + esw->offloads_inactive) + return false; + + /* already active: no change in current state */ + if (devlink_mode == DEVLINK_ESWITCH_MODE_SWITCHDEV && + !esw->offloads_inactive) + return false; + + down_write(&esw->mode_lock); + esw->offloads_inactive = !esw->offloads_inactive; + esw->eswitch_operation_in_progress = true; + up_write(&esw->mode_lock); + + if (esw->offloads_inactive) + mlx5_esw_fdb_inactive(esw); + else + mlx5_esw_fdb_active(esw); + + down_write(&esw->mode_lock); + esw->eswitch_operation_in_progress = false; + up_write(&esw->mode_lock); + return true; +} + int mlx5_devlink_eswitch_mode_set(struct devlink *devlink, u16 mode, struct netlink_ext_ack *extack) { @@ -3812,12 +3993,16 @@ int mlx5_devlink_eswitch_mode_set(struct devlink *devlink, u16 mode, if (esw_mode_from_devlink(mode, &mlx5_mode)) return -EINVAL; - if (mode == DEVLINK_ESWITCH_MODE_SWITCHDEV && mlx5_get_sd(esw->dev)) { + if (mlx5_mode == MLX5_ESWITCH_OFFLOADS && mlx5_get_sd(esw->dev)) { NL_SET_ERR_MSG_MOD(extack, "Can't change E-Switch mode to switchdev when multi-PF netdev (Socket Direct) is configured."); return -EPERM; } + /* Avoid try_lock, active/inactive mode change is not restricted */ + if (mlx5_devlink_switchdev_active_mode_change(esw, mode)) + return 0; + mlx5_lag_disable_change(esw->dev); err = mlx5_esw_try_lock(esw); if (err < 0) { @@ -3840,7 +4025,7 @@ int mlx5_devlink_eswitch_mode_set(struct devlink *devlink, u16 mode, esw->eswitch_operation_in_progress = true; up_write(&esw->mode_lock); - if (mode == DEVLINK_ESWITCH_MODE_SWITCHDEV && + if (mlx5_mode == MLX5_ESWITCH_OFFLOADS && !mlx5_devlink_netdev_netns_immutable_set(devlink, true)) { NL_SET_ERR_MSG_MOD(extack, "Can't change E-Switch mode to switchdev when netdev net namespace has diverged from the devlink's."); @@ -3848,25 +4033,27 @@ int mlx5_devlink_eswitch_mode_set(struct devlink *devlink, u16 mode, goto skip; } - if (mode == DEVLINK_ESWITCH_MODE_LEGACY) + if (mlx5_mode == MLX5_ESWITCH_LEGACY) esw->dev->priv.flags |= MLX5_PRIV_FLAGS_SWITCH_LEGACY; mlx5_eswitch_disable_locked(esw); - if (mode == DEVLINK_ESWITCH_MODE_SWITCHDEV) { + if (mlx5_mode == MLX5_ESWITCH_OFFLOADS) { if (mlx5_devlink_trap_get_num_active(esw->dev)) { NL_SET_ERR_MSG_MOD(extack, "Can't change mode while devlink traps are active"); err = -EOPNOTSUPP; goto skip; } + esw->offloads_inactive = + (mode == DEVLINK_ESWITCH_MODE_SWITCHDEV_INACTIVE); err = esw_offloads_start(esw, extack); - } else if (mode == DEVLINK_ESWITCH_MODE_LEGACY) { + } else if (mlx5_mode == MLX5_ESWITCH_LEGACY) { err = esw_offloads_stop(esw, extack); } else { err = -EINVAL; } skip: - if (mode == DEVLINK_ESWITCH_MODE_SWITCHDEV && err) + if (mlx5_mode == MLX5_ESWITCH_OFFLOADS && err) mlx5_devlink_netdev_netns_immutable_set(devlink, false); down_write(&esw->mode_lock); esw->eswitch_operation_in_progress = false; @@ -3885,7 +4072,7 @@ int mlx5_devlink_eswitch_mode_get(struct devlink *devlink, u16 *mode) if (IS_ERR(esw)) return PTR_ERR(esw); - return esw_mode_to_devlink(esw->mode, mode); + return esw_mode_to_devlink(esw, mode); } static int mlx5_esw_vports_inline_set(struct mlx5_eswitch *esw, u8 mlx5_mode, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c index 2db3ffb0a2b2..2ca3bddbdf05 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c @@ -3520,6 +3520,11 @@ static int init_fdb_root_ns(struct mlx5_flow_steering *steering) if (!steering->fdb_root_ns) return -ENOMEM; + maj_prio = fs_create_prio(&steering->fdb_root_ns->ns, FDB_DROP_ROOT, 1); + err = PTR_ERR_OR_ZERO(maj_prio); + if (err) + goto out_err; + err = create_fdb_bypass(steering); if (err) goto out_err; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/mpfs.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/mpfs.c index 99fb7a53add0..4a88a42ae4f7 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lib/mpfs.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/mpfs.c @@ -167,7 +167,7 @@ int mlx5_mpfs_add_mac(struct mlx5_core_dev *dev, u8 *mac) if (err) goto free_l2table_index; mlx5_core_dbg(dev, "MPFS entry %pM, set @index (%d)\n", - l2addr->node.addr, l2addr->index); + l2addr->node.addr, index); } l2addr->index = index; diff --git a/include/linux/mlx5/fs.h b/include/linux/mlx5/fs.h index 6ac76a0c3827..7bf2449c53b2 100644 --- a/include/linux/mlx5/fs.h +++ b/include/linux/mlx5/fs.h @@ -116,6 +116,7 @@ enum mlx5_flow_namespace_type { }; enum { + FDB_DROP_ROOT, FDB_BYPASS_PATH, FDB_CRYPTO_INGRESS, FDB_TC_OFFLOAD, From 34bff6f03c131ee7eb2ddbd2dbaa01c6497f0f8a Mon Sep 17 00:00:00 2001 From: Nathan Chancellor Date: Mon, 10 Nov 2025 13:55:34 -0700 Subject: [PATCH 0721/1381] net: netcp: ethss: Fix type of first parameter in hwtstamp stubs When building without CONFIG_TI_CPTS, there are a series of errors from -Wincompatible-pointer-types: drivers/net/ethernet/ti/netcp_ethss.c:3831:27: error: initialization of 'int (*)(void *, struct kernel_hwtstamp_config *)' from incompatible pointer type 'int (*)(struct gbe_intf *, struct kernel_hwtstamp_config *)' [-Wincompatible-pointer-types] 3831 | .hwtstamp_get = gbe_hwtstamp_get, | ^~~~~~~~~~~~~~~~ drivers/net/ethernet/ti/netcp_ethss.c:3831:27: note: (near initialization for 'gbe_module.hwtstamp_get') drivers/net/ethernet/ti/netcp_ethss.c:2758:19: note: 'gbe_hwtstamp_get' declared here 2758 | static inline int gbe_hwtstamp_get(struct gbe_intf *gbe_intf, | ^~~~~~~~~~~~~~~~ drivers/net/ethernet/ti/netcp_ethss.c:3832:27: error: initialization of 'int (*)(void *, struct kernel_hwtstamp_config *, struct netlink_ext_ack *)' from incompatible pointer type 'int (*)(struct gbe_intf *, struct kernel_hwtstamp_config *, struct netlink_ext_ack *)' [-Wincompatible-pointer-types] 3832 | .hwtstamp_set = gbe_hwtstamp_set, | ^~~~~~~~~~~~~~~~ drivers/net/ethernet/ti/netcp_ethss.c:3832:27: note: (near initialization for 'gbe_module.hwtstamp_set') drivers/net/ethernet/ti/netcp_ethss.c:2764:19: note: 'gbe_hwtstamp_set' declared here 2764 | static inline int gbe_hwtstamp_set(struct gbe_intf *gbe_intf, | ^~~~~~~~~~~~~~~~ In a recent conversion to ndo_hwtstamp, the type of the first parameter was updated for the CONFIG_TI_CPTS=y implementations of gbe_hwtstamp_get() and gbe_hwtstamp_set() but not the CONFIG_TI_CPTS=n ones. Update the type of the first parameter in the CONFIG_TI_CPTS=n stubs to resolve the errors. Fixes: 3f02b8272557 ("ti: netcp: convert to ndo_hwtstamp callbacks") Reviewed-by: Vadim Fedorenko Signed-off-by: Nathan Chancellor Link: https://patch.msgid.link/20251110-netcp_ethss-fix-cpts-stubs-clang-wifpts-v2-1-aa6204ec1f43@kernel.org Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/ti/netcp_ethss.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/ti/netcp_ethss.c b/drivers/net/ethernet/ti/netcp_ethss.c index 0ae44112812c..4f6cc6cd1f03 100644 --- a/drivers/net/ethernet/ti/netcp_ethss.c +++ b/drivers/net/ethernet/ti/netcp_ethss.c @@ -2755,13 +2755,13 @@ static inline void gbe_unregister_cpts(struct gbe_priv *gbe_dev) { } -static inline int gbe_hwtstamp_get(struct gbe_intf *gbe_intf, +static inline int gbe_hwtstamp_get(void *intf_priv, struct kernel_hwtstamp_config *cfg) { return -EOPNOTSUPP; } -static inline int gbe_hwtstamp_set(struct gbe_intf *gbe_intf, +static inline int gbe_hwtstamp_set(void *intf_priv, struct kernel_hwtstamp_config *cfg, struct netlink_ext_ack *extack) { From 12f42597ab148753959bc466c0c66abcf10e4267 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Mon, 10 Nov 2025 14:42:43 +0000 Subject: [PATCH 0722/1381] net: stmmac: meson8b: use PHY_INTF_SEL_x Use PHY_INTF_SEL_x definitions for phy_intf_sel bitfield. Reviewed-by: Martin Blumenstingl Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vIT6R-0000000DpPL-0Nli@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/dwmac-meson8b.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-meson8b.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-meson8b.c index a50782994b97..f485b9b858bf 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-meson8b.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-meson8b.c @@ -26,8 +26,8 @@ #define PRG_ETH0_RGMII_MODE BIT(0) #define PRG_ETH0_EXT_PHY_MODE_MASK GENMASK(2, 0) -#define PRG_ETH0_EXT_RGMII_MODE 1 -#define PRG_ETH0_EXT_RMII_MODE 4 +#define PRG_ETH0_EXT_RGMII_MODE PHY_INTF_SEL_RGMII +#define PRG_ETH0_EXT_RMII_MODE PHY_INTF_SEL_RMII /* mux to choose between fclk_div2 (bit unset) and mpll2 (bit set) */ #define PRG_ETH0_CLK_M250_SEL_MASK GENMASK(4, 4) From 52d639da6feea3c8787a60b1efb534bd48ea053d Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Mon, 10 Nov 2025 14:42:48 +0000 Subject: [PATCH 0723/1381] net: stmmac: meson8b: use phy_intf_sel directly Rearrange meson_axg_set_phy_mode() to use phy_intf_sel directly, converting it to the register field for meson8b_dwmac_mask_bits(). Reviewed-by: Martin Blumenstingl Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vIT6W-0000000DpPR-0tby@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../net/ethernet/stmicro/stmmac/dwmac-meson8b.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-meson8b.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-meson8b.c index f485b9b858bf..865cd6166134 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-meson8b.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-meson8b.c @@ -26,8 +26,6 @@ #define PRG_ETH0_RGMII_MODE BIT(0) #define PRG_ETH0_EXT_PHY_MODE_MASK GENMASK(2, 0) -#define PRG_ETH0_EXT_RGMII_MODE PHY_INTF_SEL_RGMII -#define PRG_ETH0_EXT_RMII_MODE PHY_INTF_SEL_RMII /* mux to choose between fclk_div2 (bit unset) and mpll2 (bit set) */ #define PRG_ETH0_CLK_M250_SEL_MASK GENMASK(4, 4) @@ -238,21 +236,19 @@ static int meson8b_set_phy_mode(struct meson8b_dwmac *dwmac) static int meson_axg_set_phy_mode(struct meson8b_dwmac *dwmac) { + int phy_intf_sel; + switch (dwmac->phy_mode) { case PHY_INTERFACE_MODE_RGMII: case PHY_INTERFACE_MODE_RGMII_RXID: case PHY_INTERFACE_MODE_RGMII_ID: case PHY_INTERFACE_MODE_RGMII_TXID: /* enable RGMII mode */ - meson8b_dwmac_mask_bits(dwmac, PRG_ETH0, - PRG_ETH0_EXT_PHY_MODE_MASK, - PRG_ETH0_EXT_RGMII_MODE); + phy_intf_sel = PHY_INTF_SEL_RGMII; break; case PHY_INTERFACE_MODE_RMII: /* disable RGMII mode -> enables RMII mode */ - meson8b_dwmac_mask_bits(dwmac, PRG_ETH0, - PRG_ETH0_EXT_PHY_MODE_MASK, - PRG_ETH0_EXT_RMII_MODE); + phy_intf_sel = PHY_INTF_SEL_RMII; break; default: dev_err(dwmac->dev, "fail to set phy-mode %s\n", @@ -260,6 +256,10 @@ static int meson_axg_set_phy_mode(struct meson8b_dwmac *dwmac) return -EINVAL; } + meson8b_dwmac_mask_bits(dwmac, PRG_ETH0, PRG_ETH0_EXT_PHY_MODE_MASK, + FIELD_PREP(PRG_ETH0_EXT_PHY_MODE_MASK, + phy_intf_sel)); + return 0; } From da3d1501235df39d9f2db092ac8592eb6f7c52b1 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Mon, 10 Nov 2025 14:42:53 +0000 Subject: [PATCH 0724/1381] net: stmmac: meson8b: use stmmac_get_phy_intf_sel() Use stmmac_get_phy_intf_sel() to decode the PHY interface mode to the phy_intf_sel value, validate the result and use that to set the control register to select the operating mode for the DWMAC core. Reviewed-by: Martin Blumenstingl Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vIT6b-0000000DpPX-1LQ0@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../ethernet/stmicro/stmmac/dwmac-meson8b.c | 18 ++++-------------- 1 file changed, 4 insertions(+), 14 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-meson8b.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-meson8b.c index 865cd6166134..e4d5c41294f4 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-meson8b.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-meson8b.c @@ -238,22 +238,12 @@ static int meson_axg_set_phy_mode(struct meson8b_dwmac *dwmac) { int phy_intf_sel; - switch (dwmac->phy_mode) { - case PHY_INTERFACE_MODE_RGMII: - case PHY_INTERFACE_MODE_RGMII_RXID: - case PHY_INTERFACE_MODE_RGMII_ID: - case PHY_INTERFACE_MODE_RGMII_TXID: - /* enable RGMII mode */ - phy_intf_sel = PHY_INTF_SEL_RGMII; - break; - case PHY_INTERFACE_MODE_RMII: - /* disable RGMII mode -> enables RMII mode */ - phy_intf_sel = PHY_INTF_SEL_RMII; - break; - default: + phy_intf_sel = stmmac_get_phy_intf_sel(dwmac->phy_mode); + if (phy_intf_sel != PHY_INTF_SEL_RGMII && + phy_intf_sel != PHY_INTF_SEL_RMII) { dev_err(dwmac->dev, "fail to set phy-mode %s\n", phy_modes(dwmac->phy_mode)); - return -EINVAL; + return phy_intf_sel < 0 ? phy_intf_sel : -EINVAL; } meson8b_dwmac_mask_bits(dwmac, PRG_ETH0, PRG_ETH0_EXT_PHY_MODE_MASK, From fc6aa0e470e092873eddb213d996a8beee86bf4d Mon Sep 17 00:00:00 2001 From: Robert Marko Date: Mon, 10 Nov 2025 13:42:53 +0100 Subject: [PATCH 0725/1381] net: sparx5/lan969x: populate netdev of_node Populate of_node for the port netdevs, to make the individual ports of_nodes available in sysfs. Signed-off-by: Robert Marko Link: https://patch.msgid.link/20251110124342.199216-1-robert.marko@sartura.hr Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/microchip/sparx5/sparx5_main.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/ethernet/microchip/sparx5/sparx5_main.c b/drivers/net/ethernet/microchip/sparx5/sparx5_main.c index 40b1bfc600a7..582145713cfd 100644 --- a/drivers/net/ethernet/microchip/sparx5/sparx5_main.c +++ b/drivers/net/ethernet/microchip/sparx5/sparx5_main.c @@ -395,6 +395,8 @@ static int sparx5_create_port(struct sparx5 *sparx5, spx5_port->phylink = phylink; + spx5_port->ndev->dev.of_node = spx5_port->of_node; + return 0; } From 68ec5df1d8946dd805d6ab67666a38331223f3a1 Mon Sep 17 00:00:00 2001 From: Bagas Sanjaya Date: Mon, 3 Nov 2025 08:50:22 +0700 Subject: [PATCH 0726/1381] Documentation: xfrm_device: Wrap iproute2 snippets in literal code block iproute2 snippets (ip x) are shown in long-running definition lists instead. Format them as literal code blocks that do the semantic job better. Reviewed-by: Randy Dunlap Tested-by: Randy Dunlap Signed-off-by: Bagas Sanjaya Signed-off-by: Steffen Klassert --- Documentation/networking/xfrm_device.rst | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/Documentation/networking/xfrm_device.rst b/Documentation/networking/xfrm_device.rst index 122204da0fff..7a13075b5bf0 100644 --- a/Documentation/networking/xfrm_device.rst +++ b/Documentation/networking/xfrm_device.rst @@ -34,7 +34,7 @@ Right now, there are two types of hardware offload that kernel supports. Userland access to the offload is typically through a system such as libreswan or KAME/raccoon, but the iproute2 'ip xfrm' command set can be handy when experimenting. An example command might look something -like this for crypto offload: +like this for crypto offload:: ip x s add proto esp dst 14.0.0.70 src 14.0.0.52 spi 0x07 mode transport \ reqid 0x07 replay-window 32 \ @@ -42,7 +42,7 @@ like this for crypto offload: sel src 14.0.0.52/24 dst 14.0.0.70/24 proto tcp \ offload dev eth4 dir in -and for packet offload +and for packet offload:: ip x s add proto esp dst 14.0.0.70 src 14.0.0.52 spi 0x07 mode transport \ reqid 0x07 replay-window 32 \ From 340e2a73866557ad1f902273d534e9a81efccbd2 Mon Sep 17 00:00:00 2001 From: Bagas Sanjaya Date: Mon, 3 Nov 2025 08:50:23 +0700 Subject: [PATCH 0727/1381] Documentation: xfrm_device: Use numbered list for offloading steps Format xfrm offloading steps as numbered list. Reviewed-by: Randy Dunlap Tested-by: Randy Dunlap Signed-off-by: Bagas Sanjaya Signed-off-by: Steffen Klassert --- Documentation/networking/xfrm_device.rst | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/Documentation/networking/xfrm_device.rst b/Documentation/networking/xfrm_device.rst index 7a13075b5bf0..86db3f42552d 100644 --- a/Documentation/networking/xfrm_device.rst +++ b/Documentation/networking/xfrm_device.rst @@ -153,26 +153,26 @@ the packet's skb. At this point the data should be decrypted but the IPsec headers are still in the packet data; they are removed later up the stack in xfrm_input(). - find and hold the SA that was used to the Rx skb:: +1. Find and hold the SA that was used to the Rx skb:: - get spi, protocol, and destination IP from packet headers + /* get spi, protocol, and destination IP from packet headers */ xs = find xs from (spi, protocol, dest_IP) xfrm_state_hold(xs); - store the state information into the skb:: +2. Store the state information into the skb:: sp = secpath_set(skb); if (!sp) return; sp->xvec[sp->len++] = xs; sp->olen++; - indicate the success and/or error status of the offload:: +3. Indicate the success and/or error status of the offload:: xo = xfrm_offload(skb); xo->flags = CRYPTO_DONE; xo->status = crypto_status; - hand the packet to napi_gro_receive() as usual +4. Hand the packet to napi_gro_receive() as usual. In ESN mode, xdo_dev_state_advance_esn() is called from xfrm_replay_advance_esn() for RX, and xfrm_replay_overflow_offload_esn for TX. From 840188d276a34e8883fd4b64a4a39f7b3eec0f28 Mon Sep 17 00:00:00 2001 From: Bagas Sanjaya Date: Mon, 3 Nov 2025 08:50:24 +0700 Subject: [PATCH 0728/1381] Documentation: xfrm_device: Separate hardware offload sublists Sublists of hardware offload type lists are rendered in combined paragraph due to lack of separator from their parent list. Add it. Reviewed-by: Randy Dunlap Tested-by: Randy Dunlap Signed-off-by: Bagas Sanjaya Signed-off-by: Steffen Klassert --- Documentation/networking/xfrm_device.rst | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/Documentation/networking/xfrm_device.rst b/Documentation/networking/xfrm_device.rst index 86db3f42552d..b0d85a5f57d1 100644 --- a/Documentation/networking/xfrm_device.rst +++ b/Documentation/networking/xfrm_device.rst @@ -20,11 +20,15 @@ can radically increase throughput and decrease CPU utilization. The XFRM Device interface allows NIC drivers to offer to the stack access to the hardware offload. -Right now, there are two types of hardware offload that kernel supports. +Right now, there are two types of hardware offload that kernel supports: + * IPsec crypto offload: + * NIC performs encrypt/decrypt * Kernel does everything else + * IPsec packet offload: + * NIC performs encrypt/decrypt * NIC does encapsulation * Kernel and NIC have SA and policy in-sync From a397b259c17379f74158b86dd7fc4e3a7d53fc57 Mon Sep 17 00:00:00 2001 From: Bagas Sanjaya Date: Mon, 3 Nov 2025 08:50:25 +0700 Subject: [PATCH 0729/1381] Documentation: xfrm_sync: Properly reindent list text List texts are currently aligned at the start of column, rather than after the list marker. Reindent them. Reviewed-by: Randy Dunlap Tested-by: Randy Dunlap Signed-off-by: Bagas Sanjaya Signed-off-by: Steffen Klassert --- Documentation/networking/xfrm_sync.rst | 77 +++++++++++++------------- 1 file changed, 40 insertions(+), 37 deletions(-) diff --git a/Documentation/networking/xfrm_sync.rst b/Documentation/networking/xfrm_sync.rst index 6246503ceab2..c811c3edfa57 100644 --- a/Documentation/networking/xfrm_sync.rst +++ b/Documentation/networking/xfrm_sync.rst @@ -88,23 +88,23 @@ to get notified of these events. a) byte value (XFRMA_LTIME_VAL) -This TLV carries the running/current counter for byte lifetime since -last event. + This TLV carries the running/current counter for byte lifetime since + last event. -b)replay value (XFRMA_REPLAY_VAL) +b) replay value (XFRMA_REPLAY_VAL) -This TLV carries the running/current counter for replay sequence since -last event. + This TLV carries the running/current counter for replay sequence since + last event. -c)replay threshold (XFRMA_REPLAY_THRESH) +c) replay threshold (XFRMA_REPLAY_THRESH) -This TLV carries the threshold being used by the kernel to trigger events -when the replay sequence is exceeded. + This TLV carries the threshold being used by the kernel to trigger events + when the replay sequence is exceeded. d) expiry timer (XFRMA_ETIMER_THRESH) -This is a timer value in milliseconds which is used as the nagle -value to rate limit the events. + This is a timer value in milliseconds which is used as the nagle + value to rate limit the events. 3) Default configurations for the parameters: --------------------------------------------- @@ -121,12 +121,14 @@ in case they are not specified. the two sysctls/proc entries are: a) /proc/sys/net/core/sysctl_xfrm_aevent_etime -used to provide default values for the XFRMA_ETIMER_THRESH in incremental -units of time of 100ms. The default is 10 (1 second) + + Used to provide default values for the XFRMA_ETIMER_THRESH in incremental + units of time of 100ms. The default is 10 (1 second) b) /proc/sys/net/core/sysctl_xfrm_aevent_rseqth -used to provide default values for XFRMA_REPLAY_THRESH parameter -in incremental packet count. The default is two packets. + + Used to provide default values for XFRMA_REPLAY_THRESH parameter + in incremental packet count. The default is two packets. 4) Message types ---------------- @@ -134,42 +136,43 @@ in incremental packet count. The default is two packets. a) XFRM_MSG_GETAE issued by user-->kernel. XFRM_MSG_GETAE does not carry any TLVs. -The response is a XFRM_MSG_NEWAE which is formatted based on what -XFRM_MSG_GETAE queried for. + The response is a XFRM_MSG_NEWAE which is formatted based on what + XFRM_MSG_GETAE queried for. -The response will always have XFRMA_LTIME_VAL and XFRMA_REPLAY_VAL TLVs. -* if XFRM_AE_RTHR flag is set, then XFRMA_REPLAY_THRESH is also retrieved -* if XFRM_AE_ETHR flag is set, then XFRMA_ETIMER_THRESH is also retrieved + The response will always have XFRMA_LTIME_VAL and XFRMA_REPLAY_VAL TLVs. + + * if XFRM_AE_RTHR flag is set, then XFRMA_REPLAY_THRESH is also retrieved + * if XFRM_AE_ETHR flag is set, then XFRMA_ETIMER_THRESH is also retrieved b) XFRM_MSG_NEWAE is issued by either user space to configure or kernel to announce events or respond to a XFRM_MSG_GETAE. -i) user --> kernel to configure a specific SA. + i) user --> kernel to configure a specific SA. -any of the values or threshold parameters can be updated by passing the -appropriate TLV. + any of the values or threshold parameters can be updated by passing the + appropriate TLV. -A response is issued back to the sender in user space to indicate success -or failure. + A response is issued back to the sender in user space to indicate success + or failure. -In the case of success, additionally an event with -XFRM_MSG_NEWAE is also issued to any listeners as described in iii). + In the case of success, additionally an event with + XFRM_MSG_NEWAE is also issued to any listeners as described in iii). -ii) kernel->user direction as a response to XFRM_MSG_GETAE + ii) kernel->user direction as a response to XFRM_MSG_GETAE -The response will always have XFRMA_LTIME_VAL and XFRMA_REPLAY_VAL TLVs. + The response will always have XFRMA_LTIME_VAL and XFRMA_REPLAY_VAL TLVs. -The threshold TLVs will be included if explicitly requested in -the XFRM_MSG_GETAE message. + The threshold TLVs will be included if explicitly requested in + the XFRM_MSG_GETAE message. -iii) kernel->user to report as event if someone sets any values or - thresholds for an SA using XFRM_MSG_NEWAE (as described in #i above). - In such a case XFRM_AE_CU flag is set to inform the user that - the change happened as a result of an update. - The message will always have XFRMA_LTIME_VAL and XFRMA_REPLAY_VAL TLVs. + iii) kernel->user to report as event if someone sets any values or + thresholds for an SA using XFRM_MSG_NEWAE (as described in #i above). + In such a case XFRM_AE_CU flag is set to inform the user that + the change happened as a result of an update. + The message will always have XFRMA_LTIME_VAL and XFRMA_REPLAY_VAL TLVs. -iv) kernel->user to report event when replay threshold or a timeout - is exceeded. + iv) kernel->user to report event when replay threshold or a timeout + is exceeded. In such a case either XFRM_AE_CR (replay exceeded) or XFRM_AE_CE (timeout happened) is set to inform the user what happened. From 01ad7831fbb28b3903fa3eb4e029f5a5d6690924 Mon Sep 17 00:00:00 2001 From: Bagas Sanjaya Date: Mon, 3 Nov 2025 08:50:26 +0700 Subject: [PATCH 0730/1381] Documentation: xfrm_sync: Trim excess section heading characters The first section "Message Structure" has excess underline, while the second and third one ("TLVS reflect the different parameters" and "Default configurations for the parameters") have trailing colon. Trim them. Reviewed-by: Randy Dunlap Tested-by: Randy Dunlap Suggested-by: Randy Dunlap Signed-off-by: Bagas Sanjaya Signed-off-by: Steffen Klassert --- Documentation/networking/xfrm_sync.rst | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/Documentation/networking/xfrm_sync.rst b/Documentation/networking/xfrm_sync.rst index c811c3edfa57..de4da4707037 100644 --- a/Documentation/networking/xfrm_sync.rst +++ b/Documentation/networking/xfrm_sync.rst @@ -36,7 +36,7 @@ is not driven by packet arrival. - the replay sequence for both inbound and outbound 1) Message Structure ----------------------- +-------------------- nlmsghdr:aevent_id:optional-TLVs. @@ -83,8 +83,8 @@ when going from kernel to user space) A program needs to subscribe to multicast group XFRMNLGRP_AEVENTS to get notified of these events. -2) TLVS reflect the different parameters: ------------------------------------------ +2) TLVS reflect the different parameters +---------------------------------------- a) byte value (XFRMA_LTIME_VAL) @@ -106,8 +106,8 @@ d) expiry timer (XFRMA_ETIMER_THRESH) This is a timer value in milliseconds which is used as the nagle value to rate limit the events. -3) Default configurations for the parameters: ---------------------------------------------- +3) Default configurations for the parameters +-------------------------------------------- By default these events should be turned off unless there is at least one listener registered to listen to the multicast From c08b786b8295fbbd1e8ac99320126fd5dd8b965c Mon Sep 17 00:00:00 2001 From: Bagas Sanjaya Date: Mon, 3 Nov 2025 08:50:27 +0700 Subject: [PATCH 0731/1381] Documentation: xfrm_sysctl: Trim trailing colon in section heading The sole section heading ("/proc/sys/net/core/xfrm_* Variables") has trailing colon. Trim it. Suggested-by: Randy Dunlap Signed-off-by: Bagas Sanjaya Reviewed-by: Randy Dunlap Tested-by: Randy Dunlap Signed-off-by: Steffen Klassert --- Documentation/networking/xfrm_sysctl.rst | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/Documentation/networking/xfrm_sysctl.rst b/Documentation/networking/xfrm_sysctl.rst index 47b9bbdd0179..7d0c4b17c0bd 100644 --- a/Documentation/networking/xfrm_sysctl.rst +++ b/Documentation/networking/xfrm_sysctl.rst @@ -4,8 +4,8 @@ XFRM Syscall ============ -/proc/sys/net/core/xfrm_* Variables: -==================================== +/proc/sys/net/core/xfrm_* Variables +=================================== xfrm_acq_expires - INTEGER default 30 - hard timeout in seconds for acquire requests From 7276e7ae569b1ce2ac2e3341107703216a290398 Mon Sep 17 00:00:00 2001 From: Bagas Sanjaya Date: Mon, 3 Nov 2025 08:50:28 +0700 Subject: [PATCH 0732/1381] Documentation: xfrm_sync: Number the fifth section Number the fifth section ("Exception to threshold settings") to be consistent with the rest of sections. Reviewed-by: Randy Dunlap Tested-by: Randy Dunlap Suggested-by: Randy Dunlap Signed-off-by: Bagas Sanjaya Signed-off-by: Steffen Klassert --- Documentation/networking/xfrm_sync.rst | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/Documentation/networking/xfrm_sync.rst b/Documentation/networking/xfrm_sync.rst index de4da4707037..112f7c102ad0 100644 --- a/Documentation/networking/xfrm_sync.rst +++ b/Documentation/networking/xfrm_sync.rst @@ -179,8 +179,8 @@ happened) is set to inform the user what happened. Note the two flags are mutually exclusive. The message will always have XFRMA_LTIME_VAL and XFRMA_REPLAY_VAL TLVs. -Exceptions to threshold settings --------------------------------- +5) Exceptions to threshold settings +----------------------------------- If you have an SA that is getting hit by traffic in bursts such that there is a period where the timer threshold expires with no packets From 03e23b18c720bdb628ccfbbb3faf4486c0413354 Mon Sep 17 00:00:00 2001 From: Bagas Sanjaya Date: Mon, 3 Nov 2025 08:50:29 +0700 Subject: [PATCH 0733/1381] net: Move XFRM documentation into its own subdirectory XFRM docs are currently reside in Documentation/networking directory, yet these are distinctive as a group of their own. Move them into xfrm subdirectory. Reviewed-by: Randy Dunlap Tested-by: Randy Dunlap Signed-off-by: Bagas Sanjaya Signed-off-by: Steffen Klassert --- Documentation/networking/index.rst | 5 +---- Documentation/networking/xfrm/index.rst | 13 +++++++++++++ Documentation/networking/{ => xfrm}/xfrm_device.rst | 0 Documentation/networking/{ => xfrm}/xfrm_proc.rst | 0 Documentation/networking/{ => xfrm}/xfrm_sync.rst | 6 +++--- Documentation/networking/{ => xfrm}/xfrm_sysctl.rst | 0 6 files changed, 17 insertions(+), 7 deletions(-) create mode 100644 Documentation/networking/xfrm/index.rst rename Documentation/networking/{ => xfrm}/xfrm_device.rst (100%) rename Documentation/networking/{ => xfrm}/xfrm_proc.rst (100%) rename Documentation/networking/{ => xfrm}/xfrm_sync.rst (99%) rename Documentation/networking/{ => xfrm}/xfrm_sysctl.rst (100%) diff --git a/Documentation/networking/index.rst b/Documentation/networking/index.rst index c775cababc8c..75db2251649b 100644 --- a/Documentation/networking/index.rst +++ b/Documentation/networking/index.rst @@ -131,10 +131,7 @@ Contents: vxlan x25 x25-iface - xfrm_device - xfrm_proc - xfrm_sync - xfrm_sysctl + xfrm/index xdp-rx-metadata xsk-tx-metadata diff --git a/Documentation/networking/xfrm/index.rst b/Documentation/networking/xfrm/index.rst new file mode 100644 index 000000000000..7d866da836fe --- /dev/null +++ b/Documentation/networking/xfrm/index.rst @@ -0,0 +1,13 @@ +.. SPDX-License-Identifier: GPL-2.0 + +============== +XFRM Framework +============== + +.. toctree:: + :maxdepth: 2 + + xfrm_device + xfrm_proc + xfrm_sync + xfrm_sysctl diff --git a/Documentation/networking/xfrm_device.rst b/Documentation/networking/xfrm/xfrm_device.rst similarity index 100% rename from Documentation/networking/xfrm_device.rst rename to Documentation/networking/xfrm/xfrm_device.rst diff --git a/Documentation/networking/xfrm_proc.rst b/Documentation/networking/xfrm/xfrm_proc.rst similarity index 100% rename from Documentation/networking/xfrm_proc.rst rename to Documentation/networking/xfrm/xfrm_proc.rst diff --git a/Documentation/networking/xfrm_sync.rst b/Documentation/networking/xfrm/xfrm_sync.rst similarity index 99% rename from Documentation/networking/xfrm_sync.rst rename to Documentation/networking/xfrm/xfrm_sync.rst index 112f7c102ad0..dfc2ec0df380 100644 --- a/Documentation/networking/xfrm_sync.rst +++ b/Documentation/networking/xfrm/xfrm_sync.rst @@ -1,8 +1,8 @@ .. SPDX-License-Identifier: GPL-2.0 -==== -XFRM -==== +========= +XFRM sync +========= The sync patches work is based on initial patches from Krisztian and others and additional patches diff --git a/Documentation/networking/xfrm_sysctl.rst b/Documentation/networking/xfrm/xfrm_sysctl.rst similarity index 100% rename from Documentation/networking/xfrm_sysctl.rst rename to Documentation/networking/xfrm/xfrm_sysctl.rst From 939ba8c5b81cbaf37781d7aa4849170860124a5e Mon Sep 17 00:00:00 2001 From: Bagas Sanjaya Date: Mon, 3 Nov 2025 08:50:30 +0700 Subject: [PATCH 0734/1381] MAINTAINERS: Add entry for XFRM documentation XFRM patches are supposed to be sent to maintainers under "NETWORKING [IPSEC]" heading, but it doesn't cover XFRM docs yet. Add the entry. Reviewed-by: Randy Dunlap Tested-by: Randy Dunlap Signed-off-by: Bagas Sanjaya Signed-off-by: Steffen Klassert --- MAINTAINERS | 1 + 1 file changed, 1 insertion(+) diff --git a/MAINTAINERS b/MAINTAINERS index d652f4f27756..4f33daad40be 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -18041,6 +18041,7 @@ L: netdev@vger.kernel.org S: Maintained T: git git://git.kernel.org/pub/scm/linux/kernel/git/klassert/ipsec.git T: git git://git.kernel.org/pub/scm/linux/kernel/git/klassert/ipsec-next.git +F: Documentation/networking/xfrm/ F: include/net/xfrm.h F: include/uapi/linux/xfrm.h F: net/ipv4/ah4.c From 5422318e27d7a4662701f518e2e51b9f73a331b1 Mon Sep 17 00:00:00 2001 From: Tariq Toukan Date: Tue, 11 Nov 2025 14:24:48 +0200 Subject: [PATCH 0735/1381] net/mlx5: Expose definition for 1600Gbps link mode This patch exposes new link mode for 1600Gbps, utilizing 8 lanes at 200Gbps per lane. Co-developed-by: Yael Chemla Reviewed-by: Shahar Shitrit Signed-off-by: Tariq Toukan Link: https://patch.msgid.link/1762863888-1092798-1-git-send-email-tariqt@nvidia.com Signed-off-by: Leon Romanovsky --- include/linux/mlx5/port.h | 1 + 1 file changed, 1 insertion(+) diff --git a/include/linux/mlx5/port.h b/include/linux/mlx5/port.h index 58770b86f793..1df9d9a57bbc 100644 --- a/include/linux/mlx5/port.h +++ b/include/linux/mlx5/port.h @@ -112,6 +112,7 @@ enum mlx5e_ext_link_mode { MLX5E_400GAUI_2_400GBASE_CR2_KR2 = 17, MLX5E_800GAUI_8_800GBASE_CR8_KR8 = 19, MLX5E_800GAUI_4_800GBASE_CR4_KR4 = 20, + MLX5E_1600TAUI_8_1600TBASE_CR8_KR8 = 23, MLX5E_EXT_LINK_MODES_NUMBER, }; From 71df9227ba9c5d18372c30904a93d360151874b2 Mon Sep 17 00:00:00 2001 From: Marc Kleine-Budde Date: Wed, 1 Oct 2025 14:40:01 +0530 Subject: [PATCH 0736/1381] can: mcp251xfd: move chip sleep mode into runtime pm This is a preparation patch to add GPIO support. Up to now, the Vdd regulator and the clocks have been managed by Runtime-PM (on systems without CONFIG_PM these remain permanently switched on). During the mcp251xfd_open() callback the mcp251xfd is powered, soft-reset and configured. In mcp251xfd_stop() the chip is shut down again. To support the on-chip GPIOs, the chip must be supplied with power while GPIOs are being requested, even if the networking interface is down. To support this, move the functions mcp251xfd_chip_softreset() and mcp251xfd_chip_clock_init() from mcp251xfd_chip_start() to mcp251xfd_runtime_resume(). Instead of setting the controller to sleep mode in mcp251xfd_chip_stop(), bring it into configuration mode. This way it doesn't take part in bus activity and doesn't enter sleep mode. Signed-off-by: Gregor Herburger Tested-by: Viken Dadhaniya Signed-off-by: Viken Dadhaniya Reviewed-by: Manivannan Sadhasivam Link: https://patch.msgid.link/20251001091006.4003841-2-viken.dadhaniya@oss.qualcomm.com Signed-off-by: Marc Kleine-Budde --- .../net/can/spi/mcp251xfd/mcp251xfd-core.c | 99 ++++++++++++------- 1 file changed, 61 insertions(+), 38 deletions(-) diff --git a/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c b/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c index c0f9d9fed02e..9104561e0995 100644 --- a/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c +++ b/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c @@ -767,21 +767,13 @@ static void mcp251xfd_chip_stop(struct mcp251xfd_priv *priv, mcp251xfd_chip_interrupts_disable(priv); mcp251xfd_chip_rx_int_disable(priv); mcp251xfd_timestamp_stop(priv); - mcp251xfd_chip_sleep(priv); + mcp251xfd_chip_set_mode(priv, MCP251XFD_REG_CON_MODE_CONFIG); } static int mcp251xfd_chip_start(struct mcp251xfd_priv *priv) { int err; - err = mcp251xfd_chip_softreset(priv); - if (err) - goto out_chip_stop; - - err = mcp251xfd_chip_clock_init(priv); - if (err) - goto out_chip_stop; - err = mcp251xfd_chip_timestamp_init(priv); if (err) goto out_chip_stop; @@ -1625,8 +1617,11 @@ static int mcp251xfd_open(struct net_device *ndev) return err; err = pm_runtime_resume_and_get(ndev->dev.parent); - if (err) + if (err) { + if (err == -ETIMEDOUT || err == -ENODEV) + pm_runtime_set_suspended(ndev->dev.parent); goto out_close_candev; + } err = mcp251xfd_ring_alloc(priv); if (err) @@ -1907,53 +1902,53 @@ static int mcp251xfd_register(struct mcp251xfd_priv *priv) struct net_device *ndev = priv->ndev; int err; + mcp251xfd_register_quirks(priv); + err = mcp251xfd_clks_and_vdd_enable(priv); if (err) return err; + err = mcp251xfd_chip_softreset(priv); + if (err == -ENODEV) + goto out_clks_and_vdd_disable; + if (err) + goto out_chip_sleep; + + err = mcp251xfd_chip_clock_init(priv); + if (err == -ENODEV) + goto out_clks_and_vdd_disable; + if (err) + goto out_chip_sleep; + pm_runtime_get_noresume(ndev->dev.parent); err = pm_runtime_set_active(ndev->dev.parent); if (err) goto out_runtime_put_noidle; pm_runtime_enable(ndev->dev.parent); - mcp251xfd_register_quirks(priv); - - err = mcp251xfd_chip_softreset(priv); - if (err == -ENODEV) - goto out_runtime_disable; - if (err) - goto out_chip_sleep; - - err = mcp251xfd_chip_clock_init(priv); - if (err == -ENODEV) - goto out_runtime_disable; - if (err) - goto out_chip_sleep; - err = mcp251xfd_register_chip_detect(priv); if (err) - goto out_chip_sleep; + goto out_runtime_disable; err = mcp251xfd_register_check_rx_int(priv); if (err) - goto out_chip_sleep; + goto out_runtime_disable; mcp251xfd_ethtool_init(priv); err = register_candev(ndev); if (err) - goto out_chip_sleep; + goto out_runtime_disable; err = mcp251xfd_register_done(priv); if (err) goto out_unregister_candev; - /* Put controller into sleep mode and let pm_runtime_put() - * disable the clocks and vdd. If CONFIG_PM is not enabled, - * the clocks and vdd will stay powered. + /* Put controller into Config mode and let pm_runtime_put() + * put in sleep mode, disable the clocks and vdd. If CONFIG_PM + * is not enabled, the clocks and vdd will stay powered. */ - err = mcp251xfd_chip_sleep(priv); + err = mcp251xfd_chip_set_mode(priv, MCP251XFD_REG_CON_MODE_CONFIG); if (err) goto out_unregister_candev; @@ -1963,12 +1958,13 @@ static int mcp251xfd_register(struct mcp251xfd_priv *priv) out_unregister_candev: unregister_candev(ndev); -out_chip_sleep: - mcp251xfd_chip_sleep(priv); out_runtime_disable: pm_runtime_disable(ndev->dev.parent); out_runtime_put_noidle: pm_runtime_put_noidle(ndev->dev.parent); +out_chip_sleep: + mcp251xfd_chip_sleep(priv); +out_clks_and_vdd_disable: mcp251xfd_clks_and_vdd_disable(priv); return err; @@ -1980,10 +1976,12 @@ static inline void mcp251xfd_unregister(struct mcp251xfd_priv *priv) unregister_candev(ndev); - if (pm_runtime_enabled(ndev->dev.parent)) + if (pm_runtime_enabled(ndev->dev.parent)) { pm_runtime_disable(ndev->dev.parent); - else + } else { + mcp251xfd_chip_sleep(priv); mcp251xfd_clks_and_vdd_disable(priv); + } } static const struct of_device_id mcp251xfd_of_match[] = { @@ -2206,16 +2204,41 @@ static void mcp251xfd_remove(struct spi_device *spi) static int __maybe_unused mcp251xfd_runtime_suspend(struct device *device) { - const struct mcp251xfd_priv *priv = dev_get_drvdata(device); + struct mcp251xfd_priv *priv = dev_get_drvdata(device); + mcp251xfd_chip_sleep(priv); return mcp251xfd_clks_and_vdd_disable(priv); } static int __maybe_unused mcp251xfd_runtime_resume(struct device *device) { - const struct mcp251xfd_priv *priv = dev_get_drvdata(device); + struct mcp251xfd_priv *priv = dev_get_drvdata(device); + int err; - return mcp251xfd_clks_and_vdd_enable(priv); + err = mcp251xfd_clks_and_vdd_enable(priv); + if (err) + return err; + + err = mcp251xfd_chip_softreset(priv); + if (err == -ENODEV) + goto out_clks_and_vdd_disable; + if (err) + goto out_chip_sleep; + + err = mcp251xfd_chip_clock_init(priv); + if (err == -ENODEV) + goto out_clks_and_vdd_disable; + if (err) + goto out_chip_sleep; + + return 0; + +out_chip_sleep: + mcp251xfd_chip_sleep(priv); +out_clks_and_vdd_disable: + mcp251xfd_clks_and_vdd_disable(priv); + + return err; } static const struct dev_pm_ops mcp251xfd_pm_ops = { From f5982a679a16a673a0fc374c1d9d77111cd86740 Mon Sep 17 00:00:00 2001 From: Gregor Herburger Date: Wed, 1 Oct 2025 14:40:02 +0530 Subject: [PATCH 0737/1381] can: mcp251xfd: utilize gather_write function for all non-CRC writes This is a preparation patch to add errata workaround for non crc writes. Currently for non-crc writes to the chip can go through the .gather_write, .write or the reg_update_bits callback. To allow the addition of the errata fix at a single location use mcp251xfd_regmap_nocrc_gather_write for all non-CRC write instructions, similar to the crc regmap. Signed-off-by: Gregor Herburger Tested-by: Viken Dadhaniya Signed-off-by: Viken Dadhaniya Reviewed-by: Manivannan Sadhasivam Link: https://patch.msgid.link/20251001091006.4003841-3-viken.dadhaniya@oss.qualcomm.com Signed-off-by: Marc Kleine-Budde --- .../net/can/spi/mcp251xfd/mcp251xfd-regmap.c | 25 ++++++++++--------- 1 file changed, 13 insertions(+), 12 deletions(-) diff --git a/drivers/net/can/spi/mcp251xfd/mcp251xfd-regmap.c b/drivers/net/can/spi/mcp251xfd/mcp251xfd-regmap.c index 8c5be8d1c519..e61cbd209955 100644 --- a/drivers/net/can/spi/mcp251xfd/mcp251xfd-regmap.c +++ b/drivers/net/can/spi/mcp251xfd/mcp251xfd-regmap.c @@ -12,14 +12,6 @@ static const struct regmap_config mcp251xfd_regmap_crc; -static int -mcp251xfd_regmap_nocrc_write(void *context, const void *data, size_t count) -{ - struct spi_device *spi = context; - - return spi_write(spi, data, count); -} - static int mcp251xfd_regmap_nocrc_gather_write(void *context, const void *reg, size_t reg_len, @@ -47,6 +39,15 @@ mcp251xfd_regmap_nocrc_gather_write(void *context, return spi_sync_transfer(spi, xfer, ARRAY_SIZE(xfer)); } +static int +mcp251xfd_regmap_nocrc_write(void *context, const void *data, size_t count) +{ + const size_t data_offset = sizeof(__be16); + + return mcp251xfd_regmap_nocrc_gather_write(context, data, data_offset, + data + data_offset, count - data_offset); +} + static inline bool mcp251xfd_update_bits_read_reg(const struct mcp251xfd_priv *priv, unsigned int reg) @@ -64,6 +65,7 @@ mcp251xfd_update_bits_read_reg(const struct mcp251xfd_priv *priv, case MCP251XFD_REG_CON: case MCP251XFD_REG_OSC: case MCP251XFD_REG_ECCCON: + case MCP251XFD_REG_IOCON: return true; default: mcp251xfd_for_each_rx_ring(priv, ring, n) { @@ -139,10 +141,9 @@ mcp251xfd_regmap_nocrc_update_bits(void *context, unsigned int reg, tmp_le32 = orig_le32 & ~mask_le32; tmp_le32 |= val_le32 & mask_le32; - mcp251xfd_spi_cmd_write_nocrc(&buf_tx->cmd, reg + first_byte); - memcpy(buf_tx->data, &tmp_le32, len); - - return spi_write(spi, buf_tx, sizeof(buf_tx->cmd) + len); + reg += first_byte; + mcp251xfd_spi_cmd_write_nocrc(&buf_tx->cmd, reg); + return mcp251xfd_regmap_nocrc_gather_write(context, &buf_tx->cmd, 2, &tmp_le32, len); } static int From d9cac93cd1bed31724f55857b934bcbbc8313b59 Mon Sep 17 00:00:00 2001 From: Bobby Eshleman Date: Sat, 8 Nov 2025 08:00:52 -0800 Subject: [PATCH 0738/1381] selftests/vsock: improve logging in vmtest.sh Improve usability of logging functions. Remove the test name prefix from logging functions so that logging calls can be made deeper into the call stack without passing down the test name or setting some global. Teach log function to accept a LOG_PREFIX variable to avoid unnecessary argument shifting. Remove log_setup() and instead use log_host(). The host/guest prefixes are useful to show whether a failure happened on the guest or host side, but "setup" doesn't really give additional useful information. Since all log_setup() calls happen on the host, lets just use log_host() instead. Signed-off-by: Bobby Eshleman Reviewed-by: Stefano Garzarella Link: https://patch.msgid.link/20251108-vsock-selftests-fixes-and-improvements-v4-1-d5e8d6c87289@meta.com Signed-off-by: Jakub Kicinski --- tools/testing/selftests/vsock/vmtest.sh | 71 +++++++++++-------------- 1 file changed, 30 insertions(+), 41 deletions(-) diff --git a/tools/testing/selftests/vsock/vmtest.sh b/tools/testing/selftests/vsock/vmtest.sh index 8ceeb8a7894f..bc16b13cdbe3 100755 --- a/tools/testing/selftests/vsock/vmtest.sh +++ b/tools/testing/selftests/vsock/vmtest.sh @@ -271,60 +271,51 @@ EOF host_wait_for_listener() { wait_for_listener "${TEST_HOST_PORT_LISTENER}" "${WAIT_PERIOD}" "${WAIT_PERIOD_MAX}" -} -__log_stdin() { - cat | awk '{ printf "%s:\t%s\n","'"${prefix}"'", $0 }' -} - -__log_args() { - echo "$*" | awk '{ printf "%s:\t%s\n","'"${prefix}"'", $0 }' } log() { - local prefix="$1" + local redirect + local prefix - shift - local redirect= if [[ ${VERBOSE} -eq 0 ]]; then redirect=/dev/null else redirect=/dev/stdout fi - if [[ "$#" -eq 0 ]]; then - __log_stdin | tee -a "${LOG}" > ${redirect} - else - __log_args "$@" | tee -a "${LOG}" > ${redirect} - fi -} + prefix="${LOG_PREFIX:-}" -log_setup() { - log "setup" "$@" + if [[ "$#" -eq 0 ]]; then + if [[ -n "${prefix}" ]]; then + awk -v prefix="${prefix}" '{printf "%s: %s\n", prefix, $0}' + else + cat + fi + else + if [[ -n "${prefix}" ]]; then + echo "${prefix}: " "$@" + else + echo "$@" + fi + fi | tee -a "${LOG}" > "${redirect}" } log_host() { - local testname=$1 - - shift - log "test:${testname}:host" "$@" + LOG_PREFIX=host log "$@" } log_guest() { - local testname=$1 - - shift - log "test:${testname}:guest" "$@" + LOG_PREFIX=guest log "$@" } test_vm_server_host_client() { - local testname="${FUNCNAME[0]#test_}" vm_ssh -- "${VSOCK_TEST}" \ --mode=server \ --control-port="${TEST_GUEST_PORT}" \ --peer-cid=2 \ - 2>&1 | log_guest "${testname}" & + 2>&1 | log_guest & vm_wait_for_listener "${TEST_GUEST_PORT}" @@ -332,18 +323,17 @@ test_vm_server_host_client() { --mode=client \ --control-host=127.0.0.1 \ --peer-cid="${VSOCK_CID}" \ - --control-port="${TEST_HOST_PORT}" 2>&1 | log_host "${testname}" + --control-port="${TEST_HOST_PORT}" 2>&1 | log_host return $? } test_vm_client_host_server() { - local testname="${FUNCNAME[0]#test_}" ${VSOCK_TEST} \ --mode "server" \ --control-port "${TEST_HOST_PORT_LISTENER}" \ - --peer-cid "${VSOCK_CID}" 2>&1 | log_host "${testname}" & + --peer-cid "${VSOCK_CID}" 2>&1 | log_host & host_wait_for_listener @@ -351,19 +341,18 @@ test_vm_client_host_server() { --mode=client \ --control-host=10.0.2.2 \ --peer-cid=2 \ - --control-port="${TEST_HOST_PORT_LISTENER}" 2>&1 | log_guest "${testname}" + --control-port="${TEST_HOST_PORT_LISTENER}" 2>&1 | log_guest return $? } test_vm_loopback() { - local testname="${FUNCNAME[0]#test_}" local port=60000 # non-forwarded local port vm_ssh -- "${VSOCK_TEST}" \ --mode=server \ --control-port="${port}" \ - --peer-cid=1 2>&1 | log_guest "${testname}" & + --peer-cid=1 2>&1 | log_guest & vm_wait_for_listener "${port}" @@ -371,7 +360,7 @@ test_vm_loopback() { --mode=client \ --control-host="127.0.0.1" \ --control-port="${port}" \ - --peer-cid=1 2>&1 | log_guest "${testname}" + --peer-cid=1 2>&1 | log_guest return $? } @@ -399,25 +388,25 @@ run_test() { host_oops_cnt_after=$(dmesg | grep -i 'Oops' | wc -l) if [[ ${host_oops_cnt_after} -gt ${host_oops_cnt_before} ]]; then - echo "FAIL: kernel oops detected on host" | log_host "${name}" + echo "FAIL: kernel oops detected on host" | log_host rc=$KSFT_FAIL fi host_warn_cnt_after=$(dmesg --level=warn | grep -c -i 'vsock') if [[ ${host_warn_cnt_after} -gt ${host_warn_cnt_before} ]]; then - echo "FAIL: kernel warning detected on host" | log_host "${name}" + echo "FAIL: kernel warning detected on host" | log_host rc=$KSFT_FAIL fi vm_oops_cnt_after=$(vm_ssh -- dmesg | grep -i 'Oops' | wc -l) if [[ ${vm_oops_cnt_after} -gt ${vm_oops_cnt_before} ]]; then - echo "FAIL: kernel oops detected on vm" | log_host "${name}" + echo "FAIL: kernel oops detected on vm" | log_host rc=$KSFT_FAIL fi vm_warn_cnt_after=$(vm_ssh -- dmesg --level=warn | grep -c -i 'vsock') if [[ ${vm_warn_cnt_after} -gt ${vm_warn_cnt_before} ]]; then - echo "FAIL: kernel warning detected on vm" | log_host "${name}" + echo "FAIL: kernel warning detected on vm" | log_host rc=$KSFT_FAIL fi @@ -452,10 +441,10 @@ handle_build echo "1..${#ARGS[@]}" -log_setup "Booting up VM" +log_host "Booting up VM" vm_start vm_wait_for_ssh -log_setup "VM booted up" +log_host "VM booted up" cnt_pass=0 cnt_fail=0 From 2ed3ce7efbd2854fb7eb76627ccec92362b2345b Mon Sep 17 00:00:00 2001 From: Bobby Eshleman Date: Sat, 8 Nov 2025 08:00:53 -0800 Subject: [PATCH 0739/1381] selftests/vsock: make wait_for_listener() work even if pipefail is on Rewrite wait_for_listener()'s pattern matching to avoid tripping the if-condition when pipefail is on. awk doesn't gracefully handle SIGPIPE with a non-zero exit code, so grep exiting upon finding a match causes false-positives when the pipefail option is used (grep exits, SIGPIPE emits, and awk complains with a non-zero exit code). Instead, move all of the pattern matching into awk so that SIGPIPE cannot happen and the correct exit code is returned. Signed-off-by: Bobby Eshleman Reviewed-by: Stefano Garzarella Link: https://patch.msgid.link/20251108-vsock-selftests-fixes-and-improvements-v4-2-d5e8d6c87289@meta.com Signed-off-by: Jakub Kicinski --- tools/testing/selftests/vsock/vmtest.sh | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/tools/testing/selftests/vsock/vmtest.sh b/tools/testing/selftests/vsock/vmtest.sh index bc16b13cdbe3..01ce16523afb 100755 --- a/tools/testing/selftests/vsock/vmtest.sh +++ b/tools/testing/selftests/vsock/vmtest.sh @@ -251,9 +251,11 @@ wait_for_listener() # for tcp protocol additionally check the socket state [ "${protocol}" = "tcp" ] && pattern="${pattern}0A" + for i in $(seq "${max_intervals}"); do - if awk '{print $2" "$4}' /proc/net/"${protocol}"* | \ - grep -q "${pattern}"; then + if awk -v pattern="${pattern}" \ + 'BEGIN {rc=1} $2" "$4 ~ pattern {rc=0} END {exit rc}' \ + /proc/net/"${protocol}"*; then break fi sleep "${interval}" From 4f76ff14d3d96a21574e66222041757d679a8a10 Mon Sep 17 00:00:00 2001 From: Bobby Eshleman Date: Sat, 8 Nov 2025 08:00:54 -0800 Subject: [PATCH 0740/1381] selftests/vsock: reuse logic for vsock_test through wrapper functions Add wrapper functions vm_vsock_test() and host_vsock_test() to invoke the vsock_test binary. This encapsulates several items of repeat logic, such as waiting for the server to reach listening state and enabling/disabling the bash option pipefail to avoid pipe-style logging from hiding failures. Signed-off-by: Bobby Eshleman Reviewed-by: Stefano Garzarella Link: https://patch.msgid.link/20251108-vsock-selftests-fixes-and-improvements-v4-3-d5e8d6c87289@meta.com Signed-off-by: Jakub Kicinski --- tools/testing/selftests/vsock/vmtest.sh | 135 +++++++++++++++++------- 1 file changed, 95 insertions(+), 40 deletions(-) diff --git a/tools/testing/selftests/vsock/vmtest.sh b/tools/testing/selftests/vsock/vmtest.sh index 01ce16523afb..3bccd9b84e4a 100755 --- a/tools/testing/selftests/vsock/vmtest.sh +++ b/tools/testing/selftests/vsock/vmtest.sh @@ -272,8 +272,81 @@ EOF } host_wait_for_listener() { - wait_for_listener "${TEST_HOST_PORT_LISTENER}" "${WAIT_PERIOD}" "${WAIT_PERIOD_MAX}" + local port=$1 + wait_for_listener "${port}" "${WAIT_PERIOD}" "${WAIT_PERIOD_MAX}" +} + +vm_vsock_test() { + local host=$1 + local cid=$2 + local port=$3 + local rc + + # log output and use pipefail to respect vsock_test errors + set -o pipefail + if [[ "${host}" != server ]]; then + vm_ssh -- "${VSOCK_TEST}" \ + --mode=client \ + --control-host="${host}" \ + --peer-cid="${cid}" \ + --control-port="${port}" \ + 2>&1 | log_guest + rc=$? + else + vm_ssh -- "${VSOCK_TEST}" \ + --mode=server \ + --peer-cid="${cid}" \ + --control-port="${port}" \ + 2>&1 | log_guest & + rc=$? + + if [[ $rc -ne 0 ]]; then + set +o pipefail + return $rc + fi + + vm_wait_for_listener "${port}" + rc=$? + fi + set +o pipefail + + return $rc +} + +host_vsock_test() { + local host=$1 + local cid=$2 + local port=$3 + local rc + + # log output and use pipefail to respect vsock_test errors + set -o pipefail + if [[ "${host}" != server ]]; then + ${VSOCK_TEST} \ + --mode=client \ + --peer-cid="${cid}" \ + --control-host="${host}" \ + --control-port="${port}" 2>&1 | log_host + rc=$? + else + ${VSOCK_TEST} \ + --mode=server \ + --peer-cid="${cid}" \ + --control-port="${port}" 2>&1 | log_host & + rc=$? + + if [[ $rc -ne 0 ]]; then + set +o pipefail + return $rc + fi + + host_wait_for_listener "${port}" + rc=$? + fi + set +o pipefail + + return $rc } log() { @@ -312,59 +385,41 @@ log_guest() { } test_vm_server_host_client() { + if ! vm_vsock_test "server" 2 "${TEST_GUEST_PORT}"; then + return "${KSFT_FAIL}" + fi - vm_ssh -- "${VSOCK_TEST}" \ - --mode=server \ - --control-port="${TEST_GUEST_PORT}" \ - --peer-cid=2 \ - 2>&1 | log_guest & + if ! host_vsock_test "127.0.0.1" "${VSOCK_CID}" "${TEST_HOST_PORT}"; then + return "${KSFT_FAIL}" + fi - vm_wait_for_listener "${TEST_GUEST_PORT}" - - ${VSOCK_TEST} \ - --mode=client \ - --control-host=127.0.0.1 \ - --peer-cid="${VSOCK_CID}" \ - --control-port="${TEST_HOST_PORT}" 2>&1 | log_host - - return $? + return "${KSFT_PASS}" } test_vm_client_host_server() { + if ! host_vsock_test "server" "${VSOCK_CID}" "${TEST_HOST_PORT_LISTENER}"; then + return "${KSFT_FAIL}" + fi - ${VSOCK_TEST} \ - --mode "server" \ - --control-port "${TEST_HOST_PORT_LISTENER}" \ - --peer-cid "${VSOCK_CID}" 2>&1 | log_host & + if ! vm_vsock_test "10.0.2.2" 2 "${TEST_HOST_PORT_LISTENER}"; then + return "${KSFT_FAIL}" + fi - host_wait_for_listener - - vm_ssh -- "${VSOCK_TEST}" \ - --mode=client \ - --control-host=10.0.2.2 \ - --peer-cid=2 \ - --control-port="${TEST_HOST_PORT_LISTENER}" 2>&1 | log_guest - - return $? + return "${KSFT_PASS}" } test_vm_loopback() { local port=60000 # non-forwarded local port - vm_ssh -- "${VSOCK_TEST}" \ - --mode=server \ - --control-port="${port}" \ - --peer-cid=1 2>&1 | log_guest & + if ! vm_vsock_test "server" 1 "${port}"; then + return "${KSFT_FAIL}" + fi - vm_wait_for_listener "${port}" + if ! vm_vsock_test "127.0.0.1" 1 "${port}"; then + return "${KSFT_FAIL}" + fi - vm_ssh -- "${VSOCK_TEST}" \ - --mode=client \ - --control-host="127.0.0.1" \ - --control-port="${port}" \ - --peer-cid=1 2>&1 | log_guest - - return $? + return "${KSFT_PASS}" } run_test() { From ac8997e943bc035171e76c8457db6fa973c8a9ab Mon Sep 17 00:00:00 2001 From: Bobby Eshleman Date: Sat, 8 Nov 2025 08:00:55 -0800 Subject: [PATCH 0741/1381] selftests/vsock: avoid multi-VM pidfile collisions with QEMU Change QEMU to use generated pidfile names instead of just a single globally-defined pidfile. This allows multiple QEMU instances to co-exist with different pidfiles. This is required for future tests that use multiple VMs to check for CID collissions. Additionally, this also places the burden of killing the QEMU process and cleaning up the pidfile on the caller of vm_start(). To help with this, a function terminate_pidfiles() is introduced that callers use to perform the cleanup. The terminate_pidfiles() function supports multiple pidfile removals because future patches will need to process two pidfiles at a time. Change QEMU_OPTS to be initialized inside the vm_start(). This allows the generated pidfile to be passed to the string assignment, and prepares for future vm-specific options as well (e.g., cid). Signed-off-by: Bobby Eshleman Reviewed-by: Stefano Garzarella Link: https://patch.msgid.link/20251108-vsock-selftests-fixes-and-improvements-v4-4-d5e8d6c87289@meta.com Signed-off-by: Jakub Kicinski --- tools/testing/selftests/vsock/vmtest.sh | 62 +++++++++++++++++-------- 1 file changed, 43 insertions(+), 19 deletions(-) diff --git a/tools/testing/selftests/vsock/vmtest.sh b/tools/testing/selftests/vsock/vmtest.sh index 3bccd9b84e4a..13b685280a67 100755 --- a/tools/testing/selftests/vsock/vmtest.sh +++ b/tools/testing/selftests/vsock/vmtest.sh @@ -23,7 +23,8 @@ readonly VSOCK_CID=1234 readonly WAIT_PERIOD=3 readonly WAIT_PERIOD_MAX=60 readonly WAIT_TOTAL=$(( WAIT_PERIOD * WAIT_PERIOD_MAX )) -readonly QEMU_PIDFILE=$(mktemp /tmp/qemu_vsock_vmtest_XXXX.pid) +readonly PIDFILE_TEMPLATE=/tmp/vsock_vmtest_XXXX.pid +declare -A PIDFILES # virtme-ng offers a netdev for ssh when using "--ssh", but we also need a # control port forwarded for vsock_test. Because virtme-ng doesn't support @@ -33,12 +34,6 @@ readonly QEMU_PIDFILE=$(mktemp /tmp/qemu_vsock_vmtest_XXXX.pid) # add the kernel cmdline options that virtme-init uses to setup the interface. readonly QEMU_TEST_PORT_FWD="hostfwd=tcp::${TEST_HOST_PORT}-:${TEST_GUEST_PORT}" readonly QEMU_SSH_PORT_FWD="hostfwd=tcp::${SSH_HOST_PORT}-:${SSH_GUEST_PORT}" -readonly QEMU_OPTS="\ - -netdev user,id=n0,${QEMU_TEST_PORT_FWD},${QEMU_SSH_PORT_FWD} \ - -device virtio-net-pci,netdev=n0 \ - -device vhost-vsock-pci,guest-cid=${VSOCK_CID} \ - --pidfile ${QEMU_PIDFILE} \ -" readonly KERNEL_CMDLINE="\ virtme.dhcp net.ifnames=0 biosdevname=0 \ virtme.ssh virtme_ssh_channel=tcp virtme_ssh_user=$USER \ @@ -90,15 +85,7 @@ vm_ssh() { } cleanup() { - if [[ -s "${QEMU_PIDFILE}" ]]; then - pkill -SIGTERM -F "${QEMU_PIDFILE}" > /dev/null 2>&1 - fi - - # If failure occurred during or before qemu start up, then we need - # to clean this up ourselves. - if [[ -e "${QEMU_PIDFILE}" ]]; then - rm "${QEMU_PIDFILE}" - fi + terminate_pidfiles "${!PIDFILES[@]}" } check_args() { @@ -188,10 +175,37 @@ handle_build() { popd &>/dev/null } +create_pidfile() { + local pidfile + + pidfile=$(mktemp "${PIDFILE_TEMPLATE}") + PIDFILES["${pidfile}"]=1 + + echo "${pidfile}" +} + +terminate_pidfiles() { + local pidfile + + for pidfile in "$@"; do + if [[ -s "${pidfile}" ]]; then + pkill -SIGTERM -F "${pidfile}" > /dev/null 2>&1 + fi + + if [[ -e "${pidfile}" ]]; then + rm -f "${pidfile}" + fi + + unset "PIDFILES[${pidfile}]" + done +} + vm_start() { + local pidfile=$1 local logfile=/dev/null local verbose_opt="" local kernel_opt="" + local qemu_opts="" local qemu qemu=$(command -v "${QEMU}") @@ -201,6 +215,13 @@ vm_start() { logfile=/dev/stdout fi + qemu_opts="\ + -netdev user,id=n0,${QEMU_TEST_PORT_FWD},${QEMU_SSH_PORT_FWD} \ + -device virtio-net-pci,netdev=n0 \ + -device vhost-vsock-pci,guest-cid=${VSOCK_CID} \ + --pidfile ${pidfile} + " + if [[ "${BUILD}" -eq 1 ]]; then kernel_opt="${KERNEL_CHECKOUT}" fi @@ -209,14 +230,14 @@ vm_start() { --run \ ${kernel_opt} \ ${verbose_opt} \ - --qemu-opts="${QEMU_OPTS}" \ + --qemu-opts="${qemu_opts}" \ --qemu="${qemu}" \ --user root \ --append "${KERNEL_CMDLINE}" \ --rw &> ${logfile} & if ! timeout ${WAIT_TOTAL} \ - bash -c 'while [[ ! -s '"${QEMU_PIDFILE}"' ]]; do sleep 1; done; exit 0'; then + bash -c 'while [[ ! -s '"${pidfile}"' ]]; do sleep 1; done; exit 0'; then die "failed to boot VM" fi } @@ -499,7 +520,8 @@ handle_build echo "1..${#ARGS[@]}" log_host "Booting up VM" -vm_start +pidfile="$(create_pidfile)" +vm_start "${pidfile}" vm_wait_for_ssh log_host "VM booted up" @@ -523,6 +545,8 @@ for arg in "${ARGS[@]}"; do cnt_total=$(( cnt_total + 1 )) done +terminate_pidfiles "${pidfile}" + echo "SUMMARY: PASS=${cnt_pass} SKIP=${cnt_skip} FAIL=${cnt_fail}" echo "Log: ${LOG}" From c7df4adc0679afa2da47c13cbf7960a26e0fe7cb Mon Sep 17 00:00:00 2001 From: Bobby Eshleman Date: Sat, 8 Nov 2025 08:00:56 -0800 Subject: [PATCH 0742/1381] selftests/vsock: do not unconditionally die if qemu fails If QEMU fails to boot, then set the returncode (via timeout) instead of unconditionally dying. This is in preparation for tests that expect QEMU to fail to boot. In that case, we just want to know if the boot failed or not so we can test the pass/fail criteria, and continue executing the next test. Reviewed-by: Simon Horman Signed-off-by: Bobby Eshleman Reviewed-by: Stefano Garzarella Link: https://patch.msgid.link/20251108-vsock-selftests-fixes-and-improvements-v4-5-d5e8d6c87289@meta.com Signed-off-by: Jakub Kicinski --- tools/testing/selftests/vsock/vmtest.sh | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/tools/testing/selftests/vsock/vmtest.sh b/tools/testing/selftests/vsock/vmtest.sh index 13b685280a67..6889bdb8a31c 100755 --- a/tools/testing/selftests/vsock/vmtest.sh +++ b/tools/testing/selftests/vsock/vmtest.sh @@ -236,10 +236,8 @@ vm_start() { --append "${KERNEL_CMDLINE}" \ --rw &> ${logfile} & - if ! timeout ${WAIT_TOTAL} \ - bash -c 'while [[ ! -s '"${pidfile}"' ]]; do sleep 1; done; exit 0'; then - die "failed to boot VM" - fi + timeout "${WAIT_TOTAL}" \ + bash -c 'while [[ ! -s '"${pidfile}"' ]]; do sleep 1; done; exit 0' } vm_wait_for_ssh() { From 9e2ad0bc3640c384799c91da0b8b794c672da4d3 Mon Sep 17 00:00:00 2001 From: Bobby Eshleman Date: Sat, 8 Nov 2025 08:00:57 -0800 Subject: [PATCH 0743/1381] selftests/vsock: speed up tests by reducing the QEMU pidfile timeout Reduce the time waiting for the QEMU pidfile from three minutes to five seconds. The three minute time window was chosen to make sure QEMU had enough time to fully boot up. This, however, is an unreasonably long delay for QEMU to write the pidfile, which happens earlier when the QEMU process starts (not after VM boot). The three minute delay becomes noticeably wasteful in future tests that expect QEMU to fail and wait a full three minutes for a pidfile that will never exist. Reviewed-by: Simon Horman Signed-off-by: Bobby Eshleman Reviewed-by: Stefano Garzarella Link: https://patch.msgid.link/20251108-vsock-selftests-fixes-and-improvements-v4-6-d5e8d6c87289@meta.com Signed-off-by: Jakub Kicinski --- tools/testing/selftests/vsock/vmtest.sh | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tools/testing/selftests/vsock/vmtest.sh b/tools/testing/selftests/vsock/vmtest.sh index 6889bdb8a31c..bd231467c66b 100755 --- a/tools/testing/selftests/vsock/vmtest.sh +++ b/tools/testing/selftests/vsock/vmtest.sh @@ -22,7 +22,7 @@ readonly SSH_HOST_PORT=2222 readonly VSOCK_CID=1234 readonly WAIT_PERIOD=3 readonly WAIT_PERIOD_MAX=60 -readonly WAIT_TOTAL=$(( WAIT_PERIOD * WAIT_PERIOD_MAX )) +readonly WAIT_QEMU=5 readonly PIDFILE_TEMPLATE=/tmp/vsock_vmtest_XXXX.pid declare -A PIDFILES @@ -236,7 +236,7 @@ vm_start() { --append "${KERNEL_CMDLINE}" \ --rw &> ${logfile} & - timeout "${WAIT_TOTAL}" \ + timeout "${WAIT_QEMU}" \ bash -c 'while [[ ! -s '"${pidfile}"' ]]; do sleep 1; done; exit 0' } From 7fea50dff98812db3ee4ddd96405c3298d9aa4cb Mon Sep 17 00:00:00 2001 From: Bobby Eshleman Date: Sat, 8 Nov 2025 08:00:58 -0800 Subject: [PATCH 0744/1381] selftests/vsock: add check_result() for pass/fail counting Add check_result() function to reuse logic for incrementing the pass/fail counters. This function will get used by different callers as we add different types of tests in future patches (namely, namespace and non-namespace tests will be called at different places, and re-use this function). Reviewed-by: Simon Horman Signed-off-by: Bobby Eshleman Reviewed-by: Stefano Garzarella Link: https://patch.msgid.link/20251108-vsock-selftests-fixes-and-improvements-v4-7-d5e8d6c87289@meta.com Signed-off-by: Jakub Kicinski --- tools/testing/selftests/vsock/vmtest.sh | 32 ++++++++++++++++--------- 1 file changed, 21 insertions(+), 11 deletions(-) diff --git a/tools/testing/selftests/vsock/vmtest.sh b/tools/testing/selftests/vsock/vmtest.sh index bd231467c66b..2dd9bbb8c4a9 100755 --- a/tools/testing/selftests/vsock/vmtest.sh +++ b/tools/testing/selftests/vsock/vmtest.sh @@ -79,6 +79,26 @@ die() { exit "${KSFT_FAIL}" } +check_result() { + local rc arg + + rc=$1 + arg=$2 + + cnt_total=$(( cnt_total + 1 )) + + if [[ ${rc} -eq ${KSFT_PASS} ]]; then + cnt_pass=$(( cnt_pass + 1 )) + echo "ok ${cnt_total} ${arg}" + elif [[ ${rc} -eq ${KSFT_SKIP} ]]; then + cnt_skip=$(( cnt_skip + 1 )) + echo "ok ${cnt_total} ${arg} # SKIP" + elif [[ ${rc} -eq ${KSFT_FAIL} ]]; then + cnt_fail=$(( cnt_fail + 1 )) + echo "not ok ${cnt_total} ${arg} # exit=${rc}" + fi +} + vm_ssh() { ssh -q -o UserKnownHostsFile=/dev/null -p ${SSH_HOST_PORT} localhost "$@" return $? @@ -530,17 +550,7 @@ cnt_total=0 for arg in "${ARGS[@]}"; do run_test "${arg}" rc=$? - if [[ ${rc} -eq $KSFT_PASS ]]; then - cnt_pass=$(( cnt_pass + 1 )) - echo "ok ${cnt_total} ${arg}" - elif [[ ${rc} -eq $KSFT_SKIP ]]; then - cnt_skip=$(( cnt_skip + 1 )) - echo "ok ${cnt_total} ${arg} # SKIP" - elif [[ ${rc} -eq $KSFT_FAIL ]]; then - cnt_fail=$(( cnt_fail + 1 )) - echo "not ok ${cnt_total} ${arg} # exit=$rc" - fi - cnt_total=$(( cnt_total + 1 )) + check_result "${rc}" "${arg}" done terminate_pidfiles "${pidfile}" From d13fb04a4bf0811b593843f7d2385fac5cf313c8 Mon Sep 17 00:00:00 2001 From: Bobby Eshleman Date: Sat, 8 Nov 2025 08:00:59 -0800 Subject: [PATCH 0745/1381] selftests/vsock: identify and execute tests that can re-use VM In preparation for future patches that introduce tests that cannot re-use the same VM, add functions to identify those that *can* re-use a VM. By continuing to re-use the same VM for these tests we can save time by avoiding the delay of booting a VM for every test. Reviewed-by: Simon Horman Signed-off-by: Bobby Eshleman Reviewed-by: Stefano Garzarella Link: https://patch.msgid.link/20251108-vsock-selftests-fixes-and-improvements-v4-8-d5e8d6c87289@meta.com Signed-off-by: Jakub Kicinski --- tools/testing/selftests/vsock/vmtest.sh | 63 ++++++++++++++++++++----- 1 file changed, 50 insertions(+), 13 deletions(-) diff --git a/tools/testing/selftests/vsock/vmtest.sh b/tools/testing/selftests/vsock/vmtest.sh index 2dd9bbb8c4a9..a1c2969c44b6 100755 --- a/tools/testing/selftests/vsock/vmtest.sh +++ b/tools/testing/selftests/vsock/vmtest.sh @@ -46,6 +46,8 @@ readonly TEST_DESCS=( "Run vsock_test using the loopback transport in the VM." ) +readonly USE_SHARED_VM=(vm_server_host_client vm_client_host_server vm_loopback) + VERBOSE=0 usage() { @@ -461,7 +463,44 @@ test_vm_loopback() { return "${KSFT_PASS}" } -run_test() { +shared_vm_test() { + local tname + + tname="${1}" + + for testname in "${USE_SHARED_VM[@]}"; do + if [[ "${tname}" == "${testname}" ]]; then + return 0 + fi + done + + return 1 +} + +shared_vm_tests_requested() { + for arg in "$@"; do + if shared_vm_test "${arg}"; then + return 0 + fi + done + + return 1 +} + +run_shared_vm_tests() { + local arg + + for arg in "$@"; do + if ! shared_vm_test "${arg}"; then + continue + fi + + run_shared_vm_test "${arg}" + check_result "$?" "${arg}" + done +} + +run_shared_vm_test() { local host_oops_cnt_before local host_warn_cnt_before local vm_oops_cnt_before @@ -537,23 +576,21 @@ handle_build echo "1..${#ARGS[@]}" -log_host "Booting up VM" -pidfile="$(create_pidfile)" -vm_start "${pidfile}" -vm_wait_for_ssh -log_host "VM booted up" - cnt_pass=0 cnt_fail=0 cnt_skip=0 cnt_total=0 -for arg in "${ARGS[@]}"; do - run_test "${arg}" - rc=$? - check_result "${rc}" "${arg}" -done -terminate_pidfiles "${pidfile}" +if shared_vm_tests_requested "${ARGS[@]}"; then + log_host "Booting up VM" + pidfile="$(create_pidfile)" + vm_start "${pidfile}" + vm_wait_for_ssh + log_host "VM booted up" + + run_shared_vm_tests "${ARGS[@]}" + terminate_pidfiles "${pidfile}" +fi echo "SUMMARY: PASS=${cnt_pass} SKIP=${cnt_skip} FAIL=${cnt_fail}" echo "Log: ${LOG}" From 592e3d14cecf8ffebe718586f02007a082aa3193 Mon Sep 17 00:00:00 2001 From: Bobby Eshleman Date: Sat, 8 Nov 2025 08:01:00 -0800 Subject: [PATCH 0746/1381] selftests/vsock: add BUILD=0 definition Add the definition for BUILD and initialize it to zero. This avoids 'bash -u vmtest.sh` from throwing 'unbound variable' when BUILD is not set to 1 and is later checked for its value. Signed-off-by: Bobby Eshleman Reviewed-by: Stefano Garzarella Link: https://patch.msgid.link/20251108-vsock-selftests-fixes-and-improvements-v4-9-d5e8d6c87289@meta.com Signed-off-by: Jakub Kicinski --- tools/testing/selftests/vsock/vmtest.sh | 1 + 1 file changed, 1 insertion(+) diff --git a/tools/testing/selftests/vsock/vmtest.sh b/tools/testing/selftests/vsock/vmtest.sh index a1c2969c44b6..e961b65b4c6e 100755 --- a/tools/testing/selftests/vsock/vmtest.sh +++ b/tools/testing/selftests/vsock/vmtest.sh @@ -548,6 +548,7 @@ run_shared_vm_test() { return "${rc}" } +BUILD=0 QEMU="qemu-system-$(uname -m)" while getopts :hvsq:b o From 67422ef38f8cc1116c28a74c144c1ab3cbbdb3bb Mon Sep 17 00:00:00 2001 From: Bobby Eshleman Date: Sat, 8 Nov 2025 08:01:01 -0800 Subject: [PATCH 0747/1381] selftests/vsock: add 1.37 to tested virtme-ng versions Testing with 1.37 shows all tests passing but emits the warning: warning: vng version 'virtme-ng 1.37' has not been tested and may not function properly. The following versions have been tested: 1.33 1.36 This patch adds 1.37 to the virtme-ng versions to get rid of the above warning. Reviewed-by: Simon Horman Reviewed-by: Stefano Garzarella Signed-off-by: Bobby Eshleman Link: https://patch.msgid.link/20251108-vsock-selftests-fixes-and-improvements-v4-10-d5e8d6c87289@meta.com Signed-off-by: Jakub Kicinski --- tools/testing/selftests/vsock/vmtest.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/testing/selftests/vsock/vmtest.sh b/tools/testing/selftests/vsock/vmtest.sh index e961b65b4c6e..b611172da09e 100755 --- a/tools/testing/selftests/vsock/vmtest.sh +++ b/tools/testing/selftests/vsock/vmtest.sh @@ -156,7 +156,7 @@ check_vng() { local version local ok - tested_versions=("1.33" "1.36") + tested_versions=("1.33" "1.36" "1.37") version="$(vng --version)" ok=0 From 338c5ddf4c889926e931067418797cd1711dd939 Mon Sep 17 00:00:00 2001 From: Bobby Eshleman Date: Sat, 8 Nov 2025 08:01:02 -0800 Subject: [PATCH 0748/1381] selftests/vsock: add vsock_loopback module loading Add vsock_loopback module loading to the loopback test so that vmtest.sh can be used for kernels built with loopback as a module. This is not technically a fix as kselftest expects loopback to be built-in already (defined in selftests/vsock/config). This is useful only for using vmtest.sh outside of kselftest. Reviewed-by: Stefano Garzarella Signed-off-by: Bobby Eshleman Link: https://patch.msgid.link/20251108-vsock-selftests-fixes-and-improvements-v4-11-d5e8d6c87289@meta.com Signed-off-by: Jakub Kicinski --- tools/testing/selftests/vsock/vmtest.sh | 2 ++ 1 file changed, 2 insertions(+) diff --git a/tools/testing/selftests/vsock/vmtest.sh b/tools/testing/selftests/vsock/vmtest.sh index b611172da09e..42e155b45602 100755 --- a/tools/testing/selftests/vsock/vmtest.sh +++ b/tools/testing/selftests/vsock/vmtest.sh @@ -452,6 +452,8 @@ test_vm_client_host_server() { test_vm_loopback() { local port=60000 # non-forwarded local port + vm_ssh -- modprobe vsock_loopback &> /dev/null || : + if ! vm_vsock_test "server" 1 "${port}"; then return "${KSFT_FAIL}" fi From 99f932c9050911fab8ac82455fc4d74b1c0200af Mon Sep 17 00:00:00 2001 From: Bobby Eshleman Date: Sat, 8 Nov 2025 08:01:03 -0800 Subject: [PATCH 0749/1381] selftests/vsock: disable shellcheck SC2317 and SC2119 Disable shellcheck rules SC2317 and SC2119. These rules are being triggered due to false positives. For SC2317, many `return "${KSFT_PASS}"` lines are reported as unreachable, even though they are executed during normal runs. For SC2119, the fact that log_guest/log_host accept either stdin or arguments triggers SC2119, despite being valid. Signed-off-by: Bobby Eshleman Reviewed-by: Stefano Garzarella Link: https://patch.msgid.link/20251108-vsock-selftests-fixes-and-improvements-v4-12-d5e8d6c87289@meta.com Signed-off-by: Jakub Kicinski --- tools/testing/selftests/vsock/vmtest.sh | 2 ++ 1 file changed, 2 insertions(+) diff --git a/tools/testing/selftests/vsock/vmtest.sh b/tools/testing/selftests/vsock/vmtest.sh index 42e155b45602..c7b270dd77a9 100755 --- a/tools/testing/selftests/vsock/vmtest.sh +++ b/tools/testing/selftests/vsock/vmtest.sh @@ -7,6 +7,8 @@ # * virtme-ng # * busybox-static (used by virtme-ng) # * qemu (used by virtme-ng) +# +# shellcheck disable=SC2317,SC2119 readonly SCRIPT_DIR="$(cd -P -- "$(dirname -- "${BASH_SOURCE[0]}")" && pwd -P)" readonly KERNEL_CHECKOUT=$(realpath "${SCRIPT_DIR}"/../../../../) From 26b8986a18c1f8c277659340e59c59d5f17d927f Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 11 Nov 2025 15:12:35 +0000 Subject: [PATCH 0750/1381] net: clear skb->sk in skb_release_head_state() skb_release_head_state() inlines skb_orphan(). We need to clear skb->sk otherwise we can freeze TCP flows on a mostly idle host, because skb_fclone_busy() would return true as long as the packet is not yet processed by skb_defer_free_flush(). Fixes: 1fcf572211da ("net: allow skb_release_head_state() to be called multiple times") Fixes: e20dfbad8aab ("net: fix napi_consume_skb() with alien skbs") Signed-off-by: Eric Dumazet Tested-by: Aditya Garg Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20251111151235.1903659-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/core/skbuff.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 4f4d7ab7057f..f34372666e67 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -1152,6 +1152,7 @@ void skb_release_head_state(struct sk_buff *skb) #endif skb->destructor = NULL; + skb->sk = NULL; } nf_reset_ct(skb); skb_ext_reset(skb); From 1479493c91fc14b8387db01c0442cd1508c325e6 Mon Sep 17 00:00:00 2001 From: javen Date: Tue, 11 Nov 2025 17:28:51 +0800 Subject: [PATCH 0751/1381] r8169: add support for RTL8125K This adds support for chip RTL8125K. Its XID is 0x68a. It is basically based on the one with XID 0x688, but with different firmware file. Signed-off-by: javen Reviewed-by: Heiner Kallweit Link: https://patch.msgid.link/20251111092851.3371-1-javen_xu@realsil.com.cn Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/realtek/r8169_main.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c index 2a4d9b548810..0b96b6aa4214 100644 --- a/drivers/net/ethernet/realtek/r8169_main.c +++ b/drivers/net/ethernet/realtek/r8169_main.c @@ -57,6 +57,7 @@ #define FIRMWARE_8125B_2 "rtl_nic/rtl8125b-2.fw" #define FIRMWARE_8125D_1 "rtl_nic/rtl8125d-1.fw" #define FIRMWARE_8125D_2 "rtl_nic/rtl8125d-2.fw" +#define FIRMWARE_8125K_1 "rtl_nic/rtl8125k-1.fw" #define FIRMWARE_8125BP_2 "rtl_nic/rtl8125bp-2.fw" #define FIRMWARE_8126A_2 "rtl_nic/rtl8126a-2.fw" #define FIRMWARE_8126A_3 "rtl_nic/rtl8126a-3.fw" @@ -110,6 +111,7 @@ static const struct rtl_chip_info { { 0x7cf, 0x681, RTL_GIGA_MAC_VER_66, "RTL8125BP", FIRMWARE_8125BP_2 }, /* 8125D family. */ + { 0x7cf, 0x68a, RTL_GIGA_MAC_VER_64, "RTL8125K", FIRMWARE_8125K_1 }, { 0x7cf, 0x689, RTL_GIGA_MAC_VER_64, "RTL8125D", FIRMWARE_8125D_2 }, { 0x7cf, 0x688, RTL_GIGA_MAC_VER_64, "RTL8125D", FIRMWARE_8125D_1 }, @@ -770,6 +772,7 @@ MODULE_FIRMWARE(FIRMWARE_8125A_3); MODULE_FIRMWARE(FIRMWARE_8125B_2); MODULE_FIRMWARE(FIRMWARE_8125D_1); MODULE_FIRMWARE(FIRMWARE_8125D_2); +MODULE_FIRMWARE(FIRMWARE_8125K_1); MODULE_FIRMWARE(FIRMWARE_8125BP_2); MODULE_FIRMWARE(FIRMWARE_8126A_2); MODULE_FIRMWARE(FIRMWARE_8126A_3); From 7e975caa0f7bf2f83a5e18b4ae69c3d7fff4eafa Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Tue, 11 Nov 2025 11:26:34 +0000 Subject: [PATCH 0752/1381] net: stmmac: improve ndev->max_mtu setup readability Improve the readibility of the code setting ndev->max_mtu. This depends on the hardware specific maximum defined by the MAC core, and also a platform provided maximum. The code was originally checking that the platform specific maximum was between ndev->min_mtu..MAC core maximum before reducing ndev->max_mtu, otherwise if the platform specific maximum was less than ndev->min_mtu, issuing a warning. Re-order the code to handle the case where the platform specific max is below ndev->min_mtu, which then means that the subsequent test is simply reducing ndev->max_mtu. Update the comment, and add a few blank lines to separate the blocks of code. Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vImWA-0000000DrIl-1HZY@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../net/ethernet/stmicro/stmmac/stmmac_main.c | 22 ++++++++++--------- 1 file changed, 12 insertions(+), 10 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index ccf383b355e7..eb4350193996 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -1392,9 +1392,9 @@ static unsigned int stmmac_rx_offset(struct stmmac_priv *priv) return NET_SKB_PAD; } -static int stmmac_set_bfsize(int mtu, int bufsize) +static int stmmac_set_bfsize(int mtu) { - int ret = bufsize; + int ret; if (mtu >= BUF_SIZE_8KiB) ret = BUF_SIZE_16KiB; @@ -3958,12 +3958,13 @@ stmmac_setup_dma_desc(struct stmmac_priv *priv, unsigned int mtu) return ERR_PTR(-ENOMEM); } + /* Returns 0 or BUF_SIZE_16KiB if mtu > 8KiB and dwmac4 or ring mode */ bfsize = stmmac_set_16kib_bfsize(priv, mtu); if (bfsize < 0) bfsize = 0; if (bfsize < BUF_SIZE_16KiB) - bfsize = stmmac_set_bfsize(mtu, 0); + bfsize = stmmac_set_bfsize(mtu); dma_conf->dma_buf_sz = bfsize; /* Chose the tx/rx size from the already defined one in the @@ -7773,22 +7774,23 @@ int stmmac_dvr_probe(struct device *device, /* MTU range: 46 - hw-specific max */ ndev->min_mtu = ETH_ZLEN - ETH_HLEN; + if (priv->plat->core_type == DWMAC_CORE_XGMAC) ndev->max_mtu = XGMAC_JUMBO_LEN; - else if ((priv->plat->enh_desc) || (priv->synopsys_id >= DWMAC_CORE_4_00)) + else if (priv->plat->enh_desc || priv->synopsys_id >= DWMAC_CORE_4_00) ndev->max_mtu = JUMBO_LEN; else ndev->max_mtu = SKB_MAX_HEAD(NET_SKB_PAD + NET_IP_ALIGN); - /* Will not overwrite ndev->max_mtu if plat->maxmtu > ndev->max_mtu - * as well as plat->maxmtu < ndev->min_mtu which is a invalid range. + + /* Warn if the platform's maxmtu is smaller than the minimum MTU, + * otherwise clamp the maximum MTU above to the platform's maxmtu. */ - if ((priv->plat->maxmtu < ndev->max_mtu) && - (priv->plat->maxmtu >= ndev->min_mtu)) - ndev->max_mtu = priv->plat->maxmtu; - else if (priv->plat->maxmtu < ndev->min_mtu) + if (priv->plat->maxmtu < ndev->min_mtu) dev_warn(priv->device, "%s: warning: maxmtu having invalid value (%d)\n", __func__, priv->plat->maxmtu); + else if (priv->plat->maxmtu < ndev->max_mtu) + ndev->max_mtu = priv->plat->maxmtu; ndev->priv_flags |= IFF_LIVE_ADDR_CHANGE; From c902835fc6eb7e13dbffc404210eed59293425c8 Mon Sep 17 00:00:00 2001 From: Gregor Herburger Date: Wed, 1 Oct 2025 14:40:03 +0530 Subject: [PATCH 0753/1381] can: mcp251xfd: add workaround for errata 5 According to Errata DS80000789E 5 writing IOCON register using one SPI write command clears LAT0/LAT1. Errata Fix/Work Around suggests to write registers with single byte write instructions. However, it seems that every write to the second byte causes the overwrite of LAT0/LAT1. Never write byte 2 of IOCON register to avoid clearing of LAT0/LAT1. Signed-off-by: Gregor Herburger Tested-by: Viken Dadhaniya Signed-off-by: Viken Dadhaniya Reviewed-by: Manivannan Sadhasivam Link: https://patch.msgid.link/20251001091006.4003841-4-viken.dadhaniya@oss.qualcomm.com [mkl: add missing MCP251XFD_REG_IOCON_GPIO_MASK] Signed-off-by: Marc Kleine-Budde --- .../net/can/spi/mcp251xfd/mcp251xfd-regmap.c | 89 +++++++++++++++++-- drivers/net/can/spi/mcp251xfd/mcp251xfd.h | 1 + 2 files changed, 84 insertions(+), 6 deletions(-) diff --git a/drivers/net/can/spi/mcp251xfd/mcp251xfd-regmap.c b/drivers/net/can/spi/mcp251xfd/mcp251xfd-regmap.c index e61cbd209955..70d5ff0ae7ac 100644 --- a/drivers/net/can/spi/mcp251xfd/mcp251xfd-regmap.c +++ b/drivers/net/can/spi/mcp251xfd/mcp251xfd-regmap.c @@ -13,9 +13,9 @@ static const struct regmap_config mcp251xfd_regmap_crc; static int -mcp251xfd_regmap_nocrc_gather_write(void *context, - const void *reg, size_t reg_len, - const void *val, size_t val_len) +_mcp251xfd_regmap_nocrc_gather_write(void *context, + const void *reg, size_t reg_len, + const void *val, size_t val_len) { struct spi_device *spi = context; struct mcp251xfd_priv *priv = spi_get_drvdata(spi); @@ -39,6 +39,45 @@ mcp251xfd_regmap_nocrc_gather_write(void *context, return spi_sync_transfer(spi, xfer, ARRAY_SIZE(xfer)); } +static int +mcp251xfd_regmap_nocrc_gather_write(void *context, + const void *reg_p, size_t reg_len, + const void *val, size_t val_len) +{ + const u16 byte_exclude = MCP251XFD_REG_IOCON + + mcp251xfd_first_byte_set(MCP251XFD_REG_IOCON_GPIO_MASK); + u16 reg = be16_to_cpu(*(__be16 *)reg_p) & MCP251XFD_SPI_ADDRESS_MASK; + int ret; + + /* Never write to bits 16..23 of IOCON register to avoid clearing of LAT0/LAT1 + * + * According to MCP2518FD Errata DS80000789E 5 writing IOCON register using one + * SPI write command clears LAT0/LAT1. + * + * Errata Fix/Work Around suggests to write registers with single byte + * write instructions. However, it seems that the byte at 0xe06(IOCON[23:16]) + * is for read-only access and writing to it causes the clearing of LAT0/LAT1. + */ + if (reg <= byte_exclude && reg + val_len > byte_exclude) { + size_t len = byte_exclude - reg; + + /* Write up to 0xe05 */ + ret = _mcp251xfd_regmap_nocrc_gather_write(context, reg_p, reg_len, val, len); + if (ret) + return ret; + + /* Write from 0xe07 on */ + reg += len + 1; + reg = (__force unsigned short)cpu_to_be16(MCP251XFD_SPI_INSTRUCTION_WRITE | reg); + return _mcp251xfd_regmap_nocrc_gather_write(context, ®, reg_len, + val + len + 1, + val_len - len - 1); + } + + return _mcp251xfd_regmap_nocrc_gather_write(context, reg_p, reg_len, + val, val_len); +} + static int mcp251xfd_regmap_nocrc_write(void *context, const void *data, size_t count) { @@ -197,9 +236,9 @@ mcp251xfd_regmap_nocrc_read(void *context, } static int -mcp251xfd_regmap_crc_gather_write(void *context, - const void *reg_p, size_t reg_len, - const void *val, size_t val_len) +_mcp251xfd_regmap_crc_gather_write(void *context, + const void *reg_p, size_t reg_len, + const void *val, size_t val_len) { struct spi_device *spi = context; struct mcp251xfd_priv *priv = spi_get_drvdata(spi); @@ -230,6 +269,44 @@ mcp251xfd_regmap_crc_gather_write(void *context, return spi_sync_transfer(spi, xfer, ARRAY_SIZE(xfer)); } +static int +mcp251xfd_regmap_crc_gather_write(void *context, + const void *reg_p, size_t reg_len, + const void *val, size_t val_len) +{ + const u16 byte_exclude = MCP251XFD_REG_IOCON + + mcp251xfd_first_byte_set(MCP251XFD_REG_IOCON_GPIO_MASK); + u16 reg = *(u16 *)reg_p; + int ret; + + /* Never write to bits 16..23 of IOCON register to avoid clearing of LAT0/LAT1 + * + * According to MCP2518FD Errata DS80000789E 5 writing IOCON register using one + * SPI write command clears LAT0/LAT1. + * + * Errata Fix/Work Around suggests to write registers with single byte + * write instructions. However, it seems that the byte at 0xe06(IOCON[23:16]) + * is for read-only access and writing to it causes the clearing of LAT0/LAT1. + */ + if (reg <= byte_exclude && reg + val_len > byte_exclude) { + size_t len = byte_exclude - reg; + + /* Write up to 0xe05 */ + ret = _mcp251xfd_regmap_crc_gather_write(context, ®, reg_len, val, len); + if (ret) + return ret; + + /* Write from 0xe07 on */ + reg += len + 1; + return _mcp251xfd_regmap_crc_gather_write(context, ®, reg_len, + val + len + 1, + val_len - len - 1); + } + + return _mcp251xfd_regmap_crc_gather_write(context, reg_p, reg_len, + val, val_len); +} + static int mcp251xfd_regmap_crc_write(void *context, const void *data, size_t count) diff --git a/drivers/net/can/spi/mcp251xfd/mcp251xfd.h b/drivers/net/can/spi/mcp251xfd/mcp251xfd.h index dcbbd2b2fae8..e63034fd5947 100644 --- a/drivers/net/can/spi/mcp251xfd/mcp251xfd.h +++ b/drivers/net/can/spi/mcp251xfd/mcp251xfd.h @@ -337,6 +337,7 @@ #define MCP251XFD_REG_IOCON_PM0 BIT(24) #define MCP251XFD_REG_IOCON_GPIO1 BIT(17) #define MCP251XFD_REG_IOCON_GPIO0 BIT(16) +#define MCP251XFD_REG_IOCON_GPIO_MASK GENMASK(17, 16) #define MCP251XFD_REG_IOCON_LAT1 BIT(9) #define MCP251XFD_REG_IOCON_LAT0 BIT(8) #define MCP251XFD_REG_IOCON_XSTBYEN BIT(6) From d35fa005f5e68b6c1730cc64438ef4c81635a733 Mon Sep 17 00:00:00 2001 From: Gregor Herburger Date: Wed, 1 Oct 2025 14:40:04 +0530 Subject: [PATCH 0754/1381] can: mcp251xfd: only configure PIN1 when rx_int is set When rx_int is used th mcp251xfd_chip_rx_int_enable and mcp251xfd_chip_rx_int_disable function configure both PIN0 and PIN1. To prepare the support of the GPIOS only configure PIN1 with regmap_update_bits. This way PIN0 can be used as GPIO while PIN1 is used as rx_int interrupt. Signed-off-by: Gregor Herburger Tested-by: Viken Dadhaniya Signed-off-by: Viken Dadhaniya Reviewed-by: Manivannan Sadhasivam Link: https://patch.msgid.link/20251001091006.4003841-5-viken.dadhaniya@oss.qualcomm.com Signed-off-by: Marc Kleine-Budde --- .../net/can/spi/mcp251xfd/mcp251xfd-core.c | 22 +++++++------------ drivers/net/can/spi/mcp251xfd/mcp251xfd.h | 5 +++++ 2 files changed, 13 insertions(+), 14 deletions(-) diff --git a/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c b/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c index 9104561e0995..565327bf9c43 100644 --- a/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c +++ b/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c @@ -608,23 +608,21 @@ static int mcp251xfd_set_bittiming(const struct mcp251xfd_priv *priv) static int mcp251xfd_chip_rx_int_enable(const struct mcp251xfd_priv *priv) { - u32 val; + u32 val, mask; if (!priv->rx_int) return 0; - /* Configure GPIOs: - * - PIN0: GPIO Input - * - PIN1: GPIO Input/RX Interrupt + /* Configure PIN1 as RX Interrupt: * * PIN1 must be Input, otherwise there is a glitch on the * rx-INT line. It happens between setting the PIN as output * (in the first byte of the SPI transfer) and configuring the * PIN as interrupt (in the last byte of the SPI transfer). */ - val = MCP251XFD_REG_IOCON_PM0 | MCP251XFD_REG_IOCON_TRIS1 | - MCP251XFD_REG_IOCON_TRIS0; - return regmap_write(priv->map_reg, MCP251XFD_REG_IOCON, val); + val = MCP251XFD_REG_IOCON_TRIS(1); + mask = MCP251XFD_REG_IOCON_TRIS(1) | MCP251XFD_REG_IOCON_PM(1); + return regmap_update_bits(priv->map_reg, MCP251XFD_REG_IOCON, mask, val); } static int mcp251xfd_chip_rx_int_disable(const struct mcp251xfd_priv *priv) @@ -634,13 +632,9 @@ static int mcp251xfd_chip_rx_int_disable(const struct mcp251xfd_priv *priv) if (!priv->rx_int) return 0; - /* Configure GPIOs: - * - PIN0: GPIO Input - * - PIN1: GPIO Input - */ - val = MCP251XFD_REG_IOCON_PM1 | MCP251XFD_REG_IOCON_PM0 | - MCP251XFD_REG_IOCON_TRIS1 | MCP251XFD_REG_IOCON_TRIS0; - return regmap_write(priv->map_reg, MCP251XFD_REG_IOCON, val); + /* Configure PIN1 as GPIO Input */ + val = MCP251XFD_REG_IOCON_PM(1) | MCP251XFD_REG_IOCON_TRIS(1); + return regmap_update_bits(priv->map_reg, MCP251XFD_REG_IOCON, val, val); } static int mcp251xfd_chip_ecc_init(struct mcp251xfd_priv *priv) diff --git a/drivers/net/can/spi/mcp251xfd/mcp251xfd.h b/drivers/net/can/spi/mcp251xfd/mcp251xfd.h index e63034fd5947..bd28510a6583 100644 --- a/drivers/net/can/spi/mcp251xfd/mcp251xfd.h +++ b/drivers/net/can/spi/mcp251xfd/mcp251xfd.h @@ -335,14 +335,19 @@ #define MCP251XFD_REG_IOCON_TXCANOD BIT(28) #define MCP251XFD_REG_IOCON_PM1 BIT(25) #define MCP251XFD_REG_IOCON_PM0 BIT(24) +#define MCP251XFD_REG_IOCON_PM(n) (MCP251XFD_REG_IOCON_PM0 << (n)) #define MCP251XFD_REG_IOCON_GPIO1 BIT(17) #define MCP251XFD_REG_IOCON_GPIO0 BIT(16) +#define MCP251XFD_REG_IOCON_GPIO(n) (MCP251XFD_REG_IOCON_GPIO0 << (n)) #define MCP251XFD_REG_IOCON_GPIO_MASK GENMASK(17, 16) #define MCP251XFD_REG_IOCON_LAT1 BIT(9) #define MCP251XFD_REG_IOCON_LAT0 BIT(8) +#define MCP251XFD_REG_IOCON_LAT(n) (MCP251XFD_REG_IOCON_LAT0 << (n)) +#define MCP251XFD_REG_IOCON_LAT_MASK GENMASK(9, 8) #define MCP251XFD_REG_IOCON_XSTBYEN BIT(6) #define MCP251XFD_REG_IOCON_TRIS1 BIT(1) #define MCP251XFD_REG_IOCON_TRIS0 BIT(0) +#define MCP251XFD_REG_IOCON_TRIS(n) (MCP251XFD_REG_IOCON_TRIS0 << (n)) #define MCP251XFD_REG_CRC 0xe08 #define MCP251XFD_REG_CRC_FERRIE BIT(25) From c6106336ec2bad2084df84e5b9ad039b0622c0b8 Mon Sep 17 00:00:00 2001 From: Gregor Herburger Date: Wed, 1 Oct 2025 14:40:05 +0530 Subject: [PATCH 0755/1381] can: mcp251xfd: add gpio functionality The mcp251xfd devices allow two pins to be configured as gpio. Add this functionality to driver. Acked-by: Bartosz Golaszewski Signed-off-by: Gregor Herburger Tested-by: Viken Dadhaniya Signed-off-by: Viken Dadhaniya Reviewed-by: Manivannan Sadhasivam Link: https://patch.msgid.link/20251001091006.4003841-6-viken.dadhaniya@oss.qualcomm.com Signed-off-by: Marc Kleine-Budde --- drivers/net/can/spi/mcp251xfd/Kconfig | 1 + .../net/can/spi/mcp251xfd/mcp251xfd-core.c | 160 ++++++++++++++++++ drivers/net/can/spi/mcp251xfd/mcp251xfd.h | 2 + 3 files changed, 163 insertions(+) diff --git a/drivers/net/can/spi/mcp251xfd/Kconfig b/drivers/net/can/spi/mcp251xfd/Kconfig index 877e4356010d..7c29846e6051 100644 --- a/drivers/net/can/spi/mcp251xfd/Kconfig +++ b/drivers/net/can/spi/mcp251xfd/Kconfig @@ -5,6 +5,7 @@ config CAN_MCP251XFD select CAN_RX_OFFLOAD select REGMAP select WANT_DEV_COREDUMP + select GPIOLIB help Driver for the Microchip MCP251XFD SPI FD-CAN controller family. diff --git a/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c b/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c index 565327bf9c43..5134ebb85880 100644 --- a/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c +++ b/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c @@ -1797,6 +1797,160 @@ static int mcp251xfd_register_check_rx_int(struct mcp251xfd_priv *priv) return 0; } +static const char * const mcp251xfd_gpio_names[] = { "GPIO0", "GPIO1" }; + +static int mcp251xfd_gpio_request(struct gpio_chip *chip, unsigned int offset) +{ + struct mcp251xfd_priv *priv = gpiochip_get_data(chip); + u32 pin_mask = MCP251XFD_REG_IOCON_PM(offset); + int ret; + + if (priv->rx_int && offset == 1) { + netdev_err(priv->ndev, "Can't use GPIO 1 with RX-INT!\n"); + return -EINVAL; + } + + ret = pm_runtime_resume_and_get(priv->ndev->dev.parent); + if (ret) + return ret; + + return regmap_update_bits(priv->map_reg, MCP251XFD_REG_IOCON, pin_mask, pin_mask); +} + +static void mcp251xfd_gpio_free(struct gpio_chip *chip, unsigned int offset) +{ + struct mcp251xfd_priv *priv = gpiochip_get_data(chip); + + pm_runtime_put(priv->ndev->dev.parent); +} + +static int mcp251xfd_gpio_get_direction(struct gpio_chip *chip, + unsigned int offset) +{ + struct mcp251xfd_priv *priv = gpiochip_get_data(chip); + u32 mask = MCP251XFD_REG_IOCON_TRIS(offset); + u32 val; + int ret; + + ret = regmap_read(priv->map_reg, MCP251XFD_REG_IOCON, &val); + if (ret) + return ret; + + if (mask & val) + return GPIO_LINE_DIRECTION_IN; + + return GPIO_LINE_DIRECTION_OUT; +} + +static int mcp251xfd_gpio_get(struct gpio_chip *chip, unsigned int offset) +{ + struct mcp251xfd_priv *priv = gpiochip_get_data(chip); + u32 mask = MCP251XFD_REG_IOCON_GPIO(offset); + u32 val; + int ret; + + ret = regmap_read(priv->map_reg, MCP251XFD_REG_IOCON, &val); + if (ret) + return ret; + + return !!(mask & val); +} + +static int mcp251xfd_gpio_get_multiple(struct gpio_chip *chip, unsigned long *mask, + unsigned long *bit) +{ + struct mcp251xfd_priv *priv = gpiochip_get_data(chip); + u32 val; + int ret; + + ret = regmap_read(priv->map_reg, MCP251XFD_REG_IOCON, &val); + if (ret) + return ret; + + *bit = FIELD_GET(MCP251XFD_REG_IOCON_GPIO_MASK, val) & *mask; + + return 0; +} + +static int mcp251xfd_gpio_direction_output(struct gpio_chip *chip, + unsigned int offset, int value) +{ + struct mcp251xfd_priv *priv = gpiochip_get_data(chip); + u32 dir_mask = MCP251XFD_REG_IOCON_TRIS(offset); + u32 val_mask = MCP251XFD_REG_IOCON_LAT(offset); + u32 val; + + if (value) + val = val_mask; + else + val = 0; + + return regmap_update_bits(priv->map_reg, MCP251XFD_REG_IOCON, + dir_mask | val_mask, val); +} + +static int mcp251xfd_gpio_direction_input(struct gpio_chip *chip, + unsigned int offset) +{ + struct mcp251xfd_priv *priv = gpiochip_get_data(chip); + u32 dir_mask = MCP251XFD_REG_IOCON_TRIS(offset); + + return regmap_update_bits(priv->map_reg, MCP251XFD_REG_IOCON, dir_mask, dir_mask); +} + +static int mcp251xfd_gpio_set(struct gpio_chip *chip, unsigned int offset, int value) +{ + struct mcp251xfd_priv *priv = gpiochip_get_data(chip); + u32 val_mask = MCP251XFD_REG_IOCON_LAT(offset); + u32 val; + + if (value) + val = val_mask; + else + val = 0; + + return regmap_update_bits(priv->map_reg, MCP251XFD_REG_IOCON, val_mask, val); +} + +static int mcp251xfd_gpio_set_multiple(struct gpio_chip *chip, unsigned long *mask, + unsigned long *bits) +{ + struct mcp251xfd_priv *priv = gpiochip_get_data(chip); + u32 val; + + val = FIELD_PREP(MCP251XFD_REG_IOCON_LAT_MASK, *bits); + + return regmap_update_bits(priv->map_reg, MCP251XFD_REG_IOCON, + MCP251XFD_REG_IOCON_LAT_MASK, val); +} + +static int mcp251fdx_gpio_setup(struct mcp251xfd_priv *priv) +{ + struct gpio_chip *gc = &priv->gc; + + if (!device_property_present(&priv->spi->dev, "gpio-controller")) + return 0; + + gc->label = dev_name(&priv->spi->dev); + gc->parent = &priv->spi->dev; + gc->owner = THIS_MODULE; + gc->request = mcp251xfd_gpio_request; + gc->free = mcp251xfd_gpio_free; + gc->get_direction = mcp251xfd_gpio_get_direction; + gc->direction_output = mcp251xfd_gpio_direction_output; + gc->direction_input = mcp251xfd_gpio_direction_input; + gc->get = mcp251xfd_gpio_get; + gc->get_multiple = mcp251xfd_gpio_get_multiple; + gc->set = mcp251xfd_gpio_set; + gc->set_multiple = mcp251xfd_gpio_set_multiple; + gc->base = -1; + gc->can_sleep = true; + gc->ngpio = ARRAY_SIZE(mcp251xfd_gpio_names); + gc->names = mcp251xfd_gpio_names; + + return devm_gpiochip_add_data(&priv->spi->dev, gc, priv); +} + static int mcp251xfd_register_get_dev_id(const struct mcp251xfd_priv *priv, u32 *dev_id, u32 *effective_speed_hz_slow, @@ -1930,6 +2084,12 @@ static int mcp251xfd_register(struct mcp251xfd_priv *priv) mcp251xfd_ethtool_init(priv); + err = mcp251fdx_gpio_setup(priv); + if (err) { + dev_err_probe(&priv->spi->dev, err, "Failed to register gpio-controller.\n"); + goto out_runtime_disable; + } + err = register_candev(ndev); if (err) goto out_runtime_disable; diff --git a/drivers/net/can/spi/mcp251xfd/mcp251xfd.h b/drivers/net/can/spi/mcp251xfd/mcp251xfd.h index bd28510a6583..085d7101e595 100644 --- a/drivers/net/can/spi/mcp251xfd/mcp251xfd.h +++ b/drivers/net/can/spi/mcp251xfd/mcp251xfd.h @@ -15,6 +15,7 @@ #include #include #include +#include #include #include #include @@ -676,6 +677,7 @@ struct mcp251xfd_priv { struct mcp251xfd_devtype_data devtype_data; struct can_berr_counter bec; + struct gpio_chip gc; }; #define MCP251XFD_IS(_model) \ From 6ece6b4c37479319410ff45b8a71b2c6e5e1c27a Mon Sep 17 00:00:00 2001 From: Gregor Herburger Date: Wed, 1 Oct 2025 14:40:06 +0530 Subject: [PATCH 0756/1381] dt-bindings: can: mcp251xfd: add gpio-controller property The mcp251xfd has two pins that can be used as gpio. Add gpio-controller property to binding description. Acked-by: Krzysztof Kozlowski Signed-off-by: Gregor Herburger Signed-off-by: Viken Dadhaniya Reviewed-by: Manivannan Sadhasivam Link: https://patch.msgid.link/20251001091006.4003841-7-viken.dadhaniya@oss.qualcomm.com Signed-off-by: Marc Kleine-Budde --- .../devicetree/bindings/net/can/microchip,mcp251xfd.yaml | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/Documentation/devicetree/bindings/net/can/microchip,mcp251xfd.yaml b/Documentation/devicetree/bindings/net/can/microchip,mcp251xfd.yaml index c155c9c6db39..2d13638ebc6a 100644 --- a/Documentation/devicetree/bindings/net/can/microchip,mcp251xfd.yaml +++ b/Documentation/devicetree/bindings/net/can/microchip,mcp251xfd.yaml @@ -49,6 +49,11 @@ properties: Must be half or less of "clocks" frequency. maximum: 20000000 + gpio-controller: true + + "#gpio-cells": + const: 2 + required: - compatible - reg From b305fbdad4ed7e66b5b3f76b15f71d05fa6af212 Mon Sep 17 00:00:00 2001 From: Chu Guangqing Date: Mon, 3 Nov 2025 15:40:09 +0800 Subject: [PATCH 0757/1381] can: bxcan: Fix a typo error for assign Fix the spelling error of "assign". Signed-off-by: Chu Guangqing Reviewed-by: Dario Binacchi Link: https://patch.msgid.link/20251103074009.4708-1-chuguangqing@inspur.com Signed-off-by: Marc Kleine-Budde --- drivers/net/can/bxcan.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/can/bxcan.c b/drivers/net/can/bxcan.c index 0b579e7bb3b6..baf494d20bef 100644 --- a/drivers/net/can/bxcan.c +++ b/drivers/net/can/bxcan.c @@ -227,7 +227,7 @@ static void bxcan_enable_filters(struct bxcan_priv *priv, enum bxcan_cfg cfg) * mask mode with 32 bits width. */ - /* Enter filter initialization mode and assing filters to CAN + /* Enter filter initialization mode and assign filters to CAN * controllers. */ regmap_update_bits(priv->gcan, BXCAN_FMR_REG, From dae8d7d63b740d8f5972b8438b139a6488e0f9fa Mon Sep 17 00:00:00 2001 From: Ping-Ke Shih Date: Tue, 11 Nov 2025 10:24:45 +0800 Subject: [PATCH 0758/1381] wifi: rtw89: pci: add to read PCI configuration space from common code Normally only access PCI device in pci.c. However for debug purpose, a set of registers predefined in firmware element including PCI configuration space should be read for diagnosis. Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251111022452.28093-2-pkshih@realtek.com --- drivers/net/wireless/realtek/rtw89/core.h | 13 +++++++++++++ drivers/net/wireless/realtek/rtw89/mac.h | 2 -- drivers/net/wireless/realtek/rtw89/pci.c | 16 ++++++++++++++++ 3 files changed, 29 insertions(+), 2 deletions(-) diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h index 71eb7078a0dc..e6584462f6c0 100644 --- a/drivers/net/wireless/realtek/rtw89/core.h +++ b/drivers/net/wireless/realtek/rtw89/core.h @@ -39,6 +39,8 @@ extern const struct ieee80211_ops rtw89_ops; #define RFREG_MASK 0xfffff #define INV_RF_DATA 0xffffffff #define BYPASS_CR_DATA 0xbabecafe +#define RTW89_R32_EA 0xEAEAEAEA +#define RTW89_R32_DEAD 0xDEADBEEF #define RTW89_TRACK_WORK_PERIOD round_jiffies_relative(HZ * 2) #define RTW89_TRACK_PS_WORK_PERIOD msecs_to_jiffies(100) @@ -3674,6 +3676,8 @@ struct rtw89_hci_ops { void (*write16)(struct rtw89_dev *rtwdev, u32 addr, u16 data); void (*write32)(struct rtw89_dev *rtwdev, u32 addr, u32 data); + u32 (*read32_pci_cfg)(struct rtw89_dev *rtwdev, u32 addr); + int (*mac_pre_init)(struct rtw89_dev *rtwdev); int (*mac_pre_deinit)(struct rtw89_dev *rtwdev); int (*mac_post_init)(struct rtw89_dev *rtwdev); @@ -6658,6 +6662,15 @@ rtw89_write_rf(struct rtw89_dev *rtwdev, enum rtw89_rf_path rf_path, mutex_unlock(&rtwdev->rf_mutex); } +static inline u32 rtw89_read32_pci_cfg(struct rtw89_dev *rtwdev, u32 addr) +{ + if (rtwdev->hci.type != RTW89_HCI_TYPE_PCIE || + !rtwdev->hci.ops->read32_pci_cfg) + return RTW89_R32_EA; + + return rtwdev->hci.ops->read32_pci_cfg(rtwdev, addr); +} + static inline struct ieee80211_txq *rtw89_txq_to_txq(struct rtw89_txq *rtwtxq) { void *p = rtwtxq; diff --git a/drivers/net/wireless/realtek/rtw89/mac.h b/drivers/net/wireless/realtek/rtw89/mac.h index 868413335f6c..3cc97fd0c0ec 100644 --- a/drivers/net/wireless/realtek/rtw89/mac.h +++ b/drivers/net/wireless/realtek/rtw89/mac.h @@ -581,8 +581,6 @@ enum rtw89_mac_bf_rrsc_rate { RTW89_MAC_BF_RRSC_MAX = 32 }; -#define RTW89_R32_EA 0xEAEAEAEA -#define RTW89_R32_DEAD 0xDEADBEEF #define MAC_REG_POOL_COUNT 10 #define ACCESS_CMAC(_addr) \ ({typeof(_addr) __addr = (_addr); \ diff --git a/drivers/net/wireless/realtek/rtw89/pci.c b/drivers/net/wireless/realtek/rtw89/pci.c index b1985193a18f..a66fcdb0293b 100644 --- a/drivers/net/wireless/realtek/rtw89/pci.c +++ b/drivers/net/wireless/realtek/rtw89/pci.c @@ -2064,6 +2064,20 @@ static void rtw89_pci_ops_write32(struct rtw89_dev *rtwdev, u32 addr, u32 data) writel(data, rtwpci->mmap + addr); } +static u32 rtw89_pci_ops_read32_pci_cfg(struct rtw89_dev *rtwdev, u32 addr) +{ + struct rtw89_pci *rtwpci = (struct rtw89_pci *)rtwdev->priv; + struct pci_dev *pdev = rtwpci->pdev; + u32 value; + int ret; + + ret = pci_read_config_dword(pdev, addr, &value); + if (ret) + return RTW89_R32_EA; + + return value; +} + static void rtw89_pci_ctrl_dma_trx(struct rtw89_dev *rtwdev, bool enable) { const struct rtw89_pci_info *info = rtwdev->pci_info; @@ -4683,6 +4697,8 @@ static const struct rtw89_hci_ops rtw89_pci_ops = { .write16 = rtw89_pci_ops_write16, .write32 = rtw89_pci_ops_write32, + .read32_pci_cfg = rtw89_pci_ops_read32_pci_cfg, + .mac_pre_init = rtw89_pci_ops_mac_pre_init, .mac_pre_deinit = rtw89_pci_ops_mac_pre_deinit, .mac_post_init = rtw89_pci_ops_mac_post_init, From de19cc7def5a9c646264a1e7a2a183a3baad112f Mon Sep 17 00:00:00 2001 From: Ping-Ke Shih Date: Tue, 11 Nov 2025 10:24:46 +0800 Subject: [PATCH 0759/1381] wifi: rtw89: fw: parse firmware element of DIAG_MAC The firmware element ID 28 is a set of rules to diagnose if MAC get abnormal. The latter patch will use these rules via debugfs to know the status. The element contains rules with their textual messages shown as below: +------------------------------------+ | | | +-----------+ | | | rule_size |-------|----------+ +----------------+-----------+-------+ -- | | rule[0] | \ | | rule[1] | | <---+ | : | / +------------------------------------+ -- | msg[0] msg[1] | each msg has variable length | msg[2] msg[3] ... | (with address align 2) | ... | +------------------------------------+ Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251111022452.28093-3-pkshih@realtek.com --- drivers/net/wireless/realtek/rtw89/core.h | 1 + drivers/net/wireless/realtek/rtw89/fw.c | 15 +++++++++++++++ drivers/net/wireless/realtek/rtw89/fw.h | 6 ++++++ 3 files changed, 22 insertions(+) diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h index e6584462f6c0..34c135269627 100644 --- a/drivers/net/wireless/realtek/rtw89/core.h +++ b/drivers/net/wireless/realtek/rtw89/core.h @@ -4728,6 +4728,7 @@ struct rtw89_fw_elm_info { struct rtw89_phy_rfk_log_fmt *rfk_log_fmt; const struct rtw89_regd_data *regd; const struct rtw89_fw_element_hdr *afe; + const struct rtw89_fw_element_hdr *diag_mac; }; enum rtw89_fw_mss_dev_type { diff --git a/drivers/net/wireless/realtek/rtw89/fw.c b/drivers/net/wireless/realtek/rtw89/fw.c index 96f0463e66d6..515e9f098380 100644 --- a/drivers/net/wireless/realtek/rtw89/fw.c +++ b/drivers/net/wireless/realtek/rtw89/fw.c @@ -1298,6 +1298,18 @@ int rtw89_build_afe_pwr_seq_from_elm(struct rtw89_dev *rtwdev, return 0; } +static +int rtw89_recognize_diag_mac_from_elm(struct rtw89_dev *rtwdev, + const struct rtw89_fw_element_hdr *elm, + const union rtw89_fw_element_arg arg) +{ + struct rtw89_fw_elm_info *elm_info = &rtwdev->fw.elm_info; + + elm_info->diag_mac = elm; + + return 0; +} + static const struct rtw89_fw_element_handler __fw_element_handlers[] = { [RTW89_FW_ELEMENT_ID_BBMCU0] = {__rtw89_fw_recognize_from_elm, { .fw_type = RTW89_FW_BBMCU0 }, NULL}, @@ -1386,6 +1398,9 @@ static const struct rtw89_fw_element_handler __fw_element_handlers[] = { [RTW89_FW_ELEMENT_ID_AFE_PWR_SEQ] = { rtw89_build_afe_pwr_seq_from_elm, {}, "AFE", }, + [RTW89_FW_ELEMENT_ID_DIAG_MAC] = { + rtw89_recognize_diag_mac_from_elm, {}, NULL, + }, }; int rtw89_fw_recognize_elements(struct rtw89_dev *rtwdev) diff --git a/drivers/net/wireless/realtek/rtw89/fw.h b/drivers/net/wireless/realtek/rtw89/fw.h index 310580206368..6620256e3f22 100644 --- a/drivers/net/wireless/realtek/rtw89/fw.h +++ b/drivers/net/wireless/realtek/rtw89/fw.h @@ -4026,6 +4026,7 @@ enum rtw89_fw_element_id { RTW89_FW_ELEMENT_ID_TXPWR_DA_LMT_RU_5GHZ = 25, RTW89_FW_ELEMENT_ID_TXPWR_DA_LMT_RU_6GHZ = 26, RTW89_FW_ELEMENT_ID_AFE_PWR_SEQ = 27, + RTW89_FW_ELEMENT_ID_DIAG_MAC = 28, RTW89_FW_ELEMENT_ID_NUM, }; @@ -4203,6 +4204,11 @@ struct rtw89_fw_element_hdr { __le32 val; } __packed infos[]; } __packed afe; + struct { + __le32 rule_size; + u8 rsvd[4]; + u8 rules_and_msgs[]; + } __packed diag_mac; struct __rtw89_fw_txpwr_element txpwr; struct __rtw89_fw_regd_element regd; } __packed u; From 7bf433c6767ffe2ad5b7ac8680c6e93e7d0be3e4 Mon Sep 17 00:00:00 2001 From: Ping-Ke Shih Date: Tue, 11 Nov 2025 10:24:47 +0800 Subject: [PATCH 0760/1381] wifi: rtw89: debug: add parser to diagnose along DIAG_MAC fw element The rules to diagnose MAC have a common header, and a cmd field is used to know the exact command and its format. The rules with the same tuple of fields {sheet, seq} can be seen as a set of compound rules, which treat it as positive rule if just one of the rules is positive. Take EQUALV rules as example, if value of {addr, mask} is equal to predefined value as field val, a rule is positive. Fields addr_name_offset and msg_offset are offsets related to textual messages for human readable. Format of common rule header (8 bytes) +-------+-----+--------+-----------+-----+---------+ | sheet | cmd | seq[2] | io / band | len | rsvd[2] | +-------+-----+--------+-----------+-----+---------+ Format of rule command is EQUALV (equal value) (24 bytes): +------+------------------+------+-----+------------+---------+ | addr | addr_name_offset | mask | val | msg_offset | rsvd[4] | +------+------------------+------+-----+------------+---------+ Format of message: +-----+----------+ | len | string[] | +-----+----------+ An example of output: Plain(Ignore)/Rules/Positive: 115(4)/86/81 Where, Plain is total rules written in firmware element. Ignore is the ignored rules, such as USB IO, but current is PCIE. Rules is number of set of compound rules. Positive is number of positive Rules. Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251111022452.28093-4-pkshih@realtek.com --- drivers/net/wireless/realtek/rtw89/debug.c | 299 +++++++++++++++++++++ 1 file changed, 299 insertions(+) diff --git a/drivers/net/wireless/realtek/rtw89/debug.c b/drivers/net/wireless/realtek/rtw89/debug.c index 3dc7981c510f..1264c2f82600 100644 --- a/drivers/net/wireless/realtek/rtw89/debug.c +++ b/drivers/net/wireless/realtek/rtw89/debug.c @@ -87,6 +87,7 @@ struct rtw89_debugfs { struct rtw89_debugfs_priv disable_dm; struct rtw89_debugfs_priv mlo_mode; struct rtw89_debugfs_priv beacon_info; + struct rtw89_debugfs_priv diag_mac; }; struct rtw89_debugfs_iter_data { @@ -4361,6 +4362,302 @@ rtw89_debug_priv_mlo_mode_set(struct rtw89_dev *rtwdev, return count; } +enum __diag_mac_cmd { + __CMD_EQUALV, + __CMD_EQUALO, + __CMD_NEQUALV, + __CMD_NEQUALO, + __CMD_SETEQUALV, + __CMD_SETEQUALO, + __CMD_CMPWCR, + __CMD_CMPWWD, + __CMD_NEQ_CMPWCR, + __CMD_NEQ_CMPWWD, + __CMD_INCREMENT, + __CMD_MESSAGE, +}; + +enum __diag_mac_io { + __IO_NORMAL, + __IO_NORMAL_PCIE, + __IO_NORMAL_USB, + __IO_NORMAL_SDIO, + __IO_PCIE_CFG, + __IO_SDIO_CCCR, +}; + +struct __diag_mac_rule_header { + u8 sheet; + u8 cmd; + u8 seq_major; + u8 seq_minor; + u8 io_band; + #define __DIAG_MAC_IO GENMASK(3, 0) + #define __DIAG_MAC_N_BAND BIT(4) + #define __DIAG_MAC_HAS_BAND BIT(5) + u8 len; /* include header. Unit: 4 bytes */ + u8 rsvd[2]; +} __packed; + +struct __diag_mac_rule_equal { + struct __diag_mac_rule_header header; + __le32 addr; + __le32 addr_name_offset; + __le32 mask; + __le32 val; + __le32 msg_offset; + u8 rsvd[4]; +} __packed; + +struct __diag_mac_rule_increment { + struct __diag_mac_rule_header header; + __le32 addr; + __le32 addr_name_offset; + __le32 mask; + __le16 sel; + __le16 delay; + __le32 msg_offset; + u8 rsvd[4]; +} __packed; + +struct __diag_mac_msg_buf { + __le16 len; + char string[]; +} __packed; + +static ssize_t rtw89_mac_diag_do_equalv(struct rtw89_dev *rtwdev, + char *buf, size_t bufsz, + const struct __diag_mac_rule_equal *r, + const void *msg_start, + u64 *positive_bmp) +{ + const struct __diag_mac_msg_buf *name = msg_start + + le32_to_cpu(r->addr_name_offset); + const struct __diag_mac_msg_buf *msg = msg_start + + le32_to_cpu(r->msg_offset); + bool want_eq = r->header.cmd == __CMD_EQUALV; + char *p = buf, *end = buf + bufsz; + bool equal = false; + u32 val; + + *positive_bmp <<= 1; + + if (u8_get_bits(r->header.io_band, __DIAG_MAC_IO) == __IO_PCIE_CFG) + val = rtw89_read32_pci_cfg(rtwdev, le32_to_cpu(r->addr)); + else + val = rtw89_read32(rtwdev, le32_to_cpu(r->addr)); + + if ((val & le32_to_cpu(r->mask)) == le32_to_cpu(r->val)) + equal = true; + + if (want_eq == equal) { + *positive_bmp |= BIT(0); + return p - buf; + } + + p += scnprintf(p, end - p, "sheet: %d, cmd: %d, Reg: %.*s => %x, %.*s\n", + r->header.sheet, r->header.cmd, le16_to_cpu(name->len), + name->string, val, le16_to_cpu(msg->len), msg->string); + + return p - buf; +} + +static ssize_t rtw89_mac_diag_do_increment(struct rtw89_dev *rtwdev, + char *buf, size_t bufsz, + const struct __diag_mac_rule_increment *r, + const void *msg_start, + u64 *positive_bmp) +{ + const struct __diag_mac_msg_buf *name = msg_start + + le32_to_cpu(r->addr_name_offset); + const struct __diag_mac_msg_buf *msg = msg_start + + le32_to_cpu(r->msg_offset); + char *p = buf, *end = buf + bufsz; + u32 addr = le32_to_cpu(r->addr); + u32 mask = le32_to_cpu(r->mask); + u16 sel = le16_to_cpu(r->sel); + u32 val1, val2; + + *positive_bmp <<= 1; + + rtw89_write32(rtwdev, addr, sel); + + if (u8_get_bits(r->header.io_band, __DIAG_MAC_IO) == __IO_PCIE_CFG) + val1 = rtw89_read32_pci_cfg(rtwdev, addr); + else + val1 = rtw89_read32(rtwdev, addr); + + mdelay(le16_to_cpu(r->delay)); + + if (u8_get_bits(r->header.io_band, __DIAG_MAC_IO) == __IO_PCIE_CFG) + val2 = rtw89_read32_pci_cfg(rtwdev, addr); + else + val2 = rtw89_read32(rtwdev, addr); + + if ((val2 & mask) > (val1 & mask)) { + *positive_bmp |= BIT(0); + return p - buf; + } + + p += scnprintf(p, end - p, "sheet: %d, cmd: %d, Reg: %.*s [%d]=> %x, %.*s\n", + r->header.sheet, r->header.cmd, le16_to_cpu(name->len), + name->string, le16_to_cpu(r->sel), val1, + le16_to_cpu(msg->len), msg->string); + + return p - buf; +} + +static bool rtw89_mac_diag_match_hci(struct rtw89_dev *rtwdev, + const struct __diag_mac_rule_header *rh) +{ + switch (u8_get_bits(rh->io_band, __DIAG_MAC_IO)) { + case __IO_NORMAL: + default: + return true; + case __IO_NORMAL_PCIE: + case __IO_PCIE_CFG: + if (rtwdev->hci.type == RTW89_HCI_TYPE_PCIE) + return true; + break; + case __IO_NORMAL_USB: + if (rtwdev->hci.type == RTW89_HCI_TYPE_USB) + return true; + break; + case __IO_NORMAL_SDIO: + case __IO_SDIO_CCCR: + if (rtwdev->hci.type == RTW89_HCI_TYPE_SDIO) + return true; + break; + } + + return false; +} + +static bool rtw89_mac_diag_match_band(struct rtw89_dev *rtwdev, + const struct __diag_mac_rule_header *rh) +{ + u8 active_bands; + bool has_band; + u8 band; + + has_band = u8_get_bits(rh->io_band, __DIAG_MAC_HAS_BAND); + if (!has_band) + return true; + + band = u8_get_bits(rh->io_band, __DIAG_MAC_N_BAND); + active_bands = rtw89_get_active_phy_bitmap(rtwdev); + + if (active_bands & BIT(band)) + return true; + + return false; +} + +static ssize_t rtw89_mac_diag_iter_all(struct rtw89_dev *rtwdev, + char *buf, size_t bufsz) +{ + const struct rtw89_fw_element_hdr *elm = rtwdev->fw.elm_info.diag_mac; + u32 n_plains = 0, n_rules = 0, n_positive = 0, n_ignore = 0; + char *p = buf, *end = buf + bufsz, *p_rewind; + const void *rule, *rule_end; + u32 elm_size, rule_size; + const void *msg_start; + u64 positive_bmp = 0; + u8 prev_sheet = 0; + u8 prev_seq = 0; + int limit; + + if (!elm) { + p += scnprintf(p, end - p, "No diag_mac entry\n"); + goto out; + } + + rule_size = le32_to_cpu(elm->u.diag_mac.rule_size); + elm_size = le32_to_cpu(elm->size); + + if (ALIGN(rule_size, 16) > elm_size) { + p += scnprintf(p, end - p, "rule size (%u) exceed elm_size (%u)\n", + ALIGN(rule_size, 16), elm_size); + goto out; + } + + rule = &elm->u.diag_mac.rules_and_msgs[0]; + rule_end = &elm->u.diag_mac.rules_and_msgs[rule_size]; + msg_start = &elm->u.diag_mac.rules_and_msgs[ALIGN(rule_size, 16)]; + + for (limit = 0; limit < 5000 && rule < rule_end; limit++) { + const struct __diag_mac_rule_header *rh = rule; + u8 sheet = rh->sheet; + u8 seq = rh->seq_major; + + if (!rtw89_mac_diag_match_hci(rtwdev, rh) || + !rtw89_mac_diag_match_band(rtwdev, rh)) { + n_ignore++; + goto next; + } + + if (!seq || prev_sheet != sheet || prev_seq != seq) { + if (positive_bmp) { + n_positive++; + /* + * discard output for negative results if one in + * a sequence set is positive. + */ + if (p_rewind) + p = p_rewind; + } + p_rewind = seq ? p : NULL; + positive_bmp = 0; + n_rules++; + } + + switch (rh->cmd) { + case __CMD_EQUALV: + case __CMD_NEQUALV: + p += rtw89_mac_diag_do_equalv(rtwdev, p, end - p, rule, + msg_start, &positive_bmp); + break; + case __CMD_INCREMENT: + p += rtw89_mac_diag_do_increment(rtwdev, p, end - p, rule, + msg_start, &positive_bmp); + break; + default: + p += scnprintf(p, end - p, "unknown rule cmd %u\n", rh->cmd); + break; + } + +next: + n_plains++; + rule += rh->len * 4; + prev_seq = seq; + prev_sheet = sheet; + } + + if (positive_bmp) { + n_positive++; + if (p_rewind) + p = p_rewind; + } + + p += scnprintf(p, end - p, "\nPlain(Ignore)/Rules/Positive: %u(%u)/%u/%u\n", + n_plains, n_ignore, n_rules, n_positive); + +out: + return p - buf; +} + +static ssize_t +rtw89_debug_priv_diag_mac_get(struct rtw89_dev *rtwdev, + struct rtw89_debugfs_priv *debugfs_priv, + char *buf, size_t bufsz) +{ + lockdep_assert_wiphy(rtwdev->hw->wiphy); + + rtw89_leave_lps(rtwdev); + + return rtw89_mac_diag_iter_all(rtwdev, buf, bufsz); +} + static ssize_t rtw89_debug_priv_beacon_info_get(struct rtw89_dev *rtwdev, struct rtw89_debugfs_priv *debugfs_priv, @@ -4478,6 +4775,7 @@ static const struct rtw89_debugfs rtw89_debugfs_templ = { .disable_dm = rtw89_debug_priv_set_and_get(disable_dm, RWLOCK), .mlo_mode = rtw89_debug_priv_set_and_get(mlo_mode, RWLOCK), .beacon_info = rtw89_debug_priv_get(beacon_info), + .diag_mac = rtw89_debug_priv_get(diag_mac, RSIZE_16K, RLOCK), }; #define rtw89_debugfs_add(name, mode, fopname, parent) \ @@ -4524,6 +4822,7 @@ void rtw89_debugfs_add_sec1(struct rtw89_dev *rtwdev, struct dentry *debugfs_top rtw89_debugfs_add_rw(disable_dm); rtw89_debugfs_add_rw(mlo_mode); rtw89_debugfs_add_r(beacon_info); + rtw89_debugfs_add_r(diag_mac); } void rtw89_debugfs_init(struct rtw89_dev *rtwdev) From b37fb77d4530f711a08b6dd860a8a4839897aba2 Mon Sep 17 00:00:00 2001 From: Ping-Ke Shih Date: Tue, 11 Nov 2025 10:24:48 +0800 Subject: [PATCH 0761/1381] wifi: rtw89: 8852c: add compensation of thermal value from efuse calibration The 3th bit of thermal value programmed in efuse means 8 (2 ^ 3) grams. Check the bit and signed bit, and add compensation to final thermal. Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251111022452.28093-5-pkshih@realtek.com --- drivers/net/wireless/realtek/rtw89/rtw8852c.c | 27 +++++++++++++++---- 1 file changed, 22 insertions(+), 5 deletions(-) diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852c.c b/drivers/net/wireless/realtek/rtw89/rtw8852c.c index 8550ae3bcb64..6be56c2fa0c7 100644 --- a/drivers/net/wireless/realtek/rtw89/rtw8852c.c +++ b/drivers/net/wireless/realtek/rtw89/rtw8852c.c @@ -670,12 +670,16 @@ static void rtw8852c_phycap_parsing_thermal_trim(struct rtw89_dev *rtwdev, } } +#define __THM_MASK_SIGN BIT(0) +#define __THM_MASK_3BITS GENMASK(3, 1) +#define __THM_MASK_VAL8 BIT(4) + static void rtw8852c_thermal_trim(struct rtw89_dev *rtwdev) { -#define __thm_setting(raw) \ -({ \ - u8 __v = (raw); \ - ((__v & 0x1) << 3) | ((__v & 0x1f) >> 1); \ +#define __thm_setting(raw) \ +({ \ + u8 __v = (raw); \ + ((__v & __THM_MASK_SIGN) << 3) | ((__v & __THM_MASK_3BITS) >> 1); \ }) struct rtw89_power_trim_info *info = &rtwdev->pwr_trim; u8 i, val; @@ -2516,13 +2520,26 @@ static void rtw8852c_bb_cfg_txrx_path(struct rtw89_dev *rtwdev) static u8 rtw8852c_get_thermal(struct rtw89_dev *rtwdev, enum rtw89_rf_path rf_path) { + struct rtw89_power_trim_info *info = &rtwdev->pwr_trim; + s8 comp = 0; + u8 val; + rtw89_write_rf(rtwdev, rf_path, RR_TM, RR_TM_TRI, 0x1); rtw89_write_rf(rtwdev, rf_path, RR_TM, RR_TM_TRI, 0x0); rtw89_write_rf(rtwdev, rf_path, RR_TM, RR_TM_TRI, 0x1); fsleep(200); - return rtw89_read_rf(rtwdev, rf_path, RR_TM, RR_TM_VAL); + val = rtw89_read_rf(rtwdev, rf_path, RR_TM, RR_TM_VAL); + + if (info->pg_thermal_trim) { + u8 trim = info->thermal_trim[rf_path]; + + if (trim & __THM_MASK_VAL8) + comp = 8 * (trim & __THM_MASK_SIGN ? -1 : 1); + } + + return val + comp; } static void rtw8852c_btc_set_rfe(struct rtw89_dev *rtwdev) From 1dd7e743dd37361c8c1ac89355a8da738e2b81fb Mon Sep 17 00:00:00 2001 From: Kuan-Chung Chen Date: Tue, 11 Nov 2025 10:24:49 +0800 Subject: [PATCH 0762/1381] wifi: rtw89: phy: fix out-of-bounds access in rtw89_phy_read_txpwr_limit() Coverity reported a potential out-of-bounds access when 'bw' exceeds the valid range for the specified band. Add a helper `rtw89_bw_is_valid()` to check bandwidth validity for each band before accessing limit tables. Addresses-Coverity-ID: 1598844 ("Out-of-bounds access") Addresses-Coverity-ID: 1598896 ("Out-of-bounds access") Signed-off-by: Kuan-Chung Chen Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251111022452.28093-6-pkshih@realtek.com --- drivers/net/wireless/realtek/rtw89/phy.c | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+) diff --git a/drivers/net/wireless/realtek/rtw89/phy.c b/drivers/net/wireless/realtek/rtw89/phy.c index 23892c1359a5..28e2b15240a7 100644 --- a/drivers/net/wireless/realtek/rtw89/phy.c +++ b/drivers/net/wireless/realtek/rtw89/phy.c @@ -2376,6 +2376,21 @@ static u8 rtw89_channel_to_idx(struct rtw89_dev *rtwdev, u8 band, u8 channel) } } +static bool rtw89_phy_validate_txpwr_limit_bw(struct rtw89_dev *rtwdev, + u8 band, u8 bw) +{ + switch (band) { + case RTW89_BAND_2G: + return bw < RTW89_2G_BW_NUM; + case RTW89_BAND_5G: + return bw < RTW89_5G_BW_NUM; + case RTW89_BAND_6G: + return bw < RTW89_6G_BW_NUM; + default: + return false; + } +} + s8 rtw89_phy_read_txpwr_limit(struct rtw89_dev *rtwdev, u8 band, u8 bw, u8 ntx, u8 rs, u8 bf, u8 ch) { @@ -2400,6 +2415,11 @@ s8 rtw89_phy_read_txpwr_limit(struct rtw89_dev *rtwdev, u8 band, }; s8 cstr; + if (!rtw89_phy_validate_txpwr_limit_bw(rtwdev, band, bw)) { + rtw89_warn(rtwdev, "invalid band %u bandwidth %u\n", band, bw); + return 0; + } + switch (band) { case RTW89_BAND_2G: if (has_ant_gain) From e96abe4fd7cd6f48434d8846aa7588bf7bdb3f0e Mon Sep 17 00:00:00 2001 From: Ping-Ke Shih Date: Tue, 11 Nov 2025 10:24:50 +0800 Subject: [PATCH 0763/1381] wifi: rtw89: consider data rate/bandwidth/GI for injected packets To send injected packets with configurable rate/bandwidth/GI, fill TXWD fields according to SKB's info->control.rates[0] annotated by mac80211. Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251111022452.28093-7-pkshih@realtek.com --- drivers/net/wireless/realtek/rtw89/core.c | 67 +++++++++++++++++++++++ drivers/net/wireless/realtek/rtw89/core.h | 2 + 2 files changed, 69 insertions(+) diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c index 1d30ff27bce1..a5c451eed5a7 100644 --- a/drivers/net/wireless/realtek/rtw89/core.c +++ b/drivers/net/wireless/realtek/rtw89/core.c @@ -321,6 +321,26 @@ static const struct ieee80211_supported_band rtw89_sband_6ghz = { .n_bitrates = ARRAY_SIZE(rtw89_bitrates) - 4, }; +static const struct rtw89_hw_rate_def { + enum rtw89_hw_rate ht; + enum rtw89_hw_rate vht[RTW89_NSS_NUM]; +} rtw89_hw_rate[RTW89_CHIP_GEN_NUM] = { + [RTW89_CHIP_AX] = { + .ht = RTW89_HW_RATE_MCS0, + .vht = {RTW89_HW_RATE_VHT_NSS1_MCS0, + RTW89_HW_RATE_VHT_NSS2_MCS0, + RTW89_HW_RATE_VHT_NSS3_MCS0, + RTW89_HW_RATE_VHT_NSS4_MCS0}, + }, + [RTW89_CHIP_BE] = { + .ht = RTW89_HW_RATE_V1_MCS0, + .vht = {RTW89_HW_RATE_V1_VHT_NSS1_MCS0, + RTW89_HW_RATE_V1_VHT_NSS2_MCS0, + RTW89_HW_RATE_V1_VHT_NSS3_MCS0, + RTW89_HW_RATE_V1_VHT_NSS4_MCS0}, + }, +}; + static void __rtw89_traffic_stats_accu(struct rtw89_traffic_stats *stats, struct sk_buff *skb, bool tx) { @@ -1097,6 +1117,44 @@ rtw89_core_tx_wake(struct rtw89_dev *rtwdev, rtw89_mac_notify_wake(rtwdev); } +static void rtw89_core_tx_update_injection(struct rtw89_dev *rtwdev, + struct rtw89_core_tx_request *tx_req, + struct ieee80211_tx_info *info) +{ + const struct rtw89_hw_rate_def *hw_rate = &rtw89_hw_rate[rtwdev->chip->chip_gen]; + enum mac80211_rate_control_flags flags = info->control.rates[0].flags; + struct rtw89_tx_desc_info *desc_info = &tx_req->desc_info; + const struct rtw89_chan *chan; + u8 idx = info->control.rates[0].idx; + u8 nss, mcs; + + desc_info->use_rate = true; + desc_info->dis_data_fb = true; + + if (flags & IEEE80211_TX_RC_160_MHZ_WIDTH) + desc_info->data_bw = 3; + else if (flags & IEEE80211_TX_RC_80_MHZ_WIDTH) + desc_info->data_bw = 2; + else if (flags & IEEE80211_TX_RC_40_MHZ_WIDTH) + desc_info->data_bw = 1; + + if (flags & IEEE80211_TX_RC_SHORT_GI) + desc_info->gi_ltf = 1; + + if (flags & IEEE80211_TX_RC_VHT_MCS) { + nss = umin(idx >> 4, ARRAY_SIZE(hw_rate->vht) - 1); + mcs = idx & 0xf; + desc_info->data_rate = hw_rate->vht[nss] + mcs; + } else if (flags & IEEE80211_TX_RC_MCS) { + desc_info->data_rate = hw_rate->ht + idx; + } else { + chan = rtw89_chan_get(rtwdev, tx_req->rtwvif_link->chanctx_idx); + + desc_info->data_rate = idx + (chan->band_type == RTW89_BAND_2G ? + RTW89_HW_RATE_CCK1 : RTW89_HW_RATE_OFDM6); + } +} + static void rtw89_core_tx_update_desc_info(struct rtw89_dev *rtwdev, struct rtw89_core_tx_request *tx_req) @@ -1157,6 +1215,9 @@ rtw89_core_tx_update_desc_info(struct rtw89_dev *rtwdev, default: break; } + + if (unlikely(info->flags & IEEE80211_TX_CTL_INJECTED)) + rtw89_core_tx_update_injection(rtwdev, tx_req, info); } static void rtw89_tx_wait_work(struct wiphy *wiphy, struct wiphy_work *work) @@ -1385,6 +1446,8 @@ static __le32 rtw89_build_txwd_body5(struct rtw89_tx_desc_info *desc_info) static __le32 rtw89_build_txwd_body7_v1(struct rtw89_tx_desc_info *desc_info) { u32 dword = FIELD_PREP(RTW89_TXWD_BODY7_USE_RATE_V1, desc_info->use_rate) | + FIELD_PREP(RTW89_TXWD_BODY7_DATA_BW, desc_info->data_bw) | + FIELD_PREP(RTW89_TXWD_BODY7_GI_LTF, desc_info->gi_ltf) | FIELD_PREP(RTW89_TXWD_BODY7_DATA_RATE, desc_info->data_rate); return cpu_to_le32(dword); @@ -1393,6 +1456,8 @@ static __le32 rtw89_build_txwd_body7_v1(struct rtw89_tx_desc_info *desc_info) static __le32 rtw89_build_txwd_info0(struct rtw89_tx_desc_info *desc_info) { u32 dword = FIELD_PREP(RTW89_TXWD_INFO0_USE_RATE, desc_info->use_rate) | + FIELD_PREP(RTW89_TXWD_INFO0_DATA_BW, desc_info->data_bw) | + FIELD_PREP(RTW89_TXWD_INFO0_GI_LTF, desc_info->gi_ltf) | FIELD_PREP(RTW89_TXWD_INFO0_DATA_RATE, desc_info->data_rate) | FIELD_PREP(RTW89_TXWD_INFO0_DATA_STBC, desc_info->stbc) | FIELD_PREP(RTW89_TXWD_INFO0_DATA_LDPC, desc_info->ldpc) | @@ -1585,6 +1650,8 @@ static __le32 rtw89_build_txwd_body6_v2(struct rtw89_tx_desc_info *desc_info) static __le32 rtw89_build_txwd_body7_v2(struct rtw89_tx_desc_info *desc_info) { u32 dword = FIELD_PREP(BE_TXD_BODY7_USERATE_SEL, desc_info->use_rate) | + FIELD_PREP(BE_TXD_BODY7_DATA_BW, desc_info->data_bw) | + FIELD_PREP(BE_TXD_BODY7_GI_LTF, desc_info->gi_ltf) | FIELD_PREP(BE_TXD_BODY7_DATA_ER, desc_info->er_cap) | FIELD_PREP(BE_TXD_BODY7_DATA_BW_ER, 0) | FIELD_PREP(BE_TXD_BODY7_DATARATE, desc_info->data_rate); diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h index 34c135269627..4da73153513a 100644 --- a/drivers/net/wireless/realtek/rtw89/core.h +++ b/drivers/net/wireless/realtek/rtw89/core.h @@ -1182,6 +1182,8 @@ struct rtw89_tx_desc_info { u8 sec_seq[6]; u16 data_rate; u16 data_retry_lowest_rate; + u8 data_bw; + u8 gi_ltf; bool fw_dl; u16 seq; bool a_ctrl_bsr; From d9204e99b69623997cba0d7472fa7df644df7cb5 Mon Sep 17 00:00:00 2001 From: Ping-Ke Shih Date: Tue, 11 Nov 2025 10:24:51 +0800 Subject: [PATCH 0764/1381] wifi: rtw89: do RF calibration once setting channel when running pure monitor mode To be able to capture and inject packets in monitor mode, do RF calibration once setting certain channel. Since calibration costs time, do not change behavior of normal usage, which do calibration only when starting as AP or going to connect AP. Since driver declares IEEE80211_HW_WANT_MONITOR_VIF, the pure monitor vif must be only one when adding interface with type NL80211_IFTYPE_MONITOR. Otherwise, monitor vif must be NULL. Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251111022452.28093-8-pkshih@realtek.com --- drivers/net/wireless/realtek/rtw89/core.c | 18 ++++++++++++++++++ drivers/net/wireless/realtek/rtw89/core.h | 1 + drivers/net/wireless/realtek/rtw89/mac80211.c | 4 ++++ 3 files changed, 23 insertions(+) diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c index a5c451eed5a7..a15a911484bb 100644 --- a/drivers/net/wireless/realtek/rtw89/core.c +++ b/drivers/net/wireless/realtek/rtw89/core.c @@ -470,6 +470,22 @@ void rtw89_core_set_chip_txpwr(struct rtw89_dev *rtwdev) __rtw89_core_set_chip_txpwr(rtwdev, chan, RTW89_PHY_1); } +static void rtw89_chip_rfk_channel_for_pure_mon_vif(struct rtw89_dev *rtwdev, + enum rtw89_phy_idx phy_idx) +{ + struct rtw89_vif *rtwvif = rtwdev->pure_monitor_mode_vif; + struct rtw89_vif_link *rtwvif_link; + + if (!rtwvif) + return; + + rtwvif_link = rtw89_vif_get_link_inst(rtwvif, phy_idx); + if (!rtwvif_link) + return; + + rtw89_chip_rfk_channel(rtwdev, rtwvif_link); +} + static void __rtw89_set_channel(struct rtw89_dev *rtwdev, const struct rtw89_chan *chan, enum rtw89_mac_idx mac_idx, @@ -498,6 +514,8 @@ static void __rtw89_set_channel(struct rtw89_dev *rtwdev, } rtw89_set_entity_state(rtwdev, phy_idx, true); + + rtw89_chip_rfk_channel_for_pure_mon_vif(rtwdev, phy_idx); } int rtw89_set_channel(struct rtw89_dev *rtwdev) diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h index 4da73153513a..f63857b389d7 100644 --- a/drivers/net/wireless/realtek/rtw89/core.h +++ b/drivers/net/wireless/realtek/rtw89/core.h @@ -6125,6 +6125,7 @@ struct rtw89_dev { int napi_budget_countdown; struct rtw89_debugfs *debugfs; + struct rtw89_vif *pure_monitor_mode_vif; /* HCI related data, keep last */ u8 priv[] __aligned(sizeof(void *)); diff --git a/drivers/net/wireless/realtek/rtw89/mac80211.c b/drivers/net/wireless/realtek/rtw89/mac80211.c index a19304ff8306..14c8864bd201 100644 --- a/drivers/net/wireless/realtek/rtw89/mac80211.c +++ b/drivers/net/wireless/realtek/rtw89/mac80211.c @@ -220,6 +220,8 @@ static int rtw89_ops_add_interface(struct ieee80211_hw *hw, if (ret) goto unset_link; + rtwdev->pure_monitor_mode_vif = vif->type == NL80211_IFTYPE_MONITOR ? + rtwvif : NULL; rtw89_recalc_lps(rtwdev); return 0; @@ -267,6 +269,8 @@ static void rtw89_ops_remove_interface(struct ieee80211_hw *hw, rtw89_core_release_bit_map(rtwdev->hw_port, port); rtw89_release_mac_id(rtwdev, macid); + rtwdev->pure_monitor_mode_vif = NULL; + rtw89_recalc_lps(rtwdev); rtw89_enter_ips_by_hwflags(rtwdev); } From 7465c0a91278b1829f7a3e5e2459e3329edb4ae4 Mon Sep 17 00:00:00 2001 From: Ping-Ke Shih Date: Tue, 11 Nov 2025 10:24:52 +0800 Subject: [PATCH 0765/1381] wifi: rtw89: configure RX antenna if chips can support If chip->ops->cfg_txrx_path is implemented, a chip can support to configure RX antenna, so accept setting via iw tool. Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251111022452.28093-9-pkshih@realtek.com --- drivers/net/wireless/realtek/rtw89/mac80211.c | 5 +++++ .../net/wireless/realtek/rtw89/rtw8852b_common.c | 6 +++++- drivers/net/wireless/realtek/rtw89/rtw8852c.c | 14 ++++++++++++-- drivers/net/wireless/realtek/rtw89/rtw8922a.c | 12 +++++++++++- 4 files changed, 33 insertions(+), 4 deletions(-) diff --git a/drivers/net/wireless/realtek/rtw89/mac80211.c b/drivers/net/wireless/realtek/rtw89/mac80211.c index 14c8864bd201..6f7571c7c274 100644 --- a/drivers/net/wireless/realtek/rtw89/mac80211.c +++ b/drivers/net/wireless/realtek/rtw89/mac80211.c @@ -1147,12 +1147,17 @@ int rtw89_ops_set_antenna(struct ieee80211_hw *hw, int radio_idx, u32 tx_ant, u3 { struct rtw89_dev *rtwdev = hw->priv; struct rtw89_hal *hal = &rtwdev->hal; + const struct rtw89_chip_info *chip; lockdep_assert_wiphy(hw->wiphy); + chip = rtwdev->chip; + if (hal->ant_diversity) { if (tx_ant != rx_ant || hweight32(tx_ant) != 1) return -EINVAL; + } else if (chip->ops->cfg_txrx_path) { + /* With cfg_txrx_path ops, chips can configure rx_ant */ } else if (rx_ant != hw->wiphy->available_antennas_rx && rx_ant != hal->antenna_rx) { return -EINVAL; } diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852b_common.c b/drivers/net/wireless/realtek/rtw89/rtw8852b_common.c index 3fb2972ae6f6..4e72f4961837 100644 --- a/drivers/net/wireless/realtek/rtw89/rtw8852b_common.c +++ b/drivers/net/wireless/realtek/rtw89/rtw8852b_common.c @@ -1747,11 +1747,15 @@ static void __rtw8852bx_bb_cfg_txrx_path(struct rtw89_dev *rtwdev) struct rtw89_hal *hal = &rtwdev->hal; const struct rtw89_chan *chan = rtw89_chan_get(rtwdev, RTW89_CHANCTX_0); enum rtw89_rf_path_bit rx_path = hal->antenna_rx ? hal->antenna_rx : RF_AB; + u8 rx_nss = rtwdev->hal.rx_nss; + + if (rx_path != RF_AB) + rx_nss = 1; rtw8852bx_bb_ctrl_rx_path(rtwdev, rx_path, chan); rtw8852bx_bb_ctrl_rf_mode_rx_path(rtwdev, rx_path); - if (rtwdev->hal.rx_nss == 1) { + if (rx_nss == 1) { rtw89_phy_write32_mask(rtwdev, R_RXHT_MCS_LIMIT, B_RXHT_MCS_LIMIT, 0); rtw89_phy_write32_mask(rtwdev, R_RXVHT_MCS_LIMIT, B_RXVHT_MCS_LIMIT, 0); rtw89_phy_write32_mask(rtwdev, R_RXHE, B_RXHE_MAX_NSS, 0); diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852c.c b/drivers/net/wireless/realtek/rtw89/rtw8852c.c index 6be56c2fa0c7..ee81a6792eee 100644 --- a/drivers/net/wireless/realtek/rtw89/rtw8852c.c +++ b/drivers/net/wireless/realtek/rtw89/rtw8852c.c @@ -2502,10 +2502,20 @@ static void rtw8852c_ctrl_nbtg_bt_tx(struct rtw89_dev *rtwdev, bool en, static void rtw8852c_bb_cfg_txrx_path(struct rtw89_dev *rtwdev) { struct rtw89_hal *hal = &rtwdev->hal; + u8 nrx_path = RF_PATH_AB; + u8 rx_nss = hal->rx_nss; - rtw8852c_bb_cfg_rx_path(rtwdev, RF_PATH_AB); + if (hal->antenna_rx == RF_A) + nrx_path = RF_PATH_A; + else if (hal->antenna_rx == RF_B) + nrx_path = RF_PATH_B; - if (hal->rx_nss == 1) { + if (nrx_path != RF_PATH_AB) + rx_nss = 1; + + rtw8852c_bb_cfg_rx_path(rtwdev, nrx_path); + + if (rx_nss == 1) { rtw89_phy_write32_mask(rtwdev, R_RXHT_MCS_LIMIT, B_RXHT_MCS_LIMIT, 0); rtw89_phy_write32_mask(rtwdev, R_RXVHT_MCS_LIMIT, B_RXVHT_MCS_LIMIT, 0); rtw89_phy_write32_mask(rtwdev, R_RXHE, B_RXHE_MAX_NSS, 0); diff --git a/drivers/net/wireless/realtek/rtw89/rtw8922a.c b/drivers/net/wireless/realtek/rtw89/rtw8922a.c index 61f3a0a3f440..11e8b50a3291 100644 --- a/drivers/net/wireless/realtek/rtw89/rtw8922a.c +++ b/drivers/net/wireless/realtek/rtw89/rtw8922a.c @@ -2347,19 +2347,29 @@ static void rtw8922a_bb_cfg_txrx_path(struct rtw89_dev *rtwdev) enum rtw89_band band = chan->band_type; struct rtw89_hal *hal = &rtwdev->hal; u8 ntx_path = RF_PATH_AB; + u8 nrx_path = RF_PATH_AB; u32 tx_en0, tx_en1; + u8 rx_nss = 2; if (hal->antenna_tx == RF_A) ntx_path = RF_PATH_A; else if (hal->antenna_tx == RF_B) ntx_path = RF_PATH_B; + if (hal->antenna_rx == RF_A) + nrx_path = RF_PATH_A; + else if (hal->antenna_rx == RF_B) + nrx_path = RF_PATH_B; + + if (nrx_path != RF_PATH_AB) + rx_nss = 1; + rtw8922a_hal_reset(rtwdev, RTW89_PHY_0, RTW89_MAC_0, band, &tx_en0, true); if (rtwdev->dbcc_en) rtw8922a_hal_reset(rtwdev, RTW89_PHY_1, RTW89_MAC_1, band, &tx_en1, true); - rtw8922a_ctrl_trx_path(rtwdev, ntx_path, 2, RF_PATH_AB, 2); + rtw8922a_ctrl_trx_path(rtwdev, ntx_path, 2, nrx_path, rx_nss); rtw8922a_hal_reset(rtwdev, RTW89_PHY_0, RTW89_MAC_0, band, &tx_en0, false); if (rtwdev->dbcc_en) From ef42cf705ca32c07a58133817567ac06ef74dc19 Mon Sep 17 00:00:00 2001 From: Dan Hamik Date: Wed, 12 Nov 2025 07:04:51 +0000 Subject: [PATCH 0766/1381] wifi: rtw89: rtw8852bu: Added dev id for ASUS AX57 NANO USB Wifi dongle Add the USB device ID 0x0b05:0x1cb6 to the rtw8852bu driver to support the ASUS AX57 Nano WiFi 6 USB adapter.This device uses the same Realtek RTL8852BU chipset as other supported models. Tested on: Linux Mint 22 with kernel 6.8.0-87-generic. The adapter initializes successfully and connects to networks. Signed-off-by: Dan Hamik Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/010f019a76e1a84b-0a6f5e9f-2a43-4a9d-9c30-de4ae6363011-000000@us-east-2.amazonses.com --- drivers/net/wireless/realtek/rtw89/rtw8852bu.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852bu.c b/drivers/net/wireless/realtek/rtw89/rtw8852bu.c index c8a644403341..980d17ef68d0 100644 --- a/drivers/net/wireless/realtek/rtw89/rtw8852bu.c +++ b/drivers/net/wireless/realtek/rtw89/rtw8852bu.c @@ -50,6 +50,8 @@ static const struct usb_device_id rtw_8852bu_id_table[] = { .driver_info = (kernel_ulong_t)&rtw89_8852bu_info }, { USB_DEVICE_AND_INTERFACE_INFO(0x0b05, 0x1a62, 0xff, 0xff, 0xff), .driver_info = (kernel_ulong_t)&rtw89_8852bu_info }, + { USB_DEVICE_AND_INTERFACE_INFO(0x0b05, 0x1cb6, 0xff, 0xff, 0xff), + .driver_info = (kernel_ulong_t)&rtw89_8852bu_info }, { USB_DEVICE_AND_INTERFACE_INFO(0x0db0, 0x6931, 0xff, 0xff, 0xff), .driver_info = (kernel_ulong_t)&rtw89_8852bu_info }, { USB_DEVICE_AND_INTERFACE_INFO(0x2001, 0x3327, 0xff, 0xff, 0xff), From f766f8cdde018967e9436f7f34d09907881fa7fa Mon Sep 17 00:00:00 2001 From: Dimitri Daskalakis Date: Tue, 11 Nov 2025 14:53:19 -0800 Subject: [PATCH 0767/1381] selftests: drv-net: Limit the max number of queues in procfs_downup_hammer For NICs with a large (1024+) number of queues, this test can cause excessive memory fragmentation. This results in OOM errors, and in the worst case driver/kernel crashes. We don't need to test with the max number of queues, just enough to create a high likelihood of races between reconfiguration and stats getting read. Signed-off-by: Dimitri Daskalakis Link: https://patch.msgid.link/20251111225319.3019542-1-dimitri.daskalakis1@gmail.com Signed-off-by: Jakub Kicinski --- tools/testing/selftests/drivers/net/stats.py | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/tools/testing/selftests/drivers/net/stats.py b/tools/testing/selftests/drivers/net/stats.py index 04d0a2a13e73..b08e4d48b15c 100755 --- a/tools/testing/selftests/drivers/net/stats.py +++ b/tools/testing/selftests/drivers/net/stats.py @@ -263,14 +263,15 @@ def procfs_downup_hammer(cfg) -> None: Reading stats via procfs only holds the RCU lock, drivers often try to sleep when reading the stats, or don't protect against races. """ - # Max out the queues, we'll flip between max and 1 + # Set a large number of queues, + # we'll flip between min(max_queues, 64) and 1 channels = ethnl.channels_get({'header': {'dev-index': cfg.ifindex}}) if channels['combined-count'] == 0: rx_type = 'rx' else: rx_type = 'combined' cur_queue_cnt = channels[f'{rx_type}-count'] - max_queue_cnt = channels[f'{rx_type}-max'] + max_queue_cnt = min(channels[f'{rx_type}-max'], 64) cmd(f"ethtool -L {cfg.ifname} {rx_type} {max_queue_cnt}") defer(cmd, f"ethtool -L {cfg.ifname} {rx_type} {cur_queue_cnt}") From 9c577f09989f921e7a7cc724949fa81f9444dffb Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 11 Nov 2025 07:52:14 -0800 Subject: [PATCH 0768/1381] tools: ynltool: correct install in Makefile Use the variable in case user has a custom install binary. Acked-by: Stanislav Fomichev Link: https://patch.msgid.link/20251111155214.2760711-1-kuba@kernel.org Signed-off-by: Jakub Kicinski --- tools/net/ynl/ynltool/Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/net/ynl/ynltool/Makefile b/tools/net/ynl/ynltool/Makefile index 86c30b7420cf..5edaa0f93237 100644 --- a/tools/net/ynl/ynltool/Makefile +++ b/tools/net/ynl/ynltool/Makefile @@ -49,7 +49,7 @@ distclean: clean bindir ?= /usr/bin install: $(YNLTOOL) - install -m 0755 $(YNLTOOL) $(DESTDIR)$(bindir)/$(YNLTOOL) + $(INSTALL) -m 0755 $(YNLTOOL) $(DESTDIR)$(bindir)/$(YNLTOOL) .PHONY: all clean distclean .DEFAULT_GOAL=all From 5d88b24c1de80037af8f46db478639679668ece5 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Tue, 11 Nov 2025 08:11:47 +0000 Subject: [PATCH 0769/1381] net: stmmac: loongson1: use PHY_INTF_SEL_x Use PHY_INTF_SEL_x definitions for phy_intf_sel bitfield. Signed-off-by: Russell King (Oracle) Reviewed-by: Maxime Chevallier Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/dwmac-loongson1.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson1.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson1.c index 32b5d1492e2e..09e2af1d778a 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson1.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson1.c @@ -38,8 +38,8 @@ #define GMAC_SHUT BIT(6) #define PHY_INTF_SELI GENMASK(30, 28) -#define PHY_INTF_MII FIELD_PREP(PHY_INTF_SELI, 0) -#define PHY_INTF_RMII FIELD_PREP(PHY_INTF_SELI, 4) +#define PHY_INTF_MII FIELD_PREP(PHY_INTF_SELI, PHY_INTF_SEL_GMII_MII) +#define PHY_INTF_RMII FIELD_PREP(PHY_INTF_SELI, PHY_INTF_SEL_RMII) struct ls1x_dwmac { struct plat_stmmacenet_data *plat_dat; From 031f7a05d4b3ad4414e646414e16dfd4ab388772 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Tue, 11 Nov 2025 08:11:52 +0000 Subject: [PATCH 0770/1381] net: stmmac: loongson1: use PHY_INTF_SEL_x directly Use the PHY_INTF_SEL_xx values directly in ls1c_dwmac_syscon_init(), converting them to the PHY_INTF_SELI bitfield when calling regmap_update_bits(). Signed-off-by: Russell King (Oracle) Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/E1vIjTk-0000000Dqt6-1gN9@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/dwmac-loongson1.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson1.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson1.c index 09e2af1d778a..5f9f66fbc191 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson1.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson1.c @@ -38,8 +38,6 @@ #define GMAC_SHUT BIT(6) #define PHY_INTF_SELI GENMASK(30, 28) -#define PHY_INTF_MII FIELD_PREP(PHY_INTF_SELI, PHY_INTF_SEL_GMII_MII) -#define PHY_INTF_RMII FIELD_PREP(PHY_INTF_SELI, PHY_INTF_SEL_RMII) struct ls1x_dwmac { struct plat_stmmacenet_data *plat_dat; @@ -140,15 +138,14 @@ static int ls1c_dwmac_syscon_init(struct platform_device *pdev, void *priv) struct ls1x_dwmac *dwmac = priv; struct plat_stmmacenet_data *plat = dwmac->plat_dat; struct regmap *regmap = dwmac->regmap; + int phy_intf_sel; switch (plat->phy_interface) { case PHY_INTERFACE_MODE_MII: - regmap_update_bits(regmap, LS1X_SYSCON1, PHY_INTF_SELI, - PHY_INTF_MII); + phy_intf_sel = PHY_INTF_SEL_GMII_MII; break; case PHY_INTERFACE_MODE_RMII: - regmap_update_bits(regmap, LS1X_SYSCON1, PHY_INTF_SELI, - PHY_INTF_RMII); + phy_intf_sel = PHY_INTF_SEL_RMII; break; default: dev_err(&pdev->dev, "Unsupported PHY-mode %u\n", @@ -156,6 +153,8 @@ static int ls1c_dwmac_syscon_init(struct platform_device *pdev, void *priv) return -EOPNOTSUPP; } + regmap_update_bits(regmap, LS1X_SYSCON1, PHY_INTF_SELI, + FIELD_PREP(PHY_INTF_SELI, phy_intf_sel)); regmap_update_bits(regmap, LS1X_SYSCON0, GMAC0_SHUT, 0); return 0; From 83eb6c7e18641f328aed5da2e9bbf63e83806ca4 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Tue, 11 Nov 2025 08:11:57 +0000 Subject: [PATCH 0771/1381] net: stmmac: loongson1: use stmmac_get_phy_intf_sel() Use stmmac_get_phy_intf_sel() to decode the PHY interface mode to the phy_intf_sel value, validate the result and use that to set the control register to select the operating mode for the DWMAC core. Note that this will allow GMII as well as MII as the phy_intf_sel value is the same for both. Signed-off-by: Russell King (Oracle) Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/E1vIjTp-0000000DqtC-2DmI@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/dwmac-loongson1.c | 11 +++-------- 1 file changed, 3 insertions(+), 8 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson1.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson1.c index 5f9f66fbc191..894ee66f5c9b 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson1.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson1.c @@ -140,14 +140,9 @@ static int ls1c_dwmac_syscon_init(struct platform_device *pdev, void *priv) struct regmap *regmap = dwmac->regmap; int phy_intf_sel; - switch (plat->phy_interface) { - case PHY_INTERFACE_MODE_MII: - phy_intf_sel = PHY_INTF_SEL_GMII_MII; - break; - case PHY_INTERFACE_MODE_RMII: - phy_intf_sel = PHY_INTF_SEL_RMII; - break; - default: + phy_intf_sel = stmmac_get_phy_intf_sel(plat->phy_interface); + if (phy_intf_sel != PHY_INTF_SEL_GMII_MII && + phy_intf_sel != PHY_INTF_SEL_RMII) { dev_err(&pdev->dev, "Unsupported PHY-mode %u\n", plat->phy_interface); return -EOPNOTSUPP; From c3308d380e2c69c5dcb85784b750c4f83a965ef4 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Tue, 11 Nov 2025 08:12:02 +0000 Subject: [PATCH 0772/1381] net: stmmac: mediatek: use PHY_INTF_SEL_x Use PHY_INTF_SEL_x definitions for the fields that correspond to the phy_intf_sel inputs to the dwmac core. Signed-off-by: Russell King (Oracle) Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/E1vIjTu-0000000DqtI-2sUB@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../ethernet/stmicro/stmmac/dwmac-mediatek.c | 18 ++++++++---------- 1 file changed, 8 insertions(+), 10 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-mediatek.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-mediatek.c index f1b36f0a401d..dcdf28418fec 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-mediatek.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-mediatek.c @@ -17,9 +17,6 @@ /* Peri Configuration register for mt2712 */ #define PERI_ETH_PHY_INTF_SEL 0x418 -#define PHY_INTF_MII 0 -#define PHY_INTF_RGMII 1 -#define PHY_INTF_RMII 4 #define RMII_CLK_SRC_RXC BIT(4) #define RMII_CLK_SRC_INTERNAL BIT(5) @@ -118,16 +115,16 @@ static int mt2712_set_interface(struct mediatek_dwmac_plat_data *plat) /* select phy interface in top control domain */ switch (plat->phy_mode) { case PHY_INTERFACE_MODE_MII: - intf_val |= PHY_INTF_MII; + intf_val |= PHY_INTF_SEL_GMII_MII; break; case PHY_INTERFACE_MODE_RMII: - intf_val |= (PHY_INTF_RMII | rmii_rxc | rmii_clk_from_mac); + intf_val |= PHY_INTF_SEL_RMII | rmii_rxc | rmii_clk_from_mac; break; case PHY_INTERFACE_MODE_RGMII: case PHY_INTERFACE_MODE_RGMII_TXID: case PHY_INTERFACE_MODE_RGMII_RXID: case PHY_INTERFACE_MODE_RGMII_ID: - intf_val |= PHY_INTF_RGMII; + intf_val |= PHY_INTF_SEL_RGMII; break; default: dev_err(plat->dev, "phy interface not supported\n"); @@ -297,17 +294,18 @@ static int mt8195_set_interface(struct mediatek_dwmac_plat_data *plat) /* select phy interface in top control domain */ switch (plat->phy_mode) { case PHY_INTERFACE_MODE_MII: - intf_val |= FIELD_PREP(MT8195_ETH_INTF_SEL, PHY_INTF_MII); + intf_val |= FIELD_PREP(MT8195_ETH_INTF_SEL, + PHY_INTF_SEL_GMII_MII); break; case PHY_INTERFACE_MODE_RMII: - intf_val |= (rmii_rxc | rmii_clk_from_mac); - intf_val |= FIELD_PREP(MT8195_ETH_INTF_SEL, PHY_INTF_RMII); + intf_val |= rmii_rxc | rmii_clk_from_mac; + intf_val |= FIELD_PREP(MT8195_ETH_INTF_SEL, PHY_INTF_SEL_RMII); break; case PHY_INTERFACE_MODE_RGMII: case PHY_INTERFACE_MODE_RGMII_TXID: case PHY_INTERFACE_MODE_RGMII_RXID: case PHY_INTERFACE_MODE_RGMII_ID: - intf_val |= FIELD_PREP(MT8195_ETH_INTF_SEL, PHY_INTF_RGMII); + intf_val |= FIELD_PREP(MT8195_ETH_INTF_SEL, PHY_INTF_SEL_RGMII); break; default: dev_err(plat->dev, "phy interface not supported\n"); From d9c7964fd93490b31a09fe08034033cb2aa747f5 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Tue, 11 Nov 2025 08:12:07 +0000 Subject: [PATCH 0773/1381] net: stmmac: mediatek: use stmmac_get_phy_intf_sel() Use stmmac_get_phy_intf_sel() to decode the PHY interface mode to the phy_intf_sel value, validate the result, and pass that into the implementation specific ->dwmac_set_phy_interface() method. Use this to configure the PHY interface selection field. Signed-off-by: Russell King (Oracle) Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/E1vIjTz-0000000DqtP-3N9v@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../ethernet/stmicro/stmmac/dwmac-mediatek.c | 43 +++++++++++-------- 1 file changed, 25 insertions(+), 18 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-mediatek.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-mediatek.c index dcdf28418fec..0f32732efb75 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-mediatek.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-mediatek.c @@ -85,7 +85,8 @@ struct mediatek_dwmac_plat_data { }; struct mediatek_dwmac_variant { - int (*dwmac_set_phy_interface)(struct mediatek_dwmac_plat_data *plat); + int (*dwmac_set_phy_interface)(struct mediatek_dwmac_plat_data *plat, + u8 phy_intf_sel); int (*dwmac_set_delay)(struct mediatek_dwmac_plat_data *plat); /* clock ids to be requested */ @@ -106,25 +107,25 @@ static const char * const mt8195_dwmac_clk_l[] = { "axi", "apb", "mac_cg", "mac_main", "ptp_ref" }; -static int mt2712_set_interface(struct mediatek_dwmac_plat_data *plat) +static int mt2712_set_interface(struct mediatek_dwmac_plat_data *plat, + u8 phy_intf_sel) { int rmii_clk_from_mac = plat->rmii_clk_from_mac ? RMII_CLK_SRC_INTERNAL : 0; int rmii_rxc = plat->rmii_rxc ? RMII_CLK_SRC_RXC : 0; - u32 intf_val = 0; + u32 intf_val; + + intf_val = phy_intf_sel; /* select phy interface in top control domain */ switch (plat->phy_mode) { - case PHY_INTERFACE_MODE_MII: - intf_val |= PHY_INTF_SEL_GMII_MII; - break; case PHY_INTERFACE_MODE_RMII: - intf_val |= PHY_INTF_SEL_RMII | rmii_rxc | rmii_clk_from_mac; + intf_val |= rmii_rxc | rmii_clk_from_mac; break; + case PHY_INTERFACE_MODE_MII: case PHY_INTERFACE_MODE_RGMII: case PHY_INTERFACE_MODE_RGMII_TXID: case PHY_INTERFACE_MODE_RGMII_RXID: case PHY_INTERFACE_MODE_RGMII_ID: - intf_val |= PHY_INTF_SEL_RGMII; break; default: dev_err(plat->dev, "phy interface not supported\n"); @@ -285,27 +286,25 @@ static const struct mediatek_dwmac_variant mt2712_gmac_variant = { .tx_delay_max = 17600, }; -static int mt8195_set_interface(struct mediatek_dwmac_plat_data *plat) +static int mt8195_set_interface(struct mediatek_dwmac_plat_data *plat, + u8 phy_intf_sel) { int rmii_clk_from_mac = plat->rmii_clk_from_mac ? MT8195_RMII_CLK_SRC_INTERNAL : 0; int rmii_rxc = plat->rmii_rxc ? MT8195_RMII_CLK_SRC_RXC : 0; - u32 intf_val = 0; + u32 intf_val; + + intf_val = FIELD_PREP(MT8195_ETH_INTF_SEL, phy_intf_sel); /* select phy interface in top control domain */ switch (plat->phy_mode) { - case PHY_INTERFACE_MODE_MII: - intf_val |= FIELD_PREP(MT8195_ETH_INTF_SEL, - PHY_INTF_SEL_GMII_MII); - break; case PHY_INTERFACE_MODE_RMII: intf_val |= rmii_rxc | rmii_clk_from_mac; - intf_val |= FIELD_PREP(MT8195_ETH_INTF_SEL, PHY_INTF_SEL_RMII); break; + case PHY_INTERFACE_MODE_MII: case PHY_INTERFACE_MODE_RGMII: case PHY_INTERFACE_MODE_RGMII_TXID: case PHY_INTERFACE_MODE_RGMII_RXID: case PHY_INTERFACE_MODE_RGMII_ID: - intf_val |= FIELD_PREP(MT8195_ETH_INTF_SEL, PHY_INTF_SEL_RGMII); break; default: dev_err(plat->dev, "phy interface not supported\n"); @@ -525,10 +524,18 @@ static int mediatek_dwmac_init(struct device *dev, void *priv) { struct mediatek_dwmac_plat_data *plat = priv; const struct mediatek_dwmac_variant *variant = plat->variant; - int ret; + int phy_intf_sel, ret; if (variant->dwmac_set_phy_interface) { - ret = variant->dwmac_set_phy_interface(plat); + phy_intf_sel = stmmac_get_phy_intf_sel(plat->phy_mode); + if (phy_intf_sel != PHY_INTF_SEL_GMII_MII && + phy_intf_sel != PHY_INTF_SEL_RGMII && + phy_intf_sel != PHY_INTF_SEL_RMII) { + dev_err(plat->dev, "phy interface not supported\n"); + return phy_intf_sel < 0 ? phy_intf_sel : -EINVAL; + } + + ret = variant->dwmac_set_phy_interface(plat, phy_intf_sel); if (ret) { dev_err(dev, "failed to set phy interface, err = %d\n", ret); return ret; From f06620091fe75015d5e97f4a0da4f5b124ca6455 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Tue, 11 Nov 2025 08:12:12 +0000 Subject: [PATCH 0774/1381] net: stmmac: mediatek: simplify set_interface() methods Use the phy_intf_sel field value when deciding what other options to apply for the configuration register. Note that this will allow GMII as well as MII as the phy_intf_sel value is the same for both. Signed-off-by: Russell King (Oracle) Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/E1vIjU4-0000000DqtV-3qsX@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../ethernet/stmicro/stmmac/dwmac-mediatek.c | 50 +++++-------------- 1 file changed, 12 insertions(+), 38 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-mediatek.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-mediatek.c index 0f32732efb75..1f2d7d19ca56 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-mediatek.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-mediatek.c @@ -110,26 +110,13 @@ static const char * const mt8195_dwmac_clk_l[] = { static int mt2712_set_interface(struct mediatek_dwmac_plat_data *plat, u8 phy_intf_sel) { - int rmii_clk_from_mac = plat->rmii_clk_from_mac ? RMII_CLK_SRC_INTERNAL : 0; - int rmii_rxc = plat->rmii_rxc ? RMII_CLK_SRC_RXC : 0; - u32 intf_val; + u32 intf_val = phy_intf_sel; - intf_val = phy_intf_sel; - - /* select phy interface in top control domain */ - switch (plat->phy_mode) { - case PHY_INTERFACE_MODE_RMII: - intf_val |= rmii_rxc | rmii_clk_from_mac; - break; - case PHY_INTERFACE_MODE_MII: - case PHY_INTERFACE_MODE_RGMII: - case PHY_INTERFACE_MODE_RGMII_TXID: - case PHY_INTERFACE_MODE_RGMII_RXID: - case PHY_INTERFACE_MODE_RGMII_ID: - break; - default: - dev_err(plat->dev, "phy interface not supported\n"); - return -EINVAL; + if (phy_intf_sel == PHY_INTF_SEL_RMII) { + if (plat->rmii_clk_from_mac) + intf_val |= RMII_CLK_SRC_INTERNAL; + if (plat->rmii_rxc) + intf_val |= RMII_CLK_SRC_RXC; } regmap_write(plat->peri_regmap, PERI_ETH_PHY_INTF_SEL, intf_val); @@ -289,26 +276,13 @@ static const struct mediatek_dwmac_variant mt2712_gmac_variant = { static int mt8195_set_interface(struct mediatek_dwmac_plat_data *plat, u8 phy_intf_sel) { - int rmii_clk_from_mac = plat->rmii_clk_from_mac ? MT8195_RMII_CLK_SRC_INTERNAL : 0; - int rmii_rxc = plat->rmii_rxc ? MT8195_RMII_CLK_SRC_RXC : 0; - u32 intf_val; + u32 intf_val = FIELD_PREP(MT8195_ETH_INTF_SEL, phy_intf_sel); - intf_val = FIELD_PREP(MT8195_ETH_INTF_SEL, phy_intf_sel); - - /* select phy interface in top control domain */ - switch (plat->phy_mode) { - case PHY_INTERFACE_MODE_RMII: - intf_val |= rmii_rxc | rmii_clk_from_mac; - break; - case PHY_INTERFACE_MODE_MII: - case PHY_INTERFACE_MODE_RGMII: - case PHY_INTERFACE_MODE_RGMII_TXID: - case PHY_INTERFACE_MODE_RGMII_RXID: - case PHY_INTERFACE_MODE_RGMII_ID: - break; - default: - dev_err(plat->dev, "phy interface not supported\n"); - return -EINVAL; + if (phy_intf_sel == PHY_INTF_SEL_RMII) { + if (plat->rmii_clk_from_mac) + intf_val |= MT8195_RMII_CLK_SRC_INTERNAL; + if (plat->rmii_rxc) + intf_val |= MT8195_RMII_CLK_SRC_RXC; } /* MT8195 only support external PHY */ From f0917b475378055c1de6c9a911da579be8726a54 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Tue, 11 Nov 2025 08:12:18 +0000 Subject: [PATCH 0775/1381] net: stmmac: starfive: use PHY_INTF_SEL_x to select PHY interface Use the common dwmac definitions for the PHY interface selection field. Signed-off-by: Russell King (Oracle) Reviewed-by: Emil Renner Berthing Link: https://patch.msgid.link/E1vIjUA-0000000Dqtb-0AfP@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/dwmac-starfive.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-starfive.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-starfive.c index 6938dd2a79b7..1ef72576c6f1 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-starfive.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-starfive.c @@ -15,8 +15,8 @@ #include "stmmac_platform.h" -#define STARFIVE_DWMAC_PHY_INFT_RGMII 0x1 -#define STARFIVE_DWMAC_PHY_INFT_RMII 0x4 +#define STARFIVE_DWMAC_PHY_INFT_RGMII PHY_INTF_SEL_RGMII +#define STARFIVE_DWMAC_PHY_INFT_RMII PHY_INTF_SEL_RMII #define STARFIVE_DWMAC_PHY_INFT_FIELD 0x7U #define JH7100_SYSMAIN_REGISTER49_DLYCHAIN 0xc8 From d22045997b538aaf801b849dfa6369d76f0152c0 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Tue, 11 Nov 2025 08:12:23 +0000 Subject: [PATCH 0776/1381] net: stmmac: starfive: use stmmac_get_phy_intf_sel() Use stmmac_get_phy_intf_sel() to decode the PHY interface mode to the phy_intf_sel value, validate the result and use that to set the control register to select the operating mode for the DWMAC core. Signed-off-by: Russell King (Oracle) Reviewed-by: Emil Renner Berthing Link: https://patch.msgid.link/E1vIjUF-0000000Dqth-0gwD@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../ethernet/stmicro/stmmac/dwmac-starfive.c | 24 +++++-------------- 1 file changed, 6 insertions(+), 18 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-starfive.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-starfive.c index 1ef72576c6f1..16b955a6d77b 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-starfive.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-starfive.c @@ -15,8 +15,6 @@ #include "stmmac_platform.h" -#define STARFIVE_DWMAC_PHY_INFT_RGMII PHY_INTF_SEL_RGMII -#define STARFIVE_DWMAC_PHY_INFT_RMII PHY_INTF_SEL_RMII #define STARFIVE_DWMAC_PHY_INFT_FIELD 0x7U #define JH7100_SYSMAIN_REGISTER49_DLYCHAIN 0xc8 @@ -35,25 +33,15 @@ static int starfive_dwmac_set_mode(struct plat_stmmacenet_data *plat_dat) struct starfive_dwmac *dwmac = plat_dat->bsp_priv; struct regmap *regmap; unsigned int args[2]; - unsigned int mode; + int phy_intf_sel; int err; - switch (plat_dat->phy_interface) { - case PHY_INTERFACE_MODE_RMII: - mode = STARFIVE_DWMAC_PHY_INFT_RMII; - break; - - case PHY_INTERFACE_MODE_RGMII: - case PHY_INTERFACE_MODE_RGMII_ID: - case PHY_INTERFACE_MODE_RGMII_RXID: - case PHY_INTERFACE_MODE_RGMII_TXID: - mode = STARFIVE_DWMAC_PHY_INFT_RGMII; - break; - - default: + phy_intf_sel = stmmac_get_phy_intf_sel(plat_dat->phy_interface); + if (phy_intf_sel != PHY_INTF_SEL_RGMII && + phy_intf_sel != PHY_INTF_SEL_RMII) { dev_err(dwmac->dev, "unsupported interface %s\n", phy_modes(plat_dat->phy_interface)); - return -EINVAL; + return phy_intf_sel < 0 ? phy_intf_sel : -EINVAL; } regmap = syscon_regmap_lookup_by_phandle_args(dwmac->dev->of_node, @@ -65,7 +53,7 @@ static int starfive_dwmac_set_mode(struct plat_stmmacenet_data *plat_dat) /* args[0]:offset args[1]: shift */ err = regmap_update_bits(regmap, args[0], STARFIVE_DWMAC_PHY_INFT_FIELD << args[1], - mode << args[1]); + phy_intf_sel << args[1]); if (err) return dev_err_probe(dwmac->dev, err, "error setting phy mode\n"); From 73130c298fa0d30f5ead5d9aeb7edc813651a8b3 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Tue, 11 Nov 2025 08:12:28 +0000 Subject: [PATCH 0777/1381] net: stmmac: stm32: use PHY_INTF_SEL_x to select PHY interface Convert dwmac-stm32 to use the PHY_INTF_SEL_x definitions. For stm32mp1, the original definitions used constant 0 (GMII, 0 << 21), BIT(21) (RGMII, 1 << 21) and BIT(23) (RMII, 4 << 21) to define these, but from the values it can be clearly seen that these are the PHY_INTF_SEL_x inputs to the dwmac. For stm32mp2, the original definitions cover a bitfield 6:4 in the SYSCFG Ethernet1 control register (according to documentation) and use the PHY_INTF_SEL_x values. Use the common dwmac definitions for the PHY interface selection field by adding the bitfield mask, and using FIELD_PREP() for the bitfield values. This removes this incorrect use of BIT(). Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vIjUK-0000000Dqtn-1AyK@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../net/ethernet/stmicro/stmmac/dwmac-stm32.c | 20 +++++++++++++------ 1 file changed, 14 insertions(+), 6 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c index 6c179911ef3f..1545772419d5 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c @@ -47,9 +47,13 @@ *------------------------------------------ */ #define SYSCFG_PMCR_ETH_SEL_MII BIT(20) -#define SYSCFG_PMCR_ETH_SEL_RGMII BIT(21) -#define SYSCFG_PMCR_ETH_SEL_RMII BIT(23) -#define SYSCFG_PMCR_ETH_SEL_GMII 0 +#define SYSCFG_PMCR_PHY_INTF_SEL_MASK GENMASK(23, 21) +#define SYSCFG_PMCR_ETH_SEL_RGMII \ + FIELD_PREP(SYSCFG_PMCR_PHY_INTF_SEL_MASK, PHY_INTF_SEL_RGMII) +#define SYSCFG_PMCR_ETH_SEL_RMII \ + FIELD_PREP(SYSCFG_PMCR_PHY_INTF_SEL_MASK, PHY_INTF_SEL_RMII) +#define SYSCFG_PMCR_ETH_SEL_GMII \ + FIELD_PREP(SYSCFG_PMCR_PHY_INTF_SEL_MASK, PHY_INTF_SEL_GMII_MII) #define SYSCFG_MCU_ETH_SEL_MII 0 #define SYSCFG_MCU_ETH_SEL_RMII 1 @@ -60,9 +64,13 @@ #define SYSCFG_ETHCR_ETH_CLK_SEL BIT(1) #define SYSCFG_ETHCR_ETH_REF_CLK_SEL BIT(0) -#define SYSCFG_ETHCR_ETH_SEL_MII 0 -#define SYSCFG_ETHCR_ETH_SEL_RGMII BIT(4) -#define SYSCFG_ETHCR_ETH_SEL_RMII BIT(6) +#define SYSCFG_ETHCR_ETH_SEL_MASK GENMASK(6, 4) +#define SYSCFG_ETHCR_ETH_SEL_MII FIELD_PREP(SYSCFG_ETHCR_ETH_SEL_MASK, \ + PHY_INTF_SEL_GMII_MII) +#define SYSCFG_ETHCR_ETH_SEL_RGMII FIELD_PREP(SYSCFG_ETHCR_ETH_SEL_MASK, \ + PHY_INTF_SEL_RGMII) +#define SYSCFG_ETHCR_ETH_SEL_RMII FIELD_PREP(SYSCFG_ETHCR_ETH_SEL_MASK, \ + PHY_INTF_SEL_RMII) /* STM32MPx register definitions * From 07669cf12ea8c7e4cb2e5f34bb8351bf65917be2 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Tue, 11 Nov 2025 08:12:33 +0000 Subject: [PATCH 0778/1381] net: stmmac: stm32: use PHY_INTF_SEL_x directly Rather than defining separate constants for each, use the PHY_INTF_SEL_x definitions in the switch()es configuring the control register, and use one FIELD_PREP() to convert phy_intf_sel to the register value. Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vIjUP-0000000Dqtt-1bYn@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../net/ethernet/stmicro/stmmac/dwmac-stm32.c | 32 ++++++++----------- 1 file changed, 13 insertions(+), 19 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c index 1545772419d5..18d26f096f5f 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c @@ -48,30 +48,17 @@ */ #define SYSCFG_PMCR_ETH_SEL_MII BIT(20) #define SYSCFG_PMCR_PHY_INTF_SEL_MASK GENMASK(23, 21) -#define SYSCFG_PMCR_ETH_SEL_RGMII \ - FIELD_PREP(SYSCFG_PMCR_PHY_INTF_SEL_MASK, PHY_INTF_SEL_RGMII) -#define SYSCFG_PMCR_ETH_SEL_RMII \ - FIELD_PREP(SYSCFG_PMCR_PHY_INTF_SEL_MASK, PHY_INTF_SEL_RMII) -#define SYSCFG_PMCR_ETH_SEL_GMII \ - FIELD_PREP(SYSCFG_PMCR_PHY_INTF_SEL_MASK, PHY_INTF_SEL_GMII_MII) #define SYSCFG_MCU_ETH_SEL_MII 0 #define SYSCFG_MCU_ETH_SEL_RMII 1 /* STM32MP2 register definitions */ #define SYSCFG_MP2_ETH_MASK GENMASK(31, 0) +#define SYSCFG_ETHCR_ETH_SEL_MASK GENMASK(6, 4) #define SYSCFG_ETHCR_ETH_PTP_CLK_SEL BIT(2) #define SYSCFG_ETHCR_ETH_CLK_SEL BIT(1) #define SYSCFG_ETHCR_ETH_REF_CLK_SEL BIT(0) -#define SYSCFG_ETHCR_ETH_SEL_MASK GENMASK(6, 4) -#define SYSCFG_ETHCR_ETH_SEL_MII FIELD_PREP(SYSCFG_ETHCR_ETH_SEL_MASK, \ - PHY_INTF_SEL_GMII_MII) -#define SYSCFG_ETHCR_ETH_SEL_RGMII FIELD_PREP(SYSCFG_ETHCR_ETH_SEL_MASK, \ - PHY_INTF_SEL_RGMII) -#define SYSCFG_ETHCR_ETH_SEL_RMII FIELD_PREP(SYSCFG_ETHCR_ETH_SEL_MASK, \ - PHY_INTF_SEL_RMII) - /* STM32MPx register definitions * * Below table summarizes the clock requirement and clock sources for @@ -244,10 +231,12 @@ static int stm32mp1_configure_pmcr(struct plat_stmmacenet_data *plat_dat) { struct stm32_dwmac *dwmac = plat_dat->bsp_priv; u32 reg = dwmac->mode_reg; + u8 phy_intf_sel; int val = 0; switch (plat_dat->phy_interface) { case PHY_INTERFACE_MODE_MII: + phy_intf_sel = PHY_INTF_SEL_GMII_MII; /* * STM32MP15xx supports both MII and GMII, STM32MP13xx MII only. * SYSCFG_PMCSETR ETH_SELMII is present only on STM32MP15xx and @@ -258,12 +247,12 @@ static int stm32mp1_configure_pmcr(struct plat_stmmacenet_data *plat_dat) val |= SYSCFG_PMCR_ETH_SEL_MII; break; case PHY_INTERFACE_MODE_GMII: - val = SYSCFG_PMCR_ETH_SEL_GMII; + phy_intf_sel = PHY_INTF_SEL_GMII_MII; if (dwmac->enable_eth_ck) val |= SYSCFG_PMCR_ETH_CLK_SEL; break; case PHY_INTERFACE_MODE_RMII: - val = SYSCFG_PMCR_ETH_SEL_RMII; + phy_intf_sel = PHY_INTF_SEL_RMII; if (dwmac->enable_eth_ck) val |= SYSCFG_PMCR_ETH_REF_CLK_SEL; break; @@ -271,7 +260,7 @@ static int stm32mp1_configure_pmcr(struct plat_stmmacenet_data *plat_dat) case PHY_INTERFACE_MODE_RGMII_ID: case PHY_INTERFACE_MODE_RGMII_RXID: case PHY_INTERFACE_MODE_RGMII_TXID: - val = SYSCFG_PMCR_ETH_SEL_RGMII; + phy_intf_sel = PHY_INTF_SEL_RGMII; if (dwmac->enable_eth_ck) val |= SYSCFG_PMCR_ETH_CLK_SEL; break; @@ -284,6 +273,8 @@ static int stm32mp1_configure_pmcr(struct plat_stmmacenet_data *plat_dat) dev_dbg(dwmac->dev, "Mode %s", phy_modes(plat_dat->phy_interface)); + val |= FIELD_PREP(SYSCFG_PMCR_PHY_INTF_SEL_MASK, phy_intf_sel); + /* Shift value at correct ethernet MAC offset in SYSCFG_PMCSETR */ val <<= ffs(dwmac->mode_mask) - ffs(SYSCFG_MP1_ETH_MASK); @@ -299,6 +290,7 @@ static int stm32mp1_configure_pmcr(struct plat_stmmacenet_data *plat_dat) static int stm32mp2_configure_syscfg(struct plat_stmmacenet_data *plat_dat) { struct stm32_dwmac *dwmac = plat_dat->bsp_priv; + u8 phy_intf_sel = PHY_INTF_SEL_GMII_MII; u32 reg = dwmac->mode_reg; int val = 0; @@ -307,7 +299,7 @@ static int stm32mp2_configure_syscfg(struct plat_stmmacenet_data *plat_dat) /* ETH_REF_CLK_SEL bit in SYSCFG register is not applicable in MII mode */ break; case PHY_INTERFACE_MODE_RMII: - val = SYSCFG_ETHCR_ETH_SEL_RMII; + phy_intf_sel = PHY_INTF_SEL_RMII; if (dwmac->enable_eth_ck) { /* Internal clock ETH_CLK of 50MHz from RCC is used */ val |= SYSCFG_ETHCR_ETH_REF_CLK_SEL; @@ -317,7 +309,7 @@ static int stm32mp2_configure_syscfg(struct plat_stmmacenet_data *plat_dat) case PHY_INTERFACE_MODE_RGMII_ID: case PHY_INTERFACE_MODE_RGMII_RXID: case PHY_INTERFACE_MODE_RGMII_TXID: - val = SYSCFG_ETHCR_ETH_SEL_RGMII; + phy_intf_sel = PHY_INTF_SEL_RGMII; fallthrough; case PHY_INTERFACE_MODE_GMII: if (dwmac->enable_eth_ck) { @@ -334,6 +326,8 @@ static int stm32mp2_configure_syscfg(struct plat_stmmacenet_data *plat_dat) dev_dbg(dwmac->dev, "Mode %s", phy_modes(plat_dat->phy_interface)); + val |= FIELD_PREP(SYSCFG_ETHCR_ETH_SEL_MASK, phy_intf_sel); + /* Select PTP (IEEE1588) clock selection from RCC (ck_ker_ethxptp) */ val |= SYSCFG_ETHCR_ETH_PTP_CLK_SEL; From 45c5e24a53f10cabb4a311c915680624b832cc2b Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Tue, 11 Nov 2025 08:12:38 +0000 Subject: [PATCH 0779/1381] net: stmmac: stm32: use stmmac_get_phy_intf_sel() Use stmmac_get_phy_intf_sel() to decode the PHY interface mode to the phy_intf_sel value. As both configure functions would end up with the same code, call this from stm32mp1_set_mode(), validate the result and pass the resulting value into the stm32 configure function. Use this value to set the operating mode for the DWMAC core. Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vIjUU-0000000Dqtz-2PwT@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../net/ethernet/stmicro/stmmac/dwmac-stm32.c | 42 ++++++++++--------- 1 file changed, 22 insertions(+), 20 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c index 18d26f096f5f..e1b260ed4790 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c @@ -227,16 +227,17 @@ static int stm32mp1_validate_ethck_rate(struct plat_stmmacenet_data *plat_dat) return -EINVAL; } -static int stm32mp1_configure_pmcr(struct plat_stmmacenet_data *plat_dat) +static int stm32mp1_configure_pmcr(struct plat_stmmacenet_data *plat_dat, + u8 phy_intf_sel) { struct stm32_dwmac *dwmac = plat_dat->bsp_priv; u32 reg = dwmac->mode_reg; - u8 phy_intf_sel; - int val = 0; + int val; + + val = FIELD_PREP(SYSCFG_PMCR_PHY_INTF_SEL_MASK, phy_intf_sel); switch (plat_dat->phy_interface) { case PHY_INTERFACE_MODE_MII: - phy_intf_sel = PHY_INTF_SEL_GMII_MII; /* * STM32MP15xx supports both MII and GMII, STM32MP13xx MII only. * SYSCFG_PMCSETR ETH_SELMII is present only on STM32MP15xx and @@ -247,12 +248,10 @@ static int stm32mp1_configure_pmcr(struct plat_stmmacenet_data *plat_dat) val |= SYSCFG_PMCR_ETH_SEL_MII; break; case PHY_INTERFACE_MODE_GMII: - phy_intf_sel = PHY_INTF_SEL_GMII_MII; if (dwmac->enable_eth_ck) val |= SYSCFG_PMCR_ETH_CLK_SEL; break; case PHY_INTERFACE_MODE_RMII: - phy_intf_sel = PHY_INTF_SEL_RMII; if (dwmac->enable_eth_ck) val |= SYSCFG_PMCR_ETH_REF_CLK_SEL; break; @@ -260,7 +259,6 @@ static int stm32mp1_configure_pmcr(struct plat_stmmacenet_data *plat_dat) case PHY_INTERFACE_MODE_RGMII_ID: case PHY_INTERFACE_MODE_RGMII_RXID: case PHY_INTERFACE_MODE_RGMII_TXID: - phy_intf_sel = PHY_INTF_SEL_RGMII; if (dwmac->enable_eth_ck) val |= SYSCFG_PMCR_ETH_CLK_SEL; break; @@ -273,8 +271,6 @@ static int stm32mp1_configure_pmcr(struct plat_stmmacenet_data *plat_dat) dev_dbg(dwmac->dev, "Mode %s", phy_modes(plat_dat->phy_interface)); - val |= FIELD_PREP(SYSCFG_PMCR_PHY_INTF_SEL_MASK, phy_intf_sel); - /* Shift value at correct ethernet MAC offset in SYSCFG_PMCSETR */ val <<= ffs(dwmac->mode_mask) - ffs(SYSCFG_MP1_ETH_MASK); @@ -287,19 +283,20 @@ static int stm32mp1_configure_pmcr(struct plat_stmmacenet_data *plat_dat) dwmac->mode_mask, val); } -static int stm32mp2_configure_syscfg(struct plat_stmmacenet_data *plat_dat) +static int stm32mp2_configure_syscfg(struct plat_stmmacenet_data *plat_dat, + u8 phy_intf_sel) { struct stm32_dwmac *dwmac = plat_dat->bsp_priv; - u8 phy_intf_sel = PHY_INTF_SEL_GMII_MII; u32 reg = dwmac->mode_reg; - int val = 0; + int val; + + val = FIELD_PREP(SYSCFG_ETHCR_ETH_SEL_MASK, phy_intf_sel); switch (plat_dat->phy_interface) { case PHY_INTERFACE_MODE_MII: /* ETH_REF_CLK_SEL bit in SYSCFG register is not applicable in MII mode */ break; case PHY_INTERFACE_MODE_RMII: - phy_intf_sel = PHY_INTF_SEL_RMII; if (dwmac->enable_eth_ck) { /* Internal clock ETH_CLK of 50MHz from RCC is used */ val |= SYSCFG_ETHCR_ETH_REF_CLK_SEL; @@ -309,8 +306,6 @@ static int stm32mp2_configure_syscfg(struct plat_stmmacenet_data *plat_dat) case PHY_INTERFACE_MODE_RGMII_ID: case PHY_INTERFACE_MODE_RGMII_RXID: case PHY_INTERFACE_MODE_RGMII_TXID: - phy_intf_sel = PHY_INTF_SEL_RGMII; - fallthrough; case PHY_INTERFACE_MODE_GMII: if (dwmac->enable_eth_ck) { /* Internal clock ETH_CLK of 125MHz from RCC is used */ @@ -326,8 +321,6 @@ static int stm32mp2_configure_syscfg(struct plat_stmmacenet_data *plat_dat) dev_dbg(dwmac->dev, "Mode %s", phy_modes(plat_dat->phy_interface)); - val |= FIELD_PREP(SYSCFG_ETHCR_ETH_SEL_MASK, phy_intf_sel); - /* Select PTP (IEEE1588) clock selection from RCC (ck_ker_ethxptp) */ val |= SYSCFG_ETHCR_ETH_PTP_CLK_SEL; @@ -339,7 +332,7 @@ static int stm32mp2_configure_syscfg(struct plat_stmmacenet_data *plat_dat) static int stm32mp1_set_mode(struct plat_stmmacenet_data *plat_dat) { struct stm32_dwmac *dwmac = plat_dat->bsp_priv; - int ret; + int phy_intf_sel, ret; ret = stm32mp1_select_ethck_external(plat_dat); if (ret) @@ -349,10 +342,19 @@ static int stm32mp1_set_mode(struct plat_stmmacenet_data *plat_dat) if (ret) return ret; + phy_intf_sel = stmmac_get_phy_intf_sel(plat_dat->phy_interface); + if (phy_intf_sel != PHY_INTF_SEL_GMII_MII && + phy_intf_sel != PHY_INTF_SEL_RGMII && + phy_intf_sel != PHY_INTF_SEL_RMII) { + dev_err(dwmac->dev, "Mode %s not supported\n", + phy_modes(plat_dat->phy_interface)); + return phy_intf_sel < 0 ? phy_intf_sel : -EINVAL; + } + if (!dwmac->ops->is_mp2) - return stm32mp1_configure_pmcr(plat_dat); + return stm32mp1_configure_pmcr(plat_dat, phy_intf_sel); else - return stm32mp2_configure_syscfg(plat_dat); + return stm32mp2_configure_syscfg(plat_dat, phy_intf_sel); } static int stm32mcu_set_mode(struct plat_stmmacenet_data *plat_dat) From bb68e0183e0452d0a1f3af747ec33976a9335ee5 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Tue, 11 Nov 2025 08:12:43 +0000 Subject: [PATCH 0780/1381] net: stmmac: visconti: use PHY_INTF_SEL_x to select PHY interface Convert dwmac-visconti to use the PHY_INTF_SEL_x definitions. The original definitions used constant 0, BIT(0) (==1) and BIT(2) (==4) to define these, but the values of the bits corresponds with the PHY_INTF_SEL_x values, so it is highly likely that these are not individual bits, but the PHY_INTF_SEL_x bitfield. This removes this incorrect use of BIT(). Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vIjUZ-0000000Dqu5-2sDI@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/dwmac-visconti.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-visconti.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-visconti.c index bd65d4239054..7b6b048e1be0 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-visconti.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-visconti.c @@ -42,9 +42,9 @@ #define ETHER_CLK_SEL_RX_TX_CLK_EN (ETHER_CLK_SEL_RX_CLK_EN | ETHER_CLK_SEL_TX_CLK_EN) -#define ETHER_CONFIG_INTF_MII 0 -#define ETHER_CONFIG_INTF_RGMII BIT(0) -#define ETHER_CONFIG_INTF_RMII BIT(2) +#define ETHER_CONFIG_INTF_MII PHY_INTF_SEL_GMII_MII +#define ETHER_CONFIG_INTF_RGMII PHY_INTF_SEL_RGMII +#define ETHER_CONFIG_INTF_RMII PHY_INTF_SEL_RMII struct visconti_eth { void __iomem *reg; From ccb4ff9f24e23f3599dd13e124f0c7cd29380df3 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Tue, 11 Nov 2025 08:12:48 +0000 Subject: [PATCH 0781/1381] net: stmmac: visconti: use stmmac_get_phy_intf_sel() Use stmmac_get_phy_intf_sel() to decode the PHY interface mode to the phy_intf_sel value, validate the result and use that to set the control register to select the operating mode for the DWMAC core. Note that this will allow GMII as well as MII as the phy_intf_sel value is the same for both. Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vIjUe-0000000DquB-3JDY@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../ethernet/stmicro/stmmac/dwmac-visconti.c | 24 ++++--------------- 1 file changed, 5 insertions(+), 19 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-visconti.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-visconti.c index 7b6b048e1be0..9497b13a5753 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-visconti.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-visconti.c @@ -42,10 +42,6 @@ #define ETHER_CLK_SEL_RX_TX_CLK_EN (ETHER_CLK_SEL_RX_CLK_EN | ETHER_CLK_SEL_TX_CLK_EN) -#define ETHER_CONFIG_INTF_MII PHY_INTF_SEL_GMII_MII -#define ETHER_CONFIG_INTF_RGMII PHY_INTF_SEL_RGMII -#define ETHER_CONFIG_INTF_RMII PHY_INTF_SEL_RMII - struct visconti_eth { void __iomem *reg; struct clk *phy_ref_clk; @@ -150,22 +146,12 @@ static int visconti_eth_init_hw(struct platform_device *pdev, struct plat_stmmac { struct visconti_eth *dwmac = plat_dat->bsp_priv; unsigned int clk_sel_val; - u32 phy_intf_sel; + int phy_intf_sel; - switch (plat_dat->phy_interface) { - case PHY_INTERFACE_MODE_RGMII: - case PHY_INTERFACE_MODE_RGMII_ID: - case PHY_INTERFACE_MODE_RGMII_RXID: - case PHY_INTERFACE_MODE_RGMII_TXID: - phy_intf_sel = ETHER_CONFIG_INTF_RGMII; - break; - case PHY_INTERFACE_MODE_MII: - phy_intf_sel = ETHER_CONFIG_INTF_MII; - break; - case PHY_INTERFACE_MODE_RMII: - phy_intf_sel = ETHER_CONFIG_INTF_RMII; - break; - default: + phy_intf_sel = stmmac_get_phy_intf_sel(plat_dat->phy_interface); + if (phy_intf_sel != PHY_INTF_SEL_GMII_MII && + phy_intf_sel != PHY_INTF_SEL_RGMII && + phy_intf_sel != PHY_INTF_SEL_RMII) { dev_err(&pdev->dev, "Unsupported phy-mode (%d)\n", plat_dat->phy_interface); return -EOPNOTSUPP; } From 9f07af1d274223a4314b5e2e6d395a78166c24c5 Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Mon, 10 Nov 2025 22:20:26 +0100 Subject: [PATCH 0782/1381] net: phy: fixed_phy: initialize the link status as up All callers initialize the link status as up. This change is in line with how of_phy_register_fixed_link() behaves. Signed-off-by: Heiner Kallweit Link: https://patch.msgid.link/45f644e8-2292-4787-a27a-f69084c93218@gmail.com Signed-off-by: Paolo Abeni --- drivers/net/phy/fixed_phy.c | 13 +++++-------- 1 file changed, 5 insertions(+), 8 deletions(-) diff --git a/drivers/net/phy/fixed_phy.c b/drivers/net/phy/fixed_phy.c index 9bd6937411e4..715f0356f895 100644 --- a/drivers/net/phy/fixed_phy.c +++ b/drivers/net/phy/fixed_phy.c @@ -174,13 +174,11 @@ struct phy_device *fixed_phy_register(const struct fixed_phy_status *status, } /* propagate the fixed link values to struct phy_device */ - phy->link = status->link; - if (status->link) { - phy->speed = status->speed; - phy->duplex = status->duplex; - phy->pause = status->pause; - phy->asym_pause = status->asym_pause; - } + phy->link = 1; + phy->speed = status->speed; + phy->duplex = status->duplex; + phy->pause = status->pause; + phy->asym_pause = status->asym_pause; of_node_get(np); phy->mdio.dev.of_node = np; @@ -224,7 +222,6 @@ EXPORT_SYMBOL_GPL(fixed_phy_register); struct phy_device *fixed_phy_register_100fd(void) { static const struct fixed_phy_status status = { - .link = true, .speed = SPEED_100, .duplex = DUPLEX_FULL, }; From d0af55df5a786cb8f11ca1923cd7dc8b2af01309 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Tue, 11 Nov 2025 11:26:39 +0000 Subject: [PATCH 0783/1381] net: stmmac: clean up stmmac_reset() stmmac_reset() takes the stmmac_priv and an ioaddr. It has one call site, which passes the priv pointer, and dereferences priv for the ioaddr. stmmac_reset() then checks whether priv is NULL. If it was, the caller would have oopsed. Remove the checks for NULL, and move the dereference for ioaddr into stmmac_reset(). Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vImWF-0000000DrIr-1fmn@rmk-PC.armlinux.org.uk Signed-off-by: Paolo Abeni --- drivers/net/ethernet/stmicro/stmmac/hwif.c | 8 +++----- drivers/net/ethernet/stmicro/stmmac/hwif.h | 2 +- drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 2 +- 3 files changed, 5 insertions(+), 7 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/hwif.c b/drivers/net/ethernet/stmicro/stmmac/hwif.c index 8212441f9826..ee612cadbd77 100644 --- a/drivers/net/ethernet/stmicro/stmmac/hwif.c +++ b/drivers/net/ethernet/stmicro/stmmac/hwif.c @@ -103,12 +103,10 @@ static int stmmac_dwxlgmac_quirks(struct stmmac_priv *priv) return 0; } -int stmmac_reset(struct stmmac_priv *priv, void __iomem *ioaddr) +int stmmac_reset(struct stmmac_priv *priv) { - struct plat_stmmacenet_data *plat = priv ? priv->plat : NULL; - - if (!priv) - return -EINVAL; + struct plat_stmmacenet_data *plat = priv->plat; + void __iomem *ioaddr = priv->ioaddr; if (plat && plat->fix_soc_reset) return plat->fix_soc_reset(priv, ioaddr); diff --git a/drivers/net/ethernet/stmicro/stmmac/hwif.h b/drivers/net/ethernet/stmicro/stmmac/hwif.h index cb8fc09caf86..d359722100fa 100644 --- a/drivers/net/ethernet/stmicro/stmmac/hwif.h +++ b/drivers/net/ethernet/stmicro/stmmac/hwif.h @@ -698,7 +698,7 @@ extern const struct stmmac_tc_ops dwmac510_tc_ops; #define GMAC_VERSION 0x00000020 /* GMAC CORE Version */ #define GMAC4_VERSION 0x00000110 /* GMAC4+ CORE Version */ -int stmmac_reset(struct stmmac_priv *priv, void __iomem *ioaddr); +int stmmac_reset(struct stmmac_priv *priv); int stmmac_hwif_init(struct stmmac_priv *priv); #endif /* __STMMAC_HWIF_H__ */ diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index eb4350193996..d202f604161e 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -3162,7 +3162,7 @@ static int stmmac_init_dma_engine(struct stmmac_priv *priv) if (ret) return ret; - ret = stmmac_reset(priv, priv->ioaddr); + ret = stmmac_reset(priv); if (ret) { netdev_err(priv->dev, "Failed to reset the dma\n"); return ret; From f694d215d34035cc64b1d176fd82db0d1f2428d4 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Tue, 11 Nov 2025 11:26:44 +0000 Subject: [PATCH 0784/1381] net: stmmac: always allocate mac_device_info The ->setup() method implemented by dwmac-loongson and dwmac-sun8i allocate the mac_device_info structure, as does stmmac_hwif_init(). This makes no sense. Have stmmac_hwif_init() always allocate this structure, and pass it to the ->setup() method to initialise when it is provided. Rename this method to "mac_setup" to more accurately describe what it is doing. Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vImWK-0000000DrIx-28vO@rmk-PC.armlinux.org.uk Signed-off-by: Paolo Abeni --- .../ethernet/stmicro/stmmac/dwmac-loongson.c | 13 ++++--------- .../net/ethernet/stmicro/stmmac/dwmac-sun8i.c | 11 +++-------- drivers/net/ethernet/stmicro/stmmac/hwif.c | 18 ++++++++++-------- include/linux/stmmac.h | 4 +++- 4 files changed, 20 insertions(+), 26 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c index 2a3ac0136cdb..dd2fc39ec3e2 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c @@ -320,10 +320,9 @@ static int loongson_dwmac_dma_interrupt(struct stmmac_priv *priv, return ret; } -static struct mac_device_info *loongson_dwmac_setup(void *apriv) +static int loongson_dwmac_setup(void *apriv, struct mac_device_info *mac) { struct stmmac_priv *priv = apriv; - struct mac_device_info *mac; struct stmmac_dma_ops *dma; struct loongson_data *ld; struct pci_dev *pdev; @@ -331,13 +330,9 @@ static struct mac_device_info *loongson_dwmac_setup(void *apriv) ld = priv->plat->bsp_priv; pdev = to_pci_dev(priv->device); - mac = devm_kzalloc(priv->device, sizeof(*mac), GFP_KERNEL); - if (!mac) - return NULL; - dma = devm_kzalloc(priv->device, sizeof(*dma), GFP_KERNEL); if (!dma) - return NULL; + return -ENOMEM; /* The Loongson GMAC and GNET devices are based on the DW GMAC * v3.50a and v3.73a IP-cores. But the HW designers have changed @@ -396,7 +391,7 @@ static struct mac_device_info *loongson_dwmac_setup(void *apriv) mac->mii.clk_csr_shift = 2; mac->mii.clk_csr_mask = GENMASK(5, 2); - return mac; + return 0; } static int loongson_dwmac_msi_config(struct pci_dev *pdev, @@ -598,7 +593,7 @@ static int loongson_dwmac_probe(struct pci_dev *pdev, const struct pci_device_id goto err_disable_device; plat->bsp_priv = ld; - plat->setup = loongson_dwmac_setup; + plat->mac_setup = loongson_dwmac_setup; plat->fix_soc_reset = loongson_dwmac_fix_reset; plat->suspend = loongson_dwmac_suspend; plat->resume = loongson_dwmac_resume; diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c index 5d871b2cd111..7434d4bbb526 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c @@ -1040,15 +1040,10 @@ static const struct stmmac_ops sun8i_dwmac_ops = { .set_mac_loopback = sun8i_dwmac_set_mac_loopback, }; -static struct mac_device_info *sun8i_dwmac_setup(void *ppriv) +static int sun8i_dwmac_setup(void *ppriv, struct mac_device_info *mac) { - struct mac_device_info *mac; struct stmmac_priv *priv = ppriv; - mac = devm_kzalloc(priv->device, sizeof(*mac), GFP_KERNEL); - if (!mac) - return NULL; - mac->pcsr = priv->ioaddr; mac->mac = &sun8i_dwmac_ops; mac->dma = &sun8i_dwmac_dma_ops; @@ -1079,7 +1074,7 @@ static struct mac_device_info *sun8i_dwmac_setup(void *ppriv) /* Synopsys Id is not available */ priv->synopsys_id = 0; - return mac; + return 0; } static struct regmap *sun8i_dwmac_get_syscon_from_dev(struct device_node *node) @@ -1192,7 +1187,7 @@ static int sun8i_dwmac_probe(struct platform_device *pdev) plat_dat->bsp_priv = gmac; plat_dat->init = sun8i_dwmac_init; plat_dat->exit = sun8i_dwmac_exit; - plat_dat->setup = sun8i_dwmac_setup; + plat_dat->mac_setup = sun8i_dwmac_setup; plat_dat->tx_fifo_size = 4096; plat_dat->rx_fifo_size = 16384; diff --git a/drivers/net/ethernet/stmicro/stmmac/hwif.c b/drivers/net/ethernet/stmicro/stmmac/hwif.c index ee612cadbd77..014f7cd79a3c 100644 --- a/drivers/net/ethernet/stmicro/stmmac/hwif.c +++ b/drivers/net/ethernet/stmicro/stmmac/hwif.c @@ -347,17 +347,19 @@ int stmmac_hwif_init(struct stmmac_priv *priv) priv->estaddr = priv->ioaddr + EST_XGMAC_OFFSET; } - /* Check for HW specific setup first */ - if (priv->plat->setup) { - mac = priv->plat->setup(priv); - needs_setup = false; - } else { - mac = devm_kzalloc(priv->device, sizeof(*mac), GFP_KERNEL); - } - + mac = devm_kzalloc(priv->device, sizeof(*mac), GFP_KERNEL); if (!mac) return -ENOMEM; + /* Check for HW specific setup first */ + if (priv->plat->mac_setup) { + ret = priv->plat->mac_setup(priv, mac); + if (ret) + return ret; + + needs_setup = false; + } + spin_lock_init(&mac->irq_ctrl_lock); /* Fallback to generic HW */ diff --git a/include/linux/stmmac.h b/include/linux/stmmac.h index 48e9f1d4e17e..4f70a6551e68 100644 --- a/include/linux/stmmac.h +++ b/include/linux/stmmac.h @@ -192,6 +192,8 @@ enum dwmac_core_type { #define STMMAC_FLAG_EN_TX_LPI_CLK_PHY_CAP BIT(12) #define STMMAC_FLAG_HWTSTAMP_CORRECT_LATENCY BIT(13) +struct mac_device_info; + struct plat_stmmacenet_data { enum dwmac_core_type core_type; int bus_id; @@ -266,7 +268,7 @@ struct plat_stmmacenet_data { void (*exit)(struct platform_device *pdev, void *priv); int (*suspend)(struct device *dev, void *priv); int (*resume)(struct device *dev, void *priv); - struct mac_device_info *(*setup)(void *priv); + int (*mac_setup)(void *priv, struct mac_device_info *mac); int (*clks_config)(void *priv, bool enabled); int (*crosststamp)(ktime_t *device, struct system_counterval_t *system, void *ctx); From c3716126cf578918bc1fe6cf4fb6624731975d56 Mon Sep 17 00:00:00 2001 From: Yue Haibing Date: Wed, 12 Nov 2025 17:20:55 +0800 Subject: [PATCH 0785/1381] vxlan: Remove unused declarations eth_vni_hash() and fdb_head_index() Commit 1f763fa808e9 ("vxlan: Convert FDB table to rhashtable") removed the implementations but leave declarations. Signed-off-by: Yue Haibing Reviewed-by: Ido Schimmel Reviewed-by: Nikolay Aleksandrov Link: https://patch.msgid.link/20251112092055.3546703-1-yuehaibing@huawei.com Signed-off-by: Jakub Kicinski --- drivers/net/vxlan/vxlan_private.h | 2 -- 1 file changed, 2 deletions(-) diff --git a/drivers/net/vxlan/vxlan_private.h b/drivers/net/vxlan/vxlan_private.h index 99fe772ad679..b1eec2216360 100644 --- a/drivers/net/vxlan/vxlan_private.h +++ b/drivers/net/vxlan/vxlan_private.h @@ -188,8 +188,6 @@ int __vxlan_fdb_delete(struct vxlan_dev *vxlan, const unsigned char *addr, union vxlan_addr ip, __be16 port, __be32 src_vni, __be32 vni, u32 ifindex, bool swdev_notify); -u32 eth_vni_hash(const unsigned char *addr, __be32 vni); -u32 fdb_head_index(struct vxlan_dev *vxlan, const u8 *mac, __be32 vni); int vxlan_fdb_update(struct vxlan_dev *vxlan, const u8 *mac, union vxlan_addr *ip, __u16 state, __u16 flags, From 205305c028ad986d0649b8b100bab6032dcd1bb5 Mon Sep 17 00:00:00 2001 From: Chen Ni Date: Wed, 12 Nov 2025 15:27:09 +0800 Subject: [PATCH 0786/1381] net/sched: act_ife: convert comma to semicolon Replace comma between expressions with semicolons. Using a ',' in place of a ';' can have unintended side effects. Although that is not the case here, it is seems best to use ';' unless ',' is intended. Found by inspection. No functional change intended. Compile tested only. Signed-off-by: Chen Ni Reviewed-by: Jamal Hadi Salim Link: https://patch.msgid.link/20251112072709.73755-1-nichen@iscas.ac.cn Signed-off-by: Jakub Kicinski --- net/sched/act_ife.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/net/sched/act_ife.c b/net/sched/act_ife.c index 7c6975632fc2..1dfdda6c2d4c 100644 --- a/net/sched/act_ife.c +++ b/net/sched/act_ife.c @@ -649,9 +649,9 @@ static int tcf_ife_dump(struct sk_buff *skb, struct tc_action *a, int bind, memset(&opt, 0, sizeof(opt)); - opt.index = ife->tcf_index, - opt.refcnt = refcount_read(&ife->tcf_refcnt) - ref, - opt.bindcnt = atomic_read(&ife->tcf_bindcnt) - bind, + opt.index = ife->tcf_index; + opt.refcnt = refcount_read(&ife->tcf_refcnt) - ref; + opt.bindcnt = atomic_read(&ife->tcf_bindcnt) - bind; spin_lock_bh(&ife->tcf_lock); opt.action = ife->tcf_action; From 992b7d5fd8a889dc77e384143ad5b4f5fe2a70de Mon Sep 17 00:00:00 2001 From: Kriish Sharma Date: Wed, 12 Nov 2025 05:56:42 +0000 Subject: [PATCH 0787/1381] dpll: zl3073x: fix kernel-doc name and missing parameter in fw.c Documentation build reported: Warning: drivers/dpll/zl3073x/fw.c:365 function parameter 'comp' not described in 'zl3073x_fw_component_flash' Warning: drivers/dpll/zl3073x/fw.c:365 expecting prototype for zl3073x_flash_bundle_flash(). Prototype was for zl3073x_fw_component_flash() instead Warning: drivers/dpll/zl3073x/fw.c:365 No description found for return value of 'zl3073x_fw_component_flash' The kernel-doc comment above `zl3073x_fw_component_flash()` used the wrong function name (`zl3073x_flash_bundle_flash`) and omitted the `@comp` parameter. Update the comment to correctly document the `zl3073x_fw_component_flash()` function and its arguments. Signed-off-by: Kriish Sharma Reviewed-by: Ivan Vecera Link: https://patch.msgid.link/20251112055642.2597450-1-kriish.sharma2006@gmail.com Signed-off-by: Jakub Kicinski --- drivers/dpll/zl3073x/fw.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/dpll/zl3073x/fw.c b/drivers/dpll/zl3073x/fw.c index def37fe8d9b0..55b638247f4b 100644 --- a/drivers/dpll/zl3073x/fw.c +++ b/drivers/dpll/zl3073x/fw.c @@ -352,12 +352,12 @@ struct zl3073x_fw *zl3073x_fw_load(struct zl3073x_dev *zldev, const char *data, } /** - * zl3073x_flash_bundle_flash - Flash all components + * zl3073x_fw_component_flash - Flash all components * @zldev: zl3073x device structure - * @components: pointer to components array + * @comp: pointer to components array * @extack: netlink extack pointer to report errors * - * Returns 0 in case of success or negative number otherwise. + * Return: 0 in case of success or negative number otherwise. */ static int zl3073x_fw_component_flash(struct zl3073x_dev *zldev, From 4aa73c6051cb65046e6fa601b7877b5c1e6edc85 Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Wed, 12 Nov 2025 21:46:24 +0100 Subject: [PATCH 0788/1381] net: dsa: remove definition of struct dsa_switch_driver Since 93e86b3bc842 ("net: dsa: Remove legacy probing support") this struct has no user any longer. Signed-off-by: Heiner Kallweit Reviewed-by: Vladimir Oltean Link: https://patch.msgid.link/4053a98f-052f-4dc1-a3d4-ed9b3d3cc7cb@gmail.com Signed-off-by: Jakub Kicinski --- include/net/dsa.h | 5 ----- 1 file changed, 5 deletions(-) diff --git a/include/net/dsa.h b/include/net/dsa.h index 2df2e2ead9a8..97d5f401cfcf 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -1314,11 +1314,6 @@ static inline int dsa_devlink_port_to_port(struct devlink_port *port) return port->index; } -struct dsa_switch_driver { - struct list_head list; - const struct dsa_switch_ops *ops; -}; - bool dsa_fdb_present_in_other_db(struct dsa_switch *ds, int port, const unsigned char *addr, u16 vid, struct dsa_db db); From 55f943c6af6d12a1b44978f7e87c90f9a9c05806 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Wed, 12 Nov 2025 23:11:18 +0200 Subject: [PATCH 0789/1381] net: pcs: xpcs-plat: fix MODULE_AUTHOR This field needs to hold just Serge's name. Signed-off-by: Vladimir Oltean Link: https://patch.msgid.link/20251112211118.700875-1-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- drivers/net/pcs/pcs-xpcs-plat.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/pcs/pcs-xpcs-plat.c b/drivers/net/pcs/pcs-xpcs-plat.c index 9e1ccc319a1d..c422e8d8b89f 100644 --- a/drivers/net/pcs/pcs-xpcs-plat.c +++ b/drivers/net/pcs/pcs-xpcs-plat.c @@ -456,5 +456,5 @@ static struct platform_driver xpcs_plat_driver = { module_platform_driver(xpcs_plat_driver); MODULE_DESCRIPTION("Synopsys DesignWare XPCS platform device driver"); -MODULE_AUTHOR("Signed-off-by: Serge Semin "); +MODULE_AUTHOR("Serge Semin "); MODULE_LICENSE("GPL"); From c294432be1509205022de2cb5bf2a8e513861f4c Mon Sep 17 00:00:00 2001 From: Felix Maurer Date: Wed, 12 Nov 2025 18:29:53 +0100 Subject: [PATCH 0790/1381] netlink: specs: rt-link: Add attributes for hsr YNL wasn't able to decode the linkinfo from hsr interfaces. Add the linkinfo attribute definitions for hsr interfaces. Example output now looks like this: $ ynl --spec Documentation/netlink/specs/rt-link.yaml --do getlink \ --json '{"ifname": "hsr0"}' --output-json | jq .linkinfo { "kind": "hsr", "data": { "slave1": 15, "slave2": 13, "supervision-addr": "01:15:4e:00:01:00", "seq-nr": 64511, "version": 1, "protocol": 0 } } Signed-off-by: Felix Maurer Link: https://patch.msgid.link/926077a70de614f1539c905d06515e258905255e.1762968225.git.fmaurer@redhat.com Signed-off-by: Jakub Kicinski --- Documentation/netlink/specs/rt-link.yaml | 32 ++++++++++++++++++++++++ 1 file changed, 32 insertions(+) diff --git a/Documentation/netlink/specs/rt-link.yaml b/Documentation/netlink/specs/rt-link.yaml index 2a23e9699c0b..e07341582771 100644 --- a/Documentation/netlink/specs/rt-link.yaml +++ b/Documentation/netlink/specs/rt-link.yaml @@ -1913,6 +1913,35 @@ attribute-sets: name: port-range type: binary struct: ifla-geneve-port-range + - + name: linkinfo-hsr-attrs + name-prefix: ifla-hsr- + attributes: + - + name: slave1 + type: u32 + - + name: slave2 + type: u32 + - + name: multicast-spec + type: u8 + - + name: supervision-addr + type: binary + display-hint: mac + - + name: seq-nr + type: u16 + - + name: version + type: u8 + - + name: protocol + type: u8 + - + name: interlink + type: u32 - name: linkinfo-iptun-attrs name-prefix: ifla-iptun- @@ -2299,6 +2328,9 @@ sub-messages: - value: geneve attribute-set: linkinfo-geneve-attrs + - + value: hsr + attribute-set: linkinfo-hsr-attrs - value: ipip attribute-set: linkinfo-iptun-attrs From de9c41624c9cda3a96a9ac6cd9b048d9a25d787d Mon Sep 17 00:00:00 2001 From: Oliver Neukum Date: Wed, 12 Nov 2025 11:25:00 +0100 Subject: [PATCH 0791/1381] net: usb: usbnet: adhere to style This satisfies the coding style. Signed-off-by: Oliver Neukum Link: https://patch.msgid.link/20251112102610.281565-1-oneukum@suse.com Signed-off-by: Jakub Kicinski --- drivers/net/usb/usbnet.c | 233 ++++++++++++++++++++------------------- 1 file changed, 120 insertions(+), 113 deletions(-) diff --git a/drivers/net/usb/usbnet.c b/drivers/net/usb/usbnet.c index 3d10cf791c51..1d9faa70ba3b 100644 --- a/drivers/net/usb/usbnet.c +++ b/drivers/net/usb/usbnet.c @@ -142,16 +142,16 @@ int usbnet_get_endpoints(struct usbnet *dev, struct usb_interface *intf) if (alt->desc.bAlternateSetting != 0 || !(dev->driver_info->flags & FLAG_NO_SETINT)) { - tmp = usb_set_interface (dev->udev, alt->desc.bInterfaceNumber, - alt->desc.bAlternateSetting); + tmp = usb_set_interface(dev->udev, alt->desc.bInterfaceNumber, + alt->desc.bAlternateSetting); if (tmp < 0) return tmp; } - dev->in = usb_rcvbulkpipe (dev->udev, - in->desc.bEndpointAddress & USB_ENDPOINT_NUMBER_MASK); - dev->out = usb_sndbulkpipe (dev->udev, - out->desc.bEndpointAddress & USB_ENDPOINT_NUMBER_MASK); + dev->in = usb_rcvbulkpipe(dev->udev, + in->desc.bEndpointAddress & USB_ENDPOINT_NUMBER_MASK); + dev->out = usb_sndbulkpipe(dev->udev, + out->desc.bEndpointAddress & USB_ENDPOINT_NUMBER_MASK); dev->status = status; return 0; } @@ -163,7 +163,7 @@ int usbnet_get_ethernet_addr(struct usbnet *dev, int iMACAddress) int tmp = -1, ret; unsigned char buf [13]; - ret = usb_string(dev->udev, iMACAddress, buf, sizeof buf); + ret = usb_string(dev->udev, iMACAddress, buf, sizeof(buf)); if (ret == 12) tmp = hex2bin(addr, buf, 6); if (tmp < 0) { @@ -215,7 +215,7 @@ static void intr_complete(struct urb *urb) break; } - status = usb_submit_urb (urb, GFP_ATOMIC); + status = usb_submit_urb(urb, GFP_ATOMIC); if (status != 0) netif_err(dev, timer, dev->net, "intr resubmit --> %d\n", status); @@ -231,24 +231,24 @@ static int init_status(struct usbnet *dev, struct usb_interface *intf) if (!dev->driver_info->status) return 0; - pipe = usb_rcvintpipe (dev->udev, - dev->status->desc.bEndpointAddress - & USB_ENDPOINT_NUMBER_MASK); + pipe = usb_rcvintpipe(dev->udev, + dev->status->desc.bEndpointAddress + & USB_ENDPOINT_NUMBER_MASK); maxp = usb_maxpacket(dev->udev, pipe); /* avoid 1 msec chatter: min 8 msec poll rate */ period = max ((int) dev->status->desc.bInterval, (dev->udev->speed == USB_SPEED_HIGH) ? 7 : 3); - buf = kmalloc (maxp, GFP_KERNEL); + buf = kmalloc(maxp, GFP_KERNEL); if (buf) { - dev->interrupt = usb_alloc_urb (0, GFP_KERNEL); + dev->interrupt = usb_alloc_urb(0, GFP_KERNEL); if (!dev->interrupt) { - kfree (buf); + kfree(buf); return -ENOMEM; } else { usb_fill_int_urb(dev->interrupt, dev->udev, pipe, - buf, maxp, intr_complete, dev, period); + buf, maxp, intr_complete, dev, period); dev->interrupt->transfer_flags |= URB_FREE_BUFFER; dev_dbg(&intf->dev, "status ep%din, %d bytes period %d\n", @@ -339,7 +339,7 @@ void usbnet_skb_return(struct usbnet *dev, struct sk_buff *skb) /* only update if unset to allow minidriver rx_fixup override */ if (skb->protocol == 0) - skb->protocol = eth_type_trans (skb, dev->net); + skb->protocol = eth_type_trans(skb, dev->net); flags = u64_stats_update_begin_irqsave(&stats64->syncp); u64_stats_inc(&stats64->rx_packets); @@ -347,8 +347,8 @@ void usbnet_skb_return(struct usbnet *dev, struct sk_buff *skb) u64_stats_update_end_irqrestore(&stats64->syncp, flags); netif_dbg(dev, rx_status, dev->net, "< rx, len %zu, type 0x%x\n", - skb->len + sizeof (struct ethhdr), skb->protocol); - memset (skb->cb, 0, sizeof (struct skb_data)); + skb->len + sizeof(struct ethhdr), skb->protocol); + memset(skb->cb, 0, sizeof(struct skb_data)); if (skb_defer_rx_timestamp(skb)) return; @@ -511,8 +511,8 @@ static int rx_submit(struct usbnet *dev, struct urb *urb, gfp_t flags) skb = __netdev_alloc_skb_ip_align(dev->net, size, flags); if (!skb) { netif_dbg(dev, rx_err, dev->net, "no rx skb\n"); - usbnet_defer_kevent (dev, EVENT_RX_MEMORY); - usb_free_urb (urb); + usbnet_defer_kevent(dev, EVENT_RX_MEMORY); + usb_free_urb(urb); return -ENOMEM; } @@ -521,27 +521,27 @@ static int rx_submit(struct usbnet *dev, struct urb *urb, gfp_t flags) entry->dev = dev; entry->length = 0; - usb_fill_bulk_urb (urb, dev->udev, dev->in, - skb->data, size, rx_complete, skb); + usb_fill_bulk_urb(urb, dev->udev, dev->in, + skb->data, size, rx_complete, skb); - spin_lock_irqsave (&dev->rxq.lock, lockflags); + spin_lock_irqsave(&dev->rxq.lock, lockflags); - if (netif_running (dev->net) && - netif_device_present (dev->net) && + if (netif_running(dev->net) && + netif_device_present(dev->net) && test_bit(EVENT_DEV_OPEN, &dev->flags) && - !test_bit (EVENT_RX_HALT, &dev->flags) && - !test_bit (EVENT_DEV_ASLEEP, &dev->flags) && + !test_bit(EVENT_RX_HALT, &dev->flags) && + !test_bit(EVENT_DEV_ASLEEP, &dev->flags) && !usbnet_going_away(dev)) { - switch (retval = usb_submit_urb (urb, GFP_ATOMIC)) { + switch (retval = usb_submit_urb(urb, GFP_ATOMIC)) { case -EPIPE: - usbnet_defer_kevent (dev, EVENT_RX_HALT); + usbnet_defer_kevent(dev, EVENT_RX_HALT); break; case -ENOMEM: - usbnet_defer_kevent (dev, EVENT_RX_MEMORY); + usbnet_defer_kevent(dev, EVENT_RX_MEMORY); break; case -ENODEV: netif_dbg(dev, ifdown, dev->net, "device gone\n"); - netif_device_detach (dev->net); + netif_device_detach(dev->net); break; case -EHOSTUNREACH: retval = -ENOLINK; @@ -558,10 +558,10 @@ static int rx_submit(struct usbnet *dev, struct urb *urb, gfp_t flags) netif_dbg(dev, ifdown, dev->net, "rx: stopped\n"); retval = -ENOLINK; } - spin_unlock_irqrestore (&dev->rxq.lock, lockflags); + spin_unlock_irqrestore(&dev->rxq.lock, lockflags); if (retval) { - dev_kfree_skb_any (skb); - usb_free_urb (urb); + dev_kfree_skb_any(skb); + usb_free_urb(urb); } return retval; } @@ -572,7 +572,7 @@ static int rx_submit(struct usbnet *dev, struct urb *urb, gfp_t flags) static inline int rx_process(struct usbnet *dev, struct sk_buff *skb) { if (dev->driver_info->rx_fixup && - !dev->driver_info->rx_fixup (dev, skb)) { + !dev->driver_info->rx_fixup(dev, skb)) { /* With RX_ASSEMBLE, rx_fixup() must update counters */ if (!(dev->driver_info->flags & FLAG_RX_ASSEMBLE)) dev->net->stats.rx_errors++; @@ -605,7 +605,7 @@ static void rx_complete(struct urb *urb) int urb_status = urb->status; enum skb_state state; - skb_put (skb, urb->actual_length); + skb_put(skb, urb->actual_length); state = rx_done; entry->urb = NULL; @@ -621,7 +621,7 @@ static void rx_complete(struct urb *urb) */ case -EPIPE: dev->net->stats.rx_errors++; - usbnet_defer_kevent (dev, EVENT_RX_HALT); + usbnet_defer_kevent(dev, EVENT_RX_HALT); fallthrough; /* software-driven interface shutdown */ @@ -639,8 +639,8 @@ static void rx_complete(struct urb *urb) case -ETIME: case -EILSEQ: dev->net->stats.rx_errors++; - if (!timer_pending (&dev->delay)) { - mod_timer (&dev->delay, jiffies + THROTTLE_JIFFIES); + if (!timer_pending(&dev->delay)) { + mod_timer(&dev->delay, jiffies + THROTTLE_JIFFIES); netif_dbg(dev, link, dev->net, "rx throttle %d\n", urb_status); } @@ -676,14 +676,14 @@ static void rx_complete(struct urb *urb) state = defer_bh(dev, skb, &dev->rxq, state); if (urb) { - if (netif_running (dev->net) && - !test_bit (EVENT_RX_HALT, &dev->flags) && + if (netif_running(dev->net) && + !test_bit(EVENT_RX_HALT, &dev->flags) && state != unlink_start) { - rx_submit (dev, urb, GFP_ATOMIC); + rx_submit(dev, urb, GFP_ATOMIC); usb_mark_last_busy(dev->udev); return; } - usb_free_urb (urb); + usb_free_urb(urb); } netif_dbg(dev, rx_err, dev->net, "no read resubmitted\n"); } @@ -761,7 +761,7 @@ static int unlink_urbs(struct usbnet *dev, struct sk_buff_head *q) spin_unlock_irqrestore(&q->lock, flags); // during some PM-driven resume scenarios, // these (async) unlinks complete immediately - retval = usb_unlink_urb (urb); + retval = usb_unlink_urb(urb); if (retval != -EINPROGRESS && retval != 0) netdev_dbg(dev->net, "unlink urb err, %d\n", retval); else @@ -769,7 +769,7 @@ static int unlink_urbs(struct usbnet *dev, struct sk_buff_head *q) usb_put_urb(urb); spin_lock_irqsave(&q->lock, flags); } - spin_unlock_irqrestore (&q->lock, flags); + spin_unlock_irqrestore(&q->lock, flags); return count; } @@ -830,7 +830,7 @@ int usbnet_stop(struct net_device *net) int retval, pm, mpn; clear_bit(EVENT_DEV_OPEN, &dev->flags); - netif_stop_queue (net); + netif_stop_queue(net); netdev_reset_queue(net); netif_info(dev, ifdown, dev->net, @@ -910,23 +910,29 @@ int usbnet_open(struct net_device *net) } // put into "known safe" state - if (info->reset && (retval = info->reset (dev)) < 0) { - netif_info(dev, ifup, dev->net, - "open reset fail (%d) usbnet usb-%s-%s, %s\n", - retval, - dev->udev->bus->bus_name, - dev->udev->devpath, - info->description); - goto done; + if (info->reset) { + retval = info->reset(dev); + if (retval < 0) { + netif_info(dev, ifup, dev->net, + "open reset fail (%d) usbnet usb-%s-%s, %s\n", + retval, + dev->udev->bus->bus_name, + dev->udev->devpath, + info->description); + goto done; + } } /* hard_mtu or rx_urb_size may change in reset() */ usbnet_update_max_qlen(dev); // insist peer be connected - if (info->check_connect && (retval = info->check_connect (dev)) < 0) { - netif_err(dev, ifup, dev->net, "can't open; %d\n", retval); - goto done; + if (info->check_connect) { + retval = info->check_connect(dev); + if (retval < 0) { + netif_err(dev, ifup, dev->net, "can't open; %d\n", retval); + goto done; + } } /* start any status interrupt transfer */ @@ -1056,7 +1062,7 @@ u32 usbnet_get_link(struct net_device *net) /* If a check_connect is defined, return its result */ if (dev->driver_info->check_connect) - return dev->driver_info->check_connect (dev) == 0; + return dev->driver_info->check_connect(dev) == 0; /* if the device has mii operations, use those */ if (dev->mii.mdio_read) @@ -1085,7 +1091,7 @@ void usbnet_get_drvinfo(struct net_device *net, struct ethtool_drvinfo *info) strscpy(info->driver, dev->driver_name, sizeof(info->driver)); strscpy(info->fw_version, dev->driver_info->description, sizeof(info->fw_version)); - usb_make_path (dev->udev, info->bus_info, sizeof info->bus_info); + usb_make_path(dev->udev, info->bus_info, sizeof(info->bus_info)); } EXPORT_SYMBOL_GPL(usbnet_get_drvinfo); @@ -1175,64 +1181,64 @@ usbnet_deferred_kevent(struct work_struct *work) int status; /* usb_clear_halt() needs a thread context */ - if (test_bit (EVENT_TX_HALT, &dev->flags)) { - unlink_urbs (dev, &dev->txq); + if (test_bit(EVENT_TX_HALT, &dev->flags)) { + unlink_urbs(dev, &dev->txq); status = usb_autopm_get_interface(dev->intf); if (status < 0) goto fail_pipe; - status = usb_clear_halt (dev->udev, dev->out); + status = usb_clear_halt(dev->udev, dev->out); usb_autopm_put_interface(dev->intf); if (status < 0 && status != -EPIPE && status != -ESHUTDOWN) { - if (netif_msg_tx_err (dev)) + if (netif_msg_tx_err(dev)) fail_pipe: netdev_err(dev->net, "can't clear tx halt, status %d\n", status); } else { - clear_bit (EVENT_TX_HALT, &dev->flags); + clear_bit(EVENT_TX_HALT, &dev->flags); if (status != -ESHUTDOWN) - netif_wake_queue (dev->net); + netif_wake_queue(dev->net); } } - if (test_bit (EVENT_RX_HALT, &dev->flags)) { - unlink_urbs (dev, &dev->rxq); + if (test_bit(EVENT_RX_HALT, &dev->flags)) { + unlink_urbs(dev, &dev->rxq); status = usb_autopm_get_interface(dev->intf); if (status < 0) goto fail_halt; - status = usb_clear_halt (dev->udev, dev->in); + status = usb_clear_halt(dev->udev, dev->in); usb_autopm_put_interface(dev->intf); if (status < 0 && status != -EPIPE && status != -ESHUTDOWN) { - if (netif_msg_rx_err (dev)) + if (netif_msg_rx_err(dev)) fail_halt: netdev_err(dev->net, "can't clear rx halt, status %d\n", status); } else { - clear_bit (EVENT_RX_HALT, &dev->flags); + clear_bit(EVENT_RX_HALT, &dev->flags); if (!usbnet_going_away(dev)) queue_work(system_bh_wq, &dev->bh_work); } } /* work could resubmit itself forever if memory is tight */ - if (test_bit (EVENT_RX_MEMORY, &dev->flags)) { + if (test_bit(EVENT_RX_MEMORY, &dev->flags)) { struct urb *urb = NULL; int resched = 1; - if (netif_running (dev->net)) - urb = usb_alloc_urb (0, GFP_KERNEL); + if (netif_running(dev->net)) + urb = usb_alloc_urb(0, GFP_KERNEL); else - clear_bit (EVENT_RX_MEMORY, &dev->flags); + clear_bit(EVENT_RX_MEMORY, &dev->flags); if (urb != NULL) { - clear_bit (EVENT_RX_MEMORY, &dev->flags); + clear_bit(EVENT_RX_MEMORY, &dev->flags); status = usb_autopm_get_interface(dev->intf); if (status < 0) { usb_free_urb(urb); goto fail_lowmem; } - if (rx_submit (dev, urb, GFP_KERNEL) == -ENOLINK) + if (rx_submit(dev, urb, GFP_KERNEL) == -ENOLINK) resched = 0; usb_autopm_put_interface(dev->intf); fail_lowmem: @@ -1246,7 +1252,7 @@ usbnet_deferred_kevent(struct work_struct *work) const struct driver_info *info = dev->driver_info; int retval = 0; - clear_bit (EVENT_LINK_RESET, &dev->flags); + clear_bit(EVENT_LINK_RESET, &dev->flags); status = usb_autopm_get_interface(dev->intf); if (status < 0) goto skip_reset; @@ -1266,10 +1272,10 @@ usbnet_deferred_kevent(struct work_struct *work) __handle_link_change(dev); } - if (test_bit (EVENT_LINK_CHANGE, &dev->flags)) + if (test_bit(EVENT_LINK_CHANGE, &dev->flags)) __handle_link_change(dev); - if (test_bit (EVENT_SET_RX_MODE, &dev->flags)) + if (test_bit(EVENT_SET_RX_MODE, &dev->flags)) __handle_set_rx_mode(dev); @@ -1298,7 +1304,7 @@ static void tx_complete(struct urb *urb) switch (urb->status) { case -EPIPE: - usbnet_defer_kevent (dev, EVENT_TX_HALT); + usbnet_defer_kevent(dev, EVENT_TX_HALT); break; /* software-driven interface shutdown */ @@ -1313,13 +1319,13 @@ static void tx_complete(struct urb *urb) case -ETIME: case -EILSEQ: usb_mark_last_busy(dev->udev); - if (!timer_pending (&dev->delay)) { - mod_timer (&dev->delay, - jiffies + THROTTLE_JIFFIES); + if (!timer_pending(&dev->delay)) { + mod_timer(&dev->delay, + jiffies + THROTTLE_JIFFIES); netif_dbg(dev, link, dev->net, "tx throttle %d\n", urb->status); } - netif_stop_queue (dev->net); + netif_stop_queue(dev->net); break; default: netif_dbg(dev, tx_err, dev->net, @@ -1338,7 +1344,7 @@ void usbnet_tx_timeout(struct net_device *net, unsigned int txqueue) { struct usbnet *dev = netdev_priv(net); - unlink_urbs (dev, &dev->txq); + unlink_urbs(dev, &dev->txq); queue_work(system_bh_wq, &dev->bh_work); /* this needs to be handled individually because the generic layer * doesn't know what is sufficient and could not restore private @@ -1400,7 +1406,7 @@ netdev_tx_t usbnet_start_xmit(struct sk_buff *skb, struct net_device *net) // some devices want funky USB-level framing, for // win32 driver (usually) and/or hardware quirks if (info->tx_fixup) { - skb = info->tx_fixup (dev, skb, GFP_ATOMIC); + skb = info->tx_fixup(dev, skb, GFP_ATOMIC); if (!skb) { /* packet collected; minidriver waiting for more */ if (info->flags & FLAG_MULTI_PACKET) @@ -1410,7 +1416,8 @@ netdev_tx_t usbnet_start_xmit(struct sk_buff *skb, struct net_device *net) } } - if (!(urb = usb_alloc_urb (0, GFP_ATOMIC))) { + urb = usb_alloc_urb(0, GFP_ATOMIC); + if (!urb) { netif_dbg(dev, tx_err, dev->net, "no urb\n"); goto drop; } @@ -1419,8 +1426,8 @@ netdev_tx_t usbnet_start_xmit(struct sk_buff *skb, struct net_device *net) entry->urb = urb; entry->dev = dev; - usb_fill_bulk_urb (urb, dev->udev, dev->out, - skb->data, skb->len, tx_complete, skb); + usb_fill_bulk_urb(urb, dev->udev, dev->out, + skb->data, skb->len, tx_complete, skb); if (dev->can_dma_sg) { if (build_dma_sg(skb, urb) < 0) goto drop; @@ -1490,8 +1497,8 @@ netdev_tx_t usbnet_start_xmit(struct sk_buff *skb, struct net_device *net) switch ((retval = usb_submit_urb (urb, GFP_ATOMIC))) { case -EPIPE: - netif_stop_queue (net); - usbnet_defer_kevent (dev, EVENT_TX_HALT); + netif_stop_queue(net); + usbnet_defer_kevent(dev, EVENT_TX_HALT); usb_autopm_put_interface_async(dev->intf); break; default: @@ -1506,7 +1513,7 @@ netdev_tx_t usbnet_start_xmit(struct sk_buff *skb, struct net_device *net) if (dev->txq.qlen >= TX_QLEN (dev)) netif_stop_queue (net); } - spin_unlock_irqrestore (&dev->txq.lock, flags); + spin_unlock_irqrestore(&dev->txq.lock, flags); if (retval) { netif_dbg(dev, tx_err, dev->net, "drop, code %d\n", retval); @@ -1514,7 +1521,7 @@ netdev_tx_t usbnet_start_xmit(struct sk_buff *skb, struct net_device *net) dev->net->stats.tx_dropped++; not_drop: if (skb) - dev_kfree_skb_any (skb); + dev_kfree_skb_any(skb); if (urb) { kfree(urb->sg); usb_free_urb(urb); @@ -1625,7 +1632,7 @@ static void usbnet_bh(struct timer_list *t) queue_work(system_bh_wq, &dev->bh_work); } if (dev->txq.qlen < TX_QLEN (dev)) - netif_wake_queue (dev->net); + netif_wake_queue(dev->net); } } @@ -1658,7 +1665,7 @@ void usbnet_disconnect(struct usb_interface *intf) return; usbnet_mark_going_away(dev); - xdev = interface_to_usbdev (intf); + xdev = interface_to_usbdev(intf); netif_info(dev, probe, dev->net, "unregister '%s' usb-%s-%s, %s\n", intf->dev.driver->name, @@ -1666,7 +1673,7 @@ void usbnet_disconnect(struct usb_interface *intf) dev->driver_info->description); net = dev->net; - unregister_netdev (net); + unregister_netdev(net); cancel_work_sync(&dev->kevent); @@ -1737,7 +1744,7 @@ usbnet_probe(struct usb_interface *udev, const struct usb_device_id *prod) dev_dbg (&udev->dev, "blacklisted by %s\n", name); return -ENODEV; } - xdev = interface_to_usbdev (udev); + xdev = interface_to_usbdev(udev); interface = udev->cur_altsetting; status = -ENOMEM; @@ -1767,10 +1774,10 @@ usbnet_probe(struct usb_interface *udev, const struct usb_device_id *prod) skb_queue_head_init(&dev->rxq_pause); spin_lock_init(&dev->bql_spinlock); INIT_WORK(&dev->bh_work, usbnet_bh_work); - INIT_WORK (&dev->kevent, usbnet_deferred_kevent); + INIT_WORK(&dev->kevent, usbnet_deferred_kevent); init_usb_anchor(&dev->deferred); timer_setup(&dev->delay, usbnet_bh, 0); - mutex_init (&dev->phy_mutex); + mutex_init(&dev->phy_mutex); mutex_init(&dev->interrupt_mutex); dev->interrupt_count = 0; @@ -1792,7 +1799,7 @@ usbnet_probe(struct usb_interface *udev, const struct usb_device_id *prod) // allow device-specific bind/init procedures // NOTE net->name still not usable ... if (info->bind) { - status = info->bind (dev, udev); + status = info->bind(dev, udev); if (status < 0) goto out1; @@ -1817,18 +1824,18 @@ usbnet_probe(struct usb_interface *udev, const struct usb_device_id *prod) if (net->mtu > (dev->hard_mtu - net->hard_header_len)) net->mtu = dev->hard_mtu - net->hard_header_len; } else if (!info->in || !info->out) - status = usbnet_get_endpoints (dev, udev); + status = usbnet_get_endpoints(dev, udev); else { u8 ep_addrs[3] = { info->in + USB_DIR_IN, info->out + USB_DIR_OUT, 0 }; - dev->in = usb_rcvbulkpipe (xdev, info->in); - dev->out = usb_sndbulkpipe (xdev, info->out); + dev->in = usb_rcvbulkpipe(xdev, info->in); + dev->out = usb_sndbulkpipe(xdev, info->out); if (!(info->flags & FLAG_NO_SETINT)) - status = usb_set_interface (xdev, - interface->desc.bInterfaceNumber, - interface->desc.bAlternateSetting); + status = usb_set_interface(xdev, + interface->desc.bInterfaceNumber, + interface->desc.bAlternateSetting); else status = 0; @@ -1836,7 +1843,7 @@ usbnet_probe(struct usb_interface *udev, const struct usb_device_id *prod) status = -EINVAL; } if (status >= 0 && dev->status) - status = init_status (dev, udev); + status = init_status(dev, udev); if (status < 0) goto out3; @@ -1870,7 +1877,7 @@ usbnet_probe(struct usb_interface *udev, const struct usb_device_id *prod) } } - status = register_netdev (net); + status = register_netdev(net); if (status) goto out5; netif_info(dev, probe, dev->net, @@ -1881,9 +1888,9 @@ usbnet_probe(struct usb_interface *udev, const struct usb_device_id *prod) net->dev_addr); // ok, it's ready to go. - usb_set_intfdata (udev, dev); + usb_set_intfdata(udev, dev); - netif_device_attach (net); + netif_device_attach(net); if (dev->driver_info->flags & FLAG_LINK_INTR) usbnet_link_change(dev, 0, 0); @@ -1896,7 +1903,7 @@ usbnet_probe(struct usb_interface *udev, const struct usb_device_id *prod) usb_free_urb(dev->interrupt); out3: if (info->unbind) - info->unbind (dev, udev); + info->unbind(dev, udev); out1: /* subdrivers must undo all they did in bind() if they * fail it, but we may fail later and a deferred kevent @@ -1938,7 +1945,7 @@ int usbnet_suspend(struct usb_interface *intf, pm_message_t message) * accelerate emptying of the rx and queues, to avoid * having everything error out. */ - netif_device_detach (dev->net); + netif_device_detach(dev->net); usbnet_terminate_urbs(dev); __usbnet_status_stop_force(dev); @@ -1946,7 +1953,7 @@ int usbnet_suspend(struct usb_interface *intf, pm_message_t message) * reattach so runtime management can use and * wake the device */ - netif_device_attach (dev->net); + netif_device_attach(dev->net); } return 0; } From c31783c2b5ae2399388b349f48fa89818b66f6d8 Mon Sep 17 00:00:00 2001 From: Lad Prabhakar Date: Wed, 12 Nov 2025 13:57:12 +0000 Subject: [PATCH 0792/1381] net: phy: mscc: Simplify LED mode update using phy_modify() The vsc85xx_led_cntl_set() function currently performs a manual read-modify-write sequence protected by the PHY lock to update the LED mode register (MSCC_PHY_LED_MODE_SEL). Replace this sequence with a call to phy_modify(), which already handles read-modify-write operations with proper locking inside the PHY core. Signed-off-by: Lad Prabhakar Reviewed-by: Andrew Lunn Reviewed-by: Russell King (Oracle) Link: https://patch.msgid.link/20251112135715.1017117-2-prabhakar.mahadev-lad.rj@bp.renesas.com Signed-off-by: Jakub Kicinski --- drivers/net/phy/mscc/mscc_main.c | 13 +++---------- 1 file changed, 3 insertions(+), 10 deletions(-) diff --git a/drivers/net/phy/mscc/mscc_main.c b/drivers/net/phy/mscc/mscc_main.c index 8678ebf89cca..032050ec0bc9 100644 --- a/drivers/net/phy/mscc/mscc_main.c +++ b/drivers/net/phy/mscc/mscc_main.c @@ -177,17 +177,10 @@ static int vsc85xx_led_cntl_set(struct phy_device *phydev, u8 led_num, u8 mode) { - int rc; - u16 reg_val; + u16 mask = LED_MODE_SEL_MASK(led_num); + u16 val = LED_MODE_SEL(led_num, mode); - mutex_lock(&phydev->lock); - reg_val = phy_read(phydev, MSCC_PHY_LED_MODE_SEL); - reg_val &= ~LED_MODE_SEL_MASK(led_num); - reg_val |= LED_MODE_SEL(led_num, (u16)mode); - rc = phy_write(phydev, MSCC_PHY_LED_MODE_SEL, reg_val); - mutex_unlock(&phydev->lock); - - return rc; + return phy_modify(phydev, MSCC_PHY_LED_MODE_SEL, mask, val); } static int vsc85xx_mdix_get(struct phy_device *phydev, u8 *mdix) From 217eb2d60f564681cf5d57e3571eb52e3ebfa465 Mon Sep 17 00:00:00 2001 From: Lad Prabhakar Date: Wed, 12 Nov 2025 13:57:13 +0000 Subject: [PATCH 0793/1381] net: phy: mscc: Consolidate probe functions into a common helper Unify the probe implementations of the VSC85xx PHY family into a single vsc85xx_probe_common() helper. The existing probe functions for the vsc85xx, vsc8514, vsc8574, and vsc8584 variants contained almost identical initialization logic, differing only in configuration parameters such as the number of LEDs, supported LED modes, hardware statistics, and PTP support. Introduce a vsc85xx_probe_config structure to describe the per-variant parameters, and move all common setup code into the shared helper. Each variant's probe function now defines a constant configuration instance and calls vsc85xx_probe_common(). Also mark the default LED mode array parameter as const to match its usage. Signed-off-by: Lad Prabhakar Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20251112135715.1017117-3-prabhakar.mahadev-lad.rj@bp.renesas.com Signed-off-by: Jakub Kicinski --- drivers/net/phy/mscc/mscc_main.c | 213 ++++++++++++++++--------------- 1 file changed, 112 insertions(+), 101 deletions(-) diff --git a/drivers/net/phy/mscc/mscc_main.c b/drivers/net/phy/mscc/mscc_main.c index 032050ec0bc9..052e68d1cd97 100644 --- a/drivers/net/phy/mscc/mscc_main.c +++ b/drivers/net/phy/mscc/mscc_main.c @@ -22,6 +22,24 @@ #include "mscc_serdes.h" #include "mscc.h" +struct vsc85xx_probe_config { + const struct vsc85xx_hw_stat *hw_stats; + size_t shared_size; + size_t nstats; + u16 supp_led_modes; + u8 nleds; + bool check_rate_magic; + bool use_package; + bool has_ptp; +}; + +static const u32 vsc85xx_default_led_modes_4[] = { + VSC8531_LINK_1000_ACTIVITY, + VSC8531_LINK_100_ACTIVITY, + VSC8531_LINK_ACTIVITY, + VSC8531_DUPLEX_COLLISION +}; + static const struct vsc85xx_hw_stat vsc85xx_hw_stats[] = { { .string = "phy_receive_errors", @@ -436,7 +454,7 @@ static int vsc85xx_dt_led_mode_get(struct phy_device *phydev, #endif /* CONFIG_OF_MDIO */ static int vsc85xx_dt_led_modes_get(struct phy_device *phydev, - u32 *default_mode) + const u32 *default_mode) { struct vsc8531_private *priv = phydev->priv; char led_dt_prop[28]; @@ -2211,70 +2229,11 @@ static int vsc85xx_config_inband(struct phy_device *phydev, unsigned int modes) reg_val); } -static int vsc8514_probe(struct phy_device *phydev) +static int vsc85xx_probe_common(struct phy_device *phydev, + const struct vsc85xx_probe_config *cfg, + const u32 *default_led_mode) { struct vsc8531_private *vsc8531; - u32 default_mode[4] = {VSC8531_LINK_1000_ACTIVITY, - VSC8531_LINK_100_ACTIVITY, VSC8531_LINK_ACTIVITY, - VSC8531_DUPLEX_COLLISION}; - - vsc8531 = devm_kzalloc(&phydev->mdio.dev, sizeof(*vsc8531), GFP_KERNEL); - if (!vsc8531) - return -ENOMEM; - - phydev->priv = vsc8531; - - vsc8584_get_base_addr(phydev); - devm_phy_package_join(&phydev->mdio.dev, phydev, - vsc8531->base_addr, 0); - - vsc8531->nleds = 4; - vsc8531->supp_led_modes = VSC85XX_SUPP_LED_MODES; - vsc8531->hw_stats = vsc85xx_hw_stats; - vsc8531->nstats = ARRAY_SIZE(vsc85xx_hw_stats); - vsc8531->stats = devm_kcalloc(&phydev->mdio.dev, vsc8531->nstats, - sizeof(u64), GFP_KERNEL); - if (!vsc8531->stats) - return -ENOMEM; - - return vsc85xx_dt_led_modes_get(phydev, default_mode); -} - -static int vsc8574_probe(struct phy_device *phydev) -{ - struct vsc8531_private *vsc8531; - u32 default_mode[4] = {VSC8531_LINK_1000_ACTIVITY, - VSC8531_LINK_100_ACTIVITY, VSC8531_LINK_ACTIVITY, - VSC8531_DUPLEX_COLLISION}; - - vsc8531 = devm_kzalloc(&phydev->mdio.dev, sizeof(*vsc8531), GFP_KERNEL); - if (!vsc8531) - return -ENOMEM; - - phydev->priv = vsc8531; - - vsc8584_get_base_addr(phydev); - devm_phy_package_join(&phydev->mdio.dev, phydev, - vsc8531->base_addr, 0); - - vsc8531->nleds = 4; - vsc8531->supp_led_modes = VSC8584_SUPP_LED_MODES; - vsc8531->hw_stats = vsc8584_hw_stats; - vsc8531->nstats = ARRAY_SIZE(vsc8584_hw_stats); - vsc8531->stats = devm_kcalloc(&phydev->mdio.dev, vsc8531->nstats, - sizeof(u64), GFP_KERNEL); - if (!vsc8531->stats) - return -ENOMEM; - - return vsc85xx_dt_led_modes_get(phydev, default_mode); -} - -static int vsc8584_probe(struct phy_device *phydev) -{ - struct vsc8531_private *vsc8531; - u32 default_mode[4] = {VSC8531_LINK_1000_ACTIVITY, - VSC8531_LINK_100_ACTIVITY, VSC8531_LINK_ACTIVITY, - VSC8531_DUPLEX_COLLISION}; int ret; vsc8531 = devm_kzalloc(&phydev->mdio.dev, sizeof(*vsc8531), GFP_KERNEL); @@ -2283,60 +2242,112 @@ static int vsc8584_probe(struct phy_device *phydev) phydev->priv = vsc8531; - vsc8584_get_base_addr(phydev); - devm_phy_package_join(&phydev->mdio.dev, phydev, vsc8531->base_addr, - sizeof(struct vsc85xx_shared_private)); + /* Check rate magic if needed (only for non-package PHYs) */ + if (cfg->check_rate_magic) { + ret = vsc85xx_edge_rate_magic_get(phydev); + if (ret < 0) + return ret; - vsc8531->nleds = 4; - vsc8531->supp_led_modes = VSC8584_SUPP_LED_MODES; - vsc8531->hw_stats = vsc8584_hw_stats; - vsc8531->nstats = ARRAY_SIZE(vsc8584_hw_stats); + vsc8531->rate_magic = ret; + } + + /* Set up package if needed */ + if (cfg->use_package) { + vsc8584_get_base_addr(phydev); + devm_phy_package_join(&phydev->mdio.dev, phydev, + vsc8531->base_addr, cfg->shared_size); + } + + /* Configure LED settings */ + vsc8531->nleds = cfg->nleds; + vsc8531->supp_led_modes = cfg->supp_led_modes; + + /* Configure hardware stats */ + vsc8531->hw_stats = cfg->hw_stats; + vsc8531->nstats = cfg->nstats; vsc8531->stats = devm_kcalloc(&phydev->mdio.dev, vsc8531->nstats, sizeof(u64), GFP_KERNEL); if (!vsc8531->stats) return -ENOMEM; - if (phy_package_probe_once(phydev)) { - ret = vsc8584_ptp_probe_once(phydev); + /* PTP setup for VSC8584 */ + if (cfg->has_ptp) { + if (phy_package_probe_once(phydev)) { + ret = vsc8584_ptp_probe_once(phydev); + if (ret) + return ret; + } + + ret = vsc8584_ptp_probe(phydev); if (ret) return ret; } - ret = vsc8584_ptp_probe(phydev); - if (ret) - return ret; + /* Parse LED modes from device tree */ + return vsc85xx_dt_led_modes_get(phydev, default_led_mode); +} - return vsc85xx_dt_led_modes_get(phydev, default_mode); +static int vsc8514_probe(struct phy_device *phydev) +{ + static const struct vsc85xx_probe_config vsc8514_cfg = { + .nleds = 4, + .supp_led_modes = VSC85XX_SUPP_LED_MODES, + .hw_stats = vsc85xx_hw_stats, + .nstats = ARRAY_SIZE(vsc85xx_hw_stats), + .use_package = true, + .shared_size = 0, + .has_ptp = false, + .check_rate_magic = false, + }; + + return vsc85xx_probe_common(phydev, &vsc8514_cfg, vsc85xx_default_led_modes_4); +} + +static int vsc8574_probe(struct phy_device *phydev) +{ + static const struct vsc85xx_probe_config vsc8574_cfg = { + .nleds = 4, + .supp_led_modes = VSC8584_SUPP_LED_MODES, + .hw_stats = vsc8584_hw_stats, + .nstats = ARRAY_SIZE(vsc8584_hw_stats), + .use_package = true, + .shared_size = 0, + .has_ptp = false, + .check_rate_magic = false, + }; + + return vsc85xx_probe_common(phydev, &vsc8574_cfg, vsc85xx_default_led_modes_4); +} + +static int vsc8584_probe(struct phy_device *phydev) +{ + static const struct vsc85xx_probe_config vsc8584_cfg = { + .nleds = 4, + .supp_led_modes = VSC8584_SUPP_LED_MODES, + .hw_stats = vsc8584_hw_stats, + .nstats = ARRAY_SIZE(vsc8584_hw_stats), + .use_package = true, + .shared_size = sizeof(struct vsc85xx_shared_private), + .has_ptp = true, + .check_rate_magic = false, + }; + + return vsc85xx_probe_common(phydev, &vsc8584_cfg, vsc85xx_default_led_modes_4); } static int vsc85xx_probe(struct phy_device *phydev) { - struct vsc8531_private *vsc8531; - int rate_magic; - u32 default_mode[2] = {VSC8531_LINK_1000_ACTIVITY, - VSC8531_LINK_100_ACTIVITY}; + static const struct vsc85xx_probe_config vsc85xx_cfg = { + .nleds = 2, + .supp_led_modes = VSC85XX_SUPP_LED_MODES, + .hw_stats = vsc85xx_hw_stats, + .nstats = ARRAY_SIZE(vsc85xx_hw_stats), + .use_package = false, + .has_ptp = false, + .check_rate_magic = true, + }; - rate_magic = vsc85xx_edge_rate_magic_get(phydev); - if (rate_magic < 0) - return rate_magic; - - vsc8531 = devm_kzalloc(&phydev->mdio.dev, sizeof(*vsc8531), GFP_KERNEL); - if (!vsc8531) - return -ENOMEM; - - phydev->priv = vsc8531; - - vsc8531->rate_magic = rate_magic; - vsc8531->nleds = 2; - vsc8531->supp_led_modes = VSC85XX_SUPP_LED_MODES; - vsc8531->hw_stats = vsc85xx_hw_stats; - vsc8531->nstats = ARRAY_SIZE(vsc85xx_hw_stats); - vsc8531->stats = devm_kcalloc(&phydev->mdio.dev, vsc8531->nstats, - sizeof(u64), GFP_KERNEL); - if (!vsc8531->stats) - return -ENOMEM; - - return vsc85xx_dt_led_modes_get(phydev, default_mode); + return vsc85xx_probe_common(phydev, &vsc85xx_cfg, vsc85xx_default_led_modes_4); } static void vsc85xx_remove(struct phy_device *phydev) From eb47c5c4888d86d4f2bfd1513f4cf79a920e65b5 Mon Sep 17 00:00:00 2001 From: Lad Prabhakar Date: Wed, 12 Nov 2025 13:57:14 +0000 Subject: [PATCH 0794/1381] net: phy: mscc: Add support for PHY LED control Add support for the PHY LED controller in the MSCC VSC85xx driver. The implementation provides LED brightness and hardware control through the LED subsystem and integrates with the standard 'netdev' trigger. Introduce new register definitions for the LED behavior register (MSCC_PHY_LED_BEHAVIOR = 30) and the LED combine disable bits, which control whether LEDs indicate link-only or combined link and activity status. Implement a helper, vsc8541_led_combine_disable_set(), to update these bits safely using phy_modify(). Add support for LED brightness control and hardware mode configuration. The new callbacks implement the standard LED class operations, allowing user control through sysfs. The brightness control maps to PHY LED force on/off modes. The hardware control get and set functions translate between the PHY-specific LED mode encodings and the LED subsystem TRIGGER_NETDEV_* rules. The combine feature is managed automatically based on the selected rules. When both RX and TX activity are disabled, the combine feature is turned off, causing LEDs to indicate link-only status. When either RX or TX activity is enabled, the combine feature remains active and LEDs indicate combined link and activity. Register the LED callbacks for all VSC85xx PHY variants so that the LED subsystem can manage their indicators consistently. Existing device tree LED configuration and default behavior are preserved. Signed-off-by: Lad Prabhakar Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20251112135715.1017117-4-prabhakar.mahadev-lad.rj@bp.renesas.com Signed-off-by: Jakub Kicinski --- drivers/net/phy/mscc/mscc.h | 4 + drivers/net/phy/mscc/mscc_main.c | 246 +++++++++++++++++++++++++++++++ 2 files changed, 250 insertions(+) diff --git a/drivers/net/phy/mscc/mscc.h b/drivers/net/phy/mscc/mscc.h index 2eef5956b9cc..65c9d7bd9315 100644 --- a/drivers/net/phy/mscc/mscc.h +++ b/drivers/net/phy/mscc/mscc.h @@ -85,6 +85,10 @@ enum rgmii_clock_delay { #define LED_MODE_SEL_MASK(x) (GENMASK(3, 0) << LED_MODE_SEL_POS(x)) #define LED_MODE_SEL(x, mode) (((mode) << LED_MODE_SEL_POS(x)) & LED_MODE_SEL_MASK(x)) +#define MSCC_PHY_LED_BEHAVIOR 30 +#define LED_COMBINE_DIS_MASK(x) BIT(x) +#define LED_COMBINE_DIS(x, dis) (((dis) ? 1 : 0) << (x)) + #define MSCC_EXT_PAGE_CSR_CNTL_17 17 #define MSCC_EXT_PAGE_CSR_CNTL_18 18 diff --git a/drivers/net/phy/mscc/mscc_main.c b/drivers/net/phy/mscc/mscc_main.c index 052e68d1cd97..21fcaf07bc6d 100644 --- a/drivers/net/phy/mscc/mscc_main.c +++ b/drivers/net/phy/mscc/mscc_main.c @@ -201,6 +201,15 @@ static int vsc85xx_led_cntl_set(struct phy_device *phydev, return phy_modify(phydev, MSCC_PHY_LED_MODE_SEL, mask, val); } +static int vsc85xx_led_combine_disable_set(struct phy_device *phydev, + u8 led_num, bool combine_disable) +{ + u16 val = LED_COMBINE_DIS(led_num, combine_disable); + u16 mask = LED_COMBINE_DIS_MASK(led_num); + + return phy_modify(phydev, MSCC_PHY_LED_BEHAVIOR, mask, val); +} + static int vsc85xx_mdix_get(struct phy_device *phydev, u8 *mdix) { u16 reg_val; @@ -2234,6 +2243,7 @@ static int vsc85xx_probe_common(struct phy_device *phydev, const u32 *default_led_mode) { struct vsc8531_private *vsc8531; + struct device_node *np; int ret; vsc8531 = devm_kzalloc(&phydev->mdio.dev, sizeof(*vsc8531), GFP_KERNEL); @@ -2283,10 +2293,186 @@ static int vsc85xx_probe_common(struct phy_device *phydev, return ret; } + /* + * Check for LED configuration in device tree if available + * or fall back to default `vsc8531,led-x-mode` DT properties. + */ + np = of_get_child_by_name(phydev->mdio.dev.of_node, "leds"); + if (np) { + of_node_put(np); + + /* Force to defaults */ + for (unsigned int i = 0; i < vsc8531->nleds; i++) + vsc8531->leds_mode[i] = default_led_mode[i]; + + return 0; + } + /* Parse LED modes from device tree */ return vsc85xx_dt_led_modes_get(phydev, default_led_mode); } +static int vsc85xx_led_brightness_set(struct phy_device *phydev, + u8 index, enum led_brightness value) +{ + struct vsc8531_private *vsc8531 = phydev->priv; + + if (index >= vsc8531->nleds) + return -EINVAL; + + return vsc85xx_led_cntl_set(phydev, index, value == LED_OFF ? + VSC8531_FORCE_LED_OFF : VSC8531_FORCE_LED_ON); +} + +static int vsc85xx_led_hw_is_supported(struct phy_device *phydev, u8 index, + unsigned long rules) +{ + static const unsigned long supported = BIT(TRIGGER_NETDEV_LINK_1000) | + BIT(TRIGGER_NETDEV_LINK_100) | + BIT(TRIGGER_NETDEV_LINK_10) | + BIT(TRIGGER_NETDEV_LINK) | + BIT(TRIGGER_NETDEV_RX) | + BIT(TRIGGER_NETDEV_TX); + struct vsc8531_private *vsc8531 = phydev->priv; + + if (index >= vsc8531->nleds) + return -EINVAL; + + if (rules & ~supported) + return -EOPNOTSUPP; + + return 0; +} + +static int vsc85xx_led_hw_control_get(struct phy_device *phydev, u8 index, + unsigned long *rules) +{ + struct vsc8531_private *vsc8531 = phydev->priv; + u8 mode, behavior; + int rc; + + if (index >= vsc8531->nleds) + return -EINVAL; + + rc = phy_read(phydev, MSCC_PHY_LED_MODE_SEL); + if (rc < 0) + return rc; + mode = (rc & LED_MODE_SEL_MASK(index)) >> LED_MODE_SEL_POS(index); + + rc = phy_read(phydev, MSCC_PHY_LED_BEHAVIOR); + if (rc < 0) + return rc; + behavior = (rc & LED_COMBINE_DIS_MASK(index)) >> index; + + switch (mode) { + case VSC8531_LINK_ACTIVITY: + case VSC8531_ACTIVITY: + *rules = BIT(TRIGGER_NETDEV_LINK); + break; + + case VSC8531_LINK_1000_ACTIVITY: + *rules = BIT(TRIGGER_NETDEV_LINK_1000) | + BIT(TRIGGER_NETDEV_LINK); + break; + + case VSC8531_LINK_100_ACTIVITY: + *rules = BIT(TRIGGER_NETDEV_LINK_100) | + BIT(TRIGGER_NETDEV_LINK); + break; + + case VSC8531_LINK_10_ACTIVITY: + *rules = BIT(TRIGGER_NETDEV_LINK_10) | + BIT(TRIGGER_NETDEV_LINK); + break; + + case VSC8531_LINK_100_1000_ACTIVITY: + *rules = BIT(TRIGGER_NETDEV_LINK_1000) | + BIT(TRIGGER_NETDEV_LINK_100) | + BIT(TRIGGER_NETDEV_LINK); + break; + + case VSC8531_LINK_10_1000_ACTIVITY: + *rules = BIT(TRIGGER_NETDEV_LINK_1000) | + BIT(TRIGGER_NETDEV_LINK_10) | + BIT(TRIGGER_NETDEV_LINK); + break; + + case VSC8531_LINK_10_100_ACTIVITY: + *rules = BIT(TRIGGER_NETDEV_LINK_100) | + BIT(TRIGGER_NETDEV_LINK_10) | + BIT(TRIGGER_NETDEV_LINK); + break; + + default: + *rules = 0; + break; + } + + if (!behavior && *rules) + *rules |= BIT(TRIGGER_NETDEV_RX) | BIT(TRIGGER_NETDEV_TX); + + return 0; +} + +static int vsc85xx_led_hw_control_set(struct phy_device *phydev, u8 index, + unsigned long rules) +{ + struct vsc8531_private *vsc8531 = phydev->priv; + u8 mode = VSC8531_FORCE_LED_ON; + bool combine_disable = false; + bool has_rx, has_tx; + int ret; + + if (index >= vsc8531->nleds) + return -EINVAL; + + if (rules & BIT(TRIGGER_NETDEV_LINK)) + mode = VSC8531_LINK_ACTIVITY; + + if (rules & BIT(TRIGGER_NETDEV_LINK_10)) + mode = VSC8531_LINK_10_ACTIVITY; + + if (rules & BIT(TRIGGER_NETDEV_LINK_100)) + mode = VSC8531_LINK_100_ACTIVITY; + + if (rules & BIT(TRIGGER_NETDEV_LINK_1000)) + mode = VSC8531_LINK_1000_ACTIVITY; + + if (rules & BIT(TRIGGER_NETDEV_LINK_100) && + rules & BIT(TRIGGER_NETDEV_LINK_1000)) + mode = VSC8531_LINK_100_1000_ACTIVITY; + + if (rules & BIT(TRIGGER_NETDEV_LINK_10) && + rules & BIT(TRIGGER_NETDEV_LINK_1000)) + mode = VSC8531_LINK_10_1000_ACTIVITY; + + if (rules & BIT(TRIGGER_NETDEV_LINK_10) && + rules & BIT(TRIGGER_NETDEV_LINK_100)) + mode = VSC8531_LINK_10_100_ACTIVITY; + + /* + * The VSC85xx PHYs provides an option to control LED behavior. By + * default, the LEDx combine function is enabled, meaning the LED + * will be on when there is link/activity or duplex/collision. If + * the combine function is disabled, the LED will be on only for + * link or duplex. + * + * To control this behavior, we check the selected rules. If both + * RX and TX activity are not selected, the LED combine function + * is disabled; otherwise, it remains enabled. + */ + has_rx = !!(rules & BIT(TRIGGER_NETDEV_RX)); + has_tx = !!(rules & BIT(TRIGGER_NETDEV_TX)); + if (!has_rx && !has_tx) + combine_disable = true; + + ret = vsc85xx_led_combine_disable_set(phydev, index, combine_disable); + if (ret < 0) + return ret; + + return vsc85xx_led_cntl_set(phydev, index, mode); +} + static int vsc8514_probe(struct phy_device *phydev) { static const struct vsc85xx_probe_config vsc8514_cfg = { @@ -2380,6 +2566,10 @@ static struct phy_driver vsc85xx_driver[] = { .get_sset_count = &vsc85xx_get_sset_count, .get_strings = &vsc85xx_get_strings, .get_stats = &vsc85xx_get_stats, + .led_brightness_set = vsc85xx_led_brightness_set, + .led_hw_is_supported = vsc85xx_led_hw_is_supported, + .led_hw_control_get = vsc85xx_led_hw_control_get, + .led_hw_control_set = vsc85xx_led_hw_control_set, }, { .phy_id = PHY_ID_VSC8502, @@ -2404,6 +2594,10 @@ static struct phy_driver vsc85xx_driver[] = { .get_sset_count = &vsc85xx_get_sset_count, .get_strings = &vsc85xx_get_strings, .get_stats = &vsc85xx_get_stats, + .led_brightness_set = vsc85xx_led_brightness_set, + .led_hw_is_supported = vsc85xx_led_hw_is_supported, + .led_hw_control_get = vsc85xx_led_hw_control_get, + .led_hw_control_set = vsc85xx_led_hw_control_set, }, { .phy_id = PHY_ID_VSC8504, @@ -2431,6 +2625,10 @@ static struct phy_driver vsc85xx_driver[] = { .get_stats = &vsc85xx_get_stats, .inband_caps = vsc85xx_inband_caps, .config_inband = vsc85xx_config_inband, + .led_brightness_set = vsc85xx_led_brightness_set, + .led_hw_is_supported = vsc85xx_led_hw_is_supported, + .led_hw_control_get = vsc85xx_led_hw_control_get, + .led_hw_control_set = vsc85xx_led_hw_control_set, }, { .phy_id = PHY_ID_VSC8514, @@ -2456,6 +2654,10 @@ static struct phy_driver vsc85xx_driver[] = { .get_stats = &vsc85xx_get_stats, .inband_caps = vsc85xx_inband_caps, .config_inband = vsc85xx_config_inband, + .led_brightness_set = vsc85xx_led_brightness_set, + .led_hw_is_supported = vsc85xx_led_hw_is_supported, + .led_hw_control_get = vsc85xx_led_hw_control_get, + .led_hw_control_set = vsc85xx_led_hw_control_set, }, { .phy_id = PHY_ID_VSC8530, @@ -2480,6 +2682,10 @@ static struct phy_driver vsc85xx_driver[] = { .get_sset_count = &vsc85xx_get_sset_count, .get_strings = &vsc85xx_get_strings, .get_stats = &vsc85xx_get_stats, + .led_brightness_set = vsc85xx_led_brightness_set, + .led_hw_is_supported = vsc85xx_led_hw_is_supported, + .led_hw_control_get = vsc85xx_led_hw_control_get, + .led_hw_control_set = vsc85xx_led_hw_control_set, }, { .phy_id = PHY_ID_VSC8531, @@ -2504,6 +2710,10 @@ static struct phy_driver vsc85xx_driver[] = { .get_sset_count = &vsc85xx_get_sset_count, .get_strings = &vsc85xx_get_strings, .get_stats = &vsc85xx_get_stats, + .led_brightness_set = vsc85xx_led_brightness_set, + .led_hw_is_supported = vsc85xx_led_hw_is_supported, + .led_hw_control_get = vsc85xx_led_hw_control_get, + .led_hw_control_set = vsc85xx_led_hw_control_set, }, { .phy_id = PHY_ID_VSC8540, @@ -2528,6 +2738,10 @@ static struct phy_driver vsc85xx_driver[] = { .get_sset_count = &vsc85xx_get_sset_count, .get_strings = &vsc85xx_get_strings, .get_stats = &vsc85xx_get_stats, + .led_brightness_set = vsc85xx_led_brightness_set, + .led_hw_is_supported = vsc85xx_led_hw_is_supported, + .led_hw_control_get = vsc85xx_led_hw_control_get, + .led_hw_control_set = vsc85xx_led_hw_control_set, }, { .phy_id = PHY_ID_VSC8541, @@ -2552,6 +2766,10 @@ static struct phy_driver vsc85xx_driver[] = { .get_sset_count = &vsc85xx_get_sset_count, .get_strings = &vsc85xx_get_strings, .get_stats = &vsc85xx_get_stats, + .led_brightness_set = vsc85xx_led_brightness_set, + .led_hw_is_supported = vsc85xx_led_hw_is_supported, + .led_hw_control_get = vsc85xx_led_hw_control_get, + .led_hw_control_set = vsc85xx_led_hw_control_set, }, { .phy_id = PHY_ID_VSC8552, @@ -2578,6 +2796,10 @@ static struct phy_driver vsc85xx_driver[] = { .get_stats = &vsc85xx_get_stats, .inband_caps = vsc85xx_inband_caps, .config_inband = vsc85xx_config_inband, + .led_brightness_set = vsc85xx_led_brightness_set, + .led_hw_is_supported = vsc85xx_led_hw_is_supported, + .led_hw_control_get = vsc85xx_led_hw_control_get, + .led_hw_control_set = vsc85xx_led_hw_control_set, }, { PHY_ID_MATCH_EXACT(PHY_ID_VSC856X), @@ -2601,6 +2823,10 @@ static struct phy_driver vsc85xx_driver[] = { .get_stats = &vsc85xx_get_stats, .inband_caps = vsc85xx_inband_caps, .config_inband = vsc85xx_config_inband, + .led_brightness_set = vsc85xx_led_brightness_set, + .led_hw_is_supported = vsc85xx_led_hw_is_supported, + .led_hw_control_get = vsc85xx_led_hw_control_get, + .led_hw_control_set = vsc85xx_led_hw_control_set, }, { .phy_id = PHY_ID_VSC8572, @@ -2629,6 +2855,10 @@ static struct phy_driver vsc85xx_driver[] = { .get_stats = &vsc85xx_get_stats, .inband_caps = vsc85xx_inband_caps, .config_inband = vsc85xx_config_inband, + .led_brightness_set = vsc85xx_led_brightness_set, + .led_hw_is_supported = vsc85xx_led_hw_is_supported, + .led_hw_control_get = vsc85xx_led_hw_control_get, + .led_hw_control_set = vsc85xx_led_hw_control_set, }, { .phy_id = PHY_ID_VSC8574, @@ -2657,6 +2887,10 @@ static struct phy_driver vsc85xx_driver[] = { .get_stats = &vsc85xx_get_stats, .inband_caps = vsc85xx_inband_caps, .config_inband = vsc85xx_config_inband, + .led_brightness_set = vsc85xx_led_brightness_set, + .led_hw_is_supported = vsc85xx_led_hw_is_supported, + .led_hw_control_get = vsc85xx_led_hw_control_get, + .led_hw_control_set = vsc85xx_led_hw_control_set, }, { PHY_ID_MATCH_EXACT(PHY_ID_VSC8575), @@ -2682,6 +2916,10 @@ static struct phy_driver vsc85xx_driver[] = { .get_stats = &vsc85xx_get_stats, .inband_caps = vsc85xx_inband_caps, .config_inband = vsc85xx_config_inband, + .led_brightness_set = vsc85xx_led_brightness_set, + .led_hw_is_supported = vsc85xx_led_hw_is_supported, + .led_hw_control_get = vsc85xx_led_hw_control_get, + .led_hw_control_set = vsc85xx_led_hw_control_set, }, { PHY_ID_MATCH_EXACT(PHY_ID_VSC8582), @@ -2707,6 +2945,10 @@ static struct phy_driver vsc85xx_driver[] = { .get_stats = &vsc85xx_get_stats, .inband_caps = vsc85xx_inband_caps, .config_inband = vsc85xx_config_inband, + .led_brightness_set = vsc85xx_led_brightness_set, + .led_hw_is_supported = vsc85xx_led_hw_is_supported, + .led_hw_control_get = vsc85xx_led_hw_control_get, + .led_hw_control_set = vsc85xx_led_hw_control_set, }, { PHY_ID_MATCH_EXACT(PHY_ID_VSC8584), @@ -2733,6 +2975,10 @@ static struct phy_driver vsc85xx_driver[] = { .link_change_notify = &vsc85xx_link_change_notify, .inband_caps = vsc85xx_inband_caps, .config_inband = vsc85xx_config_inband, + .led_brightness_set = vsc85xx_led_brightness_set, + .led_hw_is_supported = vsc85xx_led_hw_is_supported, + .led_hw_control_get = vsc85xx_led_hw_control_get, + .led_hw_control_set = vsc85xx_led_hw_control_set, } }; From df397943196f7ff396beb41449583b4629ee99e2 Mon Sep 17 00:00:00 2001 From: Lad Prabhakar Date: Wed, 12 Nov 2025 13:57:15 +0000 Subject: [PATCH 0795/1381] net: phy: mscc: Handle devm_phy_package_join() failure in vsc85xx_probe_common() devm_phy_package_join() may fail and return a negative error code. Update vsc85xx_probe_common() to properly handle this failure by checking the return value and propagating the error to the caller. Signed-off-by: Lad Prabhakar Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20251112135715.1017117-5-prabhakar.mahadev-lad.rj@bp.renesas.com Signed-off-by: Jakub Kicinski --- drivers/net/phy/mscc/mscc_main.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/drivers/net/phy/mscc/mscc_main.c b/drivers/net/phy/mscc/mscc_main.c index 21fcaf07bc6d..2b9fb8a675a6 100644 --- a/drivers/net/phy/mscc/mscc_main.c +++ b/drivers/net/phy/mscc/mscc_main.c @@ -2264,8 +2264,11 @@ static int vsc85xx_probe_common(struct phy_device *phydev, /* Set up package if needed */ if (cfg->use_package) { vsc8584_get_base_addr(phydev); - devm_phy_package_join(&phydev->mdio.dev, phydev, - vsc8531->base_addr, cfg->shared_size); + ret = devm_phy_package_join(&phydev->mdio.dev, phydev, + vsc8531->base_addr, + cfg->shared_size); + if (ret) + return ret; } /* Configure LED settings */ From c7dc5b5228822d2389e6e441f10169e460bcc67a Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 12 Nov 2025 19:17:00 -0800 Subject: [PATCH 0796/1381] ipv6: clean up routes when manually removing address with a lifetime When an IPv6 address with a finite lifetime (configured with valid_lft and preferred_lft) is manually deleted, the kernel does not clean up the associated prefix route. This results in orphaned routes (marked "proto kernel") remaining in the routing table even after their corresponding address has been deleted. This is particularly problematic on networks using combination of SLAAC and bridges. 1. Machine comes up and performs RA on eth0. 2. User creates a bridge - does an ip -6 addr flush dev eth0; - adds the eth0 under the bridge. 3. SLAAC happens on br0. Even tho the address has "moved" to br0 there will still be a route pointing to eth0, but eth0 is not usable for IP any more. Reviewed-by: David Ahern Reviewed-by: Ido Schimmel Link: https://patch.msgid.link/20251113031700.3736285-1-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ipv6/addrconf.c | 2 +- tools/testing/selftests/net/rtnetlink.sh | 20 ++++++++++++++++++++ 2 files changed, 21 insertions(+), 1 deletion(-) diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 40e9c336f6c5..b66217d1b2f8 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -1324,7 +1324,7 @@ static void ipv6_del_addr(struct inet6_ifaddr *ifp) __in6_ifa_put(ifp); } - if (ifp->flags & IFA_F_PERMANENT && !(ifp->flags & IFA_F_NOPREFIXROUTE)) + if (!(ifp->flags & IFA_F_NOPREFIXROUTE)) action = check_cleanup_prefix_route(ifp, &expires); list_del_rcu(&ifp->if_list); diff --git a/tools/testing/selftests/net/rtnetlink.sh b/tools/testing/selftests/net/rtnetlink.sh index 163a084d525d..248c2b91fe42 100755 --- a/tools/testing/selftests/net/rtnetlink.sh +++ b/tools/testing/selftests/net/rtnetlink.sh @@ -8,6 +8,7 @@ ALL_TESTS=" kci_test_polrouting kci_test_route_get kci_test_addrlft + kci_test_addrlft_route_cleanup kci_test_promote_secondaries kci_test_tc kci_test_gre @@ -323,6 +324,25 @@ kci_test_addrlft() end_test "PASS: preferred_lft addresses have expired" } +kci_test_addrlft_route_cleanup() +{ + local ret=0 + local test_addr="2001:db8:99::1/64" + local test_prefix="2001:db8:99::/64" + + run_cmd ip -6 addr add $test_addr dev "$devdummy" valid_lft 300 preferred_lft 300 + run_cmd_grep "$test_prefix proto kernel" ip -6 route show dev "$devdummy" + run_cmd ip -6 addr del $test_addr dev "$devdummy" + run_cmd_grep_fail "$test_prefix" ip -6 route show dev "$devdummy" + + if [ $ret -ne 0 ]; then + end_test "FAIL: route not cleaned up when address with valid_lft deleted" + return 1 + fi + + end_test "PASS: route cleaned up when address with valid_lft deleted" +} + kci_test_promote_secondaries() { run_cmd ifconfig "$devdummy" From e1215d1d38c00d244e5e7db4648598a0a03cc446 Mon Sep 17 00:00:00 2001 From: Dimitri Daskalakis Date: Wed, 12 Nov 2025 20:31:02 -0800 Subject: [PATCH 0797/1381] selftests: drv-net: xdp: Fix register spill error with clang 20 On clang 20.1.8 the XDP program fails to load with a register spill error. Since hdr_len is a __u32, the compiler decided it only needed the lower 32-bits of ctx->data, which later triggers the register spill verifier error. Suggested-by: Martin KaFai Lau Signed-off-by: Dimitri Daskalakis Link: https://patch.msgid.link/20251113043102.4062150-1-dimitri.daskalakis1@gmail.com Signed-off-by: Jakub Kicinski --- tools/testing/selftests/net/lib/xdp_native.bpf.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/tools/testing/selftests/net/lib/xdp_native.bpf.c b/tools/testing/selftests/net/lib/xdp_native.bpf.c index c368fc045f4b..64f05229ab24 100644 --- a/tools/testing/selftests/net/lib/xdp_native.bpf.c +++ b/tools/testing/selftests/net/lib/xdp_native.bpf.c @@ -332,7 +332,7 @@ static __u16 csum_fold_helper(__u32 csum) } static int xdp_adjst_tail_shrnk_data(struct xdp_md *ctx, __u16 offset, - __u32 hdr_len) + unsigned long hdr_len) { char tmp_buff[MAX_ADJST_OFFSET]; __u32 buff_pos, udp_csum = 0; @@ -422,8 +422,9 @@ static int xdp_adjst_tail(struct xdp_md *ctx, __u16 port) { struct udphdr *udph = NULL; __s32 *adjust_offset, *val; - __u32 key, hdr_len; + unsigned long hdr_len; void *offset_ptr; + __u32 key; __u8 tag; int ret; From eca8b8fc74cb492591f73dcdd945a5fe4612b1a4 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Thu, 13 Nov 2025 07:27:03 -0800 Subject: [PATCH 0798/1381] selftests: drv-net: xdp: make the XDP qstats tests less flaky The XDP qstats tests send 2k packets over a single socket. Looks like when netdev CI is busy running those tests in QEMU occasionally flakes. The target doesn't get to run at all before all 2000 packets are sent. Lower the number of packets to 1000 and reopen the socket every 50 packets, to give RSS a chance to spread the packets to multiple queues. For the netdev CI testing either lowering the count or using multiple sockets is enough, but let's do both for extra resiliency. Acked-by: Stanislav Fomichev Link: https://patch.msgid.link/20251113152703.3819756-1-kuba@kernel.org Signed-off-by: Jakub Kicinski --- tools/testing/selftests/drivers/net/xdp.py | 15 +++++++++------ 1 file changed, 9 insertions(+), 6 deletions(-) diff --git a/tools/testing/selftests/drivers/net/xdp.py b/tools/testing/selftests/drivers/net/xdp.py index a148004e1c36..834a37ae7d0d 100755 --- a/tools/testing/selftests/drivers/net/xdp.py +++ b/tools/testing/selftests/drivers/net/xdp.py @@ -687,9 +687,12 @@ def _test_xdp_native_ifc_stats(cfg, act): "/dev/null" # Listener runs on "remote" in case of XDP_TX rx_host = cfg.remote if act == XDPAction.TX else None - # We want to spew 2000 packets quickly, bash seems to do a good enough job - tx_udp = f"exec 5<>/dev/udp/{cfg.addr}/{port}; " \ - "for i in `seq 2000`; do echo a >&5; done; exec 5>&-" + # We want to spew 1000 packets quickly, bash seems to do a good enough job + # Each reopening of the socket gives us a differenot local port (for RSS) + tx_udp = "for _ in `seq 20`; do " \ + f"exec 5<>/dev/udp/{cfg.addr}/{port}; " \ + "for i in `seq 50`; do echo a >&5; done; " \ + "exec 5>&-; done" cfg.wait_hw_stats_settle() # Qstats have more clearly defined semantics than rtnetlink. @@ -704,11 +707,11 @@ def _test_xdp_native_ifc_stats(cfg, act): cfg.wait_hw_stats_settle() after = cfg.netnl.qstats_get({"ifindex": cfg.ifindex}, dump=True)[0] - ksft_ge(after['rx-packets'] - before['rx-packets'], 2000) + expected_pkts = 1000 + ksft_ge(after['rx-packets'] - before['rx-packets'], expected_pkts) if act == XDPAction.TX: - ksft_ge(after['tx-packets'] - before['tx-packets'], 2000) + ksft_ge(after['tx-packets'] - before['tx-packets'], expected_pkts) - expected_pkts = 2000 stats = _get_stats(prog_info["maps"]["map_xdp_stats"]) ksft_eq(stats[XDPStats.RX.value], expected_pkts, "XDP RX stats mismatch") if act == XDPAction.TX: From 6d650ae9282bcec1e76205b44cb8f17e2265052e Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 13 Nov 2025 14:03:57 +0000 Subject: [PATCH 0799/1381] tcp: gro: inline tcp_gro_pull_header() tcp_gro_pull_header() is used in GRO fast path, inline it. Signed-off-by: Eric Dumazet Link: https://patch.msgid.link/20251113140358.58242-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- include/net/gro.h | 27 +++++++++++++++++++++++++++ include/net/tcp.h | 1 - net/ipv4/tcp_offload.c | 27 --------------------------- 3 files changed, 27 insertions(+), 28 deletions(-) diff --git a/include/net/gro.h b/include/net/gro.h index e3affb2e2ca8..b65f631c521d 100644 --- a/include/net/gro.h +++ b/include/net/gro.h @@ -593,4 +593,31 @@ static inline void inet6_get_iif_sdif(const struct sk_buff *skb, int *iif, int * struct packet_offload *gro_find_receive_by_type(__be16 type); struct packet_offload *gro_find_complete_by_type(__be16 type); +static inline struct tcphdr *tcp_gro_pull_header(struct sk_buff *skb) +{ + unsigned int thlen, hlen, off; + struct tcphdr *th; + + off = skb_gro_offset(skb); + hlen = off + sizeof(*th); + th = skb_gro_header(skb, hlen, off); + if (unlikely(!th)) + return NULL; + + thlen = th->doff * 4; + if (unlikely(thlen < sizeof(*th))) + return NULL; + + hlen = off + thlen; + if (!skb_gro_may_pull(skb, hlen)) { + th = skb_gro_header_slow(skb, hlen, off); + if (unlikely(!th)) + return NULL; + } + + skb_gro_pull(skb, thlen); + + return th; +} + #endif /* _NET_GRO_H */ diff --git a/include/net/tcp.h b/include/net/tcp.h index 4833ec7903ec..0deb5e9dd911 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -2313,7 +2313,6 @@ void tcp_v4_destroy_sock(struct sock *sk); struct sk_buff *tcp_gso_segment(struct sk_buff *skb, netdev_features_t features); -struct tcphdr *tcp_gro_pull_header(struct sk_buff *skb); struct sk_buff *tcp_gro_lookup(struct list_head *head, struct tcphdr *th); struct sk_buff *tcp_gro_receive(struct list_head *head, struct sk_buff *skb, struct tcphdr *th); diff --git a/net/ipv4/tcp_offload.c b/net/ipv4/tcp_offload.c index 2cb93da93abc..fdda18b1abda 100644 --- a/net/ipv4/tcp_offload.c +++ b/net/ipv4/tcp_offload.c @@ -282,33 +282,6 @@ struct sk_buff *tcp_gro_lookup(struct list_head *head, struct tcphdr *th) return NULL; } -struct tcphdr *tcp_gro_pull_header(struct sk_buff *skb) -{ - unsigned int thlen, hlen, off; - struct tcphdr *th; - - off = skb_gro_offset(skb); - hlen = off + sizeof(*th); - th = skb_gro_header(skb, hlen, off); - if (unlikely(!th)) - return NULL; - - thlen = th->doff * 4; - if (thlen < sizeof(*th)) - return NULL; - - hlen = off + thlen; - if (!skb_gro_may_pull(skb, hlen)) { - th = skb_gro_header_slow(skb, hlen, off); - if (unlikely(!th)) - return NULL; - } - - skb_gro_pull(skb, thlen); - - return th; -} - struct sk_buff *tcp_gro_receive(struct list_head *head, struct sk_buff *skb, struct tcphdr *th) { From 06ac470658190e97518f131df01c9c530c293320 Mon Sep 17 00:00:00 2001 From: Yue Haibing Date: Thu, 13 Nov 2025 19:45:01 +0800 Subject: [PATCH 0800/1381] sctp: Remove unused declaration sctp_auth_init_hmacs() Commit bf40785fa437 ("sctp: Use HMAC-SHA1 and HMAC-SHA256 library for chunk authentication") removed the implementation but leave declaration. Signed-off-by: Yue Haibing Reviewed-by: Eric Biggers Link: https://patch.msgid.link/20251113114501.32905-1-yuehaibing@huawei.com Signed-off-by: Jakub Kicinski --- include/net/sctp/auth.h | 1 - 1 file changed, 1 deletion(-) diff --git a/include/net/sctp/auth.h b/include/net/sctp/auth.h index 3d5879e08e78..6f2cd562b1de 100644 --- a/include/net/sctp/auth.h +++ b/include/net/sctp/auth.h @@ -72,7 +72,6 @@ struct sctp_shared_key *sctp_auth_get_shkey( int sctp_auth_asoc_copy_shkeys(const struct sctp_endpoint *ep, struct sctp_association *asoc, gfp_t gfp); -int sctp_auth_init_hmacs(struct sctp_endpoint *ep, gfp_t gfp); const struct sctp_hmac *sctp_auth_get_hmac(__u16 hmac_id); const struct sctp_hmac * sctp_auth_asoc_get_hmac(const struct sctp_association *asoc); From f455d3f02d8998edb20a0accac5eeee769bd6fcd Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Thu, 13 Nov 2025 06:23:29 -0800 Subject: [PATCH 0801/1381] net: ixgbe: convert to use .get_rx_ring_count Convert the ixgbe driver to use the new .get_rx_ring_count ethtool operation for handling ETHTOOL_GRXRINGS command. This simplifies the code by extracting the ring count logic into a dedicated callback. The new callback provides the same functionality in a more direct way, following the ongoing ethtool API modernization. This was compile-tested only. Reviewed-by: Paul Menzel Reviewed-by: Aleksandr Loktionov Signed-off-by: Breno Leitao Link: https://patch.msgid.link/20251113-ixgbe_gxrings-v2-1-0ecf57808a78@debian.org Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-) diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c index 2d660e9edb80..2ad81f687a84 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c @@ -2805,6 +2805,14 @@ static int ixgbe_rss_indir_tbl_max(struct ixgbe_adapter *adapter) return 64; } +static u32 ixgbe_get_rx_ring_count(struct net_device *dev) +{ + struct ixgbe_adapter *adapter = ixgbe_from_netdev(dev); + + return min_t(u32, adapter->num_rx_queues, + ixgbe_rss_indir_tbl_max(adapter)); +} + static int ixgbe_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd, u32 *rule_locs) { @@ -2812,11 +2820,6 @@ static int ixgbe_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd, int ret = -EOPNOTSUPP; switch (cmd->cmd) { - case ETHTOOL_GRXRINGS: - cmd->data = min_t(int, adapter->num_rx_queues, - ixgbe_rss_indir_tbl_max(adapter)); - ret = 0; - break; case ETHTOOL_GRXCLSRLCNT: cmd->rule_cnt = adapter->fdir_filter_count; ret = 0; @@ -3743,6 +3746,7 @@ static const struct ethtool_ops ixgbe_ethtool_ops = { .get_ethtool_stats = ixgbe_get_ethtool_stats, .get_coalesce = ixgbe_get_coalesce, .set_coalesce = ixgbe_set_coalesce, + .get_rx_ring_count = ixgbe_get_rx_ring_count, .get_rxnfc = ixgbe_get_rxnfc, .set_rxnfc = ixgbe_set_rxnfc, .get_rxfh_indir_size = ixgbe_rss_indir_size, @@ -3791,6 +3795,7 @@ static const struct ethtool_ops ixgbe_ethtool_ops_e610 = { .get_ethtool_stats = ixgbe_get_ethtool_stats, .get_coalesce = ixgbe_get_coalesce, .set_coalesce = ixgbe_set_coalesce, + .get_rx_ring_count = ixgbe_get_rx_ring_count, .get_rxnfc = ixgbe_get_rxnfc, .set_rxnfc = ixgbe_set_rxnfc, .get_rxfh_indir_size = ixgbe_rss_indir_size, From 04ca7a69a35b060ecee60a9923e9891f5dd92eca Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Wed, 12 Nov 2025 01:50:23 -0800 Subject: [PATCH 0802/1381] net: bnx2x: convert to use get_rx_ring_count Convert the bnx2x driver to use the new .get_rx_ring_count ethtool operation instead of implementing .get_rxnfc solely for handling ETHTOOL_GRXRINGS command. This simplifies the code by replacing the switch statement with a direct return of the queue count. The new callback provides the same functionality in a more direct way, following the ongoing ethtool API modernization. Signed-off-by: Breno Leitao Reviewed-by: Simon Horman Link: https://patch.msgid.link/20251112-bnx_grxrings-v1-1-1c2cb73979e2@debian.org Signed-off-by: Jakub Kicinski --- .../net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c | 16 ++++------------ 1 file changed, 4 insertions(+), 12 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c index fc8dec37a9e4..3d853eeb976f 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c @@ -3355,19 +3355,11 @@ static int bnx2x_get_rxfh_fields(struct net_device *dev, return 0; } -static int bnx2x_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *info, - u32 *rules __always_unused) +static u32 bnx2x_get_rx_ring_count(struct net_device *dev) { struct bnx2x *bp = netdev_priv(dev); - switch (info->cmd) { - case ETHTOOL_GRXRINGS: - info->data = BNX2X_NUM_ETH_QUEUES(bp); - return 0; - default: - DP(BNX2X_MSG_ETHTOOL, "Command parameters not supported\n"); - return -EOPNOTSUPP; - } + return BNX2X_NUM_ETH_QUEUES(bp); } static int bnx2x_set_rxfh_fields(struct net_device *dev, @@ -3674,7 +3666,7 @@ static const struct ethtool_ops bnx2x_ethtool_ops = { .get_strings = bnx2x_get_strings, .set_phys_id = bnx2x_set_phys_id, .get_ethtool_stats = bnx2x_get_ethtool_stats, - .get_rxnfc = bnx2x_get_rxnfc, + .get_rx_ring_count = bnx2x_get_rx_ring_count, .get_rxfh_indir_size = bnx2x_get_rxfh_indir_size, .get_rxfh = bnx2x_get_rxfh, .set_rxfh = bnx2x_set_rxfh, @@ -3702,7 +3694,7 @@ static const struct ethtool_ops bnx2x_vf_ethtool_ops = { .get_sset_count = bnx2x_get_sset_count, .get_strings = bnx2x_get_strings, .get_ethtool_stats = bnx2x_get_ethtool_stats, - .get_rxnfc = bnx2x_get_rxnfc, + .get_rx_ring_count = bnx2x_get_rx_ring_count, .get_rxfh_indir_size = bnx2x_get_rxfh_indir_size, .get_rxfh = bnx2x_get_rxfh, .set_rxfh = bnx2x_set_rxfh, From ebb07edf97383e6821e7263d17cf6b591603c47b Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Thu, 13 Nov 2025 17:46:35 +0000 Subject: [PATCH 0803/1381] net: stmmac: rk: replace HIWORD_UPDATE() with GRF_FIELD() Provide GRF_FIELD() which takes the high/low bit numbers of the field and field value, generates the mask and passes it to FIELD_PREP_WM16. Replace all HIWORD_UPDATE() instances with this. No change to produced code on aarch64. Signed-off-by: Russell King (Oracle) Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/E1vJbP1-0000000EBqJ-1AjR@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../net/ethernet/stmicro/stmmac/dwmac-rk.c | 64 +++++++++---------- 1 file changed, 32 insertions(+), 32 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c index a5c7e03ebc63..794a7ed71451 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c @@ -149,11 +149,11 @@ static int rk_set_clk_mac_speed(struct rk_priv_data *bsp_priv, return clk_set_rate(clk_mac_speed, rate); } -#define HIWORD_UPDATE(val, mask, shift) \ - (FIELD_PREP_WM16((mask) << (shift), (val))) +#define GRF_FIELD(hi, lo, val) \ + FIELD_PREP_WM16(GENMASK_U16(hi, lo), val) -#define GRF_BIT(nr) (BIT(nr) | BIT(nr+16)) -#define GRF_CLR_BIT(nr) (BIT(nr+16)) +#define GRF_BIT(nr) (BIT(nr) | BIT(nr+16)) +#define GRF_CLR_BIT(nr) (BIT(nr+16)) #define DELAY_ENABLE(soc, tx, rx) \ (((tx) ? soc##_GMAC_TXCLK_DLY_ENABLE : soc##_GMAC_TXCLK_DLY_DISABLE) | \ @@ -168,8 +168,8 @@ static int rk_set_clk_mac_speed(struct rk_priv_data *bsp_priv, #define RK_MACPHY_DISABLE GRF_CLR_BIT(0) #define RK_MACPHY_CFG_CLK_50M GRF_BIT(14) #define RK_GMAC2PHY_RMII_MODE (GRF_BIT(6) | GRF_CLR_BIT(7)) -#define RK_GRF_CON2_MACPHY_ID HIWORD_UPDATE(0x1234, 0xffff, 0) -#define RK_GRF_CON3_MACPHY_ID HIWORD_UPDATE(0x35, 0x3f, 0) +#define RK_GRF_CON2_MACPHY_ID GRF_FIELD(15, 0, 0x1234) +#define RK_GRF_CON3_MACPHY_ID GRF_FIELD(5, 0, 0x35) static void rk_gmac_integrated_ephy_powerup(struct rk_priv_data *priv) { @@ -285,8 +285,8 @@ static const struct rk_gmac_ops px30_ops = { #define RK3128_GMAC_TXCLK_DLY_DISABLE GRF_CLR_BIT(14) #define RK3128_GMAC_RXCLK_DLY_ENABLE GRF_BIT(15) #define RK3128_GMAC_RXCLK_DLY_DISABLE GRF_CLR_BIT(15) -#define RK3128_GMAC_CLK_RX_DL_CFG(val) HIWORD_UPDATE(val, 0x7F, 7) -#define RK3128_GMAC_CLK_TX_DL_CFG(val) HIWORD_UPDATE(val, 0x7F, 0) +#define RK3128_GMAC_CLK_RX_DL_CFG(val) GRF_FIELD(13, 7, val) +#define RK3128_GMAC_CLK_TX_DL_CFG(val) GRF_FIELD(6, 0, val) /* RK3128_GRF_MAC_CON1 */ #define RK3128_GMAC_PHY_INTF_SEL_RGMII \ @@ -350,8 +350,8 @@ static const struct rk_gmac_ops rk3128_ops = { #define RK3228_GRF_CON_MUX 0x50 /* RK3228_GRF_MAC_CON0 */ -#define RK3228_GMAC_CLK_RX_DL_CFG(val) HIWORD_UPDATE(val, 0x7F, 7) -#define RK3228_GMAC_CLK_TX_DL_CFG(val) HIWORD_UPDATE(val, 0x7F, 0) +#define RK3228_GMAC_CLK_RX_DL_CFG(val) GRF_FIELD(13, 7, val) +#define RK3228_GMAC_CLK_TX_DL_CFG(val) GRF_FIELD(6, 0, val) /* RK3228_GRF_MAC_CON1 */ #define RK3228_GMAC_PHY_INTF_SEL_RGMII \ @@ -456,8 +456,8 @@ static const struct rk_gmac_ops rk3228_ops = { #define RK3288_GMAC_TXCLK_DLY_DISABLE GRF_CLR_BIT(14) #define RK3288_GMAC_RXCLK_DLY_ENABLE GRF_BIT(15) #define RK3288_GMAC_RXCLK_DLY_DISABLE GRF_CLR_BIT(15) -#define RK3288_GMAC_CLK_RX_DL_CFG(val) HIWORD_UPDATE(val, 0x7F, 7) -#define RK3288_GMAC_CLK_TX_DL_CFG(val) HIWORD_UPDATE(val, 0x7F, 0) +#define RK3288_GMAC_CLK_RX_DL_CFG(val) GRF_FIELD(13, 7, val) +#define RK3288_GMAC_CLK_TX_DL_CFG(val) GRF_FIELD(6, 0, val) static void rk3288_set_to_rgmii(struct rk_priv_data *bsp_priv, int tx_delay, int rx_delay) @@ -537,8 +537,8 @@ static const struct rk_gmac_ops rk3308_ops = { #define RK3328_GRF_MACPHY_CON1 0xb04 /* RK3328_GRF_MAC_CON0 */ -#define RK3328_GMAC_CLK_RX_DL_CFG(val) HIWORD_UPDATE(val, 0x7F, 7) -#define RK3328_GMAC_CLK_TX_DL_CFG(val) HIWORD_UPDATE(val, 0x7F, 0) +#define RK3328_GMAC_CLK_RX_DL_CFG(val) GRF_FIELD(13, 7, val) +#define RK3328_GMAC_CLK_TX_DL_CFG(val) GRF_FIELD(6, 0, val) /* RK3328_GRF_MAC_CON1 */ #define RK3328_GMAC_PHY_INTF_SEL_RGMII \ @@ -651,8 +651,8 @@ static const struct rk_gmac_ops rk3328_ops = { #define RK3366_GMAC_TXCLK_DLY_DISABLE GRF_CLR_BIT(7) #define RK3366_GMAC_RXCLK_DLY_ENABLE GRF_BIT(15) #define RK3366_GMAC_RXCLK_DLY_DISABLE GRF_CLR_BIT(15) -#define RK3366_GMAC_CLK_RX_DL_CFG(val) HIWORD_UPDATE(val, 0x7F, 8) -#define RK3366_GMAC_CLK_TX_DL_CFG(val) HIWORD_UPDATE(val, 0x7F, 0) +#define RK3366_GMAC_CLK_RX_DL_CFG(val) GRF_FIELD(14, 8, val) +#define RK3366_GMAC_CLK_TX_DL_CFG(val) GRF_FIELD(6, 0, val) static void rk3366_set_to_rgmii(struct rk_priv_data *bsp_priv, int tx_delay, int rx_delay) @@ -718,8 +718,8 @@ static const struct rk_gmac_ops rk3366_ops = { #define RK3368_GMAC_TXCLK_DLY_DISABLE GRF_CLR_BIT(7) #define RK3368_GMAC_RXCLK_DLY_ENABLE GRF_BIT(15) #define RK3368_GMAC_RXCLK_DLY_DISABLE GRF_CLR_BIT(15) -#define RK3368_GMAC_CLK_RX_DL_CFG(val) HIWORD_UPDATE(val, 0x7F, 8) -#define RK3368_GMAC_CLK_TX_DL_CFG(val) HIWORD_UPDATE(val, 0x7F, 0) +#define RK3368_GMAC_CLK_RX_DL_CFG(val) GRF_FIELD(14, 8, val) +#define RK3368_GMAC_CLK_TX_DL_CFG(val) GRF_FIELD(6, 0, val) static void rk3368_set_to_rgmii(struct rk_priv_data *bsp_priv, int tx_delay, int rx_delay) @@ -785,8 +785,8 @@ static const struct rk_gmac_ops rk3368_ops = { #define RK3399_GMAC_TXCLK_DLY_DISABLE GRF_CLR_BIT(7) #define RK3399_GMAC_RXCLK_DLY_ENABLE GRF_BIT(15) #define RK3399_GMAC_RXCLK_DLY_DISABLE GRF_CLR_BIT(15) -#define RK3399_GMAC_CLK_RX_DL_CFG(val) HIWORD_UPDATE(val, 0x7F, 8) -#define RK3399_GMAC_CLK_TX_DL_CFG(val) HIWORD_UPDATE(val, 0x7F, 0) +#define RK3399_GMAC_CLK_RX_DL_CFG(val) GRF_FIELD(14, 8, val) +#define RK3399_GMAC_CLK_TX_DL_CFG(val) GRF_FIELD(6, 0, val) static void rk3399_set_to_rgmii(struct rk_priv_data *bsp_priv, int tx_delay, int rx_delay) @@ -901,8 +901,8 @@ static const struct rk_gmac_ops rk3506_ops = { #define RK3528_GMAC_TXCLK_DLY_ENABLE GRF_BIT(14) #define RK3528_GMAC_TXCLK_DLY_DISABLE GRF_CLR_BIT(14) -#define RK3528_GMAC_CLK_RX_DL_CFG(val) HIWORD_UPDATE(val, 0xFF, 8) -#define RK3528_GMAC_CLK_TX_DL_CFG(val) HIWORD_UPDATE(val, 0xFF, 0) +#define RK3528_GMAC_CLK_RX_DL_CFG(val) GRF_FIELD(15, 8, val) +#define RK3528_GMAC_CLK_TX_DL_CFG(val) GRF_FIELD(7, 0, val) #define RK3528_GMAC0_PHY_INTF_SEL_RMII GRF_BIT(1) #define RK3528_GMAC1_PHY_INTF_SEL_RGMII GRF_CLR_BIT(8) @@ -1041,8 +1041,8 @@ static const struct rk_gmac_ops rk3528_ops = { #define RK3568_GMAC_TXCLK_DLY_DISABLE GRF_CLR_BIT(0) /* RK3568_GRF_GMAC0_CON0 && RK3568_GRF_GMAC1_CON0 */ -#define RK3568_GMAC_CLK_RX_DL_CFG(val) HIWORD_UPDATE(val, 0x7F, 8) -#define RK3568_GMAC_CLK_TX_DL_CFG(val) HIWORD_UPDATE(val, 0x7F, 0) +#define RK3568_GMAC_CLK_RX_DL_CFG(val) GRF_FIELD(14, 8, val) +#define RK3568_GMAC_CLK_TX_DL_CFG(val) GRF_FIELD(6, 0, val) static void rk3568_set_to_rgmii(struct rk_priv_data *bsp_priv, int tx_delay, int rx_delay) @@ -1096,8 +1096,8 @@ static const struct rk_gmac_ops rk3568_ops = { #define RK3576_GMAC_TXCLK_DLY_ENABLE GRF_BIT(7) #define RK3576_GMAC_TXCLK_DLY_DISABLE GRF_CLR_BIT(7) -#define RK3576_GMAC_CLK_RX_DL_CFG(val) HIWORD_UPDATE(val, 0x7F, 8) -#define RK3576_GMAC_CLK_TX_DL_CFG(val) HIWORD_UPDATE(val, 0x7F, 0) +#define RK3576_GMAC_CLK_RX_DL_CFG(val) GRF_FIELD(14, 8, val) +#define RK3576_GMAC_CLK_TX_DL_CFG(val) GRF_FIELD(6, 0, val) /* SDGMAC_GRF */ #define RK3576_GRF_GMAC_CON0 0X0020 @@ -1220,8 +1220,8 @@ static const struct rk_gmac_ops rk3576_ops = { #define RK3588_GMAC_TXCLK_DLY_ENABLE(id) GRF_BIT(2 * (id) + 2) #define RK3588_GMAC_TXCLK_DLY_DISABLE(id) GRF_CLR_BIT(2 * (id) + 2) -#define RK3588_GMAC_CLK_RX_DL_CFG(val) HIWORD_UPDATE(val, 0xFF, 8) -#define RK3588_GMAC_CLK_TX_DL_CFG(val) HIWORD_UPDATE(val, 0xFF, 0) +#define RK3588_GMAC_CLK_RX_DL_CFG(val) GRF_FIELD(15, 8, val) +#define RK3588_GMAC_CLK_TX_DL_CFG(val) GRF_FIELD(7, 0, val) /* php_grf */ #define RK3588_GRF_GMAC_CON0 0X0008 @@ -1400,11 +1400,11 @@ static const struct rk_gmac_ops rv1108_ops = { #define RV1126_GMAC_M1_TXCLK_DLY_DISABLE GRF_CLR_BIT(2) /* RV1126_GRF_GMAC_CON1 */ -#define RV1126_GMAC_M0_CLK_RX_DL_CFG(val) HIWORD_UPDATE(val, 0x7F, 8) -#define RV1126_GMAC_M0_CLK_TX_DL_CFG(val) HIWORD_UPDATE(val, 0x7F, 0) +#define RV1126_GMAC_M0_CLK_RX_DL_CFG(val) GRF_FIELD(14, 8, val) +#define RV1126_GMAC_M0_CLK_TX_DL_CFG(val) GRF_FIELD(6, 0, val) /* RV1126_GRF_GMAC_CON2 */ -#define RV1126_GMAC_M1_CLK_RX_DL_CFG(val) HIWORD_UPDATE(val, 0x7F, 8) -#define RV1126_GMAC_M1_CLK_TX_DL_CFG(val) HIWORD_UPDATE(val, 0x7F, 0) +#define RV1126_GMAC_M1_CLK_RX_DL_CFG(val) GRF_FIELD(14, 8, val) +#define RV1126_GMAC_M1_CLK_TX_DL_CFG(val) GRF_FIELD(6, 0, val) static void rv1126_set_to_rgmii(struct rk_priv_data *bsp_priv, int tx_delay, int rx_delay) From 764ebe423ef946daaa63336b14dc71a942507982 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Thu, 13 Nov 2025 17:46:40 +0000 Subject: [PATCH 0804/1381] net: stmmac: rk: convert all bitfields to GRF_FIELD*() Convert all bitfields to GRF_FIELD() or GRF_FIELD_CONST(), which makes the bitfield values more readable, and also allows the aarch64 compiler to produce better code. Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vJbP6-0000000EBqP-1cmm@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../net/ethernet/stmicro/stmmac/dwmac-rk.c | 136 ++++++++---------- 1 file changed, 57 insertions(+), 79 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c index 794a7ed71451..4257cc1f66e9 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c @@ -151,6 +151,8 @@ static int rk_set_clk_mac_speed(struct rk_priv_data *bsp_priv, #define GRF_FIELD(hi, lo, val) \ FIELD_PREP_WM16(GENMASK_U16(hi, lo), val) +#define GRF_FIELD_CONST(hi, lo, val) \ + FIELD_PREP_WM16_CONST(GENMASK_U16(hi, lo), val) #define GRF_BIT(nr) (BIT(nr) | BIT(nr+16)) #define GRF_CLR_BIT(nr) (BIT(nr+16)) @@ -167,7 +169,7 @@ static int rk_set_clk_mac_speed(struct rk_priv_data *bsp_priv, #define RK_MACPHY_ENABLE GRF_BIT(0) #define RK_MACPHY_DISABLE GRF_CLR_BIT(0) #define RK_MACPHY_CFG_CLK_50M GRF_BIT(14) -#define RK_GMAC2PHY_RMII_MODE (GRF_BIT(6) | GRF_CLR_BIT(7)) +#define RK_GMAC2PHY_RMII_MODE GRF_FIELD(7, 6, 1) #define RK_GRF_CON2_MACPHY_ID GRF_FIELD(15, 0, 0x1234) #define RK_GRF_CON3_MACPHY_ID GRF_FIELD(5, 0, 0x35) @@ -203,7 +205,7 @@ static void rk_gmac_integrated_ephy_powerdown(struct rk_priv_data *priv) #define RK_FEPHY_SHUTDOWN GRF_BIT(1) #define RK_FEPHY_POWERUP GRF_CLR_BIT(1) #define RK_FEPHY_INTERNAL_RMII_SEL GRF_BIT(6) -#define RK_FEPHY_24M_CLK_SEL (GRF_BIT(8) | GRF_BIT(9)) +#define RK_FEPHY_24M_CLK_SEL GRF_FIELD(9, 8, 3) #define RK_FEPHY_PHY_ID GRF_BIT(11) static void rk_gmac_integrated_fephy_powerup(struct rk_priv_data *priv, @@ -232,8 +234,7 @@ static void rk_gmac_integrated_fephy_powerdown(struct rk_priv_data *priv, #define PX30_GRF_GMAC_CON1 0x0904 /* PX30_GRF_GMAC_CON1 */ -#define PX30_GMAC_PHY_INTF_SEL_RMII (GRF_CLR_BIT(4) | GRF_CLR_BIT(5) | \ - GRF_BIT(6)) +#define PX30_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(6, 4, 4) #define PX30_GMAC_SPEED_10M GRF_CLR_BIT(2) #define PX30_GMAC_SPEED_100M GRF_BIT(2) @@ -289,19 +290,17 @@ static const struct rk_gmac_ops px30_ops = { #define RK3128_GMAC_CLK_TX_DL_CFG(val) GRF_FIELD(6, 0, val) /* RK3128_GRF_MAC_CON1 */ -#define RK3128_GMAC_PHY_INTF_SEL_RGMII \ - (GRF_BIT(6) | GRF_CLR_BIT(7) | GRF_CLR_BIT(8)) -#define RK3128_GMAC_PHY_INTF_SEL_RMII \ - (GRF_CLR_BIT(6) | GRF_CLR_BIT(7) | GRF_BIT(8)) +#define RK3128_GMAC_PHY_INTF_SEL_RGMII GRF_FIELD(8, 6, 1) +#define RK3128_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(8, 6, 4) #define RK3128_GMAC_FLOW_CTRL GRF_BIT(9) #define RK3128_GMAC_FLOW_CTRL_CLR GRF_CLR_BIT(9) #define RK3128_GMAC_SPEED_10M GRF_CLR_BIT(10) #define RK3128_GMAC_SPEED_100M GRF_BIT(10) #define RK3128_GMAC_RMII_CLK_25M GRF_BIT(11) #define RK3128_GMAC_RMII_CLK_2_5M GRF_CLR_BIT(11) -#define RK3128_GMAC_CLK_125M (GRF_CLR_BIT(12) | GRF_CLR_BIT(13)) -#define RK3128_GMAC_CLK_25M (GRF_BIT(12) | GRF_BIT(13)) -#define RK3128_GMAC_CLK_2_5M (GRF_CLR_BIT(12) | GRF_BIT(13)) +#define RK3128_GMAC_CLK_125M GRF_FIELD_CONST(13, 12, 0) +#define RK3128_GMAC_CLK_25M GRF_FIELD_CONST(13, 12, 3) +#define RK3128_GMAC_CLK_2_5M GRF_FIELD_CONST(13, 12, 2) #define RK3128_GMAC_RMII_MODE GRF_BIT(14) #define RK3128_GMAC_RMII_MODE_CLR GRF_CLR_BIT(14) @@ -354,19 +353,17 @@ static const struct rk_gmac_ops rk3128_ops = { #define RK3228_GMAC_CLK_TX_DL_CFG(val) GRF_FIELD(6, 0, val) /* RK3228_GRF_MAC_CON1 */ -#define RK3228_GMAC_PHY_INTF_SEL_RGMII \ - (GRF_BIT(4) | GRF_CLR_BIT(5) | GRF_CLR_BIT(6)) -#define RK3228_GMAC_PHY_INTF_SEL_RMII \ - (GRF_CLR_BIT(4) | GRF_CLR_BIT(5) | GRF_BIT(6)) +#define RK3228_GMAC_PHY_INTF_SEL_RGMII GRF_FIELD(6, 4, 1) +#define RK3228_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(6, 4, 4) #define RK3228_GMAC_FLOW_CTRL GRF_BIT(3) #define RK3228_GMAC_FLOW_CTRL_CLR GRF_CLR_BIT(3) #define RK3228_GMAC_SPEED_10M GRF_CLR_BIT(2) #define RK3228_GMAC_SPEED_100M GRF_BIT(2) #define RK3228_GMAC_RMII_CLK_25M GRF_BIT(7) #define RK3228_GMAC_RMII_CLK_2_5M GRF_CLR_BIT(7) -#define RK3228_GMAC_CLK_125M (GRF_CLR_BIT(8) | GRF_CLR_BIT(9)) -#define RK3228_GMAC_CLK_25M (GRF_BIT(8) | GRF_BIT(9)) -#define RK3228_GMAC_CLK_2_5M (GRF_CLR_BIT(8) | GRF_BIT(9)) +#define RK3228_GMAC_CLK_125M GRF_FIELD_CONST(9, 8, 0) +#define RK3228_GMAC_CLK_25M GRF_FIELD_CONST(9, 8, 3) +#define RK3228_GMAC_CLK_2_5M GRF_FIELD_CONST(9, 8, 2) #define RK3228_GMAC_RMII_MODE GRF_BIT(10) #define RK3228_GMAC_RMII_MODE_CLR GRF_CLR_BIT(10) #define RK3228_GMAC_TXCLK_DLY_ENABLE GRF_BIT(0) @@ -435,19 +432,17 @@ static const struct rk_gmac_ops rk3228_ops = { #define RK3288_GRF_SOC_CON3 0x0250 /*RK3288_GRF_SOC_CON1*/ -#define RK3288_GMAC_PHY_INTF_SEL_RGMII (GRF_BIT(6) | GRF_CLR_BIT(7) | \ - GRF_CLR_BIT(8)) -#define RK3288_GMAC_PHY_INTF_SEL_RMII (GRF_CLR_BIT(6) | GRF_CLR_BIT(7) | \ - GRF_BIT(8)) +#define RK3288_GMAC_PHY_INTF_SEL_RGMII GRF_FIELD(8, 6, 1) +#define RK3288_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(8, 6, 4) #define RK3288_GMAC_FLOW_CTRL GRF_BIT(9) #define RK3288_GMAC_FLOW_CTRL_CLR GRF_CLR_BIT(9) #define RK3288_GMAC_SPEED_10M GRF_CLR_BIT(10) #define RK3288_GMAC_SPEED_100M GRF_BIT(10) #define RK3288_GMAC_RMII_CLK_25M GRF_BIT(11) #define RK3288_GMAC_RMII_CLK_2_5M GRF_CLR_BIT(11) -#define RK3288_GMAC_CLK_125M (GRF_CLR_BIT(12) | GRF_CLR_BIT(13)) -#define RK3288_GMAC_CLK_25M (GRF_BIT(12) | GRF_BIT(13)) -#define RK3288_GMAC_CLK_2_5M (GRF_CLR_BIT(12) | GRF_BIT(13)) +#define RK3288_GMAC_CLK_125M GRF_FIELD_CONST(13, 12, 0) +#define RK3288_GMAC_CLK_25M GRF_FIELD_CONST(13, 12, 3) +#define RK3288_GMAC_CLK_2_5M GRF_FIELD_CONST(13, 12, 2) #define RK3288_GMAC_RMII_MODE GRF_BIT(14) #define RK3288_GMAC_RMII_MODE_CLR GRF_CLR_BIT(14) @@ -501,8 +496,7 @@ static const struct rk_gmac_ops rk3288_ops = { #define RK3308_GRF_MAC_CON0 0x04a0 /* RK3308_GRF_MAC_CON0 */ -#define RK3308_GMAC_PHY_INTF_SEL_RMII (GRF_CLR_BIT(2) | GRF_CLR_BIT(3) | \ - GRF_BIT(4)) +#define RK3308_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(4, 2, 4) #define RK3308_GMAC_FLOW_CTRL GRF_BIT(3) #define RK3308_GMAC_FLOW_CTRL_CLR GRF_CLR_BIT(3) #define RK3308_GMAC_SPEED_10M GRF_CLR_BIT(0) @@ -541,19 +535,17 @@ static const struct rk_gmac_ops rk3308_ops = { #define RK3328_GMAC_CLK_TX_DL_CFG(val) GRF_FIELD(6, 0, val) /* RK3328_GRF_MAC_CON1 */ -#define RK3328_GMAC_PHY_INTF_SEL_RGMII \ - (GRF_BIT(4) | GRF_CLR_BIT(5) | GRF_CLR_BIT(6)) -#define RK3328_GMAC_PHY_INTF_SEL_RMII \ - (GRF_CLR_BIT(4) | GRF_CLR_BIT(5) | GRF_BIT(6)) +#define RK3328_GMAC_PHY_INTF_SEL_RGMII GRF_FIELD(6, 4, 1) +#define RK3328_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(6, 4, 4) #define RK3328_GMAC_FLOW_CTRL GRF_BIT(3) #define RK3328_GMAC_FLOW_CTRL_CLR GRF_CLR_BIT(3) #define RK3328_GMAC_SPEED_10M GRF_CLR_BIT(2) #define RK3328_GMAC_SPEED_100M GRF_BIT(2) #define RK3328_GMAC_RMII_CLK_25M GRF_BIT(7) #define RK3328_GMAC_RMII_CLK_2_5M GRF_CLR_BIT(7) -#define RK3328_GMAC_CLK_125M (GRF_CLR_BIT(11) | GRF_CLR_BIT(12)) -#define RK3328_GMAC_CLK_25M (GRF_BIT(11) | GRF_BIT(12)) -#define RK3328_GMAC_CLK_2_5M (GRF_CLR_BIT(11) | GRF_BIT(12)) +#define RK3328_GMAC_CLK_125M GRF_FIELD_CONST(12, 11, 0) +#define RK3328_GMAC_CLK_25M GRF_FIELD_CONST(12, 11, 3) +#define RK3328_GMAC_CLK_2_5M GRF_FIELD_CONST(12, 11, 2) #define RK3328_GMAC_RMII_MODE GRF_BIT(9) #define RK3328_GMAC_RMII_MODE_CLR GRF_CLR_BIT(9) #define RK3328_GMAC_TXCLK_DLY_ENABLE GRF_BIT(0) @@ -630,19 +622,17 @@ static const struct rk_gmac_ops rk3328_ops = { #define RK3366_GRF_SOC_CON7 0x041c /* RK3366_GRF_SOC_CON6 */ -#define RK3366_GMAC_PHY_INTF_SEL_RGMII (GRF_BIT(9) | GRF_CLR_BIT(10) | \ - GRF_CLR_BIT(11)) -#define RK3366_GMAC_PHY_INTF_SEL_RMII (GRF_CLR_BIT(9) | GRF_CLR_BIT(10) | \ - GRF_BIT(11)) +#define RK3366_GMAC_PHY_INTF_SEL_RGMII GRF_FIELD(11, 9, 1) +#define RK3366_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(11, 9, 4) #define RK3366_GMAC_FLOW_CTRL GRF_BIT(8) #define RK3366_GMAC_FLOW_CTRL_CLR GRF_CLR_BIT(8) #define RK3366_GMAC_SPEED_10M GRF_CLR_BIT(7) #define RK3366_GMAC_SPEED_100M GRF_BIT(7) #define RK3366_GMAC_RMII_CLK_25M GRF_BIT(3) #define RK3366_GMAC_RMII_CLK_2_5M GRF_CLR_BIT(3) -#define RK3366_GMAC_CLK_125M (GRF_CLR_BIT(4) | GRF_CLR_BIT(5)) -#define RK3366_GMAC_CLK_25M (GRF_BIT(4) | GRF_BIT(5)) -#define RK3366_GMAC_CLK_2_5M (GRF_CLR_BIT(4) | GRF_BIT(5)) +#define RK3366_GMAC_CLK_125M GRF_FIELD_CONST(5, 4, 0) +#define RK3366_GMAC_CLK_25M GRF_FIELD_CONST(5, 4, 3) +#define RK3366_GMAC_CLK_2_5M GRF_FIELD_CONST(5, 4, 2) #define RK3366_GMAC_RMII_MODE GRF_BIT(6) #define RK3366_GMAC_RMII_MODE_CLR GRF_CLR_BIT(6) @@ -697,19 +687,17 @@ static const struct rk_gmac_ops rk3366_ops = { #define RK3368_GRF_SOC_CON16 0x0440 /* RK3368_GRF_SOC_CON15 */ -#define RK3368_GMAC_PHY_INTF_SEL_RGMII (GRF_BIT(9) | GRF_CLR_BIT(10) | \ - GRF_CLR_BIT(11)) -#define RK3368_GMAC_PHY_INTF_SEL_RMII (GRF_CLR_BIT(9) | GRF_CLR_BIT(10) | \ - GRF_BIT(11)) +#define RK3368_GMAC_PHY_INTF_SEL_RGMII GRF_FIELD(11, 9, 1) +#define RK3368_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(11, 9, 4) #define RK3368_GMAC_FLOW_CTRL GRF_BIT(8) #define RK3368_GMAC_FLOW_CTRL_CLR GRF_CLR_BIT(8) #define RK3368_GMAC_SPEED_10M GRF_CLR_BIT(7) #define RK3368_GMAC_SPEED_100M GRF_BIT(7) #define RK3368_GMAC_RMII_CLK_25M GRF_BIT(3) #define RK3368_GMAC_RMII_CLK_2_5M GRF_CLR_BIT(3) -#define RK3368_GMAC_CLK_125M (GRF_CLR_BIT(4) | GRF_CLR_BIT(5)) -#define RK3368_GMAC_CLK_25M (GRF_BIT(4) | GRF_BIT(5)) -#define RK3368_GMAC_CLK_2_5M (GRF_CLR_BIT(4) | GRF_BIT(5)) +#define RK3368_GMAC_CLK_125M GRF_FIELD_CONST(5, 4, 0) +#define RK3368_GMAC_CLK_25M GRF_FIELD_CONST(5, 4, 3) +#define RK3368_GMAC_CLK_2_5M GRF_FIELD_CONST(5, 4, 2) #define RK3368_GMAC_RMII_MODE GRF_BIT(6) #define RK3368_GMAC_RMII_MODE_CLR GRF_CLR_BIT(6) @@ -764,19 +752,17 @@ static const struct rk_gmac_ops rk3368_ops = { #define RK3399_GRF_SOC_CON6 0xc218 /* RK3399_GRF_SOC_CON5 */ -#define RK3399_GMAC_PHY_INTF_SEL_RGMII (GRF_BIT(9) | GRF_CLR_BIT(10) | \ - GRF_CLR_BIT(11)) -#define RK3399_GMAC_PHY_INTF_SEL_RMII (GRF_CLR_BIT(9) | GRF_CLR_BIT(10) | \ - GRF_BIT(11)) +#define RK3399_GMAC_PHY_INTF_SEL_RGMII GRF_FIELD(11, 9, 1) +#define RK3399_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(11, 9, 4) #define RK3399_GMAC_FLOW_CTRL GRF_BIT(8) #define RK3399_GMAC_FLOW_CTRL_CLR GRF_CLR_BIT(8) #define RK3399_GMAC_SPEED_10M GRF_CLR_BIT(7) #define RK3399_GMAC_SPEED_100M GRF_BIT(7) #define RK3399_GMAC_RMII_CLK_25M GRF_BIT(3) #define RK3399_GMAC_RMII_CLK_2_5M GRF_CLR_BIT(3) -#define RK3399_GMAC_CLK_125M (GRF_CLR_BIT(4) | GRF_CLR_BIT(5)) -#define RK3399_GMAC_CLK_25M (GRF_BIT(4) | GRF_BIT(5)) -#define RK3399_GMAC_CLK_2_5M (GRF_CLR_BIT(4) | GRF_BIT(5)) +#define RK3399_GMAC_CLK_125M GRF_FIELD_CONST(5, 4, 0) +#define RK3399_GMAC_CLK_25M GRF_FIELD_CONST(5, 4, 3) +#define RK3399_GMAC_CLK_2_5M GRF_FIELD_CONST(5, 4, 2) #define RK3399_GMAC_RMII_MODE GRF_BIT(6) #define RK3399_GMAC_RMII_MODE_CLR GRF_CLR_BIT(6) @@ -916,9 +902,9 @@ static const struct rk_gmac_ops rk3506_ops = { #define RK3528_GMAC1_CLK_RMII_DIV2 GRF_BIT(10) #define RK3528_GMAC1_CLK_RMII_DIV20 GRF_CLR_BIT(10) -#define RK3528_GMAC1_CLK_RGMII_DIV1 (GRF_CLR_BIT(11) | GRF_CLR_BIT(10)) -#define RK3528_GMAC1_CLK_RGMII_DIV5 (GRF_BIT(11) | GRF_BIT(10)) -#define RK3528_GMAC1_CLK_RGMII_DIV50 (GRF_BIT(11) | GRF_CLR_BIT(10)) +#define RK3528_GMAC1_CLK_RGMII_DIV1 GRF_FIELD_CONST(11, 10, 0) +#define RK3528_GMAC1_CLK_RGMII_DIV5 GRF_FIELD_CONST(11, 10, 3) +#define RK3528_GMAC1_CLK_RGMII_DIV50 GRF_FIELD_CONST(11, 10, 2) #define RK3528_GMAC0_CLK_RMII_GATE GRF_BIT(2) #define RK3528_GMAC0_CLK_RMII_NOGATE GRF_CLR_BIT(2) @@ -1029,10 +1015,8 @@ static const struct rk_gmac_ops rk3528_ops = { #define RK3568_GRF_GMAC1_CON1 0x038c /* RK3568_GRF_GMAC0_CON1 && RK3568_GRF_GMAC1_CON1 */ -#define RK3568_GMAC_PHY_INTF_SEL_RGMII \ - (GRF_BIT(4) | GRF_CLR_BIT(5) | GRF_CLR_BIT(6)) -#define RK3568_GMAC_PHY_INTF_SEL_RMII \ - (GRF_CLR_BIT(4) | GRF_CLR_BIT(5) | GRF_BIT(6)) +#define RK3568_GMAC_PHY_INTF_SEL_RGMII GRF_FIELD(6, 4, 1) +#define RK3568_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(6, 4, 4) #define RK3568_GMAC_FLOW_CTRL GRF_BIT(3) #define RK3568_GMAC_FLOW_CTRL_CLR GRF_CLR_BIT(3) #define RK3568_GMAC_RXCLK_DLY_ENABLE GRF_BIT(1) @@ -1112,12 +1096,9 @@ static const struct rk_gmac_ops rk3568_ops = { #define RK3576_GMAC_CLK_RMII_DIV2 GRF_BIT(5) #define RK3576_GMAC_CLK_RMII_DIV20 GRF_CLR_BIT(5) -#define RK3576_GMAC_CLK_RGMII_DIV1 \ - (GRF_CLR_BIT(6) | GRF_CLR_BIT(5)) -#define RK3576_GMAC_CLK_RGMII_DIV5 \ - (GRF_BIT(6) | GRF_BIT(5)) -#define RK3576_GMAC_CLK_RGMII_DIV50 \ - (GRF_BIT(6) | GRF_CLR_BIT(5)) +#define RK3576_GMAC_CLK_RGMII_DIV1 GRF_FIELD_CONST(6, 5, 0) +#define RK3576_GMAC_CLK_RGMII_DIV5 GRF_FIELD_CONST(6, 5, 3) +#define RK3576_GMAC_CLK_RGMII_DIV50 GRF_FIELD_CONST(6, 5, 2) #define RK3576_GMAC_CLK_RMII_GATE GRF_BIT(4) #define RK3576_GMAC_CLK_RMII_NOGATE GRF_CLR_BIT(4) @@ -1228,9 +1209,9 @@ static const struct rk_gmac_ops rk3576_ops = { #define RK3588_GRF_CLK_CON1 0X0070 #define RK3588_GMAC_PHY_INTF_SEL_RGMII(id) \ - (GRF_BIT(3 + (id) * 6) | GRF_CLR_BIT(4 + (id) * 6) | GRF_CLR_BIT(5 + (id) * 6)) + (GRF_FIELD(5, 3, 1) << ((id) * 6)) #define RK3588_GMAC_PHY_INTF_SEL_RMII(id) \ - (GRF_CLR_BIT(3 + (id) * 6) | GRF_CLR_BIT(4 + (id) * 6) | GRF_BIT(5 + (id) * 6)) + (GRF_FIELD(5, 3, 4) << ((id) * 6)) #define RK3588_GMAC_CLK_RMII_MODE(id) GRF_BIT(5 * (id)) #define RK3588_GMAC_CLK_RGMII_MODE(id) GRF_CLR_BIT(5 * (id)) @@ -1242,11 +1223,11 @@ static const struct rk_gmac_ops rk3576_ops = { #define RK3588_GMA_CLK_RMII_DIV20(id) GRF_CLR_BIT(5 * (id) + 2) #define RK3588_GMAC_CLK_RGMII_DIV1(id) \ - (GRF_CLR_BIT(5 * (id) + 2) | GRF_CLR_BIT(5 * (id) + 3)) + (GRF_FIELD_CONST(3, 2, 0) << ((id) * 5)) #define RK3588_GMAC_CLK_RGMII_DIV5(id) \ - (GRF_BIT(5 * (id) + 2) | GRF_BIT(5 * (id) + 3)) + (GRF_FIELD_CONST(3, 2, 3) << ((id) * 5)) #define RK3588_GMAC_CLK_RGMII_DIV50(id) \ - (GRF_CLR_BIT(5 * (id) + 2) | GRF_BIT(5 * (id) + 3)) + (GRF_FIELD_CONST(3, 2, 2) << ((id) * 5)) #define RK3588_GMAC_CLK_RMII_GATE(id) GRF_BIT(5 * (id) + 1) #define RK3588_GMAC_CLK_RMII_NOGATE(id) GRF_CLR_BIT(5 * (id) + 1) @@ -1347,8 +1328,7 @@ static const struct rk_gmac_ops rk3588_ops = { #define RV1108_GRF_GMAC_CON0 0X0900 /* RV1108_GRF_GMAC_CON0 */ -#define RV1108_GMAC_PHY_INTF_SEL_RMII (GRF_CLR_BIT(4) | GRF_CLR_BIT(5) | \ - GRF_BIT(6)) +#define RV1108_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(6, 4, 4) #define RV1108_GMAC_FLOW_CTRL GRF_BIT(3) #define RV1108_GMAC_FLOW_CTRL_CLR GRF_CLR_BIT(3) #define RV1108_GMAC_SPEED_10M GRF_CLR_BIT(2) @@ -1384,10 +1364,8 @@ static const struct rk_gmac_ops rv1108_ops = { #define RV1126_GRF_GMAC_CON2 0X0078 /* RV1126_GRF_GMAC_CON0 */ -#define RV1126_GMAC_PHY_INTF_SEL_RGMII \ - (GRF_BIT(4) | GRF_CLR_BIT(5) | GRF_CLR_BIT(6)) -#define RV1126_GMAC_PHY_INTF_SEL_RMII \ - (GRF_CLR_BIT(4) | GRF_CLR_BIT(5) | GRF_BIT(6)) +#define RV1126_GMAC_PHY_INTF_SEL_RGMII GRF_FIELD(6, 4, 1) +#define RV1126_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(6, 4, 4) #define RV1126_GMAC_FLOW_CTRL GRF_BIT(7) #define RV1126_GMAC_FLOW_CTRL_CLR GRF_CLR_BIT(7) #define RV1126_GMAC_M0_RXCLK_DLY_ENABLE GRF_BIT(1) From 5e37047f745b1a705ba7a1b580b5e5fe2bfd5993 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Thu, 13 Nov 2025 17:46:45 +0000 Subject: [PATCH 0805/1381] net: stmmac: rk: use PHY_INTF_SEL_x constants The values used in the xxx_GMAC_PHY_INTF_SEL_xxx() macros are the phy_intf_sel values used for the dwmac core. Use these to define these constants. No change to produced code on aarch64. Signed-off-by: Russell King (Oracle) Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/E1vJbPB-0000000EBqV-27GS@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../net/ethernet/stmicro/stmmac/dwmac-rk.c | 46 +++++++++---------- 1 file changed, 23 insertions(+), 23 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c index 4257cc1f66e9..49076ee00877 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c @@ -234,7 +234,7 @@ static void rk_gmac_integrated_fephy_powerdown(struct rk_priv_data *priv, #define PX30_GRF_GMAC_CON1 0x0904 /* PX30_GRF_GMAC_CON1 */ -#define PX30_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(6, 4, 4) +#define PX30_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(6, 4, PHY_INTF_SEL_RMII) #define PX30_GMAC_SPEED_10M GRF_CLR_BIT(2) #define PX30_GMAC_SPEED_100M GRF_BIT(2) @@ -290,8 +290,8 @@ static const struct rk_gmac_ops px30_ops = { #define RK3128_GMAC_CLK_TX_DL_CFG(val) GRF_FIELD(6, 0, val) /* RK3128_GRF_MAC_CON1 */ -#define RK3128_GMAC_PHY_INTF_SEL_RGMII GRF_FIELD(8, 6, 1) -#define RK3128_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(8, 6, 4) +#define RK3128_GMAC_PHY_INTF_SEL_RGMII GRF_FIELD(8, 6, PHY_INTF_SEL_RGMII) +#define RK3128_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(8, 6, PHY_INTF_SEL_RMII) #define RK3128_GMAC_FLOW_CTRL GRF_BIT(9) #define RK3128_GMAC_FLOW_CTRL_CLR GRF_CLR_BIT(9) #define RK3128_GMAC_SPEED_10M GRF_CLR_BIT(10) @@ -353,8 +353,8 @@ static const struct rk_gmac_ops rk3128_ops = { #define RK3228_GMAC_CLK_TX_DL_CFG(val) GRF_FIELD(6, 0, val) /* RK3228_GRF_MAC_CON1 */ -#define RK3228_GMAC_PHY_INTF_SEL_RGMII GRF_FIELD(6, 4, 1) -#define RK3228_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(6, 4, 4) +#define RK3228_GMAC_PHY_INTF_SEL_RGMII GRF_FIELD(6, 4, PHY_INTF_SEL_RGMII) +#define RK3228_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(6, 4, PHY_INTF_SEL_RMII) #define RK3228_GMAC_FLOW_CTRL GRF_BIT(3) #define RK3228_GMAC_FLOW_CTRL_CLR GRF_CLR_BIT(3) #define RK3228_GMAC_SPEED_10M GRF_CLR_BIT(2) @@ -432,8 +432,8 @@ static const struct rk_gmac_ops rk3228_ops = { #define RK3288_GRF_SOC_CON3 0x0250 /*RK3288_GRF_SOC_CON1*/ -#define RK3288_GMAC_PHY_INTF_SEL_RGMII GRF_FIELD(8, 6, 1) -#define RK3288_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(8, 6, 4) +#define RK3288_GMAC_PHY_INTF_SEL_RGMII GRF_FIELD(8, 6, PHY_INTF_SEL_RGMII) +#define RK3288_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(8, 6, PHY_INTF_SEL_RMII) #define RK3288_GMAC_FLOW_CTRL GRF_BIT(9) #define RK3288_GMAC_FLOW_CTRL_CLR GRF_CLR_BIT(9) #define RK3288_GMAC_SPEED_10M GRF_CLR_BIT(10) @@ -496,7 +496,7 @@ static const struct rk_gmac_ops rk3288_ops = { #define RK3308_GRF_MAC_CON0 0x04a0 /* RK3308_GRF_MAC_CON0 */ -#define RK3308_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(4, 2, 4) +#define RK3308_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(4, 2, PHY_INTF_SEL_RMII) #define RK3308_GMAC_FLOW_CTRL GRF_BIT(3) #define RK3308_GMAC_FLOW_CTRL_CLR GRF_CLR_BIT(3) #define RK3308_GMAC_SPEED_10M GRF_CLR_BIT(0) @@ -535,8 +535,8 @@ static const struct rk_gmac_ops rk3308_ops = { #define RK3328_GMAC_CLK_TX_DL_CFG(val) GRF_FIELD(6, 0, val) /* RK3328_GRF_MAC_CON1 */ -#define RK3328_GMAC_PHY_INTF_SEL_RGMII GRF_FIELD(6, 4, 1) -#define RK3328_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(6, 4, 4) +#define RK3328_GMAC_PHY_INTF_SEL_RGMII GRF_FIELD(6, 4, PHY_INTF_SEL_RGMII) +#define RK3328_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(6, 4, PHY_INTF_SEL_RMII) #define RK3328_GMAC_FLOW_CTRL GRF_BIT(3) #define RK3328_GMAC_FLOW_CTRL_CLR GRF_CLR_BIT(3) #define RK3328_GMAC_SPEED_10M GRF_CLR_BIT(2) @@ -622,8 +622,8 @@ static const struct rk_gmac_ops rk3328_ops = { #define RK3366_GRF_SOC_CON7 0x041c /* RK3366_GRF_SOC_CON6 */ -#define RK3366_GMAC_PHY_INTF_SEL_RGMII GRF_FIELD(11, 9, 1) -#define RK3366_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(11, 9, 4) +#define RK3366_GMAC_PHY_INTF_SEL_RGMII GRF_FIELD(11, 9, PHY_INTF_SEL_RGMII) +#define RK3366_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(11, 9, PHY_INTF_SEL_RMII) #define RK3366_GMAC_FLOW_CTRL GRF_BIT(8) #define RK3366_GMAC_FLOW_CTRL_CLR GRF_CLR_BIT(8) #define RK3366_GMAC_SPEED_10M GRF_CLR_BIT(7) @@ -687,8 +687,8 @@ static const struct rk_gmac_ops rk3366_ops = { #define RK3368_GRF_SOC_CON16 0x0440 /* RK3368_GRF_SOC_CON15 */ -#define RK3368_GMAC_PHY_INTF_SEL_RGMII GRF_FIELD(11, 9, 1) -#define RK3368_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(11, 9, 4) +#define RK3368_GMAC_PHY_INTF_SEL_RGMII GRF_FIELD(11, 9, PHY_INTF_SEL_RGMII) +#define RK3368_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(11, 9, PHY_INTF_SEL_RMII) #define RK3368_GMAC_FLOW_CTRL GRF_BIT(8) #define RK3368_GMAC_FLOW_CTRL_CLR GRF_CLR_BIT(8) #define RK3368_GMAC_SPEED_10M GRF_CLR_BIT(7) @@ -752,8 +752,8 @@ static const struct rk_gmac_ops rk3368_ops = { #define RK3399_GRF_SOC_CON6 0xc218 /* RK3399_GRF_SOC_CON5 */ -#define RK3399_GMAC_PHY_INTF_SEL_RGMII GRF_FIELD(11, 9, 1) -#define RK3399_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(11, 9, 4) +#define RK3399_GMAC_PHY_INTF_SEL_RGMII GRF_FIELD(11, 9, PHY_INTF_SEL_RGMII) +#define RK3399_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(11, 9, PHY_INTF_SEL_RMII) #define RK3399_GMAC_FLOW_CTRL GRF_BIT(8) #define RK3399_GMAC_FLOW_CTRL_CLR GRF_CLR_BIT(8) #define RK3399_GMAC_SPEED_10M GRF_CLR_BIT(7) @@ -1015,8 +1015,8 @@ static const struct rk_gmac_ops rk3528_ops = { #define RK3568_GRF_GMAC1_CON1 0x038c /* RK3568_GRF_GMAC0_CON1 && RK3568_GRF_GMAC1_CON1 */ -#define RK3568_GMAC_PHY_INTF_SEL_RGMII GRF_FIELD(6, 4, 1) -#define RK3568_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(6, 4, 4) +#define RK3568_GMAC_PHY_INTF_SEL_RGMII GRF_FIELD(6, 4, PHY_INTF_SEL_RGMII) +#define RK3568_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(6, 4, PHY_INTF_SEL_RMII) #define RK3568_GMAC_FLOW_CTRL GRF_BIT(3) #define RK3568_GMAC_FLOW_CTRL_CLR GRF_CLR_BIT(3) #define RK3568_GMAC_RXCLK_DLY_ENABLE GRF_BIT(1) @@ -1209,9 +1209,9 @@ static const struct rk_gmac_ops rk3576_ops = { #define RK3588_GRF_CLK_CON1 0X0070 #define RK3588_GMAC_PHY_INTF_SEL_RGMII(id) \ - (GRF_FIELD(5, 3, 1) << ((id) * 6)) + (GRF_FIELD(5, 3, PHY_INTF_SEL_RGMII) << ((id) * 6)) #define RK3588_GMAC_PHY_INTF_SEL_RMII(id) \ - (GRF_FIELD(5, 3, 4) << ((id) * 6)) + (GRF_FIELD(5, 3, PHY_INTF_SEL_RMII) << ((id) * 6)) #define RK3588_GMAC_CLK_RMII_MODE(id) GRF_BIT(5 * (id)) #define RK3588_GMAC_CLK_RGMII_MODE(id) GRF_CLR_BIT(5 * (id)) @@ -1328,7 +1328,7 @@ static const struct rk_gmac_ops rk3588_ops = { #define RV1108_GRF_GMAC_CON0 0X0900 /* RV1108_GRF_GMAC_CON0 */ -#define RV1108_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(6, 4, 4) +#define RV1108_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(6, 4, PHY_INTF_SEL_RMII) #define RV1108_GMAC_FLOW_CTRL GRF_BIT(3) #define RV1108_GMAC_FLOW_CTRL_CLR GRF_CLR_BIT(3) #define RV1108_GMAC_SPEED_10M GRF_CLR_BIT(2) @@ -1364,8 +1364,8 @@ static const struct rk_gmac_ops rv1108_ops = { #define RV1126_GRF_GMAC_CON2 0X0078 /* RV1126_GRF_GMAC_CON0 */ -#define RV1126_GMAC_PHY_INTF_SEL_RGMII GRF_FIELD(6, 4, 1) -#define RV1126_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(6, 4, 4) +#define RV1126_GMAC_PHY_INTF_SEL_RGMII GRF_FIELD(6, 4, PHY_INTF_SEL_RGMII) +#define RV1126_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(6, 4, PHY_INTF_SEL_RMII) #define RV1126_GMAC_FLOW_CTRL GRF_BIT(7) #define RV1126_GMAC_FLOW_CTRL_CLR GRF_CLR_BIT(7) #define RV1126_GMAC_M0_RXCLK_DLY_ENABLE GRF_BIT(1) From 1188741cb5a29187a93ee908274f7034e81486f6 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Thu, 13 Nov 2025 17:46:50 +0000 Subject: [PATCH 0806/1381] net: stmmac: rk: use PHY_INTF_SEL_x in functions Rather than defining one xxx_GMAC_PHY_INTF_SEL_xxx() for each mode, define xxx_GMAC_PHY_INTF_SEL() which takes the phy_intf_sel value. Pass the appropriate value into these new macros in the set_to_xxx() methods. No change to produced code on aarch64. Signed-off-by: Russell King (Oracle) Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/E1vJbPG-0000000EBqb-2cF2@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../net/ethernet/stmicro/stmmac/dwmac-rk.c | 91 +++++++++---------- 1 file changed, 43 insertions(+), 48 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c index 49076ee00877..6e75da577af5 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c @@ -234,14 +234,14 @@ static void rk_gmac_integrated_fephy_powerdown(struct rk_priv_data *priv, #define PX30_GRF_GMAC_CON1 0x0904 /* PX30_GRF_GMAC_CON1 */ -#define PX30_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(6, 4, PHY_INTF_SEL_RMII) +#define PX30_GMAC_PHY_INTF_SEL(val) GRF_FIELD(6, 4, val) #define PX30_GMAC_SPEED_10M GRF_CLR_BIT(2) #define PX30_GMAC_SPEED_100M GRF_BIT(2) static void px30_set_to_rmii(struct rk_priv_data *bsp_priv) { regmap_write(bsp_priv->grf, PX30_GRF_GMAC_CON1, - PX30_GMAC_PHY_INTF_SEL_RMII); + PX30_GMAC_PHY_INTF_SEL(PHY_INTF_SEL_RMII)); } static int px30_set_speed(struct rk_priv_data *bsp_priv, @@ -290,8 +290,7 @@ static const struct rk_gmac_ops px30_ops = { #define RK3128_GMAC_CLK_TX_DL_CFG(val) GRF_FIELD(6, 0, val) /* RK3128_GRF_MAC_CON1 */ -#define RK3128_GMAC_PHY_INTF_SEL_RGMII GRF_FIELD(8, 6, PHY_INTF_SEL_RGMII) -#define RK3128_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(8, 6, PHY_INTF_SEL_RMII) +#define RK3128_GMAC_PHY_INTF_SEL(val) GRF_FIELD(8, 6, val) #define RK3128_GMAC_FLOW_CTRL GRF_BIT(9) #define RK3128_GMAC_FLOW_CTRL_CLR GRF_CLR_BIT(9) #define RK3128_GMAC_SPEED_10M GRF_CLR_BIT(10) @@ -308,7 +307,7 @@ static void rk3128_set_to_rgmii(struct rk_priv_data *bsp_priv, int tx_delay, int rx_delay) { regmap_write(bsp_priv->grf, RK3128_GRF_MAC_CON1, - RK3128_GMAC_PHY_INTF_SEL_RGMII | + RK3128_GMAC_PHY_INTF_SEL(PHY_INTF_SEL_RGMII) | RK3128_GMAC_RMII_MODE_CLR); regmap_write(bsp_priv->grf, RK3128_GRF_MAC_CON0, DELAY_ENABLE(RK3128, tx_delay, rx_delay) | @@ -319,7 +318,8 @@ static void rk3128_set_to_rgmii(struct rk_priv_data *bsp_priv, static void rk3128_set_to_rmii(struct rk_priv_data *bsp_priv) { regmap_write(bsp_priv->grf, RK3128_GRF_MAC_CON1, - RK3128_GMAC_PHY_INTF_SEL_RMII | RK3128_GMAC_RMII_MODE); + RK3128_GMAC_PHY_INTF_SEL(PHY_INTF_SEL_RMII) | + RK3128_GMAC_RMII_MODE); } static const struct rk_reg_speed_data rk3128_reg_speed_data = { @@ -353,8 +353,7 @@ static const struct rk_gmac_ops rk3128_ops = { #define RK3228_GMAC_CLK_TX_DL_CFG(val) GRF_FIELD(6, 0, val) /* RK3228_GRF_MAC_CON1 */ -#define RK3228_GMAC_PHY_INTF_SEL_RGMII GRF_FIELD(6, 4, PHY_INTF_SEL_RGMII) -#define RK3228_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(6, 4, PHY_INTF_SEL_RMII) +#define RK3228_GMAC_PHY_INTF_SEL(val) GRF_FIELD(6, 4, val) #define RK3228_GMAC_FLOW_CTRL GRF_BIT(3) #define RK3228_GMAC_FLOW_CTRL_CLR GRF_CLR_BIT(3) #define RK3228_GMAC_SPEED_10M GRF_CLR_BIT(2) @@ -378,7 +377,7 @@ static void rk3228_set_to_rgmii(struct rk_priv_data *bsp_priv, int tx_delay, int rx_delay) { regmap_write(bsp_priv->grf, RK3228_GRF_MAC_CON1, - RK3228_GMAC_PHY_INTF_SEL_RGMII | + RK3228_GMAC_PHY_INTF_SEL(PHY_INTF_SEL_RGMII) | RK3228_GMAC_RMII_MODE_CLR | DELAY_ENABLE(RK3228, tx_delay, rx_delay)); @@ -390,7 +389,7 @@ static void rk3228_set_to_rgmii(struct rk_priv_data *bsp_priv, static void rk3228_set_to_rmii(struct rk_priv_data *bsp_priv) { regmap_write(bsp_priv->grf, RK3228_GRF_MAC_CON1, - RK3228_GMAC_PHY_INTF_SEL_RMII | + RK3228_GMAC_PHY_INTF_SEL(PHY_INTF_SEL_RMII) | RK3228_GMAC_RMII_MODE); /* set MAC to RMII mode */ @@ -432,8 +431,7 @@ static const struct rk_gmac_ops rk3228_ops = { #define RK3288_GRF_SOC_CON3 0x0250 /*RK3288_GRF_SOC_CON1*/ -#define RK3288_GMAC_PHY_INTF_SEL_RGMII GRF_FIELD(8, 6, PHY_INTF_SEL_RGMII) -#define RK3288_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(8, 6, PHY_INTF_SEL_RMII) +#define RK3288_GMAC_PHY_INTF_SEL(val) GRF_FIELD(8, 6, val) #define RK3288_GMAC_FLOW_CTRL GRF_BIT(9) #define RK3288_GMAC_FLOW_CTRL_CLR GRF_CLR_BIT(9) #define RK3288_GMAC_SPEED_10M GRF_CLR_BIT(10) @@ -458,7 +456,7 @@ static void rk3288_set_to_rgmii(struct rk_priv_data *bsp_priv, int tx_delay, int rx_delay) { regmap_write(bsp_priv->grf, RK3288_GRF_SOC_CON1, - RK3288_GMAC_PHY_INTF_SEL_RGMII | + RK3288_GMAC_PHY_INTF_SEL(PHY_INTF_SEL_RGMII) | RK3288_GMAC_RMII_MODE_CLR); regmap_write(bsp_priv->grf, RK3288_GRF_SOC_CON3, DELAY_ENABLE(RK3288, tx_delay, rx_delay) | @@ -469,7 +467,8 @@ static void rk3288_set_to_rgmii(struct rk_priv_data *bsp_priv, static void rk3288_set_to_rmii(struct rk_priv_data *bsp_priv) { regmap_write(bsp_priv->grf, RK3288_GRF_SOC_CON1, - RK3288_GMAC_PHY_INTF_SEL_RMII | RK3288_GMAC_RMII_MODE); + RK3288_GMAC_PHY_INTF_SEL(PHY_INTF_SEL_RMII) | + RK3288_GMAC_RMII_MODE); } static const struct rk_reg_speed_data rk3288_reg_speed_data = { @@ -496,7 +495,7 @@ static const struct rk_gmac_ops rk3288_ops = { #define RK3308_GRF_MAC_CON0 0x04a0 /* RK3308_GRF_MAC_CON0 */ -#define RK3308_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(4, 2, PHY_INTF_SEL_RMII) +#define RK3308_GMAC_PHY_INTF_SEL(val) GRF_FIELD(4, 2, val) #define RK3308_GMAC_FLOW_CTRL GRF_BIT(3) #define RK3308_GMAC_FLOW_CTRL_CLR GRF_CLR_BIT(3) #define RK3308_GMAC_SPEED_10M GRF_CLR_BIT(0) @@ -505,7 +504,7 @@ static const struct rk_gmac_ops rk3288_ops = { static void rk3308_set_to_rmii(struct rk_priv_data *bsp_priv) { regmap_write(bsp_priv->grf, RK3308_GRF_MAC_CON0, - RK3308_GMAC_PHY_INTF_SEL_RMII); + RK3308_GMAC_PHY_INTF_SEL(PHY_INTF_SEL_RMII)); } static const struct rk_reg_speed_data rk3308_reg_speed_data = { @@ -535,8 +534,7 @@ static const struct rk_gmac_ops rk3308_ops = { #define RK3328_GMAC_CLK_TX_DL_CFG(val) GRF_FIELD(6, 0, val) /* RK3328_GRF_MAC_CON1 */ -#define RK3328_GMAC_PHY_INTF_SEL_RGMII GRF_FIELD(6, 4, PHY_INTF_SEL_RGMII) -#define RK3328_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(6, 4, PHY_INTF_SEL_RMII) +#define RK3328_GMAC_PHY_INTF_SEL(val) GRF_FIELD(6, 4, val) #define RK3328_GMAC_FLOW_CTRL GRF_BIT(3) #define RK3328_GMAC_FLOW_CTRL_CLR GRF_CLR_BIT(3) #define RK3328_GMAC_SPEED_10M GRF_CLR_BIT(2) @@ -558,7 +556,7 @@ static void rk3328_set_to_rgmii(struct rk_priv_data *bsp_priv, int tx_delay, int rx_delay) { regmap_write(bsp_priv->grf, RK3328_GRF_MAC_CON1, - RK3328_GMAC_PHY_INTF_SEL_RGMII | + RK3328_GMAC_PHY_INTF_SEL(PHY_INTF_SEL_RGMII) | RK3328_GMAC_RMII_MODE_CLR | RK3328_GMAC_RXCLK_DLY_ENABLE | RK3328_GMAC_TXCLK_DLY_ENABLE); @@ -576,7 +574,7 @@ static void rk3328_set_to_rmii(struct rk_priv_data *bsp_priv) RK3328_GRF_MAC_CON1; regmap_write(bsp_priv->grf, reg, - RK3328_GMAC_PHY_INTF_SEL_RMII | + RK3328_GMAC_PHY_INTF_SEL(PHY_INTF_SEL_RMII) | RK3328_GMAC_RMII_MODE); } @@ -622,8 +620,7 @@ static const struct rk_gmac_ops rk3328_ops = { #define RK3366_GRF_SOC_CON7 0x041c /* RK3366_GRF_SOC_CON6 */ -#define RK3366_GMAC_PHY_INTF_SEL_RGMII GRF_FIELD(11, 9, PHY_INTF_SEL_RGMII) -#define RK3366_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(11, 9, PHY_INTF_SEL_RMII) +#define RK3366_GMAC_PHY_INTF_SEL(val) GRF_FIELD(11, 9, val) #define RK3366_GMAC_FLOW_CTRL GRF_BIT(8) #define RK3366_GMAC_FLOW_CTRL_CLR GRF_CLR_BIT(8) #define RK3366_GMAC_SPEED_10M GRF_CLR_BIT(7) @@ -648,7 +645,7 @@ static void rk3366_set_to_rgmii(struct rk_priv_data *bsp_priv, int tx_delay, int rx_delay) { regmap_write(bsp_priv->grf, RK3366_GRF_SOC_CON6, - RK3366_GMAC_PHY_INTF_SEL_RGMII | + RK3366_GMAC_PHY_INTF_SEL(PHY_INTF_SEL_RGMII) | RK3366_GMAC_RMII_MODE_CLR); regmap_write(bsp_priv->grf, RK3366_GRF_SOC_CON7, DELAY_ENABLE(RK3366, tx_delay, rx_delay) | @@ -659,7 +656,8 @@ static void rk3366_set_to_rgmii(struct rk_priv_data *bsp_priv, static void rk3366_set_to_rmii(struct rk_priv_data *bsp_priv) { regmap_write(bsp_priv->grf, RK3366_GRF_SOC_CON6, - RK3366_GMAC_PHY_INTF_SEL_RMII | RK3366_GMAC_RMII_MODE); + RK3366_GMAC_PHY_INTF_SEL(PHY_INTF_SEL_RMII) | + RK3366_GMAC_RMII_MODE); } static const struct rk_reg_speed_data rk3366_reg_speed_data = { @@ -687,8 +685,7 @@ static const struct rk_gmac_ops rk3366_ops = { #define RK3368_GRF_SOC_CON16 0x0440 /* RK3368_GRF_SOC_CON15 */ -#define RK3368_GMAC_PHY_INTF_SEL_RGMII GRF_FIELD(11, 9, PHY_INTF_SEL_RGMII) -#define RK3368_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(11, 9, PHY_INTF_SEL_RMII) +#define RK3368_GMAC_PHY_INTF_SEL(val) GRF_FIELD(11, 9, val) #define RK3368_GMAC_FLOW_CTRL GRF_BIT(8) #define RK3368_GMAC_FLOW_CTRL_CLR GRF_CLR_BIT(8) #define RK3368_GMAC_SPEED_10M GRF_CLR_BIT(7) @@ -713,7 +710,7 @@ static void rk3368_set_to_rgmii(struct rk_priv_data *bsp_priv, int tx_delay, int rx_delay) { regmap_write(bsp_priv->grf, RK3368_GRF_SOC_CON15, - RK3368_GMAC_PHY_INTF_SEL_RGMII | + RK3368_GMAC_PHY_INTF_SEL(PHY_INTF_SEL_RGMII) | RK3368_GMAC_RMII_MODE_CLR); regmap_write(bsp_priv->grf, RK3368_GRF_SOC_CON16, DELAY_ENABLE(RK3368, tx_delay, rx_delay) | @@ -724,7 +721,8 @@ static void rk3368_set_to_rgmii(struct rk_priv_data *bsp_priv, static void rk3368_set_to_rmii(struct rk_priv_data *bsp_priv) { regmap_write(bsp_priv->grf, RK3368_GRF_SOC_CON15, - RK3368_GMAC_PHY_INTF_SEL_RMII | RK3368_GMAC_RMII_MODE); + RK3368_GMAC_PHY_INTF_SEL(PHY_INTF_SEL_RMII) | + RK3368_GMAC_RMII_MODE); } static const struct rk_reg_speed_data rk3368_reg_speed_data = { @@ -752,8 +750,7 @@ static const struct rk_gmac_ops rk3368_ops = { #define RK3399_GRF_SOC_CON6 0xc218 /* RK3399_GRF_SOC_CON5 */ -#define RK3399_GMAC_PHY_INTF_SEL_RGMII GRF_FIELD(11, 9, PHY_INTF_SEL_RGMII) -#define RK3399_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(11, 9, PHY_INTF_SEL_RMII) +#define RK3399_GMAC_PHY_INTF_SEL(val) GRF_FIELD(11, 9, val) #define RK3399_GMAC_FLOW_CTRL GRF_BIT(8) #define RK3399_GMAC_FLOW_CTRL_CLR GRF_CLR_BIT(8) #define RK3399_GMAC_SPEED_10M GRF_CLR_BIT(7) @@ -778,7 +775,7 @@ static void rk3399_set_to_rgmii(struct rk_priv_data *bsp_priv, int tx_delay, int rx_delay) { regmap_write(bsp_priv->grf, RK3399_GRF_SOC_CON5, - RK3399_GMAC_PHY_INTF_SEL_RGMII | + RK3399_GMAC_PHY_INTF_SEL(PHY_INTF_SEL_RGMII) | RK3399_GMAC_RMII_MODE_CLR); regmap_write(bsp_priv->grf, RK3399_GRF_SOC_CON6, DELAY_ENABLE(RK3399, tx_delay, rx_delay) | @@ -789,7 +786,8 @@ static void rk3399_set_to_rgmii(struct rk_priv_data *bsp_priv, static void rk3399_set_to_rmii(struct rk_priv_data *bsp_priv) { regmap_write(bsp_priv->grf, RK3399_GRF_SOC_CON5, - RK3399_GMAC_PHY_INTF_SEL_RMII | RK3399_GMAC_RMII_MODE); + RK3399_GMAC_PHY_INTF_SEL(PHY_INTF_SEL_RMII) | + RK3399_GMAC_RMII_MODE); } static const struct rk_reg_speed_data rk3399_reg_speed_data = { @@ -1015,8 +1013,7 @@ static const struct rk_gmac_ops rk3528_ops = { #define RK3568_GRF_GMAC1_CON1 0x038c /* RK3568_GRF_GMAC0_CON1 && RK3568_GRF_GMAC1_CON1 */ -#define RK3568_GMAC_PHY_INTF_SEL_RGMII GRF_FIELD(6, 4, PHY_INTF_SEL_RGMII) -#define RK3568_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(6, 4, PHY_INTF_SEL_RMII) +#define RK3568_GMAC_PHY_INTF_SEL(val) GRF_FIELD(6, 4, val) #define RK3568_GMAC_FLOW_CTRL GRF_BIT(3) #define RK3568_GMAC_FLOW_CTRL_CLR GRF_CLR_BIT(3) #define RK3568_GMAC_RXCLK_DLY_ENABLE GRF_BIT(1) @@ -1043,7 +1040,7 @@ static void rk3568_set_to_rgmii(struct rk_priv_data *bsp_priv, RK3568_GMAC_CLK_TX_DL_CFG(tx_delay)); regmap_write(bsp_priv->grf, con1, - RK3568_GMAC_PHY_INTF_SEL_RGMII | + RK3568_GMAC_PHY_INTF_SEL(PHY_INTF_SEL_RGMII) | RK3568_GMAC_RXCLK_DLY_ENABLE | RK3568_GMAC_TXCLK_DLY_ENABLE); } @@ -1054,7 +1051,8 @@ static void rk3568_set_to_rmii(struct rk_priv_data *bsp_priv) con1 = (bsp_priv->id == 1) ? RK3568_GRF_GMAC1_CON1 : RK3568_GRF_GMAC0_CON1; - regmap_write(bsp_priv->grf, con1, RK3568_GMAC_PHY_INTF_SEL_RMII); + regmap_write(bsp_priv->grf, con1, + RK3568_GMAC_PHY_INTF_SEL(PHY_INTF_SEL_RMII)); } static const struct rk_gmac_ops rk3568_ops = { @@ -1208,10 +1206,8 @@ static const struct rk_gmac_ops rk3576_ops = { #define RK3588_GRF_GMAC_CON0 0X0008 #define RK3588_GRF_CLK_CON1 0X0070 -#define RK3588_GMAC_PHY_INTF_SEL_RGMII(id) \ - (GRF_FIELD(5, 3, PHY_INTF_SEL_RGMII) << ((id) * 6)) -#define RK3588_GMAC_PHY_INTF_SEL_RMII(id) \ - (GRF_FIELD(5, 3, PHY_INTF_SEL_RMII) << ((id) * 6)) +#define RK3588_GMAC_PHY_INTF_SEL(id, val) \ + (GRF_FIELD(5, 3, val) << ((id) * 6)) #define RK3588_GMAC_CLK_RMII_MODE(id) GRF_BIT(5 * (id)) #define RK3588_GMAC_CLK_RGMII_MODE(id) GRF_CLR_BIT(5 * (id)) @@ -1241,7 +1237,7 @@ static void rk3588_set_to_rgmii(struct rk_priv_data *bsp_priv, RK3588_GRF_GMAC_CON8; regmap_write(bsp_priv->php_grf, RK3588_GRF_GMAC_CON0, - RK3588_GMAC_PHY_INTF_SEL_RGMII(id)); + RK3588_GMAC_PHY_INTF_SEL(id, PHY_INTF_SEL_RGMII)); regmap_write(bsp_priv->php_grf, RK3588_GRF_CLK_CON1, RK3588_GMAC_CLK_RGMII_MODE(id)); @@ -1258,7 +1254,7 @@ static void rk3588_set_to_rgmii(struct rk_priv_data *bsp_priv, static void rk3588_set_to_rmii(struct rk_priv_data *bsp_priv) { regmap_write(bsp_priv->php_grf, RK3588_GRF_GMAC_CON0, - RK3588_GMAC_PHY_INTF_SEL_RMII(bsp_priv->id)); + RK3588_GMAC_PHY_INTF_SEL(bsp_priv->id, PHY_INTF_SEL_RMII)); regmap_write(bsp_priv->php_grf, RK3588_GRF_CLK_CON1, RK3588_GMAC_CLK_RMII_MODE(bsp_priv->id)); @@ -1328,7 +1324,7 @@ static const struct rk_gmac_ops rk3588_ops = { #define RV1108_GRF_GMAC_CON0 0X0900 /* RV1108_GRF_GMAC_CON0 */ -#define RV1108_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(6, 4, PHY_INTF_SEL_RMII) +#define RV1108_GMAC_PHY_INTF_SEL(val) GRF_FIELD(6, 4, val) #define RV1108_GMAC_FLOW_CTRL GRF_BIT(3) #define RV1108_GMAC_FLOW_CTRL_CLR GRF_CLR_BIT(3) #define RV1108_GMAC_SPEED_10M GRF_CLR_BIT(2) @@ -1339,7 +1335,7 @@ static const struct rk_gmac_ops rk3588_ops = { static void rv1108_set_to_rmii(struct rk_priv_data *bsp_priv) { regmap_write(bsp_priv->grf, RV1108_GRF_GMAC_CON0, - RV1108_GMAC_PHY_INTF_SEL_RMII); + RV1108_GMAC_PHY_INTF_SEL(PHY_INTF_SEL_RMII)); } static const struct rk_reg_speed_data rv1108_reg_speed_data = { @@ -1364,8 +1360,7 @@ static const struct rk_gmac_ops rv1108_ops = { #define RV1126_GRF_GMAC_CON2 0X0078 /* RV1126_GRF_GMAC_CON0 */ -#define RV1126_GMAC_PHY_INTF_SEL_RGMII GRF_FIELD(6, 4, PHY_INTF_SEL_RGMII) -#define RV1126_GMAC_PHY_INTF_SEL_RMII GRF_FIELD(6, 4, PHY_INTF_SEL_RMII) +#define RV1126_GMAC_PHY_INTF_SEL(val) GRF_FIELD(6, 4, val) #define RV1126_GMAC_FLOW_CTRL GRF_BIT(7) #define RV1126_GMAC_FLOW_CTRL_CLR GRF_CLR_BIT(7) #define RV1126_GMAC_M0_RXCLK_DLY_ENABLE GRF_BIT(1) @@ -1388,7 +1383,7 @@ static void rv1126_set_to_rgmii(struct rk_priv_data *bsp_priv, int tx_delay, int rx_delay) { regmap_write(bsp_priv->grf, RV1126_GRF_GMAC_CON0, - RV1126_GMAC_PHY_INTF_SEL_RGMII | + RV1126_GMAC_PHY_INTF_SEL(PHY_INTF_SEL_RGMII) | RV1126_GMAC_M0_RXCLK_DLY_ENABLE | RV1126_GMAC_M0_TXCLK_DLY_ENABLE | RV1126_GMAC_M1_RXCLK_DLY_ENABLE | @@ -1406,7 +1401,7 @@ static void rv1126_set_to_rgmii(struct rk_priv_data *bsp_priv, static void rv1126_set_to_rmii(struct rk_priv_data *bsp_priv) { regmap_write(bsp_priv->grf, RV1126_GRF_GMAC_CON0, - RV1126_GMAC_PHY_INTF_SEL_RMII); + RV1126_GMAC_PHY_INTF_SEL(PHY_INTF_SEL_RMII)); } static const struct rk_gmac_ops rv1126_ops = { From 467c3f008d0cafe8c21af7ad9f3036c1a15ec13d Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Thu, 13 Nov 2025 08:46:03 -0800 Subject: [PATCH 0807/1381] mlx4: extract GRXRINGS from .get_rxnfc Commit 84eaf4359c36 ("net: ethtool: add get_rx_ring_count callback to optimize RX ring queries") added specific support for GRXRINGS callback, simplifying .get_rxnfc. Remove the handling of GRXRINGS in .get_rxnfc() by moving it to the new .get_rx_ring_count(). This simplifies the RX ring count retrieval and aligns mlx4 with the new ethtool API for querying RX ring parameters. This is compiled tested only. Signed-off-by: Breno Leitao Reviewed-by: Tariq Toukan Link: https://patch.msgid.link/20251113-mlx_grxrings-v1-1-0017f2af7dd0@debian.org Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mellanox/mlx4/en_ethtool.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c index a68cd3f0304c..ad6298456639 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c @@ -1727,6 +1727,13 @@ static int mlx4_en_get_num_flows(struct mlx4_en_priv *priv) } +static u32 mlx4_en_get_rx_ring_count(struct net_device *dev) +{ + struct mlx4_en_priv *priv = netdev_priv(dev); + + return priv->rx_ring_num; +} + static int mlx4_en_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd, u32 *rule_locs) { @@ -1743,9 +1750,6 @@ static int mlx4_en_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd, return -EINVAL; switch (cmd->cmd) { - case ETHTOOL_GRXRINGS: - cmd->data = priv->rx_ring_num; - break; case ETHTOOL_GRXCLSRLCNT: cmd->rule_cnt = mlx4_en_get_num_flows(priv); break; @@ -2154,6 +2158,7 @@ const struct ethtool_ops mlx4_en_ethtool_ops = { .set_ringparam = mlx4_en_set_ringparam, .get_rxnfc = mlx4_en_get_rxnfc, .set_rxnfc = mlx4_en_set_rxnfc, + .get_rx_ring_count = mlx4_en_get_rx_ring_count, .get_rxfh_indir_size = mlx4_en_get_rxfh_indir_size, .get_rxfh_key_size = mlx4_en_get_rxfh_key_size, .get_rxfh = mlx4_en_get_rxfh, From 945499665f63197801b64fabb0bccf9d15ed09bf Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Thu, 13 Nov 2025 08:46:04 -0800 Subject: [PATCH 0808/1381] mlx5: extract GRXRINGS from .get_rxnfc Commit 84eaf4359c36 ("net: ethtool: add get_rx_ring_count callback to optimize RX ring queries") added specific support for GRXRINGS callback, simplifying .get_rxnfc. Remove the handling of GRXRINGS in .get_rxnfc() by moving it to the new .get_rx_ring_count() for both the mlx5 ethernet and IPoIB drivers. The ETHTOOL_GRXRINGS handling was previously kept in .get_rxnfc() to support "ethtool -x" when CONFIG_MLX5_EN_RXNFC=n. With the new dedicated .get_rx_ring_count() callback, this is no longer necessary. This simplifies the RX ring count retrieval and aligns mlx5 with the new ethtool API for querying RX ring parameters. Signed-off-by: Breno Leitao Reviewed-by: Tariq Toukan Link: https://patch.msgid.link/20251113-mlx_grxrings-v1-2-0017f2af7dd0@debian.org Signed-off-by: Jakub Kicinski --- .../ethernet/mellanox/mlx5/core/en_ethtool.c | 18 ++++++++---------- .../mellanox/mlx5/core/ipoib/ethtool.c | 18 ++++++++---------- 2 files changed, 16 insertions(+), 20 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c index 01b8f05a23db..939e274779b3 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -2492,21 +2492,18 @@ static int mlx5e_set_rxfh_fields(struct net_device *dev, return mlx5e_ethtool_set_rxfh_fields(priv, cmd, extack); } +static u32 mlx5e_get_rx_ring_count(struct net_device *dev) +{ + struct mlx5e_priv *priv = netdev_priv(dev); + + return priv->channels.params.num_channels; +} + static int mlx5e_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *info, u32 *rule_locs) { struct mlx5e_priv *priv = netdev_priv(dev); - /* ETHTOOL_GRXRINGS is needed by ethtool -x which is not part - * of rxnfc. We keep this logic out of mlx5e_ethtool_get_rxnfc, - * to avoid breaking "ethtool -x" when mlx5e_ethtool_get_rxnfc - * is compiled out via CONFIG_MLX5_EN_RXNFC=n. - */ - if (info->cmd == ETHTOOL_GRXRINGS) { - info->data = priv->channels.params.num_channels; - return 0; - } - return mlx5e_ethtool_get_rxnfc(priv, info, rule_locs); } @@ -2766,6 +2763,7 @@ const struct ethtool_ops mlx5e_ethtool_ops = { .remove_rxfh_context = mlx5e_remove_rxfh_context, .get_rxnfc = mlx5e_get_rxnfc, .set_rxnfc = mlx5e_set_rxnfc, + .get_rx_ring_count = mlx5e_get_rx_ring_count, .get_tunable = mlx5e_get_tunable, .set_tunable = mlx5e_set_tunable, .get_pause_stats = mlx5e_get_pause_stats, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ethtool.c index 4b3430ac3905..3b2f54ca30a8 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ethtool.c @@ -266,21 +266,18 @@ static int mlx5i_set_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd) return mlx5e_ethtool_set_rxnfc(priv, cmd); } +static u32 mlx5i_get_rx_ring_count(struct net_device *dev) +{ + struct mlx5e_priv *priv = mlx5i_epriv(dev); + + return priv->channels.params.num_channels; +} + static int mlx5i_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *info, u32 *rule_locs) { struct mlx5e_priv *priv = mlx5i_epriv(dev); - /* ETHTOOL_GRXRINGS is needed by ethtool -x which is not part - * of rxnfc. We keep this logic out of mlx5e_ethtool_get_rxnfc, - * to avoid breaking "ethtool -x" when mlx5e_ethtool_get_rxnfc - * is compiled out via CONFIG_MLX5_EN_RXNFC=n. - */ - if (info->cmd == ETHTOOL_GRXRINGS) { - info->data = priv->channels.params.num_channels; - return 0; - } - return mlx5e_ethtool_get_rxnfc(priv, info, rule_locs); } @@ -304,6 +301,7 @@ const struct ethtool_ops mlx5i_ethtool_ops = { .set_rxfh_fields = mlx5i_set_rxfh_fields, .get_rxnfc = mlx5i_get_rxnfc, .set_rxnfc = mlx5i_set_rxnfc, + .get_rx_ring_count = mlx5i_get_rx_ring_count, .get_link_ksettings = mlx5i_get_link_ksettings, .get_link = ethtool_op_get_link, }; From 0135333914d63181f823bd340ae96737c8a820ca Mon Sep 17 00:00:00 2001 From: Mohsin Bashir Date: Thu, 13 Nov 2025 15:26:10 -0800 Subject: [PATCH 0809/1381] eth: fbnic: Configure RDE settings for pause frame fbnic supports pause frames. When pause frames are enabled presumably user expects lossless operation from the NIC. Make sure we configure RDE (Rx DMA Engine) to DROP_NEVER mode to avoid discards due to delays in fetching Rx descriptors from the host. While at it enable DROP_NEVER when NIC only has a single queue configured. In this case the NIC acts as a FIFO so there's no risk of head-of-line blocking other queues by making RDE wait. If pause is disabled this just moves the packet loss from the DMA engine to the Rx buffer. Remove redundant call to fbnic_config_drop_mode_rcq(), introduced by commit 0cb4c0a13723 ("eth: fbnic: Implement Rx queue alloc/start/stop/free"). This call does not add value as fbnic_enable_rcq(), which is called immediately afterward, already handles this. Although we do not support autoneg at this time, preserve tx_pause in .mac_link_up instead of fbnic_phylink_get_pauseparam() Signed-off-by: Mohsin Bashir Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20251113232610.1151712-1-mohsin.bashr@gmail.com Signed-off-by: Jakub Kicinski --- .../net/ethernet/meta/fbnic/fbnic_netdev.h | 2 ++ .../net/ethernet/meta/fbnic/fbnic_phylink.c | 3 +++ drivers/net/ethernet/meta/fbnic/fbnic_txrx.c | 26 ++++++++++++++++--- drivers/net/ethernet/meta/fbnic/fbnic_txrx.h | 1 + 4 files changed, 28 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.h b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.h index b0a87c57910f..e6ca23a9957d 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.h +++ b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.h @@ -73,6 +73,8 @@ struct fbnic_net { /* Time stamping filter config */ struct kernel_hwtstamp_config hwtstamp_config; + + bool tx_pause; }; int __fbnic_open(struct fbnic_net *fbn); diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_phylink.c b/drivers/net/ethernet/meta/fbnic/fbnic_phylink.c index 7ce3fdd25282..62701923cfe9 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_phylink.c +++ b/drivers/net/ethernet/meta/fbnic/fbnic_phylink.c @@ -208,6 +208,9 @@ fbnic_phylink_mac_link_up(struct phylink_config *config, struct fbnic_net *fbn = netdev_priv(netdev); struct fbnic_dev *fbd = fbn->fbd; + fbn->tx_pause = tx_pause; + fbnic_config_drop_mode(fbn, tx_pause); + fbd->mac->link_up(fbd, tx_pause, rx_pause); } diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_txrx.c b/drivers/net/ethernet/meta/fbnic/fbnic_txrx.c index 57e18a68f5d2..c2d7b67fec28 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_txrx.c +++ b/drivers/net/ethernet/meta/fbnic/fbnic_txrx.c @@ -2574,11 +2574,15 @@ static void fbnic_enable_bdq(struct fbnic_ring *hpq, struct fbnic_ring *ppq) } static void fbnic_config_drop_mode_rcq(struct fbnic_napi_vector *nv, - struct fbnic_ring *rcq) + struct fbnic_ring *rcq, bool tx_pause) { + struct fbnic_net *fbn = netdev_priv(nv->napi.dev); u32 drop_mode, rcq_ctl; - drop_mode = FBNIC_QUEUE_RDE_CTL0_DROP_IMMEDIATE; + if (!tx_pause && fbn->num_rx_queues > 1) + drop_mode = FBNIC_QUEUE_RDE_CTL0_DROP_IMMEDIATE; + else + drop_mode = FBNIC_QUEUE_RDE_CTL0_DROP_NEVER; /* Specify packet layout */ rcq_ctl = FIELD_PREP(FBNIC_QUEUE_RDE_CTL0_DROP_MODE_MASK, drop_mode) | @@ -2588,6 +2592,21 @@ static void fbnic_config_drop_mode_rcq(struct fbnic_napi_vector *nv, fbnic_ring_wr32(rcq, FBNIC_QUEUE_RDE_CTL0, rcq_ctl); } +void fbnic_config_drop_mode(struct fbnic_net *fbn, bool tx_pause) +{ + int i, t; + + for (i = 0; i < fbn->num_napi; i++) { + struct fbnic_napi_vector *nv = fbn->napi[i]; + + for (t = 0; t < nv->rxt_count; t++) { + struct fbnic_q_triad *qt = &nv->qt[nv->txt_count + t]; + + fbnic_config_drop_mode_rcq(nv, &qt->cmpl, tx_pause); + } + } +} + static void fbnic_config_rim_threshold(struct fbnic_ring *rcq, u16 nv_idx, u32 rx_desc) { u32 threshold; @@ -2637,7 +2656,7 @@ static void fbnic_enable_rcq(struct fbnic_napi_vector *nv, u32 hds_thresh = fbn->hds_thresh; u32 rcq_ctl = 0; - fbnic_config_drop_mode_rcq(nv, rcq); + fbnic_config_drop_mode_rcq(nv, rcq, fbn->tx_pause); /* Force lower bound on MAX_HEADER_BYTES. Below this, all frames should * be split at L4. It would also result in the frames being split at @@ -2700,7 +2719,6 @@ static void __fbnic_nv_enable(struct fbnic_napi_vector *nv) &nv->napi); fbnic_enable_bdq(&qt->sub0, &qt->sub1); - fbnic_config_drop_mode_rcq(nv, &qt->cmpl); fbnic_enable_rcq(nv, &qt->cmpl); } } diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_txrx.h b/drivers/net/ethernet/meta/fbnic/fbnic_txrx.h index ca37da5a0b17..27776e844e29 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_txrx.h +++ b/drivers/net/ethernet/meta/fbnic/fbnic_txrx.h @@ -184,6 +184,7 @@ void fbnic_reset_netif_queues(struct fbnic_net *fbn); irqreturn_t fbnic_msix_clean_rings(int irq, void *data); void fbnic_napi_enable(struct fbnic_net *fbn); void fbnic_napi_disable(struct fbnic_net *fbn); +void fbnic_config_drop_mode(struct fbnic_net *fbn, bool tx_pause); void fbnic_enable(struct fbnic_net *fbn); void fbnic_disable(struct fbnic_net *fbn); void fbnic_flush(struct fbnic_net *fbn); From 40ea40853da9c3fa03d52c2998998d103b80f593 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Sat, 15 Nov 2025 14:55:08 -0800 Subject: [PATCH 0810/1381] tools: ynltool: remove -lmnl from link flags The libmnl dependency has been removed from libynl back in commit 73395b43819b ("tools: ynl: remove the libmnl dependency") Remove it from the ynltool Makefile. Reviewed-by: Donald Hunter Link: https://patch.msgid.link/20251115225508.1000072-1-kuba@kernel.org Signed-off-by: Jakub Kicinski --- tools/net/ynl/ynltool/Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/net/ynl/ynltool/Makefile b/tools/net/ynl/ynltool/Makefile index 5edaa0f93237..f5b1de32daa5 100644 --- a/tools/net/ynl/ynltool/Makefile +++ b/tools/net/ynl/ynltool/Makefile @@ -31,7 +31,7 @@ Q = @ $(YNLTOOL): ../libynl.a $(OBJS) $(Q)echo -e "\tLINK $@" - $(Q)$(CC) $(CFLAGS) -o $@ $(OBJS) ../libynl.a -lmnl -lm + $(Q)$(CC) $(CFLAGS) -o $@ $(OBJS) ../libynl.a -lm %.o: %.c ../libynl.a $(Q)echo -e "\tCC $@" From 5860bb1ce0ddb9c5c964f7b46a8f0c4a48f919b1 Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Wed, 12 Nov 2025 22:06:13 +0100 Subject: [PATCH 0811/1381] net: phy: fixed_phy: remove setting supported/advertised modes from fixed_phy_register This code was added with 34b31da486a5 ("phy: fixed_phy: Set supported speed in phydev") 10 yrs ago. The commit message of this change mentions a use case involving callback adjust_link of struct dsa_switch_driver. This struct doesn't exist any longer, and in general usage of the legacy fixed PHY has been removed from DSA with the switch to phylink. Note: Supported and advertised modes are now set by phy_probe() when the fixed PHY is attached to the netdev and bound to the genphy driver. Signed-off-by: Heiner Kallweit Link: https://patch.msgid.link/3abaa3c5-fbb9-4052-9346-6cb096a25878@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/phy/fixed_phy.c | 24 ------------------------ 1 file changed, 24 deletions(-) diff --git a/drivers/net/phy/fixed_phy.c b/drivers/net/phy/fixed_phy.c index 715f0356f895..1ad77f542c1f 100644 --- a/drivers/net/phy/fixed_phy.c +++ b/drivers/net/phy/fixed_phy.c @@ -19,7 +19,6 @@ #include #include #include -#include #include "swphy.h" @@ -184,29 +183,6 @@ struct phy_device *fixed_phy_register(const struct fixed_phy_status *status, phy->mdio.dev.of_node = np; phy->is_pseudo_fixed_link = true; - switch (status->speed) { - case SPEED_1000: - linkmode_set_bit(ETHTOOL_LINK_MODE_1000baseT_Half_BIT, - phy->supported); - linkmode_set_bit(ETHTOOL_LINK_MODE_1000baseT_Full_BIT, - phy->supported); - fallthrough; - case SPEED_100: - linkmode_set_bit(ETHTOOL_LINK_MODE_100baseT_Half_BIT, - phy->supported); - linkmode_set_bit(ETHTOOL_LINK_MODE_100baseT_Full_BIT, - phy->supported); - fallthrough; - case SPEED_10: - default: - linkmode_set_bit(ETHTOOL_LINK_MODE_10baseT_Half_BIT, - phy->supported); - linkmode_set_bit(ETHTOOL_LINK_MODE_10baseT_Full_BIT, - phy->supported); - } - - phy_advertise_supported(phy); - ret = phy_device_register(phy); if (ret) { phy_device_free(phy); From ca412f25d6b2c21f69a6cf12da062e0be4a5f45e Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 14 Nov 2025 13:51:41 +0000 Subject: [PATCH 0812/1381] tcp: reduce tcp_comp_sack_slack_ns default value to 10 usec net.ipv4.tcp_comp_sack_slack_ns current default value is too high. When a flow has many drops (1 % or more), and small RTT, adding 100 usec before sending SACK stalls the sender relying on getting SACK fast enough to keep the pipe busy. Decrease the default to 10 usec. This is orthogonal to Congestion Control heuristics to determine if drops are caused by congestion or not. Signed-off-by: Eric Dumazet Reviewed-by: Neal Cardwell Link: https://patch.msgid.link/20251114135141.3810964-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- Documentation/networking/ip-sysctl.rst | 3 ++- net/ipv4/tcp_ipv4.c | 2 +- 2 files changed, 3 insertions(+), 2 deletions(-) diff --git a/Documentation/networking/ip-sysctl.rst b/Documentation/networking/ip-sysctl.rst index 2bae61be1859..f4ad739a6b53 100644 --- a/Documentation/networking/ip-sysctl.rst +++ b/Documentation/networking/ip-sysctl.rst @@ -875,8 +875,9 @@ tcp_comp_sack_slack_ns - LONG INTEGER timer used by SACK compression. This gives extra time for small RTT flows, and reduces system overhead by allowing opportunistic reduction of timer interrupts. + Too big values might reduce goodput. - Default : 100,000 ns (100 us) + Default : 10,000 ns (10 us) tcp_comp_sack_nr - INTEGER Max number of SACK that can be compressed. diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index a7d9fec2950b..6fcaecb67284 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -3593,7 +3593,7 @@ static int __net_init tcp_sk_init(struct net *net) sizeof(init_net.ipv4.sysctl_tcp_wmem)); } net->ipv4.sysctl_tcp_comp_sack_delay_ns = NSEC_PER_MSEC; - net->ipv4.sysctl_tcp_comp_sack_slack_ns = 100 * NSEC_PER_USEC; + net->ipv4.sysctl_tcp_comp_sack_slack_ns = 10 * NSEC_PER_USEC; net->ipv4.sysctl_tcp_comp_sack_nr = 44; net->ipv4.sysctl_tcp_comp_sack_rtt_percent = 33; net->ipv4.sysctl_tcp_backlog_ack_defer = 1; From 3a12581fc292997f5ca6dafe3238afd39b77e798 Mon Sep 17 00:00:00 2001 From: Chih-Kang Chang Date: Fri, 14 Nov 2025 14:01:15 +0800 Subject: [PATCH 0813/1381] wifi: rtw89: flush TX queue before deleting key In the wpa_supplicant rekey flow, it sends an EAPOL packet 4/4 through nl80211_tx_control_port() and triggers wake_tx_queue() in the driver. Then, it sends nl80211_new_key() to configure a new key in mac80211. However, in wake_tx_queue(), a workqueue is used to process the TX packet, which might cause the driver to process the EAPOL packet later than nl80211_new_key(). This results in the EAPOL packet 4/4 being transmitted with the new key and IV, causing it to be dropped by the AP. Therefore, needs to flush TX queue before deleting the old key to ensure that the EAPOL 4/4 packet is transmitted using the old key. Signed-off-by: Chih-Kang Chang Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251114060128.35363-2-pkshih@realtek.com --- drivers/net/wireless/realtek/rtw89/mac80211.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/wireless/realtek/rtw89/mac80211.c b/drivers/net/wireless/realtek/rtw89/mac80211.c index 6f7571c7c274..4a7364432428 100644 --- a/drivers/net/wireless/realtek/rtw89/mac80211.c +++ b/drivers/net/wireless/realtek/rtw89/mac80211.c @@ -969,6 +969,7 @@ static int rtw89_ops_set_key(struct ieee80211_hw *hw, enum set_key_cmd cmd, } break; case DISABLE_KEY: + flush_work(&rtwdev->txq_work); rtw89_hci_flush_queues(rtwdev, BIT(rtwdev->hw->queues) - 1, false); rtw89_mac_flush_txq(rtwdev, BIT(rtwdev->hw->queues) - 1, false); From 92db331de6de948f4308ab4c80ddbe7ae27f39bb Mon Sep 17 00:00:00 2001 From: Ping-Ke Shih Date: Fri, 14 Nov 2025 14:01:16 +0800 Subject: [PATCH 0814/1381] wifi: rtw89: fw: part size to download firmware by header info The part size is the unit to download firmware piece by piece. Old chips use 2020 bytes as a piece, but in new chips the part size is defined in firmware header. Change to use the value dynamically. Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251114060128.35363-3-pkshih@realtek.com --- drivers/net/wireless/realtek/rtw89/fw.c | 31 +++++++++++++++---------- drivers/net/wireless/realtek/rtw89/fw.h | 1 + 2 files changed, 20 insertions(+), 12 deletions(-) diff --git a/drivers/net/wireless/realtek/rtw89/fw.c b/drivers/net/wireless/realtek/rtw89/fw.c index 515e9f098380..b234df109634 100644 --- a/drivers/net/wireless/realtek/rtw89/fw.c +++ b/drivers/net/wireless/realtek/rtw89/fw.c @@ -161,6 +161,11 @@ static int rtw89_fw_hdr_parser_v0(struct rtw89_dev *rtwdev, const u8 *fw, u32 le info->dynamic_hdr_en = le32_get_bits(fw_hdr->w7, FW_HDR_W7_DYN_HDR); info->idmem_share_mode = le32_get_bits(fw_hdr->w7, FW_HDR_W7_IDMEM_SHARE_MODE); + if (chip->chip_gen == RTW89_CHIP_AX) + info->part_size = FWDL_SECTION_PER_PKT_LEN; + else + info->part_size = le32_get_bits(fw_hdr->w7, FW_HDR_W7_PART_SIZE); + if (info->dynamic_hdr_en) { info->hdr_len = le32_get_bits(fw_hdr->w3, FW_HDR_W3_LEN); info->dynamic_hdr_len = info->hdr_len - base_hdr_len; @@ -439,6 +444,7 @@ static int rtw89_fw_hdr_parser_v1(struct rtw89_dev *rtwdev, const u8 *fw, u32 le struct rtw89_fw_bin_info *info) { const struct rtw89_fw_hdr_v1 *fw_hdr = (const struct rtw89_fw_hdr_v1 *)fw; + const struct rtw89_chip_info *chip = rtwdev->chip; struct rtw89_fw_hdr_section_info *section_info; const struct rtw89_fw_dynhdr_hdr *fwdynhdr; const struct rtw89_fw_hdr_section_v1 *section; @@ -455,6 +461,11 @@ static int rtw89_fw_hdr_parser_v1(struct rtw89_dev *rtwdev, const u8 *fw, u32 le info->dynamic_hdr_en = le32_get_bits(fw_hdr->w7, FW_HDR_V1_W7_DYN_HDR); info->idmem_share_mode = le32_get_bits(fw_hdr->w7, FW_HDR_V1_W7_IDMEM_SHARE_MODE); + if (chip->chip_gen == RTW89_CHIP_AX) + info->part_size = FWDL_SECTION_PER_PKT_LEN; + else + info->part_size = le32_get_bits(fw_hdr->w7, FW_HDR_V1_W7_PART_SIZE); + if (info->dynamic_hdr_en) { info->hdr_len = le32_get_bits(fw_hdr->w5, FW_HDR_V1_W5_HDR_SIZE); info->dynamic_hdr_len = info->hdr_len - base_hdr_len; @@ -1516,8 +1527,7 @@ static u32 __rtw89_fw_download_tweak_hdr_v0(struct rtw89_dev *rtwdev, struct rtw89_fw_hdr_section *section; int i; - le32p_replace_bits(&fw_hdr->w7, FWDL_SECTION_PER_PKT_LEN, - FW_HDR_W7_PART_SIZE); + le32p_replace_bits(&fw_hdr->w7, info->part_size, FW_HDR_W7_PART_SIZE); for (i = 0; i < info->section_num; i++) { section_info = &info->section_info[i]; @@ -1542,8 +1552,7 @@ static u32 __rtw89_fw_download_tweak_hdr_v1(struct rtw89_dev *rtwdev, u8 dst_sec_idx = 0; u8 sec_idx; - le32p_replace_bits(&fw_hdr->w7, FWDL_SECTION_PER_PKT_LEN, - FW_HDR_V1_W7_PART_SIZE); + le32p_replace_bits(&fw_hdr->w7, info->part_size, FW_HDR_V1_W7_PART_SIZE); for (sec_idx = 0; sec_idx < info->section_num; sec_idx++) { section_info = &info->section_info[sec_idx]; @@ -1645,7 +1654,8 @@ static int rtw89_fw_download_hdr(struct rtw89_dev *rtwdev, } static int __rtw89_fw_download_main(struct rtw89_dev *rtwdev, - struct rtw89_fw_hdr_section_info *info) + struct rtw89_fw_hdr_section_info *info, + u32 part_size) { struct sk_buff *skb; const u8 *section = info->addr; @@ -1666,20 +1676,17 @@ static int __rtw89_fw_download_main(struct rtw89_dev *rtwdev, } if (info->key_addr && info->key_len) { - if (residue_len > FWDL_SECTION_PER_PKT_LEN || info->len < info->key_len) + if (residue_len > part_size || info->len < info->key_len) rtw89_warn(rtwdev, "ignore to copy key data because of len %d, %d, %d, %d\n", - info->len, FWDL_SECTION_PER_PKT_LEN, + info->len, part_size, info->key_len, residue_len); else copy_key = true; } while (residue_len) { - if (residue_len >= FWDL_SECTION_PER_PKT_LEN) - pkt_len = FWDL_SECTION_PER_PKT_LEN; - else - pkt_len = residue_len; + pkt_len = min(residue_len, part_size); skb = rtw89_fw_h2c_alloc_skb_no_hdr(rtwdev, pkt_len); if (!skb) { @@ -1734,7 +1741,7 @@ static int rtw89_fw_download_main(struct rtw89_dev *rtwdev, int ret; while (section_num--) { - ret = __rtw89_fw_download_main(rtwdev, section_info); + ret = __rtw89_fw_download_main(rtwdev, section_info, info->part_size); if (ret) return ret; section_info++; diff --git a/drivers/net/wireless/realtek/rtw89/fw.h b/drivers/net/wireless/realtek/rtw89/fw.h index 6620256e3f22..87c55a1a9d94 100644 --- a/drivers/net/wireless/realtek/rtw89/fw.h +++ b/drivers/net/wireless/realtek/rtw89/fw.h @@ -297,6 +297,7 @@ struct rtw89_fw_hdr_section_info { struct rtw89_fw_bin_info { u8 section_num; + u32 part_size; u32 hdr_len; bool dynamic_hdr_en; u32 dynamic_hdr_len; From a2a64fe2346374940c8e8953b42c73983a84559b Mon Sep 17 00:00:00 2001 From: Ping-Ke Shih Date: Fri, 14 Nov 2025 14:01:17 +0800 Subject: [PATCH 0815/1381] wifi: rtw89: mac: separate pre-init code before downloading firmware Driver needs to initialize registers before downloading firmware, so move pre-init part (power on) from original rtw89_mac_init(). The consequence patches will add more pre-init codes before downloading firmware. Since rtw89_phy_init_bb_afe() is used by coming RTL8922D, don't change logic at all for existing chips. Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251114060128.35363-4-pkshih@realtek.com --- drivers/net/wireless/realtek/rtw89/core.c | 8 ++++++++ drivers/net/wireless/realtek/rtw89/mac.c | 15 +++++++++++---- drivers/net/wireless/realtek/rtw89/mac.h | 1 + 3 files changed, 20 insertions(+), 4 deletions(-) diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c index a15a911484bb..0cabea7dd92b 100644 --- a/drivers/net/wireless/realtek/rtw89/core.c +++ b/drivers/net/wireless/realtek/rtw89/core.c @@ -5617,8 +5617,16 @@ int rtw89_core_start(struct rtw89_dev *rtwdev) { int ret; + ret = rtw89_mac_preinit(rtwdev); + if (ret) { + rtw89_err(rtwdev, "mac preinit fail, ret: %d\n", ret); + return ret; + } + rtw89_phy_init_bb_afe(rtwdev); + /* above do preinit before downloading firmware */ + ret = rtw89_mac_init(rtwdev); if (ret) { rtw89_err(rtwdev, "mac init fail, ret:%d\n", ret); diff --git a/drivers/net/wireless/realtek/rtw89/mac.c b/drivers/net/wireless/realtek/rtw89/mac.c index dab0a76556c9..ff160d25a7c9 100644 --- a/drivers/net/wireless/realtek/rtw89/mac.c +++ b/drivers/net/wireless/realtek/rtw89/mac.c @@ -4126,6 +4126,17 @@ int rtw89_mac_partial_init(struct rtw89_dev *rtwdev, bool include_bb) return 0; } +int rtw89_mac_preinit(struct rtw89_dev *rtwdev) +{ + int ret; + + ret = rtw89_mac_pwr_on(rtwdev); + if (ret) + return ret; + + return 0; +} + int rtw89_mac_init(struct rtw89_dev *rtwdev) { const struct rtw89_mac_gen_def *mac = rtwdev->chip->mac_def; @@ -4133,10 +4144,6 @@ int rtw89_mac_init(struct rtw89_dev *rtwdev) bool include_bb = !!chip->bbmcu_nr; int ret; - ret = rtw89_mac_pwr_on(rtwdev); - if (ret) - return ret; - ret = rtw89_mac_partial_init(rtwdev, include_bb); if (ret) goto fail; diff --git a/drivers/net/wireless/realtek/rtw89/mac.h b/drivers/net/wireless/realtek/rtw89/mac.h index 3cc97fd0c0ec..d203db444a34 100644 --- a/drivers/net/wireless/realtek/rtw89/mac.h +++ b/drivers/net/wireless/realtek/rtw89/mac.h @@ -1201,6 +1201,7 @@ rtw89_write32_port_set(struct rtw89_dev *rtwdev, struct rtw89_vif_link *rtwvif_l int rtw89_mac_pwr_on(struct rtw89_dev *rtwdev); void rtw89_mac_pwr_off(struct rtw89_dev *rtwdev); int rtw89_mac_partial_init(struct rtw89_dev *rtwdev, bool include_bb); +int rtw89_mac_preinit(struct rtw89_dev *rtwdev); int rtw89_mac_init(struct rtw89_dev *rtwdev); int rtw89_mac_dle_init(struct rtw89_dev *rtwdev, enum rtw89_qta_mode mode, enum rtw89_qta_mode ext_mode); From 12e84effcb6d6d2a58e03cb237c1100c817dd479 Mon Sep 17 00:00:00 2001 From: Ping-Ke Shih Date: Fri, 14 Nov 2025 14:01:18 +0800 Subject: [PATCH 0816/1381] wifi: rtw89: phy: calling BB pre-init by chips with/without BB MCU The existing flow is doing BB pre-init before downloading BB MCU firmware, because existing chip RTL8922A has BB MCU. However, the coming chips don't have this, and BB pre-init configuring registers can affect downloading WiFi-CPU firmware. Therefore, calling BB pre-init afterward for new coming chips without BB MCU. For existing WiFi 6 chips, no BB pre-init. For RTL8922A, don't change the logic. Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251114060128.35363-5-pkshih@realtek.com --- drivers/net/wireless/realtek/rtw89/core.c | 4 ++++ drivers/net/wireless/realtek/rtw89/core.h | 11 ++++++++--- drivers/net/wireless/realtek/rtw89/mac.c | 9 ++++++--- 3 files changed, 18 insertions(+), 6 deletions(-) diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c index 0cabea7dd92b..ea22925b8772 100644 --- a/drivers/net/wireless/realtek/rtw89/core.c +++ b/drivers/net/wireless/realtek/rtw89/core.c @@ -5615,6 +5615,7 @@ EXPORT_SYMBOL(rtw89_check_quirks); int rtw89_core_start(struct rtw89_dev *rtwdev) { + bool no_bbmcu = !rtwdev->chip->bbmcu_nr; int ret; ret = rtw89_mac_preinit(rtwdev); @@ -5623,6 +5624,9 @@ int rtw89_core_start(struct rtw89_dev *rtwdev) return ret; } + if (no_bbmcu) + rtw89_chip_bb_preinit(rtwdev); + rtw89_phy_init_bb_afe(rtwdev); /* above do preinit before downloading firmware */ diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h index f63857b389d7..8943166de52c 100644 --- a/drivers/net/wireless/realtek/rtw89/core.h +++ b/drivers/net/wireless/realtek/rtw89/core.h @@ -7039,12 +7039,17 @@ static inline void rtw89_chip_rfk_hw_init(struct rtw89_dev *rtwdev) } static inline -void rtw89_chip_bb_preinit(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy_idx) +void rtw89_chip_bb_preinit(struct rtw89_dev *rtwdev) { const struct rtw89_chip_info *chip = rtwdev->chip; - if (chip->ops->bb_preinit) - chip->ops->bb_preinit(rtwdev, phy_idx); + if (!chip->ops->bb_preinit) + return; + + chip->ops->bb_preinit(rtwdev, RTW89_PHY_0); + + if (rtwdev->dbcc_en) + chip->ops->bb_preinit(rtwdev, RTW89_PHY_1); } static inline diff --git a/drivers/net/wireless/realtek/rtw89/mac.c b/drivers/net/wireless/realtek/rtw89/mac.c index ff160d25a7c9..21f5023c4f69 100644 --- a/drivers/net/wireless/realtek/rtw89/mac.c +++ b/drivers/net/wireless/realtek/rtw89/mac.c @@ -4104,9 +4104,12 @@ int rtw89_mac_partial_init(struct rtw89_dev *rtwdev, bool include_bb) rtw89_mac_ctrl_hci_dma_trx(rtwdev, true); if (include_bb) { - rtw89_chip_bb_preinit(rtwdev, RTW89_PHY_0); - if (rtwdev->dbcc_en) - rtw89_chip_bb_preinit(rtwdev, RTW89_PHY_1); + /* Only call BB preinit including configuration of BB MCU for + * the chips which need to download BB MCU firmware. Otherwise, + * calling preinit later to prevent touching registers affecting + * download firmware. + */ + rtw89_chip_bb_preinit(rtwdev); } ret = rtw89_mac_dmac_pre_init(rtwdev); From d375434cbceae79f841ecb0739ae6f5174b968e2 Mon Sep 17 00:00:00 2001 From: Ping-Ke Shih Date: Fri, 14 Nov 2025 14:01:19 +0800 Subject: [PATCH 0817/1381] wifi: rtw89: mac: remove undefined bit B_BE_PPDU_MAC_INFO The bit is defined during test chip development, but formal chips don't have this. Remove it. Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251114060128.35363-6-pkshih@realtek.com --- drivers/net/wireless/realtek/rtw89/mac_be.c | 2 +- drivers/net/wireless/realtek/rtw89/reg.h | 1 - 2 files changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/net/wireless/realtek/rtw89/mac_be.c b/drivers/net/wireless/realtek/rtw89/mac_be.c index ef69672b6862..c3620db8aff6 100644 --- a/drivers/net/wireless/realtek/rtw89/mac_be.c +++ b/drivers/net/wireless/realtek/rtw89/mac_be.c @@ -2020,7 +2020,7 @@ int rtw89_mac_cfg_ppdu_status_be(struct rtw89_dev *rtwdev, u8 mac_idx, bool enab } rtw89_write32_mask(rtwdev, R_BE_HW_PPDU_STATUS, B_BE_FWD_PPDU_STAT_MASK, 3); - rtw89_write32(rtwdev, reg, B_BE_PPDU_STAT_RPT_EN | B_BE_PPDU_MAC_INFO | + rtw89_write32(rtwdev, reg, B_BE_PPDU_STAT_RPT_EN | B_BE_APP_RX_CNT_RPT | B_BE_APP_PLCP_HDR_RPT | B_BE_PPDU_STAT_RPT_CRC32 | B_BE_PPDU_STAT_RPT_DMA); diff --git a/drivers/net/wireless/realtek/rtw89/reg.h b/drivers/net/wireless/realtek/rtw89/reg.h index ed1d958bc49e..152fd8366dd7 100644 --- a/drivers/net/wireless/realtek/rtw89/reg.h +++ b/drivers/net/wireless/realtek/rtw89/reg.h @@ -7476,7 +7476,6 @@ #define B_BE_PPDU_STAT_RPT_ADDR BIT(4) #define B_BE_APP_PLCP_HDR_RPT BIT(3) #define B_BE_APP_RX_CNT_RPT BIT(2) -#define B_BE_PPDU_MAC_INFO BIT(1) #define B_BE_PPDU_STAT_RPT_EN BIT(0) #define R_BE_RX_SR_CTRL 0x1144A From 7fefea11001d5f5cf948a069f5f2cecc8fa5633f Mon Sep 17 00:00:00 2001 From: Ping-Ke Shih Date: Fri, 14 Nov 2025 14:01:20 +0800 Subject: [PATCH 0818/1381] wifi: rtw89: mac: update wcpu_on to download firmware for RTL8922D The RTL8922D does more settings before downloading firmware, so add them accordingly. Also, update the missed settings for RTL8922A. Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251114060128.35363-7-pkshih@realtek.com --- drivers/net/wireless/realtek/rtw89/mac_be.c | 7 +++++++ drivers/net/wireless/realtek/rtw89/reg.h | 23 +++++++++++++++++++++ 2 files changed, 30 insertions(+) diff --git a/drivers/net/wireless/realtek/rtw89/mac_be.c b/drivers/net/wireless/realtek/rtw89/mac_be.c index c3620db8aff6..556e5f98e8d4 100644 --- a/drivers/net/wireless/realtek/rtw89/mac_be.c +++ b/drivers/net/wireless/realtek/rtw89/mac_be.c @@ -458,6 +458,7 @@ static void set_cpu_en(struct rtw89_dev *rtwdev, bool include_bb) static int wcpu_on(struct rtw89_dev *rtwdev, u8 boot_reason, bool dlfw) { + const struct rtw89_chip_info *chip = rtwdev->chip; u32 val32; int ret; @@ -479,6 +480,7 @@ static int wcpu_on(struct rtw89_dev *rtwdev, u8 boot_reason, bool dlfw) rtw89_write32(rtwdev, R_BE_UDM1, 0); rtw89_write32(rtwdev, R_BE_UDM2, 0); + rtw89_write32(rtwdev, R_BE_BOOT_DBG, 0x0); rtw89_write32(rtwdev, R_BE_HALT_H2C, 0); rtw89_write32(rtwdev, R_BE_HALT_C2H, 0); rtw89_write32(rtwdev, R_BE_HALT_H2C_CTRL, 0); @@ -493,6 +495,11 @@ static int wcpu_on(struct rtw89_dev *rtwdev, u8 boot_reason, bool dlfw) B_BE_WDT_WAKE_PCIE_EN | B_BE_WDT_WAKE_USB_EN); rtw89_write32_clr(rtwdev, R_BE_WCPU_FW_CTRL, B_BE_WDT_PLT_RST_EN | B_BE_WCPU_ROM_CUT_GET); + rtw89_write32(rtwdev, R_BE_SECURE_BOOT_MALLOC_INFO, 0); + rtw89_write32_clr(rtwdev, R_BE_GPIO_MUXCFG, B_BE_BOOT_MODE); + + if (chip->chip_id != RTL8922A) + rtw89_write32_set(rtwdev, R_BE_WCPU_FW_CTRL, B_BE_HOST_EXIST); rtw89_write16_mask(rtwdev, R_BE_BOOT_REASON, B_BE_BOOT_REASON_MASK, boot_reason); rtw89_write32_clr(rtwdev, R_BE_PLATFORM_ENABLE, B_BE_WCPU_EN); diff --git a/drivers/net/wireless/realtek/rtw89/reg.h b/drivers/net/wireless/realtek/rtw89/reg.h index 152fd8366dd7..5b4a459cf29c 100644 --- a/drivers/net/wireless/realtek/rtw89/reg.h +++ b/drivers/net/wireless/realtek/rtw89/reg.h @@ -3963,6 +3963,24 @@ #define R_BE_EFUSE_CTRL_1_V1 0x0034 #define B_BE_EF_DATA_MASK GENMASK(31, 0) +#define R_BE_GPIO_MUXCFG 0x0040 +#define B_BE_WCPU_AUTO_EN BIT(26) +#define B_BE_WCPU_JTAG_EN BIT(24) +#define B_BE_WCPU_DBG_EN BIT(23) +#define B_BE_JTAG_CHAIN_EN BIT(20) +#define B_BE_BOOT_MODE BIT(19) +#define B_BE_WL_EECS_EXT_32K_SEL BIT(18) +#define B_BE_WL_SEC_BONDING_OPT_STS BIT(17) +#define B_BE_SECSIC_SEL BIT(16) +#define B_BE_ENHTP BIT(14) +#define B_BE_ENSIC BIT(12) +#define B_BE_SIC_SWRST BIT(11) +#define B_BE_PINMUX_PTA_EN BIT(10) +#define B_BE_WL_BT_PTA_SEC BIT(9) +#define B_BE_ENUARTTX BIT(8) +#define B_BE_DBG_GNT_BT_S1_POLARITY BIT(4) +#define B_BE_ENUARTRX BIT(2) + #define R_BE_GPIO_EXT_CTRL 0x0060 #define B_BE_GPIO_MOD_15_TO_8_MASK GENMASK(31, 24) #define B_BE_GPIO_MOD_9 BIT(25) @@ -4323,6 +4341,7 @@ #define B_BE_RUN_ENV_MASK GENMASK(31, 30) #define B_BE_WCPU_FWDL_STATUS_MASK GENMASK(29, 26) #define B_BE_WDT_PLT_RST_EN BIT(17) +#define B_BE_HOST_EXIST BIT(16) #define B_BE_FW_SEC_AUTH_DONE BIT(14) #define B_BE_FW_CPU_UTIL_STS_EN BIT(13) #define B_BE_BBMCU1_FWDL_EN BIT(12) @@ -4599,6 +4618,10 @@ #define B_BE_HCI_RXDMA_EN BIT(1) #define B_BE_HCI_TXDMA_EN BIT(0) +#define R_BE_BOOT_DBG 0x78F0 +#define B_BE_BOOT_STATUS_MASK GENMASK(31, 16) +#define B_BE_SECUREBOOT_STATUS_MASK GENMASK(15, 0) + #define R_BE_DBG_WOW_READY 0x815E #define B_BE_DBG_WOW_READY GENMASK(7, 0) From 9df55e8d41aa6943846d526487a674316c06e307 Mon Sep 17 00:00:00 2001 From: Ping-Ke Shih Date: Fri, 14 Nov 2025 14:01:21 +0800 Subject: [PATCH 0819/1381] wifi: rtw89: phy: consider type 15 in BB gain table BB gain table is a table to configure gain for certain channels. Newly added type 15 is considered to write registers accordingly. Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251114060128.35363-8-pkshih@realtek.com --- drivers/net/wireless/realtek/rtw89/phy_be.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/net/wireless/realtek/rtw89/phy_be.c b/drivers/net/wireless/realtek/rtw89/phy_be.c index 3316a38a62d0..bd17714f13d1 100644 --- a/drivers/net/wireless/realtek/rtw89/phy_be.c +++ b/drivers/net/wireless/realtek/rtw89/phy_be.c @@ -266,6 +266,10 @@ static void rtw89_phy_config_bb_gain_be(struct rtw89_dev *rtwdev, case 3: rtw89_phy_cfg_bb_gain_op1db_be(rtwdev, arg, reg->data); break; + case 15: + rtw89_phy_write32_idx(rtwdev, reg->addr & 0xFFFFF, MASKHWORD, + reg->data, RTW89_PHY_0); + break; case 4: /* This cfg_type is only used by rfe_type >= 50 with eFEM */ if (efuse->rfe_type < 50) From 5607a141bf283960326f44780b65ab607467a14c Mon Sep 17 00:00:00 2001 From: Ping-Ke Shih Date: Fri, 14 Nov 2025 14:01:22 +0800 Subject: [PATCH 0820/1381] wifi: rtw89: phy: ignore DCFO if not defined in chip_info For WiFi 7 chips, DCFO (digital carrier frequency offset) feature isn't supported, so the corresponding registers aren't defined in chip_info. Check and ignore this feature accordingly. Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251114060128.35363-9-pkshih@realtek.com --- drivers/net/wireless/realtek/rtw89/phy.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/wireless/realtek/rtw89/phy.c b/drivers/net/wireless/realtek/rtw89/phy.c index 28e2b15240a7..08eb3fa6b347 100644 --- a/drivers/net/wireless/realtek/rtw89/phy.c +++ b/drivers/net/wireless/realtek/rtw89/phy.c @@ -4608,7 +4608,7 @@ static void rtw89_dcfo_comp(struct rtw89_dev *rtwdev, s32 curr_cfo) s32 dcfo_comp_val; int sign; - if (rtwdev->chip->chip_id == RTL8922A) + if (!dcfo_comp) return; if (!is_linked) { From 1c2ada09628b6c3de539d45015c62f10f1f00a59 Mon Sep 17 00:00:00 2001 From: Ping-Ke Shih Date: Fri, 14 Nov 2025 14:01:23 +0800 Subject: [PATCH 0821/1381] wifi: rtw89: fw: print band and port where beacon update on The C2H event of BCN_UPD_DONE is to notify driver that firmware changes beacon content on certain band/port asked by driver. Print the notification for debug purpose for now. Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251114060128.35363-10-pkshih@realtek.com --- drivers/net/wireless/realtek/rtw89/fw.h | 9 +++++++++ drivers/net/wireless/realtek/rtw89/mac.c | 13 ++++++++++++- 2 files changed, 21 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/realtek/rtw89/fw.h b/drivers/net/wireless/realtek/rtw89/fw.h index 87c55a1a9d94..bb3e7bbb41de 100644 --- a/drivers/net/wireless/realtek/rtw89/fw.h +++ b/drivers/net/wireless/realtek/rtw89/fw.h @@ -3648,6 +3648,15 @@ struct rtw89_fw_c2h_log_fmt { #define RTW89_C2H_FW_LOG_SIGNATURE 0xA5A5 #define RTW89_C2H_FW_LOG_STR_BUF_SIZE 512 +struct rtw89_c2h_bcn_upd_done { + struct rtw89_c2h_hdr hdr; + __le32 w2; +} __packed; + +#define RTW89_C2H_BCN_UPD_DONE_W2_PORT GENMASK(2, 0) +#define RTW89_C2H_BCN_UPD_DONE_W2_MBSSID GENMASK(6, 3) +#define RTW89_C2H_BCN_UPD_DONE_W2_BAND_IDX BIT(7) + struct rtw89_c2h_mac_bcnfltr_rpt { __le32 w0; __le32 w1; diff --git a/drivers/net/wireless/realtek/rtw89/mac.c b/drivers/net/wireless/realtek/rtw89/mac.c index 21f5023c4f69..386cc982d7c9 100644 --- a/drivers/net/wireless/realtek/rtw89/mac.c +++ b/drivers/net/wireless/realtek/rtw89/mac.c @@ -5309,8 +5309,19 @@ rtw89_mac_c2h_bcn_cnt(struct rtw89_dev *rtwdev, struct sk_buff *c2h, u32 len) } static void -rtw89_mac_c2h_bcn_upd_done(struct rtw89_dev *rtwdev, struct sk_buff *c2h, u32 len) +rtw89_mac_c2h_bcn_upd_done(struct rtw89_dev *rtwdev, struct sk_buff *skb_c2h, u32 len) { + const struct rtw89_c2h_bcn_upd_done *c2h = + (const struct rtw89_c2h_bcn_upd_done *)skb_c2h->data; + u8 band, port, mbssid; + + port = le32_get_bits(c2h->w2, RTW89_C2H_BCN_UPD_DONE_W2_PORT); + mbssid = le32_get_bits(c2h->w2, RTW89_C2H_BCN_UPD_DONE_W2_MBSSID); + band = le32_get_bits(c2h->w2, RTW89_C2H_BCN_UPD_DONE_W2_BAND_IDX); + + rtw89_debug(rtwdev, RTW89_DBG_FW, + "BCN update done on port:%d mbssid:%d band:%d\n", + port, mbssid, band); } static void From aa8d1a8805bb892f01283e168afbb58462f4a94c Mon Sep 17 00:00:00 2001 From: Ping-Ke Shih Date: Fri, 14 Nov 2025 14:01:24 +0800 Subject: [PATCH 0822/1381] wifi: rtw89: align RA H2C format v1 for RTL8922A For RTL8922A, the bits [16:31] of word 3 of v1 format isn't as WiFi 6 chips. Instead, it only needs to fill additional band type and partial bandwidth ER. Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251114060128.35363-11-pkshih@realtek.com --- drivers/net/wireless/realtek/rtw89/core.h | 9 ++++++- drivers/net/wireless/realtek/rtw89/fw.c | 33 ++++++++++++++--------- drivers/net/wireless/realtek/rtw89/fw.h | 7 +++++ drivers/net/wireless/realtek/rtw89/phy.c | 4 +++ 4 files changed, 39 insertions(+), 14 deletions(-) diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h index 8943166de52c..1f99d5f264d2 100644 --- a/drivers/net/wireless/realtek/rtw89/core.h +++ b/drivers/net/wireless/realtek/rtw89/core.h @@ -3384,11 +3384,18 @@ struct rtw89_ra_info { u8 cr_tbl_sel:1; u8 fix_giltf_en:1; u8 fix_giltf:3; - u8 rsvd2:1; + u8 partial_bw_er:1; u8 csi_mcs_ss_idx; u8 csi_mode:2; u8 csi_gi_ltf:3; u8 csi_bw:3; + /* after v1 */ + u8 is_noisy:1; + u8 psra_en:1; + u8 rsvd0:1; + u8 macid_msb:2; + u8 band:2; /* enum rtw89_band */ + u8 is_new_dbgreg:1; }; #define RTW89_PPDU_MAC_INFO_USR_SIZE 4 diff --git a/drivers/net/wireless/realtek/rtw89/fw.c b/drivers/net/wireless/realtek/rtw89/fw.c index b234df109634..0dc5aa5ed218 100644 --- a/drivers/net/wireless/realtek/rtw89/fw.c +++ b/drivers/net/wireless/realtek/rtw89/fw.c @@ -4739,13 +4739,16 @@ int rtw89_fw_h2c_ra(struct rtw89_dev *rtwdev, struct rtw89_ra_info *ra, bool csi struct rtw89_h2c_ra_v1 *h2c_v1; struct rtw89_h2c_ra *h2c; u32 len = sizeof(*h2c); - bool format_v1 = false; struct sk_buff *skb; + u8 ver = U8_MAX; int ret; - if (chip->chip_gen == RTW89_CHIP_BE) { + if (chip->chip_gen == RTW89_CHIP_AX) { + len = sizeof(*h2c); + ver = 0; + } else { len = sizeof(*h2c_v1); - format_v1 = true; + ver = 1; } skb = rtw89_fw_h2c_alloc_skb_with_hdr(rtwdev, len); @@ -4777,16 +4780,8 @@ int rtw89_fw_h2c_ra(struct rtw89_dev *rtwdev, struct rtw89_ra_info *ra, bool csi h2c->w3 = le32_encode_bits(ra->fix_giltf_en, RTW89_H2C_RA_W3_FIX_GILTF_EN) | le32_encode_bits(ra->fix_giltf, RTW89_H2C_RA_W3_FIX_GILTF); - if (!format_v1) - goto csi; - - h2c_v1 = (struct rtw89_h2c_ra_v1 *)h2c; - h2c_v1->w4 = le32_encode_bits(ra->mode_ctrl, RTW89_H2C_RA_V1_W4_MODE_EHT) | - le32_encode_bits(ra->bw_cap, RTW89_H2C_RA_V1_W4_BW_EHT); - -csi: - if (!csi) - goto done; + if (!csi || ver >= 1) + goto next_v1; h2c->w2 |= le32_encode_bits(1, RTW89_H2C_RA_W2_BFEE_CSI_CTL); h2c->w3 |= le32_encode_bits(ra->band_num, RTW89_H2C_RA_W3_BAND_NUM) | @@ -4798,6 +4793,18 @@ int rtw89_fw_h2c_ra(struct rtw89_dev *rtwdev, struct rtw89_ra_info *ra, bool csi le32_encode_bits(ra->csi_gi_ltf, RTW89_H2C_RA_W3_FIXED_CSI_GI_LTF) | le32_encode_bits(ra->csi_bw, RTW89_H2C_RA_W3_FIXED_CSI_BW); +next_v1: + if (ver < 1) + goto done; + + h2c->w3 |= le32_encode_bits(ra->partial_bw_er, + RTW89_H2C_RA_V1_W3_PARTIAL_BW_SU_ER) | + le32_encode_bits(ra->band, RTW89_H2C_RA_V1_W3_BAND); + + h2c_v1 = (struct rtw89_h2c_ra_v1 *)h2c; + h2c_v1->w4 = le32_encode_bits(ra->mode_ctrl, RTW89_H2C_RA_V1_W4_MODE_EHT) | + le32_encode_bits(ra->bw_cap, RTW89_H2C_RA_V1_W4_BW_EHT); + done: rtw89_h2c_pkt_set_hdr(rtwdev, skb, FWCMD_TYPE_H2C, H2C_CAT_OUTSRC, H2C_CL_OUTSRC_RA, diff --git a/drivers/net/wireless/realtek/rtw89/fw.h b/drivers/net/wireless/realtek/rtw89/fw.h index bb3e7bbb41de..e62b61f584fb 100644 --- a/drivers/net/wireless/realtek/rtw89/fw.h +++ b/drivers/net/wireless/realtek/rtw89/fw.h @@ -447,6 +447,13 @@ struct rtw89_h2c_ra { #define RTW89_H2C_RA_W3_FIXED_CSI_MODE GENMASK(25, 24) #define RTW89_H2C_RA_W3_FIXED_CSI_GI_LTF GENMASK(28, 26) #define RTW89_H2C_RA_W3_FIXED_CSI_BW GENMASK(31, 29) +#define RTW89_H2C_RA_V1_W3_PARTIAL_BW_SU_ER BIT(15) +#define RTW89_H2C_RA_V1_W3_FIXED_CSI_RATE_L GENMASK(23, 16) +#define RTW89_H2C_RA_V1_W3_IS_NOISY BIT(24) +#define RTW89_H2C_RA_V1_W3_PSRA_EN BIT(25) +#define RTW89_H2C_RA_V1_W3_MACID_MSB GENMASK(28, 27) +#define RTW89_H2C_RA_V1_W3_BAND GENMASK(30, 29) +#define RTW89_H2C_RA_V1_W3_NEW_DBGREG BIT(31) struct rtw89_h2c_ra_v1 { struct rtw89_h2c_ra v0; diff --git a/drivers/net/wireless/realtek/rtw89/phy.c b/drivers/net/wireless/realtek/rtw89/phy.c index 08eb3fa6b347..9f418b1fb7ed 100644 --- a/drivers/net/wireless/realtek/rtw89/phy.c +++ b/drivers/net/wireless/realtek/rtw89/phy.c @@ -476,6 +476,10 @@ static void rtw89_phy_ra_sta_update(struct rtw89_dev *rtwdev, ra->ra_mask = ra_mask; ra->fix_giltf_en = fix_giltf_en; ra->fix_giltf = fix_giltf; + ra->partial_bw_er = link_sta->he_cap.has_he ? + !!(link_sta->he_cap.he_cap_elem.phy_cap_info[6] & + IEEE80211_HE_PHY_CAP6_PARTIAL_BW_EXT_RANGE) : 0; + ra->band = chan->band_type; if (!csi) return; From 239dd70d776cf94f39000740b307c91e4f72d615 Mon Sep 17 00:00:00 2001 From: Ping-Ke Shih Date: Fri, 14 Nov 2025 14:01:25 +0800 Subject: [PATCH 0823/1381] wifi: rtw89: fill addr cam H2C command by struct The addr cam is used to tell firmware the MAC address and BSSID associated to connected stations. Use struct instead of macros with pointer arithmetic to fill the data. Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251114060128.35363-12-pkshih@realtek.com --- drivers/net/wireless/realtek/rtw89/cam.c | 146 ++++---- drivers/net/wireless/realtek/rtw89/cam.h | 433 +++++------------------ drivers/net/wireless/realtek/rtw89/fw.c | 17 +- 3 files changed, 172 insertions(+), 424 deletions(-) diff --git a/drivers/net/wireless/realtek/rtw89/cam.c b/drivers/net/wireless/realtek/rtw89/cam.c index 385a238fe5cc..8233d91024e8 100644 --- a/drivers/net/wireless/realtek/rtw89/cam.c +++ b/drivers/net/wireless/realtek/rtw89/cam.c @@ -760,7 +760,8 @@ int rtw89_cam_init(struct rtw89_dev *rtwdev, struct rtw89_vif_link *rtwvif_link) int rtw89_cam_fill_bssid_cam_info(struct rtw89_dev *rtwdev, struct rtw89_vif_link *rtwvif_link, - struct rtw89_sta_link *rtwsta_link, u8 *cmd) + struct rtw89_sta_link *rtwsta_link, + struct rtw89_h2c_addr_cam *h2c) { struct rtw89_bssid_cam_entry *bssid_cam = rtw89_get_bssid_cam_of(rtwvif_link, rtwsta_link); @@ -780,20 +781,19 @@ int rtw89_cam_fill_bssid_cam_info(struct rtw89_dev *rtwdev, rcu_read_unlock(); - FWCMD_SET_ADDR_BSSID_IDX(cmd, bssid_cam->bssid_cam_idx); - FWCMD_SET_ADDR_BSSID_OFFSET(cmd, bssid_cam->offset); - FWCMD_SET_ADDR_BSSID_LEN(cmd, bssid_cam->len); - FWCMD_SET_ADDR_BSSID_VALID(cmd, bssid_cam->valid); - FWCMD_SET_ADDR_BSSID_MASK(cmd, bss_mask); - FWCMD_SET_ADDR_BSSID_BB_SEL(cmd, bssid_cam->phy_idx); - FWCMD_SET_ADDR_BSSID_BSS_COLOR(cmd, bss_color); - - FWCMD_SET_ADDR_BSSID_BSSID0(cmd, bssid_cam->bssid[0]); - FWCMD_SET_ADDR_BSSID_BSSID1(cmd, bssid_cam->bssid[1]); - FWCMD_SET_ADDR_BSSID_BSSID2(cmd, bssid_cam->bssid[2]); - FWCMD_SET_ADDR_BSSID_BSSID3(cmd, bssid_cam->bssid[3]); - FWCMD_SET_ADDR_BSSID_BSSID4(cmd, bssid_cam->bssid[4]); - FWCMD_SET_ADDR_BSSID_BSSID5(cmd, bssid_cam->bssid[5]); + h2c->w12 = le32_encode_bits(bssid_cam->bssid_cam_idx, ADDR_CAM_W12_BSSID_IDX) | + le32_encode_bits(bssid_cam->offset, ADDR_CAM_W12_BSSID_OFFSET) | + le32_encode_bits(bssid_cam->len, ADDR_CAM_W12_BSSID_LEN); + h2c->w13 = le32_encode_bits(bssid_cam->valid, ADDR_CAM_W13_BSSID_VALID) | + le32_encode_bits(bss_mask, ADDR_CAM_W13_BSSID_MASK) | + le32_encode_bits(bssid_cam->phy_idx, ADDR_CAM_W13_BSSID_BB_SEL) | + le32_encode_bits(bss_color, ADDR_CAM_W13_BSSID_BSS_COLOR) | + le32_encode_bits(bssid_cam->bssid[0], ADDR_CAM_W13_BSSID_BSSID0) | + le32_encode_bits(bssid_cam->bssid[1], ADDR_CAM_W13_BSSID_BSSID1); + h2c->w14 = le32_encode_bits(bssid_cam->bssid[2], ADDR_CAM_W14_BSSID_BSSID2) | + le32_encode_bits(bssid_cam->bssid[3], ADDR_CAM_W14_BSSID_BSSID3) | + le32_encode_bits(bssid_cam->bssid[4], ADDR_CAM_W14_BSSID_BSSID4) | + le32_encode_bits(bssid_cam->bssid[5], ADDR_CAM_W14_BSSID_BSSID5); return 0; } @@ -813,7 +813,7 @@ void rtw89_cam_fill_addr_cam_info(struct rtw89_dev *rtwdev, struct rtw89_vif_link *rtwvif_link, struct rtw89_sta_link *rtwsta_link, const u8 *scan_mac_addr, - u8 *cmd) + struct rtw89_h2c_addr_cam *h2c) { struct ieee80211_vif *vif = rtwvif_link_to_vif(rtwvif_link); struct rtw89_addr_cam_entry *addr_cam = @@ -845,69 +845,65 @@ void rtw89_cam_fill_addr_cam_info(struct rtw89_dev *rtwdev, sma_hash = rtw89_cam_addr_hash(sma_start, sma); tma_hash = rtw89_cam_addr_hash(tma_start, tma); - FWCMD_SET_ADDR_IDX(cmd, addr_cam->addr_cam_idx); - FWCMD_SET_ADDR_OFFSET(cmd, addr_cam->offset); - FWCMD_SET_ADDR_LEN(cmd, addr_cam->len); + h2c->w1 = le32_encode_bits(addr_cam->addr_cam_idx, ADDR_CAM_W1_IDX) | + le32_encode_bits(addr_cam->offset, ADDR_CAM_W1_OFFSET) | + le32_encode_bits(addr_cam->len, ADDR_CAM_W1_LEN); + h2c->w2 = le32_encode_bits(addr_cam->valid, ADDR_CAM_W2_VALID) | + le32_encode_bits(rtwvif_link->net_type, ADDR_CAM_W2_NET_TYPE) | + le32_encode_bits(rtwvif_link->bcn_hit_cond, ADDR_CAM_W2_BCN_HIT_COND) | + le32_encode_bits(rtwvif_link->hit_rule, ADDR_CAM_W2_HIT_RULE) | + le32_encode_bits(rtwvif_link->phy_idx, ADDR_CAM_W2_BB_SEL) | + le32_encode_bits(addr_cam->addr_mask, ADDR_CAM_W2_ADDR_MASK) | + le32_encode_bits(addr_cam->mask_sel, ADDR_CAM_W2_MASK_SEL) | + le32_encode_bits(sma_hash, ADDR_CAM_W2_SMA_HASH) | + le32_encode_bits(tma_hash, ADDR_CAM_W2_TMA_HASH); + h2c->w3 = le32_encode_bits(addr_cam->bssid_cam_idx, ADDR_CAM_W3_BSSID_CAM_IDX); + h2c->w4 = le32_encode_bits(sma[0], ADDR_CAM_W4_SMA0) | + le32_encode_bits(sma[1], ADDR_CAM_W4_SMA1) | + le32_encode_bits(sma[2], ADDR_CAM_W4_SMA2) | + le32_encode_bits(sma[3], ADDR_CAM_W4_SMA3); + h2c->w5 = le32_encode_bits(sma[4], ADDR_CAM_W5_SMA4) | + le32_encode_bits(sma[5], ADDR_CAM_W5_SMA5) | + le32_encode_bits(tma[0], ADDR_CAM_W5_TMA0) | + le32_encode_bits(tma[1], ADDR_CAM_W5_TMA1); + h2c->w6 = le32_encode_bits(tma[2], ADDR_CAM_W6_TMA2) | + le32_encode_bits(tma[3], ADDR_CAM_W6_TMA3) | + le32_encode_bits(tma[4], ADDR_CAM_W6_TMA4) | + le32_encode_bits(tma[5], ADDR_CAM_W6_TMA5); + h2c->w8 = le32_encode_bits(rtwvif_link->port, ADDR_CAM_W8_PORT_INT) | + le32_encode_bits(rtwvif_link->port, ADDR_CAM_W8_TSF_SYNC) | + le32_encode_bits(rtwvif_link->trigger, ADDR_CAM_W8_TF_TRS) | + le32_encode_bits(rtwvif_link->lsig_txop, ADDR_CAM_W8_LSIG_TXOP) | + le32_encode_bits(rtwvif_link->tgt_ind, ADDR_CAM_W8_TGT_IND) | + le32_encode_bits(rtwvif_link->frm_tgt_ind, ADDR_CAM_W8_FRM_TGT_IND) | + le32_encode_bits(rtwsta_link ? rtwsta_link->mac_id : + rtwvif_link->mac_id, ADDR_CAM_W8_MACID); - FWCMD_SET_ADDR_VALID(cmd, addr_cam->valid); - FWCMD_SET_ADDR_NET_TYPE(cmd, rtwvif_link->net_type); - FWCMD_SET_ADDR_BCN_HIT_COND(cmd, rtwvif_link->bcn_hit_cond); - FWCMD_SET_ADDR_HIT_RULE(cmd, rtwvif_link->hit_rule); - FWCMD_SET_ADDR_BB_SEL(cmd, rtwvif_link->phy_idx); - FWCMD_SET_ADDR_ADDR_MASK(cmd, addr_cam->addr_mask); - FWCMD_SET_ADDR_MASK_SEL(cmd, addr_cam->mask_sel); - FWCMD_SET_ADDR_SMA_HASH(cmd, sma_hash); - FWCMD_SET_ADDR_TMA_HASH(cmd, tma_hash); - - FWCMD_SET_ADDR_BSSID_CAM_IDX(cmd, addr_cam->bssid_cam_idx); - - FWCMD_SET_ADDR_SMA0(cmd, sma[0]); - FWCMD_SET_ADDR_SMA1(cmd, sma[1]); - FWCMD_SET_ADDR_SMA2(cmd, sma[2]); - FWCMD_SET_ADDR_SMA3(cmd, sma[3]); - FWCMD_SET_ADDR_SMA4(cmd, sma[4]); - FWCMD_SET_ADDR_SMA5(cmd, sma[5]); - - FWCMD_SET_ADDR_TMA0(cmd, tma[0]); - FWCMD_SET_ADDR_TMA1(cmd, tma[1]); - FWCMD_SET_ADDR_TMA2(cmd, tma[2]); - FWCMD_SET_ADDR_TMA3(cmd, tma[3]); - FWCMD_SET_ADDR_TMA4(cmd, tma[4]); - FWCMD_SET_ADDR_TMA5(cmd, tma[5]); - - FWCMD_SET_ADDR_PORT_INT(cmd, rtwvif_link->port); - FWCMD_SET_ADDR_TSF_SYNC(cmd, rtwvif_link->port); - FWCMD_SET_ADDR_TF_TRS(cmd, rtwvif_link->trigger); - FWCMD_SET_ADDR_LSIG_TXOP(cmd, rtwvif_link->lsig_txop); - FWCMD_SET_ADDR_TGT_IND(cmd, rtwvif_link->tgt_ind); - FWCMD_SET_ADDR_FRM_TGT_IND(cmd, rtwvif_link->frm_tgt_ind); - FWCMD_SET_ADDR_MACID(cmd, rtwsta_link ? rtwsta_link->mac_id : - rtwvif_link->mac_id); if (rtwvif_link->net_type == RTW89_NET_TYPE_INFRA) - FWCMD_SET_ADDR_AID12(cmd, vif->cfg.aid & 0xfff); + h2c->w9 = le32_encode_bits(vif->cfg.aid & 0xfff, ADDR_CAM_W9_AID12); else if (rtwvif_link->net_type == RTW89_NET_TYPE_AP_MODE) - FWCMD_SET_ADDR_AID12(cmd, sta ? sta->aid & 0xfff : 0); - FWCMD_SET_ADDR_WOL_PATTERN(cmd, rtwvif_link->wowlan_pattern); - FWCMD_SET_ADDR_WOL_UC(cmd, rtwvif_link->wowlan_uc); - FWCMD_SET_ADDR_WOL_MAGIC(cmd, rtwvif_link->wowlan_magic); - FWCMD_SET_ADDR_WAPI(cmd, addr_cam->wapi); - FWCMD_SET_ADDR_SEC_ENT_MODE(cmd, addr_cam->sec_ent_mode); - FWCMD_SET_ADDR_SEC_ENT0_KEYID(cmd, addr_cam->sec_ent_keyid[0]); - FWCMD_SET_ADDR_SEC_ENT1_KEYID(cmd, addr_cam->sec_ent_keyid[1]); - FWCMD_SET_ADDR_SEC_ENT2_KEYID(cmd, addr_cam->sec_ent_keyid[2]); - FWCMD_SET_ADDR_SEC_ENT3_KEYID(cmd, addr_cam->sec_ent_keyid[3]); - FWCMD_SET_ADDR_SEC_ENT4_KEYID(cmd, addr_cam->sec_ent_keyid[4]); - FWCMD_SET_ADDR_SEC_ENT5_KEYID(cmd, addr_cam->sec_ent_keyid[5]); - FWCMD_SET_ADDR_SEC_ENT6_KEYID(cmd, addr_cam->sec_ent_keyid[6]); + h2c->w9 = le32_encode_bits(sta ? sta->aid & 0xfff : 0, ADDR_CAM_W9_AID12); - FWCMD_SET_ADDR_SEC_ENT_VALID(cmd, addr_cam->sec_cam_map[0] & 0xff); - FWCMD_SET_ADDR_SEC_ENT0(cmd, addr_cam->sec_ent[0]); - FWCMD_SET_ADDR_SEC_ENT1(cmd, addr_cam->sec_ent[1]); - FWCMD_SET_ADDR_SEC_ENT2(cmd, addr_cam->sec_ent[2]); - FWCMD_SET_ADDR_SEC_ENT3(cmd, addr_cam->sec_ent[3]); - FWCMD_SET_ADDR_SEC_ENT4(cmd, addr_cam->sec_ent[4]); - FWCMD_SET_ADDR_SEC_ENT5(cmd, addr_cam->sec_ent[5]); - FWCMD_SET_ADDR_SEC_ENT6(cmd, addr_cam->sec_ent[6]); + h2c->w9 |= le32_encode_bits(rtwvif_link->wowlan_pattern, ADDR_CAM_W9_WOL_PATTERN) | + le32_encode_bits(rtwvif_link->wowlan_uc, ADDR_CAM_W9_WOL_UC) | + le32_encode_bits(rtwvif_link->wowlan_magic, ADDR_CAM_W9_WOL_MAGIC) | + le32_encode_bits(addr_cam->wapi, ADDR_CAM_W9_WAPI) | + le32_encode_bits(addr_cam->sec_ent_mode, ADDR_CAM_W9_SEC_ENT_MODE) | + le32_encode_bits(addr_cam->sec_ent_keyid[0], ADDR_CAM_W9_SEC_ENT0_KEYID) | + le32_encode_bits(addr_cam->sec_ent_keyid[1], ADDR_CAM_W9_SEC_ENT1_KEYID) | + le32_encode_bits(addr_cam->sec_ent_keyid[2], ADDR_CAM_W9_SEC_ENT2_KEYID) | + le32_encode_bits(addr_cam->sec_ent_keyid[3], ADDR_CAM_W9_SEC_ENT3_KEYID) | + le32_encode_bits(addr_cam->sec_ent_keyid[4], ADDR_CAM_W9_SEC_ENT4_KEYID) | + le32_encode_bits(addr_cam->sec_ent_keyid[5], ADDR_CAM_W9_SEC_ENT5_KEYID) | + le32_encode_bits(addr_cam->sec_ent_keyid[6], ADDR_CAM_W9_SEC_ENT6_KEYID); + h2c->w10 = le32_encode_bits(addr_cam->sec_cam_map[0] & 0xff, ADDR_CAM_W10_SEC_ENT_VALID) | + le32_encode_bits(addr_cam->sec_ent[0], ADDR_CAM_W10_SEC_ENT0) | + le32_encode_bits(addr_cam->sec_ent[1], ADDR_CAM_W10_SEC_ENT1) | + le32_encode_bits(addr_cam->sec_ent[2], ADDR_CAM_W10_SEC_ENT2); + h2c->w11 = le32_encode_bits(addr_cam->sec_ent[3], ADDR_CAM_W11_SEC_ENT3) | + le32_encode_bits(addr_cam->sec_ent[4], ADDR_CAM_W11_SEC_ENT4) | + le32_encode_bits(addr_cam->sec_ent[5], ADDR_CAM_W11_SEC_ENT5) | + le32_encode_bits(addr_cam->sec_ent[6], ADDR_CAM_W11_SEC_ENT6); rcu_read_unlock(); } diff --git a/drivers/net/wireless/realtek/rtw89/cam.h b/drivers/net/wireless/realtek/rtw89/cam.h index 8fd2d776408e..2bc8fbf79c0b 100644 --- a/drivers/net/wireless/realtek/rtw89/cam.h +++ b/drivers/net/wireless/realtek/rtw89/cam.h @@ -12,345 +12,92 @@ #define RTW89_BSSID_MATCH_ALL GENMASK(5, 0) #define RTW89_BSSID_MATCH_5_BYTES GENMASK(4, 0) -static inline void FWCMD_SET_ADDR_IDX(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 1, value, GENMASK(7, 0)); -} - -static inline void FWCMD_SET_ADDR_OFFSET(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 1, value, GENMASK(15, 8)); -} - -static inline void FWCMD_SET_ADDR_LEN(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 1, value, GENMASK(23, 16)); -} - -static inline void FWCMD_SET_ADDR_VALID(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 2, value, BIT(0)); -} - -static inline void FWCMD_SET_ADDR_NET_TYPE(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 2, value, GENMASK(2, 1)); -} - -static inline void FWCMD_SET_ADDR_BCN_HIT_COND(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 2, value, GENMASK(4, 3)); -} - -static inline void FWCMD_SET_ADDR_HIT_RULE(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 2, value, GENMASK(6, 5)); -} - -static inline void FWCMD_SET_ADDR_BB_SEL(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 2, value, BIT(7)); -} - -static inline void FWCMD_SET_ADDR_ADDR_MASK(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 2, value, GENMASK(13, 8)); -} - -static inline void FWCMD_SET_ADDR_MASK_SEL(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 2, value, GENMASK(15, 14)); -} - -static inline void FWCMD_SET_ADDR_SMA_HASH(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 2, value, GENMASK(23, 16)); -} - -static inline void FWCMD_SET_ADDR_TMA_HASH(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 2, value, GENMASK(31, 24)); -} - -static inline void FWCMD_SET_ADDR_BSSID_CAM_IDX(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 3, value, GENMASK(5, 0)); -} - -static inline void FWCMD_SET_ADDR_SMA0(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 4, value, GENMASK(7, 0)); -} - -static inline void FWCMD_SET_ADDR_SMA1(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 4, value, GENMASK(15, 8)); -} - -static inline void FWCMD_SET_ADDR_SMA2(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 4, value, GENMASK(23, 16)); -} - -static inline void FWCMD_SET_ADDR_SMA3(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 4, value, GENMASK(31, 24)); -} - -static inline void FWCMD_SET_ADDR_SMA4(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 5, value, GENMASK(7, 0)); -} - -static inline void FWCMD_SET_ADDR_SMA5(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 5, value, GENMASK(15, 8)); -} - -static inline void FWCMD_SET_ADDR_TMA0(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 5, value, GENMASK(23, 16)); -} - -static inline void FWCMD_SET_ADDR_TMA1(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 5, value, GENMASK(31, 24)); -} - -static inline void FWCMD_SET_ADDR_TMA2(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 6, value, GENMASK(7, 0)); -} - -static inline void FWCMD_SET_ADDR_TMA3(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 6, value, GENMASK(15, 8)); -} - -static inline void FWCMD_SET_ADDR_TMA4(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 6, value, GENMASK(23, 16)); -} - -static inline void FWCMD_SET_ADDR_TMA5(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 6, value, GENMASK(31, 24)); -} - -static inline void FWCMD_SET_ADDR_MACID(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 8, value, GENMASK(7, 0)); -} - -static inline void FWCMD_SET_ADDR_PORT_INT(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 8, value, GENMASK(10, 8)); -} - -static inline void FWCMD_SET_ADDR_TSF_SYNC(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 8, value, GENMASK(13, 11)); -} - -static inline void FWCMD_SET_ADDR_TF_TRS(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 8, value, BIT(14)); -} - -static inline void FWCMD_SET_ADDR_LSIG_TXOP(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 8, value, BIT(15)); -} - -static inline void FWCMD_SET_ADDR_TGT_IND(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 8, value, GENMASK(26, 24)); -} - -static inline void FWCMD_SET_ADDR_FRM_TGT_IND(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 8, value, GENMASK(29, 27)); -} - -static inline void FWCMD_SET_ADDR_AID12(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 9, value, GENMASK(11, 0)); -} - -static inline void FWCMD_SET_ADDR_AID12_0(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 9, value, GENMASK(7, 0)); -} - -static inline void FWCMD_SET_ADDR_AID12_1(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 9, value, GENMASK(11, 8)); -} - -static inline void FWCMD_SET_ADDR_WOL_PATTERN(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 9, value, BIT(12)); -} - -static inline void FWCMD_SET_ADDR_WOL_UC(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 9, value, BIT(13)); -} - -static inline void FWCMD_SET_ADDR_WOL_MAGIC(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 9, value, BIT(14)); -} - -static inline void FWCMD_SET_ADDR_WAPI(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 9, value, BIT(15)); -} - -static inline void FWCMD_SET_ADDR_SEC_ENT_MODE(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 9, value, GENMASK(17, 16)); -} - -static inline void FWCMD_SET_ADDR_SEC_ENT0_KEYID(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 9, value, GENMASK(19, 18)); -} - -static inline void FWCMD_SET_ADDR_SEC_ENT1_KEYID(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 9, value, GENMASK(21, 20)); -} - -static inline void FWCMD_SET_ADDR_SEC_ENT2_KEYID(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 9, value, GENMASK(23, 22)); -} - -static inline void FWCMD_SET_ADDR_SEC_ENT3_KEYID(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 9, value, GENMASK(25, 24)); -} - -static inline void FWCMD_SET_ADDR_SEC_ENT4_KEYID(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 9, value, GENMASK(27, 26)); -} - -static inline void FWCMD_SET_ADDR_SEC_ENT5_KEYID(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 9, value, GENMASK(29, 28)); -} - -static inline void FWCMD_SET_ADDR_SEC_ENT6_KEYID(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 9, value, GENMASK(31, 30)); -} - -static inline void FWCMD_SET_ADDR_SEC_ENT_VALID(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 10, value, GENMASK(7, 0)); -} - -static inline void FWCMD_SET_ADDR_SEC_ENT0(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 10, value, GENMASK(15, 8)); -} - -static inline void FWCMD_SET_ADDR_SEC_ENT1(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 10, value, GENMASK(23, 16)); -} - -static inline void FWCMD_SET_ADDR_SEC_ENT2(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 10, value, GENMASK(31, 24)); -} - -static inline void FWCMD_SET_ADDR_SEC_ENT3(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 11, value, GENMASK(7, 0)); -} - -static inline void FWCMD_SET_ADDR_SEC_ENT4(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 11, value, GENMASK(15, 8)); -} - -static inline void FWCMD_SET_ADDR_SEC_ENT5(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 11, value, GENMASK(23, 16)); -} - -static inline void FWCMD_SET_ADDR_SEC_ENT6(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 11, value, GENMASK(31, 24)); -} - -static inline void FWCMD_SET_ADDR_BSSID_IDX(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 12, value, GENMASK(7, 0)); -} - -static inline void FWCMD_SET_ADDR_BSSID_OFFSET(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 12, value, GENMASK(15, 8)); -} - -static inline void FWCMD_SET_ADDR_BSSID_LEN(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 12, value, GENMASK(23, 16)); -} - -static inline void FWCMD_SET_ADDR_BSSID_VALID(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 13, value, BIT(0)); -} - -static inline void FWCMD_SET_ADDR_BSSID_BB_SEL(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 13, value, BIT(1)); -} - -static inline void FWCMD_SET_ADDR_BSSID_MASK(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 13, value, GENMASK(7, 2)); -} - -static inline void FWCMD_SET_ADDR_BSSID_BSS_COLOR(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 13, value, GENMASK(13, 8)); -} - -static inline void FWCMD_SET_ADDR_BSSID_BSSID0(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 13, value, GENMASK(23, 16)); -} - -static inline void FWCMD_SET_ADDR_BSSID_BSSID1(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 13, value, GENMASK(31, 24)); -} - -static inline void FWCMD_SET_ADDR_BSSID_BSSID2(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 14, value, GENMASK(7, 0)); -} - -static inline void FWCMD_SET_ADDR_BSSID_BSSID3(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 14, value, GENMASK(15, 8)); -} - -static inline void FWCMD_SET_ADDR_BSSID_BSSID4(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 14, value, GENMASK(23, 16)); -} - -static inline void FWCMD_SET_ADDR_BSSID_BSSID5(void *cmd, u32 value) -{ - le32p_replace_bits((__le32 *)(cmd) + 14, value, GENMASK(31, 24)); -} +struct rtw89_h2c_addr_cam { + __le32 w0; + __le32 w1; + __le32 w2; + __le32 w3; + __le32 w4; + __le32 w5; + __le32 w6; + __le32 w7; + __le32 w8; + __le32 w9; + __le32 w10; + __le32 w11; + __le32 w12; + __le32 w13; + __le32 w14; +} __packed; + +#define ADDR_CAM_W1_IDX GENMASK(7, 0) +#define ADDR_CAM_W1_OFFSET GENMASK(15, 8) +#define ADDR_CAM_W1_LEN GENMASK(23, 16) +#define ADDR_CAM_W2_VALID BIT(0) +#define ADDR_CAM_W2_NET_TYPE GENMASK(2, 1) +#define ADDR_CAM_W2_BCN_HIT_COND GENMASK(4, 3) +#define ADDR_CAM_W2_HIT_RULE GENMASK(6, 5) +#define ADDR_CAM_W2_BB_SEL BIT(7) +#define ADDR_CAM_W2_ADDR_MASK GENMASK(13, 8) +#define ADDR_CAM_W2_MASK_SEL GENMASK(15, 14) +#define ADDR_CAM_W2_SMA_HASH GENMASK(23, 16) +#define ADDR_CAM_W2_TMA_HASH GENMASK(31, 24) +#define ADDR_CAM_W3_BSSID_CAM_IDX GENMASK(5, 0) +#define ADDR_CAM_W4_SMA0 GENMASK(7, 0) +#define ADDR_CAM_W4_SMA1 GENMASK(15, 8) +#define ADDR_CAM_W4_SMA2 GENMASK(23, 16) +#define ADDR_CAM_W4_SMA3 GENMASK(31, 24) +#define ADDR_CAM_W5_SMA4 GENMASK(7, 0) +#define ADDR_CAM_W5_SMA5 GENMASK(15, 8) +#define ADDR_CAM_W5_TMA0 GENMASK(23, 16) +#define ADDR_CAM_W5_TMA1 GENMASK(31, 24) +#define ADDR_CAM_W6_TMA2 GENMASK(7, 0) +#define ADDR_CAM_W6_TMA3 GENMASK(15, 8) +#define ADDR_CAM_W6_TMA4 GENMASK(23, 16) +#define ADDR_CAM_W6_TMA5 GENMASK(31, 24) +#define ADDR_CAM_W8_MACID GENMASK(7, 0) +#define ADDR_CAM_W8_PORT_INT GENMASK(10, 8) +#define ADDR_CAM_W8_TSF_SYNC GENMASK(13, 11) +#define ADDR_CAM_W8_TF_TRS BIT(14) +#define ADDR_CAM_W8_LSIG_TXOP BIT(15) +#define ADDR_CAM_W8_TGT_IND GENMASK(26, 24) +#define ADDR_CAM_W8_FRM_TGT_IND GENMASK(29, 27) +#define ADDR_CAM_W9_AID12 GENMASK(11, 0) +#define ADDR_CAM_W9_AID12_0 GENMASK(7, 0) +#define ADDR_CAM_W9_AID12_1 GENMASK(11, 8) +#define ADDR_CAM_W9_WOL_PATTERN BIT(12) +#define ADDR_CAM_W9_WOL_UC BIT(13) +#define ADDR_CAM_W9_WOL_MAGIC BIT(14) +#define ADDR_CAM_W9_WAPI BIT(15) +#define ADDR_CAM_W9_SEC_ENT_MODE GENMASK(17, 16) +#define ADDR_CAM_W9_SEC_ENT0_KEYID GENMASK(19, 18) +#define ADDR_CAM_W9_SEC_ENT1_KEYID GENMASK(21, 20) +#define ADDR_CAM_W9_SEC_ENT2_KEYID GENMASK(23, 22) +#define ADDR_CAM_W9_SEC_ENT3_KEYID GENMASK(25, 24) +#define ADDR_CAM_W9_SEC_ENT4_KEYID GENMASK(27, 26) +#define ADDR_CAM_W9_SEC_ENT5_KEYID GENMASK(29, 28) +#define ADDR_CAM_W9_SEC_ENT6_KEYID GENMASK(31, 30) +#define ADDR_CAM_W10_SEC_ENT_VALID GENMASK(7, 0) +#define ADDR_CAM_W10_SEC_ENT0 GENMASK(15, 8) +#define ADDR_CAM_W10_SEC_ENT1 GENMASK(23, 16) +#define ADDR_CAM_W10_SEC_ENT2 GENMASK(31, 24) +#define ADDR_CAM_W11_SEC_ENT3 GENMASK(7, 0) +#define ADDR_CAM_W11_SEC_ENT4 GENMASK(15, 8) +#define ADDR_CAM_W11_SEC_ENT5 GENMASK(23, 16) +#define ADDR_CAM_W11_SEC_ENT6 GENMASK(31, 24) +#define ADDR_CAM_W12_BSSID_IDX GENMASK(7, 0) +#define ADDR_CAM_W12_BSSID_OFFSET GENMASK(15, 8) +#define ADDR_CAM_W12_BSSID_LEN GENMASK(23, 16) +#define ADDR_CAM_W13_BSSID_VALID BIT(0) +#define ADDR_CAM_W13_BSSID_BB_SEL BIT(1) +#define ADDR_CAM_W13_BSSID_MASK GENMASK(7, 2) +#define ADDR_CAM_W13_BSSID_BSS_COLOR GENMASK(13, 8) +#define ADDR_CAM_W13_BSSID_BSSID0 GENMASK(23, 16) +#define ADDR_CAM_W13_BSSID_BSSID1 GENMASK(31, 24) +#define ADDR_CAM_W14_BSSID_BSSID2 GENMASK(7, 0) +#define ADDR_CAM_W14_BSSID_BSSID3 GENMASK(15, 8) +#define ADDR_CAM_W14_BSSID_BSSID4 GENMASK(23, 16) +#define ADDR_CAM_W14_BSSID_BSSID5 GENMASK(31, 24) struct rtw89_h2c_dctlinfo_ud_v1 { __le32 c0; @@ -552,9 +299,10 @@ int rtw89_cam_init_bssid_cam(struct rtw89_dev *rtwdev, void rtw89_cam_deinit_bssid_cam(struct rtw89_dev *rtwdev, struct rtw89_bssid_cam_entry *bssid_cam); void rtw89_cam_fill_addr_cam_info(struct rtw89_dev *rtwdev, - struct rtw89_vif_link *vif, + struct rtw89_vif_link *rtwvif_link, struct rtw89_sta_link *rtwsta_link, - const u8 *scan_mac_addr, u8 *cmd); + const u8 *scan_mac_addr, + struct rtw89_h2c_addr_cam *h2c); void rtw89_cam_fill_dctl_sec_cam_info_v1(struct rtw89_dev *rtwdev, struct rtw89_vif_link *rtwvif_link, struct rtw89_sta_link *rtwsta_link, @@ -565,7 +313,8 @@ void rtw89_cam_fill_dctl_sec_cam_info_v2(struct rtw89_dev *rtwdev, struct rtw89_h2c_dctlinfo_ud_v2 *h2c); int rtw89_cam_fill_bssid_cam_info(struct rtw89_dev *rtwdev, struct rtw89_vif_link *rtwvif_link, - struct rtw89_sta_link *rtwsta_link, u8 *cmd); + struct rtw89_sta_link *rtwsta_link, + struct rtw89_h2c_addr_cam *h2c); int rtw89_cam_sec_key_add(struct rtw89_dev *rtwdev, struct ieee80211_vif *vif, struct ieee80211_sta *sta, diff --git a/drivers/net/wireless/realtek/rtw89/fw.c b/drivers/net/wireless/realtek/rtw89/fw.c index 0dc5aa5ed218..9f132a469629 100644 --- a/drivers/net/wireless/realtek/rtw89/fw.c +++ b/drivers/net/wireless/realtek/rtw89/fw.c @@ -2132,28 +2132,31 @@ void rtw89_fw_log_dump(struct rtw89_dev *rtwdev, u8 *buf, u32 len) } -#define H2C_CAM_LEN 60 int rtw89_fw_h2c_cam(struct rtw89_dev *rtwdev, struct rtw89_vif_link *rtwvif_link, struct rtw89_sta_link *rtwsta_link, const u8 *scan_mac_addr) { + struct rtw89_h2c_addr_cam *h2c; + u32 len = sizeof(*h2c); struct sk_buff *skb; int ret; - skb = rtw89_fw_h2c_alloc_skb_with_hdr(rtwdev, H2C_CAM_LEN); + skb = rtw89_fw_h2c_alloc_skb_with_hdr(rtwdev, len); if (!skb) { rtw89_err(rtwdev, "failed to alloc skb for fw dl\n"); return -ENOMEM; } - skb_put(skb, H2C_CAM_LEN); - rtw89_cam_fill_addr_cam_info(rtwdev, rtwvif_link, rtwsta_link, scan_mac_addr, - skb->data); - rtw89_cam_fill_bssid_cam_info(rtwdev, rtwvif_link, rtwsta_link, skb->data); + skb_put(skb, len); + h2c = (struct rtw89_h2c_addr_cam *)skb->data; + + rtw89_cam_fill_addr_cam_info(rtwdev, rtwvif_link, rtwsta_link, + scan_mac_addr, h2c); + rtw89_cam_fill_bssid_cam_info(rtwdev, rtwvif_link, rtwsta_link, h2c); rtw89_h2c_pkt_set_hdr(rtwdev, skb, FWCMD_TYPE_H2C, H2C_CAT_MAC, H2C_CL_MAC_ADDR_CAM_UPDATE, H2C_FUNC_MAC_ADDR_CAM_UPD, 0, 1, - H2C_CAM_LEN); + len); ret = rtw89_h2c_tx(rtwdev, skb, false); if (ret) { From fc2e8c873f4620f07a34f1acd5697187c3ca8a8c Mon Sep 17 00:00:00 2001 From: Ping-Ke Shih Date: Fri, 14 Nov 2025 14:01:26 +0800 Subject: [PATCH 0824/1381] wifi: rtw89: add addr cam H2C command v1 The coming RTL8922D uses different format of address CAM command, so add the new format accordingly. Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251114060128.35363-13-pkshih@realtek.com --- drivers/net/wireless/realtek/rtw89/cam.c | 39 +++++++++++++------ drivers/net/wireless/realtek/rtw89/cam.h | 11 ++++++ drivers/net/wireless/realtek/rtw89/core.h | 1 + drivers/net/wireless/realtek/rtw89/rtw8851b.c | 1 + drivers/net/wireless/realtek/rtw89/rtw8852a.c | 1 + drivers/net/wireless/realtek/rtw89/rtw8852b.c | 1 + .../net/wireless/realtek/rtw89/rtw8852bt.c | 1 + drivers/net/wireless/realtek/rtw89/rtw8852c.c | 1 + drivers/net/wireless/realtek/rtw89/rtw8922a.c | 1 + 9 files changed, 46 insertions(+), 11 deletions(-) diff --git a/drivers/net/wireless/realtek/rtw89/cam.c b/drivers/net/wireless/realtek/rtw89/cam.c index 8233d91024e8..5f282a02a356 100644 --- a/drivers/net/wireless/realtek/rtw89/cam.c +++ b/drivers/net/wireless/realtek/rtw89/cam.c @@ -819,12 +819,15 @@ void rtw89_cam_fill_addr_cam_info(struct rtw89_dev *rtwdev, struct rtw89_addr_cam_entry *addr_cam = rtw89_get_addr_cam_of(rtwvif_link, rtwsta_link); struct ieee80211_sta *sta = rtwsta_link_to_sta_safe(rtwsta_link); + const struct rtw89_chip_info *chip = rtwdev->chip; struct ieee80211_link_sta *link_sta; const u8 *sma = scan_mac_addr ? scan_mac_addr : rtwvif_link->mac_addr; u8 sma_hash, tma_hash, addr_msk_start; + u8 ver = chip->addrcam_ver; u8 sma_start = 0; u8 tma_start = 0; const u8 *tma; + u8 mac_id; rcu_read_lock(); @@ -845,9 +848,17 @@ void rtw89_cam_fill_addr_cam_info(struct rtw89_dev *rtwdev, sma_hash = rtw89_cam_addr_hash(sma_start, sma); tma_hash = rtw89_cam_addr_hash(tma_start, tma); - h2c->w1 = le32_encode_bits(addr_cam->addr_cam_idx, ADDR_CAM_W1_IDX) | - le32_encode_bits(addr_cam->offset, ADDR_CAM_W1_OFFSET) | - le32_encode_bits(addr_cam->len, ADDR_CAM_W1_LEN); + mac_id = rtwsta_link ? rtwsta_link->mac_id : rtwvif_link->mac_id; + + if (ver == 0) + h2c->w1 = le32_encode_bits(addr_cam->addr_cam_idx, ADDR_CAM_W1_IDX) | + le32_encode_bits(addr_cam->offset, ADDR_CAM_W1_OFFSET) | + le32_encode_bits(addr_cam->len, ADDR_CAM_W1_LEN); + else + h2c->w1 = le32_encode_bits(addr_cam->addr_cam_idx, ADDR_CAM_W1_V1_IDX) | + le32_encode_bits(addr_cam->offset, ADDR_CAM_W1_V1_OFFSET) | + le32_encode_bits(addr_cam->len, ADDR_CAM_W1_V1_LEN); + h2c->w2 = le32_encode_bits(addr_cam->valid, ADDR_CAM_W2_VALID) | le32_encode_bits(rtwvif_link->net_type, ADDR_CAM_W2_NET_TYPE) | le32_encode_bits(rtwvif_link->bcn_hit_cond, ADDR_CAM_W2_BCN_HIT_COND) | @@ -870,14 +881,20 @@ void rtw89_cam_fill_addr_cam_info(struct rtw89_dev *rtwdev, le32_encode_bits(tma[3], ADDR_CAM_W6_TMA3) | le32_encode_bits(tma[4], ADDR_CAM_W6_TMA4) | le32_encode_bits(tma[5], ADDR_CAM_W6_TMA5); - h2c->w8 = le32_encode_bits(rtwvif_link->port, ADDR_CAM_W8_PORT_INT) | - le32_encode_bits(rtwvif_link->port, ADDR_CAM_W8_TSF_SYNC) | - le32_encode_bits(rtwvif_link->trigger, ADDR_CAM_W8_TF_TRS) | - le32_encode_bits(rtwvif_link->lsig_txop, ADDR_CAM_W8_LSIG_TXOP) | - le32_encode_bits(rtwvif_link->tgt_ind, ADDR_CAM_W8_TGT_IND) | - le32_encode_bits(rtwvif_link->frm_tgt_ind, ADDR_CAM_W8_FRM_TGT_IND) | - le32_encode_bits(rtwsta_link ? rtwsta_link->mac_id : - rtwvif_link->mac_id, ADDR_CAM_W8_MACID); + if (ver == 0) + h2c->w8 = le32_encode_bits(rtwvif_link->port, ADDR_CAM_W8_PORT_INT) | + le32_encode_bits(rtwvif_link->port, ADDR_CAM_W8_TSF_SYNC) | + le32_encode_bits(rtwvif_link->trigger, ADDR_CAM_W8_TF_TRS) | + le32_encode_bits(rtwvif_link->lsig_txop, ADDR_CAM_W8_LSIG_TXOP) | + le32_encode_bits(rtwvif_link->tgt_ind, ADDR_CAM_W8_TGT_IND) | + le32_encode_bits(rtwvif_link->frm_tgt_ind, ADDR_CAM_W8_FRM_TGT_IND) | + le32_encode_bits(mac_id, ADDR_CAM_W8_MACID); + else + h2c->w8 = le32_encode_bits(rtwvif_link->port, ADDR_CAM_W8_V1_PORT_INT) | + le32_encode_bits(rtwvif_link->port, ADDR_CAM_W8_V1_TSF_SYNC) | + le32_encode_bits(rtwvif_link->trigger, ADDR_CAM_W8_V1_TF_TRS) | + le32_encode_bits(rtwvif_link->lsig_txop, ADDR_CAM_W8_V1_LSIG_TXOP) | + le32_encode_bits(mac_id, ADDR_CAM_W8_V1_MACID); if (rtwvif_link->net_type == RTW89_NET_TYPE_INFRA) h2c->w9 = le32_encode_bits(vif->cfg.aid & 0xfff, ADDR_CAM_W9_AID12); diff --git a/drivers/net/wireless/realtek/rtw89/cam.h b/drivers/net/wireless/realtek/rtw89/cam.h index 2bc8fbf79c0b..833a956f0176 100644 --- a/drivers/net/wireless/realtek/rtw89/cam.h +++ b/drivers/net/wireless/realtek/rtw89/cam.h @@ -33,6 +33,9 @@ struct rtw89_h2c_addr_cam { #define ADDR_CAM_W1_IDX GENMASK(7, 0) #define ADDR_CAM_W1_OFFSET GENMASK(15, 8) #define ADDR_CAM_W1_LEN GENMASK(23, 16) +#define ADDR_CAM_W1_V1_IDX GENMASK(9, 0) +#define ADDR_CAM_W1_V1_OFFSET GENMASK(23, 16) +#define ADDR_CAM_W1_V1_LEN GENMASK(31, 24) #define ADDR_CAM_W2_VALID BIT(0) #define ADDR_CAM_W2_NET_TYPE GENMASK(2, 1) #define ADDR_CAM_W2_BCN_HIT_COND GENMASK(4, 3) @@ -62,6 +65,14 @@ struct rtw89_h2c_addr_cam { #define ADDR_CAM_W8_LSIG_TXOP BIT(15) #define ADDR_CAM_W8_TGT_IND GENMASK(26, 24) #define ADDR_CAM_W8_FRM_TGT_IND GENMASK(29, 27) +#define ADDR_CAM_W8_V1_MACID GENMASK(9, 0) +#define ADDR_CAM_W8_V1_PORT_INT GENMASK(18, 16) +#define ADDR_CAM_W8_V1_TSF_SYNC GENMASK(21, 19) +#define ADDR_CAM_W8_V1_TF_TRS BIT(22) +#define ADDR_CAM_W8_V1_LSIG_TXOP BIT(23) +#define ADDR_CAM_W8_V1_TB_RANGING BIT(24) +#define ADDR_CAM_W8_V1_TB_SENSING BIT(25) +#define ADDR_CAM_W8_V1_SENS_EN BIT(26) #define ADDR_CAM_W9_AID12 GENMASK(11, 0) #define ADDR_CAM_W9_AID12_0 GENMASK(7, 0) #define ADDR_CAM_W9_AID12_1 GENMASK(11, 8) diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h index 1f99d5f264d2..c33ff7c0090f 100644 --- a/drivers/net/wireless/realtek/rtw89/core.h +++ b/drivers/net/wireless/realtek/rtw89/core.h @@ -4458,6 +4458,7 @@ struct rtw89_chip_info { u8 bacam_num; u8 bacam_dynamic_num; enum rtw89_bacam_ver bacam_ver; + u8 addrcam_ver; u8 ppdu_max_usr; u8 sec_ctrl_efuse_size; diff --git a/drivers/net/wireless/realtek/rtw89/rtw8851b.c b/drivers/net/wireless/realtek/rtw89/rtw8851b.c index 2019f6022cbb..84b628d23882 100644 --- a/drivers/net/wireless/realtek/rtw89/rtw8851b.c +++ b/drivers/net/wireless/realtek/rtw89/rtw8851b.c @@ -2648,6 +2648,7 @@ const struct rtw89_chip_info rtw8851b_chip_info = { .bacam_num = 2, .bacam_dynamic_num = 4, .bacam_ver = RTW89_BACAM_V0, + .addrcam_ver = 0, .ppdu_max_usr = 4, .sec_ctrl_efuse_size = 4, .physical_efuse_size = 1216, diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852a.c b/drivers/net/wireless/realtek/rtw89/rtw8852a.c index 48205aa4a980..8677723e3561 100644 --- a/drivers/net/wireless/realtek/rtw89/rtw8852a.c +++ b/drivers/net/wireless/realtek/rtw89/rtw8852a.c @@ -2334,6 +2334,7 @@ const struct rtw89_chip_info rtw8852a_chip_info = { .bacam_num = 2, .bacam_dynamic_num = 4, .bacam_ver = RTW89_BACAM_V0, + .addrcam_ver = 0, .ppdu_max_usr = 4, .sec_ctrl_efuse_size = 4, .physical_efuse_size = 1216, diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852b.c b/drivers/net/wireless/realtek/rtw89/rtw8852b.c index 38cd151f8c3f..70fb05bc5e98 100644 --- a/drivers/net/wireless/realtek/rtw89/rtw8852b.c +++ b/drivers/net/wireless/realtek/rtw89/rtw8852b.c @@ -959,6 +959,7 @@ const struct rtw89_chip_info rtw8852b_chip_info = { .bacam_num = 2, .bacam_dynamic_num = 4, .bacam_ver = RTW89_BACAM_V0, + .addrcam_ver = 0, .ppdu_max_usr = 4, .sec_ctrl_efuse_size = 4, .physical_efuse_size = 1216, diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852bt.c b/drivers/net/wireless/realtek/rtw89/rtw8852bt.c index 15ba780492d6..f956474c3b72 100644 --- a/drivers/net/wireless/realtek/rtw89/rtw8852bt.c +++ b/drivers/net/wireless/realtek/rtw89/rtw8852bt.c @@ -818,6 +818,7 @@ const struct rtw89_chip_info rtw8852bt_chip_info = { .bacam_num = 2, .bacam_dynamic_num = 4, .bacam_ver = RTW89_BACAM_V0, + .addrcam_ver = 0, .ppdu_max_usr = 4, .sec_ctrl_efuse_size = 4, .physical_efuse_size = 1216, diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852c.c b/drivers/net/wireless/realtek/rtw89/rtw8852c.c index ee81a6792eee..db99450e9158 100644 --- a/drivers/net/wireless/realtek/rtw89/rtw8852c.c +++ b/drivers/net/wireless/realtek/rtw89/rtw8852c.c @@ -3178,6 +3178,7 @@ const struct rtw89_chip_info rtw8852c_chip_info = { .bacam_num = 8, .bacam_dynamic_num = 8, .bacam_ver = RTW89_BACAM_V0_EXT, + .addrcam_ver = 0, .ppdu_max_usr = 8, .sec_ctrl_efuse_size = 4, .physical_efuse_size = 1216, diff --git a/drivers/net/wireless/realtek/rtw89/rtw8922a.c b/drivers/net/wireless/realtek/rtw89/rtw8922a.c index 11e8b50a3291..4437279c554b 100644 --- a/drivers/net/wireless/realtek/rtw89/rtw8922a.c +++ b/drivers/net/wireless/realtek/rtw89/rtw8922a.c @@ -2931,6 +2931,7 @@ const struct rtw89_chip_info rtw8922a_chip_info = { .bacam_num = 24, .bacam_dynamic_num = 8, .bacam_ver = RTW89_BACAM_V1, + .addrcam_ver = 0, .ppdu_max_usr = 16, .sec_ctrl_efuse_size = 4, .physical_efuse_size = 0x1300, From 9dab26b9fa457ee538be650ca0bddf352c059e79 Mon Sep 17 00:00:00 2001 From: Chih-Kang Chang Date: Fri, 14 Nov 2025 14:01:27 +0800 Subject: [PATCH 0825/1381] wifi: rtw89: update format of addr cam H2C command The addr cam H2C command is to tell firmware the addr related info. For RTL8922D and RTL8922A after firmware version 0.35.84.0, the addr cam must be updated with update mode to avoid clearing previously set fields. Update it accordingly. Signed-off-by: Chih-Kang Chang Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251114060128.35363-14-pkshih@realtek.com --- drivers/net/wireless/realtek/rtw89/cam.c | 10 ++++--- drivers/net/wireless/realtek/rtw89/cam.h | 12 ++++++--- drivers/net/wireless/realtek/rtw89/core.c | 12 ++++++--- drivers/net/wireless/realtek/rtw89/core.h | 1 + drivers/net/wireless/realtek/rtw89/fw.c | 26 ++++++++++++++++--- drivers/net/wireless/realtek/rtw89/fw.h | 3 ++- drivers/net/wireless/realtek/rtw89/mac.c | 4 +-- drivers/net/wireless/realtek/rtw89/mac80211.c | 4 +-- drivers/net/wireless/realtek/rtw89/wow.c | 6 +++-- 9 files changed, 56 insertions(+), 22 deletions(-) diff --git a/drivers/net/wireless/realtek/rtw89/cam.c b/drivers/net/wireless/realtek/rtw89/cam.c index 5f282a02a356..9370cbda945c 100644 --- a/drivers/net/wireless/realtek/rtw89/cam.c +++ b/drivers/net/wireless/realtek/rtw89/cam.c @@ -236,7 +236,8 @@ static int __rtw89_cam_detach_sec_cam(struct rtw89_dev *rtwdev, if (ret) rtw89_err(rtwdev, "failed to update dctl cam del key: %d\n", ret); - ret = rtw89_fw_h2c_cam(rtwdev, rtwvif_link, rtwsta_link, NULL); + ret = rtw89_fw_h2c_cam(rtwdev, rtwvif_link, rtwsta_link, NULL, + RTW89_ROLE_INFO_CHANGE); if (ret) rtw89_err(rtwdev, "failed to update cam del key: %d\n", ret); } @@ -276,7 +277,8 @@ static int __rtw89_cam_attach_sec_cam(struct rtw89_dev *rtwdev, ret); return ret; } - ret = rtw89_fw_h2c_cam(rtwdev, rtwvif_link, rtwsta_link, NULL); + ret = rtw89_fw_h2c_cam(rtwdev, rtwvif_link, rtwsta_link, NULL, + RTW89_ROLE_INFO_CHANGE); if (ret) { rtw89_err(rtwdev, "failed to update addr cam sec entry: %d\n", ret); @@ -761,7 +763,7 @@ int rtw89_cam_init(struct rtw89_dev *rtwdev, struct rtw89_vif_link *rtwvif_link) int rtw89_cam_fill_bssid_cam_info(struct rtw89_dev *rtwdev, struct rtw89_vif_link *rtwvif_link, struct rtw89_sta_link *rtwsta_link, - struct rtw89_h2c_addr_cam *h2c) + struct rtw89_h2c_addr_cam_v0 *h2c) { struct rtw89_bssid_cam_entry *bssid_cam = rtw89_get_bssid_cam_of(rtwvif_link, rtwsta_link); @@ -813,7 +815,7 @@ void rtw89_cam_fill_addr_cam_info(struct rtw89_dev *rtwdev, struct rtw89_vif_link *rtwvif_link, struct rtw89_sta_link *rtwsta_link, const u8 *scan_mac_addr, - struct rtw89_h2c_addr_cam *h2c) + struct rtw89_h2c_addr_cam_v0 *h2c) { struct ieee80211_vif *vif = rtwvif_link_to_vif(rtwvif_link); struct rtw89_addr_cam_entry *addr_cam = diff --git a/drivers/net/wireless/realtek/rtw89/cam.h b/drivers/net/wireless/realtek/rtw89/cam.h index 833a956f0176..c46b6f91bbdb 100644 --- a/drivers/net/wireless/realtek/rtw89/cam.h +++ b/drivers/net/wireless/realtek/rtw89/cam.h @@ -12,7 +12,7 @@ #define RTW89_BSSID_MATCH_ALL GENMASK(5, 0) #define RTW89_BSSID_MATCH_5_BYTES GENMASK(4, 0) -struct rtw89_h2c_addr_cam { +struct rtw89_h2c_addr_cam_v0 { __le32 w0; __le32 w1; __le32 w2; @@ -30,6 +30,11 @@ struct rtw89_h2c_addr_cam { __le32 w14; } __packed; +struct rtw89_h2c_addr_cam { + struct rtw89_h2c_addr_cam_v0 v0; + __le32 w15; +} __packed; + #define ADDR_CAM_W1_IDX GENMASK(7, 0) #define ADDR_CAM_W1_OFFSET GENMASK(15, 8) #define ADDR_CAM_W1_LEN GENMASK(23, 16) @@ -109,6 +114,7 @@ struct rtw89_h2c_addr_cam { #define ADDR_CAM_W14_BSSID_BSSID3 GENMASK(15, 8) #define ADDR_CAM_W14_BSSID_BSSID4 GENMASK(23, 16) #define ADDR_CAM_W14_BSSID_BSSID5 GENMASK(31, 24) +#define ADDR_CAM_W15_UPD_MODE GENMASK(2, 0) struct rtw89_h2c_dctlinfo_ud_v1 { __le32 c0; @@ -313,7 +319,7 @@ void rtw89_cam_fill_addr_cam_info(struct rtw89_dev *rtwdev, struct rtw89_vif_link *rtwvif_link, struct rtw89_sta_link *rtwsta_link, const u8 *scan_mac_addr, - struct rtw89_h2c_addr_cam *h2c); + struct rtw89_h2c_addr_cam_v0 *h2c); void rtw89_cam_fill_dctl_sec_cam_info_v1(struct rtw89_dev *rtwdev, struct rtw89_vif_link *rtwvif_link, struct rtw89_sta_link *rtwsta_link, @@ -325,7 +331,7 @@ void rtw89_cam_fill_dctl_sec_cam_info_v2(struct rtw89_dev *rtwdev, int rtw89_cam_fill_bssid_cam_info(struct rtw89_dev *rtwdev, struct rtw89_vif_link *rtwvif_link, struct rtw89_sta_link *rtwsta_link, - struct rtw89_h2c_addr_cam *h2c); + struct rtw89_h2c_addr_cam_v0 *h2c); int rtw89_cam_sec_key_add(struct rtw89_dev *rtwdev, struct ieee80211_vif *vif, struct ieee80211_sta *sta, diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c index ea22925b8772..a14edc321ca7 100644 --- a/drivers/net/wireless/realtek/rtw89/core.c +++ b/drivers/net/wireless/realtek/rtw89/core.c @@ -4823,7 +4823,8 @@ int rtw89_core_sta_link_disconnect(struct rtw89_dev *rtwdev, } /* update cam aid mac_id net_type */ - ret = rtw89_fw_h2c_cam(rtwdev, rtwvif_link, rtwsta_link, NULL); + ret = rtw89_fw_h2c_cam(rtwdev, rtwvif_link, rtwsta_link, NULL, + RTW89_ROLE_CON_DISCONN); if (ret) { rtw89_warn(rtwdev, "failed to send h2c cam\n"); return ret; @@ -4897,7 +4898,8 @@ int rtw89_core_sta_link_assoc(struct rtw89_dev *rtwdev, } /* update cam aid mac_id net_type */ - ret = rtw89_fw_h2c_cam(rtwdev, rtwvif_link, rtwsta_link, NULL); + ret = rtw89_fw_h2c_cam(rtwdev, rtwvif_link, rtwsta_link, NULL, + RTW89_ROLE_CON_DISCONN); if (ret) { rtw89_warn(rtwdev, "failed to send h2c cam\n"); return ret; @@ -6029,7 +6031,8 @@ void rtw89_core_scan_start(struct rtw89_dev *rtwdev, struct rtw89_vif_link *rtwv rtw89_phy_config_edcca(rtwdev, bb, true); rtw89_tas_scan(rtwdev, true); - rtw89_fw_h2c_cam(rtwdev, rtwvif_link, NULL, mac_addr); + rtw89_fw_h2c_cam(rtwdev, rtwvif_link, NULL, mac_addr, + RTW89_ROLE_INFO_CHANGE); } void rtw89_core_scan_complete(struct rtw89_dev *rtwdev, @@ -6049,7 +6052,8 @@ void rtw89_core_scan_complete(struct rtw89_dev *rtwdev, rcu_read_unlock(); - rtw89_fw_h2c_cam(rtwdev, rtwvif_link, NULL, NULL); + rtw89_fw_h2c_cam(rtwdev, rtwvif_link, NULL, NULL, + RTW89_ROLE_INFO_CHANGE); rtw89_chip_rfk_scan(rtwdev, rtwvif_link, false); rtw89_btc_ntfy_scan_finish(rtwdev, rtwvif_link->phy_idx); diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h index c33ff7c0090f..ab71489b89a3 100644 --- a/drivers/net/wireless/realtek/rtw89/core.h +++ b/drivers/net/wireless/realtek/rtw89/core.h @@ -4678,6 +4678,7 @@ enum rtw89_fw_feature { RTW89_FW_FEATURE_RFK_NTFY_MCC_V0, RTW89_FW_FEATURE_LPS_DACK_BY_C2H_REG, RTW89_FW_FEATURE_BEACON_TRACKING, + RTW89_FW_FEATURE_ADDR_CAM_V0, }; struct rtw89_fw_suit { diff --git a/drivers/net/wireless/realtek/rtw89/fw.c b/drivers/net/wireless/realtek/rtw89/fw.c index 9f132a469629..18eab404885c 100644 --- a/drivers/net/wireless/realtek/rtw89/fw.c +++ b/drivers/net/wireless/realtek/rtw89/fw.c @@ -881,6 +881,7 @@ static const struct __fw_feat_cfg fw_feat_tbl[] = { __CFG_FW_FEAT(RTL8922A, ge, 0, 35, 76, 0, LPS_DACK_BY_C2H_REG), __CFG_FW_FEAT(RTL8922A, ge, 0, 35, 79, 0, CRASH_TRIGGER_TYPE_1), __CFG_FW_FEAT(RTL8922A, ge, 0, 35, 80, 0, BEACON_TRACKING), + __CFG_FW_FEAT(RTL8922A, lt, 0, 35, 84, 0, ADDR_CAM_V0), }; static void rtw89_fw_iterate_feature_cfg(struct rtw89_fw_info *fw, @@ -2133,25 +2134,42 @@ void rtw89_fw_log_dump(struct rtw89_dev *rtwdev, u8 *buf, u32 len) } int rtw89_fw_h2c_cam(struct rtw89_dev *rtwdev, struct rtw89_vif_link *rtwvif_link, - struct rtw89_sta_link *rtwsta_link, const u8 *scan_mac_addr) + struct rtw89_sta_link *rtwsta_link, const u8 *scan_mac_addr, + enum rtw89_upd_mode upd_mode) { + const struct rtw89_chip_info *chip = rtwdev->chip; + struct rtw89_h2c_addr_cam_v0 *h2c_v0; struct rtw89_h2c_addr_cam *h2c; u32 len = sizeof(*h2c); struct sk_buff *skb; + u8 ver = U8_MAX; int ret; + if (RTW89_CHK_FW_FEATURE(ADDR_CAM_V0, &rtwdev->fw) || + chip->chip_gen == RTW89_CHIP_AX) { + len = sizeof(*h2c_v0); + ver = 0; + } + skb = rtw89_fw_h2c_alloc_skb_with_hdr(rtwdev, len); if (!skb) { rtw89_err(rtwdev, "failed to alloc skb for fw dl\n"); return -ENOMEM; } skb_put(skb, len); - h2c = (struct rtw89_h2c_addr_cam *)skb->data; + h2c_v0 = (struct rtw89_h2c_addr_cam_v0 *)skb->data; rtw89_cam_fill_addr_cam_info(rtwdev, rtwvif_link, rtwsta_link, - scan_mac_addr, h2c); - rtw89_cam_fill_bssid_cam_info(rtwdev, rtwvif_link, rtwsta_link, h2c); + scan_mac_addr, h2c_v0); + rtw89_cam_fill_bssid_cam_info(rtwdev, rtwvif_link, rtwsta_link, h2c_v0); + if (ver == 0) + goto hdr; + + h2c = (struct rtw89_h2c_addr_cam *)skb->data; + h2c->w15 = le32_encode_bits(upd_mode, ADDR_CAM_W15_UPD_MODE); + +hdr: rtw89_h2c_pkt_set_hdr(rtwdev, skb, FWCMD_TYPE_H2C, H2C_CAT_MAC, H2C_CL_MAC_ADDR_CAM_UPDATE, diff --git a/drivers/net/wireless/realtek/rtw89/fw.h b/drivers/net/wireless/realtek/rtw89/fw.h index e62b61f584fb..cedb4a47a769 100644 --- a/drivers/net/wireless/realtek/rtw89/fw.h +++ b/drivers/net/wireless/realtek/rtw89/fw.h @@ -4887,7 +4887,8 @@ int rtw89_fw_h2c_tbtt_tuning(struct rtw89_dev *rtwdev, struct rtw89_vif_link *rtwvif_link, u32 offset); int rtw89_fw_h2c_pwr_lvl(struct rtw89_dev *rtwdev, struct rtw89_vif_link *rtwvif_link); int rtw89_fw_h2c_cam(struct rtw89_dev *rtwdev, struct rtw89_vif_link *vif, - struct rtw89_sta_link *rtwsta_link, const u8 *scan_mac_addr); + struct rtw89_sta_link *rtwsta_link, const u8 *scan_mac_addr, + enum rtw89_upd_mode upd_mode); int rtw89_fw_h2c_dctl_sec_cam_v1(struct rtw89_dev *rtwdev, struct rtw89_vif_link *rtwvif_link, struct rtw89_sta_link *rtwsta_link); diff --git a/drivers/net/wireless/realtek/rtw89/mac.c b/drivers/net/wireless/realtek/rtw89/mac.c index 386cc982d7c9..bde3791dd330 100644 --- a/drivers/net/wireless/realtek/rtw89/mac.c +++ b/drivers/net/wireless/realtek/rtw89/mac.c @@ -4835,7 +4835,7 @@ int rtw89_mac_vif_init(struct rtw89_dev *rtwdev, struct rtw89_vif_link *rtwvif_l if (ret) return ret; - ret = rtw89_fw_h2c_cam(rtwdev, rtwvif_link, NULL, NULL); + ret = rtw89_fw_h2c_cam(rtwdev, rtwvif_link, NULL, NULL, RTW89_ROLE_CREATE); if (ret) return ret; @@ -4860,7 +4860,7 @@ int rtw89_mac_vif_deinit(struct rtw89_dev *rtwdev, struct rtw89_vif_link *rtwvif rtw89_cam_deinit(rtwdev, rtwvif_link); - ret = rtw89_fw_h2c_cam(rtwdev, rtwvif_link, NULL, NULL); + ret = rtw89_fw_h2c_cam(rtwdev, rtwvif_link, NULL, NULL, RTW89_ROLE_REMOVE); if (ret) return ret; diff --git a/drivers/net/wireless/realtek/rtw89/mac80211.c b/drivers/net/wireless/realtek/rtw89/mac80211.c index 4a7364432428..82598673ddec 100644 --- a/drivers/net/wireless/realtek/rtw89/mac80211.c +++ b/drivers/net/wireless/realtek/rtw89/mac80211.c @@ -759,7 +759,7 @@ static void rtw89_ops_link_info_changed(struct ieee80211_hw *hw, if (changed & BSS_CHANGED_BSSID) { ether_addr_copy(rtwvif_link->bssid, conf->bssid); rtw89_cam_bssid_changed(rtwdev, rtwvif_link); - rtw89_fw_h2c_cam(rtwdev, rtwvif_link, NULL, NULL); + rtw89_fw_h2c_cam(rtwdev, rtwvif_link, NULL, NULL, RTW89_ROLE_INFO_CHANGE); WRITE_ONCE(rtwvif_link->sync_bcn_tsf, 0); } @@ -818,7 +818,7 @@ static int rtw89_ops_start_ap(struct ieee80211_hw *hw, rtw89_chip_h2c_assoc_cmac_tbl(rtwdev, rtwvif_link, NULL); rtw89_fw_h2c_role_maintain(rtwdev, rtwvif_link, NULL, RTW89_ROLE_TYPE_CHANGE); rtw89_fw_h2c_join_info(rtwdev, rtwvif_link, NULL, true); - rtw89_fw_h2c_cam(rtwdev, rtwvif_link, NULL, NULL); + rtw89_fw_h2c_cam(rtwdev, rtwvif_link, NULL, NULL, RTW89_ROLE_TYPE_CHANGE); rtw89_chip_rfk_channel(rtwdev, rtwvif_link); if (RTW89_CHK_FW_FEATURE(NOTIFY_AP_INFO, &rtwdev->fw)) { diff --git a/drivers/net/wireless/realtek/rtw89/wow.c b/drivers/net/wireless/realtek/rtw89/wow.c index b9fce0be3ab7..46aba4cb2ee9 100644 --- a/drivers/net/wireless/realtek/rtw89/wow.c +++ b/drivers/net/wireless/realtek/rtw89/wow.c @@ -1221,7 +1221,8 @@ static int rtw89_wow_cfg_wake(struct rtw89_dev *rtwdev, bool wow) } } - ret = rtw89_fw_h2c_cam(rtwdev, rtwvif_link, rtwsta_link, NULL); + ret = rtw89_fw_h2c_cam(rtwdev, rtwvif_link, rtwsta_link, NULL, + RTW89_ROLE_INFO_CHANGE); if (ret) { rtw89_warn(rtwdev, "failed to send h2c cam\n"); return ret; @@ -1318,7 +1319,8 @@ static int rtw89_wow_swap_fw(struct rtw89_dev *rtwdev, bool wow) return ret; } - ret = rtw89_fw_h2c_cam(rtwdev, rtwvif_link, rtwsta_link, NULL); + ret = rtw89_fw_h2c_cam(rtwdev, rtwvif_link, rtwsta_link, NULL, + RTW89_ROLE_FW_RESTORE); if (ret) { rtw89_warn(rtwdev, "failed to send h2c cam\n"); return ret; From 09afd209a84421948ac9bf2076459a0418e102ec Mon Sep 17 00:00:00 2001 From: Chih-Kang Chang Date: Fri, 14 Nov 2025 14:01:28 +0800 Subject: [PATCH 0826/1381] wifi: rtw89: correct user macid mask of RX info for RTL8922D The user MAC ID mask of RX info in MAC PPDU for RTL8922A and RTL8922D is different, correct it accordingly. Signed-off-by: Chih-Kang Chang Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251114060128.35363-15-pkshih@realtek.com --- drivers/net/wireless/realtek/rtw89/core.c | 6 +++++- drivers/net/wireless/realtek/rtw89/core.h | 1 + drivers/net/wireless/realtek/rtw89/txrx.h | 1 + 3 files changed, 7 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c index a14edc321ca7..8078c9aa0ac8 100644 --- a/drivers/net/wireless/realtek/rtw89/core.c +++ b/drivers/net/wireless/realtek/rtw89/core.c @@ -1834,9 +1834,13 @@ static int rtw89_core_rx_process_mac_ppdu(struct rtw89_dev *rtwdev, /* For WiFi 7 chips, RXWD.mac_id of PPDU status is not set * by hardware, so update mac_id by rxinfo_user[].mac_id. */ - if (chip_gen == RTW89_CHIP_BE) + if (chip->chip_id == RTL8922A) phy_ppdu->mac_id = le32_get_bits(user->w0, RTW89_RXINFO_USER_MACID); + else if (chip->chip_id == RTL8922D) + phy_ppdu->mac_id = + le32_get_bits(user->w0, RTW89_RXINFO_USER_MACID_V1); + phy_ppdu->has_data = le32_get_bits(user->w0, RTW89_RXINFO_USER_DATA); phy_ppdu->has_bcn = diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h index ab71489b89a3..a9cb47ea0b93 100644 --- a/drivers/net/wireless/realtek/rtw89/core.h +++ b/drivers/net/wireless/realtek/rtw89/core.h @@ -155,6 +155,7 @@ enum rtw89_core_chip_id { RTL8852C, RTL8851B, RTL8922A, + RTL8922D, }; enum rtw89_chip_gen { diff --git a/drivers/net/wireless/realtek/rtw89/txrx.h b/drivers/net/wireless/realtek/rtw89/txrx.h index b37dbac7b790..fa324b4a1dde 100644 --- a/drivers/net/wireless/realtek/rtw89/txrx.h +++ b/drivers/net/wireless/realtek/rtw89/txrx.h @@ -421,6 +421,7 @@ struct rtw89_rxinfo_user { #define RTW89_RXINFO_USER_MGMT BIT(3) #define RTW89_RXINFO_USER_BCN BIT(4) #define RTW89_RXINFO_USER_MACID GENMASK(15, 8) +#define RTW89_RXINFO_USER_MACID_V1 GENMASK(31, 20) struct rtw89_rxinfo { __le32 w0; From c8d212fa81c5e0cad15655918b5fb8fcf6e6fe57 Mon Sep 17 00:00:00 2001 From: Ping-Ke Shih Date: Thu, 13 Nov 2025 13:34:59 +0800 Subject: [PATCH 0827/1381] wifi: rtw89: use separated function to set RX filter Normally use rtw89_write32_mask() with a consecutive bit mask, but mask of RX filter B_AX_RX_FLTR_CFG_MASK is bits 31-22 and 15-0, which excludes bits of B_AX_RX_MPDU_MAX_LEN_MASK (bits 21-16). Though the original logic is well to set RX filter, change it to a separate function to avoid the tricky design. Cc: Geert Uytterhoeven Signed-off-by: Ping-Ke Shih Reviewed-by: Geert Uytterhoeven Link: https://patch.msgid.link/20251113053459.34995-1-pkshih@realtek.com --- drivers/net/wireless/realtek/rtw89/core.c | 5 +---- drivers/net/wireless/realtek/rtw89/fw.c | 10 ++-------- drivers/net/wireless/realtek/rtw89/mac.c | 14 ++++++++++++++ drivers/net/wireless/realtek/rtw89/mac.h | 1 + drivers/net/wireless/realtek/rtw89/mac80211.c | 11 ++--------- 5 files changed, 20 insertions(+), 21 deletions(-) diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c index 8078c9aa0ac8..0824940c91ae 100644 --- a/drivers/net/wireless/realtek/rtw89/core.c +++ b/drivers/net/wireless/realtek/rtw89/core.c @@ -4173,12 +4173,10 @@ void rtw89_roc_start(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwvif) void rtw89_roc_end(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwvif) { - const struct rtw89_mac_gen_def *mac = rtwdev->chip->mac_def; struct ieee80211_hw *hw = rtwdev->hw; struct rtw89_roc *roc = &rtwvif->roc; struct rtw89_vif_link *rtwvif_link; struct rtw89_vif *tmp_vif; - u32 reg; int ret; lockdep_assert_wiphy(hw->wiphy); @@ -4195,8 +4193,7 @@ void rtw89_roc_end(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwvif) return; } - reg = rtw89_mac_reg_by_idx(rtwdev, mac->rx_fltr, rtwvif_link->mac_idx); - rtw89_write32_mask(rtwdev, reg, B_AX_RX_FLTR_CFG_MASK, rtwdev->hal.rx_fltr); + rtw89_mac_set_rx_fltr(rtwdev, rtwvif_link->mac_idx, rtwdev->hal.rx_fltr); roc->state = RTW89_ROC_IDLE; rtw89_config_roc_chandef(rtwdev, rtwvif_link, NULL); diff --git a/drivers/net/wireless/realtek/rtw89/fw.c b/drivers/net/wireless/realtek/rtw89/fw.c index 18eab404885c..d50c35328363 100644 --- a/drivers/net/wireless/realtek/rtw89/fw.c +++ b/drivers/net/wireless/realtek/rtw89/fw.c @@ -8103,7 +8103,6 @@ int rtw89_hw_scan_start(struct rtw89_dev *rtwdev, struct rtw89_vif_link *rtwvif_link, struct ieee80211_scan_request *scan_req) { - const struct rtw89_mac_gen_def *mac = rtwdev->chip->mac_def; enum rtw89_entity_mode mode = rtw89_get_entity_mode(rtwdev); struct cfg80211_scan_request *req = &scan_req->req; const struct rtw89_chan *chan = rtw89_chan_get(rtwdev, @@ -8115,7 +8114,6 @@ int rtw89_hw_scan_start(struct rtw89_dev *rtwdev, }; u32 rx_fltr = rtwdev->hal.rx_fltr; u8 mac_addr[ETH_ALEN]; - u32 reg; int ret; /* clone op and keep it during scan */ @@ -8155,8 +8153,7 @@ int rtw89_hw_scan_start(struct rtw89_dev *rtwdev, rx_fltr &= ~B_AX_A_BC; rx_fltr &= ~B_AX_A_A1_MATCH; - reg = rtw89_mac_reg_by_idx(rtwdev, mac->rx_fltr, rtwvif_link->mac_idx); - rtw89_write32_mask(rtwdev, reg, B_AX_RX_FLTR_CFG_MASK, rx_fltr); + rtw89_mac_set_rx_fltr(rtwdev, rtwvif_link->mac_idx, rx_fltr); rtw89_chanctx_pause(rtwdev, &pause_parm); rtw89_phy_dig_suspend(rtwdev); @@ -8174,20 +8171,17 @@ struct rtw89_hw_scan_complete_cb_data { static int rtw89_hw_scan_complete_cb(struct rtw89_dev *rtwdev, void *data) { - const struct rtw89_mac_gen_def *mac = rtwdev->chip->mac_def; enum rtw89_entity_mode mode = rtw89_get_entity_mode(rtwdev); struct rtw89_hw_scan_complete_cb_data *cb_data = data; struct rtw89_vif_link *rtwvif_link = cb_data->rtwvif_link; struct cfg80211_scan_info info = { .aborted = cb_data->aborted, }; - u32 reg; if (!rtwvif_link) return -EINVAL; - reg = rtw89_mac_reg_by_idx(rtwdev, mac->rx_fltr, rtwvif_link->mac_idx); - rtw89_write32_mask(rtwdev, reg, B_AX_RX_FLTR_CFG_MASK, rtwdev->hal.rx_fltr); + rtw89_mac_set_rx_fltr(rtwdev, rtwvif_link->mac_idx, rtwdev->hal.rx_fltr); rtw89_core_scan_complete(rtwdev, rtwvif_link, true); ieee80211_scan_completed(rtwdev->hw, &info); diff --git a/drivers/net/wireless/realtek/rtw89/mac.c b/drivers/net/wireless/realtek/rtw89/mac.c index bde3791dd330..d78fbe73e365 100644 --- a/drivers/net/wireless/realtek/rtw89/mac.c +++ b/drivers/net/wireless/realtek/rtw89/mac.c @@ -2543,6 +2543,20 @@ static int rtw89_mac_typ_fltr_opt_ax(struct rtw89_dev *rtwdev, return 0; } +void rtw89_mac_set_rx_fltr(struct rtw89_dev *rtwdev, u8 mac_idx, u32 rx_fltr) +{ + const struct rtw89_mac_gen_def *mac = rtwdev->chip->mac_def; + u32 reg; + u32 val; + + reg = rtw89_mac_reg_by_idx(rtwdev, mac->rx_fltr, mac_idx); + + val = rtw89_read32(rtwdev, reg); + /* B_AX_RX_FLTR_CFG_MASK is not a consecutive bit mask */ + val = (val & ~B_AX_RX_FLTR_CFG_MASK) | (rx_fltr & B_AX_RX_FLTR_CFG_MASK); + rtw89_write32(rtwdev, reg, val); +} + static int rx_fltr_init_ax(struct rtw89_dev *rtwdev, u8 mac_idx) { int ret, i; diff --git a/drivers/net/wireless/realtek/rtw89/mac.h b/drivers/net/wireless/realtek/rtw89/mac.h index d203db444a34..0007229d6753 100644 --- a/drivers/net/wireless/realtek/rtw89/mac.h +++ b/drivers/net/wireless/realtek/rtw89/mac.h @@ -1340,6 +1340,7 @@ int rtw89_mac_cfg_ppdu_status_bands(struct rtw89_dev *rtwdev, bool enable) return rtw89_mac_cfg_ppdu_status(rtwdev, RTW89_MAC_1, enable); } +void rtw89_mac_set_rx_fltr(struct rtw89_dev *rtwdev, u8 mac_idx, u32 rx_fltr); void rtw89_mac_update_rts_threshold(struct rtw89_dev *rtwdev); void rtw89_mac_flush_txq(struct rtw89_dev *rtwdev, u32 queues, bool drop); int rtw89_mac_coex_init(struct rtw89_dev *rtwdev, const struct rtw89_mac_ax_coex *coex); diff --git a/drivers/net/wireless/realtek/rtw89/mac80211.c b/drivers/net/wireless/realtek/rtw89/mac80211.c index 82598673ddec..f39ca1c2ed10 100644 --- a/drivers/net/wireless/realtek/rtw89/mac80211.c +++ b/drivers/net/wireless/realtek/rtw89/mac80211.c @@ -307,7 +307,6 @@ static void rtw89_ops_configure_filter(struct ieee80211_hw *hw, u64 multicast) { struct rtw89_dev *rtwdev = hw->priv; - const struct rtw89_mac_gen_def *mac = rtwdev->chip->mac_def; u32 rx_fltr; lockdep_assert_wiphy(hw->wiphy); @@ -369,16 +368,10 @@ static void rtw89_ops_configure_filter(struct ieee80211_hw *hw, rx_fltr &= ~B_AX_A_A1_MATCH; } - rtw89_write32_mask(rtwdev, - rtw89_mac_reg_by_idx(rtwdev, mac->rx_fltr, RTW89_MAC_0), - B_AX_RX_FLTR_CFG_MASK, - rx_fltr); + rtw89_mac_set_rx_fltr(rtwdev, RTW89_MAC_0, rx_fltr); if (!rtwdev->dbcc_en) return; - rtw89_write32_mask(rtwdev, - rtw89_mac_reg_by_idx(rtwdev, mac->rx_fltr, RTW89_MAC_1), - B_AX_RX_FLTR_CFG_MASK, - rx_fltr); + rtw89_mac_set_rx_fltr(rtwdev, RTW89_MAC_1, rx_fltr); } static const u8 ac_to_fw_idx[IEEE80211_NUM_ACS] = { From 9b5b9c042b30befc5b37e4539ace95af70843473 Mon Sep 17 00:00:00 2001 From: Abdun Nihaal Date: Fri, 14 Nov 2025 15:15:26 +0530 Subject: [PATCH 0828/1381] wifi: rtl818x: Fix potential memory leaks in rtl8180_init_rx_ring() In rtl8180_init_rx_ring(), memory is allocated for skb packets and DMA allocations in a loop. When an allocation fails, the previously successful allocations are not freed on exit. Fix that by jumping to err_free_rings label on error, which calls rtl8180_free_rx_ring() to free the allocations. Remove the free of rx_ring in rtl8180_init_rx_ring() error path, and set the freed priv->rx_buf entry to null, to avoid double free. Fixes: f653211197f3 ("Add rtl8180 wireless driver") Signed-off-by: Abdun Nihaal Reviewed-by: Ping-Ke Shih Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251114094527.79842-1-nihaal@cse.iitm.ac.in --- drivers/net/wireless/realtek/rtl818x/rtl8180/dev.c | 9 ++------- 1 file changed, 2 insertions(+), 7 deletions(-) diff --git a/drivers/net/wireless/realtek/rtl818x/rtl8180/dev.c b/drivers/net/wireless/realtek/rtl818x/rtl8180/dev.c index 2905baea6239..070c0431c482 100644 --- a/drivers/net/wireless/realtek/rtl818x/rtl8180/dev.c +++ b/drivers/net/wireless/realtek/rtl818x/rtl8180/dev.c @@ -1023,9 +1023,6 @@ static int rtl8180_init_rx_ring(struct ieee80211_hw *dev) dma_addr_t *mapping; entry = priv->rx_ring + priv->rx_ring_sz*i; if (!skb) { - dma_free_coherent(&priv->pdev->dev, - priv->rx_ring_sz * 32, - priv->rx_ring, priv->rx_ring_dma); wiphy_err(dev->wiphy, "Cannot allocate RX skb\n"); return -ENOMEM; } @@ -1037,9 +1034,7 @@ static int rtl8180_init_rx_ring(struct ieee80211_hw *dev) if (dma_mapping_error(&priv->pdev->dev, *mapping)) { kfree_skb(skb); - dma_free_coherent(&priv->pdev->dev, - priv->rx_ring_sz * 32, - priv->rx_ring, priv->rx_ring_dma); + priv->rx_buf[i] = NULL; wiphy_err(dev->wiphy, "Cannot map DMA for RX skb\n"); return -ENOMEM; } @@ -1130,7 +1125,7 @@ static int rtl8180_start(struct ieee80211_hw *dev) ret = rtl8180_init_rx_ring(dev); if (ret) - return ret; + goto err_free_rings; for (i = 0; i < (dev->queues + 1); i++) if ((ret = rtl8180_init_tx_ring(dev, i, 16))) From c453d9c5bb8aa1510078a53eef37ea437c21ab07 Mon Sep 17 00:00:00 2001 From: Zenm Chen Date: Mon, 17 Nov 2025 07:44:03 +0800 Subject: [PATCH 0829/1381] wifi: rtw89: Add default ID 0bda:b831 for RTL8831BU Add 0bda:b831 for RTL8831BU-based adapters that use this default ID. Tested with TOTOLINK X900USM in station mode very briefly. Signed-off-by: Zenm Chen Acked-by: Ping-Ke Shih Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251116234403.8803-1-zenmchen@gmail.com --- drivers/net/wireless/realtek/rtw89/rtw8851bu.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/wireless/realtek/rtw89/rtw8851bu.c b/drivers/net/wireless/realtek/rtw89/rtw8851bu.c index b8799a37add3..959d62aefdd8 100644 --- a/drivers/net/wireless/realtek/rtw89/rtw8851bu.c +++ b/drivers/net/wireless/realtek/rtw89/rtw8851bu.c @@ -36,6 +36,8 @@ static const struct rtw89_driver_info rtw89_8851bu_info = { }; static const struct usb_device_id rtw_8851bu_id_table[] = { + { USB_DEVICE_AND_INTERFACE_INFO(0x0bda, 0xb831, 0xff, 0xff, 0xff), + .driver_info = (kernel_ulong_t)&rtw89_8851bu_info }, { USB_DEVICE_AND_INTERFACE_INFO(0x0bda, 0xb851, 0xff, 0xff, 0xff), .driver_info = (kernel_ulong_t)&rtw89_8851bu_info }, /* D-Link AX9U rev. A1 */ From d3305c016a6a978688ca40422a539d0c46c59366 Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" Date: Fri, 14 Nov 2025 19:12:05 +0100 Subject: [PATCH 0830/1381] selftests: mptcp: lib: introduce 'nstat_{init,get}' These new helpers are easier to read than the long and multi lines commands. Plus it will ease the addition of new features related to that in the next commits. No behavioural changes intended. Acked-by: Paolo Abeni Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20251114-net-next-mptcp-sft-count-cache-stats-timeout-v1-1-863cb04e1b7b@kernel.org Signed-off-by: Jakub Kicinski --- tools/testing/selftests/net/mptcp/mptcp_connect.sh | 12 ++++-------- tools/testing/selftests/net/mptcp/mptcp_join.sh | 12 ++++-------- tools/testing/selftests/net/mptcp/mptcp_lib.sh | 14 ++++++++++++++ tools/testing/selftests/net/mptcp/mptcp_sockopt.sh | 12 ++++-------- tools/testing/selftests/net/mptcp/simult_flows.sh | 12 ++++-------- 5 files changed, 30 insertions(+), 32 deletions(-) diff --git a/tools/testing/selftests/net/mptcp/mptcp_connect.sh b/tools/testing/selftests/net/mptcp/mptcp_connect.sh index 9b7b93f8eb0c..e1f34508490a 100755 --- a/tools/testing/selftests/net/mptcp/mptcp_connect.sh +++ b/tools/testing/selftests/net/mptcp/mptcp_connect.sh @@ -384,11 +384,9 @@ do_transfer() sleep 1 fi - NSTAT_HISTORY=/tmp/${listener_ns}.nstat ip netns exec ${listener_ns} \ - nstat -n + mptcp_lib_nstat_init "${listener_ns}" if [ ${listener_ns} != ${connector_ns} ]; then - NSTAT_HISTORY=/tmp/${connector_ns}.nstat ip netns exec ${connector_ns} \ - nstat -n + mptcp_lib_nstat_init "${connector_ns}" fi local stat_synrx_last_l @@ -436,11 +434,9 @@ do_transfer() kill ${cappid_connector} fi - NSTAT_HISTORY=/tmp/${listener_ns}.nstat ip netns exec ${listener_ns} \ - nstat | grep Tcp > /tmp/${listener_ns}.out + mptcp_lib_nstat_get "${listener_ns}" if [ ${listener_ns} != ${connector_ns} ]; then - NSTAT_HISTORY=/tmp/${connector_ns}.nstat ip netns exec ${connector_ns} \ - nstat | grep Tcp > /tmp/${connector_ns}.out + mptcp_lib_nstat_get "${connector_ns}" fi local duration diff --git a/tools/testing/selftests/net/mptcp/mptcp_join.sh b/tools/testing/selftests/net/mptcp/mptcp_join.sh index f0efbf985625..f51c9bf2d37f 100755 --- a/tools/testing/selftests/net/mptcp/mptcp_join.sh +++ b/tools/testing/selftests/net/mptcp/mptcp_join.sh @@ -983,10 +983,8 @@ do_transfer() cond_start_capture ${listener_ns} - NSTAT_HISTORY=/tmp/${listener_ns}.nstat ip netns exec ${listener_ns} \ - nstat -n - NSTAT_HISTORY=/tmp/${connector_ns}.nstat ip netns exec ${connector_ns} \ - nstat -n + mptcp_lib_nstat_init "${listener_ns}" + mptcp_lib_nstat_init "${connector_ns}" local extra_args if [ $speed = "fast" ]; then @@ -1068,10 +1066,8 @@ do_transfer() cond_stop_capture - NSTAT_HISTORY=/tmp/${listener_ns}.nstat ip netns exec ${listener_ns} \ - nstat | grep Tcp > /tmp/${listener_ns}.out - NSTAT_HISTORY=/tmp/${connector_ns}.nstat ip netns exec ${connector_ns} \ - nstat | grep Tcp > /tmp/${connector_ns}.out + mptcp_lib_nstat_get "${listener_ns}" + mptcp_lib_nstat_get "${connector_ns}" if [ ${rets} -ne 0 ] || [ ${retc} -ne 0 ]; then fail_test "client exit code $retc, server $rets" diff --git a/tools/testing/selftests/net/mptcp/mptcp_lib.sh b/tools/testing/selftests/net/mptcp/mptcp_lib.sh index f4388900016a..1f31746cc519 100644 --- a/tools/testing/selftests/net/mptcp/mptcp_lib.sh +++ b/tools/testing/selftests/net/mptcp/mptcp_lib.sh @@ -376,6 +376,20 @@ mptcp_lib_is_v6() { [ -z "${1##*:*}" ] } +mptcp_lib_nstat_init() { + local ns="${1}" + + NSTAT_HISTORY="/tmp/${ns}.nstat" ip netns exec "${ns}" nstat -n +} + +mptcp_lib_nstat_get() { + local ns="${1}" + + # filter out non-*TCP stats + NSTAT_HISTORY="/tmp/${ns}.nstat" ip netns exec "${ns}" nstat | + grep Tcp > "/tmp/${ns}.out" +} + # $1: ns, $2: MIB counter mptcp_lib_get_counter() { local ns="${1}" diff --git a/tools/testing/selftests/net/mptcp/mptcp_sockopt.sh b/tools/testing/selftests/net/mptcp/mptcp_sockopt.sh index f01989be6e9b..343615908558 100755 --- a/tools/testing/selftests/net/mptcp/mptcp_sockopt.sh +++ b/tools/testing/selftests/net/mptcp/mptcp_sockopt.sh @@ -169,10 +169,8 @@ do_transfer() cmsg+=",TCPINQ" fi - NSTAT_HISTORY=/tmp/${listener_ns}.nstat ip netns exec ${listener_ns} \ - nstat -n - NSTAT_HISTORY=/tmp/${connector_ns}.nstat ip netns exec ${connector_ns} \ - nstat -n + mptcp_lib_nstat_init "${listener_ns}" + mptcp_lib_nstat_init "${connector_ns}" timeout ${timeout_test} \ ip netns exec ${listener_ns} \ @@ -194,10 +192,8 @@ do_transfer() wait $spid local rets=$? - NSTAT_HISTORY=/tmp/${listener_ns}.nstat ip netns exec ${listener_ns} \ - nstat | grep Tcp > /tmp/${listener_ns}.out - NSTAT_HISTORY=/tmp/${connector_ns}.nstat ip netns exec ${connector_ns} \ - nstat | grep Tcp > /tmp/${connector_ns}.out + mptcp_lib_nstat_get "${listener_ns}" + mptcp_lib_nstat_get "${connector_ns}" print_title "Transfer ${ip:2}" if [ ${rets} -ne 0 ] || [ ${retc} -ne 0 ]; then diff --git a/tools/testing/selftests/net/mptcp/simult_flows.sh b/tools/testing/selftests/net/mptcp/simult_flows.sh index 1903e8e84a31..fbd1685d1adc 100755 --- a/tools/testing/selftests/net/mptcp/simult_flows.sh +++ b/tools/testing/selftests/net/mptcp/simult_flows.sh @@ -155,10 +155,8 @@ do_transfer() sleep 1 fi - NSTAT_HISTORY=/tmp/${ns3}.nstat ip netns exec ${ns3} \ - nstat -n - NSTAT_HISTORY=/tmp/${ns1}.nstat ip netns exec ${ns1} \ - nstat -n + mptcp_lib_nstat_init "${ns3}" + mptcp_lib_nstat_init "${ns1}" timeout ${timeout_test} \ ip netns exec ${ns3} \ @@ -185,10 +183,8 @@ do_transfer() kill ${cappid_connector} fi - NSTAT_HISTORY=/tmp/${ns3}.nstat ip netns exec ${ns3} \ - nstat | grep Tcp > /tmp/${ns3}.out - NSTAT_HISTORY=/tmp/${ns1}.nstat ip netns exec ${ns1} \ - nstat | grep Tcp > /tmp/${ns1}.out + mptcp_lib_nstat_get "${ns3}" + mptcp_lib_nstat_get "${ns1}" cmp $sin $cout > /dev/null 2>&1 local cmps=$? From a89fc262b6260c1ef69439a9fc233d84e58357d3 Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" Date: Fri, 14 Nov 2025 19:12:06 +0100 Subject: [PATCH 0831/1381] selftests: mptcp: lib: remove stats files args Now that these files are written from MPTCP lib helpers, the stats file paths are uniformed. Then, no need to specify them from the each selftest. No behavioural changes intended. Acked-by: Paolo Abeni Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20251114-net-next-mptcp-sft-count-cache-stats-timeout-v1-2-863cb04e1b7b@kernel.org Signed-off-by: Jakub Kicinski --- .../testing/selftests/net/mptcp/mptcp_connect.sh | 3 +-- tools/testing/selftests/net/mptcp/mptcp_join.sh | 3 +-- tools/testing/selftests/net/mptcp/mptcp_lib.sh | 15 ++++++++++----- .../testing/selftests/net/mptcp/mptcp_sockopt.sh | 3 +-- tools/testing/selftests/net/mptcp/simult_flows.sh | 3 +-- 5 files changed, 14 insertions(+), 13 deletions(-) diff --git a/tools/testing/selftests/net/mptcp/mptcp_connect.sh b/tools/testing/selftests/net/mptcp/mptcp_connect.sh index e1f34508490a..4ff2d02429d3 100755 --- a/tools/testing/selftests/net/mptcp/mptcp_connect.sh +++ b/tools/testing/selftests/net/mptcp/mptcp_connect.sh @@ -444,8 +444,7 @@ do_transfer() printf "(duration %05sms) " "${duration}" if [ ${rets} -ne 0 ] || [ ${retc} -ne 0 ]; then mptcp_lib_pr_fail "client exit code $retc, server $rets" - mptcp_lib_pr_err_stats "${listener_ns}" "${connector_ns}" "${port}" \ - "/tmp/${listener_ns}.out" "/tmp/${connector_ns}.out" + mptcp_lib_pr_err_stats "${listener_ns}" "${connector_ns}" "${port}" echo cat "$capout" diff --git a/tools/testing/selftests/net/mptcp/mptcp_join.sh b/tools/testing/selftests/net/mptcp/mptcp_join.sh index f51c9bf2d37f..aee215d73b7c 100755 --- a/tools/testing/selftests/net/mptcp/mptcp_join.sh +++ b/tools/testing/selftests/net/mptcp/mptcp_join.sh @@ -1071,8 +1071,7 @@ do_transfer() if [ ${rets} -ne 0 ] || [ ${retc} -ne 0 ]; then fail_test "client exit code $retc, server $rets" - mptcp_lib_pr_err_stats "${listener_ns}" "${connector_ns}" "${port}" \ - "/tmp/${listener_ns}.out" "/tmp/${connector_ns}.out" + mptcp_lib_pr_err_stats "${listener_ns}" "${connector_ns}" "${port}" return 1 fi diff --git a/tools/testing/selftests/net/mptcp/mptcp_lib.sh b/tools/testing/selftests/net/mptcp/mptcp_lib.sh index 1f31746cc519..4a26d4150603 100644 --- a/tools/testing/selftests/net/mptcp/mptcp_lib.sh +++ b/tools/testing/selftests/net/mptcp/mptcp_lib.sh @@ -106,23 +106,28 @@ mptcp_lib_pr_info() { mptcp_lib_print_info "INFO: ${*}" } -# $1-2: listener/connector ns ; $3 port ; $4-5 listener/connector stat file +mptcp_lib_pr_nstat() { + local ns="${1}" + local hist="/tmp/${ns}.out" + + cat "${hist}" +} + +# $1-2: listener/connector ns ; $3 port mptcp_lib_pr_err_stats() { local lns="${1}" local cns="${2}" local port="${3}" - local lstat="${4}" - local cstat="${5}" echo -en "${MPTCP_LIB_COLOR_RED}" { printf "\nnetns %s (listener) socket stat for %d:\n" "${lns}" "${port}" ip netns exec "${lns}" ss -Menitam -o "sport = :${port}" - cat "${lstat}" + mptcp_lib_pr_nstat "${lns}" printf "\nnetns %s (connector) socket stat for %d:\n" "${cns}" "${port}" ip netns exec "${cns}" ss -Menitam -o "dport = :${port}" - [ "${lstat}" != "${cstat}" ] && cat "${cstat}" + [ "${lns}" != "${cns}" ] && mptcp_lib_pr_nstat "${cns}" } 1>&2 echo -en "${MPTCP_LIB_COLOR_RESET}" } diff --git a/tools/testing/selftests/net/mptcp/mptcp_sockopt.sh b/tools/testing/selftests/net/mptcp/mptcp_sockopt.sh index 343615908558..42d533b95ec7 100755 --- a/tools/testing/selftests/net/mptcp/mptcp_sockopt.sh +++ b/tools/testing/selftests/net/mptcp/mptcp_sockopt.sh @@ -198,8 +198,7 @@ do_transfer() print_title "Transfer ${ip:2}" if [ ${rets} -ne 0 ] || [ ${retc} -ne 0 ]; then mptcp_lib_pr_fail "client exit code $retc, server $rets" - mptcp_lib_pr_err_stats "${listener_ns}" "${connector_ns}" "${port}" \ - "/tmp/${listener_ns}.out" "/tmp/${connector_ns}.out" + mptcp_lib_pr_err_stats "${listener_ns}" "${connector_ns}" "${port}" mptcp_lib_result_fail "transfer ${ip}" diff --git a/tools/testing/selftests/net/mptcp/simult_flows.sh b/tools/testing/selftests/net/mptcp/simult_flows.sh index fbd1685d1adc..503cb59571a4 100755 --- a/tools/testing/selftests/net/mptcp/simult_flows.sh +++ b/tools/testing/selftests/net/mptcp/simult_flows.sh @@ -200,8 +200,7 @@ do_transfer() fi mptcp_lib_pr_fail "client exit code $retc, server $rets" - mptcp_lib_pr_err_stats "${ns3}" "${ns1}" "${port}" \ - "/tmp/${ns3}.out" "/tmp/${ns1}.out" + mptcp_lib_pr_err_stats "${ns3}" "${ns1}" "${port}" ls -l $sin $cout ls -l $cin $sout From 2e6daf6b9bc86e3fe0e3217def16aaaaf0924801 Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" Date: Fri, 14 Nov 2025 19:12:07 +0100 Subject: [PATCH 0832/1381] selftests: mptcp: lib: stats: remove nstat rate columns With the MPTCP selftests, the nstat daemon is not used. It means that the last column (the rate) is always 0.0, and that's not something interesting to display. Then, this last column can be filtered out. Acked-by: Paolo Abeni Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20251114-net-next-mptcp-sft-count-cache-stats-timeout-v1-3-863cb04e1b7b@kernel.org Signed-off-by: Jakub Kicinski --- tools/testing/selftests/net/mptcp/mptcp_lib.sh | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tools/testing/selftests/net/mptcp/mptcp_lib.sh b/tools/testing/selftests/net/mptcp/mptcp_lib.sh index 4a26d4150603..c5571100f797 100644 --- a/tools/testing/selftests/net/mptcp/mptcp_lib.sh +++ b/tools/testing/selftests/net/mptcp/mptcp_lib.sh @@ -390,9 +390,9 @@ mptcp_lib_nstat_init() { mptcp_lib_nstat_get() { local ns="${1}" - # filter out non-*TCP stats + # filter out non-*TCP stats, and the rate (last column) NSTAT_HISTORY="/tmp/${ns}.nstat" ip netns exec "${ns}" nstat | - grep Tcp > "/tmp/${ns}.out" + grep -o ".*Tcp\S\+\s\+[0-9]\+" > "/tmp/${ns}.out" } # $1: ns, $2: MIB counter From 658e531417801cd8d9aac630cee712e160ab5380 Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" Date: Fri, 14 Nov 2025 19:12:08 +0100 Subject: [PATCH 0833/1381] selftests: mptcp: join: dump stats from history In case of errors, dump the stats from history instead of using nstat. There are multiple advantages to that: - The same filters from pr_err_stats are used, e.g. the unused 'rate' column is not displayed. - The counters are closer to the ones from when the test stopped. - While at it, the errors can be better presented: error colours, a small indentation to distinguish the different parts, extra new lines. Even if it should only happen in rare cases -- internal errors, or netns issues -- if no history is available, 'nstat' is used like before, just in case. Acked-by: Paolo Abeni Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20251114-net-next-mptcp-sft-count-cache-stats-timeout-v1-4-863cb04e1b7b@kernel.org Signed-off-by: Jakub Kicinski --- tools/testing/selftests/net/mptcp/mptcp_join.sh | 16 ++++++++++++---- tools/testing/selftests/net/mptcp/mptcp_lib.sh | 6 +++++- 2 files changed, 17 insertions(+), 5 deletions(-) diff --git a/tools/testing/selftests/net/mptcp/mptcp_join.sh b/tools/testing/selftests/net/mptcp/mptcp_join.sh index aee215d73b7c..54bac074f184 100755 --- a/tools/testing/selftests/net/mptcp/mptcp_join.sh +++ b/tools/testing/selftests/net/mptcp/mptcp_join.sh @@ -1146,12 +1146,20 @@ run_tests() do_transfer ${listener_ns} ${connector_ns} MPTCP MPTCP ${connect_addr} } +_dump_stats() +{ + local ns="${1}" + local side="${2}" + + mptcp_lib_print_err "${side} ns stats (${ns2})" + mptcp_lib_pr_nstat "${ns}" + echo +} + dump_stats() { - echo Server ns stats - ip netns exec $ns1 nstat -as | grep Tcp - echo Client ns stats - ip netns exec $ns2 nstat -as | grep Tcp + _dump_stats "${ns1}" "Server" + _dump_stats "${ns2}" "Client" } chk_csum_nr() diff --git a/tools/testing/selftests/net/mptcp/mptcp_lib.sh b/tools/testing/selftests/net/mptcp/mptcp_lib.sh index c5571100f797..fa91eebdbc47 100644 --- a/tools/testing/selftests/net/mptcp/mptcp_lib.sh +++ b/tools/testing/selftests/net/mptcp/mptcp_lib.sh @@ -110,7 +110,11 @@ mptcp_lib_pr_nstat() { local ns="${1}" local hist="/tmp/${ns}.out" - cat "${hist}" + if [ -f "${hist}" ]; then + awk '{ print " "$0 }' "${hist}" + else + ip netns exec "${ns}" nstat -as | grep Tcp + fi } # $1-2: listener/connector ns ; $3 port From 71388a9f331da20bab70e09bdf010f925ad7ce79 Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" Date: Fri, 14 Nov 2025 19:12:09 +0100 Subject: [PATCH 0834/1381] selftests: mptcp: lib: get counters from nstat history Before, 'nstat' was used to retrieve each individual counter: this means querying 4 different sources from /proc/net and iterating over 100+ counters each time. Instead, the stats could be retrieved once, and the output file could be parsed for each counter. Even better, such file is already present: the nstat history file. To be able to get this working, the nstat history file also needs to contains zero counters too, so it is still possible to know if a counter is missing or set to 0. This also simplifies mptcp_connect.sh: instead of checking multiple counters before and after a test to compute the difference, the stats history files can be reset before each test, and nstat can display only the difference. mptcp_lib_get_counter() continues to work when no history file is available: by fetching nstat directly, like before. This is the case in diag.sh and userspace_pm.sh where there is no need to save the history file. This is also the case in mptcp_join.sh, when 'run_tests' is executed in the background: easier to continue fetching counters than updating the history each time it is needed. Note: 'nstat' is called with '-s' in mptcp_lib_nstat_get(), so this helper can be called multiple times during the test if needed. Acked-by: Paolo Abeni Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20251114-net-next-mptcp-sft-count-cache-stats-timeout-v1-5-863cb04e1b7b@kernel.org Signed-off-by: Jakub Kicinski --- .../selftests/net/mptcp/mptcp_connect.sh | 85 ++++++++----------- .../testing/selftests/net/mptcp/mptcp_lib.sh | 16 +++- 2 files changed, 46 insertions(+), 55 deletions(-) diff --git a/tools/testing/selftests/net/mptcp/mptcp_connect.sh b/tools/testing/selftests/net/mptcp/mptcp_connect.sh index 4ff2d02429d3..3a804abebd2c 100755 --- a/tools/testing/selftests/net/mptcp/mptcp_connect.sh +++ b/tools/testing/selftests/net/mptcp/mptcp_connect.sh @@ -389,21 +389,6 @@ do_transfer() mptcp_lib_nstat_init "${connector_ns}" fi - local stat_synrx_last_l - local stat_ackrx_last_l - local stat_cookietx_last - local stat_cookierx_last - local stat_csum_err_s - local stat_csum_err_c - local stat_tcpfb_last_l - stat_synrx_last_l=$(mptcp_lib_get_counter "${listener_ns}" "MPTcpExtMPCapableSYNRX") - stat_ackrx_last_l=$(mptcp_lib_get_counter "${listener_ns}" "MPTcpExtMPCapableACKRX") - stat_cookietx_last=$(mptcp_lib_get_counter "${listener_ns}" "TcpExtSyncookiesSent") - stat_cookierx_last=$(mptcp_lib_get_counter "${listener_ns}" "TcpExtSyncookiesRecv") - stat_csum_err_s=$(mptcp_lib_get_counter "${listener_ns}" "MPTcpExtDataCsumErr") - stat_csum_err_c=$(mptcp_lib_get_counter "${connector_ns}" "MPTcpExtDataCsumErr") - stat_tcpfb_last_l=$(mptcp_lib_get_counter "${listener_ns}" "MPTcpExtMPCapableFallbackACK") - timeout ${timeout_test} \ ip netns exec ${listener_ns} \ ./mptcp_connect -t ${timeout_poll} -l -p $port -s ${srv_proto} \ @@ -458,38 +443,38 @@ do_transfer() rets=$? local extra="" - local stat_synrx_now_l - local stat_ackrx_now_l - local stat_cookietx_now - local stat_cookierx_now - local stat_ooo_now - local stat_tcpfb_now_l - stat_synrx_now_l=$(mptcp_lib_get_counter "${listener_ns}" "MPTcpExtMPCapableSYNRX") - stat_ackrx_now_l=$(mptcp_lib_get_counter "${listener_ns}" "MPTcpExtMPCapableACKRX") - stat_cookietx_now=$(mptcp_lib_get_counter "${listener_ns}" "TcpExtSyncookiesSent") - stat_cookierx_now=$(mptcp_lib_get_counter "${listener_ns}" "TcpExtSyncookiesRecv") - stat_ooo_now=$(mptcp_lib_get_counter "${listener_ns}" "TcpExtTCPOFOQueue") - stat_tcpfb_now_l=$(mptcp_lib_get_counter "${listener_ns}" "MPTcpExtMPCapableFallbackACK") + local stat_synrx + local stat_ackrx + local stat_cookietx + local stat_cookierx + local stat_ooo + local stat_tcpfb + stat_synrx=$(mptcp_lib_get_counter "${listener_ns}" "MPTcpExtMPCapableSYNRX") + stat_ackrx=$(mptcp_lib_get_counter "${listener_ns}" "MPTcpExtMPCapableACKRX") + stat_cookietx=$(mptcp_lib_get_counter "${listener_ns}" "TcpExtSyncookiesSent") + stat_cookierx=$(mptcp_lib_get_counter "${listener_ns}" "TcpExtSyncookiesRecv") + stat_ooo=$(mptcp_lib_get_counter "${listener_ns}" "TcpExtTCPOFOQueue") + stat_tcpfb=$(mptcp_lib_get_counter "${listener_ns}" "MPTcpExtMPCapableFallbackACK") - expect_synrx=$((stat_synrx_last_l)) - expect_ackrx=$((stat_ackrx_last_l)) + expect_synrx=0 + expect_ackrx=0 cookies=$(ip netns exec ${listener_ns} sysctl net.ipv4.tcp_syncookies) cookies=${cookies##*=} if [ ${cl_proto} = "MPTCP" ] && [ ${srv_proto} = "MPTCP" ]; then - expect_synrx=$((stat_synrx_last_l+connect_per_transfer)) - expect_ackrx=$((stat_ackrx_last_l+connect_per_transfer)) + expect_synrx=${connect_per_transfer} + expect_ackrx=${connect_per_transfer} fi - if [ ${stat_synrx_now_l} -lt ${expect_synrx} ]; then - mptcp_lib_pr_fail "lower MPC SYN rx (${stat_synrx_now_l})" \ + if [ ${stat_synrx} -lt ${expect_synrx} ]; then + mptcp_lib_pr_fail "lower MPC SYN rx (${stat_synrx})" \ "than expected (${expect_synrx})" retc=1 fi - if [ ${stat_ackrx_now_l} -lt ${expect_ackrx} ]; then - if [ ${stat_ooo_now} -eq 0 ]; then - mptcp_lib_pr_fail "lower MPC ACK rx (${stat_ackrx_now_l})" \ + if [ ${stat_ackrx} -lt ${expect_ackrx} ]; then + if [ ${stat_ooo} -eq 0 ]; then + mptcp_lib_pr_fail "lower MPC ACK rx (${stat_ackrx})" \ "than expected (${expect_ackrx})" rets=1 else @@ -503,47 +488,45 @@ do_transfer() csum_err_s=$(mptcp_lib_get_counter "${listener_ns}" "MPTcpExtDataCsumErr") csum_err_c=$(mptcp_lib_get_counter "${connector_ns}" "MPTcpExtDataCsumErr") - local csum_err_s_nr=$((csum_err_s - stat_csum_err_s)) - if [ $csum_err_s_nr -gt 0 ]; then - mptcp_lib_pr_fail "server got ${csum_err_s_nr} data checksum error[s]" + if [ $csum_err_s -gt 0 ]; then + mptcp_lib_pr_fail "server got ${csum_err_s} data checksum error[s]" rets=1 fi - local csum_err_c_nr=$((csum_err_c - stat_csum_err_c)) - if [ $csum_err_c_nr -gt 0 ]; then - mptcp_lib_pr_fail "client got ${csum_err_c_nr} data checksum error[s]" + if [ $csum_err_c -gt 0 ]; then + mptcp_lib_pr_fail "client got ${csum_err_c} data checksum error[s]" retc=1 fi fi - if [ ${stat_ooo_now} -eq 0 ] && [ ${stat_tcpfb_last_l} -ne ${stat_tcpfb_now_l} ]; then + if [ ${stat_ooo} -eq 0 ] && [ ${stat_tcpfb} -gt 0 ]; then mptcp_lib_pr_fail "unexpected fallback to TCP" rets=1 fi if [ $cookies -eq 2 ];then - if [ $stat_cookietx_last -ge $stat_cookietx_now ] ;then + if [ $stat_cookietx -eq 0 ] ;then extra+=" WARN: CookieSent: did not advance" fi - if [ $stat_cookierx_last -ge $stat_cookierx_now ] ;then + if [ $stat_cookierx -eq 0 ] ;then extra+=" WARN: CookieRecv: did not advance" fi else - if [ $stat_cookietx_last -ne $stat_cookietx_now ] ;then + if [ $stat_cookietx -gt 0 ] ;then extra+=" WARN: CookieSent: changed" fi - if [ $stat_cookierx_last -ne $stat_cookierx_now ] ;then + if [ $stat_cookierx -gt 0 ] ;then extra+=" WARN: CookieRecv: changed" fi fi - if [ ${stat_synrx_now_l} -gt ${expect_synrx} ]; then + if [ ${stat_synrx} -gt ${expect_synrx} ]; then extra+=" WARN: SYNRX: expect ${expect_synrx}," - extra+=" got ${stat_synrx_now_l} (probably retransmissions)" + extra+=" got ${stat_synrx} (probably retransmissions)" fi - if [ ${stat_ackrx_now_l} -gt ${expect_ackrx} ]; then + if [ ${stat_ackrx} -gt ${expect_ackrx} ]; then extra+=" WARN: ACKRX: expect ${expect_ackrx}," - extra+=" got ${stat_ackrx_now_l} (probably retransmissions)" + extra+=" got ${stat_ackrx} (probably retransmissions)" fi if [ $retc -eq 0 ] && [ $rets -eq 0 ]; then diff --git a/tools/testing/selftests/net/mptcp/mptcp_lib.sh b/tools/testing/selftests/net/mptcp/mptcp_lib.sh index fa91eebdbc47..91ec75ddcb96 100644 --- a/tools/testing/selftests/net/mptcp/mptcp_lib.sh +++ b/tools/testing/selftests/net/mptcp/mptcp_lib.sh @@ -111,7 +111,7 @@ mptcp_lib_pr_nstat() { local hist="/tmp/${ns}.out" if [ -f "${hist}" ]; then - awk '{ print " "$0 }' "${hist}" + awk '$2 != 0 { print " "$0 }' "${hist}" else ip netns exec "${ns}" nstat -as | grep Tcp fi @@ -388,6 +388,7 @@ mptcp_lib_is_v6() { mptcp_lib_nstat_init() { local ns="${1}" + rm -f "/tmp/${ns}."{nstat,out} NSTAT_HISTORY="/tmp/${ns}.nstat" ip netns exec "${ns}" nstat -n } @@ -395,18 +396,25 @@ mptcp_lib_nstat_get() { local ns="${1}" # filter out non-*TCP stats, and the rate (last column) - NSTAT_HISTORY="/tmp/${ns}.nstat" ip netns exec "${ns}" nstat | + NSTAT_HISTORY="/tmp/${ns}.nstat" ip netns exec "${ns}" nstat -sz | grep -o ".*Tcp\S\+\s\+[0-9]\+" > "/tmp/${ns}.out" } # $1: ns, $2: MIB counter +# Get the counter from the history (mptcp_lib_nstat_{init,get}()) if available. +# If not, get the counter from nstat ignoring any history. mptcp_lib_get_counter() { local ns="${1}" local counter="${2}" + local hist="/tmp/${ns}.out" local count - count=$(ip netns exec "${ns}" nstat -asz "${counter}" | - awk 'NR==1 {next} {print $2}') + if [[ -s "${hist}" && "${counter}" == *"Tcp"* ]]; then + count=$(awk "/^${counter} / {print \$2; exit}" "${hist}") + else + count=$(ip netns exec "${ns}" nstat -asz "${counter}" | + awk 'NR==1 {next} {print $2}') + fi if [ -z "${count}" ]; then mptcp_lib_fail_if_expected_feature "${counter} counter" return 1 From 8c1fe0a500519d21e9777433aa67d482862b2f3e Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" Date: Fri, 14 Nov 2025 19:12:10 +0100 Subject: [PATCH 0835/1381] selftests: mptcp: connect: avoid double packet traces When the same netns is used for the listener and the connector, no need to take exactly the same packet trace twice, one is enough. This avoids confusions when the traces are the same, and wasting resources which might not help reproducing an issue. While at it, avoid long lines and double spaces now that these lines are no longer aligned. Reviewed-by: Geliang Tang Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20251114-net-next-mptcp-sft-count-cache-stats-timeout-v1-6-863cb04e1b7b@kernel.org Signed-off-by: Jakub Kicinski --- tools/testing/selftests/net/mptcp/mptcp_connect.sh | 14 ++++++++++---- 1 file changed, 10 insertions(+), 4 deletions(-) diff --git a/tools/testing/selftests/net/mptcp/mptcp_connect.sh b/tools/testing/selftests/net/mptcp/mptcp_connect.sh index 3a804abebd2c..e0bfd9b4730c 100755 --- a/tools/testing/selftests/net/mptcp/mptcp_connect.sh +++ b/tools/testing/selftests/net/mptcp/mptcp_connect.sh @@ -375,11 +375,15 @@ do_transfer() local capfile="${rndh}-${connector_ns:0:3}-${listener_ns:0:3}-${cl_proto}-${srv_proto}-${connect_addr}-${port}" local capopt="-i any -s 65535 -B 32768 ${capuser}" - ip netns exec ${listener_ns} tcpdump ${capopt} -w "${capfile}-listener.pcap" >> "${capout}" 2>&1 & + ip netns exec ${listener_ns} tcpdump ${capopt} \ + -w "${capfile}-listener.pcap" >> "${capout}" 2>&1 & local cappid_listener=$! - ip netns exec ${connector_ns} tcpdump ${capopt} -w "${capfile}-connector.pcap" >> "${capout}" 2>&1 & - local cappid_connector=$! + if [ ${listener_ns} != ${connector_ns} ]; then + ip netns exec ${connector_ns} tcpdump ${capopt} \ + -w "${capfile}-connector.pcap" >> "${capout}" 2>&1 & + local cappid_connector=$! + fi sleep 1 fi @@ -416,7 +420,9 @@ do_transfer() if $capture; then sleep 1 kill ${cappid_listener} - kill ${cappid_connector} + if [ ${listener_ns} != ${connector_ns} ]; then + kill ${cappid_connector} + fi fi mptcp_lib_nstat_get "${listener_ns}" From 39348f5f2f13c2da122b5049952066c2e802d3a0 Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" Date: Fri, 14 Nov 2025 19:12:11 +0100 Subject: [PATCH 0836/1381] selftests: mptcp: wait for port instead of sleep After having started mptcp_connect in listening mode, 'mptcp_lib_wait_local_port_listen' can be used to wait for the listening socket to be ready. This is better than using the 'sleep' command, not to pause for a fixed amount of time, but waiting for an event. This helper is used in all other MPTCP selftests, but not in these two. Reviewed-by: Geliang Tang Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20251114-net-next-mptcp-sft-count-cache-stats-timeout-v1-7-863cb04e1b7b@kernel.org Signed-off-by: Jakub Kicinski --- tools/testing/selftests/net/mptcp/mptcp_sockopt.sh | 2 +- tools/testing/selftests/net/mptcp/userspace_pm.sh | 3 ++- 2 files changed, 3 insertions(+), 2 deletions(-) diff --git a/tools/testing/selftests/net/mptcp/mptcp_sockopt.sh b/tools/testing/selftests/net/mptcp/mptcp_sockopt.sh index 42d533b95ec7..6cde7429104b 100755 --- a/tools/testing/selftests/net/mptcp/mptcp_sockopt.sh +++ b/tools/testing/selftests/net/mptcp/mptcp_sockopt.sh @@ -178,7 +178,7 @@ do_transfer() ${local_addr} < "$sin" > "$sout" & local spid=$! - sleep 1 + mptcp_lib_wait_local_port_listen "${listener_ns}" "${port}" timeout ${timeout_test} \ ip netns exec ${connector_ns} \ diff --git a/tools/testing/selftests/net/mptcp/userspace_pm.sh b/tools/testing/selftests/net/mptcp/userspace_pm.sh index 87323942cb8a..e9ae1806ab07 100755 --- a/tools/testing/selftests/net/mptcp/userspace_pm.sh +++ b/tools/testing/selftests/net/mptcp/userspace_pm.sh @@ -211,7 +211,8 @@ make_connection() ip netns exec "$ns1" \ ./mptcp_connect -s MPTCP -w 300 -p $app_port -l $listen_addr > /dev/null 2>&1 & local server_pid=$! - sleep 0.5 + + mptcp_lib_wait_local_port_listen "${ns1}" "${port}" # Run the client, transfer $file and stay connected to the server # to conduct tests From eea2f44870d297f0535944bf4dbf49d8b90c838e Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" Date: Fri, 14 Nov 2025 19:12:12 +0100 Subject: [PATCH 0837/1381] selftests: mptcp: get stats just before timing out Recently, some debugging happened around a test that was timing out. The stats were showing connections being closed which was confusing because the closing state was caused by the timeout stopping the transfer. To avoid such confusion, the timeout is no longer done per mptcp_connect process, but separately. In case of timeout, the stats are now printed, then the apps are killed. The stats will still be printed after the kill, but that's fine, and this might even be useful, just in case. Timeout should be exceptional. Reviewed-by: Geliang Tang Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20251114-net-next-mptcp-sft-count-cache-stats-timeout-v1-8-863cb04e1b7b@kernel.org Signed-off-by: Jakub Kicinski --- .../selftests/net/mptcp/mptcp_connect.sh | 26 +++++++++----- .../testing/selftests/net/mptcp/mptcp_join.sh | 34 +++++++++++-------- .../testing/selftests/net/mptcp/mptcp_lib.sh | 13 +++++++ .../selftests/net/mptcp/mptcp_sockopt.sh | 26 +++++++++----- .../selftests/net/mptcp/simult_flows.sh | 29 ++++++++++------ 5 files changed, 86 insertions(+), 42 deletions(-) diff --git a/tools/testing/selftests/net/mptcp/mptcp_connect.sh b/tools/testing/selftests/net/mptcp/mptcp_connect.sh index e0bfd9b4730c..a6447f7a31fe 100755 --- a/tools/testing/selftests/net/mptcp/mptcp_connect.sh +++ b/tools/testing/selftests/net/mptcp/mptcp_connect.sh @@ -393,27 +393,35 @@ do_transfer() mptcp_lib_nstat_init "${connector_ns}" fi - timeout ${timeout_test} \ - ip netns exec ${listener_ns} \ - ./mptcp_connect -t ${timeout_poll} -l -p $port -s ${srv_proto} \ - $extra_args $local_addr < "$sin" > "$sout" & + ip netns exec ${listener_ns} \ + ./mptcp_connect -t ${timeout_poll} -l -p $port -s ${srv_proto} \ + $extra_args $local_addr < "$sin" > "$sout" & local spid=$! mptcp_lib_wait_local_port_listen "${listener_ns}" "${port}" local start start=$(date +%s%3N) - timeout ${timeout_test} \ - ip netns exec ${connector_ns} \ - ./mptcp_connect -t ${timeout_poll} -p $port -s ${cl_proto} \ - $extra_args $connect_addr < "$cin" > "$cout" & + ip netns exec ${connector_ns} \ + ./mptcp_connect -t ${timeout_poll} -p $port -s ${cl_proto} \ + $extra_args $connect_addr < "$cin" > "$cout" & local cpid=$! + mptcp_lib_wait_timeout "${timeout_test}" "${listener_ns}" \ + "${connector_ns}" "${port}" "${cpid}" "${spid}" & + local timeout_pid=$! + wait $cpid local retc=$? wait $spid local rets=$? + if kill -0 $timeout_pid; then + # Finished before the timeout: kill the background job + mptcp_lib_kill_group_wait $timeout_pid + timeout_pid=0 + fi + local stop stop=$(date +%s%3N) @@ -433,7 +441,7 @@ do_transfer() local duration duration=$((stop-start)) printf "(duration %05sms) " "${duration}" - if [ ${rets} -ne 0 ] || [ ${retc} -ne 0 ]; then + if [ ${rets} -ne 0 ] || [ ${retc} -ne 0 ] || [ ${timeout_pid} -ne 0 ]; then mptcp_lib_pr_fail "client exit code $retc, server $rets" mptcp_lib_pr_err_stats "${listener_ns}" "${connector_ns}" "${port}" diff --git a/tools/testing/selftests/net/mptcp/mptcp_join.sh b/tools/testing/selftests/net/mptcp/mptcp_join.sh index 54bac074f184..be61612f68fe 100755 --- a/tools/testing/selftests/net/mptcp/mptcp_join.sh +++ b/tools/testing/selftests/net/mptcp/mptcp_join.sh @@ -1024,38 +1024,38 @@ do_transfer() if [ "$test_linkfail" -gt 1 ];then listener_in="${sinfail}" fi - timeout ${timeout_test} \ - ip netns exec ${listener_ns} \ - ./mptcp_connect -t ${timeout_poll} -l -p ${port} -s ${srv_proto} \ - ${extra_srv_args} "${bind_addr}" < "${listener_in}" > "${sout}" & + ip netns exec ${listener_ns} \ + ./mptcp_connect -t ${timeout_poll} -l -p ${port} -s ${srv_proto} \ + ${extra_srv_args} "${bind_addr}" < "${listener_in}" > "${sout}" & local spid=$! mptcp_lib_wait_local_port_listen "${listener_ns}" "${port}" extra_cl_args="$extra_args $extra_cl_args" if [ "$test_linkfail" -eq 0 ];then - timeout ${timeout_test} \ - ip netns exec ${connector_ns} \ - ./mptcp_connect -t ${timeout_poll} -p $port -s ${cl_proto} \ - $extra_cl_args $connect_addr < "$cin" > "$cout" & + ip netns exec ${connector_ns} \ + ./mptcp_connect -t ${timeout_poll} -p $port -s ${cl_proto} \ + $extra_cl_args $connect_addr < "$cin" > "$cout" & elif [ "$test_linkfail" -eq 1 ] || [ "$test_linkfail" -eq 2 ];then connector_in="${cinsent}" ( cat "$cinfail" ; sleep 2; link_failure $listener_ns ; cat "$cinfail" ) | \ tee "$cinsent" | \ - timeout ${timeout_test} \ ip netns exec ${connector_ns} \ ./mptcp_connect -t ${timeout_poll} -p $port -s ${cl_proto} \ $extra_cl_args $connect_addr > "$cout" & else connector_in="${cinsent}" tee "$cinsent" < "$cinfail" | \ - timeout ${timeout_test} \ - ip netns exec ${connector_ns} \ - ./mptcp_connect -t ${timeout_poll} -p $port -s ${cl_proto} \ - $extra_cl_args $connect_addr > "$cout" & + ip netns exec ${connector_ns} \ + ./mptcp_connect -t ${timeout_poll} -p $port -s ${cl_proto} \ + $extra_cl_args $connect_addr > "$cout" & fi local cpid=$! + mptcp_lib_wait_timeout "${timeout_test}" "${listener_ns}" \ + "${connector_ns}" "${port}" "${cpid}" "${spid}" & + local timeout_pid=$! + pm_nl_set_endpoint $listener_ns $connector_ns $connect_addr check_cestab $listener_ns $connector_ns @@ -1064,12 +1064,18 @@ do_transfer() wait $spid local rets=$? + if kill -0 $timeout_pid; then + # Finished before the timeout: kill the background job + mptcp_lib_kill_group_wait $timeout_pid + timeout_pid=0 + fi + cond_stop_capture mptcp_lib_nstat_get "${listener_ns}" mptcp_lib_nstat_get "${connector_ns}" - if [ ${rets} -ne 0 ] || [ ${retc} -ne 0 ]; then + if [ ${rets} -ne 0 ] || [ ${retc} -ne 0 ] || [ ${timeout_pid} -ne 0 ]; then fail_test "client exit code $retc, server $rets" mptcp_lib_pr_err_stats "${listener_ns}" "${connector_ns}" "${port}" return 1 diff --git a/tools/testing/selftests/net/mptcp/mptcp_lib.sh b/tools/testing/selftests/net/mptcp/mptcp_lib.sh index 91ec75ddcb96..5fea7e7df628 100644 --- a/tools/testing/selftests/net/mptcp/mptcp_lib.sh +++ b/tools/testing/selftests/net/mptcp/mptcp_lib.sh @@ -350,6 +350,19 @@ mptcp_lib_evts_get_info() { mptcp_lib_get_info_value "${1}" "^type:${3:-1}," } +mptcp_lib_wait_timeout() { + local timeout_test="${1}" + local listener_ns="${2}" + local connector_ns="${3}" + local port="${4}" + shift 4 # rest are PIDs + + sleep "${timeout_test}" + mptcp_lib_print_err "timeout" + mptcp_lib_pr_err_stats "${listener_ns}" "${connector_ns}" "${port}" + kill "${@}" 2>/dev/null +} + # $1: PID mptcp_lib_kill_wait() { [ "${1}" -eq 0 ] && return 0 diff --git a/tools/testing/selftests/net/mptcp/mptcp_sockopt.sh b/tools/testing/selftests/net/mptcp/mptcp_sockopt.sh index 6cde7429104b..ab8bce06b262 100755 --- a/tools/testing/selftests/net/mptcp/mptcp_sockopt.sh +++ b/tools/testing/selftests/net/mptcp/mptcp_sockopt.sh @@ -172,31 +172,39 @@ do_transfer() mptcp_lib_nstat_init "${listener_ns}" mptcp_lib_nstat_init "${connector_ns}" - timeout ${timeout_test} \ - ip netns exec ${listener_ns} \ - $mptcp_connect -t ${timeout_poll} -l -M 1 -p $port -s ${srv_proto} -c "${cmsg}" \ - ${local_addr} < "$sin" > "$sout" & + ip netns exec ${listener_ns} \ + $mptcp_connect -t ${timeout_poll} -l -M 1 -p $port -s ${srv_proto} -c "${cmsg}" \ + ${local_addr} < "$sin" > "$sout" & local spid=$! mptcp_lib_wait_local_port_listen "${listener_ns}" "${port}" - timeout ${timeout_test} \ - ip netns exec ${connector_ns} \ - $mptcp_connect -t ${timeout_poll} -M 2 -p $port -s ${cl_proto} -c "${cmsg}" \ - $connect_addr < "$cin" > "$cout" & + ip netns exec ${connector_ns} \ + $mptcp_connect -t ${timeout_poll} -M 2 -p $port -s ${cl_proto} -c "${cmsg}" \ + $connect_addr < "$cin" > "$cout" & local cpid=$! + mptcp_lib_wait_timeout "${timeout_test}" "${listener_ns}" \ + "${connector_ns}" "${port}" "${cpid}" "${spid}" & + local timeout_pid=$! + wait $cpid local retc=$? wait $spid local rets=$? + if kill -0 $timeout_pid; then + # Finished before the timeout: kill the background job + mptcp_lib_kill_group_wait $timeout_pid + timeout_pid=0 + fi + mptcp_lib_nstat_get "${listener_ns}" mptcp_lib_nstat_get "${connector_ns}" print_title "Transfer ${ip:2}" - if [ ${rets} -ne 0 ] || [ ${retc} -ne 0 ]; then + if [ ${rets} -ne 0 ] || [ ${retc} -ne 0 ] || [ ${timeout_pid} -ne 0 ]; then mptcp_lib_pr_fail "client exit code $retc, server $rets" mptcp_lib_pr_err_stats "${listener_ns}" "${connector_ns}" "${port}" diff --git a/tools/testing/selftests/net/mptcp/simult_flows.sh b/tools/testing/selftests/net/mptcp/simult_flows.sh index 503cb59571a4..806aaa7d2d61 100755 --- a/tools/testing/selftests/net/mptcp/simult_flows.sh +++ b/tools/testing/selftests/net/mptcp/simult_flows.sh @@ -158,25 +158,33 @@ do_transfer() mptcp_lib_nstat_init "${ns3}" mptcp_lib_nstat_init "${ns1}" - timeout ${timeout_test} \ - ip netns exec ${ns3} \ - ./mptcp_connect -jt ${timeout_poll} -l -p $port -T $max_time \ - 0.0.0.0 < "$sin" > "$sout" & + ip netns exec ${ns3} \ + ./mptcp_connect -jt ${timeout_poll} -l -p $port -T $max_time \ + 0.0.0.0 < "$sin" > "$sout" & local spid=$! mptcp_lib_wait_local_port_listen "${ns3}" "${port}" - timeout ${timeout_test} \ - ip netns exec ${ns1} \ - ./mptcp_connect -jt ${timeout_poll} -p $port -T $max_time \ - 10.0.3.3 < "$cin" > "$cout" & + ip netns exec ${ns1} \ + ./mptcp_connect -jt ${timeout_poll} -p $port -T $max_time \ + 10.0.3.3 < "$cin" > "$cout" & local cpid=$! + mptcp_lib_wait_timeout "${timeout_test}" "${ns3}" "${ns1}" "${port}" \ + "${cpid}" "${spid}" & + local timeout_pid=$! + wait $cpid local retc=$? wait $spid local rets=$? + if kill -0 $timeout_pid; then + # Finished before the timeout: kill the background job + mptcp_lib_kill_group_wait $timeout_pid + timeout_pid=0 + fi + if $capture; then sleep 1 kill ${cappid_listener} @@ -191,8 +199,9 @@ do_transfer() cmp $cin $sout > /dev/null 2>&1 local cmpc=$? - if [ $retc -eq 0 ] && [ $rets -eq 0 ] && \ - [ $cmpc -eq 0 ] && [ $cmps -eq 0 ]; then + if [ $retc -eq 0 ] && [ $rets -eq 0 ] && + [ $cmpc -eq 0 ] && [ $cmps -eq 0 ] && + [ $timeout_pid -eq 0 ]; then printf "%-16s" " max $max_time " mptcp_lib_pr_ok cat "$capout" From 889e6af87734389068fd924431f2141ddacad65e Mon Sep 17 00:00:00 2001 From: Vadim Fedorenko Date: Sun, 16 Nov 2025 09:46:09 +0000 Subject: [PATCH 0838/1381] bnx2x: convert to use ndo_hwtstamp callbacks The driver implemented SIOCSHWTSTAMP ioctl command only, but at the same time it has configuration stored in a private structure. Implement both ndo_hwtstamp_set and ndo_hwtstamp_get callback using stored info. ndo_hwtstamp_set callback implements a check for unsupported 1-step timestamping. The same check is removed from bnx2x_configure_ptp_filters function as it's not needed anymore. Another call site of bnx2x_configure_ptp_filters has hwtstamp_ioctl_called guard. Reviewed-by: Kory Maincent Signed-off-by: Vadim Fedorenko Link: https://patch.msgid.link/20251116094610.3932005-2-vadim.fedorenko@linux.dev Signed-off-by: Jakub Kicinski --- .../net/ethernet/broadcom/bnx2x/bnx2x_main.c | 70 +++++++++++++------ 1 file changed, 48 insertions(+), 22 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c index f0f05d7315ac..aca4267babc8 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c @@ -308,8 +308,11 @@ static int bnx2x_set_storm_rx_mode(struct bnx2x *bp); /**************************************************************************** * General service functions ****************************************************************************/ - -static int bnx2x_hwtstamp_ioctl(struct bnx2x *bp, struct ifreq *ifr); +static int bnx2x_hwtstamp_set(struct net_device *dev, + struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack); +static int bnx2x_hwtstamp_get(struct net_device *dev, + struct kernel_hwtstamp_config *config); static void __storm_memset_dma_mapping(struct bnx2x *bp, u32 addr, dma_addr_t mapping) @@ -12813,14 +12816,9 @@ static int bnx2x_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) if (!netif_running(dev)) return -EAGAIN; - switch (cmd) { - case SIOCSHWTSTAMP: - return bnx2x_hwtstamp_ioctl(bp, ifr); - default: - DP(NETIF_MSG_LINK, "ioctl: phy id 0x%x, reg 0x%x, val_in 0x%x\n", - mdio->phy_id, mdio->reg_num, mdio->val_in); - return mdio_mii_ioctl(&bp->mdio, mdio, cmd); - } + DP(NETIF_MSG_LINK, "ioctl: phy id 0x%x, reg 0x%x, val_in 0x%x\n", + mdio->phy_id, mdio->reg_num, mdio->val_in); + return mdio_mii_ioctl(&bp->mdio, mdio, cmd); } static int bnx2x_validate_addr(struct net_device *dev) @@ -13036,6 +13034,8 @@ static const struct net_device_ops bnx2x_netdev_ops = { .ndo_get_phys_port_id = bnx2x_get_phys_port_id, .ndo_set_vf_link_state = bnx2x_set_vf_link_state, .ndo_features_check = bnx2x_features_check, + .ndo_hwtstamp_get = bnx2x_hwtstamp_get, + .ndo_hwtstamp_set = bnx2x_hwtstamp_set, }; static int bnx2x_init_dev(struct bnx2x *bp, struct pci_dev *pdev, @@ -15350,31 +15350,57 @@ int bnx2x_configure_ptp_filters(struct bnx2x *bp) return 0; } -static int bnx2x_hwtstamp_ioctl(struct bnx2x *bp, struct ifreq *ifr) +static int bnx2x_hwtstamp_set(struct net_device *dev, + struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack) { - struct hwtstamp_config config; + struct bnx2x *bp = netdev_priv(dev); int rc; - DP(BNX2X_MSG_PTP, "HWTSTAMP IOCTL called\n"); + DP(BNX2X_MSG_PTP, "HWTSTAMP SET called\n"); - if (copy_from_user(&config, ifr->ifr_data, sizeof(config))) - return -EFAULT; + if (!netif_running(dev)) { + NL_SET_ERR_MSG_MOD(extack, "Device is down"); + return -EAGAIN; + } DP(BNX2X_MSG_PTP, "Requested tx_type: %d, requested rx_filters = %d\n", - config.tx_type, config.rx_filter); + config->tx_type, config->rx_filter); + + switch (config->tx_type) { + case HWTSTAMP_TX_ON: + case HWTSTAMP_TX_OFF: + break; + default: + NL_SET_ERR_MSG_MOD(extack, + "One-step timestamping is not supported"); + return -ERANGE; + } bp->hwtstamp_ioctl_called = true; - bp->tx_type = config.tx_type; - bp->rx_filter = config.rx_filter; + bp->tx_type = config->tx_type; + bp->rx_filter = config->rx_filter; rc = bnx2x_configure_ptp_filters(bp); - if (rc) + if (rc) { + NL_SET_ERR_MSG_MOD(extack, "HW configuration failure"); return rc; + } - config.rx_filter = bp->rx_filter; + config->rx_filter = bp->rx_filter; - return copy_to_user(ifr->ifr_data, &config, sizeof(config)) ? - -EFAULT : 0; + return 0; +} + +static int bnx2x_hwtstamp_get(struct net_device *dev, + struct kernel_hwtstamp_config *config) +{ + struct bnx2x *bp = netdev_priv(dev); + + config->rx_filter = bp->rx_filter; + config->tx_type = bp->tx_type; + + return 0; } /* Configures HW for PTP */ From 89ae72f21be369a3f245a65f3c44a0f4ffde55bd Mon Sep 17 00:00:00 2001 From: Vadim Fedorenko Date: Sun, 16 Nov 2025 09:46:10 +0000 Subject: [PATCH 0839/1381] qede: convert to use ndo_hwtstamp callbacks The driver implemented SIOCSHWTSTAMP ioctl cmd only, but it stores configuration in private structure, so it can be reported back to users. Implement both ndo_hwtstamp_set and ndo_hwtstamp_set callbacks. ndo_hwtstamp_set implements a check of unsupported 1-step timestamping and qede_ptp_cfg_filters() becomes void as it cannot fail anymore. Reviewed-by: Kory Maincent Signed-off-by: Vadim Fedorenko Link: https://patch.msgid.link/20251116094610.3932005-3-vadim.fedorenko@linux.dev Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/qlogic/qede/qede_main.c | 22 +----- drivers/net/ethernet/qlogic/qede/qede_ptp.c | 80 ++++++++++++-------- drivers/net/ethernet/qlogic/qede/qede_ptp.h | 6 +- 3 files changed, 56 insertions(+), 52 deletions(-) diff --git a/drivers/net/ethernet/qlogic/qede/qede_main.c b/drivers/net/ethernet/qlogic/qede/qede_main.c index b5d744d2586f..66ab1b9d65a1 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_main.c +++ b/drivers/net/ethernet/qlogic/qede/qede_main.c @@ -506,25 +506,6 @@ static int qede_set_vf_trust(struct net_device *dev, int vfidx, bool setting) } #endif -static int qede_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) -{ - struct qede_dev *edev = netdev_priv(dev); - - if (!netif_running(dev)) - return -EAGAIN; - - switch (cmd) { - case SIOCSHWTSTAMP: - return qede_ptp_hw_ts(edev, ifr); - default: - DP_VERBOSE(edev, QED_MSG_DEBUG, - "default IOCTL cmd 0x%x\n", cmd); - return -EOPNOTSUPP; - } - - return 0; -} - static void qede_fp_sb_dump(struct qede_dev *edev, struct qede_fastpath *fp) { char *p_sb = (char *)fp->sb_info->sb_virt; @@ -717,7 +698,6 @@ static const struct net_device_ops qede_netdev_ops = { .ndo_set_mac_address = qede_set_mac_addr, .ndo_validate_addr = eth_validate_addr, .ndo_change_mtu = qede_change_mtu, - .ndo_eth_ioctl = qede_ioctl, .ndo_tx_timeout = qede_tx_timeout, #ifdef CONFIG_QED_SRIOV .ndo_set_vf_mac = qede_set_vf_mac, @@ -742,6 +722,8 @@ static const struct net_device_ops qede_netdev_ops = { #endif .ndo_xdp_xmit = qede_xdp_transmit, .ndo_setup_tc = qede_setup_tc_offload, + .ndo_hwtstamp_get = qede_hwtstamp_get, + .ndo_hwtstamp_set = qede_hwtstamp_set, }; static const struct net_device_ops qede_netdev_vf_ops = { diff --git a/drivers/net/ethernet/qlogic/qede/qede_ptp.c b/drivers/net/ethernet/qlogic/qede/qede_ptp.c index a38f1e72c62b..d351be5fbda1 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_ptp.c +++ b/drivers/net/ethernet/qlogic/qede/qede_ptp.c @@ -199,18 +199,15 @@ static u64 qede_ptp_read_cc(struct cyclecounter *cc) return phc_cycles; } -static int qede_ptp_cfg_filters(struct qede_dev *edev) +static void qede_ptp_cfg_filters(struct qede_dev *edev) { enum qed_ptp_hwtstamp_tx_type tx_type = QED_PTP_HWTSTAMP_TX_ON; enum qed_ptp_filter_type rx_filter = QED_PTP_FILTER_NONE; struct qede_ptp *ptp = edev->ptp; - if (!ptp) - return -EIO; - if (!ptp->hw_ts_ioctl_called) { DP_INFO(edev, "TS IOCTL not called\n"); - return 0; + return; } switch (ptp->tx_type) { @@ -223,11 +220,6 @@ static int qede_ptp_cfg_filters(struct qede_dev *edev) clear_bit(QEDE_FLAGS_TX_TIMESTAMPING_EN, &edev->flags); tx_type = QED_PTP_HWTSTAMP_TX_OFF; break; - - case HWTSTAMP_TX_ONESTEP_SYNC: - case HWTSTAMP_TX_ONESTEP_P2P: - DP_ERR(edev, "One-step timestamping is not supported\n"); - return -ERANGE; } spin_lock_bh(&ptp->lock); @@ -286,39 +278,65 @@ static int qede_ptp_cfg_filters(struct qede_dev *edev) ptp->ops->cfg_filters(edev->cdev, rx_filter, tx_type); spin_unlock_bh(&ptp->lock); +} + +int qede_hwtstamp_set(struct net_device *netdev, + struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack) +{ + struct qede_dev *edev = netdev_priv(netdev); + struct qede_ptp *ptp; + + if (!netif_running(netdev)) { + NL_SET_ERR_MSG_MOD(extack, "Device is down"); + return -EAGAIN; + } + + ptp = edev->ptp; + if (!ptp) { + NL_SET_ERR_MSG_MOD(extack, "HW timestamping is not supported"); + return -EIO; + } + + DP_VERBOSE(edev, QED_MSG_DEBUG, + "HWTSTAMP SET: Requested tx_type = %d, requested rx_filters = %d\n", + config->tx_type, config->rx_filter); + + switch (config->tx_type) { + case HWTSTAMP_TX_ON: + case HWTSTAMP_TX_OFF: + break; + default: + NL_SET_ERR_MSG_MOD(extack, + "One-step timestamping is not supported"); + return -ERANGE; + } + + ptp->hw_ts_ioctl_called = 1; + ptp->tx_type = config->tx_type; + ptp->rx_filter = config->rx_filter; + + qede_ptp_cfg_filters(edev); + + config->rx_filter = ptp->rx_filter; return 0; } -int qede_ptp_hw_ts(struct qede_dev *edev, struct ifreq *ifr) +int qede_hwtstamp_get(struct net_device *netdev, + struct kernel_hwtstamp_config *config) { - struct hwtstamp_config config; + struct qede_dev *edev = netdev_priv(netdev); struct qede_ptp *ptp; - int rc; ptp = edev->ptp; if (!ptp) return -EIO; - if (copy_from_user(&config, ifr->ifr_data, sizeof(config))) - return -EFAULT; + config->tx_type = ptp->tx_type; + config->rx_filter = ptp->rx_filter; - DP_VERBOSE(edev, QED_MSG_DEBUG, - "HWTSTAMP IOCTL: Requested tx_type = %d, requested rx_filters = %d\n", - config.tx_type, config.rx_filter); - - ptp->hw_ts_ioctl_called = 1; - ptp->tx_type = config.tx_type; - ptp->rx_filter = config.rx_filter; - - rc = qede_ptp_cfg_filters(edev); - if (rc) - return rc; - - config.rx_filter = ptp->rx_filter; - - return copy_to_user(ifr->ifr_data, &config, - sizeof(config)) ? -EFAULT : 0; + return 0; } int qede_ptp_get_ts_info(struct qede_dev *edev, struct kernel_ethtool_ts_info *info) diff --git a/drivers/net/ethernet/qlogic/qede/qede_ptp.h b/drivers/net/ethernet/qlogic/qede/qede_ptp.h index adafc894797e..88f168395812 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_ptp.h +++ b/drivers/net/ethernet/qlogic/qede/qede_ptp.h @@ -14,7 +14,11 @@ void qede_ptp_rx_ts(struct qede_dev *edev, struct sk_buff *skb); void qede_ptp_tx_ts(struct qede_dev *edev, struct sk_buff *skb); -int qede_ptp_hw_ts(struct qede_dev *edev, struct ifreq *req); +int qede_hwtstamp_get(struct net_device *netdev, + struct kernel_hwtstamp_config *config); +int qede_hwtstamp_set(struct net_device *netdev, + struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack); void qede_ptp_disable(struct qede_dev *edev); int qede_ptp_enable(struct qede_dev *edev); int qede_ptp_get_ts_info(struct qede_dev *edev, struct kernel_ethtool_ts_info *ts); From 7cf3ac8a9c0aeef2803018d3d371f886839d0b52 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Sat, 15 Nov 2025 23:09:59 -0800 Subject: [PATCH 0840/1381] NFC: mei_phy: fix kernel-doc warnings Fix kernel-doc warnings in mei_phy.h to avoid build warnings and to improve and documentation: mei_phy.h:15: warning: missing initial short description on line: * struct nfc_mei_phy mei_phy.h:19: warning: bad line: Signed-off-by: Randy Dunlap Reviewed-by: Krzysztof Kozlowski Link: https://patch.msgid.link/20251116070959.85055-1-rdunlap@infradead.org Signed-off-by: Jakub Kicinski --- drivers/nfc/mei_phy.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/nfc/mei_phy.h b/drivers/nfc/mei_phy.h index 2b1edb3eba15..9b9c5eb54e07 100644 --- a/drivers/nfc/mei_phy.h +++ b/drivers/nfc/mei_phy.h @@ -12,11 +12,11 @@ #define MEI_NFC_MAX_HCI_PAYLOAD 300 /** - * struct nfc_mei_phy + * struct nfc_mei_phy - NFC description of the MEI PHY and interface functions * * @cldev: mei client device * @hdev: nfc hci device - + * * @send_wq: send completion wait queue * @fw_ivn: NFC Interface Version Number * @vendor_id: NFC manufacturer ID From 511171e47f8bfd922ce76d10ce1487ae469cdfda Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Fri, 14 Nov 2025 15:28:24 +0000 Subject: [PATCH 0841/1381] net: stmmac: add stmmac_plat_dat_alloc() Add a function to allocate and initialise the plat_stmmacenet_data structure with default values. Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vJviq-0000000EVjP-0c0l@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c | 2 +- drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c | 2 +- drivers/net/ethernet/stmicro/stmmac/stmmac.h | 2 ++ drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 12 ++++++++++++ drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c | 2 +- .../net/ethernet/stmicro/stmmac/stmmac_platform.c | 2 +- 6 files changed, 18 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c index b2194e414ec1..7e56fbc3e141 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c @@ -1286,7 +1286,7 @@ static int intel_eth_pci_probe(struct pci_dev *pdev, if (!intel_priv) return -ENOMEM; - plat = devm_kzalloc(&pdev->dev, sizeof(*plat), GFP_KERNEL); + plat = stmmac_plat_dat_alloc(&pdev->dev); if (!plat) return -ENOMEM; diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c index dd2fc39ec3e2..2d803fa37e21 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c @@ -559,7 +559,7 @@ static int loongson_dwmac_probe(struct pci_dev *pdev, const struct pci_device_id struct loongson_data *ld; int ret; - plat = devm_kzalloc(&pdev->dev, sizeof(*plat), GFP_KERNEL); + plat = stmmac_plat_dat_alloc(&pdev->dev); if (!plat) return -ENOMEM; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac.h b/drivers/net/ethernet/stmicro/stmmac/stmmac.h index 0ea74c88a779..e9ed5086c049 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac.h +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac.h @@ -408,6 +408,8 @@ int stmmac_reinit_ringparam(struct net_device *dev, u32 rx_size, u32 tx_size); int stmmac_set_clk_tx_rate(void *bsp_priv, struct clk *clk_tx_i, phy_interface_t interface, int speed); +struct plat_stmmacenet_data *stmmac_plat_dat_alloc(struct device *dev); + static inline bool stmmac_xdp_is_enabled(struct stmmac_priv *priv) { return !!priv->xdp_prog; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index d202f604161e..400b4b955820 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -7555,6 +7555,18 @@ static void stmmac_unregister_devlink(struct stmmac_priv *priv) devlink_free(priv->devlink); } +struct plat_stmmacenet_data *stmmac_plat_dat_alloc(struct device *dev) +{ + struct plat_stmmacenet_data *plat_dat; + + plat_dat = devm_kzalloc(dev, sizeof(*plat_dat), GFP_KERNEL); + if (!plat_dat) + return NULL; + + return plat_dat; +} +EXPORT_SYMBOL_GPL(stmmac_plat_dat_alloc); + /** * stmmac_dvr_probe * @device: device pointer diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c index 94b3a3b27270..622cdbeca20f 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c @@ -191,7 +191,7 @@ static int stmmac_pci_probe(struct pci_dev *pdev, int ret; int i; - plat = devm_kzalloc(&pdev->dev, sizeof(*plat), GFP_KERNEL); + plat = stmmac_plat_dat_alloc(&pdev->dev); if (!plat) return -ENOMEM; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c index 6483d52b4c0f..38d574907a04 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c @@ -436,7 +436,7 @@ stmmac_probe_config_dt(struct platform_device *pdev, u8 *mac) void *ret; int rc; - plat = devm_kzalloc(&pdev->dev, sizeof(*plat), GFP_KERNEL); + plat = stmmac_plat_dat_alloc(&pdev->dev); if (!plat) return ERR_PTR(-ENOMEM); From 99e6ddaabdb4a4562d18eb7d2225f98801235904 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Fri, 14 Nov 2025 15:28:29 +0000 Subject: [PATCH 0842/1381] net: stmmac: move initialisation of phy_addr to stmmac_plat_dat_alloc() Move the default initialisation of plat_dat->phy_addr to stmmac_plat_dat_alloc(). Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vJviv-0000000EVjV-1CLF@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c | 3 --- drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 5 +++++ drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c | 1 - drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c | 2 -- 4 files changed, 5 insertions(+), 6 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c index 2d803fa37e21..8593411844bc 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c @@ -114,9 +114,6 @@ static void loongson_default_data(struct pci_dev *pdev, plat->clk_ref_rate = 125000000; plat->clk_ptp_rate = 125000000; - /* Default to phy auto-detection */ - plat->phy_addr = -1; - plat->dma_cfg->pbl = 32; plat->dma_cfg->pblx8 = true; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index 400b4b955820..1851f7d0702d 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -7563,6 +7563,11 @@ struct plat_stmmacenet_data *stmmac_plat_dat_alloc(struct device *dev) if (!plat_dat) return NULL; + /* Set the defaults: + * - phy autodetection + */ + plat_dat->phy_addr = -1; + return plat_dat; } EXPORT_SYMBOL_GPL(stmmac_plat_dat_alloc); diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c index 622cdbeca20f..b981a9dd511d 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c @@ -112,7 +112,6 @@ static int snps_gmac5_default_data(struct pci_dev *pdev, } plat->bus_id = 1; - plat->phy_addr = -1; plat->phy_interface = PHY_INTERFACE_MODE_GMII; plat->dma_cfg->pbl = 32; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c index 38d574907a04..745032fd46ce 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c @@ -480,8 +480,6 @@ stmmac_probe_config_dt(struct platform_device *pdev, u8 *mac) plat->bus_id = ++bus_id; } - /* Default to phy auto-detection */ - plat->phy_addr = -1; /* Default to get clk_csr from stmmac_clk_csr_set(), * or get clk_csr from device tree. From ae4f29712bf3468111b9ca65b6ecd94c062cb98f Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Fri, 14 Nov 2025 15:28:34 +0000 Subject: [PATCH 0843/1381] net: stmmac: move initialisation of clk_csr to stmmac_plat_dat_alloc() Move the default initialisation of plat_dat->clk_csr to stmmac_plat_dat_alloc(). Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vJvj0-0000000EVjb-1jDh@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 2 ++ drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c | 5 ----- 2 files changed, 2 insertions(+), 5 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index 1851f7d0702d..a36e8a90fcaa 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -7565,8 +7565,10 @@ struct plat_stmmacenet_data *stmmac_plat_dat_alloc(struct device *dev) /* Set the defaults: * - phy autodetection + * - determine GMII_Address CR field from CSR clock */ plat_dat->phy_addr = -1; + plat_dat->clk_csr = -1; return plat_dat; } diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c index 745032fd46ce..fe3d95274fd6 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c @@ -480,11 +480,6 @@ stmmac_probe_config_dt(struct platform_device *pdev, u8 *mac) plat->bus_id = ++bus_id; } - - /* Default to get clk_csr from stmmac_clk_csr_set(), - * or get clk_csr from device tree. - */ - plat->clk_csr = -1; if (of_property_read_u32(np, "snps,clk-csr", &plat->clk_csr)) of_property_read_u32(np, "clk_csr", &plat->clk_csr); From 528478a746a514a542f769a93cc3f0b5b847fe27 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Fri, 14 Nov 2025 15:28:39 +0000 Subject: [PATCH 0844/1381] net: stmmac: move initialisation of maxmtu to stmmac_plat_dat_alloc() Move the default initialisation of plat_dat->maxmtu to JUMBO_LEN to stmmac_plat_dat_alloc(). Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vJvj5-0000000EVji-2EYA@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c | 6 ------ drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c | 3 --- drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 2 ++ drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c | 6 ------ drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c | 5 ----- 5 files changed, 2 insertions(+), 20 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c index 7e56fbc3e141..cf69e659c415 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c @@ -576,9 +576,6 @@ static void common_default_data(struct plat_stmmacenet_data *plat) /* Set default value for unicast filter entries */ plat->unicast_filter_entries = 1; - /* Set the maxmtu to a default of JUMBO_LEN */ - plat->maxmtu = JUMBO_LEN; - /* Set default number of RX and TX queues to use */ plat->tx_queues_to_use = 1; plat->rx_queues_to_use = 1; @@ -712,9 +709,6 @@ static int intel_mgbe_common_data(struct pci_dev *pdev, /* Set default value for unicast filter entries */ plat->unicast_filter_entries = 1; - /* Set the maxmtu to a default of JUMBO_LEN */ - plat->maxmtu = JUMBO_LEN; - plat->flags |= STMMAC_FLAG_VLAN_FAIL_Q_EN; /* Use the last Rx queue */ diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c index 8593411844bc..89232c788c61 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c @@ -101,9 +101,6 @@ static void loongson_default_data(struct pci_dev *pdev, /* Set default value for unicast filter entries */ plat->unicast_filter_entries = 1; - /* Set the maxmtu to a default of JUMBO_LEN */ - plat->maxmtu = JUMBO_LEN; - /* Disable Priority config by default */ plat->tx_queues_cfg[0].use_prio = false; plat->rx_queues_cfg[0].use_prio = false; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index a36e8a90fcaa..0763ed06715e 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -7566,9 +7566,11 @@ struct plat_stmmacenet_data *stmmac_plat_dat_alloc(struct device *dev) /* Set the defaults: * - phy autodetection * - determine GMII_Address CR field from CSR clock + * - allow MTU up to JUMBO_LEN */ plat_dat->phy_addr = -1; plat_dat->clk_csr = -1; + plat_dat->maxmtu = JUMBO_LEN; return plat_dat; } diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c index b981a9dd511d..b0b4358e0adf 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c @@ -34,9 +34,6 @@ static void common_default_data(struct plat_stmmacenet_data *plat) /* Set default value for unicast filter entries */ plat->unicast_filter_entries = 1; - /* Set the maxmtu to a default of JUMBO_LEN */ - plat->maxmtu = JUMBO_LEN; - /* Set default number of RX and TX queues to use */ plat->tx_queues_to_use = 1; plat->rx_queues_to_use = 1; @@ -87,9 +84,6 @@ static int snps_gmac5_default_data(struct pci_dev *pdev, /* Set default value for unicast filter entries */ plat->unicast_filter_entries = 1; - /* Set the maxmtu to a default of JUMBO_LEN */ - plat->maxmtu = JUMBO_LEN; - /* Set default number of RX and TX queues to use */ plat->tx_queues_to_use = 4; plat->rx_queues_to_use = 4; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c index fe3d95274fd6..81a599475577 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c @@ -508,11 +508,6 @@ stmmac_probe_config_dt(struct platform_device *pdev, u8 *mac) plat->flags |= STMMAC_FLAG_EN_TX_LPI_CLOCKGATING; } - /* Set the maxmtu to a default of JUMBO_LEN in case the - * parameter is not present in the device tree. - */ - plat->maxmtu = JUMBO_LEN; - /* Set default value for multicast hash bins */ plat->multicast_filter_bins = HASH_TABLE_SIZE; From 07cedb9eed41067b059b4fe8149b5c60080add87 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Fri, 14 Nov 2025 15:28:44 +0000 Subject: [PATCH 0845/1381] net: stmmac: move initialisation of multicast_filter_bins to stmmac_plat_dat_alloc() Move the default initialisation of plat_dat->multicast_filter_bins to HASH_TABLE_SIZE to stmmac_plat_dat_alloc(). This means platform glue only needs to override this if different. Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vJvjA-0000000EVjo-2qVn@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c | 6 ------ drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c | 2 +- drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 2 ++ drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c | 6 ------ drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c | 3 --- 5 files changed, 3 insertions(+), 16 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c index cf69e659c415..a4ef3ec070e2 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c @@ -570,9 +570,6 @@ static void common_default_data(struct plat_stmmacenet_data *plat) plat->mdio_bus_data->needs_reset = true; - /* Set default value for multicast hash bins */ - plat->multicast_filter_bins = HASH_TABLE_SIZE; - /* Set default value for unicast filter entries */ plat->unicast_filter_entries = 1; @@ -703,9 +700,6 @@ static int intel_mgbe_common_data(struct pci_dev *pdev, plat->ptp_clk_freq_config = intel_mgbe_ptp_clk_freq_config; - /* Set default value for multicast hash bins */ - plat->multicast_filter_bins = HASH_TABLE_SIZE; - /* Set default value for unicast filter entries */ plat->unicast_filter_entries = 1; diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c index 89232c788c61..cfdda9e82a19 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c @@ -95,7 +95,7 @@ static void loongson_default_data(struct pci_dev *pdev, plat->core_type = DWMAC_CORE_GMAC; plat->force_sf_dma_mode = 1; - /* Set default value for multicast hash bins */ + /* Increase the default value for multicast hash bins */ plat->multicast_filter_bins = 256; /* Set default value for unicast filter entries */ diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index 0763ed06715e..a7393a3e792f 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -7567,10 +7567,12 @@ struct plat_stmmacenet_data *stmmac_plat_dat_alloc(struct device *dev) * - phy autodetection * - determine GMII_Address CR field from CSR clock * - allow MTU up to JUMBO_LEN + * - hash table size */ plat_dat->phy_addr = -1; plat_dat->clk_csr = -1; plat_dat->maxmtu = JUMBO_LEN; + plat_dat->multicast_filter_bins = HASH_TABLE_SIZE; return plat_dat; } diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c index b0b4358e0adf..0c65b24480ae 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c @@ -28,9 +28,6 @@ static void common_default_data(struct plat_stmmacenet_data *plat) plat->mdio_bus_data->needs_reset = true; - /* Set default value for multicast hash bins */ - plat->multicast_filter_bins = HASH_TABLE_SIZE; - /* Set default value for unicast filter entries */ plat->unicast_filter_entries = 1; @@ -78,9 +75,6 @@ static int snps_gmac5_default_data(struct pci_dev *pdev, plat->flags |= STMMAC_FLAG_TSO_EN; plat->pmt = 1; - /* Set default value for multicast hash bins */ - plat->multicast_filter_bins = HASH_TABLE_SIZE; - /* Set default value for unicast filter entries */ plat->unicast_filter_entries = 1; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c index 81a599475577..9982aaa19519 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c @@ -508,9 +508,6 @@ stmmac_probe_config_dt(struct platform_device *pdev, u8 *mac) plat->flags |= STMMAC_FLAG_EN_TX_LPI_CLOCKGATING; } - /* Set default value for multicast hash bins */ - plat->multicast_filter_bins = HASH_TABLE_SIZE; - /* Set default value for unicast filter entries */ plat->unicast_filter_entries = 1; From bcb145c696903712a82a1234c6848f49188b9cd3 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Fri, 14 Nov 2025 15:28:49 +0000 Subject: [PATCH 0846/1381] net: stmmac: move initialisation of unicast_filter_entries to stmmac_plat_dat_alloc() Move the default initialisation of plat_dat->unicast_filter_entries to 1 to stmmac_plat_dat_alloc(). This means platform glue only needs to override this if different. Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vJvjF-0000000EVju-3LfS@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c | 6 ------ drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c | 3 --- drivers/net/ethernet/stmicro/stmmac/dwmac-sophgo.c | 1 - drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 2 ++ drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c | 6 ------ drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c | 3 --- 6 files changed, 2 insertions(+), 19 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c index a4ef3ec070e2..de07cca2e625 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c @@ -570,9 +570,6 @@ static void common_default_data(struct plat_stmmacenet_data *plat) plat->mdio_bus_data->needs_reset = true; - /* Set default value for unicast filter entries */ - plat->unicast_filter_entries = 1; - /* Set default number of RX and TX queues to use */ plat->tx_queues_to_use = 1; plat->rx_queues_to_use = 1; @@ -700,9 +697,6 @@ static int intel_mgbe_common_data(struct pci_dev *pdev, plat->ptp_clk_freq_config = intel_mgbe_ptp_clk_freq_config; - /* Set default value for unicast filter entries */ - plat->unicast_filter_entries = 1; - plat->flags |= STMMAC_FLAG_VLAN_FAIL_Q_EN; /* Use the last Rx queue */ diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c index cfdda9e82a19..99b2d2deaceb 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c @@ -98,9 +98,6 @@ static void loongson_default_data(struct pci_dev *pdev, /* Increase the default value for multicast hash bins */ plat->multicast_filter_bins = 256; - /* Set default value for unicast filter entries */ - plat->unicast_filter_entries = 1; - /* Disable Priority config by default */ plat->tx_queues_cfg[0].use_prio = false; plat->rx_queues_cfg[0].use_prio = false; diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-sophgo.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-sophgo.c index 3b7947a7a7ba..24ce17ea35c4 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-sophgo.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-sophgo.c @@ -24,7 +24,6 @@ static int sophgo_sg2044_dwmac_init(struct platform_device *pdev, plat_dat->flags |= STMMAC_FLAG_SPH_DISABLE; plat_dat->set_clk_tx_rate = stmmac_set_clk_tx_rate; plat_dat->multicast_filter_bins = 0; - plat_dat->unicast_filter_entries = 1; return 0; } diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index a7393a3e792f..1d12835d14ce 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -7568,11 +7568,13 @@ struct plat_stmmacenet_data *stmmac_plat_dat_alloc(struct device *dev) * - determine GMII_Address CR field from CSR clock * - allow MTU up to JUMBO_LEN * - hash table size + * - one unicast filter entry */ plat_dat->phy_addr = -1; plat_dat->clk_csr = -1; plat_dat->maxmtu = JUMBO_LEN; plat_dat->multicast_filter_bins = HASH_TABLE_SIZE; + plat_dat->unicast_filter_entries = 1; return plat_dat; } diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c index 0c65b24480ae..aea615e76dac 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c @@ -28,9 +28,6 @@ static void common_default_data(struct plat_stmmacenet_data *plat) plat->mdio_bus_data->needs_reset = true; - /* Set default value for unicast filter entries */ - plat->unicast_filter_entries = 1; - /* Set default number of RX and TX queues to use */ plat->tx_queues_to_use = 1; plat->rx_queues_to_use = 1; @@ -75,9 +72,6 @@ static int snps_gmac5_default_data(struct pci_dev *pdev, plat->flags |= STMMAC_FLAG_TSO_EN; plat->pmt = 1; - /* Set default value for unicast filter entries */ - plat->unicast_filter_entries = 1; - /* Set default number of RX and TX queues to use */ plat->tx_queues_to_use = 4; plat->rx_queues_to_use = 4; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c index 9982aaa19519..314cb3e720fd 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c @@ -508,9 +508,6 @@ stmmac_probe_config_dt(struct platform_device *pdev, u8 *mac) plat->flags |= STMMAC_FLAG_EN_TX_LPI_CLOCKGATING; } - /* Set default value for unicast filter entries */ - plat->unicast_filter_entries = 1; - /* * Currently only the properties needed on SPEAr600 * are provided. All other properties should be added From d5e788e86fe3dcbbe8fcad41663f7756f1eee1a3 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Fri, 14 Nov 2025 15:28:54 +0000 Subject: [PATCH 0847/1381] net: stmmac: move initialisation of queues_to_use to stmmac_plat_dat_alloc() Move the default initialisation of plat_dat->tx_queues_to_use and plat_dat->rx_queues_to_use to 1 to stmmac_plat_dat_alloc(). This means platform glue only needs to override this if different. Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vJvjK-0000000EVk0-3qb2@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../net/ethernet/stmicro/stmmac/dwmac-intel.c | 4 ---- .../ethernet/stmicro/stmmac/dwmac-loongson.c | 2 -- .../net/ethernet/stmicro/stmmac/stmmac_main.c | 4 ++++ .../net/ethernet/stmicro/stmmac/stmmac_pci.c | 4 ---- .../ethernet/stmicro/stmmac/stmmac_platform.c | 17 ++++------------- 5 files changed, 8 insertions(+), 23 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c index de07cca2e625..55f97b2f4e04 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c @@ -570,10 +570,6 @@ static void common_default_data(struct plat_stmmacenet_data *plat) plat->mdio_bus_data->needs_reset = true; - /* Set default number of RX and TX queues to use */ - plat->tx_queues_to_use = 1; - plat->rx_queues_to_use = 1; - /* Disable Priority config by default */ plat->tx_queues_cfg[0].use_prio = false; plat->rx_queues_cfg[0].use_prio = false; diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c index 99b2d2deaceb..ed5e9ca738bf 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c @@ -131,8 +131,6 @@ static void loongson_default_data(struct pci_dev *pdev, break; default: ld->multichan = 0; - plat->tx_queues_to_use = 1; - plat->rx_queues_to_use = 1; break; } } diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index 1d12835d14ce..c7763db011d6 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -7576,6 +7576,10 @@ struct plat_stmmacenet_data *stmmac_plat_dat_alloc(struct device *dev) plat_dat->multicast_filter_bins = HASH_TABLE_SIZE; plat_dat->unicast_filter_entries = 1; + /* Set the mtl defaults */ + plat_dat->tx_queues_to_use = 1; + plat_dat->rx_queues_to_use = 1; + return plat_dat; } EXPORT_SYMBOL_GPL(stmmac_plat_dat_alloc); diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c index aea615e76dac..8c7188ff658b 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c @@ -28,10 +28,6 @@ static void common_default_data(struct plat_stmmacenet_data *plat) plat->mdio_bus_data->needs_reset = true; - /* Set default number of RX and TX queues to use */ - plat->tx_queues_to_use = 1; - plat->rx_queues_to_use = 1; - /* Disable Priority config by default */ plat->tx_queues_cfg[0].use_prio = false; plat->rx_queues_cfg[0].use_prio = false; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c index 314cb3e720fd..e1e23ee0b48e 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c @@ -137,13 +137,6 @@ static int stmmac_mtl_setup(struct platform_device *pdev, u8 queue = 0; int ret = 0; - /* For backwards-compatibility with device trees that don't have any - * snps,mtl-rx-config or snps,mtl-tx-config properties, we fall back - * to one RX and TX queues each. - */ - plat->rx_queues_to_use = 1; - plat->tx_queues_to_use = 1; - /* First Queue must always be in DCB mode. As MTL_QUEUE_DCB = 1 we need * to always set this, otherwise Queue will be classified as AVB * (because MTL_QUEUE_AVB = 0). @@ -162,9 +155,8 @@ static int stmmac_mtl_setup(struct platform_device *pdev, } /* Processing RX queues common config */ - if (of_property_read_u32(rx_node, "snps,rx-queues-to-use", - &plat->rx_queues_to_use)) - plat->rx_queues_to_use = 1; + of_property_read_u32(rx_node, "snps,rx-queues-to-use", + &plat->rx_queues_to_use); if (of_property_read_bool(rx_node, "snps,rx-sched-sp")) plat->rx_sched_algorithm = MTL_RX_ALGORITHM_SP; @@ -221,9 +213,8 @@ static int stmmac_mtl_setup(struct platform_device *pdev, } /* Processing TX queues common config */ - if (of_property_read_u32(tx_node, "snps,tx-queues-to-use", - &plat->tx_queues_to_use)) - plat->tx_queues_to_use = 1; + of_property_read_u32(tx_node, "snps,tx-queues-to-use", + &plat->tx_queues_to_use); if (of_property_read_bool(tx_node, "snps,tx-sched-wrr")) plat->tx_sched_algorithm = MTL_TX_ALGORITHM_WRR; From b6d013b3260be4c82a7ef9b1cb98b874b209a50c Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Fri, 14 Nov 2025 15:29:00 +0000 Subject: [PATCH 0848/1381] net: stmmac: setup default RX channel map in stmmac_plat_dat_alloc() Setup the default 1:1 RX channel map in stmmac_plat_dat_alloc() and remove 1:1 initialisations from platform glue drivers. Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vJvjQ-0000000EVk6-05z7@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c | 1 - drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 5 +++++ drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c | 1 - drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c | 5 ++--- 4 files changed, 7 insertions(+), 5 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c index 55f97b2f4e04..1fd6faa0c70c 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c @@ -618,7 +618,6 @@ static int intel_mgbe_common_data(struct pci_dev *pdev, for (i = 0; i < plat->rx_queues_to_use; i++) { plat->rx_queues_cfg[i].mode_to_use = MTL_QUEUE_DCB; - plat->rx_queues_cfg[i].chan = i; /* Disable Priority config by default */ plat->rx_queues_cfg[i].use_prio = false; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index c7763db011d6..d08ff8f5ff15 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -7558,6 +7558,7 @@ static void stmmac_unregister_devlink(struct stmmac_priv *priv) struct plat_stmmacenet_data *stmmac_plat_dat_alloc(struct device *dev) { struct plat_stmmacenet_data *plat_dat; + int i; plat_dat = devm_kzalloc(dev, sizeof(*plat_dat), GFP_KERNEL); if (!plat_dat) @@ -7580,6 +7581,10 @@ struct plat_stmmacenet_data *stmmac_plat_dat_alloc(struct device *dev) plat_dat->tx_queues_to_use = 1; plat_dat->rx_queues_to_use = 1; + /* Setup the default RX queue channel map */ + for (i = 0; i < ARRAY_SIZE(plat_dat->rx_queues_cfg); i++) + plat_dat->rx_queues_cfg[i].chan = i; + return plat_dat; } EXPORT_SYMBOL_GPL(stmmac_plat_dat_alloc); diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c index 8c7188ff658b..ded44846f74a 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c @@ -86,7 +86,6 @@ static int snps_gmac5_default_data(struct pci_dev *pdev, plat->rx_queues_cfg[i].use_prio = false; plat->rx_queues_cfg[i].mode_to_use = MTL_QUEUE_DCB; plat->rx_queues_cfg[i].pkt_route = 0x0; - plat->rx_queues_cfg[i].chan = i; } plat->bus_id = 1; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c index e1e23ee0b48e..7eb22511acf5 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c @@ -177,9 +177,8 @@ static int stmmac_mtl_setup(struct platform_device *pdev, else plat->rx_queues_cfg[queue].mode_to_use = MTL_QUEUE_DCB; - if (of_property_read_u32(q_node, "snps,map-to-dma-channel", - &plat->rx_queues_cfg[queue].chan)) - plat->rx_queues_cfg[queue].chan = queue; + of_property_read_u32(q_node, "snps,map-to-dma-channel", + &plat->rx_queues_cfg[queue].chan); /* TODO: Dynamic mapping to be included in the future */ if (of_property_read_u32(q_node, "snps,priority", From c03101cb1bf020fb6897864afe2a63f54c65b778 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Fri, 14 Nov 2025 15:29:05 +0000 Subject: [PATCH 0849/1381] net: stmmac: remove unnecessary .use_prio queue initialisation Several drivers (see below) explicitly set the queue .use_prio configuration to false. However, as this structure is allocated using devm_kzalloc(), all members default to zero unless otherwise explicitly initialised. .use_prio isn't, so defaults to false. Remove these unnecessary initialisations, leaving stmmac_platform.c as the only file that .use_prio is set true. $ grep 'use_prio =' *.c dwmac-intel.c: plat->tx_queues_cfg[0].use_prio = false; dwmac-intel.c: plat->rx_queues_cfg[0].use_prio = false; dwmac-intel.c: plat->rx_queues_cfg[i].use_prio = false; dwmac-intel.c: plat->tx_queues_cfg[i].use_prio = false; dwmac-loongson.c: plat->tx_queues_cfg[0].use_prio = false; dwmac-loongson.c: plat->rx_queues_cfg[0].use_prio = false; stmmac_pci.c: plat->tx_queues_cfg[0].use_prio = false; stmmac_pci.c: plat->rx_queues_cfg[0].use_prio = false; stmmac_pci.c: plat->tx_queues_cfg[i].use_prio = false; stmmac_pci.c: plat->rx_queues_cfg[i].use_prio = false; stmmac_platform.c: plat->rx_queues_cfg[queue].use_prio = false; stmmac_platform.c: plat->rx_queues_cfg[queue].use_prio = true; stmmac_platform.c: plat->tx_queues_cfg[queue].use_prio = false; stmmac_platform.c: plat->tx_queues_cfg[queue].use_prio = true; Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vJvjV-0000000EVkC-0WAV@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c | 9 --------- drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c | 4 ---- drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c | 6 ------ drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c | 2 -- 4 files changed, 21 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c index 1fd6faa0c70c..72f6acde544f 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c @@ -570,10 +570,6 @@ static void common_default_data(struct plat_stmmacenet_data *plat) plat->mdio_bus_data->needs_reset = true; - /* Disable Priority config by default */ - plat->tx_queues_cfg[0].use_prio = false; - plat->rx_queues_cfg[0].use_prio = false; - /* Disable RX queues routing by default */ plat->rx_queues_cfg[0].pkt_route = 0x0; } @@ -619,9 +615,6 @@ static int intel_mgbe_common_data(struct pci_dev *pdev, for (i = 0; i < plat->rx_queues_to_use; i++) { plat->rx_queues_cfg[i].mode_to_use = MTL_QUEUE_DCB; - /* Disable Priority config by default */ - plat->rx_queues_cfg[i].use_prio = false; - /* Disable RX queues routing by default */ plat->rx_queues_cfg[i].pkt_route = 0x0; } @@ -629,8 +622,6 @@ static int intel_mgbe_common_data(struct pci_dev *pdev, for (i = 0; i < plat->tx_queues_to_use; i++) { plat->tx_queues_cfg[i].mode_to_use = MTL_QUEUE_DCB; - /* Disable Priority config by default */ - plat->tx_queues_cfg[i].use_prio = false; /* Default TX Q0 to use TSO and rest TXQ for TBS */ if (i > 0) plat->tx_queues_cfg[i].tbs_en = 1; diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c index ed5e9ca738bf..c64a24bb060f 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c @@ -98,10 +98,6 @@ static void loongson_default_data(struct pci_dev *pdev, /* Increase the default value for multicast hash bins */ plat->multicast_filter_bins = 256; - /* Disable Priority config by default */ - plat->tx_queues_cfg[0].use_prio = false; - plat->rx_queues_cfg[0].use_prio = false; - /* Disable RX queues routing by default */ plat->rx_queues_cfg[0].pkt_route = 0x0; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c index ded44846f74a..2f45b7986903 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c @@ -28,10 +28,6 @@ static void common_default_data(struct plat_stmmacenet_data *plat) plat->mdio_bus_data->needs_reset = true; - /* Disable Priority config by default */ - plat->tx_queues_cfg[0].use_prio = false; - plat->rx_queues_cfg[0].use_prio = false; - /* Disable RX queues routing by default */ plat->rx_queues_cfg[0].pkt_route = 0x0; } @@ -74,7 +70,6 @@ static int snps_gmac5_default_data(struct pci_dev *pdev, plat->tx_sched_algorithm = MTL_TX_ALGORITHM_WRR; for (i = 0; i < plat->tx_queues_to_use; i++) { - plat->tx_queues_cfg[i].use_prio = false; plat->tx_queues_cfg[i].mode_to_use = MTL_QUEUE_DCB; plat->tx_queues_cfg[i].weight = 25; if (i > 0) @@ -83,7 +78,6 @@ static int snps_gmac5_default_data(struct pci_dev *pdev, plat->rx_sched_algorithm = MTL_RX_ALGORITHM_SP; for (i = 0; i < plat->rx_queues_to_use; i++) { - plat->rx_queues_cfg[i].use_prio = false; plat->rx_queues_cfg[i].mode_to_use = MTL_QUEUE_DCB; plat->rx_queues_cfg[i].pkt_route = 0x0; } diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c index 7eb22511acf5..4750843cf102 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c @@ -184,7 +184,6 @@ static int stmmac_mtl_setup(struct platform_device *pdev, if (of_property_read_u32(q_node, "snps,priority", &plat->rx_queues_cfg[queue].prio)) { plat->rx_queues_cfg[queue].prio = 0; - plat->rx_queues_cfg[queue].use_prio = false; } else { plat->rx_queues_cfg[queue].use_prio = true; } @@ -261,7 +260,6 @@ static int stmmac_mtl_setup(struct platform_device *pdev, if (of_property_read_u32(q_node, "snps,priority", &plat->tx_queues_cfg[queue].prio)) { plat->tx_queues_cfg[queue].prio = 0; - plat->tx_queues_cfg[queue].use_prio = false; } else { plat->tx_queues_cfg[queue].use_prio = true; } From 0a20999ed4521cd18348a3ca0a55f3406f879f7a Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Fri, 14 Nov 2025 15:29:10 +0000 Subject: [PATCH 0850/1381] net: stmmac: remove unnecessary .prio queue initialisation stmmac_platform.c explicitly sets .prio to zero if the snps,priority property is not present in DT for the queue. However, as the struct is allocated using devm_kzalloc(), all members default to zero unless explicitly initialised, and of_property_read_u32() will not write to its argument if the property is not found. Thus, explicitly setting these to zero is unnecessary. Remove these. $ grep '\.prio =' *.c stmmac_platform.c: plat->rx_queues_cfg[queue].prio = 0; stmmac_platform.c: plat->tx_queues_cfg[queue].prio = 0; Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vJvja-0000000EVkI-0zUH@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../net/ethernet/stmicro/stmmac/stmmac_platform.c | 14 ++++---------- 1 file changed, 4 insertions(+), 10 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c index 4750843cf102..e769638586fe 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c @@ -181,12 +181,9 @@ static int stmmac_mtl_setup(struct platform_device *pdev, &plat->rx_queues_cfg[queue].chan); /* TODO: Dynamic mapping to be included in the future */ - if (of_property_read_u32(q_node, "snps,priority", - &plat->rx_queues_cfg[queue].prio)) { - plat->rx_queues_cfg[queue].prio = 0; - } else { + if (!of_property_read_u32(q_node, "snps,priority", + &plat->rx_queues_cfg[queue].prio)) plat->rx_queues_cfg[queue].use_prio = true; - } /* RX queue specific packet type routing */ if (of_property_read_bool(q_node, "snps,route-avcp")) @@ -257,12 +254,9 @@ static int stmmac_mtl_setup(struct platform_device *pdev, plat->tx_queues_cfg[queue].mode_to_use = MTL_QUEUE_DCB; } - if (of_property_read_u32(q_node, "snps,priority", - &plat->tx_queues_cfg[queue].prio)) { - plat->tx_queues_cfg[queue].prio = 0; - } else { + if (!of_property_read_u32(q_node, "snps,priority", + &plat->tx_queues_cfg[queue].prio)) plat->tx_queues_cfg[queue].use_prio = true; - } plat->tx_queues_cfg[queue].coe_unsupported = of_property_read_bool(q_node, "snps,coe-unsupported"); From 6409249ccc152a3170cdd14f0f660d1d70046601 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Fri, 14 Nov 2025 15:29:15 +0000 Subject: [PATCH 0851/1381] net: stmmac: remove unnecessary .pkt_route queue initialisation PCI drivers explicitly set .pkt_route to zero. However, as the struct is allocated using devm_kzalloc(), all members default to zero unless explicitly initialised. Thus, explicitly setting these to zero is unnecessary. Remove these. This leaves only stmmac_platform.c where this is explicitly initialised depending on DT properties. $ grep '\.pkt_route =' *.c dwmac-intel.c: plat->rx_queues_cfg[0].pkt_route = 0x0; dwmac-intel.c: plat->rx_queues_cfg[i].pkt_route = 0x0; dwmac-loongson.c: plat->rx_queues_cfg[0].pkt_route = 0x0; stmmac_main.c: if (priv->plat->rx_queues_cfg[queue].pkt_route == 0x0) stmmac_pci.c: plat->rx_queues_cfg[0].pkt_route = 0x0; stmmac_pci.c: plat->rx_queues_cfg[i].pkt_route = 0x0; stmmac_platform.c: plat->rx_queues_cfg[queue].pkt_route = PACKET_AVCPQ; stmmac_platform.c: plat->rx_queues_cfg[queue].pkt_route = PACKET_PTPQ; stmmac_platform.c: plat->rx_queues_cfg[queue].pkt_route = PACKET_DCBCPQ; stmmac_platform.c: plat->rx_queues_cfg[queue].pkt_route = PACKET_UPQ; stmmac_platform.c: plat->rx_queues_cfg[queue].pkt_route = PACKET_MCBCQ; stmmac_platform.c: plat->rx_queues_cfg[queue].pkt_route = 0x0; Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vJvjf-0000000EVkO-1ZaO@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c | 9 +-------- drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c | 3 --- drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c | 7 +------ drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c | 2 -- 4 files changed, 2 insertions(+), 19 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c index 72f6acde544f..8938e7a59925 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c @@ -569,9 +569,6 @@ static void common_default_data(struct plat_stmmacenet_data *plat) plat->force_sf_dma_mode = 1; plat->mdio_bus_data->needs_reset = true; - - /* Disable RX queues routing by default */ - plat->rx_queues_cfg[0].pkt_route = 0x0; } static struct phylink_pcs *intel_mgbe_select_pcs(struct stmmac_priv *priv, @@ -612,13 +609,9 @@ static int intel_mgbe_common_data(struct pci_dev *pdev, plat->rx_sched_algorithm = MTL_RX_ALGORITHM_SP; - for (i = 0; i < plat->rx_queues_to_use; i++) { + for (i = 0; i < plat->rx_queues_to_use; i++) plat->rx_queues_cfg[i].mode_to_use = MTL_QUEUE_DCB; - /* Disable RX queues routing by default */ - plat->rx_queues_cfg[i].pkt_route = 0x0; - } - for (i = 0; i < plat->tx_queues_to_use; i++) { plat->tx_queues_cfg[i].mode_to_use = MTL_QUEUE_DCB; diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c index c64a24bb060f..5f9472f47e35 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c @@ -98,9 +98,6 @@ static void loongson_default_data(struct pci_dev *pdev, /* Increase the default value for multicast hash bins */ plat->multicast_filter_bins = 256; - /* Disable RX queues routing by default */ - plat->rx_queues_cfg[0].pkt_route = 0x0; - plat->clk_ref_rate = 125000000; plat->clk_ptp_rate = 125000000; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c index 2f45b7986903..2fd4660838bb 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c @@ -27,9 +27,6 @@ static void common_default_data(struct plat_stmmacenet_data *plat) plat->force_sf_dma_mode = 1; plat->mdio_bus_data->needs_reset = true; - - /* Disable RX queues routing by default */ - plat->rx_queues_cfg[0].pkt_route = 0x0; } static int stmmac_default_data(struct pci_dev *pdev, @@ -77,10 +74,8 @@ static int snps_gmac5_default_data(struct pci_dev *pdev, } plat->rx_sched_algorithm = MTL_RX_ALGORITHM_SP; - for (i = 0; i < plat->rx_queues_to_use; i++) { + for (i = 0; i < plat->rx_queues_to_use; i++) plat->rx_queues_cfg[i].mode_to_use = MTL_QUEUE_DCB; - plat->rx_queues_cfg[i].pkt_route = 0x0; - } plat->bus_id = 1; plat->phy_interface = PHY_INTERFACE_MODE_GMII; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c index e769638586fe..1fefa6c55db1 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c @@ -196,8 +196,6 @@ static int stmmac_mtl_setup(struct platform_device *pdev, plat->rx_queues_cfg[queue].pkt_route = PACKET_UPQ; else if (of_property_read_bool(q_node, "snps,route-multi-broad")) plat->rx_queues_cfg[queue].pkt_route = PACKET_MCBCQ; - else - plat->rx_queues_cfg[queue].pkt_route = 0x0; queue++; } From e275d9091c01b3b46f3ec534ce4ac77cffc9e3ae Mon Sep 17 00:00:00 2001 From: Erni Sri Satya Vennela Date: Fri, 14 Nov 2025 03:43:18 -0800 Subject: [PATCH 0852/1381] net: mana: Move hardware counter stats from per-port to per-VF context Move hardware counter (HC) statistics from mana_port_context to mana_context to enable sharing stats across multiple network ports on the same MANA VF. Previously, each network port queried hardware counters independently using MANA_QUERY_GF_STAT command (GF = Generic Function stats from GDMA hardware), resulting in redundant queries when multiple ports existed on the same device. Isolate hardware counter stats by introducing mana_ethtool_hc_stats in mana_context and update the code to ensure all stats are properly reported via ethtool -S , maintaining consistency with previous behavior. Signed-off-by: Erni Sri Satya Vennela Reviewed-by: Haiyang Zhang Link: https://patch.msgid.link/1763120599-6331-2-git-send-email-ernis@linux.microsoft.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/microsoft/mana/mana_en.c | 67 +++++----- .../ethernet/microsoft/mana/mana_ethtool.c | 121 ++++++++++-------- include/net/mana/mana.h | 14 +- 3 files changed, 108 insertions(+), 94 deletions(-) diff --git a/drivers/net/ethernet/microsoft/mana/mana_en.c b/drivers/net/ethernet/microsoft/mana/mana_en.c index cccd5b63cee6..d8ce4402c696 100644 --- a/drivers/net/ethernet/microsoft/mana/mana_en.c +++ b/drivers/net/ethernet/microsoft/mana/mana_en.c @@ -2809,11 +2809,12 @@ int mana_config_rss(struct mana_port_context *apc, enum TRI_STATE rx, return 0; } -void mana_query_gf_stats(struct mana_port_context *apc) +void mana_query_gf_stats(struct mana_context *ac) { + struct gdma_context *gc = ac->gdma_dev->gdma_context; struct mana_query_gf_stat_resp resp = {}; struct mana_query_gf_stat_req req = {}; - struct net_device *ndev = apc->ndev; + struct device *dev = gc->dev; int err; mana_gd_init_req_hdr(&req.hdr, MANA_QUERY_GF_STAT, @@ -2847,52 +2848,52 @@ void mana_query_gf_stats(struct mana_port_context *apc) STATISTICS_FLAGS_HC_TX_BCAST_BYTES | STATISTICS_FLAGS_TX_ERRORS_GDMA_ERROR; - err = mana_send_request(apc->ac, &req, sizeof(req), &resp, + err = mana_send_request(ac, &req, sizeof(req), &resp, sizeof(resp)); if (err) { - netdev_err(ndev, "Failed to query GF stats: %d\n", err); + dev_err(dev, "Failed to query GF stats: %d\n", err); return; } err = mana_verify_resp_hdr(&resp.hdr, MANA_QUERY_GF_STAT, sizeof(resp)); if (err || resp.hdr.status) { - netdev_err(ndev, "Failed to query GF stats: %d, 0x%x\n", err, - resp.hdr.status); + dev_err(dev, "Failed to query GF stats: %d, 0x%x\n", err, + resp.hdr.status); return; } - apc->eth_stats.hc_rx_discards_no_wqe = resp.rx_discards_nowqe; - apc->eth_stats.hc_rx_err_vport_disabled = resp.rx_err_vport_disabled; - apc->eth_stats.hc_rx_bytes = resp.hc_rx_bytes; - apc->eth_stats.hc_rx_ucast_pkts = resp.hc_rx_ucast_pkts; - apc->eth_stats.hc_rx_ucast_bytes = resp.hc_rx_ucast_bytes; - apc->eth_stats.hc_rx_bcast_pkts = resp.hc_rx_bcast_pkts; - apc->eth_stats.hc_rx_bcast_bytes = resp.hc_rx_bcast_bytes; - apc->eth_stats.hc_rx_mcast_pkts = resp.hc_rx_mcast_pkts; - apc->eth_stats.hc_rx_mcast_bytes = resp.hc_rx_mcast_bytes; - apc->eth_stats.hc_tx_err_gf_disabled = resp.tx_err_gf_disabled; - apc->eth_stats.hc_tx_err_vport_disabled = resp.tx_err_vport_disabled; - apc->eth_stats.hc_tx_err_inval_vportoffset_pkt = + ac->hc_stats.hc_rx_discards_no_wqe = resp.rx_discards_nowqe; + ac->hc_stats.hc_rx_err_vport_disabled = resp.rx_err_vport_disabled; + ac->hc_stats.hc_rx_bytes = resp.hc_rx_bytes; + ac->hc_stats.hc_rx_ucast_pkts = resp.hc_rx_ucast_pkts; + ac->hc_stats.hc_rx_ucast_bytes = resp.hc_rx_ucast_bytes; + ac->hc_stats.hc_rx_bcast_pkts = resp.hc_rx_bcast_pkts; + ac->hc_stats.hc_rx_bcast_bytes = resp.hc_rx_bcast_bytes; + ac->hc_stats.hc_rx_mcast_pkts = resp.hc_rx_mcast_pkts; + ac->hc_stats.hc_rx_mcast_bytes = resp.hc_rx_mcast_bytes; + ac->hc_stats.hc_tx_err_gf_disabled = resp.tx_err_gf_disabled; + ac->hc_stats.hc_tx_err_vport_disabled = resp.tx_err_vport_disabled; + ac->hc_stats.hc_tx_err_inval_vportoffset_pkt = resp.tx_err_inval_vport_offset_pkt; - apc->eth_stats.hc_tx_err_vlan_enforcement = + ac->hc_stats.hc_tx_err_vlan_enforcement = resp.tx_err_vlan_enforcement; - apc->eth_stats.hc_tx_err_eth_type_enforcement = + ac->hc_stats.hc_tx_err_eth_type_enforcement = resp.tx_err_ethtype_enforcement; - apc->eth_stats.hc_tx_err_sa_enforcement = resp.tx_err_SA_enforcement; - apc->eth_stats.hc_tx_err_sqpdid_enforcement = + ac->hc_stats.hc_tx_err_sa_enforcement = resp.tx_err_SA_enforcement; + ac->hc_stats.hc_tx_err_sqpdid_enforcement = resp.tx_err_SQPDID_enforcement; - apc->eth_stats.hc_tx_err_cqpdid_enforcement = + ac->hc_stats.hc_tx_err_cqpdid_enforcement = resp.tx_err_CQPDID_enforcement; - apc->eth_stats.hc_tx_err_mtu_violation = resp.tx_err_mtu_violation; - apc->eth_stats.hc_tx_err_inval_oob = resp.tx_err_inval_oob; - apc->eth_stats.hc_tx_bytes = resp.hc_tx_bytes; - apc->eth_stats.hc_tx_ucast_pkts = resp.hc_tx_ucast_pkts; - apc->eth_stats.hc_tx_ucast_bytes = resp.hc_tx_ucast_bytes; - apc->eth_stats.hc_tx_bcast_pkts = resp.hc_tx_bcast_pkts; - apc->eth_stats.hc_tx_bcast_bytes = resp.hc_tx_bcast_bytes; - apc->eth_stats.hc_tx_mcast_pkts = resp.hc_tx_mcast_pkts; - apc->eth_stats.hc_tx_mcast_bytes = resp.hc_tx_mcast_bytes; - apc->eth_stats.hc_tx_err_gdma = resp.tx_err_gdma; + ac->hc_stats.hc_tx_err_mtu_violation = resp.tx_err_mtu_violation; + ac->hc_stats.hc_tx_err_inval_oob = resp.tx_err_inval_oob; + ac->hc_stats.hc_tx_bytes = resp.hc_tx_bytes; + ac->hc_stats.hc_tx_ucast_pkts = resp.hc_tx_ucast_pkts; + ac->hc_stats.hc_tx_ucast_bytes = resp.hc_tx_ucast_bytes; + ac->hc_stats.hc_tx_bcast_pkts = resp.hc_tx_bcast_pkts; + ac->hc_stats.hc_tx_bcast_bytes = resp.hc_tx_bcast_bytes; + ac->hc_stats.hc_tx_mcast_pkts = resp.hc_tx_mcast_pkts; + ac->hc_stats.hc_tx_mcast_bytes = resp.hc_tx_mcast_bytes; + ac->hc_stats.hc_tx_err_gdma = resp.tx_err_gdma; } void mana_query_phy_stats(struct mana_port_context *apc) diff --git a/drivers/net/ethernet/microsoft/mana/mana_ethtool.c b/drivers/net/ethernet/microsoft/mana/mana_ethtool.c index a1afa75a9463..3dfd96146424 100644 --- a/drivers/net/ethernet/microsoft/mana/mana_ethtool.c +++ b/drivers/net/ethernet/microsoft/mana/mana_ethtool.c @@ -15,59 +15,6 @@ struct mana_stats_desc { static const struct mana_stats_desc mana_eth_stats[] = { {"stop_queue", offsetof(struct mana_ethtool_stats, stop_queue)}, {"wake_queue", offsetof(struct mana_ethtool_stats, wake_queue)}, - {"hc_rx_discards_no_wqe", offsetof(struct mana_ethtool_stats, - hc_rx_discards_no_wqe)}, - {"hc_rx_err_vport_disabled", offsetof(struct mana_ethtool_stats, - hc_rx_err_vport_disabled)}, - {"hc_rx_bytes", offsetof(struct mana_ethtool_stats, hc_rx_bytes)}, - {"hc_rx_ucast_pkts", offsetof(struct mana_ethtool_stats, - hc_rx_ucast_pkts)}, - {"hc_rx_ucast_bytes", offsetof(struct mana_ethtool_stats, - hc_rx_ucast_bytes)}, - {"hc_rx_bcast_pkts", offsetof(struct mana_ethtool_stats, - hc_rx_bcast_pkts)}, - {"hc_rx_bcast_bytes", offsetof(struct mana_ethtool_stats, - hc_rx_bcast_bytes)}, - {"hc_rx_mcast_pkts", offsetof(struct mana_ethtool_stats, - hc_rx_mcast_pkts)}, - {"hc_rx_mcast_bytes", offsetof(struct mana_ethtool_stats, - hc_rx_mcast_bytes)}, - {"hc_tx_err_gf_disabled", offsetof(struct mana_ethtool_stats, - hc_tx_err_gf_disabled)}, - {"hc_tx_err_vport_disabled", offsetof(struct mana_ethtool_stats, - hc_tx_err_vport_disabled)}, - {"hc_tx_err_inval_vportoffset_pkt", - offsetof(struct mana_ethtool_stats, - hc_tx_err_inval_vportoffset_pkt)}, - {"hc_tx_err_vlan_enforcement", offsetof(struct mana_ethtool_stats, - hc_tx_err_vlan_enforcement)}, - {"hc_tx_err_eth_type_enforcement", - offsetof(struct mana_ethtool_stats, hc_tx_err_eth_type_enforcement)}, - {"hc_tx_err_sa_enforcement", offsetof(struct mana_ethtool_stats, - hc_tx_err_sa_enforcement)}, - {"hc_tx_err_sqpdid_enforcement", - offsetof(struct mana_ethtool_stats, hc_tx_err_sqpdid_enforcement)}, - {"hc_tx_err_cqpdid_enforcement", - offsetof(struct mana_ethtool_stats, hc_tx_err_cqpdid_enforcement)}, - {"hc_tx_err_mtu_violation", offsetof(struct mana_ethtool_stats, - hc_tx_err_mtu_violation)}, - {"hc_tx_err_inval_oob", offsetof(struct mana_ethtool_stats, - hc_tx_err_inval_oob)}, - {"hc_tx_err_gdma", offsetof(struct mana_ethtool_stats, - hc_tx_err_gdma)}, - {"hc_tx_bytes", offsetof(struct mana_ethtool_stats, hc_tx_bytes)}, - {"hc_tx_ucast_pkts", offsetof(struct mana_ethtool_stats, - hc_tx_ucast_pkts)}, - {"hc_tx_ucast_bytes", offsetof(struct mana_ethtool_stats, - hc_tx_ucast_bytes)}, - {"hc_tx_bcast_pkts", offsetof(struct mana_ethtool_stats, - hc_tx_bcast_pkts)}, - {"hc_tx_bcast_bytes", offsetof(struct mana_ethtool_stats, - hc_tx_bcast_bytes)}, - {"hc_tx_mcast_pkts", offsetof(struct mana_ethtool_stats, - hc_tx_mcast_pkts)}, - {"hc_tx_mcast_bytes", offsetof(struct mana_ethtool_stats, - hc_tx_mcast_bytes)}, {"tx_cq_err", offsetof(struct mana_ethtool_stats, tx_cqe_err)}, {"tx_cqe_unknown_type", offsetof(struct mana_ethtool_stats, tx_cqe_unknown_type)}, @@ -77,6 +24,62 @@ static const struct mana_stats_desc mana_eth_stats[] = { rx_cqe_unknown_type)}, }; +static const struct mana_stats_desc mana_hc_stats[] = { + {"hc_rx_discards_no_wqe", offsetof(struct mana_ethtool_hc_stats, + hc_rx_discards_no_wqe)}, + {"hc_rx_err_vport_disabled", offsetof(struct mana_ethtool_hc_stats, + hc_rx_err_vport_disabled)}, + {"hc_rx_bytes", offsetof(struct mana_ethtool_hc_stats, hc_rx_bytes)}, + {"hc_rx_ucast_pkts", offsetof(struct mana_ethtool_hc_stats, + hc_rx_ucast_pkts)}, + {"hc_rx_ucast_bytes", offsetof(struct mana_ethtool_hc_stats, + hc_rx_ucast_bytes)}, + {"hc_rx_bcast_pkts", offsetof(struct mana_ethtool_hc_stats, + hc_rx_bcast_pkts)}, + {"hc_rx_bcast_bytes", offsetof(struct mana_ethtool_hc_stats, + hc_rx_bcast_bytes)}, + {"hc_rx_mcast_pkts", offsetof(struct mana_ethtool_hc_stats, + hc_rx_mcast_pkts)}, + {"hc_rx_mcast_bytes", offsetof(struct mana_ethtool_hc_stats, + hc_rx_mcast_bytes)}, + {"hc_tx_err_gf_disabled", offsetof(struct mana_ethtool_hc_stats, + hc_tx_err_gf_disabled)}, + {"hc_tx_err_vport_disabled", offsetof(struct mana_ethtool_hc_stats, + hc_tx_err_vport_disabled)}, + {"hc_tx_err_inval_vportoffset_pkt", + offsetof(struct mana_ethtool_hc_stats, + hc_tx_err_inval_vportoffset_pkt)}, + {"hc_tx_err_vlan_enforcement", offsetof(struct mana_ethtool_hc_stats, + hc_tx_err_vlan_enforcement)}, + {"hc_tx_err_eth_type_enforcement", + offsetof(struct mana_ethtool_hc_stats, hc_tx_err_eth_type_enforcement)}, + {"hc_tx_err_sa_enforcement", offsetof(struct mana_ethtool_hc_stats, + hc_tx_err_sa_enforcement)}, + {"hc_tx_err_sqpdid_enforcement", + offsetof(struct mana_ethtool_hc_stats, hc_tx_err_sqpdid_enforcement)}, + {"hc_tx_err_cqpdid_enforcement", + offsetof(struct mana_ethtool_hc_stats, hc_tx_err_cqpdid_enforcement)}, + {"hc_tx_err_mtu_violation", offsetof(struct mana_ethtool_hc_stats, + hc_tx_err_mtu_violation)}, + {"hc_tx_err_inval_oob", offsetof(struct mana_ethtool_hc_stats, + hc_tx_err_inval_oob)}, + {"hc_tx_err_gdma", offsetof(struct mana_ethtool_hc_stats, + hc_tx_err_gdma)}, + {"hc_tx_bytes", offsetof(struct mana_ethtool_hc_stats, hc_tx_bytes)}, + {"hc_tx_ucast_pkts", offsetof(struct mana_ethtool_hc_stats, + hc_tx_ucast_pkts)}, + {"hc_tx_ucast_bytes", offsetof(struct mana_ethtool_hc_stats, + hc_tx_ucast_bytes)}, + {"hc_tx_bcast_pkts", offsetof(struct mana_ethtool_hc_stats, + hc_tx_bcast_pkts)}, + {"hc_tx_bcast_bytes", offsetof(struct mana_ethtool_hc_stats, + hc_tx_bcast_bytes)}, + {"hc_tx_mcast_pkts", offsetof(struct mana_ethtool_hc_stats, + hc_tx_mcast_pkts)}, + {"hc_tx_mcast_bytes", offsetof(struct mana_ethtool_hc_stats, + hc_tx_mcast_bytes)}, +}; + static const struct mana_stats_desc mana_phy_stats[] = { { "hc_rx_pkt_drop_phy", offsetof(struct mana_ethtool_phy_stats, rx_pkt_drop_phy) }, { "hc_tx_pkt_drop_phy", offsetof(struct mana_ethtool_phy_stats, tx_pkt_drop_phy) }, @@ -138,7 +141,7 @@ static int mana_get_sset_count(struct net_device *ndev, int stringset) if (stringset != ETH_SS_STATS) return -EINVAL; - return ARRAY_SIZE(mana_eth_stats) + ARRAY_SIZE(mana_phy_stats) + + return ARRAY_SIZE(mana_eth_stats) + ARRAY_SIZE(mana_phy_stats) + ARRAY_SIZE(mana_hc_stats) + num_queues * (MANA_STATS_RX_COUNT + MANA_STATS_TX_COUNT); } @@ -150,10 +153,12 @@ static void mana_get_strings(struct net_device *ndev, u32 stringset, u8 *data) if (stringset != ETH_SS_STATS) return; - for (i = 0; i < ARRAY_SIZE(mana_eth_stats); i++) ethtool_puts(&data, mana_eth_stats[i].name); + for (i = 0; i < ARRAY_SIZE(mana_hc_stats); i++) + ethtool_puts(&data, mana_hc_stats[i].name); + for (i = 0; i < ARRAY_SIZE(mana_phy_stats); i++) ethtool_puts(&data, mana_phy_stats[i].name); @@ -186,6 +191,7 @@ static void mana_get_ethtool_stats(struct net_device *ndev, struct mana_port_context *apc = netdev_priv(ndev); unsigned int num_queues = apc->num_queues; void *eth_stats = &apc->eth_stats; + void *hc_stats = &apc->ac->hc_stats; void *phy_stats = &apc->phy_stats; struct mana_stats_rx *rx_stats; struct mana_stats_tx *tx_stats; @@ -208,7 +214,7 @@ static void mana_get_ethtool_stats(struct net_device *ndev, if (!apc->port_is_up) return; /* we call mana function to update stats from GDMA */ - mana_query_gf_stats(apc); + mana_query_gf_stats(apc->ac); /* We call this mana function to get the phy stats from GDMA and includes * aggregate tx/rx drop counters, Per-TC(Traffic Channel) tx/rx and pause @@ -219,6 +225,9 @@ static void mana_get_ethtool_stats(struct net_device *ndev, for (q = 0; q < ARRAY_SIZE(mana_eth_stats); q++) data[i++] = *(u64 *)(eth_stats + mana_eth_stats[q].offset); + for (q = 0; q < ARRAY_SIZE(mana_hc_stats); q++) + data[i++] = *(u64 *)(hc_stats + mana_hc_stats[q].offset); + for (q = 0; q < ARRAY_SIZE(mana_phy_stats); q++) data[i++] = *(u64 *)(phy_stats + mana_phy_stats[q].offset); diff --git a/include/net/mana/mana.h b/include/net/mana/mana.h index 8906901535f5..3484f42803e3 100644 --- a/include/net/mana/mana.h +++ b/include/net/mana/mana.h @@ -375,6 +375,13 @@ struct mana_tx_qp { struct mana_ethtool_stats { u64 stop_queue; u64 wake_queue; + u64 tx_cqe_err; + u64 tx_cqe_unknown_type; + u64 rx_coalesced_err; + u64 rx_cqe_unknown_type; +}; + +struct mana_ethtool_hc_stats { u64 hc_rx_discards_no_wqe; u64 hc_rx_err_vport_disabled; u64 hc_rx_bytes; @@ -402,10 +409,6 @@ struct mana_ethtool_stats { u64 hc_tx_mcast_pkts; u64 hc_tx_mcast_bytes; u64 hc_tx_err_gdma; - u64 tx_cqe_err; - u64 tx_cqe_unknown_type; - u64 rx_coalesced_err; - u64 rx_cqe_unknown_type; }; struct mana_ethtool_phy_stats { @@ -473,6 +476,7 @@ struct mana_context { u16 num_ports; u8 bm_hostmode; + struct mana_ethtool_hc_stats hc_stats; struct mana_eq *eqs; struct dentry *mana_eqs_debugfs; @@ -577,7 +581,7 @@ u32 mana_run_xdp(struct net_device *ndev, struct mana_rxq *rxq, struct bpf_prog *mana_xdp_get(struct mana_port_context *apc); void mana_chn_setxdp(struct mana_port_context *apc, struct bpf_prog *prog); int mana_bpf(struct net_device *ndev, struct netdev_bpf *bpf); -void mana_query_gf_stats(struct mana_port_context *apc); +void mana_query_gf_stats(struct mana_context *ac); int mana_query_link_cfg(struct mana_port_context *apc); int mana_set_bw_clamp(struct mana_port_context *apc, u32 speed, int enable_clamping); From be4f1d67ec56f23f37714ac73c01094e63c7ff28 Mon Sep 17 00:00:00 2001 From: Erni Sri Satya Vennela Date: Fri, 14 Nov 2025 03:43:19 -0800 Subject: [PATCH 0853/1381] net: mana: Add standard counter rx_missed_errors Report standard counter stats->rx_missed_errors using hc_rx_discards_no_wqe from the hardware. Add a global workqueue to periodically run mana_query_gf_stats every 2 seconds to get the latest info in eth_stats and define a driver capability flag to notify hardware of the periodic queries. To avoid repeated failures and log flooding, the workqueue is not rescheduled if mana_query_gf_stats fails on HWC timeout error and the stats are reset to 0. Other errors are transient which will not need a VF reset for recovery. Signed-off-by: Erni Sri Satya Vennela Reviewed-by: Haiyang Zhang Link: https://patch.msgid.link/1763120599-6331-3-git-send-email-ernis@linux.microsoft.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/microsoft/mana/mana_en.c | 36 +++++++++++++++++-- .../ethernet/microsoft/mana/mana_ethtool.c | 2 -- include/net/mana/gdma.h | 6 +++- include/net/mana/mana.h | 6 +++- 4 files changed, 43 insertions(+), 7 deletions(-) diff --git a/drivers/net/ethernet/microsoft/mana/mana_en.c b/drivers/net/ethernet/microsoft/mana/mana_en.c index d8ce4402c696..13f47be7aca6 100644 --- a/drivers/net/ethernet/microsoft/mana/mana_en.c +++ b/drivers/net/ethernet/microsoft/mana/mana_en.c @@ -534,6 +534,11 @@ static void mana_get_stats64(struct net_device *ndev, netdev_stats_to_stats64(st, &ndev->stats); + if (apc->ac->hwc_timeout_occurred) + netdev_warn_once(ndev, "HWC timeout occurred\n"); + + st->rx_missed_errors = apc->ac->hc_stats.hc_rx_discards_no_wqe; + for (q = 0; q < num_queues; q++) { rx_stats = &apc->rxqs[q]->stats; @@ -2809,7 +2814,7 @@ int mana_config_rss(struct mana_port_context *apc, enum TRI_STATE rx, return 0; } -void mana_query_gf_stats(struct mana_context *ac) +int mana_query_gf_stats(struct mana_context *ac) { struct gdma_context *gc = ac->gdma_dev->gdma_context; struct mana_query_gf_stat_resp resp = {}; @@ -2852,14 +2857,14 @@ void mana_query_gf_stats(struct mana_context *ac) sizeof(resp)); if (err) { dev_err(dev, "Failed to query GF stats: %d\n", err); - return; + return err; } err = mana_verify_resp_hdr(&resp.hdr, MANA_QUERY_GF_STAT, sizeof(resp)); if (err || resp.hdr.status) { dev_err(dev, "Failed to query GF stats: %d, 0x%x\n", err, resp.hdr.status); - return; + return err; } ac->hc_stats.hc_rx_discards_no_wqe = resp.rx_discards_nowqe; @@ -2894,6 +2899,8 @@ void mana_query_gf_stats(struct mana_context *ac) ac->hc_stats.hc_tx_mcast_pkts = resp.hc_tx_mcast_pkts; ac->hc_stats.hc_tx_mcast_bytes = resp.hc_tx_mcast_bytes; ac->hc_stats.hc_tx_err_gdma = resp.tx_err_gdma; + + return 0; } void mana_query_phy_stats(struct mana_port_context *apc) @@ -3428,6 +3435,24 @@ int mana_rdma_service_event(struct gdma_context *gc, enum gdma_service_type even return 0; } +#define MANA_GF_STATS_PERIOD (2 * HZ) + +static void mana_gf_stats_work_handler(struct work_struct *work) +{ + struct mana_context *ac = + container_of(to_delayed_work(work), struct mana_context, gf_stats_work); + int err; + + err = mana_query_gf_stats(ac); + if (err == -ETIMEDOUT) { + /* HWC timeout detected - reset stats and stop rescheduling */ + ac->hwc_timeout_occurred = true; + memset(&ac->hc_stats, 0, sizeof(ac->hc_stats)); + return; + } + schedule_delayed_work(&ac->gf_stats_work, MANA_GF_STATS_PERIOD); +} + int mana_probe(struct gdma_dev *gd, bool resuming) { struct gdma_context *gc = gd->gdma_context; @@ -3520,6 +3545,10 @@ int mana_probe(struct gdma_dev *gd, bool resuming) } err = add_adev(gd, "eth"); + + INIT_DELAYED_WORK(&ac->gf_stats_work, mana_gf_stats_work_handler); + schedule_delayed_work(&ac->gf_stats_work, MANA_GF_STATS_PERIOD); + out: if (err) { mana_remove(gd, false); @@ -3544,6 +3573,7 @@ void mana_remove(struct gdma_dev *gd, bool suspending) int i; disable_work_sync(&ac->link_change_work); + cancel_delayed_work_sync(&ac->gf_stats_work); /* adev currently doesn't support suspending, always remove it */ if (gd->adev) diff --git a/drivers/net/ethernet/microsoft/mana/mana_ethtool.c b/drivers/net/ethernet/microsoft/mana/mana_ethtool.c index 3dfd96146424..99e811208683 100644 --- a/drivers/net/ethernet/microsoft/mana/mana_ethtool.c +++ b/drivers/net/ethernet/microsoft/mana/mana_ethtool.c @@ -213,8 +213,6 @@ static void mana_get_ethtool_stats(struct net_device *ndev, if (!apc->port_is_up) return; - /* we call mana function to update stats from GDMA */ - mana_query_gf_stats(apc->ac); /* We call this mana function to get the phy stats from GDMA and includes * aggregate tx/rx drop counters, Per-TC(Traffic Channel) tx/rx and pause diff --git a/include/net/mana/gdma.h b/include/net/mana/gdma.h index 637f42485dba..2e4f2f3175e5 100644 --- a/include/net/mana/gdma.h +++ b/include/net/mana/gdma.h @@ -592,6 +592,9 @@ enum { #define GDMA_DRV_CAP_FLAG_1_HANDLE_RECONFIG_EQE BIT(17) #define GDMA_DRV_CAP_FLAG_1_HW_VPORT_LINK_AWARE BIT(6) +/* Driver can send HWC periodically to query stats */ +#define GDMA_DRV_CAP_FLAG_1_PERIODIC_STATS_QUERY BIT(21) + #define GDMA_DRV_CAP_FLAGS1 \ (GDMA_DRV_CAP_FLAG_1_EQ_SHARING_MULTI_VPORT | \ GDMA_DRV_CAP_FLAG_1_NAPI_WKDONE_FIX | \ @@ -601,7 +604,8 @@ enum { GDMA_DRV_CAP_FLAG_1_DYNAMIC_IRQ_ALLOC_SUPPORT | \ GDMA_DRV_CAP_FLAG_1_SELF_RESET_ON_EQE | \ GDMA_DRV_CAP_FLAG_1_HANDLE_RECONFIG_EQE | \ - GDMA_DRV_CAP_FLAG_1_HW_VPORT_LINK_AWARE) + GDMA_DRV_CAP_FLAG_1_HW_VPORT_LINK_AWARE | \ + GDMA_DRV_CAP_FLAG_1_PERIODIC_STATS_QUERY) #define GDMA_DRV_CAP_FLAGS2 0 diff --git a/include/net/mana/mana.h b/include/net/mana/mana.h index 3484f42803e3..d37f4cea0ac3 100644 --- a/include/net/mana/mana.h +++ b/include/net/mana/mana.h @@ -480,6 +480,10 @@ struct mana_context { struct mana_eq *eqs; struct dentry *mana_eqs_debugfs; + /* Workqueue for querying hardware stats */ + struct delayed_work gf_stats_work; + bool hwc_timeout_occurred; + struct net_device *ports[MAX_PORTS_IN_MANA_DEV]; /* Link state change work */ @@ -581,7 +585,7 @@ u32 mana_run_xdp(struct net_device *ndev, struct mana_rxq *rxq, struct bpf_prog *mana_xdp_get(struct mana_port_context *apc); void mana_chn_setxdp(struct mana_port_context *apc, struct bpf_prog *prog); int mana_bpf(struct net_device *ndev, struct netdev_bpf *bpf); -void mana_query_gf_stats(struct mana_context *ac); +int mana_query_gf_stats(struct mana_context *ac); int mana_query_link_cfg(struct mana_port_context *apc); int mana_set_bw_clamp(struct mana_port_context *apc, u32 speed, int enable_clamping); From 6b1aa3c87fcbf06b29b1a7123c386ad5cf2c8e9b Mon Sep 17 00:00:00 2001 From: Inochi Amaoto Date: Fri, 14 Nov 2025 08:38:03 +0800 Subject: [PATCH 0854/1381] dt-bindings: net: sophgo,sg2044-dwmac: add phy mode restriction As the ethernet controller of SG2044 and SG2042 only supports RGMII phy. Add phy-mode property to restrict the value. Also, since SG2042 has internal rx delay in its mac, make only "rgmii-txid" and "rgmii-id" valid for phy-mode. Fixes: e281c48a7336 ("dt-bindings: net: sophgo,sg2044-dwmac: Add support for Sophgo SG2042 dwmac") Signed-off-by: Inochi Amaoto Acked-by: Conor Dooley Link: https://patch.msgid.link/20251114003805.494387-2-inochiama@gmail.com Signed-off-by: Jakub Kicinski --- .../bindings/net/sophgo,sg2044-dwmac.yaml | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+) diff --git a/Documentation/devicetree/bindings/net/sophgo,sg2044-dwmac.yaml b/Documentation/devicetree/bindings/net/sophgo,sg2044-dwmac.yaml index ce21979a2d9a..e8d3814db0e9 100644 --- a/Documentation/devicetree/bindings/net/sophgo,sg2044-dwmac.yaml +++ b/Documentation/devicetree/bindings/net/sophgo,sg2044-dwmac.yaml @@ -70,6 +70,25 @@ required: allOf: - $ref: snps,dwmac.yaml# + - if: + properties: + compatible: + contains: + const: sophgo,sg2042-dwmac + then: + properties: + phy-mode: + enum: + - rgmii-rxid + - rgmii-id + else: + properties: + phy-mode: + enum: + - rgmii + - rgmii-rxid + - rgmii-txid + - rgmii-id unevaluatedProperties: false From 24afd7827efb7c69adfc41835390470e3eec4740 Mon Sep 17 00:00:00 2001 From: Inochi Amaoto Date: Fri, 14 Nov 2025 08:38:04 +0800 Subject: [PATCH 0855/1381] net: phy: Add helper for fixing RGMII PHY mode based on internal mac delay The "phy-mode" property of devicetree indicates whether the PCB has delay now, which means the mac needs to modify the PHY mode based on whether there is an internal delay in the mac. This modification is similar for many ethernet drivers. To simplify code, define the helper phy_fix_phy_mode_for_mac_delays(speed, mac_txid, mac_rxid) to fix PHY mode based on whether mac adds internal delay. Suggested-by: Russell King (Oracle) Signed-off-by: Inochi Amaoto Reviewed-by: Maxime Chevallier Reviewed-by: Simon Horman Link: https://patch.msgid.link/20251114003805.494387-3-inochiama@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/phy/phy-core.c | 43 ++++++++++++++++++++++++++++++++++++++ include/linux/phy.h | 3 +++ 2 files changed, 46 insertions(+) diff --git a/drivers/net/phy/phy-core.c b/drivers/net/phy/phy-core.c index 605ca20ae192..0c63e6ba2cb0 100644 --- a/drivers/net/phy/phy-core.c +++ b/drivers/net/phy/phy-core.c @@ -101,6 +101,49 @@ const char *phy_rate_matching_to_str(int rate_matching) } EXPORT_SYMBOL_GPL(phy_rate_matching_to_str); +/** + * phy_fix_phy_mode_for_mac_delays - Convenience function for fixing PHY + * mode based on whether mac adds internal delay + * + * @interface: The current interface mode of the port + * @mac_txid: True if the mac adds internal tx delay + * @mac_rxid: True if the mac adds internal rx delay + * + * Return: fixed PHY mode, or PHY_INTERFACE_MODE_NA if the interface can + * not apply the internal delay + */ +phy_interface_t phy_fix_phy_mode_for_mac_delays(phy_interface_t interface, + bool mac_txid, bool mac_rxid) +{ + if (!phy_interface_mode_is_rgmii(interface)) + return interface; + + if (mac_txid && mac_rxid) { + if (interface == PHY_INTERFACE_MODE_RGMII_ID) + return PHY_INTERFACE_MODE_RGMII; + return PHY_INTERFACE_MODE_NA; + } + + if (mac_txid) { + if (interface == PHY_INTERFACE_MODE_RGMII_ID) + return PHY_INTERFACE_MODE_RGMII_RXID; + if (interface == PHY_INTERFACE_MODE_RGMII_TXID) + return PHY_INTERFACE_MODE_RGMII; + return PHY_INTERFACE_MODE_NA; + } + + if (mac_rxid) { + if (interface == PHY_INTERFACE_MODE_RGMII_ID) + return PHY_INTERFACE_MODE_RGMII_TXID; + if (interface == PHY_INTERFACE_MODE_RGMII_RXID) + return PHY_INTERFACE_MODE_RGMII; + return PHY_INTERFACE_MODE_NA; + } + + return interface; +} +EXPORT_SYMBOL_GPL(phy_fix_phy_mode_for_mac_delays); + /** * phy_interface_num_ports - Return the number of links that can be carried by * a given MAC-PHY physical link. Returns 0 if this is diff --git a/include/linux/phy.h b/include/linux/phy.h index bf5457341ca8..65b0c3ca6a2b 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -2040,6 +2040,9 @@ static inline bool phy_is_pseudo_fixed_link(struct phy_device *phydev) return phydev->is_pseudo_fixed_link; } +phy_interface_t phy_fix_phy_mode_for_mac_delays(phy_interface_t interface, + bool mac_txid, bool mac_rxid); + int phy_save_page(struct phy_device *phydev); int phy_select_page(struct phy_device *phydev, int page); int phy_restore_page(struct phy_device *phydev, int oldpage, int ret); From db37c6e510deabc9b0ee27c08f1c5aaa19f2e8ef Mon Sep 17 00:00:00 2001 From: Inochi Amaoto Date: Fri, 14 Nov 2025 08:38:05 +0800 Subject: [PATCH 0856/1381] net: stmmac: dwmac-sophgo: Add phy interface filter As the SG2042 has an internal rx delay, the delay should be removed when initializing the mac, otherwise the phy will be misconfigurated. Fixes: 543009e2d4cd ("net: stmmac: dwmac-sophgo: Add support for Sophgo SG2042 SoC") Signed-off-by: Inochi Amaoto Tested-by: Han Gao Reviewed-by: Andrew Lunn Reviewed-by: Simon Horman Link: https://patch.msgid.link/20251114003805.494387-4-inochiama@gmail.com Signed-off-by: Jakub Kicinski --- .../ethernet/stmicro/stmmac/dwmac-sophgo.c | 20 ++++++++++++++++++- 1 file changed, 19 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-sophgo.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-sophgo.c index 24ce17ea35c4..44d4ceb8415f 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-sophgo.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-sophgo.c @@ -7,11 +7,16 @@ #include #include +#include #include #include #include "stmmac_platform.h" +struct sophgo_dwmac_data { + bool has_internal_rx_delay; +}; + static int sophgo_sg2044_dwmac_init(struct platform_device *pdev, struct plat_stmmacenet_data *plat_dat, struct stmmac_resources *stmmac_res) @@ -31,6 +36,7 @@ static int sophgo_sg2044_dwmac_init(struct platform_device *pdev, static int sophgo_dwmac_probe(struct platform_device *pdev) { struct plat_stmmacenet_data *plat_dat; + const struct sophgo_dwmac_data *data; struct stmmac_resources stmmac_res; struct device *dev = &pdev->dev; int ret; @@ -49,11 +55,23 @@ static int sophgo_dwmac_probe(struct platform_device *pdev) if (ret) return ret; + data = device_get_match_data(&pdev->dev); + if (data && data->has_internal_rx_delay) { + plat_dat->phy_interface = phy_fix_phy_mode_for_mac_delays(plat_dat->phy_interface, + false, true); + if (plat_dat->phy_interface == PHY_INTERFACE_MODE_NA) + return -EINVAL; + } + return stmmac_dvr_probe(dev, plat_dat, &stmmac_res); } +static const struct sophgo_dwmac_data sg2042_dwmac_data = { + .has_internal_rx_delay = true, +}; + static const struct of_device_id sophgo_dwmac_match[] = { - { .compatible = "sophgo,sg2042-dwmac" }, + { .compatible = "sophgo,sg2042-dwmac", .data = &sg2042_dwmac_data }, { .compatible = "sophgo,sg2044-dwmac" }, { /* sentinel */ } }; From 28c0074fd4b79c1bae561ca8ea2bfa6bb47a9f46 Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Thu, 13 Nov 2025 22:09:08 +0100 Subject: [PATCH 0857/1381] r8169: bail out from probe if fiber mode is detected on RTL8127AF It was reported that on a card with RTL8127AF (SFP + DAC) link-up isn't detected. Realtek hides the SFP behind the internal PHY, which isn't behaving fully compliance with clause 22 any longer in fiber mode. Due to not having access to chip documentation there isn't much I can do for now. Instead of silently failing to detect link-up in fiber mode, inform the user that fiber mode isn't support and bail out. The logic to detect fiber mode is borrowed from Realtek's r8127 driver. Signed-off-by: Heiner Kallweit Link: https://patch.msgid.link/fab6605a-54e2-4f54-b194-11c2b9caaaa9@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/realtek/r8169_main.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c index 0b96b6aa4214..de304d1eb477 100644 --- a/drivers/net/ethernet/realtek/r8169_main.c +++ b/drivers/net/ethernet/realtek/r8169_main.c @@ -5451,6 +5451,15 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) } tp->aspm_manageable = !rc; + /* Fiber mode on RTL8127AF isn't supported */ + if (rtl_is_8125(tp)) { + u16 data = r8168_mac_ocp_read(tp, 0xd006); + + if ((data & 0xff) == 0x07) + return dev_err_probe(&pdev->dev, -ENODEV, + "Fiber mode not supported\n"); + } + tp->dash_type = rtl_get_dash_type(tp); tp->dash_enabled = rtl_dash_is_enabled(tp); From eef1f5ae73d17a770f1019b33566507f926a1377 Mon Sep 17 00:00:00 2001 From: Aswin Karuvally Date: Thu, 13 Nov 2025 15:42:08 +0100 Subject: [PATCH 0858/1381] s390/qeth: Move all OSA RCs to single enum OSA Express defines a number of return codes whose meaning is determined by the issuing command, making them ambiguous. Move definitions of all return codes including the ambiguous ones to a single enum block to aid readability and maintainability. Reviewed-by: Alexandra Winter Signed-off-by: Aswin Karuvally Link: https://patch.msgid.link/20251113144209.2140061-2-aswin@linux.ibm.com Signed-off-by: Jakub Kicinski --- drivers/s390/net/qeth_core_mpc.h | 15 +++++---------- 1 file changed, 5 insertions(+), 10 deletions(-) diff --git a/drivers/s390/net/qeth_core_mpc.h b/drivers/s390/net/qeth_core_mpc.h index 6257f00786b3..e6904ca9defa 100644 --- a/drivers/s390/net/qeth_core_mpc.h +++ b/drivers/s390/net/qeth_core_mpc.h @@ -159,13 +159,17 @@ enum qeth_ipa_return_codes { IPA_RC_SUCCESS = 0x0000, IPA_RC_NOTSUPP = 0x0001, IPA_RC_IP_TABLE_FULL = 0x0002, + IPA_RC_INVALID_SUBCMD = 0x0002, IPA_RC_UNKNOWN_ERROR = 0x0003, + IPA_RC_HARDWARE_AUTH_ERROR = 0x0003, IPA_RC_UNSUPPORTED_COMMAND = 0x0004, IPA_RC_TRACE_ALREADY_ACTIVE = 0x0005, + IPA_RC_VNICC_OOSEQ = 0x0005, IPA_RC_INVALID_FORMAT = 0x0006, IPA_RC_DUP_IPV6_REMOTE = 0x0008, IPA_RC_SBP_IQD_NOT_CONFIGURED = 0x000C, IPA_RC_DUP_IPV6_HOME = 0x0010, + IPA_RC_SBP_IQD_OS_MISMATCH = 0x0010, IPA_RC_UNREGISTERED_ADDR = 0x0011, IPA_RC_NO_ID_AVAILABLE = 0x0012, IPA_RC_ID_NOT_FOUND = 0x0013, @@ -173,6 +177,7 @@ enum qeth_ipa_return_codes { IPA_RC_SBP_IQD_CURRENT_SECOND = 0x0018, IPA_RC_SBP_IQD_LIMIT_SECOND = 0x001C, IPA_RC_INVALID_IP_VERSION = 0x0020, + IPA_RC_SBP_IQD_NOT_AUTHD_BY_ZMAN = 0x0020, IPA_RC_SBP_IQD_CURRENT_PRIMARY = 0x0024, IPA_RC_LAN_FRAME_MISMATCH = 0x0040, IPA_RC_SBP_IQD_NO_QDIO_QUEUES = 0x00EB, @@ -220,16 +225,6 @@ enum qeth_ipa_return_codes { IPA_RC_INVALID_IP_VERSION2 = 0xf001, IPA_RC_FFFF = 0xffff }; -/* for VNIC Characteristics */ -#define IPA_RC_VNICC_OOSEQ 0x0005 - -/* for SET_DIAGNOSTIC_ASSIST */ -#define IPA_RC_INVALID_SUBCMD IPA_RC_IP_TABLE_FULL -#define IPA_RC_HARDWARE_AUTH_ERROR IPA_RC_UNKNOWN_ERROR - -/* for SETBRIDGEPORT (double occupancies) */ -#define IPA_RC_SBP_IQD_OS_MISMATCH IPA_RC_DUP_IPV6_HOME -#define IPA_RC_SBP_IQD_NOT_AUTHD_BY_ZMAN IPA_RC_INVALID_IP_VERSION /* IPA function flags; each flag marks availability of respective function */ enum qeth_ipa_funcs { From 53e58437b470227b56478a17c6498bc26c03f65c Mon Sep 17 00:00:00 2001 From: Aswin Karuvally Date: Thu, 13 Nov 2025 15:42:09 +0100 Subject: [PATCH 0859/1381] s390/qeth: Handle ambiguous OSA RCs in s390dbf OSA Express defines a number of return codes whose meaning is determined by the issuing command, making them ambiguous. The important ones are reported as debug messages through the s390 debug feature. The qeth driver currently does not take the issuing command into account when interpreting the return code which sometimes leads to incorrect debug messages. Implement a mechanism to interpret and report these return codes properly. While at it, remove extern keyword and fix indentation for function declarations to be in line with Linux kernel coding style. Suggested-by: Alexandra Winter Reviewed-by: Alexandra Winter Signed-off-by: Aswin Karuvally Link: https://patch.msgid.link/20251113144209.2140061-3-aswin@linux.ibm.com Signed-off-by: Jakub Kicinski --- drivers/s390/net/qeth_core_main.c | 2 +- drivers/s390/net/qeth_core_mpc.c | 247 ++++++++++++++++++++++++------ drivers/s390/net/qeth_core_mpc.h | 5 +- 3 files changed, 205 insertions(+), 49 deletions(-) diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c index edc0bcd46923..10b53bba373c 100644 --- a/drivers/s390/net/qeth_core_main.c +++ b/drivers/s390/net/qeth_core_main.c @@ -761,7 +761,7 @@ static void qeth_issue_ipa_msg(struct qeth_ipa_cmd *cmd, int rc, if (rc) QETH_DBF_MESSAGE(2, "IPA: %s(%#x) for device %x returned %#x \"%s\"\n", ipa_name, com, CARD_DEVID(card), rc, - qeth_get_ipa_msg(rc)); + qeth_get_ipa_msg(com, rc)); else QETH_DBF_MESSAGE(5, "IPA: %s(%#x) for device %x succeeded\n", ipa_name, com, CARD_DEVID(card)); diff --git a/drivers/s390/net/qeth_core_mpc.c b/drivers/s390/net/qeth_core_mpc.c index d9266f7d8187..1add124e033b 100644 --- a/drivers/s390/net/qeth_core_mpc.c +++ b/drivers/s390/net/qeth_core_mpc.c @@ -139,39 +139,64 @@ struct ipa_rc_msg { const char *msg; }; -static const struct ipa_rc_msg qeth_ipa_rc_msg[] = { +static const struct ipa_rc_msg qeth_ipa_rc_def_msg[] = { {IPA_RC_SUCCESS, "success"}, {IPA_RC_NOTSUPP, "Command not supported"}, - {IPA_RC_IP_TABLE_FULL, "Add Addr IP Table Full - ipv6"}, - {IPA_RC_UNKNOWN_ERROR, "IPA command failed - reason unknown"}, {IPA_RC_UNSUPPORTED_COMMAND, "Command not supported"}, - {IPA_RC_VNICC_OOSEQ, "Command issued out of sequence"}, - {IPA_RC_INVALID_FORMAT, "invalid format or length"}, {IPA_RC_DUP_IPV6_REMOTE, "ipv6 address already registered remote"}, - {IPA_RC_SBP_IQD_NOT_CONFIGURED, "Not configured for bridgeport"}, {IPA_RC_DUP_IPV6_HOME, "ipv6 address already registered"}, {IPA_RC_UNREGISTERED_ADDR, "Address not registered"}, - {IPA_RC_NO_ID_AVAILABLE, "No identifiers available"}, {IPA_RC_ID_NOT_FOUND, "Identifier not found"}, + {IPA_RC_LAN_FRAME_MISMATCH, "LAN and frame mismatch"}, + {IPA_RC_L2_UNSUPPORTED_CMD, "Unsupported layer 2 command"}, + {IPA_RC_L2_ADDR_TABLE_FULL, "Layer2 address table full"}, + {IPA_RC_L2_MAC_NOT_AUTH_BY_ADP, "L2 mac not authorized by adapter"}, + {IPA_RC_DATA_MISMATCH, "Data field mismatch (v4/v6 mixed)"}, + {IPA_RC_INVALID_MTU_SIZE, "Invalid MTU size"}, + {IPA_RC_INVALID_LANTYPE, "Invalid LAN type"}, + {IPA_RC_INVALID_LANNUM, "Invalid LAN num"}, + {IPA_RC_LAN_PORT_STATE_ERROR, "LAN port state error"}, + {IPA_RC_SETIP_NO_STARTLAN, "Setip no startlan received"}, + {IPA_RC_SETIP_ALREADY_RECEIVED, "Setip already received"}, + {IPA_RC_SETIP_INVALID_VERSION, "SETIP invalid IP version"}, + {IPA_RC_UNSUPPORTED_SUBCMD, "Unsupported assist subcommand"}, + {IPA_RC_ARP_ASSIST_NO_ENABLE, "Only partial success, no enable"}, + {IPA_RC_INVALID_IP_VERSION2, "Invalid IP version"}, + /* default for qeth_get_ipa_msg(): */ + {IPA_RC_FFFF, "Unknown Error"} +}; + +static const struct ipa_rc_msg qeth_ipa_rc_adp_parms_msg[] = { + {IPA_RC_IP_TABLE_FULL, "Add Addr IP Table Full - ipv6"} +}; + +static const struct ipa_rc_msg qeth_ipa_rc_diag_ass_msg[] = { + {IPA_RC_INVALID_FORMAT, "invalid format or length"} +}; + +static const struct ipa_rc_msg qeth_ipa_rc_addr_msg[] = { + {IPA_RC_UNKNOWN_ERROR, "IPA command failed - reason unknown"}, + {IPA_RC_NO_ID_AVAILABLE, "No identifiers available"}, + {IPA_RC_INVALID_IP_VERSION, "IP version incorrect"} +}; + +static const struct ipa_rc_msg qeth_ipa_rc_vnicc_msg[] = { + {IPA_RC_VNICC_OOSEQ, "Command issued out of sequence"}, + {IPA_RC_VNICC_VNICBP, "VNIC is BridgePort"} +}; + +static const struct ipa_rc_msg qeth_ipa_rc_sbp_iqd_msg[] = { + {IPA_RC_SBP_IQD_NOT_CONFIGURED, "Not configured for bridgeport"}, + {IPA_RC_SBP_IQD_OS_MISMATCH, "OS mismatch"}, {IPA_RC_SBP_IQD_ANO_DEV_PRIMARY, "Primary bridgeport exists already"}, {IPA_RC_SBP_IQD_CURRENT_SECOND, "Bridgeport is currently secondary"}, {IPA_RC_SBP_IQD_LIMIT_SECOND, "Limit of secondary bridgeports reached"}, - {IPA_RC_INVALID_IP_VERSION, "IP version incorrect"}, + {IPA_RC_SBP_IQD_NOT_AUTHD_BY_ZMAN, "Not authorized by zManager"}, {IPA_RC_SBP_IQD_CURRENT_PRIMARY, "Bridgeport is currently primary"}, - {IPA_RC_LAN_FRAME_MISMATCH, "LAN and frame mismatch"}, - {IPA_RC_SBP_IQD_NO_QDIO_QUEUES, "QDIO queues not established"}, - {IPA_RC_L2_UNSUPPORTED_CMD, "Unsupported layer 2 command"}, - {IPA_RC_L2_DUP_MAC, "Duplicate MAC address"}, - {IPA_RC_L2_ADDR_TABLE_FULL, "Layer2 address table full"}, - {IPA_RC_L2_DUP_LAYER3_MAC, "Duplicate with layer 3 MAC"}, - {IPA_RC_L2_GMAC_NOT_FOUND, "GMAC not found"}, - {IPA_RC_L2_MAC_NOT_AUTH_BY_HYP, "L2 mac not authorized by hypervisor"}, - {IPA_RC_L2_MAC_NOT_AUTH_BY_ADP, "L2 mac not authorized by adapter"}, - {IPA_RC_L2_MAC_NOT_FOUND, "L2 mac address not found"}, - {IPA_RC_L2_INVALID_VLAN_ID, "L2 invalid vlan id"}, - {IPA_RC_L2_DUP_VLAN_ID, "L2 duplicate vlan id"}, - {IPA_RC_L2_VLAN_ID_NOT_FOUND, "L2 vlan id not found"}, - {IPA_RC_VNICC_VNICBP, "VNIC is BridgePort"}, + {IPA_RC_SBP_IQD_NO_QDIO_QUEUES, "QDIO queues not established"} +}; + +static const struct ipa_rc_msg qeth_ipa_rc_sbp_osa_msg[] = { {IPA_RC_SBP_OSA_NOT_CONFIGURED, "Not configured for bridgeport"}, {IPA_RC_SBP_OSA_OS_MISMATCH, "OS mismatch"}, {IPA_RC_SBP_OSA_ANO_DEV_PRIMARY, "Primary bridgeport exists already"}, @@ -179,43 +204,173 @@ static const struct ipa_rc_msg qeth_ipa_rc_msg[] = { {IPA_RC_SBP_OSA_LIMIT_SECOND, "Limit of secondary bridgeports reached"}, {IPA_RC_SBP_OSA_NOT_AUTHD_BY_ZMAN, "Not authorized by zManager"}, {IPA_RC_SBP_OSA_CURRENT_PRIMARY, "Bridgeport is currently primary"}, - {IPA_RC_SBP_OSA_NO_QDIO_QUEUES, "QDIO queues not established"}, - {IPA_RC_DATA_MISMATCH, "Data field mismatch (v4/v6 mixed)"}, - {IPA_RC_INVALID_MTU_SIZE, "Invalid MTU size"}, - {IPA_RC_INVALID_LANTYPE, "Invalid LAN type"}, - {IPA_RC_INVALID_LANNUM, "Invalid LAN num"}, + {IPA_RC_SBP_OSA_NO_QDIO_QUEUES, "QDIO queues not established"} +}; + +static const struct ipa_rc_msg qeth_ipa_rc_mac_msg[] = { + {IPA_RC_L2_DUP_MAC, "Duplicate MAC address"}, + {IPA_RC_L2_DUP_LAYER3_MAC, "Duplicate with layer 3 MAC"}, + {IPA_RC_L2_GMAC_NOT_FOUND, "GMAC not found"}, + {IPA_RC_L2_MAC_NOT_AUTH_BY_HYP, "L2 mac not authorized by hypervisor"}, + {IPA_RC_L2_MAC_NOT_FOUND, "L2 mac address not found"} +}; + +static const struct ipa_rc_msg qeth_ipa_rc_ip_msg[] = { {IPA_RC_DUPLICATE_IP_ADDRESS, "Address already registered"}, {IPA_RC_IP_ADDR_TABLE_FULL, "IP address table full"}, - {IPA_RC_LAN_PORT_STATE_ERROR, "LAN port state error"}, - {IPA_RC_SETIP_NO_STARTLAN, "Setip no startlan received"}, - {IPA_RC_SETIP_ALREADY_RECEIVED, "Setip already received"}, {IPA_RC_IP_ADDR_ALREADY_USED, "IP address already in use on LAN"}, - {IPA_RC_MC_ADDR_NOT_FOUND, "Multicast address not found"}, - {IPA_RC_SETIP_INVALID_VERSION, "SETIP invalid IP version"}, - {IPA_RC_UNSUPPORTED_SUBCMD, "Unsupported assist subcommand"}, - {IPA_RC_ARP_ASSIST_NO_ENABLE, "Only partial success, no enable"}, + {IPA_RC_MC_ADDR_NOT_FOUND, "Multicast address not found"} +}; + +static const struct ipa_rc_msg qeth_ipa_rc_lan_msg[] = { + {IPA_RC_LAN_OFFLINE, "STRTLAN_LAN_DISABLED - LAN offline"}, + {IPA_RC_VEPA_TO_VEB_TRANSITION, "Adj. switch disabled port mode RR"}, +}; + +static const struct ipa_rc_msg qeth_ipa_rc_vlan_msg[] = { + {IPA_RC_L2_INVALID_VLAN_ID, "L2 invalid vlan id"}, + {IPA_RC_L2_DUP_VLAN_ID, "L2 duplicate vlan id"}, + {IPA_RC_L2_VLAN_ID_NOT_FOUND, "L2 vlan id not found"} +}; + +static const struct ipa_rc_msg qeth_ipa_rc_rtg_msg[] = { {IPA_RC_PRIMARY_ALREADY_DEFINED, "Primary already defined"}, {IPA_RC_SECOND_ALREADY_DEFINED, "Secondary already defined"}, {IPA_RC_INVALID_SETRTG_INDICATOR, "Invalid SETRTG indicator"}, - {IPA_RC_MC_ADDR_ALREADY_DEFINED, "Multicast address already defined"}, - {IPA_RC_LAN_OFFLINE, "STRTLAN_LAN_DISABLED - LAN offline"}, - {IPA_RC_VEPA_TO_VEB_TRANSITION, "Adj. switch disabled port mode RR"}, - {IPA_RC_INVALID_IP_VERSION2, "Invalid IP version"}, - /* default for qeth_get_ipa_msg(): */ - {IPA_RC_FFFF, "Unknown Error"} + {IPA_RC_MC_ADDR_ALREADY_DEFINED, "Multicast address already defined"} }; -const char *qeth_get_ipa_msg(enum qeth_ipa_return_codes rc) +struct ipa_cmd_rc_map { + enum qeth_ipa_cmds cmd; + const struct ipa_rc_msg *msg_arr; + const size_t arr_len; +}; + +static const struct ipa_cmd_rc_map qeth_ipa_cmd_rc_map[] = { + { + .cmd = IPA_CMD_SETADAPTERPARMS, + .msg_arr = qeth_ipa_rc_adp_parms_msg, + .arr_len = ARRAY_SIZE(qeth_ipa_rc_adp_parms_msg) + }, + { + .cmd = IPA_CMD_SET_DIAG_ASS, + .msg_arr = qeth_ipa_rc_diag_ass_msg, + .arr_len = ARRAY_SIZE(qeth_ipa_rc_diag_ass_msg) + }, + { + .cmd = IPA_CMD_CREATE_ADDR, + .msg_arr = qeth_ipa_rc_addr_msg, + .arr_len = ARRAY_SIZE(qeth_ipa_rc_addr_msg) + }, + { + .cmd = IPA_CMD_DESTROY_ADDR, + .msg_arr = qeth_ipa_rc_addr_msg, + .arr_len = ARRAY_SIZE(qeth_ipa_rc_addr_msg) + }, + { + .cmd = IPA_CMD_VNICC, + .msg_arr = qeth_ipa_rc_vnicc_msg, + .arr_len = ARRAY_SIZE(qeth_ipa_rc_vnicc_msg) + }, + { + .cmd = IPA_CMD_SETBRIDGEPORT_IQD, + .msg_arr = qeth_ipa_rc_sbp_iqd_msg, + .arr_len = ARRAY_SIZE(qeth_ipa_rc_sbp_iqd_msg) + }, + { + .cmd = IPA_CMD_SETBRIDGEPORT_OSA, + .msg_arr = qeth_ipa_rc_sbp_osa_msg, + .arr_len = ARRAY_SIZE(qeth_ipa_rc_sbp_osa_msg) + }, + { + .cmd = IPA_CMD_SETVMAC, + .msg_arr = qeth_ipa_rc_mac_msg, + .arr_len = ARRAY_SIZE(qeth_ipa_rc_mac_msg) + }, + { + .cmd = IPA_CMD_DELVMAC, + .msg_arr = qeth_ipa_rc_mac_msg, + .arr_len = ARRAY_SIZE(qeth_ipa_rc_mac_msg) + }, + { + .cmd = IPA_CMD_SETGMAC, + .msg_arr = qeth_ipa_rc_mac_msg, + .arr_len = ARRAY_SIZE(qeth_ipa_rc_mac_msg) + }, + { + .cmd = IPA_CMD_DELGMAC, + .msg_arr = qeth_ipa_rc_mac_msg, + .arr_len = ARRAY_SIZE(qeth_ipa_rc_mac_msg) + }, + { + .cmd = IPA_CMD_SETIP, + .msg_arr = qeth_ipa_rc_ip_msg, + .arr_len = ARRAY_SIZE(qeth_ipa_rc_ip_msg) + }, + { + .cmd = IPA_CMD_SETIPM, + .msg_arr = qeth_ipa_rc_ip_msg, + .arr_len = ARRAY_SIZE(qeth_ipa_rc_ip_msg) + }, + { + .cmd = IPA_CMD_DELIPM, + .msg_arr = qeth_ipa_rc_ip_msg, + .arr_len = ARRAY_SIZE(qeth_ipa_rc_ip_msg) + }, + { + .cmd = IPA_CMD_STARTLAN, + .msg_arr = qeth_ipa_rc_lan_msg, + .arr_len = ARRAY_SIZE(qeth_ipa_rc_lan_msg) + }, + { + .cmd = IPA_CMD_STOPLAN, + .msg_arr = qeth_ipa_rc_lan_msg, + .arr_len = ARRAY_SIZE(qeth_ipa_rc_lan_msg) + }, + { + .cmd = IPA_CMD_SETVLAN, + .msg_arr = qeth_ipa_rc_vlan_msg, + .arr_len = ARRAY_SIZE(qeth_ipa_rc_vlan_msg) + }, + { + .cmd = IPA_CMD_DELVLAN, + .msg_arr = qeth_ipa_rc_vlan_msg, + .arr_len = ARRAY_SIZE(qeth_ipa_rc_vlan_msg) + }, + { + .cmd = IPA_CMD_SETRTG, + .msg_arr = qeth_ipa_rc_rtg_msg, + .arr_len = ARRAY_SIZE(qeth_ipa_rc_rtg_msg) + } +}; + +const char *qeth_get_ipa_msg(enum qeth_ipa_cmds cmd, + enum qeth_ipa_return_codes rc) { int x; + const struct ipa_rc_msg *msg_arr = NULL; + size_t arr_len = 0; - for (x = 0; x < ARRAY_SIZE(qeth_ipa_rc_msg) - 1; x++) - if (qeth_ipa_rc_msg[x].rc == rc) - return qeth_ipa_rc_msg[x].msg; - return qeth_ipa_rc_msg[x].msg; + for (x = 0; x < ARRAY_SIZE(qeth_ipa_cmd_rc_map); x++) { + if (qeth_ipa_cmd_rc_map[x].cmd == cmd) { + msg_arr = qeth_ipa_cmd_rc_map[x].msg_arr; + arr_len = qeth_ipa_cmd_rc_map[x].arr_len; + break; + } + } + + for (x = 0; x < arr_len; x++) { + if (msg_arr[x].rc == rc) + return msg_arr[x].msg; + } + + for (x = 0; x < ARRAY_SIZE(qeth_ipa_rc_def_msg) - 1; x++) { + if (qeth_ipa_rc_def_msg[x].rc == rc) + return qeth_ipa_rc_def_msg[x].msg; + } + return qeth_ipa_rc_def_msg[x].msg; } - struct ipa_cmd_names { enum qeth_ipa_cmds cmd; const char *name; diff --git a/drivers/s390/net/qeth_core_mpc.h b/drivers/s390/net/qeth_core_mpc.h index e6904ca9defa..252fc84e6eca 100644 --- a/drivers/s390/net/qeth_core_mpc.h +++ b/drivers/s390/net/qeth_core_mpc.h @@ -857,8 +857,9 @@ enum qeth_ipa_arp_return_codes { QETH_IPA_ARP_RC_Q_NO_DATA = 0x0008, }; -extern const char *qeth_get_ipa_msg(enum qeth_ipa_return_codes rc); -extern const char *qeth_get_ipa_cmd_name(enum qeth_ipa_cmds cmd); +const char *qeth_get_ipa_msg(enum qeth_ipa_cmds cmd, + enum qeth_ipa_return_codes rc); +const char *qeth_get_ipa_cmd_name(enum qeth_ipa_cmds cmd); /* Helper functions */ #define IS_IPA_REPLY(cmd) ((cmd)->hdr.initiator == IPA_CMD_INITIATOR_HOST) From 58fb88d30b0250f928e1afa0eaa4547770d86229 Mon Sep 17 00:00:00 2001 From: Ivan Vecera Date: Thu, 13 Nov 2025 08:41:00 +0100 Subject: [PATCH 0860/1381] dpll: zl3073x: Store raw register values instead of parsed state The zl3073x_ref, zl3073x_out and zl3073x_synth structures previously stored state that was parsed from register reads. This included values like boolean 'enabled' flags, synthesizer selections, and pre-calculated frequencies. This commit refactors the state management to store the raw register values directly in these structures. The various inline helper functions are updated to parse these raw values on-demand using FIELD_GET. Reviewed-by: Petr Oros Tested-by: Prathosh Satish Signed-off-by: Ivan Vecera Link: https://patch.msgid.link/20251113074105.141379-2-ivecera@redhat.com Signed-off-by: Jakub Kicinski --- drivers/dpll/zl3073x/core.c | 81 ++++++++++++------------------------- drivers/dpll/zl3073x/core.h | 61 ++++++++++++++++------------ 2 files changed, 60 insertions(+), 82 deletions(-) diff --git a/drivers/dpll/zl3073x/core.c b/drivers/dpll/zl3073x/core.c index e42e527813cf..50c1fe59bc7f 100644 --- a/drivers/dpll/zl3073x/core.c +++ b/drivers/dpll/zl3073x/core.c @@ -598,25 +598,22 @@ int zl3073x_write_hwreg_seq(struct zl3073x_dev *zldev, * @zldev: pointer to zl3073x_dev structure * @index: input reference index to fetch state for * - * Function fetches information for the given input reference that are - * invariant and stores them for later use. + * Function fetches state for the given input reference and stores it for + * later user. * * Return: 0 on success, <0 on error */ static int zl3073x_ref_state_fetch(struct zl3073x_dev *zldev, u8 index) { - struct zl3073x_ref *input = &zldev->ref[index]; - u8 ref_config; + struct zl3073x_ref *ref = &zldev->ref[index]; int rc; /* If the input is differential then the configuration for N-pin * reference is ignored and P-pin config is used for both. */ - if (zl3073x_is_n_pin(index) && - zl3073x_ref_is_diff(zldev, index - 1)) { - input->enabled = zl3073x_ref_is_enabled(zldev, index - 1); - input->diff = true; + if (zl3073x_is_n_pin(index) && zl3073x_ref_is_diff(zldev, index - 1)) { + memcpy(ref, &zldev->ref[index - 1], sizeof(*ref)); return 0; } @@ -630,16 +627,14 @@ zl3073x_ref_state_fetch(struct zl3073x_dev *zldev, u8 index) return rc; /* Read ref_config register */ - rc = zl3073x_read_u8(zldev, ZL_REG_REF_CONFIG, &ref_config); + rc = zl3073x_read_u8(zldev, ZL_REG_REF_CONFIG, &ref->config); if (rc) return rc; - input->enabled = FIELD_GET(ZL_REF_CONFIG_ENABLE, ref_config); - input->diff = FIELD_GET(ZL_REF_CONFIG_DIFF_EN, ref_config); - dev_dbg(zldev->dev, "REF%u is %s and configured as %s\n", index, - str_enabled_disabled(input->enabled), - input->diff ? "differential" : "single-ended"); + str_enabled_disabled(zl3073x_ref_is_enabled(zldev, index)), + zl3073x_ref_is_diff(zldev, index) + ? "differential" : "single-ended"); return rc; } @@ -649,8 +644,8 @@ zl3073x_ref_state_fetch(struct zl3073x_dev *zldev, u8 index) * @zldev: pointer to zl3073x_dev structure * @index: output index to fetch state for * - * Function fetches information for the given output (not output pin) - * that are invariant and stores them for later use. + * Function fetches state of the given output (not output pin) and stores it + * for later use. * * Return: 0 on success, <0 on error */ @@ -658,22 +653,16 @@ static int zl3073x_out_state_fetch(struct zl3073x_dev *zldev, u8 index) { struct zl3073x_out *out = &zldev->out[index]; - u8 output_ctrl, output_mode; int rc; /* Read output configuration */ - rc = zl3073x_read_u8(zldev, ZL_REG_OUTPUT_CTRL(index), &output_ctrl); + rc = zl3073x_read_u8(zldev, ZL_REG_OUTPUT_CTRL(index), &out->ctrl); if (rc) return rc; - /* Store info about output enablement and synthesizer the output - * is connected to. - */ - out->enabled = FIELD_GET(ZL_OUTPUT_CTRL_EN, output_ctrl); - out->synth = FIELD_GET(ZL_OUTPUT_CTRL_SYNTH_SEL, output_ctrl); - dev_dbg(zldev->dev, "OUT%u is %s and connected to SYNTH%u\n", index, - str_enabled_disabled(out->enabled), out->synth); + str_enabled_disabled(zl3073x_out_is_enabled(zldev, index)), + zl3073x_out_synth_get(zldev, index)); guard(mutex)(&zldev->multiop_lock); @@ -683,17 +672,13 @@ zl3073x_out_state_fetch(struct zl3073x_dev *zldev, u8 index) if (rc) return rc; - /* Read output_mode */ - rc = zl3073x_read_u8(zldev, ZL_REG_OUTPUT_MODE, &output_mode); + /* Read output mode */ + rc = zl3073x_read_u8(zldev, ZL_REG_OUTPUT_MODE, &out->mode); if (rc) return rc; - /* Extract and store output signal format */ - out->signal_format = FIELD_GET(ZL_OUTPUT_MODE_SIGNAL_FORMAT, - output_mode); - dev_dbg(zldev->dev, "OUT%u has signal format 0x%02x\n", index, - out->signal_format); + zl3073x_out_signal_format_get(zldev, index)); return rc; } @@ -703,8 +688,7 @@ zl3073x_out_state_fetch(struct zl3073x_dev *zldev, u8 index) * @zldev: pointer to zl3073x_dev structure * @index: synth index to fetch state for * - * Function fetches information for the given synthesizer that are - * invariant and stores them for later use. + * Function fetches state of the given synthesizer and stores it for later use. * * Return: 0 on success, <0 on error */ @@ -712,25 +696,13 @@ static int zl3073x_synth_state_fetch(struct zl3073x_dev *zldev, u8 index) { struct zl3073x_synth *synth = &zldev->synth[index]; - u16 base, m, n; - u8 synth_ctrl; - u32 mult; int rc; /* Read synth control register */ - rc = zl3073x_read_u8(zldev, ZL_REG_SYNTH_CTRL(index), &synth_ctrl); + rc = zl3073x_read_u8(zldev, ZL_REG_SYNTH_CTRL(index), &synth->ctrl); if (rc) return rc; - /* Store info about synth enablement and DPLL channel the synth is - * driven by. - */ - synth->enabled = FIELD_GET(ZL_SYNTH_CTRL_EN, synth_ctrl); - synth->dpll = FIELD_GET(ZL_SYNTH_CTRL_DPLL_SEL, synth_ctrl); - - dev_dbg(zldev->dev, "SYNTH%u is %s and driven by DPLL%u\n", index, - str_enabled_disabled(synth->enabled), synth->dpll); - guard(mutex)(&zldev->multiop_lock); /* Read synth configuration */ @@ -744,35 +716,32 @@ zl3073x_synth_state_fetch(struct zl3073x_dev *zldev, u8 index) * * Read registers with these values */ - rc = zl3073x_read_u16(zldev, ZL_REG_SYNTH_FREQ_BASE, &base); + rc = zl3073x_read_u16(zldev, ZL_REG_SYNTH_FREQ_BASE, &synth->freq_base); if (rc) return rc; - rc = zl3073x_read_u32(zldev, ZL_REG_SYNTH_FREQ_MULT, &mult); + rc = zl3073x_read_u32(zldev, ZL_REG_SYNTH_FREQ_MULT, &synth->freq_mult); if (rc) return rc; - rc = zl3073x_read_u16(zldev, ZL_REG_SYNTH_FREQ_M, &m); + rc = zl3073x_read_u16(zldev, ZL_REG_SYNTH_FREQ_M, &synth->freq_m); if (rc) return rc; - rc = zl3073x_read_u16(zldev, ZL_REG_SYNTH_FREQ_N, &n); + rc = zl3073x_read_u16(zldev, ZL_REG_SYNTH_FREQ_N, &synth->freq_n); if (rc) return rc; /* Check denominator for zero to avoid div by 0 */ - if (!n) { + if (!synth->freq_n) { dev_err(zldev->dev, "Zero divisor for SYNTH%u retrieved from device\n", index); return -EINVAL; } - /* Compute and store synth frequency */ - zldev->synth[index].freq = div_u64(mul_u32_u32(base * m, mult), n); - dev_dbg(zldev->dev, "SYNTH%u frequency: %u Hz\n", index, - zldev->synth[index].freq); + zl3073x_synth_freq_get(zldev, index)); return rc; } diff --git a/drivers/dpll/zl3073x/core.h b/drivers/dpll/zl3073x/core.h index 1dca4ddcf235..51d0fd6cfabf 100644 --- a/drivers/dpll/zl3073x/core.h +++ b/drivers/dpll/zl3073x/core.h @@ -29,38 +29,38 @@ struct zl3073x_dpll; /** * struct zl3073x_ref - input reference invariant info - * @enabled: input reference is enabled or disabled - * @diff: true if input reference is differential * @ffo: current fractional frequency offset + * @config: reference config */ struct zl3073x_ref { - bool enabled; - bool diff; s64 ffo; + u8 config; }; /** * struct zl3073x_out - output invariant info - * @enabled: out is enabled or disabled - * @synth: synthesizer the out is connected to - * @signal_format: out signal format + * @ctrl: output control + * @mode: output mode */ struct zl3073x_out { - bool enabled; - u8 synth; - u8 signal_format; + u8 ctrl; + u8 mode; }; /** * struct zl3073x_synth - synthesizer invariant info - * @freq: synthesizer frequency - * @dpll: ID of DPLL the synthesizer is driven by - * @enabled: synth is enabled or disabled + * @freq_mult: frequency multiplier + * @freq_base: frequency base + * @freq_m: frequency numerator + * @freq_n: frequency denominator + * @ctrl: synth control */ struct zl3073x_synth { - u32 freq; - u8 dpll; - bool enabled; + u32 freq_mult; + u16 freq_base; + u16 freq_m; + u16 freq_n; + u8 ctrl; }; /** @@ -239,7 +239,10 @@ zl3073x_ref_ffo_get(struct zl3073x_dev *zldev, u8 index) static inline bool zl3073x_ref_is_diff(struct zl3073x_dev *zldev, u8 index) { - return zldev->ref[index].diff; + if (FIELD_GET(ZL_REF_CONFIG_DIFF_EN, zldev->ref[index].config)) + return true; + + return false; } /** @@ -252,7 +255,10 @@ zl3073x_ref_is_diff(struct zl3073x_dev *zldev, u8 index) static inline bool zl3073x_ref_is_enabled(struct zl3073x_dev *zldev, u8 index) { - return zldev->ref[index].enabled; + if (FIELD_GET(ZL_REF_CONFIG_ENABLE, zldev->ref[index].config)) + return true; + + return false; } /** @@ -265,7 +271,7 @@ zl3073x_ref_is_enabled(struct zl3073x_dev *zldev, u8 index) static inline u8 zl3073x_synth_dpll_get(struct zl3073x_dev *zldev, u8 index) { - return zldev->synth[index].dpll; + return FIELD_GET(ZL_SYNTH_CTRL_DPLL_SEL, zldev->synth[index].ctrl); } /** @@ -278,7 +284,10 @@ zl3073x_synth_dpll_get(struct zl3073x_dev *zldev, u8 index) static inline u32 zl3073x_synth_freq_get(struct zl3073x_dev *zldev, u8 index) { - return zldev->synth[index].freq; + struct zl3073x_synth *synth = &zldev->synth[index]; + + return mul_u64_u32_div(synth->freq_base * synth->freq_m, + synth->freq_mult, synth->freq_n); } /** @@ -291,7 +300,7 @@ zl3073x_synth_freq_get(struct zl3073x_dev *zldev, u8 index) static inline bool zl3073x_synth_is_enabled(struct zl3073x_dev *zldev, u8 index) { - return zldev->synth[index].enabled; + return FIELD_GET(ZL_SYNTH_CTRL_EN, zldev->synth[index].ctrl); } /** @@ -304,7 +313,7 @@ zl3073x_synth_is_enabled(struct zl3073x_dev *zldev, u8 index) static inline u8 zl3073x_out_synth_get(struct zl3073x_dev *zldev, u8 index) { - return zldev->out[index].synth; + return FIELD_GET(ZL_OUTPUT_CTRL_SYNTH_SEL, zldev->out[index].ctrl); } /** @@ -321,10 +330,10 @@ zl3073x_out_is_enabled(struct zl3073x_dev *zldev, u8 index) /* Output is enabled only if associated synth is enabled */ synth = zl3073x_out_synth_get(zldev, index); - if (zl3073x_synth_is_enabled(zldev, synth)) - return zldev->out[index].enabled; + if (!zl3073x_synth_is_enabled(zldev, synth)) + return false; - return false; + return FIELD_GET(ZL_OUTPUT_CTRL_EN, zldev->out[index].ctrl); } /** @@ -337,7 +346,7 @@ zl3073x_out_is_enabled(struct zl3073x_dev *zldev, u8 index) static inline u8 zl3073x_out_signal_format_get(struct zl3073x_dev *zldev, u8 index) { - return zldev->out[index].signal_format; + return FIELD_GET(ZL_OUTPUT_MODE_SIGNAL_FORMAT, zldev->out[index].mode); } /** From 607f2c00c61faa3b437dbb0d38287e7a9d398a52 Mon Sep 17 00:00:00 2001 From: Ivan Vecera Date: Thu, 13 Nov 2025 08:41:01 +0100 Subject: [PATCH 0861/1381] dpll: zl3073x: Split ref, out, and synth logic from core Refactor the zl3073x driver by splitting the logic for input references, outputs and synthesizers out of the monolithic core.[ch] files. Move the logic for each functional block into its own dedicated files: ref.[ch], out.[ch] and synth.[ch]. Specifically: - Move state structures (zl3073x_ref, zl3073x_out, zl3073x_synth) from core.h into their respective new headers - Move state-fetching functions (..._state_fetch) from core.c to their new .c files - Move the zl3073x_ref_freq_factorize helper from core.c to ref.c - Introduce a new helper layer to decouple the core device logic from the state-parsing logic: 1. Move the original inline helpers (e.g., zl3073x_ref_is_enabled) to the new headers (ref.h, etc.) and make them operate on a const struct ... * pointer. 2. Create new zl3073x_dev_... prefixed functions in core.h (e.g., zl3073x_dev_ref_is_enabled) and Implement these _dev_ functions to fetch state using a new ..._state_get() helper and then call the non-prefixed helper. 3. Update all driver-internal callers (in dpll.c, prop.c, etc.) to use the new zl3073x_dev_... functions. Reviewed-by: Petr Oros Tested-by: Prathosh Satish Signed-off-by: Ivan Vecera Link: https://patch.msgid.link/20251113074105.141379-3-ivecera@redhat.com Signed-off-by: Jakub Kicinski --- drivers/dpll/zl3073x/Makefile | 3 +- drivers/dpll/zl3073x/core.c | 194 ---------------------------------- drivers/dpll/zl3073x/core.h | 170 +++++++++++++---------------- drivers/dpll/zl3073x/dpll.c | 36 +++---- drivers/dpll/zl3073x/out.c | 67 ++++++++++++ drivers/dpll/zl3073x/out.h | 80 ++++++++++++++ drivers/dpll/zl3073x/prop.c | 12 +-- drivers/dpll/zl3073x/ref.c | 112 ++++++++++++++++++++ drivers/dpll/zl3073x/ref.h | 66 ++++++++++++ drivers/dpll/zl3073x/synth.c | 87 +++++++++++++++ drivers/dpll/zl3073x/synth.h | 72 +++++++++++++ 11 files changed, 584 insertions(+), 315 deletions(-) create mode 100644 drivers/dpll/zl3073x/out.c create mode 100644 drivers/dpll/zl3073x/out.h create mode 100644 drivers/dpll/zl3073x/ref.c create mode 100644 drivers/dpll/zl3073x/ref.h create mode 100644 drivers/dpll/zl3073x/synth.c create mode 100644 drivers/dpll/zl3073x/synth.h diff --git a/drivers/dpll/zl3073x/Makefile b/drivers/dpll/zl3073x/Makefile index 84e22aae57e5..bd324c7fe710 100644 --- a/drivers/dpll/zl3073x/Makefile +++ b/drivers/dpll/zl3073x/Makefile @@ -1,7 +1,8 @@ # SPDX-License-Identifier: GPL-2.0 obj-$(CONFIG_ZL3073X) += zl3073x.o -zl3073x-objs := core.o devlink.o dpll.o flash.o fw.o prop.o +zl3073x-objs := core.o devlink.o dpll.o flash.o fw.o \ + out.o prop.o ref.o synth.o obj-$(CONFIG_ZL3073X_I2C) += zl3073x_i2c.o zl3073x_i2c-objs := i2c.o diff --git a/drivers/dpll/zl3073x/core.c b/drivers/dpll/zl3073x/core.c index 50c1fe59bc7f..2f340f7eb9ec 100644 --- a/drivers/dpll/zl3073x/core.c +++ b/drivers/dpll/zl3073x/core.c @@ -129,47 +129,6 @@ const struct regmap_config zl3073x_regmap_config = { }; EXPORT_SYMBOL_NS_GPL(zl3073x_regmap_config, "ZL3073X"); -/** - * zl3073x_ref_freq_factorize - factorize given frequency - * @freq: input frequency - * @base: base frequency - * @mult: multiplier - * - * Checks if the given frequency can be factorized using one of the - * supported base frequencies. If so the base frequency and multiplier - * are stored into appropriate parameters if they are not NULL. - * - * Return: 0 on success, -EINVAL if the frequency cannot be factorized - */ -int -zl3073x_ref_freq_factorize(u32 freq, u16 *base, u16 *mult) -{ - static const u16 base_freqs[] = { - 1, 2, 4, 5, 8, 10, 16, 20, 25, 32, 40, 50, 64, 80, 100, 125, - 128, 160, 200, 250, 256, 320, 400, 500, 625, 640, 800, 1000, - 1250, 1280, 1600, 2000, 2500, 3125, 3200, 4000, 5000, 6250, - 6400, 8000, 10000, 12500, 15625, 16000, 20000, 25000, 31250, - 32000, 40000, 50000, 62500, - }; - u32 div; - int i; - - for (i = 0; i < ARRAY_SIZE(base_freqs); i++) { - div = freq / base_freqs[i]; - - if (div <= U16_MAX && (freq % base_freqs[i]) == 0) { - if (base) - *base = base_freqs[i]; - if (mult) - *mult = div; - - return 0; - } - } - - return -EINVAL; -} - static bool zl3073x_check_reg(struct zl3073x_dev *zldev, unsigned int reg, size_t size) { @@ -593,159 +552,6 @@ int zl3073x_write_hwreg_seq(struct zl3073x_dev *zldev, return rc; } -/** - * zl3073x_ref_state_fetch - get input reference state - * @zldev: pointer to zl3073x_dev structure - * @index: input reference index to fetch state for - * - * Function fetches state for the given input reference and stores it for - * later user. - * - * Return: 0 on success, <0 on error - */ -static int -zl3073x_ref_state_fetch(struct zl3073x_dev *zldev, u8 index) -{ - struct zl3073x_ref *ref = &zldev->ref[index]; - int rc; - - /* If the input is differential then the configuration for N-pin - * reference is ignored and P-pin config is used for both. - */ - if (zl3073x_is_n_pin(index) && zl3073x_ref_is_diff(zldev, index - 1)) { - memcpy(ref, &zldev->ref[index - 1], sizeof(*ref)); - - return 0; - } - - guard(mutex)(&zldev->multiop_lock); - - /* Read reference configuration */ - rc = zl3073x_mb_op(zldev, ZL_REG_REF_MB_SEM, ZL_REF_MB_SEM_RD, - ZL_REG_REF_MB_MASK, BIT(index)); - if (rc) - return rc; - - /* Read ref_config register */ - rc = zl3073x_read_u8(zldev, ZL_REG_REF_CONFIG, &ref->config); - if (rc) - return rc; - - dev_dbg(zldev->dev, "REF%u is %s and configured as %s\n", index, - str_enabled_disabled(zl3073x_ref_is_enabled(zldev, index)), - zl3073x_ref_is_diff(zldev, index) - ? "differential" : "single-ended"); - - return rc; -} - -/** - * zl3073x_out_state_fetch - get output state - * @zldev: pointer to zl3073x_dev structure - * @index: output index to fetch state for - * - * Function fetches state of the given output (not output pin) and stores it - * for later use. - * - * Return: 0 on success, <0 on error - */ -static int -zl3073x_out_state_fetch(struct zl3073x_dev *zldev, u8 index) -{ - struct zl3073x_out *out = &zldev->out[index]; - int rc; - - /* Read output configuration */ - rc = zl3073x_read_u8(zldev, ZL_REG_OUTPUT_CTRL(index), &out->ctrl); - if (rc) - return rc; - - dev_dbg(zldev->dev, "OUT%u is %s and connected to SYNTH%u\n", index, - str_enabled_disabled(zl3073x_out_is_enabled(zldev, index)), - zl3073x_out_synth_get(zldev, index)); - - guard(mutex)(&zldev->multiop_lock); - - /* Read output configuration */ - rc = zl3073x_mb_op(zldev, ZL_REG_OUTPUT_MB_SEM, ZL_OUTPUT_MB_SEM_RD, - ZL_REG_OUTPUT_MB_MASK, BIT(index)); - if (rc) - return rc; - - /* Read output mode */ - rc = zl3073x_read_u8(zldev, ZL_REG_OUTPUT_MODE, &out->mode); - if (rc) - return rc; - - dev_dbg(zldev->dev, "OUT%u has signal format 0x%02x\n", index, - zl3073x_out_signal_format_get(zldev, index)); - - return rc; -} - -/** - * zl3073x_synth_state_fetch - get synth state - * @zldev: pointer to zl3073x_dev structure - * @index: synth index to fetch state for - * - * Function fetches state of the given synthesizer and stores it for later use. - * - * Return: 0 on success, <0 on error - */ -static int -zl3073x_synth_state_fetch(struct zl3073x_dev *zldev, u8 index) -{ - struct zl3073x_synth *synth = &zldev->synth[index]; - int rc; - - /* Read synth control register */ - rc = zl3073x_read_u8(zldev, ZL_REG_SYNTH_CTRL(index), &synth->ctrl); - if (rc) - return rc; - - guard(mutex)(&zldev->multiop_lock); - - /* Read synth configuration */ - rc = zl3073x_mb_op(zldev, ZL_REG_SYNTH_MB_SEM, ZL_SYNTH_MB_SEM_RD, - ZL_REG_SYNTH_MB_MASK, BIT(index)); - if (rc) - return rc; - - /* The output frequency is determined by the following formula: - * base * multiplier * numerator / denominator - * - * Read registers with these values - */ - rc = zl3073x_read_u16(zldev, ZL_REG_SYNTH_FREQ_BASE, &synth->freq_base); - if (rc) - return rc; - - rc = zl3073x_read_u32(zldev, ZL_REG_SYNTH_FREQ_MULT, &synth->freq_mult); - if (rc) - return rc; - - rc = zl3073x_read_u16(zldev, ZL_REG_SYNTH_FREQ_M, &synth->freq_m); - if (rc) - return rc; - - rc = zl3073x_read_u16(zldev, ZL_REG_SYNTH_FREQ_N, &synth->freq_n); - if (rc) - return rc; - - /* Check denominator for zero to avoid div by 0 */ - if (!synth->freq_n) { - dev_err(zldev->dev, - "Zero divisor for SYNTH%u retrieved from device\n", - index); - return -EINVAL; - } - - dev_dbg(zldev->dev, "SYNTH%u frequency: %u Hz\n", index, - zl3073x_synth_freq_get(zldev, index)); - - return rc; -} - static int zl3073x_dev_state_fetch(struct zl3073x_dev *zldev) { diff --git a/drivers/dpll/zl3073x/core.h b/drivers/dpll/zl3073x/core.h index 51d0fd6cfabf..fe779fc77dd0 100644 --- a/drivers/dpll/zl3073x/core.h +++ b/drivers/dpll/zl3073x/core.h @@ -9,7 +9,10 @@ #include #include +#include "out.h" +#include "ref.h" #include "regs.h" +#include "synth.h" struct device; struct regmap; @@ -27,42 +30,6 @@ struct zl3073x_dpll; #define ZL3073X_NUM_PINS (ZL3073X_NUM_INPUT_PINS + \ ZL3073X_NUM_OUTPUT_PINS) -/** - * struct zl3073x_ref - input reference invariant info - * @ffo: current fractional frequency offset - * @config: reference config - */ -struct zl3073x_ref { - s64 ffo; - u8 config; -}; - -/** - * struct zl3073x_out - output invariant info - * @ctrl: output control - * @mode: output mode - */ -struct zl3073x_out { - u8 ctrl; - u8 mode; -}; - -/** - * struct zl3073x_synth - synthesizer invariant info - * @freq_mult: frequency multiplier - * @freq_base: frequency base - * @freq_m: frequency numerator - * @freq_n: frequency denominator - * @ctrl: synth control - */ -struct zl3073x_synth { - u32 freq_mult; - u16 freq_base; - u16 freq_m; - u16 freq_n; - u8 ctrl; -}; - /** * struct zl3073x_dev - zl3073x device * @dev: pointer to device @@ -175,7 +142,6 @@ int zl3073x_write_hwreg_seq(struct zl3073x_dev *zldev, * Misc operations *****************/ -int zl3073x_ref_freq_factorize(u32 freq, u16 *base, u16 *mult); int zl3073x_ref_phase_offsets_update(struct zl3073x_dev *zldev, int channel); static inline bool @@ -217,181 +183,188 @@ zl3073x_output_pin_out_get(u8 id) } /** - * zl3073x_ref_ffo_get - get current fractional frequency offset + * zl3073x_dev_ref_ffo_get - get current fractional frequency offset * @zldev: pointer to zl3073x device * @index: input reference index * * Return: the latest measured fractional frequency offset */ static inline s64 -zl3073x_ref_ffo_get(struct zl3073x_dev *zldev, u8 index) +zl3073x_dev_ref_ffo_get(struct zl3073x_dev *zldev, u8 index) { - return zldev->ref[index].ffo; + const struct zl3073x_ref *ref = zl3073x_ref_state_get(zldev, index); + + return zl3073x_ref_ffo_get(ref); } /** - * zl3073x_ref_is_diff - check if the given input reference is differential + * zl3073x_dev_ref_is_diff - check if the given input reference is differential * @zldev: pointer to zl3073x device * @index: input reference index * * Return: true if reference is differential, false if reference is single-ended */ static inline bool -zl3073x_ref_is_diff(struct zl3073x_dev *zldev, u8 index) +zl3073x_dev_ref_is_diff(struct zl3073x_dev *zldev, u8 index) { - if (FIELD_GET(ZL_REF_CONFIG_DIFF_EN, zldev->ref[index].config)) - return true; + const struct zl3073x_ref *ref = zl3073x_ref_state_get(zldev, index); - return false; + return zl3073x_ref_is_diff(ref); } /** - * zl3073x_ref_is_enabled - check if the given input reference is enabled + * zl3073x_dev_ref_is_enabled - check if the given input reference is enabled * @zldev: pointer to zl3073x device * @index: input reference index * * Return: true if input refernce is enabled, false otherwise */ static inline bool -zl3073x_ref_is_enabled(struct zl3073x_dev *zldev, u8 index) +zl3073x_dev_ref_is_enabled(struct zl3073x_dev *zldev, u8 index) { - if (FIELD_GET(ZL_REF_CONFIG_ENABLE, zldev->ref[index].config)) - return true; + const struct zl3073x_ref *ref = zl3073x_ref_state_get(zldev, index); - return false; + return zl3073x_ref_is_enabled(ref); } /** - * zl3073x_synth_dpll_get - get DPLL ID the synth is driven by + * zl3073x_dev_synth_dpll_get - get DPLL ID the synth is driven by * @zldev: pointer to zl3073x device * @index: synth index * * Return: ID of DPLL the given synthetizer is driven by */ static inline u8 -zl3073x_synth_dpll_get(struct zl3073x_dev *zldev, u8 index) +zl3073x_dev_synth_dpll_get(struct zl3073x_dev *zldev, u8 index) { - return FIELD_GET(ZL_SYNTH_CTRL_DPLL_SEL, zldev->synth[index].ctrl); + const struct zl3073x_synth *synth; + + synth = zl3073x_synth_state_get(zldev, index); + return zl3073x_synth_dpll_get(synth); } /** - * zl3073x_synth_freq_get - get synth current freq + * zl3073x_dev_synth_freq_get - get synth current freq * @zldev: pointer to zl3073x device * @index: synth index * * Return: frequency of given synthetizer */ static inline u32 -zl3073x_synth_freq_get(struct zl3073x_dev *zldev, u8 index) +zl3073x_dev_synth_freq_get(struct zl3073x_dev *zldev, u8 index) { - struct zl3073x_synth *synth = &zldev->synth[index]; + const struct zl3073x_synth *synth; - return mul_u64_u32_div(synth->freq_base * synth->freq_m, - synth->freq_mult, synth->freq_n); + synth = zl3073x_synth_state_get(zldev, index); + return zl3073x_synth_freq_get(synth); } /** - * zl3073x_synth_is_enabled - check if the given synth is enabled + * zl3073x_dev_synth_is_enabled - check if the given synth is enabled * @zldev: pointer to zl3073x device * @index: synth index * * Return: true if synth is enabled, false otherwise */ static inline bool -zl3073x_synth_is_enabled(struct zl3073x_dev *zldev, u8 index) +zl3073x_dev_synth_is_enabled(struct zl3073x_dev *zldev, u8 index) { - return FIELD_GET(ZL_SYNTH_CTRL_EN, zldev->synth[index].ctrl); + const struct zl3073x_synth *synth; + + synth = zl3073x_synth_state_get(zldev, index); + return zl3073x_synth_is_enabled(synth); } /** - * zl3073x_out_synth_get - get synth connected to given output + * zl3073x_dev_out_synth_get - get synth connected to given output * @zldev: pointer to zl3073x device * @index: output index * * Return: index of synth connected to given output. */ static inline u8 -zl3073x_out_synth_get(struct zl3073x_dev *zldev, u8 index) +zl3073x_dev_out_synth_get(struct zl3073x_dev *zldev, u8 index) { - return FIELD_GET(ZL_OUTPUT_CTRL_SYNTH_SEL, zldev->out[index].ctrl); + const struct zl3073x_out *out = zl3073x_out_state_get(zldev, index); + + return zl3073x_out_synth_get(out); } /** - * zl3073x_out_is_enabled - check if the given output is enabled + * zl3073x_dev_out_is_enabled - check if the given output is enabled * @zldev: pointer to zl3073x device * @index: output index * * Return: true if the output is enabled, false otherwise */ static inline bool -zl3073x_out_is_enabled(struct zl3073x_dev *zldev, u8 index) +zl3073x_dev_out_is_enabled(struct zl3073x_dev *zldev, u8 index) { - u8 synth; + const struct zl3073x_out *out = zl3073x_out_state_get(zldev, index); + const struct zl3073x_synth *synth; + u8 synth_id; /* Output is enabled only if associated synth is enabled */ - synth = zl3073x_out_synth_get(zldev, index); - if (!zl3073x_synth_is_enabled(zldev, synth)) - return false; + synth_id = zl3073x_out_synth_get(out); + synth = zl3073x_synth_state_get(zldev, synth_id); - return FIELD_GET(ZL_OUTPUT_CTRL_EN, zldev->out[index].ctrl); + return zl3073x_synth_is_enabled(synth) && zl3073x_out_is_enabled(out); } /** - * zl3073x_out_signal_format_get - get output signal format + * zl3073x_dev_out_signal_format_get - get output signal format * @zldev: pointer to zl3073x device * @index: output index * * Return: signal format of given output */ static inline u8 -zl3073x_out_signal_format_get(struct zl3073x_dev *zldev, u8 index) +zl3073x_dev_out_signal_format_get(struct zl3073x_dev *zldev, u8 index) { - return FIELD_GET(ZL_OUTPUT_MODE_SIGNAL_FORMAT, zldev->out[index].mode); + const struct zl3073x_out *out = zl3073x_out_state_get(zldev, index); + + return zl3073x_out_signal_format_get(out); } /** - * zl3073x_out_dpll_get - get DPLL ID the output is driven by + * zl3073x_dev_out_dpll_get - get DPLL ID the output is driven by * @zldev: pointer to zl3073x device * @index: output index * * Return: ID of DPLL the given output is driven by */ static inline -u8 zl3073x_out_dpll_get(struct zl3073x_dev *zldev, u8 index) +u8 zl3073x_dev_out_dpll_get(struct zl3073x_dev *zldev, u8 index) { - u8 synth; + const struct zl3073x_out *out = zl3073x_out_state_get(zldev, index); + const struct zl3073x_synth *synth; + u8 synth_id; /* Get synthesizer connected to given output */ - synth = zl3073x_out_synth_get(zldev, index); + synth_id = zl3073x_out_synth_get(out); + synth = zl3073x_synth_state_get(zldev, synth_id); /* Return DPLL that drives the synth */ - return zl3073x_synth_dpll_get(zldev, synth); + return zl3073x_synth_dpll_get(synth); } /** - * zl3073x_out_is_diff - check if the given output is differential + * zl3073x_dev_out_is_diff - check if the given output is differential * @zldev: pointer to zl3073x device * @index: output index * * Return: true if output is differential, false if output is single-ended */ static inline bool -zl3073x_out_is_diff(struct zl3073x_dev *zldev, u8 index) +zl3073x_dev_out_is_diff(struct zl3073x_dev *zldev, u8 index) { - switch (zl3073x_out_signal_format_get(zldev, index)) { - case ZL_OUTPUT_MODE_SIGNAL_FORMAT_LVDS: - case ZL_OUTPUT_MODE_SIGNAL_FORMAT_DIFF: - case ZL_OUTPUT_MODE_SIGNAL_FORMAT_LOWVCM: - return true; - default: - break; - } + const struct zl3073x_out *out = zl3073x_out_state_get(zldev, index); - return false; + return zl3073x_out_is_diff(out); } /** - * zl3073x_output_pin_is_enabled - check if the given output pin is enabled + * zl3073x_dev_output_pin_is_enabled - check if the given output pin is enabled * @zldev: pointer to zl3073x device * @id: output pin id * @@ -401,16 +374,21 @@ zl3073x_out_is_diff(struct zl3073x_dev *zldev, u8 index) * Return: true if output pin is enabled, false if output pin is disabled */ static inline bool -zl3073x_output_pin_is_enabled(struct zl3073x_dev *zldev, u8 id) +zl3073x_dev_output_pin_is_enabled(struct zl3073x_dev *zldev, u8 id) { - u8 output = zl3073x_output_pin_out_get(id); + u8 out_id = zl3073x_output_pin_out_get(id); + const struct zl3073x_out *out; - /* Check if the whole output is enabled */ - if (!zl3073x_out_is_enabled(zldev, output)) + out = zl3073x_out_state_get(zldev, out_id); + + /* Check if the output is enabled - call _dev_ helper that + * additionally checks for attached synth enablement. + */ + if (!zl3073x_dev_out_is_enabled(zldev, out_id)) return false; /* Check signal format */ - switch (zl3073x_out_signal_format_get(zldev, output)) { + switch (zl3073x_out_signal_format_get(out)) { case ZL_OUTPUT_MODE_SIGNAL_FORMAT_DISABLED: /* Both output pins are disabled by signal format */ return false; diff --git a/drivers/dpll/zl3073x/dpll.c b/drivers/dpll/zl3073x/dpll.c index d90150671d37..62996f26e065 100644 --- a/drivers/dpll/zl3073x/dpll.c +++ b/drivers/dpll/zl3073x/dpll.c @@ -967,7 +967,7 @@ zl3073x_dpll_output_pin_esync_get(const struct dpll_pin *dpll_pin, * for N-division is also used for the esync divider so both cannot * be used. */ - switch (zl3073x_out_signal_format_get(zldev, out)) { + switch (zl3073x_dev_out_signal_format_get(zldev, out)) { case ZL_OUTPUT_MODE_SIGNAL_FORMAT_2_NDIV: case ZL_OUTPUT_MODE_SIGNAL_FORMAT_2_NDIV_INV: return -EOPNOTSUPP; @@ -1001,10 +1001,10 @@ zl3073x_dpll_output_pin_esync_get(const struct dpll_pin *dpll_pin, } /* Get synth attached to output pin */ - synth = zl3073x_out_synth_get(zldev, out); + synth = zl3073x_dev_out_synth_get(zldev, out); /* Get synth frequency */ - synth_freq = zl3073x_synth_freq_get(zldev, synth); + synth_freq = zl3073x_dev_synth_freq_get(zldev, synth); clock_type = FIELD_GET(ZL_OUTPUT_MODE_CLOCK_TYPE, output_mode); if (clock_type != ZL_OUTPUT_MODE_CLOCK_TYPE_ESYNC) { @@ -1078,7 +1078,7 @@ zl3073x_dpll_output_pin_esync_set(const struct dpll_pin *dpll_pin, * for N-division is also used for the esync divider so both cannot * be used. */ - switch (zl3073x_out_signal_format_get(zldev, out)) { + switch (zl3073x_dev_out_signal_format_get(zldev, out)) { case ZL_OUTPUT_MODE_SIGNAL_FORMAT_2_NDIV: case ZL_OUTPUT_MODE_SIGNAL_FORMAT_2_NDIV_INV: return -EOPNOTSUPP; @@ -1117,10 +1117,10 @@ zl3073x_dpll_output_pin_esync_set(const struct dpll_pin *dpll_pin, goto write_mailbox; /* Get synth attached to output pin */ - synth = zl3073x_out_synth_get(zldev, out); + synth = zl3073x_dev_out_synth_get(zldev, out); /* Get synth frequency */ - synth_freq = zl3073x_synth_freq_get(zldev, synth); + synth_freq = zl3073x_dev_synth_freq_get(zldev, synth); rc = zl3073x_read_u32(zldev, ZL_REG_OUTPUT_DIV, &output_div); if (rc) @@ -1172,8 +1172,8 @@ zl3073x_dpll_output_pin_frequency_get(const struct dpll_pin *dpll_pin, int rc; out = zl3073x_output_pin_out_get(pin->id); - synth = zl3073x_out_synth_get(zldev, out); - synth_freq = zl3073x_synth_freq_get(zldev, synth); + synth = zl3073x_dev_out_synth_get(zldev, out); + synth_freq = zl3073x_dev_synth_freq_get(zldev, synth); guard(mutex)(&zldev->multiop_lock); @@ -1195,7 +1195,7 @@ zl3073x_dpll_output_pin_frequency_get(const struct dpll_pin *dpll_pin, } /* Read used signal format for the given output */ - signal_format = zl3073x_out_signal_format_get(zldev, out); + signal_format = zl3073x_dev_out_signal_format_get(zldev, out); switch (signal_format) { case ZL_OUTPUT_MODE_SIGNAL_FORMAT_2_NDIV: @@ -1263,12 +1263,12 @@ zl3073x_dpll_output_pin_frequency_set(const struct dpll_pin *dpll_pin, int rc; out = zl3073x_output_pin_out_get(pin->id); - synth = zl3073x_out_synth_get(zldev, out); - synth_freq = zl3073x_synth_freq_get(zldev, synth); + synth = zl3073x_dev_out_synth_get(zldev, out); + synth_freq = zl3073x_dev_synth_freq_get(zldev, synth); new_div = synth_freq / (u32)frequency; /* Get used signal format for the given output */ - signal_format = zl3073x_out_signal_format_get(zldev, out); + signal_format = zl3073x_dev_out_signal_format_get(zldev, out); guard(mutex)(&zldev->multiop_lock); @@ -1856,8 +1856,8 @@ zl3073x_dpll_pin_is_registrable(struct zl3073x_dpll *zldpll, if (zldpll->refsel_mode == ZL_DPLL_MODE_REFSEL_MODE_NCO) return false; - is_diff = zl3073x_ref_is_diff(zldev, ref); - is_enabled = zl3073x_ref_is_enabled(zldev, ref); + is_diff = zl3073x_dev_ref_is_diff(zldev, ref); + is_enabled = zl3073x_dev_ref_is_enabled(zldev, ref); } else { /* Output P&N pair shares single HW output */ u8 out = zl3073x_output_pin_out_get(index); @@ -1865,7 +1865,7 @@ zl3073x_dpll_pin_is_registrable(struct zl3073x_dpll *zldpll, name = "OUT"; /* Skip the pin if it is connected to different DPLL channel */ - if (zl3073x_out_dpll_get(zldev, out) != zldpll->id) { + if (zl3073x_dev_out_dpll_get(zldev, out) != zldpll->id) { dev_dbg(zldev->dev, "%s%u is driven by different DPLL\n", name, out); @@ -1873,8 +1873,8 @@ zl3073x_dpll_pin_is_registrable(struct zl3073x_dpll *zldpll, return false; } - is_diff = zl3073x_out_is_diff(zldev, out); - is_enabled = zl3073x_output_pin_is_enabled(zldev, index); + is_diff = zl3073x_dev_out_is_diff(zldev, out); + is_enabled = zl3073x_dev_output_pin_is_enabled(zldev, index); } /* Skip N-pin if the corresponding input/output is differential */ @@ -2124,7 +2124,7 @@ zl3073x_dpll_pin_ffo_check(struct zl3073x_dpll_pin *pin) return false; /* Get the latest measured ref's ffo */ - ffo = zl3073x_ref_ffo_get(zldev, ref); + ffo = zl3073x_dev_ref_ffo_get(zldev, ref); /* Compare with previous value */ if (pin->freq_offset != ffo) { diff --git a/drivers/dpll/zl3073x/out.c b/drivers/dpll/zl3073x/out.c new file mode 100644 index 000000000000..a48f6917b39f --- /dev/null +++ b/drivers/dpll/zl3073x/out.c @@ -0,0 +1,67 @@ +// SPDX-License-Identifier: GPL-2.0-only + +#include +#include +#include +#include +#include +#include + +#include "core.h" +#include "out.h" + +/** + * zl3073x_out_state_fetch - fetch output state from hardware + * @zldev: pointer to zl3073x_dev structure + * @index: output index to fetch state for + * + * Function fetches state of the given output from hardware and stores it + * for later use. + * + * Return: 0 on success, <0 on error + */ +int zl3073x_out_state_fetch(struct zl3073x_dev *zldev, u8 index) +{ + struct zl3073x_out *out = &zldev->out[index]; + int rc; + + /* Read output configuration */ + rc = zl3073x_read_u8(zldev, ZL_REG_OUTPUT_CTRL(index), &out->ctrl); + if (rc) + return rc; + + dev_dbg(zldev->dev, "OUT%u is %s and connected to SYNTH%u\n", index, + str_enabled_disabled(zl3073x_out_is_enabled(out)), + zl3073x_out_synth_get(out)); + + guard(mutex)(&zldev->multiop_lock); + + /* Read output configuration */ + rc = zl3073x_mb_op(zldev, ZL_REG_OUTPUT_MB_SEM, ZL_OUTPUT_MB_SEM_RD, + ZL_REG_OUTPUT_MB_MASK, BIT(index)); + if (rc) + return rc; + + /* Read output mode */ + rc = zl3073x_read_u8(zldev, ZL_REG_OUTPUT_MODE, &out->mode); + if (rc) + return rc; + + dev_dbg(zldev->dev, "OUT%u has signal format 0x%02x\n", index, + zl3073x_out_signal_format_get(out)); + + return rc; +} + +/** + * zl3073x_out_state_get - get current output state + * @zldev: pointer to zl3073x_dev structure + * @index: output index to get state for + * + * Return: pointer to given output state + */ +const struct zl3073x_out *zl3073x_out_state_get(struct zl3073x_dev *zldev, + u8 index) +{ + return &zldev->out[index]; +} diff --git a/drivers/dpll/zl3073x/out.h b/drivers/dpll/zl3073x/out.h new file mode 100644 index 000000000000..986aa046221d --- /dev/null +++ b/drivers/dpll/zl3073x/out.h @@ -0,0 +1,80 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ + +#ifndef _ZL3073X_OUT_H +#define _ZL3073X_OUT_H + +#include +#include + +#include "regs.h" + +struct zl3073x_dev; + +/** + * struct zl3073x_out - output state + * @ctrl: output control + * @mode: output mode + */ +struct zl3073x_out { + u8 ctrl; + u8 mode; +}; + +int zl3073x_out_state_fetch(struct zl3073x_dev *zldev, u8 index); +const struct zl3073x_out *zl3073x_out_state_get(struct zl3073x_dev *zldev, + u8 index); + +/** + * zl3073x_out_signal_format_get - get output signal format + * @out: pointer to out state + * + * Return: signal format of given output + */ +static inline u8 zl3073x_out_signal_format_get(const struct zl3073x_out *out) +{ + return FIELD_GET(ZL_OUTPUT_MODE_SIGNAL_FORMAT, out->mode); +} + +/** + * zl3073x_out_is_diff - check if the given output is differential + * @out: pointer to out state + * + * Return: true if output is differential, false if output is single-ended + */ +static inline bool zl3073x_out_is_diff(const struct zl3073x_out *out) +{ + switch (zl3073x_out_signal_format_get(out)) { + case ZL_OUTPUT_MODE_SIGNAL_FORMAT_LVDS: + case ZL_OUTPUT_MODE_SIGNAL_FORMAT_DIFF: + case ZL_OUTPUT_MODE_SIGNAL_FORMAT_LOWVCM: + return true; + default: + break; + } + + return false; +} + +/** + * zl3073x_out_is_enabled - check if the given output is enabled + * @out: pointer to out state + * + * Return: true if output is enabled, false if output is disabled + */ +static inline bool zl3073x_out_is_enabled(const struct zl3073x_out *out) +{ + return !!FIELD_GET(ZL_OUTPUT_CTRL_EN, out->ctrl); +} + +/** + * zl3073x_out_synth_get - get synth connected to given output + * @out: pointer to out state + * + * Return: index of synth connected to given output. + */ +static inline u8 zl3073x_out_synth_get(const struct zl3073x_out *out) +{ + return FIELD_GET(ZL_OUTPUT_CTRL_SYNTH_SEL, out->ctrl); +} + +#endif /* _ZL3073X_OUT_H */ diff --git a/drivers/dpll/zl3073x/prop.c b/drivers/dpll/zl3073x/prop.c index 9e1fca5cdaf1..4ed153087570 100644 --- a/drivers/dpll/zl3073x/prop.c +++ b/drivers/dpll/zl3073x/prop.c @@ -46,10 +46,10 @@ zl3073x_pin_check_freq(struct zl3073x_dev *zldev, enum dpll_pin_direction dir, /* Get output pin synthesizer */ out = zl3073x_output_pin_out_get(id); - synth = zl3073x_out_synth_get(zldev, out); + synth = zl3073x_dev_out_synth_get(zldev, out); /* Get synth frequency */ - synth_freq = zl3073x_synth_freq_get(zldev, synth); + synth_freq = zl3073x_dev_synth_freq_get(zldev, synth); /* Check the frequency divides synth frequency */ if (synth_freq % (u32)freq) @@ -93,13 +93,13 @@ zl3073x_prop_pin_package_label_set(struct zl3073x_dev *zldev, prefix = "REF"; ref = zl3073x_input_pin_ref_get(id); - is_diff = zl3073x_ref_is_diff(zldev, ref); + is_diff = zl3073x_dev_ref_is_diff(zldev, ref); } else { u8 out; prefix = "OUT"; out = zl3073x_output_pin_out_get(id); - is_diff = zl3073x_out_is_diff(zldev, out); + is_diff = zl3073x_dev_out_is_diff(zldev, out); } if (!is_diff) @@ -217,8 +217,8 @@ struct zl3073x_pin_props *zl3073x_pin_props_get(struct zl3073x_dev *zldev, * the synth frequency count. */ out = zl3073x_output_pin_out_get(index); - synth = zl3073x_out_synth_get(zldev, out); - f = 2 * zl3073x_synth_freq_get(zldev, synth); + synth = zl3073x_dev_out_synth_get(zldev, out); + f = 2 * zl3073x_dev_synth_freq_get(zldev, synth); props->dpll_props.phase_gran = f ? div_u64(PSEC_PER_SEC, f) : 1; } diff --git a/drivers/dpll/zl3073x/ref.c b/drivers/dpll/zl3073x/ref.c new file mode 100644 index 000000000000..6abd6288a02a --- /dev/null +++ b/drivers/dpll/zl3073x/ref.c @@ -0,0 +1,112 @@ +// SPDX-License-Identifier: GPL-2.0-only + +#include +#include +#include +#include +#include +#include + +#include "core.h" +#include "ref.h" + +/** + * zl3073x_ref_freq_factorize - factorize given frequency + * @freq: input frequency + * @base: base frequency + * @mult: multiplier + * + * Checks if the given frequency can be factorized using one of the + * supported base frequencies. If so the base frequency and multiplier + * are stored into appropriate parameters if they are not NULL. + * + * Return: 0 on success, -EINVAL if the frequency cannot be factorized + */ +int +zl3073x_ref_freq_factorize(u32 freq, u16 *base, u16 *mult) +{ + static const u16 base_freqs[] = { + 1, 2, 4, 5, 8, 10, 16, 20, 25, 32, 40, 50, 64, 80, 100, 125, + 128, 160, 200, 250, 256, 320, 400, 500, 625, 640, 800, 1000, + 1250, 1280, 1600, 2000, 2500, 3125, 3200, 4000, 5000, 6250, + 6400, 8000, 10000, 12500, 15625, 16000, 20000, 25000, 31250, + 32000, 40000, 50000, 62500, + }; + u32 div; + int i; + + for (i = 0; i < ARRAY_SIZE(base_freqs); i++) { + div = freq / base_freqs[i]; + + if (div <= U16_MAX && (freq % base_freqs[i]) == 0) { + if (base) + *base = base_freqs[i]; + if (mult) + *mult = div; + + return 0; + } + } + + return -EINVAL; +} + +/** + * zl3073x_ref_state_fetch - fetch input reference state from hardware + * @zldev: pointer to zl3073x_dev structure + * @index: input reference index to fetch state for + * + * Function fetches state for the given input reference from hardware and + * stores it for later use. + * + * Return: 0 on success, <0 on error + */ +int zl3073x_ref_state_fetch(struct zl3073x_dev *zldev, u8 index) +{ + struct zl3073x_ref *ref = &zldev->ref[index]; + int rc; + + /* For differential type inputs the N-pin reference shares + * part of the configuration with the P-pin counterpart. + */ + if (zl3073x_is_n_pin(index) && zl3073x_ref_is_diff(ref - 1)) { + struct zl3073x_ref *p_ref = &zldev->ref[index - 1]; + + /* Copy the shared items from the P-pin */ + ref->config = p_ref->config; + + return 0; /* Finish - no non-shared items for now */ + } + + guard(mutex)(&zldev->multiop_lock); + + /* Read reference configuration */ + rc = zl3073x_mb_op(zldev, ZL_REG_REF_MB_SEM, ZL_REF_MB_SEM_RD, + ZL_REG_REF_MB_MASK, BIT(index)); + if (rc) + return rc; + + /* Read ref_config register */ + rc = zl3073x_read_u8(zldev, ZL_REG_REF_CONFIG, &ref->config); + if (rc) + return rc; + + dev_dbg(zldev->dev, "REF%u is %s and configured as %s\n", index, + str_enabled_disabled(zl3073x_ref_is_enabled(ref)), + zl3073x_ref_is_diff(ref) ? "differential" : "single-ended"); + + return rc; +} + +/** + * zl3073x_ref_state_get - get current input reference state + * @zldev: pointer to zl3073x_dev structure + * @index: input reference index to get state for + * + * Return: pointer to given input reference state + */ +const struct zl3073x_ref * +zl3073x_ref_state_get(struct zl3073x_dev *zldev, u8 index) +{ + return &zldev->ref[index]; +} diff --git a/drivers/dpll/zl3073x/ref.h b/drivers/dpll/zl3073x/ref.h new file mode 100644 index 000000000000..e72f2c875087 --- /dev/null +++ b/drivers/dpll/zl3073x/ref.h @@ -0,0 +1,66 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ + +#ifndef _ZL3073X_REF_H +#define _ZL3073X_REF_H + +#include +#include + +#include "regs.h" + +struct zl3073x_dev; + +/** + * struct zl3073x_ref - input reference state + * @ffo: current fractional frequency offset + * @config: reference config + */ +struct zl3073x_ref { + s64 ffo; + u8 config; +}; + +int zl3073x_ref_state_fetch(struct zl3073x_dev *zldev, u8 index); + +const struct zl3073x_ref *zl3073x_ref_state_get(struct zl3073x_dev *zldev, + u8 index); + +int zl3073x_ref_freq_factorize(u32 freq, u16 *base, u16 *mult); + +/** + * zl3073x_ref_ffo_get - get current fractional frequency offset + * @ref: pointer to ref state + * + * Return: the latest measured fractional frequency offset + */ +static inline s64 +zl3073x_ref_ffo_get(const struct zl3073x_ref *ref) +{ + return ref->ffo; +} + +/** + * zl3073x_ref_is_diff - check if the given input reference is differential + * @ref: pointer to ref state + * + * Return: true if reference is differential, false if reference is single-ended + */ +static inline bool +zl3073x_ref_is_diff(const struct zl3073x_ref *ref) +{ + return !!FIELD_GET(ZL_REF_CONFIG_DIFF_EN, ref->config); +} + +/** + * zl3073x_ref_is_enabled - check if the given input reference is enabled + * @ref: pointer to ref state + * + * Return: true if input refernce is enabled, false otherwise + */ +static inline bool +zl3073x_ref_is_enabled(const struct zl3073x_ref *ref) +{ + return !!FIELD_GET(ZL_REF_CONFIG_ENABLE, ref->config); +} + +#endif /* _ZL3073X_REF_H */ diff --git a/drivers/dpll/zl3073x/synth.c b/drivers/dpll/zl3073x/synth.c new file mode 100644 index 000000000000..da839572dab2 --- /dev/null +++ b/drivers/dpll/zl3073x/synth.c @@ -0,0 +1,87 @@ +// SPDX-License-Identifier: GPL-2.0-only + +#include +#include +#include +#include +#include +#include + +#include "core.h" +#include "synth.h" + +/** + * zl3073x_synth_state_fetch - fetch synth state from hardware + * @zldev: pointer to zl3073x_dev structure + * @index: synth index to fetch state for + * + * Function fetches state of the given synthesizer from the hardware and + * stores it for later use. + * + * Return: 0 on success, <0 on error + */ +int zl3073x_synth_state_fetch(struct zl3073x_dev *zldev, u8 index) +{ + struct zl3073x_synth *synth = &zldev->synth[index]; + int rc; + + /* Read synth control register */ + rc = zl3073x_read_u8(zldev, ZL_REG_SYNTH_CTRL(index), &synth->ctrl); + if (rc) + return rc; + + guard(mutex)(&zldev->multiop_lock); + + /* Read synth configuration */ + rc = zl3073x_mb_op(zldev, ZL_REG_SYNTH_MB_SEM, ZL_SYNTH_MB_SEM_RD, + ZL_REG_SYNTH_MB_MASK, BIT(index)); + if (rc) + return rc; + + /* The output frequency is determined by the following formula: + * base * multiplier * numerator / denominator + * + * Read registers with these values + */ + rc = zl3073x_read_u16(zldev, ZL_REG_SYNTH_FREQ_BASE, &synth->freq_base); + if (rc) + return rc; + + rc = zl3073x_read_u32(zldev, ZL_REG_SYNTH_FREQ_MULT, &synth->freq_mult); + if (rc) + return rc; + + rc = zl3073x_read_u16(zldev, ZL_REG_SYNTH_FREQ_M, &synth->freq_m); + if (rc) + return rc; + + rc = zl3073x_read_u16(zldev, ZL_REG_SYNTH_FREQ_N, &synth->freq_n); + if (rc) + return rc; + + /* Check denominator for zero to avoid div by 0 */ + if (!synth->freq_n) { + dev_err(zldev->dev, + "Zero divisor for SYNTH%u retrieved from device\n", + index); + return -EINVAL; + } + + dev_dbg(zldev->dev, "SYNTH%u frequency: %u Hz\n", index, + zl3073x_synth_freq_get(synth)); + + return rc; +} + +/** + * zl3073x_synth_state_get - get current synth state + * @zldev: pointer to zl3073x_dev structure + * @index: synth index to get state for + * + * Return: pointer to given synth state + */ +const struct zl3073x_synth *zl3073x_synth_state_get(struct zl3073x_dev *zldev, + u8 index) +{ + return &zldev->synth[index]; +} diff --git a/drivers/dpll/zl3073x/synth.h b/drivers/dpll/zl3073x/synth.h new file mode 100644 index 000000000000..6c55eb8a888c --- /dev/null +++ b/drivers/dpll/zl3073x/synth.h @@ -0,0 +1,72 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ + +#ifndef _ZL3073X_SYNTH_H +#define _ZL3073X_SYNTH_H + +#include +#include +#include + +#include "regs.h" + +struct zl3073x_dev; + +/** + * struct zl3073x_synth - synthesizer state + * @freq_mult: frequency multiplier + * @freq_base: frequency base + * @freq_m: frequency numerator + * @freq_n: frequency denominator + * @ctrl: synth control + */ +struct zl3073x_synth { + u32 freq_mult; + u16 freq_base; + u16 freq_m; + u16 freq_n; + u8 ctrl; +}; + +int zl3073x_synth_state_fetch(struct zl3073x_dev *zldev, u8 synth_id); + +const struct zl3073x_synth *zl3073x_synth_state_get(struct zl3073x_dev *zldev, + u8 synth_id); + +int zl3073x_synth_state_set(struct zl3073x_dev *zldev, u8 synth_id, + const struct zl3073x_synth *synth); + +/** + * zl3073x_synth_dpll_get - get DPLL ID the synth is driven by + * @synth: pointer to synth state + * + * Return: ID of DPLL the given synthetizer is driven by + */ +static inline u8 zl3073x_synth_dpll_get(const struct zl3073x_synth *synth) +{ + return FIELD_GET(ZL_SYNTH_CTRL_DPLL_SEL, synth->ctrl); +} + +/** + * zl3073x_synth_freq_get - get synth current freq + * @synth: pointer to synth state + * + * Return: frequency of given synthetizer + */ +static inline u32 zl3073x_synth_freq_get(const struct zl3073x_synth *synth) +{ + return mul_u64_u32_div(synth->freq_base * synth->freq_m, + synth->freq_mult, synth->freq_n); +} + +/** + * zl3073x_synth_is_enabled - check if the given synth is enabled + * @synth: pointer to synth state + * + * Return: true if synth is enabled, false otherwise + */ +static inline bool zl3073x_synth_is_enabled(const struct zl3073x_synth *synth) +{ + return FIELD_GET(ZL_SYNTH_CTRL_EN, synth->ctrl); +} + +#endif /* _ZL3073X_SYNTH_H */ From 5534a8202d7cf9b2e4557517745583c43662b6a8 Mon Sep 17 00:00:00 2001 From: Ivan Vecera Date: Thu, 13 Nov 2025 08:41:02 +0100 Subject: [PATCH 0862/1381] dpll: zl3073x: Cache reference monitor status Instead of reading the ZL_REG_REF_MON_STATUS register every time the reference status is needed, cache this value in the zl3073x_ref struct. This is achieved by: * Adding a mon_status field to struct zl3073x_ref * Introducing zl3073x_dev_ref_status_update() to read the status for all references into this new cache field * Calling this update function from the periodic work handler * Adding zl3073x_ref_is_status_ok() and zl3073x_dev_ref_is_status_ok() helpers to check the cached value * Refactoring all callers in dpll.c to use the new zl3073x_dev_ref_is_status_ok() helper, removing direct register reads This change consolidates all status register reads into a single periodic function and reduces I/O bus traffic in dpll callbacks. Reviewed-by: Petr Oros Tested-by: Prathosh Satish Signed-off-by: Ivan Vecera Link: https://patch.msgid.link/20251113074105.141379-4-ivecera@redhat.com Signed-off-by: Jakub Kicinski --- drivers/dpll/zl3073x/core.c | 18 +++++++ drivers/dpll/zl3073x/core.h | 15 ++++++ drivers/dpll/zl3073x/dpll.c | 96 ++++++++----------------------------- drivers/dpll/zl3073x/ref.h | 14 ++++++ 4 files changed, 68 insertions(+), 75 deletions(-) diff --git a/drivers/dpll/zl3073x/core.c b/drivers/dpll/zl3073x/core.c index 2f340f7eb9ec..383e2397dd03 100644 --- a/drivers/dpll/zl3073x/core.c +++ b/drivers/dpll/zl3073x/core.c @@ -591,6 +591,21 @@ zl3073x_dev_state_fetch(struct zl3073x_dev *zldev) return rc; } +static void +zl3073x_dev_ref_status_update(struct zl3073x_dev *zldev) +{ + int i, rc; + + for (i = 0; i < ZL3073X_NUM_REFS; i++) { + rc = zl3073x_read_u8(zldev, ZL_REG_REF_MON_STATUS(i), + &zldev->ref[i].mon_status); + if (rc) + dev_warn(zldev->dev, + "Failed to get REF%u status: %pe\n", i, + ERR_PTR(rc)); + } +} + /** * zl3073x_ref_phase_offsets_update - update reference phase offsets * @zldev: pointer to zl3073x_dev structure @@ -710,6 +725,9 @@ zl3073x_dev_periodic_work(struct kthread_work *work) struct zl3073x_dpll *zldpll; int rc; + /* Update input references status */ + zl3073x_dev_ref_status_update(zldev); + /* Update DPLL-to-connected-ref phase offsets registers */ rc = zl3073x_ref_phase_offsets_update(zldev, -1); if (rc) diff --git a/drivers/dpll/zl3073x/core.h b/drivers/dpll/zl3073x/core.h index fe779fc77dd0..4148580d1f34 100644 --- a/drivers/dpll/zl3073x/core.h +++ b/drivers/dpll/zl3073x/core.h @@ -227,6 +227,21 @@ zl3073x_dev_ref_is_enabled(struct zl3073x_dev *zldev, u8 index) return zl3073x_ref_is_enabled(ref); } +/* + * zl3073x_dev_ref_is_status_ok - check the given input reference status + * @zldev: pointer to zl3073x device + * @index: input reference index + * + * Return: true if the status is ok, false otherwise + */ +static inline bool +zl3073x_dev_ref_is_status_ok(struct zl3073x_dev *zldev, u8 index) +{ + const struct zl3073x_ref *ref = zl3073x_ref_state_get(zldev, index); + + return zl3073x_ref_is_status_ok(ref); +} + /** * zl3073x_dev_synth_dpll_get - get DPLL ID the synth is driven by * @zldev: pointer to zl3073x device diff --git a/drivers/dpll/zl3073x/dpll.c b/drivers/dpll/zl3073x/dpll.c index 62996f26e065..20c921d6f42c 100644 --- a/drivers/dpll/zl3073x/dpll.c +++ b/drivers/dpll/zl3073x/dpll.c @@ -497,19 +497,10 @@ zl3073x_dpll_connected_ref_get(struct zl3073x_dpll *zldpll, u8 *ref) if (rc) return rc; - if (ZL3073X_DPLL_REF_IS_VALID(*ref)) { - u8 ref_status; - - /* Read the reference monitor status */ - rc = zl3073x_read_u8(zldev, ZL_REG_REF_MON_STATUS(*ref), - &ref_status); - if (rc) - return rc; - - /* If the monitor indicates an error nothing is connected */ - if (ref_status != ZL_REF_MON_STATUS_OK) - *ref = ZL3073X_DPLL_REF_NONE; - } + /* If the monitor indicates an error nothing is connected */ + if (ZL3073X_DPLL_REF_IS_VALID(*ref) && + !zl3073x_dev_ref_is_status_ok(zldev, *ref)) + *ref = ZL3073X_DPLL_REF_NONE; return 0; } @@ -524,7 +515,7 @@ zl3073x_dpll_input_pin_phase_offset_get(const struct dpll_pin *dpll_pin, struct zl3073x_dpll *zldpll = dpll_priv; struct zl3073x_dev *zldev = zldpll->dev; struct zl3073x_dpll_pin *pin = pin_priv; - u8 conn_ref, ref, ref_status; + u8 conn_ref, ref; s64 ref_phase; int rc; @@ -537,21 +528,9 @@ zl3073x_dpll_input_pin_phase_offset_get(const struct dpll_pin *dpll_pin, * monitor feature is disabled. */ ref = zl3073x_input_pin_ref_get(pin->id); - if (!zldpll->phase_monitor && ref != conn_ref) { + if ((!zldpll->phase_monitor && ref != conn_ref) || + !zl3073x_dev_ref_is_status_ok(zldev, ref)) { *phase_offset = 0; - - return 0; - } - - /* Get this pin monitor status */ - rc = zl3073x_read_u8(zldev, ZL_REG_REF_MON_STATUS(ref), &ref_status); - if (rc) - return rc; - - /* Report phase offset only if the input pin signal is present */ - if (ref_status != ZL_REF_MON_STATUS_OK) { - *phase_offset = 0; - return 0; } @@ -777,7 +756,7 @@ zl3073x_dpll_ref_state_get(struct zl3073x_dpll_pin *pin, { struct zl3073x_dpll *zldpll = pin->dpll; struct zl3073x_dev *zldev = zldpll->dev; - u8 ref, ref_conn, status; + u8 ref, ref_conn; int rc; ref = zl3073x_input_pin_ref_get(pin->id); @@ -797,20 +776,9 @@ zl3073x_dpll_ref_state_get(struct zl3073x_dpll_pin *pin, * pin as selectable. */ if (zldpll->refsel_mode == ZL_DPLL_MODE_REFSEL_MODE_AUTO && - pin->selectable) { - /* Read reference monitor status */ - rc = zl3073x_read_u8(zldev, ZL_REG_REF_MON_STATUS(ref), - &status); - if (rc) - return rc; - - /* If the monitor indicates errors report the reference - * as disconnected - */ - if (status == ZL_REF_MON_STATUS_OK) { - *state = DPLL_PIN_STATE_SELECTABLE; - return 0; - } + zl3073x_dev_ref_is_status_ok(zldev, ref) && pin->selectable) { + *state = DPLL_PIN_STATE_SELECTABLE; + return 0; } /* Otherwise report the pin as disconnected */ @@ -2036,37 +2004,23 @@ zl3073x_dpll_pin_phase_offset_check(struct zl3073x_dpll_pin *pin) ref = zl3073x_input_pin_ref_get(pin->id); + /* No phase offset if the ref monitor reports signal errors */ + if (!zl3073x_dev_ref_is_status_ok(zldev, ref)) + return false; + /* Select register to read phase offset value depending on pin and * phase monitor state: * 1) For connected pin use dpll_phase_err_data register * 2) For other pins use appropriate ref_phase register if the phase - * monitor feature is enabled and reference monitor does not - * report signal errors for given input pin + * monitor feature is enabled. */ - if (pin->pin_state == DPLL_PIN_STATE_CONNECTED) { + if (pin->pin_state == DPLL_PIN_STATE_CONNECTED) reg = ZL_REG_DPLL_PHASE_ERR_DATA(zldpll->id); - } else if (zldpll->phase_monitor) { - u8 status; - - /* Get reference monitor status */ - rc = zl3073x_read_u8(zldev, ZL_REG_REF_MON_STATUS(ref), - &status); - if (rc) { - dev_err(zldev->dev, - "Failed to read %s refmon status: %pe\n", - pin->label, ERR_PTR(rc)); - - return false; - } - - if (status != ZL_REF_MON_STATUS_OK) - return false; - + else if (zldpll->phase_monitor) reg = ZL_REG_REF_PHASE(ref); - } else { + else /* The pin is not connected or phase monitor disabled */ return false; - } /* Read measured phase offset value */ rc = zl3073x_read_u48(zldev, reg, &phase_offset); @@ -2105,22 +2059,14 @@ zl3073x_dpll_pin_ffo_check(struct zl3073x_dpll_pin *pin) { struct zl3073x_dpll *zldpll = pin->dpll; struct zl3073x_dev *zldev = zldpll->dev; - u8 ref, status; s64 ffo; - int rc; + u8 ref; /* Get reference monitor status */ ref = zl3073x_input_pin_ref_get(pin->id); - rc = zl3073x_read_u8(zldev, ZL_REG_REF_MON_STATUS(ref), &status); - if (rc) { - dev_err(zldev->dev, "Failed to read %s refmon status: %pe\n", - pin->label, ERR_PTR(rc)); - - return false; - } /* Do not report ffo changes if the reference monitor report errors */ - if (status != ZL_REF_MON_STATUS_OK) + if (!zl3073x_dev_ref_is_status_ok(zldev, ref)) return false; /* Get the latest measured ref's ffo */ diff --git a/drivers/dpll/zl3073x/ref.h b/drivers/dpll/zl3073x/ref.h index e72f2c875087..c4931e545d24 100644 --- a/drivers/dpll/zl3073x/ref.h +++ b/drivers/dpll/zl3073x/ref.h @@ -14,10 +14,12 @@ struct zl3073x_dev; * struct zl3073x_ref - input reference state * @ffo: current fractional frequency offset * @config: reference config + * @mon_status: reference monitor status */ struct zl3073x_ref { s64 ffo; u8 config; + u8 mon_status; }; int zl3073x_ref_state_fetch(struct zl3073x_dev *zldev, u8 index); @@ -63,4 +65,16 @@ zl3073x_ref_is_enabled(const struct zl3073x_ref *ref) return !!FIELD_GET(ZL_REF_CONFIG_ENABLE, ref->config); } +/** + * zl3073x_ref_is_status_ok - check the given input reference status + * @ref: pointer to ref state + * + * Return: true if the status is ok, false otherwise + */ +static inline bool +zl3073x_ref_is_status_ok(const struct zl3073x_ref *ref) +{ + return ref->mon_status == ZL_REF_MON_STATUS_OK; +} + #endif /* _ZL3073X_REF_H */ From 5bc02b190a3fb703bf8cadc4d778fc22cd4d1e78 Mon Sep 17 00:00:00 2001 From: Ivan Vecera Date: Thu, 13 Nov 2025 08:41:03 +0100 Subject: [PATCH 0863/1381] dpll: zl3073x: Cache all reference properties in zl3073x_ref Expand the zl3073x_ref structure to cache all reference-related hardware registers, including frequency components, embedded-sync settings and phase compensation. Previously, these registers were read on-demand from various functions in dpll.c leading to frequent mailbox operations. Modify zl3073x_ref_state_fetch() to read and populate all these new fields at once. Refactor all "getter" functions in dpll.c to read from this new cached state instead of performing direct register access. Remove the standalone zl3073x_dpll_input_ref_frequency_get() helper, as its functionality is now replaced by zl3073x_ref_freq_get() which operates on the cached state and add a corresponding zl3073x_dev_... wrapper. Introduce a new function, zl3073x_ref_state_set(), to handle writing changes back to the hardware. This function compares the provided state with the current cached state and writes *only* the modified register values to the device via a single mailbox sequence before updating the local cache. Refactor all dpll "setter" functions to modify a local copy of the ref state and then call zl3073x_ref_state_set() to commit the changes. As a cleanup, update callers in dpll.c that already have a struct zl3073x_ref * to use the direct helpers instead of the zl3073x_dev_... wrappers. This change centralizes all reference-related register I/O into ref.c, significantly reduces bus traffic, and simplifies the logic in dpll.c. Reviewed-by: Petr Oros Tested-by: Prathosh Satish Signed-off-by: Ivan Vecera Link: https://patch.msgid.link/20251113074105.141379-5-ivecera@redhat.com Signed-off-by: Jakub Kicinski --- drivers/dpll/zl3073x/core.h | 15 ++ drivers/dpll/zl3073x/dpll.c | 308 +++++++++--------------------------- drivers/dpll/zl3073x/ref.c | 94 ++++++++++- drivers/dpll/zl3073x/ref.h | 54 +++++++ 4 files changed, 239 insertions(+), 232 deletions(-) diff --git a/drivers/dpll/zl3073x/core.h b/drivers/dpll/zl3073x/core.h index 4148580d1f34..fe8b70e25d3c 100644 --- a/drivers/dpll/zl3073x/core.h +++ b/drivers/dpll/zl3073x/core.h @@ -197,6 +197,21 @@ zl3073x_dev_ref_ffo_get(struct zl3073x_dev *zldev, u8 index) return zl3073x_ref_ffo_get(ref); } +/** + * zl3073x_dev_ref_freq_get - get input reference frequency + * @zldev: pointer to zl3073x device + * @index: input reference index + * + * Return: frequency of given input reference + */ +static inline u32 +zl3073x_dev_ref_freq_get(struct zl3073x_dev *zldev, u8 index) +{ + const struct zl3073x_ref *ref = zl3073x_ref_state_get(zldev, index); + + return zl3073x_ref_freq_get(ref); +} + /** * zl3073x_dev_ref_is_diff - check if the given input reference is differential * @zldev: pointer to zl3073x device diff --git a/drivers/dpll/zl3073x/dpll.c b/drivers/dpll/zl3073x/dpll.c index 20c921d6f42c..e0d696bc1909 100644 --- a/drivers/dpll/zl3073x/dpll.c +++ b/drivers/dpll/zl3073x/dpll.c @@ -100,60 +100,6 @@ zl3073x_dpll_pin_direction_get(const struct dpll_pin *dpll_pin, void *pin_priv, return 0; } -/** - * zl3073x_dpll_input_ref_frequency_get - get input reference frequency - * @zldpll: pointer to zl3073x_dpll - * @ref_id: reference id - * @frequency: pointer to variable to store frequency - * - * Reads frequency of given input reference. - * - * Return: 0 on success, <0 on error - */ -static int -zl3073x_dpll_input_ref_frequency_get(struct zl3073x_dpll *zldpll, u8 ref_id, - u32 *frequency) -{ - struct zl3073x_dev *zldev = zldpll->dev; - u16 base, mult, num, denom; - int rc; - - guard(mutex)(&zldev->multiop_lock); - - /* Read reference configuration */ - rc = zl3073x_mb_op(zldev, ZL_REG_REF_MB_SEM, ZL_REF_MB_SEM_RD, - ZL_REG_REF_MB_MASK, BIT(ref_id)); - if (rc) - return rc; - - /* Read registers to compute resulting frequency */ - rc = zl3073x_read_u16(zldev, ZL_REG_REF_FREQ_BASE, &base); - if (rc) - return rc; - rc = zl3073x_read_u16(zldev, ZL_REG_REF_FREQ_MULT, &mult); - if (rc) - return rc; - rc = zl3073x_read_u16(zldev, ZL_REG_REF_RATIO_M, &num); - if (rc) - return rc; - rc = zl3073x_read_u16(zldev, ZL_REG_REF_RATIO_N, &denom); - if (rc) - return rc; - - /* Sanity check that HW has not returned zero denominator */ - if (!denom) { - dev_err(zldev->dev, - "Zero divisor for ref %u frequency got from device\n", - ref_id); - return -EINVAL; - } - - /* Compute the frequency */ - *frequency = mul_u64_u32_div(base * mult, num, denom); - - return rc; -} - static int zl3073x_dpll_input_pin_esync_get(const struct dpll_pin *dpll_pin, void *pin_priv, @@ -165,39 +111,15 @@ zl3073x_dpll_input_pin_esync_get(const struct dpll_pin *dpll_pin, struct zl3073x_dpll *zldpll = dpll_priv; struct zl3073x_dev *zldev = zldpll->dev; struct zl3073x_dpll_pin *pin = pin_priv; - u8 ref, ref_sync_ctrl, sync_mode; - u32 esync_div, ref_freq; - int rc; + const struct zl3073x_ref *ref; + u8 ref_id; - /* Get reference frequency */ - ref = zl3073x_input_pin_ref_get(pin->id); - rc = zl3073x_dpll_input_ref_frequency_get(zldpll, pin->id, &ref_freq); - if (rc) - return rc; + ref_id = zl3073x_input_pin_ref_get(pin->id); + ref = zl3073x_ref_state_get(zldev, ref_id); - guard(mutex)(&zldev->multiop_lock); - - /* Read reference configuration into mailbox */ - rc = zl3073x_mb_op(zldev, ZL_REG_REF_MB_SEM, ZL_REF_MB_SEM_RD, - ZL_REG_REF_MB_MASK, BIT(ref)); - if (rc) - return rc; - - /* Get ref sync mode */ - rc = zl3073x_read_u8(zldev, ZL_REG_REF_SYNC_CTRL, &ref_sync_ctrl); - if (rc) - return rc; - - /* Get esync divisor */ - rc = zl3073x_read_u32(zldev, ZL_REG_REF_ESYNC_DIV, &esync_div); - if (rc) - return rc; - - sync_mode = FIELD_GET(ZL_REF_SYNC_CTRL_MODE, ref_sync_ctrl); - - switch (sync_mode) { + switch (FIELD_GET(ZL_REF_SYNC_CTRL_MODE, ref->sync_ctrl)) { case ZL_REF_SYNC_CTRL_MODE_50_50_ESYNC_25_75: - esync->freq = (esync_div == ZL_REF_ESYNC_DIV_1HZ) ? 1 : 0; + esync->freq = ref->esync_n_div == ZL_REF_ESYNC_DIV_1HZ ? 1 : 0; esync->pulse = 25; break; default: @@ -209,7 +131,7 @@ zl3073x_dpll_input_pin_esync_get(const struct dpll_pin *dpll_pin, /* If the pin supports esync control expose its range but only * if the current reference frequency is > 1 Hz. */ - if (pin->esync_control && ref_freq > 1) { + if (pin->esync_control && zl3073x_ref_freq_get(ref) > 1) { esync->range = esync_freq_ranges; esync->range_num = ARRAY_SIZE(esync_freq_ranges); } else { @@ -217,7 +139,7 @@ zl3073x_dpll_input_pin_esync_get(const struct dpll_pin *dpll_pin, esync->range_num = 0; } - return rc; + return 0; } static int @@ -230,22 +152,11 @@ zl3073x_dpll_input_pin_esync_set(const struct dpll_pin *dpll_pin, struct zl3073x_dpll *zldpll = dpll_priv; struct zl3073x_dev *zldev = zldpll->dev; struct zl3073x_dpll_pin *pin = pin_priv; - u8 ref, ref_sync_ctrl, sync_mode; - int rc; + struct zl3073x_ref ref; + u8 ref_id, sync_mode; - guard(mutex)(&zldev->multiop_lock); - - /* Read reference configuration into mailbox */ - ref = zl3073x_input_pin_ref_get(pin->id); - rc = zl3073x_mb_op(zldev, ZL_REG_REF_MB_SEM, ZL_REF_MB_SEM_RD, - ZL_REG_REF_MB_MASK, BIT(ref)); - if (rc) - return rc; - - /* Get ref sync mode */ - rc = zl3073x_read_u8(zldev, ZL_REG_REF_SYNC_CTRL, &ref_sync_ctrl); - if (rc) - return rc; + ref_id = zl3073x_input_pin_ref_get(pin->id); + ref = *zl3073x_ref_state_get(zldev, ref_id); /* Use freq == 0 to disable esync */ if (!freq) @@ -253,25 +164,16 @@ zl3073x_dpll_input_pin_esync_set(const struct dpll_pin *dpll_pin, else sync_mode = ZL_REF_SYNC_CTRL_MODE_50_50_ESYNC_25_75; - ref_sync_ctrl &= ~ZL_REF_SYNC_CTRL_MODE; - ref_sync_ctrl |= FIELD_PREP(ZL_REF_SYNC_CTRL_MODE, sync_mode); - - /* Update ref sync control register */ - rc = zl3073x_write_u8(zldev, ZL_REG_REF_SYNC_CTRL, ref_sync_ctrl); - if (rc) - return rc; + ref.sync_ctrl &= ~ZL_REF_SYNC_CTRL_MODE; + ref.sync_ctrl |= FIELD_PREP(ZL_REF_SYNC_CTRL_MODE, sync_mode); if (freq) { - /* 1 Hz is only supported frequnecy currently */ - rc = zl3073x_write_u32(zldev, ZL_REG_REF_ESYNC_DIV, - ZL_REF_ESYNC_DIV_1HZ); - if (rc) - return rc; + /* 1 Hz is only supported frequency now */ + ref.esync_n_div = ZL_REF_ESYNC_DIV_1HZ; } - /* Commit reference configuration */ - return zl3073x_mb_op(zldev, ZL_REG_REF_MB_SEM, ZL_REF_MB_SEM_WR, - ZL_REG_REF_MB_MASK, BIT(ref)); + /* Update reference configuration */ + return zl3073x_ref_state_set(zldev, ref_id, &ref); } static int @@ -295,17 +197,12 @@ zl3073x_dpll_input_pin_frequency_get(const struct dpll_pin *dpll_pin, { struct zl3073x_dpll *zldpll = dpll_priv; struct zl3073x_dpll_pin *pin = pin_priv; - u32 ref_freq; - u8 ref; - int rc; + u8 ref_id; - /* Read and return ref frequency */ - ref = zl3073x_input_pin_ref_get(pin->id); - rc = zl3073x_dpll_input_ref_frequency_get(zldpll, ref, &ref_freq); - if (!rc) - *frequency = ref_freq; + ref_id = zl3073x_input_pin_ref_get(pin->id); + *frequency = zl3073x_dev_ref_freq_get(zldpll->dev, ref_id); - return rc; + return 0; } static int @@ -318,39 +215,18 @@ zl3073x_dpll_input_pin_frequency_set(const struct dpll_pin *dpll_pin, struct zl3073x_dpll *zldpll = dpll_priv; struct zl3073x_dev *zldev = zldpll->dev; struct zl3073x_dpll_pin *pin = pin_priv; - u16 base, mult; - u8 ref; - int rc; + struct zl3073x_ref ref; + u8 ref_id; - /* Get base frequency and multiplier for the requested frequency */ - rc = zl3073x_ref_freq_factorize(frequency, &base, &mult); - if (rc) - return rc; + /* Get reference state */ + ref_id = zl3073x_input_pin_ref_get(pin->id); + ref = *zl3073x_ref_state_get(zldev, ref_id); - guard(mutex)(&zldev->multiop_lock); + /* Update frequency */ + zl3073x_ref_freq_set(&ref, frequency); - /* Load reference configuration */ - ref = zl3073x_input_pin_ref_get(pin->id); - rc = zl3073x_mb_op(zldev, ZL_REG_REF_MB_SEM, ZL_REF_MB_SEM_RD, - ZL_REG_REF_MB_MASK, BIT(ref)); - - /* Update base frequency, multiplier, numerator & denominator */ - rc = zl3073x_write_u16(zldev, ZL_REG_REF_FREQ_BASE, base); - if (rc) - return rc; - rc = zl3073x_write_u16(zldev, ZL_REG_REF_FREQ_MULT, mult); - if (rc) - return rc; - rc = zl3073x_write_u16(zldev, ZL_REG_REF_RATIO_M, 1); - if (rc) - return rc; - rc = zl3073x_write_u16(zldev, ZL_REG_REF_RATIO_N, 1); - if (rc) - return rc; - - /* Commit reference configuration */ - return zl3073x_mb_op(zldev, ZL_REG_REF_MB_SEM, ZL_REF_MB_SEM_WR, - ZL_REG_REF_MB_MASK, BIT(ref)); + /* Commit reference state */ + return zl3073x_ref_state_set(zldev, ref_id, &ref); } /** @@ -515,21 +391,24 @@ zl3073x_dpll_input_pin_phase_offset_get(const struct dpll_pin *dpll_pin, struct zl3073x_dpll *zldpll = dpll_priv; struct zl3073x_dev *zldev = zldpll->dev; struct zl3073x_dpll_pin *pin = pin_priv; - u8 conn_ref, ref; + const struct zl3073x_ref *ref; + u8 conn_id, ref_id; s64 ref_phase; int rc; /* Get currently connected reference */ - rc = zl3073x_dpll_connected_ref_get(zldpll, &conn_ref); + rc = zl3073x_dpll_connected_ref_get(zldpll, &conn_id); if (rc) return rc; /* Report phase offset only for currently connected pin if the phase - * monitor feature is disabled. + * monitor feature is disabled and only if the input pin signal is + * present. */ - ref = zl3073x_input_pin_ref_get(pin->id); - if ((!zldpll->phase_monitor && ref != conn_ref) || - !zl3073x_dev_ref_is_status_ok(zldev, ref)) { + ref_id = zl3073x_input_pin_ref_get(pin->id); + ref = zl3073x_ref_state_get(zldev, ref_id); + if ((!zldpll->phase_monitor && ref_id != conn_id) || + !zl3073x_ref_is_status_ok(ref)) { *phase_offset = 0; return 0; } @@ -540,20 +419,12 @@ zl3073x_dpll_input_pin_phase_offset_get(const struct dpll_pin *dpll_pin, * the phase offset is modded to the period of the signal * the dpll is locked to. */ - if (ZL3073X_DPLL_REF_IS_VALID(conn_ref) && conn_ref != ref) { + if (ZL3073X_DPLL_REF_IS_VALID(conn_id) && conn_id != ref_id) { u32 conn_freq, ref_freq; - /* Get frequency of connected ref */ - rc = zl3073x_dpll_input_ref_frequency_get(zldpll, conn_ref, - &conn_freq); - if (rc) - return rc; - - /* Get frequency of given ref */ - rc = zl3073x_dpll_input_ref_frequency_get(zldpll, ref, - &ref_freq); - if (rc) - return rc; + /* Get frequency of connected and given ref */ + conn_freq = zl3073x_dev_ref_freq_get(zldev, conn_id); + ref_freq = zl3073x_ref_freq_get(ref); if (conn_freq > ref_freq) { s64 conn_period, div_factor; @@ -580,33 +451,23 @@ zl3073x_dpll_input_pin_phase_adjust_get(const struct dpll_pin *dpll_pin, struct zl3073x_dpll *zldpll = dpll_priv; struct zl3073x_dev *zldev = zldpll->dev; struct zl3073x_dpll_pin *pin = pin_priv; + const struct zl3073x_ref *ref; s64 phase_comp; - u8 ref; - int rc; - - guard(mutex)(&zldev->multiop_lock); + u8 ref_id; /* Read reference configuration */ - ref = zl3073x_input_pin_ref_get(pin->id); - rc = zl3073x_mb_op(zldev, ZL_REG_REF_MB_SEM, ZL_REF_MB_SEM_RD, - ZL_REG_REF_MB_MASK, BIT(ref)); - if (rc) - return rc; - - /* Read current phase offset compensation */ - rc = zl3073x_read_u48(zldev, ZL_REG_REF_PHASE_OFFSET_COMP, &phase_comp); - if (rc) - return rc; + ref_id = zl3073x_input_pin_ref_get(pin->id); + ref = zl3073x_ref_state_get(zldev, ref_id); /* Perform sign extension for 48bit signed value */ - phase_comp = sign_extend64(phase_comp, 47); + phase_comp = sign_extend64(ref->phase_comp, 47); /* Reverse two's complement negation applied during set and convert * to 32bit signed int */ *phase_adjust = (s32)-phase_comp; - return rc; + return 0; } static int @@ -620,32 +481,20 @@ zl3073x_dpll_input_pin_phase_adjust_set(const struct dpll_pin *dpll_pin, struct zl3073x_dpll *zldpll = dpll_priv; struct zl3073x_dev *zldev = zldpll->dev; struct zl3073x_dpll_pin *pin = pin_priv; - s64 phase_comp; - u8 ref; - int rc; + struct zl3073x_ref ref; + u8 ref_id; + + /* Read reference configuration */ + ref_id = zl3073x_input_pin_ref_get(pin->id); + ref = *zl3073x_ref_state_get(zldev, ref_id); /* The value in the register is stored as two's complement negation * of requested value. */ - phase_comp = -phase_adjust; + ref.phase_comp = -phase_adjust; - guard(mutex)(&zldev->multiop_lock); - - /* Read reference configuration */ - ref = zl3073x_input_pin_ref_get(pin->id); - rc = zl3073x_mb_op(zldev, ZL_REG_REF_MB_SEM, ZL_REF_MB_SEM_RD, - ZL_REG_REF_MB_MASK, BIT(ref)); - if (rc) - return rc; - - /* Write the requested value into the compensation register */ - rc = zl3073x_write_u48(zldev, ZL_REG_REF_PHASE_OFFSET_COMP, phase_comp); - if (rc) - return rc; - - /* Commit reference configuration */ - return zl3073x_mb_op(zldev, ZL_REG_REF_MB_SEM, ZL_REF_MB_SEM_WR, - ZL_REG_REF_MB_MASK, BIT(ref)); + /* Update reference configuration */ + return zl3073x_ref_state_set(zldev, ref_id, &ref); } /** @@ -1816,16 +1665,17 @@ zl3073x_dpll_pin_is_registrable(struct zl3073x_dpll *zldpll, const char *name; if (dir == DPLL_PIN_DIRECTION_INPUT) { - u8 ref = zl3073x_input_pin_ref_get(index); - - name = "REF"; + u8 ref_id = zl3073x_input_pin_ref_get(index); + const struct zl3073x_ref *ref; /* Skip the pin if the DPLL is running in NCO mode */ if (zldpll->refsel_mode == ZL_DPLL_MODE_REFSEL_MODE_NCO) return false; - is_diff = zl3073x_dev_ref_is_diff(zldev, ref); - is_enabled = zl3073x_dev_ref_is_enabled(zldev, ref); + name = "REF"; + ref = zl3073x_ref_state_get(zldev, ref_id); + is_diff = zl3073x_ref_is_diff(ref); + is_enabled = zl3073x_ref_is_enabled(ref); } else { /* Output P&N pair shares single HW output */ u8 out = zl3073x_output_pin_out_get(index); @@ -1999,13 +1849,12 @@ zl3073x_dpll_pin_phase_offset_check(struct zl3073x_dpll_pin *pin) struct zl3073x_dev *zldev = zldpll->dev; unsigned int reg; s64 phase_offset; - u8 ref; + u8 ref_id; int rc; - ref = zl3073x_input_pin_ref_get(pin->id); - /* No phase offset if the ref monitor reports signal errors */ - if (!zl3073x_dev_ref_is_status_ok(zldev, ref)) + ref_id = zl3073x_input_pin_ref_get(pin->id); + if (!zl3073x_dev_ref_is_status_ok(zldev, ref_id)) return false; /* Select register to read phase offset value depending on pin and @@ -2017,9 +1866,8 @@ zl3073x_dpll_pin_phase_offset_check(struct zl3073x_dpll_pin *pin) if (pin->pin_state == DPLL_PIN_STATE_CONNECTED) reg = ZL_REG_DPLL_PHASE_ERR_DATA(zldpll->id); else if (zldpll->phase_monitor) - reg = ZL_REG_REF_PHASE(ref); + reg = ZL_REG_REF_PHASE(ref_id); else - /* The pin is not connected or phase monitor disabled */ return false; /* Read measured phase offset value */ @@ -2059,24 +1907,22 @@ zl3073x_dpll_pin_ffo_check(struct zl3073x_dpll_pin *pin) { struct zl3073x_dpll *zldpll = pin->dpll; struct zl3073x_dev *zldev = zldpll->dev; - s64 ffo; - u8 ref; + const struct zl3073x_ref *ref; + u8 ref_id; /* Get reference monitor status */ - ref = zl3073x_input_pin_ref_get(pin->id); + ref_id = zl3073x_input_pin_ref_get(pin->id); + ref = zl3073x_ref_state_get(zldev, ref_id); /* Do not report ffo changes if the reference monitor report errors */ - if (!zl3073x_dev_ref_is_status_ok(zldev, ref)) + if (!zl3073x_ref_is_status_ok(ref)) return false; - /* Get the latest measured ref's ffo */ - ffo = zl3073x_dev_ref_ffo_get(zldev, ref); - /* Compare with previous value */ - if (pin->freq_offset != ffo) { + if (pin->freq_offset != ref->ffo) { dev_dbg(zldev->dev, "%s freq offset changed: %lld -> %lld\n", - pin->label, pin->freq_offset, ffo); - pin->freq_offset = ffo; + pin->label, pin->freq_offset, ref->ffo); + pin->freq_offset = ref->ffo; return true; } diff --git a/drivers/dpll/zl3073x/ref.c b/drivers/dpll/zl3073x/ref.c index 6abd6288a02a..aa2de13effa8 100644 --- a/drivers/dpll/zl3073x/ref.c +++ b/drivers/dpll/zl3073x/ref.c @@ -70,10 +70,17 @@ int zl3073x_ref_state_fetch(struct zl3073x_dev *zldev, u8 index) * part of the configuration with the P-pin counterpart. */ if (zl3073x_is_n_pin(index) && zl3073x_ref_is_diff(ref - 1)) { - struct zl3073x_ref *p_ref = &zldev->ref[index - 1]; + struct zl3073x_ref *p_ref = ref - 1; /* P-pin counterpart*/ /* Copy the shared items from the P-pin */ ref->config = p_ref->config; + ref->esync_n_div = p_ref->esync_n_div; + ref->freq_base = p_ref->freq_base; + ref->freq_mult = p_ref->freq_mult; + ref->freq_ratio_m = p_ref->freq_ratio_m; + ref->freq_ratio_n = p_ref->freq_ratio_n; + ref->phase_comp = p_ref->phase_comp; + ref->sync_ctrl = p_ref->sync_ctrl; return 0; /* Finish - no non-shared items for now */ } @@ -91,6 +98,34 @@ int zl3073x_ref_state_fetch(struct zl3073x_dev *zldev, u8 index) if (rc) return rc; + /* Read frequency related registers */ + rc = zl3073x_read_u16(zldev, ZL_REG_REF_FREQ_BASE, &ref->freq_base); + if (rc) + return rc; + rc = zl3073x_read_u16(zldev, ZL_REG_REF_FREQ_MULT, &ref->freq_mult); + if (rc) + return rc; + rc = zl3073x_read_u16(zldev, ZL_REG_REF_RATIO_M, &ref->freq_ratio_m); + if (rc) + return rc; + rc = zl3073x_read_u16(zldev, ZL_REG_REF_RATIO_N, &ref->freq_ratio_n); + if (rc) + return rc; + + /* Read eSync and N-div rated registers */ + rc = zl3073x_read_u32(zldev, ZL_REG_REF_ESYNC_DIV, &ref->esync_n_div); + if (rc) + return rc; + rc = zl3073x_read_u8(zldev, ZL_REG_REF_SYNC_CTRL, &ref->sync_ctrl); + if (rc) + return rc; + + /* Read phase compensation register */ + rc = zl3073x_read_u48(zldev, ZL_REG_REF_PHASE_OFFSET_COMP, + &ref->phase_comp); + if (rc) + return rc; + dev_dbg(zldev->dev, "REF%u is %s and configured as %s\n", index, str_enabled_disabled(zl3073x_ref_is_enabled(ref)), zl3073x_ref_is_diff(ref) ? "differential" : "single-ended"); @@ -110,3 +145,60 @@ zl3073x_ref_state_get(struct zl3073x_dev *zldev, u8 index) { return &zldev->ref[index]; } + +int zl3073x_ref_state_set(struct zl3073x_dev *zldev, u8 index, + const struct zl3073x_ref *ref) +{ + struct zl3073x_ref *dref = &zldev->ref[index]; + int rc; + + guard(mutex)(&zldev->multiop_lock); + + /* Read reference configuration into mailbox */ + rc = zl3073x_mb_op(zldev, ZL_REG_REF_MB_SEM, ZL_REF_MB_SEM_RD, + ZL_REG_REF_MB_MASK, BIT(index)); + if (rc) + return rc; + + /* Update mailbox with changed values */ + if (dref->freq_base != ref->freq_base) + rc = zl3073x_write_u16(zldev, ZL_REG_REF_FREQ_BASE, + ref->freq_base); + if (!rc && dref->freq_mult != ref->freq_mult) + rc = zl3073x_write_u16(zldev, ZL_REG_REF_FREQ_MULT, + ref->freq_mult); + if (!rc && dref->freq_ratio_m != ref->freq_ratio_m) + rc = zl3073x_write_u16(zldev, ZL_REG_REF_RATIO_M, + ref->freq_ratio_m); + if (!rc && dref->freq_ratio_n != ref->freq_ratio_n) + rc = zl3073x_write_u16(zldev, ZL_REG_REF_RATIO_N, + ref->freq_ratio_n); + if (!rc && dref->esync_n_div != ref->esync_n_div) + rc = zl3073x_write_u32(zldev, ZL_REG_REF_ESYNC_DIV, + ref->esync_n_div); + if (!rc && dref->sync_ctrl != ref->sync_ctrl) + rc = zl3073x_write_u8(zldev, ZL_REG_REF_SYNC_CTRL, + ref->sync_ctrl); + if (!rc && dref->phase_comp != ref->phase_comp) + rc = zl3073x_write_u48(zldev, ZL_REG_REF_PHASE_OFFSET_COMP, + ref->phase_comp); + if (rc) + return rc; + + /* Commit reference configuration */ + rc = zl3073x_mb_op(zldev, ZL_REG_REF_MB_SEM, ZL_REF_MB_SEM_WR, + ZL_REG_REF_MB_MASK, BIT(index)); + if (rc) + return rc; + + /* After successful commit store new state */ + dref->freq_base = ref->freq_base; + dref->freq_mult = ref->freq_mult; + dref->freq_ratio_m = ref->freq_ratio_m; + dref->freq_ratio_n = ref->freq_ratio_n; + dref->esync_n_div = ref->esync_n_div; + dref->sync_ctrl = ref->sync_ctrl; + dref->phase_comp = ref->phase_comp; + + return 0; +} diff --git a/drivers/dpll/zl3073x/ref.h b/drivers/dpll/zl3073x/ref.h index c4931e545d24..efc7f59cd9f9 100644 --- a/drivers/dpll/zl3073x/ref.h +++ b/drivers/dpll/zl3073x/ref.h @@ -4,6 +4,7 @@ #define _ZL3073X_REF_H #include +#include #include #include "regs.h" @@ -13,12 +14,26 @@ struct zl3073x_dev; /** * struct zl3073x_ref - input reference state * @ffo: current fractional frequency offset + * @phase_comp: phase compensation + * @esync_n_div: divisor for embedded sync or n-divided signal formats + * @freq_base: frequency base + * @freq_mult: frequnecy multiplier + * @freq_ratio_m: FEC mode multiplier + * @freq_ratio_n: FEC mode divisor * @config: reference config + * @sync_ctrl: reference sync control * @mon_status: reference monitor status */ struct zl3073x_ref { s64 ffo; + u64 phase_comp; + u32 esync_n_div; + u16 freq_base; + u16 freq_mult; + u16 freq_ratio_m; + u16 freq_ratio_n; u8 config; + u8 sync_ctrl; u8 mon_status; }; @@ -27,6 +42,9 @@ int zl3073x_ref_state_fetch(struct zl3073x_dev *zldev, u8 index); const struct zl3073x_ref *zl3073x_ref_state_get(struct zl3073x_dev *zldev, u8 index); +int zl3073x_ref_state_set(struct zl3073x_dev *zldev, u8 index, + const struct zl3073x_ref *ref); + int zl3073x_ref_freq_factorize(u32 freq, u16 *base, u16 *mult); /** @@ -41,6 +59,42 @@ zl3073x_ref_ffo_get(const struct zl3073x_ref *ref) return ref->ffo; } +/** + * zl3073x_ref_freq_get - get given input reference frequency + * @ref: pointer to ref state + * + * Return: frequency of the given input reference + */ +static inline u32 +zl3073x_ref_freq_get(const struct zl3073x_ref *ref) +{ + return mul_u64_u32_div(ref->freq_base * ref->freq_mult, + ref->freq_ratio_m, ref->freq_ratio_n); +} + +/** + * zl3073x_ref_freq_set - set given input reference frequency + * @ref: pointer to ref state + * @freq: frequency to be set + * + * Return: 0 on success, <0 when frequency cannot be factorized + */ +static inline int +zl3073x_ref_freq_set(struct zl3073x_ref *ref, u32 freq) +{ + u16 base, mult; + int rc; + + rc = zl3073x_ref_freq_factorize(freq, &base, &mult); + if (rc) + return rc; + + ref->freq_base = base; + ref->freq_mult = mult; + + return 0; +} + /** * zl3073x_ref_is_diff - check if the given input reference is differential * @ref: pointer to ref state From 5fb9b0d411f81ec46833ea8e43c0263515060c64 Mon Sep 17 00:00:00 2001 From: Ivan Vecera Date: Thu, 13 Nov 2025 08:41:04 +0100 Subject: [PATCH 0864/1381] dpll: zl3073x: Cache all output properties in zl3073x_out Expand the zl3073x_out structure to cache all output-related hardware registers, including divisors, widths, embedded-sync parameters and phase compensation. Modify zl3073x_out_state_fetch() to read and populate all these new fields at once, including zero-divisor checks. Refactor all dpll "getter" functions in dpll.c to read from this new cached state instead of performing direct register access. Introduce a new function, zl3073x_out_state_set(), to handle writing changes back to the hardware. This function compares the provided state with the current cached state and writes *only* the modified register values via a single mailbox sequence before updating the local cache. Refactor all dpll "setter" functions to modify a local copy of the output state and then call zl3073x_out_state_set() to commit the changes. This change centralizes all output-related register I/O into out.c, significantly reduces bus traffic, and simplifies the logic in dpll.c. Reviewed-by: Petr Oros Tested-by: Prathosh Satish Signed-off-by: Ivan Vecera Link: https://patch.msgid.link/20251113074105.141379-6-ivecera@redhat.com Signed-off-by: Jakub Kicinski --- drivers/dpll/zl3073x/dpll.c | 380 +++++++++--------------------------- drivers/dpll/zl3073x/out.c | 90 +++++++++ drivers/dpll/zl3073x/out.h | 13 ++ 3 files changed, 193 insertions(+), 290 deletions(-) diff --git a/drivers/dpll/zl3073x/dpll.c b/drivers/dpll/zl3073x/dpll.c index e0d696bc1909..9879d85d29af 100644 --- a/drivers/dpll/zl3073x/dpll.c +++ b/drivers/dpll/zl3073x/dpll.c @@ -770,21 +770,19 @@ zl3073x_dpll_output_pin_esync_get(const struct dpll_pin *dpll_pin, struct zl3073x_dpll *zldpll = dpll_priv; struct zl3073x_dev *zldev = zldpll->dev; struct zl3073x_dpll_pin *pin = pin_priv; - struct device *dev = zldev->dev; - u32 esync_period, esync_width; - u8 clock_type, synth; - u8 out, output_mode; - u32 output_div; + const struct zl3073x_synth *synth; + const struct zl3073x_out *out; + u8 clock_type, out_id; u32 synth_freq; - int rc; - out = zl3073x_output_pin_out_get(pin->id); + out_id = zl3073x_output_pin_out_get(pin->id); + out = zl3073x_out_state_get(zldev, out_id); /* If N-division is enabled, esync is not supported. The register used * for N-division is also used for the esync divider so both cannot * be used. */ - switch (zl3073x_dev_out_signal_format_get(zldev, out)) { + switch (zl3073x_out_signal_format_get(out)) { case ZL_OUTPUT_MODE_SIGNAL_FORMAT_2_NDIV: case ZL_OUTPUT_MODE_SIGNAL_FORMAT_2_NDIV_INV: return -EOPNOTSUPP; @@ -792,38 +790,11 @@ zl3073x_dpll_output_pin_esync_get(const struct dpll_pin *dpll_pin, break; } - guard(mutex)(&zldev->multiop_lock); + /* Get attached synth frequency */ + synth = zl3073x_synth_state_get(zldev, zl3073x_out_synth_get(out)); + synth_freq = zl3073x_synth_freq_get(synth); - /* Read output configuration into mailbox */ - rc = zl3073x_mb_op(zldev, ZL_REG_OUTPUT_MB_SEM, ZL_OUTPUT_MB_SEM_RD, - ZL_REG_OUTPUT_MB_MASK, BIT(out)); - if (rc) - return rc; - - /* Read output mode */ - rc = zl3073x_read_u8(zldev, ZL_REG_OUTPUT_MODE, &output_mode); - if (rc) - return rc; - - /* Read output divisor */ - rc = zl3073x_read_u32(zldev, ZL_REG_OUTPUT_DIV, &output_div); - if (rc) - return rc; - - /* Check output divisor for zero */ - if (!output_div) { - dev_err(dev, "Zero divisor for OUTPUT%u got from device\n", - out); - return -EINVAL; - } - - /* Get synth attached to output pin */ - synth = zl3073x_dev_out_synth_get(zldev, out); - - /* Get synth frequency */ - synth_freq = zl3073x_dev_synth_freq_get(zldev, synth); - - clock_type = FIELD_GET(ZL_OUTPUT_MODE_CLOCK_TYPE, output_mode); + clock_type = FIELD_GET(ZL_OUTPUT_MODE_CLOCK_TYPE, out->mode); if (clock_type != ZL_OUTPUT_MODE_CLOCK_TYPE_ESYNC) { /* No need to read esync data if it is not enabled */ esync->freq = 0; @@ -832,38 +803,21 @@ zl3073x_dpll_output_pin_esync_get(const struct dpll_pin *dpll_pin, goto finish; } - /* Read esync period */ - rc = zl3073x_read_u32(zldev, ZL_REG_OUTPUT_ESYNC_PERIOD, &esync_period); - if (rc) - return rc; - - /* Check esync divisor for zero */ - if (!esync_period) { - dev_err(dev, "Zero esync divisor for OUTPUT%u got from device\n", - out); - return -EINVAL; - } - - /* Get esync pulse width in units of half synth cycles */ - rc = zl3073x_read_u32(zldev, ZL_REG_OUTPUT_ESYNC_WIDTH, &esync_width); - if (rc) - return rc; - /* Compute esync frequency */ - esync->freq = synth_freq / output_div / esync_period; + esync->freq = synth_freq / out->div / out->esync_n_period; /* By comparing the esync_pulse_width to the half of the pulse width * the esync pulse percentage can be determined. * Note that half pulse width is in units of half synth cycles, which * is why it reduces down to be output_div. */ - esync->pulse = (50 * esync_width) / output_div; + esync->pulse = (50 * out->esync_n_width) / out->div; finish: /* Set supported esync ranges if the pin supports esync control and * if the output frequency is > 1 Hz. */ - if (pin->esync_control && (synth_freq / output_div) > 1) { + if (pin->esync_control && (synth_freq / out->div) > 1) { esync->range = esync_freq_ranges; esync->range_num = ARRAY_SIZE(esync_freq_ranges); } else { @@ -881,21 +835,22 @@ zl3073x_dpll_output_pin_esync_set(const struct dpll_pin *dpll_pin, void *dpll_priv, u64 freq, struct netlink_ext_ack *extack) { - u32 esync_period, esync_width, output_div; struct zl3073x_dpll *zldpll = dpll_priv; struct zl3073x_dev *zldev = zldpll->dev; struct zl3073x_dpll_pin *pin = pin_priv; - u8 clock_type, out, output_mode, synth; + const struct zl3073x_synth *synth; + struct zl3073x_out out; + u8 clock_type, out_id; u32 synth_freq; - int rc; - out = zl3073x_output_pin_out_get(pin->id); + out_id = zl3073x_output_pin_out_get(pin->id); + out = *zl3073x_out_state_get(zldev, out_id); /* If N-division is enabled, esync is not supported. The register used * for N-division is also used for the esync divider so both cannot * be used. */ - switch (zl3073x_dev_out_signal_format_get(zldev, out)) { + switch (zl3073x_out_signal_format_get(&out)) { case ZL_OUTPUT_MODE_SIGNAL_FORMAT_2_NDIV: case ZL_OUTPUT_MODE_SIGNAL_FORMAT_2_NDIV_INV: return -EOPNOTSUPP; @@ -903,19 +858,6 @@ zl3073x_dpll_output_pin_esync_set(const struct dpll_pin *dpll_pin, break; } - guard(mutex)(&zldev->multiop_lock); - - /* Read output configuration into mailbox */ - rc = zl3073x_mb_op(zldev, ZL_REG_OUTPUT_MB_SEM, ZL_OUTPUT_MB_SEM_RD, - ZL_REG_OUTPUT_MB_MASK, BIT(out)); - if (rc) - return rc; - - /* Read output mode */ - rc = zl3073x_read_u8(zldev, ZL_REG_OUTPUT_MODE, &output_mode); - if (rc) - return rc; - /* Select clock type */ if (freq) clock_type = ZL_OUTPUT_MODE_CLOCK_TYPE_ESYNC; @@ -923,38 +865,19 @@ zl3073x_dpll_output_pin_esync_set(const struct dpll_pin *dpll_pin, clock_type = ZL_OUTPUT_MODE_CLOCK_TYPE_NORMAL; /* Update clock type in output mode */ - output_mode &= ~ZL_OUTPUT_MODE_CLOCK_TYPE; - output_mode |= FIELD_PREP(ZL_OUTPUT_MODE_CLOCK_TYPE, clock_type); - rc = zl3073x_write_u8(zldev, ZL_REG_OUTPUT_MODE, output_mode); - if (rc) - return rc; + out.mode &= ~ZL_OUTPUT_MODE_CLOCK_TYPE; + out.mode |= FIELD_PREP(ZL_OUTPUT_MODE_CLOCK_TYPE, clock_type); /* If esync is being disabled just write mailbox and finish */ if (!freq) goto write_mailbox; - /* Get synth attached to output pin */ - synth = zl3073x_dev_out_synth_get(zldev, out); - - /* Get synth frequency */ - synth_freq = zl3073x_dev_synth_freq_get(zldev, synth); - - rc = zl3073x_read_u32(zldev, ZL_REG_OUTPUT_DIV, &output_div); - if (rc) - return rc; - - /* Check output divisor for zero */ - if (!output_div) { - dev_err(zldev->dev, - "Zero divisor for OUTPUT%u got from device\n", out); - return -EINVAL; - } + /* Get attached synth frequency */ + synth = zl3073x_synth_state_get(zldev, zl3073x_out_synth_get(&out)); + synth_freq = zl3073x_synth_freq_get(synth); /* Compute and update esync period */ - esync_period = synth_freq / (u32)freq / output_div; - rc = zl3073x_write_u32(zldev, ZL_REG_OUTPUT_ESYNC_PERIOD, esync_period); - if (rc) - return rc; + out.esync_n_period = synth_freq / (u32)freq / out.div; /* Half of the period in units of 1/2 synth cycle can be represented by * the output_div. To get the supported esync pulse width of 25% of the @@ -962,15 +885,11 @@ zl3073x_dpll_output_pin_esync_set(const struct dpll_pin *dpll_pin, * assumes that output_div is even, otherwise some resolution will be * lost. */ - esync_width = output_div / 2; - rc = zl3073x_write_u32(zldev, ZL_REG_OUTPUT_ESYNC_WIDTH, esync_width); - if (rc) - return rc; + out.esync_n_width = out.div / 2; write_mailbox: /* Commit output configuration */ - return zl3073x_mb_op(zldev, ZL_REG_OUTPUT_MB_SEM, ZL_OUTPUT_MB_SEM_WR, - ZL_REG_OUTPUT_MB_MASK, BIT(out)); + return zl3073x_out_state_set(zldev, out_id, &out); } static int @@ -983,83 +902,46 @@ zl3073x_dpll_output_pin_frequency_get(const struct dpll_pin *dpll_pin, struct zl3073x_dpll *zldpll = dpll_priv; struct zl3073x_dev *zldev = zldpll->dev; struct zl3073x_dpll_pin *pin = pin_priv; - struct device *dev = zldev->dev; - u8 out, signal_format, synth; - u32 output_div, synth_freq; - int rc; + const struct zl3073x_synth *synth; + const struct zl3073x_out *out; + u32 synth_freq; + u8 out_id; - out = zl3073x_output_pin_out_get(pin->id); - synth = zl3073x_dev_out_synth_get(zldev, out); - synth_freq = zl3073x_dev_synth_freq_get(zldev, synth); + out_id = zl3073x_output_pin_out_get(pin->id); + out = zl3073x_out_state_get(zldev, out_id); - guard(mutex)(&zldev->multiop_lock); + /* Get attached synth frequency */ + synth = zl3073x_synth_state_get(zldev, zl3073x_out_synth_get(out)); + synth_freq = zl3073x_synth_freq_get(synth); - /* Read output configuration into mailbox */ - rc = zl3073x_mb_op(zldev, ZL_REG_OUTPUT_MB_SEM, ZL_OUTPUT_MB_SEM_RD, - ZL_REG_OUTPUT_MB_MASK, BIT(out)); - if (rc) - return rc; - - rc = zl3073x_read_u32(zldev, ZL_REG_OUTPUT_DIV, &output_div); - if (rc) - return rc; - - /* Check output divisor for zero */ - if (!output_div) { - dev_err(dev, "Zero divisor for output %u got from device\n", - out); - return -EINVAL; - } - - /* Read used signal format for the given output */ - signal_format = zl3073x_dev_out_signal_format_get(zldev, out); - - switch (signal_format) { + switch (zl3073x_out_signal_format_get(out)) { case ZL_OUTPUT_MODE_SIGNAL_FORMAT_2_NDIV: case ZL_OUTPUT_MODE_SIGNAL_FORMAT_2_NDIV_INV: /* In case of divided format we have to distiguish between * given output pin type. + * + * For P-pin the resulting frequency is computed as simple + * division of synth frequency and output divisor. + * + * For N-pin we have to divide additionally by divisor stored + * in esync_n_period output mailbox register that is used as + * N-pin divisor for these modes. */ - if (zl3073x_dpll_is_p_pin(pin)) { - /* For P-pin the resulting frequency is computed as - * simple division of synth frequency and output - * divisor. - */ - *frequency = synth_freq / output_div; - } else { - /* For N-pin we have to divide additionally by - * divisor stored in esync_period output mailbox - * register that is used as N-pin divisor for these - * modes. - */ - u32 ndiv; + *frequency = synth_freq / out->div; - rc = zl3073x_read_u32(zldev, ZL_REG_OUTPUT_ESYNC_PERIOD, - &ndiv); - if (rc) - return rc; + if (!zl3073x_dpll_is_p_pin(pin)) + *frequency = (u32)*frequency / out->esync_n_period; - /* Check N-pin divisor for zero */ - if (!ndiv) { - dev_err(dev, - "Zero N-pin divisor for output %u got from device\n", - out); - return -EINVAL; - } - - /* Compute final divisor for N-pin */ - *frequency = synth_freq / output_div / ndiv; - } break; default: /* In other modes the resulting frequency is computed as * division of synth frequency and output divisor. */ - *frequency = synth_freq / output_div; + *frequency = synth_freq / out->div; break; } - return rc; + return 0; } static int @@ -1072,28 +954,21 @@ zl3073x_dpll_output_pin_frequency_set(const struct dpll_pin *dpll_pin, struct zl3073x_dpll *zldpll = dpll_priv; struct zl3073x_dev *zldev = zldpll->dev; struct zl3073x_dpll_pin *pin = pin_priv; - struct device *dev = zldev->dev; - u32 output_n_freq, output_p_freq; - u8 out, signal_format, synth; - u32 cur_div, new_div, ndiv; - u32 synth_freq; - int rc; + const struct zl3073x_synth *synth; + u8 out_id, signal_format; + u32 new_div, synth_freq; + struct zl3073x_out out; - out = zl3073x_output_pin_out_get(pin->id); - synth = zl3073x_dev_out_synth_get(zldev, out); - synth_freq = zl3073x_dev_synth_freq_get(zldev, synth); + out_id = zl3073x_output_pin_out_get(pin->id); + out = *zl3073x_out_state_get(zldev, out_id); + + /* Get attached synth frequency and compute new divisor */ + synth = zl3073x_synth_state_get(zldev, zl3073x_out_synth_get(&out)); + synth_freq = zl3073x_synth_freq_get(synth); new_div = synth_freq / (u32)frequency; /* Get used signal format for the given output */ - signal_format = zl3073x_dev_out_signal_format_get(zldev, out); - - guard(mutex)(&zldev->multiop_lock); - - /* Load output configuration */ - rc = zl3073x_mb_op(zldev, ZL_REG_OUTPUT_MB_SEM, ZL_OUTPUT_MB_SEM_RD, - ZL_REG_OUTPUT_MB_MASK, BIT(out)); - if (rc) - return rc; + signal_format = zl3073x_out_signal_format_get(&out); /* Check signal format */ if (signal_format != ZL_OUTPUT_MODE_SIGNAL_FORMAT_2_NDIV && @@ -1101,99 +976,50 @@ zl3073x_dpll_output_pin_frequency_set(const struct dpll_pin *dpll_pin, /* For non N-divided signal formats the frequency is computed * as division of synth frequency and output divisor. */ - rc = zl3073x_write_u32(zldev, ZL_REG_OUTPUT_DIV, new_div); - if (rc) - return rc; + out.div = new_div; /* For 50/50 duty cycle the divisor is equal to width */ - rc = zl3073x_write_u32(zldev, ZL_REG_OUTPUT_WIDTH, new_div); - if (rc) - return rc; + out.width = new_div; /* Commit output configuration */ - return zl3073x_mb_op(zldev, - ZL_REG_OUTPUT_MB_SEM, ZL_OUTPUT_MB_SEM_WR, - ZL_REG_OUTPUT_MB_MASK, BIT(out)); + return zl3073x_out_state_set(zldev, out_id, &out); } - /* For N-divided signal format get current divisor */ - rc = zl3073x_read_u32(zldev, ZL_REG_OUTPUT_DIV, &cur_div); - if (rc) - return rc; - - /* Check output divisor for zero */ - if (!cur_div) { - dev_err(dev, "Zero divisor for output %u got from device\n", - out); - return -EINVAL; - } - - /* Get N-pin divisor (shares the same register with esync */ - rc = zl3073x_read_u32(zldev, ZL_REG_OUTPUT_ESYNC_PERIOD, &ndiv); - if (rc) - return rc; - - /* Check N-pin divisor for zero */ - if (!ndiv) { - dev_err(dev, - "Zero N-pin divisor for output %u got from device\n", - out); - return -EINVAL; - } - - /* Compute current output frequency for P-pin */ - output_p_freq = synth_freq / cur_div; - - /* Compute current N-pin frequency */ - output_n_freq = output_p_freq / ndiv; - if (zl3073x_dpll_is_p_pin(pin)) { /* We are going to change output frequency for P-pin but * if the requested frequency is less than current N-pin * frequency then indicate a failure as we are not able * to compute N-pin divisor to keep its frequency unchanged. + * + * Update divisor for N-pin to keep N-pin frequency. */ - if (frequency <= output_n_freq) + out.esync_n_period = (out.esync_n_period * out.div) / new_div; + if (!out.esync_n_period) return -EINVAL; /* Update the output divisor */ - rc = zl3073x_write_u32(zldev, ZL_REG_OUTPUT_DIV, new_div); - if (rc) - return rc; + out.div = new_div; /* For 50/50 duty cycle the divisor is equal to width */ - rc = zl3073x_write_u32(zldev, ZL_REG_OUTPUT_WIDTH, new_div); - if (rc) - return rc; - - /* Compute new divisor for N-pin */ - ndiv = (u32)frequency / output_n_freq; + out.width = out.div; } else { /* We are going to change frequency of N-pin but if * the requested freq is greater or equal than freq of P-pin * in the output pair we cannot compute divisor for the N-pin. * In this case indicate a failure. + * + * Update divisor for N-pin */ - if (output_p_freq <= frequency) + out.esync_n_period = div64_u64(synth_freq, frequency * out.div); + if (!out.esync_n_period) return -EINVAL; - - /* Compute new divisor for N-pin */ - ndiv = output_p_freq / (u32)frequency; } - /* Update divisor for the N-pin */ - rc = zl3073x_write_u32(zldev, ZL_REG_OUTPUT_ESYNC_PERIOD, ndiv); - if (rc) - return rc; - /* For 50/50 duty cycle the divisor is equal to width */ - rc = zl3073x_write_u32(zldev, ZL_REG_OUTPUT_ESYNC_WIDTH, ndiv); - if (rc) - return rc; + out.esync_n_width = out.esync_n_period; /* Commit output configuration */ - return zl3073x_mb_op(zldev, ZL_REG_OUTPUT_MB_SEM, ZL_OUTPUT_MB_SEM_WR, - ZL_REG_OUTPUT_MB_MASK, BIT(out)); + return zl3073x_out_state_set(zldev, out_id, &out); } static int @@ -1207,30 +1033,18 @@ zl3073x_dpll_output_pin_phase_adjust_get(const struct dpll_pin *dpll_pin, struct zl3073x_dpll *zldpll = dpll_priv; struct zl3073x_dev *zldev = zldpll->dev; struct zl3073x_dpll_pin *pin = pin_priv; - s32 phase_comp; - u8 out; - int rc; + const struct zl3073x_out *out; + u8 out_id; - guard(mutex)(&zldev->multiop_lock); - - /* Read output configuration */ - out = zl3073x_output_pin_out_get(pin->id); - rc = zl3073x_mb_op(zldev, ZL_REG_OUTPUT_MB_SEM, ZL_OUTPUT_MB_SEM_RD, - ZL_REG_OUTPUT_MB_MASK, BIT(out)); - if (rc) - return rc; - - /* Read current output phase compensation */ - rc = zl3073x_read_u32(zldev, ZL_REG_OUTPUT_PHASE_COMP, &phase_comp); - if (rc) - return rc; + out_id = zl3073x_output_pin_out_get(pin->id); + out = zl3073x_out_state_get(zldev, out_id); /* Convert value to ps and reverse two's complement negation applied * during 'set' */ - *phase_adjust = -phase_comp * pin->phase_gran; + *phase_adjust = -out->phase_comp * pin->phase_gran; - return rc; + return 0; } static int @@ -1244,31 +1058,19 @@ zl3073x_dpll_output_pin_phase_adjust_set(const struct dpll_pin *dpll_pin, struct zl3073x_dpll *zldpll = dpll_priv; struct zl3073x_dev *zldev = zldpll->dev; struct zl3073x_dpll_pin *pin = pin_priv; - u8 out; - int rc; + struct zl3073x_out out; + u8 out_id; + + out_id = zl3073x_output_pin_out_get(pin->id); + out = *zl3073x_out_state_get(zldev, out_id); /* The value in the register is stored as two's complement negation * of requested value and expressed in half synth clock cycles. */ - phase_adjust = -phase_adjust / pin->phase_gran; - - guard(mutex)(&zldev->multiop_lock); - - /* Read output configuration */ - out = zl3073x_output_pin_out_get(pin->id); - rc = zl3073x_mb_op(zldev, ZL_REG_OUTPUT_MB_SEM, ZL_OUTPUT_MB_SEM_RD, - ZL_REG_OUTPUT_MB_MASK, BIT(out)); - if (rc) - return rc; - - /* Write the requested value into the compensation register */ - rc = zl3073x_write_u32(zldev, ZL_REG_OUTPUT_PHASE_COMP, phase_adjust); - if (rc) - return rc; + out.phase_comp = -phase_adjust / pin->phase_gran; /* Update output configuration from mailbox */ - return zl3073x_mb_op(zldev, ZL_REG_OUTPUT_MB_SEM, ZL_OUTPUT_MB_SEM_WR, - ZL_REG_OUTPUT_MB_MASK, BIT(out)); + return zl3073x_out_state_set(zldev, out_id, &out); } static int @@ -1680,17 +1482,15 @@ zl3073x_dpll_pin_is_registrable(struct zl3073x_dpll *zldpll, /* Output P&N pair shares single HW output */ u8 out = zl3073x_output_pin_out_get(index); - name = "OUT"; - /* Skip the pin if it is connected to different DPLL channel */ if (zl3073x_dev_out_dpll_get(zldev, out) != zldpll->id) { dev_dbg(zldev->dev, - "%s%u is driven by different DPLL\n", name, - out); + "OUT%u is driven by different DPLL\n", out); return false; } + name = "OUT"; is_diff = zl3073x_dev_out_is_diff(zldev, out); is_enabled = zl3073x_dev_output_pin_is_enabled(zldev, index); } diff --git a/drivers/dpll/zl3073x/out.c b/drivers/dpll/zl3073x/out.c index a48f6917b39f..86829a0c1c02 100644 --- a/drivers/dpll/zl3073x/out.c +++ b/drivers/dpll/zl3073x/out.c @@ -50,6 +50,46 @@ int zl3073x_out_state_fetch(struct zl3073x_dev *zldev, u8 index) dev_dbg(zldev->dev, "OUT%u has signal format 0x%02x\n", index, zl3073x_out_signal_format_get(out)); + /* Read output divisor */ + rc = zl3073x_read_u32(zldev, ZL_REG_OUTPUT_DIV, &out->div); + if (rc) + return rc; + + if (!out->div) { + dev_err(zldev->dev, "Zero divisor for OUT%u got from device\n", + index); + return -EINVAL; + } + + dev_dbg(zldev->dev, "OUT%u divisor: %u\n", index, out->div); + + /* Read output width */ + rc = zl3073x_read_u32(zldev, ZL_REG_OUTPUT_WIDTH, &out->width); + if (rc) + return rc; + + rc = zl3073x_read_u32(zldev, ZL_REG_OUTPUT_ESYNC_PERIOD, + &out->esync_n_period); + if (rc) + return rc; + + if (!out->esync_n_period) { + dev_err(zldev->dev, + "Zero esync divisor for OUT%u got from device\n", + index); + return -EINVAL; + } + + rc = zl3073x_read_u32(zldev, ZL_REG_OUTPUT_ESYNC_WIDTH, + &out->esync_n_width); + if (rc) + return rc; + + rc = zl3073x_read_u32(zldev, ZL_REG_OUTPUT_PHASE_COMP, + &out->phase_comp); + if (rc) + return rc; + return rc; } @@ -65,3 +105,53 @@ const struct zl3073x_out *zl3073x_out_state_get(struct zl3073x_dev *zldev, { return &zldev->out[index]; } + +int zl3073x_out_state_set(struct zl3073x_dev *zldev, u8 index, + const struct zl3073x_out *out) +{ + struct zl3073x_out *dout = &zldev->out[index]; + int rc; + + guard(mutex)(&zldev->multiop_lock); + + /* Read output configuration into mailbox */ + rc = zl3073x_mb_op(zldev, ZL_REG_OUTPUT_MB_SEM, ZL_OUTPUT_MB_SEM_RD, + ZL_REG_OUTPUT_MB_MASK, BIT(index)); + if (rc) + return rc; + + /* Update mailbox with changed values */ + if (dout->div != out->div) + rc = zl3073x_write_u32(zldev, ZL_REG_OUTPUT_DIV, out->div); + if (!rc && dout->width != out->width) + rc = zl3073x_write_u32(zldev, ZL_REG_OUTPUT_WIDTH, out->width); + if (!rc && dout->esync_n_period != out->esync_n_period) + rc = zl3073x_write_u32(zldev, ZL_REG_OUTPUT_ESYNC_PERIOD, + out->esync_n_period); + if (!rc && dout->esync_n_width != out->esync_n_width) + rc = zl3073x_write_u32(zldev, ZL_REG_OUTPUT_ESYNC_WIDTH, + out->esync_n_width); + if (!rc && dout->mode != out->mode) + rc = zl3073x_write_u8(zldev, ZL_REG_OUTPUT_MODE, out->mode); + if (!rc && dout->phase_comp != out->phase_comp) + rc = zl3073x_write_u32(zldev, ZL_REG_OUTPUT_PHASE_COMP, + out->phase_comp); + if (rc) + return rc; + + /* Commit output configuration */ + rc = zl3073x_mb_op(zldev, ZL_REG_OUTPUT_MB_SEM, ZL_OUTPUT_MB_SEM_WR, + ZL_REG_OUTPUT_MB_MASK, BIT(index)); + if (rc) + return rc; + + /* After successful commit store new state */ + dout->div = out->div; + dout->width = out->width; + dout->esync_n_period = out->esync_n_period; + dout->esync_n_width = out->esync_n_width; + dout->mode = out->mode; + dout->phase_comp = out->phase_comp; + + return 0; +} diff --git a/drivers/dpll/zl3073x/out.h b/drivers/dpll/zl3073x/out.h index 986aa046221d..e8ea7a0e0f07 100644 --- a/drivers/dpll/zl3073x/out.h +++ b/drivers/dpll/zl3073x/out.h @@ -12,10 +12,20 @@ struct zl3073x_dev; /** * struct zl3073x_out - output state + * @div: output divisor + * @width: output pulse width + * @esync_n_period: embedded sync or n-pin period (for n-div formats) + * @esync_n_width: embedded sync or n-pin pulse width + * @phase_comp: phase compensation * @ctrl: output control * @mode: output mode */ struct zl3073x_out { + u32 div; + u32 width; + u32 esync_n_period; + u32 esync_n_width; + s32 phase_comp; u8 ctrl; u8 mode; }; @@ -24,6 +34,9 @@ int zl3073x_out_state_fetch(struct zl3073x_dev *zldev, u8 index); const struct zl3073x_out *zl3073x_out_state_get(struct zl3073x_dev *zldev, u8 index); +int zl3073x_out_state_set(struct zl3073x_dev *zldev, u8 index, + const struct zl3073x_out *out); + /** * zl3073x_out_signal_format_get - get output signal format * @out: pointer to out state From 01e0e8b6a2d798e7403f4f636b15bcf644a7f110 Mon Sep 17 00:00:00 2001 From: Ivan Vecera Date: Thu, 13 Nov 2025 08:41:05 +0100 Subject: [PATCH 0865/1381] dpll: zl3073x: Remove unused dev wrappers Remove several zl3073x_dev_... inline wrapper functions from core.h as they are no longer used by any callers. Removed functions: * zl3073x_dev_ref_ffo_get * zl3073x_dev_ref_is_enabled * zl3073x_dev_synth_dpll_get * zl3073x_dev_synth_is_enabled * zl3073x_dev_out_signal_format_get This is a cleanup after recent refactoring, as the remaining callers now fetch the state object and use the base helpers directly. Reviewed-by: Petr Oros Tested-by: Prathosh Satish Signed-off-by: Ivan Vecera Link: https://patch.msgid.link/20251113074105.141379-7-ivecera@redhat.com Signed-off-by: Jakub Kicinski --- drivers/dpll/zl3073x/core.h | 77 ------------------------------------- 1 file changed, 77 deletions(-) diff --git a/drivers/dpll/zl3073x/core.h b/drivers/dpll/zl3073x/core.h index fe8b70e25d3c..09bca2d0926d 100644 --- a/drivers/dpll/zl3073x/core.h +++ b/drivers/dpll/zl3073x/core.h @@ -182,21 +182,6 @@ zl3073x_output_pin_out_get(u8 id) return id / 2; } -/** - * zl3073x_dev_ref_ffo_get - get current fractional frequency offset - * @zldev: pointer to zl3073x device - * @index: input reference index - * - * Return: the latest measured fractional frequency offset - */ -static inline s64 -zl3073x_dev_ref_ffo_get(struct zl3073x_dev *zldev, u8 index) -{ - const struct zl3073x_ref *ref = zl3073x_ref_state_get(zldev, index); - - return zl3073x_ref_ffo_get(ref); -} - /** * zl3073x_dev_ref_freq_get - get input reference frequency * @zldev: pointer to zl3073x device @@ -227,21 +212,6 @@ zl3073x_dev_ref_is_diff(struct zl3073x_dev *zldev, u8 index) return zl3073x_ref_is_diff(ref); } -/** - * zl3073x_dev_ref_is_enabled - check if the given input reference is enabled - * @zldev: pointer to zl3073x device - * @index: input reference index - * - * Return: true if input refernce is enabled, false otherwise - */ -static inline bool -zl3073x_dev_ref_is_enabled(struct zl3073x_dev *zldev, u8 index) -{ - const struct zl3073x_ref *ref = zl3073x_ref_state_get(zldev, index); - - return zl3073x_ref_is_enabled(ref); -} - /* * zl3073x_dev_ref_is_status_ok - check the given input reference status * @zldev: pointer to zl3073x device @@ -257,22 +227,6 @@ zl3073x_dev_ref_is_status_ok(struct zl3073x_dev *zldev, u8 index) return zl3073x_ref_is_status_ok(ref); } -/** - * zl3073x_dev_synth_dpll_get - get DPLL ID the synth is driven by - * @zldev: pointer to zl3073x device - * @index: synth index - * - * Return: ID of DPLL the given synthetizer is driven by - */ -static inline u8 -zl3073x_dev_synth_dpll_get(struct zl3073x_dev *zldev, u8 index) -{ - const struct zl3073x_synth *synth; - - synth = zl3073x_synth_state_get(zldev, index); - return zl3073x_synth_dpll_get(synth); -} - /** * zl3073x_dev_synth_freq_get - get synth current freq * @zldev: pointer to zl3073x device @@ -289,22 +243,6 @@ zl3073x_dev_synth_freq_get(struct zl3073x_dev *zldev, u8 index) return zl3073x_synth_freq_get(synth); } -/** - * zl3073x_dev_synth_is_enabled - check if the given synth is enabled - * @zldev: pointer to zl3073x device - * @index: synth index - * - * Return: true if synth is enabled, false otherwise - */ -static inline bool -zl3073x_dev_synth_is_enabled(struct zl3073x_dev *zldev, u8 index) -{ - const struct zl3073x_synth *synth; - - synth = zl3073x_synth_state_get(zldev, index); - return zl3073x_synth_is_enabled(synth); -} - /** * zl3073x_dev_out_synth_get - get synth connected to given output * @zldev: pointer to zl3073x device @@ -341,21 +279,6 @@ zl3073x_dev_out_is_enabled(struct zl3073x_dev *zldev, u8 index) return zl3073x_synth_is_enabled(synth) && zl3073x_out_is_enabled(out); } -/** - * zl3073x_dev_out_signal_format_get - get output signal format - * @zldev: pointer to zl3073x device - * @index: output index - * - * Return: signal format of given output - */ -static inline u8 -zl3073x_dev_out_signal_format_get(struct zl3073x_dev *zldev, u8 index) -{ - const struct zl3073x_out *out = zl3073x_out_state_get(zldev, index); - - return zl3073x_out_signal_format_get(out); -} - /** * zl3073x_dev_out_dpll_get - get DPLL ID the output is driven by * @zldev: pointer to zl3073x device From d9db25723677c3741a0cf3643f7f7429fc983921 Mon Sep 17 00:00:00 2001 From: Ovidiu Panait Date: Thu, 13 Nov 2025 11:27:20 +0000 Subject: [PATCH 0866/1381] net: stmmac: Fix VLAN 0 deletion in vlan_del_hw_rx_fltr() When the "rx-vlan-filter" feature is enabled on a network device, the 8021q module automatically adds a VLAN 0 hardware filter when the device is brought administratively up. For stmmac, this causes vlan_add_hw_rx_fltr() to create a new entry for VID 0 in the mac_device_info->vlan_filter array, in the following format: VLAN_TAG_DATA_ETV | VLAN_TAG_DATA_VEN | vid Here, VLAN_TAG_DATA_VEN indicates that the hardware filter is enabled for that VID. However, on the delete path, vlan_del_hw_rx_fltr() searches the vlan_filter array by VID only, without verifying whether a VLAN entry is enabled. As a result, when the 8021q module attempts to remove VLAN 0, the function may mistakenly match a zero-initialized slot rather than the actual VLAN 0 entry, causing incorrect deletions and leaving stale entries in the hardware table. Fix this by verifying that the VLAN entry's enable bit (VLAN_TAG_DATA_VEN) is set before matching and deleting by VID. This ensures only active VLAN entries are removed and avoids leaving stale entries in the VLAN filter table, particularly for VLAN ID 0. Fixes: ed64639bc1e08 ("net: stmmac: Add support for VLAN Rx filtering") Signed-off-by: Ovidiu Panait Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/20251113112721.70500-2-ovidiu.panait.rb@renesas.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/stmicro/stmmac/stmmac_vlan.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_vlan.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_vlan.c index ff02a79c00d4..b18404dd5a8b 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_vlan.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_vlan.c @@ -122,7 +122,8 @@ static int vlan_del_hw_rx_fltr(struct net_device *dev, /* Extended Rx VLAN Filter Enable */ for (i = 0; i < hw->num_vlan; i++) { - if ((hw->vlan_filter[i] & VLAN_TAG_DATA_VID) == vid) { + if ((hw->vlan_filter[i] & VLAN_TAG_DATA_VEN) && + ((hw->vlan_filter[i] & VLAN_TAG_DATA_VID) == vid)) { ret = vlan_write_filter(dev, hw, i, 0); if (!ret) From c171e679ee66d7c0e2b58db9531af96797a76bca Mon Sep 17 00:00:00 2001 From: Ovidiu Panait Date: Thu, 13 Nov 2025 11:27:21 +0000 Subject: [PATCH 0867/1381] net: stmmac: Disable EEE RX clock stop when VLAN is enabled On the Renesas RZ/V2H EVK platform, where the stmmac MAC is connected to a Microchip KSZ9131RNXI PHY, creating or deleting VLAN interfaces may fail with timeouts: # ip link add link end1 name end1.5 type vlan id 5 15c40000.ethernet end1: Timeout accessing MAC_VLAN_Tag_Filter RTNETLINK answers: Device or resource busy Disabling EEE at runtime avoids the problem: # ethtool --set-eee end1 eee off # ip link add link end1 name end1.5 type vlan id 5 # ip link del end1.5 The stmmac hardware requires the receive clock to be running when writing certain registers, such as those used for MAC address configuration or VLAN filtering. However, by default the driver enables Energy Efficient Ethernet (EEE) and allows the PHY to stop the receive clock when the link is idle. As a result, the RX clock might be stopped when attempting to access these registers, leading to timeouts and other issues. Commit dd557266cf5fb ("net: stmmac: block PHY RXC clock-stop") addressed this issue for most register accesses by wrapping them in phylink_rx_clk_stop_block()/phylink_rx_clk_stop_unblock() calls. However, VLAN add/delete operations may be invoked with bottom halves disabled, where sleeping is not allowed, so using these helpers is not possible. Therefore, to fix this, disable the RX clock stop feature in the phylink configuration if VLAN features are set. This ensures the RX clock remains active and register accesses succeed during VLAN operations. Signed-off-by: Ovidiu Panait Reviewed-by: Russell King (Oracle) Link: https://patch.msgid.link/20251113112721.70500-3-ovidiu.panait.rb@renesas.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index d08ff8f5ff15..a862d5f91fce 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -1245,7 +1245,11 @@ static int stmmac_phylink_setup(struct stmmac_priv *priv) /* Stmmac always requires an RX clock for hardware initialization */ config->mac_requires_rxc = true; - if (!(priv->plat->flags & STMMAC_FLAG_RX_CLK_RUNS_IN_LPI)) + /* Disable EEE RX clock stop to ensure VLAN register access works + * correctly. + */ + if (!(priv->plat->flags & STMMAC_FLAG_RX_CLK_RUNS_IN_LPI) && + !(priv->dev->features & NETIF_F_VLAN_FEATURES)) config->eee_rx_clk_stop_enable = true; /* Set the default transmit clock stop bit based on the platform glue */ From 45a1cd8346ca245a1ca475b26eb6ceb9d8b7c6f0 Mon Sep 17 00:00:00 2001 From: Sunday Adelodun Date: Thu, 13 Nov 2025 12:28:02 +0100 Subject: [PATCH 0868/1381] selftests: af_unix: Add tests for ECONNRESET and EOF semantics MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Add selftests to verify and document Linux’s intended behaviour for UNIX domain sockets (SOCK_STREAM and SOCK_DGRAM) when a peer closes. The tests verify that: 1. SOCK_STREAM returns EOF when the peer closes normally. 2. SOCK_STREAM returns ECONNRESET if the peer closes with unread data. 3. SOCK_SEQPACKET returns EOF when the peer closes normally. 4. SOCK_SEQPACKET returns ECONNRESET if the peer closes with unread data. 5. SOCK_DGRAM does not return ECONNRESET when the peer closes. This follows up on review feedback suggesting a selftest to clarify Linux’s semantics. Suggested-by: Kuniyuki Iwashima Signed-off-by: Sunday Adelodun Link: https://patch.msgid.link/20251113112802.44657-1-adelodunolaoluwa@yahoo.com Signed-off-by: Paolo Abeni --- tools/testing/selftests/net/.gitignore | 1 + tools/testing/selftests/net/af_unix/Makefile | 1 + .../selftests/net/af_unix/unix_connreset.c | 177 ++++++++++++++++++ 3 files changed, 179 insertions(+) create mode 100644 tools/testing/selftests/net/af_unix/unix_connreset.c diff --git a/tools/testing/selftests/net/.gitignore b/tools/testing/selftests/net/.gitignore index 439101b518ee..e89a60581a13 100644 --- a/tools/testing/selftests/net/.gitignore +++ b/tools/testing/selftests/net/.gitignore @@ -65,3 +65,4 @@ udpgso udpgso_bench_rx udpgso_bench_tx unix_connect +unix_connreset diff --git a/tools/testing/selftests/net/af_unix/Makefile b/tools/testing/selftests/net/af_unix/Makefile index de805cbbdf69..5826a8372451 100644 --- a/tools/testing/selftests/net/af_unix/Makefile +++ b/tools/testing/selftests/net/af_unix/Makefile @@ -7,6 +7,7 @@ TEST_GEN_PROGS := \ scm_pidfd \ scm_rights \ unix_connect \ + unix_connreset \ # end of TEST_GEN_PROGS include ../../lib.mk diff --git a/tools/testing/selftests/net/af_unix/unix_connreset.c b/tools/testing/selftests/net/af_unix/unix_connreset.c new file mode 100644 index 000000000000..bffef2b54bfd --- /dev/null +++ b/tools/testing/selftests/net/af_unix/unix_connreset.c @@ -0,0 +1,177 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Selftest for AF_UNIX socket close and ECONNRESET behaviour. + * + * This test verifies: + * 1. SOCK_STREAM returns EOF when the peer closes normally. + * 2. SOCK_STREAM returns ECONNRESET if peer closes with unread data. + * 3. SOCK_SEQPACKET returns EOF when the peer closes normally. + * 4. SOCK_SEQPACKET returns ECONNRESET if the peer closes with unread data. + * 5. SOCK_DGRAM does not return ECONNRESET when the peer closes. + * + * These tests document the intended Linux behaviour. + * + */ + +#define _GNU_SOURCE +#include +#include +#include +#include +#include +#include +#include +#include "../../kselftest_harness.h" + +#define SOCK_PATH "/tmp/af_unix_connreset.sock" + +static void remove_socket_file(void) +{ + unlink(SOCK_PATH); +} + +FIXTURE(unix_sock) +{ + int server; + int client; + int child; +}; + +FIXTURE_VARIANT(unix_sock) +{ + int socket_type; + const char *name; +}; + +FIXTURE_VARIANT_ADD(unix_sock, stream) { + .socket_type = SOCK_STREAM, + .name = "SOCK_STREAM", +}; + +FIXTURE_VARIANT_ADD(unix_sock, dgram) { + .socket_type = SOCK_DGRAM, + .name = "SOCK_DGRAM", +}; + +FIXTURE_VARIANT_ADD(unix_sock, seqpacket) { + .socket_type = SOCK_SEQPACKET, + .name = "SOCK_SEQPACKET", +}; + +FIXTURE_SETUP(unix_sock) +{ + struct sockaddr_un addr = {}; + int err; + + addr.sun_family = AF_UNIX; + strcpy(addr.sun_path, SOCK_PATH); + remove_socket_file(); + + self->server = socket(AF_UNIX, variant->socket_type, 0); + ASSERT_LT(-1, self->server); + + err = bind(self->server, (struct sockaddr *)&addr, sizeof(addr)); + ASSERT_EQ(0, err); + + if (variant->socket_type == SOCK_STREAM || + variant->socket_type == SOCK_SEQPACKET) { + err = listen(self->server, 1); + ASSERT_EQ(0, err); + } + + self->client = socket(AF_UNIX, variant->socket_type | SOCK_NONBLOCK, 0); + ASSERT_LT(-1, self->client); + + err = connect(self->client, (struct sockaddr *)&addr, sizeof(addr)); + ASSERT_EQ(0, err); +} + +FIXTURE_TEARDOWN(unix_sock) +{ + if (variant->socket_type == SOCK_STREAM || + variant->socket_type == SOCK_SEQPACKET) + close(self->child); + + close(self->client); + close(self->server); + remove_socket_file(); +} + +/* Test 1: peer closes normally */ +TEST_F(unix_sock, eof) +{ + char buf[16] = {}; + ssize_t n; + + if (variant->socket_type == SOCK_STREAM || + variant->socket_type == SOCK_SEQPACKET) { + self->child = accept(self->server, NULL, NULL); + ASSERT_LT(-1, self->child); + + close(self->child); + } else { + close(self->server); + } + + n = recv(self->client, buf, sizeof(buf), 0); + + if (variant->socket_type == SOCK_STREAM || + variant->socket_type == SOCK_SEQPACKET) { + ASSERT_EQ(0, n); + } else { + ASSERT_EQ(-1, n); + ASSERT_EQ(EAGAIN, errno); + } +} + +/* Test 2: peer closes with unread data */ +TEST_F(unix_sock, reset_unread_behavior) +{ + char buf[16] = {}; + ssize_t n; + + /* Send data that will remain unread */ + send(self->client, "hello", 5, 0); + + if (variant->socket_type == SOCK_DGRAM) { + /* No real connection, just close the server */ + close(self->server); + } else { + self->child = accept(self->server, NULL, NULL); + ASSERT_LT(-1, self->child); + + /* Peer closes before client reads */ + close(self->child); + } + + n = recv(self->client, buf, sizeof(buf), 0); + ASSERT_EQ(-1, n); + + if (variant->socket_type == SOCK_STREAM || + variant->socket_type == SOCK_SEQPACKET) { + ASSERT_EQ(ECONNRESET, errno); + } else { + ASSERT_EQ(EAGAIN, errno); + } +} + +/* Test 3: closing unaccepted (embryo) server socket should reset client. */ +TEST_F(unix_sock, reset_closed_embryo) +{ + char buf[16] = {}; + ssize_t n; + + if (variant->socket_type == SOCK_DGRAM) + SKIP(return, "This test only applies to SOCK_STREAM and SOCK_SEQPACKET"); + + /* Close server without accept()ing */ + close(self->server); + + n = recv(self->client, buf, sizeof(buf), 0); + + ASSERT_EQ(-1, n); + ASSERT_EQ(ECONNRESET, errno); +} + +TEST_HARNESS_MAIN + From 19f1d6c7230b7da3775a2db915dc8eb9d3151658 Mon Sep 17 00:00:00 2001 From: Horatiu Vultur Date: Fri, 14 Nov 2025 09:42:24 +0100 Subject: [PATCH 0869/1381] net: phy: micrel: lan8814: Enable in-band auto-negotiation The lan8814 supports two interfaces towards the host (QSGMII and QUSGMII). Currently the lan8814 disables the auto-negotiation towards the host side. So, extend this to allow to configure to use in-band auto-negotiation. I have tested this only with the QSGMII interface. Signed-off-by: Horatiu Vultur Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/20251114084224.3268928-1-horatiu.vultur@microchip.com Signed-off-by: Paolo Abeni --- drivers/net/phy/micrel.c | 26 +++++++++++++++++++++----- 1 file changed, 21 insertions(+), 5 deletions(-) diff --git a/drivers/net/phy/micrel.c b/drivers/net/phy/micrel.c index 57ea947369fe..5d90ccc20df7 100644 --- a/drivers/net/phy/micrel.c +++ b/drivers/net/phy/micrel.c @@ -2988,6 +2988,8 @@ static int ksz886x_cable_test_get_status(struct phy_device *phydev, #define LAN_EXT_PAGE_ACCESS_ADDRESS_DATA 0x17 #define LAN_EXT_PAGE_ACCESS_CTRL_EP_FUNC 0x4000 +#define LAN8814_QSGMII_TX_CONFIG 0x35 +#define LAN8814_QSGMII_TX_CONFIG_QSGMII BIT(3) #define LAN8814_QSGMII_SOFT_RESET 0x43 #define LAN8814_QSGMII_SOFT_RESET_BIT BIT(0) #define LAN8814_QSGMII_PCS1G_ANEG_CONFIG 0x13 @@ -4501,12 +4503,24 @@ static void lan8814_setup_led(struct phy_device *phydev, int val) static int lan8814_config_init(struct phy_device *phydev) { struct kszphy_priv *lan8814 = phydev->priv; + int ret; - /* Disable ANEG with QSGMII PCS Host side */ - lanphy_modify_page_reg(phydev, LAN8814_PAGE_PORT_REGS, - LAN8814_QSGMII_PCS1G_ANEG_CONFIG, - LAN8814_QSGMII_PCS1G_ANEG_CONFIG_ANEG_ENA, - 0); + /* Based on the interface type select how the advertise ability is + * encoded, to set as SGMII or as USGMII. + */ + if (phydev->interface == PHY_INTERFACE_MODE_QSGMII) + ret = lanphy_modify_page_reg(phydev, LAN8814_PAGE_COMMON_REGS, + LAN8814_QSGMII_TX_CONFIG, + LAN8814_QSGMII_TX_CONFIG_QSGMII, + LAN8814_QSGMII_TX_CONFIG_QSGMII); + else + ret = lanphy_modify_page_reg(phydev, LAN8814_PAGE_COMMON_REGS, + LAN8814_QSGMII_TX_CONFIG, + LAN8814_QSGMII_TX_CONFIG_QSGMII, + 0); + + if (ret < 0) + return ret; /* MDI-X setting for swap A,B transmit */ lanphy_modify_page_reg(phydev, LAN8814_PAGE_PCS_DIGITAL, LAN8814_ALIGN_SWAP, @@ -6640,6 +6654,8 @@ static struct phy_driver ksphy_driver[] = { .suspend = genphy_suspend, .resume = kszphy_resume, .config_intr = lan8814_config_intr, + .inband_caps = lan8842_inband_caps, + .config_inband = lan8842_config_inband, .handle_interrupt = lan8814_handle_interrupt, .cable_test_start = lan8814_cable_test_start, .cable_test_get_status = ksz886x_cable_test_get_status, From 46e7860ef941fa92fcc61c79caac315839f86df5 Mon Sep 17 00:00:00 2001 From: Tim Hostetler Date: Fri, 14 Nov 2025 13:11:43 -0800 Subject: [PATCH 0870/1381] gve: Move ptp_schedule_worker to gve_init_clock Previously, gve had been only initializing ptp aux work when hardware timestamping was initialized through ndo_hwtsatmp_set. As this patch series introduces XDP hardware timestamp metadata which will require the ptp aux work, the work can't be gated on the kernel_hwtstamp_config being set and must be initialized elsewhere. For simplicity, ptp_schedule_worker is invoked right after the ptp_clock is registered with the kernel (which happens during gve_probe or following reset). The worker is scheduled in GVE_NIC_TS_SYNC_INTERVAL_MS as the synchronous call to gve_clock_nic_ts_read makes the worker redundant if scheduled immediately. If gve cannot read the device clock immediately, it errors out of gve_init_clock. Signed-off-by: Tim Hostetler Reviewed-by: Willem de Bruijn Reviewed-by: Harshitha Ramamurthy Signed-off-by: Joshua Washington Link: https://patch.msgid.link/20251114211146.292068-2-joshwash@google.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/google/gve/gve_main.c | 4 ---- drivers/net/ethernet/google/gve/gve_ptp.c | 12 ++++++++++++ 2 files changed, 12 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/google/gve/gve_main.c b/drivers/net/ethernet/google/gve/gve_main.c index 6fb8fbb38a7d..2b41a42fb516 100644 --- a/drivers/net/ethernet/google/gve/gve_main.c +++ b/drivers/net/ethernet/google/gve/gve_main.c @@ -2188,10 +2188,6 @@ static int gve_set_ts_config(struct net_device *dev, } kernel_config->rx_filter = HWTSTAMP_FILTER_ALL; - gve_clock_nic_ts_read(priv); - ptp_schedule_worker(priv->ptp->clock, 0); - } else { - ptp_cancel_worker_sync(priv->ptp->clock); } priv->ts_config.rx_filter = kernel_config->rx_filter; diff --git a/drivers/net/ethernet/google/gve/gve_ptp.c b/drivers/net/ethernet/google/gve/gve_ptp.c index a384a9ed4914..073677d82ee8 100644 --- a/drivers/net/ethernet/google/gve/gve_ptp.c +++ b/drivers/net/ethernet/google/gve/gve_ptp.c @@ -133,9 +133,21 @@ int gve_init_clock(struct gve_priv *priv) err = -ENOMEM; goto release_ptp; } + err = gve_clock_nic_ts_read(priv); + if (err) { + dev_err(&priv->pdev->dev, "failed to read NIC clock %d\n", err); + goto release_nic_ts_report; + } + ptp_schedule_worker(priv->ptp->clock, + msecs_to_jiffies(GVE_NIC_TS_SYNC_INTERVAL_MS)); return 0; +release_nic_ts_report: + dma_free_coherent(&priv->pdev->dev, + sizeof(struct gve_nic_ts_report), + priv->nic_ts_report, priv->nic_ts_report_bus); + priv->nic_ts_report = NULL; release_ptp: gve_ptp_release(priv); return err; From f356a66b87bb19e5370dee469ec32e0094ac9da3 Mon Sep 17 00:00:00 2001 From: Tim Hostetler Date: Fri, 14 Nov 2025 13:11:44 -0800 Subject: [PATCH 0871/1381] gve: Wrap struct xdp_buff RX timestamping will need to keep track of extra temporary information per-packet. In preparation for this, introduce gve_xdp_buff to wrap the xdp_buff. This is similar in function to stmmac_xdp_buff and ice_xdp_buff. Signed-off-by: Tim Hostetler Reviewed-by: Willem de Bruijn Reviewed-by: Harshitha Ramamurthy Signed-off-by: Joshua Washington Link: https://patch.msgid.link/20251114211146.292068-3-joshwash@google.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/google/gve/gve.h | 5 +++++ drivers/net/ethernet/google/gve/gve_rx_dqo.c | 16 ++++++++-------- 2 files changed, 13 insertions(+), 8 deletions(-) diff --git a/drivers/net/ethernet/google/gve/gve.h b/drivers/net/ethernet/google/gve/gve.h index a33b44c1eb86..a21e599cf710 100644 --- a/drivers/net/ethernet/google/gve/gve.h +++ b/drivers/net/ethernet/google/gve/gve.h @@ -205,6 +205,11 @@ struct gve_rx_buf_state_dqo { s16 next; }; +/* Wrapper for XDP Rx metadata */ +struct gve_xdp_buff { + struct xdp_buff xdp; +}; + /* `head` and `tail` are indices into an array, or -1 if empty. */ struct gve_index_list { s16 head; diff --git a/drivers/net/ethernet/google/gve/gve_rx_dqo.c b/drivers/net/ethernet/google/gve/gve_rx_dqo.c index 1aff3bbb8cfc..76b26896f572 100644 --- a/drivers/net/ethernet/google/gve/gve_rx_dqo.c +++ b/drivers/net/ethernet/google/gve/gve_rx_dqo.c @@ -840,23 +840,23 @@ static int gve_rx_dqo(struct napi_struct *napi, struct gve_rx_ring *rx, } if (xprog) { - struct xdp_buff xdp; + struct gve_xdp_buff gve_xdp; void *old_data; int xdp_act; - xdp_init_buff(&xdp, buf_state->page_info.buf_size, + xdp_init_buff(&gve_xdp.xdp, buf_state->page_info.buf_size, &rx->xdp_rxq); - xdp_prepare_buff(&xdp, + xdp_prepare_buff(&gve_xdp.xdp, buf_state->page_info.page_address + buf_state->page_info.page_offset, buf_state->page_info.pad, buf_len, false); - old_data = xdp.data; - xdp_act = bpf_prog_run_xdp(xprog, &xdp); - buf_state->page_info.pad += xdp.data - old_data; - buf_len = xdp.data_end - xdp.data; + old_data = gve_xdp.xdp.data; + xdp_act = bpf_prog_run_xdp(xprog, &gve_xdp.xdp); + buf_state->page_info.pad += gve_xdp.xdp.data - old_data; + buf_len = gve_xdp.xdp.data_end - gve_xdp.xdp.data; if (xdp_act != XDP_PASS) { - gve_xdp_done_dqo(priv, rx, &xdp, xprog, xdp_act, + gve_xdp_done_dqo(priv, rx, &gve_xdp.xdp, xprog, xdp_act, buf_state); return 0; } From 66adaf1021282edcb9be50bea5ea5aa78afda527 Mon Sep 17 00:00:00 2001 From: Tim Hostetler Date: Fri, 14 Nov 2025 13:11:45 -0800 Subject: [PATCH 0872/1381] gve: Prepare bpf_xdp_metadata_rx_timestamp support Support populating XDP RX metadata with hardware RX timestamps. This patch utilizes the same underlying logic to calculate hardware timestamps as the regular RX path. xdp_metadata_ops is registered with the net_device in a future patch. gve_rx_calculate_hwtstamp was pulled out so as to not duplicate logic between gve_xdp_rx_timestamp and gve_rx_hwtstamp. Signed-off-by: Tim Hostetler Reviewed-by: Willem de Bruijn Reviewed-by: Harshitha Ramamurthy Signed-off-by: Joshua Washington Link: https://patch.msgid.link/20251114211146.292068-4-joshwash@google.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/google/gve/gve.h | 2 + drivers/net/ethernet/google/gve/gve_dqo.h | 1 + drivers/net/ethernet/google/gve/gve_rx_dqo.c | 39 +++++++++++++++----- 3 files changed, 33 insertions(+), 9 deletions(-) diff --git a/drivers/net/ethernet/google/gve/gve.h b/drivers/net/ethernet/google/gve/gve.h index a21e599cf710..970d5ca8cdde 100644 --- a/drivers/net/ethernet/google/gve/gve.h +++ b/drivers/net/ethernet/google/gve/gve.h @@ -208,6 +208,8 @@ struct gve_rx_buf_state_dqo { /* Wrapper for XDP Rx metadata */ struct gve_xdp_buff { struct xdp_buff xdp; + struct gve_priv *gve; + const struct gve_rx_compl_desc_dqo *compl_desc; }; /* `head` and `tail` are indices into an array, or -1 if empty. */ diff --git a/drivers/net/ethernet/google/gve/gve_dqo.h b/drivers/net/ethernet/google/gve/gve_dqo.h index 6eb442096e02..5871f773f0c7 100644 --- a/drivers/net/ethernet/google/gve/gve_dqo.h +++ b/drivers/net/ethernet/google/gve/gve_dqo.h @@ -36,6 +36,7 @@ netdev_tx_t gve_tx_dqo(struct sk_buff *skb, struct net_device *dev); netdev_features_t gve_features_check_dqo(struct sk_buff *skb, struct net_device *dev, netdev_features_t features); +int gve_xdp_rx_timestamp(const struct xdp_md *_ctx, u64 *timestamp); bool gve_tx_poll_dqo(struct gve_notify_block *block, bool do_clean); bool gve_xdp_poll_dqo(struct gve_notify_block *block); bool gve_xsk_tx_poll_dqo(struct gve_notify_block *block, int budget); diff --git a/drivers/net/ethernet/google/gve/gve_rx_dqo.c b/drivers/net/ethernet/google/gve/gve_rx_dqo.c index 76b26896f572..f20d1b1d06e6 100644 --- a/drivers/net/ethernet/google/gve/gve_rx_dqo.c +++ b/drivers/net/ethernet/google/gve/gve_rx_dqo.c @@ -456,20 +456,38 @@ static void gve_rx_skb_hash(struct sk_buff *skb, * Note that this means if the time delta between packet reception and the last * clock read is greater than ~2 seconds, this will provide invalid results. */ +static ktime_t gve_rx_get_hwtstamp(struct gve_priv *gve, u32 hwts) +{ + u64 last_read = READ_ONCE(gve->last_sync_nic_counter); + u32 low = (u32)last_read; + s32 diff = hwts - low; + + return ns_to_ktime(last_read + diff); +} + static void gve_rx_skb_hwtstamp(struct gve_rx_ring *rx, const struct gve_rx_compl_desc_dqo *desc) { - u64 last_read = READ_ONCE(rx->gve->last_sync_nic_counter); struct sk_buff *skb = rx->ctx.skb_head; - u32 ts, low; - s32 diff; - if (desc->ts_sub_nsecs_low & GVE_DQO_RX_HWTSTAMP_VALID) { - ts = le32_to_cpu(desc->ts); - low = (u32)last_read; - diff = ts - low; - skb_hwtstamps(skb)->hwtstamp = ns_to_ktime(last_read + diff); - } + if (desc->ts_sub_nsecs_low & GVE_DQO_RX_HWTSTAMP_VALID) + skb_hwtstamps(skb)->hwtstamp = + gve_rx_get_hwtstamp(rx->gve, le32_to_cpu(desc->ts)); +} + +int gve_xdp_rx_timestamp(const struct xdp_md *_ctx, u64 *timestamp) +{ + const struct gve_xdp_buff *ctx = (void *)_ctx; + + if (!ctx->gve->nic_ts_report) + return -ENODATA; + + if (!(ctx->compl_desc->ts_sub_nsecs_low & GVE_DQO_RX_HWTSTAMP_VALID)) + return -ENODATA; + + *timestamp = gve_rx_get_hwtstamp(ctx->gve, + le32_to_cpu(ctx->compl_desc->ts)); + return 0; } static void gve_rx_free_skb(struct napi_struct *napi, struct gve_rx_ring *rx) @@ -851,6 +869,9 @@ static int gve_rx_dqo(struct napi_struct *napi, struct gve_rx_ring *rx, buf_state->page_info.page_offset, buf_state->page_info.pad, buf_len, false); + gve_xdp.gve = priv; + gve_xdp.compl_desc = compl_desc; + old_data = gve_xdp.xdp.data; xdp_act = bpf_prog_run_xdp(xprog, &gve_xdp.xdp); buf_state->page_info.pad += gve_xdp.xdp.data - old_data; From 1b42e07af1ee65d10ee72f90d6fbd02a9b53208c Mon Sep 17 00:00:00 2001 From: Tim Hostetler Date: Fri, 14 Nov 2025 13:11:46 -0800 Subject: [PATCH 0873/1381] gve: Add Rx HWTS metadata to AF_XDP ZC mode By overlaying the struct gve_xdp_buff on top of the struct xdp_buff_xsk that AF_XDP utilizes, the driver records the 32 bit timestamp via the completion descriptor and the cached 64 bit NIC timestamp via gve_priv. The driver's implementation of xmo_rx_timestamp extends the timestamp to the full and up to date 64 bit timestamp and returns it to the user. gve_rx_xsk_dqo is modified to accept a pointer to the completion descriptor and no longer takes a buf_len explicitly as it can be pulled out of the descriptor. With this patch gve now supports bpf_xdp_metadata_rx_timestamp. Signed-off-by: Tim Hostetler Reviewed-by: Willem de Bruijn Reviewed-by: Harshitha Ramamurthy Signed-off-by: Joshua Washington Link: https://patch.msgid.link/20251114211146.292068-5-joshwash@google.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/google/gve/gve_main.c | 7 +++++++ drivers/net/ethernet/google/gve/gve_rx_dqo.c | 16 ++++++++++++++-- 2 files changed, 21 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/google/gve/gve_main.c b/drivers/net/ethernet/google/gve/gve_main.c index 2b41a42fb516..a5a2b18d309b 100644 --- a/drivers/net/ethernet/google/gve/gve_main.c +++ b/drivers/net/ethernet/google/gve/gve_main.c @@ -2348,6 +2348,10 @@ static void gve_set_netdev_xdp_features(struct gve_priv *priv) xdp_set_features_flag_locked(priv->dev, xdp_features); } +static const struct xdp_metadata_ops gve_xdp_metadata_ops = { + .xmo_rx_timestamp = gve_xdp_rx_timestamp, +}; + static int gve_init_priv(struct gve_priv *priv, bool skip_describe_device) { int num_ntfy; @@ -2443,6 +2447,9 @@ static int gve_init_priv(struct gve_priv *priv, bool skip_describe_device) } gve_set_netdev_xdp_features(priv); + if (!gve_is_gqi(priv)) + priv->dev->xdp_metadata_ops = &gve_xdp_metadata_ops; + err = gve_setup_device_resources(priv); if (err) goto err_free_xsk_bitmap; diff --git a/drivers/net/ethernet/google/gve/gve_rx_dqo.c b/drivers/net/ethernet/google/gve/gve_rx_dqo.c index f20d1b1d06e6..f1bd8f5d5732 100644 --- a/drivers/net/ethernet/google/gve/gve_rx_dqo.c +++ b/drivers/net/ethernet/google/gve/gve_rx_dqo.c @@ -240,6 +240,11 @@ int gve_rx_alloc_ring_dqo(struct gve_priv *priv, rx->rx_headroom = 0; } + /* struct gve_xdp_buff is overlaid on struct xdp_buff_xsk and utilizes + * the 24 byte field cb to store gve specific data. + */ + XSK_CHECK_PRIV_TYPE(struct gve_xdp_buff); + rx->dqo.num_buf_states = cfg->raw_addressing ? buffer_queue_slots : gve_get_rx_pages_per_qpl_dqo(cfg->ring_size); rx->dqo.buf_states = kvcalloc_node(rx->dqo.num_buf_states, @@ -701,16 +706,23 @@ static void gve_xdp_done_dqo(struct gve_priv *priv, struct gve_rx_ring *rx, } static int gve_rx_xsk_dqo(struct napi_struct *napi, struct gve_rx_ring *rx, - struct gve_rx_buf_state_dqo *buf_state, int buf_len, + const struct gve_rx_compl_desc_dqo *compl_desc, + struct gve_rx_buf_state_dqo *buf_state, struct bpf_prog *xprog) { struct xdp_buff *xdp = buf_state->xsk_buff; + int buf_len = compl_desc->packet_len; struct gve_priv *priv = rx->gve; + struct gve_xdp_buff *gve_xdp; int xdp_act; xdp->data_end = xdp->data + buf_len; xsk_buff_dma_sync_for_cpu(xdp); + gve_xdp = (void *)xdp; + gve_xdp->gve = priv; + gve_xdp->compl_desc = compl_desc; + if (xprog) { xdp_act = bpf_prog_run_xdp(xprog, xdp); buf_len = xdp->data_end - xdp->data; @@ -800,7 +812,7 @@ static int gve_rx_dqo(struct napi_struct *napi, struct gve_rx_ring *rx, xprog = READ_ONCE(priv->xdp_prog); if (buf_state->xsk_buff) - return gve_rx_xsk_dqo(napi, rx, buf_state, buf_len, xprog); + return gve_rx_xsk_dqo(napi, rx, compl_desc, buf_state, xprog); /* Page might have not been used for awhile and was likely last written * by a different thread. From 6770eaad75ae039b620e9dc05f62dcbcfc2e5085 Mon Sep 17 00:00:00 2001 From: Donald Hunter Date: Mon, 17 Nov 2025 14:31:54 +0000 Subject: [PATCH 0874/1381] tools: ynltool: ignore *.d deps files Add *.d to gitignore for ynltool Signed-off-by: Donald Hunter Link: https://patch.msgid.link/20251117143155.44806-1-donald.hunter@gmail.com Signed-off-by: Jakub Kicinski --- tools/net/ynl/ynltool/.gitignore | 1 + 1 file changed, 1 insertion(+) diff --git a/tools/net/ynl/ynltool/.gitignore b/tools/net/ynl/ynltool/.gitignore index f38848dbb0d3..690d399c921a 100644 --- a/tools/net/ynl/ynltool/.gitignore +++ b/tools/net/ynl/ynltool/.gitignore @@ -1 +1,2 @@ ynltool +*.d From f4e3402f59ac5e6e629d8b94c3760414fb0f37bf Mon Sep 17 00:00:00 2001 From: Pierre-Henry Moussay Date: Mon, 17 Nov 2025 16:24:33 +0000 Subject: [PATCH 0875/1381] dt-bindings: net: cdns,macb: Add pic64gx compatibility The pic64gx uses an identical integration of the macb IP to mpfs. Signed-off-by: Pierre-Henry Moussay Signed-off-by: Conor Dooley Acked-by: Nicolas Ferre Acked-by: Rob Herring (Arm) Link: https://patch.msgid.link/20251117-easter-machine-37851f20aaf3@spud Signed-off-by: Jakub Kicinski --- Documentation/devicetree/bindings/net/cdns,macb.yaml | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/Documentation/devicetree/bindings/net/cdns,macb.yaml b/Documentation/devicetree/bindings/net/cdns,macb.yaml index ea8337846ab2..cb14c35ba996 100644 --- a/Documentation/devicetree/bindings/net/cdns,macb.yaml +++ b/Documentation/devicetree/bindings/net/cdns,macb.yaml @@ -38,7 +38,10 @@ properties: - cdns,sam9x60-macb # Microchip sam9x60 SoC - microchip,mpfs-macb # Microchip PolarFire SoC - const: cdns,macb # Generic - + - items: + - const: microchip,pic64gx-macb # Microchip PIC64GX SoC + - const: microchip,mpfs-macb # Microchip PolarFire SoC + - const: cdns,macb # Generic - items: - enum: - atmel,sama5d3-macb # 10/100Mbit IP on Atmel sama5d3 SoCs From 932478bf9f6d3c7da9635c2fd0f87766c72ae683 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Tue, 18 Nov 2025 00:04:40 +0000 Subject: [PATCH 0876/1381] tcp: Don't reinitialise tw->tw_transparent in tcp_time_wait(). tw->tw_transparent is initialised twice in inet_twsk_alloc() and tcp_time_wait(). Let's remove the latter. Signed-off-by: Kuniyuki Iwashima Reviewed-by: Jason Xing Link: https://patch.msgid.link/20251118000445.4091280-1-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/tcp_minisocks.c | 1 - 1 file changed, 1 deletion(-) diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index d8f4d813e8dd..bd5462154f97 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -337,7 +337,6 @@ void tcp_time_wait(struct sock *sk, int state, int timeo) struct tcp_timewait_sock *tcptw = tcp_twsk((struct sock *)tw); const int rto = (icsk->icsk_rto << 2) - (icsk->icsk_rto >> 1); - tw->tw_transparent = inet_test_bit(TRANSPARENT, sk); tw->tw_mark = sk->sk_mark; tw->tw_priority = READ_ONCE(sk->sk_priority); tw->tw_rcv_wscale = tp->rx_opt.rcv_wscale; From efb238160e889e377578411d55abf47b1073492a Mon Sep 17 00:00:00 2001 From: Thorsten Blum Date: Sun, 16 Nov 2025 14:56:14 +0100 Subject: [PATCH 0877/1381] kcm: Fix typo and add hyphen in Kconfig help text s/connectons/connections/ and s/message based/message-based/ Signed-off-by: Thorsten Blum Link: https://patch.msgid.link/20251116135616.106079-2-thorsten.blum@linux.dev Signed-off-by: Jakub Kicinski --- net/kcm/Kconfig | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/kcm/Kconfig b/net/kcm/Kconfig index 16f39f2565d9..66660a06cacf 100644 --- a/net/kcm/Kconfig +++ b/net/kcm/Kconfig @@ -7,5 +7,5 @@ config AF_KCM select STREAM_PARSER help KCM (Kernel Connection Multiplexor) sockets provide a method - for multiplexing messages of a message based application - protocol over kernel connectons (e.g. TCP connections). + for multiplexing messages of a message-based application + protocol over kernel connections (e.g. TCP connections). From eb74ae2f87d254e54ab15429e845c2c46f8e970f Mon Sep 17 00:00:00 2001 From: Alok Tiwari Date: Mon, 17 Nov 2025 01:53:50 -0800 Subject: [PATCH 0878/1381] net: dsa: ks8995: Fix incorrect OF match table name The driver declares an OF match table named ks8895_spi_of_match, even though it describes compatible strings for the KS8995 and related Micrel switches. This is a leftover typo, the correct name should match the chip family handled by this driver ks8995, and also match the variable used in spi_driver.of_match_table. Signed-off-by: Alok Tiwari Reviewed-by: Andrew Lunn Reviewed-by: Linus Walleij Link: https://patch.msgid.link/20251117095356.2099772-1-alok.a.tiwari@oracle.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/ks8995.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/net/dsa/ks8995.c b/drivers/net/dsa/ks8995.c index 5c4c83e00477..77d8b842693c 100644 --- a/drivers/net/dsa/ks8995.c +++ b/drivers/net/dsa/ks8995.c @@ -203,13 +203,13 @@ static const struct spi_device_id ks8995_id[] = { }; MODULE_DEVICE_TABLE(spi, ks8995_id); -static const struct of_device_id ks8895_spi_of_match[] = { +static const struct of_device_id ks8995_spi_of_match[] = { { .compatible = "micrel,ks8995" }, { .compatible = "micrel,ksz8864" }, { .compatible = "micrel,ksz8795" }, { }, }; -MODULE_DEVICE_TABLE(of, ks8895_spi_of_match); +MODULE_DEVICE_TABLE(of, ks8995_spi_of_match); static inline u8 get_chip_id(u8 val) { @@ -842,7 +842,7 @@ static void ks8995_remove(struct spi_device *spi) static struct spi_driver ks8995_driver = { .driver = { .name = "spi-ks8995", - .of_match_table = ks8895_spi_of_match, + .of_match_table = ks8995_spi_of_match, }, .probe = ks8995_probe, .remove = ks8995_remove, From dac0236075684b9edfd71492e79a9122f3ef45cc Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sun, 16 Nov 2025 20:27:15 +0000 Subject: [PATCH 0879/1381] net: add a new @alloc parameter to napi_skb_cache_get() We want to be able in the series last patch to get an skb from napi_skb_cache from process context, if there is one available. Signed-off-by: Eric Dumazet Reviewed-by: Jason Xing Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20251116202717.1542829-2-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/core/skbuff.c | 15 ++++++++------- 1 file changed, 8 insertions(+), 7 deletions(-) diff --git a/net/core/skbuff.c b/net/core/skbuff.c index f34372666e67..88b5530f9c46 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -280,17 +280,18 @@ EXPORT_SYMBOL(__netdev_alloc_frag_align); */ static u32 skbuff_cache_size __read_mostly; -static struct sk_buff *napi_skb_cache_get(void) +static struct sk_buff *napi_skb_cache_get(bool alloc) { struct napi_alloc_cache *nc = this_cpu_ptr(&napi_alloc_cache); struct sk_buff *skb; local_lock_nested_bh(&napi_alloc_cache.bh_lock); if (unlikely(!nc->skb_count)) { - nc->skb_count = kmem_cache_alloc_bulk(net_hotdata.skbuff_cache, - GFP_ATOMIC | __GFP_NOWARN, - NAPI_SKB_CACHE_BULK, - nc->skb_cache); + if (alloc) + nc->skb_count = kmem_cache_alloc_bulk(net_hotdata.skbuff_cache, + GFP_ATOMIC | __GFP_NOWARN, + NAPI_SKB_CACHE_BULK, + nc->skb_cache); if (unlikely(!nc->skb_count)) { local_unlock_nested_bh(&napi_alloc_cache.bh_lock); return NULL; @@ -530,7 +531,7 @@ static struct sk_buff *__napi_build_skb(void *data, unsigned int frag_size) { struct sk_buff *skb; - skb = napi_skb_cache_get(); + skb = napi_skb_cache_get(true); if (unlikely(!skb)) return NULL; @@ -659,7 +660,7 @@ struct sk_buff *__alloc_skb(unsigned int size, gfp_t gfp_mask, /* Get the HEAD */ if ((flags & (SKB_ALLOC_FCLONE | SKB_ALLOC_NAPI)) == SKB_ALLOC_NAPI && likely(node == NUMA_NO_NODE || node == numa_mem_id())) - skb = napi_skb_cache_get(); + skb = napi_skb_cache_get(true); else skb = kmem_cache_alloc_node(cache, gfp_mask & ~GFP_DMA, node); if (unlikely(!skb)) From 294e63825966319e815e469931cfbfc437004d0f Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sun, 16 Nov 2025 20:27:16 +0000 Subject: [PATCH 0880/1381] net: __alloc_skb() cleanup This patch refactors __alloc_skb() to prepare the following one, and does not change functionality. Signed-off-by: Eric Dumazet Reviewed-by: Jason Xing Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20251116202717.1542829-3-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/core/skbuff.c | 28 ++++++++++++++++++---------- 1 file changed, 18 insertions(+), 10 deletions(-) diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 88b5530f9c46..8a0a4ca7fa5d 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -646,25 +646,33 @@ static void *kmalloc_reserve(unsigned int *size, gfp_t flags, int node, struct sk_buff *__alloc_skb(unsigned int size, gfp_t gfp_mask, int flags, int node) { + struct sk_buff *skb = NULL; struct kmem_cache *cache; - struct sk_buff *skb; bool pfmemalloc; u8 *data; - cache = (flags & SKB_ALLOC_FCLONE) - ? net_hotdata.skbuff_fclone_cache : net_hotdata.skbuff_cache; - if (sk_memalloc_socks() && (flags & SKB_ALLOC_RX)) gfp_mask |= __GFP_MEMALLOC; - /* Get the HEAD */ - if ((flags & (SKB_ALLOC_FCLONE | SKB_ALLOC_NAPI)) == SKB_ALLOC_NAPI && - likely(node == NUMA_NO_NODE || node == numa_mem_id())) + if (flags & SKB_ALLOC_FCLONE) { + cache = net_hotdata.skbuff_fclone_cache; + goto fallback; + } + cache = net_hotdata.skbuff_cache; + if (unlikely(node != NUMA_NO_NODE && node != numa_mem_id())) + goto fallback; + + if (flags & SKB_ALLOC_NAPI) { skb = napi_skb_cache_get(true); - else + if (unlikely(!skb)) + return NULL; + } + if (!skb) { +fallback: skb = kmem_cache_alloc_node(cache, gfp_mask & ~GFP_DMA, node); - if (unlikely(!skb)) - return NULL; + if (unlikely(!skb)) + return NULL; + } prefetchw(skb); /* We do our best to align skb_shared_info on a separate cache From 21664814b89e1268bc48e9f641b813746a7dbaae Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sun, 16 Nov 2025 20:27:17 +0000 Subject: [PATCH 0881/1381] net: use napi_skb_cache even in process context This is a followup of commit e20dfbad8aab ("net: fix napi_consume_skb() with alien skbs"). Now the per-cpu napi_skb_cache is populated from TX completion path, we can make use of this cache, especially for cpus not used from a driver NAPI poll (primary user of napi_cache). We can use the napi_skb_cache only if current context is not from hard irq. With this patch, I consistently reach 130 Mpps on my UDP tx stress test and reduce SLUB spinlock contention to smaller values. Note there is still some SLUB contention for skb->head allocations. I had to tune /sys/kernel/slab/skbuff_small_head/cpu_partial and /sys/kernel/slab/skbuff_small_head/min_partial depending on the platform taxonomy. Signed-off-by: Eric Dumazet Reviewed-by: Jason Xing Tested-by: Jason Xing Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20251116202717.1542829-4-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/core/skbuff.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 8a0a4ca7fa5d..9feea830a4db 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -666,7 +666,12 @@ struct sk_buff *__alloc_skb(unsigned int size, gfp_t gfp_mask, skb = napi_skb_cache_get(true); if (unlikely(!skb)) return NULL; + } else if (!in_hardirq() && !irqs_disabled()) { + local_bh_disable(); + skb = napi_skb_cache_get(false); + local_bh_enable(); } + if (!skb) { fallback: skb = kmem_cache_alloc_node(cache, gfp_mask & ~GFP_DMA, node); From 4abe51dba69f1d72a13ba3b5d0dcc237cdd71404 Mon Sep 17 00:00:00 2001 From: Hangbin Liu Date: Mon, 17 Nov 2025 02:44:55 +0000 Subject: [PATCH 0882/1381] tools: ynl: Add MAC address parsing support Add missing support for parsing MAC addresses when display_hint is 'mac' in the YNL library. This enables YNL CLI to accept MAC address strings for attributes like lladdr in rt-neigh operations. Reviewed-by: Donald Hunter Signed-off-by: Hangbin Liu Link: https://patch.msgid.link/20251117024457.3034-2-liuhangbin@gmail.com Signed-off-by: Jakub Kicinski --- tools/net/ynl/pyynl/lib/ynl.py | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/tools/net/ynl/pyynl/lib/ynl.py b/tools/net/ynl/pyynl/lib/ynl.py index 225baad3c8f8..36d36eb7e3b8 100644 --- a/tools/net/ynl/pyynl/lib/ynl.py +++ b/tools/net/ynl/pyynl/lib/ynl.py @@ -985,6 +985,15 @@ class YnlFamily(SpecFamily): raw = bytes.fromhex(string) else: raw = int(string, 16) + elif attr_spec.display_hint == 'mac': + # Parse MAC address in format "00:11:22:33:44:55" or "001122334455" + if ':' in string: + mac_bytes = [int(x, 16) for x in string.split(':')] + else: + if len(string) % 2 != 0: + raise Exception(f"Invalid MAC address format: {string}") + mac_bytes = [int(string[i:i+2], 16) for i in range(0, len(string), 2)] + raw = bytes(mac_bytes) else: raise Exception(f"Display hint '{attr_spec.display_hint}' not implemented" f" when parsing '{attr_spec['name']}'") From 1064d521d177023358b3b516f250faa714415a6d Mon Sep 17 00:00:00 2001 From: Hangbin Liu Date: Mon, 17 Nov 2025 02:44:56 +0000 Subject: [PATCH 0883/1381] netlink: specs: support ipv4-or-v6 for dual-stack fields MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Since commit 1b255e1beabf ("tools: ynl: add ipv4-or-v6 display hint"), we can display either IPv4 or IPv6 addresses for a single field based on the address family. However, most dual-stack fields still use the ipv4 display hint. This update changes them to use the new ipv4-or-v6 display hint and converts IPv4-only fields to use the u32 type. Field changes: - v4-or-v6 - IFA_ADDRESS, IFA_LOCAL - IFLA_GRE_LOCAL, IFLA_GRE_REMOTE - IFLA_VTI_LOCAL, IFLA_VTI_REMOTE - IFLA_IPTUN_LOCAL, IFLA_IPTUN_REMOTE - NDA_DST - RTA_DST, RTA_SRC, RTA_GATEWAY, RTA_PREFSRC - FRA_SRC, FRA_DST - ipv4 - IFA_BROADCAST - IFLA_GENEVE_REMOTE - IFLA_IPTUN_6RD_RELAY_PREFIX Reviewed-by: Asbjørn Sloth Tønnesen Reviewed-by: Donald Hunter Signed-off-by: Hangbin Liu Link: https://patch.msgid.link/20251117024457.3034-3-liuhangbin@gmail.com Signed-off-by: Jakub Kicinski --- Documentation/netlink/genetlink-c.yaml | 2 +- Documentation/netlink/genetlink.yaml | 2 +- Documentation/netlink/netlink-raw.yaml | 2 +- Documentation/netlink/specs/rt-addr.yaml | 6 +++--- Documentation/netlink/specs/rt-link.yaml | 16 ++++++++-------- Documentation/netlink/specs/rt-neigh.yaml | 2 +- Documentation/netlink/specs/rt-route.yaml | 8 ++++---- Documentation/netlink/specs/rt-rule.yaml | 6 ++++-- 8 files changed, 23 insertions(+), 21 deletions(-) diff --git a/Documentation/netlink/genetlink-c.yaml b/Documentation/netlink/genetlink-c.yaml index 5a234e9b5fa2..57f59fe23e3f 100644 --- a/Documentation/netlink/genetlink-c.yaml +++ b/Documentation/netlink/genetlink-c.yaml @@ -227,7 +227,7 @@ properties: Optional format indicator that is intended only for choosing the right formatting mechanism when displaying values of this type. - enum: [ hex, mac, fddi, ipv4, ipv6, uuid ] + enum: [ hex, mac, fddi, ipv4, ipv6, ipv4-or-v6, uuid ] # Start genetlink-c name-prefix: type: string diff --git a/Documentation/netlink/genetlink.yaml b/Documentation/netlink/genetlink.yaml index 7b1ec153e834..b020a537d8ac 100644 --- a/Documentation/netlink/genetlink.yaml +++ b/Documentation/netlink/genetlink.yaml @@ -185,7 +185,7 @@ properties: Optional format indicator that is intended only for choosing the right formatting mechanism when displaying values of this type. - enum: [ hex, mac, fddi, ipv4, ipv6, uuid ] + enum: [ hex, mac, fddi, ipv4, ipv6, ipv4-or-v6, uuid ] # Make sure name-prefix does not appear in subsets (subsets inherit naming) dependencies: diff --git a/Documentation/netlink/netlink-raw.yaml b/Documentation/netlink/netlink-raw.yaml index 246fa07bccf6..0166a7e4afbb 100644 --- a/Documentation/netlink/netlink-raw.yaml +++ b/Documentation/netlink/netlink-raw.yaml @@ -157,7 +157,7 @@ properties: Optional format indicator that is intended only for choosing the right formatting mechanism when displaying values of this type. - enum: [ hex, mac, fddi, ipv4, ipv6, uuid ] + enum: [ hex, mac, fddi, ipv4, ipv6, ipv4-or-v6, uuid ] struct: description: Name of the nested struct type. type: string diff --git a/Documentation/netlink/specs/rt-addr.yaml b/Documentation/netlink/specs/rt-addr.yaml index 3a582eac1629..abcbaa73fa9d 100644 --- a/Documentation/netlink/specs/rt-addr.yaml +++ b/Documentation/netlink/specs/rt-addr.yaml @@ -86,17 +86,17 @@ attribute-sets: - name: address type: binary - display-hint: ipv4 + display-hint: ipv4-or-v6 - name: local type: binary - display-hint: ipv4 + display-hint: ipv4-or-v6 - name: label type: string - name: broadcast - type: binary + type: u32 display-hint: ipv4 - name: anycast diff --git a/Documentation/netlink/specs/rt-link.yaml b/Documentation/netlink/specs/rt-link.yaml index e07341582771..ca22c68ca691 100644 --- a/Documentation/netlink/specs/rt-link.yaml +++ b/Documentation/netlink/specs/rt-link.yaml @@ -1707,11 +1707,11 @@ attribute-sets: - name: local type: binary - display-hint: ipv4 + display-hint: ipv4-or-v6 - name: remote type: binary - display-hint: ipv4 + display-hint: ipv4-or-v6 - name: ttl type: u8 @@ -1833,11 +1833,11 @@ attribute-sets: - name: local type: binary - display-hint: ipv4 + display-hint: ipv4-or-v6 - name: remote type: binary - display-hint: ipv4 + display-hint: ipv4-or-v6 - name: fwmark type: u32 @@ -1868,7 +1868,7 @@ attribute-sets: type: u32 - name: remote - type: binary + type: u32 display-hint: ipv4 - name: ttl @@ -1952,11 +1952,11 @@ attribute-sets: - name: local type: binary - display-hint: ipv4 + display-hint: ipv4-or-v6 - name: remote type: binary - display-hint: ipv4 + display-hint: ipv4-or-v6 - name: ttl type: u8 @@ -1986,7 +1986,7 @@ attribute-sets: display-hint: ipv6 - name: 6rd-relay-prefix - type: binary + type: u32 display-hint: ipv4 - name: 6rd-prefixlen diff --git a/Documentation/netlink/specs/rt-neigh.yaml b/Documentation/netlink/specs/rt-neigh.yaml index 2f568a6231c9..0f46ef313590 100644 --- a/Documentation/netlink/specs/rt-neigh.yaml +++ b/Documentation/netlink/specs/rt-neigh.yaml @@ -194,7 +194,7 @@ attribute-sets: - name: dst type: binary - display-hint: ipv4 + display-hint: ipv4-or-v6 - name: lladdr type: binary diff --git a/Documentation/netlink/specs/rt-route.yaml b/Documentation/netlink/specs/rt-route.yaml index 1ecb3fadc067..33195db96746 100644 --- a/Documentation/netlink/specs/rt-route.yaml +++ b/Documentation/netlink/specs/rt-route.yaml @@ -87,11 +87,11 @@ attribute-sets: - name: dst type: binary - display-hint: ipv4 + display-hint: ipv4-or-v6 - name: src type: binary - display-hint: ipv4 + display-hint: ipv4-or-v6 - name: iif type: u32 @@ -101,14 +101,14 @@ attribute-sets: - name: gateway type: binary - display-hint: ipv4 + display-hint: ipv4-or-v6 - name: priority type: u32 - name: prefsrc type: binary - display-hint: ipv4 + display-hint: ipv4-or-v6 - name: metrics type: nest diff --git a/Documentation/netlink/specs/rt-rule.yaml b/Documentation/netlink/specs/rt-rule.yaml index bebee452a950..7f03a44ab036 100644 --- a/Documentation/netlink/specs/rt-rule.yaml +++ b/Documentation/netlink/specs/rt-rule.yaml @@ -96,10 +96,12 @@ attribute-sets: attributes: - name: dst - type: u32 + type: binary + display-hint: ipv4-or-v6 - name: src - type: u32 + type: binary + display-hint: ipv4-or-v6 - name: iifname type: string From 2e4c44b12f4da60d3e8dcbc1ccf38bb28a878050 Mon Sep 17 00:00:00 2001 From: Gal Pressman Date: Mon, 17 Nov 2025 23:42:05 +0200 Subject: [PATCH 0884/1381] net/mlx5: Refactor EEPROM query error handling to return status separately Matthew and Jakub reported [1] issues where inventory automation tools are calling EEPROM query repeatedly on a port that doesn't have an SFP connected, resulting in millions of error prints. Move MCIA register status extraction from the query functions to the callers, allowing use of extack reporting instead of a dmesg print when using the netlink API. [1] https://lore.kernel.org/netdev/20251028194011.39877-1-mattc@purestorage.com/ Cc: Matthew W Carlis Signed-off-by: Gal Pressman Reviewed-by: Jianbo Liu Signed-off-by: Tariq Toukan Link: https://patch.msgid.link/1763415729-1238421-2-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- .../ethernet/mellanox/mlx5/core/en_ethtool.c | 19 +++++----- .../ethernet/mellanox/mlx5/core/mlx5_core.h | 4 +-- .../net/ethernet/mellanox/mlx5/core/port.c | 35 +++++++++---------- 3 files changed, 30 insertions(+), 28 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c index 939e274779b3..727d7a833110 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -2027,7 +2027,7 @@ static int mlx5e_get_module_info(struct net_device *netdev, int size_read = 0; u8 data[4] = {0}; - size_read = mlx5_query_module_eeprom(dev, 0, 2, data); + size_read = mlx5_query_module_eeprom(dev, 0, 2, data, NULL); if (size_read < 2) return -EIO; @@ -2069,6 +2069,7 @@ static int mlx5e_get_module_eeprom(struct net_device *netdev, struct mlx5_core_dev *mdev = priv->mdev; int offset = ee->offset; int size_read; + u8 status = 0; int i = 0; if (!ee->len) @@ -2078,15 +2079,15 @@ static int mlx5e_get_module_eeprom(struct net_device *netdev, while (i < ee->len) { size_read = mlx5_query_module_eeprom(mdev, offset, ee->len - i, - data + i); - + data + i, &status); if (!size_read) /* Done reading */ return 0; if (size_read < 0) { - netdev_err(priv->netdev, "%s: mlx5_query_eeprom failed:0x%x\n", - __func__, size_read); + netdev_err(netdev, + "%s: mlx5_query_eeprom failed:0x%x, status %u\n", + __func__, size_read, status); return size_read; } @@ -2106,6 +2107,7 @@ static int mlx5e_get_module_eeprom_by_page(struct net_device *netdev, struct mlx5_core_dev *mdev = priv->mdev; u8 *data = page_data->data; int size_read; + u8 status = 0; int i = 0; if (!page_data->length) @@ -2119,7 +2121,8 @@ static int mlx5e_get_module_eeprom_by_page(struct net_device *netdev, query.page = page_data->page; while (i < page_data->length) { query.size = page_data->length - i; - size_read = mlx5_query_module_eeprom_by_page(mdev, &query, data + i); + size_read = mlx5_query_module_eeprom_by_page(mdev, &query, + data + i, &status); /* Done reading, return how many bytes was read */ if (!size_read) @@ -2128,8 +2131,8 @@ static int mlx5e_get_module_eeprom_by_page(struct net_device *netdev, if (size_read < 0) { NL_SET_ERR_MSG_FMT_MOD( extack, - "Query module eeprom by page failed, read %u bytes, err %d", - i, size_read); + "Query module eeprom by page failed, read %u bytes, err %d, status %u", + i, size_read, status); return size_read; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h index acef7d0ffa09..cfebc110c02f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h @@ -357,11 +357,11 @@ int mlx5_set_port_fcs(struct mlx5_core_dev *mdev, u8 enable); void mlx5_query_port_fcs(struct mlx5_core_dev *mdev, bool *supported, bool *enabled); int mlx5_query_module_eeprom(struct mlx5_core_dev *dev, - u16 offset, u16 size, u8 *data); + u16 offset, u16 size, u8 *data, u8 *status); int mlx5_query_module_eeprom_by_page(struct mlx5_core_dev *dev, struct mlx5_module_eeprom_query_params *params, - u8 *data); + u8 *data, u8 *status); int mlx5_query_port_dcbx_param(struct mlx5_core_dev *mdev, u32 *out); int mlx5_set_port_dcbx_param(struct mlx5_core_dev *mdev, u32 *in); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/port.c b/drivers/net/ethernet/mellanox/mlx5/core/port.c index aa9f2b0a77d3..e4b1dfafb41f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/port.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/port.c @@ -289,11 +289,11 @@ int mlx5_query_module_num(struct mlx5_core_dev *dev, int *module_num) } static int mlx5_query_module_id(struct mlx5_core_dev *dev, int module_num, - u8 *module_id) + u8 *module_id, u8 *status) { u32 in[MLX5_ST_SZ_DW(mcia_reg)] = {}; u32 out[MLX5_ST_SZ_DW(mcia_reg)]; - int err, status; + int err; u8 *ptr; MLX5_SET(mcia_reg, in, i2c_device_address, MLX5_I2C_ADDR_LOW); @@ -308,12 +308,12 @@ static int mlx5_query_module_id(struct mlx5_core_dev *dev, int module_num, if (err) return err; - status = MLX5_GET(mcia_reg, out, status); - if (status) { - mlx5_core_err(dev, "query_mcia_reg failed: status: 0x%x\n", - status); + if (MLX5_GET(mcia_reg, out, status)) { + if (status) + *status = MLX5_GET(mcia_reg, out, status); return -EIO; } + ptr = MLX5_ADDR_OF(mcia_reg, out, dword_0); *module_id = ptr[0]; @@ -370,13 +370,14 @@ static int mlx5_mcia_max_bytes(struct mlx5_core_dev *dev) } static int mlx5_query_mcia(struct mlx5_core_dev *dev, - struct mlx5_module_eeprom_query_params *params, u8 *data) + struct mlx5_module_eeprom_query_params *params, + u8 *data, u8 *status) { u32 in[MLX5_ST_SZ_DW(mcia_reg)] = {}; u32 out[MLX5_ST_SZ_DW(mcia_reg)]; - int status, err; void *ptr; u16 size; + int err; size = min_t(int, params->size, mlx5_mcia_max_bytes(dev)); @@ -392,12 +393,9 @@ static int mlx5_query_mcia(struct mlx5_core_dev *dev, if (err) return err; - status = MLX5_GET(mcia_reg, out, status); - if (status) { - mlx5_core_err(dev, "query_mcia_reg failed: status: 0x%x\n", - status); + *status = MLX5_GET(mcia_reg, out, status); + if (*status) return -EIO; - } ptr = MLX5_ADDR_OF(mcia_reg, out, dword_0); memcpy(data, ptr, size); @@ -406,7 +404,7 @@ static int mlx5_query_mcia(struct mlx5_core_dev *dev, } int mlx5_query_module_eeprom(struct mlx5_core_dev *dev, - u16 offset, u16 size, u8 *data) + u16 offset, u16 size, u8 *data, u8 *status) { struct mlx5_module_eeprom_query_params query = {0}; u8 module_id; @@ -416,7 +414,8 @@ int mlx5_query_module_eeprom(struct mlx5_core_dev *dev, if (err) return err; - err = mlx5_query_module_id(dev, query.module_number, &module_id); + err = mlx5_query_module_id(dev, query.module_number, &module_id, + status); if (err) return err; @@ -441,12 +440,12 @@ int mlx5_query_module_eeprom(struct mlx5_core_dev *dev, query.size = size; query.offset = offset; - return mlx5_query_mcia(dev, &query, data); + return mlx5_query_mcia(dev, &query, data, status); } int mlx5_query_module_eeprom_by_page(struct mlx5_core_dev *dev, struct mlx5_module_eeprom_query_params *params, - u8 *data) + u8 *data, u8 *status) { int err; @@ -460,7 +459,7 @@ int mlx5_query_module_eeprom_by_page(struct mlx5_core_dev *dev, return -EINVAL; } - return mlx5_query_mcia(dev, params, data); + return mlx5_query_mcia(dev, params, data, status); } static int mlx5_query_port_pvlc(struct mlx5_core_dev *dev, u32 *pvlc, From 391dad2e686f214932f769847cc8603a7df389eb Mon Sep 17 00:00:00 2001 From: Carolina Jubran Date: Mon, 17 Nov 2025 23:42:06 +0200 Subject: [PATCH 0885/1381] net/mlx5e: Recover SQ on excessive PTP TX timestamp delta Extend the TX timestamp handler to recover the SQ when the difference between the port and CQE TX timestamps is abnormally large. The current logic aborts timestamp delivery if the delta exceeds 1/128 seconds, which matches the maximum expected packet interval in ptp4l. A larger delta makes the timestamps unreliable. This change adds recovery if the delta exceeds 0.5 seconds. Such a large gap should not occur in normal operation and indicates that firmware is stuck or metadata tracking is out of sync, leading to stale or mismatched timestamps. Recovering the SQ ensures forward progress and avoids silently dropping invalid timestamps. The timestamp handler now takes mlx5e_ptpsq directly to access both CQ stats and the recovery state. Signed-off-by: Carolina Jubran Reviewed-by: Shahar Shitrit Signed-off-by: Tariq Toukan Link: https://patch.msgid.link/1763415729-1238421-3-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- .../net/ethernet/mellanox/mlx5/core/en/ptp.c | 21 +++++++++++++------ .../net/ethernet/mellanox/mlx5/core/en/ptp.h | 2 +- .../net/ethernet/mellanox/mlx5/core/en_tx.c | 2 +- 3 files changed, 17 insertions(+), 8 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c b/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c index 12e10feb30f0..424f8a2728a3 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c @@ -82,7 +82,7 @@ static struct mlx5e_skb_cb_hwtstamp *mlx5e_skb_cb_get_hwts(struct sk_buff *skb) } static void mlx5e_skb_cb_hwtstamp_tx(struct sk_buff *skb, - struct mlx5e_ptp_cq_stats *cq_stats) + struct mlx5e_ptpsq *ptpsq) { struct skb_shared_hwtstamps hwts = {}; ktime_t diff; @@ -92,8 +92,17 @@ static void mlx5e_skb_cb_hwtstamp_tx(struct sk_buff *skb, /* Maximal allowed diff is 1 / 128 second */ if (diff > (NSEC_PER_SEC >> 7)) { - cq_stats->abort++; - cq_stats->abort_abs_diff_ns += diff; + struct mlx5e_txqsq *sq = &ptpsq->txqsq; + + ptpsq->cq_stats->abort++; + ptpsq->cq_stats->abort_abs_diff_ns += diff; + if (diff > (NSEC_PER_SEC >> 1) && + !test_and_set_bit(MLX5E_SQ_STATE_RECOVERING, &sq->state)) { + netdev_warn(sq->channel->netdev, + "PTP TX timestamp difference between CQE and port exceeds threshold: %lld ns, recovering SQ %u\n", + (s64)diff, sq->sqn); + queue_work(sq->priv->wq, &ptpsq->report_unhealthy_work); + } return; } @@ -103,7 +112,7 @@ static void mlx5e_skb_cb_hwtstamp_tx(struct sk_buff *skb, void mlx5e_skb_cb_hwtstamp_handler(struct sk_buff *skb, int hwtstamp_type, ktime_t hwtstamp, - struct mlx5e_ptp_cq_stats *cq_stats) + struct mlx5e_ptpsq *ptpsq) { switch (hwtstamp_type) { case (MLX5E_SKB_CB_CQE_HWTSTAMP): @@ -121,7 +130,7 @@ void mlx5e_skb_cb_hwtstamp_handler(struct sk_buff *skb, int hwtstamp_type, !mlx5e_skb_cb_get_hwts(skb)->port_hwtstamp) return; - mlx5e_skb_cb_hwtstamp_tx(skb, cq_stats); + mlx5e_skb_cb_hwtstamp_tx(skb, ptpsq); memset(skb->cb, 0, sizeof(struct mlx5e_skb_cb_hwtstamp)); } @@ -209,7 +218,7 @@ static void mlx5e_ptp_handle_ts_cqe(struct mlx5e_ptpsq *ptpsq, hwtstamp = mlx5e_cqe_ts_to_ns(sq->ptp_cyc2time, sq->clock, get_cqe_ts(cqe)); mlx5e_skb_cb_hwtstamp_handler(skb, MLX5E_SKB_CB_PORT_HWTSTAMP, - hwtstamp, ptpsq->cq_stats); + hwtstamp, ptpsq); ptpsq->cq_stats->cqe++; mlx5e_ptpsq_mark_ts_cqes_undelivered(ptpsq, hwtstamp); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.h b/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.h index 1c0e0a86a9ac..2a457a2ed707 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.h @@ -147,7 +147,7 @@ enum { void mlx5e_skb_cb_hwtstamp_handler(struct sk_buff *skb, int hwtstamp_type, ktime_t hwtstamp, - struct mlx5e_ptp_cq_stats *cq_stats); + struct mlx5e_ptpsq *ptpsq); void mlx5e_skb_cb_hwtstamp_init(struct sk_buff *skb); #endif /* __MLX5_EN_PTP_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c index 2702b3885f06..14884b9ea7f3 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c @@ -755,7 +755,7 @@ static void mlx5e_consume_skb(struct mlx5e_txqsq *sq, struct sk_buff *skb, hwts.hwtstamp = mlx5e_cqe_ts_to_ns(sq->ptp_cyc2time, sq->clock, ts); if (sq->ptpsq) { mlx5e_skb_cb_hwtstamp_handler(skb, MLX5E_SKB_CB_CQE_HWTSTAMP, - hwts.hwtstamp, sq->ptpsq->cq_stats); + hwts.hwtstamp, sq->ptpsq); } else { skb_tstamp_tx(skb, &hwts); sq->stats->timestamps++; From ea3270351c792632db5722ea3ca83b468cebb531 Mon Sep 17 00:00:00 2001 From: Carolina Jubran Date: Mon, 17 Nov 2025 23:42:07 +0200 Subject: [PATCH 0886/1381] net/mlx5: Remove redundant bw_share minimal value assignment Remove unnecessary logic that sets bw_share to minimal value, when parent has bw_share configured but nodes don't have min_rate. This check is redundant because the parent bandwidth acts as the upper bound regardless, and the firmware always enforces the topmost bandwidth constraint. Signed-off-by: Carolina Jubran Reviewed-by: Cosmin Ratiu Signed-off-by: Tariq Toukan Link: https://patch.msgid.link/1763415729-1238421-4-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c | 7 ------- 1 file changed, 7 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c index 56e6f54b1e2e..4278bcb04c72 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c @@ -341,13 +341,6 @@ static u32 esw_qos_calculate_min_rate_divider(struct mlx5_eswitch *esw, if (max_guarantee) return max_t(u32, max_guarantee / fw_max_bw_share, 1); - /* If nodes max min_rate divider is 0 but their parent has bw_share - * configured, then set bw_share for nodes to minimal value. - */ - - if (parent && parent->bw_share) - return 1; - /* If the node nodes has min_rate configured, a divider of 0 sets all * nodes' bw_share to 0, effectively disabling min guarantees. */ From fbb9933666e31f84c62e9620e9ec4d220ee31ab4 Mon Sep 17 00:00:00 2001 From: Saeed Mahameed Date: Mon, 17 Nov 2025 23:42:08 +0200 Subject: [PATCH 0887/1381] net/mlx5: Abort new commands if all command slots are stalled In case of a FW issue, FW might be not responding to FW commands, causing kernel lockout for a long period of time, e.g. rtnl_lock held while ethtool is trying to collect stats waiting for FW to respond to multiple commands, when all of them will timeout. While there's no immediate indication of the FW lockout, we can safely assume that something is wrong when all command slots are busy and in a timeout state and no FW completion was received on any of them. In such case, start immediately failing new commands. Signed-off-by: Saeed Mahameed Reviewed-by: Moshe Shemesh Signed-off-by: Tariq Toukan Link: https://patch.msgid.link/1763415729-1238421-5-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mellanox/mlx5/core/cmd.c | 55 +++++++++++++++++++ include/linux/mlx5/driver.h | 1 + 2 files changed, 56 insertions(+) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c index 722282cebce9..5b08e5ffe0e2 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c @@ -181,6 +181,7 @@ static int cmd_alloc_index(struct mlx5_cmd *cmd, struct mlx5_cmd_work_ent *ent) static void cmd_free_index(struct mlx5_cmd *cmd, int idx) { lockdep_assert_held(&cmd->alloc_lock); + cmd->ent_arr[idx] = NULL; set_bit(idx, &cmd->vars.bitmask); } @@ -1200,6 +1201,44 @@ static int wait_func(struct mlx5_core_dev *dev, struct mlx5_cmd_work_ent *ent) return err; } +/* Check if all command slots are stalled (timed out and not recovered). + * returns true if all slots timed out on a recent command and have not been + * completed by FW yet. (stalled state) + * false otherwise (at least one slot is not stalled). + * + * In such odd situation "all_stalled", this serves as a protection mechanism + * to avoid blocking the kernel for long periods of time in case FW is not + * responding to commands. + */ +static bool mlx5_cmd_all_stalled(struct mlx5_core_dev *dev) +{ + struct mlx5_cmd *cmd = &dev->cmd; + bool all_stalled = true; + unsigned long flags; + int i; + + spin_lock_irqsave(&cmd->alloc_lock, flags); + + /* at least one command slot is free */ + if (bitmap_weight(&cmd->vars.bitmask, cmd->vars.max_reg_cmds) > 0) { + all_stalled = false; + goto out; + } + + for_each_clear_bit(i, &cmd->vars.bitmask, cmd->vars.max_reg_cmds) { + struct mlx5_cmd_work_ent *ent = dev->cmd.ent_arr[i]; + + if (!test_bit(MLX5_CMD_ENT_STATE_TIMEDOUT, &ent->state)) { + all_stalled = false; + break; + } + } +out: + spin_unlock_irqrestore(&cmd->alloc_lock, flags); + + return all_stalled; +} + /* Notes: * 1. Callback functions may not sleep * 2. page queue commands do not support asynchrous completion @@ -1230,6 +1269,15 @@ static int mlx5_cmd_invoke(struct mlx5_core_dev *dev, struct mlx5_cmd_msg *in, if (callback && page_queue) return -EINVAL; + if (!page_queue && mlx5_cmd_all_stalled(dev)) { + mlx5_core_err_rl(dev, + "All CMD slots are stalled, aborting command\n"); + /* there's no reason to wait and block the whole kernel if FW + * isn't currently responding to all slots, fail immediately + */ + return -EAGAIN; + } + ent = cmd_alloc_ent(cmd, in, out, uout, uout_size, callback, context, page_queue); if (IS_ERR(ent)) @@ -1700,6 +1748,13 @@ static void mlx5_cmd_comp_handler(struct mlx5_core_dev *dev, u64 vec, bool force if (test_bit(i, &vector)) { ent = cmd->ent_arr[i]; + if (forced && ent->ret == -ETIMEDOUT) + set_bit(MLX5_CMD_ENT_STATE_TIMEDOUT, + &ent->state); + else if (!forced) /* real FW completion */ + clear_bit(MLX5_CMD_ENT_STATE_TIMEDOUT, + &ent->state); + /* if we already completed the command, ignore it */ if (!test_and_clear_bit(MLX5_CMD_ENT_STATE_PENDING_COMP, &ent->state)) { diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index 046396269ccf..7aec53371cf0 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -819,6 +819,7 @@ typedef void (*mlx5_cmd_cbk_t)(int status, void *context); enum { MLX5_CMD_ENT_STATE_PENDING_COMP, + MLX5_CMD_ENT_STATE_TIMEDOUT, }; struct mlx5_cmd_work_ent { From 70ca239b612cd154c9828fe4d0093fb9bd02a6c7 Mon Sep 17 00:00:00 2001 From: Tariq Toukan Date: Mon, 17 Nov 2025 23:42:09 +0200 Subject: [PATCH 0888/1381] net/mlx5: Use EOPNOTSUPP instead of ENOTSUPP Per Documentation/dev-tools/checkpatch.rst, ENOTSUPP is not a standard error code and should be avoided. EOPNOTSUPP should be used instead. Signed-off-by: Tariq Toukan Reviewed-by: Gal Pressman Link: https://patch.msgid.link/1763415729-1238421-6-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c | 2 +- drivers/net/ethernet/mellanox/mlx5/core/en_fs_ethtool.c | 2 +- drivers/net/ethernet/mellanox/mlx5/core/fpga/core.c | 2 +- drivers/net/ethernet/mellanox/mlx5/core/lib/vxlan.c | 2 +- .../ethernet/mellanox/mlx5/core/steering/sws/dr_domain.c | 8 ++++---- 5 files changed, 8 insertions(+), 8 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c index 080e7eab52c7..7bcf822a89f9 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c @@ -54,7 +54,7 @@ static int mlx5_query_mtrc_caps(struct mlx5_fw_tracer *tracer) if (!MLX5_GET(mtrc_cap, out, trace_to_memory)) { mlx5_core_dbg(dev, "FWTracer: Device does not support logging traces to memory\n"); - return -ENOTSUPP; + return -EOPNOTSUPP; } tracer->trc_ver = MLX5_GET(mtrc_cap, out, trc_ver); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_fs_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_fs_ethtool.c index 79916f1abd14..63bdef5b4ba5 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_fs_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_fs_ethtool.c @@ -704,7 +704,7 @@ static int validate_flow(struct mlx5e_priv *priv, num_tuples += ret; break; default: - return -ENOTSUPP; + return -EOPNOTSUPP; } if ((fs->flow_type & FLOW_EXT)) { ret = validate_vlan(fs); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fpga/core.c b/drivers/net/ethernet/mellanox/mlx5/core/fpga/core.c index e5c1012921d2..1ec61164e6b5 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fpga/core.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fpga/core.c @@ -211,7 +211,7 @@ int mlx5_fpga_device_start(struct mlx5_core_dev *mdev) max_num_qps = MLX5_CAP_FPGA(mdev, shell_caps.max_num_qps); if (!max_num_qps) { mlx5_fpga_err(fdev, "FPGA reports 0 QPs in SHELL_CAPS\n"); - err = -ENOTSUPP; + err = -EOPNOTSUPP; goto out; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/vxlan.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/vxlan.c index d55e15c1f380..304912637c35 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lib/vxlan.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/vxlan.c @@ -149,7 +149,7 @@ struct mlx5_vxlan *mlx5_vxlan_create(struct mlx5_core_dev *mdev) struct mlx5_vxlan *vxlan; if (!MLX5_CAP_ETH(mdev, tunnel_stateless_vxlan) || !mlx5_core_is_pf(mdev)) - return ERR_PTR(-ENOTSUPP); + return ERR_PTR(-EOPNOTSUPP); vxlan = kzalloc(sizeof(*vxlan), GFP_KERNEL); if (!vxlan) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/sws/dr_domain.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/sws/dr_domain.c index 65740bb68b09..e8c67ed9f748 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/sws/dr_domain.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/sws/dr_domain.c @@ -410,7 +410,7 @@ static int dr_domain_caps_init(struct mlx5_core_dev *mdev, switch (dmn->type) { case MLX5DR_DOMAIN_TYPE_NIC_RX: if (!DR_DOMAIN_SW_STEERING_SUPPORTED(dmn, rx)) - return -ENOTSUPP; + return -EOPNOTSUPP; dmn->info.supp_sw_steering = true; dmn->info.rx.type = DR_DOMAIN_NIC_TYPE_RX; @@ -419,7 +419,7 @@ static int dr_domain_caps_init(struct mlx5_core_dev *mdev, break; case MLX5DR_DOMAIN_TYPE_NIC_TX: if (!DR_DOMAIN_SW_STEERING_SUPPORTED(dmn, tx)) - return -ENOTSUPP; + return -EOPNOTSUPP; dmn->info.supp_sw_steering = true; dmn->info.tx.type = DR_DOMAIN_NIC_TYPE_TX; @@ -428,10 +428,10 @@ static int dr_domain_caps_init(struct mlx5_core_dev *mdev, break; case MLX5DR_DOMAIN_TYPE_FDB: if (!dmn->info.caps.eswitch_manager) - return -ENOTSUPP; + return -EOPNOTSUPP; if (!DR_DOMAIN_SW_STEERING_SUPPORTED(dmn, fdb)) - return -ENOTSUPP; + return -EOPNOTSUPP; dmn->info.rx.type = DR_DOMAIN_NIC_TYPE_RX; dmn->info.tx.type = DR_DOMAIN_NIC_TYPE_TX; From 58b47c713711b8afbf68e3158d4d5acdead00e9b Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Sat, 15 Nov 2025 02:08:32 +0000 Subject: [PATCH 0889/1381] af_unix: Count cyclic SCC. __unix_walk_scc() and unix_walk_scc_fast() call unix_scc_cyclic() for each SCC to check if it forms a cyclic reference, so that we can skip GC at the following invocations in case all SCCs do not have any cycles. If we count the number of cyclic SCCs in __unix_walk_scc(), we can simplify unix_walk_scc_fast() because the number of cyclic SCCs only changes when it garbage-collects a SCC. So, let's count cyclic SCC in __unix_walk_scc() and decrement it in unix_walk_scc_fast() when performing garbage collection. Note that we will use this counter in a later patch to check if a cycle existed in the previous GC run. Signed-off-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20251115020935.2643121-2-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/unix/garbage.c | 31 +++++++++++++++++++++---------- 1 file changed, 21 insertions(+), 10 deletions(-) diff --git a/net/unix/garbage.c b/net/unix/garbage.c index 65396a4e1b07..9f62d5097973 100644 --- a/net/unix/garbage.c +++ b/net/unix/garbage.c @@ -404,9 +404,11 @@ static bool unix_scc_cyclic(struct list_head *scc) static LIST_HEAD(unix_visited_vertices); static unsigned long unix_vertex_grouped_index = UNIX_VERTEX_INDEX_MARK2; -static void __unix_walk_scc(struct unix_vertex *vertex, unsigned long *last_index, - struct sk_buff_head *hitlist) +static unsigned long __unix_walk_scc(struct unix_vertex *vertex, + unsigned long *last_index, + struct sk_buff_head *hitlist) { + unsigned long cyclic_sccs = 0; LIST_HEAD(vertex_stack); struct unix_edge *edge; LIST_HEAD(edge_stack); @@ -497,8 +499,8 @@ static void __unix_walk_scc(struct unix_vertex *vertex, unsigned long *last_inde if (unix_vertex_max_scc_index < vertex->scc_index) unix_vertex_max_scc_index = vertex->scc_index; - if (!unix_graph_maybe_cyclic) - unix_graph_maybe_cyclic = unix_scc_cyclic(&scc); + if (unix_scc_cyclic(&scc)) + cyclic_sccs++; } list_del(&scc); @@ -507,13 +509,17 @@ static void __unix_walk_scc(struct unix_vertex *vertex, unsigned long *last_inde /* Need backtracking ? */ if (!list_empty(&edge_stack)) goto prev_vertex; + + return cyclic_sccs; } +static unsigned long unix_graph_cyclic_sccs; + static void unix_walk_scc(struct sk_buff_head *hitlist) { unsigned long last_index = UNIX_VERTEX_INDEX_START; + unsigned long cyclic_sccs = 0; - unix_graph_maybe_cyclic = false; unix_vertex_max_scc_index = UNIX_VERTEX_INDEX_START; /* Visit every vertex exactly once. @@ -523,18 +529,20 @@ static void unix_walk_scc(struct sk_buff_head *hitlist) struct unix_vertex *vertex; vertex = list_first_entry(&unix_unvisited_vertices, typeof(*vertex), entry); - __unix_walk_scc(vertex, &last_index, hitlist); + cyclic_sccs += __unix_walk_scc(vertex, &last_index, hitlist); } list_replace_init(&unix_visited_vertices, &unix_unvisited_vertices); swap(unix_vertex_unvisited_index, unix_vertex_grouped_index); + unix_graph_cyclic_sccs = cyclic_sccs; + unix_graph_maybe_cyclic = !!unix_graph_cyclic_sccs; unix_graph_grouped = true; } static void unix_walk_scc_fast(struct sk_buff_head *hitlist) { - unix_graph_maybe_cyclic = false; + unsigned long cyclic_sccs = unix_graph_cyclic_sccs; while (!list_empty(&unix_unvisited_vertices)) { struct unix_vertex *vertex; @@ -551,15 +559,18 @@ static void unix_walk_scc_fast(struct sk_buff_head *hitlist) scc_dead = unix_vertex_dead(vertex); } - if (scc_dead) + if (scc_dead) { + cyclic_sccs--; unix_collect_skb(&scc, hitlist); - else if (!unix_graph_maybe_cyclic) - unix_graph_maybe_cyclic = unix_scc_cyclic(&scc); + } list_del(&scc); } list_replace_init(&unix_visited_vertices, &unix_unvisited_vertices); + + unix_graph_cyclic_sccs = cyclic_sccs; + unix_graph_maybe_cyclic = !!unix_graph_cyclic_sccs; } static bool gc_in_progress; From 6b6f3c71fe568aa8ed3e16e9135d88a5f4fd3e84 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Sat, 15 Nov 2025 02:08:33 +0000 Subject: [PATCH 0890/1381] af_unix: Simplify GC state. GC manages its state by two variables, unix_graph_maybe_cyclic and unix_graph_grouped, both of which are set to false in the initial state. When an AF_UNIX socket is passed to an in-flight AF_UNIX socket, unix_update_graph() sets unix_graph_maybe_cyclic to true and unix_graph_grouped to false, making the next GC invocation call unix_walk_scc() to group SCCs. Once unix_walk_scc() finishes, sockets in the same SCC are linked via vertex->scc_entry. Then, unix_graph_grouped is set to true so that the following GC invocations can skip Tarjan's algorithm and simply iterate through the list in unix_walk_scc_fast(). In addition, if we know there is at least one cyclic reference, we set unix_graph_maybe_cyclic to true so that we do not skip GC. So the state transitions as follows: (unix_graph_maybe_cyclic, unix_graph_grouped) = (false, false) -> (true, false) -> (true, true) or (false, true) ^.______________/________________/ There is no transition to the initial state where both variables are false. If we consider the initial state as grouped, we can see that the GC actually has a tristate. Let's consolidate two variables into one enum. Signed-off-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20251115020935.2643121-3-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/unix/garbage.c | 21 ++++++++++++--------- 1 file changed, 12 insertions(+), 9 deletions(-) diff --git a/net/unix/garbage.c b/net/unix/garbage.c index 9f62d5097973..7528e2db1293 100644 --- a/net/unix/garbage.c +++ b/net/unix/garbage.c @@ -121,8 +121,13 @@ static struct unix_vertex *unix_edge_successor(struct unix_edge *edge) return edge->successor->vertex; } -static bool unix_graph_maybe_cyclic; -static bool unix_graph_grouped; +enum { + UNIX_GRAPH_NOT_CYCLIC, + UNIX_GRAPH_MAYBE_CYCLIC, + UNIX_GRAPH_CYCLIC, +}; + +static unsigned char unix_graph_state; static void unix_update_graph(struct unix_vertex *vertex) { @@ -132,8 +137,7 @@ static void unix_update_graph(struct unix_vertex *vertex) if (!vertex) return; - unix_graph_maybe_cyclic = true; - unix_graph_grouped = false; + unix_graph_state = UNIX_GRAPH_MAYBE_CYCLIC; } static LIST_HEAD(unix_unvisited_vertices); @@ -536,8 +540,7 @@ static void unix_walk_scc(struct sk_buff_head *hitlist) swap(unix_vertex_unvisited_index, unix_vertex_grouped_index); unix_graph_cyclic_sccs = cyclic_sccs; - unix_graph_maybe_cyclic = !!unix_graph_cyclic_sccs; - unix_graph_grouped = true; + unix_graph_state = cyclic_sccs ? UNIX_GRAPH_CYCLIC : UNIX_GRAPH_NOT_CYCLIC; } static void unix_walk_scc_fast(struct sk_buff_head *hitlist) @@ -570,7 +573,7 @@ static void unix_walk_scc_fast(struct sk_buff_head *hitlist) list_replace_init(&unix_visited_vertices, &unix_unvisited_vertices); unix_graph_cyclic_sccs = cyclic_sccs; - unix_graph_maybe_cyclic = !!unix_graph_cyclic_sccs; + unix_graph_state = cyclic_sccs ? UNIX_GRAPH_CYCLIC : UNIX_GRAPH_NOT_CYCLIC; } static bool gc_in_progress; @@ -582,14 +585,14 @@ static void __unix_gc(struct work_struct *work) spin_lock(&unix_gc_lock); - if (!unix_graph_maybe_cyclic) { + if (unix_graph_state == UNIX_GRAPH_NOT_CYCLIC) { spin_unlock(&unix_gc_lock); goto skip_gc; } __skb_queue_head_init(&hitlist); - if (unix_graph_grouped) + if (unix_graph_state == UNIX_GRAPH_CYCLIC) unix_walk_scc_fast(&hitlist); else unix_walk_scc(&hitlist); From da8fc7a39be897426e1ac05aa90263abf40621b7 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Sat, 15 Nov 2025 02:08:34 +0000 Subject: [PATCH 0891/1381] af_unix: Don't trigger GC from close() if unnecessary. We have been triggering GC on every close() if there is even one inflight AF_UNIX socket. This is because the old GC implementation had no idea of the graph shape formed by SCM_RIGHTS references. The new GC knows whether there could be a cyclic reference or not, and we can do better. Let's not trigger GC from close() if there is no cyclic reference or GC is already in progress. While at it, unix_gc() is renamed to unix_schedule_gc() as it does not actually perform GC since commit 8b90a9f819dc ("af_unix: Run GC on only one CPU."). Signed-off-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20251115020935.2643121-4-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/unix/af_unix.c | 3 +-- net/unix/af_unix.h | 3 +-- net/unix/garbage.c | 27 +++++++++++++++++---------- 3 files changed, 19 insertions(+), 14 deletions(-) diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index 3b44cadaed96..4a80dac56bbd 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -733,8 +733,7 @@ static void unix_release_sock(struct sock *sk, int embrion) /* ---- Socket is dead now and most probably destroyed ---- */ - if (READ_ONCE(unix_tot_inflight)) - unix_gc(); /* Garbage collect fds */ + unix_schedule_gc(); } struct unix_peercred { diff --git a/net/unix/af_unix.h b/net/unix/af_unix.h index 59db179df9bb..0fb5b348ad94 100644 --- a/net/unix/af_unix.h +++ b/net/unix/af_unix.h @@ -24,13 +24,12 @@ struct unix_skb_parms { #define UNIXCB(skb) (*(struct unix_skb_parms *)&((skb)->cb)) /* GC for SCM_RIGHTS */ -extern unsigned int unix_tot_inflight; void unix_add_edges(struct scm_fp_list *fpl, struct unix_sock *receiver); void unix_del_edges(struct scm_fp_list *fpl); void unix_update_edges(struct unix_sock *receiver); int unix_prepare_fpl(struct scm_fp_list *fpl); void unix_destroy_fpl(struct scm_fp_list *fpl); -void unix_gc(void); +void unix_schedule_gc(void); void wait_for_unix_gc(struct scm_fp_list *fpl); /* SOCK_DIAG */ diff --git a/net/unix/garbage.c b/net/unix/garbage.c index 7528e2db1293..190dea73f0ab 100644 --- a/net/unix/garbage.c +++ b/net/unix/garbage.c @@ -137,7 +137,7 @@ static void unix_update_graph(struct unix_vertex *vertex) if (!vertex) return; - unix_graph_state = UNIX_GRAPH_MAYBE_CYCLIC; + WRITE_ONCE(unix_graph_state, UNIX_GRAPH_MAYBE_CYCLIC); } static LIST_HEAD(unix_unvisited_vertices); @@ -200,7 +200,7 @@ static void unix_free_vertices(struct scm_fp_list *fpl) } static DEFINE_SPINLOCK(unix_gc_lock); -unsigned int unix_tot_inflight; +static unsigned int unix_tot_inflight; void unix_add_edges(struct scm_fp_list *fpl, struct unix_sock *receiver) { @@ -540,7 +540,8 @@ static void unix_walk_scc(struct sk_buff_head *hitlist) swap(unix_vertex_unvisited_index, unix_vertex_grouped_index); unix_graph_cyclic_sccs = cyclic_sccs; - unix_graph_state = cyclic_sccs ? UNIX_GRAPH_CYCLIC : UNIX_GRAPH_NOT_CYCLIC; + WRITE_ONCE(unix_graph_state, + cyclic_sccs ? UNIX_GRAPH_CYCLIC : UNIX_GRAPH_NOT_CYCLIC); } static void unix_walk_scc_fast(struct sk_buff_head *hitlist) @@ -573,12 +574,13 @@ static void unix_walk_scc_fast(struct sk_buff_head *hitlist) list_replace_init(&unix_visited_vertices, &unix_unvisited_vertices); unix_graph_cyclic_sccs = cyclic_sccs; - unix_graph_state = cyclic_sccs ? UNIX_GRAPH_CYCLIC : UNIX_GRAPH_NOT_CYCLIC; + WRITE_ONCE(unix_graph_state, + cyclic_sccs ? UNIX_GRAPH_CYCLIC : UNIX_GRAPH_NOT_CYCLIC); } static bool gc_in_progress; -static void __unix_gc(struct work_struct *work) +static void unix_gc(struct work_struct *work) { struct sk_buff_head hitlist; struct sk_buff *skb; @@ -609,10 +611,16 @@ static void __unix_gc(struct work_struct *work) WRITE_ONCE(gc_in_progress, false); } -static DECLARE_WORK(unix_gc_work, __unix_gc); +static DECLARE_WORK(unix_gc_work, unix_gc); -void unix_gc(void) +void unix_schedule_gc(void) { + if (READ_ONCE(unix_graph_state) == UNIX_GRAPH_NOT_CYCLIC) + return; + + if (READ_ONCE(gc_in_progress)) + return; + WRITE_ONCE(gc_in_progress, true); queue_work(system_dfl_wq, &unix_gc_work); } @@ -628,9 +636,8 @@ void wait_for_unix_gc(struct scm_fp_list *fpl) * Paired with the WRITE_ONCE() in unix_inflight(), * unix_notinflight(), and __unix_gc(). */ - if (READ_ONCE(unix_tot_inflight) > UNIX_INFLIGHT_TRIGGER_GC && - !READ_ONCE(gc_in_progress)) - unix_gc(); + if (READ_ONCE(unix_tot_inflight) > UNIX_INFLIGHT_TRIGGER_GC) + unix_schedule_gc(); /* Penalise users who want to send AF_UNIX sockets * but whose sockets have not been received yet. From 384900542dc85f3aac7918fea8e7ef62141e3ea6 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Sat, 15 Nov 2025 02:08:35 +0000 Subject: [PATCH 0892/1381] af_unix: Don't call wait_for_unix_gc() on every sendmsg(). We have been calling wait_for_unix_gc() on every sendmsg() in case there are too many inflight AF_UNIX sockets. This is also because the old GC implementation had poor knowledge of the inflight sockets and had to suspect every sendmsg(). This was improved by commit d9f21b361333 ("af_unix: Try to run GC async."), but we do not even need to call wait_for_unix_gc() if the process is not sending AF_UNIX sockets. The wait_for_unix_gc() call only helps when a malicious process continues to create cyclic references, and we can detect that in a better place and slow it down. Let's move wait_for_unix_gc() to unix_prepare_fpl() that is called only when AF_UNIX socket fd is passed via SCM_RIGHTS. Signed-off-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20251115020935.2643121-5-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/unix/af_unix.c | 4 ---- net/unix/af_unix.h | 1 - net/unix/garbage.c | 9 ++++++--- 3 files changed, 6 insertions(+), 8 deletions(-) diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index 4a80dac56bbd..34952242bd81 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -2098,8 +2098,6 @@ static int unix_dgram_sendmsg(struct socket *sock, struct msghdr *msg, if (err < 0) return err; - wait_for_unix_gc(scm.fp); - if (msg->msg_flags & MSG_OOB) { err = -EOPNOTSUPP; goto out; @@ -2393,8 +2391,6 @@ static int unix_stream_sendmsg(struct socket *sock, struct msghdr *msg, if (err < 0) return err; - wait_for_unix_gc(scm.fp); - if (msg->msg_flags & MSG_OOB) { err = -EOPNOTSUPP; #if IS_ENABLED(CONFIG_AF_UNIX_OOB) diff --git a/net/unix/af_unix.h b/net/unix/af_unix.h index 0fb5b348ad94..2f1bfe3217c1 100644 --- a/net/unix/af_unix.h +++ b/net/unix/af_unix.h @@ -30,7 +30,6 @@ void unix_update_edges(struct unix_sock *receiver); int unix_prepare_fpl(struct scm_fp_list *fpl); void unix_destroy_fpl(struct scm_fp_list *fpl); void unix_schedule_gc(void); -void wait_for_unix_gc(struct scm_fp_list *fpl); /* SOCK_DIAG */ long unix_inq_len(struct sock *sk); diff --git a/net/unix/garbage.c b/net/unix/garbage.c index 190dea73f0ab..280b9b07b1c0 100644 --- a/net/unix/garbage.c +++ b/net/unix/garbage.c @@ -282,6 +282,8 @@ void unix_update_edges(struct unix_sock *receiver) } } +static void wait_for_unix_gc(struct scm_fp_list *fpl); + int unix_prepare_fpl(struct scm_fp_list *fpl) { struct unix_vertex *vertex; @@ -303,6 +305,8 @@ int unix_prepare_fpl(struct scm_fp_list *fpl) if (!fpl->edges) goto err; + wait_for_unix_gc(fpl); + return 0; err: @@ -628,7 +632,7 @@ void unix_schedule_gc(void) #define UNIX_INFLIGHT_TRIGGER_GC 16000 #define UNIX_INFLIGHT_SANE_USER (SCM_MAX_FD * 8) -void wait_for_unix_gc(struct scm_fp_list *fpl) +static void wait_for_unix_gc(struct scm_fp_list *fpl) { /* If number of inflight sockets is insane, * force a garbage collect right now. @@ -642,8 +646,7 @@ void wait_for_unix_gc(struct scm_fp_list *fpl) /* Penalise users who want to send AF_UNIX sockets * but whose sockets have not been received yet. */ - if (!fpl || !fpl->count_unix || - READ_ONCE(fpl->user->unix_inflight) < UNIX_INFLIGHT_SANE_USER) + if (READ_ONCE(fpl->user->unix_inflight) < UNIX_INFLIGHT_SANE_USER) return; if (READ_ONCE(gc_in_progress)) From e29c7a4cec867f9d860b8ff3da0fc44c7177876a Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Sat, 15 Nov 2025 02:08:36 +0000 Subject: [PATCH 0893/1381] af_unix: Refine wait_for_unix_gc(). unix_tot_inflight is a poor metric, only telling the number of inflight AF_UNXI sockets, and we should use unix_graph_state instead. Also, if the receiver is catching up with the passed fds, the sender does not need to schedule GC. GC only helps unreferenced cyclic SCM_RIGHTS references, and in such a situation, the malicious sendmsg() will continue to call wait_for_unix_gc() and hit the UNIX_INFLIGHT_SANE_USER condition. Let's make only malicious users schedule GC and wait for it to finish if a cyclic reference exists during the previous GC run. Then, sane users will pay almost no cost for wait_for_unix_gc(). Signed-off-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20251115020935.2643121-6-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/unix/garbage.c | 21 ++++++++------------- 1 file changed, 8 insertions(+), 13 deletions(-) diff --git a/net/unix/garbage.c b/net/unix/garbage.c index 280b9b07b1c0..a6929226d40d 100644 --- a/net/unix/garbage.c +++ b/net/unix/garbage.c @@ -543,7 +543,7 @@ static void unix_walk_scc(struct sk_buff_head *hitlist) list_replace_init(&unix_visited_vertices, &unix_unvisited_vertices); swap(unix_vertex_unvisited_index, unix_vertex_grouped_index); - unix_graph_cyclic_sccs = cyclic_sccs; + WRITE_ONCE(unix_graph_cyclic_sccs, cyclic_sccs); WRITE_ONCE(unix_graph_state, cyclic_sccs ? UNIX_GRAPH_CYCLIC : UNIX_GRAPH_NOT_CYCLIC); } @@ -577,7 +577,7 @@ static void unix_walk_scc_fast(struct sk_buff_head *hitlist) list_replace_init(&unix_visited_vertices, &unix_unvisited_vertices); - unix_graph_cyclic_sccs = cyclic_sccs; + WRITE_ONCE(unix_graph_cyclic_sccs, cyclic_sccs); WRITE_ONCE(unix_graph_state, cyclic_sccs ? UNIX_GRAPH_CYCLIC : UNIX_GRAPH_NOT_CYCLIC); } @@ -629,19 +629,12 @@ void unix_schedule_gc(void) queue_work(system_dfl_wq, &unix_gc_work); } -#define UNIX_INFLIGHT_TRIGGER_GC 16000 -#define UNIX_INFLIGHT_SANE_USER (SCM_MAX_FD * 8) +#define UNIX_INFLIGHT_SANE_USER (SCM_MAX_FD * 8) static void wait_for_unix_gc(struct scm_fp_list *fpl) { - /* If number of inflight sockets is insane, - * force a garbage collect right now. - * - * Paired with the WRITE_ONCE() in unix_inflight(), - * unix_notinflight(), and __unix_gc(). - */ - if (READ_ONCE(unix_tot_inflight) > UNIX_INFLIGHT_TRIGGER_GC) - unix_schedule_gc(); + if (READ_ONCE(unix_graph_state) == UNIX_GRAPH_NOT_CYCLIC) + return; /* Penalise users who want to send AF_UNIX sockets * but whose sockets have not been received yet. @@ -649,6 +642,8 @@ static void wait_for_unix_gc(struct scm_fp_list *fpl) if (READ_ONCE(fpl->user->unix_inflight) < UNIX_INFLIGHT_SANE_USER) return; - if (READ_ONCE(gc_in_progress)) + unix_schedule_gc(); + + if (READ_ONCE(unix_graph_cyclic_sccs)) flush_work(&unix_gc_work); } From ab8b23150abccd34fddc3effe7776ad32c44b6c9 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Sat, 15 Nov 2025 02:08:37 +0000 Subject: [PATCH 0894/1381] af_unix: Remove unix_tot_inflight. unix_tot_inflight is no longer used. Let's remove it. Signed-off-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20251115020935.2643121-7-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/unix/garbage.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/net/unix/garbage.c b/net/unix/garbage.c index a6929226d40d..fe1f74345b66 100644 --- a/net/unix/garbage.c +++ b/net/unix/garbage.c @@ -200,7 +200,6 @@ static void unix_free_vertices(struct scm_fp_list *fpl) } static DEFINE_SPINLOCK(unix_gc_lock); -static unsigned int unix_tot_inflight; void unix_add_edges(struct scm_fp_list *fpl, struct unix_sock *receiver) { @@ -226,7 +225,6 @@ void unix_add_edges(struct scm_fp_list *fpl, struct unix_sock *receiver) } while (i < fpl->count_unix); receiver->scm_stat.nr_unix_fds += fpl->count_unix; - WRITE_ONCE(unix_tot_inflight, unix_tot_inflight + fpl->count_unix); out: WRITE_ONCE(fpl->user->unix_inflight, fpl->user->unix_inflight + fpl->count); @@ -257,7 +255,6 @@ void unix_del_edges(struct scm_fp_list *fpl) receiver = fpl->edges[0].successor; receiver->scm_stat.nr_unix_fds -= fpl->count_unix; } - WRITE_ONCE(unix_tot_inflight, unix_tot_inflight - fpl->count_unix); out: WRITE_ONCE(fpl->user->unix_inflight, fpl->user->unix_inflight - fpl->count); From 24fa77dad25c2f55cc4615c09df2201ef72c66f4 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Sat, 15 Nov 2025 02:08:38 +0000 Subject: [PATCH 0895/1381] af_unix: Consolidate unix_schedule_gc() and wait_for_unix_gc(). unix_schedule_gc() and wait_for_unix_gc() share some code. Let's consolidate the two. Signed-off-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20251115020935.2643121-8-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/unix/af_unix.c | 2 +- net/unix/af_unix.h | 2 +- net/unix/garbage.c | 28 +++++++++------------------- 3 files changed, 11 insertions(+), 21 deletions(-) diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index 34952242bd81..e518116f8171 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -733,7 +733,7 @@ static void unix_release_sock(struct sock *sk, int embrion) /* ---- Socket is dead now and most probably destroyed ---- */ - unix_schedule_gc(); + unix_schedule_gc(NULL); } struct unix_peercred { diff --git a/net/unix/af_unix.h b/net/unix/af_unix.h index 2f1bfe3217c1..c4f1b2da363d 100644 --- a/net/unix/af_unix.h +++ b/net/unix/af_unix.h @@ -29,7 +29,7 @@ void unix_del_edges(struct scm_fp_list *fpl); void unix_update_edges(struct unix_sock *receiver); int unix_prepare_fpl(struct scm_fp_list *fpl); void unix_destroy_fpl(struct scm_fp_list *fpl); -void unix_schedule_gc(void); +void unix_schedule_gc(struct user_struct *user); /* SOCK_DIAG */ long unix_inq_len(struct sock *sk); diff --git a/net/unix/garbage.c b/net/unix/garbage.c index fe1f74345b66..78323d43e63e 100644 --- a/net/unix/garbage.c +++ b/net/unix/garbage.c @@ -279,8 +279,6 @@ void unix_update_edges(struct unix_sock *receiver) } } -static void wait_for_unix_gc(struct scm_fp_list *fpl); - int unix_prepare_fpl(struct scm_fp_list *fpl) { struct unix_vertex *vertex; @@ -302,7 +300,7 @@ int unix_prepare_fpl(struct scm_fp_list *fpl) if (!fpl->edges) goto err; - wait_for_unix_gc(fpl); + unix_schedule_gc(fpl->user); return 0; @@ -614,21 +612,9 @@ static void unix_gc(struct work_struct *work) static DECLARE_WORK(unix_gc_work, unix_gc); -void unix_schedule_gc(void) -{ - if (READ_ONCE(unix_graph_state) == UNIX_GRAPH_NOT_CYCLIC) - return; - - if (READ_ONCE(gc_in_progress)) - return; - - WRITE_ONCE(gc_in_progress, true); - queue_work(system_dfl_wq, &unix_gc_work); -} - #define UNIX_INFLIGHT_SANE_USER (SCM_MAX_FD * 8) -static void wait_for_unix_gc(struct scm_fp_list *fpl) +void unix_schedule_gc(struct user_struct *user) { if (READ_ONCE(unix_graph_state) == UNIX_GRAPH_NOT_CYCLIC) return; @@ -636,11 +622,15 @@ static void wait_for_unix_gc(struct scm_fp_list *fpl) /* Penalise users who want to send AF_UNIX sockets * but whose sockets have not been received yet. */ - if (READ_ONCE(fpl->user->unix_inflight) < UNIX_INFLIGHT_SANE_USER) + if (user && + READ_ONCE(user->unix_inflight) < UNIX_INFLIGHT_SANE_USER) return; - unix_schedule_gc(); + if (!READ_ONCE(gc_in_progress)) { + WRITE_ONCE(gc_in_progress, true); + queue_work(system_dfl_wq, &unix_gc_work); + } - if (READ_ONCE(unix_graph_cyclic_sccs)) + if (user && READ_ONCE(unix_graph_cyclic_sccs)) flush_work(&unix_gc_work); } From f72514b3c5698e4b900b25345e09f9ed33123de6 Mon Sep 17 00:00:00 2001 From: Fernando Fernandez Mancera Date: Sat, 15 Nov 2025 10:59:38 +0100 Subject: [PATCH 0896/1381] ipv6: clear RA flags when adding a static route When an IPv6 Router Advertisement (RA) is received for a prefix, the kernel creates the corresponding on-link route with flags RTF_ADDRCONF and RTF_PREFIX_RT configured and RTF_EXPIRES if lifetime is set. If later a user configures a static IPv6 address on the same prefix the kernel clears the RTF_EXPIRES flag but it doesn't clear the RTF_ADDRCONF and RTF_PREFIX_RT. When the next RA for that prefix is received, the kernel sees the route as RA-learned and wrongly configures back the lifetime. This is problematic because if the route expires, the static address won't have the corresponding on-link route. This fix clears the RTF_ADDRCONF and RTF_PREFIX_RT flags preventing that the lifetime is configured when the next RA arrives. If the static address is deleted, the route becomes RA-learned again. Fixes: 14ef37b6d00e ("ipv6: fix route lookup in addrconf_prefix_rcv()") Reported-by: Garri Djavadyan Closes: https://lore.kernel.org/netdev/ba807d39aca5b4dcf395cc11dca61a130a52cfd3.camel@gmail.com/ Signed-off-by: Fernando Fernandez Mancera Reviewed-by: David Ahern Link: https://patch.msgid.link/20251115095939.6967-1-fmancera@suse.de Signed-off-by: Jakub Kicinski --- net/ipv6/ip6_fib.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index 02c16909f618..2111af022d94 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -1138,6 +1138,10 @@ static int fib6_add_rt2node(struct fib6_node *fn, struct fib6_info *rt, fib6_set_expires(iter, rt->expires); fib6_add_gc_list(iter); } + if (!(rt->fib6_flags & (RTF_ADDRCONF | RTF_PREFIX_RT))) { + iter->fib6_flags &= ~RTF_ADDRCONF; + iter->fib6_flags &= ~RTF_PREFIX_RT; + } if (rt->fib6_pmtu) fib6_metric_set(iter, RTAX_MTU, From d7dbda87892039f19805b896d1fe0fa75aca1b06 Mon Sep 17 00:00:00 2001 From: Fernando Fernandez Mancera Date: Sat, 15 Nov 2025 10:59:39 +0100 Subject: [PATCH 0897/1381] selftests: fib_tests: add fib6 from ra to static test The new test checks that a route that has been promoted from RA-learned to static does not switch back when a new RA message arrives. In addition, it checks that the route is owned by RA again when the static address is removed. Signed-off-by: Fernando Fernandez Mancera Link: https://patch.msgid.link/20251115095939.6967-2-fmancera@suse.de Signed-off-by: Jakub Kicinski --- tools/testing/selftests/net/fib_tests.sh | 66 +++++++++++++++++++++++- 1 file changed, 65 insertions(+), 1 deletion(-) diff --git a/tools/testing/selftests/net/fib_tests.sh b/tools/testing/selftests/net/fib_tests.sh index a94b73a53f72..a88f797c549a 100755 --- a/tools/testing/selftests/net/fib_tests.sh +++ b/tools/testing/selftests/net/fib_tests.sh @@ -11,7 +11,8 @@ TESTS="unregister down carrier nexthop suppress ipv6_notify ipv4_notify \ ipv6_rt ipv4_rt ipv6_addr_metric ipv4_addr_metric ipv6_route_metrics \ ipv4_route_metrics ipv4_route_v6_gw rp_filter ipv4_del_addr \ ipv6_del_addr ipv4_mangle ipv6_mangle ipv4_bcast_neigh fib6_gc_test \ - ipv4_mpath_list ipv6_mpath_list ipv4_mpath_balance ipv6_mpath_balance" + ipv4_mpath_list ipv6_mpath_list ipv4_mpath_balance ipv6_mpath_balance \ + fib6_ra_to_static" VERBOSE=0 PAUSE_ON_FAIL=no @@ -1476,6 +1477,68 @@ ipv6_route_metrics_test() route_cleanup } +fib6_ra_to_static() +{ + setup + + echo + echo "Fib6 route promotion from RA-learned to static test" + set -e + + # ra6 is required for the test. (ipv6toolkit) + if [ ! -x "$(command -v ra6)" ]; then + echo "SKIP: ra6 not found." + set +e + cleanup &> /dev/null + return + fi + + # Create a pair of veth devices to send a RA message from one + # device to another. + $IP link add veth1 type veth peer name veth2 + $IP link set dev veth1 up + $IP link set dev veth2 up + $IP -6 address add 2001:10::1/64 dev veth1 nodad + $IP -6 address add 2001:10::2/64 dev veth2 nodad + + # Make veth1 ready to receive RA messages. + $NS_EXEC sysctl -wq net.ipv6.conf.veth1.accept_ra=2 + + # Send a RA message with a prefix from veth2. + $NS_EXEC ra6 -i veth2 -d 2001:10::1 -P 2001:12::/64\#LA\#120\#60 + + # Wait for the RA message. + sleep 1 + + # systemd may mess up the test. Make sure that + # systemd-networkd.service and systemd-networkd.socket are stopped. + check_rt_num_clean 2 $($IP -6 route list|grep expires|wc -l) || return + + # Configure static address on the same prefix + $IP -6 address add 2001:12::dead/64 dev veth1 nodad + + # On-link route won't expire anymore, default route still owned by RA + check_rt_num 1 $($IP -6 route list |grep expires|wc -l) + + # Send a second RA message with a prefix from veth2. + $NS_EXEC ra6 -i veth2 -d 2001:10::1 -P 2001:12::/64\#LA\#120\#60 + sleep 1 + + # Expire is not back, on-link route is still static + check_rt_num 1 $($IP -6 route list |grep expires|wc -l) + + $IP -6 address del 2001:12::dead/64 dev veth1 nodad + + # Expire is back, on-link route is now owned by RA again + check_rt_num 2 $($IP -6 route list |grep expires|wc -l) + + log_test $ret 0 "ipv6 promote RA route to static" + + set +e + + cleanup &> /dev/null +} + # add route for a prefix, flushing any existing routes first # expected to be the first step of a test add_route() @@ -2798,6 +2861,7 @@ do ipv6_mpath_list) ipv6_mpath_list_test;; ipv4_mpath_balance) ipv4_mpath_balance_test;; ipv6_mpath_balance) ipv6_mpath_balance_test;; + fib6_ra_to_static) fib6_ra_to_static;; help) echo "Test names: $TESTS"; exit 0;; esac From 7ac60a14d3fce87f0bfd0e50a7bfd5e683c33817 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Tue, 18 Nov 2025 09:41:33 +0000 Subject: [PATCH 0898/1381] net: stmmac: convert priv->sph* to boolean and rename priv->sph* only have 'true' and 'false' used with them, yet they are an int. Change their type to a bool, and rename to make their usage more clear. Signed-off-by: Russell King (Oracle) Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/E1vLIDN-0000000Evur-2NLU@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/stmmac.h | 4 +-- .../net/ethernet/stmicro/stmmac/stmmac_main.c | 26 +++++++++---------- .../stmicro/stmmac/stmmac_selftests.c | 2 +- .../net/ethernet/stmicro/stmmac/stmmac_xdp.c | 2 +- 4 files changed, 17 insertions(+), 17 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac.h b/drivers/net/ethernet/stmicro/stmmac/stmmac.h index e9ed5086c049..012b0a477255 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac.h +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac.h @@ -254,8 +254,8 @@ struct stmmac_priv { int hwts_tx_en; bool tx_path_in_lpi_mode; bool tso; - int sph; - int sph_cap; + bool sph_active; + bool sph_capable; u32 sarc_type; u32 rx_riwt[MTL_MAX_RX_QUEUES]; int hwts_rx_en; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index a862d5f91fce..32edeeef9d8f 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -1527,7 +1527,7 @@ static int stmmac_init_rx_buffers(struct stmmac_priv *priv, buf->page_offset = stmmac_rx_offset(priv); } - if (priv->sph && !buf->sec_page) { + if (priv->sph_active && !buf->sec_page) { buf->sec_page = page_pool_alloc_pages(rx_q->page_pool, gfp); if (!buf->sec_page) return -ENOMEM; @@ -2113,7 +2113,7 @@ static int __alloc_dma_rx_desc_resources(struct stmmac_priv *priv, pp_params.offset = stmmac_rx_offset(priv); pp_params.max_len = dma_conf->dma_buf_sz; - if (priv->sph) { + if (priv->sph_active) { pp_params.offset = 0; pp_params.max_len += stmmac_rx_offset(priv); } @@ -3607,7 +3607,7 @@ static int stmmac_hw_setup(struct net_device *dev) } /* Enable Split Header */ - sph_en = (priv->hw->rx_csum > 0) && priv->sph; + sph_en = (priv->hw->rx_csum > 0) && priv->sph_active; for (chan = 0; chan < rx_cnt; chan++) stmmac_enable_sph(priv, priv->ioaddr, sph_en, chan); @@ -4899,7 +4899,7 @@ static inline void stmmac_rx_refill(struct stmmac_priv *priv, u32 queue) break; } - if (priv->sph && !buf->sec_page) { + if (priv->sph_active && !buf->sec_page) { buf->sec_page = page_pool_alloc_pages(rx_q->page_pool, gfp); if (!buf->sec_page) break; @@ -4910,7 +4910,7 @@ static inline void stmmac_rx_refill(struct stmmac_priv *priv, u32 queue) buf->addr = page_pool_get_dma_addr(buf->page) + buf->page_offset; stmmac_set_desc_addr(priv, p, buf->addr); - if (priv->sph) + if (priv->sph_active) stmmac_set_desc_sec_addr(priv, p, buf->sec_addr, true); else stmmac_set_desc_sec_addr(priv, p, buf->sec_addr, false); @@ -4945,12 +4945,12 @@ static unsigned int stmmac_rx_buf1_len(struct stmmac_priv *priv, int coe = priv->hw->rx_csum; /* Not first descriptor, buffer is always zero */ - if (priv->sph && len) + if (priv->sph_active && len) return 0; /* First descriptor, get split header length */ stmmac_get_rx_header_len(priv, p, &hlen); - if (priv->sph && hlen) { + if (priv->sph_active && hlen) { priv->xstats.rx_split_hdr_pkt_n++; return hlen; } @@ -4973,7 +4973,7 @@ static unsigned int stmmac_rx_buf2_len(struct stmmac_priv *priv, unsigned int plen = 0; /* Not split header, buffer is not available */ - if (!priv->sph) + if (!priv->sph_active) return 0; /* Not last descriptor */ @@ -6041,8 +6041,8 @@ static int stmmac_set_features(struct net_device *netdev, */ stmmac_rx_ipc(priv, priv->hw); - if (priv->sph_cap) { - bool sph_en = (priv->hw->rx_csum > 0) && priv->sph; + if (priv->sph_capable) { + bool sph_en = (priv->hw->rx_csum > 0) && priv->sph_active; u32 chan; for (chan = 0; chan < priv->plat->rx_queues_to_use; chan++) @@ -6991,7 +6991,7 @@ int stmmac_xdp_open(struct net_device *dev) } /* Adjust Split header */ - sph_en = (priv->hw->rx_csum > 0) && priv->sph; + sph_en = (priv->hw->rx_csum > 0) && priv->sph_active; /* DMA RX Channel Configuration */ for (chan = 0; chan < rx_cnt; chan++) { @@ -7740,8 +7740,8 @@ int stmmac_dvr_probe(struct device *device, if (priv->dma_cap.sphen && !(priv->plat->flags & STMMAC_FLAG_SPH_DISABLE)) { ndev->hw_features |= NETIF_F_GRO; - priv->sph_cap = true; - priv->sph = priv->sph_cap; + priv->sph_capable = true; + priv->sph_active = priv->sph_capable; dev_info(priv->device, "SPH feature enabled\n"); } diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_selftests.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_selftests.c index a01bc394d1ac..e90a2c469b9a 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_selftests.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_selftests.c @@ -1721,7 +1721,7 @@ static int stmmac_test_sph(struct stmmac_priv *priv) struct stmmac_packet_attrs attr = { }; int ret; - if (!priv->sph) + if (!priv->sph_active) return -EOPNOTSUPP; /* Check for UDP first */ diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_xdp.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_xdp.c index aa6f16d3df64..d7e4db7224b0 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_xdp.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_xdp.c @@ -129,7 +129,7 @@ int stmmac_xdp_set_prog(struct stmmac_priv *priv, struct bpf_prog *prog, bpf_prog_put(old_prog); /* Disable RX SPH for XDP operation */ - priv->sph = priv->sph_cap && !stmmac_xdp_is_enabled(priv); + priv->sph_active = priv->sph_capable && !stmmac_xdp_is_enabled(priv); if (if_running && need_update) stmmac_xdp_open(dev); From b5adada61e02c19fce48641283f0031d28981066 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Tue, 18 Nov 2025 10:01:15 +0000 Subject: [PATCH 0899/1381] net: stmmac: stmmac_is_jumbo_frm() len should be unsigned stmmac_is_jumbo_frm() and the is_jumbo_frm() methods take skb->len which is an unsigned int. Avoid an implicit cast to "int" via the method parameter and then incorrectly doing signed comparisons on this unsigned value. Signed-off-by: Russell King (Oracle) Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/E1vLIWR-0000000Ewkf-1Tdx@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/chain_mode.c | 2 +- drivers/net/ethernet/stmicro/stmmac/hwif.h | 2 +- drivers/net/ethernet/stmicro/stmmac/ring_mode.c | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/chain_mode.c b/drivers/net/ethernet/stmicro/stmmac/chain_mode.c index fb55efd52240..d14b56e5ed40 100644 --- a/drivers/net/ethernet/stmicro/stmmac/chain_mode.c +++ b/drivers/net/ethernet/stmicro/stmmac/chain_mode.c @@ -83,7 +83,7 @@ static int jumbo_frm(struct stmmac_tx_queue *tx_q, struct sk_buff *skb, return entry; } -static unsigned int is_jumbo_frm(int len, int enh_desc) +static unsigned int is_jumbo_frm(unsigned int len, int enh_desc) { unsigned int ret = 0; diff --git a/drivers/net/ethernet/stmicro/stmmac/hwif.h b/drivers/net/ethernet/stmicro/stmmac/hwif.h index d359722100fa..4953e0fab547 100644 --- a/drivers/net/ethernet/stmicro/stmmac/hwif.h +++ b/drivers/net/ethernet/stmicro/stmmac/hwif.h @@ -541,7 +541,7 @@ struct stmmac_rx_queue; struct stmmac_mode_ops { void (*init) (void *des, dma_addr_t phy_addr, unsigned int size, unsigned int extend_desc); - unsigned int (*is_jumbo_frm) (int len, int ehn_desc); + unsigned int (*is_jumbo_frm)(unsigned int len, int ehn_desc); int (*jumbo_frm)(struct stmmac_tx_queue *tx_q, struct sk_buff *skb, int csum); int (*set_16kib_bfsize)(int mtu); diff --git a/drivers/net/ethernet/stmicro/stmmac/ring_mode.c b/drivers/net/ethernet/stmicro/stmmac/ring_mode.c index d218412ca832..039903c424df 100644 --- a/drivers/net/ethernet/stmicro/stmmac/ring_mode.c +++ b/drivers/net/ethernet/stmicro/stmmac/ring_mode.c @@ -91,7 +91,7 @@ static int jumbo_frm(struct stmmac_tx_queue *tx_q, struct sk_buff *skb, return entry; } -static unsigned int is_jumbo_frm(int len, int enh_desc) +static unsigned int is_jumbo_frm(unsigned int len, int enh_desc) { unsigned int ret = 0; From bf351bbec57fd4d09177974f762cf553a5ef8663 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Tue, 18 Nov 2025 10:01:20 +0000 Subject: [PATCH 0900/1381] net: stmmac: stmmac_is_jumbo_frm() returns boolean stmmac_is_jumbo_frm() returns whether the driver considers the frame size to be a jumbo frame, and thus returns 0/1 values. This is boolean, so convert it to return a boolean and use false/true instead. Also convert stmmac_xmit()'s is_jumbo to be bool, which causes several variables to be repositioned to keep it in reverse Christmas-tree order. Signed-off-by: Russell King (Oracle) Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/E1vLIWW-0000000Ewkl-21Ia@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/chain_mode.c | 9 ++++----- drivers/net/ethernet/stmicro/stmmac/hwif.h | 2 +- drivers/net/ethernet/stmicro/stmmac/ring_mode.c | 9 ++------- drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 6 +++--- 4 files changed, 10 insertions(+), 16 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/chain_mode.c b/drivers/net/ethernet/stmicro/stmmac/chain_mode.c index d14b56e5ed40..120a009c9992 100644 --- a/drivers/net/ethernet/stmicro/stmmac/chain_mode.c +++ b/drivers/net/ethernet/stmicro/stmmac/chain_mode.c @@ -83,14 +83,13 @@ static int jumbo_frm(struct stmmac_tx_queue *tx_q, struct sk_buff *skb, return entry; } -static unsigned int is_jumbo_frm(unsigned int len, int enh_desc) +static bool is_jumbo_frm(unsigned int len, bool enh_desc) { - unsigned int ret = 0; + bool ret = false; if ((enh_desc && (len > BUF_SIZE_8KiB)) || - (!enh_desc && (len > BUF_SIZE_2KiB))) { - ret = 1; - } + (!enh_desc && (len > BUF_SIZE_2KiB))) + ret = true; return ret; } diff --git a/drivers/net/ethernet/stmicro/stmmac/hwif.h b/drivers/net/ethernet/stmicro/stmmac/hwif.h index 4953e0fab547..f257ce4b6c66 100644 --- a/drivers/net/ethernet/stmicro/stmmac/hwif.h +++ b/drivers/net/ethernet/stmicro/stmmac/hwif.h @@ -541,7 +541,7 @@ struct stmmac_rx_queue; struct stmmac_mode_ops { void (*init) (void *des, dma_addr_t phy_addr, unsigned int size, unsigned int extend_desc); - unsigned int (*is_jumbo_frm)(unsigned int len, int ehn_desc); + bool (*is_jumbo_frm)(unsigned int len, bool enh_desc); int (*jumbo_frm)(struct stmmac_tx_queue *tx_q, struct sk_buff *skb, int csum); int (*set_16kib_bfsize)(int mtu); diff --git a/drivers/net/ethernet/stmicro/stmmac/ring_mode.c b/drivers/net/ethernet/stmicro/stmmac/ring_mode.c index 039903c424df..382d94a3b972 100644 --- a/drivers/net/ethernet/stmicro/stmmac/ring_mode.c +++ b/drivers/net/ethernet/stmicro/stmmac/ring_mode.c @@ -91,14 +91,9 @@ static int jumbo_frm(struct stmmac_tx_queue *tx_q, struct sk_buff *skb, return entry; } -static unsigned int is_jumbo_frm(unsigned int len, int enh_desc) +static bool is_jumbo_frm(unsigned int len, bool enh_desc) { - unsigned int ret = 0; - - if (len >= BUF_SIZE_4KiB) - ret = 1; - - return ret; + return len >= BUF_SIZE_4KiB; } static void refill_desc3(struct stmmac_rx_queue *rx_q, struct dma_desc *p) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index 32edeeef9d8f..15b0c08ebd87 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -4583,18 +4583,18 @@ static bool stmmac_has_ip_ethertype(struct sk_buff *skb) */ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev) { - unsigned int first_entry, tx_packets, enh_desc; + bool enh_desc, has_vlan, set_ic, is_jumbo = false; struct stmmac_priv *priv = netdev_priv(dev); unsigned int nopaged_len = skb_headlen(skb); - int i, csum_insertion = 0, is_jumbo = 0; u32 queue = skb_get_queue_mapping(skb); int nfrags = skb_shinfo(skb)->nr_frags; + unsigned int first_entry, tx_packets; int gso = skb_shinfo(skb)->gso_type; struct stmmac_txq_stats *txq_stats; struct dma_edesc *tbs_desc = NULL; struct dma_desc *desc, *first; struct stmmac_tx_queue *tx_q; - bool has_vlan, set_ic; + int i, csum_insertion = 0; int entry, first_tx; dma_addr_t des; u32 sdu_len; From bc41fbbf6faa9ffeaf0148019ed631077f7f150f Mon Sep 17 00:00:00 2001 From: Sjoerd Simons Date: Sat, 15 Nov 2025 21:58:09 +0100 Subject: [PATCH 0901/1381] dt-bindings: net: mediatek,net: Correct bindings for MT7981 Different SoCs have different numbers of Wireless Ethernet Dispatch (WED) units: - MT7981: Has 1 WED unit - MT7986: Has 2 WED units - MT7988: Has 2 WED units Update the binding to reflect these hardware differences. The MT7981 also uses infracfg for PHY switching, so allow that property. Signed-off-by: Sjoerd Simons Acked-by: Conor Dooley Link: https://patch.msgid.link/20251115-openwrt-one-network-v4-6-48cbda2969ac@collabora.com Signed-off-by: Jakub Kicinski --- .../devicetree/bindings/net/mediatek,net.yaml | 26 ++++++++++++++++--- 1 file changed, 23 insertions(+), 3 deletions(-) diff --git a/Documentation/devicetree/bindings/net/mediatek,net.yaml b/Documentation/devicetree/bindings/net/mediatek,net.yaml index b45f67f92e80..cc346946291a 100644 --- a/Documentation/devicetree/bindings/net/mediatek,net.yaml +++ b/Documentation/devicetree/bindings/net/mediatek,net.yaml @@ -112,7 +112,7 @@ properties: mediatek,wed: $ref: /schemas/types.yaml#/definitions/phandle-array - minItems: 2 + minItems: 1 maxItems: 2 items: maxItems: 1 @@ -249,6 +249,9 @@ allOf: minItems: 1 maxItems: 1 + mediatek,wed: + minItems: 2 + mediatek,wed-pcie: false else: properties: @@ -338,12 +341,13 @@ allOf: - const: netsys0 - const: netsys1 - mediatek,infracfg: false - mediatek,sgmiisys: minItems: 2 maxItems: 2 + mediatek,wed: + maxItems: 1 + - if: properties: compatible: @@ -385,6 +389,9 @@ allOf: minItems: 2 maxItems: 2 + mediatek,wed: + minItems: 2 + - if: properties: compatible: @@ -429,6 +436,19 @@ allOf: - const: xgp2 - const: xgp3 + mediatek,wed: + minItems: 2 + + - if: + properties: + compatible: + contains: + const: ralink,rt5350-eth + then: + properties: + mediatek,wed: + minItems: 2 + patternProperties: "^mac@[0-2]$": type: object From 929ca3bceab8eaeeb452c0574999c7c6d12b3772 Mon Sep 17 00:00:00 2001 From: Anshumali Gaur Date: Tue, 18 Nov 2025 11:12:34 +0530 Subject: [PATCH 0902/1381] octeontx2-af: Skip TM tree print for disabled SQs Currently, the TM tree is printing all SQ topology including those which are not enabled, this results in redundant output for SQs which are not active. This patch adds a check in print_tm_tree() to skip printing the TM tree hierarchy if the SQ is not enabled. Signed-off-by: Anshumali Gaur Reviewed-by: Simon Horman Link: https://patch.msgid.link/20251118054235.1599714-1-agaur@marvell.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c index 7370812ece2a..15d3cb0b9da6 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c @@ -1663,6 +1663,9 @@ static void print_tm_tree(struct seq_file *m, int blkaddr; u64 cfg; + if (!sq_ctx->ena) + return; + blkaddr = nix_hw->blkaddr; schq = sq_ctx->smq; From 934fa943b53795339486cc0026b3ab7ad39dc600 Mon Sep 17 00:00:00 2001 From: Aditya Garg Date: Tue, 18 Nov 2025 03:11:08 -0800 Subject: [PATCH 0903/1381] net: mana: Handle SKB if TX SGEs exceed hardware limit The MANA hardware supports a maximum of 30 scatter-gather entries (SGEs) per TX WQE. Exceeding this limit can cause TX failures. Add ndo_features_check() callback to validate SKB layout before transmission. For GSO SKBs that would exceed the hardware SGE limit, clear NETIF_F_GSO_MASK to enforce software segmentation in the stack. Add a fallback in mana_start_xmit() to linearize non-GSO SKBs that still exceed the SGE limit. Also, Add ethtool counter for SKBs linearized Co-developed-by: Dipayaan Roy Signed-off-by: Dipayaan Roy Signed-off-by: Aditya Garg Reviewed-by: Eric Dumazet Reviewed-by: Haiyang Zhang Link: https://patch.msgid.link/1763464269-10431-2-git-send-email-gargaditya@linux.microsoft.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/microsoft/mana/mana_en.c | 40 ++++++++++++++++++- .../ethernet/microsoft/mana/mana_ethtool.c | 2 + include/net/mana/gdma.h | 8 +++- include/net/mana/mana.h | 1 + 4 files changed, 48 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/microsoft/mana/mana_en.c b/drivers/net/ethernet/microsoft/mana/mana_en.c index 13f47be7aca6..7b49ab005e2d 100644 --- a/drivers/net/ethernet/microsoft/mana/mana_en.c +++ b/drivers/net/ethernet/microsoft/mana/mana_en.c @@ -11,6 +11,7 @@ #include #include #include +#include #include #include @@ -329,6 +330,21 @@ netdev_tx_t mana_start_xmit(struct sk_buff *skb, struct net_device *ndev) cq = &apc->tx_qp[txq_idx].tx_cq; tx_stats = &txq->stats; + BUILD_BUG_ON(MAX_TX_WQE_SGL_ENTRIES != MANA_MAX_TX_WQE_SGL_ENTRIES); + if (MAX_SKB_FRAGS + 2 > MAX_TX_WQE_SGL_ENTRIES && + skb_shinfo(skb)->nr_frags + 2 > MAX_TX_WQE_SGL_ENTRIES) { + /* GSO skb with Hardware SGE limit exceeded is not expected here + * as they are handled in mana_features_check() callback + */ + if (skb_linearize(skb)) { + netdev_warn_once(ndev, "Failed to linearize skb with nr_frags=%d and is_gso=%d\n", + skb_shinfo(skb)->nr_frags, + skb_is_gso(skb)); + goto tx_drop_count; + } + apc->eth_stats.tx_linear_pkt_cnt++; + } + pkg.tx_oob.s_oob.vcq_num = cq->gdma_id; pkg.tx_oob.s_oob.vsq_frame = txq->vsq_frame; @@ -442,8 +458,6 @@ netdev_tx_t mana_start_xmit(struct sk_buff *skb, struct net_device *ndev) } } - WARN_ON_ONCE(pkg.wqe_req.num_sge > MAX_TX_WQE_SGL_ENTRIES); - if (pkg.wqe_req.num_sge <= ARRAY_SIZE(pkg.sgl_array)) { pkg.wqe_req.sgl = pkg.sgl_array; } else { @@ -518,6 +532,25 @@ netdev_tx_t mana_start_xmit(struct sk_buff *skb, struct net_device *ndev) return NETDEV_TX_OK; } +#if (MAX_SKB_FRAGS + 2 > MANA_MAX_TX_WQE_SGL_ENTRIES) +static netdev_features_t mana_features_check(struct sk_buff *skb, + struct net_device *ndev, + netdev_features_t features) +{ + if (skb_shinfo(skb)->nr_frags + 2 > MAX_TX_WQE_SGL_ENTRIES) { + /* Exceeds HW SGE limit. + * GSO case: + * Disable GSO so the stack will software-segment the skb + * into smaller skbs that fit the SGE budget. + * Non-GSO case: + * The xmit path will attempt skb_linearize() as a fallback. + */ + features &= ~NETIF_F_GSO_MASK; + } + return features; +} +#endif + static void mana_get_stats64(struct net_device *ndev, struct rtnl_link_stats64 *st) { @@ -883,6 +916,9 @@ static const struct net_device_ops mana_devops = { .ndo_open = mana_open, .ndo_stop = mana_close, .ndo_select_queue = mana_select_queue, +#if (MAX_SKB_FRAGS + 2 > MANA_MAX_TX_WQE_SGL_ENTRIES) + .ndo_features_check = mana_features_check, +#endif .ndo_start_xmit = mana_start_xmit, .ndo_validate_addr = eth_validate_addr, .ndo_get_stats64 = mana_get_stats64, diff --git a/drivers/net/ethernet/microsoft/mana/mana_ethtool.c b/drivers/net/ethernet/microsoft/mana/mana_ethtool.c index 99e811208683..0e2f4343ac67 100644 --- a/drivers/net/ethernet/microsoft/mana/mana_ethtool.c +++ b/drivers/net/ethernet/microsoft/mana/mana_ethtool.c @@ -18,6 +18,8 @@ static const struct mana_stats_desc mana_eth_stats[] = { {"tx_cq_err", offsetof(struct mana_ethtool_stats, tx_cqe_err)}, {"tx_cqe_unknown_type", offsetof(struct mana_ethtool_stats, tx_cqe_unknown_type)}, + {"tx_linear_pkt_cnt", offsetof(struct mana_ethtool_stats, + tx_linear_pkt_cnt)}, {"rx_coalesced_err", offsetof(struct mana_ethtool_stats, rx_coalesced_err)}, {"rx_cqe_unknown_type", offsetof(struct mana_ethtool_stats, diff --git a/include/net/mana/gdma.h b/include/net/mana/gdma.h index 2e4f2f3175e5..a4cf307859f8 100644 --- a/include/net/mana/gdma.h +++ b/include/net/mana/gdma.h @@ -486,6 +486,8 @@ struct gdma_wqe { #define INLINE_OOB_SMALL_SIZE 8 #define INLINE_OOB_LARGE_SIZE 24 +#define MANA_MAX_TX_WQE_SGL_ENTRIES 30 + #define MAX_TX_WQE_SIZE 512 #define MAX_RX_WQE_SIZE 256 @@ -592,6 +594,9 @@ enum { #define GDMA_DRV_CAP_FLAG_1_HANDLE_RECONFIG_EQE BIT(17) #define GDMA_DRV_CAP_FLAG_1_HW_VPORT_LINK_AWARE BIT(6) +/* Driver supports linearizing the skb when num_sge exceeds hardware limit */ +#define GDMA_DRV_CAP_FLAG_1_SKB_LINEARIZE BIT(20) + /* Driver can send HWC periodically to query stats */ #define GDMA_DRV_CAP_FLAG_1_PERIODIC_STATS_QUERY BIT(21) @@ -605,7 +610,8 @@ enum { GDMA_DRV_CAP_FLAG_1_SELF_RESET_ON_EQE | \ GDMA_DRV_CAP_FLAG_1_HANDLE_RECONFIG_EQE | \ GDMA_DRV_CAP_FLAG_1_HW_VPORT_LINK_AWARE | \ - GDMA_DRV_CAP_FLAG_1_PERIODIC_STATS_QUERY) + GDMA_DRV_CAP_FLAG_1_PERIODIC_STATS_QUERY | \ + GDMA_DRV_CAP_FLAG_1_SKB_LINEARIZE) #define GDMA_DRV_CAP_FLAGS2 0 diff --git a/include/net/mana/mana.h b/include/net/mana/mana.h index d37f4cea0ac3..fb28b3cac067 100644 --- a/include/net/mana/mana.h +++ b/include/net/mana/mana.h @@ -377,6 +377,7 @@ struct mana_ethtool_stats { u64 wake_queue; u64 tx_cqe_err; u64 tx_cqe_unknown_type; + u64 tx_linear_pkt_cnt; u64 rx_coalesced_err; u64 rx_cqe_unknown_type; }; From 45120304e84171fd215c1b57b15b285446d15106 Mon Sep 17 00:00:00 2001 From: Aditya Garg Date: Tue, 18 Nov 2025 03:11:09 -0800 Subject: [PATCH 0904/1381] net: mana: Drop TX skb on post_work_request failure and unmap resources Drop TX packets when posting the work request fails and ensure DMA mappings are always cleaned up. Signed-off-by: Aditya Garg Reviewed-by: Haiyang Zhang Link: https://patch.msgid.link/1763464269-10431-3-git-send-email-gargaditya@linux.microsoft.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/microsoft/mana/gdma_main.c | 6 +----- drivers/net/ethernet/microsoft/mana/mana_en.c | 7 +++---- include/net/mana/mana.h | 1 + 3 files changed, 5 insertions(+), 9 deletions(-) diff --git a/drivers/net/ethernet/microsoft/mana/gdma_main.c b/drivers/net/ethernet/microsoft/mana/gdma_main.c index effe0a2f207a..8fd70b34807a 100644 --- a/drivers/net/ethernet/microsoft/mana/gdma_main.c +++ b/drivers/net/ethernet/microsoft/mana/gdma_main.c @@ -1300,7 +1300,6 @@ int mana_gd_post_work_request(struct gdma_queue *wq, struct gdma_posted_wqe_info *wqe_info) { u32 client_oob_size = wqe_req->inline_oob_size; - struct gdma_context *gc; u32 sgl_data_size; u32 max_wqe_size; u32 wqe_size; @@ -1330,11 +1329,8 @@ int mana_gd_post_work_request(struct gdma_queue *wq, if (wqe_size > max_wqe_size) return -EINVAL; - if (wq->monitor_avl_buf && wqe_size > mana_gd_wq_avail_space(wq)) { - gc = wq->gdma_dev->gdma_context; - dev_err(gc->dev, "unsuccessful flow control!\n"); + if (wq->monitor_avl_buf && wqe_size > mana_gd_wq_avail_space(wq)) return -ENOSPC; - } if (wqe_info) wqe_info->wqe_size_in_bu = wqe_size / GDMA_WQE_BU_SIZE; diff --git a/drivers/net/ethernet/microsoft/mana/mana_en.c b/drivers/net/ethernet/microsoft/mana/mana_en.c index 7b49ab005e2d..1ad154f9db1a 100644 --- a/drivers/net/ethernet/microsoft/mana/mana_en.c +++ b/drivers/net/ethernet/microsoft/mana/mana_en.c @@ -492,9 +492,9 @@ netdev_tx_t mana_start_xmit(struct sk_buff *skb, struct net_device *ndev) if (err) { (void)skb_dequeue_tail(&txq->pending_skbs); + mana_unmap_skb(skb, apc); netdev_warn(ndev, "Failed to post TX OOB: %d\n", err); - err = NETDEV_TX_BUSY; - goto tx_busy; + goto free_sgl_ptr; } err = NETDEV_TX_OK; @@ -514,7 +514,6 @@ netdev_tx_t mana_start_xmit(struct sk_buff *skb, struct net_device *ndev) tx_stats->bytes += len + ((num_gso_seg - 1) * gso_hs); u64_stats_update_end(&tx_stats->syncp); -tx_busy: if (netif_tx_queue_stopped(net_txq) && mana_can_tx(gdma_sq)) { netif_tx_wake_queue(net_txq); apc->eth_stats.wake_queue++; @@ -1687,7 +1686,7 @@ static int mana_move_wq_tail(struct gdma_queue *wq, u32 num_units) return 0; } -static void mana_unmap_skb(struct sk_buff *skb, struct mana_port_context *apc) +void mana_unmap_skb(struct sk_buff *skb, struct mana_port_context *apc) { struct mana_skb_head *ash = (struct mana_skb_head *)skb->head; struct gdma_context *gc = apc->ac->gdma_dev->gdma_context; diff --git a/include/net/mana/mana.h b/include/net/mana/mana.h index fb28b3cac067..d7e089c6b694 100644 --- a/include/net/mana/mana.h +++ b/include/net/mana/mana.h @@ -593,6 +593,7 @@ int mana_set_bw_clamp(struct mana_port_context *apc, u32 speed, void mana_query_phy_stats(struct mana_port_context *apc); int mana_pre_alloc_rxbufs(struct mana_port_context *apc, int mtu, int num_queues); void mana_pre_dealloc_rxbufs(struct mana_port_context *apc); +void mana_unmap_skb(struct sk_buff *skb, struct mana_port_context *apc); extern const struct ethtool_ops mana_ethtool_ops; extern struct dentry *mana_debugfs_root; From 672cb5c2a9584802ad43d61209d5e24cccee823d Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Tue, 18 Nov 2025 01:44:56 -0800 Subject: [PATCH 0905/1381] net: vmxnet3: convert to use .get_rx_ring_count Convert the vmxnet3 driver to use the new .get_rx_ring_count ethtool operation instead of implementing .get_rxnfc solely for handling ETHTOOL_GRXRINGS command. This simplifies the code by removing the switch statement and replacing it with a direct return of the queue count. The new callback provides the same functionality in a more direct way, following the ongoing ethtool API modernization. Signed-off-by: Breno Leitao Link: https://patch.msgid.link/20251118-vmxnet3_grxrings-v1-1-ed8abddd2d52@debian.org Signed-off-by: Jakub Kicinski --- drivers/net/vmxnet3/vmxnet3_ethtool.c | 18 +++--------------- 1 file changed, 3 insertions(+), 15 deletions(-) diff --git a/drivers/net/vmxnet3/vmxnet3_ethtool.c b/drivers/net/vmxnet3/vmxnet3_ethtool.c index cc4d7573839d..a14d0ad978e1 100644 --- a/drivers/net/vmxnet3/vmxnet3_ethtool.c +++ b/drivers/net/vmxnet3/vmxnet3_ethtool.c @@ -1081,23 +1081,11 @@ vmxnet3_set_rss_hash_opt(struct net_device *netdev, return 0; } -static int -vmxnet3_get_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *info, - u32 *rules) +static u32 vmxnet3_get_rx_ring_count(struct net_device *netdev) { struct vmxnet3_adapter *adapter = netdev_priv(netdev); - int err = 0; - switch (info->cmd) { - case ETHTOOL_GRXRINGS: - info->data = adapter->num_rx_queues; - break; - default: - err = -EOPNOTSUPP; - break; - } - - return err; + return adapter->num_rx_queues; } #ifdef VMXNET3_RSS @@ -1335,7 +1323,7 @@ static const struct ethtool_ops vmxnet3_ethtool_ops = { .get_ethtool_stats = vmxnet3_get_ethtool_stats, .get_ringparam = vmxnet3_get_ringparam, .set_ringparam = vmxnet3_set_ringparam, - .get_rxnfc = vmxnet3_get_rxnfc, + .get_rx_ring_count = vmxnet3_get_rx_ring_count, #ifdef VMXNET3_RSS .get_rxfh_indir_size = vmxnet3_get_rss_indir_size, .get_rxfh = vmxnet3_get_rss, From 8e982441ba601d982dd0739972115d85ae01d99b Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Tue, 18 Nov 2025 01:40:28 +0200 Subject: [PATCH 0906/1381] net: phy: realtek: create rtl8211f_config_rgmii_delay() The control flow in rtl8211f_config_init() has some pitfalls which were probably unintended. Specifically it has an early return: switch (phydev->interface) { ... default: /* the rest of the modes imply leaving delay as is. */ return 0; } which exits the entire config_init() function. This means it also skips doing things such as disabling CLKOUT or disabling PHY-mode EEE. For the RTL8211FS, which uses PHY_INTERFACE_MODE_SGMII, this might be a problem. However, I don't know that it is, so there is no Fixes: tag. The issue was observed through code inspection. Signed-off-by: Vladimir Oltean Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20251117234033.345679-2-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- drivers/net/phy/realtek/realtek_main.c | 65 +++++++++++++++----------- 1 file changed, 39 insertions(+), 26 deletions(-) diff --git a/drivers/net/phy/realtek/realtek_main.c b/drivers/net/phy/realtek/realtek_main.c index 417f9a88aab6..896351022682 100644 --- a/drivers/net/phy/realtek/realtek_main.c +++ b/drivers/net/phy/realtek/realtek_main.c @@ -587,22 +587,11 @@ static int rtl8211c_config_init(struct phy_device *phydev) CTL1000_ENABLE_MASTER | CTL1000_AS_MASTER); } -static int rtl8211f_config_init(struct phy_device *phydev) +static int rtl8211f_config_rgmii_delay(struct phy_device *phydev) { - struct rtl821x_priv *priv = phydev->priv; - struct device *dev = &phydev->mdio.dev; u16 val_txdly, val_rxdly; int ret; - ret = phy_modify_paged_changed(phydev, RTL8211F_PHYCR_PAGE, RTL8211F_PHYCR1, - RTL8211F_ALDPS_PLL_OFF | RTL8211F_ALDPS_ENABLE | RTL8211F_ALDPS_XTAL_OFF, - priv->phycr1); - if (ret < 0) { - dev_err(dev, "aldps mode configuration failed: %pe\n", - ERR_PTR(ret)); - return ret; - } - switch (phydev->interface) { case PHY_INTERFACE_MODE_RGMII: val_txdly = 0; @@ -632,34 +621,58 @@ static int rtl8211f_config_init(struct phy_device *phydev) RTL8211F_TXCR, RTL8211F_TX_DELAY, val_txdly); if (ret < 0) { - dev_err(dev, "Failed to update the TX delay register\n"); + phydev_err(phydev, "Failed to update the TX delay register: %pe\n", + ERR_PTR(ret)); return ret; } else if (ret) { - dev_dbg(dev, - "%s 2ns TX delay (and changing the value from pin-strapping RXD1 or the bootloader)\n", - str_enable_disable(val_txdly)); + phydev_dbg(phydev, + "%s 2ns TX delay (and changing the value from pin-strapping RXD1 or the bootloader)\n", + str_enable_disable(val_txdly)); } else { - dev_dbg(dev, - "2ns TX delay was already %s (by pin-strapping RXD1 or bootloader configuration)\n", - str_enabled_disabled(val_txdly)); + phydev_dbg(phydev, + "2ns TX delay was already %s (by pin-strapping RXD1 or bootloader configuration)\n", + str_enabled_disabled(val_txdly)); } ret = phy_modify_paged_changed(phydev, RTL8211F_RGMII_PAGE, RTL8211F_RXCR, RTL8211F_RX_DELAY, val_rxdly); if (ret < 0) { - dev_err(dev, "Failed to update the RX delay register\n"); + phydev_err(phydev, "Failed to update the RX delay register: %pe\n", + ERR_PTR(ret)); return ret; } else if (ret) { - dev_dbg(dev, - "%s 2ns RX delay (and changing the value from pin-strapping RXD0 or the bootloader)\n", - str_enable_disable(val_rxdly)); + phydev_dbg(phydev, + "%s 2ns RX delay (and changing the value from pin-strapping RXD0 or the bootloader)\n", + str_enable_disable(val_rxdly)); } else { - dev_dbg(dev, - "2ns RX delay was already %s (by pin-strapping RXD0 or bootloader configuration)\n", - str_enabled_disabled(val_rxdly)); + phydev_dbg(phydev, + "2ns RX delay was already %s (by pin-strapping RXD0 or bootloader configuration)\n", + str_enabled_disabled(val_rxdly)); } + return 0; +} + +static int rtl8211f_config_init(struct phy_device *phydev) +{ + struct rtl821x_priv *priv = phydev->priv; + struct device *dev = &phydev->mdio.dev; + int ret; + + ret = phy_modify_paged_changed(phydev, RTL8211F_PHYCR_PAGE, RTL8211F_PHYCR1, + RTL8211F_ALDPS_PLL_OFF | RTL8211F_ALDPS_ENABLE | RTL8211F_ALDPS_XTAL_OFF, + priv->phycr1); + if (ret < 0) { + dev_err(dev, "aldps mode configuration failed: %pe\n", + ERR_PTR(ret)); + return ret; + } + + ret = rtl8211f_config_rgmii_delay(phydev); + if (ret) + return ret; + if (!priv->has_phycr2) return 0; From 27033d06917758d47162581da7e9de8004049dee Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Tue, 18 Nov 2025 01:40:29 +0200 Subject: [PATCH 0907/1381] net: phy: realtek: eliminate priv->phycr2 variable The RTL8211F(D)(I)-VD-CG PHY also has support for disabling the CLKOUT, and we'd like to introduce the "realtek,clkout-disable" property for that. But it isn't done through the PHYCR2 register, and it becomes awkward to have the driver pretend that it is. So just replace the machine-level "u16 phycr2" variable with a logical "bool disable_clk_out", which scales better to the other PHY as well. The change is a complete functional equivalent. Before, if the device tree property was absent, priv->phycr2 would contain the RTL8211F_CLKOUT_EN bit as read from hardware. Now, we don't save priv->phycr2, but we just don't call phy_modify_paged() on it. Also, we can simply call phy_modify_paged() with the "set" argument to 0. Signed-off-by: Vladimir Oltean Link: https://patch.msgid.link/20251117234033.345679-3-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- drivers/net/phy/realtek/realtek_main.c | 38 ++++++++++++++++---------- 1 file changed, 23 insertions(+), 15 deletions(-) diff --git a/drivers/net/phy/realtek/realtek_main.c b/drivers/net/phy/realtek/realtek_main.c index 896351022682..eed939ef4e18 100644 --- a/drivers/net/phy/realtek/realtek_main.c +++ b/drivers/net/phy/realtek/realtek_main.c @@ -194,8 +194,8 @@ MODULE_LICENSE("GPL"); struct rtl821x_priv { u16 phycr1; - u16 phycr2; bool has_phycr2; + bool disable_clk_out; struct clk *clk; /* rtl8211f */ u16 iner; @@ -266,15 +266,8 @@ static int rtl821x_probe(struct phy_device *phydev) priv->phycr1 |= RTL8211F_ALDPS_PLL_OFF | RTL8211F_ALDPS_ENABLE | RTL8211F_ALDPS_XTAL_OFF; priv->has_phycr2 = !(phy_id == RTL_8211FVD_PHYID); - if (priv->has_phycr2) { - ret = phy_read_paged(phydev, RTL8211F_PHYCR_PAGE, RTL8211F_PHYCR2); - if (ret < 0) - return ret; - - priv->phycr2 = ret & RTL8211F_CLKOUT_EN; - if (of_property_read_bool(dev->of_node, "realtek,clkout-disable")) - priv->phycr2 &= ~RTL8211F_CLKOUT_EN; - } + priv->disable_clk_out = of_property_read_bool(dev->of_node, + "realtek,clkout-disable"); phydev->priv = priv; @@ -654,6 +647,23 @@ static int rtl8211f_config_rgmii_delay(struct phy_device *phydev) return 0; } +static int rtl8211f_config_clk_out(struct phy_device *phydev) +{ + struct rtl821x_priv *priv = phydev->priv; + int ret; + + /* The value is preserved if the device tree property is absent */ + if (!priv->disable_clk_out) + return 0; + + ret = phy_modify_paged(phydev, RTL8211F_PHYCR_PAGE, + RTL8211F_PHYCR2, RTL8211F_CLKOUT_EN, 0); + if (ret) + return ret; + + return genphy_soft_reset(phydev); +} + static int rtl8211f_config_init(struct phy_device *phydev) { struct rtl821x_priv *priv = phydev->priv; @@ -682,16 +692,14 @@ static int rtl8211f_config_init(struct phy_device *phydev) if (ret) return ret; - ret = phy_modify_paged(phydev, RTL8211F_PHYCR_PAGE, - RTL8211F_PHYCR2, RTL8211F_CLKOUT_EN, - priv->phycr2); - if (ret < 0) { + ret = rtl8211f_config_clk_out(phydev); + if (ret) { dev_err(dev, "clkout configuration failed: %pe\n", ERR_PTR(ret)); return ret; } - return genphy_soft_reset(phydev); + return 0; } static int rtl821x_suspend(struct phy_device *phydev) From 910ac7bfb1af1ae4cd141ef80e03a6729213c189 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Tue, 18 Nov 2025 01:40:30 +0200 Subject: [PATCH 0908/1381] net: phy: realtek: eliminate has_phycr2 variable This variable is assigned in rtl821x_probe() and used in rtl8211f_config_init(), which is more complex than it needs to be. Simply testing the same condition from rtl821x_probe() in rtl8211f_config_init() yields the same result (the PHY driver ID is a runtime invariant), but with one temporary variable less. Signed-off-by: Vladimir Oltean Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20251117234033.345679-4-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- drivers/net/phy/realtek/realtek_main.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/drivers/net/phy/realtek/realtek_main.c b/drivers/net/phy/realtek/realtek_main.c index eed939ef4e18..da1db6499c38 100644 --- a/drivers/net/phy/realtek/realtek_main.c +++ b/drivers/net/phy/realtek/realtek_main.c @@ -194,7 +194,6 @@ MODULE_LICENSE("GPL"); struct rtl821x_priv { u16 phycr1; - bool has_phycr2; bool disable_clk_out; struct clk *clk; /* rtl8211f */ @@ -245,7 +244,6 @@ static int rtl821x_probe(struct phy_device *phydev) { struct device *dev = &phydev->mdio.dev; struct rtl821x_priv *priv; - u32 phy_id = phydev->drv->phy_id; int ret; priv = devm_kzalloc(dev, sizeof(*priv), GFP_KERNEL); @@ -265,7 +263,6 @@ static int rtl821x_probe(struct phy_device *phydev) if (of_property_read_bool(dev->of_node, "realtek,aldps-enable")) priv->phycr1 |= RTL8211F_ALDPS_PLL_OFF | RTL8211F_ALDPS_ENABLE | RTL8211F_ALDPS_XTAL_OFF; - priv->has_phycr2 = !(phy_id == RTL_8211FVD_PHYID); priv->disable_clk_out = of_property_read_bool(dev->of_node, "realtek,clkout-disable"); @@ -683,7 +680,8 @@ static int rtl8211f_config_init(struct phy_device *phydev) if (ret) return ret; - if (!priv->has_phycr2) + /* RTL8211FVD has no PHYCR2 register */ + if (phydev->drv->phy_id == RTL_8211FVD_PHYID) return 0; /* Disable PHY-mode EEE so LPI is passed to the MAC */ From e1a31c41bef678afe0d99b7f0dc3711a80c68447 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Tue, 18 Nov 2025 01:40:31 +0200 Subject: [PATCH 0909/1381] net: phy: realtek: allow CLKOUT to be disabled on RTL8211F(D)(I)-VD-CG Add CLKOUT disable support for RTL8211F(D)(I)-VD-CG. Like with other PHY variants, this feature might be requested by customers when the clock output is not used, in order to reduce electromagnetic interference (EMI). In the common driver, the CLKOUT configuration is done through PHYCR2. The RTL_8211FVD_PHYID is singled out as not having that register, and execution in rtl8211f_config_init() returns early after commit 2c67301584f2 ("net: phy: realtek: Avoid PHYCR2 access if PHYCR2 not present"). But actually CLKOUT is configured through a different register for this PHY. Instead of pretending this is PHYCR2 (which it is not), just add some code for modifying this register inside the rtl8211f_disable_clk_out() function, and move that outside the code portion that runs only if PHYCR2 exists. In practice this reorders the PHYCR2 writes to disable PHY-mode EEE and to disable the CLKOUT for the normal RTL8211F variants, but this should be perfectly fine. It was not noted that RTL8211F(D)(I)-VD-CG would need a genphy_soft_reset() call after disabling the CLKOUT. Despite that, we do it out of caution and for symmetry with the other RTL8211F models. Co-developed-by: Clark Wang Signed-off-by: Clark Wang Signed-off-by: Vladimir Oltean Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20251117234033.345679-5-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- drivers/net/phy/realtek/realtek_main.c | 31 ++++++++++++++++++-------- 1 file changed, 22 insertions(+), 9 deletions(-) diff --git a/drivers/net/phy/realtek/realtek_main.c b/drivers/net/phy/realtek/realtek_main.c index da1db6499c38..da30cdb522a3 100644 --- a/drivers/net/phy/realtek/realtek_main.c +++ b/drivers/net/phy/realtek/realtek_main.c @@ -90,6 +90,14 @@ #define RTL8211F_LEDCR_MASK GENMASK(4, 0) #define RTL8211F_LEDCR_SHIFT 5 +/* RTL8211F(D)(I)-VD-CG CLKOUT configuration is specified via magic values + * to undocumented register pages. The names here do not reflect the datasheet. + * Unlike other PHY models, CLKOUT configuration does not go through PHYCR2. + */ +#define RTL8211FVD_CLKOUT_PAGE 0xd05 +#define RTL8211FVD_CLKOUT_REG 0x11 +#define RTL8211FVD_CLKOUT_EN BIT(8) + /* RTL8211F RGMII configuration */ #define RTL8211F_RGMII_PAGE 0xd08 @@ -653,8 +661,13 @@ static int rtl8211f_config_clk_out(struct phy_device *phydev) if (!priv->disable_clk_out) return 0; - ret = phy_modify_paged(phydev, RTL8211F_PHYCR_PAGE, - RTL8211F_PHYCR2, RTL8211F_CLKOUT_EN, 0); + if (phydev->drv->phy_id == RTL_8211FVD_PHYID) + ret = phy_modify_paged(phydev, RTL8211FVD_CLKOUT_PAGE, + RTL8211FVD_CLKOUT_REG, + RTL8211FVD_CLKOUT_EN, 0); + else + ret = phy_modify_paged(phydev, RTL8211F_PHYCR_PAGE, + RTL8211F_PHYCR2, RTL8211F_CLKOUT_EN, 0); if (ret) return ret; @@ -680,6 +693,13 @@ static int rtl8211f_config_init(struct phy_device *phydev) if (ret) return ret; + ret = rtl8211f_config_clk_out(phydev); + if (ret) { + dev_err(dev, "clkout configuration failed: %pe\n", + ERR_PTR(ret)); + return ret; + } + /* RTL8211FVD has no PHYCR2 register */ if (phydev->drv->phy_id == RTL_8211FVD_PHYID) return 0; @@ -690,13 +710,6 @@ static int rtl8211f_config_init(struct phy_device *phydev) if (ret) return ret; - ret = rtl8211f_config_clk_out(phydev); - if (ret) { - dev_err(dev, "clkout configuration failed: %pe\n", - ERR_PTR(ret)); - return ret; - } - return 0; } From bb78b71faf60d11a15f07e3390fcfd31e5e523bb Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Tue, 18 Nov 2025 01:40:32 +0200 Subject: [PATCH 0910/1381] net: phy: realtek: eliminate priv->phycr1 variable Previous changes have replaced the machine-level priv->phycr2 with a high-level priv->disable_clk_out. This created a discrepancy with priv->phycr1 which is resolved here, for uniformity. One advantage of this new implementation is that we don't read priv->phycr1 in rtl821x_probe() if we're never going to modify it. We never test the positive return code from phy_modify_mmd_changed(), so we could just as well use phy_modify_mmd(). I took the ALDPS feature description from commit d90db36a9e74 ("net: phy: realtek: add dt property to enable ALDPS mode") and transformed it into a function comment - the feature is sufficiently non-obvious to deserve that. Signed-off-by: Vladimir Oltean Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20251117234033.345679-6-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- drivers/net/phy/realtek/realtek_main.c | 44 ++++++++++++++++---------- 1 file changed, 28 insertions(+), 16 deletions(-) diff --git a/drivers/net/phy/realtek/realtek_main.c b/drivers/net/phy/realtek/realtek_main.c index da30cdb522a3..2ecdcea53a11 100644 --- a/drivers/net/phy/realtek/realtek_main.c +++ b/drivers/net/phy/realtek/realtek_main.c @@ -201,7 +201,7 @@ MODULE_AUTHOR("Johnson Leung"); MODULE_LICENSE("GPL"); struct rtl821x_priv { - u16 phycr1; + bool enable_aldps; bool disable_clk_out; struct clk *clk; /* rtl8211f */ @@ -252,7 +252,6 @@ static int rtl821x_probe(struct phy_device *phydev) { struct device *dev = &phydev->mdio.dev; struct rtl821x_priv *priv; - int ret; priv = devm_kzalloc(dev, sizeof(*priv), GFP_KERNEL); if (!priv) @@ -263,14 +262,8 @@ static int rtl821x_probe(struct phy_device *phydev) return dev_err_probe(dev, PTR_ERR(priv->clk), "failed to get phy clock\n"); - ret = phy_read_paged(phydev, RTL8211F_PHYCR_PAGE, RTL8211F_PHYCR1); - if (ret < 0) - return ret; - - priv->phycr1 = ret & (RTL8211F_ALDPS_PLL_OFF | RTL8211F_ALDPS_ENABLE | RTL8211F_ALDPS_XTAL_OFF); - if (of_property_read_bool(dev->of_node, "realtek,aldps-enable")) - priv->phycr1 |= RTL8211F_ALDPS_PLL_OFF | RTL8211F_ALDPS_ENABLE | RTL8211F_ALDPS_XTAL_OFF; - + priv->enable_aldps = of_property_read_bool(dev->of_node, + "realtek,aldps-enable"); priv->disable_clk_out = of_property_read_bool(dev->of_node, "realtek,clkout-disable"); @@ -674,17 +667,36 @@ static int rtl8211f_config_clk_out(struct phy_device *phydev) return genphy_soft_reset(phydev); } -static int rtl8211f_config_init(struct phy_device *phydev) +/* Advance Link Down Power Saving (ALDPS) mode changes crystal/clock behaviour, + * which causes the RXC clock signal to stop for tens to hundreds of + * milliseconds. + * + * Some MACs need the RXC clock to support their internal RX logic, so ALDPS is + * only enabled based on an opt-in device tree property. + */ +static int rtl8211f_config_aldps(struct phy_device *phydev) { struct rtl821x_priv *priv = phydev->priv; + u16 mask = RTL8211F_ALDPS_PLL_OFF | + RTL8211F_ALDPS_ENABLE | + RTL8211F_ALDPS_XTAL_OFF; + + /* The value is preserved if the device tree property is absent */ + if (!priv->enable_aldps) + return 0; + + return phy_modify_paged(phydev, RTL8211F_PHYCR_PAGE, RTL8211F_PHYCR1, + mask, mask); +} + +static int rtl8211f_config_init(struct phy_device *phydev) +{ struct device *dev = &phydev->mdio.dev; int ret; - ret = phy_modify_paged_changed(phydev, RTL8211F_PHYCR_PAGE, RTL8211F_PHYCR1, - RTL8211F_ALDPS_PLL_OFF | RTL8211F_ALDPS_ENABLE | RTL8211F_ALDPS_XTAL_OFF, - priv->phycr1); - if (ret < 0) { - dev_err(dev, "aldps mode configuration failed: %pe\n", + ret = rtl8211f_config_aldps(phydev); + if (ret) { + dev_err(dev, "aldps mode configuration failed: %pe\n", ERR_PTR(ret)); return ret; } From 4465ae435ddc0162d5033a543658449d53d46d08 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Tue, 18 Nov 2025 01:40:33 +0200 Subject: [PATCH 0911/1381] net: phy: realtek: create rtl8211f_config_phy_eee() helper To simplify the rtl8211f_config_init() control flow and get rid of "early" returns for PHYs where the PHYCR2 register is absent, move the entire logic sub-block that deals with disabling PHY-mode EEE to a separate function. There, it is much more obvious what the early "return 0" skips, and it becomes more difficult to accidentally skip unintended stuff. Signed-off-by: Vladimir Oltean Link: https://patch.msgid.link/20251117234033.345679-7-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- drivers/net/phy/realtek/realtek_main.c | 23 ++++++++++++----------- 1 file changed, 12 insertions(+), 11 deletions(-) diff --git a/drivers/net/phy/realtek/realtek_main.c b/drivers/net/phy/realtek/realtek_main.c index 2ecdcea53a11..67ecf3d4af2b 100644 --- a/drivers/net/phy/realtek/realtek_main.c +++ b/drivers/net/phy/realtek/realtek_main.c @@ -689,6 +689,17 @@ static int rtl8211f_config_aldps(struct phy_device *phydev) mask, mask); } +static int rtl8211f_config_phy_eee(struct phy_device *phydev) +{ + /* RTL8211FVD has no PHYCR2 register */ + if (phydev->drv->phy_id == RTL_8211FVD_PHYID) + return 0; + + /* Disable PHY-mode EEE so LPI is passed to the MAC */ + return phy_modify_paged(phydev, RTL8211F_PHYCR_PAGE, RTL8211F_PHYCR2, + RTL8211F_PHYCR2_PHY_EEE_ENABLE, 0); +} + static int rtl8211f_config_init(struct phy_device *phydev) { struct device *dev = &phydev->mdio.dev; @@ -712,17 +723,7 @@ static int rtl8211f_config_init(struct phy_device *phydev) return ret; } - /* RTL8211FVD has no PHYCR2 register */ - if (phydev->drv->phy_id == RTL_8211FVD_PHYID) - return 0; - - /* Disable PHY-mode EEE so LPI is passed to the MAC */ - ret = phy_modify_paged(phydev, RTL8211F_PHYCR_PAGE, RTL8211F_PHYCR2, - RTL8211F_PHYCR2_PHY_EEE_ENABLE, 0); - if (ret) - return ret; - - return 0; + return rtl8211f_config_phy_eee(phydev); } static int rtl821x_suspend(struct phy_device *phydev) From 3505730d9042a8d26e89288ecab04e5d32794e4a Mon Sep 17 00:00:00 2001 From: Jason Xing Date: Tue, 18 Nov 2025 15:06:43 +0800 Subject: [PATCH 0912/1381] net: increase default NAPI_SKB_CACHE_SIZE to 128 After commit b61785852ed0 ("net: increase skb_defer_max default to 128") changed the value sysctl_skb_defer_max to avoid many calls to kick_defer_list_purge(), the same situation can be applied to NAPI_SKB_CACHE_SIZE that was proposed in 2016. It's a trade-off between using pre-allocated memory in skb_cache and saving more a bit heavy function calls in the softirq context. With this patch applied, we can have more skbs per-cpu to accelerate the sending path that needs to acquire new skbs. Suggested-by: Eric Dumazet Signed-off-by: Jason Xing Reviewed-by: Eric Dumazet Reviewed-by: Alexander Lobakin Link: https://patch.msgid.link/20251118070646.61344-2-kerneljasonxing@gmail.com Signed-off-by: Jakub Kicinski --- net/core/skbuff.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 9feea830a4db..e4abf0e56776 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -223,7 +223,7 @@ static void skb_under_panic(struct sk_buff *skb, unsigned int sz, void *addr) skb_panic(skb, sz, addr, __func__); } -#define NAPI_SKB_CACHE_SIZE 64 +#define NAPI_SKB_CACHE_SIZE 128 #define NAPI_SKB_CACHE_BULK 16 #define NAPI_SKB_CACHE_HALF (NAPI_SKB_CACHE_SIZE / 2) From 01d7385618992a9017198534870aec62f54f963a Mon Sep 17 00:00:00 2001 From: Jason Xing Date: Tue, 18 Nov 2025 15:06:44 +0800 Subject: [PATCH 0913/1381] net: increase default NAPI_SKB_CACHE_BULK to 32 The previous value 16 is a bit conservative, so adjust it along with NAPI_SKB_CACHE_SIZE, which can minimize triggering memory allocation in napi_skb_cache_get*(). Suggested-by: Eric Dumazet Signed-off-by: Jason Xing Reviewed-by: Eric Dumazet Reviewed-by: Alexander Lobakin Link: https://patch.msgid.link/20251118070646.61344-3-kerneljasonxing@gmail.com Signed-off-by: Jakub Kicinski --- net/core/skbuff.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/core/skbuff.c b/net/core/skbuff.c index e4abf0e56776..b6fe7ab85c4a 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -224,7 +224,7 @@ static void skb_under_panic(struct sk_buff *skb, unsigned int sz, void *addr) } #define NAPI_SKB_CACHE_SIZE 128 -#define NAPI_SKB_CACHE_BULK 16 +#define NAPI_SKB_CACHE_BULK 32 #define NAPI_SKB_CACHE_HALF (NAPI_SKB_CACHE_SIZE / 2) struct napi_alloc_cache { From 2d67b5c5c67f934c54a55e00ee291a8587c5b4fe Mon Sep 17 00:00:00 2001 From: Jason Xing Date: Tue, 18 Nov 2025 15:06:45 +0800 Subject: [PATCH 0914/1381] net: use NAPI_SKB_CACHE_FREE to keep 32 as default to do bulk free - Replace NAPI_SKB_CACHE_HALF with NAPI_SKB_CACHE_FREE - Only free 32 skbs in napi_skb_cache_put() Since the first patch adjusting NAPI_SKB_CACHE_SIZE to 128, the number of packets to be freed in the softirq was increased from 32 to 64. Considering a subsequent net_rx_action() calling napi_poll() a few times can easily consume the 64 available slots and we can afford keeping a higher value of sk_buffs in per-cpu storage, decrease NAPI_SKB_CACHE_FREE to 32 like before. So now the logic is 1) keeping 96 skbs, 2) freeing 32 skbs at one time. Suggested-by: Eric Dumazet Signed-off-by: Jason Xing Reviewed-by: Eric Dumazet Reviewed-by: Alexander Lobakin Link: https://patch.msgid.link/20251118070646.61344-4-kerneljasonxing@gmail.com Signed-off-by: Jakub Kicinski --- net/core/skbuff.c | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-) diff --git a/net/core/skbuff.c b/net/core/skbuff.c index b6fe7ab85c4a..d81ac78c32ff 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -225,7 +225,7 @@ static void skb_under_panic(struct sk_buff *skb, unsigned int sz, void *addr) #define NAPI_SKB_CACHE_SIZE 128 #define NAPI_SKB_CACHE_BULK 32 -#define NAPI_SKB_CACHE_HALF (NAPI_SKB_CACHE_SIZE / 2) +#define NAPI_SKB_CACHE_FREE 32 struct napi_alloc_cache { local_lock_t bh_lock; @@ -1445,7 +1445,6 @@ void __consume_stateless_skb(struct sk_buff *skb) static void napi_skb_cache_put(struct sk_buff *skb) { struct napi_alloc_cache *nc = this_cpu_ptr(&napi_alloc_cache); - u32 i; if (!kasan_mempool_poison_object(skb)) return; @@ -1454,13 +1453,16 @@ static void napi_skb_cache_put(struct sk_buff *skb) nc->skb_cache[nc->skb_count++] = skb; if (unlikely(nc->skb_count == NAPI_SKB_CACHE_SIZE)) { - for (i = NAPI_SKB_CACHE_HALF; i < NAPI_SKB_CACHE_SIZE; i++) + u32 i, remaining = NAPI_SKB_CACHE_SIZE - NAPI_SKB_CACHE_FREE; + + for (i = remaining; i < NAPI_SKB_CACHE_SIZE; i++) kasan_mempool_unpoison_object(nc->skb_cache[i], skbuff_cache_size); - kmem_cache_free_bulk(net_hotdata.skbuff_cache, NAPI_SKB_CACHE_HALF, - nc->skb_cache + NAPI_SKB_CACHE_HALF); - nc->skb_count = NAPI_SKB_CACHE_HALF; + kmem_cache_free_bulk(net_hotdata.skbuff_cache, + NAPI_SKB_CACHE_FREE, + nc->skb_cache + remaining); + nc->skb_count = remaining; } local_unlock_nested_bh(&napi_alloc_cache.bh_lock); } From 5d7fc63ab84182f75d5e73e299ce0e7c9c20c092 Mon Sep 17 00:00:00 2001 From: Jason Xing Date: Tue, 18 Nov 2025 15:06:46 +0800 Subject: [PATCH 0915/1381] net: prefetch the next skb in napi_skb_cache_get() After getting the current skb in napi_skb_cache_get(), the next skb in cache is highly likely to be used soon, so prefetch would be helpful. Suggested-by: Eric Dumazet Signed-off-by: Jason Xing Reviewed-by: Eric Dumazet Reviewed-by: Alexander Lobakin Link: https://patch.msgid.link/20251118070646.61344-5-kerneljasonxing@gmail.com Signed-off-by: Jakub Kicinski --- net/core/skbuff.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/core/skbuff.c b/net/core/skbuff.c index d81ac78c32ff..5a1d123e7ef7 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -299,6 +299,8 @@ static struct sk_buff *napi_skb_cache_get(bool alloc) } skb = nc->skb_cache[--nc->skb_count]; + if (nc->skb_count) + prefetch(nc->skb_cache[nc->skb_count - 1]); local_unlock_nested_bh(&napi_alloc_cache.bh_lock); kasan_mempool_unpoison_object(skb, skbuff_cache_size); From b6b03097f9826db72aeb3f751774c5e9edd9a5b3 Mon Sep 17 00:00:00 2001 From: Cosmin Ratiu Date: Sun, 16 Nov 2025 22:45:35 +0200 Subject: [PATCH 0916/1381] net/mlx5: Initialize events outside devlink lock Move event init/cleanup outside of mlx5_init_one() / mlx5_uninit_one() and into the mlx5_mdev_init() / mlx5_mdev_uninit() functions. By doing this, we avoid the events being reinitialized on devlink reload and, more importantly, the events->sw_nh notifier chain becomes available earlier in the init procedure, which will be used in subsequent patches. This makes sense because the events struct is pure software, independent of any HW details. Signed-off-by: Cosmin Ratiu Reviewed-by: Carolina Jubran Signed-off-by: Tariq Toukan Link: https://patch.msgid.link/1763325940-1231508-2-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- .../net/ethernet/mellanox/mlx5/core/main.c | 34 +++++++++++++------ 1 file changed, 24 insertions(+), 10 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c index c904696cbc3a..612fc4de9d3c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c @@ -1010,16 +1010,10 @@ static int mlx5_init_once(struct mlx5_core_dev *dev) goto err_irq_cleanup; } - err = mlx5_events_init(dev); - if (err) { - mlx5_core_err(dev, "failed to initialize events\n"); - goto err_eq_cleanup; - } - err = mlx5_fw_reset_init(dev); if (err) { mlx5_core_err(dev, "failed to initialize fw reset events\n"); - goto err_events_cleanup; + goto err_eq_cleanup; } mlx5_cq_debugfs_init(dev); @@ -1121,8 +1115,6 @@ static int mlx5_init_once(struct mlx5_core_dev *dev) mlx5_cleanup_reserved_gids(dev); mlx5_cq_debugfs_cleanup(dev); mlx5_fw_reset_cleanup(dev); -err_events_cleanup: - mlx5_events_cleanup(dev); err_eq_cleanup: mlx5_eq_table_cleanup(dev); err_irq_cleanup: @@ -1155,7 +1147,6 @@ static void mlx5_cleanup_once(struct mlx5_core_dev *dev) mlx5_cleanup_reserved_gids(dev); mlx5_cq_debugfs_cleanup(dev); mlx5_fw_reset_cleanup(dev); - mlx5_events_cleanup(dev); mlx5_eq_table_cleanup(dev); mlx5_irq_table_cleanup(dev); mlx5_devcom_unregister_device(dev->priv.devc); @@ -1833,6 +1824,24 @@ static int vhca_id_show(struct seq_file *file, void *priv) DEFINE_SHOW_ATTRIBUTE(vhca_id); +static int mlx5_notifiers_init(struct mlx5_core_dev *dev) +{ + int err; + + err = mlx5_events_init(dev); + if (err) { + mlx5_core_err(dev, "failed to initialize events\n"); + return err; + } + + return 0; +} + +static void mlx5_notifiers_cleanup(struct mlx5_core_dev *dev) +{ + mlx5_events_cleanup(dev); +} + int mlx5_mdev_init(struct mlx5_core_dev *dev, int profile_idx) { struct mlx5_priv *priv = &dev->priv; @@ -1888,6 +1897,10 @@ int mlx5_mdev_init(struct mlx5_core_dev *dev, int profile_idx) if (err) goto err_hca_caps; + err = mlx5_notifiers_init(dev); + if (err) + goto err_hca_caps; + /* The conjunction of sw_vhca_id with sw_owner_id will be a global * unique id per function which uses mlx5_core. * Those values are supplied to FW as part of the init HCA command to @@ -1930,6 +1943,7 @@ void mlx5_mdev_uninit(struct mlx5_core_dev *dev) if (priv->sw_vhca_id > 0) ida_free(&sw_vhca_ida, dev->priv.sw_vhca_id); + mlx5_notifiers_cleanup(dev); mlx5_hca_caps_free(dev); mlx5_adev_cleanup(dev); mlx5_pagealloc_cleanup(dev); From 3fee828789b1cf294a8fc83ad8a37f644c174fae Mon Sep 17 00:00:00 2001 From: Cosmin Ratiu Date: Sun, 16 Nov 2025 22:45:36 +0200 Subject: [PATCH 0917/1381] net/mlx5: Move the esw mode notifier chain outside the devlink lock The esw mode change notifier chain is initialized/cleaned up in mlx5_init_one() / mlx5_uninit_one() with the devlink lock held. Move the notifier head from the eswitch struct into mlx5_priv directly, and initialize it outside the critical section. This will allow notifier registration to happen earlier in the init procedure in subsequent patches. Signed-off-by: Cosmin Ratiu Reviewed-by: Carolina Jubran Signed-off-by: Tariq Toukan Link: https://patch.msgid.link/1763325940-1231508-3-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mellanox/mlx5/core/eswitch.c | 13 +++++++------ drivers/net/ethernet/mellanox/mlx5/core/eswitch.h | 7 ++++--- drivers/net/ethernet/mellanox/mlx5/core/main.c | 2 ++ .../net/ethernet/mellanox/mlx5/core/sf/devlink.c | 6 +++--- include/linux/mlx5/driver.h | 1 + 5 files changed, 17 insertions(+), 12 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c index 25af8bd7f077..3adf2b1cd26a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c @@ -1474,7 +1474,7 @@ static void mlx5_esw_mode_change_notify(struct mlx5_eswitch *esw, u16 mode) info.new_mode = mode; - blocking_notifier_call_chain(&esw->n_head, 0, &info); + blocking_notifier_call_chain(&esw->dev->priv.esw_n_head, 0, &info); } static int mlx5_esw_egress_acls_init(struct mlx5_core_dev *dev) @@ -2050,7 +2050,6 @@ int mlx5_eswitch_init(struct mlx5_core_dev *dev) esw->offloads.encap = DEVLINK_ESWITCH_ENCAP_MODE_BASIC; else esw->offloads.encap = DEVLINK_ESWITCH_ENCAP_MODE_NONE; - BLOCKING_INIT_NOTIFIER_HEAD(&esw->n_head); esw_info(dev, "Total vports %d, per vport: max uc(%d) max mc(%d)\n", @@ -2379,14 +2378,16 @@ bool mlx5_esw_multipath_prereq(struct mlx5_core_dev *dev0, dev1->priv.eswitch->mode == MLX5_ESWITCH_OFFLOADS); } -int mlx5_esw_event_notifier_register(struct mlx5_eswitch *esw, struct notifier_block *nb) +int mlx5_esw_event_notifier_register(struct mlx5_core_dev *dev, + struct notifier_block *nb) { - return blocking_notifier_chain_register(&esw->n_head, nb); + return blocking_notifier_chain_register(&dev->priv.esw_n_head, nb); } -void mlx5_esw_event_notifier_unregister(struct mlx5_eswitch *esw, struct notifier_block *nb) +void mlx5_esw_event_notifier_unregister(struct mlx5_core_dev *dev, + struct notifier_block *nb) { - blocking_notifier_chain_unregister(&esw->n_head, nb); + blocking_notifier_chain_unregister(&dev->priv.esw_n_head, nb); } /** diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h index beaec450a734..ad1073f7b79f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h @@ -403,7 +403,6 @@ struct mlx5_eswitch { struct { u32 large_group_num; } params; - struct blocking_notifier_head n_head; struct xarray paired; struct mlx5_devcom_comp_dev *devcom; u16 enabled_ipsec_vf_count; @@ -864,8 +863,10 @@ struct mlx5_esw_event_info { u16 new_mode; }; -int mlx5_esw_event_notifier_register(struct mlx5_eswitch *esw, struct notifier_block *n); -void mlx5_esw_event_notifier_unregister(struct mlx5_eswitch *esw, struct notifier_block *n); +int mlx5_esw_event_notifier_register(struct mlx5_core_dev *dev, + struct notifier_block *n); +void mlx5_esw_event_notifier_unregister(struct mlx5_core_dev *dev, + struct notifier_block *n); bool mlx5_esw_hold(struct mlx5_core_dev *dev); void mlx5_esw_release(struct mlx5_core_dev *dev); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c index 612fc4de9d3c..05f16f3e9c4f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c @@ -1834,6 +1834,8 @@ static int mlx5_notifiers_init(struct mlx5_core_dev *dev) return err; } + BLOCKING_INIT_NOTIFIER_HEAD(&dev->priv.esw_n_head); + return 0; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c b/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c index 3304f25cc805..2ece4983d33f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c @@ -481,7 +481,7 @@ int mlx5_sf_table_init(struct mlx5_core_dev *dev) xa_init(&table->function_ids); dev->priv.sf_table = table; table->esw_nb.notifier_call = mlx5_sf_esw_event; - err = mlx5_esw_event_notifier_register(dev->priv.eswitch, &table->esw_nb); + err = mlx5_esw_event_notifier_register(dev, &table->esw_nb); if (err) goto reg_err; @@ -496,7 +496,7 @@ int mlx5_sf_table_init(struct mlx5_core_dev *dev) return 0; vhca_err: - mlx5_esw_event_notifier_unregister(dev->priv.eswitch, &table->esw_nb); + mlx5_esw_event_notifier_unregister(dev, &table->esw_nb); reg_err: mutex_destroy(&table->sf_state_lock); kfree(table); @@ -513,7 +513,7 @@ void mlx5_sf_table_cleanup(struct mlx5_core_dev *dev) mlx5_blocking_notifier_unregister(dev, &table->mdev_nb); mlx5_vhca_event_notifier_unregister(table->dev, &table->vhca_nb); - mlx5_esw_event_notifier_unregister(dev->priv.eswitch, &table->esw_nb); + mlx5_esw_event_notifier_unregister(dev, &table->esw_nb); mutex_destroy(&table->sf_state_lock); WARN_ON(!xa_empty(&table->function_ids)); kfree(table); diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index 7aec53371cf0..9a4a5112a59e 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -599,6 +599,7 @@ struct mlx5_priv { struct mlx5_flow_steering *steering; struct mlx5_mpfs *mpfs; + struct blocking_notifier_head esw_n_head; struct mlx5_eswitch *eswitch; struct mlx5_core_sriov sriov; struct mlx5_lag *lag; From d3a356db853bc2dfb51034eacafd41aca7dd4c37 Mon Sep 17 00:00:00 2001 From: Cosmin Ratiu Date: Sun, 16 Nov 2025 22:45:37 +0200 Subject: [PATCH 0918/1381] net/mlx5: Move the vhca event notifier outside of the devlink lock The vhca event notifier consists of an atomic notifier for vhca state changes (used for SF events), multiple workqueues and a blocking notifier chain for delivering the vhca state change events for further processing. This patch moves the vhca notifier head outside of mlx5_init_one() / mlx5_uninit_one() and into the mlx5_mdev_init() / mlx5_mdev_uninit() functions. This allows called notifiers to grab the PF devlink lock which was previously impossible because it would create a circular lock dependency. mlx5_vhca_event_stop() is now called earlier in the cleanup phase and flushes the workqueues to ensure that after the call, there are no pending events. This simplifies the cleanup flow for vhca event consumers. Signed-off-by: Cosmin Ratiu Reviewed-by: Carolina Jubran Signed-off-by: Tariq Toukan Link: https://patch.msgid.link/1763325940-1231508-4-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- .../net/ethernet/mellanox/mlx5/core/main.c | 3 +- .../ethernet/mellanox/mlx5/core/sf/dev/dev.c | 1 - .../ethernet/mellanox/mlx5/core/sf/hw_table.c | 1 + .../mellanox/mlx5/core/sf/vhca_event.c | 69 +++++++------------ .../mellanox/mlx5/core/sf/vhca_event.h | 5 ++ include/linux/mlx5/driver.h | 4 +- 6 files changed, 35 insertions(+), 48 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c index 05f16f3e9c4f..097ba7ab90a4 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c @@ -1438,12 +1438,12 @@ static void mlx5_unload(struct mlx5_core_dev *dev) { mlx5_eswitch_disable(dev->priv.eswitch); mlx5_devlink_traps_unregister(priv_to_devlink(dev)); + mlx5_vhca_event_stop(dev); mlx5_sf_dev_table_destroy(dev); mlx5_sriov_detach(dev); mlx5_lag_remove_mdev(dev); mlx5_ec_cleanup(dev); mlx5_sf_hw_table_destroy(dev); - mlx5_vhca_event_stop(dev); mlx5_fs_core_cleanup(dev); mlx5_fpga_device_stop(dev); mlx5_rsc_dump_cleanup(dev); @@ -1835,6 +1835,7 @@ static int mlx5_notifiers_init(struct mlx5_core_dev *dev) } BLOCKING_INIT_NOTIFIER_HEAD(&dev->priv.esw_n_head); + mlx5_vhca_state_notifier_init(dev); return 0; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/dev.c b/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/dev.c index 99219ea52c4b..a68a8ee24dce 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/dev.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/dev.c @@ -381,7 +381,6 @@ void mlx5_sf_dev_table_destroy(struct mlx5_core_dev *dev) mlx5_sf_dev_destroy_active_works(table); mlx5_vhca_event_notifier_unregister(dev, &table->nb); - mlx5_vhca_event_work_queues_flush(dev); /* Now that event handler is not running, it is safe to destroy * the sf device without race. diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/hw_table.c b/drivers/net/ethernet/mellanox/mlx5/core/sf/hw_table.c index 1f613320fe07..a14b1aa5fb5a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/sf/hw_table.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/hw_table.c @@ -389,6 +389,7 @@ void mlx5_sf_hw_table_destroy(struct mlx5_core_dev *dev) return; mlx5_vhca_event_notifier_unregister(dev, &table->vhca_nb); + /* Dealloc SFs whose firmware event has been missed. */ mlx5_sf_hw_table_dealloc_all(table); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/vhca_event.c b/drivers/net/ethernet/mellanox/mlx5/core/sf/vhca_event.c index cda01ba441ae..b04cf6cf8956 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/sf/vhca_event.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/vhca_event.c @@ -9,15 +9,9 @@ #define CREATE_TRACE_POINTS #include "diag/vhca_tracepoint.h" -struct mlx5_vhca_state_notifier { - struct mlx5_core_dev *dev; - struct mlx5_nb nb; - struct blocking_notifier_head n_head; -}; - struct mlx5_vhca_event_work { struct work_struct work; - struct mlx5_vhca_state_notifier *notifier; + struct mlx5_core_dev *dev; struct mlx5_vhca_state_event event; }; @@ -95,16 +89,14 @@ mlx5_vhca_event_notify(struct mlx5_core_dev *dev, struct mlx5_vhca_state_event * mlx5_vhca_event_arm(dev, event->function_id); trace_mlx5_sf_vhca_event(dev, event); - blocking_notifier_call_chain(&dev->priv.vhca_state_notifier->n_head, 0, event); + blocking_notifier_call_chain(&dev->priv.vhca_state_n_head, 0, event); } static void mlx5_vhca_state_work_handler(struct work_struct *_work) { struct mlx5_vhca_event_work *work = container_of(_work, struct mlx5_vhca_event_work, work); - struct mlx5_vhca_state_notifier *notifier = work->notifier; - struct mlx5_core_dev *dev = notifier->dev; - mlx5_vhca_event_notify(dev, &work->event); + mlx5_vhca_event_notify(work->dev, &work->event); kfree(work); } @@ -116,8 +108,8 @@ void mlx5_vhca_events_work_enqueue(struct mlx5_core_dev *dev, int idx, struct wo static int mlx5_vhca_state_change_notifier(struct notifier_block *nb, unsigned long type, void *data) { - struct mlx5_vhca_state_notifier *notifier = - mlx5_nb_cof(nb, struct mlx5_vhca_state_notifier, nb); + struct mlx5_core_dev *dev = mlx5_nb_cof(nb, struct mlx5_core_dev, + priv.vhca_state_nb); struct mlx5_vhca_event_work *work; struct mlx5_eqe *eqe = data; int wq_idx; @@ -126,10 +118,10 @@ mlx5_vhca_state_change_notifier(struct notifier_block *nb, unsigned long type, v if (!work) return NOTIFY_DONE; INIT_WORK(&work->work, &mlx5_vhca_state_work_handler); - work->notifier = notifier; + work->dev = dev; work->event.function_id = be16_to_cpu(eqe->data.vhca_state.function_id); wq_idx = work->event.function_id % MLX5_DEV_MAX_WQS; - mlx5_vhca_events_work_enqueue(notifier->dev, wq_idx, &work->work); + mlx5_vhca_events_work_enqueue(dev, wq_idx, &work->work); return NOTIFY_OK; } @@ -145,9 +137,15 @@ void mlx5_vhca_state_cap_handle(struct mlx5_core_dev *dev, void *set_hca_cap) MLX5_SET(cmd_hca_cap, set_hca_cap, event_on_vhca_state_teardown_request, 1); } +void mlx5_vhca_state_notifier_init(struct mlx5_core_dev *dev) +{ + BLOCKING_INIT_NOTIFIER_HEAD(&dev->priv.vhca_state_n_head); + MLX5_NB_INIT(&dev->priv.vhca_state_nb, mlx5_vhca_state_change_notifier, + VHCA_STATE_CHANGE); +} + int mlx5_vhca_event_init(struct mlx5_core_dev *dev) { - struct mlx5_vhca_state_notifier *notifier; char wq_name[MLX5_CMD_WQ_MAX_NAME]; struct mlx5_vhca_events *events; int err, i; @@ -160,7 +158,6 @@ int mlx5_vhca_event_init(struct mlx5_core_dev *dev) return -ENOMEM; events->dev = dev; - dev->priv.vhca_events = events; for (i = 0; i < MLX5_DEV_MAX_WQS; i++) { snprintf(wq_name, MLX5_CMD_WQ_MAX_NAME, "mlx5_vhca_event%d", i); events->handler[i].wq = create_singlethread_workqueue(wq_name); @@ -169,20 +166,10 @@ int mlx5_vhca_event_init(struct mlx5_core_dev *dev) goto err_create_wq; } } + dev->priv.vhca_events = events; - notifier = kzalloc(sizeof(*notifier), GFP_KERNEL); - if (!notifier) { - err = -ENOMEM; - goto err_notifier; - } - - dev->priv.vhca_state_notifier = notifier; - notifier->dev = dev; - BLOCKING_INIT_NOTIFIER_HEAD(¬ifier->n_head); - MLX5_NB_INIT(¬ifier->nb, mlx5_vhca_state_change_notifier, VHCA_STATE_CHANGE); return 0; -err_notifier: err_create_wq: for (--i; i >= 0; i--) destroy_workqueue(events->handler[i].wq); @@ -211,8 +198,6 @@ void mlx5_vhca_event_cleanup(struct mlx5_core_dev *dev) if (!mlx5_vhca_event_supported(dev)) return; - kfree(dev->priv.vhca_state_notifier); - dev->priv.vhca_state_notifier = NULL; vhca_events = dev->priv.vhca_events; for (i = 0; i < MLX5_DEV_MAX_WQS; i++) destroy_workqueue(vhca_events->handler[i].wq); @@ -221,34 +206,30 @@ void mlx5_vhca_event_cleanup(struct mlx5_core_dev *dev) void mlx5_vhca_event_start(struct mlx5_core_dev *dev) { - struct mlx5_vhca_state_notifier *notifier; - - if (!dev->priv.vhca_state_notifier) + if (!mlx5_vhca_event_supported(dev)) return; - notifier = dev->priv.vhca_state_notifier; - mlx5_eq_notifier_register(dev, ¬ifier->nb); + mlx5_eq_notifier_register(dev, &dev->priv.vhca_state_nb); } void mlx5_vhca_event_stop(struct mlx5_core_dev *dev) { - struct mlx5_vhca_state_notifier *notifier; - - if (!dev->priv.vhca_state_notifier) + if (!mlx5_vhca_event_supported(dev)) return; - notifier = dev->priv.vhca_state_notifier; - mlx5_eq_notifier_unregister(dev, ¬ifier->nb); + mlx5_eq_notifier_unregister(dev, &dev->priv.vhca_state_nb); + + /* Flush workqueues of all pending events. */ + mlx5_vhca_event_work_queues_flush(dev); } int mlx5_vhca_event_notifier_register(struct mlx5_core_dev *dev, struct notifier_block *nb) { - if (!dev->priv.vhca_state_notifier) - return -EOPNOTSUPP; - return blocking_notifier_chain_register(&dev->priv.vhca_state_notifier->n_head, nb); + return blocking_notifier_chain_register(&dev->priv.vhca_state_n_head, + nb); } void mlx5_vhca_event_notifier_unregister(struct mlx5_core_dev *dev, struct notifier_block *nb) { - blocking_notifier_chain_unregister(&dev->priv.vhca_state_notifier->n_head, nb); + blocking_notifier_chain_unregister(&dev->priv.vhca_state_n_head, nb); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/vhca_event.h b/drivers/net/ethernet/mellanox/mlx5/core/sf/vhca_event.h index 1725ba64f8af..52790423874c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/sf/vhca_event.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/vhca_event.h @@ -18,6 +18,7 @@ static inline bool mlx5_vhca_event_supported(const struct mlx5_core_dev *dev) } void mlx5_vhca_state_cap_handle(struct mlx5_core_dev *dev, void *set_hca_cap); +void mlx5_vhca_state_notifier_init(struct mlx5_core_dev *dev); int mlx5_vhca_event_init(struct mlx5_core_dev *dev); void mlx5_vhca_event_cleanup(struct mlx5_core_dev *dev); void mlx5_vhca_event_start(struct mlx5_core_dev *dev); @@ -37,6 +38,10 @@ static inline void mlx5_vhca_state_cap_handle(struct mlx5_core_dev *dev, void *s { } +static inline void mlx5_vhca_state_notifier_init(struct mlx5_core_dev *dev) +{ +} + static inline int mlx5_vhca_event_init(struct mlx5_core_dev *dev) { return 0; diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index 9a4a5112a59e..88afb2788dc9 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -488,7 +488,6 @@ struct mlx5_devcom_dev; struct mlx5_fw_reset; struct mlx5_eq_table; struct mlx5_irq_table; -struct mlx5_vhca_state_notifier; struct mlx5_sf_dev_table; struct mlx5_sf_hw_table; struct mlx5_sf_table; @@ -615,7 +614,8 @@ struct mlx5_priv { struct mlx5_bfreg_data bfregs; struct mlx5_sq_bfreg bfreg; #ifdef CONFIG_MLX5_SF - struct mlx5_vhca_state_notifier *vhca_state_notifier; + struct mlx5_nb vhca_state_nb; + struct blocking_notifier_head vhca_state_n_head; struct mlx5_sf_dev_table *sf_dev_table; struct mlx5_core_dev *parent_mdev; #endif From e63c9c5f0a4802deea81a48c2c40d0af56153e8a Mon Sep 17 00:00:00 2001 From: Cosmin Ratiu Date: Sun, 16 Nov 2025 22:45:38 +0200 Subject: [PATCH 0919/1381] net/mlx5: Move the SF HW table notifier outside the devlink lock Move the SF HW table notifier registration/unregistration outside of mlx5_init_one() / mlx5_uninit_one() and into the mlx5_mdev_init() / mlx5_mdev_uninit() functions. This is only done for non-SFs, since SFs do not have a SF HW table themselves. Signed-off-by: Cosmin Ratiu Reviewed-by: Carolina Jubran Signed-off-by: Tariq Toukan Link: https://patch.msgid.link/1763325940-1231508-5-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- .../net/ethernet/mellanox/mlx5/core/main.c | 17 ++--- .../ethernet/mellanox/mlx5/core/sf/hw_table.c | 62 +++++++++++-------- .../net/ethernet/mellanox/mlx5/core/sf/sf.h | 9 ++- include/linux/mlx5/driver.h | 1 + 4 files changed, 54 insertions(+), 35 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c index 097ba7ab90a4..843ee452239f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c @@ -1377,12 +1377,6 @@ static int mlx5_load(struct mlx5_core_dev *dev) mlx5_vhca_event_start(dev); - err = mlx5_sf_hw_table_create(dev); - if (err) { - mlx5_core_err(dev, "sf table create failed %d\n", err); - goto err_vhca; - } - err = mlx5_ec_init(dev); if (err) { mlx5_core_err(dev, "Failed to init embedded CPU\n"); @@ -1411,8 +1405,6 @@ static int mlx5_load(struct mlx5_core_dev *dev) mlx5_lag_remove_mdev(dev); mlx5_ec_cleanup(dev); err_ec: - mlx5_sf_hw_table_destroy(dev); -err_vhca: mlx5_vhca_event_stop(dev); err_set_hca: mlx5_fs_core_cleanup(dev); @@ -1837,11 +1829,20 @@ static int mlx5_notifiers_init(struct mlx5_core_dev *dev) BLOCKING_INIT_NOTIFIER_HEAD(&dev->priv.esw_n_head); mlx5_vhca_state_notifier_init(dev); + err = mlx5_sf_hw_notifier_init(dev); + if (err) + goto err_sf_hw_notifier; + return 0; + +err_sf_hw_notifier: + mlx5_events_cleanup(dev); + return err; } static void mlx5_notifiers_cleanup(struct mlx5_core_dev *dev) { + mlx5_sf_hw_notifier_cleanup(dev); mlx5_events_cleanup(dev); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/hw_table.c b/drivers/net/ethernet/mellanox/mlx5/core/sf/hw_table.c index a14b1aa5fb5a..bd968f3b3855 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/sf/hw_table.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/hw_table.c @@ -30,9 +30,7 @@ enum mlx5_sf_hwc_index { }; struct mlx5_sf_hw_table { - struct mlx5_core_dev *dev; struct mutex table_lock; /* Serializes sf deletion and vhca state change handler. */ - struct notifier_block vhca_nb; struct mlx5_sf_hwc_table hwc[MLX5_SF_HWC_MAX]; }; @@ -71,14 +69,16 @@ mlx5_sf_table_fn_to_hwc(struct mlx5_sf_hw_table *table, u16 fn_id) return NULL; } -static int mlx5_sf_hw_table_id_alloc(struct mlx5_sf_hw_table *table, u32 controller, +static int mlx5_sf_hw_table_id_alloc(struct mlx5_core_dev *dev, + struct mlx5_sf_hw_table *table, + u32 controller, u32 usr_sfnum) { struct mlx5_sf_hwc_table *hwc; int free_idx = -1; int i; - hwc = mlx5_sf_controller_to_hwc(table->dev, controller); + hwc = mlx5_sf_controller_to_hwc(dev, controller); if (!hwc->sfs) return -ENOSPC; @@ -100,11 +100,13 @@ static int mlx5_sf_hw_table_id_alloc(struct mlx5_sf_hw_table *table, u32 control return free_idx; } -static void mlx5_sf_hw_table_id_free(struct mlx5_sf_hw_table *table, u32 controller, int id) +static void mlx5_sf_hw_table_id_free(struct mlx5_core_dev *dev, + struct mlx5_sf_hw_table *table, + u32 controller, int id) { struct mlx5_sf_hwc_table *hwc; - hwc = mlx5_sf_controller_to_hwc(table->dev, controller); + hwc = mlx5_sf_controller_to_hwc(dev, controller); hwc->sfs[id].allocated = false; hwc->sfs[id].pending_delete = false; } @@ -120,7 +122,7 @@ int mlx5_sf_hw_table_sf_alloc(struct mlx5_core_dev *dev, u32 controller, u32 usr return -EOPNOTSUPP; mutex_lock(&table->table_lock); - sw_id = mlx5_sf_hw_table_id_alloc(table, controller, usr_sfnum); + sw_id = mlx5_sf_hw_table_id_alloc(dev, table, controller, usr_sfnum); if (sw_id < 0) { err = sw_id; goto exist_err; @@ -151,7 +153,7 @@ int mlx5_sf_hw_table_sf_alloc(struct mlx5_core_dev *dev, u32 controller, u32 usr vhca_err: mlx5_cmd_dealloc_sf(dev, hw_fn_id); err: - mlx5_sf_hw_table_id_free(table, controller, sw_id); + mlx5_sf_hw_table_id_free(dev, table, controller, sw_id); exist_err: mutex_unlock(&table->table_lock); return err; @@ -165,7 +167,7 @@ void mlx5_sf_hw_table_sf_free(struct mlx5_core_dev *dev, u32 controller, u16 id) mutex_lock(&table->table_lock); hw_fn_id = mlx5_sf_sw_to_hw_id(dev, controller, id); mlx5_cmd_dealloc_sf(dev, hw_fn_id); - mlx5_sf_hw_table_id_free(table, controller, id); + mlx5_sf_hw_table_id_free(dev, table, controller, id); mutex_unlock(&table->table_lock); } @@ -216,10 +218,12 @@ static void mlx5_sf_hw_table_hwc_dealloc_all(struct mlx5_core_dev *dev, } } -static void mlx5_sf_hw_table_dealloc_all(struct mlx5_sf_hw_table *table) +static void mlx5_sf_hw_table_dealloc_all(struct mlx5_core_dev *dev, + struct mlx5_sf_hw_table *table) { - mlx5_sf_hw_table_hwc_dealloc_all(table->dev, &table->hwc[MLX5_SF_HWC_EXTERNAL]); - mlx5_sf_hw_table_hwc_dealloc_all(table->dev, &table->hwc[MLX5_SF_HWC_LOCAL]); + mlx5_sf_hw_table_hwc_dealloc_all(dev, + &table->hwc[MLX5_SF_HWC_EXTERNAL]); + mlx5_sf_hw_table_hwc_dealloc_all(dev, &table->hwc[MLX5_SF_HWC_LOCAL]); } static int mlx5_sf_hw_table_hwc_init(struct mlx5_sf_hwc_table *hwc, u16 max_fn, u16 base_id) @@ -301,7 +305,6 @@ int mlx5_sf_hw_table_init(struct mlx5_core_dev *dev) } mutex_init(&table->table_lock); - table->dev = dev; dev->priv.sf_hw_table = table; base_id = mlx5_sf_start_function_id(dev); @@ -338,19 +341,22 @@ void mlx5_sf_hw_table_cleanup(struct mlx5_core_dev *dev) mlx5_sf_hw_table_hwc_cleanup(&table->hwc[MLX5_SF_HWC_LOCAL]); mutex_destroy(&table->table_lock); kfree(table); + dev->priv.sf_hw_table = NULL; res_unregister: mlx5_sf_hw_table_res_unregister(dev); } static int mlx5_sf_hw_vhca_event(struct notifier_block *nb, unsigned long opcode, void *data) { - struct mlx5_sf_hw_table *table = container_of(nb, struct mlx5_sf_hw_table, vhca_nb); + struct mlx5_core_dev *dev = container_of(nb, struct mlx5_core_dev, + priv.sf_hw_table_vhca_nb); + struct mlx5_sf_hw_table *table = dev->priv.sf_hw_table; const struct mlx5_vhca_state_event *event = data; struct mlx5_sf_hwc_table *hwc; struct mlx5_sf_hw *sf_hw; u16 sw_id; - if (event->new_vhca_state != MLX5_VHCA_STATE_ALLOCATED) + if (!table || event->new_vhca_state != MLX5_VHCA_STATE_ALLOCATED) return 0; hwc = mlx5_sf_table_fn_to_hwc(table, event->function_id); @@ -365,20 +371,28 @@ static int mlx5_sf_hw_vhca_event(struct notifier_block *nb, unsigned long opcode * Hence recycle the sf hardware id for reuse. */ if (sf_hw->allocated && sf_hw->pending_delete) - mlx5_sf_hw_table_hwc_sf_free(table->dev, hwc, sw_id); + mlx5_sf_hw_table_hwc_sf_free(dev, hwc, sw_id); mutex_unlock(&table->table_lock); return 0; } -int mlx5_sf_hw_table_create(struct mlx5_core_dev *dev) +int mlx5_sf_hw_notifier_init(struct mlx5_core_dev *dev) { - struct mlx5_sf_hw_table *table = dev->priv.sf_hw_table; - - if (!table) + if (mlx5_core_is_sf(dev)) return 0; - table->vhca_nb.notifier_call = mlx5_sf_hw_vhca_event; - return mlx5_vhca_event_notifier_register(dev, &table->vhca_nb); + dev->priv.sf_hw_table_vhca_nb.notifier_call = mlx5_sf_hw_vhca_event; + return mlx5_vhca_event_notifier_register(dev, + &dev->priv.sf_hw_table_vhca_nb); +} + +void mlx5_sf_hw_notifier_cleanup(struct mlx5_core_dev *dev) +{ + if (mlx5_core_is_sf(dev)) + return; + + mlx5_vhca_event_notifier_unregister(dev, + &dev->priv.sf_hw_table_vhca_nb); } void mlx5_sf_hw_table_destroy(struct mlx5_core_dev *dev) @@ -388,10 +402,8 @@ void mlx5_sf_hw_table_destroy(struct mlx5_core_dev *dev) if (!table) return; - mlx5_vhca_event_notifier_unregister(dev, &table->vhca_nb); - /* Dealloc SFs whose firmware event has been missed. */ - mlx5_sf_hw_table_dealloc_all(table); + mlx5_sf_hw_table_dealloc_all(dev, table); } bool mlx5_sf_hw_table_supported(const struct mlx5_core_dev *dev) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/sf.h b/drivers/net/ethernet/mellanox/mlx5/core/sf/sf.h index 89559a37997a..3922dacffae8 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/sf/sf.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/sf.h @@ -12,7 +12,8 @@ int mlx5_sf_hw_table_init(struct mlx5_core_dev *dev); void mlx5_sf_hw_table_cleanup(struct mlx5_core_dev *dev); -int mlx5_sf_hw_table_create(struct mlx5_core_dev *dev); +int mlx5_sf_hw_notifier_init(struct mlx5_core_dev *dev); +void mlx5_sf_hw_notifier_cleanup(struct mlx5_core_dev *dev); void mlx5_sf_hw_table_destroy(struct mlx5_core_dev *dev); int mlx5_sf_table_init(struct mlx5_core_dev *dev); @@ -44,11 +45,15 @@ static inline void mlx5_sf_hw_table_cleanup(struct mlx5_core_dev *dev) { } -static inline int mlx5_sf_hw_table_create(struct mlx5_core_dev *dev) +static inline int mlx5_sf_hw_notifier_init(struct mlx5_core_dev *dev) { return 0; } +static inline void mlx5_sf_hw_notifier_cleanup(struct mlx5_core_dev *dev) +{ +} + static inline void mlx5_sf_hw_table_destroy(struct mlx5_core_dev *dev) { } diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index 88afb2788dc9..d6c5bcebdaca 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -620,6 +620,7 @@ struct mlx5_priv { struct mlx5_core_dev *parent_mdev; #endif #ifdef CONFIG_MLX5_SF_MANAGER + struct notifier_block sf_hw_table_vhca_nb; struct mlx5_sf_hw_table *sf_hw_table; struct mlx5_sf_table *sf_table; #endif From d4a0acbd94c2a93bf308a9fde9ab6719f5d98c7a Mon Sep 17 00:00:00 2001 From: Cosmin Ratiu Date: Sun, 16 Nov 2025 22:45:39 +0200 Subject: [PATCH 0920/1381] net/mlx5: Move the SF table notifiers outside the devlink lock Move the SF table notifiers registration/unregistration outside of mlx5_init_one() / mlx5_uninit_one() and into the mlx5_mdev_init() / mlx5_mdev_uninit() functions. This is only done for non-SFs, since SFs do not have a SF table themselves and thus don't need notifiers. Signed-off-by: Cosmin Ratiu Reviewed-by: Carolina Jubran Signed-off-by: Tariq Toukan Link: https://patch.msgid.link/1763325940-1231508-6-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- .../net/ethernet/mellanox/mlx5/core/main.c | 7 ++ .../ethernet/mellanox/mlx5/core/sf/devlink.c | 90 ++++++++++++------- .../net/ethernet/mellanox/mlx5/core/sf/sf.h | 11 +++ include/linux/mlx5/driver.h | 3 + 4 files changed, 78 insertions(+), 33 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c index 843ee452239f..0c3613ef39b1 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c @@ -1833,8 +1833,14 @@ static int mlx5_notifiers_init(struct mlx5_core_dev *dev) if (err) goto err_sf_hw_notifier; + err = mlx5_sf_notifiers_init(dev); + if (err) + goto err_sf_notifiers; + return 0; +err_sf_notifiers: + mlx5_sf_hw_notifier_cleanup(dev); err_sf_hw_notifier: mlx5_events_cleanup(dev); return err; @@ -1842,6 +1848,7 @@ static int mlx5_notifiers_init(struct mlx5_core_dev *dev) static void mlx5_notifiers_cleanup(struct mlx5_core_dev *dev) { + mlx5_sf_notifiers_cleanup(dev); mlx5_sf_hw_notifier_cleanup(dev); mlx5_events_cleanup(dev); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c b/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c index 2ece4983d33f..b82323b8449e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c @@ -31,9 +31,6 @@ struct mlx5_sf_table { struct mlx5_core_dev *dev; /* To refer from notifier context. */ struct xarray function_ids; /* function id based lookup. */ struct mutex sf_state_lock; /* Serializes sf state among user cmds & vhca event handler. */ - struct notifier_block esw_nb; - struct notifier_block vhca_nb; - struct notifier_block mdev_nb; }; static struct mlx5_sf * @@ -391,11 +388,16 @@ static bool mlx5_sf_state_update_check(const struct mlx5_sf *sf, u8 new_state) static int mlx5_sf_vhca_event(struct notifier_block *nb, unsigned long opcode, void *data) { - struct mlx5_sf_table *table = container_of(nb, struct mlx5_sf_table, vhca_nb); + struct mlx5_core_dev *dev = container_of(nb, struct mlx5_core_dev, + priv.sf_table_vhca_nb); + struct mlx5_sf_table *table = dev->priv.sf_table; const struct mlx5_vhca_state_event *event = data; bool update = false; struct mlx5_sf *sf; + if (!table) + return 0; + mutex_lock(&table->sf_state_lock); sf = mlx5_sf_lookup_by_function_id(table, event->function_id); if (!sf) @@ -407,7 +409,7 @@ static int mlx5_sf_vhca_event(struct notifier_block *nb, unsigned long opcode, v update = mlx5_sf_state_update_check(sf, event->new_vhca_state); if (update) sf->hw_state = event->new_vhca_state; - trace_mlx5_sf_update_state(table->dev, sf->port_index, sf->controller, + trace_mlx5_sf_update_state(dev, sf->port_index, sf->controller, sf->hw_fn_id, sf->hw_state); unlock: mutex_unlock(&table->sf_state_lock); @@ -425,12 +427,16 @@ static void mlx5_sf_del_all(struct mlx5_sf_table *table) static int mlx5_sf_esw_event(struct notifier_block *nb, unsigned long event, void *data) { - struct mlx5_sf_table *table = container_of(nb, struct mlx5_sf_table, esw_nb); + struct mlx5_core_dev *dev = container_of(nb, struct mlx5_core_dev, + priv.sf_table_esw_nb); const struct mlx5_esw_event_info *mode = data; + if (!dev->priv.sf_table) + return 0; + switch (mode->new_mode) { case MLX5_ESWITCH_LEGACY: - mlx5_sf_del_all(table); + mlx5_sf_del_all(dev->priv.sf_table); break; default: break; @@ -441,15 +447,16 @@ static int mlx5_sf_esw_event(struct notifier_block *nb, unsigned long event, voi static int mlx5_sf_mdev_event(struct notifier_block *nb, unsigned long event, void *data) { - struct mlx5_sf_table *table = container_of(nb, struct mlx5_sf_table, mdev_nb); + struct mlx5_core_dev *dev = container_of(nb, struct mlx5_core_dev, + priv.sf_table_mdev_nb); struct mlx5_sf_peer_devlink_event_ctx *event_ctx = data; + struct mlx5_sf_table *table = dev->priv.sf_table; int ret = NOTIFY_DONE; struct mlx5_sf *sf; - if (event != MLX5_DRIVER_EVENT_SF_PEER_DEVLINK) + if (!table || event != MLX5_DRIVER_EVENT_SF_PEER_DEVLINK) return NOTIFY_DONE; - mutex_lock(&table->sf_state_lock); sf = mlx5_sf_lookup_by_function_id(table, event_ctx->fn_id); if (!sf) @@ -464,10 +471,40 @@ static int mlx5_sf_mdev_event(struct notifier_block *nb, unsigned long event, vo return ret; } +int mlx5_sf_notifiers_init(struct mlx5_core_dev *dev) +{ + int err; + + if (mlx5_core_is_sf(dev)) + return 0; + + dev->priv.sf_table_esw_nb.notifier_call = mlx5_sf_esw_event; + err = mlx5_esw_event_notifier_register(dev, &dev->priv.sf_table_esw_nb); + if (err) + return err; + + dev->priv.sf_table_vhca_nb.notifier_call = mlx5_sf_vhca_event; + err = mlx5_vhca_event_notifier_register(dev, + &dev->priv.sf_table_vhca_nb); + if (err) + goto vhca_err; + + dev->priv.sf_table_mdev_nb.notifier_call = mlx5_sf_mdev_event; + err = mlx5_blocking_notifier_register(dev, &dev->priv.sf_table_mdev_nb); + if (err) + goto mdev_err; + + return 0; +mdev_err: + mlx5_vhca_event_notifier_unregister(dev, &dev->priv.sf_table_vhca_nb); +vhca_err: + mlx5_esw_event_notifier_unregister(dev, &dev->priv.sf_table_esw_nb); + return err; +} + int mlx5_sf_table_init(struct mlx5_core_dev *dev) { struct mlx5_sf_table *table; - int err; if (!mlx5_sf_table_supported(dev) || !mlx5_vhca_event_supported(dev)) return 0; @@ -480,28 +517,18 @@ int mlx5_sf_table_init(struct mlx5_core_dev *dev) table->dev = dev; xa_init(&table->function_ids); dev->priv.sf_table = table; - table->esw_nb.notifier_call = mlx5_sf_esw_event; - err = mlx5_esw_event_notifier_register(dev, &table->esw_nb); - if (err) - goto reg_err; - - table->vhca_nb.notifier_call = mlx5_sf_vhca_event; - err = mlx5_vhca_event_notifier_register(table->dev, &table->vhca_nb); - if (err) - goto vhca_err; - - table->mdev_nb.notifier_call = mlx5_sf_mdev_event; - mlx5_blocking_notifier_register(dev, &table->mdev_nb); return 0; +} -vhca_err: - mlx5_esw_event_notifier_unregister(dev, &table->esw_nb); -reg_err: - mutex_destroy(&table->sf_state_lock); - kfree(table); - dev->priv.sf_table = NULL; - return err; +void mlx5_sf_notifiers_cleanup(struct mlx5_core_dev *dev) +{ + if (mlx5_core_is_sf(dev)) + return; + + mlx5_blocking_notifier_unregister(dev, &dev->priv.sf_table_mdev_nb); + mlx5_vhca_event_notifier_unregister(dev, &dev->priv.sf_table_vhca_nb); + mlx5_esw_event_notifier_unregister(dev, &dev->priv.sf_table_esw_nb); } void mlx5_sf_table_cleanup(struct mlx5_core_dev *dev) @@ -511,9 +538,6 @@ void mlx5_sf_table_cleanup(struct mlx5_core_dev *dev) if (!table) return; - mlx5_blocking_notifier_unregister(dev, &table->mdev_nb); - mlx5_vhca_event_notifier_unregister(table->dev, &table->vhca_nb); - mlx5_esw_event_notifier_unregister(dev, &table->esw_nb); mutex_destroy(&table->sf_state_lock); WARN_ON(!xa_empty(&table->function_ids)); kfree(table); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/sf.h b/drivers/net/ethernet/mellanox/mlx5/core/sf/sf.h index 3922dacffae8..d8a934a0e968 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/sf/sf.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/sf.h @@ -16,7 +16,9 @@ int mlx5_sf_hw_notifier_init(struct mlx5_core_dev *dev); void mlx5_sf_hw_notifier_cleanup(struct mlx5_core_dev *dev); void mlx5_sf_hw_table_destroy(struct mlx5_core_dev *dev); +int mlx5_sf_notifiers_init(struct mlx5_core_dev *dev); int mlx5_sf_table_init(struct mlx5_core_dev *dev); +void mlx5_sf_notifiers_cleanup(struct mlx5_core_dev *dev); void mlx5_sf_table_cleanup(struct mlx5_core_dev *dev); bool mlx5_sf_table_empty(const struct mlx5_core_dev *dev); @@ -58,11 +60,20 @@ static inline void mlx5_sf_hw_table_destroy(struct mlx5_core_dev *dev) { } +static inline int mlx5_sf_notifiers_init(struct mlx5_core_dev *dev) +{ + return 0; +} + static inline int mlx5_sf_table_init(struct mlx5_core_dev *dev) { return 0; } +static inline void mlx5_sf_notifiers_cleanup(struct mlx5_core_dev *dev) +{ +} + static inline void mlx5_sf_table_cleanup(struct mlx5_core_dev *dev) { } diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index d6c5bcebdaca..6af62047a614 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -622,6 +622,9 @@ struct mlx5_priv { #ifdef CONFIG_MLX5_SF_MANAGER struct notifier_block sf_hw_table_vhca_nb; struct mlx5_sf_hw_table *sf_hw_table; + struct notifier_block sf_table_esw_nb; + struct notifier_block sf_table_vhca_nb; + struct notifier_block sf_table_mdev_nb; struct mlx5_sf_table *sf_table; #endif struct blocking_notifier_head lag_nh; From 64ad6470c882fcaecfa4a1da96ea94de7ca0dc80 Mon Sep 17 00:00:00 2001 From: Cosmin Ratiu Date: Sun, 16 Nov 2025 22:45:40 +0200 Subject: [PATCH 0921/1381] net/mlx5: Move SF dev table notifier registration outside the PF devlink lock This completes the previous patches by moving notifier registration for SF dev tables outside the devlink locked critical section in mlx5_init_one() / mlx5_uninit_one() and into the mlx5_mdev_init() / mlx5_mdev_uninit() functions. This is only done for non-SFs, since SFs do not have a SF HW table themselves. After this patch, notifiers can grab the PF devlink lock (soon to be necessary) without creating a locking cycle. Signed-off-by: Cosmin Ratiu Reviewed-by: Carolina Jubran Signed-off-by: Tariq Toukan Link: https://patch.msgid.link/1763325940-1231508-7-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- .../net/ethernet/mellanox/mlx5/core/main.c | 7 +++ .../ethernet/mellanox/mlx5/core/sf/dev/dev.c | 47 ++++++++++++------- .../ethernet/mellanox/mlx5/core/sf/dev/dev.h | 11 +++++ include/linux/mlx5/driver.h | 1 + 4 files changed, 49 insertions(+), 17 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c index 0c3613ef39b1..024339ce41f1 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c @@ -1837,8 +1837,14 @@ static int mlx5_notifiers_init(struct mlx5_core_dev *dev) if (err) goto err_sf_notifiers; + err = mlx5_sf_dev_notifier_init(dev); + if (err) + goto err_sf_dev_notifier; + return 0; +err_sf_dev_notifier: + mlx5_sf_notifiers_cleanup(dev); err_sf_notifiers: mlx5_sf_hw_notifier_cleanup(dev); err_sf_hw_notifier: @@ -1848,6 +1854,7 @@ static int mlx5_notifiers_init(struct mlx5_core_dev *dev) static void mlx5_notifiers_cleanup(struct mlx5_core_dev *dev) { + mlx5_sf_dev_notifier_cleanup(dev); mlx5_sf_notifiers_cleanup(dev); mlx5_sf_hw_notifier_cleanup(dev); mlx5_events_cleanup(dev); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/dev.c b/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/dev.c index a68a8ee24dce..f310bde3d11f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/dev.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/dev.c @@ -16,7 +16,6 @@ struct mlx5_sf_dev_table { struct xarray devices; phys_addr_t base_address; u64 sf_bar_length; - struct notifier_block nb; struct workqueue_struct *active_wq; struct work_struct work; u8 stop_active_wq:1; @@ -156,18 +155,23 @@ static void mlx5_sf_dev_del(struct mlx5_core_dev *dev, struct mlx5_sf_dev *sf_de static int mlx5_sf_dev_state_change_handler(struct notifier_block *nb, unsigned long event_code, void *data) { - struct mlx5_sf_dev_table *table = container_of(nb, struct mlx5_sf_dev_table, nb); + struct mlx5_core_dev *dev = container_of(nb, struct mlx5_core_dev, + priv.sf_dev_nb); + struct mlx5_sf_dev_table *table = dev->priv.sf_dev_table; const struct mlx5_vhca_state_event *event = data; struct mlx5_sf_dev *sf_dev; u16 max_functions; u16 sf_index; u16 base_id; - max_functions = mlx5_sf_max_functions(table->dev); + if (!table) + return 0; + + max_functions = mlx5_sf_max_functions(dev); if (!max_functions) return 0; - base_id = mlx5_sf_start_function_id(table->dev); + base_id = mlx5_sf_start_function_id(dev); if (event->function_id < base_id || event->function_id >= (base_id + max_functions)) return 0; @@ -177,19 +181,19 @@ mlx5_sf_dev_state_change_handler(struct notifier_block *nb, unsigned long event_ case MLX5_VHCA_STATE_INVALID: case MLX5_VHCA_STATE_ALLOCATED: if (sf_dev) - mlx5_sf_dev_del(table->dev, sf_dev, sf_index); + mlx5_sf_dev_del(dev, sf_dev, sf_index); break; case MLX5_VHCA_STATE_TEARDOWN_REQUEST: if (sf_dev) - mlx5_sf_dev_del(table->dev, sf_dev, sf_index); + mlx5_sf_dev_del(dev, sf_dev, sf_index); else - mlx5_core_err(table->dev, + mlx5_core_err(dev, "SF DEV: teardown state for invalid dev index=%d sfnum=0x%x\n", sf_index, event->sw_function_id); break; case MLX5_VHCA_STATE_ACTIVE: if (!sf_dev) - mlx5_sf_dev_add(table->dev, sf_index, event->function_id, + mlx5_sf_dev_add(dev, sf_index, event->function_id, event->sw_function_id); break; default: @@ -315,6 +319,15 @@ static void mlx5_sf_dev_destroy_active_works(struct mlx5_sf_dev_table *table) } } +int mlx5_sf_dev_notifier_init(struct mlx5_core_dev *dev) +{ + if (mlx5_core_is_sf(dev)) + return 0; + + dev->priv.sf_dev_nb.notifier_call = mlx5_sf_dev_state_change_handler; + return mlx5_vhca_event_notifier_register(dev, &dev->priv.sf_dev_nb); +} + void mlx5_sf_dev_table_create(struct mlx5_core_dev *dev) { struct mlx5_sf_dev_table *table; @@ -329,17 +342,12 @@ void mlx5_sf_dev_table_create(struct mlx5_core_dev *dev) goto table_err; } - table->nb.notifier_call = mlx5_sf_dev_state_change_handler; table->dev = dev; table->sf_bar_length = 1 << (MLX5_CAP_GEN(dev, log_min_sf_size) + 12); table->base_address = pci_resource_start(dev->pdev, 2); xa_init(&table->devices); dev->priv.sf_dev_table = table; - err = mlx5_vhca_event_notifier_register(dev, &table->nb); - if (err) - goto vhca_err; - err = mlx5_sf_dev_create_active_works(table); if (err) goto add_active_err; @@ -351,10 +359,8 @@ void mlx5_sf_dev_table_create(struct mlx5_core_dev *dev) arm_err: mlx5_sf_dev_destroy_active_works(table); -add_active_err: - mlx5_vhca_event_notifier_unregister(dev, &table->nb); mlx5_vhca_event_work_queues_flush(dev); -vhca_err: +add_active_err: kfree(table); dev->priv.sf_dev_table = NULL; table_err: @@ -372,6 +378,14 @@ static void mlx5_sf_dev_destroy_all(struct mlx5_sf_dev_table *table) } } +void mlx5_sf_dev_notifier_cleanup(struct mlx5_core_dev *dev) +{ + if (mlx5_core_is_sf(dev)) + return; + + mlx5_vhca_event_notifier_unregister(dev, &dev->priv.sf_dev_nb); +} + void mlx5_sf_dev_table_destroy(struct mlx5_core_dev *dev) { struct mlx5_sf_dev_table *table = dev->priv.sf_dev_table; @@ -380,7 +394,6 @@ void mlx5_sf_dev_table_destroy(struct mlx5_core_dev *dev) return; mlx5_sf_dev_destroy_active_works(table); - mlx5_vhca_event_notifier_unregister(dev, &table->nb); /* Now that event handler is not running, it is safe to destroy * the sf device without race. diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/dev.h b/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/dev.h index b99131e95e37..3ab0449c770c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/dev.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/dev.h @@ -25,7 +25,9 @@ struct mlx5_sf_peer_devlink_event_ctx { int err; }; +int mlx5_sf_dev_notifier_init(struct mlx5_core_dev *dev); void mlx5_sf_dev_table_create(struct mlx5_core_dev *dev); +void mlx5_sf_dev_notifier_cleanup(struct mlx5_core_dev *dev); void mlx5_sf_dev_table_destroy(struct mlx5_core_dev *dev); int mlx5_sf_driver_register(void); @@ -35,10 +37,19 @@ bool mlx5_sf_dev_allocated(const struct mlx5_core_dev *dev); #else +static inline int mlx5_sf_dev_notifier_init(struct mlx5_core_dev *dev) +{ + return 0; +} + static inline void mlx5_sf_dev_table_create(struct mlx5_core_dev *dev) { } +static inline void mlx5_sf_dev_notifier_cleanup(struct mlx5_core_dev *dev) +{ +} + static inline void mlx5_sf_dev_table_destroy(struct mlx5_core_dev *dev) { } diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index 6af62047a614..1c54aa6f74fb 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -616,6 +616,7 @@ struct mlx5_priv { #ifdef CONFIG_MLX5_SF struct mlx5_nb vhca_state_nb; struct blocking_notifier_head vhca_state_n_head; + struct notifier_block sf_dev_nb; struct mlx5_sf_dev_table *sf_dev_table; struct mlx5_core_dev *parent_mdev; #endif From a77f0ad44fde89874654ba48f461209fb0382107 Mon Sep 17 00:00:00 2001 From: Pagadala Yesu Anjaneyulu Date: Wed, 12 Nov 2025 11:10:23 +0200 Subject: [PATCH 0922/1381] wifi: cfg80211: Add support for 6GHz AP role not relevant AP type Add IEEE80211_6GHZ_CTRL_REG_AP_ROLE_NOT_RELEVANT and map it to IEEE80211_REG_LPI_AP for safe regulatory compliance when AP role classification is not applicable. Use LPI as safe fallback to prevent power limit violations. Signed-off-by: Pagadala Yesu Anjaneyulu Reviewed-by: Johannes Berg Signed-off-by: Miri Korenblit Link: https://patch.msgid.link/20251112110828.856283677cc7.I36138a34847c3b4e680974bf347dde844448f3bc@changeid Signed-off-by: Johannes Berg --- include/linux/ieee80211-he.h | 1 + include/net/cfg80211.h | 1 + 2 files changed, 2 insertions(+) diff --git a/include/linux/ieee80211-he.h b/include/linux/ieee80211-he.h index 904d50db5bb8..a08c446fbb04 100644 --- a/include/linux/ieee80211-he.h +++ b/include/linux/ieee80211-he.h @@ -548,6 +548,7 @@ static inline bool ieee80211_he_capa_size_ok(const u8 *data, u8 len) #define IEEE80211_6GHZ_CTRL_REG_VLP_AP 2 #define IEEE80211_6GHZ_CTRL_REG_INDOOR_LPI_AP 3 #define IEEE80211_6GHZ_CTRL_REG_INDOOR_SP_AP_OLD 4 +#define IEEE80211_6GHZ_CTRL_REG_AP_ROLE_NOT_RELEVANT 7 #define IEEE80211_6GHZ_CTRL_REG_INDOOR_SP_AP 8 /** diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index 625cb2c78361..3d3ed1932262 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -10148,6 +10148,7 @@ cfg80211_6ghz_power_type(u8 control, u32 client_flags) switch (u8_get_bits(control, IEEE80211_HE_6GHZ_OPER_CTRL_REG_INFO)) { case IEEE80211_6GHZ_CTRL_REG_LPI_AP: case IEEE80211_6GHZ_CTRL_REG_INDOOR_LPI_AP: + case IEEE80211_6GHZ_CTRL_REG_AP_ROLE_NOT_RELEVANT: return IEEE80211_REG_LPI_AP; case IEEE80211_6GHZ_CTRL_REG_SP_AP: case IEEE80211_6GHZ_CTRL_REG_INDOOR_SP_AP_OLD: From 5d5c121c4fb6b96c0b1ee9b1862047b09e0830aa Mon Sep 17 00:00:00 2001 From: Marco Crivellari Date: Thu, 13 Nov 2025 17:00:35 +0100 Subject: [PATCH 0923/1381] wifi: qtnfmac: add WQ_PERCPU to alloc_workqueue users MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Currently if a user enqueues a work item using schedule_delayed_work() the used wq is "system_wq" (per-cpu wq) while queue_delayed_work() use WORK_CPU_UNBOUND (used when a cpu is not specified). The same applies to schedule_work() that is using system_wq and queue_work(), that makes use again of WORK_CPU_UNBOUND. This lack of consistency cannot be addressed without refactoring the API. For more details see the Link tag below. alloc_workqueue() treats all queues as per-CPU by default, while unbound workqueues must opt-in via WQ_UNBOUND. This default is suboptimal: most workloads benefit from unbound queues, allowing the scheduler to place worker threads where they’re needed and reducing noise when CPUs are isolated. This continues the effort to refactor workqueue APIs, which began with the introduction of new workqueues and a new alloc_workqueue flag in: commit 128ea9f6ccfb ("workqueue: Add system_percpu_wq and system_dfl_wq") commit 930c2ea566af ("workqueue: Add new WQ_PERCPU flag") This change adds a new WQ_PERCPU flag to explicitly request alloc_workqueue() to be per-cpu when WQ_UNBOUND has not been specified. With the introduction of the WQ_PERCPU flag (equivalent to !WQ_UNBOUND), any alloc_workqueue() caller that doesn’t explicitly specify WQ_UNBOUND must now use WQ_PERCPU. Once migration is complete, WQ_UNBOUND can be removed and unbound will become the implicit default. Suggested-by: Tejun Heo Signed-off-by: Marco Crivellari Link: https://lore.kernel.org/all/20250221112003.1dSuoGyc@linutronix.de/ Link: https://patch.msgid.link/20251113160035.376524-1-marco.crivellari@suse.com Signed-off-by: Johannes Berg --- drivers/net/wireless/quantenna/qtnfmac/core.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/quantenna/qtnfmac/core.c b/drivers/net/wireless/quantenna/qtnfmac/core.c index 825b05dd3271..38af6cdc2843 100644 --- a/drivers/net/wireless/quantenna/qtnfmac/core.c +++ b/drivers/net/wireless/quantenna/qtnfmac/core.c @@ -714,7 +714,8 @@ int qtnf_core_attach(struct qtnf_bus *bus) goto error; } - bus->hprio_workqueue = alloc_workqueue("QTNF_HPRI", WQ_HIGHPRI, 0); + bus->hprio_workqueue = alloc_workqueue("QTNF_HPRI", + WQ_HIGHPRI | WQ_PERCPU, 0); if (!bus->hprio_workqueue) { pr_err("failed to alloc high prio workqueue\n"); ret = -ENOMEM; From 2112519bdf05685c603c62bf633ab1c8f0edd3c6 Mon Sep 17 00:00:00 2001 From: Marco Crivellari Date: Thu, 13 Nov 2025 17:08:25 +0100 Subject: [PATCH 0924/1381] wifi: wfx: add WQ_PERCPU to alloc_workqueue users MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Currently if a user enqueues a work item using schedule_delayed_work() the used wq is "system_wq" (per-cpu wq) while queue_delayed_work() use WORK_CPU_UNBOUND (used when a cpu is not specified). The same applies to schedule_work() that is using system_wq and queue_work(), that makes use again of WORK_CPU_UNBOUND. This lack of consistency cannot be addressed without refactoring the API. For more details see the Link tag below. alloc_workqueue() treats all queues as per-CPU by default, while unbound workqueues must opt-in via WQ_UNBOUND. This default is suboptimal: most workloads benefit from unbound queues, allowing the scheduler to place worker threads where they’re needed and reducing noise when CPUs are isolated. This continues the effort to refactor workqueue APIs, which began with the introduction of new workqueues and a new alloc_workqueue flag in: commit 128ea9f6ccfb ("workqueue: Add system_percpu_wq and system_dfl_wq") commit 930c2ea566af ("workqueue: Add new WQ_PERCPU flag") This change adds a new WQ_PERCPU flag to explicitly request alloc_workqueue() to be per-cpu when WQ_UNBOUND has not been specified. With the introduction of the WQ_PERCPU flag (equivalent to !WQ_UNBOUND), any alloc_workqueue() caller that doesn’t explicitly specify WQ_UNBOUND must now use WQ_PERCPU. Once migration is complete, WQ_UNBOUND can be removed and unbound will become the implicit default. Suggested-by: Tejun Heo Signed-off-by: Marco Crivellari Link: https://lore.kernel.org/all/20250221112003.1dSuoGyc@linutronix.de/ Reviewed-by: Jérôme Pouiller Link: https://patch.msgid.link/20251113160825.383883-1-marco.crivellari@suse.com Signed-off-by: Johannes Berg --- drivers/net/wireless/silabs/wfx/main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/wireless/silabs/wfx/main.c b/drivers/net/wireless/silabs/wfx/main.c index a61128debbad..dda36e41eed1 100644 --- a/drivers/net/wireless/silabs/wfx/main.c +++ b/drivers/net/wireless/silabs/wfx/main.c @@ -364,7 +364,7 @@ int wfx_probe(struct wfx_dev *wdev) wdev->pdata.gpio_wakeup = NULL; wdev->poll_irq = true; - wdev->bh_wq = alloc_workqueue("wfx_bh_wq", WQ_HIGHPRI, 0); + wdev->bh_wq = alloc_workqueue("wfx_bh_wq", WQ_HIGHPRI | WQ_PERCPU, 0); if (!wdev->bh_wq) return -ENOMEM; From e4582bdf5132ce80e7432d8143f2c321d258e60f Mon Sep 17 00:00:00 2001 From: Marco Crivellari Date: Thu, 13 Nov 2025 17:20:31 +0100 Subject: [PATCH 0925/1381] wifi: cw1200: add WQ_PERCPU to alloc_workqueue users MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Currently if a user enqueues a work item using schedule_delayed_work() the used wq is "system_wq" (per-cpu wq) while queue_delayed_work() use WORK_CPU_UNBOUND (used when a cpu is not specified). The same applies to schedule_work() that is using system_wq and queue_work(), that makes use again of WORK_CPU_UNBOUND. This lack of consistency cannot be addressed without refactoring the API. For more details see the Link tag below. alloc_workqueue() treats all queues as per-CPU by default, while unbound workqueues must opt-in via WQ_UNBOUND. This default is suboptimal: most workloads benefit from unbound queues, allowing the scheduler to place worker threads where they’re needed and reducing noise when CPUs are isolated. This continues the effort to refactor workqueue APIs, which began with the introduction of new workqueues and a new alloc_workqueue flag in: commit 128ea9f6ccfb ("workqueue: Add system_percpu_wq and system_dfl_wq") commit 930c2ea566af ("workqueue: Add new WQ_PERCPU flag") This change adds a new WQ_PERCPU flag to explicitly request alloc_workqueue() to be per-cpu when WQ_UNBOUND has not been specified. With the introduction of the WQ_PERCPU flag (equivalent to !WQ_UNBOUND), any alloc_workqueue() caller that doesn’t explicitly specify WQ_UNBOUND must now use WQ_PERCPU. Once migration is complete, WQ_UNBOUND can be removed and unbound will become the implicit default. Suggested-by: Tejun Heo Signed-off-by: Marco Crivellari Link: https://lore.kernel.org/all/20250221112003.1dSuoGyc@linutronix.de/ Link: https://patch.msgid.link/20251113162032.394804-3-marco.crivellari@suse.com Signed-off-by: Johannes Berg --- drivers/net/wireless/st/cw1200/bh.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/net/wireless/st/cw1200/bh.c b/drivers/net/wireless/st/cw1200/bh.c index 37232ee22037..b034bab4b489 100644 --- a/drivers/net/wireless/st/cw1200/bh.c +++ b/drivers/net/wireless/st/cw1200/bh.c @@ -54,8 +54,9 @@ int cw1200_register_bh(struct cw1200_common *priv) int err = 0; /* Realtime workqueue */ priv->bh_workqueue = alloc_workqueue("cw1200_bh", - WQ_MEM_RECLAIM | WQ_HIGHPRI - | WQ_CPU_INTENSIVE, 1); + WQ_MEM_RECLAIM | WQ_HIGHPRI | + WQ_CPU_INTENSIVE | WQ_PERCPU, + 1); if (!priv->bh_workqueue) return -ENOMEM; From fec4d9d6ffd297b1ce8f19a8f723f5b0aaac5b3e Mon Sep 17 00:00:00 2001 From: Marco Crivellari Date: Thu, 13 Nov 2025 17:20:32 +0100 Subject: [PATCH 0926/1381] wifi: cfg80211: replace use of system_unbound_wq with system_dfl_wq Currently if a user enqueues a work item using schedule_delayed_work() the used wq is "system_wq" (per-cpu wq) while queue_delayed_work() use WORK_CPU_UNBOUND (used when a cpu is not specified). The same applies to schedule_work() that is using system_wq and queue_work(), that makes use again of WORK_CPU_UNBOUND. This lack of consistency cannot be addressed without refactoring the API. For more details see the Link tag below. This continues the effort to refactor workqueue APIs, which began with the introduction of new workqueues and a new alloc_workqueue flag in: commit 128ea9f6ccfb ("workqueue: Add system_percpu_wq and system_dfl_wq") commit 930c2ea566af ("workqueue: Add new WQ_PERCPU flag") Switch to using system_dfl_wq because system_unbound_wq is going away as part of a workqueue restructuring. Suggested-by: Tejun Heo Signed-off-by: Marco Crivellari Link: https://lore.kernel.org/all/20250221112003.1dSuoGyc@linutronix.de/ Link: https://patch.msgid.link/20251113162032.394804-4-marco.crivellari@suse.com Signed-off-by: Johannes Berg --- net/wireless/core.c | 4 ++-- net/wireless/sysfs.c | 2 +- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/net/wireless/core.c b/net/wireless/core.c index aa78dbaaf093..7be25b9ae36e 100644 --- a/net/wireless/core.c +++ b/net/wireless/core.c @@ -431,7 +431,7 @@ static void cfg80211_wiphy_work(struct work_struct *work) if (wk) { list_del_init(&wk->entry); if (!list_empty(&rdev->wiphy_work_list)) - queue_work(system_unbound_wq, work); + queue_work(system_dfl_wq, work); spin_unlock_irq(&rdev->wiphy_work_lock); trace_wiphy_work_run(&rdev->wiphy, wk); @@ -1713,7 +1713,7 @@ void wiphy_work_queue(struct wiphy *wiphy, struct wiphy_work *work) list_add_tail(&work->entry, &rdev->wiphy_work_list); spin_unlock_irqrestore(&rdev->wiphy_work_lock, flags); - queue_work(system_unbound_wq, &rdev->wiphy_work); + queue_work(system_dfl_wq, &rdev->wiphy_work); } EXPORT_SYMBOL_GPL(wiphy_work_queue); diff --git a/net/wireless/sysfs.c b/net/wireless/sysfs.c index 62f26618f674..8d142856e385 100644 --- a/net/wireless/sysfs.c +++ b/net/wireless/sysfs.c @@ -137,7 +137,7 @@ static int wiphy_resume(struct device *dev) if (rdev->wiphy.registered && rdev->ops->resume) ret = rdev_resume(rdev); rdev->suspended = false; - queue_work(system_unbound_wq, &rdev->wiphy_work); + queue_work(system_dfl_wq, &rdev->wiphy_work); wiphy_unlock(&rdev->wiphy); if (ret) From 799e98708f748253cbeda75d633da81324d7e7db Mon Sep 17 00:00:00 2001 From: Marco Crivellari Date: Thu, 20 Nov 2025 10:45:24 +0100 Subject: [PATCH 0927/1381] wifi: ipw2x00: replace use of system_wq with system_percpu_wq This patch continues the effort to refactor workqueue APIs, which has begun with the changes introducing new workqueues and a new alloc_workqueue flag: commit 128ea9f6ccfb ("workqueue: Add system_percpu_wq and system_dfl_wq") commit 930c2ea566af ("workqueue: Add new WQ_PERCPU flag") The point of the refactoring is to eventually alter the default behavior of workqueues to become unbound by default so that their workload placement is optimized by the scheduler. Before that to happen after a careful review and conversion of each individual case, workqueue users must be converted to the better named new workqueues with no intended behaviour changes: system_wq -> system_percpu_wq system_unbound_wq -> systemd_dfl_wq This way the old obsolete workqueues (system_wq, system_unbound_wq) can be removed in the future. Suggested-by: Tejun Heo Signed-off-by: Marco Crivellari Link: https://patch.msgid.link/20251120094524.45264-1-marco.crivellari@suse.com Signed-off-by: Johannes Berg --- drivers/net/wireless/intel/ipw2x00/ipw2100.c | 6 +++--- drivers/net/wireless/intel/ipw2x00/ipw2200.c | 2 +- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/net/wireless/intel/ipw2x00/ipw2100.c b/drivers/net/wireless/intel/ipw2x00/ipw2100.c index 215814861cbd..c7c5bc0f1650 100644 --- a/drivers/net/wireless/intel/ipw2x00/ipw2100.c +++ b/drivers/net/wireless/intel/ipw2x00/ipw2100.c @@ -2143,7 +2143,7 @@ static void isr_indicate_rf_kill(struct ipw2100_priv *priv, u32 status) /* Make sure the RF Kill check timer is running */ priv->stop_rf_kill = 0; - mod_delayed_work(system_wq, &priv->rf_kill, round_jiffies_relative(HZ)); + mod_delayed_work(system_percpu_wq, &priv->rf_kill, round_jiffies_relative(HZ)); } static void ipw2100_scan_event(struct work_struct *work) @@ -2170,7 +2170,7 @@ static void isr_scan_complete(struct ipw2100_priv *priv, u32 status) round_jiffies_relative(msecs_to_jiffies(4000))); } else { priv->user_requested_scan = 0; - mod_delayed_work(system_wq, &priv->scan_event, 0); + mod_delayed_work(system_percpu_wq, &priv->scan_event, 0); } } @@ -4252,7 +4252,7 @@ static int ipw_radio_kill_sw(struct ipw2100_priv *priv, int disable_radio) "disabled by HW switch\n"); /* Make sure the RF_KILL check timer is running */ priv->stop_rf_kill = 0; - mod_delayed_work(system_wq, &priv->rf_kill, + mod_delayed_work(system_percpu_wq, &priv->rf_kill, round_jiffies_relative(HZ)); } else schedule_reset(priv); diff --git a/drivers/net/wireless/intel/ipw2x00/ipw2200.c b/drivers/net/wireless/intel/ipw2x00/ipw2200.c index 24a5624ef207..09035a77e775 100644 --- a/drivers/net/wireless/intel/ipw2x00/ipw2200.c +++ b/drivers/net/wireless/intel/ipw2x00/ipw2200.c @@ -4415,7 +4415,7 @@ static void handle_scan_event(struct ipw_priv *priv) round_jiffies_relative(msecs_to_jiffies(4000))); } else { priv->user_requested_scan = 0; - mod_delayed_work(system_wq, &priv->scan_event, 0); + mod_delayed_work(system_percpu_wq, &priv->scan_event, 0); } } From 353cda30d30e5dc7cacf8de5d2546724708ae3bb Mon Sep 17 00:00:00 2001 From: Chien Wong Date: Thu, 13 Nov 2025 22:05:07 +0800 Subject: [PATCH 0928/1381] wifi: mac80211: fix CMAC functions not handling errors The called hash functions could fail thus we should check return values. Fixes: 26717828b75d ("mac80211: aes-cmac: switch to shash CMAC driver") Signed-off-by: Chien Wong Link: https://patch.msgid.link/20251113140511.48658-2-m@xv97.com Signed-off-by: Johannes Berg --- net/mac80211/aes_cmac.c | 63 +++++++++++++++++++++++++++++------------ net/mac80211/aes_cmac.h | 8 +++--- net/mac80211/wpa.c | 20 +++++++------ 3 files changed, 61 insertions(+), 30 deletions(-) diff --git a/net/mac80211/aes_cmac.c b/net/mac80211/aes_cmac.c index 48c04f89de20..65989c7dfc68 100644 --- a/net/mac80211/aes_cmac.c +++ b/net/mac80211/aes_cmac.c @@ -22,50 +22,77 @@ static const u8 zero[CMAC_TLEN_256]; -void ieee80211_aes_cmac(struct crypto_shash *tfm, const u8 *aad, - const u8 *data, size_t data_len, u8 *mic) +int ieee80211_aes_cmac(struct crypto_shash *tfm, const u8 *aad, + const u8 *data, size_t data_len, u8 *mic) { + int err; SHASH_DESC_ON_STACK(desc, tfm); u8 out[AES_BLOCK_SIZE]; const __le16 *fc; desc->tfm = tfm; - crypto_shash_init(desc); - crypto_shash_update(desc, aad, AAD_LEN); + err = crypto_shash_init(desc); + if (err) + return err; + err = crypto_shash_update(desc, aad, AAD_LEN); + if (err) + return err; fc = (const __le16 *)aad; if (ieee80211_is_beacon(*fc)) { /* mask Timestamp field to zero */ - crypto_shash_update(desc, zero, 8); - crypto_shash_update(desc, data + 8, data_len - 8 - CMAC_TLEN); + err = crypto_shash_update(desc, zero, 8); + if (err) + return err; + err = crypto_shash_update(desc, data + 8, + data_len - 8 - CMAC_TLEN); + if (err) + return err; } else { - crypto_shash_update(desc, data, data_len - CMAC_TLEN); + err = crypto_shash_update(desc, data, + data_len - CMAC_TLEN); + if (err) + return err; } - crypto_shash_finup(desc, zero, CMAC_TLEN, out); - + err = crypto_shash_finup(desc, zero, CMAC_TLEN, out); + if (err) + return err; memcpy(mic, out, CMAC_TLEN); + + return 0; } -void ieee80211_aes_cmac_256(struct crypto_shash *tfm, const u8 *aad, - const u8 *data, size_t data_len, u8 *mic) +int ieee80211_aes_cmac_256(struct crypto_shash *tfm, const u8 *aad, + const u8 *data, size_t data_len, u8 *mic) { + int err; SHASH_DESC_ON_STACK(desc, tfm); const __le16 *fc; desc->tfm = tfm; - crypto_shash_init(desc); - crypto_shash_update(desc, aad, AAD_LEN); + err = crypto_shash_init(desc); + if (err) + return err; + err = crypto_shash_update(desc, aad, AAD_LEN); + if (err) + return err; fc = (const __le16 *)aad; if (ieee80211_is_beacon(*fc)) { /* mask Timestamp field to zero */ - crypto_shash_update(desc, zero, 8); - crypto_shash_update(desc, data + 8, - data_len - 8 - CMAC_TLEN_256); + err = crypto_shash_update(desc, zero, 8); + if (err) + return err; + err = crypto_shash_update(desc, data + 8, + data_len - 8 - CMAC_TLEN_256); + if (err) + return err; } else { - crypto_shash_update(desc, data, data_len - CMAC_TLEN_256); + err = crypto_shash_update(desc, data, data_len - CMAC_TLEN_256); + if (err) + return err; } - crypto_shash_finup(desc, zero, CMAC_TLEN_256, mic); + return crypto_shash_finup(desc, zero, CMAC_TLEN_256, mic); } struct crypto_shash *ieee80211_aes_cmac_key_setup(const u8 key[], diff --git a/net/mac80211/aes_cmac.h b/net/mac80211/aes_cmac.h index 76817446fb83..f74150542142 100644 --- a/net/mac80211/aes_cmac.h +++ b/net/mac80211/aes_cmac.h @@ -11,10 +11,10 @@ struct crypto_shash *ieee80211_aes_cmac_key_setup(const u8 key[], size_t key_len); -void ieee80211_aes_cmac(struct crypto_shash *tfm, const u8 *aad, - const u8 *data, size_t data_len, u8 *mic); -void ieee80211_aes_cmac_256(struct crypto_shash *tfm, const u8 *aad, - const u8 *data, size_t data_len, u8 *mic); +int ieee80211_aes_cmac(struct crypto_shash *tfm, const u8 *aad, + const u8 *data, size_t data_len, u8 *mic); +int ieee80211_aes_cmac_256(struct crypto_shash *tfm, const u8 *aad, + const u8 *data, size_t data_len, u8 *mic); void ieee80211_aes_cmac_key_free(struct crypto_shash *tfm); #endif /* AES_CMAC_H */ diff --git a/net/mac80211/wpa.c b/net/mac80211/wpa.c index 40d5d9e48479..bb0fa505cdca 100644 --- a/net/mac80211/wpa.c +++ b/net/mac80211/wpa.c @@ -869,8 +869,9 @@ ieee80211_crypto_aes_cmac_encrypt(struct ieee80211_tx_data *tx) /* * MIC = AES-128-CMAC(IGTK, AAD || Management Frame Body || MMIE, 64) */ - ieee80211_aes_cmac(key->u.aes_cmac.tfm, aad, - skb->data + 24, skb->len - 24, mmie->mic); + if (ieee80211_aes_cmac(key->u.aes_cmac.tfm, aad, + skb->data + 24, skb->len - 24, mmie->mic)) + return TX_DROP; return TX_CONTINUE; } @@ -916,8 +917,9 @@ ieee80211_crypto_aes_cmac_256_encrypt(struct ieee80211_tx_data *tx) /* MIC = AES-256-CMAC(IGTK, AAD || Management Frame Body || MMIE, 128) */ - ieee80211_aes_cmac_256(key->u.aes_cmac.tfm, aad, - skb->data + 24, skb->len - 24, mmie->mic); + if (ieee80211_aes_cmac_256(key->u.aes_cmac.tfm, aad, + skb->data + 24, skb->len - 24, mmie->mic)) + return TX_DROP; return TX_CONTINUE; } @@ -956,8 +958,9 @@ ieee80211_crypto_aes_cmac_decrypt(struct ieee80211_rx_data *rx) if (!(status->flag & RX_FLAG_DECRYPTED)) { /* hardware didn't decrypt/verify MIC */ bip_aad(skb, aad); - ieee80211_aes_cmac(key->u.aes_cmac.tfm, aad, - skb->data + 24, skb->len - 24, mic); + if (ieee80211_aes_cmac(key->u.aes_cmac.tfm, aad, + skb->data + 24, skb->len - 24, mic)) + return RX_DROP_U_DECRYPT_FAIL; if (crypto_memneq(mic, mmie->mic, sizeof(mmie->mic))) { key->u.aes_cmac.icverrors++; return RX_DROP_U_MIC_FAIL; @@ -1006,8 +1009,9 @@ ieee80211_crypto_aes_cmac_256_decrypt(struct ieee80211_rx_data *rx) if (!(status->flag & RX_FLAG_DECRYPTED)) { /* hardware didn't decrypt/verify MIC */ bip_aad(skb, aad); - ieee80211_aes_cmac_256(key->u.aes_cmac.tfm, aad, - skb->data + 24, skb->len - 24, mic); + if (ieee80211_aes_cmac_256(key->u.aes_cmac.tfm, aad, + skb->data + 24, skb->len - 24, mic)) + return RX_DROP_U_DECRYPT_FAIL; if (crypto_memneq(mic, mmie->mic, sizeof(mmie->mic))) { key->u.aes_cmac.icverrors++; return RX_DROP_U_MIC_FAIL; From 4255545a28f75fb6082b6f91d1e7ada28383ab22 Mon Sep 17 00:00:00 2001 From: Chien Wong Date: Thu, 13 Nov 2025 22:05:08 +0800 Subject: [PATCH 0929/1381] wifi: mac80211: add generic MMIE struct defines The added struct is needed when writing generic handler for both CMAC-128 and CMAC-256. Signed-off-by: Chien Wong Link: https://patch.msgid.link/20251113140511.48658-3-m@xv97.com Signed-off-by: Johannes Berg --- include/linux/ieee80211.h | 14 +++++++++++++- 1 file changed, 13 insertions(+), 1 deletion(-) diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h index 6d4bc80caf96..d55d8ea3a8be 100644 --- a/include/linux/ieee80211.h +++ b/include/linux/ieee80211.h @@ -1207,7 +1207,7 @@ struct ieee80211_mgmt { #define IEEE80211_MIN_ACTION_SIZE offsetof(struct ieee80211_mgmt, u.action.u) -/* Management MIC information element (IEEE 802.11w) */ +/* Management MIC information element (IEEE 802.11w) for CMAC */ struct ieee80211_mmie { u8 element_id; u8 length; @@ -1225,6 +1225,15 @@ struct ieee80211_mmie_16 { u8 mic[16]; } __packed; +/* Management MIC information element (IEEE 802.11w) for all variants */ +struct ieee80211_mmie_var { + u8 element_id; + u8 length; + __le16 key_id; + u8 sequence_number[6]; + u8 mic[]; /* 8 or 16 bytes */ +} __packed; + struct ieee80211_vendor_ie { u8 element_id; u8 len; @@ -1889,6 +1898,9 @@ enum ieee80211_radio_measurement_actioncode { #define IEEE80211_GCMP_HDR_LEN 8 #define IEEE80211_GCMP_MIC_LEN 16 #define IEEE80211_GCMP_PN_LEN 6 +#define IEEE80211_CMAC_128_MIC_LEN 8 +#define IEEE80211_CMAC_256_MIC_LEN 16 +#define IEEE80211_GMAC_MIC_LEN 16 #define FILS_NONCE_LEN 16 #define FILS_MAX_KEK_LEN 64 From edf62602fcbbc0440565ed18372a01e4c7bc9a9d Mon Sep 17 00:00:00 2001 From: Chien Wong Date: Thu, 13 Nov 2025 22:05:09 +0800 Subject: [PATCH 0930/1381] wifi: mac80211: utilize the newly defined CMAC constants Make use of the added constants to reduce duplication. Signed-off-by: Chien Wong Link: https://patch.msgid.link/20251113140511.48658-4-m@xv97.com Signed-off-by: Johannes Berg --- net/mac80211/aes_cmac.c | 25 ++++++++++++------------- net/mac80211/aes_gmac.c | 22 ++++++++++++---------- net/mac80211/aes_gmac.h | 1 - net/mac80211/wpa.c | 2 +- 4 files changed, 25 insertions(+), 25 deletions(-) diff --git a/net/mac80211/aes_cmac.c b/net/mac80211/aes_cmac.c index 65989c7dfc68..672ed80ee4ff 100644 --- a/net/mac80211/aes_cmac.c +++ b/net/mac80211/aes_cmac.c @@ -16,11 +16,9 @@ #include "key.h" #include "aes_cmac.h" -#define CMAC_TLEN 8 /* CMAC TLen = 64 bits (8 octets) */ -#define CMAC_TLEN_256 16 /* CMAC TLen = 128 bits (16 octets) */ #define AAD_LEN 20 -static const u8 zero[CMAC_TLEN_256]; +static const u8 zero[IEEE80211_CMAC_256_MIC_LEN]; int ieee80211_aes_cmac(struct crypto_shash *tfm, const u8 *aad, const u8 *data, size_t data_len, u8 *mic) @@ -44,20 +42,20 @@ int ieee80211_aes_cmac(struct crypto_shash *tfm, const u8 *aad, err = crypto_shash_update(desc, zero, 8); if (err) return err; - err = crypto_shash_update(desc, data + 8, - data_len - 8 - CMAC_TLEN); + err = crypto_shash_update(desc, data + 8, data_len - 8 - + IEEE80211_CMAC_128_MIC_LEN); if (err) return err; } else { - err = crypto_shash_update(desc, data, - data_len - CMAC_TLEN); + err = crypto_shash_update(desc, data, data_len - + IEEE80211_CMAC_128_MIC_LEN); if (err) return err; } - err = crypto_shash_finup(desc, zero, CMAC_TLEN, out); + err = crypto_shash_finup(desc, zero, IEEE80211_CMAC_128_MIC_LEN, out); if (err) return err; - memcpy(mic, out, CMAC_TLEN); + memcpy(mic, out, IEEE80211_CMAC_128_MIC_LEN); return 0; } @@ -83,16 +81,17 @@ int ieee80211_aes_cmac_256(struct crypto_shash *tfm, const u8 *aad, err = crypto_shash_update(desc, zero, 8); if (err) return err; - err = crypto_shash_update(desc, data + 8, - data_len - 8 - CMAC_TLEN_256); + err = crypto_shash_update(desc, data + 8, data_len - 8 - + IEEE80211_CMAC_256_MIC_LEN); if (err) return err; } else { - err = crypto_shash_update(desc, data, data_len - CMAC_TLEN_256); + err = crypto_shash_update(desc, data, data_len - + IEEE80211_CMAC_256_MIC_LEN); if (err) return err; } - return crypto_shash_finup(desc, zero, CMAC_TLEN_256, mic); + return crypto_shash_finup(desc, zero, IEEE80211_CMAC_256_MIC_LEN, mic); } struct crypto_shash *ieee80211_aes_cmac_key_setup(const u8 key[], diff --git a/net/mac80211/aes_gmac.c b/net/mac80211/aes_gmac.c index 512cab073f2e..811a83d8d525 100644 --- a/net/mac80211/aes_gmac.c +++ b/net/mac80211/aes_gmac.c @@ -24,15 +24,16 @@ int ieee80211_aes_gmac(struct crypto_aead *tfm, const u8 *aad, u8 *nonce, const __le16 *fc; int ret; - if (data_len < GMAC_MIC_LEN) + if (data_len < IEEE80211_GMAC_MIC_LEN) return -EINVAL; - aead_req = kzalloc(reqsize + GMAC_MIC_LEN + GMAC_AAD_LEN, GFP_ATOMIC); + aead_req = kzalloc(reqsize + IEEE80211_GMAC_MIC_LEN + GMAC_AAD_LEN, + GFP_ATOMIC); if (!aead_req) return -ENOMEM; zero = (u8 *)aead_req + reqsize; - __aad = zero + GMAC_MIC_LEN; + __aad = zero + IEEE80211_GMAC_MIC_LEN; memcpy(__aad, aad, GMAC_AAD_LEN); fc = (const __le16 *)aad; @@ -41,15 +42,16 @@ int ieee80211_aes_gmac(struct crypto_aead *tfm, const u8 *aad, u8 *nonce, sg_init_table(sg, 5); sg_set_buf(&sg[0], __aad, GMAC_AAD_LEN); sg_set_buf(&sg[1], zero, 8); - sg_set_buf(&sg[2], data + 8, data_len - 8 - GMAC_MIC_LEN); - sg_set_buf(&sg[3], zero, GMAC_MIC_LEN); - sg_set_buf(&sg[4], mic, GMAC_MIC_LEN); + sg_set_buf(&sg[2], data + 8, + data_len - 8 - IEEE80211_GMAC_MIC_LEN); + sg_set_buf(&sg[3], zero, IEEE80211_GMAC_MIC_LEN); + sg_set_buf(&sg[4], mic, IEEE80211_GMAC_MIC_LEN); } else { sg_init_table(sg, 4); sg_set_buf(&sg[0], __aad, GMAC_AAD_LEN); - sg_set_buf(&sg[1], data, data_len - GMAC_MIC_LEN); - sg_set_buf(&sg[2], zero, GMAC_MIC_LEN); - sg_set_buf(&sg[3], mic, GMAC_MIC_LEN); + sg_set_buf(&sg[1], data, data_len - IEEE80211_GMAC_MIC_LEN); + sg_set_buf(&sg[2], zero, IEEE80211_GMAC_MIC_LEN); + sg_set_buf(&sg[3], mic, IEEE80211_GMAC_MIC_LEN); } memcpy(iv, nonce, GMAC_NONCE_LEN); @@ -78,7 +80,7 @@ struct crypto_aead *ieee80211_aes_gmac_key_setup(const u8 key[], err = crypto_aead_setkey(tfm, key, key_len); if (!err) - err = crypto_aead_setauthsize(tfm, GMAC_MIC_LEN); + err = crypto_aead_setauthsize(tfm, IEEE80211_GMAC_MIC_LEN); if (!err) return tfm; diff --git a/net/mac80211/aes_gmac.h b/net/mac80211/aes_gmac.h index c739356bae2a..206136b60bca 100644 --- a/net/mac80211/aes_gmac.h +++ b/net/mac80211/aes_gmac.h @@ -9,7 +9,6 @@ #include #define GMAC_AAD_LEN 20 -#define GMAC_MIC_LEN 16 #define GMAC_NONCE_LEN 12 struct crypto_aead *ieee80211_aes_gmac_key_setup(const u8 key[], diff --git a/net/mac80211/wpa.c b/net/mac80211/wpa.c index bb0fa505cdca..3d5efd8c6e2d 100644 --- a/net/mac80211/wpa.c +++ b/net/mac80211/wpa.c @@ -1117,7 +1117,7 @@ ieee80211_crypto_aes_gmac_decrypt(struct ieee80211_rx_data *rx) memcpy(nonce, hdr->addr2, ETH_ALEN); memcpy(nonce + ETH_ALEN, ipn, 6); - mic = kmalloc(GMAC_MIC_LEN, GFP_ATOMIC); + mic = kmalloc(IEEE80211_GMAC_MIC_LEN, GFP_ATOMIC); if (!mic) return RX_DROP_U_OOM; if (ieee80211_aes_gmac(key->u.aes_gmac.tfm, aad, nonce, From a22fb19244cb2425b075106541bd5b3eee42730c Mon Sep 17 00:00:00 2001 From: Chien Wong Date: Thu, 13 Nov 2025 22:05:10 +0800 Subject: [PATCH 0931/1381] wifi: mac80211: refactor CMAC crypt functions ieee80211_aes_cmac() and ieee80211_aes_cmac_256() are almost the same. Merge them. This removes duplication. Signed-off-by: Chien Wong Link: https://patch.msgid.link/20251113140511.48658-5-m@xv97.com Signed-off-by: Johannes Berg --- net/mac80211/aes_cmac.c | 48 ++++++----------------------------------- net/mac80211/aes_cmac.h | 5 ++--- net/mac80211/wpa.c | 16 ++++++++------ 3 files changed, 19 insertions(+), 50 deletions(-) diff --git a/net/mac80211/aes_cmac.c b/net/mac80211/aes_cmac.c index 672ed80ee4ff..0827965455dc 100644 --- a/net/mac80211/aes_cmac.c +++ b/net/mac80211/aes_cmac.c @@ -21,7 +21,8 @@ static const u8 zero[IEEE80211_CMAC_256_MIC_LEN]; int ieee80211_aes_cmac(struct crypto_shash *tfm, const u8 *aad, - const u8 *data, size_t data_len, u8 *mic) + const u8 *data, size_t data_len, u8 *mic, + unsigned int mic_len) { int err; SHASH_DESC_ON_STACK(desc, tfm); @@ -42,58 +43,23 @@ int ieee80211_aes_cmac(struct crypto_shash *tfm, const u8 *aad, err = crypto_shash_update(desc, zero, 8); if (err) return err; - err = crypto_shash_update(desc, data + 8, data_len - 8 - - IEEE80211_CMAC_128_MIC_LEN); + err = crypto_shash_update(desc, data + 8, + data_len - 8 - mic_len); if (err) return err; } else { - err = crypto_shash_update(desc, data, data_len - - IEEE80211_CMAC_128_MIC_LEN); + err = crypto_shash_update(desc, data, data_len - mic_len); if (err) return err; } - err = crypto_shash_finup(desc, zero, IEEE80211_CMAC_128_MIC_LEN, out); + err = crypto_shash_finup(desc, zero, mic_len, out); if (err) return err; - memcpy(mic, out, IEEE80211_CMAC_128_MIC_LEN); + memcpy(mic, out, mic_len); return 0; } -int ieee80211_aes_cmac_256(struct crypto_shash *tfm, const u8 *aad, - const u8 *data, size_t data_len, u8 *mic) -{ - int err; - SHASH_DESC_ON_STACK(desc, tfm); - const __le16 *fc; - - desc->tfm = tfm; - - err = crypto_shash_init(desc); - if (err) - return err; - err = crypto_shash_update(desc, aad, AAD_LEN); - if (err) - return err; - fc = (const __le16 *)aad; - if (ieee80211_is_beacon(*fc)) { - /* mask Timestamp field to zero */ - err = crypto_shash_update(desc, zero, 8); - if (err) - return err; - err = crypto_shash_update(desc, data + 8, data_len - 8 - - IEEE80211_CMAC_256_MIC_LEN); - if (err) - return err; - } else { - err = crypto_shash_update(desc, data, data_len - - IEEE80211_CMAC_256_MIC_LEN); - if (err) - return err; - } - return crypto_shash_finup(desc, zero, IEEE80211_CMAC_256_MIC_LEN, mic); -} - struct crypto_shash *ieee80211_aes_cmac_key_setup(const u8 key[], size_t key_len) { diff --git a/net/mac80211/aes_cmac.h b/net/mac80211/aes_cmac.h index f74150542142..5f971a8298cb 100644 --- a/net/mac80211/aes_cmac.h +++ b/net/mac80211/aes_cmac.h @@ -12,9 +12,8 @@ struct crypto_shash *ieee80211_aes_cmac_key_setup(const u8 key[], size_t key_len); int ieee80211_aes_cmac(struct crypto_shash *tfm, const u8 *aad, - const u8 *data, size_t data_len, u8 *mic); -int ieee80211_aes_cmac_256(struct crypto_shash *tfm, const u8 *aad, - const u8 *data, size_t data_len, u8 *mic); + const u8 *data, size_t data_len, u8 *mic, + unsigned int mic_len); void ieee80211_aes_cmac_key_free(struct crypto_shash *tfm); #endif /* AES_CMAC_H */ diff --git a/net/mac80211/wpa.c b/net/mac80211/wpa.c index 3d5efd8c6e2d..7431ccecd17f 100644 --- a/net/mac80211/wpa.c +++ b/net/mac80211/wpa.c @@ -870,7 +870,8 @@ ieee80211_crypto_aes_cmac_encrypt(struct ieee80211_tx_data *tx) * MIC = AES-128-CMAC(IGTK, AAD || Management Frame Body || MMIE, 64) */ if (ieee80211_aes_cmac(key->u.aes_cmac.tfm, aad, - skb->data + 24, skb->len - 24, mmie->mic)) + skb->data + 24, skb->len - 24, mmie->mic, + IEEE80211_CMAC_128_MIC_LEN)) return TX_DROP; return TX_CONTINUE; @@ -917,8 +918,9 @@ ieee80211_crypto_aes_cmac_256_encrypt(struct ieee80211_tx_data *tx) /* MIC = AES-256-CMAC(IGTK, AAD || Management Frame Body || MMIE, 128) */ - if (ieee80211_aes_cmac_256(key->u.aes_cmac.tfm, aad, - skb->data + 24, skb->len - 24, mmie->mic)) + if (ieee80211_aes_cmac(key->u.aes_cmac.tfm, aad, + skb->data + 24, skb->len - 24, mmie->mic, + IEEE80211_CMAC_256_MIC_LEN)) return TX_DROP; return TX_CONTINUE; @@ -959,7 +961,8 @@ ieee80211_crypto_aes_cmac_decrypt(struct ieee80211_rx_data *rx) /* hardware didn't decrypt/verify MIC */ bip_aad(skb, aad); if (ieee80211_aes_cmac(key->u.aes_cmac.tfm, aad, - skb->data + 24, skb->len - 24, mic)) + skb->data + 24, skb->len - 24, mic, + IEEE80211_CMAC_128_MIC_LEN)) return RX_DROP_U_DECRYPT_FAIL; if (crypto_memneq(mic, mmie->mic, sizeof(mmie->mic))) { key->u.aes_cmac.icverrors++; @@ -1009,8 +1012,9 @@ ieee80211_crypto_aes_cmac_256_decrypt(struct ieee80211_rx_data *rx) if (!(status->flag & RX_FLAG_DECRYPTED)) { /* hardware didn't decrypt/verify MIC */ bip_aad(skb, aad); - if (ieee80211_aes_cmac_256(key->u.aes_cmac.tfm, aad, - skb->data + 24, skb->len - 24, mic)) + if (ieee80211_aes_cmac(key->u.aes_cmac.tfm, aad, + skb->data + 24, skb->len - 24, mic, + IEEE80211_CMAC_256_MIC_LEN)) return RX_DROP_U_DECRYPT_FAIL; if (crypto_memneq(mic, mmie->mic, sizeof(mmie->mic))) { key->u.aes_cmac.icverrors++; From b688a9447be360d09c0982234fcd703ac7dd8c15 Mon Sep 17 00:00:00 2001 From: Chien Wong Date: Thu, 13 Nov 2025 22:05:11 +0800 Subject: [PATCH 0932/1381] wifi: mac80211: refactor CMAC packet handlers Merge CMAC-128 and CMAC-256 handlers since they are almost the same. This removes duplication. The comment 'MIC = AES-128-CMAC(IGTK, AAD ...' is out-dated since CMAC is also used with BIGTK, as is the comment for CMAC-256. Simply remove the comments. Tested-on: mac80211_hwsim Signed-off-by: Chien Wong Link: https://patch.msgid.link/20251113140511.48658-6-m@xv97.com Signed-off-by: Johannes Berg --- net/mac80211/rx.c | 6 +- net/mac80211/tx.c | 6 +- net/mac80211/wpa.c | 150 ++++++++------------------------------------- net/mac80211/wpa.h | 10 ++- 4 files changed, 39 insertions(+), 133 deletions(-) diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c index cc26c279984d..6a1899512d07 100644 --- a/net/mac80211/rx.c +++ b/net/mac80211/rx.c @@ -2215,10 +2215,12 @@ ieee80211_rx_h_decrypt(struct ieee80211_rx_data *rx) rx, IEEE80211_CCMP_256_MIC_LEN); break; case WLAN_CIPHER_SUITE_AES_CMAC: - result = ieee80211_crypto_aes_cmac_decrypt(rx); + result = ieee80211_crypto_aes_cmac_decrypt( + rx, IEEE80211_CMAC_128_MIC_LEN); break; case WLAN_CIPHER_SUITE_BIP_CMAC_256: - result = ieee80211_crypto_aes_cmac_256_decrypt(rx); + result = ieee80211_crypto_aes_cmac_decrypt( + rx, IEEE80211_CMAC_256_MIC_LEN); break; case WLAN_CIPHER_SUITE_BIP_GMAC_128: case WLAN_CIPHER_SUITE_BIP_GMAC_256: diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index e7b141c55f7a..9d8b0a25f73c 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -1062,9 +1062,11 @@ ieee80211_tx_h_encrypt(struct ieee80211_tx_data *tx) return ieee80211_crypto_ccmp_encrypt( tx, IEEE80211_CCMP_256_MIC_LEN); case WLAN_CIPHER_SUITE_AES_CMAC: - return ieee80211_crypto_aes_cmac_encrypt(tx); + return ieee80211_crypto_aes_cmac_encrypt( + tx, IEEE80211_CMAC_128_MIC_LEN); case WLAN_CIPHER_SUITE_BIP_CMAC_256: - return ieee80211_crypto_aes_cmac_256_encrypt(tx); + return ieee80211_crypto_aes_cmac_encrypt( + tx, IEEE80211_CMAC_256_MIC_LEN); case WLAN_CIPHER_SUITE_BIP_GMAC_128: case WLAN_CIPHER_SUITE_BIP_GMAC_256: return ieee80211_crypto_aes_gmac_encrypt(tx); diff --git a/net/mac80211/wpa.c b/net/mac80211/wpa.c index 7431ccecd17f..4a858112e4ef 100644 --- a/net/mac80211/wpa.c +++ b/net/mac80211/wpa.c @@ -828,12 +828,14 @@ static inline void bip_ipn_swap(u8 *d, const u8 *s) ieee80211_tx_result -ieee80211_crypto_aes_cmac_encrypt(struct ieee80211_tx_data *tx) +ieee80211_crypto_aes_cmac_encrypt(struct ieee80211_tx_data *tx, + unsigned int mic_len) { struct sk_buff *skb; struct ieee80211_tx_info *info; struct ieee80211_key *key = tx->key; - struct ieee80211_mmie *mmie; + struct ieee80211_mmie_var *mmie; + size_t mmie_len; u8 aad[20]; u64 pn64; @@ -848,12 +850,14 @@ ieee80211_crypto_aes_cmac_encrypt(struct ieee80211_tx_data *tx) !(key->conf.flags & IEEE80211_KEY_FLAG_GENERATE_MMIE)) return TX_CONTINUE; - if (WARN_ON(skb_tailroom(skb) < sizeof(*mmie))) + mmie_len = sizeof(*mmie) + mic_len; + + if (WARN_ON(skb_tailroom(skb) < mmie_len)) return TX_DROP; - mmie = skb_put(skb, sizeof(*mmie)); + mmie = skb_put(skb, mmie_len); mmie->element_id = WLAN_EID_MMIE; - mmie->length = sizeof(*mmie) - 2; + mmie->length = mmie_len - 2; mmie->key_id = cpu_to_le16(key->conf.keyidx); /* PN = PN + 1 */ @@ -866,88 +870,40 @@ ieee80211_crypto_aes_cmac_encrypt(struct ieee80211_tx_data *tx) bip_aad(skb, aad); - /* - * MIC = AES-128-CMAC(IGTK, AAD || Management Frame Body || MMIE, 64) - */ if (ieee80211_aes_cmac(key->u.aes_cmac.tfm, aad, - skb->data + 24, skb->len - 24, mmie->mic, - IEEE80211_CMAC_128_MIC_LEN)) - return TX_DROP; - - return TX_CONTINUE; -} - -ieee80211_tx_result -ieee80211_crypto_aes_cmac_256_encrypt(struct ieee80211_tx_data *tx) -{ - struct sk_buff *skb; - struct ieee80211_tx_info *info; - struct ieee80211_key *key = tx->key; - struct ieee80211_mmie_16 *mmie; - u8 aad[20]; - u64 pn64; - - if (WARN_ON(skb_queue_len(&tx->skbs) != 1)) - return TX_DROP; - - skb = skb_peek(&tx->skbs); - - info = IEEE80211_SKB_CB(skb); - - if (info->control.hw_key && - !(key->conf.flags & IEEE80211_KEY_FLAG_GENERATE_MMIE)) - return TX_CONTINUE; - - if (WARN_ON(skb_tailroom(skb) < sizeof(*mmie))) - return TX_DROP; - - mmie = skb_put(skb, sizeof(*mmie)); - mmie->element_id = WLAN_EID_MMIE; - mmie->length = sizeof(*mmie) - 2; - mmie->key_id = cpu_to_le16(key->conf.keyidx); - - /* PN = PN + 1 */ - pn64 = atomic64_inc_return(&key->conf.tx_pn); - - bip_ipn_set64(mmie->sequence_number, pn64); - - if (info->control.hw_key) - return TX_CONTINUE; - - bip_aad(skb, aad); - - /* MIC = AES-256-CMAC(IGTK, AAD || Management Frame Body || MMIE, 128) - */ - if (ieee80211_aes_cmac(key->u.aes_cmac.tfm, aad, - skb->data + 24, skb->len - 24, mmie->mic, - IEEE80211_CMAC_256_MIC_LEN)) + skb->data + 24, skb->len - 24, + mmie->mic, mic_len)) return TX_DROP; return TX_CONTINUE; } ieee80211_rx_result -ieee80211_crypto_aes_cmac_decrypt(struct ieee80211_rx_data *rx) +ieee80211_crypto_aes_cmac_decrypt(struct ieee80211_rx_data *rx, + unsigned int mic_len) { struct sk_buff *skb = rx->skb; struct ieee80211_rx_status *status = IEEE80211_SKB_RXCB(skb); struct ieee80211_key *key = rx->key; - struct ieee80211_mmie *mmie; - u8 aad[20], mic[8], ipn[6]; + struct ieee80211_mmie_var *mmie; + size_t mmie_len; + u8 aad[20], mic[IEEE80211_CMAC_256_MIC_LEN], ipn[6]; struct ieee80211_hdr *hdr = (struct ieee80211_hdr *) skb->data; if (!ieee80211_is_mgmt(hdr->frame_control)) return RX_CONTINUE; + mmie_len = sizeof(*mmie) + mic_len; + /* management frames are already linear */ - if (skb->len < 24 + sizeof(*mmie)) - return RX_DROP_U_SHORT_CMAC; + if (skb->len < 24 + mmie_len) + return mic_len == IEEE80211_CMAC_128_MIC_LEN ? + RX_DROP_U_SHORT_CMAC : RX_DROP_U_SHORT_CMAC256; - mmie = (struct ieee80211_mmie *) - (skb->data + skb->len - sizeof(*mmie)); + mmie = (struct ieee80211_mmie_var *)(skb->data + skb->len - mmie_len); if (mmie->element_id != WLAN_EID_MMIE || - mmie->length != sizeof(*mmie) - 2) + mmie->length != mmie_len - 2) return RX_DROP_U_BAD_MMIE; /* Invalid MMIE */ bip_ipn_swap(ipn, mmie->sequence_number); @@ -961,10 +917,10 @@ ieee80211_crypto_aes_cmac_decrypt(struct ieee80211_rx_data *rx) /* hardware didn't decrypt/verify MIC */ bip_aad(skb, aad); if (ieee80211_aes_cmac(key->u.aes_cmac.tfm, aad, - skb->data + 24, skb->len - 24, mic, - IEEE80211_CMAC_128_MIC_LEN)) + skb->data + 24, skb->len - 24, + mic, mic_len)) return RX_DROP_U_DECRYPT_FAIL; - if (crypto_memneq(mic, mmie->mic, sizeof(mmie->mic))) { + if (crypto_memneq(mic, mmie->mic, mic_len)) { key->u.aes_cmac.icverrors++; return RX_DROP_U_MIC_FAIL; } @@ -973,59 +929,7 @@ ieee80211_crypto_aes_cmac_decrypt(struct ieee80211_rx_data *rx) memcpy(key->u.aes_cmac.rx_pn, ipn, 6); /* Remove MMIE */ - skb_trim(skb, skb->len - sizeof(*mmie)); - - return RX_CONTINUE; -} - -ieee80211_rx_result -ieee80211_crypto_aes_cmac_256_decrypt(struct ieee80211_rx_data *rx) -{ - struct sk_buff *skb = rx->skb; - struct ieee80211_rx_status *status = IEEE80211_SKB_RXCB(skb); - struct ieee80211_key *key = rx->key; - struct ieee80211_mmie_16 *mmie; - u8 aad[20], mic[16], ipn[6]; - struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)skb->data; - - if (!ieee80211_is_mgmt(hdr->frame_control)) - return RX_CONTINUE; - - /* management frames are already linear */ - - if (skb->len < 24 + sizeof(*mmie)) - return RX_DROP_U_SHORT_CMAC256; - - mmie = (struct ieee80211_mmie_16 *) - (skb->data + skb->len - sizeof(*mmie)); - if (mmie->element_id != WLAN_EID_MMIE || - mmie->length != sizeof(*mmie) - 2) - return RX_DROP_U_BAD_MMIE; /* Invalid MMIE */ - - bip_ipn_swap(ipn, mmie->sequence_number); - - if (memcmp(ipn, key->u.aes_cmac.rx_pn, 6) <= 0) { - key->u.aes_cmac.replays++; - return RX_DROP_U_REPLAY; - } - - if (!(status->flag & RX_FLAG_DECRYPTED)) { - /* hardware didn't decrypt/verify MIC */ - bip_aad(skb, aad); - if (ieee80211_aes_cmac(key->u.aes_cmac.tfm, aad, - skb->data + 24, skb->len - 24, mic, - IEEE80211_CMAC_256_MIC_LEN)) - return RX_DROP_U_DECRYPT_FAIL; - if (crypto_memneq(mic, mmie->mic, sizeof(mmie->mic))) { - key->u.aes_cmac.icverrors++; - return RX_DROP_U_MIC_FAIL; - } - } - - memcpy(key->u.aes_cmac.rx_pn, ipn, 6); - - /* Remove MMIE */ - skb_trim(skb, skb->len - sizeof(*mmie)); + skb_trim(skb, skb->len - mmie_len); return RX_CONTINUE; } diff --git a/net/mac80211/wpa.h b/net/mac80211/wpa.h index a9a81abb5479..6e8846dfe710 100644 --- a/net/mac80211/wpa.h +++ b/net/mac80211/wpa.h @@ -29,13 +29,11 @@ ieee80211_crypto_ccmp_decrypt(struct ieee80211_rx_data *rx, unsigned int mic_len); ieee80211_tx_result -ieee80211_crypto_aes_cmac_encrypt(struct ieee80211_tx_data *tx); -ieee80211_tx_result -ieee80211_crypto_aes_cmac_256_encrypt(struct ieee80211_tx_data *tx); +ieee80211_crypto_aes_cmac_encrypt(struct ieee80211_tx_data *tx, + unsigned int mic_len); ieee80211_rx_result -ieee80211_crypto_aes_cmac_decrypt(struct ieee80211_rx_data *rx); -ieee80211_rx_result -ieee80211_crypto_aes_cmac_256_decrypt(struct ieee80211_rx_data *rx); +ieee80211_crypto_aes_cmac_decrypt(struct ieee80211_rx_data *rx, + unsigned int mic_len); ieee80211_tx_result ieee80211_crypto_aes_gmac_encrypt(struct ieee80211_tx_data *tx); ieee80211_rx_result From 354d128aa7212c53ffc7127877953264a445f5af Mon Sep 17 00:00:00 2001 From: Jiawen Wu Date: Tue, 18 Nov 2025 16:02:55 +0800 Subject: [PATCH 0933/1381] net: txgbe: support CR modules for AML devices Support to identify 25G/10G CR modules for AML devices. Autoneg is enbaled by default in CR mode. Signed-off-by: Jiawen Wu Link: https://patch.msgid.link/20251118080259.24676-2-jiawenwu@trustnetic.com Signed-off-by: Paolo Abeni --- .../net/ethernet/wangxun/txgbe/txgbe_aml.c | 59 +++++++++++++------ .../ethernet/wangxun/txgbe/txgbe_ethtool.c | 3 +- 2 files changed, 44 insertions(+), 18 deletions(-) diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_aml.c b/drivers/net/ethernet/wangxun/txgbe/txgbe_aml.c index 35eebdb07761..a7650f548fa4 100644 --- a/drivers/net/ethernet/wangxun/txgbe/txgbe_aml.c +++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_aml.c @@ -104,7 +104,8 @@ static int txgbe_set_phy_link_hostif(struct wx *wx, int speed, int autoneg, int WX_HI_COMMAND_TIMEOUT, true); } -static void txgbe_get_link_capabilities(struct wx *wx, int *speed, int *duplex) +static void txgbe_get_link_capabilities(struct wx *wx, int *speed, + int *autoneg, int *duplex) { struct txgbe *txgbe = wx->priv; @@ -115,6 +116,7 @@ static void txgbe_get_link_capabilities(struct wx *wx, int *speed, int *duplex) else *speed = SPEED_UNKNOWN; + *autoneg = phylink_test(txgbe->advertising, Autoneg); *duplex = *speed == SPEED_UNKNOWN ? DUPLEX_HALF : DUPLEX_FULL; } @@ -135,11 +137,11 @@ static void txgbe_get_mac_link(struct wx *wx, int *speed) int txgbe_set_phy_link(struct wx *wx) { - int speed, duplex, err; + int speed, autoneg, duplex, err; - txgbe_get_link_capabilities(wx, &speed, &duplex); + txgbe_get_link_capabilities(wx, &speed, &autoneg, &duplex); - err = txgbe_set_phy_link_hostif(wx, speed, 0, duplex); + err = txgbe_set_phy_link_hostif(wx, speed, autoneg, duplex); if (err) { wx_err(wx, "Failed to setup link\n"); return err; @@ -154,19 +156,43 @@ static int txgbe_sfp_to_linkmodes(struct wx *wx, struct txgbe_sfp_id *id) DECLARE_PHY_INTERFACE_MASK(interfaces); struct txgbe *txgbe = wx->priv; - if (id->com_25g_code & (TXGBE_SFF_25GBASESR_CAPABLE | - TXGBE_SFF_25GBASEER_CAPABLE | - TXGBE_SFF_25GBASELR_CAPABLE)) { - phylink_set(modes, 25000baseSR_Full); + if (id->cable_tech & TXGBE_SFF_DA_PASSIVE_CABLE) { + txgbe->link_port = PORT_DA; + phylink_set(modes, Autoneg); + if (id->com_25g_code == TXGBE_SFF_25GBASECR_91FEC || + id->com_25g_code == TXGBE_SFF_25GBASECR_74FEC || + id->com_25g_code == TXGBE_SFF_25GBASECR_NOFEC) { + phylink_set(modes, 25000baseCR_Full); + phylink_set(modes, 10000baseCR_Full); + __set_bit(PHY_INTERFACE_MODE_25GBASER, interfaces); + __set_bit(PHY_INTERFACE_MODE_10GBASER, interfaces); + } else { + phylink_set(modes, 10000baseCR_Full); + __set_bit(PHY_INTERFACE_MODE_10GBASER, interfaces); + } + } else if (id->cable_tech & TXGBE_SFF_DA_ACTIVE_CABLE) { + txgbe->link_port = PORT_DA; + phylink_set(modes, Autoneg); + phylink_set(modes, 25000baseCR_Full); __set_bit(PHY_INTERFACE_MODE_25GBASER, interfaces); - } - if (id->com_10g_code & TXGBE_SFF_10GBASESR_CAPABLE) { - phylink_set(modes, 10000baseSR_Full); - __set_bit(PHY_INTERFACE_MODE_10GBASER, interfaces); - } - if (id->com_10g_code & TXGBE_SFF_10GBASELR_CAPABLE) { - phylink_set(modes, 10000baseLR_Full); - __set_bit(PHY_INTERFACE_MODE_10GBASER, interfaces); + } else { + if (id->com_25g_code == TXGBE_SFF_25GBASESR_CAPABLE || + id->com_25g_code == TXGBE_SFF_25GBASEER_CAPABLE || + id->com_25g_code == TXGBE_SFF_25GBASELR_CAPABLE) { + txgbe->link_port = PORT_FIBRE; + phylink_set(modes, 25000baseSR_Full); + __set_bit(PHY_INTERFACE_MODE_25GBASER, interfaces); + } + if (id->com_10g_code & TXGBE_SFF_10GBASESR_CAPABLE) { + txgbe->link_port = PORT_FIBRE; + phylink_set(modes, 10000baseSR_Full); + __set_bit(PHY_INTERFACE_MODE_10GBASER, interfaces); + } + if (id->com_10g_code & TXGBE_SFF_10GBASELR_CAPABLE) { + txgbe->link_port = PORT_FIBRE; + phylink_set(modes, 10000baseLR_Full); + __set_bit(PHY_INTERFACE_MODE_10GBASER, interfaces); + } } if (phy_interface_empty(interfaces)) { @@ -177,7 +203,6 @@ static int txgbe_sfp_to_linkmodes(struct wx *wx, struct txgbe_sfp_id *id) phylink_set(modes, Pause); phylink_set(modes, Asym_Pause); phylink_set(modes, FIBRE); - txgbe->link_port = PORT_FIBRE; if (!linkmode_equal(txgbe->sfp_support, modes)) { linkmode_copy(txgbe->sfp_support, modes); diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_ethtool.c b/drivers/net/ethernet/wangxun/txgbe/txgbe_ethtool.c index e285b088c7b2..e8dd277a35c7 100644 --- a/drivers/net/ethernet/wangxun/txgbe/txgbe_ethtool.c +++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_ethtool.c @@ -30,7 +30,8 @@ int txgbe_get_link_ksettings(struct net_device *netdev, return 0; cmd->base.port = txgbe->link_port; - cmd->base.autoneg = AUTONEG_DISABLE; + cmd->base.autoneg = phylink_test(txgbe->advertising, Autoneg) ? + AUTONEG_ENABLE : AUTONEG_DISABLE; linkmode_copy(cmd->link_modes.supported, txgbe->sfp_support); linkmode_copy(cmd->link_modes.advertising, txgbe->advertising); From dbba6b7a47cba914d48890da7233a64c7b9f3ccc Mon Sep 17 00:00:00 2001 From: Jiawen Wu Date: Tue, 18 Nov 2025 16:02:56 +0800 Subject: [PATCH 0934/1381] net: txgbe: rename the SFP related QSFP supported will be introduced for AML 40G devices, the code related to identify various modules should be renamed to more appropriate names. And struct txgbe_hic_i2c_read used to get module information is renamed as struct txgbe_hic_get_module_info, because another SW-FW command to read I2C will be added later. Signed-off-by: Jiawen Wu Link: https://patch.msgid.link/20251118080259.24676-3-jiawenwu@trustnetic.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/wangxun/libwx/wx_type.h | 2 +- .../net/ethernet/wangxun/txgbe/txgbe_aml.c | 39 ++++++++++--------- .../net/ethernet/wangxun/txgbe/txgbe_aml.h | 2 +- .../ethernet/wangxun/txgbe/txgbe_ethtool.c | 2 +- .../net/ethernet/wangxun/txgbe/txgbe_main.c | 12 +++--- .../net/ethernet/wangxun/txgbe/txgbe_type.h | 12 +++--- 6 files changed, 35 insertions(+), 34 deletions(-) diff --git a/drivers/net/ethernet/wangxun/libwx/wx_type.h b/drivers/net/ethernet/wangxun/libwx/wx_type.h index b1a6ef5709a9..29e5c5470c94 100644 --- a/drivers/net/ethernet/wangxun/libwx/wx_type.h +++ b/drivers/net/ethernet/wangxun/libwx/wx_type.h @@ -1249,7 +1249,7 @@ enum wx_pf_flags { WX_FLAG_RX_HWTSTAMP_IN_REGISTER, WX_FLAG_PTP_PPS_ENABLED, WX_FLAG_NEED_LINK_CONFIG, - WX_FLAG_NEED_SFP_RESET, + WX_FLAG_NEED_MODULE_RESET, WX_FLAG_NEED_UPDATE_LINK, WX_FLAG_NEED_DO_RESET, WX_FLAG_RX_MERGE_ENABLED, diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_aml.c b/drivers/net/ethernet/wangxun/txgbe/txgbe_aml.c index a7650f548fa4..432880ccc640 100644 --- a/drivers/net/ethernet/wangxun/txgbe/txgbe_aml.c +++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_aml.c @@ -38,7 +38,7 @@ irqreturn_t txgbe_gpio_irq_handler_aml(int irq, void *data) wr32(wx, WX_GPIO_INTMASK, 0xFF); status = rd32(wx, WX_GPIO_INTSTATUS); if (status & TXGBE_GPIOBIT_2) { - set_bit(WX_FLAG_NEED_SFP_RESET, wx->flags); + set_bit(WX_FLAG_NEED_MODULE_RESET, wx->flags); wr32(wx, WX_GPIO_EOI, TXGBE_GPIOBIT_2); wx_service_event_schedule(wx); } @@ -63,15 +63,16 @@ int txgbe_test_hostif(struct wx *wx) WX_HI_COMMAND_TIMEOUT, true); } -static int txgbe_identify_sfp_hostif(struct wx *wx, struct txgbe_hic_i2c_read *buffer) +static int txgbe_identify_module_hostif(struct wx *wx, + struct txgbe_hic_get_module_info *buffer) { - buffer->hdr.cmd = FW_READ_SFP_INFO_CMD; - buffer->hdr.buf_len = sizeof(struct txgbe_hic_i2c_read) - + buffer->hdr.cmd = FW_GET_MODULE_INFO_CMD; + buffer->hdr.buf_len = sizeof(struct txgbe_hic_get_module_info) - sizeof(struct wx_hic_hdr); buffer->hdr.cmd_or_resp.cmd_resv = FW_CEM_CMD_RESERVED; return wx_host_interface_command(wx, (u32 *)buffer, - sizeof(struct txgbe_hic_i2c_read), + sizeof(struct txgbe_hic_get_module_info), WX_HI_COMMAND_TIMEOUT, true); } @@ -109,9 +110,9 @@ static void txgbe_get_link_capabilities(struct wx *wx, int *speed, { struct txgbe *txgbe = wx->priv; - if (test_bit(PHY_INTERFACE_MODE_25GBASER, txgbe->sfp_interfaces)) + if (test_bit(PHY_INTERFACE_MODE_25GBASER, txgbe->link_interfaces)) *speed = SPEED_25000; - else if (test_bit(PHY_INTERFACE_MODE_10GBASER, txgbe->sfp_interfaces)) + else if (test_bit(PHY_INTERFACE_MODE_10GBASER, txgbe->link_interfaces)) *speed = SPEED_10000; else *speed = SPEED_UNKNOWN; @@ -150,7 +151,7 @@ int txgbe_set_phy_link(struct wx *wx) return 0; } -static int txgbe_sfp_to_linkmodes(struct wx *wx, struct txgbe_sfp_id *id) +static int txgbe_sfp_to_linkmodes(struct wx *wx, struct txgbe_sff_id *id) { __ETHTOOL_DECLARE_LINK_MODE_MASK(modes) = { 0, }; DECLARE_PHY_INTERFACE_MASK(interfaces); @@ -204,9 +205,9 @@ static int txgbe_sfp_to_linkmodes(struct wx *wx, struct txgbe_sfp_id *id) phylink_set(modes, Asym_Pause); phylink_set(modes, FIBRE); - if (!linkmode_equal(txgbe->sfp_support, modes)) { - linkmode_copy(txgbe->sfp_support, modes); - phy_interface_and(txgbe->sfp_interfaces, + if (!linkmode_equal(txgbe->link_support, modes)) { + linkmode_copy(txgbe->link_support, modes); + phy_interface_and(txgbe->link_interfaces, wx->phylink_config.supported_interfaces, interfaces); linkmode_copy(txgbe->advertising, modes); @@ -217,10 +218,10 @@ static int txgbe_sfp_to_linkmodes(struct wx *wx, struct txgbe_sfp_id *id) return 0; } -int txgbe_identify_sfp(struct wx *wx) +int txgbe_identify_module(struct wx *wx) { - struct txgbe_hic_i2c_read buffer; - struct txgbe_sfp_id *id; + struct txgbe_hic_get_module_info buffer; + struct txgbe_sff_id *id; int err = 0; u32 gpio; @@ -228,9 +229,9 @@ int txgbe_identify_sfp(struct wx *wx) if (gpio & TXGBE_GPIOBIT_2) return -ENODEV; - err = txgbe_identify_sfp_hostif(wx, &buffer); + err = txgbe_identify_module_hostif(wx, &buffer); if (err) { - wx_err(wx, "Failed to identify SFP module\n"); + wx_err(wx, "Failed to identify module\n"); return err; } @@ -247,10 +248,10 @@ void txgbe_setup_link(struct wx *wx) { struct txgbe *txgbe = wx->priv; - phy_interface_zero(txgbe->sfp_interfaces); - linkmode_zero(txgbe->sfp_support); + phy_interface_zero(txgbe->link_interfaces); + linkmode_zero(txgbe->link_support); - txgbe_identify_sfp(wx); + txgbe_identify_module(wx); } static void txgbe_get_link_state(struct phylink_config *config, diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_aml.h b/drivers/net/ethernet/wangxun/txgbe/txgbe_aml.h index 25d4971ca0d9..7c8fa48e68d3 100644 --- a/drivers/net/ethernet/wangxun/txgbe/txgbe_aml.h +++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_aml.h @@ -8,7 +8,7 @@ void txgbe_gpio_init_aml(struct wx *wx); irqreturn_t txgbe_gpio_irq_handler_aml(int irq, void *data); int txgbe_test_hostif(struct wx *wx); int txgbe_set_phy_link(struct wx *wx); -int txgbe_identify_sfp(struct wx *wx); +int txgbe_identify_module(struct wx *wx); void txgbe_setup_link(struct wx *wx); int txgbe_phylink_init_aml(struct txgbe *txgbe); diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_ethtool.c b/drivers/net/ethernet/wangxun/txgbe/txgbe_ethtool.c index e8dd277a35c7..d7f905359458 100644 --- a/drivers/net/ethernet/wangxun/txgbe/txgbe_ethtool.c +++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_ethtool.c @@ -32,7 +32,7 @@ int txgbe_get_link_ksettings(struct net_device *netdev, cmd->base.port = txgbe->link_port; cmd->base.autoneg = phylink_test(txgbe->advertising, Autoneg) ? AUTONEG_ENABLE : AUTONEG_DISABLE; - linkmode_copy(cmd->link_modes.supported, txgbe->sfp_support); + linkmode_copy(cmd->link_modes.supported, txgbe->link_support); linkmode_copy(cmd->link_modes.advertising, txgbe->advertising); return 0; diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c b/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c index daa761e48f9d..91d1b4e68126 100644 --- a/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c +++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c @@ -89,21 +89,21 @@ static int txgbe_enumerate_functions(struct wx *wx) return physfns; } -static void txgbe_sfp_detection_subtask(struct wx *wx) +static void txgbe_module_detection_subtask(struct wx *wx) { int err; - if (!test_bit(WX_FLAG_NEED_SFP_RESET, wx->flags)) + if (!test_bit(WX_FLAG_NEED_MODULE_RESET, wx->flags)) return; - /* wait for SFP module ready */ + /* wait for SFF module ready */ msleep(200); - err = txgbe_identify_sfp(wx); + err = txgbe_identify_module(wx); if (err) return; - clear_bit(WX_FLAG_NEED_SFP_RESET, wx->flags); + clear_bit(WX_FLAG_NEED_MODULE_RESET, wx->flags); } static void txgbe_link_config_subtask(struct wx *wx) @@ -128,7 +128,7 @@ static void txgbe_service_task(struct work_struct *work) { struct wx *wx = container_of(work, struct wx, service_task); - txgbe_sfp_detection_subtask(wx); + txgbe_module_detection_subtask(wx); txgbe_link_config_subtask(wx); wx_service_event_complete(wx); diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h b/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h index b9a4ba48f5b9..c115ed659544 100644 --- a/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h +++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h @@ -341,9 +341,9 @@ void txgbe_do_reset(struct net_device *netdev); #define FW_PHY_GET_LINK_CMD 0xC0 #define FW_PHY_SET_LINK_CMD 0xC1 -#define FW_READ_SFP_INFO_CMD 0xC5 +#define FW_GET_MODULE_INFO_CMD 0xC5 -struct txgbe_sfp_id { +struct txgbe_sff_id { u8 identifier; /* A0H 0x00 */ u8 com_1g_code; /* A0H 0x06 */ u8 com_10g_code; /* A0H 0x03 */ @@ -358,9 +358,9 @@ struct txgbe_sfp_id { u8 reserved[5]; }; -struct txgbe_hic_i2c_read { +struct txgbe_hic_get_module_info { struct wx_hic_hdr hdr; - struct txgbe_sfp_id id; + struct txgbe_sff_id id; }; struct txgbe_hic_ephy_setlink { @@ -451,8 +451,8 @@ struct txgbe { int fdir_filter_count; spinlock_t fdir_perfect_lock; /* spinlock for FDIR */ - DECLARE_PHY_INTERFACE_MASK(sfp_interfaces); - __ETHTOOL_DECLARE_LINK_MODE_MASK(sfp_support); + DECLARE_PHY_INTERFACE_MASK(link_interfaces); + __ETHTOOL_DECLARE_LINK_MODE_MASK(link_support); __ETHTOOL_DECLARE_LINK_MODE_MASK(advertising); u8 link_port; }; From 57d39faed4c942ccd611588ce0065d1516778aab Mon Sep 17 00:00:00 2001 From: Jiawen Wu Date: Tue, 18 Nov 2025 16:02:57 +0800 Subject: [PATCH 0935/1381] net: txgbe: improve functions of AML 40G devices Support to identify QSFP modules for AML 40G devices. The definition of GPIO pins follows the design of the QSFP modules, and TXGBE_GPIOBIT_4 is used for module present. Meanwhile, implement phylink in XLGMII mode by default, and get the link state from MAC link. Signed-off-by: Jiawen Wu Link: https://patch.msgid.link/20251118080259.24676-4-jiawenwu@trustnetic.com Signed-off-by: Paolo Abeni --- .../net/ethernet/wangxun/libwx/wx_ethtool.c | 12 -- .../net/ethernet/wangxun/txgbe/txgbe_aml.c | 136 ++++++++++++++++-- .../ethernet/wangxun/txgbe/txgbe_ethtool.c | 3 - .../net/ethernet/wangxun/txgbe/txgbe_irq.c | 10 +- .../net/ethernet/wangxun/txgbe/txgbe_main.c | 11 +- .../net/ethernet/wangxun/txgbe/txgbe_phy.c | 2 - .../net/ethernet/wangxun/txgbe/txgbe_type.h | 11 ++ 7 files changed, 137 insertions(+), 48 deletions(-) diff --git a/drivers/net/ethernet/wangxun/libwx/wx_ethtool.c b/drivers/net/ethernet/wangxun/libwx/wx_ethtool.c index 9aa3964187e1..f362e51c73ee 100644 --- a/drivers/net/ethernet/wangxun/libwx/wx_ethtool.c +++ b/drivers/net/ethernet/wangxun/libwx/wx_ethtool.c @@ -240,9 +240,6 @@ int wx_nway_reset(struct net_device *netdev) { struct wx *wx = netdev_priv(netdev); - if (wx->mac.type == wx_mac_aml40) - return -EOPNOTSUPP; - return phylink_ethtool_nway_reset(wx->phylink); } EXPORT_SYMBOL(wx_nway_reset); @@ -261,9 +258,6 @@ int wx_set_link_ksettings(struct net_device *netdev, { struct wx *wx = netdev_priv(netdev); - if (wx->mac.type == wx_mac_aml40) - return -EOPNOTSUPP; - return phylink_ethtool_ksettings_set(wx->phylink, cmd); } EXPORT_SYMBOL(wx_set_link_ksettings); @@ -273,9 +267,6 @@ void wx_get_pauseparam(struct net_device *netdev, { struct wx *wx = netdev_priv(netdev); - if (wx->mac.type == wx_mac_aml40) - return; - phylink_ethtool_get_pauseparam(wx->phylink, pause); } EXPORT_SYMBOL(wx_get_pauseparam); @@ -285,9 +276,6 @@ int wx_set_pauseparam(struct net_device *netdev, { struct wx *wx = netdev_priv(netdev); - if (wx->mac.type == wx_mac_aml40) - return -EOPNOTSUPP; - return phylink_ethtool_set_pauseparam(wx->phylink, pause); } EXPORT_SYMBOL(wx_set_pauseparam); diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_aml.c b/drivers/net/ethernet/wangxun/txgbe/txgbe_aml.c index 432880ccc640..5725e9557669 100644 --- a/drivers/net/ethernet/wangxun/txgbe/txgbe_aml.c +++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_aml.c @@ -17,10 +17,15 @@ void txgbe_gpio_init_aml(struct wx *wx) { - u32 status; + u32 status, mod_rst; - wr32(wx, WX_GPIO_INTTYPE_LEVEL, TXGBE_GPIOBIT_2); - wr32(wx, WX_GPIO_INTEN, TXGBE_GPIOBIT_2); + if (wx->mac.type == wx_mac_aml40) + mod_rst = TXGBE_GPIOBIT_4; + else + mod_rst = TXGBE_GPIOBIT_2; + + wr32(wx, WX_GPIO_INTTYPE_LEVEL, mod_rst); + wr32(wx, WX_GPIO_INTEN, mod_rst); status = rd32(wx, WX_GPIO_INTSTATUS); for (int i = 0; i < 6; i++) { @@ -33,13 +38,18 @@ irqreturn_t txgbe_gpio_irq_handler_aml(int irq, void *data) { struct txgbe *txgbe = data; struct wx *wx = txgbe->wx; - u32 status; + u32 status, mod_rst; + + if (wx->mac.type == wx_mac_aml40) + mod_rst = TXGBE_GPIOBIT_4; + else + mod_rst = TXGBE_GPIOBIT_2; wr32(wx, WX_GPIO_INTMASK, 0xFF); status = rd32(wx, WX_GPIO_INTSTATUS); - if (status & TXGBE_GPIOBIT_2) { + if (status & mod_rst) { set_bit(WX_FLAG_NEED_MODULE_RESET, wx->flags); - wr32(wx, WX_GPIO_EOI, TXGBE_GPIOBIT_2); + wr32(wx, WX_GPIO_EOI, mod_rst); wx_service_event_schedule(wx); } @@ -51,7 +61,7 @@ int txgbe_test_hostif(struct wx *wx) { struct txgbe_hic_ephy_getlink buffer; - if (wx->mac.type != wx_mac_aml) + if (wx->mac.type == wx_mac_sp) return 0; buffer.hdr.cmd = FW_PHY_GET_LINK_CMD; @@ -86,6 +96,9 @@ static int txgbe_set_phy_link_hostif(struct wx *wx, int speed, int autoneg, int buffer.hdr.cmd_or_resp.cmd_resv = FW_CEM_CMD_RESERVED; switch (speed) { + case SPEED_40000: + buffer.speed = TXGBE_LINK_SPEED_40GB_FULL; + break; case SPEED_25000: buffer.speed = TXGBE_LINK_SPEED_25GB_FULL; break; @@ -110,7 +123,9 @@ static void txgbe_get_link_capabilities(struct wx *wx, int *speed, { struct txgbe *txgbe = wx->priv; - if (test_bit(PHY_INTERFACE_MODE_25GBASER, txgbe->link_interfaces)) + if (test_bit(PHY_INTERFACE_MODE_XLGMII, txgbe->link_interfaces)) + *speed = SPEED_40000; + else if (test_bit(PHY_INTERFACE_MODE_25GBASER, txgbe->link_interfaces)) *speed = SPEED_25000; else if (test_bit(PHY_INTERFACE_MODE_10GBASER, txgbe->link_interfaces)) *speed = SPEED_10000; @@ -128,6 +143,8 @@ static void txgbe_get_mac_link(struct wx *wx, int *speed) status = rd32(wx, TXGBE_CFG_PORT_ST); if (!(status & TXGBE_CFG_PORT_ST_LINK_UP)) *speed = SPEED_UNKNOWN; + else if (status & TXGBE_CFG_PORT_ST_LINK_AML_40G) + *speed = SPEED_40000; else if (status & TXGBE_CFG_PORT_ST_LINK_AML_25G) *speed = SPEED_25000; else if (status & TXGBE_CFG_PORT_ST_LINK_AML_10G) @@ -218,15 +235,86 @@ static int txgbe_sfp_to_linkmodes(struct wx *wx, struct txgbe_sff_id *id) return 0; } +static int txgbe_qsfp_to_linkmodes(struct wx *wx, struct txgbe_sff_id *id) +{ + __ETHTOOL_DECLARE_LINK_MODE_MASK(modes) = { 0, }; + DECLARE_PHY_INTERFACE_MASK(interfaces); + struct txgbe *txgbe = wx->priv; + + if (id->transceiver_type & TXGBE_SFF_ETHERNET_40G_CR4) { + txgbe->link_port = PORT_DA; + phylink_set(modes, Autoneg); + phylink_set(modes, 40000baseCR4_Full); + phylink_set(modes, 10000baseCR_Full); + __set_bit(PHY_INTERFACE_MODE_XLGMII, interfaces); + __set_bit(PHY_INTERFACE_MODE_10GBASER, interfaces); + } + if (id->transceiver_type & TXGBE_SFF_ETHERNET_40G_SR4) { + txgbe->link_port = PORT_FIBRE; + phylink_set(modes, 40000baseSR4_Full); + __set_bit(PHY_INTERFACE_MODE_XLGMII, interfaces); + } + if (id->transceiver_type & TXGBE_SFF_ETHERNET_40G_LR4) { + txgbe->link_port = PORT_FIBRE; + phylink_set(modes, 40000baseLR4_Full); + __set_bit(PHY_INTERFACE_MODE_XLGMII, interfaces); + } + if (id->transceiver_type & TXGBE_SFF_ETHERNET_40G_ACTIVE) { + txgbe->link_port = PORT_DA; + phylink_set(modes, Autoneg); + phylink_set(modes, 40000baseCR4_Full); + __set_bit(PHY_INTERFACE_MODE_XLGMII, interfaces); + } + if (id->transceiver_type & TXGBE_SFF_ETHERNET_RSRVD) { + if (id->sff_opt1 & TXGBE_SFF_ETHERNET_100G_CR4) { + txgbe->link_port = PORT_DA; + phylink_set(modes, Autoneg); + phylink_set(modes, 40000baseCR4_Full); + phylink_set(modes, 25000baseCR_Full); + phylink_set(modes, 10000baseCR_Full); + __set_bit(PHY_INTERFACE_MODE_XLGMII, interfaces); + __set_bit(PHY_INTERFACE_MODE_25GBASER, interfaces); + __set_bit(PHY_INTERFACE_MODE_10GBASER, interfaces); + } + } + + if (phy_interface_empty(interfaces)) { + wx_err(wx, "unsupported QSFP module\n"); + return -EINVAL; + } + + phylink_set(modes, Pause); + phylink_set(modes, Asym_Pause); + phylink_set(modes, FIBRE); + + if (!linkmode_equal(txgbe->link_support, modes)) { + linkmode_copy(txgbe->link_support, modes); + phy_interface_and(txgbe->link_interfaces, + wx->phylink_config.supported_interfaces, + interfaces); + linkmode_copy(txgbe->advertising, modes); + + set_bit(WX_FLAG_NEED_LINK_CONFIG, wx->flags); + } + + return 0; +} + int txgbe_identify_module(struct wx *wx) { struct txgbe_hic_get_module_info buffer; struct txgbe_sff_id *id; int err = 0; + u32 mod_abs; u32 gpio; + if (wx->mac.type == wx_mac_aml40) + mod_abs = TXGBE_GPIOBIT_4; + else + mod_abs = TXGBE_GPIOBIT_2; + gpio = rd32(wx, WX_GPIO_EXT); - if (gpio & TXGBE_GPIOBIT_2) + if (gpio & mod_abs) return -ENODEV; err = txgbe_identify_module_hostif(wx, &buffer); @@ -236,12 +324,18 @@ int txgbe_identify_module(struct wx *wx) } id = &buffer.id; - if (id->identifier != TXGBE_SFF_IDENTIFIER_SFP) { - wx_err(wx, "Invalid SFP module\n"); + if (id->identifier != TXGBE_SFF_IDENTIFIER_SFP && + id->identifier != TXGBE_SFF_IDENTIFIER_QSFP && + id->identifier != TXGBE_SFF_IDENTIFIER_QSFP_PLUS && + id->identifier != TXGBE_SFF_IDENTIFIER_QSFP28) { + wx_err(wx, "Invalid module\n"); return -ENODEV; } - return txgbe_sfp_to_linkmodes(wx, id); + if (id->transceiver_type == 0xFF) + return txgbe_sfp_to_linkmodes(wx, id); + + return txgbe_qsfp_to_linkmodes(wx, id); } void txgbe_setup_link(struct wx *wx) @@ -304,6 +398,9 @@ static void txgbe_mac_link_up_aml(struct phylink_config *config, txcfg &= ~TXGBE_AML_MAC_TX_CFG_SPEED_MASK; switch (speed) { + case SPEED_40000: + txcfg |= TXGBE_AML_MAC_TX_CFG_SPEED_40G; + break; case SPEED_25000: txcfg |= TXGBE_AML_MAC_TX_CFG_SPEED_25G; break; @@ -368,7 +465,18 @@ int txgbe_phylink_init_aml(struct txgbe *txgbe) MAC_SYM_PAUSE | MAC_ASYM_PAUSE; config->get_fixed_state = txgbe_get_link_state; - phy_mode = PHY_INTERFACE_MODE_25GBASER; + if (wx->mac.type == wx_mac_aml40) { + config->mac_capabilities |= MAC_40000FD; + phy_mode = PHY_INTERFACE_MODE_XLGMII; + __set_bit(PHY_INTERFACE_MODE_XLGMII, config->supported_interfaces); + state.speed = SPEED_40000; + state.duplex = DUPLEX_FULL; + } else { + phy_mode = PHY_INTERFACE_MODE_25GBASER; + state.speed = SPEED_25000; + state.duplex = DUPLEX_FULL; + } + __set_bit(PHY_INTERFACE_MODE_25GBASER, config->supported_interfaces); __set_bit(PHY_INTERFACE_MODE_10GBASER, config->supported_interfaces); @@ -376,8 +484,6 @@ int txgbe_phylink_init_aml(struct txgbe *txgbe) if (IS_ERR(phylink)) return PTR_ERR(phylink); - state.speed = SPEED_25000; - state.duplex = DUPLEX_FULL; err = phylink_set_fixed_link(phylink, &state); if (err) { wx_err(wx, "Failed to set fixed link\n"); diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_ethtool.c b/drivers/net/ethernet/wangxun/txgbe/txgbe_ethtool.c index d7f905359458..f553ec5f8238 100644 --- a/drivers/net/ethernet/wangxun/txgbe/txgbe_ethtool.c +++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_ethtool.c @@ -19,9 +19,6 @@ int txgbe_get_link_ksettings(struct net_device *netdev, struct txgbe *txgbe = wx->priv; int err; - if (wx->mac.type == wx_mac_aml40) - return -EOPNOTSUPP; - err = wx_get_link_ksettings(netdev, cmd); if (err) return err; diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_irq.c b/drivers/net/ethernet/wangxun/txgbe/txgbe_irq.c index 3885283681ec..aa14958d439a 100644 --- a/drivers/net/ethernet/wangxun/txgbe/txgbe_irq.c +++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_irq.c @@ -23,7 +23,7 @@ void txgbe_irq_enable(struct wx *wx, bool queues) { u32 misc_ien = TXGBE_PX_MISC_IEN_MASK; - if (wx->mac.type == wx_mac_aml) { + if (wx->mac.type != wx_mac_sp) { misc_ien |= TXGBE_PX_MISC_GPIO; txgbe_gpio_init_aml(wx); } @@ -201,10 +201,7 @@ static void txgbe_del_irq_domain(struct txgbe *txgbe) void txgbe_free_misc_irq(struct txgbe *txgbe) { - if (txgbe->wx->mac.type == wx_mac_aml40) - return; - - if (txgbe->wx->mac.type == wx_mac_aml) + if (txgbe->wx->mac.type != wx_mac_sp) free_irq(txgbe->gpio_irq, txgbe); free_irq(txgbe->link_irq, txgbe); @@ -219,9 +216,6 @@ int txgbe_setup_misc_irq(struct txgbe *txgbe) struct wx *wx = txgbe->wx; int hwirq, err; - if (wx->mac.type == wx_mac_aml40) - goto skip_sp_irq; - txgbe->misc.nirqs = TXGBE_IRQ_MAX; txgbe->misc.domain = irq_domain_create_simple(NULL, txgbe->misc.nirqs, 0, &txgbe_misc_irq_domain_ops, txgbe); diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c b/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c index 91d1b4e68126..0de051450a82 100644 --- a/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c +++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c @@ -144,7 +144,6 @@ static void txgbe_init_service(struct wx *wx) static void txgbe_up_complete(struct wx *wx) { struct net_device *netdev = wx->netdev; - u32 reg; wx_control_hw(wx, true); wx_configure_vectors(wx); @@ -155,12 +154,8 @@ static void txgbe_up_complete(struct wx *wx) switch (wx->mac.type) { case wx_mac_aml40: - reg = rd32(wx, TXGBE_AML_MAC_TX_CFG); - reg &= ~TXGBE_AML_MAC_TX_CFG_SPEED_MASK; - reg |= TXGBE_AML_MAC_TX_CFG_SPEED_40G; - wr32(wx, WX_MAC_TX_CFG, reg); - txgbe_enable_sec_tx_path(wx); - netif_carrier_on(wx->netdev); + txgbe_setup_link(wx); + phylink_start(wx->phylink); break; case wx_mac_aml: /* Enable TX laser */ @@ -276,7 +271,7 @@ void txgbe_down(struct wx *wx) switch (wx->mac.type) { case wx_mac_aml40: - netif_carrier_off(wx->netdev); + phylink_stop(wx->phylink); break; case wx_mac_aml: phylink_stop(wx->phylink); diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_phy.c b/drivers/net/ethernet/wangxun/txgbe/txgbe_phy.c index 03f1b9bc604d..8ea7aa07ae4e 100644 --- a/drivers/net/ethernet/wangxun/txgbe/txgbe_phy.c +++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_phy.c @@ -579,7 +579,6 @@ int txgbe_init_phy(struct txgbe *txgbe) switch (wx->mac.type) { case wx_mac_aml40: - return 0; case wx_mac_aml: return txgbe_phylink_init_aml(txgbe); case wx_mac_sp: @@ -653,7 +652,6 @@ void txgbe_remove_phy(struct txgbe *txgbe) { switch (txgbe->wx->mac.type) { case wx_mac_aml40: - return; case wx_mac_aml: phylink_destroy(txgbe->wx->phylink); return; diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h b/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h index c115ed659544..e72edb9ef084 100644 --- a/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h +++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h @@ -98,6 +98,7 @@ /* Port cfg registers */ #define TXGBE_CFG_PORT_ST 0x14404 #define TXGBE_CFG_PORT_ST_LINK_UP BIT(0) +#define TXGBE_CFG_PORT_ST_LINK_AML_40G BIT(2) #define TXGBE_CFG_PORT_ST_LINK_AML_25G BIT(3) #define TXGBE_CFG_PORT_ST_LINK_AML_10G BIT(4) #define TXGBE_CFG_VXLAN 0x14410 @@ -317,8 +318,12 @@ void txgbe_do_reset(struct net_device *netdev); #define TXGBE_LINK_SPEED_UNKNOWN 0 #define TXGBE_LINK_SPEED_10GB_FULL 4 #define TXGBE_LINK_SPEED_25GB_FULL 0x10 +#define TXGBE_LINK_SPEED_40GB_FULL 0x20 #define TXGBE_SFF_IDENTIFIER_SFP 0x3 +#define TXGBE_SFF_IDENTIFIER_QSFP 0xC +#define TXGBE_SFF_IDENTIFIER_QSFP_PLUS 0xD +#define TXGBE_SFF_IDENTIFIER_QSFP28 0x11 #define TXGBE_SFF_DA_PASSIVE_CABLE 0x4 #define TXGBE_SFF_DA_ACTIVE_CABLE 0x8 #define TXGBE_SFF_DA_SPEC_ACTIVE_LIMIT 0x4 @@ -331,6 +336,12 @@ void txgbe_do_reset(struct net_device *netdev); #define TXGBE_SFF_25GBASECR_91FEC 0xB #define TXGBE_SFF_25GBASECR_74FEC 0xC #define TXGBE_SFF_25GBASECR_NOFEC 0xD +#define TXGBE_SFF_ETHERNET_RSRVD BIT(7) +#define TXGBE_SFF_ETHERNET_40G_CR4 BIT(3) +#define TXGBE_SFF_ETHERNET_40G_SR4 BIT(2) +#define TXGBE_SFF_ETHERNET_40G_LR4 BIT(1) +#define TXGBE_SFF_ETHERNET_40G_ACTIVE BIT(0) +#define TXGBE_SFF_ETHERNET_100G_CR4 0xB #define TXGBE_PHY_FEC_RS BIT(0) #define TXGBE_PHY_FEC_BASER BIT(1) From c6e97daec549c2c01434d2978ae48b1a2f71a893 Mon Sep 17 00:00:00 2001 From: Jiawen Wu Date: Tue, 18 Nov 2025 16:02:58 +0800 Subject: [PATCH 0936/1381] net: txgbe: delay to identify modules in .ndo_open For QSFP modules, there is a possibility that the module cannot be identified when read I2C immediately in .ndo_open. So just set the flag WX_FLAG_NEED_MODULE_RESET and do it in the subtask, which always wait 200 ms to identify the module. And this change has no impact on the original adaptation. Signed-off-by: Jiawen Wu Link: https://patch.msgid.link/20251118080259.24676-5-jiawenwu@trustnetic.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/wangxun/txgbe/txgbe_aml.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_aml.c b/drivers/net/ethernet/wangxun/txgbe/txgbe_aml.c index 5725e9557669..3b6ea456fbf7 100644 --- a/drivers/net/ethernet/wangxun/txgbe/txgbe_aml.c +++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_aml.c @@ -345,7 +345,8 @@ void txgbe_setup_link(struct wx *wx) phy_interface_zero(txgbe->link_interfaces); linkmode_zero(txgbe->link_support); - txgbe_identify_module(wx); + set_bit(WX_FLAG_NEED_MODULE_RESET, wx->flags); + wx_service_event_schedule(wx); } static void txgbe_get_link_state(struct phylink_config *config, From 9b97b6b5635b3de18debd07a3adcf25198ec8299 Mon Sep 17 00:00:00 2001 From: Jiawen Wu Date: Tue, 18 Nov 2025 16:02:59 +0800 Subject: [PATCH 0937/1381] net: txgbe: support getting module EEPROM by page Getting module EEPROM has been supported in TXGBE SP devices, since SFP driver has already implemented it. Now add support to read module EEPROM for AML devices. Towards this, add a new firmware mailbox command to get the page data. Signed-off-by: Jiawen Wu Link: https://patch.msgid.link/20251118080259.24676-6-jiawenwu@trustnetic.com Signed-off-by: Paolo Abeni --- .../net/ethernet/wangxun/txgbe/txgbe_aml.c | 33 +++++++++++++++++++ .../net/ethernet/wangxun/txgbe/txgbe_aml.h | 3 ++ .../ethernet/wangxun/txgbe/txgbe_ethtool.c | 30 +++++++++++++++++ .../net/ethernet/wangxun/txgbe/txgbe_type.h | 11 +++++++ 4 files changed, 77 insertions(+) diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_aml.c b/drivers/net/ethernet/wangxun/txgbe/txgbe_aml.c index 3b6ea456fbf7..62d7f47d4f8d 100644 --- a/drivers/net/ethernet/wangxun/txgbe/txgbe_aml.c +++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_aml.c @@ -73,6 +73,39 @@ int txgbe_test_hostif(struct wx *wx) WX_HI_COMMAND_TIMEOUT, true); } +int txgbe_read_eeprom_hostif(struct wx *wx, + struct txgbe_hic_i2c_read *buffer, + u32 length, u8 *data) +{ + u32 dword_len, offset, value, i; + int err; + + buffer->hdr.cmd = FW_READ_EEPROM_CMD; + buffer->hdr.buf_len = sizeof(struct txgbe_hic_i2c_read) - + sizeof(struct wx_hic_hdr); + buffer->hdr.cmd_or_resp.cmd_resv = FW_CEM_CMD_RESERVED; + + err = wx_host_interface_command(wx, (u32 *)buffer, + sizeof(struct txgbe_hic_i2c_read), + WX_HI_COMMAND_TIMEOUT, false); + if (err != 0) + return err; + + /* buffer length offset to read return data */ + offset = sizeof(struct txgbe_hic_i2c_read) >> 2; + dword_len = round_up(length, 4) >> 2; + + for (i = 0; i < dword_len; i++) { + value = rd32a(wx, WX_FW2SW_MBOX, i + offset); + le32_to_cpus(&value); + + memcpy(data, &value, 4); + data += 4; + } + + return 0; +} + static int txgbe_identify_module_hostif(struct wx *wx, struct txgbe_hic_get_module_info *buffer) { diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_aml.h b/drivers/net/ethernet/wangxun/txgbe/txgbe_aml.h index 7c8fa48e68d3..4f6df0ee860b 100644 --- a/drivers/net/ethernet/wangxun/txgbe/txgbe_aml.h +++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_aml.h @@ -7,6 +7,9 @@ void txgbe_gpio_init_aml(struct wx *wx); irqreturn_t txgbe_gpio_irq_handler_aml(int irq, void *data); int txgbe_test_hostif(struct wx *wx); +int txgbe_read_eeprom_hostif(struct wx *wx, + struct txgbe_hic_i2c_read *buffer, + u32 length, u8 *data); int txgbe_set_phy_link(struct wx *wx); int txgbe_identify_module(struct wx *wx); void txgbe_setup_link(struct wx *wx); diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_ethtool.c b/drivers/net/ethernet/wangxun/txgbe/txgbe_ethtool.c index f553ec5f8238..f3cb00109529 100644 --- a/drivers/net/ethernet/wangxun/txgbe/txgbe_ethtool.c +++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_ethtool.c @@ -10,6 +10,7 @@ #include "../libwx/wx_lib.h" #include "txgbe_type.h" #include "txgbe_fdir.h" +#include "txgbe_aml.h" #include "txgbe_ethtool.h" int txgbe_get_link_ksettings(struct net_device *netdev, @@ -534,6 +535,34 @@ static int txgbe_set_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd) return ret; } +static int +txgbe_get_module_eeprom_by_page(struct net_device *netdev, + const struct ethtool_module_eeprom *page_data, + struct netlink_ext_ack *extack) +{ + struct wx *wx = netdev_priv(netdev); + struct txgbe_hic_i2c_read buffer; + int err; + + if (!test_bit(WX_FLAG_SWFW_RING, wx->flags)) + return -EOPNOTSUPP; + + buffer.length = cpu_to_be32(page_data->length); + buffer.offset = cpu_to_be32(page_data->offset); + buffer.page = page_data->page; + buffer.bank = page_data->bank; + buffer.i2c_address = page_data->i2c_address; + + err = txgbe_read_eeprom_hostif(wx, &buffer, page_data->length, + page_data->data); + if (err) { + wx_err(wx, "Failed to read module EEPROM\n"); + return err; + } + + return page_data->length; +} + static const struct ethtool_ops txgbe_ethtool_ops = { .supported_coalesce_params = ETHTOOL_COALESCE_USECS | ETHTOOL_COALESCE_TX_MAX_FRAMES_IRQ | @@ -568,6 +597,7 @@ static const struct ethtool_ops txgbe_ethtool_ops = { .set_msglevel = wx_set_msglevel, .get_ts_info = wx_get_ts_info, .get_ts_stats = wx_get_ptp_stats, + .get_module_eeprom_by_page = txgbe_get_module_eeprom_by_page, }; void txgbe_set_ethtool_ops(struct net_device *netdev) diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h b/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h index e72edb9ef084..82433e9cb0e3 100644 --- a/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h +++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h @@ -353,6 +353,7 @@ void txgbe_do_reset(struct net_device *netdev); #define FW_PHY_GET_LINK_CMD 0xC0 #define FW_PHY_SET_LINK_CMD 0xC1 #define FW_GET_MODULE_INFO_CMD 0xC5 +#define FW_READ_EEPROM_CMD 0xC6 struct txgbe_sff_id { u8 identifier; /* A0H 0x00 */ @@ -394,6 +395,16 @@ struct txgbe_hic_ephy_getlink { u8 resv[6]; }; +struct txgbe_hic_i2c_read { + struct wx_hic_hdr hdr; + __be32 offset; + __be32 length; + u8 page; + u8 bank; + u8 i2c_address; + u8 resv; +}; + #define NODE_PROP(_NAME, _PROP) \ (const struct software_node) { \ .name = _NAME, \ From 41dde7f1d01343ebebe265edd8f4b59bca591db8 Mon Sep 17 00:00:00 2001 From: Meghana Malladi Date: Tue, 18 Nov 2025 19:25:37 +0530 Subject: [PATCH 0938/1381] net: ti: icssg-prueth: Add functions to create and destroy Rx/Tx queues Each port for a given ICSSG instance has their own set of Tx and Rx queues. Add functions to create and destroy these queues, which will be further used while performing ndo_bpf operations to set up XSK Tx/Rx queues for a given port. In the destroy Rx queue sequence add teardown wait to ensure that all the descriptors including the TDCM (teardown completion marker) have been serviced and freed to avoid any sort of descriptor leaks. Reviewed-by: Jacob Keller Signed-off-by: Meghana Malladi Link: https://patch.msgid.link/20251118135542.380574-2-m-malladi@ti.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/ti/icssg/icssg_common.c | 10 +- drivers/net/ethernet/ti/icssg/icssg_prueth.c | 177 ++++++++++++------- drivers/net/ethernet/ti/icssg/icssg_prueth.h | 2 + 3 files changed, 127 insertions(+), 62 deletions(-) diff --git a/drivers/net/ethernet/ti/icssg/icssg_common.c b/drivers/net/ethernet/ti/icssg/icssg_common.c index 0eed29d6187a..94021751b6b7 100644 --- a/drivers/net/ethernet/ti/icssg/icssg_common.c +++ b/drivers/net/ethernet/ti/icssg/icssg_common.c @@ -719,8 +719,10 @@ static int emac_rx_packet(struct prueth_emac *emac, u32 flow_id, u32 *xdp_state) return ret; } - if (cppi5_desc_is_tdcm(desc_dma)) /* Teardown ? */ + if (cppi5_desc_is_tdcm(desc_dma)) { + complete(&emac->tdown_complete); return 0; + } desc_rx = k3_cppi_desc_pool_dma2virt(rx_chn->desc_pool, desc_dma); swdata = cppi5_hdesc_get_swdata(desc_rx); @@ -804,7 +806,7 @@ static int emac_rx_packet(struct prueth_emac *emac, u32 flow_id, u32 *xdp_state) return ret; } -static void prueth_rx_cleanup(void *data, dma_addr_t desc_dma) +void prueth_rx_cleanup(void *data, dma_addr_t desc_dma) { struct prueth_rx_chn *rx_chn = data; struct cppi5_host_desc_t *desc_rx; @@ -822,6 +824,7 @@ static void prueth_rx_cleanup(void *data, dma_addr_t desc_dma) k3_cppi_desc_pool_free(rx_chn->desc_pool, desc_rx); } +EXPORT_SYMBOL_GPL(prueth_rx_cleanup); static int prueth_tx_ts_cookie_get(struct prueth_emac *emac) { @@ -1025,7 +1028,7 @@ enum netdev_tx icssg_ndo_start_xmit(struct sk_buff *skb, struct net_device *ndev } EXPORT_SYMBOL_GPL(icssg_ndo_start_xmit); -static void prueth_tx_cleanup(void *data, dma_addr_t desc_dma) +void prueth_tx_cleanup(void *data, dma_addr_t desc_dma) { struct prueth_tx_chn *tx_chn = data; struct cppi5_host_desc_t *desc_tx; @@ -1051,6 +1054,7 @@ static void prueth_tx_cleanup(void *data, dma_addr_t desc_dma) prueth_xmit_free(tx_chn, desc_tx); } +EXPORT_SYMBOL_GPL(prueth_tx_cleanup); irqreturn_t prueth_rx_irq(int irq, void *dev_id) { diff --git a/drivers/net/ethernet/ti/icssg/icssg_prueth.c b/drivers/net/ethernet/ti/icssg/icssg_prueth.c index 57a7d1ceab08..b66ffbfb499c 100644 --- a/drivers/net/ethernet/ti/icssg/icssg_prueth.c +++ b/drivers/net/ethernet/ti/icssg/icssg_prueth.c @@ -735,6 +735,114 @@ static int icssg_update_vlan_mcast(struct net_device *vdev, int vid, return 0; } +static void prueth_destroy_txq(struct prueth_emac *emac) +{ + int ret, i; + + atomic_set(&emac->tdown_cnt, emac->tx_ch_num); + /* ensure new tdown_cnt value is visible */ + smp_mb__after_atomic(); + /* tear down and disable UDMA channels */ + reinit_completion(&emac->tdown_complete); + for (i = 0; i < emac->tx_ch_num; i++) + k3_udma_glue_tdown_tx_chn(emac->tx_chns[i].tx_chn, false); + + ret = wait_for_completion_timeout(&emac->tdown_complete, + msecs_to_jiffies(1000)); + if (!ret) + netdev_err(emac->ndev, "tx teardown timeout\n"); + + for (i = 0; i < emac->tx_ch_num; i++) { + napi_disable(&emac->tx_chns[i].napi_tx); + hrtimer_cancel(&emac->tx_chns[i].tx_hrtimer); + k3_udma_glue_reset_tx_chn(emac->tx_chns[i].tx_chn, + &emac->tx_chns[i], + prueth_tx_cleanup); + k3_udma_glue_disable_tx_chn(emac->tx_chns[i].tx_chn); + } +} + +static void prueth_destroy_rxq(struct prueth_emac *emac) +{ + int i, ret; + + /* tear down and disable UDMA channels */ + reinit_completion(&emac->tdown_complete); + k3_udma_glue_tdown_rx_chn(emac->rx_chns.rx_chn, true); + + /* When RX DMA Channel Teardown is initiated, it will result in an + * interrupt and a Teardown Completion Marker (TDCM) is queued into + * the RX Completion queue. Acknowledging the interrupt involves + * popping the TDCM descriptor from the RX Completion queue via the + * RX NAPI Handler. To avoid timing out when waiting for the TDCM to + * be popped, schedule the RX NAPI handler to run immediately. + */ + if (!napi_if_scheduled_mark_missed(&emac->napi_rx)) { + if (napi_schedule_prep(&emac->napi_rx)) + __napi_schedule(&emac->napi_rx); + } + + ret = wait_for_completion_timeout(&emac->tdown_complete, + msecs_to_jiffies(1000)); + if (!ret) + netdev_err(emac->ndev, "rx teardown timeout\n"); + + for (i = 0; i < PRUETH_MAX_RX_FLOWS; i++) { + napi_disable(&emac->napi_rx); + hrtimer_cancel(&emac->rx_hrtimer); + k3_udma_glue_reset_rx_chn(emac->rx_chns.rx_chn, i, + &emac->rx_chns, + prueth_rx_cleanup); + } + + prueth_destroy_xdp_rxqs(emac); + k3_udma_glue_disable_rx_chn(emac->rx_chns.rx_chn); +} + +static int prueth_create_txq(struct prueth_emac *emac) +{ + int ret, i; + + for (i = 0; i < emac->tx_ch_num; i++) { + ret = k3_udma_glue_enable_tx_chn(emac->tx_chns[i].tx_chn); + if (ret) + goto reset_tx_chan; + napi_enable(&emac->tx_chns[i].napi_tx); + } + return 0; + +reset_tx_chan: + /* Since interface is not yet up, there is wouldn't be + * any SKB for completion. So set false to free_skb + */ + prueth_reset_tx_chan(emac, i, false); + return ret; +} + +static int prueth_create_rxq(struct prueth_emac *emac) +{ + int ret; + + ret = prueth_prepare_rx_chan(emac, &emac->rx_chns, PRUETH_MAX_PKT_SIZE); + if (ret) + return ret; + + ret = k3_udma_glue_enable_rx_chn(emac->rx_chns.rx_chn); + if (ret) + goto reset_rx_chn; + + ret = prueth_create_xdp_rxqs(emac); + if (ret) + goto reset_rx_chn; + + napi_enable(&emac->napi_rx); + return 0; + +reset_rx_chn: + prueth_reset_rx_chan(&emac->rx_chns, PRUETH_MAX_RX_FLOWS, false); + return ret; +} + /** * emac_ndo_open - EMAC device open * @ndev: network adapter device @@ -746,7 +854,7 @@ static int icssg_update_vlan_mcast(struct net_device *vdev, int vid, static int emac_ndo_open(struct net_device *ndev) { struct prueth_emac *emac = netdev_priv(ndev); - int ret, i, num_data_chn = emac->tx_ch_num; + int ret, num_data_chn = emac->tx_ch_num; struct icssg_flow_cfg __iomem *flow_cfg; struct prueth *prueth = emac->prueth; int slice = prueth_emac_slice(emac); @@ -819,28 +927,13 @@ static int emac_ndo_open(struct net_device *ndev) goto stop; /* Prepare RX */ - ret = prueth_prepare_rx_chan(emac, &emac->rx_chns, PRUETH_MAX_PKT_SIZE); + ret = prueth_create_rxq(emac); if (ret) goto free_tx_ts_irq; - ret = prueth_create_xdp_rxqs(emac); + ret = prueth_create_txq(emac); if (ret) - goto reset_rx_chn; - - ret = k3_udma_glue_enable_rx_chn(emac->rx_chns.rx_chn); - if (ret) - goto destroy_xdp_rxqs; - - for (i = 0; i < emac->tx_ch_num; i++) { - ret = k3_udma_glue_enable_tx_chn(emac->tx_chns[i].tx_chn); - if (ret) - goto reset_tx_chan; - } - - /* Enable NAPI in Tx and Rx direction */ - for (i = 0; i < emac->tx_ch_num; i++) - napi_enable(&emac->tx_chns[i].napi_tx); - napi_enable(&emac->napi_rx); + goto destroy_rxq; /* start PHY */ phy_start(ndev->phydev); @@ -851,15 +944,8 @@ static int emac_ndo_open(struct net_device *ndev) return 0; -reset_tx_chan: - /* Since interface is not yet up, there is wouldn't be - * any SKB for completion. So set false to free_skb - */ - prueth_reset_tx_chan(emac, i, false); -destroy_xdp_rxqs: - prueth_destroy_xdp_rxqs(emac); -reset_rx_chn: - prueth_reset_rx_chan(&emac->rx_chns, max_rx_flows, false); +destroy_rxq: + prueth_destroy_rxq(emac); free_tx_ts_irq: free_irq(emac->tx_ts_irq, emac); stop: @@ -889,9 +975,6 @@ static int emac_ndo_stop(struct net_device *ndev) { struct prueth_emac *emac = netdev_priv(ndev); struct prueth *prueth = emac->prueth; - int rx_flow = PRUETH_RX_FLOW_DATA; - int max_rx_flows; - int ret, i; /* inform the upper layers. */ netif_tx_stop_all_queues(ndev); @@ -905,32 +988,8 @@ static int emac_ndo_stop(struct net_device *ndev) else __dev_mc_unsync(ndev, icssg_prueth_del_mcast); - atomic_set(&emac->tdown_cnt, emac->tx_ch_num); - /* ensure new tdown_cnt value is visible */ - smp_mb__after_atomic(); - /* tear down and disable UDMA channels */ - reinit_completion(&emac->tdown_complete); - for (i = 0; i < emac->tx_ch_num; i++) - k3_udma_glue_tdown_tx_chn(emac->tx_chns[i].tx_chn, false); - - ret = wait_for_completion_timeout(&emac->tdown_complete, - msecs_to_jiffies(1000)); - if (!ret) - netdev_err(ndev, "tx teardown timeout\n"); - - prueth_reset_tx_chan(emac, emac->tx_ch_num, true); - for (i = 0; i < emac->tx_ch_num; i++) { - napi_disable(&emac->tx_chns[i].napi_tx); - hrtimer_cancel(&emac->tx_chns[i].tx_hrtimer); - } - - max_rx_flows = PRUETH_MAX_RX_FLOWS; - k3_udma_glue_tdown_rx_chn(emac->rx_chns.rx_chn, true); - - prueth_reset_rx_chan(&emac->rx_chns, max_rx_flows, true); - prueth_destroy_xdp_rxqs(emac); - napi_disable(&emac->napi_rx); - hrtimer_cancel(&emac->rx_hrtimer); + prueth_destroy_txq(emac); + prueth_destroy_rxq(emac); cancel_work_sync(&emac->rx_mode_work); @@ -943,10 +1002,10 @@ static int emac_ndo_stop(struct net_device *ndev) free_irq(emac->tx_ts_irq, emac); - free_irq(emac->rx_chns.irq[rx_flow], emac); + free_irq(emac->rx_chns.irq[PRUETH_RX_FLOW_DATA], emac); prueth_ndev_del_tx_napi(emac, emac->tx_ch_num); - prueth_cleanup_rx_chns(emac, &emac->rx_chns, max_rx_flows); + prueth_cleanup_rx_chns(emac, &emac->rx_chns, PRUETH_MAX_RX_FLOWS); prueth_cleanup_tx_chns(emac); prueth->emacs_initialized--; diff --git a/drivers/net/ethernet/ti/icssg/icssg_prueth.h b/drivers/net/ethernet/ti/icssg/icssg_prueth.h index f0fa9688d9a0..5cc90b66035a 100644 --- a/drivers/net/ethernet/ti/icssg/icssg_prueth.h +++ b/drivers/net/ethernet/ti/icssg/icssg_prueth.h @@ -501,5 +501,7 @@ u32 emac_xmit_xdp_frame(struct prueth_emac *emac, struct xdp_frame *xdpf, struct page *page, unsigned int q_idx); +void prueth_rx_cleanup(void *data, dma_addr_t desc_dma); +void prueth_tx_cleanup(void *data, dma_addr_t desc_dma); #endif /* __NET_TI_ICSSG_PRUETH_H */ From 7dfd7597911f52c4d2e0e7b8c710f726ecc9d87f Mon Sep 17 00:00:00 2001 From: Meghana Malladi Date: Tue, 18 Nov 2025 19:25:38 +0530 Subject: [PATCH 0939/1381] net: ti: icssg-prueth: Add XSK pool helpers Implement XSK NDOs (setup, wakeup) and create XSK Rx and Tx queues. xsk_qid stores the queue id for a given port which has been registered for zero copy AF_XDP and used to acquire UMEM pointer if registered. Based on the xsk_qid and the xsk_pool (umem) the driver is either in copy or zero copy mode. In case of copy mode the xsk_qid value will be invalid and will be set to valid queue id when enabling zero copy. To enable zero copy, the Rx queues are destroyed, i.e., descriptors pushed to fq and cq are freed to remap them to xdp buffers from the umem. Reviewed-by: Jacob Keller Signed-off-by: Meghana Malladi Link: https://patch.msgid.link/20251118135542.380574-3-m-malladi@ti.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/ti/icssg/icssg_common.c | 2 +- drivers/net/ethernet/ti/icssg/icssg_prueth.c | 138 +++++++++++++++++++ drivers/net/ethernet/ti/icssg/icssg_prueth.h | 10 ++ 3 files changed, 149 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/ti/icssg/icssg_common.c b/drivers/net/ethernet/ti/icssg/icssg_common.c index 94021751b6b7..cc52cff70d7e 100644 --- a/drivers/net/ethernet/ti/icssg/icssg_common.c +++ b/drivers/net/ethernet/ti/icssg/icssg_common.c @@ -754,7 +754,7 @@ static int emac_rx_packet(struct prueth_emac *emac, u32 flow_id, u32 *xdp_state) } pa = page_address(page); - if (emac->xdp_prog) { + if (prueth_xdp_is_enabled(emac)) { xdp_init_buff(&xdp, PAGE_SIZE, &rx_chn->xdp_rxq); xdp_prepare_buff(&xdp, pa, PRUETH_HEADROOM, pkt_len, false); diff --git a/drivers/net/ethernet/ti/icssg/icssg_prueth.c b/drivers/net/ethernet/ti/icssg/icssg_prueth.c index b66ffbfb499c..e4c3b6b152ea 100644 --- a/drivers/net/ethernet/ti/icssg/icssg_prueth.c +++ b/drivers/net/ethernet/ti/icssg/icssg_prueth.c @@ -47,6 +47,9 @@ NETIF_F_HW_HSR_TAG_INS | \ NETIF_F_HW_HSR_TAG_RM) +#define PRUETH_RX_DMA_ATTR (DMA_ATTR_SKIP_CPU_SYNC |\ + DMA_ATTR_WEAK_ORDERING) + /* CTRLMMR_ICSSG_RGMII_CTRL register bits */ #define ICSSG_CTRL_RGMII_ID_MODE BIT(24) @@ -735,6 +738,20 @@ static int icssg_update_vlan_mcast(struct net_device *vdev, int vid, return 0; } +static void prueth_set_xsk_pool(struct prueth_emac *emac, u16 queue_id) +{ + struct prueth_tx_chn *tx_chn = &emac->tx_chns[queue_id]; + struct prueth_rx_chn *rx_chn = &emac->rx_chns; + + if (emac->xsk_qid != queue_id) { + rx_chn->xsk_pool = NULL; + tx_chn->xsk_pool = NULL; + } else { + rx_chn->xsk_pool = xsk_get_pool_from_qid(emac->ndev, queue_id); + tx_chn->xsk_pool = xsk_get_pool_from_qid(emac->ndev, queue_id); + } +} + static void prueth_destroy_txq(struct prueth_emac *emac) { int ret, i; @@ -875,6 +892,7 @@ static int emac_ndo_open(struct net_device *ndev) return ret; } + emac->xsk_qid = -EINVAL; init_completion(&emac->cmd_complete); ret = prueth_init_tx_chns(emac); if (ret) { @@ -1200,6 +1218,109 @@ static int emac_xdp_setup(struct prueth_emac *emac, struct netdev_bpf *bpf) return 0; } +static int prueth_xsk_pool_enable(struct prueth_emac *emac, + struct xsk_buff_pool *pool, u16 queue_id) +{ + struct prueth_rx_chn *rx_chn = &emac->rx_chns; + u32 frame_size; + int ret; + + if (queue_id >= PRUETH_MAX_RX_FLOWS || + queue_id >= emac->tx_ch_num) { + netdev_err(emac->ndev, "Invalid XSK queue ID %d\n", queue_id); + return -EINVAL; + } + + frame_size = xsk_pool_get_rx_frame_size(pool); + if (frame_size < PRUETH_MAX_PKT_SIZE) + return -EOPNOTSUPP; + + ret = xsk_pool_dma_map(pool, rx_chn->dma_dev, PRUETH_RX_DMA_ATTR); + if (ret) { + netdev_err(emac->ndev, "Failed to map XSK pool: %d\n", ret); + return ret; + } + + if (netif_running(emac->ndev)) { + /* stop packets from wire for graceful teardown */ + ret = icssg_set_port_state(emac, ICSSG_EMAC_PORT_DISABLE); + if (ret) + return ret; + prueth_destroy_rxq(emac); + } + + emac->xsk_qid = queue_id; + prueth_set_xsk_pool(emac, queue_id); + + if (netif_running(emac->ndev)) { + ret = prueth_create_rxq(emac); + if (ret) { + netdev_err(emac->ndev, "Failed to create RX queue: %d\n", ret); + return ret; + } + ret = icssg_set_port_state(emac, ICSSG_EMAC_PORT_FORWARD); + if (ret) { + prueth_destroy_rxq(emac); + return ret; + } + ret = prueth_xsk_wakeup(emac->ndev, queue_id, XDP_WAKEUP_RX); + if (ret) + return ret; + } + + return 0; +} + +static int prueth_xsk_pool_disable(struct prueth_emac *emac, u16 queue_id) +{ + struct xsk_buff_pool *pool; + int ret; + + if (queue_id >= PRUETH_MAX_RX_FLOWS || + queue_id >= emac->tx_ch_num) { + netdev_err(emac->ndev, "Invalid XSK queue ID %d\n", queue_id); + return -EINVAL; + } + + if (emac->xsk_qid != queue_id) { + netdev_err(emac->ndev, "XSK queue ID %d not registered\n", queue_id); + return -EINVAL; + } + + pool = xsk_get_pool_from_qid(emac->ndev, queue_id); + if (!pool) { + netdev_err(emac->ndev, "No XSK pool registered for queue %d\n", queue_id); + return -EINVAL; + } + + if (netif_running(emac->ndev)) { + /* stop packets from wire for graceful teardown */ + ret = icssg_set_port_state(emac, ICSSG_EMAC_PORT_DISABLE); + if (ret) + return ret; + prueth_destroy_rxq(emac); + } + + xsk_pool_dma_unmap(pool, PRUETH_RX_DMA_ATTR); + emac->xsk_qid = -EINVAL; + prueth_set_xsk_pool(emac, queue_id); + + if (netif_running(emac->ndev)) { + ret = prueth_create_rxq(emac); + if (ret) { + netdev_err(emac->ndev, "Failed to create RX queue: %d\n", ret); + return ret; + } + ret = icssg_set_port_state(emac, ICSSG_EMAC_PORT_FORWARD); + if (ret) { + prueth_destroy_rxq(emac); + return ret; + } + } + + return 0; +} + /** * emac_ndo_bpf - implements ndo_bpf for icssg_prueth * @ndev: network adapter device @@ -1214,11 +1335,27 @@ static int emac_ndo_bpf(struct net_device *ndev, struct netdev_bpf *bpf) switch (bpf->command) { case XDP_SETUP_PROG: return emac_xdp_setup(emac, bpf); + case XDP_SETUP_XSK_POOL: + return bpf->xsk.pool ? + prueth_xsk_pool_enable(emac, bpf->xsk.pool, bpf->xsk.queue_id) : + prueth_xsk_pool_disable(emac, bpf->xsk.queue_id); default: return -EINVAL; } } +int prueth_xsk_wakeup(struct net_device *ndev, u32 qid, u32 flags) +{ + struct prueth_emac *emac = netdev_priv(ndev); + + if (qid >= PRUETH_MAX_RX_FLOWS || qid >= emac->tx_ch_num) { + netdev_err(ndev, "Invalid XSK queue ID %d\n", qid); + return -EINVAL; + } + + return 0; +} + static const struct net_device_ops emac_netdev_ops = { .ndo_open = emac_ndo_open, .ndo_stop = emac_ndo_stop, @@ -1237,6 +1374,7 @@ static const struct net_device_ops emac_netdev_ops = { .ndo_xdp_xmit = emac_xdp_xmit, .ndo_hwtstamp_get = icssg_ndo_get_ts_config, .ndo_hwtstamp_set = icssg_ndo_set_ts_config, + .ndo_xsk_wakeup = prueth_xsk_wakeup, }; static int prueth_netdev_init(struct prueth *prueth, diff --git a/drivers/net/ethernet/ti/icssg/icssg_prueth.h b/drivers/net/ethernet/ti/icssg/icssg_prueth.h index 5cc90b66035a..a5e3774b0388 100644 --- a/drivers/net/ethernet/ti/icssg/icssg_prueth.h +++ b/drivers/net/ethernet/ti/icssg/icssg_prueth.h @@ -38,6 +38,8 @@ #include #include #include +#include +#include #include "icssg_config.h" #include "icss_iep.h" @@ -126,6 +128,7 @@ struct prueth_tx_chn { char name[32]; struct hrtimer tx_hrtimer; unsigned long tx_pace_timeout_ns; + struct xsk_buff_pool *xsk_pool; }; struct prueth_rx_chn { @@ -138,6 +141,7 @@ struct prueth_rx_chn { char name[32]; struct page_pool *pg_pool; struct xdp_rxq_info xdp_rxq; + struct xsk_buff_pool *xsk_pool; }; enum prueth_swdata_type { @@ -241,6 +245,7 @@ struct prueth_emac { struct netdev_hw_addr_list vlan_mcast_list[MAX_VLAN_ID]; struct bpf_prog *xdp_prog; struct xdp_attachment_info xdpi; + int xsk_qid; }; /* The buf includes headroom compatible with both skb and xdpf */ @@ -503,5 +508,10 @@ u32 emac_xmit_xdp_frame(struct prueth_emac *emac, unsigned int q_idx); void prueth_rx_cleanup(void *data, dma_addr_t desc_dma); void prueth_tx_cleanup(void *data, dma_addr_t desc_dma); +int prueth_xsk_wakeup(struct net_device *ndev, u32 qid, u32 flags); +static inline bool prueth_xdp_is_enabled(struct prueth_emac *emac) +{ + return !!READ_ONCE(emac->xdp_prog); +} #endif /* __NET_TI_ICSSG_PRUETH_H */ From 8756ef2eb078d9c92f7578c308e1650bb51c7f45 Mon Sep 17 00:00:00 2001 From: Meghana Malladi Date: Tue, 18 Nov 2025 19:25:39 +0530 Subject: [PATCH 0940/1381] net: ti: icssg-prueth: Add AF_XDP zero copy for TX Use xsk_pool inside tx_chn to check if a given Tx queue id is registered for xsk zero copy, which gets populated during xsk enable If xsk_pool is set, get frames from the pool in NAPI context and submit them to the Tx channel. Tx completion is also handled in the NAPI context. Use PRUETH_SWDATA_XSK to recycle xsk buffers back to the umem pool. Add XDP_WAKEUP_TX support to enable xsk_wakeup for Tx. Reviewed-by: Jacob Keller Signed-off-by: Meghana Malladi Link: https://patch.msgid.link/20251118135542.380574-4-m-malladi@ti.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/ti/icssg/icssg_common.c | 112 ++++++++++++++++++- drivers/net/ethernet/ti/icssg/icssg_prueth.c | 13 +++ drivers/net/ethernet/ti/icssg/icssg_prueth.h | 2 + 3 files changed, 125 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/ti/icssg/icssg_common.c b/drivers/net/ethernet/ti/icssg/icssg_common.c index cc52cff70d7e..d7469ad457fd 100644 --- a/drivers/net/ethernet/ti/icssg/icssg_common.c +++ b/drivers/net/ethernet/ti/icssg/icssg_common.c @@ -93,15 +93,91 @@ void prueth_ndev_del_tx_napi(struct prueth_emac *emac, int num) } EXPORT_SYMBOL_GPL(prueth_ndev_del_tx_napi); +static int emac_xsk_xmit_zc(struct prueth_emac *emac, + unsigned int q_idx) +{ + struct prueth_tx_chn *tx_chn = &emac->tx_chns[q_idx]; + struct xsk_buff_pool *pool = tx_chn->xsk_pool; + struct net_device *ndev = emac->ndev; + struct cppi5_host_desc_t *host_desc; + dma_addr_t dma_desc, dma_buf; + struct prueth_swdata *swdata; + struct xdp_desc xdp_desc; + int num_tx = 0, pkt_len; + int descs_avail, ret; + u32 *epib; + int i; + + descs_avail = k3_cppi_desc_pool_avail(tx_chn->desc_pool); + /* ensure that TX ring is not filled up by XDP, always MAX_SKB_FRAGS + * will be available for normal TX path and queue is stopped there if + * necessary + */ + if (descs_avail <= MAX_SKB_FRAGS) + return 0; + + descs_avail -= MAX_SKB_FRAGS; + + for (i = 0; i < descs_avail; i++) { + if (!xsk_tx_peek_desc(pool, &xdp_desc)) + break; + + dma_buf = xsk_buff_raw_get_dma(pool, xdp_desc.addr); + pkt_len = xdp_desc.len; + xsk_buff_raw_dma_sync_for_device(pool, dma_buf, pkt_len); + + host_desc = k3_cppi_desc_pool_alloc(tx_chn->desc_pool); + if (unlikely(!host_desc)) + break; + + cppi5_hdesc_init(host_desc, CPPI5_INFO0_HDESC_EPIB_PRESENT, + PRUETH_NAV_PS_DATA_SIZE); + cppi5_hdesc_set_pkttype(host_desc, 0); + epib = host_desc->epib; + epib[0] = 0; + epib[1] = 0; + cppi5_hdesc_set_pktlen(host_desc, pkt_len); + cppi5_desc_set_tags_ids(&host_desc->hdr, 0, + (emac->port_id | (q_idx << 8))); + + k3_udma_glue_tx_dma_to_cppi5_addr(tx_chn->tx_chn, &dma_buf); + cppi5_hdesc_attach_buf(host_desc, dma_buf, pkt_len, dma_buf, + pkt_len); + + swdata = cppi5_hdesc_get_swdata(host_desc); + swdata->type = PRUETH_SWDATA_XSK; + + dma_desc = k3_cppi_desc_pool_virt2dma(tx_chn->desc_pool, + host_desc); + ret = k3_udma_glue_push_tx_chn(tx_chn->tx_chn, + host_desc, dma_desc); + + if (ret) { + ndev->stats.tx_errors++; + k3_cppi_desc_pool_free(tx_chn->desc_pool, host_desc); + break; + } + + num_tx++; + } + + xsk_tx_release(tx_chn->xsk_pool); + return num_tx; +} + void prueth_xmit_free(struct prueth_tx_chn *tx_chn, struct cppi5_host_desc_t *desc) { struct cppi5_host_desc_t *first_desc, *next_desc; dma_addr_t buf_dma, next_desc_dma; + struct prueth_swdata *swdata; u32 buf_dma_len; first_desc = desc; next_desc = first_desc; + swdata = cppi5_hdesc_get_swdata(first_desc); + if (swdata->type == PRUETH_SWDATA_XSK) + goto free_pool; cppi5_hdesc_get_obuf(first_desc, &buf_dma, &buf_dma_len); k3_udma_glue_tx_cppi5_to_dma_addr(tx_chn->tx_chn, &buf_dma); @@ -126,6 +202,7 @@ void prueth_xmit_free(struct prueth_tx_chn *tx_chn, k3_cppi_desc_pool_free(tx_chn->desc_pool, next_desc); } +free_pool: k3_cppi_desc_pool_free(tx_chn->desc_pool, first_desc); } EXPORT_SYMBOL_GPL(prueth_xmit_free); @@ -139,7 +216,9 @@ int emac_tx_complete_packets(struct prueth_emac *emac, int chn, struct prueth_swdata *swdata; struct prueth_tx_chn *tx_chn; unsigned int total_bytes = 0; + int xsk_frames_done = 0; struct xdp_frame *xdpf; + unsigned int pkt_len; struct sk_buff *skb; dma_addr_t desc_dma; int res, num_tx = 0; @@ -176,6 +255,11 @@ int emac_tx_complete_packets(struct prueth_emac *emac, int chn, total_bytes += xdpf->len; xdp_return_frame(xdpf); break; + case PRUETH_SWDATA_XSK: + pkt_len = cppi5_hdesc_get_pktlen(desc_tx); + dev_sw_netstats_tx_add(ndev, 1, pkt_len); + xsk_frames_done++; + break; default: prueth_xmit_free(tx_chn, desc_tx); ndev->stats.tx_dropped++; @@ -204,6 +288,18 @@ int emac_tx_complete_packets(struct prueth_emac *emac, int chn, __netif_tx_unlock(netif_txq); } + if (tx_chn->xsk_pool) { + if (xsk_frames_done) + xsk_tx_completed(tx_chn->xsk_pool, xsk_frames_done); + + if (xsk_uses_need_wakeup(tx_chn->xsk_pool)) + xsk_set_tx_need_wakeup(tx_chn->xsk_pool); + + netif_txq = netdev_get_tx_queue(ndev, chn); + txq_trans_cond_update(netif_txq); + emac_xsk_xmit_zc(emac, chn); + } + return num_tx; } @@ -212,7 +308,10 @@ static enum hrtimer_restart emac_tx_timer_callback(struct hrtimer *timer) struct prueth_tx_chn *tx_chns = container_of(timer, struct prueth_tx_chn, tx_hrtimer); - enable_irq(tx_chns->irq); + if (tx_chns->irq_disabled) { + tx_chns->irq_disabled = false; + enable_irq(tx_chns->irq); + } return HRTIMER_NORESTART; } @@ -235,7 +334,10 @@ static int emac_napi_tx_poll(struct napi_struct *napi_tx, int budget) ns_to_ktime(tx_chn->tx_pace_timeout_ns), HRTIMER_MODE_REL_PINNED); } else { - enable_irq(tx_chn->irq); + if (tx_chn->irq_disabled) { + tx_chn->irq_disabled = false; + enable_irq(tx_chn->irq); + } } } @@ -246,6 +348,7 @@ static irqreturn_t prueth_tx_irq(int irq, void *dev_id) { struct prueth_tx_chn *tx_chn = dev_id; + tx_chn->irq_disabled = true; disable_irq_nosync(irq); napi_schedule(&tx_chn->napi_tx); @@ -1032,6 +1135,7 @@ void prueth_tx_cleanup(void *data, dma_addr_t desc_dma) { struct prueth_tx_chn *tx_chn = data; struct cppi5_host_desc_t *desc_tx; + struct xsk_buff_pool *xsk_pool; struct prueth_swdata *swdata; struct xdp_frame *xdpf; struct sk_buff *skb; @@ -1048,6 +1152,10 @@ void prueth_tx_cleanup(void *data, dma_addr_t desc_dma) xdpf = swdata->data.xdpf; xdp_return_frame(xdpf); break; + case PRUETH_SWDATA_XSK: + xsk_pool = tx_chn->xsk_pool; + xsk_tx_completed(xsk_pool, 1); + break; default: break; } diff --git a/drivers/net/ethernet/ti/icssg/icssg_prueth.c b/drivers/net/ethernet/ti/icssg/icssg_prueth.c index e4c3b6b152ea..bdce5a40defe 100644 --- a/drivers/net/ethernet/ti/icssg/icssg_prueth.c +++ b/drivers/net/ethernet/ti/icssg/icssg_prueth.c @@ -1347,12 +1347,25 @@ static int emac_ndo_bpf(struct net_device *ndev, struct netdev_bpf *bpf) int prueth_xsk_wakeup(struct net_device *ndev, u32 qid, u32 flags) { struct prueth_emac *emac = netdev_priv(ndev); + struct prueth_tx_chn *tx_chn = &emac->tx_chns[qid]; if (qid >= PRUETH_MAX_RX_FLOWS || qid >= emac->tx_ch_num) { netdev_err(ndev, "Invalid XSK queue ID %d\n", qid); return -EINVAL; } + if (!tx_chn->xsk_pool) { + netdev_err(ndev, "XSK pool not registered for queue %d\n", qid); + return -EINVAL; + } + + if (flags & XDP_WAKEUP_TX) { + if (!napi_if_scheduled_mark_missed(&tx_chn->napi_tx)) { + if (likely(napi_schedule_prep(&tx_chn->napi_tx))) + __napi_schedule(&tx_chn->napi_tx); + } + } + return 0; } diff --git a/drivers/net/ethernet/ti/icssg/icssg_prueth.h b/drivers/net/ethernet/ti/icssg/icssg_prueth.h index a5e3774b0388..67339cdf2ddf 100644 --- a/drivers/net/ethernet/ti/icssg/icssg_prueth.h +++ b/drivers/net/ethernet/ti/icssg/icssg_prueth.h @@ -129,6 +129,7 @@ struct prueth_tx_chn { struct hrtimer tx_hrtimer; unsigned long tx_pace_timeout_ns; struct xsk_buff_pool *xsk_pool; + bool irq_disabled; }; struct prueth_rx_chn { @@ -150,6 +151,7 @@ enum prueth_swdata_type { PRUETH_SWDATA_PAGE, PRUETH_SWDATA_CMD, PRUETH_SWDATA_XDPF, + PRUETH_SWDATA_XSK, }; struct prueth_swdata { From 121133163c9ff0d1e6a2f7d4a461e535088fc2e3 Mon Sep 17 00:00:00 2001 From: Meghana Malladi Date: Tue, 18 Nov 2025 19:25:40 +0530 Subject: [PATCH 0941/1381] net: ti: icssg-prueth: Make emac_run_xdp function independent of page emac_run_xdp function runs xdp program, at a given hook point in the Rx path of the driver in NAPI context and returns XDP return codes. In zero copy mode the driver receives packets using UMEM frames instead of pages (native XDP). Decouple the usage of page in this function. Reviewed-by: Jacob Keller Signed-off-by: Meghana Malladi Link: https://patch.msgid.link/20251118135542.380574-5-m-malladi@ti.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/ti/icssg/icssg_common.c | 26 ++++++++++++-------- drivers/net/ethernet/ti/icssg/icssg_prueth.c | 3 ++- drivers/net/ethernet/ti/icssg/icssg_prueth.h | 9 +++++-- 3 files changed, 25 insertions(+), 13 deletions(-) diff --git a/drivers/net/ethernet/ti/icssg/icssg_common.c b/drivers/net/ethernet/ti/icssg/icssg_common.c index d7469ad457fd..b88cfe99e8b7 100644 --- a/drivers/net/ethernet/ti/icssg/icssg_common.c +++ b/drivers/net/ethernet/ti/icssg/icssg_common.c @@ -647,15 +647,15 @@ void emac_rx_timestamp(struct prueth_emac *emac, * emac_xmit_xdp_frame - transmits an XDP frame * @emac: emac device * @xdpf: data to transmit - * @page: page from page pool if already DMA mapped * @q_idx: queue id + * @buff_type: Type of buffer to be transmitted * * Return: XDP state */ u32 emac_xmit_xdp_frame(struct prueth_emac *emac, struct xdp_frame *xdpf, - struct page *page, - unsigned int q_idx) + unsigned int q_idx, + enum prueth_tx_buff_type buff_type) { struct cppi5_host_desc_t *first_desc; struct net_device *ndev = emac->ndev; @@ -663,6 +663,7 @@ u32 emac_xmit_xdp_frame(struct prueth_emac *emac, struct prueth_tx_chn *tx_chn; dma_addr_t desc_dma, buf_dma; struct prueth_swdata *swdata; + struct page *page; u32 *epib; int ret; @@ -679,7 +680,12 @@ u32 emac_xmit_xdp_frame(struct prueth_emac *emac, return ICSSG_XDP_CONSUMED; /* drop */ } - if (page) { /* already DMA mapped by page_pool */ + if (buff_type == PRUETH_TX_BUFF_TYPE_XDP_TX) { /* already DMA mapped by page_pool */ + page = virt_to_head_page(xdpf->data); + if (unlikely(!page)) { + netdev_err(ndev, "xdp tx: failed to get page from xdpf\n"); + goto drop_free_descs; + } buf_dma = page_pool_get_dma_addr(page); buf_dma += xdpf->headroom + sizeof(struct xdp_frame); } else { /* Map the linear buffer */ @@ -734,13 +740,11 @@ EXPORT_SYMBOL_GPL(emac_xmit_xdp_frame); * emac_run_xdp - run an XDP program * @emac: emac device * @xdp: XDP buffer containing the frame - * @page: page with RX data if already DMA mapped * @len: Rx descriptor packet length * * Return: XDP state */ -static u32 emac_run_xdp(struct prueth_emac *emac, struct xdp_buff *xdp, - struct page *page, u32 *len) +static u32 emac_run_xdp(struct prueth_emac *emac, struct xdp_buff *xdp, u32 *len) { struct net_device *ndev = emac->ndev; struct netdev_queue *netif_txq; @@ -767,7 +771,8 @@ static u32 emac_run_xdp(struct prueth_emac *emac, struct xdp_buff *xdp, q_idx = cpu % emac->tx_ch_num; netif_txq = netdev_get_tx_queue(ndev, q_idx); __netif_tx_lock(netif_txq, cpu); - result = emac_xmit_xdp_frame(emac, xdpf, page, q_idx); + result = emac_xmit_xdp_frame(emac, xdpf, q_idx, + PRUETH_TX_BUFF_TYPE_XDP_TX); __netif_tx_unlock(netif_txq); if (result == ICSSG_XDP_CONSUMED) { ndev->stats.tx_dropped++; @@ -792,7 +797,8 @@ static u32 emac_run_xdp(struct prueth_emac *emac, struct xdp_buff *xdp, fallthrough; /* handle aborts by dropping packet */ case XDP_DROP: ndev->stats.rx_dropped++; - page_pool_recycle_direct(emac->rx_chns.pg_pool, page); + page_pool_recycle_direct(emac->rx_chns.pg_pool, + virt_to_head_page(xdp->data)); return ICSSG_XDP_CONSUMED; } } @@ -861,7 +867,7 @@ static int emac_rx_packet(struct prueth_emac *emac, u32 flow_id, u32 *xdp_state) xdp_init_buff(&xdp, PAGE_SIZE, &rx_chn->xdp_rxq); xdp_prepare_buff(&xdp, pa, PRUETH_HEADROOM, pkt_len, false); - *xdp_state = emac_run_xdp(emac, &xdp, page, &pkt_len); + *xdp_state = emac_run_xdp(emac, &xdp, &pkt_len); if (*xdp_state != ICSSG_XDP_PASS) goto requeue; headroom = xdp.data - xdp.data_hard_start; diff --git a/drivers/net/ethernet/ti/icssg/icssg_prueth.c b/drivers/net/ethernet/ti/icssg/icssg_prueth.c index bdce5a40defe..bb8d42ba0102 100644 --- a/drivers/net/ethernet/ti/icssg/icssg_prueth.c +++ b/drivers/net/ethernet/ti/icssg/icssg_prueth.c @@ -1185,7 +1185,8 @@ static int emac_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frame __netif_tx_lock(netif_txq, cpu); for (i = 0; i < n; i++) { xdpf = frames[i]; - err = emac_xmit_xdp_frame(emac, xdpf, NULL, q_idx); + err = emac_xmit_xdp_frame(emac, xdpf, q_idx, + PRUETH_TX_BUFF_TYPE_XDP_NDO); if (err != ICSSG_XDP_TX) { ndev->stats.tx_dropped++; break; diff --git a/drivers/net/ethernet/ti/icssg/icssg_prueth.h b/drivers/net/ethernet/ti/icssg/icssg_prueth.h index 67339cdf2ddf..3147a1d8f59a 100644 --- a/drivers/net/ethernet/ti/icssg/icssg_prueth.h +++ b/drivers/net/ethernet/ti/icssg/icssg_prueth.h @@ -154,6 +154,11 @@ enum prueth_swdata_type { PRUETH_SWDATA_XSK, }; +enum prueth_tx_buff_type { + PRUETH_TX_BUFF_TYPE_XDP_TX, + PRUETH_TX_BUFF_TYPE_XDP_NDO, +}; + struct prueth_swdata { enum prueth_swdata_type type; union prueth_data { @@ -506,8 +511,8 @@ void prueth_put_cores(struct prueth *prueth, int slice); u64 icssg_ts_to_ns(u32 hi_sw, u32 hi, u32 lo, u32 cycle_time_ns); u32 emac_xmit_xdp_frame(struct prueth_emac *emac, struct xdp_frame *xdpf, - struct page *page, - unsigned int q_idx); + unsigned int q_idx, + enum prueth_tx_buff_type buff_type); void prueth_rx_cleanup(void *data, dma_addr_t desc_dma); void prueth_tx_cleanup(void *data, dma_addr_t desc_dma); int prueth_xsk_wakeup(struct net_device *ndev, u32 qid, u32 flags); From 7a64bb388df3cf091afdd047c701039a13acd3b4 Mon Sep 17 00:00:00 2001 From: Meghana Malladi Date: Tue, 18 Nov 2025 19:25:41 +0530 Subject: [PATCH 0942/1381] net: ti: icssg-prueth: Add AF_XDP zero copy for RX Use xsk_pool inside rx_chn to check if a given Rx queue id is registered for xsk zero copy, which gets populated during xsk enable. Update prueth_create_xdp_rxqs to register and support two different memory models (xsk and page) for a given Rx queue, if registered for zero copy. If xsk_pool is registered, allocate buffers from UMEM and map them to the hardware Rx descriptors. In NAPI context, run the XDP program for each packet and process the xsk buffer according to the XDP result codes. Also allocate new set of buffers from UMEM for the next batch of NAPI Rx processing. Add XDK_WAKEUP_RX support to support xsk wakeup for Rx. Move prueth_create_page_pool to prueth_init_rx_chns to avoid freeing and re-allocating the system memory every time there is a transition from zero copy to copy and prevents any type of memory fragmentation or leak. Reviewed-by: Jacob Keller Signed-off-by: Meghana Malladi Link: https://patch.msgid.link/20251118135542.380574-6-m-malladi@ti.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/ti/icssg/icssg_common.c | 321 +++++++++++++++---- drivers/net/ethernet/ti/icssg/icssg_prueth.c | 60 +++- drivers/net/ethernet/ti/icssg/icssg_prueth.h | 2 + 3 files changed, 312 insertions(+), 71 deletions(-) diff --git a/drivers/net/ethernet/ti/icssg/icssg_common.c b/drivers/net/ethernet/ti/icssg/icssg_common.c index b88cfe99e8b7..090aa74d3ce7 100644 --- a/drivers/net/ethernet/ti/icssg/icssg_common.c +++ b/drivers/net/ethernet/ti/icssg/icssg_common.c @@ -465,6 +465,29 @@ int prueth_init_tx_chns(struct prueth_emac *emac) } EXPORT_SYMBOL_GPL(prueth_init_tx_chns); +static struct page_pool *prueth_create_page_pool(struct prueth_emac *emac, + struct device *dma_dev, + int size) +{ + struct page_pool_params pp_params = { 0 }; + struct page_pool *pool; + + pp_params.order = 0; + pp_params.flags = PP_FLAG_DMA_MAP | PP_FLAG_DMA_SYNC_DEV; + pp_params.pool_size = size; + pp_params.nid = dev_to_node(emac->prueth->dev); + pp_params.dma_dir = DMA_BIDIRECTIONAL; + pp_params.dev = dma_dev; + pp_params.napi = &emac->napi_rx; + pp_params.max_len = PAGE_SIZE; + + pool = page_pool_create(&pp_params); + if (IS_ERR(pool)) + netdev_err(emac->ndev, "cannot create rx page pool\n"); + + return pool; +} + int prueth_init_rx_chns(struct prueth_emac *emac, struct prueth_rx_chn *rx_chn, char *name, u32 max_rflows, @@ -474,6 +497,7 @@ int prueth_init_rx_chns(struct prueth_emac *emac, struct device *dev = emac->prueth->dev; struct net_device *ndev = emac->ndev; u32 fdqring_id, hdesc_size; + struct page_pool *pool; int i, ret = 0, slice; int flow_id_base; @@ -516,6 +540,14 @@ int prueth_init_rx_chns(struct prueth_emac *emac, goto fail; } + pool = prueth_create_page_pool(emac, rx_chn->dma_dev, rx_chn->descs_num); + if (IS_ERR(pool)) { + ret = PTR_ERR(pool); + goto fail; + } + + rx_chn->pg_pool = pool; + flow_id_base = k3_udma_glue_rx_get_flow_id_base(rx_chn->rx_chn); if (emac->is_sr1 && !strcmp(name, "rxmgm")) { emac->rx_mgm_flow_id_base = flow_id_base; @@ -797,12 +829,188 @@ static u32 emac_run_xdp(struct prueth_emac *emac, struct xdp_buff *xdp, u32 *len fallthrough; /* handle aborts by dropping packet */ case XDP_DROP: ndev->stats.rx_dropped++; - page_pool_recycle_direct(emac->rx_chns.pg_pool, - virt_to_head_page(xdp->data)); return ICSSG_XDP_CONSUMED; } } +static int prueth_dma_rx_push_mapped_zc(struct prueth_emac *emac, + struct prueth_rx_chn *rx_chn, + struct xdp_buff *xdp) +{ + struct net_device *ndev = emac->ndev; + struct cppi5_host_desc_t *desc_rx; + struct prueth_swdata *swdata; + dma_addr_t desc_dma; + dma_addr_t buf_dma; + int buf_len; + + buf_dma = xsk_buff_xdp_get_dma(xdp); + desc_rx = k3_cppi_desc_pool_alloc(rx_chn->desc_pool); + if (!desc_rx) { + netdev_err(ndev, "rx push: failed to allocate descriptor\n"); + return -ENOMEM; + } + desc_dma = k3_cppi_desc_pool_virt2dma(rx_chn->desc_pool, desc_rx); + + cppi5_hdesc_init(desc_rx, CPPI5_INFO0_HDESC_EPIB_PRESENT, + PRUETH_NAV_PS_DATA_SIZE); + k3_udma_glue_rx_dma_to_cppi5_addr(rx_chn->rx_chn, &buf_dma); + buf_len = xsk_pool_get_rx_frame_size(rx_chn->xsk_pool); + cppi5_hdesc_attach_buf(desc_rx, buf_dma, buf_len, buf_dma, buf_len); + swdata = cppi5_hdesc_get_swdata(desc_rx); + swdata->type = PRUETH_SWDATA_XSK; + swdata->data.xdp = xdp; + + return k3_udma_glue_push_rx_chn(rx_chn->rx_chn, PRUETH_RX_FLOW_DATA, + desc_rx, desc_dma); +} + +static int prueth_rx_alloc_zc(struct prueth_emac *emac, int budget) +{ + struct prueth_rx_chn *rx_chn = &emac->rx_chns; + struct xdp_buff *xdp; + int i, ret; + + for (i = 0; i < budget; i++) { + xdp = xsk_buff_alloc(rx_chn->xsk_pool); + if (!xdp) + break; + + ret = prueth_dma_rx_push_mapped_zc(emac, rx_chn, xdp); + if (ret) { + netdev_err(emac->ndev, "rx alloc: failed to map descriptors to xdp buff\n"); + xsk_buff_free(xdp); + break; + } + } + + return i; +} + +static void emac_dispatch_skb_zc(struct prueth_emac *emac, struct xdp_buff *xdp, u32 *psdata) +{ + unsigned int headroom = xdp->data - xdp->data_hard_start; + unsigned int pkt_len = xdp->data_end - xdp->data; + struct net_device *ndev = emac->ndev; + struct sk_buff *skb; + + skb = napi_alloc_skb(&emac->napi_rx, xdp->data_end - xdp->data_hard_start); + if (unlikely(!skb)) { + ndev->stats.rx_dropped++; + return; + } + + skb_reserve(skb, headroom); + skb_put(skb, pkt_len); + skb->dev = ndev; + + /* RX HW timestamp */ + if (emac->rx_ts_enabled) + emac_rx_timestamp(emac, skb, psdata); + + if (emac->prueth->is_switch_mode) + skb->offload_fwd_mark = emac->offload_fwd_mark; + skb->protocol = eth_type_trans(skb, ndev); + + skb_mark_for_recycle(skb); + napi_gro_receive(&emac->napi_rx, skb); + ndev->stats.rx_bytes += pkt_len; + ndev->stats.rx_packets++; +} + +static int emac_rx_packet_zc(struct prueth_emac *emac, u32 flow_id, + int budget) +{ + struct prueth_rx_chn *rx_chn = &emac->rx_chns; + u32 buf_dma_len, pkt_len, port_id = 0; + struct net_device *ndev = emac->ndev; + struct cppi5_host_desc_t *desc_rx; + struct prueth_swdata *swdata; + dma_addr_t desc_dma, buf_dma; + struct xdp_buff *xdp; + int xdp_status = 0; + int count = 0; + u32 *psdata; + int ret; + + while (count < budget) { + ret = k3_udma_glue_pop_rx_chn(rx_chn->rx_chn, flow_id, &desc_dma); + if (ret) { + if (ret != -ENODATA) + netdev_err(ndev, "rx pop: failed: %d\n", ret); + break; + } + + if (cppi5_desc_is_tdcm(desc_dma)) { + complete(&emac->tdown_complete); + break; + } + + desc_rx = k3_cppi_desc_pool_dma2virt(rx_chn->desc_pool, desc_dma); + swdata = cppi5_hdesc_get_swdata(desc_rx); + if (swdata->type != PRUETH_SWDATA_XSK) { + netdev_err(ndev, "rx_pkt: invalid swdata->type %d\n", swdata->type); + k3_cppi_desc_pool_free(rx_chn->desc_pool, desc_rx); + break; + } + + xdp = swdata->data.xdp; + cppi5_hdesc_get_obuf(desc_rx, &buf_dma, &buf_dma_len); + k3_udma_glue_rx_cppi5_to_dma_addr(rx_chn->rx_chn, &buf_dma); + pkt_len = cppi5_hdesc_get_pktlen(desc_rx); + /* firmware adds 4 CRC bytes, strip them */ + pkt_len -= 4; + cppi5_desc_get_tags_ids(&desc_rx->hdr, &port_id, NULL); + psdata = cppi5_hdesc_get_psdata(desc_rx); + k3_cppi_desc_pool_free(rx_chn->desc_pool, desc_rx); + count++; + xsk_buff_set_size(xdp, pkt_len); + xsk_buff_dma_sync_for_cpu(xdp); + + if (prueth_xdp_is_enabled(emac)) { + ret = emac_run_xdp(emac, xdp, &pkt_len); + switch (ret) { + case ICSSG_XDP_PASS: + /* prepare skb and send to n/w stack */ + emac_dispatch_skb_zc(emac, xdp, psdata); + xsk_buff_free(xdp); + break; + case ICSSG_XDP_CONSUMED: + xsk_buff_free(xdp); + break; + case ICSSG_XDP_TX: + case ICSSG_XDP_REDIR: + xdp_status |= ret; + break; + } + } else { + /* prepare skb and send to n/w stack */ + emac_dispatch_skb_zc(emac, xdp, psdata); + xsk_buff_free(xdp); + } + } + + if (xdp_status & ICSSG_XDP_REDIR) + xdp_do_flush(); + + /* Allocate xsk buffers from the pool for the "count" number of + * packets processed in order to be able to receive more packets. + */ + ret = prueth_rx_alloc_zc(emac, count); + + if (xsk_uses_need_wakeup(rx_chn->xsk_pool)) { + /* If the user space doesn't provide enough buffers then it must + * explicitly wake up the kernel when new buffers are available + */ + if (ret < count) + xsk_set_rx_need_wakeup(rx_chn->xsk_pool); + else + xsk_clear_rx_need_wakeup(rx_chn->xsk_pool); + } + + return count; +} + static int emac_rx_packet(struct prueth_emac *emac, u32 flow_id, u32 *xdp_state) { struct prueth_rx_chn *rx_chn = &emac->rx_chns; @@ -849,7 +1057,6 @@ static int emac_rx_packet(struct prueth_emac *emac, u32 flow_id, u32 *xdp_state) /* firmware adds 4 CRC bytes, strip them */ pkt_len -= 4; cppi5_desc_get_tags_ids(&desc_rx->hdr, &port_id, NULL); - k3_cppi_desc_pool_free(rx_chn->desc_pool, desc_rx); /* if allocation fails we drop the packet but push the @@ -921,12 +1128,16 @@ void prueth_rx_cleanup(void *data, dma_addr_t desc_dma) struct cppi5_host_desc_t *desc_rx; struct prueth_swdata *swdata; struct page_pool *pool; + struct xdp_buff *xdp; struct page *page; pool = rx_chn->pg_pool; desc_rx = k3_cppi_desc_pool_dma2virt(rx_chn->desc_pool, desc_dma); swdata = cppi5_hdesc_get_swdata(desc_rx); - if (swdata->type == PRUETH_SWDATA_PAGE) { + if (rx_chn->xsk_pool) { + xdp = swdata->data.xdp; + xsk_buff_free(xdp); + } else { page = swdata->data.page; page_pool_recycle_direct(pool, page); } @@ -1174,6 +1385,7 @@ irqreturn_t prueth_rx_irq(int irq, void *dev_id) { struct prueth_emac *emac = dev_id; + emac->rx_chns.irq_disabled = true; disable_irq_nosync(irq); napi_schedule(&emac->napi_rx); @@ -1201,6 +1413,7 @@ int icssg_napi_rx_poll(struct napi_struct *napi_rx, int budget) PRUETH_RX_FLOW_DATA_SR1 : PRUETH_RX_FLOW_DATA; int flow = emac->is_sr1 ? PRUETH_MAX_RX_FLOWS_SR1 : PRUETH_MAX_RX_FLOWS; + struct prueth_rx_chn *rx_chn = &emac->rx_chns; int xdp_state_or = 0; int num_rx = 0; int cur_budget; @@ -1208,14 +1421,18 @@ int icssg_napi_rx_poll(struct napi_struct *napi_rx, int budget) int ret; while (flow--) { - cur_budget = budget - num_rx; + if (rx_chn->xsk_pool) { + num_rx = emac_rx_packet_zc(emac, flow, budget); + } else { + cur_budget = budget - num_rx; - while (cur_budget--) { - ret = emac_rx_packet(emac, flow, &xdp_state); - xdp_state_or |= xdp_state; - if (ret) - break; - num_rx++; + while (cur_budget--) { + ret = emac_rx_packet(emac, flow, &xdp_state); + xdp_state_or |= xdp_state; + if (ret) + break; + num_rx++; + } } if (num_rx >= budget) @@ -1231,7 +1448,11 @@ int icssg_napi_rx_poll(struct napi_struct *napi_rx, int budget) ns_to_ktime(emac->rx_pace_timeout_ns), HRTIMER_MODE_REL_PINNED); } else { - enable_irq(emac->rx_chns.irq[rx_flow]); + if (emac->rx_chns.irq_disabled) { + /* re-enable the RX IRQ */ + emac->rx_chns.irq_disabled = false; + enable_irq(emac->rx_chns.irq[rx_flow]); + } } } @@ -1239,62 +1460,48 @@ int icssg_napi_rx_poll(struct napi_struct *napi_rx, int budget) } EXPORT_SYMBOL_GPL(icssg_napi_rx_poll); -static struct page_pool *prueth_create_page_pool(struct prueth_emac *emac, - struct device *dma_dev, - int size) -{ - struct page_pool_params pp_params = { 0 }; - struct page_pool *pool; - - pp_params.order = 0; - pp_params.flags = PP_FLAG_DMA_MAP | PP_FLAG_DMA_SYNC_DEV; - pp_params.pool_size = size; - pp_params.nid = dev_to_node(emac->prueth->dev); - pp_params.dma_dir = DMA_BIDIRECTIONAL; - pp_params.dev = dma_dev; - pp_params.napi = &emac->napi_rx; - pp_params.max_len = PAGE_SIZE; - - pool = page_pool_create(&pp_params); - if (IS_ERR(pool)) - netdev_err(emac->ndev, "cannot create rx page pool\n"); - - return pool; -} - int prueth_prepare_rx_chan(struct prueth_emac *emac, struct prueth_rx_chn *chn, int buf_size) { - struct page_pool *pool; struct page *page; + int desc_avail; int i, ret; - pool = prueth_create_page_pool(emac, chn->dma_dev, chn->descs_num); - if (IS_ERR(pool)) - return PTR_ERR(pool); + desc_avail = k3_cppi_desc_pool_avail(chn->desc_pool); + if (desc_avail < chn->descs_num) + netdev_warn(emac->ndev, + "not enough RX descriptors available %d < %d\n", + desc_avail, chn->descs_num); - chn->pg_pool = pool; - - for (i = 0; i < chn->descs_num; i++) { - /* NOTE: we're not using memory efficiently here. - * 1 full page (4KB?) used here instead of - * PRUETH_MAX_PKT_SIZE (~1.5KB?) + if (chn->xsk_pool) { + /* get pages from xsk_pool and push to RX ring + * queue as much as possible */ - page = page_pool_dev_alloc_pages(pool); - if (!page) { - netdev_err(emac->ndev, "couldn't allocate rx page\n"); - ret = -ENOMEM; + ret = prueth_rx_alloc_zc(emac, desc_avail); + if (!ret) goto recycle_alloc_pg; - } + } else { + for (i = 0; i < desc_avail; i++) { + /* NOTE: we're not using memory efficiently here. + * 1 full page (4KB?) used here instead of + * PRUETH_MAX_PKT_SIZE (~1.5KB?) + */ + page = page_pool_dev_alloc_pages(chn->pg_pool); + if (!page) { + netdev_err(emac->ndev, "couldn't allocate rx page\n"); + ret = -ENOMEM; + goto recycle_alloc_pg; + } - ret = prueth_dma_rx_push_mapped(emac, chn, page, buf_size); - if (ret < 0) { - netdev_err(emac->ndev, - "cannot submit page for rx chan %s ret %d\n", - chn->name, ret); - page_pool_recycle_direct(pool, page); - goto recycle_alloc_pg; + ret = prueth_dma_rx_push_mapped(emac, chn, page, buf_size); + if (ret < 0) { + netdev_err(emac->ndev, + "cannot submit page for rx chan %s ret %d\n", + chn->name, ret); + page_pool_recycle_direct(chn->pg_pool, page); + goto recycle_alloc_pg; + } } } diff --git a/drivers/net/ethernet/ti/icssg/icssg_prueth.c b/drivers/net/ethernet/ti/icssg/icssg_prueth.c index bb8d42ba0102..22de04ac18cb 100644 --- a/drivers/net/ethernet/ti/icssg/icssg_prueth.c +++ b/drivers/net/ethernet/ti/icssg/icssg_prueth.c @@ -395,7 +395,11 @@ static enum hrtimer_restart emac_rx_timer_callback(struct hrtimer *timer) container_of(timer, struct prueth_emac, rx_hrtimer); int rx_flow = PRUETH_RX_FLOW_DATA; - enable_irq(emac->rx_chns.irq[rx_flow]); + if (emac->rx_chns.irq_disabled) { + /* re-enable the RX IRQ */ + emac->rx_chns.irq_disabled = false; + enable_irq(emac->rx_chns.irq[rx_flow]); + } return HRTIMER_NORESTART; } @@ -569,33 +573,43 @@ const struct icss_iep_clockops prueth_iep_clockops = { .perout_enable = prueth_perout_enable, }; +static void prueth_destroy_xdp_rxqs(struct prueth_emac *emac) +{ + struct xdp_rxq_info *rxq = &emac->rx_chns.xdp_rxq; + + if (xdp_rxq_info_is_reg(rxq)) + xdp_rxq_info_unreg(rxq); +} + static int prueth_create_xdp_rxqs(struct prueth_emac *emac) { struct xdp_rxq_info *rxq = &emac->rx_chns.xdp_rxq; struct page_pool *pool = emac->rx_chns.pg_pool; + struct prueth_rx_chn *rx_chn = &emac->rx_chns; int ret; ret = xdp_rxq_info_reg(rxq, emac->ndev, 0, emac->napi_rx.napi_id); if (ret) return ret; - ret = xdp_rxq_info_reg_mem_model(rxq, MEM_TYPE_PAGE_POOL, pool); - if (ret) - xdp_rxq_info_unreg(rxq); + if (rx_chn->xsk_pool) { + ret = xdp_rxq_info_reg_mem_model(rxq, MEM_TYPE_XSK_BUFF_POOL, NULL); + if (ret) + goto xdp_unreg; + xsk_pool_set_rxq_info(rx_chn->xsk_pool, rxq); + } else { + ret = xdp_rxq_info_reg_mem_model(rxq, MEM_TYPE_PAGE_POOL, pool); + if (ret) + goto xdp_unreg; + } + return 0; + +xdp_unreg: + prueth_destroy_xdp_rxqs(emac); return ret; } -static void prueth_destroy_xdp_rxqs(struct prueth_emac *emac) -{ - struct xdp_rxq_info *rxq = &emac->rx_chns.xdp_rxq; - - if (!xdp_rxq_info_is_reg(rxq)) - return; - - xdp_rxq_info_unreg(rxq); -} - static int icssg_prueth_add_mcast(struct net_device *ndev, const u8 *addr) { struct net_device *real_dev; @@ -1349,6 +1363,12 @@ int prueth_xsk_wakeup(struct net_device *ndev, u32 qid, u32 flags) { struct prueth_emac *emac = netdev_priv(ndev); struct prueth_tx_chn *tx_chn = &emac->tx_chns[qid]; + struct prueth_rx_chn *rx_chn = &emac->rx_chns; + + if (emac->xsk_qid != qid) { + netdev_err(ndev, "XSK queue %d not registered\n", qid); + return -EINVAL; + } if (qid >= PRUETH_MAX_RX_FLOWS || qid >= emac->tx_ch_num) { netdev_err(ndev, "Invalid XSK queue ID %d\n", qid); @@ -1360,6 +1380,11 @@ int prueth_xsk_wakeup(struct net_device *ndev, u32 qid, u32 flags) return -EINVAL; } + if (!rx_chn->xsk_pool) { + netdev_err(ndev, "XSK pool not registered for RX queue %d\n", qid); + return -EINVAL; + } + if (flags & XDP_WAKEUP_TX) { if (!napi_if_scheduled_mark_missed(&tx_chn->napi_tx)) { if (likely(napi_schedule_prep(&tx_chn->napi_tx))) @@ -1367,6 +1392,13 @@ int prueth_xsk_wakeup(struct net_device *ndev, u32 qid, u32 flags) } } + if (flags & XDP_WAKEUP_RX) { + if (!napi_if_scheduled_mark_missed(&emac->napi_rx)) { + if (likely(napi_schedule_prep(&emac->napi_rx))) + __napi_schedule(&emac->napi_rx); + } + } + return 0; } diff --git a/drivers/net/ethernet/ti/icssg/icssg_prueth.h b/drivers/net/ethernet/ti/icssg/icssg_prueth.h index 3147a1d8f59a..10eadd356650 100644 --- a/drivers/net/ethernet/ti/icssg/icssg_prueth.h +++ b/drivers/net/ethernet/ti/icssg/icssg_prueth.h @@ -143,6 +143,7 @@ struct prueth_rx_chn { struct page_pool *pg_pool; struct xdp_rxq_info xdp_rxq; struct xsk_buff_pool *xsk_pool; + bool irq_disabled; }; enum prueth_swdata_type { @@ -166,6 +167,7 @@ struct prueth_swdata { struct page *page; u32 cmd; struct xdp_frame *xdpf; + struct xdp_buff *xdp; } data; }; From c6a1ec1870e623ff74e70a481d11ff82932f31ff Mon Sep 17 00:00:00 2001 From: Meghana Malladi Date: Tue, 18 Nov 2025 19:25:42 +0530 Subject: [PATCH 0943/1381] net: ti: icssg-prueth: Enable zero copy in XDP features Enable the zero copy feature flag in xdp_set_features_flag() for a given ndev to get the AF-XDP zero copy support running for both Tx and Rx. Reviewed-by: Jacob Keller Signed-off-by: Meghana Malladi Link: https://patch.msgid.link/20251118135542.380574-7-m-malladi@ti.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/ti/icssg/icssg_prueth.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/ti/icssg/icssg_prueth.c b/drivers/net/ethernet/ti/icssg/icssg_prueth.c index 22de04ac18cb..f65041662173 100644 --- a/drivers/net/ethernet/ti/icssg/icssg_prueth.c +++ b/drivers/net/ethernet/ti/icssg/icssg_prueth.c @@ -1554,7 +1554,8 @@ static int prueth_netdev_init(struct prueth *prueth, xdp_set_features_flag(ndev, NETDEV_XDP_ACT_BASIC | NETDEV_XDP_ACT_REDIRECT | - NETDEV_XDP_ACT_NDO_XMIT); + NETDEV_XDP_ACT_NDO_XMIT | + NETDEV_XDP_ACT_XSK_ZEROCOPY); netif_napi_add(ndev, &emac->napi_rx, icssg_napi_rx_poll); hrtimer_setup(&emac->rx_hrtimer, &emac_rx_timer_callback, CLOCK_MONOTONIC, From 2a2d5a3392b63ef5141c158117dba4f3b9b3ac22 Mon Sep 17 00:00:00 2001 From: Gal Pressman Date: Tue, 18 Nov 2025 16:32:06 +0200 Subject: [PATCH 0944/1381] tools: ynl: cli: Add --list-attrs option to show operation attributes Add a --list-attrs option to the YNL CLI that displays information about netlink operations, including request and reply attributes. This eliminates the need to manually inspect YAML spec files to determine the JSON structure required for operations, or understand the structure of the reply. Example usage: # ./cli.py --family netdev --list-attrs dev-get Operation: dev-get Get / dump information about a netdev. Do request attributes: - ifindex: u32 netdev ifindex Do reply attributes: - ifindex: u32 netdev ifindex - xdp-features: u64 (enum: xdp-act) Bitmask of enabled xdp-features. - xdp-zc-max-segs: u32 max fragment count supported by ZC driver - xdp-rx-metadata-features: u64 (enum: xdp-rx-metadata) Bitmask of supported XDP receive metadata features. See Documentation/networking/xdp-rx-metadata.rst for more details. - xsk-features: u64 (enum: xsk-flags) Bitmask of enabled AF_XDP features. Dump reply attributes: - ifindex: u32 netdev ifindex - xdp-features: u64 (enum: xdp-act) Bitmask of enabled xdp-features. - xdp-zc-max-segs: u32 max fragment count supported by ZC driver - xdp-rx-metadata-features: u64 (enum: xdp-rx-metadata) Bitmask of supported XDP receive metadata features. See Documentation/networking/xdp-rx-metadata.rst for more details. - xsk-features: u64 (enum: xsk-flags) Bitmask of enabled AF_XDP features. Reviewed-by: Nimrod Oren Signed-off-by: Gal Pressman Link: https://patch.msgid.link/20251118143208.2380814-2-gal@nvidia.com Signed-off-by: Paolo Abeni --- tools/net/ynl/pyynl/cli.py | 56 ++++++++++++++++++++++++++++++++++++++ 1 file changed, 56 insertions(+) diff --git a/tools/net/ynl/pyynl/cli.py b/tools/net/ynl/pyynl/cli.py index 8c192e900bd3..4d91a2cee381 100755 --- a/tools/net/ynl/pyynl/cli.py +++ b/tools/net/ynl/pyynl/cli.py @@ -7,6 +7,7 @@ import os import pathlib import pprint import sys +import textwrap sys.path.append(pathlib.Path(__file__).resolve().parent.as_posix()) from lib import YnlFamily, Netlink, NlError @@ -39,6 +40,37 @@ class YnlEncoder(json.JSONEncoder): return json.JSONEncoder.default(self, obj) +def print_attr_list(attr_names, attr_set): + """Print a list of attributes with their types and documentation.""" + for attr_name in attr_names: + if attr_name in attr_set.attrs: + attr = attr_set.attrs[attr_name] + attr_info = f' - {attr_name}: {attr.type}' + if 'enum' in attr.yaml: + attr_info += f" (enum: {attr.yaml['enum']})" + if attr.yaml.get('doc'): + doc_text = textwrap.indent(attr.yaml['doc'], ' ') + attr_info += f"\n{doc_text}" + print(attr_info) + + +def print_mode_attrs(mode, mode_spec, attr_set, print_request=True): + """Print a given mode (do/dump/event/notify).""" + mode_title = mode.capitalize() + + if print_request and 'request' in mode_spec and 'attributes' in mode_spec['request']: + print(f'\n{mode_title} request attributes:') + print_attr_list(mode_spec['request']['attributes'], attr_set) + + if 'reply' in mode_spec and 'attributes' in mode_spec['reply']: + print(f'\n{mode_title} reply attributes:') + print_attr_list(mode_spec['reply']['attributes'], attr_set) + + if 'attributes' in mode_spec: + print(f'\n{mode_title} attributes:') + print_attr_list(mode_spec['attributes'], attr_set) + + def main(): description = """ YNL CLI utility - a general purpose netlink utility that uses YAML @@ -70,6 +102,8 @@ def main(): group.add_argument('--dump', dest='dump', metavar='DUMP-OPERATION', type=str) group.add_argument('--list-ops', action='store_true') group.add_argument('--list-msgs', action='store_true') + group.add_argument('--list-attrs', dest='list_attrs', metavar='OPERATION', type=str, + help='List attributes for an operation') parser.add_argument('--duration', dest='duration', type=int, help='when subscribed, watch for DURATION seconds') @@ -135,6 +169,28 @@ def main(): for op_name, op in ynl.msgs.items(): print(op_name, " [", ", ".join(op.modes), "]") + if args.list_attrs: + op = ynl.msgs.get(args.list_attrs) + if not op: + print(f'Operation {args.list_attrs} not found') + exit(1) + + print(f'Operation: {op.name}') + print(op.yaml['doc']) + + for mode in ['do', 'dump', 'event']: + if mode in op.yaml: + print_mode_attrs(mode, op.yaml[mode], op.attr_set, True) + + if 'notify' in op.yaml: + mode_spec = op.yaml['notify'] + ref_spec = ynl.msgs.get(mode_spec).yaml.get('do') + if ref_spec: + print_mode_attrs('notify', ref_spec, op.attr_set, False) + + if 'mcgrp' in op.yaml: + print(f"\nMulticast group: {op.yaml['mcgrp']}") + try: if args.do: reply = ynl.do(args.do, attrs, args.flags) From bc1bc1b357cdae75d17ecde0aabe1b95ce08c7a5 Mon Sep 17 00:00:00 2001 From: Gal Pressman Date: Tue, 18 Nov 2025 16:32:07 +0200 Subject: [PATCH 0945/1381] tools: ynl: cli: Parse nested attributes in --list-attrs output Enhance the --list-attrs option to recursively display nested attributes instead of just showing "nest" as the type. Nested attributes now show their attribute set name and expand to display their contents. # ./cli.py --family ethtool --list-attrs rss-get [..] Do request attributes: - header: nest -> header - dev-index: u32 - dev-name: string - flags: u32 (enum: header-flags) - phy-index: u32 - context: u32 [..] Reviewed-by: Nimrod Oren Signed-off-by: Gal Pressman Reviewed-by: Donald Hunter Link: https://patch.msgid.link/20251118143208.2380814-3-gal@nvidia.com Signed-off-by: Paolo Abeni --- tools/net/ynl/pyynl/cli.py | 35 ++++++++++++++++++++++++++--------- 1 file changed, 26 insertions(+), 9 deletions(-) diff --git a/tools/net/ynl/pyynl/cli.py b/tools/net/ynl/pyynl/cli.py index 4d91a2cee381..6655ee61081a 100755 --- a/tools/net/ynl/pyynl/cli.py +++ b/tools/net/ynl/pyynl/cli.py @@ -40,35 +40,52 @@ class YnlEncoder(json.JSONEncoder): return json.JSONEncoder.default(self, obj) -def print_attr_list(attr_names, attr_set): +def print_attr_list(ynl, attr_names, attr_set, indent=2): """Print a list of attributes with their types and documentation.""" + prefix = ' ' * indent for attr_name in attr_names: if attr_name in attr_set.attrs: attr = attr_set.attrs[attr_name] - attr_info = f' - {attr_name}: {attr.type}' + attr_info = f'{prefix}- {attr_name}: {attr.type}' if 'enum' in attr.yaml: attr_info += f" (enum: {attr.yaml['enum']})" + + # Show nested attributes reference and recursively display them + nested_set_name = None + if attr.type == 'nest' and 'nested-attributes' in attr.yaml: + nested_set_name = attr.yaml['nested-attributes'] + attr_info += f" -> {nested_set_name}" + if attr.yaml.get('doc'): - doc_text = textwrap.indent(attr.yaml['doc'], ' ') + doc_text = textwrap.indent(attr.yaml['doc'], prefix + ' ') attr_info += f"\n{doc_text}" print(attr_info) + # Recursively show nested attributes + if nested_set_name in ynl.attr_sets: + nested_set = ynl.attr_sets[nested_set_name] + # Filter out 'unspec' and other unused attrs + nested_names = [n for n in nested_set.attrs.keys() + if nested_set.attrs[n].type != 'unused'] + if nested_names: + print_attr_list(ynl, nested_names, nested_set, indent + 4) -def print_mode_attrs(mode, mode_spec, attr_set, print_request=True): + +def print_mode_attrs(ynl, mode, mode_spec, attr_set, print_request=True): """Print a given mode (do/dump/event/notify).""" mode_title = mode.capitalize() if print_request and 'request' in mode_spec and 'attributes' in mode_spec['request']: print(f'\n{mode_title} request attributes:') - print_attr_list(mode_spec['request']['attributes'], attr_set) + print_attr_list(ynl, mode_spec['request']['attributes'], attr_set) if 'reply' in mode_spec and 'attributes' in mode_spec['reply']: print(f'\n{mode_title} reply attributes:') - print_attr_list(mode_spec['reply']['attributes'], attr_set) + print_attr_list(ynl, mode_spec['reply']['attributes'], attr_set) if 'attributes' in mode_spec: print(f'\n{mode_title} attributes:') - print_attr_list(mode_spec['attributes'], attr_set) + print_attr_list(ynl, mode_spec['attributes'], attr_set) def main(): @@ -180,13 +197,13 @@ def main(): for mode in ['do', 'dump', 'event']: if mode in op.yaml: - print_mode_attrs(mode, op.yaml[mode], op.attr_set, True) + print_mode_attrs(ynl, mode, op.yaml[mode], op.attr_set, True) if 'notify' in op.yaml: mode_spec = op.yaml['notify'] ref_spec = ynl.msgs.get(mode_spec).yaml.get('do') if ref_spec: - print_mode_attrs('notify', ref_spec, op.attr_set, False) + print_mode_attrs(ynl, 'notify', ref_spec, op.attr_set, False) if 'mcgrp' in op.yaml: print(f"\nMulticast group: {op.yaml['mcgrp']}") From 6c10f1a1c08acd948e4aa7064656ccc24ddabfe3 Mon Sep 17 00:00:00 2001 From: Gal Pressman Date: Tue, 18 Nov 2025 16:32:08 +0200 Subject: [PATCH 0946/1381] tools: ynl: cli: Display enum values in --list-attrs output When listing attributes with --list-attrs, display the actual enum values for attributes that reference an enum type. # ./cli.py --family netdev --list-attrs dev-get [..] - xdp-features: u64 (enum: xdp-act) Flags: basic, redirect, ndo-xmit, xsk-zerocopy, hw-offload, rx-sg, ndo-xmit-sg Bitmask of enabled xdp-features. [..] Reviewed-by: Nimrod Oren Signed-off-by: Gal Pressman Link: https://patch.msgid.link/20251118143208.2380814-4-gal@nvidia.com Signed-off-by: Paolo Abeni --- tools/net/ynl/pyynl/cli.py | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/tools/net/ynl/pyynl/cli.py b/tools/net/ynl/pyynl/cli.py index 6655ee61081a..ff81ff083764 100755 --- a/tools/net/ynl/pyynl/cli.py +++ b/tools/net/ynl/pyynl/cli.py @@ -48,7 +48,13 @@ def print_attr_list(ynl, attr_names, attr_set, indent=2): attr = attr_set.attrs[attr_name] attr_info = f'{prefix}- {attr_name}: {attr.type}' if 'enum' in attr.yaml: - attr_info += f" (enum: {attr.yaml['enum']})" + enum_name = attr.yaml['enum'] + attr_info += f" (enum: {enum_name})" + # Print enum values if available + if enum_name in ynl.consts: + const = ynl.consts[enum_name] + enum_values = list(const.entries.keys()) + attr_info += f"\n{prefix} {const.type.capitalize()}: {', '.join(enum_values)}" # Show nested attributes reference and recursively display them nested_set_name = None From 02aeff20e8f5ab5f7ebc64858239787d420d73c4 Mon Sep 17 00:00:00 2001 From: Buday Csaba Date: Tue, 18 Nov 2025 14:58:52 +0100 Subject: [PATCH 0947/1381] net: mdio: move device reset functions to mdio_device.c The functions mdiobus_register_gpiod() and mdiobus_register_reset() handle the mdio device reset initialization, which belong to mdio_device.c. Move them from mdio_bus.c to mdio_device.c, and rename them to match the corresponding source file: mdio_device_register_gpio() and mdio_device_register_reset(). Remove 'static' qualifiers and declare them in drivers/net/phy/mdio-private.h (new header file). Signed-off-by: Buday Csaba Link: https://patch.msgid.link/5f684838ee897130f21b21beb07695eea4af8988.1763473655.git.buday.csaba@prolan.hu Signed-off-by: Jakub Kicinski --- drivers/net/phy/mdio-private.h | 11 +++++++++++ drivers/net/phy/mdio_bus.c | 32 +++----------------------------- drivers/net/phy/mdio_device.c | 28 ++++++++++++++++++++++++++++ 3 files changed, 42 insertions(+), 29 deletions(-) create mode 100644 drivers/net/phy/mdio-private.h diff --git a/drivers/net/phy/mdio-private.h b/drivers/net/phy/mdio-private.h new file mode 100644 index 000000000000..44e2e0907253 --- /dev/null +++ b/drivers/net/phy/mdio-private.h @@ -0,0 +1,11 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ +#ifndef __MDIO_PRIVATE_H +#define __MDIO_PRIVATE_H + +/* MDIO internal helpers + */ + +int mdio_device_register_reset(struct mdio_device *mdiodev); +int mdio_device_register_gpiod(struct mdio_device *mdiodev); + +#endif /* __MDIO_PRIVATE_H */ diff --git a/drivers/net/phy/mdio_bus.c b/drivers/net/phy/mdio_bus.c index 4354241137d5..575b8bb5bb7a 100644 --- a/drivers/net/phy/mdio_bus.c +++ b/drivers/net/phy/mdio_bus.c @@ -29,37 +29,11 @@ #include #include #include +#include "mdio-private.h" #define CREATE_TRACE_POINTS #include -static int mdiobus_register_gpiod(struct mdio_device *mdiodev) -{ - /* Deassert the optional reset signal */ - mdiodev->reset_gpio = gpiod_get_optional(&mdiodev->dev, - "reset", GPIOD_OUT_LOW); - if (IS_ERR(mdiodev->reset_gpio)) - return PTR_ERR(mdiodev->reset_gpio); - - if (mdiodev->reset_gpio) - gpiod_set_consumer_name(mdiodev->reset_gpio, "PHY reset"); - - return 0; -} - -static int mdiobus_register_reset(struct mdio_device *mdiodev) -{ - struct reset_control *reset; - - reset = reset_control_get_optional_exclusive(&mdiodev->dev, "phy"); - if (IS_ERR(reset)) - return PTR_ERR(reset); - - mdiodev->reset_ctrl = reset; - - return 0; -} - int mdiobus_register_device(struct mdio_device *mdiodev) { int err; @@ -68,11 +42,11 @@ int mdiobus_register_device(struct mdio_device *mdiodev) return -EBUSY; if (mdiodev->flags & MDIO_DEVICE_FLAG_PHY) { - err = mdiobus_register_gpiod(mdiodev); + err = mdio_device_register_gpiod(mdiodev); if (err) return err; - err = mdiobus_register_reset(mdiodev); + err = mdio_device_register_reset(mdiodev); if (err) { gpiod_put(mdiodev->reset_gpio); mdiodev->reset_gpio = NULL; diff --git a/drivers/net/phy/mdio_device.c b/drivers/net/phy/mdio_device.c index f64176e0e197..0e04bddd3619 100644 --- a/drivers/net/phy/mdio_device.c +++ b/drivers/net/phy/mdio_device.c @@ -22,6 +22,7 @@ #include #include #include +#include "mdio-private.h" void mdio_device_free(struct mdio_device *mdiodev) { @@ -118,6 +119,33 @@ void mdio_device_remove(struct mdio_device *mdiodev) } EXPORT_SYMBOL(mdio_device_remove); +int mdio_device_register_gpiod(struct mdio_device *mdiodev) +{ + /* Deassert the optional reset signal */ + mdiodev->reset_gpio = gpiod_get_optional(&mdiodev->dev, + "reset", GPIOD_OUT_LOW); + if (IS_ERR(mdiodev->reset_gpio)) + return PTR_ERR(mdiodev->reset_gpio); + + if (mdiodev->reset_gpio) + gpiod_set_consumer_name(mdiodev->reset_gpio, "PHY reset"); + + return 0; +} + +int mdio_device_register_reset(struct mdio_device *mdiodev) +{ + struct reset_control *reset; + + reset = reset_control_get_optional_exclusive(&mdiodev->dev, "phy"); + if (IS_ERR(reset)) + return PTR_ERR(reset); + + mdiodev->reset_ctrl = reset; + + return 0; +} + void mdio_device_reset(struct mdio_device *mdiodev, int value) { unsigned int d; From acde7ad968f6c536397b39b23e4c70dc466fbc59 Mon Sep 17 00:00:00 2001 From: Buday Csaba Date: Tue, 18 Nov 2025 14:58:53 +0100 Subject: [PATCH 0948/1381] net: mdio: common handling of phy device reset properties Unify the handling of the per device reset properties for `mdio_device`. Merge mdio_device_register_gpiod() and mdio_device_register_reset() into mdio_device_register_reset(), that handles both reset-controllers and reset-gpios. Move reading of the reset firmware properties (reset-assert-us, reset-deassert-us) from fwnode_mdio.c to mdio_device_register_reset(), so all reset related initialization code is kept in one place. Introduce mdio_device_unregister_reset() to release the associated resources. These changes make tracking the reset properties easier. Added kernel-doc for mdio_device_register/unregister_reset(). Signed-off-by: Buday Csaba Link: https://patch.msgid.link/17c216efd7a47be17db104378b6aacfc8741d8b9.1763473655.git.buday.csaba@prolan.hu Signed-off-by: Jakub Kicinski --- drivers/net/mdio/fwnode_mdio.c | 5 ----- drivers/net/phy/mdio-private.h | 2 +- drivers/net/phy/mdio_bus.c | 12 ++-------- drivers/net/phy/mdio_device.c | 40 ++++++++++++++++++++++++++-------- 4 files changed, 34 insertions(+), 25 deletions(-) diff --git a/drivers/net/mdio/fwnode_mdio.c b/drivers/net/mdio/fwnode_mdio.c index 9b41d4697a40..ba7091518265 100644 --- a/drivers/net/mdio/fwnode_mdio.c +++ b/drivers/net/mdio/fwnode_mdio.c @@ -92,11 +92,6 @@ int fwnode_mdiobus_phy_device_register(struct mii_bus *mdio, if (fwnode_property_read_bool(child, "broken-turn-around")) mdio->phy_ignore_ta_mask |= 1 << addr; - fwnode_property_read_u32(child, "reset-assert-us", - &phy->mdio.reset_assert_delay); - fwnode_property_read_u32(child, "reset-deassert-us", - &phy->mdio.reset_deassert_delay); - /* Associate the fwnode with the device structure so it * can be looked up later */ diff --git a/drivers/net/phy/mdio-private.h b/drivers/net/phy/mdio-private.h index 44e2e0907253..8bc6d9088af1 100644 --- a/drivers/net/phy/mdio-private.h +++ b/drivers/net/phy/mdio-private.h @@ -6,6 +6,6 @@ */ int mdio_device_register_reset(struct mdio_device *mdiodev); -int mdio_device_register_gpiod(struct mdio_device *mdiodev); +void mdio_device_unregister_reset(struct mdio_device *mdiodev); #endif /* __MDIO_PRIVATE_H */ diff --git a/drivers/net/phy/mdio_bus.c b/drivers/net/phy/mdio_bus.c index 575b8bb5bb7a..ef041ad66852 100644 --- a/drivers/net/phy/mdio_bus.c +++ b/drivers/net/phy/mdio_bus.c @@ -42,17 +42,10 @@ int mdiobus_register_device(struct mdio_device *mdiodev) return -EBUSY; if (mdiodev->flags & MDIO_DEVICE_FLAG_PHY) { - err = mdio_device_register_gpiod(mdiodev); + err = mdio_device_register_reset(mdiodev); if (err) return err; - err = mdio_device_register_reset(mdiodev); - if (err) { - gpiod_put(mdiodev->reset_gpio); - mdiodev->reset_gpio = NULL; - return err; - } - /* Assert the reset signal */ mdio_device_reset(mdiodev, 1); } @@ -68,8 +61,7 @@ int mdiobus_unregister_device(struct mdio_device *mdiodev) if (mdiodev->bus->mdio_map[mdiodev->addr] != mdiodev) return -EINVAL; - gpiod_put(mdiodev->reset_gpio); - reset_control_put(mdiodev->reset_ctrl); + mdio_device_unregister_reset(mdiodev); mdiodev->bus->mdio_map[mdiodev->addr] = NULL; diff --git a/drivers/net/phy/mdio_device.c b/drivers/net/phy/mdio_device.c index 0e04bddd3619..a4d9c6ccf876 100644 --- a/drivers/net/phy/mdio_device.c +++ b/drivers/net/phy/mdio_device.c @@ -119,8 +119,17 @@ void mdio_device_remove(struct mdio_device *mdiodev) } EXPORT_SYMBOL(mdio_device_remove); -int mdio_device_register_gpiod(struct mdio_device *mdiodev) +/** + * mdio_device_register_reset - Read and initialize the reset properties of + * an mdio device + * @mdiodev: mdio_device structure + * + * Return: Zero if successful, negative error code on failure + */ +int mdio_device_register_reset(struct mdio_device *mdiodev) { + struct reset_control *reset; + /* Deassert the optional reset signal */ mdiodev->reset_gpio = gpiod_get_optional(&mdiodev->dev, "reset", GPIOD_OUT_LOW); @@ -130,22 +139,35 @@ int mdio_device_register_gpiod(struct mdio_device *mdiodev) if (mdiodev->reset_gpio) gpiod_set_consumer_name(mdiodev->reset_gpio, "PHY reset"); - return 0; -} - -int mdio_device_register_reset(struct mdio_device *mdiodev) -{ - struct reset_control *reset; - reset = reset_control_get_optional_exclusive(&mdiodev->dev, "phy"); - if (IS_ERR(reset)) + if (IS_ERR(reset)) { + gpiod_put(mdiodev->reset_gpio); + mdiodev->reset_gpio = NULL; return PTR_ERR(reset); + } mdiodev->reset_ctrl = reset; + /* Read optional firmware properties */ + fwnode_property_read_u32(dev_fwnode(&mdiodev->dev), "reset-assert-us", + &mdiodev->reset_assert_delay); + fwnode_property_read_u32(dev_fwnode(&mdiodev->dev), "reset-deassert-us", + &mdiodev->reset_deassert_delay); + return 0; } +/** + * mdio_device_unregister_reset - uninitialize the reset properties of + * an mdio device + * @mdiodev: mdio_device structure + */ +void mdio_device_unregister_reset(struct mdio_device *mdiodev) +{ + gpiod_put(mdiodev->reset_gpio); + reset_control_put(mdiodev->reset_ctrl); +} + void mdio_device_reset(struct mdio_device *mdiodev, int value) { unsigned int d; From e5a440bf020ef9ee11d442159bd394d66250db30 Mon Sep 17 00:00:00 2001 From: Buday Csaba Date: Tue, 18 Nov 2025 14:58:54 +0100 Subject: [PATCH 0949/1381] net: mdio: improve reset handling in mdio_device.c Change fwnode_property_read_u32() in mdio_device_register_reset() to device_property_read_u32(), which is more appropriate here. Make mdio_device_unregister_reset() truly reverse mdio_device_register_reset() by setting the internal fields to their default values. Signed-off-by: Buday Csaba Link: https://patch.msgid.link/641df1488517ae71ba10158ec1e38424211d8651.1763473655.git.buday.csaba@prolan.hu Signed-off-by: Jakub Kicinski --- drivers/net/phy/mdio_device.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/drivers/net/phy/mdio_device.c b/drivers/net/phy/mdio_device.c index a4d9c6ccf876..fd0e16dbc5cc 100644 --- a/drivers/net/phy/mdio_device.c +++ b/drivers/net/phy/mdio_device.c @@ -149,9 +149,9 @@ int mdio_device_register_reset(struct mdio_device *mdiodev) mdiodev->reset_ctrl = reset; /* Read optional firmware properties */ - fwnode_property_read_u32(dev_fwnode(&mdiodev->dev), "reset-assert-us", + device_property_read_u32(&mdiodev->dev, "reset-assert-us", &mdiodev->reset_assert_delay); - fwnode_property_read_u32(dev_fwnode(&mdiodev->dev), "reset-deassert-us", + device_property_read_u32(&mdiodev->dev, "reset-deassert-us", &mdiodev->reset_deassert_delay); return 0; @@ -165,7 +165,11 @@ int mdio_device_register_reset(struct mdio_device *mdiodev) void mdio_device_unregister_reset(struct mdio_device *mdiodev) { gpiod_put(mdiodev->reset_gpio); + mdiodev->reset_gpio = NULL; reset_control_put(mdiodev->reset_ctrl); + mdiodev->reset_ctrl = NULL; + mdiodev->reset_assert_delay = 0; + mdiodev->reset_deassert_delay = 0; } void mdio_device_reset(struct mdio_device *mdiodev, int value) From 6d5dea68246ecb190a50a7fecbaf7f8c1ddb15e4 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 19 Nov 2025 08:48:12 +0000 Subject: [PATCH 0950/1381] tcp: tcp_moderate_rcvbuf is only used in rx path sysctl_tcp_moderate_rcvbuf is only used from tcp_rcvbuf_grow(). Move it to netns_ipv4_read_rx group. Remove various CACHELINE_ASSERT_GROUP_SIZE() from netns_ipv4_struct_check(), as they have no real benefit but cause pain for all changes. Signed-off-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20251119084813.3684576-2-edumazet@google.com Signed-off-by: Jakub Kicinski --- .../networking/net_cachelines/netns_ipv4_sysctl.rst | 2 +- include/net/netns/ipv4.h | 2 +- net/core/net_namespace.c | 9 ++------- 3 files changed, 4 insertions(+), 9 deletions(-) diff --git a/Documentation/networking/net_cachelines/netns_ipv4_sysctl.rst b/Documentation/networking/net_cachelines/netns_ipv4_sysctl.rst index 6e7b20afd2d4..5d5d54fb6ab1 100644 --- a/Documentation/networking/net_cachelines/netns_ipv4_sysctl.rst +++ b/Documentation/networking/net_cachelines/netns_ipv4_sysctl.rst @@ -102,7 +102,7 @@ u8 sysctl_tcp_app_win u8 sysctl_tcp_frto tcp_enter_loss u8 sysctl_tcp_nometrics_save TCP_LAST_ACK/tcp_update_metrics u8 sysctl_tcp_no_ssthresh_metrics_save TCP_LAST_ACK/tcp_(update/init)_metrics -u8 sysctl_tcp_moderate_rcvbuf read_mostly read_mostly tcp_tso_should_defer(tx);tcp_rcv_space_adjust(rx) +u8 sysctl_tcp_moderate_rcvbuf read_mostly tcp_rcvbuf_grow() u8 sysctl_tcp_tso_win_divisor read_mostly tcp_tso_should_defer(tcp_write_xmit) u8 sysctl_tcp_workaround_signed_windows tcp_select_window int sysctl_tcp_limit_output_bytes read_mostly tcp_small_queue_check(tcp_write_xmit) diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index de9d36acc8e2..11837d3ccc0a 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -74,11 +74,11 @@ struct netns_ipv4 { /* TXRX readonly hotpath cache lines */ __cacheline_group_begin(netns_ipv4_read_txrx); - u8 sysctl_tcp_moderate_rcvbuf; __cacheline_group_end(netns_ipv4_read_txrx); /* RX readonly hotpath cache line */ __cacheline_group_begin(netns_ipv4_read_rx); + u8 sysctl_tcp_moderate_rcvbuf; u8 sysctl_ip_early_demux; u8 sysctl_tcp_early_demux; u8 sysctl_tcp_l3mdev_accept; diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c index adcfef55a66f..c8adbbe01451 100644 --- a/net/core/net_namespace.c +++ b/net/core/net_namespace.c @@ -1223,14 +1223,10 @@ static void __init netns_ipv4_struct_check(void) sysctl_tcp_wmem); CACHELINE_ASSERT_GROUP_MEMBER(struct netns_ipv4, netns_ipv4_read_tx, sysctl_ip_fwd_use_pmtu); - CACHELINE_ASSERT_GROUP_SIZE(struct netns_ipv4, netns_ipv4_read_tx, 33); - - /* TXRX readonly hotpath cache lines */ - CACHELINE_ASSERT_GROUP_MEMBER(struct netns_ipv4, netns_ipv4_read_txrx, - sysctl_tcp_moderate_rcvbuf); - CACHELINE_ASSERT_GROUP_SIZE(struct netns_ipv4, netns_ipv4_read_txrx, 1); /* RX readonly hotpath cache line */ + CACHELINE_ASSERT_GROUP_MEMBER(struct netns_ipv4, netns_ipv4_read_rx, + sysctl_tcp_moderate_rcvbuf); CACHELINE_ASSERT_GROUP_MEMBER(struct netns_ipv4, netns_ipv4_read_rx, sysctl_ip_early_demux); CACHELINE_ASSERT_GROUP_MEMBER(struct netns_ipv4, netns_ipv4_read_rx, @@ -1241,7 +1237,6 @@ static void __init netns_ipv4_struct_check(void) sysctl_tcp_reordering); CACHELINE_ASSERT_GROUP_MEMBER(struct netns_ipv4, netns_ipv4_read_rx, sysctl_tcp_rmem); - CACHELINE_ASSERT_GROUP_SIZE(struct netns_ipv4, netns_ipv4_read_rx, 22); } #endif From ecfea98b7d0d56c5bf2df3fc02c5501afa5cef6f Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 19 Nov 2025 08:48:13 +0000 Subject: [PATCH 0951/1381] tcp: add net.ipv4.tcp_rcvbuf_low_rtt This is a follow up of commit aa251c84636c ("tcp: fix too slow tcp_rcvbuf_grow() action") which brought again the issue that I tried to fix in commit 65c5287892e9 ("tcp: fix sk_rcvbuf overshoot") We also recently increased tcp_rmem[2] to 32 MB in commit 572be9bf9d0d ("tcp: increase tcp_rmem[2] to 32 MB") Idea of this patch is to not let tcp_rcvbuf_grow() grow sk->sk_rcvbuf too fast for small RTT flows. If sk->sk_rcvbuf is too big, this can force NIC driver to not recycle pages from their page pool, and also can cause cache evictions for DDIO enabled cpus/NIC, as receivers are usually slower than senders. Add net.ipv4.tcp_rcvbuf_low_rtt sysctl, set by default to 1000 usec (1 ms) If RTT if smaller than the sysctl value, use the RTT/tcp_rcvbuf_low_rtt ratio to control sk_rcvbuf inflation. Tested: Pair of hosts with a 200Gbit IDPF NIC. Using netperf/netserver Client initiates 8 TCP bulk flows, asking netserver to use CPU #10 only. super_netperf 8 -H server -T,10 -l 30 On server, use perf -e tcp:tcp_rcvbuf_grow while test is running. Before: sysctl -w net.ipv4.tcp_rcvbuf_low_rtt=1 perf record -a -e tcp:tcp_rcvbuf_grow sleep 30 ; perf script|tail -20|cut -c30-230 1153.051201: tcp:tcp_rcvbuf_grow: time=398 rtt_us=382 copied=6905856 inq=180224 space=6115328 ooo=0 scaling_ratio=240 rcvbuf=27666235 rcv_ssthresh=25878235 window_clamp=25937095 rcv_wnd=25600000 famil 1153.138752: tcp:tcp_rcvbuf_grow: time=446 rtt_us=413 copied=5529600 inq=180224 space=4505600 ooo=0 scaling_ratio=240 rcvbuf=23068672 rcv_ssthresh=21571860 window_clamp=21626880 rcv_wnd=21286912 famil 1153.361484: tcp:tcp_rcvbuf_grow: time=415 rtt_us=380 copied=7061504 inq=204800 space=6725632 ooo=0 scaling_ratio=240 rcvbuf=27666235 rcv_ssthresh=25878235 window_clamp=25937095 rcv_wnd=25600000 famil 1153.457642: tcp:tcp_rcvbuf_grow: time=483 rtt_us=421 copied=5885952 inq=720896 space=4407296 ooo=0 scaling_ratio=240 rcvbuf=23763511 rcv_ssthresh=22223271 window_clamp=22278291 rcv_wnd=21430272 famil 1153.466002: tcp:tcp_rcvbuf_grow: time=308 rtt_us=281 copied=3244032 inq=180224 space=2883584 ooo=0 scaling_ratio=240 rcvbuf=44854314 rcv_ssthresh=41992059 window_clamp=42050919 rcv_wnd=41713664 famil 1153.747792: tcp:tcp_rcvbuf_grow: time=394 rtt_us=332 copied=4460544 inq=585728 space=3063808 ooo=0 scaling_ratio=240 rcvbuf=44854314 rcv_ssthresh=41992059 window_clamp=42050919 rcv_wnd=41373696 famil 1154.260747: tcp:tcp_rcvbuf_grow: time=652 rtt_us=226 copied=10977280 inq=737280 space=9486336 ooo=0 scaling_ratio=240 rcvbuf=31165538 rcv_ssthresh=29197743 window_clamp=29217691 rcv_wnd=28368896 fami 1154.375019: tcp:tcp_rcvbuf_grow: time=461 rtt_us=443 copied=7573504 inq=507904 space=6856704 ooo=0 scaling_ratio=240 rcvbuf=27666235 rcv_ssthresh=25878235 window_clamp=25937095 rcv_wnd=25288704 famil 1154.463072: tcp:tcp_rcvbuf_grow: time=494 rtt_us=408 copied=7983104 inq=200704 space=7065600 ooo=0 scaling_ratio=240 rcvbuf=27666235 rcv_ssthresh=25878235 window_clamp=25937095 rcv_wnd=25579520 famil 1154.474658: tcp:tcp_rcvbuf_grow: time=507 rtt_us=459 copied=5586944 inq=540672 space=4718592 ooo=0 scaling_ratio=240 rcvbuf=17852266 rcv_ssthresh=16692999 window_clamp=16736499 rcv_wnd=16056320 famil 1154.584657: tcp:tcp_rcvbuf_grow: time=494 rtt_us=427 copied=8126464 inq=204800 space=7782400 ooo=0 scaling_ratio=240 rcvbuf=27666235 rcv_ssthresh=25878235 window_clamp=25937095 rcv_wnd=25600000 famil 1154.702117: tcp:tcp_rcvbuf_grow: time=480 rtt_us=406 copied=5734400 inq=180224 space=5349376 ooo=0 scaling_ratio=240 rcvbuf=23068672 rcv_ssthresh=21571860 window_clamp=21626880 rcv_wnd=21286912 famil 1155.941595: tcp:tcp_rcvbuf_grow: time=717 rtt_us=670 copied=11042816 inq=3784704 space=7159808 ooo=0 scaling_ratio=240 rcvbuf=19581357 rcv_ssthresh=18333222 window_clamp=18357522 rcv_wnd=14614528 fam 1156.384735: tcp:tcp_rcvbuf_grow: time=529 rtt_us=473 copied=9011200 inq=180224 space=7258112 ooo=0 scaling_ratio=240 rcvbuf=19581357 rcv_ssthresh=18333222 window_clamp=18357522 rcv_wnd=18018304 famil 1157.821676: tcp:tcp_rcvbuf_grow: time=529 rtt_us=272 copied=8224768 inq=602112 space=6545408 ooo=0 scaling_ratio=240 rcvbuf=67000000 rcv_ssthresh=62793576 window_clamp=62812500 rcv_wnd=62115840 famil 1158.906379: tcp:tcp_rcvbuf_grow: time=710 rtt_us=445 copied=11845632 inq=540672 space=10240000 ooo=0 scaling_ratio=240 rcvbuf=31165538 rcv_ssthresh=29205935 window_clamp=29217691 rcv_wnd=28536832 fam 1164.600160: tcp:tcp_rcvbuf_grow: time=841 rtt_us=430 copied=12976128 inq=1290240 space=11304960 ooo=0 scaling_ratio=240 rcvbuf=31165538 rcv_ssthresh=29212591 window_clamp=29217691 rcv_wnd=27856896 fa 1165.163572: tcp:tcp_rcvbuf_grow: time=845 rtt_us=800 copied=12632064 inq=540672 space=7921664 ooo=0 scaling_ratio=240 rcvbuf=27666235 rcv_ssthresh=25912795 window_clamp=25937095 rcv_wnd=25260032 fami 1165.653464: tcp:tcp_rcvbuf_grow: time=388 rtt_us=309 copied=4493312 inq=180224 space=3874816 ooo=0 scaling_ratio=240 rcvbuf=44854314 rcv_ssthresh=41995899 window_clamp=42050919 rcv_wnd=41713664 famil 1166.651211: tcp:tcp_rcvbuf_grow: time=556 rtt_us=553 copied=6328320 inq=540672 space=5554176 ooo=0 scaling_ratio=240 rcvbuf=23068672 rcv_ssthresh=21571860 window_clamp=21626880 rcv_wnd=20946944 famil After: sysctl -w net.ipv4.tcp_rcvbuf_low_rtt=1000 perf record -a -e tcp:tcp_rcvbuf_grow sleep 30 ; perf script|tail -20|cut -c30-230 1457.053149: tcp:tcp_rcvbuf_grow: time=128 rtt_us=24 copied=1441792 inq=40960 space=1269760 ooo=0 scaling_ratio=240 rcvbuf=2960741 rcv_ssthresh=2605474 window_clamp=2775694 rcv_wnd=2568192 family=AF_I 1458.000778: tcp:tcp_rcvbuf_grow: time=128 rtt_us=31 copied=1441792 inq=24576 space=1400832 ooo=0 scaling_ratio=240 rcvbuf=3060163 rcv_ssthresh=2810042 window_clamp=2868902 rcv_wnd=2674688 family=AF_I 1458.088059: tcp:tcp_rcvbuf_grow: time=190 rtt_us=110 copied=3227648 inq=385024 space=2781184 ooo=0 scaling_ratio=240 rcvbuf=6728240 rcv_ssthresh=6252705 window_clamp=6307725 rcv_wnd=5799936 family=AF 1458.148549: tcp:tcp_rcvbuf_grow: time=232 rtt_us=129 copied=3956736 inq=237568 space=2842624 ooo=0 scaling_ratio=240 rcvbuf=6731333 rcv_ssthresh=6252705 window_clamp=6310624 rcv_wnd=5918720 family=AF 1458.466861: tcp:tcp_rcvbuf_grow: time=193 rtt_us=83 copied=2949120 inq=180224 space=2457600 ooo=0 scaling_ratio=240 rcvbuf=5751438 rcv_ssthresh=5357689 window_clamp=5391973 rcv_wnd=5054464 family=AF_ 1458.775476: tcp:tcp_rcvbuf_grow: time=257 rtt_us=127 copied=4304896 inq=352256 space=3346432 ooo=0 scaling_ratio=240 rcvbuf=8067131 rcv_ssthresh=7523275 window_clamp=7562935 rcv_wnd=7061504 family=AF 1458.776631: tcp:tcp_rcvbuf_grow: time=200 rtt_us=96 copied=3260416 inq=143360 space=2768896 ooo=0 scaling_ratio=240 rcvbuf=6397256 rcv_ssthresh=5938567 window_clamp=5997427 rcv_wnd=5828608 family=AF_ 1459.707973: tcp:tcp_rcvbuf_grow: time=215 rtt_us=96 copied=2506752 inq=163840 space=1388544 ooo=0 scaling_ratio=240 rcvbuf=3068867 rcv_ssthresh=2768282 window_clamp=2877062 rcv_wnd=2555904 family=AF_ 1460.246494: tcp:tcp_rcvbuf_grow: time=231 rtt_us=80 copied=3756032 inq=204800 space=3117056 ooo=0 scaling_ratio=240 rcvbuf=7288091 rcv_ssthresh=6773725 window_clamp=6832585 rcv_wnd=6471680 family=AF_ 1460.714596: tcp:tcp_rcvbuf_grow: time=270 rtt_us=110 copied=4714496 inq=311296 space=3719168 ooo=0 scaling_ratio=240 rcvbuf=8957739 rcv_ssthresh=8339020 window_clamp=8397880 rcv_wnd=7933952 family=AF 1462.029977: tcp:tcp_rcvbuf_grow: time=101 rtt_us=19 copied=1105920 inq=40960 space=1036288 ooo=0 scaling_ratio=240 rcvbuf=2338970 rcv_ssthresh=2091684 window_clamp=2192784 rcv_wnd=1986560 family=AF_I 1462.802385: tcp:tcp_rcvbuf_grow: time=89 rtt_us=45 copied=1069056 inq=0 space=1064960 ooo=0 scaling_ratio=240 rcvbuf=2338970 rcv_ssthresh=2091684 window_clamp=2192784 rcv_wnd=2035712 family=AF_INET6 1462.918648: tcp:tcp_rcvbuf_grow: time=105 rtt_us=33 copied=1441792 inq=180224 space=1069056 ooo=0 scaling_ratio=240 rcvbuf=2383282 rcv_ssthresh=2091684 window_clamp=2234326 rcv_wnd=1896448 family=AF_ 1463.222533: tcp:tcp_rcvbuf_grow: time=273 rtt_us=144 copied=4603904 inq=385024 space=3469312 ooo=0 scaling_ratio=240 rcvbuf=8422564 rcv_ssthresh=7891053 window_clamp=7896153 rcv_wnd=7409664 family=AF 1466.519312: tcp:tcp_rcvbuf_grow: time=130 rtt_us=23 copied=1343488 inq=0 space=1261568 ooo=0 scaling_ratio=240 rcvbuf=2780158 rcv_ssthresh=2493778 window_clamp=2606398 rcv_wnd=2494464 family=AF_INET6 1466.681003: tcp:tcp_rcvbuf_grow: time=128 rtt_us=21 copied=1441792 inq=12288 space=1343488 ooo=0 scaling_ratio=240 rcvbuf=2932027 rcv_ssthresh=2578555 window_clamp=2748775 rcv_wnd=2568192 family=AF_I 1470.689959: tcp:tcp_rcvbuf_grow: time=255 rtt_us=122 copied=3932160 inq=204800 space=3551232 ooo=0 scaling_ratio=240 rcvbuf=8182038 rcv_ssthresh=7647384 window_clamp=7670660 rcv_wnd=7442432 family=AF 1471.754154: tcp:tcp_rcvbuf_grow: time=188 rtt_us=95 copied=2138112 inq=577536 space=1429504 ooo=0 scaling_ratio=240 rcvbuf=3113650 rcv_ssthresh=2806426 window_clamp=2919046 rcv_wnd=2248704 family=AF_ 1476.813542: tcp:tcp_rcvbuf_grow: time=269 rtt_us=99 copied=3088384 inq=180224 space=2564096 ooo=0 scaling_ratio=240 rcvbuf=6219470 rcv_ssthresh=5771893 window_clamp=5830753 rcv_wnd=5509120 family=AF_ 1477.738309: tcp:tcp_rcvbuf_grow: time=166 rtt_us=54 copied=1777664 inq=180224 space=1417216 ooo=0 scaling_ratio=240 rcvbuf=3117118 rcv_ssthresh=2874958 window_clamp=2922298 rcv_wnd=2613248 family=AF_ We can see sk_rcvbuf values are much smaller, and that rtt_us (estimation of rtt from a receiver point of view) is kept small, instead of being bloated. No difference in throughput. Signed-off-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Tested-by: Paolo Abeni Link: https://patch.msgid.link/20251119084813.3684576-3-edumazet@google.com Signed-off-by: Jakub Kicinski --- Documentation/networking/ip-sysctl.rst | 10 ++++++++++ .../net_cachelines/netns_ipv4_sysctl.rst | 1 + include/net/netns/ipv4.h | 1 + net/core/net_namespace.c | 2 ++ net/ipv4/sysctl_net_ipv4.c | 9 +++++++++ net/ipv4/tcp_input.c | 18 ++++++++++++++---- net/ipv4/tcp_ipv4.c | 1 + 7 files changed, 38 insertions(+), 4 deletions(-) diff --git a/Documentation/networking/ip-sysctl.rst b/Documentation/networking/ip-sysctl.rst index f4ad739a6b53..bc9a01606daf 100644 --- a/Documentation/networking/ip-sysctl.rst +++ b/Documentation/networking/ip-sysctl.rst @@ -673,6 +673,16 @@ tcp_moderate_rcvbuf - BOOLEAN Default: 1 (enabled) +tcp_rcvbuf_low_rtt - INTEGER + rcvbuf autotuning can over estimate final socket rcvbuf, which + can lead to cache trashing for high throughput flows. + + For small RTT flows (below tcp_rcvbuf_low_rtt usecs), we can relax + rcvbuf growth: Few additional ms to reach the final (and smaller) + rcvbuf is a good tradeoff. + + Default : 1000 (1 ms) + tcp_mtu_probing - INTEGER Controls TCP Packetization-Layer Path MTU Discovery. Takes three values: diff --git a/Documentation/networking/net_cachelines/netns_ipv4_sysctl.rst b/Documentation/networking/net_cachelines/netns_ipv4_sysctl.rst index 5d5d54fb6ab1..beaf1880a19b 100644 --- a/Documentation/networking/net_cachelines/netns_ipv4_sysctl.rst +++ b/Documentation/networking/net_cachelines/netns_ipv4_sysctl.rst @@ -103,6 +103,7 @@ u8 sysctl_tcp_frto u8 sysctl_tcp_nometrics_save TCP_LAST_ACK/tcp_update_metrics u8 sysctl_tcp_no_ssthresh_metrics_save TCP_LAST_ACK/tcp_(update/init)_metrics u8 sysctl_tcp_moderate_rcvbuf read_mostly tcp_rcvbuf_grow() +u32 sysctl_tcp_rcvbuf_low_rtt read_mostly tcp_rcvbuf_grow() u8 sysctl_tcp_tso_win_divisor read_mostly tcp_tso_should_defer(tcp_write_xmit) u8 sysctl_tcp_workaround_signed_windows tcp_select_window int sysctl_tcp_limit_output_bytes read_mostly tcp_small_queue_check(tcp_write_xmit) diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index 11837d3ccc0a..2dbd46fc4734 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -85,6 +85,7 @@ struct netns_ipv4 { /* 3 bytes hole, try to pack */ int sysctl_tcp_reordering; int sysctl_tcp_rmem[3]; + int sysctl_tcp_rcvbuf_low_rtt; __cacheline_group_end(netns_ipv4_read_rx); struct inet_timewait_death_row tcp_death_row; diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c index c8adbbe01451..dfad7c03b809 100644 --- a/net/core/net_namespace.c +++ b/net/core/net_namespace.c @@ -1227,6 +1227,8 @@ static void __init netns_ipv4_struct_check(void) /* RX readonly hotpath cache line */ CACHELINE_ASSERT_GROUP_MEMBER(struct netns_ipv4, netns_ipv4_read_rx, sysctl_tcp_moderate_rcvbuf); + CACHELINE_ASSERT_GROUP_MEMBER(struct netns_ipv4, netns_ipv4_read_rx, + sysctl_tcp_rcvbuf_low_rtt); CACHELINE_ASSERT_GROUP_MEMBER(struct netns_ipv4, netns_ipv4_read_rx, sysctl_ip_early_demux); CACHELINE_ASSERT_GROUP_MEMBER(struct netns_ipv4, netns_ipv4_read_rx, diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 35367f8e2da3..a1a50a5c80dc 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -1342,6 +1342,15 @@ static struct ctl_table ipv4_net_table[] = { .mode = 0644, .proc_handler = proc_dou8vec_minmax, }, + { + .procname = "tcp_rcvbuf_low_rtt", + .data = &init_net.ipv4.sysctl_tcp_rcvbuf_low_rtt, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = SYSCTL_ZERO, + .extra2 = SYSCTL_INT_MAX, + }, { .procname = "tcp_tso_win_divisor", .data = &init_net.ipv4.sysctl_tcp_tso_win_divisor, diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 9df5d7515605..198f8a0d37be 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -896,6 +896,7 @@ void tcp_rcvbuf_grow(struct sock *sk, u32 newval) const struct net *net = sock_net(sk); struct tcp_sock *tp = tcp_sk(sk); u32 rcvwin, rcvbuf, cap, oldval; + u32 rtt_threshold, rtt_us; u64 grow; oldval = tp->rcvq_space.space; @@ -908,10 +909,19 @@ void tcp_rcvbuf_grow(struct sock *sk, u32 newval) /* DRS is always one RTT late. */ rcvwin = newval << 1; - /* slow start: allow the sender to double its rate. */ - grow = (u64)rcvwin * (newval - oldval); - do_div(grow, oldval); - rcvwin += grow << 1; + rtt_us = tp->rcv_rtt_est.rtt_us >> 3; + rtt_threshold = READ_ONCE(net->ipv4.sysctl_tcp_rcvbuf_low_rtt); + if (rtt_us < rtt_threshold) { + /* For small RTT, we set @grow to rcvwin * rtt_us/rtt_threshold. + * It might take few additional ms to reach 'line rate', + * but will avoid sk_rcvbuf inflation and poor cache use. + */ + grow = div_u64((u64)rcvwin * rtt_us, rtt_threshold); + } else { + /* slow start: allow the sender to double its rate. */ + grow = div_u64(((u64)rcvwin << 1) * (newval - oldval), oldval); + } + rcvwin += grow; if (!RB_EMPTY_ROOT(&tp->out_of_order_queue)) rcvwin += TCP_SKB_CB(tp->ooo_last_skb)->end_seq - tp->rcv_nxt; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 6fcaecb67284..e0bb8d9e2d9c 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -3566,6 +3566,7 @@ static int __net_init tcp_sk_init(struct net *net) net->ipv4.sysctl_tcp_adv_win_scale = 1; net->ipv4.sysctl_tcp_frto = 2; net->ipv4.sysctl_tcp_moderate_rcvbuf = 1; + net->ipv4.sysctl_tcp_rcvbuf_low_rtt = USEC_PER_MSEC; /* This limits the percentage of the congestion window which we * will allow a single TSO frame to consume. Building TSO frames * which are too large can cause TCP streams to be bursty. From 85081acc6b1188f2a6e5e605dc644225fcdf327f Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Wed, 19 Nov 2025 10:03:50 +0000 Subject: [PATCH 0952/1381] net: stmmac: pass struct device to init()/exit() methods As struct plat_stmmacenet_data is not platform_device specific, pass a struct device into the init() and exit() methods to allow them to become independent of the underlying device. Signed-off-by: Russell King (Oracle) Acked-by: Chen-Yu Tsai Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/E1vLf2U-0000000FMN2-0SLg@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../ethernet/stmicro/stmmac/dwmac-anarion.c | 4 +-- .../ethernet/stmicro/stmmac/dwmac-eic7700.c | 4 +-- .../ethernet/stmicro/stmmac/dwmac-loongson1.c | 12 ++++---- .../stmicro/stmmac/dwmac-renesas-gbeth.c | 4 +-- .../net/ethernet/stmicro/stmmac/dwmac-s32.c | 14 +++++----- .../ethernet/stmicro/stmmac/dwmac-socfpga.c | 2 +- .../net/ethernet/stmicro/stmmac/dwmac-sti.c | 4 +-- .../net/ethernet/stmicro/stmmac/dwmac-sun8i.c | 10 +++---- .../net/ethernet/stmicro/stmmac/dwmac-sunxi.c | 4 +-- .../net/ethernet/stmicro/stmmac/dwmac-thead.c | 2 +- .../ethernet/stmicro/stmmac/stmmac_platform.c | 28 ++++++++++--------- include/linux/stmmac.h | 4 +-- 12 files changed, 47 insertions(+), 45 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-anarion.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-anarion.c index 84072c8ed741..5e0fc31762d9 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-anarion.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-anarion.c @@ -34,7 +34,7 @@ static void gmac_write_reg(struct anarion_gmac *gmac, uint8_t reg, uint32_t val) writel(val, gmac->ctl_block + reg); } -static int anarion_gmac_init(struct platform_device *pdev, void *priv) +static int anarion_gmac_init(struct device *dev, void *priv) { uint32_t sw_config; struct anarion_gmac *gmac = priv; @@ -52,7 +52,7 @@ static int anarion_gmac_init(struct platform_device *pdev, void *priv) return 0; } -static void anarion_gmac_exit(struct platform_device *pdev, void *priv) +static void anarion_gmac_exit(struct device *dev, void *priv) { struct anarion_gmac *gmac = priv; diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-eic7700.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-eic7700.c index 1dcf2037001e..bcb8e000e720 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-eic7700.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-eic7700.c @@ -58,14 +58,14 @@ static int eic7700_clks_config(void *priv, bool enabled) return ret; } -static int eic7700_dwmac_init(struct platform_device *pdev, void *priv) +static int eic7700_dwmac_init(struct device *dev, void *priv) { struct eic7700_qos_priv *dwc = priv; return eic7700_clks_config(dwc, true); } -static void eic7700_dwmac_exit(struct platform_device *pdev, void *priv) +static void eic7700_dwmac_exit(struct device *dev, void *priv) { struct eic7700_qos_priv *dwc = priv; diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson1.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson1.c index 894ee66f5c9b..de9aba756aac 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson1.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson1.c @@ -48,7 +48,7 @@ struct ls1x_dwmac { struct ls1x_data { int (*setup)(struct platform_device *pdev, struct plat_stmmacenet_data *plat_dat); - int (*init)(struct platform_device *pdev, void *bsp_priv); + int (*init)(struct device *dev, void *bsp_priv); }; static int ls1b_dwmac_setup(struct platform_device *pdev, @@ -79,7 +79,7 @@ static int ls1b_dwmac_setup(struct platform_device *pdev, return 0; } -static int ls1b_dwmac_syscon_init(struct platform_device *pdev, void *priv) +static int ls1b_dwmac_syscon_init(struct device *dev, void *priv) { struct ls1x_dwmac *dwmac = priv; struct plat_stmmacenet_data *plat = dwmac->plat_dat; @@ -98,7 +98,7 @@ static int ls1b_dwmac_syscon_init(struct platform_device *pdev, void *priv) GMAC0_USE_TXCLK | GMAC0_USE_PWM01); break; default: - dev_err(&pdev->dev, "Unsupported PHY mode %u\n", + dev_err(dev, "Unsupported PHY mode %u\n", plat->phy_interface); return -EOPNOTSUPP; } @@ -122,7 +122,7 @@ static int ls1b_dwmac_syscon_init(struct platform_device *pdev, void *priv) GMAC1_USE_TXCLK | GMAC1_USE_PWM23); break; default: - dev_err(&pdev->dev, "Unsupported PHY mode %u\n", + dev_err(dev, "Unsupported PHY mode %u\n", plat->phy_interface); return -EOPNOTSUPP; } @@ -133,7 +133,7 @@ static int ls1b_dwmac_syscon_init(struct platform_device *pdev, void *priv) return 0; } -static int ls1c_dwmac_syscon_init(struct platform_device *pdev, void *priv) +static int ls1c_dwmac_syscon_init(struct device *dev, void *priv) { struct ls1x_dwmac *dwmac = priv; struct plat_stmmacenet_data *plat = dwmac->plat_dat; @@ -143,7 +143,7 @@ static int ls1c_dwmac_syscon_init(struct platform_device *pdev, void *priv) phy_intf_sel = stmmac_get_phy_intf_sel(plat->phy_interface); if (phy_intf_sel != PHY_INTF_SEL_GMII_MII && phy_intf_sel != PHY_INTF_SEL_RMII) { - dev_err(&pdev->dev, "Unsupported PHY-mode %u\n", + dev_err(dev, "Unsupported PHY-mode %u\n", plat->phy_interface); return -EOPNOTSUPP; } diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-renesas-gbeth.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-renesas-gbeth.c index bc7bb975803c..be7f5eb2cdcf 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-renesas-gbeth.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-renesas-gbeth.c @@ -91,7 +91,7 @@ static struct phylink_pcs *renesas_gmac_select_pcs(struct stmmac_priv *priv, return priv->hw->phylink_pcs; } -static int renesas_gbeth_init(struct platform_device *pdev, void *priv) +static int renesas_gbeth_init(struct device *dev, void *priv) { struct plat_stmmacenet_data *plat_dat; struct renesas_gbeth *gbeth = priv; @@ -113,7 +113,7 @@ static int renesas_gbeth_init(struct platform_device *pdev, void *priv) return ret; } -static void renesas_gbeth_exit(struct platform_device *pdev, void *priv) +static void renesas_gbeth_exit(struct device *dev, void *priv) { struct plat_stmmacenet_data *plat_dat; struct renesas_gbeth *gbeth = priv; diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-s32.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-s32.c index 2b7ad64bfdf7..5a485ee98fa7 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-s32.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-s32.c @@ -47,7 +47,7 @@ static int s32_gmac_write_phy_intf_select(struct s32_priv_data *gmac) return 0; } -static int s32_gmac_init(struct platform_device *pdev, void *priv) +static int s32_gmac_init(struct device *dev, void *priv) { struct s32_priv_data *gmac = priv; int ret; @@ -55,31 +55,31 @@ static int s32_gmac_init(struct platform_device *pdev, void *priv) /* Set initial TX interface clock */ ret = clk_prepare_enable(gmac->tx_clk); if (ret) { - dev_err(&pdev->dev, "Can't enable tx clock\n"); + dev_err(dev, "Can't enable tx clock\n"); return ret; } ret = clk_set_rate(gmac->tx_clk, GMAC_INTF_RATE_125M); if (ret) { - dev_err(&pdev->dev, "Can't set tx clock\n"); + dev_err(dev, "Can't set tx clock\n"); goto err_tx_disable; } /* Set initial RX interface clock */ ret = clk_prepare_enable(gmac->rx_clk); if (ret) { - dev_err(&pdev->dev, "Can't enable rx clock\n"); + dev_err(dev, "Can't enable rx clock\n"); goto err_tx_disable; } ret = clk_set_rate(gmac->rx_clk, GMAC_INTF_RATE_125M); if (ret) { - dev_err(&pdev->dev, "Can't set rx clock\n"); + dev_err(dev, "Can't set rx clock\n"); goto err_txrx_disable; } /* Set interface mode */ ret = s32_gmac_write_phy_intf_select(gmac); if (ret) { - dev_err(&pdev->dev, "Can't set PHY interface mode\n"); + dev_err(dev, "Can't set PHY interface mode\n"); goto err_txrx_disable; } @@ -92,7 +92,7 @@ static int s32_gmac_init(struct platform_device *pdev, void *priv) return ret; } -static void s32_gmac_exit(struct platform_device *pdev, void *priv) +static void s32_gmac_exit(struct device *dev, void *priv) { struct s32_priv_data *gmac = priv; diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c index 49d651948e2b..a2b52d2c4eb6 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c @@ -551,7 +551,7 @@ static struct phylink_pcs *socfpga_dwmac_select_pcs(struct stmmac_priv *priv, return priv->hw->phylink_pcs; } -static int socfpga_dwmac_init(struct platform_device *pdev, void *bsp_priv) +static int socfpga_dwmac_init(struct device *dev, void *bsp_priv) { struct socfpga_dwmac *dwmac = bsp_priv; diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-sti.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-sti.c index b0509ab6b31c..f50547b67fbc 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-sti.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-sti.c @@ -229,14 +229,14 @@ static int sti_dwmac_parse_data(struct sti_dwmac *dwmac, return 0; } -static int sti_dwmac_init(struct platform_device *pdev, void *bsp_priv) +static int sti_dwmac_init(struct device *dev, void *bsp_priv) { struct sti_dwmac *dwmac = bsp_priv; return clk_prepare_enable(dwmac->clk); } -static void sti_dwmac_exit(struct platform_device *pdev, void *bsp_priv) +static void sti_dwmac_exit(struct device *dev, void *bsp_priv) { struct sti_dwmac *dwmac = bsp_priv; diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c index 7434d4bbb526..8aa496ac85cc 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c @@ -571,16 +571,16 @@ static const struct stmmac_dma_ops sun8i_dwmac_dma_ops = { static int sun8i_dwmac_power_internal_phy(struct stmmac_priv *priv); -static int sun8i_dwmac_init(struct platform_device *pdev, void *priv) +static int sun8i_dwmac_init(struct device *dev, void *priv) { - struct net_device *ndev = platform_get_drvdata(pdev); + struct net_device *ndev = dev_get_drvdata(dev); struct sunxi_priv_data *gmac = priv; int ret; if (gmac->regulator) { ret = regulator_enable(gmac->regulator); if (ret) { - dev_err(&pdev->dev, "Fail to enable regulator\n"); + dev_err(dev, "Fail to enable regulator\n"); return ret; } } @@ -1005,7 +1005,7 @@ static void sun8i_dwmac_unset_syscon(struct sunxi_priv_data *gmac) (H3_EPHY_SHUTDOWN | H3_EPHY_SELECT)); } -static void sun8i_dwmac_exit(struct platform_device *pdev, void *priv) +static void sun8i_dwmac_exit(struct device *dev, void *priv) { struct sunxi_priv_data *gmac = priv; @@ -1265,7 +1265,7 @@ static void sun8i_dwmac_shutdown(struct platform_device *pdev) struct stmmac_priv *priv = netdev_priv(ndev); struct sunxi_priv_data *gmac = priv->plat->bsp_priv; - sun8i_dwmac_exit(pdev, gmac); + sun8i_dwmac_exit(&pdev->dev, gmac); } static const struct of_device_id sun8i_dwmac_match[] = { diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-sunxi.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-sunxi.c index 7f560d78209d..52593ba3a3a3 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-sunxi.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-sunxi.c @@ -27,7 +27,7 @@ struct sunxi_priv_data { #define SUN7I_GMAC_GMII_RGMII_RATE 125000000 #define SUN7I_GMAC_MII_RATE 25000000 -static int sun7i_gmac_init(struct platform_device *pdev, void *priv) +static int sun7i_gmac_init(struct device *dev, void *priv) { struct sunxi_priv_data *gmac = priv; int ret = 0; @@ -58,7 +58,7 @@ static int sun7i_gmac_init(struct platform_device *pdev, void *priv) return ret; } -static void sun7i_gmac_exit(struct platform_device *pdev, void *priv) +static void sun7i_gmac_exit(struct device *dev, void *priv) { struct sunxi_priv_data *gmac = priv; diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-thead.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-thead.c index a3378046b061..e291028ba56e 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-thead.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-thead.c @@ -186,7 +186,7 @@ static int thead_dwmac_enable_clk(struct plat_stmmacenet_data *plat) return 0; } -static int thead_dwmac_init(struct platform_device *pdev, void *priv) +static int thead_dwmac_init(struct device *dev, void *priv) { struct thead_dwmac *dwmac = priv; unsigned int reg; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c index 1fefa6c55db1..feccb8a3e7e8 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c @@ -747,40 +747,40 @@ EXPORT_SYMBOL_GPL(stmmac_get_platform_resources); /** * stmmac_pltfr_init - * @pdev: pointer to the platform device + * @dev: pointer to the device structure * @plat: driver data platform structure * Description: Call the platform's init callback (if any) and propagate * the return value. */ -static int stmmac_pltfr_init(struct platform_device *pdev, +static int stmmac_pltfr_init(struct device *dev, struct plat_stmmacenet_data *plat) { int ret = 0; if (plat->init) - ret = plat->init(pdev, plat->bsp_priv); + ret = plat->init(dev, plat->bsp_priv); return ret; } /** * stmmac_pltfr_exit - * @pdev: pointer to the platform device + * @dev: pointer to the device structure * @plat: driver data platform structure * Description: Call the platform's exit callback (if any). */ -static void stmmac_pltfr_exit(struct platform_device *pdev, +static void stmmac_pltfr_exit(struct device *dev, struct plat_stmmacenet_data *plat) { if (plat->exit) - plat->exit(pdev, plat->bsp_priv); + plat->exit(dev, plat->bsp_priv); } static int stmmac_plat_suspend(struct device *dev, void *bsp_priv) { struct stmmac_priv *priv = netdev_priv(dev_get_drvdata(dev)); - stmmac_pltfr_exit(to_platform_device(dev), priv->plat); + stmmac_pltfr_exit(dev, priv->plat); return 0; } @@ -789,7 +789,7 @@ static int stmmac_plat_resume(struct device *dev, void *bsp_priv) { struct stmmac_priv *priv = netdev_priv(dev_get_drvdata(dev)); - return stmmac_pltfr_init(to_platform_device(dev), priv->plat); + return stmmac_pltfr_init(dev, priv->plat); } /** @@ -804,6 +804,7 @@ int stmmac_pltfr_probe(struct platform_device *pdev, struct plat_stmmacenet_data *plat, struct stmmac_resources *res) { + struct device *dev = &pdev->dev; int ret; if (!plat->suspend && plat->exit) @@ -811,13 +812,13 @@ int stmmac_pltfr_probe(struct platform_device *pdev, if (!plat->resume && plat->init) plat->resume = stmmac_plat_resume; - ret = stmmac_pltfr_init(pdev, plat); + ret = stmmac_pltfr_init(dev, plat); if (ret) return ret; - ret = stmmac_dvr_probe(&pdev->dev, plat, res); + ret = stmmac_dvr_probe(dev, plat, res); if (ret) { - stmmac_pltfr_exit(pdev, plat); + stmmac_pltfr_exit(dev, plat); return ret; } @@ -866,9 +867,10 @@ void stmmac_pltfr_remove(struct platform_device *pdev) struct net_device *ndev = platform_get_drvdata(pdev); struct stmmac_priv *priv = netdev_priv(ndev); struct plat_stmmacenet_data *plat = priv->plat; + struct device *dev = &pdev->dev; - stmmac_dvr_remove(&pdev->dev); - stmmac_pltfr_exit(pdev, plat); + stmmac_dvr_remove(dev); + stmmac_pltfr_exit(dev, plat); } EXPORT_SYMBOL_GPL(stmmac_pltfr_remove); diff --git a/include/linux/stmmac.h b/include/linux/stmmac.h index 4f70a6551e68..673b068fdadf 100644 --- a/include/linux/stmmac.h +++ b/include/linux/stmmac.h @@ -264,8 +264,8 @@ struct plat_stmmacenet_data { unsigned int mode, phy_interface_t interface); void (*ptp_clk_freq_config)(struct stmmac_priv *priv); - int (*init)(struct platform_device *pdev, void *priv); - void (*exit)(struct platform_device *pdev, void *priv); + int (*init)(struct device *dev, void *priv); + void (*exit)(struct device *dev, void *priv); int (*suspend)(struct device *dev, void *priv); int (*resume)(struct device *dev, void *priv); int (*mac_setup)(void *priv, struct mac_device_info *mac); From 32da89a840e852b1c33df2b8e5b74ba4da468b62 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Wed, 19 Nov 2025 10:03:55 +0000 Subject: [PATCH 0953/1381] net: stmmac: move probe/remove calling of init/exit Move the probe/remove time calling of the init()/exit() methods in the platform data to the main driver probe/remove functions. This allows them to be used by non-platform_device based drivers. Signed-off-by: Russell King (Oracle) Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/E1vLf2Z-0000000FMNH-0xPV@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../net/ethernet/stmicro/stmmac/stmmac_main.c | 47 ++++++++++++++----- .../ethernet/stmicro/stmmac/stmmac_platform.c | 23 +-------- 2 files changed, 36 insertions(+), 34 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index 15b0c08ebd87..e9c3c18b9fc1 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -7593,19 +7593,9 @@ struct plat_stmmacenet_data *stmmac_plat_dat_alloc(struct device *dev) } EXPORT_SYMBOL_GPL(stmmac_plat_dat_alloc); -/** - * stmmac_dvr_probe - * @device: device pointer - * @plat_dat: platform data pointer - * @res: stmmac resource pointer - * Description: this is the main probe function used to - * call the alloc_etherdev, allocate the priv structure. - * Return: - * returns 0 on success, otherwise errno. - */ -int stmmac_dvr_probe(struct device *device, - struct plat_stmmacenet_data *plat_dat, - struct stmmac_resources *res) +static int __stmmac_dvr_probe(struct device *device, + struct plat_stmmacenet_data *plat_dat, + struct stmmac_resources *res) { struct net_device *ndev = NULL; struct stmmac_priv *priv; @@ -7906,6 +7896,34 @@ int stmmac_dvr_probe(struct device *device, return ret; } + +/** + * stmmac_dvr_probe + * @dev: device pointer + * @plat_dat: platform data pointer + * @res: stmmac resource pointer + * Description: this is the main probe function used to + * call the alloc_etherdev, allocate the priv structure. + * Return: + * returns 0 on success, otherwise errno. + */ +int stmmac_dvr_probe(struct device *dev, struct plat_stmmacenet_data *plat_dat, + struct stmmac_resources *res) +{ + int ret; + + if (plat_dat->init) { + ret = plat_dat->init(dev, plat_dat->bsp_priv); + if (ret) + return ret; + } + + ret = __stmmac_dvr_probe(dev, plat_dat, res); + if (ret && plat_dat->exit) + plat_dat->exit(dev, plat_dat->bsp_priv); + + return ret; +} EXPORT_SYMBOL_GPL(stmmac_dvr_probe); /** @@ -7944,6 +7962,9 @@ void stmmac_dvr_remove(struct device *dev) pm_runtime_disable(dev); pm_runtime_put_noidle(dev); + + if (priv->plat->exit) + priv->plat->exit(dev, priv->plat->bsp_priv); } EXPORT_SYMBOL_GPL(stmmac_dvr_remove); diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c index feccb8a3e7e8..9015b7f80d1b 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c @@ -804,25 +804,12 @@ int stmmac_pltfr_probe(struct platform_device *pdev, struct plat_stmmacenet_data *plat, struct stmmac_resources *res) { - struct device *dev = &pdev->dev; - int ret; - if (!plat->suspend && plat->exit) plat->suspend = stmmac_plat_suspend; if (!plat->resume && plat->init) plat->resume = stmmac_plat_resume; - ret = stmmac_pltfr_init(dev, plat); - if (ret) - return ret; - - ret = stmmac_dvr_probe(dev, plat, res); - if (ret) { - stmmac_pltfr_exit(dev, plat); - return ret; - } - - return ret; + return stmmac_dvr_probe(&pdev->dev, plat, res); } EXPORT_SYMBOL_GPL(stmmac_pltfr_probe); @@ -864,13 +851,7 @@ EXPORT_SYMBOL_GPL(devm_stmmac_pltfr_probe); */ void stmmac_pltfr_remove(struct platform_device *pdev) { - struct net_device *ndev = platform_get_drvdata(pdev); - struct stmmac_priv *priv = netdev_priv(ndev); - struct plat_stmmacenet_data *plat = priv->plat; - struct device *dev = &pdev->dev; - - stmmac_dvr_remove(dev); - stmmac_pltfr_exit(dev, plat); + stmmac_dvr_remove(&pdev->dev); } EXPORT_SYMBOL_GPL(stmmac_pltfr_remove); From 1a62894e040dad73782e0c1a4b0e8a0d5480d29b Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Wed, 19 Nov 2025 10:04:00 +0000 Subject: [PATCH 0954/1381] net: stmmac: rk: convert to init()/exit() methods Convert rk to use the init() and exit() methods for powering up and down the device. This allows us to use the pltfr versions of probe() and remove(). Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vLf2e-0000000FMNN-1Xnh@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../net/ethernet/stmicro/stmmac/dwmac-rk.c | 47 ++++++++----------- 1 file changed, 19 insertions(+), 28 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c index 6e75da577af5..6bfe82d7001a 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c @@ -1763,6 +1763,22 @@ static int rk_gmac_resume(struct device *dev, void *bsp_priv_) return 0; } +static int rk_gmac_init(struct device *dev, void *bsp_priv) +{ + return rk_gmac_powerup(bsp_priv); +} + +static void rk_gmac_exit(struct device *dev, void *bsp_priv_) +{ + struct stmmac_priv *priv = netdev_priv(dev_get_drvdata(dev)); + struct rk_priv_data *bsp_priv = bsp_priv_; + + rk_gmac_powerdown(bsp_priv); + + if (priv->plat->phy_node && bsp_priv->integrated_phy) + clk_put(bsp_priv->clk_phy); +} + static int rk_gmac_probe(struct platform_device *pdev) { struct plat_stmmacenet_data *plat_dat; @@ -1795,6 +1811,8 @@ static int rk_gmac_probe(struct platform_device *pdev) plat_dat->get_interfaces = rk_get_interfaces; plat_dat->set_clk_tx_rate = rk_set_clk_tx_rate; + plat_dat->init = rk_gmac_init; + plat_dat->exit = rk_gmac_exit; plat_dat->suspend = rk_gmac_suspend; plat_dat->resume = rk_gmac_resume; @@ -1806,33 +1824,7 @@ static int rk_gmac_probe(struct platform_device *pdev) if (ret) return ret; - ret = rk_gmac_powerup(plat_dat->bsp_priv); - if (ret) - return ret; - - ret = stmmac_dvr_probe(&pdev->dev, plat_dat, &stmmac_res); - if (ret) - goto err_gmac_powerdown; - - return 0; - -err_gmac_powerdown: - rk_gmac_powerdown(plat_dat->bsp_priv); - - return ret; -} - -static void rk_gmac_remove(struct platform_device *pdev) -{ - struct stmmac_priv *priv = netdev_priv(platform_get_drvdata(pdev)); - struct rk_priv_data *bsp_priv = priv->plat->bsp_priv; - - stmmac_dvr_remove(&pdev->dev); - - rk_gmac_powerdown(bsp_priv); - - if (priv->plat->phy_node && bsp_priv->integrated_phy) - clk_put(bsp_priv->clk_phy); + return devm_stmmac_pltfr_probe(pdev, plat_dat, &stmmac_res); } static const struct of_device_id rk_gmac_dwmac_match[] = { @@ -1858,7 +1850,6 @@ MODULE_DEVICE_TABLE(of, rk_gmac_dwmac_match); static struct platform_driver rk_gmac_dwmac_driver = { .probe = rk_gmac_probe, - .remove = rk_gmac_remove, .driver = { .name = "rk_gmac-dwmac", .pm = &stmmac_simple_pm_ops, From f15bcd071913dfa754cabb85e9fff4eded4e35b2 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Wed, 19 Nov 2025 11:29:16 +0000 Subject: [PATCH 0955/1381] net: stmmac: rk: use phylink's interface mode for set_clk_tx_rate() rk_set_clk_tx_rate() is passed the interface mode from phylink which will be the same as bsp_priv->phy_iface. Use the passed-in interface mode rather than bsp_priv->phy_iface. Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vLgNA-0000000FMjN-0DSS@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c index 6bfe82d7001a..0a95f54e725e 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c @@ -1735,8 +1735,7 @@ static int rk_set_clk_tx_rate(void *bsp_priv_, struct clk *clk_tx_i, struct rk_priv_data *bsp_priv = bsp_priv_; if (bsp_priv->ops->set_speed) - return bsp_priv->ops->set_speed(bsp_priv, bsp_priv->phy_iface, - speed); + return bsp_priv->ops->set_speed(bsp_priv, interface, speed); return -EINVAL; } From f7ac9a0bbe3f15d0164ac09688e336933cb54a03 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Wed, 19 Nov 2025 10:23:14 +0000 Subject: [PATCH 0956/1381] net: stmmac: dwc-qos-eth: simplify switch() in dwc_eth_dwmac_config_dt() Simplify the switch() statement in dwc_eth_dwmac_config_dt(). Although this is not speed-critical, simplifying it can make it more readable. This also drastically improves the code emitted by the compiler. On aarch64, with the original code, the compiler loads registers with every possible value, and then has a tree of test-and-branch statements to work out which register to store. With the simplified code, the compiler can load a register with '4' and shift it appropriately. This shrinks the text size on aarch64 from 4289 bytes to 4153 bytes, a reduction of 3%. Reviewed-by: Maxime Chevallier Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vLfLG-0000000FMai-3fKz@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../stmicro/stmmac/dwmac-dwc-qos-eth.c | 26 +++---------------- 1 file changed, 3 insertions(+), 23 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-dwc-qos-eth.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-dwc-qos-eth.c index c7cd6497d42d..e6d5893c5905 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-dwc-qos-eth.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-dwc-qos-eth.c @@ -84,29 +84,9 @@ static int dwc_eth_dwmac_config_dt(struct platform_device *pdev, device_property_read_u32(dev, "snps,burst-map", &burst_map); /* converts burst-map bitmask to burst array */ - for (bit_index = 0; bit_index < 7; bit_index++) { - if (burst_map & (1 << bit_index)) { - switch (bit_index) { - case 0: - plat_dat->axi->axi_blen[a_index] = 4; break; - case 1: - plat_dat->axi->axi_blen[a_index] = 8; break; - case 2: - plat_dat->axi->axi_blen[a_index] = 16; break; - case 3: - plat_dat->axi->axi_blen[a_index] = 32; break; - case 4: - plat_dat->axi->axi_blen[a_index] = 64; break; - case 5: - plat_dat->axi->axi_blen[a_index] = 128; break; - case 6: - plat_dat->axi->axi_blen[a_index] = 256; break; - default: - break; - } - a_index++; - } - } + for (bit_index = 0; bit_index < 7; bit_index++) + if (burst_map & (1 << bit_index)) + plat_dat->axi->axi_blen[a_index++] = 4 << bit_index; /* dwc-qos needs GMAC4, AAL, TSO and PMT */ plat_dat->core_type = DWMAC_CORE_GMAC4; From 8c696659f47aa4f4e5f35808b416664378930dd9 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Wed, 19 Nov 2025 10:23:19 +0000 Subject: [PATCH 0957/1381] net: stmmac: move common DMA AXI register bits to common.h Move the common DMA AXI register bits to common.h so they can be shared and we can provide a common function to convert the axi->dma_blen[] array to the format needed for this register. Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vLfLL-0000000FMap-49gf@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/common.h | 10 ++++++++++ drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.h | 9 +-------- drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h | 11 ++--------- drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h | 16 ++++++++-------- 4 files changed, 21 insertions(+), 25 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/common.h b/drivers/net/ethernet/stmicro/stmmac/common.h index 7395bbb94aea..3c6e7fe7b999 100644 --- a/drivers/net/ethernet/stmicro/stmmac/common.h +++ b/drivers/net/ethernet/stmicro/stmmac/common.h @@ -548,6 +548,16 @@ struct dma_features { #define LPI_CTRL_STATUS_TLPIEX BIT(1) /* Transmit LPI Exit */ #define LPI_CTRL_STATUS_TLPIEN BIT(0) /* Transmit LPI Entry */ +/* Common definitions for AXI Master Bus Mode */ +#define DMA_AXI_AAL BIT(12) +#define DMA_AXI_BLEN256 BIT(7) +#define DMA_AXI_BLEN128 BIT(6) +#define DMA_AXI_BLEN64 BIT(5) +#define DMA_AXI_BLEN32 BIT(4) +#define DMA_AXI_BLEN16 BIT(3) +#define DMA_AXI_BLEN8 BIT(2) +#define DMA_AXI_BLEN4 BIT(1) + #define STMMAC_CHAIN_MODE 0x1 #define STMMAC_RING_MODE 0x2 diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.h b/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.h index 4f980dcd3958..dfcb7ce79e76 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.h +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.h @@ -69,15 +69,8 @@ #define DMA_SYS_BUS_MB BIT(14) #define DMA_AXI_1KBBE BIT(13) -#define DMA_SYS_BUS_AAL BIT(12) +#define DMA_SYS_BUS_AAL DMA_AXI_AAL #define DMA_SYS_BUS_EAME BIT(11) -#define DMA_AXI_BLEN256 BIT(7) -#define DMA_AXI_BLEN128 BIT(6) -#define DMA_AXI_BLEN64 BIT(5) -#define DMA_AXI_BLEN32 BIT(4) -#define DMA_AXI_BLEN16 BIT(3) -#define DMA_AXI_BLEN8 BIT(2) -#define DMA_AXI_BLEN4 BIT(1) #define DMA_SYS_BUS_FB BIT(0) #define DMA_BURST_LEN_DEFAULT (DMA_AXI_BLEN256 | DMA_AXI_BLEN128 | \ diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h b/drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h index 5d9c18f5bbf5..967a735e9a0b 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h @@ -68,20 +68,13 @@ static inline u32 dma_chan_base_addr(u32 base, u32 chan) #define DMA_AXI_OSR_MAX 0xf #define DMA_AXI_MAX_OSR_LIMIT ((DMA_AXI_OSR_MAX << DMA_AXI_WR_OSR_LMT_SHIFT) | \ (DMA_AXI_OSR_MAX << DMA_AXI_RD_OSR_LMT_SHIFT)) -#define DMA_AXI_1KBBE BIT(13) -#define DMA_AXI_AAL BIT(12) -#define DMA_AXI_BLEN256 BIT(7) -#define DMA_AXI_BLEN128 BIT(6) -#define DMA_AXI_BLEN64 BIT(5) -#define DMA_AXI_BLEN32 BIT(4) -#define DMA_AXI_BLEN16 BIT(3) -#define DMA_AXI_BLEN8 BIT(2) -#define DMA_AXI_BLEN4 BIT(1) #define DMA_BURST_LEN_DEFAULT (DMA_AXI_BLEN256 | DMA_AXI_BLEN128 | \ DMA_AXI_BLEN64 | DMA_AXI_BLEN32 | \ DMA_AXI_BLEN16 | DMA_AXI_BLEN8 | \ DMA_AXI_BLEN4) +#define DMA_AXI_1KBBE BIT(13) + #define DMA_AXI_UNDEF BIT(0) #define DMA_AXI_BURST_LEN_MASK 0x000000FE diff --git a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h index e48cfa05000c..16c6d03fc929 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h +++ b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h @@ -338,16 +338,16 @@ #define XGMAC_RD_OSR_LMT_SHIFT 16 #define XGMAC_EN_LPI BIT(15) #define XGMAC_LPI_XIT_PKT BIT(14) -#define XGMAC_AAL BIT(12) +#define XGMAC_AAL DMA_AXI_AAL #define XGMAC_EAME BIT(11) #define XGMAC_BLEN GENMASK(7, 1) -#define XGMAC_BLEN256 BIT(7) -#define XGMAC_BLEN128 BIT(6) -#define XGMAC_BLEN64 BIT(5) -#define XGMAC_BLEN32 BIT(4) -#define XGMAC_BLEN16 BIT(3) -#define XGMAC_BLEN8 BIT(2) -#define XGMAC_BLEN4 BIT(1) +#define XGMAC_BLEN256 DMA_AXI_BLEN256 +#define XGMAC_BLEN128 DMA_AXI_BLEN128 +#define XGMAC_BLEN64 DMA_AXI_BLEN64 +#define XGMAC_BLEN32 DMA_AXI_BLEN32 +#define XGMAC_BLEN16 DMA_AXI_BLEN16 +#define XGMAC_BLEN8 DMA_AXI_BLEN8 +#define XGMAC_BLEN4 DMA_AXI_BLEN4 #define XGMAC_UNDEF BIT(0) #define XGMAC_TX_EDMA_CTRL 0x00003040 #define XGMAC_TDPS GENMASK(29, 0) From 2704af20c8e5e7f5f1cfb56003d246a7936458f6 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Wed, 19 Nov 2025 10:23:25 +0000 Subject: [PATCH 0958/1381] net: stmmac: provide common stmmac_axi_blen_to_mask() Provide a common stmmac_axi_blen_to_mask() function to translate the burst length array to the value for the AXI bus mode register, and use it for dwmac, dwmac4 and dwxgmac2. Remove the now unnecessary XGMAC_BLEN* definitions. Note that stmmac_axi_blen_to_dma_mask() is coded to be more efficient than the original three implementations, and verifies the contents of the burst length array. Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vLfLR-0000000FMav-0VL6@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/common.h | 3 ++ .../ethernet/stmicro/stmmac/dwmac1000_dma.c | 31 ++-------------- .../net/ethernet/stmicro/stmmac/dwmac4_dma.c | 31 ++-------------- .../net/ethernet/stmicro/stmmac/dwmac4_dma.h | 2 - .../net/ethernet/stmicro/stmmac/dwmac_dma.h | 2 - .../net/ethernet/stmicro/stmmac/dwxgmac2.h | 9 +---- .../ethernet/stmicro/stmmac/dwxgmac2_dma.c | 34 ++++------------- .../net/ethernet/stmicro/stmmac/stmmac_main.c | 37 +++++++++++++++++++ 8 files changed, 56 insertions(+), 93 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/common.h b/drivers/net/ethernet/stmicro/stmmac/common.h index 3c6e7fe7b999..49df46be3669 100644 --- a/drivers/net/ethernet/stmicro/stmmac/common.h +++ b/drivers/net/ethernet/stmicro/stmmac/common.h @@ -557,6 +557,9 @@ struct dma_features { #define DMA_AXI_BLEN16 BIT(3) #define DMA_AXI_BLEN8 BIT(2) #define DMA_AXI_BLEN4 BIT(1) +#define DMA_AXI_BLEN_MASK GENMASK(7, 1) + +void stmmac_axi_blen_to_mask(u32 *regval, const u32 *blen, size_t len); #define STMMAC_CHAIN_MODE 0x1 #define STMMAC_RING_MODE 0x2 diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c index 118a22406a2e..b6476a1bfeab 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c @@ -19,7 +19,6 @@ static void dwmac1000_dma_axi(void __iomem *ioaddr, struct stmmac_axi *axi) { u32 value = readl(ioaddr + DMA_AXI_BUS_MODE); - int i; pr_info("dwmac1000: Master AXI performs %s burst length\n", !(value & DMA_AXI_UNDEF) ? "fixed" : "any"); @@ -39,33 +38,11 @@ static void dwmac1000_dma_axi(void __iomem *ioaddr, struct stmmac_axi *axi) /* Depending on the UNDEF bit the Master AXI will perform any burst * length according to the BLEN programmed (by default all BLEN are - * set). + * set). Note that the UNDEF bit is readonly, and is the inverse of + * Bus Mode bit 16. */ - for (i = 0; i < AXI_BLEN; i++) { - switch (axi->axi_blen[i]) { - case 256: - value |= DMA_AXI_BLEN256; - break; - case 128: - value |= DMA_AXI_BLEN128; - break; - case 64: - value |= DMA_AXI_BLEN64; - break; - case 32: - value |= DMA_AXI_BLEN32; - break; - case 16: - value |= DMA_AXI_BLEN16; - break; - case 8: - value |= DMA_AXI_BLEN8; - break; - case 4: - value |= DMA_AXI_BLEN4; - break; - } - } + stmmac_axi_blen_to_mask(&value, axi->axi_blen, + ARRAY_SIZE(axi->axi_blen)); writel(value, ioaddr + DMA_AXI_BUS_MODE); } diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.c b/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.c index d87a8b595e6a..90d03c7b29f4 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.c @@ -18,7 +18,6 @@ static void dwmac4_dma_axi(void __iomem *ioaddr, struct stmmac_axi *axi) { u32 value = readl(ioaddr + DMA_SYS_BUS_MODE); - int i; pr_info("dwmac4: Master AXI performs %s burst length\n", (value & DMA_SYS_BUS_FB) ? "fixed" : "any"); @@ -38,33 +37,11 @@ static void dwmac4_dma_axi(void __iomem *ioaddr, struct stmmac_axi *axi) /* Depending on the UNDEF bit the Master AXI will perform any burst * length according to the BLEN programmed (by default all BLEN are - * set). + * set). Note that the UNDEF bit is readonly, and is the inverse of + * Bus Mode bit 16. */ - for (i = 0; i < AXI_BLEN; i++) { - switch (axi->axi_blen[i]) { - case 256: - value |= DMA_AXI_BLEN256; - break; - case 128: - value |= DMA_AXI_BLEN128; - break; - case 64: - value |= DMA_AXI_BLEN64; - break; - case 32: - value |= DMA_AXI_BLEN32; - break; - case 16: - value |= DMA_AXI_BLEN16; - break; - case 8: - value |= DMA_AXI_BLEN8; - break; - case 4: - value |= DMA_AXI_BLEN4; - break; - } - } + stmmac_axi_blen_to_mask(&value, axi->axi_blen, + ARRAY_SIZE(axi->axi_blen)); writel(value, ioaddr + DMA_SYS_BUS_MODE); } diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.h b/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.h index dfcb7ce79e76..f27126f05551 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.h +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.h @@ -78,8 +78,6 @@ DMA_AXI_BLEN16 | DMA_AXI_BLEN8 | \ DMA_AXI_BLEN4) -#define DMA_AXI_BURST_LEN_MASK 0x000000FE - /* DMA TBS Control */ #define DMA_TBS_FTOS GENMASK(31, 8) #define DMA_TBS_FTOV BIT(0) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h b/drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h index 967a735e9a0b..d1c149f7a3dd 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h @@ -77,8 +77,6 @@ static inline u32 dma_chan_base_addr(u32 base, u32 chan) #define DMA_AXI_UNDEF BIT(0) -#define DMA_AXI_BURST_LEN_MASK 0x000000FE - #define DMA_CUR_TX_BUF_ADDR 0x00001050 /* Current Host Tx Buffer */ #define DMA_CUR_RX_BUF_ADDR 0x00001054 /* Current Host Rx Buffer */ #define DMA_HW_FEATURE 0x00001058 /* HW Feature Register */ diff --git a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h index 16c6d03fc929..fecda3034d36 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h +++ b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h @@ -340,14 +340,7 @@ #define XGMAC_LPI_XIT_PKT BIT(14) #define XGMAC_AAL DMA_AXI_AAL #define XGMAC_EAME BIT(11) -#define XGMAC_BLEN GENMASK(7, 1) -#define XGMAC_BLEN256 DMA_AXI_BLEN256 -#define XGMAC_BLEN128 DMA_AXI_BLEN128 -#define XGMAC_BLEN64 DMA_AXI_BLEN64 -#define XGMAC_BLEN32 DMA_AXI_BLEN32 -#define XGMAC_BLEN16 DMA_AXI_BLEN16 -#define XGMAC_BLEN8 DMA_AXI_BLEN8 -#define XGMAC_BLEN4 DMA_AXI_BLEN4 +/* XGMAC_BLEN* are now defined as DMA_AXI_BLEN* in common.h */ #define XGMAC_UNDEF BIT(0) #define XGMAC_TX_EDMA_CTRL 0x00003040 #define XGMAC_TDPS GENMASK(29, 0) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c index 4d6bb995d8d8..8a2cb6ca9588 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c @@ -84,7 +84,6 @@ static void dwxgmac2_dma_init_tx_chan(struct stmmac_priv *priv, static void dwxgmac2_dma_axi(void __iomem *ioaddr, struct stmmac_axi *axi) { u32 value = readl(ioaddr + XGMAC_DMA_SYSBUS_MODE); - int i; if (axi->axi_lpi_en) value |= XGMAC_EN_LPI; @@ -102,32 +101,13 @@ static void dwxgmac2_dma_axi(void __iomem *ioaddr, struct stmmac_axi *axi) if (!axi->axi_fb) value |= XGMAC_UNDEF; - value &= ~XGMAC_BLEN; - for (i = 0; i < AXI_BLEN; i++) { - switch (axi->axi_blen[i]) { - case 256: - value |= XGMAC_BLEN256; - break; - case 128: - value |= XGMAC_BLEN128; - break; - case 64: - value |= XGMAC_BLEN64; - break; - case 32: - value |= XGMAC_BLEN32; - break; - case 16: - value |= XGMAC_BLEN16; - break; - case 8: - value |= XGMAC_BLEN8; - break; - case 4: - value |= XGMAC_BLEN4; - break; - } - } + /* Depending on the UNDEF bit the Master AXI will perform any burst + * length according to the BLEN programmed (by default all BLEN are + * set). Note that the UNDEF bit is readonly, and is the inverse of + * Bus Mode bit 16. + */ + stmmac_axi_blen_to_mask(&value, axi->axi_blen, + ARRAY_SIZE(axi->axi_blen)); writel(value, ioaddr + XGMAC_DMA_SYSBUS_MODE); writel(XGMAC_TDPS, ioaddr + XGMAC_TX_EDMA_CTRL); diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index e9c3c18b9fc1..0b1e571f70f0 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -189,6 +189,43 @@ int stmmac_set_clk_tx_rate(void *bsp_priv, struct clk *clk_tx_i, } EXPORT_SYMBOL_GPL(stmmac_set_clk_tx_rate); +/** + * stmmac_axi_blen_to_mask() - convert a burst length array to reg value + * @regval: pointer to a u32 for the resulting register value + * @blen: pointer to an array of u32 containing the burst length values in bytes + * @len: the number of entries in the @blen array + */ +void stmmac_axi_blen_to_mask(u32 *regval, const u32 *blen, size_t len) +{ + size_t i; + u32 val; + + for (val = i = 0; i < len; i++) { + u32 burst = blen[i]; + + /* Burst values of zero must be skipped. */ + if (!burst) + continue; + + /* The valid range for the burst length is 4 to 256 inclusive, + * and it must be a power of two. + */ + if (burst < 4 || burst > 256 || !is_power_of_2(burst)) { + pr_err("stmmac: invalid burst length %u at index %zu\n", + burst, i); + continue; + } + + /* Since burst is a power of two, and the register field starts + * with burst = 4, shift right by two bits so bit 0 of the field + * corresponds with the minimum value. + */ + val |= burst >> 2; + } + + u32p_replace_bits(regval, val, DMA_AXI_BLEN_MASK); +} + /** * stmmac_verify_args - verify the driver parameters. * Description: it checks the driver parameters and set a default in case of From 6ff3310ca28298e363c78143b6a2f20312421f4e Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Wed, 19 Nov 2025 10:23:30 +0000 Subject: [PATCH 0959/1381] net: stmmac: move stmmac_axi_blen_to_mask() to stmmac_main.c Move the call to stmmac_axi_blen_to_mask() out of the individual MAC version drivers into the main code in stmmac_init_dma_engine(), passing the resulting value through a new member, axi_blen_regval, in the struct stmmac_axi structure. There is now no need for stmmac_axi_blen_to_dma_mask() to use u32p_replace_bits(), so use FIELD_PREP() instead. Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vLfLW-0000000FMb1-0zKV@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c | 3 +-- drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.c | 3 +-- drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c | 3 +-- drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 8 ++++++-- include/linux/stmmac.h | 1 + 5 files changed, 10 insertions(+), 8 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c index b6476a1bfeab..6d9b8fac3c6d 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c @@ -41,8 +41,7 @@ static void dwmac1000_dma_axi(void __iomem *ioaddr, struct stmmac_axi *axi) * set). Note that the UNDEF bit is readonly, and is the inverse of * Bus Mode bit 16. */ - stmmac_axi_blen_to_mask(&value, axi->axi_blen, - ARRAY_SIZE(axi->axi_blen)); + value = (value & ~DMA_AXI_BLEN_MASK) | axi->axi_blen_regval; writel(value, ioaddr + DMA_AXI_BUS_MODE); } diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.c b/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.c index 90d03c7b29f4..7b513324cfb0 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_dma.c @@ -40,8 +40,7 @@ static void dwmac4_dma_axi(void __iomem *ioaddr, struct stmmac_axi *axi) * set). Note that the UNDEF bit is readonly, and is the inverse of * Bus Mode bit 16. */ - stmmac_axi_blen_to_mask(&value, axi->axi_blen, - ARRAY_SIZE(axi->axi_blen)); + value = (value & ~DMA_AXI_BLEN_MASK) | axi->axi_blen_regval; writel(value, ioaddr + DMA_SYS_BUS_MODE); } diff --git a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c index 8a2cb6ca9588..cc1bdc0975d5 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c @@ -106,8 +106,7 @@ static void dwxgmac2_dma_axi(void __iomem *ioaddr, struct stmmac_axi *axi) * set). Note that the UNDEF bit is readonly, and is the inverse of * Bus Mode bit 16. */ - stmmac_axi_blen_to_mask(&value, axi->axi_blen, - ARRAY_SIZE(axi->axi_blen)); + value = (value & ~DMA_AXI_BLEN_MASK) | axi->axi_blen_regval; writel(value, ioaddr + XGMAC_DMA_SYSBUS_MODE); writel(XGMAC_TDPS, ioaddr + XGMAC_TX_EDMA_CTRL); diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index 0b1e571f70f0..aac82ddfb8c0 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -223,7 +223,7 @@ void stmmac_axi_blen_to_mask(u32 *regval, const u32 *blen, size_t len) val |= burst >> 2; } - u32p_replace_bits(regval, val, DMA_AXI_BLEN_MASK); + *regval = FIELD_PREP(DMA_AXI_BLEN_MASK, val); } /** @@ -3212,8 +3212,12 @@ static int stmmac_init_dma_engine(struct stmmac_priv *priv) /* DMA Configuration */ stmmac_dma_init(priv, priv->ioaddr, priv->plat->dma_cfg); - if (priv->plat->axi) + if (priv->plat->axi) { + /* Encode the AXI burst length to a register value */ + stmmac_axi_blen_to_mask(&priv->plat->axi->axi_blen_regval, + priv->plat->axi->axi_blen, AXI_BLEN); stmmac_axi(priv, priv->ioaddr, priv->plat->axi); + } /* DMA CSR Channel configuration */ for (chan = 0; chan < dma_csr_ch; chan++) { diff --git a/include/linux/stmmac.h b/include/linux/stmmac.h index 673b068fdadf..d1a41fe0825f 100644 --- a/include/linux/stmmac.h +++ b/include/linux/stmmac.h @@ -113,6 +113,7 @@ struct stmmac_axi { u32 axi_wr_osr_lmt; u32 axi_rd_osr_lmt; bool axi_kbbe; + u32 axi_blen_regval; u32 axi_blen[AXI_BLEN]; bool axi_fb; bool axi_mb; From e676cc8561c86799f9e12f8cea550c226130821a Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Wed, 19 Nov 2025 10:23:35 +0000 Subject: [PATCH 0960/1381] net: stmmac: move stmmac_axi_blen_to_mask() to axi_blen init sites Move stmmac_axi_blen_to_mask() to the axi->axi_blen array init sites to prepare for the removal of axi_blen. For sites which initialise axi->axi_blen with constant data, initialise axi->axi_blen_regval using the DMA_AXI_BLENx constants. Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vLfLb-0000000FMb7-1SgG@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/dwmac-dwc-qos-eth.c | 3 +++ drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c | 2 ++ drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 7 ++----- drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c | 2 ++ drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c | 1 + 5 files changed, 10 insertions(+), 5 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-dwc-qos-eth.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-dwc-qos-eth.c index e6d5893c5905..bd06f26a27b4 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-dwc-qos-eth.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-dwc-qos-eth.c @@ -88,6 +88,9 @@ static int dwc_eth_dwmac_config_dt(struct platform_device *pdev, if (burst_map & (1 << bit_index)) plat_dat->axi->axi_blen[a_index++] = 4 << bit_index; + stmmac_axi_blen_to_mask(&plat_dat->axi->axi_blen_regval, + plat_dat->axi->axi_blen, a_index); + /* dwc-qos needs GMAC4, AAL, TSO and PMT */ plat_dat->core_type = DWMAC_CORE_GMAC4; plat_dat->dma_cfg->aal = 1; diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c index 8938e7a59925..e94605d3d185 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c @@ -650,6 +650,8 @@ static int intel_mgbe_common_data(struct pci_dev *pdev, plat->axi->axi_xit_frm = 0; plat->axi->axi_wr_osr_lmt = 1; plat->axi->axi_rd_osr_lmt = 1; + plat->axi->axi_blen_regval = DMA_AXI_BLEN4 | DMA_AXI_BLEN8 | + DMA_AXI_BLEN16; plat->axi->axi_blen[0] = 4; plat->axi->axi_blen[1] = 8; plat->axi->axi_blen[2] = 16; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index aac82ddfb8c0..b1aa236d8051 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -225,6 +225,7 @@ void stmmac_axi_blen_to_mask(u32 *regval, const u32 *blen, size_t len) *regval = FIELD_PREP(DMA_AXI_BLEN_MASK, val); } +EXPORT_SYMBOL_GPL(stmmac_axi_blen_to_mask); /** * stmmac_verify_args - verify the driver parameters. @@ -3212,12 +3213,8 @@ static int stmmac_init_dma_engine(struct stmmac_priv *priv) /* DMA Configuration */ stmmac_dma_init(priv, priv->ioaddr, priv->plat->dma_cfg); - if (priv->plat->axi) { - /* Encode the AXI burst length to a register value */ - stmmac_axi_blen_to_mask(&priv->plat->axi->axi_blen_regval, - priv->plat->axi->axi_blen, AXI_BLEN); + if (priv->plat->axi) stmmac_axi(priv, priv->ioaddr, priv->plat->axi); - } /* DMA CSR Channel configuration */ for (chan = 0; chan < dma_csr_ch; chan++) { diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c index 2fd4660838bb..e1036150fae2 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c @@ -92,6 +92,8 @@ static int snps_gmac5_default_data(struct pci_dev *pdev, plat->axi->axi_rd_osr_lmt = 31; plat->axi->axi_fb = false; + plat->axi->axi_blen_regval = DMA_AXI_BLEN4 | DMA_AXI_BLEN8 | + DMA_AXI_BLEN16 | DMA_AXI_BLEN32; plat->axi->axi_blen[0] = 4; plat->axi->axi_blen[1] = 8; plat->axi->axi_blen[2] = 16; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c index 9015b7f80d1b..656d4adedabe 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c @@ -118,6 +118,7 @@ static struct stmmac_axi *stmmac_axi_setup(struct platform_device *pdev) if (of_property_read_u32(np, "snps,rd_osr_lmt", &axi->axi_rd_osr_lmt)) axi->axi_rd_osr_lmt = 1; of_property_read_u32_array(np, "snps,blen", axi->axi_blen, AXI_BLEN); + stmmac_axi_blen_to_mask(&axi->axi_blen_regval, axi->axi_blen, AXI_BLEN); of_node_put(np); return axi; From efd3c8cc52bb9583183ebb83c8c55b23bf97cb2f Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Wed, 19 Nov 2025 10:23:40 +0000 Subject: [PATCH 0961/1381] net: stmmac: remove axi_blen array Remove the axi_blen array from struct stmmac_axi as we set this array, and then immediately convert it ot the register value, never looking at the array again. Thus, the array can be function local rather than part of a run-time allocated long-lived struct. Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vLfLg-0000000FMbD-1vmh@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../net/ethernet/stmicro/stmmac/dwmac-dwc-qos-eth.c | 11 ++--------- drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c | 3 --- drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c | 4 ---- drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c | 5 +++-- include/linux/stmmac.h | 1 - 5 files changed, 5 insertions(+), 19 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-dwc-qos-eth.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-dwc-qos-eth.c index bd06f26a27b4..d043bad4a862 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-dwc-qos-eth.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-dwc-qos-eth.c @@ -38,8 +38,6 @@ static int dwc_eth_dwmac_config_dt(struct platform_device *pdev, { struct device *dev = &pdev->dev; u32 burst_map = 0; - u32 bit_index = 0; - u32 a_index = 0; if (!plat_dat->axi) { plat_dat->axi = devm_kzalloc(&pdev->dev, @@ -83,13 +81,8 @@ static int dwc_eth_dwmac_config_dt(struct platform_device *pdev, } device_property_read_u32(dev, "snps,burst-map", &burst_map); - /* converts burst-map bitmask to burst array */ - for (bit_index = 0; bit_index < 7; bit_index++) - if (burst_map & (1 << bit_index)) - plat_dat->axi->axi_blen[a_index++] = 4 << bit_index; - - stmmac_axi_blen_to_mask(&plat_dat->axi->axi_blen_regval, - plat_dat->axi->axi_blen, a_index); + plat_dat->axi->axi_blen_regval = FIELD_PREP(DMA_AXI_BLEN_MASK, + burst_map); /* dwc-qos needs GMAC4, AAL, TSO and PMT */ plat_dat->core_type = DWMAC_CORE_GMAC4; diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c index e94605d3d185..aad1be1ec4c1 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c @@ -652,9 +652,6 @@ static int intel_mgbe_common_data(struct pci_dev *pdev, plat->axi->axi_rd_osr_lmt = 1; plat->axi->axi_blen_regval = DMA_AXI_BLEN4 | DMA_AXI_BLEN8 | DMA_AXI_BLEN16; - plat->axi->axi_blen[0] = 4; - plat->axi->axi_blen[1] = 8; - plat->axi->axi_blen[2] = 16; plat->ptp_max_adj = plat->clk_ptp_rate; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c index e1036150fae2..afb1c53ca6f8 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c @@ -94,10 +94,6 @@ static int snps_gmac5_default_data(struct pci_dev *pdev, plat->axi->axi_fb = false; plat->axi->axi_blen_regval = DMA_AXI_BLEN4 | DMA_AXI_BLEN8 | DMA_AXI_BLEN16 | DMA_AXI_BLEN32; - plat->axi->axi_blen[0] = 4; - plat->axi->axi_blen[1] = 8; - plat->axi->axi_blen[2] = 16; - plat->axi->axi_blen[3] = 32; return 0; } diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c index 656d4adedabe..8979a50b5507 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c @@ -95,6 +95,7 @@ static struct stmmac_axi *stmmac_axi_setup(struct platform_device *pdev) { struct device_node *np; struct stmmac_axi *axi; + u32 axi_blen[AXI_BLEN]; np = of_parse_phandle(pdev->dev.of_node, "snps,axi-config", 0); if (!np) @@ -117,8 +118,8 @@ static struct stmmac_axi *stmmac_axi_setup(struct platform_device *pdev) axi->axi_wr_osr_lmt = 1; if (of_property_read_u32(np, "snps,rd_osr_lmt", &axi->axi_rd_osr_lmt)) axi->axi_rd_osr_lmt = 1; - of_property_read_u32_array(np, "snps,blen", axi->axi_blen, AXI_BLEN); - stmmac_axi_blen_to_mask(&axi->axi_blen_regval, axi->axi_blen, AXI_BLEN); + of_property_read_u32_array(np, "snps,blen", axi_blen, AXI_BLEN); + stmmac_axi_blen_to_mask(&axi->axi_blen_regval, axi_blen, AXI_BLEN); of_node_put(np); return axi; diff --git a/include/linux/stmmac.h b/include/linux/stmmac.h index d1a41fe0825f..f1054b9c2d8a 100644 --- a/include/linux/stmmac.h +++ b/include/linux/stmmac.h @@ -114,7 +114,6 @@ struct stmmac_axi { u32 axi_rd_osr_lmt; bool axi_kbbe; u32 axi_blen_regval; - u32 axi_blen[AXI_BLEN]; bool axi_fb; bool axi_mb; bool axi_rb; From bccaf1fe08f2c9f96f6bc38391d41e67f6bf38e3 Mon Sep 17 00:00:00 2001 From: Alexander Dahl Date: Wed, 19 Nov 2025 13:47:36 +0100 Subject: [PATCH 0962/1381] net: phy: adin1100: Fix software power-down ready condition MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Value CRSM_SFT_PD written to Software Power-Down Control Register (CRSM_SFT_PD_CNTRL) is 0x01 and therefor different to value CRSM_SFT_PD_RDY (0x02) read from System Status Register (CRSM_STAT) for confirmation powerdown has been reached. The condition could have only worked when disabling powerdown (both 0x00), but never when enabling it (0x01 != 0x02). Result is a timeout, like so: $ ifdown eth0 macb f802c000.ethernet eth0: Link is Down ADIN1100 f802c000.ethernet-ffffffff:01: adin_set_powerdown_mode failed: -110 ADIN1100 f802c000.ethernet-ffffffff:01: adin_set_powerdown_mode failed: -110 Fixes: 7eaf9132996a ("net: phy: adin1100: Add initial support for ADIN1100 industrial PHY") Signed-off-by: Alexander Dahl Reviewed-by: Russell King (Oracle) Acked-by: Nuno Sá Link: https://patch.msgid.link/20251119124737.280939-2-ada@thorsis.com Signed-off-by: Jakub Kicinski --- drivers/net/phy/adin1100.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/phy/adin1100.c b/drivers/net/phy/adin1100.c index bd7a47a903ac..10b796c2daee 100644 --- a/drivers/net/phy/adin1100.c +++ b/drivers/net/phy/adin1100.c @@ -201,7 +201,7 @@ static int adin_set_powerdown_mode(struct phy_device *phydev, bool en) return ret; return phy_read_mmd_poll_timeout(phydev, MDIO_MMD_VEND1, ADIN_CRSM_STAT, ret, - (ret & ADIN_CRSM_SFT_PD_RDY) == val, + !!(ret & ADIN_CRSM_SFT_PD_RDY) == en, 1000, 30000, true); } From 5894cab4e1b97365c0bb5ca2c29fe84e46b00d26 Mon Sep 17 00:00:00 2001 From: Alexander Dahl Date: Wed, 19 Nov 2025 13:47:37 +0100 Subject: [PATCH 0963/1381] net: phy: adin1100: Simplify register value passing MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The additional use case for that variable is gone, the expression is simple enough to pass it inline now. Signed-off-by: Alexander Dahl Reviewed-by: Russell King (Oracle) Acked-by: Nuno Sá Link: https://patch.msgid.link/20251119124737.280939-3-ada@thorsis.com Signed-off-by: Jakub Kicinski --- drivers/net/phy/adin1100.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/drivers/net/phy/adin1100.c b/drivers/net/phy/adin1100.c index 10b796c2daee..8f9753d4318c 100644 --- a/drivers/net/phy/adin1100.c +++ b/drivers/net/phy/adin1100.c @@ -192,11 +192,10 @@ static irqreturn_t adin_phy_handle_interrupt(struct phy_device *phydev) static int adin_set_powerdown_mode(struct phy_device *phydev, bool en) { int ret; - int val; - val = en ? ADIN_CRSM_SFT_PD_CNTRL_EN : 0; ret = phy_write_mmd(phydev, MDIO_MMD_VEND1, - ADIN_CRSM_SFT_PD_CNTRL, val); + ADIN_CRSM_SFT_PD_CNTRL, + en ? ADIN_CRSM_SFT_PD_CNTRL_EN : 0); if (ret < 0) return ret; From bd048f8ce6ce8d5e16415fd6f7506e916cb3d418 Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Wed, 19 Nov 2025 08:05:45 +0100 Subject: [PATCH 0964/1381] net: phy: fixed_phy: fix missing initialization of fixed phy link Original change remove the link initialization from the passed struct fixed_phy_status, but @status is also passed to __fixed_phy_add(), where it is saved. Make sure that copy also has link set to 1. Fixes: 9f07af1d2742 ("net: phy: fixed_phy: initialize the link status as up") Signed-off-by: Heiner Kallweit Link: https://patch.msgid.link/dab6c10e-725e-4648-9662-39cc821723d0@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/phy/fixed_phy.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/phy/fixed_phy.c b/drivers/net/phy/fixed_phy.c index 1ad77f542c1f..1ea67119f08f 100644 --- a/drivers/net/phy/fixed_phy.c +++ b/drivers/net/phy/fixed_phy.c @@ -124,6 +124,7 @@ static int __fixed_phy_add(int phy_addr, fp->addr = phy_addr; fp->status = *status; + fp->status.link = true; list_add_tail(&fp->node, &fmb_phys); From d99b408ed8e5f7c070274f4d6c73023b2aae22b3 Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Wed, 19 Nov 2025 07:55:47 +0100 Subject: [PATCH 0965/1381] net: phy: fixed_phy: remove not needed initialization of phy_device members All these members are populated by the phylib state machine once the PHY has been started, based on the fixed autoneg results. Signed-off-by: Heiner Kallweit Link: https://patch.msgid.link/bc666a53-5469-4e9c-85a1-dd285aadfe4f@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/phy/fixed_phy.c | 7 ------- 1 file changed, 7 deletions(-) diff --git a/drivers/net/phy/fixed_phy.c b/drivers/net/phy/fixed_phy.c index 1ea67119f08f..50684271f81a 100644 --- a/drivers/net/phy/fixed_phy.c +++ b/drivers/net/phy/fixed_phy.c @@ -173,13 +173,6 @@ struct phy_device *fixed_phy_register(const struct fixed_phy_status *status, return ERR_PTR(-EINVAL); } - /* propagate the fixed link values to struct phy_device */ - phy->link = 1; - phy->speed = status->speed; - phy->duplex = status->duplex; - phy->pause = status->pause; - phy->asym_pause = status->asym_pause; - of_node_get(np); phy->mdio.dev.of_node = np; phy->is_pseudo_fixed_link = true; From 5cb7b71b76f1235ee2064c30922864f24e31a076 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 19 Nov 2025 18:10:13 -0800 Subject: [PATCH 0966/1381] selftests: net: py: coding style improvements We're about to add more features here and finding new issues with old ones in place is hard. Address ruff checks: - bare exceptions - f-string with no params - unused import We need to use BaseException when handling defer(), as Petr points out. This retains the old behavior of ignoring SIGTERM while running cleanups. Reviewed-by: Willem de Bruijn Reviewed-by: Petr Machata Link: https://patch.msgid.link/20251120021024.2944527-2-kuba@kernel.org Signed-off-by: Jakub Kicinski --- tools/testing/selftests/net/lib/py/ksft.py | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/tools/testing/selftests/net/lib/py/ksft.py b/tools/testing/selftests/net/lib/py/ksft.py index 83b1574f7719..2eed24244b8a 100644 --- a/tools/testing/selftests/net/lib/py/ksft.py +++ b/tools/testing/selftests/net/lib/py/ksft.py @@ -1,6 +1,5 @@ # SPDX-License-Identifier: GPL-2.0 -import builtins import functools import inspect import signal @@ -163,7 +162,7 @@ def ksft_flush_defer(): entry = global_defer_queue.pop() try: entry.exec_only() - except: + except BaseException: ksft_pr(f"Exception while handling defer / cleanup (callback {i} of {qlen_start})!") tb = traceback.format_exc() for line in tb.strip().split('\n'): @@ -181,7 +180,7 @@ def ksft_disruptive(func): @functools.wraps(func) def wrapper(*args, **kwargs): if not KSFT_DISRUPTIVE: - raise KsftSkipEx(f"marked as disruptive") + raise KsftSkipEx("marked as disruptive") return func(*args, **kwargs) return wrapper @@ -199,7 +198,7 @@ def ksft_setup(env): return False try: return bool(int(value)) - except: + except Exception: raise Exception(f"failed to parse {name}") if "DISRUPTIVE" in env: From 80970e0fc07ecb1fbaefeb2c912aa2b0c04ed557 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 19 Nov 2025 18:10:14 -0800 Subject: [PATCH 0967/1381] selftests: net: py: extract the case generation logic In preparation for adding test variants move the test case collection logic to a dedicated function. New helper returns (function, args, name, ) tuples. The main test loop can simply run them, not much logic or discernment needed. Reviewed-by: Petr Machata Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20251120021024.2944527-3-kuba@kernel.org Signed-off-by: Jakub Kicinski --- tools/testing/selftests/net/lib/py/ksft.py | 31 +++++++++++++++------- 1 file changed, 22 insertions(+), 9 deletions(-) diff --git a/tools/testing/selftests/net/lib/py/ksft.py b/tools/testing/selftests/net/lib/py/ksft.py index 2eed24244b8a..3ef39436302a 100644 --- a/tools/testing/selftests/net/lib/py/ksft.py +++ b/tools/testing/selftests/net/lib/py/ksft.py @@ -135,7 +135,7 @@ def ksft_busy_wait(cond, sleep=0.005, deadline=1, comment=""): time.sleep(sleep) -def ktap_result(ok, cnt=1, case="", comment=""): +def ktap_result(ok, cnt=1, case_name="", comment=""): global KSFT_RESULT_ALL KSFT_RESULT_ALL = KSFT_RESULT_ALL and ok @@ -145,8 +145,8 @@ def ktap_result(ok, cnt=1, case="", comment=""): res += "ok " res += str(cnt) + " " res += KSFT_MAIN_NAME - if case: - res += "." + str(case.__name__) + if case_name: + res += "." + case_name if comment: res += " # " + comment print(res, flush=True) @@ -219,9 +219,13 @@ def _ksft_intr(signum, frame): ksft_pr(f"Ignoring SIGTERM (cnt: {term_cnt}), already exiting...") -def ksft_run(cases=None, globs=None, case_pfx=None, args=()): - cases = cases or [] +def _ksft_generate_test_cases(cases, globs, case_pfx, args): + """Generate a flat list of (func, args, name) tuples""" + cases = cases or [] + test_cases = [] + + # If using the globs method find all relevant functions if globs and case_pfx: for key, value in globs.items(): if not callable(value): @@ -231,6 +235,15 @@ def ksft_run(cases=None, globs=None, case_pfx=None, args=()): cases.append(value) break + for func in cases: + test_cases.append((func, args, func.__name__)) + + return test_cases + + +def ksft_run(cases=None, globs=None, case_pfx=None, args=()): + test_cases = _ksft_generate_test_cases(cases, globs, case_pfx, args) + global term_cnt term_cnt = 0 prev_sigterm = signal.signal(signal.SIGTERM, _ksft_intr) @@ -238,19 +251,19 @@ def ksft_run(cases=None, globs=None, case_pfx=None, args=()): totals = {"pass": 0, "fail": 0, "skip": 0, "xfail": 0} print("TAP version 13", flush=True) - print("1.." + str(len(cases)), flush=True) + print("1.." + str(len(test_cases)), flush=True) global KSFT_RESULT cnt = 0 stop = False - for case in cases: + for func, args, name in test_cases: KSFT_RESULT = True cnt += 1 comment = "" cnt_key = "" try: - case(*args) + func(*args) except KsftSkipEx as e: comment = "SKIP " + str(e) cnt_key = 'skip' @@ -272,7 +285,7 @@ def ksft_run(cases=None, globs=None, case_pfx=None, args=()): if not cnt_key: cnt_key = 'pass' if KSFT_RESULT else 'fail' - ktap_result(KSFT_RESULT, cnt, case, comment=comment) + ktap_result(KSFT_RESULT, cnt, name, comment=comment) totals[cnt_key] += 1 if stop: From 6ae67f115986734bc24dfed5cee1daafb5351623 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 19 Nov 2025 18:10:15 -0800 Subject: [PATCH 0968/1381] selftests: net: py: add test variants There's a lot of cases where we try to re-run the same code with different parameters. We currently need to either use a generator method or create a "main" case implementation which then gets called by trivial case functions: def _test(x, y, z): ... def case_int(): _test(1, 2, 3) def case_str(): _test('a', 'b', 'c') Add support for variants, similar to kselftests_harness.h and a lot of other frameworks. Variants can be added as decorator to test functions: @ksft_variants([(1, 2, 3), ('a', 'b', 'c')]) def case(x, y, z): ... ksft_run() will auto-generate case names: case.1_2_3 case.a_b_c Because the names may not always be pretty (and to avoid forcing classes to implement case-friendly __str__()) add a wrapper class KsftNamedVariant which lets the user specify the name for the variant. Note that ksft_run's args are still supported. ksft_run splices args and variant params together. Reviewed-by: Willem de Bruijn Reviewed-by: Petr Machata Link: https://patch.msgid.link/20251120021024.2944527-4-kuba@kernel.org Signed-off-by: Jakub Kicinski --- .../drivers/net/hw/lib/py/__init__.py | 4 +- .../selftests/drivers/net/lib/py/__init__.py | 4 +- .../testing/selftests/net/lib/py/__init__.py | 5 +- tools/testing/selftests/net/lib/py/ksft.py | 55 ++++++++++++++++++- 4 files changed, 61 insertions(+), 7 deletions(-) diff --git a/tools/testing/selftests/drivers/net/hw/lib/py/__init__.py b/tools/testing/selftests/drivers/net/hw/lib/py/__init__.py index fb010a48a5a1..0c61debf86fb 100644 --- a/tools/testing/selftests/drivers/net/hw/lib/py/__init__.py +++ b/tools/testing/selftests/drivers/net/hw/lib/py/__init__.py @@ -25,7 +25,7 @@ try: fd_read_timeout, ip, rand_port, wait_port_listen, wait_file from net.lib.py import KsftSkipEx, KsftFailEx, KsftXfailEx from net.lib.py import ksft_disruptive, ksft_exit, ksft_pr, ksft_run, \ - ksft_setup + ksft_setup, ksft_variants, KsftNamedVariant from net.lib.py import ksft_eq, ksft_ge, ksft_in, ksft_is, ksft_lt, \ ksft_ne, ksft_not_in, ksft_raises, ksft_true, ksft_gt, ksft_not_none from drivers.net.lib.py import GenerateTraffic, Remote @@ -40,7 +40,7 @@ try: "wait_port_listen", "wait_file", "KsftSkipEx", "KsftFailEx", "KsftXfailEx", "ksft_disruptive", "ksft_exit", "ksft_pr", "ksft_run", - "ksft_setup", + "ksft_setup", "ksft_variants", "KsftNamedVariant", "ksft_eq", "ksft_ge", "ksft_in", "ksft_is", "ksft_lt", "ksft_ne", "ksft_not_in", "ksft_raises", "ksft_true", "ksft_gt", "ksft_not_none", "ksft_not_none", diff --git a/tools/testing/selftests/drivers/net/lib/py/__init__.py b/tools/testing/selftests/drivers/net/lib/py/__init__.py index b0c6300150fb..d9d035634a31 100644 --- a/tools/testing/selftests/drivers/net/lib/py/__init__.py +++ b/tools/testing/selftests/drivers/net/lib/py/__init__.py @@ -25,7 +25,7 @@ try: fd_read_timeout, ip, rand_port, wait_port_listen, wait_file from net.lib.py import KsftSkipEx, KsftFailEx, KsftXfailEx from net.lib.py import ksft_disruptive, ksft_exit, ksft_pr, ksft_run, \ - ksft_setup + ksft_setup, ksft_variants, KsftNamedVariant from net.lib.py import ksft_eq, ksft_ge, ksft_in, ksft_is, ksft_lt, \ ksft_ne, ksft_not_in, ksft_raises, ksft_true, ksft_gt, ksft_not_none @@ -38,7 +38,7 @@ try: "wait_port_listen", "wait_file", "KsftSkipEx", "KsftFailEx", "KsftXfailEx", "ksft_disruptive", "ksft_exit", "ksft_pr", "ksft_run", - "ksft_setup", + "ksft_setup", "ksft_variants", "KsftNamedVariant", "ksft_eq", "ksft_ge", "ksft_in", "ksft_is", "ksft_lt", "ksft_ne", "ksft_not_in", "ksft_raises", "ksft_true", "ksft_gt", "ksft_not_none", "ksft_not_none"] diff --git a/tools/testing/selftests/net/lib/py/__init__.py b/tools/testing/selftests/net/lib/py/__init__.py index 97b7cf2b20eb..40f9ce307dd1 100644 --- a/tools/testing/selftests/net/lib/py/__init__.py +++ b/tools/testing/selftests/net/lib/py/__init__.py @@ -8,7 +8,8 @@ from .consts import KSRC from .ksft import KsftFailEx, KsftSkipEx, KsftXfailEx, ksft_pr, ksft_eq, \ ksft_ne, ksft_true, ksft_not_none, ksft_in, ksft_not_in, ksft_is, \ ksft_ge, ksft_gt, ksft_lt, ksft_raises, ksft_busy_wait, \ - ktap_result, ksft_disruptive, ksft_setup, ksft_run, ksft_exit + ktap_result, ksft_disruptive, ksft_setup, ksft_run, ksft_exit, \ + ksft_variants, KsftNamedVariant from .netns import NetNS, NetNSEnter from .nsim import NetdevSim, NetdevSimDev from .utils import CmdExitFailure, fd_read_timeout, cmd, bkg, defer, \ @@ -21,7 +22,7 @@ __all__ = ["KSRC", "ksft_ne", "ksft_true", "ksft_not_none", "ksft_in", "ksft_not_in", "ksft_is", "ksft_ge", "ksft_gt", "ksft_lt", "ksft_raises", "ksft_busy_wait", "ktap_result", "ksft_disruptive", "ksft_setup", - "ksft_run", "ksft_exit", + "ksft_run", "ksft_exit", "ksft_variants", "KsftNamedVariant", "NetNS", "NetNSEnter", "CmdExitFailure", "fd_read_timeout", "cmd", "bkg", "defer", "bpftool", "ip", "ethtool", "bpftrace", "rand_port", diff --git a/tools/testing/selftests/net/lib/py/ksft.py b/tools/testing/selftests/net/lib/py/ksft.py index 3ef39436302a..ebd82940ee50 100644 --- a/tools/testing/selftests/net/lib/py/ksft.py +++ b/tools/testing/selftests/net/lib/py/ksft.py @@ -6,6 +6,7 @@ import signal import sys import time import traceback +from collections import namedtuple from .consts import KSFT_MAIN_NAME from .utils import global_defer_queue @@ -170,6 +171,10 @@ def ksft_flush_defer(): KSFT_RESULT = False +KsftCaseFunction = namedtuple("KsftCaseFunction", + ['name', 'original_func', 'variants']) + + def ksft_disruptive(func): """ Decorator that marks the test as disruptive (e.g. the test @@ -185,6 +190,42 @@ def ksft_disruptive(func): return wrapper +class KsftNamedVariant: + """ Named string name + argument list tuple for @ksft_variants """ + + def __init__(self, name, *params): + self.params = params + self.name = name or "_".join([str(x) for x in self.params]) + + +def ksft_variants(params): + """ + Decorator defining the sets of inputs for a test. + The parameters will be included in the name of the resulting sub-case. + Parameters can be either single object, tuple or a KsftNamedVariant. + The argument can be a list or a generator. + + Example: + + @ksft_variants([ + (1, "a"), + (2, "b"), + KsftNamedVariant("three", 3, "c"), + ]) + def my_case(cfg, a, b): + pass # ... + + ksft_run(cases=[my_case], args=(cfg, )) + + Will generate cases: + my_case.1_a + my_case.2_b + my_case.three + """ + + return lambda func: KsftCaseFunction(func.__name__, func, params) + + def ksft_setup(env): """ Setup test framework global state from the environment. @@ -236,7 +277,19 @@ def _ksft_generate_test_cases(cases, globs, case_pfx, args): break for func in cases: - test_cases.append((func, args, func.__name__)) + if isinstance(func, KsftCaseFunction): + # Parametrized test - create case for each param + for param in func.variants: + if not isinstance(param, KsftNamedVariant): + if not isinstance(param, tuple): + param = (param, ) + param = KsftNamedVariant(None, *param) + + test_cases.append((func.original_func, + (*args, *param.params), + func.name + "." + param.name)) + else: + test_cases.append((func, args, func.__name__)) return test_cases From 173227d7d6c4902a017ba9bbf3fb0d99f021f478 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 19 Nov 2025 18:10:16 -0800 Subject: [PATCH 0969/1381] selftests: drv-net: xdp: use variants for qstat tests Use just-added ksft variants for XDP qstat tests. While at it correct the number of packets, we're sending 1000 packets now. Reviewed-by: Petr Machata Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20251120021024.2944527-5-kuba@kernel.org Signed-off-by: Jakub Kicinski --- tools/testing/selftests/drivers/net/xdp.py | 42 ++++++++-------------- 1 file changed, 14 insertions(+), 28 deletions(-) diff --git a/tools/testing/selftests/drivers/net/xdp.py b/tools/testing/selftests/drivers/net/xdp.py index 834a37ae7d0d..e54df158dfe9 100755 --- a/tools/testing/selftests/drivers/net/xdp.py +++ b/tools/testing/selftests/drivers/net/xdp.py @@ -12,6 +12,7 @@ from dataclasses import dataclass from enum import Enum from lib.py import ksft_run, ksft_exit, ksft_eq, ksft_ge, ksft_ne, ksft_pr +from lib.py import KsftNamedVariant, ksft_variants from lib.py import KsftFailEx, NetDrvEpEnv from lib.py import EthtoolFamily, NetdevFamily, NlError from lib.py import bkg, cmd, rand_port, wait_port_listen @@ -672,7 +673,18 @@ def test_xdp_native_adjst_head_shrnk_data(cfg): _validate_res(res, offset_lst, pkt_sz_lst) -def _test_xdp_native_ifc_stats(cfg, act): +@ksft_variants([ + KsftNamedVariant("pass", XDPAction.PASS), + KsftNamedVariant("drop", XDPAction.DROP), + KsftNamedVariant("tx", XDPAction.TX), +]) +def test_xdp_native_qstats(cfg, act): + """ + Send 1000 messages. Expect XDP action specified in @act. + Make sure the packets were counted to interface level qstats + (Rx, and Tx if act is TX). + """ + cfg.require_cmd("socat") bpf_info = BPFProgInfo("xdp_prog", "xdp_native.bpf.o", "xdp", 1500) @@ -733,30 +745,6 @@ def _test_xdp_native_ifc_stats(cfg, act): ksft_ge(after['tx-packets'], before['tx-packets']) -def test_xdp_native_qstats_pass(cfg): - """ - Send 2000 messages, expect XDP_PASS, make sure the packets were counted - to interface level qstats (Rx). - """ - _test_xdp_native_ifc_stats(cfg, XDPAction.PASS) - - -def test_xdp_native_qstats_drop(cfg): - """ - Send 2000 messages, expect XDP_DROP, make sure the packets were counted - to interface level qstats (Rx). - """ - _test_xdp_native_ifc_stats(cfg, XDPAction.DROP) - - -def test_xdp_native_qstats_tx(cfg): - """ - Send 2000 messages, expect XDP_TX, make sure the packets were counted - to interface level qstats (Rx and Tx) - """ - _test_xdp_native_ifc_stats(cfg, XDPAction.TX) - - def main(): """ Main function to execute the XDP tests. @@ -781,9 +769,7 @@ def main(): test_xdp_native_adjst_tail_shrnk_data, test_xdp_native_adjst_head_grow_data, test_xdp_native_adjst_head_shrnk_data, - test_xdp_native_qstats_pass, - test_xdp_native_qstats_drop, - test_xdp_native_qstats_tx, + test_xdp_native_qstats, ], args=(cfg,)) ksft_exit() From 89268f7dbca1842cd38d3a4578ba7e3b6789c5ad Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 19 Nov 2025 18:10:17 -0800 Subject: [PATCH 0970/1381] selftests: net: relocate gro and toeplitz tests to drivers/net The GRO test can run on a real device or a veth. The Toeplitz hash test can only run on a real device. Move them from net/ to drivers/net/ and drivers/net/hw/ respectively. There are two scripts which set up the environment for these tests setup_loopback.sh and setup_veth.sh. Move those scripts to net/lib. The paths to the setup files are a little ugly but they will be deleted shortly. toeplitz_client.sh is not a test in itself, but rather a helper to send traffic, so add it to TEST_FILES rather than TEST_PROGS. Reviewed-by: Petr Machata Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20251120021024.2944527-6-kuba@kernel.org Signed-off-by: Jakub Kicinski --- tools/testing/selftests/drivers/net/.gitignore | 1 + tools/testing/selftests/drivers/net/Makefile | 2 ++ tools/testing/selftests/{ => drivers}/net/gro.c | 2 +- tools/testing/selftests/{ => drivers}/net/gro.sh | 4 ++-- tools/testing/selftests/drivers/net/hw/.gitignore | 1 + tools/testing/selftests/drivers/net/hw/Makefile | 7 ++++++- tools/testing/selftests/{net => drivers/net/hw}/toeplitz.c | 2 +- .../testing/selftests/{net => drivers/net/hw}/toeplitz.sh | 2 +- .../selftests/{net => drivers/net/hw}/toeplitz_client.sh | 0 tools/testing/selftests/net/.gitignore | 2 -- tools/testing/selftests/net/Makefile | 7 ------- tools/testing/selftests/net/lib/Makefile | 2 ++ tools/testing/selftests/net/{ => lib}/setup_loopback.sh | 0 tools/testing/selftests/net/{ => lib}/setup_veth.sh | 0 14 files changed, 17 insertions(+), 15 deletions(-) rename tools/testing/selftests/{ => drivers}/net/gro.c (99%) rename tools/testing/selftests/{ => drivers}/net/gro.sh (95%) rename tools/testing/selftests/{net => drivers/net/hw}/toeplitz.c (99%) rename tools/testing/selftests/{net => drivers/net/hw}/toeplitz.sh (98%) rename tools/testing/selftests/{net => drivers/net/hw}/toeplitz_client.sh (100%) rename tools/testing/selftests/net/{ => lib}/setup_loopback.sh (100%) rename tools/testing/selftests/net/{ => lib}/setup_veth.sh (100%) diff --git a/tools/testing/selftests/drivers/net/.gitignore b/tools/testing/selftests/drivers/net/.gitignore index 585ecb4d5dc4..3633c7a3ed65 100644 --- a/tools/testing/selftests/drivers/net/.gitignore +++ b/tools/testing/selftests/drivers/net/.gitignore @@ -1,3 +1,4 @@ # SPDX-License-Identifier: GPL-2.0-only +gro napi_id_helper psp_responder diff --git a/tools/testing/selftests/drivers/net/Makefile b/tools/testing/selftests/drivers/net/Makefile index 33f4816216ec..7083a8707c4e 100644 --- a/tools/testing/selftests/drivers/net/Makefile +++ b/tools/testing/selftests/drivers/net/Makefile @@ -6,10 +6,12 @@ TEST_INCLUDES := $(wildcard lib/py/*.py) \ ../../net/lib.sh \ TEST_GEN_FILES := \ + gro \ napi_id_helper \ # end of TEST_GEN_FILES TEST_PROGS := \ + gro.sh \ hds.py \ napi_id.py \ napi_threaded.py \ diff --git a/tools/testing/selftests/net/gro.c b/tools/testing/selftests/drivers/net/gro.c similarity index 99% rename from tools/testing/selftests/net/gro.c rename to tools/testing/selftests/drivers/net/gro.c index cfc39f70635d..9b9be0cf8f7f 100644 --- a/tools/testing/selftests/net/gro.c +++ b/tools/testing/selftests/drivers/net/gro.c @@ -57,7 +57,7 @@ #include #include -#include "../kselftest.h" +#include "../../kselftest.h" #define DPORT 8000 #define SPORT 1500 diff --git a/tools/testing/selftests/net/gro.sh b/tools/testing/selftests/drivers/net/gro.sh similarity index 95% rename from tools/testing/selftests/net/gro.sh rename to tools/testing/selftests/drivers/net/gro.sh index 4c5144c6f652..bd3cf6d02eda 100755 --- a/tools/testing/selftests/net/gro.sh +++ b/tools/testing/selftests/drivers/net/gro.sh @@ -90,9 +90,9 @@ while getopts "i:t:p:" opt; do done if [ -n "$dev" ]; then - source setup_loopback.sh + source $(dirname $0)/../../net/lib/setup_loopback.sh else - source setup_veth.sh + source $(dirname $0)/../../net/lib/setup_veth.sh fi setup diff --git a/tools/testing/selftests/drivers/net/hw/.gitignore b/tools/testing/selftests/drivers/net/hw/.gitignore index 6942bf575497..46540468a775 100644 --- a/tools/testing/selftests/drivers/net/hw/.gitignore +++ b/tools/testing/selftests/drivers/net/hw/.gitignore @@ -1,3 +1,4 @@ # SPDX-License-Identifier: GPL-2.0-only iou-zcrx ncdevmem +toeplitz diff --git a/tools/testing/selftests/drivers/net/hw/Makefile b/tools/testing/selftests/drivers/net/hw/Makefile index 8133d1a0051c..c9dced8c934a 100644 --- a/tools/testing/selftests/drivers/net/hw/Makefile +++ b/tools/testing/selftests/drivers/net/hw/Makefile @@ -1,6 +1,9 @@ # SPDX-License-Identifier: GPL-2.0+ OR MIT -TEST_GEN_FILES = iou-zcrx +TEST_GEN_FILES := \ + iou-zcrx \ + toeplitz \ +# end of TEST_GEN_FILES TEST_PROGS = \ csum.py \ @@ -21,12 +24,14 @@ TEST_PROGS = \ rss_ctx.py \ rss_flow_label.py \ rss_input_xfrm.py \ + toeplitz.sh \ tso.py \ xsk_reconfig.py \ # TEST_FILES := \ ethtool_lib.sh \ + toeplitz_client.sh \ # TEST_INCLUDES := \ diff --git a/tools/testing/selftests/net/toeplitz.c b/tools/testing/selftests/drivers/net/hw/toeplitz.c similarity index 99% rename from tools/testing/selftests/net/toeplitz.c rename to tools/testing/selftests/drivers/net/hw/toeplitz.c index 9ba03164d73a..bf74aa25345d 100644 --- a/tools/testing/selftests/net/toeplitz.c +++ b/tools/testing/selftests/drivers/net/hw/toeplitz.c @@ -52,7 +52,7 @@ #include #include -#include "../kselftest.h" +#include "../../../kselftest.h" #define TOEPLITZ_KEY_MIN_LEN 40 #define TOEPLITZ_KEY_MAX_LEN 60 diff --git a/tools/testing/selftests/net/toeplitz.sh b/tools/testing/selftests/drivers/net/hw/toeplitz.sh similarity index 98% rename from tools/testing/selftests/net/toeplitz.sh rename to tools/testing/selftests/drivers/net/hw/toeplitz.sh index 8ff172f7bb1b..d236b666dd3b 100755 --- a/tools/testing/selftests/net/toeplitz.sh +++ b/tools/testing/selftests/drivers/net/hw/toeplitz.sh @@ -11,7 +11,7 @@ # invoke as ./toeplitz.sh (-i ) -u|-t -4|-6 \ # [(-rss -irq_prefix )|(-rps )] -source setup_loopback.sh +source $(dirname $0)/../../../net/lib/setup_loopback.sh readonly SERVER_IP4="192.168.1.200/24" readonly SERVER_IP6="fda8::1/64" readonly SERVER_MAC="aa:00:00:00:00:02" diff --git a/tools/testing/selftests/net/toeplitz_client.sh b/tools/testing/selftests/drivers/net/hw/toeplitz_client.sh similarity index 100% rename from tools/testing/selftests/net/toeplitz_client.sh rename to tools/testing/selftests/drivers/net/hw/toeplitz_client.sh diff --git a/tools/testing/selftests/net/.gitignore b/tools/testing/selftests/net/.gitignore index f627f1a2a2b8..283ca5ffc244 100644 --- a/tools/testing/selftests/net/.gitignore +++ b/tools/testing/selftests/net/.gitignore @@ -7,7 +7,6 @@ cmsg_sender diag_uid epoll_busy_poll fin_ack_lat -gro hwtstamp_config io_uring_zerocopy_tx ioam6_parser @@ -57,7 +56,6 @@ tcp_port_share tfo timestamping tls -toeplitz tools tun txring_overwrite diff --git a/tools/testing/selftests/net/Makefile b/tools/testing/selftests/net/Makefile index b5127e968108..b66ba04f19d9 100644 --- a/tools/testing/selftests/net/Makefile +++ b/tools/testing/selftests/net/Makefile @@ -38,7 +38,6 @@ TEST_PROGS := \ fq_band_pktlimit.sh \ gre_gso.sh \ gre_ipv6_lladdr.sh \ - gro.sh \ icmp.sh \ icmp_redirect.sh \ io_uring_zerocopy_tx.sh \ @@ -121,8 +120,6 @@ TEST_PROGS := \ # end of TEST_PROGS TEST_PROGS_EXTENDED := \ - toeplitz.sh \ - toeplitz_client.sh \ xfrm_policy_add_speed.sh \ # end of TEST_PROGS_EXTENDED @@ -130,7 +127,6 @@ TEST_GEN_FILES := \ bind_bhash \ cmsg_sender \ fin_ack_lat \ - gro \ hwtstamp_config \ io_uring_zerocopy_tx \ ioam6_parser \ @@ -159,7 +155,6 @@ TEST_GEN_FILES := \ tcp_mmap \ tfo \ timestamping \ - toeplitz \ txring_overwrite \ txtimestamp \ udpgso \ @@ -193,8 +188,6 @@ TEST_FILES := \ in_netns.sh \ lib.sh \ settings \ - setup_loopback.sh \ - setup_veth.sh \ # end of TEST_FILES # YNL files, must be before "include ..lib.mk" diff --git a/tools/testing/selftests/net/lib/Makefile b/tools/testing/selftests/net/lib/Makefile index ce795bc0a1af..c10796933d42 100644 --- a/tools/testing/selftests/net/lib/Makefile +++ b/tools/testing/selftests/net/lib/Makefile @@ -8,6 +8,8 @@ CFLAGS += -I../../ TEST_FILES := \ ../../../../net/ynl \ ../../../../../Documentation/netlink/specs \ + setup_loopback.sh \ + setup_veth.sh \ # end of TEST_FILES TEST_GEN_FILES := \ diff --git a/tools/testing/selftests/net/setup_loopback.sh b/tools/testing/selftests/net/lib/setup_loopback.sh similarity index 100% rename from tools/testing/selftests/net/setup_loopback.sh rename to tools/testing/selftests/net/lib/setup_loopback.sh diff --git a/tools/testing/selftests/net/setup_veth.sh b/tools/testing/selftests/net/lib/setup_veth.sh similarity index 100% rename from tools/testing/selftests/net/setup_veth.sh rename to tools/testing/selftests/net/lib/setup_veth.sh From e02b52ecef5ba468f55ed0dc38de6016424d9703 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 19 Nov 2025 18:10:18 -0800 Subject: [PATCH 0971/1381] selftests: net: py: support ksft ready without wait There's a common synchronization problem when a script (Python test) uses a C program to set up some state (usually start a receiving process for traffic). The script needs to know when the process has fully initialized. The inverse of the problem exists for shutting the process down - we need a reliable way to tell the process to exit. We added helpers to do this safely in commit 71477137994f ("selftests: drv-net: add a way to wait for a local process") unfortunately the two operations (wait for init, and shutdown) are controlled by a single parameter (ksft_wait). Add support for using ksft_ready without using the second fd for exit. This is useful for programs which wait for a specific number of packets to rx so exit_wait is a good match, but we still need to wait for init. Reviewed-by: Petr Machata Reviewed-by: Willem de Bruijn Reviewed-by: breno Leitao Link: https://patch.msgid.link/20251120021024.2944527-7-kuba@kernel.org Signed-off-by: Jakub Kicinski --- tools/testing/selftests/net/lib/py/utils.py | 20 ++++++++++++-------- 1 file changed, 12 insertions(+), 8 deletions(-) diff --git a/tools/testing/selftests/net/lib/py/utils.py b/tools/testing/selftests/net/lib/py/utils.py index cb40ecef9456..106ee1f2df86 100644 --- a/tools/testing/selftests/net/lib/py/utils.py +++ b/tools/testing/selftests/net/lib/py/utils.py @@ -32,7 +32,7 @@ class cmd: Use bkg() instead to run a command in the background. """ def __init__(self, comm, shell=None, fail=True, ns=None, background=False, - host=None, timeout=5, ksft_wait=None): + host=None, timeout=5, ksft_ready=None, ksft_wait=None): if ns: comm = f'ip netns exec {ns} ' + comm @@ -52,21 +52,25 @@ class cmd: # ksft_wait lets us wait for the background process to fully start, # we pass an FD to the child process, and wait for it to write back. # Similarly term_fd tells child it's time to exit. - pass_fds = () + pass_fds = [] env = os.environ.copy() if ksft_wait is not None: - rfd, ready_fd = os.pipe() wait_fd, self.ksft_term_fd = os.pipe() - pass_fds = (ready_fd, wait_fd, ) - env["KSFT_READY_FD"] = str(ready_fd) + pass_fds.append(wait_fd) env["KSFT_WAIT_FD"] = str(wait_fd) + ksft_ready = True # ksft_wait implies ready + if ksft_ready is not None: + rfd, ready_fd = os.pipe() + pass_fds.append(ready_fd) + env["KSFT_READY_FD"] = str(ready_fd) self.proc = subprocess.Popen(comm, shell=shell, stdout=subprocess.PIPE, stderr=subprocess.PIPE, pass_fds=pass_fds, env=env) if ksft_wait is not None: - os.close(ready_fd) os.close(wait_fd) + if ksft_ready is not None: + os.close(ready_fd) msg = fd_read_timeout(rfd, ksft_wait) os.close(rfd) if not msg: @@ -116,10 +120,10 @@ class bkg(cmd): with bkg("my_binary", ksft_wait=5): """ def __init__(self, comm, shell=None, fail=None, ns=None, host=None, - exit_wait=False, ksft_wait=None): + exit_wait=False, ksft_ready=None, ksft_wait=None): super().__init__(comm, background=True, shell=shell, fail=fail, ns=ns, host=host, - ksft_wait=ksft_wait) + ksft_ready=ksft_ready, ksft_wait=ksft_wait) self.terminate = not exit_wait and not ksft_wait self._exit_wait = exit_wait self.check_fail = fail From 15011a57d0ec95d7ee7d5c1a4b108b9ee3c9ee31 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 19 Nov 2025 18:10:19 -0800 Subject: [PATCH 0972/1381] selftests: net: py: read ip link info about remote dev We're already saving the info about the local dev in env.dev for the tests, save remote dev as well. This is more symmetric, env generally provides the same info for local and remote end. While at it make sure that we reliably get the detailed info about the local dev. nsim used to read the dev info without -d. Reviewed-by: Petr Machata Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20251120021024.2944527-8-kuba@kernel.org Signed-off-by: Jakub Kicinski --- tools/testing/selftests/drivers/net/lib/py/env.py | 2 ++ tools/testing/selftests/net/lib/py/nsim.py | 2 +- 2 files changed, 3 insertions(+), 1 deletion(-) diff --git a/tools/testing/selftests/drivers/net/lib/py/env.py b/tools/testing/selftests/drivers/net/lib/py/env.py index 01be3d9b9720..8b644fd84ff2 100644 --- a/tools/testing/selftests/drivers/net/lib/py/env.py +++ b/tools/testing/selftests/drivers/net/lib/py/env.py @@ -168,6 +168,8 @@ class NetDrvEpEnv(NetDrvEnvBase): # resolve remote interface name self.remote_ifname = self.resolve_remote_ifc() + self.remote_dev = ip("-d link show dev " + self.remote_ifname, + host=self.remote, json=True)[0] self._required_cmd = {} diff --git a/tools/testing/selftests/net/lib/py/nsim.py b/tools/testing/selftests/net/lib/py/nsim.py index 1a8cbe9acc48..7c640ed64c0b 100644 --- a/tools/testing/selftests/net/lib/py/nsim.py +++ b/tools/testing/selftests/net/lib/py/nsim.py @@ -27,7 +27,7 @@ class NetdevSim: self.port_index = port_index self.ns = ns self.dfs_dir = "%s/ports/%u/" % (nsimdev.dfs_dir, port_index) - ret = ip("-j link show dev %s" % ifname, ns=ns) + ret = ip("-d -j link show dev %s" % ifname, ns=ns) self.dev = json.loads(ret.stdout)[0] self.ifindex = self.dev["ifindex"] From 40dd789bc5a76066fe3dc40328b609f8302f6216 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 19 Nov 2025 18:10:20 -0800 Subject: [PATCH 0973/1381] netdevsim: pass packets thru GRO on Rx To replace veth in software GRO testing with netdevsim we need GRO support in netdevsim. Luckily we already have NAPI support so this change is trivial (compared to veth). Reviewed-by: Petr Machata Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20251120021024.2944527-9-kuba@kernel.org Signed-off-by: Jakub Kicinski --- drivers/net/netdevsim/netdev.c | 9 ++------- 1 file changed, 2 insertions(+), 7 deletions(-) diff --git a/drivers/net/netdevsim/netdev.c b/drivers/net/netdevsim/netdev.c index fa1d97885caa..2b713db16cd0 100644 --- a/drivers/net/netdevsim/netdev.c +++ b/drivers/net/netdevsim/netdev.c @@ -433,13 +433,8 @@ static int nsim_rcv(struct nsim_rq *rq, int budget) } /* skb might be discard at netif_receive_skb, save the len */ - skblen = skb->len; - skb_mark_napi_id(skb, &rq->napi); - ret = netif_receive_skb(skb); - if (ret == NET_RX_SUCCESS) - dev_dstats_rx_add(dev, skblen); - else - dev_dstats_rx_dropped(dev); + dev_dstats_rx_add(dev, skb->len); + napi_gro_receive(&rq->napi, skb); } nsim_start_peer_tx_queue(dev, rq); From fdb0267d565ab51b7b81d4c530135a5e77e34ff2 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 19 Nov 2025 18:10:21 -0800 Subject: [PATCH 0974/1381] selftests: drv-net: add a Python version of the GRO test Rewrite the existing gro.sh test in Python. The conversion not exact, the changes are related to integrating the test with our "remote endpoint" paradigm. The test now reads the IP addresses from the user config. It resolves the MAC address (including running over Layer 3 networks). Reviewed-by: Petr Machata Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20251120021024.2944527-10-kuba@kernel.org Signed-off-by: Jakub Kicinski --- tools/testing/selftests/drivers/net/Makefile | 2 +- tools/testing/selftests/drivers/net/gro.c | 3 + tools/testing/selftests/drivers/net/gro.py | 164 +++++++++++++++++++ tools/testing/selftests/drivers/net/gro.sh | 105 ------------ 4 files changed, 168 insertions(+), 106 deletions(-) create mode 100755 tools/testing/selftests/drivers/net/gro.py delete mode 100755 tools/testing/selftests/drivers/net/gro.sh diff --git a/tools/testing/selftests/drivers/net/Makefile b/tools/testing/selftests/drivers/net/Makefile index 7083a8707c4e..f5c71d993750 100644 --- a/tools/testing/selftests/drivers/net/Makefile +++ b/tools/testing/selftests/drivers/net/Makefile @@ -11,7 +11,7 @@ TEST_GEN_FILES := \ # end of TEST_GEN_FILES TEST_PROGS := \ - gro.sh \ + gro.py \ hds.py \ napi_id.py \ napi_threaded.py \ diff --git a/tools/testing/selftests/drivers/net/gro.c b/tools/testing/selftests/drivers/net/gro.c index 9b9be0cf8f7f..995b492f5bcb 100644 --- a/tools/testing/selftests/drivers/net/gro.c +++ b/tools/testing/selftests/drivers/net/gro.c @@ -58,6 +58,7 @@ #include #include "../../kselftest.h" +#include "../../net/lib/ksft.h" #define DPORT 8000 #define SPORT 1500 @@ -1127,6 +1128,8 @@ static void gro_receiver(void) set_timeout(rxfd); bind_packetsocket(rxfd); + ksft_ready(); + memset(correct_payload, 0, sizeof(correct_payload)); if (strcmp(testname, "data") == 0) { diff --git a/tools/testing/selftests/drivers/net/gro.py b/tools/testing/selftests/drivers/net/gro.py new file mode 100755 index 000000000000..ba83713bf7b5 --- /dev/null +++ b/tools/testing/selftests/drivers/net/gro.py @@ -0,0 +1,164 @@ +#!/usr/bin/env python3 +# SPDX-License-Identifier: GPL-2.0 + +""" +GRO (Generic Receive Offload) conformance tests. + +Validates that GRO coalescing works correctly by running the gro +binary in different configurations and checking for correct packet +coalescing behavior. + +Test cases: + - data: Data packets with same size/headers and correct seq numbers coalesce + - ack: Pure ACK packets do not coalesce + - flags: Packets with PSH, SYN, URG, RST flags do not coalesce + - tcp: Packets with incorrect checksum, non-consecutive seqno don't coalesce + - ip: Packets with different ECN, TTL, TOS, or IP options don't coalesce + - large: Packets larger than GRO_MAX_SIZE don't coalesce +""" + +import os +from lib.py import ksft_run, ksft_exit, ksft_pr +from lib.py import NetDrvEpEnv, KsftXfailEx +from lib.py import cmd, defer, bkg, ip +from lib.py import ksft_variants + + +def _resolve_dmac(cfg, ipver): + """ + Find the destination MAC address remote host should use to send packets + towards the local host. It may be a router / gateway address. + """ + + attr = "dmac" + ipver + # Cache the response across test cases + if hasattr(cfg, attr): + return getattr(cfg, attr) + + route = ip(f"-{ipver} route get {cfg.addr_v[ipver]}", + json=True, host=cfg.remote)[0] + gw = route.get("gateway") + # Local L2 segment, address directly + if not gw: + setattr(cfg, attr, cfg.dev['address']) + return getattr(cfg, attr) + + # ping to make sure neighbor is resolved, + # bind to an interface, for v6 the GW is likely link local + cmd(f"ping -c1 -W0 -I{cfg.remote_ifname} {gw}", host=cfg.remote) + + neigh = ip(f"neigh get {gw} dev {cfg.remote_ifname}", + json=True, host=cfg.remote)[0] + setattr(cfg, attr, neigh['lladdr']) + return getattr(cfg, attr) + + +def _write_defer_restore(cfg, path, val, defer_undo=False): + with open(path, "r", encoding="utf-8") as fp: + orig_val = fp.read().strip() + if str(val) == orig_val: + return + with open(path, "w", encoding="utf-8") as fp: + fp.write(val) + if defer_undo: + defer(_write_defer_restore, cfg, path, orig_val) + + +def _set_mtu_restore(dev, mtu, host): + if dev['mtu'] < mtu: + ip(f"link set dev {dev['ifname']} mtu {mtu}", host=host) + defer(ip, f"link set dev {dev['ifname']} mtu {dev['mtu']}", host=host) + + +def _setup(cfg, test_name): + """ Setup hardware loopback mode for GRO testing. """ + + if not hasattr(cfg, "bin_remote"): + cfg.bin_local = cfg.test_dir / "gro" + cfg.bin_remote = cfg.remote.deploy(cfg.bin_local) + + # "large" test needs at least 4k MTU + if test_name == "large": + _set_mtu_restore(cfg.dev, 4096, None) + _set_mtu_restore(cfg.remote_dev, 4096, cfg.remote) + + flush_path = f"/sys/class/net/{cfg.ifname}/gro_flush_timeout" + irq_path = f"/sys/class/net/{cfg.ifname}/napi_defer_hard_irqs" + + _write_defer_restore(cfg, flush_path, "200000", defer_undo=True) + _write_defer_restore(cfg, irq_path, "10", defer_undo=True) + + try: + # Disable TSO for local tests + cfg.require_nsim() # will raise KsftXfailEx if not running on nsim + + cmd(f"ethtool -K {cfg.ifname} gro on tso off") + cmd(f"ethtool -K {cfg.remote_ifname} gro on tso off", host=cfg.remote) + except KsftXfailEx: + pass + +def _gro_variants(): + """Generator that yields all combinations of protocol and test types.""" + + for protocol in ["ipv4", "ipv6", "ipip"]: + for test_name in ["data", "ack", "flags", "tcp", "ip", "large"]: + yield protocol, test_name + + +@ksft_variants(_gro_variants()) +def test(cfg, protocol, test_name): + """Run a single GRO test with retries.""" + + ipver = "6" if protocol[-1] == "6" else "4" + cfg.require_ipver(ipver) + + _setup(cfg, test_name) + + base_cmd_args = [ + f"--{protocol}", + f"--dmac {_resolve_dmac(cfg, ipver)}", + f"--smac {cfg.remote_dev['address']}", + f"--daddr {cfg.addr_v[ipver]}", + f"--saddr {cfg.remote_addr_v[ipver]}", + f"--test {test_name}", + "--verbose" + ] + base_args = " ".join(base_cmd_args) + + # Each test is run 6 times to deflake, because given the receive timing, + # not all packets that should coalesce will be considered in the same flow + # on every try. + max_retries = 6 + for attempt in range(max_retries): + rx_cmd = f"{cfg.bin_local} {base_args} --rx --iface {cfg.ifname}" + tx_cmd = f"{cfg.bin_remote} {base_args} --iface {cfg.remote_ifname}" + + fail_now = attempt >= max_retries - 1 + + with bkg(rx_cmd, ksft_ready=True, exit_wait=True, + fail=fail_now) as rx_proc: + cmd(tx_cmd, host=cfg.remote) + + if rx_proc.ret == 0: + return + + ksft_pr(rx_proc.stdout.strip().replace('\n', '\n# ')) + ksft_pr(rx_proc.stderr.strip().replace('\n', '\n# ')) + + if test_name == "large" and os.environ.get("KSFT_MACHINE_SLOW"): + ksft_pr(f"Ignoring {protocol}/{test_name} failure due to slow environment") + return + + ksft_pr(f"Attempt {attempt + 1}/{max_retries} failed, retrying...") + + +def main() -> None: + """ Ksft boiler plate main """ + + with NetDrvEpEnv(__file__) as cfg: + ksft_run(cases=[test], args=(cfg,)) + ksft_exit() + + +if __name__ == "__main__": + main() diff --git a/tools/testing/selftests/drivers/net/gro.sh b/tools/testing/selftests/drivers/net/gro.sh deleted file mode 100755 index bd3cf6d02eda..000000000000 --- a/tools/testing/selftests/drivers/net/gro.sh +++ /dev/null @@ -1,105 +0,0 @@ -#!/bin/bash -# SPDX-License-Identifier: GPL-2.0 - -readonly SERVER_MAC="aa:00:00:00:00:02" -readonly CLIENT_MAC="aa:00:00:00:00:01" -readonly TESTS=("data" "ack" "flags" "tcp" "ip" "large") -readonly PROTOS=("ipv4" "ipv6" "ipip") -dev="" -test="all" -proto="ipv4" - -run_test() { - local server_pid=0 - local exit_code=0 - local protocol=$1 - local test=$2 - local ARGS=( "--${protocol}" "--dmac" "${SERVER_MAC}" \ - "--smac" "${CLIENT_MAC}" "--test" "${test}" "--verbose" ) - - setup_ns - # Each test is run 6 times to deflake, because given the receive timing, - # not all packets that should coalesce will be considered in the same flow - # on every try. - for tries in {1..6}; do - # Actual test starts here - ip netns exec $server_ns ./gro "${ARGS[@]}" "--rx" "--iface" "server" \ - 1>>log.txt & - server_pid=$! - sleep 0.5 # to allow for socket init - ip netns exec $client_ns ./gro "${ARGS[@]}" "--iface" "client" \ - 1>>log.txt - wait "${server_pid}" - exit_code=$? - if [[ ${test} == "large" && -n "${KSFT_MACHINE_SLOW}" && \ - ${exit_code} -ne 0 ]]; then - echo "Ignoring errors due to slow environment" 1>&2 - exit_code=0 - fi - if [[ "${exit_code}" -eq 0 ]]; then - break; - fi - done - cleanup_ns - echo ${exit_code} -} - -run_all_tests() { - local failed_tests=() - for proto in "${PROTOS[@]}"; do - for test in "${TESTS[@]}"; do - echo "running test ${proto} ${test}" >&2 - exit_code=$(run_test $proto $test) - if [[ "${exit_code}" -ne 0 ]]; then - failed_tests+=("${proto}_${test}") - fi; - done; - done - if [[ ${#failed_tests[@]} -ne 0 ]]; then - echo "failed tests: ${failed_tests[*]}. \ - Please see log.txt for more logs" - exit 1 - else - echo "All Tests Succeeded!" - fi; -} - -usage() { - echo "Usage: $0 \ - [-i ] \ - [-t data|ack|flags|tcp|ip|large] \ - [-p ]" 1>&2; - exit 1; -} - -while getopts "i:t:p:" opt; do - case "${opt}" in - i) - dev="${OPTARG}" - ;; - t) - test="${OPTARG}" - ;; - p) - proto="${OPTARG}" - ;; - *) - usage - ;; - esac -done - -if [ -n "$dev" ]; then - source $(dirname $0)/../../net/lib/setup_loopback.sh -else - source $(dirname $0)/../../net/lib/setup_veth.sh -fi - -setup -trap cleanup EXIT -if [[ "${test}" == "all" ]]; then - run_all_tests -else - exit_code=$(run_test "${proto}" "${test}") - exit $exit_code -fi; From 9cf9aa77a1f6339ac97ac4afe44939cf0974ef96 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 19 Nov 2025 18:10:22 -0800 Subject: [PATCH 0975/1381] selftests: drv-net: hw: convert the Toeplitz test to Python Rewrite the existing toeplitz.sh test in Python. The conversion is a lot less exact than the GRO one. We use Netlink APIs to get the device RSS and IRQ information. We expect that the device has neither RPS nor RFS configured, and set RPS up as part of the test. Reviewed-by: Petr Machata Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20251120021024.2944527-11-kuba@kernel.org Signed-off-by: Jakub Kicinski --- .../testing/selftests/drivers/net/hw/Makefile | 3 +- .../selftests/drivers/net/hw/toeplitz.c | 5 + .../selftests/drivers/net/hw/toeplitz.py | 209 ++++++++++++++++++ .../selftests/drivers/net/hw/toeplitz.sh | 199 ----------------- .../drivers/net/hw/toeplitz_client.sh | 28 --- 5 files changed, 215 insertions(+), 229 deletions(-) create mode 100755 tools/testing/selftests/drivers/net/hw/toeplitz.py delete mode 100755 tools/testing/selftests/drivers/net/hw/toeplitz.sh delete mode 100755 tools/testing/selftests/drivers/net/hw/toeplitz_client.sh diff --git a/tools/testing/selftests/drivers/net/hw/Makefile b/tools/testing/selftests/drivers/net/hw/Makefile index c9dced8c934a..1760238e9d4f 100644 --- a/tools/testing/selftests/drivers/net/hw/Makefile +++ b/tools/testing/selftests/drivers/net/hw/Makefile @@ -24,14 +24,13 @@ TEST_PROGS = \ rss_ctx.py \ rss_flow_label.py \ rss_input_xfrm.py \ - toeplitz.sh \ + toeplitz.py \ tso.py \ xsk_reconfig.py \ # TEST_FILES := \ ethtool_lib.sh \ - toeplitz_client.sh \ # TEST_INCLUDES := \ diff --git a/tools/testing/selftests/drivers/net/hw/toeplitz.c b/tools/testing/selftests/drivers/net/hw/toeplitz.c index bf74aa25345d..afc5f910b006 100644 --- a/tools/testing/selftests/drivers/net/hw/toeplitz.c +++ b/tools/testing/selftests/drivers/net/hw/toeplitz.c @@ -53,6 +53,7 @@ #include #include "../../../kselftest.h" +#include "../../../net/lib/ksft.h" #define TOEPLITZ_KEY_MIN_LEN 40 #define TOEPLITZ_KEY_MAX_LEN 60 @@ -576,6 +577,10 @@ int main(int argc, char **argv) fd_sink = setup_sink(); setup_rings(); + + /* Signal to test framework that we're ready to receive */ + ksft_ready(); + process_rings(); cleanup_rings(); diff --git a/tools/testing/selftests/drivers/net/hw/toeplitz.py b/tools/testing/selftests/drivers/net/hw/toeplitz.py new file mode 100755 index 000000000000..9019a8c1ff62 --- /dev/null +++ b/tools/testing/selftests/drivers/net/hw/toeplitz.py @@ -0,0 +1,209 @@ +#!/usr/bin/env python3 +# SPDX-License-Identifier: GPL-2.0 + +""" +Toeplitz Rx hashing test: + - rxhash (the hash value calculation itself); + - RSS mapping from rxhash to rx queue; + - RPS mapping from rxhash to cpu. +""" + +import glob +import os +import socket +from lib.py import ksft_run, ksft_exit, ksft_pr +from lib.py import NetDrvEpEnv, EthtoolFamily, NetdevFamily +from lib.py import cmd, bkg, rand_port, defer +from lib.py import ksft_in +from lib.py import ksft_variants, KsftNamedVariant, KsftSkipEx, KsftFailEx + + +def _check_rps_and_rfs_not_configured(cfg): + """Verify that RPS is not already configured.""" + + for rps_file in glob.glob(f"/sys/class/net/{cfg.ifname}/queues/rx-*/rps_cpus"): + with open(rps_file, "r", encoding="utf-8") as fp: + val = fp.read().strip() + if set(val) - {"0", ","}: + raise KsftSkipEx(f"RPS already configured on {rps_file}: {val}") + + rfs_file = "/proc/sys/net/core/rps_sock_flow_entries" + with open(rfs_file, "r", encoding="utf-8") as fp: + val = fp.read().strip() + if val != "0": + raise KsftSkipEx(f"RFS already configured {rfs_file}: {val}") + + +def _get_rss_key(cfg): + """ + Read the RSS key from the device. + Return a string in the traditional %02x:%02x:%02x:.. format. + """ + + rss = cfg.ethnl.rss_get({"header": {"dev-index": cfg.ifindex}}) + return ':'.join(f'{b:02x}' for b in rss["hkey"]) + + +def _get_cpu_for_irq(irq): + with open(f"/proc/irq/{irq}/smp_affinity_list", "r", + encoding="utf-8") as fp: + data = fp.read().strip() + if "," in data or "-" in data: + raise KsftFailEx(f"IRQ{irq} is not mapped to a single core: {data}") + return int(data) + + +def _get_irq_cpus(cfg): + """ + Read the list of IRQs for the device Rx queues. + """ + queues = cfg.netnl.queue_get({"ifindex": cfg.ifindex}, dump=True) + napis = cfg.netnl.napi_get({"ifindex": cfg.ifindex}, dump=True) + + # Remap into ID-based dicts + napis = {n["id"]: n for n in napis} + queues = {f"{q['type']}{q['id']}": q for q in queues} + + cpus = [] + for rx in range(9999): + name = f"rx{rx}" + if name not in queues: + break + cpus.append(_get_cpu_for_irq(napis[queues[name]["napi-id"]]["irq"])) + + return cpus + + +def _get_unused_cpus(cfg, count=2): + """ + Get CPUs that are not used by Rx queues. + Returns a list of at least 'count' CPU numbers. + """ + + # Get CPUs used by Rx queues + rx_cpus = set(_get_irq_cpus(cfg)) + + # Get total number of CPUs + num_cpus = os.cpu_count() + + # Find unused CPUs + unused_cpus = [cpu for cpu in range(num_cpus) if cpu not in rx_cpus] + + if len(unused_cpus) < count: + raise KsftSkipEx(f"Need at {count} CPUs not used by Rx queues, found {len(unused_cpus)}") + + return unused_cpus[:count] + + +def _configure_rps(cfg, rps_cpus): + """Configure RPS for all Rx queues.""" + + mask = 0 + for cpu in rps_cpus: + mask |= (1 << cpu) + mask = hex(mask)[2:] + + # Set RPS bitmap for all rx queues + for rps_file in glob.glob(f"/sys/class/net/{cfg.ifname}/queues/rx-*/rps_cpus"): + with open(rps_file, "w", encoding="utf-8") as fp: + fp.write(mask) + + return mask + + +def _send_traffic(cfg, proto_flag, ipver, port): + """Send 20 packets of requested type.""" + + # Determine protocol and IP version for socat + if proto_flag == "-u": + proto = "UDP" + else: + proto = "TCP" + + baddr = f"[{cfg.addr_v['6']}]" if ipver == "6" else cfg.addr_v["4"] + + # Run socat in a loop to send traffic periodically + # Use sh -c with a loop similar to toeplitz_client.sh + socat_cmd = f""" + for i in `seq 20`; do + echo "msg $i" | socat -{ipver} -t 0.1 - {proto}:{baddr}:{port}; + sleep 0.001; + done + """ + + cmd(socat_cmd, shell=True, host=cfg.remote) + + +def _test_variants(): + for grp in ["", "rss", "rps"]: + for l4 in ["tcp", "udp"]: + for l3 in ["4", "6"]: + name = f"{l4}_ipv{l3}" + if grp: + name = f"{grp}_{name}" + yield KsftNamedVariant(name, "-" + l4[0], l3, grp) + + +@ksft_variants(_test_variants()) +def test(cfg, proto_flag, ipver, grp): + """Run a single toeplitz test.""" + + cfg.require_ipver(ipver) + + # Check that rxhash is enabled + ksft_in("receive-hashing: on", cmd(f"ethtool -k {cfg.ifname}").stdout) + + port = rand_port(socket.SOCK_DGRAM) + key = _get_rss_key(cfg) + + toeplitz_path = cfg.test_dir / "toeplitz" + rx_cmd = [ + str(toeplitz_path), + "-" + ipver, + proto_flag, + "-d", str(port), + "-i", cfg.ifname, + "-k", key, + "-T", "1000", + "-s", + "-v" + ] + + if grp: + _check_rps_and_rfs_not_configured(cfg) + if grp == "rss": + irq_cpus = ",".join([str(x) for x in _get_irq_cpus(cfg)]) + rx_cmd += ["-C", irq_cpus] + ksft_pr(f"RSS using CPUs: {irq_cpus}") + elif grp == "rps": + # Get CPUs not used by Rx queues and configure them for RPS + rps_cpus = _get_unused_cpus(cfg, count=2) + rps_mask = _configure_rps(cfg, rps_cpus) + defer(_configure_rps, cfg, []) + rx_cmd += ["-r", rps_mask] + ksft_pr(f"RPS using CPUs: {rps_cpus}, mask: {rps_mask}") + + # Run rx in background, it will exit once it has seen enough packets + with bkg(" ".join(rx_cmd), ksft_ready=True, exit_wait=True) as rx_proc: + while rx_proc.proc.poll() is None: + _send_traffic(cfg, proto_flag, ipver, port) + + # Check rx result + ksft_pr("Receiver output:") + ksft_pr(rx_proc.stdout.strip().replace('\n', '\n# ')) + if rx_proc.stderr: + ksft_pr(rx_proc.stderr.strip().replace('\n', '\n# ')) + + +def main() -> None: + """Ksft boilerplate main.""" + + with NetDrvEpEnv(__file__) as cfg: + cfg.ethnl = EthtoolFamily() + cfg.netnl = NetdevFamily() + ksft_run(cases=[test], args=(cfg,)) + ksft_exit() + + +if __name__ == "__main__": + main() diff --git a/tools/testing/selftests/drivers/net/hw/toeplitz.sh b/tools/testing/selftests/drivers/net/hw/toeplitz.sh deleted file mode 100755 index d236b666dd3b..000000000000 --- a/tools/testing/selftests/drivers/net/hw/toeplitz.sh +++ /dev/null @@ -1,199 +0,0 @@ -#!/bin/bash -# SPDX-License-Identifier: GPL-2.0 -# -# extended toeplitz test: test rxhash plus, optionally, either (1) rss mapping -# from rxhash to rx queue ('-rss') or (2) rps mapping from rxhash to cpu -# ('-rps ') -# -# irq-pattern-prefix can be derived from /sys/kernel/irq/*/action, -# which is a driver-specific encoding. -# -# invoke as ./toeplitz.sh (-i ) -u|-t -4|-6 \ -# [(-rss -irq_prefix )|(-rps )] - -source $(dirname $0)/../../../net/lib/setup_loopback.sh -readonly SERVER_IP4="192.168.1.200/24" -readonly SERVER_IP6="fda8::1/64" -readonly SERVER_MAC="aa:00:00:00:00:02" - -readonly CLIENT_IP4="192.168.1.100/24" -readonly CLIENT_IP6="fda8::2/64" -readonly CLIENT_MAC="aa:00:00:00:00:01" - -PORT=8000 -KEY="$( /proc/sys/net/core/rps_sock_flow_entries;" -} - -get_set_rps_bitmaps_cmd() { - CMD="" - for i in /sys/class/net/${DEV}/queues/rx-*/rps_cpus - do - CMD="${CMD} echo $1 > ${i};" - done - - echo "${CMD}" -} - -get_disable_rps_cmd() { - echo "$(get_set_rps_bitmaps_cmd 0)" -} - -die() { - echo "$1" - exit 1 -} - -check_nic_rxhash_enabled() { - local -r pattern="receive-hashing:\ on" - - ethtool -k "${DEV}" | grep -q "${pattern}" || die "rxhash must be enabled" -} - -parse_opts() { - local prog=$0 - shift 1 - - while [[ "$1" =~ "-" ]]; do - if [[ "$1" = "-irq_prefix" ]]; then - shift - IRQ_PATTERN="^$1-[0-9]*$" - elif [[ "$1" = "-u" || "$1" = "-t" ]]; then - PROTO_FLAG="$1" - elif [[ "$1" = "-4" ]]; then - IP_FLAG="$1" - SERVER_IP="${SERVER_IP4}" - CLIENT_IP="${CLIENT_IP4}" - elif [[ "$1" = "-6" ]]; then - IP_FLAG="$1" - SERVER_IP="${SERVER_IP6}" - CLIENT_IP="${CLIENT_IP6}" - elif [[ "$1" = "-rss" ]]; then - TEST_RSS=true - elif [[ "$1" = "-rps" ]]; then - shift - RPS_MAP="$1" - elif [[ "$1" = "-i" ]]; then - shift - DEV="$1" - else - die "Usage: ${prog} (-i ) -u|-t -4|-6 \ - [(-rss -irq_prefix )|(-rps )]" - fi - shift - done -} - -setup() { - setup_loopback_environment "${DEV}" - - # Set up server_ns namespace and client_ns namespace - setup_macvlan_ns "${DEV}" $server_ns server \ - "${SERVER_MAC}" "${SERVER_IP}" - setup_macvlan_ns "${DEV}" $client_ns client \ - "${CLIENT_MAC}" "${CLIENT_IP}" -} - -cleanup() { - cleanup_macvlan_ns $server_ns server $client_ns client - cleanup_loopback "${DEV}" -} - -parse_opts $0 $@ - -setup -trap cleanup EXIT - -check_nic_rxhash_enabled - -# Actual test starts here -if [[ "${TEST_RSS}" = true ]]; then - # RPS/RFS must be disabled because they move packets between cpus, - # which breaks the PACKET_FANOUT_CPU identification of RSS decisions. - eval "$(get_disable_rfs_cmd) $(get_disable_rps_cmd)" \ - ip netns exec $server_ns ./toeplitz "${IP_FLAG}" "${PROTO_FLAG}" \ - -d "${PORT}" -i "${DEV}" -k "${KEY}" -T 1000 \ - -C "$(get_rx_irq_cpus)" -s -v & -elif [[ ! -z "${RPS_MAP}" ]]; then - eval "$(get_disable_rfs_cmd) $(get_set_rps_bitmaps_cmd ${RPS_MAP})" \ - ip netns exec $server_ns ./toeplitz "${IP_FLAG}" "${PROTO_FLAG}" \ - -d "${PORT}" -i "${DEV}" -k "${KEY}" -T 1000 \ - -r "0x${RPS_MAP}" -s -v & -else - ip netns exec $server_ns ./toeplitz "${IP_FLAG}" "${PROTO_FLAG}" \ - -d "${PORT}" -i "${DEV}" -k "${KEY}" -T 1000 -s -v & -fi - -server_pid=$! - -ip netns exec $client_ns ./toeplitz_client.sh "${PROTO_FLAG}" \ - "${IP_FLAG}" "${SERVER_IP%%/*}" "${PORT}" & - -client_pid=$! - -wait "${server_pid}" -exit_code=$? -kill -9 "${client_pid}" -if [[ "${exit_code}" -eq 0 ]]; then - echo "Test Succeeded!" -fi -exit "${exit_code}" diff --git a/tools/testing/selftests/drivers/net/hw/toeplitz_client.sh b/tools/testing/selftests/drivers/net/hw/toeplitz_client.sh deleted file mode 100755 index 2fef34f4aba1..000000000000 --- a/tools/testing/selftests/drivers/net/hw/toeplitz_client.sh +++ /dev/null @@ -1,28 +0,0 @@ -#!/bin/bash -# SPDX-License-Identifier: GPL-2.0 -# -# A simple program for generating traffic for the toeplitz test. -# -# This program sends packets periodically for, conservatively, 20 seconds. The -# intent is for the calling program to kill this program once it is no longer -# needed, rather than waiting for the 20 second expiration. - -send_traffic() { - expiration=$((SECONDS+20)) - while [[ "${SECONDS}" -lt "${expiration}" ]] - do - if [[ "${PROTO}" == "-u" ]]; then - echo "msg $i" | nc "${IPVER}" -u -w 0 "${ADDR}" "${PORT}" - else - echo "msg $i" | nc "${IPVER}" -w 0 "${ADDR}" "${PORT}" - fi - sleep 0.001 - done -} - -PROTO=$1 -IPVER=$2 -ADDR=$3 -PORT=$4 - -send_traffic From 358008f41d9b893f0c843cb8b941954e0acbf8ec Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 19 Nov 2025 18:10:23 -0800 Subject: [PATCH 0976/1381] netdevsim: add loopback support Support device loopback. Apparently this mode has been historically supported by the toeplitz test and I don't have any HW which lets me test the conversion.. Reviewed-by: Petr Machata Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20251120021024.2944527-12-kuba@kernel.org Signed-off-by: Jakub Kicinski --- drivers/net/netdevsim/netdev.c | 17 ++++++++++++----- 1 file changed, 12 insertions(+), 5 deletions(-) diff --git a/drivers/net/netdevsim/netdev.c b/drivers/net/netdevsim/netdev.c index 2b713db16cd0..6927c1962277 100644 --- a/drivers/net/netdevsim/netdev.c +++ b/drivers/net/netdevsim/netdev.c @@ -133,15 +133,21 @@ static netdev_tx_t nsim_start_xmit(struct sk_buff *skb, struct net_device *dev) if (!nsim_ipsec_tx(ns, skb)) goto out_drop_any; - peer_ns = rcu_dereference(ns->peer); - if (!peer_ns) - goto out_drop_any; + /* Check if loopback mode is enabled */ + if (dev->features & NETIF_F_LOOPBACK) { + peer_ns = ns; + peer_dev = dev; + } else { + peer_ns = rcu_dereference(ns->peer); + if (!peer_ns) + goto out_drop_any; + peer_dev = peer_ns->netdev; + } dr = nsim_do_psp(skb, ns, peer_ns, &psp_ext); if (dr) goto out_drop_free; - peer_dev = peer_ns->netdev; rxq = skb_get_queue_mapping(skb); if (rxq >= peer_dev->num_rx_queues) rxq = rxq % peer_dev->num_rx_queues; @@ -976,7 +982,8 @@ static void nsim_setup(struct net_device *dev) NETIF_F_FRAGLIST | NETIF_F_HW_CSUM | NETIF_F_LRO | - NETIF_F_TSO; + NETIF_F_TSO | + NETIF_F_LOOPBACK; dev->pcpu_stat_type = NETDEV_PCPU_STAT_DSTATS; dev->max_mtu = ETH_MAX_MTU; dev->xdp_features = NETDEV_XDP_ACT_BASIC | NETDEV_XDP_ACT_HW_OFFLOAD; From bd28e5bddc1a75089fd1c45bd58892d9399097c4 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 19 Nov 2025 18:10:24 -0800 Subject: [PATCH 0977/1381] selftests: net: remove old setup_* scripts gro.sh and toeplitz.sh used to source in one of two setup scripts depending on whether the test was expected to be run against veth or a real device. veth testing is replaced by netdevsim and existing "remote endpoint" support in our Python tests. Add a script which sets up loopback mode. The usage is a little bit more complicated than running the scripts used to be. Testing used to work like this: ./../gro.sh -i eth0 ... now the "setup script" has to be run explicitly: NETIF=eth0 ./../ksft_setup_loopback.sh ./../gro.sh But the functionality itself is retained. Reviewed-by: Petr Machata Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20251120021024.2944527-13-kuba@kernel.org Signed-off-by: Jakub Kicinski --- tools/testing/selftests/net/lib/Makefile | 3 +- .../selftests/net/lib/ksft_setup_loopback.sh | 111 ++++++++++++++++ .../selftests/net/lib/setup_loopback.sh | 120 ------------------ tools/testing/selftests/net/lib/setup_veth.sh | 45 ------- 4 files changed, 112 insertions(+), 167 deletions(-) create mode 100755 tools/testing/selftests/net/lib/ksft_setup_loopback.sh delete mode 100644 tools/testing/selftests/net/lib/setup_loopback.sh delete mode 100644 tools/testing/selftests/net/lib/setup_veth.sh diff --git a/tools/testing/selftests/net/lib/Makefile b/tools/testing/selftests/net/lib/Makefile index c10796933d42..5339f56329e1 100644 --- a/tools/testing/selftests/net/lib/Makefile +++ b/tools/testing/selftests/net/lib/Makefile @@ -8,8 +8,7 @@ CFLAGS += -I../../ TEST_FILES := \ ../../../../net/ynl \ ../../../../../Documentation/netlink/specs \ - setup_loopback.sh \ - setup_veth.sh \ + ksft_setup_loopback.sh \ # end of TEST_FILES TEST_GEN_FILES := \ diff --git a/tools/testing/selftests/net/lib/ksft_setup_loopback.sh b/tools/testing/selftests/net/lib/ksft_setup_loopback.sh new file mode 100755 index 000000000000..3defbb1919c5 --- /dev/null +++ b/tools/testing/selftests/net/lib/ksft_setup_loopback.sh @@ -0,0 +1,111 @@ +#!/bin/bash +# SPDX-License-Identifier: GPL-2.0 + +# Setup script for running ksft tests over a real interface in loopback mode. +# This scripts replaces the historical setup_loopback.sh. It puts +# a (presumably) real hardware interface into loopback mode, creates macvlan +# interfaces on top and places them in a network namespace for isolation. +# +# NETIF env variable must be exported to indicate the real target device. +# Note that the test will override NETIF with one of the macvlans, the +# actual ksft test will only see the macvlans. +# +# Example use: +# export NETIF=eth0 +# ./net/lib/ksft_setup_loopback.sh ./drivers/net/gro.py + +if [ -z "$NETIF" ]; then + echo "Error: NETIF variable not set" + exit 1 +fi +if ! [ -d "/sys/class/net/$NETIF" ]; then + echo "Error: Can't find $NETIF, invalid netdevice" + exit 1 +fi + +# Save original settings for cleanup +readonly FLUSH_PATH="/sys/class/net/${NETIF}/gro_flush_timeout" +readonly IRQ_PATH="/sys/class/net/${NETIF}/napi_defer_hard_irqs" +FLUSH_TIMEOUT="$(< "${FLUSH_PATH}")" +readonly FLUSH_TIMEOUT +HARD_IRQS="$(< "${IRQ_PATH}")" +readonly HARD_IRQS + +SERVER_NS=$(mktemp -u server-XXXXXXXX) +readonly SERVER_NS +CLIENT_NS=$(mktemp -u client-XXXXXXXX) +readonly CLIENT_NS +readonly SERVER_MAC="aa:00:00:00:00:02" +readonly CLIENT_MAC="aa:00:00:00:00:01" + +# ksft expects addresses to communicate with remote +export LOCAL_V6=2001:db8:1::1 +export REMOTE_V6=2001:db8:1::2 + +cleanup() { + local exit_code=$? + + echo "Cleaning up..." + + # Remove macvlan interfaces and namespaces + ip -netns "${SERVER_NS}" link del dev server 2>/dev/null || true + ip netns del "${SERVER_NS}" 2>/dev/null || true + ip -netns "${CLIENT_NS}" link del dev client 2>/dev/null || true + ip netns del "${CLIENT_NS}" 2>/dev/null || true + + # Disable loopback + ethtool -K "${NETIF}" loopback off 2>/dev/null || true + sleep 1 + + echo "${FLUSH_TIMEOUT}" >"${FLUSH_PATH}" + echo "${HARD_IRQS}" >"${IRQ_PATH}" + + exit $exit_code +} + +trap cleanup EXIT INT TERM + +# Enable loopback mode +echo "Enabling loopback on ${NETIF}..." +ethtool -K "${NETIF}" loopback on || { + echo "Failed to enable loopback mode" + exit 1 +} +# The interface may need time to get carrier back, but selftests +# will wait for carrier, so no need to wait / sleep here. + +# Use timer on host to trigger the network stack +# Also disable device interrupt to not depend on NIC interrupt +# Reduce test flakiness caused by unexpected interrupts +echo 100000 >"${FLUSH_PATH}" +echo 50 >"${IRQ_PATH}" + +# Create server namespace with macvlan +ip netns add "${SERVER_NS}" +ip link add link "${NETIF}" dev server address "${SERVER_MAC}" type macvlan +ip link set dev server netns "${SERVER_NS}" +ip -netns "${SERVER_NS}" link set dev server up +ip -netns "${SERVER_NS}" addr add $LOCAL_V6/64 dev server +ip -netns "${SERVER_NS}" link set dev lo up + +# Create client namespace with macvlan +ip netns add "${CLIENT_NS}" +ip link add link "${NETIF}" dev client address "${CLIENT_MAC}" type macvlan +ip link set dev client netns "${CLIENT_NS}" +ip -netns "${CLIENT_NS}" link set dev client up +ip -netns "${CLIENT_NS}" addr add $REMOTE_V6/64 dev client +ip -netns "${CLIENT_NS}" link set dev lo up + +echo "Setup complete!" +echo " Device: ${NETIF}" +echo " Server NS: ${SERVER_NS}" +echo " Client NS: ${CLIENT_NS}" +echo "" + +# Setup environment variables for tests +export NETIF=server +export REMOTE_TYPE=netns +export REMOTE_ARGS="${CLIENT_NS}" + +# Run the command +ip netns exec "${SERVER_NS}" "$@" diff --git a/tools/testing/selftests/net/lib/setup_loopback.sh b/tools/testing/selftests/net/lib/setup_loopback.sh deleted file mode 100644 index 2070b57849de..000000000000 --- a/tools/testing/selftests/net/lib/setup_loopback.sh +++ /dev/null @@ -1,120 +0,0 @@ -#!/bin/bash -# SPDX-License-Identifier: GPL-2.0 - -readonly FLUSH_PATH="/sys/class/net/${dev}/gro_flush_timeout" -readonly IRQ_PATH="/sys/class/net/${dev}/napi_defer_hard_irqs" -readonly FLUSH_TIMEOUT="$(< ${FLUSH_PATH})" -readonly HARD_IRQS="$(< ${IRQ_PATH})" -readonly server_ns=$(mktemp -u server-XXXXXXXX) -readonly client_ns=$(mktemp -u client-XXXXXXXX) - -netdev_check_for_carrier() { - local -r dev="$1" - - for i in {1..5}; do - carrier="$(cat /sys/class/net/${dev}/carrier)" - if [[ "${carrier}" -ne 1 ]] ; then - echo "carrier not ready yet..." >&2 - sleep 1 - else - echo "carrier ready" >&2 - break - fi - done - echo "${carrier}" -} - -# Assumes that there is no existing ipvlan device on the physical device -setup_loopback_environment() { - local dev="$1" - - # Fail hard if cannot turn on loopback mode for current NIC - ethtool -K "${dev}" loopback on || exit 1 - sleep 1 - - # Check for the carrier - carrier=$(netdev_check_for_carrier ${dev}) - if [[ "${carrier}" -ne 1 ]] ; then - echo "setup_loopback_environment failed" - exit 1 - fi -} - -setup_macvlan_ns(){ - local -r link_dev="$1" - local -r ns_name="$2" - local -r ns_dev="$3" - local -r ns_mac="$4" - local -r addr="$5" - - ip link add link "${link_dev}" dev "${ns_dev}" \ - address "${ns_mac}" type macvlan - exit_code=$? - if [[ "${exit_code}" -ne 0 ]]; then - echo "setup_macvlan_ns failed" - exit $exit_code - fi - - [[ -e /var/run/netns/"${ns_name}" ]] || ip netns add "${ns_name}" - ip link set dev "${ns_dev}" netns "${ns_name}" - ip -netns "${ns_name}" link set dev "${ns_dev}" up - if [[ -n "${addr}" ]]; then - ip -netns "${ns_name}" addr add dev "${ns_dev}" "${addr}" - fi - - sleep 1 -} - -cleanup_macvlan_ns(){ - while (( $# >= 2 )); do - ns_name="$1" - ns_dev="$2" - ip -netns "${ns_name}" link del dev "${ns_dev}" - ip netns del "${ns_name}" - shift 2 - done -} - -cleanup_loopback(){ - local -r dev="$1" - - ethtool -K "${dev}" loopback off - sleep 1 - - # Check for the carrier - carrier=$(netdev_check_for_carrier ${dev}) - if [[ "${carrier}" -ne 1 ]] ; then - echo "setup_loopback_environment failed" - exit 1 - fi -} - -setup_interrupt() { - # Use timer on host to trigger the network stack - # Also disable device interrupt to not depend on NIC interrupt - # Reduce test flakiness caused by unexpected interrupts - echo 100000 >"${FLUSH_PATH}" - echo 50 >"${IRQ_PATH}" -} - -setup_ns() { - # Set up server_ns namespace and client_ns namespace - setup_macvlan_ns "${dev}" ${server_ns} server "${SERVER_MAC}" - setup_macvlan_ns "${dev}" ${client_ns} client "${CLIENT_MAC}" -} - -cleanup_ns() { - cleanup_macvlan_ns ${server_ns} server ${client_ns} client -} - -setup() { - setup_loopback_environment "${dev}" - setup_interrupt -} - -cleanup() { - cleanup_loopback "${dev}" - - echo "${FLUSH_TIMEOUT}" >"${FLUSH_PATH}" - echo "${HARD_IRQS}" >"${IRQ_PATH}" -} diff --git a/tools/testing/selftests/net/lib/setup_veth.sh b/tools/testing/selftests/net/lib/setup_veth.sh deleted file mode 100644 index 152bf4c65747..000000000000 --- a/tools/testing/selftests/net/lib/setup_veth.sh +++ /dev/null @@ -1,45 +0,0 @@ -#!/bin/bash -# SPDX-License-Identifier: GPL-2.0 - -readonly server_ns=$(mktemp -u server-XXXXXXXX) -readonly client_ns=$(mktemp -u client-XXXXXXXX) - -setup_veth_ns() { - local -r link_dev="$1" - local -r ns_name="$2" - local -r ns_dev="$3" - local -r ns_mac="$4" - - [[ -e /var/run/netns/"${ns_name}" ]] || ip netns add "${ns_name}" - echo 200000 > "/sys/class/net/${ns_dev}/gro_flush_timeout" - echo 1 > "/sys/class/net/${ns_dev}/napi_defer_hard_irqs" - ip link set dev "${ns_dev}" netns "${ns_name}" mtu 65535 - ip -netns "${ns_name}" link set dev "${ns_dev}" up - - ip netns exec "${ns_name}" ethtool -K "${ns_dev}" gro on tso off -} - -setup_ns() { - # Set up server_ns namespace and client_ns namespace - ip link add name server type veth peer name client - - setup_veth_ns "${dev}" ${server_ns} server "${SERVER_MAC}" - setup_veth_ns "${dev}" ${client_ns} client "${CLIENT_MAC}" -} - -cleanup_ns() { - local ns_name - - for ns_name in ${client_ns} ${server_ns}; do - [[ -e /var/run/netns/"${ns_name}" ]] && ip netns del "${ns_name}" - done -} - -setup() { - # no global init setup step needed - : -} - -cleanup() { - cleanup_ns -} From 8b4e023d79b760d217dd1c462848c4a27fcc7677 Mon Sep 17 00:00:00 2001 From: Zahari Doychev Date: Wed, 19 Nov 2025 21:36:18 +0100 Subject: [PATCH 0978/1381] ynl: samples: add tc filter example Add a sample tool demonstrating how to add, dump, and delete a flower filter with two VLAN push actions. The example can be invoked as: # samples/tc-filter-add p2 flower pref 1 proto: 0x8100 flower: vlan_id: 100 vlan_prio: 5 num_of_vlans: 3 action order: 1 vlan push id 200 protocol 0x8100 priority 0 action order: 2 vlan push id 300 protocol 0x8100 priority 0 This verifies correct handling of tc action attributes for multiple VLAN push actions. The tc action indexed arrays start from index 1, and the index defines the action order. This behavior differs from the YNL specification, which expects arrays to be zero-based. To accommodate this, the example adds a dummy action at index 0, which is ignored by the kernel. Signed-off-by: Zahari Doychev Link: https://patch.msgid.link/20251119203618.263780-2-zahari.doychev@linux.com Signed-off-by: Jakub Kicinski --- tools/net/ynl/samples/.gitignore | 1 + tools/net/ynl/samples/Makefile | 1 + tools/net/ynl/samples/tc-filter-add.c | 335 ++++++++++++++++++++++++++ 3 files changed, 337 insertions(+) create mode 100644 tools/net/ynl/samples/tc-filter-add.c diff --git a/tools/net/ynl/samples/.gitignore b/tools/net/ynl/samples/.gitignore index 7f5fca7682d7..05087ee323ba 100644 --- a/tools/net/ynl/samples/.gitignore +++ b/tools/net/ynl/samples/.gitignore @@ -7,3 +7,4 @@ rt-addr rt-link rt-route tc +tc-filter-add diff --git a/tools/net/ynl/samples/Makefile b/tools/net/ynl/samples/Makefile index c9494a564da4..d76cbd41cbb1 100644 --- a/tools/net/ynl/samples/Makefile +++ b/tools/net/ynl/samples/Makefile @@ -19,6 +19,7 @@ include $(wildcard *.d) all: $(BINS) CFLAGS_page-pool=$(CFLAGS_netdev) +CFLAGS_tc-filter-add:=$(CFLAGS_tc) $(BINS): ../lib/ynl.a ../generated/protos.a $(SRCS) @echo -e '\tCC sample $@' diff --git a/tools/net/ynl/samples/tc-filter-add.c b/tools/net/ynl/samples/tc-filter-add.c new file mode 100644 index 000000000000..1f9cd3f62df6 --- /dev/null +++ b/tools/net/ynl/samples/tc-filter-add.c @@ -0,0 +1,335 @@ +// SPDX-License-Identifier: GPL-2.0 +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include + +#include "tc-user.h" + +#define TC_HANDLE (0xFFFF << 16) + +const char *vlan_act_name(struct tc_vlan *p) +{ + switch (p->v_action) { + case TCA_VLAN_ACT_POP: + return "pop"; + case TCA_VLAN_ACT_PUSH: + return "push"; + case TCA_VLAN_ACT_MODIFY: + return "modify"; + default: + break; + } + + return "not supported"; +} + +const char *gact_act_name(struct tc_gact *p) +{ + switch (p->action) { + case TC_ACT_SHOT: + return "drop"; + case TC_ACT_OK: + return "ok"; + case TC_ACT_PIPE: + return "pipe"; + default: + break; + } + + return "not supported"; +} + +static void print_vlan(struct tc_act_vlan_attrs *vlan) +{ + printf("%s ", vlan_act_name(vlan->parms)); + if (vlan->_present.push_vlan_id) + printf("id %u ", vlan->push_vlan_id); + if (vlan->_present.push_vlan_protocol) + printf("protocol %#x ", ntohs(vlan->push_vlan_protocol)); + if (vlan->_present.push_vlan_priority) + printf("priority %u ", vlan->push_vlan_priority); +} + +static void print_gact(struct tc_act_gact_attrs *gact) +{ + struct tc_gact *p = gact->parms; + + printf("%s ", gact_act_name(p)); +} + +static void flower_print(struct tc_flower_attrs *flower, const char *kind) +{ + struct tc_act_attrs *a; + unsigned int i; + + printf("%s:\n", kind); + + if (flower->_present.key_vlan_id) + printf(" vlan_id: %u\n", flower->key_vlan_id); + if (flower->_present.key_vlan_prio) + printf(" vlan_prio: %u\n", flower->key_vlan_prio); + if (flower->_present.key_num_of_vlans) + printf(" num_of_vlans: %u\n", flower->key_num_of_vlans); + + for (i = 0; i < flower->_count.act; i++) { + a = &flower->act[i]; + printf("action order: %i %s ", i + 1, a->kind); + if (a->options._present.vlan) + print_vlan(&a->options.vlan); + else if (a->options._present.gact) + print_gact(&a->options.gact); + printf("\n"); + } + printf("\n"); +} + +static void tc_filter_print(struct tc_gettfilter_rsp *f) +{ + struct tc_options_msg *opt = &f->options; + + if (opt->_present.flower) + flower_print(&opt->flower, f->kind); + else if (f->_len.kind) + printf("%s pref %u proto: %#x\n", f->kind, + (f->_hdr.tcm_info >> 16), + ntohs(TC_H_MIN(f->_hdr.tcm_info))); +} + +static int tc_filter_add(struct ynl_sock *ys, int ifi) +{ + struct tc_newtfilter_req *req; + struct tc_act_attrs *acts; + struct tc_vlan p = { + .action = TC_ACT_PIPE, + .v_action = TCA_VLAN_ACT_PUSH + }; + __u16 flags = NLM_F_REQUEST | NLM_F_EXCL | NLM_F_CREATE; + int ret; + + req = tc_newtfilter_req_alloc(); + if (!req) { + fprintf(stderr, "tc_newtfilter_req_alloc failed\n"); + return -1; + } + memset(req, 0, sizeof(*req)); + + acts = tc_act_attrs_alloc(3); + if (!acts) { + fprintf(stderr, "tc_act_attrs_alloc\n"); + tc_newtfilter_req_free(req); + return -1; + } + memset(acts, 0, sizeof(*acts) * 3); + + req->_hdr.tcm_ifindex = ifi; + req->_hdr.tcm_parent = TC_H_MAKE(TC_H_CLSACT, TC_H_MIN_INGRESS); + req->_hdr.tcm_info = TC_H_MAKE(1 << 16, htons(ETH_P_8021Q)); + req->chain = 0; + + tc_newtfilter_req_set_nlflags(req, flags); + tc_newtfilter_req_set_kind(req, "flower"); + tc_newtfilter_req_set_options_flower_key_vlan_id(req, 100); + tc_newtfilter_req_set_options_flower_key_vlan_prio(req, 5); + tc_newtfilter_req_set_options_flower_key_num_of_vlans(req, 3); + + __tc_newtfilter_req_set_options_flower_act(req, acts, 3); + + /* Skip action at index 0 because in TC, the action array + * index starts at 1, with each index defining the action's + * order. In contrast, in YNL indexed arrays start at index 0. + */ + tc_act_attrs_set_kind(&acts[1], "vlan"); + tc_act_attrs_set_options_vlan_parms(&acts[1], &p, sizeof(p)); + tc_act_attrs_set_options_vlan_push_vlan_id(&acts[1], 200); + tc_act_attrs_set_kind(&acts[2], "vlan"); + tc_act_attrs_set_options_vlan_parms(&acts[2], &p, sizeof(p)); + tc_act_attrs_set_options_vlan_push_vlan_id(&acts[2], 300); + + tc_newtfilter_req_set_options_flower_flags(req, 0); + tc_newtfilter_req_set_options_flower_key_eth_type(req, htons(0x8100)); + + ret = tc_newtfilter(ys, req); + if (ret) + fprintf(stderr, "tc_newtfilter: %s\n", ys->err.msg); + + tc_newtfilter_req_free(req); + + return ret; +} + +static int tc_filter_show(struct ynl_sock *ys, int ifi) +{ + struct tc_gettfilter_req_dump *req; + struct tc_gettfilter_list *rsp; + + req = tc_gettfilter_req_dump_alloc(); + if (!req) { + fprintf(stderr, "tc_gettfilter_req_dump_alloc failed\n"); + return -1; + } + memset(req, 0, sizeof(*req)); + + req->_hdr.tcm_ifindex = ifi; + req->_hdr.tcm_parent = TC_H_MAKE(TC_H_CLSACT, TC_H_MIN_INGRESS); + req->_present.chain = 1; + req->chain = 0; + + rsp = tc_gettfilter_dump(ys, req); + tc_gettfilter_req_dump_free(req); + if (!rsp) { + fprintf(stderr, "YNL: %s\n", ys->err.msg); + return -1; + } + + if (ynl_dump_empty(rsp)) + fprintf(stderr, "Error: no filters reported\n"); + else + ynl_dump_foreach(rsp, flt) tc_filter_print(flt); + + tc_gettfilter_list_free(rsp); + + return 0; +} + +static int tc_filter_del(struct ynl_sock *ys, int ifi) +{ + struct tc_deltfilter_req *req; + __u16 flags = NLM_F_REQUEST; + int ret; + + req = tc_deltfilter_req_alloc(); + if (!req) { + fprintf(stderr, "tc_deltfilter_req_alloc failedq\n"); + return -1; + } + memset(req, 0, sizeof(*req)); + + req->_hdr.tcm_ifindex = ifi; + req->_hdr.tcm_parent = TC_H_MAKE(TC_H_CLSACT, TC_H_MIN_INGRESS); + req->_hdr.tcm_info = TC_H_MAKE(1 << 16, htons(ETH_P_8021Q)); + tc_deltfilter_req_set_nlflags(req, flags); + + ret = tc_deltfilter(ys, req); + if (ret) + fprintf(stderr, "tc_deltfilter failed: %s\n", ys->err.msg); + + tc_deltfilter_req_free(req); + + return ret; +} + +static int tc_clsact_add(struct ynl_sock *ys, int ifi) +{ + struct tc_newqdisc_req *req; + __u16 flags = NLM_F_REQUEST | NLM_F_EXCL | NLM_F_CREATE; + int ret; + + req = tc_newqdisc_req_alloc(); + if (!req) { + fprintf(stderr, "tc_newqdisc_req_alloc failed\n"); + return -1; + } + memset(req, 0, sizeof(*req)); + + req->_hdr.tcm_ifindex = ifi; + req->_hdr.tcm_parent = TC_H_CLSACT; + req->_hdr.tcm_handle = TC_HANDLE; + tc_newqdisc_req_set_nlflags(req, flags); + tc_newqdisc_req_set_kind(req, "clsact"); + + ret = tc_newqdisc(ys, req); + if (ret) + fprintf(stderr, "tc_newqdisc failed: %s\n", ys->err.msg); + + tc_newqdisc_req_free(req); + + return ret; +} + +static int tc_clsact_del(struct ynl_sock *ys, int ifi) +{ + struct tc_delqdisc_req *req; + __u16 flags = NLM_F_REQUEST; + int ret; + + req = tc_delqdisc_req_alloc(); + if (!req) { + fprintf(stderr, "tc_delqdisc_req_alloc failed\n"); + return -1; + } + memset(req, 0, sizeof(*req)); + + req->_hdr.tcm_ifindex = ifi; + req->_hdr.tcm_parent = TC_H_CLSACT; + req->_hdr.tcm_handle = TC_HANDLE; + tc_delqdisc_req_set_nlflags(req, flags); + + ret = tc_delqdisc(ys, req); + if (ret) + fprintf(stderr, "tc_delqdisc failed: %s\n", ys->err.msg); + + tc_delqdisc_req_free(req); + + return ret; +} + +static int tc_filter_config(struct ynl_sock *ys, int ifi) +{ + int ret = 0; + + if (tc_filter_add(ys, ifi)) + return -1; + + ret = tc_filter_show(ys, ifi); + + if (tc_filter_del(ys, ifi)) + return -1; + + return ret; +} + +int main(int argc, char **argv) +{ + struct ynl_error yerr; + struct ynl_sock *ys; + int ifi, ret = 0; + + if (argc < 2) { + fprintf(stderr, "Usage: %s \n", argv[0]); + return 1; + } + ifi = if_nametoindex(argv[1]); + if (!ifi) { + perror("if_nametoindex"); + return 1; + } + + ys = ynl_sock_create(&ynl_tc_family, &yerr); + if (!ys) { + fprintf(stderr, "YNL: %s\n", yerr.msg); + return 1; + } + + if (tc_clsact_add(ys, ifi)) { + ret = 2; + goto err_destroy; + } + + if (tc_filter_config(ys, ifi)) + ret = 3; + + if (tc_clsact_del(ys, ifi)) + ret = 4; + +err_destroy: + ynl_sock_destroy(ys); + return ret; +} From 491c5dc98b848c4781addd514caed95039e5366c Mon Sep 17 00:00:00 2001 From: Yael Chemla Date: Wed, 19 Nov 2025 22:48:15 +0200 Subject: [PATCH 0979/1381] net: ethtool: Add support for 1600Gbps speed Add support for 1600Gbps link modes based on 200Gbps per lane [1]. This includes the adopted IEEE 802.3dj copper and optical PMDs that use 200G/lane signaling [2]. Add the following PMD types: - KR8 (backplane) - CR8 (copper cable) - DR8 (SMF 500m) - DR8-2 (SMF 2km) These modes are defined in the 802.3dj specifications. References: [1] https://www.ieee802.org/3/dj/public/23_03/opsasnick_3dj_01a_2303.pdf [2] https://www.ieee802.org/3/dj/projdoc/objectives_P802d3dj_240314.pdf Signed-off-by: Yael Chemla Reviewed-by: Shahar Shitrit Signed-off-by: Tariq Toukan Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/1763585297-1243980-2-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- drivers/net/phy/phy-caps.h | 1 + drivers/net/phy/phy-core.c | 4 +++- drivers/net/phy/phy_caps.c | 2 ++ include/uapi/linux/ethtool.h | 5 +++++ net/ethtool/common.c | 8 ++++++++ 5 files changed, 19 insertions(+), 1 deletion(-) diff --git a/drivers/net/phy/phy-caps.h b/drivers/net/phy/phy-caps.h index b7f0c6a3037a..4951a39f3828 100644 --- a/drivers/net/phy/phy-caps.h +++ b/drivers/net/phy/phy-caps.h @@ -29,6 +29,7 @@ enum { LINK_CAPA_200000FD, LINK_CAPA_400000FD, LINK_CAPA_800000FD, + LINK_CAPA_1600000FD, __LINK_CAPA_MAX, }; diff --git a/drivers/net/phy/phy-core.c b/drivers/net/phy/phy-core.c index 0c63e6ba2cb0..277c034bc32f 100644 --- a/drivers/net/phy/phy-core.c +++ b/drivers/net/phy/phy-core.c @@ -17,7 +17,7 @@ */ const char *phy_speed_to_str(int speed) { - BUILD_BUG_ON_MSG(__ETHTOOL_LINK_MODE_MASK_NBITS != 121, + BUILD_BUG_ON_MSG(__ETHTOOL_LINK_MODE_MASK_NBITS != 125, "Enum ethtool_link_mode_bit_indices and phylib are out of sync. " "If a speed or mode has been added please update phy_speed_to_str " "and the PHY settings array.\n"); @@ -55,6 +55,8 @@ const char *phy_speed_to_str(int speed) return "400Gbps"; case SPEED_800000: return "800Gbps"; + case SPEED_1600000: + return "1600Gbps"; case SPEED_UNKNOWN: return "Unknown"; default: diff --git a/drivers/net/phy/phy_caps.c b/drivers/net/phy/phy_caps.c index 23c808b59b6f..3a05982b39bf 100644 --- a/drivers/net/phy/phy_caps.c +++ b/drivers/net/phy/phy_caps.c @@ -25,6 +25,7 @@ static struct link_capabilities link_caps[__LINK_CAPA_MAX] __ro_after_init = { { SPEED_200000, DUPLEX_FULL, {0} }, /* LINK_CAPA_200000FD */ { SPEED_400000, DUPLEX_FULL, {0} }, /* LINK_CAPA_400000FD */ { SPEED_800000, DUPLEX_FULL, {0} }, /* LINK_CAPA_800000FD */ + { SPEED_1600000, DUPLEX_FULL, {0} }, /* LINK_CAPA_1600000FD */ }; static int speed_duplex_to_capa(int speed, unsigned int duplex) @@ -52,6 +53,7 @@ static int speed_duplex_to_capa(int speed, unsigned int duplex) case SPEED_200000: return LINK_CAPA_200000FD; case SPEED_400000: return LINK_CAPA_400000FD; case SPEED_800000: return LINK_CAPA_800000FD; + case SPEED_1600000: return LINK_CAPA_1600000FD; } return -EINVAL; diff --git a/include/uapi/linux/ethtool.h b/include/uapi/linux/ethtool.h index 8bd5ea5469d9..eb7ff2602fbb 100644 --- a/include/uapi/linux/ethtool.h +++ b/include/uapi/linux/ethtool.h @@ -2077,6 +2077,10 @@ enum ethtool_link_mode_bit_indices { ETHTOOL_LINK_MODE_800000baseDR4_2_Full_BIT = 118, ETHTOOL_LINK_MODE_800000baseSR4_Full_BIT = 119, ETHTOOL_LINK_MODE_800000baseVR4_Full_BIT = 120, + ETHTOOL_LINK_MODE_1600000baseCR8_Full_BIT = 121, + ETHTOOL_LINK_MODE_1600000baseKR8_Full_BIT = 122, + ETHTOOL_LINK_MODE_1600000baseDR8_Full_BIT = 123, + ETHTOOL_LINK_MODE_1600000baseDR8_2_Full_BIT = 124, /* must be last entry */ __ETHTOOL_LINK_MODE_MASK_NBITS @@ -2190,6 +2194,7 @@ enum ethtool_link_mode_bit_indices { #define SPEED_200000 200000 #define SPEED_400000 400000 #define SPEED_800000 800000 +#define SPEED_1600000 1600000 #define SPEED_UNKNOWN -1 diff --git a/net/ethtool/common.c b/net/ethtool/common.c index 55223ebc2a7e..369c05cf8163 100644 --- a/net/ethtool/common.c +++ b/net/ethtool/common.c @@ -233,6 +233,10 @@ const char link_mode_names[][ETH_GSTRING_LEN] = { __DEFINE_LINK_MODE_NAME(800000, DR4_2, Full), __DEFINE_LINK_MODE_NAME(800000, SR4, Full), __DEFINE_LINK_MODE_NAME(800000, VR4, Full), + __DEFINE_LINK_MODE_NAME(1600000, CR8, Full), + __DEFINE_LINK_MODE_NAME(1600000, KR8, Full), + __DEFINE_LINK_MODE_NAME(1600000, DR8, Full), + __DEFINE_LINK_MODE_NAME(1600000, DR8_2, Full), }; static_assert(ARRAY_SIZE(link_mode_names) == __ETHTOOL_LINK_MODE_MASK_NBITS); @@ -422,6 +426,10 @@ const struct link_mode_info link_mode_params[] = { __DEFINE_LINK_MODE_PARAMS(800000, DR4_2, Full), __DEFINE_LINK_MODE_PARAMS(800000, SR4, Full), __DEFINE_LINK_MODE_PARAMS(800000, VR4, Full), + __DEFINE_LINK_MODE_PARAMS(1600000, CR8, Full), + __DEFINE_LINK_MODE_PARAMS(1600000, KR8, Full), + __DEFINE_LINK_MODE_PARAMS(1600000, DR8, Full), + __DEFINE_LINK_MODE_PARAMS(1600000, DR8_2, Full), }; static_assert(ARRAY_SIZE(link_mode_params) == __ETHTOOL_LINK_MODE_MASK_NBITS); EXPORT_SYMBOL_GPL(link_mode_params); From be3a435df74b3e84037fcc23ade68e01c4ed541f Mon Sep 17 00:00:00 2001 From: Yael Chemla Date: Wed, 19 Nov 2025 22:48:16 +0200 Subject: [PATCH 0980/1381] net/mlx5e: Add 1600Gbps link modes Introduce support for a 1600Gbps link mode, utilizing 8 lanes at 200Gbps per lane. Signed-off-by: Yael Chemla Reviewed-by: Shahar Shitrit Reviewed-by: Leon Romanovsky Signed-off-by: Tariq Toukan Link: https://patch.msgid.link/1763585297-1243980-3-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c | 5 +++++ drivers/net/ethernet/mellanox/mlx5/core/port.c | 1 + 2 files changed, 6 insertions(+) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c index 727d7a833110..fe67c73849f9 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -261,6 +261,11 @@ void mlx5e_build_ptys2ethtool_map(void) ETHTOOL_LINK_MODE_800000baseDR4_2_Full_BIT, ETHTOOL_LINK_MODE_800000baseSR4_Full_BIT, ETHTOOL_LINK_MODE_800000baseVR4_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_1600TAUI_8_1600TBASE_CR8_KR8, ext, + ETHTOOL_LINK_MODE_1600000baseCR8_Full_BIT, + ETHTOOL_LINK_MODE_1600000baseKR8_Full_BIT, + ETHTOOL_LINK_MODE_1600000baseDR8_Full_BIT, + ETHTOOL_LINK_MODE_1600000baseDR8_2_Full_BIT); } static void mlx5e_ethtool_get_speed_arr(bool ext, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/port.c b/drivers/net/ethernet/mellanox/mlx5/core/port.c index e4b1dfafb41f..85a9e534f442 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/port.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/port.c @@ -1108,6 +1108,7 @@ mlx5e_ext_link_info[MLX5E_EXT_LINK_MODES_NUMBER] = { [MLX5E_200GAUI_1_200GBASE_CR1_KR1] = {.speed = 200000, .lanes = 1}, [MLX5E_400GAUI_2_400GBASE_CR2_KR2] = {.speed = 400000, .lanes = 2}, [MLX5E_800GAUI_4_800GBASE_CR4_KR4] = {.speed = 800000, .lanes = 4}, + [MLX5E_1600TAUI_8_1600TBASE_CR8_KR8] = {.speed = 1600000, .lanes = 8}, }; int mlx5_port_query_eth_proto(struct mlx5_core_dev *dev, u8 port, bool ext, From 5fb9a0b89e2ab959547cd684f497bd31bcb3261f Mon Sep 17 00:00:00 2001 From: Yael Chemla Date: Wed, 19 Nov 2025 22:48:17 +0200 Subject: [PATCH 0981/1381] bonding: 3ad: Add support for 1600G speed Add support for 1600Gbps speed to allow using 3ad mode with 1600G devices. Signed-off-by: Yael Chemla Reviewed-by: Shahar Shitrit Signed-off-by: Tariq Toukan Link: https://patch.msgid.link/1763585297-1243980-4-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- drivers/net/bonding/bond_3ad.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/drivers/net/bonding/bond_3ad.c b/drivers/net/bonding/bond_3ad.c index 49717b7b82a2..1a8de2bf8655 100644 --- a/drivers/net/bonding/bond_3ad.c +++ b/drivers/net/bonding/bond_3ad.c @@ -76,6 +76,7 @@ enum ad_link_speed_type { AD_LINK_SPEED_200000MBPS, AD_LINK_SPEED_400000MBPS, AD_LINK_SPEED_800000MBPS, + AD_LINK_SPEED_1600000MBPS, }; /* compare MAC addresses */ @@ -300,6 +301,7 @@ static inline int __check_agg_selection_timer(struct port *port) * %AD_LINK_SPEED_200000MBPS * %AD_LINK_SPEED_400000MBPS * %AD_LINK_SPEED_800000MBPS + * %AD_LINK_SPEED_1600000MBPS */ static u16 __get_link_speed(struct port *port) { @@ -379,6 +381,10 @@ static u16 __get_link_speed(struct port *port) speed = AD_LINK_SPEED_800000MBPS; break; + case SPEED_1600000: + speed = AD_LINK_SPEED_1600000MBPS; + break; + default: /* unknown speed value from ethtool. shouldn't happen */ if (slave->speed != SPEED_UNKNOWN) @@ -822,6 +828,9 @@ static u32 __get_agg_bandwidth(struct aggregator *aggregator) case AD_LINK_SPEED_800000MBPS: bandwidth = nports * 800000; break; + case AD_LINK_SPEED_1600000MBPS: + bandwidth = nports * 1600000; + break; default: bandwidth = 0; /* to silence the compiler */ } From 3eea593b5597e580def9658aa6397e1057f27acd Mon Sep 17 00:00:00 2001 From: Wei Fang Date: Wed, 19 Nov 2025 10:51:44 +0800 Subject: [PATCH 0982/1381] net: fec: remove useless conditional preprocessor directives The conditional preprocessor directive was added to fix build errors on the MCF5272 platform, see commit d13919301d9a ("net: fec: Fix build for MCF5272"). The compilation errors were originally caused by some register macros not being defined on that platform. The driver now uses quirks to dynamically handle platform differences, and for MCF5272, its quirks is 0, so it does not support RACC and GBIT Ethernet. So these preprocessor directives are no longer required and can be safely removed without causing build or functional issue. Signed-off-by: Wei Fang Link: https://patch.msgid.link/20251119025148.2817602-2-wei.fang@nxp.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/freescale/fec_main.c | 6 ------ 1 file changed, 6 deletions(-) diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c index b6fbb84cfb06..c2a307c6e774 100644 --- a/drivers/net/ethernet/freescale/fec_main.c +++ b/drivers/net/ethernet/freescale/fec_main.c @@ -1773,7 +1773,6 @@ fec_enet_rx_queue(struct net_device *ndev, u16 queue_id, int budget) __fec32 cbd_bufaddr; u32 sub_len = 4; -#if !defined(CONFIG_M5272) /*If it has the FEC_QUIRK_HAS_RACC quirk property, the bit of * FEC_RACC_SHIFT16 is set by default in the probe function. */ @@ -1781,7 +1780,6 @@ fec_enet_rx_queue(struct net_device *ndev, u16 queue_id, int budget) data_start += 2; sub_len += 2; } -#endif #if defined(CONFIG_COLDFIRE) && !defined(CONFIG_COLDFIRE_COHERENT_DMA) /* @@ -2517,9 +2515,7 @@ static int fec_enet_mii_probe(struct net_device *ndev) phy_set_max_speed(phy_dev, 1000); phy_remove_link_mode(phy_dev, ETHTOOL_LINK_MODE_1000baseT_Half_BIT); -#if !defined(CONFIG_M5272) phy_support_sym_pause(phy_dev); -#endif } else phy_set_max_speed(phy_dev, 100); @@ -4402,11 +4398,9 @@ fec_probe(struct platform_device *pdev) fep->num_rx_queues = num_rx_qs; fep->num_tx_queues = num_tx_qs; -#if !defined(CONFIG_M5272) /* default enable pause frame auto negotiation */ if (fep->quirks & FEC_QUIRK_HAS_GBIT) fep->pause_flag |= FEC_PAUSE_FLAG_AUTONEG; -#endif /* Select default pin state */ pinctrl_pm_select_default_state(&pdev->dev); From eef7b786bdab146047d0372eff9f6bd16cffe415 Mon Sep 17 00:00:00 2001 From: Wei Fang Date: Wed, 19 Nov 2025 10:51:45 +0800 Subject: [PATCH 0983/1381] net: fec: simplify the conditional preprocessor directives From the Kconfig file, we can see CONFIG_FEC depends on the following platform-related options. ColdFire: M523x, M527x, M5272, M528x, M520x and M532x S32: ARCH_S32 (ARM64) i.MX: SOC_IMX28 and ARCH_MXC (ARM and ARM64) Based on the code of fec driver, only some macro definitions on the M5272 platform are different from those on other platforms. Therefore, we can simplify the following complex preprocessor directives to "if !defined(CONFIG_M5272)". "#if defined(CONFIG_M523x) || defined(CONFIG_M527x) || \ defined(CONFIG_M528x) || defined(CONFIG_M520x) || \ defined(CONFIG_M532x) || defined(CONFIG_ARM) || \ defined(CONFIG_ARM64)" Signed-off-by: Wei Fang Link: https://patch.msgid.link/20251119025148.2817602-3-wei.fang@nxp.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/freescale/fec.h | 4 +--- drivers/net/ethernet/freescale/fec_main.c | 26 ++++++----------------- 2 files changed, 8 insertions(+), 22 deletions(-) diff --git a/drivers/net/ethernet/freescale/fec.h b/drivers/net/ethernet/freescale/fec.h index 41e0d85d15da..8e438f6e7ec4 100644 --- a/drivers/net/ethernet/freescale/fec.h +++ b/drivers/net/ethernet/freescale/fec.h @@ -24,9 +24,7 @@ #include #include -#if defined(CONFIG_M523x) || defined(CONFIG_M527x) || defined(CONFIG_M528x) || \ - defined(CONFIG_M520x) || defined(CONFIG_M532x) || defined(CONFIG_ARM) || \ - defined(CONFIG_ARM64) || defined(CONFIG_COMPILE_TEST) +#if !defined(CONFIG_M5272) || defined(CONFIG_COMPILE_TEST) /* * Just figures, Motorola would have to change the offsets for * registers in the same peripheral device on different models diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c index c2a307c6e774..4193559c6b9c 100644 --- a/drivers/net/ethernet/freescale/fec_main.c +++ b/drivers/net/ethernet/freescale/fec_main.c @@ -253,9 +253,7 @@ MODULE_PARM_DESC(macaddr, "FEC Ethernet MAC address"); * size bits. Other FEC hardware does not, so we need to take that into * account when setting it. */ -#if defined(CONFIG_M523x) || defined(CONFIG_M527x) || defined(CONFIG_M528x) || \ - defined(CONFIG_M520x) || defined(CONFIG_M532x) || defined(CONFIG_ARM) || \ - defined(CONFIG_ARM64) +#ifndef CONFIG_M5272 #define OPT_ARCH_HAS_MAX_FL 1 #else #define OPT_ARCH_HAS_MAX_FL 0 @@ -2706,9 +2704,7 @@ static int fec_enet_get_regs_len(struct net_device *ndev) } /* List of registers that can be safety be read to dump them with ethtool */ -#if defined(CONFIG_M523x) || defined(CONFIG_M527x) || defined(CONFIG_M528x) || \ - defined(CONFIG_M520x) || defined(CONFIG_M532x) || defined(CONFIG_ARM) || \ - defined(CONFIG_ARM64) || defined(CONFIG_COMPILE_TEST) +#if !defined(CONFIG_M5272) || defined(CONFIG_COMPILE_TEST) static __u32 fec_enet_register_version = 2; static u32 fec_enet_register_offset[] = { FEC_IEVENT, FEC_IMASK, FEC_R_DES_ACTIVE_0, FEC_X_DES_ACTIVE_0, @@ -2782,30 +2778,22 @@ static u32 fec_enet_register_offset[] = { static void fec_enet_get_regs(struct net_device *ndev, struct ethtool_regs *regs, void *regbuf) { + u32 reg_cnt = ARRAY_SIZE(fec_enet_register_offset); struct fec_enet_private *fep = netdev_priv(ndev); u32 __iomem *theregs = (u32 __iomem *)fep->hwp; + u32 *reg_list = fec_enet_register_offset; struct device *dev = &fep->pdev->dev; u32 *buf = (u32 *)regbuf; u32 i, off; int ret; -#if defined(CONFIG_M523x) || defined(CONFIG_M527x) || defined(CONFIG_M528x) || \ - defined(CONFIG_M520x) || defined(CONFIG_M532x) || defined(CONFIG_ARM) || \ - defined(CONFIG_ARM64) || defined(CONFIG_COMPILE_TEST) - u32 *reg_list; - u32 reg_cnt; - if (!of_machine_is_compatible("fsl,imx6ul")) { - reg_list = fec_enet_register_offset; - reg_cnt = ARRAY_SIZE(fec_enet_register_offset); - } else { +#if !defined(CONFIG_M5272) || defined(CONFIG_COMPILE_TEST) + if (of_machine_is_compatible("fsl,imx6ul")) { reg_list = fec_enet_register_offset_6ul; reg_cnt = ARRAY_SIZE(fec_enet_register_offset_6ul); } -#else - /* coldfire */ - static u32 *reg_list = fec_enet_register_offset; - static const u32 reg_cnt = ARRAY_SIZE(fec_enet_register_offset); #endif + ret = pm_runtime_resume_and_get(dev); if (ret < 0) return; From 63083d597adabffd12ecb879dbae3556e87ee277 Mon Sep 17 00:00:00 2001 From: Wei Fang Date: Wed, 19 Nov 2025 10:51:46 +0800 Subject: [PATCH 0984/1381] net: fec: remove struct fec_enet_priv_txrx_info The struct fec_enet_priv_txrx_info has three members: offset, page and skb. The offset is only initialized in the driver and is not used, the skb is never initialized and used in the driver. The both will not be used in the future. Therefore, replace struct fec_enet_priv_txrx_info directly with struct page. Signed-off-by: Wei Fang Reviewed-by: Frank Li Link: https://patch.msgid.link/20251119025148.2817602-4-wei.fang@nxp.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/freescale/fec.h | 8 +------- drivers/net/ethernet/freescale/fec_main.c | 11 +++++------ 2 files changed, 6 insertions(+), 13 deletions(-) diff --git a/drivers/net/ethernet/freescale/fec.h b/drivers/net/ethernet/freescale/fec.h index 8e438f6e7ec4..c5bbc2c16a4f 100644 --- a/drivers/net/ethernet/freescale/fec.h +++ b/drivers/net/ethernet/freescale/fec.h @@ -528,12 +528,6 @@ struct bufdesc_prop { unsigned char dsize_log2; }; -struct fec_enet_priv_txrx_info { - int offset; - struct page *page; - struct sk_buff *skb; -}; - enum { RX_XDP_REDIRECT = 0, RX_XDP_PASS, @@ -573,7 +567,7 @@ struct fec_enet_priv_tx_q { struct fec_enet_priv_rx_q { struct bufdesc_prop bd; - struct fec_enet_priv_txrx_info rx_skb_info[RX_RING_SIZE]; + struct page *rx_buf[RX_RING_SIZE]; /* page_pool */ struct page_pool *page_pool; diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c index 4193559c6b9c..6c19be0618ae 100644 --- a/drivers/net/ethernet/freescale/fec_main.c +++ b/drivers/net/ethernet/freescale/fec_main.c @@ -1655,8 +1655,7 @@ static int fec_enet_update_cbd(struct fec_enet_priv_rx_q *rxq, if (unlikely(!new_page)) return -ENOMEM; - rxq->rx_skb_info[index].page = new_page; - rxq->rx_skb_info[index].offset = FEC_ENET_XDP_HEADROOM; + rxq->rx_buf[index] = new_page; phys_addr = page_pool_get_dma_addr(new_page) + FEC_ENET_XDP_HEADROOM; bdp->cbd_bufaddr = cpu_to_fec32(phys_addr); @@ -1836,7 +1835,7 @@ fec_enet_rx_queue(struct net_device *ndev, u16 queue_id, int budget) ndev->stats.rx_bytes -= 2; index = fec_enet_get_bd_index(bdp, &rxq->bd); - page = rxq->rx_skb_info[index].page; + page = rxq->rx_buf[index]; cbd_bufaddr = bdp->cbd_bufaddr; if (fec_enet_update_cbd(rxq, bdp, index)) { ndev->stats.rx_dropped++; @@ -3312,7 +3311,8 @@ static void fec_enet_free_buffers(struct net_device *ndev) for (q = 0; q < fep->num_rx_queues; q++) { rxq = fep->rx_queue[q]; for (i = 0; i < rxq->bd.ring_size; i++) - page_pool_put_full_page(rxq->page_pool, rxq->rx_skb_info[i].page, false); + page_pool_put_full_page(rxq->page_pool, rxq->rx_buf[i], + false); for (i = 0; i < XDP_STATS_TOTAL; i++) rxq->stats[i] = 0; @@ -3446,8 +3446,7 @@ fec_enet_alloc_rxq_buffers(struct net_device *ndev, unsigned int queue) phys_addr = page_pool_get_dma_addr(page) + FEC_ENET_XDP_HEADROOM; bdp->cbd_bufaddr = cpu_to_fec32(phys_addr); - rxq->rx_skb_info[i].page = page; - rxq->rx_skb_info[i].offset = FEC_ENET_XDP_HEADROOM; + rxq->rx_buf[i] = page; bdp->cbd_sc = cpu_to_fec16(BD_ENET_RX_EMPTY); if (fep->bufdesc_ex) { From 3bb06c8a461b8c8714eef5094b1fadd2d066dd1a Mon Sep 17 00:00:00 2001 From: Wei Fang Date: Wed, 19 Nov 2025 10:51:47 +0800 Subject: [PATCH 0985/1381] net: fec: remove rx_align from fec_enet_private The rx_align was introduced by the commit 41ef84ce4c72 ("net: fec: change FEC alignment according to i.mx6 sx requirement"). Because the i.MX6 SX requires RX buffer must be 64 bytes alignment. Since the commit 95698ff6177b ("net: fec: using page pool to manage RX buffers"), the address of the RX buffer is always the page address plus FEC_ENET_XDP_HEADROOM which is 256 bytes, so the RX buffer is always 64-byte aligned. Therefore, rx_align has no effect since that commit, and we can safely remove it. In addition, to prevent future modifications to FEC_ENET_XDP_HEADROOM, a BUILD_BUG_ON() test has been added to the driver, which ensures that FEC_ENET_XDP_HEADROOM provides the required alignment. Signed-off-by: Wei Fang Link: https://patch.msgid.link/20251119025148.2817602-5-wei.fang@nxp.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/freescale/fec.h | 1 - drivers/net/ethernet/freescale/fec_main.c | 19 ++++++++++++++----- 2 files changed, 14 insertions(+), 6 deletions(-) diff --git a/drivers/net/ethernet/freescale/fec.h b/drivers/net/ethernet/freescale/fec.h index c5bbc2c16a4f..a25dca9c7d71 100644 --- a/drivers/net/ethernet/freescale/fec.h +++ b/drivers/net/ethernet/freescale/fec.h @@ -660,7 +660,6 @@ struct fec_enet_private { struct pm_qos_request pm_qos_req; unsigned int tx_align; - unsigned int rx_align; /* hw interrupt coalesce */ unsigned int rx_pkts_itr; diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c index 6c19be0618ae..c82be43b19ab 100644 --- a/drivers/net/ethernet/freescale/fec_main.c +++ b/drivers/net/ethernet/freescale/fec_main.c @@ -3438,6 +3438,19 @@ fec_enet_alloc_rxq_buffers(struct net_device *ndev, unsigned int queue) return err; } + /* Some platforms require the RX buffer must be 64 bytes alignment. + * Some platforms require 16 bytes alignment. And some platforms + * require 4 bytes alignment. But since the page pool have been + * introduced into the driver, the address of RX buffer is always + * the page address plus FEC_ENET_XDP_HEADROOM, and + * FEC_ENET_XDP_HEADROOM is 256 bytes. Therefore, this address can + * satisfy all platforms. To prevent future modifications to + * FEC_ENET_XDP_HEADROOM from ignoring this hardware limitation, a + * BUILD_BUG_ON() test has been added, which ensures that + * FEC_ENET_XDP_HEADROOM provides the required alignment. + */ + BUILD_BUG_ON(FEC_ENET_XDP_HEADROOM & 0x3f); + for (i = 0; i < rxq->bd.ring_size; i++) { page = page_pool_dev_alloc_pages(rxq->page_pool); if (!page) @@ -4072,10 +4085,8 @@ static int fec_enet_init(struct net_device *ndev) WARN_ON(dsize != (1 << dsize_log2)); #if defined(CONFIG_ARM) || defined(CONFIG_ARM64) - fep->rx_align = 0xf; fep->tx_align = 0xf; #else - fep->rx_align = 0x3; fep->tx_align = 0x3; #endif fep->rx_pkts_itr = FEC_ITR_ICFT_DEFAULT; @@ -4164,10 +4175,8 @@ static int fec_enet_init(struct net_device *ndev) fep->csum_flags |= FLAG_RX_CSUM_ENABLED; } - if (fep->quirks & FEC_QUIRK_HAS_MULTI_QUEUES) { + if (fep->quirks & FEC_QUIRK_HAS_MULTI_QUEUES) fep->tx_align = 0; - fep->rx_align = 0x3f; - } ndev->hw_features = ndev->features; From bd31490718b47d91b0926619635ba967505457e7 Mon Sep 17 00:00:00 2001 From: Wei Fang Date: Wed, 19 Nov 2025 10:51:48 +0800 Subject: [PATCH 0986/1381] net: fec: remove duplicate macros of the BD status There are two sets of macros used to define the status bits of TX and RX BDs, one is the BD_SC_xx macros, the other one is the BD_ENET_xx macros. For the BD_SC_xx macros, only BD_SC_WRAP is used in the driver. But the BD_ENET_xx macros are more widely used in the driver, and they define more bits of the BD status. Therefore, remove the BD_SC_xx macros from now on. Signed-off-by: Wei Fang Reviewed-by: Frank Li Link: https://patch.msgid.link/20251119025148.2817602-6-wei.fang@nxp.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/freescale/fec.h | 17 ----------------- drivers/net/ethernet/freescale/fec_main.c | 8 ++++---- 2 files changed, 4 insertions(+), 21 deletions(-) diff --git a/drivers/net/ethernet/freescale/fec.h b/drivers/net/ethernet/freescale/fec.h index a25dca9c7d71..7b4d1fc8e7eb 100644 --- a/drivers/net/ethernet/freescale/fec.h +++ b/drivers/net/ethernet/freescale/fec.h @@ -240,23 +240,6 @@ struct bufdesc_ex { __fec16 res0[4]; }; -/* - * The following definitions courtesy of commproc.h, which where - * Copyright (c) 1997 Dan Malek (dmalek@jlc.net). - */ -#define BD_SC_EMPTY ((ushort)0x8000) /* Receive is empty */ -#define BD_SC_READY ((ushort)0x8000) /* Transmit is ready */ -#define BD_SC_WRAP ((ushort)0x2000) /* Last buffer descriptor */ -#define BD_SC_INTRPT ((ushort)0x1000) /* Interrupt on change */ -#define BD_SC_CM ((ushort)0x0200) /* Continuous mode */ -#define BD_SC_ID ((ushort)0x0100) /* Rec'd too many idles */ -#define BD_SC_P ((ushort)0x0100) /* xmt preamble */ -#define BD_SC_BR ((ushort)0x0020) /* Break received */ -#define BD_SC_FR ((ushort)0x0010) /* Framing error */ -#define BD_SC_PR ((ushort)0x0008) /* Parity error */ -#define BD_SC_OV ((ushort)0x0002) /* Overrun */ -#define BD_SC_CD ((ushort)0x0001) /* ?? */ - /* Buffer descriptor control/status used by Ethernet receive. */ #define BD_ENET_RX_EMPTY ((ushort)0x8000) diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c index c82be43b19ab..c685a5c0cc51 100644 --- a/drivers/net/ethernet/freescale/fec_main.c +++ b/drivers/net/ethernet/freescale/fec_main.c @@ -1010,7 +1010,7 @@ static void fec_enet_bd_init(struct net_device *dev) /* Set the last buffer to wrap */ bdp = fec_enet_get_prevdesc(bdp, &rxq->bd); - bdp->cbd_sc |= cpu_to_fec16(BD_SC_WRAP); + bdp->cbd_sc |= cpu_to_fec16(BD_ENET_RX_WRAP); rxq->bd.cur = rxq->bd.base; } @@ -1060,7 +1060,7 @@ static void fec_enet_bd_init(struct net_device *dev) /* Set the last buffer to wrap */ bdp = fec_enet_get_prevdesc(bdp, &txq->bd); - bdp->cbd_sc |= cpu_to_fec16(BD_SC_WRAP); + bdp->cbd_sc |= cpu_to_fec16(BD_ENET_TX_WRAP); txq->dirty_tx = bdp; } } @@ -3472,7 +3472,7 @@ fec_enet_alloc_rxq_buffers(struct net_device *ndev, unsigned int queue) /* Set the last buffer to wrap. */ bdp = fec_enet_get_prevdesc(bdp, &rxq->bd); - bdp->cbd_sc |= cpu_to_fec16(BD_SC_WRAP); + bdp->cbd_sc |= cpu_to_fec16(BD_ENET_RX_WRAP); return 0; err_alloc: @@ -3508,7 +3508,7 @@ fec_enet_alloc_txq_buffers(struct net_device *ndev, unsigned int queue) /* Set the last buffer to wrap. */ bdp = fec_enet_get_prevdesc(bdp, &txq->bd); - bdp->cbd_sc |= cpu_to_fec16(BD_SC_WRAP); + bdp->cbd_sc |= cpu_to_fec16(BD_ENET_TX_WRAP); return 0; From 920fa394dcdaf4aa9c89433d07c3c17fd17dee25 Mon Sep 17 00:00:00 2001 From: Byungchul Park Date: Thu, 20 Nov 2025 10:11:18 +0900 Subject: [PATCH 0987/1381] eth: fbnic: access @pp through netmem_desc instead of page To eliminate the use of struct page in page pool, the page pool users should use netmem descriptor and APIs instead. Make fbnic access @pp through netmem_desc instead of page. Signed-off-by: Byungchul Park Link: https://patch.msgid.link/20251120011118.73253-1-byungchul@sk.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/meta/fbnic/fbnic_txrx.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_txrx.c b/drivers/net/ethernet/meta/fbnic/fbnic_txrx.c index c2d7b67fec28..56744e3a14ec 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_txrx.c +++ b/drivers/net/ethernet/meta/fbnic/fbnic_txrx.c @@ -653,7 +653,8 @@ static void fbnic_clean_twq1(struct fbnic_napi_vector *nv, bool pp_allow_direct, FBNIC_TWD_TYPE_AL; total_bytes += FIELD_GET(FBNIC_TWD_LEN_MASK, twd); - page_pool_put_page(page->pp, page, -1, pp_allow_direct); + page_pool_put_page(pp_page_to_nmdesc(page)->pp, page, -1, + pp_allow_direct); next_desc: head++; head &= ring->size_mask; From 7279b718b431701137f9943434229915222a70b8 Mon Sep 17 00:00:00 2001 From: Gustavo Luiz Duarte Date: Wed, 19 Nov 2025 16:14:49 -0800 Subject: [PATCH 0988/1381] netconsole: Simplify send_fragmented_body() Refactor send_fragmented_body() to use separate offset tracking for msgbody, and extradata instead of complex conditional logic. The previous implementation used boolean flags and calculated offsets which made the code harder to follow. The new implementation maintains independent offset counters (msgbody_offset, extradata_offset) and processes each section sequentially, making the data flow more straightforward and the code easier to maintain. This is a preparatory refactoring with no functional changes, which will allow easily splitting extradata_complete into separate userdata and sysdata buffers in the next patch. Signed-off-by: Gustavo Luiz Duarte Reviewed-by: Breno Leitao Link: https://patch.msgid.link/20251119-netconsole_dynamic_extradata-v3-1-497ac3191707@meta.com Signed-off-by: Jakub Kicinski --- drivers/net/netconsole.c | 109 ++++++++++++++++----------------------- 1 file changed, 45 insertions(+), 64 deletions(-) diff --git a/drivers/net/netconsole.c b/drivers/net/netconsole.c index bb6e03a92956..5fe5896d6ff5 100644 --- a/drivers/net/netconsole.c +++ b/drivers/net/netconsole.c @@ -1560,89 +1560,70 @@ static void append_release(char *buf) } static void send_fragmented_body(struct netconsole_target *nt, - const char *msgbody, int header_len, + const char *msgbody_ptr, int header_len, int msgbody_len, int extradata_len) { - int sent_extradata, preceding_bytes; - const char *extradata = NULL; - int body_len, offset = 0; + const char *extradata_ptr = NULL; + int data_len, data_sent = 0; + int extradata_offset = 0; + int msgbody_offset = 0; #ifdef CONFIG_NETCONSOLE_DYNAMIC - extradata = nt->extradata_complete; + extradata_ptr = nt->extradata_complete; #endif + if (WARN_ON_ONCE(!extradata_ptr && extradata_len != 0)) + return; - /* body_len represents the number of bytes that will be sent. This is + /* data_len represents the number of bytes that will be sent. This is * bigger than MAX_PRINT_CHUNK, thus, it will be split in multiple * packets */ - body_len = msgbody_len + extradata_len; + data_len = msgbody_len + extradata_len; /* In each iteration of the while loop below, we send a packet - * containing the header and a portion of the body. The body is + * containing the header and a portion of the data. The data is * composed of two parts: msgbody and extradata. We keep track of how - * many bytes have been sent so far using the offset variable, which - * ranges from 0 to the total length of the body. + * many bytes have been sent so far using the data_sent variable, which + * ranges from 0 to the total bytes to be sent. */ - while (offset < body_len) { - int this_header = header_len; - bool msgbody_written = false; - int this_offset = 0; + while (data_sent < data_len) { + int extradata_left = extradata_len - extradata_offset; + int msgbody_left = msgbody_len - msgbody_offset; + int buf_offset = 0; int this_chunk = 0; - this_header += scnprintf(nt->buf + this_header, - MAX_PRINT_CHUNK - this_header, - ",ncfrag=%d/%d;", offset, - body_len); + /* header is already populated in nt->buf, just append to it */ + buf_offset = header_len; - /* Not all msgbody data has been written yet */ - if (offset < msgbody_len) { - this_chunk = min(msgbody_len - offset, - MAX_PRINT_CHUNK - this_header); - if (WARN_ON_ONCE(this_chunk <= 0)) - return; - memcpy(nt->buf + this_header, msgbody + offset, - this_chunk); - this_offset += this_chunk; + buf_offset += scnprintf(nt->buf + buf_offset, + MAX_PRINT_CHUNK - buf_offset, + ",ncfrag=%d/%d;", data_sent, + data_len); + + /* append msgbody first */ + this_chunk = min(msgbody_left, MAX_PRINT_CHUNK - buf_offset); + memcpy(nt->buf + buf_offset, msgbody_ptr + msgbody_offset, + this_chunk); + msgbody_offset += this_chunk; + buf_offset += this_chunk; + data_sent += this_chunk; + + /* after msgbody, append extradata */ + if (extradata_ptr && extradata_left) { + this_chunk = min(extradata_left, + MAX_PRINT_CHUNK - buf_offset); + memcpy(nt->buf + buf_offset, + extradata_ptr + extradata_offset, this_chunk); + extradata_offset += this_chunk; + buf_offset += this_chunk; + data_sent += this_chunk; } - /* msgbody was finally written, either in the previous - * messages and/or in the current buf. Time to write - * the extradata. - */ - msgbody_written |= offset + this_offset >= msgbody_len; + /* if all is good, send the packet out */ + if (WARN_ON_ONCE(data_sent > data_len)) + return; - /* Msg body is fully written and there is pending extradata to - * write, append extradata in this chunk - */ - if (msgbody_written && offset + this_offset < body_len) { - /* Track how much user data was already sent. First - * time here, sent_userdata is zero - */ - sent_extradata = (offset + this_offset) - msgbody_len; - /* offset of bytes used in current buf */ - preceding_bytes = this_chunk + this_header; - - if (WARN_ON_ONCE(sent_extradata < 0)) - return; - - this_chunk = min(extradata_len - sent_extradata, - MAX_PRINT_CHUNK - preceding_bytes); - if (WARN_ON_ONCE(this_chunk < 0)) - /* this_chunk could be zero if all the previous - * message used all the buffer. This is not a - * problem, extradata will be sent in the next - * iteration - */ - return; - - memcpy(nt->buf + this_header + this_offset, - extradata + sent_extradata, - this_chunk); - this_offset += this_chunk; - } - - send_udp(nt, nt->buf, this_header + this_offset); - offset += this_offset; + send_udp(nt, nt->buf, buf_offset); } } From 9dc10f50c430d861d4f3d5d0b38efd675f6d76ae Mon Sep 17 00:00:00 2001 From: Gustavo Luiz Duarte Date: Wed, 19 Nov 2025 16:14:50 -0800 Subject: [PATCH 0989/1381] netconsole: Split userdata and sysdata Separate userdata and sysdata into distinct buffers to enable independent management. Previously, both were stored in a single extradata_complete buffer with a fixed size that accommodated both types of data. This separation allows: - userdata to grow dynamically (in subsequent patch) - sysdata to remain in a small static buffer - removal of complex entry counting logic that tracked both types together The split also simplifies the code by eliminating the need to check total entry count across both userdata and sysdata when enabling features, which allows to drop holding su_mutex on sysdata_*_enabled_store(). No functional change in this patch, just structural preparation for dynamic userdata allocation. Signed-off-by: Gustavo Luiz Duarte Reviewed-by: Breno Leitao Link: https://patch.msgid.link/20251119-netconsole_dynamic_extradata-v3-2-497ac3191707@meta.com Signed-off-by: Jakub Kicinski --- drivers/net/netconsole.c | 215 ++++++++++++++++++--------------------- 1 file changed, 98 insertions(+), 117 deletions(-) diff --git a/drivers/net/netconsole.c b/drivers/net/netconsole.c index 5fe5896d6ff5..1bd811714322 100644 --- a/drivers/net/netconsole.c +++ b/drivers/net/netconsole.c @@ -50,7 +50,7 @@ MODULE_LICENSE("GPL"); /* The number 3 comes from userdata entry format characters (' ', '=', '\n') */ #define MAX_EXTRADATA_NAME_LEN (MAX_EXTRADATA_ENTRY_LEN - \ MAX_EXTRADATA_VALUE_LEN - 3) -#define MAX_EXTRADATA_ITEMS 16 +#define MAX_USERDATA_ITEMS 16 #define MAX_PRINT_CHUNK 1000 static char config[MAX_PARAM_LENGTH]; @@ -115,6 +115,8 @@ enum sysdata_feature { SYSDATA_RELEASE = BIT(2), /* Include a per-target message ID as part of sysdata */ SYSDATA_MSGID = BIT(3), + /* Sentinel: highest bit position */ + MAX_SYSDATA_ITEMS = 4, }; /** @@ -122,8 +124,9 @@ enum sysdata_feature { * @list: Links this target into the target_list. * @group: Links us into the configfs subsystem hierarchy. * @userdata_group: Links to the userdata configfs hierarchy - * @extradata_complete: Cached, formatted string of append - * @userdata_length: String length of usedata in extradata_complete. + * @userdata: Cached, formatted string of append + * @userdata_length: String length of userdata. + * @sysdata: Cached, formatted string of append * @sysdata_fields: Sysdata features enabled. * @msgcounter: Message sent counter. * @stats: Packet send stats for the target. Used for debugging. @@ -152,8 +155,10 @@ struct netconsole_target { #ifdef CONFIG_NETCONSOLE_DYNAMIC struct config_group group; struct config_group userdata_group; - char extradata_complete[MAX_EXTRADATA_ENTRY_LEN * MAX_EXTRADATA_ITEMS]; + char userdata[MAX_EXTRADATA_ENTRY_LEN * MAX_USERDATA_ITEMS]; size_t userdata_length; + char sysdata[MAX_EXTRADATA_ENTRY_LEN * MAX_SYSDATA_ITEMS]; + /* bit-wise with sysdata_feature bits */ u32 sysdata_fields; /* protected by target_list_lock */ @@ -802,28 +807,14 @@ static ssize_t remote_ip_store(struct config_item *item, const char *buf, return ret; } -/* Count number of entries we have in extradata. - * This is important because the extradata_complete only supports - * MAX_EXTRADATA_ITEMS entries. Before enabling any new {user,sys}data - * feature, number of entries needs to checked for available space. +/* Count number of entries we have in userdata. + * This is important because userdata only supports MAX_USERDATA_ITEMS + * entries. Before enabling any new userdata feature, number of entries needs + * to checked for available space. */ -static size_t count_extradata_entries(struct netconsole_target *nt) +static size_t count_userdata_entries(struct netconsole_target *nt) { - size_t entries; - - /* Userdata entries */ - entries = list_count_nodes(&nt->userdata_group.cg_children); - /* Plus sysdata entries */ - if (nt->sysdata_fields & SYSDATA_CPU_NR) - entries += 1; - if (nt->sysdata_fields & SYSDATA_TASKNAME) - entries += 1; - if (nt->sysdata_fields & SYSDATA_RELEASE) - entries += 1; - if (nt->sysdata_fields & SYSDATA_MSGID) - entries += 1; - - return entries; + return list_count_nodes(&nt->userdata_group.cg_children); } static ssize_t remote_mac_store(struct config_item *item, const char *buf, @@ -894,13 +885,13 @@ static void update_userdata(struct netconsole_target *nt) /* Clear the current string in case the last userdatum was deleted */ nt->userdata_length = 0; - nt->extradata_complete[0] = 0; + nt->userdata[0] = 0; list_for_each(entry, &nt->userdata_group.cg_children) { struct userdatum *udm_item; struct config_item *item; - if (child_count >= MAX_EXTRADATA_ITEMS) { + if (child_count >= MAX_USERDATA_ITEMS) { spin_unlock_irqrestore(&target_list_lock, flags); WARN_ON_ONCE(1); return; @@ -914,11 +905,11 @@ static void update_userdata(struct netconsole_target *nt) if (strnlen(udm_item->value, MAX_EXTRADATA_VALUE_LEN) == 0) continue; - /* This doesn't overflow extradata_complete since it will write - * one entry length (1/MAX_EXTRADATA_ITEMS long), entry count is + /* This doesn't overflow userdata since it will write + * one entry length (1/MAX_USERDATA_ITEMS long), entry count is * checked to not exceed MAX items with child_count above */ - nt->userdata_length += scnprintf(&nt->extradata_complete[nt->userdata_length], + nt->userdata_length += scnprintf(&nt->userdata[nt->userdata_length], MAX_EXTRADATA_ENTRY_LEN, " %s=%s\n", item->ci_name, udm_item->value); } @@ -962,7 +953,7 @@ static void disable_sysdata_feature(struct netconsole_target *nt, enum sysdata_feature feature) { nt->sysdata_fields &= ~feature; - nt->extradata_complete[nt->userdata_length] = 0; + nt->sysdata[0] = 0; } static ssize_t sysdata_msgid_enabled_store(struct config_item *item, @@ -982,12 +973,6 @@ static ssize_t sysdata_msgid_enabled_store(struct config_item *item, if (msgid_enabled == curr) goto unlock_ok; - if (msgid_enabled && - count_extradata_entries(nt) >= MAX_EXTRADATA_ITEMS) { - ret = -ENOSPC; - goto unlock; - } - if (msgid_enabled) nt->sysdata_fields |= SYSDATA_MSGID; else @@ -995,7 +980,6 @@ static ssize_t sysdata_msgid_enabled_store(struct config_item *item, unlock_ok: ret = strnlen(buf, count); -unlock: mutex_unlock(&dynamic_netconsole_mutex); mutex_unlock(&netconsole_subsys.su_mutex); return ret; @@ -1018,12 +1002,6 @@ static ssize_t sysdata_release_enabled_store(struct config_item *item, if (release_enabled == curr) goto unlock_ok; - if (release_enabled && - count_extradata_entries(nt) >= MAX_EXTRADATA_ITEMS) { - ret = -ENOSPC; - goto unlock; - } - if (release_enabled) nt->sysdata_fields |= SYSDATA_RELEASE; else @@ -1031,7 +1009,6 @@ static ssize_t sysdata_release_enabled_store(struct config_item *item, unlock_ok: ret = strnlen(buf, count); -unlock: mutex_unlock(&dynamic_netconsole_mutex); mutex_unlock(&netconsole_subsys.su_mutex); return ret; @@ -1054,12 +1031,6 @@ static ssize_t sysdata_taskname_enabled_store(struct config_item *item, if (taskname_enabled == curr) goto unlock_ok; - if (taskname_enabled && - count_extradata_entries(nt) >= MAX_EXTRADATA_ITEMS) { - ret = -ENOSPC; - goto unlock; - } - if (taskname_enabled) nt->sysdata_fields |= SYSDATA_TASKNAME; else @@ -1067,7 +1038,6 @@ static ssize_t sysdata_taskname_enabled_store(struct config_item *item, unlock_ok: ret = strnlen(buf, count); -unlock: mutex_unlock(&dynamic_netconsole_mutex); mutex_unlock(&netconsole_subsys.su_mutex); return ret; @@ -1092,27 +1062,16 @@ static ssize_t sysdata_cpu_nr_enabled_store(struct config_item *item, /* no change requested */ goto unlock_ok; - if (cpu_nr_enabled && - count_extradata_entries(nt) >= MAX_EXTRADATA_ITEMS) { - /* user wants the new feature, but there is no space in the - * buffer. - */ - ret = -ENOSPC; - goto unlock; - } - if (cpu_nr_enabled) nt->sysdata_fields |= SYSDATA_CPU_NR; else - /* This is special because extradata_complete might have - * remaining data from previous sysdata, and it needs to be - * cleaned. + /* This is special because sysdata might have remaining data + * from previous sysdata, and it needs to be cleaned. */ disable_sysdata_feature(nt, SYSDATA_CPU_NR); unlock_ok: ret = strnlen(buf, count); -unlock: mutex_unlock(&dynamic_netconsole_mutex); mutex_unlock(&netconsole_subsys.su_mutex); return ret; @@ -1156,7 +1115,7 @@ static struct config_item *userdatum_make_item(struct config_group *group, ud = to_userdata(&group->cg_item); nt = userdata_to_target(ud); - if (count_extradata_entries(nt) >= MAX_EXTRADATA_ITEMS) + if (count_userdata_entries(nt) >= MAX_USERDATA_ITEMS) return ERR_PTR(-ENOSPC); udm = kzalloc(sizeof(*udm), GFP_KERNEL); @@ -1363,22 +1322,21 @@ static void populate_configfs_item(struct netconsole_target *nt, static int sysdata_append_cpu_nr(struct netconsole_target *nt, int offset) { - /* Append cpu=%d at extradata_complete after userdata str */ - return scnprintf(&nt->extradata_complete[offset], + return scnprintf(&nt->sysdata[offset], MAX_EXTRADATA_ENTRY_LEN, " cpu=%u\n", raw_smp_processor_id()); } static int sysdata_append_taskname(struct netconsole_target *nt, int offset) { - return scnprintf(&nt->extradata_complete[offset], + return scnprintf(&nt->sysdata[offset], MAX_EXTRADATA_ENTRY_LEN, " taskname=%s\n", current->comm); } static int sysdata_append_release(struct netconsole_target *nt, int offset) { - return scnprintf(&nt->extradata_complete[offset], + return scnprintf(&nt->sysdata[offset], MAX_EXTRADATA_ENTRY_LEN, " release=%s\n", init_utsname()->release); } @@ -1386,46 +1344,36 @@ static int sysdata_append_release(struct netconsole_target *nt, int offset) static int sysdata_append_msgid(struct netconsole_target *nt, int offset) { wrapping_assign_add(nt->msgcounter, 1); - return scnprintf(&nt->extradata_complete[offset], + return scnprintf(&nt->sysdata[offset], MAX_EXTRADATA_ENTRY_LEN, " msgid=%u\n", nt->msgcounter); } /* - * prepare_extradata - append sysdata at extradata_complete in runtime + * prepare_sysdata - append sysdata in runtime * @nt: target to send message to */ -static int prepare_extradata(struct netconsole_target *nt) +static int prepare_sysdata(struct netconsole_target *nt) { - int extradata_len; - - /* userdata was appended when configfs write helper was called - * by update_userdata(). - */ - extradata_len = nt->userdata_length; + int sysdata_len = 0; if (!nt->sysdata_fields) goto out; if (nt->sysdata_fields & SYSDATA_CPU_NR) - extradata_len += sysdata_append_cpu_nr(nt, extradata_len); + sysdata_len += sysdata_append_cpu_nr(nt, sysdata_len); if (nt->sysdata_fields & SYSDATA_TASKNAME) - extradata_len += sysdata_append_taskname(nt, extradata_len); + sysdata_len += sysdata_append_taskname(nt, sysdata_len); if (nt->sysdata_fields & SYSDATA_RELEASE) - extradata_len += sysdata_append_release(nt, extradata_len); + sysdata_len += sysdata_append_release(nt, sysdata_len); if (nt->sysdata_fields & SYSDATA_MSGID) - extradata_len += sysdata_append_msgid(nt, extradata_len); + sysdata_len += sysdata_append_msgid(nt, sysdata_len); - WARN_ON_ONCE(extradata_len > - MAX_EXTRADATA_ENTRY_LEN * MAX_EXTRADATA_ITEMS); + WARN_ON_ONCE(sysdata_len > + MAX_EXTRADATA_ENTRY_LEN * MAX_SYSDATA_ITEMS); out: - return extradata_len; -} -#else /* CONFIG_NETCONSOLE_DYNAMIC not set */ -static int prepare_extradata(struct netconsole_target *nt) -{ - return 0; + return sysdata_len; } #endif /* CONFIG_NETCONSOLE_DYNAMIC */ @@ -1527,11 +1475,13 @@ static void send_msg_no_fragmentation(struct netconsole_target *nt, int msg_len, int release_len) { - const char *extradata = NULL; + const char *userdata = NULL; + const char *sysdata = NULL; const char *release; #ifdef CONFIG_NETCONSOLE_DYNAMIC - extradata = nt->extradata_complete; + userdata = nt->userdata; + sysdata = nt->sysdata; #endif if (release_len) { @@ -1543,10 +1493,15 @@ static void send_msg_no_fragmentation(struct netconsole_target *nt, memcpy(nt->buf, msg, msg_len); } - if (extradata) + if (userdata) msg_len += scnprintf(&nt->buf[msg_len], - MAX_PRINT_CHUNK - msg_len, - "%s", extradata); + MAX_PRINT_CHUNK - msg_len, "%s", + userdata); + + if (sysdata) + msg_len += scnprintf(&nt->buf[msg_len], + MAX_PRINT_CHUNK - msg_len, "%s", + sysdata); send_udp(nt, nt->buf, msg_len); } @@ -1561,33 +1516,45 @@ static void append_release(char *buf) static void send_fragmented_body(struct netconsole_target *nt, const char *msgbody_ptr, int header_len, - int msgbody_len, int extradata_len) + int msgbody_len, int sysdata_len) { - const char *extradata_ptr = NULL; + const char *userdata_ptr = NULL; + const char *sysdata_ptr = NULL; int data_len, data_sent = 0; - int extradata_offset = 0; + int userdata_offset = 0; + int sysdata_offset = 0; int msgbody_offset = 0; + int userdata_len = 0; #ifdef CONFIG_NETCONSOLE_DYNAMIC - extradata_ptr = nt->extradata_complete; + userdata_ptr = nt->userdata; + sysdata_ptr = nt->sysdata; + userdata_len = nt->userdata_length; #endif - if (WARN_ON_ONCE(!extradata_ptr && extradata_len != 0)) + if (WARN_ON_ONCE(!userdata_ptr && userdata_len != 0)) + return; + + if (WARN_ON_ONCE(!sysdata_ptr && sysdata_len != 0)) return; /* data_len represents the number of bytes that will be sent. This is * bigger than MAX_PRINT_CHUNK, thus, it will be split in multiple * packets */ - data_len = msgbody_len + extradata_len; + data_len = msgbody_len + userdata_len + sysdata_len; /* In each iteration of the while loop below, we send a packet * containing the header and a portion of the data. The data is - * composed of two parts: msgbody and extradata. We keep track of how - * many bytes have been sent so far using the data_sent variable, which - * ranges from 0 to the total bytes to be sent. + * composed of three parts: msgbody, userdata, and sysdata. + * We keep track of how many bytes have been sent from each part using + * the *_offset variables. + * We keep track of how many bytes have been sent overall using the + * data_sent variable, which ranges from 0 to the total bytes to be + * sent. */ while (data_sent < data_len) { - int extradata_left = extradata_len - extradata_offset; + int userdata_left = userdata_len - userdata_offset; + int sysdata_left = sysdata_len - sysdata_offset; int msgbody_left = msgbody_len - msgbody_offset; int buf_offset = 0; int this_chunk = 0; @@ -1608,13 +1575,24 @@ static void send_fragmented_body(struct netconsole_target *nt, buf_offset += this_chunk; data_sent += this_chunk; - /* after msgbody, append extradata */ - if (extradata_ptr && extradata_left) { - this_chunk = min(extradata_left, + /* after msgbody, append userdata */ + if (userdata_ptr && userdata_left) { + this_chunk = min(userdata_left, MAX_PRINT_CHUNK - buf_offset); memcpy(nt->buf + buf_offset, - extradata_ptr + extradata_offset, this_chunk); - extradata_offset += this_chunk; + userdata_ptr + userdata_offset, this_chunk); + userdata_offset += this_chunk; + buf_offset += this_chunk; + data_sent += this_chunk; + } + + /* after userdata, append sysdata */ + if (sysdata_ptr && sysdata_left) { + this_chunk = min(sysdata_left, + MAX_PRINT_CHUNK - buf_offset); + memcpy(nt->buf + buf_offset, + sysdata_ptr + sysdata_offset, this_chunk); + sysdata_offset += this_chunk; buf_offset += this_chunk; data_sent += this_chunk; } @@ -1631,7 +1609,7 @@ static void send_msg_fragmented(struct netconsole_target *nt, const char *msg, int msg_len, int release_len, - int extradata_len) + int sysdata_len) { int header_len, msgbody_len; const char *msgbody; @@ -1660,7 +1638,7 @@ static void send_msg_fragmented(struct netconsole_target *nt, * will be replaced */ send_fragmented_body(nt, msgbody, header_len, msgbody_len, - extradata_len); + sysdata_len); } /** @@ -1676,19 +1654,22 @@ static void send_msg_fragmented(struct netconsole_target *nt, static void send_ext_msg_udp(struct netconsole_target *nt, const char *msg, int msg_len) { + int userdata_len = 0; int release_len = 0; - int extradata_len; - - extradata_len = prepare_extradata(nt); + int sysdata_len = 0; +#ifdef CONFIG_NETCONSOLE_DYNAMIC + sysdata_len = prepare_sysdata(nt); + userdata_len = nt->userdata_length; +#endif if (nt->release) release_len = strlen(init_utsname()->release) + 1; - if (msg_len + release_len + extradata_len <= MAX_PRINT_CHUNK) + if (msg_len + release_len + sysdata_len + userdata_len <= MAX_PRINT_CHUNK) return send_msg_no_fragmentation(nt, msg, msg_len, release_len); return send_msg_fragmented(nt, msg, msg_len, release_len, - extradata_len); + sysdata_len); } static void write_ext_msg(struct console *con, const char *msg, From eb83801af2dcf60dc10d7c0f5d9cb8289b2af260 Mon Sep 17 00:00:00 2001 From: Gustavo Luiz Duarte Date: Wed, 19 Nov 2025 16:14:51 -0800 Subject: [PATCH 0990/1381] netconsole: Dynamic allocation of userdata buffer The userdata buffer in struct netconsole_target is currently statically allocated with a size of MAX_USERDATA_ITEMS * MAX_EXTRADATA_ENTRY_LEN (16 * 256 = 4096 bytes). This wastes memory when userdata entries are not used or when only a few entries are configured, which is common in typical usage scenarios. It also forces us to keep MAX_USERDATA_ITEMS small to limit the memory wasted. Change the userdata buffer from a static array to a dynamically allocated pointer. The buffer is now allocated on-demand in update_userdata() whenever userdata entries are added, modified, or removed via configfs. The implementation calculates the exact size needed for all current userdata entries, allocates a new buffer of that size, formats the entries into it, and atomically swaps it with the old buffer. This approach provides several benefits: - Memory efficiency: Targets with no userdata use zero bytes instead of 4KB, and targets with userdata only allocate what they need; - Scalability: Makes it practical to increase MAX_USERDATA_ITEMS to a much larger value without imposing a fixed memory cost on every target; - No hot-path overhead: Allocation occurs during configuration (write to configfs), not during message transmission If memory allocation fails during userdata update, -ENOMEM is returned to userspace through the configfs attribute write operation. The sysdata buffer remains statically allocated since it has a smaller fixed size (MAX_SYSDATA_ITEMS * MAX_EXTRADATA_ENTRY_LEN = 4 * 256 = 1024 bytes) and its content length is less predictable. Signed-off-by: Gustavo Luiz Duarte Reviewed-by: Breno Leitao Link: https://patch.msgid.link/20251119-netconsole_dynamic_extradata-v3-3-497ac3191707@meta.com Signed-off-by: Jakub Kicinski --- drivers/net/netconsole.c | 106 +++++++++++++++++++++++++++------------ 1 file changed, 73 insertions(+), 33 deletions(-) diff --git a/drivers/net/netconsole.c b/drivers/net/netconsole.c index 1bd811714322..0b350f82d915 100644 --- a/drivers/net/netconsole.c +++ b/drivers/net/netconsole.c @@ -155,7 +155,7 @@ struct netconsole_target { #ifdef CONFIG_NETCONSOLE_DYNAMIC struct config_group group; struct config_group userdata_group; - char userdata[MAX_EXTRADATA_ENTRY_LEN * MAX_USERDATA_ITEMS]; + char *userdata; size_t userdata_length; char sysdata[MAX_EXTRADATA_ENTRY_LEN * MAX_SYSDATA_ITEMS]; @@ -875,45 +875,77 @@ static ssize_t userdatum_value_show(struct config_item *item, char *buf) return sysfs_emit(buf, "%s\n", &(to_userdatum(item)->value[0])); } -static void update_userdata(struct netconsole_target *nt) +/* Navigate configfs and calculate the lentgh of the formatted string + * representing userdata. + * Must be called holding netconsole_subsys.su_mutex + */ +static int calc_userdata_len(struct netconsole_target *nt) { + struct userdatum *udm_item; + struct config_item *item; struct list_head *entry; - int child_count = 0; - unsigned long flags; - - spin_lock_irqsave(&target_list_lock, flags); - - /* Clear the current string in case the last userdatum was deleted */ - nt->userdata_length = 0; - nt->userdata[0] = 0; + int len = 0; list_for_each(entry, &nt->userdata_group.cg_children) { - struct userdatum *udm_item; - struct config_item *item; - - if (child_count >= MAX_USERDATA_ITEMS) { - spin_unlock_irqrestore(&target_list_lock, flags); - WARN_ON_ONCE(1); - return; - } - child_count++; - item = container_of(entry, struct config_item, ci_entry); udm_item = to_userdatum(item); - /* Skip userdata with no value set */ - if (strnlen(udm_item->value, MAX_EXTRADATA_VALUE_LEN) == 0) - continue; - - /* This doesn't overflow userdata since it will write - * one entry length (1/MAX_USERDATA_ITEMS long), entry count is - * checked to not exceed MAX items with child_count above - */ - nt->userdata_length += scnprintf(&nt->userdata[nt->userdata_length], - MAX_EXTRADATA_ENTRY_LEN, " %s=%s\n", - item->ci_name, udm_item->value); + if (udm_item->value[0]) { + len += snprintf(NULL, 0, " %s=%s\n", item->ci_name, + udm_item->value); + } } + return len; +} + +static int update_userdata(struct netconsole_target *nt) +{ + struct userdatum *udm_item; + struct config_item *item; + struct list_head *entry; + char *old_buf = NULL; + char *new_buf = NULL; + unsigned long flags; + int offset = 0; + int len; + + /* Calculate required buffer size */ + len = calc_userdata_len(nt); + + if (WARN_ON_ONCE(len > MAX_EXTRADATA_ENTRY_LEN * MAX_USERDATA_ITEMS)) + return -ENOSPC; + + /* Allocate new buffer */ + if (len) { + new_buf = kmalloc(len + 1, GFP_KERNEL); + if (!new_buf) + return -ENOMEM; + } + + /* Write userdata to new buffer */ + list_for_each(entry, &nt->userdata_group.cg_children) { + item = container_of(entry, struct config_item, ci_entry); + udm_item = to_userdatum(item); + /* Skip userdata with no value set */ + if (udm_item->value[0]) { + offset += scnprintf(&new_buf[offset], len + 1 - offset, + " %s=%s\n", item->ci_name, + udm_item->value); + } + } + + WARN_ON_ONCE(offset != len); + + /* Switch to new buffer and free old buffer */ + spin_lock_irqsave(&target_list_lock, flags); + old_buf = nt->userdata; + nt->userdata = new_buf; + nt->userdata_length = offset; spin_unlock_irqrestore(&target_list_lock, flags); + + kfree(old_buf); + + return 0; } static ssize_t userdatum_value_store(struct config_item *item, const char *buf, @@ -937,7 +969,9 @@ static ssize_t userdatum_value_store(struct config_item *item, const char *buf, ud = to_userdata(item->ci_parent); nt = userdata_to_target(ud); - update_userdata(nt); + ret = update_userdata(nt); + if (ret < 0) + goto out_unlock; ret = count; out_unlock: mutex_unlock(&dynamic_netconsole_mutex); @@ -1193,7 +1227,10 @@ static struct configfs_attribute *netconsole_target_attrs[] = { static void netconsole_target_release(struct config_item *item) { - kfree(to_target(item)); + struct netconsole_target *nt = to_target(item); + + kfree(nt->userdata); + kfree(nt); } static struct configfs_item_operations netconsole_target_item_ops = { @@ -1874,6 +1911,9 @@ static struct netconsole_target *alloc_param_target(char *target_config, static void free_param_target(struct netconsole_target *nt) { netpoll_cleanup(&nt->np); +#ifdef CONFIG_NETCONSOLE_DYNAMIC + kfree(nt->userdata); +#endif kfree(nt); } From 5ad9945341725c575a58a35e154a08ea81b3439a Mon Sep 17 00:00:00 2001 From: Gustavo Luiz Duarte Date: Wed, 19 Nov 2025 16:14:52 -0800 Subject: [PATCH 0991/1381] netconsole: Increase MAX_USERDATA_ITEMS Increase MAX_USERDATA_ITEMS from 16 to 256 entries now that the userdata buffer is allocated dynamically. The previous limit of 16 was necessary because the buffer was statically allocated for all targets. With dynamic allocation, we can support more entries without wasting memory on targets that don't use userdata. This allows users to attach more metadata to their netconsole messages, which is useful for complex debugging and logging scenarios. Also update the testcase accordingly. Signed-off-by: Gustavo Luiz Duarte Reviewed-by: Breno Leitao Link: https://patch.msgid.link/20251119-netconsole_dynamic_extradata-v3-4-497ac3191707@meta.com Signed-off-by: Jakub Kicinski --- drivers/net/netconsole.c | 2 +- tools/testing/selftests/drivers/net/netcons_overflow.sh | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/netconsole.c b/drivers/net/netconsole.c index 0b350f82d915..9cb4dfc242f5 100644 --- a/drivers/net/netconsole.c +++ b/drivers/net/netconsole.c @@ -50,7 +50,7 @@ MODULE_LICENSE("GPL"); /* The number 3 comes from userdata entry format characters (' ', '=', '\n') */ #define MAX_EXTRADATA_NAME_LEN (MAX_EXTRADATA_ENTRY_LEN - \ MAX_EXTRADATA_VALUE_LEN - 3) -#define MAX_USERDATA_ITEMS 16 +#define MAX_USERDATA_ITEMS 256 #define MAX_PRINT_CHUNK 1000 static char config[MAX_PARAM_LENGTH]; diff --git a/tools/testing/selftests/drivers/net/netcons_overflow.sh b/tools/testing/selftests/drivers/net/netcons_overflow.sh index 29bad56448a2..06089643b771 100755 --- a/tools/testing/selftests/drivers/net/netcons_overflow.sh +++ b/tools/testing/selftests/drivers/net/netcons_overflow.sh @@ -15,7 +15,7 @@ SCRIPTDIR=$(dirname "$(readlink -e "${BASH_SOURCE[0]}")") source "${SCRIPTDIR}"/lib/sh/lib_netcons.sh # This is coming from netconsole code. Check for it in drivers/net/netconsole.c -MAX_USERDATA_ITEMS=16 +MAX_USERDATA_ITEMS=256 # Function to create userdata entries function create_userdata_max_entries() { From 011d133bb988f80d597a9cbdab659414ba7ff72b Mon Sep 17 00:00:00 2001 From: Daniel Zahka Date: Tue, 18 Nov 2025 18:50:31 -0800 Subject: [PATCH 0992/1381] devlink: pass extack through to devlink_param::get() Allow devlink_param::get() handlers to report error messages via extack. This function is called in a few different contexts, but not all of them will have an valid extack to use. When devlink_param::get() is called from param_get_doit or param_get_dumpit contexts, pass the extack through so that drivers can report errors when retrieving param values. devlink_param::get() is called from the context of devlink_param_notify(), pass NULL in for the extack. Reviewed-by: Saeed Mahameed Reviewed-by: Aleksandr Loktionov Signed-off-by: Daniel Zahka Link: https://patch.msgid.link/20251119025038.651131-2-daniel.zahka@gmail.com Signed-off-by: Jakub Kicinski --- .../marvell/octeontx2/otx2_cpt_devlink.c | 6 ++++-- drivers/net/ethernet/amd/pds_core/core.h | 3 ++- drivers/net/ethernet/amd/pds_core/devlink.c | 3 ++- .../net/ethernet/broadcom/bnxt/bnxt_devlink.c | 6 ++++-- .../net/ethernet/intel/i40e/i40e_devlink.c | 3 ++- .../net/ethernet/intel/ice/devlink/devlink.c | 14 ++++++++++---- .../marvell/octeontx2/af/rvu_devlink.c | 15 ++++++++++----- .../marvell/octeontx2/nic/otx2_devlink.c | 6 ++++-- drivers/net/ethernet/mellanox/mlx4/main.c | 6 ++++-- .../net/ethernet/mellanox/mlx5/core/eswitch.c | 3 ++- .../mellanox/mlx5/core/eswitch_offloads.c | 3 ++- .../net/ethernet/mellanox/mlx5/core/fs_core.c | 3 ++- .../ethernet/mellanox/mlx5/core/fw_reset.c | 3 ++- .../mellanox/mlx5/core/lib/nv_param.c | 9 ++++++--- .../mellanox/mlxsw/spectrum_acl_tcam.c | 3 ++- .../ethernet/netronome/nfp/devlink_param.c | 3 ++- drivers/net/ethernet/qlogic/qed/qed_devlink.c | 3 ++- .../net/ethernet/stmicro/stmmac/stmmac_main.c | 3 ++- drivers/net/ethernet/ti/am65-cpsw-nuss.c | 3 ++- drivers/net/ethernet/ti/cpsw_new.c | 6 ++++-- drivers/net/wwan/iosm/iosm_ipc_devlink.c | 3 ++- include/net/devlink.h | 3 ++- include/net/dsa.h | 3 ++- net/devlink/param.c | 19 +++++++++++-------- net/dsa/devlink.c | 3 ++- 25 files changed, 89 insertions(+), 46 deletions(-) diff --git a/drivers/crypto/marvell/octeontx2/otx2_cpt_devlink.c b/drivers/crypto/marvell/octeontx2/otx2_cpt_devlink.c index 215a1a8ba7e9..07a74f702c3a 100644 --- a/drivers/crypto/marvell/octeontx2/otx2_cpt_devlink.c +++ b/drivers/crypto/marvell/octeontx2/otx2_cpt_devlink.c @@ -24,7 +24,8 @@ static int otx2_cpt_dl_egrp_delete(struct devlink *dl, u32 id, } static int otx2_cpt_dl_uc_info(struct devlink *dl, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { ctx->val.vstr[0] = '\0'; @@ -32,7 +33,8 @@ static int otx2_cpt_dl_uc_info(struct devlink *dl, u32 id, } static int otx2_cpt_dl_t106_mode_get(struct devlink *dl, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct otx2_cpt_devlink *cpt_dl = devlink_priv(dl); struct otx2_cptpf_dev *cptpf = cpt_dl->cptpf; diff --git a/drivers/net/ethernet/amd/pds_core/core.h b/drivers/net/ethernet/amd/pds_core/core.h index 0b53a1fab46d..4a6b35c84dab 100644 --- a/drivers/net/ethernet/amd/pds_core/core.h +++ b/drivers/net/ethernet/amd/pds_core/core.h @@ -255,7 +255,8 @@ int pdsc_dl_flash_update(struct devlink *dl, struct devlink_flash_update_params *params, struct netlink_ext_ack *extack); int pdsc_dl_enable_get(struct devlink *dl, u32 id, - struct devlink_param_gset_ctx *ctx); + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack); int pdsc_dl_enable_set(struct devlink *dl, u32 id, struct devlink_param_gset_ctx *ctx, struct netlink_ext_ack *extack); diff --git a/drivers/net/ethernet/amd/pds_core/devlink.c b/drivers/net/ethernet/amd/pds_core/devlink.c index d8dc39da4161..b576be626a29 100644 --- a/drivers/net/ethernet/amd/pds_core/devlink.c +++ b/drivers/net/ethernet/amd/pds_core/devlink.c @@ -22,7 +22,8 @@ pdsc_viftype *pdsc_dl_find_viftype_by_id(struct pdsc *pdsc, } int pdsc_dl_enable_get(struct devlink *dl, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct pdsc *pdsc = devlink_priv(dl); struct pdsc_viftype *vt_entry; diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c index 67ca02d84c97..15de802bbac4 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c @@ -1086,7 +1086,8 @@ static int bnxt_hwrm_nvm_req(struct bnxt *bp, u32 param_id, void *msg, } static int bnxt_dl_nvm_param_get(struct devlink *dl, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct bnxt *bp = bnxt_get_bp_from_dl(dl); struct hwrm_nvm_get_variable_input *req; @@ -1168,7 +1169,8 @@ static int bnxt_dl_msix_validate(struct devlink *dl, u32 id, } static int bnxt_remote_dev_reset_get(struct devlink *dl, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct bnxt *bp = bnxt_get_bp_from_dl(dl); diff --git a/drivers/net/ethernet/intel/i40e/i40e_devlink.c b/drivers/net/ethernet/intel/i40e/i40e_devlink.c index bc205e3077c7..229179ccc131 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_devlink.c +++ b/drivers/net/ethernet/intel/i40e/i40e_devlink.c @@ -24,7 +24,8 @@ static int i40e_max_mac_per_vf_set(struct devlink *devlink, static int i40e_max_mac_per_vf_get(struct devlink *devlink, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct i40e_pf *pf = devlink_priv(devlink); diff --git a/drivers/net/ethernet/intel/ice/devlink/devlink.c b/drivers/net/ethernet/intel/ice/devlink/devlink.c index 938914abbe06..d88b7f3fd1f9 100644 --- a/drivers/net/ethernet/intel/ice/devlink/devlink.c +++ b/drivers/net/ethernet/intel/ice/devlink/devlink.c @@ -610,11 +610,13 @@ static int ice_update_tx_topo_user_sel(struct ice_pf *pf, int layers) * @devlink: pointer to the devlink instance * @id: the parameter ID to set * @ctx: context to store the parameter value + * @extack: netlink extended ACK structure * * Return: zero on success and negative value on failure. */ static int ice_devlink_tx_sched_layers_get(struct devlink *devlink, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct ice_pf *pf = devlink_priv(devlink); int err; @@ -1349,7 +1351,8 @@ static const struct devlink_ops ice_sf_devlink_ops; static int ice_devlink_enable_roce_get(struct devlink *devlink, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct ice_pf *pf = devlink_priv(devlink); struct iidc_rdma_core_dev_info *cdev; @@ -1415,7 +1418,8 @@ ice_devlink_enable_roce_validate(struct devlink *devlink, u32 id, static int ice_devlink_enable_iw_get(struct devlink *devlink, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct ice_pf *pf = devlink_priv(devlink); struct iidc_rdma_core_dev_info *cdev; @@ -1522,11 +1526,13 @@ static int ice_devlink_local_fwd_str_to_mode(const char *mode_str) * @devlink: Pointer to the devlink instance. * @id: The parameter ID to set. * @ctx: Context to store the parameter value. + * @extack: netlink extended ACK structure * * Return: Zero. */ static int ice_devlink_local_fwd_get(struct devlink *devlink, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct ice_pf *pf = devlink_priv(devlink); struct ice_port_info *pi; diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_devlink.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_devlink.c index 3735372539bd..0f9953eaf1b0 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_devlink.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_devlink.c @@ -1233,7 +1233,8 @@ static int rvu_af_dl_dwrr_mtu_set(struct devlink *devlink, u32 id, } static int rvu_af_dl_dwrr_mtu_get(struct devlink *devlink, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct rvu_devlink *rvu_dl = devlink_priv(devlink); struct rvu *rvu = rvu_dl->rvu; @@ -1259,7 +1260,8 @@ enum rvu_af_dl_param_id { }; static int rvu_af_npc_exact_feature_get(struct devlink *devlink, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct rvu_devlink *rvu_dl = devlink_priv(devlink); struct rvu *rvu = rvu_dl->rvu; @@ -1314,7 +1316,8 @@ static int rvu_af_npc_exact_feature_validate(struct devlink *devlink, u32 id, } static int rvu_af_dl_npc_mcam_high_zone_percent_get(struct devlink *devlink, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct rvu_devlink *rvu_dl = devlink_priv(devlink); struct rvu *rvu = rvu_dl->rvu; @@ -1376,7 +1379,8 @@ static int rvu_af_dl_npc_mcam_high_zone_percent_validate(struct devlink *devlink } static int rvu_af_dl_npc_def_rule_cntr_get(struct devlink *devlink, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct rvu_devlink *rvu_dl = devlink_priv(devlink); struct rvu *rvu = rvu_dl->rvu; @@ -1402,7 +1406,8 @@ static int rvu_af_dl_npc_def_rule_cntr_set(struct devlink *devlink, u32 id, } static int rvu_af_dl_nix_maxlf_get(struct devlink *devlink, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct rvu_devlink *rvu_dl = devlink_priv(devlink); struct rvu *rvu = rvu_dl->rvu; diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_devlink.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_devlink.c index e13ae5484c19..a72694219df4 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_devlink.c +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_devlink.c @@ -48,7 +48,8 @@ static int otx2_dl_mcam_count_set(struct devlink *devlink, u32 id, } static int otx2_dl_mcam_count_get(struct devlink *devlink, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct otx2_devlink *otx2_dl = devlink_priv(devlink); struct otx2_nic *pfvf = otx2_dl->pfvf; @@ -84,7 +85,8 @@ static int otx2_dl_ucast_flt_cnt_set(struct devlink *devlink, u32 id, } static int otx2_dl_ucast_flt_cnt_get(struct devlink *devlink, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct otx2_devlink *otx2_dl = devlink_priv(devlink); struct otx2_nic *pfvf = otx2_dl->pfvf; diff --git a/drivers/net/ethernet/mellanox/mlx4/main.c b/drivers/net/ethernet/mellanox/mlx4/main.c index 03d2fc7d9b09..2de226951e19 100644 --- a/drivers/net/ethernet/mellanox/mlx4/main.c +++ b/drivers/net/ethernet/mellanox/mlx4/main.c @@ -174,7 +174,8 @@ MODULE_PARM_DESC(port_type_array, "Array of port types: HW_DEFAULT (0) is defaul static atomic_t pf_loading = ATOMIC_INIT(0); static int mlx4_devlink_ierr_reset_get(struct devlink *devlink, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { ctx->val.vbool = !!mlx4_internal_err_reset; return 0; @@ -189,7 +190,8 @@ static int mlx4_devlink_ierr_reset_set(struct devlink *devlink, u32 id, } static int mlx4_devlink_crdump_snapshot_get(struct devlink *devlink, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct mlx4_priv *priv = devlink_priv(devlink); struct mlx4_dev *dev = &priv->dev; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c index 3adf2b1cd26a..4b7a1ce7f406 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c @@ -1969,7 +1969,8 @@ static int mlx5_devlink_esw_multiport_set(struct devlink *devlink, u32 id, } static int mlx5_devlink_esw_multiport_get(struct devlink *devlink, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct mlx5_core_dev *dev = devlink_priv(devlink); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c index 8ebca0d17f65..8de6c7f6c294 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -2618,7 +2618,8 @@ static int esw_port_metadata_set(struct devlink *devlink, u32 id, } static int esw_port_metadata_get(struct devlink *devlink, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct mlx5_core_dev *dev = devlink_priv(devlink); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c index 2b755a0035ce..0a6031a64c6f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c @@ -3833,7 +3833,8 @@ static int mlx5_fs_mode_set(struct devlink *devlink, u32 id, } static int mlx5_fs_mode_get(struct devlink *devlink, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct mlx5_core_dev *dev = devlink_priv(devlink); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c b/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c index 89e399606877..2bceb42c98cc 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c @@ -73,7 +73,8 @@ static int mlx5_fw_reset_enable_remote_dev_reset_set(struct devlink *devlink, u3 } static int mlx5_fw_reset_enable_remote_dev_reset_get(struct devlink *devlink, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct mlx5_core_dev *dev = devlink_priv(devlink); struct mlx5_fw_reset *fw_reset; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/nv_param.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/nv_param.c index 459a0b4d08e6..70cb22fa96db 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lib/nv_param.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/nv_param.c @@ -200,7 +200,8 @@ static const char *const static int mlx5_nv_param_devlink_cqe_compress_get(struct devlink *devlink, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct mlx5_core_dev *dev = devlink_priv(devlink); u32 mnvda[MLX5_ST_SZ_DW(mnvda_reg)] = {}; @@ -302,7 +303,8 @@ static int mlx5_nv_param_read_per_host_pf_conf(struct mlx5_core_dev *dev, } static int mlx5_devlink_enable_sriov_get(struct devlink *devlink, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct mlx5_core_dev *dev = devlink_priv(devlink); u32 mnvda[MLX5_ST_SZ_DW(mnvda_reg)] = {}; @@ -413,7 +415,8 @@ static int mlx5_devlink_enable_sriov_set(struct devlink *devlink, u32 id, } static int mlx5_devlink_total_vfs_get(struct devlink *devlink, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct mlx5_core_dev *dev = devlink_priv(devlink); u32 mnvda[MLX5_ST_SZ_DW(mnvda_reg)] = {}; diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c index b1d08e958bf9..69f9da9fb305 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c @@ -1489,7 +1489,8 @@ mlxsw_sp_acl_tcam_vregion_rehash(struct mlxsw_sp *mlxsw_sp, static int mlxsw_sp_acl_tcam_region_rehash_intrvl_get(struct devlink *devlink, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct mlxsw_core *mlxsw_core = devlink_priv(devlink); struct mlxsw_sp_acl_tcam *tcam; diff --git a/drivers/net/ethernet/netronome/nfp/devlink_param.c b/drivers/net/ethernet/netronome/nfp/devlink_param.c index 0e1a3800f371..85e3b19e6165 100644 --- a/drivers/net/ethernet/netronome/nfp/devlink_param.c +++ b/drivers/net/ethernet/netronome/nfp/devlink_param.c @@ -81,7 +81,8 @@ static const struct nfp_devlink_param_u8_arg nfp_devlink_u8_args[] = { static int nfp_devlink_param_u8_get(struct devlink *devlink, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { const struct nfp_devlink_param_u8_arg *arg; struct nfp_pf *pf = devlink_priv(devlink); diff --git a/drivers/net/ethernet/qlogic/qed/qed_devlink.c b/drivers/net/ethernet/qlogic/qed/qed_devlink.c index 94c5689b5abd..0c5278c0598c 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_devlink.c +++ b/drivers/net/ethernet/qlogic/qed/qed_devlink.c @@ -121,7 +121,8 @@ void qed_fw_reporters_destroy(struct devlink *devlink) } static int qed_dl_param_get(struct devlink *dl, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct qed_devlink *qed_dl = devlink_priv(dl); struct qed_dev *cdev; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index b1aa236d8051..6cacedb2c9b3 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -7531,7 +7531,8 @@ static int stmmac_dl_ts_coarse_set(struct devlink *dl, u32 id, } static int stmmac_dl_ts_coarse_get(struct devlink *dl, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct stmmac_devlink_priv *dl_priv = devlink_priv(dl); struct stmmac_priv *priv = dl_priv->stmmac_priv; diff --git a/drivers/net/ethernet/ti/am65-cpsw-nuss.c b/drivers/net/ethernet/ti/am65-cpsw-nuss.c index d5f358ec9820..5924db6be3fe 100644 --- a/drivers/net/ethernet/ti/am65-cpsw-nuss.c +++ b/drivers/net/ethernet/ti/am65-cpsw-nuss.c @@ -3068,7 +3068,8 @@ static void am65_cpsw_init_host_port_emac(struct am65_cpsw_common *common) } static int am65_cpsw_dl_switch_mode_get(struct devlink *dl, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct am65_cpsw_devlink *dl_priv = devlink_priv(dl); struct am65_cpsw_common *common = dl_priv->common; diff --git a/drivers/net/ethernet/ti/cpsw_new.c b/drivers/net/ethernet/ti/cpsw_new.c index 8b9e2078c602..ab88d4c02cbd 100644 --- a/drivers/net/ethernet/ti/cpsw_new.c +++ b/drivers/net/ethernet/ti/cpsw_new.c @@ -1618,7 +1618,8 @@ static const struct devlink_ops cpsw_devlink_ops = { }; static int cpsw_dl_switch_mode_get(struct devlink *dl, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct cpsw_devlink *dl_priv = devlink_priv(dl); struct cpsw_common *cpsw = dl_priv->cpsw; @@ -1753,7 +1754,8 @@ static int cpsw_dl_switch_mode_set(struct devlink *dl, u32 id, } static int cpsw_dl_ale_ctrl_get(struct devlink *dl, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct cpsw_devlink *dl_priv = devlink_priv(dl); struct cpsw_common *cpsw = dl_priv->cpsw; diff --git a/drivers/net/wwan/iosm/iosm_ipc_devlink.c b/drivers/net/wwan/iosm/iosm_ipc_devlink.c index 33d6342124bc..301a9d294d30 100644 --- a/drivers/net/wwan/iosm/iosm_ipc_devlink.c +++ b/drivers/net/wwan/iosm/iosm_ipc_devlink.c @@ -21,7 +21,8 @@ static struct iosm_coredump_file_info list[IOSM_NOF_CD_REGION] = { /* Get the param values for the specific param ID's */ static int ipc_devlink_get_param(struct devlink *dl, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct iosm_devlink *ipc_devlink = devlink_priv(dl); diff --git a/include/net/devlink.h b/include/net/devlink.h index d01046ef0577..5f479227144d 100644 --- a/include/net/devlink.h +++ b/include/net/devlink.h @@ -490,7 +490,8 @@ struct devlink_param { enum devlink_param_type type; unsigned long supported_cmodes; int (*get)(struct devlink *devlink, u32 id, - struct devlink_param_gset_ctx *ctx); + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack); int (*set)(struct devlink *devlink, u32 id, struct devlink_param_gset_ctx *ctx, struct netlink_ext_ack *extack); diff --git a/include/net/dsa.h b/include/net/dsa.h index 97d5f401cfcf..e40cdc12f7f3 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -1251,7 +1251,8 @@ struct dsa_switch_ops { dsa_devlink_param_get, dsa_devlink_param_set, NULL) int dsa_devlink_param_get(struct devlink *dl, u32 id, - struct devlink_param_gset_ctx *ctx); + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack); int dsa_devlink_param_set(struct devlink *dl, u32 id, struct devlink_param_gset_ctx *ctx, struct netlink_ext_ack *extack); diff --git a/net/devlink/param.c b/net/devlink/param.c index 6b233b13b69a..3dbd023e4c36 100644 --- a/net/devlink/param.c +++ b/net/devlink/param.c @@ -174,11 +174,12 @@ devlink_param_cmode_is_supported(const struct devlink_param *param, static int devlink_param_get(struct devlink *devlink, const struct devlink_param *param, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { if (!param->get) return -EOPNOTSUPP; - return param->get(devlink, param->id, ctx); + return param->get(devlink, param->id, ctx, extack); } static int devlink_param_set(struct devlink *devlink, @@ -250,7 +251,8 @@ static int devlink_nl_param_fill(struct sk_buff *msg, struct devlink *devlink, unsigned int port_index, struct devlink_param_item *param_item, enum devlink_command cmd, - u32 portid, u32 seq, int flags) + u32 portid, u32 seq, int flags, + struct netlink_ext_ack *extack) { union devlink_param_value param_value[DEVLINK_PARAM_CMODE_MAX + 1]; bool param_value_set[DEVLINK_PARAM_CMODE_MAX + 1] = {}; @@ -275,7 +277,7 @@ static int devlink_nl_param_fill(struct sk_buff *msg, struct devlink *devlink, return -EOPNOTSUPP; } else { ctx.cmode = i; - err = devlink_param_get(devlink, param, &ctx); + err = devlink_param_get(devlink, param, &ctx, extack); if (err) return err; param_value[i] = ctx.val; @@ -357,7 +359,7 @@ static void devlink_param_notify(struct devlink *devlink, if (!msg) return; err = devlink_nl_param_fill(msg, devlink, port_index, param_item, cmd, - 0, 0, 0); + 0, 0, 0, NULL); if (err) { nlmsg_free(msg); return; @@ -400,7 +402,8 @@ static int devlink_nl_param_get_dump_one(struct sk_buff *msg, err = devlink_nl_param_fill(msg, devlink, 0, param_item, DEVLINK_CMD_PARAM_GET, NETLINK_CB(cb->skb).portid, - cb->nlh->nlmsg_seq, flags); + cb->nlh->nlmsg_seq, flags, + cb->extack); if (err == -EOPNOTSUPP) { err = 0; } else if (err) { @@ -509,8 +512,8 @@ int devlink_nl_param_get_doit(struct sk_buff *skb, return -ENOMEM; err = devlink_nl_param_fill(msg, devlink, 0, param_item, - DEVLINK_CMD_PARAM_GET, - info->snd_portid, info->snd_seq, 0); + DEVLINK_CMD_PARAM_GET, info->snd_portid, + info->snd_seq, 0, info->extack); if (err) { nlmsg_free(msg); return err; diff --git a/net/dsa/devlink.c b/net/dsa/devlink.c index f41f9fc2194e..ed342f345692 100644 --- a/net/dsa/devlink.c +++ b/net/dsa/devlink.c @@ -182,7 +182,8 @@ static const struct devlink_ops dsa_devlink_ops = { }; int dsa_devlink_param_get(struct devlink *dl, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct dsa_switch *ds = dsa_devlink_to_ds(dl); From 17a42aa465c000fccdcf581950d2c6f4e0c343ca Mon Sep 17 00:00:00 2001 From: Daniel Zahka Date: Tue, 18 Nov 2025 18:50:32 -0800 Subject: [PATCH 0993/1381] devlink: refactor devlink_nl_param_value_fill_one() Lift the param type demux and value attr placement into a separate function. This new function, devlink_nl_param_put(), can be used to place additional types values in the value array, e.g., default, current, next values. This commit has no functional change. Signed-off-by: Daniel Zahka Link: https://patch.msgid.link/20251119025038.651131-3-daniel.zahka@gmail.com Signed-off-by: Jakub Kicinski --- net/devlink/param.c | 72 +++++++++++++++++++++++++-------------------- 1 file changed, 40 insertions(+), 32 deletions(-) diff --git a/net/devlink/param.c b/net/devlink/param.c index 3dbd023e4c36..3aa14ef345f0 100644 --- a/net/devlink/param.c +++ b/net/devlink/param.c @@ -192,6 +192,39 @@ static int devlink_param_set(struct devlink *devlink, return param->set(devlink, param->id, ctx, extack); } +static int +devlink_nl_param_value_put(struct sk_buff *msg, enum devlink_param_type type, + int nla_type, union devlink_param_value val) +{ + switch (type) { + case DEVLINK_PARAM_TYPE_U8: + if (nla_put_u8(msg, nla_type, val.vu8)) + return -EMSGSIZE; + break; + case DEVLINK_PARAM_TYPE_U16: + if (nla_put_u16(msg, nla_type, val.vu16)) + return -EMSGSIZE; + break; + case DEVLINK_PARAM_TYPE_U32: + if (nla_put_u32(msg, nla_type, val.vu32)) + return -EMSGSIZE; + break; + case DEVLINK_PARAM_TYPE_U64: + if (devlink_nl_put_u64(msg, nla_type, val.vu64)) + return -EMSGSIZE; + break; + case DEVLINK_PARAM_TYPE_STRING: + if (nla_put_string(msg, nla_type, val.vstr)) + return -EMSGSIZE; + break; + case DEVLINK_PARAM_TYPE_BOOL: + if (val.vbool && nla_put_flag(msg, nla_type)) + return -EMSGSIZE; + break; + } + return 0; +} + static int devlink_nl_param_value_fill_one(struct sk_buff *msg, enum devlink_param_type type, @@ -199,52 +232,27 @@ devlink_nl_param_value_fill_one(struct sk_buff *msg, union devlink_param_value val) { struct nlattr *param_value_attr; + int err = -EMSGSIZE; param_value_attr = nla_nest_start_noflag(msg, DEVLINK_ATTR_PARAM_VALUE); if (!param_value_attr) - goto nla_put_failure; + return -EMSGSIZE; if (nla_put_u8(msg, DEVLINK_ATTR_PARAM_VALUE_CMODE, cmode)) goto value_nest_cancel; - switch (type) { - case DEVLINK_PARAM_TYPE_U8: - if (nla_put_u8(msg, DEVLINK_ATTR_PARAM_VALUE_DATA, val.vu8)) - goto value_nest_cancel; - break; - case DEVLINK_PARAM_TYPE_U16: - if (nla_put_u16(msg, DEVLINK_ATTR_PARAM_VALUE_DATA, val.vu16)) - goto value_nest_cancel; - break; - case DEVLINK_PARAM_TYPE_U32: - if (nla_put_u32(msg, DEVLINK_ATTR_PARAM_VALUE_DATA, val.vu32)) - goto value_nest_cancel; - break; - case DEVLINK_PARAM_TYPE_U64: - if (devlink_nl_put_u64(msg, DEVLINK_ATTR_PARAM_VALUE_DATA, - val.vu64)) - goto value_nest_cancel; - break; - case DEVLINK_PARAM_TYPE_STRING: - if (nla_put_string(msg, DEVLINK_ATTR_PARAM_VALUE_DATA, - val.vstr)) - goto value_nest_cancel; - break; - case DEVLINK_PARAM_TYPE_BOOL: - if (val.vbool && - nla_put_flag(msg, DEVLINK_ATTR_PARAM_VALUE_DATA)) - goto value_nest_cancel; - break; - } + err = devlink_nl_param_value_put(msg, type, + DEVLINK_ATTR_PARAM_VALUE_DATA, val); + if (err) + goto value_nest_cancel; nla_nest_end(msg, param_value_attr); return 0; value_nest_cancel: nla_nest_cancel(msg, param_value_attr); -nla_put_failure: - return -EMSGSIZE; + return err; } static int devlink_nl_param_fill(struct sk_buff *msg, struct devlink *devlink, From 2a367002ed321e884276c3d7232a362ddd1bf7d6 Mon Sep 17 00:00:00 2001 From: Daniel Zahka Date: Tue, 18 Nov 2025 18:50:33 -0800 Subject: [PATCH 0994/1381] devlink: support default values for param-get and param-set Support querying and resetting to default param values. Introduce two new devlink netlink attrs: DEVLINK_ATTR_PARAM_VALUE_DEFAULT and DEVLINK_ATTR_PARAM_RESET_DEFAULT. The former is used to contain an optional parameter value inside of the param_value nested attribute. The latter is used in param-set requests from userspace to indicate that the driver should reset the param to its default value. To implement this, two new functions are added to the devlink driver api: devlink_param::get_default() and devlink_param::reset_default(). These callbacks allow drivers to implement default param actions for runtime and permanent cmodes. For driverinit params, the core latches the last value set by a driver via devl_param_driverinit_value_set(), and uses that as the default value for a param. Because default parameter values are optional, it would be impossible to discern whether or not a param of type bool has default value of false or not provided if the default value is encoded using a netlink flag type. For this reason, when a DEVLINK_PARAM_TYPE_BOOL has an associated default value, the default value is encoded using a u8 type. Signed-off-by: Daniel Zahka Link: https://patch.msgid.link/20251119025038.651131-4-daniel.zahka@gmail.com Signed-off-by: Jakub Kicinski --- Documentation/netlink/specs/devlink.yaml | 9 ++ .../networking/devlink/devlink-params.rst | 10 ++ include/net/devlink.h | 42 +++++++ include/uapi/linux/devlink.h | 3 + net/devlink/netlink_gen.c | 5 +- net/devlink/param.c | 105 ++++++++++++++++-- 6 files changed, 160 insertions(+), 14 deletions(-) diff --git a/Documentation/netlink/specs/devlink.yaml b/Documentation/netlink/specs/devlink.yaml index 426d5aa7d955..837112da6738 100644 --- a/Documentation/netlink/specs/devlink.yaml +++ b/Documentation/netlink/specs/devlink.yaml @@ -859,6 +859,14 @@ attribute-sets: name: health-reporter-burst-period type: u64 doc: Time (in msec) for recoveries before starting the grace period. + + # TODO: fill in the attributes in between + + - + name: param-reset-default + type: flag + doc: Request restoring parameter to its default value. + value: 183 - name: dl-dev-stats subset-of: devlink @@ -1793,6 +1801,7 @@ operations: - param-type # param-value-data is missing here as the type is variable - param-value-cmode + - param-reset-default - name: region-get diff --git a/Documentation/networking/devlink/devlink-params.rst b/Documentation/networking/devlink/devlink-params.rst index c0597d456641..ea17756dcda6 100644 --- a/Documentation/networking/devlink/devlink-params.rst +++ b/Documentation/networking/devlink/devlink-params.rst @@ -41,6 +41,16 @@ In order for ``driverinit`` parameters to take effect, the driver must support reloading via the ``devlink-reload`` command. This command will request a reload of the device driver. +Default parameter values +========================= + +Drivers may optionally export default values for parameters of cmode +``runtime`` and ``permanent``. For ``driverinit`` parameters, the last +value set by the driver will be used as the default value. Drivers can +also support resetting params with cmode ``runtime`` and ``permanent`` +to their default values. Resetting ``driverinit`` params is supported +by devlink core without additional driver support needed. + .. _devlink_params_generic: Generic configuration parameters diff --git a/include/net/devlink.h b/include/net/devlink.h index 5f479227144d..cb839e0435a1 100644 --- a/include/net/devlink.h +++ b/include/net/devlink.h @@ -479,6 +479,10 @@ struct devlink_flash_notify { * @set: set parameter value, used for runtime and permanent * configuration modes * @validate: validate input value is applicable (within value range, etc.) + * @get_default: get parameter default value, used for runtime and permanent + * configuration modes + * @reset_default: reset parameter to default value, used for runtime and permanent + * configuration modes * * This struct should be used by the driver to fill the data for * a parameter it registers. @@ -498,6 +502,12 @@ struct devlink_param { int (*validate)(struct devlink *devlink, u32 id, union devlink_param_value val, struct netlink_ext_ack *extack); + int (*get_default)(struct devlink *devlink, u32 id, + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack); + int (*reset_default)(struct devlink *devlink, u32 id, + enum devlink_param_cmode cmode, + struct netlink_ext_ack *extack); }; struct devlink_param_item { @@ -509,6 +519,7 @@ struct devlink_param_item { * until reload. */ bool driverinit_value_new_valid; + union devlink_param_value driverinit_default; }; enum devlink_param_generic_id { @@ -630,6 +641,37 @@ enum devlink_param_generic_id { .validate = _validate, \ } +#define DEVLINK_PARAM_GENERIC_WITH_DEFAULTS(_id, _cmodes, _get, _set, \ + _validate, _get_default, \ + _reset_default) \ +{ \ + .id = DEVLINK_PARAM_GENERIC_ID_##_id, \ + .name = DEVLINK_PARAM_GENERIC_##_id##_NAME, \ + .type = DEVLINK_PARAM_GENERIC_##_id##_TYPE, \ + .generic = true, \ + .supported_cmodes = _cmodes, \ + .get = _get, \ + .set = _set, \ + .validate = _validate, \ + .get_default = _get_default, \ + .reset_default = _reset_default, \ +} + +#define DEVLINK_PARAM_DRIVER_WITH_DEFAULTS(_id, _name, _type, _cmodes, \ + _get, _set, _validate, \ + _get_default, _reset_default) \ +{ \ + .id = _id, \ + .name = _name, \ + .type = _type, \ + .supported_cmodes = _cmodes, \ + .get = _get, \ + .set = _set, \ + .validate = _validate, \ + .get_default = _get_default, \ + .reset_default = _reset_default, \ +} + /* Identifier of board design */ #define DEVLINK_INFO_VERSION_GENERIC_BOARD_ID "board.id" /* Revision of board design */ diff --git a/include/uapi/linux/devlink.h b/include/uapi/linux/devlink.h index 157f11d3fb72..e7d6b6d13470 100644 --- a/include/uapi/linux/devlink.h +++ b/include/uapi/linux/devlink.h @@ -639,6 +639,9 @@ enum devlink_attr { DEVLINK_ATTR_HEALTH_REPORTER_BURST_PERIOD, /* u64 */ + DEVLINK_ATTR_PARAM_VALUE_DEFAULT, /* dynamic */ + DEVLINK_ATTR_PARAM_RESET_DEFAULT, /* flag */ + /* Add new attributes above here, update the spec in * Documentation/netlink/specs/devlink.yaml and re-generate * net/devlink/netlink_gen.c. diff --git a/net/devlink/netlink_gen.c b/net/devlink/netlink_gen.c index 5ad435aee29d..580985025f49 100644 --- a/net/devlink/netlink_gen.c +++ b/net/devlink/netlink_gen.c @@ -301,12 +301,13 @@ static const struct nla_policy devlink_param_get_dump_nl_policy[DEVLINK_ATTR_DEV }; /* DEVLINK_CMD_PARAM_SET - do */ -static const struct nla_policy devlink_param_set_nl_policy[DEVLINK_ATTR_PARAM_VALUE_CMODE + 1] = { +static const struct nla_policy devlink_param_set_nl_policy[DEVLINK_ATTR_PARAM_RESET_DEFAULT + 1] = { [DEVLINK_ATTR_BUS_NAME] = { .type = NLA_NUL_STRING, }, [DEVLINK_ATTR_DEV_NAME] = { .type = NLA_NUL_STRING, }, [DEVLINK_ATTR_PARAM_NAME] = { .type = NLA_NUL_STRING, }, [DEVLINK_ATTR_PARAM_TYPE] = NLA_POLICY_VALIDATE_FN(NLA_U8, &devlink_attr_param_type_validate), [DEVLINK_ATTR_PARAM_VALUE_CMODE] = NLA_POLICY_MAX(NLA_U8, 2), + [DEVLINK_ATTR_PARAM_RESET_DEFAULT] = { .type = NLA_FLAG, }, }; /* DEVLINK_CMD_REGION_GET - do */ @@ -919,7 +920,7 @@ const struct genl_split_ops devlink_nl_ops[74] = { .doit = devlink_nl_param_set_doit, .post_doit = devlink_nl_post_doit, .policy = devlink_param_set_nl_policy, - .maxattr = DEVLINK_ATTR_PARAM_VALUE_CMODE, + .maxattr = DEVLINK_ATTR_PARAM_RESET_DEFAULT, .flags = GENL_ADMIN_PERM | GENL_CMD_CAP_DO, }, { diff --git a/net/devlink/param.c b/net/devlink/param.c index 3aa14ef345f0..e0ea93eded43 100644 --- a/net/devlink/param.c +++ b/net/devlink/param.c @@ -192,9 +192,32 @@ static int devlink_param_set(struct devlink *devlink, return param->set(devlink, param->id, ctx, extack); } +static int devlink_param_get_default(struct devlink *devlink, + const struct devlink_param *param, + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) +{ + if (!param->get_default) + return -EOPNOTSUPP; + + return param->get_default(devlink, param->id, ctx, extack); +} + +static int devlink_param_reset_default(struct devlink *devlink, + const struct devlink_param *param, + enum devlink_param_cmode cmode, + struct netlink_ext_ack *extack) +{ + if (!param->reset_default) + return -EOPNOTSUPP; + + return param->reset_default(devlink, param->id, cmode, extack); +} + static int devlink_nl_param_value_put(struct sk_buff *msg, enum devlink_param_type type, - int nla_type, union devlink_param_value val) + int nla_type, union devlink_param_value val, + bool flag_as_u8) { switch (type) { case DEVLINK_PARAM_TYPE_U8: @@ -218,8 +241,16 @@ devlink_nl_param_value_put(struct sk_buff *msg, enum devlink_param_type type, return -EMSGSIZE; break; case DEVLINK_PARAM_TYPE_BOOL: - if (val.vbool && nla_put_flag(msg, nla_type)) - return -EMSGSIZE; + /* default values of type bool are encoded with u8, so that + * false can be distinguished from not present + */ + if (flag_as_u8) { + if (nla_put_u8(msg, nla_type, val.vbool)) + return -EMSGSIZE; + } else { + if (val.vbool && nla_put_flag(msg, nla_type)) + return -EMSGSIZE; + } break; } return 0; @@ -229,7 +260,9 @@ static int devlink_nl_param_value_fill_one(struct sk_buff *msg, enum devlink_param_type type, enum devlink_param_cmode cmode, - union devlink_param_value val) + union devlink_param_value val, + union devlink_param_value default_val, + bool has_default) { struct nlattr *param_value_attr; int err = -EMSGSIZE; @@ -243,10 +276,19 @@ devlink_nl_param_value_fill_one(struct sk_buff *msg, goto value_nest_cancel; err = devlink_nl_param_value_put(msg, type, - DEVLINK_ATTR_PARAM_VALUE_DATA, val); + DEVLINK_ATTR_PARAM_VALUE_DATA, + val, false); if (err) goto value_nest_cancel; + if (has_default) { + err = devlink_nl_param_value_put(msg, type, + DEVLINK_ATTR_PARAM_VALUE_DEFAULT, + default_val, true); + if (err) + goto value_nest_cancel; + } + nla_nest_end(msg, param_value_attr); return 0; @@ -262,7 +304,9 @@ static int devlink_nl_param_fill(struct sk_buff *msg, struct devlink *devlink, u32 portid, u32 seq, int flags, struct netlink_ext_ack *extack) { + union devlink_param_value default_value[DEVLINK_PARAM_CMODE_MAX + 1]; union devlink_param_value param_value[DEVLINK_PARAM_CMODE_MAX + 1]; + bool default_value_set[DEVLINK_PARAM_CMODE_MAX + 1] = {}; bool param_value_set[DEVLINK_PARAM_CMODE_MAX + 1] = {}; const struct devlink_param *param = param_item->param; struct devlink_param_gset_ctx ctx; @@ -283,12 +327,26 @@ static int devlink_nl_param_fill(struct sk_buff *msg, struct devlink *devlink, param_value[i] = param_item->driverinit_value; else return -EOPNOTSUPP; + + if (param_item->driverinit_value_valid) { + default_value[i] = param_item->driverinit_default; + default_value_set[i] = true; + } } else { ctx.cmode = i; err = devlink_param_get(devlink, param, &ctx, extack); if (err) return err; param_value[i] = ctx.val; + + err = devlink_param_get_default(devlink, param, &ctx, + extack); + if (!err) { + default_value[i] = ctx.val; + default_value_set[i] = true; + } else if (err != -EOPNOTSUPP) { + return err; + } } param_value_set[i] = true; } @@ -325,7 +383,9 @@ static int devlink_nl_param_fill(struct sk_buff *msg, struct devlink *devlink, if (!param_value_set[i]) continue; err = devlink_nl_param_value_fill_one(msg, param->type, - i, param_value[i]); + i, param_value[i], + default_value[i], + default_value_set[i]); if (err) goto values_list_nest_cancel; } @@ -542,6 +602,7 @@ static int __devlink_nl_cmd_param_set_doit(struct devlink *devlink, struct devlink_param_item *param_item; const struct devlink_param *param; union devlink_param_value value; + bool reset_default; int err = 0; param_item = devlink_param_get_from_info(params, info); @@ -553,13 +614,18 @@ static int __devlink_nl_cmd_param_set_doit(struct devlink *devlink, return err; if (param_type != param->type) return -EINVAL; - err = devlink_param_value_get_from_info(param, info, &value); - if (err) - return err; - if (param->validate) { - err = param->validate(devlink, param->id, value, info->extack); + + reset_default = info->attrs[DEVLINK_ATTR_PARAM_RESET_DEFAULT]; + if (!reset_default) { + err = devlink_param_value_get_from_info(param, info, &value); if (err) return err; + if (param->validate) { + err = param->validate(devlink, param->id, value, + info->extack); + if (err) + return err; + } } if (GENL_REQ_ATTR_CHECK(info, DEVLINK_ATTR_PARAM_VALUE_CMODE)) @@ -569,6 +635,15 @@ static int __devlink_nl_cmd_param_set_doit(struct devlink *devlink, return -EOPNOTSUPP; if (cmode == DEVLINK_PARAM_CMODE_DRIVERINIT) { + if (reset_default) { + if (!param_item->driverinit_value_valid) { + NL_SET_ERR_MSG(info->extack, + "Default value not available"); + return -EOPNOTSUPP; + } + value = param_item->driverinit_default; + } + param_item->driverinit_value_new = value; param_item->driverinit_value_new_valid = true; } else { @@ -576,7 +651,12 @@ static int __devlink_nl_cmd_param_set_doit(struct devlink *devlink, return -EOPNOTSUPP; ctx.val = value; ctx.cmode = cmode; - err = devlink_param_set(devlink, param, &ctx, info->extack); + if (reset_default) + err = devlink_param_reset_default(devlink, param, cmode, + info->extack); + else + err = devlink_param_set(devlink, param, &ctx, + info->extack); if (err) return err; } @@ -824,6 +904,7 @@ void devl_param_driverinit_value_set(struct devlink *devlink, u32 param_id, param_item->driverinit_value = init_val; param_item->driverinit_value_valid = true; + param_item->driverinit_default = init_val; devlink_param_notify(devlink, 0, param_item, DEVLINK_CMD_PARAM_NEW); } From b11d358bf8c306b18a86322d0d8d2549e971be97 Mon Sep 17 00:00:00 2001 From: Daniel Zahka Date: Tue, 18 Nov 2025 18:50:34 -0800 Subject: [PATCH 0995/1381] net/mlx5: implement swp_l4_csum_mode via devlink params swp_l4_csum_mode controls how L4 transmit checksums are computed when using Software Parser (SWP) hints for header locations. Supported values: 1. default: device will choose between full_csum or l4_only. Driver will discover the device's choice during initialization. 2. full_csum: calculate L4 checksum with the pseudo-header. 3. l4_only: calculate L4 checksum without the pseudo-header. Only available when swp_l4_csum_mode_l4_only is set in mlx5_ifc_nv_sw_offload_cap_bits. Note that 'default' might be returned from the device and passed to userspace, and it might also be set during a devlink_param::reset_default() call, but attempts to set a value of default directly with param-set will be rejected. The l4_only setting is a dependency for PSP initialization in mlx5e_psp_init(). Reviewed-by: Aleksandr Loktionov Signed-off-by: Daniel Zahka Link: https://patch.msgid.link/20251119025038.651131-5-daniel.zahka@gmail.com Signed-off-by: Jakub Kicinski --- Documentation/networking/devlink/mlx5.rst | 14 ++ .../net/ethernet/mellanox/mlx5/core/devlink.h | 3 +- .../mellanox/mlx5/core/lib/nv_param.c | 229 ++++++++++++++++++ 3 files changed, 245 insertions(+), 1 deletion(-) diff --git a/Documentation/networking/devlink/mlx5.rst b/Documentation/networking/devlink/mlx5.rst index 0e5f9c76e514..4bba4d780a4a 100644 --- a/Documentation/networking/devlink/mlx5.rst +++ b/Documentation/networking/devlink/mlx5.rst @@ -218,6 +218,20 @@ parameters. * ``balanced`` : Merges fewer CQEs, resulting in a moderate compression ratio but maintaining a balance between bandwidth savings and performance * ``aggressive`` : Merges more CQEs into a single entry, achieving a higher compression rate and maximizing performance, particularly under high traffic loads + * - ``swp_l4_csum_mode`` + - string + - permanent + - Configure how the L4 checksum is calculated by the device when using + Software Parser (SWP) hints for header locations. + + * ``default`` : Use the device's default checksum calculation + mode. The driver will discover during init whether or + full_csum or l4_only is in use. Setting this value explicitly + from userspace is not allowed, but some firmware versions may + return this value on param read. + * ``full_csum`` : Calculate full checksum including the pseudo-header + * ``l4_only`` : Calculate L4-only checksum, excluding the pseudo-header + The ``mlx5`` driver supports reloading via ``DEVLINK_CMD_RELOAD`` Info versions diff --git a/drivers/net/ethernet/mellanox/mlx5/core/devlink.h b/drivers/net/ethernet/mellanox/mlx5/core/devlink.h index c9555119a661..43b9bf8829cf 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/devlink.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/devlink.h @@ -26,7 +26,8 @@ enum mlx5_devlink_param_id { MLX5_DEVLINK_PARAM_ID_PCIE_CONG_IN_HIGH, MLX5_DEVLINK_PARAM_ID_PCIE_CONG_OUT_LOW, MLX5_DEVLINK_PARAM_ID_PCIE_CONG_OUT_HIGH, - MLX5_DEVLINK_PARAM_ID_CQE_COMPRESSION_TYPE + MLX5_DEVLINK_PARAM_ID_CQE_COMPRESSION_TYPE, + MLX5_DEVLINK_PARAM_ID_SWP_L4_CSUM_MODE, }; struct mlx5_trap_ctx { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/nv_param.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/nv_param.c index 70cb22fa96db..19bb620b7436 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lib/nv_param.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/nv_param.c @@ -8,6 +8,8 @@ enum { MLX5_CLASS_0_CTRL_ID_NV_GLOBAL_PCI_CONF = 0x80, MLX5_CLASS_0_CTRL_ID_NV_GLOBAL_PCI_CAP = 0x81, MLX5_CLASS_0_CTRL_ID_NV_SW_OFFLOAD_CONFIG = 0x10a, + MLX5_CLASS_0_CTRL_ID_NV_SW_OFFLOAD_CAP = 0x10b, + MLX5_CLASS_0_CTRL_ID_NV_SW_ACCELERATE_CONF = 0x11d, MLX5_CLASS_3_CTRL_ID_NV_PF_PCI_CONF = 0x80, }; @@ -32,6 +34,12 @@ union mlx5_ifc_config_item_type_auto_bits { u8 reserved_at_0[0x20]; }; +enum { + MLX5_ACCESS_MODE_NEXT = 0, + MLX5_ACCESS_MODE_CURRENT, + MLX5_ACCESS_MODE_DEFAULT, +}; + struct mlx5_ifc_config_item_bits { u8 valid[0x2]; u8 priority[0x2]; @@ -123,6 +131,17 @@ struct mlx5_ifc_nv_sw_offload_conf_bits { u8 lro_log_timeout0[0x4]; }; +struct mlx5_ifc_nv_sw_offload_cap_bits { + u8 reserved_at_0[0x19]; + u8 swp_l4_csum_mode_l4_only[0x1]; + u8 reserved_at_1a[0x6]; +}; + +struct mlx5_ifc_nv_sw_accelerate_conf_bits { + u8 swp_l4_csum_mode[0x2]; + u8 reserved_at_2[0x3e]; +}; + #define MNVDA_HDR_SZ \ (MLX5_ST_SZ_BYTES(mnvda_reg) - \ MLX5_BYTE_OFF(mnvda_reg, configuration_item_data)) @@ -195,6 +214,32 @@ mlx5_nv_param_read_sw_offload_conf(struct mlx5_core_dev *dev, void *mnvda, return mlx5_nv_param_read(dev, mnvda, len); } +static int +mlx5_nv_param_read_sw_offload_cap(struct mlx5_core_dev *dev, void *mnvda, + size_t len) +{ + MLX5_SET_CFG_ITEM_TYPE(global, mnvda, type_class, 0); + MLX5_SET_CFG_ITEM_TYPE(global, mnvda, parameter_index, + MLX5_CLASS_0_CTRL_ID_NV_SW_OFFLOAD_CAP); + MLX5_SET_CFG_HDR_LEN(mnvda, nv_sw_offload_cap); + + return mlx5_nv_param_read(dev, mnvda, len); +} + +static int +mlx5_nv_param_read_sw_accelerate_conf(struct mlx5_core_dev *dev, void *mnvda, + size_t len, int access_mode) +{ + MLX5_SET_CFG_ITEM_TYPE(global, mnvda, type_class, 0); + MLX5_SET_CFG_ITEM_TYPE(global, mnvda, parameter_index, + MLX5_CLASS_0_CTRL_ID_NV_SW_ACCELERATE_CONF); + MLX5_SET_CFG_HDR_LEN(mnvda, nv_sw_accelerate_conf); + MLX5_SET(mnvda_reg, mnvda, configuration_item_header.access_mode, + access_mode); + + return mlx5_nv_param_read(dev, mnvda, len); +} + static const char *const cqe_compress_str[] = { "balanced", "aggressive" }; @@ -269,6 +314,182 @@ mlx5_nv_param_devlink_cqe_compress_set(struct devlink *devlink, u32 id, return mlx5_nv_param_write(dev, mnvda, sizeof(mnvda)); } +enum swp_l4_csum_mode { + SWP_L4_CSUM_MODE_DEFAULT = 0, + SWP_L4_CSUM_MODE_FULL_CSUM = 1, + SWP_L4_CSUM_MODE_L4_ONLY = 2, +}; + +static const char *const + swp_l4_csum_mode_str[] = { "default", "full_csum", "l4_only" }; + +static int +mlx5_swp_l4_csum_mode_get(struct devlink *devlink, u32 id, + int access_mode, u8 *value, + struct netlink_ext_ack *extack) +{ + struct mlx5_core_dev *dev = devlink_priv(devlink); + u32 mnvda[MLX5_ST_SZ_DW(mnvda_reg)] = {}; + void *data; + int err; + + err = mlx5_nv_param_read_sw_accelerate_conf(dev, mnvda, sizeof(mnvda), + access_mode); + if (err) { + NL_SET_ERR_MSG_MOD(extack, + "Failed to read sw_accelerate_conf mnvda reg"); + return err; + } + + data = MLX5_ADDR_OF(mnvda_reg, mnvda, configuration_item_data); + *value = MLX5_GET(nv_sw_accelerate_conf, data, swp_l4_csum_mode); + + if (*value >= ARRAY_SIZE(swp_l4_csum_mode_str)) { + NL_SET_ERR_MSG_FMT_MOD(extack, + "Invalid swp_l4_csum_mode value %u read from device", + *value); + return -EINVAL; + } + + return 0; +} + +static int +mlx5_devlink_swp_l4_csum_mode_get(struct devlink *devlink, u32 id, + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) +{ + u8 value; + int err; + + err = mlx5_swp_l4_csum_mode_get(devlink, id, MLX5_ACCESS_MODE_NEXT, + &value, extack); + if (err) + return err; + + strscpy(ctx->val.vstr, swp_l4_csum_mode_str[value], + sizeof(ctx->val.vstr)); + return 0; +} + +static int +mlx5_devlink_swp_l4_csum_mode_validate(struct devlink *devlink, u32 id, + union devlink_param_value val, + struct netlink_ext_ack *extack) +{ + struct mlx5_core_dev *dev = devlink_priv(devlink); + u32 cap[MLX5_ST_SZ_DW(mnvda_reg)] = {}; + void *data; + int err, i; + + for (i = 0; i < ARRAY_SIZE(swp_l4_csum_mode_str); i++) { + if (!strcmp(val.vstr, swp_l4_csum_mode_str[i])) + break; + } + + if (i >= ARRAY_SIZE(swp_l4_csum_mode_str) || + i == SWP_L4_CSUM_MODE_DEFAULT) { + NL_SET_ERR_MSG_MOD(extack, + "Invalid value, supported values are full_csum/l4_only"); + return -EINVAL; + } + + if (i == SWP_L4_CSUM_MODE_L4_ONLY) { + err = mlx5_nv_param_read_sw_offload_cap(dev, cap, sizeof(cap)); + if (err) { + NL_SET_ERR_MSG_MOD(extack, + "Failed to read sw_offload_cap"); + return err; + } + + data = MLX5_ADDR_OF(mnvda_reg, cap, configuration_item_data); + if (!MLX5_GET(nv_sw_offload_cap, data, swp_l4_csum_mode_l4_only)) { + NL_SET_ERR_MSG_MOD(extack, + "l4_only mode is not supported on this device"); + return -EOPNOTSUPP; + } + } + + return 0; +} + +static int +mlx5_swp_l4_csum_mode_set(struct devlink *devlink, u32 id, u8 value, + struct netlink_ext_ack *extack) +{ + struct mlx5_core_dev *dev = devlink_priv(devlink); + u32 mnvda[MLX5_ST_SZ_DW(mnvda_reg)] = {}; + void *data; + int err; + + err = mlx5_nv_param_read_sw_accelerate_conf(dev, mnvda, sizeof(mnvda), + MLX5_ACCESS_MODE_NEXT); + if (err) { + NL_SET_ERR_MSG_MOD(extack, + "Failed to read sw_accelerate_conf mnvda reg"); + return err; + } + + data = MLX5_ADDR_OF(mnvda_reg, mnvda, configuration_item_data); + MLX5_SET(nv_sw_accelerate_conf, data, swp_l4_csum_mode, value); + + err = mlx5_nv_param_write(dev, mnvda, sizeof(mnvda)); + if (err) + NL_SET_ERR_MSG_MOD(extack, + "Failed to write sw_accelerate_conf mnvda reg"); + + return err; +} + +static int +mlx5_devlink_swp_l4_csum_mode_set(struct devlink *devlink, u32 id, + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) +{ + u8 value; + + if (!strcmp(ctx->val.vstr, "full_csum")) + value = SWP_L4_CSUM_MODE_FULL_CSUM; + else + value = SWP_L4_CSUM_MODE_L4_ONLY; + + return mlx5_swp_l4_csum_mode_set(devlink, id, value, extack); +} + +static int +mlx5_devlink_swp_l4_csum_mode_get_default(struct devlink *devlink, u32 id, + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) +{ + u8 value; + int err; + + err = mlx5_swp_l4_csum_mode_get(devlink, id, MLX5_ACCESS_MODE_DEFAULT, + &value, extack); + if (err) + return err; + + strscpy(ctx->val.vstr, swp_l4_csum_mode_str[value], + sizeof(ctx->val.vstr)); + return 0; +} + +static int +mlx5_devlink_swp_l4_csum_mode_set_default(struct devlink *devlink, u32 id, + enum devlink_param_cmode cmode, + struct netlink_ext_ack *extack) +{ + u8 value; + int err; + + err = mlx5_swp_l4_csum_mode_get(devlink, id, MLX5_ACCESS_MODE_DEFAULT, + &value, extack); + if (err) + return err; + + return mlx5_swp_l4_csum_mode_set(devlink, id, value, extack); +} + static int mlx5_nv_param_read_global_pci_conf(struct mlx5_core_dev *dev, void *mnvda, size_t len) { @@ -548,6 +769,14 @@ static const struct devlink_param mlx5_nv_param_devlink_params[] = { mlx5_nv_param_devlink_cqe_compress_get, mlx5_nv_param_devlink_cqe_compress_set, mlx5_nv_param_devlink_cqe_compress_validate), + DEVLINK_PARAM_DRIVER_WITH_DEFAULTS(MLX5_DEVLINK_PARAM_ID_SWP_L4_CSUM_MODE, + "swp_l4_csum_mode", DEVLINK_PARAM_TYPE_STRING, + BIT(DEVLINK_PARAM_CMODE_PERMANENT), + mlx5_devlink_swp_l4_csum_mode_get, + mlx5_devlink_swp_l4_csum_mode_set, + mlx5_devlink_swp_l4_csum_mode_validate, + mlx5_devlink_swp_l4_csum_mode_get_default, + mlx5_devlink_swp_l4_csum_mode_set_default), }; int mlx5_nv_param_register_dl_params(struct devlink *devlink) From 72924056ebacbfcdc8de7a1f7e8769ce2198a987 Mon Sep 17 00:00:00 2001 From: Daniel Zahka Date: Tue, 18 Nov 2025 18:50:35 -0800 Subject: [PATCH 0996/1381] netdevsim: register a new devlink param with default value interface Create a new devlink param, test2, that supports default param actions via the devlink_param::get_default() and devlink_param::reset_default() functions. Reviewed-by: Aleksandr Loktionov Signed-off-by: Daniel Zahka Link: https://patch.msgid.link/20251119025038.651131-6-daniel.zahka@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/netdevsim/dev.c | 56 +++++++++++++++++++++++++++++++ drivers/net/netdevsim/netdevsim.h | 1 + 2 files changed, 57 insertions(+) diff --git a/drivers/net/netdevsim/dev.c b/drivers/net/netdevsim/dev.c index 95f66c1f59db..2683a989873e 100644 --- a/drivers/net/netdevsim/dev.c +++ b/drivers/net/netdevsim/dev.c @@ -320,6 +320,8 @@ static int nsim_dev_debugfs_init(struct nsim_dev *nsim_dev) &nsim_dev->max_macs); debugfs_create_bool("test1", 0600, nsim_dev->ddir, &nsim_dev->test1); + debugfs_create_u32("test2", 0600, nsim_dev->ddir, + &nsim_dev->test2); nsim_dev->take_snapshot = debugfs_create_file("take_snapshot", 0200, nsim_dev->ddir, @@ -521,8 +523,53 @@ static int nsim_dev_resources_register(struct devlink *devlink) enum nsim_devlink_param_id { NSIM_DEVLINK_PARAM_ID_BASE = DEVLINK_PARAM_GENERIC_ID_MAX, NSIM_DEVLINK_PARAM_ID_TEST1, + NSIM_DEVLINK_PARAM_ID_TEST2, }; +static int +nsim_devlink_param_test2_get(struct devlink *devlink, u32 id, + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) +{ + struct nsim_dev *nsim_dev = devlink_priv(devlink); + + ctx->val.vu32 = nsim_dev->test2; + return 0; +} + +static int +nsim_devlink_param_test2_set(struct devlink *devlink, u32 id, + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) +{ + struct nsim_dev *nsim_dev = devlink_priv(devlink); + + nsim_dev->test2 = ctx->val.vu32; + return 0; +} + +#define NSIM_DEV_TEST2_DEFAULT 1234 + +static int +nsim_devlink_param_test2_get_default(struct devlink *devlink, u32 id, + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) +{ + ctx->val.vu32 = NSIM_DEV_TEST2_DEFAULT; + return 0; +} + +static int +nsim_devlink_param_test2_reset_default(struct devlink *devlink, u32 id, + enum devlink_param_cmode cmode, + struct netlink_ext_ack *extack) +{ + struct nsim_dev *nsim_dev = devlink_priv(devlink); + + nsim_dev->test2 = NSIM_DEV_TEST2_DEFAULT; + return 0; +} + static const struct devlink_param nsim_devlink_params[] = { DEVLINK_PARAM_GENERIC(MAX_MACS, BIT(DEVLINK_PARAM_CMODE_DRIVERINIT), @@ -531,6 +578,14 @@ static const struct devlink_param nsim_devlink_params[] = { "test1", DEVLINK_PARAM_TYPE_BOOL, BIT(DEVLINK_PARAM_CMODE_DRIVERINIT), NULL, NULL, NULL), + DEVLINK_PARAM_DRIVER_WITH_DEFAULTS(NSIM_DEVLINK_PARAM_ID_TEST2, + "test2", DEVLINK_PARAM_TYPE_U32, + BIT(DEVLINK_PARAM_CMODE_RUNTIME), + nsim_devlink_param_test2_get, + nsim_devlink_param_test2_set, + NULL, + nsim_devlink_param_test2_get_default, + nsim_devlink_param_test2_reset_default), }; static void nsim_devlink_set_params_init_values(struct nsim_dev *nsim_dev, @@ -1590,6 +1645,7 @@ int nsim_drv_probe(struct nsim_bus_dev *nsim_bus_dev) nsim_dev->fw_update_flash_chunk_time_ms = NSIM_DEV_FLASH_CHUNK_TIME_MS_DEFAULT; nsim_dev->max_macs = NSIM_DEV_MAX_MACS_DEFAULT; nsim_dev->test1 = NSIM_DEV_TEST1_DEFAULT; + nsim_dev->test2 = NSIM_DEV_TEST2_DEFAULT; spin_lock_init(&nsim_dev->fa_cookie_lock); dev_set_drvdata(&nsim_bus_dev->dev, nsim_dev); diff --git a/drivers/net/netdevsim/netdevsim.h b/drivers/net/netdevsim/netdevsim.h index af6fcfcda8ba..d1a941e2b18f 100644 --- a/drivers/net/netdevsim/netdevsim.h +++ b/drivers/net/netdevsim/netdevsim.h @@ -331,6 +331,7 @@ struct nsim_dev { u32 fw_update_flash_chunk_time_ms; u32 max_macs; bool test1; + u32 test2; bool dont_allow_reload; bool fail_reload; struct devlink_region *dummy_region; From 8be656cfb931f79b3971407622422c62575abf90 Mon Sep 17 00:00:00 2001 From: Daniel Zahka Date: Tue, 18 Nov 2025 18:50:36 -0800 Subject: [PATCH 0997/1381] selftest: netdevsim: test devlink default params Test querying default values and resetting to default values for netdevsim devlink params. This should cover the basic paths of interest: driverinit and non-driverinit cmodes, as well as bool and non-bool value type. Default param values of type bool are encoded with u8 netlink type as opposed to flag type, so that userspace can distinguish "not-present" from false. Signed-off-by: Daniel Zahka Link: https://patch.msgid.link/20251119025038.651131-7-daniel.zahka@gmail.com Signed-off-by: Jakub Kicinski --- .../drivers/net/netdevsim/devlink.sh | 116 +++++++++++++++++- 1 file changed, 110 insertions(+), 6 deletions(-) diff --git a/tools/testing/selftests/drivers/net/netdevsim/devlink.sh b/tools/testing/selftests/drivers/net/netdevsim/devlink.sh index 030762b203d7..1b529ccaf050 100755 --- a/tools/testing/selftests/drivers/net/netdevsim/devlink.sh +++ b/tools/testing/selftests/drivers/net/netdevsim/devlink.sh @@ -3,7 +3,8 @@ lib_dir=$(dirname $0)/../../../net/forwarding -ALL_TESTS="fw_flash_test params_test regions_test reload_test \ +ALL_TESTS="fw_flash_test params_test \ + params_default_test regions_test reload_test \ netns_reload_test resource_test dev_info_test \ empty_reporter_test dummy_reporter_test rate_test" NUM_NETIFS=0 @@ -78,17 +79,28 @@ fw_flash_test() param_get() { local name=$1 + local attr=${2:-value} + local cmode=${3:-driverinit} cmd_jq "devlink dev param show $DL_HANDLE name $name -j" \ - '.[][][].values[] | select(.cmode == "driverinit").value' + '.[][][].values[] | select(.cmode == "'"$cmode"'").'"$attr" } param_set() { local name=$1 local value=$2 + local cmode=${3:-driverinit} - devlink dev param set $DL_HANDLE name $name cmode driverinit value $value + devlink dev param set $DL_HANDLE name $name cmode $cmode value $value +} + +param_set_default() +{ + local name=$1 + local cmode=${2:-driverinit} + + devlink dev param set $DL_HANDLE name $name default cmode $cmode } check_value() @@ -97,12 +109,18 @@ check_value() local phase_name=$2 local expected_param_value=$3 local expected_debugfs_value=$4 + local cmode=${5:-driverinit} local value + local attr="value" - value=$(param_get $name) - check_err $? "Failed to get $name param value" + if [[ "$phase_name" == *"default"* ]]; then + attr="default" + fi + + value=$(param_get $name $attr $cmode) + check_err $? "Failed to get $name param $attr" [ "$value" == "$expected_param_value" ] - check_err $? "Unexpected $phase_name $name param value" + check_err $? "Unexpected $phase_name $name param $attr" value=$(<$DEBUGFS_DIR/$name) check_err $? "Failed to get $name debugfs value" [ "$value" == "$expected_debugfs_value" ] @@ -135,6 +153,92 @@ params_test() log_test "params test" } +value_to_debugfs() +{ + local value=$1 + + case "$value" in + true) + echo "Y" + ;; + false) + echo "N" + ;; + *) + echo "$value" + ;; + esac +} + +test_default() +{ + local param_name=$1 + local new_value=$2 + local expected_default=$3 + local cmode=${4:-driverinit} + local default_debugfs + local new_debugfs + local expected_debugfs + + default_debugfs=$(value_to_debugfs $expected_default) + new_debugfs=$(value_to_debugfs $new_value) + + expected_debugfs=$default_debugfs + check_value $param_name initial-default $expected_default $expected_debugfs $cmode + + param_set $param_name $new_value $cmode + check_err $? "Failed to set $param_name to $new_value" + + expected_debugfs=$([ "$cmode" == "runtime" ] && echo "$new_debugfs" || echo "$default_debugfs") + check_value $param_name post-set $new_value $expected_debugfs $cmode + + devlink dev reload $DL_HANDLE + check_err $? "Failed to reload device" + + expected_debugfs=$new_debugfs + check_value $param_name post-reload-new-value $new_value $expected_debugfs $cmode + + param_set_default $param_name $cmode + check_err $? "Failed to set $param_name to default" + + expected_debugfs=$([ "$cmode" == "runtime" ] && echo "$default_debugfs" || echo "$new_debugfs") + check_value $param_name post-set-default $expected_default $expected_debugfs $cmode + + devlink dev reload $DL_HANDLE + check_err $? "Failed to reload device" + + expected_debugfs=$default_debugfs + check_value $param_name post-reload-default $expected_default $expected_debugfs $cmode +} + +params_default_test() +{ + RET=0 + + if ! devlink dev param help 2>&1 | grep -q "value VALUE | default"; then + echo "SKIP: devlink cli missing default feature" + return + fi + + # Remove side effects of previous tests. Use plain param_set, because + # param_set_default is a feature under test here. + param_set max_macs 32 driverinit + check_err $? "Failed to reset max_macs to default value" + param_set test1 true driverinit + check_err $? "Failed to reset test1 to default value" + param_set test2 1234 runtime + check_err $? "Failed to reset test2 to default value" + + devlink dev reload $DL_HANDLE + check_err $? "Failed to reload device for clean state" + + test_default max_macs 16 32 driverinit + test_default test1 false true driverinit + test_default test2 100 1234 runtime + + log_test "params default test" +} + check_region_size() { local name=$1 From b647d2574e4583c2e3b0ab35568f60c88e910840 Mon Sep 17 00:00:00 2001 From: Seungjin Bae Date: Mon, 17 Nov 2025 20:32:59 -0500 Subject: [PATCH 0998/1381] wifi: rtl818x: rtl8187: Fix potential buffer underflow in rtl8187_rx_cb() The rtl8187_rx_cb() calculates the rx descriptor header address by subtracting its size from the skb tail pointer. However, it does not validate if the received packet (skb->len from urb->actual_length) is large enough to contain this header. If a truncated packet is received, this will lead to a buffer underflow, reading memory before the start of the skb data area, and causing a kernel panic. Add length checks for both rtl8187 and rtl8187b descriptor headers before attempting to access them, dropping the packet cleanly if the check fails. Fixes: 6f7853f3cbe4 ("rtl8187: change rtl8187_dev.c to support RTL8187B (part 2)") Signed-off-by: Seungjin Bae Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251118013258.1789949-2-eeodqql09@gmail.com --- .../wireless/realtek/rtl818x/rtl8187/dev.c | 27 +++++++++++++------ 1 file changed, 19 insertions(+), 8 deletions(-) diff --git a/drivers/net/wireless/realtek/rtl818x/rtl8187/dev.c b/drivers/net/wireless/realtek/rtl818x/rtl8187/dev.c index 0c5c66401daa..7aa2da0cd63c 100644 --- a/drivers/net/wireless/realtek/rtl818x/rtl8187/dev.c +++ b/drivers/net/wireless/realtek/rtl818x/rtl8187/dev.c @@ -338,14 +338,16 @@ static void rtl8187_rx_cb(struct urb *urb) spin_unlock_irqrestore(&priv->rx_queue.lock, f); skb_put(skb, urb->actual_length); - if (unlikely(urb->status)) { - dev_kfree_skb_irq(skb); - return; - } + if (unlikely(urb->status)) + goto free_skb; if (!priv->is_rtl8187b) { - struct rtl8187_rx_hdr *hdr = - (typeof(hdr))(skb_tail_pointer(skb) - sizeof(*hdr)); + struct rtl8187_rx_hdr *hdr; + + if (skb->len < sizeof(struct rtl8187_rx_hdr)) + goto free_skb; + + hdr = (typeof(hdr))(skb_tail_pointer(skb) - sizeof(*hdr)); flags = le32_to_cpu(hdr->flags); /* As with the RTL8187B below, the AGC is used to calculate * signal strength. In this case, the scaling @@ -355,8 +357,12 @@ static void rtl8187_rx_cb(struct urb *urb) rx_status.antenna = (hdr->signal >> 7) & 1; rx_status.mactime = le64_to_cpu(hdr->mac_time); } else { - struct rtl8187b_rx_hdr *hdr = - (typeof(hdr))(skb_tail_pointer(skb) - sizeof(*hdr)); + struct rtl8187b_rx_hdr *hdr; + + if (skb->len < sizeof(struct rtl8187b_rx_hdr)) + goto free_skb; + + hdr = (typeof(hdr))(skb_tail_pointer(skb) - sizeof(*hdr)); /* The Realtek datasheet for the RTL8187B shows that the RX * header contains the following quantities: signal quality, * RSSI, AGC, the received power in dB, and the measured SNR. @@ -409,6 +415,11 @@ static void rtl8187_rx_cb(struct urb *urb) skb_unlink(skb, &priv->rx_queue); dev_kfree_skb_irq(skb); } + return; + +free_skb: + dev_kfree_skb_irq(skb); + return; } static int rtl8187_init_urbs(struct ieee80211_hw *dev) From 7cbec00dc731e9ddb3cdb350d05cbc26d45b3273 Mon Sep 17 00:00:00 2001 From: Marco Crivellari Date: Tue, 18 Nov 2025 11:20:31 +0100 Subject: [PATCH 0999/1381] wifi: rtlwifi: add WQ_UNBOUND to alloc_workqueue users MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Currently if a user enqueues a work item using schedule_delayed_work() the used wq is "system_wq" (per-cpu wq) while queue_delayed_work() use WORK_CPU_UNBOUND (used when a cpu is not specified). The same applies to schedule_work() that is using system_wq and queue_work(), that makes use again of WORK_CPU_UNBOUND. This lack of consistency cannot be addressed without refactoring the API. For more details see the Link tag below. alloc_workqueue() treats all queues as per-CPU by default, while unbound workqueues must opt-in via WQ_UNBOUND. This default is suboptimal: most workloads benefit from unbound queues, allowing the scheduler to place worker threads where they’re needed and reducing noise when CPUs are isolated. This continues the effort to refactor workqueue APIs, which began with the introduction of new workqueues and a new alloc_workqueue flag in: commit 128ea9f6ccfb ("workqueue: Add system_percpu_wq and system_dfl_wq") commit 930c2ea566af ("workqueue: Add new WQ_PERCPU flag") This change adds the WQ_UNBOUND flag to explicitly request alloc_workqueue() to be unbound, because this specific workload has no benefit being per-cpu. With the introduction of the WQ_PERCPU flag (equivalent to !WQ_UNBOUND), any alloc_workqueue() caller that doesn’t explicitly specify WQ_UNBOUND must now use WQ_PERCPU. Once migration is complete, WQ_UNBOUND can be removed and unbound will become the implicit default. Suggested-by: Tejun Heo Signed-off-by: Marco Crivellari Link: https://lore.kernel.org/all/20250221112003.1dSuoGyc@linutronix.de/ Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251118102032.54375-2-marco.crivellari@suse.com --- drivers/net/wireless/realtek/rtlwifi/base.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/wireless/realtek/rtlwifi/base.c b/drivers/net/wireless/realtek/rtlwifi/base.c index e26feb8de658..09e5a16d7252 100644 --- a/drivers/net/wireless/realtek/rtlwifi/base.c +++ b/drivers/net/wireless/realtek/rtlwifi/base.c @@ -445,7 +445,7 @@ static int _rtl_init_deferred_work(struct ieee80211_hw *hw) struct rtl_priv *rtlpriv = rtl_priv(hw); struct workqueue_struct *wq; - wq = alloc_workqueue("%s", 0, 0, rtlpriv->cfg->name); + wq = alloc_workqueue("%s", WQ_UNBOUND, 0, rtlpriv->cfg->name); if (!wq) return -ENOMEM; From 9c194fe4625db18f93d5abcfb7f7997557a0b29d Mon Sep 17 00:00:00 2001 From: Marco Crivellari Date: Tue, 18 Nov 2025 11:20:32 +0100 Subject: [PATCH 1000/1381] wifi: rtw88: add WQ_UNBOUND to alloc_workqueue users MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Currently if a user enqueues a work item using schedule_delayed_work() the used wq is "system_wq" (per-cpu wq) while queue_delayed_work() use WORK_CPU_UNBOUND (used when a cpu is not specified). The same applies to schedule_work() that is using system_wq and queue_work(), that makes use again of WORK_CPU_UNBOUND. This lack of consistency cannot be addressed without refactoring the API. For more details see the Link tag below. alloc_workqueue() treats all queues as per-CPU by default, while unbound workqueues must opt-in via WQ_UNBOUND. This default is suboptimal: most workloads benefit from unbound queues, allowing the scheduler to place worker threads where they’re needed and reducing noise when CPUs are isolated. This continues the effort to refactor workqueue APIs, which began with the introduction of new workqueues and a new alloc_workqueue flag in: commit 128ea9f6ccfb ("workqueue: Add system_percpu_wq and system_dfl_wq") commit 930c2ea566af ("workqueue: Add new WQ_PERCPU flag") This change adds the WQ_UNBOUND flag to explicitly request alloc_workqueue() to be unbound, because this specific workload has no benefit being per-cpu. With the introduction of the WQ_PERCPU flag (equivalent to !WQ_UNBOUND), any alloc_workqueue() caller that doesn’t explicitly specify WQ_UNBOUND must now use WQ_PERCPU. Once migration is complete, WQ_UNBOUND can be removed and unbound will become the implicit default. Suggested-by: Tejun Heo Signed-off-by: Marco Crivellari Link: https://lore.kernel.org/all/20250221112003.1dSuoGyc@linutronix.de/ Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251118102032.54375-3-marco.crivellari@suse.com --- drivers/net/wireless/realtek/rtw88/usb.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/realtek/rtw88/usb.c b/drivers/net/wireless/realtek/rtw88/usb.c index 3b5126ffc81a..009202c627d2 100644 --- a/drivers/net/wireless/realtek/rtw88/usb.c +++ b/drivers/net/wireless/realtek/rtw88/usb.c @@ -965,7 +965,8 @@ static int rtw_usb_init_rx(struct rtw_dev *rtwdev) struct sk_buff *rx_skb; int i; - rtwusb->rxwq = alloc_workqueue("rtw88_usb: rx wq", WQ_BH, 0); + rtwusb->rxwq = alloc_workqueue("rtw88_usb: rx wq", WQ_BH | WQ_UNBOUND, + 0); if (!rtwusb->rxwq) { rtw_err(rtwdev, "failed to create RX work queue\n"); return -ENOMEM; From 2a2aae3655343b753dfef86b341be99f7273f2ba Mon Sep 17 00:00:00 2001 From: Ping-Ke Shih Date: Thu, 20 Nov 2025 11:10:43 +0800 Subject: [PATCH 1001/1381] wifi: rtw89: 8852a: correct field mask of reset DAC/ADC FIFO The field mask should be bits 16-31, but suddenly use wrong bits 24-31, rarely causing a little performance degraded if DAC/DAC FIFO stays on an unexpected state. Found this by Geert who works on bit field functions. Reported-by: Geert Uytterhoeven Closes: https://lore.kernel.org/linux-wireless/CAMuHMdVt+5yOA6tuasX4KQgZud5wtRwu0A15UkEfQJbcd_xvVw@mail.gmail.com/ Cc: Geert Uytterhoeven Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/20251120031044.12493-2-pkshih@realtek.com --- drivers/net/wireless/realtek/rtw89/rtw8852a_rfk.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852a_rfk.c b/drivers/net/wireless/realtek/rtw89/rtw8852a_rfk.c index e74257d19412..463399413318 100644 --- a/drivers/net/wireless/realtek/rtw89/rtw8852a_rfk.c +++ b/drivers/net/wireless/realtek/rtw89/rtw8852a_rfk.c @@ -756,8 +756,8 @@ static void _iqk_rxk_setting(struct rtw89_dev *rtwdev, u8 path) rtw89_phy_write32_mask(rtwdev, R_ANAPAR, B_ANAPAR_FLTRST, 0x1); rtw89_phy_write32_mask(rtwdev, R_ANAPAR_PW15, B_ANAPAR_PW15_H2, 0x0); udelay(1); - rtw89_phy_write32_mask(rtwdev, R_ADC_FIFO, B_ADC_FIFO_RST, 0x0303); - rtw89_phy_write32_mask(rtwdev, R_ADC_FIFO, B_ADC_FIFO_RST, 0x0000); + rtw89_phy_write32_mask(rtwdev, R_ADC_FIFO, B_ADC_FIFO_RXK, 0x0303); + rtw89_phy_write32_mask(rtwdev, R_ADC_FIFO, B_ADC_FIFO_RXK, 0x0000); switch (iqk_info->iqk_band[path]) { case RTW89_BAND_2G: @@ -1239,8 +1239,8 @@ static void _iqk_txk_setting(struct rtw89_dev *rtwdev, u8 path) udelay(1); rtw89_phy_write32_mask(rtwdev, R_ANAPAR, B_ANAPAR_15, 0x0041); udelay(1); - rtw89_phy_write32_mask(rtwdev, R_ADC_FIFO, B_ADC_FIFO_RST, 0x0303); - rtw89_phy_write32_mask(rtwdev, R_ADC_FIFO, B_ADC_FIFO_RST, 0x0000); + rtw89_phy_write32_mask(rtwdev, R_ADC_FIFO, B_ADC_FIFO_RXK, 0x0303); + rtw89_phy_write32_mask(rtwdev, R_ADC_FIFO, B_ADC_FIFO_RXK, 0x0000); switch (iqk_info->iqk_band[path]) { case RTW89_BAND_2G: rtw89_write_rf(rtwdev, path, RR_XALNA2, RR_XALNA2_SW, 0x00); From 5511ba3de434892e5ef3594d6eabbd12b1629356 Mon Sep 17 00:00:00 2001 From: Bitterblue Smith Date: Thu, 20 Nov 2025 16:10:01 +0200 Subject: [PATCH 1002/1381] wifi: rtl8xxxu: Fix HT40 channel config for RTL8192CU, RTL8723AU Flip the response rate subchannel. It was backwards, causing low speeds when using 40 MHz channel width. "iw dev ... station dump" showed a low RX rate, 11M or less. Also fix the channel width field of RF6052_REG_MODE_AG. Tested only with RTL8192CU, but these settings are identical for RTL8723AU. Signed-off-by: Bitterblue Smith Reviewed-by: Ping-Ke Shih Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/1f46571d-855b-43e1-8bfc-abacceb96043@gmail.com --- drivers/net/wireless/realtek/rtl8xxxu/core.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/drivers/net/wireless/realtek/rtl8xxxu/core.c b/drivers/net/wireless/realtek/rtl8xxxu/core.c index d82cb7714314..be8ee6c30034 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/core.c +++ b/drivers/net/wireless/realtek/rtl8xxxu/core.c @@ -1252,7 +1252,7 @@ void rtl8xxxu_gen1_config_channel(struct ieee80211_hw *hw) opmode &= ~BW_OPMODE_20MHZ; rtl8xxxu_write8(priv, REG_BW_OPMODE, opmode); rsr &= ~RSR_RSC_BANDWIDTH_40M; - if (sec_ch_above) + if (!sec_ch_above) rsr |= RSR_RSC_UPPER_SUB_CHANNEL; else rsr |= RSR_RSC_LOWER_SUB_CHANNEL; @@ -1321,9 +1321,8 @@ void rtl8xxxu_gen1_config_channel(struct ieee80211_hw *hw) for (i = RF_A; i < priv->rf_paths; i++) { val32 = rtl8xxxu_read_rfreg(priv, i, RF6052_REG_MODE_AG); - if (hw->conf.chandef.width == NL80211_CHAN_WIDTH_40) - val32 &= ~MODE_AG_CHANNEL_20MHZ; - else + val32 &= ~MODE_AG_BW_MASK; + if (hw->conf.chandef.width != NL80211_CHAN_WIDTH_40) val32 |= MODE_AG_CHANNEL_20MHZ; rtl8xxxu_write_rfreg(priv, i, RF6052_REG_MODE_AG, val32); } From bdb4c850c368d4009a0b109b6135042540aca749 Mon Sep 17 00:00:00 2001 From: Bitterblue Smith Date: Thu, 20 Nov 2025 16:11:58 +0200 Subject: [PATCH 1003/1381] wifi: rtl8xxxu: Make RTL8192CU, RTL8723AU TX with 40 MHz width Set the required fields in the TX descriptor to allow these chips to transmit with 40 MHz channel width when the access point supports it. Tested only with RTL8192CU, but these settings are identical for RTL8723AU. Signed-off-by: Bitterblue Smith Reviewed-by: Ping-Ke Shih Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/30d95228-69b2-48f9-8854-c98d2408c4d3@gmail.com --- drivers/net/wireless/realtek/rtl8xxxu/core.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/realtek/rtl8xxxu/core.c b/drivers/net/wireless/realtek/rtl8xxxu/core.c index be8ee6c30034..7700e4074dc3 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/core.c +++ b/drivers/net/wireless/realtek/rtl8xxxu/core.c @@ -5221,9 +5221,19 @@ rtl8xxxu_fill_txdesc_v1(struct ieee80211_hw *hw, struct ieee80211_hdr *hdr, tx_desc->txdw5 |= cpu_to_le32(TXDESC32_RETRY_LIMIT_ENABLE); } - if (ieee80211_is_data_qos(hdr->frame_control)) + if (ieee80211_is_data_qos(hdr->frame_control)) { tx_desc->txdw4 |= cpu_to_le32(TXDESC32_QOS); + if (conf_is_ht40(&hw->conf)) { + tx_desc->txdw4 |= cpu_to_le32(TXDESC_DATA_BW); + + if (conf_is_ht40_minus(&hw->conf)) + tx_desc->txdw4 |= cpu_to_le32(TXDESC_PRIME_CH_OFF_UPPER); + else + tx_desc->txdw4 |= cpu_to_le32(TXDESC_PRIME_CH_OFF_LOWER); + } + } + if (short_preamble) tx_desc->txdw4 |= cpu_to_le32(TXDESC32_SHORT_PREAMBLE); From fc44314a377ace14c76b7502cf0d785eb35e5cc4 Mon Sep 17 00:00:00 2001 From: Bitterblue Smith Date: Thu, 20 Nov 2025 16:12:35 +0200 Subject: [PATCH 1004/1381] wifi: rtl8xxxu: Fix the 40 MHz subchannel for RTL8192EU, RTL8723BU rtl8xxxu_gen2_config_channel() was missing the subchannel setting. This function is used by RTL8192EU and RTL8723BU. This change seems to make no difference in my testing on channel 13 with either chip. Signed-off-by: Bitterblue Smith Reviewed-by: Ping-Ke Shih Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/a5de8d39-45c1-4667-ab4c-7109de6eb13d@gmail.com --- drivers/net/wireless/realtek/rtl8xxxu/core.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/wireless/realtek/rtl8xxxu/core.c b/drivers/net/wireless/realtek/rtl8xxxu/core.c index 7700e4074dc3..739456640ca8 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/core.c +++ b/drivers/net/wireless/realtek/rtl8xxxu/core.c @@ -1373,9 +1373,11 @@ void rtl8xxxu_gen2_config_channel(struct ieee80211_hw *hw) hw->conf.chandef.chan->center_freq) { sec_ch_above = 1; channel += 2; + subchannel = 2; } else { sec_ch_above = 0; channel -= 2; + subchannel = 1; } val32 = rtl8xxxu_read32(priv, REG_FPGA0_RF_MODE); From 41a21d0ff3a0b05de78f9cbeb29945d0eb18102a Mon Sep 17 00:00:00 2001 From: Bitterblue Smith Date: Thu, 20 Nov 2025 16:13:17 +0200 Subject: [PATCH 1005/1381] wifi: rtl8xxxu: Fix RX channel width reported by RTL8192FU The other chips report the RX channel width in the RX descriptor, but this one doesn't. Get the RX channel width from the PHY status. Signed-off-by: Bitterblue Smith Reviewed-by: Ping-Ke Shih Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/1c6c1fd4-92f6-4327-a24e-f0747ab21819@gmail.com --- drivers/net/wireless/realtek/rtl8xxxu/core.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/realtek/rtl8xxxu/core.c b/drivers/net/wireless/realtek/rtl8xxxu/core.c index 739456640ca8..a18f73aea837 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/core.c +++ b/drivers/net/wireless/realtek/rtl8xxxu/core.c @@ -5701,7 +5701,7 @@ static void jaguar2_rx_parse_phystats_type1(struct rtl8xxxu_priv *priv, !rtl8xxxu_is_sta_sta(priv) && (rtl8xxxu_is_packet_match_bssid(priv, hdr, 0) || rtl8xxxu_is_packet_match_bssid(priv, hdr, 1)); - u8 pwdb_max = 0; + u8 pwdb_max = 0, rxsc; int rx_path; if (parse_cfo) { @@ -5716,6 +5716,16 @@ static void jaguar2_rx_parse_phystats_type1(struct rtl8xxxu_priv *priv, pwdb_max = max(pwdb_max, phy_stats1->pwdb[rx_path]); rx_status->signal = pwdb_max - 110; + + if (rxmcs >= DESC_RATE_6M && rxmcs <= DESC_RATE_54M) + rxsc = phy_stats1->l_rxsc; + else + rxsc = phy_stats1->ht_rxsc; + + if (phy_stats1->rf_mode == 0 || rxsc == 1 || rxsc == 2) + rx_status->bw = RATE_INFO_BW_20; + else + rx_status->bw = RATE_INFO_BW_40; } static void jaguar2_rx_parse_phystats_type2(struct rtl8xxxu_priv *priv, From dbf9b7bb0edfa192d43ebb41dd0e1041f8a9c5b0 Mon Sep 17 00:00:00 2001 From: Bitterblue Smith Date: Thu, 20 Nov 2025 16:13:49 +0200 Subject: [PATCH 1006/1381] wifi: rtl8xxxu: Enable 40 MHz width by default 40 MHz support is hidden behind the ht40_2g module parameter with this comment: /* * Some APs will negotiate HT20_40 in a noisy environment leading * to miserable performance. Rather than defaulting to this, only * enable it if explicitly requested at module load time. */ This parameter was added in commit 26f1fad29ad9 ("New driver: rtl8xxxu (mac80211)"). Back then rtl8xxxu only supported RTL8723AU and the RTL8192CU family. It's entirely possible the miserable performance was due to mistakes in the channel switching function, which were fixed in a previous patch. Delete the ht40_2g module parameter. If someone still needs to disable 40 MHz support, cfg80211 has the module parameter cfg80211_disable_40mhz_24ghz. That works too. Signed-off-by: Bitterblue Smith Reviewed-by: Ping-Ke Shih Signed-off-by: Ping-Ke Shih Link: https://patch.msgid.link/4f053103-adfd-4ead-acb3-ef69127a4bab@gmail.com --- drivers/net/wireless/realtek/rtl8xxxu/core.c | 19 ++++--------------- 1 file changed, 4 insertions(+), 15 deletions(-) diff --git a/drivers/net/wireless/realtek/rtl8xxxu/core.c b/drivers/net/wireless/realtek/rtl8xxxu/core.c index a18f73aea837..c06ad064f37c 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/core.c +++ b/drivers/net/wireless/realtek/rtl8xxxu/core.c @@ -20,7 +20,6 @@ #define DRIVER_NAME "rtl8xxxu" int rtl8xxxu_debug; -static bool rtl8xxxu_ht40_2g; static bool rtl8xxxu_dma_aggregation; static int rtl8xxxu_dma_agg_timeout = -1; static int rtl8xxxu_dma_agg_pages = -1; @@ -45,8 +44,6 @@ MODULE_FIRMWARE("rtlwifi/rtl8192fufw.bin"); module_param_named(debug, rtl8xxxu_debug, int, 0600); MODULE_PARM_DESC(debug, "Set debug mask"); -module_param_named(ht40_2g, rtl8xxxu_ht40_2g, bool, 0600); -MODULE_PARM_DESC(ht40_2g, "Enable HT40 support on the 2.4GHz band"); module_param_named(dma_aggregation, rtl8xxxu_dma_aggregation, bool, 0600); MODULE_PARM_DESC(dma_aggregation, "Enable DMA packet aggregation"); module_param_named(dma_agg_timeout, rtl8xxxu_dma_agg_timeout, int, 0600); @@ -4896,8 +4893,7 @@ rtl8xxxu_bss_info_changed(struct ieee80211_hw *hw, struct ieee80211_vif *vif, sgi = 1; highest_rate = fls(ramask) - 1; - if (rtl8xxxu_ht40_2g && - (sta->deflink.ht_cap.cap & IEEE80211_HT_CAP_SUP_WIDTH_20_40)) + if (sta->deflink.ht_cap.cap & IEEE80211_HT_CAP_SUP_WIDTH_20_40) bw = RATE_INFO_BW_40; else bw = RATE_INFO_BW_20; @@ -7851,7 +7847,8 @@ static int rtl8xxxu_probe(struct usb_interface *interface, sband->ht_cap.ht_supported = true; sband->ht_cap.ampdu_factor = IEEE80211_HT_MAX_AMPDU_64K; sband->ht_cap.ampdu_density = IEEE80211_HT_MPDU_DENSITY_16; - sband->ht_cap.cap = IEEE80211_HT_CAP_SGI_20 | IEEE80211_HT_CAP_SGI_40; + sband->ht_cap.cap = IEEE80211_HT_CAP_SGI_20 | IEEE80211_HT_CAP_SGI_40 | + IEEE80211_HT_CAP_SUP_WIDTH_20_40; memset(&sband->ht_cap.mcs, 0, sizeof(sband->ht_cap.mcs)); sband->ht_cap.mcs.rx_mask[0] = 0xff; sband->ht_cap.mcs.rx_mask[4] = 0x01; @@ -7860,15 +7857,7 @@ static int rtl8xxxu_probe(struct usb_interface *interface, sband->ht_cap.cap |= IEEE80211_HT_CAP_SGI_40; } sband->ht_cap.mcs.tx_params = IEEE80211_HT_MCS_TX_DEFINED; - /* - * Some APs will negotiate HT20_40 in a noisy environment leading - * to miserable performance. Rather than defaulting to this, only - * enable it if explicitly requested at module load time. - */ - if (rtl8xxxu_ht40_2g) { - dev_info(&udev->dev, "Enabling HT_20_40 on the 2.4GHz band\n"); - sband->ht_cap.cap |= IEEE80211_HT_CAP_SUP_WIDTH_20_40; - } + hw->wiphy->bands[NL80211_BAND_2GHZ] = sband; hw->wiphy->rts_threshold = 2347; From 501253b61d843745e6e1f1fd3b39a49785ffcbf4 Mon Sep 17 00:00:00 2001 From: Slark Xiao Date: Thu, 20 Nov 2025 19:52:08 +0800 Subject: [PATCH 1007/1381] net: wwan: t7xx: Make local function static This function was used in t7xx_hif_cldma.c only. Make it static as it should be. Signed-off-by: Slark Xiao Reviewed-by: Loic Poulain Link: https://patch.msgid.link/20251120115208.345578-1-slark_xiao@163.com Signed-off-by: Jakub Kicinski --- drivers/net/wwan/t7xx/t7xx_hif_cldma.c | 2 +- drivers/net/wwan/t7xx/t7xx_hif_cldma.h | 2 -- 2 files changed, 1 insertion(+), 3 deletions(-) diff --git a/drivers/net/wwan/t7xx/t7xx_hif_cldma.c b/drivers/net/wwan/t7xx/t7xx_hif_cldma.c index 689c920ca898..43ac1c3f1ad0 100644 --- a/drivers/net/wwan/t7xx/t7xx_hif_cldma.c +++ b/drivers/net/wwan/t7xx/t7xx_hif_cldma.c @@ -897,7 +897,7 @@ static void t7xx_cldma_hw_start_send(struct cldma_ctrl *md_ctrl, int qno, * @queue: CLDMA queue. * @recv_skb: Receiving skb callback. */ -void t7xx_cldma_set_recv_skb(struct cldma_queue *queue, +static void t7xx_cldma_set_recv_skb(struct cldma_queue *queue, int (*recv_skb)(struct cldma_queue *queue, struct sk_buff *skb)) { queue->recv_skb = recv_skb; diff --git a/drivers/net/wwan/t7xx/t7xx_hif_cldma.h b/drivers/net/wwan/t7xx/t7xx_hif_cldma.h index f2d9941be9c8..9d0107e18a7b 100644 --- a/drivers/net/wwan/t7xx/t7xx_hif_cldma.h +++ b/drivers/net/wwan/t7xx/t7xx_hif_cldma.h @@ -126,8 +126,6 @@ void t7xx_cldma_switch_cfg(struct cldma_ctrl *md_ctrl, enum cldma_cfg cfg_id); void t7xx_cldma_start(struct cldma_ctrl *md_ctrl); int t7xx_cldma_stop(struct cldma_ctrl *md_ctrl); void t7xx_cldma_reset(struct cldma_ctrl *md_ctrl); -void t7xx_cldma_set_recv_skb(struct cldma_queue *queue, - int (*recv_skb)(struct cldma_queue *queue, struct sk_buff *skb)); int t7xx_cldma_send_skb(struct cldma_ctrl *md_ctrl, int qno, struct sk_buff *skb); void t7xx_cldma_stop_all_qs(struct cldma_ctrl *md_ctrl, enum mtk_txrx tx_rx); void t7xx_cldma_clear_all_qs(struct cldma_ctrl *md_ctrl, enum mtk_txrx tx_rx); From f54bbd390f5fc038cf6ad75432c83b3cedf9bef4 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Thu, 20 Nov 2025 11:25:16 +0000 Subject: [PATCH 1008/1381] net: stmmac: qcom-ethqos: use u32 for rgmii read/write/update readl() returns a u32, and writel() takes a "u32" for the value. These are used in rgmii_readl()() and rgmii_writel(), but the value and return are "int". As these are 32-bit register values which are not signed, use "u32". These changes do not cause generated code changes. Update rgmii_updatel() to use u32 for mask and val. Changing "mask" to "u32" also does not cause generated code changes. However, changing "val" causes the generated assembly to be re-ordered for aarch64. Update the temporary variables used with the rgmii functions to use u32. Signed-off-by: Russell King (Oracle) Reviewed-by: Konrad Dybcio Link: https://patch.msgid.link/E1vM2mq-0000000FRTi-3y5F@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c index 1a616a71c36a..ae3cf163005b 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c @@ -116,21 +116,21 @@ struct qcom_ethqos { bool needs_sgmii_loopback; }; -static int rgmii_readl(struct qcom_ethqos *ethqos, unsigned int offset) +static u32 rgmii_readl(struct qcom_ethqos *ethqos, unsigned int offset) { return readl(ethqos->rgmii_base + offset); } -static void rgmii_writel(struct qcom_ethqos *ethqos, - int value, unsigned int offset) +static void rgmii_writel(struct qcom_ethqos *ethqos, u32 value, + unsigned int offset) { writel(value, ethqos->rgmii_base + offset); } -static void rgmii_updatel(struct qcom_ethqos *ethqos, - int mask, int val, unsigned int offset) +static void rgmii_updatel(struct qcom_ethqos *ethqos, u32 mask, u32 val, + unsigned int offset) { - unsigned int temp; + u32 temp; temp = rgmii_readl(ethqos, offset); temp = (temp & ~(mask)) | val; @@ -300,8 +300,8 @@ static const struct ethqos_emac_driver_data emac_v4_0_0_data = { static int ethqos_dll_configure(struct qcom_ethqos *ethqos) { struct device *dev = ðqos->pdev->dev; - unsigned int val; int retry = 1000; + u32 val; /* Set CDR_EN */ rgmii_updatel(ethqos, SDCC_DLL_CONFIG_CDR_EN, @@ -535,7 +535,7 @@ static int ethqos_rgmii_macro_init(struct qcom_ethqos *ethqos, int speed) static int ethqos_configure_rgmii(struct qcom_ethqos *ethqos, int speed) { struct device *dev = ðqos->pdev->dev; - volatile unsigned int dll_lock; + volatile u32 dll_lock; unsigned int i, retry = 1000; /* Reset to POR values and enable clk */ From 819212185ae5460e63485ff42ef92a3d1b9ec0c7 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Thu, 20 Nov 2025 11:25:22 +0000 Subject: [PATCH 1009/1381] net: stmmac: qcom-ethqos: add rgmii set/clear functions The driver has a lot of bit manipulation of the RGMII registers. Add a pair of helpers to set bits and clear bits, converting the various calls to rgmii_updatel() as appropriate. Most of the change was done via this sed script: /rgmii_updatel/ { N /,$/N /mask, / ! { s|rgmii_updatel\(([^,]*,\s+([^,]*),\s+)\2,\s+|rgmii_setmask(\1| s|rgmii_updatel\(([^,]*,\s+([^,]*),\s+)0,\s+|rgmii_clrmask(\1| s|^\s+$|| } } and then formatting tweaked where necessary. Signed-off-by: Russell King (Oracle) Reviewed-by: Konrad Dybcio Link: https://patch.msgid.link/E1vM2mw-0000000FRTo-0End@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../stmicro/stmmac/dwmac-qcom-ethqos.c | 187 +++++++++--------- 1 file changed, 89 insertions(+), 98 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c index ae3cf163005b..1f84bd821c4e 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c @@ -137,6 +137,18 @@ static void rgmii_updatel(struct qcom_ethqos *ethqos, u32 mask, u32 val, rgmii_writel(ethqos, temp, offset); } +static void rgmii_setmask(struct qcom_ethqos *ethqos, u32 mask, + unsigned int offset) +{ + rgmii_updatel(ethqos, mask, mask, offset); +} + +static void rgmii_clrmask(struct qcom_ethqos *ethqos, u32 mask, + unsigned int offset) +{ + rgmii_updatel(ethqos, mask, 0, offset); +} + static void rgmii_dump(void *priv) { struct qcom_ethqos *ethqos = priv; @@ -194,8 +206,7 @@ qcom_ethqos_set_sgmii_loopback(struct qcom_ethqos *ethqos, bool enable) static void ethqos_set_func_clk_en(struct qcom_ethqos *ethqos) { qcom_ethqos_set_sgmii_loopback(ethqos, true); - rgmii_updatel(ethqos, RGMII_CONFIG_FUNC_CLK_EN, - RGMII_CONFIG_FUNC_CLK_EN, RGMII_IO_MACRO_CONFIG); + rgmii_setmask(ethqos, RGMII_CONFIG_FUNC_CLK_EN, RGMII_IO_MACRO_CONFIG); } static const struct ethqos_emac_por emac_v2_3_0_por[] = { @@ -304,27 +315,25 @@ static int ethqos_dll_configure(struct qcom_ethqos *ethqos) u32 val; /* Set CDR_EN */ - rgmii_updatel(ethqos, SDCC_DLL_CONFIG_CDR_EN, - SDCC_DLL_CONFIG_CDR_EN, SDCC_HC_REG_DLL_CONFIG); + rgmii_setmask(ethqos, SDCC_DLL_CONFIG_CDR_EN, SDCC_HC_REG_DLL_CONFIG); /* Set CDR_EXT_EN */ - rgmii_updatel(ethqos, SDCC_DLL_CONFIG_CDR_EXT_EN, - SDCC_DLL_CONFIG_CDR_EXT_EN, SDCC_HC_REG_DLL_CONFIG); + rgmii_setmask(ethqos, SDCC_DLL_CONFIG_CDR_EXT_EN, + SDCC_HC_REG_DLL_CONFIG); /* Clear CK_OUT_EN */ - rgmii_updatel(ethqos, SDCC_DLL_CONFIG_CK_OUT_EN, - 0, SDCC_HC_REG_DLL_CONFIG); + rgmii_clrmask(ethqos, SDCC_DLL_CONFIG_CK_OUT_EN, + SDCC_HC_REG_DLL_CONFIG); /* Set DLL_EN */ - rgmii_updatel(ethqos, SDCC_DLL_CONFIG_DLL_EN, - SDCC_DLL_CONFIG_DLL_EN, SDCC_HC_REG_DLL_CONFIG); + rgmii_setmask(ethqos, SDCC_DLL_CONFIG_DLL_EN, SDCC_HC_REG_DLL_CONFIG); if (!ethqos->has_emac_ge_3) { - rgmii_updatel(ethqos, SDCC_DLL_MCLK_GATING_EN, - 0, SDCC_HC_REG_DLL_CONFIG); + rgmii_clrmask(ethqos, SDCC_DLL_MCLK_GATING_EN, + SDCC_HC_REG_DLL_CONFIG); - rgmii_updatel(ethqos, SDCC_DLL_CDR_FINE_PHASE, - 0, SDCC_HC_REG_DLL_CONFIG); + rgmii_clrmask(ethqos, SDCC_DLL_CDR_FINE_PHASE, + SDCC_HC_REG_DLL_CONFIG); } /* Wait for CK_OUT_EN clear */ @@ -340,8 +349,8 @@ static int ethqos_dll_configure(struct qcom_ethqos *ethqos) dev_err(dev, "Clear CK_OUT_EN timedout\n"); /* Set CK_OUT_EN */ - rgmii_updatel(ethqos, SDCC_DLL_CONFIG_CK_OUT_EN, - SDCC_DLL_CONFIG_CK_OUT_EN, SDCC_HC_REG_DLL_CONFIG); + rgmii_setmask(ethqos, SDCC_DLL_CONFIG_CK_OUT_EN, + SDCC_HC_REG_DLL_CONFIG); /* Wait for CK_OUT_EN set */ retry = 1000; @@ -357,12 +366,12 @@ static int ethqos_dll_configure(struct qcom_ethqos *ethqos) dev_err(dev, "Set CK_OUT_EN timedout\n"); /* Set DDR_CAL_EN */ - rgmii_updatel(ethqos, SDCC_DLL_CONFIG2_DDR_CAL_EN, - SDCC_DLL_CONFIG2_DDR_CAL_EN, SDCC_HC_REG_DLL_CONFIG2); + rgmii_setmask(ethqos, SDCC_DLL_CONFIG2_DDR_CAL_EN, + SDCC_HC_REG_DLL_CONFIG2); if (!ethqos->has_emac_ge_3) { - rgmii_updatel(ethqos, SDCC_DLL_CONFIG2_DLL_CLOCK_DIS, - 0, SDCC_HC_REG_DLL_CONFIG2); + rgmii_clrmask(ethqos, SDCC_DLL_CONFIG2_DLL_CLOCK_DIS, + SDCC_HC_REG_DLL_CONFIG2); rgmii_updatel(ethqos, SDCC_DLL_CONFIG2_MCLK_FREQ_CALC, 0x1A << 10, SDCC_HC_REG_DLL_CONFIG2); @@ -370,8 +379,7 @@ static int ethqos_dll_configure(struct qcom_ethqos *ethqos) rgmii_updatel(ethqos, SDCC_DLL_CONFIG2_DDR_TRAFFIC_INIT_SEL, BIT(2), SDCC_HC_REG_DLL_CONFIG2); - rgmii_updatel(ethqos, SDCC_DLL_CONFIG2_DDR_TRAFFIC_INIT_SW, - SDCC_DLL_CONFIG2_DDR_TRAFFIC_INIT_SW, + rgmii_setmask(ethqos, SDCC_DLL_CONFIG2_DDR_TRAFFIC_INIT_SW, SDCC_HC_REG_DLL_CONFIG2); } @@ -392,8 +400,8 @@ static int ethqos_rgmii_macro_init(struct qcom_ethqos *ethqos, int speed) phase_shift = RGMII_CONFIG2_TX_CLK_PHASE_SHIFT_EN; /* Disable loopback mode */ - rgmii_updatel(ethqos, RGMII_CONFIG2_TX_TO_RX_LOOPBACK_EN, - 0, RGMII_IO_MACRO_CONFIG2); + rgmii_clrmask(ethqos, RGMII_CONFIG2_TX_TO_RX_LOOPBACK_EN, + RGMII_IO_MACRO_CONFIG2); /* Determine if this platform wants loopback enabled after programming */ if (ethqos->rgmii_config_loopback_en) @@ -402,29 +410,26 @@ static int ethqos_rgmii_macro_init(struct qcom_ethqos *ethqos, int speed) loopback = 0; /* Select RGMII, write 0 to interface select */ - rgmii_updatel(ethqos, RGMII_CONFIG_INTF_SEL, - 0, RGMII_IO_MACRO_CONFIG); + rgmii_clrmask(ethqos, RGMII_CONFIG_INTF_SEL, RGMII_IO_MACRO_CONFIG); switch (speed) { case SPEED_1000: - rgmii_updatel(ethqos, RGMII_CONFIG_DDR_MODE, - RGMII_CONFIG_DDR_MODE, RGMII_IO_MACRO_CONFIG); - rgmii_updatel(ethqos, RGMII_CONFIG_BYPASS_TX_ID_EN, - 0, RGMII_IO_MACRO_CONFIG); - rgmii_updatel(ethqos, RGMII_CONFIG_POS_NEG_DATA_SEL, - RGMII_CONFIG_POS_NEG_DATA_SEL, + rgmii_setmask(ethqos, RGMII_CONFIG_DDR_MODE, RGMII_IO_MACRO_CONFIG); - rgmii_updatel(ethqos, RGMII_CONFIG_PROG_SWAP, - RGMII_CONFIG_PROG_SWAP, RGMII_IO_MACRO_CONFIG); - rgmii_updatel(ethqos, RGMII_CONFIG2_DATA_DIVIDE_CLK_SEL, - 0, RGMII_IO_MACRO_CONFIG2); + rgmii_clrmask(ethqos, RGMII_CONFIG_BYPASS_TX_ID_EN, + RGMII_IO_MACRO_CONFIG); + rgmii_setmask(ethqos, RGMII_CONFIG_POS_NEG_DATA_SEL, + RGMII_IO_MACRO_CONFIG); + rgmii_setmask(ethqos, RGMII_CONFIG_PROG_SWAP, + RGMII_IO_MACRO_CONFIG); + rgmii_clrmask(ethqos, RGMII_CONFIG2_DATA_DIVIDE_CLK_SEL, + RGMII_IO_MACRO_CONFIG2); rgmii_updatel(ethqos, RGMII_CONFIG2_TX_CLK_PHASE_SHIFT_EN, phase_shift, RGMII_IO_MACRO_CONFIG2); - rgmii_updatel(ethqos, RGMII_CONFIG2_RSVD_CONFIG15, - 0, RGMII_IO_MACRO_CONFIG2); - rgmii_updatel(ethqos, RGMII_CONFIG2_RX_PROG_SWAP, - RGMII_CONFIG2_RX_PROG_SWAP, + rgmii_clrmask(ethqos, RGMII_CONFIG2_RSVD_CONFIG15, + RGMII_IO_MACRO_CONFIG2); + rgmii_setmask(ethqos, RGMII_CONFIG2_RX_PROG_SWAP, RGMII_IO_MACRO_CONFIG2); /* PRG_RCLK_DLY = TCXO period * TCXO_CYCLES_CNT / 2 * RX delay ns, @@ -439,87 +444,78 @@ static int ethqos_rgmii_macro_init(struct qcom_ethqos *ethqos, int speed) rgmii_updatel(ethqos, SDCC_DDR_CONFIG_PRG_RCLK_DLY, 57, SDCC_HC_REG_DDR_CONFIG); } - rgmii_updatel(ethqos, SDCC_DDR_CONFIG_PRG_DLY_EN, - SDCC_DDR_CONFIG_PRG_DLY_EN, + rgmii_setmask(ethqos, SDCC_DDR_CONFIG_PRG_DLY_EN, SDCC_HC_REG_DDR_CONFIG); rgmii_updatel(ethqos, RGMII_CONFIG_LOOPBACK_EN, loopback, RGMII_IO_MACRO_CONFIG); break; case SPEED_100: - rgmii_updatel(ethqos, RGMII_CONFIG_DDR_MODE, - RGMII_CONFIG_DDR_MODE, RGMII_IO_MACRO_CONFIG); - rgmii_updatel(ethqos, RGMII_CONFIG_BYPASS_TX_ID_EN, - RGMII_CONFIG_BYPASS_TX_ID_EN, + rgmii_setmask(ethqos, RGMII_CONFIG_DDR_MODE, RGMII_IO_MACRO_CONFIG); - rgmii_updatel(ethqos, RGMII_CONFIG_POS_NEG_DATA_SEL, - 0, RGMII_IO_MACRO_CONFIG); - rgmii_updatel(ethqos, RGMII_CONFIG_PROG_SWAP, - 0, RGMII_IO_MACRO_CONFIG); - rgmii_updatel(ethqos, RGMII_CONFIG2_DATA_DIVIDE_CLK_SEL, - 0, RGMII_IO_MACRO_CONFIG2); + rgmii_setmask(ethqos, RGMII_CONFIG_BYPASS_TX_ID_EN, + RGMII_IO_MACRO_CONFIG); + rgmii_clrmask(ethqos, RGMII_CONFIG_POS_NEG_DATA_SEL, + RGMII_IO_MACRO_CONFIG); + rgmii_clrmask(ethqos, RGMII_CONFIG_PROG_SWAP, + RGMII_IO_MACRO_CONFIG); + rgmii_clrmask(ethqos, RGMII_CONFIG2_DATA_DIVIDE_CLK_SEL, + RGMII_IO_MACRO_CONFIG2); rgmii_updatel(ethqos, RGMII_CONFIG2_TX_CLK_PHASE_SHIFT_EN, phase_shift, RGMII_IO_MACRO_CONFIG2); rgmii_updatel(ethqos, RGMII_CONFIG_MAX_SPD_PRG_2, BIT(6), RGMII_IO_MACRO_CONFIG); - rgmii_updatel(ethqos, RGMII_CONFIG2_RSVD_CONFIG15, - 0, RGMII_IO_MACRO_CONFIG2); + rgmii_clrmask(ethqos, RGMII_CONFIG2_RSVD_CONFIG15, + RGMII_IO_MACRO_CONFIG2); if (ethqos->has_emac_ge_3) - rgmii_updatel(ethqos, RGMII_CONFIG2_RX_PROG_SWAP, - RGMII_CONFIG2_RX_PROG_SWAP, + rgmii_setmask(ethqos, RGMII_CONFIG2_RX_PROG_SWAP, RGMII_IO_MACRO_CONFIG2); else - rgmii_updatel(ethqos, RGMII_CONFIG2_RX_PROG_SWAP, - 0, RGMII_IO_MACRO_CONFIG2); + rgmii_clrmask(ethqos, RGMII_CONFIG2_RX_PROG_SWAP, + RGMII_IO_MACRO_CONFIG2); /* Write 0x5 to PRG_RCLK_DLY_CODE */ rgmii_updatel(ethqos, SDCC_DDR_CONFIG_EXT_PRG_RCLK_DLY_CODE, (BIT(29) | BIT(27)), SDCC_HC_REG_DDR_CONFIG); - rgmii_updatel(ethqos, SDCC_DDR_CONFIG_EXT_PRG_RCLK_DLY, - SDCC_DDR_CONFIG_EXT_PRG_RCLK_DLY, + rgmii_setmask(ethqos, SDCC_DDR_CONFIG_EXT_PRG_RCLK_DLY, SDCC_HC_REG_DDR_CONFIG); - rgmii_updatel(ethqos, SDCC_DDR_CONFIG_EXT_PRG_RCLK_DLY_EN, - SDCC_DDR_CONFIG_EXT_PRG_RCLK_DLY_EN, + rgmii_setmask(ethqos, SDCC_DDR_CONFIG_EXT_PRG_RCLK_DLY_EN, SDCC_HC_REG_DDR_CONFIG); rgmii_updatel(ethqos, RGMII_CONFIG_LOOPBACK_EN, loopback, RGMII_IO_MACRO_CONFIG); break; case SPEED_10: - rgmii_updatel(ethqos, RGMII_CONFIG_DDR_MODE, - RGMII_CONFIG_DDR_MODE, RGMII_IO_MACRO_CONFIG); - rgmii_updatel(ethqos, RGMII_CONFIG_BYPASS_TX_ID_EN, - RGMII_CONFIG_BYPASS_TX_ID_EN, + rgmii_setmask(ethqos, RGMII_CONFIG_DDR_MODE, RGMII_IO_MACRO_CONFIG); - rgmii_updatel(ethqos, RGMII_CONFIG_POS_NEG_DATA_SEL, - 0, RGMII_IO_MACRO_CONFIG); - rgmii_updatel(ethqos, RGMII_CONFIG_PROG_SWAP, - 0, RGMII_IO_MACRO_CONFIG); - rgmii_updatel(ethqos, RGMII_CONFIG2_DATA_DIVIDE_CLK_SEL, - 0, RGMII_IO_MACRO_CONFIG2); + rgmii_setmask(ethqos, RGMII_CONFIG_BYPASS_TX_ID_EN, + RGMII_IO_MACRO_CONFIG); + rgmii_clrmask(ethqos, RGMII_CONFIG_POS_NEG_DATA_SEL, + RGMII_IO_MACRO_CONFIG); + rgmii_clrmask(ethqos, RGMII_CONFIG_PROG_SWAP, + RGMII_IO_MACRO_CONFIG); + rgmii_clrmask(ethqos, RGMII_CONFIG2_DATA_DIVIDE_CLK_SEL, + RGMII_IO_MACRO_CONFIG2); rgmii_updatel(ethqos, RGMII_CONFIG2_TX_CLK_PHASE_SHIFT_EN, phase_shift, RGMII_IO_MACRO_CONFIG2); rgmii_updatel(ethqos, RGMII_CONFIG_MAX_SPD_PRG_9, BIT(12) | GENMASK(9, 8), RGMII_IO_MACRO_CONFIG); - rgmii_updatel(ethqos, RGMII_CONFIG2_RSVD_CONFIG15, - 0, RGMII_IO_MACRO_CONFIG2); + rgmii_clrmask(ethqos, RGMII_CONFIG2_RSVD_CONFIG15, + RGMII_IO_MACRO_CONFIG2); if (ethqos->has_emac_ge_3) - rgmii_updatel(ethqos, RGMII_CONFIG2_RX_PROG_SWAP, - RGMII_CONFIG2_RX_PROG_SWAP, + rgmii_setmask(ethqos, RGMII_CONFIG2_RX_PROG_SWAP, RGMII_IO_MACRO_CONFIG2); else - rgmii_updatel(ethqos, RGMII_CONFIG2_RX_PROG_SWAP, - 0, RGMII_IO_MACRO_CONFIG2); + rgmii_clrmask(ethqos, RGMII_CONFIG2_RX_PROG_SWAP, + RGMII_IO_MACRO_CONFIG2); /* Write 0x5 to PRG_RCLK_DLY_CODE */ rgmii_updatel(ethqos, SDCC_DDR_CONFIG_EXT_PRG_RCLK_DLY_CODE, (BIT(29) | BIT(27)), SDCC_HC_REG_DDR_CONFIG); - rgmii_updatel(ethqos, SDCC_DDR_CONFIG_EXT_PRG_RCLK_DLY, - SDCC_DDR_CONFIG_EXT_PRG_RCLK_DLY, + rgmii_setmask(ethqos, SDCC_DDR_CONFIG_EXT_PRG_RCLK_DLY, SDCC_HC_REG_DDR_CONFIG); - rgmii_updatel(ethqos, SDCC_DDR_CONFIG_EXT_PRG_RCLK_DLY_EN, - SDCC_DDR_CONFIG_EXT_PRG_RCLK_DLY_EN, + rgmii_setmask(ethqos, SDCC_DDR_CONFIG_EXT_PRG_RCLK_DLY_EN, SDCC_HC_REG_DDR_CONFIG); rgmii_updatel(ethqos, RGMII_CONFIG_LOOPBACK_EN, loopback, RGMII_IO_MACRO_CONFIG); @@ -547,12 +543,12 @@ static int ethqos_configure_rgmii(struct qcom_ethqos *ethqos, int speed) /* Initialize the DLL first */ /* Set DLL_RST */ - rgmii_updatel(ethqos, SDCC_DLL_CONFIG_DLL_RST, - SDCC_DLL_CONFIG_DLL_RST, SDCC_HC_REG_DLL_CONFIG); + rgmii_setmask(ethqos, SDCC_DLL_CONFIG_DLL_RST, + SDCC_HC_REG_DLL_CONFIG); /* Set PDN */ - rgmii_updatel(ethqos, SDCC_DLL_CONFIG_PDN, - SDCC_DLL_CONFIG_PDN, SDCC_HC_REG_DLL_CONFIG); + rgmii_setmask(ethqos, SDCC_DLL_CONFIG_PDN, + SDCC_HC_REG_DLL_CONFIG); if (ethqos->has_emac_ge_3) { if (speed == SPEED_1000) { @@ -566,21 +562,18 @@ static int ethqos_configure_rgmii(struct qcom_ethqos *ethqos, int speed) } /* Clear DLL_RST */ - rgmii_updatel(ethqos, SDCC_DLL_CONFIG_DLL_RST, 0, - SDCC_HC_REG_DLL_CONFIG); + rgmii_clrmask(ethqos, SDCC_DLL_CONFIG_DLL_RST, SDCC_HC_REG_DLL_CONFIG); /* Clear PDN */ - rgmii_updatel(ethqos, SDCC_DLL_CONFIG_PDN, 0, - SDCC_HC_REG_DLL_CONFIG); + rgmii_clrmask(ethqos, SDCC_DLL_CONFIG_PDN, SDCC_HC_REG_DLL_CONFIG); if (speed != SPEED_100 && speed != SPEED_10) { /* Set DLL_EN */ - rgmii_updatel(ethqos, SDCC_DLL_CONFIG_DLL_EN, - SDCC_DLL_CONFIG_DLL_EN, SDCC_HC_REG_DLL_CONFIG); + rgmii_setmask(ethqos, SDCC_DLL_CONFIG_DLL_EN, + SDCC_HC_REG_DLL_CONFIG); /* Set CK_OUT_EN */ - rgmii_updatel(ethqos, SDCC_DLL_CONFIG_CK_OUT_EN, - SDCC_DLL_CONFIG_CK_OUT_EN, + rgmii_setmask(ethqos, SDCC_DLL_CONFIG_CK_OUT_EN, SDCC_HC_REG_DLL_CONFIG); /* Set USR_CTL bit 26 with mask of 3 bits */ @@ -631,15 +624,13 @@ static int ethqos_configure_sgmii(struct qcom_ethqos *ethqos, int speed) switch (speed) { case SPEED_2500: - rgmii_updatel(ethqos, RGMII_CONFIG2_RGMII_CLK_SEL_CFG, - RGMII_CONFIG2_RGMII_CLK_SEL_CFG, + rgmii_setmask(ethqos, RGMII_CONFIG2_RGMII_CLK_SEL_CFG, RGMII_IO_MACRO_CONFIG2); ethqos_set_serdes_speed(ethqos, SPEED_2500); ethqos_pcs_set_inband(priv, false); break; case SPEED_1000: - rgmii_updatel(ethqos, RGMII_CONFIG2_RGMII_CLK_SEL_CFG, - RGMII_CONFIG2_RGMII_CLK_SEL_CFG, + rgmii_setmask(ethqos, RGMII_CONFIG2_RGMII_CLK_SEL_CFG, RGMII_IO_MACRO_CONFIG2); ethqos_set_serdes_speed(ethqos, SPEED_1000); ethqos_pcs_set_inband(priv, true); From 9b60ba512c7f82739b705d94c41c38d4bf6b0235 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Thu, 20 Nov 2025 11:25:27 +0000 Subject: [PATCH 1010/1381] net: stmmac: qcom-ethqos: use read_poll_timeout_atomic() Use read_poll_timeout_atomic() to poll the rgmii registers rather than open-coding the polling. Reviewed-by: Konrad Dybcio Signed-off-by: Russell King (Oracle) Link: https://patch.msgid.link/E1vM2n1-0000000FRTu-0js9@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- .../stmicro/stmmac/dwmac-qcom-ethqos.c | 44 ++++++------------- 1 file changed, 14 insertions(+), 30 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c index 1f84bd821c4e..0826a7bd32ff 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c @@ -311,7 +311,6 @@ static const struct ethqos_emac_driver_data emac_v4_0_0_data = { static int ethqos_dll_configure(struct qcom_ethqos *ethqos) { struct device *dev = ðqos->pdev->dev; - int retry = 1000; u32 val; /* Set CDR_EN */ @@ -337,15 +336,10 @@ static int ethqos_dll_configure(struct qcom_ethqos *ethqos) } /* Wait for CK_OUT_EN clear */ - do { - val = rgmii_readl(ethqos, SDCC_HC_REG_DLL_CONFIG); - val &= SDCC_DLL_CONFIG_CK_OUT_EN; - if (!val) - break; - mdelay(1); - retry--; - } while (retry > 0); - if (!retry) + if (read_poll_timeout_atomic(rgmii_readl, val, + !(val & SDCC_DLL_CONFIG_CK_OUT_EN), + 1000, 1000000, false, + ethqos, SDCC_HC_REG_DLL_CONFIG)) dev_err(dev, "Clear CK_OUT_EN timedout\n"); /* Set CK_OUT_EN */ @@ -353,16 +347,10 @@ static int ethqos_dll_configure(struct qcom_ethqos *ethqos) SDCC_HC_REG_DLL_CONFIG); /* Wait for CK_OUT_EN set */ - retry = 1000; - do { - val = rgmii_readl(ethqos, SDCC_HC_REG_DLL_CONFIG); - val &= SDCC_DLL_CONFIG_CK_OUT_EN; - if (val) - break; - mdelay(1); - retry--; - } while (retry > 0); - if (!retry) + if (read_poll_timeout_atomic(rgmii_readl, val, + val & SDCC_DLL_CONFIG_CK_OUT_EN, + 1000, 1000000, false, + ethqos, SDCC_HC_REG_DLL_CONFIG)) dev_err(dev, "Set CK_OUT_EN timedout\n"); /* Set DDR_CAL_EN */ @@ -531,8 +519,8 @@ static int ethqos_rgmii_macro_init(struct qcom_ethqos *ethqos, int speed) static int ethqos_configure_rgmii(struct qcom_ethqos *ethqos, int speed) { struct device *dev = ðqos->pdev->dev; - volatile u32 dll_lock; - unsigned int i, retry = 1000; + unsigned int i; + u32 val; /* Reset to POR values and enable clk */ for (i = 0; i < ethqos->num_por; i++) @@ -582,14 +570,10 @@ static int ethqos_configure_rgmii(struct qcom_ethqos *ethqos, int speed) SDCC_USR_CTL); /* wait for DLL LOCK */ - do { - mdelay(1); - dll_lock = rgmii_readl(ethqos, SDC4_STATUS); - if (dll_lock & SDC4_STATUS_DLL_LOCK) - break; - retry--; - } while (retry > 0); - if (!retry) + if (read_poll_timeout_atomic(rgmii_readl, val, + val & SDC4_STATUS_DLL_LOCK, + 1000, 1000000, true, + ethqos, SDC4_STATUS)) dev_err(dev, "Timeout while waiting for DLL lock\n"); } From de3c5142349c6cd67fbd20afc5b3ae320df436b7 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Fri, 21 Nov 2025 11:37:34 +0100 Subject: [PATCH 1011/1381] wifi: mac80211: fix channel switching code My prior commit here introduced a bug due to copy/paste, it was iterating the links assigned to 'ctx->replace_ctx' and I replaced it by iterating links assigned to 'ctx' by accident, then modified it for the iteration later. Fix it to iterate the users of the correct chanctx, i.e. 'ctx->replace_ctx'. Ultimately, this issue led to a crash in a hwsim test (multi_ap_wps_shared_apdev_csa) because it would actually do the switch (rather than refuse here) and then later have a double-free of the original chanctx, because it was still in use by another interface yet freed as part of the switching. Fixes: a1dc648aa76d ("wifi: mac80211: remove chanctx to link back-references") Link: https://patch.msgid.link/20251121113733.7710a58d45eb.Ie9ec010b52b1baed93dbe44f968c2119b6b5d98d@changeid Signed-off-by: Johannes Berg --- net/mac80211/chan.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/mac80211/chan.c b/net/mac80211/chan.c index 6aa305839f53..c8aba4183c9a 100644 --- a/net/mac80211/chan.c +++ b/net/mac80211/chan.c @@ -1715,7 +1715,7 @@ static int ieee80211_vif_use_reserved_switch(struct ieee80211_local *local) n_reserved = 0; n_ready = 0; - for_each_chanctx_user_assigned(local, ctx, &iter) { + for_each_chanctx_user_assigned(local, ctx->replace_ctx, &iter) { n_assigned++; if (iter.link->reserved_chanctx) { n_reserved++; From 7a27b73943a70ee226fa125327101fb18e94701d Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Fri, 21 Nov 2025 17:40:20 +0100 Subject: [PATCH 1012/1381] wifi: cfg80211: use cfg80211_leave() in iftype change When changing the interface type, all activity on the interface has to be stopped first. This was done independent of existing code in cfg80211_leave(), so didn't handle e.g. background radar detection. Use cfg80211_leave() to handle it the same way. Note that cfg80211_leave() behaves slightly differently for IBSS in wireless extensions, it won't send an event in that case. We could handle that, but since nl80211 was used to change the type, IBSS is rare, and wext is already a corner case, it doesn't seem worth it. Link: https://patch.msgid.link/20251121174021.922ef48ce007.I970c8514252ef8a864a7fbdab9591b71031dee03@changeid Signed-off-by: Johannes Berg --- net/wireless/util.c | 23 +---------------------- 1 file changed, 1 insertion(+), 22 deletions(-) diff --git a/net/wireless/util.c b/net/wireless/util.c index 97f40c6d1e9d..27e8a2f52f04 100644 --- a/net/wireless/util.c +++ b/net/wireless/util.c @@ -1203,28 +1203,7 @@ int cfg80211_change_iface(struct cfg80211_registered_device *rdev, dev->ieee80211_ptr->use_4addr = false; rdev_set_qos_map(rdev, dev, NULL); - switch (otype) { - case NL80211_IFTYPE_AP: - case NL80211_IFTYPE_P2P_GO: - cfg80211_stop_ap(rdev, dev, -1, true); - break; - case NL80211_IFTYPE_ADHOC: - cfg80211_leave_ibss(rdev, dev, false); - break; - case NL80211_IFTYPE_STATION: - case NL80211_IFTYPE_P2P_CLIENT: - cfg80211_disconnect(rdev, dev, - WLAN_REASON_DEAUTH_LEAVING, true); - break; - case NL80211_IFTYPE_MESH_POINT: - /* mesh should be handled? */ - break; - case NL80211_IFTYPE_OCB: - cfg80211_leave_ocb(rdev, dev); - break; - default: - break; - } + cfg80211_leave(rdev, dev->ieee80211_ptr); cfg80211_process_rdev_events(rdev); cfg80211_mlme_purge_registrations(dev->ieee80211_ptr); From 9f33477b9a31a1edfe2df9f1a0359cccb0e16b4c Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Fri, 21 Nov 2025 17:40:21 +0100 Subject: [PATCH 1013/1381] wifi: cfg80211: stop radar detection in cfg80211_leave() If an interface is set down or, per the previous patch, changes type, radar detection for it should be cancelled. This is done for AP mode in mac80211 (somewhat needlessly, since cfg80211 can do it, but didn't until now), but wasn't handled for mesh, so if radar detection was started and then the interface set down or its type switched (the latter sometimes happning in the hwsim test 'mesh_peer_connected_dfs'), radar detection would be around with the interface unknown to the driver, later leading to some warnings around chanctx usage. Link: https://patch.msgid.link/20251121174021.290120e419e3.I2a5650c9062e29c988992dd8ce0d8eb570d23267@changeid Signed-off-by: Johannes Berg --- net/wireless/core.c | 1 + net/wireless/core.h | 1 + net/wireless/mlme.c | 19 +++++++++++++++++++ 3 files changed, 21 insertions(+) diff --git a/net/wireless/core.c b/net/wireless/core.c index 7be25b9ae36e..9a420d627d3c 100644 --- a/net/wireless/core.c +++ b/net/wireless/core.c @@ -1380,6 +1380,7 @@ void cfg80211_leave(struct cfg80211_registered_device *rdev, cfg80211_pmsr_wdev_down(wdev); + cfg80211_stop_radar_detection(wdev); cfg80211_stop_background_radar_detection(wdev); switch (wdev->iftype) { diff --git a/net/wireless/core.h b/net/wireless/core.h index 82f343663e8f..63dcf315dba7 100644 --- a/net/wireless/core.h +++ b/net/wireless/core.h @@ -489,6 +489,7 @@ cfg80211_start_background_radar_detection(struct cfg80211_registered_device *rde struct wireless_dev *wdev, struct cfg80211_chan_def *chandef); +void cfg80211_stop_radar_detection(struct wireless_dev *wdev); void cfg80211_stop_background_radar_detection(struct wireless_dev *wdev); void cfg80211_background_cac_done_wk(struct work_struct *work); diff --git a/net/wireless/mlme.c b/net/wireless/mlme.c index 46394eb2086f..3fc175f9f868 100644 --- a/net/wireless/mlme.c +++ b/net/wireless/mlme.c @@ -1295,6 +1295,25 @@ cfg80211_start_background_radar_detection(struct cfg80211_registered_device *rde return 0; } +void cfg80211_stop_radar_detection(struct wireless_dev *wdev) +{ + struct wiphy *wiphy = wdev->wiphy; + struct cfg80211_registered_device *rdev = wiphy_to_rdev(wiphy); + int link_id; + + for_each_valid_link(wdev, link_id) { + struct cfg80211_chan_def chandef; + + if (!wdev->links[link_id].cac_started) + continue; + + chandef = *wdev_chandef(wdev, link_id); + rdev_end_cac(rdev, wdev->netdev, link_id); + nl80211_radar_notify(rdev, &chandef, NL80211_RADAR_CAC_ABORTED, + wdev->netdev, GFP_KERNEL); + } +} + void cfg80211_stop_background_radar_detection(struct wireless_dev *wdev) { struct wiphy *wiphy = wdev->wiphy; From a8e5a110c0c38e08e5dd66356cd1156e91cf88e1 Mon Sep 17 00:00:00 2001 From: Hans de Goede Date: Mon, 3 Nov 2025 11:03:14 +0100 Subject: [PATCH 1014/1381] wifi: brcmfmac: Add DMI nvram filename quirk for Acer A1 840 tablet The Acer A1 840 tablet contains quite generic names in the sys_vendor and product_name DMI strings, without this patch brcmfmac will try to load: brcmfmac43340-sdio.Insyde-BayTrail.txt as nvram file which is a bit too generic. Add a DMI quirk so that a unique and clearly identifiable nvram file name is used on the Acer A1 840 tablet. Acked-by: Arend van Spriel Signed-off-by: Hans de Goede Link: https://patch.msgid.link/20251103100314.353826-1-hansg@kernel.org Signed-off-by: Johannes Berg --- .../net/wireless/broadcom/brcm80211/brcmfmac/dmi.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+) diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/dmi.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/dmi.c index c3a602197662..abe7f6501e5e 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/dmi.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/dmi.c @@ -24,6 +24,10 @@ static const struct brcmf_dmi_data acepc_t8_data = { BRCM_CC_4345_CHIP_ID, 6, "acepc-t8" }; +static const struct brcmf_dmi_data acer_a1_840_data = { + BRCM_CC_43340_CHIP_ID, 2, "acer-a1-840" +}; + /* The Chuwi Hi8 Pro uses the same Ampak AP6212 module as the Chuwi Vi8 Plus * and the nvram for the Vi8 Plus is already in linux-firmware, so use that. */ @@ -91,6 +95,16 @@ static const struct dmi_system_id dmi_platform_data[] = { }, .driver_data = (void *)&acepc_t8_data, }, + { + /* Acer Iconia One 8 A1-840 (non FHD version) */ + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Insyde"), + DMI_MATCH(DMI_PRODUCT_NAME, "BayTrail"), + /* Above strings are too generic also match BIOS date */ + DMI_MATCH(DMI_BIOS_DATE, "04/01/2014"), + }, + .driver_data = (void *)&acer_a1_840_data, + }, { /* Chuwi Hi8 Pro with D2D3_Hi8Pro.233 BIOS */ .matches = { From 79277f8ad15ec5f255ed0e1427c7a8a3e94e7f52 Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Mon, 29 Sep 2025 13:17:23 +0200 Subject: [PATCH 1015/1381] wifi: mt76: mt7996: fix null pointer deref in mt7996_conf_tx() If a link does not have an assigned channel yet, mt7996_vif_link returns NULL. We still need to store the updated queue settings in that case, and apply them later. Move the location of the queue params to within struct mt7996_vif_link. Fixes: c0df2f0caa8d ("wifi: mt76: mt7996: prepare mt7996_mcu_set_tx for MLO support") Acked-by: Lorenzo Bianconi Link: https://patch.msgid.link/20250929111723.52486-1-nbd@nbd.name Signed-off-by: Felix Fietkau --- drivers/net/wireless/mediatek/mt76/mt7996/main.c | 6 +++--- drivers/net/wireless/mediatek/mt76/mt7996/mcu.c | 5 ++++- drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h | 7 ++++++- 3 files changed, 13 insertions(+), 5 deletions(-) diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/main.c b/drivers/net/wireless/mediatek/mt76/mt7996/main.c index 581314368c5b..b53ca702591c 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/main.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/main.c @@ -665,8 +665,8 @@ mt7996_conf_tx(struct ieee80211_hw *hw, struct ieee80211_vif *vif, unsigned int link_id, u16 queue, const struct ieee80211_tx_queue_params *params) { - struct mt7996_dev *dev = mt7996_hw_dev(hw); - struct mt7996_vif_link *mlink = mt7996_vif_link(dev, vif, link_id); + struct mt7996_vif *mvif = (struct mt7996_vif *)vif->drv_priv; + struct mt7996_vif_link_info *link_info = &mvif->link_info[link_id]; static const u8 mq_to_aci[] = { [IEEE80211_AC_VO] = 3, [IEEE80211_AC_VI] = 2, @@ -675,7 +675,7 @@ mt7996_conf_tx(struct ieee80211_hw *hw, struct ieee80211_vif *vif, }; /* firmware uses access class index */ - mlink->queue_params[mq_to_aci[queue]] = *params; + link_info->queue_params[mq_to_aci[queue]] = *params; /* no need to update right away, we'll get BSS_CHANGED_QOS */ return 0; diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c index 0347ee0c2dd7..afa6a43bd51e 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c @@ -3414,6 +3414,9 @@ int mt7996_mcu_set_tx(struct mt7996_dev *dev, struct ieee80211_vif *vif, #define WMM_PARAM_SET (WMM_AIFS_SET | WMM_CW_MIN_SET | \ WMM_CW_MAX_SET | WMM_TXOP_SET) struct mt7996_vif_link *link = mt7996_vif_conf_link(dev, vif, link_conf); + struct mt7996_vif *mvif = (struct mt7996_vif *)vif->drv_priv; + unsigned int link_id = link_conf->link_id; + struct mt7996_vif_link_info *link_info = &mvif->link_info[link_id]; struct { u8 bss_idx; u8 __rsv[3]; @@ -3431,7 +3434,7 @@ int mt7996_mcu_set_tx(struct mt7996_dev *dev, struct ieee80211_vif *vif, skb_put_data(skb, &hdr, sizeof(hdr)); for (ac = 0; ac < IEEE80211_NUM_ACS; ac++) { - struct ieee80211_tx_queue_params *q = &link->queue_params[ac]; + struct ieee80211_tx_queue_params *q = &link_info->queue_params[ac]; struct edca *e; struct tlv *tlv; diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h b/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h index 8ec2acdb3319..718e4d4ad85f 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h +++ b/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h @@ -253,16 +253,21 @@ struct mt7996_vif_link { struct mt7996_sta_link msta_link; struct mt7996_phy *phy; - struct ieee80211_tx_queue_params queue_params[IEEE80211_NUM_ACS]; struct cfg80211_bitrate_mask bitrate_mask; u8 mld_idx; }; +struct mt7996_vif_link_info { + struct ieee80211_tx_queue_params queue_params[IEEE80211_NUM_ACS]; +}; + struct mt7996_vif { struct mt7996_vif_link deflink; /* must be first */ struct mt76_vif_data mt76; + struct mt7996_vif_link_info link_info[IEEE80211_MLD_MAX_NUM_LINKS]; + u8 mld_group_idx; u8 mld_remap_idx; }; From 084922069ceac4d594c06b76a80352139fd15f4d Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Wed, 24 Sep 2025 15:51:39 +0200 Subject: [PATCH 1016/1381] wifi: mt76: mt7996: Remove unnecessary link_id checks in mt7996_tx Remove unnecessary link_id checks in mt7996_tx routine since if the link identifier provided by mac80211 is unspecified the value will be overwritten at the beginning on the function. Fixes: f940c9b7aef6 ("wifi: mt76: mt7996: Set proper link destination address in mt7996_tx()") Signed-off-by: Lorenzo Bianconi Link: https://patch.msgid.link/20250924-mt76_tx_unnecessary-check-v1-1-e595930a5662@kernel.org Signed-off-by: Felix Fietkau --- drivers/net/wireless/mediatek/mt76/mt7996/main.c | 12 +++++------- 1 file changed, 5 insertions(+), 7 deletions(-) diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/main.c b/drivers/net/wireless/mediatek/mt76/mt7996/main.c index b53ca702591c..2b52d057287a 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/main.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/main.c @@ -1339,12 +1339,10 @@ static void mt7996_tx(struct ieee80211_hw *hw, } if (mvif) { - struct mt76_vif_link *mlink = &mvif->deflink.mt76; + struct mt76_vif_link *mlink; - if (link_id < IEEE80211_LINK_UNSPECIFIED) - mlink = rcu_dereference(mvif->mt76.link[link_id]); - - if (mlink->wcid) + mlink = rcu_dereference(mvif->mt76.link[link_id]); + if (mlink && mlink->wcid) wcid = mlink->wcid; if (mvif->mt76.roc_phy && @@ -1352,7 +1350,7 @@ static void mt7996_tx(struct ieee80211_hw *hw, mphy = mvif->mt76.roc_phy; if (mphy->roc_link) wcid = mphy->roc_link->wcid; - } else { + } else if (mlink) { mphy = mt76_vif_link_phy(mlink); } } @@ -1362,7 +1360,7 @@ static void mt7996_tx(struct ieee80211_hw *hw, goto unlock; } - if (msta && link_id < IEEE80211_LINK_UNSPECIFIED) { + if (msta) { struct mt7996_sta_link *msta_link; msta_link = rcu_dereference(msta->link[link_id]); From 8c5b063061326b84192625c62e963f1e998193a8 Mon Sep 17 00:00:00 2001 From: Thorsten Blum Date: Tue, 23 Sep 2025 23:38:31 +0200 Subject: [PATCH 1017/1381] wifi: mt76: connac: Replace memcpy + hard-coded size with strscpy Replace memcpy() and the hard-coded string length with strscpy() to safely copy the string and improve mt76_connac_mcu_chip_config(). No functional changes. Signed-off-by: Thorsten Blum Link: https://patch.msgid.link/20250923213831.1896823-2-thorsten.blum@linux.dev Signed-off-by: Felix Fietkau --- drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c index fc3e6728fcfb..451375f2901b 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c +++ b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c @@ -1974,7 +1974,7 @@ int mt76_connac_mcu_chip_config(struct mt76_dev *dev) .resp_type = 0, }; - memcpy(req.data, "assert", 7); + strscpy(req.data, "assert"); return mt76_mcu_send_msg(dev, MCU_CE_CMD(CHIP_CONFIG), &req, sizeof(req), false); From 38b845e1f9e810869b0a0b69f202b877b7b7fb12 Mon Sep 17 00:00:00 2001 From: "Sven Eckelmann (Plasma Cloud)" Date: Fri, 26 Sep 2025 11:32:54 +0200 Subject: [PATCH 1018/1381] wifi: mt76: Fix DTS power-limits on little endian systems The power-limits for ru and mcs and stored in the devicetree as bytewise array (often with sizes which are not a multiple of 4). These arrays have a prefix which defines for how many modes a line is applied. This prefix is also only a byte - but the code still tried to fix the endianness of this byte with a be32 operation. As result, loading was mostly failing or was sending completely unexpected values to the firmware. Since the other rates are also stored in the devicetree as bytewise arrays, just drop the u32 access + be32_to_cpu conversion and directly access them as bytes arrays. Cc: stable@vger.kernel.org Fixes: 22b980badc0f ("mt76: add functions for parsing rate power limits from DT") Fixes: a9627d992b5e ("mt76: extend DT rate power limits to support 11ax devices") Signed-off-by: Sven Eckelmann (Plasma Cloud) Signed-off-by: Felix Fietkau --- drivers/net/wireless/mediatek/mt76/eeprom.c | 37 +++++++++++++-------- 1 file changed, 24 insertions(+), 13 deletions(-) diff --git a/drivers/net/wireless/mediatek/mt76/eeprom.c b/drivers/net/wireless/mediatek/mt76/eeprom.c index a987c5e4eff6..6ce8e4af18fe 100644 --- a/drivers/net/wireless/mediatek/mt76/eeprom.c +++ b/drivers/net/wireless/mediatek/mt76/eeprom.c @@ -253,6 +253,19 @@ mt76_get_of_array(struct device_node *np, char *name, size_t *len, int min) return prop->value; } +static const s8 * +mt76_get_of_array_s8(struct device_node *np, char *name, size_t *len, int min) +{ + struct property *prop = of_find_property(np, name, NULL); + + if (!prop || !prop->value || prop->length < min) + return NULL; + + *len = prop->length; + + return prop->value; +} + struct device_node * mt76_find_channel_node(struct device_node *np, struct ieee80211_channel *chan) { @@ -294,7 +307,7 @@ mt76_get_txs_delta(struct device_node *np, u8 nss) } static void -mt76_apply_array_limit(s8 *pwr, size_t pwr_len, const __be32 *data, +mt76_apply_array_limit(s8 *pwr, size_t pwr_len, const s8 *data, s8 target_power, s8 nss_delta, s8 *max_power) { int i; @@ -303,15 +316,14 @@ mt76_apply_array_limit(s8 *pwr, size_t pwr_len, const __be32 *data, return; for (i = 0; i < pwr_len; i++) { - pwr[i] = min_t(s8, target_power, - be32_to_cpu(data[i]) + nss_delta); + pwr[i] = min_t(s8, target_power, data[i] + nss_delta); *max_power = max(*max_power, pwr[i]); } } static void mt76_apply_multi_array_limit(s8 *pwr, size_t pwr_len, s8 pwr_num, - const __be32 *data, size_t len, s8 target_power, + const s8 *data, size_t len, s8 target_power, s8 nss_delta, s8 *max_power) { int i, cur; @@ -319,8 +331,7 @@ mt76_apply_multi_array_limit(s8 *pwr, size_t pwr_len, s8 pwr_num, if (!data) return; - len /= 4; - cur = be32_to_cpu(data[0]); + cur = data[0]; for (i = 0; i < pwr_num; i++) { if (len < pwr_len + 1) break; @@ -335,7 +346,7 @@ mt76_apply_multi_array_limit(s8 *pwr, size_t pwr_len, s8 pwr_num, if (!len) break; - cur = be32_to_cpu(data[0]); + cur = data[0]; } } @@ -346,7 +357,7 @@ s8 mt76_get_rate_power_limits(struct mt76_phy *phy, { struct mt76_dev *dev = phy->dev; struct device_node *np; - const __be32 *val; + const s8 *val; char name[16]; u32 mcs_rates = dev->drv->mcs_rates; u32 ru_rates = ARRAY_SIZE(dest->ru[0]); @@ -392,21 +403,21 @@ s8 mt76_get_rate_power_limits(struct mt76_phy *phy, txs_delta = mt76_get_txs_delta(np, hweight16(phy->chainmask)); - val = mt76_get_of_array(np, "rates-cck", &len, ARRAY_SIZE(dest->cck)); + val = mt76_get_of_array_s8(np, "rates-cck", &len, ARRAY_SIZE(dest->cck)); mt76_apply_array_limit(dest->cck, ARRAY_SIZE(dest->cck), val, target_power, txs_delta, &max_power); - val = mt76_get_of_array(np, "rates-ofdm", - &len, ARRAY_SIZE(dest->ofdm)); + val = mt76_get_of_array_s8(np, "rates-ofdm", + &len, ARRAY_SIZE(dest->ofdm)); mt76_apply_array_limit(dest->ofdm, ARRAY_SIZE(dest->ofdm), val, target_power, txs_delta, &max_power); - val = mt76_get_of_array(np, "rates-mcs", &len, mcs_rates + 1); + val = mt76_get_of_array_s8(np, "rates-mcs", &len, mcs_rates + 1); mt76_apply_multi_array_limit(dest->mcs[0], ARRAY_SIZE(dest->mcs[0]), ARRAY_SIZE(dest->mcs), val, len, target_power, txs_delta, &max_power); - val = mt76_get_of_array(np, "rates-ru", &len, ru_rates + 1); + val = mt76_get_of_array_s8(np, "rates-ru", &len, ru_rates + 1); mt76_apply_multi_array_limit(dest->ru[0], ARRAY_SIZE(dest->ru[0]), ARRAY_SIZE(dest->ru), val, len, target_power, txs_delta, &max_power); From 9a04a69b68ad10dee793c887b367760651777c42 Mon Sep 17 00:00:00 2001 From: "Sven Eckelmann (Plasma Cloud)" Date: Tue, 7 Oct 2025 10:11:14 +0200 Subject: [PATCH 1019/1381] dt-bindings: net: wireless: mt76: Document power-limits country property The commit 22b980badc0f ("mt76: add functions for parsing rate power limits from DT") added filtering of the power limits based on two properties: * regdomain * country If either the country or the regdomain matches, the power limits are applied and the search is aborted. If none of the two is defined for the power limit, it is a global (or "fallback") power limit. The last "fallback" power limit in the list will be returned when not matching regdomain or country was found. The idea is here to allow to specify "overwriting" country limits in front of the list - just in case a regdomain is shared but a country has additional limitations. But this property was forgotten to be defined in commit 2de6ccebe0e7 ("dt-bindings:net:wireless:mediatek,mt76: introduce power-limits node"). Signed-off-by: Sven Eckelmann (Plasma Cloud) Reviewed-by: Rob Herring (Arm) Link: https://patch.msgid.link/20251007-backoff-table-support-v3-1-fd6e2684988f@simonwunderlich.de Signed-off-by: Felix Fietkau --- .../devicetree/bindings/net/wireless/mediatek,mt76.yaml | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/Documentation/devicetree/bindings/net/wireless/mediatek,mt76.yaml b/Documentation/devicetree/bindings/net/wireless/mediatek,mt76.yaml index eabceb849537..82b224c2f6f7 100644 --- a/Documentation/devicetree/bindings/net/wireless/mediatek,mt76.yaml +++ b/Documentation/devicetree/bindings/net/wireless/mediatek,mt76.yaml @@ -151,6 +151,12 @@ properties: - ETSI - JP + country: + $ref: /schemas/types.yaml#/definitions/string + pattern: '^[A-Z]{2}$' + description: + ISO 3166-1 alpha-2 country code for power limits + patternProperties: "^txpower-[256]g$": type: object From 6b9833c611d2e149292cc1c542dccfda1ca98c1f Mon Sep 17 00:00:00 2001 From: "Sven Eckelmann (Plasma Cloud)" Date: Tue, 7 Oct 2025 10:11:15 +0200 Subject: [PATCH 1020/1381] dt-bindings: net: wireless: mt76: introduce backoff limit properties Introduce path backoff limit properties in mt76 binding in order to specify beamforming and non-beamforming backoff limits for 802.11n/ac/ax. Reviewed-by: Rob Herring (Arm) Signed-off-by: Sven Eckelmann (Plasma Cloud) Link: https://patch.msgid.link/20251007-backoff-table-support-v3-2-fd6e2684988f@simonwunderlich.de Signed-off-by: Felix Fietkau --- .../bindings/net/wireless/mediatek,mt76.yaml | 60 +++++++++++++++++++ 1 file changed, 60 insertions(+) diff --git a/Documentation/devicetree/bindings/net/wireless/mediatek,mt76.yaml b/Documentation/devicetree/bindings/net/wireless/mediatek,mt76.yaml index 82b224c2f6f7..ae6b97cdc44b 100644 --- a/Documentation/devicetree/bindings/net/wireless/mediatek,mt76.yaml +++ b/Documentation/devicetree/bindings/net/wireless/mediatek,mt76.yaml @@ -216,6 +216,66 @@ properties: minItems: 13 maxItems: 13 + paths-cck: + $ref: /schemas/types.yaml#/definitions/uint8-array + minItems: 4 + maxItems: 4 + description: + 4 half-dBm backoff values (1 - 4 antennas, single spacial + stream) + + paths-ofdm: + $ref: /schemas/types.yaml#/definitions/uint8-array + minItems: 4 + maxItems: 4 + description: + 4 half-dBm backoff values (1 - 4 antennas, single spacial + stream) + + paths-ofdm-bf: + $ref: /schemas/types.yaml#/definitions/uint8-array + minItems: 4 + maxItems: 4 + description: + 4 half-dBm backoff values for beamforming + (1 - 4 antennas, single spacial stream) + + paths-ru: + $ref: /schemas/types.yaml#/definitions/uint8-matrix + description: + Sets of half-dBm backoff values for 802.11ax rates for + 1T1ss (aka 1 transmitting antenna with 1 spacial stream), + 2T1ss, 3T1ss, 4T1ss, 2T2ss, 3T2ss, 4T2ss, 3T3ss, 4T3ss + and 4T4ss. + Each set starts with the number of channel bandwidth or + resource unit settings for which the rate set applies, + followed by 10 power limit values. The order of the + channel resource unit settings is RU26, RU52, RU106, + RU242/SU20, RU484/SU40, RU996/SU80 and RU2x996/SU160. + minItems: 1 + maxItems: 7 + items: + minItems: 11 + maxItems: 11 + + paths-ru-bf: + $ref: /schemas/types.yaml#/definitions/uint8-matrix + description: + Sets of half-dBm backoff (beamforming) values for 802.11ax + rates for 1T1ss (aka 1 transmitting antenna with 1 spacial + stream), 2T1ss, 3T1ss, 4T1ss, 2T2ss, 3T2ss, 4T2ss, 3T3ss, + 4T3ss and 4T4ss. + Each set starts with the number of channel bandwidth or + resource unit settings for which the rate set applies, + followed by 10 power limit values. The order of the + channel resource unit settings is RU26, RU52, RU106, + RU242/SU20, RU484/SU40, RU996/SU80 and RU2x996/SU160. + minItems: 1 + maxItems: 7 + items: + minItems: 11 + maxItems: 11 + txs-delta: $ref: /schemas/types.yaml#/definitions/uint32-array description: From b05ab4be9fd779115635144ebb0691a0c3dd3943 Mon Sep 17 00:00:00 2001 From: Shayne Chen Date: Tue, 7 Oct 2025 10:11:16 +0200 Subject: [PATCH 1021/1381] wifi: mt76: mt7915: add bf backoff limit table support The commit 22b980badc0f ("mt76: add functions for parsing rate power limits from DT") introduced generic support for rates limits in the devicetree. But the mt7915 supports beamforming and has another table for configuring the backoff limits. These can be configured in the DT with the paths-* properties. The path-*-bf are the ones relevant for beamforming and the ones without -bf suffix for "traditional" path backoff. Signed-off-by: Shayne Chen Signed-off-by: Sven Eckelmann (Plasma Cloud) Link: https://patch.msgid.link/20251007-backoff-table-support-v3-3-fd6e2684988f@simonwunderlich.de Signed-off-by: Felix Fietkau --- drivers/net/wireless/mediatek/mt76/debugfs.c | 4 +- drivers/net/wireless/mediatek/mt76/eeprom.c | 38 +++- drivers/net/wireless/mediatek/mt76/mt76.h | 8 + .../wireless/mediatek/mt76/mt7915/debugfs.c | 74 ++++++- .../net/wireless/mediatek/mt76/mt7915/init.c | 7 + .../net/wireless/mediatek/mt76/mt7915/main.c | 2 +- .../net/wireless/mediatek/mt76/mt7915/mcu.c | 182 ++++++++++++++---- .../net/wireless/mediatek/mt76/mt7915/mcu.h | 6 + .../wireless/mediatek/mt76/mt7915/mt7915.h | 9 +- .../wireless/mediatek/mt76/mt7915/testmode.c | 2 +- 10 files changed, 277 insertions(+), 55 deletions(-) diff --git a/drivers/net/wireless/mediatek/mt76/debugfs.c b/drivers/net/wireless/mediatek/mt76/debugfs.c index b6a2746c187d..bee1177594d3 100644 --- a/drivers/net/wireless/mediatek/mt76/debugfs.c +++ b/drivers/net/wireless/mediatek/mt76/debugfs.c @@ -93,9 +93,9 @@ void mt76_seq_puts_array(struct seq_file *file, const char *str, { int i; - seq_printf(file, "%10s:", str); + seq_printf(file, "%16s:", str); for (i = 0; i < len; i++) - seq_printf(file, " %2d", val[i]); + seq_printf(file, " %4d", val[i]); seq_puts(file, "\n"); } EXPORT_SYMBOL_GPL(mt76_seq_puts_array); diff --git a/drivers/net/wireless/mediatek/mt76/eeprom.c b/drivers/net/wireless/mediatek/mt76/eeprom.c index 6ce8e4af18fe..cbda19a86882 100644 --- a/drivers/net/wireless/mediatek/mt76/eeprom.c +++ b/drivers/net/wireless/mediatek/mt76/eeprom.c @@ -324,9 +324,10 @@ mt76_apply_array_limit(s8 *pwr, size_t pwr_len, const s8 *data, static void mt76_apply_multi_array_limit(s8 *pwr, size_t pwr_len, s8 pwr_num, const s8 *data, size_t len, s8 target_power, - s8 nss_delta, s8 *max_power) + s8 nss_delta) { int i, cur; + s8 max_power = -128; if (!data) return; @@ -337,7 +338,7 @@ mt76_apply_multi_array_limit(s8 *pwr, size_t pwr_len, s8 pwr_num, break; mt76_apply_array_limit(pwr + pwr_len * i, pwr_len, data + 1, - target_power, nss_delta, max_power); + target_power, nss_delta, &max_power); if (--cur > 0) continue; @@ -364,12 +365,16 @@ s8 mt76_get_rate_power_limits(struct mt76_phy *phy, char band; size_t len; s8 max_power = 0; + s8 max_power_backoff = -127; s8 txs_delta; + int n_chains = hweight16(phy->chainmask); + s8 target_power_combine = target_power + mt76_tx_power_path_delta(n_chains); if (!mcs_rates) mcs_rates = 10; - memset(dest, target_power, sizeof(*dest)); + memset(dest, target_power, sizeof(*dest) - sizeof(dest->path)); + memset(&dest->path, 0, sizeof(dest->path)); if (!IS_ENABLED(CONFIG_OF)) return target_power; @@ -415,12 +420,35 @@ s8 mt76_get_rate_power_limits(struct mt76_phy *phy, val = mt76_get_of_array_s8(np, "rates-mcs", &len, mcs_rates + 1); mt76_apply_multi_array_limit(dest->mcs[0], ARRAY_SIZE(dest->mcs[0]), ARRAY_SIZE(dest->mcs), val, len, - target_power, txs_delta, &max_power); + target_power, txs_delta); val = mt76_get_of_array_s8(np, "rates-ru", &len, ru_rates + 1); mt76_apply_multi_array_limit(dest->ru[0], ARRAY_SIZE(dest->ru[0]), ARRAY_SIZE(dest->ru), val, len, - target_power, txs_delta, &max_power); + target_power, txs_delta); + + max_power_backoff = max_power; + val = mt76_get_of_array_s8(np, "paths-cck", &len, ARRAY_SIZE(dest->path.cck)); + mt76_apply_array_limit(dest->path.cck, ARRAY_SIZE(dest->path.cck), val, + target_power_combine, txs_delta, &max_power_backoff); + + val = mt76_get_of_array_s8(np, "paths-ofdm", &len, ARRAY_SIZE(dest->path.ofdm)); + mt76_apply_array_limit(dest->path.ofdm, ARRAY_SIZE(dest->path.ofdm), val, + target_power_combine, txs_delta, &max_power_backoff); + + val = mt76_get_of_array_s8(np, "paths-ofdm-bf", &len, ARRAY_SIZE(dest->path.ofdm_bf)); + mt76_apply_array_limit(dest->path.ofdm_bf, ARRAY_SIZE(dest->path.ofdm_bf), val, + target_power_combine, txs_delta, &max_power_backoff); + + val = mt76_get_of_array_s8(np, "paths-ru", &len, ARRAY_SIZE(dest->path.ru[0]) + 1); + mt76_apply_multi_array_limit(dest->path.ru[0], ARRAY_SIZE(dest->path.ru[0]), + ARRAY_SIZE(dest->path.ru), val, len, + target_power_combine, txs_delta); + + val = mt76_get_of_array_s8(np, "paths-ru-bf", &len, ARRAY_SIZE(dest->path.ru_bf[0]) + 1); + mt76_apply_multi_array_limit(dest->path.ru_bf[0], ARRAY_SIZE(dest->path.ru_bf[0]), + ARRAY_SIZE(dest->path.ru_bf), val, len, + target_power_combine, txs_delta); return max_power; } diff --git a/drivers/net/wireless/mediatek/mt76/mt76.h b/drivers/net/wireless/mediatek/mt76/mt76.h index e0d50b58cd01..86b812f68c97 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76.h +++ b/drivers/net/wireless/mediatek/mt76/mt76.h @@ -1113,6 +1113,14 @@ struct mt76_power_limits { s8 mcs[4][10]; s8 ru[7][12]; s8 eht[16][16]; + + struct { + s8 cck[4]; + s8 ofdm[4]; + s8 ofdm_bf[4]; + s8 ru[7][10]; + s8 ru_bf[7][10]; + } path; }; struct mt76_ethtool_worker_info { diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/debugfs.c b/drivers/net/wireless/mediatek/mt76/mt7915/debugfs.c index b287b7d9394e..6ff0d262c28a 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7915/debugfs.c +++ b/drivers/net/wireless/mediatek/mt76/mt7915/debugfs.c @@ -1008,7 +1008,7 @@ mt7915_rate_txpower_get(struct file *file, char __user *user_buf, if (!buf) return -ENOMEM; - ret = mt7915_mcu_get_txpower_sku(phy, txpwr, sizeof(txpwr)); + ret = mt7915_mcu_get_txpower_sku(phy, txpwr, sizeof(txpwr), TX_POWER_INFO_RATE); if (ret) goto out; @@ -1118,7 +1118,7 @@ mt7915_rate_txpower_set(struct file *file, const char __user *user_buf, mutex_lock(&dev->mt76.mutex); ret = mt7915_mcu_get_txpower_sku(phy, req.txpower_sku, - sizeof(req.txpower_sku)); + sizeof(req.txpower_sku), TX_POWER_INFO_RATE); if (ret) goto out; @@ -1160,7 +1160,7 @@ mt7915_rate_txpower_set(struct file *file, const char __user *user_buf, return ret ? ret : count; } -static const struct file_operations mt7915_rate_txpower_fops = { +static const struct file_operations mt7915_txpower_fops = { .write = mt7915_rate_txpower_set, .read = mt7915_rate_txpower_get, .open = simple_open, @@ -1168,6 +1168,70 @@ static const struct file_operations mt7915_rate_txpower_fops = { .llseek = default_llseek, }; +static int +mt7915_path_txpower_show(struct seq_file *file) +{ + struct mt7915_phy *phy = file->private; + s8 txpower[MT7915_SKU_PATH_NUM], *buf = txpower; + int ret; + +#define PATH_POWER_SHOW(_name, _len, _skip) do { \ + size_t __len = (_len); \ + if (_skip) { \ + buf -= 1; \ + *buf = 0; \ + } \ + mt76_seq_puts_array(file, _name, buf, __len); \ + buf += __len; \ + } while (0) + + seq_printf(file, "\n%*c", 18, ' '); + seq_puts(file, "1T1S/2T1S/3T1S/4T1S/2T2S/3T2S/4T2S/3T3S/4T3S/4T4S\n"); + ret = mt7915_mcu_get_txpower_sku(phy, txpower, sizeof(txpower), + TX_POWER_INFO_PATH); + if (ret) + return ret; + + PATH_POWER_SHOW("CCK", 4, 0); + PATH_POWER_SHOW("OFDM", 4, 0); + PATH_POWER_SHOW("BF-OFDM", 4, 1); + + PATH_POWER_SHOW("HT/VHT20", 10, 0); + PATH_POWER_SHOW("BF-HT/VHT20", 10, 1); + PATH_POWER_SHOW("HT/VHT40", 10, 0); + PATH_POWER_SHOW("BF-HT/VHT40", 10, 1); + + PATH_POWER_SHOW("BW20/RU242", 10, 0); + PATH_POWER_SHOW("BF-BW20/RU242", 10, 1); + PATH_POWER_SHOW("BW40/RU484", 10, 0); + PATH_POWER_SHOW("BF-BW40/RU484", 10, 1); + PATH_POWER_SHOW("BW80/RU996", 10, 0); + PATH_POWER_SHOW("BF-BW80/RU996", 10, 1); + PATH_POWER_SHOW("BW160/RU2x996", 10, 0); + PATH_POWER_SHOW("BF-BW160/RU2x996", 10, 1); + PATH_POWER_SHOW("RU26", 10, 0); + PATH_POWER_SHOW("BF-RU26", 10, 0); + PATH_POWER_SHOW("RU52", 10, 0); + PATH_POWER_SHOW("BF-RU52", 10, 0); + PATH_POWER_SHOW("RU106", 10, 0); + PATH_POWER_SHOW("BF-RU106", 10, 0); +#undef PATH_POWER_SHOW + + return 0; +} + +static int +mt7915_txpower_path_show(struct seq_file *file, void *data) +{ + struct mt7915_phy *phy = file->private; + + seq_printf(file, "\nBand %d\n", phy != &phy->dev->phy); + + return mt7915_path_txpower_show(file); +} + +DEFINE_SHOW_ATTRIBUTE(mt7915_txpower_path); + static int mt7915_twt_stats(struct seq_file *s, void *data) { @@ -1254,7 +1318,9 @@ int mt7915_init_debugfs(struct mt7915_phy *phy) debugfs_create_file("implicit_txbf", 0600, dir, dev, &fops_implicit_txbf); debugfs_create_file("txpower_sku", 0400, dir, phy, - &mt7915_rate_txpower_fops); + &mt7915_txpower_fops); + debugfs_create_file("txpower_path", 0400, dir, phy, + &mt7915_txpower_path_fops); debugfs_create_devm_seqfile(dev->mt76.dev, "twt_stats", dir, mt7915_twt_stats); debugfs_create_file("rf_regval", 0600, dir, dev, &fops_rf_regval); diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/init.c b/drivers/net/wireless/mediatek/mt76/mt7915/init.c index 5ea8b46e092e..1ac7ee2922b0 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7915/init.c +++ b/drivers/net/wireless/mediatek/mt76/mt7915/init.c @@ -289,6 +289,8 @@ static void __mt7915_init_txpower(struct mt7915_phy *phy, int pwr_delta = mt7915_eeprom_get_power_delta(dev, sband->band); struct mt76_power_limits limits; + phy->sku_limit_en = true; + phy->sku_path_en = true; for (i = 0; i < sband->n_channels; i++) { struct ieee80211_channel *chan = &sband->channels[i]; u32 target_power = 0; @@ -305,6 +307,11 @@ static void __mt7915_init_txpower(struct mt7915_phy *phy, target_power = mt76_get_rate_power_limits(phy->mt76, chan, &limits, target_power); + + /* MT7915N can not enable Backoff table without setting value in dts */ + if (!limits.path.ofdm[0]) + phy->sku_path_en = false; + target_power += path_delta; target_power = DIV_ROUND_UP(target_power, 2); chan->max_power = min_t(int, chan->max_reg_power, diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/main.c b/drivers/net/wireless/mediatek/mt76/mt7915/main.c index fe0639c14bf9..fa135447a263 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7915/main.c +++ b/drivers/net/wireless/mediatek/mt76/mt7915/main.c @@ -73,7 +73,7 @@ int mt7915_run(struct ieee80211_hw *hw) if (ret) goto out; - ret = mt7915_mcu_set_sku_en(phy, true); + ret = mt7915_mcu_set_sku_en(phy); if (ret) goto out; diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c index c1fdd3c4f1ba..673338d508a8 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c +++ b/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c @@ -3336,7 +3336,8 @@ int mt7915_mcu_set_txpower_frame(struct mt7915_phy *phy, int ret; s8 txpower_sku[MT7915_SKU_RATE_NUM]; - ret = mt7915_mcu_get_txpower_sku(phy, txpower_sku, sizeof(txpower_sku)); + ret = mt7915_mcu_get_txpower_sku(phy, txpower_sku, sizeof(txpower_sku), + TX_POWER_INFO_RATE); if (ret) return ret; @@ -3376,51 +3377,136 @@ int mt7915_mcu_set_txpower_frame(struct mt7915_phy *phy, sizeof(req), true); } +static void +mt7915_update_txpower(struct mt7915_phy *phy, int tx_power) +{ + struct mt76_phy *mphy = phy->mt76; + struct ieee80211_channel *chan = mphy->main_chandef.chan; + int chain_idx, val, e2p_power_limit = 0; + + if (!chan) { + mphy->txpower_cur = tx_power; + return; + } + + for (chain_idx = 0; chain_idx < hweight16(mphy->chainmask); chain_idx++) { + val = mt7915_eeprom_get_target_power(phy->dev, chan, chain_idx); + val += mt7915_eeprom_get_power_delta(phy->dev, chan->band); + + e2p_power_limit = max_t(int, e2p_power_limit, val); + } + + if (phy->sku_limit_en) + mphy->txpower_cur = min_t(int, e2p_power_limit, tx_power); + else + mphy->txpower_cur = e2p_power_limit; +} + int mt7915_mcu_set_txpower_sku(struct mt7915_phy *phy) { +#define TX_POWER_LIMIT_TABLE_RATE 0 +#define TX_POWER_LIMIT_TABLE_PATH 1 struct mt7915_dev *dev = phy->dev; struct mt76_phy *mphy = phy->mt76; struct ieee80211_hw *hw = mphy->hw; - struct mt7915_mcu_txpower_sku req = { + struct mt7915_sku_val { + u8 format_id; + u8 limit_type; + u8 band_idx; + } __packed hdr = { .format_id = TX_POWER_LIMIT_TABLE, + .limit_type = TX_POWER_LIMIT_TABLE_RATE, .band_idx = phy->mt76->band_idx, }; - struct mt76_power_limits limits_array; - s8 *la = (s8 *)&limits_array; - int i, idx; - int tx_power; + int i, ret, tx_power; + const u8 *len = mt7915_sku_group_len; + struct mt76_power_limits la = {}; + struct sk_buff *skb; tx_power = mt76_get_power_bound(mphy, hw->conf.power_level); - tx_power = mt76_get_rate_power_limits(mphy, mphy->chandef.chan, - &limits_array, tx_power); - mphy->txpower_cur = tx_power; - - for (i = 0, idx = 0; i < ARRAY_SIZE(mt7915_sku_group_len); i++) { - u8 mcs_num, len = mt7915_sku_group_len[i]; - int j; - - if (i >= SKU_HT_BW20 && i <= SKU_VHT_BW160) { - mcs_num = 10; - - if (i == SKU_HT_BW20 || i == SKU_VHT_BW20) - la = (s8 *)&limits_array + 12; - } else { - mcs_num = len; - } - - for (j = 0; j < min_t(u8, mcs_num, len); j++) - req.txpower_sku[idx + j] = la[j]; - - la += mcs_num; - idx += len; + if (phy->sku_limit_en) { + tx_power = mt76_get_rate_power_limits(mphy, mphy->chandef.chan, + &la, tx_power); + mt7915_update_txpower(phy, tx_power); + } else { + mt7915_update_txpower(phy, tx_power); + return 0; } - return mt76_mcu_send_msg(&dev->mt76, - MCU_EXT_CMD(TX_POWER_FEATURE_CTRL), &req, - sizeof(req), true); + skb = mt76_mcu_msg_alloc(&dev->mt76, NULL, + sizeof(hdr) + MT7915_SKU_RATE_NUM); + if (!skb) + return -ENOMEM; + + skb_put_data(skb, &hdr, sizeof(hdr)); + skb_put_data(skb, &la.cck, len[SKU_CCK] + len[SKU_OFDM]); + skb_put_data(skb, &la.mcs[0], len[SKU_HT_BW20]); + skb_put_data(skb, &la.mcs[1], len[SKU_HT_BW40]); + + /* vht */ + for (i = 0; i < 4; i++) { + skb_put_data(skb, &la.mcs[i], sizeof(la.mcs[i])); + skb_put_zero(skb, 2); /* padding */ + } + + /* he */ + skb_put_data(skb, &la.ru[0], sizeof(la.ru)); + ret = mt76_mcu_skb_send_msg(&dev->mt76, skb, + MCU_EXT_CMD(TX_POWER_FEATURE_CTRL), true); + if (ret) + return ret; + + /* only set per-path power table when it's configured */ + if (!phy->sku_path_en) + return 0; + + skb = mt76_mcu_msg_alloc(&dev->mt76, NULL, + sizeof(hdr) + MT7915_SKU_PATH_NUM); + if (!skb) + return -ENOMEM; + + hdr.limit_type = TX_POWER_LIMIT_TABLE_PATH; + skb_put_data(skb, &hdr, sizeof(hdr)); + skb_put_data(skb, &la.path.cck, sizeof(la.path.cck)); + skb_put_data(skb, &la.path.ofdm, sizeof(la.path.ofdm)); + skb_put_data(skb, &la.path.ofdm_bf[1], sizeof(la.path.ofdm_bf) - 1); + + /* HT20 and HT40 */ + skb_put_data(skb, &la.path.ru[3], sizeof(la.path.ru[3])); + skb_put_data(skb, &la.path.ru_bf[3][1], sizeof(la.path.ru_bf[3]) - 1); + skb_put_data(skb, &la.path.ru[4], sizeof(la.path.ru[4])); + skb_put_data(skb, &la.path.ru_bf[4][1], sizeof(la.path.ru_bf[4]) - 1); + + /* start from non-bf and bf fields of + * BW20/RU242, BW40/RU484, BW80/RU996, BW160/RU2x996, + * RU26, RU52, and RU106 + */ + + for (i = 0; i < 8; i++) { + bool bf = i % 2; + u8 idx = (i + 6) / 2; + s8 *buf = bf ? la.path.ru_bf[idx] : la.path.ru[idx]; + /* The non-bf fields of RU26 to RU106 are special cases */ + if (bf) + skb_put_data(skb, buf + 1, 9); + else + skb_put_data(skb, buf, 10); + } + + for (i = 0; i < 6; i++) { + bool bf = i % 2; + u8 idx = i / 2; + s8 *buf = bf ? la.path.ru_bf[idx] : la.path.ru[idx]; + + skb_put_data(skb, buf, 10); + } + + return mt76_mcu_skb_send_msg(&dev->mt76, skb, + MCU_EXT_CMD(TX_POWER_FEATURE_CTRL), true); } -int mt7915_mcu_get_txpower_sku(struct mt7915_phy *phy, s8 *txpower, int len) +int mt7915_mcu_get_txpower_sku(struct mt7915_phy *phy, s8 *txpower, int len, + u8 category) { #define RATE_POWER_INFO 2 struct mt7915_dev *dev = phy->dev; @@ -3431,10 +3517,9 @@ int mt7915_mcu_get_txpower_sku(struct mt7915_phy *phy, s8 *txpower, int len) u8 _rsv; } __packed req = { .format_id = TX_POWER_LIMIT_INFO, - .category = RATE_POWER_INFO, + .category = category, .band_idx = phy->mt76->band_idx, }; - s8 txpower_sku[MT7915_SKU_RATE_NUM][2]; struct sk_buff *skb; int ret, i; @@ -3444,9 +3529,15 @@ int mt7915_mcu_get_txpower_sku(struct mt7915_phy *phy, s8 *txpower, int len) if (ret) return ret; - memcpy(txpower_sku, skb->data + 4, sizeof(txpower_sku)); - for (i = 0; i < len; i++) - txpower[i] = txpower_sku[i][req.band_idx]; + if (category == TX_POWER_INFO_RATE) { + s8 res[MT7915_SKU_RATE_NUM][2]; + + memcpy(res, skb->data + 4, sizeof(res)); + for (i = 0; i < len; i++) + txpower[i] = res[i][req.band_idx]; + } else if (category == TX_POWER_INFO_PATH) { + memcpy(txpower, skb->data + 4, len); + } dev_kfree_skb(skb); @@ -3475,7 +3566,7 @@ int mt7915_mcu_set_test_param(struct mt7915_dev *dev, u8 param, bool test_mode, sizeof(req), false); } -int mt7915_mcu_set_sku_en(struct mt7915_phy *phy, bool enable) +int mt7915_mcu_set_sku_en(struct mt7915_phy *phy) { struct mt7915_dev *dev = phy->dev; struct mt7915_sku { @@ -3484,10 +3575,21 @@ int mt7915_mcu_set_sku_en(struct mt7915_phy *phy, bool enable) u8 band_idx; u8 rsv; } __packed req = { - .format_id = TX_POWER_LIMIT_ENABLE, .band_idx = phy->mt76->band_idx, - .sku_enable = enable, }; + int ret; + + req.sku_enable = phy->sku_limit_en; + req.format_id = TX_POWER_LIMIT_ENABLE; + + ret = mt76_mcu_send_msg(&dev->mt76, + MCU_EXT_CMD(TX_POWER_FEATURE_CTRL), &req, + sizeof(req), true); + if (ret) + return ret; + + req.sku_enable = phy->sku_path_en; + req.format_id = TX_POWER_LIMIT_PATH_ENABLE; return mt76_mcu_send_msg(&dev->mt76, MCU_EXT_CMD(TX_POWER_FEATURE_CTRL), &req, diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/mcu.h b/drivers/net/wireless/mediatek/mt76/mt7915/mcu.h index 086ad89ecd91..b72535efc6d7 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7915/mcu.h +++ b/drivers/net/wireless/mediatek/mt76/mt7915/mcu.h @@ -429,12 +429,18 @@ enum { enum { TX_POWER_LIMIT_ENABLE, + TX_POWER_LIMIT_PATH_ENABLE = 0x3, TX_POWER_LIMIT_TABLE = 0x4, TX_POWER_LIMIT_INFO = 0x7, TX_POWER_LIMIT_FRAME = 0x11, TX_POWER_LIMIT_FRAME_MIN = 0x12, }; +enum { + TX_POWER_INFO_PATH = 1, + TX_POWER_INFO_RATE, +}; + enum { SPR_ENABLE = 0x1, SPR_ENABLE_SD = 0x3, diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/mt7915.h b/drivers/net/wireless/mediatek/mt76/mt7915/mt7915.h index 2e94347c46d6..b15d31d36a87 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7915/mt7915.h +++ b/drivers/net/wireless/mediatek/mt76/mt7915/mt7915.h @@ -70,6 +70,7 @@ #define MT7915_CDEV_THROTTLE_MAX 99 #define MT7915_SKU_RATE_NUM 161 +#define MT7915_SKU_PATH_NUM 185 #define MT7915_MAX_TWT_AGRT 16 #define MT7915_MAX_STA_TWT_AGRT 8 @@ -223,6 +224,9 @@ struct mt7915_phy { struct mt76_mib_stats mib; struct mt76_channel_state state_ts; + bool sku_limit_en:1; + bool sku_path_en:1; + #ifdef CONFIG_NL80211_TESTMODE struct { u32 *reg_backup; @@ -491,9 +495,10 @@ int mt7915_mcu_set_mac(struct mt7915_dev *dev, int band, bool enable, int mt7915_mcu_set_test_param(struct mt7915_dev *dev, u8 param, bool test_mode, u8 en); int mt7915_mcu_set_ser(struct mt7915_dev *dev, u8 action, u8 set, u8 band); -int mt7915_mcu_set_sku_en(struct mt7915_phy *phy, bool enable); +int mt7915_mcu_set_sku_en(struct mt7915_phy *phy); int mt7915_mcu_set_txpower_sku(struct mt7915_phy *phy); -int mt7915_mcu_get_txpower_sku(struct mt7915_phy *phy, s8 *txpower, int len); +int mt7915_mcu_get_txpower_sku(struct mt7915_phy *phy, s8 *txpower, int len, + u8 category); int mt7915_mcu_set_txpower_frame_min(struct mt7915_phy *phy, s8 txpower); int mt7915_mcu_set_txpower_frame(struct mt7915_phy *phy, struct ieee80211_vif *vif, diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/testmode.c b/drivers/net/wireless/mediatek/mt76/mt7915/testmode.c index d534fff5c952..5836b9733f27 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7915/testmode.c +++ b/drivers/net/wireless/mediatek/mt76/mt7915/testmode.c @@ -409,7 +409,7 @@ mt7915_tm_init(struct mt7915_phy *phy, bool en) if (!test_bit(MT76_STATE_RUNNING, &phy->mt76->state)) return; - mt7915_mcu_set_sku_en(phy, !en); + mt7915_mcu_set_sku_en(phy); mt7915_tm_mode_ctrl(dev, en); mt7915_tm_reg_backup_restore(phy); From 385aab8fccd7a8746b9f1a17f3c1e38498a14bc7 Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Wed, 8 Oct 2025 12:41:48 +0200 Subject: [PATCH 1022/1381] wifi: mt76: wed: use proper wed reference in mt76 wed driver callabacks MT7996 driver can use both wed and wed_hif2 devices to offload traffic from/to the wireless NIC. In the current codebase we assume to always use the primary wed device in wed callbacks resulting in the following crash if the hw runs wed_hif2 (e.g. 6GHz link). [ 297.455876] Unable to handle kernel read from unreadable memory at virtual address 000000000000080a [ 297.464928] Mem abort info: [ 297.467722] ESR = 0x0000000096000005 [ 297.471461] EC = 0x25: DABT (current EL), IL = 32 bits [ 297.476766] SET = 0, FnV = 0 [ 297.479809] EA = 0, S1PTW = 0 [ 297.482940] FSC = 0x05: level 1 translation fault [ 297.487809] Data abort info: [ 297.490679] ISV = 0, ISS = 0x00000005, ISS2 = 0x00000000 [ 297.496156] CM = 0, WnR = 0, TnD = 0, TagAccess = 0 [ 297.501196] GCS = 0, Overlay = 0, DirtyBit = 0, Xs = 0 [ 297.506500] user pgtable: 4k pages, 39-bit VAs, pgdp=0000000107480000 [ 297.512927] [000000000000080a] pgd=08000001097fb003, p4d=08000001097fb003, pud=08000001097fb003, pmd=0000000000000000 [ 297.523532] Internal error: Oops: 0000000096000005 [#1] SMP [ 297.715393] CPU: 2 UID: 0 PID: 45 Comm: kworker/u16:2 Tainted: G O 6.12.50 #0 [ 297.723908] Tainted: [O]=OOT_MODULE [ 297.727384] Hardware name: Banana Pi BPI-R4 (2x SFP+) (DT) [ 297.732857] Workqueue: nf_ft_offload_del nf_flow_rule_route_ipv6 [nf_flow_table] [ 297.740254] pstate: 60400005 (nZCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--) [ 297.747205] pc : mt76_wed_offload_disable+0x64/0xa0 [mt76] [ 297.752688] lr : mtk_wed_flow_remove+0x58/0x80 [ 297.757126] sp : ffffffc080fe3ae0 [ 297.760430] x29: ffffffc080fe3ae0 x28: ffffffc080fe3be0 x27: 00000000deadbef7 [ 297.767557] x26: ffffff80c5ebca00 x25: 0000000000000001 x24: ffffff80c85f4c00 [ 297.774683] x23: ffffff80c1875b78 x22: ffffffc080d42cd0 x21: ffffffc080660018 [ 297.781809] x20: ffffff80c6a076d0 x19: ffffff80c6a043c8 x18: 0000000000000000 [ 297.788935] x17: 0000000000000000 x16: 0000000000000001 x15: 0000000000000000 [ 297.796060] x14: 0000000000000019 x13: ffffff80c0ad8ec0 x12: 00000000fa83b2da [ 297.803185] x11: ffffff80c02700c0 x10: ffffff80c0ad8ec0 x9 : ffffff81fef96200 [ 297.810311] x8 : ffffff80c02700c0 x7 : ffffff80c02700d0 x6 : 0000000000000002 [ 297.817435] x5 : 0000000000000400 x4 : 0000000000000000 x3 : 0000000000000000 [ 297.824561] x2 : 0000000000000001 x1 : 0000000000000800 x0 : ffffff80c6a063c8 [ 297.831686] Call trace: [ 297.834123] mt76_wed_offload_disable+0x64/0xa0 [mt76] [ 297.839254] mtk_wed_flow_remove+0x58/0x80 [ 297.843342] mtk_flow_offload_cmd+0x434/0x574 [ 297.847689] mtk_wed_setup_tc_block_cb+0x30/0x40 [ 297.852295] nf_flow_offload_ipv6_hook+0x7f4/0x964 [nf_flow_table] [ 297.858466] nf_flow_rule_route_ipv6+0x438/0x4a4 [nf_flow_table] [ 297.864463] process_one_work+0x174/0x300 [ 297.868465] worker_thread+0x278/0x430 [ 297.872204] kthread+0xd8/0xdc [ 297.875251] ret_from_fork+0x10/0x20 [ 297.878820] Code: 928b5ae0 8b000273 91400a60 f943fa61 (79401421) [ 297.884901] ---[ end trace 0000000000000000 ]--- Fix the issue detecting the proper wed reference to use running wed callabacks. Fixes: 83eafc9251d6 ("wifi: mt76: mt7996: add wed tx support") Tested-by: Daniel Pawlik Tested-by: Matteo Croce Signed-off-by: Lorenzo Bianconi Link: https://patch.msgid.link/20251008-wed-fixes-v1-1-8f7678583385@kernel.org Signed-off-by: Felix Fietkau --- drivers/net/wireless/mediatek/mt76/mt76.h | 9 +++++++++ drivers/net/wireless/mediatek/mt76/mt7996/mmio.c | 1 + drivers/net/wireless/mediatek/mt76/wed.c | 10 +++++----- include/linux/soc/mediatek/mtk_wed.h | 1 + 4 files changed, 16 insertions(+), 5 deletions(-) diff --git a/drivers/net/wireless/mediatek/mt76/mt76.h b/drivers/net/wireless/mediatek/mt76/mt76.h index 86b812f68c97..e0a036b03b03 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76.h +++ b/drivers/net/wireless/mediatek/mt76/mt76.h @@ -1260,6 +1260,15 @@ static inline int mt76_wed_dma_setup(struct mt76_dev *dev, struct mt76_queue *q, #define mt76_dereference(p, dev) \ rcu_dereference_protected(p, lockdep_is_held(&(dev)->mutex)) +static inline struct mt76_dev *mt76_wed_to_dev(struct mtk_wed_device *wed) +{ +#ifdef CONFIG_NET_MEDIATEK_SOC_WED + if (wed->wlan.hif2) + return container_of(wed, struct mt76_dev, mmio.wed_hif2); +#endif /* CONFIG_NET_MEDIATEK_SOC_WED */ + return container_of(wed, struct mt76_dev, mmio.wed); +} + static inline struct mt76_wcid * __mt76_wcid_ptr(struct mt76_dev *dev, u16 idx) { diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mmio.c b/drivers/net/wireless/mediatek/mt76/mt7996/mmio.c index d14b626ee511..80db102ed809 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/mmio.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/mmio.c @@ -595,6 +595,7 @@ int mt7996_mmio_wed_init(struct mt7996_dev *dev, void *pdev_ptr, wed->wlan.nbuf = MT7996_HW_TOKEN_SIZE; wed->wlan.token_start = MT7996_TOKEN_SIZE - wed->wlan.nbuf; + wed->wlan.hif2 = hif2; wed->wlan.amsdu_max_subframes = 8; wed->wlan.amsdu_max_len = 1536; diff --git a/drivers/net/wireless/mediatek/mt76/wed.c b/drivers/net/wireless/mediatek/mt76/wed.c index 907a8e43e72a..fbd7e59c73aa 100644 --- a/drivers/net/wireless/mediatek/mt76/wed.c +++ b/drivers/net/wireless/mediatek/mt76/wed.c @@ -8,7 +8,7 @@ void mt76_wed_release_rx_buf(struct mtk_wed_device *wed) { - struct mt76_dev *dev = container_of(wed, struct mt76_dev, mmio.wed); + struct mt76_dev *dev = mt76_wed_to_dev(wed); int i; for (i = 0; i < dev->rx_token_size; i++) { @@ -31,8 +31,8 @@ EXPORT_SYMBOL_GPL(mt76_wed_release_rx_buf); #ifdef CONFIG_NET_MEDIATEK_SOC_WED u32 mt76_wed_init_rx_buf(struct mtk_wed_device *wed, int size) { - struct mt76_dev *dev = container_of(wed, struct mt76_dev, mmio.wed); struct mtk_wed_bm_desc *desc = wed->rx_buf_ring.desc; + struct mt76_dev *dev = mt76_wed_to_dev(wed); struct mt76_queue *q = &dev->q_rx[MT_RXQ_MAIN]; struct mt76_txwi_cache *t = NULL; int i; @@ -80,7 +80,7 @@ EXPORT_SYMBOL_GPL(mt76_wed_init_rx_buf); int mt76_wed_offload_enable(struct mtk_wed_device *wed) { - struct mt76_dev *dev = container_of(wed, struct mt76_dev, mmio.wed); + struct mt76_dev *dev = mt76_wed_to_dev(wed); spin_lock_bh(&dev->token_lock); dev->token_size = wed->wlan.token_start; @@ -164,7 +164,7 @@ EXPORT_SYMBOL_GPL(mt76_wed_dma_setup); void mt76_wed_offload_disable(struct mtk_wed_device *wed) { - struct mt76_dev *dev = container_of(wed, struct mt76_dev, mmio.wed); + struct mt76_dev *dev = mt76_wed_to_dev(wed); spin_lock_bh(&dev->token_lock); dev->token_size = dev->drv->token_size; @@ -174,7 +174,7 @@ EXPORT_SYMBOL_GPL(mt76_wed_offload_disable); void mt76_wed_reset_complete(struct mtk_wed_device *wed) { - struct mt76_dev *dev = container_of(wed, struct mt76_dev, mmio.wed); + struct mt76_dev *dev = mt76_wed_to_dev(wed); complete(&dev->mmio.wed_reset_complete); } diff --git a/include/linux/soc/mediatek/mtk_wed.h b/include/linux/soc/mediatek/mtk_wed.h index c4ff6bab176d..3fa93bd65004 100644 --- a/include/linux/soc/mediatek/mtk_wed.h +++ b/include/linux/soc/mediatek/mtk_wed.h @@ -154,6 +154,7 @@ struct mtk_wed_device { bool wcid_512; bool hw_rro; bool msi; + bool hif2; u16 token_start; unsigned int nbuf; From a9730354ca26e16f80ddda3c88fd08075afe7078 Mon Sep 17 00:00:00 2001 From: Peter Chiu Date: Wed, 8 Oct 2025 12:41:49 +0200 Subject: [PATCH 1023/1381] wifi: mt76: use GFP_DMA32 for page_pool buffer allocation Set GFP_DMA32 flag for page_pool buffers allocation since the hw relies on 32-bit DMA addresses for WED offloading. Tested-by: Daniel Pawlik Tested-by: Matteo Croce Signed-off-by: Peter Chiu Co-developed-by: Lorenzo Bianconi Signed-off-by: Lorenzo Bianconi Link: https://patch.msgid.link/20251008-wed-fixes-v1-2-8f7678583385@kernel.org Signed-off-by: Felix Fietkau --- drivers/net/wireless/mediatek/mt76/mt76.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/mediatek/mt76/mt76.h b/drivers/net/wireless/mediatek/mt76/mt76.h index e0a036b03b03..37e1d9e24932 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76.h +++ b/drivers/net/wireless/mediatek/mt76/mt76.h @@ -1877,7 +1877,8 @@ mt76_get_page_pool_buf(struct mt76_queue *q, u32 *offset, u32 size) { struct page *page; - page = page_pool_dev_alloc_frag(q->page_pool, offset, size); + page = page_pool_alloc_frag(q->page_pool, offset, size, + GFP_ATOMIC | __GFP_NOWARN | GFP_DMA32); if (!page) return NULL; From 909675fd4344f73aad5f75f123bd271ada2ab9fb Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Wed, 8 Oct 2025 12:42:49 +0200 Subject: [PATCH 1024/1381] wifi: mt76: fix license/copyright of util.h The extra copyright line for Ivo van Doorn and GPL license was only there because of code that I had already removed before the initial upstream submission of mt76. Remove it and make this header file use ISC license like the rest of the source files. Link: https://patch.msgid.link/20251008104250.46292-1-nbd@nbd.name Signed-off-by: Felix Fietkau --- drivers/net/wireless/mediatek/mt76/util.h | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/net/wireless/mediatek/mt76/util.h b/drivers/net/wireless/mediatek/mt76/util.h index 260965dde94c..037d39a55f71 100644 --- a/drivers/net/wireless/mediatek/mt76/util.h +++ b/drivers/net/wireless/mediatek/mt76/util.h @@ -1,7 +1,6 @@ -/* SPDX-License-Identifier: GPL-2.0-only */ +/* SPDX-License-Identifier: ISC */ /* * Copyright (C) 2016 Felix Fietkau - * Copyright (C) 2004 - 2009 Ivo van Doorn */ #ifndef __MT76_UTIL_H From a96fed2825d8dfb068bf640419c619b5f2df4218 Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Wed, 8 Oct 2025 12:42:50 +0200 Subject: [PATCH 1025/1381] wifi: mt76: relicense to BSD-3-Clause-Clear MediaTek has asked to switch from the ISC license to BSD-3-Clause-Clear, in order to improve clarity and the legal integrity of the code. The BSD-3-Clause license includes the "no endorsement" clause, which is important for protecting the reputation of the original authors and contributors by preventing unauthorized use of their names for endorsement purposes. This clause is absent in the BSD-2-Clause license, which is more permissive but lacks this specific protection. This change also cleans up the license of some Kconfig/Makefile files, which were accidentally marked as GPL. The GPL 2.0 remains in use on mt76x0, as well as two source files in mt7615 for which the license situation still needs to be clarified. Link: https://patch.msgid.link/20251008104250.46292-2-nbd@nbd.name Signed-off-by: Felix Fietkau --- drivers/net/wireless/mediatek/mt76/Kconfig | 2 +- drivers/net/wireless/mediatek/mt76/Makefile | 2 +- drivers/net/wireless/mediatek/mt76/agg-rx.c | 2 +- drivers/net/wireless/mediatek/mt76/channel.c | 2 +- drivers/net/wireless/mediatek/mt76/debugfs.c | 2 +- drivers/net/wireless/mediatek/mt76/dma.c | 2 +- drivers/net/wireless/mediatek/mt76/dma.h | 2 +- drivers/net/wireless/mediatek/mt76/eeprom.c | 2 +- drivers/net/wireless/mediatek/mt76/mac80211.c | 2 +- drivers/net/wireless/mediatek/mt76/mcu.c | 2 +- drivers/net/wireless/mediatek/mt76/mmio.c | 2 +- drivers/net/wireless/mediatek/mt76/mt76.h | 2 +- drivers/net/wireless/mediatek/mt76/mt7603/Kconfig | 2 +- drivers/net/wireless/mediatek/mt76/mt7603/Makefile | 2 +- drivers/net/wireless/mediatek/mt76/mt7603/beacon.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7603/core.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7603/debugfs.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7603/dma.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7603/eeprom.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7603/eeprom.h | 2 +- drivers/net/wireless/mediatek/mt76/mt7603/init.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7603/mac.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7603/mac.h | 2 +- drivers/net/wireless/mediatek/mt76/mt7603/main.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7603/mcu.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7603/mcu.h | 2 +- drivers/net/wireless/mediatek/mt76/mt7603/mt7603.h | 2 +- drivers/net/wireless/mediatek/mt76/mt7603/pci.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7603/regs.h | 2 +- drivers/net/wireless/mediatek/mt76/mt7603/soc.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7615/Kconfig | 2 +- drivers/net/wireless/mediatek/mt76/mt7615/Makefile | 2 +- drivers/net/wireless/mediatek/mt76/mt7615/debugfs.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7615/dma.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7615/eeprom.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7615/eeprom.h | 2 +- drivers/net/wireless/mediatek/mt76/mt7615/init.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7615/mac.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7615/mac.h | 2 +- drivers/net/wireless/mediatek/mt76/mt7615/main.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7615/mcu.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7615/mcu.h | 2 +- drivers/net/wireless/mediatek/mt76/mt7615/mmio.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7615/mt7615.h | 2 +- drivers/net/wireless/mediatek/mt76/mt7615/mt7615_trace.h | 2 +- drivers/net/wireless/mediatek/mt76/mt7615/pci.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7615/pci_init.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7615/pci_mac.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7615/regs.h | 2 +- drivers/net/wireless/mediatek/mt76/mt7615/sdio.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7615/soc.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7615/testmode.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7615/trace.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7615/usb.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7615/usb_sdio.c | 2 +- drivers/net/wireless/mediatek/mt76/mt76_connac.h | 2 +- drivers/net/wireless/mediatek/mt76/mt76_connac2_mac.h | 2 +- drivers/net/wireless/mediatek/mt76/mt76_connac3_mac.c | 2 +- drivers/net/wireless/mediatek/mt76/mt76_connac3_mac.h | 2 +- drivers/net/wireless/mediatek/mt76/mt76_connac_mac.c | 2 +- drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c | 2 +- drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h | 2 +- drivers/net/wireless/mediatek/mt76/mt76x0/pci.c | 2 +- drivers/net/wireless/mediatek/mt76/mt76x0/pci_mcu.c | 2 +- drivers/net/wireless/mediatek/mt76/mt76x0/usb_mcu.c | 2 +- drivers/net/wireless/mediatek/mt76/mt76x02.h | 2 +- drivers/net/wireless/mediatek/mt76/mt76x02_beacon.c | 2 +- drivers/net/wireless/mediatek/mt76/mt76x02_debugfs.c | 2 +- drivers/net/wireless/mediatek/mt76/mt76x02_dfs.c | 2 +- drivers/net/wireless/mediatek/mt76/mt76x02_dfs.h | 2 +- drivers/net/wireless/mediatek/mt76/mt76x02_dma.h | 2 +- drivers/net/wireless/mediatek/mt76/mt76x02_eeprom.c | 2 +- drivers/net/wireless/mediatek/mt76/mt76x02_eeprom.h | 2 +- drivers/net/wireless/mediatek/mt76/mt76x02_mac.c | 2 +- drivers/net/wireless/mediatek/mt76/mt76x02_mac.h | 2 +- drivers/net/wireless/mediatek/mt76/mt76x02_mcu.c | 2 +- drivers/net/wireless/mediatek/mt76/mt76x02_mcu.h | 2 +- drivers/net/wireless/mediatek/mt76/mt76x02_mmio.c | 2 +- drivers/net/wireless/mediatek/mt76/mt76x02_phy.c | 2 +- drivers/net/wireless/mediatek/mt76/mt76x02_phy.h | 2 +- drivers/net/wireless/mediatek/mt76/mt76x02_regs.h | 2 +- drivers/net/wireless/mediatek/mt76/mt76x02_trace.c | 2 +- drivers/net/wireless/mediatek/mt76/mt76x02_trace.h | 2 +- drivers/net/wireless/mediatek/mt76/mt76x02_txrx.c | 2 +- drivers/net/wireless/mediatek/mt76/mt76x02_usb.h | 2 +- drivers/net/wireless/mediatek/mt76/mt76x02_usb_core.c | 2 +- drivers/net/wireless/mediatek/mt76/mt76x02_usb_mcu.c | 2 +- drivers/net/wireless/mediatek/mt76/mt76x02_util.c | 2 +- drivers/net/wireless/mediatek/mt76/mt76x2/Kconfig | 2 +- drivers/net/wireless/mediatek/mt76/mt76x2/Makefile | 2 +- drivers/net/wireless/mediatek/mt76/mt76x2/eeprom.c | 2 +- drivers/net/wireless/mediatek/mt76/mt76x2/eeprom.h | 2 +- drivers/net/wireless/mediatek/mt76/mt76x2/init.c | 2 +- drivers/net/wireless/mediatek/mt76/mt76x2/mac.c | 2 +- drivers/net/wireless/mediatek/mt76/mt76x2/mac.h | 2 +- drivers/net/wireless/mediatek/mt76/mt76x2/mcu.c | 2 +- drivers/net/wireless/mediatek/mt76/mt76x2/mcu.h | 2 +- drivers/net/wireless/mediatek/mt76/mt76x2/mt76x2.h | 2 +- drivers/net/wireless/mediatek/mt76/mt76x2/mt76x2u.h | 2 +- drivers/net/wireless/mediatek/mt76/mt76x2/pci.c | 2 +- drivers/net/wireless/mediatek/mt76/mt76x2/pci_init.c | 2 +- drivers/net/wireless/mediatek/mt76/mt76x2/pci_main.c | 2 +- drivers/net/wireless/mediatek/mt76/mt76x2/pci_mcu.c | 2 +- drivers/net/wireless/mediatek/mt76/mt76x2/pci_phy.c | 2 +- drivers/net/wireless/mediatek/mt76/mt76x2/phy.c | 2 +- drivers/net/wireless/mediatek/mt76/mt76x2/usb.c | 2 +- drivers/net/wireless/mediatek/mt76/mt76x2/usb_init.c | 2 +- drivers/net/wireless/mediatek/mt76/mt76x2/usb_mac.c | 2 +- drivers/net/wireless/mediatek/mt76/mt76x2/usb_main.c | 2 +- drivers/net/wireless/mediatek/mt76/mt76x2/usb_mcu.c | 2 +- drivers/net/wireless/mediatek/mt76/mt76x2/usb_phy.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7915/Kconfig | 2 +- drivers/net/wireless/mediatek/mt76/mt7915/Makefile | 2 +- drivers/net/wireless/mediatek/mt76/mt7915/coredump.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7915/coredump.h | 2 +- drivers/net/wireless/mediatek/mt76/mt7915/debugfs.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7915/dma.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7915/eeprom.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7915/eeprom.h | 2 +- drivers/net/wireless/mediatek/mt76/mt7915/init.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7915/mac.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7915/mac.h | 2 +- drivers/net/wireless/mediatek/mt76/mt7915/main.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7915/mcu.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7915/mcu.h | 2 +- drivers/net/wireless/mediatek/mt76/mt7915/mmio.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7915/mt7915.h | 2 +- drivers/net/wireless/mediatek/mt76/mt7915/pci.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7915/regs.h | 2 +- drivers/net/wireless/mediatek/mt76/mt7915/soc.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7915/testmode.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7915/testmode.h | 2 +- drivers/net/wireless/mediatek/mt76/mt7921/Kconfig | 2 +- drivers/net/wireless/mediatek/mt76/mt7921/Makefile | 2 +- drivers/net/wireless/mediatek/mt76/mt7921/debugfs.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7921/init.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7921/mac.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7921/main.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7921/mcu.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7921/mcu.h | 2 +- drivers/net/wireless/mediatek/mt76/mt7921/mt7921.h | 2 +- drivers/net/wireless/mediatek/mt76/mt7921/pci.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7921/pci_mac.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7921/pci_mcu.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7921/regs.h | 2 +- drivers/net/wireless/mediatek/mt76/mt7921/sdio.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7921/sdio_mac.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7921/sdio_mcu.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7921/testmode.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7921/usb.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7925/Kconfig | 2 +- drivers/net/wireless/mediatek/mt76/mt7925/Makefile | 2 +- drivers/net/wireless/mediatek/mt76/mt7925/debugfs.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7925/init.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7925/mac.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7925/mac.h | 2 +- drivers/net/wireless/mediatek/mt76/mt7925/main.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7925/mcu.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7925/mcu.h | 2 +- drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h | 2 +- drivers/net/wireless/mediatek/mt76/mt7925/pci.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7925/pci_mac.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7925/pci_mcu.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7925/regs.h | 2 +- drivers/net/wireless/mediatek/mt76/mt7925/testmode.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7925/usb.c | 2 +- drivers/net/wireless/mediatek/mt76/mt792x.h | 2 +- drivers/net/wireless/mediatek/mt76/mt792x_acpi_sar.c | 2 +- drivers/net/wireless/mediatek/mt76/mt792x_acpi_sar.h | 2 +- drivers/net/wireless/mediatek/mt76/mt792x_core.c | 2 +- drivers/net/wireless/mediatek/mt76/mt792x_debugfs.c | 2 +- drivers/net/wireless/mediatek/mt76/mt792x_dma.c | 2 +- drivers/net/wireless/mediatek/mt76/mt792x_mac.c | 2 +- drivers/net/wireless/mediatek/mt76/mt792x_regs.h | 2 +- drivers/net/wireless/mediatek/mt76/mt792x_trace.c | 2 +- drivers/net/wireless/mediatek/mt76/mt792x_trace.h | 2 +- drivers/net/wireless/mediatek/mt76/mt792x_usb.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7996/Kconfig | 2 +- drivers/net/wireless/mediatek/mt76/mt7996/Makefile | 2 +- drivers/net/wireless/mediatek/mt76/mt7996/coredump.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7996/coredump.h | 2 +- drivers/net/wireless/mediatek/mt76/mt7996/debugfs.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7996/dma.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7996/eeprom.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7996/eeprom.h | 2 +- drivers/net/wireless/mediatek/mt76/mt7996/init.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7996/mac.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7996/mac.h | 2 +- drivers/net/wireless/mediatek/mt76/mt7996/main.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7996/mcu.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7996/mcu.h | 2 +- drivers/net/wireless/mediatek/mt76/mt7996/mmio.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h | 2 +- drivers/net/wireless/mediatek/mt76/mt7996/pci.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7996/regs.h | 2 +- drivers/net/wireless/mediatek/mt76/pci.c | 2 +- drivers/net/wireless/mediatek/mt76/scan.c | 2 +- drivers/net/wireless/mediatek/mt76/sdio.c | 2 +- drivers/net/wireless/mediatek/mt76/sdio.h | 2 +- drivers/net/wireless/mediatek/mt76/sdio_txrx.c | 2 +- drivers/net/wireless/mediatek/mt76/testmode.c | 2 +- drivers/net/wireless/mediatek/mt76/testmode.h | 2 +- drivers/net/wireless/mediatek/mt76/trace.c | 2 +- drivers/net/wireless/mediatek/mt76/trace.h | 2 +- drivers/net/wireless/mediatek/mt76/tx.c | 2 +- drivers/net/wireless/mediatek/mt76/usb.c | 2 +- drivers/net/wireless/mediatek/mt76/usb_trace.c | 2 +- drivers/net/wireless/mediatek/mt76/usb_trace.h | 2 +- drivers/net/wireless/mediatek/mt76/util.c | 2 +- drivers/net/wireless/mediatek/mt76/util.h | 2 +- drivers/net/wireless/mediatek/mt76/wed.c | 2 +- 211 files changed, 211 insertions(+), 211 deletions(-) diff --git a/drivers/net/wireless/mediatek/mt76/Kconfig b/drivers/net/wireless/mediatek/mt76/Kconfig index a86f800b8bf5..d0aff94075d6 100644 --- a/drivers/net/wireless/mediatek/mt76/Kconfig +++ b/drivers/net/wireless/mediatek/mt76/Kconfig @@ -1,4 +1,4 @@ -# SPDX-License-Identifier: GPL-2.0-only +# SPDX-License-Identifier: BSD-3-Clause-Clear config MT76_CORE tristate select PAGE_POOL diff --git a/drivers/net/wireless/mediatek/mt76/Makefile b/drivers/net/wireless/mediatek/mt76/Makefile index 87512d101a91..53e7bcefe770 100644 --- a/drivers/net/wireless/mediatek/mt76/Makefile +++ b/drivers/net/wireless/mediatek/mt76/Makefile @@ -1,4 +1,4 @@ -# SPDX-License-Identifier: GPL-2.0-only +# SPDX-License-Identifier: BSD-3-Clause-Clear obj-$(CONFIG_MT76_CORE) += mt76.o obj-$(CONFIG_MT76_USB) += mt76-usb.o obj-$(CONFIG_MT76_SDIO) += mt76-sdio.o diff --git a/drivers/net/wireless/mediatek/mt76/agg-rx.c b/drivers/net/wireless/mediatek/mt76/agg-rx.c index 936ab1ca9246..3d34caf7e4f7 100644 --- a/drivers/net/wireless/mediatek/mt76/agg-rx.c +++ b/drivers/net/wireless/mediatek/mt76/agg-rx.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2018 Felix Fietkau */ diff --git a/drivers/net/wireless/mediatek/mt76/channel.c b/drivers/net/wireless/mediatek/mt76/channel.c index 130af1b254db..2b705bdb7993 100644 --- a/drivers/net/wireless/mediatek/mt76/channel.c +++ b/drivers/net/wireless/mediatek/mt76/channel.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2024 Felix Fietkau */ diff --git a/drivers/net/wireless/mediatek/mt76/debugfs.c b/drivers/net/wireless/mediatek/mt76/debugfs.c index bee1177594d3..a5ac6ca86735 100644 --- a/drivers/net/wireless/mediatek/mt76/debugfs.c +++ b/drivers/net/wireless/mediatek/mt76/debugfs.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2016 Felix Fietkau */ diff --git a/drivers/net/wireless/mediatek/mt76/dma.c b/drivers/net/wireless/mediatek/mt76/dma.c index 1fa7de1d2c45..27aa591ff2ea 100644 --- a/drivers/net/wireless/mediatek/mt76/dma.c +++ b/drivers/net/wireless/mediatek/mt76/dma.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2016 Felix Fietkau */ diff --git a/drivers/net/wireless/mediatek/mt76/dma.h b/drivers/net/wireless/mediatek/mt76/dma.h index 17a80e1757fc..d41f978d4d7d 100644 --- a/drivers/net/wireless/mediatek/mt76/dma.h +++ b/drivers/net/wireless/mediatek/mt76/dma.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* * Copyright (C) 2016 Felix Fietkau */ diff --git a/drivers/net/wireless/mediatek/mt76/eeprom.c b/drivers/net/wireless/mediatek/mt76/eeprom.c index cbda19a86882..573400d57ce7 100644 --- a/drivers/net/wireless/mediatek/mt76/eeprom.c +++ b/drivers/net/wireless/mediatek/mt76/eeprom.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2016 Felix Fietkau */ diff --git a/drivers/net/wireless/mediatek/mt76/mac80211.c b/drivers/net/wireless/mediatek/mt76/mac80211.c index 5ceaf78c9ea0..4be47aea2253 100644 --- a/drivers/net/wireless/mediatek/mt76/mac80211.c +++ b/drivers/net/wireless/mediatek/mt76/mac80211.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2016 Felix Fietkau */ diff --git a/drivers/net/wireless/mediatek/mt76/mcu.c b/drivers/net/wireless/mediatek/mt76/mcu.c index 65d4c2adb538..535c3d8a9cc0 100644 --- a/drivers/net/wireless/mediatek/mt76/mcu.c +++ b/drivers/net/wireless/mediatek/mt76/mcu.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2019 Lorenzo Bianconi */ diff --git a/drivers/net/wireless/mediatek/mt76/mmio.c b/drivers/net/wireless/mediatek/mt76/mmio.c index cd2e9737c3bf..0ea97295f5c4 100644 --- a/drivers/net/wireless/mediatek/mt76/mmio.c +++ b/drivers/net/wireless/mediatek/mt76/mmio.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2016 Felix Fietkau */ diff --git a/drivers/net/wireless/mediatek/mt76/mt76.h b/drivers/net/wireless/mediatek/mt76/mt76.h index 37e1d9e24932..e7a00669fe01 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76.h +++ b/drivers/net/wireless/mediatek/mt76/mt76.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* * Copyright (C) 2016 Felix Fietkau */ diff --git a/drivers/net/wireless/mediatek/mt76/mt7603/Kconfig b/drivers/net/wireless/mediatek/mt76/mt7603/Kconfig index dd16acfd9735..ae40a596e49c 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7603/Kconfig +++ b/drivers/net/wireless/mediatek/mt76/mt7603/Kconfig @@ -1,4 +1,4 @@ -# SPDX-License-Identifier: GPL-2.0-only +# SPDX-License-Identifier: BSD-3-Clause-Clear config MT7603E tristate "MediaTek MT7603E (PCIe) and MT76x8 WLAN support" select MT76_CORE diff --git a/drivers/net/wireless/mediatek/mt76/mt7603/Makefile b/drivers/net/wireless/mediatek/mt76/mt7603/Makefile index 6878e305c24d..e954165ee133 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7603/Makefile +++ b/drivers/net/wireless/mediatek/mt76/mt7603/Makefile @@ -1,4 +1,4 @@ -# SPDX-License-Identifier: GPL-2.0-only +# SPDX-License-Identifier: BSD-3-Clause-Clear obj-$(CONFIG_MT7603E) += mt7603e.o mt7603e-y := \ diff --git a/drivers/net/wireless/mediatek/mt76/mt7603/beacon.c b/drivers/net/wireless/mediatek/mt76/mt7603/beacon.c index 6457ee06bb5a..300a7f9c2ef1 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7603/beacon.c +++ b/drivers/net/wireless/mediatek/mt76/mt7603/beacon.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear #include "mt7603.h" diff --git a/drivers/net/wireless/mediatek/mt76/mt7603/core.c b/drivers/net/wireless/mediatek/mt76/mt7603/core.c index 915b8349146a..9c2943fd904e 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7603/core.c +++ b/drivers/net/wireless/mediatek/mt76/mt7603/core.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear #include "mt7603.h" #include "../trace.h" diff --git a/drivers/net/wireless/mediatek/mt76/mt7603/debugfs.c b/drivers/net/wireless/mediatek/mt76/mt7603/debugfs.c index 3967f2f05774..c891ad5498e6 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7603/debugfs.c +++ b/drivers/net/wireless/mediatek/mt76/mt7603/debugfs.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear #include "mt7603.h" diff --git a/drivers/net/wireless/mediatek/mt76/mt7603/dma.c b/drivers/net/wireless/mediatek/mt76/mt7603/dma.c index e26cc78fff94..3a16851524a0 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7603/dma.c +++ b/drivers/net/wireless/mediatek/mt76/mt7603/dma.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear #include "mt7603.h" #include "mac.h" diff --git a/drivers/net/wireless/mediatek/mt76/mt7603/eeprom.c b/drivers/net/wireless/mediatek/mt76/mt7603/eeprom.c index 88382b537a33..b89db2db6573 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7603/eeprom.c +++ b/drivers/net/wireless/mediatek/mt76/mt7603/eeprom.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear #include #include "mt7603.h" diff --git a/drivers/net/wireless/mediatek/mt76/mt7603/eeprom.h b/drivers/net/wireless/mediatek/mt76/mt7603/eeprom.h index 4687d6dc00dc..b6b746d1e56f 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7603/eeprom.h +++ b/drivers/net/wireless/mediatek/mt76/mt7603/eeprom.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ #ifndef __MT7603_EEPROM_H #define __MT7603_EEPROM_H diff --git a/drivers/net/wireless/mediatek/mt76/mt7603/init.c b/drivers/net/wireless/mediatek/mt76/mt7603/init.c index 86617a3e4328..10f2ec70c792 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7603/init.c +++ b/drivers/net/wireless/mediatek/mt76/mt7603/init.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear #include #include "mt7603.h" diff --git a/drivers/net/wireless/mediatek/mt76/mt7603/mac.c b/drivers/net/wireless/mediatek/mt76/mt7603/mac.c index 6387f9e61060..d3110eeb45d7 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7603/mac.c +++ b/drivers/net/wireless/mediatek/mt76/mt7603/mac.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear #include #include diff --git a/drivers/net/wireless/mediatek/mt76/mt7603/mac.h b/drivers/net/wireless/mediatek/mt76/mt7603/mac.h index 17e34ecf2bfb..9f5fab51ff83 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7603/mac.h +++ b/drivers/net/wireless/mediatek/mt76/mt7603/mac.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ #ifndef __MT7603_MAC_H #define __MT7603_MAC_H diff --git a/drivers/net/wireless/mediatek/mt76/mt7603/main.c b/drivers/net/wireless/mediatek/mt76/mt7603/main.c index 0d7c84941cd0..0f3a7508996c 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7603/main.c +++ b/drivers/net/wireless/mediatek/mt76/mt7603/main.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear #include #include diff --git a/drivers/net/wireless/mediatek/mt76/mt7603/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7603/mcu.c index 301668c3cc92..e432cce97640 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7603/mcu.c +++ b/drivers/net/wireless/mediatek/mt76/mt7603/mcu.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear #include #include "mt7603.h" diff --git a/drivers/net/wireless/mediatek/mt76/mt7603/mcu.h b/drivers/net/wireless/mediatek/mt76/mt7603/mcu.h index 30df8a3fd11a..7debe76cd092 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7603/mcu.h +++ b/drivers/net/wireless/mediatek/mt76/mt7603/mcu.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ #ifndef __MT7603_MCU_H #define __MT7603_MCU_H diff --git a/drivers/net/wireless/mediatek/mt76/mt7603/mt7603.h b/drivers/net/wireless/mediatek/mt76/mt7603/mt7603.h index 55a034ccbacd..071bfab3af7c 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7603/mt7603.h +++ b/drivers/net/wireless/mediatek/mt76/mt7603/mt7603.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ #ifndef __MT7603_H #define __MT7603_H diff --git a/drivers/net/wireless/mediatek/mt76/mt7603/pci.c b/drivers/net/wireless/mediatek/mt76/mt7603/pci.c index 3d94cdb4314a..5fee610597a4 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7603/pci.c +++ b/drivers/net/wireless/mediatek/mt76/mt7603/pci.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear #include #include diff --git a/drivers/net/wireless/mediatek/mt76/mt7603/regs.h b/drivers/net/wireless/mediatek/mt76/mt7603/regs.h index 524bceb8e958..97942f5ebdb4 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7603/regs.h +++ b/drivers/net/wireless/mediatek/mt76/mt7603/regs.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ #ifndef __MT7603_REGS_H #define __MT7603_REGS_H diff --git a/drivers/net/wireless/mediatek/mt76/mt7603/soc.c b/drivers/net/wireless/mediatek/mt76/mt7603/soc.c index 1dd372372048..b74256efba55 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7603/soc.c +++ b/drivers/net/wireless/mediatek/mt76/mt7603/soc.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear #include #include diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/Kconfig b/drivers/net/wireless/mediatek/mt76/mt7615/Kconfig index 1ab1439143f4..8cc0c1b5c24e 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7615/Kconfig +++ b/drivers/net/wireless/mediatek/mt76/mt7615/Kconfig @@ -1,4 +1,4 @@ -# SPDX-License-Identifier: GPL-2.0-only +# SPDX-License-Identifier: BSD-3-Clause-Clear config MT7615_COMMON tristate diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/Makefile b/drivers/net/wireless/mediatek/mt76/mt7615/Makefile index 2b97b9dde477..4def3b13eae1 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7615/Makefile +++ b/drivers/net/wireless/mediatek/mt76/mt7615/Makefile @@ -1,4 +1,4 @@ -# SPDX-License-Identifier: ISC +# SPDX-License-Identifier: BSD-3-Clause-Clear obj-$(CONFIG_MT7615_COMMON) += mt7615-common.o obj-$(CONFIG_MT7615E) += mt7615e.o diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/debugfs.c b/drivers/net/wireless/mediatek/mt76/mt7615/debugfs.c index 2a6d317db5e0..0f7b20152279 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7615/debugfs.c +++ b/drivers/net/wireless/mediatek/mt76/mt7615/debugfs.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear #include "mt7615.h" diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/dma.c b/drivers/net/wireless/mediatek/mt76/mt7615/dma.c index bcf7864312d7..59d2b3e8696b 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7615/dma.c +++ b/drivers/net/wireless/mediatek/mt76/mt7615/dma.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2019 MediaTek Inc. * * Author: Ryder Lee diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/eeprom.c b/drivers/net/wireless/mediatek/mt76/mt7615/eeprom.c index d4bc7e11e772..d6828e1cda19 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7615/eeprom.c +++ b/drivers/net/wireless/mediatek/mt76/mt7615/eeprom.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2019 MediaTek Inc. * * Author: Ryder Lee diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/eeprom.h b/drivers/net/wireless/mediatek/mt76/mt7615/eeprom.h index a67fbb90f5b3..6aed52e14181 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7615/eeprom.h +++ b/drivers/net/wireless/mediatek/mt76/mt7615/eeprom.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* Copyright (C) 2019 MediaTek Inc. */ #ifndef __MT7615_EEPROM_H diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/init.c b/drivers/net/wireless/mediatek/mt76/mt7615/init.c index 3e7af3e58736..42e11ba1206e 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7615/init.c +++ b/drivers/net/wireless/mediatek/mt76/mt7615/init.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2019 MediaTek Inc. * * Author: Roy Luo diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/mac.c b/drivers/net/wireless/mediatek/mt76/mt7615/mac.c index f8d2cc94b742..bd56cdb022a2 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7615/mac.c +++ b/drivers/net/wireless/mediatek/mt76/mt7615/mac.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2019 MediaTek Inc. * * Author: Ryder Lee diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/mac.h b/drivers/net/wireless/mediatek/mt76/mt7615/mac.h index d08fbe64c262..336ebce5db5f 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7615/mac.h +++ b/drivers/net/wireless/mediatek/mt76/mt7615/mac.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* Copyright (C) 2019 MediaTek Inc. */ #ifndef __MT7615_MAC_H diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/main.c b/drivers/net/wireless/mediatek/mt76/mt7615/main.c index 15fe155ac3f3..727266892c3d 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7615/main.c +++ b/drivers/net/wireless/mediatek/mt76/mt7615/main.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2019 MediaTek Inc. * * Author: Roy Luo diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7615/mcu.c index 4064e193d4de..52bbedda672b 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7615/mcu.c +++ b/drivers/net/wireless/mediatek/mt76/mt7615/mcu.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2019 MediaTek Inc. * * Author: Roy Luo diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/mcu.h b/drivers/net/wireless/mediatek/mt76/mt7615/mcu.h index 8e9604be0792..851b0e4839b5 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7615/mcu.h +++ b/drivers/net/wireless/mediatek/mt76/mt7615/mcu.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* Copyright (C) 2019 MediaTek Inc. */ #ifndef __MT7615_MCU_H diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/mmio.c b/drivers/net/wireless/mediatek/mt76/mt7615/mmio.c index dbb2c82407df..da7edd48ce2c 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7615/mmio.c +++ b/drivers/net/wireless/mediatek/mt76/mt7615/mmio.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2020 MediaTek Inc. */ #include diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/mt7615.h b/drivers/net/wireless/mediatek/mt76/mt7615/mt7615.h index 9bdd29e8d25e..e16865dd8e52 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7615/mt7615.h +++ b/drivers/net/wireless/mediatek/mt76/mt7615/mt7615.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* Copyright (C) 2019 MediaTek Inc. */ #ifndef __MT7615_H diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/mt7615_trace.h b/drivers/net/wireless/mediatek/mt76/mt7615/mt7615_trace.h index 9be5a58a4e6d..697fc5d225de 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7615/mt7615_trace.h +++ b/drivers/net/wireless/mediatek/mt76/mt7615/mt7615_trace.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* * Copyright (C) 2019 Lorenzo Bianconi */ diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/pci.c b/drivers/net/wireless/mediatek/mt76/mt7615/pci.c index 68010e27f065..f5018bfa317a 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7615/pci.c +++ b/drivers/net/wireless/mediatek/mt76/mt7615/pci.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2019 MediaTek Inc. * * Author: Ryder Lee diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/pci_init.c b/drivers/net/wireless/mediatek/mt76/mt7615/pci_init.c index f607eee3fb47..a9178e077fd6 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7615/pci_init.c +++ b/drivers/net/wireless/mediatek/mt76/mt7615/pci_init.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2019 MediaTek Inc. * * Author: Roy Luo diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/pci_mac.c b/drivers/net/wireless/mediatek/mt76/mt7615/pci_mac.c index b795d11d943d..53cb1eed1e4f 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7615/pci_mac.c +++ b/drivers/net/wireless/mediatek/mt76/mt7615/pci_mac.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2020 MediaTek Inc. * * Author: Ryder Lee diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/regs.h b/drivers/net/wireless/mediatek/mt76/mt7615/regs.h index 806b3887c541..eb3c24d51987 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7615/regs.h +++ b/drivers/net/wireless/mediatek/mt76/mt7615/regs.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* Copyright (C) 2019 MediaTek Inc. */ #ifndef __MT7615_REGS_H diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/sdio.c b/drivers/net/wireless/mediatek/mt76/mt7615/sdio.c index f56038cd4d3a..46188951ad19 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7615/sdio.c +++ b/drivers/net/wireless/mediatek/mt76/mt7615/sdio.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2020 MediaTek Inc. * * Author: Felix Fietkau diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/soc.c b/drivers/net/wireless/mediatek/mt76/mt7615/soc.c index 06a0f2a141e8..4bd189dd67e3 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7615/soc.c +++ b/drivers/net/wireless/mediatek/mt76/mt7615/soc.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2019 MediaTek Inc. * * Author: Ryder Lee diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/testmode.c b/drivers/net/wireless/mediatek/mt76/mt7615/testmode.c index 03f5af84424b..6eb97b7eba2d 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7615/testmode.c +++ b/drivers/net/wireless/mediatek/mt76/mt7615/testmode.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2020 Felix Fietkau */ #include "mt7615.h" diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/trace.c b/drivers/net/wireless/mediatek/mt76/mt7615/trace.c index 6c02d5aff68f..7ec39e0b3fb2 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7615/trace.c +++ b/drivers/net/wireless/mediatek/mt76/mt7615/trace.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2019 Lorenzo Bianconi */ diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/usb.c b/drivers/net/wireless/mediatek/mt76/mt7615/usb.c index d96e06b4fee1..d91feffadda9 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7615/usb.c +++ b/drivers/net/wireless/mediatek/mt76/mt7615/usb.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2019 MediaTek Inc. * * Author: Felix Fietkau diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/usb_sdio.c b/drivers/net/wireless/mediatek/mt76/mt7615/usb_sdio.c index 820b39590027..f4169de939c4 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7615/usb_sdio.c +++ b/drivers/net/wireless/mediatek/mt76/mt7615/usb_sdio.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2020 MediaTek Inc. * * Author: Lorenzo Bianconi diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac.h b/drivers/net/wireless/mediatek/mt76/mt76_connac.h index 192dcc374a64..813d61bffc2c 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76_connac.h +++ b/drivers/net/wireless/mediatek/mt76/mt76_connac.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* Copyright (C) 2020 MediaTek Inc. */ #ifndef __MT76_CONNAC_H diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac2_mac.h b/drivers/net/wireless/mediatek/mt76/mt76_connac2_mac.h index eb4765365b8c..d4e2c3140441 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76_connac2_mac.h +++ b/drivers/net/wireless/mediatek/mt76/mt76_connac2_mac.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* Copyright (C) 2022 MediaTek Inc. */ #ifndef __MT76_CONNAC2_MAC_H diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac3_mac.c b/drivers/net/wireless/mediatek/mt76/mt76_connac3_mac.c index 2d300948308d..651fcd4169f4 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76_connac3_mac.c +++ b/drivers/net/wireless/mediatek/mt76/mt76_connac3_mac.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2023 MediaTek Inc. */ #include "mt76_connac.h" diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac3_mac.h b/drivers/net/wireless/mediatek/mt76/mt76_connac3_mac.h index c5eaedca11e0..247e2e7a47d8 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76_connac3_mac.h +++ b/drivers/net/wireless/mediatek/mt76/mt76_connac3_mac.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* Copyright (C) 2023 MediaTek Inc. */ #ifndef __MT76_CONNAC3_MAC_H diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac_mac.c b/drivers/net/wireless/mediatek/mt76/mt76_connac_mac.c index 0db00efe88b0..7af6a578562f 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76_connac_mac.c +++ b/drivers/net/wireless/mediatek/mt76/mt76_connac_mac.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2020 MediaTek Inc. */ #include "mt76_connac.h" diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c index 451375f2901b..fba7025ffd3f 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c +++ b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2020 MediaTek Inc. */ #include diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h index 27daf419560a..927138130842 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h +++ b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* Copyright (C) 2020 MediaTek Inc. */ #ifndef __MT76_CONNAC_MCU_H diff --git a/drivers/net/wireless/mediatek/mt76/mt76x0/pci.c b/drivers/net/wireless/mediatek/mt76/mt76x0/pci.c index 11c16d1fc70f..f8d206a07f99 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x0/pci.c +++ b/drivers/net/wireless/mediatek/mt76/mt76x0/pci.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2016 Felix Fietkau */ diff --git a/drivers/net/wireless/mediatek/mt76/mt76x0/pci_mcu.c b/drivers/net/wireless/mediatek/mt76/mt76x0/pci_mcu.c index f0962507f72f..efa549dc68ec 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x0/pci_mcu.c +++ b/drivers/net/wireless/mediatek/mt76/mt76x0/pci_mcu.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2018 Lorenzo Bianconi */ diff --git a/drivers/net/wireless/mediatek/mt76/mt76x0/usb_mcu.c b/drivers/net/wireless/mediatek/mt76/mt76x0/usb_mcu.c index 6dc1f51f5658..20a8f3659490 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x0/usb_mcu.c +++ b/drivers/net/wireless/mediatek/mt76/mt76x0/usb_mcu.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2018 Lorenzo Bianconi */ diff --git a/drivers/net/wireless/mediatek/mt76/mt76x02.h b/drivers/net/wireless/mediatek/mt76/mt76x02.h index 8d06ef8c7c62..3c98808ccf26 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x02.h +++ b/drivers/net/wireless/mediatek/mt76/mt76x02.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* * Copyright (C) 2016 Felix Fietkau * Copyright (C) 2018 Stanislaw Gruszka diff --git a/drivers/net/wireless/mediatek/mt76/mt76x02_beacon.c b/drivers/net/wireless/mediatek/mt76/mt76x02_beacon.c index d570b99bccb9..7c9b16942275 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x02_beacon.c +++ b/drivers/net/wireless/mediatek/mt76/mt76x02_beacon.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2016 Felix Fietkau * Copyright (C) 2018 Lorenzo Bianconi diff --git a/drivers/net/wireless/mediatek/mt76/mt76x02_debugfs.c b/drivers/net/wireless/mediatek/mt76/mt76x02_debugfs.c index 8ce4bf44733d..d81f696b32c7 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x02_debugfs.c +++ b/drivers/net/wireless/mediatek/mt76/mt76x02_debugfs.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2016 Felix Fietkau */ diff --git a/drivers/net/wireless/mediatek/mt76/mt76x02_dfs.c b/drivers/net/wireless/mediatek/mt76/mt76x02_dfs.c index 7a07636d09c6..21c99ad7ef59 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x02_dfs.c +++ b/drivers/net/wireless/mediatek/mt76/mt76x02_dfs.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2016 Lorenzo Bianconi */ diff --git a/drivers/net/wireless/mediatek/mt76/mt76x02_dfs.h b/drivers/net/wireless/mediatek/mt76/mt76x02_dfs.h index 491010a32247..d40051992586 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x02_dfs.h +++ b/drivers/net/wireless/mediatek/mt76/mt76x02_dfs.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* * Copyright (C) 2016 Lorenzo Bianconi */ diff --git a/drivers/net/wireless/mediatek/mt76/mt76x02_dma.h b/drivers/net/wireless/mediatek/mt76/mt76x02_dma.h index 23b0e7d10d57..2f6ba8cf51e8 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x02_dma.h +++ b/drivers/net/wireless/mediatek/mt76/mt76x02_dma.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* * Copyright (C) 2018 Lorenzo Bianconi */ diff --git a/drivers/net/wireless/mediatek/mt76/mt76x02_eeprom.c b/drivers/net/wireless/mediatek/mt76/mt76x02_eeprom.c index a5e3392c0b48..d16be0cb0dc7 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x02_eeprom.c +++ b/drivers/net/wireless/mediatek/mt76/mt76x02_eeprom.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2016 Felix Fietkau * Copyright (C) 2018 Lorenzo Bianconi diff --git a/drivers/net/wireless/mediatek/mt76/mt76x02_eeprom.h b/drivers/net/wireless/mediatek/mt76/mt76x02_eeprom.h index 13fa70853b0d..3cbb2977f375 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x02_eeprom.h +++ b/drivers/net/wireless/mediatek/mt76/mt76x02_eeprom.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* * Copyright (C) 2016 Felix Fietkau * Copyright (C) 2018 Lorenzo Bianconi diff --git a/drivers/net/wireless/mediatek/mt76/mt76x02_mac.c b/drivers/net/wireless/mediatek/mt76/mt76x02_mac.c index 83488b2d6efb..14ee5b3b94d3 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x02_mac.c +++ b/drivers/net/wireless/mediatek/mt76/mt76x02_mac.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2016 Felix Fietkau * Copyright (C) 2018 Stanislaw Gruszka diff --git a/drivers/net/wireless/mediatek/mt76/mt76x02_mac.h b/drivers/net/wireless/mediatek/mt76/mt76x02_mac.h index 5dc6c834111e..778454ac8974 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x02_mac.h +++ b/drivers/net/wireless/mediatek/mt76/mt76x02_mac.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* * Copyright (C) 2016 Felix Fietkau * Copyright (C) 2018 Stanislaw Gruszka diff --git a/drivers/net/wireless/mediatek/mt76/mt76x02_mcu.c b/drivers/net/wireless/mediatek/mt76/mt76x02_mcu.c index 75978820a260..e16f06a284ca 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x02_mcu.c +++ b/drivers/net/wireless/mediatek/mt76/mt76x02_mcu.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2016 Felix Fietkau * Copyright (C) 2018 Lorenzo Bianconi diff --git a/drivers/net/wireless/mediatek/mt76/mt76x02_mcu.h b/drivers/net/wireless/mediatek/mt76/mt76x02_mcu.h index e187ed52968e..a422cdc520cb 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x02_mcu.h +++ b/drivers/net/wireless/mediatek/mt76/mt76x02_mcu.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* * Copyright (C) 2018 Lorenzo Bianconi */ diff --git a/drivers/net/wireless/mediatek/mt76/mt76x02_mmio.c b/drivers/net/wireless/mediatek/mt76/mt76x02_mmio.c index a683d53c7ceb..dd71c1c95cc9 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x02_mmio.c +++ b/drivers/net/wireless/mediatek/mt76/mt76x02_mmio.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2016 Felix Fietkau * Copyright (C) 2018 Lorenzo Bianconi diff --git a/drivers/net/wireless/mediatek/mt76/mt76x02_phy.c b/drivers/net/wireless/mediatek/mt76/mt76x02_phy.c index cbe7e6f0c29a..557380c9bfab 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x02_phy.c +++ b/drivers/net/wireless/mediatek/mt76/mt76x02_phy.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2016 Felix Fietkau * Copyright (C) 2018 Lorenzo Bianconi diff --git a/drivers/net/wireless/mediatek/mt76/mt76x02_phy.h b/drivers/net/wireless/mediatek/mt76/mt76x02_phy.h index 84d8a6138b3e..09e8edee2195 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x02_phy.h +++ b/drivers/net/wireless/mediatek/mt76/mt76x02_phy.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* * Copyright (C) 2018 Lorenzo Bianconi */ diff --git a/drivers/net/wireless/mediatek/mt76/mt76x02_regs.h b/drivers/net/wireless/mediatek/mt76/mt76x02_regs.h index fe0c5e3298bc..e87d3f8a1de9 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x02_regs.h +++ b/drivers/net/wireless/mediatek/mt76/mt76x02_regs.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* * Copyright (C) 2016 Felix Fietkau */ diff --git a/drivers/net/wireless/mediatek/mt76/mt76x02_trace.c b/drivers/net/wireless/mediatek/mt76/mt76x02_trace.c index a812c3a1e258..a92b2b7391ff 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x02_trace.c +++ b/drivers/net/wireless/mediatek/mt76/mt76x02_trace.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2016 Felix Fietkau */ diff --git a/drivers/net/wireless/mediatek/mt76/mt76x02_trace.h b/drivers/net/wireless/mediatek/mt76/mt76x02_trace.h index 11d119cd0f6f..56eea2f68983 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x02_trace.h +++ b/drivers/net/wireless/mediatek/mt76/mt76x02_trace.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* * Copyright (C) 2016 Felix Fietkau */ diff --git a/drivers/net/wireless/mediatek/mt76/mt76x02_txrx.c b/drivers/net/wireless/mediatek/mt76/mt76x02_txrx.c index d8bc4ae185f5..301b43180006 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x02_txrx.c +++ b/drivers/net/wireless/mediatek/mt76/mt76x02_txrx.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2016 Felix Fietkau * Copyright (C) 2018 Lorenzo Bianconi diff --git a/drivers/net/wireless/mediatek/mt76/mt76x02_usb.h b/drivers/net/wireless/mediatek/mt76/mt76x02_usb.h index b5be884b3549..49ab05c1fe73 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x02_usb.h +++ b/drivers/net/wireless/mediatek/mt76/mt76x02_usb.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* * Copyright (C) 2018 Lorenzo Bianconi */ diff --git a/drivers/net/wireless/mediatek/mt76/mt76x02_usb_core.c b/drivers/net/wireless/mediatek/mt76/mt76x02_usb_core.c index 4840d0b500b3..3a28a8cc1338 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x02_usb_core.c +++ b/drivers/net/wireless/mediatek/mt76/mt76x02_usb_core.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2018 Lorenzo Bianconi */ diff --git a/drivers/net/wireless/mediatek/mt76/mt76x02_usb_mcu.c b/drivers/net/wireless/mediatek/mt76/mt76x02_usb_mcu.c index b2cc44914294..968c73e06a5f 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x02_usb_mcu.c +++ b/drivers/net/wireless/mediatek/mt76/mt76x02_usb_mcu.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2018 Lorenzo Bianconi */ diff --git a/drivers/net/wireless/mediatek/mt76/mt76x02_util.c b/drivers/net/wireless/mediatek/mt76/mt76x02_util.c index 7dfcb20c692c..e5d9d1bc9415 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x02_util.c +++ b/drivers/net/wireless/mediatek/mt76/mt76x02_util.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2018 Stanislaw Gruszka * Copyright (C) 2016 Felix Fietkau diff --git a/drivers/net/wireless/mediatek/mt76/mt76x2/Kconfig b/drivers/net/wireless/mediatek/mt76/mt76x2/Kconfig index 482a32b70ddf..d820510cb4bb 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x2/Kconfig +++ b/drivers/net/wireless/mediatek/mt76/mt76x2/Kconfig @@ -1,4 +1,4 @@ -# SPDX-License-Identifier: GPL-2.0-only +# SPDX-License-Identifier: BSD-3-Clause-Clear config MT76x2_COMMON tristate select MT76x02_LIB diff --git a/drivers/net/wireless/mediatek/mt76/mt76x2/Makefile b/drivers/net/wireless/mediatek/mt76/mt76x2/Makefile index caf089538c11..cbc90a9616a6 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x2/Makefile +++ b/drivers/net/wireless/mediatek/mt76/mt76x2/Makefile @@ -1,4 +1,4 @@ -# SPDX-License-Identifier: GPL-2.0-only +# SPDX-License-Identifier: BSD-3-Clause-Clear obj-$(CONFIG_MT76x2_COMMON) += mt76x2-common.o obj-$(CONFIG_MT76x2E) += mt76x2e.o obj-$(CONFIG_MT76x2U) += mt76x2u.o diff --git a/drivers/net/wireless/mediatek/mt76/mt76x2/eeprom.c b/drivers/net/wireless/mediatek/mt76/mt76x2/eeprom.c index 221805deb42f..782813aadc0a 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x2/eeprom.c +++ b/drivers/net/wireless/mediatek/mt76/mt76x2/eeprom.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2016 Felix Fietkau */ diff --git a/drivers/net/wireless/mediatek/mt76/mt76x2/eeprom.h b/drivers/net/wireless/mediatek/mt76/mt76x2/eeprom.h index 43430ef98b11..1ee8be389b24 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x2/eeprom.h +++ b/drivers/net/wireless/mediatek/mt76/mt76x2/eeprom.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* * Copyright (C) 2016 Felix Fietkau */ diff --git a/drivers/net/wireless/mediatek/mt76/mt76x2/init.c b/drivers/net/wireless/mediatek/mt76/mt76x2/init.c index 19c139290adb..408dc08b6457 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x2/init.c +++ b/drivers/net/wireless/mediatek/mt76/mt76x2/init.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2016 Felix Fietkau * Copyright (C) 2018 Lorenzo Bianconi diff --git a/drivers/net/wireless/mediatek/mt76/mt76x2/mac.c b/drivers/net/wireless/mediatek/mt76/mt76x2/mac.c index e08740ca3d0c..2fa34ca69095 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x2/mac.c +++ b/drivers/net/wireless/mediatek/mt76/mt76x2/mac.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2016 Felix Fietkau * Copyright (C) 2018 Lorenzo Bianconi diff --git a/drivers/net/wireless/mediatek/mt76/mt76x2/mac.h b/drivers/net/wireless/mediatek/mt76/mt76x2/mac.h index d5c3d26b94c1..f8ea70074c41 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x2/mac.h +++ b/drivers/net/wireless/mediatek/mt76/mt76x2/mac.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* * Copyright (C) 2016 Felix Fietkau */ diff --git a/drivers/net/wireless/mediatek/mt76/mt76x2/mcu.c b/drivers/net/wireless/mediatek/mt76/mt76x2/mcu.c index ac83ce5f3e8b..769d924220e3 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x2/mcu.c +++ b/drivers/net/wireless/mediatek/mt76/mt76x2/mcu.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2016 Felix Fietkau * Copyright (C) 2018 Lorenzo Bianconi diff --git a/drivers/net/wireless/mediatek/mt76/mt76x2/mcu.h b/drivers/net/wireless/mediatek/mt76/mt76x2/mcu.h index 41fd66563e82..16a4386aa754 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x2/mcu.h +++ b/drivers/net/wireless/mediatek/mt76/mt76x2/mcu.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* * Copyright (C) 2016 Felix Fietkau */ diff --git a/drivers/net/wireless/mediatek/mt76/mt76x2/mt76x2.h b/drivers/net/wireless/mediatek/mt76/mt76x2/mt76x2.h index f051721bb00e..984756c81bdc 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x2/mt76x2.h +++ b/drivers/net/wireless/mediatek/mt76/mt76x2/mt76x2.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* * Copyright (C) 2016 Felix Fietkau */ diff --git a/drivers/net/wireless/mediatek/mt76/mt76x2/mt76x2u.h b/drivers/net/wireless/mediatek/mt76/mt76x2/mt76x2u.h index f9d37c6cf1f0..27e478ab5b15 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x2/mt76x2u.h +++ b/drivers/net/wireless/mediatek/mt76/mt76x2/mt76x2u.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* * Copyright (C) 2018 Lorenzo Bianconi */ diff --git a/drivers/net/wireless/mediatek/mt76/mt76x2/pci.c b/drivers/net/wireless/mediatek/mt76/mt76x2/pci.c index 2303019670e2..491a32921a06 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x2/pci.c +++ b/drivers/net/wireless/mediatek/mt76/mt76x2/pci.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2016 Felix Fietkau */ diff --git a/drivers/net/wireless/mediatek/mt76/mt76x2/pci_init.c b/drivers/net/wireless/mediatek/mt76/mt76x2/pci_init.c index e38e8e5685c2..bec84f932311 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x2/pci_init.c +++ b/drivers/net/wireless/mediatek/mt76/mt76x2/pci_init.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2016 Felix Fietkau */ diff --git a/drivers/net/wireless/mediatek/mt76/mt76x2/pci_main.c b/drivers/net/wireless/mediatek/mt76/mt76x2/pci_main.c index c5dfb06d81e8..550644676201 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x2/pci_main.c +++ b/drivers/net/wireless/mediatek/mt76/mt76x2/pci_main.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2016 Felix Fietkau */ diff --git a/drivers/net/wireless/mediatek/mt76/mt76x2/pci_mcu.c b/drivers/net/wireless/mediatek/mt76/mt76x2/pci_mcu.c index e5b6282d1a6c..daba163802b6 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x2/pci_mcu.c +++ b/drivers/net/wireless/mediatek/mt76/mt76x2/pci_mcu.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2016 Felix Fietkau */ diff --git a/drivers/net/wireless/mediatek/mt76/mt76x2/pci_phy.c b/drivers/net/wireless/mediatek/mt76/mt76x2/pci_phy.c index 8831337df23e..dcf4328c1cac 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x2/pci_phy.c +++ b/drivers/net/wireless/mediatek/mt76/mt76x2/pci_phy.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2016 Felix Fietkau */ diff --git a/drivers/net/wireless/mediatek/mt76/mt76x2/phy.c b/drivers/net/wireless/mediatek/mt76/mt76x2/phy.c index e2b4cf30dc44..a5efa13a892f 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x2/phy.c +++ b/drivers/net/wireless/mediatek/mt76/mt76x2/phy.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2016 Felix Fietkau * Copyright (C) 2018 Lorenzo Bianconi diff --git a/drivers/net/wireless/mediatek/mt76/mt76x2/usb.c b/drivers/net/wireless/mediatek/mt76/mt76x2/usb.c index 96cecc576a98..01cb3b2830f3 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x2/usb.c +++ b/drivers/net/wireless/mediatek/mt76/mt76x2/usb.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2018 Lorenzo Bianconi */ diff --git a/drivers/net/wireless/mediatek/mt76/mt76x2/usb_init.c b/drivers/net/wireless/mediatek/mt76/mt76x2/usb_init.c index 3b5562811511..41778a8ef02c 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x2/usb_init.c +++ b/drivers/net/wireless/mediatek/mt76/mt76x2/usb_init.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2018 Lorenzo Bianconi */ diff --git a/drivers/net/wireless/mediatek/mt76/mt76x2/usb_mac.c b/drivers/net/wireless/mediatek/mt76/mt76x2/usb_mac.c index eaa622833f85..d0cb511ac6a2 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x2/usb_mac.c +++ b/drivers/net/wireless/mediatek/mt76/mt76x2/usb_mac.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2018 Lorenzo Bianconi */ diff --git a/drivers/net/wireless/mediatek/mt76/mt76x2/usb_main.c b/drivers/net/wireless/mediatek/mt76/mt76x2/usb_main.c index 6671c53faf9f..66b06a493d95 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x2/usb_main.c +++ b/drivers/net/wireless/mediatek/mt76/mt76x2/usb_main.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2018 Lorenzo Bianconi */ diff --git a/drivers/net/wireless/mediatek/mt76/mt76x2/usb_mcu.c b/drivers/net/wireless/mediatek/mt76/mt76x2/usb_mcu.c index dd22d8af0901..9102be1803b7 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x2/usb_mcu.c +++ b/drivers/net/wireless/mediatek/mt76/mt76x2/usb_mcu.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2018 Lorenzo Bianconi */ diff --git a/drivers/net/wireless/mediatek/mt76/mt76x2/usb_phy.c b/drivers/net/wireless/mediatek/mt76/mt76x2/usb_phy.c index a04a98f5ce1e..b63dd7f3ee80 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76x2/usb_phy.c +++ b/drivers/net/wireless/mediatek/mt76/mt76x2/usb_phy.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2018 Lorenzo Bianconi */ diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/Kconfig b/drivers/net/wireless/mediatek/mt76/mt7915/Kconfig index 193112c49bd1..c24be8227f11 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7915/Kconfig +++ b/drivers/net/wireless/mediatek/mt76/mt7915/Kconfig @@ -1,4 +1,4 @@ -# SPDX-License-Identifier: ISC +# SPDX-License-Identifier: BSD-3-Clause-Clear config MT7915E tristate "MediaTek MT7915E (PCIe) support" select MT76_CONNAC_LIB diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/Makefile b/drivers/net/wireless/mediatek/mt76/mt7915/Makefile index e0ca638c91a5..963fb3109af3 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7915/Makefile +++ b/drivers/net/wireless/mediatek/mt76/mt7915/Makefile @@ -1,4 +1,4 @@ -# SPDX-License-Identifier: ISC +# SPDX-License-Identifier: BSD-3-Clause-Clear obj-$(CONFIG_MT7915E) += mt7915e.o diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/coredump.c b/drivers/net/wireless/mediatek/mt76/mt7915/coredump.c index 5daf2258dfe6..6c7273e5b0bc 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7915/coredump.c +++ b/drivers/net/wireless/mediatek/mt76/mt7915/coredump.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2022 MediaTek Inc. */ #include diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/coredump.h b/drivers/net/wireless/mediatek/mt76/mt7915/coredump.h index 709f8e9c795c..eb3991eda2e5 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7915/coredump.h +++ b/drivers/net/wireless/mediatek/mt76/mt7915/coredump.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* Copyright (C) 2022 MediaTek Inc. */ #ifndef _COREDUMP_H_ diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/debugfs.c b/drivers/net/wireless/mediatek/mt76/mt7915/debugfs.c index 6ff0d262c28a..26ed3745af43 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7915/debugfs.c +++ b/drivers/net/wireless/mediatek/mt76/mt7915/debugfs.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2020 MediaTek Inc. */ #include diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/dma.c b/drivers/net/wireless/mediatek/mt76/mt7915/dma.c index 009ef713f437..aabd37366e86 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7915/dma.c +++ b/drivers/net/wireless/mediatek/mt76/mt7915/dma.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2020 MediaTek Inc. */ #include "mt7915.h" diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/eeprom.c b/drivers/net/wireless/mediatek/mt76/mt7915/eeprom.c index 38dfd5de365c..eb92cbf1a284 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7915/eeprom.c +++ b/drivers/net/wireless/mediatek/mt76/mt7915/eeprom.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2020 MediaTek Inc. */ #include diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/eeprom.h b/drivers/net/wireless/mediatek/mt76/mt7915/eeprom.h index 73611c9d26e1..1dc285c72991 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7915/eeprom.h +++ b/drivers/net/wireless/mediatek/mt76/mt7915/eeprom.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* Copyright (C) 2020 MediaTek Inc. */ #ifndef __MT7915_EEPROM_H diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/init.c b/drivers/net/wireless/mediatek/mt76/mt7915/init.c index 1ac7ee2922b0..22443cbc74ad 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7915/init.c +++ b/drivers/net/wireless/mediatek/mt76/mt7915/init.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2020 MediaTek Inc. */ #include diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/mac.c b/drivers/net/wireless/mediatek/mt76/mt7915/mac.c index 1c0d310146d6..cfbe457039ce 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7915/mac.c +++ b/drivers/net/wireless/mediatek/mt76/mt7915/mac.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2020 MediaTek Inc. */ #include diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/mac.h b/drivers/net/wireless/mediatek/mt76/mt7915/mac.h index 448b1b380190..e39f96e00ba4 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7915/mac.h +++ b/drivers/net/wireless/mediatek/mt76/mt7915/mac.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* Copyright (C) 2020 MediaTek Inc. */ #ifndef __MT7915_MAC_H diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/main.c b/drivers/net/wireless/mediatek/mt76/mt7915/main.c index fa135447a263..90d5e79fbf74 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7915/main.c +++ b/drivers/net/wireless/mediatek/mt76/mt7915/main.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2020 MediaTek Inc. */ #include diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c index 673338d508a8..00bff4d3aab8 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c +++ b/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2020 MediaTek Inc. */ #include diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/mcu.h b/drivers/net/wireless/mediatek/mt76/mt7915/mcu.h index b72535efc6d7..3af11a075a2f 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7915/mcu.h +++ b/drivers/net/wireless/mediatek/mt76/mt7915/mcu.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* Copyright (C) 2020 MediaTek Inc. */ #ifndef __MT7915_MCU_H diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/mmio.c b/drivers/net/wireless/mediatek/mt76/mt7915/mmio.c index 36488aa6cc20..2708b1556f40 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7915/mmio.c +++ b/drivers/net/wireless/mediatek/mt76/mt7915/mmio.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2020 MediaTek Inc. */ #include diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/mt7915.h b/drivers/net/wireless/mediatek/mt76/mt7915/mt7915.h index b15d31d36a87..b5c06201b707 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7915/mt7915.h +++ b/drivers/net/wireless/mediatek/mt76/mt7915/mt7915.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* Copyright (C) 2020 MediaTek Inc. */ #ifndef __MT7915_H diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/pci.c b/drivers/net/wireless/mediatek/mt76/mt7915/pci.c index 07b0a5766eab..f6b03211a879 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7915/pci.c +++ b/drivers/net/wireless/mediatek/mt76/mt7915/pci.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2020 MediaTek Inc. * * Author: Ryder Lee diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/regs.h b/drivers/net/wireless/mediatek/mt76/mt7915/regs.h index c5ec63a25a42..307bf6a75674 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7915/regs.h +++ b/drivers/net/wireless/mediatek/mt76/mt7915/regs.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* Copyright (C) 2020 MediaTek Inc. */ #ifndef __MT7915_REGS_H diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/soc.c b/drivers/net/wireless/mediatek/mt76/mt7915/soc.c index c823a7554a3a..e22a737bd9ab 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7915/soc.c +++ b/drivers/net/wireless/mediatek/mt76/mt7915/soc.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2022 MediaTek Inc. */ #include diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/testmode.c b/drivers/net/wireless/mediatek/mt76/mt7915/testmode.c index 5836b9733f27..618a5c2bdd29 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7915/testmode.c +++ b/drivers/net/wireless/mediatek/mt76/mt7915/testmode.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2020 MediaTek Inc. */ #include "mt7915.h" diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/testmode.h b/drivers/net/wireless/mediatek/mt76/mt7915/testmode.h index 5573ac309363..bb1bc568751b 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7915/testmode.h +++ b/drivers/net/wireless/mediatek/mt76/mt7915/testmode.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* Copyright (C) 2020 MediaTek Inc. */ #ifndef __MT7915_TESTMODE_H diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/Kconfig b/drivers/net/wireless/mediatek/mt76/mt7921/Kconfig index 7ed51e057857..37b5f46e76f4 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7921/Kconfig +++ b/drivers/net/wireless/mediatek/mt76/mt7921/Kconfig @@ -1,4 +1,4 @@ -# SPDX-License-Identifier: ISC +# SPDX-License-Identifier: BSD-3-Clause-Clear config MT7921_COMMON tristate select MT792x_LIB diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/Makefile b/drivers/net/wireless/mediatek/mt76/mt7921/Makefile index 849be9e848e0..2ad3c1cc3779 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7921/Makefile +++ b/drivers/net/wireless/mediatek/mt76/mt7921/Makefile @@ -1,4 +1,4 @@ -# SPDX-License-Identifier: ISC +# SPDX-License-Identifier: BSD-3-Clause-Clear obj-$(CONFIG_MT7921_COMMON) += mt7921-common.o obj-$(CONFIG_MT7921E) += mt7921e.o diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/debugfs.c b/drivers/net/wireless/mediatek/mt76/mt7921/debugfs.c index 616b66a3fde2..4333005b3ad9 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7921/debugfs.c +++ b/drivers/net/wireless/mediatek/mt76/mt7921/debugfs.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2020 MediaTek Inc. */ #include "mt7921.h" diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/init.c b/drivers/net/wireless/mediatek/mt76/mt7921/init.c index b9098a7331b1..d8410c77abe4 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7921/init.c +++ b/drivers/net/wireless/mediatek/mt76/mt7921/init.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2020 MediaTek Inc. */ #include diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/mac.c b/drivers/net/wireless/mediatek/mt76/mt7921/mac.c index bce26389ab18..03b4960db73f 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7921/mac.c +++ b/drivers/net/wireless/mediatek/mt76/mt7921/mac.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2020 MediaTek Inc. */ #include diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/main.c b/drivers/net/wireless/mediatek/mt76/mt7921/main.c index 67383c41a319..5fae9a6e273c 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7921/main.c +++ b/drivers/net/wireless/mediatek/mt76/mt7921/main.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2020 MediaTek Inc. */ #include diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7921/mcu.c index 86bd33b916a9..8c5e6bdc0fc9 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7921/mcu.c +++ b/drivers/net/wireless/mediatek/mt76/mt7921/mcu.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2020 MediaTek Inc. */ #include diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/mcu.h b/drivers/net/wireless/mediatek/mt76/mt7921/mcu.h index 2834c6c53e58..de676b83b89c 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7921/mcu.h +++ b/drivers/net/wireless/mediatek/mt76/mt7921/mcu.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* Copyright (C) 2020 MediaTek Inc. */ #ifndef __MT7921_MCU_H diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/mt7921.h b/drivers/net/wireless/mediatek/mt76/mt7921/mt7921.h index c88793fcec64..83fc7f49ff84 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7921/mt7921.h +++ b/drivers/net/wireless/mediatek/mt76/mt7921/mt7921.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* Copyright (C) 2020 MediaTek Inc. */ #ifndef __MT7921_H diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/pci.c b/drivers/net/wireless/mediatek/mt76/mt7921/pci.c index a0c9df3c2cc7..ec9686183251 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7921/pci.c +++ b/drivers/net/wireless/mediatek/mt76/mt7921/pci.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2020 MediaTek Inc. * */ diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/pci_mac.c b/drivers/net/wireless/mediatek/mt76/mt7921/pci_mac.c index 881812ba03ff..5ec084432ae3 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7921/pci_mac.c +++ b/drivers/net/wireless/mediatek/mt76/mt7921/pci_mac.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2021 MediaTek Inc. */ #include "mt7921.h" diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/pci_mcu.c b/drivers/net/wireless/mediatek/mt76/mt7921/pci_mcu.c index 4cf1f2f0f968..8439c849a7a6 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7921/pci_mcu.c +++ b/drivers/net/wireless/mediatek/mt76/mt7921/pci_mcu.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2021 MediaTek Inc. */ #include "mt7921.h" diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/regs.h b/drivers/net/wireless/mediatek/mt76/mt7921/regs.h index 43427a3a48af..4d9eaf1e0692 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7921/regs.h +++ b/drivers/net/wireless/mediatek/mt76/mt7921/regs.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* Copyright (C) 2020 MediaTek Inc. */ #ifndef __MT7921_REGS_H diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/sdio.c b/drivers/net/wireless/mediatek/mt76/mt7921/sdio.c index d8d36b3c3068..3421e53dc948 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7921/sdio.c +++ b/drivers/net/wireless/mediatek/mt76/mt7921/sdio.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2021 MediaTek Inc. * */ diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/sdio_mac.c b/drivers/net/wireless/mediatek/mt76/mt7921/sdio_mac.c index a9eb6252a904..416d49e53499 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7921/sdio_mac.c +++ b/drivers/net/wireless/mediatek/mt76/mt7921/sdio_mac.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2021 MediaTek Inc. */ #include diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/sdio_mcu.c b/drivers/net/wireless/mediatek/mt76/mt7921/sdio_mcu.c index 5e4501d7f1c0..14e66f3f5aad 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7921/sdio_mcu.c +++ b/drivers/net/wireless/mediatek/mt76/mt7921/sdio_mcu.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2021 MediaTek Inc. */ #include diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/testmode.c b/drivers/net/wireless/mediatek/mt76/mt7921/testmode.c index e838d93477c1..e60ee992edf8 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7921/testmode.c +++ b/drivers/net/wireless/mediatek/mt76/mt7921/testmode.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear #include "mt7921.h" #include "mcu.h" diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/usb.c b/drivers/net/wireless/mediatek/mt76/mt7921/usb.c index 100bdba32ba5..17057e68bf21 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7921/usb.c +++ b/drivers/net/wireless/mediatek/mt76/mt7921/usb.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2022 MediaTek Inc. * * Author: Lorenzo Bianconi diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/Kconfig b/drivers/net/wireless/mediatek/mt76/mt7925/Kconfig index 5854e95e68a5..f4f7c93c2ea7 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/Kconfig +++ b/drivers/net/wireless/mediatek/mt76/mt7925/Kconfig @@ -1,4 +1,4 @@ -# SPDX-License-Identifier: ISC +# SPDX-License-Identifier: BSD-3-Clause-Clear config MT7925_COMMON tristate select MT792x_LIB diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/Makefile b/drivers/net/wireless/mediatek/mt76/mt7925/Makefile index ade5e647c941..5d769de979c0 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/Makefile +++ b/drivers/net/wireless/mediatek/mt76/mt7925/Makefile @@ -1,4 +1,4 @@ -# SPDX-License-Identifier: ISC +# SPDX-License-Identifier: BSD-3-Clause-Clear obj-$(CONFIG_MT7925_COMMON) += mt7925-common.o obj-$(CONFIG_MT7925E) += mt7925e.o diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/debugfs.c b/drivers/net/wireless/mediatek/mt76/mt7925/debugfs.c index 1e2fc6577e78..e2498659c884 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/debugfs.c +++ b/drivers/net/wireless/mediatek/mt76/mt7925/debugfs.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2023 MediaTek Inc. */ #include "mt7925.h" diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/init.c b/drivers/net/wireless/mediatek/mt76/mt7925/init.c index d7d5afe365ed..ea397ad48c30 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/init.c +++ b/drivers/net/wireless/mediatek/mt76/mt7925/init.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2023 MediaTek Inc. */ #include diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mac.c b/drivers/net/wireless/mediatek/mt76/mt7925/mac.c index 1e44e96f034e..19c9bec7c7e6 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/mac.c +++ b/drivers/net/wireless/mediatek/mt76/mt7925/mac.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2023 MediaTek Inc. */ #include diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mac.h b/drivers/net/wireless/mediatek/mt76/mt7925/mac.h index b10a993326b9..83ea9021daea 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/mac.h +++ b/drivers/net/wireless/mediatek/mt76/mt7925/mac.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* Copyright (C) 2023 MediaTek Inc. */ #ifndef __MT7925_MAC_H diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c index ac3d485a2f78..13424e2fe447 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c +++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2023 MediaTek Inc. */ #include diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c index 8eda407e4135..620f61abee0b 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c +++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2023 MediaTek Inc. */ #include diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h index a40764d89a1f..1a11e977a244 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h +++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* Copyright (C) 2023 MediaTek Inc. */ #ifndef __MT7925_MCU_H diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h b/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h index 1b165d0d8bd3..f46eadce31fc 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h +++ b/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* Copyright (C) 2023 MediaTek Inc. */ #ifndef __MT7925_H diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/pci.c b/drivers/net/wireless/mediatek/mt76/mt7925/pci.c index 8eb1fe1082d1..1de5e41a4061 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/pci.c +++ b/drivers/net/wireless/mediatek/mt76/mt7925/pci.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2023 MediaTek Inc. */ #include diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/pci_mac.c b/drivers/net/wireless/mediatek/mt76/mt7925/pci_mac.c index 4578d16bf456..3072850c2752 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/pci_mac.c +++ b/drivers/net/wireless/mediatek/mt76/mt7925/pci_mac.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2023 MediaTek Inc. */ #include "mt7925.h" diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/pci_mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/pci_mcu.c index f95bc5dcd830..6cceff88c656 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/pci_mcu.c +++ b/drivers/net/wireless/mediatek/mt76/mt7925/pci_mcu.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2023 MediaTek Inc. */ #include "mt7925.h" diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/regs.h b/drivers/net/wireless/mediatek/mt76/mt7925/regs.h index 341987e47f67..24985bba1b90 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/regs.h +++ b/drivers/net/wireless/mediatek/mt76/mt7925/regs.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* Copyright (C) 2023 MediaTek Inc. */ #ifndef __MT7925_REGS_H diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/testmode.c b/drivers/net/wireless/mediatek/mt76/mt7925/testmode.c index a3c97164ba21..3d40aacfc011 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/testmode.c +++ b/drivers/net/wireless/mediatek/mt76/mt7925/testmode.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear #include "mt7925.h" #include "mcu.h" diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/usb.c b/drivers/net/wireless/mediatek/mt76/mt7925/usb.c index bf040f34e4b9..d9968f03856d 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/usb.c +++ b/drivers/net/wireless/mediatek/mt76/mt7925/usb.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2023 MediaTek Inc. */ #include diff --git a/drivers/net/wireless/mediatek/mt76/mt792x.h b/drivers/net/wireless/mediatek/mt76/mt792x.h index f2c8b9e4aa0f..da7831f9efec 100644 --- a/drivers/net/wireless/mediatek/mt76/mt792x.h +++ b/drivers/net/wireless/mediatek/mt76/mt792x.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* Copyright (C) 2023 MediaTek Inc. */ #ifndef __MT792X_H diff --git a/drivers/net/wireless/mediatek/mt76/mt792x_acpi_sar.c b/drivers/net/wireless/mediatek/mt76/mt792x_acpi_sar.c index d1aebadd50aa..946dd7956e4a 100644 --- a/drivers/net/wireless/mediatek/mt76/mt792x_acpi_sar.c +++ b/drivers/net/wireless/mediatek/mt76/mt792x_acpi_sar.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2023 MediaTek Inc. */ #include diff --git a/drivers/net/wireless/mediatek/mt76/mt792x_acpi_sar.h b/drivers/net/wireless/mediatek/mt76/mt792x_acpi_sar.h index e45dcd7fbdb1..474033073831 100644 --- a/drivers/net/wireless/mediatek/mt76/mt792x_acpi_sar.h +++ b/drivers/net/wireless/mediatek/mt76/mt792x_acpi_sar.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* Copyright (C) 2023 MediaTek Inc. */ #ifndef __MT7921_ACPI_SAR_H diff --git a/drivers/net/wireless/mediatek/mt76/mt792x_core.c b/drivers/net/wireless/mediatek/mt76/mt792x_core.c index c0e56541a954..5b50bf94996e 100644 --- a/drivers/net/wireless/mediatek/mt76/mt792x_core.c +++ b/drivers/net/wireless/mediatek/mt76/mt792x_core.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2023 MediaTek Inc. */ #include diff --git a/drivers/net/wireless/mediatek/mt76/mt792x_debugfs.c b/drivers/net/wireless/mediatek/mt76/mt792x_debugfs.c index 9858d9a93851..65c37e0cef8f 100644 --- a/drivers/net/wireless/mediatek/mt76/mt792x_debugfs.c +++ b/drivers/net/wireless/mediatek/mt76/mt792x_debugfs.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2023 MediaTek Inc. */ #include "mt792x.h" diff --git a/drivers/net/wireless/mediatek/mt76/mt792x_dma.c b/drivers/net/wireless/mediatek/mt76/mt792x_dma.c index 69217ce91130..1ddec7788b66 100644 --- a/drivers/net/wireless/mediatek/mt76/mt792x_dma.c +++ b/drivers/net/wireless/mediatek/mt76/mt792x_dma.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2023 MediaTek Inc. */ #include diff --git a/drivers/net/wireless/mediatek/mt76/mt792x_mac.c b/drivers/net/wireless/mediatek/mt76/mt792x_mac.c index 3f1d9ba49076..71dec93094eb 100644 --- a/drivers/net/wireless/mediatek/mt76/mt792x_mac.c +++ b/drivers/net/wireless/mediatek/mt76/mt792x_mac.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2023 MediaTek Inc. */ #include diff --git a/drivers/net/wireless/mediatek/mt76/mt792x_regs.h b/drivers/net/wireless/mediatek/mt76/mt792x_regs.h index 458cfd0260b1..acf627aed609 100644 --- a/drivers/net/wireless/mediatek/mt76/mt792x_regs.h +++ b/drivers/net/wireless/mediatek/mt76/mt792x_regs.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* Copyright (C) 2023 MediaTek Inc. */ #ifndef __MT792X_REGS_H diff --git a/drivers/net/wireless/mediatek/mt76/mt792x_trace.c b/drivers/net/wireless/mediatek/mt76/mt792x_trace.c index b6f284fb929d..ffc77d3944bd 100644 --- a/drivers/net/wireless/mediatek/mt76/mt792x_trace.c +++ b/drivers/net/wireless/mediatek/mt76/mt792x_trace.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2023 Lorenzo Bianconi */ diff --git a/drivers/net/wireless/mediatek/mt76/mt792x_trace.h b/drivers/net/wireless/mediatek/mt76/mt792x_trace.h index 61f2aa260656..7b0e3f00b194 100644 --- a/drivers/net/wireless/mediatek/mt76/mt792x_trace.h +++ b/drivers/net/wireless/mediatek/mt76/mt792x_trace.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* * Copyright (C) 2023 Lorenzo Bianconi */ diff --git a/drivers/net/wireless/mediatek/mt76/mt792x_usb.c b/drivers/net/wireless/mediatek/mt76/mt792x_usb.c index 76272a03b22e..552808458138 100644 --- a/drivers/net/wireless/mediatek/mt76/mt792x_usb.c +++ b/drivers/net/wireless/mediatek/mt76/mt792x_usb.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2023 MediaTek Inc. * * Author: Lorenzo Bianconi diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/Kconfig b/drivers/net/wireless/mediatek/mt76/mt7996/Kconfig index bb44d4a5e2dc..196e1cbab813 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/Kconfig +++ b/drivers/net/wireless/mediatek/mt76/mt7996/Kconfig @@ -1,4 +1,4 @@ -# SPDX-License-Identifier: ISC +# SPDX-License-Identifier: BSD-3-Clause-Clear config MT7996E tristate "MediaTek MT7996 (PCIe) support" select MT76_CONNAC_LIB diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/Makefile b/drivers/net/wireless/mediatek/mt76/mt7996/Makefile index 07c8b555c1ac..19c0bc02b584 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/Makefile +++ b/drivers/net/wireless/mediatek/mt76/mt7996/Makefile @@ -1,4 +1,4 @@ -# SPDX-License-Identifier: ISC +# SPDX-License-Identifier: BSD-3-Clause-Clear obj-$(CONFIG_MT7996E) += mt7996e.o diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/coredump.c b/drivers/net/wireless/mediatek/mt76/mt7996/coredump.c index 303d6e80a666..5c293ae965cd 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/coredump.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/coredump.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2023 MediaTek Inc. */ #include diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/coredump.h b/drivers/net/wireless/mediatek/mt76/mt7996/coredump.h index af2ba219b1b5..baa2f6f50832 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/coredump.h +++ b/drivers/net/wireless/mediatek/mt76/mt7996/coredump.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* Copyright (C) 2023 MediaTek Inc. */ #ifndef _COREDUMP_H_ diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/debugfs.c b/drivers/net/wireless/mediatek/mt76/mt7996/debugfs.c index 0ab827f52fd7..a000275b369b 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/debugfs.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/debugfs.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2022 MediaTek Inc. */ diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/dma.c b/drivers/net/wireless/mediatek/mt76/mt7996/dma.c index 659015f93d32..20aa67eff76f 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/dma.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/dma.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2022 MediaTek Inc. */ diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/eeprom.c b/drivers/net/wireless/mediatek/mt76/mt7996/eeprom.c index da3231c9aa11..8f60772913b4 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/eeprom.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/eeprom.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2022 MediaTek Inc. */ diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/eeprom.h b/drivers/net/wireless/mediatek/mt76/mt7996/eeprom.h index 7a771ca2434c..9e6f0e04caf9 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/eeprom.h +++ b/drivers/net/wireless/mediatek/mt76/mt7996/eeprom.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* * Copyright (C) 2022 MediaTek Inc. */ diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/init.c b/drivers/net/wireless/mediatek/mt76/mt7996/init.c index 5e95a36b42d1..5f2a2da5c326 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/init.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/init.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2022 MediaTek Inc. */ diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c index 9501def3e0e3..fd3fabc60026 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2022 MediaTek Inc. */ diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mac.h b/drivers/net/wireless/mediatek/mt76/mt7996/mac.h index e629324a5617..4eca37b013fc 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/mac.h +++ b/drivers/net/wireless/mediatek/mt76/mt7996/mac.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* * Copyright (C) 2022 MediaTek Inc. */ diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/main.c b/drivers/net/wireless/mediatek/mt76/mt7996/main.c index 2b52d057287a..848593043781 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/main.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/main.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2022 MediaTek Inc. */ diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c index afa6a43bd51e..dda01fa6d7bb 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2022 MediaTek Inc. */ diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.h b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.h index c841da1c60e5..e0b83ac9f5e2 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.h +++ b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* * Copyright (C) 2022 MediaTek Inc. */ diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mmio.c b/drivers/net/wireless/mediatek/mt76/mt7996/mmio.c index 80db102ed809..b795a72c2c82 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/mmio.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/mmio.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2022 MediaTek Inc. */ diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h b/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h index 718e4d4ad85f..d52264b23163 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h +++ b/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* * Copyright (C) 2022 MediaTek Inc. */ diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/pci.c b/drivers/net/wireless/mediatek/mt76/mt7996/pci.c index 3f49bbbba3b9..4eb55ffdf1a0 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/pci.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/pci.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2022 MediaTek Inc. */ diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/regs.h b/drivers/net/wireless/mediatek/mt76/mt7996/regs.h index 0fa325f87fcd..e48e0e575b64 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/regs.h +++ b/drivers/net/wireless/mediatek/mt76/mt7996/regs.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* * Copyright (C) 2022 MediaTek Inc. */ diff --git a/drivers/net/wireless/mediatek/mt76/pci.c b/drivers/net/wireless/mediatek/mt76/pci.c index b5031ca7f73f..833923ab2483 100644 --- a/drivers/net/wireless/mediatek/mt76/pci.c +++ b/drivers/net/wireless/mediatek/mt76/pci.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2019 Lorenzo Bianconi */ diff --git a/drivers/net/wireless/mediatek/mt76/scan.c b/drivers/net/wireless/mediatek/mt76/scan.c index 5a875aac410f..ff9176cdee3d 100644 --- a/drivers/net/wireless/mediatek/mt76/scan.c +++ b/drivers/net/wireless/mediatek/mt76/scan.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2024 Felix Fietkau */ diff --git a/drivers/net/wireless/mediatek/mt76/sdio.c b/drivers/net/wireless/mediatek/mt76/sdio.c index 8e9576747052..8bae77c761be 100644 --- a/drivers/net/wireless/mediatek/mt76/sdio.c +++ b/drivers/net/wireless/mediatek/mt76/sdio.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2020 MediaTek Inc. * * This file is written based on mt76/usb.c. diff --git a/drivers/net/wireless/mediatek/mt76/sdio.h b/drivers/net/wireless/mediatek/mt76/sdio.h index 27d5d2077eba..41b89f3de86b 100644 --- a/drivers/net/wireless/mediatek/mt76/sdio.h +++ b/drivers/net/wireless/mediatek/mt76/sdio.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* Copyright (C) 2020 MediaTek Inc. * * Author: Sean Wang diff --git a/drivers/net/wireless/mediatek/mt76/sdio_txrx.c b/drivers/net/wireless/mediatek/mt76/sdio_txrx.c index f882d21c9f63..3f314e8e1e69 100644 --- a/drivers/net/wireless/mediatek/mt76/sdio_txrx.c +++ b/drivers/net/wireless/mediatek/mt76/sdio_txrx.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2020 MediaTek Inc. * * Author: Felix Fietkau diff --git a/drivers/net/wireless/mediatek/mt76/testmode.c b/drivers/net/wireless/mediatek/mt76/testmode.c index ca4feccf38ca..6ee160bda882 100644 --- a/drivers/net/wireless/mediatek/mt76/testmode.c +++ b/drivers/net/wireless/mediatek/mt76/testmode.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* Copyright (C) 2020 Felix Fietkau */ #include diff --git a/drivers/net/wireless/mediatek/mt76/testmode.h b/drivers/net/wireless/mediatek/mt76/testmode.h index 0590c35c7126..bed1ba40ba94 100644 --- a/drivers/net/wireless/mediatek/mt76/testmode.h +++ b/drivers/net/wireless/mediatek/mt76/testmode.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* * Copyright (C) 2020 Felix Fietkau */ diff --git a/drivers/net/wireless/mediatek/mt76/trace.c b/drivers/net/wireless/mediatek/mt76/trace.c index f199fcd2a63d..f17cc01017f3 100644 --- a/drivers/net/wireless/mediatek/mt76/trace.c +++ b/drivers/net/wireless/mediatek/mt76/trace.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2016 Felix Fietkau */ diff --git a/drivers/net/wireless/mediatek/mt76/trace.h b/drivers/net/wireless/mediatek/mt76/trace.h index 109a07f9733a..794b957ac79d 100644 --- a/drivers/net/wireless/mediatek/mt76/trace.h +++ b/drivers/net/wireless/mediatek/mt76/trace.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* * Copyright (C) 2016 Felix Fietkau */ diff --git a/drivers/net/wireless/mediatek/mt76/tx.c b/drivers/net/wireless/mediatek/mt76/tx.c index b78ae6a34b65..b09b7dfab416 100644 --- a/drivers/net/wireless/mediatek/mt76/tx.c +++ b/drivers/net/wireless/mediatek/mt76/tx.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2016 Felix Fietkau */ diff --git a/drivers/net/wireless/mediatek/mt76/usb.c b/drivers/net/wireless/mediatek/mt76/usb.c index f9e67b8c3b3c..632ae755c7a6 100644 --- a/drivers/net/wireless/mediatek/mt76/usb.c +++ b/drivers/net/wireless/mediatek/mt76/usb.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2018 Lorenzo Bianconi */ diff --git a/drivers/net/wireless/mediatek/mt76/usb_trace.c b/drivers/net/wireless/mediatek/mt76/usb_trace.c index 9942bdd6177b..a04585b4b778 100644 --- a/drivers/net/wireless/mediatek/mt76/usb_trace.c +++ b/drivers/net/wireless/mediatek/mt76/usb_trace.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2018 Lorenzo Bianconi */ diff --git a/drivers/net/wireless/mediatek/mt76/usb_trace.h b/drivers/net/wireless/mediatek/mt76/usb_trace.h index 7b261ddb2ac6..93bb69c65a4f 100644 --- a/drivers/net/wireless/mediatek/mt76/usb_trace.h +++ b/drivers/net/wireless/mediatek/mt76/usb_trace.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* * Copyright (C) 2018 Lorenzo Bianconi */ diff --git a/drivers/net/wireless/mediatek/mt76/util.c b/drivers/net/wireless/mediatek/mt76/util.c index 97249ebb4bc8..83d3dc42e534 100644 --- a/drivers/net/wireless/mediatek/mt76/util.c +++ b/drivers/net/wireless/mediatek/mt76/util.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2016 Felix Fietkau */ diff --git a/drivers/net/wireless/mediatek/mt76/util.h b/drivers/net/wireless/mediatek/mt76/util.h index 037d39a55f71..617966e8de76 100644 --- a/drivers/net/wireless/mediatek/mt76/util.h +++ b/drivers/net/wireless/mediatek/mt76/util.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: ISC */ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* * Copyright (C) 2016 Felix Fietkau */ diff --git a/drivers/net/wireless/mediatek/mt76/wed.c b/drivers/net/wireless/mediatek/mt76/wed.c index fbd7e59c73aa..ed657d952de2 100644 --- a/drivers/net/wireless/mediatek/mt76/wed.c +++ b/drivers/net/wireless/mediatek/mt76/wed.c @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: ISC +// SPDX-License-Identifier: BSD-3-Clause-Clear /* * Copyright (C) 2023 Lorenzo Bianconi */ From 2157e49892c5eae210b8fa6ee8672bd9d0ffa4b5 Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Tue, 14 Oct 2025 15:28:53 +0200 Subject: [PATCH 1026/1381] wifi: mt76: mt7996: Remove useless check in mt7996_msdu_page_get_from_cache() Get rid of useless null-pointer check in mt7996_msdu_page_get_from_cache since we have already verfied the list is not empty. Fixes: b1e58e137b616 ("wifi: mt76: mt7996: Introduce RRO MSDU callbacks") Reported-by: kernel test robot Closes: https://lore.kernel.org/r/202510100155.MS0IXhzm-lkp@intel.com/ Signed-off-by: Lorenzo Bianconi Link: https://patch.msgid.link/20251014-mt7996_msdu_page_get_from_cache-remove-null-ptr-check-v1-1-fbeb7881e192@kernel.org Signed-off-by: Felix Fietkau --- drivers/net/wireless/mediatek/mt76/mt7996/mac.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c index fd3fabc60026..ace23fd67c8d 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c @@ -1681,8 +1681,7 @@ mt7996_msdu_page_get_from_cache(struct mt7996_dev *dev) if (!list_empty(&dev->wed_rro.page_cache)) { p = list_first_entry(&dev->wed_rro.page_cache, struct mt7996_msdu_page, list); - if (p) - list_del(&p->list); + list_del(&p->list); } spin_unlock(&dev->wed_rro.lock); From e627439aecf358944261cf2ffb4449c61a7e5e9f Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Fri, 17 Oct 2025 10:50:29 +0200 Subject: [PATCH 1027/1381] wifi: mt76: Move Q_READ/Q_WRITE definitions in dma.h This is a preliminary patch to enable traffic forward offloading between the MT76 NIC and the Airoha ethernet one via the Airoha NPU module available on the Airoha EN7581 SoC. Signed-off-by: Lorenzo Bianconi Link: https://patch.msgid.link/20251017-mt76-npu-devel-v2-1-ddaa90901723@kernel.org Signed-off-by: Felix Fietkau --- drivers/net/wireless/mediatek/mt76/dma.c | 31 ------------------------ drivers/net/wireless/mediatek/mt76/dma.h | 31 ++++++++++++++++++++++++ 2 files changed, 31 insertions(+), 31 deletions(-) diff --git a/drivers/net/wireless/mediatek/mt76/dma.c b/drivers/net/wireless/mediatek/mt76/dma.c index 27aa591ff2ea..2dd3bd990fba 100644 --- a/drivers/net/wireless/mediatek/mt76/dma.c +++ b/drivers/net/wireless/mediatek/mt76/dma.c @@ -7,37 +7,6 @@ #include "mt76.h" #include "dma.h" -#if IS_ENABLED(CONFIG_NET_MEDIATEK_SOC_WED) - -#define Q_READ(_q, _field) ({ \ - u32 _offset = offsetof(struct mt76_queue_regs, _field); \ - u32 _val; \ - if ((_q)->flags & MT_QFLAG_WED) \ - _val = mtk_wed_device_reg_read((_q)->wed, \ - ((_q)->wed_regs + \ - _offset)); \ - else \ - _val = readl(&(_q)->regs->_field); \ - _val; \ -}) - -#define Q_WRITE(_q, _field, _val) do { \ - u32 _offset = offsetof(struct mt76_queue_regs, _field); \ - if ((_q)->flags & MT_QFLAG_WED) \ - mtk_wed_device_reg_write((_q)->wed, \ - ((_q)->wed_regs + _offset), \ - _val); \ - else \ - writel(_val, &(_q)->regs->_field); \ -} while (0) - -#else - -#define Q_READ(_q, _field) readl(&(_q)->regs->_field) -#define Q_WRITE(_q, _field, _val) writel(_val, &(_q)->regs->_field) - -#endif - static struct mt76_txwi_cache * mt76_alloc_txwi(struct mt76_dev *dev) { diff --git a/drivers/net/wireless/mediatek/mt76/dma.h b/drivers/net/wireless/mediatek/mt76/dma.h index d41f978d4d7d..19bc768913ff 100644 --- a/drivers/net/wireless/mediatek/mt76/dma.h +++ b/drivers/net/wireless/mediatek/mt76/dma.h @@ -46,6 +46,37 @@ #define MT_FCE_INFO_LEN 4 #define MT_RX_RXWI_LEN 32 +#if IS_ENABLED(CONFIG_NET_MEDIATEK_SOC_WED) + +#define Q_READ(_q, _field) ({ \ + u32 _offset = offsetof(struct mt76_queue_regs, _field); \ + u32 _val; \ + if ((_q)->flags & MT_QFLAG_WED) \ + _val = mtk_wed_device_reg_read((_q)->wed, \ + ((_q)->wed_regs + \ + _offset)); \ + else \ + _val = readl(&(_q)->regs->_field); \ + _val; \ +}) + +#define Q_WRITE(_q, _field, _val) do { \ + u32 _offset = offsetof(struct mt76_queue_regs, _field); \ + if ((_q)->flags & MT_QFLAG_WED) \ + mtk_wed_device_reg_write((_q)->wed, \ + ((_q)->wed_regs + _offset), \ + _val); \ + else \ + writel(_val, &(_q)->regs->_field); \ +} while (0) + +#else + +#define Q_READ(_q, _field) readl(&(_q)->regs->_field) +#define Q_WRITE(_q, _field, _val) writel(_val, &(_q)->regs->_field) + +#endif + struct mt76_desc { __le32 buf0; __le32 ctrl; From a7fb9aac3e65e8b3189d4b4088ad5e0d10628876 Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Fri, 17 Oct 2025 10:50:30 +0200 Subject: [PATCH 1028/1381] wifi: mt76: Add mt76_dev pointer in mt76_queue struct. This is a preliminary patch to enable traffic forward offloading via the Airoha NPU module available on the Airoha EN7581 SoC. Signed-off-by: Lorenzo Bianconi Link: https://patch.msgid.link/20251017-mt76-npu-devel-v2-2-ddaa90901723@kernel.org Signed-off-by: Felix Fietkau --- drivers/net/wireless/mediatek/mt76/dma.c | 1 + drivers/net/wireless/mediatek/mt76/mt76.h | 1 + 2 files changed, 2 insertions(+) diff --git a/drivers/net/wireless/mediatek/mt76/dma.c b/drivers/net/wireless/mediatek/mt76/dma.c index 2dd3bd990fba..a4c1b6c66488 100644 --- a/drivers/net/wireless/mediatek/mt76/dma.c +++ b/drivers/net/wireless/mediatek/mt76/dma.c @@ -794,6 +794,7 @@ mt76_dma_alloc_queue(struct mt76_dev *dev, struct mt76_queue *q, q->ndesc = n_desc; q->buf_size = bufsize; q->hw_idx = idx; + q->dev = dev; size = mt76_queue_is_wed_rro_ind(q) ? sizeof(struct mt76_wed_rro_desc) : sizeof(struct mt76_desc); diff --git a/drivers/net/wireless/mediatek/mt76/mt76.h b/drivers/net/wireless/mediatek/mt76/mt76.h index e7a00669fe01..b0ff65427088 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76.h +++ b/drivers/net/wireless/mediatek/mt76/mt76.h @@ -247,6 +247,7 @@ struct mt76_queue { __le16 *emi_cpu_idx; struct mtk_wed_device *wed; + struct mt76_dev *dev; u32 wed_regs; dma_addr_t desc_dma; From f7632a7fdda861e1215a6b284de738d7ab575837 Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Fri, 17 Oct 2025 10:50:31 +0200 Subject: [PATCH 1029/1381] wifi: mt76: Add the capability to set TX token start ID This is a preliminary patch to enable traffic forward offloading via the Airoha NPU module available on the Airoha EN7581 SoC. Signed-off-by: Lorenzo Bianconi Link: https://patch.msgid.link/20251017-mt76-npu-devel-v2-3-ddaa90901723@kernel.org Signed-off-by: Felix Fietkau --- drivers/net/wireless/mediatek/mt76/mt76.h | 1 + drivers/net/wireless/mediatek/mt76/tx.c | 6 ++++-- 2 files changed, 5 insertions(+), 2 deletions(-) diff --git a/drivers/net/wireless/mediatek/mt76/mt76.h b/drivers/net/wireless/mediatek/mt76/mt76.h index b0ff65427088..a66b89bc4b4a 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76.h +++ b/drivers/net/wireless/mediatek/mt76/mt76.h @@ -944,6 +944,7 @@ struct mt76_dev { struct idr token; u16 wed_token_count; u16 token_count; + u16 token_start; u16 token_size; spinlock_t rx_token_lock; diff --git a/drivers/net/wireless/mediatek/mt76/tx.c b/drivers/net/wireless/mediatek/mt76/tx.c index b09b7dfab416..9ec6d0b53a84 100644 --- a/drivers/net/wireless/mediatek/mt76/tx.c +++ b/drivers/net/wireless/mediatek/mt76/tx.c @@ -847,8 +847,10 @@ int mt76_token_consume(struct mt76_dev *dev, struct mt76_txwi_cache **ptxwi) spin_lock_bh(&dev->token_lock); - token = idr_alloc(&dev->token, *ptxwi, 0, dev->token_size, GFP_ATOMIC); - if (token >= 0) + token = idr_alloc(&dev->token, *ptxwi, dev->token_start, + dev->token_start + dev->token_size, + GFP_ATOMIC); + if (token >= dev->token_start) dev->token_count++; #ifdef CONFIG_NET_MEDIATEK_SOC_WED From 7fb554b1b623c7da845521604bd05fa9570d07bc Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Fri, 17 Oct 2025 10:50:32 +0200 Subject: [PATCH 1030/1381] wifi: mt76: Introduce the NPU generic layer Add the NPU generic layer in mt76 module. NPU will be used to enable traffic forward offloading between the MT76 NIC and the Airoha ethernet one available on the Airoha EN7581 SoC using Netfilter Flowtable APIs. Signed-off-by: Lorenzo Bianconi Link: https://patch.msgid.link/20251017-mt76-npu-devel-v2-4-ddaa90901723@kernel.org Signed-off-by: Felix Fietkau --- drivers/net/wireless/mediatek/mt76/Kconfig | 4 + drivers/net/wireless/mediatek/mt76/Makefile | 1 + drivers/net/wireless/mediatek/mt76/dma.c | 41 +- drivers/net/wireless/mediatek/mt76/dma.h | 36 ++ drivers/net/wireless/mediatek/mt76/mac80211.c | 6 +- drivers/net/wireless/mediatek/mt76/mt76.h | 135 +++++ drivers/net/wireless/mediatek/mt76/npu.c | 501 ++++++++++++++++++ include/linux/soc/airoha/airoha_offload.h | 1 + 8 files changed, 718 insertions(+), 7 deletions(-) create mode 100644 drivers/net/wireless/mediatek/mt76/npu.c diff --git a/drivers/net/wireless/mediatek/mt76/Kconfig b/drivers/net/wireless/mediatek/mt76/Kconfig index d0aff94075d6..502303622a53 100644 --- a/drivers/net/wireless/mediatek/mt76/Kconfig +++ b/drivers/net/wireless/mediatek/mt76/Kconfig @@ -37,6 +37,10 @@ config MT792x_USB tristate select MT76_USB +config MT76_NPU + bool + depends on MT76_CORE + source "drivers/net/wireless/mediatek/mt76/mt76x0/Kconfig" source "drivers/net/wireless/mediatek/mt76/mt76x2/Kconfig" source "drivers/net/wireless/mediatek/mt76/mt7603/Kconfig" diff --git a/drivers/net/wireless/mediatek/mt76/Makefile b/drivers/net/wireless/mediatek/mt76/Makefile index 53e7bcefe770..1d42adfe8030 100644 --- a/drivers/net/wireless/mediatek/mt76/Makefile +++ b/drivers/net/wireless/mediatek/mt76/Makefile @@ -12,6 +12,7 @@ mt76-y := \ mmio.o util.o trace.o dma.o mac80211.o debugfs.o eeprom.o \ tx.o agg-rx.o mcu.o wed.o scan.o channel.o +mt76-$(CONFIG_MT76_NPU) += npu.o mt76-$(CONFIG_PCI) += pci.o mt76-$(CONFIG_NL80211_TESTMODE) += testmode.o diff --git a/drivers/net/wireless/mediatek/mt76/dma.c b/drivers/net/wireless/mediatek/mt76/dma.c index a4c1b6c66488..f240016ed9f0 100644 --- a/drivers/net/wireless/mediatek/mt76/dma.c +++ b/drivers/net/wireless/mediatek/mt76/dma.c @@ -189,10 +189,15 @@ static void mt76_dma_sync_idx(struct mt76_dev *dev, struct mt76_queue *q) { Q_WRITE(q, desc_base, q->desc_dma); - if (q->flags & MT_QFLAG_WED_RRO_EN) + if ((q->flags & MT_QFLAG_WED_RRO_EN) && !mt76_npu_device_active(dev)) Q_WRITE(q, ring_size, MT_DMA_RRO_EN | q->ndesc); else Q_WRITE(q, ring_size, q->ndesc); + + if (mt76_queue_is_npu_tx(q)) { + writel(q->desc_dma, &q->regs->desc_base); + writel(q->ndesc, &q->regs->ring_size); + } q->head = Q_READ(q, dma_idx); q->tail = q->head; } @@ -204,7 +209,7 @@ void mt76_dma_queue_reset(struct mt76_dev *dev, struct mt76_queue *q, return; if (!mt76_queue_is_wed_rro_ind(q) && - !mt76_queue_is_wed_rro_rxdmad_c(q)) { + !mt76_queue_is_wed_rro_rxdmad_c(q) && !mt76_queue_is_npu(q)) { int i; /* clear descriptors */ @@ -415,6 +420,7 @@ mt76_dma_tx_cleanup(struct mt76_dev *dev, struct mt76_queue *q, bool flush) while (q->queued > 0 && q->tail != last) { mt76_dma_tx_cleanup_idx(dev, q, q->tail, &entry); + mt76_npu_txdesc_cleanup(q, q->tail); mt76_queue_tx_complete(dev, q, &entry); if (entry.txwi) { @@ -649,6 +655,10 @@ mt76_dma_tx_queue_skb(struct mt76_phy *phy, struct mt76_queue *q, if (test_bit(MT76_RESET, &phy->state)) goto free_skb; + /* TODO: Take into account unlinear skbs */ + if (mt76_npu_device_active(dev) && skb_linearize(skb)) + goto free_skb; + t = mt76_get_txwi(dev); if (!t) goto free_skb; @@ -696,6 +706,9 @@ mt76_dma_tx_queue_skb(struct mt76_phy *phy, struct mt76_queue *q, if (ret < 0) goto unmap; + if (mt76_npu_device_active(dev)) + return mt76_npu_dma_add_buf(phy, q, skb, &tx_info.buf[1], txwi); + return mt76_dma_add_buf(dev, q, tx_info.buf, tx_info.nbuf, tx_info.info, tx_info.skb, t); @@ -796,8 +809,15 @@ mt76_dma_alloc_queue(struct mt76_dev *dev, struct mt76_queue *q, q->hw_idx = idx; q->dev = dev; - size = mt76_queue_is_wed_rro_ind(q) ? sizeof(struct mt76_wed_rro_desc) - : sizeof(struct mt76_desc); + if (mt76_queue_is_wed_rro_ind(q)) + size = sizeof(struct mt76_wed_rro_desc); + else if (mt76_queue_is_npu_tx(q)) + size = sizeof(struct airoha_npu_tx_dma_desc); + else if (mt76_queue_is_npu_rx(q)) + size = sizeof(struct airoha_npu_rx_dma_desc); + else + size = sizeof(struct mt76_desc); + q->desc = dmam_alloc_coherent(dev->dma_dev, q->ndesc * size, &q->desc_dma, GFP_KERNEL); if (!q->desc) @@ -813,6 +833,7 @@ mt76_dma_alloc_queue(struct mt76_dev *dev, struct mt76_queue *q, if (ret) return ret; + mt76_npu_queue_setup(dev, q); ret = mt76_wed_dma_setup(dev, q, false); if (ret) return ret; @@ -840,6 +861,11 @@ mt76_dma_rx_cleanup(struct mt76_dev *dev, struct mt76_queue *q) if (!q->ndesc) return; + if (mt76_queue_is_npu(q)) { + mt76_npu_queue_cleanup(dev, q); + return; + } + do { spin_lock_bh(&q->lock); buf = mt76_dma_dequeue(dev, q, true, NULL, NULL, &more, NULL); @@ -870,7 +896,7 @@ mt76_dma_rx_reset(struct mt76_dev *dev, enum mt76_rxq_id qid) return; if (!mt76_queue_is_wed_rro_ind(q) && - !mt76_queue_is_wed_rro_rxdmad_c(q)) { + !mt76_queue_is_wed_rro_rxdmad_c(q) && !mt76_queue_is_npu(q)) { int i; for (i = 0; i < q->ndesc; i++) @@ -890,7 +916,10 @@ mt76_dma_rx_reset(struct mt76_dev *dev, enum mt76_rxq_id qid) return; mt76_dma_sync_idx(dev, q); - mt76_dma_rx_fill_buf(dev, q, false); + if (mt76_queue_is_npu(q)) + mt76_npu_fill_rx_queue(dev, q); + else + mt76_dma_rx_fill(dev, q, false); } static void diff --git a/drivers/net/wireless/mediatek/mt76/dma.h b/drivers/net/wireless/mediatek/mt76/dma.h index 19bc768913ff..4a63de6c5bf5 100644 --- a/drivers/net/wireless/mediatek/mt76/dma.h +++ b/drivers/net/wireless/mediatek/mt76/dma.h @@ -70,6 +70,42 @@ writel(_val, &(_q)->regs->_field); \ } while (0) +#elif IS_ENABLED(CONFIG_MT76_NPU) + +#define Q_READ(_q, _field) ({ \ + u32 _offset = offsetof(struct mt76_queue_regs, _field); \ + u32 _val = 0; \ + if ((_q)->flags & MT_QFLAG_NPU) { \ + struct airoha_npu *npu; \ + \ + rcu_read_lock(); \ + npu = rcu_dereference(q->dev->mmio.npu); \ + if (npu) \ + regmap_read(npu->regmap, \ + ((_q)->wed_regs + _offset), &_val); \ + rcu_read_unlock(); \ + } else { \ + _val = readl(&(_q)->regs->_field); \ + } \ + _val; \ +}) + +#define Q_WRITE(_q, _field, _val) do { \ + u32 _offset = offsetof(struct mt76_queue_regs, _field); \ + if ((_q)->flags & MT_QFLAG_NPU) { \ + struct airoha_npu *npu; \ + \ + rcu_read_lock(); \ + npu = rcu_dereference(q->dev->mmio.npu); \ + if (npu) \ + regmap_write(npu->regmap, \ + ((_q)->wed_regs + _offset), _val); \ + rcu_read_unlock(); \ + } else { \ + writel(_val, &(_q)->regs->_field); \ + } \ +} while (0) + #else #define Q_READ(_q, _field) readl(&(_q)->regs->_field) diff --git a/drivers/net/wireless/mediatek/mt76/mac80211.c b/drivers/net/wireless/mediatek/mt76/mac80211.c index 4be47aea2253..72ed3c825fa6 100644 --- a/drivers/net/wireless/mediatek/mt76/mac80211.c +++ b/drivers/net/wireless/mediatek/mt76/mac80211.c @@ -630,6 +630,8 @@ int mt76_create_page_pool(struct mt76_dev *dev, struct mt76_queue *q) case MT_RXQ_MAIN: case MT_RXQ_BAND1: case MT_RXQ_BAND2: + case MT_RXQ_NPU0: + case MT_RXQ_NPU1: pp_params.pool_size = 256; break; default: @@ -814,6 +816,7 @@ void mt76_free_device(struct mt76_dev *dev) destroy_workqueue(dev->wq); dev->wq = NULL; } + mt76_npu_deinit(dev); ieee80211_free_hw(dev->hw); } EXPORT_SYMBOL_GPL(mt76_free_device); @@ -1553,7 +1556,8 @@ void mt76_rx_poll_complete(struct mt76_dev *dev, enum mt76_rxq_id q, while ((skb = __skb_dequeue(&dev->rx_skb[q])) != NULL) { mt76_check_sta(dev, skb); - if (mtk_wed_device_active(&dev->mmio.wed)) + if (mtk_wed_device_active(&dev->mmio.wed) || + mt76_npu_device_active(dev)) __skb_queue_tail(&frames, skb); else mt76_rx_aggr_reorder(skb, &frames); diff --git a/drivers/net/wireless/mediatek/mt76/mt76.h b/drivers/net/wireless/mediatek/mt76/mt76.h index a66b89bc4b4a..d05e83ea1cac 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76.h +++ b/drivers/net/wireless/mediatek/mt76/mt76.h @@ -13,6 +13,7 @@ #include #include #include +#include #include #include #include @@ -34,6 +35,7 @@ #define MT_QFLAG_WED_RRO BIT(6) #define MT_QFLAG_WED_RRO_EN BIT(7) #define MT_QFLAG_EMI_EN BIT(8) +#define MT_QFLAG_NPU BIT(9) #define __MT_WED_Q(_type, _n) (MT_QFLAG_WED | \ FIELD_PREP(MT_QFLAG_WED_TYPE, _type) | \ @@ -48,6 +50,12 @@ #define MT_WED_RRO_Q_IND __MT_WED_RRO_Q(MT76_WED_RRO_Q_IND, 0) #define MT_WED_RRO_Q_RXDMAD_C __MT_WED_RRO_Q(MT76_WED_RRO_Q_RXDMAD_C, 0) +#define __MT_NPU_Q(_type, _n) (MT_QFLAG_NPU | \ + FIELD_PREP(MT_QFLAG_WED_TYPE, _type) | \ + FIELD_PREP(MT_QFLAG_WED_RING, _n)) +#define MT_NPU_Q_TX(_n) __MT_NPU_Q(MT76_WED_Q_TX, _n) +#define MT_NPU_Q_RX(_n) __MT_NPU_Q(MT76_WED_Q_RX, _n) + struct mt76_dev; struct mt76_phy; struct mt76_wcid; @@ -139,6 +147,8 @@ enum mt76_rxq_id { MT_RXQ_TXFREE_BAND2, MT_RXQ_RRO_IND, MT_RXQ_RRO_RXDMAD_C, + MT_RXQ_NPU0, + MT_RXQ_NPU1, __MT_RXQ_MAX }; @@ -707,6 +717,11 @@ struct mt76_mmio { struct mtk_wed_device wed_hif2; struct completion wed_reset; struct completion wed_reset_complete; + + struct airoha_ppe_dev __rcu *ppe_dev; + struct airoha_npu __rcu *npu; + phys_addr_t phy_addr; + int npu_type; }; struct mt76_rx_status { @@ -1617,6 +1632,109 @@ int mt76_testmode_dump(struct ieee80211_hw *hw, struct sk_buff *skb, int mt76_testmode_set_state(struct mt76_phy *phy, enum mt76_testmode_state state); int mt76_testmode_alloc_skb(struct mt76_phy *phy, u32 len); +#ifdef CONFIG_MT76_NPU +void mt76_npu_check_ppe(struct mt76_dev *dev, struct sk_buff *skb, + u32 info); +int mt76_npu_dma_add_buf(struct mt76_phy *phy, struct mt76_queue *q, + struct sk_buff *skb, struct mt76_queue_buf *buf, + void *txwi_ptr); +int mt76_npu_rx_queue_init(struct mt76_dev *dev, struct mt76_queue *q); +int mt76_npu_fill_rx_queue(struct mt76_dev *dev, struct mt76_queue *q); +void mt76_npu_queue_cleanup(struct mt76_dev *dev, struct mt76_queue *q); +void mt76_npu_disable_irqs(struct mt76_dev *dev); +int mt76_npu_init(struct mt76_dev *dev, phys_addr_t phy_addr, int type); +void mt76_npu_deinit(struct mt76_dev *dev); +void mt76_npu_queue_setup(struct mt76_dev *dev, struct mt76_queue *q); +void mt76_npu_txdesc_cleanup(struct mt76_queue *q, int index); +int mt76_npu_net_setup_tc(struct ieee80211_hw *hw, struct ieee80211_vif *vif, + struct net_device *dev, enum tc_setup_type type, + void *type_data); +#else +static inline void mt76_npu_check_ppe(struct mt76_dev *dev, + struct sk_buff *skb, u32 info) +{ +} + +static inline int mt76_npu_dma_add_buf(struct mt76_phy *phy, + struct mt76_queue *q, + struct sk_buff *skb, + struct mt76_queue_buf *buf, + void *txwi_ptr) +{ + return -EOPNOTSUPP; +} + +static inline int mt76_npu_fill_rx_queue(struct mt76_dev *dev, + struct mt76_queue *q) +{ + return 0; +} + +static inline void mt76_npu_queue_cleanup(struct mt76_dev *dev, + struct mt76_queue *q) +{ +} + +static inline void mt76_npu_disable_irqs(struct mt76_dev *dev) +{ +} + +static inline int mt76_npu_init(struct mt76_dev *dev, phys_addr_t phy_addr, + int type) +{ + return 0; +} + +static inline void mt76_npu_deinit(struct mt76_dev *dev) +{ +} + +static inline void mt76_npu_queue_setup(struct mt76_dev *dev, + struct mt76_queue *q) +{ +} + +static inline void mt76_npu_txdesc_cleanup(struct mt76_queue *q, + int index) +{ +} + +static inline int mt76_npu_net_setup_tc(struct ieee80211_hw *hw, + struct ieee80211_vif *vif, + struct net_device *dev, + enum tc_setup_type type, + void *type_data) +{ + return -EOPNOTSUPP; +} +#endif /* CONFIG_MT76_NPU */ + +static inline bool mt76_npu_device_active(struct mt76_dev *dev) +{ + return !!rcu_access_pointer(dev->mmio.npu); +} + +static inline bool mt76_ppe_device_active(struct mt76_dev *dev) +{ + return !!rcu_access_pointer(dev->mmio.ppe_dev); +} + +static inline int mt76_npu_send_msg(struct airoha_npu *npu, int ifindex, + enum airoha_npu_wlan_set_cmd cmd, + u32 val, gfp_t gfp) +{ + return airoha_npu_wlan_send_msg(npu, ifindex, cmd, &val, sizeof(val), + gfp); +} + +static inline int mt76_npu_get_msg(struct airoha_npu *npu, int ifindex, + enum airoha_npu_wlan_get_cmd cmd, + u32 *val, gfp_t gfp) +{ + return airoha_npu_wlan_get_msg(npu, ifindex, cmd, val, sizeof(*val), + gfp); +} + static inline void mt76_testmode_reset(struct mt76_phy *phy, bool disable) { #ifdef CONFIG_NL80211_TESTMODE @@ -1858,6 +1976,23 @@ static inline bool mt76_queue_is_emi(struct mt76_queue *q) return q->flags & MT_QFLAG_EMI_EN; } +static inline bool mt76_queue_is_npu(struct mt76_queue *q) +{ + return q->flags & MT_QFLAG_NPU; +} + +static inline bool mt76_queue_is_npu_tx(struct mt76_queue *q) +{ + return mt76_queue_is_npu(q) && + FIELD_GET(MT_QFLAG_WED_TYPE, q->flags) == MT76_WED_Q_TX; +} + +static inline bool mt76_queue_is_npu_rx(struct mt76_queue *q) +{ + return mt76_queue_is_npu(q) && + FIELD_GET(MT_QFLAG_WED_TYPE, q->flags) == MT76_WED_Q_RX; +} + struct mt76_txwi_cache * mt76_token_release(struct mt76_dev *dev, int token, bool *wake); int mt76_token_consume(struct mt76_dev *dev, struct mt76_txwi_cache **ptxwi); diff --git a/drivers/net/wireless/mediatek/mt76/npu.c b/drivers/net/wireless/mediatek/mt76/npu.c new file mode 100644 index 000000000000..ec36975f6dc9 --- /dev/null +++ b/drivers/net/wireless/mediatek/mt76/npu.c @@ -0,0 +1,501 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2025 AIROHA Inc + * Author: Lorenzo Bianconi + */ +#include +#include +#include + +#include "mt76.h" +#include "dma.h" +#include "mt76_connac.h" + +#define MT76_NPU_RX_BUF_SIZE (1800 + \ + SKB_DATA_ALIGN(sizeof(struct skb_shared_info))) + +int mt76_npu_fill_rx_queue(struct mt76_dev *dev, struct mt76_queue *q) +{ + int nframes = 0; + + while (q->queued < q->ndesc - 1) { + struct airoha_npu_rx_dma_desc *desc = (void *)q->desc; + struct mt76_queue_entry *e = &q->entry[q->head]; + struct page *page; + int offset; + + e->buf = mt76_get_page_pool_buf(q, &offset, q->buf_size); + if (!e->buf) + break; + + e->dma_len[0] = SKB_WITH_OVERHEAD(q->buf_size); + page = virt_to_head_page(e->buf); + e->dma_addr[0] = page_pool_get_dma_addr(page) + offset; + + memset(&desc[q->head], 0, sizeof(*desc)); + desc[q->head].addr = e->dma_addr[0]; + + q->head = (q->head + 1) % q->ndesc; + q->queued++; + nframes++; + } + + return nframes; +} + +void mt76_npu_queue_cleanup(struct mt76_dev *dev, struct mt76_queue *q) +{ + spin_lock_bh(&q->lock); + while (q->queued > 0) { + struct mt76_queue_entry *e = &q->entry[q->tail]; + + dma_sync_single_for_cpu(dev->dma_dev, e->dma_addr[0], + e->dma_len[0], + page_pool_get_dma_dir(q->page_pool)); + mt76_put_page_pool_buf(e->buf, false); + q->tail = (q->tail + 1) % q->ndesc; + q->queued--; + } + spin_unlock_bh(&q->lock); +} + +static struct sk_buff *mt76_npu_dequeue(struct mt76_dev *dev, + struct mt76_queue *q, + u32 *info) +{ + struct airoha_npu_rx_dma_desc *desc = (void *)q->desc; + int i, nframes, index = q->tail; + struct sk_buff *skb = NULL; + + nframes = FIELD_GET(NPU_RX_DMA_PKT_COUNT_MASK, desc[index].info); + nframes = max_t(int, nframes, 1); + + for (i = 0; i < nframes; i++) { + struct mt76_queue_entry *e = &q->entry[index]; + int len = FIELD_GET(NPU_RX_DMA_DESC_CUR_LEN_MASK, + desc[index].ctrl); + + if (!FIELD_GET(NPU_RX_DMA_DESC_DONE_MASK, desc[index].ctrl)) { + dev_kfree_skb(skb); + return NULL; + } + + dma_sync_single_for_cpu(dev->dma_dev, e->dma_addr[0], + e->dma_len[0], + page_pool_get_dma_dir(q->page_pool)); + + if (!skb) { + skb = napi_build_skb(e->buf, q->buf_size); + if (!skb) + return NULL; + + __skb_put(skb, len); + skb_reset_mac_header(skb); + skb_mark_for_recycle(skb); + } else { + struct skb_shared_info *shinfo = skb_shinfo(skb); + struct page *page = virt_to_head_page(e->buf); + int nr_frags = shinfo->nr_frags; + + if (nr_frags < ARRAY_SIZE(shinfo->frags)) + skb_add_rx_frag(skb, nr_frags, page, + e->buf - page_address(page), + len, q->buf_size); + } + + *info = desc[index].info; + index = (index + 1) % q->ndesc; + } + q->tail = index; + q->queued -= i; + Q_WRITE(q, dma_idx, q->tail); + + return skb; +} + +void mt76_npu_check_ppe(struct mt76_dev *dev, struct sk_buff *skb, + u32 info) +{ + struct airoha_ppe_dev *ppe_dev; + u16 reason, hash; + + if (!mt76_npu_device_active(dev)) + return; + + rcu_read_lock(); + + ppe_dev = rcu_dereference(dev->mmio.ppe_dev); + if (!ppe_dev) + goto out; + + hash = FIELD_GET(NPU_RX_DMA_FOE_ID_MASK, info); + skb_set_hash(skb, hash, PKT_HASH_TYPE_L4); + + reason = FIELD_GET(NPU_RX_DMA_CRSN_MASK, info); + if (reason == PPE_CPU_REASON_HIT_UNBIND_RATE_REACHED) { + skb_set_mac_header(skb, 0); + airoha_ppe_dev_check_skb(ppe_dev, skb, hash, true); + } +out: + rcu_read_unlock(); +} +EXPORT_SYMBOL_GPL(mt76_npu_check_ppe); + +static int mt76_npu_rx_poll(struct napi_struct *napi, int budget) +{ + struct mt76_dev *dev = mt76_priv(napi->dev); + enum mt76_rxq_id qid = napi - dev->napi; + struct airoha_npu *npu; + int done = 0; + + rcu_read_lock(); + + npu = rcu_dereference(dev->mmio.npu); + if (!npu) + goto out; + + while (done < budget) { + struct sk_buff *skb; + u32 info = 0; + + skb = mt76_npu_dequeue(dev, &dev->q_rx[qid], &info); + if (!skb) + break; + + dev->drv->rx_skb(dev, qid, skb, &info); + mt76_rx_poll_complete(dev, qid, napi); + done++; + } + + mt76_npu_fill_rx_queue(dev, &dev->q_rx[qid]); +out: + if (done < budget && napi_complete(napi)) + dev->drv->rx_poll_complete(dev, qid); + + rcu_read_unlock(); + + return done; +} + +static irqreturn_t mt76_npu_irq_handler(int irq, void *q_instance) +{ + struct mt76_queue *q = q_instance; + struct mt76_dev *dev = q->dev; + int qid = q - &dev->q_rx[0]; + int index = qid - MT_RXQ_NPU0; + struct airoha_npu *npu; + u32 status; + + rcu_read_lock(); + + npu = rcu_dereference(dev->mmio.npu); + if (!npu) + goto out; + + status = airoha_npu_wlan_get_irq_status(npu, index); + airoha_npu_wlan_set_irq_status(npu, status); + + airoha_npu_wlan_disable_irq(npu, index); + napi_schedule(&dev->napi[qid]); +out: + rcu_read_unlock(); + + return IRQ_HANDLED; +} + +int mt76_npu_dma_add_buf(struct mt76_phy *phy, struct mt76_queue *q, + struct sk_buff *skb, struct mt76_queue_buf *buf, + void *txwi_ptr) +{ + u16 txwi_len = min_t(u16, phy->dev->drv->txwi_size, NPU_TXWI_LEN); + struct airoha_npu_tx_dma_desc *desc = (void *)q->desc; + int ret; + + /* TODO: Take into account unlinear skbs */ + memcpy(desc[q->head].txwi, txwi_ptr, txwi_len); + desc[q->head].addr = buf->addr; + desc[q->head].ctrl = FIELD_PREP(NPU_TX_DMA_DESC_VEND_LEN_MASK, txwi_len) | + FIELD_PREP(NPU_TX_DMA_DESC_LEN_MASK, skb->len) | + NPU_TX_DMA_DESC_DONE_MASK; + + ret = q->head; + q->entry[q->head].skip_buf0 = true; + q->entry[q->head].skip_buf1 = true; + q->entry[q->head].txwi = NULL; + q->entry[q->head].skb = NULL; + q->entry[q->head].wcid = 0xffff; + + q->head = (q->head + 1) % q->ndesc; + q->queued++; + + return ret; +} + +void mt76_npu_txdesc_cleanup(struct mt76_queue *q, int index) +{ + struct airoha_npu_tx_dma_desc *desc = (void *)q->desc; + + if (!mt76_queue_is_npu_tx(q)) + return; + + desc[index].ctrl &= ~NPU_TX_DMA_DESC_DONE_MASK; +} + +void mt76_npu_queue_setup(struct mt76_dev *dev, struct mt76_queue *q) +{ + int qid = FIELD_GET(MT_QFLAG_WED_RING, q->flags); + bool xmit = mt76_queue_is_npu_tx(q); + struct airoha_npu *npu; + + if (!mt76_queue_is_npu(q)) + return; + + npu = rcu_dereference_protected(dev->mmio.npu, &dev->mutex); + if (npu) + q->wed_regs = airoha_npu_wlan_get_queue_addr(npu, qid, xmit); +} + +int mt76_npu_rx_queue_init(struct mt76_dev *dev, struct mt76_queue *q) +{ + int err, irq, qid = q - &dev->q_rx[0]; + int size, index = qid - MT_RXQ_NPU0; + struct airoha_npu *npu; + const char *name; + + mutex_lock(&dev->mutex); + + npu = rcu_dereference_protected(dev->mmio.npu, &dev->mutex); + irq = npu && index < ARRAY_SIZE(npu->irqs) ? npu->irqs[index] + : -EINVAL; + if (irq < 0) { + err = irq; + goto out; + } + + q->flags = MT_NPU_Q_RX(index); + size = qid == MT_RXQ_NPU1 ? NPU_RX1_DESC_NUM : NPU_RX0_DESC_NUM; + err = dev->queue_ops->alloc(dev, q, 0, size, + MT76_NPU_RX_BUF_SIZE, 0); + if (err) + goto out; + + name = devm_kasprintf(dev->dev, GFP_KERNEL, "mt76-npu.%d", index); + if (!name) { + err = -ENOMEM; + goto out; + } + + err = devm_request_irq(dev->dev, irq, mt76_npu_irq_handler, + IRQF_SHARED, name, q); + if (err) + goto out; + + netif_napi_add(dev->napi_dev, &dev->napi[qid], mt76_npu_rx_poll); + mt76_npu_fill_rx_queue(dev, q); + napi_enable(&dev->napi[qid]); +out: + mutex_unlock(&dev->mutex); + + return err; +} +EXPORT_SYMBOL_GPL(mt76_npu_rx_queue_init); + +static int mt76_npu_setup_tc_block_cb(enum tc_setup_type type, + void *type_data, void *cb_priv) +{ + struct mt76_phy *phy = cb_priv; + struct mt76_dev *dev = phy->dev; + struct airoha_ppe_dev *ppe_dev; + int err = -EOPNOTSUPP; + + if (type != TC_SETUP_CLSFLOWER) + return -EOPNOTSUPP; + + mutex_lock(&dev->mutex); + + ppe_dev = rcu_dereference_protected(dev->mmio.ppe_dev, &dev->mutex); + if (ppe_dev) + err = airoha_ppe_dev_setup_tc_block_cb(ppe_dev, type_data); + + mutex_unlock(&dev->mutex); + + return err; +} + +static int mt76_npu_setup_tc_block(struct mt76_phy *phy, + struct net_device *dev, + struct flow_block_offload *f) +{ + flow_setup_cb_t *cb = mt76_npu_setup_tc_block_cb; + static LIST_HEAD(block_cb_list); + struct flow_block_cb *block_cb; + + if (f->binder_type != FLOW_BLOCK_BINDER_TYPE_CLSACT_INGRESS) + return -EOPNOTSUPP; + + if (!tc_can_offload(dev)) + return -EOPNOTSUPP; + + f->driver_block_list = &block_cb_list; + switch (f->command) { + case FLOW_BLOCK_BIND: + block_cb = flow_block_cb_lookup(f->block, cb, dev); + if (block_cb) { + flow_block_cb_incref(block_cb); + return 0; + } + + block_cb = flow_block_cb_alloc(cb, dev, phy, NULL); + if (IS_ERR(block_cb)) + return PTR_ERR(block_cb); + + flow_block_cb_incref(block_cb); + flow_block_cb_add(block_cb, f); + list_add_tail(&block_cb->driver_list, &block_cb_list); + return 0; + case FLOW_BLOCK_UNBIND: + block_cb = flow_block_cb_lookup(f->block, cb, dev); + if (!block_cb) + return -ENOENT; + + if (!flow_block_cb_decref(block_cb)) { + flow_block_cb_remove(block_cb, f); + list_del(&block_cb->driver_list); + } + return 0; + default: + return -EOPNOTSUPP; + } +} + +int mt76_npu_net_setup_tc(struct ieee80211_hw *hw, struct ieee80211_vif *vif, + struct net_device *dev, enum tc_setup_type type, + void *type_data) +{ + struct mt76_phy *phy = hw->priv; + + if (!tc_can_offload(dev)) + return -EOPNOTSUPP; + + if (!mt76_npu_device_active(phy->dev)) + return -EOPNOTSUPP; + + switch (type) { + case TC_SETUP_BLOCK: + case TC_SETUP_FT: + return mt76_npu_setup_tc_block(phy, dev, type_data); + default: + return -EOPNOTSUPP; + } +} +EXPORT_SYMBOL_GPL(mt76_npu_net_setup_tc); + +void mt76_npu_disable_irqs(struct mt76_dev *dev) +{ + struct airoha_npu *npu; + int i; + + rcu_read_lock(); + + npu = rcu_dereference(dev->mmio.npu); + if (!npu) + goto unlock; + + for (i = MT_RXQ_NPU0; i <= MT_RXQ_NPU1; i++) { + int qid = i - MT_RXQ_NPU0; + u32 status; + + status = airoha_npu_wlan_get_irq_status(npu, qid); + airoha_npu_wlan_set_irq_status(npu, status); + airoha_npu_wlan_disable_irq(npu, qid); + } +unlock: + rcu_read_unlock(); +} +EXPORT_SYMBOL_GPL(mt76_npu_disable_irqs); + +int mt76_npu_init(struct mt76_dev *dev, phys_addr_t phy_addr, int type) +{ + struct airoha_ppe_dev *ppe_dev; + struct airoha_npu *npu; + int err = 0; + + /* NPU offloading is only supported by MT7992 */ + if (!is_mt7992(dev)) + return 0; + + mutex_lock(&dev->mutex); + + npu = airoha_npu_get(dev->dev); + if (IS_ERR(npu)) { + request_module("airoha-npu"); + npu = airoha_npu_get(dev->dev); + } + + if (IS_ERR(npu)) { + err = PTR_ERR(npu); + goto error_unlock; + } + + ppe_dev = airoha_ppe_get_dev(dev->dev); + if (IS_ERR(ppe_dev)) { + request_module("airoha-eth"); + ppe_dev = airoha_ppe_get_dev(dev->dev); + } + + if (IS_ERR(ppe_dev)) { + err = PTR_ERR(ppe_dev); + goto error_npu_put; + } + + err = airoha_npu_wlan_init_reserved_memory(npu); + if (err) + goto error_ppe_put; + + dev->dma_dev = npu->dev; + dev->mmio.phy_addr = phy_addr; + dev->mmio.npu_type = type; + /* NPU offloading requires HW-RRO for RX packet reordering. */ + dev->hwrro_mode = MT76_HWRRO_V3_1; + + rcu_assign_pointer(dev->mmio.npu, npu); + rcu_assign_pointer(dev->mmio.ppe_dev, ppe_dev); + synchronize_rcu(); + + mutex_unlock(&dev->mutex); + + return 0; + +error_ppe_put: + airoha_ppe_put_dev(ppe_dev); +error_npu_put: + airoha_npu_put(npu); +error_unlock: + mutex_unlock(&dev->mutex); + + return err; +} +EXPORT_SYMBOL_GPL(mt76_npu_init); + +void mt76_npu_deinit(struct mt76_dev *dev) +{ + struct airoha_ppe_dev *ppe_dev; + struct airoha_npu *npu; + + mutex_lock(&dev->mutex); + + npu = rcu_replace_pointer(dev->mmio.npu, NULL, + lockdep_is_held(&dev->mutex)); + if (npu) + airoha_npu_put(npu); + + ppe_dev = rcu_replace_pointer(dev->mmio.ppe_dev, NULL, + lockdep_is_held(&dev->mutex)); + if (ppe_dev) + airoha_ppe_put_dev(ppe_dev); + + mutex_unlock(&dev->mutex); + + mt76_npu_queue_cleanup(dev, &dev->q_rx[MT_RXQ_NPU0]); + mt76_npu_queue_cleanup(dev, &dev->q_rx[MT_RXQ_NPU1]); +} diff --git a/include/linux/soc/airoha/airoha_offload.h b/include/linux/soc/airoha/airoha_offload.h index 6f66eb339b3f..4d23cbb7d407 100644 --- a/include/linux/soc/airoha/airoha_offload.h +++ b/include/linux/soc/airoha/airoha_offload.h @@ -6,6 +6,7 @@ #ifndef AIROHA_OFFLOAD_H #define AIROHA_OFFLOAD_H +#include #include #include From 377aa17d2aedc98d1c3bbe7e49b22a0c76308b0e Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Fri, 17 Oct 2025 10:50:33 +0200 Subject: [PATCH 1031/1381] wifi: mt76: mt7996: Add NPU offload support to MT7996 driver Introduce Airoha NPU support to MT7996 driver. NPU is used to enable traffic forward offloading between the MT76 NIC and the Airoha ethernet one available on the Airoha EN7581 SoC using Netfilter Flowtable APIs. Signed-off-by: Lorenzo Bianconi Link: https://patch.msgid.link/20251017-mt76-npu-devel-v2-5-ddaa90901723@kernel.org Signed-off-by: Felix Fietkau --- .../net/wireless/mediatek/mt76/mt7996/Kconfig | 7 + .../wireless/mediatek/mt76/mt7996/Makefile | 1 + .../net/wireless/mediatek/mt76/mt7996/dma.c | 16 +- .../net/wireless/mediatek/mt76/mt7996/init.c | 13 +- .../net/wireless/mediatek/mt76/mt7996/mac.c | 4 + .../net/wireless/mediatek/mt76/mt7996/main.c | 26 +- .../net/wireless/mediatek/mt76/mt7996/mmio.c | 13 +- .../wireless/mediatek/mt76/mt7996/mt7996.h | 21 ++ .../net/wireless/mediatek/mt76/mt7996/npu.c | 352 ++++++++++++++++++ .../net/wireless/mediatek/mt76/mt7996/pci.c | 3 + 10 files changed, 438 insertions(+), 18 deletions(-) create mode 100644 drivers/net/wireless/mediatek/mt76/mt7996/npu.c diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/Kconfig b/drivers/net/wireless/mediatek/mt76/mt7996/Kconfig index 196e1cbab813..5503d03bf62c 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/Kconfig +++ b/drivers/net/wireless/mediatek/mt76/mt7996/Kconfig @@ -12,3 +12,10 @@ config MT7996E and 2.4GHz IEEE 802.11be 4x4:4SS 4096-QAM, 320MHz channels. To compile this driver as a module, choose M here. + +config MT7996_NPU + bool "MT7996 (PCIe) NPU support" + depends on MT7996E + depends on NET_AIROHA_NPU=y || MT7996E=NET_AIROHA_NPU + select MT76_NPU + default n diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/Makefile b/drivers/net/wireless/mediatek/mt76/mt7996/Makefile index 19c0bc02b584..69d2d4bb9e69 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/Makefile +++ b/drivers/net/wireless/mediatek/mt76/mt7996/Makefile @@ -5,4 +5,5 @@ obj-$(CONFIG_MT7996E) += mt7996e.o mt7996e-y := pci.o init.o dma.o eeprom.o main.o mcu.o mac.o \ debugfs.o mmio.o +mt7996e-$(CONFIG_MT7996_NPU) += npu.o mt7996e-$(CONFIG_DEV_COREDUMP) += coredump.o diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/dma.c b/drivers/net/wireless/mediatek/mt76/mt7996/dma.c index 20aa67eff76f..7ac4defca29d 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/dma.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/dma.c @@ -23,6 +23,9 @@ int mt7996_init_tx_queues(struct mt7996_phy *phy, int idx, int n_desc, flags = MT_WED_Q_TX(idx); } + if (mt76_npu_device_active(&dev->mt76)) + flags = MT_NPU_Q_TX(phy->mt76->band_idx); + return mt76_connac_init_tx_queues(phy->mt76, idx, n_desc, ring_base, wed, flags); } @@ -344,7 +347,7 @@ void mt7996_dma_start(struct mt7996_dev *dev, bool reset, bool wed_reset) mtk_wed_device_start(wed, wed_irq_mask); } - if (!mt7996_has_wa(dev)) + if (!mt7996_has_wa(dev) || mt76_npu_device_active(&dev->mt76)) irq_mask &= ~(MT_INT_RX(MT_RXQ_MAIN_WA) | MT_INT_RX(MT_RXQ_BAND1_WA)); irq_mask = reset ? MT_INT_MCU_CMD : irq_mask; @@ -502,7 +505,7 @@ int mt7996_dma_rro_init(struct mt7996_dev *dev) mdev->q_rx[MT_RXQ_RRO_RXDMAD_C].flags = MT_WED_RRO_Q_RXDMAD_C; if (mtk_wed_device_active(&mdev->mmio.wed)) mdev->q_rx[MT_RXQ_RRO_RXDMAD_C].wed = &mdev->mmio.wed; - else + else if (!mt76_npu_device_active(&dev->mt76)) mdev->q_rx[MT_RXQ_RRO_RXDMAD_C].flags |= MT_QFLAG_EMI_EN; ret = mt76_queue_alloc(dev, &mdev->q_rx[MT_RXQ_RRO_RXDMAD_C], MT_RXQ_ID(MT_RXQ_RRO_RXDMAD_C), @@ -610,7 +613,9 @@ int mt7996_dma_rro_init(struct mt7996_dev *dev) mt76_queue_rx_init(dev, MT_RXQ_MSDU_PAGE_BAND0, mt76_dma_rx_poll); } - mt7996_irq_enable(dev, MT_INT_RRO_RX_DONE); + + if (!mt76_npu_device_active(&dev->mt76)) + mt7996_irq_enable(dev, MT_INT_RRO_RX_DONE); } return 0; @@ -884,6 +889,10 @@ int mt7996_dma_init(struct mt7996_dev *dev) if (ret < 0) return ret; + ret = mt7996_npu_rx_queues_init(dev); + if (ret) + return ret; + netif_napi_add_tx(dev->mt76.tx_napi_dev, &dev->mt76.tx_napi, mt7996_poll_tx); napi_enable(&dev->mt76.tx_napi); @@ -941,6 +950,7 @@ void mt7996_dma_reset(struct mt7996_dev *dev, bool force) if (mtk_wed_device_active(&dev->mt76.mmio.wed)) mtk_wed_device_dma_reset(&dev->mt76.mmio.wed); + mt76_npu_disable_irqs(&dev->mt76); mt7996_dma_disable(dev, force); mt76_wed_dma_reset(&dev->mt76); diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/init.c b/drivers/net/wireless/mediatek/mt76/mt7996/init.c index 5f2a2da5c326..bd7b720c64c5 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/init.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/init.c @@ -475,7 +475,7 @@ mt7996_init_wiphy(struct ieee80211_hw *hw, struct mtk_wed_device *wed) hw->max_tx_aggregation_subframes = 512; hw->netdev_features = NETIF_F_RXCSUM; - if (mtk_wed_device_active(wed)) + if (mtk_wed_device_active(wed) || mt76_npu_device_active(mdev)) hw->netdev_features |= NETIF_F_HW_TC; hw->radiotap_timestamp.units_pos = @@ -830,7 +830,8 @@ void mt7996_rro_hw_init(struct mt7996_dev *dev) MT_RRO_3_0_EMU_CONF_EN_MASK); mt76_set(dev, MT_RRO_3_1_GLOBAL_CONFIG, MT_RRO_3_1_GLOBAL_CONFIG_RXDMAD_SEL); - if (!mtk_wed_device_active(&dev->mt76.mmio.wed)) { + if (!mtk_wed_device_active(&dev->mt76.mmio.wed) && + !mt76_npu_device_active(&dev->mt76)) { mt76_set(dev, MT_RRO_3_1_GLOBAL_CONFIG, MT_RRO_3_1_GLOBAL_CONFIG_RX_DIDX_WR_EN | MT_RRO_3_1_GLOBAL_CONFIG_RX_CIDX_RD_EN); @@ -1067,6 +1068,9 @@ static void mt7996_wed_rro_work(struct work_struct *work) list); list_del_init(&e->list); + if (mt76_npu_device_active(&dev->mt76)) + goto reset_session; + for (i = 0; i < MT7996_RRO_WINDOW_MAX_LEN; i++) { void *ptr = dev->wed_rro.session.ptr; struct mt7996_wed_rro_addr *elem; @@ -1087,6 +1091,7 @@ static void mt7996_wed_rro_work(struct work_struct *work) elem = ptr + elem_id * sizeof(*elem); elem->data |= cpu_to_le32(val); } +reset_session: mt7996_mcu_wed_rro_reset_sessions(dev, e->id); out: kfree(e); @@ -1674,6 +1679,10 @@ int mt7996_register_device(struct mt7996_dev *dev) if (ret) return ret; + ret = mt7996_npu_hw_init(dev); + if (ret) + return ret; + ret = mt76_register_device(&dev->mt76, true, mt76_rates, ARRAY_SIZE(mt76_rates)); if (ret) diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c index ace23fd67c8d..6ec17d364e83 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c @@ -718,6 +718,7 @@ mt7996_mac_fill_rx(struct mt7996_dev *dev, enum mt76_rxq_id q, status->flag |= RX_FLAG_8023; mt7996_wed_check_ppe(dev, &dev->mt76.q_rx[q], msta, skb, *info); + mt76_npu_check_ppe(&dev->mt76, skb, *info); } if (rxv && !(status->flag & RX_FLAG_8023)) { @@ -2535,6 +2536,8 @@ void mt7996_mac_reset_work(struct work_struct *work) mutex_lock(&dev->mt76.mutex); + mt7996_npu_hw_stop(dev); + mt76_wr(dev, MT_MCU_INT_EVENT, MT_MCU_INT_EVENT_DMA_STOPPED); if (mt7996_wait_reset_state(dev, MT_MCU_CMD_RESET_DONE)) { @@ -2601,6 +2604,7 @@ void mt7996_mac_reset_work(struct work_struct *work) mutex_unlock(&dev->mt76.mutex); + mt7996_npu_hw_init(dev); mt7996_update_beacons(dev); mt7996_for_each_phy(dev, phy) diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/main.c b/drivers/net/wireless/mediatek/mt76/mt7996/main.c index 848593043781..08c4d9c8791b 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/main.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/main.c @@ -2157,7 +2157,6 @@ mt7996_set_radar_background(struct ieee80211_hw *hw, return ret; } -#ifdef CONFIG_NET_MEDIATEK_SOC_WED static int mt7996_net_fill_forward_path(struct ieee80211_hw *hw, struct ieee80211_vif *vif, @@ -2165,15 +2164,14 @@ mt7996_net_fill_forward_path(struct ieee80211_hw *hw, struct net_device_path_ctx *ctx, struct net_device_path *path) { - struct mt7996_vif *mvif = (struct mt7996_vif *)vif->drv_priv; struct mt7996_sta *msta = (struct mt7996_sta *)sta->drv_priv; struct mt7996_dev *dev = mt7996_hw_dev(hw); struct mtk_wed_device *wed = &dev->mt76.mmio.wed; struct mt7996_sta_link *msta_link; - struct mt76_vif_link *mlink; + struct mt7996_vif_link *link; - mlink = rcu_dereference(mvif->mt76.link[msta->deflink_id]); - if (!mlink) + link = mt7996_vif_link(dev, vif, msta->deflink_id); + if (!link) return -EIO; msta_link = rcu_dereference(msta->link[msta->deflink_id]); @@ -2188,13 +2186,19 @@ mt7996_net_fill_forward_path(struct ieee80211_hw *hw, (is_mt7992(&dev->mt76) && msta_link->wcid.phy_idx == MT_BAND1))) wed = &dev->mt76.mmio.wed_hif2; - if (!mtk_wed_device_active(wed)) + if (!mtk_wed_device_active(wed) && + !mt76_npu_device_active(&dev->mt76)) return -ENODEV; path->type = DEV_PATH_MTK_WDMA; path->dev = ctx->dev; - path->mtk_wdma.wdma_idx = wed->wdma_idx; - path->mtk_wdma.bss = mlink->idx; +#ifdef CONFIG_NET_MEDIATEK_SOC_WED + if (mtk_wed_device_active(wed)) + path->mtk_wdma.wdma_idx = wed->wdma_idx; + else +#endif + path->mtk_wdma.wdma_idx = link->mt76.band_idx; + path->mtk_wdma.bss = link->mt76.idx; path->mtk_wdma.queue = 0; path->mtk_wdma.wcid = msta_link->wcid.idx; @@ -2208,8 +2212,6 @@ mt7996_net_fill_forward_path(struct ieee80211_hw *hw, return 0; } -#endif - static int mt7996_change_vif_links(struct ieee80211_hw *hw, struct ieee80211_vif *vif, u16 old_links, u16 new_links, @@ -2283,9 +2285,11 @@ const struct ieee80211_ops mt7996_ops = { .sta_add_debugfs = mt7996_sta_add_debugfs, #endif .set_radar_background = mt7996_set_radar_background, -#ifdef CONFIG_NET_MEDIATEK_SOC_WED .net_fill_forward_path = mt7996_net_fill_forward_path, +#ifdef CONFIG_NET_MEDIATEK_SOC_WED .net_setup_tc = mt76_wed_net_setup_tc, +#elif defined(CONFIG_MT7996_NPU) + .net_setup_tc = mt76_npu_net_setup_tc, #endif .change_vif_links = mt7996_change_vif_links, .change_sta_links = mt7996_mac_sta_change_links, diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mmio.c b/drivers/net/wireless/mediatek/mt76/mt7996/mmio.c index b795a72c2c82..d9780bb425a7 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/mmio.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/mmio.c @@ -707,9 +707,18 @@ void mt7996_dual_hif_set_irq_mask(struct mt7996_dev *dev, bool write_reg, static void mt7996_rx_poll_complete(struct mt76_dev *mdev, enum mt76_rxq_id q) { - struct mt7996_dev *dev = container_of(mdev, struct mt7996_dev, mt76); + if (q == MT_RXQ_NPU0 || q == MT_RXQ_NPU1) { + struct airoha_npu *npu; - mt7996_irq_enable(dev, MT_INT_RX(q)); + npu = rcu_dereference(mdev->mmio.npu); + if (npu) + airoha_npu_wlan_enable_irq(npu, q - MT_RXQ_NPU0); + } else { + struct mt7996_dev *dev = container_of(mdev, struct mt7996_dev, + mt76); + + mt7996_irq_enable(dev, MT_INT_RX(q)); + } } /* TODO: support 2/4/6/8 MSI-X vectors */ diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h b/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h index d52264b23163..87730d26669d 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h +++ b/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h @@ -874,4 +874,25 @@ int mt7996_mtk_init_debugfs(struct mt7996_phy *phy, struct dentry *dir); int mt7996_dma_rro_init(struct mt7996_dev *dev); +#ifdef CONFIG_MT7996_NPU +int mt7996_npu_hw_init(struct mt7996_dev *dev); +int mt7996_npu_hw_stop(struct mt7996_dev *dev); +int mt7996_npu_rx_queues_init(struct mt7996_dev *dev); +#else +static inline int mt7996_npu_hw_init(struct mt7996_dev *dev) +{ + return 0; +} + +static inline int mt7996_npu_hw_stop(struct mt7996_dev *dev) +{ + return 0; +} + +static inline int mt7996_npu_rx_queues_init(struct mt7996_dev *dev) +{ + return 0; +} +#endif /* CONFIG_MT7996_NPU */ + #endif diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/npu.c b/drivers/net/wireless/mediatek/mt76/mt7996/npu.c new file mode 100644 index 000000000000..29bb735da4cb --- /dev/null +++ b/drivers/net/wireless/mediatek/mt76/mt7996/npu.c @@ -0,0 +1,352 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2025 AIROHA Inc + * Author: Lorenzo Bianconi + */ +#include +#include + +#include "mt7996.h" + +static int mt7996_npu_offload_init(struct mt7996_dev *dev, + struct airoha_npu *npu) +{ + phys_addr_t phy_addr = dev->mt76.mmio.phy_addr; + u32 val, hif1_ofs = 0, dma_addr; + int i, err; + + err = mt76_npu_get_msg(npu, 0, WLAN_FUNC_GET_WAIT_NPU_VERSION, + &val, GFP_KERNEL); + if (err) { + dev_warn(dev->mt76.dev, "failed getting NPU fw version\n"); + return err; + } + + dev_info(dev->mt76.dev, "NPU version: %0d.%d\n", + (val >> 16) & 0xffff, val & 0xffff); + + err = mt76_npu_send_msg(npu, 0, WLAN_FUNC_SET_WAIT_PCIE_PORT_TYPE, + dev->mt76.mmio.npu_type, GFP_KERNEL); + if (err) { + dev_warn(dev->mt76.dev, + "failed setting NPU wlan PCIe port type\n"); + return err; + } + + if (dev->hif2) + hif1_ofs = MT_WFDMA0_PCIE1(0) - MT_WFDMA0(0); + + for (i = MT_BAND0; i < MT_BAND2; i++) { + dma_addr = phy_addr; + if (i) + dma_addr += MT_RXQ_RING_BASE(MT_RXQ_RRO_BAND1) + 0x90 + + hif1_ofs; + else + dma_addr += MT_RXQ_RING_BASE(MT_RXQ_RRO_BAND0) + 0x80; + + err = mt76_npu_send_msg(npu, i, WLAN_FUNC_SET_WAIT_PCIE_ADDR, + dma_addr, GFP_KERNEL); + if (err) { + dev_warn(dev->mt76.dev, + "failed setting NPU wlan PCIe desc addr\n"); + return err; + } + + err = mt76_npu_send_msg(npu, i, WLAN_FUNC_SET_WAIT_DESC, + MT7996_RX_RING_SIZE, GFP_KERNEL); + if (err) { + dev_warn(dev->mt76.dev, + "failed setting NPU wlan PCIe desc size\n"); + return err; + } + + dma_addr = phy_addr; + if (i) + dma_addr += MT_TXQ_RING_BASE(0) + 0x150 + hif1_ofs; + else + dma_addr += MT_TXQ_RING_BASE(0) + 0x120; + + err = mt76_npu_send_msg(npu, i, + WLAN_FUNC_SET_WAIT_TX_RING_PCIE_ADDR, + dma_addr, GFP_KERNEL); + if (err) { + dev_warn(dev->mt76.dev, + "failed setting NPU wlan tx desc addr\n"); + return err; + } + } + + err = mt76_npu_send_msg(npu, 9, WLAN_FUNC_SET_WAIT_PCIE_ADDR, + phy_addr + MT_RXQ_RRO_AP_RING_BASE, + GFP_KERNEL); + if (err) { + dev_warn(dev->mt76.dev, + "failed setting NPU wlan rxdmad_c addr\n"); + return err; + } + + err = mt76_npu_send_msg(npu, 9, WLAN_FUNC_SET_WAIT_DESC, + MT7996_RX_RING_SIZE, GFP_KERNEL); + if (err) { + dev_warn(dev->mt76.dev, + "failed setting NPU wlan rxdmad_c desc size\n"); + return err; + } + + err = mt76_npu_send_msg(npu, 2, WLAN_FUNC_SET_WAIT_TX_RING_PCIE_ADDR, + phy_addr + MT_RRO_ACK_SN_CTRL, GFP_KERNEL); + if (err) { + dev_warn(dev->mt76.dev, + "failed setting NPU wlan rro_ack_sn desc addr\n"); + return err; + } + + err = mt76_npu_send_msg(npu, 0, WLAN_FUNC_SET_WAIT_TOKEN_ID_SIZE, + MT7996_HW_TOKEN_SIZE, GFP_KERNEL); + if (err) + return err; + + dev->mt76.token_start = MT7996_HW_TOKEN_SIZE; + + return 0; +} + +static int mt7996_npu_rxd_init(struct mt7996_dev *dev, struct airoha_npu *npu) +{ + u32 val; + int err; + + err = mt76_npu_get_msg(npu, 0, WLAN_FUNC_GET_WAIT_RXDESC_BASE, + &val, GFP_KERNEL); + if (err) { + dev_warn(dev->mt76.dev, + "failed retriving NPU wlan rx ring0 addr\n"); + return err; + } + writel(val, &dev->mt76.q_rx[MT_RXQ_RRO_BAND0].regs->desc_base); + + err = mt76_npu_get_msg(npu, 1, WLAN_FUNC_GET_WAIT_RXDESC_BASE, + &val, GFP_KERNEL); + if (err) { + dev_warn(dev->mt76.dev, + "failed retriving NPU wlan rx ring1 addr\n"); + return err; + } + writel(val, &dev->mt76.q_rx[MT_RXQ_RRO_BAND1].regs->desc_base); + + err = mt76_npu_get_msg(npu, 9, WLAN_FUNC_GET_WAIT_RXDESC_BASE, + &val, GFP_KERNEL); + if (err) { + dev_warn(dev->mt76.dev, + "failed retriving NPU wlan rxdmad_c ring addr\n"); + return err; + } + writel(val, &dev->mt76.q_rx[MT_RXQ_RRO_RXDMAD_C].regs->desc_base); + + return 0; +} + +static int mt7996_npu_txd_init(struct mt7996_dev *dev, struct airoha_npu *npu) +{ + int i, err; + + for (i = MT_BAND0; i < MT_BAND2; i++) { + dma_addr_t dma_addr; + u32 val; + + err = mt76_npu_get_msg(npu, i + 5, + WLAN_FUNC_GET_WAIT_RXDESC_BASE, + &val, GFP_KERNEL); + if (err) { + dev_warn(dev->mt76.dev, + "failed retriving NPU wlan tx ring addr\n"); + return err; + } + writel(val, &dev->mt76.phys[i]->q_tx[0]->regs->desc_base); + + if (!dmam_alloc_coherent(dev->mt76.dma_dev, + 256 * MT7996_TX_RING_SIZE, + &dma_addr, GFP_KERNEL)) + return -ENOMEM; + + err = mt76_npu_send_msg(npu, i, + WLAN_FUNC_SET_WAIT_TX_BUF_SPACE_HW_BASE, + dma_addr, GFP_KERNEL); + if (err) { + dev_warn(dev->mt76.dev, + "failed setting NPU wlan queue buf addr\n"); + return err; + } + + if (!dmam_alloc_coherent(dev->mt76.dma_dev, + 256 * MT7996_TX_RING_SIZE, + &dma_addr, GFP_KERNEL)) + return -ENOMEM; + + err = mt76_npu_send_msg(npu, i + 5, + WLAN_FUNC_SET_WAIT_TX_BUF_SPACE_HW_BASE, + dma_addr, GFP_KERNEL); + if (err) { + dev_warn(dev->mt76.dev, + "failed setting NPU wlan tx buf addr\n"); + return err; + } + + if (!dmam_alloc_coherent(dev->mt76.dma_dev, 256 * 1024, + &dma_addr, GFP_KERNEL)) + return -ENOMEM; + + err = mt76_npu_send_msg(npu, i + 10, + WLAN_FUNC_SET_WAIT_TX_BUF_SPACE_HW_BASE, + dma_addr, GFP_KERNEL); + if (err) { + dev_warn(dev->mt76.dev, + "failed setting NPU wlan tx buf base\n"); + return err; + } + } + + return 0; +} + +static int mt7996_npu_rx_event_init(struct mt7996_dev *dev, + struct airoha_npu *npu) +{ + struct mt76_queue *q = &dev->mt76.q_rx[MT_RXQ_MAIN_WA]; + phys_addr_t phy_addr = dev->mt76.mmio.phy_addr; + int err; + + err = mt76_npu_send_msg(npu, 0, + WLAN_FUNC_SET_WAIT_RX_RING_FOR_TXDONE_HW_BASE, + q->desc_dma, GFP_KERNEL); + if (err) { + dev_warn(dev->mt76.dev, + "failed setting NPU wlan tx-done ring\n"); + return err; + } + + err = mt76_npu_send_msg(npu, 10, WLAN_FUNC_SET_WAIT_DESC, + MT7996_RX_MCU_RING_SIZE, GFP_KERNEL); + if (err) { + dev_warn(dev->mt76.dev, + "failed setting NPU wlan descriptors\n"); + return err; + } + + phy_addr += MT_RXQ_RING_BASE(MT_RXQ_MAIN_WA) + 0x20; + err = mt76_npu_send_msg(npu, 10, WLAN_FUNC_SET_WAIT_PCIE_ADDR, + phy_addr, GFP_KERNEL); + if (err) + dev_warn(dev->mt76.dev, + "failed setting NPU wlan rx pcie address\n"); + return err; +} + +static int mt7996_npu_tx_done_init(struct mt7996_dev *dev, + struct airoha_npu *npu) +{ + int err; + + err = mt76_npu_send_msg(npu, 2, WLAN_FUNC_SET_WAIT_INODE_TXRX_REG_ADDR, + 0, GFP_KERNEL); + if (err) { + dev_warn(dev->mt76.dev, "failed setting NPU wlan txrx addr2\n"); + return err; + } + + err = mt76_npu_send_msg(npu, 7, WLAN_FUNC_SET_WAIT_INODE_TXRX_REG_ADDR, + 0, GFP_KERNEL); + if (err) + dev_warn(dev->mt76.dev, "failed setting NPU wlan txrx addr7\n"); + + return err; +} + +int mt7996_npu_rx_queues_init(struct mt7996_dev *dev) +{ + int err; + + if (!mt76_npu_device_active(&dev->mt76)) + return 0; + + err = mt76_npu_rx_queue_init(&dev->mt76, + &dev->mt76.q_rx[MT_RXQ_NPU0]); + if (err) + return err; + + return mt76_npu_rx_queue_init(&dev->mt76, + &dev->mt76.q_rx[MT_RXQ_NPU1]); +} + +int mt7996_npu_hw_init(struct mt7996_dev *dev) +{ + struct airoha_npu *npu; + int i, err = 0; + + mutex_lock(&dev->mt76.mutex); + + npu = rcu_dereference_protected(dev->mt76.mmio.npu, &dev->mt76.mutex); + if (!npu) + goto unlock; + + err = mt7996_npu_offload_init(dev, npu); + if (err) + goto unlock; + + err = mt7996_npu_rxd_init(dev, npu); + if (err) + goto unlock; + + err = mt7996_npu_txd_init(dev, npu); + if (err) + goto unlock; + + err = mt7996_npu_rx_event_init(dev, npu); + if (err) + goto unlock; + + err = mt7996_npu_tx_done_init(dev, npu); + if (err) + goto unlock; + + for (i = MT_RXQ_NPU0; i <= MT_RXQ_NPU1; i++) + airoha_npu_wlan_enable_irq(npu, i - MT_RXQ_NPU0); +unlock: + mutex_unlock(&dev->mt76.mutex); + + return err; +} + +int mt7996_npu_hw_stop(struct mt7996_dev *dev) +{ + struct airoha_npu *npu; + int i, err; + u32 info; + + npu = rcu_dereference_protected(dev->mt76.mmio.npu, &dev->mt76.mutex); + if (!npu) + return 0; + + err = mt76_npu_send_msg(npu, 4, WLAN_FUNC_SET_WAIT_INODE_TXRX_REG_ADDR, + 0, GFP_KERNEL); + if (err) + return err; + + for (i = 0; i < 10; i++) { + err = mt76_npu_get_msg(npu, 3, WLAN_FUNC_GET_WAIT_NPU_INFO, + &info, GFP_KERNEL); + if (err) + continue; + + if (info) { + err = -ETIMEDOUT; + continue; + } + } + + if (!err) + err = mt76_npu_send_msg(npu, 6, + WLAN_FUNC_SET_WAIT_INODE_TXRX_REG_ADDR, + 0, GFP_KERNEL); + return err; +} diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/pci.c b/drivers/net/wireless/mediatek/mt76/mt7996/pci.c index 4eb55ffdf1a0..024d8b217e62 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/pci.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/pci.c @@ -140,6 +140,9 @@ static int mt7996_pci_probe(struct pci_dev *pdev, hif2 = mt7996_pci_init_hif2(pdev); dev->hif2 = hif2; + mt76_npu_init(mdev, pci_resource_start(pdev, 0), + pdev->bus && pci_domain_nr(pdev->bus) ? 3 : 2); + ret = mt7996_mmio_wed_init(dev, pdev, false, &irq); if (ret < 0) goto free_wed_or_irq_vector; From 9ba77f1a634a89893022640c7d241c0781d48663 Mon Sep 17 00:00:00 2001 From: Alok Tiwari Date: Mon, 20 Oct 2025 23:48:09 -0700 Subject: [PATCH 1032/1381] wifi: mt76: mt7996: fix typos in comments Fix two minor comment typos in the mt7996 driver: - "Tx/Tx" -> "Rx/Tx" - "tnterrupt" -> "interrupt" Signed-off-by: Alok Tiwari Link: https://patch.msgid.link/20251021064812.1778297-1-alok.a.tiwari@oracle.com Signed-off-by: Felix Fietkau --- drivers/net/wireless/mediatek/mt76/mt7996/mac.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7996/pci.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c index 6ec17d364e83..7d5f9d9b3b62 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c @@ -2553,7 +2553,7 @@ void mt7996_mac_reset_work(struct work_struct *work) mt76_wr(dev, MT_MCU_INT_EVENT, MT_MCU_INT_EVENT_RESET_DONE); mt7996_wait_reset_state(dev, MT_MCU_CMD_NORMAL_STATE); - /* enable DMA Tx/Tx and interrupt */ + /* enable DMA Rx/Tx and interrupt */ mt7996_dma_start(dev, false, false); if (!is_mt7996(&dev->mt76) && dev->mt76.hwrro_mode == MT76_HWRRO_V3) diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/pci.c b/drivers/net/wireless/mediatek/mt76/mt7996/pci.c index 024d8b217e62..12523ddba630 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/pci.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/pci.c @@ -161,7 +161,7 @@ static int mt7996_pci_probe(struct pci_dev *pdev, goto free_wed_or_irq_vector; mt76_wr(dev, MT_INT_MASK_CSR, 0); - /* master switch of PCIe tnterrupt enable */ + /* master switch of PCIe interrupt enable */ mt76_wr(dev, MT_PCIE_MAC_INT_ENABLE, 0xff); if (hif2) { From c1d8beea631ce0c30028c680075f8a0c9b65ad4a Mon Sep 17 00:00:00 2001 From: "Rob Herring (Arm)" Date: Fri, 31 Oct 2025 12:51:13 -0500 Subject: [PATCH 1033/1381] wifi: mt76: Use of_reserved_mem_region_to_resource() for "memory-region" Use the newly added of_reserved_mem_region_to_resource() function to handle "memory-region" properties. Signed-off-by: Rob Herring (Arm) Reviewed-by: AngeloGioacchino Del Regno Link: https://patch.msgid.link/20251031175113.1453100-1-robh@kernel.org Signed-off-by: Felix Fietkau --- .../net/wireless/mediatek/mt76/mt7915/soc.c | 21 +++++++------------ 1 file changed, 8 insertions(+), 13 deletions(-) diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/soc.c b/drivers/net/wireless/mediatek/mt76/mt7915/soc.c index e22a737bd9ab..54ff6de96f3e 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7915/soc.c +++ b/drivers/net/wireless/mediatek/mt76/mt7915/soc.c @@ -284,20 +284,15 @@ static int mt798x_wmac_coninfra_check(struct mt7915_dev *dev) static int mt798x_wmac_coninfra_setup(struct mt7915_dev *dev) { struct device *pdev = dev->mt76.dev; - struct reserved_mem *rmem; - struct device_node *np; + struct resource res; u32 val; + int ret; - np = of_parse_phandle(pdev->of_node, "memory-region", 0); - if (!np) - return -EINVAL; + ret = of_reserved_mem_region_to_resource(pdev->of_node, 0, &res); + if (ret) + return ret; - rmem = of_reserved_mem_lookup(np); - of_node_put(np); - if (!rmem) - return -EINVAL; - - val = (rmem->base >> 16) & MT_TOP_MCU_EMI_BASE_MASK; + val = (res.start >> 16) & MT_TOP_MCU_EMI_BASE_MASK; if (is_mt7986(&dev->mt76)) { /* Set conninfra subsys PLL check */ @@ -318,8 +313,8 @@ static int mt798x_wmac_coninfra_setup(struct mt7915_dev *dev) MT_TOP_EFUSE_BASE_MASK, 0x11f20000 >> 16); } - mt76_wr(dev, MT_INFRA_BUS_EMI_START, rmem->base); - mt76_wr(dev, MT_INFRA_BUS_EMI_END, rmem->size); + mt76_wr(dev, MT_INFRA_BUS_EMI_START, res.start); + mt76_wr(dev, MT_INFRA_BUS_EMI_END, resource_size(&res)); mt76_rr(dev, MT_CONN_INFRA_EFUSE); From 87c394127049538d950422cdd23fd7aa2e71152a Mon Sep 17 00:00:00 2001 From: Ming Yen Hsieh Date: Fri, 31 Oct 2025 17:03:47 +0800 Subject: [PATCH 1034/1381] wifi: mt76: mt7925: refactor regulatory domain handling to regd.[ch] Move regd logic to regd.c and regd.h files Signed-off-by: Ming Yen Hsieh Link: https://patch.msgid.link/20251031090352.1400079-2-mingyen.hsieh@mediatek.com Signed-off-by: Felix Fietkau --- .../wireless/mediatek/mt76/mt7925/Makefile | 2 +- .../net/wireless/mediatek/mt76/mt7925/init.c | 146 +---------------- .../net/wireless/mediatek/mt76/mt7925/main.c | 1 + .../wireless/mediatek/mt76/mt7925/mt7925.h | 2 - .../net/wireless/mediatek/mt76/mt7925/pci.c | 1 + .../net/wireless/mediatek/mt76/mt7925/regd.c | 149 ++++++++++++++++++ .../net/wireless/mediatek/mt76/mt7925/regd.h | 14 ++ 7 files changed, 167 insertions(+), 148 deletions(-) create mode 100644 drivers/net/wireless/mediatek/mt76/mt7925/regd.c create mode 100644 drivers/net/wireless/mediatek/mt76/mt7925/regd.h diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/Makefile b/drivers/net/wireless/mediatek/mt76/mt7925/Makefile index 5d769de979c0..8f1078ce3231 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/Makefile +++ b/drivers/net/wireless/mediatek/mt76/mt7925/Makefile @@ -4,7 +4,7 @@ obj-$(CONFIG_MT7925_COMMON) += mt7925-common.o obj-$(CONFIG_MT7925E) += mt7925e.o obj-$(CONFIG_MT7925U) += mt7925u.o -mt7925-common-y := mac.o mcu.o main.o init.o debugfs.o +mt7925-common-y := mac.o mcu.o regd.o main.o init.o debugfs.o mt7925-common-$(CONFIG_NL80211_TESTMODE) += testmode.o mt7925e-y := pci.o pci_mac.o pci_mcu.o mt7925u-y := usb.o diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/init.c b/drivers/net/wireless/mediatek/mt76/mt7925/init.c index ea397ad48c30..741e38b29b77 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/init.c +++ b/drivers/net/wireless/mediatek/mt76/mt7925/init.c @@ -7,6 +7,7 @@ #include #include #include "mt7925.h" +#include "regd.h" #include "mac.h" #include "mcu.h" @@ -60,151 +61,6 @@ static int mt7925_thermal_init(struct mt792x_phy *phy) return PTR_ERR_OR_ZERO(hwmon); } -void mt7925_regd_be_ctrl(struct mt792x_dev *dev, u8 *alpha2) -{ - struct mt792x_phy *phy = &dev->phy; - struct mt7925_clc_rule_v2 *rule; - struct mt7925_clc *clc; - bool old = dev->has_eht, new = true; - u32 mtcl_conf = mt792x_acpi_get_mtcl_conf(&dev->phy, alpha2); - u8 *pos; - - if (mtcl_conf != MT792X_ACPI_MTCL_INVALID && - (((mtcl_conf >> 4) & 0x3) == 0)) { - new = false; - goto out; - } - - if (!phy->clc[MT792x_CLC_BE_CTRL]) - goto out; - - clc = (struct mt7925_clc *)phy->clc[MT792x_CLC_BE_CTRL]; - pos = clc->data; - - while (1) { - rule = (struct mt7925_clc_rule_v2 *)pos; - - if (rule->alpha2[0] == alpha2[0] && - rule->alpha2[1] == alpha2[1]) { - new = false; - break; - } - - /* Check the last one */ - if (rule->flag & BIT(0)) - break; - - pos += sizeof(*rule); - } - -out: - if (old == new) - return; - - dev->has_eht = new; - mt7925_set_stream_he_eht_caps(phy); -} - -static void -mt7925_regd_channel_update(struct wiphy *wiphy, struct mt792x_dev *dev) -{ -#define IS_UNII_INVALID(idx, sfreq, efreq, cfreq) \ - (!(dev->phy.clc_chan_conf & BIT(idx)) && (cfreq) >= (sfreq) && (cfreq) <= (efreq)) -#define MT7925_UNII_59G_IS_VALID 0x1 -#define MT7925_UNII_6G_IS_VALID 0x1e - struct ieee80211_supported_band *sband; - struct mt76_dev *mdev = &dev->mt76; - struct ieee80211_channel *ch; - u32 mtcl_conf = mt792x_acpi_get_mtcl_conf(&dev->phy, mdev->alpha2); - int i; - - if (mtcl_conf != MT792X_ACPI_MTCL_INVALID) { - if ((mtcl_conf & 0x3) == 0) - dev->phy.clc_chan_conf &= ~MT7925_UNII_59G_IS_VALID; - if (((mtcl_conf >> 2) & 0x3) == 0) - dev->phy.clc_chan_conf &= ~MT7925_UNII_6G_IS_VALID; - } - - sband = wiphy->bands[NL80211_BAND_5GHZ]; - if (!sband) - return; - - for (i = 0; i < sband->n_channels; i++) { - ch = &sband->channels[i]; - - /* UNII-4 */ - if (IS_UNII_INVALID(0, 5845, 5925, ch->center_freq)) - ch->flags |= IEEE80211_CHAN_DISABLED; - } - - sband = wiphy->bands[NL80211_BAND_6GHZ]; - if (!sband) - return; - - for (i = 0; i < sband->n_channels; i++) { - ch = &sband->channels[i]; - - /* UNII-5/6/7/8 */ - if (IS_UNII_INVALID(1, 5925, 6425, ch->center_freq) || - IS_UNII_INVALID(2, 6425, 6525, ch->center_freq) || - IS_UNII_INVALID(3, 6525, 6875, ch->center_freq) || - IS_UNII_INVALID(4, 6875, 7125, ch->center_freq)) - ch->flags |= IEEE80211_CHAN_DISABLED; - } -} - -void mt7925_regd_update(struct mt792x_dev *dev) -{ - struct mt76_dev *mdev = &dev->mt76; - struct ieee80211_hw *hw = mdev->hw; - struct wiphy *wiphy = hw->wiphy; - - if (!dev->regd_change) - return; - - mt7925_mcu_set_clc(dev, mdev->alpha2, dev->country_ie_env); - mt7925_regd_channel_update(wiphy, dev); - mt7925_mcu_set_channel_domain(hw->priv); - mt7925_set_tx_sar_pwr(hw, NULL); - dev->regd_change = false; -} -EXPORT_SYMBOL_GPL(mt7925_regd_update); - -static void -mt7925_regd_notifier(struct wiphy *wiphy, - struct regulatory_request *req) -{ - struct ieee80211_hw *hw = wiphy_to_ieee80211_hw(wiphy); - struct mt792x_dev *dev = mt792x_hw_dev(hw); - struct mt76_dev *mdev = &dev->mt76; - struct mt76_connac_pm *pm = &dev->pm; - - /* allow world regdom at the first boot only */ - if (!memcmp(req->alpha2, "00", 2) && - mdev->alpha2[0] && mdev->alpha2[1]) - return; - - /* do not need to update the same country twice */ - if (!memcmp(req->alpha2, mdev->alpha2, 2) && - dev->country_ie_env == req->country_ie_env) - return; - - memcpy(mdev->alpha2, req->alpha2, 2); - mdev->region = req->dfs_region; - dev->country_ie_env = req->country_ie_env; - dev->regd_change = true; - - if (pm->suspended) - return; - - dev->regd_in_progress = true; - mt792x_mutex_acquire(dev); - mt7925_regd_update(dev); - mt792x_mutex_release(dev); - dev->regd_in_progress = false; - wake_up(&dev->wait); -} - static void mt7925_mac_init_basic_rates(struct mt792x_dev *dev) { int i; diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c index 13424e2fe447..687f58f115b0 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c +++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c @@ -8,6 +8,7 @@ #include #include #include "mt7925.h" +#include "regd.h" #include "mcu.h" #include "mac.h" diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h b/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h index f46eadce31fc..6412969f0ac6 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h +++ b/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h @@ -257,8 +257,6 @@ int mt7925_mcu_chip_config(struct mt792x_dev *dev, const char *cmd); int mt7925_mcu_set_rxfilter(struct mt792x_dev *dev, u32 fif, u8 bit_op, u32 bit_map); -void mt7925_regd_be_ctrl(struct mt792x_dev *dev, u8 *alpha2); -void mt7925_regd_update(struct mt792x_dev *dev); int mt7925_mac_init(struct mt792x_dev *dev); int mt7925_mac_sta_add(struct mt76_dev *mdev, struct ieee80211_vif *vif, struct ieee80211_sta *sta); diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/pci.c b/drivers/net/wireless/mediatek/mt76/mt7925/pci.c index 1de5e41a4061..c10f14386571 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/pci.c +++ b/drivers/net/wireless/mediatek/mt76/mt7925/pci.c @@ -8,6 +8,7 @@ #include "mt7925.h" #include "mac.h" #include "mcu.h" +#include "regd.h" #include "../dma.h" static const struct pci_device_id mt7925_pci_device_table[] = { diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/regd.c b/drivers/net/wireless/mediatek/mt76/mt7925/regd.c new file mode 100644 index 000000000000..8d7ce274adcd --- /dev/null +++ b/drivers/net/wireless/mediatek/mt76/mt7925/regd.c @@ -0,0 +1,149 @@ +// SPDX-License-Identifier: BSD-3-Clause-Clear +/* Copyright (C) 2025 MediaTek Inc. */ + +#include "mt7925.h" +#include "regd.h" +#include "mcu.h" + +void mt7925_regd_be_ctrl(struct mt792x_dev *dev, u8 *alpha2) +{ + struct mt792x_phy *phy = &dev->phy; + struct mt7925_clc_rule_v2 *rule; + struct mt7925_clc *clc; + bool old = dev->has_eht, new = true; + u32 mtcl_conf = mt792x_acpi_get_mtcl_conf(&dev->phy, alpha2); + u8 *pos; + + if (mtcl_conf != MT792X_ACPI_MTCL_INVALID && + (((mtcl_conf >> 4) & 0x3) == 0)) { + new = false; + goto out; + } + + if (!phy->clc[MT792x_CLC_BE_CTRL]) + goto out; + + clc = (struct mt7925_clc *)phy->clc[MT792x_CLC_BE_CTRL]; + pos = clc->data; + + while (1) { + rule = (struct mt7925_clc_rule_v2 *)pos; + + if (rule->alpha2[0] == alpha2[0] && + rule->alpha2[1] == alpha2[1]) { + new = false; + break; + } + + /* Check the last one */ + if (rule->flag & BIT(0)) + break; + + pos += sizeof(*rule); + } + +out: + if (old == new) + return; + + dev->has_eht = new; + mt7925_set_stream_he_eht_caps(phy); +} + +static void +mt7925_regd_channel_update(struct wiphy *wiphy, struct mt792x_dev *dev) +{ +#define IS_UNII_INVALID(idx, sfreq, efreq, cfreq) \ + (!(dev->phy.clc_chan_conf & BIT(idx)) && (cfreq) >= (sfreq) && (cfreq) <= (efreq)) +#define MT7925_UNII_59G_IS_VALID 0x1 +#define MT7925_UNII_6G_IS_VALID 0x1e + struct ieee80211_supported_band *sband; + struct mt76_dev *mdev = &dev->mt76; + struct ieee80211_channel *ch; + u32 mtcl_conf = mt792x_acpi_get_mtcl_conf(&dev->phy, mdev->alpha2); + int i; + + if (mtcl_conf != MT792X_ACPI_MTCL_INVALID) { + if ((mtcl_conf & 0x3) == 0) + dev->phy.clc_chan_conf &= ~MT7925_UNII_59G_IS_VALID; + if (((mtcl_conf >> 2) & 0x3) == 0) + dev->phy.clc_chan_conf &= ~MT7925_UNII_6G_IS_VALID; + } + + sband = wiphy->bands[NL80211_BAND_5GHZ]; + if (!sband) + return; + + for (i = 0; i < sband->n_channels; i++) { + ch = &sband->channels[i]; + + /* UNII-4 */ + if (IS_UNII_INVALID(0, 5845, 5925, ch->center_freq)) + ch->flags |= IEEE80211_CHAN_DISABLED; + } + + sband = wiphy->bands[NL80211_BAND_6GHZ]; + if (!sband) + return; + + for (i = 0; i < sband->n_channels; i++) { + ch = &sband->channels[i]; + + /* UNII-5/6/7/8 */ + if (IS_UNII_INVALID(1, 5925, 6425, ch->center_freq) || + IS_UNII_INVALID(2, 6425, 6525, ch->center_freq) || + IS_UNII_INVALID(3, 6525, 6875, ch->center_freq) || + IS_UNII_INVALID(4, 6875, 7125, ch->center_freq)) + ch->flags |= IEEE80211_CHAN_DISABLED; + } +} + +void mt7925_regd_update(struct mt792x_dev *dev) +{ + struct mt76_dev *mdev = &dev->mt76; + struct ieee80211_hw *hw = mdev->hw; + struct wiphy *wiphy = hw->wiphy; + + if (!dev->regd_change) + return; + + mt7925_mcu_set_clc(dev, mdev->alpha2, dev->country_ie_env); + mt7925_regd_channel_update(wiphy, dev); + mt7925_mcu_set_channel_domain(hw->priv); + mt7925_set_tx_sar_pwr(hw, NULL); + dev->regd_change = false; +} +EXPORT_SYMBOL_GPL(mt7925_regd_update); + +void mt7925_regd_notifier(struct wiphy *wiphy, struct regulatory_request *req) +{ + struct ieee80211_hw *hw = wiphy_to_ieee80211_hw(wiphy); + struct mt792x_dev *dev = mt792x_hw_dev(hw); + struct mt76_dev *mdev = &dev->mt76; + struct mt76_connac_pm *pm = &dev->pm; + + /* allow world regdom at the first boot only */ + if (!memcmp(req->alpha2, "00", 2) && + mdev->alpha2[0] && mdev->alpha2[1]) + return; + + /* do not need to update the same country twice */ + if (!memcmp(req->alpha2, mdev->alpha2, 2) && + dev->country_ie_env == req->country_ie_env) + return; + + memcpy(mdev->alpha2, req->alpha2, 2); + mdev->region = req->dfs_region; + dev->country_ie_env = req->country_ie_env; + dev->regd_change = true; + + if (pm->suspended) + return; + + dev->regd_in_progress = true; + mt792x_mutex_acquire(dev); + mt7925_regd_update(dev); + mt792x_mutex_release(dev); + dev->regd_in_progress = false; + wake_up(&dev->wait); +} diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/regd.h b/drivers/net/wireless/mediatek/mt76/mt7925/regd.h new file mode 100644 index 000000000000..a504e71472fd --- /dev/null +++ b/drivers/net/wireless/mediatek/mt76/mt7925/regd.h @@ -0,0 +1,14 @@ +/* SPDX-License-Identifier: BSD-3-Clause-Clear */ +/* Copyright (C) 2025 MediaTek Inc. */ + +#ifndef __MT7925_REGD_H +#define __MT7925_REGD_H + +#include "mt7925.h" + +void mt7925_regd_be_ctrl(struct mt792x_dev *dev, u8 *alpha2); +void mt7925_regd_update(struct mt792x_dev *dev); +void mt7925_regd_notifier(struct wiphy *wiphy, struct regulatory_request *req); + +#endif + From e323b841270a0cd17d95872035facc4c7fe03b44 Mon Sep 17 00:00:00 2001 From: Ming Yen Hsieh Date: Fri, 31 Oct 2025 17:03:48 +0800 Subject: [PATCH 1035/1381] wifi: mt76: mt7925: refactor CLC support check flow Move the disable_clc module parameter to regd.c and introduce mt7925_regd_clc_supported() to centralize CLC support checks. Signed-off-by: Ming Yen Hsieh Link: https://patch.msgid.link/20251031090352.1400079-3-mingyen.hsieh@mediatek.com Signed-off-by: Felix Fietkau --- drivers/net/wireless/mediatek/mt76/mt7925/mcu.c | 11 +++++------ drivers/net/wireless/mediatek/mt76/mt7925/regd.c | 13 +++++++++++++ drivers/net/wireless/mediatek/mt76/mt7925/regd.h | 1 + 3 files changed, 19 insertions(+), 6 deletions(-) diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c index 620f61abee0b..983766faa972 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c +++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c @@ -4,16 +4,13 @@ #include #include #include "mt7925.h" +#include "regd.h" #include "mcu.h" #include "mac.h" #define MT_STA_BFER BIT(0) #define MT_STA_BFEE BIT(1) -static bool mt7925_disable_clc; -module_param_named(disable_clc, mt7925_disable_clc, bool, 0644); -MODULE_PARM_DESC(disable_clc, "disable CLC support"); - int mt7925_mcu_parse_response(struct mt76_dev *mdev, int cmd, struct sk_buff *skb, int seq) { @@ -688,8 +685,7 @@ static int mt7925_load_clc(struct mt792x_dev *dev, const char *fw_name) int ret, i, len, offset = 0; dev->phy.clc_chan_conf = 0xff; - if (mt7925_disable_clc || - mt76_is_usb(&dev->mt76)) + if (!mt7925_regd_clc_supported(dev)) return 0; if (mt76_is_mmio(&dev->mt76)) { @@ -3383,6 +3379,9 @@ int mt7925_mcu_set_clc(struct mt792x_dev *dev, u8 *alpha2, struct mt792x_phy *phy = (struct mt792x_phy *)&dev->phy; int i, ret; + if (!ARRAY_SIZE(phy->clc)) + return -ESRCH; + /* submit all clc config */ for (i = 0; i < ARRAY_SIZE(phy->clc); i++) { if (i == MT792x_CLC_BE_CTRL) diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/regd.c b/drivers/net/wireless/mediatek/mt76/mt7925/regd.c index 8d7ce274adcd..9e7b468a8228 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/regd.c +++ b/drivers/net/wireless/mediatek/mt76/mt7925/regd.c @@ -5,6 +5,19 @@ #include "regd.h" #include "mcu.h" +static bool mt7925_disable_clc; +module_param_named(disable_clc, mt7925_disable_clc, bool, 0644); +MODULE_PARM_DESC(disable_clc, "disable CLC support"); + +bool mt7925_regd_clc_supported(struct mt792x_dev *dev) +{ + if (mt7925_disable_clc || + mt76_is_usb(&dev->mt76)) + return false; + + return true; +} + void mt7925_regd_be_ctrl(struct mt792x_dev *dev, u8 *alpha2) { struct mt792x_phy *phy = &dev->phy; diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/regd.h b/drivers/net/wireless/mediatek/mt76/mt7925/regd.h index a504e71472fd..5977e068c1c6 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/regd.h +++ b/drivers/net/wireless/mediatek/mt76/mt7925/regd.h @@ -9,6 +9,7 @@ void mt7925_regd_be_ctrl(struct mt792x_dev *dev, u8 *alpha2); void mt7925_regd_update(struct mt792x_dev *dev); void mt7925_regd_notifier(struct wiphy *wiphy, struct regulatory_request *req); +bool mt7925_regd_clc_supported(struct mt792x_dev *dev); #endif From 330510085907a2f235b0ae3385b5b24c3a9c17ac Mon Sep 17 00:00:00 2001 From: Ming Yen Hsieh Date: Fri, 31 Oct 2025 17:03:49 +0800 Subject: [PATCH 1036/1381] wifi: mt76: mt7925: refactor regulatory notifier flow Rename mt7925_regd_update() to mt7925_mcu_regd_update() to centralize regd updates with error handling. Signed-off-by: Ming Yen Hsieh Link: https://patch.msgid.link/20251031090352.1400079-4-mingyen.hsieh@mediatek.com Signed-off-by: Felix Fietkau --- .../net/wireless/mediatek/mt76/mt7925/pci.c | 2 +- .../net/wireless/mediatek/mt76/mt7925/regd.c | 49 +++++++++++++------ .../net/wireless/mediatek/mt76/mt7925/regd.h | 4 +- 3 files changed, 38 insertions(+), 17 deletions(-) diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/pci.c b/drivers/net/wireless/mediatek/mt76/mt7925/pci.c index c10f14386571..c4161754c01d 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/pci.c +++ b/drivers/net/wireless/mediatek/mt76/mt7925/pci.c @@ -585,7 +585,7 @@ static int _mt7925_pci_resume(struct device *device, bool restore) if (!pm->ds_enable) mt7925_mcu_set_deep_sleep(dev, false); - mt7925_regd_update(dev); + mt7925_mcu_regd_update(dev, mdev->alpha2, dev->country_ie_env); failed: pm->suspended = false; diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/regd.c b/drivers/net/wireless/mediatek/mt76/mt7925/regd.c index 9e7b468a8228..b36c6a6f5da6 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/regd.c +++ b/drivers/net/wireless/mediatek/mt76/mt7925/regd.c @@ -111,29 +111,49 @@ mt7925_regd_channel_update(struct wiphy *wiphy, struct mt792x_dev *dev) } } -void mt7925_regd_update(struct mt792x_dev *dev) +int mt7925_mcu_regd_update(struct mt792x_dev *dev, u8 *alpha2, + enum environment_cap country_ie_env) { - struct mt76_dev *mdev = &dev->mt76; - struct ieee80211_hw *hw = mdev->hw; + struct ieee80211_hw *hw = mt76_hw(dev); struct wiphy *wiphy = hw->wiphy; + int ret = 0; + dev->regd_in_progress = true; + + mt792x_mutex_acquire(dev); if (!dev->regd_change) - return; + goto err; + + ret = mt7925_mcu_set_clc(dev, alpha2, country_ie_env); + if (ret < 0) + goto err; - mt7925_mcu_set_clc(dev, mdev->alpha2, dev->country_ie_env); mt7925_regd_channel_update(wiphy, dev); - mt7925_mcu_set_channel_domain(hw->priv); - mt7925_set_tx_sar_pwr(hw, NULL); + + ret = mt7925_mcu_set_channel_domain(hw->priv); + if (ret < 0) + goto err; + + ret = mt7925_set_tx_sar_pwr(hw, NULL); + if (ret < 0) + goto err; + +err: + mt792x_mutex_release(dev); dev->regd_change = false; + dev->regd_in_progress = false; + wake_up(&dev->wait); + + return ret; } -EXPORT_SYMBOL_GPL(mt7925_regd_update); +EXPORT_SYMBOL_GPL(mt7925_mcu_regd_update); void mt7925_regd_notifier(struct wiphy *wiphy, struct regulatory_request *req) { struct ieee80211_hw *hw = wiphy_to_ieee80211_hw(wiphy); struct mt792x_dev *dev = mt792x_hw_dev(hw); - struct mt76_dev *mdev = &dev->mt76; struct mt76_connac_pm *pm = &dev->pm; + struct mt76_dev *mdev = &dev->mt76; /* allow world regdom at the first boot only */ if (!memcmp(req->alpha2, "00", 2) && @@ -148,15 +168,14 @@ void mt7925_regd_notifier(struct wiphy *wiphy, struct regulatory_request *req) memcpy(mdev->alpha2, req->alpha2, 2); mdev->region = req->dfs_region; dev->country_ie_env = req->country_ie_env; + dev->regd_change = true; if (pm->suspended) + /* postpone the mcu update to resume */ return; - dev->regd_in_progress = true; - mt792x_mutex_acquire(dev); - mt7925_regd_update(dev); - mt792x_mutex_release(dev); - dev->regd_in_progress = false; - wake_up(&dev->wait); + mt7925_mcu_regd_update(dev, req->alpha2, + req->country_ie_env); + return; } diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/regd.h b/drivers/net/wireless/mediatek/mt76/mt7925/regd.h index 5977e068c1c6..edcda72b12b0 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/regd.h +++ b/drivers/net/wireless/mediatek/mt76/mt7925/regd.h @@ -6,8 +6,10 @@ #include "mt7925.h" +int mt7925_mcu_regd_update(struct mt792x_dev *dev, u8 *alpha2, + enum environment_cap country_ie_env); + void mt7925_regd_be_ctrl(struct mt792x_dev *dev, u8 *alpha2); -void mt7925_regd_update(struct mt792x_dev *dev); void mt7925_regd_notifier(struct wiphy *wiphy, struct regulatory_request *req); bool mt7925_regd_clc_supported(struct mt792x_dev *dev); From 6338709a4f8636911b4eacfdd75cc12e28702285 Mon Sep 17 00:00:00 2001 From: Ming Yen Hsieh Date: Fri, 31 Oct 2025 17:03:50 +0800 Subject: [PATCH 1037/1381] wifi: mt76: mt7925: improve EHT capability control in regulatory flow Move EHT flag handling into mt7925_regd_channel_update() to ensure correct channel capability reporting. Signed-off-by: Ming Yen Hsieh Link: https://patch.msgid.link/20251031090352.1400079-5-mingyen.hsieh@mediatek.com Signed-off-by: Felix Fietkau --- .../net/wireless/mediatek/mt76/mt7925/main.c | 2 -- .../net/wireless/mediatek/mt76/mt7925/regd.c | 18 ++++++++++++++++++ 2 files changed, 18 insertions(+), 2 deletions(-) diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c index 687f58f115b0..641be23011b1 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c +++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c @@ -1369,8 +1369,6 @@ void mt7925_scan_work(struct work_struct *work) if (!is_valid_alpha2(evt->alpha2)) break; - mt7925_regd_be_ctrl(phy->dev, evt->alpha2); - if (mdev->alpha2[0] != '0' && mdev->alpha2[1] != '0') break; diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/regd.c b/drivers/net/wireless/mediatek/mt76/mt7925/regd.c index b36c6a6f5da6..12d8aac16e34 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/regd.c +++ b/drivers/net/wireless/mediatek/mt76/mt7925/regd.c @@ -83,6 +83,17 @@ mt7925_regd_channel_update(struct wiphy *wiphy, struct mt792x_dev *dev) dev->phy.clc_chan_conf &= ~MT7925_UNII_6G_IS_VALID; } + sband = wiphy->bands[NL80211_BAND_2GHZ]; + if (!sband) + return; + + for (i = 0; i < sband->n_channels; i++) { + ch = &sband->channels[i]; + + if (!dev->has_eht) + ch->flags |= IEEE80211_CHAN_NO_EHT; + } + sband = wiphy->bands[NL80211_BAND_5GHZ]; if (!sband) return; @@ -93,6 +104,9 @@ mt7925_regd_channel_update(struct wiphy *wiphy, struct mt792x_dev *dev) /* UNII-4 */ if (IS_UNII_INVALID(0, 5845, 5925, ch->center_freq)) ch->flags |= IEEE80211_CHAN_DISABLED; + + if (!dev->has_eht) + ch->flags |= IEEE80211_CHAN_NO_EHT; } sband = wiphy->bands[NL80211_BAND_6GHZ]; @@ -108,6 +122,9 @@ mt7925_regd_channel_update(struct wiphy *wiphy, struct mt792x_dev *dev) IS_UNII_INVALID(3, 6525, 6875, ch->center_freq) || IS_UNII_INVALID(4, 6875, 7125, ch->center_freq)) ch->flags |= IEEE80211_CHAN_DISABLED; + + if (!dev->has_eht) + ch->flags |= IEEE80211_CHAN_NO_EHT; } } @@ -128,6 +145,7 @@ int mt7925_mcu_regd_update(struct mt792x_dev *dev, u8 *alpha2, if (ret < 0) goto err; + mt7925_regd_be_ctrl(dev, alpha2); mt7925_regd_channel_update(wiphy, dev); ret = mt7925_mcu_set_channel_domain(hw->priv); From 3bc62aa4484dcb9529fea70e756743769b1145e6 Mon Sep 17 00:00:00 2001 From: Ming Yen Hsieh Date: Fri, 31 Oct 2025 17:03:51 +0800 Subject: [PATCH 1038/1381] wifi: mt76: mt7925: add auto regdomain switch support Implement 802.11d-based automatic regulatory domain switching to dynamically determine the regulatory domain at runtime. Signed-off-by: Ming Yen Hsieh Link: https://patch.msgid.link/20251031090352.1400079-6-mingyen.hsieh@mediatek.com Signed-off-by: Felix Fietkau --- .../net/wireless/mediatek/mt76/mt7925/init.c | 2 - .../net/wireless/mediatek/mt76/mt7925/mac.c | 5 +- .../net/wireless/mediatek/mt76/mt7925/main.c | 23 +------ .../net/wireless/mediatek/mt76/mt7925/mcu.c | 1 + .../net/wireless/mediatek/mt76/mt7925/regd.c | 61 +++++++++++++++++++ .../net/wireless/mediatek/mt76/mt7925/regd.h | 2 + drivers/net/wireless/mediatek/mt76/mt792x.h | 1 + 7 files changed, 68 insertions(+), 27 deletions(-) diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/init.c b/drivers/net/wireless/mediatek/mt76/mt7925/init.c index 741e38b29b77..36f7e98f256c 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/init.c +++ b/drivers/net/wireless/mediatek/mt76/mt7925/init.c @@ -91,8 +91,6 @@ int mt7925_mac_init(struct mt792x_dev *dev) mt7925_mac_init_basic_rates(dev); - memzero_explicit(&dev->mt76.alpha2, sizeof(dev->mt76.alpha2)); - return 0; } EXPORT_SYMBOL_GPL(mt7925_mac_init); diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mac.c b/drivers/net/wireless/mediatek/mt76/mt7925/mac.c index 19c9bec7c7e6..871b67101976 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/mac.c +++ b/drivers/net/wireless/mediatek/mt76/mt7925/mac.c @@ -6,6 +6,7 @@ #include #include "mt7925.h" #include "../dma.h" +#include "regd.h" #include "mac.h" #include "mcu.h" @@ -1329,9 +1330,7 @@ void mt7925_mac_reset_work(struct work_struct *work) mt7925_vif_connect_iter, NULL); mt76_connac_power_save_sched(&dev->mt76.phy, pm); - mt792x_mutex_acquire(dev); - mt7925_mcu_set_clc(dev, "00", ENVIRON_INDOOR); - mt792x_mutex_release(dev); + mt7925_regd_change(&dev->phy, "00"); } void mt7925_coredump_work(struct work_struct *work) diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c index 641be23011b1..cb913dfedcda 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c +++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c @@ -1313,20 +1313,6 @@ void mt7925_mlo_pm_work(struct work_struct *work) mt7925_mlo_pm_iter, dev); } -static bool is_valid_alpha2(const char *alpha2) -{ - if (!alpha2) - return false; - - if (alpha2[0] == '0' && alpha2[1] == '0') - return true; - - if (isalpha(alpha2[0]) && isalpha(alpha2[1])) - return true; - - return false; -} - void mt7925_scan_work(struct work_struct *work) { struct mt792x_phy *phy; @@ -1335,7 +1321,6 @@ void mt7925_scan_work(struct work_struct *work) scan_work.work); while (true) { - struct mt76_dev *mdev = &phy->dev->mt76; struct sk_buff *skb; struct tlv *tlv; int tlv_len; @@ -1366,13 +1351,7 @@ void mt7925_scan_work(struct work_struct *work) case UNI_EVENT_SCAN_DONE_CHNLINFO: evt = (struct mt7925_mcu_scan_chinfo_event *)tlv->data; - if (!is_valid_alpha2(evt->alpha2)) - break; - - if (mdev->alpha2[0] != '0' && mdev->alpha2[1] != '0') - break; - - mt7925_mcu_set_clc(phy->dev, evt->alpha2, ENVIRON_INDOOR); + mt7925_regd_change(phy, evt->alpha2); break; case UNI_EVENT_SCAN_DONE_NLO: diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c index 983766faa972..54c4ab2f322c 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c +++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c @@ -755,6 +755,7 @@ static int mt7925_load_clc(struct mt792x_dev *dev, const char *fw_name) } } + ret = mt7925_regd_init(phy); out: release_firmware(fw); diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/regd.c b/drivers/net/wireless/mediatek/mt76/mt7925/regd.c index 12d8aac16e34..4565e1132b36 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/regd.c +++ b/drivers/net/wireless/mediatek/mt76/mt7925/regd.c @@ -197,3 +197,64 @@ void mt7925_regd_notifier(struct wiphy *wiphy, struct regulatory_request *req) req->country_ie_env); return; } + +static bool +mt7925_regd_is_valid_alpha2(const char *alpha2) +{ + if (!alpha2) + return false; + + if (alpha2[0] == '0' && alpha2[1] == '0') + return true; + + if (isalpha(alpha2[0]) && isalpha(alpha2[1])) + return true; + + return false; +} + +int mt7925_regd_change(struct mt792x_phy *phy, char *alpha2) +{ + struct wiphy *wiphy = phy->mt76->hw->wiphy; + struct ieee80211_hw *hw = wiphy_to_ieee80211_hw(wiphy); + struct mt792x_dev *dev = mt792x_hw_dev(hw); + struct mt76_dev *mdev = &dev->mt76; + + if (dev->hw_full_reset) + return 0; + + if (!mt7925_regd_is_valid_alpha2(alpha2) || + !mt7925_regd_clc_supported(dev)) + return -EINVAL; + + if (mdev->alpha2[0] != '0' && mdev->alpha2[1] != '0') + return 0; + + /* do not need to update the same country twice */ + if (!memcmp(alpha2, mdev->alpha2, 2)) + return 0; + + if (phy->chip_cap & MT792x_CHIP_CAP_11D_EN) { + return regulatory_hint(wiphy, alpha2); + } else { + return mt7925_mcu_set_clc(dev, alpha2, ENVIRON_INDOOR); + } +} +EXPORT_SYMBOL_GPL(mt7925_regd_change); + +int mt7925_regd_init(struct mt792x_phy *phy) +{ + struct wiphy *wiphy = phy->mt76->hw->wiphy; + struct ieee80211_hw *hw = wiphy_to_ieee80211_hw(wiphy); + struct mt792x_dev *dev = mt792x_hw_dev(hw); + struct mt76_dev *mdev = &dev->mt76; + + if (phy->chip_cap & MT792x_CHIP_CAP_11D_EN) { + wiphy->regulatory_flags |= REGULATORY_COUNTRY_IE_IGNORE | + REGULATORY_DISABLE_BEACON_HINTS; + } else { + memzero_explicit(&mdev->alpha2, sizeof(mdev->alpha2)); + } + + return 0; +} diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/regd.h b/drivers/net/wireless/mediatek/mt76/mt7925/regd.h index edcda72b12b0..0767f078862e 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/regd.h +++ b/drivers/net/wireless/mediatek/mt76/mt7925/regd.h @@ -12,6 +12,8 @@ int mt7925_mcu_regd_update(struct mt792x_dev *dev, u8 *alpha2, void mt7925_regd_be_ctrl(struct mt792x_dev *dev, u8 *alpha2); void mt7925_regd_notifier(struct wiphy *wiphy, struct regulatory_request *req); bool mt7925_regd_clc_supported(struct mt792x_dev *dev); +int mt7925_regd_change(struct mt792x_phy *phy, char *alpha2); +int mt7925_regd_init(struct mt792x_phy *phy); #endif diff --git a/drivers/net/wireless/mediatek/mt76/mt792x.h b/drivers/net/wireless/mediatek/mt76/mt792x.h index da7831f9efec..ed2606e9251a 100644 --- a/drivers/net/wireless/mediatek/mt76/mt792x.h +++ b/drivers/net/wireless/mediatek/mt76/mt792x.h @@ -28,6 +28,7 @@ #define MT792x_CHIP_CAP_CLC_EVT_EN BIT(0) #define MT792x_CHIP_CAP_RSSI_NOTIFY_EVT_EN BIT(1) #define MT792x_CHIP_CAP_WF_RF_PIN_CTRL_EVT_EN BIT(3) +#define MT792x_CHIP_CAP_11D_EN BIT(4) #define MT792x_CHIP_CAP_MLO_EN BIT(8) #define MT792x_CHIP_CAP_MLO_EML_EN BIT(9) From 992c304112631498fd9bdc7d9bcf3840c12e304f Mon Sep 17 00:00:00 2001 From: Ming Yen Hsieh Date: Fri, 31 Oct 2025 17:03:52 +0800 Subject: [PATCH 1039/1381] wifi: mt76: mt7925: disable auto regd changes after user set Add regd_user flag to block automatic regulatory domain updates if set by user. Signed-off-by: Ming Yen Hsieh Link: https://patch.msgid.link/20251031090352.1400079-7-mingyen.hsieh@mediatek.com Signed-off-by: Felix Fietkau --- drivers/net/wireless/mediatek/mt76/mt7925/mcu.c | 1 + drivers/net/wireless/mediatek/mt76/mt7925/regd.c | 7 ++++++- drivers/net/wireless/mediatek/mt76/mt792x.h | 1 + 3 files changed, 8 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c index 54c4ab2f322c..b6ef9d9e5f7f 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c +++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c @@ -685,6 +685,7 @@ static int mt7925_load_clc(struct mt792x_dev *dev, const char *fw_name) int ret, i, len, offset = 0; dev->phy.clc_chan_conf = 0xff; + dev->regd_user = false; if (!mt7925_regd_clc_supported(dev)) return 0; diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/regd.c b/drivers/net/wireless/mediatek/mt76/mt7925/regd.c index 4565e1132b36..292087e882d1 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/regd.c +++ b/drivers/net/wireless/mediatek/mt76/mt7925/regd.c @@ -173,6 +173,10 @@ void mt7925_regd_notifier(struct wiphy *wiphy, struct regulatory_request *req) struct mt76_connac_pm *pm = &dev->pm; struct mt76_dev *mdev = &dev->mt76; + if (req->initiator == NL80211_REGDOM_SET_BY_USER && + !dev->regd_user) + dev->regd_user = true; + /* allow world regdom at the first boot only */ if (!memcmp(req->alpha2, "00", 2) && mdev->alpha2[0] && mdev->alpha2[1]) @@ -224,7 +228,8 @@ int mt7925_regd_change(struct mt792x_phy *phy, char *alpha2) return 0; if (!mt7925_regd_is_valid_alpha2(alpha2) || - !mt7925_regd_clc_supported(dev)) + !mt7925_regd_clc_supported(dev) || + dev->regd_user) return -EINVAL; if (mdev->alpha2[0] != '0' && mdev->alpha2[1] != '0') diff --git a/drivers/net/wireless/mediatek/mt76/mt792x.h b/drivers/net/wireless/mediatek/mt76/mt792x.h index ed2606e9251a..8388638ed550 100644 --- a/drivers/net/wireless/mediatek/mt76/mt792x.h +++ b/drivers/net/wireless/mediatek/mt76/mt792x.h @@ -231,6 +231,7 @@ struct mt792x_dev { bool hw_init_done:1; bool fw_assert:1; bool has_eht:1; + bool regd_user:1; bool regd_in_progress:1; bool aspm_supported:1; bool hif_idle:1; From 2df00805f7dbaa46b60c682aad0d76270b7ba266 Mon Sep 17 00:00:00 2001 From: Caleb James DeLisle Date: Wed, 29 Oct 2025 18:41:43 +0000 Subject: [PATCH 1040/1381] wifi: mt76: mmio_*_copy fix byte order and alignment Update functions which copy to and from MMIO to load bytes as Little Endian, and also support unaligned buffers. PCI devices almost universally use Little Endian ordering for MMIO registers, mt76 is no exception. PCI hardware that is designed to work with Big Endian CPUs often (but not always) "helps" by transparently byte-swapping MMIO reads and writes on the wire. If this is enabled then it cannot be turned off for a single write. On hardware which does not support this, writel() does the swap in software. When we are transferring arbitrary bytes to MMIO space, we need them to arrive in the same order they were in memory, so when the hardware swaps them this is a problem. Rather than care about how our PCI host controller works, we instead load bytes as Little Endian - so on a Big Endian machine this will reverse them, then we use writel() which will put them back in the right order again. This way we do not make it our business whether the swapping is done in software or hardware. Furthermore, inspection of the code shows that these functions are often called with stack-allocated u8 arrays which have no alignment guarantees so we now use (get|put)_unaligned_le32(). Fixes this issue: mt76x2e 0000:02:00.0: ROM patch build: 20141115060606a mt76x2e 0000:02:00.0: Firmware Version: 0.0.00 mt76x2e 0000:02:00.0: Build: 1 mt76x2e 0000:02:00.0: Build Time: 201607111443____ mt76x2e 0000:02:00.0: Firmware failed to start mt76x2e 0000:02:00.0: probe with driver mt76x2e failed with error -145 Tested on: SmartFiber XP8421-B (Big Endian MIPS 34Kc) - MT7612 -> 5g / ap / psk2 - MT7603 -> 2g / sta / psk2 - MT7603 -> 2g / ap / psk2 TpLink Archer v1200v-v2 (Big Endian MIPS 34Kc) - MT7613 -> 5g / ap / psk2 - MT7603 -> 2g / sta / psk2 Signed-off-by: Caleb James DeLisle Link: https://patch.msgid.link/20251029184143.3991388-1-cjd@cjdns.fr Signed-off-by: Felix Fietkau --- drivers/net/wireless/mediatek/mt76/mmio.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) diff --git a/drivers/net/wireless/mediatek/mt76/mmio.c b/drivers/net/wireless/mediatek/mt76/mmio.c index 0ea97295f5c4..05d74cd7248e 100644 --- a/drivers/net/wireless/mediatek/mt76/mmio.c +++ b/drivers/net/wireless/mediatek/mt76/mmio.c @@ -33,13 +33,21 @@ static u32 mt76_mmio_rmw(struct mt76_dev *dev, u32 offset, u32 mask, u32 val) static void mt76_mmio_write_copy(struct mt76_dev *dev, u32 offset, const void *data, int len) { - __iowrite32_copy(dev->mmio.regs + offset, data, DIV_ROUND_UP(len, 4)); + int i; + + for (i = 0; i < ALIGN(len, 4); i += 4) + writel(get_unaligned_le32(data + i), + dev->mmio.regs + offset + i); } static void mt76_mmio_read_copy(struct mt76_dev *dev, u32 offset, void *data, int len) { - __ioread32_copy(data, dev->mmio.regs + offset, DIV_ROUND_UP(len, 4)); + int i; + + for (i = 0; i < ALIGN(len, 4); i += 4) + put_unaligned_le32(readl(dev->mmio.regs + offset + i), + data + i); } static int mt76_mmio_wr_rp(struct mt76_dev *dev, u32 base, From cdb2941a516cf06929293604e2e0f4c1d6f3541e Mon Sep 17 00:00:00 2001 From: Fedor Pchelkin Date: Mon, 27 Oct 2025 14:18:39 +0300 Subject: [PATCH 1041/1381] Revert "wifi: mt76: mt792x: improve monitor interface handling" This reverts commit 55e95ce469d0c61041bae48b2ebb7fcbf6d1ba7f. mt792x drivers don't seem to support multi-radio devices yet. At least they don't mess with `struct wiphy_radio` at the moment. Packet capturing on monitor interface doesn't work after the blamed patch: tcpdump -i wls6mon -n -vvv Revert the NO_VIRTUAL_MONITOR feature for now to resolve the issue. Found by Linux Verification Center (linuxtesting.org). Fixes: 55e95ce469d0 ("wifi: mt76: mt792x: improve monitor interface handling") Signed-off-by: Fedor Pchelkin Link: https://patch.msgid.link/20251027111843.38975-1-pchelkin@ispras.ru Signed-off-by: Felix Fietkau --- drivers/net/wireless/mediatek/mt76/mt792x_core.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/net/wireless/mediatek/mt76/mt792x_core.c b/drivers/net/wireless/mediatek/mt76/mt792x_core.c index 5b50bf94996e..f2ed16feb6c1 100644 --- a/drivers/net/wireless/mediatek/mt76/mt792x_core.c +++ b/drivers/net/wireless/mediatek/mt76/mt792x_core.c @@ -688,7 +688,6 @@ int mt792x_init_wiphy(struct ieee80211_hw *hw) ieee80211_hw_set(hw, SUPPORTS_DYNAMIC_PS); ieee80211_hw_set(hw, SUPPORTS_VHT_EXT_NSS_BW); ieee80211_hw_set(hw, CONNECTION_MONITOR); - ieee80211_hw_set(hw, NO_VIRTUAL_MONITOR); ieee80211_hw_set(hw, SUPPORTS_MULTI_BSSID); ieee80211_hw_set(hw, SUPPORTS_ONLY_HE_MULTI_BSSID); From 5ef6de7bad7255621f5f542d8d313b93662b7524 Mon Sep 17 00:00:00 2001 From: Fedor Pchelkin Date: Mon, 27 Oct 2025 14:18:40 +0300 Subject: [PATCH 1042/1381] wifi: mt76: adjust BSS conf pointer handling Passing a BSS conf pointer to mt76_connac2_mac_tx_rate_val() currently has two inconsistencies. When called from mt76_connac2_mac_write_txwi(), BSS conf is got by dereferencing a vif pointer. A NULL vif isn't accounted for though the function itself supposes it to be NULL and tries to handle this case in previous checks. This looks like a cosmetic change since the drivers calling the function (namely, mt7915 and mt7921) set WANT_MONITOR_VIF flag so judging by info->control.vif initialization inside ieee80211_tx_*() routines it can't actually come as NULL here. The same holds for the BSS conf pointer handling inside mt76_connac2_mac_tx_rate_val(). It is dereferenced before being checked for NULL. The function supposes to handle the case so reorder the check and dereference of the pointer. Again, this looks like a syntax issue only. Found by Linux Verification Center (linuxtesting.org) with SVACE static analysis tool. Co-developed-by: Matvey Kovalev Signed-off-by: Matvey Kovalev Signed-off-by: Fedor Pchelkin Link: https://patch.msgid.link/20251027111843.38975-2-pchelkin@ispras.ru Signed-off-by: Felix Fietkau --- .../wireless/mediatek/mt76/mt76_connac_mac.c | 19 +++++++++++-------- 1 file changed, 11 insertions(+), 8 deletions(-) diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac_mac.c b/drivers/net/wireless/mediatek/mt76/mt76_connac_mac.c index 7af6a578562f..3304b5971be0 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76_connac_mac.c +++ b/drivers/net/wireless/mediatek/mt76/mt76_connac_mac.c @@ -297,16 +297,18 @@ u16 mt76_connac2_mac_tx_rate_val(struct mt76_phy *mphy, struct ieee80211_bss_conf *conf, bool beacon, bool mcast) { - struct mt76_vif_link *mvif = mt76_vif_conf_link(mphy->dev, conf->vif, conf); - struct cfg80211_chan_def *chandef = mvif->ctx ? - &mvif->ctx->def : &mphy->chandef; - u8 nss = 0, mode = 0, band = chandef->chan->band; - int rateidx = 0, mcast_rate; - int offset = 0; + u8 nss = 0, mode = 0, band = NL80211_BAND_2GHZ; + int rateidx = 0, offset = 0, mcast_rate; + struct cfg80211_chan_def *chandef; + struct mt76_vif_link *mvif; if (!conf) goto legacy; + mvif = mt76_vif_conf_link(mphy->dev, conf->vif, conf); + chandef = mvif->ctx ? &mvif->ctx->def : &mphy->chandef; + band = chandef->chan->band; + if (is_mt7921(mphy->dev)) { rateidx = ffs(conf->basic_rates) - 1; goto legacy; @@ -584,8 +586,9 @@ void mt76_connac2_mac_write_txwi(struct mt76_dev *dev, __le32 *txwi, struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)skb->data; bool multicast = ieee80211_is_data(hdr->frame_control) && is_multicast_ether_addr(hdr->addr1); - u16 rate = mt76_connac2_mac_tx_rate_val(mphy, &vif->bss_conf, beacon, - multicast); + u16 rate = mt76_connac2_mac_tx_rate_val(mphy, + vif ? &vif->bss_conf : NULL, + beacon, multicast); u32 val = MT_TXD6_FIXED_BW; /* hardware won't add HTC for mgmt/ctrl frame */ From ee518f914cd901c16fa08a6f228b266551f72a6a Mon Sep 17 00:00:00 2001 From: Marco Crivellari Date: Mon, 22 Sep 2025 11:32:43 +0200 Subject: [PATCH 1043/1381] wifi: mt76: replace use of system_wq with system_percpu_wq Currently if a user enqueue a work item using schedule_delayed_work() the used wq is "system_wq" (per-cpu wq) while queue_delayed_work() use WORK_CPU_UNBOUND (used when a cpu is not specified). The same applies to schedule_work() that is using system_wq and queue_work(), that makes use again of WORK_CPU_UNBOUND. This lack of consistentcy cannot be addressed without refactoring the API. For more details see the Link tag below. This continues the effort to refactor workqueue APIs, which began with the introduction of new workqueues and a new alloc_workqueue flag in: commit 128ea9f6ccfb ("workqueue: Add system_percpu_wq and system_dfl_wq") commit 930c2ea566af ("workqueue: Add new WQ_PERCPU flag") Switch to using system_percpu_wq because system_wq is going away as part of a workqueue restructuring. Suggested-by: Tejun Heo Signed-off-by: Marco Crivellari Link: https://patch.msgid.link/20250922093243.140946-2-marco.crivellari@suse.com Signed-off-by: Felix Fietkau --- drivers/net/wireless/mediatek/mt76/mt7921/init.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7925/init.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/init.c b/drivers/net/wireless/mediatek/mt76/mt7921/init.c index d8410c77abe4..29732315af1c 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7921/init.c +++ b/drivers/net/wireless/mediatek/mt76/mt7921/init.c @@ -343,7 +343,7 @@ int mt7921_register_device(struct mt792x_dev *dev) dev->mphy.hw->wiphy->available_antennas_rx = dev->mphy.chainmask; dev->mphy.hw->wiphy->available_antennas_tx = dev->mphy.chainmask; - queue_work(system_wq, &dev->init_work); + queue_work(system_percpu_wq, &dev->init_work); return 0; } diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/init.c b/drivers/net/wireless/mediatek/mt76/mt7925/init.c index 36f7e98f256c..3ce5d6fcc69d 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/init.c +++ b/drivers/net/wireless/mediatek/mt76/mt7925/init.c @@ -274,7 +274,7 @@ int mt7925_register_device(struct mt792x_dev *dev) dev->mphy.hw->wiphy->available_antennas_rx = dev->mphy.chainmask; dev->mphy.hw->wiphy->available_antennas_tx = dev->mphy.chainmask; - queue_work(system_wq, &dev->init_work); + queue_work(system_percpu_wq, &dev->init_work); return 0; } From 2a035ae2062f38dc5183002d1c5a64ca682170c1 Mon Sep 17 00:00:00 2001 From: Jack Kao Date: Wed, 1 Oct 2025 09:25:06 +0800 Subject: [PATCH 1044/1381] wifi: mt76: mt7925: cqm rssi low/high event notify The implementation amounts to setting the driver flag IEEE80211_VIF_SUPPORTS_CQM_RSSI, and then providing mechanisms for continuously updating enough information to be able to provide notifications to userspace when RSSI drops below a certain threshold Signed-off-by: Jack Kao Signed-off-by: Ming Yen Hsieh Link: https://patch.msgid.link/20251001012506.2168037-1-mingyen.hsieh@mediatek.com Signed-off-by: Felix Fietkau --- .../wireless/mediatek/mt76/mt76_connac_mcu.h | 2 + .../net/wireless/mediatek/mt76/mt7925/main.c | 6 ++ .../net/wireless/mediatek/mt76/mt7925/mcu.c | 82 +++++++++++++++++++ .../net/wireless/mediatek/mt76/mt7925/mcu.h | 8 ++ .../wireless/mediatek/mt76/mt7925/mt7925.h | 7 ++ 5 files changed, 105 insertions(+) diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h index 927138130842..8d59cf43f0e2 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h +++ b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h @@ -1062,6 +1062,7 @@ enum { MCU_UNI_EVENT_ROC = 0x27, MCU_UNI_EVENT_TX_DONE = 0x2d, MCU_UNI_EVENT_THERMAL = 0x35, + MCU_UNI_EVENT_RSSI_MONITOR = 0x41, MCU_UNI_EVENT_NIC_CAPAB = 0x43, MCU_UNI_EVENT_WED_RRO = 0x57, MCU_UNI_EVENT_PER_STA_INFO = 0x6d, @@ -1300,6 +1301,7 @@ enum { MCU_UNI_CMD_THERMAL = 0x35, MCU_UNI_CMD_VOW = 0x37, MCU_UNI_CMD_FIXED_RATE_TABLE = 0x40, + MCU_UNI_CMD_RSSI_MONITOR = 0x41, MCU_UNI_CMD_TESTMODE_CTRL = 0x46, MCU_UNI_CMD_RRO = 0x57, MCU_UNI_CMD_OFFCH_SCAN_CTRL = 0x58, diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c index cb913dfedcda..add4c8d59e93 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c +++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c @@ -431,6 +431,9 @@ mt7925_add_interface(struct ieee80211_hw *hw, struct ieee80211_vif *vif) goto out; vif->driver_flags |= IEEE80211_VIF_BEACON_FILTER; + if (phy->chip_cap & MT792x_CHIP_CAP_RSSI_NOTIFY_EVT_EN) + vif->driver_flags |= IEEE80211_VIF_SUPPORTS_CQM_RSSI; + out: mt792x_mutex_release(dev); @@ -1936,6 +1939,9 @@ static void mt7925_link_info_changed(struct ieee80211_hw *hw, mt7925_mcu_set_eht_pp(mvif->phy->mt76, &mconf->mt76, link_conf, NULL); + if (changed & BSS_CHANGED_CQM) + mt7925_mcu_set_rssimonitor(dev, vif); + mt792x_mutex_release(dev); } diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c index b6ef9d9e5f7f..d8f5ecebf5dc 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c +++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c @@ -447,6 +447,56 @@ mt7925_mcu_tx_done_event(struct mt792x_dev *dev, struct sk_buff *skb) } } +static void +mt7925_mcu_rssi_monitor_iter(void *priv, u8 *mac, + struct ieee80211_vif *vif) +{ + struct mt7925_uni_rssi_monitor_event *event = priv; + enum nl80211_cqm_rssi_threshold_event nl_event; + s32 rssi = le32_to_cpu(event->rssi); + + if (vif->type != NL80211_IFTYPE_STATION) + return; + + if (!(vif->driver_flags & IEEE80211_VIF_SUPPORTS_CQM_RSSI)) + return; + + if (rssi > vif->bss_conf.cqm_rssi_thold) + nl_event = NL80211_CQM_RSSI_THRESHOLD_EVENT_HIGH; + else + nl_event = NL80211_CQM_RSSI_THRESHOLD_EVENT_LOW; + + ieee80211_cqm_rssi_notify(vif, nl_event, rssi, GFP_KERNEL); +} + +static void +mt7925_mcu_rssi_monitor_event(struct mt792x_dev *dev, struct sk_buff *skb) +{ + struct tlv *tlv; + u32 tlv_len; + struct mt7925_uni_rssi_monitor_event *event; + + skb_pull(skb, sizeof(struct mt7925_mcu_rxd) + 4); + tlv = (struct tlv *)skb->data; + tlv_len = skb->len; + + while (tlv_len > 0 && le16_to_cpu(tlv->len) <= tlv_len) { + switch (le16_to_cpu(tlv->tag)) { + case UNI_EVENT_RSSI_MONITOR_INFO: + event = (struct mt7925_uni_rssi_monitor_event *)skb->data; + ieee80211_iterate_active_interfaces_atomic(dev->mt76.hw, + IEEE80211_IFACE_ITER_RESUME_ALL, + mt7925_mcu_rssi_monitor_iter, + event); + break; + default: + break; + } + tlv_len -= le16_to_cpu(tlv->len); + tlv = (struct tlv *)((char *)(tlv) + le16_to_cpu(tlv->len)); + } +} + static void mt7925_mcu_uni_debug_msg_event(struct mt792x_dev *dev, struct sk_buff *skb) { @@ -543,6 +593,9 @@ mt7925_mcu_uni_rx_unsolicited_event(struct mt792x_dev *dev, case MCU_UNI_EVENT_BSS_BEACON_LOSS: mt7925_mcu_connection_loss_event(dev, skb); break; + case MCU_UNI_EVENT_RSSI_MONITOR: + mt7925_mcu_rssi_monitor_event(dev, skb); + break; case MCU_UNI_EVENT_COREDUMP: dev->fw_assert = true; mt76_connac_mcu_coredump_event(&dev->mt76, skb, &dev->coredump); @@ -3819,3 +3872,32 @@ int mt7925_mcu_set_rxfilter(struct mt792x_dev *dev, u32 fif, return mt76_mcu_send_msg(&phy->dev->mt76, MCU_UNI_CMD(BAND_CONFIG), &req, sizeof(req), true); } + +int mt7925_mcu_set_rssimonitor(struct mt792x_dev *dev, struct ieee80211_vif *vif) +{ + struct mt792x_bss_conf *mconf = mt792x_link_conf_to_mconf(&vif->bss_conf); + struct { + struct { + u8 bss_idx; + u8 pad[3]; + } __packed hdr; + __le16 tag; + __le16 len; + u8 enable; + s8 cqm_rssi_high; + s8 cqm_rssi_low; + u8 rsv; + } req = { + .hdr = { + .bss_idx = mconf->mt76.idx, + }, + .tag = cpu_to_le16(UNI_CMD_RSSI_MONITOR_SET), + .len = cpu_to_le16(sizeof(req) - 4), + .enable = vif->cfg.assoc, + .cqm_rssi_high = (s8)(vif->bss_conf.cqm_rssi_thold + vif->bss_conf.cqm_rssi_hyst), + .cqm_rssi_low = (s8)(vif->bss_conf.cqm_rssi_thold - vif->bss_conf.cqm_rssi_hyst), + }; + + return mt76_mcu_send_msg(&dev->mt76, MCU_UNI_CMD(RSSI_MONITOR), &req, + sizeof(req), false); +} diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h index 1a11e977a244..e09e0600534a 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h +++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h @@ -152,6 +152,14 @@ enum { UNI_EVENT_SCAN_DONE_NLO = 3, }; +enum { + UNI_CMD_RSSI_MONITOR_SET = 0, +}; + +enum { + UNI_EVENT_RSSI_MONITOR_INFO = 0, +}; + enum connac3_mcu_cipher_type { CONNAC3_CIPHER_NONE = 0, CONNAC3_CIPHER_WEP40 = 1, diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h b/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h index 6412969f0ac6..6b9bf1b89032 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h +++ b/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h @@ -103,6 +103,12 @@ struct mt7925_uni_beacon_loss_event { struct mt7925_beacon_loss_tlv beacon_loss; } __packed; +struct mt7925_uni_rssi_monitor_event { + __le16 tag; + __le16 len; + __le32 rssi; +} __packed; + #define to_rssi(field, rxv) ((FIELD_GET(field, rxv) - 220) / 2) #define to_rcpi(rssi) (2 * (rssi) + 220) @@ -370,4 +376,5 @@ int mt7925_testmode_cmd(struct ieee80211_hw *hw, struct ieee80211_vif *vif, int mt7925_testmode_dump(struct ieee80211_hw *hw, struct sk_buff *msg, struct netlink_callback *cb, void *data, int len); +int mt7925_mcu_set_rssimonitor(struct mt792x_dev *dev, struct ieee80211_vif *vif); #endif From 361b59b6be7c33c43b619d5cada394efc0f3b398 Mon Sep 17 00:00:00 2001 From: StanleyYP Wang Date: Thu, 6 Nov 2025 14:41:52 +0800 Subject: [PATCH 1045/1381] wifi: mt76: mt7996: fix max nss value when getting rx chainmask Since wiphy->available_antennas_tx now accumulates the chainmask of all the radios of a wiphy, use phy->orig_antenna_mask to get the original max nss for comparison. Fixes: 69d54ce7491d ("wifi: mt76: mt7996: switch to single multi-radio wiphy") Signed-off-by: StanleyYP Wang Signed-off-by: Shayne Chen Link: https://patch.msgid.link/20251106064203.1000505-1-shayne.chen@mediatek.com Signed-off-by: Felix Fietkau --- drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h b/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h index 87730d26669d..891b187566b3 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h +++ b/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h @@ -786,7 +786,7 @@ void mt7996_memcpy_fromio(struct mt7996_dev *dev, void *buf, u32 offset, static inline u16 mt7996_rx_chainmask(struct mt7996_phy *phy) { - int max_nss = hweight8(phy->mt76->hw->wiphy->available_antennas_tx); + int max_nss = hweight16(phy->orig_antenna_mask); int cur_nss = hweight8(phy->mt76->antenna_mask); u16 tx_chainmask = phy->mt76->chainmask; From dab5b2025452f90673329dbc03513377b624ed3c Mon Sep 17 00:00:00 2001 From: Peter Chiu Date: Thu, 6 Nov 2025 14:41:53 +0800 Subject: [PATCH 1046/1381] wifi: mt76: mt7996: no need to wait ACK event for SDO command For the SDO unified command, driver does not need to wait for ACK event so do not mark MCU_CMD_ACK in command header. Signed-off-by: Peter Chiu Signed-off-by: Shayne Chen Link: https://patch.msgid.link/20251106064203.1000505-2-shayne.chen@mediatek.com Signed-off-by: Felix Fietkau --- drivers/net/wireless/mediatek/mt76/mt7996/mcu.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c index dda01fa6d7bb..e97984693290 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c @@ -318,6 +318,9 @@ mt7996_mcu_send_message(struct mt76_dev *mdev, struct sk_buff *skb, else uni_txd->option = MCU_CMD_UNI_EXT_ACK; + if (mcu_cmd == MCU_UNI_CMD_SDO) + uni_txd->option &= ~MCU_CMD_ACK; + if ((cmd & __MCU_CMD_FIELD_WA) && (cmd & __MCU_CMD_FIELD_WM)) uni_txd->s2d_index = MCU_S2D_H2CN; else if (cmd & __MCU_CMD_FIELD_WA) From 5d86765828b47444908a8689f2625872e8dac48f Mon Sep 17 00:00:00 2001 From: Howard Hsu Date: Thu, 6 Nov 2025 14:41:54 +0800 Subject: [PATCH 1047/1381] wifi: mt76: mt7996: fix implicit beamforming support for mt7992 Fix the ibf_timeout field for mt7996, mt7992 and mt7990 chipsets. For the mt7992, this value shall be set as 0xff, while the others shall be set as 0x18. Fixes: ad4c9a8a9803 ("wifi: mt76: mt7996: add implicit beamforming support for mt7992") Signed-off-by: Howard Hsu Signed-off-by: Shayne Chen Link: https://patch.msgid.link/20251106064203.1000505-3-shayne.chen@mediatek.com Signed-off-by: Felix Fietkau --- drivers/net/wireless/mediatek/mt76/mt7996/mcu.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c index e97984693290..82dd34a8780f 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c @@ -1825,8 +1825,8 @@ mt7996_mcu_sta_bfer_tlv(struct mt7996_dev *dev, struct sk_buff *skb, bf->ibf_nrow = tx_ant; if (link_sta->eht_cap.has_eht || link_sta->he_cap.has_he) - bf->ibf_timeout = is_mt7996(&dev->mt76) ? MT7996_IBF_TIMEOUT : - MT7992_IBF_TIMEOUT; + bf->ibf_timeout = is_mt7992(&dev->mt76) ? MT7992_IBF_TIMEOUT : + MT7996_IBF_TIMEOUT; else if (!ebf && link_sta->bandwidth <= IEEE80211_STA_RX_BW_40 && !bf->ncol) bf->ibf_timeout = MT7996_IBF_TIMEOUT_LEGACY; else From feb06d4556203cd27cf3fa31147d43f28f329653 Mon Sep 17 00:00:00 2001 From: Shayne Chen Date: Thu, 6 Nov 2025 14:41:55 +0800 Subject: [PATCH 1048/1381] wifi: mt76: mt7996: support fixed rate for link station Introduce mt7996_link_sta_add_debugfs() to extend fixed rate support for MLO link station. Co-developed-by: Howard Hsu Signed-off-by: Howard Hsu Signed-off-by: Shayne Chen Link: https://patch.msgid.link/20251106064203.1000505-4-shayne.chen@mediatek.com Signed-off-by: Felix Fietkau --- .../wireless/mediatek/mt76/mt7996/debugfs.c | 72 +++++++++++-------- .../net/wireless/mediatek/mt76/mt7996/main.c | 1 + .../wireless/mediatek/mt76/mt7996/mt7996.h | 3 + 3 files changed, 48 insertions(+), 28 deletions(-) diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/debugfs.c b/drivers/net/wireless/mediatek/mt76/mt7996/debugfs.c index a000275b369b..76d623b2cafb 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/debugfs.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/debugfs.c @@ -953,16 +953,34 @@ bool mt7996_debugfs_rx_log(struct mt7996_dev *dev, const void *data, int len) #ifdef CONFIG_MAC80211_DEBUGFS /** per-station debugfs **/ -static ssize_t mt7996_sta_fixed_rate_set(struct file *file, - const char __user *user_buf, - size_t count, loff_t *ppos) +static int +mt7996_queues_show(struct seq_file *s, void *data) +{ + struct ieee80211_sta *sta = s->private; + + mt7996_sta_hw_queue_read(s, sta); + + return 0; +} + +DEFINE_SHOW_ATTRIBUTE(mt7996_queues); + +void mt7996_sta_add_debugfs(struct ieee80211_hw *hw, struct ieee80211_vif *vif, + struct ieee80211_sta *sta, struct dentry *dir) +{ + debugfs_create_file("hw-queues", 0400, dir, sta, &mt7996_queues_fops); +} + +static ssize_t mt7996_link_sta_fixed_rate_set(struct file *file, + const char __user *user_buf, + size_t count, loff_t *ppos) { #define SHORT_PREAMBLE 0 #define LONG_PREAMBLE 1 - struct ieee80211_sta *sta = file->private_data; - struct mt7996_sta *msta = (struct mt7996_sta *)sta->drv_priv; + struct ieee80211_link_sta *link_sta = file->private_data; + struct mt7996_sta *msta = (struct mt7996_sta *)link_sta->sta->drv_priv; struct mt7996_dev *dev = msta->vif->deflink.phy->dev; - struct mt7996_sta_link *msta_link = &msta->deflink; + struct mt7996_sta_link *msta_link; struct ra_rate phy = {}; char buf[100]; int ret; @@ -981,12 +999,13 @@ static ssize_t mt7996_sta_fixed_rate_set(struct file *file, /* mode - cck: 0, ofdm: 1, ht: 2, gf: 3, vht: 4, he_su: 8, he_er: 9 EHT: 15 * bw - bw20: 0, bw40: 1, bw80: 2, bw160: 3, BW320: 4 - * nss - vht: 1~4, he: 1~4, eht: 1~4, others: ignore * mcs - cck: 0~4, ofdm: 0~7, ht: 0~32, vht: 0~9, he_su: 0~11, he_er: 0~2, eht: 0~13 + * nss - vht: 1~4, he: 1~4, eht: 1~4, others: ignore * gi - (ht/vht) lgi: 0, sgi: 1; (he) 0.8us: 0, 1.6us: 1, 3.2us: 2 * preamble - short: 1, long: 0 - * ldpc - off: 0, on: 1 * stbc - off: 0, on: 1 + * ldpc - off: 0, on: 1 + * spe - off: 0, on: 1 * ltf - 1xltf: 0, 2xltf: 1, 4xltf: 2 */ if (sscanf(buf, "%hhu %hhu %hhu %hhu %hu %hhu %hhu %hhu %hhu %hu", @@ -994,9 +1013,16 @@ static ssize_t mt7996_sta_fixed_rate_set(struct file *file, &phy.preamble, &phy.stbc, &phy.ldpc, &phy.spe, <f) != 10) { dev_warn(dev->mt76.dev, "format: Mode BW MCS NSS GI Preamble STBC LDPC SPE ltf\n"); - goto out; + return -EINVAL; } + mutex_lock(&dev->mt76.mutex); + + msta_link = mt76_dereference(msta->link[link_sta->link_id], &dev->mt76); + if (!msta_link) { + ret = -EINVAL; + goto out; + } phy.wlan_idx = cpu_to_le16(msta_link->wcid.idx); phy.gi = cpu_to_le16(gi); phy.ltf = cpu_to_le16(ltf); @@ -1005,36 +1031,26 @@ static ssize_t mt7996_sta_fixed_rate_set(struct file *file, ret = mt7996_mcu_set_fixed_rate_ctrl(dev, &phy, 0); if (ret) - return -EFAULT; + goto out; + ret = count; out: - return count; + mutex_unlock(&dev->mt76.mutex); + return ret; } static const struct file_operations fops_fixed_rate = { - .write = mt7996_sta_fixed_rate_set, + .write = mt7996_link_sta_fixed_rate_set, .open = simple_open, .owner = THIS_MODULE, .llseek = default_llseek, }; -static int -mt7996_queues_show(struct seq_file *s, void *data) +void mt7996_link_sta_add_debugfs(struct ieee80211_hw *hw, struct ieee80211_vif *vif, + struct ieee80211_link_sta *link_sta, + struct dentry *dir) { - struct ieee80211_sta *sta = s->private; - - mt7996_sta_hw_queue_read(s, sta); - - return 0; -} - -DEFINE_SHOW_ATTRIBUTE(mt7996_queues); - -void mt7996_sta_add_debugfs(struct ieee80211_hw *hw, struct ieee80211_vif *vif, - struct ieee80211_sta *sta, struct dentry *dir) -{ - debugfs_create_file("fixed_rate", 0600, dir, sta, &fops_fixed_rate); - debugfs_create_file("hw-queues", 0400, dir, sta, &mt7996_queues_fops); + debugfs_create_file("fixed_rate", 0600, dir, link_sta, &fops_fixed_rate); } #endif diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/main.c b/drivers/net/wireless/mediatek/mt76/mt7996/main.c index 08c4d9c8791b..5e0b57657e7e 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/main.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/main.c @@ -2283,6 +2283,7 @@ const struct ieee80211_ops mt7996_ops = { .twt_teardown_request = mt7996_twt_teardown_request, #ifdef CONFIG_MAC80211_DEBUGFS .sta_add_debugfs = mt7996_sta_add_debugfs, + .link_sta_add_debugfs = mt7996_link_sta_add_debugfs, #endif .set_radar_background = mt7996_set_radar_background, .net_fill_forward_path = mt7996_net_fill_forward_path, diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h b/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h index 891b187566b3..65eeb37ab8c7 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h +++ b/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h @@ -863,6 +863,9 @@ int mt7996_mcu_cp_support(struct mt7996_dev *dev, u8 mode); #ifdef CONFIG_MAC80211_DEBUGFS void mt7996_sta_add_debugfs(struct ieee80211_hw *hw, struct ieee80211_vif *vif, struct ieee80211_sta *sta, struct dentry *dir); +void mt7996_link_sta_add_debugfs(struct ieee80211_hw *hw, struct ieee80211_vif *vif, + struct ieee80211_link_sta *link_sta, + struct dentry *dir); #endif int mt7996_mmio_wed_init(struct mt7996_dev *dev, void *pdev_ptr, bool hif2, int *irq); From bb705a606734e1ce0ff17a4f368a896757ba686d Mon Sep 17 00:00:00 2001 From: Shayne Chen Date: Thu, 6 Nov 2025 14:41:56 +0800 Subject: [PATCH 1049/1381] wifi: mt76: mt7996: fix several fields in mt7996_mcu_bss_basic_tlv() Fix several fields in mt7996_mcu_bss_basic_tlv() that were not obtained from the correct link. Without this patch, the MLD station interface does not function properly. Fixes: 34a41bfbcb71 ("wifi: mt76: mt7996: prepare mt7996_mcu_add_dev/bss_info for MLO support") Signed-off-by: Shayne Chen Acked-by: Lorenzo Bianconi Link: https://patch.msgid.link/20251106064203.1000505-5-shayne.chen@mediatek.com Signed-off-by: Felix Fietkau --- .../net/wireless/mediatek/mt76/mt7996/mcu.c | 25 ++++++++++--------- 1 file changed, 13 insertions(+), 12 deletions(-) diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c index 82dd34a8780f..f1892aaf6a91 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c @@ -1037,7 +1037,6 @@ mt7996_mcu_bss_basic_tlv(struct sk_buff *skb, struct mt76_connac_bss_basic_tlv *bss; u32 type = CONNECTION_INFRA_AP; u16 sta_wlan_idx = wlan_idx; - struct ieee80211_sta *sta; struct tlv *tlv; int idx; @@ -1048,14 +1047,18 @@ mt7996_mcu_bss_basic_tlv(struct sk_buff *skb, break; case NL80211_IFTYPE_STATION: if (enable) { - rcu_read_lock(); - sta = ieee80211_find_sta(vif, vif->bss_conf.bssid); - /* TODO: enable BSS_INFO_UAPSD & BSS_INFO_PM */ - if (sta) { - struct mt76_wcid *wcid; + struct ieee80211_sta *sta; - wcid = (struct mt76_wcid *)sta->drv_priv; - sta_wlan_idx = wcid->idx; + rcu_read_lock(); + sta = ieee80211_find_sta(vif, link_conf->bssid); + if (sta) { + struct mt7996_sta *msta = (void *)sta->drv_priv; + struct mt7996_sta_link *msta_link; + int link_id = link_conf->link_id; + + msta_link = rcu_dereference(msta->link[link_id]); + if (msta_link) + sta_wlan_idx = msta_link->wcid.idx; } rcu_read_unlock(); } @@ -1072,8 +1075,6 @@ mt7996_mcu_bss_basic_tlv(struct sk_buff *skb, tlv = mt7996_mcu_add_uni_tlv(skb, UNI_BSS_INFO_BASIC, sizeof(*bss)); bss = (struct mt76_connac_bss_basic_tlv *)tlv; - bss->bcn_interval = cpu_to_le16(link_conf->beacon_int); - bss->dtim_period = link_conf->dtim_period; bss->bmc_tx_wlan_idx = cpu_to_le16(wlan_idx); bss->sta_idx = cpu_to_le16(sta_wlan_idx); bss->conn_type = cpu_to_le32(type); @@ -1093,10 +1094,10 @@ mt7996_mcu_bss_basic_tlv(struct sk_buff *skb, memcpy(bss->bssid, link_conf->bssid, ETH_ALEN); bss->bcn_interval = cpu_to_le16(link_conf->beacon_int); - bss->dtim_period = vif->bss_conf.dtim_period; + bss->dtim_period = link_conf->dtim_period; bss->phymode = mt76_connac_get_phy_mode(phy, vif, chandef->chan->band, NULL); - bss->phymode_ext = mt76_connac_get_phy_mode_ext(phy, &vif->bss_conf, + bss->phymode_ext = mt76_connac_get_phy_mode_ext(phy, link_conf, chandef->chan->band); return 0; From e077071e7ac48d5453072f615d51629891c5b90d Mon Sep 17 00:00:00 2001 From: Shayne Chen Date: Thu, 6 Nov 2025 14:41:57 +0800 Subject: [PATCH 1050/1381] wifi: mt76: mt7996: fix teardown command for an MLD peer For an MLD peer, we only need to call the teardown command when removing the last link, and there's no need to call mt7996_mcu_add_sta() for the earlier links. Fixes: c1d6dd5d03eb ("wifi: mt76: mt7996: Add mt7996_mcu_teardown_mld_sta rouine") Signed-off-by: Shayne Chen Acked-by: Lorenzo Bianconi Link: https://patch.msgid.link/20251106064203.1000505-6-shayne.chen@mediatek.com Signed-off-by: Felix Fietkau --- drivers/net/wireless/mediatek/mt76/mt7996/main.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/main.c b/drivers/net/wireless/mediatek/mt76/mt7996/main.c index 5e0b57657e7e..685be98b9f27 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/main.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/main.c @@ -1206,13 +1206,13 @@ mt7996_mac_sta_event(struct mt7996_dev *dev, struct ieee80211_vif *vif, mt7996_mac_twt_teardown_flow(dev, link, msta_link, i); - if (sta->mlo && links == BIT(link_id)) /* last link */ - mt7996_mcu_teardown_mld_sta(dev, link, - msta_link); - else + if (!sta->mlo) mt7996_mcu_add_sta(dev, link_conf, link_sta, link, msta_link, CONN_STATE_DISCONNECT, false); + else if (sta->mlo && links == BIT(link_id)) /* last link */ + mt7996_mcu_teardown_mld_sta(dev, link, + msta_link); msta_link->wcid.sta_disabled = 1; msta_link->wcid.sta = 0; links = links & ~BIT(link_id); From 7eaea3a8ba1e9bb58f87e3030f6ce18537e57e1f Mon Sep 17 00:00:00 2001 From: Shayne Chen Date: Thu, 6 Nov 2025 14:41:58 +0800 Subject: [PATCH 1051/1381] wifi: mt76: mt7996: set link_valid field when initializing wcid This ensures the upper layer uses the correct link ID during packet processing. Fixes: dd82a9e02c05 ("wifi: mt76: mt7996: Rely on mt7996_sta_link in sta_add/sta_remove callbacks") Signed-off-by: Shayne Chen Acked-by: Lorenzo Bianconi Link: https://patch.msgid.link/20251106064203.1000505-7-shayne.chen@mediatek.com Signed-off-by: Felix Fietkau --- drivers/net/wireless/mediatek/mt76/mt7996/main.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/main.c b/drivers/net/wireless/mediatek/mt76/mt7996/main.c index 685be98b9f27..c26f8f49ce44 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/main.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/main.c @@ -343,6 +343,7 @@ int mt7996_vif_link_add(struct mt76_phy *mphy, struct ieee80211_vif *vif, INIT_LIST_HEAD(&msta_link->rc_list); msta_link->wcid.idx = idx; msta_link->wcid.link_id = link_conf->link_id; + msta_link->wcid.link_valid = ieee80211_vif_is_mld(vif); msta_link->wcid.tx_info |= MT_WCID_TX_INFO_SET; mt76_wcid_init(&msta_link->wcid, band_idx); @@ -984,6 +985,7 @@ mt7996_mac_sta_init_link(struct mt7996_dev *dev, msta_link->wcid.sta = 1; msta_link->wcid.idx = idx; msta_link->wcid.link_id = link_id; + msta_link->wcid.link_valid = !!sta->valid_links; msta_link->wcid.def_wcid = &msta->deflink.wcid; ewma_avg_signal_init(&msta_link->avg_ack_signal); From 85cd5534a3f2ec93e7d88713a77df5b4255520df Mon Sep 17 00:00:00 2001 From: Shayne Chen Date: Thu, 6 Nov 2025 14:41:59 +0800 Subject: [PATCH 1052/1381] wifi: mt76: mt7996: use correct link_id when filling TXD and TXP Obtain the correct link ID and, if needed, switch to the corresponding wcid before populating the TX descriptor and TX payload. Rules for link id: - For QoS data of MLD peers (excluding EAPOL), select the primary or secondary wcid based on whether the TID is odd or even to meet FW/HW requirements - For other packets, use IEEE80211_TX_CTRL_MLO_LINK if specified (such as multicast and broadcast packets) Signed-off-by: Shayne Chen Acked-by: Lorenzo Bianconi Link: https://patch.msgid.link/20251106064203.1000505-8-shayne.chen@mediatek.com Signed-off-by: Felix Fietkau --- .../net/wireless/mediatek/mt76/mt7996/mac.c | 34 ++++++++++++++++--- .../net/wireless/mediatek/mt76/mt7996/main.c | 9 +++++ .../net/wireless/mediatek/mt76/mt7996/mcu.c | 4 +-- .../wireless/mediatek/mt76/mt7996/mt7996.h | 1 + 4 files changed, 42 insertions(+), 6 deletions(-) diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c index 7d5f9d9b3b62..579084cbb30f 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c @@ -1035,15 +1035,20 @@ int mt7996_tx_prepare_skb(struct mt76_dev *mdev, void *txwi_ptr, struct ieee80211_sta *sta, struct mt76_tx_info *tx_info) { + struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)tx_info->skb->data; struct mt7996_dev *dev = container_of(mdev, struct mt7996_dev, mt76); struct ieee80211_tx_info *info = IEEE80211_SKB_CB(tx_info->skb); struct ieee80211_key_conf *key = info->control.hw_key; struct ieee80211_vif *vif = info->control.vif; + struct mt7996_vif *mvif = vif ? (struct mt7996_vif *)vif->drv_priv : NULL; + struct mt7996_sta *msta = sta ? (struct mt7996_sta *)sta->drv_priv : NULL; + struct mt76_vif_link *mlink = NULL; struct mt76_txwi_cache *t; int id, i, pid, nbuf = tx_info->nbuf - 1; bool is_8023 = info->flags & IEEE80211_TX_CTL_HW_80211_ENCAP; __le32 *ptr = (__le32 *)txwi_ptr; u8 *txwi = (u8 *)txwi_ptr; + u8 link_id; if (unlikely(tx_info->skb->len <= ETH_HLEN)) return -EINVAL; @@ -1051,6 +1056,30 @@ int mt7996_tx_prepare_skb(struct mt76_dev *mdev, void *txwi_ptr, if (!wcid) wcid = &dev->mt76.global_wcid; + if ((is_8023 || ieee80211_is_data_qos(hdr->frame_control)) && sta->mlo && + likely(tx_info->skb->protocol != cpu_to_be16(ETH_P_PAE))) { + u8 tid = tx_info->skb->priority & IEEE80211_QOS_CTL_TID_MASK; + + link_id = (tid % 2) ? msta->seclink_id : msta->deflink_id; + } else { + link_id = u32_get_bits(info->control.flags, + IEEE80211_TX_CTRL_MLO_LINK); + } + + if (link_id != wcid->link_id && link_id != IEEE80211_LINK_UNSPECIFIED) { + if (msta) { + struct mt7996_sta_link *msta_link = + rcu_dereference(msta->link[link_id]); + + if (msta_link) + wcid = &msta_link->wcid; + } else if (mvif) { + mlink = rcu_dereference(mvif->mt76.link[link_id]); + if (mlink && mlink->wcid) + wcid = mlink->wcid; + } + } + t = (struct mt76_txwi_cache *)(txwi + mdev->drv->txwi_size); t->skb = tx_info->skb; @@ -1155,10 +1184,7 @@ int mt7996_tx_prepare_skb(struct mt76_dev *mdev, void *txwi_ptr, if (!is_8023 && mt7996_tx_use_mgmt(dev, tx_info->skb)) txp->fw.flags |= cpu_to_le16(MT_CT_INFO_MGMT_FRAME); - if (vif) { - struct mt7996_vif *mvif = (struct mt7996_vif *)vif->drv_priv; - struct mt76_vif_link *mlink = NULL; - + if (mvif) { if (wcid->offchannel) mlink = rcu_dereference(mvif->mt76.offchannel_link); if (!mlink) diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/main.c b/drivers/net/wireless/mediatek/mt76/mt7996/main.c index c26f8f49ce44..ef605e81cba9 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/main.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/main.c @@ -963,6 +963,7 @@ mt7996_mac_sta_init_link(struct mt7996_dev *dev, msta_link = &msta->deflink; msta->deflink_id = link_id; + msta->seclink_id = msta->deflink_id; for (i = 0; i < ARRAY_SIZE(sta->txq); i++) { struct mt76_txq *mtxq; @@ -977,6 +978,11 @@ mt7996_mac_sta_init_link(struct mt7996_dev *dev, msta_link = kzalloc(sizeof(*msta_link), GFP_KERNEL); if (!msta_link) return -ENOMEM; + + if (msta->seclink_id == msta->deflink_id && + (sta->valid_links & ~BIT(msta->deflink_id))) + msta->seclink_id = __ffs(sta->valid_links & + ~BIT(msta->deflink_id)); } INIT_LIST_HEAD(&msta_link->rc_list); @@ -1051,6 +1057,8 @@ mt7996_mac_sta_remove_links(struct mt7996_dev *dev, struct ieee80211_vif *vif, if (msta->deflink_id == link_id) { msta->deflink_id = IEEE80211_LINK_UNSPECIFIED; continue; + } else if (msta->seclink_id == link_id) { + msta->seclink_id = IEEE80211_LINK_UNSPECIFIED; } kfree_rcu(msta_link, rcu_head); @@ -1146,6 +1154,7 @@ mt7996_mac_sta_add(struct mt7996_dev *dev, struct ieee80211_vif *vif, mutex_lock(&dev->mt76.mutex); msta->deflink_id = IEEE80211_LINK_UNSPECIFIED; + msta->seclink_id = IEEE80211_LINK_UNSPECIFIED; msta->vif = mvif; err = mt7996_mac_sta_add_links(dev, vif, sta, links); diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c index f1892aaf6a91..2bb98df317a6 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c @@ -2394,8 +2394,8 @@ mt7996_mcu_sta_mld_setup_tlv(struct mt7996_dev *dev, struct sk_buff *skb, mld_setup->primary_id = cpu_to_le16(msta_link->wcid.idx); if (nlinks > 1) { - link_id = __ffs(sta->valid_links & ~BIT(msta->deflink_id)); - msta_link = mt76_dereference(msta->link[link_id], &dev->mt76); + msta_link = mt76_dereference(msta->link[msta->seclink_id], + &dev->mt76); if (!msta_link) return; } diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h b/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h index 65eeb37ab8c7..c47820699ec8 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h +++ b/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h @@ -243,6 +243,7 @@ struct mt7996_sta { struct mt7996_sta_link deflink; /* must be first */ struct mt7996_sta_link __rcu *link[IEEE80211_MLD_MAX_NUM_LINKS]; u8 deflink_id; + u8 seclink_id; struct mt7996_vif *vif; }; From 4fb3b4e7d1ca5453c6167816230370afc15f26bf Mon Sep 17 00:00:00 2001 From: Shayne Chen Date: Thu, 6 Nov 2025 14:42:00 +0800 Subject: [PATCH 1053/1381] wifi: mt76: mt7996: fix MLD group index assignment Fix extender mode and MBSS issues caused by incorrect assignment of the MLD group and remap indices. Fixes: ed01c310eca9 ("wifi: mt76: mt7996: Fix mt7996_mcu_bss_mld_tlv routine") Signed-off-by: Shayne Chen Acked-by: Lorenzo Bianconi Link: https://patch.msgid.link/20251106064203.1000505-9-shayne.chen@mediatek.com Signed-off-by: Felix Fietkau --- .../net/wireless/mediatek/mt76/mt7996/main.c | 58 +++++++++++++------ 1 file changed, 40 insertions(+), 18 deletions(-) diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/main.c b/drivers/net/wireless/mediatek/mt76/mt7996/main.c index ef605e81cba9..66e22f5576f8 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/main.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/main.c @@ -90,9 +90,11 @@ static void mt7996_stop(struct ieee80211_hw *hw, bool suspend) { } -static inline int get_free_idx(u32 mask, u8 start, u8 end) +static inline int get_free_idx(u64 mask, u8 start, u8 end) { - return ffs(~mask & GENMASK(end, start)); + if (~mask & GENMASK_ULL(end, start)) + return __ffs64(~mask & GENMASK_ULL(end, start)) + 1; + return 0; } static int get_omac_idx(enum nl80211_iftype type, u64 mask) @@ -308,12 +310,6 @@ int mt7996_vif_link_add(struct mt76_phy *mphy, struct ieee80211_vif *vif, if (idx < 0) return -ENOSPC; - if (!dev->mld_idx_mask) { /* first link in the group */ - mvif->mld_group_idx = get_own_mld_idx(dev->mld_idx_mask, true); - mvif->mld_remap_idx = get_free_idx(dev->mld_remap_idx_mask, - 0, 15); - } - mld_idx = get_own_mld_idx(dev->mld_idx_mask, false); if (mld_idx < 0) return -ENOSPC; @@ -331,10 +327,6 @@ int mt7996_vif_link_add(struct mt76_phy *mphy, struct ieee80211_vif *vif, return ret; dev->mt76.vif_mask |= BIT_ULL(mlink->idx); - if (!dev->mld_idx_mask) { - dev->mld_idx_mask |= BIT_ULL(mvif->mld_group_idx); - dev->mld_remap_idx_mask |= BIT_ULL(mvif->mld_remap_idx); - } dev->mld_idx_mask |= BIT_ULL(link->mld_idx); phy->omac_mask |= BIT_ULL(mlink->omac_idx); @@ -424,11 +416,6 @@ void mt7996_vif_link_remove(struct mt76_phy *mphy, struct ieee80211_vif *vif, dev->mt76.vif_mask &= ~BIT_ULL(mlink->idx); dev->mld_idx_mask &= ~BIT_ULL(link->mld_idx); phy->omac_mask &= ~BIT_ULL(mlink->omac_idx); - if (!(dev->mld_idx_mask & ~BIT_ULL(mvif->mld_group_idx))) { - /* last link */ - dev->mld_idx_mask &= ~BIT_ULL(mvif->mld_group_idx); - dev->mld_remap_idx_mask &= ~BIT_ULL(mvif->mld_remap_idx); - } spin_lock_bh(&dev->mt76.sta_poll_lock); if (!list_empty(&msta_link->wcid.poll_list)) @@ -2228,7 +2215,42 @@ mt7996_change_vif_links(struct ieee80211_hw *hw, struct ieee80211_vif *vif, u16 old_links, u16 new_links, struct ieee80211_bss_conf *old[IEEE80211_MLD_MAX_NUM_LINKS]) { - return 0; + struct mt7996_dev *dev = mt7996_hw_dev(hw); + struct mt7996_vif *mvif = (struct mt7996_vif *)vif->drv_priv; + int ret = 0; + + mutex_lock(&dev->mt76.mutex); + + if (!old_links) { + int idx; + + idx = get_own_mld_idx(dev->mld_idx_mask, true); + if (idx < 0) { + ret = -ENOSPC; + goto out; + } + mvif->mld_group_idx = idx; + dev->mld_idx_mask |= BIT_ULL(mvif->mld_group_idx); + + idx = get_free_idx(dev->mld_remap_idx_mask, 0, 15) - 1; + if (idx < 0) { + ret = -ENOSPC; + goto out; + } + mvif->mld_remap_idx = idx; + dev->mld_remap_idx_mask |= BIT_ULL(mvif->mld_remap_idx); + } + + if (new_links) + goto out; + + dev->mld_idx_mask &= ~BIT_ULL(mvif->mld_group_idx); + dev->mld_remap_idx_mask &= ~BIT_ULL(mvif->mld_remap_idx); + +out: + mutex_unlock(&dev->mt76.mutex); + + return ret; } static void From e11be918d91e7d33ac4bad41dbe666a9abf1cfaa Mon Sep 17 00:00:00 2001 From: Shayne Chen Date: Thu, 6 Nov 2025 14:42:01 +0800 Subject: [PATCH 1054/1381] wifi: mt76: mt7996: fix MLO set key and group key issues This patch fixes the following key issues: - Pass correct link BSS to mt7996_mcu_add_key(), and use HW beacon protection mode for mt7990 chipset - Do not do group key deletion for GTK and IGTK due to FW design, the delete key command will delete all group keys of a link BSS - For deleting BIGTK, FW adds a new flow, but the "sec->add" field should be filled with "SET_KEY". Note that if BIGTK is not deleted, it will cause beacon decryption issue when switching from an AP interface to a station interface Fixes: 0c45d52276fd ("wifi: mt76: mt7996: fix setting beacon protection keys") Co-developed-by: Allen Ye Signed-off-by: Allen Ye Co-developed-by: Peter Chiu Signed-off-by: Peter Chiu Signed-off-by: Shayne Chen Link: https://patch.msgid.link/20251106064203.1000505-10-shayne.chen@mediatek.com Signed-off-by: Felix Fietkau --- .../net/wireless/mediatek/mt76/mt7996/mac.c | 7 +++- .../net/wireless/mediatek/mt76/mt7996/main.c | 5 +-- .../net/wireless/mediatek/mt76/mt7996/mcu.c | 35 +++++++++++++------ .../wireless/mediatek/mt76/mt7996/mt7996.h | 2 +- 4 files changed, 34 insertions(+), 15 deletions(-) diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c index 579084cbb30f..c1f2938d1f21 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c @@ -795,6 +795,7 @@ mt7996_mac_write_txwi_80211(struct mt7996_dev *dev, __le32 *txwi, u8 tid = skb->priority & IEEE80211_QOS_CTL_TID_MASK; __le16 fc = hdr->frame_control, sc = hdr->seq_ctrl; u16 seqno = le16_to_cpu(sc); + bool hw_bigtk = false; u8 fc_type, fc_stype; u32 val; @@ -820,7 +821,11 @@ mt7996_mac_write_txwi_80211(struct mt7996_dev *dev, __le32 *txwi, info->flags & IEEE80211_TX_CTL_USE_MINRATE) val |= MT_TXD1_FIXED_RATE; - if (key && multicast && ieee80211_is_robust_mgmt_frame(skb)) { + if (is_mt7990(&dev->mt76) && ieee80211_is_beacon(fc) && + (wcid->hw_key_idx2 == 6 || wcid->hw_key_idx2 == 7)) + hw_bigtk = true; + + if ((key && multicast && ieee80211_is_robust_mgmt_frame(skb)) || hw_bigtk) { val |= MT_TXD1_BIP; txwi[3] &= ~cpu_to_le32(MT_TXD3_PROTECT_FRAME); } diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/main.c b/drivers/net/wireless/mediatek/mt76/mt7996/main.c index 66e22f5576f8..68ce6f96348b 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/main.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/main.c @@ -249,12 +249,13 @@ mt7996_set_hw_key(struct ieee80211_hw *hw, enum set_key_cmd cmd, else if (idx == *wcid_keyidx) *wcid_keyidx = -1; - if (cmd != SET_KEY && sta) + /* only do remove key for BIGTK */ + if (cmd != SET_KEY && !is_bigtk) return 0; mt76_wcid_key_setup(&dev->mt76, &msta_link->wcid, key); - err = mt7996_mcu_add_key(&dev->mt76, vif, key, + err = mt7996_mcu_add_key(&dev->mt76, link, key, MCU_WMWA_UNI_CMD(STA_REC_UPDATE), &msta_link->wcid, cmd); diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c index 2bb98df317a6..14a88ef79b6c 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c @@ -2530,7 +2530,7 @@ int mt7996_mcu_teardown_mld_sta(struct mt7996_dev *dev, } static int -mt7996_mcu_sta_key_tlv(struct mt76_wcid *wcid, +mt7996_mcu_sta_key_tlv(struct mt76_dev *dev, struct mt76_wcid *wcid, struct sk_buff *skb, struct ieee80211_key_conf *key, enum set_key_cmd cmd) @@ -2542,7 +2542,10 @@ mt7996_mcu_sta_key_tlv(struct mt76_wcid *wcid, tlv = mt76_connac_mcu_add_tlv(skb, STA_REC_KEY_V2, sizeof(*sec)); sec = (struct sta_rec_sec_uni *)tlv; - sec->add = 0; + /* due to connac3 FW design, we only do remove key for BIGTK; even for + * removal, the field should be filled with SET_KEY + */ + sec->add = SET_KEY; sec->n_cipher = 1; sec_key = &sec->key[0]; sec_key->wlan_idx = cpu_to_le16(wcid->idx); @@ -2582,29 +2585,33 @@ mt7996_mcu_sta_key_tlv(struct mt76_wcid *wcid, case WLAN_CIPHER_SUITE_BIP_GMAC_256: sec_key->cipher_id = MCU_CIPHER_BCN_PROT_GMAC_256; break; + case WLAN_CIPHER_SUITE_BIP_CMAC_256: + if (!is_mt7990(dev)) + return -EOPNOTSUPP; + sec_key->cipher_id = MCU_CIPHER_BCN_PROT_CMAC_256; + break; default: return -EOPNOTSUPP; } - sec_key->bcn_mode = BP_SW_MODE; + sec_key->bcn_mode = is_mt7990(dev) ? BP_HW_MODE : BP_SW_MODE; return 0; } -int mt7996_mcu_add_key(struct mt76_dev *dev, struct ieee80211_vif *vif, +int mt7996_mcu_add_key(struct mt76_dev *dev, struct mt7996_vif_link *link, struct ieee80211_key_conf *key, int mcu_cmd, struct mt76_wcid *wcid, enum set_key_cmd cmd) { - struct mt76_vif_link *mvif = (struct mt76_vif_link *)vif->drv_priv; struct sk_buff *skb; int ret; - skb = __mt76_connac_mcu_alloc_sta_req(dev, mvif, wcid, - MT7996_STA_UPDATE_MAX_SIZE); + skb = __mt76_connac_mcu_alloc_sta_req(dev, (struct mt76_vif_link *)link, + wcid, MT7996_STA_UPDATE_MAX_SIZE); if (IS_ERR(skb)) return PTR_ERR(skb); - ret = mt7996_mcu_sta_key_tlv(wcid, skb, key, cmd); + ret = mt7996_mcu_sta_key_tlv(dev, wcid, skb, key, cmd); if (ret) { dev_kfree_skb(skb); return ret; @@ -2724,12 +2731,18 @@ mt7996_mcu_beacon_mbss(struct sk_buff *rskb, struct sk_buff *skb, static void mt7996_mcu_beacon_cont(struct mt7996_dev *dev, struct ieee80211_bss_conf *link_conf, + struct mt7996_vif_link *link, struct sk_buff *rskb, struct sk_buff *skb, struct bss_bcn_content_tlv *bcn, struct ieee80211_mutable_offsets *offs) { - struct mt76_wcid *wcid = &dev->mt76.global_wcid; - u8 *buf; + u8 *buf, keyidx = link->msta_link.wcid.hw_key_idx2; + struct mt76_wcid *wcid; + + if (is_mt7990(&dev->mt76) && (keyidx == 6 || keyidx == 7)) + wcid = &link->msta_link.wcid; + else + wcid = &dev->mt76.global_wcid; bcn->pkt_len = cpu_to_le16(MT_TXD_SIZE + skb->len); bcn->tim_ie_pos = cpu_to_le16(offs->tim_offset); @@ -2804,7 +2817,7 @@ int mt7996_mcu_add_beacon(struct ieee80211_hw *hw, struct ieee80211_vif *vif, info = IEEE80211_SKB_CB(skb); info->hw_queue |= FIELD_PREP(MT_TX_HW_QUEUE_PHY, mlink->band_idx); - mt7996_mcu_beacon_cont(dev, link_conf, rskb, skb, bcn, &offs); + mt7996_mcu_beacon_cont(dev, link_conf, link, rskb, skb, bcn, &offs); if (link_conf->bssid_indicator) mt7996_mcu_beacon_mbss(rskb, skb, bcn, &offs); mt7996_mcu_beacon_cntdwn(rskb, skb, &offs, link_conf->csa_active); diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h b/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h index c47820699ec8..7a884311800e 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h +++ b/drivers/net/wireless/mediatek/mt76/mt7996/mt7996.h @@ -849,7 +849,7 @@ void mt7996_update_channel(struct mt76_phy *mphy); int mt7996_init_debugfs(struct mt7996_dev *dev); void mt7996_debugfs_rx_fw_monitor(struct mt7996_dev *dev, const void *data, int len); bool mt7996_debugfs_rx_log(struct mt7996_dev *dev, const void *data, int len); -int mt7996_mcu_add_key(struct mt76_dev *dev, struct ieee80211_vif *vif, +int mt7996_mcu_add_key(struct mt76_dev *dev, struct mt7996_vif_link *link, struct ieee80211_key_conf *key, int mcu_cmd, struct mt76_wcid *wcid, enum set_key_cmd cmd); int mt7996_mcu_bcn_prot_enable(struct mt7996_dev *dev, From f1e9f369ae42ee433836b24467e645192d046a51 Mon Sep 17 00:00:00 2001 From: Shayne Chen Date: Thu, 6 Nov 2025 14:42:02 +0800 Subject: [PATCH 1055/1381] wifi: mt76: mt7996: fix using wrong phy to start in mt7996_mac_restart() Pass the correct mt7996_phy to mt7996_run(). Fixes: 0a5df0ec47f7 ("wifi: mt76: mt7996: remove redundant per-phy mac80211 calls during restart") Signed-off-by: Shayne Chen Acked-by: Lorenzo Bianconi Link: https://patch.msgid.link/20251106064203.1000505-11-shayne.chen@mediatek.com Signed-off-by: Felix Fietkau --- drivers/net/wireless/mediatek/mt76/mt7996/mac.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c index c1f2938d1f21..2c869f710193 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c @@ -2368,7 +2368,7 @@ mt7996_mac_restart(struct mt7996_dev *dev) if (!test_bit(MT76_STATE_RUNNING, &phy->mt76->state)) continue; - ret = mt7996_run(&dev->phy); + ret = mt7996_run(phy); if (ret) goto out; } From a4031fec9d0d230224a7edcefa3368c06c317148 Mon Sep 17 00:00:00 2001 From: Shayne Chen Date: Thu, 6 Nov 2025 14:42:03 +0800 Subject: [PATCH 1056/1381] wifi: mt76: mt7996: fix EMI rings for RRO The RRO EMI rings only need to be allocated when WED is not active. This patch fixes command timeout issue for the setting of WED off and RRO on. Fixes: 3a29164425e9 ("wifi: mt76: mt7996: Add SW path for HW-RRO v3.1") Co-developed-by: Rex Lu Signed-off-by: Rex Lu Signed-off-by: Shayne Chen Acked-by: Lorenzo Bianconi Link: https://patch.msgid.link/20251106064203.1000505-12-shayne.chen@mediatek.com Signed-off-by: Felix Fietkau --- .../net/wireless/mediatek/mt76/mt7996/dma.c | 15 +++++++++------ .../net/wireless/mediatek/mt76/mt7996/init.c | 19 ++++++++++++++++--- 2 files changed, 25 insertions(+), 9 deletions(-) diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/dma.c b/drivers/net/wireless/mediatek/mt76/mt7996/dma.c index 7ac4defca29d..274b273df1ee 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/dma.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/dma.c @@ -515,12 +515,15 @@ int mt7996_dma_rro_init(struct mt7996_dev *dev) if (ret) return ret; - /* We need to set cpu idx pointer before resetting the EMI - * queues. - */ - mdev->q_rx[MT_RXQ_RRO_RXDMAD_C].emi_cpu_idx = - &dev->wed_rro.emi_rings_cpu.ptr->ring[0].idx; - mt76_queue_reset(dev, &mdev->q_rx[MT_RXQ_RRO_RXDMAD_C], true); + if (!mtk_wed_device_active(&mdev->mmio.wed)) { + /* We need to set cpu idx pointer before resetting the + * EMI queues. + */ + mdev->q_rx[MT_RXQ_RRO_RXDMAD_C].emi_cpu_idx = + &dev->wed_rro.emi_rings_cpu.ptr->ring[0].idx; + mt76_queue_reset(dev, &mdev->q_rx[MT_RXQ_RRO_RXDMAD_C], + true); + } goto start_hw_rro; } diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/init.c b/drivers/net/wireless/mediatek/mt76/mt7996/init.c index bd7b720c64c5..00a8286bd136 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/init.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/init.c @@ -960,9 +960,10 @@ static int mt7996_wed_rro_init(struct mt7996_dev *dev) MT7996_RRO_MSDU_PG_SIZE_PER_CR); } - if (dev->mt76.hwrro_mode == MT76_HWRRO_V3_1) { + if (!mtk_wed_device_active(&dev->mt76.mmio.wed) && + dev->mt76.hwrro_mode == MT76_HWRRO_V3_1) { ptr = dmam_alloc_coherent(dev->mt76.dma_dev, - sizeof(dev->wed_rro.emi_rings_cpu.ptr), + sizeof(*dev->wed_rro.emi_rings_cpu.ptr), &dev->wed_rro.emi_rings_cpu.phy_addr, GFP_KERNEL); if (!ptr) @@ -971,7 +972,7 @@ static int mt7996_wed_rro_init(struct mt7996_dev *dev) dev->wed_rro.emi_rings_cpu.ptr = ptr; ptr = dmam_alloc_coherent(dev->mt76.dma_dev, - sizeof(dev->wed_rro.emi_rings_dma.ptr), + sizeof(*dev->wed_rro.emi_rings_dma.ptr), &dev->wed_rro.emi_rings_dma.phy_addr, GFP_KERNEL); if (!ptr) @@ -1037,6 +1038,18 @@ static void mt7996_wed_rro_free(struct mt7996_dev *dev) dev->wed_rro.msdu_pg[i].phy_addr); } + if (dev->wed_rro.emi_rings_cpu.ptr) + dmam_free_coherent(dev->mt76.dma_dev, + sizeof(*dev->wed_rro.emi_rings_cpu.ptr), + dev->wed_rro.emi_rings_cpu.ptr, + dev->wed_rro.emi_rings_cpu.phy_addr); + + if (dev->wed_rro.emi_rings_dma.ptr) + dmam_free_coherent(dev->mt76.dma_dev, + sizeof(*dev->wed_rro.emi_rings_dma.ptr), + dev->wed_rro.emi_rings_dma.ptr, + dev->wed_rro.emi_rings_dma.phy_addr); + if (!dev->wed_rro.session.ptr) return; From 2ccbea08dddaacb1b9b5573ecce58c31eab3c967 Mon Sep 17 00:00:00 2001 From: Michael Lo Date: Thu, 6 Nov 2025 17:21:51 +0800 Subject: [PATCH 1057/1381] wifi: mt76: mt7925: ensure the 6GHz A-MPDU density cap from the hardware. Set the 6GHz HE A-MPDU density from the hardware capability instead of a hardcoded value, ensuring accurate capability reporting. Signed-off-by: Michael Lo Signed-off-by: Ming Yen Hsieh Link: https://patch.msgid.link/20251106092151.1061648-1-mingyen.hsieh@mediatek.com Signed-off-by: Felix Fietkau --- drivers/net/wireless/mediatek/mt76/mt7925/main.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c index add4c8d59e93..2d358a96640c 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c +++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c @@ -139,10 +139,14 @@ mt7925_init_he_caps(struct mt792x_phy *phy, enum nl80211_band band, } if (band == NL80211_BAND_6GHZ) { + struct ieee80211_supported_band *sband = + &phy->mt76->sband_5g.sband; + struct ieee80211_sta_ht_cap *ht_cap = &sband->ht_cap; + u16 cap = IEEE80211_HE_6GHZ_CAP_TX_ANTPAT_CONS | IEEE80211_HE_6GHZ_CAP_RX_ANTPAT_CONS; - cap |= u16_encode_bits(IEEE80211_HT_MPDU_DENSITY_0_5, + cap |= u16_encode_bits(ht_cap->ampdu_density, IEEE80211_HE_6GHZ_CAP_MIN_MPDU_START) | u16_encode_bits(IEEE80211_VHT_MAX_AMPDU_1024K, IEEE80211_HE_6GHZ_CAP_MAX_AMPDU_LEN_EXP) | From 5a4bcba26e9fbea87507a81ad891e70bb525014f Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Fri, 14 Nov 2025 14:16:21 +0100 Subject: [PATCH 1058/1381] wifi: mt76: mt7996: grab mt76 mutex in mt7996_mac_sta_event() Grab mt76 mutex in mt7996_mac_sta_event routine in order to rely on mt76_dereference() utility macro. Fixes: ecd72f9695e7e ("wifi: mt76: mt7996: Support MLO in mt7996_mac_sta_event()") Signed-off-by: Lorenzo Bianconi Tested-by: Ben Greear Link: https://patch.msgid.link/20251114-mt76-fix-missing-mtx-v1-1-259ebf11f654@kernel.org Signed-off-by: Felix Fietkau --- drivers/net/wireless/mediatek/mt76/mt7996/main.c | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-) diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/main.c b/drivers/net/wireless/mediatek/mt76/mt7996/main.c index 68ce6f96348b..79e5731d8f0d 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/main.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/main.c @@ -1159,12 +1159,15 @@ mt7996_mac_sta_event(struct mt7996_dev *dev, struct ieee80211_vif *vif, unsigned long links = sta->valid_links; struct ieee80211_link_sta *link_sta; unsigned int link_id; + int err = 0; + + mutex_lock(&dev->mt76.mutex); for_each_sta_active_link(vif, sta, link_sta, link_id) { struct ieee80211_bss_conf *link_conf; struct mt7996_sta_link *msta_link; struct mt7996_vif_link *link; - int i, err; + int i; link_conf = link_conf_dereference_protected(vif, link_id); if (!link_conf) @@ -1184,12 +1187,12 @@ mt7996_mac_sta_event(struct mt7996_dev *dev, struct ieee80211_vif *vif, link, msta_link, CONN_STATE_CONNECT, true); if (err) - return err; + goto unlock; err = mt7996_mcu_add_rate_ctrl(dev, msta_link->sta, vif, link_id, false); if (err) - return err; + goto unlock; msta_link->wcid.tx_info |= MT_WCID_TX_INFO_SET; break; @@ -1198,7 +1201,7 @@ mt7996_mac_sta_event(struct mt7996_dev *dev, struct ieee80211_vif *vif, link, msta_link, CONN_STATE_PORT_SECURE, false); if (err) - return err; + goto unlock; break; case MT76_STA_EVENT_DISASSOC: for (i = 0; i < ARRAY_SIZE(msta_link->twt.flow); i++) @@ -1218,8 +1221,10 @@ mt7996_mac_sta_event(struct mt7996_dev *dev, struct ieee80211_vif *vif, break; } } +unlock: + mutex_unlock(&dev->mt76.mutex); - return 0; + return err; } static void From a84b172cca90a3a6b97afac9113daf3bf1b172b3 Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Fri, 14 Nov 2025 14:16:22 +0100 Subject: [PATCH 1059/1381] wifi: mt76: mt7996: move mt7996_update_beacons under mt76 mutex Move mt7996_update_beacons routine inside mt76 mutex critical section in mt7996_mac_reset_work() in order to run mt7996_vif_conf_link() in mt7996_mcu_add_beacon routine. Fixes: f30906c55a400 ("wifi: mt76: mt7996: disable beacons when going offchannel") Signed-off-by: Lorenzo Bianconi Tested-by: Ben Greear Link: https://patch.msgid.link/20251114-mt76-fix-missing-mtx-v1-2-259ebf11f654@kernel.org Signed-off-by: Felix Fietkau --- drivers/net/wireless/mediatek/mt76/mt7996/mac.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c index 2c869f710193..a3420f683d27 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c @@ -2632,11 +2632,11 @@ void mt7996_mac_reset_work(struct work_struct *work) local_bh_enable(); ieee80211_wake_queues(hw); + mt7996_update_beacons(dev); mutex_unlock(&dev->mt76.mutex); mt7996_npu_hw_init(dev); - mt7996_update_beacons(dev); mt7996_for_each_phy(dev, phy) ieee80211_queue_delayed_work(hw, &phy->mt76->mac_work, From 6aaaaeacf18b2dc2b0f78f241800e0ea680938c7 Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Fri, 14 Nov 2025 14:16:23 +0100 Subject: [PATCH 1060/1381] wifi: mt76: Move mt76_abort_scan out of mt76_reset_device() Move mt76_abort_scan routine out of mt76_reset_device() in order to avoid a possible deadlock since mt76_reset_device routine is running with mt76 mutex help and mt76_abort_scan_complete() can grab mt76 mutex in some cases. Fixes: b36d55610215a ("wifi: mt76: abort scan/roc on hw restart") Signed-off-by: Lorenzo Bianconi Tested-by: Ben Greear Link: https://patch.msgid.link/20251114-mt76-fix-missing-mtx-v1-3-259ebf11f654@kernel.org Signed-off-by: Felix Fietkau --- drivers/net/wireless/mediatek/mt76/mac80211.c | 2 -- drivers/net/wireless/mediatek/mt76/mt7915/mac.c | 2 ++ drivers/net/wireless/mediatek/mt76/mt7996/mac.c | 2 ++ 3 files changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/net/wireless/mediatek/mt76/mac80211.c b/drivers/net/wireless/mediatek/mt76/mac80211.c index 72ed3c825fa6..75772979f438 100644 --- a/drivers/net/wireless/mediatek/mt76/mac80211.c +++ b/drivers/net/wireless/mediatek/mt76/mac80211.c @@ -850,8 +850,6 @@ void mt76_reset_device(struct mt76_dev *dev) } rcu_read_unlock(); - mt76_abort_scan(dev); - INIT_LIST_HEAD(&dev->wcid_list); INIT_LIST_HEAD(&dev->sta_poll_list); dev->vif_mask = 0; diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/mac.c b/drivers/net/wireless/mediatek/mt76/mt7915/mac.c index cfbe457039ce..cefe56c05731 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7915/mac.c +++ b/drivers/net/wireless/mediatek/mt76/mt7915/mac.c @@ -1451,6 +1451,8 @@ mt7915_mac_full_reset(struct mt7915_dev *dev) if (ext_phy) cancel_delayed_work_sync(&ext_phy->mac_work); + mt76_abort_scan(&dev->mt76); + mutex_lock(&dev->mt76.mutex); for (i = 0; i < 10; i++) { if (!mt7915_mac_restart(dev)) diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c index a3420f683d27..8ddb6d69f0da 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c @@ -2451,6 +2451,8 @@ mt7996_mac_full_reset(struct mt7996_dev *dev) mt7996_for_each_phy(dev, phy) cancel_delayed_work_sync(&phy->mt76->mac_work); + mt76_abort_scan(&dev->mt76); + mutex_lock(&dev->mt76.mutex); for (i = 0; i < 10; i++) { if (!mt7996_mac_restart(dev)) From 4fe823b9ee0317b04ddc6d9e00fea892498aa0f2 Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Fri, 14 Nov 2025 14:16:24 +0100 Subject: [PATCH 1061/1381] wifi: mt76: mt7996: skip deflink accounting for offchannel links Do not take into account offchannel links for deflink accounting. Fixes: a3316d2fc669f ("wifi: mt76: mt7996: set vif default link_id adding/removing vif links") Signed-off-by: Lorenzo Bianconi Tested-by: Ben Greear Link: https://patch.msgid.link/20251114-mt76-fix-missing-mtx-v1-4-259ebf11f654@kernel.org Signed-off-by: Felix Fietkau --- drivers/net/wireless/mediatek/mt76/mt7996/main.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/main.c b/drivers/net/wireless/mediatek/mt76/mt7996/main.c index 79e5731d8f0d..2a45db398fd5 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/main.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/main.c @@ -370,7 +370,8 @@ int mt7996_vif_link_add(struct mt76_phy *mphy, struct ieee80211_vif *vif, ieee80211_iter_keys(mphy->hw, vif, mt7996_key_iter, &it); - if (mvif->mt76.deflink_id == IEEE80211_LINK_UNSPECIFIED) + if (!mlink->wcid->offchannel && + mvif->mt76.deflink_id == IEEE80211_LINK_UNSPECIFIED) mvif->mt76.deflink_id = link_conf->link_id; return 0; @@ -401,7 +402,8 @@ void mt7996_vif_link_remove(struct mt76_phy *mphy, struct ieee80211_vif *vif, rcu_assign_pointer(dev->mt76.wcid[idx], NULL); - if (mvif->mt76.deflink_id == link_conf->link_id) { + if (!mlink->wcid->offchannel && + mvif->mt76.deflink_id == link_conf->link_id) { struct ieee80211_bss_conf *iter; unsigned int link_id; From 2a432a6d0066d4ce05a2d0eec1da9e061eb70c49 Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Sat, 15 Nov 2025 11:41:00 +0100 Subject: [PATCH 1062/1381] wifi: mt76: mt7996: skip ieee80211_iter_keys() on scanning link remove mt7996_vif_link_remove routine is executed by mt76_scan_complete() without holding the wiphy mutex triggering the following lockdep warning. WARNING: CPU: 0 PID: 72 at net/mac80211/key.c:1029 ieee80211_iter_keys+0xe4/0x1a0 [mac80211] CPU: 0 UID: 0 PID: 72 Comm: kworker/u32:2 Tainted: G S 6.18.0-rc5+ #27 PREEMPT(full) Tainted: [S]=CPU_OUT_OF_SPEC Hardware name: Default string Default string/SKYBAY, BIOS 5.12 02/15/2023 Workqueue: phy3 mt76_scan_work [mt76] RIP: 0010:ieee80211_iter_keys+0xe4/0x1a0 [mac80211] Code: 4c 48 83 c4 10 5b 5d 41 5c 41 5d 41 5e 41 5f c3 48 8b 47 48 be ff ff ff ff 48 8d 78 68 e8 b4 eb 1e e1 85 c0 0f 85 49 ff ff ff 4c 8b ab 90 1a 00 00 48 8d 83 90 RSP: 0018:ffffc900002f7cb0 EFLAGS: 00010246 RAX: 0000000000000000 RBX: ffff888127e00ee0 RCX: 0000000000000000 RDX: 0000000000000000 RSI: ffff888127e00788 RDI: ffff88811132b5c8 RBP: ffffffffa0ddf400 R08: 0000000000000001 R09: 000000009dcc1dac R10: 0000000000000001 R11: ffff88811132b5a0 R12: ffffc900002f7d00 R13: ffff8882581e6a80 R14: ffff888127e0afc8 R15: ffff888158832038 FS: 0000000000000000(0000) GS:ffff8884da486000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000030a0fd90 CR3: 0000000002c52004 CR4: 00000000003706f0 Call Trace: ? lock_acquire+0xc2/0x2c0 mt7996_vif_link_remove+0x64/0x2b0 [mt7996e] mt76_put_vif_phy_link+0x41/0x50 [mt76] mt76_scan_complete+0x77/0x100 [mt76] mt76_scan_work+0x2eb/0x3f0 [mt76] ? process_one_work+0x1e5/0x6d0 process_one_work+0x221/0x6d0 worker_thread+0x19a/0x340 ? rescuer_thread+0x450/0x450 kthread+0x108/0x220 ? kthreads_online_cpu+0x110/0x110 ret_from_fork+0x1c6/0x220 ? kthreads_online_cpu+0x110/0x110 ret_from_fork_asm+0x11/0x20 irq event stamp: 45471 hardirqs last enabled at (45477): [] __up_console_sem+0x5e/0x70 hardirqs last disabled at (45482): [] __up_console_sem+0x43/0x70 softirqs last enabled at (44500): [] napi_pp_put_page+0xac/0xd0 softirqs last disabled at (44498): [] page_pool_put_unrefed_netmem+0x290/0x3d0 ---[ end trace 0000000000000000 ]--- Fix the issue skipping ieee80211_iter_keys() for scanning links in mt7996_vif_link_remove routine since we have not uploaded any hw keys for these links. Fixes: 04414d7bba78 ("wifi: mt76: mt7996: delete vif keys when requested") Signed-off-by: Lorenzo Bianconi Tested-by: Ben Greear Link: https://patch.msgid.link/20251115-mt7996-key-iter-link-remove-fix-v1-1-4f3f4e1eaa78@kernel.org Signed-off-by: Felix Fietkau --- drivers/net/wireless/mediatek/mt76/mt7996/main.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/main.c b/drivers/net/wireless/mediatek/mt76/mt7996/main.c index 2a45db398fd5..beed795edb24 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/main.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/main.c @@ -392,7 +392,8 @@ void mt7996_vif_link_remove(struct mt76_phy *mphy, struct ieee80211_vif *vif, }; int idx = msta_link->wcid.idx; - ieee80211_iter_keys(mphy->hw, vif, mt7996_key_iter, &it); + if (!mlink->wcid->offchannel) + ieee80211_iter_keys(mphy->hw, vif, mt7996_key_iter, &it); mt7996_mcu_add_sta(dev, link_conf, NULL, link, NULL, CONN_STATE_DISCONNECT, false); From 7545551631fa63101f97974f49ac0b564814f703 Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Tue, 18 Nov 2025 10:30:26 +0100 Subject: [PATCH 1063/1381] wifi: mt76: mt7996: Add missing locking in mt7996_mac_sta_rc_work() Grab the mt76 mutex running mt7996_mac_sta_rc_work() since it is required by mt7996_mcu_add_rate_ctrl routine. Fixes: 28d519d0d493a ("wifi: mt76: Move RCU section in mt7996_mcu_add_rate_ctrl_fixed()") Signed-off-by: Lorenzo Bianconi Link: https://patch.msgid.link/20251118-mt7996-rc-work-missing-mtx-v1-1-0739c493a6cb@kernel.org Signed-off-by: Felix Fietkau --- drivers/net/wireless/mediatek/mt76/mt7996/mac.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c index 8ddb6d69f0da..2560e2f46e89 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c @@ -2890,6 +2890,8 @@ void mt7996_mac_sta_rc_work(struct work_struct *work) LIST_HEAD(list); u32 changed; + mutex_lock(&dev->mt76.mutex); + spin_lock_bh(&dev->mt76.sta_poll_lock); list_splice_init(&dev->sta_rc_list, &list); @@ -2922,6 +2924,8 @@ void mt7996_mac_sta_rc_work(struct work_struct *work) } spin_unlock_bh(&dev->mt76.sta_poll_lock); + + mutex_unlock(&dev->mt76.mutex); } void mt7996_mac_work(struct work_struct *work) From f804a5895ebad2b2d4fb8a3688d2115926e993d5 Mon Sep 17 00:00:00 2001 From: "Mario Limonciello (AMD)" Date: Thu, 20 Nov 2025 09:58:27 -0600 Subject: [PATCH 1064/1381] wifi: mt76: Strip whitespace from build ddate On systems I have with mt7925 cards I've been noticing a blank line in my kernel logs. IE: ``` [ 17.294105] mt7925e 0000:c3:00.0: HW/SW Version: 0x8a108a10, Build Time: 20250721232852a [ 17.314233] r8169 0000:c4:00.0 enp196s0f0: Link is Down ``` This is because the build_date from the header has a newline character as does the dev_info() print. As the firmware isn't guaranteed to always have a newline but the print is, copy the firmware build date to a temporary variable and strip any whitespace from it before showing it in the logs. Signed-off-by: Mario Limonciello (AMD) Link: https://patch.msgid.link/20251120155829.3494747-1-superm1@kernel.org Signed-off-by: Felix Fietkau --- drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c index fba7025ffd3f..ea99167765b0 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c +++ b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c @@ -3101,6 +3101,7 @@ int mt76_connac2_load_patch(struct mt76_dev *dev, const char *fw_name) int i, ret, sem, max_len = mt76_is_sdio(dev) ? 2048 : 4096; const struct mt76_connac2_patch_hdr *hdr; const struct firmware *fw = NULL; + char build_date[17]; sem = mt76_connac_mcu_patch_sem_ctrl(dev, true); switch (sem) { @@ -3124,8 +3125,11 @@ int mt76_connac2_load_patch(struct mt76_dev *dev, const char *fw_name) } hdr = (const void *)fw->data; + strscpy(build_date, hdr->build_date, sizeof(build_date)); + build_date[16] = '\0'; + strim(build_date); dev_info(dev->dev, "HW/SW Version: 0x%x, Build Time: %.16s\n", - be32_to_cpu(hdr->hw_sw_ver), hdr->build_date); + be32_to_cpu(hdr->hw_sw_ver), build_date); for (i = 0; i < be32_to_cpu(hdr->desc.n_region); i++) { struct mt76_connac2_patch_sec *sec; From 066f417be5fd8c7fe581c5550206364735dad7a3 Mon Sep 17 00:00:00 2001 From: Quan Zhou Date: Tue, 18 Nov 2025 19:54:54 +0800 Subject: [PATCH 1065/1381] wifi: mt76: mt792x: fix wifi init fail by setting MCU_RUNNING after CLC load Set the MT76_STATE_MCU_RUNNING bit only after mt7921_load_clc() has successfully completed. Previously, the MCU_RUNNING state was set before loading CLC, which could cause conflict between chip mcu_init retry and mac_reset flow, result in chip init fail and chip abnormal status. By moving the state set after CLC load, firmware initialization becomes robust and resolves init fail issue. Signed-off-by: Quan Zhou Reviewed-by: druth@chromium.org Link: https://patch.msgid.link/19ec8e4465142e774f17801025accd0ae2214092.1763465933.git.quan.zhou@mediatek.com Signed-off-by: Felix Fietkau --- drivers/net/wireless/mediatek/mt76/mt7921/mcu.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7925/mcu.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7921/mcu.c index 8c5e6bdc0fc9..833d0ab64230 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7921/mcu.c +++ b/drivers/net/wireless/mediatek/mt76/mt7921/mcu.c @@ -646,10 +646,10 @@ int mt7921_run_firmware(struct mt792x_dev *dev) if (err) return err; - set_bit(MT76_STATE_MCU_RUNNING, &dev->mphy.state); err = mt7921_load_clc(dev, mt792x_ram_name(dev)); if (err) return err; + set_bit(MT76_STATE_MCU_RUNNING, &dev->mphy.state); return mt7921_mcu_fw_log_2_host(dev, 1); } diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c index d8f5ecebf5dc..cf0fdea45cf7 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c +++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c @@ -1054,10 +1054,10 @@ int mt7925_run_firmware(struct mt792x_dev *dev) if (err) return err; - set_bit(MT76_STATE_MCU_RUNNING, &dev->mphy.state); err = mt7925_load_clc(dev, mt792x_ram_name(dev)); if (err) return err; + set_bit(MT76_STATE_MCU_RUNNING, &dev->mphy.state); return mt7925_mcu_fw_log_2_host(dev, 1); } From 53d1548612670aa8b5d89745116cc33d9d172863 Mon Sep 17 00:00:00 2001 From: Zilin Guan Date: Thu, 13 Nov 2025 06:24:15 +0000 Subject: [PATCH 1066/1381] mt76: mt7615: Fix memory leak in mt7615_mcu_wtbl_sta_add() In mt7615_mcu_wtbl_sta_add(), an skb sskb is allocated. If the subsequent call to mt76_connac_mcu_alloc_wtbl_req() fails, the function returns an error without freeing sskb, leading to a memory leak. Fix this by calling dev_kfree_skb() on sskb in the error handling path to ensure it is properly released. Fixes: 99c457d902cf9 ("mt76: mt7615: move mt7615_mcu_set_bmc to mt7615_mcu_ops") Signed-off-by: Zilin Guan Acked-by: Lorenzo Bianconi Link: https://patch.msgid.link/20251113062415.103611-1-zilin@seu.edu.cn Signed-off-by: Felix Fietkau --- drivers/net/wireless/mediatek/mt76/mt7615/mcu.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7615/mcu.c index 52bbedda672b..fc0054f8bd60 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7615/mcu.c +++ b/drivers/net/wireless/mediatek/mt76/mt7615/mcu.c @@ -874,8 +874,10 @@ mt7615_mcu_wtbl_sta_add(struct mt7615_phy *phy, struct ieee80211_vif *vif, wtbl_hdr = mt76_connac_mcu_alloc_wtbl_req(&dev->mt76, &msta->wcid, WTBL_RESET_AND_SET, NULL, &wskb); - if (IS_ERR(wtbl_hdr)) + if (IS_ERR(wtbl_hdr)) { + dev_kfree_skb(sskb); return PTR_ERR(wtbl_hdr); + } if (enable) { mt76_connac_mcu_wtbl_generic_tlv(&dev->mt76, wskb, vif, sta, From ef0b78b5b6cb139af1273fc5f2720201556b2650 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Fri, 21 Nov 2025 16:35:10 +0300 Subject: [PATCH 1067/1381] i40e: delete a stray tab This return statement is indented one tab too far. Delete a tab. Signed-off-by: Dan Carpenter Reviewed-by: Tony Nguyen Reviewed-by: Aleksandr Loktionov Link: https://patch.msgid.link/aSBqjtA8oF25G1OG@stanley.mountain Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c index 9d91a382612d..8b30a3accd31 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c +++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c @@ -2967,7 +2967,7 @@ static inline int i40e_check_vf_permission(struct i40e_vf *vf, dev_err(&pf->pdev->dev, "Cannot add more MAC addresses: VF reached its maximum allowed limit (%d)\n", mac_add_max); - return -EPERM; + return -EPERM; } if (!vf_trusted) { dev_err(&pf->pdev->dev, From f81171fecd0c33f442fe3aa94cff650d0069152a Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Thu, 20 Nov 2025 20:02:55 -0800 Subject: [PATCH 1068/1381] selftests: hw-net: auto-disable building the iouring C code Looks like the liburing is not updated by distros very aggressively. Presumably because a lot of packages depend on it. I just updated to Fedora 43 and it's still on liburing 2.9. The test is 9mo old, at this stage I think this warrants handling the build failure more gracefully. Detect if iouring is recent enough and if not print a warning and exclude the C prog from build. The Python test will just fail since the binary won't exist. But it removes the major annoyance of having to update liburing from sources when developing other tests. Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20251121040259.3647749-2-kuba@kernel.org Signed-off-by: Jakub Kicinski --- tools/testing/selftests/drivers/net/hw/Makefile | 17 ++++++++++++++++- 1 file changed, 16 insertions(+), 1 deletion(-) diff --git a/tools/testing/selftests/drivers/net/hw/Makefile b/tools/testing/selftests/drivers/net/hw/Makefile index 1760238e9d4f..949aeeeb357d 100644 --- a/tools/testing/selftests/drivers/net/hw/Makefile +++ b/tools/testing/selftests/drivers/net/hw/Makefile @@ -1,7 +1,20 @@ # SPDX-License-Identifier: GPL-2.0+ OR MIT +# Check if io_uring supports zero-copy receive +HAS_IOURING_ZCRX := $(shell \ + echo -e '#include \n' \ + 'void *func = (void *)io_uring_register_ifq;\n' \ + 'int main() {return 0;}' | \ + $(CC) -luring -x c - -o /dev/null 2>&1 && echo y) + +ifeq ($(HAS_IOURING_ZCRX),y) +COND_GEN_FILES += iou-zcrx +else +$(warning excluding iouring tests, liburing not installed or too old) +endif + TEST_GEN_FILES := \ - iou-zcrx \ + $(COND_GEN_FILES) \ toeplitz \ # end of TEST_GEN_FILES @@ -58,4 +71,6 @@ include ../../../net/ynl.mk include ../../../net/bpf.mk +ifeq ($(HAS_IOURING_ZCRX),y) $(OUTPUT)/iou-zcrx: LDLIBS += -luring +endif From 27c512af190e037a6f330e9461fc4741fb77da45 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Thu, 20 Nov 2025 20:02:56 -0800 Subject: [PATCH 1069/1381] selftests: hw-net: toeplitz: make sure NICs have pure Toeplitz configured Make sure that the NIC under test is configured for pure Toeplitz hashing, and no input key transform (no symmetric hashing). Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20251121040259.3647749-3-kuba@kernel.org Signed-off-by: Jakub Kicinski --- .../selftests/drivers/net/hw/toeplitz.py | 29 ++++++++++++------- 1 file changed, 18 insertions(+), 11 deletions(-) diff --git a/tools/testing/selftests/drivers/net/hw/toeplitz.py b/tools/testing/selftests/drivers/net/hw/toeplitz.py index 9019a8c1ff62..642a5cc385b6 100755 --- a/tools/testing/selftests/drivers/net/hw/toeplitz.py +++ b/tools/testing/selftests/drivers/net/hw/toeplitz.py @@ -17,6 +17,9 @@ from lib.py import cmd, bkg, rand_port, defer from lib.py import ksft_in from lib.py import ksft_variants, KsftNamedVariant, KsftSkipEx, KsftFailEx +# "define" for the ID of the Toeplitz hash function +ETH_RSS_HASH_TOP = 1 + def _check_rps_and_rfs_not_configured(cfg): """Verify that RPS is not already configured.""" @@ -34,16 +37,6 @@ def _check_rps_and_rfs_not_configured(cfg): raise KsftSkipEx(f"RFS already configured {rfs_file}: {val}") -def _get_rss_key(cfg): - """ - Read the RSS key from the device. - Return a string in the traditional %02x:%02x:%02x:.. format. - """ - - rss = cfg.ethnl.rss_get({"header": {"dev-index": cfg.ifindex}}) - return ':'.join(f'{b:02x}' for b in rss["hkey"]) - - def _get_cpu_for_irq(irq): with open(f"/proc/irq/{irq}/smp_affinity_list", "r", encoding="utf-8") as fp: @@ -153,8 +146,22 @@ def test(cfg, proto_flag, ipver, grp): # Check that rxhash is enabled ksft_in("receive-hashing: on", cmd(f"ethtool -k {cfg.ifname}").stdout) + rss = cfg.ethnl.rss_get({"header": {"dev-index": cfg.ifindex}}) + # Make sure NIC is configured to use Toeplitz hash, and no key xfrm. + if rss.get('hfunc') != ETH_RSS_HASH_TOP or rss.get('input-xfrm'): + cfg.ethnl.rss_set({"header": {"dev-index": cfg.ifindex}, + "hfunc": ETH_RSS_HASH_TOP, + "input-xfrm": {}}) + defer(cfg.ethnl.rss_set, {"header": {"dev-index": cfg.ifindex}, + "hfunc": rss.get('hfunc'), + "input-xfrm": rss.get('input-xfrm', {}) + }) + # Refresh in case changing hfunc changes things. + rss = cfg.ethnl.rss_get({"header": {"dev-index": cfg.ifindex}}) + + key = ':'.join(f'{b:02x}' for b in rss["hkey"]) + port = rand_port(socket.SOCK_DGRAM) - key = _get_rss_key(cfg) toeplitz_path = cfg.test_dir / "toeplitz" rx_cmd = [ From aa91dbf3eda2977e3046d4838eadf7af4dbd47ec Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Thu, 20 Nov 2025 20:02:57 -0800 Subject: [PATCH 1070/1381] selftests: hw-net: toeplitz: read the RSS key directly from C Now that we have YNL support for RSS accessing the RSS info from C is very easy. Instead of passing the RSS key from Python do it directly in the C code. Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20251121040259.3647749-4-kuba@kernel.org Signed-off-by: Jakub Kicinski --- .../testing/selftests/drivers/net/hw/Makefile | 6 ++- .../selftests/drivers/net/hw/toeplitz.c | 41 ++++++++++++++++++- .../selftests/drivers/net/hw/toeplitz.py | 5 --- 3 files changed, 44 insertions(+), 8 deletions(-) diff --git a/tools/testing/selftests/drivers/net/hw/Makefile b/tools/testing/selftests/drivers/net/hw/Makefile index 949aeeeb357d..7c819fdfa107 100644 --- a/tools/testing/selftests/drivers/net/hw/Makefile +++ b/tools/testing/selftests/drivers/net/hw/Makefile @@ -15,7 +15,6 @@ endif TEST_GEN_FILES := \ $(COND_GEN_FILES) \ - toeplitz \ # end of TEST_GEN_FILES TEST_PROGS = \ @@ -55,7 +54,10 @@ TEST_INCLUDES := \ # # YNL files, must be before "include ..lib.mk" -YNL_GEN_FILES := ncdevmem +YNL_GEN_FILES := \ + ncdevmem \ + toeplitz \ +# end of YNL_GEN_FILES TEST_GEN_FILES += $(YNL_GEN_FILES) TEST_GEN_FILES += $(patsubst %.c,%.o,$(wildcard *.bpf.c)) diff --git a/tools/testing/selftests/drivers/net/hw/toeplitz.c b/tools/testing/selftests/drivers/net/hw/toeplitz.c index afc5f910b006..7420a4e201cc 100644 --- a/tools/testing/selftests/drivers/net/hw/toeplitz.c +++ b/tools/testing/selftests/drivers/net/hw/toeplitz.c @@ -52,6 +52,9 @@ #include #include +#include +#include "ethtool-user.h" + #include "../../../kselftest.h" #include "../../../net/lib/ksft.h" @@ -483,6 +486,42 @@ static void parse_rps_bitmap(const char *arg) rps_silo_to_cpu[cfg_num_rps_cpus++] = i; } +static void read_rss_dev_info_ynl(void) +{ + struct ethtool_rss_get_req *req; + struct ethtool_rss_get_rsp *rsp; + struct ynl_sock *ys; + + ys = ynl_sock_create(&ynl_ethtool_family, NULL); + if (!ys) + error(1, errno, "ynl_sock_create failed"); + + req = ethtool_rss_get_req_alloc(); + if (!req) + error(1, errno, "ethtool_rss_get_req_alloc failed"); + + ethtool_rss_get_req_set_header_dev_name(req, cfg_ifname); + + rsp = ethtool_rss_get(ys, req); + if (!rsp) + error(1, ys->err.code, "YNL: %s", ys->err.msg); + + if (!rsp->_len.hkey) + error(1, 0, "RSS key not available for %s", cfg_ifname); + + if (rsp->_len.hkey < TOEPLITZ_KEY_MIN_LEN || + rsp->_len.hkey > TOEPLITZ_KEY_MAX_LEN) + error(1, 0, "RSS key length %u out of bounds [%u, %u]", + rsp->_len.hkey, TOEPLITZ_KEY_MIN_LEN, + TOEPLITZ_KEY_MAX_LEN); + + memcpy(toeplitz_key, rsp->hkey, rsp->_len.hkey); + + ethtool_rss_get_rsp_free(rsp); + ethtool_rss_get_req_free(req); + ynl_sock_destroy(ys); +} + static void parse_opts(int argc, char **argv) { static struct option long_options[] = { @@ -551,7 +590,7 @@ static void parse_opts(int argc, char **argv) } if (!have_toeplitz) - error(1, 0, "Must supply rss key ('-k')"); + read_rss_dev_info_ynl(); num_cpus = get_nprocs(); if (num_cpus > RSS_MAX_CPUS) diff --git a/tools/testing/selftests/drivers/net/hw/toeplitz.py b/tools/testing/selftests/drivers/net/hw/toeplitz.py index 642a5cc385b6..945c58d23310 100755 --- a/tools/testing/selftests/drivers/net/hw/toeplitz.py +++ b/tools/testing/selftests/drivers/net/hw/toeplitz.py @@ -156,10 +156,6 @@ def test(cfg, proto_flag, ipver, grp): "hfunc": rss.get('hfunc'), "input-xfrm": rss.get('input-xfrm', {}) }) - # Refresh in case changing hfunc changes things. - rss = cfg.ethnl.rss_get({"header": {"dev-index": cfg.ifindex}}) - - key = ':'.join(f'{b:02x}' for b in rss["hkey"]) port = rand_port(socket.SOCK_DGRAM) @@ -170,7 +166,6 @@ def test(cfg, proto_flag, ipver, grp): proto_flag, "-d", str(port), "-i", cfg.ifname, - "-k", key, "-T", "1000", "-s", "-v" From c0105ffc5056852b16824bea9e8e0cd9a598d715 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Thu, 20 Nov 2025 20:02:58 -0800 Subject: [PATCH 1071/1381] selftests: hw-net: toeplitz: read indirection table from the device Replace the simple modulo math with the real indirection table read from the device. This makes the tests pass for mlx5 and bnxt NICs. Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20251121040259.3647749-5-kuba@kernel.org Signed-off-by: Jakub Kicinski --- .../selftests/drivers/net/hw/toeplitz.c | 24 ++++++++++++++++++- 1 file changed, 23 insertions(+), 1 deletion(-) diff --git a/tools/testing/selftests/drivers/net/hw/toeplitz.c b/tools/testing/selftests/drivers/net/hw/toeplitz.c index 7420a4e201cc..a4d04438c313 100644 --- a/tools/testing/selftests/drivers/net/hw/toeplitz.c +++ b/tools/testing/selftests/drivers/net/hw/toeplitz.c @@ -68,6 +68,7 @@ #define FOUR_TUPLE_MAX_LEN ((sizeof(struct in6_addr) * 2) + (sizeof(uint16_t) * 2)) #define RSS_MAX_CPUS (1 << 16) /* real constraint is PACKET_FANOUT_MAX */ +#define RSS_MAX_INDIR (1 << 16) #define RPS_MAX_CPUS 16UL /* must be a power of 2 */ @@ -105,6 +106,8 @@ struct ring_state { static unsigned int rx_irq_cpus[RSS_MAX_CPUS]; /* map from rxq to cpu */ static int rps_silo_to_cpu[RPS_MAX_CPUS]; static unsigned char toeplitz_key[TOEPLITZ_KEY_MAX_LEN]; +static unsigned int rss_indir_tbl[RSS_MAX_INDIR]; +static unsigned int rss_indir_tbl_size; static struct ring_state rings[RSS_MAX_CPUS]; static inline uint32_t toeplitz(const unsigned char *four_tuple, @@ -133,7 +136,12 @@ static inline uint32_t toeplitz(const unsigned char *four_tuple, /* Compare computed cpu with arrival cpu from packet_fanout_cpu */ static void verify_rss(uint32_t rx_hash, int cpu) { - int queue = rx_hash % cfg_num_queues; + int queue; + + if (rss_indir_tbl_size) + queue = rss_indir_tbl[rx_hash % rss_indir_tbl_size]; + else + queue = rx_hash % cfg_num_queues; log_verbose(" rxq %d (cpu %d)", queue, rx_irq_cpus[queue]); if (rx_irq_cpus[queue] != cpu) { @@ -517,6 +525,20 @@ static void read_rss_dev_info_ynl(void) memcpy(toeplitz_key, rsp->hkey, rsp->_len.hkey); + if (rsp->_count.indir > RSS_MAX_INDIR) + error(1, 0, "RSS indirection table too large (%u > %u)", + rsp->_count.indir, RSS_MAX_INDIR); + + /* If indir table not available we'll fallback to simple modulo math */ + if (rsp->_count.indir) { + memcpy(rss_indir_tbl, rsp->indir, + rsp->_count.indir * sizeof(rss_indir_tbl[0])); + rss_indir_tbl_size = rsp->_count.indir; + + log_verbose("RSS indirection table size: %u\n", + rss_indir_tbl_size); + } + ethtool_rss_get_rsp_free(rsp); ethtool_rss_get_req_free(req); ynl_sock_destroy(ys); From 5aadc155849eb85d799604939fd9e9024b7392a3 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Thu, 20 Nov 2025 20:02:59 -0800 Subject: [PATCH 1072/1381] selftests: hw-net: toeplitz: give the test up to 4 seconds Increase the receiver timeout. When running between machines in different geographic regions the test needs more than a second to SSH across and send the frames. The bkg() command that runs the receiver defaults to 5 sec timeout, so using 4 sec sounds like a reasonable value for the receiver itself. Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20251121040259.3647749-6-kuba@kernel.org Signed-off-by: Jakub Kicinski --- tools/testing/selftests/drivers/net/hw/toeplitz.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/testing/selftests/drivers/net/hw/toeplitz.py b/tools/testing/selftests/drivers/net/hw/toeplitz.py index 945c58d23310..d2db5ee9e358 100755 --- a/tools/testing/selftests/drivers/net/hw/toeplitz.py +++ b/tools/testing/selftests/drivers/net/hw/toeplitz.py @@ -166,7 +166,7 @@ def test(cfg, proto_flag, ipver, grp): proto_flag, "-d", str(port), "-i", cfg.ifname, - "-T", "1000", + "-T", "4000", "-s", "-v" ] From 00f3b3251814bebe17e1f52da09d0da165be07e8 Mon Sep 17 00:00:00 2001 From: Andre Carvalho Date: Fri, 21 Nov 2025 15:00:22 +0000 Subject: [PATCH 1073/1381] selftests: netconsole: ensure required log level is set on netcons_basic This commit ensures that the required log level is set at the start of the test iteration. Part of the cleanup performed at the end of each test iteration resets the log level (do_cleanup in lib_netcons.sh) to the values defined at the time test script started. This may cause further test iterations to fail if the default values are not sufficient. Signed-off-by: Andre Carvalho Reviewed-by: Breno Leitao Link: https://patch.msgid.link/20251121-netcons-basic-loglevel-v1-1-577f8586159c@gmail.com Signed-off-by: Jakub Kicinski --- tools/testing/selftests/drivers/net/netcons_basic.sh | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/tools/testing/selftests/drivers/net/netcons_basic.sh b/tools/testing/selftests/drivers/net/netcons_basic.sh index a3446b569976..2022f3061738 100755 --- a/tools/testing/selftests/drivers/net/netcons_basic.sh +++ b/tools/testing/selftests/drivers/net/netcons_basic.sh @@ -28,8 +28,6 @@ OUTPUT_FILE="/tmp/${TARGET}" # Check for basic system dependency and exit if not found check_for_dependencies -# Set current loglevel to KERN_INFO(6), and default to KERN_NOTICE(5) -echo "6 5" > /proc/sys/kernel/printk # Remove the namespace, interfaces and netconsole target on exit trap cleanup EXIT @@ -39,6 +37,9 @@ do for IP_VERSION in "ipv6" "ipv4" do echo "Running with target mode: ${FORMAT} (${IP_VERSION})" + # Set current loglevel to KERN_INFO(6), and default to + # KERN_NOTICE(5) + echo "6 5" > /proc/sys/kernel/printk # Create one namespace and two interfaces set_network "${IP_VERSION}" # Create a dynamic target for netconsole From e254c212cd9cabb9469a1282c4c91ccc6180bf51 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Sat, 22 Nov 2025 18:16:01 -0800 Subject: [PATCH 1074/1381] selftests: af_unix: don't use SKIP for expected failures netdev CI reserves SKIP in selftests for cases which can't be executed due to setup issues, like missing or old commands. Tests which are expected to fail must use XFAIL. Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20251123021601.158709-1-kuba@kernel.org Signed-off-by: Jakub Kicinski --- tools/testing/selftests/net/af_unix/unix_connreset.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/tools/testing/selftests/net/af_unix/unix_connreset.c b/tools/testing/selftests/net/af_unix/unix_connreset.c index bffef2b54bfd..6eb936207b31 100644 --- a/tools/testing/selftests/net/af_unix/unix_connreset.c +++ b/tools/testing/selftests/net/af_unix/unix_connreset.c @@ -161,8 +161,12 @@ TEST_F(unix_sock, reset_closed_embryo) char buf[16] = {}; ssize_t n; - if (variant->socket_type == SOCK_DGRAM) - SKIP(return, "This test only applies to SOCK_STREAM and SOCK_SEQPACKET"); + if (variant->socket_type == SOCK_DGRAM) { + snprintf(_metadata->results->reason, + sizeof(_metadata->results->reason), + "Test only applies to SOCK_STREAM and SOCK_SEQPACKET"); + exit(KSFT_XFAIL); + } /* Close server without accept()ing */ close(self->server); From ec1e48e97febbf7f87fc150793af2eda7223815c Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 21 Nov 2025 06:17:25 +0000 Subject: [PATCH 1075/1381] net: optimize eth_type_trans() vs CONFIG_STACKPROTECTOR_STRONG=y Some platforms exhibit very high costs with CONFIG_STACKPROTECTOR_STRONG=y when a function needs to pass the address of a local variable to external functions. eth_type_trans() (and its callers) is showing this anomaly on AMD EPYC 7B12 platforms (and maybe others). We could : 1) inline eth_type_trans() This would help if its callers also has the same issue, and the canary cost would be paid by the callers already. This is a bit cumbersome because netdev_uses_dsa() is pulling whole definitions. 2) Compile net/ethernet/eth.c with -fno-stack-protector This would weaken security. 3) Hack eth_type_trans() to temporarily use skb->dev as a place holder if skb_header_pointer() needs to pull 2 bytes not present in skb->head. This patch implements 3), and brings a 5% improvement on TX/RX intensive workload (tcp_rr 10,000 flows) on AMD EPYC 7B12. Removing CONFIG_STACKPROTECTOR_STRONG on this platform can improve performance by 25 %. This means eth_type_trans() issue is not an isolated artifact. Signed-off-by: Eric Dumazet Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20251121061725.206675-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ethernet/eth.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/net/ethernet/eth.c b/net/ethernet/eth.c index 43e211e611b1..13a63b48b7ee 100644 --- a/net/ethernet/eth.c +++ b/net/ethernet/eth.c @@ -154,9 +154,9 @@ EXPORT_SYMBOL(eth_get_headlen); */ __be16 eth_type_trans(struct sk_buff *skb, struct net_device *dev) { - unsigned short _service_access_point; const unsigned short *sap; const struct ethhdr *eth; + __be16 res; skb->dev = dev; skb_reset_mac_header(skb); @@ -181,15 +181,15 @@ __be16 eth_type_trans(struct sk_buff *skb, struct net_device *dev) * the protocol design and runs IPX over 802.3 without an 802.2 LLC * layer. We look for FFFF which isn't a used 802.2 SSAP/DSAP. This * won't work for fault tolerant netware but does for the rest. + * We use skb->dev as temporary storage to not hit + * CONFIG_STACKPROTECTOR_STRONG=y costs on some platforms. */ - sap = skb_header_pointer(skb, 0, sizeof(*sap), &_service_access_point); - if (sap && *sap == 0xFFFF) - return htons(ETH_P_802_3); + sap = skb_header_pointer(skb, 0, sizeof(*sap), &skb->dev); + res = (sap && *sap == 0xFFFF) ? htons(ETH_P_802_3) : htons(ETH_P_802_2); - /* - * Real 802.2 LLC - */ - return htons(ETH_P_802_2); + /* restore skb->dev in case it was mangled by skb_header_pointer(). */ + skb->dev = dev; + return res; } EXPORT_SYMBOL(eth_type_trans); From a8ff4842da50b5f4ea03d5d38a2ee93b701fe8e6 Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Fri, 21 Nov 2025 01:59:23 -0800 Subject: [PATCH 1076/1381] net: hyperv: convert to use .get_rx_ring_count Convert the hyperv netvsc driver to use the new .get_rx_ring_count ethtool operation instead of implementing .get_rxnfc solely for handling ETHTOOL_GRXRINGS command. This simplifies the code by replacing the switch statement with a direct return of the queue count. The new callback provides the same functionality in a more direct way, following the ongoing ethtool API modernization. Signed-off-by: Breno Leitao Link: https://patch.msgid.link/20251121-hyperv_gxrings-v1-1-31293104953b@debian.org Signed-off-by: Jakub Kicinski --- drivers/net/hyperv/netvsc_drv.c | 15 ++++----------- 1 file changed, 4 insertions(+), 11 deletions(-) diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv/netvsc_drv.c index 39c892e46cb0..3d47d749ef9f 100644 --- a/drivers/net/hyperv/netvsc_drv.c +++ b/drivers/net/hyperv/netvsc_drv.c @@ -1624,22 +1624,15 @@ netvsc_get_rxfh_fields(struct net_device *ndev, return 0; } -static int -netvsc_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *info, - u32 *rules) +static u32 netvsc_get_rx_ring_count(struct net_device *dev) { struct net_device_context *ndc = netdev_priv(dev); struct netvsc_device *nvdev = rtnl_dereference(ndc->nvdev); if (!nvdev) - return -ENODEV; - - switch (info->cmd) { - case ETHTOOL_GRXRINGS: - info->data = nvdev->num_chn; return 0; - } - return -EOPNOTSUPP; + + return nvdev->num_chn; } static int @@ -1969,7 +1962,7 @@ static const struct ethtool_ops ethtool_ops = { .get_channels = netvsc_get_channels, .set_channels = netvsc_set_channels, .get_ts_info = ethtool_op_get_ts_info, - .get_rxnfc = netvsc_get_rxnfc, + .get_rx_ring_count = netvsc_get_rx_ring_count, .get_rxfh_key_size = netvsc_get_rxfh_key_size, .get_rxfh_indir_size = netvsc_rss_indir_size, .get_rxfh = netvsc_get_rxfh, From 737e14c5dce3764125bb56de1b343f0e4d21492d Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Fri, 21 Nov 2025 09:02:35 -0800 Subject: [PATCH 1077/1381] net: mvneta: convert to use .get_rx_ring_count Convert the mvneta driver to use the new .get_rx_ring_count ethtool operation instead of implementing .get_rxnfc solely for handling ETHTOOL_GRXRINGS command. This simplifies the code by removing the switch statement and replacing it with a direct return of the queue count. The new callback provides the same functionality in a more direct way, following the ongoing ethtool API modernization. Signed-off-by: Breno Leitao Link: https://patch.msgid.link/20251121-marvell-v1-1-8338f3e55a4c@debian.org Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/marvell/mvneta.c | 14 +++----------- 1 file changed, 3 insertions(+), 11 deletions(-) diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c index 89ccb8eb82c7..7af44f858fa3 100644 --- a/drivers/net/ethernet/marvell/mvneta.c +++ b/drivers/net/ethernet/marvell/mvneta.c @@ -5012,17 +5012,9 @@ static u32 mvneta_ethtool_get_rxfh_indir_size(struct net_device *dev) return MVNETA_RSS_LU_TABLE_SIZE; } -static int mvneta_ethtool_get_rxnfc(struct net_device *dev, - struct ethtool_rxnfc *info, - u32 *rules __always_unused) +static u32 mvneta_ethtool_get_rx_ring_count(struct net_device *dev) { - switch (info->cmd) { - case ETHTOOL_GRXRINGS: - info->data = rxq_number; - return 0; - default: - return -EOPNOTSUPP; - } + return rxq_number; } static int mvneta_config_rss(struct mvneta_port *pp) @@ -5356,7 +5348,7 @@ static const struct ethtool_ops mvneta_eth_tool_ops = { .get_ethtool_stats = mvneta_ethtool_get_stats, .get_sset_count = mvneta_ethtool_get_sset_count, .get_rxfh_indir_size = mvneta_ethtool_get_rxfh_indir_size, - .get_rxnfc = mvneta_ethtool_get_rxnfc, + .get_rx_ring_count = mvneta_ethtool_get_rx_ring_count, .get_rxfh = mvneta_ethtool_get_rxfh, .set_rxfh = mvneta_ethtool_set_rxfh, .get_link_ksettings = mvneta_ethtool_get_link_ksettings, From 20c20f05cf50719a667f1a1747a0267fa09addc4 Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Fri, 21 Nov 2025 09:02:36 -0800 Subject: [PATCH 1078/1381] net: mvpp2: extract GRXRINGS from .get_rxnfc Commit 84eaf4359c36 ("net: ethtool: add get_rx_ring_count callback to optimize RX ring queries") added specific support for GRXRINGS callback, simplifying .get_rxnfc. Remove the handling of GRXRINGS in .get_rxnfc() by moving it to the new .get_rx_ring_count() for the mvpp2 driver. This simplifies the RX ring count retrieval and aligns mvpp2 with the new ethtool API for querying RX ring parameters, while keeping the other rxnfc handlers (GRXCLSRLCNT, GRXCLSRULE, GRXCLSRLALL) intact. Signed-off-by: Breno Leitao Link: https://patch.msgid.link/20251121-marvell-v1-2-8338f3e55a4c@debian.org Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c b/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c index ab0c99aa9f9a..33426fded919 100644 --- a/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c +++ b/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c @@ -5580,6 +5580,13 @@ static int mvpp2_ethtool_set_link_ksettings(struct net_device *dev, return phylink_ethtool_ksettings_set(port->phylink, cmd); } +static u32 mvpp2_ethtool_get_rx_ring_count(struct net_device *dev) +{ + struct mvpp2_port *port = netdev_priv(dev); + + return port->nrxqs; +} + static int mvpp2_ethtool_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *info, u32 *rules) { @@ -5590,9 +5597,6 @@ static int mvpp2_ethtool_get_rxnfc(struct net_device *dev, return -EOPNOTSUPP; switch (info->cmd) { - case ETHTOOL_GRXRINGS: - info->data = port->nrxqs; - break; case ETHTOOL_GRXCLSRLCNT: info->rule_cnt = port->n_rfs_rules; break; @@ -5827,6 +5831,7 @@ static const struct ethtool_ops mvpp2_eth_tool_ops = { .set_pauseparam = mvpp2_ethtool_set_pause_param, .get_link_ksettings = mvpp2_ethtool_get_link_ksettings, .set_link_ksettings = mvpp2_ethtool_set_link_ksettings, + .get_rx_ring_count = mvpp2_ethtool_get_rx_ring_count, .get_rxnfc = mvpp2_ethtool_get_rxnfc, .set_rxnfc = mvpp2_ethtool_set_rxnfc, .get_rxfh_indir_size = mvpp2_ethtool_get_rxfh_indir_size, From f296b73d17a486fb547caa9c7718401735c8c99c Mon Sep 17 00:00:00 2001 From: Dmitry Skorodumov Date: Fri, 21 Nov 2025 18:51:08 +0300 Subject: [PATCH 1079/1381] ipvlan: fix sparse warning about __be32 -> u32 Fixed a sparse warning: ipvlan_core.c:56: warning: incorrect type in argument 1 (different base types) expected unsigned int [usertype] a got restricted __be32 const [usertype] s_addr Force cast the s_addr to u32 Signed-off-by: Dmitry Skorodumov Link: https://patch.msgid.link/20251121155112.4182007-1-skorodumov.dmitry@huawei.com Signed-off-by: Jakub Kicinski --- drivers/net/ipvlan/ipvlan_core.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ipvlan/ipvlan_core.c b/drivers/net/ipvlan/ipvlan_core.c index d7e3ddbcab6f..dea411e132db 100644 --- a/drivers/net/ipvlan/ipvlan_core.c +++ b/drivers/net/ipvlan/ipvlan_core.c @@ -52,8 +52,8 @@ static u8 ipvlan_get_v4_hash(const void *iaddr) { const struct in_addr *ip4_addr = iaddr; - return jhash_1word(ip4_addr->s_addr, ipvlan_jhash_secret) & - IPVLAN_HASH_MASK; + return jhash_1word((__force u32)ip4_addr->s_addr, ipvlan_jhash_secret) & + IPVLAN_HASH_MASK; } static bool addr_equal(bool is_v6, struct ipvl_addr *addr, const void *iaddr) From 075b19c211dfeea5f27075293ddf8795b78c9bd9 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Fri, 21 Nov 2025 18:02:00 +0100 Subject: [PATCH 1080/1381] net: factor-out _sk_charge() helper Move out of __inet_accept() the code dealing charging newly accepted socket to memcg. MPTCP will soon use it to on a per subflow basis, in different contexts. No functional changes intended. Signed-off-by: Paolo Abeni Acked-by: Geliang Tang Acked-by: Matthieu Baerts (NGI0) Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20251121-net-next-mptcp-memcg-backlog-imp-v1-1-1f34b6c1e0b1@kernel.org Signed-off-by: Jakub Kicinski --- include/net/sock.h | 2 ++ net/core/sock.c | 18 ++++++++++++++++++ net/ipv4/af_inet.c | 17 +---------------- 3 files changed, 21 insertions(+), 16 deletions(-) diff --git a/include/net/sock.h b/include/net/sock.h index a5f36ea9d46f..38d48cfe0741 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -1631,6 +1631,8 @@ static inline void sk_mem_uncharge(struct sock *sk, int size) sk_mem_reclaim(sk); } +void __sk_charge(struct sock *sk, gfp_t gfp); + #if IS_ENABLED(CONFIG_PROVE_LOCKING) && IS_ENABLED(CONFIG_MODULES) static inline void sk_owner_set(struct sock *sk, struct module *owner) { diff --git a/net/core/sock.c b/net/core/sock.c index 3b74fc71f51c..b26a6cdc9bcd 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -3448,6 +3448,24 @@ void __sk_mem_reclaim(struct sock *sk, int amount) } EXPORT_SYMBOL(__sk_mem_reclaim); +void __sk_charge(struct sock *sk, gfp_t gfp) +{ + int amt; + + gfp |= __GFP_NOFAIL; + if (mem_cgroup_from_sk(sk)) { + /* The socket has not been accepted yet, no need + * to look at newsk->sk_wmem_queued. + */ + amt = sk_mem_pages(sk->sk_forward_alloc + + atomic_read(&sk->sk_rmem_alloc)); + if (amt) + mem_cgroup_sk_charge(sk, amt, gfp); + } + + kmem_cache_charge(sk, gfp); +} + int sk_set_peek_off(struct sock *sk, int val) { WRITE_ONCE(sk->sk_peek_off, val); diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index a31b94ce8968..08d811f11896 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -756,23 +756,8 @@ EXPORT_SYMBOL(inet_stream_connect); void __inet_accept(struct socket *sock, struct socket *newsock, struct sock *newsk) { if (mem_cgroup_sockets_enabled) { - gfp_t gfp = GFP_KERNEL | __GFP_NOFAIL; - mem_cgroup_sk_alloc(newsk); - - if (mem_cgroup_from_sk(newsk)) { - int amt; - - /* The socket has not been accepted yet, no need - * to look at newsk->sk_wmem_queued. - */ - amt = sk_mem_pages(newsk->sk_forward_alloc + - atomic_read(&newsk->sk_rmem_alloc)); - if (amt) - mem_cgroup_sk_charge(newsk, amt, gfp); - } - - kmem_cache_charge(newsk, gfp); + __sk_charge(newsk, GFP_KERNEL); } sock_rps_record_flow(newsk); From bd92dd8e03d98a499c8525e3748f046aaca2090d Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Fri, 21 Nov 2025 18:02:01 +0100 Subject: [PATCH 1081/1381] mptcp: factor-out cgroup data inherit helper MPTCP will soon need the same functionality for passive sockets, factor them out in a common helper. No functional change intended. Signed-off-by: Paolo Abeni Reviewed-by: Geliang Tang Reviewed-by: Matthieu Baerts (NGI0) Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20251121-net-next-mptcp-memcg-backlog-imp-v1-2-1f34b6c1e0b1@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/protocol.h | 2 ++ net/mptcp/subflow.c | 20 ++++++++++++-------- 2 files changed, 14 insertions(+), 8 deletions(-) diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h index a23780ff670f..6d9de13c1f9c 100644 --- a/net/mptcp/protocol.h +++ b/net/mptcp/protocol.h @@ -707,6 +707,8 @@ mptcp_subflow_delegated_next(struct mptcp_delegated_action *delegated) return ret; } +void __mptcp_inherit_cgrp_data(struct sock *sk, struct sock *ssk); + int mptcp_is_enabled(const struct net *net); unsigned int mptcp_get_add_addr_timeout(const struct net *net); int mptcp_is_checksum_enabled(const struct net *net); diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c index ddd0fc6fcf45..d98d151392d2 100644 --- a/net/mptcp/subflow.c +++ b/net/mptcp/subflow.c @@ -1712,21 +1712,25 @@ int __mptcp_subflow_connect(struct sock *sk, const struct mptcp_pm_local *local, return err; } -static void mptcp_attach_cgroup(struct sock *parent, struct sock *child) +void __mptcp_inherit_cgrp_data(struct sock *sk, struct sock *ssk) { #ifdef CONFIG_SOCK_CGROUP_DATA - struct sock_cgroup_data *parent_skcd = &parent->sk_cgrp_data, - *child_skcd = &child->sk_cgrp_data; + struct sock_cgroup_data *sk_cd = &sk->sk_cgrp_data, + *ssk_cd = &ssk->sk_cgrp_data; /* only the additional subflows created by kworkers have to be modified */ - if (cgroup_id(sock_cgroup_ptr(parent_skcd)) != - cgroup_id(sock_cgroup_ptr(child_skcd))) { - cgroup_sk_free(child_skcd); - *child_skcd = *parent_skcd; - cgroup_sk_clone(child_skcd); + if (cgroup_id(sock_cgroup_ptr(sk_cd)) != + cgroup_id(sock_cgroup_ptr(ssk_cd))) { + cgroup_sk_free(ssk_cd); + *ssk_cd = *sk_cd; + cgroup_sk_clone(sk_cd); } #endif /* CONFIG_SOCK_CGROUP_DATA */ +} +static void mptcp_attach_cgroup(struct sock *parent, struct sock *child) +{ + __mptcp_inherit_cgrp_data(parent, child); if (mem_cgroup_sockets_enabled) mem_cgroup_sk_inherit(parent, child); } From e777a7fb06b17cd62f6935313eef0206174ed71d Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Fri, 21 Nov 2025 18:02:02 +0100 Subject: [PATCH 1082/1381] mptcp: grafting MPJ subflow earlier Later patches need to ensure that all MPJ subflows are grafted to the msk socket before accept() completion. Currently the grafting happens under the msk socket lock: potentially at msk release_cb time which make satisfying the above condition a bit tricky. Move the MPJ subflow grafting earlier, under the msk data lock, so that we can use such lock as a synchronization point. Signed-off-by: Paolo Abeni Reviewed-by: Matthieu Baerts (NGI0) Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20251121-net-next-mptcp-memcg-backlog-imp-v1-3-1f34b6c1e0b1@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/protocol.c | 30 +++++++++++++++++++++++------- 1 file changed, 23 insertions(+), 7 deletions(-) diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index 75bb1199bed9..2104ab1eda1d 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -895,12 +895,6 @@ static bool __mptcp_finish_join(struct mptcp_sock *msk, struct sock *ssk) mptcp_subflow_joined(msk, ssk); spin_unlock_bh(&msk->fallback_lock); - /* attach to msk socket only after we are sure we will deal with it - * at close time - */ - if (sk->sk_socket && !ssk->sk_socket) - mptcp_sock_graft(ssk, sk->sk_socket); - mptcp_subflow_ctx(ssk)->subflow_id = msk->subflow_id++; mptcp_sockopt_sync_locked(msk, ssk); mptcp_stop_tout_timer(sk); @@ -3647,6 +3641,20 @@ void mptcp_sock_graft(struct sock *sk, struct socket *parent) write_unlock_bh(&sk->sk_callback_lock); } +/* Can be called without holding the msk socket lock; use the callback lock + * to avoid {READ_,WRITE_}ONCE annotations on sk_socket. + */ +static void mptcp_sock_check_graft(struct sock *sk, struct sock *ssk) +{ + struct socket *sock; + + write_lock_bh(&sk->sk_callback_lock); + sock = sk->sk_socket; + write_unlock_bh(&sk->sk_callback_lock); + if (sock) + mptcp_sock_graft(ssk, sock); +} + bool mptcp_finish_join(struct sock *ssk) { struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(ssk); @@ -3662,7 +3670,9 @@ bool mptcp_finish_join(struct sock *ssk) return false; } - /* active subflow, already present inside the conn_list */ + /* Active subflow, already present inside the conn_list; is grafted + * either by __mptcp_subflow_connect() or accept. + */ if (!list_empty(&subflow->node)) { spin_lock_bh(&msk->fallback_lock); if (!msk->allow_subflows) { @@ -3689,11 +3699,17 @@ bool mptcp_finish_join(struct sock *ssk) if (ret) { sock_hold(ssk); list_add_tail(&subflow->node, &msk->conn_list); + mptcp_sock_check_graft(parent, ssk); } } else { sock_hold(ssk); list_add_tail(&subflow->node, &msk->join_list); __set_bit(MPTCP_FLUSH_JOIN_LIST, &msk->cb_flags); + + /* In case of later failures, __mptcp_flush_join_list() will + * properly orphan the ssk via mptcp_close_ssk(). + */ + mptcp_sock_check_graft(parent, ssk); } mptcp_data_unlock(parent); From 68c7c386714512219e3517ec1097d164aee102fb Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Fri, 21 Nov 2025 18:02:03 +0100 Subject: [PATCH 1083/1381] mptcp: fix memcg accounting for passive sockets The passive sockets never got proper memcg accounting: the msk socket is associated with the memcg at accept time, but the passive subflows never got it right. At accept time, traverse the subflows list and associate each of them with the msk memcg, and try to do the same at join completion time, if the msk has been already accepted. Fixes: cf7da0d66cc1 ("mptcp: Create SUBFLOW socket for incoming connections") Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/298 Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/597 Signed-off-by: Paolo Abeni Reviewed-by: Matthieu Baerts (NGI0) Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20251121-net-next-mptcp-memcg-backlog-imp-v1-4-1f34b6c1e0b1@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/protocol.c | 38 +++++++++++++++++++++++++++----------- net/mptcp/protocol.h | 1 + net/mptcp/subflow.c | 10 ++++++++++ 3 files changed, 38 insertions(+), 11 deletions(-) diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index 2104ab1eda1d..67732d3c502c 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -3651,8 +3651,11 @@ static void mptcp_sock_check_graft(struct sock *sk, struct sock *ssk) write_lock_bh(&sk->sk_callback_lock); sock = sk->sk_socket; write_unlock_bh(&sk->sk_callback_lock); - if (sock) + if (sock) { mptcp_sock_graft(ssk, sock); + __mptcp_inherit_cgrp_data(sk, ssk); + __mptcp_inherit_memcg(sk, ssk, GFP_ATOMIC); + } } bool mptcp_finish_join(struct sock *ssk) @@ -3970,6 +3973,28 @@ static int mptcp_listen(struct socket *sock, int backlog) return err; } +static void mptcp_graft_subflows(struct sock *sk) +{ + struct mptcp_subflow_context *subflow; + struct mptcp_sock *msk = mptcp_sk(sk); + + mptcp_for_each_subflow(msk, subflow) { + struct sock *ssk = mptcp_subflow_tcp_sock(subflow); + + lock_sock(ssk); + + /* Set ssk->sk_socket of accept()ed flows to mptcp socket. + * This is needed so NOSPACE flag can be set from tcp stack. + */ + if (!ssk->sk_socket) + mptcp_sock_graft(ssk, sk->sk_socket); + + __mptcp_inherit_cgrp_data(sk, ssk); + __mptcp_inherit_memcg(sk, ssk, GFP_KERNEL); + release_sock(ssk); + } +} + static int mptcp_stream_accept(struct socket *sock, struct socket *newsock, struct proto_accept_arg *arg) { @@ -4017,16 +4042,7 @@ static int mptcp_stream_accept(struct socket *sock, struct socket *newsock, msk = mptcp_sk(newsk); msk->in_accept_queue = 0; - /* set ssk->sk_socket of accept()ed flows to mptcp socket. - * This is needed so NOSPACE flag can be set from tcp stack. - */ - mptcp_for_each_subflow(msk, subflow) { - struct sock *ssk = mptcp_subflow_tcp_sock(subflow); - - if (!ssk->sk_socket) - mptcp_sock_graft(ssk, newsock); - } - + mptcp_graft_subflows(newsk); mptcp_rps_record_subflows(msk); /* Do late cleanup for the first subflow as necessary. Also diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h index 6d9de13c1f9c..8c27f4b1789f 100644 --- a/net/mptcp/protocol.h +++ b/net/mptcp/protocol.h @@ -707,6 +707,7 @@ mptcp_subflow_delegated_next(struct mptcp_delegated_action *delegated) return ret; } +void __mptcp_inherit_memcg(struct sock *sk, struct sock *ssk, gfp_t gfp); void __mptcp_inherit_cgrp_data(struct sock *sk, struct sock *ssk); int mptcp_is_enabled(const struct net *net); diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c index d98d151392d2..72b7efe388db 100644 --- a/net/mptcp/subflow.c +++ b/net/mptcp/subflow.c @@ -1712,6 +1712,16 @@ int __mptcp_subflow_connect(struct sock *sk, const struct mptcp_pm_local *local, return err; } +void __mptcp_inherit_memcg(struct sock *sk, struct sock *ssk, gfp_t gfp) +{ + /* Only if the msk has been accepted already (and not orphaned).*/ + if (!mem_cgroup_sockets_enabled || !sk->sk_socket) + return; + + mem_cgroup_sk_inherit(sk, ssk); + __sk_charge(ssk, gfp); +} + void __mptcp_inherit_cgrp_data(struct sock *sk, struct sock *ssk) { #ifdef CONFIG_SOCK_CGROUP_DATA From 85f22b8e1e9db00a7003bf3c0e0a60d19718e083 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Fri, 21 Nov 2025 18:02:04 +0100 Subject: [PATCH 1084/1381] mptcp: cleanup fallback data fin reception MPTCP currently generate a dummy data_fin for fallback socket when the fallback subflow has completed data reception using the current ack_seq. We are going to introduce backlog usage for the msk soon, even for fallback sockets: the ack_seq value will not match the most recent sequence number seen by the fallback subflow socket, as it will ignore data_seq sitting in the backlog. Instead use the last map sequence number to set the data_fin, as fallback (dummy) map sequences are always in sequence. Reviewed-by: Geliang Tang Tested-by: Geliang Tang Signed-off-by: Paolo Abeni Reviewed-by: Mat Martineau Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20251121-net-next-mptcp-memcg-backlog-imp-v1-5-1f34b6c1e0b1@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/subflow.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c index 72b7efe388db..1f7311afd48d 100644 --- a/net/mptcp/subflow.c +++ b/net/mptcp/subflow.c @@ -1285,6 +1285,7 @@ static bool subflow_is_done(const struct sock *sk) /* sched mptcp worker for subflow cleanup if no more data is pending */ static void subflow_sched_work_if_closed(struct mptcp_sock *msk, struct sock *ssk) { + const struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(ssk); struct sock *sk = (struct sock *)msk; if (likely(ssk->sk_state != TCP_CLOSE && @@ -1303,7 +1304,8 @@ static void subflow_sched_work_if_closed(struct mptcp_sock *msk, struct sock *ss */ if (__mptcp_check_fallback(msk) && subflow_is_done(ssk) && msk->first == ssk && - mptcp_update_rcv_data_fin(msk, READ_ONCE(msk->ack_seq), true)) + mptcp_update_rcv_data_fin(msk, subflow->map_seq + + subflow->map_data_len, true)) mptcp_schedule_work(sk); } From 2834f8edd74d5dda368087a654c0e52b141e9893 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Fri, 21 Nov 2025 18:02:05 +0100 Subject: [PATCH 1085/1381] mptcp: cleanup fallback dummy mapping generation MPTCP currently access ack_seq outside the msk socket log scope to generate the dummy mapping for fallback socket. Soon we are going to introduce backlog usage and even for fallback socket the ack_seq value will be significantly off outside of the msk socket lock scope. Avoid relying on ack_seq for dummy mapping generation, using instead the subflow sequence number. Note that in case of disconnect() and (re)connect() we must ensure that any previous state is re-set. Signed-off-by: Paolo Abeni Reviewed-by: Geliang Tang Tested-by: Geliang Tang Reviewed-by: Mat Martineau Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20251121-net-next-mptcp-memcg-backlog-imp-v1-6-1f34b6c1e0b1@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/protocol.c | 3 +++ net/mptcp/subflow.c | 8 +++++++- 2 files changed, 10 insertions(+), 1 deletion(-) diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index 67732d3c502c..df4be41ed3fe 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -3274,6 +3274,9 @@ static int mptcp_disconnect(struct sock *sk, int flags) msk->bytes_retrans = 0; msk->rcvspace_init = 0; + /* for fallback's sake */ + WRITE_ONCE(msk->ack_seq, 0); + WRITE_ONCE(sk->sk_shutdown, 0); sk_error_report(sk); return 0; diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c index 1f7311afd48d..86ce58ae533d 100644 --- a/net/mptcp/subflow.c +++ b/net/mptcp/subflow.c @@ -491,6 +491,9 @@ static void subflow_set_remote_key(struct mptcp_sock *msk, mptcp_crypto_key_sha(subflow->remote_key, NULL, &subflow->iasn); subflow->iasn++; + /* for fallback's sake */ + subflow->map_seq = subflow->iasn; + WRITE_ONCE(msk->remote_key, subflow->remote_key); WRITE_ONCE(msk->ack_seq, subflow->iasn); WRITE_ONCE(msk->can_ack, true); @@ -1435,9 +1438,12 @@ static bool subflow_check_data_avail(struct sock *ssk) skb = skb_peek(&ssk->sk_receive_queue); subflow->map_valid = 1; - subflow->map_seq = READ_ONCE(msk->ack_seq); subflow->map_data_len = skb->len; subflow->map_subflow_seq = tcp_sk(ssk)->copied_seq - subflow->ssn_offset; + subflow->map_seq = __mptcp_expand_seq(subflow->map_seq, + subflow->iasn + + TCP_SKB_CB(skb)->seq - + subflow->ssn_offset - 1); WRITE_ONCE(subflow->data_avail, true); return true; } From 2ca1b8926fda1468513bb3b8731422d5b2aa0ab1 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Fri, 21 Nov 2025 18:02:06 +0100 Subject: [PATCH 1086/1381] mptcp: ensure the kernel PM does not take action too late The PM hooks can currently take place when the msk is already shutting down. Subflow creation will fail, thanks to the existing check at join time, but we can entirely avoid starting the to be failed operations. Signed-off-by: Paolo Abeni Reviewed-by: Geliang Tang Tested-by: Geliang Tang Reviewed-by: Mat Martineau Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20251121-net-next-mptcp-memcg-backlog-imp-v1-7-1f34b6c1e0b1@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/pm.c | 4 +++- net/mptcp/pm_kernel.c | 2 ++ 2 files changed, 5 insertions(+), 1 deletion(-) diff --git a/net/mptcp/pm.c b/net/mptcp/pm.c index 9604b91902b8..e2040c327af6 100644 --- a/net/mptcp/pm.c +++ b/net/mptcp/pm.c @@ -594,6 +594,7 @@ void mptcp_pm_subflow_established(struct mptcp_sock *msk) void mptcp_pm_subflow_check_next(struct mptcp_sock *msk, const struct mptcp_subflow_context *subflow) { + struct sock *sk = (struct sock *)msk; struct mptcp_pm_data *pm = &msk->pm; bool update_subflows; @@ -617,7 +618,8 @@ void mptcp_pm_subflow_check_next(struct mptcp_sock *msk, /* Even if this subflow is not really established, tell the PM to try * to pick the next ones, if possible. */ - if (mptcp_pm_nl_check_work_pending(msk)) + if (mptcp_is_fully_established(sk) && + mptcp_pm_nl_check_work_pending(msk)) mptcp_pm_schedule_work(msk, MPTCP_PM_SUBFLOW_ESTABLISHED); spin_unlock_bh(&pm->lock); diff --git a/net/mptcp/pm_kernel.c b/net/mptcp/pm_kernel.c index 5c1dc13efa94..57570a44e418 100644 --- a/net/mptcp/pm_kernel.c +++ b/net/mptcp/pm_kernel.c @@ -337,6 +337,8 @@ static void mptcp_pm_create_subflow_or_signal_addr(struct mptcp_sock *msk) struct mptcp_pm_local local; mptcp_mpc_endpoint_setup(msk); + if (!mptcp_is_fully_established(sk)) + return; pr_debug("local %d:%d signal %d:%d subflows %d:%d\n", msk->pm.local_addr_used, endp_subflow_max, From 48a395605e088e1365bcc0b94e53746537c21969 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Fri, 21 Nov 2025 18:02:07 +0100 Subject: [PATCH 1087/1381] mptcp: do not miss early first subflow close event notification The MPTCP protocol is not currently emitting the NL event when the first subflow is closed before msk accept() time. By replacing the in use close helper is such scenario, implicitly introduce the missing notification. Note that in such scenario we want to be sure that mptcp_close_ssk() will not trigger any PM work, move the msk state change update earlier, so that the previous patch will offer such guarantee. Signed-off-by: Paolo Abeni Reviewed-by: Geliang Tang Tested-by: Geliang Tang Reviewed-by: Mat Martineau Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20251121-net-next-mptcp-memcg-backlog-imp-v1-8-1f34b6c1e0b1@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/protocol.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index df4be41ed3fe..2ee76c8c5167 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -4052,10 +4052,10 @@ static int mptcp_stream_accept(struct socket *sock, struct socket *newsock, * deal with bad peers not doing a complete shutdown. */ if (unlikely(inet_sk_state_load(msk->first) == TCP_CLOSE)) { - __mptcp_close_ssk(newsk, msk->first, - mptcp_subflow_ctx(msk->first), 0); if (unlikely(list_is_singular(&msk->conn_list))) mptcp_set_state(newsk, TCP_CLOSE); + mptcp_close_ssk(newsk, msk->first, + mptcp_subflow_ctx(msk->first)); } } else { tcpfallback: From 9d8295960300788b64d7027ff07ad8c15503121a Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Fri, 21 Nov 2025 18:02:08 +0100 Subject: [PATCH 1088/1381] mptcp: make mptcp_destroy_common() static Such function is only used inside protocol.c, there is no need to expose it to the whole stack. Note that the function definition most be moved earlier to avoid forward declaration. Signed-off-by: Paolo Abeni Reviewed-by: Geliang Tang Tested-by: Geliang Tang Reviewed-by: Mat Martineau Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20251121-net-next-mptcp-memcg-backlog-imp-v1-9-1f34b6c1e0b1@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/protocol.c | 42 +++++++++++++++++++++--------------------- net/mptcp/protocol.h | 2 -- 2 files changed, 21 insertions(+), 23 deletions(-) diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index 2ee76c8c5167..29e5bda0e913 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -3222,6 +3222,27 @@ static void mptcp_copy_inaddrs(struct sock *msk, const struct sock *ssk) inet_sk(msk)->inet_rcv_saddr = inet_sk(ssk)->inet_rcv_saddr; } +static void mptcp_destroy_common(struct mptcp_sock *msk) +{ + struct mptcp_subflow_context *subflow, *tmp; + struct sock *sk = (struct sock *)msk; + + __mptcp_clear_xmit(sk); + + /* join list will be eventually flushed (with rst) at sock lock release time */ + mptcp_for_each_subflow_safe(msk, subflow, tmp) + __mptcp_close_ssk(sk, mptcp_subflow_tcp_sock(subflow), subflow, 0); + + __skb_queue_purge(&sk->sk_receive_queue); + skb_rbtree_purge(&msk->out_of_order_queue); + + /* move all the rx fwd alloc into the sk_mem_reclaim_final in + * inet_sock_destruct() will dispose it + */ + mptcp_token_destroy(msk); + mptcp_pm_destroy(msk); +} + static int mptcp_disconnect(struct sock *sk, int flags) { struct mptcp_sock *msk = mptcp_sk(sk); @@ -3427,27 +3448,6 @@ void mptcp_rcv_space_init(struct mptcp_sock *msk, const struct sock *ssk) msk->rcvq_space.space = TCP_INIT_CWND * TCP_MSS_DEFAULT; } -void mptcp_destroy_common(struct mptcp_sock *msk) -{ - struct mptcp_subflow_context *subflow, *tmp; - struct sock *sk = (struct sock *)msk; - - __mptcp_clear_xmit(sk); - - /* join list will be eventually flushed (with rst) at sock lock release time */ - mptcp_for_each_subflow_safe(msk, subflow, tmp) - __mptcp_close_ssk(sk, mptcp_subflow_tcp_sock(subflow), subflow, 0); - - __skb_queue_purge(&sk->sk_receive_queue); - skb_rbtree_purge(&msk->out_of_order_queue); - - /* move all the rx fwd alloc into the sk_mem_reclaim_final in - * inet_sock_destruct() will dispose it - */ - mptcp_token_destroy(msk); - mptcp_pm_destroy(msk); -} - static void mptcp_destroy(struct sock *sk) { struct mptcp_sock *msk = mptcp_sk(sk); diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h index 8c27f4b1789f..3d2892cc0ef2 100644 --- a/net/mptcp/protocol.h +++ b/net/mptcp/protocol.h @@ -980,8 +980,6 @@ static inline void mptcp_propagate_sndbuf(struct sock *sk, struct sock *ssk) local_bh_enable(); } -void mptcp_destroy_common(struct mptcp_sock *msk); - #define MPTCP_TOKEN_MAX_RETRIES 4 void __init mptcp_token_init(void); From 38a4a469c850fc007d6fe2429b1f7f492e50e7ad Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Fri, 21 Nov 2025 18:02:09 +0100 Subject: [PATCH 1089/1381] mptcp: drop the __mptcp_data_ready() helper It adds little clarity and there is a single user of such helper, just inline it in the caller. Signed-off-by: Paolo Abeni Reviewed-by: Geliang Tang Tested-by: Geliang Tang Reviewed-by: Mat Martineau Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20251121-net-next-mptcp-memcg-backlog-imp-v1-10-1f34b6c1e0b1@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/protocol.c | 19 +++++++------------ 1 file changed, 7 insertions(+), 12 deletions(-) diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index 29e5bda0e913..ba1237853ebf 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -845,18 +845,10 @@ static bool move_skbs_to_msk(struct mptcp_sock *msk, struct sock *ssk) return moved; } -static void __mptcp_data_ready(struct sock *sk, struct sock *ssk) -{ - struct mptcp_sock *msk = mptcp_sk(sk); - - /* Wake-up the reader only for in-sequence data */ - if (move_skbs_to_msk(msk, ssk) && mptcp_epollin_ready(sk)) - sk->sk_data_ready(sk); -} - void mptcp_data_ready(struct sock *sk, struct sock *ssk) { struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(ssk); + struct mptcp_sock *msk = mptcp_sk(sk); /* The peer can send data while we are shutting down this * subflow at msk destruction time, but we must avoid enqueuing @@ -866,10 +858,13 @@ void mptcp_data_ready(struct sock *sk, struct sock *ssk) return; mptcp_data_lock(sk); - if (!sock_owned_by_user(sk)) - __mptcp_data_ready(sk, ssk); - else + if (!sock_owned_by_user(sk)) { + /* Wake-up the reader only for in-sequence data */ + if (move_skbs_to_msk(msk, ssk) && mptcp_epollin_ready(sk)) + sk->sk_data_ready(sk); + } else { __set_bit(MPTCP_DEQUEUE, &mptcp_sk(sk)->cb_flags); + } mptcp_data_unlock(sk); } From 0eeb372deebce6c25b9afc09e35d6c75a744299a Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Fri, 21 Nov 2025 18:02:10 +0100 Subject: [PATCH 1090/1381] mptcp: handle first subflow closing consistently Currently, as soon as the PM closes a subflow, the msk stops accepting data from it, even if the TCP socket could be still formally open in the incoming direction, with the notable exception of the first subflow. The root cause of such behavior is that code currently piggy back two separate semantic on the subflow->disposable bit: the subflow context must be released and that the subflow must stop accepting incoming data. The first subflow is never disposed, so it also never stop accepting incoming data. Use a separate bit to mark the latter status and set such bit in __mptcp_close_ssk() for all subflows. Beyond making per subflow behaviour more consistent this will also simplify the next patch. Signed-off-by: Paolo Abeni Reviewed-by: Mat Martineau Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20251121-net-next-mptcp-memcg-backlog-imp-v1-11-1f34b6c1e0b1@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/protocol.c | 14 +++++++++----- net/mptcp/protocol.h | 3 ++- 2 files changed, 11 insertions(+), 6 deletions(-) diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index ba1237853ebf..d22f792f4760 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -851,10 +851,10 @@ void mptcp_data_ready(struct sock *sk, struct sock *ssk) struct mptcp_sock *msk = mptcp_sk(sk); /* The peer can send data while we are shutting down this - * subflow at msk destruction time, but we must avoid enqueuing + * subflow at subflow destruction time, but we must avoid enqueuing * more data to the msk receive queue */ - if (unlikely(subflow->disposable)) + if (unlikely(subflow->closing)) return; mptcp_data_lock(sk); @@ -2437,6 +2437,13 @@ static void __mptcp_close_ssk(struct sock *sk, struct sock *ssk, struct mptcp_sock *msk = mptcp_sk(sk); bool dispose_it, need_push = false; + /* Do not pass RX data to the msk, even if the subflow socket is not + * going to be freed (i.e. even for the first subflow on graceful + * subflow close. + */ + lock_sock_nested(ssk, SINGLE_DEPTH_NESTING); + subflow->closing = 1; + /* If the first subflow moved to a close state before accept, e.g. due * to an incoming reset or listener shutdown, the subflow socket is * already deleted by inet_child_forget() and the mptcp socket can't @@ -2447,7 +2454,6 @@ static void __mptcp_close_ssk(struct sock *sk, struct sock *ssk, /* ensure later check in mptcp_worker() will dispose the msk */ sock_set_flag(sk, SOCK_DEAD); mptcp_set_close_tout(sk, tcp_jiffies32 - (mptcp_close_timeout(sk) + 1)); - lock_sock_nested(ssk, SINGLE_DEPTH_NESTING); mptcp_subflow_drop_ctx(ssk); goto out_release; } @@ -2456,8 +2462,6 @@ static void __mptcp_close_ssk(struct sock *sk, struct sock *ssk, if (dispose_it) list_del(&subflow->node); - lock_sock_nested(ssk, SINGLE_DEPTH_NESTING); - if (subflow->send_fastclose && ssk->sk_state != TCP_CLOSE) tcp_set_state(ssk, TCP_CLOSE); diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h index 3d2892cc0ef2..d30806b287d2 100644 --- a/net/mptcp/protocol.h +++ b/net/mptcp/protocol.h @@ -536,12 +536,13 @@ struct mptcp_subflow_context { send_infinite_map : 1, remote_key_valid : 1, /* received the peer key from */ disposable : 1, /* ctx can be free at ulp release time */ + closing : 1, /* must not pass rx data to msk anymore */ stale : 1, /* unable to snd/rcv data, do not use for xmit */ valid_csum_seen : 1, /* at least one csum validated */ is_mptfo : 1, /* subflow is doing TFO */ close_event_done : 1, /* has done the post-closed part */ mpc_drop : 1, /* the MPC option has been dropped in a rtx */ - __unused : 9; + __unused : 8; bool data_avail; bool scheduled; bool pm_listener; /* a listener managed by the kernel PM? */ From 9db5b3cec4ec1c0cd3239689f5c8653d691a1754 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Fri, 21 Nov 2025 18:02:11 +0100 Subject: [PATCH 1091/1381] mptcp: borrow forward memory from subflow In the MPTCP receive path, we release the subflow allocated fwd memory just to allocate it again shortly after for the msk. That could increases the failures chances, especially when we will add backlog processing, with other actions could consume the just released memory before the msk socket has a chance to do the rcv allocation. Replace the skb_orphan() call with an open-coded variant that explicitly borrows, the fwd memory from the subflow socket instead of releasing it. The borrowed memory does not have PAGE_SIZE granularity; rounding to the page size will make the fwd allocated memory higher than what is strictly required and could make the incoming subflow fwd mem consistently negative. Instead, keep track of the accumulated frag and borrow the full page at subflow close time. This allow removing the last drop in the TCP to MPTCP transition and the associated, now unused, MIB. Signed-off-by: Paolo Abeni Reviewed-by: Mat Martineau Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20251121-net-next-mptcp-memcg-backlog-imp-v1-12-1f34b6c1e0b1@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/fastopen.c | 4 +++- net/mptcp/mib.c | 1 - net/mptcp/mib.h | 1 - net/mptcp/protocol.c | 23 +++++++++++++++-------- net/mptcp/protocol.h | 28 ++++++++++++++++++++++++++++ 5 files changed, 46 insertions(+), 11 deletions(-) diff --git a/net/mptcp/fastopen.c b/net/mptcp/fastopen.c index b9e451197902..82ec15bcfd7f 100644 --- a/net/mptcp/fastopen.c +++ b/net/mptcp/fastopen.c @@ -32,7 +32,8 @@ void mptcp_fastopen_subflow_synack_set_params(struct mptcp_subflow_context *subf /* dequeue the skb from sk receive queue */ __skb_unlink(skb, &ssk->sk_receive_queue); skb_ext_reset(skb); - skb_orphan(skb); + + mptcp_subflow_lend_fwdmem(subflow, skb); /* We copy the fastopen data, but that don't belong to the mptcp sequence * space, need to offset it in the subflow sequence, see mptcp_subflow_get_map_offset() @@ -50,6 +51,7 @@ void mptcp_fastopen_subflow_synack_set_params(struct mptcp_subflow_context *subf mptcp_data_lock(sk); DEBUG_NET_WARN_ON_ONCE(sock_owned_by_user_nocheck(sk)); + mptcp_borrow_fwdmem(sk, skb); skb_set_owner_r(skb, sk); __skb_queue_tail(&sk->sk_receive_queue, skb); mptcp_sk(sk)->bytes_received += skb->len; diff --git a/net/mptcp/mib.c b/net/mptcp/mib.c index 171643815076..f23fda0c55a7 100644 --- a/net/mptcp/mib.c +++ b/net/mptcp/mib.c @@ -71,7 +71,6 @@ static const struct snmp_mib mptcp_snmp_list[] = { SNMP_MIB_ITEM("MPFastcloseRx", MPTCP_MIB_MPFASTCLOSERX), SNMP_MIB_ITEM("MPRstTx", MPTCP_MIB_MPRSTTX), SNMP_MIB_ITEM("MPRstRx", MPTCP_MIB_MPRSTRX), - SNMP_MIB_ITEM("RcvPruned", MPTCP_MIB_RCVPRUNED), SNMP_MIB_ITEM("SubflowStale", MPTCP_MIB_SUBFLOWSTALE), SNMP_MIB_ITEM("SubflowRecover", MPTCP_MIB_SUBFLOWRECOVER), SNMP_MIB_ITEM("SndWndShared", MPTCP_MIB_SNDWNDSHARED), diff --git a/net/mptcp/mib.h b/net/mptcp/mib.h index a1d3e9369fbb..812218b5ed2b 100644 --- a/net/mptcp/mib.h +++ b/net/mptcp/mib.h @@ -70,7 +70,6 @@ enum linux_mptcp_mib_field { MPTCP_MIB_MPFASTCLOSERX, /* Received a MP_FASTCLOSE */ MPTCP_MIB_MPRSTTX, /* Transmit a MP_RST */ MPTCP_MIB_MPRSTRX, /* Received a MP_RST */ - MPTCP_MIB_RCVPRUNED, /* Incoming packet dropped due to memory limit */ MPTCP_MIB_SUBFLOWSTALE, /* Subflows entered 'stale' status */ MPTCP_MIB_SUBFLOWRECOVER, /* Subflows returned to active status after being stale */ MPTCP_MIB_SNDWNDSHARED, /* Subflow snd wnd is overridden by msk's one */ diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index d22f792f4760..f5526855a2e5 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -358,7 +358,7 @@ static void mptcp_data_queue_ofo(struct mptcp_sock *msk, struct sk_buff *skb) static void mptcp_init_skb(struct sock *ssk, struct sk_buff *skb, int offset, int copy_len) { - const struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(ssk); + struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(ssk); bool has_rxtstamp = TCP_SKB_CB(skb)->has_rxtstamp; /* the skb map_seq accounts for the skb offset: @@ -383,11 +383,7 @@ static bool __mptcp_move_skb(struct sock *sk, struct sk_buff *skb) struct mptcp_sock *msk = mptcp_sk(sk); struct sk_buff *tail; - /* try to fetch required memory from subflow */ - if (!sk_rmem_schedule(sk, skb, skb->truesize)) { - MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_RCVPRUNED); - goto drop; - } + mptcp_borrow_fwdmem(sk, skb); if (MPTCP_SKB_CB(skb)->map_seq == msk->ack_seq) { /* in sequence */ @@ -409,7 +405,6 @@ static bool __mptcp_move_skb(struct sock *sk, struct sk_buff *skb) * will retransmit as needed, if needed. */ MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_DUPDATA); -drop: mptcp_drop(sk, skb); return false; } @@ -710,7 +705,7 @@ static bool __mptcp_move_skbs_from_subflow(struct mptcp_sock *msk, size_t len = skb->len - offset; mptcp_init_skb(ssk, skb, offset, len); - skb_orphan(skb); + mptcp_subflow_lend_fwdmem(subflow, skb); ret = __mptcp_move_skb(sk, skb) || ret; seq += len; @@ -2436,6 +2431,7 @@ static void __mptcp_close_ssk(struct sock *sk, struct sock *ssk, { struct mptcp_sock *msk = mptcp_sk(sk); bool dispose_it, need_push = false; + int fwd_remaining; /* Do not pass RX data to the msk, even if the subflow socket is not * going to be freed (i.e. even for the first subflow on graceful @@ -2444,6 +2440,17 @@ static void __mptcp_close_ssk(struct sock *sk, struct sock *ssk, lock_sock_nested(ssk, SINGLE_DEPTH_NESTING); subflow->closing = 1; + /* Borrow the fwd allocated page left-over; fwd memory for the subflow + * could be negative at this point, but will be reach zero soon - when + * the data allocated using such fragment will be freed. + */ + if (subflow->lent_mem_frag) { + fwd_remaining = PAGE_SIZE - subflow->lent_mem_frag; + sk_forward_alloc_add(sk, fwd_remaining); + sk_forward_alloc_add(ssk, -fwd_remaining); + subflow->lent_mem_frag = 0; + } + /* If the first subflow moved to a close state before accept, e.g. due * to an incoming reset or listener shutdown, the subflow socket is * already deleted by inet_child_forget() and the mptcp socket can't diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h index d30806b287d2..5e2749d92a49 100644 --- a/net/mptcp/protocol.h +++ b/net/mptcp/protocol.h @@ -547,6 +547,7 @@ struct mptcp_subflow_context { bool scheduled; bool pm_listener; /* a listener managed by the kernel PM? */ bool fully_established; /* path validated */ + u32 lent_mem_frag; u32 remote_nonce; u64 thmac; u32 local_nonce; @@ -646,6 +647,33 @@ mptcp_send_active_reset_reason(struct sock *sk) tcp_send_active_reset(sk, GFP_ATOMIC, reason); } +/* Made the fwd mem carried by the given skb available to the msk, + * To be paired with a previous mptcp_subflow_lend_fwdmem() before freeing + * the skb or setting the skb ownership. + */ +static inline void mptcp_borrow_fwdmem(struct sock *sk, struct sk_buff *skb) +{ + struct sock *ssk = skb->sk; + + /* The subflow just lend the skb fwd memory, and we know that the skb + * is only accounted on the incoming subflow rcvbuf. + */ + DEBUG_NET_WARN_ON_ONCE(skb->destructor); + skb->sk = NULL; + sk_forward_alloc_add(sk, skb->truesize); + atomic_sub(skb->truesize, &ssk->sk_rmem_alloc); +} + +static inline void +mptcp_subflow_lend_fwdmem(struct mptcp_subflow_context *subflow, + struct sk_buff *skb) +{ + int frag = (subflow->lent_mem_frag + skb->truesize) & (PAGE_SIZE - 1); + + skb->destructor = NULL; + subflow->lent_mem_frag = frag; +} + static inline u64 mptcp_subflow_get_map_offset(const struct mptcp_subflow_context *subflow) { From ee458a3f314e9c669ddd227bf5ab08354d9e75cc Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Fri, 21 Nov 2025 18:02:12 +0100 Subject: [PATCH 1092/1381] mptcp: introduce mptcp-level backlog We are soon using it for incoming data processing. MPTCP can't leverage the sk_backlog, as the latter is processed before the release callback, and such callback for MPTCP releases and re-acquire the socket spinlock, breaking the sk_backlog processing assumption. Add a skb backlog list inside the mptcp sock struct, and implement basic helper to transfer packet to and purge such list. Packets in the backlog are memory accounted and still use the incoming subflow receive memory, to allow back-pressure. The backlog size is implicitly bounded to the sum of subflows rcvbuf. When a subflow is closed, references from the backlog to such sock are removed. No packet is currently added to the backlog, so no functional changes intended here. Signed-off-by: Paolo Abeni Reviewed-by: Mat Martineau Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20251121-net-next-mptcp-memcg-backlog-imp-v1-13-1f34b6c1e0b1@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/mptcp_diag.c | 3 +- net/mptcp/protocol.c | 78 ++++++++++++++++++++++++++++++++++++++++-- net/mptcp/protocol.h | 25 ++++++++++---- 3 files changed, 97 insertions(+), 9 deletions(-) diff --git a/net/mptcp/mptcp_diag.c b/net/mptcp/mptcp_diag.c index ac974299de71..136c2d05c0ee 100644 --- a/net/mptcp/mptcp_diag.c +++ b/net/mptcp/mptcp_diag.c @@ -195,7 +195,8 @@ static void mptcp_diag_get_info(struct sock *sk, struct inet_diag_msg *r, struct mptcp_sock *msk = mptcp_sk(sk); struct mptcp_info *info = _info; - r->idiag_rqueue = sk_rmem_alloc_get(sk); + r->idiag_rqueue = sk_rmem_alloc_get(sk) + + READ_ONCE(mptcp_sk(sk)->backlog_len); r->idiag_wqueue = sk_wmem_alloc_get(sk); if (inet_sk_state_load(sk) == TCP_LISTEN) { diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index f5526855a2e5..dfed036e0591 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -659,6 +659,39 @@ static void mptcp_dss_corruption(struct mptcp_sock *msk, struct sock *ssk) } } +static void __mptcp_add_backlog(struct sock *sk, + struct mptcp_subflow_context *subflow, + struct sk_buff *skb) +{ + struct mptcp_sock *msk = mptcp_sk(sk); + struct sk_buff *tail = NULL; + bool fragstolen; + int delta; + + if (unlikely(sk->sk_state == TCP_CLOSE)) { + kfree_skb_reason(skb, SKB_DROP_REASON_SOCKET_CLOSE); + return; + } + + /* Try to coalesce with the last skb in our backlog */ + if (!list_empty(&msk->backlog_list)) + tail = list_last_entry(&msk->backlog_list, struct sk_buff, list); + + if (tail && MPTCP_SKB_CB(skb)->map_seq == MPTCP_SKB_CB(tail)->end_seq && + skb->sk == tail->sk && + __mptcp_try_coalesce(sk, tail, skb, &fragstolen, &delta)) { + skb->truesize -= delta; + kfree_skb_partial(skb, fragstolen); + __mptcp_subflow_lend_fwdmem(subflow, delta); + WRITE_ONCE(msk->backlog_len, msk->backlog_len + delta); + return; + } + + list_add_tail(&skb->list, &msk->backlog_list); + mptcp_subflow_lend_fwdmem(subflow, skb); + WRITE_ONCE(msk->backlog_len, msk->backlog_len + skb->truesize); +} + static bool __mptcp_move_skbs_from_subflow(struct mptcp_sock *msk, struct sock *ssk) { @@ -705,8 +738,13 @@ static bool __mptcp_move_skbs_from_subflow(struct mptcp_sock *msk, size_t len = skb->len - offset; mptcp_init_skb(ssk, skb, offset, len); - mptcp_subflow_lend_fwdmem(subflow, skb); - ret = __mptcp_move_skb(sk, skb) || ret; + + if (true) { + mptcp_subflow_lend_fwdmem(subflow, skb); + ret |= __mptcp_move_skb(sk, skb); + } else { + __mptcp_add_backlog(sk, subflow, skb); + } seq += len; if (unlikely(map_remaining < len)) { @@ -2531,6 +2569,9 @@ static void __mptcp_close_ssk(struct sock *sk, struct sock *ssk, void mptcp_close_ssk(struct sock *sk, struct sock *ssk, struct mptcp_subflow_context *subflow) { + struct mptcp_sock *msk = mptcp_sk(sk); + struct sk_buff *skb; + /* The first subflow can already be closed and still in the list */ if (subflow->close_event_done) return; @@ -2540,6 +2581,17 @@ void mptcp_close_ssk(struct sock *sk, struct sock *ssk, if (sk->sk_state == TCP_ESTABLISHED) mptcp_event(MPTCP_EVENT_SUB_CLOSED, mptcp_sk(sk), ssk, GFP_KERNEL); + /* Remove any reference from the backlog to this ssk; backlog skbs consume + * space in the msk receive queue, no need to touch sk->sk_rmem_alloc + */ + list_for_each_entry(skb, &msk->backlog_list, list) { + if (skb->sk != ssk) + continue; + + atomic_sub(skb->truesize, &skb->sk->sk_rmem_alloc); + skb->sk = NULL; + } + /* subflow aborted before reaching the fully_established status * attempt the creation of the next subflow */ @@ -2769,12 +2821,31 @@ static void mptcp_mp_fail_no_response(struct mptcp_sock *msk) unlock_sock_fast(ssk, slow); } +static void mptcp_backlog_purge(struct sock *sk) +{ + struct mptcp_sock *msk = mptcp_sk(sk); + struct sk_buff *tmp, *skb; + LIST_HEAD(backlog); + + mptcp_data_lock(sk); + list_splice_init(&msk->backlog_list, &backlog); + msk->backlog_len = 0; + mptcp_data_unlock(sk); + + list_for_each_entry_safe(skb, tmp, &backlog, list) { + mptcp_borrow_fwdmem(sk, skb); + kfree_skb_reason(skb, SKB_DROP_REASON_SOCKET_CLOSE); + } + sk_mem_reclaim(sk); +} + static void mptcp_do_fastclose(struct sock *sk) { struct mptcp_subflow_context *subflow, *tmp; struct mptcp_sock *msk = mptcp_sk(sk); mptcp_set_state(sk, TCP_CLOSE); + mptcp_backlog_purge(sk); /* Explicitly send the fastclose reset as need */ if (__mptcp_check_fallback(msk)) @@ -2853,11 +2924,13 @@ static void __mptcp_init_sock(struct sock *sk) INIT_LIST_HEAD(&msk->conn_list); INIT_LIST_HEAD(&msk->join_list); INIT_LIST_HEAD(&msk->rtx_queue); + INIT_LIST_HEAD(&msk->backlog_list); INIT_WORK(&msk->work, mptcp_worker); msk->out_of_order_queue = RB_ROOT; msk->first_pending = NULL; msk->timer_ival = TCP_RTO_MIN; msk->scaling_ratio = TCP_DEFAULT_SCALING_RATIO; + msk->backlog_len = 0; WRITE_ONCE(msk->first, NULL); inet_csk(sk)->icsk_sync_mss = mptcp_sync_mss; @@ -3234,6 +3307,7 @@ static void mptcp_destroy_common(struct mptcp_sock *msk) struct sock *sk = (struct sock *)msk; __mptcp_clear_xmit(sk); + mptcp_backlog_purge(sk); /* join list will be eventually flushed (with rst) at sock lock release time */ mptcp_for_each_subflow_safe(msk, subflow, tmp) diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h index 5e2749d92a49..fe0dca4122f2 100644 --- a/net/mptcp/protocol.h +++ b/net/mptcp/protocol.h @@ -357,6 +357,9 @@ struct mptcp_sock { * allow_infinite_fallback and * allow_join */ + + struct list_head backlog_list; /* protected by the data lock */ + u32 backlog_len; }; #define mptcp_data_lock(sk) spin_lock_bh(&(sk)->sk_lock.slock) @@ -407,6 +410,7 @@ static inline int mptcp_space_from_win(const struct sock *sk, int win) static inline int __mptcp_space(const struct sock *sk) { return mptcp_win_from_space(sk, READ_ONCE(sk->sk_rcvbuf) - + READ_ONCE(mptcp_sk(sk)->backlog_len) - sk_rmem_alloc_get(sk)); } @@ -655,23 +659,32 @@ static inline void mptcp_borrow_fwdmem(struct sock *sk, struct sk_buff *skb) { struct sock *ssk = skb->sk; - /* The subflow just lend the skb fwd memory, and we know that the skb - * is only accounted on the incoming subflow rcvbuf. + /* The subflow just lend the skb fwd memory; if the subflow meanwhile + * closed, mptcp_close_ssk() already released the ssk rcv memory. */ DEBUG_NET_WARN_ON_ONCE(skb->destructor); - skb->sk = NULL; sk_forward_alloc_add(sk, skb->truesize); + if (!ssk) + return; + atomic_sub(skb->truesize, &ssk->sk_rmem_alloc); + skb->sk = NULL; +} + +static inline void +__mptcp_subflow_lend_fwdmem(struct mptcp_subflow_context *subflow, int size) +{ + int frag = (subflow->lent_mem_frag + size) & (PAGE_SIZE - 1); + + subflow->lent_mem_frag = frag; } static inline void mptcp_subflow_lend_fwdmem(struct mptcp_subflow_context *subflow, struct sk_buff *skb) { - int frag = (subflow->lent_mem_frag + skb->truesize) & (PAGE_SIZE - 1); - + __mptcp_subflow_lend_fwdmem(subflow, skb->truesize); skb->destructor = NULL; - subflow->lent_mem_frag = frag; } static inline u64 From 6228efe0cc01d155a99fdb95891b0e20d14e7a85 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Fri, 21 Nov 2025 18:02:13 +0100 Subject: [PATCH 1093/1381] mptcp: leverage the backlog for RX packet processing When the msk socket is owned or the msk receive buffer is full, move the incoming skbs in a msk level backlog list. This avoid traversing the joined subflows and acquiring the subflow level socket lock at reception time, improving the RX performances. When processing the backlog, use the fwd alloc memory borrowed from the incoming subflow. skbs exceeding the msk receive space are not dropped; instead they are kept into the backlog until the receive buffer is freed. Dropping packets already acked at the TCP level is explicitly discouraged by the RFC and would corrupt the data stream for fallback sockets. Special care is needed to avoid adding skbs to the backlog of a closed msk and to avoid leaving dangling references into the backlog at subflow closing time. Signed-off-by: Paolo Abeni Reviewed-by: Mat Martineau Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20251121-net-next-mptcp-memcg-backlog-imp-v1-14-1f34b6c1e0b1@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/protocol.c | 191 +++++++++++++++++++++++++++++-------------- net/mptcp/protocol.h | 2 +- 2 files changed, 132 insertions(+), 61 deletions(-) diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index dfed036e0591..e4ccc57b6f57 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -665,6 +665,7 @@ static void __mptcp_add_backlog(struct sock *sk, { struct mptcp_sock *msk = mptcp_sk(sk); struct sk_buff *tail = NULL; + struct sock *ssk = skb->sk; bool fragstolen; int delta; @@ -678,22 +679,30 @@ static void __mptcp_add_backlog(struct sock *sk, tail = list_last_entry(&msk->backlog_list, struct sk_buff, list); if (tail && MPTCP_SKB_CB(skb)->map_seq == MPTCP_SKB_CB(tail)->end_seq && - skb->sk == tail->sk && + ssk == tail->sk && __mptcp_try_coalesce(sk, tail, skb, &fragstolen, &delta)) { skb->truesize -= delta; kfree_skb_partial(skb, fragstolen); __mptcp_subflow_lend_fwdmem(subflow, delta); - WRITE_ONCE(msk->backlog_len, msk->backlog_len + delta); - return; + goto account; } list_add_tail(&skb->list, &msk->backlog_list); mptcp_subflow_lend_fwdmem(subflow, skb); - WRITE_ONCE(msk->backlog_len, msk->backlog_len + skb->truesize); + delta = skb->truesize; + +account: + WRITE_ONCE(msk->backlog_len, msk->backlog_len + delta); + + /* Possibly not accept()ed yet, keep track of memory not CG + * accounted, mptcp_graft_subflows() will handle it. + */ + if (!mem_cgroup_from_sk(ssk)) + msk->backlog_unaccounted += delta; } static bool __mptcp_move_skbs_from_subflow(struct mptcp_sock *msk, - struct sock *ssk) + struct sock *ssk, bool own_msk) { struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(ssk); struct sock *sk = (struct sock *)msk; @@ -709,9 +718,6 @@ static bool __mptcp_move_skbs_from_subflow(struct mptcp_sock *msk, struct sk_buff *skb; bool fin; - if (sk_rmem_alloc_get(sk) > sk->sk_rcvbuf) - break; - /* try to move as much data as available */ map_remaining = subflow->map_data_len - mptcp_subflow_get_map_offset(subflow); @@ -739,7 +745,7 @@ static bool __mptcp_move_skbs_from_subflow(struct mptcp_sock *msk, mptcp_init_skb(ssk, skb, offset, len); - if (true) { + if (own_msk && sk_rmem_alloc_get(sk) < sk->sk_rcvbuf) { mptcp_subflow_lend_fwdmem(subflow, skb); ret |= __mptcp_move_skb(sk, skb); } else { @@ -863,7 +869,7 @@ static bool move_skbs_to_msk(struct mptcp_sock *msk, struct sock *ssk) struct sock *sk = (struct sock *)msk; bool moved; - moved = __mptcp_move_skbs_from_subflow(msk, ssk); + moved = __mptcp_move_skbs_from_subflow(msk, ssk, true); __mptcp_ofo_queue(msk); if (unlikely(ssk->sk_err)) __mptcp_subflow_error_report(sk, ssk); @@ -896,7 +902,7 @@ void mptcp_data_ready(struct sock *sk, struct sock *ssk) if (move_skbs_to_msk(msk, ssk) && mptcp_epollin_ready(sk)) sk->sk_data_ready(sk); } else { - __set_bit(MPTCP_DEQUEUE, &mptcp_sk(sk)->cb_flags); + __mptcp_move_skbs_from_subflow(msk, ssk, false); } mptcp_data_unlock(sk); } @@ -2136,60 +2142,80 @@ static void mptcp_rcv_space_adjust(struct mptcp_sock *msk, int copied) msk->rcvq_space.time = mstamp; } -static struct mptcp_subflow_context * -__mptcp_first_ready_from(struct mptcp_sock *msk, - struct mptcp_subflow_context *subflow) +static bool __mptcp_move_skbs(struct sock *sk, struct list_head *skbs, u32 *delta) { - struct mptcp_subflow_context *start_subflow = subflow; - - while (!READ_ONCE(subflow->data_avail)) { - subflow = mptcp_next_subflow(msk, subflow); - if (subflow == start_subflow) - return NULL; - } - return subflow; -} - -static bool __mptcp_move_skbs(struct sock *sk) -{ - struct mptcp_subflow_context *subflow; + struct sk_buff *skb = list_first_entry(skbs, struct sk_buff, list); struct mptcp_sock *msk = mptcp_sk(sk); - bool ret = false; + bool moved = false; - if (list_empty(&msk->conn_list)) - return false; - - subflow = list_first_entry(&msk->conn_list, - struct mptcp_subflow_context, node); - for (;;) { - struct sock *ssk; - bool slowpath; - - /* - * As an optimization avoid traversing the subflows list - * and ev. acquiring the subflow socket lock before baling out - */ + *delta = 0; + while (1) { + /* If the msk recvbuf is full stop, don't drop */ if (sk_rmem_alloc_get(sk) > sk->sk_rcvbuf) break; - subflow = __mptcp_first_ready_from(msk, subflow); - if (!subflow) + prefetch(skb->next); + list_del(&skb->list); + *delta += skb->truesize; + + moved |= __mptcp_move_skb(sk, skb); + if (list_empty(skbs)) break; - ssk = mptcp_subflow_tcp_sock(subflow); - slowpath = lock_sock_fast(ssk); - ret = __mptcp_move_skbs_from_subflow(msk, ssk) || ret; - if (unlikely(ssk->sk_err)) - __mptcp_error_report(sk); - unlock_sock_fast(ssk, slowpath); - - subflow = mptcp_next_subflow(msk, subflow); + skb = list_first_entry(skbs, struct sk_buff, list); } __mptcp_ofo_queue(msk); - if (ret) + if (moved) mptcp_check_data_fin((struct sock *)msk); - return ret; + return moved; +} + +static bool mptcp_can_spool_backlog(struct sock *sk, struct list_head *skbs) +{ + struct mptcp_sock *msk = mptcp_sk(sk); + + /* After CG initialization, subflows should never add skb before + * gaining the CG themself. + */ + DEBUG_NET_WARN_ON_ONCE(msk->backlog_unaccounted && sk->sk_socket && + mem_cgroup_from_sk(sk)); + + /* Don't spool the backlog if the rcvbuf is full. */ + if (list_empty(&msk->backlog_list) || + sk_rmem_alloc_get(sk) > sk->sk_rcvbuf) + return false; + + INIT_LIST_HEAD(skbs); + list_splice_init(&msk->backlog_list, skbs); + return true; +} + +static void mptcp_backlog_spooled(struct sock *sk, u32 moved, + struct list_head *skbs) +{ + struct mptcp_sock *msk = mptcp_sk(sk); + + WRITE_ONCE(msk->backlog_len, msk->backlog_len - moved); + list_splice(skbs, &msk->backlog_list); +} + +static bool mptcp_move_skbs(struct sock *sk) +{ + struct list_head skbs; + bool enqueued = false; + u32 moved; + + mptcp_data_lock(sk); + while (mptcp_can_spool_backlog(sk, &skbs)) { + mptcp_data_unlock(sk); + enqueued |= __mptcp_move_skbs(sk, &skbs, &moved); + + mptcp_data_lock(sk); + mptcp_backlog_spooled(sk, moved, &skbs); + } + mptcp_data_unlock(sk); + return enqueued; } static unsigned int mptcp_inq_hint(const struct sock *sk) @@ -2255,7 +2281,7 @@ static int mptcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, copied += bytes_read; - if (skb_queue_empty(&sk->sk_receive_queue) && __mptcp_move_skbs(sk)) + if (!list_empty(&msk->backlog_list) && mptcp_move_skbs(sk)) continue; /* only the MPTCP socket status is relevant here. The exit @@ -3556,8 +3582,7 @@ void __mptcp_check_push(struct sock *sk, struct sock *ssk) #define MPTCP_FLAGS_PROCESS_CTX_NEED (BIT(MPTCP_PUSH_PENDING) | \ BIT(MPTCP_RETRANSMIT) | \ - BIT(MPTCP_FLUSH_JOIN_LIST) | \ - BIT(MPTCP_DEQUEUE)) + BIT(MPTCP_FLUSH_JOIN_LIST)) /* processes deferred events and flush wmem */ static void mptcp_release_cb(struct sock *sk) @@ -3567,9 +3592,12 @@ static void mptcp_release_cb(struct sock *sk) for (;;) { unsigned long flags = (msk->cb_flags & MPTCP_FLAGS_PROCESS_CTX_NEED); - struct list_head join_list; + struct list_head join_list, skbs; + bool spool_bl; + u32 moved; - if (!flags) + spool_bl = mptcp_can_spool_backlog(sk, &skbs); + if (!flags && !spool_bl) break; INIT_LIST_HEAD(&join_list); @@ -3591,7 +3619,7 @@ static void mptcp_release_cb(struct sock *sk) __mptcp_push_pending(sk, 0); if (flags & BIT(MPTCP_RETRANSMIT)) __mptcp_retrans(sk); - if ((flags & BIT(MPTCP_DEQUEUE)) && __mptcp_move_skbs(sk)) { + if (spool_bl && __mptcp_move_skbs(sk, &skbs, &moved)) { /* notify ack seq update */ mptcp_cleanup_rbuf(msk, 0); sk->sk_data_ready(sk); @@ -3599,6 +3627,8 @@ static void mptcp_release_cb(struct sock *sk) cond_resched(); spin_lock_bh(&sk->sk_lock.slock); + if (spool_bl) + mptcp_backlog_spooled(sk, moved, &skbs); } if (__test_and_clear_bit(MPTCP_CLEAN_UNA, &msk->cb_flags)) @@ -3856,7 +3886,7 @@ static int mptcp_ioctl(struct sock *sk, int cmd, int *karg) return -EINVAL; lock_sock(sk); - if (__mptcp_move_skbs(sk)) + if (mptcp_move_skbs(sk)) mptcp_cleanup_rbuf(msk, 0); *karg = mptcp_inq_hint(sk); release_sock(sk); @@ -4061,6 +4091,22 @@ static void mptcp_graft_subflows(struct sock *sk) struct mptcp_subflow_context *subflow; struct mptcp_sock *msk = mptcp_sk(sk); + if (mem_cgroup_sockets_enabled) { + LIST_HEAD(join_list); + + /* Subflows joining after __inet_accept() will get the + * mem CG properly initialized at mptcp_finish_join() time, + * but subflows pending in join_list need explicit + * initialization before flushing `backlog_unaccounted` + * or MPTCP can later unexpectedly observe unaccounted memory. + */ + mptcp_data_lock(sk); + list_splice_init(&msk->join_list, &join_list); + mptcp_data_unlock(sk); + + __mptcp_flush_join_list(sk, &join_list); + } + mptcp_for_each_subflow(msk, subflow) { struct sock *ssk = mptcp_subflow_tcp_sock(subflow); @@ -4072,10 +4118,35 @@ static void mptcp_graft_subflows(struct sock *sk) if (!ssk->sk_socket) mptcp_sock_graft(ssk, sk->sk_socket); + if (!mem_cgroup_sk_enabled(sk)) + goto unlock; + __mptcp_inherit_cgrp_data(sk, ssk); __mptcp_inherit_memcg(sk, ssk, GFP_KERNEL); + +unlock: release_sock(ssk); } + + if (mem_cgroup_sk_enabled(sk)) { + gfp_t gfp = GFP_KERNEL | __GFP_NOFAIL; + int amt; + + /* Account the backlog memory; prior accept() is aware of + * fwd and rmem only. + */ + mptcp_data_lock(sk); + amt = sk_mem_pages(sk->sk_forward_alloc + + msk->backlog_unaccounted + + atomic_read(&sk->sk_rmem_alloc)) - + sk_mem_pages(sk->sk_forward_alloc + + atomic_read(&sk->sk_rmem_alloc)); + msk->backlog_unaccounted = 0; + mptcp_data_unlock(sk); + + if (amt) + mem_cgroup_sk_charge(sk, amt, gfp); + } } static int mptcp_stream_accept(struct socket *sock, struct socket *newsock, diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h index fe0dca4122f2..313da78e2b75 100644 --- a/net/mptcp/protocol.h +++ b/net/mptcp/protocol.h @@ -124,7 +124,6 @@ #define MPTCP_FLUSH_JOIN_LIST 5 #define MPTCP_SYNC_STATE 6 #define MPTCP_SYNC_SNDBUF 7 -#define MPTCP_DEQUEUE 8 struct mptcp_skb_cb { u64 map_seq; @@ -360,6 +359,7 @@ struct mptcp_sock { struct list_head backlog_list; /* protected by the data lock */ u32 backlog_len; + u32 backlog_unaccounted; }; #define mptcp_data_lock(sk) spin_lock_bh(&(sk)->sk_lock.slock) From 24d4da5c2565313c2ad3c43449937a9351a64407 Mon Sep 17 00:00:00 2001 From: Ria Thomas Date: Mon, 24 Nov 2025 18:26:37 +0530 Subject: [PATCH 1094/1381] wifi: ieee80211: correct FILS status codes The FILS status codes are set to 108/109, but the IEEE 802.11-2020 spec defines them as 112/113. Update the enum so it matches the specification and keeps the kernel consistent with standard values. Fixes: a3caf7440ded ("cfg80211: Add support for FILS shared key authentication offload") Signed-off-by: Ria Thomas Reviewed-by: Jeff Johnson Link: https://patch.msgid.link/20251124125637.3936154-1-ria.thomas@morsemicro.com Signed-off-by: Johannes Berg --- include/linux/ieee80211.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h index d55d8ea3a8be..96439de55f07 100644 --- a/include/linux/ieee80211.h +++ b/include/linux/ieee80211.h @@ -1493,8 +1493,8 @@ enum ieee80211_statuscode { WLAN_STATUS_DENIED_WITH_SUGGESTED_BAND_AND_CHANNEL = 99, WLAN_STATUS_DENIED_DUE_TO_SPECTRUM_MANAGEMENT = 103, /* 802.11ai */ - WLAN_STATUS_FILS_AUTHENTICATION_FAILURE = 108, - WLAN_STATUS_UNKNOWN_AUTHENTICATION_SERVER = 109, + WLAN_STATUS_FILS_AUTHENTICATION_FAILURE = 112, + WLAN_STATUS_UNKNOWN_AUTHENTICATION_SERVER = 113, WLAN_STATUS_SAE_HASH_TO_ELEMENT = 126, WLAN_STATUS_SAE_PK = 127, WLAN_STATUS_DENIED_TID_TO_LINK_MAPPING = 133, From 3fc830cd8c9d0e5efae64da38a1a5eac01584b2f Mon Sep 17 00:00:00 2001 From: Lachlan Hodges Date: Tue, 25 Nov 2025 13:59:25 +1100 Subject: [PATCH 1095/1381] wifi: cfg80211: include s1g_primary_2mhz when sending chandef The chandef now includes a flag denoting the use of a 2MHz primary channel for S1G interfaces, include this when sending the chandef. Signed-off-by: Lachlan Hodges Link: https://patch.msgid.link/20251125025927.245280-2-lachlan.hodges@morsemicro.com Signed-off-by: Johannes Berg --- net/wireless/nl80211.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index 29c92bc8291b..c961cd42a832 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -4178,6 +4178,9 @@ int nl80211_send_chandef(struct sk_buff *msg, const struct cfg80211_chan_def *ch if (chandef->punctured && nla_put_u32(msg, NL80211_ATTR_PUNCT_BITMAP, chandef->punctured)) return -ENOBUFS; + if (chandef->s1g_primary_2mhz && + nla_put_flag(msg, NL80211_ATTR_S1G_PRIMARY_2MHZ)) + return -ENOBUFS; return 0; } From cba1ba11c1bae87de9c2e13d342bfbd6a3c1cf63 Mon Sep 17 00:00:00 2001 From: Lachlan Hodges Date: Tue, 25 Nov 2025 13:59:26 +1100 Subject: [PATCH 1096/1381] wifi: cfg80211: include s1g_primary_2mhz when comparing chandefs When comparing chandefs, ensure we include s1g_primary_2mhz. Signed-off-by: Lachlan Hodges Link: https://patch.msgid.link/20251125025927.245280-3-lachlan.hodges@morsemicro.com Signed-off-by: Johannes Berg --- include/net/cfg80211.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index 3d3ed1932262..899f267b7cf9 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -974,7 +974,8 @@ cfg80211_chandef_identical(const struct cfg80211_chan_def *chandef1, chandef1->center_freq1 == chandef2->center_freq1 && chandef1->freq1_offset == chandef2->freq1_offset && chandef1->center_freq2 == chandef2->center_freq2 && - chandef1->punctured == chandef2->punctured); + chandef1->punctured == chandef2->punctured && + chandef1->s1g_primary_2mhz == chandef2->s1g_primary_2mhz); } /** From 1c6a92a5a5de7ebf94526dee7068926e6d5b1b01 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Mon, 24 Nov 2025 18:28:34 -0800 Subject: [PATCH 1097/1381] wifi: nl80211: vendor-cmd: intel: fix a blank kernel-doc line warning Delete an empty line prevent a kernel-doc warning: Warning: ../include/uapi/linux/nl80211-vnd-intel.h:86 bad line: Fixes: 3d2a2544eae9 ("nl80211: vendor-cmd: add Intel vendor commands for iwlmei usage") Signed-off-by: Randy Dunlap Link: https://patch.msgid.link/20251125022834.3171742-1-rdunlap@infradead.org Signed-off-by: Johannes Berg --- include/uapi/linux/nl80211-vnd-intel.h | 1 - 1 file changed, 1 deletion(-) diff --git a/include/uapi/linux/nl80211-vnd-intel.h b/include/uapi/linux/nl80211-vnd-intel.h index 4ed7d0b24512..79ccc9401d50 100644 --- a/include/uapi/linux/nl80211-vnd-intel.h +++ b/include/uapi/linux/nl80211-vnd-intel.h @@ -84,7 +84,6 @@ enum iwl_vendor_auth_akm_mode { * * @NUM_IWL_MVM_VENDOR_ATTR: number of vendor attributes * @MAX_IWL_MVM_VENDOR_ATTR: highest vendor attribute number - */ enum iwl_mvm_vendor_attr { __IWL_MVM_VENDOR_ATTR_INVALID = 0x00, From e3daf0e7fe9758613bec324fd606ed9caa187f74 Mon Sep 17 00:00:00 2001 From: Jacky Chou Date: Thu, 20 Nov 2025 11:52:03 +0800 Subject: [PATCH 1098/1381] dt-bindings: net: aspeed: add AST2700 MDIO compatible Add "aspeed,ast2700-mdio" compatible to the binding schema with a fallback to "aspeed,ast2600-mdio". Although the MDIO controller on AST2700 is functionally the same as the one on AST2600, it's good practice to add a SoC-specific compatible for new silicon. This allows future driver updates to handle any 2700-specific integration issues without requiring devicetree changes or complex runtime detection logic. For now, the driver continues to bind via the existing "aspeed,ast2600-mdio" compatible, so no driver changes are needed. Acked-by: Conor Dooley Signed-off-by: Jacky Chou Link: https://patch.msgid.link/20251120-aspeed_mdio_ast2700-v2-1-0d722bfb2c54@aspeedtech.com Signed-off-by: Paolo Abeni --- .../devicetree/bindings/net/aspeed,ast2600-mdio.yaml | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/Documentation/devicetree/bindings/net/aspeed,ast2600-mdio.yaml b/Documentation/devicetree/bindings/net/aspeed,ast2600-mdio.yaml index d6ef468495c5..a105dc07ed12 100644 --- a/Documentation/devicetree/bindings/net/aspeed,ast2600-mdio.yaml +++ b/Documentation/devicetree/bindings/net/aspeed,ast2600-mdio.yaml @@ -19,7 +19,12 @@ allOf: properties: compatible: - const: aspeed,ast2600-mdio + oneOf: + - const: aspeed,ast2600-mdio + - items: + - enum: + - aspeed,ast2700-mdio + - const: aspeed,ast2600-mdio reg: maxItems: 1 From b2a38f6df9dab0b05858746edcbe2403f8f4e4ec Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 21 Nov 2025 08:32:43 +0000 Subject: [PATCH 1099/1381] net_sched: make room for (struct qdisc_skb_cb)->pkt_segs Add a new u16 field, next to pkt_len : pkt_segs This will cache shinfo->gso_segs to speed up qdisc deqeue(). Move slave_dev_queue_mapping at the end of qdisc_skb_cb, and move three bits from tc_skb_cb : - post_ct - post_ct_snat - post_ct_dnat Signed-off-by: Eric Dumazet Link: https://patch.msgid.link/20251121083256.674562-2-edumazet@google.com Signed-off-by: Paolo Abeni --- include/net/sch_generic.h | 18 +++++++++--------- net/core/dev.c | 2 +- net/sched/act_ct.c | 8 ++++---- net/sched/cls_api.c | 6 +++--- net/sched/cls_flower.c | 2 +- 5 files changed, 18 insertions(+), 18 deletions(-) diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index 94966692ccdf..9cd8b5d4b236 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -429,13 +429,16 @@ struct tcf_proto { }; struct qdisc_skb_cb { - struct { - unsigned int pkt_len; - u16 slave_dev_queue_mapping; - u16 tc_classid; - }; + unsigned int pkt_len; + u16 pkt_segs; + u16 tc_classid; #define QDISC_CB_PRIV_LEN 20 unsigned char data[QDISC_CB_PRIV_LEN]; + + u16 slave_dev_queue_mapping; + u8 post_ct:1; + u8 post_ct_snat:1; + u8 post_ct_dnat:1; }; typedef void tcf_chain_head_change_t(struct tcf_proto *tp_head, void *priv); @@ -1064,11 +1067,8 @@ struct tc_skb_cb { struct qdisc_skb_cb qdisc_cb; u32 drop_reason; - u16 zone; /* Only valid if post_ct = true */ + u16 zone; /* Only valid if qdisc_skb_cb(skb)->post_ct = true */ u16 mru; - u8 post_ct:1; - u8 post_ct_snat:1; - u8 post_ct_dnat:1; }; static inline struct tc_skb_cb *tc_skb_cb(const struct sk_buff *skb) diff --git a/net/core/dev.c b/net/core/dev.c index 69515edd17bc..46ce6c610780 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4355,7 +4355,7 @@ static int tc_run(struct tcx_entry *entry, struct sk_buff *skb, return ret; tc_skb_cb(skb)->mru = 0; - tc_skb_cb(skb)->post_ct = false; + qdisc_skb_cb(skb)->post_ct = false; tcf_set_drop_reason(skb, *drop_reason); mini_qdisc_bstats_cpu_update(miniq, skb); diff --git a/net/sched/act_ct.c b/net/sched/act_ct.c index 6749a4a9a9cd..2b6ac7069dc1 100644 --- a/net/sched/act_ct.c +++ b/net/sched/act_ct.c @@ -948,9 +948,9 @@ static int tcf_ct_act_nat(struct sk_buff *skb, return err & NF_VERDICT_MASK; if (action & BIT(NF_NAT_MANIP_SRC)) - tc_skb_cb(skb)->post_ct_snat = 1; + qdisc_skb_cb(skb)->post_ct_snat = 1; if (action & BIT(NF_NAT_MANIP_DST)) - tc_skb_cb(skb)->post_ct_dnat = 1; + qdisc_skb_cb(skb)->post_ct_dnat = 1; return err; #else @@ -986,7 +986,7 @@ TC_INDIRECT_SCOPE int tcf_ct_act(struct sk_buff *skb, const struct tc_action *a, tcf_action_update_bstats(&c->common, skb); if (clear) { - tc_skb_cb(skb)->post_ct = false; + qdisc_skb_cb(skb)->post_ct = false; ct = nf_ct_get(skb, &ctinfo); if (ct) { nf_ct_put(ct); @@ -1097,7 +1097,7 @@ TC_INDIRECT_SCOPE int tcf_ct_act(struct sk_buff *skb, const struct tc_action *a, out_push: skb_push_rcsum(skb, nh_ofs); - tc_skb_cb(skb)->post_ct = true; + qdisc_skb_cb(skb)->post_ct = true; tc_skb_cb(skb)->zone = p->zone; out_clear: if (defrag) diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index f751cd5eeac8..ebca4b926dcf 100644 --- a/net/sched/cls_api.c +++ b/net/sched/cls_api.c @@ -1872,9 +1872,9 @@ int tcf_classify(struct sk_buff *skb, } ext->chain = last_executed_chain; ext->mru = cb->mru; - ext->post_ct = cb->post_ct; - ext->post_ct_snat = cb->post_ct_snat; - ext->post_ct_dnat = cb->post_ct_dnat; + ext->post_ct = qdisc_skb_cb(skb)->post_ct; + ext->post_ct_snat = qdisc_skb_cb(skb)->post_ct_snat; + ext->post_ct_dnat = qdisc_skb_cb(skb)->post_ct_dnat; ext->zone = cb->zone; } } diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c index 099ff6a3e1f5..7669371c1354 100644 --- a/net/sched/cls_flower.c +++ b/net/sched/cls_flower.c @@ -326,7 +326,7 @@ TC_INDIRECT_SCOPE int fl_classify(struct sk_buff *skb, struct tcf_result *res) { struct cls_fl_head *head = rcu_dereference_bh(tp->root); - bool post_ct = tc_skb_cb(skb)->post_ct; + bool post_ct = qdisc_skb_cb(skb)->post_ct; u16 zone = tc_skb_cb(skb)->zone; struct fl_flow_key skb_key; struct fl_flow_mask *mask; From be1b70ab21cbe8f8bb8ecf39eb34880882252773 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 21 Nov 2025 08:32:44 +0000 Subject: [PATCH 1100/1381] net: init shinfo->gso_segs from qdisc_pkt_len_init() Qdisc use shinfo->gso_segs for their pkts stats in bstats_update(), but this field needs to be initialized for SKB_GSO_DODGY users. Signed-off-by: Eric Dumazet Link: https://patch.msgid.link/20251121083256.674562-3-edumazet@google.com Signed-off-by: Paolo Abeni --- net/core/dev.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/core/dev.c b/net/core/dev.c index 46ce6c610780..dba9eef8bd83 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4071,7 +4071,7 @@ EXPORT_SYMBOL_GPL(validate_xmit_skb_list); static void qdisc_pkt_len_init(struct sk_buff *skb) { - const struct skb_shared_info *shinfo = skb_shinfo(skb); + struct skb_shared_info *shinfo = skb_shinfo(skb); qdisc_skb_cb(skb)->pkt_len = skb->len; @@ -4112,6 +4112,7 @@ static void qdisc_pkt_len_init(struct sk_buff *skb) if (payload <= 0) return; gso_segs = DIV_ROUND_UP(payload, shinfo->gso_size); + shinfo->gso_segs = gso_segs; } qdisc_skb_cb(skb)->pkt_len += (gso_segs - 1) * hdr_len; } From 874c1928d37297a02452a404c8b496aad9fee146 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 21 Nov 2025 08:32:45 +0000 Subject: [PATCH 1101/1381] net_sched: initialize qdisc_skb_cb(skb)->pkt_segs in qdisc_pkt_len_init() qdisc_pkt_len_init() is currently initalizing qdisc_skb_cb(skb)->pkt_len. Add qdisc_skb_cb(skb)->pkt_segs initialization and rename this function to qdisc_pkt_len_segs_init(). Signed-off-by: Eric Dumazet Link: https://patch.msgid.link/20251121083256.674562-4-edumazet@google.com Signed-off-by: Paolo Abeni --- net/core/dev.c | 15 +++++++++++---- net/sched/sch_cake.c | 2 +- 2 files changed, 12 insertions(+), 5 deletions(-) diff --git a/net/core/dev.c b/net/core/dev.c index dba9eef8bd83..895c3e37e686 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4069,17 +4069,23 @@ struct sk_buff *validate_xmit_skb_list(struct sk_buff *skb, struct net_device *d } EXPORT_SYMBOL_GPL(validate_xmit_skb_list); -static void qdisc_pkt_len_init(struct sk_buff *skb) +static void qdisc_pkt_len_segs_init(struct sk_buff *skb) { struct skb_shared_info *shinfo = skb_shinfo(skb); + u16 gso_segs; qdisc_skb_cb(skb)->pkt_len = skb->len; + if (!shinfo->gso_size) { + qdisc_skb_cb(skb)->pkt_segs = 1; + return; + } + + qdisc_skb_cb(skb)->pkt_segs = gso_segs = shinfo->gso_segs; /* To get more precise estimation of bytes sent on wire, * we add to pkt_len the headers size of all segments */ - if (shinfo->gso_size && skb_transport_header_was_set(skb)) { - u16 gso_segs = shinfo->gso_segs; + if (skb_transport_header_was_set(skb)) { unsigned int hdr_len; /* mac layer + network layer */ @@ -4113,6 +4119,7 @@ static void qdisc_pkt_len_init(struct sk_buff *skb) return; gso_segs = DIV_ROUND_UP(payload, shinfo->gso_size); shinfo->gso_segs = gso_segs; + qdisc_skb_cb(skb)->pkt_segs = gso_segs; } qdisc_skb_cb(skb)->pkt_len += (gso_segs - 1) * hdr_len; } @@ -4738,7 +4745,7 @@ int __dev_queue_xmit(struct sk_buff *skb, struct net_device *sb_dev) skb_update_prio(skb); - qdisc_pkt_len_init(skb); + qdisc_pkt_len_segs_init(skb); tcx_set_ingress(skb, false); #ifdef CONFIG_NET_EGRESS if (static_branch_unlikely(&egress_needed_key)) { diff --git a/net/sched/sch_cake.c b/net/sched/sch_cake.c index 32bacfc314c2..9213129f0de1 100644 --- a/net/sched/sch_cake.c +++ b/net/sched/sch_cake.c @@ -1406,7 +1406,7 @@ static u32 cake_overhead(struct cake_sched_data *q, const struct sk_buff *skb) if (!shinfo->gso_size) return cake_calc_overhead(q, len, off); - /* borrowed from qdisc_pkt_len_init() */ + /* borrowed from qdisc_pkt_len_segs_init() */ if (!skb->encapsulation) hdr_len = skb_transport_offset(skb); else From f9e00e51e391d08de31ca98d9f8609a1bceec2d2 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 21 Nov 2025 08:32:46 +0000 Subject: [PATCH 1102/1381] net: use qdisc_pkt_len_segs_init() in sch_handle_ingress() sch_handle_ingress() sets qdisc_skb_cb(skb)->pkt_len. We also need to initialize qdisc_skb_cb(skb)->pkt_segs. Signed-off-by: Eric Dumazet Link: https://patch.msgid.link/20251121083256.674562-5-edumazet@google.com Signed-off-by: Paolo Abeni --- net/core/dev.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/core/dev.c b/net/core/dev.c index 895c3e37e686..e19eb4e9d77c 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4434,7 +4434,7 @@ sch_handle_ingress(struct sk_buff *skb, struct packet_type **pt_prev, int *ret, *pt_prev = NULL; } - qdisc_skb_cb(skb)->pkt_len = skb->len; + qdisc_pkt_len_segs_init(skb); tcx_set_ingress(skb, true); if (static_branch_unlikely(&tcx_needed_key)) { From 2773cb0b3120eb5c4b66d949eb99853d5bae1221 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 21 Nov 2025 08:32:47 +0000 Subject: [PATCH 1103/1381] net_sched: use qdisc_skb_cb(skb)->pkt_segs in bstats_update() Avoid up to two cache line misses in qdisc dequeue() to fetch skb_shinfo(skb)->gso_segs/gso_size while qdisc spinlock is held. This gives a 5 % improvement in a TX intensive workload. Signed-off-by: Eric Dumazet Link: https://patch.msgid.link/20251121083256.674562-6-edumazet@google.com Signed-off-by: Paolo Abeni --- include/net/sch_generic.h | 13 ++++++++++--- net/sched/sch_cake.c | 1 + net/sched/sch_dualpi2.c | 1 + net/sched/sch_netem.c | 1 + net/sched/sch_qfq.c | 2 +- net/sched/sch_taprio.c | 1 + net/sched/sch_tbf.c | 1 + 7 files changed, 16 insertions(+), 4 deletions(-) diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index 9cd8b5d4b236..cdf7a58ebcf5 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -829,6 +829,15 @@ static inline unsigned int qdisc_pkt_len(const struct sk_buff *skb) return qdisc_skb_cb(skb)->pkt_len; } +static inline unsigned int qdisc_pkt_segs(const struct sk_buff *skb) +{ + u32 pkt_segs = qdisc_skb_cb(skb)->pkt_segs; + + DEBUG_NET_WARN_ON_ONCE(pkt_segs != + (skb_is_gso(skb) ? skb_shinfo(skb)->gso_segs : 1)); + return pkt_segs; +} + /* additional qdisc xmit flags (NET_XMIT_MASK in linux/netdevice.h) */ enum net_xmit_qdisc_t { __NET_XMIT_STOLEN = 0x00010000, @@ -870,9 +879,7 @@ static inline void _bstats_update(struct gnet_stats_basic_sync *bstats, static inline void bstats_update(struct gnet_stats_basic_sync *bstats, const struct sk_buff *skb) { - _bstats_update(bstats, - qdisc_pkt_len(skb), - skb_is_gso(skb) ? skb_shinfo(skb)->gso_segs : 1); + _bstats_update(bstats, qdisc_pkt_len(skb), qdisc_pkt_segs(skb)); } static inline void qdisc_bstats_cpu_update(struct Qdisc *sch, diff --git a/net/sched/sch_cake.c b/net/sched/sch_cake.c index 9213129f0de1..a20880034aa5 100644 --- a/net/sched/sch_cake.c +++ b/net/sched/sch_cake.c @@ -1800,6 +1800,7 @@ static s32 cake_enqueue(struct sk_buff *skb, struct Qdisc *sch, skb_list_walk_safe(segs, segs, nskb) { skb_mark_not_on_list(segs); qdisc_skb_cb(segs)->pkt_len = segs->len; + qdisc_skb_cb(segs)->pkt_segs = 1; cobalt_set_enqueue_time(segs, now); get_cobalt_cb(segs)->adjusted_len = cake_overhead(q, segs); diff --git a/net/sched/sch_dualpi2.c b/net/sched/sch_dualpi2.c index 4b975feb52b1..6d7e6389758d 100644 --- a/net/sched/sch_dualpi2.c +++ b/net/sched/sch_dualpi2.c @@ -475,6 +475,7 @@ static int dualpi2_qdisc_enqueue(struct sk_buff *skb, struct Qdisc *sch, * (3) Enqueue fragment & set ts in dualpi2_enqueue_skb */ qdisc_skb_cb(nskb)->pkt_len = nskb->len; + qdisc_skb_cb(nskb)->pkt_segs = 1; dualpi2_skb_cb(nskb)->classified = dualpi2_skb_cb(skb)->classified; dualpi2_skb_cb(nskb)->ect = dualpi2_skb_cb(skb)->ect; diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c index eafc316ae319..32a5f3304046 100644 --- a/net/sched/sch_netem.c +++ b/net/sched/sch_netem.c @@ -429,6 +429,7 @@ static struct sk_buff *netem_segment(struct sk_buff *skb, struct Qdisc *sch, struct sk_buff *segs; netdev_features_t features = netif_skb_features(skb); + qdisc_skb_cb(skb)->pkt_segs = 1; segs = skb_gso_segment(skb, features & ~NETIF_F_GSO_MASK); if (IS_ERR_OR_NULL(segs)) { diff --git a/net/sched/sch_qfq.c b/net/sched/sch_qfq.c index 2255355e51d3..d920f57dc6d7 100644 --- a/net/sched/sch_qfq.c +++ b/net/sched/sch_qfq.c @@ -1250,7 +1250,7 @@ static int qfq_enqueue(struct sk_buff *skb, struct Qdisc *sch, } } - gso_segs = skb_is_gso(skb) ? skb_shinfo(skb)->gso_segs : 1; + gso_segs = qdisc_pkt_segs(skb); err = qdisc_enqueue(skb, cl->qdisc, to_free); if (unlikely(err != NET_XMIT_SUCCESS)) { pr_debug("qfq_enqueue: enqueue failed %d\n", err); diff --git a/net/sched/sch_taprio.c b/net/sched/sch_taprio.c index 39b735386996..300d577b3286 100644 --- a/net/sched/sch_taprio.c +++ b/net/sched/sch_taprio.c @@ -595,6 +595,7 @@ static int taprio_enqueue_segmented(struct sk_buff *skb, struct Qdisc *sch, skb_list_walk_safe(segs, segs, nskb) { skb_mark_not_on_list(segs); qdisc_skb_cb(segs)->pkt_len = segs->len; + qdisc_skb_cb(segs)->pkt_segs = 1; slen += segs->len; /* FIXME: we should be segmenting to a smaller size diff --git a/net/sched/sch_tbf.c b/net/sched/sch_tbf.c index 4c977f049670..f2340164f579 100644 --- a/net/sched/sch_tbf.c +++ b/net/sched/sch_tbf.c @@ -221,6 +221,7 @@ static int tbf_segment(struct sk_buff *skb, struct Qdisc *sch, skb_mark_not_on_list(segs); seg_len = segs->len; qdisc_skb_cb(segs)->pkt_len = seg_len; + qdisc_skb_cb(segs)->pkt_segs = 1; ret = qdisc_enqueue(segs, q->qdisc, to_free); if (ret != NET_XMIT_SUCCESS) { if (net_xmit_drop_count(ret)) From c5d34f4583ea883b8d3441ded83e3a7207e0182d Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 21 Nov 2025 08:32:48 +0000 Subject: [PATCH 1104/1381] net_sched: cake: use qdisc_pkt_segs() Use new qdisc_pkt_segs() to avoid a cache line miss in cake_enqueue() for non GSO packets. cake_overhead() does not have to recompute it. Signed-off-by: Eric Dumazet Link: https://patch.msgid.link/20251121083256.674562-7-edumazet@google.com Signed-off-by: Paolo Abeni --- net/sched/sch_cake.c | 12 +++--------- 1 file changed, 3 insertions(+), 9 deletions(-) diff --git a/net/sched/sch_cake.c b/net/sched/sch_cake.c index a20880034aa5..5948a149129c 100644 --- a/net/sched/sch_cake.c +++ b/net/sched/sch_cake.c @@ -1398,12 +1398,12 @@ static u32 cake_overhead(struct cake_sched_data *q, const struct sk_buff *skb) const struct skb_shared_info *shinfo = skb_shinfo(skb); unsigned int hdr_len, last_len = 0; u32 off = skb_network_offset(skb); + u16 segs = qdisc_pkt_segs(skb); u32 len = qdisc_pkt_len(skb); - u16 segs = 1; q->avg_netoff = cake_ewma(q->avg_netoff, off << 16, 8); - if (!shinfo->gso_size) + if (segs == 1) return cake_calc_overhead(q, len, off); /* borrowed from qdisc_pkt_len_segs_init() */ @@ -1430,12 +1430,6 @@ static u32 cake_overhead(struct cake_sched_data *q, const struct sk_buff *skb) hdr_len += sizeof(struct udphdr); } - if (unlikely(shinfo->gso_type & SKB_GSO_DODGY)) - segs = DIV_ROUND_UP(skb->len - hdr_len, - shinfo->gso_size); - else - segs = shinfo->gso_segs; - len = shinfo->gso_size + hdr_len; last_len = skb->len - shinfo->gso_size * (segs - 1); @@ -1788,7 +1782,7 @@ static s32 cake_enqueue(struct sk_buff *skb, struct Qdisc *sch, if (unlikely(len > b->max_skblen)) b->max_skblen = len; - if (skb_is_gso(skb) && q->rate_flags & CAKE_FLAG_SPLIT_GSO) { + if (qdisc_pkt_segs(skb) > 1 && q->rate_flags & CAKE_FLAG_SPLIT_GSO) { struct sk_buff *segs, *nskb; netdev_features_t features = netif_skb_features(skb); unsigned int slen = 0, numsegs = 0; From ad50d5a3fc20327e133e2db849c6e67fc49650e6 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 21 Nov 2025 08:32:49 +0000 Subject: [PATCH 1105/1381] net_sched: add Qdisc_read_mostly and Qdisc_write groups It is possible to reorg Qdisc to avoid always dirtying 2 cache lines in fast path by reducing this to a single dirtied cache line. In current layout, we change only four/six fields in the first cache line: - q.spinlock - q.qlen - bstats.bytes - bstats.packets - some Qdisc also change q.next/q.prev In the second cache line we change in the fast path: - running - state - qstats.backlog /* --- cacheline 2 boundary (128 bytes) --- */ struct sk_buff_head gso_skb __attribute__((__aligned__(64))); /* 0x80 0x18 */ struct qdisc_skb_head q; /* 0x98 0x18 */ struct gnet_stats_basic_sync bstats __attribute__((__aligned__(16))); /* 0xb0 0x10 */ /* --- cacheline 3 boundary (192 bytes) --- */ struct gnet_stats_queue qstats; /* 0xc0 0x14 */ bool running; /* 0xd4 0x1 */ /* XXX 3 bytes hole, try to pack */ unsigned long state; /* 0xd8 0x8 */ struct Qdisc * next_sched; /* 0xe0 0x8 */ struct sk_buff_head skb_bad_txq; /* 0xe8 0x18 */ /* --- cacheline 4 boundary (256 bytes) --- */ Reorganize things to have a first cache line mostly read, then a mostly written one. This gives a ~3% increase of performance under tx stress. Note that there is an additional hole because @qstats now spans over a third cache line. /* --- cacheline 2 boundary (128 bytes) --- */ __u8 __cacheline_group_begin__Qdisc_read_mostly[0] __attribute__((__aligned__(64))); /* 0x80 0 */ struct sk_buff_head gso_skb; /* 0x80 0x18 */ struct Qdisc * next_sched; /* 0x98 0x8 */ struct sk_buff_head skb_bad_txq; /* 0xa0 0x18 */ __u8 __cacheline_group_end__Qdisc_read_mostly[0]; /* 0xb8 0 */ /* XXX 8 bytes hole, try to pack */ /* --- cacheline 3 boundary (192 bytes) --- */ __u8 __cacheline_group_begin__Qdisc_write[0] __attribute__((__aligned__(64))); /* 0xc0 0 */ struct qdisc_skb_head q; /* 0xc0 0x18 */ unsigned long state; /* 0xd8 0x8 */ struct gnet_stats_basic_sync bstats __attribute__((__aligned__(16))); /* 0xe0 0x10 */ bool running; /* 0xf0 0x1 */ /* XXX 3 bytes hole, try to pack */ struct gnet_stats_queue qstats; /* 0xf4 0x14 */ /* --- cacheline 4 boundary (256 bytes) was 8 bytes ago --- */ __u8 __cacheline_group_end__Qdisc_write[0]; /* 0x108 0 */ /* XXX 56 bytes hole, try to pack */ Signed-off-by: Eric Dumazet Link: https://patch.msgid.link/20251121083256.674562-8-edumazet@google.com Signed-off-by: Paolo Abeni --- include/net/sch_generic.h | 29 ++++++++++++++++++----------- 1 file changed, 18 insertions(+), 11 deletions(-) diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index cdf7a58ebcf5..79501499dafb 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -103,17 +103,24 @@ struct Qdisc { int pad; refcount_t refcnt; - /* - * For performance sake on SMP, we put highly modified fields at the end - */ - struct sk_buff_head gso_skb ____cacheline_aligned_in_smp; - struct qdisc_skb_head q; - struct gnet_stats_basic_sync bstats; - struct gnet_stats_queue qstats; - bool running; /* must be written under qdisc spinlock */ - unsigned long state; - struct Qdisc *next_sched; - struct sk_buff_head skb_bad_txq; + /* Cache line potentially dirtied in dequeue() or __netif_reschedule(). */ + __cacheline_group_begin(Qdisc_read_mostly) ____cacheline_aligned; + struct sk_buff_head gso_skb; + struct Qdisc *next_sched; + struct sk_buff_head skb_bad_txq; + __cacheline_group_end(Qdisc_read_mostly); + + /* Fields dirtied in dequeue() fast path. */ + __cacheline_group_begin(Qdisc_write) ____cacheline_aligned; + struct qdisc_skb_head q; + unsigned long state; + struct gnet_stats_basic_sync bstats; + bool running; /* must be written under qdisc spinlock */ + + /* Note : we only change qstats.backlog in fast path. */ + struct gnet_stats_queue qstats; + __cacheline_group_end(Qdisc_write); + atomic_long_t defer_count ____cacheline_aligned_in_smp; struct llist_head defer_list; From 3c1100f042c006cae6c241028cc4c69e1a70483f Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 21 Nov 2025 08:32:50 +0000 Subject: [PATCH 1106/1381] net_sched: sch_fq: move qdisc_bstats_update() to fq_dequeue_skb() Group together changes to qdisc fields to reduce chances of false sharing if another cpu attempts to acquire the qdisc spinlock. qdisc_qstats_backlog_dec(sch, skb); sch->q.qlen--; qdisc_bstats_update(sch, skb); Signed-off-by: Eric Dumazet Link: https://patch.msgid.link/20251121083256.674562-9-edumazet@google.com Signed-off-by: Paolo Abeni --- net/sched/sch_fq.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/sched/sch_fq.c b/net/sched/sch_fq.c index fee922da2f99..0b0ca1aa9251 100644 --- a/net/sched/sch_fq.c +++ b/net/sched/sch_fq.c @@ -497,6 +497,7 @@ static void fq_dequeue_skb(struct Qdisc *sch, struct fq_flow *flow, skb_mark_not_on_list(skb); qdisc_qstats_backlog_dec(sch, skb); sch->q.qlen--; + qdisc_bstats_update(sch, skb); } static void flow_queue_add(struct fq_flow *flow, struct sk_buff *skb) @@ -776,7 +777,6 @@ static struct sk_buff *fq_dequeue(struct Qdisc *sch) f->time_next_packet = now + len; } out: - qdisc_bstats_update(sch, skb); return skb; } From 2f9babc04d74cbf984f0cb5b6e20bd78fdf32997 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 21 Nov 2025 08:32:51 +0000 Subject: [PATCH 1107/1381] net_sched: sch_fq: prefetch one skb ahead in dequeue() prefetch the skb that we are likely to dequeue at the next dequeue(). Also call fq_dequeue_skb() a bit sooner in fq_dequeue(). This reduces the window between read of q.qlen and changes of fields in the cache line that could be dirtied by another cpu trying to queue a packet. Signed-off-by: Eric Dumazet Link: https://patch.msgid.link/20251121083256.674562-10-edumazet@google.com Signed-off-by: Paolo Abeni --- net/sched/sch_fq.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/net/sched/sch_fq.c b/net/sched/sch_fq.c index 0b0ca1aa9251..6e5f2f4f2415 100644 --- a/net/sched/sch_fq.c +++ b/net/sched/sch_fq.c @@ -480,7 +480,10 @@ static void fq_erase_head(struct Qdisc *sch, struct fq_flow *flow, struct sk_buff *skb) { if (skb == flow->head) { - flow->head = skb->next; + struct sk_buff *next = skb->next; + + prefetch(next); + flow->head = next; } else { rb_erase(&skb->rbnode, &flow->t_root); skb->dev = qdisc_dev(sch); @@ -712,6 +715,7 @@ static struct sk_buff *fq_dequeue(struct Qdisc *sch) goto begin; } prefetch(&skb->end); + fq_dequeue_skb(sch, f, skb); if ((s64)(now - time_next_packet - q->ce_threshold) > 0) { INET_ECN_set_ce(skb); q->stat_ce_mark++; @@ -719,7 +723,6 @@ static struct sk_buff *fq_dequeue(struct Qdisc *sch) if (--f->qlen == 0) q->inactive_flows++; q->band_pkt_count[fq_skb_cb(skb)->band]--; - fq_dequeue_skb(sch, f, skb); } else { head->first = f->next; /* force a pass through old_flows to prevent starvation */ From b2e9821cff6c3c9ac107fce5327070f4462bf8a7 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 21 Nov 2025 08:32:52 +0000 Subject: [PATCH 1108/1381] net: prefech skb->priority in __dev_xmit_skb() Most qdiscs need to read skb->priority at enqueue time(). In commit 100dfa74cad9 ("net: dev_queue_xmit() llist adoption") I added a prefetch(next), lets add another one for the second half of skb. Note that skb->priority and skb->hash share a common cache line, so this patch helps qdiscs needing both fields. Signed-off-by: Eric Dumazet Link: https://patch.msgid.link/20251121083256.674562-11-edumazet@google.com Signed-off-by: Paolo Abeni --- net/core/dev.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/core/dev.c b/net/core/dev.c index e19eb4e9d77c..53e2496dc429 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4246,6 +4246,7 @@ static inline int __dev_xmit_skb(struct sk_buff *skb, struct Qdisc *q, llist_for_each_entry_safe(skb, next, ll_list, ll_node) { prefetch(next); + prefetch(&next->priority); skb_mark_not_on_list(skb); rc = dev_qdisc_enqueue(skb, q, &to_free, txq); count++; From 4792c3a4c1470202b0d5bf44b6058cb0fb050ba8 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 21 Nov 2025 08:32:53 +0000 Subject: [PATCH 1109/1381] net: annotate a data-race in __dev_xmit_skb() q->limit is read locklessly, add a READ_ONCE(). Fixes: 100dfa74cad9 ("net: dev_queue_xmit() llist adoption") Signed-off-by: Eric Dumazet Link: https://patch.msgid.link/20251121083256.674562-12-edumazet@google.com Signed-off-by: Paolo Abeni --- net/core/dev.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/core/dev.c b/net/core/dev.c index 53e2496dc429..10042139dbb0 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4194,7 +4194,7 @@ static inline int __dev_xmit_skb(struct sk_buff *skb, struct Qdisc *q, do { if (first_n && !defer_count) { defer_count = atomic_long_inc_return(&q->defer_count); - if (unlikely(defer_count > q->limit)) { + if (unlikely(defer_count > READ_ONCE(q->limit))) { kfree_skb_reason(skb, SKB_DROP_REASON_QDISC_DROP); return NET_XMIT_DROP; } From 0170d7f47c8bb0311bc802bad52245c045f151fe Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 21 Nov 2025 08:32:54 +0000 Subject: [PATCH 1110/1381] net_sched: add tcf_kfree_skb_list() helper Using kfree_skb_list_reason() to free list of skbs from qdisc operations seems wrong as each skb might have a different drop reason. Cleanup __dev_xmit_skb() to call tcf_kfree_skb_list() once in preparation of the following patch. Signed-off-by: Eric Dumazet Link: https://patch.msgid.link/20251121083256.674562-13-edumazet@google.com Signed-off-by: Paolo Abeni --- include/net/sch_generic.h | 11 +++++++++++ net/core/dev.c | 15 +++++---------- 2 files changed, 16 insertions(+), 10 deletions(-) diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index 79501499dafb..b8092d0378a0 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -1105,6 +1105,17 @@ static inline void tcf_set_drop_reason(const struct sk_buff *skb, tc_skb_cb(skb)->drop_reason = reason; } +static inline void tcf_kfree_skb_list(struct sk_buff *skb) +{ + while (unlikely(skb)) { + struct sk_buff *next = skb->next; + + prefetch(next); + kfree_skb_reason(skb, tcf_get_drop_reason(skb)); + skb = next; + } +} + /* Instead of calling kfree_skb() while root qdisc lock is held, * queue the skb for future freeing at end of __dev_xmit_skb() */ diff --git a/net/core/dev.c b/net/core/dev.c index 10042139dbb0..e865cdb9b696 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4162,7 +4162,7 @@ static inline int __dev_xmit_skb(struct sk_buff *skb, struct Qdisc *q, __qdisc_run(q); qdisc_run_end(q); - goto no_lock_out; + goto free_skbs; } qdisc_bstats_cpu_update(q, skb); @@ -4176,12 +4176,7 @@ static inline int __dev_xmit_skb(struct sk_buff *skb, struct Qdisc *q, rc = dev_qdisc_enqueue(skb, q, &to_free, txq); qdisc_run(q); - -no_lock_out: - if (unlikely(to_free)) - kfree_skb_list_reason(to_free, - tcf_get_drop_reason(to_free)); - return rc; + goto free_skbs; } /* Open code llist_add(&skb->ll_node, &q->defer_list) + queue limit. @@ -4257,9 +4252,9 @@ static inline int __dev_xmit_skb(struct sk_buff *skb, struct Qdisc *q, } unlock: spin_unlock(root_lock); - if (unlikely(to_free)) - kfree_skb_list_reason(to_free, - tcf_get_drop_reason(to_free)); + +free_skbs: + tcf_kfree_skb_list(to_free); return rc; } From 191ff13e42a7b7824fec5b2ed84fd6481356754d Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 21 Nov 2025 08:32:55 +0000 Subject: [PATCH 1111/1381] net_sched: add qdisc_dequeue_drop() helper Some qdisc like cake, codel, fq_codel might drop packets in their dequeue() method. This is currently problematic because dequeue() runs with the qdisc spinlock held. Freeing skbs can be extremely expensive. Add qdisc_dequeue_drop() method and a new TCQ_F_DEQUEUE_DROPS so that these qdiscs can opt-in to defer the skb frees after the socket spinlock is released. TCQ_F_DEQUEUE_DROPS is an attempt to not penalize other qdiscs with an extra cache line miss. Signed-off-by: Eric Dumazet Link: https://patch.msgid.link/20251121083256.674562-14-edumazet@google.com Signed-off-by: Paolo Abeni --- include/net/pkt_sched.h | 5 +++-- include/net/sch_generic.h | 30 +++++++++++++++++++++++++++--- net/core/dev.c | 22 +++++++++++++--------- 3 files changed, 43 insertions(+), 14 deletions(-) diff --git a/include/net/pkt_sched.h b/include/net/pkt_sched.h index 4678db45832a..e703c507d0da 100644 --- a/include/net/pkt_sched.h +++ b/include/net/pkt_sched.h @@ -114,12 +114,13 @@ bool sch_direct_xmit(struct sk_buff *skb, struct Qdisc *q, void __qdisc_run(struct Qdisc *q); -static inline void qdisc_run(struct Qdisc *q) +static inline struct sk_buff *qdisc_run(struct Qdisc *q) { if (qdisc_run_begin(q)) { __qdisc_run(q); - qdisc_run_end(q); + return qdisc_run_end(q); } + return NULL; } extern const struct nla_policy rtm_tca_policy[TCA_MAX + 1]; diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index b8092d0378a0..c3a7268b567e 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -88,6 +88,8 @@ struct Qdisc { #define TCQ_F_INVISIBLE 0x80 /* invisible by default in dump */ #define TCQ_F_NOLOCK 0x100 /* qdisc does not require locking */ #define TCQ_F_OFFLOADED 0x200 /* qdisc is offloaded to HW */ +#define TCQ_F_DEQUEUE_DROPS 0x400 /* ->dequeue() can drop packets in q->to_free */ + u32 limit; const struct Qdisc_ops *ops; struct qdisc_size_table __rcu *stab; @@ -119,6 +121,8 @@ struct Qdisc { /* Note : we only change qstats.backlog in fast path. */ struct gnet_stats_queue qstats; + + struct sk_buff *to_free; __cacheline_group_end(Qdisc_write); @@ -218,8 +222,10 @@ static inline bool qdisc_run_begin(struct Qdisc *qdisc) return true; } -static inline void qdisc_run_end(struct Qdisc *qdisc) +static inline struct sk_buff *qdisc_run_end(struct Qdisc *qdisc) { + struct sk_buff *to_free = NULL; + if (qdisc->flags & TCQ_F_NOLOCK) { spin_unlock(&qdisc->seqlock); @@ -232,9 +238,16 @@ static inline void qdisc_run_end(struct Qdisc *qdisc) if (unlikely(test_bit(__QDISC_STATE_MISSED, &qdisc->state))) __netif_schedule(qdisc); - } else { - WRITE_ONCE(qdisc->running, false); + return NULL; } + + if (qdisc->flags & TCQ_F_DEQUEUE_DROPS) { + to_free = qdisc->to_free; + if (to_free) + qdisc->to_free = NULL; + } + WRITE_ONCE(qdisc->running, false); + return to_free; } static inline bool qdisc_may_bulk(const struct Qdisc *qdisc) @@ -1116,6 +1129,17 @@ static inline void tcf_kfree_skb_list(struct sk_buff *skb) } } +static inline void qdisc_dequeue_drop(struct Qdisc *q, struct sk_buff *skb, + enum skb_drop_reason reason) +{ + DEBUG_NET_WARN_ON_ONCE(!(q->flags & TCQ_F_DEQUEUE_DROPS)); + DEBUG_NET_WARN_ON_ONCE(q->flags & TCQ_F_NOLOCK); + + tcf_set_drop_reason(skb, reason); + skb->next = q->to_free; + q->to_free = skb; +} + /* Instead of calling kfree_skb() while root qdisc lock is held, * queue the skb for future freeing at end of __dev_xmit_skb() */ diff --git a/net/core/dev.c b/net/core/dev.c index e865cdb9b696..9094c0fb8c68 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4141,7 +4141,7 @@ static inline int __dev_xmit_skb(struct sk_buff *skb, struct Qdisc *q, struct net_device *dev, struct netdev_queue *txq) { - struct sk_buff *next, *to_free = NULL; + struct sk_buff *next, *to_free = NULL, *to_free2 = NULL; spinlock_t *root_lock = qdisc_lock(q); struct llist_node *ll_list, *first_n; unsigned long defer_count = 0; @@ -4160,7 +4160,7 @@ static inline int __dev_xmit_skb(struct sk_buff *skb, struct Qdisc *q, if (unlikely(!nolock_qdisc_is_empty(q))) { rc = dev_qdisc_enqueue(skb, q, &to_free, txq); __qdisc_run(q); - qdisc_run_end(q); + to_free2 = qdisc_run_end(q); goto free_skbs; } @@ -4170,12 +4170,13 @@ static inline int __dev_xmit_skb(struct sk_buff *skb, struct Qdisc *q, !nolock_qdisc_is_empty(q)) __qdisc_run(q); - qdisc_run_end(q); - return NET_XMIT_SUCCESS; + to_free2 = qdisc_run_end(q); + rc = NET_XMIT_SUCCESS; + goto free_skbs; } rc = dev_qdisc_enqueue(skb, q, &to_free, txq); - qdisc_run(q); + to_free2 = qdisc_run(q); goto free_skbs; } @@ -4234,7 +4235,7 @@ static inline int __dev_xmit_skb(struct sk_buff *skb, struct Qdisc *q, qdisc_bstats_update(q, skb); if (sch_direct_xmit(skb, q, dev, txq, root_lock, true)) __qdisc_run(q); - qdisc_run_end(q); + to_free2 = qdisc_run_end(q); rc = NET_XMIT_SUCCESS; } else { int count = 0; @@ -4246,7 +4247,7 @@ static inline int __dev_xmit_skb(struct sk_buff *skb, struct Qdisc *q, rc = dev_qdisc_enqueue(skb, q, &to_free, txq); count++; } - qdisc_run(q); + to_free2 = qdisc_run(q); if (count != 1) rc = NET_XMIT_SUCCESS; } @@ -4255,6 +4256,7 @@ static inline int __dev_xmit_skb(struct sk_buff *skb, struct Qdisc *q, free_skbs: tcf_kfree_skb_list(to_free); + tcf_kfree_skb_list(to_free2); return rc; } @@ -5747,8 +5749,9 @@ static __latent_entropy void net_tx_action(void) rcu_read_lock(); while (head) { - struct Qdisc *q = head; spinlock_t *root_lock = NULL; + struct sk_buff *to_free; + struct Qdisc *q = head; head = head->next_sched; @@ -5775,9 +5778,10 @@ static __latent_entropy void net_tx_action(void) } clear_bit(__QDISC_STATE_SCHED, &q->state); - qdisc_run(q); + to_free = qdisc_run(q); if (root_lock) spin_unlock(root_lock); + tcf_kfree_skb_list(to_free); } rcu_read_unlock(); From a6efc273ab8245722eee2150fa12cf75781dc410 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 21 Nov 2025 08:32:56 +0000 Subject: [PATCH 1112/1381] net_sched: use qdisc_dequeue_drop() in cake, codel, fq_codel cake, codel and fq_codel can drop many packets from dequeue(). Use qdisc_dequeue_drop() so that the freeing can happen outside of the qdisc spinlock scope. Add TCQ_F_DEQUEUE_DROPS to sch->flags. Signed-off-by: Eric Dumazet Link: https://patch.msgid.link/20251121083256.674562-15-edumazet@google.com Signed-off-by: Paolo Abeni --- net/sched/sch_cake.c | 4 +++- net/sched/sch_codel.c | 4 +++- net/sched/sch_fq_codel.c | 5 ++++- 3 files changed, 10 insertions(+), 3 deletions(-) diff --git a/net/sched/sch_cake.c b/net/sched/sch_cake.c index 5948a149129c..0ea9440f68c6 100644 --- a/net/sched/sch_cake.c +++ b/net/sched/sch_cake.c @@ -2183,7 +2183,7 @@ static struct sk_buff *cake_dequeue(struct Qdisc *sch) b->tin_dropped++; qdisc_tree_reduce_backlog(sch, 1, qdisc_pkt_len(skb)); qdisc_qstats_drop(sch); - kfree_skb_reason(skb, reason); + qdisc_dequeue_drop(sch, skb, reason); if (q->rate_flags & CAKE_FLAG_INGRESS) goto retry; } @@ -2724,6 +2724,8 @@ static int cake_init(struct Qdisc *sch, struct nlattr *opt, int i, j, err; sch->limit = 10240; + sch->flags |= TCQ_F_DEQUEUE_DROPS; + q->tin_mode = CAKE_DIFFSERV_DIFFSERV3; q->flow_mode = CAKE_FLOW_TRIPLE; diff --git a/net/sched/sch_codel.c b/net/sched/sch_codel.c index fa0314679e43..c6551578f1cf 100644 --- a/net/sched/sch_codel.c +++ b/net/sched/sch_codel.c @@ -52,7 +52,7 @@ static void drop_func(struct sk_buff *skb, void *ctx) { struct Qdisc *sch = ctx; - kfree_skb_reason(skb, SKB_DROP_REASON_QDISC_CONGESTED); + qdisc_dequeue_drop(sch, skb, SKB_DROP_REASON_QDISC_CONGESTED); qdisc_qstats_drop(sch); } @@ -182,6 +182,8 @@ static int codel_init(struct Qdisc *sch, struct nlattr *opt, else sch->flags &= ~TCQ_F_CAN_BYPASS; + sch->flags |= TCQ_F_DEQUEUE_DROPS; + return 0; } diff --git a/net/sched/sch_fq_codel.c b/net/sched/sch_fq_codel.c index a14142392939..dc187c7f06b1 100644 --- a/net/sched/sch_fq_codel.c +++ b/net/sched/sch_fq_codel.c @@ -275,7 +275,7 @@ static void drop_func(struct sk_buff *skb, void *ctx) { struct Qdisc *sch = ctx; - kfree_skb_reason(skb, SKB_DROP_REASON_QDISC_CONGESTED); + qdisc_dequeue_drop(sch, skb, SKB_DROP_REASON_QDISC_CONGESTED); qdisc_qstats_drop(sch); } @@ -519,6 +519,9 @@ static int fq_codel_init(struct Qdisc *sch, struct nlattr *opt, sch->flags |= TCQ_F_CAN_BYPASS; else sch->flags &= ~TCQ_F_CAN_BYPASS; + + sch->flags |= TCQ_F_DEQUEUE_DROPS; + return 0; alloc_failure: From d6eea0048bc38c259fa4121646e54e2c3bd08121 Mon Sep 17 00:00:00 2001 From: Javen Xu Date: Fri, 21 Nov 2025 17:01:04 +0800 Subject: [PATCH 1113/1381] r8169: add support for RTL9151A This adds support for chip RTL9151A. Its XID is 0x68b. It is bascially basd on the one with XID 0x688, but with different firmware file. Signed-off-by: Javen Xu Reviewed-by: Heiner Kallweit Link: https://patch.msgid.link/20251121090104.3753-1-javen_xu@realsil.com.cn Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/realtek/r8169_main.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c index de304d1eb477..9fd0ca399d5f 100644 --- a/drivers/net/ethernet/realtek/r8169_main.c +++ b/drivers/net/ethernet/realtek/r8169_main.c @@ -59,6 +59,7 @@ #define FIRMWARE_8125D_2 "rtl_nic/rtl8125d-2.fw" #define FIRMWARE_8125K_1 "rtl_nic/rtl8125k-1.fw" #define FIRMWARE_8125BP_2 "rtl_nic/rtl8125bp-2.fw" +#define FIRMWARE_9151A_1 "rtl_nic/rtl9151a-1.fw" #define FIRMWARE_8126A_2 "rtl_nic/rtl8126a-2.fw" #define FIRMWARE_8126A_3 "rtl_nic/rtl8126a-3.fw" #define FIRMWARE_8127A_1 "rtl_nic/rtl8127a-1.fw" @@ -111,6 +112,7 @@ static const struct rtl_chip_info { { 0x7cf, 0x681, RTL_GIGA_MAC_VER_66, "RTL8125BP", FIRMWARE_8125BP_2 }, /* 8125D family. */ + { 0x7cf, 0x68b, RTL_GIGA_MAC_VER_64, "RTL9151A", FIRMWARE_9151A_1 }, { 0x7cf, 0x68a, RTL_GIGA_MAC_VER_64, "RTL8125K", FIRMWARE_8125K_1 }, { 0x7cf, 0x689, RTL_GIGA_MAC_VER_64, "RTL8125D", FIRMWARE_8125D_2 }, { 0x7cf, 0x688, RTL_GIGA_MAC_VER_64, "RTL8125D", FIRMWARE_8125D_1 }, @@ -774,6 +776,7 @@ MODULE_FIRMWARE(FIRMWARE_8125D_1); MODULE_FIRMWARE(FIRMWARE_8125D_2); MODULE_FIRMWARE(FIRMWARE_8125K_1); MODULE_FIRMWARE(FIRMWARE_8125BP_2); +MODULE_FIRMWARE(FIRMWARE_9151A_1); MODULE_FIRMWARE(FIRMWARE_8126A_2); MODULE_FIRMWARE(FIRMWARE_8126A_3); MODULE_FIRMWARE(FIRMWARE_8127A_1); From 380d19db6e6c089c7d2902e02a85fd2bde3e519b Mon Sep 17 00:00:00 2001 From: Alan Maguire Date: Fri, 21 Nov 2025 18:12:31 +0000 Subject: [PATCH 1114/1381] cxgb4: Rename sched_class to avoid type clash drivers/net/ethernet/chelsio/cxgb4/sched.h declares a sched_class struct which has a type name clash with struct sched_class in kernel/sched/sched.h (a type used in a field in task_struct). When cxgb4 is a builtin we end up with both sched_class types, and as a result of this we wind up with DWARF (and derived from that BTF) with a duplicate incorrect task_struct representation. When cxgb4 is built-in this type clash can cause kernel builds to fail as resolve_btfids will fail when confused which task_struct to use. See [1] for more details. As such, renaming sched_class to ch_sched_class (in line with other structs like ch_sched_flowc) makes sense. [1] https://lore.kernel.org/bpf/2412725b-916c-47bd-91c3-c2d57e3e6c7b@acm.org/ Reported-by: Bart Van Assche Reviewed-by: Bart Van Assche Signed-off-by: Alan Maguire Acked-by: Potnuri Bharat Teja Link: https://patch.msgid.link/20251121181231.64337-1-alan.maguire@oracle.com Signed-off-by: Jakub Kicinski --- .../net/ethernet/chelsio/cxgb4/cxgb4_main.c | 2 +- .../chelsio/cxgb4/cxgb4_tc_matchall.c | 4 +- .../ethernet/chelsio/cxgb4/cxgb4_tc_mqprio.c | 2 +- drivers/net/ethernet/chelsio/cxgb4/sched.c | 44 +++++++++---------- drivers/net/ethernet/chelsio/cxgb4/sched.h | 12 ++--- 5 files changed, 32 insertions(+), 32 deletions(-) diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c index 7e2283c95b97..66b8854e059f 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c @@ -3485,7 +3485,7 @@ static int cxgb_set_tx_maxrate(struct net_device *dev, int index, u32 rate) struct adapter *adap = pi->adapter; struct ch_sched_queue qe = { 0 }; struct ch_sched_params p = { 0 }; - struct sched_class *e; + struct ch_sched_class *e; u32 req_rate; int err = 0; diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_matchall.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_matchall.c index 1672d3afe5be..f8dcf0b4abcd 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_matchall.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_matchall.c @@ -56,7 +56,7 @@ static int cxgb4_matchall_egress_validate(struct net_device *dev, struct port_info *pi = netdev2pinfo(dev); struct flow_action_entry *entry; struct ch_sched_queue qe; - struct sched_class *e; + struct ch_sched_class *e; u64 max_link_rate; u32 i, speed; int ret; @@ -180,7 +180,7 @@ static int cxgb4_matchall_alloc_tc(struct net_device *dev, struct port_info *pi = netdev2pinfo(dev); struct adapter *adap = netdev2adap(dev); struct flow_action_entry *entry; - struct sched_class *e; + struct ch_sched_class *e; int ret; u32 i; diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_mqprio.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_mqprio.c index 338b04f339b3..a2dcd2e24263 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_mqprio.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_mqprio.c @@ -330,7 +330,7 @@ static int cxgb4_mqprio_alloc_tc(struct net_device *dev, struct cxgb4_tc_port_mqprio *tc_port_mqprio; struct port_info *pi = netdev2pinfo(dev); struct adapter *adap = netdev2adap(dev); - struct sched_class *e; + struct ch_sched_class *e; int ret; u8 i; diff --git a/drivers/net/ethernet/chelsio/cxgb4/sched.c b/drivers/net/ethernet/chelsio/cxgb4/sched.c index a1b14468d1ff..38a30aeee122 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/sched.c +++ b/drivers/net/ethernet/chelsio/cxgb4/sched.c @@ -44,7 +44,7 @@ static int t4_sched_class_fw_cmd(struct port_info *pi, { struct adapter *adap = pi->adapter; struct sched_table *s = pi->sched_tbl; - struct sched_class *e; + struct ch_sched_class *e; int err = 0; e = &s->tab[p->u.params.class]; @@ -122,7 +122,7 @@ static void *t4_sched_entry_lookup(struct port_info *pi, const u32 val) { struct sched_table *s = pi->sched_tbl; - struct sched_class *e, *end; + struct ch_sched_class *e, *end; void *found = NULL; /* Look for an entry with matching @val */ @@ -166,8 +166,8 @@ static void *t4_sched_entry_lookup(struct port_info *pi, return found; } -struct sched_class *cxgb4_sched_queue_lookup(struct net_device *dev, - struct ch_sched_queue *p) +struct ch_sched_class *cxgb4_sched_queue_lookup(struct net_device *dev, + struct ch_sched_queue *p) { struct port_info *pi = netdev2pinfo(dev); struct sched_queue_entry *qe = NULL; @@ -187,7 +187,7 @@ static int t4_sched_queue_unbind(struct port_info *pi, struct ch_sched_queue *p) struct sched_queue_entry *qe = NULL; struct adapter *adap = pi->adapter; struct sge_eth_txq *txq; - struct sched_class *e; + struct ch_sched_class *e; int err = 0; if (p->queue < 0 || p->queue >= pi->nqsets) @@ -218,7 +218,7 @@ static int t4_sched_queue_bind(struct port_info *pi, struct ch_sched_queue *p) struct sched_queue_entry *qe = NULL; struct adapter *adap = pi->adapter; struct sge_eth_txq *txq; - struct sched_class *e; + struct ch_sched_class *e; unsigned int qid; int err = 0; @@ -260,7 +260,7 @@ static int t4_sched_flowc_unbind(struct port_info *pi, struct ch_sched_flowc *p) { struct sched_flowc_entry *fe = NULL; struct adapter *adap = pi->adapter; - struct sched_class *e; + struct ch_sched_class *e; int err = 0; if (p->tid < 0 || p->tid >= adap->tids.neotids) @@ -288,7 +288,7 @@ static int t4_sched_flowc_bind(struct port_info *pi, struct ch_sched_flowc *p) struct sched_table *s = pi->sched_tbl; struct sched_flowc_entry *fe = NULL; struct adapter *adap = pi->adapter; - struct sched_class *e; + struct ch_sched_class *e; int err = 0; if (p->tid < 0 || p->tid >= adap->tids.neotids) @@ -322,7 +322,7 @@ static int t4_sched_flowc_bind(struct port_info *pi, struct ch_sched_flowc *p) } static void t4_sched_class_unbind_all(struct port_info *pi, - struct sched_class *e, + struct ch_sched_class *e, enum sched_bind_type type) { if (!e) @@ -476,12 +476,12 @@ int cxgb4_sched_class_unbind(struct net_device *dev, void *arg, } /* If @p is NULL, fetch any available unused class */ -static struct sched_class *t4_sched_class_lookup(struct port_info *pi, - const struct ch_sched_params *p) +static struct ch_sched_class *t4_sched_class_lookup(struct port_info *pi, + const struct ch_sched_params *p) { struct sched_table *s = pi->sched_tbl; - struct sched_class *found = NULL; - struct sched_class *e, *end; + struct ch_sched_class *found = NULL; + struct ch_sched_class *e, *end; if (!p) { /* Get any available unused class */ @@ -522,10 +522,10 @@ static struct sched_class *t4_sched_class_lookup(struct port_info *pi, return found; } -static struct sched_class *t4_sched_class_alloc(struct port_info *pi, - struct ch_sched_params *p) +static struct ch_sched_class *t4_sched_class_alloc(struct port_info *pi, + struct ch_sched_params *p) { - struct sched_class *e = NULL; + struct ch_sched_class *e = NULL; u8 class_id; int err; @@ -579,8 +579,8 @@ static struct sched_class *t4_sched_class_alloc(struct port_info *pi, * scheduling class with matching @p is found, then the matching class is * returned. */ -struct sched_class *cxgb4_sched_class_alloc(struct net_device *dev, - struct ch_sched_params *p) +struct ch_sched_class *cxgb4_sched_class_alloc(struct net_device *dev, + struct ch_sched_params *p) { struct port_info *pi = netdev2pinfo(dev); u8 class_id; @@ -607,7 +607,7 @@ void cxgb4_sched_class_free(struct net_device *dev, u8 classid) struct port_info *pi = netdev2pinfo(dev); struct sched_table *s = pi->sched_tbl; struct ch_sched_params p; - struct sched_class *e; + struct ch_sched_class *e; u32 speed; int ret; @@ -640,7 +640,7 @@ void cxgb4_sched_class_free(struct net_device *dev, u8 classid) } } -static void t4_sched_class_free(struct net_device *dev, struct sched_class *e) +static void t4_sched_class_free(struct net_device *dev, struct ch_sched_class *e) { struct port_info *pi = netdev2pinfo(dev); @@ -660,7 +660,7 @@ struct sched_table *t4_init_sched(unsigned int sched_size) s->sched_size = sched_size; for (i = 0; i < s->sched_size; i++) { - memset(&s->tab[i], 0, sizeof(struct sched_class)); + memset(&s->tab[i], 0, sizeof(struct ch_sched_class)); s->tab[i].idx = i; s->tab[i].state = SCHED_STATE_UNUSED; INIT_LIST_HEAD(&s->tab[i].entry_list); @@ -682,7 +682,7 @@ void t4_cleanup_sched(struct adapter *adap) continue; for (i = 0; i < s->sched_size; i++) { - struct sched_class *e; + struct ch_sched_class *e; e = &s->tab[i]; if (e->state == SCHED_STATE_ACTIVE) diff --git a/drivers/net/ethernet/chelsio/cxgb4/sched.h b/drivers/net/ethernet/chelsio/cxgb4/sched.h index 6b3c778815f0..4d3b5a757536 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/sched.h +++ b/drivers/net/ethernet/chelsio/cxgb4/sched.h @@ -71,7 +71,7 @@ struct sched_flowc_entry { struct ch_sched_flowc param; }; -struct sched_class { +struct ch_sched_class { u8 state; u8 idx; struct ch_sched_params info; @@ -82,7 +82,7 @@ struct sched_class { struct sched_table { /* per port scheduling table */ u8 sched_size; - struct sched_class tab[] __counted_by(sched_size); + struct ch_sched_class tab[] __counted_by(sched_size); }; static inline bool can_sched(struct net_device *dev) @@ -103,15 +103,15 @@ static inline bool valid_class_id(struct net_device *dev, u8 class_id) return true; } -struct sched_class *cxgb4_sched_queue_lookup(struct net_device *dev, - struct ch_sched_queue *p); +struct ch_sched_class *cxgb4_sched_queue_lookup(struct net_device *dev, + struct ch_sched_queue *p); int cxgb4_sched_class_bind(struct net_device *dev, void *arg, enum sched_bind_type type); int cxgb4_sched_class_unbind(struct net_device *dev, void *arg, enum sched_bind_type type); -struct sched_class *cxgb4_sched_class_alloc(struct net_device *dev, - struct ch_sched_params *p); +struct ch_sched_class *cxgb4_sched_class_alloc(struct net_device *dev, + struct ch_sched_params *p); void cxgb4_sched_class_free(struct net_device *dev, u8 classid); struct sched_table *t4_init_sched(unsigned int size); From eba81b0a6de39e2466d37e410003642282b4e546 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Sat, 22 Nov 2025 13:23:09 +0200 Subject: [PATCH 1115/1381] net: dsa: cpu_dp->orig_ethtool_ops might be NULL In theory this would have been seen by now, but it seems that all drivers used as DSA conduit interfaces thus far have had ethtool_ops set, and it's hard to even find modern Ethernet drivers (and not VF ones) which don't use ethtool. Here is the unfiltered list of drivers which register any sort of net_device but don't set its ethtool_ops pointer. I don't think any of them 'risks' being used as a DSA conduit, maybe except for moxart, rnpbge and icssm, I'm not sure. - drivers/net/can/dev/dev.c - drivers/net/wwan/qcom_bam_dmux.c - drivers/net/wwan/t7xx/t7xx_netdev.c - drivers/net/arcnet/arcnet.c - drivers/net/hamradio/ - drivers/net/slip/slip.c - drivers/net/ethernet/ezchip/nps_enet.c - drivers/net/ethernet/moxa/moxart_ether.c - drivers/net/ethernet/wangxun/txgbevf/txgbevf_main.c - drivers/net/ethernet/wangxun/ngbevf/ngbevf_main.c - drivers/net/ethernet/huawei/hinic3/hinic3_main.c - drivers/net/ethernet/i825xx/ - drivers/net/ethernet/ti/icssm/icssm_prueth.c - drivers/net/ethernet/seeq/ - drivers/net/ethernet/litex/litex_liteeth.c - drivers/net/ethernet/sunplus/spl2sw_driver.c - drivers/net/ethernet/mucse/rnpgbe/rnpgbe_main.c - drivers/net/ipa/ - drivers/net/wireless/microchip/wilc1000/ - drivers/net/wireless/mediatek/mt76/dma.c - drivers/net/wireless/ath/ath12k/ - drivers/net/wireless/ath/ath11k/ - drivers/net/wireless/ath/ath6kl/ - drivers/net/wireless/ath/ath10k/ - drivers/net/wireless/intel/iwlwifi/pcie/gen1_2/trans.c - drivers/net/wireless/virtual/mac80211_hwsim.c - drivers/net/wireless/quantenna/qtnfmac/pcie/pcie.c - drivers/net/wireless/realtek/rtw89/core.c - drivers/net/wireless/realtek/rtw88/pci.c - drivers/net/caif/ - drivers/net/plip/ - drivers/net/wan/ - drivers/net/mctp/ - drivers/net/ppp/ - drivers/net/thunderbolt/ Nonetheless, it's good for the framework not to make such assumptions, and not panic when coming across such kind of host device in the future. Signed-off-by: Vladimir Oltean Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20251122112311.138784-2-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- net/dsa/conduit.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/net/dsa/conduit.c b/net/dsa/conduit.c index 4ae255cfb23f..f80795b3d046 100644 --- a/net/dsa/conduit.c +++ b/net/dsa/conduit.c @@ -26,7 +26,7 @@ static int dsa_conduit_get_regs_len(struct net_device *dev) int ret = 0; int len; - if (ops->get_regs_len) { + if (ops && ops->get_regs_len) { netdev_lock_ops(dev); len = ops->get_regs_len(dev); netdev_unlock_ops(dev); @@ -59,7 +59,7 @@ static void dsa_conduit_get_regs(struct net_device *dev, int port = cpu_dp->index; int len; - if (ops->get_regs_len && ops->get_regs) { + if (ops && ops->get_regs_len && ops->get_regs) { netdev_lock_ops(dev); len = ops->get_regs_len(dev); if (len < 0) { @@ -97,7 +97,7 @@ static void dsa_conduit_get_ethtool_stats(struct net_device *dev, int port = cpu_dp->index; int count = 0; - if (ops->get_sset_count && ops->get_ethtool_stats) { + if (ops && ops->get_sset_count && ops->get_ethtool_stats) { netdev_lock_ops(dev); count = ops->get_sset_count(dev, ETH_SS_STATS); ops->get_ethtool_stats(dev, stats, data); @@ -118,11 +118,11 @@ static void dsa_conduit_get_ethtool_phy_stats(struct net_device *dev, int port = cpu_dp->index; int count = 0; - if (dev->phydev && !ops->get_ethtool_phy_stats) { + if (dev->phydev && (!ops || !ops->get_ethtool_phy_stats)) { count = phy_ethtool_get_sset_count(dev->phydev); if (count >= 0) phy_ethtool_get_stats(dev->phydev, stats, data); - } else if (ops->get_sset_count && ops->get_ethtool_phy_stats) { + } else if (ops && ops->get_sset_count && ops->get_ethtool_phy_stats) { netdev_lock_ops(dev); count = ops->get_sset_count(dev, ETH_SS_PHY_STATS); ops->get_ethtool_phy_stats(dev, stats, data); @@ -145,9 +145,9 @@ static int dsa_conduit_get_sset_count(struct net_device *dev, int sset) netdev_lock_ops(dev); if (sset == ETH_SS_PHY_STATS && dev->phydev && - !ops->get_ethtool_phy_stats) + (!ops || !ops->get_ethtool_phy_stats)) count = phy_ethtool_get_sset_count(dev->phydev); - else if (ops->get_sset_count) + else if (ops && ops->get_sset_count) count = ops->get_sset_count(dev, sset); netdev_unlock_ops(dev); From 8afabd27fe46ebf991b4aea20b74e08196c15c0c Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Sat, 22 Nov 2025 13:23:10 +0200 Subject: [PATCH 1116/1381] net: dsa: use kernel data types for ethtool ops on conduit Suppress some checkpatch 'CHECK' messages about u8 being preferable over uint8_t, etc. No functional change. Signed-off-by: Vladimir Oltean Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20251122112311.138784-3-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- net/dsa/conduit.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) diff --git a/net/dsa/conduit.c b/net/dsa/conduit.c index f80795b3d046..c210e3129655 100644 --- a/net/dsa/conduit.c +++ b/net/dsa/conduit.c @@ -89,7 +89,7 @@ static void dsa_conduit_get_regs(struct net_device *dev, static void dsa_conduit_get_ethtool_stats(struct net_device *dev, struct ethtool_stats *stats, - uint64_t *data) + u64 *data) { struct dsa_port *cpu_dp = dev->dsa_ptr; const struct ethtool_ops *ops = cpu_dp->orig_ethtool_ops; @@ -110,7 +110,7 @@ static void dsa_conduit_get_ethtool_stats(struct net_device *dev, static void dsa_conduit_get_ethtool_phy_stats(struct net_device *dev, struct ethtool_stats *stats, - uint64_t *data) + u64 *data) { struct dsa_port *cpu_dp = dev->dsa_ptr; const struct ethtool_ops *ops = cpu_dp->orig_ethtool_ops; @@ -160,8 +160,8 @@ static int dsa_conduit_get_sset_count(struct net_device *dev, int sset) return count; } -static void dsa_conduit_get_strings(struct net_device *dev, uint32_t stringset, - uint8_t *data) +static void dsa_conduit_get_strings(struct net_device *dev, u32 stringset, + u8 *data) { struct dsa_port *cpu_dp = dev->dsa_ptr; const struct ethtool_ops *ops = cpu_dp->orig_ethtool_ops; @@ -169,8 +169,7 @@ static void dsa_conduit_get_strings(struct net_device *dev, uint32_t stringset, int port = cpu_dp->index; int len = ETH_GSTRING_LEN; int mcount = 0, count, i; - uint8_t pfx[4]; - uint8_t *ndata; + u8 pfx[4], *ndata; snprintf(pfx, sizeof(pfx), "p%.2d", port); /* We do not want to be NULL-terminated, since this is a prefix */ From f647ed2ca78ec4efcc436915b441da9de0974926 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Sat, 22 Nov 2025 13:23:11 +0200 Subject: [PATCH 1117/1381] net: dsa: append ethtool counters of all hidden ports to conduit Currently there is no way to see packet counters on cascade ports, and no clarity on how the API for that would look like. Because it's something that is currently needed, just extend the hack where ethtool -S on the conduit interface dumps CPU port counters, and also use it to dump counters of cascade ports. Note that the "pXX_" naming convention changes to "sXX_pYY", to distinguish between ports having the same index but belonging to different switches. This has a slight chance of causing regressions to existing tooling: - grepping for "p04_counter_name" still works, but might return more than one string now - grepping for " p04_counter_name" no longer works Signed-off-by: Vladimir Oltean Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20251122112311.138784-4-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- net/dsa/conduit.c | 128 ++++++++++++++++++++++++++++++++++------------ 1 file changed, 94 insertions(+), 34 deletions(-) diff --git a/net/dsa/conduit.c b/net/dsa/conduit.c index c210e3129655..a1b044467bd6 100644 --- a/net/dsa/conduit.c +++ b/net/dsa/conduit.c @@ -87,25 +87,51 @@ static void dsa_conduit_get_regs(struct net_device *dev, } } +static ssize_t dsa_conduit_append_port_stats(struct dsa_switch *ds, int port, + u64 *data, size_t start) +{ + int count; + + if (!ds->ops->get_sset_count) + return 0; + + count = ds->ops->get_sset_count(ds, port, ETH_SS_STATS); + if (count < 0) + return count; + + if (ds->ops->get_ethtool_stats) + ds->ops->get_ethtool_stats(ds, port, data + start); + + return count; +} + static void dsa_conduit_get_ethtool_stats(struct net_device *dev, struct ethtool_stats *stats, u64 *data) { - struct dsa_port *cpu_dp = dev->dsa_ptr; + struct dsa_port *dp, *cpu_dp = dev->dsa_ptr; const struct ethtool_ops *ops = cpu_dp->orig_ethtool_ops; - struct dsa_switch *ds = cpu_dp->ds; - int port = cpu_dp->index; - int count = 0; + struct dsa_switch_tree *dst = cpu_dp->dst; + int count, mcount = 0; if (ops && ops->get_sset_count && ops->get_ethtool_stats) { netdev_lock_ops(dev); - count = ops->get_sset_count(dev, ETH_SS_STATS); + mcount = ops->get_sset_count(dev, ETH_SS_STATS); ops->get_ethtool_stats(dev, stats, data); netdev_unlock_ops(dev); } - if (ds->ops->get_ethtool_stats) - ds->ops->get_ethtool_stats(ds, port, data + count); + list_for_each_entry(dp, &dst->ports, list) { + if (!dsa_port_is_dsa(dp) && !dsa_port_is_cpu(dp)) + continue; + + count = dsa_conduit_append_port_stats(dp->ds, dp->index, + data, mcount); + if (count < 0) + return; + + mcount += count; + } } static void dsa_conduit_get_ethtool_phy_stats(struct net_device *dev, @@ -136,11 +162,18 @@ static void dsa_conduit_get_ethtool_phy_stats(struct net_device *dev, ds->ops->get_ethtool_phy_stats(ds, port, data + count); } +static void dsa_conduit_append_port_sset_count(struct dsa_switch *ds, int port, + int sset, int *count) +{ + if (ds->ops->get_sset_count) + *count += ds->ops->get_sset_count(ds, port, sset); +} + static int dsa_conduit_get_sset_count(struct net_device *dev, int sset) { - struct dsa_port *cpu_dp = dev->dsa_ptr; + struct dsa_port *dp, *cpu_dp = dev->dsa_ptr; const struct ethtool_ops *ops = cpu_dp->orig_ethtool_ops; - struct dsa_switch *ds = cpu_dp->ds; + struct dsa_switch_tree *dst = cpu_dp->dst; int count = 0; netdev_lock_ops(dev); @@ -154,8 +187,46 @@ static int dsa_conduit_get_sset_count(struct net_device *dev, int sset) if (count < 0) count = 0; - if (ds->ops->get_sset_count) - count += ds->ops->get_sset_count(ds, cpu_dp->index, sset); + list_for_each_entry(dp, &dst->ports, list) { + if (!dsa_port_is_dsa(dp) && !dsa_port_is_cpu(dp)) + continue; + + dsa_conduit_append_port_sset_count(dp->ds, dp->index, sset, + &count); + } + + return count; +} + +static ssize_t dsa_conduit_append_port_strings(struct dsa_switch *ds, int port, + u32 stringset, u8 *data, + size_t start) +{ + int len = ETH_GSTRING_LEN; + u8 pfx[8], *ndata; + int count, i; + + if (!ds->ops->get_strings) + return 0; + + snprintf(pfx, sizeof(pfx), "s%.2d_p%.2d", ds->index, port); + /* We do not want to be NULL-terminated, since this is a prefix */ + pfx[sizeof(pfx) - 1] = '_'; + ndata = data + start * len; + /* This function copies ETH_GSTRINGS_LEN bytes, we will mangle + * the output after to prepend our CPU port prefix we + * constructed earlier + */ + ds->ops->get_strings(ds, port, stringset, ndata); + count = ds->ops->get_sset_count(ds, port, stringset); + if (count < 0) + return count; + + for (i = 0; i < count; i++) { + memmove(ndata + (i * len + sizeof(pfx)), + ndata + i * len, len - sizeof(pfx)); + memcpy(ndata + i * len, pfx, sizeof(pfx)); + } return count; } @@ -163,17 +234,10 @@ static int dsa_conduit_get_sset_count(struct net_device *dev, int sset) static void dsa_conduit_get_strings(struct net_device *dev, u32 stringset, u8 *data) { - struct dsa_port *cpu_dp = dev->dsa_ptr; + struct dsa_port *dp, *cpu_dp = dev->dsa_ptr; const struct ethtool_ops *ops = cpu_dp->orig_ethtool_ops; - struct dsa_switch *ds = cpu_dp->ds; - int port = cpu_dp->index; - int len = ETH_GSTRING_LEN; - int mcount = 0, count, i; - u8 pfx[4], *ndata; - - snprintf(pfx, sizeof(pfx), "p%.2d", port); - /* We do not want to be NULL-terminated, since this is a prefix */ - pfx[sizeof(pfx) - 1] = '_'; + struct dsa_switch_tree *dst = cpu_dp->dst; + int count, mcount = 0; netdev_lock_ops(dev); if (stringset == ETH_SS_PHY_STATS && dev->phydev && @@ -191,21 +255,17 @@ static void dsa_conduit_get_strings(struct net_device *dev, u32 stringset, } netdev_unlock_ops(dev); - if (ds->ops->get_strings) { - ndata = data + mcount * len; - /* This function copies ETH_GSTRINGS_LEN bytes, we will mangle - * the output after to prepend our CPU port prefix we - * constructed earlier - */ - ds->ops->get_strings(ds, port, stringset, ndata); - count = ds->ops->get_sset_count(ds, port, stringset); + list_for_each_entry(dp, &dst->ports, list) { + if (!dsa_port_is_dsa(dp) && !dsa_port_is_cpu(dp)) + continue; + + count = dsa_conduit_append_port_strings(dp->ds, dp->index, + stringset, data, + mcount); if (count < 0) return; - for (i = 0; i < count; i++) { - memmove(ndata + (i * len + sizeof(pfx)), - ndata + i * len, len - sizeof(pfx)); - memcpy(ndata + i * len, pfx, sizeof(pfx)); - } + + mcount += count; } } From 6633df05f3ade81474d55acd712431b63182d858 Mon Sep 17 00:00:00 2001 From: Wei Fang Date: Wed, 19 Nov 2025 18:25:55 +0800 Subject: [PATCH 1118/1381] net: enetc: set the external PHY address in IERB for port MDIO usage The ENETC supports managing its own external PHY through its port MDIO functionality. To use this function, the PHY address needs be set in the corresponding LaBCR register in the Integrated Endpoint Register Block (IERB), which is used for pre-boot initialization of NETC PCIe functions. The port MDIO can only work properly when the PHY address accessed by the port MDIO matches the corresponding LaBCR[MDIO_PHYAD_PRTAD] value. Because the ENETC driver only registers the MDIO bus (port MDIO bus) when it detects an MDIO child node in its node, similarly, the netc-blk-ctrl driver only resolves the PHY address and sets it in the corresponding LaBCR when it detects an MDIO child node in the ENETC node. Co-developed-by: Aziz Sellami Signed-off-by: Aziz Sellami Signed-off-by: Wei Fang Reviewed-by: Claudiu Manoil Link: https://patch.msgid.link/20251119102557.1041881-2-wei.fang@nxp.com Signed-off-by: Jakub Kicinski --- .../ethernet/freescale/enetc/netc_blk_ctrl.c | 141 +++++++++++++++++- 1 file changed, 140 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/freescale/enetc/netc_blk_ctrl.c b/drivers/net/ethernet/freescale/enetc/netc_blk_ctrl.c index d7aee3c934d3..6dd54b0d9616 100644 --- a/drivers/net/ethernet/freescale/enetc/netc_blk_ctrl.c +++ b/drivers/net/ethernet/freescale/enetc/netc_blk_ctrl.c @@ -67,6 +67,9 @@ #define IERB_EMDIOFAUXR 0x344 #define IERB_T0FAUXR 0x444 #define IERB_ETBCR(a) (0x300c + 0x100 * (a)) +#define IERB_LBCR(a) (0x1010 + 0x40 * (a)) +#define LBCR_MDIO_PHYAD_PRTAD(addr) (((addr) & 0x1f) << 8) + #define IERB_EFAUXR(a) (0x3044 + 0x100 * (a)) #define IERB_VFAUXR(a) (0x4004 + 0x40 * (a)) #define FAUXR_LDID GENMASK(3, 0) @@ -322,6 +325,142 @@ static int netc_unlock_ierb_with_warm_reset(struct netc_blk_ctrl *priv) 1000, 100000, true, priv->prb, PRB_NETCRR); } +static int netc_get_phy_addr(struct device_node *np) +{ + struct device_node *mdio_node, *phy_node; + u32 addr = 0; + int err = 0; + + mdio_node = of_get_child_by_name(np, "mdio"); + if (!mdio_node) + return 0; + + phy_node = of_get_next_child(mdio_node, NULL); + if (!phy_node) + goto of_put_mdio_node; + + err = of_property_read_u32(phy_node, "reg", &addr); + if (err) + goto of_put_phy_node; + + if (addr >= PHY_MAX_ADDR) + err = -EINVAL; + +of_put_phy_node: + of_node_put(phy_node); + +of_put_mdio_node: + of_node_put(mdio_node); + + return err ? err : addr; +} + +static int netc_parse_emdio_phy_mask(struct device_node *np, u32 *phy_mask) +{ + u32 mask = 0; + + for_each_child_of_node_scoped(np, child) { + u32 addr; + int err; + + err = of_property_read_u32(child, "reg", &addr); + if (err) + return err; + + if (addr >= PHY_MAX_ADDR) + return -EINVAL; + + mask |= BIT(addr); + } + + *phy_mask = mask; + + return 0; +} + +static int netc_get_emdio_phy_mask(struct device_node *np, u32 *phy_mask) +{ + for_each_child_of_node_scoped(np, child) { + for_each_child_of_node_scoped(child, gchild) { + if (!of_device_is_compatible(gchild, "pci1131,ee00")) + continue; + + return netc_parse_emdio_phy_mask(gchild, phy_mask); + } + } + + return 0; +} + +static int imx95_enetc_mdio_phyaddr_config(struct platform_device *pdev) +{ + struct netc_blk_ctrl *priv = platform_get_drvdata(pdev); + struct device_node *np = pdev->dev.of_node; + struct device *dev = &pdev->dev; + int bus_devfn, addr, err; + u32 phy_mask = 0; + + err = netc_get_emdio_phy_mask(np, &phy_mask); + if (err) { + dev_err(dev, "Failed to get PHY address mask\n"); + return err; + } + + /* Update the port EMDIO PHY address through parsing phy properties. + * This is needed when using the port EMDIO but it's harmless when + * using the central EMDIO. So apply it on all cases. + */ + for_each_child_of_node_scoped(np, child) { + for_each_child_of_node_scoped(child, gchild) { + if (!of_device_is_compatible(gchild, "pci1131,e101")) + continue; + + bus_devfn = netc_of_pci_get_bus_devfn(gchild); + if (bus_devfn < 0) { + dev_err(dev, "Failed to get BDF number\n"); + return bus_devfn; + } + + addr = netc_get_phy_addr(gchild); + if (addr < 0) { + dev_err(dev, "Failed to get PHY address\n"); + return addr; + } + + if (phy_mask & BIT(addr)) { + dev_err(dev, + "Find same PHY address in EMDIO and ENETC node\n"); + return -EINVAL; + } + + /* The default value of LaBCR[MDIO_PHYAD_PRTAD ] is + * 0, so no need to set the register. + */ + if (!addr) + continue; + + switch (bus_devfn) { + case IMX95_ENETC0_BUS_DEVFN: + netc_reg_write(priv->ierb, IERB_LBCR(0), + LBCR_MDIO_PHYAD_PRTAD(addr)); + break; + case IMX95_ENETC1_BUS_DEVFN: + netc_reg_write(priv->ierb, IERB_LBCR(1), + LBCR_MDIO_PHYAD_PRTAD(addr)); + break; + case IMX95_ENETC2_BUS_DEVFN: + netc_reg_write(priv->ierb, IERB_LBCR(2), + LBCR_MDIO_PHYAD_PRTAD(addr)); + break; + default: + break; + } + } + } + + return 0; +} + static int imx95_ierb_init(struct platform_device *pdev) { struct netc_blk_ctrl *priv = platform_get_drvdata(pdev); @@ -349,7 +488,7 @@ static int imx95_ierb_init(struct platform_device *pdev) /* NETC TIMER */ netc_reg_write(priv->ierb, IERB_T0FAUXR, 7); - return 0; + return imx95_enetc_mdio_phyaddr_config(pdev); } static int imx94_get_enetc_id(struct device_node *np) From 50bfd9c06f0ff80e3ab6cfe6169d5ae2fe1afaa2 Mon Sep 17 00:00:00 2001 From: Wei Fang Date: Wed, 19 Nov 2025 18:25:56 +0800 Subject: [PATCH 1119/1381] net: enetc: set external PHY address in IERB for i.MX94 ENETC NETC IP has only one external master MDIO interface (eMDIO) for managing the external PHYs. ENETC can use the interfaces provided by the EMDIO function or its port MDIO to access and manage its external PHY. Both the EMDIO function and the port MDIO are all virtual ports of the eMDIO. The difference is that the EMDIO function is a 'global port', it can access all the PHYs on the eMDIO, but port MDIO can only access its own PHY. To ensure that ENETC can only access its own PHY through port MDIO, LaBCR[MDIO_PHYAD_PRTAD] needs to be set, which represents the address of the external PHY connected to ENETC. If the accessed PHY address is not consistent with LaBCR[MDIO_PHYAD_PRTAD], then the MDIO access initiated by port MDIO will be invalid. Signed-off-by: Wei Fang Reviewed-by: Claudiu Manoil Link: https://patch.msgid.link/20251119102557.1041881-3-wei.fang@nxp.com Signed-off-by: Jakub Kicinski --- .../ethernet/freescale/enetc/netc_blk_ctrl.c | 57 +++++++++++++++++++ 1 file changed, 57 insertions(+) diff --git a/drivers/net/ethernet/freescale/enetc/netc_blk_ctrl.c b/drivers/net/ethernet/freescale/enetc/netc_blk_ctrl.c index 6dd54b0d9616..443983fdecd9 100644 --- a/drivers/net/ethernet/freescale/enetc/netc_blk_ctrl.c +++ b/drivers/net/ethernet/freescale/enetc/netc_blk_ctrl.c @@ -563,12 +563,64 @@ static int imx94_enetc_update_tid(struct netc_blk_ctrl *priv, return 0; } +static int imx94_enetc_mdio_phyaddr_config(struct netc_blk_ctrl *priv, + struct device_node *np, + u32 phy_mask) +{ + struct device *dev = &priv->pdev->dev; + int bus_devfn, addr; + + bus_devfn = netc_of_pci_get_bus_devfn(np); + if (bus_devfn < 0) { + dev_err(dev, "Failed to get BDF number\n"); + return bus_devfn; + } + + addr = netc_get_phy_addr(np); + if (addr <= 0) { + dev_err(dev, "Failed to get PHY address\n"); + return addr; + } + + if (phy_mask & BIT(addr)) { + dev_err(dev, + "Find same PHY address in EMDIO and ENETC node\n"); + return -EINVAL; + } + + switch (bus_devfn) { + case IMX94_ENETC0_BUS_DEVFN: + netc_reg_write(priv->ierb, IERB_LBCR(IMX94_ENETC0_LINK), + LBCR_MDIO_PHYAD_PRTAD(addr)); + break; + case IMX94_ENETC1_BUS_DEVFN: + netc_reg_write(priv->ierb, IERB_LBCR(IMX94_ENETC1_LINK), + LBCR_MDIO_PHYAD_PRTAD(addr)); + break; + case IMX94_ENETC2_BUS_DEVFN: + netc_reg_write(priv->ierb, IERB_LBCR(IMX94_ENETC2_LINK), + LBCR_MDIO_PHYAD_PRTAD(addr)); + break; + default: + break; + } + + return 0; +} + static int imx94_ierb_init(struct platform_device *pdev) { struct netc_blk_ctrl *priv = platform_get_drvdata(pdev); struct device_node *np = pdev->dev.of_node; + u32 phy_mask = 0; int err; + err = netc_get_emdio_phy_mask(np, &phy_mask); + if (err) { + dev_err(&pdev->dev, "Failed to get PHY address mask\n"); + return err; + } + for_each_child_of_node_scoped(np, child) { for_each_child_of_node_scoped(child, gchild) { if (!of_device_is_compatible(gchild, "pci1131,e101")) @@ -577,6 +629,11 @@ static int imx94_ierb_init(struct platform_device *pdev) err = imx94_enetc_update_tid(priv, gchild); if (err) return err; + + err = imx94_enetc_mdio_phyaddr_config(priv, gchild, + phy_mask); + if (err) + return err; } } From 10ba23a7f6cc4afbe7f1740b12b1ceb55fc57c00 Mon Sep 17 00:00:00 2001 From: Wei Fang Date: Wed, 19 Nov 2025 18:25:57 +0800 Subject: [PATCH 1120/1381] net: enetc: update the base address of port MDIO registers for ENETC v4 Each ENETC has a set of external MDIO registers to access its external PHY based on its port EMDIO bus, these registers are used for MDIO bus access, such as setting the PHY address, PHY register address and value, read or write operations, C22 or C45 format, etc. The base address of this set of registers has been modified in ENETC v4 and is different from that in ENETC v1. So the base address needs to be updated so that ENETC v4 can use port MDIO to manage its own external PHY. Additionally, if ENETC has the PCS layer, it also has a set of internal MDIO registers for managing its on-die PHY (PCS/Serdes). The base address of this set of registers is also different from that of ENETC v1, so the base address also needs to be updated so that ENETC v4 can support the management of on-die PHY through the internal MDIO bus. Signed-off-by: Wei Fang Reviewed-by: Claudiu Manoil Link: https://patch.msgid.link/20251119102557.1041881-4-wei.fang@nxp.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/freescale/enetc/enetc4_hw.h | 6 ++++++ .../net/ethernet/freescale/enetc/enetc_pf_common.c | 14 ++++++++++++-- 2 files changed, 18 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/freescale/enetc/enetc4_hw.h b/drivers/net/ethernet/freescale/enetc/enetc4_hw.h index ebea4298791c..3ed0f7a02767 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc4_hw.h +++ b/drivers/net/ethernet/freescale/enetc/enetc4_hw.h @@ -170,6 +170,9 @@ /* Port MAC 0/1 Maximum Frame Length Register */ #define ENETC4_PM_MAXFRM(mac) (0x5014 + (mac) * 0x400) +/* Port internal MDIO base address, use to access PCS */ +#define ENETC4_PM_IMDIO_BASE 0x5030 + /* Port MAC 0/1 Pause Quanta Register */ #define ENETC4_PM_PAUSE_QUANTA(mac) (0x5054 + (mac) * 0x400) @@ -198,6 +201,9 @@ #define SSP_1G 2 #define PM_IF_MODE_ENA BIT(15) +/* Port external MDIO Base address, use to access off-chip PHY */ +#define ENETC4_EMDIO_BASE 0x5c00 + /**********************ENETC Pseudo MAC port registers************************/ /* Port pseudo MAC receive octets counter (64-bit) */ #define ENETC4_PPMROCR 0x5080 diff --git a/drivers/net/ethernet/freescale/enetc/enetc_pf_common.c b/drivers/net/ethernet/freescale/enetc/enetc_pf_common.c index 9c634205e2a7..76263b8566bb 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc_pf_common.c +++ b/drivers/net/ethernet/freescale/enetc/enetc_pf_common.c @@ -176,7 +176,12 @@ static int enetc_mdio_probe(struct enetc_pf *pf, struct device_node *np) bus->parent = dev; mdio_priv = bus->priv; mdio_priv->hw = &pf->si->hw; - mdio_priv->mdio_base = ENETC_EMDIO_BASE; + + if (is_enetc_rev1(pf->si)) + mdio_priv->mdio_base = ENETC_EMDIO_BASE; + else + mdio_priv->mdio_base = ENETC4_EMDIO_BASE; + snprintf(bus->id, MII_BUS_ID_SIZE, "%s", dev_name(dev)); err = of_mdiobus_register(bus, np); @@ -221,7 +226,12 @@ static int enetc_imdio_create(struct enetc_pf *pf) bus->phy_mask = ~0; mdio_priv = bus->priv; mdio_priv->hw = &pf->si->hw; - mdio_priv->mdio_base = ENETC_PM_IMDIO_BASE; + + if (is_enetc_rev1(pf->si)) + mdio_priv->mdio_base = ENETC_PM_IMDIO_BASE; + else + mdio_priv->mdio_base = ENETC4_PM_IMDIO_BASE; + snprintf(bus->id, MII_BUS_ID_SIZE, "%s-imdio", dev_name(dev)); err = mdiobus_register(bus); From a11e0d467da2dcbe3f199a0f94769ed333b88288 Mon Sep 17 00:00:00 2001 From: Buday Csaba Date: Mon, 24 Nov 2025 08:19:15 +0100 Subject: [PATCH 1121/1381] net: mdio: eliminate kdoc warnings in mdio_device.c and mdio_bus.c Fix all warnings reported by scripts/kernel-doc in mdio_device.c and mdio_bus.c Signed-off-by: Buday Csaba Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/7ef7b80669da2b899d38afdb6c45e122229c3d8c.1763968667.git.buday.csaba@prolan.hu Signed-off-by: Jakub Kicinski --- drivers/net/phy/mdio_bus.c | 56 ++++++++++++++++++++++++++++++----- drivers/net/phy/mdio_device.c | 6 ++++ 2 files changed, 55 insertions(+), 7 deletions(-) diff --git a/drivers/net/phy/mdio_bus.c b/drivers/net/phy/mdio_bus.c index ef041ad66852..afdf1ad6c0e6 100644 --- a/drivers/net/phy/mdio_bus.c +++ b/drivers/net/phy/mdio_bus.c @@ -339,7 +339,7 @@ EXPORT_SYMBOL_GPL(mdio_bus_class); * mdio_find_bus - Given the name of a mdiobus, find the mii_bus. * @mdio_name: The name of a mdiobus. * - * Returns a reference to the mii_bus, or NULL if none found. The + * Return: a reference to the mii_bus, or NULL if none found. The * embedded struct device will have its reference count incremented, * and this must be put_deviced'ed once the bus is finished with. */ @@ -357,7 +357,7 @@ EXPORT_SYMBOL(mdio_find_bus); * of_mdio_find_bus - Given an mii_bus node, find the mii_bus. * @mdio_bus_np: Pointer to the mii_bus. * - * Returns a reference to the mii_bus, or NULL if none found. The + * Return: a reference to the mii_bus, or NULL if none found. The * embedded struct device will have its reference count incremented, * and this must be put once the bus is finished with. * @@ -405,6 +405,8 @@ static void mdiobus_stats_acct(struct mdio_bus_stats *stats, bool op, int ret) * @addr: the phy address * @regnum: register number to read * + * Return: The register value if successful, negative error code on failure + * * Read a MDIO bus register. Caller must hold the mdio bus lock. * * NOTE: MUST NOT be called from interrupt context. @@ -437,6 +439,8 @@ EXPORT_SYMBOL(__mdiobus_read); * @regnum: register number to write * @val: value to write to @regnum * + * Return: Zero if successful, negative error code on failure + * * Write a MDIO bus register. Caller must hold the mdio bus lock. * * NOTE: MUST NOT be called from interrupt context. @@ -470,8 +474,11 @@ EXPORT_SYMBOL(__mdiobus_write); * @mask: bit mask of bits to clear * @set: bit mask of bits to set * + * Return: 1 if the register was modified, 0 if no change was needed, + * negative on any error condition + * * Read, modify, and if any change, write the register value back to the - * device. Any error returns a negative number. + * device. * * NOTE: MUST NOT be called from interrupt context. */ @@ -501,6 +508,8 @@ EXPORT_SYMBOL_GPL(__mdiobus_modify_changed); * @devad: device address to read * @regnum: register number to read * + * Return: The register value if successful, negative error code on failure + * * Read a MDIO bus register. Caller must hold the mdio bus lock. * * NOTE: MUST NOT be called from interrupt context. @@ -534,6 +543,8 @@ EXPORT_SYMBOL(__mdiobus_c45_read); * @regnum: register number to write * @val: value to write to @regnum * + * Return: Zero if successful, negative error code on failure + * * Write a MDIO bus register. Caller must hold the mdio bus lock. * * NOTE: MUST NOT be called from interrupt context. @@ -569,6 +580,9 @@ EXPORT_SYMBOL(__mdiobus_c45_write); * @mask: bit mask of bits to clear * @set: bit mask of bits to set * + * Return: 1 if the register was modified, 0 if no change was needed, + * negative on any error condition + * * Read, modify, and if any change, write the register value back to the * device. Any error returns a negative number. * @@ -599,6 +613,8 @@ static int __mdiobus_c45_modify_changed(struct mii_bus *bus, int addr, * @addr: the phy address * @regnum: register number to read * + * Return: The register value if successful, negative error code on failure + * * In case of nested MDIO bus access avoid lockdep false positives by * using mutex_lock_nested(). * @@ -624,6 +640,8 @@ EXPORT_SYMBOL(mdiobus_read_nested); * @addr: the phy address * @regnum: register number to read * + * Return: The register value if successful, negative error code on failure + * * NOTE: MUST NOT be called from interrupt context, * because the bus read/write functions may wait for an interrupt * to conclude the operation. @@ -647,6 +665,8 @@ EXPORT_SYMBOL(mdiobus_read); * @devad: device address to read * @regnum: register number to read * + * Return: The register value if successful, negative error code on failure + * * NOTE: MUST NOT be called from interrupt context, * because the bus read/write functions may wait for an interrupt * to conclude the operation. @@ -670,6 +690,8 @@ EXPORT_SYMBOL(mdiobus_c45_read); * @devad: device address to read * @regnum: register number to read * + * Return: The register value if successful, negative error code on failure + * * In case of nested MDIO bus access avoid lockdep false positives by * using mutex_lock_nested(). * @@ -697,6 +719,8 @@ EXPORT_SYMBOL(mdiobus_c45_read_nested); * @regnum: register number to write * @val: value to write to @regnum * + * Return: Zero if successful, negative error code on failure + * * In case of nested MDIO bus access avoid lockdep false positives by * using mutex_lock_nested(). * @@ -723,6 +747,8 @@ EXPORT_SYMBOL(mdiobus_write_nested); * @regnum: register number to write * @val: value to write to @regnum * + * Return: Zero if successful, negative error code on failure + * * NOTE: MUST NOT be called from interrupt context, * because the bus read/write functions may wait for an interrupt * to conclude the operation. @@ -747,6 +773,8 @@ EXPORT_SYMBOL(mdiobus_write); * @regnum: register number to write * @val: value to write to @regnum * + * Return: Zero if successful, negative error code on failure + * * NOTE: MUST NOT be called from interrupt context, * because the bus read/write functions may wait for an interrupt * to conclude the operation. @@ -772,6 +800,8 @@ EXPORT_SYMBOL(mdiobus_c45_write); * @regnum: register number to write * @val: value to write to @regnum * + * Return: Zero if successful, negative error code on failure + * * In case of nested MDIO bus access avoid lockdep false positives by * using mutex_lock_nested(). * @@ -800,6 +830,8 @@ EXPORT_SYMBOL(mdiobus_c45_write_nested); * @regnum: register number to write * @mask: bit mask of bits to clear * @set: bit mask of bits to set + * + * Return: 0 on success, negative on any error condition */ int __mdiobus_modify(struct mii_bus *bus, int addr, u32 regnum, u16 mask, u16 set) @@ -820,6 +852,8 @@ EXPORT_SYMBOL_GPL(__mdiobus_modify); * @regnum: register number to write * @mask: bit mask of bits to clear * @set: bit mask of bits to set + * + * Return: 0 on success, negative on any error condition */ int mdiobus_modify(struct mii_bus *bus, int addr, u32 regnum, u16 mask, u16 set) { @@ -842,6 +876,8 @@ EXPORT_SYMBOL_GPL(mdiobus_modify); * @regnum: register number to write * @mask: bit mask of bits to clear * @set: bit mask of bits to set + * + * Return: 0 on success, negative on any error condition */ int mdiobus_c45_modify(struct mii_bus *bus, int addr, int devad, u32 regnum, u16 mask, u16 set) @@ -865,6 +901,9 @@ EXPORT_SYMBOL_GPL(mdiobus_c45_modify); * @regnum: register number to write * @mask: bit mask of bits to clear * @set: bit mask of bits to set + * + * Return: 1 if the register was modified, 0 if no change was needed, + * negative on any error condition */ int mdiobus_modify_changed(struct mii_bus *bus, int addr, u32 regnum, u16 mask, u16 set) @@ -888,6 +927,9 @@ EXPORT_SYMBOL_GPL(mdiobus_modify_changed); * @regnum: register number to write * @mask: bit mask of bits to clear * @set: bit mask of bits to set + * + * Return: 1 if the register was modified, 0 if no change was needed, + * negative on any error condition */ int mdiobus_c45_modify_changed(struct mii_bus *bus, int addr, int devad, u32 regnum, u16 mask, u16 set) @@ -908,10 +950,10 @@ EXPORT_SYMBOL_GPL(mdiobus_c45_modify_changed); * @dev: target MDIO device * @drv: given MDIO driver * - * Description: Given a MDIO device, and a MDIO driver, return 1 if - * the driver supports the device. Otherwise, return 0. This may - * require calling the devices own match function, since different classes - * of MDIO devices have different match criteria. + * Return: 1 if the driver supports the device, 0 otherwise + * + * Description: This may require calling the devices own match function, + * since different classes of MDIO devices have different match criteria. */ static int mdio_bus_match(struct device *dev, const struct device_driver *drv) { diff --git a/drivers/net/phy/mdio_device.c b/drivers/net/phy/mdio_device.c index fd0e16dbc5cc..6e90ed42cd98 100644 --- a/drivers/net/phy/mdio_device.c +++ b/drivers/net/phy/mdio_device.c @@ -78,6 +78,8 @@ EXPORT_SYMBOL(mdio_device_create); /** * mdio_device_register - Register the mdio device on the MDIO bus * @mdiodev: mdio_device structure to be added to the MDIO bus + * + * Return: Zero if successful, negative error code on failure */ int mdio_device_register(struct mdio_device *mdiodev) { @@ -206,6 +208,8 @@ EXPORT_SYMBOL(mdio_device_reset); * * Description: Take care of setting up the mdio_device structure * and calling the driver to probe the device. + * + * Return: Zero if successful, negative error code on failure */ static int mdio_probe(struct device *dev) { @@ -256,6 +260,8 @@ static void mdio_shutdown(struct device *dev) /** * mdio_driver_register - register an mdio_driver with the MDIO layer * @drv: new mdio_driver to register + * + * Return: Zero if successful, negative error code on failure */ int mdio_driver_register(struct mdio_driver *drv) { From ce28e333d6286cd10a27c8f22d3ac2b758f5fa0c Mon Sep 17 00:00:00 2001 From: Buday Csaba Date: Mon, 24 Nov 2025 15:50:44 +0100 Subject: [PATCH 1122/1381] net: mdio: remove redundant fwnode cleanup Remove redundant fwnode cleanup in of_mdiobus_register_device() and xpcs_plat_init_dev(). mdio_device_free() eventually calls mdio_device_release(), which already performs fwnode_handle_put(), making the manual cleanup unnecessary. Combine fwnode_handle_get() with device_set_node() in of_mdiobus_register_device() for clarity. Signed-off-by: Buday Csaba Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/00847693daa8f7c8ff5dfa19dd35fc712fa4e2b5.1763995734.git.buday.csaba@prolan.hu Signed-off-by: Jakub Kicinski --- drivers/net/mdio/of_mdio.c | 5 +---- drivers/net/pcs/pcs-xpcs-plat.c | 3 --- 2 files changed, 1 insertion(+), 7 deletions(-) diff --git a/drivers/net/mdio/of_mdio.c b/drivers/net/mdio/of_mdio.c index 1357348e01d5..b8d298c04d3f 100644 --- a/drivers/net/mdio/of_mdio.c +++ b/drivers/net/mdio/of_mdio.c @@ -63,14 +63,11 @@ static int of_mdiobus_register_device(struct mii_bus *mdio, /* Associate the OF node with the device structure so it * can be looked up later. */ - fwnode_handle_get(fwnode); - device_set_node(&mdiodev->dev, fwnode); + device_set_node(&mdiodev->dev, fwnode_handle_get(fwnode)); /* All data is now stored in the mdiodev struct; register it. */ rc = mdio_device_register(mdiodev); if (rc) { - device_set_node(&mdiodev->dev, NULL); - fwnode_handle_put(fwnode); mdio_device_free(mdiodev); return rc; } diff --git a/drivers/net/pcs/pcs-xpcs-plat.c b/drivers/net/pcs/pcs-xpcs-plat.c index c422e8d8b89f..b8c48f9effbf 100644 --- a/drivers/net/pcs/pcs-xpcs-plat.c +++ b/drivers/net/pcs/pcs-xpcs-plat.c @@ -365,9 +365,6 @@ static int xpcs_plat_init_dev(struct dw_xpcs_plat *pxpcs) err_clean_data: mdiodev->dev.platform_data = NULL; - fwnode_handle_put(dev_fwnode(&mdiodev->dev)); - device_set_node(&mdiodev->dev, NULL); - mdio_device_free(mdiodev); return ret; From 9d844da693d6d0813714d9b5b7a58ac05c4cf7f0 Mon Sep 17 00:00:00 2001 From: Chad Monroe Date: Sat, 22 Nov 2025 13:32:15 +0000 Subject: [PATCH 1123/1381] net: phy: mxl-gpy: add support for MxL86211C MxL86211C is a smaller and more efficient version of the GPY211C. Add the PHY ID and phy_driver instance to the mxl-gpy driver. Signed-off-by: Chad Monroe Signed-off-by: Daniel Golle Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/cabf3559d6511bed6b8a925f540e3162efc20f6b.1763818120.git.daniel@makrotopia.org Signed-off-by: Jakub Kicinski --- drivers/net/phy/mxl-gpy.c | 24 ++++++++++++++++++++++++ 1 file changed, 24 insertions(+) diff --git a/drivers/net/phy/mxl-gpy.c b/drivers/net/phy/mxl-gpy.c index 0c8dc16ee7bd..bea9072ac4a8 100644 --- a/drivers/net/phy/mxl-gpy.c +++ b/drivers/net/phy/mxl-gpy.c @@ -30,6 +30,7 @@ #define PHY_ID_GPY241B 0x67C9DE40 #define PHY_ID_GPY241BM 0x67C9DE80 #define PHY_ID_GPY245B 0x67C9DEC0 +#define PHY_ID_MXL86211C 0xC1335400 #define PHY_CTL1 0x13 #define PHY_CTL1_MDICD BIT(3) @@ -1268,6 +1269,28 @@ static struct phy_driver gpy_drivers[] = { .get_wol = gpy_get_wol, .set_loopback = gpy_loopback, }, + { + PHY_ID_MATCH_MODEL(PHY_ID_MXL86211C), + .name = "Maxlinear Ethernet MxL86211C", + .get_features = genphy_c45_pma_read_abilities, + .config_init = gpy_config_init, + .probe = gpy_probe, + .suspend = genphy_suspend, + .resume = genphy_resume, + .config_aneg = gpy_config_aneg, + .aneg_done = genphy_c45_aneg_done, + .read_status = gpy_read_status, + .config_intr = gpy_config_intr, + .handle_interrupt = gpy_handle_interrupt, + .set_wol = gpy_set_wol, + .get_wol = gpy_get_wol, + .set_loopback = gpy_loopback, + .led_brightness_set = gpy_led_brightness_set, + .led_hw_is_supported = gpy_led_hw_is_supported, + .led_hw_control_get = gpy_led_hw_control_get, + .led_hw_control_set = gpy_led_hw_control_set, + .led_polarity_set = gpy_led_polarity_set, + }, }; module_phy_driver(gpy_drivers); @@ -1284,6 +1307,7 @@ static const struct mdio_device_id __maybe_unused gpy_tbl[] = { {PHY_ID_MATCH_MODEL(PHY_ID_GPY241B)}, {PHY_ID_MATCH_MODEL(PHY_ID_GPY241BM)}, {PHY_ID_MATCH_MODEL(PHY_ID_GPY245B)}, + {PHY_ID_MATCH_MODEL(PHY_ID_MXL86211C)}, { } }; MODULE_DEVICE_TABLE(mdio, gpy_tbl); From de1e5c9333f426348571f7a3b034f99490d3f926 Mon Sep 17 00:00:00 2001 From: Daniel Golle Date: Sat, 22 Nov 2025 13:33:47 +0000 Subject: [PATCH 1124/1381] net: phy: mxl-gpy: add support for MxL86252 and MxL86282 Add PHY driver support for Maxlinear MxL86252 and MxL86282 switches. The PHYs built-into those switches are just like any other GPY 2.5G PHYs with the exception of the temperature sensor data being encoded in a different way. Signed-off-by: Daniel Golle Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/a6cd7fe461b011cec2b59dffaf34e9c8b0819059.1763818120.git.daniel@makrotopia.org Signed-off-by: Jakub Kicinski --- drivers/net/phy/mxl-gpy.c | 91 ++++++++++++++++++++++++++++++++++++++- 1 file changed, 89 insertions(+), 2 deletions(-) diff --git a/drivers/net/phy/mxl-gpy.c b/drivers/net/phy/mxl-gpy.c index bea9072ac4a8..58d8dd718445 100644 --- a/drivers/net/phy/mxl-gpy.c +++ b/drivers/net/phy/mxl-gpy.c @@ -31,6 +31,8 @@ #define PHY_ID_GPY241BM 0x67C9DE80 #define PHY_ID_GPY245B 0x67C9DEC0 #define PHY_ID_MXL86211C 0xC1335400 +#define PHY_ID_MXL86252 0xC1335520 +#define PHY_ID_MXL86282 0xC1335500 #define PHY_CTL1 0x13 #define PHY_CTL1_MDICD BIT(3) @@ -200,6 +202,29 @@ static int gpy_hwmon_read(struct device *dev, return 0; } +static int mxl862x2_hwmon_read(struct device *dev, + enum hwmon_sensor_types type, + u32 attr, int channel, long *value) +{ + struct phy_device *phydev = dev_get_drvdata(dev); + long tmp; + int ret; + + ret = phy_read_mmd(phydev, MDIO_MMD_VEND1, VSPEC1_TEMP_STA); + if (ret < 0) + return ret; + if (!ret) + return -ENODATA; + + tmp = (s16)ret; + tmp *= 78125; + tmp /= 10000; + + *value = tmp; + + return 0; +} + static umode_t gpy_hwmon_is_visible(const void *data, enum hwmon_sensor_types type, u32 attr, int channel) @@ -217,19 +242,35 @@ static const struct hwmon_ops gpy_hwmon_hwmon_ops = { .read = gpy_hwmon_read, }; +static const struct hwmon_ops mxl862x2_hwmon_hwmon_ops = { + .is_visible = gpy_hwmon_is_visible, + .read = mxl862x2_hwmon_read, +}; + static const struct hwmon_chip_info gpy_hwmon_chip_info = { .ops = &gpy_hwmon_hwmon_ops, .info = gpy_hwmon_info, }; +static const struct hwmon_chip_info mxl862x2_hwmon_chip_info = { + .ops = &mxl862x2_hwmon_hwmon_ops, + .info = gpy_hwmon_info, +}; + static int gpy_hwmon_register(struct phy_device *phydev) { struct device *dev = &phydev->mdio.dev; + const struct hwmon_chip_info *info; struct device *hwmon_dev; + if (phy_id_compare_model(phydev->phy_id, PHY_ID_MXL86252) || + phy_id_compare_model(phydev->phy_id, PHY_ID_MXL86282)) + info = &mxl862x2_hwmon_chip_info; + else + info = &gpy_hwmon_chip_info; + hwmon_dev = devm_hwmon_device_register_with_info(dev, NULL, phydev, - &gpy_hwmon_chip_info, - NULL); + info, NULL); return PTR_ERR_OR_ZERO(hwmon_dev); } @@ -1291,6 +1332,50 @@ static struct phy_driver gpy_drivers[] = { .led_hw_control_set = gpy_led_hw_control_set, .led_polarity_set = gpy_led_polarity_set, }, + { + PHY_ID_MATCH_MODEL(PHY_ID_MXL86252), + .name = "MaxLinear Ethernet MxL86252", + .get_features = genphy_c45_pma_read_abilities, + .config_init = gpy_config_init, + .probe = gpy_probe, + .suspend = genphy_suspend, + .resume = genphy_resume, + .config_aneg = gpy_config_aneg, + .aneg_done = genphy_c45_aneg_done, + .read_status = gpy_read_status, + .config_intr = gpy_config_intr, + .handle_interrupt = gpy_handle_interrupt, + .set_wol = gpy_set_wol, + .get_wol = gpy_get_wol, + .set_loopback = gpy_loopback, + .led_brightness_set = gpy_led_brightness_set, + .led_hw_is_supported = gpy_led_hw_is_supported, + .led_hw_control_get = gpy_led_hw_control_get, + .led_hw_control_set = gpy_led_hw_control_set, + .led_polarity_set = gpy_led_polarity_set, + }, + { + PHY_ID_MATCH_MODEL(PHY_ID_MXL86282), + .name = "MaxLinear Ethernet MxL86282", + .get_features = genphy_c45_pma_read_abilities, + .config_init = gpy_config_init, + .probe = gpy_probe, + .suspend = genphy_suspend, + .resume = genphy_resume, + .config_aneg = gpy_config_aneg, + .aneg_done = genphy_c45_aneg_done, + .read_status = gpy_read_status, + .config_intr = gpy_config_intr, + .handle_interrupt = gpy_handle_interrupt, + .set_wol = gpy_set_wol, + .get_wol = gpy_get_wol, + .set_loopback = gpy_loopback, + .led_brightness_set = gpy_led_brightness_set, + .led_hw_is_supported = gpy_led_hw_is_supported, + .led_hw_control_get = gpy_led_hw_control_get, + .led_hw_control_set = gpy_led_hw_control_set, + .led_polarity_set = gpy_led_polarity_set, + }, }; module_phy_driver(gpy_drivers); @@ -1308,6 +1393,8 @@ static const struct mdio_device_id __maybe_unused gpy_tbl[] = { {PHY_ID_MATCH_MODEL(PHY_ID_GPY241BM)}, {PHY_ID_MATCH_MODEL(PHY_ID_GPY245B)}, {PHY_ID_MATCH_MODEL(PHY_ID_MXL86211C)}, + {PHY_ID_MATCH_MODEL(PHY_ID_MXL86252)}, + {PHY_ID_MATCH_MODEL(PHY_ID_MXL86282)}, { } }; MODULE_DEVICE_TABLE(mdio, gpy_tbl); From 87ad869feaedb0ebd61b2d254f6952f7e4ceb624 Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Mon, 24 Nov 2025 08:37:53 +0100 Subject: [PATCH 1125/1381] r8169: improve MAC EEE handling Let phydev->enable_tx_lpi control whether MAC enables TX LPI, instead of enabling it unconditionally. This way TX LPI is disabled if e.g. link partner doesn't support EEE. This helps to avoid potential issues like link flaps. Signed-off-by: Heiner Kallweit Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/91bcb837-3fab-4b4e-b495-038df0932e44@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/realtek/r8169_main.c | 73 +++++++++++------------ 1 file changed, 36 insertions(+), 37 deletions(-) diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c index 9fd0ca399d5f..3f5ffaa55c85 100644 --- a/drivers/net/ethernet/realtek/r8169_main.c +++ b/drivers/net/ethernet/realtek/r8169_main.c @@ -2379,26 +2379,6 @@ void r8169_apply_firmware(struct rtl8169_private *tp) } } -static void rtl8168_config_eee_mac(struct rtl8169_private *tp) -{ - /* Adjust EEE LED frequency */ - if (tp->mac_version != RTL_GIGA_MAC_VER_38) - RTL_W8(tp, EEE_LED, RTL_R8(tp, EEE_LED) & ~0x07); - - rtl_eri_set_bits(tp, 0x1b0, 0x0003); -} - -static void rtl8125a_config_eee_mac(struct rtl8169_private *tp) -{ - r8168_mac_ocp_modify(tp, 0xe040, 0, BIT(1) | BIT(0)); - r8168_mac_ocp_modify(tp, 0xeb62, 0, BIT(2) | BIT(1)); -} - -static void rtl8125b_config_eee_mac(struct rtl8169_private *tp) -{ - r8168_mac_ocp_modify(tp, 0xe040, 0, BIT(1) | BIT(0)); -} - static void rtl_rar_exgmac_set(struct rtl8169_private *tp, const u8 *addr) { rtl_eri_write(tp, 0xe0, ERIAR_MASK_1111, get_unaligned_le32(addr)); @@ -3176,8 +3156,6 @@ static void rtl_hw_start_8168e_2(struct rtl8169_private *tp) RTL_W8(tp, MCU, RTL_R8(tp, MCU) & ~NOW_IS_OOB); - rtl8168_config_eee_mac(tp); - RTL_W8(tp, DLLPR, RTL_R8(tp, DLLPR) | PFM_EN); RTL_W32(tp, MISC, RTL_R32(tp, MISC) | PWM_EN); rtl_mod_config5(tp, Spi_en, 0); @@ -3202,8 +3180,6 @@ static void rtl_hw_start_8168f(struct rtl8169_private *tp) RTL_W8(tp, DLLPR, RTL_R8(tp, DLLPR) | PFM_EN); RTL_W32(tp, MISC, RTL_R32(tp, MISC) | PWM_EN); rtl_mod_config5(tp, Spi_en, 0); - - rtl8168_config_eee_mac(tp); } static void rtl_hw_start_8168f_1(struct rtl8169_private *tp) @@ -3253,8 +3229,6 @@ static void rtl_hw_start_8168g(struct rtl8169_private *tp) rtl_eri_write(tp, 0xc0, ERIAR_MASK_0011, 0x0000); rtl_eri_write(tp, 0xb8, ERIAR_MASK_0011, 0x0000); - rtl8168_config_eee_mac(tp); - rtl_w0w1_eri(tp, 0x2fc, 0x01, 0x06); rtl_eri_clear_bits(tp, 0x1b0, BIT(12)); @@ -3395,8 +3369,6 @@ static void rtl_hw_start_8168h_1(struct rtl8169_private *tp) rtl_eri_write(tp, 0xc0, ERIAR_MASK_0011, 0x0000); rtl_eri_write(tp, 0xb8, ERIAR_MASK_0011, 0x0000); - rtl8168_config_eee_mac(tp); - RTL_W8(tp, DLLPR, RTL_R8(tp, DLLPR) & ~PFM_EN); RTL_W8(tp, MISC_1, RTL_R8(tp, MISC_1) & ~PFM_D3COLD_EN); @@ -3444,8 +3416,6 @@ static void rtl_hw_start_8168ep(struct rtl8169_private *tp) rtl_eri_write(tp, 0xc0, ERIAR_MASK_0011, 0x0000); rtl_eri_write(tp, 0xb8, ERIAR_MASK_0011, 0x0000); - rtl8168_config_eee_mac(tp); - rtl_w0w1_eri(tp, 0x2fc, 0x01, 0x06); RTL_W8(tp, DLLPR, RTL_R8(tp, DLLPR) & ~TX_10M_PS_EN); @@ -3501,8 +3471,6 @@ static void rtl_hw_start_8117(struct rtl8169_private *tp) rtl_eri_write(tp, 0xc0, ERIAR_MASK_0011, 0x0000); rtl_eri_write(tp, 0xb8, ERIAR_MASK_0011, 0x0000); - rtl8168_config_eee_mac(tp); - RTL_W8(tp, DLLPR, RTL_R8(tp, DLLPR) & ~PFM_EN); RTL_W8(tp, MISC_1, RTL_R8(tp, MISC_1) & ~PFM_D3COLD_EN); @@ -3743,11 +3711,6 @@ static void rtl_hw_start_8125_common(struct rtl8169_private *tp) rtl_loop_wait_low(tp, &rtl_mac_ocp_e00e_cond, 1000, 10); - if (tp->mac_version == RTL_GIGA_MAC_VER_61) - rtl8125a_config_eee_mac(tp); - else - rtl8125b_config_eee_mac(tp); - rtl_disable_rxdvgate(tp); } @@ -4750,6 +4713,41 @@ static int rtl8169_poll(struct napi_struct *napi, int budget) return work_done; } +static void rtl_enable_tx_lpi(struct rtl8169_private *tp, bool enable) +{ + if (!rtl_supports_eee(tp)) + return; + + switch (tp->mac_version) { + case RTL_GIGA_MAC_VER_34 ... RTL_GIGA_MAC_VER_52: + /* Adjust EEE LED frequency */ + if (tp->mac_version != RTL_GIGA_MAC_VER_38) + RTL_W8(tp, EEE_LED, RTL_R8(tp, EEE_LED) & ~0x07); + if (enable) + rtl_eri_set_bits(tp, 0x1b0, 0x0003); + else + rtl_eri_clear_bits(tp, 0x1b0, 0x0003); + break; + case RTL_GIGA_MAC_VER_61: + if (enable) { + r8168_mac_ocp_modify(tp, 0xe040, 0, 0x0003); + r8168_mac_ocp_modify(tp, 0xeb62, 0, 0x0006); + } else { + r8168_mac_ocp_modify(tp, 0xe040, 0x0003, 0); + r8168_mac_ocp_modify(tp, 0xeb62, 0x0006, 0); + } + break; + case RTL_GIGA_MAC_VER_63 ... RTL_GIGA_MAC_VER_LAST: + if (enable) + r8168_mac_ocp_modify(tp, 0xe040, 0, 0x0003); + else + r8168_mac_ocp_modify(tp, 0xe040, 0x0003, 0); + break; + default: + break; + } +} + static void r8169_phylink_handler(struct net_device *ndev) { struct rtl8169_private *tp = netdev_priv(ndev); @@ -4757,6 +4755,7 @@ static void r8169_phylink_handler(struct net_device *ndev) if (netif_carrier_ok(ndev)) { rtl_link_chg_patch(tp); + rtl_enable_tx_lpi(tp, tp->phydev->enable_tx_lpi); pm_request_resume(d); } else { pm_runtime_idle(d); From b796632fc83c8f9e0d97443f4f389ef9dddadb11 Mon Sep 17 00:00:00 2001 From: Michal Luczaj Date: Sun, 23 Nov 2025 22:43:59 +0100 Subject: [PATCH 1126/1381] vsock/test: Extend transport change null-ptr-deref test syzkaller reported a lockdep lock order inversion warning[1] due to commit 687aa0c5581b ("vsock: Fix transport_* TOCTOU"). This was fixed in commit f7c877e75352 ("vsock: fix lock inversion in vsock_assign_transport()"). Redo syzkaller's repro by piggybacking on a somewhat related test implemented in commit 3a764d93385c ("vsock/test: Add test for null ptr deref when transport changes"). [1]: https://lore.kernel.org/netdev/68f6cdb0.a70a0220.205af.0039.GAE@google.com/ Signed-off-by: Michal Luczaj Reviewed-by: Stefano Garzarella Link: https://patch.msgid.link/20251123-vsock_test-linger-lockdep-warn-v1-1-4b1edf9d8cdc@rbox.co Signed-off-by: Jakub Kicinski --- tools/testing/vsock/vsock_test.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/tools/testing/vsock/vsock_test.c b/tools/testing/vsock/vsock_test.c index d4517386e551..9e1250790f33 100644 --- a/tools/testing/vsock/vsock_test.c +++ b/tools/testing/vsock/vsock_test.c @@ -2015,6 +2015,11 @@ static void test_stream_transport_change_client(const struct test_opts *opts) exit(EXIT_FAILURE); } + /* Although setting SO_LINGER does not affect the original test + * for null-ptr-deref, it may trigger a lockdep warning. + */ + enable_so_linger(s, 1); + ret = connect(s, (struct sockaddr *)&sa, sizeof(sa)); /* The connect can fail due to signals coming from the thread, * or because the receiver connection queue is full. @@ -2352,7 +2357,7 @@ static struct test_case test_cases[] = { .run_server = test_stream_nolinger_server, }, { - .name = "SOCK_STREAM transport change null-ptr-deref", + .name = "SOCK_STREAM transport change null-ptr-deref, lockdep warn", .run_client = test_stream_transport_change_client, .run_server = test_stream_transport_change_server, }, From 622cc66ed72c2da2294de21da5976a087bcc5748 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Mon, 24 Nov 2025 09:45:45 +0100 Subject: [PATCH 1127/1381] ptp: ocp: Refactor signal_show() and fix %ptT misuse Refactor signal_show() to avoid sequential calls to sysfs_emit*() and use the same pattern to get the index of a signal as it's done in signal_store(). While at it, fix wrong use of %ptT against struct timespec64. It's kinda lucky that it worked just because the first member there 64-bit and it's of time64_t type. Now with %ptS it may be used correctly. Reviewed-by: Vadim Fedorenko Signed-off-by: Andy Shevchenko Link: https://patch.msgid.link/20251124084816.205035-2-andriy.shevchenko@linux.intel.com Signed-off-by: Jakub Kicinski --- drivers/ptp/ptp_ocp.c | 14 +++++--------- 1 file changed, 5 insertions(+), 9 deletions(-) diff --git a/drivers/ptp/ptp_ocp.c b/drivers/ptp/ptp_ocp.c index eeebe4d149f7..95889f85ffb2 100644 --- a/drivers/ptp/ptp_ocp.c +++ b/drivers/ptp/ptp_ocp.c @@ -3250,20 +3250,16 @@ signal_show(struct device *dev, struct device_attribute *attr, char *buf) struct dev_ext_attribute *ea = to_ext_attr(attr); struct ptp_ocp *bp = dev_get_drvdata(dev); struct ptp_ocp_signal *signal; + int gen = (uintptr_t)ea->var; struct timespec64 ts; - ssize_t count; - int i; - i = (uintptr_t)ea->var; - signal = &bp->signal[i]; - - count = sysfs_emit(buf, "%llu %d %llu %d", signal->period, - signal->duty, signal->phase, signal->polarity); + signal = &bp->signal[gen]; ts = ktime_to_timespec64(signal->start); - count += sysfs_emit_at(buf, count, " %ptT TAI\n", &ts); - return count; + return sysfs_emit(buf, "%llu %d %llu %d %ptT TAI\n", + signal->period, signal->duty, signal->phase, signal->polarity, + &ts.tv_sec); } static EXT_ATTR_RW(signal, signal, 0); static EXT_ATTR_RW(signal, signal, 1); From 590f5d1fa6ee22b1ec07283d5fa870326f47b2f6 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Mon, 24 Nov 2025 09:45:46 +0100 Subject: [PATCH 1128/1381] ptp: ocp: Make ptp_ocp_unregister_ext() NULL-aware It's a common practice to make resource release functions be NULL-aware. Make ptp_ocp_unregister_ext() NULL-aware. Reviewed-by: Vadim Fedorenko Signed-off-by: Andy Shevchenko Link: https://patch.msgid.link/20251124084816.205035-3-andriy.shevchenko@linux.intel.com Signed-off-by: Jakub Kicinski --- drivers/ptp/ptp_ocp.c | 24 ++++++++++-------------- 1 file changed, 10 insertions(+), 14 deletions(-) diff --git a/drivers/ptp/ptp_ocp.c b/drivers/ptp/ptp_ocp.c index 95889f85ffb2..28243fb1d78f 100644 --- a/drivers/ptp/ptp_ocp.c +++ b/drivers/ptp/ptp_ocp.c @@ -2225,6 +2225,9 @@ ptp_ocp_ts_enable(void *priv, u32 req, bool enable) static void ptp_ocp_unregister_ext(struct ptp_ocp_ext_src *ext) { + if (!ext) + return; + ext->info->enable(ext, ~0, false); pci_free_irq(ext->bp->pdev, ext->irq_vec, ext); kfree(ext); @@ -4558,21 +4561,14 @@ ptp_ocp_detach(struct ptp_ocp *bp) ptp_ocp_detach_sysfs(bp); ptp_ocp_attr_group_del(bp); timer_delete_sync(&bp->watchdog); - if (bp->ts0) - ptp_ocp_unregister_ext(bp->ts0); - if (bp->ts1) - ptp_ocp_unregister_ext(bp->ts1); - if (bp->ts2) - ptp_ocp_unregister_ext(bp->ts2); - if (bp->ts3) - ptp_ocp_unregister_ext(bp->ts3); - if (bp->ts4) - ptp_ocp_unregister_ext(bp->ts4); - if (bp->pps) - ptp_ocp_unregister_ext(bp->pps); + ptp_ocp_unregister_ext(bp->ts0); + ptp_ocp_unregister_ext(bp->ts1); + ptp_ocp_unregister_ext(bp->ts2); + ptp_ocp_unregister_ext(bp->ts3); + ptp_ocp_unregister_ext(bp->ts4); + ptp_ocp_unregister_ext(bp->pps); for (i = 0; i < 4; i++) - if (bp->signal_out[i]) - ptp_ocp_unregister_ext(bp->signal_out[i]); + ptp_ocp_unregister_ext(bp->signal_out[i]); for (i = 0; i < __PORT_COUNT; i++) if (bp->port[i].line != -1) serial8250_unregister_port(bp->port[i].line); From 4c84a5c7b095ea52e65ace483177fce23e55fa8f Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Mon, 24 Nov 2025 09:45:47 +0100 Subject: [PATCH 1129/1381] ptp: ocp: Apply standard pattern for cleaning up loop The while (i--) is a standard pattern for the cleaning up loops. Apply this pattern where it makes sense in the driver. Reviewed-by: Vadim Fedorenko Signed-off-by: Andy Shevchenko Link: https://patch.msgid.link/20251124084816.205035-4-andriy.shevchenko@linux.intel.com Signed-off-by: Jakub Kicinski --- drivers/ptp/ptp_ocp.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/ptp/ptp_ocp.c b/drivers/ptp/ptp_ocp.c index 28243fb1d78f..c1d190a09c33 100644 --- a/drivers/ptp/ptp_ocp.c +++ b/drivers/ptp/ptp_ocp.c @@ -4821,8 +4821,7 @@ ptp_ocp_probe(struct pci_dev *pdev, const struct pci_device_id *id) return 0; out_dpll: - while (i) { - --i; + while (i--) { dpll_pin_unregister(bp->dpll, bp->sma[i].dpll_pin, &dpll_pins_ops, &bp->sma[i]); dpll_pin_put(bp->sma[i].dpll_pin); } From 648282e2d1e5508de3662e8d2680f64e8586fc58 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Mon, 24 Nov 2025 09:45:48 +0100 Subject: [PATCH 1130/1381] ptp: ocp: Reuse META's PCI vendor ID The META's PCI vendor ID is listed already in the pci_ids.h. Reuse it here. Reviewed-by: Vadim Fedorenko Signed-off-by: Andy Shevchenko Link: https://patch.msgid.link/20251124084816.205035-5-andriy.shevchenko@linux.intel.com Signed-off-by: Jakub Kicinski --- drivers/ptp/ptp_ocp.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/drivers/ptp/ptp_ocp.c b/drivers/ptp/ptp_ocp.c index c1d190a09c33..bf8187c963aa 100644 --- a/drivers/ptp/ptp_ocp.c +++ b/drivers/ptp/ptp_ocp.c @@ -25,8 +25,7 @@ #include #include -#define PCI_VENDOR_ID_FACEBOOK 0x1d9b -#define PCI_DEVICE_ID_FACEBOOK_TIMECARD 0x0400 +#define PCI_DEVICE_ID_META_TIMECARD 0x0400 #define PCI_VENDOR_ID_CELESTICA 0x18d4 #define PCI_DEVICE_ID_CELESTICA_TIMECARD 0x1008 @@ -1030,7 +1029,7 @@ static struct ocp_resource ocp_adva_resource[] = { }; static const struct pci_device_id ptp_ocp_pcidev_id[] = { - { PCI_DEVICE_DATA(FACEBOOK, TIMECARD, &ocp_fb_resource) }, + { PCI_DEVICE_DATA(META, TIMECARD, &ocp_fb_resource) }, { PCI_DEVICE_DATA(CELESTICA, TIMECARD, &ocp_fb_resource) }, { PCI_DEVICE_DATA(OROLIA, ARTCARD, &ocp_art_resource) }, { PCI_DEVICE_DATA(ADVA, TIMECARD, &ocp_adva_resource) }, From 17fa6ee35bd4b78752a2d33b92614a1c230a1ced Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Asbj=C3=B8rn=20Sloth=20T=C3=B8nnesen?= Date: Thu, 20 Nov 2025 17:44:26 +0000 Subject: [PATCH 1131/1381] tools: ynl-gen: add function prefix argument MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This patch adds a new CLI argument for overriding the default function prefix, as used for naming the doit/dumpit functions in the generated kernel code. When not specified the default "$(FAMILY)-nl" is used. This can also be specified persistently in generated files: /* YNL-ARG --function-prefix wg */ In the above example it causes the following changes: wireguard_nl_get_device_dumpit() -> wg_get_device_dumpit() wireguard_nl_get_device_doit() -> wg_get_device_doit() The variable name fn_prefix, was chosen as it relates to op_prefix which is used to prefix the UAPI commands enum entries. Link: https://lore.kernel.org/r/aRvWzC8qz3iXDAb3@zx2c4.com/ Suggested-by: Jason A. Donenfeld Signed-off-by: Asbjørn Sloth Tønnesen Link: https://patch.msgid.link/20251120174429.390574-2-ast@fiberby.net Signed-off-by: Jakub Kicinski --- tools/net/ynl/pyynl/ynl_gen_c.py | 25 ++++++++++++++++--------- 1 file changed, 16 insertions(+), 9 deletions(-) diff --git a/tools/net/ynl/pyynl/ynl_gen_c.py b/tools/net/ynl/pyynl/ynl_gen_c.py index aadeb3abcad8..0dd589e502cb 100755 --- a/tools/net/ynl/pyynl/ynl_gen_c.py +++ b/tools/net/ynl/pyynl/ynl_gen_c.py @@ -1205,7 +1205,7 @@ class SubMessage(SpecSubMessage): class Family(SpecFamily): - def __init__(self, file_name, exclude_ops): + def __init__(self, file_name, exclude_ops, fn_prefix): # Added by resolve: self.c_name = None delattr(self, "c_name") @@ -1237,6 +1237,8 @@ class Family(SpecFamily): else: self.uapi_header_name = self.ident_name + self.fn_prefix = fn_prefix if fn_prefix else f'{self.ident_name}-nl' + def resolve(self): self.resolve_up(super()) @@ -2911,12 +2913,12 @@ def print_kernel_op_table_fwd(family, cw, terminate): continue if 'do' in op: - name = c_lower(f"{family.ident_name}-nl-{op_name}-doit") + name = c_lower(f"{family.fn_prefix}-{op_name}-doit") cw.write_func_prot('int', name, ['struct sk_buff *skb', 'struct genl_info *info'], suffix=';') if 'dump' in op: - name = c_lower(f"{family.ident_name}-nl-{op_name}-dumpit") + name = c_lower(f"{family.fn_prefix}-{op_name}-dumpit") cw.write_func_prot('int', name, ['struct sk_buff *skb', 'struct netlink_callback *cb'], suffix=';') cw.nl() @@ -2942,7 +2944,7 @@ def print_kernel_op_table(family, cw): for x in op['dont-validate']])), ) for op_mode in ['do', 'dump']: if op_mode in op: - name = c_lower(f"{family.ident_name}-nl-{op_name}-{op_mode}it") + name = c_lower(f"{family.fn_prefix}-{op_name}-{op_mode}it") members.append((op_mode + 'it', name)) if family.kernel_policy == 'per-op': struct = Struct(family, op['attribute-set'], @@ -2980,7 +2982,7 @@ def print_kernel_op_table(family, cw): members.append(('validate', ' | '.join([c_upper('genl-dont-validate-' + x) for x in dont_validate])), ) - name = c_lower(f"{family.ident_name}-nl-{op_name}-{op_mode}it") + name = c_lower(f"{family.fn_prefix}-{op_name}-{op_mode}it") if 'pre' in op[op_mode]: members.append((cb_names[op_mode]['pre'], c_lower(op[op_mode]['pre']))) members.append((op_mode + 'it', name)) @@ -3402,6 +3404,7 @@ def main(): help='Do not overwrite the output file if the new output is identical to the old') parser.add_argument('--exclude-op', action='append', default=[]) parser.add_argument('-o', dest='out_file', type=str, default=None) + parser.add_argument('--function-prefix', dest='fn_prefix', type=str) args = parser.parse_args() if args.header is None: @@ -3410,7 +3413,7 @@ def main(): exclude_ops = [re.compile(expr) for expr in args.exclude_op] try: - parsed = Family(args.spec, exclude_ops) + parsed = Family(args.spec, exclude_ops, args.fn_prefix) if parsed.license != '((GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause)': print('Spec license:', parsed.license) print('License must be: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause)') @@ -3430,10 +3433,14 @@ def main(): cw.p("/* Do not edit directly, auto-generated from: */") cw.p(f"/*\t{spec_kernel} */") cw.p(f"/* YNL-GEN {args.mode} {'header' if args.header else 'source'} */") - if args.exclude_op or args.user_header: + if args.exclude_op or args.user_header or args.fn_prefix: line = '' - line += ' --user-header '.join([''] + args.user_header) - line += ' --exclude-op '.join([''] + args.exclude_op) + if args.user_header: + line += ' --user-header '.join([''] + args.user_header) + if args.exclude_op: + line += ' --exclude-op '.join([''] + args.exclude_op) + if args.fn_prefix: + line += f' --function-prefix {args.fn_prefix}' cw.p(f'/* YNL-ARG{line} */') cw.nl() From 68e83f3472667aac18d577587102f4bf77d0bd06 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Asbj=C3=B8rn=20Sloth=20T=C3=B8nnesen?= Date: Thu, 20 Nov 2025 17:44:27 +0000 Subject: [PATCH 1132/1381] tools: ynl-gen: add regeneration comment MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Add a comment on regeneration to the generated files. The comment is placed after the YNL-GEN line[1], as to not interfere with ynl-regen.sh's detection logic. [1] and after the optional YNL-ARG line. Link: https://lore.kernel.org/r/aR5m174O7pklKrMR@zx2c4.com/ Suggested-by: Jason A. Donenfeld Signed-off-by: Asbjørn Sloth Tønnesen Acked-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20251120174429.390574-3-ast@fiberby.net Signed-off-by: Jakub Kicinski --- drivers/android/binder_netlink.c | 1 + drivers/android/binder_netlink.h | 1 + drivers/dpll/dpll_nl.c | 1 + drivers/dpll/dpll_nl.h | 1 + drivers/net/ovpn/netlink-gen.c | 1 + drivers/net/ovpn/netlink-gen.h | 1 + drivers/net/team/team_nl.c | 1 + drivers/net/team/team_nl.h | 1 + fs/lockd/netlink.c | 1 + fs/lockd/netlink.h | 1 + fs/nfsd/netlink.c | 1 + fs/nfsd/netlink.h | 1 + include/uapi/linux/android/binder_netlink.h | 1 + include/uapi/linux/dpll.h | 1 + include/uapi/linux/ethtool_netlink_generated.h | 1 + include/uapi/linux/fou.h | 1 + include/uapi/linux/handshake.h | 1 + include/uapi/linux/if_team.h | 1 + include/uapi/linux/lockd_netlink.h | 1 + include/uapi/linux/mptcp_pm.h | 1 + include/uapi/linux/net_shaper.h | 1 + include/uapi/linux/netdev.h | 1 + include/uapi/linux/nfsd_netlink.h | 1 + include/uapi/linux/ovpn.h | 1 + include/uapi/linux/psp.h | 1 + net/core/netdev-genl-gen.c | 1 + net/core/netdev-genl-gen.h | 1 + net/devlink/netlink_gen.c | 1 + net/devlink/netlink_gen.h | 1 + net/handshake/genl.c | 1 + net/handshake/genl.h | 1 + net/ipv4/fou_nl.c | 1 + net/ipv4/fou_nl.h | 1 + net/mptcp/mptcp_pm_gen.c | 1 + net/mptcp/mptcp_pm_gen.h | 1 + net/psp/psp-nl-gen.c | 1 + net/psp/psp-nl-gen.h | 1 + net/shaper/shaper_nl_gen.c | 1 + net/shaper/shaper_nl_gen.h | 1 + tools/include/uapi/linux/netdev.h | 1 + tools/net/ynl/pyynl/ynl_gen_c.py | 1 + 41 files changed, 41 insertions(+) diff --git a/drivers/android/binder_netlink.c b/drivers/android/binder_netlink.c index d05397a50ca6..81e8432b5904 100644 --- a/drivers/android/binder_netlink.c +++ b/drivers/android/binder_netlink.c @@ -2,6 +2,7 @@ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/binder.yaml */ /* YNL-GEN kernel source */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ #include #include diff --git a/drivers/android/binder_netlink.h b/drivers/android/binder_netlink.h index 882c7a6b537e..57399942a5e3 100644 --- a/drivers/android/binder_netlink.h +++ b/drivers/android/binder_netlink.h @@ -2,6 +2,7 @@ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/binder.yaml */ /* YNL-GEN kernel header */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ #ifndef _LINUX_BINDER_GEN_H #define _LINUX_BINDER_GEN_H diff --git a/drivers/dpll/dpll_nl.c b/drivers/dpll/dpll_nl.c index 3c6d570babf8..36d11ff195df 100644 --- a/drivers/dpll/dpll_nl.c +++ b/drivers/dpll/dpll_nl.c @@ -2,6 +2,7 @@ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/dpll.yaml */ /* YNL-GEN kernel source */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ #include #include diff --git a/drivers/dpll/dpll_nl.h b/drivers/dpll/dpll_nl.h index 3da10cfe9a6e..7419679b6977 100644 --- a/drivers/dpll/dpll_nl.h +++ b/drivers/dpll/dpll_nl.h @@ -2,6 +2,7 @@ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/dpll.yaml */ /* YNL-GEN kernel header */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ #ifndef _LINUX_DPLL_GEN_H #define _LINUX_DPLL_GEN_H diff --git a/drivers/net/ovpn/netlink-gen.c b/drivers/net/ovpn/netlink-gen.c index 14298188c5f1..ecbe9dcf4f7d 100644 --- a/drivers/net/ovpn/netlink-gen.c +++ b/drivers/net/ovpn/netlink-gen.c @@ -2,6 +2,7 @@ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/ovpn.yaml */ /* YNL-GEN kernel source */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ #include #include diff --git a/drivers/net/ovpn/netlink-gen.h b/drivers/net/ovpn/netlink-gen.h index 220b5b2fdd4f..b2301580770f 100644 --- a/drivers/net/ovpn/netlink-gen.h +++ b/drivers/net/ovpn/netlink-gen.h @@ -2,6 +2,7 @@ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/ovpn.yaml */ /* YNL-GEN kernel header */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ #ifndef _LINUX_OVPN_GEN_H #define _LINUX_OVPN_GEN_H diff --git a/drivers/net/team/team_nl.c b/drivers/net/team/team_nl.c index 208424ab78f5..6db21725f9cc 100644 --- a/drivers/net/team/team_nl.c +++ b/drivers/net/team/team_nl.c @@ -2,6 +2,7 @@ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/team.yaml */ /* YNL-GEN kernel source */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ #include #include diff --git a/drivers/net/team/team_nl.h b/drivers/net/team/team_nl.h index c9ec1b22ac4d..74816b193475 100644 --- a/drivers/net/team/team_nl.h +++ b/drivers/net/team/team_nl.h @@ -2,6 +2,7 @@ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/team.yaml */ /* YNL-GEN kernel header */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ #ifndef _LINUX_TEAM_GEN_H #define _LINUX_TEAM_GEN_H diff --git a/fs/lockd/netlink.c b/fs/lockd/netlink.c index 6e00b02cad90..880c42b4f8c3 100644 --- a/fs/lockd/netlink.c +++ b/fs/lockd/netlink.c @@ -2,6 +2,7 @@ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/lockd.yaml */ /* YNL-GEN kernel source */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ #include #include diff --git a/fs/lockd/netlink.h b/fs/lockd/netlink.h index 1920543a7955..d8408f077dd8 100644 --- a/fs/lockd/netlink.h +++ b/fs/lockd/netlink.h @@ -2,6 +2,7 @@ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/lockd.yaml */ /* YNL-GEN kernel header */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ #ifndef _LINUX_LOCKD_GEN_H #define _LINUX_LOCKD_GEN_H diff --git a/fs/nfsd/netlink.c b/fs/nfsd/netlink.c index ca54aa583530..ac51a44e1065 100644 --- a/fs/nfsd/netlink.c +++ b/fs/nfsd/netlink.c @@ -2,6 +2,7 @@ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/nfsd.yaml */ /* YNL-GEN kernel source */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ #include #include diff --git a/fs/nfsd/netlink.h b/fs/nfsd/netlink.h index 8eb903f24c41..478117ff6b8c 100644 --- a/fs/nfsd/netlink.h +++ b/fs/nfsd/netlink.h @@ -2,6 +2,7 @@ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/nfsd.yaml */ /* YNL-GEN kernel header */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ #ifndef _LINUX_NFSD_GEN_H #define _LINUX_NFSD_GEN_H diff --git a/include/uapi/linux/android/binder_netlink.h b/include/uapi/linux/android/binder_netlink.h index b218f96d6668..bf69833c9a19 100644 --- a/include/uapi/linux/android/binder_netlink.h +++ b/include/uapi/linux/android/binder_netlink.h @@ -2,6 +2,7 @@ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/binder.yaml */ /* YNL-GEN uapi header */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ #ifndef _UAPI_LINUX_ANDROID_BINDER_NETLINK_H #define _UAPI_LINUX_ANDROID_BINDER_NETLINK_H diff --git a/include/uapi/linux/dpll.h b/include/uapi/linux/dpll.h index 69d35570ac4f..b7ff9c44f9aa 100644 --- a/include/uapi/linux/dpll.h +++ b/include/uapi/linux/dpll.h @@ -2,6 +2,7 @@ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/dpll.yaml */ /* YNL-GEN uapi header */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ #ifndef _UAPI_LINUX_DPLL_H #define _UAPI_LINUX_DPLL_H diff --git a/include/uapi/linux/ethtool_netlink_generated.h b/include/uapi/linux/ethtool_netlink_generated.h index b71b175df46d..556a0c834df5 100644 --- a/include/uapi/linux/ethtool_netlink_generated.h +++ b/include/uapi/linux/ethtool_netlink_generated.h @@ -2,6 +2,7 @@ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/ethtool.yaml */ /* YNL-GEN uapi header */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ #ifndef _UAPI_LINUX_ETHTOOL_NETLINK_GENERATED_H #define _UAPI_LINUX_ETHTOOL_NETLINK_GENERATED_H diff --git a/include/uapi/linux/fou.h b/include/uapi/linux/fou.h index b5cd3e7b3775..bb6bef74d2d1 100644 --- a/include/uapi/linux/fou.h +++ b/include/uapi/linux/fou.h @@ -2,6 +2,7 @@ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/fou.yaml */ /* YNL-GEN uapi header */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ #ifndef _UAPI_LINUX_FOU_H #define _UAPI_LINUX_FOU_H diff --git a/include/uapi/linux/handshake.h b/include/uapi/linux/handshake.h index 662e7de46c54..d7e40f594888 100644 --- a/include/uapi/linux/handshake.h +++ b/include/uapi/linux/handshake.h @@ -2,6 +2,7 @@ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/handshake.yaml */ /* YNL-GEN uapi header */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ #ifndef _UAPI_LINUX_HANDSHAKE_H #define _UAPI_LINUX_HANDSHAKE_H diff --git a/include/uapi/linux/if_team.h b/include/uapi/linux/if_team.h index a5c06243a435..f4cd839ae725 100644 --- a/include/uapi/linux/if_team.h +++ b/include/uapi/linux/if_team.h @@ -2,6 +2,7 @@ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/team.yaml */ /* YNL-GEN uapi header */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ #ifndef _UAPI_LINUX_IF_TEAM_H #define _UAPI_LINUX_IF_TEAM_H diff --git a/include/uapi/linux/lockd_netlink.h b/include/uapi/linux/lockd_netlink.h index 21c65aec3bc6..2d766a0fa6ea 100644 --- a/include/uapi/linux/lockd_netlink.h +++ b/include/uapi/linux/lockd_netlink.h @@ -2,6 +2,7 @@ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/lockd.yaml */ /* YNL-GEN uapi header */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ #ifndef _UAPI_LINUX_LOCKD_NETLINK_H #define _UAPI_LINUX_LOCKD_NETLINK_H diff --git a/include/uapi/linux/mptcp_pm.h b/include/uapi/linux/mptcp_pm.h index bf44a5cf5b5a..c97d060ee90b 100644 --- a/include/uapi/linux/mptcp_pm.h +++ b/include/uapi/linux/mptcp_pm.h @@ -2,6 +2,7 @@ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/mptcp_pm.yaml */ /* YNL-GEN uapi header */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ #ifndef _UAPI_LINUX_MPTCP_PM_H #define _UAPI_LINUX_MPTCP_PM_H diff --git a/include/uapi/linux/net_shaper.h b/include/uapi/linux/net_shaper.h index d8834b59f7d7..3dd22c2930d9 100644 --- a/include/uapi/linux/net_shaper.h +++ b/include/uapi/linux/net_shaper.h @@ -2,6 +2,7 @@ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/net_shaper.yaml */ /* YNL-GEN uapi header */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ #ifndef _UAPI_LINUX_NET_SHAPER_H #define _UAPI_LINUX_NET_SHAPER_H diff --git a/include/uapi/linux/netdev.h b/include/uapi/linux/netdev.h index 048c8de1a130..e0b579a1df4f 100644 --- a/include/uapi/linux/netdev.h +++ b/include/uapi/linux/netdev.h @@ -2,6 +2,7 @@ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/netdev.yaml */ /* YNL-GEN uapi header */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ #ifndef _UAPI_LINUX_NETDEV_H #define _UAPI_LINUX_NETDEV_H diff --git a/include/uapi/linux/nfsd_netlink.h b/include/uapi/linux/nfsd_netlink.h index 887cbd12b695..e157e2009ea8 100644 --- a/include/uapi/linux/nfsd_netlink.h +++ b/include/uapi/linux/nfsd_netlink.h @@ -2,6 +2,7 @@ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/nfsd.yaml */ /* YNL-GEN uapi header */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ #ifndef _UAPI_LINUX_NFSD_NETLINK_H #define _UAPI_LINUX_NFSD_NETLINK_H diff --git a/include/uapi/linux/ovpn.h b/include/uapi/linux/ovpn.h index 680d1522dc87..959b41def61f 100644 --- a/include/uapi/linux/ovpn.h +++ b/include/uapi/linux/ovpn.h @@ -2,6 +2,7 @@ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/ovpn.yaml */ /* YNL-GEN uapi header */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ #ifndef _UAPI_LINUX_OVPN_H #define _UAPI_LINUX_OVPN_H diff --git a/include/uapi/linux/psp.h b/include/uapi/linux/psp.h index d8449c043ba1..a3a336488dc3 100644 --- a/include/uapi/linux/psp.h +++ b/include/uapi/linux/psp.h @@ -2,6 +2,7 @@ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/psp.yaml */ /* YNL-GEN uapi header */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ #ifndef _UAPI_LINUX_PSP_H #define _UAPI_LINUX_PSP_H diff --git a/net/core/netdev-genl-gen.c b/net/core/netdev-genl-gen.c index ff20435c45d2..ba673e81716f 100644 --- a/net/core/netdev-genl-gen.c +++ b/net/core/netdev-genl-gen.c @@ -2,6 +2,7 @@ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/netdev.yaml */ /* YNL-GEN kernel source */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ #include #include diff --git a/net/core/netdev-genl-gen.h b/net/core/netdev-genl-gen.h index cf3fad74511f..cffc08517a41 100644 --- a/net/core/netdev-genl-gen.h +++ b/net/core/netdev-genl-gen.h @@ -2,6 +2,7 @@ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/netdev.yaml */ /* YNL-GEN kernel header */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ #ifndef _LINUX_NETDEV_GEN_H #define _LINUX_NETDEV_GEN_H diff --git a/net/devlink/netlink_gen.c b/net/devlink/netlink_gen.c index 580985025f49..f4c61c2b4f22 100644 --- a/net/devlink/netlink_gen.c +++ b/net/devlink/netlink_gen.c @@ -2,6 +2,7 @@ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/devlink.yaml */ /* YNL-GEN kernel source */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ #include #include diff --git a/net/devlink/netlink_gen.h b/net/devlink/netlink_gen.h index 09cc6f264ccf..2817d53a0eba 100644 --- a/net/devlink/netlink_gen.h +++ b/net/devlink/netlink_gen.h @@ -2,6 +2,7 @@ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/devlink.yaml */ /* YNL-GEN kernel header */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ #ifndef _LINUX_DEVLINK_GEN_H #define _LINUX_DEVLINK_GEN_H diff --git a/net/handshake/genl.c b/net/handshake/genl.c index f55d14d7b726..870612609491 100644 --- a/net/handshake/genl.c +++ b/net/handshake/genl.c @@ -2,6 +2,7 @@ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/handshake.yaml */ /* YNL-GEN kernel source */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ #include #include diff --git a/net/handshake/genl.h b/net/handshake/genl.h index ae72a596f6cc..8d3e18672daf 100644 --- a/net/handshake/genl.h +++ b/net/handshake/genl.h @@ -2,6 +2,7 @@ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/handshake.yaml */ /* YNL-GEN kernel header */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ #ifndef _LINUX_HANDSHAKE_GEN_H #define _LINUX_HANDSHAKE_GEN_H diff --git a/net/ipv4/fou_nl.c b/net/ipv4/fou_nl.c index 506260b4a4dc..7a99639204b1 100644 --- a/net/ipv4/fou_nl.c +++ b/net/ipv4/fou_nl.c @@ -2,6 +2,7 @@ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/fou.yaml */ /* YNL-GEN kernel source */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ #include #include diff --git a/net/ipv4/fou_nl.h b/net/ipv4/fou_nl.h index 63a6c4ed803d..438342dc8507 100644 --- a/net/ipv4/fou_nl.h +++ b/net/ipv4/fou_nl.h @@ -2,6 +2,7 @@ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/fou.yaml */ /* YNL-GEN kernel header */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ #ifndef _LINUX_FOU_GEN_H #define _LINUX_FOU_GEN_H diff --git a/net/mptcp/mptcp_pm_gen.c b/net/mptcp/mptcp_pm_gen.c index dcffd847af33..c180930a8e0e 100644 --- a/net/mptcp/mptcp_pm_gen.c +++ b/net/mptcp/mptcp_pm_gen.c @@ -2,6 +2,7 @@ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/mptcp_pm.yaml */ /* YNL-GEN kernel source */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ #include #include diff --git a/net/mptcp/mptcp_pm_gen.h b/net/mptcp/mptcp_pm_gen.h index e24258f6f819..b9280bcb43f5 100644 --- a/net/mptcp/mptcp_pm_gen.h +++ b/net/mptcp/mptcp_pm_gen.h @@ -2,6 +2,7 @@ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/mptcp_pm.yaml */ /* YNL-GEN kernel header */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ #ifndef _LINUX_MPTCP_PM_GEN_H #define _LINUX_MPTCP_PM_GEN_H diff --git a/net/psp/psp-nl-gen.c b/net/psp/psp-nl-gen.c index 73f8b06d66f0..22a48d0fa378 100644 --- a/net/psp/psp-nl-gen.c +++ b/net/psp/psp-nl-gen.c @@ -2,6 +2,7 @@ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/psp.yaml */ /* YNL-GEN kernel source */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ #include #include diff --git a/net/psp/psp-nl-gen.h b/net/psp/psp-nl-gen.h index 5bc3b5d5a53e..599c5f1c82f2 100644 --- a/net/psp/psp-nl-gen.h +++ b/net/psp/psp-nl-gen.h @@ -2,6 +2,7 @@ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/psp.yaml */ /* YNL-GEN kernel header */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ #ifndef _LINUX_PSP_GEN_H #define _LINUX_PSP_GEN_H diff --git a/net/shaper/shaper_nl_gen.c b/net/shaper/shaper_nl_gen.c index 204c8ae8c7b1..e8cccc4c1180 100644 --- a/net/shaper/shaper_nl_gen.c +++ b/net/shaper/shaper_nl_gen.c @@ -2,6 +2,7 @@ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/net_shaper.yaml */ /* YNL-GEN kernel source */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ #include #include diff --git a/net/shaper/shaper_nl_gen.h b/net/shaper/shaper_nl_gen.h index cb7f9026fc23..ec41c90431a4 100644 --- a/net/shaper/shaper_nl_gen.h +++ b/net/shaper/shaper_nl_gen.h @@ -2,6 +2,7 @@ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/net_shaper.yaml */ /* YNL-GEN kernel header */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ #ifndef _LINUX_NET_SHAPER_GEN_H #define _LINUX_NET_SHAPER_GEN_H diff --git a/tools/include/uapi/linux/netdev.h b/tools/include/uapi/linux/netdev.h index 048c8de1a130..e0b579a1df4f 100644 --- a/tools/include/uapi/linux/netdev.h +++ b/tools/include/uapi/linux/netdev.h @@ -2,6 +2,7 @@ /* Do not edit directly, auto-generated from: */ /* Documentation/netlink/specs/netdev.yaml */ /* YNL-GEN uapi header */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ #ifndef _UAPI_LINUX_NETDEV_H #define _UAPI_LINUX_NETDEV_H diff --git a/tools/net/ynl/pyynl/ynl_gen_c.py b/tools/net/ynl/pyynl/ynl_gen_c.py index 0dd589e502cb..b517d0c605ad 100755 --- a/tools/net/ynl/pyynl/ynl_gen_c.py +++ b/tools/net/ynl/pyynl/ynl_gen_c.py @@ -3442,6 +3442,7 @@ def main(): if args.fn_prefix: line += f' --function-prefix {args.fn_prefix}' cw.p(f'/* YNL-ARG{line} */') + cw.p('/* To regenerate run: tools/net/ynl/ynl-regen.sh */') cw.nl() if args.mode == 'uapi': From d696c73716147cb3158f1da1b89d7e75af3aa285 Mon Sep 17 00:00:00 2001 From: "Gustavo A. R. Silva" Date: Mon, 24 Nov 2025 18:42:08 +0900 Subject: [PATCH 1133/1381] chtls: Avoid -Wflex-array-member-not-at-end warning -Wflex-array-member-not-at-end was introduced in GCC-14, and we are getting ready to enable it, globally. Use the `DEFINE_RAW_FLEX()` helper for on-stack definitions of a flexible structure where the size of the flexible-array member is known at compile-time, and refactor the rest of the code, accordingly. So, with these changes, fix the following warning: drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_io.c:163:36: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end] Signed-off-by: Gustavo A. R. Silva Link: https://patch.msgid.link/aSQocKoJGkN0wzEj@kspp Signed-off-by: Jakub Kicinski --- .../net/ethernet/chelsio/inline_crypto/chtls/chtls_io.c | 8 +------- 1 file changed, 1 insertion(+), 7 deletions(-) diff --git a/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_io.c b/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_io.c index 4036db466e18..ee19933e2cca 100644 --- a/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_io.c +++ b/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_io.c @@ -159,19 +159,13 @@ static u8 tcp_state_to_flowc_state(u8 state) int send_tx_flowc_wr(struct sock *sk, int compl, u32 snd_nxt, u32 rcv_nxt) { - struct flowc_packed { - struct fw_flowc_wr fc; - struct fw_flowc_mnemval mnemval[FW_FLOWC_MNEM_MAX]; - } __packed sflowc; + DEFINE_RAW_FLEX(struct fw_flowc_wr, flowc, mnemval, FW_FLOWC_MNEM_MAX); int nparams, paramidx, flowclen16, flowclen; - struct fw_flowc_wr *flowc; struct chtls_sock *csk; struct tcp_sock *tp; csk = rcu_dereference_sk_user_data(sk); tp = tcp_sk(sk); - memset(&sflowc, 0, sizeof(sflowc)); - flowc = &sflowc.fc; #define FLOWC_PARAM(__m, __v) \ do { \ From 436fa8e7d1a119ec90f39e7c2ce0f3a146652899 Mon Sep 17 00:00:00 2001 From: Alexander Lobakin Date: Mon, 24 Nov 2025 18:07:35 +0100 Subject: [PATCH 1134/1381] ice: fix broken Rx on VFs Since the tagged commit, ice stopped respecting Rx buffer length passed from VFs. At that point, the buffer length was hardcoded in ice, so VFs still worked up to some point (until, for example, a VF wanted an MTU larger than its PF). The next commit 93f53db9f9dc ("ice: switch to Page Pool"), broke Rx on VFs completely since ice started accounting per-queue buffer lengths again, but now VF queues always had their length zeroed, as ice was already ignoring what iavf was passing to it. Restore the line that initializes the buffer length on VF queues basing on the virtchnl messages. Fixes: 3a4f419f7509 ("ice: drop page splitting and recycling") Reported-by: Jakub Slepecki Reviewed-by: Tony Nguyen Signed-off-by: Alexander Lobakin Reviewed-by: Aleksandr Loktionov Tested-by: Jakub Slepecki Link: https://patch.msgid.link/20251124170735.3077425-1-aleksander.lobakin@intel.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/intel/ice/virt/queues.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/net/ethernet/intel/ice/virt/queues.c b/drivers/net/ethernet/intel/ice/virt/queues.c index 7928f4e8e788..f73d5a3e83d4 100644 --- a/drivers/net/ethernet/intel/ice/virt/queues.c +++ b/drivers/net/ethernet/intel/ice/virt/queues.c @@ -842,6 +842,9 @@ int ice_vc_cfg_qs_msg(struct ice_vf *vf, u8 *msg) (qpi->rxq.databuffer_size > ((16 * 1024) - 128) || qpi->rxq.databuffer_size < 1024)) goto error_param; + + ring->rx_buf_len = qpi->rxq.databuffer_size; + if (qpi->rxq.max_pkt_size > max_frame_size || qpi->rxq.max_pkt_size < 64) goto error_param; From 3a6e8fd0bf4042c572dc52e634878b9aca02970d Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 24 Nov 2025 17:50:10 +0000 Subject: [PATCH 1135/1381] tcp: rename icsk_timeout() to tcp_timeout_expires() In preparation of sk->tcp_timeout_timer introduction, rename icsk_timeout() helper and change its argument to plain 'const struct sock *sk'. Signed-off-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20251124175013.1473655-2-edumazet@google.com Signed-off-by: Jakub Kicinski --- include/net/inet_connection_sock.h | 5 ++--- net/ipv4/inet_diag.c | 4 ++-- net/ipv4/tcp_ipv4.c | 4 ++-- net/ipv4/tcp_timer.c | 6 +++--- net/ipv6/tcp_ipv6.c | 4 ++-- net/mptcp/protocol.c | 2 +- 6 files changed, 12 insertions(+), 13 deletions(-) diff --git a/include/net/inet_connection_sock.h b/include/net/inet_connection_sock.h index fd40af2221b9..765c2149d678 100644 --- a/include/net/inet_connection_sock.h +++ b/include/net/inet_connection_sock.h @@ -184,10 +184,9 @@ static inline void inet_csk_delack_init(struct sock *sk) memset(&inet_csk(sk)->icsk_ack, 0, sizeof(inet_csk(sk)->icsk_ack)); } -static inline unsigned long -icsk_timeout(const struct inet_connection_sock *icsk) +static inline unsigned long tcp_timeout_expires(const struct sock *sk) { - return READ_ONCE(icsk->icsk_retransmit_timer.expires); + return READ_ONCE(inet_csk(sk)->icsk_retransmit_timer.expires); } static inline unsigned long diff --git a/net/ipv4/inet_diag.c b/net/ipv4/inet_diag.c index f0b6c5a411a2..9f63c09439a0 100644 --- a/net/ipv4/inet_diag.c +++ b/net/ipv4/inet_diag.c @@ -287,12 +287,12 @@ int inet_sk_diag_fill(struct sock *sk, struct inet_connection_sock *icsk, r->idiag_timer = 1; r->idiag_retrans = READ_ONCE(icsk->icsk_retransmits); r->idiag_expires = - jiffies_delta_to_msecs(icsk_timeout(icsk) - jiffies); + jiffies_delta_to_msecs(tcp_timeout_expires(sk) - jiffies); } else if (icsk_pending == ICSK_TIME_PROBE0) { r->idiag_timer = 4; r->idiag_retrans = READ_ONCE(icsk->icsk_probes_out); r->idiag_expires = - jiffies_delta_to_msecs(icsk_timeout(icsk) - jiffies); + jiffies_delta_to_msecs(tcp_timeout_expires(sk) - jiffies); } else if (timer_pending(&sk->sk_timer)) { r->idiag_timer = 2; r->idiag_retrans = READ_ONCE(icsk->icsk_probes_out); diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index e0bb8d9e2d9c..7b8af2c8d03a 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2869,10 +2869,10 @@ static void get_tcp4_sock(struct sock *sk, struct seq_file *f, int i) icsk_pending == ICSK_TIME_REO_TIMEOUT || icsk_pending == ICSK_TIME_LOSS_PROBE) { timer_active = 1; - timer_expires = icsk_timeout(icsk); + timer_expires = tcp_timeout_expires(sk); } else if (icsk_pending == ICSK_TIME_PROBE0) { timer_active = 4; - timer_expires = icsk_timeout(icsk); + timer_expires = tcp_timeout_expires(sk); } else if (timer_pending(&sk->sk_timer)) { timer_active = 2; timer_expires = sk->sk_timer.expires; diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index 0672c3d8f4f1..afbd901e610e 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -510,7 +510,7 @@ static bool tcp_rtx_probe0_timed_out(const struct sock *sk, * and tp->rcv_tstamp might very well have been written recently. * rcv_delta can thus be negative. */ - rcv_delta = icsk_timeout(icsk) - tp->rcv_tstamp; + rcv_delta = tcp_timeout_expires(sk) - tp->rcv_tstamp; if (rcv_delta <= timeout) return false; @@ -697,9 +697,9 @@ void tcp_write_timer_handler(struct sock *sk) !icsk->icsk_pending) return; - if (time_after(icsk_timeout(icsk), jiffies)) { + if (time_after(tcp_timeout_expires(sk), jiffies)) { sk_reset_timer(sk, &icsk->icsk_retransmit_timer, - icsk_timeout(icsk)); + tcp_timeout_expires(sk)); return; } tcp_mstamp_refresh(tcp_sk(sk)); diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 08113f430124..33c76c3a6da7 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -2163,10 +2163,10 @@ static void get_tcp6_sock(struct seq_file *seq, struct sock *sp, int i) icsk_pending == ICSK_TIME_REO_TIMEOUT || icsk_pending == ICSK_TIME_LOSS_PROBE) { timer_active = 1; - timer_expires = icsk_timeout(icsk); + timer_expires = tcp_timeout_expires(sp); } else if (icsk_pending == ICSK_TIME_PROBE0) { timer_active = 4; - timer_expires = icsk_timeout(icsk); + timer_expires = tcp_timeout_expires(sp); } else if (timer_pending(&sp->sk_timer)) { timer_active = 2; timer_expires = sp->sk_timer.expires; diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index e4ccc57b6f57..4288b6a53b6e 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -519,7 +519,7 @@ static long mptcp_timeout_from_subflow(const struct mptcp_subflow_context *subfl const struct sock *ssk = mptcp_subflow_tcp_sock(subflow); return inet_csk(ssk)->icsk_pending && !subflow->stale_count ? - icsk_timeout(inet_csk(ssk)) - jiffies : 0; + tcp_timeout_expires(ssk) - jiffies : 0; } static void mptcp_set_timeout(struct sock *sk) From 27e8257a86516682e2ec5d7543a8909c37ae8b00 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 24 Nov 2025 17:50:11 +0000 Subject: [PATCH 1136/1381] net: move sk_dst_pending_confirm and sk_pacing_status to sock_read_tx group These two fields are mostly read in TCP tx path, move them in an more appropriate group for better cache locality. Signed-off-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20251124175013.1473655-3-edumazet@google.com Signed-off-by: Jakub Kicinski --- include/net/sock.h | 4 ++-- net/core/sock.c | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/include/net/sock.h b/include/net/sock.h index 38d48cfe0741..a89aa97151f5 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -481,8 +481,6 @@ struct sock { struct rb_root tcp_rtx_queue; }; struct sk_buff_head sk_write_queue; - u32 sk_dst_pending_confirm; - u32 sk_pacing_status; /* see enum sk_pacing */ struct page_frag sk_frag; struct timer_list sk_timer; @@ -493,6 +491,8 @@ struct sock { __cacheline_group_end(sock_write_tx); __cacheline_group_begin(sock_read_tx); + u32 sk_dst_pending_confirm; + u32 sk_pacing_status; /* see enum sk_pacing */ unsigned long sk_max_pacing_rate; long sk_sndtimeo; u32 sk_priority; diff --git a/net/core/sock.c b/net/core/sock.c index b26a6cdc9bcd..45c98bf524b2 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -4519,14 +4519,14 @@ static int __init sock_struct_check(void) CACHELINE_ASSERT_GROUP_MEMBER(struct sock, sock_write_tx, sk_send_head); CACHELINE_ASSERT_GROUP_MEMBER(struct sock, sock_write_tx, sk_write_queue); CACHELINE_ASSERT_GROUP_MEMBER(struct sock, sock_write_tx, sk_write_pending); - CACHELINE_ASSERT_GROUP_MEMBER(struct sock, sock_write_tx, sk_dst_pending_confirm); - CACHELINE_ASSERT_GROUP_MEMBER(struct sock, sock_write_tx, sk_pacing_status); CACHELINE_ASSERT_GROUP_MEMBER(struct sock, sock_write_tx, sk_frag); CACHELINE_ASSERT_GROUP_MEMBER(struct sock, sock_write_tx, sk_timer); CACHELINE_ASSERT_GROUP_MEMBER(struct sock, sock_write_tx, sk_pacing_rate); CACHELINE_ASSERT_GROUP_MEMBER(struct sock, sock_write_tx, sk_zckey); CACHELINE_ASSERT_GROUP_MEMBER(struct sock, sock_write_tx, sk_tskey); + CACHELINE_ASSERT_GROUP_MEMBER(struct sock, sock_read_tx, sk_dst_pending_confirm); + CACHELINE_ASSERT_GROUP_MEMBER(struct sock, sock_read_tx, sk_pacing_status); CACHELINE_ASSERT_GROUP_MEMBER(struct sock, sock_read_tx, sk_max_pacing_rate); CACHELINE_ASSERT_GROUP_MEMBER(struct sock, sock_read_tx, sk_sndtimeo); CACHELINE_ASSERT_GROUP_MEMBER(struct sock, sock_read_tx, sk_priority); From 08dfe370239e53494453cee1e2ded2cdaa1efd12 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 24 Nov 2025 17:50:12 +0000 Subject: [PATCH 1137/1381] tcp: introduce icsk->icsk_keepalive_timer sk->sk_timer has been used for TCP keepalives. Keepalive timers are not in fast path, we want to use sk->sk_timer storage for retransmit timers, for better cache locality. Create icsk->icsk_keepalive_timer and change keepalive code to no longer use sk->sk_timer. Added space is reclaimed in the following patch. This includes changes to MPTCP, which was also using sk_timer. Alias icsk->mptcp_tout_timer and icsk->icsk_keepalive_timer for inet_sk_diag_fill() sake. Signed-off-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20251124175013.1473655-4-edumazet@google.com Signed-off-by: Jakub Kicinski --- .../net_cachelines/inet_connection_sock.rst | 1 + include/net/inet_connection_sock.h | 11 +++++++++-- net/ipv4/inet_connection_sock.c | 6 +++--- net/ipv4/inet_diag.c | 4 ++-- net/ipv4/tcp_ipv4.c | 4 ++-- net/ipv4/tcp_timer.c | 9 +++++---- net/ipv6/tcp_ipv6.c | 4 ++-- net/mptcp/protocol.c | 10 ++++++---- net/mptcp/protocol.h | 2 +- tools/testing/selftests/bpf/progs/bpf_iter_tcp4.c | 4 ++-- tools/testing/selftests/bpf/progs/bpf_iter_tcp6.c | 4 ++-- 11 files changed, 35 insertions(+), 24 deletions(-) diff --git a/Documentation/networking/net_cachelines/inet_connection_sock.rst b/Documentation/networking/net_cachelines/inet_connection_sock.rst index 8fae85ebb773..4f65de2def8c 100644 --- a/Documentation/networking/net_cachelines/inet_connection_sock.rst +++ b/Documentation/networking/net_cachelines/inet_connection_sock.rst @@ -14,6 +14,7 @@ struct inet_bind_bucket icsk_bind_hash read_mostly struct inet_bind2_bucket icsk_bind2_hash read_mostly tcp_set_state,inet_put_port struct timer_list icsk_retransmit_timer read_write inet_csk_reset_xmit_timer,tcp_connect struct timer_list icsk_delack_timer read_mostly inet_csk_reset_xmit_timer,tcp_connect +struct timer_list icsk_keepalive_timer u32 icsk_rto read_write tcp_cwnd_validate,tcp_schedule_loss_probe,tcp_connect_init,tcp_connect,tcp_write_xmit,tcp_push_one u32 icsk_rto_min u32 icsk_rto_max read_mostly tcp_reset_xmit_timer diff --git a/include/net/inet_connection_sock.h b/include/net/inet_connection_sock.h index 765c2149d678..e0d90b996348 100644 --- a/include/net/inet_connection_sock.h +++ b/include/net/inet_connection_sock.h @@ -57,6 +57,9 @@ struct inet_connection_sock_af_ops { * @icsk_bind_hash: Bind node * @icsk_bind2_hash: Bind node in the bhash2 table * @icsk_retransmit_timer: Resend (no ack) + * @icsk_delack_timer: Delayed ACK timer + * @icsk_keepalive_timer: Keepalive timer + * @mptcp_tout_timer: mptcp timer * @icsk_rto: Retransmit timeout * @icsk_pmtu_cookie Last pmtu seen by socket * @icsk_ca_ops Pluggable congestion control hook @@ -81,8 +84,12 @@ struct inet_connection_sock { struct request_sock_queue icsk_accept_queue; struct inet_bind_bucket *icsk_bind_hash; struct inet_bind2_bucket *icsk_bind2_hash; - struct timer_list icsk_retransmit_timer; - struct timer_list icsk_delack_timer; + struct timer_list icsk_retransmit_timer; + struct timer_list icsk_delack_timer; + union { + struct timer_list icsk_keepalive_timer; + struct timer_list mptcp_tout_timer; + }; __u32 icsk_rto; __u32 icsk_rto_min; u32 icsk_rto_max; diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index b4eae731c9ba..4fc09f9bf25d 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -739,7 +739,7 @@ void inet_csk_init_xmit_timers(struct sock *sk, timer_setup(&icsk->icsk_retransmit_timer, retransmit_handler, 0); timer_setup(&icsk->icsk_delack_timer, delack_handler, 0); - timer_setup(&sk->sk_timer, keepalive_handler, 0); + timer_setup(&icsk->icsk_keepalive_timer, keepalive_handler, 0); icsk->icsk_pending = icsk->icsk_ack.pending = 0; } @@ -752,7 +752,7 @@ void inet_csk_clear_xmit_timers(struct sock *sk) sk_stop_timer(sk, &icsk->icsk_retransmit_timer); sk_stop_timer(sk, &icsk->icsk_delack_timer); - sk_stop_timer(sk, &sk->sk_timer); + sk_stop_timer(sk, &icsk->icsk_keepalive_timer); } void inet_csk_clear_xmit_timers_sync(struct sock *sk) @@ -767,7 +767,7 @@ void inet_csk_clear_xmit_timers_sync(struct sock *sk) sk_stop_timer_sync(sk, &icsk->icsk_retransmit_timer); sk_stop_timer_sync(sk, &icsk->icsk_delack_timer); - sk_stop_timer_sync(sk, &sk->sk_timer); + sk_stop_timer_sync(sk, &icsk->icsk_keepalive_timer); } struct dst_entry *inet_csk_route_req(const struct sock *sk, diff --git a/net/ipv4/inet_diag.c b/net/ipv4/inet_diag.c index 9f63c09439a0..3f5b1418a610 100644 --- a/net/ipv4/inet_diag.c +++ b/net/ipv4/inet_diag.c @@ -293,11 +293,11 @@ int inet_sk_diag_fill(struct sock *sk, struct inet_connection_sock *icsk, r->idiag_retrans = READ_ONCE(icsk->icsk_probes_out); r->idiag_expires = jiffies_delta_to_msecs(tcp_timeout_expires(sk) - jiffies); - } else if (timer_pending(&sk->sk_timer)) { + } else if (timer_pending(&icsk->icsk_keepalive_timer)) { r->idiag_timer = 2; r->idiag_retrans = READ_ONCE(icsk->icsk_probes_out); r->idiag_expires = - jiffies_delta_to_msecs(sk->sk_timer.expires - jiffies); + jiffies_delta_to_msecs(icsk->icsk_keepalive_timer.expires - jiffies); } if ((ext & (1 << (INET_DIAG_INFO - 1))) && handler->idiag_info_size) { diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 7b8af2c8d03a..f8a9596e8f4d 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2873,9 +2873,9 @@ static void get_tcp4_sock(struct sock *sk, struct seq_file *f, int i) } else if (icsk_pending == ICSK_TIME_PROBE0) { timer_active = 4; timer_expires = tcp_timeout_expires(sk); - } else if (timer_pending(&sk->sk_timer)) { + } else if (timer_pending(&icsk->icsk_keepalive_timer)) { timer_active = 2; - timer_expires = sk->sk_timer.expires; + timer_expires = icsk->icsk_keepalive_timer.expires; } else { timer_active = 0; timer_expires = jiffies; diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index afbd901e610e..d2678dfd8118 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -755,12 +755,12 @@ void tcp_syn_ack_timeout(const struct request_sock *req) void tcp_reset_keepalive_timer(struct sock *sk, unsigned long len) { - sk_reset_timer(sk, &sk->sk_timer, jiffies + len); + sk_reset_timer(sk, &inet_csk(sk)->icsk_keepalive_timer, jiffies + len); } static void tcp_delete_keepalive_timer(struct sock *sk) { - sk_stop_timer(sk, &sk->sk_timer); + sk_stop_timer(sk, &inet_csk(sk)->icsk_keepalive_timer); } void tcp_set_keepalive(struct sock *sk, int val) @@ -777,8 +777,9 @@ EXPORT_IPV6_MOD_GPL(tcp_set_keepalive); static void tcp_keepalive_timer(struct timer_list *t) { - struct sock *sk = timer_container_of(sk, t, sk_timer); - struct inet_connection_sock *icsk = inet_csk(sk); + struct inet_connection_sock *icsk = + timer_container_of(icsk, t, icsk_keepalive_timer); + struct sock *sk = &icsk->icsk_inet.sk; struct tcp_sock *tp = tcp_sk(sk); u32 elapsed; diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 33c76c3a6da7..280fe5978559 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -2167,9 +2167,9 @@ static void get_tcp6_sock(struct seq_file *seq, struct sock *sp, int i) } else if (icsk_pending == ICSK_TIME_PROBE0) { timer_active = 4; timer_expires = tcp_timeout_expires(sp); - } else if (timer_pending(&sp->sk_timer)) { + } else if (timer_pending(&icsk->icsk_keepalive_timer)) { timer_active = 2; - timer_expires = sp->sk_timer.expires; + timer_expires = icsk->icsk_keepalive_timer.expires; } else { timer_active = 0; timer_expires = jiffies; diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index 4288b6a53b6e..89a5f63921e6 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -2374,7 +2374,9 @@ static void mptcp_retransmit_timer(struct timer_list *t) static void mptcp_tout_timer(struct timer_list *t) { - struct sock *sk = timer_container_of(sk, t, sk_timer); + struct inet_connection_sock *icsk = + timer_container_of(icsk, t, mptcp_tout_timer); + struct sock *sk = &icsk->icsk_inet.sk; mptcp_schedule_work(sk); sock_put(sk); @@ -2828,7 +2830,7 @@ void mptcp_reset_tout_timer(struct mptcp_sock *msk, unsigned long fail_tout) */ timeout = inet_csk(sk)->icsk_mtup.probe_timestamp ? close_timeout : fail_tout; - sk_reset_timer(sk, &sk->sk_timer, timeout); + sk_reset_timer(sk, &inet_csk(sk)->mptcp_tout_timer, timeout); } static void mptcp_mp_fail_no_response(struct mptcp_sock *msk) @@ -2974,7 +2976,7 @@ static void __mptcp_init_sock(struct sock *sk) /* re-use the csk retrans timer for MPTCP-level retrans */ timer_setup(&msk->sk.icsk_retransmit_timer, mptcp_retransmit_timer, 0); - timer_setup(&sk->sk_timer, mptcp_tout_timer, 0); + timer_setup(&msk->sk.mptcp_tout_timer, mptcp_tout_timer, 0); } static void mptcp_ca_reset(struct sock *sk) @@ -3176,7 +3178,7 @@ static void __mptcp_destroy_sock(struct sock *sk) might_sleep(); mptcp_stop_rtx_timer(sk); - sk_stop_timer(sk, &sk->sk_timer); + sk_stop_timer(sk, &inet_csk(sk)->mptcp_tout_timer); msk->pm.status = 0; mptcp_release_sched(msk); diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h index 313da78e2b75..9c0d17876b22 100644 --- a/net/mptcp/protocol.h +++ b/net/mptcp/protocol.h @@ -892,7 +892,7 @@ static inline void mptcp_stop_tout_timer(struct sock *sk) if (!inet_csk(sk)->icsk_mtup.probe_timestamp) return; - sk_stop_timer(sk, &sk->sk_timer); + sk_stop_timer(sk, &inet_csk(sk)->mptcp_tout_timer); inet_csk(sk)->icsk_mtup.probe_timestamp = 0; } diff --git a/tools/testing/selftests/bpf/progs/bpf_iter_tcp4.c b/tools/testing/selftests/bpf/progs/bpf_iter_tcp4.c index 164640db3a29..685811326a04 100644 --- a/tools/testing/selftests/bpf/progs/bpf_iter_tcp4.c +++ b/tools/testing/selftests/bpf/progs/bpf_iter_tcp4.c @@ -103,9 +103,9 @@ static int dump_tcp_sock(struct seq_file *seq, struct tcp_sock *tp, } else if (icsk->icsk_pending == ICSK_TIME_PROBE0) { timer_active = 4; timer_expires = icsk->icsk_retransmit_timer.expires; - } else if (timer_pending(&sp->sk_timer)) { + } else if (timer_pending(&icsk->icsk_keepalive_timer)) { timer_active = 2; - timer_expires = sp->sk_timer.expires; + timer_expires = icsk->icsk_keepalive_timer.expires; } else { timer_active = 0; timer_expires = bpf_jiffies64(); diff --git a/tools/testing/selftests/bpf/progs/bpf_iter_tcp6.c b/tools/testing/selftests/bpf/progs/bpf_iter_tcp6.c index 591c703f5032..0f4a92712751 100644 --- a/tools/testing/selftests/bpf/progs/bpf_iter_tcp6.c +++ b/tools/testing/selftests/bpf/progs/bpf_iter_tcp6.c @@ -103,9 +103,9 @@ static int dump_tcp6_sock(struct seq_file *seq, struct tcp6_sock *tp, } else if (icsk->icsk_pending == ICSK_TIME_PROBE0) { timer_active = 4; timer_expires = icsk->icsk_retransmit_timer.expires; - } else if (timer_pending(&sp->sk_timer)) { + } else if (timer_pending(&icsk->icsk_keepalive_timer)) { timer_active = 2; - timer_expires = sp->sk_timer.expires; + timer_expires = icsk->icsk_keepalive_timer.expires; } else { timer_active = 0; timer_expires = bpf_jiffies64(); From 9a5e5334adc039fa652aa071ea95b18db0bc1f43 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 24 Nov 2025 17:50:13 +0000 Subject: [PATCH 1138/1381] tcp: remove icsk->icsk_retransmit_timer Now sk->sk_timer is no longer used by TCP keepalive, we can use its storage for TCP and MPTCP retransmit timers for better cache locality. Signed-off-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20251124175013.1473655-5-edumazet@google.com Signed-off-by: Jakub Kicinski --- .../net_cachelines/inet_connection_sock.rst | 1 - include/net/inet_connection_sock.h | 8 +++----- include/net/sock.h | 9 +++++++-- net/ipv4/inet_connection_sock.c | 6 +++--- net/ipv4/tcp_timer.c | 8 +++----- net/mptcp/protocol.c | 15 +++++---------- tools/testing/selftests/bpf/progs/bpf_iter_tcp4.c | 4 ++-- tools/testing/selftests/bpf/progs/bpf_iter_tcp6.c | 4 ++-- 8 files changed, 25 insertions(+), 30 deletions(-) diff --git a/Documentation/networking/net_cachelines/inet_connection_sock.rst b/Documentation/networking/net_cachelines/inet_connection_sock.rst index 4f65de2def8c..cc2000f55c29 100644 --- a/Documentation/networking/net_cachelines/inet_connection_sock.rst +++ b/Documentation/networking/net_cachelines/inet_connection_sock.rst @@ -12,7 +12,6 @@ struct inet_sock icsk_inet read_mostly r struct request_sock_queue icsk_accept_queue struct inet_bind_bucket icsk_bind_hash read_mostly tcp_set_state struct inet_bind2_bucket icsk_bind2_hash read_mostly tcp_set_state,inet_put_port -struct timer_list icsk_retransmit_timer read_write inet_csk_reset_xmit_timer,tcp_connect struct timer_list icsk_delack_timer read_mostly inet_csk_reset_xmit_timer,tcp_connect struct timer_list icsk_keepalive_timer u32 icsk_rto read_write tcp_cwnd_validate,tcp_schedule_loss_probe,tcp_connect_init,tcp_connect,tcp_write_xmit,tcp_push_one diff --git a/include/net/inet_connection_sock.h b/include/net/inet_connection_sock.h index e0d90b996348..ecb362025c4e 100644 --- a/include/net/inet_connection_sock.h +++ b/include/net/inet_connection_sock.h @@ -56,7 +56,6 @@ struct inet_connection_sock_af_ops { * @icsk_accept_queue: FIFO of established children * @icsk_bind_hash: Bind node * @icsk_bind2_hash: Bind node in the bhash2 table - * @icsk_retransmit_timer: Resend (no ack) * @icsk_delack_timer: Delayed ACK timer * @icsk_keepalive_timer: Keepalive timer * @mptcp_tout_timer: mptcp timer @@ -84,7 +83,6 @@ struct inet_connection_sock { struct request_sock_queue icsk_accept_queue; struct inet_bind_bucket *icsk_bind_hash; struct inet_bind2_bucket *icsk_bind2_hash; - struct timer_list icsk_retransmit_timer; struct timer_list icsk_delack_timer; union { struct timer_list icsk_keepalive_timer; @@ -193,7 +191,7 @@ static inline void inet_csk_delack_init(struct sock *sk) static inline unsigned long tcp_timeout_expires(const struct sock *sk) { - return READ_ONCE(inet_csk(sk)->icsk_retransmit_timer.expires); + return READ_ONCE(sk->tcp_retransmit_timer.expires); } static inline unsigned long @@ -209,7 +207,7 @@ static inline void inet_csk_clear_xmit_timer(struct sock *sk, const int what) if (what == ICSK_TIME_RETRANS || what == ICSK_TIME_PROBE0) { smp_store_release(&icsk->icsk_pending, 0); #ifdef INET_CSK_CLEAR_TIMERS - sk_stop_timer(sk, &icsk->icsk_retransmit_timer); + sk_stop_timer(sk, &sk->tcp_retransmit_timer); #endif } else if (what == ICSK_TIME_DACK) { smp_store_release(&icsk->icsk_ack.pending, 0); @@ -241,7 +239,7 @@ static inline void inet_csk_reset_xmit_timer(struct sock *sk, const int what, if (what == ICSK_TIME_RETRANS || what == ICSK_TIME_PROBE0 || what == ICSK_TIME_LOSS_PROBE || what == ICSK_TIME_REO_TIMEOUT) { smp_store_release(&icsk->icsk_pending, what); - sk_reset_timer(sk, &icsk->icsk_retransmit_timer, when); + sk_reset_timer(sk, &sk->tcp_retransmit_timer, when); } else if (what == ICSK_TIME_DACK) { smp_store_release(&icsk->icsk_ack.pending, icsk->icsk_ack.pending | ICSK_ACK_TIMER); diff --git a/include/net/sock.h b/include/net/sock.h index a89aa97151f5..02253c6a578b 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -305,6 +305,8 @@ struct sk_filter; * @sk_txrehash: enable TX hash rethink * @sk_filter: socket filtering instructions * @sk_timer: sock cleanup timer + * @tcp_retransmit_timer: tcp retransmit timer + * @mptcp_retransmit_timer: mptcp retransmit timer * @sk_stamp: time stamp of last packet received * @sk_stamp_seq: lock for accessing sk_stamp on 32 bit architectures only * @sk_tsflags: SO_TIMESTAMPING flags @@ -482,8 +484,11 @@ struct sock { }; struct sk_buff_head sk_write_queue; struct page_frag sk_frag; - struct timer_list sk_timer; - + union { + struct timer_list sk_timer; + struct timer_list tcp_retransmit_timer; + struct timer_list mptcp_retransmit_timer; + }; unsigned long sk_pacing_rate; /* bytes per second */ atomic_t sk_zckey; atomic_t sk_tskey; diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index 4fc09f9bf25d..97d57c52b9ad 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -737,7 +737,7 @@ void inet_csk_init_xmit_timers(struct sock *sk, { struct inet_connection_sock *icsk = inet_csk(sk); - timer_setup(&icsk->icsk_retransmit_timer, retransmit_handler, 0); + timer_setup(&sk->tcp_retransmit_timer, retransmit_handler, 0); timer_setup(&icsk->icsk_delack_timer, delack_handler, 0); timer_setup(&icsk->icsk_keepalive_timer, keepalive_handler, 0); icsk->icsk_pending = icsk->icsk_ack.pending = 0; @@ -750,7 +750,7 @@ void inet_csk_clear_xmit_timers(struct sock *sk) smp_store_release(&icsk->icsk_pending, 0); smp_store_release(&icsk->icsk_ack.pending, 0); - sk_stop_timer(sk, &icsk->icsk_retransmit_timer); + sk_stop_timer(sk, &sk->tcp_retransmit_timer); sk_stop_timer(sk, &icsk->icsk_delack_timer); sk_stop_timer(sk, &icsk->icsk_keepalive_timer); } @@ -765,7 +765,7 @@ void inet_csk_clear_xmit_timers_sync(struct sock *sk) smp_store_release(&icsk->icsk_pending, 0); smp_store_release(&icsk->icsk_ack.pending, 0); - sk_stop_timer_sync(sk, &icsk->icsk_retransmit_timer); + sk_stop_timer_sync(sk, &sk->tcp_retransmit_timer); sk_stop_timer_sync(sk, &icsk->icsk_delack_timer); sk_stop_timer_sync(sk, &icsk->icsk_keepalive_timer); } diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index d2678dfd8118..160080c9021d 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -698,7 +698,7 @@ void tcp_write_timer_handler(struct sock *sk) return; if (time_after(tcp_timeout_expires(sk), jiffies)) { - sk_reset_timer(sk, &icsk->icsk_retransmit_timer, + sk_reset_timer(sk, &sk->tcp_retransmit_timer, tcp_timeout_expires(sk)); return; } @@ -725,12 +725,10 @@ void tcp_write_timer_handler(struct sock *sk) static void tcp_write_timer(struct timer_list *t) { - struct inet_connection_sock *icsk = - timer_container_of(icsk, t, icsk_retransmit_timer); - struct sock *sk = &icsk->icsk_inet.sk; + struct sock *sk = timer_container_of(sk, t, tcp_retransmit_timer); /* Avoid locking the socket when there is no pending event. */ - if (!smp_load_acquire(&icsk->icsk_pending)) + if (!smp_load_acquire(&inet_csk(sk)->icsk_pending)) goto out; bh_lock_sock(sk); diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index 89a5f63921e6..bb7d634cf312 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -411,9 +411,7 @@ static bool __mptcp_move_skb(struct sock *sk, struct sk_buff *skb) static void mptcp_stop_rtx_timer(struct sock *sk) { - struct inet_connection_sock *icsk = inet_csk(sk); - - sk_stop_timer(sk, &icsk->icsk_retransmit_timer); + sk_stop_timer(sk, &sk->mptcp_retransmit_timer); mptcp_sk(sk)->timer_ival = 0; } @@ -954,12 +952,11 @@ static void __mptcp_flush_join_list(struct sock *sk, struct list_head *join_list static bool mptcp_rtx_timer_pending(struct sock *sk) { - return timer_pending(&inet_csk(sk)->icsk_retransmit_timer); + return timer_pending(&sk->mptcp_retransmit_timer); } static void mptcp_reset_rtx_timer(struct sock *sk) { - struct inet_connection_sock *icsk = inet_csk(sk); unsigned long tout; /* prevent rescheduling on close */ @@ -967,7 +964,7 @@ static void mptcp_reset_rtx_timer(struct sock *sk) return; tout = mptcp_sk(sk)->timer_ival; - sk_reset_timer(sk, &icsk->icsk_retransmit_timer, jiffies + tout); + sk_reset_timer(sk, &sk->mptcp_retransmit_timer, jiffies + tout); } bool mptcp_schedule_work(struct sock *sk) @@ -2354,9 +2351,7 @@ static int mptcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, static void mptcp_retransmit_timer(struct timer_list *t) { - struct inet_connection_sock *icsk = timer_container_of(icsk, t, - icsk_retransmit_timer); - struct sock *sk = &icsk->icsk_inet.sk; + struct sock *sk = timer_container_of(sk, t, mptcp_retransmit_timer); struct mptcp_sock *msk = mptcp_sk(sk); bh_lock_sock(sk); @@ -2975,7 +2970,7 @@ static void __mptcp_init_sock(struct sock *sk) spin_lock_init(&msk->fallback_lock); /* re-use the csk retrans timer for MPTCP-level retrans */ - timer_setup(&msk->sk.icsk_retransmit_timer, mptcp_retransmit_timer, 0); + timer_setup(&sk->mptcp_retransmit_timer, mptcp_retransmit_timer, 0); timer_setup(&msk->sk.mptcp_tout_timer, mptcp_tout_timer, 0); } diff --git a/tools/testing/selftests/bpf/progs/bpf_iter_tcp4.c b/tools/testing/selftests/bpf/progs/bpf_iter_tcp4.c index 685811326a04..b1e509b231cd 100644 --- a/tools/testing/selftests/bpf/progs/bpf_iter_tcp4.c +++ b/tools/testing/selftests/bpf/progs/bpf_iter_tcp4.c @@ -99,10 +99,10 @@ static int dump_tcp_sock(struct seq_file *seq, struct tcp_sock *tp, icsk->icsk_pending == ICSK_TIME_REO_TIMEOUT || icsk->icsk_pending == ICSK_TIME_LOSS_PROBE) { timer_active = 1; - timer_expires = icsk->icsk_retransmit_timer.expires; + timer_expires = sp->tcp_retransmit_timer.expires; } else if (icsk->icsk_pending == ICSK_TIME_PROBE0) { timer_active = 4; - timer_expires = icsk->icsk_retransmit_timer.expires; + timer_expires = sp->tcp_retransmit_timer.expires; } else if (timer_pending(&icsk->icsk_keepalive_timer)) { timer_active = 2; timer_expires = icsk->icsk_keepalive_timer.expires; diff --git a/tools/testing/selftests/bpf/progs/bpf_iter_tcp6.c b/tools/testing/selftests/bpf/progs/bpf_iter_tcp6.c index 0f4a92712751..dbc7166aee91 100644 --- a/tools/testing/selftests/bpf/progs/bpf_iter_tcp6.c +++ b/tools/testing/selftests/bpf/progs/bpf_iter_tcp6.c @@ -99,10 +99,10 @@ static int dump_tcp6_sock(struct seq_file *seq, struct tcp6_sock *tp, icsk->icsk_pending == ICSK_TIME_REO_TIMEOUT || icsk->icsk_pending == ICSK_TIME_LOSS_PROBE) { timer_active = 1; - timer_expires = icsk->icsk_retransmit_timer.expires; + timer_expires = sp->tcp_retransmit_timer.expires; } else if (icsk->icsk_pending == ICSK_TIME_PROBE0) { timer_active = 4; - timer_expires = icsk->icsk_retransmit_timer.expires; + timer_expires = sp->tcp_retransmit_timer.expires; } else if (timer_pending(&icsk->icsk_keepalive_timer)) { timer_active = 2; timer_expires = icsk->icsk_keepalive_timer.expires; From cfeb7cd80f40dbe02ee95cd175dec341c42abae1 Mon Sep 17 00:00:00 2001 From: Liming Wu Date: Thu, 20 Nov 2025 09:53:20 +0800 Subject: [PATCH 1139/1381] virtio_net: enhance wake/stop tx queue statistics accounting This patch refines and strengthens the statistics collection of TX queue wake/stop events introduced by commit c39add9b2423 ("virtio_net: Add TX stopped and wake counters"). Previously, the driver only recorded partial wake/stop statistics for TX queues. Some wake events triggered by 'skb_xmit_done()' or resume operations were not counted, which made the per-queue metrics incomplete. Signed-off-by: Liming Wu Acked-by: Michael S. Tsirkin Link: https://patch.msgid.link/20251120015320.1418-1-liming.wu@jaguarmicro.com Signed-off-by: Jakub Kicinski --- drivers/net/virtio_net.c | 44 ++++++++++++++++++++++++---------------- 1 file changed, 26 insertions(+), 18 deletions(-) diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index cfa006b88688..b0947e15895f 100644 --- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -775,10 +775,26 @@ static bool virtqueue_napi_complete(struct napi_struct *napi, return false; } +static void virtnet_tx_wake_queue(struct virtnet_info *vi, + struct send_queue *sq) +{ + unsigned int index = vq2txq(sq->vq); + struct netdev_queue *txq = netdev_get_tx_queue(vi->dev, index); + + if (netif_tx_queue_stopped(txq)) { + u64_stats_update_begin(&sq->stats.syncp); + u64_stats_inc(&sq->stats.wake); + u64_stats_update_end(&sq->stats.syncp); + netif_tx_wake_queue(txq); + } +} + static void skb_xmit_done(struct virtqueue *vq) { struct virtnet_info *vi = vq->vdev->priv; - struct napi_struct *napi = &vi->sq[vq2txq(vq)].napi; + unsigned int index = vq2txq(vq); + struct send_queue *sq = &vi->sq[index]; + struct napi_struct *napi = &sq->napi; /* Suppress further interrupts. */ virtqueue_disable_cb(vq); @@ -786,8 +802,7 @@ static void skb_xmit_done(struct virtqueue *vq) if (napi->weight) virtqueue_napi_schedule(napi, vq); else - /* We were probably waiting for more output buffers. */ - netif_wake_subqueue(vi->dev, vq2txq(vq)); + virtnet_tx_wake_queue(vi, sq); } #define MRG_CTX_HEADER_SHIFT 22 @@ -3080,13 +3095,8 @@ static void virtnet_poll_cleantx(struct receive_queue *rq, int budget) free_old_xmit(sq, txq, !!budget); } while (unlikely(!virtqueue_enable_cb_delayed(sq->vq))); - if (sq->vq->num_free >= MAX_SKB_FRAGS + 2 && - netif_tx_queue_stopped(txq)) { - u64_stats_update_begin(&sq->stats.syncp); - u64_stats_inc(&sq->stats.wake); - u64_stats_update_end(&sq->stats.syncp); - netif_tx_wake_queue(txq); - } + if (sq->vq->num_free >= MAX_SKB_FRAGS + 2) + virtnet_tx_wake_queue(vi, sq); __netif_tx_unlock(txq); } @@ -3276,13 +3286,8 @@ static int virtnet_poll_tx(struct napi_struct *napi, int budget) else free_old_xmit(sq, txq, !!budget); - if (sq->vq->num_free >= MAX_SKB_FRAGS + 2 && - netif_tx_queue_stopped(txq)) { - u64_stats_update_begin(&sq->stats.syncp); - u64_stats_inc(&sq->stats.wake); - u64_stats_update_end(&sq->stats.syncp); - netif_tx_wake_queue(txq); - } + if (sq->vq->num_free >= MAX_SKB_FRAGS + 2) + virtnet_tx_wake_queue(vi, sq); if (xsk_done >= budget) { __netif_tx_unlock(txq); @@ -3537,6 +3542,9 @@ static void virtnet_tx_pause(struct virtnet_info *vi, struct send_queue *sq) /* Prevent the upper layer from trying to send packets. */ netif_stop_subqueue(vi->dev, qindex); + u64_stats_update_begin(&sq->stats.syncp); + u64_stats_inc(&sq->stats.stop); + u64_stats_update_end(&sq->stats.syncp); __netif_tx_unlock_bh(txq); } @@ -3553,7 +3561,7 @@ static void virtnet_tx_resume(struct virtnet_info *vi, struct send_queue *sq) __netif_tx_lock_bh(txq); sq->reset = false; - netif_tx_wake_queue(txq); + virtnet_tx_wake_queue(vi, sq); __netif_tx_unlock_bh(txq); if (running) From adb6b68c50604f3113d62758e839cc0186b94ce8 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 24 Nov 2025 21:26:39 +0000 Subject: [PATCH 1140/1381] selftest: af_unix: Create its own .gitignore. Somehow AF_UNIX tests have reused ../.gitignore, but now NIPA warns about it. Let's create .gitignore under af_unix/. Signed-off-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20251124212805.486235-2-kuniyu@google.com Signed-off-by: Jakub Kicinski --- tools/testing/selftests/net/.gitignore | 8 -------- tools/testing/selftests/net/af_unix/.gitignore | 8 ++++++++ 2 files changed, 8 insertions(+), 8 deletions(-) create mode 100644 tools/testing/selftests/net/af_unix/.gitignore diff --git a/tools/testing/selftests/net/.gitignore b/tools/testing/selftests/net/.gitignore index 283ca5ffc244..6930fe926c58 100644 --- a/tools/testing/selftests/net/.gitignore +++ b/tools/testing/selftests/net/.gitignore @@ -4,7 +4,6 @@ bind_timewait bind_wildcard busy_poller cmsg_sender -diag_uid epoll_busy_poll fin_ack_lat hwtstamp_config @@ -17,7 +16,6 @@ ipv6_flowlabel ipv6_flowlabel_mgr ipv6_fragmentation log.txt -msg_oob msg_zerocopy netlink-dumps nettest @@ -34,9 +32,6 @@ reuseport_bpf_numa reuseport_dualstack rxtimestamp sctp_hello -scm_inq -scm_pidfd -scm_rights sk_bind_sendto_listen sk_connect_zero_addr sk_so_peek_off @@ -44,7 +39,6 @@ skf_net_off socket so_incoming_cpu so_netns_cookie -so_peek_off so_txtime so_rcv_listener stress_reuseport_listen @@ -63,5 +57,3 @@ txtimestamp udpgso udpgso_bench_rx udpgso_bench_tx -unix_connect -unix_connreset diff --git a/tools/testing/selftests/net/af_unix/.gitignore b/tools/testing/selftests/net/af_unix/.gitignore new file mode 100644 index 000000000000..240b26740c9e --- /dev/null +++ b/tools/testing/selftests/net/af_unix/.gitignore @@ -0,0 +1,8 @@ +diag_uid +msg_oob +scm_inq +scm_pidfd +scm_rights +so_peek_off +unix_connect +unix_connreset From ebe2f0b3cfe380120adec59264d58d5b26b931ed Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 24 Nov 2025 21:26:40 +0000 Subject: [PATCH 1141/1381] selftest: af_unix: Extend recv() timeout in so_peek_off.c. so_peek_off.c is reported to be flaky on NIPA: # # so_peek_off.c:149:two_chunks_overlap_blocking:Expected -1 (-1) != bytes (-1) # # two_chunks_overlap_blocking: Test terminated by assertion # # FAIL so_peek_off.stream.two_chunks_overlap_blocking The test fork()s a child process to send() data after 1ms to wake up the parent process being blocked (up to 3ms) on recv(). But, from the log, the parent woke up after 3ms timeout, so it could be too short when the host is overloaded. Let's extend it to 5s. Reported-by: Jakub Kicinski Closes: https://lore.kernel.org/netdev/20251124070722.1e828c53@kernel.org/ Signed-off-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20251124212805.486235-3-kuniyu@google.com Signed-off-by: Jakub Kicinski --- tools/testing/selftests/net/af_unix/so_peek_off.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tools/testing/selftests/net/af_unix/so_peek_off.c b/tools/testing/selftests/net/af_unix/so_peek_off.c index 1a77728128e5..86e7b0fb522d 100644 --- a/tools/testing/selftests/net/af_unix/so_peek_off.c +++ b/tools/testing/selftests/net/af_unix/so_peek_off.c @@ -36,8 +36,8 @@ FIXTURE_VARIANT_ADD(so_peek_off, seqpacket) FIXTURE_SETUP(so_peek_off) { struct timeval timeout = { - .tv_sec = 0, - .tv_usec = 3000, + .tv_sec = 5, + .tv_usec = 0, }; int ret; From ab084f0b8d6d2ee4b1c6a28f39a2a7430bdfa7f0 Mon Sep 17 00:00:00 2001 From: Dimitri Daskalakis Date: Mon, 24 Nov 2025 12:05:18 -0800 Subject: [PATCH 1142/1381] drivers: net: fbnic: Return the true error in fbnic_alloc_napi_vectors. The error case in fbnic_alloc_napi_vectors defaulted to returning ENOMEM. This can mask the true error case, causing confusion. Signed-off-by: Dimitri Daskalakis Link: https://patch.msgid.link/20251124200518.1848029-1-dimitri.daskalakis1@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/meta/fbnic/fbnic_txrx.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_txrx.c b/drivers/net/ethernet/meta/fbnic/fbnic_txrx.c index 56744e3a14ec..13d508ce637f 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_txrx.c +++ b/drivers/net/ethernet/meta/fbnic/fbnic_txrx.c @@ -1808,7 +1808,7 @@ int fbnic_alloc_napi_vectors(struct fbnic_net *fbn) free_vectors: fbnic_free_napi_vectors(fbn); - return -ENOMEM; + return err; } static void fbnic_free_ring_resources(struct device *dev, From f9e788c5fd3a23edecd808ebb354e2cb1aef87c3 Mon Sep 17 00:00:00 2001 From: Lachlan Hodges Date: Wed, 26 Nov 2025 12:57:58 +1100 Subject: [PATCH 1143/1381] wifi: mac80211: allow sharing identical chanctx for S1G interfaces Introduce support for sharing identical channel contexts for S1G interfaces. Additionally, do not downgrade channel requests for S1G interfaces. Signed-off-by: Lachlan Hodges Link: https://patch.msgid.link/20251126015758.149034-1-lachlan.hodges@morsemicro.com Signed-off-by: Johannes Berg --- net/mac80211/chan.c | 13 ++++++++++++- net/mac80211/mlme.c | 5 +++-- 2 files changed, 15 insertions(+), 3 deletions(-) diff --git a/net/mac80211/chan.c b/net/mac80211/chan.c index c8aba4183c9a..d0bfb1216401 100644 --- a/net/mac80211/chan.c +++ b/net/mac80211/chan.c @@ -654,8 +654,19 @@ static void _ieee80211_change_chanctx(struct ieee80211_local *local, }; u32 changed = 0; - /* expected to handle only 20/40/80/160/320 channel widths */ + /* 5/10 MHz not handled here */ switch (chandef->width) { + case NL80211_CHAN_WIDTH_1: + case NL80211_CHAN_WIDTH_2: + case NL80211_CHAN_WIDTH_4: + case NL80211_CHAN_WIDTH_8: + case NL80211_CHAN_WIDTH_16: + /* + * mac80211 currently only supports sharing identical + * chanctx's for S1G interfaces. + */ + WARN_ON(!ieee80211_chanreq_identical(&ctx_req, chanreq)); + return; case NL80211_CHAN_WIDTH_20_NOHT: case NL80211_CHAN_WIDTH_20: case NL80211_CHAN_WIDTH_40: diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c index c705d3f45aff..e56ad4b9330f 100644 --- a/net/mac80211/mlme.c +++ b/net/mac80211/mlme.c @@ -6108,9 +6108,10 @@ static int ieee80211_prep_channel(struct ieee80211_sub_if_data *sdata, ret = ieee80211_link_use_channel(link, &chanreq, IEEE80211_CHANCTX_SHARED); - /* don't downgrade for 5 and 10 MHz channels, though. */ + /* don't downgrade for 5/10/S1G MHz channels, though. */ if (chanreq.oper.width == NL80211_CHAN_WIDTH_5 || - chanreq.oper.width == NL80211_CHAN_WIDTH_10) + chanreq.oper.width == NL80211_CHAN_WIDTH_10 || + cfg80211_chandef_is_s1g(&chanreq.oper)) return ret; while (ret && chanreq.oper.width != NL80211_CHAN_WIDTH_20_NOHT) { From 66e75b2758d69dce908fe98da72c345fa18a46f4 Mon Sep 17 00:00:00 2001 From: Oliver Hartkopp Date: Wed, 26 Nov 2025 11:16:02 +0100 Subject: [PATCH 1144/1381] can: dev: can_get_ctrlmode_str: use capitalized ctrlmode strings Unify the ctrlmode related strings to the command line options of the 'ip' tool from the iproute2 package. The capitalized strings are also shown when the detailed interface configuration is printed by 'ip'. Suggested-by: Stephane Grosjean Signed-off-by: Oliver Hartkopp Link: https://patch.msgid.link/20251126-canxl-v8-1-e7e3eb74f889@pengutronix.de Signed-off-by: Marc Kleine-Budde --- drivers/net/can/dev/dev.c | 24 ++++++++++++------------ 1 file changed, 12 insertions(+), 12 deletions(-) diff --git a/drivers/net/can/dev/dev.c b/drivers/net/can/dev/dev.c index 80e1ab18de87..b392483e4499 100644 --- a/drivers/net/can/dev/dev.c +++ b/drivers/net/can/dev/dev.c @@ -92,29 +92,29 @@ const char *can_get_ctrlmode_str(u32 ctrlmode) { switch (ctrlmode & ~(ctrlmode - 1)) { case 0: - return "none"; + return "(none)"; case CAN_CTRLMODE_LOOPBACK: - return "loopback"; + return "LOOPBACK"; case CAN_CTRLMODE_LISTENONLY: - return "listen-only"; + return "LISTEN-ONLY"; case CAN_CTRLMODE_3_SAMPLES: - return "triple-sampling"; + return "TRIPLE-SAMPLING"; case CAN_CTRLMODE_ONE_SHOT: - return "one-shot"; + return "ONE-SHOT"; case CAN_CTRLMODE_BERR_REPORTING: - return "berr-reporting"; + return "BERR-REPORTING"; case CAN_CTRLMODE_FD: - return "fd"; + return "FD"; case CAN_CTRLMODE_PRESUME_ACK: - return "presume-ack"; + return "PRESUME-ACK"; case CAN_CTRLMODE_FD_NON_ISO: - return "fd-non-iso"; + return "FD-NON-ISO"; case CAN_CTRLMODE_CC_LEN8_DLC: - return "cc-len8-dlc"; + return "CC-LEN8-DLC"; case CAN_CTRLMODE_TDC_AUTO: - return "fd-tdc-auto"; + return "TDC-AUTO"; case CAN_CTRLMODE_TDC_MANUAL: - return "fd-tdc-manual"; + return "TDC-MANUAL"; default: return ""; } From 585a4f22c4f9d85e32d42be65e67c232e82e5b3a Mon Sep 17 00:00:00 2001 From: Vincent Mailhol Date: Wed, 26 Nov 2025 11:16:03 +0100 Subject: [PATCH 1145/1381] can: bittiming: apply NL_SET_ERR_MSG() to can_calc_bittiming() When CONFIG_CAN_CALC_BITTIMING is disabled, the can_calc_bittiming() functions can not be used and the user needs to provide all the bittiming parameters. Currently, can_calc_bittiming() prints an error message to the kernel log. Instead use NL_SET_ERR_MSG() to make it return the error message through the netlink interface so that the user can directly see it. Signed-off-by: Vincent Mailhol Signed-off-by: Oliver Hartkopp Link: https://patch.msgid.link/20251126-canxl-v8-2-e7e3eb74f889@pengutronix.de Signed-off-by: Marc Kleine-Budde --- include/linux/can/bittiming.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/can/bittiming.h b/include/linux/can/bittiming.h index d30816dd93c7..3926c78b2222 100644 --- a/include/linux/can/bittiming.h +++ b/include/linux/can/bittiming.h @@ -141,7 +141,7 @@ static inline int can_calc_bittiming(const struct net_device *dev, struct can_bittiming *bt, const struct can_bittiming_const *btc, struct netlink_ext_ack *extack) { - netdev_err(dev, "bit-timing calculation not available\n"); + NL_SET_ERR_MSG(extack, "bit-timing calculation not available\n"); return -EINVAL; } From d037d05c2e32792a6fa572b0aa3c92a8ac78589d Mon Sep 17 00:00:00 2001 From: Vincent Mailhol Date: Wed, 26 Nov 2025 11:16:04 +0100 Subject: [PATCH 1146/1381] can: dev: can_dev_dropped_skb: drop CAN FD skbs if FD is off Currently, the CAN FD skb validation logic is based on the MTU: the interface is deemed FD capable if and only if its MTU is greater or equal to CANFD_MTU. This logic is showing its limit with the introduction of CAN XL. For example, consider the two scenarios below: 1. An interface configured with CAN FD on and CAN XL on 2. An interface configured with CAN FD off and CAN XL on In those two scenarios, the interfaces would have the same MTU: CANXL_MTU making it impossible to differentiate which one has CAN FD turned on and which one has it off. Because of the limitation, the only non-UAPI-breaking workaround is to do the check at the device level using the can_priv->ctrlmode flags. Unfortunately, the virtual interfaces (vcan, vxcan), which do not have a can_priv, are left behind. Add a check on the CAN_CTRLMODE_FD flag in can_dev_dropped_skb() and drop FD frames whenever the feature is turned off. Signed-off-by: Vincent Mailhol Signed-off-by: Oliver Hartkopp Link: https://patch.msgid.link/20251126-canxl-v8-3-e7e3eb74f889@pengutronix.de Signed-off-by: Marc Kleine-Budde --- include/linux/can/dev.h | 14 +++++++++++--- 1 file changed, 11 insertions(+), 3 deletions(-) diff --git a/include/linux/can/dev.h b/include/linux/can/dev.h index bd7410b5d8a6..a7a39a6101d9 100644 --- a/include/linux/can/dev.h +++ b/include/linux/can/dev.h @@ -103,12 +103,20 @@ static inline bool can_dev_dropped_skb(struct net_device *dev, struct sk_buff *s if (priv->ctrlmode & CAN_CTRLMODE_LISTENONLY) { netdev_info_once(dev, "interface in listen only mode, dropping skb\n"); - kfree_skb(skb); - dev->stats.tx_dropped++; - return true; + goto invalid_skb; + } + + if (!(priv->ctrlmode & CAN_CTRLMODE_FD) && can_is_canfd_skb(skb)) { + netdev_info_once(dev, "CAN FD is disabled, dropping skb\n"); + goto invalid_skb; } return can_dropped_invalid_skb(dev, skb); + +invalid_skb: + kfree_skb(skb); + dev->stats.tx_dropped++; + return true; } void can_setup(struct net_device *dev); From 60f511f443e552ef5b5cd79ec2b881f4323e19c9 Mon Sep 17 00:00:00 2001 From: Vincent Mailhol Date: Wed, 26 Nov 2025 11:16:05 +0100 Subject: [PATCH 1147/1381] can: netlink: add CAN_CTRLMODE_RESTRICTED ISO 11898-1:2024 adds a new restricted operation mode. This mode is added as a mandatory feature for nodes which support CAN XL and is retrofitted as optional for legacy nodes (i.e. the ones which only support Classical CAN and CAN FD). The restricted operation mode is nearly the same as the listen only mode: the node can not send data frames or remote frames and can not send dominant bits if an error occurs. The only exception is that the node shall still send the acknowledgment bit. A second niche exception is that the node may still send a data frame containing a time reference message if the node is a primary time provider, but because the time provider feature is not yet implemented in the kernel, this second exception is not relevant to us at the moment. Add the CAN_CTRLMODE_RESTRICTED control mode flag and update the can_dev_dropped_skb() helper function accordingly. Finally, bail out if both CAN_CTRLMODE_LISTENONLY and CAN_CTRLMODE_RESTRICTED are provided. Signed-off-by: Vincent Mailhol Signed-off-by: Oliver Hartkopp Link: https://patch.msgid.link/20251126-canxl-v8-4-e7e3eb74f889@pengutronix.de Signed-off-by: Marc Kleine-Budde --- drivers/net/can/dev/dev.c | 2 ++ drivers/net/can/dev/netlink.c | 7 +++++ include/linux/can/dev.h | 50 +++++++++++++++++--------------- include/uapi/linux/can/netlink.h | 1 + 4 files changed, 36 insertions(+), 24 deletions(-) diff --git a/drivers/net/can/dev/dev.c b/drivers/net/can/dev/dev.c index b392483e4499..b6980d32e5b4 100644 --- a/drivers/net/can/dev/dev.c +++ b/drivers/net/can/dev/dev.c @@ -115,6 +115,8 @@ const char *can_get_ctrlmode_str(u32 ctrlmode) return "TDC-AUTO"; case CAN_CTRLMODE_TDC_MANUAL: return "TDC-MANUAL"; + case CAN_CTRLMODE_RESTRICTED: + return "RESTRICTED"; default: return ""; } diff --git a/drivers/net/can/dev/netlink.c b/drivers/net/can/dev/netlink.c index 6f83b87d54fc..87e731527dd7 100644 --- a/drivers/net/can/dev/netlink.c +++ b/drivers/net/can/dev/netlink.c @@ -188,6 +188,13 @@ static int can_validate(struct nlattr *tb[], struct nlattr *data[], struct can_ctrlmode *cm = nla_data(data[IFLA_CAN_CTRLMODE]); flags = cm->flags & cm->mask; + + if ((flags & CAN_CTRLMODE_LISTENONLY) && + (flags & CAN_CTRLMODE_RESTRICTED)) { + NL_SET_ERR_MSG(extack, + "LISTEN-ONLY and RESTRICTED modes are mutually exclusive"); + return -EOPNOTSUPP; + } } err = can_validate_bittiming(data, extack, IFLA_CAN_BITTIMING); diff --git a/include/linux/can/dev.h b/include/linux/can/dev.h index a7a39a6101d9..ab11c0e9111b 100644 --- a/include/linux/can/dev.h +++ b/include/linux/can/dev.h @@ -95,30 +95,6 @@ static inline bool can_is_canxl_dev_mtu(unsigned int mtu) return (mtu >= CANXL_MIN_MTU && mtu <= CANXL_MAX_MTU); } -/* drop skb if it does not contain a valid CAN frame for sending */ -static inline bool can_dev_dropped_skb(struct net_device *dev, struct sk_buff *skb) -{ - struct can_priv *priv = netdev_priv(dev); - - if (priv->ctrlmode & CAN_CTRLMODE_LISTENONLY) { - netdev_info_once(dev, - "interface in listen only mode, dropping skb\n"); - goto invalid_skb; - } - - if (!(priv->ctrlmode & CAN_CTRLMODE_FD) && can_is_canfd_skb(skb)) { - netdev_info_once(dev, "CAN FD is disabled, dropping skb\n"); - goto invalid_skb; - } - - return can_dropped_invalid_skb(dev, skb); - -invalid_skb: - kfree_skb(skb); - dev->stats.tx_dropped++; - return true; -} - void can_setup(struct net_device *dev); struct net_device *alloc_candev_mqs(int sizeof_priv, unsigned int echo_skb_max, @@ -154,6 +130,32 @@ void can_bus_off(struct net_device *dev); const char *can_get_state_str(const enum can_state state); const char *can_get_ctrlmode_str(u32 ctrlmode); +/* drop skb if it does not contain a valid CAN frame for sending */ +static inline bool can_dev_dropped_skb(struct net_device *dev, struct sk_buff *skb) +{ + struct can_priv *priv = netdev_priv(dev); + u32 silent_mode = priv->ctrlmode & (CAN_CTRLMODE_LISTENONLY | + CAN_CTRLMODE_RESTRICTED); + + if (silent_mode) { + netdev_info_once(dev, "interface in %s mode, dropping skb\n", + can_get_ctrlmode_str(silent_mode)); + goto invalid_skb; + } + + if (!(priv->ctrlmode & CAN_CTRLMODE_FD) && can_is_canfd_skb(skb)) { + netdev_info_once(dev, "CAN FD is disabled, dropping skb\n"); + goto invalid_skb; + } + + return can_dropped_invalid_skb(dev, skb); + +invalid_skb: + kfree_skb(skb); + dev->stats.tx_dropped++; + return true; +} + void can_state_get_by_berr_counter(const struct net_device *dev, const struct can_berr_counter *bec, enum can_state *tx_state, diff --git a/include/uapi/linux/can/netlink.h b/include/uapi/linux/can/netlink.h index ef62f56eaaef..fafd1cce4798 100644 --- a/include/uapi/linux/can/netlink.h +++ b/include/uapi/linux/can/netlink.h @@ -103,6 +103,7 @@ struct can_ctrlmode { #define CAN_CTRLMODE_CC_LEN8_DLC 0x100 /* Classic CAN DLC option */ #define CAN_CTRLMODE_TDC_AUTO 0x200 /* FD transceiver automatically calculates TDCV */ #define CAN_CTRLMODE_TDC_MANUAL 0x400 /* FD TDCV is manually set up by user */ +#define CAN_CTRLMODE_RESTRICTED 0x800 /* Restricted operation mode */ /* * CAN device statistics From e63281614747c73f25b708c75bc696c4e76f5588 Mon Sep 17 00:00:00 2001 From: Vincent Mailhol Date: Wed, 26 Nov 2025 11:16:06 +0100 Subject: [PATCH 1148/1381] can: netlink: add initial CAN XL support CAN XL uses bittiming parameters different from Classical CAN and CAN FD. Thus, all the data bittiming parameters, including TDC, need to be duplicated for CAN XL. Add the CAN XL netlink interface for all the features which are common with CAN FD. Any new CAN XL specific features are added later on. The first time CAN XL is activated, the MTU is set by default to CANXL_MAX_MTU. The user may then configure a custom MTU within the CANXL_MIN_MTU to CANXL_MAX_MTU range, in which case, the custom MTU value will be kept as long as CAN XL remains active. Signed-off-by: Vincent Mailhol Signed-off-by: Oliver Hartkopp Link: https://patch.msgid.link/20251126-canxl-v8-5-e7e3eb74f889@pengutronix.de Signed-off-by: Marc Kleine-Budde --- drivers/net/can/dev/dev.c | 14 +++++- drivers/net/can/dev/netlink.c | 76 +++++++++++++++++++++++++------- include/linux/can/bittiming.h | 6 ++- include/linux/can/dev.h | 7 ++- include/uapi/linux/can/netlink.h | 7 +++ 5 files changed, 90 insertions(+), 20 deletions(-) diff --git a/drivers/net/can/dev/dev.c b/drivers/net/can/dev/dev.c index b6980d32e5b4..bdec2c52c8ec 100644 --- a/drivers/net/can/dev/dev.c +++ b/drivers/net/can/dev/dev.c @@ -117,6 +117,12 @@ const char *can_get_ctrlmode_str(u32 ctrlmode) return "TDC-MANUAL"; case CAN_CTRLMODE_RESTRICTED: return "RESTRICTED"; + case CAN_CTRLMODE_XL: + return "XL"; + case CAN_CTRLMODE_XL_TDC_AUTO: + return "XL-TDC-AUTO"; + case CAN_CTRLMODE_XL_TDC_MANUAL: + return "XL-TDC-MANUAL"; default: return ""; } @@ -350,7 +356,13 @@ void can_set_default_mtu(struct net_device *dev) { struct can_priv *priv = netdev_priv(dev); - if (priv->ctrlmode & CAN_CTRLMODE_FD) { + if (priv->ctrlmode & CAN_CTRLMODE_XL) { + if (can_is_canxl_dev_mtu(dev->mtu)) + return; + dev->mtu = CANXL_MTU; + dev->min_mtu = CANXL_MIN_MTU; + dev->max_mtu = CANXL_MAX_MTU; + } else if (priv->ctrlmode & CAN_CTRLMODE_FD) { dev->mtu = CANFD_MTU; dev->min_mtu = CANFD_MTU; dev->max_mtu = CANFD_MTU; diff --git a/drivers/net/can/dev/netlink.c b/drivers/net/can/dev/netlink.c index 87e731527dd7..fdd1fa7cf93a 100644 --- a/drivers/net/can/dev/netlink.c +++ b/drivers/net/can/dev/netlink.c @@ -2,7 +2,7 @@ /* Copyright (C) 2005 Marc Kleine-Budde, Pengutronix * Copyright (C) 2006 Andrey Volkov, Varma Electronics * Copyright (C) 2008-2009 Wolfgang Grandegger - * Copyright (C) 2021 Vincent Mailhol + * Copyright (C) 2021-2025 Vincent Mailhol */ #include @@ -22,6 +22,9 @@ static const struct nla_policy can_policy[IFLA_CAN_MAX + 1] = { [IFLA_CAN_TERMINATION] = { .type = NLA_U16 }, [IFLA_CAN_TDC] = { .type = NLA_NESTED }, [IFLA_CAN_CTRLMODE_EXT] = { .type = NLA_NESTED }, + [IFLA_CAN_XL_DATA_BITTIMING] = { .len = sizeof(struct can_bittiming) }, + [IFLA_CAN_XL_DATA_BITTIMING_CONST] = { .len = sizeof(struct can_bittiming_const) }, + [IFLA_CAN_XL_TDC] = { .type = NLA_NESTED }, }; static const struct nla_policy can_tdc_policy[IFLA_CAN_TDC_MAX + 1] = { @@ -70,7 +73,7 @@ static int can_validate_tdc(struct nlattr *data_tdc, return -EOPNOTSUPP; } - /* If one of the CAN_CTRLMODE_TDC_* flag is set then TDC + /* If one of the CAN_CTRLMODE_{,XL}_TDC_* flags is set then TDC * must be set and vice-versa */ if ((tdc_auto || tdc_manual) && !data_tdc) { @@ -82,8 +85,8 @@ static int can_validate_tdc(struct nlattr *data_tdc, return -EOPNOTSUPP; } - /* If providing TDC parameters, at least TDCO is needed. TDCV - * is needed if and only if CAN_CTRLMODE_TDC_MANUAL is set + /* If providing TDC parameters, at least TDCO is needed. TDCV is + * needed if and only if CAN_CTRLMODE_{,XL}_TDC_MANUAL is set */ if (data_tdc) { struct nlattr *tb_tdc[IFLA_CAN_TDC_MAX + 1]; @@ -126,10 +129,10 @@ static int can_validate_databittiming(struct nlattr *data[], bool is_on; int err; - /* Make sure that valid CAN FD configurations always consist of + /* Make sure that valid CAN FD/XL configurations always consist of * - nominal/arbitration bittiming * - data bittiming - * - control mode with CAN_CTRLMODE_FD set + * - control mode with CAN_CTRLMODE_{FD,XL} set * - TDC parameters are coherent (details in can_validate_tdc()) */ @@ -139,7 +142,10 @@ static int can_validate_databittiming(struct nlattr *data[], is_on = flags & CAN_CTRLMODE_FD; type = "FD"; } else { - return -EOPNOTSUPP; /* Place holder for CAN XL */ + data_tdc = data[IFLA_CAN_XL_TDC]; + tdc_flags = flags & CAN_CTRLMODE_XL_TDC_MASK; + is_on = flags & CAN_CTRLMODE_XL; + type = "XL"; } if (is_on) { @@ -206,6 +212,11 @@ static int can_validate(struct nlattr *tb[], struct nlattr *data[], if (err) return err; + err = can_validate_databittiming(data, extack, + IFLA_CAN_XL_DATA_BITTIMING, flags); + if (err) + return err; + return 0; } @@ -251,18 +262,26 @@ static int can_ctrlmode_changelink(struct net_device *dev, /* If a top dependency flag is provided, reset all its dependencies */ if (cm->mask & CAN_CTRLMODE_FD) priv->ctrlmode &= ~CAN_CTRLMODE_FD_TDC_MASK; + if (cm->mask & CAN_CTRLMODE_XL) + priv->ctrlmode &= ~(CAN_CTRLMODE_XL_TDC_MASK); /* clear bits to be modified and copy the flag values */ priv->ctrlmode &= ~cm->mask; priv->ctrlmode |= maskedflags; - /* Wipe potential leftovers from previous CAN FD config */ + /* Wipe potential leftovers from previous CAN FD/XL config */ if (!(priv->ctrlmode & CAN_CTRLMODE_FD)) { memset(&priv->fd.data_bittiming, 0, sizeof(priv->fd.data_bittiming)); priv->ctrlmode &= ~CAN_CTRLMODE_FD_TDC_MASK; memset(&priv->fd.tdc, 0, sizeof(priv->fd.tdc)); } + if (!(priv->ctrlmode & CAN_CTRLMODE_XL)) { + memset(&priv->xl.data_bittiming, 0, + sizeof(priv->fd.data_bittiming)); + priv->ctrlmode &= ~CAN_CTRLMODE_XL_TDC_MASK; + memset(&priv->xl.tdc, 0, sizeof(priv->xl.tdc)); + } can_set_default_mtu(dev); @@ -337,7 +356,10 @@ static int can_dbt_changelink(struct net_device *dev, struct nlattr *data[], dbt_params = &priv->fd; tdc_mask = CAN_CTRLMODE_FD_TDC_MASK; } else { - return -EOPNOTSUPP; /* Place holder for CAN XL */ + data_bittiming = data[IFLA_CAN_XL_DATA_BITTIMING]; + data_tdc = data[IFLA_CAN_XL_TDC]; + dbt_params = &priv->xl; + tdc_mask = CAN_CTRLMODE_XL_TDC_MASK; } if (!data_bittiming) @@ -388,7 +410,7 @@ static int can_dbt_changelink(struct net_device *dev, struct nlattr *data[], */ can_calc_tdco(&dbt_params->tdc, dbt_params->tdc_const, &dbt, tdc_mask, &priv->ctrlmode, priv->ctrlmode_supported); - } /* else: both CAN_CTRLMODE_TDC_{AUTO,MANUAL} are explicitly + } /* else: both CAN_CTRLMODE_{,XL}_TDC_{AUTO,MANUAL} are explicitly * turned off. TDC is disabled: do nothing */ @@ -493,6 +515,11 @@ static int can_changelink(struct net_device *dev, struct nlattr *tb[], if (err) return err; + /* CAN XL */ + err = can_dbt_changelink(dev, data, false, extack); + if (err) + return err; + if (data[IFLA_CAN_TERMINATION]) { const u16 termval = nla_get_u16(data[IFLA_CAN_TERMINATION]); const unsigned int num_term = priv->termination_const_cnt; @@ -560,14 +587,14 @@ static size_t can_data_bittiming_get_size(struct data_bittiming_params *dbt_para { size_t size = 0; - if (dbt_params->data_bittiming.bitrate) /* IFLA_CAN_DATA_BITTIMING */ + if (dbt_params->data_bittiming.bitrate) /* IFLA_CAN_{,XL}_DATA_BITTIMING */ size += nla_total_size(sizeof(dbt_params->data_bittiming)); - if (dbt_params->data_bittiming_const) /* IFLA_CAN_DATA_BITTIMING_CONST */ + if (dbt_params->data_bittiming_const) /* IFLA_CAN_{,XL}_DATA_BITTIMING_CONST */ size += nla_total_size(sizeof(*dbt_params->data_bittiming_const)); - if (dbt_params->data_bitrate_const) /* IFLA_CAN_DATA_BITRATE_CONST */ + if (dbt_params->data_bitrate_const) /* IFLA_CAN_{,XL}_DATA_BITRATE_CONST */ size += nla_total_size(sizeof(*dbt_params->data_bitrate_const) * dbt_params->data_bitrate_const_cnt); - size += can_tdc_get_size(dbt_params, tdc_flags);/* IFLA_CAN_TDC */ + size += can_tdc_get_size(dbt_params, tdc_flags);/* IFLA_CAN_{,XL}_TDC */ return size; } @@ -607,6 +634,9 @@ static size_t can_get_size(const struct net_device *dev) size += can_data_bittiming_get_size(&priv->fd, priv->ctrlmode & CAN_CTRLMODE_FD_TDC_MASK); + size += can_data_bittiming_get_size(&priv->xl, + priv->ctrlmode & CAN_CTRLMODE_XL_TDC_MASK); + return size; } @@ -651,7 +681,9 @@ static int can_tdc_fill_info(struct sk_buff *skb, const struct net_device *dev, tdc_is_enabled = can_fd_tdc_is_enabled(priv); tdc_manual = priv->ctrlmode & CAN_CTRLMODE_TDC_MANUAL; } else { - return -EOPNOTSUPP; /* Place holder for CAN XL */ + dbt_params = &priv->xl; + tdc_is_enabled = can_xl_tdc_is_enabled(priv); + tdc_manual = priv->ctrlmode & CAN_CTRLMODE_XL_TDC_MANUAL; } tdc_const = dbt_params->tdc_const; tdc = &dbt_params->tdc; @@ -773,7 +805,19 @@ static int can_fill_info(struct sk_buff *skb, const struct net_device *dev) can_tdc_fill_info(skb, dev, IFLA_CAN_TDC) || - can_ctrlmode_ext_fill_info(skb, priv) + can_ctrlmode_ext_fill_info(skb, priv) || + + can_bittiming_fill_info(skb, IFLA_CAN_XL_DATA_BITTIMING, + &priv->xl.data_bittiming) || + + can_bittiming_const_fill_info(skb, IFLA_CAN_XL_DATA_BITTIMING_CONST, + priv->xl.data_bittiming_const) || + + can_bitrate_const_fill_info(skb, IFLA_CAN_XL_DATA_BITRATE_CONST, + priv->xl.data_bitrate_const, + priv->xl.data_bitrate_const_cnt) || + + can_tdc_fill_info(skb, dev, IFLA_CAN_XL_TDC) ) return -EMSGSIZE; diff --git a/include/linux/can/bittiming.h b/include/linux/can/bittiming.h index 3926c78b2222..b6cd2476ffd7 100644 --- a/include/linux/can/bittiming.h +++ b/include/linux/can/bittiming.h @@ -16,10 +16,12 @@ #define CAN_CTRLMODE_FD_TDC_MASK \ (CAN_CTRLMODE_TDC_AUTO | CAN_CTRLMODE_TDC_MANUAL) +#define CAN_CTRLMODE_XL_TDC_MASK \ + (CAN_CTRLMODE_XL_TDC_AUTO | CAN_CTRLMODE_XL_TDC_MANUAL) #define CAN_CTRLMODE_TDC_AUTO_MASK \ - (CAN_CTRLMODE_TDC_AUTO) + (CAN_CTRLMODE_TDC_AUTO | CAN_CTRLMODE_XL_TDC_AUTO) #define CAN_CTRLMODE_TDC_MANUAL_MASK \ - (CAN_CTRLMODE_TDC_MANUAL) + (CAN_CTRLMODE_TDC_MANUAL | CAN_CTRLMODE_XL_TDC_MANUAL) /* * struct can_tdc - CAN FD Transmission Delay Compensation parameters diff --git a/include/linux/can/dev.h b/include/linux/can/dev.h index ab11c0e9111b..f15879bd818d 100644 --- a/include/linux/can/dev.h +++ b/include/linux/can/dev.h @@ -47,7 +47,7 @@ struct can_priv { const struct can_bittiming_const *bittiming_const; struct can_bittiming bittiming; - struct data_bittiming_params fd; + struct data_bittiming_params fd, xl; unsigned int bitrate_const_cnt; const u32 *bitrate_const; u32 bitrate_max; @@ -85,6 +85,11 @@ static inline bool can_fd_tdc_is_enabled(const struct can_priv *priv) return !!(priv->ctrlmode & CAN_CTRLMODE_FD_TDC_MASK); } +static inline bool can_xl_tdc_is_enabled(const struct can_priv *priv) +{ + return !!(priv->ctrlmode & CAN_CTRLMODE_XL_TDC_MASK); +} + static inline u32 can_get_static_ctrlmode(struct can_priv *priv) { return priv->ctrlmode & ~priv->ctrlmode_supported; diff --git a/include/uapi/linux/can/netlink.h b/include/uapi/linux/can/netlink.h index fafd1cce4798..c2c96c5978a8 100644 --- a/include/uapi/linux/can/netlink.h +++ b/include/uapi/linux/can/netlink.h @@ -104,6 +104,9 @@ struct can_ctrlmode { #define CAN_CTRLMODE_TDC_AUTO 0x200 /* FD transceiver automatically calculates TDCV */ #define CAN_CTRLMODE_TDC_MANUAL 0x400 /* FD TDCV is manually set up by user */ #define CAN_CTRLMODE_RESTRICTED 0x800 /* Restricted operation mode */ +#define CAN_CTRLMODE_XL 0x1000 /* CAN XL mode */ +#define CAN_CTRLMODE_XL_TDC_AUTO 0x2000 /* XL transceiver automatically calculates TDCV */ +#define CAN_CTRLMODE_XL_TDC_MANUAL 0x4000 /* XL TDCV is manually set up by user */ /* * CAN device statistics @@ -139,6 +142,10 @@ enum { IFLA_CAN_BITRATE_MAX, IFLA_CAN_TDC, /* FD */ IFLA_CAN_CTRLMODE_EXT, + IFLA_CAN_XL_DATA_BITTIMING, + IFLA_CAN_XL_DATA_BITTIMING_CONST, + IFLA_CAN_XL_DATA_BITRATE_CONST, + IFLA_CAN_XL_TDC, /* add new constants above here */ __IFLA_CAN_MAX, From 233134af208689c2d5d40896f5740473a74e3cb2 Mon Sep 17 00:00:00 2001 From: Vincent Mailhol Date: Wed, 26 Nov 2025 11:16:07 +0100 Subject: [PATCH 1149/1381] can: netlink: add CAN_CTRLMODE_XL_TMS flag The Transceiver Mode Switching (TMS) indicates whether the CAN XL controller shall use the PWM or NRZ encoding during the data phase. The term "transceiver mode switching" is used in both ISO 11898-1 and CiA 612-2 (although only the latter one uses the abbreviation TMS). We adopt the same naming convention here for consistency. Add the CAN_CTRLMODE_XL_TMS flag to the list of the CAN control modes. Add can_validate_xl_flags() to check the coherency of the TMS flag. That function will be reused in upcoming changes to validate the other CAN XL flags. Signed-off-by: Vincent Mailhol Signed-off-by: Oliver Hartkopp Link: https://patch.msgid.link/20251126-canxl-v8-6-e7e3eb74f889@pengutronix.de Signed-off-by: Marc Kleine-Budde --- drivers/net/can/dev/dev.c | 2 ++ drivers/net/can/dev/netlink.c | 48 ++++++++++++++++++++++++++++++-- include/uapi/linux/can/netlink.h | 1 + 3 files changed, 48 insertions(+), 3 deletions(-) diff --git a/drivers/net/can/dev/dev.c b/drivers/net/can/dev/dev.c index bdec2c52c8ec..091f30e94c61 100644 --- a/drivers/net/can/dev/dev.c +++ b/drivers/net/can/dev/dev.c @@ -123,6 +123,8 @@ const char *can_get_ctrlmode_str(u32 ctrlmode) return "XL-TDC-AUTO"; case CAN_CTRLMODE_XL_TDC_MANUAL: return "XL-TDC-MANUAL"; + case CAN_CTRLMODE_XL_TMS: + return "TMS"; default: return ""; } diff --git a/drivers/net/can/dev/netlink.c b/drivers/net/can/dev/netlink.c index fdd1fa7cf93a..b2c24439abba 100644 --- a/drivers/net/can/dev/netlink.c +++ b/drivers/net/can/dev/netlink.c @@ -181,6 +181,32 @@ static int can_validate_databittiming(struct nlattr *data[], return 0; } +static int can_validate_xl_flags(struct netlink_ext_ack *extack, + u32 masked_flags, u32 mask) +{ + if (masked_flags & CAN_CTRLMODE_XL) { + if (masked_flags & CAN_CTRLMODE_XL_TMS) { + const u32 tms_conflicts_mask = CAN_CTRLMODE_FD | + CAN_CTRLMODE_XL_TDC_MASK; + u32 tms_conflicts = masked_flags & tms_conflicts_mask; + + if (tms_conflicts) { + NL_SET_ERR_MSG_FMT(extack, + "TMS and %s are mutually exclusive", + can_get_ctrlmode_str(tms_conflicts)); + return -EOPNOTSUPP; + } + } + } else { + if (mask & CAN_CTRLMODE_XL_TMS) { + NL_SET_ERR_MSG(extack, "TMS requires CAN XL"); + return -EOPNOTSUPP; + } + } + + return 0; +} + static int can_validate(struct nlattr *tb[], struct nlattr *data[], struct netlink_ext_ack *extack) { @@ -201,6 +227,10 @@ static int can_validate(struct nlattr *tb[], struct nlattr *data[], "LISTEN-ONLY and RESTRICTED modes are mutually exclusive"); return -EOPNOTSUPP; } + + err = can_validate_xl_flags(extack, flags, cm->mask); + if (err) + return err; } err = can_validate_bittiming(data, extack, IFLA_CAN_BITTIMING); @@ -226,7 +256,7 @@ static int can_ctrlmode_changelink(struct net_device *dev, { struct can_priv *priv = netdev_priv(dev); struct can_ctrlmode *cm; - u32 ctrlstatic, maskedflags, notsupp, ctrlstatic_missing; + u32 ctrlstatic, maskedflags, deactivated, notsupp, ctrlstatic_missing; if (!data[IFLA_CAN_CTRLMODE]) return 0; @@ -238,6 +268,7 @@ static int can_ctrlmode_changelink(struct net_device *dev, cm = nla_data(data[IFLA_CAN_CTRLMODE]); ctrlstatic = can_get_static_ctrlmode(priv); maskedflags = cm->flags & cm->mask; + deactivated = ~cm->flags & cm->mask; notsupp = maskedflags & ~(priv->ctrlmode_supported | ctrlstatic); ctrlstatic_missing = (maskedflags & ctrlstatic) ^ ctrlstatic; @@ -259,11 +290,21 @@ static int can_ctrlmode_changelink(struct net_device *dev, return -EOPNOTSUPP; } + /* If FD was active and is not turned off, check for XL conflicts */ + if (priv->ctrlmode & CAN_CTRLMODE_FD & ~deactivated) { + if (maskedflags & CAN_CTRLMODE_XL_TMS) { + NL_SET_ERR_MSG(extack, + "TMS can not be activated while CAN FD is on"); + return -EOPNOTSUPP; + } + } + /* If a top dependency flag is provided, reset all its dependencies */ if (cm->mask & CAN_CTRLMODE_FD) priv->ctrlmode &= ~CAN_CTRLMODE_FD_TDC_MASK; if (cm->mask & CAN_CTRLMODE_XL) - priv->ctrlmode &= ~(CAN_CTRLMODE_XL_TDC_MASK); + priv->ctrlmode &= ~(CAN_CTRLMODE_XL_TDC_MASK | + CAN_CTRLMODE_XL_TMS); /* clear bits to be modified and copy the flag values */ priv->ctrlmode &= ~cm->mask; @@ -395,7 +436,8 @@ static int can_dbt_changelink(struct net_device *dev, struct nlattr *data[], if (data[IFLA_CAN_CTRLMODE]) { struct can_ctrlmode *cm = nla_data(data[IFLA_CAN_CTRLMODE]); - need_tdc_calc = !(cm->mask & tdc_mask); + if (fd || !(priv->ctrlmode & CAN_CTRLMODE_XL_TMS)) + need_tdc_calc = !(cm->mask & tdc_mask); } if (data_tdc) { /* TDC parameters are provided: use them */ diff --git a/include/uapi/linux/can/netlink.h b/include/uapi/linux/can/netlink.h index c2c96c5978a8..ebafb091d80f 100644 --- a/include/uapi/linux/can/netlink.h +++ b/include/uapi/linux/can/netlink.h @@ -107,6 +107,7 @@ struct can_ctrlmode { #define CAN_CTRLMODE_XL 0x1000 /* CAN XL mode */ #define CAN_CTRLMODE_XL_TDC_AUTO 0x2000 /* XL transceiver automatically calculates TDCV */ #define CAN_CTRLMODE_XL_TDC_MANUAL 0x4000 /* XL TDCV is manually set up by user */ +#define CAN_CTRLMODE_XL_TMS 0x8000 /* Transceiver Mode Switching */ /* * CAN device statistics From 6df01533e535d21cac779ff35cc25c43304035c3 Mon Sep 17 00:00:00 2001 From: Oliver Hartkopp Date: Wed, 26 Nov 2025 11:16:08 +0100 Subject: [PATCH 1150/1381] can: dev: can_dev_dropped_skb: drop CC/FD frames in CANXL-only mode The error-signalling (ES) is a mandatory functionality for CAN CC and CAN FD to report CAN frame format violations by sending an error-frame signal on the bus. A so-called 'mixed-mode' is intended to have (XL-tolerant) CAN FD nodes and CAN XL nodes on one CAN segment, where the FD-controllers can talk CC/FD and the XL-controllers can talk CC/FD/XL. This mixed-mode utilizes the error-signalling for sending CC/FD/XL frames. The CANXL-only mode disables the error-signalling in the CAN XL controller. This mode does not allow CC/FD frames to be sent but additionally offers a CAN XL transceiver mode switching (TMS). Configured with CAN_CTRLMODE_FD and CAN_CTRLMODE_XL this leads to: FD=0 XL=0 CC-only mode (ES=1) FD=1 XL=0 FD/CC mixed-mode (ES=1) FD=1 XL=1 XL/FD/CC mixed-mode (ES=1) FD=0 XL=1 XL-only mode (ES=0, TMS optional) The helper function can_dev_in_xl_only_mode() determines the required value to disable error signalling in the CAN XL controller. Signed-off-by: Oliver Hartkopp Link: https://patch.msgid.link/20251126-canxl-v8-7-e7e3eb74f889@pengutronix.de Signed-off-by: Marc Kleine-Budde --- include/linux/can/dev.h | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+) diff --git a/include/linux/can/dev.h b/include/linux/can/dev.h index f15879bd818d..52c8be5c160e 100644 --- a/include/linux/can/dev.h +++ b/include/linux/can/dev.h @@ -135,6 +135,19 @@ void can_bus_off(struct net_device *dev); const char *can_get_state_str(const enum can_state state); const char *can_get_ctrlmode_str(u32 ctrlmode); +static inline bool can_dev_in_xl_only_mode(struct can_priv *priv) +{ + const u32 mixed_mode = CAN_CTRLMODE_FD | CAN_CTRLMODE_XL; + + /* When CAN XL is enabled but FD is disabled we are running in + * the so-called 'CANXL-only mode' where the error signalling is + * disabled. This helper function determines the required value + * to disable error signalling in the CAN XL controller. + * The so-called CC/FD/XL 'mixed mode' requires error signalling. + */ + return ((priv->ctrlmode & mixed_mode) == CAN_CTRLMODE_XL); +} + /* drop skb if it does not contain a valid CAN frame for sending */ static inline bool can_dev_dropped_skb(struct net_device *dev, struct sk_buff *skb) { @@ -153,6 +166,12 @@ static inline bool can_dev_dropped_skb(struct net_device *dev, struct sk_buff *s goto invalid_skb; } + if (can_dev_in_xl_only_mode(priv) && !can_is_canxl_skb(skb)) { + netdev_info_once(dev, + "Error signaling is disabled, dropping skb\n"); + goto invalid_skb; + } + return can_dropped_invalid_skb(dev, skb); invalid_skb: From f6ccc2b293ba27e9171c63e456d9cba664fa2337 Mon Sep 17 00:00:00 2001 From: Vincent Mailhol Date: Wed, 26 Nov 2025 11:16:09 +0100 Subject: [PATCH 1151/1381] can: bittiming: add PWM parameters In CAN XL, higher data bit rates require the CAN transceiver to switch its operation mode to use Pulse-Width Modulation (PWM) transmission mode instead of the classic dominant/recessive transmission mode. The PWM parameters are: - PWMS: pulse width modulation short phase - PWML: pulse width modulation long phase - PWMO: pulse width modulation offset CiA 612-2 specifies PWMS and PWML to be at least 1 (arguably, PWML shall be at least 2 to respect the PWMS < PWML rule). PWMO's minimum is expected to always be zero. It is added more for consistency than anything else. Add struct can_pwm_const so that the different devices can provide their minimum and maximum values. When TMS is on, the runtime PWMS, PWML and PWMO are needed (either calculated or provided by the user): add struct can_pwm to store these. TDC and PWM can not be used at the same time (TDC can only be used when TMS is off and PWM only when TMS is on). struct can_pwm is thus put together with struct can_tdc inside a union to save some space. The netlink logic will be added in an upcoming change. Signed-off-by: Vincent Mailhol Signed-off-by: Oliver Hartkopp Link: https://patch.msgid.link/20251126-canxl-v8-8-e7e3eb74f889@pengutronix.de Signed-off-by: Marc Kleine-Budde --- include/linux/can/bittiming.h | 41 +++++++++++++++++++++++++++++++++-- 1 file changed, 39 insertions(+), 2 deletions(-) diff --git a/include/linux/can/bittiming.h b/include/linux/can/bittiming.h index b6cd2476ffd7..967d76689c4f 100644 --- a/include/linux/can/bittiming.h +++ b/include/linux/can/bittiming.h @@ -1,6 +1,6 @@ /* SPDX-License-Identifier: GPL-2.0-only */ /* Copyright (c) 2020 Pengutronix, Marc Kleine-Budde - * Copyright (c) 2021 Vincent Mailhol + * Copyright (c) 2021-2025 Vincent Mailhol */ #ifndef _CAN_BITTIMING_H @@ -120,11 +120,48 @@ struct can_tdc_const { u32 tdcf_max; }; +/* + * struct can_pwm - CAN Pulse-Width Modulation (PWM) parameters + * + * @pwms: pulse width modulation short phase + * @pwml: pulse width modulation long phase + * @pwmo: pulse width modulation offset + */ +struct can_pwm { + u32 pwms; + u32 pwml; + u32 pwmo; +}; + +/* + * struct can_pwm - CAN hardware-dependent constants for Pulse-Width + * Modulation (PWM) + * + * @pwms_min: PWM short phase minimum value. Must be at least 1. + * @pwms_max: PWM short phase maximum value + * @pwml_min: PWM long phase minimum value. Must be at least 1. + * @pwml_max: PWM long phase maximum value + * @pwmo_min: PWM offset phase minimum value + * @pwmo_max: PWM offset phase maximum value + */ +struct can_pwm_const { + u32 pwms_min; + u32 pwms_max; + u32 pwml_min; + u32 pwml_max; + u32 pwmo_min; + u32 pwmo_max; +}; + struct data_bittiming_params { const struct can_bittiming_const *data_bittiming_const; struct can_bittiming data_bittiming; const struct can_tdc_const *tdc_const; - struct can_tdc tdc; + const struct can_pwm_const *pwm_const; + union { + struct can_tdc tdc; + struct can_pwm pwm; + }; const u32 *data_bitrate_const; unsigned int data_bitrate_const_cnt; int (*do_set_data_bittiming)(struct net_device *dev); From 8e2a2885a2a6217190065d1aae98fe88a670cc28 Mon Sep 17 00:00:00 2001 From: Vincent Mailhol Date: Wed, 26 Nov 2025 11:16:10 +0100 Subject: [PATCH 1152/1381] can: bittiming: add PWM validation Add can_validate_pwm() to validate the values pwms, pwml and pwml. Error messages are added to each of the checks to inform the user on what went wrong. Refer to those error messages to understand the validation logic. The boundary values CAN_PWM_DECODE_NS (the transceiver minimum decoding margin) and CAN_PWM_NS_MAX (the maximum PWM symbol duration) are hardcoded for the moment. Note that a transceiver capable of bitrates higher than 20 Mbps may be able to handle a CAN_PWM_DECODE_NS below 5 ns. If such transceivers become commercially available, this code could be revisited to make this parameter configurable. For now, leave it static. Signed-off-by: Vincent Mailhol Signed-off-by: Oliver Hartkopp Link: https://patch.msgid.link/20251126-canxl-v8-9-e7e3eb74f889@pengutronix.de Signed-off-by: Marc Kleine-Budde --- drivers/net/can/dev/bittiming.c | 63 +++++++++++++++++++++++++++++++++ include/linux/can/bittiming.h | 22 ++++++++++++ 2 files changed, 85 insertions(+) diff --git a/drivers/net/can/dev/bittiming.c b/drivers/net/can/dev/bittiming.c index 0b93900b1dfa..8f82418230ce 100644 --- a/drivers/net/can/dev/bittiming.c +++ b/drivers/net/can/dev/bittiming.c @@ -2,6 +2,7 @@ /* Copyright (C) 2005 Marc Kleine-Budde, Pengutronix * Copyright (C) 2006 Andrey Volkov, Varma Electronics * Copyright (C) 2008-2009 Wolfgang Grandegger + * Copyright (c) 2025 Vincent Mailhol */ #include @@ -151,3 +152,65 @@ int can_get_bittiming(const struct net_device *dev, struct can_bittiming *bt, return -EINVAL; } + +int can_validate_pwm_bittiming(const struct net_device *dev, + const struct can_pwm *pwm, + struct netlink_ext_ack *extack) +{ + const struct can_priv *priv = netdev_priv(dev); + u32 xl_bit_time_tqmin = can_bit_time_tqmin(&priv->xl.data_bittiming); + u32 nom_bit_time_tqmin = can_bit_time_tqmin(&priv->bittiming); + u32 pwms_ns = can_tqmin_to_ns(pwm->pwms, priv->clock.freq); + u32 pwml_ns = can_tqmin_to_ns(pwm->pwml, priv->clock.freq); + + if (pwms_ns + pwml_ns > CAN_PWM_NS_MAX) { + NL_SET_ERR_MSG_FMT(extack, + "The PWM symbol duration: %u ns may not exceed %u ns", + pwms_ns + pwml_ns, CAN_PWM_NS_MAX); + return -EINVAL; + } + + if (pwms_ns < CAN_PWM_DECODE_NS) { + NL_SET_ERR_MSG_FMT(extack, + "PWMS: %u ns shall be at least %u ns", + pwms_ns, CAN_PWM_DECODE_NS); + return -EINVAL; + } + + if (pwm->pwms >= pwm->pwml) { + NL_SET_ERR_MSG_FMT(extack, + "PWMS: %u tqmin shall be smaller than PWML: %u tqmin", + pwm->pwms, pwm->pwml); + return -EINVAL; + } + + if (pwml_ns - pwms_ns < 2 * CAN_PWM_DECODE_NS) { + NL_SET_ERR_MSG_FMT(extack, + "At least %u ns shall separate PWMS: %u ns from PMWL: %u ns", + 2 * CAN_PWM_DECODE_NS, pwms_ns, pwml_ns); + return -EINVAL; + } + + if (xl_bit_time_tqmin % (pwm->pwms + pwm->pwml) != 0) { + NL_SET_ERR_MSG_FMT(extack, + "PWM duration: %u tqmin does not divide XL's bit time: %u tqmin", + pwm->pwms + pwm->pwml, xl_bit_time_tqmin); + return -EINVAL; + } + + if (pwm->pwmo >= pwm->pwms + pwm->pwml) { + NL_SET_ERR_MSG_FMT(extack, + "PWMO: %u tqmin can not be greater than PWMS + PWML: %u tqmin", + pwm->pwmo, pwm->pwms + pwm->pwml); + return -EINVAL; + } + + if (nom_bit_time_tqmin % (pwm->pwms + pwm->pwml) != pwm->pwmo) { + NL_SET_ERR_MSG_FMT(extack, + "Can not assemble nominal bit time: %u tqmin out of PWMS + PMWL and PWMO", + nom_bit_time_tqmin); + return -EINVAL; + } + + return 0; +} diff --git a/include/linux/can/bittiming.h b/include/linux/can/bittiming.h index 967d76689c4f..2504fafc72e4 100644 --- a/include/linux/can/bittiming.h +++ b/include/linux/can/bittiming.h @@ -87,6 +87,11 @@ struct can_tdc { u32 tdcf; }; +/* The transceiver decoding margin corresponds to t_Decode in ISO 11898-2 */ +#define CAN_PWM_DECODE_NS 5 +/* Maximum PWM symbol duration. Corresponds to t_SymbolNom_MAX - t_Decode */ +#define CAN_PWM_NS_MAX (205 - CAN_PWM_DECODE_NS) + /* * struct can_tdc_const - CAN hardware-dependent constant for * Transmission Delay Compensation @@ -203,6 +208,10 @@ int can_get_bittiming(const struct net_device *dev, struct can_bittiming *bt, const unsigned int bitrate_const_cnt, struct netlink_ext_ack *extack); +int can_validate_pwm_bittiming(const struct net_device *dev, + const struct can_pwm *pwm, + struct netlink_ext_ack *extack); + /* * can_get_relative_tdco() - TDCO relative to the sample point * @@ -245,4 +254,17 @@ static inline unsigned int can_bit_time(const struct can_bittiming *bt) return CAN_SYNC_SEG + bt->prop_seg + bt->phase_seg1 + bt->phase_seg2; } +/* Duration of one bit in minimum time quantum */ +static inline unsigned int can_bit_time_tqmin(const struct can_bittiming *bt) +{ + return can_bit_time(bt) * bt->brp; +} + +/* Convert a duration from minimum a minimum time quantum to nano seconds */ +static inline u32 can_tqmin_to_ns(u32 tqmin, u32 clock_freq) +{ + return DIV_U64_ROUND_CLOSEST(mul_u32_u32(tqmin, NSEC_PER_SEC), + clock_freq); +} + #endif /* !_CAN_BITTIMING_H */ From 9892339cf0348730e82383d4de9d9387b9d63925 Mon Sep 17 00:00:00 2001 From: Vincent Mailhol Date: Wed, 26 Nov 2025 11:16:11 +0100 Subject: [PATCH 1153/1381] can: calc_bittiming: add PWM calculation Perform the PWM calculation according to CiA recommendations. Note that for databitrates greater than 5 MBPS, tqmin is less than CAN_PWM_NS_MAX (which is defined to 200 nano seconds), consequently, the result of the division: DIV_ROUND_UP(xl_ns, CAN_PWM_NS_MAX) is one and thus the for loop automatically stops on the first iteration giving a single PWM symbol per bit as expected. Because of that, there is no actual need for a separate conditional branch for when the databitrate is greater than 5 MBPS. Signed-off-by: Vincent Mailhol Signed-off-by: Oliver Hartkopp Link: https://patch.msgid.link/20251126-canxl-v8-10-e7e3eb74f889@pengutronix.de Signed-off-by: Marc Kleine-Budde --- drivers/net/can/dev/calc_bittiming.c | 36 ++++++++++++++++++++++++++++ include/linux/can/bittiming.h | 10 ++++++++ 2 files changed, 46 insertions(+) diff --git a/drivers/net/can/dev/calc_bittiming.c b/drivers/net/can/dev/calc_bittiming.c index 394d6974f481..268ec6fa7c49 100644 --- a/drivers/net/can/dev/calc_bittiming.c +++ b/drivers/net/can/dev/calc_bittiming.c @@ -2,6 +2,7 @@ /* Copyright (C) 2005 Marc Kleine-Budde, Pengutronix * Copyright (C) 2006 Andrey Volkov, Varma Electronics * Copyright (C) 2008-2009 Wolfgang Grandegger + * Copyright (C) 2021-2025 Vincent Mailhol */ #include @@ -198,3 +199,38 @@ void can_calc_tdco(struct can_tdc *tdc, const struct can_tdc_const *tdc_const, *ctrlmode |= tdc_auto; } } + +int can_calc_pwm(struct net_device *dev, struct netlink_ext_ack *extack) +{ + struct can_priv *priv = netdev_priv(dev); + const struct can_pwm_const *pwm_const = priv->xl.pwm_const; + struct can_pwm *pwm = &priv->xl.pwm; + u32 xl_tqmin = can_bit_time_tqmin(&priv->xl.data_bittiming); + u32 xl_ns = can_tqmin_to_ns(xl_tqmin, priv->clock.freq); + u32 nom_tqmin = can_bit_time_tqmin(&priv->bittiming); + int pwm_per_bit_max = xl_tqmin / (pwm_const->pwms_min + pwm_const->pwml_min); + int pwm_per_bit; + u32 pwm_tqmin; + + /* For 5 MB/s databitrate or greater, xl_ns < CAN_PWM_NS_MAX + * giving us a pwm_per_bit of 1 and the loop immediately breaks + */ + for (pwm_per_bit = DIV_ROUND_UP(xl_ns, CAN_PWM_NS_MAX); + pwm_per_bit <= pwm_per_bit_max; pwm_per_bit++) + if (xl_tqmin % pwm_per_bit == 0) + break; + + if (pwm_per_bit > pwm_per_bit_max) { + NL_SET_ERR_MSG_FMT(extack, + "Can not divide the XL data phase's bit time: %u tqmin into multiple PWM symbols", + xl_tqmin); + return -EINVAL; + } + + pwm_tqmin = xl_tqmin / pwm_per_bit; + pwm->pwms = DIV_ROUND_UP_POW2(pwm_tqmin, 4); + pwm->pwml = pwm_tqmin - pwm->pwms; + pwm->pwmo = nom_tqmin % pwm_tqmin; + + return 0; +} diff --git a/include/linux/can/bittiming.h b/include/linux/can/bittiming.h index 2504fafc72e4..726d909e87ce 100644 --- a/include/linux/can/bittiming.h +++ b/include/linux/can/bittiming.h @@ -180,6 +180,8 @@ int can_calc_bittiming(const struct net_device *dev, struct can_bittiming *bt, void can_calc_tdco(struct can_tdc *tdc, const struct can_tdc_const *tdc_const, const struct can_bittiming *dbt, u32 tdc_mask, u32 *ctrlmode, u32 ctrlmode_supported); + +int can_calc_pwm(struct net_device *dev, struct netlink_ext_ack *extack); #else /* !CONFIG_CAN_CALC_BITTIMING */ static inline int can_calc_bittiming(const struct net_device *dev, struct can_bittiming *bt, @@ -195,6 +197,14 @@ can_calc_tdco(struct can_tdc *tdc, const struct can_tdc_const *tdc_const, u32 tdc_mask, u32 *ctrlmode, u32 ctrlmode_supported) { } + +static inline int +can_calc_pwm(struct net_device *dev, struct netlink_ext_ack *extack) +{ + NL_SET_ERR_MSG(extack, + "bit-timing calculation not available: manually provide PWML and PWMS\n"); + return -EINVAL; +} #endif /* CONFIG_CAN_CALC_BITTIMING */ void can_sjw_set_default(struct can_bittiming *bt); From 46552323fa6779beb1ea558254dfd56021174c93 Mon Sep 17 00:00:00 2001 From: Vincent Mailhol Date: Wed, 26 Nov 2025 11:16:12 +0100 Subject: [PATCH 1154/1381] can: netlink: add PWM netlink interface When the TMS is switched on, the node uses PWM (Pulse Width Modulation) during the data phase instead of the classic NRZ (Non Return to Zero) encoding. PWM is configured by three parameters: - PWMS: Pulse Width Modulation Short phase - PWML: Pulse Width Modulation Long phase - PWMO: Pulse Width Modulation Offset time For each of these parameters, define three IFLA symbols: - IFLA_CAN_PWM_PWM*_MIN: the minimum allowed value. - IFLA_CAN_PWM_PWM*_MAX: the maximum allowed value. - IFLA_CAN_PWM_PWM*: the runtime value. This results in a total of nine IFLA symbols which are all nested in a parent IFLA_CAN_XL_PWM symbol. IFLA_CAN_PWM_PWM*_MIN and IFLA_CAN_PWM_PWM*_MAX define the range of allowed values and will match the value statically configured by the device in struct can_pwm_const. IFLA_CAN_PWM_PWM* match the runtime values stored in struct can_pwm. Those parameters may only be configured when the tms mode is on. If the PWMS, PWML and PWMO parameters are provided, check that all the needed parameters are present using can_validate_pwm(), then check their value using can_validate_pwm_bittiming(). PWMO defaults to zero if omitted. Otherwise, if CAN_CTRLMODE_XL_TMS is true but none of the PWM parameters are provided, calculate them using can_calc_pwm(). Signed-off-by: Vincent Mailhol Signed-off-by: Oliver Hartkopp Link: https://patch.msgid.link/20251126-canxl-v8-11-e7e3eb74f889@pengutronix.de Signed-off-by: Marc Kleine-Budde --- drivers/net/can/dev/netlink.c | 192 ++++++++++++++++++++++++++++++- include/uapi/linux/can/netlink.h | 25 ++++ 2 files changed, 215 insertions(+), 2 deletions(-) diff --git a/drivers/net/can/dev/netlink.c b/drivers/net/can/dev/netlink.c index b2c24439abba..d6b0e686fb11 100644 --- a/drivers/net/can/dev/netlink.c +++ b/drivers/net/can/dev/netlink.c @@ -25,6 +25,7 @@ static const struct nla_policy can_policy[IFLA_CAN_MAX + 1] = { [IFLA_CAN_XL_DATA_BITTIMING] = { .len = sizeof(struct can_bittiming) }, [IFLA_CAN_XL_DATA_BITTIMING_CONST] = { .len = sizeof(struct can_bittiming_const) }, [IFLA_CAN_XL_TDC] = { .type = NLA_NESTED }, + [IFLA_CAN_XL_PWM] = { .type = NLA_NESTED }, }; static const struct nla_policy can_tdc_policy[IFLA_CAN_TDC_MAX + 1] = { @@ -39,6 +40,18 @@ static const struct nla_policy can_tdc_policy[IFLA_CAN_TDC_MAX + 1] = { [IFLA_CAN_TDC_TDCF] = { .type = NLA_U32 }, }; +static const struct nla_policy can_pwm_policy[IFLA_CAN_PWM_MAX + 1] = { + [IFLA_CAN_PWM_PWMS_MIN] = { .type = NLA_U32 }, + [IFLA_CAN_PWM_PWMS_MAX] = { .type = NLA_U32 }, + [IFLA_CAN_PWM_PWML_MIN] = { .type = NLA_U32 }, + [IFLA_CAN_PWM_PWML_MAX] = { .type = NLA_U32 }, + [IFLA_CAN_PWM_PWMO_MIN] = { .type = NLA_U32 }, + [IFLA_CAN_PWM_PWMO_MAX] = { .type = NLA_U32 }, + [IFLA_CAN_PWM_PWMS] = { .type = NLA_U32 }, + [IFLA_CAN_PWM_PWML] = { .type = NLA_U32 }, + [IFLA_CAN_PWM_PWMO] = { .type = NLA_U32 }, +}; + static int can_validate_bittiming(struct nlattr *data[], struct netlink_ext_ack *extack, int ifla_can_bittiming) @@ -119,6 +132,40 @@ static int can_validate_tdc(struct nlattr *data_tdc, return 0; } +static int can_validate_pwm(struct nlattr *data[], + struct netlink_ext_ack *extack, u32 flags) +{ + struct nlattr *tb_pwm[IFLA_CAN_PWM_MAX + 1]; + int err; + + if (!data[IFLA_CAN_XL_PWM]) + return 0; + + if (!(flags & CAN_CTRLMODE_XL_TMS)) { + NL_SET_ERR_MSG(extack, "PWM requires TMS"); + return -EOPNOTSUPP; + } + + err = nla_parse_nested(tb_pwm, IFLA_CAN_PWM_MAX, data[IFLA_CAN_XL_PWM], + can_pwm_policy, extack); + if (err) + return err; + + if (!tb_pwm[IFLA_CAN_PWM_PWMS] != !tb_pwm[IFLA_CAN_PWM_PWML]) { + NL_SET_ERR_MSG(extack, + "Provide either both PWMS and PWML, or none for automatic calculation"); + return -EOPNOTSUPP; + } + + if (tb_pwm[IFLA_CAN_PWM_PWMO] && + (!tb_pwm[IFLA_CAN_PWM_PWMS] || !tb_pwm[IFLA_CAN_PWM_PWML])) { + NL_SET_ERR_MSG(extack, "PWMO requires both PWMS and PWML"); + return -EOPNOTSUPP; + } + + return 0; +} + static int can_validate_databittiming(struct nlattr *data[], struct netlink_ext_ack *extack, int ifla_can_data_bittiming, u32 flags) @@ -247,6 +294,10 @@ static int can_validate(struct nlattr *tb[], struct nlattr *data[], if (err) return err; + err = can_validate_pwm(data, extack, flags); + if (err) + return err; + return 0; } @@ -322,6 +373,7 @@ static int can_ctrlmode_changelink(struct net_device *dev, sizeof(priv->fd.data_bittiming)); priv->ctrlmode &= ~CAN_CTRLMODE_XL_TDC_MASK; memset(&priv->xl.tdc, 0, sizeof(priv->xl.tdc)); + memset(&priv->xl.pwm, 0, sizeof(priv->xl.pwm)); } can_set_default_mtu(dev); @@ -468,6 +520,76 @@ static int can_dbt_changelink(struct net_device *dev, struct nlattr *data[], return 0; } +static int can_pwm_changelink(struct net_device *dev, + const struct nlattr *pwm_nla, + struct netlink_ext_ack *extack) +{ + struct can_priv *priv = netdev_priv(dev); + const struct can_pwm_const *pwm_const = priv->xl.pwm_const; + struct nlattr *tb_pwm[IFLA_CAN_PWM_MAX + 1]; + struct can_pwm pwm = { 0 }; + int err; + + if (!(priv->ctrlmode & CAN_CTRLMODE_XL_TMS)) + return 0; + + if (!pwm_const) { + NL_SET_ERR_MSG(extack, "The device does not support PWM"); + return -EOPNOTSUPP; + } + + if (!pwm_nla) + return can_calc_pwm(dev, extack); + + err = nla_parse_nested(tb_pwm, IFLA_CAN_PWM_MAX, pwm_nla, + can_pwm_policy, extack); + if (err) + return err; + + if (tb_pwm[IFLA_CAN_PWM_PWMS]) { + pwm.pwms = nla_get_u32(tb_pwm[IFLA_CAN_PWM_PWMS]); + if (pwm.pwms < pwm_const->pwms_min || + pwm.pwms > pwm_const->pwms_max) { + NL_SET_ERR_MSG_FMT(extack, + "PWMS: %u tqmin is out of range: %u...%u", + pwm.pwms, pwm_const->pwms_min, + pwm_const->pwms_max); + return -EINVAL; + } + } + + if (tb_pwm[IFLA_CAN_PWM_PWML]) { + pwm.pwml = nla_get_u32(tb_pwm[IFLA_CAN_PWM_PWML]); + if (pwm.pwml < pwm_const->pwml_min || + pwm.pwml > pwm_const->pwml_max) { + NL_SET_ERR_MSG_FMT(extack, + "PWML: %u tqmin is out of range: %u...%u", + pwm.pwml, pwm_const->pwml_min, + pwm_const->pwml_max); + return -EINVAL; + } + } + + if (tb_pwm[IFLA_CAN_PWM_PWMO]) { + pwm.pwmo = nla_get_u32(tb_pwm[IFLA_CAN_PWM_PWMO]); + if (pwm.pwmo < pwm_const->pwmo_min || + pwm.pwmo > pwm_const->pwmo_max) { + NL_SET_ERR_MSG_FMT(extack, + "PWMO: %u tqmin is out of range: %u...%u", + pwm.pwmo, pwm_const->pwmo_min, + pwm_const->pwmo_max); + return -EINVAL; + } + } + + err = can_validate_pwm_bittiming(dev, &pwm, extack); + if (err) + return err; + + priv->xl.pwm = pwm; + return 0; +} + static int can_changelink(struct net_device *dev, struct nlattr *tb[], struct nlattr *data[], struct netlink_ext_ack *extack) @@ -559,6 +681,9 @@ static int can_changelink(struct net_device *dev, struct nlattr *tb[], /* CAN XL */ err = can_dbt_changelink(dev, data, false, extack); + if (err) + return err; + err = can_pwm_changelink(dev, data[IFLA_CAN_XL_PWM], extack); if (err) return err; @@ -647,6 +772,30 @@ static size_t can_ctrlmode_ext_get_size(void) nla_total_size(sizeof(u32)); /* IFLA_CAN_CTRLMODE_SUPPORTED */ } +static size_t can_pwm_get_size(const struct can_pwm_const *pwm_const, + bool pwm_on) +{ + size_t size; + + if (!pwm_const || !pwm_on) + return 0; + + size = nla_total_size(0); /* nest IFLA_CAN_PWM */ + + size += nla_total_size(sizeof(u32)); /* IFLA_CAN_PWM_PWMS_MIN */ + size += nla_total_size(sizeof(u32)); /* IFLA_CAN_PWM_PWMS_MAX */ + size += nla_total_size(sizeof(u32)); /* IFLA_CAN_PWM_PWML_MIN */ + size += nla_total_size(sizeof(u32)); /* IFLA_CAN_PWM_PWML_MAX */ + size += nla_total_size(sizeof(u32)); /* IFLA_CAN_PWM_PWMO_MIN */ + size += nla_total_size(sizeof(u32)); /* IFLA_CAN_PWM_PWMO_MAX */ + + size += nla_total_size(sizeof(u32)); /* IFLA_CAN_PWM_PWMS */ + size += nla_total_size(sizeof(u32)); /* IFLA_CAN_PWM_PWML */ + size += nla_total_size(sizeof(u32)); /* IFLA_CAN_PWM_PWMO */ + + return size; +} + static size_t can_get_size(const struct net_device *dev) { struct can_priv *priv = netdev_priv(dev); @@ -678,6 +827,8 @@ static size_t can_get_size(const struct net_device *dev) size += can_data_bittiming_get_size(&priv->xl, priv->ctrlmode & CAN_CTRLMODE_XL_TDC_MASK); + size += can_pwm_get_size(priv->xl.pwm_const, /* IFLA_CAN_XL_PWM */ + priv->ctrlmode & CAN_CTRLMODE_XL_TMS); return size; } @@ -776,6 +927,42 @@ static int can_tdc_fill_info(struct sk_buff *skb, const struct net_device *dev, return -EMSGSIZE; } +static int can_pwm_fill_info(struct sk_buff *skb, const struct can_priv *priv) +{ + const struct can_pwm_const *pwm_const = priv->xl.pwm_const; + const struct can_pwm *pwm = &priv->xl.pwm; + struct nlattr *nest; + + if (!pwm_const) + return 0; + + nest = nla_nest_start(skb, IFLA_CAN_XL_PWM); + if (!nest) + return -EMSGSIZE; + + if (nla_put_u32(skb, IFLA_CAN_PWM_PWMS_MIN, pwm_const->pwms_min) || + nla_put_u32(skb, IFLA_CAN_PWM_PWMS_MAX, pwm_const->pwms_max) || + nla_put_u32(skb, IFLA_CAN_PWM_PWML_MIN, pwm_const->pwml_min) || + nla_put_u32(skb, IFLA_CAN_PWM_PWML_MAX, pwm_const->pwml_max) || + nla_put_u32(skb, IFLA_CAN_PWM_PWMO_MIN, pwm_const->pwmo_min) || + nla_put_u32(skb, IFLA_CAN_PWM_PWMO_MAX, pwm_const->pwmo_max)) + goto err_cancel; + + if (priv->ctrlmode & CAN_CTRLMODE_XL_TMS) { + if (nla_put_u32(skb, IFLA_CAN_PWM_PWMS, pwm->pwms) || + nla_put_u32(skb, IFLA_CAN_PWM_PWML, pwm->pwml) || + nla_put_u32(skb, IFLA_CAN_PWM_PWMO, pwm->pwmo)) + goto err_cancel; + } + + nla_nest_end(skb, nest); + return 0; + +err_cancel: + nla_nest_cancel(skb, nest); + return -EMSGSIZE; +} + static int can_ctrlmode_ext_fill_info(struct sk_buff *skb, const struct can_priv *priv) { @@ -859,9 +1046,10 @@ static int can_fill_info(struct sk_buff *skb, const struct net_device *dev) priv->xl.data_bitrate_const, priv->xl.data_bitrate_const_cnt) || - can_tdc_fill_info(skb, dev, IFLA_CAN_XL_TDC) - ) + can_tdc_fill_info(skb, dev, IFLA_CAN_XL_TDC) || + can_pwm_fill_info(skb, priv) + ) return -EMSGSIZE; return 0; diff --git a/include/uapi/linux/can/netlink.h b/include/uapi/linux/can/netlink.h index ebafb091d80f..c30d16746159 100644 --- a/include/uapi/linux/can/netlink.h +++ b/include/uapi/linux/can/netlink.h @@ -5,6 +5,7 @@ * Definitions for the CAN netlink interface * * Copyright (c) 2009 Wolfgang Grandegger + * Copyright (c) 2021-2025 Vincent Mailhol * * This program is free software; you can redistribute it and/or modify * it under the terms of the version 2 of the GNU General Public License @@ -147,6 +148,7 @@ enum { IFLA_CAN_XL_DATA_BITTIMING_CONST, IFLA_CAN_XL_DATA_BITRATE_CONST, IFLA_CAN_XL_TDC, + IFLA_CAN_XL_PWM, /* add new constants above here */ __IFLA_CAN_MAX, @@ -188,6 +190,29 @@ enum { IFLA_CAN_CTRLMODE_MAX = __IFLA_CAN_CTRLMODE - 1 }; +/* + * CAN FD/XL Pulse-Width Modulation (PWM) + * + * Please refer to struct can_pwm_const and can_pwm in + * include/linux/can/bittiming.h for further details. + */ +enum { + IFLA_CAN_PWM_UNSPEC, + IFLA_CAN_PWM_PWMS_MIN, /* u32 */ + IFLA_CAN_PWM_PWMS_MAX, /* u32 */ + IFLA_CAN_PWM_PWML_MIN, /* u32 */ + IFLA_CAN_PWM_PWML_MAX, /* u32 */ + IFLA_CAN_PWM_PWMO_MIN, /* u32 */ + IFLA_CAN_PWM_PWMO_MAX, /* u32 */ + IFLA_CAN_PWM_PWMS, /* u32 */ + IFLA_CAN_PWM_PWML, /* u32 */ + IFLA_CAN_PWM_PWMO, /* u32 */ + + /* add new constants above here */ + __IFLA_CAN_PWM, + IFLA_CAN_PWM_MAX = __IFLA_CAN_PWM - 1 +}; + /* u16 termination range: 1..65535 Ohms */ #define CAN_TERMINATION_DISABLED 0 From 1d147cb7c51d6e994ba740709072adf270d7b878 Mon Sep 17 00:00:00 2001 From: Vincent Mailhol Date: Wed, 26 Nov 2025 11:16:13 +0100 Subject: [PATCH 1155/1381] can: calc_bittiming: replace misleading "nominal" by "reference" The functions can_update_sample_point() and can_calc_bittiming() are generic and meant to be used for both the nominal and the data bittiming calculation. However, those functions use misleading terminologies such as "bitrate nominal" or "sample point nominal". Replace all places where the word "nominal" appears with "reference" in order to better distinguish it from the calculated values. Signed-off-by: Vincent Mailhol Signed-off-by: Oliver Hartkopp Link: https://patch.msgid.link/20251126-canxl-v8-12-e7e3eb74f889@pengutronix.de Signed-off-by: Marc Kleine-Budde --- drivers/net/can/dev/calc_bittiming.c | 26 +++++++++++++------------- 1 file changed, 13 insertions(+), 13 deletions(-) diff --git a/drivers/net/can/dev/calc_bittiming.c b/drivers/net/can/dev/calc_bittiming.c index 268ec6fa7c49..c8c166b383cd 100644 --- a/drivers/net/can/dev/calc_bittiming.c +++ b/drivers/net/can/dev/calc_bittiming.c @@ -24,7 +24,7 @@ */ static int can_update_sample_point(const struct can_bittiming_const *btc, - const unsigned int sample_point_nominal, const unsigned int tseg, + const unsigned int sample_point_reference, const unsigned int tseg, unsigned int *tseg1_ptr, unsigned int *tseg2_ptr, unsigned int *sample_point_error_ptr) { @@ -35,7 +35,7 @@ can_update_sample_point(const struct can_bittiming_const *btc, for (i = 0; i <= 1; i++) { tseg2 = tseg + CAN_SYNC_SEG - - (sample_point_nominal * (tseg + CAN_SYNC_SEG)) / + (sample_point_reference * (tseg + CAN_SYNC_SEG)) / 1000 - i; tseg2 = clamp(tseg2, btc->tseg2_min, btc->tseg2_max); tseg1 = tseg - tseg2; @@ -46,9 +46,9 @@ can_update_sample_point(const struct can_bittiming_const *btc, sample_point = 1000 * (tseg + CAN_SYNC_SEG - tseg2) / (tseg + CAN_SYNC_SEG); - sample_point_error = abs(sample_point_nominal - sample_point); + sample_point_error = abs(sample_point_reference - sample_point); - if (sample_point <= sample_point_nominal && + if (sample_point <= sample_point_reference && sample_point_error < best_sample_point_error) { best_sample_point = sample_point; best_sample_point_error = sample_point_error; @@ -68,11 +68,11 @@ int can_calc_bittiming(const struct net_device *dev, struct can_bittiming *bt, { struct can_priv *priv = netdev_priv(dev); unsigned int bitrate; /* current bitrate */ - unsigned int bitrate_error; /* difference between current and nominal value */ + unsigned int bitrate_error; /* diff between calculated and reference value */ unsigned int best_bitrate_error = UINT_MAX; - unsigned int sample_point_error; /* difference between current and nominal value */ + unsigned int sample_point_error; /* diff between calculated and reference value */ unsigned int best_sample_point_error = UINT_MAX; - unsigned int sample_point_nominal; /* nominal sample point */ + unsigned int sample_point_reference; /* reference sample point */ unsigned int best_tseg = 0; /* current best value for tseg */ unsigned int best_brp = 0; /* current best value for brp */ unsigned int brp, tsegall, tseg, tseg1 = 0, tseg2 = 0; @@ -81,14 +81,14 @@ int can_calc_bittiming(const struct net_device *dev, struct can_bittiming *bt, /* Use CiA recommended sample points */ if (bt->sample_point) { - sample_point_nominal = bt->sample_point; + sample_point_reference = bt->sample_point; } else { if (bt->bitrate > 800 * KILO /* BPS */) - sample_point_nominal = 750; + sample_point_reference = 750; else if (bt->bitrate > 500 * KILO /* BPS */) - sample_point_nominal = 800; + sample_point_reference = 800; else - sample_point_nominal = 875; + sample_point_reference = 875; } /* tseg even = round down, odd = round up */ @@ -115,7 +115,7 @@ int can_calc_bittiming(const struct net_device *dev, struct can_bittiming *bt, if (bitrate_error < best_bitrate_error) best_sample_point_error = UINT_MAX; - can_update_sample_point(btc, sample_point_nominal, tseg / 2, + can_update_sample_point(btc, sample_point_reference, tseg / 2, &tseg1, &tseg2, &sample_point_error); if (sample_point_error >= best_sample_point_error) continue; @@ -146,7 +146,7 @@ int can_calc_bittiming(const struct net_device *dev, struct can_bittiming *bt, } /* real sample point */ - bt->sample_point = can_update_sample_point(btc, sample_point_nominal, + bt->sample_point = can_update_sample_point(btc, sample_point_reference, best_tseg, &tseg1, &tseg2, NULL); From a6ddf91a4f9718cec712785904c57b7117f61d6c Mon Sep 17 00:00:00 2001 From: Vincent Mailhol Date: Wed, 26 Nov 2025 11:16:14 +0100 Subject: [PATCH 1156/1381] can: calc_bittiming: add can_calc_sample_point_nrz() CAN XL optimal sample point for PWM encoding (when TMS is on) differs from the NRZ optimal one. There is thus a need to calculate a different sample point depending whether TMS is on or off. This is a preparation change: move the sample point calculation from can_calc_bittiming() into the new can_calc_sample_point_nrz() function. In an upcoming change, a function will be added to calculate the sample point for PWM encoding. Signed-off-by: Vincent Mailhol Signed-off-by: Oliver Hartkopp Link: https://patch.msgid.link/20251126-canxl-v8-13-e7e3eb74f889@pengutronix.de Signed-off-by: Marc Kleine-Budde --- drivers/net/can/dev/calc_bittiming.c | 25 +++++++++++++++---------- 1 file changed, 15 insertions(+), 10 deletions(-) diff --git a/drivers/net/can/dev/calc_bittiming.c b/drivers/net/can/dev/calc_bittiming.c index c8c166b383cd..bacdf3b218d3 100644 --- a/drivers/net/can/dev/calc_bittiming.c +++ b/drivers/net/can/dev/calc_bittiming.c @@ -10,6 +10,18 @@ #define CAN_CALC_MAX_ERROR 50 /* in one-tenth of a percent */ +/* CiA recommended sample points for Non Return to Zero encoding. */ +static int can_calc_sample_point_nrz(const struct can_bittiming *bt) +{ + if (bt->bitrate > 800 * KILO /* BPS */) + return 750; + + if (bt->bitrate > 500 * KILO /* BPS */) + return 800; + + return 875; +} + /* Bit-timing calculation derived from: * * Code based on LinCAN sources and H8S2638 project @@ -79,17 +91,10 @@ int can_calc_bittiming(const struct net_device *dev, struct can_bittiming *bt, u64 v64; int err; - /* Use CiA recommended sample points */ - if (bt->sample_point) { + if (bt->sample_point) sample_point_reference = bt->sample_point; - } else { - if (bt->bitrate > 800 * KILO /* BPS */) - sample_point_reference = 750; - else if (bt->bitrate > 500 * KILO /* BPS */) - sample_point_reference = 800; - else - sample_point_reference = 875; - } + else + sample_point_reference = can_calc_sample_point_nrz(bt); /* tseg even = round down, odd = round up */ for (tseg = (btc->tseg1_max + btc->tseg2_max) * 2 + 1; From f5de373ae455f1db6cf15033d660ac0046bcb0ff Mon Sep 17 00:00:00 2001 From: Vincent Mailhol Date: Wed, 26 Nov 2025 11:16:15 +0100 Subject: [PATCH 1157/1381] can: calc_bittiming: add can_calc_sample_point_pwm() The optimum sample point value depends on the bit symmetry. The more asymmetric the bit is, the more the sample point would be located towards the end of the bit. On the contrary, if the transceiver only has a small asymmetry, the optimal sample point would be slightly after the centre of the bit. For NRZ encoding (used by Classical CAN, CAN FD and CAN XL with TMS off), the optimum sample points values are above 70% as implemented in can_calc_sample_point_nrz(). When TMS is on, CAN XL optimum sample points are near to 50% or 60% [1]. Add can_calc_sample_point_pwm() which returns a sample point which is suitable for PWM encoding. We crafted the formula to make it return the same values as below table (source: table 3 of [1]). Bit rate (Mbits/s) Sample point ------------------------------------- 2.0 51.3% 5.0 53.1% 8.0 55.0% 10.0 56.3% 12.3 53.8% 13.3 58.3% 14.5 54.5% 16.0 60.0% 17.7 55.6% 20.0 62.5% The calculation simply consists of setting a slightly too high sample point and then letting can_update_sample_point() correct the values. For now, it is just a formula up our sleeves which matches the empirical observations of [1]. Once CiA recommendations become available, can_calc_sample_point_pwm() should be updated accordingly. [1] CAN XL system design: Clock tolerances and edge deviations edge deviations Link: https://www.can-cia.org/fileadmin/cia/documents/publications/cnlm/december_2024/cnlm_24-4_p18_can_xl_system_design_clock_tolerances_and_edge_deviations_dr_arthur_mutter_bosch.pdf Signed-off-by: Vincent Mailhol Signed-off-by: Oliver Hartkopp Link: https://patch.msgid.link/20251126-canxl-v8-14-e7e3eb74f889@pengutronix.de Signed-off-by: Marc Kleine-Budde --- drivers/net/can/dev/calc_bittiming.c | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+) diff --git a/drivers/net/can/dev/calc_bittiming.c b/drivers/net/can/dev/calc_bittiming.c index bacdf3b218d3..60a505ce69de 100644 --- a/drivers/net/can/dev/calc_bittiming.c +++ b/drivers/net/can/dev/calc_bittiming.c @@ -22,6 +22,21 @@ static int can_calc_sample_point_nrz(const struct can_bittiming *bt) return 875; } +/* Sample points for Pulse-Width Modulation encoding. */ +static int can_calc_sample_point_pwm(const struct can_bittiming *bt) +{ + if (bt->bitrate > 15 * MEGA /* BPS */) + return 625; + + if (bt->bitrate > 9 * MEGA /* BPS */) + return 600; + + if (bt->bitrate > 4 * MEGA /* BPS */) + return 560; + + return 520; +} + /* Bit-timing calculation derived from: * * Code based on LinCAN sources and H8S2638 project @@ -93,6 +108,9 @@ int can_calc_bittiming(const struct net_device *dev, struct can_bittiming *bt, if (bt->sample_point) sample_point_reference = bt->sample_point; + else if (btc == priv->xl.data_bittiming_const && + (priv->ctrlmode & CAN_CTRLMODE_XL_TMS)) + sample_point_reference = can_calc_sample_point_pwm(bt); else sample_point_reference = can_calc_sample_point_nrz(bt); From 816cf430e84b4628dba665491e78ce081a468fcb Mon Sep 17 00:00:00 2001 From: Vincent Mailhol Date: Wed, 26 Nov 2025 11:16:16 +0100 Subject: [PATCH 1158/1381] can: add dummy_can driver During the development of CAN XL, we found the need of creating a dummy CAN XL driver in order to test the new netlink interface. While this code was initially intended to be some throwaway, it received some positive feedback. Add the dummy_can driver. This driver acts similarly to the vcan interface in the sense that it will echo back any packet it receives. The difference is that it exposes a set on bittiming parameters as a real device would and thus must be configured as if it was a real physical interface. The driver comes with a debug mode. If debug message are enabled (for example by enabling CONFIG_CAN_DEBUG_DEVICES), it will print in the kernel log all the bittiming values, similar to what a: ip --details link show can0 would do. This driver is mostly intended for debugging and testing, but some developers also may want to look at it as a simple reference implementation. Signed-off-by: Vincent Mailhol Signed-off-by: Oliver Hartkopp Link: https://patch.msgid.link/20251126-canxl-v8-15-e7e3eb74f889@pengutronix.de Signed-off-by: Marc Kleine-Budde --- drivers/net/can/Kconfig | 17 +++ drivers/net/can/Makefile | 1 + drivers/net/can/dummy_can.c | 285 ++++++++++++++++++++++++++++++++++++ 3 files changed, 303 insertions(+) create mode 100644 drivers/net/can/dummy_can.c diff --git a/drivers/net/can/Kconfig b/drivers/net/can/Kconfig index d43d56694667..e15e320db476 100644 --- a/drivers/net/can/Kconfig +++ b/drivers/net/can/Kconfig @@ -124,6 +124,23 @@ config CAN_CAN327 If this driver is built as a module, it will be called can327. +config CAN_DUMMY + tristate "Dummy CAN" + help + A dummy CAN module supporting Classical CAN, CAN FD and CAN XL. It + exposes bittiming values which can be configured through the netlink + interface. + + The module will simply echo any frame sent to it. If debug messages + are activated, it prints all the CAN bittiming information in the + kernel log. Aside from that it does nothing. + + This is convenient for testing the CAN netlink interface. Most of the + users will never need this. If unsure, say NO. + + To compile this driver as a module, choose M here: the module will be + called dummy-can. + config CAN_FLEXCAN tristate "Support for Freescale FLEXCAN based chips" depends on OF || COLDFIRE || COMPILE_TEST diff --git a/drivers/net/can/Makefile b/drivers/net/can/Makefile index 56138d8ddfd2..d7bc10a6b8ea 100644 --- a/drivers/net/can/Makefile +++ b/drivers/net/can/Makefile @@ -21,6 +21,7 @@ obj-$(CONFIG_CAN_CAN327) += can327.o obj-$(CONFIG_CAN_CC770) += cc770/ obj-$(CONFIG_CAN_C_CAN) += c_can/ obj-$(CONFIG_CAN_CTUCANFD) += ctucanfd/ +obj-$(CONFIG_CAN_DUMMY) += dummy_can.o obj-$(CONFIG_CAN_FLEXCAN) += flexcan/ obj-$(CONFIG_CAN_GRCAN) += grcan.o obj-$(CONFIG_CAN_IFI_CANFD) += ifi_canfd/ diff --git a/drivers/net/can/dummy_can.c b/drivers/net/can/dummy_can.c new file mode 100644 index 000000000000..41953655e3d3 --- /dev/null +++ b/drivers/net/can/dummy_can.c @@ -0,0 +1,285 @@ +// SPDX-License-Identifier: GPL-2.0 + +/* Copyright (c) 2025 Vincent Mailhol */ + +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include + +struct dummy_can { + struct can_priv can; + struct net_device *dev; +}; + +static struct dummy_can *dummy_can; + +static const struct can_bittiming_const dummy_can_bittiming_const = { + .name = "dummy_can CC", + .tseg1_min = 2, + .tseg1_max = 256, + .tseg2_min = 2, + .tseg2_max = 128, + .sjw_max = 128, + .brp_min = 1, + .brp_max = 512, + .brp_inc = 1 +}; + +static const struct can_bittiming_const dummy_can_fd_databittiming_const = { + .name = "dummy_can FD", + .tseg1_min = 2, + .tseg1_max = 256, + .tseg2_min = 2, + .tseg2_max = 128, + .sjw_max = 128, + .brp_min = 1, + .brp_max = 512, + .brp_inc = 1 +}; + +static const struct can_tdc_const dummy_can_fd_tdc_const = { + .tdcv_min = 0, + .tdcv_max = 0, /* Manual mode not supported. */ + .tdco_min = 0, + .tdco_max = 127, + .tdcf_min = 0, + .tdcf_max = 127 +}; + +static const struct can_bittiming_const dummy_can_xl_databittiming_const = { + .name = "dummy_can XL", + .tseg1_min = 2, + .tseg1_max = 256, + .tseg2_min = 2, + .tseg2_max = 128, + .sjw_max = 128, + .brp_min = 1, + .brp_max = 512, + .brp_inc = 1 +}; + +static const struct can_tdc_const dummy_can_xl_tdc_const = { + .tdcv_min = 0, + .tdcv_max = 0, /* Manual mode not supported. */ + .tdco_min = 0, + .tdco_max = 127, + .tdcf_min = 0, + .tdcf_max = 127 +}; + +static const struct can_pwm_const dummy_can_pwm_const = { + .pwms_min = 1, + .pwms_max = 8, + .pwml_min = 2, + .pwml_max = 24, + .pwmo_min = 0, + .pwmo_max = 16, +}; + +static void dummy_can_print_bittiming(struct net_device *dev, + struct can_bittiming *bt) +{ + netdev_dbg(dev, "\tbitrate: %u\n", bt->bitrate); + netdev_dbg(dev, "\tsample_point: %u\n", bt->sample_point); + netdev_dbg(dev, "\ttq: %u\n", bt->tq); + netdev_dbg(dev, "\tprop_seg: %u\n", bt->prop_seg); + netdev_dbg(dev, "\tphase_seg1: %u\n", bt->phase_seg1); + netdev_dbg(dev, "\tphase_seg2: %u\n", bt->phase_seg2); + netdev_dbg(dev, "\tsjw: %u\n", bt->sjw); + netdev_dbg(dev, "\tbrp: %u\n", bt->brp); +} + +static void dummy_can_print_tdc(struct net_device *dev, struct can_tdc *tdc) +{ + netdev_dbg(dev, "\t\ttdcv: %u\n", tdc->tdcv); + netdev_dbg(dev, "\t\ttdco: %u\n", tdc->tdco); + netdev_dbg(dev, "\t\ttdcf: %u\n", tdc->tdcf); +} + +static void dummy_can_print_pwm(struct net_device *dev, struct can_pwm *pwm, + struct can_bittiming *dbt) +{ + netdev_dbg(dev, "\t\tpwms: %u\n", pwm->pwms); + netdev_dbg(dev, "\t\tpwml: %u\n", pwm->pwml); + netdev_dbg(dev, "\t\tpwmo: %u\n", pwm->pwmo); +} + +static void dummy_can_print_ctrlmode(struct net_device *dev) +{ + struct dummy_can *priv = netdev_priv(dev); + struct can_priv *can_priv = &priv->can; + unsigned long supported = can_priv->ctrlmode_supported; + u32 enabled = can_priv->ctrlmode; + + netdev_dbg(dev, "Control modes:\n"); + netdev_dbg(dev, "\tsupported: 0x%08x\n", (u32)supported); + netdev_dbg(dev, "\tenabled: 0x%08x\n", enabled); + + if (supported) { + int idx; + + netdev_dbg(dev, "\tlist:"); + for_each_set_bit(idx, &supported, BITS_PER_TYPE(u32)) + netdev_dbg(dev, "\t\t%s: %s\n", + can_get_ctrlmode_str(BIT(idx)), + enabled & BIT(idx) ? "on" : "off"); + } +} + +static void dummy_can_print_bittiming_info(struct net_device *dev) +{ + struct dummy_can *priv = netdev_priv(dev); + struct can_priv *can_priv = &priv->can; + + netdev_dbg(dev, "Clock frequency: %u\n", can_priv->clock.freq); + netdev_dbg(dev, "Maximum bitrate: %u\n", can_priv->bitrate_max); + netdev_dbg(dev, "MTU: %u\n", dev->mtu); + netdev_dbg(dev, "\n"); + + dummy_can_print_ctrlmode(dev); + netdev_dbg(dev, "\n"); + + netdev_dbg(dev, "Classical CAN nominal bittiming:\n"); + dummy_can_print_bittiming(dev, &can_priv->bittiming); + netdev_dbg(dev, "\n"); + + if (can_priv->ctrlmode & CAN_CTRLMODE_FD) { + netdev_dbg(dev, "CAN FD databittiming:\n"); + dummy_can_print_bittiming(dev, &can_priv->fd.data_bittiming); + if (can_fd_tdc_is_enabled(can_priv)) { + netdev_dbg(dev, "\tCAN FD TDC:\n"); + dummy_can_print_tdc(dev, &can_priv->fd.tdc); + } + } + netdev_dbg(dev, "\n"); + + if (can_priv->ctrlmode & CAN_CTRLMODE_XL) { + netdev_dbg(dev, "CAN XL databittiming:\n"); + dummy_can_print_bittiming(dev, &can_priv->xl.data_bittiming); + if (can_xl_tdc_is_enabled(can_priv)) { + netdev_dbg(dev, "\tCAN XL TDC:\n"); + dummy_can_print_tdc(dev, &can_priv->xl.tdc); + } + if (can_priv->ctrlmode & CAN_CTRLMODE_XL_TMS) { + netdev_dbg(dev, "\tCAN XL PWM:\n"); + dummy_can_print_pwm(dev, &can_priv->xl.pwm, + &can_priv->xl.data_bittiming); + } + } + netdev_dbg(dev, "\n"); +} + +static int dummy_can_netdev_open(struct net_device *dev) +{ + int ret; + struct can_priv *priv = netdev_priv(dev); + + dummy_can_print_bittiming_info(dev); + netdev_dbg(dev, "error-signalling is %s\n", + str_enabled_disabled(!can_dev_in_xl_only_mode(priv))); + + ret = open_candev(dev); + if (ret) + return ret; + netif_start_queue(dev); + netdev_dbg(dev, "dummy-can is up\n"); + + return 0; +} + +static int dummy_can_netdev_close(struct net_device *dev) +{ + netif_stop_queue(dev); + close_candev(dev); + netdev_dbg(dev, "dummy-can is down\n"); + + return 0; +} + +static netdev_tx_t dummy_can_start_xmit(struct sk_buff *skb, + struct net_device *dev) +{ + if (can_dev_dropped_skb(dev, skb)) + return NETDEV_TX_OK; + + can_put_echo_skb(skb, dev, 0, 0); + dev->stats.tx_packets++; + dev->stats.tx_bytes += can_get_echo_skb(dev, 0, NULL); + + return NETDEV_TX_OK; +} + +static const struct net_device_ops dummy_can_netdev_ops = { + .ndo_open = dummy_can_netdev_open, + .ndo_stop = dummy_can_netdev_close, + .ndo_start_xmit = dummy_can_start_xmit, +}; + +static const struct ethtool_ops dummy_can_ethtool_ops = { + .get_ts_info = ethtool_op_get_ts_info, +}; + +static int __init dummy_can_init(void) +{ + struct net_device *dev; + struct dummy_can *priv; + int ret; + + dev = alloc_candev(sizeof(*priv), 1); + if (!dev) + return -ENOMEM; + + dev->netdev_ops = &dummy_can_netdev_ops; + dev->ethtool_ops = &dummy_can_ethtool_ops; + priv = netdev_priv(dev); + priv->can.bittiming_const = &dummy_can_bittiming_const; + priv->can.bitrate_max = 20 * MEGA /* BPS */; + priv->can.clock.freq = 160 * MEGA /* Hz */; + priv->can.fd.data_bittiming_const = &dummy_can_fd_databittiming_const; + priv->can.fd.tdc_const = &dummy_can_fd_tdc_const; + priv->can.xl.data_bittiming_const = &dummy_can_xl_databittiming_const; + priv->can.xl.tdc_const = &dummy_can_xl_tdc_const; + priv->can.xl.pwm_const = &dummy_can_pwm_const; + priv->can.ctrlmode_supported = CAN_CTRLMODE_LISTENONLY | + CAN_CTRLMODE_FD | CAN_CTRLMODE_TDC_AUTO | + CAN_CTRLMODE_RESTRICTED | CAN_CTRLMODE_XL | + CAN_CTRLMODE_XL_TDC_AUTO | CAN_CTRLMODE_XL_TMS; + priv->dev = dev; + + ret = register_candev(priv->dev); + if (ret) { + free_candev(priv->dev); + return ret; + } + + dummy_can = priv; + netdev_dbg(dev, "dummy-can ready\n"); + + return 0; +} + +static void __exit dummy_can_exit(void) +{ + struct net_device *dev = dummy_can->dev; + + netdev_dbg(dev, "dummy-can bye bye\n"); + unregister_candev(dev); + free_candev(dev); +} + +module_init(dummy_can_init); +module_exit(dummy_can_exit); + +MODULE_DESCRIPTION("A dummy CAN driver, mainly to test the netlink interface"); +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Vincent Mailhol "); From 1a620a723853a0f49703c317d52dc6b9602cbaa8 Mon Sep 17 00:00:00 2001 From: Oliver Hartkopp Date: Wed, 26 Nov 2025 11:16:17 +0100 Subject: [PATCH 1159/1381] can: raw: instantly reject unsupported CAN frames For real CAN interfaces the CAN_CTRLMODE_FD and CAN_CTRLMODE_XL control modes indicate whether an interface can handle those CAN FD/XL frames. In the case a CAN XL interface is configured in CANXL-only mode with disabled error-signalling neither CAN CC nor CAN FD frames can be sent. The checks are performed on CAN_RAW sockets to give an instant feedback to the user when writing unsupported CAN frames to the interface. Signed-off-by: Oliver Hartkopp Link: https://patch.msgid.link/20251126-canxl-v8-16-e7e3eb74f889@pengutronix.de Signed-off-by: Marc Kleine-Budde --- net/can/raw.c | 54 +++++++++++++++++++++++++++++++++++++++++++-------- 1 file changed, 46 insertions(+), 8 deletions(-) diff --git a/net/can/raw.c b/net/can/raw.c index f36a83d3447c..be1ef7cf4204 100644 --- a/net/can/raw.c +++ b/net/can/raw.c @@ -892,20 +892,58 @@ static void raw_put_canxl_vcid(struct raw_sock *ro, struct sk_buff *skb) } } -static unsigned int raw_check_txframe(struct raw_sock *ro, struct sk_buff *skb, int mtu) +static inline bool raw_dev_cc_enabled(struct net_device *dev, + struct can_priv *priv) { - /* Classical CAN -> no checks for flags and device capabilities */ - if (can_is_can_skb(skb)) + /* The CANXL-only mode disables error-signalling on the CAN bus + * which is needed to send CAN CC/FD frames + */ + if (priv) + return !can_dev_in_xl_only_mode(priv); + + /* virtual CAN interfaces always support CAN CC */ + return true; +} + +static inline bool raw_dev_fd_enabled(struct net_device *dev, + struct can_priv *priv) +{ + /* check FD ctrlmode on real CAN interfaces */ + if (priv) + return (priv->ctrlmode & CAN_CTRLMODE_FD); + + /* check MTU for virtual CAN FD interfaces */ + return (READ_ONCE(dev->mtu) >= CANFD_MTU); +} + +static inline bool raw_dev_xl_enabled(struct net_device *dev, + struct can_priv *priv) +{ + /* check XL ctrlmode on real CAN interfaces */ + if (priv) + return (priv->ctrlmode & CAN_CTRLMODE_XL); + + /* check MTU for virtual CAN XL interfaces */ + return can_is_canxl_dev_mtu(READ_ONCE(dev->mtu)); +} + +static unsigned int raw_check_txframe(struct raw_sock *ro, struct sk_buff *skb, + struct net_device *dev) +{ + struct can_priv *priv = safe_candev_priv(dev); + + /* Classical CAN */ + if (can_is_can_skb(skb) && raw_dev_cc_enabled(dev, priv)) return CAN_MTU; - /* CAN FD -> needs to be enabled and a CAN FD or CAN XL device */ + /* CAN FD */ if (ro->fd_frames && can_is_canfd_skb(skb) && - (mtu == CANFD_MTU || can_is_canxl_dev_mtu(mtu))) + raw_dev_fd_enabled(dev, priv)) return CANFD_MTU; - /* CAN XL -> needs to be enabled and a CAN XL device */ + /* CAN XL */ if (ro->xl_frames && can_is_canxl_skb(skb) && - can_is_canxl_dev_mtu(mtu)) + raw_dev_xl_enabled(dev, priv)) return CANXL_MTU; return 0; @@ -961,7 +999,7 @@ static int raw_sendmsg(struct socket *sock, struct msghdr *msg, size_t size) err = -EINVAL; /* check for valid CAN (CC/FD/XL) frame content */ - txmtu = raw_check_txframe(ro, skb, READ_ONCE(dev->mtu)); + txmtu = raw_check_txframe(ro, skb, dev); if (!txmtu) goto free_skb; From b360a13d44db148f6d72fe4f73356726c0663f6a Mon Sep 17 00:00:00 2001 From: Oliver Hartkopp Date: Wed, 26 Nov 2025 11:16:18 +0100 Subject: [PATCH 1160/1381] can: dev: print bitrate error with two decimal digits Increase the resolution when printing the bitrate error and round-up the value to 0.01% in the case the resolution would still provide values which would lead to 0.00%. Suggested-by: Vincent Mailhol Signed-off-by: Oliver Hartkopp Link: https://patch.msgid.link/20251126-canxl-v8-17-e7e3eb74f889@pengutronix.de Signed-off-by: Marc Kleine-Budde --- drivers/net/can/dev/calc_bittiming.c | 15 +++++++++------ 1 file changed, 9 insertions(+), 6 deletions(-) diff --git a/drivers/net/can/dev/calc_bittiming.c b/drivers/net/can/dev/calc_bittiming.c index 60a505ce69de..cc4022241553 100644 --- a/drivers/net/can/dev/calc_bittiming.c +++ b/drivers/net/can/dev/calc_bittiming.c @@ -153,19 +153,22 @@ int can_calc_bittiming(const struct net_device *dev, struct can_bittiming *bt, } if (best_bitrate_error) { - /* Error in one-tenth of a percent */ - v64 = (u64)best_bitrate_error * 1000; + /* Error in one-hundredth of a percent */ + v64 = (u64)best_bitrate_error * 10000; do_div(v64, bt->bitrate); bitrate_error = (u32)v64; + /* print at least 0.01% if the error is smaller */ + bitrate_error = max(bitrate_error, 1U); if (bitrate_error > CAN_CALC_MAX_ERROR) { NL_SET_ERR_MSG_FMT(extack, - "bitrate error: %u.%u%% too high", - bitrate_error / 10, bitrate_error % 10); + "bitrate error: %u.%02u%% too high", + bitrate_error / 100, + bitrate_error % 100); return -EINVAL; } NL_SET_ERR_MSG_FMT(extack, - "bitrate error: %u.%u%%", - bitrate_error / 10, bitrate_error % 10); + "bitrate error: %u.%02u%%", + bitrate_error / 100, bitrate_error % 100); } /* real sample point */ From 41c13eaf39932fc79aa1ac245a9b97090fe23d5e Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Mon, 24 Nov 2025 10:28:26 +0000 Subject: [PATCH 1161/1381] can: rcar_canfd: Invert reset assert order The two resets are asserted during cleanup in the same order as they were deasserted during probe. Invert the order to restore symmetry. Signed-off-by: Geert Uytterhoeven Reviewed-by: Vincent Mailhol Reviewed-by: Biju Das Signed-off-by: Biju Das Link: https://patch.msgid.link/20251124102837.106973-2-biju.das.jz@bp.renesas.com Signed-off-by: Marc Kleine-Budde --- drivers/net/can/rcar/rcar_canfd.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/can/rcar/rcar_canfd.c b/drivers/net/can/rcar/rcar_canfd.c index 49ab65274b51..cbcc971cda9a 100644 --- a/drivers/net/can/rcar/rcar_canfd.c +++ b/drivers/net/can/rcar/rcar_canfd.c @@ -2192,8 +2192,8 @@ static int rcar_canfd_probe(struct platform_device *pdev) fail_clk: clk_disable_unprepare(gpriv->clkp); fail_reset: - reset_control_assert(gpriv->rstc1); reset_control_assert(gpriv->rstc2); + reset_control_assert(gpriv->rstc1); fail_dev: return err; } @@ -2214,8 +2214,8 @@ static void rcar_canfd_remove(struct platform_device *pdev) /* Enter global sleep mode */ rcar_canfd_set_bit(gpriv->base, RCANFD_GCTR, RCANFD_GCTR_GSLPR); clk_disable_unprepare(gpriv->clkp); - reset_control_assert(gpriv->rstc1); reset_control_assert(gpriv->rstc2); + reset_control_assert(gpriv->rstc1); } static int __maybe_unused rcar_canfd_suspend(struct device *dev) From 790ec4c453890f1221ea595674a1206bbee41dc4 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Mon, 24 Nov 2025 10:28:27 +0000 Subject: [PATCH 1162/1381] can: rcar_canfd: Invert global vs. channel teardown Global state is initialized and torn down before per-channel state. Invert the order to restore symmetry. Signed-off-by: Geert Uytterhoeven Reviewed-by: Vincent Mailhol Reviewed-by: Biju Das Signed-off-by: Biju Das Link: https://patch.msgid.link/20251124102837.106973-3-biju.das.jz@bp.renesas.com Signed-off-by: Marc Kleine-Budde --- drivers/net/can/rcar/rcar_canfd.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/net/can/rcar/rcar_canfd.c b/drivers/net/can/rcar/rcar_canfd.c index cbcc971cda9a..15358eefa8fa 100644 --- a/drivers/net/can/rcar/rcar_canfd.c +++ b/drivers/net/can/rcar/rcar_canfd.c @@ -2203,14 +2203,14 @@ static void rcar_canfd_remove(struct platform_device *pdev) struct rcar_canfd_global *gpriv = platform_get_drvdata(pdev); u32 ch; - rcar_canfd_reset_controller(gpriv); - rcar_canfd_disable_global_interrupts(gpriv); - for_each_set_bit(ch, &gpriv->channels_mask, gpriv->info->max_channels) { rcar_canfd_disable_channel_interrupts(gpriv->ch[ch]); rcar_canfd_channel_remove(gpriv, ch); } + rcar_canfd_disable_global_interrupts(gpriv); + rcar_canfd_reset_controller(gpriv); + /* Enter global sleep mode */ rcar_canfd_set_bit(gpriv->base, RCANFD_GCTR, RCANFD_GCTR_GSLPR); clk_disable_unprepare(gpriv->clkp); From eda3d6c8d784835cec86f42b3f8118c9eb0cc58c Mon Sep 17 00:00:00 2001 From: Biju Das Date: Mon, 24 Nov 2025 10:28:28 +0000 Subject: [PATCH 1163/1381] can: rcar_canfd: Use devm_clk_get_optional() for RAM clk Replace devm_clk_get_optional_enabled()->devm_clk_get_optional() as the RAM clk needs to be enabled in resume for proper operation in STR mode for RZ/G3E SoC. Signed-off-by: Biju Das Reviewed-by: Geert Uytterhoeven Link: https://patch.msgid.link/20251124102837.106973-4-biju.das.jz@bp.renesas.com Signed-off-by: Marc Kleine-Budde --- drivers/net/can/rcar/rcar_canfd.c | 23 +++++++++++++++++------ 1 file changed, 17 insertions(+), 6 deletions(-) diff --git a/drivers/net/can/rcar/rcar_canfd.c b/drivers/net/can/rcar/rcar_canfd.c index 15358eefa8fa..3f71adce6f56 100644 --- a/drivers/net/can/rcar/rcar_canfd.c +++ b/drivers/net/can/rcar/rcar_canfd.c @@ -468,6 +468,7 @@ struct rcar_canfd_global { struct platform_device *pdev; /* Respective platform device */ struct clk *clkp; /* Peripheral clock */ struct clk *can_clk; /* fCAN clock */ + struct clk *clk_ram; /* Clock RAM */ unsigned long channels_mask; /* Enabled channels mask */ bool extclk; /* CANFD or Ext clock */ bool fdmode; /* CAN FD or Classical CAN only mode */ @@ -1975,7 +1976,6 @@ static int rcar_canfd_probe(struct platform_device *pdev) u32 rule_entry = 0; bool fdmode = true; /* CAN FD only mode - default */ char name[9] = "channelX"; - struct clk *clk_ram; int i; info = of_device_get_match_data(dev); @@ -2065,10 +2065,10 @@ static int rcar_canfd_probe(struct platform_device *pdev) gpriv->extclk = gpriv->info->external_clk; } - clk_ram = devm_clk_get_optional_enabled(dev, "ram_clk"); - if (IS_ERR(clk_ram)) - return dev_err_probe(dev, PTR_ERR(clk_ram), - "cannot get enabled ram clock\n"); + gpriv->clk_ram = devm_clk_get_optional(dev, "ram_clk"); + if (IS_ERR(gpriv->clk_ram)) + return dev_err_probe(dev, PTR_ERR(gpriv->clk_ram), + "cannot get ram clock\n"); addr = devm_platform_ioremap_resource(pdev, 0); if (IS_ERR(addr)) { @@ -2134,10 +2134,18 @@ static int rcar_canfd_probe(struct platform_device *pdev) goto fail_reset; } + /* Enable RAM clock */ + err = clk_prepare_enable(gpriv->clk_ram); + if (err) { + dev_err(dev, "failed to enable RAM clock: %pe\n", + ERR_PTR(err)); + goto fail_clk; + } + err = rcar_canfd_reset_controller(gpriv); if (err) { dev_err(dev, "reset controller failed: %pe\n", ERR_PTR(err)); - goto fail_clk; + goto fail_ram_clk; } /* Controller in Global reset & Channel reset mode */ @@ -2189,6 +2197,8 @@ static int rcar_canfd_probe(struct platform_device *pdev) rcar_canfd_channel_remove(gpriv, ch); fail_mode: rcar_canfd_disable_global_interrupts(gpriv); +fail_ram_clk: + clk_disable_unprepare(gpriv->clk_ram); fail_clk: clk_disable_unprepare(gpriv->clkp); fail_reset: @@ -2213,6 +2223,7 @@ static void rcar_canfd_remove(struct platform_device *pdev) /* Enter global sleep mode */ rcar_canfd_set_bit(gpriv->base, RCANFD_GCTR, RCANFD_GCTR_GSLPR); + clk_disable_unprepare(gpriv->clk_ram); clk_disable_unprepare(gpriv->clkp); reset_control_assert(gpriv->rstc2); reset_control_assert(gpriv->rstc1); From fa5f4ec8fff8bc587a2cbf7101303306e045c11f Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Mon, 24 Nov 2025 10:28:29 +0000 Subject: [PATCH 1164/1381] can: rcar_canfd: Extract rcar_canfd_global_{,de}init() Extract the code to (de)initialize global state into separate functions, for future reuse. Signed-off-by: Geert Uytterhoeven Signed-off-by: Biju Das Link: https://patch.msgid.link/20251124102837.106973-5-biju.das.jz@bp.renesas.com Signed-off-by: Marc Kleine-Budde --- drivers/net/can/rcar/rcar_canfd.c | 182 +++++++++++++++++------------- 1 file changed, 104 insertions(+), 78 deletions(-) diff --git a/drivers/net/can/rcar/rcar_canfd.c b/drivers/net/can/rcar/rcar_canfd.c index 3f71adce6f56..f18efaf81061 100644 --- a/drivers/net/can/rcar/rcar_canfd.c +++ b/drivers/net/can/rcar/rcar_canfd.c @@ -1961,21 +1961,120 @@ static void rcar_canfd_channel_remove(struct rcar_canfd_global *gpriv, u32 ch) } } +static int rcar_canfd_global_init(struct rcar_canfd_global *gpriv) +{ + struct device *dev = &gpriv->pdev->dev; + u32 rule_entry = 0; + u32 ch, sts; + int err; + + err = reset_control_reset(gpriv->rstc1); + if (err) + return err; + + err = reset_control_reset(gpriv->rstc2); + if (err) + goto fail_reset1; + + /* Enable peripheral clock for register access */ + err = clk_prepare_enable(gpriv->clkp); + if (err) { + dev_err(dev, "failed to enable peripheral clock: %pe\n", + ERR_PTR(err)); + goto fail_reset2; + } + + /* Enable RAM clock */ + err = clk_prepare_enable(gpriv->clk_ram); + if (err) { + dev_err(dev, + "failed to enable RAM clock, error %d\n", err); + goto fail_clk; + } + + err = rcar_canfd_reset_controller(gpriv); + if (err) { + dev_err(dev, "reset controller failed: %pe\n", ERR_PTR(err)); + goto fail_ram_clk; + } + + /* Controller in Global reset & Channel reset mode */ + rcar_canfd_configure_controller(gpriv); + + /* Configure per channel attributes */ + for_each_set_bit(ch, &gpriv->channels_mask, gpriv->info->max_channels) { + /* Configure Channel's Rx fifo */ + rcar_canfd_configure_rx(gpriv, ch); + + /* Configure Channel's Tx (Common) fifo */ + rcar_canfd_configure_tx(gpriv, ch); + + /* Configure receive rules */ + rcar_canfd_configure_afl_rules(gpriv, ch, rule_entry); + rule_entry += RCANFD_CHANNEL_NUMRULES; + } + + /* Configure common interrupts */ + rcar_canfd_enable_global_interrupts(gpriv); + + /* Start Global operation mode */ + rcar_canfd_update_bit(gpriv->base, RCANFD_GCTR, RCANFD_GCTR_GMDC_MASK, + RCANFD_GCTR_GMDC_GOPM); + + /* Verify mode change */ + err = readl_poll_timeout((gpriv->base + RCANFD_GSTS), sts, + !(sts & RCANFD_GSTS_GNOPM), 2, 500000); + if (err) { + dev_err(dev, "global operational mode failed\n"); + goto fail_mode; + } + + return 0; + +fail_mode: + rcar_canfd_disable_global_interrupts(gpriv); +fail_ram_clk: + clk_disable_unprepare(gpriv->clk_ram); +fail_clk: + clk_disable_unprepare(gpriv->clkp); +fail_reset2: + reset_control_assert(gpriv->rstc2); +fail_reset1: + reset_control_assert(gpriv->rstc1); + return err; +} + +static void rcar_canfd_global_deinit(struct rcar_canfd_global *gpriv, bool full) +{ + rcar_canfd_disable_global_interrupts(gpriv); + + if (full) { + rcar_canfd_reset_controller(gpriv); + + /* Enter global sleep mode */ + rcar_canfd_set_bit(gpriv->base, RCANFD_GCTR, RCANFD_GCTR_GSLPR); + } + + clk_disable_unprepare(gpriv->clk_ram); + clk_disable_unprepare(gpriv->clkp); + reset_control_assert(gpriv->rstc2); + reset_control_assert(gpriv->rstc1); +} + static int rcar_canfd_probe(struct platform_device *pdev) { struct phy *transceivers[RCANFD_NUM_CHANNELS] = { NULL, }; const struct rcar_canfd_hw_info *info; struct device *dev = &pdev->dev; void __iomem *addr; - u32 sts, ch, fcan_freq; struct rcar_canfd_global *gpriv; struct device_node *of_child; unsigned long channels_mask = 0; int err, ch_irq, g_irq; int g_err_irq, g_recc_irq; - u32 rule_entry = 0; bool fdmode = true; /* CAN FD only mode - default */ char name[9] = "channelX"; + u32 ch, fcan_freq; int i; info = of_device_get_match_data(dev); @@ -2117,67 +2216,9 @@ static int rcar_canfd_probe(struct platform_device *pdev) } } - err = reset_control_reset(gpriv->rstc1); + err = rcar_canfd_global_init(gpriv); if (err) - goto fail_dev; - err = reset_control_reset(gpriv->rstc2); - if (err) { - reset_control_assert(gpriv->rstc1); - goto fail_dev; - } - - /* Enable peripheral clock for register access */ - err = clk_prepare_enable(gpriv->clkp); - if (err) { - dev_err(dev, "failed to enable peripheral clock: %pe\n", - ERR_PTR(err)); - goto fail_reset; - } - - /* Enable RAM clock */ - err = clk_prepare_enable(gpriv->clk_ram); - if (err) { - dev_err(dev, "failed to enable RAM clock: %pe\n", - ERR_PTR(err)); - goto fail_clk; - } - - err = rcar_canfd_reset_controller(gpriv); - if (err) { - dev_err(dev, "reset controller failed: %pe\n", ERR_PTR(err)); - goto fail_ram_clk; - } - - /* Controller in Global reset & Channel reset mode */ - rcar_canfd_configure_controller(gpriv); - - /* Configure per channel attributes */ - for_each_set_bit(ch, &gpriv->channels_mask, info->max_channels) { - /* Configure Channel's Rx fifo */ - rcar_canfd_configure_rx(gpriv, ch); - - /* Configure Channel's Tx (Common) fifo */ - rcar_canfd_configure_tx(gpriv, ch); - - /* Configure receive rules */ - rcar_canfd_configure_afl_rules(gpriv, ch, rule_entry); - rule_entry += RCANFD_CHANNEL_NUMRULES; - } - - /* Configure common interrupts */ - rcar_canfd_enable_global_interrupts(gpriv); - - /* Start Global operation mode */ - rcar_canfd_update_bit(gpriv->base, RCANFD_GCTR, RCANFD_GCTR_GMDC_MASK, - RCANFD_GCTR_GMDC_GOPM); - - /* Verify mode change */ - err = readl_poll_timeout((gpriv->base + RCANFD_GSTS), sts, - !(sts & RCANFD_GSTS_GNOPM), 2, 500000); - if (err) { - dev_err(dev, "global operational mode failed\n"); goto fail_mode; - } for_each_set_bit(ch, &gpriv->channels_mask, info->max_channels) { err = rcar_canfd_channel_probe(gpriv, ch, fcan_freq, @@ -2196,14 +2237,7 @@ static int rcar_canfd_probe(struct platform_device *pdev) for_each_set_bit(ch, &gpriv->channels_mask, info->max_channels) rcar_canfd_channel_remove(gpriv, ch); fail_mode: - rcar_canfd_disable_global_interrupts(gpriv); -fail_ram_clk: - clk_disable_unprepare(gpriv->clk_ram); -fail_clk: - clk_disable_unprepare(gpriv->clkp); -fail_reset: - reset_control_assert(gpriv->rstc2); - reset_control_assert(gpriv->rstc1); + rcar_canfd_global_deinit(gpriv, false); fail_dev: return err; } @@ -2218,15 +2252,7 @@ static void rcar_canfd_remove(struct platform_device *pdev) rcar_canfd_channel_remove(gpriv, ch); } - rcar_canfd_disable_global_interrupts(gpriv); - rcar_canfd_reset_controller(gpriv); - - /* Enter global sleep mode */ - rcar_canfd_set_bit(gpriv->base, RCANFD_GCTR, RCANFD_GCTR_GSLPR); - clk_disable_unprepare(gpriv->clk_ram); - clk_disable_unprepare(gpriv->clkp); - reset_control_assert(gpriv->rstc2); - reset_control_assert(gpriv->rstc1); + rcar_canfd_global_deinit(gpriv, true); } static int __maybe_unused rcar_canfd_suspend(struct device *dev) From ddf9bbf22b70d6b1a10e7d45de8ec3b94de7c54c Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Mon, 24 Nov 2025 10:28:30 +0000 Subject: [PATCH 1165/1381] can: rcar_canfd: Invert CAN clock and close_candev() order The CAN clock is enabled before calling open_candev(), and disabled before calling close_candev(). Invert the order of the latter, to restore symmetry. Signed-off-by: Geert Uytterhoeven Reviewed-by: Biju Das Signed-off-by: Biju Das Link: https://patch.msgid.link/20251124102837.106973-6-biju.das.jz@bp.renesas.com Signed-off-by: Marc Kleine-Budde --- drivers/net/can/rcar/rcar_canfd.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/can/rcar/rcar_canfd.c b/drivers/net/can/rcar/rcar_canfd.c index f18efaf81061..9550f1e74994 100644 --- a/drivers/net/can/rcar/rcar_canfd.c +++ b/drivers/net/can/rcar/rcar_canfd.c @@ -1570,8 +1570,8 @@ static int rcar_canfd_close(struct net_device *ndev) netif_stop_queue(ndev); rcar_canfd_stop(ndev); napi_disable(&priv->napi); - clk_disable_unprepare(gpriv->can_clk); close_candev(ndev); + clk_disable_unprepare(gpriv->can_clk); phy_power_off(priv->transceiver); return 0; } From 161266c754e71d979be994967984c9fdcab74090 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Mon, 24 Nov 2025 10:28:31 +0000 Subject: [PATCH 1166/1381] can: rcar_canfd: Convert to DEFINE_SIMPLE_DEV_PM_OPS() Convert the Renesas R-Car CAN-FD driver from SIMPLE_DEV_PM_OPS() to DEFINE_SIMPLE_DEV_PM_OPS() and pm_sleep_ptr(). This lets us drop the __maybe_unused annotations from its suspend and resume callbacks, and reduces kernel size in case CONFIG_PM or CONFIG_PM_SLEEP is disabled. Signed-off-by: Geert Uytterhoeven Reviewed-by: Biju Das Signed-off-by: Biju Das Link: https://patch.msgid.link/20251124102837.106973-7-biju.das.jz@bp.renesas.com Signed-off-by: Marc Kleine-Budde --- drivers/net/can/rcar/rcar_canfd.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/net/can/rcar/rcar_canfd.c b/drivers/net/can/rcar/rcar_canfd.c index 9550f1e74994..1d6b2f7efa8c 100644 --- a/drivers/net/can/rcar/rcar_canfd.c +++ b/drivers/net/can/rcar/rcar_canfd.c @@ -2255,18 +2255,18 @@ static void rcar_canfd_remove(struct platform_device *pdev) rcar_canfd_global_deinit(gpriv, true); } -static int __maybe_unused rcar_canfd_suspend(struct device *dev) +static int rcar_canfd_suspend(struct device *dev) { return 0; } -static int __maybe_unused rcar_canfd_resume(struct device *dev) +static int rcar_canfd_resume(struct device *dev) { return 0; } -static SIMPLE_DEV_PM_OPS(rcar_canfd_pm_ops, rcar_canfd_suspend, - rcar_canfd_resume); +static DEFINE_SIMPLE_DEV_PM_OPS(rcar_canfd_pm_ops, rcar_canfd_suspend, + rcar_canfd_resume); static const __maybe_unused struct of_device_id rcar_canfd_of_table[] = { { .compatible = "renesas,r8a779a0-canfd", .data = &rcar_gen4_hw_info }, @@ -2283,7 +2283,7 @@ static struct platform_driver rcar_canfd_driver = { .driver = { .name = RCANFD_DRV_NAME, .of_match_table = of_match_ptr(rcar_canfd_of_table), - .pm = &rcar_canfd_pm_ops, + .pm = pm_sleep_ptr(&rcar_canfd_pm_ops), }, .probe = rcar_canfd_probe, .remove = rcar_canfd_remove, From 3a34330f6339641b7f5b76b066385f56c114490f Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Mon, 24 Nov 2025 10:28:32 +0000 Subject: [PATCH 1167/1381] can: rcar_canfd: Add suspend/resume support On R-Car Gen3 using PSCI, s2ram powers down the SoC. After resume, the CAN-FD interface no longer works. Trying to bring it up again fails: # ip link set can0 up RTNETLINK answers: Connection timed out # dmesg ... channel 0 communication state failed Fix this by populating the (currently empty) suspend and resume callbacks, to stop/start the individual CAN-FD channels, and (de)initialize the CAN-FD controller. Signed-off-by: Geert Uytterhoeven Tested-by: Biju Das Signed-off-by: Biju Das Link: https://patch.msgid.link/20251124102837.106973-8-biju.das.jz@bp.renesas.com Signed-off-by: Marc Kleine-Budde --- drivers/net/can/rcar/rcar_canfd.c | 53 +++++++++++++++++++++++++++++++ 1 file changed, 53 insertions(+) diff --git a/drivers/net/can/rcar/rcar_canfd.c b/drivers/net/can/rcar/rcar_canfd.c index 1d6b2f7efa8c..ce75948e6af6 100644 --- a/drivers/net/can/rcar/rcar_canfd.c +++ b/drivers/net/can/rcar/rcar_canfd.c @@ -2257,11 +2257,64 @@ static void rcar_canfd_remove(struct platform_device *pdev) static int rcar_canfd_suspend(struct device *dev) { + struct rcar_canfd_global *gpriv = dev_get_drvdata(dev); + int err; + u32 ch; + + for_each_set_bit(ch, &gpriv->channels_mask, gpriv->info->max_channels) { + struct rcar_canfd_channel *priv = gpriv->ch[ch]; + struct net_device *ndev = priv->ndev; + + if (!netif_running(ndev)) + continue; + + netif_device_detach(ndev); + + err = rcar_canfd_close(ndev); + if (err) { + netdev_err(ndev, "rcar_canfd_close() failed %pe\n", + ERR_PTR(err)); + return err; + } + + priv->can.state = CAN_STATE_SLEEPING; + } + + /* TODO Skip if wake-up (which is not yet supported) is enabled */ + rcar_canfd_global_deinit(gpriv, false); + return 0; } static int rcar_canfd_resume(struct device *dev) { + struct rcar_canfd_global *gpriv = dev_get_drvdata(dev); + int err; + u32 ch; + + err = rcar_canfd_global_init(gpriv); + if (err) { + dev_err(dev, "rcar_canfd_global_init() failed %pe\n", ERR_PTR(err)); + return err; + } + + for_each_set_bit(ch, &gpriv->channels_mask, gpriv->info->max_channels) { + struct rcar_canfd_channel *priv = gpriv->ch[ch]; + struct net_device *ndev = priv->ndev; + + if (!netif_running(ndev)) + continue; + + err = rcar_canfd_open(ndev); + if (err) { + netdev_err(ndev, "rcar_canfd_open() failed %pe\n", + ERR_PTR(err)); + return err; + } + + netif_device_attach(ndev); + } + return 0; } From 07688a882f8ee6228e63774f23dccab6977f1bdf Mon Sep 17 00:00:00 2001 From: Markus Schneider-Pargmann Date: Wed, 19 Nov 2025 14:50:02 +0100 Subject: [PATCH 1168/1381] MAINTAINERS: Add myself as m_can maintainer As I have contributed to the m_can driver over the past years, I would like to continue maintaining the driver. As Chandrasekar is currently not responsive, I will replace him as the maintainer of the driver. Signed-off-by: Markus Schneider-Pargmann Link: https://patch.msgid.link/20251119-topic-mcan-reviewer-v6-18-v2-1-f842c3094b18@baylibre.com Signed-off-by: Marc Kleine-Budde --- MAINTAINERS | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index e9a8d945632b..92951d196d27 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -15413,7 +15413,7 @@ F: drivers/net/phy/mxl-86110.c F: drivers/net/phy/mxl-gpy.c MCAN MMIO DEVICE DRIVER -M: Chandrasekar Ramakrishnan +M: Markus Schneider-Pargmann L: linux-can@vger.kernel.org S: Maintained F: Documentation/devicetree/bindings/net/can/bosch,m_can.yaml From d20103d8f880b0d716ae83dba6d1d150f5790a9c Mon Sep 17 00:00:00 2001 From: Markus Schneider-Pargmann Date: Wed, 19 Nov 2025 14:50:03 +0100 Subject: [PATCH 1169/1381] MAINTAINERS: Simplify m_can section Simplify the section by using the whole m_can directory. This includes a few new files, e.g. Kconfig, Makefile, m_can_pci.c and tcan4x5x* which are all closely coupled to the m_can driver core. Signed-off-by: Markus Schneider-Pargmann Link: https://patch.msgid.link/20251119-topic-mcan-reviewer-v6-18-v2-2-f842c3094b18@baylibre.com Signed-off-by: Marc Kleine-Budde --- MAINTAINERS | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/MAINTAINERS b/MAINTAINERS index 92951d196d27..cb7168eb0ce4 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -15412,14 +15412,12 @@ S: Supported F: drivers/net/phy/mxl-86110.c F: drivers/net/phy/mxl-gpy.c -MCAN MMIO DEVICE DRIVER +MCAN DEVICE DRIVER M: Markus Schneider-Pargmann L: linux-can@vger.kernel.org S: Maintained F: Documentation/devicetree/bindings/net/can/bosch,m_can.yaml -F: drivers/net/can/m_can/m_can.c -F: drivers/net/can/m_can/m_can.h -F: drivers/net/can/m_can/m_can_platform.c +F: drivers/net/can/m_can/ MCBA MICROCHIP CAN BUS ANALYZER TOOL DRIVER R: Yasushi SHOJI From 9aea35eb98a6560daf85a2ae9cbd482a66e4d076 Mon Sep 17 00:00:00 2001 From: Conor Dooley Date: Fri, 21 Nov 2025 13:42:30 +0000 Subject: [PATCH 1170/1381] dt-bindings: can: mpfs: document resets The CAN cores on Polarfire SoC both have a reset. The platform firmware brings both cores out of reset, but the linux driver must use them during normal operation. The resets should have been made required, but this is one of the things that can happen when the binding is written without driver support. Fixes: c878d518d7b6 ("dt-bindings: can: mpfs: document the mpfs CAN controller") Signed-off-by: Conor Dooley Reviewed-by: Krzysztof Kozlowski Link: https://patch.msgid.link/20251121-sample-footsore-743d81772efc@spud Signed-off-by: Marc Kleine-Budde --- .../devicetree/bindings/net/can/microchip,mpfs-can.yaml | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/Documentation/devicetree/bindings/net/can/microchip,mpfs-can.yaml b/Documentation/devicetree/bindings/net/can/microchip,mpfs-can.yaml index 1219c5cb601f..519a11fbe972 100644 --- a/Documentation/devicetree/bindings/net/can/microchip,mpfs-can.yaml +++ b/Documentation/devicetree/bindings/net/can/microchip,mpfs-can.yaml @@ -32,11 +32,15 @@ properties: - description: AHB peripheral clock - description: CAN bus clock + resets: + maxItems: 1 + required: - compatible - reg - interrupts - clocks + - resets additionalProperties: false @@ -46,6 +50,7 @@ examples: compatible = "microchip,mpfs-can"; reg = <0x2010c000 0x1000>; clocks = <&clkcfg 17>, <&clkcfg 37>; + resets = <&clkcfg 17>; interrupt-parent = <&plic>; interrupts = <56>; }; From af7273cc7ae01f5b3e34e62f59588ce79fe50f79 Mon Sep 17 00:00:00 2001 From: Ankit Khushwaha Date: Tue, 25 Nov 2025 22:23:02 +0530 Subject: [PATCH 1171/1381] selftests/net: initialize char variable to null char variable in 'so_txtime.c' & 'txtimestamp.c' were left uninitilized when switch default case taken. which raises following warning. txtimestamp.c:240:2: warning: variable 'tsname' is used uninitialized whenever switch default is taken [-Wsometimes-uninitialized] so_txtime.c:210:3: warning: variable 'reason' is used uninitialized whenever switch default is taken [-Wsometimes-uninitialized] initializing these variables to NULL to fix this. Signed-off-by: Ankit Khushwaha Link: https://patch.msgid.link/20251125165302.20079-1-ankitkhushwaha.linux@gmail.com Signed-off-by: Jakub Kicinski --- tools/testing/selftests/net/so_txtime.c | 2 +- tools/testing/selftests/net/txtimestamp.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/tools/testing/selftests/net/so_txtime.c b/tools/testing/selftests/net/so_txtime.c index 8457b7ccbc09..b76df1efc2ef 100644 --- a/tools/testing/selftests/net/so_txtime.c +++ b/tools/testing/selftests/net/so_txtime.c @@ -174,7 +174,7 @@ static int do_recv_errqueue_timeout(int fdt) msg.msg_controllen = sizeof(control); while (1) { - const char *reason; + const char *reason = NULL; ret = recvmsg(fdt, &msg, MSG_ERRQUEUE); if (ret == -1 && errno == EAGAIN) diff --git a/tools/testing/selftests/net/txtimestamp.c b/tools/testing/selftests/net/txtimestamp.c index dae91eb97d69..bcc14688661d 100644 --- a/tools/testing/selftests/net/txtimestamp.c +++ b/tools/testing/selftests/net/txtimestamp.c @@ -217,7 +217,7 @@ static void print_timestamp_usr(void) static void print_timestamp(struct scm_timestamping *tss, int tstype, int tskey, int payload_len) { - const char *tsname; + const char *tsname = NULL; validate_key(tskey, tstype); From c01a6e5b2e4f21d31cf725b9f3803cb0280b1b8d Mon Sep 17 00:00:00 2001 From: Willem de Bruijn Date: Tue, 25 Nov 2025 18:35:05 -0500 Subject: [PATCH 1172/1381] selftests/net: packetdrill: pass send_omit_free to MSG_ZEROCOPY tests The --send_omit_free flag is needed for TCP zero copy tests, to ensure that packetdrill doesn't free the send() buffer after the send() call. Fixes: 1e42f73fd3c2 ("selftests/net: packetdrill: import tcp/zerocopy") Closes: https://lore.kernel.org/netdev/20251124071831.4cbbf412@kernel.org/ Suggested-by: Neal Cardwell Signed-off-by: Willem de Bruijn Link: https://patch.msgid.link/20251125234029.1320984-1-willemdebruijn.kernel@gmail.com Signed-off-by: Jakub Kicinski --- .../net/packetdrill/tcp_syscall_bad_arg_sendmsg-empty-iov.pkt | 4 ++++ .../testing/selftests/net/packetdrill/tcp_zerocopy_basic.pkt | 2 ++ .../testing/selftests/net/packetdrill/tcp_zerocopy_batch.pkt | 2 ++ .../testing/selftests/net/packetdrill/tcp_zerocopy_client.pkt | 2 ++ .../testing/selftests/net/packetdrill/tcp_zerocopy_closed.pkt | 2 ++ .../selftests/net/packetdrill/tcp_zerocopy_epoll_edge.pkt | 3 +++ .../net/packetdrill/tcp_zerocopy_epoll_exclusive.pkt | 3 +++ .../selftests/net/packetdrill/tcp_zerocopy_epoll_oneshot.pkt | 3 +++ .../net/packetdrill/tcp_zerocopy_fastopen-client.pkt | 2 ++ .../net/packetdrill/tcp_zerocopy_fastopen-server.pkt | 2 ++ .../selftests/net/packetdrill/tcp_zerocopy_maxfrags.pkt | 2 ++ .../testing/selftests/net/packetdrill/tcp_zerocopy_small.pkt | 2 ++ 12 files changed, 29 insertions(+) diff --git a/tools/testing/selftests/net/packetdrill/tcp_syscall_bad_arg_sendmsg-empty-iov.pkt b/tools/testing/selftests/net/packetdrill/tcp_syscall_bad_arg_sendmsg-empty-iov.pkt index b2b2cdf27e20..454441e7ecff 100644 --- a/tools/testing/selftests/net/packetdrill/tcp_syscall_bad_arg_sendmsg-empty-iov.pkt +++ b/tools/testing/selftests/net/packetdrill/tcp_syscall_bad_arg_sendmsg-empty-iov.pkt @@ -1,6 +1,10 @@ // SPDX-License-Identifier: GPL-2.0 // Test that we correctly skip zero-length IOVs. + +--send_omit_free // do not reuse send buffers with zerocopy + `./defaults.sh` + 0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3 +0 setsockopt(3, SOL_SOCKET, SO_ZEROCOPY, [1], 4) = 0 +0 setsockopt(3, SOL_SOCKET, SO_REUSEADDR, [1], 4) = 0 diff --git a/tools/testing/selftests/net/packetdrill/tcp_zerocopy_basic.pkt b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_basic.pkt index a82c8899d36b..0a0700afdaa3 100644 --- a/tools/testing/selftests/net/packetdrill/tcp_zerocopy_basic.pkt +++ b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_basic.pkt @@ -4,6 +4,8 @@ // send a packet with MSG_ZEROCOPY and receive the notification ID // repeat and verify IDs are consecutive +--send_omit_free // do not reuse send buffers with zerocopy + `./defaults.sh` 0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3 diff --git a/tools/testing/selftests/net/packetdrill/tcp_zerocopy_batch.pkt b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_batch.pkt index c01915e7f4a1..df91675d2991 100644 --- a/tools/testing/selftests/net/packetdrill/tcp_zerocopy_batch.pkt +++ b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_batch.pkt @@ -3,6 +3,8 @@ // // send multiple packets, then read one range of all notifications. +--send_omit_free // do not reuse send buffers with zerocopy + `./defaults.sh` 0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3 diff --git a/tools/testing/selftests/net/packetdrill/tcp_zerocopy_client.pkt b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_client.pkt index 6509882932e9..2963cfcb14df 100644 --- a/tools/testing/selftests/net/packetdrill/tcp_zerocopy_client.pkt +++ b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_client.pkt @@ -1,6 +1,8 @@ // SPDX-License-Identifier: GPL-2.0 // Minimal client-side zerocopy test +--send_omit_free // do not reuse send buffers with zerocopy + `./defaults.sh` 0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 4 diff --git a/tools/testing/selftests/net/packetdrill/tcp_zerocopy_closed.pkt b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_closed.pkt index 2cd78755cb2a..ea0c2fa73c2d 100644 --- a/tools/testing/selftests/net/packetdrill/tcp_zerocopy_closed.pkt +++ b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_closed.pkt @@ -7,6 +7,8 @@ // First send on a closed socket and wait for (absent) notification. // Then connect and send and verify that notification nr. is zero. +--send_omit_free // do not reuse send buffers with zerocopy + `./defaults.sh` 0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 4 diff --git a/tools/testing/selftests/net/packetdrill/tcp_zerocopy_epoll_edge.pkt b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_epoll_edge.pkt index 7671c20e01cf..4df978a9b82e 100644 --- a/tools/testing/selftests/net/packetdrill/tcp_zerocopy_epoll_edge.pkt +++ b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_epoll_edge.pkt @@ -7,6 +7,9 @@ // fire two sends with MSG_ZEROCOPY and receive the acks. confirm that EPOLLERR // is correctly fired only once, when EPOLLET is set. send another packet with // MSG_ZEROCOPY. confirm that EPOLLERR is correctly fired again only once. + +--send_omit_free // do not reuse send buffers with zerocopy + `./defaults.sh` 0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3 diff --git a/tools/testing/selftests/net/packetdrill/tcp_zerocopy_epoll_exclusive.pkt b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_epoll_exclusive.pkt index fadc480fdb7f..36b6edc4858c 100644 --- a/tools/testing/selftests/net/packetdrill/tcp_zerocopy_epoll_exclusive.pkt +++ b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_epoll_exclusive.pkt @@ -8,6 +8,9 @@ // fire two sends with MSG_ZEROCOPY and receive the acks. confirm that EPOLLERR // is correctly fired only once, when EPOLLET is set. send another packet with // MSG_ZEROCOPY. confirm that EPOLLERR is correctly fired again only once. + +--send_omit_free // do not reuse send buffers with zerocopy + `./defaults.sh` 0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3 diff --git a/tools/testing/selftests/net/packetdrill/tcp_zerocopy_epoll_oneshot.pkt b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_epoll_oneshot.pkt index 5bfa0d1d2f4a..1bea6f3b4558 100644 --- a/tools/testing/selftests/net/packetdrill/tcp_zerocopy_epoll_oneshot.pkt +++ b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_epoll_oneshot.pkt @@ -8,6 +8,9 @@ // is correctly fired only once, when EPOLLONESHOT is set. send another packet // with MSG_ZEROCOPY. confirm that EPOLLERR is not fired. Rearm the FD and // confirm that EPOLLERR is correctly set. + +--send_omit_free // do not reuse send buffers with zerocopy + `./defaults.sh` 0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3 diff --git a/tools/testing/selftests/net/packetdrill/tcp_zerocopy_fastopen-client.pkt b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_fastopen-client.pkt index 4a73bbf46961..e27c21ff5d18 100644 --- a/tools/testing/selftests/net/packetdrill/tcp_zerocopy_fastopen-client.pkt +++ b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_fastopen-client.pkt @@ -8,6 +8,8 @@ // one will have no data in the initial send. On return 0 the // zerocopy notification counter is not incremented. Verify this too. +--send_omit_free // do not reuse send buffers with zerocopy + `./defaults.sh` // Send a FastOpen request, no cookie yet so no data in SYN diff --git a/tools/testing/selftests/net/packetdrill/tcp_zerocopy_fastopen-server.pkt b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_fastopen-server.pkt index 36086c5877ce..b1fa77c77dfa 100644 --- a/tools/testing/selftests/net/packetdrill/tcp_zerocopy_fastopen-server.pkt +++ b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_fastopen-server.pkt @@ -4,6 +4,8 @@ // send data with MSG_FASTOPEN | MSG_ZEROCOPY and verify that the // kernel returns the notification ID. +--send_omit_free // do not reuse send buffers with zerocopy + `./defaults.sh ./set_sysctls.py /proc/sys/net/ipv4/tcp_fastopen=0x207` diff --git a/tools/testing/selftests/net/packetdrill/tcp_zerocopy_maxfrags.pkt b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_maxfrags.pkt index 672f817faca0..2f5317d0a9fa 100644 --- a/tools/testing/selftests/net/packetdrill/tcp_zerocopy_maxfrags.pkt +++ b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_maxfrags.pkt @@ -7,6 +7,8 @@ // because each iovec element becomes a frag // 3) the PSH bit is set on an skb when it runs out of fragments +--send_omit_free // do not reuse send buffers with zerocopy + `./defaults.sh` 0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3 diff --git a/tools/testing/selftests/net/packetdrill/tcp_zerocopy_small.pkt b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_small.pkt index a9a1ac0aea4f..9d5272c6b207 100644 --- a/tools/testing/selftests/net/packetdrill/tcp_zerocopy_small.pkt +++ b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_small.pkt @@ -4,6 +4,8 @@ // verify that SO_EE_CODE_ZEROCOPY_COPIED is set on zerocopy // packets of all sizes, including the smallest payload, 1B. +--send_omit_free // do not reuse send buffers with zerocopy + `./defaults.sh` 0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3 From 6aac2aa2dfae38b60f22c3dfe4103ceefbe2d761 Mon Sep 17 00:00:00 2001 From: Vadim Fedorenko Date: Mon, 24 Nov 2025 18:11:45 +0000 Subject: [PATCH 1173/1381] phy: rename hwtstamp callback to hwtstamp_set PHY devices has hwtstamp callback which actually performs set operation. Rename it to better reflect the action. Reviewed-by: Russell King (Oracle) Reviewed-by: Kory Maincent Reviewed-by: Andrew Lunn Signed-off-by: Vadim Fedorenko Link: https://patch.msgid.link/20251124181151.277256-2-vadim.fedorenko@linux.dev Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/ti/netcp_ethss.c | 2 +- drivers/net/phy/bcm-phy-ptp.c | 8 ++++---- drivers/net/phy/dp83640.c | 8 ++++---- drivers/net/phy/micrel.c | 16 ++++++++-------- drivers/net/phy/microchip_rds_ptp.c | 8 ++++---- drivers/net/phy/mscc/mscc_ptp.c | 8 ++++---- drivers/net/phy/nxp-c45-tja11xx.c | 8 ++++---- drivers/net/phy/phy.c | 11 +++++++---- drivers/ptp/ptp_ines.c | 8 ++++---- include/linux/mii_timestamper.h | 8 ++++---- include/linux/phy.h | 4 ++-- 11 files changed, 46 insertions(+), 43 deletions(-) diff --git a/drivers/net/ethernet/ti/netcp_ethss.c b/drivers/net/ethernet/ti/netcp_ethss.c index 4f6cc6cd1f03..8f46e9be76b1 100644 --- a/drivers/net/ethernet/ti/netcp_ethss.c +++ b/drivers/net/ethernet/ti/netcp_ethss.c @@ -2657,7 +2657,7 @@ static int gbe_hwtstamp_set(void *intf_priv, struct kernel_hwtstamp_config *cfg, phy = gbe_intf->slave->phy; if (phy_has_hwtstamp(phy)) - return phy->mii_ts->hwtstamp(phy->mii_ts, cfg, extack); + return phy->mii_ts->hwtstamp_set(phy->mii_ts, cfg, extack); switch (cfg->tx_type) { case HWTSTAMP_TX_OFF: diff --git a/drivers/net/phy/bcm-phy-ptp.c b/drivers/net/phy/bcm-phy-ptp.c index d3501f8487d9..6815e844a62e 100644 --- a/drivers/net/phy/bcm-phy-ptp.c +++ b/drivers/net/phy/bcm-phy-ptp.c @@ -780,9 +780,9 @@ static void bcm_ptp_txtstamp(struct mii_timestamper *mii_ts, kfree_skb(skb); } -static int bcm_ptp_hwtstamp(struct mii_timestamper *mii_ts, - struct kernel_hwtstamp_config *cfg, - struct netlink_ext_ack *extack) +static int bcm_ptp_hwtstamp_set(struct mii_timestamper *mii_ts, + struct kernel_hwtstamp_config *cfg, + struct netlink_ext_ack *extack) { struct bcm_ptp_private *priv = mii2priv(mii_ts); u16 mode, ctrl; @@ -898,7 +898,7 @@ static void bcm_ptp_init(struct bcm_ptp_private *priv) priv->mii_ts.rxtstamp = bcm_ptp_rxtstamp; priv->mii_ts.txtstamp = bcm_ptp_txtstamp; - priv->mii_ts.hwtstamp = bcm_ptp_hwtstamp; + priv->mii_ts.hwtstamp_set = bcm_ptp_hwtstamp_set; priv->mii_ts.ts_info = bcm_ptp_ts_info; priv->phydev->mii_ts = &priv->mii_ts; diff --git a/drivers/net/phy/dp83640.c b/drivers/net/phy/dp83640.c index 74396453f5bb..f733a8b72d40 100644 --- a/drivers/net/phy/dp83640.c +++ b/drivers/net/phy/dp83640.c @@ -1176,9 +1176,9 @@ static irqreturn_t dp83640_handle_interrupt(struct phy_device *phydev) return IRQ_HANDLED; } -static int dp83640_hwtstamp(struct mii_timestamper *mii_ts, - struct kernel_hwtstamp_config *cfg, - struct netlink_ext_ack *extack) +static int dp83640_hwtstamp_set(struct mii_timestamper *mii_ts, + struct kernel_hwtstamp_config *cfg, + struct netlink_ext_ack *extack) { struct dp83640_private *dp83640 = container_of(mii_ts, struct dp83640_private, mii_ts); @@ -1407,7 +1407,7 @@ static int dp83640_probe(struct phy_device *phydev) dp83640->phydev = phydev; dp83640->mii_ts.rxtstamp = dp83640_rxtstamp; dp83640->mii_ts.txtstamp = dp83640_txtstamp; - dp83640->mii_ts.hwtstamp = dp83640_hwtstamp; + dp83640->mii_ts.hwtstamp_set = dp83640_hwtstamp_set; dp83640->mii_ts.ts_info = dp83640_ts_info; INIT_DELAYED_WORK(&dp83640->ts_work, rx_timestamp_work); diff --git a/drivers/net/phy/micrel.c b/drivers/net/phy/micrel.c index 5d90ccc20df7..05de68b9f719 100644 --- a/drivers/net/phy/micrel.c +++ b/drivers/net/phy/micrel.c @@ -3147,9 +3147,9 @@ static void lan8814_flush_fifo(struct phy_device *phydev, bool egress) lanphy_read_page_reg(phydev, LAN8814_PAGE_PORT_REGS, PTP_TSU_INT_STS); } -static int lan8814_hwtstamp(struct mii_timestamper *mii_ts, - struct kernel_hwtstamp_config *config, - struct netlink_ext_ack *extack) +static int lan8814_hwtstamp_set(struct mii_timestamper *mii_ts, + struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack) { struct kszphy_ptp_priv *ptp_priv = container_of(mii_ts, struct kszphy_ptp_priv, mii_ts); @@ -4389,7 +4389,7 @@ static void lan8814_ptp_init(struct phy_device *phydev) ptp_priv->mii_ts.rxtstamp = lan8814_rxtstamp; ptp_priv->mii_ts.txtstamp = lan8814_txtstamp; - ptp_priv->mii_ts.hwtstamp = lan8814_hwtstamp; + ptp_priv->mii_ts.hwtstamp_set = lan8814_hwtstamp_set; ptp_priv->mii_ts.ts_info = lan8814_ts_info; phydev->mii_ts = &ptp_priv->mii_ts; @@ -5042,9 +5042,9 @@ static void lan8841_ptp_enable_processing(struct kszphy_ptp_priv *ptp_priv, #define LAN8841_PTP_TX_TIMESTAMP_EN 443 #define LAN8841_PTP_TX_MOD 445 -static int lan8841_hwtstamp(struct mii_timestamper *mii_ts, - struct kernel_hwtstamp_config *config, - struct netlink_ext_ack *extack) +static int lan8841_hwtstamp_set(struct mii_timestamper *mii_ts, + struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack) { struct kszphy_ptp_priv *ptp_priv = container_of(mii_ts, struct kszphy_ptp_priv, mii_ts); struct phy_device *phydev = ptp_priv->phydev; @@ -5924,7 +5924,7 @@ static int lan8841_probe(struct phy_device *phydev) ptp_priv->mii_ts.rxtstamp = lan8841_rxtstamp; ptp_priv->mii_ts.txtstamp = lan8814_txtstamp; - ptp_priv->mii_ts.hwtstamp = lan8841_hwtstamp; + ptp_priv->mii_ts.hwtstamp_set = lan8841_hwtstamp_set; ptp_priv->mii_ts.ts_info = lan8841_ts_info; phydev->mii_ts = &ptp_priv->mii_ts; diff --git a/drivers/net/phy/microchip_rds_ptp.c b/drivers/net/phy/microchip_rds_ptp.c index e6514ce04c29..4c6326b0ceaf 100644 --- a/drivers/net/phy/microchip_rds_ptp.c +++ b/drivers/net/phy/microchip_rds_ptp.c @@ -476,9 +476,9 @@ static bool mchp_rds_ptp_rxtstamp(struct mii_timestamper *mii_ts, return true; } -static int mchp_rds_ptp_hwtstamp(struct mii_timestamper *mii_ts, - struct kernel_hwtstamp_config *config, - struct netlink_ext_ack *extack) +static int mchp_rds_ptp_hwtstamp_set(struct mii_timestamper *mii_ts, + struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack) { struct mchp_rds_ptp_clock *clock = container_of(mii_ts, struct mchp_rds_ptp_clock, @@ -1281,7 +1281,7 @@ struct mchp_rds_ptp_clock *mchp_rds_ptp_probe(struct phy_device *phydev, u8 mmd, clock->mii_ts.rxtstamp = mchp_rds_ptp_rxtstamp; clock->mii_ts.txtstamp = mchp_rds_ptp_txtstamp; - clock->mii_ts.hwtstamp = mchp_rds_ptp_hwtstamp; + clock->mii_ts.hwtstamp_set = mchp_rds_ptp_hwtstamp_set; clock->mii_ts.ts_info = mchp_rds_ptp_ts_info; phydev->mii_ts = &clock->mii_ts; diff --git a/drivers/net/phy/mscc/mscc_ptp.c b/drivers/net/phy/mscc/mscc_ptp.c index d692df7d975c..dc06614222f6 100644 --- a/drivers/net/phy/mscc/mscc_ptp.c +++ b/drivers/net/phy/mscc/mscc_ptp.c @@ -1051,9 +1051,9 @@ static void vsc85xx_ts_reset_fifo(struct phy_device *phydev) val); } -static int vsc85xx_hwtstamp(struct mii_timestamper *mii_ts, - struct kernel_hwtstamp_config *cfg, - struct netlink_ext_ack *extack) +static int vsc85xx_hwtstamp_set(struct mii_timestamper *mii_ts, + struct kernel_hwtstamp_config *cfg, + struct netlink_ext_ack *extack) { struct vsc8531_private *vsc8531 = container_of(mii_ts, struct vsc8531_private, mii_ts); @@ -1611,7 +1611,7 @@ int vsc8584_ptp_probe(struct phy_device *phydev) vsc8531->mii_ts.rxtstamp = vsc85xx_rxtstamp; vsc8531->mii_ts.txtstamp = vsc85xx_txtstamp; - vsc8531->mii_ts.hwtstamp = vsc85xx_hwtstamp; + vsc8531->mii_ts.hwtstamp_set = vsc85xx_hwtstamp_set; vsc8531->mii_ts.ts_info = vsc85xx_ts_info; phydev->mii_ts = &vsc8531->mii_ts; diff --git a/drivers/net/phy/nxp-c45-tja11xx.c b/drivers/net/phy/nxp-c45-tja11xx.c index 87adb6508017..13a8fac223a9 100644 --- a/drivers/net/phy/nxp-c45-tja11xx.c +++ b/drivers/net/phy/nxp-c45-tja11xx.c @@ -1012,9 +1012,9 @@ static bool nxp_c45_rxtstamp(struct mii_timestamper *mii_ts, return true; } -static int nxp_c45_hwtstamp(struct mii_timestamper *mii_ts, - struct kernel_hwtstamp_config *cfg, - struct netlink_ext_ack *extack) +static int nxp_c45_hwtstamp_set(struct mii_timestamper *mii_ts, + struct kernel_hwtstamp_config *cfg, + struct netlink_ext_ack *extack) { struct nxp_c45_phy *priv = container_of(mii_ts, struct nxp_c45_phy, mii_ts); @@ -1749,7 +1749,7 @@ static int nxp_c45_probe(struct phy_device *phydev) IS_ENABLED(CONFIG_NETWORK_PHY_TIMESTAMPING)) { priv->mii_ts.rxtstamp = nxp_c45_rxtstamp; priv->mii_ts.txtstamp = nxp_c45_txtstamp; - priv->mii_ts.hwtstamp = nxp_c45_hwtstamp; + priv->mii_ts.hwtstamp_set = nxp_c45_hwtstamp_set; priv->mii_ts.ts_info = nxp_c45_ts_info; phydev->mii_ts = &priv->mii_ts; ret = nxp_c45_init_ptp_clock(priv); diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c index 02da4a203ddd..350bc23c1fdb 100644 --- a/drivers/net/phy/phy.c +++ b/drivers/net/phy/phy.c @@ -405,12 +405,14 @@ int phy_mii_ioctl(struct phy_device *phydev, struct ifreq *ifr, int cmd) return 0; case SIOCSHWTSTAMP: - if (phydev->mii_ts && phydev->mii_ts->hwtstamp) { + if (phydev->mii_ts && phydev->mii_ts->hwtstamp_set) { if (copy_from_user(&cfg, ifr->ifr_data, sizeof(cfg))) return -EFAULT; hwtstamp_config_to_kernel(&kernel_cfg, &cfg); - ret = phydev->mii_ts->hwtstamp(phydev->mii_ts, &kernel_cfg, &extack); + ret = phydev->mii_ts->hwtstamp_set(phydev->mii_ts, + &kernel_cfg, + &extack); if (ret) return ret; @@ -493,8 +495,9 @@ int __phy_hwtstamp_set(struct phy_device *phydev, if (!phydev) return -ENODEV; - if (phydev->mii_ts && phydev->mii_ts->hwtstamp) - return phydev->mii_ts->hwtstamp(phydev->mii_ts, config, extack); + if (phydev->mii_ts && phydev->mii_ts->hwtstamp_set) + return phydev->mii_ts->hwtstamp_set(phydev->mii_ts, config, + extack); return -EOPNOTSUPP; } diff --git a/drivers/ptp/ptp_ines.c b/drivers/ptp/ptp_ines.c index 68f1f7fdaa9d..56c798e77f20 100644 --- a/drivers/ptp/ptp_ines.c +++ b/drivers/ptp/ptp_ines.c @@ -328,9 +328,9 @@ static u64 ines_find_txts(struct ines_port *port, struct sk_buff *skb) return ns; } -static int ines_hwtstamp(struct mii_timestamper *mii_ts, - struct kernel_hwtstamp_config *cfg, - struct netlink_ext_ack *extack) +static int ines_hwtstamp_set(struct mii_timestamper *mii_ts, + struct kernel_hwtstamp_config *cfg, + struct netlink_ext_ack *extack) { struct ines_port *port = container_of(mii_ts, struct ines_port, mii_ts); u32 cm_one_step = 0, port_conf, ts_stat_rx, ts_stat_tx; @@ -709,7 +709,7 @@ static struct mii_timestamper *ines_ptp_probe_channel(struct device *device, } port->mii_ts.rxtstamp = ines_rxtstamp; port->mii_ts.txtstamp = ines_txtstamp; - port->mii_ts.hwtstamp = ines_hwtstamp; + port->mii_ts.hwtstamp_set = ines_hwtstamp_set; port->mii_ts.link_state = ines_link_state; port->mii_ts.ts_info = ines_ts_info; diff --git a/include/linux/mii_timestamper.h b/include/linux/mii_timestamper.h index 995db62570f9..08863c0e9ea3 100644 --- a/include/linux/mii_timestamper.h +++ b/include/linux/mii_timestamper.h @@ -27,7 +27,7 @@ struct phy_device; * as soon as a timestamp becomes available. One of the PTP_CLASS_ * values is passed in 'type'. * - * @hwtstamp: Handles SIOCSHWTSTAMP ioctl for hardware time stamping. + * @hwtstamp_set: Handles SIOCSHWTSTAMP ioctl for hardware time stamping. * * @link_state: Allows the device to respond to changes in the link * state. The caller invokes this function while holding @@ -51,9 +51,9 @@ struct mii_timestamper { void (*txtstamp)(struct mii_timestamper *mii_ts, struct sk_buff *skb, int type); - int (*hwtstamp)(struct mii_timestamper *mii_ts, - struct kernel_hwtstamp_config *kernel_config, - struct netlink_ext_ack *extack); + int (*hwtstamp_set)(struct mii_timestamper *mii_ts, + struct kernel_hwtstamp_config *kernel_config, + struct netlink_ext_ack *extack); void (*link_state)(struct mii_timestamper *mii_ts, struct phy_device *phydev); diff --git a/include/linux/phy.h b/include/linux/phy.h index 65b0c3ca6a2b..059a104223c4 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -1915,7 +1915,7 @@ static inline bool phy_polling_mode(struct phy_device *phydev) */ static inline bool phy_has_hwtstamp(struct phy_device *phydev) { - return phydev && phydev->mii_ts && phydev->mii_ts->hwtstamp; + return phydev && phydev->mii_ts && phydev->mii_ts->hwtstamp_set; } /** @@ -1950,7 +1950,7 @@ static inline int phy_hwtstamp(struct phy_device *phydev, struct kernel_hwtstamp_config *cfg, struct netlink_ext_ack *extack) { - return phydev->mii_ts->hwtstamp(phydev->mii_ts, cfg, extack); + return phydev->mii_ts->hwtstamp_set(phydev->mii_ts, cfg, extack); } static inline bool phy_rxtstamp(struct phy_device *phydev, struct sk_buff *skb, From f467777efbfb8034d813b601b961b25f777b3d37 Mon Sep 17 00:00:00 2001 From: Vadim Fedorenko Date: Mon, 24 Nov 2025 18:11:46 +0000 Subject: [PATCH 1174/1381] phy: add hwtstamp_get callback to phy drivers PHY devices had lack of hwtstamp_get callback even though most of them are tracking configuration info. Introduce new call back to mii_timestamper. Reviewed-by: Russell King (Oracle) Reviewed-by: Kory Maincent Signed-off-by: Vadim Fedorenko Link: https://patch.msgid.link/20251124181151.277256-3-vadim.fedorenko@linux.dev Signed-off-by: Jakub Kicinski --- drivers/net/phy/phy.c | 3 +++ include/linux/mii_timestamper.h | 5 +++++ net/core/dev_ioctl.c | 9 +++++---- 3 files changed, 13 insertions(+), 4 deletions(-) diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c index 350bc23c1fdb..13dd1691886d 100644 --- a/drivers/net/phy/phy.c +++ b/drivers/net/phy/phy.c @@ -478,6 +478,9 @@ int __phy_hwtstamp_get(struct phy_device *phydev, if (!phydev) return -ENODEV; + if (phydev->mii_ts && phydev->mii_ts->hwtstamp_get) + return phydev->mii_ts->hwtstamp_get(phydev->mii_ts, config); + return -EOPNOTSUPP; } diff --git a/include/linux/mii_timestamper.h b/include/linux/mii_timestamper.h index 08863c0e9ea3..3102c425c8e0 100644 --- a/include/linux/mii_timestamper.h +++ b/include/linux/mii_timestamper.h @@ -29,6 +29,8 @@ struct phy_device; * * @hwtstamp_set: Handles SIOCSHWTSTAMP ioctl for hardware time stamping. * + * @hwtstamp_get: Handles SIOCGHWTSTAMP ioctl for hardware time stamping. + * * @link_state: Allows the device to respond to changes in the link * state. The caller invokes this function while holding * the phy_device mutex. @@ -55,6 +57,9 @@ struct mii_timestamper { struct kernel_hwtstamp_config *kernel_config, struct netlink_ext_ack *extack); + int (*hwtstamp_get)(struct mii_timestamper *mii_ts, + struct kernel_hwtstamp_config *kernel_config); + void (*link_state)(struct mii_timestamper *mii_ts, struct phy_device *phydev); diff --git a/net/core/dev_ioctl.c b/net/core/dev_ioctl.c index 0720ccc14df9..53a53357cfef 100644 --- a/net/core/dev_ioctl.c +++ b/net/core/dev_ioctl.c @@ -249,10 +249,11 @@ int net_hwtstamp_validate(const struct kernel_hwtstamp_config *cfg) * * Helper for calling the default hardware provider timestamping. * - * Note: phy_mii_ioctl() only handles SIOCSHWTSTAMP (not SIOCGHWTSTAMP), and - * there only exists a phydev->mii_ts->hwtstamp() method. So this will return - * -EOPNOTSUPP for phylib for now, which is still more accurate than letting - * the netdev handle the GET request. + * Note: phy_mii_ioctl() only handles SIOCSHWTSTAMP (not SIOCGHWTSTAMP), but + * phydev->mii_ts has both hwtstamp_get() and hwtstamp_set() methods. So this + * will return -EOPNOTSUPP for phylib only if hwtstamp_get() is not + * implemented for now, which is still more accurate than letting the netdev + * handle the GET request. */ int dev_get_hwtstamp_phylib(struct net_device *dev, struct kernel_hwtstamp_config *cfg) From 1cff8392df0cb43716204948cc82633af14e67fd Mon Sep 17 00:00:00 2001 From: Vadim Fedorenko Date: Mon, 24 Nov 2025 18:11:47 +0000 Subject: [PATCH 1175/1381] net: phy: broadcom: add HW timestamp configuration reporting The driver stores configuration information and can technically report it. Implement hwtstamp_get callback to report the configuration. Reviewed-by: Kory Maincent Signed-off-by: Vadim Fedorenko Reviewed-by: Florian Fainelli Link: https://patch.msgid.link/20251124181151.277256-4-vadim.fedorenko@linux.dev Signed-off-by: Jakub Kicinski --- drivers/net/phy/bcm-phy-ptp.c | 13 +++++++++++++ 1 file changed, 13 insertions(+) diff --git a/drivers/net/phy/bcm-phy-ptp.c b/drivers/net/phy/bcm-phy-ptp.c index 6815e844a62e..65d609ed69fb 100644 --- a/drivers/net/phy/bcm-phy-ptp.c +++ b/drivers/net/phy/bcm-phy-ptp.c @@ -780,6 +780,18 @@ static void bcm_ptp_txtstamp(struct mii_timestamper *mii_ts, kfree_skb(skb); } +static int bcm_ptp_hwtstamp_get(struct mii_timestamper *mii_ts, + struct kernel_hwtstamp_config *cfg) +{ + struct bcm_ptp_private *priv = mii2priv(mii_ts); + + cfg->rx_filter = priv->hwts_rx ? HWTSTAMP_FILTER_PTP_V2_EVENT + : HWTSTAMP_FILTER_NONE; + cfg->tx_type = priv->tx_type; + + return 0; +} + static int bcm_ptp_hwtstamp_set(struct mii_timestamper *mii_ts, struct kernel_hwtstamp_config *cfg, struct netlink_ext_ack *extack) @@ -899,6 +911,7 @@ static void bcm_ptp_init(struct bcm_ptp_private *priv) priv->mii_ts.rxtstamp = bcm_ptp_rxtstamp; priv->mii_ts.txtstamp = bcm_ptp_txtstamp; priv->mii_ts.hwtstamp_set = bcm_ptp_hwtstamp_set; + priv->mii_ts.hwtstamp_get = bcm_ptp_hwtstamp_get; priv->mii_ts.ts_info = bcm_ptp_ts_info; priv->phydev->mii_ts = &priv->mii_ts; From 036bb4a5372ee13543d3c67959a8f071e8a4308f Mon Sep 17 00:00:00 2001 From: Vadim Fedorenko Date: Mon, 24 Nov 2025 18:11:48 +0000 Subject: [PATCH 1176/1381] net: phy: dp83640: add HW timestamp configuration reporting The driver stores configuration of TX timestamping and can technically report it. Patch RX timestamp configuration storage to be more precise on reporting and add callback to actually report it. Reviewed-by: Kory Maincent Signed-off-by: Vadim Fedorenko Link: https://patch.msgid.link/20251124181151.277256-5-vadim.fedorenko@linux.dev Signed-off-by: Jakub Kicinski --- drivers/net/phy/dp83640.c | 21 +++++++++++++++++---- 1 file changed, 17 insertions(+), 4 deletions(-) diff --git a/drivers/net/phy/dp83640.c b/drivers/net/phy/dp83640.c index f733a8b72d40..b950acc9c49b 100644 --- a/drivers/net/phy/dp83640.c +++ b/drivers/net/phy/dp83640.c @@ -1176,6 +1176,18 @@ static irqreturn_t dp83640_handle_interrupt(struct phy_device *phydev) return IRQ_HANDLED; } +static int dp83640_hwtstamp_get(struct mii_timestamper *mii_ts, + struct kernel_hwtstamp_config *cfg) +{ + struct dp83640_private *dp83640 = + container_of(mii_ts, struct dp83640_private, mii_ts); + + cfg->rx_filter = dp83640->hwts_rx_en; + cfg->tx_type = dp83640->hwts_tx_en; + + return 0; +} + static int dp83640_hwtstamp_set(struct mii_timestamper *mii_ts, struct kernel_hwtstamp_config *cfg, struct netlink_ext_ack *extack) @@ -1198,7 +1210,7 @@ static int dp83640_hwtstamp_set(struct mii_timestamper *mii_ts, case HWTSTAMP_FILTER_PTP_V1_L4_EVENT: case HWTSTAMP_FILTER_PTP_V1_L4_SYNC: case HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ: - dp83640->hwts_rx_en = 1; + dp83640->hwts_rx_en = HWTSTAMP_FILTER_PTP_V1_L4_EVENT; dp83640->layer = PTP_CLASS_L4; dp83640->version = PTP_CLASS_V1; cfg->rx_filter = HWTSTAMP_FILTER_PTP_V1_L4_EVENT; @@ -1206,7 +1218,7 @@ static int dp83640_hwtstamp_set(struct mii_timestamper *mii_ts, case HWTSTAMP_FILTER_PTP_V2_L4_EVENT: case HWTSTAMP_FILTER_PTP_V2_L4_SYNC: case HWTSTAMP_FILTER_PTP_V2_L4_DELAY_REQ: - dp83640->hwts_rx_en = 1; + dp83640->hwts_rx_en = HWTSTAMP_FILTER_PTP_V2_L4_EVENT; dp83640->layer = PTP_CLASS_L4; dp83640->version = PTP_CLASS_V2; cfg->rx_filter = HWTSTAMP_FILTER_PTP_V2_L4_EVENT; @@ -1214,7 +1226,7 @@ static int dp83640_hwtstamp_set(struct mii_timestamper *mii_ts, case HWTSTAMP_FILTER_PTP_V2_L2_EVENT: case HWTSTAMP_FILTER_PTP_V2_L2_SYNC: case HWTSTAMP_FILTER_PTP_V2_L2_DELAY_REQ: - dp83640->hwts_rx_en = 1; + dp83640->hwts_rx_en = HWTSTAMP_FILTER_PTP_V2_L2_EVENT; dp83640->layer = PTP_CLASS_L2; dp83640->version = PTP_CLASS_V2; cfg->rx_filter = HWTSTAMP_FILTER_PTP_V2_L2_EVENT; @@ -1222,7 +1234,7 @@ static int dp83640_hwtstamp_set(struct mii_timestamper *mii_ts, case HWTSTAMP_FILTER_PTP_V2_EVENT: case HWTSTAMP_FILTER_PTP_V2_SYNC: case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ: - dp83640->hwts_rx_en = 1; + dp83640->hwts_rx_en = HWTSTAMP_FILTER_PTP_V2_EVENT; dp83640->layer = PTP_CLASS_L4 | PTP_CLASS_L2; dp83640->version = PTP_CLASS_V2; cfg->rx_filter = HWTSTAMP_FILTER_PTP_V2_EVENT; @@ -1408,6 +1420,7 @@ static int dp83640_probe(struct phy_device *phydev) dp83640->mii_ts.rxtstamp = dp83640_rxtstamp; dp83640->mii_ts.txtstamp = dp83640_txtstamp; dp83640->mii_ts.hwtstamp_set = dp83640_hwtstamp_set; + dp83640->mii_ts.hwtstamp_get = dp83640_hwtstamp_get; dp83640->mii_ts.ts_info = dp83640_ts_info; INIT_DELAYED_WORK(&dp83640->ts_work, rx_timestamp_work); From ab95392ab5d3000d68806d77e6d7ae32a8d7edec Mon Sep 17 00:00:00 2001 From: Vadim Fedorenko Date: Mon, 24 Nov 2025 18:11:49 +0000 Subject: [PATCH 1177/1381] phy: mscc: add HW timestamp configuration reporting The driver stores HW configuration and can technically report it. Add callback to do it. Reviewed-by: Kory Maincent Signed-off-by: Vadim Fedorenko Link: https://patch.msgid.link/20251124181151.277256-6-vadim.fedorenko@linux.dev Signed-off-by: Jakub Kicinski --- drivers/net/phy/mscc/mscc_ptp.c | 13 +++++++++++++ 1 file changed, 13 insertions(+) diff --git a/drivers/net/phy/mscc/mscc_ptp.c b/drivers/net/phy/mscc/mscc_ptp.c index dc06614222f6..4865eac74b0e 100644 --- a/drivers/net/phy/mscc/mscc_ptp.c +++ b/drivers/net/phy/mscc/mscc_ptp.c @@ -1051,6 +1051,18 @@ static void vsc85xx_ts_reset_fifo(struct phy_device *phydev) val); } +static int vsc85xx_hwtstamp_get(struct mii_timestamper *mii_ts, + struct kernel_hwtstamp_config *cfg) +{ + struct vsc8531_private *vsc8531 = + container_of(mii_ts, struct vsc8531_private, mii_ts); + + cfg->tx_type = vsc8531->ptp->tx_type; + cfg->rx_filter = vsc8531->ptp->rx_filter; + + return 0; +} + static int vsc85xx_hwtstamp_set(struct mii_timestamper *mii_ts, struct kernel_hwtstamp_config *cfg, struct netlink_ext_ack *extack) @@ -1612,6 +1624,7 @@ int vsc8584_ptp_probe(struct phy_device *phydev) vsc8531->mii_ts.rxtstamp = vsc85xx_rxtstamp; vsc8531->mii_ts.txtstamp = vsc85xx_txtstamp; vsc8531->mii_ts.hwtstamp_set = vsc85xx_hwtstamp_set; + vsc8531->mii_ts.hwtstamp_get = vsc85xx_hwtstamp_get; vsc8531->mii_ts.ts_info = vsc85xx_ts_info; phydev->mii_ts = &vsc8531->mii_ts; From d51de60b8edb8f333bdf84677ae273b6731ee52c Mon Sep 17 00:00:00 2001 From: Vadim Fedorenko Date: Mon, 24 Nov 2025 18:11:50 +0000 Subject: [PATCH 1178/1381] net: phy: nxp-c45-tja11xx: add HW timestamp configuration reporting The driver stores HW timestamping configuration and can technically report it. Add callback to do it. Reviewed-by: Kory Maincent Signed-off-by: Vadim Fedorenko Link: https://patch.msgid.link/20251124181151.277256-7-vadim.fedorenko@linux.dev Signed-off-by: Jakub Kicinski --- drivers/net/phy/nxp-c45-tja11xx.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+) diff --git a/drivers/net/phy/nxp-c45-tja11xx.c b/drivers/net/phy/nxp-c45-tja11xx.c index 13a8fac223a9..f526528d2e32 100644 --- a/drivers/net/phy/nxp-c45-tja11xx.c +++ b/drivers/net/phy/nxp-c45-tja11xx.c @@ -1012,6 +1012,19 @@ static bool nxp_c45_rxtstamp(struct mii_timestamper *mii_ts, return true; } +static int nxp_c45_hwtstamp_get(struct mii_timestamper *mii_ts, + struct kernel_hwtstamp_config *cfg) +{ + struct nxp_c45_phy *priv = container_of(mii_ts, struct nxp_c45_phy, + mii_ts); + + cfg->tx_type = priv->hwts_tx; + cfg->rx_filter = priv->hwts_rx ? HWTSTAMP_FILTER_PTP_V2_L2_EVENT + : HWTSTAMP_FILTER_NONE; + + return 0; +} + static int nxp_c45_hwtstamp_set(struct mii_timestamper *mii_ts, struct kernel_hwtstamp_config *cfg, struct netlink_ext_ack *extack) @@ -1750,6 +1763,7 @@ static int nxp_c45_probe(struct phy_device *phydev) priv->mii_ts.rxtstamp = nxp_c45_rxtstamp; priv->mii_ts.txtstamp = nxp_c45_txtstamp; priv->mii_ts.hwtstamp_set = nxp_c45_hwtstamp_set; + priv->mii_ts.hwtstamp_get = nxp_c45_hwtstamp_get; priv->mii_ts.ts_info = nxp_c45_ts_info; phydev->mii_ts = &priv->mii_ts; ret = nxp_c45_init_ptp_clock(priv); From dadc51871d768268ee29580fcdf21ac51b306c21 Mon Sep 17 00:00:00 2001 From: Vadim Fedorenko Date: Mon, 24 Nov 2025 18:11:51 +0000 Subject: [PATCH 1179/1381] ptp: ptp_ines: add HW timestamp configuration reporting The driver partially stores HW timestamping configuration, but missing pieces can be read from HW. Add callback to report configuration. Reviewed-by: Kory Maincent Signed-off-by: Vadim Fedorenko Link: https://patch.msgid.link/20251124181151.277256-8-vadim.fedorenko@linux.dev Signed-off-by: Jakub Kicinski --- drivers/ptp/ptp_ines.c | 23 +++++++++++++++++++++++ 1 file changed, 23 insertions(+) diff --git a/drivers/ptp/ptp_ines.c b/drivers/ptp/ptp_ines.c index 56c798e77f20..790eb42b78db 100644 --- a/drivers/ptp/ptp_ines.c +++ b/drivers/ptp/ptp_ines.c @@ -328,6 +328,28 @@ static u64 ines_find_txts(struct ines_port *port, struct sk_buff *skb) return ns; } +static int ines_hwtstamp_get(struct mii_timestamper *mii_ts, + struct kernel_hwtstamp_config *cfg) +{ + struct ines_port *port = container_of(mii_ts, struct ines_port, mii_ts); + unsigned long flags; + u32 port_conf; + + cfg->rx_filter = port->rxts_enabled ? HWTSTAMP_FILTER_PTP_V2_EVENT + : HWTSTAMP_FILTER_NONE; + if (port->txts_enabled) { + spin_lock_irqsave(&port->lock, flags); + port_conf = ines_read32(port, port_conf); + spin_unlock_irqrestore(&port->lock, flags); + cfg->tx_type = (port_conf & CM_ONE_STEP) ? HWTSTAMP_TX_ONESTEP_P2P + : HWTSTAMP_TX_OFF; + } else { + cfg->tx_type = HWTSTAMP_TX_OFF; + } + + return 0; +} + static int ines_hwtstamp_set(struct mii_timestamper *mii_ts, struct kernel_hwtstamp_config *cfg, struct netlink_ext_ack *extack) @@ -710,6 +732,7 @@ static struct mii_timestamper *ines_ptp_probe_channel(struct device *device, port->mii_ts.rxtstamp = ines_rxtstamp; port->mii_ts.txtstamp = ines_txtstamp; port->mii_ts.hwtstamp_set = ines_hwtstamp_set; + port->mii_ts.hwtstamp_get = ines_hwtstamp_get; port->mii_ts.link_state = ines_link_state; port->mii_ts.ts_info = ines_ts_info; From 4440bf5f2e75a6ed5d451569e317c24e6e222e09 Mon Sep 17 00:00:00 2001 From: Yao Zi Date: Mon, 24 Nov 2025 16:04:15 +0000 Subject: [PATCH 1180/1381] net: stmmac: Add generic suspend/resume helper for PCI-based controllers Most glue driver for PCI-based DWMAC controllers utilize similar platform suspend/resume routines. Add a generic implementation to reduce duplicated code. Signed-off-by: Yao Zi Reviewed-by: Russell King (Oracle) Reviewed-by: Yanteng Si Link: https://patch.msgid.link/20251124160417.51514-2-ziyao@disroot.org Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/Kconfig | 5 ++ drivers/net/ethernet/stmicro/stmmac/Makefile | 1 + .../ethernet/stmicro/stmmac/stmmac_libpci.c | 48 +++++++++++++++++++ .../ethernet/stmicro/stmmac/stmmac_libpci.h | 12 +++++ 4 files changed, 66 insertions(+) create mode 100644 drivers/net/ethernet/stmicro/stmmac/stmmac_libpci.c create mode 100644 drivers/net/ethernet/stmicro/stmmac/stmmac_libpci.h diff --git a/drivers/net/ethernet/stmicro/stmmac/Kconfig b/drivers/net/ethernet/stmicro/stmmac/Kconfig index 87c5bea6c2a2..15ed5c1d071a 100644 --- a/drivers/net/ethernet/stmicro/stmmac/Kconfig +++ b/drivers/net/ethernet/stmicro/stmmac/Kconfig @@ -349,6 +349,11 @@ config DWMAC_VISCONTI endif +config STMMAC_LIBPCI + tristate + help + This option enables the PCI bus helpers for the stmmac driver. + config DWMAC_INTEL tristate "Intel GMAC support" default X86 diff --git a/drivers/net/ethernet/stmicro/stmmac/Makefile b/drivers/net/ethernet/stmicro/stmmac/Makefile index 1681a8a28313..7bf528731034 100644 --- a/drivers/net/ethernet/stmicro/stmmac/Makefile +++ b/drivers/net/ethernet/stmicro/stmmac/Makefile @@ -44,6 +44,7 @@ obj-$(CONFIG_DWMAC_VISCONTI) += dwmac-visconti.o stmmac-platform-objs:= stmmac_platform.o dwmac-altr-socfpga-objs := dwmac-socfpga.o +obj-$(CONFIG_STMMAC_LIBPCI) += stmmac_libpci.o obj-$(CONFIG_STMMAC_PCI) += stmmac-pci.o obj-$(CONFIG_DWMAC_INTEL) += dwmac-intel.o obj-$(CONFIG_DWMAC_LOONGSON) += dwmac-loongson.o diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_libpci.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_libpci.c new file mode 100644 index 000000000000..5c5dd502f79a --- /dev/null +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_libpci.c @@ -0,0 +1,48 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * PCI bus helpers for STMMAC driver + * Copyright (C) 2025 Yao Zi + */ + +#include +#include + +#include "stmmac_libpci.h" + +int stmmac_pci_plat_suspend(struct device *dev, void *bsp_priv) +{ + struct pci_dev *pdev = to_pci_dev(dev); + int ret; + + ret = pci_save_state(pdev); + if (ret) + return ret; + + pci_disable_device(pdev); + pci_wake_from_d3(pdev, true); + + return 0; +} +EXPORT_SYMBOL_GPL(stmmac_pci_plat_suspend); + +int stmmac_pci_plat_resume(struct device *dev, void *bsp_priv) +{ + struct pci_dev *pdev = to_pci_dev(dev); + int ret; + + pci_restore_state(pdev); + pci_set_power_state(pdev, PCI_D0); + + ret = pci_enable_device(pdev); + if (ret) + return ret; + + pci_set_master(pdev); + + return 0; +} +EXPORT_SYMBOL_GPL(stmmac_pci_plat_resume); + +MODULE_DESCRIPTION("STMMAC PCI helper library"); +MODULE_AUTHOR("Yao Zi "); +MODULE_LICENSE("GPL"); diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_libpci.h b/drivers/net/ethernet/stmicro/stmmac/stmmac_libpci.h new file mode 100644 index 000000000000..71553184f982 --- /dev/null +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_libpci.h @@ -0,0 +1,12 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (C) 2025 Yao Zi + */ + +#ifndef __STMMAC_LIBPCI_H__ +#define __STMMAC_LIBPCI_H__ + +int stmmac_pci_plat_suspend(struct device *dev, void *bsp_priv); +int stmmac_pci_plat_resume(struct device *dev, void *bsp_priv); + +#endif /* __STMMAC_LIBPCI_H__ */ From c4064af1c7e37d81d9556057e8d0295f0be9eef4 Mon Sep 17 00:00:00 2001 From: Yao Zi Date: Mon, 24 Nov 2025 16:04:16 +0000 Subject: [PATCH 1181/1381] net: stmmac: loongson: Use generic PCI suspend/resume routines Convert glue driver for Loongson DWMAC controller to use the generic platform suspend/resume routines for PCI controllers, instead of implementing its own one. Signed-off-by: Yao Zi Reviewed-by: Russell King (Oracle) Acked-by: Yanteng Si Link: https://patch.msgid.link/20251124160417.51514-3-ziyao@disroot.org Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/Kconfig | 3 +- .../ethernet/stmicro/stmmac/dwmac-loongson.c | 36 ++----------------- 2 files changed, 5 insertions(+), 34 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/Kconfig b/drivers/net/ethernet/stmicro/stmmac/Kconfig index 15ed5c1d071a..ad7ae3618099 100644 --- a/drivers/net/ethernet/stmicro/stmmac/Kconfig +++ b/drivers/net/ethernet/stmicro/stmmac/Kconfig @@ -367,8 +367,9 @@ config DWMAC_INTEL config DWMAC_LOONGSON tristate "Loongson PCI DWMAC support" default MACH_LOONGSON64 - depends on (MACH_LOONGSON64 || COMPILE_TEST) && STMMAC_ETH && PCI + depends on (MACH_LOONGSON64 || COMPILE_TEST) && PCI depends on COMMON_CLK + select STMMAC_LIBPCI help This selects the LOONGSON PCI bus support for the stmmac driver, Support for ethernet controller on Loongson-2K1000 SoC and LS7A1000 bridge. diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c index 5f9472f47e35..107a7c84ace8 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c @@ -8,6 +8,7 @@ #include #include #include "stmmac.h" +#include "stmmac_libpci.h" #include "dwmac_dma.h" #include "dwmac1000.h" @@ -502,37 +503,6 @@ static int loongson_dwmac_fix_reset(struct stmmac_priv *priv, void __iomem *ioad 10000, 2000000); } -static int loongson_dwmac_suspend(struct device *dev, void *bsp_priv) -{ - struct pci_dev *pdev = to_pci_dev(dev); - int ret; - - ret = pci_save_state(pdev); - if (ret) - return ret; - - pci_disable_device(pdev); - pci_wake_from_d3(pdev, true); - return 0; -} - -static int loongson_dwmac_resume(struct device *dev, void *bsp_priv) -{ - struct pci_dev *pdev = to_pci_dev(dev); - int ret; - - pci_restore_state(pdev); - pci_set_power_state(pdev, PCI_D0); - - ret = pci_enable_device(pdev); - if (ret) - return ret; - - pci_set_master(pdev); - - return 0; -} - static int loongson_dwmac_probe(struct pci_dev *pdev, const struct pci_device_id *id) { struct plat_stmmacenet_data *plat; @@ -577,8 +547,8 @@ static int loongson_dwmac_probe(struct pci_dev *pdev, const struct pci_device_id plat->bsp_priv = ld; plat->mac_setup = loongson_dwmac_setup; plat->fix_soc_reset = loongson_dwmac_fix_reset; - plat->suspend = loongson_dwmac_suspend; - plat->resume = loongson_dwmac_resume; + plat->suspend = stmmac_pci_plat_suspend; + plat->resume = stmmac_pci_plat_resume; ld->dev = &pdev->dev; ld->loongson_id = readl(res.addr + GMAC_VERSION) & 0xff; From b35e94edf2290ee72ddbbacc29adc2b7eab7d7f3 Mon Sep 17 00:00:00 2001 From: Yao Zi Date: Mon, 24 Nov 2025 16:04:17 +0000 Subject: [PATCH 1182/1381] net: stmmac: pci: Use generic PCI suspend/resume routines Convert STMMAC PCI glue driver to use the generic platform suspend/resume routines for PCI controllers, instead of implementing its own one. Signed-off-by: Yao Zi Reviewed-by: Russell King (Oracle) Reviewed-by: Yanteng Si Link: https://patch.msgid.link/20251124160417.51514-4-ziyao@disroot.org Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/Kconfig | 3 +- .../net/ethernet/stmicro/stmmac/stmmac_pci.c | 36 ++----------------- 2 files changed, 5 insertions(+), 34 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/Kconfig b/drivers/net/ethernet/stmicro/stmmac/Kconfig index ad7ae3618099..907fe2e927f0 100644 --- a/drivers/net/ethernet/stmicro/stmmac/Kconfig +++ b/drivers/net/ethernet/stmicro/stmmac/Kconfig @@ -376,8 +376,9 @@ config DWMAC_LOONGSON config STMMAC_PCI tristate "STMMAC PCI bus support" - depends on STMMAC_ETH && PCI + depends on PCI depends on COMMON_CLK + select STMMAC_LIBPCI help This selects the platform specific bus support for the stmmac driver. This driver was tested on XLINX XC2V3000 FF1152AMT0221 diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c index afb1c53ca6f8..270ad066ced3 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c @@ -14,6 +14,7 @@ #include #include "stmmac.h" +#include "stmmac_libpci.h" struct stmmac_pci_info { int (*setup)(struct pci_dev *pdev, struct plat_stmmacenet_data *plat); @@ -102,37 +103,6 @@ static const struct stmmac_pci_info snps_gmac5_pci_info = { .setup = snps_gmac5_default_data, }; -static int stmmac_pci_suspend(struct device *dev, void *bsp_priv) -{ - struct pci_dev *pdev = to_pci_dev(dev); - int ret; - - ret = pci_save_state(pdev); - if (ret) - return ret; - - pci_disable_device(pdev); - pci_wake_from_d3(pdev, true); - return 0; -} - -static int stmmac_pci_resume(struct device *dev, void *bsp_priv) -{ - struct pci_dev *pdev = to_pci_dev(dev); - int ret; - - pci_restore_state(pdev); - pci_set_power_state(pdev, PCI_D0); - - ret = pci_enable_device(pdev); - if (ret) - return ret; - - pci_set_master(pdev); - - return 0; -} - /** * stmmac_pci_probe * @@ -212,8 +182,8 @@ static int stmmac_pci_probe(struct pci_dev *pdev, plat->safety_feat_cfg->prtyen = 1; plat->safety_feat_cfg->tmouten = 1; - plat->suspend = stmmac_pci_suspend; - plat->resume = stmmac_pci_resume; + plat->suspend = stmmac_pci_plat_suspend; + plat->resume = stmmac_pci_plat_resume; return stmmac_dvr_probe(&pdev->dev, plat, &res); } From a8acbcbaf6d0c33555cc043ce3f006aaec0ccb3f Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Tue, 25 Nov 2025 02:19:44 -0800 Subject: [PATCH 1183/1381] i40e: extract GRXRINGS from .get_rxnfc Commit 84eaf4359c36 ("net: ethtool: add get_rx_ring_count callback to optimize RX ring queries") added specific support for GRXRINGS callback, simplifying .get_rxnfc. Remove the handling of GRXRINGS in .get_rxnfc() by moving it to the new .get_rx_ring_count(). This simplifies the RX ring count retrieval and aligns i40e with the new ethtool API for querying RX ring parameters. Signed-off-by: Breno Leitao Signed-off-by: Aleksandr Loktionov Link: https://patch.msgid.link/20251125-gxring_intel-v2-1-f55cd022d28b@debian.org Signed-off-by: Jakub Kicinski --- .../net/ethernet/intel/i40e/i40e_ethtool.c | 19 +++++++++++++++---- 1 file changed, 15 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c index 86c72596617a..f2c2646ea298 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c +++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c @@ -3521,6 +3521,20 @@ static int i40e_get_ethtool_fdir_entry(struct i40e_pf *pf, return 0; } +/** + * i40e_get_rx_ring_count - get RX ring count + * @netdev: network interface device structure + * + * Return: number of RX rings. + **/ +static u32 i40e_get_rx_ring_count(struct net_device *netdev) +{ + struct i40e_netdev_priv *np = netdev_priv(netdev); + struct i40e_vsi *vsi = np->vsi; + + return vsi->rss_size; +} + /** * i40e_get_rxnfc - command to get RX flow classification rules * @netdev: network interface device structure @@ -3538,10 +3552,6 @@ static int i40e_get_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *cmd, int ret = -EOPNOTSUPP; switch (cmd->cmd) { - case ETHTOOL_GRXRINGS: - cmd->data = vsi->rss_size; - ret = 0; - break; case ETHTOOL_GRXCLSRLCNT: cmd->rule_cnt = pf->fdir_pf_active_filters; /* report total rule count */ @@ -5819,6 +5829,7 @@ static const struct ethtool_ops i40e_ethtool_ops = { .set_msglevel = i40e_set_msglevel, .get_rxnfc = i40e_get_rxnfc, .set_rxnfc = i40e_set_rxnfc, + .get_rx_ring_count = i40e_get_rx_ring_count, .self_test = i40e_diag_test, .get_strings = i40e_get_strings, .get_eee = i40e_get_eee, From fe0a3d7d1dca4b8d37710becf10ffd7fcfe2741e Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Tue, 25 Nov 2025 02:19:45 -0800 Subject: [PATCH 1184/1381] iavf: extract GRXRINGS from .get_rxnfc Commit 84eaf4359c36 ("net: ethtool: add get_rx_ring_count callback to optimize RX ring queries") added specific support for GRXRINGS callback, simplifying .get_rxnfc. Remove the handling of GRXRINGS in .get_rxnfc() by moving it to the new .get_rx_ring_count(). This simplifies the RX ring count retrieval and aligns iavf with the new ethtool API for querying RX ring parameters. Signed-off-by: Breno Leitao Signed-off-by: Aleksandr Loktionov Link: https://patch.msgid.link/20251125-gxring_intel-v2-2-f55cd022d28b@debian.org Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/intel/iavf/iavf_ethtool.c | 18 ++++++++++++++---- 1 file changed, 14 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/intel/iavf/iavf_ethtool.c b/drivers/net/ethernet/intel/iavf/iavf_ethtool.c index a3f8ced23266..2cc21289a707 100644 --- a/drivers/net/ethernet/intel/iavf/iavf_ethtool.c +++ b/drivers/net/ethernet/intel/iavf/iavf_ethtool.c @@ -1638,6 +1638,19 @@ static int iavf_set_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *cmd) return ret; } +/** + * iavf_get_rx_ring_count - get RX ring count + * @netdev: network interface device structure + * + * Return: number of RX rings. + **/ +static u32 iavf_get_rx_ring_count(struct net_device *netdev) +{ + struct iavf_adapter *adapter = netdev_priv(netdev); + + return adapter->num_active_queues; +} + /** * iavf_get_rxnfc - command to get RX flow classification rules * @netdev: network interface device structure @@ -1653,10 +1666,6 @@ static int iavf_get_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *cmd, int ret = -EOPNOTSUPP; switch (cmd->cmd) { - case ETHTOOL_GRXRINGS: - cmd->data = adapter->num_active_queues; - ret = 0; - break; case ETHTOOL_GRXCLSRLCNT: if (!(adapter->flags & IAVF_FLAG_FDIR_ENABLED)) break; @@ -1866,6 +1875,7 @@ static const struct ethtool_ops iavf_ethtool_ops = { .set_per_queue_coalesce = iavf_set_per_queue_coalesce, .set_rxnfc = iavf_set_rxnfc, .get_rxnfc = iavf_get_rxnfc, + .get_rx_ring_count = iavf_get_rx_ring_count, .get_rxfh_indir_size = iavf_get_rxfh_indir_size, .get_rxfh = iavf_get_rxfh, .set_rxfh = iavf_set_rxfh, From 8e8c00e1d2130c7d174b799cd5ebc4b41f325aa8 Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Tue, 25 Nov 2025 02:19:46 -0800 Subject: [PATCH 1185/1381] ice: extract GRXRINGS from .get_rxnfc Commit 84eaf4359c36 ("net: ethtool: add get_rx_ring_count callback to optimize RX ring queries") added specific support for GRXRINGS callback, simplifying .get_rxnfc. Remove the handling of GRXRINGS in .get_rxnfc() by moving it to the new .get_rx_ring_count(). This simplifies the RX ring count retrieval and aligns ice with the new ethtool API for querying RX ring parameters. Signed-off-by: Breno Leitao Signed-off-by: Aleksandr Loktionov Link: https://patch.msgid.link/20251125-gxring_intel-v2-3-f55cd022d28b@debian.org Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/intel/ice/ice_ethtool.c | 19 +++++++++++++++---- 1 file changed, 15 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/intel/ice/ice_ethtool.c b/drivers/net/ethernet/intel/ice/ice_ethtool.c index a1d9abee97e5..969d4f8f9c02 100644 --- a/drivers/net/ethernet/intel/ice/ice_ethtool.c +++ b/drivers/net/ethernet/intel/ice/ice_ethtool.c @@ -3083,6 +3083,20 @@ static int ice_set_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *cmd) return -EOPNOTSUPP; } +/** + * ice_get_rx_ring_count - get RX ring count + * @netdev: network interface device structure + * + * Return: number of RX rings. + */ +static u32 ice_get_rx_ring_count(struct net_device *netdev) +{ + struct ice_netdev_priv *np = netdev_priv(netdev); + struct ice_vsi *vsi = np->vsi; + + return vsi->rss_size; +} + /** * ice_get_rxnfc - command to get Rx flow classification rules * @netdev: network interface device structure @@ -3103,10 +3117,6 @@ ice_get_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *cmd, hw = &vsi->back->hw; switch (cmd->cmd) { - case ETHTOOL_GRXRINGS: - cmd->data = vsi->rss_size; - ret = 0; - break; case ETHTOOL_GRXCLSRLCNT: cmd->rule_cnt = hw->fdir_active_fltr; /* report total rule count */ @@ -4853,6 +4863,7 @@ static const struct ethtool_ops ice_ethtool_ops = { .get_sset_count = ice_get_sset_count, .get_rxnfc = ice_get_rxnfc, .set_rxnfc = ice_set_rxnfc, + .get_rx_ring_count = ice_get_rx_ring_count, .get_ringparam = ice_get_ringparam, .set_ringparam = ice_set_ringparam, .nway_reset = ice_nway_reset, From 873a1942fbc622694ef21b43eb567bdcd163ab86 Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Tue, 25 Nov 2025 02:19:47 -0800 Subject: [PATCH 1186/1381] idpf: extract GRXRINGS from .get_rxnfc Commit 84eaf4359c36 ("net: ethtool: add get_rx_ring_count callback to optimize RX ring queries") added specific support for GRXRINGS callback, simplifying .get_rxnfc. Remove the handling of GRXRINGS in .get_rxnfc() by moving it to the new .get_rx_ring_count(). This simplifies the RX ring count retrieval and aligns idpf with the new ethtool API for querying RX ring parameters. I was not totally convinced I needed to have the lock, but, I decided to be on the safe side and get the exact same behaviour it was before. Signed-off-by: Breno Leitao Signed-off-by: Aleksandr Loktionov Link: https://patch.msgid.link/20251125-gxring_intel-v2-4-f55cd022d28b@debian.org Signed-off-by: Jakub Kicinski --- .../net/ethernet/intel/idpf/idpf_ethtool.c | 23 ++++++++++++++++--- 1 file changed, 20 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/intel/idpf/idpf_ethtool.c b/drivers/net/ethernet/intel/idpf/idpf_ethtool.c index a5a1eec9ade8..5dd82b1349c5 100644 --- a/drivers/net/ethernet/intel/idpf/idpf_ethtool.c +++ b/drivers/net/ethernet/intel/idpf/idpf_ethtool.c @@ -5,6 +5,25 @@ #include "idpf_ptp.h" #include "idpf_virtchnl.h" +/** + * idpf_get_rx_ring_count - get RX ring count + * @netdev: network interface device structure + * + * Return: number of RX rings. + */ +static u32 idpf_get_rx_ring_count(struct net_device *netdev) +{ + struct idpf_vport *vport; + u32 num_rxq; + + idpf_vport_ctrl_lock(netdev); + vport = idpf_netdev_to_vport(netdev); + num_rxq = vport->num_rxq; + idpf_vport_ctrl_unlock(netdev); + + return num_rxq; +} + /** * idpf_get_rxnfc - command to get RX flow classification rules * @netdev: network interface device structure @@ -28,9 +47,6 @@ static int idpf_get_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *cmd, user_config = &np->adapter->vport_config[np->vport_idx]->user_config; switch (cmd->cmd) { - case ETHTOOL_GRXRINGS: - cmd->data = vport->num_rxq; - break; case ETHTOOL_GRXCLSRLCNT: cmd->rule_cnt = user_config->num_fsteer_fltrs; cmd->data = idpf_fsteer_max_rules(vport); @@ -1757,6 +1773,7 @@ static const struct ethtool_ops idpf_ethtool_ops = { .get_channels = idpf_get_channels, .get_rxnfc = idpf_get_rxnfc, .set_rxnfc = idpf_set_rxnfc, + .get_rx_ring_count = idpf_get_rx_ring_count, .get_rxfh_key_size = idpf_get_rxfh_key_size, .get_rxfh_indir_size = idpf_get_rxfh_indir_size, .get_rxfh = idpf_get_rxfh, From d6c744f46816994d9e8b7e9f37065caba6664e86 Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Tue, 25 Nov 2025 02:19:48 -0800 Subject: [PATCH 1187/1381] igb: extract GRXRINGS from .get_rxnfc Commit 84eaf4359c36 ("net: ethtool: add get_rx_ring_count callback to optimize RX ring queries") added specific support for GRXRINGS callback, simplifying .get_rxnfc. Remove the handling of GRXRINGS in .get_rxnfc() by moving it to the new .get_rx_ring_count(). This simplifies the RX ring count retrieval and aligns igb with the new ethtool API for querying RX ring parameters. Signed-off-by: Breno Leitao Signed-off-by: Aleksandr Loktionov Link: https://patch.msgid.link/20251125-gxring_intel-v2-5-f55cd022d28b@debian.org Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/intel/igb/igb_ethtool.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/intel/igb/igb_ethtool.c b/drivers/net/ethernet/intel/igb/igb_ethtool.c index 10e2445e0ded..b507576b28b2 100644 --- a/drivers/net/ethernet/intel/igb/igb_ethtool.c +++ b/drivers/net/ethernet/intel/igb/igb_ethtool.c @@ -2541,6 +2541,13 @@ static int igb_get_rxfh_fields(struct net_device *dev, return 0; } +static u32 igb_get_rx_ring_count(struct net_device *dev) +{ + struct igb_adapter *adapter = netdev_priv(dev); + + return adapter->num_rx_queues; +} + static int igb_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd, u32 *rule_locs) { @@ -2548,10 +2555,6 @@ static int igb_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd, int ret = -EOPNOTSUPP; switch (cmd->cmd) { - case ETHTOOL_GRXRINGS: - cmd->data = adapter->num_rx_queues; - ret = 0; - break; case ETHTOOL_GRXCLSRLCNT: cmd->rule_cnt = adapter->nfc_filter_count; ret = 0; @@ -3473,6 +3476,7 @@ static const struct ethtool_ops igb_ethtool_ops = { .get_ts_info = igb_get_ts_info, .get_rxnfc = igb_get_rxnfc, .set_rxnfc = igb_set_rxnfc, + .get_rx_ring_count = igb_get_rx_ring_count, .get_eee = igb_get_eee, .set_eee = igb_set_eee, .get_module_info = igb_get_module_info, From 768ce58dddb26c591a2b77a659f779e730f1796c Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Tue, 25 Nov 2025 02:19:49 -0800 Subject: [PATCH 1188/1381] igc: extract GRXRINGS from .get_rxnfc Commit 84eaf4359c36 ("net: ethtool: add get_rx_ring_count callback to optimize RX ring queries") added specific support for GRXRINGS callback, simplifying .get_rxnfc. Remove the handling of GRXRINGS in .get_rxnfc() by moving it to the new .get_rx_ring_count(). This simplifies the RX ring count retrieval and aligns igc with the new ethtool API for querying RX ring parameters. Signed-off-by: Breno Leitao Signed-off-by: Aleksandr Loktionov Link: https://patch.msgid.link/20251125-gxring_intel-v2-6-f55cd022d28b@debian.org Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/intel/igc/igc_ethtool.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/intel/igc/igc_ethtool.c b/drivers/net/ethernet/intel/igc/igc_ethtool.c index bb783042d1af..e94c1922b97a 100644 --- a/drivers/net/ethernet/intel/igc/igc_ethtool.c +++ b/drivers/net/ethernet/intel/igc/igc_ethtool.c @@ -1091,15 +1091,19 @@ static int igc_ethtool_get_rxfh_fields(struct net_device *dev, return 0; } +static u32 igc_ethtool_get_rx_ring_count(struct net_device *dev) +{ + struct igc_adapter *adapter = netdev_priv(dev); + + return adapter->num_rx_queues; +} + static int igc_ethtool_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd, u32 *rule_locs) { struct igc_adapter *adapter = netdev_priv(dev); switch (cmd->cmd) { - case ETHTOOL_GRXRINGS: - cmd->data = adapter->num_rx_queues; - return 0; case ETHTOOL_GRXCLSRLCNT: cmd->rule_cnt = adapter->nfc_rule_count; return 0; @@ -2170,6 +2174,7 @@ static const struct ethtool_ops igc_ethtool_ops = { .set_coalesce = igc_ethtool_set_coalesce, .get_rxnfc = igc_ethtool_get_rxnfc, .set_rxnfc = igc_ethtool_set_rxnfc, + .get_rx_ring_count = igc_ethtool_get_rx_ring_count, .get_rxfh_indir_size = igc_ethtool_get_rxfh_indir_size, .get_rxfh = igc_ethtool_get_rxfh, .set_rxfh = igc_ethtool_set_rxfh, From 3399fd519dd4f8fd8451cf18eb7f2eef0b47e39a Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Tue, 25 Nov 2025 02:19:50 -0800 Subject: [PATCH 1189/1381] ixgbevf: extract GRXRINGS from .get_rxnfc Commit 84eaf4359c36 ("net: ethtool: add get_rx_ring_count callback to optimize RX ring queries") added specific support for GRXRINGS callback, simplifying .get_rxnfc. Remove the handling of GRXRINGS in .get_rxnfc() by moving it to the new .get_rx_ring_count(). This simplifies the RX ring count retrieval and aligns ixgbevf with the new ethtool API for querying RX ring parameters. Signed-off-by: Breno Leitao Signed-off-by: Aleksandr Loktionov Link: https://patch.msgid.link/20251125-gxring_intel-v2-7-f55cd022d28b@debian.org Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/intel/ixgbevf/ethtool.c | 14 +++----------- 1 file changed, 3 insertions(+), 11 deletions(-) diff --git a/drivers/net/ethernet/intel/ixgbevf/ethtool.c b/drivers/net/ethernet/intel/ixgbevf/ethtool.c index bebad564188e..537a60d5276f 100644 --- a/drivers/net/ethernet/intel/ixgbevf/ethtool.c +++ b/drivers/net/ethernet/intel/ixgbevf/ethtool.c @@ -867,19 +867,11 @@ static int ixgbevf_set_coalesce(struct net_device *netdev, return 0; } -static int ixgbevf_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *info, - u32 *rules __always_unused) +static u32 ixgbevf_get_rx_ring_count(struct net_device *dev) { struct ixgbevf_adapter *adapter = netdev_priv(dev); - switch (info->cmd) { - case ETHTOOL_GRXRINGS: - info->data = adapter->num_rx_queues; - return 0; - default: - hw_dbg(&adapter->hw, "Command parameters not supported\n"); - return -EOPNOTSUPP; - } + return adapter->num_rx_queues; } static u32 ixgbevf_get_rxfh_indir_size(struct net_device *netdev) @@ -987,7 +979,7 @@ static const struct ethtool_ops ixgbevf_ethtool_ops = { .get_ethtool_stats = ixgbevf_get_ethtool_stats, .get_coalesce = ixgbevf_get_coalesce, .set_coalesce = ixgbevf_set_coalesce, - .get_rxnfc = ixgbevf_get_rxnfc, + .get_rx_ring_count = ixgbevf_get_rx_ring_count, .get_rxfh_indir_size = ixgbevf_get_rxfh_indir_size, .get_rxfh_key_size = ixgbevf_get_rxfh_key_size, .get_rxfh = ixgbevf_get_rxfh, From 73d834cd17741839a3bb2a753cc83dfd3fdda13a Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Tue, 25 Nov 2025 02:19:51 -0800 Subject: [PATCH 1190/1381] fm10k: extract GRXRINGS from .get_rxnfc Commit 84eaf4359c36 ("net: ethtool: add get_rx_ring_count callback to optimize RX ring queries") added specific support for GRXRINGS callback, simplifying .get_rxnfc. Remove the handling of GRXRINGS in .get_rxnfc() by moving it to the new .get_rx_ring_count(). This simplifies the RX ring count retrieval and aligns fm10k with the new ethtool API for querying RX ring parameters. Signed-off-by: Breno Leitao Signed-off-by: Aleksandr Loktionov Link: https://patch.msgid.link/20251125-gxring_intel-v2-8-f55cd022d28b@debian.org Signed-off-by: Jakub Kicinski --- .../net/ethernet/intel/fm10k/fm10k_ethtool.c | 17 +++-------------- 1 file changed, 3 insertions(+), 14 deletions(-) diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_ethtool.c b/drivers/net/ethernet/intel/fm10k/fm10k_ethtool.c index bf2029144c1d..76e42abca965 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_ethtool.c +++ b/drivers/net/ethernet/intel/fm10k/fm10k_ethtool.c @@ -734,22 +734,11 @@ static int fm10k_get_rssh_fields(struct net_device *dev, return 0; } -static int fm10k_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd, - u32 __always_unused *rule_locs) +static u32 fm10k_get_rx_ring_count(struct net_device *dev) { struct fm10k_intfc *interface = netdev_priv(dev); - int ret = -EOPNOTSUPP; - switch (cmd->cmd) { - case ETHTOOL_GRXRINGS: - cmd->data = interface->num_rx_queues; - ret = 0; - break; - default: - break; - } - - return ret; + return interface->num_rx_queues; } static int fm10k_set_rssh_fields(struct net_device *dev, @@ -1160,7 +1149,7 @@ static const struct ethtool_ops fm10k_ethtool_ops = { .set_ringparam = fm10k_set_ringparam, .get_coalesce = fm10k_get_coalesce, .set_coalesce = fm10k_set_coalesce, - .get_rxnfc = fm10k_get_rxnfc, + .get_rx_ring_count = fm10k_get_rx_ring_count, .get_regs = fm10k_get_regs, .get_regs_len = fm10k_get_regs_len, .self_test = fm10k_self_test, From 651765e8d527427e1d91fb7f606c5506f437f622 Mon Sep 17 00:00:00 2001 From: Hangbin Liu Date: Tue, 25 Nov 2025 11:20:48 +0000 Subject: [PATCH 1191/1381] netlink: specs: add big-endian byte-order for u32 IPv4 addresses MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The fix commit converted several IPv4 address attributes from binary to u32, but forgot to specify byte-order: big-endian. Without this, YNL tools display IPv4 addresses incorrectly due to host-endian interpretation. Add the missing byte-order: big-endian to all affected u32 IPv4 address fields to ensure correct parsing and display. Fixes: 1064d521d177 ("netlink: specs: support ipv4-or-v6 for dual-stack fields") Reported-by: Paolo Abeni Signed-off-by: Hangbin Liu Reviewed-by: Asbjørn Sloth Tønnesen Link: https://patch.msgid.link/20251125112048.37631-1-liuhangbin@gmail.com Signed-off-by: Jakub Kicinski --- Documentation/netlink/specs/rt-addr.yaml | 1 + Documentation/netlink/specs/rt-link.yaml | 2 ++ 2 files changed, 3 insertions(+) diff --git a/Documentation/netlink/specs/rt-addr.yaml b/Documentation/netlink/specs/rt-addr.yaml index abcbaa73fa9d..163a106c41bb 100644 --- a/Documentation/netlink/specs/rt-addr.yaml +++ b/Documentation/netlink/specs/rt-addr.yaml @@ -97,6 +97,7 @@ attribute-sets: - name: broadcast type: u32 + byte-order: big-endian display-hint: ipv4 - name: anycast diff --git a/Documentation/netlink/specs/rt-link.yaml b/Documentation/netlink/specs/rt-link.yaml index ca22c68ca691..6beeb6ee5adf 100644 --- a/Documentation/netlink/specs/rt-link.yaml +++ b/Documentation/netlink/specs/rt-link.yaml @@ -1869,6 +1869,7 @@ attribute-sets: - name: remote type: u32 + byte-order: big-endian display-hint: ipv4 - name: ttl @@ -1987,6 +1988,7 @@ attribute-sets: - name: 6rd-relay-prefix type: u32 + byte-order: big-endian display-hint: ipv4 - name: 6rd-prefixlen From 308b7dee3e5c767e88bbecceff3883c8b17c55b6 Mon Sep 17 00:00:00 2001 From: Hangbin Liu Date: Mon, 24 Nov 2025 02:20:55 +0000 Subject: [PATCH 1192/1381] tools: ynl: add YNL test framework Add a test framework for YAML Netlink (YNL) tools, covering both CLI and ethtool functionality. The framework includes: 1) cli: family listing, netdev, ethtool, rt-* families, and nlctrl operations 2) ethtool: device info, statistics, ring/coalesce/pause parameters, and feature gettings The current YNL syntax is a bit obscure, and end users may not always know how to use it. This test framework provides usage examples and also serves as a regression test to catch potential breakages caused by future changes. Reviewed-by: Donald Hunter Acked-by: Matthieu Baerts (NGI0) Signed-off-by: Hangbin Liu Link: https://patch.msgid.link/20251124022055.33389-1-liuhangbin@gmail.com Signed-off-by: Jakub Kicinski --- tools/net/ynl/Makefile | 8 +- tools/net/ynl/tests/Makefile | 32 +++ tools/net/ynl/tests/config | 6 + tools/net/ynl/tests/test_ynl_cli.sh | 327 ++++++++++++++++++++++++ tools/net/ynl/tests/test_ynl_ethtool.sh | 222 ++++++++++++++++ 5 files changed, 593 insertions(+), 2 deletions(-) create mode 100644 tools/net/ynl/tests/Makefile create mode 100644 tools/net/ynl/tests/config create mode 100755 tools/net/ynl/tests/test_ynl_cli.sh create mode 100755 tools/net/ynl/tests/test_ynl_ethtool.sh diff --git a/tools/net/ynl/Makefile b/tools/net/ynl/Makefile index 31ed20c0f3f8..a40591e513b7 100644 --- a/tools/net/ynl/Makefile +++ b/tools/net/ynl/Makefile @@ -12,7 +12,7 @@ endif libdir ?= $(prefix)/$(libdir_relative) includedir ?= $(prefix)/include -SUBDIRS = lib generated samples ynltool +SUBDIRS = lib generated samples ynltool tests all: $(SUBDIRS) libynl.a @@ -49,5 +49,9 @@ install: libynl.a lib/*.h @echo -e "\tINSTALL pyynl" @pip install --prefix=$(DESTDIR)$(prefix) . @make -C generated install + @make -C tests install -.PHONY: all clean distclean install $(SUBDIRS) +run_tests: + @$(MAKE) -C tests run_tests + +.PHONY: all clean distclean install run_tests $(SUBDIRS) diff --git a/tools/net/ynl/tests/Makefile b/tools/net/ynl/tests/Makefile new file mode 100644 index 000000000000..c1df2e001255 --- /dev/null +++ b/tools/net/ynl/tests/Makefile @@ -0,0 +1,32 @@ +# SPDX-License-Identifier: GPL-2.0 +# Makefile for YNL tests + +TESTS := \ + test_ynl_cli.sh \ + test_ynl_ethtool.sh \ +# end of TESTS + +all: $(TESTS) + +run_tests: + @for test in $(TESTS); do \ + ./$$test; \ + done + +install: $(TESTS) + @mkdir -p $(DESTDIR)/usr/bin + @mkdir -p $(DESTDIR)/usr/share/kselftest + @cp ../../../testing/selftests/kselftest/ktap_helpers.sh $(DESTDIR)/usr/share/kselftest/ + @for test in $(TESTS); do \ + name=$$(basename $$test .sh); \ + sed -e 's|^ynl=.*|ynl="ynl"|' \ + -e 's|^ynl_ethtool=.*|ynl_ethtool="ynl-ethtool"|' \ + -e 's|KSELFTEST_KTAP_HELPERS=.*|KSELFTEST_KTAP_HELPERS="/usr/share/kselftest/ktap_helpers.sh"|' \ + $$test > $(DESTDIR)/usr/bin/$$name; \ + chmod +x $(DESTDIR)/usr/bin/$$name; \ + done + +clean distclean: + @# Nothing to clean + +.PHONY: all install clean run_tests diff --git a/tools/net/ynl/tests/config b/tools/net/ynl/tests/config new file mode 100644 index 000000000000..339f1309c03f --- /dev/null +++ b/tools/net/ynl/tests/config @@ -0,0 +1,6 @@ +CONFIG_DUMMY=m +CONFIG_INET_DIAG=y +CONFIG_IPV6=y +CONFIG_NET_NS=y +CONFIG_NETDEVSIM=m +CONFIG_VETH=m diff --git a/tools/net/ynl/tests/test_ynl_cli.sh b/tools/net/ynl/tests/test_ynl_cli.sh new file mode 100755 index 000000000000..7c0722a08117 --- /dev/null +++ b/tools/net/ynl/tests/test_ynl_cli.sh @@ -0,0 +1,327 @@ +#!/bin/bash +# SPDX-License-Identifier: GPL-2.0 +# Test YNL CLI functionality + +# Load KTAP test helpers +KSELFTEST_KTAP_HELPERS="$(dirname "$(realpath "$0")")/../../../testing/selftests/kselftest/ktap_helpers.sh" +# shellcheck source=../../../testing/selftests/kselftest/ktap_helpers.sh +source "$KSELFTEST_KTAP_HELPERS" + +# Default ynl path for direct execution, can be overridden by make install +ynl="../pyynl/cli.py" + +readonly NSIM_ID="1338" +readonly NSIM_DEV_NAME="nsim${NSIM_ID}" +readonly VETH_A="veth_a" +readonly VETH_B="veth_b" + +testns="ynl-$(mktemp -u XXXXXX)" +TESTS_NO=0 + +# Test listing available families +cli_list_families() +{ + if $ynl --list-families &>/dev/null; then + ktap_test_pass "YNL CLI list families" + else + ktap_test_fail "YNL CLI list families" + fi +} +TESTS_NO=$((TESTS_NO + 1)) + +# Test netdev family operations (dev-get, queue-get) +cli_netdev_ops() +{ + local dev_output + local ifindex + + ifindex=$(ip netns exec "$testns" cat /sys/class/net/"$NSIM_DEV_NAME"/ifindex 2>/dev/null) + + dev_output=$(ip netns exec "$testns" $ynl --family netdev \ + --do dev-get --json "{\"ifindex\": $ifindex}" 2>/dev/null) + + if ! echo "$dev_output" | grep -q "ifindex"; then + ktap_test_fail "YNL CLI netdev operations (netdev dev-get output missing ifindex)" + return + fi + + if ! ip netns exec "$testns" $ynl --family netdev \ + --dump queue-get --json "{\"ifindex\": $ifindex}" &>/dev/null; then + ktap_test_fail "YNL CLI netdev operations (failed to get netdev queue info)" + return + fi + + ktap_test_pass "YNL CLI netdev operations" +} +TESTS_NO=$((TESTS_NO + 1)) + +# Test ethtool family operations (rings-get, linkinfo-get) +cli_ethtool_ops() +{ + local rings_output + local linkinfo_output + + rings_output=$(ip netns exec "$testns" $ynl --family ethtool \ + --do rings-get --json "{\"header\": {\"dev-name\": \"$NSIM_DEV_NAME\"}}" 2>/dev/null) + + if ! echo "$rings_output" | grep -q "header"; then + ktap_test_fail "YNL CLI ethtool operations (ethtool rings-get output missing header)" + return + fi + + linkinfo_output=$(ip netns exec "$testns" $ynl --family ethtool \ + --do linkinfo-get --json "{\"header\": {\"dev-name\": \"$VETH_A\"}}" 2>/dev/null) + + if ! echo "$linkinfo_output" | grep -q "header"; then + ktap_test_fail "YNL CLI ethtool operations (ethtool linkinfo-get output missing header)" + return + fi + + ktap_test_pass "YNL CLI ethtool operations" +} +TESTS_NO=$((TESTS_NO + 1)) + +# Test rt-route family operations +cli_rt_route_ops() +{ + local ifindex + + if ! $ynl --list-families 2>/dev/null | grep -q "rt-route"; then + ktap_test_skip "YNL CLI rt-route operations (rt-route family not available)" + return + fi + + ifindex=$(ip netns exec "$testns" cat /sys/class/net/"$NSIM_DEV_NAME"/ifindex 2>/dev/null) + + # Add route: 192.0.2.0/24 dev $dev scope link + if ! ip netns exec "$testns" $ynl --family rt-route --do newroute --create \ + --json "{\"dst\": \"192.0.2.0\", \"oif\": $ifindex, \"rtm-dst-len\": 24, \"rtm-family\": 2, \"rtm-scope\": 253, \"rtm-type\": 1, \"rtm-protocol\": 3, \"rtm-table\": 254}" &>/dev/null; then + ktap_test_fail "YNL CLI rt-route operations (failed to add route)" + return + fi + + local route_output + route_output=$(ip netns exec "$testns" $ynl --family rt-route --dump getroute 2>/dev/null) + if echo "$route_output" | grep -q "192.0.2.0"; then + ktap_test_pass "YNL CLI rt-route operations" + else + ktap_test_fail "YNL CLI rt-route operations (failed to verify route)" + fi + + ip netns exec "$testns" $ynl --family rt-route --do delroute \ + --json "{\"dst\": \"192.0.2.0\", \"oif\": $ifindex, \"rtm-dst-len\": 24, \"rtm-family\": 2, \"rtm-scope\": 253, \"rtm-type\": 1, \"rtm-protocol\": 3, \"rtm-table\": 254}" &>/dev/null +} +TESTS_NO=$((TESTS_NO + 1)) + +# Test rt-addr family operations +cli_rt_addr_ops() +{ + local ifindex + + if ! $ynl --list-families 2>/dev/null | grep -q "rt-addr"; then + ktap_test_skip "YNL CLI rt-addr operations (rt-addr family not available)" + return + fi + + ifindex=$(ip netns exec "$testns" cat /sys/class/net/"$NSIM_DEV_NAME"/ifindex 2>/dev/null) + + if ! ip netns exec "$testns" $ynl --family rt-addr --do newaddr \ + --json "{\"ifa-index\": $ifindex, \"local\": \"192.0.2.100\", \"ifa-prefixlen\": 24, \"ifa-family\": 2}" &>/dev/null; then + ktap_test_fail "YNL CLI rt-addr operations (failed to add address)" + return + fi + + local addr_output + addr_output=$(ip netns exec "$testns" $ynl --family rt-addr --dump getaddr 2>/dev/null) + if echo "$addr_output" | grep -q "192.0.2.100"; then + ktap_test_pass "YNL CLI rt-addr operations" + else + ktap_test_fail "YNL CLI rt-addr operations (failed to verify address)" + fi + + ip netns exec "$testns" $ynl --family rt-addr --do deladdr \ + --json "{\"ifa-index\": $ifindex, \"local\": \"192.0.2.100\", \"ifa-prefixlen\": 24, \"ifa-family\": 2}" &>/dev/null +} +TESTS_NO=$((TESTS_NO + 1)) + +# Test rt-link family operations +cli_rt_link_ops() +{ + if ! $ynl --list-families 2>/dev/null | grep -q "rt-link"; then + ktap_test_skip "YNL CLI rt-link operations (rt-link family not available)" + return + fi + + if ! ip netns exec "$testns" $ynl --family rt-link --do newlink --create \ + --json "{\"ifname\": \"dummy0\", \"linkinfo\": {\"kind\": \"dummy\"}}" &>/dev/null; then + ktap_test_fail "YNL CLI rt-link operations (failed to add link)" + return + fi + + local link_output + link_output=$(ip netns exec "$testns" $ynl --family rt-link --dump getlink 2>/dev/null) + if echo "$link_output" | grep -q "$NSIM_DEV_NAME" && echo "$link_output" | grep -q "dummy0"; then + ktap_test_pass "YNL CLI rt-link operations" + else + ktap_test_fail "YNL CLI rt-link operations (failed to verify link)" + fi + + ip netns exec "$testns" $ynl --family rt-link --do dellink \ + --json "{\"ifname\": \"dummy0\"}" &>/dev/null +} +TESTS_NO=$((TESTS_NO + 1)) + +# Test rt-neigh family operations +cli_rt_neigh_ops() +{ + local ifindex + + if ! $ynl --list-families 2>/dev/null | grep -q "rt-neigh"; then + ktap_test_skip "YNL CLI rt-neigh operations (rt-neigh family not available)" + return + fi + + ifindex=$(ip netns exec "$testns" cat /sys/class/net/"$NSIM_DEV_NAME"/ifindex 2>/dev/null) + + # Add neighbor: 192.0.2.1 dev nsim1338 lladdr 11:22:33:44:55:66 PERMANENT + if ! ip netns exec "$testns" $ynl --family rt-neigh --do newneigh --create \ + --json "{\"ndm-ifindex\": $ifindex, \"dst\": \"192.0.2.1\", \"lladdr\": \"11:22:33:44:55:66\", \"ndm-family\": 2, \"ndm-state\": 128}" &>/dev/null; then + ktap_test_fail "YNL CLI rt-neigh operations (failed to add neighbor)" + fi + + local neigh_output + neigh_output=$(ip netns exec "$testns" $ynl --family rt-neigh --dump getneigh 2>/dev/null) + if echo "$neigh_output" | grep -q "192.0.2.1"; then + ktap_test_pass "YNL CLI rt-neigh operations" + else + ktap_test_fail "YNL CLI rt-neigh operations (failed to verify neighbor)" + fi + + ip netns exec "$testns" $ynl --family rt-neigh --do delneigh \ + --json "{\"ndm-ifindex\": $ifindex, \"dst\": \"192.0.2.1\", \"lladdr\": \"11:22:33:44:55:66\", \"ndm-family\": 2}" &>/dev/null +} +TESTS_NO=$((TESTS_NO + 1)) + +# Test rt-rule family operations +cli_rt_rule_ops() +{ + if ! $ynl --list-families 2>/dev/null | grep -q "rt-rule"; then + ktap_test_skip "YNL CLI rt-rule operations (rt-rule family not available)" + return + fi + + # Add rule: from 192.0.2.0/24 lookup 100 none + if ! ip netns exec "$testns" $ynl --family rt-rule --do newrule \ + --json "{\"family\": 2, \"src-len\": 24, \"src\": \"192.0.2.0\", \"table\": 100}" &>/dev/null; then + ktap_test_fail "YNL CLI rt-rule operations (failed to add rule)" + return + fi + + local rule_output + rule_output=$(ip netns exec "$testns" $ynl --family rt-rule --dump getrule 2>/dev/null) + if echo "$rule_output" | grep -q "192.0.2.0"; then + ktap_test_pass "YNL CLI rt-rule operations" + else + ktap_test_fail "YNL CLI rt-rule operations (failed to verify rule)" + fi + + ip netns exec "$testns" $ynl --family rt-rule --do delrule \ + --json "{\"family\": 2, \"src-len\": 24, \"src\": \"192.0.2.0\", \"table\": 100}" &>/dev/null +} +TESTS_NO=$((TESTS_NO + 1)) + +# Test nlctrl family operations +cli_nlctrl_ops() +{ + local family_output + + if ! family_output=$($ynl --family nlctrl \ + --do getfamily --json "{\"family-name\": \"netdev\"}" 2>/dev/null); then + ktap_test_fail "YNL CLI nlctrl getfamily (failed to get nlctrl family info)" + return + fi + + if ! echo "$family_output" | grep -q "family-name"; then + ktap_test_fail "YNL CLI nlctrl getfamily (nlctrl getfamily output missing family-name)" + return + fi + + if ! echo "$family_output" | grep -q "family-id"; then + ktap_test_fail "YNL CLI nlctrl getfamily (nlctrl getfamily output missing family-id)" + return + fi + + ktap_test_pass "YNL CLI nlctrl getfamily" +} +TESTS_NO=$((TESTS_NO + 1)) + +setup() +{ + modprobe netdevsim &> /dev/null + if ! [ -f /sys/bus/netdevsim/new_device ]; then + ktap_skip_all "netdevsim module not available" + exit "$KSFT_SKIP" + fi + + if ! ip netns add "$testns" 2>/dev/null; then + ktap_skip_all "failed to create test namespace" + exit "$KSFT_SKIP" + fi + + echo "$NSIM_ID 1" | ip netns exec "$testns" tee /sys/bus/netdevsim/new_device >/dev/null 2>&1 || { + ktap_skip_all "failed to create netdevsim device" + exit "$KSFT_SKIP" + } + + local dev + dev=$(ip netns exec "$testns" ls /sys/bus/netdevsim/devices/netdevsim$NSIM_ID/net 2>/dev/null | head -1) + if [[ -z "$dev" ]]; then + ktap_skip_all "failed to find netdevsim device" + exit "$KSFT_SKIP" + fi + + ip -netns "$testns" link set dev "$dev" name "$NSIM_DEV_NAME" 2>/dev/null || { + ktap_skip_all "failed to rename netdevsim device" + exit "$KSFT_SKIP" + } + + ip -netns "$testns" link set dev "$NSIM_DEV_NAME" up 2>/dev/null + + if ! ip -n "$testns" link add "$VETH_A" type veth peer name "$VETH_B" 2>/dev/null; then + ktap_skip_all "failed to create veth pair" + exit "$KSFT_SKIP" + fi + + ip -n "$testns" link set "$VETH_A" up 2>/dev/null + ip -n "$testns" link set "$VETH_B" up 2>/dev/null +} + +cleanup() +{ + ip netns exec "$testns" bash -c "echo $NSIM_ID > /sys/bus/netdevsim/del_device" 2>/dev/null || true + ip netns del "$testns" 2>/dev/null || true +} + +# Check if ynl command is available +if ! command -v $ynl &>/dev/null && [[ ! -x $ynl ]]; then + ktap_skip_all "ynl command not found: $ynl" + exit "$KSFT_SKIP" +fi + +trap cleanup EXIT + +ktap_print_header +setup +ktap_set_plan "${TESTS_NO}" + +cli_list_families +cli_netdev_ops +cli_ethtool_ops +cli_rt_route_ops +cli_rt_addr_ops +cli_rt_link_ops +cli_rt_neigh_ops +cli_rt_rule_ops +cli_nlctrl_ops + +ktap_finished diff --git a/tools/net/ynl/tests/test_ynl_ethtool.sh b/tools/net/ynl/tests/test_ynl_ethtool.sh new file mode 100755 index 000000000000..b826269017f4 --- /dev/null +++ b/tools/net/ynl/tests/test_ynl_ethtool.sh @@ -0,0 +1,222 @@ +#!/bin/bash +# SPDX-License-Identifier: GPL-2.0 +# Test YNL ethtool functionality + +# Load KTAP test helpers +KSELFTEST_KTAP_HELPERS="$(dirname "$(realpath "$0")")/../../../testing/selftests/kselftest/ktap_helpers.sh" +# shellcheck source=../../../testing/selftests/kselftest/ktap_helpers.sh +source "$KSELFTEST_KTAP_HELPERS" + +# Default ynl-ethtool path for direct execution, can be overridden by make install +ynl_ethtool="../pyynl/ethtool.py" + +readonly NSIM_ID="1337" +readonly NSIM_DEV_NAME="nsim${NSIM_ID}" +readonly VETH_A="veth_a" +readonly VETH_B="veth_b" + +testns="ynl-ethtool-$(mktemp -u XXXXXX)" +TESTS_NO=0 + +# Uses veth device as netdevsim doesn't support basic ethtool device info +ethtool_device_info() +{ + local info_output + + info_output=$(ip netns exec "$testns" $ynl_ethtool "$VETH_A" 2>/dev/null) + + if ! echo "$info_output" | grep -q "Settings for"; then + ktap_test_fail "YNL ethtool device info (device info output missing expected content)" + return + fi + + ktap_test_pass "YNL ethtool device info" +} +TESTS_NO=$((TESTS_NO + 1)) + +ethtool_statistics() +{ + local stats_output + + stats_output=$(ip netns exec "$testns" $ynl_ethtool --statistics "$NSIM_DEV_NAME" 2>/dev/null) + + if ! echo "$stats_output" | grep -q -E "(NIC statistics|packets|bytes)"; then + ktap_test_fail "YNL ethtool statistics (statistics output missing expected content)" + return + fi + + ktap_test_pass "YNL ethtool statistics" +} +TESTS_NO=$((TESTS_NO + 1)) + +ethtool_ring_params() +{ + local ring_output + + ring_output=$(ip netns exec "$testns" $ynl_ethtool --show-ring "$NSIM_DEV_NAME" 2>/dev/null) + + if ! echo "$ring_output" | grep -q -E "(Ring parameters|RX|TX)"; then + ktap_test_fail "YNL ethtool ring parameters (ring parameters output missing expected content)" + return + fi + + if ! ip netns exec "$testns" $ynl_ethtool --set-ring "$NSIM_DEV_NAME" rx 64 2>/dev/null; then + ktap_test_fail "YNL ethtool ring parameters (set-ring command failed unexpectedly)" + return + fi + + ktap_test_pass "YNL ethtool ring parameters (show/set)" +} +TESTS_NO=$((TESTS_NO + 1)) + +ethtool_coalesce_params() +{ + if ! ip netns exec "$testns" $ynl_ethtool --show-coalesce "$NSIM_DEV_NAME" &>/dev/null; then + ktap_test_fail "YNL ethtool coalesce parameters (failed to get coalesce parameters)" + return + fi + + if ! ip netns exec "$testns" $ynl_ethtool --set-coalesce "$NSIM_DEV_NAME" rx-usecs 50 2>/dev/null; then + ktap_test_fail "YNL ethtool coalesce parameters (set-coalesce command failed unexpectedly)" + return + fi + + ktap_test_pass "YNL ethtool coalesce parameters (show/set)" +} +TESTS_NO=$((TESTS_NO + 1)) + +ethtool_pause_params() +{ + if ! ip netns exec "$testns" $ynl_ethtool --show-pause "$NSIM_DEV_NAME" &>/dev/null; then + ktap_test_fail "YNL ethtool pause parameters (failed to get pause parameters)" + return + fi + + if ! ip netns exec "$testns" $ynl_ethtool --set-pause "$NSIM_DEV_NAME" tx 1 rx 1 2>/dev/null; then + ktap_test_fail "YNL ethtool pause parameters (set-pause command failed unexpectedly)" + return + fi + + ktap_test_pass "YNL ethtool pause parameters (show/set)" +} +TESTS_NO=$((TESTS_NO + 1)) + +ethtool_features_info() +{ + local features_output + + features_output=$(ip netns exec "$testns" $ynl_ethtool --show-features "$NSIM_DEV_NAME" 2>/dev/null) + + if ! echo "$features_output" | grep -q -E "(Features|offload)"; then + ktap_test_fail "YNL ethtool features info (features output missing expected content)" + return + fi + + ktap_test_pass "YNL ethtool features info (show/set)" +} +TESTS_NO=$((TESTS_NO + 1)) + +ethtool_channels_info() +{ + local channels_output + + channels_output=$(ip netns exec "$testns" $ynl_ethtool --show-channels "$NSIM_DEV_NAME" 2>/dev/null) + + if ! echo "$channels_output" | grep -q -E "(Channel|Combined|RX|TX)"; then + ktap_test_fail "YNL ethtool channels info (channels output missing expected content)" + return + fi + + if ! ip netns exec "$testns" $ynl_ethtool --set-channels "$NSIM_DEV_NAME" combined-count 1 2>/dev/null; then + ktap_test_fail "YNL ethtool channels info (set-channels command failed unexpectedly)" + return + fi + + ktap_test_pass "YNL ethtool channels info (show/set)" +} +TESTS_NO=$((TESTS_NO + 1)) + +ethtool_time_stamping() +{ + local ts_output + + ts_output=$(ip netns exec "$testns" $ynl_ethtool --show-time-stamping "$NSIM_DEV_NAME" 2>/dev/null) + + if ! echo "$ts_output" | grep -q -E "(Time stamping|timestamping|SOF_TIMESTAMPING)"; then + ktap_test_fail "YNL ethtool time stamping (time stamping output missing expected content)" + return + fi + + ktap_test_pass "YNL ethtool time stamping" +} +TESTS_NO=$((TESTS_NO + 1)) + +setup() +{ + modprobe netdevsim &> /dev/null + if ! [ -f /sys/bus/netdevsim/new_device ]; then + ktap_skip_all "netdevsim module not available" + exit "$KSFT_SKIP" + fi + + if ! ip netns add "$testns" 2>/dev/null; then + ktap_skip_all "failed to create test namespace" + exit "$KSFT_SKIP" + fi + + echo "$NSIM_ID 1" | ip netns exec "$testns" tee /sys/bus/netdevsim/new_device >/dev/null 2>&1 || { + ktap_skip_all "failed to create netdevsim device" + exit "$KSFT_SKIP" + } + + local dev + dev=$(ip netns exec "$testns" ls /sys/bus/netdevsim/devices/netdevsim$NSIM_ID/net 2>/dev/null | head -1) + if [[ -z "$dev" ]]; then + ktap_skip_all "failed to find netdevsim device" + exit "$KSFT_SKIP" + fi + + ip -netns "$testns" link set dev "$dev" name "$NSIM_DEV_NAME" 2>/dev/null || { + ktap_skip_all "failed to rename netdevsim device" + exit "$KSFT_SKIP" + } + + ip -netns "$testns" link set dev "$NSIM_DEV_NAME" up 2>/dev/null + + if ! ip -n "$testns" link add "$VETH_A" type veth peer name "$VETH_B" 2>/dev/null; then + ktap_skip_all "failed to create veth pair" + exit "$KSFT_SKIP" + fi + + ip -n "$testns" link set "$VETH_A" up 2>/dev/null + ip -n "$testns" link set "$VETH_B" up 2>/dev/null +} + +cleanup() +{ + ip netns exec "$testns" bash -c "echo $NSIM_ID > /sys/bus/netdevsim/del_device" 2>/dev/null || true + ip netns del "$testns" 2>/dev/null || true +} + +# Check if ynl-ethtool command is available +if ! command -v $ynl_ethtool &>/dev/null && [[ ! -x $ynl_ethtool ]]; then + ktap_skip_all "ynl-ethtool command not found: $ynl_ethtool" + exit "$KSFT_SKIP" +fi + +trap cleanup EXIT + +ktap_print_header +setup +ktap_set_plan "${TESTS_NO}" + +ethtool_device_info +ethtool_statistics +ethtool_ring_params +ethtool_coalesce_params +ethtool_pause_params +ethtool_features_info +ethtool_channels_info +ethtool_time_stamping + +ktap_finished From 002373a8b01df23be072bb91741c2e8b2b70ccc1 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Sat, 22 Nov 2025 13:04:27 +0200 Subject: [PATCH 1193/1381] net: phy: dp83867: implement configurability for SGMII in-band auto-negotiation Implement the inband_caps() and config_inband() PHY driver methods, to allow working with PCS devices that do not support or want in-band to be used. There is a complication due to existing logic from commit c76acfb7e19d ("net: phy: dp83867: retrigger SGMII AN when link change") which might re-enable what dp83867_config_inband() has disabled. So we need to modify dp83867_link_change_notify() to use phy_modify_changed() when temporarily disabling in-band autoneg. If the return code is 0, it means the original in-band was disabled and we need to keep it disabled. If the return code is 1, the original was enabled and we need to re-enable it. If negative, there was an error, which was silent before, and remains silent now. dp83867_config_inband() and dp83867_link_change_notify() are serialized by the phydev->lock. Signed-off-by: Vladimir Oltean Link: https://patch.msgid.link/20251122110427.133035-1-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- drivers/net/phy/dp83867.c | 36 +++++++++++++++++++++++++++++------- 1 file changed, 29 insertions(+), 7 deletions(-) diff --git a/drivers/net/phy/dp83867.c b/drivers/net/phy/dp83867.c index 36a0c1b7f59c..5f5de01c41e1 100644 --- a/drivers/net/phy/dp83867.c +++ b/drivers/net/phy/dp83867.c @@ -937,15 +937,15 @@ static void dp83867_link_change_notify(struct phy_device *phydev) * whenever there is a link change. */ if (phydev->interface == PHY_INTERFACE_MODE_SGMII) { - int val = 0; + int val; - val = phy_clear_bits(phydev, DP83867_CFG2, + val = phy_modify_changed(phydev, DP83867_CFG2, + DP83867_SGMII_AUTONEG_EN, 0); + + /* Keep the in-band setting made by dp83867_config_inband() */ + if (val != 0) + phy_set_bits(phydev, DP83867_CFG2, DP83867_SGMII_AUTONEG_EN); - if (val < 0) - return; - - phy_set_bits(phydev, DP83867_CFG2, - DP83867_SGMII_AUTONEG_EN); } } @@ -1116,6 +1116,25 @@ static int dp83867_led_polarity_set(struct phy_device *phydev, int index, DP83867_LED_POLARITY(index), polarity); } +static unsigned int dp83867_inband_caps(struct phy_device *phydev, + phy_interface_t interface) +{ + if (interface == PHY_INTERFACE_MODE_SGMII) + return LINK_INBAND_ENABLE | LINK_INBAND_DISABLE; + + return 0; +} + +static int dp83867_config_inband(struct phy_device *phydev, unsigned int modes) +{ + int val = 0; + + if (modes == LINK_INBAND_ENABLE) + val = DP83867_SGMII_AUTONEG_EN; + + return phy_modify(phydev, DP83867_CFG2, DP83867_SGMII_AUTONEG_EN, val); +} + static struct phy_driver dp83867_driver[] = { { .phy_id = DP83867_PHY_ID, @@ -1149,6 +1168,9 @@ static struct phy_driver dp83867_driver[] = { .led_hw_control_set = dp83867_led_hw_control_set, .led_hw_control_get = dp83867_led_hw_control_get, .led_polarity_set = dp83867_led_polarity_set, + + .inband_caps = dp83867_inband_caps, + .config_inband = dp83867_config_inband, }, }; module_phy_driver(dp83867_driver); From 7241d80e77066dacc2fd7ad5ce29e524548a4940 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Sat, 22 Nov 2025 13:55:23 +0200 Subject: [PATCH 1194/1381] net: dpaa: fman_memac: complete phylink support with 2500base-x The DPAA phylink conversion in the following commits partially developed code for handling the 2500base-x host interface mode (called "2.5G SGMII" in LS1043A/LS1046A reference manuals). - 0fc83bd79589 ("net: fman: memac: Add serdes support") - 5d93cfcf7360 ("net: dpaa: Convert to phylink") In principle, having phy-interface-mode = "2500base-x" and a pcsphy-handle (unnamed or with pcs-handle-names = "sgmii") in the MAC device tree node results in PHY_INTERFACE_MODE_2500BASEX being set in phylink_config :: supported_interfaces, but this isn't sufficient. Because memac_select_pcs() returns no PCS for PHY_INTERFACE_MODE_2500BASEX, the Lynx PCS code never engages. There's a chance the PCS driver doesn't have any configuration to change for 2500base-x fixed-link (based on bootloader pre-initialization), but there's an even higher chance that this is not the case, and the PCS remains misconfigured. More importantly, memac_if_mode() does not handle PHY_INTERFACE_MODE_2500BASEX, and it should be telling the mEMAC to configure itself in GMII mode (which is upclocked by the PCS). Currently it prints a WARN_ON() and returns zero, aka IF_MODE_10G (incorrect). The additional case statement in memac_prepare() for calling phy_set_mode_ext() does not make any difference, because there is no generic PHY driver for the Lynx 10G SerDes from LS1043A/LS1046A. But we add it nonetheless, for consistency. Regarding the question "did 2500base-x ever work with the FMan mEMAC mainline code prior to the phylink conversion?" - the answer is more nuanced. For context, the previous phylib-based implementation was unable to describe the fixed-link speed as 2500, because the software PHY implementation is limited to 1G. However, improperly describing the link as an sgmii fixed-link with speed = <1000> would have resulted in a functional 2.5G speed, because there is no other difference than the SerDes lane clock net frequency (3.125 GHz for 2500base-x) - all the other higher-level settings are the same, and the SerDes lane frequency is currently handled by the RCW. But this hack cannot be extended towards a phylib PHY such as Aquantia operating in OCSGMII, because the latter requires phy-mode = "2500base-x", which the mEMAC driver did not support prior to the phylink conversion. So I do not really consider this a regression, just completing support for a missing feature. The FMan mEMAC driver sets phylink's "default_an_inband" property to true, making it as if the device tree node had the managed = "in-band-status" property anyway. This default made sense for SGMII, where it was added to avoid regressions, but for 2500base-x we learned only recently how to enable in-band autoneg: https://lore.kernel.org/netdev/20251122113433.141930-1-vladimir.oltean@nxp.com/ so the driver needs to opt out of this default in-band enabled behaviour, and only enable in-band based on the device tree property. Suggested-by: Russell King (Oracle) Link: https://lore.kernel.org/netdev/aIyx0OLWGw5zKarX@shell.armlinux.org.uk/#t Signed-off-by: Vladimir Oltean Link: https://patch.msgid.link/20251122115523.150260-1-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/freescale/fman/fman_memac.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/net/ethernet/freescale/fman/fman_memac.c b/drivers/net/ethernet/freescale/fman/fman_memac.c index 0291093f2e4e..d32ffd6be7b1 100644 --- a/drivers/net/ethernet/freescale/fman/fman_memac.c +++ b/drivers/net/ethernet/freescale/fman/fman_memac.c @@ -649,6 +649,7 @@ static u32 memac_if_mode(phy_interface_t interface) return IF_MODE_GMII | IF_MODE_RGMII; case PHY_INTERFACE_MODE_SGMII: case PHY_INTERFACE_MODE_1000BASEX: + case PHY_INTERFACE_MODE_2500BASEX: case PHY_INTERFACE_MODE_QSGMII: return IF_MODE_GMII; case PHY_INTERFACE_MODE_10GBASER: @@ -667,6 +668,7 @@ static struct phylink_pcs *memac_select_pcs(struct phylink_config *config, switch (iface) { case PHY_INTERFACE_MODE_SGMII: case PHY_INTERFACE_MODE_1000BASEX: + case PHY_INTERFACE_MODE_2500BASEX: return memac->sgmii_pcs; case PHY_INTERFACE_MODE_QSGMII: return memac->qsgmii_pcs; @@ -685,6 +687,7 @@ static int memac_prepare(struct phylink_config *config, unsigned int mode, switch (iface) { case PHY_INTERFACE_MODE_SGMII: case PHY_INTERFACE_MODE_1000BASEX: + case PHY_INTERFACE_MODE_2500BASEX: case PHY_INTERFACE_MODE_QSGMII: case PHY_INTERFACE_MODE_10GBASER: return phy_set_mode_ext(memac->serdes, PHY_MODE_ETHERNET, @@ -1226,6 +1229,7 @@ int memac_initialization(struct mac_device *mac_dev, * those configurations modes don't use in-band autonegotiation. */ if (!of_property_present(mac_node, "managed") && + mac_dev->phy_if != PHY_INTERFACE_MODE_2500BASEX && mac_dev->phy_if != PHY_INTERFACE_MODE_MII && !phy_interface_mode_is_rgmii(mac_dev->phy_if)) mac_dev->phylink_config.default_an_inband = true; From 37a96c2009f75d25c33b81edcb016a48e1342846 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Sat, 22 Nov 2025 13:59:31 +0200 Subject: [PATCH 1195/1381] net: fman_memac: report structured ethtool counters The FMan driver has support for 2 MACs: mEMAC (newer, present on Layerscape and PowerPC T series) and dTSEC/TGEC (older, present on PowerPC P series). I only have handy access to the mEMAC, and this adds support for MAC counters for those platforms. MAC counters are necessary for any kind of low-level debugging, and currently there is no mechanism to dump them. Signed-off-by: Vladimir Oltean Link: https://patch.msgid.link/20251122115931.151719-1-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- .../ethernet/freescale/dpaa/dpaa_ethtool.c | 45 ++++++++++ .../net/ethernet/freescale/fman/fman_memac.c | 87 +++++++++++++++++++ drivers/net/ethernet/freescale/fman/mac.h | 14 +++ 3 files changed, 146 insertions(+) diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_ethtool.c b/drivers/net/ethernet/freescale/dpaa/dpaa_ethtool.c index d09e456f14c0..ed3fa80af8c3 100644 --- a/drivers/net/ethernet/freescale/dpaa/dpaa_ethtool.c +++ b/drivers/net/ethernet/freescale/dpaa/dpaa_ethtool.c @@ -467,6 +467,47 @@ static int dpaa_set_coalesce(struct net_device *dev, return res; } +static void dpaa_get_pause_stats(struct net_device *net_dev, + struct ethtool_pause_stats *s) +{ + struct dpaa_priv *priv = netdev_priv(net_dev); + struct mac_device *mac_dev = priv->mac_dev; + + if (mac_dev->get_pause_stats) + mac_dev->get_pause_stats(mac_dev->fman_mac, s); +} + +static void dpaa_get_rmon_stats(struct net_device *net_dev, + struct ethtool_rmon_stats *s, + const struct ethtool_rmon_hist_range **ranges) +{ + struct dpaa_priv *priv = netdev_priv(net_dev); + struct mac_device *mac_dev = priv->mac_dev; + + if (mac_dev->get_rmon_stats) + mac_dev->get_rmon_stats(mac_dev->fman_mac, s, ranges); +} + +static void dpaa_get_eth_ctrl_stats(struct net_device *net_dev, + struct ethtool_eth_ctrl_stats *s) +{ + struct dpaa_priv *priv = netdev_priv(net_dev); + struct mac_device *mac_dev = priv->mac_dev; + + if (mac_dev->get_eth_ctrl_stats) + mac_dev->get_eth_ctrl_stats(mac_dev->fman_mac, s); +} + +static void dpaa_get_eth_mac_stats(struct net_device *net_dev, + struct ethtool_eth_mac_stats *s) +{ + struct dpaa_priv *priv = netdev_priv(net_dev); + struct mac_device *mac_dev = priv->mac_dev; + + if (mac_dev->get_eth_mac_stats) + mac_dev->get_eth_mac_stats(mac_dev->fman_mac, s); +} + const struct ethtool_ops dpaa_ethtool_ops = { .supported_coalesce_params = ETHTOOL_COALESCE_RX_USECS | ETHTOOL_COALESCE_RX_MAX_FRAMES, @@ -487,4 +528,8 @@ const struct ethtool_ops dpaa_ethtool_ops = { .get_ts_info = dpaa_get_ts_info, .get_coalesce = dpaa_get_coalesce, .set_coalesce = dpaa_set_coalesce, + .get_pause_stats = dpaa_get_pause_stats, + .get_rmon_stats = dpaa_get_rmon_stats, + .get_eth_ctrl_stats = dpaa_get_eth_ctrl_stats, + .get_eth_mac_stats = dpaa_get_eth_mac_stats, }; diff --git a/drivers/net/ethernet/freescale/fman/fman_memac.c b/drivers/net/ethernet/freescale/fman/fman_memac.c index d32ffd6be7b1..c84f0336c94c 100644 --- a/drivers/net/ethernet/freescale/fman/fman_memac.c +++ b/drivers/net/ethernet/freescale/fman/fman_memac.c @@ -900,6 +900,89 @@ static int memac_set_exception(struct fman_mac *memac, return 0; } +static u64 memac_read64(void __iomem *reg) +{ + u32 low, high, tmp; + + do { + high = ioread32be(reg + 4); + low = ioread32be(reg); + tmp = ioread32be(reg + 4); + } while (high != tmp); + + return ((u64)high << 32) | low; +} + +static void memac_get_pause_stats(struct fman_mac *memac, + struct ethtool_pause_stats *s) +{ + s->tx_pause_frames = memac_read64(&memac->regs->txpf_l); + s->rx_pause_frames = memac_read64(&memac->regs->rxpf_l); +} + +static const struct ethtool_rmon_hist_range memac_rmon_ranges[] = { + { 64, 64 }, + { 65, 127 }, + { 128, 255 }, + { 256, 511 }, + { 512, 1023 }, + { 1024, 1518 }, + { 1519, 9600 }, + {}, +}; + +static void memac_get_rmon_stats(struct fman_mac *memac, + struct ethtool_rmon_stats *s, + const struct ethtool_rmon_hist_range **ranges) +{ + s->undersize_pkts = memac_read64(&memac->regs->rund_l); + s->oversize_pkts = memac_read64(&memac->regs->rovr_l); + s->fragments = memac_read64(&memac->regs->rfrg_l); + s->jabbers = memac_read64(&memac->regs->rjbr_l); + + s->hist[0] = memac_read64(&memac->regs->r64_l); + s->hist[1] = memac_read64(&memac->regs->r127_l); + s->hist[2] = memac_read64(&memac->regs->r255_l); + s->hist[3] = memac_read64(&memac->regs->r511_l); + s->hist[4] = memac_read64(&memac->regs->r1023_l); + s->hist[5] = memac_read64(&memac->regs->r1518_l); + s->hist[6] = memac_read64(&memac->regs->r1519x_l); + + s->hist_tx[0] = memac_read64(&memac->regs->t64_l); + s->hist_tx[1] = memac_read64(&memac->regs->t127_l); + s->hist_tx[2] = memac_read64(&memac->regs->t255_l); + s->hist_tx[3] = memac_read64(&memac->regs->t511_l); + s->hist_tx[4] = memac_read64(&memac->regs->t1023_l); + s->hist_tx[5] = memac_read64(&memac->regs->t1518_l); + s->hist_tx[6] = memac_read64(&memac->regs->t1519x_l); + + *ranges = memac_rmon_ranges; +} + +static void memac_get_eth_ctrl_stats(struct fman_mac *memac, + struct ethtool_eth_ctrl_stats *s) +{ + s->MACControlFramesTransmitted = memac_read64(&memac->regs->tcnp_l); + s->MACControlFramesReceived = memac_read64(&memac->regs->rcnp_l); +} + +static void memac_get_eth_mac_stats(struct fman_mac *memac, + struct ethtool_eth_mac_stats *s) +{ + s->FramesTransmittedOK = memac_read64(&memac->regs->tfrm_l); + s->FramesReceivedOK = memac_read64(&memac->regs->rfrm_l); + s->FrameCheckSequenceErrors = memac_read64(&memac->regs->rfcs_l); + s->AlignmentErrors = memac_read64(&memac->regs->raln_l); + s->OctetsTransmittedOK = memac_read64(&memac->regs->teoct_l); + s->FramesLostDueToIntMACXmitError = memac_read64(&memac->regs->terr_l); + s->OctetsReceivedOK = memac_read64(&memac->regs->reoct_l); + s->FramesLostDueToIntMACRcvError = memac_read64(&memac->regs->rdrntp_l); + s->MulticastFramesXmittedOK = memac_read64(&memac->regs->tmca_l); + s->BroadcastFramesXmittedOK = memac_read64(&memac->regs->tbca_l); + s->MulticastFramesReceivedOK = memac_read64(&memac->regs->rmca_l); + s->BroadcastFramesReceivedOK = memac_read64(&memac->regs->rbca_l); +} + static int memac_init(struct fman_mac *memac) { struct memac_cfg *memac_drv_param; @@ -1092,6 +1175,10 @@ int memac_initialization(struct mac_device *mac_dev, mac_dev->set_tstamp = memac_set_tstamp; mac_dev->enable = memac_enable; mac_dev->disable = memac_disable; + mac_dev->get_pause_stats = memac_get_pause_stats; + mac_dev->get_rmon_stats = memac_get_rmon_stats; + mac_dev->get_eth_ctrl_stats = memac_get_eth_ctrl_stats; + mac_dev->get_eth_mac_stats = memac_get_eth_mac_stats; mac_dev->fman_mac = memac_config(mac_dev, params); if (!mac_dev->fman_mac) diff --git a/drivers/net/ethernet/freescale/fman/mac.h b/drivers/net/ethernet/freescale/fman/mac.h index 955ace338965..63c2c5b4f99e 100644 --- a/drivers/net/ethernet/freescale/fman/mac.h +++ b/drivers/net/ethernet/freescale/fman/mac.h @@ -16,6 +16,11 @@ #include "fman.h" #include "fman_mac.h" +struct ethtool_eth_ctrl_stats; +struct ethtool_eth_mac_stats; +struct ethtool_pause_stats; +struct ethtool_rmon_stats; +struct ethtool_rmon_hist_range; struct fman_mac; struct mac_priv_s; @@ -46,6 +51,15 @@ struct mac_device { enet_addr_t *eth_addr); int (*remove_hash_mac_addr)(struct fman_mac *mac_dev, enet_addr_t *eth_addr); + void (*get_pause_stats)(struct fman_mac *memac, + struct ethtool_pause_stats *s); + void (*get_rmon_stats)(struct fman_mac *memac, + struct ethtool_rmon_stats *s, + const struct ethtool_rmon_hist_range **ranges); + void (*get_eth_ctrl_stats)(struct fman_mac *memac, + struct ethtool_eth_ctrl_stats *s); + void (*get_eth_mac_stats)(struct fman_mac *memac, + struct ethtool_eth_mac_stats *s); void (*update_speed)(struct mac_device *mac_dev, int speed); From 91f3305b97fc41b42b67b9ff58441aa4d72cde2e Mon Sep 17 00:00:00 2001 From: Tao Lan Date: Sat, 22 Nov 2025 11:46:55 +0800 Subject: [PATCH 1196/1381] net: hibmcge: add support for tracepoint to dump some fields of rx_desc add support for tracepoint to dump some fields of rx_desc Signed-off-by: Tao Lan Signed-off-by: Jijie Shao Link: https://patch.msgid.link/20251122034657.3373143-2-shaojijie@huawei.com Signed-off-by: Jakub Kicinski --- .../net/ethernet/hisilicon/hibmcge/Makefile | 1 + .../net/ethernet/hisilicon/hibmcge/hbg_reg.h | 4 + .../ethernet/hisilicon/hibmcge/hbg_trace.h | 84 +++++++++++++++++++ .../net/ethernet/hisilicon/hibmcge/hbg_txrx.c | 4 + 4 files changed, 93 insertions(+) create mode 100644 drivers/net/ethernet/hisilicon/hibmcge/hbg_trace.h diff --git a/drivers/net/ethernet/hisilicon/hibmcge/Makefile b/drivers/net/ethernet/hisilicon/hibmcge/Makefile index 1a9da564b306..d6610ba16855 100644 --- a/drivers/net/ethernet/hisilicon/hibmcge/Makefile +++ b/drivers/net/ethernet/hisilicon/hibmcge/Makefile @@ -3,6 +3,7 @@ # Makefile for the HISILICON BMC GE network device drivers. # +ccflags-y += -I$(src) obj-$(CONFIG_HIBMCGE) += hibmcge.o hibmcge-objs = hbg_main.o hbg_hw.o hbg_mdio.o hbg_irq.o hbg_txrx.o hbg_ethtool.o \ diff --git a/drivers/net/ethernet/hisilicon/hibmcge/hbg_reg.h b/drivers/net/ethernet/hisilicon/hibmcge/hbg_reg.h index a39d1e796e4a..30b3903c8f2d 100644 --- a/drivers/net/ethernet/hisilicon/hibmcge/hbg_reg.h +++ b/drivers/net/ethernet/hisilicon/hibmcge/hbg_reg.h @@ -252,6 +252,8 @@ struct hbg_rx_desc { #define HBG_RX_DESC_W2_PKT_LEN_M GENMASK(31, 16) #define HBG_RX_DESC_W2_PORT_NUM_M GENMASK(15, 12) +#define HBG_RX_DESC_W3_IP_OFFSET_M GENMASK(23, 16) +#define HBG_RX_DESC_W3_VLAN_M GENMASK(15, 0) #define HBG_RX_DESC_W4_IP_TCP_UDP_M GENMASK(31, 30) #define HBG_RX_DESC_W4_IPSEC_B BIT(29) #define HBG_RX_DESC_W4_IP_VERSION_B BIT(28) @@ -269,6 +271,8 @@ struct hbg_rx_desc { #define HBG_RX_DESC_W4_L3_ERR_CODE_M GENMASK(12, 9) #define HBG_RX_DESC_W4_L2_ERR_B BIT(8) #define HBG_RX_DESC_W4_IDX_MATCH_B BIT(7) +#define HBG_RX_DESC_W4_PARSE_MODE_M GENMASK(6, 5) +#define HBG_RX_DESC_W5_VALID_SIZE_M GENMASK(15, 0) enum hbg_l3_err_code { HBG_L3_OK = 0, diff --git a/drivers/net/ethernet/hisilicon/hibmcge/hbg_trace.h b/drivers/net/ethernet/hisilicon/hibmcge/hbg_trace.h new file mode 100644 index 000000000000..b70fd960da8d --- /dev/null +++ b/drivers/net/ethernet/hisilicon/hibmcge/hbg_trace.h @@ -0,0 +1,84 @@ +/* SPDX-License-Identifier: GPL-2.0+ */ +/* Copyright (c) 2025 Hisilicon Limited. */ + +/* This must be outside ifdef _HBG_TRACE_H */ +#undef TRACE_SYSTEM +#define TRACE_SYSTEM hibmcge + +#if !defined(_HBG_TRACE_H_) || defined(TRACE_HEADER_MULTI_READ) +#define _HBG_TRACE_H_ + +#include +#include +#include +#include +#include "hbg_reg.h" + +TRACE_EVENT(hbg_rx_desc, + TP_PROTO(struct hbg_priv *priv, u32 index, + struct hbg_rx_desc *rx_desc), + TP_ARGS(priv, index, rx_desc), + + TP_STRUCT__entry(__field(u32, index) + __field(u8, port_num) + __field(u8, ip_offset) + __field(u8, parse_mode) + __field(u8, l4_error_code) + __field(u8, l3_error_code) + __field(u8, l2_error_code) + __field(u16, packet_len) + __field(u16, valid_size) + __field(u16, vlan) + __string(pciname, pci_name(priv->pdev)) + __string(devname, priv->netdev->name) + ), + + TP_fast_assign(__entry->index = index, + __entry->packet_len = + FIELD_GET(HBG_RX_DESC_W2_PKT_LEN_M, + rx_desc->word2); + __entry->port_num = + FIELD_GET(HBG_RX_DESC_W2_PORT_NUM_M, + rx_desc->word2); + __entry->ip_offset = + FIELD_GET(HBG_RX_DESC_W3_IP_OFFSET_M, + rx_desc->word3); + __entry->vlan = + FIELD_GET(HBG_RX_DESC_W3_VLAN_M, + rx_desc->word3); + __entry->parse_mode = + FIELD_GET(HBG_RX_DESC_W4_PARSE_MODE_M, + rx_desc->word4); + __entry->l4_error_code = + FIELD_GET(HBG_RX_DESC_W4_L4_ERR_CODE_M, + rx_desc->word4); + __entry->l3_error_code = + FIELD_GET(HBG_RX_DESC_W4_L3_ERR_CODE_M, + rx_desc->word4); + __entry->l2_error_code = + FIELD_GET(HBG_RX_DESC_W4_L2_ERR_B, + rx_desc->word4); + __entry->valid_size = + FIELD_GET(HBG_RX_DESC_W5_VALID_SIZE_M, + rx_desc->word5); + __assign_str(pciname); + __assign_str(devname); + ), + + TP_printk("%s %s index:%u, port num:%u, len:%u, valid size:%u, ip_offset:%u, vlan:0x%04x, parse mode:%u, l4_err:0x%x, l3_err:0x%x, l2_err:0x%x", + __get_str(pciname), __get_str(devname), __entry->index, + __entry->port_num, __entry->packet_len, + __entry->valid_size, __entry->ip_offset, __entry->vlan, + __entry->parse_mode, __entry->l4_error_code, + __entry->l3_error_code, __entry->l2_error_code + ) +); + +#endif /* _HBG_TRACE_H_ */ + +/* This must be outside ifdef _HBG_TRACE_H */ +#undef TRACE_INCLUDE_PATH +#define TRACE_INCLUDE_PATH . +#undef TRACE_INCLUDE_FILE +#define TRACE_INCLUDE_FILE hbg_trace +#include diff --git a/drivers/net/ethernet/hisilicon/hibmcge/hbg_txrx.c b/drivers/net/ethernet/hisilicon/hibmcge/hbg_txrx.c index 8d814c8f19ea..5f2e48f1dd25 100644 --- a/drivers/net/ethernet/hisilicon/hibmcge/hbg_txrx.c +++ b/drivers/net/ethernet/hisilicon/hibmcge/hbg_txrx.c @@ -7,6 +7,9 @@ #include "hbg_reg.h" #include "hbg_txrx.h" +#define CREATE_TRACE_POINTS +#include "hbg_trace.h" + #define netdev_get_tx_ring(netdev) \ (&(((struct hbg_priv *)netdev_priv(netdev))->tx_ring)) @@ -429,6 +432,7 @@ static int hbg_napi_rx_poll(struct napi_struct *napi, int budget) break; rx_desc = (struct hbg_rx_desc *)buffer->skb->data; pkt_len = FIELD_GET(HBG_RX_DESC_W2_PKT_LEN_M, rx_desc->word2); + trace_hbg_rx_desc(priv, ring->ntc, rx_desc); if (unlikely(!hbg_rx_pkt_check(priv, rx_desc, buffer->skb))) { hbg_buffer_free(buffer); From 2e68bb2e0f7712239711cabc22ceeb56c011c69d Mon Sep 17 00:00:00 2001 From: Jijie Shao Date: Sat, 22 Nov 2025 11:46:56 +0800 Subject: [PATCH 1197/1381] net: hibmcge: reduce packet drop under stress testing Under stress test scenarios, hibmcge driver may not receive packets in a timely manner, which can lead to the buffer of the hardware queue being exhausted, resulting in packet drop. This patch doubles the software queue depth and uses half of the buffer to fill the hardware queue before receiving packets, thus preventing packet loss caused by the hardware queue buffer being exhausted. Signed-off-by: Jijie Shao Link: https://patch.msgid.link/20251122034657.3373143-3-shaojijie@huawei.com Signed-off-by: Jakub Kicinski --- .../net/ethernet/hisilicon/hibmcge/hbg_txrx.c | 47 +++++++++++++++---- 1 file changed, 37 insertions(+), 10 deletions(-) diff --git a/drivers/net/ethernet/hisilicon/hibmcge/hbg_txrx.c b/drivers/net/ethernet/hisilicon/hibmcge/hbg_txrx.c index 5f2e48f1dd25..ea691d564161 100644 --- a/drivers/net/ethernet/hisilicon/hibmcge/hbg_txrx.c +++ b/drivers/net/ethernet/hisilicon/hibmcge/hbg_txrx.c @@ -377,7 +377,8 @@ static int hbg_rx_fill_one_buffer(struct hbg_priv *priv) struct hbg_buffer *buffer; int ret; - if (hbg_queue_is_full(ring->ntc, ring->ntu, ring)) + if (hbg_queue_is_full(ring->ntc, ring->ntu, ring) || + hbg_fifo_is_full(priv, ring->dir)) return 0; buffer = &ring->queue[ring->ntu]; @@ -396,6 +397,26 @@ static int hbg_rx_fill_one_buffer(struct hbg_priv *priv) return 0; } +static int hbg_rx_fill_buffers(struct hbg_priv *priv) +{ + u32 remained = hbg_hw_get_fifo_used_num(priv, HBG_DIR_RX); + u32 max_count = priv->dev_specs.rx_fifo_num; + u32 refill_count; + int ret; + + if (unlikely(remained >= max_count)) + return 0; + + refill_count = max_count - remained; + while (refill_count--) { + ret = hbg_rx_fill_one_buffer(priv); + if (unlikely(ret)) + break; + } + + return ret; +} + static bool hbg_sync_data_from_hw(struct hbg_priv *priv, struct hbg_buffer *buffer) { @@ -420,6 +441,7 @@ static int hbg_napi_rx_poll(struct napi_struct *napi, int budget) u32 packet_done = 0; u32 pkt_len; + hbg_rx_fill_buffers(priv); while (packet_done < budget) { if (unlikely(hbg_queue_is_empty(ring->ntc, ring->ntu, ring))) break; @@ -497,6 +519,16 @@ static int hbg_ring_init(struct hbg_priv *priv, struct hbg_ring *ring, u32 i, len; len = hbg_get_spec_fifo_max_num(priv, dir) + 1; + /* To improve receiving performance under high-stress scenarios, + * in the `hbg_napi_rx_poll()`, we first use the other half of + * the buffer to receive packets from the hardware via the + * `hbg_rx_fill_buffers()`, and then process the packets in the + * original half of the buffer to avoid packet loss caused by + * hardware overflow as much as possible. + */ + if (dir == HBG_DIR_RX) + len += hbg_get_spec_fifo_max_num(priv, dir); + ring->queue = dma_alloc_coherent(&priv->pdev->dev, len * sizeof(*ring->queue), &ring->queue_dma, GFP_KERNEL); @@ -545,21 +577,16 @@ static int hbg_tx_ring_init(struct hbg_priv *priv) static int hbg_rx_ring_init(struct hbg_priv *priv) { int ret; - u32 i; ret = hbg_ring_init(priv, &priv->rx_ring, hbg_napi_rx_poll, HBG_DIR_RX); if (ret) return ret; - for (i = 0; i < priv->rx_ring.len - 1; i++) { - ret = hbg_rx_fill_one_buffer(priv); - if (ret) { - hbg_ring_uninit(&priv->rx_ring); - return ret; - } - } + ret = hbg_rx_fill_buffers(priv); + if (ret) + hbg_ring_uninit(&priv->rx_ring); - return 0; + return ret; } int hbg_txrx_init(struct hbg_priv *priv) From c305959175856b40dea4d9ec10bb2c72add22910 Mon Sep 17 00:00:00 2001 From: Jijie Shao Date: Sat, 22 Nov 2025 11:46:57 +0800 Subject: [PATCH 1198/1381] net: hibmcge: add support for pagepool on rx add support for pagepool on rx, and remove the legacy path Signed-off-by: Jijie Shao Link: https://patch.msgid.link/20251122034657.3373143-4-shaojijie@huawei.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/hisilicon/Kconfig | 1 + .../ethernet/hisilicon/hibmcge/hbg_common.h | 8 + .../net/ethernet/hisilicon/hibmcge/hbg_txrx.c | 168 ++++++++++++++---- 3 files changed, 143 insertions(+), 34 deletions(-) diff --git a/drivers/net/ethernet/hisilicon/Kconfig b/drivers/net/ethernet/hisilicon/Kconfig index 38875c196cb6..18eca7d12c20 100644 --- a/drivers/net/ethernet/hisilicon/Kconfig +++ b/drivers/net/ethernet/hisilicon/Kconfig @@ -151,6 +151,7 @@ config HIBMCGE select FIXED_PHY select MOTORCOMM_PHY select REALTEK_PHY + select PAGE_POOL help If you wish to compile a kernel for a BMC with HIBMC-xx_gmac then you should answer Y to this. This makes this driver suitable for use diff --git a/drivers/net/ethernet/hisilicon/hibmcge/hbg_common.h b/drivers/net/ethernet/hisilicon/hibmcge/hbg_common.h index 2097e4c2b3d7..8e134da3e217 100644 --- a/drivers/net/ethernet/hisilicon/hibmcge/hbg_common.h +++ b/drivers/net/ethernet/hisilicon/hibmcge/hbg_common.h @@ -7,6 +7,7 @@ #include #include #include +#include #include "hbg_reg.h" #define HBG_STATUS_DISABLE 0x0 @@ -55,6 +56,12 @@ struct hbg_buffer { dma_addr_t skb_dma; u32 skb_len; + struct page *page; + void *page_addr; + dma_addr_t page_dma; + u32 page_size; + u32 page_offset; + enum hbg_dir dir; struct hbg_ring *ring; struct hbg_priv *priv; @@ -78,6 +85,7 @@ struct hbg_ring { struct hbg_priv *priv; struct napi_struct napi; char *tout_log_buf; /* tx timeout log buffer */ + struct page_pool *page_pool; /* only for rx */ }; enum hbg_hw_event_type { diff --git a/drivers/net/ethernet/hisilicon/hibmcge/hbg_txrx.c b/drivers/net/ethernet/hisilicon/hibmcge/hbg_txrx.c index ea691d564161..a4ea92c31c2f 100644 --- a/drivers/net/ethernet/hisilicon/hibmcge/hbg_txrx.c +++ b/drivers/net/ethernet/hisilicon/hibmcge/hbg_txrx.c @@ -31,6 +31,11 @@ typeof(ring) _ring = (ring); \ _ring->p = hbg_queue_next_prt(_ring->p, _ring); }) +#define hbg_get_page_order(ring) ({ \ + typeof(ring) _ring = (ring); \ + get_order(hbg_spec_max_frame_len(_ring->priv, _ring->dir)); }) +#define hbg_get_page_size(ring) (PAGE_SIZE << hbg_get_page_order((ring))) + #define HBG_TX_STOP_THRS 2 #define HBG_TX_START_THRS (2 * HBG_TX_STOP_THRS) @@ -65,6 +70,43 @@ static void hbg_dma_unmap(struct hbg_buffer *buffer) buffer->skb_dma = 0; } +static void hbg_buffer_free_page(struct hbg_buffer *buffer) +{ + struct hbg_ring *ring = buffer->ring; + + if (unlikely(!buffer->page)) + return; + + page_pool_put_full_page(ring->page_pool, buffer->page, false); + + buffer->page = NULL; + buffer->page_dma = 0; + buffer->page_addr = NULL; + buffer->page_size = 0; + buffer->page_offset = 0; +} + +static int hbg_buffer_alloc_page(struct hbg_buffer *buffer) +{ + struct hbg_ring *ring = buffer->ring; + u32 len = hbg_get_page_size(ring); + u32 offset; + + if (unlikely(!ring->page_pool)) + return 0; + + buffer->page = page_pool_dev_alloc_frag(ring->page_pool, &offset, len); + if (unlikely(!buffer->page)) + return -ENOMEM; + + buffer->page_dma = page_pool_get_dma_addr(buffer->page) + offset; + buffer->page_addr = page_address(buffer->page) + offset; + buffer->page_size = len; + buffer->page_offset = offset; + + return 0; +} + static void hbg_init_tx_desc(struct hbg_buffer *buffer, struct hbg_tx_desc *tx_desc) { @@ -138,24 +180,14 @@ static void hbg_buffer_free_skb(struct hbg_buffer *buffer) buffer->skb = NULL; } -static int hbg_buffer_alloc_skb(struct hbg_buffer *buffer) -{ - u32 len = hbg_spec_max_frame_len(buffer->priv, buffer->dir); - struct hbg_priv *priv = buffer->priv; - - buffer->skb = netdev_alloc_skb(priv->netdev, len); - if (unlikely(!buffer->skb)) - return -ENOMEM; - - buffer->skb_len = len; - memset(buffer->skb->data, 0, HBG_PACKET_HEAD_SIZE); - return 0; -} - static void hbg_buffer_free(struct hbg_buffer *buffer) { - hbg_dma_unmap(buffer); - hbg_buffer_free_skb(buffer); + if (buffer->skb) { + hbg_dma_unmap(buffer); + return hbg_buffer_free_skb(buffer); + } + + hbg_buffer_free_page(buffer); } static int hbg_napi_tx_recycle(struct napi_struct *napi, int budget) @@ -382,17 +414,15 @@ static int hbg_rx_fill_one_buffer(struct hbg_priv *priv) return 0; buffer = &ring->queue[ring->ntu]; - ret = hbg_buffer_alloc_skb(buffer); + ret = hbg_buffer_alloc_page(buffer); if (unlikely(ret)) return ret; - ret = hbg_dma_map(buffer); - if (unlikely(ret)) { - hbg_buffer_free_skb(buffer); - return ret; - } + memset(buffer->page_addr, 0, HBG_PACKET_HEAD_SIZE); + dma_sync_single_for_device(&priv->pdev->dev, buffer->page_dma, + HBG_PACKET_HEAD_SIZE, DMA_TO_DEVICE); - hbg_hw_fill_buffer(priv, buffer->skb_dma); + hbg_hw_fill_buffer(priv, buffer->page_dma); hbg_queue_move_next(ntu, ring); return 0; } @@ -425,13 +455,29 @@ static bool hbg_sync_data_from_hw(struct hbg_priv *priv, /* make sure HW write desc complete */ dma_rmb(); - dma_sync_single_for_cpu(&priv->pdev->dev, buffer->skb_dma, - buffer->skb_len, DMA_FROM_DEVICE); + dma_sync_single_for_cpu(&priv->pdev->dev, buffer->page_dma, + buffer->page_size, DMA_FROM_DEVICE); - rx_desc = (struct hbg_rx_desc *)buffer->skb->data; + rx_desc = (struct hbg_rx_desc *)buffer->page_addr; return FIELD_GET(HBG_RX_DESC_W2_PKT_LEN_M, rx_desc->word2) != 0; } +static int hbg_build_skb(struct hbg_priv *priv, + struct hbg_buffer *buffer, u32 pkt_len) +{ + net_prefetch(buffer->page_addr); + + buffer->skb = napi_build_skb(buffer->page_addr, buffer->page_size); + if (unlikely(!buffer->skb)) + return -ENOMEM; + skb_mark_for_recycle(buffer->skb); + + /* page will be freed together with the skb */ + buffer->page = NULL; + + return 0; +} + static int hbg_napi_rx_poll(struct napi_struct *napi, int budget) { struct hbg_ring *ring = container_of(napi, struct hbg_ring, napi); @@ -447,29 +493,33 @@ static int hbg_napi_rx_poll(struct napi_struct *napi, int budget) break; buffer = &ring->queue[ring->ntc]; - if (unlikely(!buffer->skb)) + if (unlikely(!buffer->page)) goto next_buffer; if (unlikely(!hbg_sync_data_from_hw(priv, buffer))) break; - rx_desc = (struct hbg_rx_desc *)buffer->skb->data; + rx_desc = (struct hbg_rx_desc *)buffer->page_addr; pkt_len = FIELD_GET(HBG_RX_DESC_W2_PKT_LEN_M, rx_desc->word2); trace_hbg_rx_desc(priv, ring->ntc, rx_desc); - if (unlikely(!hbg_rx_pkt_check(priv, rx_desc, buffer->skb))) { - hbg_buffer_free(buffer); + if (unlikely(hbg_build_skb(priv, buffer, pkt_len))) { + hbg_buffer_free_page(buffer); + goto next_buffer; + } + + if (unlikely(!hbg_rx_pkt_check(priv, rx_desc, buffer->skb))) { + hbg_buffer_free_skb(buffer); goto next_buffer; } - hbg_dma_unmap(buffer); skb_reserve(buffer->skb, HBG_PACKET_HEAD_SIZE + NET_IP_ALIGN); skb_put(buffer->skb, pkt_len); buffer->skb->protocol = eth_type_trans(buffer->skb, priv->netdev); - dev_sw_netstats_rx_add(priv->netdev, pkt_len); napi_gro_receive(napi, buffer->skb); buffer->skb = NULL; + buffer->page = NULL; next_buffer: hbg_rx_fill_one_buffer(priv); @@ -484,6 +534,42 @@ static int hbg_napi_rx_poll(struct napi_struct *napi, int budget) return packet_done; } +static void hbg_ring_page_pool_destory(struct hbg_ring *ring) +{ + if (!ring->page_pool) + return; + + page_pool_destroy(ring->page_pool); + ring->page_pool = NULL; +} + +static int hbg_ring_page_pool_init(struct hbg_priv *priv, struct hbg_ring *ring) +{ + u32 buf_size = hbg_spec_max_frame_len(priv, ring->dir); + struct page_pool_params pp_params = { + .flags = PP_FLAG_DMA_MAP | PP_FLAG_DMA_SYNC_DEV, + .order = hbg_get_page_order(ring), + .pool_size = ring->len * buf_size / hbg_get_page_size(ring), + .nid = dev_to_node(&priv->pdev->dev), + .dev = &priv->pdev->dev, + .napi = &ring->napi, + .dma_dir = DMA_FROM_DEVICE, + .offset = 0, + .max_len = hbg_get_page_size(ring), + }; + int ret = 0; + + ring->page_pool = page_pool_create(&pp_params); + if (IS_ERR(ring->page_pool)) { + ret = PTR_ERR(ring->page_pool); + dev_err(&priv->pdev->dev, + "failed to create page pool, ret = %d\n", ret); + ring->page_pool = NULL; + } + + return ret; +} + static void hbg_ring_uninit(struct hbg_ring *ring) { struct hbg_buffer *buffer; @@ -502,6 +588,7 @@ static void hbg_ring_uninit(struct hbg_ring *ring) buffer->priv = NULL; } + hbg_ring_page_pool_destory(ring); dma_free_coherent(&ring->priv->pdev->dev, ring->len * sizeof(*ring->queue), ring->queue, ring->queue_dma); @@ -517,6 +604,7 @@ static int hbg_ring_init(struct hbg_priv *priv, struct hbg_ring *ring, { struct hbg_buffer *buffer; u32 i, len; + int ret; len = hbg_get_spec_fifo_max_num(priv, dir) + 1; /* To improve receiving performance under high-stress scenarios, @@ -550,11 +638,23 @@ static int hbg_ring_init(struct hbg_priv *priv, struct hbg_ring *ring, ring->ntu = 0; ring->len = len; - if (dir == HBG_DIR_TX) + if (dir == HBG_DIR_TX) { netif_napi_add_tx(priv->netdev, &ring->napi, napi_poll); - else + } else { netif_napi_add(priv->netdev, &ring->napi, napi_poll); + ret = hbg_ring_page_pool_init(priv, ring); + if (ret) { + netif_napi_del(&ring->napi); + dma_free_coherent(&ring->priv->pdev->dev, + ring->len * sizeof(*ring->queue), + ring->queue, ring->queue_dma); + ring->queue = NULL; + ring->len = 0; + return ret; + } + } + napi_enable(&ring->napi); return 0; } From f93505f35745637b6d94efe8effa97ef26819784 Mon Sep 17 00:00:00 2001 From: Raju Rangoju Date: Sun, 23 Nov 2025 22:07:21 +0530 Subject: [PATCH 1199/1381] amd-xgbe: let the MAC manage PHY PM Use the MAC managed PM flag to indicate that MAC driver takes care of suspending/resuming the PHY, and reset it when the device is brought up. Signed-off-by: Raju Rangoju Link: https://patch.msgid.link/20251123163721.442162-1-Raju.Rangoju@amd.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/amd/xgbe/xgbe-drv.c | 14 +++++++++----- drivers/net/ethernet/amd/xgbe/xgbe-phy-v2.c | 1 + 2 files changed, 10 insertions(+), 5 deletions(-) diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c index f3adf29b222b..0653e69f0ef7 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c @@ -1259,6 +1259,11 @@ static int xgbe_start(struct xgbe_prv_data *pdata) udp_tunnel_nic_reset_ntf(netdev); + /* Reset the phy settings */ + ret = xgbe_phy_reset(pdata); + if (ret) + goto err_txrx; + netif_tx_start_all_queues(netdev); xgbe_start_timers(pdata); @@ -1268,6 +1273,10 @@ static int xgbe_start(struct xgbe_prv_data *pdata) return 0; +err_txrx: + hw_if->disable_rx(pdata); + hw_if->disable_tx(pdata); + err_irqs: xgbe_free_irqs(pdata); @@ -1574,11 +1583,6 @@ static int xgbe_open(struct net_device *netdev) goto err_dev_wq; } - /* Reset the phy settings */ - ret = xgbe_phy_reset(pdata); - if (ret) - goto err_an_wq; - /* Enable the clocks */ ret = clk_prepare_enable(pdata->sysclk); if (ret) { diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-phy-v2.c b/drivers/net/ethernet/amd/xgbe/xgbe-phy-v2.c index 35a381a83647..a68757e8fd22 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-phy-v2.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-phy-v2.c @@ -989,6 +989,7 @@ static int xgbe_phy_find_phy_device(struct xgbe_prv_data *pdata) return ret; } phy_data->phydev = phydev; + phy_data->phydev->mac_managed_pm = true; xgbe_phy_external_phy_quirks(pdata); From e6c43c95009035a63091cd49736886f883127510 Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Fri, 21 Nov 2025 08:39:55 -0800 Subject: [PATCH 1200/1381] net: phy: Add MDIO_PMA_CTRL1_SPEED for 2.5G and 5G to reflect PMA values The 2.5G and 5G values are not consistent between the PCS CTRL1 and PMA CTRL1 values. In order to avoid confusion between the two I am updating the values to include "PMA" in the name similar to values used in similar places. To avoid breaking UAPI I have retained the original macros and just defined them as the new PMA based defines. Signed-off-by: Alexander Duyck Link: https://patch.msgid.link/176374319569.959489.6610469879021800710.stgit@ahduyck-xeon-server.home.arpa Signed-off-by: Paolo Abeni --- drivers/net/phy/phy-c45.c | 8 ++++---- include/uapi/linux/mdio.h | 12 ++++++++++-- 2 files changed, 14 insertions(+), 6 deletions(-) diff --git a/drivers/net/phy/phy-c45.c b/drivers/net/phy/phy-c45.c index e8e5be4684ab..f5e23b53994f 100644 --- a/drivers/net/phy/phy-c45.c +++ b/drivers/net/phy/phy-c45.c @@ -148,12 +148,12 @@ int genphy_c45_pma_setup_forced(struct phy_device *phydev) ctrl2 |= MDIO_PMA_CTRL2_1000BT; break; case SPEED_2500: - ctrl1 |= MDIO_CTRL1_SPEED2_5G; + ctrl1 |= MDIO_PMA_CTRL1_SPEED2_5G; /* Assume 2.5Gbase-T */ ctrl2 |= MDIO_PMA_CTRL2_2_5GBT; break; case SPEED_5000: - ctrl1 |= MDIO_CTRL1_SPEED5G; + ctrl1 |= MDIO_PMA_CTRL1_SPEED5G; /* Assume 5Gbase-T */ ctrl2 |= MDIO_PMA_CTRL2_5GBT; break; @@ -618,10 +618,10 @@ int genphy_c45_read_pma(struct phy_device *phydev) case MDIO_PMA_CTRL1_SPEED1000: phydev->speed = SPEED_1000; break; - case MDIO_CTRL1_SPEED2_5G: + case MDIO_PMA_CTRL1_SPEED2_5G: phydev->speed = SPEED_2500; break; - case MDIO_CTRL1_SPEED5G: + case MDIO_PMA_CTRL1_SPEED5G: phydev->speed = SPEED_5000; break; case MDIO_CTRL1_SPEED10G: diff --git a/include/uapi/linux/mdio.h b/include/uapi/linux/mdio.h index 6975f182b22c..9ee6eeae64b8 100644 --- a/include/uapi/linux/mdio.h +++ b/include/uapi/linux/mdio.h @@ -116,10 +116,18 @@ #define MDIO_CTRL1_SPEED10G (MDIO_CTRL1_SPEEDSELEXT | 0x00) /* 10PASS-TS/2BASE-TL */ #define MDIO_CTRL1_SPEED10P2B (MDIO_CTRL1_SPEEDSELEXT | 0x04) +/* Note: the MDIO_CTRL1_SPEED_XXX values for everything past 10PASS-TS/2BASE-TL + * do not match between the PCS and PMA values. Any additions past this point + * should be PMA or PCS specific. The following 2 defines are workarounds for + * values added before this was caught. They should be considered deprecated. + */ +#define MDIO_CTRL1_SPEED2_5G MDIO_PMA_CTRL1_SPEED2_5G +#define MDIO_CTRL1_SPEED5G MDIO_PMA_CTRL1_SPEED5G /* 2.5 Gb/s */ -#define MDIO_CTRL1_SPEED2_5G (MDIO_CTRL1_SPEEDSELEXT | 0x18) +#define MDIO_PMA_CTRL1_SPEED2_5G (MDIO_CTRL1_SPEEDSELEXT | 0x18) /* 5 Gb/s */ -#define MDIO_CTRL1_SPEED5G (MDIO_CTRL1_SPEEDSELEXT | 0x1c) +#define MDIO_PMA_CTRL1_SPEED5G (MDIO_CTRL1_SPEEDSELEXT | 0x1c) + /* Status register 1. */ #define MDIO_STAT1_LPOWERABLE 0x0002 /* Low-power ability */ From 7622d55276932bfeb947b7b6cbf7ea0aa41feeb8 Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Fri, 21 Nov 2025 08:40:02 -0800 Subject: [PATCH 1201/1381] net: pcs: xpcs: Add support for 25G, 50G, and 100G interfaces With this change we are adding support for 25G, 50G, and 100G interface types to the XPCS driver. This had supposedly been enabled with the addition of XLGMII but I don't see any capability for configuration there so I suspect it may need to be refactored in the future. With this change we can enable the XPCS driver with the selected interface and it should be able to detect link, speed, and report the link status to the phylink interface. Signed-off-by: Alexander Duyck Link: https://patch.msgid.link/176374320248.959489.11649590675011158859.stgit@ahduyck-xeon-server.home.arpa Signed-off-by: Paolo Abeni --- drivers/net/pcs/pcs-xpcs.c | 105 +++++++++++++++++++++++++++++++++++-- include/uapi/linux/mdio.h | 6 +++ 2 files changed, 107 insertions(+), 4 deletions(-) diff --git a/drivers/net/pcs/pcs-xpcs.c b/drivers/net/pcs/pcs-xpcs.c index 3d1bd5aac093..9fb9d4fd2a5b 100644 --- a/drivers/net/pcs/pcs-xpcs.c +++ b/drivers/net/pcs/pcs-xpcs.c @@ -37,6 +37,16 @@ static const int xpcs_10gkr_features[] = { __ETHTOOL_LINK_MODE_MASK_NBITS, }; +static const int xpcs_25gbaser_features[] = { + ETHTOOL_LINK_MODE_MII_BIT, + ETHTOOL_LINK_MODE_Pause_BIT, + ETHTOOL_LINK_MODE_Asym_Pause_BIT, + ETHTOOL_LINK_MODE_25000baseCR_Full_BIT, + ETHTOOL_LINK_MODE_25000baseKR_Full_BIT, + ETHTOOL_LINK_MODE_25000baseSR_Full_BIT, + __ETHTOOL_LINK_MODE_MASK_NBITS, +}; + static const int xpcs_xlgmii_features[] = { ETHTOOL_LINK_MODE_Pause_BIT, ETHTOOL_LINK_MODE_Asym_Pause_BIT, @@ -67,6 +77,40 @@ static const int xpcs_xlgmii_features[] = { __ETHTOOL_LINK_MODE_MASK_NBITS, }; +static const int xpcs_50gbaser_features[] = { + ETHTOOL_LINK_MODE_MII_BIT, + ETHTOOL_LINK_MODE_Pause_BIT, + ETHTOOL_LINK_MODE_Asym_Pause_BIT, + ETHTOOL_LINK_MODE_50000baseKR_Full_BIT, + ETHTOOL_LINK_MODE_50000baseSR_Full_BIT, + ETHTOOL_LINK_MODE_50000baseCR_Full_BIT, + ETHTOOL_LINK_MODE_50000baseLR_ER_FR_Full_BIT, + ETHTOOL_LINK_MODE_50000baseDR_Full_BIT, + __ETHTOOL_LINK_MODE_MASK_NBITS, +}; + +static const int xpcs_50gbaser2_features[] = { + ETHTOOL_LINK_MODE_MII_BIT, + ETHTOOL_LINK_MODE_Pause_BIT, + ETHTOOL_LINK_MODE_Asym_Pause_BIT, + ETHTOOL_LINK_MODE_50000baseCR2_Full_BIT, + ETHTOOL_LINK_MODE_50000baseKR2_Full_BIT, + ETHTOOL_LINK_MODE_50000baseSR2_Full_BIT, + __ETHTOOL_LINK_MODE_MASK_NBITS, +}; + +static const int xpcs_100gbasep_features[] = { + ETHTOOL_LINK_MODE_MII_BIT, + ETHTOOL_LINK_MODE_Pause_BIT, + ETHTOOL_LINK_MODE_Asym_Pause_BIT, + ETHTOOL_LINK_MODE_100000baseKR2_Full_BIT, + ETHTOOL_LINK_MODE_100000baseSR2_Full_BIT, + ETHTOOL_LINK_MODE_100000baseCR2_Full_BIT, + ETHTOOL_LINK_MODE_100000baseLR2_ER2_FR2_Full_BIT, + ETHTOOL_LINK_MODE_100000baseDR2_Full_BIT, + __ETHTOOL_LINK_MODE_MASK_NBITS, +}; + static const int xpcs_10gbaser_features[] = { ETHTOOL_LINK_MODE_Pause_BIT, ETHTOOL_LINK_MODE_Asym_Pause_BIT, @@ -523,9 +567,38 @@ static int xpcs_get_max_xlgmii_speed(struct dw_xpcs *xpcs, return speed; } -static void xpcs_resolve_pma(struct dw_xpcs *xpcs, - struct phylink_link_state *state) +static int xpcs_c45_read_pcs_speed(struct dw_xpcs *xpcs, + struct phylink_link_state *state) { + int pcs_ctrl1; + + pcs_ctrl1 = xpcs_read(xpcs, MDIO_MMD_PCS, MDIO_CTRL1); + if (pcs_ctrl1 < 0) + return pcs_ctrl1; + + switch (pcs_ctrl1 & MDIO_CTRL1_SPEEDSEL) { + case MDIO_PCS_CTRL1_SPEED25G: + state->speed = SPEED_25000; + break; + case MDIO_PCS_CTRL1_SPEED50G: + state->speed = SPEED_50000; + break; + case MDIO_PCS_CTRL1_SPEED100G: + state->speed = SPEED_100000; + break; + default: + state->speed = SPEED_UNKNOWN; + break; + } + + return 0; +} + +static int xpcs_resolve_pma(struct dw_xpcs *xpcs, + struct phylink_link_state *state) +{ + int err = 0; + state->pause = MLO_PAUSE_TX | MLO_PAUSE_RX; state->duplex = DUPLEX_FULL; @@ -536,10 +609,18 @@ static void xpcs_resolve_pma(struct dw_xpcs *xpcs, case PHY_INTERFACE_MODE_XLGMII: state->speed = xpcs_get_max_xlgmii_speed(xpcs, state); break; + case PHY_INTERFACE_MODE_100GBASEP: + case PHY_INTERFACE_MODE_LAUI: + case PHY_INTERFACE_MODE_50GBASER: + case PHY_INTERFACE_MODE_25GBASER: + err = xpcs_c45_read_pcs_speed(xpcs, state); + break; default: state->speed = SPEED_UNKNOWN; break; } + + return err; } static int xpcs_validate(struct phylink_pcs *pcs, unsigned long *supported, @@ -945,10 +1026,10 @@ static int xpcs_get_state_c73(struct dw_xpcs *xpcs, phylink_resolve_c73(state); } else { - xpcs_resolve_pma(xpcs, state); + ret = xpcs_resolve_pma(xpcs, state); } - return 0; + return ret; } static int xpcs_get_state_c37_sgmii(struct dw_xpcs *xpcs, @@ -1312,10 +1393,26 @@ static const struct dw_xpcs_compat synopsys_xpcs_compat[] = { .interface = PHY_INTERFACE_MODE_10GKR, .supported = xpcs_10gkr_features, .an_mode = DW_AN_C73, + }, { + .interface = PHY_INTERFACE_MODE_25GBASER, + .supported = xpcs_25gbaser_features, + .an_mode = DW_AN_C73, }, { .interface = PHY_INTERFACE_MODE_XLGMII, .supported = xpcs_xlgmii_features, .an_mode = DW_AN_C73, + }, { + .interface = PHY_INTERFACE_MODE_50GBASER, + .supported = xpcs_50gbaser_features, + .an_mode = DW_AN_C73, + }, { + .interface = PHY_INTERFACE_MODE_LAUI, + .supported = xpcs_50gbaser2_features, + .an_mode = DW_AN_C73, + }, { + .interface = PHY_INTERFACE_MODE_100GBASEP, + .supported = xpcs_100gbasep_features, + .an_mode = DW_AN_C73, }, { .interface = PHY_INTERFACE_MODE_10GBASER, .supported = xpcs_10gbaser_features, diff --git a/include/uapi/linux/mdio.h b/include/uapi/linux/mdio.h index 9ee6eeae64b8..f23cab33e586 100644 --- a/include/uapi/linux/mdio.h +++ b/include/uapi/linux/mdio.h @@ -123,6 +123,12 @@ */ #define MDIO_CTRL1_SPEED2_5G MDIO_PMA_CTRL1_SPEED2_5G #define MDIO_CTRL1_SPEED5G MDIO_PMA_CTRL1_SPEED5G +/* 100 Gb/s */ +#define MDIO_PCS_CTRL1_SPEED100G (MDIO_CTRL1_SPEEDSELEXT | 0x10) +/* 25 Gb/s */ +#define MDIO_PCS_CTRL1_SPEED25G (MDIO_CTRL1_SPEEDSELEXT | 0x14) +/* 50 Gb/s */ +#define MDIO_PCS_CTRL1_SPEED50G (MDIO_CTRL1_SPEEDSELEXT | 0x18) /* 2.5 Gb/s */ #define MDIO_PMA_CTRL1_SPEED2_5G (MDIO_CTRL1_SPEEDSELEXT | 0x18) /* 5 Gb/s */ From 39e138173ae7641e952b456d2de7ad2ac03e8d88 Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Fri, 21 Nov 2025 08:40:09 -0800 Subject: [PATCH 1202/1381] net: pcs: xpcs: Fix PMA identifier handling in XPCS The XPCS driver was mangling the PMA identifier as the original code appears to have been focused on just capturing the OUI. Rather than store a mangled ID it is better to work with the actual PMA ID and instead just mask out the values that don't apply rather than shifting them and reordering them as you still don't get the original OUI for the NIC without having to bitswap the values as per the definition of the layout in IEEE 802.3-2022 22.2.4.3.1. By laying it out as it was in the hardware it is also less likely for us to have an unintentional collision as the enum values will occupy the revision number area while the OUI occupies the upper 22 bits. Signed-off-by: Alexander Duyck Link: https://patch.msgid.link/176374320920.959489.17267159479370601070.stgit@ahduyck-xeon-server.home.arpa Signed-off-by: Paolo Abeni --- drivers/net/pcs/pcs-xpcs.c | 9 ++++----- include/linux/pcs/pcs-xpcs.h | 2 +- include/uapi/linux/mdio.h | 5 +++++ 3 files changed, 10 insertions(+), 6 deletions(-) diff --git a/drivers/net/pcs/pcs-xpcs.c b/drivers/net/pcs/pcs-xpcs.c index 9fb9d4fd2a5b..a94a7cb93664 100644 --- a/drivers/net/pcs/pcs-xpcs.c +++ b/drivers/net/pcs/pcs-xpcs.c @@ -1365,17 +1365,16 @@ static int xpcs_read_ids(struct dw_xpcs *xpcs) if (ret < 0) return ret; - id = ret; + id = ret << 16; ret = xpcs_read(xpcs, MDIO_MMD_PMAPMD, MDIO_DEVID2); if (ret < 0) return ret; - /* Note the inverted dword order and masked out Model/Revision numbers - * with respect to what is done with the PCS ID... + /* For now we only record the OUI for the PMAPMD, we may want to + * add the model number at some point in the future. */ - ret = (ret >> 10) & 0x3F; - id |= ret << 16; + id |= ret & MDIO_DEVID2_OUI; /* Set the PMA ID if it hasn't been pre-initialized */ if (xpcs->info.pma == DW_XPCS_PMA_ID_NATIVE) diff --git a/include/linux/pcs/pcs-xpcs.h b/include/linux/pcs/pcs-xpcs.h index e40f554ff717..4cf6bd611e5a 100644 --- a/include/linux/pcs/pcs-xpcs.h +++ b/include/linux/pcs/pcs-xpcs.h @@ -38,7 +38,7 @@ enum dw_xpcs_pma_id { DW_XPCS_PMA_GEN4_6G_ID, DW_XPCS_PMA_GEN5_10G_ID, DW_XPCS_PMA_GEN5_12G_ID, - WX_TXGBE_XPCS_PMA_10G_ID = 0x0018fc80, + WX_TXGBE_XPCS_PMA_10G_ID = 0xfc806000, }; struct dw_xpcs_info { diff --git a/include/uapi/linux/mdio.h b/include/uapi/linux/mdio.h index f23cab33e586..8d769f100de6 100644 --- a/include/uapi/linux/mdio.h +++ b/include/uapi/linux/mdio.h @@ -147,6 +147,11 @@ #define MDIO_AN_STAT1_PAGE 0x0040 /* Page received */ #define MDIO_AN_STAT1_XNP 0x0080 /* Extended next page status */ +/* Device Identifier 2 */ +#define MDIO_DEVID2_OUI 0xfc00 /* OUI Portion of PHY ID */ +#define MDIO_DEVID2_MODEL_NUM 0x03f0 /* Manufacturer's Model Number */ +#define MDIO_DEVID2_REV_NUM 0x000f /* Revision Number */ + /* Speed register. */ #define MDIO_SPEED_10G 0x0001 /* 10G capable */ #define MDIO_PMA_SPEED_2B 0x0002 /* 2BASE-TL capable */ From 3f29dd34f75a09ee7f8333305618edb44617d835 Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Fri, 21 Nov 2025 08:40:16 -0800 Subject: [PATCH 1203/1381] net: pcs: xpcs: Add support for FBNIC 25G, 50G, 100G PMD The fbnic driver is planning to make use of the XPCS driver to enable support for PCS and better integration with phylink. To do this though we will need to enable several workarounds since the PMD interface for fbnic is likely to be unique since it is a mix of two different vendor products with a unique wrapper around the IP. I have generated a PHY identifier based on IEEE 802.3-2022 22.2.4.3.1 using an OUI belonging to Meta Platforms and used with our NICs. Using this we will provide it as the PMD ID via the SW based MDIO interface so that the fbnic device can be identified and necessary workarounds enabled in the XPCS driver. As an initial workaround this change adds an exception so that soft_reset is not set when the driver is initially bound to the PCS. In addition I have added logic to integrate the PMD Rx signal detect state into the link state for the PCS. With this we can avoid the link coming up too soon on the FBNIC PMD and as a result of it being in the training state so we can avoid link flaps. Signed-off-by: Alexander Duyck Link: https://patch.msgid.link/176374321695.959489.6648161125012056619.stgit@ahduyck-xeon-server.home.arpa Signed-off-by: Paolo Abeni --- drivers/net/pcs/pcs-xpcs.c | 24 ++++++++++++++++++++++-- include/linux/pcs/pcs-xpcs.h | 2 ++ 2 files changed, 24 insertions(+), 2 deletions(-) diff --git a/drivers/net/pcs/pcs-xpcs.c b/drivers/net/pcs/pcs-xpcs.c index a94a7cb93664..9679f2b35a44 100644 --- a/drivers/net/pcs/pcs-xpcs.c +++ b/drivers/net/pcs/pcs-xpcs.c @@ -597,7 +597,26 @@ static int xpcs_c45_read_pcs_speed(struct dw_xpcs *xpcs, static int xpcs_resolve_pma(struct dw_xpcs *xpcs, struct phylink_link_state *state) { - int err = 0; + int pmd_rxdet, err = 0; + + /* The Meta Platforms FBNIC PMD will go into a training state for + * about 4 seconds when the link first comes up. During this time the + * PCS link will bounce. To avoid reporting link up too soon we include + * the PMD state provided by the driver. + */ + if (xpcs->info.pma == MP_FBNIC_XPCS_PMA_100G_ID) { + pmd_rxdet = xpcs_read(xpcs, MDIO_MMD_PMAPMD, MDIO_PMA_RXDET); + if (pmd_rxdet < 0) { + state->link = false; + return pmd_rxdet; + } + + /* Verify Rx lanes are trained before reporting link up */ + if (!(pmd_rxdet & MDIO_PMD_RXDET_GLOBAL)) { + state->link = false; + return 0; + } + } state->pause = MLO_PAUSE_TX | MLO_PAUSE_RX; state->duplex = DUPLEX_FULL; @@ -1591,7 +1610,8 @@ static struct dw_xpcs *xpcs_create(struct mdio_device *mdiodev) xpcs_get_interfaces(xpcs, xpcs->pcs.supported_interfaces); - if (xpcs->info.pma == WX_TXGBE_XPCS_PMA_10G_ID) + if (xpcs->info.pma == WX_TXGBE_XPCS_PMA_10G_ID || + xpcs->info.pma == MP_FBNIC_XPCS_PMA_100G_ID) xpcs->pcs.poll = false; else xpcs->need_reset = true; diff --git a/include/linux/pcs/pcs-xpcs.h b/include/linux/pcs/pcs-xpcs.h index 4cf6bd611e5a..36073f7b6bb4 100644 --- a/include/linux/pcs/pcs-xpcs.h +++ b/include/linux/pcs/pcs-xpcs.h @@ -39,6 +39,8 @@ enum dw_xpcs_pma_id { DW_XPCS_PMA_GEN5_10G_ID, DW_XPCS_PMA_GEN5_12G_ID, WX_TXGBE_XPCS_PMA_10G_ID = 0xfc806000, + /* Meta Platforms OUI 88:25:08, model 0, revision 0 */ + MP_FBNIC_XPCS_PMA_100G_ID = 0x46904000, }; struct dw_xpcs_info { From f18dd1b15f7a0bea5743fcbf3a62b7d2940e2dd5 Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Fri, 21 Nov 2025 08:40:23 -0800 Subject: [PATCH 1204/1381] fbnic: Rename PCS IRQ to MAC IRQ as it is actually a MAC interrupt Throughout several spots in the code I had called out the IRQ as being related to the PCS. However the actual IRQ is a part of the MAC and it is just exposing PCS data. To more accurately reflect the owner of the calls this change makes it so that we rename the functions and values that are taking in the interrupt value and processing it to reflect that it is a MAC call and not a PCS one. This change is mostly motivated by the fact that we will be moving the handling of this interrupt from being PCS focused to being more PMA/PMD focused as this will drive the phydev driver that I am adding instead of driving the PCS directly. Signed-off-by: Alexander Duyck Link: https://patch.msgid.link/176374322373.959489.12018231545479053860.stgit@ahduyck-xeon-server.home.arpa Signed-off-by: Paolo Abeni --- drivers/net/ethernet/meta/fbnic/fbnic.h | 6 ++-- drivers/net/ethernet/meta/fbnic/fbnic_irq.c | 30 +++++++++---------- drivers/net/ethernet/meta/fbnic/fbnic_mac.c | 14 ++++----- drivers/net/ethernet/meta/fbnic/fbnic_mac.h | 8 ++--- .../net/ethernet/meta/fbnic/fbnic_netdev.c | 4 +-- .../net/ethernet/meta/fbnic/fbnic_phylink.c | 2 +- 6 files changed, 32 insertions(+), 32 deletions(-) diff --git a/drivers/net/ethernet/meta/fbnic/fbnic.h b/drivers/net/ethernet/meta/fbnic/fbnic.h index b03e5a3d5144..98929add5f21 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic.h +++ b/drivers/net/ethernet/meta/fbnic/fbnic.h @@ -34,7 +34,7 @@ struct fbnic_dev { u32 __iomem *uc_addr4; const struct fbnic_mac *mac; unsigned int fw_msix_vector; - unsigned int pcs_msix_vector; + unsigned int mac_msix_vector; unsigned short num_irqs; struct { @@ -175,8 +175,8 @@ void fbnic_fw_free_mbx(struct fbnic_dev *fbd); void fbnic_hwmon_register(struct fbnic_dev *fbd); void fbnic_hwmon_unregister(struct fbnic_dev *fbd); -int fbnic_pcs_request_irq(struct fbnic_dev *fbd); -void fbnic_pcs_free_irq(struct fbnic_dev *fbd); +int fbnic_mac_request_irq(struct fbnic_dev *fbd); +void fbnic_mac_free_irq(struct fbnic_dev *fbd); void fbnic_napi_name_irqs(struct fbnic_dev *fbd); int fbnic_napi_request_irq(struct fbnic_dev *fbd, diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_irq.c b/drivers/net/ethernet/meta/fbnic/fbnic_irq.c index 1c88a2bf3a7a..40947e142c5d 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_irq.c +++ b/drivers/net/ethernet/meta/fbnic/fbnic_irq.c @@ -118,12 +118,12 @@ void fbnic_fw_free_mbx(struct fbnic_dev *fbd) fbd->fw_msix_vector = 0; } -static irqreturn_t fbnic_pcs_msix_intr(int __always_unused irq, void *data) +static irqreturn_t fbnic_mac_msix_intr(int __always_unused irq, void *data) { struct fbnic_dev *fbd = data; struct fbnic_net *fbn; - if (fbd->mac->pcs_get_link_event(fbd) == FBNIC_LINK_EVENT_NONE) { + if (fbd->mac->get_link_event(fbd) == FBNIC_LINK_EVENT_NONE) { fbnic_wr32(fbd, FBNIC_INTR_MASK_CLEAR(0), 1u << FBNIC_PCS_MSIX_ENTRY); return IRQ_HANDLED; @@ -137,20 +137,20 @@ static irqreturn_t fbnic_pcs_msix_intr(int __always_unused irq, void *data) } /** - * fbnic_pcs_request_irq - Configure the PCS to enable it to advertise link + * fbnic_mac_request_irq - Configure the MAC to enable it to advertise link * @fbd: Pointer to device to initialize * - * This function provides basic bringup for the MAC/PCS IRQ. For now the IRQ + * This function provides basic bringup for the MAC/PHY IRQ. For now the IRQ * will remain disabled until we start the MAC/PCS/PHY logic via phylink. * * Return: non-zero on failure. **/ -int fbnic_pcs_request_irq(struct fbnic_dev *fbd) +int fbnic_mac_request_irq(struct fbnic_dev *fbd) { struct pci_dev *pdev = to_pci_dev(fbd->dev); int vector, err; - WARN_ON(fbd->pcs_msix_vector); + WARN_ON(fbd->mac_msix_vector); vector = pci_irq_vector(pdev, FBNIC_PCS_MSIX_ENTRY); if (vector < 0) @@ -159,7 +159,7 @@ int fbnic_pcs_request_irq(struct fbnic_dev *fbd) /* Request the IRQ for PCS link vector. * Map PCS cause to it, and unmask it */ - err = request_irq(vector, &fbnic_pcs_msix_intr, 0, + err = request_irq(vector, &fbnic_mac_msix_intr, 0, fbd->netdev->name, fbd); if (err) return err; @@ -168,22 +168,22 @@ int fbnic_pcs_request_irq(struct fbnic_dev *fbd) fbnic_wr32(fbd, FBNIC_INTR_MSIX_CTRL(FBNIC_INTR_MSIX_CTRL_PCS_IDX), FBNIC_PCS_MSIX_ENTRY | FBNIC_INTR_MSIX_CTRL_ENABLE); - fbd->pcs_msix_vector = vector; + fbd->mac_msix_vector = vector; return 0; } /** - * fbnic_pcs_free_irq - Teardown the PCS IRQ to prepare for stopping + * fbnic_mac_free_irq - Teardown the MAC IRQ to prepare for stopping * @fbd: Pointer to device that is stopping * - * This function undoes the work done in fbnic_pcs_request_irq and prepares + * This function undoes the work done in fbnic_mac_request_irq and prepares * the device to no longer receive traffic on the host interface. **/ -void fbnic_pcs_free_irq(struct fbnic_dev *fbd) +void fbnic_mac_free_irq(struct fbnic_dev *fbd) { /* Vector has already been freed */ - if (!fbd->pcs_msix_vector) + if (!fbd->mac_msix_vector) return; /* Disable interrupt */ @@ -192,14 +192,14 @@ void fbnic_pcs_free_irq(struct fbnic_dev *fbd) fbnic_wrfl(fbd); /* Synchronize IRQ to prevent race that would unmask vector */ - synchronize_irq(fbd->pcs_msix_vector); + synchronize_irq(fbd->mac_msix_vector); /* Mask the vector */ fbnic_wr32(fbd, FBNIC_INTR_MASK_SET(0), 1u << FBNIC_PCS_MSIX_ENTRY); /* Free the vector */ - free_irq(fbd->pcs_msix_vector, fbd); - fbd->pcs_msix_vector = 0; + free_irq(fbd->mac_msix_vector, fbd); + fbd->mac_msix_vector = 0; } void fbnic_synchronize_irq(struct fbnic_dev *fbd, int nr) diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_mac.c b/drivers/net/ethernet/meta/fbnic/fbnic_mac.c index 2a84bd1d7e26..28a2e1fd3760 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_mac.c +++ b/drivers/net/ethernet/meta/fbnic/fbnic_mac.c @@ -434,14 +434,14 @@ static void fbnic_mac_tx_pause_config(struct fbnic_dev *fbd, bool tx_pause) wr32(fbd, FBNIC_RXB_PAUSE_DROP_CTRL, rxb_pause_ctrl); } -static int fbnic_pcs_get_link_event_asic(struct fbnic_dev *fbd) +static int fbnic_mac_get_link_event(struct fbnic_dev *fbd) { - u32 pcs_intr_mask = rd32(fbd, FBNIC_SIG_PCS_INTR_STS); + u32 intr_mask = rd32(fbd, FBNIC_SIG_PCS_INTR_STS); - if (pcs_intr_mask & FBNIC_SIG_PCS_INTR_LINK_DOWN) + if (intr_mask & FBNIC_SIG_PCS_INTR_LINK_DOWN) return FBNIC_LINK_EVENT_DOWN; - return (pcs_intr_mask & FBNIC_SIG_PCS_INTR_LINK_UP) ? + return (intr_mask & FBNIC_SIG_PCS_INTR_LINK_UP) ? FBNIC_LINK_EVENT_UP : FBNIC_LINK_EVENT_NONE; } @@ -521,7 +521,7 @@ static bool fbnic_mac_get_pcs_link_status(struct fbnic_dev *fbd) return !lane_mask; } -static bool fbnic_pcs_get_link_asic(struct fbnic_dev *fbd) +static bool fbnic_mac_get_link(struct fbnic_dev *fbd) { bool link; @@ -869,8 +869,8 @@ static const struct fbnic_mac fbnic_mac_asic = { .init_regs = fbnic_mac_init_regs, .pcs_enable = fbnic_pcs_enable_asic, .pcs_disable = fbnic_pcs_disable_asic, - .pcs_get_link = fbnic_pcs_get_link_asic, - .pcs_get_link_event = fbnic_pcs_get_link_event_asic, + .get_link = fbnic_mac_get_link, + .get_link_event = fbnic_mac_get_link_event, .get_fec_stats = fbnic_mac_get_fec_stats, .get_pcs_stats = fbnic_mac_get_pcs_stats, .get_eth_mac_stats = fbnic_mac_get_eth_mac_stats, diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_mac.h b/drivers/net/ethernet/meta/fbnic/fbnic_mac.h index ede5ff0dae22..414c170abcba 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_mac.h +++ b/drivers/net/ethernet/meta/fbnic/fbnic_mac.h @@ -59,9 +59,9 @@ enum fbnic_sensor_id { * Configure and enable PCS to enable link if not already enabled * void (*pcs_disable)(struct fbnic_dev *fbd); * Shutdown the link if we are the only consumer of it. - * bool (*pcs_get_link)(struct fbnic_dev *fbd); + * bool (*get_link)(struct fbnic_dev *fbd); * Check PCS link status - * int (*pcs_get_link_event)(struct fbnic_dev *fbd) + * int (*get_link_event)(struct fbnic_dev *fbd) * Get the current link event status, reports true if link has * changed to either FBNIC_LINK_EVENT_DOWN or FBNIC_LINK_EVENT_UP * @@ -76,8 +76,8 @@ struct fbnic_mac { int (*pcs_enable)(struct fbnic_dev *fbd); void (*pcs_disable)(struct fbnic_dev *fbd); - bool (*pcs_get_link)(struct fbnic_dev *fbd); - int (*pcs_get_link_event)(struct fbnic_dev *fbd); + bool (*get_link)(struct fbnic_dev *fbd); + int (*get_link_event)(struct fbnic_dev *fbd); void (*get_fec_stats)(struct fbnic_dev *fbd, bool reset, struct fbnic_fec_stats *fec_stats); diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c index e95be0e7bd9e..2d5ae89b4a15 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c +++ b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c @@ -44,7 +44,7 @@ int __fbnic_open(struct fbnic_net *fbn) if (err) goto time_stop; - err = fbnic_pcs_request_irq(fbd); + err = fbnic_mac_request_irq(fbd); if (err) goto time_stop; @@ -89,7 +89,7 @@ static int fbnic_stop(struct net_device *netdev) phylink_suspend(fbn->phylink, fbnic_bmc_present(fbn->fbd)); fbnic_down(fbn); - fbnic_pcs_free_irq(fbn->fbd); + fbnic_mac_free_irq(fbn->fbd); fbnic_time_stop(fbn); fbnic_fw_xmit_ownership_msg(fbn->fbd, false); diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_phylink.c b/drivers/net/ethernet/meta/fbnic/fbnic_phylink.c index 62701923cfe9..99fa79acd2df 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_phylink.c +++ b/drivers/net/ethernet/meta/fbnic/fbnic_phylink.c @@ -132,7 +132,7 @@ fbnic_phylink_pcs_get_state(struct phylink_pcs *pcs, unsigned int neg_mode, state->duplex = DUPLEX_FULL; - state->link = fbd->mac->pcs_get_link(fbd); + state->link = fbd->mac->get_link(fbd); } static int From 9963117a2b9b29074763ccb62fe8ee4d78ca46f2 Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Fri, 21 Nov 2025 08:40:31 -0800 Subject: [PATCH 1205/1381] fbnic: Add logic to track PMD state via MAC/PCS signals One complication with the design of our part is that the PMD doesn't provide a direct signal to the host. Instead we have visibility to signals that the PCS provides to the MAC that allow us to check the link state through that. We will need to account for several things in the PMD and firmware when managing the link. Specifically when the link first starts to come up the PMD will cause the link to flap. This is due to the firmware starting a training cycle when the link is first detected. This will cause link flapping if we were to immediately report link up when the PCS first detects it. To address that we are adding a pmd_state variable that is meant to be a countdown of sorts indicating the state of the PMD. If the link is down or has been reconfigured the PMD will start out in the initialize state. By default the link is assumed to be in the SEND_DATA state if it is available on initial link inspection. If link is detected while in the initialize state the PMD state will switch to training, and if after 4 seconds the link is still stable we will transition to link_ready, and finally the send_data state. With this we can avoid link flapping when a cable is first connected to the NIC. One side effect of this is that we need to pull the link state away from the PCS. For now we use a union of the PCS link state register value and the pmd_state. The plan is to add a PMD register to report the pmd_state to the phylink interface. With that we can then look at switching over to the use of the XPCS driver for fbnic instead of having an internal one. Signed-off-by: Alexander Duyck Link: https://patch.msgid.link/176374323107.959489.14951134213387615059.stgit@ahduyck-xeon-server.home.arpa Signed-off-by: Paolo Abeni --- drivers/net/ethernet/meta/fbnic/fbnic.h | 4 + drivers/net/ethernet/meta/fbnic/fbnic_csr.h | 2 + drivers/net/ethernet/meta/fbnic/fbnic_irq.c | 4 +- drivers/net/ethernet/meta/fbnic/fbnic_mac.c | 69 ++++++++---- drivers/net/ethernet/meta/fbnic/fbnic_mac.h | 36 +++++-- .../net/ethernet/meta/fbnic/fbnic_netdev.c | 2 +- .../net/ethernet/meta/fbnic/fbnic_netdev.h | 2 +- drivers/net/ethernet/meta/fbnic/fbnic_pci.c | 6 +- .../net/ethernet/meta/fbnic/fbnic_phylink.c | 102 ++++++++++++++---- 9 files changed, 175 insertions(+), 52 deletions(-) diff --git a/drivers/net/ethernet/meta/fbnic/fbnic.h b/drivers/net/ethernet/meta/fbnic/fbnic.h index 98929add5f21..fac1283d0ade 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic.h +++ b/drivers/net/ethernet/meta/fbnic/fbnic.h @@ -83,6 +83,10 @@ struct fbnic_dev { /* Last @time_high refresh time in jiffies (to catch stalls) */ unsigned long last_read; + /* PMD specific data */ + unsigned long end_of_pmd_training; + u8 pmd_state; + /* Local copy of hardware statistics */ struct fbnic_hw_stats hw_stats; diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_csr.h b/drivers/net/ethernet/meta/fbnic/fbnic_csr.h index d3a7ad921f18..422265dc7abd 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_csr.h +++ b/drivers/net/ethernet/meta/fbnic/fbnic_csr.h @@ -787,6 +787,8 @@ enum { /* MAC PCS registers */ #define FBNIC_CSR_START_PCS 0x10000 /* CSR section delimiter */ +#define FBNIC_PCS_PAGE(n) (0x10000 + 0x400 * (n)) /* 0x40000 + 1024*n */ +#define FBNIC_PCS(reg, n) ((reg) + FBNIC_PCS_PAGE(n)) #define FBNIC_CSR_END_PCS 0x10668 /* CSR section delimiter */ #define FBNIC_CSR_START_RSFEC 0x10800 /* CSR section delimiter */ diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_irq.c b/drivers/net/ethernet/meta/fbnic/fbnic_irq.c index 40947e142c5d..9b068b82f30a 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_irq.c +++ b/drivers/net/ethernet/meta/fbnic/fbnic_irq.c @@ -131,7 +131,9 @@ static irqreturn_t fbnic_mac_msix_intr(int __always_unused irq, void *data) fbn = netdev_priv(fbd->netdev); - phylink_pcs_change(&fbn->phylink_pcs, false); + /* Record link down events */ + if (!fbd->mac->get_link(fbd, fbn->aui, fbn->fec)) + phylink_pcs_change(&fbn->phylink_pcs, false); return IRQ_HANDLED; } diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_mac.c b/drivers/net/ethernet/meta/fbnic/fbnic_mac.c index 28a2e1fd3760..fc7abea4ef5b 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_mac.c +++ b/drivers/net/ethernet/meta/fbnic/fbnic_mac.c @@ -466,9 +466,8 @@ static u32 __fbnic_mac_cmd_config_asic(struct fbnic_dev *fbd, return command_config; } -static bool fbnic_mac_get_pcs_link_status(struct fbnic_dev *fbd) +static bool fbnic_mac_get_link_status(struct fbnic_dev *fbd, u8 aui, u8 fec) { - struct fbnic_net *fbn = netdev_priv(fbd->netdev); u32 pcs_status, lane_mask = ~0; pcs_status = rd32(fbd, FBNIC_SIG_PCS_OUT0); @@ -476,7 +475,7 @@ static bool fbnic_mac_get_pcs_link_status(struct fbnic_dev *fbd) return false; /* Define the expected lane mask for the status bits we need to check */ - switch (fbn->aui) { + switch (aui) { case FBNIC_AUI_100GAUI2: lane_mask = 0xf; break; @@ -484,7 +483,7 @@ static bool fbnic_mac_get_pcs_link_status(struct fbnic_dev *fbd) lane_mask = 3; break; case FBNIC_AUI_LAUI2: - switch (fbn->fec) { + switch (fec) { case FBNIC_FEC_OFF: lane_mask = 0x63; break; @@ -502,7 +501,7 @@ static bool fbnic_mac_get_pcs_link_status(struct fbnic_dev *fbd) } /* Use an XOR to remove the bits we expect to see set */ - switch (fbn->fec) { + switch (fec) { case FBNIC_FEC_OFF: lane_mask ^= FIELD_GET(FBNIC_SIG_PCS_OUT0_BLOCK_LOCK, pcs_status); @@ -521,7 +520,46 @@ static bool fbnic_mac_get_pcs_link_status(struct fbnic_dev *fbd) return !lane_mask; } -static bool fbnic_mac_get_link(struct fbnic_dev *fbd) +static bool fbnic_pmd_update_state(struct fbnic_dev *fbd, bool signal_detect) +{ + /* Delay link up for 4 seconds to allow for link training. + * The state transitions for this are as follows: + * + * All states have the following two transitions in common: + * Loss of signal -> FBNIC_PMD_INITIALIZE + * The condition handled below (!signal) + * Reconfiguration -> FBNIC_PMD_INITIALIZE + * Occurs when mac_prepare starts a PHY reconfig + * FBNIC_PMD_TRAINING: + * signal still detected && 4s have passed -> Report link up + * When link is brought up in link_up -> FBNIC_PMD_SEND_DATA + * FBNIC_PMD_INITIALIZE: + * signal detected -> FBNIC_PMD_TRAINING + */ + if (!signal_detect) { + fbd->pmd_state = FBNIC_PMD_INITIALIZE; + return false; + } + + switch (fbd->pmd_state) { + case FBNIC_PMD_TRAINING: + return time_before(fbd->end_of_pmd_training, jiffies); + case FBNIC_PMD_LINK_READY: + case FBNIC_PMD_SEND_DATA: + return true; + } + + fbd->end_of_pmd_training = jiffies + 4 * HZ; + + /* Ensure end_of_training is visible before the state change */ + smp_wmb(); + + fbd->pmd_state = FBNIC_PMD_TRAINING; + + return false; +} + +static bool fbnic_mac_get_link(struct fbnic_dev *fbd, u8 aui, u8 fec) { bool link; @@ -538,7 +576,8 @@ static bool fbnic_mac_get_link(struct fbnic_dev *fbd) wr32(fbd, FBNIC_SIG_PCS_INTR_STS, FBNIC_SIG_PCS_INTR_LINK_DOWN | FBNIC_SIG_PCS_INTR_LINK_UP); - link = fbnic_mac_get_pcs_link_status(fbd); + link = fbnic_mac_get_link_status(fbd, aui, fec); + link = fbnic_pmd_update_state(fbd, link); /* Enable interrupt to only capture changes in link state */ wr32(fbd, FBNIC_SIG_PCS_INTR_MASK, @@ -586,20 +625,15 @@ void fbnic_mac_get_fw_settings(struct fbnic_dev *fbd, u8 *aui, u8 *fec) } } -static int fbnic_pcs_enable_asic(struct fbnic_dev *fbd) +static void fbnic_mac_prepare(struct fbnic_dev *fbd, u8 aui, u8 fec) { /* Mask and clear the PCS interrupt, will be enabled by link handler */ wr32(fbd, FBNIC_SIG_PCS_INTR_MASK, ~0); wr32(fbd, FBNIC_SIG_PCS_INTR_STS, ~0); - return 0; -} - -static void fbnic_pcs_disable_asic(struct fbnic_dev *fbd) -{ - /* Mask and clear the PCS interrupt */ - wr32(fbd, FBNIC_SIG_PCS_INTR_MASK, ~0); - wr32(fbd, FBNIC_SIG_PCS_INTR_STS, ~0); + /* If we don't have link tear it all down and start over */ + if (!fbnic_mac_get_link_status(fbd, aui, fec)) + fbd->pmd_state = FBNIC_PMD_INITIALIZE; } static void fbnic_mac_link_down_asic(struct fbnic_dev *fbd) @@ -867,10 +901,9 @@ static int fbnic_mac_get_sensor_asic(struct fbnic_dev *fbd, int id, static const struct fbnic_mac fbnic_mac_asic = { .init_regs = fbnic_mac_init_regs, - .pcs_enable = fbnic_pcs_enable_asic, - .pcs_disable = fbnic_pcs_disable_asic, .get_link = fbnic_mac_get_link, .get_link_event = fbnic_mac_get_link_event, + .prepare = fbnic_mac_prepare, .get_fec_stats = fbnic_mac_get_fec_stats, .get_pcs_stats = fbnic_mac_get_pcs_stats, .get_eth_mac_stats = fbnic_mac_get_eth_mac_stats, diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_mac.h b/drivers/net/ethernet/meta/fbnic/fbnic_mac.h index 414c170abcba..9b4fb0586dff 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_mac.h +++ b/drivers/net/ethernet/meta/fbnic/fbnic_mac.h @@ -10,6 +10,24 @@ struct fbnic_dev; #define FBNIC_MAX_JUMBO_FRAME_SIZE 9742 +/* States loosely based on section 136.8.11.7.5 of IEEE 802.3-2022 Ethernet + * Standard. These are needed to track the state of the PHY as it has a delay + * of several seconds from the time link comes up until it has completed + * training that we need to wait to report the link. + * + * Currently we treat training as a single block as this is managed by the + * firmware. + * + * We have FBNIC_PMD_SEND_DATA set to 0 as the expected default at driver load + * and we initialize the structure containing it to zero at allocation. + */ +enum { + FBNIC_PMD_SEND_DATA = 0x0, + FBNIC_PMD_INITIALIZE = 0x1, + FBNIC_PMD_TRAINING = 0x2, + FBNIC_PMD_LINK_READY = 0x3, +}; + enum { FBNIC_LINK_EVENT_NONE = 0, FBNIC_LINK_EVENT_UP = 1, @@ -55,15 +73,15 @@ enum fbnic_sensor_id { * void (*init_regs)(struct fbnic_dev *fbd); * Initialize MAC registers to enable Tx/Rx paths and FIFOs. * - * void (*pcs_enable)(struct fbnic_dev *fbd); - * Configure and enable PCS to enable link if not already enabled - * void (*pcs_disable)(struct fbnic_dev *fbd); - * Shutdown the link if we are the only consumer of it. - * bool (*get_link)(struct fbnic_dev *fbd); - * Check PCS link status * int (*get_link_event)(struct fbnic_dev *fbd) * Get the current link event status, reports true if link has * changed to either FBNIC_LINK_EVENT_DOWN or FBNIC_LINK_EVENT_UP + * bool (*get_link)(struct fbnic_dev *fbd, u8 aui, u8 fec); + * Check link status + * + * void (*prepare)(struct fbnic_dev *fbd, u8 aui, u8 fec); + * Prepare PHY for init by fetching settings, disabling interrupts, + * and sending an updated PHY config to FW if needed. * * void (*link_down)(struct fbnic_dev *fbd); * Configure MAC for link down event @@ -74,10 +92,10 @@ enum fbnic_sensor_id { struct fbnic_mac { void (*init_regs)(struct fbnic_dev *fbd); - int (*pcs_enable)(struct fbnic_dev *fbd); - void (*pcs_disable)(struct fbnic_dev *fbd); - bool (*get_link)(struct fbnic_dev *fbd); int (*get_link_event)(struct fbnic_dev *fbd); + bool (*get_link)(struct fbnic_dev *fbd, u8 aui, u8 fec); + + void (*prepare)(struct fbnic_dev *fbd, u8 aui, u8 fec); void (*get_fec_stats)(struct fbnic_dev *fbd, bool reset, struct fbnic_fec_stats *fec_stats); diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c index 2d5ae89b4a15..65318a5b466e 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c +++ b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c @@ -86,10 +86,10 @@ static int fbnic_stop(struct net_device *netdev) { struct fbnic_net *fbn = netdev_priv(netdev); + fbnic_mac_free_irq(fbn->fbd); phylink_suspend(fbn->phylink, fbnic_bmc_present(fbn->fbd)); fbnic_down(fbn); - fbnic_mac_free_irq(fbn->fbd); fbnic_time_stop(fbn); fbnic_fw_xmit_ownership_msg(fbn->fbd, false); diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.h b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.h index e6ca23a9957d..deab789b2a6c 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.h +++ b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.h @@ -109,7 +109,7 @@ int fbnic_phylink_ethtool_ksettings_get(struct net_device *netdev, int fbnic_phylink_get_fecparam(struct net_device *netdev, struct ethtool_fecparam *fecparam); int fbnic_phylink_init(struct net_device *netdev); - +void fbnic_phylink_pmd_training_complete_notify(struct net_device *netdev); bool fbnic_check_split_frames(struct bpf_prog *prog, unsigned int mtu, u32 hds_threshold); #endif /* _FBNIC_NETDEV_H_ */ diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_pci.c b/drivers/net/ethernet/meta/fbnic/fbnic_pci.c index 4620f1847f2e..dea5367d2190 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_pci.c +++ b/drivers/net/ethernet/meta/fbnic/fbnic_pci.c @@ -207,6 +207,10 @@ static void fbnic_service_task(struct work_struct *work) { struct fbnic_dev *fbd = container_of(to_delayed_work(work), struct fbnic_dev, service_task); + struct net_device *netdev = fbd->netdev; + + if (netif_running(netdev)) + fbnic_phylink_pmd_training_complete_notify(netdev); rtnl_lock(); @@ -224,7 +228,7 @@ static void fbnic_service_task(struct work_struct *work) netdev_unlock(fbd->netdev); } - if (netif_running(fbd->netdev)) + if (netif_running(netdev)) schedule_delayed_work(&fbd->service_task, HZ); rtnl_unlock(); diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_phylink.c b/drivers/net/ethernet/meta/fbnic/fbnic_phylink.c index 99fa79acd2df..20f88c8dcc79 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_phylink.c +++ b/drivers/net/ethernet/meta/fbnic/fbnic_phylink.c @@ -132,25 +132,9 @@ fbnic_phylink_pcs_get_state(struct phylink_pcs *pcs, unsigned int neg_mode, state->duplex = DUPLEX_FULL; - state->link = fbd->mac->get_link(fbd); -} - -static int -fbnic_phylink_pcs_enable(struct phylink_pcs *pcs) -{ - struct fbnic_net *fbn = fbnic_pcs_to_net(pcs); - struct fbnic_dev *fbd = fbn->fbd; - - return fbd->mac->pcs_enable(fbd); -} - -static void -fbnic_phylink_pcs_disable(struct phylink_pcs *pcs) -{ - struct fbnic_net *fbn = fbnic_pcs_to_net(pcs); - struct fbnic_dev *fbd = fbn->fbd; - - return fbd->mac->pcs_disable(fbd); + state->link = (fbd->pmd_state == FBNIC_PMD_SEND_DATA) && + (rd32(fbd, FBNIC_PCS(MDIO_STAT1, 0)) & + MDIO_STAT1_LSTATUS); } static int @@ -164,8 +148,6 @@ fbnic_phylink_pcs_config(struct phylink_pcs *pcs, unsigned int neg_mode, static const struct phylink_pcs_ops fbnic_phylink_pcs_ops = { .pcs_config = fbnic_phylink_pcs_config, - .pcs_enable = fbnic_phylink_pcs_enable, - .pcs_disable = fbnic_phylink_pcs_disable, .pcs_get_state = fbnic_phylink_pcs_get_state, }; @@ -179,12 +161,39 @@ fbnic_phylink_mac_select_pcs(struct phylink_config *config, return &fbn->phylink_pcs; } +static int +fbnic_phylink_mac_prepare(struct phylink_config *config, unsigned int mode, + phy_interface_t iface) +{ + struct net_device *netdev = to_net_dev(config->dev); + struct fbnic_net *fbn = netdev_priv(netdev); + struct fbnic_dev *fbd = fbn->fbd; + + fbd->mac->prepare(fbd, fbn->aui, fbn->fec); + + return 0; +} + static void fbnic_phylink_mac_config(struct phylink_config *config, unsigned int mode, const struct phylink_link_state *state) { } +static int +fbnic_phylink_mac_finish(struct phylink_config *config, unsigned int mode, + phy_interface_t iface) +{ + struct net_device *netdev = to_net_dev(config->dev); + struct fbnic_net *fbn = netdev_priv(netdev); + struct fbnic_dev *fbd = fbn->fbd; + + /* Retest the link state and restart interrupts */ + fbd->mac->get_link(fbd, fbn->aui, fbn->fec); + + return 0; +} + static void fbnic_phylink_mac_link_down(struct phylink_config *config, unsigned int mode, phy_interface_t interface) @@ -216,7 +225,9 @@ fbnic_phylink_mac_link_up(struct phylink_config *config, static const struct phylink_mac_ops fbnic_phylink_mac_ops = { .mac_select_pcs = fbnic_phylink_mac_select_pcs, + .mac_prepare = fbnic_phylink_mac_prepare, .mac_config = fbnic_phylink_mac_config, + .mac_finish = fbnic_phylink_mac_finish, .mac_link_down = fbnic_phylink_mac_link_down, .mac_link_up = fbnic_phylink_mac_link_up, }; @@ -257,3 +268,52 @@ int fbnic_phylink_init(struct net_device *netdev) return 0; } + +/** + * fbnic_phylink_pmd_training_complete_notify - PMD training complete notifier + * @netdev: Netdev struct phylink device attached to + * + * When the link first comes up the PMD will have a period of 2 to 3 seconds + * where the link will flutter due to link training. To avoid spamming the + * kernel log with messages about this we add a delay of 4 seconds from the + * time of the last PCS report of link so that we can guarantee we are unlikely + * to see any further link loss events due to link training. + **/ +void fbnic_phylink_pmd_training_complete_notify(struct net_device *netdev) +{ + struct fbnic_net *fbn = netdev_priv(netdev); + struct fbnic_dev *fbd = fbn->fbd; + + if (fbd->pmd_state != FBNIC_PMD_TRAINING) + return; + + /* Prevent reading end_of_pmd_training until we verified state */ + smp_rmb(); + + if (!time_before(READ_ONCE(fbd->end_of_pmd_training), jiffies)) + return; + + /* At this point we have verified that the link has been up for + * the full training duration. As a first step we will try + * transitioning to link ready. + */ + if (cmpxchg(&fbd->pmd_state, FBNIC_PMD_TRAINING, + FBNIC_PMD_LINK_READY) != FBNIC_PMD_TRAINING) + return; + + /* Perform a follow-up check to verify that the link didn't flap + * just before our transition by rechecking the training timer. + */ + if (!time_before(READ_ONCE(fbd->end_of_pmd_training), jiffies)) + return; + + /* The training timeout has been completed. We are good to swap out + * link_ready for send_data assuming no other events have occurred + * that would have pulled us back into initialization or training. + */ + if (cmpxchg(&fbd->pmd_state, FBNIC_PMD_LINK_READY, + FBNIC_PMD_SEND_DATA) != FBNIC_PMD_LINK_READY) + return; + + phylink_pcs_change(&fbn->phylink_pcs, false); +} From 1fe7978329d736e90600b16b34c7656073dc7a5a Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Fri, 21 Nov 2025 08:40:38 -0800 Subject: [PATCH 1206/1381] fbnic: Add handler for reporting link down event statistics We were previously not displaying the number of link_down_events tracked by the device. With this change we should now be able to display the value. The value itself tracks the calls from the phylink interface to the mac_link_down call. Signed-off-by: Alexander Duyck Link: https://patch.msgid.link/176374323824.959489.6915296616773178954.stgit@ahduyck-xeon-server.home.arpa Signed-off-by: Paolo Abeni --- drivers/net/ethernet/meta/fbnic/fbnic_ethtool.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_ethtool.c b/drivers/net/ethernet/meta/fbnic/fbnic_ethtool.c index 95fac020eb93..693ebdf38705 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_ethtool.c +++ b/drivers/net/ethernet/meta/fbnic/fbnic_ethtool.c @@ -1863,6 +1863,14 @@ fbnic_get_rmon_stats(struct net_device *netdev, *ranges = fbnic_rmon_ranges; } +static void fbnic_get_link_ext_stats(struct net_device *netdev, + struct ethtool_link_ext_stats *stats) +{ + struct fbnic_net *fbn = netdev_priv(netdev); + + stats->link_down_events = fbn->link_down_events; +} + static const struct ethtool_ops fbnic_ethtool_ops = { .cap_link_lanes_supported = true, .supported_coalesce_params = ETHTOOL_COALESCE_USECS | @@ -1874,6 +1882,7 @@ static const struct ethtool_ops fbnic_ethtool_ops = { .get_regs_len = fbnic_get_regs_len, .get_regs = fbnic_get_regs, .get_link = ethtool_op_get_link, + .get_link_ext_stats = fbnic_get_link_ext_stats, .get_coalesce = fbnic_get_coalesce, .set_coalesce = fbnic_set_coalesce, .get_ringparam = fbnic_get_ringparam, From d0ce9fd7eae048fbf4004fb4632b878807b18c57 Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Fri, 21 Nov 2025 08:40:45 -0800 Subject: [PATCH 1207/1381] fbnic: Add SW shim for MDIO interface to PMD and PCS In order for us to support a PCS device we need to add an MDIO bus to allow the drivers to have access to the registers for the device. This change adds such an interface. The interface will consist of 2 PHY addrs, the first one consisting of a PMD and PCS, and the second just being a PCS. There is a need for 2 PHYs addrs due to the fact that in order to support the 50GBase-CR2 mode we will need to access and configure the PCS vendor registers and RSFEC registers from the second lane identical to the first. Signed-off-by: Alexander Duyck Link: https://patch.msgid.link/176374324532.959489.15389723111560978054.stgit@ahduyck-xeon-server.home.arpa Signed-off-by: Paolo Abeni --- drivers/net/ethernet/meta/fbnic/Makefile | 1 + drivers/net/ethernet/meta/fbnic/fbnic.h | 5 + drivers/net/ethernet/meta/fbnic/fbnic_mac.h | 1 + drivers/net/ethernet/meta/fbnic/fbnic_mdio.c | 195 +++++++++++++++++++ drivers/net/ethernet/meta/fbnic/fbnic_pci.c | 3 + 5 files changed, 205 insertions(+) create mode 100644 drivers/net/ethernet/meta/fbnic/fbnic_mdio.c diff --git a/drivers/net/ethernet/meta/fbnic/Makefile b/drivers/net/ethernet/meta/fbnic/Makefile index 15e8ff649615..72c41af65364 100644 --- a/drivers/net/ethernet/meta/fbnic/Makefile +++ b/drivers/net/ethernet/meta/fbnic/Makefile @@ -21,6 +21,7 @@ fbnic-y := fbnic_csr.o \ fbnic_pci.o \ fbnic_phylink.o \ fbnic_rpc.o \ + fbnic_mdio.o \ fbnic_time.o \ fbnic_tlv.o \ fbnic_txrx.o \ diff --git a/drivers/net/ethernet/meta/fbnic/fbnic.h b/drivers/net/ethernet/meta/fbnic/fbnic.h index fac1283d0ade..779a083b9215 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic.h +++ b/drivers/net/ethernet/meta/fbnic/fbnic.h @@ -95,6 +95,9 @@ struct fbnic_dev { u64 prev_firmware_time; struct fbnic_fw_log fw_log; + + /* MDIO bus for PHYs */ + struct mii_bus *mdio_bus; }; /* Reserve entry 0 in the MSI-X "others" array until we have filled all @@ -204,6 +207,8 @@ void fbnic_dbg_exit(void); void fbnic_rpc_reset_valid_entries(struct fbnic_dev *fbd); +int fbnic_mdiobus_create(struct fbnic_dev *fbd); + void fbnic_csr_get_regs(struct fbnic_dev *fbd, u32 *data, u32 *regs_version); int fbnic_csr_regs_len(struct fbnic_dev *fbd); diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_mac.h b/drivers/net/ethernet/meta/fbnic/fbnic_mac.h index 9b4fb0586dff..f08fe8b7c497 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_mac.h +++ b/drivers/net/ethernet/meta/fbnic/fbnic_mac.h @@ -56,6 +56,7 @@ enum { FBNIC_AUI_50GAUI1 = 2, /* 53.125GBd 53.125 * 1 */ FBNIC_AUI_100GAUI2 = 3, /* 106.25GBd 53.125 * 2 */ FBNIC_AUI_UNKNOWN = 4, + __FBNIC_AUI_MAX__ }; #define FBNIC_AUI_MODE_R2 (FBNIC_AUI_LAUI2) diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_mdio.c b/drivers/net/ethernet/meta/fbnic/fbnic_mdio.c new file mode 100644 index 000000000000..709041f7fc43 --- /dev/null +++ b/drivers/net/ethernet/meta/fbnic/fbnic_mdio.c @@ -0,0 +1,195 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) Meta Platforms, Inc. and affiliates. */ + +#include +#include + +#include "fbnic.h" +#include "fbnic_netdev.h" + +#define DW_VENDOR BIT(15) +#define FBNIC_PCS_VENDOR BIT(9) +#define FBNIC_PCS_ZERO_MASK (DW_VENDOR - FBNIC_PCS_VENDOR) + +static int +fbnic_mdio_read_pmd(struct fbnic_dev *fbd, int addr, int regnum) +{ + u8 aui = FBNIC_AUI_UNKNOWN; + struct fbnic_net *fbn; + int ret = 0; + + /* We don't need a second PMD, just one can handle both lanes */ + if (addr) + return 0; + + if (fbd->netdev) { + fbn = netdev_priv(fbd->netdev); + if (fbn->aui < FBNIC_AUI_UNKNOWN) + aui = fbn->aui; + } + + switch (regnum) { + case MDIO_DEVID1: + ret = MP_FBNIC_XPCS_PMA_100G_ID >> 16; + break; + case MDIO_DEVID2: + ret = MP_FBNIC_XPCS_PMA_100G_ID & 0xffff; + break; + case MDIO_DEVS1: + ret = MDIO_DEVS_PMAPMD | MDIO_DEVS_PCS; + break; + case MDIO_STAT2: + ret = MDIO_STAT2_DEVPRST_VAL; + break; + case MDIO_PMA_RXDET: + /* If training isn't complete default to 0 */ + if (fbd->pmd_state != FBNIC_PMD_SEND_DATA) + break; + /* Report either 1 or 2 lanes detected depending on config */ + ret = (MDIO_PMD_RXDET_GLOBAL | MDIO_PMD_RXDET_0) | + ((aui & FBNIC_AUI_MODE_R2) * + (MDIO_PMD_RXDET_1 / FBNIC_AUI_MODE_R2)); + break; + default: + break; + } + + dev_dbg(fbd->dev, + "SWMII PMD Rd: Addr: %d RegNum: %d Value: 0x%04x\n", + addr, regnum, ret); + + return ret; +} + +static int +fbnic_mdio_read_pcs(struct fbnic_dev *fbd, int addr, int regnum) +{ + int ret, offset = 0; + + /* We will need access to both PCS instances to get config info */ + if (addr >= 2) + return 0; + + /* Report 0 for reserved registers */ + if (regnum & FBNIC_PCS_ZERO_MASK) + return 0; + + /* Intercept and return correct ID for PCS */ + if (regnum == MDIO_DEVID1) + return DW_XPCS_ID >> 16; + if (regnum == MDIO_DEVID2) + return DW_XPCS_ID & 0xffff; + if (regnum == MDIO_DEVS1) + return MDIO_DEVS_PMAPMD | MDIO_DEVS_PCS; + + /* Swap vendor page bit for FBNIC PCS vendor page bit */ + if (regnum & DW_VENDOR) + offset ^= DW_VENDOR | FBNIC_PCS_VENDOR; + + ret = fbnic_rd32(fbd, FBNIC_PCS_PAGE(addr) + (regnum ^ offset)); + + dev_dbg(fbd->dev, + "SWMII PCS Rd: Addr: %d RegNum: %d Value: 0x%04x\n", + addr, regnum, ret); + + return ret; +} + +static int +fbnic_mdio_read_c45(struct mii_bus *bus, int addr, int devnum, int regnum) +{ + struct fbnic_dev *fbd = bus->priv; + + if (devnum == MDIO_MMD_PMAPMD) + return fbnic_mdio_read_pmd(fbd, addr, regnum); + + if (devnum == MDIO_MMD_PCS) + return fbnic_mdio_read_pcs(fbd, addr, regnum); + + return 0; +} + +static void +fbnic_mdio_write_pmd(struct fbnic_dev *fbd, int addr, int regnum, u16 val) +{ + dev_dbg(fbd->dev, + "SWMII PMD Wr: Addr: %d RegNum: %d Value: 0x%04x\n", + addr, regnum, val); +} + +static void +fbnic_mdio_write_pcs(struct fbnic_dev *fbd, int addr, int regnum, u16 val) +{ + dev_dbg(fbd->dev, + "SWMII PCS Wr: Addr: %d RegNum: %d Value: 0x%04x\n", + addr, regnum, val); + + /* Allow access to both halves of PCS for 50R2 config */ + if (addr > 2) + return; + + /* Skip write for reserved registers */ + if (regnum & FBNIC_PCS_ZERO_MASK) + return; + + /* Swap vendor page bit for FBNIC PCS vendor page bit */ + if (regnum & DW_VENDOR) + regnum ^= DW_VENDOR | FBNIC_PCS_VENDOR; + + fbnic_wr32(fbd, FBNIC_PCS_PAGE(addr) + regnum, val); +} + +static int +fbnic_mdio_write_c45(struct mii_bus *bus, int addr, int devnum, + int regnum, u16 val) +{ + struct fbnic_dev *fbd = bus->priv; + + if (devnum == MDIO_MMD_PMAPMD) + fbnic_mdio_write_pmd(fbd, addr, regnum, val); + + if (devnum == MDIO_MMD_PCS) + fbnic_mdio_write_pcs(fbd, addr, regnum, val); + + return 0; +} + +/** + * fbnic_mdiobus_create - Create an MDIO bus to allow interfacing w/ PHYs + * @fbd: Pointer to FBNIC device structure to populate bus on + * + * Initialize an MDIO bus and place a pointer to it on the fbd struct. This bus + * will be used to interface with the PMA/PMD and PCS. + * + * Return: 0 on success, negative on failure + **/ +int fbnic_mdiobus_create(struct fbnic_dev *fbd) +{ + struct mii_bus *bus; + int err; + + bus = devm_mdiobus_alloc(fbd->dev); + if (!bus) + return -ENOMEM; + + bus->name = "fbnic_mii_bus"; + bus->read_c45 = &fbnic_mdio_read_c45; + bus->write_c45 = &fbnic_mdio_write_c45; + + /* Disable PHY auto probing. We will add PCS manually */ + bus->phy_mask = ~0; + + bus->parent = fbd->dev; + bus->priv = fbd; + snprintf(bus->id, MII_BUS_ID_SIZE, "%s-mii", dev_name(fbd->dev)); + + err = devm_mdiobus_register(fbd->dev, bus); + if (err) { + dev_err(fbd->dev, "Failed to create MDIO bus: %d\n", err); + return err; + } + + fbd->mdio_bus = bus; + + return 0; +} diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_pci.c b/drivers/net/ethernet/meta/fbnic/fbnic_pci.c index dea5367d2190..861d98099c44 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_pci.c +++ b/drivers/net/ethernet/meta/fbnic/fbnic_pci.c @@ -339,6 +339,9 @@ static int fbnic_probe(struct pci_dev *pdev, const struct pci_device_id *ent) goto init_failure_mode; } + if (fbnic_mdiobus_create(fbd)) + goto init_failure_mode; + netdev = fbnic_netdev_alloc(fbd); if (!netdev) { dev_err(&pdev->dev, "Netdev allocation failed\n"); From d0fe7104c79569ebd66d3dcefb9f0acd831d4413 Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Fri, 21 Nov 2025 08:40:52 -0800 Subject: [PATCH 1208/1381] fbnic: Replace use of internal PCS w/ Designware XPCS As we have exposed the PCS registers via the SWMII we can now start looking at connecting the XPCS driver to those registers and let it mange the PCS instead of us doing it directly from the fbnic driver. For now this just gets us the ability to detect link. The hope is in the future to add some of the vendor specific registers to begin enabling XPCS configuration of the interface. Signed-off-by: Alexander Duyck Link: https://patch.msgid.link/176374325295.959489.14521115864034905277.stgit@ahduyck-xeon-server.home.arpa Signed-off-by: Paolo Abeni --- drivers/net/ethernet/meta/Kconfig | 1 + drivers/net/ethernet/meta/fbnic/fbnic_irq.c | 2 +- .../net/ethernet/meta/fbnic/fbnic_netdev.c | 7 +- .../net/ethernet/meta/fbnic/fbnic_netdev.h | 4 +- .../net/ethernet/meta/fbnic/fbnic_phylink.c | 104 ++++++++---------- 5 files changed, 55 insertions(+), 63 deletions(-) diff --git a/drivers/net/ethernet/meta/Kconfig b/drivers/net/ethernet/meta/Kconfig index dff51f23d295..ca5c7ac2a5bc 100644 --- a/drivers/net/ethernet/meta/Kconfig +++ b/drivers/net/ethernet/meta/Kconfig @@ -26,6 +26,7 @@ config FBNIC depends on PTP_1588_CLOCK_OPTIONAL select NET_DEVLINK select PAGE_POOL + select PCS_XPCS select PHYLINK select PLDMFW help diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_irq.c b/drivers/net/ethernet/meta/fbnic/fbnic_irq.c index 9b068b82f30a..02e8b0b257fe 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_irq.c +++ b/drivers/net/ethernet/meta/fbnic/fbnic_irq.c @@ -133,7 +133,7 @@ static irqreturn_t fbnic_mac_msix_intr(int __always_unused irq, void *data) /* Record link down events */ if (!fbd->mac->get_link(fbd, fbn->aui, fbn->fec)) - phylink_pcs_change(&fbn->phylink_pcs, false); + phylink_pcs_change(fbn->pcs, false); return IRQ_HANDLED; } diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c index 65318a5b466e..81c9d5c9a4b2 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c +++ b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c @@ -697,10 +697,7 @@ void fbnic_reset_queues(struct fbnic_net *fbn, **/ void fbnic_netdev_free(struct fbnic_dev *fbd) { - struct fbnic_net *fbn = netdev_priv(fbd->netdev); - - if (fbn->phylink) - phylink_destroy(fbn->phylink); + fbnic_phylink_destroy(fbd->netdev); free_netdev(fbd->netdev); fbd->netdev = NULL; @@ -802,7 +799,7 @@ struct net_device *fbnic_netdev_alloc(struct fbnic_dev *fbd) netif_tx_stop_all_queues(netdev); - if (fbnic_phylink_init(netdev)) { + if (fbnic_phylink_create(netdev)) { fbnic_netdev_free(fbd); return NULL; } diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.h b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.h index deab789b2a6c..9129a658f8fa 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.h +++ b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.h @@ -44,7 +44,7 @@ struct fbnic_net { struct phylink *phylink; struct phylink_config phylink_config; - struct phylink_pcs phylink_pcs; + struct phylink_pcs *pcs; u8 aui; u8 fec; @@ -108,6 +108,8 @@ int fbnic_phylink_ethtool_ksettings_get(struct net_device *netdev, struct ethtool_link_ksettings *cmd); int fbnic_phylink_get_fecparam(struct net_device *netdev, struct ethtool_fecparam *fecparam); +int fbnic_phylink_create(struct net_device *netdev); +void fbnic_phylink_destroy(struct net_device *netdev); int fbnic_phylink_init(struct net_device *netdev); void fbnic_phylink_pmd_training_complete_notify(struct net_device *netdev); bool fbnic_check_split_frames(struct bpf_prog *prog, diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_phylink.c b/drivers/net/ethernet/meta/fbnic/fbnic_phylink.c index 20f88c8dcc79..09c5225111be 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_phylink.c +++ b/drivers/net/ethernet/meta/fbnic/fbnic_phylink.c @@ -1,6 +1,7 @@ // SPDX-License-Identifier: GPL-2.0 /* Copyright (c) Meta Platforms, Inc. and affiliates. */ +#include #include #include @@ -101,56 +102,6 @@ int fbnic_phylink_get_fecparam(struct net_device *netdev, return 0; } -static struct fbnic_net * -fbnic_pcs_to_net(struct phylink_pcs *pcs) -{ - return container_of(pcs, struct fbnic_net, phylink_pcs); -} - -static void -fbnic_phylink_pcs_get_state(struct phylink_pcs *pcs, unsigned int neg_mode, - struct phylink_link_state *state) -{ - struct fbnic_net *fbn = fbnic_pcs_to_net(pcs); - struct fbnic_dev *fbd = fbn->fbd; - - switch (fbn->aui) { - case FBNIC_AUI_25GAUI: - state->speed = SPEED_25000; - break; - case FBNIC_AUI_LAUI2: - case FBNIC_AUI_50GAUI1: - state->speed = SPEED_50000; - break; - case FBNIC_AUI_100GAUI2: - state->speed = SPEED_100000; - break; - default: - state->link = 0; - return; - } - - state->duplex = DUPLEX_FULL; - - state->link = (fbd->pmd_state == FBNIC_PMD_SEND_DATA) && - (rd32(fbd, FBNIC_PCS(MDIO_STAT1, 0)) & - MDIO_STAT1_LSTATUS); -} - -static int -fbnic_phylink_pcs_config(struct phylink_pcs *pcs, unsigned int neg_mode, - phy_interface_t interface, - const unsigned long *advertising, - bool permit_pause_to_mac) -{ - return 0; -} - -static const struct phylink_pcs_ops fbnic_phylink_pcs_ops = { - .pcs_config = fbnic_phylink_pcs_config, - .pcs_get_state = fbnic_phylink_pcs_get_state, -}; - static struct phylink_pcs * fbnic_phylink_mac_select_pcs(struct phylink_config *config, phy_interface_t interface) @@ -158,7 +109,7 @@ fbnic_phylink_mac_select_pcs(struct phylink_config *config, struct net_device *netdev = to_net_dev(config->dev); struct fbnic_net *fbn = netdev_priv(netdev); - return &fbn->phylink_pcs; + return fbn->pcs; } static int @@ -232,13 +183,33 @@ static const struct phylink_mac_ops fbnic_phylink_mac_ops = { .mac_link_up = fbnic_phylink_mac_link_up, }; -int fbnic_phylink_init(struct net_device *netdev) +/** + * fbnic_phylink_create - Phylink device creation + * @netdev: Network Device struct to attach phylink device + * + * Initialize and attach a phylink instance to the device. The phylink + * device will make use of the netdev struct to track carrier and will + * eventually be used to expose the current state of the MAC and PCS + * setup. + * + * Return: 0 on success, negative on failure + **/ +int fbnic_phylink_create(struct net_device *netdev) { struct fbnic_net *fbn = netdev_priv(netdev); struct fbnic_dev *fbd = fbn->fbd; + struct phylink_pcs *pcs; struct phylink *phylink; + int err; - fbn->phylink_pcs.ops = &fbnic_phylink_pcs_ops; + pcs = xpcs_create_pcs_mdiodev(fbd->mdio_bus, 0); + if (IS_ERR(pcs)) { + err = PTR_ERR(pcs); + dev_err(fbd->dev, "Failed to create PCS device: %d\n", err); + return err; + } + + fbn->pcs = pcs; fbn->phylink_config.dev = &netdev->dev; fbn->phylink_config.type = PHYLINK_NETDEV; @@ -261,14 +232,35 @@ int fbnic_phylink_init(struct net_device *netdev) phylink = phylink_create(&fbn->phylink_config, NULL, fbnic_phylink_select_interface(fbn->aui), &fbnic_phylink_mac_ops); - if (IS_ERR(phylink)) - return PTR_ERR(phylink); + if (IS_ERR(phylink)) { + err = PTR_ERR(phylink); + dev_err(netdev->dev.parent, + "Failed to create Phylink interface, err: %d\n", err); + xpcs_destroy_pcs(pcs); + return err; + } fbn->phylink = phylink; return 0; } +/** + * fbnic_phylink_destroy - Teardown phylink related interfaces + * @netdev: Network Device struct containing phylink device + * + * Detach and free resources related to phylink interface. + **/ +void fbnic_phylink_destroy(struct net_device *netdev) +{ + struct fbnic_net *fbn = netdev_priv(netdev); + + if (fbn->phylink) + phylink_destroy(fbn->phylink); + if (fbn->pcs) + xpcs_destroy_pcs(fbn->pcs); +} + /** * fbnic_phylink_pmd_training_complete_notify - PMD training complete notifier * @netdev: Netdev struct phylink device attached to @@ -315,5 +307,5 @@ void fbnic_phylink_pmd_training_complete_notify(struct net_device *netdev) FBNIC_PMD_SEND_DATA) != FBNIC_PMD_LINK_READY) return; - phylink_pcs_change(&fbn->phylink_pcs, false); + phylink_pcs_change(fbn->pcs, false); } From 5c9c1e78de1e41b5321583139851797580c077c3 Mon Sep 17 00:00:00 2001 From: Sunday Adelodun Date: Tue, 25 Nov 2025 12:36:48 +0100 Subject: [PATCH 1209/1381] selftests: af_unix: remove unused stdlib.h include The unix_connreset.c test included , but no symbol from that header is used. This causes a fatal build error under certain linux-next configurations where stdlib.h is not available. Remove the unused include to fix the build. Reported-by: kernel test robot Closes: https://lore.kernel.org/r/202511221800.hcgCKvVa-lkp@intel.com/ Signed-off-by: Sunday Adelodun Link: https://patch.msgid.link/20251125113648.25903-1-adelodunolaoluwa@yahoo.com Signed-off-by: Paolo Abeni --- tools/testing/selftests/net/af_unix/unix_connreset.c | 1 - 1 file changed, 1 deletion(-) diff --git a/tools/testing/selftests/net/af_unix/unix_connreset.c b/tools/testing/selftests/net/af_unix/unix_connreset.c index 6eb936207b31..08c1de8f5a98 100644 --- a/tools/testing/selftests/net/af_unix/unix_connreset.c +++ b/tools/testing/selftests/net/af_unix/unix_connreset.c @@ -14,7 +14,6 @@ */ #define _GNU_SOURCE -#include #include #include #include From 45d100ee0d6e8b4b4ba6c48f54decd62f875cf70 Mon Sep 17 00:00:00 2001 From: Rohan G Thomas Date: Wed, 26 Nov 2025 00:37:12 +0800 Subject: [PATCH 1210/1381] net: stmmac: dwmac: Disable flushing frames on Rx Buffer Unavailable In Store and Forward mode, flushing frames when the receive buffer is unavailable, can cause the MTL Rx FIFO to go out of sync. This results in buffering of a few frames in the FIFO without triggering Rx DMA from transferring the data to the system memory until another packet is received. Once the issue happens, for a ping request, the packet is forwarded to the system memory only after we receive another packet and hece we observe a latency equivalent to the ping interval. 64 bytes from 192.168.2.100: seq=1 ttl=64 time=1000.344 ms Also, we can observe constant gmacgrp_debug register value of 0x00000120, which indicates "Reading frame data". The issue is not reproducible after disabling frame flushing when Rx buffer is unavailable. But in that case, the Rx DMA enters a suspend state due to buffer unavailability. To resume operation, software must write to the receive_poll_demand register after adding new descriptors, which reactivates the Rx DMA. This issue is observed in the socfpga platforms which has dwmac1000 IP like Arria 10, Cyclone V and Agilex 7. Issue is reproducible after running iperf3 server at the DUT for UDP lower packet sizes. Signed-off-by: Rohan G Thomas Reviewed-by: Matthew Gerlach Tested-by: Maxime Chevallier Reviewed-by: Russell King (Oracle) Link: https://patch.msgid.link/20251126-a10_ext_fix-v1-1-d163507f646f@altera.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c | 5 +++-- drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h | 1 + drivers/net/ethernet/stmicro/stmmac/dwmac_lib.c | 5 +++++ drivers/net/ethernet/stmicro/stmmac/hwif.h | 3 +++ drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 2 ++ 5 files changed, 14 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c index 6d9b8fac3c6d..5877fec9f6c3 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c @@ -135,10 +135,10 @@ static void dwmac1000_dma_operation_mode_rx(struct stmmac_priv *priv, if (mode == SF_DMA_MODE) { pr_debug("GMAC: enable RX store and forward mode\n"); - csr6 |= DMA_CONTROL_RSF; + csr6 |= DMA_CONTROL_RSF | DMA_CONTROL_DFF; } else { pr_debug("GMAC: disable RX SF mode (threshold %d)\n", mode); - csr6 &= ~DMA_CONTROL_RSF; + csr6 &= ~(DMA_CONTROL_RSF | DMA_CONTROL_DFF); csr6 &= DMA_CONTROL_TC_RX_MASK; if (mode <= 32) csr6 |= DMA_CONTROL_RTC_32; @@ -262,6 +262,7 @@ const struct stmmac_dma_ops dwmac1000_dma_ops = { .dma_rx_mode = dwmac1000_dma_operation_mode_rx, .dma_tx_mode = dwmac1000_dma_operation_mode_tx, .enable_dma_transmission = dwmac_enable_dma_transmission, + .enable_dma_reception = dwmac_enable_dma_reception, .enable_dma_irq = dwmac_enable_dma_irq, .disable_dma_irq = dwmac_disable_dma_irq, .start_tx = dwmac_dma_start_tx, diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h b/drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h index d1c149f7a3dd..054ecb20ce3f 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h @@ -169,6 +169,7 @@ static inline u32 dma_chan_base_addr(u32 base, u32 chan) #define NUM_DWMAC4_DMA_REGS 27 void dwmac_enable_dma_transmission(void __iomem *ioaddr, u32 chan); +void dwmac_enable_dma_reception(void __iomem *ioaddr, u32 chan); void dwmac_enable_dma_irq(struct stmmac_priv *priv, void __iomem *ioaddr, u32 chan, bool rx, bool tx); void dwmac_disable_dma_irq(struct stmmac_priv *priv, void __iomem *ioaddr, diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac_lib.c b/drivers/net/ethernet/stmicro/stmmac/dwmac_lib.c index 467f1a05747e..97a803d68e3a 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac_lib.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac_lib.c @@ -33,6 +33,11 @@ void dwmac_enable_dma_transmission(void __iomem *ioaddr, u32 chan) writel(1, ioaddr + DMA_CHAN_XMT_POLL_DEMAND(chan)); } +void dwmac_enable_dma_reception(void __iomem *ioaddr, u32 chan) +{ + writel(1, ioaddr + DMA_CHAN_RCV_POLL_DEMAND(chan)); +} + void dwmac_enable_dma_irq(struct stmmac_priv *priv, void __iomem *ioaddr, u32 chan, bool rx, bool tx) { diff --git a/drivers/net/ethernet/stmicro/stmmac/hwif.h b/drivers/net/ethernet/stmicro/stmmac/hwif.h index f257ce4b6c66..df6e8a567b1f 100644 --- a/drivers/net/ethernet/stmicro/stmmac/hwif.h +++ b/drivers/net/ethernet/stmicro/stmmac/hwif.h @@ -201,6 +201,7 @@ struct stmmac_dma_ops { void (*dma_diagnostic_fr)(struct stmmac_extra_stats *x, void __iomem *ioaddr); void (*enable_dma_transmission)(void __iomem *ioaddr, u32 chan); + void (*enable_dma_reception)(void __iomem *ioaddr, u32 chan); void (*enable_dma_irq)(struct stmmac_priv *priv, void __iomem *ioaddr, u32 chan, bool rx, bool tx); void (*disable_dma_irq)(struct stmmac_priv *priv, void __iomem *ioaddr, @@ -261,6 +262,8 @@ struct stmmac_dma_ops { stmmac_do_void_callback(__priv, dma, dma_diagnostic_fr, __args) #define stmmac_enable_dma_transmission(__priv, __args...) \ stmmac_do_void_callback(__priv, dma, enable_dma_transmission, __args) +#define stmmac_enable_dma_reception(__priv, __args...) \ + stmmac_do_void_callback(__priv, dma, enable_dma_reception, __args) #define stmmac_enable_dma_irq(__priv, __args...) \ stmmac_do_void_callback(__priv, dma, enable_dma_irq, __priv, __args) #define stmmac_disable_dma_irq(__priv, __args...) \ diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index 6cacedb2c9b3..1ecca60baf74 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -4973,6 +4973,8 @@ static inline void stmmac_rx_refill(struct stmmac_priv *priv, u32 queue) rx_q->rx_tail_addr = rx_q->dma_rx_phy + (rx_q->dirty_rx * sizeof(struct dma_desc)); stmmac_set_rx_tail_ptr(priv, priv->ioaddr, rx_q->rx_tail_addr, queue); + /* Wake up Rx DMA from the suspend state if required */ + stmmac_enable_dma_reception(priv, priv->ioaddr, queue); } static unsigned int stmmac_rx_buf1_len(struct stmmac_priv *priv, From 634f3853cc98d73bdec8918010ee29b06981583e Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Thu, 27 Nov 2025 23:26:22 +0000 Subject: [PATCH 1211/1381] netfilter: flowtable: check for maximum number of encapsulations in bridge vlan Add a sanity check to skip path discovery if the maximum number of encapsulation is reached. While at it, check for underflow too. Fixes: 26267bf9bb57 ("netfilter: flowtable: bridge vlan hardware offload and switchdev") Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nft_flow_offload.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/net/netfilter/nft_flow_offload.c b/net/netfilter/nft_flow_offload.c index 14dd1c0698c3..e95e5f59a3d6 100644 --- a/net/netfilter/nft_flow_offload.c +++ b/net/netfilter/nft_flow_offload.c @@ -141,12 +141,19 @@ static void nft_dev_path_info(const struct net_device_path_stack *stack, info->ingress_vlans |= BIT(info->num_encaps - 1); break; case DEV_PATH_BR_VLAN_TAG: + if (info->num_encaps >= NF_FLOW_TABLE_ENCAP_MAX) { + info->indev = NULL; + break; + } info->encap[info->num_encaps].id = path->bridge.vlan_id; info->encap[info->num_encaps].proto = path->bridge.vlan_proto; info->num_encaps++; break; case DEV_PATH_BR_VLAN_UNTAG: - info->num_encaps--; + if (WARN_ON_ONCE(info->num_encaps-- == 0)) { + info->indev = NULL; + break; + } break; case DEV_PATH_BR_VLAN_KEEP: break; From 93d7a7ed07342f5e3da2d250cfd67f899d0b5318 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Fri, 10 Oct 2025 12:32:23 +0200 Subject: [PATCH 1212/1381] netfilter: flowtable: move path discovery infrastructure to its own file This file contains the path discovery that is run from the forward chain for the packet offloading the flow into the flowtable. This consists of a series of calls to dev_fill_forward_path() for each device stack. More topologies may be supported in the future, so move this code to its own file to separate it from the nftables flow_offload expression. Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_flow_table.h | 6 + net/netfilter/Makefile | 1 + net/netfilter/nf_flow_table_path.c | 274 ++++++++++++++++++++++++++ net/netfilter/nft_flow_offload.c | 259 ------------------------ 4 files changed, 281 insertions(+), 259 deletions(-) create mode 100644 net/netfilter/nf_flow_table_path.c diff --git a/include/net/netfilter/nf_flow_table.h b/include/net/netfilter/nf_flow_table.h index c003cd194fa2..e9f72d2558e9 100644 --- a/include/net/netfilter/nf_flow_table.h +++ b/include/net/netfilter/nf_flow_table.h @@ -222,6 +222,12 @@ struct nf_flow_route { struct flow_offload *flow_offload_alloc(struct nf_conn *ct); void flow_offload_free(struct flow_offload *flow); +struct nft_flowtable; +struct nft_pktinfo; +int nft_flow_route(const struct nft_pktinfo *pkt, const struct nf_conn *ct, + struct nf_flow_route *route, enum ip_conntrack_dir dir, + struct nft_flowtable *ft); + static inline int nf_flow_table_offload_add_cb(struct nf_flowtable *flow_table, flow_setup_cb_t *cb, void *cb_priv) diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index e43e20f529f8..6bfc250e474f 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -141,6 +141,7 @@ obj-$(CONFIG_NFT_FWD_NETDEV) += nft_fwd_netdev.o # flow table infrastructure obj-$(CONFIG_NF_FLOW_TABLE) += nf_flow_table.o nf_flow_table-objs := nf_flow_table_core.o nf_flow_table_ip.o \ + nf_flow_table_path.o \ nf_flow_table_offload.o nf_flow_table_xdp.o nf_flow_table-$(CONFIG_NF_FLOW_TABLE_PROCFS) += nf_flow_table_procfs.o ifeq ($(CONFIG_NF_FLOW_TABLE),m) diff --git a/net/netfilter/nf_flow_table_path.c b/net/netfilter/nf_flow_table_path.c new file mode 100644 index 000000000000..e525e3745651 --- /dev/null +++ b/net/netfilter/nf_flow_table_path.c @@ -0,0 +1,274 @@ +// SPDX-License-Identifier: GPL-2.0-only +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +static enum flow_offload_xmit_type nft_xmit_type(struct dst_entry *dst) +{ + if (dst_xfrm(dst)) + return FLOW_OFFLOAD_XMIT_XFRM; + + return FLOW_OFFLOAD_XMIT_NEIGH; +} + +static void nft_default_forward_path(struct nf_flow_route *route, + struct dst_entry *dst_cache, + enum ip_conntrack_dir dir) +{ + route->tuple[!dir].in.ifindex = dst_cache->dev->ifindex; + route->tuple[dir].dst = dst_cache; + route->tuple[dir].xmit_type = nft_xmit_type(dst_cache); +} + +static bool nft_is_valid_ether_device(const struct net_device *dev) +{ + if (!dev || (dev->flags & IFF_LOOPBACK) || dev->type != ARPHRD_ETHER || + dev->addr_len != ETH_ALEN || !is_valid_ether_addr(dev->dev_addr)) + return false; + + return true; +} + +static int nft_dev_fill_forward_path(const struct nf_flow_route *route, + const struct dst_entry *dst_cache, + const struct nf_conn *ct, + enum ip_conntrack_dir dir, u8 *ha, + struct net_device_path_stack *stack) +{ + const void *daddr = &ct->tuplehash[!dir].tuple.src.u3; + struct net_device *dev = dst_cache->dev; + struct neighbour *n; + u8 nud_state; + + if (!nft_is_valid_ether_device(dev)) + goto out; + + n = dst_neigh_lookup(dst_cache, daddr); + if (!n) + return -1; + + read_lock_bh(&n->lock); + nud_state = n->nud_state; + ether_addr_copy(ha, n->ha); + read_unlock_bh(&n->lock); + neigh_release(n); + + if (!(nud_state & NUD_VALID)) + return -1; + +out: + return dev_fill_forward_path(dev, ha, stack); +} + +struct nft_forward_info { + const struct net_device *indev; + const struct net_device *outdev; + const struct net_device *hw_outdev; + struct id { + __u16 id; + __be16 proto; + } encap[NF_FLOW_TABLE_ENCAP_MAX]; + u8 num_encaps; + u8 ingress_vlans; + u8 h_source[ETH_ALEN]; + u8 h_dest[ETH_ALEN]; + enum flow_offload_xmit_type xmit_type; +}; + +static void nft_dev_path_info(const struct net_device_path_stack *stack, + struct nft_forward_info *info, + unsigned char *ha, struct nf_flowtable *flowtable) +{ + const struct net_device_path *path; + int i; + + memcpy(info->h_dest, ha, ETH_ALEN); + + for (i = 0; i < stack->num_paths; i++) { + path = &stack->path[i]; + switch (path->type) { + case DEV_PATH_ETHERNET: + case DEV_PATH_DSA: + case DEV_PATH_VLAN: + case DEV_PATH_PPPOE: + info->indev = path->dev; + if (is_zero_ether_addr(info->h_source)) + memcpy(info->h_source, path->dev->dev_addr, ETH_ALEN); + + if (path->type == DEV_PATH_ETHERNET) + break; + if (path->type == DEV_PATH_DSA) { + i = stack->num_paths; + break; + } + + /* DEV_PATH_VLAN and DEV_PATH_PPPOE */ + if (info->num_encaps >= NF_FLOW_TABLE_ENCAP_MAX) { + info->indev = NULL; + break; + } + if (!info->outdev) + info->outdev = path->dev; + info->encap[info->num_encaps].id = path->encap.id; + info->encap[info->num_encaps].proto = path->encap.proto; + info->num_encaps++; + if (path->type == DEV_PATH_PPPOE) + memcpy(info->h_dest, path->encap.h_dest, ETH_ALEN); + break; + case DEV_PATH_BRIDGE: + if (is_zero_ether_addr(info->h_source)) + memcpy(info->h_source, path->dev->dev_addr, ETH_ALEN); + + switch (path->bridge.vlan_mode) { + case DEV_PATH_BR_VLAN_UNTAG_HW: + info->ingress_vlans |= BIT(info->num_encaps - 1); + break; + case DEV_PATH_BR_VLAN_TAG: + if (info->num_encaps >= NF_FLOW_TABLE_ENCAP_MAX) { + info->indev = NULL; + break; + } + info->encap[info->num_encaps].id = path->bridge.vlan_id; + info->encap[info->num_encaps].proto = path->bridge.vlan_proto; + info->num_encaps++; + break; + case DEV_PATH_BR_VLAN_UNTAG: + if (WARN_ON_ONCE(info->num_encaps-- == 0)) { + info->indev = NULL; + break; + } + break; + case DEV_PATH_BR_VLAN_KEEP: + break; + } + info->xmit_type = FLOW_OFFLOAD_XMIT_DIRECT; + break; + default: + info->indev = NULL; + break; + } + } + if (!info->outdev) + info->outdev = info->indev; + + info->hw_outdev = info->indev; + + if (nf_flowtable_hw_offload(flowtable) && + nft_is_valid_ether_device(info->indev)) + info->xmit_type = FLOW_OFFLOAD_XMIT_DIRECT; +} + +static bool nft_flowtable_find_dev(const struct net_device *dev, + struct nft_flowtable *ft) +{ + struct nft_hook *hook; + bool found = false; + + list_for_each_entry_rcu(hook, &ft->hook_list, list) { + if (!nft_hook_find_ops_rcu(hook, dev)) + continue; + + found = true; + break; + } + + return found; +} + +static void nft_dev_forward_path(struct nf_flow_route *route, + const struct nf_conn *ct, + enum ip_conntrack_dir dir, + struct nft_flowtable *ft) +{ + const struct dst_entry *dst = route->tuple[dir].dst; + struct net_device_path_stack stack; + struct nft_forward_info info = {}; + unsigned char ha[ETH_ALEN]; + int i; + + if (nft_dev_fill_forward_path(route, dst, ct, dir, ha, &stack) >= 0) + nft_dev_path_info(&stack, &info, ha, &ft->data); + + if (!info.indev || !nft_flowtable_find_dev(info.indev, ft)) + return; + + route->tuple[!dir].in.ifindex = info.indev->ifindex; + for (i = 0; i < info.num_encaps; i++) { + route->tuple[!dir].in.encap[i].id = info.encap[i].id; + route->tuple[!dir].in.encap[i].proto = info.encap[i].proto; + } + route->tuple[!dir].in.num_encaps = info.num_encaps; + route->tuple[!dir].in.ingress_vlans = info.ingress_vlans; + + if (info.xmit_type == FLOW_OFFLOAD_XMIT_DIRECT) { + memcpy(route->tuple[dir].out.h_source, info.h_source, ETH_ALEN); + memcpy(route->tuple[dir].out.h_dest, info.h_dest, ETH_ALEN); + route->tuple[dir].out.ifindex = info.outdev->ifindex; + route->tuple[dir].out.hw_ifindex = info.hw_outdev->ifindex; + route->tuple[dir].xmit_type = info.xmit_type; + } +} + +int nft_flow_route(const struct nft_pktinfo *pkt, const struct nf_conn *ct, + struct nf_flow_route *route, enum ip_conntrack_dir dir, + struct nft_flowtable *ft) +{ + struct dst_entry *this_dst = skb_dst(pkt->skb); + struct dst_entry *other_dst = NULL; + struct flowi fl; + + memset(&fl, 0, sizeof(fl)); + switch (nft_pf(pkt)) { + case NFPROTO_IPV4: + fl.u.ip4.daddr = ct->tuplehash[dir].tuple.src.u3.ip; + fl.u.ip4.saddr = ct->tuplehash[!dir].tuple.src.u3.ip; + fl.u.ip4.flowi4_oif = nft_in(pkt)->ifindex; + fl.u.ip4.flowi4_iif = this_dst->dev->ifindex; + fl.u.ip4.flowi4_dscp = ip4h_dscp(ip_hdr(pkt->skb)); + fl.u.ip4.flowi4_mark = pkt->skb->mark; + fl.u.ip4.flowi4_flags = FLOWI_FLAG_ANYSRC; + break; + case NFPROTO_IPV6: + fl.u.ip6.daddr = ct->tuplehash[dir].tuple.src.u3.in6; + fl.u.ip6.saddr = ct->tuplehash[!dir].tuple.src.u3.in6; + fl.u.ip6.flowi6_oif = nft_in(pkt)->ifindex; + fl.u.ip6.flowi6_iif = this_dst->dev->ifindex; + fl.u.ip6.flowlabel = ip6_flowinfo(ipv6_hdr(pkt->skb)); + fl.u.ip6.flowi6_mark = pkt->skb->mark; + fl.u.ip6.flowi6_flags = FLOWI_FLAG_ANYSRC; + break; + } + + if (!dst_hold_safe(this_dst)) + return -ENOENT; + + nf_route(nft_net(pkt), &other_dst, &fl, false, nft_pf(pkt)); + if (!other_dst) { + dst_release(this_dst); + return -ENOENT; + } + + nft_default_forward_path(route, this_dst, dir); + nft_default_forward_path(route, other_dst, !dir); + + if (route->tuple[dir].xmit_type == FLOW_OFFLOAD_XMIT_NEIGH && + route->tuple[!dir].xmit_type == FLOW_OFFLOAD_XMIT_NEIGH) { + nft_dev_forward_path(route, ct, dir, ft); + nft_dev_forward_path(route, ct, !dir, ft); + } + + return 0; +} +EXPORT_SYMBOL_GPL(nft_flow_route); diff --git a/net/netfilter/nft_flow_offload.c b/net/netfilter/nft_flow_offload.c index e95e5f59a3d6..b8f76c9057fd 100644 --- a/net/netfilter/nft_flow_offload.c +++ b/net/netfilter/nft_flow_offload.c @@ -20,265 +20,6 @@ struct nft_flow_offload { struct nft_flowtable *flowtable; }; -static enum flow_offload_xmit_type nft_xmit_type(struct dst_entry *dst) -{ - if (dst_xfrm(dst)) - return FLOW_OFFLOAD_XMIT_XFRM; - - return FLOW_OFFLOAD_XMIT_NEIGH; -} - -static void nft_default_forward_path(struct nf_flow_route *route, - struct dst_entry *dst_cache, - enum ip_conntrack_dir dir) -{ - route->tuple[!dir].in.ifindex = dst_cache->dev->ifindex; - route->tuple[dir].dst = dst_cache; - route->tuple[dir].xmit_type = nft_xmit_type(dst_cache); -} - -static bool nft_is_valid_ether_device(const struct net_device *dev) -{ - if (!dev || (dev->flags & IFF_LOOPBACK) || dev->type != ARPHRD_ETHER || - dev->addr_len != ETH_ALEN || !is_valid_ether_addr(dev->dev_addr)) - return false; - - return true; -} - -static int nft_dev_fill_forward_path(const struct nf_flow_route *route, - const struct dst_entry *dst_cache, - const struct nf_conn *ct, - enum ip_conntrack_dir dir, u8 *ha, - struct net_device_path_stack *stack) -{ - const void *daddr = &ct->tuplehash[!dir].tuple.src.u3; - struct net_device *dev = dst_cache->dev; - struct neighbour *n; - u8 nud_state; - - if (!nft_is_valid_ether_device(dev)) - goto out; - - n = dst_neigh_lookup(dst_cache, daddr); - if (!n) - return -1; - - read_lock_bh(&n->lock); - nud_state = n->nud_state; - ether_addr_copy(ha, n->ha); - read_unlock_bh(&n->lock); - neigh_release(n); - - if (!(nud_state & NUD_VALID)) - return -1; - -out: - return dev_fill_forward_path(dev, ha, stack); -} - -struct nft_forward_info { - const struct net_device *indev; - const struct net_device *outdev; - const struct net_device *hw_outdev; - struct id { - __u16 id; - __be16 proto; - } encap[NF_FLOW_TABLE_ENCAP_MAX]; - u8 num_encaps; - u8 ingress_vlans; - u8 h_source[ETH_ALEN]; - u8 h_dest[ETH_ALEN]; - enum flow_offload_xmit_type xmit_type; -}; - -static void nft_dev_path_info(const struct net_device_path_stack *stack, - struct nft_forward_info *info, - unsigned char *ha, struct nf_flowtable *flowtable) -{ - const struct net_device_path *path; - int i; - - memcpy(info->h_dest, ha, ETH_ALEN); - - for (i = 0; i < stack->num_paths; i++) { - path = &stack->path[i]; - switch (path->type) { - case DEV_PATH_ETHERNET: - case DEV_PATH_DSA: - case DEV_PATH_VLAN: - case DEV_PATH_PPPOE: - info->indev = path->dev; - if (is_zero_ether_addr(info->h_source)) - memcpy(info->h_source, path->dev->dev_addr, ETH_ALEN); - - if (path->type == DEV_PATH_ETHERNET) - break; - if (path->type == DEV_PATH_DSA) { - i = stack->num_paths; - break; - } - - /* DEV_PATH_VLAN and DEV_PATH_PPPOE */ - if (info->num_encaps >= NF_FLOW_TABLE_ENCAP_MAX) { - info->indev = NULL; - break; - } - if (!info->outdev) - info->outdev = path->dev; - info->encap[info->num_encaps].id = path->encap.id; - info->encap[info->num_encaps].proto = path->encap.proto; - info->num_encaps++; - if (path->type == DEV_PATH_PPPOE) - memcpy(info->h_dest, path->encap.h_dest, ETH_ALEN); - break; - case DEV_PATH_BRIDGE: - if (is_zero_ether_addr(info->h_source)) - memcpy(info->h_source, path->dev->dev_addr, ETH_ALEN); - - switch (path->bridge.vlan_mode) { - case DEV_PATH_BR_VLAN_UNTAG_HW: - info->ingress_vlans |= BIT(info->num_encaps - 1); - break; - case DEV_PATH_BR_VLAN_TAG: - if (info->num_encaps >= NF_FLOW_TABLE_ENCAP_MAX) { - info->indev = NULL; - break; - } - info->encap[info->num_encaps].id = path->bridge.vlan_id; - info->encap[info->num_encaps].proto = path->bridge.vlan_proto; - info->num_encaps++; - break; - case DEV_PATH_BR_VLAN_UNTAG: - if (WARN_ON_ONCE(info->num_encaps-- == 0)) { - info->indev = NULL; - break; - } - break; - case DEV_PATH_BR_VLAN_KEEP: - break; - } - info->xmit_type = FLOW_OFFLOAD_XMIT_DIRECT; - break; - default: - info->indev = NULL; - break; - } - } - if (!info->outdev) - info->outdev = info->indev; - - info->hw_outdev = info->indev; - - if (nf_flowtable_hw_offload(flowtable) && - nft_is_valid_ether_device(info->indev)) - info->xmit_type = FLOW_OFFLOAD_XMIT_DIRECT; -} - -static bool nft_flowtable_find_dev(const struct net_device *dev, - struct nft_flowtable *ft) -{ - struct nft_hook *hook; - bool found = false; - - list_for_each_entry_rcu(hook, &ft->hook_list, list) { - if (!nft_hook_find_ops_rcu(hook, dev)) - continue; - - found = true; - break; - } - - return found; -} - -static void nft_dev_forward_path(struct nf_flow_route *route, - const struct nf_conn *ct, - enum ip_conntrack_dir dir, - struct nft_flowtable *ft) -{ - const struct dst_entry *dst = route->tuple[dir].dst; - struct net_device_path_stack stack; - struct nft_forward_info info = {}; - unsigned char ha[ETH_ALEN]; - int i; - - if (nft_dev_fill_forward_path(route, dst, ct, dir, ha, &stack) >= 0) - nft_dev_path_info(&stack, &info, ha, &ft->data); - - if (!info.indev || !nft_flowtable_find_dev(info.indev, ft)) - return; - - route->tuple[!dir].in.ifindex = info.indev->ifindex; - for (i = 0; i < info.num_encaps; i++) { - route->tuple[!dir].in.encap[i].id = info.encap[i].id; - route->tuple[!dir].in.encap[i].proto = info.encap[i].proto; - } - route->tuple[!dir].in.num_encaps = info.num_encaps; - route->tuple[!dir].in.ingress_vlans = info.ingress_vlans; - - if (info.xmit_type == FLOW_OFFLOAD_XMIT_DIRECT) { - memcpy(route->tuple[dir].out.h_source, info.h_source, ETH_ALEN); - memcpy(route->tuple[dir].out.h_dest, info.h_dest, ETH_ALEN); - route->tuple[dir].out.ifindex = info.outdev->ifindex; - route->tuple[dir].out.hw_ifindex = info.hw_outdev->ifindex; - route->tuple[dir].xmit_type = info.xmit_type; - } -} - -static int nft_flow_route(const struct nft_pktinfo *pkt, - const struct nf_conn *ct, - struct nf_flow_route *route, - enum ip_conntrack_dir dir, - struct nft_flowtable *ft) -{ - struct dst_entry *this_dst = skb_dst(pkt->skb); - struct dst_entry *other_dst = NULL; - struct flowi fl; - - memset(&fl, 0, sizeof(fl)); - switch (nft_pf(pkt)) { - case NFPROTO_IPV4: - fl.u.ip4.daddr = ct->tuplehash[dir].tuple.src.u3.ip; - fl.u.ip4.saddr = ct->tuplehash[!dir].tuple.src.u3.ip; - fl.u.ip4.flowi4_oif = nft_in(pkt)->ifindex; - fl.u.ip4.flowi4_iif = this_dst->dev->ifindex; - fl.u.ip4.flowi4_dscp = ip4h_dscp(ip_hdr(pkt->skb)); - fl.u.ip4.flowi4_mark = pkt->skb->mark; - fl.u.ip4.flowi4_flags = FLOWI_FLAG_ANYSRC; - break; - case NFPROTO_IPV6: - fl.u.ip6.daddr = ct->tuplehash[dir].tuple.src.u3.in6; - fl.u.ip6.saddr = ct->tuplehash[!dir].tuple.src.u3.in6; - fl.u.ip6.flowi6_oif = nft_in(pkt)->ifindex; - fl.u.ip6.flowi6_iif = this_dst->dev->ifindex; - fl.u.ip6.flowlabel = ip6_flowinfo(ipv6_hdr(pkt->skb)); - fl.u.ip6.flowi6_mark = pkt->skb->mark; - fl.u.ip6.flowi6_flags = FLOWI_FLAG_ANYSRC; - break; - } - - if (!dst_hold_safe(this_dst)) - return -ENOENT; - - nf_route(nft_net(pkt), &other_dst, &fl, false, nft_pf(pkt)); - if (!other_dst) { - dst_release(this_dst); - return -ENOENT; - } - - nft_default_forward_path(route, this_dst, dir); - nft_default_forward_path(route, other_dst, !dir); - - if (route->tuple[dir].xmit_type == FLOW_OFFLOAD_XMIT_NEIGH && - route->tuple[!dir].xmit_type == FLOW_OFFLOAD_XMIT_NEIGH) { - nft_dev_forward_path(route, ct, dir, ft); - nft_dev_forward_path(route, ct, !dir, ft); - } - - return 0; -} - static bool nft_flow_offload_skip(struct sk_buff *skb, int family) { if (skb_sec_path(skb)) From b5964aac51e0c286a50e68225e0dfcf11fb554cb Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Fri, 10 Oct 2025 12:32:35 +0200 Subject: [PATCH 1213/1381] netfilter: flowtable: consolidate xmit path Use dev_queue_xmit() for the XMIT_NEIGH case. Store the interface index of the real device behind the vlan/pppoe device, this introduces an extra lookup for the real device in the xmit path because rt->dst.dev provides the vlan/pppoe device. XMIT_NEIGH now looks more similar to XMIT_DIRECT but the check for stale dst and the neighbour lookup still remain in place which is convenient to deal with network topology changes. Note that nft_flow_route() needs to relax the check for _XMIT_NEIGH so the existing basic xfrm offload (which only works in one direction) does not break. Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_flow_table.h | 1 + net/netfilter/nf_flow_table_core.c | 1 + net/netfilter/nf_flow_table_ip.c | 87 ++++++++++++++++----------- net/netfilter/nf_flow_table_path.c | 7 +-- 4 files changed, 57 insertions(+), 39 deletions(-) diff --git a/include/net/netfilter/nf_flow_table.h b/include/net/netfilter/nf_flow_table.h index e9f72d2558e9..7c330caae52b 100644 --- a/include/net/netfilter/nf_flow_table.h +++ b/include/net/netfilter/nf_flow_table.h @@ -141,6 +141,7 @@ struct flow_offload_tuple { union { struct { struct dst_entry *dst_cache; + u32 ifidx; u32 dst_cookie; }; struct { diff --git a/net/netfilter/nf_flow_table_core.c b/net/netfilter/nf_flow_table_core.c index 9441ac3d8c1a..98d7b3708602 100644 --- a/net/netfilter/nf_flow_table_core.c +++ b/net/netfilter/nf_flow_table_core.c @@ -132,6 +132,7 @@ static int flow_offload_fill_route(struct flow_offload *flow, break; case FLOW_OFFLOAD_XMIT_XFRM: case FLOW_OFFLOAD_XMIT_NEIGH: + flow_tuple->ifidx = route->tuple[dir].out.ifindex; flow_tuple->dst_cache = dst; flow_tuple->dst_cookie = flow_offload_dst_cookie(flow_tuple); break; diff --git a/net/netfilter/nf_flow_table_ip.c b/net/netfilter/nf_flow_table_ip.c index 8cd4cf7ae211..eb4f6a11e779 100644 --- a/net/netfilter/nf_flow_table_ip.c +++ b/net/netfilter/nf_flow_table_ip.c @@ -333,19 +333,18 @@ static void nf_flow_encap_pop(struct sk_buff *skb, } } +struct nf_flow_xmit { + const void *dest; + const void *source; + struct net_device *outdev; +}; + static unsigned int nf_flow_queue_xmit(struct net *net, struct sk_buff *skb, - const struct flow_offload_tuple_rhash *tuplehash, - unsigned short type) + struct nf_flow_xmit *xmit) { - struct net_device *outdev; - - outdev = dev_get_by_index_rcu(net, tuplehash->tuple.out.ifidx); - if (!outdev) - return NF_DROP; - - skb->dev = outdev; - dev_hard_header(skb, skb->dev, type, tuplehash->tuple.out.h_dest, - tuplehash->tuple.out.h_source, skb->len); + skb->dev = xmit->outdev; + dev_hard_header(skb, skb->dev, ntohs(skb->protocol), + xmit->dest, xmit->source, skb->len); dev_queue_xmit(skb); return NF_STOLEN; @@ -424,10 +423,10 @@ nf_flow_offload_ip_hook(void *priv, struct sk_buff *skb, struct nf_flowtable_ctx ctx = { .in = state->in, }; + struct nf_flow_xmit xmit = {}; struct flow_offload *flow; - struct net_device *outdev; + struct neighbour *neigh; struct rtable *rt; - __be32 nexthop; int ret; tuplehash = nf_flow_offload_lookup(&ctx, flow_table, skb); @@ -454,25 +453,34 @@ nf_flow_offload_ip_hook(void *priv, struct sk_buff *skb, switch (tuplehash->tuple.xmit_type) { case FLOW_OFFLOAD_XMIT_NEIGH: rt = dst_rtable(tuplehash->tuple.dst_cache); - outdev = rt->dst.dev; - skb->dev = outdev; - nexthop = rt_nexthop(rt, flow->tuplehash[!dir].tuple.src_v4.s_addr); + xmit.outdev = dev_get_by_index_rcu(state->net, tuplehash->tuple.ifidx); + if (!xmit.outdev) { + flow_offload_teardown(flow); + return NF_DROP; + } + neigh = ip_neigh_gw4(rt->dst.dev, rt_nexthop(rt, flow->tuplehash[!dir].tuple.src_v4.s_addr)); + if (IS_ERR(neigh)) { + flow_offload_teardown(flow); + return NF_DROP; + } + xmit.dest = neigh->ha; skb_dst_set_noref(skb, &rt->dst); - neigh_xmit(NEIGH_ARP_TABLE, outdev, &nexthop, skb); - ret = NF_STOLEN; break; case FLOW_OFFLOAD_XMIT_DIRECT: - ret = nf_flow_queue_xmit(state->net, skb, tuplehash, ETH_P_IP); - if (ret == NF_DROP) + xmit.outdev = dev_get_by_index_rcu(state->net, tuplehash->tuple.out.ifidx); + if (!xmit.outdev) { flow_offload_teardown(flow); + return NF_DROP; + } + xmit.dest = tuplehash->tuple.out.h_dest; + xmit.source = tuplehash->tuple.out.h_source; break; default: WARN_ON_ONCE(1); - ret = NF_DROP; - break; + return NF_DROP; } - return ret; + return nf_flow_queue_xmit(state->net, skb, &xmit); } EXPORT_SYMBOL_GPL(nf_flow_offload_ip_hook); @@ -719,9 +727,9 @@ nf_flow_offload_ipv6_hook(void *priv, struct sk_buff *skb, struct nf_flowtable_ctx ctx = { .in = state->in, }; - const struct in6_addr *nexthop; + struct nf_flow_xmit xmit = {}; struct flow_offload *flow; - struct net_device *outdev; + struct neighbour *neigh; struct rt6_info *rt; int ret; @@ -749,24 +757,33 @@ nf_flow_offload_ipv6_hook(void *priv, struct sk_buff *skb, switch (tuplehash->tuple.xmit_type) { case FLOW_OFFLOAD_XMIT_NEIGH: rt = dst_rt6_info(tuplehash->tuple.dst_cache); - outdev = rt->dst.dev; - skb->dev = outdev; - nexthop = rt6_nexthop(rt, &flow->tuplehash[!dir].tuple.src_v6); + xmit.outdev = dev_get_by_index_rcu(state->net, tuplehash->tuple.ifidx); + if (!xmit.outdev) { + flow_offload_teardown(flow); + return NF_DROP; + } + neigh = ip_neigh_gw6(rt->dst.dev, rt6_nexthop(rt, &flow->tuplehash[!dir].tuple.src_v6)); + if (IS_ERR(neigh)) { + flow_offload_teardown(flow); + return NF_DROP; + } + xmit.dest = neigh->ha; skb_dst_set_noref(skb, &rt->dst); - neigh_xmit(NEIGH_ND_TABLE, outdev, nexthop, skb); - ret = NF_STOLEN; break; case FLOW_OFFLOAD_XMIT_DIRECT: - ret = nf_flow_queue_xmit(state->net, skb, tuplehash, ETH_P_IPV6); - if (ret == NF_DROP) + xmit.outdev = dev_get_by_index_rcu(state->net, tuplehash->tuple.out.ifidx); + if (!xmit.outdev) { flow_offload_teardown(flow); + return NF_DROP; + } + xmit.dest = tuplehash->tuple.out.h_dest; + xmit.source = tuplehash->tuple.out.h_source; break; default: WARN_ON_ONCE(1); - ret = NF_DROP; - break; + return NF_DROP; } - return ret; + return nf_flow_queue_xmit(state->net, skb, &xmit); } EXPORT_SYMBOL_GPL(nf_flow_offload_ipv6_hook); diff --git a/net/netfilter/nf_flow_table_path.c b/net/netfilter/nf_flow_table_path.c index e525e3745651..e0c69fea2e0c 100644 --- a/net/netfilter/nf_flow_table_path.c +++ b/net/netfilter/nf_flow_table_path.c @@ -211,11 +211,11 @@ static void nft_dev_forward_path(struct nf_flow_route *route, } route->tuple[!dir].in.num_encaps = info.num_encaps; route->tuple[!dir].in.ingress_vlans = info.ingress_vlans; + route->tuple[dir].out.ifindex = info.outdev->ifindex; if (info.xmit_type == FLOW_OFFLOAD_XMIT_DIRECT) { memcpy(route->tuple[dir].out.h_source, info.h_source, ETH_ALEN); memcpy(route->tuple[dir].out.h_dest, info.h_dest, ETH_ALEN); - route->tuple[dir].out.ifindex = info.outdev->ifindex; route->tuple[dir].out.hw_ifindex = info.hw_outdev->ifindex; route->tuple[dir].xmit_type = info.xmit_type; } @@ -263,11 +263,10 @@ int nft_flow_route(const struct nft_pktinfo *pkt, const struct nf_conn *ct, nft_default_forward_path(route, this_dst, dir); nft_default_forward_path(route, other_dst, !dir); - if (route->tuple[dir].xmit_type == FLOW_OFFLOAD_XMIT_NEIGH && - route->tuple[!dir].xmit_type == FLOW_OFFLOAD_XMIT_NEIGH) { + if (route->tuple[dir].xmit_type == FLOW_OFFLOAD_XMIT_NEIGH) nft_dev_forward_path(route, ct, dir, ft); + if (route->tuple[!dir].xmit_type == FLOW_OFFLOAD_XMIT_NEIGH) nft_dev_forward_path(route, ct, !dir, ft); - } return 0; } From c653d5a78f3463cfd4c80909425510fb930b0641 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Fri, 10 Oct 2025 12:39:15 +0200 Subject: [PATCH 1214/1381] netfilter: flowtable: inline vlan encapsulation in xmit path Push the vlan header from the flowtable xmit path, instead of passing the packet to the vlan device. This is based on a patch originally written by wenxu. Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_flow_table_ip.c | 25 +++++++++++++++++++++++++ net/netfilter/nf_flow_table_path.c | 7 ++++--- 2 files changed, 29 insertions(+), 3 deletions(-) diff --git a/net/netfilter/nf_flow_table_ip.c b/net/netfilter/nf_flow_table_ip.c index eb4f6a11e779..aef799edca9f 100644 --- a/net/netfilter/nf_flow_table_ip.c +++ b/net/netfilter/nf_flow_table_ip.c @@ -413,6 +413,25 @@ static int nf_flow_offload_forward(struct nf_flowtable_ctx *ctx, return 1; } +static int nf_flow_encap_push(struct sk_buff *skb, + struct flow_offload_tuple *tuple) +{ + int i; + + for (i = 0; i < tuple->encap_num; i++) { + switch (tuple->encap[i].proto) { + case htons(ETH_P_8021Q): + case htons(ETH_P_8021AD): + if (skb_vlan_push(skb, tuple->encap[i].proto, + tuple->encap[i].id) < 0) + return -1; + break; + } + } + + return 0; +} + unsigned int nf_flow_offload_ip_hook(void *priv, struct sk_buff *skb, const struct nf_hook_state *state) @@ -450,6 +469,9 @@ nf_flow_offload_ip_hook(void *priv, struct sk_buff *skb, dir = tuplehash->tuple.dir; flow = container_of(tuplehash, struct flow_offload, tuplehash[dir]); + if (nf_flow_encap_push(skb, &flow->tuplehash[!dir].tuple) < 0) + return NF_DROP; + switch (tuplehash->tuple.xmit_type) { case FLOW_OFFLOAD_XMIT_NEIGH: rt = dst_rtable(tuplehash->tuple.dst_cache); @@ -754,6 +776,9 @@ nf_flow_offload_ipv6_hook(void *priv, struct sk_buff *skb, dir = tuplehash->tuple.dir; flow = container_of(tuplehash, struct flow_offload, tuplehash[dir]); + if (nf_flow_encap_push(skb, &flow->tuplehash[!dir].tuple) < 0) + return NF_DROP; + switch (tuplehash->tuple.xmit_type) { case FLOW_OFFLOAD_XMIT_NEIGH: rt = dst_rt6_info(tuplehash->tuple.dst_cache); diff --git a/net/netfilter/nf_flow_table_path.c b/net/netfilter/nf_flow_table_path.c index e0c69fea2e0c..38ebda1afdce 100644 --- a/net/netfilter/nf_flow_table_path.c +++ b/net/netfilter/nf_flow_table_path.c @@ -119,13 +119,14 @@ static void nft_dev_path_info(const struct net_device_path_stack *stack, info->indev = NULL; break; } - if (!info->outdev) - info->outdev = path->dev; info->encap[info->num_encaps].id = path->encap.id; info->encap[info->num_encaps].proto = path->encap.proto; info->num_encaps++; - if (path->type == DEV_PATH_PPPOE) + if (path->type == DEV_PATH_PPPOE) { + if (!info->outdev) + info->outdev = path->dev; memcpy(info->h_dest, path->encap.h_dest, ETH_ALEN); + } break; case DEV_PATH_BRIDGE: if (is_zero_ether_addr(info->h_source)) From 18d27bed0880b1ac8638a6f4f2edc2ecc6b5c6a1 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Tue, 18 Nov 2025 23:59:47 +0100 Subject: [PATCH 1215/1381] netfilter: flowtable: inline pppoe encapsulation in xmit path Push the pppoe header from the flowtable xmit path, inlining is faster than the original xmit path because it can avoid some locking. This is based on a patch originally written by wenxu. Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_flow_table_ip.c | 42 ++++++++++++++++++++++++++++++ net/netfilter/nf_flow_table_path.c | 9 ++----- 2 files changed, 44 insertions(+), 7 deletions(-) diff --git a/net/netfilter/nf_flow_table_ip.c b/net/netfilter/nf_flow_table_ip.c index aef799edca9f..ee6ec63257d1 100644 --- a/net/netfilter/nf_flow_table_ip.c +++ b/net/netfilter/nf_flow_table_ip.c @@ -413,6 +413,44 @@ static int nf_flow_offload_forward(struct nf_flowtable_ctx *ctx, return 1; } +static int nf_flow_pppoe_push(struct sk_buff *skb, u16 id) +{ + int data_len = skb->len + sizeof(__be16); + struct ppp_hdr { + struct pppoe_hdr hdr; + __be16 proto; + } *ph; + __be16 proto; + + if (skb_cow_head(skb, PPPOE_SES_HLEN)) + return -1; + + switch (skb->protocol) { + case htons(ETH_P_IP): + proto = htons(PPP_IP); + break; + case htons(ETH_P_IPV6): + proto = htons(PPP_IPV6); + break; + default: + return -1; + } + + __skb_push(skb, PPPOE_SES_HLEN); + skb_reset_network_header(skb); + + ph = (struct ppp_hdr *)(skb->data); + ph->hdr.ver = 1; + ph->hdr.type = 1; + ph->hdr.code = 0; + ph->hdr.sid = htons(id); + ph->hdr.length = htons(data_len); + ph->proto = proto; + skb->protocol = htons(ETH_P_PPP_SES); + + return 0; +} + static int nf_flow_encap_push(struct sk_buff *skb, struct flow_offload_tuple *tuple) { @@ -426,6 +464,10 @@ static int nf_flow_encap_push(struct sk_buff *skb, tuple->encap[i].id) < 0) return -1; break; + case htons(ETH_P_PPP_SES): + if (nf_flow_pppoe_push(skb, tuple->encap[i].id) < 0) + return -1; + break; } } diff --git a/net/netfilter/nf_flow_table_path.c b/net/netfilter/nf_flow_table_path.c index 38ebda1afdce..c51e310bb2ab 100644 --- a/net/netfilter/nf_flow_table_path.c +++ b/net/netfilter/nf_flow_table_path.c @@ -122,11 +122,8 @@ static void nft_dev_path_info(const struct net_device_path_stack *stack, info->encap[info->num_encaps].id = path->encap.id; info->encap[info->num_encaps].proto = path->encap.proto; info->num_encaps++; - if (path->type == DEV_PATH_PPPOE) { - if (!info->outdev) - info->outdev = path->dev; + if (path->type == DEV_PATH_PPPOE) memcpy(info->h_dest, path->encap.h_dest, ETH_ALEN); - } break; case DEV_PATH_BRIDGE: if (is_zero_ether_addr(info->h_source)) @@ -161,9 +158,7 @@ static void nft_dev_path_info(const struct net_device_path_stack *stack, break; } } - if (!info->outdev) - info->outdev = info->indev; - + info->outdev = info->indev; info->hw_outdev = info->indev; if (nf_flowtable_hw_offload(flowtable) && From 030feea3097c41ed268c81240e5c334d9977b1c4 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Fri, 10 Oct 2025 12:50:07 +0200 Subject: [PATCH 1216/1381] netfilter: flowtable: remove hw_ifidx hw_ifidx was originally introduced to store the real netdevice as a requirement for the hardware offload support in: 73f97025a972 ("netfilter: nft_flow_offload: use direct xmit if hardware offload is enabled") Since ("netfilter: flowtable: consolidate xmit path"), ifidx and hw_ifidx points to the real device in the xmit path, remove it. Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_flow_table.h | 1 - net/netfilter/nf_flow_table_core.c | 1 - net/netfilter/nf_flow_table_offload.c | 2 +- net/netfilter/nf_flow_table_path.c | 3 --- 4 files changed, 1 insertion(+), 6 deletions(-) diff --git a/include/net/netfilter/nf_flow_table.h b/include/net/netfilter/nf_flow_table.h index 7c330caae52b..f7306276ece7 100644 --- a/include/net/netfilter/nf_flow_table.h +++ b/include/net/netfilter/nf_flow_table.h @@ -146,7 +146,6 @@ struct flow_offload_tuple { }; struct { u32 ifidx; - u32 hw_ifidx; u8 h_source[ETH_ALEN]; u8 h_dest[ETH_ALEN]; } out; diff --git a/net/netfilter/nf_flow_table_core.c b/net/netfilter/nf_flow_table_core.c index 98d7b3708602..6c6a5165f993 100644 --- a/net/netfilter/nf_flow_table_core.c +++ b/net/netfilter/nf_flow_table_core.c @@ -127,7 +127,6 @@ static int flow_offload_fill_route(struct flow_offload *flow, memcpy(flow_tuple->out.h_source, route->tuple[dir].out.h_source, ETH_ALEN); flow_tuple->out.ifidx = route->tuple[dir].out.ifindex; - flow_tuple->out.hw_ifidx = route->tuple[dir].out.hw_ifindex; dst_release(dst); break; case FLOW_OFFLOAD_XMIT_XFRM: diff --git a/net/netfilter/nf_flow_table_offload.c b/net/netfilter/nf_flow_table_offload.c index e06bc36f49fe..d8f7bfd60ac6 100644 --- a/net/netfilter/nf_flow_table_offload.c +++ b/net/netfilter/nf_flow_table_offload.c @@ -555,7 +555,7 @@ static void flow_offload_redirect(struct net *net, switch (this_tuple->xmit_type) { case FLOW_OFFLOAD_XMIT_DIRECT: this_tuple = &flow->tuplehash[dir].tuple; - ifindex = this_tuple->out.hw_ifidx; + ifindex = this_tuple->out.ifidx; break; case FLOW_OFFLOAD_XMIT_NEIGH: other_tuple = &flow->tuplehash[!dir].tuple; diff --git a/net/netfilter/nf_flow_table_path.c b/net/netfilter/nf_flow_table_path.c index c51e310bb2ab..eb9b33a1873a 100644 --- a/net/netfilter/nf_flow_table_path.c +++ b/net/netfilter/nf_flow_table_path.c @@ -75,7 +75,6 @@ static int nft_dev_fill_forward_path(const struct nf_flow_route *route, struct nft_forward_info { const struct net_device *indev; const struct net_device *outdev; - const struct net_device *hw_outdev; struct id { __u16 id; __be16 proto; @@ -159,7 +158,6 @@ static void nft_dev_path_info(const struct net_device_path_stack *stack, } } info->outdev = info->indev; - info->hw_outdev = info->indev; if (nf_flowtable_hw_offload(flowtable) && nft_is_valid_ether_device(info->indev)) @@ -212,7 +210,6 @@ static void nft_dev_forward_path(struct nf_flow_route *route, if (info.xmit_type == FLOW_OFFLOAD_XMIT_DIRECT) { memcpy(route->tuple[dir].out.h_source, info.h_source, ETH_ALEN); memcpy(route->tuple[dir].out.h_dest, info.h_dest, ETH_ALEN); - route->tuple[dir].out.hw_ifindex = info.hw_outdev->ifindex; route->tuple[dir].xmit_type = info.xmit_type; } } From a0d98b641d676e9fc5c458b14aee8ee874dd7298 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Wed, 12 Nov 2025 22:54:24 +0100 Subject: [PATCH 1217/1381] netfilter: flowtable: use tuple address to calculate next hop This simplifies IPIP tunnel support coming in follow up patches. No function changes are intended. Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_flow_table_ip.c | 16 ++++++++++++---- 1 file changed, 12 insertions(+), 4 deletions(-) diff --git a/net/netfilter/nf_flow_table_ip.c b/net/netfilter/nf_flow_table_ip.c index ee6ec63257d1..083ceb64ac17 100644 --- a/net/netfilter/nf_flow_table_ip.c +++ b/net/netfilter/nf_flow_table_ip.c @@ -480,6 +480,7 @@ nf_flow_offload_ip_hook(void *priv, struct sk_buff *skb, { struct flow_offload_tuple_rhash *tuplehash; struct nf_flowtable *flow_table = priv; + struct flow_offload_tuple *other_tuple; enum flow_offload_tuple_dir dir; struct nf_flowtable_ctx ctx = { .in = state->in, @@ -488,6 +489,7 @@ nf_flow_offload_ip_hook(void *priv, struct sk_buff *skb, struct flow_offload *flow; struct neighbour *neigh; struct rtable *rt; + __be32 ip_daddr; int ret; tuplehash = nf_flow_offload_lookup(&ctx, flow_table, skb); @@ -510,8 +512,10 @@ nf_flow_offload_ip_hook(void *priv, struct sk_buff *skb, dir = tuplehash->tuple.dir; flow = container_of(tuplehash, struct flow_offload, tuplehash[dir]); + other_tuple = &flow->tuplehash[!dir].tuple; + ip_daddr = other_tuple->src_v4.s_addr; - if (nf_flow_encap_push(skb, &flow->tuplehash[!dir].tuple) < 0) + if (nf_flow_encap_push(skb, other_tuple) < 0) return NF_DROP; switch (tuplehash->tuple.xmit_type) { @@ -522,7 +526,7 @@ nf_flow_offload_ip_hook(void *priv, struct sk_buff *skb, flow_offload_teardown(flow); return NF_DROP; } - neigh = ip_neigh_gw4(rt->dst.dev, rt_nexthop(rt, flow->tuplehash[!dir].tuple.src_v4.s_addr)); + neigh = ip_neigh_gw4(rt->dst.dev, rt_nexthop(rt, ip_daddr)); if (IS_ERR(neigh)) { flow_offload_teardown(flow); return NF_DROP; @@ -787,11 +791,13 @@ nf_flow_offload_ipv6_hook(void *priv, struct sk_buff *skb, { struct flow_offload_tuple_rhash *tuplehash; struct nf_flowtable *flow_table = priv; + struct flow_offload_tuple *other_tuple; enum flow_offload_tuple_dir dir; struct nf_flowtable_ctx ctx = { .in = state->in, }; struct nf_flow_xmit xmit = {}; + struct in6_addr *ip6_daddr; struct flow_offload *flow; struct neighbour *neigh; struct rt6_info *rt; @@ -817,8 +823,10 @@ nf_flow_offload_ipv6_hook(void *priv, struct sk_buff *skb, dir = tuplehash->tuple.dir; flow = container_of(tuplehash, struct flow_offload, tuplehash[dir]); + other_tuple = &flow->tuplehash[!dir].tuple; + ip6_daddr = &other_tuple->src_v6; - if (nf_flow_encap_push(skb, &flow->tuplehash[!dir].tuple) < 0) + if (nf_flow_encap_push(skb, other_tuple) < 0) return NF_DROP; switch (tuplehash->tuple.xmit_type) { @@ -829,7 +837,7 @@ nf_flow_offload_ipv6_hook(void *priv, struct sk_buff *skb, flow_offload_teardown(flow); return NF_DROP; } - neigh = ip_neigh_gw6(rt->dst.dev, rt6_nexthop(rt, &flow->tuplehash[!dir].tuple.src_v6)); + neigh = ip_neigh_gw6(rt->dst.dev, rt6_nexthop(rt, ip6_daddr)); if (IS_ERR(neigh)) { flow_offload_teardown(flow); return NF_DROP; From ab427db17885814069bae891834f20842f0ac3a4 Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Fri, 7 Nov 2025 12:14:46 +0100 Subject: [PATCH 1218/1381] netfilter: flowtable: Add IPIP rx sw acceleration Introduce sw acceleration for rx path of IPIP tunnels relying on the netfilter flowtable infrastructure. Subsequent patches will add sw acceleration for IPIP tunnels tx path. This series introduces basic infrastructure to accelerate other tunnel types (e.g. IP6IP6). IPIP rx sw acceleration can be tested running the following scenario where the traffic is forwarded between two NICs (eth0 and eth1) and an IPIP tunnel is used to access a remote site (using eth1 as the underlay device): ETH0 -- TUN0 <==> ETH1 -- [IP network] -- TUN1 (192.168.100.2) $ip addr show 6: eth0: mtu 1500 qdisc noqueue state UP group default qlen 1000 link/ether 00:00:22:33:11:55 brd ff:ff:ff:ff:ff:ff inet 192.168.0.2/24 scope global eth0 valid_lft forever preferred_lft forever 7: eth1: mtu 1500 qdisc noqueue state UP group default qlen 1000 link/ether 00:11:22:33:11:55 brd ff:ff:ff:ff:ff:ff inet 192.168.1.1/24 scope global eth1 valid_lft forever preferred_lft forever 8: tun0@NONE: mtu 1480 qdisc noqueue state UNKNOWN group default qlen 1000 link/ipip 192.168.1.1 peer 192.168.1.2 inet 192.168.100.1/24 scope global tun0 valid_lft forever preferred_lft forever $ip route show default via 192.168.100.2 dev tun0 192.168.0.0/24 dev eth0 proto kernel scope link src 192.168.0.2 192.168.1.0/24 dev eth1 proto kernel scope link src 192.168.1.1 192.168.100.0/24 dev tun0 proto kernel scope link src 192.168.100.1 $nft list ruleset table inet filter { flowtable ft { hook ingress priority filter devices = { eth0, eth1 } } chain forward { type filter hook forward priority filter; policy accept; meta l4proto { tcp, udp } flow add @ft } } Reproducing the scenario described above using veths I got the following results: - TCP stream received from the IPIP tunnel: - net-next: (baseline) ~ 71Gbps - net-next + IPIP flowtbale support: ~101Gbps Signed-off-by: Lorenzo Bianconi Signed-off-by: Pablo Neira Ayuso --- include/linux/netdevice.h | 13 +++++ include/net/netfilter/nf_flow_table.h | 18 +++++++ net/ipv4/ipip.c | 25 ++++++++++ net/netfilter/nf_flow_table_core.c | 3 ++ net/netfilter/nf_flow_table_ip.c | 69 ++++++++++++++++++++++++--- net/netfilter/nf_flow_table_path.c | 38 ++++++++++++--- 6 files changed, 153 insertions(+), 13 deletions(-) diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index e808071dbb7d..bf99fe8622da 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -877,6 +877,7 @@ enum net_device_path_type { DEV_PATH_PPPOE, DEV_PATH_DSA, DEV_PATH_MTK_WDMA, + DEV_PATH_TUN, }; struct net_device_path { @@ -888,6 +889,18 @@ struct net_device_path { __be16 proto; u8 h_dest[ETH_ALEN]; } encap; + struct { + union { + struct in_addr src_v4; + struct in6_addr src_v6; + }; + union { + struct in_addr dst_v4; + struct in6_addr dst_v6; + }; + + u8 l3_proto; + } tun; struct { enum { DEV_PATH_BR_VLAN_KEEP, diff --git a/include/net/netfilter/nf_flow_table.h b/include/net/netfilter/nf_flow_table.h index f7306276ece7..b09c11c048d5 100644 --- a/include/net/netfilter/nf_flow_table.h +++ b/include/net/netfilter/nf_flow_table.h @@ -107,6 +107,19 @@ enum flow_offload_xmit_type { #define NF_FLOW_TABLE_ENCAP_MAX 2 +struct flow_offload_tunnel { + union { + struct in_addr src_v4; + struct in6_addr src_v6; + }; + union { + struct in_addr dst_v4; + struct in6_addr dst_v6; + }; + + u8 l3_proto; +}; + struct flow_offload_tuple { union { struct in_addr src_v4; @@ -130,12 +143,15 @@ struct flow_offload_tuple { __be16 proto; } encap[NF_FLOW_TABLE_ENCAP_MAX]; + struct flow_offload_tunnel tun; + /* All members above are keys for lookups, see flow_offload_hash(). */ struct { } __hash; u8 dir:2, xmit_type:3, encap_num:2, + tun_num:2, in_vlan_ingress:2; u16 mtu; union { @@ -206,7 +222,9 @@ struct nf_flow_route { u16 id; __be16 proto; } encap[NF_FLOW_TABLE_ENCAP_MAX]; + struct flow_offload_tunnel tun; u8 num_encaps:2, + num_tuns:2, ingress_vlans:2; } in; struct { diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c index 3e03af073a1c..ff95b1b9908e 100644 --- a/net/ipv4/ipip.c +++ b/net/ipv4/ipip.c @@ -353,6 +353,30 @@ ipip_tunnel_ctl(struct net_device *dev, struct ip_tunnel_parm_kern *p, int cmd) return ip_tunnel_ctl(dev, p, cmd); } +static int ipip_fill_forward_path(struct net_device_path_ctx *ctx, + struct net_device_path *path) +{ + struct ip_tunnel *tunnel = netdev_priv(ctx->dev); + const struct iphdr *tiph = &tunnel->parms.iph; + struct rtable *rt; + + rt = ip_route_output(dev_net(ctx->dev), tiph->daddr, 0, 0, 0, + RT_SCOPE_UNIVERSE); + if (IS_ERR(rt)) + return PTR_ERR(rt); + + path->type = DEV_PATH_TUN; + path->tun.src_v4.s_addr = tiph->saddr; + path->tun.dst_v4.s_addr = tiph->daddr; + path->tun.l3_proto = IPPROTO_IPIP; + path->dev = ctx->dev; + + ctx->dev = rt->dst.dev; + ip_rt_put(rt); + + return 0; +} + static const struct net_device_ops ipip_netdev_ops = { .ndo_init = ipip_tunnel_init, .ndo_uninit = ip_tunnel_uninit, @@ -362,6 +386,7 @@ static const struct net_device_ops ipip_netdev_ops = { .ndo_get_stats64 = dev_get_tstats64, .ndo_get_iflink = ip_tunnel_get_iflink, .ndo_tunnel_ctl = ipip_tunnel_ctl, + .ndo_fill_forward_path = ipip_fill_forward_path, }; #define IPIP_FEATURES (NETIF_F_SG | \ diff --git a/net/netfilter/nf_flow_table_core.c b/net/netfilter/nf_flow_table_core.c index 6c6a5165f993..06e8251a6644 100644 --- a/net/netfilter/nf_flow_table_core.c +++ b/net/netfilter/nf_flow_table_core.c @@ -118,7 +118,10 @@ static int flow_offload_fill_route(struct flow_offload *flow, flow_tuple->in_vlan_ingress |= BIT(j); j++; } + + flow_tuple->tun = route->tuple[dir].in.tun; flow_tuple->encap_num = route->tuple[dir].in.num_encaps; + flow_tuple->tun_num = route->tuple[dir].in.num_tuns; switch (route->tuple[dir].xmit_type) { case FLOW_OFFLOAD_XMIT_DIRECT: diff --git a/net/netfilter/nf_flow_table_ip.c b/net/netfilter/nf_flow_table_ip.c index 083ceb64ac17..d6a1f0cda189 100644 --- a/net/netfilter/nf_flow_table_ip.c +++ b/net/netfilter/nf_flow_table_ip.c @@ -145,8 +145,11 @@ static bool ip_has_options(unsigned int thoff) static void nf_flow_tuple_encap(struct sk_buff *skb, struct flow_offload_tuple *tuple) { + __be16 inner_proto = skb->protocol; struct vlan_ethhdr *veth; struct pppoe_hdr *phdr; + struct iphdr *iph; + u16 offset = 0; int i = 0; if (skb_vlan_tag_present(skb)) { @@ -159,13 +162,26 @@ static void nf_flow_tuple_encap(struct sk_buff *skb, veth = (struct vlan_ethhdr *)skb_mac_header(skb); tuple->encap[i].id = ntohs(veth->h_vlan_TCI); tuple->encap[i].proto = skb->protocol; + inner_proto = veth->h_vlan_encapsulated_proto; + offset += VLAN_HLEN; break; case htons(ETH_P_PPP_SES): phdr = (struct pppoe_hdr *)skb_network_header(skb); tuple->encap[i].id = ntohs(phdr->sid); tuple->encap[i].proto = skb->protocol; + inner_proto = *((__be16 *)(phdr + 1)); + offset += PPPOE_SES_HLEN; break; } + + if (inner_proto == htons(ETH_P_IP)) { + iph = (struct iphdr *)(skb_network_header(skb) + offset); + if (iph->protocol == IPPROTO_IPIP) { + tuple->tun.dst_v4.s_addr = iph->daddr; + tuple->tun.src_v4.s_addr = iph->saddr; + tuple->tun.l3_proto = IPPROTO_IPIP; + } + } } struct nf_flowtable_ctx { @@ -277,11 +293,46 @@ static unsigned int nf_flow_xmit_xfrm(struct sk_buff *skb, return NF_STOLEN; } +static bool nf_flow_ip4_tunnel_proto(struct sk_buff *skb, u32 *psize) +{ + struct iphdr *iph; + u16 size; + + if (!pskb_may_pull(skb, sizeof(*iph) + *psize)) + return false; + + iph = (struct iphdr *)(skb_network_header(skb) + *psize); + size = iph->ihl << 2; + + if (ip_is_fragment(iph) || unlikely(ip_has_options(size))) + return false; + + if (iph->ttl <= 1) + return false; + + if (iph->protocol == IPPROTO_IPIP) + *psize += size; + + return true; +} + +static void nf_flow_ip4_tunnel_pop(struct sk_buff *skb) +{ + struct iphdr *iph = (struct iphdr *)skb_network_header(skb); + + if (iph->protocol != IPPROTO_IPIP) + return; + + skb_pull(skb, iph->ihl << 2); + skb_reset_network_header(skb); +} + static bool nf_flow_skb_encap_protocol(struct sk_buff *skb, __be16 proto, u32 *offset) { + __be16 inner_proto = skb->protocol; struct vlan_ethhdr *veth; - __be16 inner_proto; + bool ret = false; switch (skb->protocol) { case htons(ETH_P_8021Q): @@ -291,19 +342,23 @@ static bool nf_flow_skb_encap_protocol(struct sk_buff *skb, __be16 proto, veth = (struct vlan_ethhdr *)skb_mac_header(skb); if (veth->h_vlan_encapsulated_proto == proto) { *offset += VLAN_HLEN; - return true; + inner_proto = proto; + ret = true; } break; case htons(ETH_P_PPP_SES): if (nf_flow_pppoe_proto(skb, &inner_proto) && inner_proto == proto) { *offset += PPPOE_SES_HLEN; - return true; + ret = true; } break; } - return false; + if (inner_proto == htons(ETH_P_IP)) + ret = nf_flow_ip4_tunnel_proto(skb, offset); + + return ret; } static void nf_flow_encap_pop(struct sk_buff *skb, @@ -331,6 +386,9 @@ static void nf_flow_encap_pop(struct sk_buff *skb, break; } } + + if (skb->protocol == htons(ETH_P_IP)) + nf_flow_ip4_tunnel_pop(skb); } struct nf_flow_xmit { @@ -356,8 +414,7 @@ nf_flow_offload_lookup(struct nf_flowtable_ctx *ctx, { struct flow_offload_tuple tuple = {}; - if (skb->protocol != htons(ETH_P_IP) && - !nf_flow_skb_encap_protocol(skb, htons(ETH_P_IP), &ctx->offset)) + if (!nf_flow_skb_encap_protocol(skb, htons(ETH_P_IP), &ctx->offset)) return NULL; if (nf_flow_tuple_ip(ctx, skb, &tuple) < 0) diff --git a/net/netfilter/nf_flow_table_path.c b/net/netfilter/nf_flow_table_path.c index eb9b33a1873a..73717cc32df5 100644 --- a/net/netfilter/nf_flow_table_path.c +++ b/net/netfilter/nf_flow_table_path.c @@ -80,6 +80,8 @@ struct nft_forward_info { __be16 proto; } encap[NF_FLOW_TABLE_ENCAP_MAX]; u8 num_encaps; + struct flow_offload_tunnel tun; + u8 num_tuns; u8 ingress_vlans; u8 h_source[ETH_ALEN]; u8 h_dest[ETH_ALEN]; @@ -102,6 +104,7 @@ static void nft_dev_path_info(const struct net_device_path_stack *stack, case DEV_PATH_DSA: case DEV_PATH_VLAN: case DEV_PATH_PPPOE: + case DEV_PATH_TUN: info->indev = path->dev; if (is_zero_ether_addr(info->h_source)) memcpy(info->h_source, path->dev->dev_addr, ETH_ALEN); @@ -113,14 +116,27 @@ static void nft_dev_path_info(const struct net_device_path_stack *stack, break; } - /* DEV_PATH_VLAN and DEV_PATH_PPPOE */ - if (info->num_encaps >= NF_FLOW_TABLE_ENCAP_MAX) { - info->indev = NULL; - break; + /* DEV_PATH_VLAN, DEV_PATH_PPPOE and DEV_PATH_TUN */ + if (path->type == DEV_PATH_TUN) { + if (info->num_tuns) { + info->indev = NULL; + break; + } + info->tun.src_v6 = path->tun.src_v6; + info->tun.dst_v6 = path->tun.dst_v6; + info->tun.l3_proto = path->tun.l3_proto; + info->num_tuns++; + } else { + if (info->num_encaps >= NF_FLOW_TABLE_ENCAP_MAX) { + info->indev = NULL; + break; + } + info->encap[info->num_encaps].id = + path->encap.id; + info->encap[info->num_encaps].proto = + path->encap.proto; + info->num_encaps++; } - info->encap[info->num_encaps].id = path->encap.id; - info->encap[info->num_encaps].proto = path->encap.proto; - info->num_encaps++; if (path->type == DEV_PATH_PPPOE) memcpy(info->h_dest, path->encap.h_dest, ETH_ALEN); break; @@ -203,6 +219,14 @@ static void nft_dev_forward_path(struct nf_flow_route *route, route->tuple[!dir].in.encap[i].id = info.encap[i].id; route->tuple[!dir].in.encap[i].proto = info.encap[i].proto; } + + if (info.num_tuns) { + route->tuple[!dir].in.tun.src_v6 = info.tun.dst_v6; + route->tuple[!dir].in.tun.dst_v6 = info.tun.src_v6; + route->tuple[!dir].in.tun.l3_proto = info.tun.l3_proto; + route->tuple[!dir].in.num_tuns = info.num_tuns; + } + route->tuple[!dir].in.num_encaps = info.num_encaps; route->tuple[!dir].in.ingress_vlans = info.ingress_vlans; route->tuple[dir].out.ifindex = info.outdev->ifindex; From d30301ba4b07ac92eb38353a111833b009003170 Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Fri, 7 Nov 2025 12:14:47 +0100 Subject: [PATCH 1219/1381] netfilter: flowtable: Add IPIP tx sw acceleration Introduce sw acceleration for tx path of IPIP tunnels relying on the netfilter flowtable infrastructure. This patch introduces basic infrastructure to accelerate other tunnel types (e.g. IP6IP6). IPIP sw tx acceleration can be tested running the following scenario where the traffic is forwarded between two NICs (eth0 and eth1) and an IPIP tunnel is used to access a remote site (using eth1 as the underlay device): ETH0 -- TUN0 <==> ETH1 -- [IP network] -- TUN1 (192.168.100.2) $ip addr show 6: eth0: mtu 1500 qdisc noqueue state UP group default qlen 1000 link/ether 00:00:22:33:11:55 brd ff:ff:ff:ff:ff:ff inet 192.168.0.2/24 scope global eth0 valid_lft forever preferred_lft forever 7: eth1: mtu 1500 qdisc noqueue state UP group default qlen 1000 link/ether 00:11:22:33:11:55 brd ff:ff:ff:ff:ff:ff inet 192.168.1.1/24 scope global eth1 valid_lft forever preferred_lft forever 8: tun0@NONE: mtu 1480 qdisc noqueue state UNKNOWN group default qlen 1000 link/ipip 192.168.1.1 peer 192.168.1.2 inet 192.168.100.1/24 scope global tun0 valid_lft forever preferred_lft forever $ip route show default via 192.168.100.2 dev tun0 192.168.0.0/24 dev eth0 proto kernel scope link src 192.168.0.2 192.168.1.0/24 dev eth1 proto kernel scope link src 192.168.1.1 192.168.100.0/24 dev tun0 proto kernel scope link src 192.168.100.1 $nft list ruleset table inet filter { flowtable ft { hook ingress priority filter devices = { eth0, eth1 } } chain forward { type filter hook forward priority filter; policy accept; meta l4proto { tcp, udp } flow add @ft } } Reproducing the scenario described above using veths I got the following results: - TCP stream trasmitted into the IPIP tunnel: - net-next: (baseline) ~ 85Gbps - net-next + IPIP flowtable support: ~102Gbps Co-developed-by: Pablo Neira Ayuso Signed-off-by: Lorenzo Bianconi Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_flow_table_ip.c | 62 ++++++++++++++++++++++++++++++ net/netfilter/nf_flow_table_path.c | 48 +++++++++++++++++++++-- 2 files changed, 106 insertions(+), 4 deletions(-) diff --git a/net/netfilter/nf_flow_table_ip.c b/net/netfilter/nf_flow_table_ip.c index d6a1f0cda189..78883343e5d6 100644 --- a/net/netfilter/nf_flow_table_ip.c +++ b/net/netfilter/nf_flow_table_ip.c @@ -437,6 +437,9 @@ static int nf_flow_offload_forward(struct nf_flowtable_ctx *ctx, flow = container_of(tuplehash, struct flow_offload, tuplehash[dir]); mtu = flow->tuplehash[dir].tuple.mtu + ctx->offset; + if (flow->tuplehash[!dir].tuple.tun_num) + mtu -= sizeof(*iph); + if (unlikely(nf_flow_exceeds_mtu(skb, mtu))) return 0; @@ -508,6 +511,62 @@ static int nf_flow_pppoe_push(struct sk_buff *skb, u16 id) return 0; } +static int nf_flow_tunnel_ipip_push(struct net *net, struct sk_buff *skb, + struct flow_offload_tuple *tuple, + __be32 *ip_daddr) +{ + struct iphdr *iph = (struct iphdr *)skb_network_header(skb); + struct rtable *rt = dst_rtable(tuple->dst_cache); + u8 tos = iph->tos, ttl = iph->ttl; + __be16 frag_off = iph->frag_off; + u32 headroom = sizeof(*iph); + int err; + + err = iptunnel_handle_offloads(skb, SKB_GSO_IPXIP4); + if (err) + return err; + + skb_set_inner_ipproto(skb, IPPROTO_IPIP); + headroom += LL_RESERVED_SPACE(rt->dst.dev) + rt->dst.header_len; + err = skb_cow_head(skb, headroom); + if (err) + return err; + + skb_scrub_packet(skb, true); + skb_clear_hash_if_not_l4(skb); + + /* Push down and install the IP header. */ + skb_push(skb, sizeof(*iph)); + skb_reset_network_header(skb); + + iph = ip_hdr(skb); + iph->version = 4; + iph->ihl = sizeof(*iph) >> 2; + iph->frag_off = ip_mtu_locked(&rt->dst) ? 0 : frag_off; + iph->protocol = tuple->tun.l3_proto; + iph->tos = tos; + iph->daddr = tuple->tun.src_v4.s_addr; + iph->saddr = tuple->tun.dst_v4.s_addr; + iph->ttl = ttl; + iph->tot_len = htons(skb->len); + __ip_select_ident(net, iph, skb_shinfo(skb)->gso_segs ?: 1); + ip_send_check(iph); + + *ip_daddr = tuple->tun.src_v4.s_addr; + + return 0; +} + +static int nf_flow_tunnel_v4_push(struct net *net, struct sk_buff *skb, + struct flow_offload_tuple *tuple, + __be32 *ip_daddr) +{ + if (tuple->tun_num) + return nf_flow_tunnel_ipip_push(net, skb, tuple, ip_daddr); + + return 0; +} + static int nf_flow_encap_push(struct sk_buff *skb, struct flow_offload_tuple *tuple) { @@ -572,6 +631,9 @@ nf_flow_offload_ip_hook(void *priv, struct sk_buff *skb, other_tuple = &flow->tuplehash[!dir].tuple; ip_daddr = other_tuple->src_v4.s_addr; + if (nf_flow_tunnel_v4_push(state->net, skb, other_tuple, &ip_daddr) < 0) + return NF_DROP; + if (nf_flow_encap_push(skb, other_tuple) < 0) return NF_DROP; diff --git a/net/netfilter/nf_flow_table_path.c b/net/netfilter/nf_flow_table_path.c index 73717cc32df5..f0984cf69a09 100644 --- a/net/netfilter/nf_flow_table_path.c +++ b/net/netfilter/nf_flow_table_path.c @@ -197,7 +197,46 @@ static bool nft_flowtable_find_dev(const struct net_device *dev, return found; } -static void nft_dev_forward_path(struct nf_flow_route *route, +static int nft_flow_tunnel_update_route(const struct nft_pktinfo *pkt, + struct flow_offload_tunnel *tun, + struct nf_flow_route *route, + enum ip_conntrack_dir dir) +{ + struct dst_entry *cur_dst = route->tuple[dir].dst; + struct dst_entry *tun_dst = NULL; + struct flowi fl = {}; + + switch (nft_pf(pkt)) { + case NFPROTO_IPV4: + fl.u.ip4.daddr = tun->dst_v4.s_addr; + fl.u.ip4.saddr = tun->src_v4.s_addr; + fl.u.ip4.flowi4_iif = nft_in(pkt)->ifindex; + fl.u.ip4.flowi4_dscp = ip4h_dscp(ip_hdr(pkt->skb)); + fl.u.ip4.flowi4_mark = pkt->skb->mark; + fl.u.ip4.flowi4_flags = FLOWI_FLAG_ANYSRC; + break; + case NFPROTO_IPV6: + fl.u.ip6.daddr = tun->dst_v6; + fl.u.ip6.saddr = tun->src_v6; + fl.u.ip6.flowi6_iif = nft_in(pkt)->ifindex; + fl.u.ip6.flowlabel = ip6_flowinfo(ipv6_hdr(pkt->skb)); + fl.u.ip6.flowi6_mark = pkt->skb->mark; + fl.u.ip6.flowi6_flags = FLOWI_FLAG_ANYSRC; + break; + } + + nf_route(nft_net(pkt), &tun_dst, &fl, false, nft_pf(pkt)); + if (!tun_dst) + return -ENOENT; + + route->tuple[dir].dst = tun_dst; + dst_release(cur_dst); + + return 0; +} + +static void nft_dev_forward_path(const struct nft_pktinfo *pkt, + struct nf_flow_route *route, const struct nf_conn *ct, enum ip_conntrack_dir dir, struct nft_flowtable *ft) @@ -220,7 +259,8 @@ static void nft_dev_forward_path(struct nf_flow_route *route, route->tuple[!dir].in.encap[i].proto = info.encap[i].proto; } - if (info.num_tuns) { + if (info.num_tuns && + !nft_flow_tunnel_update_route(pkt, &info.tun, route, dir)) { route->tuple[!dir].in.tun.src_v6 = info.tun.dst_v6; route->tuple[!dir].in.tun.dst_v6 = info.tun.src_v6; route->tuple[!dir].in.tun.l3_proto = info.tun.l3_proto; @@ -281,9 +321,9 @@ int nft_flow_route(const struct nft_pktinfo *pkt, const struct nf_conn *ct, nft_default_forward_path(route, other_dst, !dir); if (route->tuple[dir].xmit_type == FLOW_OFFLOAD_XMIT_NEIGH) - nft_dev_forward_path(route, ct, dir, ft); + nft_dev_forward_path(pkt, route, ct, dir, ft); if (route->tuple[!dir].xmit_type == FLOW_OFFLOAD_XMIT_NEIGH) - nft_dev_forward_path(route, ct, !dir, ft); + nft_dev_forward_path(pkt, route, ct, !dir, ft); return 0; } From fe8313316eaf0f3b052d5a9464ca6aa630229400 Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Fri, 7 Nov 2025 12:14:48 +0100 Subject: [PATCH 1220/1381] selftests: netfilter: nft_flowtable.sh: Add IPIP flowtable selftest Introduce specific selftest for IPIP flowtable SW acceleration in nft_flowtable.sh Signed-off-by: Lorenzo Bianconi Signed-off-by: Pablo Neira Ayuso --- .../selftests/net/netfilter/nft_flowtable.sh | 69 +++++++++++++++++++ 1 file changed, 69 insertions(+) diff --git a/tools/testing/selftests/net/netfilter/nft_flowtable.sh b/tools/testing/selftests/net/netfilter/nft_flowtable.sh index 45832df98295..1fbfc8ad8dcd 100755 --- a/tools/testing/selftests/net/netfilter/nft_flowtable.sh +++ b/tools/testing/selftests/net/netfilter/nft_flowtable.sh @@ -558,6 +558,73 @@ if ! test_tcp_forwarding_nat "$ns1" "$ns2" 1 ""; then ip netns exec "$nsr1" nft list ruleset fi +# IPIP tunnel test: +# Add IPIP tunnel interfaces and check flowtable acceleration. +test_ipip() { +if ! ip -net "$nsr1" link add name tun0 type ipip \ + local 192.168.10.1 remote 192.168.10.2 >/dev/null;then + echo "SKIP: could not add ipip tunnel" + [ "$ret" -eq 0 ] && ret=$ksft_skip + return +fi +ip -net "$nsr1" link set tun0 up +ip -net "$nsr1" addr add 192.168.100.1/24 dev tun0 +ip netns exec "$nsr1" sysctl net.ipv4.conf.tun0.forwarding=1 > /dev/null + +ip -net "$nsr2" link add name tun0 type ipip local 192.168.10.2 remote 192.168.10.1 +ip -net "$nsr2" link set tun0 up +ip -net "$nsr2" addr add 192.168.100.2/24 dev tun0 +ip netns exec "$nsr2" sysctl net.ipv4.conf.tun0.forwarding=1 > /dev/null + +ip -net "$nsr1" route change default via 192.168.100.2 +ip -net "$nsr2" route change default via 192.168.100.1 +ip -net "$ns2" route add default via 10.0.2.1 + +ip netns exec "$nsr1" nft -a insert rule inet filter forward 'meta oif tun0 accept' +ip netns exec "$nsr1" nft -a insert rule inet filter forward \ + 'meta oif "veth0" tcp sport 12345 ct mark set 1 flow add @f1 counter name routed_repl accept' + +if ! test_tcp_forwarding_nat "$ns1" "$ns2" 1 "IPIP tunnel"; then + echo "FAIL: flow offload for ns1/ns2 with IPIP tunnel" 1>&2 + ip netns exec "$nsr1" nft list ruleset + ret=1 +fi + +# Create vlan tagged devices for IPIP traffic. +ip -net "$nsr1" link add link veth1 name veth1.10 type vlan id 10 +ip -net "$nsr1" link set veth1.10 up +ip -net "$nsr1" addr add 192.168.20.1/24 dev veth1.10 +ip netns exec "$nsr1" sysctl net.ipv4.conf.veth1/10.forwarding=1 > /dev/null +ip netns exec "$nsr1" nft -a insert rule inet filter forward 'meta oif veth1.10 accept' +ip -net "$nsr1" link add name tun1 type ipip local 192.168.20.1 remote 192.168.20.2 +ip -net "$nsr1" link set tun1 up +ip -net "$nsr1" addr add 192.168.200.1/24 dev tun1 +ip -net "$nsr1" route change default via 192.168.200.2 +ip netns exec "$nsr1" sysctl net.ipv4.conf.tun1.forwarding=1 > /dev/null +ip netns exec "$nsr1" nft -a insert rule inet filter forward 'meta oif tun1 accept' + +ip -net "$nsr2" link add link veth0 name veth0.10 type vlan id 10 +ip -net "$nsr2" link set veth0.10 up +ip -net "$nsr2" addr add 192.168.20.2/24 dev veth0.10 +ip netns exec "$nsr2" sysctl net.ipv4.conf.veth0/10.forwarding=1 > /dev/null +ip -net "$nsr2" link add name tun1 type ipip local 192.168.20.2 remote 192.168.20.1 +ip -net "$nsr2" link set tun1 up +ip -net "$nsr2" addr add 192.168.200.2/24 dev tun1 +ip -net "$nsr2" route change default via 192.168.200.1 +ip netns exec "$nsr2" sysctl net.ipv4.conf.tun1.forwarding=1 > /dev/null + +if ! test_tcp_forwarding_nat "$ns1" "$ns2" 1 "IPIP tunnel over vlan"; then + echo "FAIL: flow offload for ns1/ns2 with IPIP tunnel over vlan" 1>&2 + ip netns exec "$nsr1" nft list ruleset + ret=1 +fi + +# Restore the previous configuration +ip -net "$nsr1" route change default via 192.168.10.2 +ip -net "$nsr2" route change default via 192.168.10.1 +ip -net "$ns2" route del default via 10.0.2.1 +} + # Another test: # Add bridge interface br0 to Router1, with NAT enabled. test_bridge() { @@ -643,6 +710,8 @@ ip -net "$nsr1" addr add dead:1::1/64 dev veth0 nodad ip -net "$nsr1" link set up dev veth0 } +test_ipip + test_bridge KEY_SHA="0x"$(ps -af | sha1sum | cut -d " " -f 1) From be102eb6a0e7c03db00e50540622f4e43b2d2844 Mon Sep 17 00:00:00 2001 From: Fernando Fernandez Mancera Date: Fri, 21 Nov 2025 01:14:30 +0100 Subject: [PATCH 1221/1381] netfilter: nf_conncount: rework API to use sk_buff directly When using nf_conncount infrastructure for non-confirmed connections a duplicated track is possible due to an optimization introduced since commit d265929930e2 ("netfilter: nf_conncount: reduce unnecessary GC"). In order to fix this introduce a new conncount API that receives directly an sk_buff struct. It fetches the tuple and zone and the corresponding ct from it. It comes with both existing conncount variants nf_conncount_count_skb() and nf_conncount_add_skb(). In addition remove the old API and adjust all the users to use the new one. This way, for each sk_buff struct it is possible to check if there is a ct present and already confirmed. If so, skip the add operation. Fixes: d265929930e2 ("netfilter: nf_conncount: reduce unnecessary GC") Signed-off-by: Fernando Fernandez Mancera Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_conntrack_count.h | 15 +- net/netfilter/nf_conncount.c | 185 ++++++++++++++------- net/netfilter/nft_connlimit.c | 21 +-- net/netfilter/xt_connlimit.c | 14 +- net/openvswitch/conntrack.c | 16 +- 5 files changed, 145 insertions(+), 106 deletions(-) diff --git a/include/net/netfilter/nf_conntrack_count.h b/include/net/netfilter/nf_conntrack_count.h index 1b58b5b91ff6..52a06de41aa0 100644 --- a/include/net/netfilter/nf_conntrack_count.h +++ b/include/net/netfilter/nf_conntrack_count.h @@ -18,15 +18,14 @@ struct nf_conncount_list { struct nf_conncount_data *nf_conncount_init(struct net *net, unsigned int keylen); void nf_conncount_destroy(struct net *net, struct nf_conncount_data *data); -unsigned int nf_conncount_count(struct net *net, - struct nf_conncount_data *data, - const u32 *key, - const struct nf_conntrack_tuple *tuple, - const struct nf_conntrack_zone *zone); +unsigned int nf_conncount_count_skb(struct net *net, + const struct sk_buff *skb, + u16 l3num, + struct nf_conncount_data *data, + const u32 *key); -int nf_conncount_add(struct net *net, struct nf_conncount_list *list, - const struct nf_conntrack_tuple *tuple, - const struct nf_conntrack_zone *zone); +int nf_conncount_add_skb(struct net *net, const struct sk_buff *skb, + u16 l3num, struct nf_conncount_list *list); void nf_conncount_list_init(struct nf_conncount_list *list); diff --git a/net/netfilter/nf_conncount.c b/net/netfilter/nf_conncount.c index 913ede2f57f9..0ffc5ff78a71 100644 --- a/net/netfilter/nf_conncount.c +++ b/net/netfilter/nf_conncount.c @@ -122,15 +122,65 @@ find_or_evict(struct net *net, struct nf_conncount_list *list, return ERR_PTR(-EAGAIN); } -static int __nf_conncount_add(struct net *net, - struct nf_conncount_list *list, - const struct nf_conntrack_tuple *tuple, - const struct nf_conntrack_zone *zone) +static bool get_ct_or_tuple_from_skb(struct net *net, + const struct sk_buff *skb, + u16 l3num, + struct nf_conn **ct, + struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_zone **zone, + bool *refcounted) { + const struct nf_conntrack_tuple_hash *h; + enum ip_conntrack_info ctinfo; + struct nf_conn *found_ct; + + found_ct = nf_ct_get(skb, &ctinfo); + if (found_ct && !nf_ct_is_template(found_ct)) { + *tuple = found_ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple; + *zone = nf_ct_zone(found_ct); + *ct = found_ct; + return true; + } + + if (!nf_ct_get_tuplepr(skb, skb_network_offset(skb), l3num, net, tuple)) + return false; + + if (found_ct) + *zone = nf_ct_zone(found_ct); + + h = nf_conntrack_find_get(net, *zone, tuple); + if (!h) + return true; + + found_ct = nf_ct_tuplehash_to_ctrack(h); + *refcounted = true; + *ct = found_ct; + + return true; +} + +static int __nf_conncount_add(struct net *net, + const struct sk_buff *skb, + u16 l3num, + struct nf_conncount_list *list) +{ + const struct nf_conntrack_zone *zone = &nf_ct_zone_dflt; const struct nf_conntrack_tuple_hash *found; struct nf_conncount_tuple *conn, *conn_n; + struct nf_conntrack_tuple tuple; + struct nf_conn *ct = NULL; struct nf_conn *found_ct; unsigned int collect = 0; + bool refcounted = false; + + if (!get_ct_or_tuple_from_skb(net, skb, l3num, &ct, &tuple, &zone, &refcounted)) + return -ENOENT; + + if (ct && nf_ct_is_confirmed(ct)) { + if (refcounted) + nf_ct_put(ct); + return 0; + } if ((u32)jiffies == list->last_gc) goto add_new_node; @@ -144,10 +194,10 @@ static int __nf_conncount_add(struct net *net, if (IS_ERR(found)) { /* Not found, but might be about to be confirmed */ if (PTR_ERR(found) == -EAGAIN) { - if (nf_ct_tuple_equal(&conn->tuple, tuple) && + if (nf_ct_tuple_equal(&conn->tuple, &tuple) && nf_ct_zone_id(&conn->zone, conn->zone.dir) == nf_ct_zone_id(zone, zone->dir)) - return 0; /* already exists */ + goto out_put; /* already exists */ } else { collect++; } @@ -156,7 +206,7 @@ static int __nf_conncount_add(struct net *net, found_ct = nf_ct_tuplehash_to_ctrack(found); - if (nf_ct_tuple_equal(&conn->tuple, tuple) && + if (nf_ct_tuple_equal(&conn->tuple, &tuple) && nf_ct_zone_equal(found_ct, zone, zone->dir)) { /* * We should not see tuples twice unless someone hooks @@ -165,7 +215,7 @@ static int __nf_conncount_add(struct net *net, * Attempt to avoid a re-add in this case. */ nf_ct_put(found_ct); - return 0; + goto out_put; } else if (already_closed(found_ct)) { /* * we do not care about connections which are @@ -188,31 +238,35 @@ static int __nf_conncount_add(struct net *net, if (conn == NULL) return -ENOMEM; - conn->tuple = *tuple; + conn->tuple = tuple; conn->zone = *zone; conn->cpu = raw_smp_processor_id(); conn->jiffies32 = (u32)jiffies; list_add_tail(&conn->node, &list->head); list->count++; list->last_gc = (u32)jiffies; + +out_put: + if (refcounted) + nf_ct_put(ct); return 0; } -int nf_conncount_add(struct net *net, - struct nf_conncount_list *list, - const struct nf_conntrack_tuple *tuple, - const struct nf_conntrack_zone *zone) +int nf_conncount_add_skb(struct net *net, + const struct sk_buff *skb, + u16 l3num, + struct nf_conncount_list *list) { int ret; /* check the saved connections */ spin_lock_bh(&list->list_lock); - ret = __nf_conncount_add(net, list, tuple, zone); + ret = __nf_conncount_add(net, skb, l3num, list); spin_unlock_bh(&list->list_lock); return ret; } -EXPORT_SYMBOL_GPL(nf_conncount_add); +EXPORT_SYMBOL_GPL(nf_conncount_add_skb); void nf_conncount_list_init(struct nf_conncount_list *list) { @@ -309,19 +363,22 @@ static void schedule_gc_worker(struct nf_conncount_data *data, int tree) static unsigned int insert_tree(struct net *net, + const struct sk_buff *skb, + u16 l3num, struct nf_conncount_data *data, struct rb_root *root, unsigned int hash, - const u32 *key, - const struct nf_conntrack_tuple *tuple, - const struct nf_conntrack_zone *zone) + const u32 *key) { struct nf_conncount_rb *gc_nodes[CONNCOUNT_GC_MAX_NODES]; - struct rb_node **rbnode, *parent; - struct nf_conncount_rb *rbconn; - struct nf_conncount_tuple *conn; + const struct nf_conntrack_zone *zone = &nf_ct_zone_dflt; + bool do_gc = true, refcounted = false; unsigned int count = 0, gc_count = 0; - bool do_gc = true; + struct rb_node **rbnode, *parent; + struct nf_conntrack_tuple tuple; + struct nf_conncount_tuple *conn; + struct nf_conncount_rb *rbconn; + struct nf_conn *ct = NULL; spin_lock_bh(&nf_conncount_locks[hash]); restart: @@ -340,7 +397,7 @@ insert_tree(struct net *net, } else { int ret; - ret = nf_conncount_add(net, &rbconn->list, tuple, zone); + ret = nf_conncount_add_skb(net, skb, l3num, &rbconn->list); if (ret) count = 0; /* hotdrop */ else @@ -364,30 +421,35 @@ insert_tree(struct net *net, goto restart; } - /* expected case: match, insert new node */ - rbconn = kmem_cache_alloc(conncount_rb_cachep, GFP_ATOMIC); - if (rbconn == NULL) - goto out_unlock; + if (get_ct_or_tuple_from_skb(net, skb, l3num, &ct, &tuple, &zone, &refcounted)) { + /* expected case: match, insert new node */ + rbconn = kmem_cache_alloc(conncount_rb_cachep, GFP_ATOMIC); + if (rbconn == NULL) + goto out_unlock; - conn = kmem_cache_alloc(conncount_conn_cachep, GFP_ATOMIC); - if (conn == NULL) { - kmem_cache_free(conncount_rb_cachep, rbconn); - goto out_unlock; + conn = kmem_cache_alloc(conncount_conn_cachep, GFP_ATOMIC); + if (conn == NULL) { + kmem_cache_free(conncount_rb_cachep, rbconn); + goto out_unlock; + } + + conn->tuple = tuple; + conn->zone = *zone; + conn->cpu = raw_smp_processor_id(); + conn->jiffies32 = (u32)jiffies; + memcpy(rbconn->key, key, sizeof(u32) * data->keylen); + + nf_conncount_list_init(&rbconn->list); + list_add(&conn->node, &rbconn->list.head); + count = 1; + rbconn->list.count = count; + + rb_link_node_rcu(&rbconn->node, parent, rbnode); + rb_insert_color(&rbconn->node, root); + + if (refcounted) + nf_ct_put(ct); } - - conn->tuple = *tuple; - conn->zone = *zone; - conn->cpu = raw_smp_processor_id(); - conn->jiffies32 = (u32)jiffies; - memcpy(rbconn->key, key, sizeof(u32) * data->keylen); - - nf_conncount_list_init(&rbconn->list); - list_add(&conn->node, &rbconn->list.head); - count = 1; - rbconn->list.count = count; - - rb_link_node_rcu(&rbconn->node, parent, rbnode); - rb_insert_color(&rbconn->node, root); out_unlock: spin_unlock_bh(&nf_conncount_locks[hash]); return count; @@ -395,10 +457,10 @@ insert_tree(struct net *net, static unsigned int count_tree(struct net *net, + const struct sk_buff *skb, + u16 l3num, struct nf_conncount_data *data, - const u32 *key, - const struct nf_conntrack_tuple *tuple, - const struct nf_conntrack_zone *zone) + const u32 *key) { struct rb_root *root; struct rb_node *parent; @@ -422,7 +484,7 @@ count_tree(struct net *net, } else { int ret; - if (!tuple) { + if (!skb) { nf_conncount_gc_list(net, &rbconn->list); return rbconn->list.count; } @@ -437,7 +499,7 @@ count_tree(struct net *net, } /* same source network -> be counted! */ - ret = __nf_conncount_add(net, &rbconn->list, tuple, zone); + ret = __nf_conncount_add(net, skb, l3num, &rbconn->list); spin_unlock_bh(&rbconn->list.list_lock); if (ret) return 0; /* hotdrop */ @@ -446,10 +508,10 @@ count_tree(struct net *net, } } - if (!tuple) + if (!skb) return 0; - return insert_tree(net, data, root, hash, key, tuple, zone); + return insert_tree(net, skb, l3num, data, root, hash, key); } static void tree_gc_worker(struct work_struct *work) @@ -511,18 +573,19 @@ static void tree_gc_worker(struct work_struct *work) } /* Count and return number of conntrack entries in 'net' with particular 'key'. - * If 'tuple' is not null, insert it into the accounting data structure. - * Call with RCU read lock. + * If 'skb' is not null, insert the corresponding tuple into the accounting + * data structure. Call with RCU read lock. */ -unsigned int nf_conncount_count(struct net *net, - struct nf_conncount_data *data, - const u32 *key, - const struct nf_conntrack_tuple *tuple, - const struct nf_conntrack_zone *zone) +unsigned int nf_conncount_count_skb(struct net *net, + const struct sk_buff *skb, + u16 l3num, + struct nf_conncount_data *data, + const u32 *key) { - return count_tree(net, data, key, tuple, zone); + return count_tree(net, skb, l3num, data, key); + } -EXPORT_SYMBOL_GPL(nf_conncount_count); +EXPORT_SYMBOL_GPL(nf_conncount_count_skb); struct nf_conncount_data *nf_conncount_init(struct net *net, unsigned int keylen) { diff --git a/net/netfilter/nft_connlimit.c b/net/netfilter/nft_connlimit.c index fc35a11cdca2..5df7134131d2 100644 --- a/net/netfilter/nft_connlimit.c +++ b/net/netfilter/nft_connlimit.c @@ -24,26 +24,11 @@ static inline void nft_connlimit_do_eval(struct nft_connlimit *priv, const struct nft_pktinfo *pkt, const struct nft_set_ext *ext) { - const struct nf_conntrack_zone *zone = &nf_ct_zone_dflt; - const struct nf_conntrack_tuple *tuple_ptr; - struct nf_conntrack_tuple tuple; - enum ip_conntrack_info ctinfo; - const struct nf_conn *ct; unsigned int count; + int err; - tuple_ptr = &tuple; - - ct = nf_ct_get(pkt->skb, &ctinfo); - if (ct != NULL) { - tuple_ptr = &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple; - zone = nf_ct_zone(ct); - } else if (!nf_ct_get_tuplepr(pkt->skb, skb_network_offset(pkt->skb), - nft_pf(pkt), nft_net(pkt), &tuple)) { - regs->verdict.code = NF_DROP; - return; - } - - if (nf_conncount_add(nft_net(pkt), priv->list, tuple_ptr, zone)) { + err = nf_conncount_add_skb(nft_net(pkt), pkt->skb, nft_pf(pkt), priv->list); + if (err) { regs->verdict.code = NF_DROP; return; } diff --git a/net/netfilter/xt_connlimit.c b/net/netfilter/xt_connlimit.c index 0189f8b6b0bd..848287ab79cf 100644 --- a/net/netfilter/xt_connlimit.c +++ b/net/netfilter/xt_connlimit.c @@ -31,8 +31,6 @@ connlimit_mt(const struct sk_buff *skb, struct xt_action_param *par) { struct net *net = xt_net(par); const struct xt_connlimit_info *info = par->matchinfo; - struct nf_conntrack_tuple tuple; - const struct nf_conntrack_tuple *tuple_ptr = &tuple; const struct nf_conntrack_zone *zone = &nf_ct_zone_dflt; enum ip_conntrack_info ctinfo; const struct nf_conn *ct; @@ -40,13 +38,8 @@ connlimit_mt(const struct sk_buff *skb, struct xt_action_param *par) u32 key[5]; ct = nf_ct_get(skb, &ctinfo); - if (ct != NULL) { - tuple_ptr = &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple; + if (ct) zone = nf_ct_zone(ct); - } else if (!nf_ct_get_tuplepr(skb, skb_network_offset(skb), - xt_family(par), net, &tuple)) { - goto hotdrop; - } if (xt_family(par) == NFPROTO_IPV6) { const struct ipv6hdr *iph = ipv6_hdr(skb); @@ -69,10 +62,9 @@ connlimit_mt(const struct sk_buff *skb, struct xt_action_param *par) key[1] = zone->id; } - connections = nf_conncount_count(net, info->data, key, tuple_ptr, - zone); + connections = nf_conncount_count_skb(net, skb, xt_family(par), info->data, key); if (connections == 0) - /* kmalloc failed, drop it entirely */ + /* kmalloc failed or tuple couldn't be found, drop it entirely */ goto hotdrop; return (connections > info->limit) ^ !!(info->flags & XT_CONNLIMIT_INVERT); diff --git a/net/openvswitch/conntrack.c b/net/openvswitch/conntrack.c index e573e9221302..a0811e1fba65 100644 --- a/net/openvswitch/conntrack.c +++ b/net/openvswitch/conntrack.c @@ -928,8 +928,8 @@ static u32 ct_limit_get(const struct ovs_ct_limit_info *info, u16 zone) } static int ovs_ct_check_limit(struct net *net, - const struct ovs_conntrack_info *info, - const struct nf_conntrack_tuple *tuple) + const struct sk_buff *skb, + const struct ovs_conntrack_info *info) { struct ovs_net *ovs_net = net_generic(net, ovs_net_id); const struct ovs_ct_limit_info *ct_limit_info = ovs_net->ct_limit_info; @@ -942,8 +942,9 @@ static int ovs_ct_check_limit(struct net *net, if (per_zone_limit == OVS_CT_LIMIT_UNLIMITED) return 0; - connections = nf_conncount_count(net, ct_limit_info->data, - &conncount_key, tuple, &info->zone); + connections = nf_conncount_count_skb(net, skb, info->family, + ct_limit_info->data, + &conncount_key); if (connections > per_zone_limit) return -ENOMEM; @@ -972,8 +973,7 @@ static int ovs_ct_commit(struct net *net, struct sw_flow_key *key, #if IS_ENABLED(CONFIG_NETFILTER_CONNCOUNT) if (static_branch_unlikely(&ovs_ct_limit_enabled)) { if (!nf_ct_is_confirmed(ct)) { - err = ovs_ct_check_limit(net, info, - &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple); + err = ovs_ct_check_limit(net, skb, info); if (err) { net_warn_ratelimited("openvswitch: zone: %u " "exceeds conntrack limit\n", @@ -1770,8 +1770,8 @@ static int __ovs_ct_limit_get_zone_limit(struct net *net, zone_limit.limit = limit; nf_ct_zone_init(&ct_zone, zone_id, NF_CT_DEFAULT_ZONE_DIR, 0); - zone_limit.count = nf_conncount_count(net, data, &conncount_key, NULL, - &ct_zone); + zone_limit.count = nf_conncount_count_skb(net, NULL, 0, data, + &conncount_key); return nla_put_nohdr(reply, sizeof(zone_limit), &zone_limit); } From c0362b5748282e22fa1592a8d3474f726ad964c2 Mon Sep 17 00:00:00 2001 From: Fernando Fernandez Mancera Date: Fri, 21 Nov 2025 01:14:31 +0100 Subject: [PATCH 1222/1381] netfilter: nf_conncount: make nf_conncount_gc_list() to disable BH For convenience when performing GC over the connection list, make nf_conncount_gc_list() to disable BH. This unifies the behavior with nf_conncount_add() and nf_conncount_count(). Signed-off-by: Fernando Fernandez Mancera Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_conncount.c | 24 +++++++++++++++++------- net/netfilter/nft_connlimit.c | 7 +------ 2 files changed, 18 insertions(+), 13 deletions(-) diff --git a/net/netfilter/nf_conncount.c b/net/netfilter/nf_conncount.c index 0ffc5ff78a71..d8893e172444 100644 --- a/net/netfilter/nf_conncount.c +++ b/net/netfilter/nf_conncount.c @@ -278,8 +278,8 @@ void nf_conncount_list_init(struct nf_conncount_list *list) EXPORT_SYMBOL_GPL(nf_conncount_list_init); /* Return true if the list is empty. Must be called with BH disabled. */ -bool nf_conncount_gc_list(struct net *net, - struct nf_conncount_list *list) +static bool __nf_conncount_gc_list(struct net *net, + struct nf_conncount_list *list) { const struct nf_conntrack_tuple_hash *found; struct nf_conncount_tuple *conn, *conn_n; @@ -291,10 +291,6 @@ bool nf_conncount_gc_list(struct net *net, if ((u32)jiffies == READ_ONCE(list->last_gc)) return false; - /* don't bother if other cpu is already doing GC */ - if (!spin_trylock(&list->list_lock)) - return false; - list_for_each_entry_safe(conn, conn_n, &list->head, node) { found = find_or_evict(net, list, conn); if (IS_ERR(found)) { @@ -323,7 +319,21 @@ bool nf_conncount_gc_list(struct net *net, if (!list->count) ret = true; list->last_gc = (u32)jiffies; - spin_unlock(&list->list_lock); + + return ret; +} + +bool nf_conncount_gc_list(struct net *net, + struct nf_conncount_list *list) +{ + bool ret; + + /* don't bother if other cpu is already doing GC */ + if (!spin_trylock_bh(&list->list_lock)) + return false; + + ret = __nf_conncount_gc_list(net, list); + spin_unlock_bh(&list->list_lock); return ret; } diff --git a/net/netfilter/nft_connlimit.c b/net/netfilter/nft_connlimit.c index 5df7134131d2..41770bde39d3 100644 --- a/net/netfilter/nft_connlimit.c +++ b/net/netfilter/nft_connlimit.c @@ -223,13 +223,8 @@ static void nft_connlimit_destroy_clone(const struct nft_ctx *ctx, static bool nft_connlimit_gc(struct net *net, const struct nft_expr *expr) { struct nft_connlimit *priv = nft_expr_priv(expr); - bool ret; - local_bh_disable(); - ret = nf_conncount_gc_list(net, priv->list); - local_bh_enable(); - - return ret; + return nf_conncount_gc_list(net, priv->list); } static struct nft_expr_type nft_connlimit_type; From 69894e5b4c5e28cda5f32af33d4a92b7a4b93b0e Mon Sep 17 00:00:00 2001 From: Fernando Fernandez Mancera Date: Fri, 21 Nov 2025 01:14:32 +0100 Subject: [PATCH 1223/1381] netfilter: nft_connlimit: update the count if add was skipped Connlimit expression can be used for all kind of packets and not only for packets with connection state new. See this ruleset as example: table ip filter { chain input { type filter hook input priority filter; policy accept; tcp dport 22 ct count over 4 counter } } Currently, if the connection count goes over the limit the counter will count the packets. When a connection is closed, the connection count won't decrement as it should because it is only updated for new connections due to an optimization on __nf_conncount_add() that prevents updating the list if the connection is duplicated. To solve this problem, check whether the connection was skipped and if so, update the list. Adjust count_tree() too so the same fix is applied for xt_connlimit. Fixes: 976afca1ceba ("netfilter: nf_conncount: Early exit in nf_conncount_lookup() and cleanup") Closes: https://lore.kernel.org/netfilter/trinity-85c72a88-d762-46c3-be97-36f10e5d9796-1761173693813@3c-app-mailcom-bs12/ Signed-off-by: Fernando Fernandez Mancera Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_conncount.c | 12 ++++++++---- net/netfilter/nft_connlimit.c | 13 +++++++++++-- 2 files changed, 19 insertions(+), 6 deletions(-) diff --git a/net/netfilter/nf_conncount.c b/net/netfilter/nf_conncount.c index d8893e172444..f1be4dd5cf85 100644 --- a/net/netfilter/nf_conncount.c +++ b/net/netfilter/nf_conncount.c @@ -179,7 +179,7 @@ static int __nf_conncount_add(struct net *net, if (ct && nf_ct_is_confirmed(ct)) { if (refcounted) nf_ct_put(ct); - return 0; + return -EEXIST; } if ((u32)jiffies == list->last_gc) @@ -408,7 +408,7 @@ insert_tree(struct net *net, int ret; ret = nf_conncount_add_skb(net, skb, l3num, &rbconn->list); - if (ret) + if (ret && ret != -EEXIST) count = 0; /* hotdrop */ else count = rbconn->list.count; @@ -511,10 +511,14 @@ count_tree(struct net *net, /* same source network -> be counted! */ ret = __nf_conncount_add(net, skb, l3num, &rbconn->list); spin_unlock_bh(&rbconn->list.list_lock); - if (ret) + if (ret && ret != -EEXIST) { return 0; /* hotdrop */ - else + } else { + /* -EEXIST means add was skipped, update the list */ + if (ret == -EEXIST) + nf_conncount_gc_list(net, &rbconn->list); return rbconn->list.count; + } } } diff --git a/net/netfilter/nft_connlimit.c b/net/netfilter/nft_connlimit.c index 41770bde39d3..714a59485935 100644 --- a/net/netfilter/nft_connlimit.c +++ b/net/netfilter/nft_connlimit.c @@ -29,8 +29,17 @@ static inline void nft_connlimit_do_eval(struct nft_connlimit *priv, err = nf_conncount_add_skb(nft_net(pkt), pkt->skb, nft_pf(pkt), priv->list); if (err) { - regs->verdict.code = NF_DROP; - return; + if (err == -EEXIST) { + /* Call gc to update the list count if any connection has + * been closed already. This is useful for softlimit + * connections like limiting bandwidth based on a number + * of open connections. + */ + nf_conncount_gc_list(nft_net(pkt), priv->list); + } else { + regs->verdict.code = NF_DROP; + return; + } } count = READ_ONCE(priv->list->count); From c4cbe4a4df39a2cf80593f87d129cd4b04ea568d Mon Sep 17 00:00:00 2001 From: Fernando Fernandez Mancera Date: Mon, 24 Nov 2025 17:36:58 +0100 Subject: [PATCH 1224/1381] netfilter: nft_connlimit: add support to object update operation This is useful to update the limit or flags without clearing the connections tracked. Use READ_ONCE() on packetpath as it can be modified on controlplane. Signed-off-by: Fernando Fernandez Mancera Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nft_connlimit.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-) diff --git a/net/netfilter/nft_connlimit.c b/net/netfilter/nft_connlimit.c index 714a59485935..657764774a2d 100644 --- a/net/netfilter/nft_connlimit.c +++ b/net/netfilter/nft_connlimit.c @@ -44,7 +44,7 @@ static inline void nft_connlimit_do_eval(struct nft_connlimit *priv, count = READ_ONCE(priv->list->count); - if ((count > priv->limit) ^ priv->invert) { + if ((count > READ_ONCE(priv->limit)) ^ READ_ONCE(priv->invert)) { regs->verdict.code = NFT_BREAK; return; } @@ -131,6 +131,16 @@ static int nft_connlimit_obj_init(const struct nft_ctx *ctx, return nft_connlimit_do_init(ctx, tb, priv); } +static void nft_connlimit_obj_update(struct nft_object *obj, + struct nft_object *newobj) +{ + struct nft_connlimit *newpriv = nft_obj_data(newobj); + struct nft_connlimit *priv = nft_obj_data(obj); + + WRITE_ONCE(priv->limit, newpriv->limit); + WRITE_ONCE(priv->invert, newpriv->invert); +} + static void nft_connlimit_obj_destroy(const struct nft_ctx *ctx, struct nft_object *obj) { @@ -160,6 +170,7 @@ static const struct nft_object_ops nft_connlimit_obj_ops = { .init = nft_connlimit_obj_init, .destroy = nft_connlimit_obj_destroy, .dump = nft_connlimit_obj_dump, + .update = nft_connlimit_obj_update, }; static struct nft_object_type nft_connlimit_obj_type __read_mostly = { From c0bd21682aedeb5d1367ee8026f71dc32624091a Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Thu, 27 Nov 2025 23:21:43 +0100 Subject: [PATCH 1225/1381] selftests: netfilter: nft_flowtable.sh: Add the capability to send IPv6 TCP traffic Introduce the capability to send TCP traffic over IPv6 to nft_flowtable netfilter selftest. Reviewed-by: Simon Horman Signed-off-by: Lorenzo Bianconi Signed-off-by: Pablo Neira Ayuso --- .../selftests/net/netfilter/nft_flowtable.sh | 57 ++++++++++++++----- 1 file changed, 43 insertions(+), 14 deletions(-) diff --git a/tools/testing/selftests/net/netfilter/nft_flowtable.sh b/tools/testing/selftests/net/netfilter/nft_flowtable.sh index 1fbfc8ad8dcd..a68bc882fa4e 100755 --- a/tools/testing/selftests/net/netfilter/nft_flowtable.sh +++ b/tools/testing/selftests/net/netfilter/nft_flowtable.sh @@ -127,6 +127,8 @@ ip -net "$nsr1" addr add fee1:2::1/64 dev veth1 nodad ip -net "$nsr2" addr add 192.168.10.2/24 dev veth0 ip -net "$nsr2" addr add fee1:2::2/64 dev veth0 nodad +ip netns exec "$nsr1" sysctl net.ipv6.conf.all.forwarding=1 > /dev/null +ip netns exec "$nsr2" sysctl net.ipv6.conf.all.forwarding=1 > /dev/null for i in 0 1; do ip netns exec "$nsr1" sysctl net.ipv4.conf.veth$i.forwarding=1 > /dev/null ip netns exec "$nsr2" sysctl net.ipv4.conf.veth$i.forwarding=1 > /dev/null @@ -153,7 +155,9 @@ ip -net "$ns1" route add default via dead:1::1 ip -net "$ns2" route add default via dead:2::1 ip -net "$nsr1" route add default via 192.168.10.2 +ip -6 -net "$nsr1" route add default via fee1:2::2 ip -net "$nsr2" route add default via 192.168.10.1 +ip -6 -net "$nsr2" route add default via fee1:2::1 ip netns exec "$nsr1" nft -f - < "$ns2out" & + timeout "$SOCAT_TIMEOUT" ip netns exec "$nsb" socat -${proto} \ + TCP"${proto}"-LISTEN:12345,reuseaddr STDIO < "$infile" > "$ns2out" & lpid=$! busywait 1000 listener_ready - timeout "$SOCAT_TIMEOUT" ip netns exec "$nsa" socat -4 TCP:"$dstip":"$dstport" STDIO < "$infile" > "$ns1out" + timeout "$SOCAT_TIMEOUT" ip netns exec "$nsa" socat -${proto} \ + TCP"${proto}":"$dstip":"$dstport" STDIO < "$infile" > "$ns1out" socatc=$? wait $lpid @@ -394,8 +401,11 @@ test_tcp_forwarding_ip() test_tcp_forwarding() { local pmtu="$3" + local proto="$4" + local dstip="$5" + local dstport="$6" - test_tcp_forwarding_ip "$1" "$2" "$pmtu" 10.0.2.99 12345 + test_tcp_forwarding_ip "$1" "$2" "$pmtu" "$proto" "$dstip" "$dstport" return $? } @@ -403,6 +413,9 @@ test_tcp_forwarding() test_tcp_forwarding_set_dscp() { local pmtu="$3" + local proto="$4" + local dstip="$5" + local dstport="$6" ip netns exec "$nsr1" nft -f - <&2 @@ -495,6 +508,14 @@ else ret=1 fi +if test_tcp_forwarding "$ns1" "$ns2" 0 6 "[dead:2::99]" 12345; then + echo "PASS: IPv6 flow offloaded for ns1/ns2" +else + echo "FAIL: IPv6 flow offload for ns1/ns2:" 1>&2 + ip netns exec "$nsr1" nft list ruleset + ret=1 +fi + # delete default route, i.e. ns2 won't be able to reach ns1 and # will depend on ns1 being masqueraded in nsr1. # expect ns1 has nsr1 address. @@ -520,7 +541,7 @@ table ip nat { EOF check_dscp "dscp_none" "0" -if ! test_tcp_forwarding_set_dscp "$ns1" "$ns2" 0 ""; then +if ! test_tcp_forwarding_set_dscp "$ns1" "$ns2" 0 4 10.0.2.99 12345; then echo "FAIL: flow offload for ns1/ns2 with dscp update and no pmtu discovery" 1>&2 exit 0 fi @@ -546,7 +567,7 @@ ip netns exec "$ns2" sysctl net.ipv4.ip_no_pmtu_disc=0 > /dev/null ip netns exec "$nsr1" nft reset counters table inet filter >/dev/null ip netns exec "$ns2" nft reset counters table inet filter >/dev/null -if ! test_tcp_forwarding_set_dscp "$ns1" "$ns2" 1 ""; then +if ! test_tcp_forwarding_set_dscp "$ns1" "$ns2" 1 4 10.0.2.99 12345; then echo "FAIL: flow offload for ns1/ns2 with dscp update and pmtu discovery" 1>&2 exit 0 fi @@ -752,7 +773,7 @@ ip -net "$ns2" route del 192.168.10.1 via 10.0.2.1 ip -net "$ns2" route add default via 10.0.2.1 ip -net "$ns2" route add default via dead:2::1 -if test_tcp_forwarding "$ns1" "$ns2" 1; then +if test_tcp_forwarding "$ns1" "$ns2" 1 4 10.0.2.99 12345; then check_counters "ipsec tunnel mode for ns1/ns2" else echo "FAIL: ipsec tunnel mode for ns1/ns2" @@ -760,6 +781,14 @@ else ip netns exec "$nsr1" cat /proc/net/xfrm_stat 1>&2 fi +if test_tcp_forwarding "$ns1" "$ns2" 1 6 "[dead:2::99]" 12345; then + check_counters "IPv6 ipsec tunnel mode for ns1/ns2" +else + echo "FAIL: IPv6 ipsec tunnel mode for ns1/ns2" + ip netns exec "$nsr1" nft list ruleset 1>&2 + ip netns exec "$nsr1" cat /proc/net/xfrm_stat 1>&2 +fi + if [ "$1" = "" ]; then low=1280 mtu=$((65536 - low)) From c4f0ab06e1e0c1331e6febd03538a7f621f15134 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Sat, 1 Nov 2025 12:20:50 -0700 Subject: [PATCH 1226/1381] netfilter: ip6t_srh: fix UAPI kernel-doc comments format Fix the kernel-doc format for struct members to be "@member" instead of "@ member" to avoid kernel-doc warnings. Warning: ip6t_srh.h:60 struct member 'next_hdr' not described in 'ip6t_srh' Warning: ip6t_srh.h:60 struct member 'hdr_len' not described in 'ip6t_srh' Warning: ip6t_srh.h:60 struct member 'segs_left' not described in 'ip6t_srh' Warning: ip6t_srh.h:60 struct member 'last_entry' not described in 'ip6t_srh' Warning: ip6t_srh.h:60 struct member 'tag' not described in 'ip6t_srh' Warning: ip6t_srh.h:60 struct member 'mt_flags' not described in 'ip6t_srh' Warning: ip6t_srh.h:60 struct member 'mt_invflags' not described in 'ip6t_srh' Warning: ip6t_srh.h:93 struct member 'next_hdr' not described in 'ip6t_srh1' Warning: ip6t_srh.h:93 struct member 'hdr_len' not described in 'ip6t_srh1' Warning: ip6t_srh.h:93 struct member 'segs_left' not described in 'ip6t_srh1' Warning: ip6t_srh.h:93 struct member 'last_entry' not described in 'ip6t_srh1' Warning: ip6t_srh.h:93 struct member 'tag' not described in 'ip6t_srh1' Warning: ip6t_srh.h:93 struct member 'psid_addr' not described in 'ip6t_srh1' Warning: ip6t_srh.h:93 struct member 'nsid_addr' not described in 'ip6t_srh1' Warning: ip6t_srh.h:93 struct member 'lsid_addr' not described in 'ip6t_srh1' Warning: ip6t_srh.h:93 struct member 'psid_msk' not described in 'ip6t_srh1' Warning: ip6t_srh.h:93 struct member 'nsid_msk' not described in 'ip6t_srh1' Warning: ip6t_srh.h:93 struct member 'lsid_msk' not described in 'ip6t_srh1' Warning: ip6t_srh.h:93 struct member 'mt_flags' not described in 'ip6t_srh1' Warning: ip6t_srh.h:93 struct member 'mt_invflags' not described in 'ip6t_srh1' Signed-off-by: Randy Dunlap Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/uapi/linux/netfilter_ipv6/ip6t_srh.h | 40 ++++++++++---------- 1 file changed, 20 insertions(+), 20 deletions(-) diff --git a/include/uapi/linux/netfilter_ipv6/ip6t_srh.h b/include/uapi/linux/netfilter_ipv6/ip6t_srh.h index 54ed83360dac..80c66c8ece82 100644 --- a/include/uapi/linux/netfilter_ipv6/ip6t_srh.h +++ b/include/uapi/linux/netfilter_ipv6/ip6t_srh.h @@ -41,13 +41,13 @@ /** * struct ip6t_srh - SRH match options - * @ next_hdr: Next header field of SRH - * @ hdr_len: Extension header length field of SRH - * @ segs_left: Segments left field of SRH - * @ last_entry: Last entry field of SRH - * @ tag: Tag field of SRH - * @ mt_flags: match options - * @ mt_invflags: Invert the sense of match options + * @next_hdr: Next header field of SRH + * @hdr_len: Extension header length field of SRH + * @segs_left: Segments left field of SRH + * @last_entry: Last entry field of SRH + * @tag: Tag field of SRH + * @mt_flags: match options + * @mt_invflags: Invert the sense of match options */ struct ip6t_srh { @@ -62,19 +62,19 @@ struct ip6t_srh { /** * struct ip6t_srh1 - SRH match options (revision 1) - * @ next_hdr: Next header field of SRH - * @ hdr_len: Extension header length field of SRH - * @ segs_left: Segments left field of SRH - * @ last_entry: Last entry field of SRH - * @ tag: Tag field of SRH - * @ psid_addr: Address of previous SID in SRH SID list - * @ nsid_addr: Address of NEXT SID in SRH SID list - * @ lsid_addr: Address of LAST SID in SRH SID list - * @ psid_msk: Mask of previous SID in SRH SID list - * @ nsid_msk: Mask of next SID in SRH SID list - * @ lsid_msk: MAsk of last SID in SRH SID list - * @ mt_flags: match options - * @ mt_invflags: Invert the sense of match options + * @next_hdr: Next header field of SRH + * @hdr_len: Extension header length field of SRH + * @segs_left: Segments left field of SRH + * @last_entry: Last entry field of SRH + * @tag: Tag field of SRH + * @psid_addr: Address of previous SID in SRH SID list + * @nsid_addr: Address of NEXT SID in SRH SID list + * @lsid_addr: Address of LAST SID in SRH SID list + * @psid_msk: Mask of previous SID in SRH SID list + * @nsid_msk: Mask of next SID in SRH SID list + * @lsid_msk: MAsk of last SID in SRH SID list + * @mt_flags: match options + * @mt_invflags: Invert the sense of match options */ struct ip6t_srh1 { From d3a439e55c193b930e0007967cf8d7a29890449b Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Sat, 1 Nov 2025 12:20:38 -0700 Subject: [PATCH 1227/1381] netfilter: nf_tables: improve UAPI kernel-doc comments In include/uapi/linux/netfilter/nf_tables.h, correct the kernel-doc comments for mistyped enum names and enum values to avoid these kernel-doc warnings and improve the documentation: nf_tables.h:896: warning: Enum value 'NFT_EXTHDR_OP_TCPOPT' not described in enum 'nft_exthdr_op' nf_tables.h:896: warning: Excess enum value 'NFT_EXTHDR_OP_TCP' description in 'nft_exthdr_op' nf_tables.h:1210: warning: expecting prototype for enum nft_flow_attributes. Prototype was for enum nft_offload_attributes instead nf_tables.h:1428: warning: expecting prototype for enum nft_reject_code. Prototype was for enum nft_reject_inet_code instead (add beginning '@' to each enum value description:) nf_tables.h:1493: warning: Enum value 'NFTA_TPROXY_FAMILY' not described in enum 'nft_tproxy_attributes' nf_tables.h:1493: warning: Enum value 'NFTA_TPROXY_REG_ADDR' not described in enum 'nft_tproxy_attributes' nf_tables.h:1493: warning: Enum value 'NFTA_TPROXY_REG_PORT' not described in enum 'nft_tproxy_attributes' nf_tables.h:1796: warning: expecting prototype for enum nft_device_attributes. Prototype was for enum nft_devices_attributes instead Signed-off-by: Randy Dunlap Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/uapi/linux/netfilter/nf_tables.h | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/include/uapi/linux/netfilter/nf_tables.h b/include/uapi/linux/netfilter/nf_tables.h index 7c0c915f0306..45c71f7d21c2 100644 --- a/include/uapi/linux/netfilter/nf_tables.h +++ b/include/uapi/linux/netfilter/nf_tables.h @@ -881,7 +881,7 @@ enum nft_exthdr_flags { * enum nft_exthdr_op - nf_tables match options * * @NFT_EXTHDR_OP_IPV6: match against ipv6 extension headers - * @NFT_EXTHDR_OP_TCP: match against tcp options + * @NFT_EXTHDR_OP_TCPOPT: match against tcp options * @NFT_EXTHDR_OP_IPV4: match against ipv4 options * @NFT_EXTHDR_OP_SCTP: match against sctp chunks * @NFT_EXTHDR_OP_DCCP: match against dccp otions @@ -1200,7 +1200,7 @@ enum nft_ct_attributes { #define NFTA_CT_MAX (__NFTA_CT_MAX - 1) /** - * enum nft_flow_attributes - ct offload expression attributes + * enum nft_offload_attributes - ct offload expression attributes * @NFTA_FLOW_TABLE_NAME: flow table name (NLA_STRING) */ enum nft_offload_attributes { @@ -1410,7 +1410,7 @@ enum nft_reject_types { }; /** - * enum nft_reject_code - Generic reject codes for IPv4/IPv6 + * enum nft_reject_inet_code - Generic reject codes for IPv4/IPv6 * * @NFT_REJECT_ICMPX_NO_ROUTE: no route to host / network unreachable * @NFT_REJECT_ICMPX_PORT_UNREACH: port unreachable @@ -1480,9 +1480,9 @@ enum nft_nat_attributes { /** * enum nft_tproxy_attributes - nf_tables tproxy expression netlink attributes * - * NFTA_TPROXY_FAMILY: Target address family (NLA_U32: nft_registers) - * NFTA_TPROXY_REG_ADDR: Target address register (NLA_U32: nft_registers) - * NFTA_TPROXY_REG_PORT: Target port register (NLA_U32: nft_registers) + * @NFTA_TPROXY_FAMILY: Target address family (NLA_U32: nft_registers) + * @NFTA_TPROXY_REG_ADDR: Target address register (NLA_U32: nft_registers) + * @NFTA_TPROXY_REG_PORT: Target port register (NLA_U32: nft_registers) */ enum nft_tproxy_attributes { NFTA_TPROXY_UNSPEC, @@ -1783,7 +1783,7 @@ enum nft_synproxy_attributes { #define NFTA_SYNPROXY_MAX (__NFTA_SYNPROXY_MAX - 1) /** - * enum nft_device_attributes - nf_tables device netlink attributes + * enum nft_devices_attributes - nf_tables device netlink attributes * * @NFTA_DEVICE_NAME: name of this device (NLA_STRING) * @NFTA_DEVICE_PREFIX: device name prefix, a simple wildcard (NLA_STRING) From df59bb5b9af3fc24d957261e9f80f0c0dec151a4 Mon Sep 17 00:00:00 2001 From: Byungchul Park Date: Wed, 26 Nov 2025 13:36:46 +0900 Subject: [PATCH 1228/1381] netmem, devmem, tcp: access pp fields through @desc in net_iov Convert all the legacy code directly accessing the pp fields in net_iov to access them through @desc in net_iov. Signed-off-by: Byungchul Park Signed-off-by: Jakub Kicinski --- include/linux/skbuff.h | 4 ++-- net/core/devmem.c | 6 +++--- net/ipv4/tcp.c | 2 +- 3 files changed, 6 insertions(+), 6 deletions(-) diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index ff90281ddf90..86737076101d 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -3778,8 +3778,8 @@ static inline dma_addr_t __skb_frag_dma_map(struct device *dev, enum dma_data_direction dir) { if (skb_frag_is_net_iov(frag)) { - return netmem_to_net_iov(frag->netmem)->dma_addr + offset + - frag->offset; + return netmem_to_net_iov(frag->netmem)->desc.dma_addr + + offset + frag->offset; } return dma_map_page(dev, skb_frag_page(frag), skb_frag_off(frag) + offset, size, dir); diff --git a/net/core/devmem.c b/net/core/devmem.c index 1d04754bc756..ec4217d6c0b4 100644 --- a/net/core/devmem.c +++ b/net/core/devmem.c @@ -97,9 +97,9 @@ net_devmem_alloc_dmabuf(struct net_devmem_dmabuf_binding *binding) index = offset / PAGE_SIZE; niov = &owner->area.niovs[index]; - niov->pp_magic = 0; - niov->pp = NULL; - atomic_long_set(&niov->pp_ref_count, 0); + niov->desc.pp_magic = 0; + niov->desc.pp = NULL; + atomic_long_set(&niov->desc.pp_ref_count, 0); return niov; } diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index dee578aad690..f035440c475a 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -2587,7 +2587,7 @@ static int tcp_recvmsg_dmabuf(struct sock *sk, const struct sk_buff *skb, if (err) goto out; - atomic_long_inc(&niov->pp_ref_count); + atomic_long_inc(&niov->desc.pp_ref_count); tcp_xa_pool.netmems[tcp_xa_pool.idx++] = skb_frag_netmem(frag); sent += copy; From 362a161b2582f8d6a34b3b0940f8323adc2f292f Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 25 Nov 2025 19:48:19 -0800 Subject: [PATCH 1229/1381] eth: bnxt: make use of napi_consume_skb() As those following recent changes from Eric know very well using NAPI skb cache is crucial to achieve good perf, at least on recent AMD platforms. Make sure bnxt feeds the skb cache with Tx skbs. Reviewed-by: Michael Chan Reviewed-by: Eric Dumazet Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index a625e7c311dd..8f95f5f29e00 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -877,7 +877,7 @@ static bool __bnxt_tx_int(struct bnxt *bp, struct bnxt_tx_ring_info *txr, next_tx_int: cons = NEXT_TX(cons); - dev_consume_skb_any(skb); + napi_consume_skb(skb, budget); } WRITE_ONCE(txr->tx_cons, cons); From 4c03592689bc19df9deda7a33d56c6ac0cec8651 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 26 Nov 2025 17:43:11 -0800 Subject: [PATCH 1230/1381] net: restore napi_consume_skb()'s NULL-handling Commit e20dfbad8aab ("net: fix napi_consume_skb() with alien skbs") added a skb->cpu check to napi_consume_skb(), before the point where napi_consume_skb() validated skb is not NULL. Add an explicit check to the early exit condition. Reviewed-by: Eric Dumazet Signed-off-by: Jakub Kicinski --- net/core/skbuff.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 5a1d123e7ef7..a00808f7be6a 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -1490,7 +1490,7 @@ void napi_skb_free_stolen_head(struct sk_buff *skb) void napi_consume_skb(struct sk_buff *skb, int budget) { /* Zero budget indicate non-NAPI context called us, like netpoll */ - if (unlikely(!budget)) { + if (unlikely(!budget || !skb)) { dev_consume_skb_any(skb); return; } From 858b1d07e49106a302cc7c7fbeee4fb2698573c9 Mon Sep 17 00:00:00 2001 From: Max Yuan Date: Thu, 27 Nov 2025 00:07:51 +0000 Subject: [PATCH 1231/1381] gve: Fix race condition on tx->dropped_pkt update The tx->dropped_pkt counter is a 64-bit integer that is incremented directly. On 32-bit architectures, this operation is not atomic and can lead to read/write tearing if a reader accesses the counter during the update. This can result in incorrect values being reported for dropped packets. To prevent this potential data corruption, wrap the increment operation with u64_stats_update_begin() and u64_stats_update_end(). This ensures that updates to the 64-bit counter are atomic, even on 32-bit systems, by using a sequence lock. The u64_stats_sync API requires the writer to have exclusive access, which is already provided in this context by the network stack's serialization of the transmit path (net_device_ops::ndo_start_xmit [1]) for a given queue. [1]: https://www.kernel.org/doc/Documentation/networking/netdevices.txt Signed-off-by: Max Yuan Reviewed-by: Jordan Rhee Signed-off-by: Harshitha Ramamurthy Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/google/gve/gve_tx.c | 2 ++ drivers/net/ethernet/google/gve/gve_tx_dqo.c | 6 ++++++ 2 files changed, 8 insertions(+) diff --git a/drivers/net/ethernet/google/gve/gve_tx.c b/drivers/net/ethernet/google/gve/gve_tx.c index c6ff0968929d..97efc8d27e6f 100644 --- a/drivers/net/ethernet/google/gve/gve_tx.c +++ b/drivers/net/ethernet/google/gve/gve_tx.c @@ -730,7 +730,9 @@ static int gve_tx_add_skb_no_copy(struct gve_priv *priv, struct gve_tx_ring *tx, gve_tx_unmap_buf(tx->dev, &tx->info[idx & tx->mask]); } drop: + u64_stats_update_begin(&tx->statss); tx->dropped_pkt++; + u64_stats_update_end(&tx->statss); return 0; } diff --git a/drivers/net/ethernet/google/gve/gve_tx_dqo.c b/drivers/net/ethernet/google/gve/gve_tx_dqo.c index 6f1d515673d2..40b89b3e5a31 100644 --- a/drivers/net/ethernet/google/gve/gve_tx_dqo.c +++ b/drivers/net/ethernet/google/gve/gve_tx_dqo.c @@ -1002,7 +1002,9 @@ static int gve_try_tx_skb(struct gve_priv *priv, struct gve_tx_ring *tx, return 0; drop: + u64_stats_update_begin(&tx->statss); tx->dropped_pkt++; + u64_stats_update_end(&tx->statss); dev_kfree_skb_any(skb); return 0; } @@ -1324,7 +1326,11 @@ static void remove_miss_completions(struct gve_priv *priv, /* This indicates the packet was dropped. */ dev_kfree_skb_any(pending_packet->skb); pending_packet->skb = NULL; + + u64_stats_update_begin(&tx->statss); tx->dropped_pkt++; + u64_stats_update_end(&tx->statss); + net_err_ratelimited("%s: No reinjection completion was received for: %d.\n", priv->dev->name, (int)(pending_packet - tx->dqo.pending_packets)); From eeecf5d3a3a484cedfa3f2f87e6d51a7390ed960 Mon Sep 17 00:00:00 2001 From: "Gustavo A. R. Silva" Date: Thu, 27 Nov 2025 12:12:03 +0900 Subject: [PATCH 1232/1381] net: wwan: mhi_wwan_mbim: Avoid -Wflex-array-member-not-at-end warning Use DEFINE_RAW_FLEX() to avoid a -Wflex-array-member-not-at-end warning. Remove fixed-size array struct usb_cdc_ncm_dpe16 dpe16[2]; from struct mbim_tx_hdr, so that flex-array member struct mbim_tx_hdr::ndp16.dpe16[] ends last in this structure. Compensate for this by using the DEFINE_RAW_FLEX() helper to declare the on-stack struct instance that contains struct usb_cdc_ncm_ndp16 as a member. Adjust the rest of the code, accordingly. So, with these changes fix the following warning: drivers/net/wwan/mhi_wwan_mbim.c:81:34: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end] Reviewed-by: Loic Poulain Signed-off-by: Gustavo A. R. Silva Signed-off-by: Jakub Kicinski --- drivers/net/wwan/mhi_wwan_mbim.c | 17 +++++++++-------- 1 file changed, 9 insertions(+), 8 deletions(-) diff --git a/drivers/net/wwan/mhi_wwan_mbim.c b/drivers/net/wwan/mhi_wwan_mbim.c index f8bc9a39bfa3..0dace12f5ad0 100644 --- a/drivers/net/wwan/mhi_wwan_mbim.c +++ b/drivers/net/wwan/mhi_wwan_mbim.c @@ -78,8 +78,9 @@ struct mhi_mbim_context { struct mbim_tx_hdr { struct usb_cdc_ncm_nth16 nth16; + + /* Must be last as it ends in a flexible-array member. */ struct usb_cdc_ncm_ndp16 ndp16; - struct usb_cdc_ncm_dpe16 dpe16[2]; } __packed; static struct mhi_mbim_link *mhi_mbim_get_link_rcu(struct mhi_mbim_context *mbim, @@ -107,20 +108,20 @@ static int mhi_mbim_get_link_mux_id(struct mhi_controller *cntrl) static struct sk_buff *mbim_tx_fixup(struct sk_buff *skb, unsigned int session, u16 tx_seq) { + DEFINE_RAW_FLEX(struct mbim_tx_hdr, mbim_hdr, ndp16.dpe16, 2); unsigned int dgram_size = skb->len; struct usb_cdc_ncm_nth16 *nth16; struct usb_cdc_ncm_ndp16 *ndp16; - struct mbim_tx_hdr *mbim_hdr; /* Only one NDP is sent, containing the IP packet (no aggregation) */ /* Ensure we have enough headroom for crafting MBIM header */ - if (skb_cow_head(skb, sizeof(struct mbim_tx_hdr))) { + if (skb_cow_head(skb, __struct_size(mbim_hdr))) { dev_kfree_skb_any(skb); return NULL; } - mbim_hdr = skb_push(skb, sizeof(struct mbim_tx_hdr)); + mbim_hdr = skb_push(skb, __struct_size(mbim_hdr)); /* Fill NTB header */ nth16 = &mbim_hdr->nth16; @@ -133,12 +134,11 @@ static struct sk_buff *mbim_tx_fixup(struct sk_buff *skb, unsigned int session, /* Fill the unique NDP */ ndp16 = &mbim_hdr->ndp16; ndp16->dwSignature = cpu_to_le32(USB_CDC_MBIM_NDP16_IPS_SIGN | (session << 24)); - ndp16->wLength = cpu_to_le16(sizeof(struct usb_cdc_ncm_ndp16) - + sizeof(struct usb_cdc_ncm_dpe16) * 2); + ndp16->wLength = cpu_to_le16(struct_size(ndp16, dpe16, 2)); ndp16->wNextNdpIndex = 0; /* Datagram follows the mbim header */ - ndp16->dpe16[0].wDatagramIndex = cpu_to_le16(sizeof(struct mbim_tx_hdr)); + ndp16->dpe16[0].wDatagramIndex = cpu_to_le16(__struct_size(mbim_hdr)); ndp16->dpe16[0].wDatagramLength = cpu_to_le16(dgram_size); /* null termination */ @@ -584,7 +584,8 @@ static void mhi_mbim_setup(struct net_device *ndev) { ndev->header_ops = NULL; /* No header */ ndev->type = ARPHRD_RAWIP; - ndev->needed_headroom = sizeof(struct mbim_tx_hdr); + ndev->needed_headroom = + struct_size_t(struct mbim_tx_hdr, ndp16.dpe16, 2); ndev->hard_header_len = 0; ndev->addr_len = 0; ndev->flags = IFF_POINTOPOINT | IFF_NOARP; From 6557cae0a2a1952645e5df50e1d6eb7267ea2131 Mon Sep 17 00:00:00 2001 From: Peter Enderborg Date: Wed, 26 Nov 2025 14:54:06 +0100 Subject: [PATCH 1233/1381] if_ether.h: Clarify ethertype validity for gsw1xx dsa This 0x88C3 is registered to Infineon Technologies Corporate Research ST and are used by MaxLinear. Infineon made a spin off called Lantiq. Lantiq was acquired by Intel MaxLinear acquired Intels Connected Home division. The product FAQ from MaxLinear describes it's history from the F24S. The driver for the gsw1xx is based on Lantiq showing it's similarities. Ref https://standards-oui.ieee.org/ethertype/eth.txt Signed-off-by: Peter Enderborg Reviewed-by: Andrew Lunn Signed-off-by: Jakub Kicinski --- include/uapi/linux/if_ether.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/include/uapi/linux/if_ether.h b/include/uapi/linux/if_ether.h index 2c93b7b731c8..df9d44a11540 100644 --- a/include/uapi/linux/if_ether.h +++ b/include/uapi/linux/if_ether.h @@ -92,7 +92,9 @@ #define ETH_P_ETHERCAT 0x88A4 /* EtherCAT */ #define ETH_P_8021AD 0x88A8 /* 802.1ad Service VLAN */ #define ETH_P_802_EX1 0x88B5 /* 802.1 Local Experimental 1. */ -#define ETH_P_MXLGSW 0x88C3 /* MaxLinear GSW DSA [ NOT AN OFFICIALLY REGISTERED ID ] */ +#define ETH_P_MXLGSW 0x88C3 /* Infineon Technologies Corporate Research ST + * Used by MaxLinear GSW DSA + */ #define ETH_P_PREAUTH 0x88C7 /* 802.11 Preauthentication */ #define ETH_P_TIPC 0x88CA /* TIPC */ #define ETH_P_LLDP 0x88CC /* Link Layer Discovery Protocol */ From 1e43ebcd5152b3e681a334cc6542fb21770c3a2e Mon Sep 17 00:00:00 2001 From: Michal Schmidt Date: Wed, 26 Nov 2025 10:48:49 +0100 Subject: [PATCH 1234/1381] iavf: Implement settime64 with -EOPNOTSUPP ptp_clock_settime() assumes every ptp_clock has implemented settime64(). Stub it with -EOPNOTSUPP to prevent a NULL dereference. The fix is similar to commit 329d050bbe63 ("gve: Implement settime64 with -EOPNOTSUPP"). Fixes: d734223b2f0d ("iavf: add initial framework for registering PTP clock") Signed-off-by: Michal Schmidt Reviewed-by: Aleksandr Loktionov Reviewed-by: Tim Hostetler Link: https://patch.msgid.link/20251126094850.2842557-1-mschmidt@redhat.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/intel/iavf/iavf_ptp.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/net/ethernet/intel/iavf/iavf_ptp.c b/drivers/net/ethernet/intel/iavf/iavf_ptp.c index b4d5eda2e84f..9cbd8c154031 100644 --- a/drivers/net/ethernet/intel/iavf/iavf_ptp.c +++ b/drivers/net/ethernet/intel/iavf/iavf_ptp.c @@ -252,6 +252,12 @@ static int iavf_ptp_gettimex64(struct ptp_clock_info *info, return iavf_read_phc_indirect(adapter, ts, sts); } +static int iavf_ptp_settime64(struct ptp_clock_info *info, + const struct timespec64 *ts) +{ + return -EOPNOTSUPP; +} + /** * iavf_ptp_cache_phc_time - Cache PHC time for performing timestamp extension * @adapter: private adapter structure @@ -320,6 +326,7 @@ static int iavf_ptp_register_clock(struct iavf_adapter *adapter) KBUILD_MODNAME, dev_name(dev)); ptp_info->owner = THIS_MODULE; ptp_info->gettimex64 = iavf_ptp_gettimex64; + ptp_info->settime64 = iavf_ptp_settime64; ptp_info->do_aux_work = iavf_ptp_do_aux_work; clock = ptp_clock_register(ptp_info, dev); From 1f73a56f986005f0bc64ed23873930e2ee4f5911 Mon Sep 17 00:00:00 2001 From: Antoine Tenart Date: Wed, 26 Nov 2025 11:26:25 +0100 Subject: [PATCH 1235/1381] net: vxlan: prevent NULL deref in vxlan_xmit_one Neither sock4 nor sock6 pointers are guaranteed to be non-NULL in vxlan_xmit_one, e.g. if the iface is brought down. This can lead to the following NULL dereference: BUG: kernel NULL pointer dereference, address: 0000000000000010 Oops: Oops: 0000 [#1] SMP NOPTI RIP: 0010:vxlan_xmit_one+0xbb3/0x1580 Call Trace: vxlan_xmit+0x429/0x610 dev_hard_start_xmit+0x55/0xa0 __dev_queue_xmit+0x6d0/0x7f0 ip_finish_output2+0x24b/0x590 ip_output+0x63/0x110 Mentioned commits changed the code path in vxlan_xmit_one and as a side effect the sock4/6 pointer validity checks in vxlan(6)_get_route were lost. Fix this by adding back checks. Since both commits being fixed were released in the same version (v6.7) and are strongly related, bundle the fixes in a single commit. Reported-by: Liang Li Fixes: 6f19b2c136d9 ("vxlan: use generic function for tunnel IPv4 route lookup") Fixes: 2aceb896ee18 ("vxlan: use generic function for tunnel IPv6 route lookup") Cc: Beniamino Galvani Signed-off-by: Antoine Tenart Reviewed-by: Ido Schimmel Tested-by: Ido Schimmel Link: https://patch.msgid.link/20251126102627.74223-1-atenart@kernel.org Signed-off-by: Jakub Kicinski --- drivers/net/vxlan/vxlan_core.c | 18 +++++++++++++++--- 1 file changed, 15 insertions(+), 3 deletions(-) diff --git a/drivers/net/vxlan/vxlan_core.c b/drivers/net/vxlan/vxlan_core.c index a5c55e7e4d79..e957aa12a8a4 100644 --- a/drivers/net/vxlan/vxlan_core.c +++ b/drivers/net/vxlan/vxlan_core.c @@ -2349,7 +2349,7 @@ void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev, int addr_family; __u8 tos, ttl; int ifindex; - int err; + int err = 0; u32 flags = vxlan->cfg.flags; bool use_cache; bool udp_sum = false; @@ -2454,12 +2454,18 @@ void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev, rcu_read_lock(); if (addr_family == AF_INET) { - struct vxlan_sock *sock4 = rcu_dereference(vxlan->vn4_sock); + struct vxlan_sock *sock4; u16 ipcb_flags = 0; struct rtable *rt; __be16 df = 0; __be32 saddr; + sock4 = rcu_dereference(vxlan->vn4_sock); + if (unlikely(!sock4)) { + reason = SKB_DROP_REASON_DEV_READY; + goto tx_error; + } + if (!ifindex) ifindex = sock4->sock->sk->sk_bound_dev_if; @@ -2534,10 +2540,16 @@ void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev, ipcb_flags); #if IS_ENABLED(CONFIG_IPV6) } else { - struct vxlan_sock *sock6 = rcu_dereference(vxlan->vn6_sock); + struct vxlan_sock *sock6; struct in6_addr saddr; u16 ip6cb_flags = 0; + sock6 = rcu_dereference(vxlan->vn6_sock); + if (unlikely(!sock6)) { + reason = SKB_DROP_REASON_DEV_READY; + goto tx_error; + } + if (!ifindex) ifindex = sock6->sock->sk->sk_bound_dev_if; From 17e9f841dd227a4dc976b22d000d5f669bc14493 Mon Sep 17 00:00:00 2001 From: Javen Xu Date: Wed, 26 Nov 2025 13:59:50 +0800 Subject: [PATCH 1236/1381] r8169: add DASH support for RTL8127AP This adds DASH support for chip RTL8127AP. Its mac version is RTL_GIGA_MAC_VER_80. DASH is a standard for remote management of network device, allowing out-of-band control. Signed-off-by: Javen Xu Link: https://patch.msgid.link/20251126055950.2050-1-javen_xu@realsil.com.cn Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/realtek/r8169_main.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c index 405e91eb3141..630319604211 100644 --- a/drivers/net/ethernet/realtek/r8169_main.c +++ b/drivers/net/ethernet/realtek/r8169_main.c @@ -1512,6 +1512,7 @@ static enum rtl_dash_type rtl_get_dash_type(struct rtl8169_private *tp) case RTL_GIGA_MAC_VER_51 ... RTL_GIGA_MAC_VER_52: return RTL_DASH_EP; case RTL_GIGA_MAC_VER_66: + case RTL_GIGA_MAC_VER_80: return RTL_DASH_25_BP; default: return RTL_DASH_NONE; From 510026a39849ae7066dce17343caba4550e2a977 Mon Sep 17 00:00:00 2001 From: David Yang Date: Wed, 26 Nov 2025 16:40:19 +0800 Subject: [PATCH 1237/1381] net: dsa: yt921x: Fix parsing MIB attributes There are hard-to-find unused fields in the MIB table I didn't notice in the example driver code, causing wrong interpretation of the MIB data. For some 64-bit attributes, the current (wrong) implementation took the correct lower 32 bits, but messed up the upper 32 bits, so it would work accidentally until 32-bit overflows happen. Fix that too. Fixes: 186623f4aa72 ("net: dsa: yt921x: Add support for Motorcomm YT921x") Signed-off-by: David Yang Link: https://patch.msgid.link/20251126084024.2843851-2-mmyangfl@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/yt921x.c | 22 +++++++++++----------- 1 file changed, 11 insertions(+), 11 deletions(-) diff --git a/drivers/net/dsa/yt921x.c b/drivers/net/dsa/yt921x.c index 944988e29127..97fc6085f4d0 100644 --- a/drivers/net/dsa/yt921x.c +++ b/drivers/net/dsa/yt921x.c @@ -56,13 +56,13 @@ static const struct yt921x_mib_desc yt921x_mib_descs[] = { MIB_DESC(1, 0x30, NULL), /* RxPktSz1024To1518 */ MIB_DESC(1, 0x34, NULL), /* RxPktSz1519ToMax */ - MIB_DESC(2, 0x38, NULL), /* RxGoodBytes */ - /* 0x3c */ + /* 0x38 unused */ + MIB_DESC(2, 0x3c, NULL), /* RxGoodBytes */ - MIB_DESC(2, 0x40, "RxBadBytes"), - /* 0x44 */ - MIB_DESC(2, 0x48, NULL), /* RxOverSzErr */ - /* 0x4c */ + /* 0x40 */ + MIB_DESC(2, 0x44, "RxBadBytes"), + /* 0x48 */ + MIB_DESC(1, 0x4c, NULL), /* RxOverSzErr */ MIB_DESC(1, 0x50, NULL), /* RxDropped */ MIB_DESC(1, 0x54, NULL), /* TxBroadcast */ @@ -79,10 +79,10 @@ static const struct yt921x_mib_desc yt921x_mib_descs[] = { MIB_DESC(1, 0x78, NULL), /* TxPktSz1024To1518 */ MIB_DESC(1, 0x7c, NULL), /* TxPktSz1519ToMax */ - MIB_DESC(2, 0x80, NULL), /* TxGoodBytes */ - /* 0x84 */ - MIB_DESC(2, 0x88, NULL), /* TxCollision */ - /* 0x8c */ + /* 0x80 unused */ + MIB_DESC(2, 0x84, NULL), /* TxGoodBytes */ + /* 0x88 */ + MIB_DESC(1, 0x8c, NULL), /* TxCollision */ MIB_DESC(1, 0x90, NULL), /* TxExcessiveCollistion */ MIB_DESC(1, 0x94, NULL), /* TxMultipleCollision */ @@ -705,7 +705,7 @@ static int yt921x_read_mib(struct yt921x_priv *priv, int port) res = yt921x_reg_read(priv, reg + 4, &val1); if (res) break; - val = ((u64)val0 << 32) | val1; + val = ((u64)val1 << 32) | val0; } WRITE_ONCE(*valp, val); From fbce7b36c8c64b2bab68f2ec878b5aa48b6f0d75 Mon Sep 17 00:00:00 2001 From: David Yang Date: Wed, 26 Nov 2025 16:40:20 +0800 Subject: [PATCH 1238/1381] net: dsa: yt921x: Use macros for MIB locations Extract MIB constants into the header file to improve code style. This patch will not change the behavior of the function. Signed-off-by: David Yang Link: https://patch.msgid.link/20251126084024.2843851-3-mmyangfl@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/yt921x.c | 83 +++++++++++++++++++--------------------- drivers/net/dsa/yt921x.h | 54 ++++++++++++++++++++++++++ 2 files changed, 93 insertions(+), 44 deletions(-) diff --git a/drivers/net/dsa/yt921x.c b/drivers/net/dsa/yt921x.c index 97fc6085f4d0..ebfd34f72314 100644 --- a/drivers/net/dsa/yt921x.c +++ b/drivers/net/dsa/yt921x.c @@ -39,60 +39,55 @@ struct yt921x_mib_desc { * to perform 32bit MIB overflow wraparound. */ static const struct yt921x_mib_desc yt921x_mib_descs[] = { - MIB_DESC(1, 0x00, NULL), /* RxBroadcast */ - MIB_DESC(1, 0x04, NULL), /* RxPause */ - MIB_DESC(1, 0x08, NULL), /* RxMulticast */ - MIB_DESC(1, 0x0c, NULL), /* RxCrcErr */ + MIB_DESC(1, YT921X_MIB_DATA_RX_BROADCAST, NULL), + MIB_DESC(1, YT921X_MIB_DATA_RX_PAUSE, NULL), + MIB_DESC(1, YT921X_MIB_DATA_RX_MULTICAST, NULL), + MIB_DESC(1, YT921X_MIB_DATA_RX_CRC_ERR, NULL), - MIB_DESC(1, 0x10, NULL), /* RxAlignErr */ - MIB_DESC(1, 0x14, NULL), /* RxUnderSizeErr */ - MIB_DESC(1, 0x18, NULL), /* RxFragErr */ - MIB_DESC(1, 0x1c, NULL), /* RxPktSz64 */ + MIB_DESC(1, YT921X_MIB_DATA_RX_ALIGN_ERR, NULL), + MIB_DESC(1, YT921X_MIB_DATA_RX_UNDERSIZE_ERR, NULL), + MIB_DESC(1, YT921X_MIB_DATA_RX_FRAG_ERR, NULL), + MIB_DESC(1, YT921X_MIB_DATA_RX_PKT_SZ_64, NULL), - MIB_DESC(1, 0x20, NULL), /* RxPktSz65To127 */ - MIB_DESC(1, 0x24, NULL), /* RxPktSz128To255 */ - MIB_DESC(1, 0x28, NULL), /* RxPktSz256To511 */ - MIB_DESC(1, 0x2c, NULL), /* RxPktSz512To1023 */ + MIB_DESC(1, YT921X_MIB_DATA_RX_PKT_SZ_65_TO_127, NULL), + MIB_DESC(1, YT921X_MIB_DATA_RX_PKT_SZ_128_TO_255, NULL), + MIB_DESC(1, YT921X_MIB_DATA_RX_PKT_SZ_256_TO_511, NULL), + MIB_DESC(1, YT921X_MIB_DATA_RX_PKT_SZ_512_TO_1023, NULL), - MIB_DESC(1, 0x30, NULL), /* RxPktSz1024To1518 */ - MIB_DESC(1, 0x34, NULL), /* RxPktSz1519ToMax */ - /* 0x38 unused */ - MIB_DESC(2, 0x3c, NULL), /* RxGoodBytes */ + MIB_DESC(1, YT921X_MIB_DATA_RX_PKT_SZ_1024_TO_1518, NULL), + MIB_DESC(1, YT921X_MIB_DATA_RX_PKT_SZ_1519_TO_MAX, NULL), + MIB_DESC(2, YT921X_MIB_DATA_RX_GOOD_BYTES, NULL), - /* 0x40 */ - MIB_DESC(2, 0x44, "RxBadBytes"), - /* 0x48 */ - MIB_DESC(1, 0x4c, NULL), /* RxOverSzErr */ + MIB_DESC(2, YT921X_MIB_DATA_RX_BAD_BYTES, "RxBadBytes"), + MIB_DESC(1, YT921X_MIB_DATA_RX_OVERSIZE_ERR, NULL), - MIB_DESC(1, 0x50, NULL), /* RxDropped */ - MIB_DESC(1, 0x54, NULL), /* TxBroadcast */ - MIB_DESC(1, 0x58, NULL), /* TxPause */ - MIB_DESC(1, 0x5c, NULL), /* TxMulticast */ + MIB_DESC(1, YT921X_MIB_DATA_RX_DROPPED, NULL), + MIB_DESC(1, YT921X_MIB_DATA_TX_BROADCAST, NULL), + MIB_DESC(1, YT921X_MIB_DATA_TX_PAUSE, NULL), + MIB_DESC(1, YT921X_MIB_DATA_TX_MULTICAST, NULL), - MIB_DESC(1, 0x60, NULL), /* TxUnderSizeErr */ - MIB_DESC(1, 0x64, NULL), /* TxPktSz64 */ - MIB_DESC(1, 0x68, NULL), /* TxPktSz65To127 */ - MIB_DESC(1, 0x6c, NULL), /* TxPktSz128To255 */ + MIB_DESC(1, YT921X_MIB_DATA_TX_UNDERSIZE_ERR, NULL), + MIB_DESC(1, YT921X_MIB_DATA_TX_PKT_SZ_64, NULL), + MIB_DESC(1, YT921X_MIB_DATA_TX_PKT_SZ_65_TO_127, NULL), + MIB_DESC(1, YT921X_MIB_DATA_TX_PKT_SZ_128_TO_255, NULL), - MIB_DESC(1, 0x70, NULL), /* TxPktSz256To511 */ - MIB_DESC(1, 0x74, NULL), /* TxPktSz512To1023 */ - MIB_DESC(1, 0x78, NULL), /* TxPktSz1024To1518 */ - MIB_DESC(1, 0x7c, NULL), /* TxPktSz1519ToMax */ + MIB_DESC(1, YT921X_MIB_DATA_TX_PKT_SZ_256_TO_511, NULL), + MIB_DESC(1, YT921X_MIB_DATA_TX_PKT_SZ_512_TO_1023, NULL), + MIB_DESC(1, YT921X_MIB_DATA_TX_PKT_SZ_1024_TO_1518, NULL), + MIB_DESC(1, YT921X_MIB_DATA_TX_PKT_SZ_1519_TO_MAX, NULL), - /* 0x80 unused */ - MIB_DESC(2, 0x84, NULL), /* TxGoodBytes */ - /* 0x88 */ - MIB_DESC(1, 0x8c, NULL), /* TxCollision */ + MIB_DESC(2, YT921X_MIB_DATA_TX_GOOD_BYTES, NULL), + MIB_DESC(1, YT921X_MIB_DATA_TX_COLLISION, NULL), - MIB_DESC(1, 0x90, NULL), /* TxExcessiveCollistion */ - MIB_DESC(1, 0x94, NULL), /* TxMultipleCollision */ - MIB_DESC(1, 0x98, NULL), /* TxSingleCollision */ - MIB_DESC(1, 0x9c, NULL), /* TxPkt */ + MIB_DESC(1, YT921X_MIB_DATA_TX_EXCESSIVE_COLLISION, NULL), + MIB_DESC(1, YT921X_MIB_DATA_TX_MULTIPLE_COLLISION, NULL), + MIB_DESC(1, YT921X_MIB_DATA_TX_SINGLE_COLLISION, NULL), + MIB_DESC(1, YT921X_MIB_DATA_TX_PKT, NULL), - MIB_DESC(1, 0xa0, NULL), /* TxDeferred */ - MIB_DESC(1, 0xa4, NULL), /* TxLateCollision */ - MIB_DESC(1, 0xa8, "RxOAM"), - MIB_DESC(1, 0xac, "TxOAM"), + MIB_DESC(1, YT921X_MIB_DATA_TX_DEFERRED, NULL), + MIB_DESC(1, YT921X_MIB_DATA_TX_LATE_COLLISION, NULL), + MIB_DESC(1, YT921X_MIB_DATA_RX_OAM, "RxOAM"), + MIB_DESC(1, YT921X_MIB_DATA_TX_OAM, "TxOAM"), }; struct yt921x_info { diff --git a/drivers/net/dsa/yt921x.h b/drivers/net/dsa/yt921x.h index 3e85d90826fb..44719d841d40 100644 --- a/drivers/net/dsa/yt921x.h +++ b/drivers/net/dsa/yt921x.h @@ -173,6 +173,60 @@ #define YT921X_MIB_CTRL_ALL_PORT BIT(0) #define YT921X_MIBn_DATA0(port) (0xc0100 + 0x100 * (port)) #define YT921X_MIBn_DATAm(port, x) (YT921X_MIBn_DATA0(port) + 4 * (x)) +#define YT921X_MIB_DATA_RX_BROADCAST 0x00 +#define YT921X_MIB_DATA_RX_PAUSE 0x04 +#define YT921X_MIB_DATA_RX_MULTICAST 0x08 +#define YT921X_MIB_DATA_RX_CRC_ERR 0x0c + +#define YT921X_MIB_DATA_RX_ALIGN_ERR 0x10 +#define YT921X_MIB_DATA_RX_UNDERSIZE_ERR 0x14 +#define YT921X_MIB_DATA_RX_FRAG_ERR 0x18 +#define YT921X_MIB_DATA_RX_PKT_SZ_64 0x1c + +#define YT921X_MIB_DATA_RX_PKT_SZ_65_TO_127 0x20 +#define YT921X_MIB_DATA_RX_PKT_SZ_128_TO_255 0x24 +#define YT921X_MIB_DATA_RX_PKT_SZ_256_TO_511 0x28 +#define YT921X_MIB_DATA_RX_PKT_SZ_512_TO_1023 0x2c + +#define YT921X_MIB_DATA_RX_PKT_SZ_1024_TO_1518 0x30 +#define YT921X_MIB_DATA_RX_PKT_SZ_1519_TO_MAX 0x34 +/* 0x38: unused */ +#define YT921X_MIB_DATA_RX_GOOD_BYTES 0x3c + +/* 0x40: 64 bytes */ +#define YT921X_MIB_DATA_RX_BAD_BYTES 0x44 +/* 0x48: 64 bytes */ +#define YT921X_MIB_DATA_RX_OVERSIZE_ERR 0x4c + +#define YT921X_MIB_DATA_RX_DROPPED 0x50 +#define YT921X_MIB_DATA_TX_BROADCAST 0x54 +#define YT921X_MIB_DATA_TX_PAUSE 0x58 +#define YT921X_MIB_DATA_TX_MULTICAST 0x5c + +#define YT921X_MIB_DATA_TX_UNDERSIZE_ERR 0x60 +#define YT921X_MIB_DATA_TX_PKT_SZ_64 0x64 +#define YT921X_MIB_DATA_TX_PKT_SZ_65_TO_127 0x68 +#define YT921X_MIB_DATA_TX_PKT_SZ_128_TO_255 0x6c + +#define YT921X_MIB_DATA_TX_PKT_SZ_256_TO_511 0x70 +#define YT921X_MIB_DATA_TX_PKT_SZ_512_TO_1023 0x74 +#define YT921X_MIB_DATA_TX_PKT_SZ_1024_TO_1518 0x78 +#define YT921X_MIB_DATA_TX_PKT_SZ_1519_TO_MAX 0x7c + +/* 0x80: unused */ +#define YT921X_MIB_DATA_TX_GOOD_BYTES 0x84 +/* 0x88: 64 bytes */ +#define YT921X_MIB_DATA_TX_COLLISION 0x8c + +#define YT921X_MIB_DATA_TX_EXCESSIVE_COLLISION 0x90 +#define YT921X_MIB_DATA_TX_MULTIPLE_COLLISION 0x94 +#define YT921X_MIB_DATA_TX_SINGLE_COLLISION 0x98 +#define YT921X_MIB_DATA_TX_PKT 0x9c + +#define YT921X_MIB_DATA_TX_DEFERRED 0xa0 +#define YT921X_MIB_DATA_TX_LATE_COLLISION 0xa4 +#define YT921X_MIB_DATA_RX_OAM 0xa8 +#define YT921X_MIB_DATA_TX_OAM 0xac #define YT921X_EDATA_CTRL 0xe0000 #define YT921X_EDATA_CTRL_ADDR_M GENMASK(15, 8) From 8048168df56e225c94e50b04cb7b0514135d7a1c Mon Sep 17 00:00:00 2001 From: Alexey Kodanev Date: Wed, 26 Nov 2025 10:43:27 +0000 Subject: [PATCH 1239/1381] net: stmmac: fix rx limit check in stmmac_rx_zc() The extra "count >= limit" check in stmmac_rx_zc() is redundant and has no effect because the value of "count" doesn't change after the while condition at this point. However, it can change after "read_again:" label: while (count < limit) { ... if (count >= limit) break; read_again: ... /* XSK pool expects RX frame 1:1 mapped to XSK buffer */ if (likely(status & rx_not_ls)) { xsk_buff_free(buf->xdp); buf->xdp = NULL; dirty++; count++; goto read_again; } ... This patch addresses the same issue previously resolved in stmmac_rx() by commit fa02de9e7588 ("net: stmmac: fix rx budget limit check"). The fix is the same: move the check after the label to ensure that it bounds the goto loop. Fixes: bba2556efad6 ("net: stmmac: Enable RX via AF_XDP zero-copy") Signed-off-by: Alexey Kodanev Reviewed-by: Russell King (Oracle) Link: https://patch.msgid.link/20251126104327.175590-1-aleksei.kodanev@bell-sw.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index 1ecca60baf74..da206b24aaed 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -5396,10 +5396,10 @@ static int stmmac_rx_zc(struct stmmac_priv *priv, int limit, u32 queue) len = 0; } +read_again: if (count >= limit) break; -read_again: buf1_len = 0; entry = next_entry; buf = &rx_q->buf_pool[entry]; From 73880e66b79a6c950c0388559e2d425da2c37bd6 Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Wed, 26 Nov 2025 02:54:40 -0800 Subject: [PATCH 1240/1381] net: thunder: convert to use .get_rx_ring_count Convert the Cavium Thunder NIC VF driver to use the new .get_rx_ring_count ethtool operation instead of implementing .get_rxnfc solely for handling ETHTOOL_GRXRINGS command. This simplifies the code by removing the switch statement and replacing it with a direct return of the queue count. The new callback provides the same functionality in a more direct way, following the ongoing ethtool API modernization. Signed-off-by: Breno Leitao Link: https://patch.msgid.link/20251126-gxring_cavium-v1-1-a066c0c9e0c6@debian.org Signed-off-by: Jakub Kicinski --- .../net/ethernet/cavium/thunder/nicvf_ethtool.c | 16 +++------------- 1 file changed, 3 insertions(+), 13 deletions(-) diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_ethtool.c b/drivers/net/ethernet/cavium/thunder/nicvf_ethtool.c index fc6053414b7d..413028bdcacb 100644 --- a/drivers/net/ethernet/cavium/thunder/nicvf_ethtool.c +++ b/drivers/net/ethernet/cavium/thunder/nicvf_ethtool.c @@ -541,21 +541,11 @@ static int nicvf_get_rxfh_fields(struct net_device *dev, return 0; } -static int nicvf_get_rxnfc(struct net_device *dev, - struct ethtool_rxnfc *info, u32 *rules) +static u32 nicvf_get_rx_ring_count(struct net_device *dev) { struct nicvf *nic = netdev_priv(dev); - int ret = -EOPNOTSUPP; - switch (info->cmd) { - case ETHTOOL_GRXRINGS: - info->data = nic->rx_queues; - ret = 0; - break; - default: - break; - } - return ret; + return nic->rx_queues; } static int nicvf_set_rxfh_fields(struct net_device *dev, @@ -861,7 +851,7 @@ static const struct ethtool_ops nicvf_ethtool_ops = { .get_coalesce = nicvf_get_coalesce, .get_ringparam = nicvf_get_ringparam, .set_ringparam = nicvf_set_ringparam, - .get_rxnfc = nicvf_get_rxnfc, + .get_rx_ring_count = nicvf_get_rx_ring_count, .get_rxfh_key_size = nicvf_get_rxfh_key_size, .get_rxfh_indir_size = nicvf_get_rxfh_indir_size, .get_rxfh = nicvf_get_rxfh, From 4636b4e797f71ad68332e174d8b50a65e7d5cf8a Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Wed, 26 Nov 2025 15:22:42 +0100 Subject: [PATCH 1241/1381] dibs: Remove KMSG_COMPONENT macro The KMSG_COMPONENT macro is a leftover of the s390 specific "kernel message catalog" from 2008 [1] which never made it upstream. The macro was added to s390 code to allow for an out-of-tree patch which used this to generate unique message ids. Also this out-of-tree doesn't exist anymore. The pattern of how the KMSG_COMPONENT is used was partially also used for non s390 specific code, for whatever reasons. Remove the macro in order to get rid of a pointless indirection. [1] https://lwn.net/Articles/292650/ Signed-off-by: Heiko Carstens Acked-by: Alexandra Winter Link: https://patch.msgid.link/20251126142242.2124317-1-hca@linux.ibm.com Signed-off-by: Jakub Kicinski --- drivers/dibs/dibs_main.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/dibs/dibs_main.c b/drivers/dibs/dibs_main.c index dac14d843af7..b8c16586706c 100644 --- a/drivers/dibs/dibs_main.c +++ b/drivers/dibs/dibs_main.c @@ -6,8 +6,7 @@ * * Copyright IBM Corp. 2025 */ -#define KMSG_COMPONENT "dibs" -#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt +#define pr_fmt(fmt) "dibs: " fmt #include #include From 1645759a0405aaeaa9fa5d0a4167a3cb626808bb Mon Sep 17 00:00:00 2001 From: Natalia Wochtman Date: Tue, 25 Nov 2025 14:36:23 -0800 Subject: [PATCH 1242/1381] ixgbevf: ixgbevf_q_vector clean up Flex array should be at the end of the structure and use [] syntax Remove unused fields of ixgbevf_q_vector. They aren't used since busy poll was moved to core code in commit 508aac6dee02 ("ixgbevf: get rid of custom busy polling code"). Reviewed-by: Przemek Kitszel Reviewed-by: Aleksandr Loktionov Signed-off-by: Natalia Wochtman Reviewed-by: Jacob Keller Tested-by: Rafal Romanowski Signed-off-by: Tony Nguyen Link: https://patch.msgid.link/20251125223632.1857532-5-anthony.l.nguyen@intel.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/intel/ixgbevf/ixgbevf.h | 18 +----------------- 1 file changed, 1 insertion(+), 17 deletions(-) diff --git a/drivers/net/ethernet/intel/ixgbevf/ixgbevf.h b/drivers/net/ethernet/intel/ixgbevf/ixgbevf.h index 039187607e98..516a6fdd23d0 100644 --- a/drivers/net/ethernet/intel/ixgbevf/ixgbevf.h +++ b/drivers/net/ethernet/intel/ixgbevf/ixgbevf.h @@ -241,23 +241,7 @@ struct ixgbevf_q_vector { char name[IFNAMSIZ + 9]; /* for dynamic allocation of rings associated with this q_vector */ - struct ixgbevf_ring ring[0] ____cacheline_internodealigned_in_smp; -#ifdef CONFIG_NET_RX_BUSY_POLL - unsigned int state; -#define IXGBEVF_QV_STATE_IDLE 0 -#define IXGBEVF_QV_STATE_NAPI 1 /* NAPI owns this QV */ -#define IXGBEVF_QV_STATE_POLL 2 /* poll owns this QV */ -#define IXGBEVF_QV_STATE_DISABLED 4 /* QV is disabled */ -#define IXGBEVF_QV_OWNED (IXGBEVF_QV_STATE_NAPI | IXGBEVF_QV_STATE_POLL) -#define IXGBEVF_QV_LOCKED (IXGBEVF_QV_OWNED | IXGBEVF_QV_STATE_DISABLED) -#define IXGBEVF_QV_STATE_NAPI_YIELD 8 /* NAPI yielded this QV */ -#define IXGBEVF_QV_STATE_POLL_YIELD 16 /* poll yielded this QV */ -#define IXGBEVF_QV_YIELD (IXGBEVF_QV_STATE_NAPI_YIELD | \ - IXGBEVF_QV_STATE_POLL_YIELD) -#define IXGBEVF_QV_USER_PEND (IXGBEVF_QV_STATE_POLL | \ - IXGBEVF_QV_STATE_POLL_YIELD) - spinlock_t lock; -#endif /* CONFIG_NET_RX_BUSY_POLL */ + struct ixgbevf_ring ring[] ____cacheline_internodealigned_in_smp; }; /* microsecond values for various ITR rates shifted by 2 to fit itr register From 8dd72ebc73f37b216410db17340f15e6fb2cdb7b Mon Sep 17 00:00:00 2001 From: Emil Tantilov Date: Tue, 25 Nov 2025 14:36:24 -0800 Subject: [PATCH 1243/1381] idpf: convert vport state to bitmap Convert vport state to a bitmap and remove the DOWN state which is redundant in the existing logic. There are no functional changes aside from the use of bitwise operations when setting and checking the states. Removed the double underscore to be consistent with the naming of other bitmaps in the header and renamed current_state to vport_is_up to match the meaning of the new variable. Reviewed-by: Przemek Kitszel Reviewed-by: Aleksandr Loktionov Reviewed-by: Chittim Madhu Signed-off-by: Emil Tantilov Tested-by: Samuel Salin Signed-off-by: Tony Nguyen Link: https://patch.msgid.link/20251125223632.1857532-6-anthony.l.nguyen@intel.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/intel/idpf/idpf.h | 12 ++++------ .../net/ethernet/intel/idpf/idpf_ethtool.c | 12 +++++----- drivers/net/ethernet/intel/idpf/idpf_lib.c | 24 +++++++++---------- .../ethernet/intel/idpf/idpf_singleq_txrx.c | 2 +- drivers/net/ethernet/intel/idpf/idpf_txrx.c | 2 +- .../net/ethernet/intel/idpf/idpf_virtchnl.c | 4 ++-- drivers/net/ethernet/intel/idpf/xdp.c | 2 +- 7 files changed, 28 insertions(+), 30 deletions(-) diff --git a/drivers/net/ethernet/intel/idpf/idpf.h b/drivers/net/ethernet/intel/idpf/idpf.h index 50fa7be0c00d..8cfc68cbfa06 100644 --- a/drivers/net/ethernet/intel/idpf/idpf.h +++ b/drivers/net/ethernet/intel/idpf/idpf.h @@ -131,14 +131,12 @@ enum idpf_cap_field { /** * enum idpf_vport_state - Current vport state - * @__IDPF_VPORT_DOWN: Vport is down - * @__IDPF_VPORT_UP: Vport is up - * @__IDPF_VPORT_STATE_LAST: Must be last, number of states + * @IDPF_VPORT_UP: Vport is up + * @IDPF_VPORT_STATE_NBITS: Must be last, number of states */ enum idpf_vport_state { - __IDPF_VPORT_DOWN, - __IDPF_VPORT_UP, - __IDPF_VPORT_STATE_LAST, + IDPF_VPORT_UP, + IDPF_VPORT_STATE_NBITS }; /** @@ -162,7 +160,7 @@ struct idpf_netdev_priv { u16 vport_idx; u16 max_tx_hdr_size; u16 tx_max_bufs; - enum idpf_vport_state state; + DECLARE_BITMAP(state, IDPF_VPORT_STATE_NBITS); struct rtnl_link_stats64 netstats; spinlock_t stats_lock; }; diff --git a/drivers/net/ethernet/intel/idpf/idpf_ethtool.c b/drivers/net/ethernet/intel/idpf/idpf_ethtool.c index 5dd82b1349c5..2589e124e41c 100644 --- a/drivers/net/ethernet/intel/idpf/idpf_ethtool.c +++ b/drivers/net/ethernet/intel/idpf/idpf_ethtool.c @@ -402,7 +402,7 @@ static int idpf_get_rxfh(struct net_device *netdev, } rss_data = &adapter->vport_config[np->vport_idx]->user_config.rss_data; - if (np->state != __IDPF_VPORT_UP) + if (!test_bit(IDPF_VPORT_UP, np->state)) goto unlock_mutex; rxfh->hfunc = ETH_RSS_HASH_TOP; @@ -452,7 +452,7 @@ static int idpf_set_rxfh(struct net_device *netdev, } rss_data = &adapter->vport_config[vport->idx]->user_config.rss_data; - if (np->state != __IDPF_VPORT_UP) + if (!test_bit(IDPF_VPORT_UP, np->state)) goto unlock_mutex; if (rxfh->hfunc != ETH_RSS_HASH_NO_CHANGE && @@ -1183,7 +1183,7 @@ static void idpf_get_ethtool_stats(struct net_device *netdev, idpf_vport_ctrl_lock(netdev); vport = idpf_netdev_to_vport(netdev); - if (np->state != __IDPF_VPORT_UP) { + if (!test_bit(IDPF_VPORT_UP, np->state)) { idpf_vport_ctrl_unlock(netdev); return; @@ -1335,7 +1335,7 @@ static int idpf_get_q_coalesce(struct net_device *netdev, idpf_vport_ctrl_lock(netdev); vport = idpf_netdev_to_vport(netdev); - if (np->state != __IDPF_VPORT_UP) + if (!test_bit(IDPF_VPORT_UP, np->state)) goto unlock_mutex; if (q_num >= vport->num_rxq && q_num >= vport->num_txq) { @@ -1523,7 +1523,7 @@ static int idpf_set_coalesce(struct net_device *netdev, idpf_vport_ctrl_lock(netdev); vport = idpf_netdev_to_vport(netdev); - if (np->state != __IDPF_VPORT_UP) + if (!test_bit(IDPF_VPORT_UP, np->state)) goto unlock_mutex; for (i = 0; i < vport->num_txq; i++) { @@ -1726,7 +1726,7 @@ static void idpf_get_ts_stats(struct net_device *netdev, ts_stats->err = u64_stats_read(&vport->tstamp_stats.discarded); } while (u64_stats_fetch_retry(&vport->tstamp_stats.stats_sync, start)); - if (np->state != __IDPF_VPORT_UP) + if (!test_bit(IDPF_VPORT_UP, np->state)) goto exit; for (u16 i = 0; i < vport->num_txq_grp; i++) { diff --git a/drivers/net/ethernet/intel/idpf/idpf_lib.c b/drivers/net/ethernet/intel/idpf/idpf_lib.c index 8a941f0fb048..7a7e101afeb6 100644 --- a/drivers/net/ethernet/intel/idpf/idpf_lib.c +++ b/drivers/net/ethernet/intel/idpf/idpf_lib.c @@ -519,7 +519,7 @@ static int idpf_del_mac_filter(struct idpf_vport *vport, } spin_unlock_bh(&vport_config->mac_filter_list_lock); - if (np->state == __IDPF_VPORT_UP) { + if (test_bit(IDPF_VPORT_UP, np->state)) { int err; err = idpf_add_del_mac_filters(vport, np, false, async); @@ -590,7 +590,7 @@ static int idpf_add_mac_filter(struct idpf_vport *vport, if (err) return err; - if (np->state == __IDPF_VPORT_UP) + if (test_bit(IDPF_VPORT_UP, np->state)) err = idpf_add_del_mac_filters(vport, np, true, async); return err; @@ -894,7 +894,7 @@ static void idpf_vport_stop(struct idpf_vport *vport, bool rtnl) { struct idpf_netdev_priv *np = netdev_priv(vport->netdev); - if (np->state <= __IDPF_VPORT_DOWN) + if (!test_bit(IDPF_VPORT_UP, np->state)) return; if (rtnl) @@ -921,7 +921,7 @@ static void idpf_vport_stop(struct idpf_vport *vport, bool rtnl) idpf_xdp_rxq_info_deinit_all(vport); idpf_vport_queues_rel(vport); idpf_vport_intr_rel(vport); - np->state = __IDPF_VPORT_DOWN; + clear_bit(IDPF_VPORT_UP, np->state); if (rtnl) rtnl_unlock(); @@ -1345,7 +1345,7 @@ static int idpf_up_complete(struct idpf_vport *vport) netif_tx_start_all_queues(vport->netdev); } - np->state = __IDPF_VPORT_UP; + set_bit(IDPF_VPORT_UP, np->state); return 0; } @@ -1391,7 +1391,7 @@ static int idpf_vport_open(struct idpf_vport *vport, bool rtnl) struct idpf_vport_config *vport_config; int err; - if (np->state != __IDPF_VPORT_DOWN) + if (test_bit(IDPF_VPORT_UP, np->state)) return -EBUSY; if (rtnl) @@ -1602,7 +1602,7 @@ void idpf_init_task(struct work_struct *work) /* Once state is put into DOWN, driver is ready for dev_open */ np = netdev_priv(vport->netdev); - np->state = __IDPF_VPORT_DOWN; + clear_bit(IDPF_VPORT_UP, np->state); if (test_and_clear_bit(IDPF_VPORT_UP_REQUESTED, vport_config->flags)) idpf_vport_open(vport, true); @@ -1801,7 +1801,7 @@ static void idpf_set_vport_state(struct idpf_adapter *adapter) continue; np = netdev_priv(adapter->netdevs[i]); - if (np->state == __IDPF_VPORT_UP) + if (test_bit(IDPF_VPORT_UP, np->state)) set_bit(IDPF_VPORT_UP_REQUESTED, adapter->vport_config[i]->flags); } @@ -1939,7 +1939,7 @@ int idpf_initiate_soft_reset(struct idpf_vport *vport, enum idpf_vport_reset_cause reset_cause) { struct idpf_netdev_priv *np = netdev_priv(vport->netdev); - enum idpf_vport_state current_state = np->state; + bool vport_is_up = test_bit(IDPF_VPORT_UP, np->state); struct idpf_adapter *adapter = vport->adapter; struct idpf_vport *new_vport; int err; @@ -1990,7 +1990,7 @@ int idpf_initiate_soft_reset(struct idpf_vport *vport, goto free_vport; } - if (current_state <= __IDPF_VPORT_DOWN) { + if (!vport_is_up) { idpf_send_delete_queues_msg(vport); } else { set_bit(IDPF_VPORT_DEL_QUEUES, vport->flags); @@ -2023,7 +2023,7 @@ int idpf_initiate_soft_reset(struct idpf_vport *vport, if (err) goto err_open; - if (current_state == __IDPF_VPORT_UP) + if (vport_is_up) err = idpf_vport_open(vport, false); goto free_vport; @@ -2033,7 +2033,7 @@ int idpf_initiate_soft_reset(struct idpf_vport *vport, vport->num_rxq, vport->num_bufq); err_open: - if (current_state == __IDPF_VPORT_UP) + if (vport_is_up) idpf_vport_open(vport, false); free_vport: diff --git a/drivers/net/ethernet/intel/idpf/idpf_singleq_txrx.c b/drivers/net/ethernet/intel/idpf/idpf_singleq_txrx.c index 61e613066140..e3ddf18dcbf5 100644 --- a/drivers/net/ethernet/intel/idpf/idpf_singleq_txrx.c +++ b/drivers/net/ethernet/intel/idpf/idpf_singleq_txrx.c @@ -570,7 +570,7 @@ static bool idpf_tx_singleq_clean(struct idpf_tx_queue *tx_q, int napi_budget, np = netdev_priv(tx_q->netdev); nq = netdev_get_tx_queue(tx_q->netdev, tx_q->idx); - dont_wake = np->state != __IDPF_VPORT_UP || + dont_wake = !test_bit(IDPF_VPORT_UP, np->state) || !netif_carrier_ok(tx_q->netdev); __netif_txq_completed_wake(nq, ss.packets, ss.bytes, IDPF_DESC_UNUSED(tx_q), IDPF_TX_WAKE_THRESH, diff --git a/drivers/net/ethernet/intel/idpf/idpf_txrx.c b/drivers/net/ethernet/intel/idpf/idpf_txrx.c index 828f7c444d30..1993a3b0da59 100644 --- a/drivers/net/ethernet/intel/idpf/idpf_txrx.c +++ b/drivers/net/ethernet/intel/idpf/idpf_txrx.c @@ -2275,7 +2275,7 @@ static bool idpf_tx_clean_complq(struct idpf_compl_queue *complq, int budget, /* Update BQL */ nq = netdev_get_tx_queue(tx_q->netdev, tx_q->idx); - dont_wake = !complq_ok || np->state != __IDPF_VPORT_UP || + dont_wake = !complq_ok || !test_bit(IDPF_VPORT_UP, np->state) || !netif_carrier_ok(tx_q->netdev); /* Check if the TXQ needs to and can be restarted */ __netif_txq_completed_wake(nq, tx_q->cleaned_pkts, tx_q->cleaned_bytes, diff --git a/drivers/net/ethernet/intel/idpf/idpf_virtchnl.c b/drivers/net/ethernet/intel/idpf/idpf_virtchnl.c index cbb5fa30f5a0..44cd4b466c48 100644 --- a/drivers/net/ethernet/intel/idpf/idpf_virtchnl.c +++ b/drivers/net/ethernet/intel/idpf/idpf_virtchnl.c @@ -68,7 +68,7 @@ static void idpf_handle_event_link(struct idpf_adapter *adapter, vport->link_up = v2e->link_status; - if (np->state != __IDPF_VPORT_UP) + if (!test_bit(IDPF_VPORT_UP, np->state)) return; if (vport->link_up) { @@ -2755,7 +2755,7 @@ int idpf_send_get_stats_msg(struct idpf_vport *vport) /* Don't send get_stats message if the link is down */ - if (np->state <= __IDPF_VPORT_DOWN) + if (!test_bit(IDPF_VPORT_UP, np->state)) return 0; stats_msg.vport_id = cpu_to_le32(vport->vport_id); diff --git a/drivers/net/ethernet/intel/idpf/xdp.c b/drivers/net/ethernet/intel/idpf/xdp.c index 21ce25b0567f..958d16f87424 100644 --- a/drivers/net/ethernet/intel/idpf/xdp.c +++ b/drivers/net/ethernet/intel/idpf/xdp.c @@ -418,7 +418,7 @@ static int idpf_xdp_setup_prog(struct idpf_vport *vport, if (test_bit(IDPF_REMOVE_IN_PROG, vport->adapter->flags) || !test_bit(IDPF_VPORT_REG_NETDEV, cfg->flags) || !!vport->xdp_prog == !!prog) { - if (np->state == __IDPF_VPORT_UP) + if (test_bit(IDPF_VPORT_UP, np->state)) idpf_xdp_copy_prog_to_rqs(vport, prog); old = xchg(&vport->xdp_prog, prog); From 954ba97cca1614bdc6d5dbd37fa357da9febb59d Mon Sep 17 00:00:00 2001 From: Tony Nguyen Date: Tue, 25 Nov 2025 14:36:25 -0800 Subject: [PATCH 1244/1381] e1000e: Remove unneeded checks The caller, ethtool_set_eeprom(), already performs the same checks so these are unnecessary in the driver. This reverts commit 90fb7db49c6d ("e1000e: fix heap overflow in e1000_set_eeprom"), however, corrections for RCT have been kept. Link: https://lore.kernel.org/all/db92fcc8-114d-4e85-9d15-7860545bc65e@suse.de/ Signed-off-by: Tony Nguyen Link: https://patch.msgid.link/20251125223632.1857532-7-anthony.l.nguyen@intel.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/intel/e1000e/ethtool.c | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) diff --git a/drivers/net/ethernet/intel/e1000e/ethtool.c b/drivers/net/ethernet/intel/e1000e/ethtool.c index cee57a2149ab..7b1ac90b3de4 100644 --- a/drivers/net/ethernet/intel/e1000e/ethtool.c +++ b/drivers/net/ethernet/intel/e1000e/ethtool.c @@ -551,9 +551,9 @@ static int e1000_set_eeprom(struct net_device *netdev, { struct e1000_adapter *adapter = netdev_priv(netdev); struct e1000_hw *hw = &adapter->hw; - size_t total_len, max_len; u16 *eeprom_buff; int ret_val = 0; + size_t max_len; int first_word; int last_word; void *ptr; @@ -571,10 +571,6 @@ static int e1000_set_eeprom(struct net_device *netdev, max_len = hw->nvm.word_size * 2; - if (check_add_overflow(eeprom->offset, eeprom->len, &total_len) || - total_len > max_len) - return -EFBIG; - first_word = eeprom->offset >> 1; last_word = (eeprom->offset + eeprom->len - 1) >> 1; eeprom_buff = kmalloc(max_len, GFP_KERNEL); From 5849b56addbfc060a012c7326d3bac279c303f43 Mon Sep 17 00:00:00 2001 From: Alok Tiwari Date: Tue, 25 Nov 2025 14:36:26 -0800 Subject: [PATCH 1245/1381] ixgbe: avoid redundant call to ixgbe_non_sfp_link_config() ixgbe_non_sfp_link_config() is called twice in ixgbe_open() once to assign its return value to err and again in the conditional check. This patch uses the stored err value instead of calling the function a second time. This avoids redundant work and ensures consistent error reporting. Also fix a small typo in the ixgbe_remove() comment: "The could be caused" -> "This could be caused". Signed-off-by: Alok Tiwari Reviewed-by: Jedrzej Jagielski Reviewed-by: Paul Menzel Tested-by: Rinitha S (A Contingent worker at Intel) Signed-off-by: Tony Nguyen Link: https://patch.msgid.link/20251125223632.1857532-8-anthony.l.nguyen@intel.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c index 3190ce7e44c7..4af3b3e71ff1 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c @@ -7449,7 +7449,7 @@ int ixgbe_open(struct net_device *netdev) adapter->hw.link.link_info.link_cfg_err); err = ixgbe_non_sfp_link_config(&adapter->hw); - if (ixgbe_non_sfp_link_config(&adapter->hw)) + if (err) e_dev_err("Link setup failed, err %d.\n", err); } @@ -12046,7 +12046,7 @@ static int ixgbe_probe(struct pci_dev *pdev, const struct pci_device_id *ent) * @pdev: PCI device information struct * * ixgbe_remove is called by the PCI subsystem to alert the driver - * that it should release a PCI device. The could be caused by a + * that it should release a PCI device. This could be caused by a * Hot-Plug event, or because the driver is going to be removed from * memory. **/ From d89a5c27e4f395a0d93652f01c959ce4b1ed42b7 Mon Sep 17 00:00:00 2001 From: Alok Tiwari Date: Tue, 25 Nov 2025 14:36:27 -0800 Subject: [PATCH 1246/1381] idpf: use desc_ring when checking completion queue DMA allocation idpf_compl_queue uses a union for comp, comp_4b, and desc_ring. The release path should check complq->desc_ring to determine whether the DMA descriptor ring is allocated. The current check against comp works but is leftover from a previous commit and is misleading in this context. Switching the check to desc_ring improves readability and more directly reflects the intended meaning, since desc_ring is the field representing the allocated DMA-backed descriptor ring. No functional change. Signed-off-by: Alok Tiwari Reviewed-by: Aleksandr Loktionov Signed-off-by: Tony Nguyen Link: https://patch.msgid.link/20251125223632.1857532-9-anthony.l.nguyen@intel.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/intel/idpf/idpf_txrx.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/intel/idpf/idpf_txrx.c b/drivers/net/ethernet/intel/idpf/idpf_txrx.c index 1993a3b0da59..e2b6b9e26102 100644 --- a/drivers/net/ethernet/intel/idpf/idpf_txrx.c +++ b/drivers/net/ethernet/intel/idpf/idpf_txrx.c @@ -134,7 +134,7 @@ static void idpf_compl_desc_rel(struct idpf_compl_queue *complq) { idpf_xsk_clear_queue(complq, VIRTCHNL2_QUEUE_TYPE_TX_COMPLETION); - if (!complq->comp) + if (!complq->desc_ring) return; dma_free_coherent(complq->netdev->dev.parent, complq->size, From 79bb84758f2c6dda7133c5ac391acd61ae4326fd Mon Sep 17 00:00:00 2001 From: Alok Tiwari Date: Tue, 25 Nov 2025 14:36:28 -0800 Subject: [PATCH 1247/1381] idpf: correct queue index in Rx allocation error messages The error messages in idpf_rx_desc_alloc_all() used the group index i when reporting memory allocation failures for individual Rx and Rx buffer queues. This is incorrect. Update the messages to use the correct queue index j and include the queue group index i for clearer identification of the affected Rx and Rx buffer queues. Signed-off-by: Alok Tiwari Reviewed-by: Simon Horman Reviewed-by: Aleksandr Loktionov Signed-off-by: Tony Nguyen Link: https://patch.msgid.link/20251125223632.1857532-10-anthony.l.nguyen@intel.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/intel/idpf/idpf_txrx.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/intel/idpf/idpf_txrx.c b/drivers/net/ethernet/intel/idpf/idpf_txrx.c index e2b6b9e26102..1d91c56f7469 100644 --- a/drivers/net/ethernet/intel/idpf/idpf_txrx.c +++ b/drivers/net/ethernet/intel/idpf/idpf_txrx.c @@ -922,8 +922,8 @@ static int idpf_rx_desc_alloc_all(struct idpf_vport *vport) err = idpf_rx_desc_alloc(vport, q); if (err) { pci_err(vport->adapter->pdev, - "Memory allocation for Rx Queue %u failed\n", - i); + "Memory allocation for Rx queue %u from queue group %u failed\n", + j, i); goto err_out; } } @@ -939,8 +939,8 @@ static int idpf_rx_desc_alloc_all(struct idpf_vport *vport) err = idpf_bufq_desc_alloc(vport, q); if (err) { pci_err(vport->adapter->pdev, - "Memory allocation for Rx Buffer Queue %u failed\n", - i); + "Memory allocation for Rx Buffer Queue %u from queue group %u failed\n", + j, i); goto err_out; } } From 1105a7a12051c24e942b382b6e4d895d267dda1e Mon Sep 17 00:00:00 2001 From: Alok Tiwari Date: Tue, 25 Nov 2025 14:36:29 -0800 Subject: [PATCH 1248/1381] ice: fix comment typo and correct module format string - Fix a typo in the ice_fdir_has_frag() kernel-doc comment ("is" -> "if") - Correct the NVM erase error message format string from "0x02%x" to "0x%02x" so the module value is printed correctly. Signed-off-by: Alok Tiwari Reviewed-by: Aleksandr Loktionov Reviewed-by: Paul Menzel Signed-off-by: Tony Nguyen Link: https://patch.msgid.link/20251125223632.1857532-11-anthony.l.nguyen@intel.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/intel/ice/ice_fdir.c | 2 +- drivers/net/ethernet/intel/ice/ice_fw_update.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/intel/ice/ice_fdir.c b/drivers/net/ethernet/intel/ice/ice_fdir.c index 26b357c0ae15..b29fbdec9442 100644 --- a/drivers/net/ethernet/intel/ice/ice_fdir.c +++ b/drivers/net/ethernet/intel/ice/ice_fdir.c @@ -1121,7 +1121,7 @@ ice_fdir_get_gen_prgm_pkt(struct ice_hw *hw, struct ice_fdir_fltr *input, * ice_fdir_has_frag - does flow type have 2 ptypes * @flow: flow ptype * - * returns true is there is a fragment packet for this ptype + * Return: true if there is a fragment packet for this ptype */ bool ice_fdir_has_frag(enum ice_fltr_ptype flow) { diff --git a/drivers/net/ethernet/intel/ice/ice_fw_update.c b/drivers/net/ethernet/intel/ice/ice_fw_update.c index d86db081579f..973a13d3d92a 100644 --- a/drivers/net/ethernet/intel/ice/ice_fw_update.c +++ b/drivers/net/ethernet/intel/ice/ice_fw_update.c @@ -534,7 +534,7 @@ ice_erase_nvm_module(struct ice_pf *pf, u16 module, const char *component, } if (completion_retval) { - dev_err(dev, "Firmware failed to erase %s (module 0x02%x), aq_err %s\n", + dev_err(dev, "Firmware failed to erase %s (module 0x%02x), aq_err %s\n", component, module, libie_aq_str((enum libie_aq_err)completion_retval)); NL_SET_ERR_MSG_MOD(extack, "Firmware failed to erase flash"); From 57bb13d7eb5051563d1e29f930a72079d95ea28c Mon Sep 17 00:00:00 2001 From: Alok Tiwari Date: Tue, 25 Nov 2025 14:36:30 -0800 Subject: [PATCH 1249/1381] iavf: clarify VLAN add/delete log messages and lower log level The current dev_warn messages for too many VLAN changes are confusing and one place incorrectly references "add" instead of "delete" VLANs due to copy-paste errors. - Use dev_info instead of dev_warn to lower the log level. - Rephrase the message to: "virtchnl: Too many VLAN [add|delete] ([v1|v2]) requests; splitting into multiple messages to PF\n". Suggested-by: Przemek Kitszel Signed-off-by: Alok Tiwari Reviewed-by: Przemek Kitszel Reviewed-by: Aleksandr Loktionov Signed-off-by: Tony Nguyen Link: https://patch.msgid.link/20251125223632.1857532-12-anthony.l.nguyen@intel.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/intel/iavf/iavf_virtchnl.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/intel/iavf/iavf_virtchnl.c b/drivers/net/ethernet/intel/iavf/iavf_virtchnl.c index 34a422a4a29c..88156082a41d 100644 --- a/drivers/net/ethernet/intel/iavf/iavf_virtchnl.c +++ b/drivers/net/ethernet/intel/iavf/iavf_virtchnl.c @@ -793,7 +793,8 @@ void iavf_add_vlans(struct iavf_adapter *adapter) len = virtchnl_struct_size(vvfl, vlan_id, count); if (len > IAVF_MAX_AQ_BUF_SIZE) { - dev_warn(&adapter->pdev->dev, "Too many add VLAN changes in one request\n"); + dev_info(&adapter->pdev->dev, + "virtchnl: Too many VLAN add (v1) requests; splitting into multiple messages to PF\n"); while (len > IAVF_MAX_AQ_BUF_SIZE) len = virtchnl_struct_size(vvfl, vlan_id, --count); @@ -838,7 +839,8 @@ void iavf_add_vlans(struct iavf_adapter *adapter) len = virtchnl_struct_size(vvfl_v2, filters, count); if (len > IAVF_MAX_AQ_BUF_SIZE) { - dev_warn(&adapter->pdev->dev, "Too many add VLAN changes in one request\n"); + dev_info(&adapter->pdev->dev, + "virtchnl: Too many VLAN add (v2) requests; splitting into multiple messages to PF\n"); while (len > IAVF_MAX_AQ_BUF_SIZE) len = virtchnl_struct_size(vvfl_v2, filters, --count); @@ -941,7 +943,8 @@ void iavf_del_vlans(struct iavf_adapter *adapter) len = virtchnl_struct_size(vvfl, vlan_id, count); if (len > IAVF_MAX_AQ_BUF_SIZE) { - dev_warn(&adapter->pdev->dev, "Too many delete VLAN changes in one request\n"); + dev_info(&adapter->pdev->dev, + "virtchnl: Too many VLAN delete (v1) requests; splitting into multiple messages to PF\n"); while (len > IAVF_MAX_AQ_BUF_SIZE) len = virtchnl_struct_size(vvfl, vlan_id, --count); @@ -987,7 +990,8 @@ void iavf_del_vlans(struct iavf_adapter *adapter) len = virtchnl_struct_size(vvfl_v2, filters, count); if (len > IAVF_MAX_AQ_BUF_SIZE) { - dev_warn(&adapter->pdev->dev, "Too many add VLAN changes in one request\n"); + dev_info(&adapter->pdev->dev, + "virtchnl: Too many VLAN delete (v2) requests; splitting into multiple messages to PF\n"); while (len > IAVF_MAX_AQ_BUF_SIZE) len = virtchnl_struct_size(vvfl_v2, filters, --count); From caa343e9a4a171db045c8e734d68cb10ad0dd65c Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Wed, 26 Nov 2025 13:56:42 -0800 Subject: [PATCH 1250/1381] bnxt_en: Enhance TX pri counters The priority packet and byte counters in ethtool -S are returned by the driver based on the pri2cos mapping. The assumption is that each priority is mapped to one and only one hardware CoS queue. In a special RoCE configuration, the FW uses combined CoS queue 0 and CoS queue 1 for the priority mapped to CoS queue 0. In this special case, we need to add the CoS queue 0 and CoS queue 1 counters for the priority packet and byte counters. Reviewed-by: Andy Gospodarek Signed-off-by: Michael Chan Link: https://patch.msgid.link/20251126215648.1885936-2-michael.chan@broadcom.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 5 +++++ drivers/net/ethernet/broadcom/bnxt/bnxt.h | 1 + drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c | 14 ++++++++++---- 3 files changed, 16 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 8f95f5f29e00..c289d1f35e16 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -8506,6 +8506,11 @@ static int bnxt_hwrm_func_qcfg(struct bnxt *bp) if (flags & FUNC_QCFG_RESP_FLAGS_ENABLE_RDMA_SRIOV) bp->fw_cap |= BNXT_FW_CAP_ENABLE_RDMA_SRIOV; + if (resp->roce_bidi_opt_mode & + FUNC_QCFG_RESP_ROCE_BIDI_OPT_MODE_DEDICATED) + bp->cos0_cos1_shared = 1; + else + bp->cos0_cos1_shared = 0; switch (resp->port_partition_type) { case FUNC_QCFG_RESP_PORT_PARTITION_TYPE_NPAR1_0: diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h index 3613a172483a..aedb059f4ce5 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h @@ -2424,6 +2424,7 @@ struct bnxt { u8 tc_to_qidx[BNXT_MAX_QUEUE]; u8 q_ids[BNXT_MAX_QUEUE]; u8 max_q; + u8 cos0_cos1_shared; u8 num_tc; u16 max_pfcwd_tmo_ms; diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c index 41686a6f84b5..baac639f9c94 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c @@ -688,16 +688,22 @@ static void bnxt_get_ethtool_stats(struct net_device *dev, buf[j] = *(rx_port_stats_ext + n); } for (i = 0; i < 8; i++, j++) { - long n = bnxt_tx_bytes_pri_arr[i].base_off + - bp->pri2cos_idx[i]; + u8 cos_idx = bp->pri2cos_idx[i]; + long n; + n = bnxt_tx_bytes_pri_arr[i].base_off + cos_idx; buf[j] = *(tx_port_stats_ext + n); + if (bp->cos0_cos1_shared && !cos_idx) + buf[j] += *(tx_port_stats_ext + n + 1); } for (i = 0; i < 8; i++, j++) { - long n = bnxt_tx_pkts_pri_arr[i].base_off + - bp->pri2cos_idx[i]; + u8 cos_idx = bp->pri2cos_idx[i]; + long n; + n = bnxt_tx_pkts_pri_arr[i].base_off + cos_idx; buf[j] = *(tx_port_stats_ext + n); + if (bp->cos0_cos1_shared && !cos_idx) + buf[j] += *(tx_port_stats_ext + n + 1); } } } From f3d88fe6354fcbc38d872029564e2b71f533ca9e Mon Sep 17 00:00:00 2001 From: Gautam R A Date: Wed, 26 Nov 2025 13:56:43 -0800 Subject: [PATCH 1251/1381] bnxt_en: Enhance log message in bnxt_get_module_status() Rturn early with -EOPNOTSUPP and an extack message if the PHY type is BaseT since module status is not available for BaseT. Reviewed-by: Somnath Kotur Signed-off-by: Gautam R A Signed-off-by: Michael Chan Link: https://patch.msgid.link/20251126215648.1885936-3-michael.chan@broadcom.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c index baac639f9c94..efb9bf20e66b 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c @@ -4623,6 +4623,11 @@ static int bnxt_get_module_status(struct bnxt *bp, struct netlink_ext_ack *extac PORT_PHY_QCFG_RESP_MODULE_STATUS_WARNINGMSG) return 0; + if (bp->link_info.phy_type == PORT_PHY_QCFG_RESP_PHY_TYPE_BASET || + bp->link_info.phy_type == PORT_PHY_QCFG_RESP_PHY_TYPE_BASETE){ + NL_SET_ERR_MSG_MOD(extack, "Operation not supported as PHY type is Base-T"); + return -EOPNOTSUPP; + } switch (bp->link_info.module_status) { case PORT_PHY_QCFG_RESP_MODULE_STATUS_PWRDOWN: NL_SET_ERR_MSG_MOD(extack, "Transceiver module is powering down"); From bba2a0577e3c20a5393f9906fe066d500657f1c6 Mon Sep 17 00:00:00 2001 From: Kalesh AP Date: Wed, 26 Nov 2025 13:56:44 -0800 Subject: [PATCH 1252/1381] bnxt_en: Remove the redundant BNXT_EN_FLAG_MSIX_REQUESTED flag MSIX is always requested when the RoCE driver calls bnxt_register_dev(). We already check bnxt_ulp_registered(), so checking the flag is redundant. It was a left-over flag after converting to auxbus, so remove it. Reviewed-by: Pavan Chebbi Reviewed-by: Somnath Kotur Signed-off-by: Kalesh AP Signed-off-by: Michael Chan Link: https://patch.msgid.link/20251126215648.1885936-4-michael.chan@broadcom.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c | 7 ++----- drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.h | 1 - 2 files changed, 2 insertions(+), 6 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c index f8c2c72b382d..927971c362f1 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c @@ -142,7 +142,6 @@ int bnxt_register_dev(struct bnxt_en_dev *edev, edev->ulp_tbl->msix_requested = bnxt_get_ulp_msix_num(bp); bnxt_fill_msix_vecs(bp, bp->edev->msix_entries); - edev->flags |= BNXT_EN_FLAG_MSIX_REQUESTED; exit: mutex_unlock(&edev->en_dev_lock); netdev_unlock(dev); @@ -159,8 +158,6 @@ void bnxt_unregister_dev(struct bnxt_en_dev *edev) ulp = edev->ulp_tbl; netdev_lock(dev); mutex_lock(&edev->en_dev_lock); - if (ulp->msix_requested) - edev->flags &= ~BNXT_EN_FLAG_MSIX_REQUESTED; edev->ulp_tbl->msix_requested = 0; if (ulp->max_async_event_id) @@ -298,7 +295,7 @@ void bnxt_ulp_irq_stop(struct bnxt *bp) struct bnxt_ulp_ops *ops; bool reset = false; - if (!edev || !(edev->flags & BNXT_EN_FLAG_MSIX_REQUESTED)) + if (!edev) return; if (bnxt_ulp_registered(bp->edev)) { @@ -321,7 +318,7 @@ void bnxt_ulp_irq_restart(struct bnxt *bp, int err) struct bnxt_en_dev *edev = bp->edev; struct bnxt_ulp_ops *ops; - if (!edev || !(edev->flags & BNXT_EN_FLAG_MSIX_REQUESTED)) + if (!edev) return; if (bnxt_ulp_registered(bp->edev)) { diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.h index 7b9dd8ebe4bc..3c5b8a53f715 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.h @@ -58,7 +58,6 @@ struct bnxt_en_dev { #define BNXT_EN_FLAG_ROCEV2_CAP 0x2 #define BNXT_EN_FLAG_ROCE_CAP (BNXT_EN_FLAG_ROCEV1_CAP | \ BNXT_EN_FLAG_ROCEV2_CAP) - #define BNXT_EN_FLAG_MSIX_REQUESTED 0x4 #define BNXT_EN_FLAG_ULP_STOPPED 0x8 #define BNXT_EN_FLAG_VF 0x10 #define BNXT_EN_VF(edev) ((edev)->flags & BNXT_EN_FLAG_VF) From b1e7f9566f322b53e81f7bfb6b4361e52e44b287 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Wed, 26 Nov 2025 13:56:45 -0800 Subject: [PATCH 1253/1381] bnxt_en: Add CQ ring dump to bnxt_dump_cp_sw_state() On newer chips that use NQs and CQs, add the CQ ring dump to bnxt_dump_cp_sw_state() to make it more complete. Signed-off-by: Michael Chan Link: https://patch.msgid.link/20251126215648.1885936-5-michael.chan@broadcom.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index c289d1f35e16..a267bb49ed88 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -14025,11 +14025,19 @@ static void bnxt_dump_rx_sw_state(struct bnxt_napi *bnapi) static void bnxt_dump_cp_sw_state(struct bnxt_napi *bnapi) { - struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring; - int i = bnapi->index; + struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring, *cpr2; + int i = bnapi->index, j; netdev_info(bnapi->bp->dev, "[%d]: cp{fw_ring: %d raw_cons: %x}\n", i, cpr->cp_ring_struct.fw_ring_id, cpr->cp_raw_cons); + for (j = 0; j < cpr->cp_ring_count; j++) { + cpr2 = &cpr->cp_ring_arr[j]; + if (!cpr2->bnapi) + continue; + netdev_info(bnapi->bp->dev, "[%d.%d]: cp{fw_ring: %d raw_cons: %x}\n", + i, j, cpr2->cp_ring_struct.fw_ring_id, + cpr2->cp_raw_cons); + } } static void bnxt_dbg_dump_states(struct bnxt *bp) From 30f253f8d9a01d532fdb7ec6c8a9d4c15fe29241 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Wed, 26 Nov 2025 13:56:46 -0800 Subject: [PATCH 1254/1381] bnxt_en: Do not set EOP on RX AGG BDs on 5760X chips With End-of-Packet padding (EOP) set, the chip will disable Relaxed Ordering (RO) of TPA data packets. A TPA segment with EOP set will be padded to the next cache boundary and can potentially overwrite the beginning bytes of the next TPA segment when RO is enabled on 5760X. To prevent that, the chip disables RO for TPA when EOP is set. To take advantge of RO and higher performance, do not set EOP on 5760X chips when TPA is enabled. Define a proper RX_BD_FLAGS_AGG_EOP constant to make it clear that we are setting EOP. Reviewed-by: Andy Gospodarek Reviewed-by: Somnath Kotur Signed-off-by: Michael Chan Link: https://patch.msgid.link/20251126215648.1885936-6-michael.chan@broadcom.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 9 ++++++++- drivers/net/ethernet/broadcom/bnxt/bnxt.h | 1 + 2 files changed, 9 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index a267bb49ed88..d226a4117958 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -4479,7 +4479,14 @@ static void bnxt_init_one_rx_agg_ring_rxbd(struct bnxt *bp, ring->fw_ring_id = INVALID_HW_RING_ID; if ((bp->flags & BNXT_FLAG_AGG_RINGS)) { type = ((u32)BNXT_RX_PAGE_SIZE << RX_BD_LEN_SHIFT) | - RX_BD_TYPE_RX_AGG_BD | RX_BD_FLAGS_SOP; + RX_BD_TYPE_RX_AGG_BD; + + /* On P7, setting EOP will cause the chip to disable + * Relaxed Ordering (RO) for TPA data. Disable EOP for + * potentially higher performance with RO. + */ + if (BNXT_CHIP_P5_AND_MINUS(bp) || !(bp->flags & BNXT_FLAG_TPA)) + type |= RX_BD_FLAGS_AGG_EOP; bnxt_init_rxbd_pages(ring, type); } diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h index aedb059f4ce5..bb12cebd40e1 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h @@ -131,6 +131,7 @@ struct rx_bd { #define RX_BD_TYPE_48B_BD_SIZE (2 << 4) #define RX_BD_TYPE_64B_BD_SIZE (3 << 4) #define RX_BD_FLAGS_SOP (1 << 6) + #define RX_BD_FLAGS_AGG_EOP (1 << 6) #define RX_BD_FLAGS_EOP (1 << 7) #define RX_BD_FLAGS_BUFFERS (3 << 8) #define RX_BD_FLAGS_1_BUFFER_PACKET (0 << 8) From 72405298e92b590160ea21558b8105d18c107f3c Mon Sep 17 00:00:00 2001 From: Rob Miller Date: Wed, 26 Nov 2025 13:56:47 -0800 Subject: [PATCH 1255/1381] bnxt_en: Add Virtual Admin Link State Support for VFs The firmware can now cache the virtual link admin state (auto/on/off) of all VFs and as such, the PF driver no longer has to intercept the VF driver's port_phy_qcfg() call and then provide the link admin state. If the FW does not have this capability, fall back to the existing interception method. The initial default link admin state (auto) is also set initially when the VFs are created. Reviewed-by: Pavan Chebbi Reviewed-by: Kalesh AP Reviewed-by: Mohammad Shuab Siddique Signed-off-by: Rob Miller Signed-off-by: Michael Chan Link: https://patch.msgid.link/20251126215648.1885936-7-michael.chan@broadcom.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 6 ++ drivers/net/ethernet/broadcom/bnxt/bnxt.h | 1 + .../net/ethernet/broadcom/bnxt/bnxt_sriov.c | 55 +++++++++++++++++-- 3 files changed, 57 insertions(+), 5 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index d226a4117958..d17d0ea89c36 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -5695,6 +5695,10 @@ int bnxt_hwrm_func_drv_rgtr(struct bnxt *bp, unsigned long *bmap, int bmap_size, u16 cmd = bnxt_vf_req_snif[i]; unsigned int bit, idx; + if ((bp->fw_cap & BNXT_FW_CAP_LINK_ADMIN) && + cmd == HWRM_PORT_PHY_QCFG) + continue; + idx = cmd / 32; bit = cmd % 32; data[idx] |= 1 << bit; @@ -9665,6 +9669,8 @@ static int __bnxt_hwrm_func_qcaps(struct bnxt *bp) bp->flags |= BNXT_FLAG_ROCEV1_CAP; if (flags & FUNC_QCAPS_RESP_FLAGS_ROCE_V2_SUPPORTED) bp->flags |= BNXT_FLAG_ROCEV2_CAP; + if (flags & FUNC_QCAPS_RESP_FLAGS_LINK_ADMIN_STATUS_SUPPORTED) + bp->fw_cap |= BNXT_FW_CAP_LINK_ADMIN; if (flags & FUNC_QCAPS_RESP_FLAGS_PCIE_STATS_SUPPORTED) bp->fw_cap |= BNXT_FW_CAP_PCIE_STATS_SUPPORTED; if (flags & FUNC_QCAPS_RESP_FLAGS_HOT_RESET_CAPABLE) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h index bb12cebd40e1..f5f07a7e6b29 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h @@ -2484,6 +2484,7 @@ struct bnxt { #define BNXT_FW_CAP_ROCE_VF_RESC_MGMT_SUPPORTED BIT_ULL(6) #define BNXT_FW_CAP_KONG_MB_CHNL BIT_ULL(7) #define BNXT_FW_CAP_ROCE_VF_DYN_ALLOC_SUPPORT BIT_ULL(8) + #define BNXT_FW_CAP_LINK_ADMIN BIT_ULL(9) #define BNXT_FW_CAP_OVS_64BIT_HANDLE BIT_ULL(10) #define BNXT_FW_CAP_TRUSTED_VF BIT_ULL(11) #define BNXT_FW_CAP_ERROR_RECOVERY BIT_ULL(13) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c index 80fed2c07b9e..be7deb9cc410 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c @@ -332,6 +332,38 @@ int bnxt_set_vf_bw(struct net_device *dev, int vf_id, int min_tx_rate, return rc; } +static int bnxt_set_vf_link_admin_state(struct bnxt *bp, int vf_id) +{ + struct hwrm_func_cfg_input *req; + struct bnxt_vf_info *vf; + int rc; + + if (!(bp->fw_cap & BNXT_FW_CAP_LINK_ADMIN)) + return 0; + + vf = &bp->pf.vf[vf_id]; + + rc = bnxt_hwrm_func_cfg_short_req_init(bp, &req); + if (rc) + return rc; + + req->fid = cpu_to_le16(vf->fw_fid); + switch (vf->flags & (BNXT_VF_LINK_FORCED | BNXT_VF_LINK_UP)) { + case BNXT_VF_LINK_FORCED: + req->options = + FUNC_CFG_REQ_OPTIONS_LINK_ADMIN_STATE_FORCED_DOWN; + break; + case (BNXT_VF_LINK_FORCED | BNXT_VF_LINK_UP): + req->options = FUNC_CFG_REQ_OPTIONS_LINK_ADMIN_STATE_FORCED_UP; + break; + default: + req->options = FUNC_CFG_REQ_OPTIONS_LINK_ADMIN_STATE_AUTO; + break; + } + req->enables = cpu_to_le32(FUNC_CFG_REQ_ENABLES_ADMIN_LINK_STATE); + return hwrm_req_send(bp, req); +} + int bnxt_set_vf_link_state(struct net_device *dev, int vf_id, int link) { struct bnxt *bp = netdev_priv(dev); @@ -357,10 +389,11 @@ int bnxt_set_vf_link_state(struct net_device *dev, int vf_id, int link) break; default: netdev_err(bp->dev, "Invalid link option\n"); - rc = -EINVAL; - break; + return -EINVAL; } - if (vf->flags & (BNXT_VF_LINK_UP | BNXT_VF_LINK_FORCED)) + if (bp->fw_cap & BNXT_FW_CAP_LINK_ADMIN) + rc = bnxt_set_vf_link_admin_state(bp, vf_id); + else if (vf->flags & (BNXT_VF_LINK_UP | BNXT_VF_LINK_FORCED)) rc = bnxt_hwrm_fwd_async_event_cmpl(bp, vf, ASYNC_EVENT_CMPL_EVENT_ID_LINK_STATUS_CHANGE); return rc; @@ -666,15 +699,21 @@ static int bnxt_hwrm_func_vf_resc_cfg(struct bnxt *bp, int num_vfs, bool reset) hwrm_req_hold(bp, req); for (i = 0; i < num_vfs; i++) { + struct bnxt_vf_info *vf = &pf->vf[i]; + + vf->fw_fid = pf->first_vf_id + i; + rc = bnxt_set_vf_link_admin_state(bp, i); + if (rc) + break; + if (reset) __bnxt_set_vf_params(bp, i); - req->vf_id = cpu_to_le16(pf->first_vf_id + i); + req->vf_id = cpu_to_le16(vf->fw_fid); rc = hwrm_req_send(bp, req); if (rc) break; pf->active_vfs = i + 1; - pf->vf[i].fw_fid = pf->first_vf_id + i; } if (pf->active_vfs) { @@ -741,6 +780,12 @@ static int bnxt_hwrm_func_cfg(struct bnxt *bp, int num_vfs) FUNC_CFG_REQ_ENABLES_NUM_VNICS | FUNC_CFG_REQ_ENABLES_NUM_HW_RING_GRPS); + if (bp->fw_cap & BNXT_FW_CAP_LINK_ADMIN) { + req->options = FUNC_CFG_REQ_OPTIONS_LINK_ADMIN_STATE_AUTO; + req->enables |= + cpu_to_le32(FUNC_CFG_REQ_ENABLES_ADMIN_LINK_STATE); + } + mtu = bp->dev->mtu + VLAN_ETH_HLEN; req->mru = cpu_to_le16(mtu); req->admin_mtu = cpu_to_le16(mtu); From c940be4c7c75684799d5bff495ac9c48ca19f183 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Wed, 26 Nov 2025 15:07:05 +0100 Subject: [PATCH 1256/1381] net: Remove KMSG_COMPONENT macro The KMSG_COMPONENT macro is a leftover of the s390 specific "kernel message catalog" from 2008 [1] which never made it upstream. The macro was added to s390 code to allow for an out-of-tree patch which used this to generate unique message ids. Also this out-of-tree patch doesn't exist anymore. The pattern of how the KMSG_COMPONENT macro is used can also be found at some non s390 specific code, for whatever reasons. Besides adding an indirection it is unused. Remove the macro in order to get rid of a pointless indirection. Replace all users with the string it defines. In all cases this leads to a simple replacement like this: - #define KMSG_COMPONENT "af_iucv" - #define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + #define pr_fmt(fmt) "af_iucv: " fmt [1] https://lwn.net/Articles/292650/ Signed-off-by: Heiko Carstens Acked-by: Alexandra Winter Acked-by: Julian Anastasov Acked-by: Sidraya Jayagond Link: https://patch.msgid.link/20251126140705.1944278-1-hca@linux.ibm.com Signed-off-by: Jakub Kicinski --- net/iucv/af_iucv.c | 3 +-- net/iucv/iucv.c | 3 +-- net/netfilter/ipvs/ip_vs_app.c | 3 +-- net/netfilter/ipvs/ip_vs_conn.c | 3 +-- net/netfilter/ipvs/ip_vs_core.c | 3 +-- net/netfilter/ipvs/ip_vs_ctl.c | 3 +-- net/netfilter/ipvs/ip_vs_dh.c | 3 +-- net/netfilter/ipvs/ip_vs_est.c | 3 +-- net/netfilter/ipvs/ip_vs_fo.c | 3 +-- net/netfilter/ipvs/ip_vs_ftp.c | 3 +-- net/netfilter/ipvs/ip_vs_lblc.c | 3 +-- net/netfilter/ipvs/ip_vs_lblcr.c | 3 +-- net/netfilter/ipvs/ip_vs_lc.c | 3 +-- net/netfilter/ipvs/ip_vs_mh.c | 3 +-- net/netfilter/ipvs/ip_vs_nfct.c | 3 +-- net/netfilter/ipvs/ip_vs_nq.c | 3 +-- net/netfilter/ipvs/ip_vs_ovf.c | 3 +-- net/netfilter/ipvs/ip_vs_pe.c | 3 +-- net/netfilter/ipvs/ip_vs_pe_sip.c | 3 +-- net/netfilter/ipvs/ip_vs_proto.c | 3 +-- net/netfilter/ipvs/ip_vs_proto_ah_esp.c | 3 +-- net/netfilter/ipvs/ip_vs_proto_tcp.c | 3 +-- net/netfilter/ipvs/ip_vs_proto_udp.c | 3 +-- net/netfilter/ipvs/ip_vs_rr.c | 3 +-- net/netfilter/ipvs/ip_vs_sched.c | 3 +-- net/netfilter/ipvs/ip_vs_sed.c | 3 +-- net/netfilter/ipvs/ip_vs_sh.c | 3 +-- net/netfilter/ipvs/ip_vs_sync.c | 3 +-- net/netfilter/ipvs/ip_vs_twos.c | 3 +-- net/netfilter/ipvs/ip_vs_wlc.c | 3 +-- net/netfilter/ipvs/ip_vs_wrr.c | 3 +-- net/netfilter/ipvs/ip_vs_xmit.c | 3 +-- net/smc/af_smc.c | 3 +-- 33 files changed, 33 insertions(+), 66 deletions(-) diff --git a/net/iucv/af_iucv.c b/net/iucv/af_iucv.c index a4f1df92417d..1e62fbc22cb7 100644 --- a/net/iucv/af_iucv.c +++ b/net/iucv/af_iucv.c @@ -10,8 +10,7 @@ * Ursula Braun */ -#define KMSG_COMPONENT "af_iucv" -#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt +#define pr_fmt(fmt) "af_iucv: " fmt #include #include diff --git a/net/iucv/iucv.c b/net/iucv/iucv.c index 008be0abe3a5..da2af413c89d 100644 --- a/net/iucv/iucv.c +++ b/net/iucv/iucv.c @@ -20,8 +20,7 @@ * CP Programming Service, IBM document # SC24-5760 */ -#define KMSG_COMPONENT "iucv" -#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt +#define pr_fmt(fmt) "iucv: " fmt #include #include diff --git a/net/netfilter/ipvs/ip_vs_app.c b/net/netfilter/ipvs/ip_vs_app.c index fdacbc3c15be..d54d7da58334 100644 --- a/net/netfilter/ipvs/ip_vs_app.c +++ b/net/netfilter/ipvs/ip_vs_app.c @@ -13,8 +13,7 @@ * Author: Juan Jose Ciarlante, */ -#define KMSG_COMPONENT "IPVS" -#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt +#define pr_fmt(fmt) "IPVS: " fmt #include #include diff --git a/net/netfilter/ipvs/ip_vs_conn.c b/net/netfilter/ipvs/ip_vs_conn.c index 37ebb0cb62b8..50cc492c7553 100644 --- a/net/netfilter/ipvs/ip_vs_conn.c +++ b/net/netfilter/ipvs/ip_vs_conn.c @@ -17,8 +17,7 @@ * Changes: */ -#define KMSG_COMPONENT "IPVS" -#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt +#define pr_fmt(fmt) "IPVS: " fmt #include #include diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c index 5ea7ab8bf4dc..90d56f92c0f6 100644 --- a/net/netfilter/ipvs/ip_vs_core.c +++ b/net/netfilter/ipvs/ip_vs_core.c @@ -19,8 +19,7 @@ * Harald Welte don't use nfcache */ -#define KMSG_COMPONENT "IPVS" -#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt +#define pr_fmt(fmt) "IPVS: " fmt #include #include diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index 4c8fa22be88a..068702894377 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -13,8 +13,7 @@ * Changes: */ -#define KMSG_COMPONENT "IPVS" -#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt +#define pr_fmt(fmt) "IPVS: " fmt #include #include diff --git a/net/netfilter/ipvs/ip_vs_dh.c b/net/netfilter/ipvs/ip_vs_dh.c index 75f4c231f4a0..bb7aca4601ff 100644 --- a/net/netfilter/ipvs/ip_vs_dh.c +++ b/net/netfilter/ipvs/ip_vs_dh.c @@ -30,8 +30,7 @@ * */ -#define KMSG_COMPONENT "IPVS" -#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt +#define pr_fmt(fmt) "IPVS: " fmt #include #include diff --git a/net/netfilter/ipvs/ip_vs_est.c b/net/netfilter/ipvs/ip_vs_est.c index 93a925f1ed9b..77f4f637ff67 100644 --- a/net/netfilter/ipvs/ip_vs_est.c +++ b/net/netfilter/ipvs/ip_vs_est.c @@ -12,8 +12,7 @@ * get_stats()) do the per cpu summing. */ -#define KMSG_COMPONENT "IPVS" -#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt +#define pr_fmt(fmt) "IPVS: " fmt #include #include diff --git a/net/netfilter/ipvs/ip_vs_fo.c b/net/netfilter/ipvs/ip_vs_fo.c index ab117e5bc34e..d657b47c6511 100644 --- a/net/netfilter/ipvs/ip_vs_fo.c +++ b/net/netfilter/ipvs/ip_vs_fo.c @@ -8,8 +8,7 @@ * Kenny Mathis : added initial functionality based on weight */ -#define KMSG_COMPONENT "IPVS" -#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt +#define pr_fmt(fmt) "IPVS: " fmt #include #include diff --git a/net/netfilter/ipvs/ip_vs_ftp.c b/net/netfilter/ipvs/ip_vs_ftp.c index 206c6700e200..b315c608fda4 100644 --- a/net/netfilter/ipvs/ip_vs_ftp.c +++ b/net/netfilter/ipvs/ip_vs_ftp.c @@ -16,8 +16,7 @@ * Author: Wouter Gadeyne */ -#define KMSG_COMPONENT "IPVS" -#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt +#define pr_fmt(fmt) "IPVS: " fmt #include #include diff --git a/net/netfilter/ipvs/ip_vs_lblc.c b/net/netfilter/ipvs/ip_vs_lblc.c index 156181a3bacd..e6c8ed0c92f6 100644 --- a/net/netfilter/ipvs/ip_vs_lblc.c +++ b/net/netfilter/ipvs/ip_vs_lblc.c @@ -34,8 +34,7 @@ * me to write this module. */ -#define KMSG_COMPONENT "IPVS" -#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt +#define pr_fmt(fmt) "IPVS: " fmt #include #include diff --git a/net/netfilter/ipvs/ip_vs_lblcr.c b/net/netfilter/ipvs/ip_vs_lblcr.c index a021e6aba3d7..a25cf7bb6185 100644 --- a/net/netfilter/ipvs/ip_vs_lblcr.c +++ b/net/netfilter/ipvs/ip_vs_lblcr.c @@ -32,8 +32,7 @@ * */ -#define KMSG_COMPONENT "IPVS" -#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt +#define pr_fmt(fmt) "IPVS: " fmt #include #include diff --git a/net/netfilter/ipvs/ip_vs_lc.c b/net/netfilter/ipvs/ip_vs_lc.c index c2764505e380..38cc38c5d8bb 100644 --- a/net/netfilter/ipvs/ip_vs_lc.c +++ b/net/netfilter/ipvs/ip_vs_lc.c @@ -9,8 +9,7 @@ * Wensong Zhang : added any dest with weight=0 is quiesced */ -#define KMSG_COMPONENT "IPVS" -#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt +#define pr_fmt(fmt) "IPVS: " fmt #include #include diff --git a/net/netfilter/ipvs/ip_vs_mh.c b/net/netfilter/ipvs/ip_vs_mh.c index e3d7f5c879ce..f61f54004c9e 100644 --- a/net/netfilter/ipvs/ip_vs_mh.c +++ b/net/netfilter/ipvs/ip_vs_mh.c @@ -17,8 +17,7 @@ * */ -#define KMSG_COMPONENT "IPVS" -#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt +#define pr_fmt(fmt) "IPVS: " fmt #include #include diff --git a/net/netfilter/ipvs/ip_vs_nfct.c b/net/netfilter/ipvs/ip_vs_nfct.c index 08adcb222986..81974f69e5bb 100644 --- a/net/netfilter/ipvs/ip_vs_nfct.c +++ b/net/netfilter/ipvs/ip_vs_nfct.c @@ -30,8 +30,7 @@ * PASV response can not be NAT-ed) but Active FTP should work */ -#define KMSG_COMPONENT "IPVS" -#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt +#define pr_fmt(fmt) "IPVS: " fmt #include #include diff --git a/net/netfilter/ipvs/ip_vs_nq.c b/net/netfilter/ipvs/ip_vs_nq.c index ed7f5c889b41..ada158c610ce 100644 --- a/net/netfilter/ipvs/ip_vs_nq.c +++ b/net/netfilter/ipvs/ip_vs_nq.c @@ -26,8 +26,7 @@ * */ -#define KMSG_COMPONENT "IPVS" -#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt +#define pr_fmt(fmt) "IPVS: " fmt #include #include diff --git a/net/netfilter/ipvs/ip_vs_ovf.c b/net/netfilter/ipvs/ip_vs_ovf.c index c7708b809700..c5c67df80a0b 100644 --- a/net/netfilter/ipvs/ip_vs_ovf.c +++ b/net/netfilter/ipvs/ip_vs_ovf.c @@ -12,8 +12,7 @@ * active connections */ -#define KMSG_COMPONENT "IPVS" -#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt +#define pr_fmt(fmt) "IPVS: " fmt #include #include diff --git a/net/netfilter/ipvs/ip_vs_pe.c b/net/netfilter/ipvs/ip_vs_pe.c index 166c669f0763..3035079ebd99 100644 --- a/net/netfilter/ipvs/ip_vs_pe.c +++ b/net/netfilter/ipvs/ip_vs_pe.c @@ -1,6 +1,5 @@ // SPDX-License-Identifier: GPL-2.0-only -#define KMSG_COMPONENT "IPVS" -#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt +#define pr_fmt(fmt) "IPVS: " fmt #include #include diff --git a/net/netfilter/ipvs/ip_vs_pe_sip.c b/net/netfilter/ipvs/ip_vs_pe_sip.c index e4ce1d9a63f9..85f31d71e29a 100644 --- a/net/netfilter/ipvs/ip_vs_pe_sip.c +++ b/net/netfilter/ipvs/ip_vs_pe_sip.c @@ -1,6 +1,5 @@ // SPDX-License-Identifier: GPL-2.0-only -#define KMSG_COMPONENT "IPVS" -#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt +#define pr_fmt(fmt) "IPVS: " fmt #include #include diff --git a/net/netfilter/ipvs/ip_vs_proto.c b/net/netfilter/ipvs/ip_vs_proto.c index a9fd1d3fc2cb..fd9dbca24c85 100644 --- a/net/netfilter/ipvs/ip_vs_proto.c +++ b/net/netfilter/ipvs/ip_vs_proto.c @@ -8,8 +8,7 @@ * Changes: */ -#define KMSG_COMPONENT "IPVS" -#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt +#define pr_fmt(fmt) "IPVS: " fmt #include #include diff --git a/net/netfilter/ipvs/ip_vs_proto_ah_esp.c b/net/netfilter/ipvs/ip_vs_proto_ah_esp.c index 89602c16f6b6..44e14acc187e 100644 --- a/net/netfilter/ipvs/ip_vs_proto_ah_esp.c +++ b/net/netfilter/ipvs/ip_vs_proto_ah_esp.c @@ -6,8 +6,7 @@ * Wensong Zhang */ -#define KMSG_COMPONENT "IPVS" -#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt +#define pr_fmt(fmt) "IPVS: " fmt #include #include diff --git a/net/netfilter/ipvs/ip_vs_proto_tcp.c b/net/netfilter/ipvs/ip_vs_proto_tcp.c index 7da51390cea6..f68a1533ee45 100644 --- a/net/netfilter/ipvs/ip_vs_proto_tcp.c +++ b/net/netfilter/ipvs/ip_vs_proto_tcp.c @@ -13,8 +13,7 @@ * protocol ip_vs_proto_data and is handled by netns */ -#define KMSG_COMPONENT "IPVS" -#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt +#define pr_fmt(fmt) "IPVS: " fmt #include #include diff --git a/net/netfilter/ipvs/ip_vs_proto_udp.c b/net/netfilter/ipvs/ip_vs_proto_udp.c index 68260d91c988..0f0107c80dd2 100644 --- a/net/netfilter/ipvs/ip_vs_proto_udp.c +++ b/net/netfilter/ipvs/ip_vs_proto_udp.c @@ -9,8 +9,7 @@ * Network name space (netns) aware. */ -#define KMSG_COMPONENT "IPVS" -#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt +#define pr_fmt(fmt) "IPVS: " fmt #include #include diff --git a/net/netfilter/ipvs/ip_vs_rr.c b/net/netfilter/ipvs/ip_vs_rr.c index 6baa34dff9f0..4125ee561cdc 100644 --- a/net/netfilter/ipvs/ip_vs_rr.c +++ b/net/netfilter/ipvs/ip_vs_rr.c @@ -14,8 +14,7 @@ * Wensong Zhang : added any dest with weight=0 is quiesced */ -#define KMSG_COMPONENT "IPVS" -#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt +#define pr_fmt(fmt) "IPVS: " fmt #include #include diff --git a/net/netfilter/ipvs/ip_vs_sched.c b/net/netfilter/ipvs/ip_vs_sched.c index d4903723be7e..c6e421c4e299 100644 --- a/net/netfilter/ipvs/ip_vs_sched.c +++ b/net/netfilter/ipvs/ip_vs_sched.c @@ -12,8 +12,7 @@ * Changes: */ -#define KMSG_COMPONENT "IPVS" -#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt +#define pr_fmt(fmt) "IPVS: " fmt #include #include diff --git a/net/netfilter/ipvs/ip_vs_sed.c b/net/netfilter/ipvs/ip_vs_sed.c index a46f99a56618..245a323c84cd 100644 --- a/net/netfilter/ipvs/ip_vs_sed.c +++ b/net/netfilter/ipvs/ip_vs_sed.c @@ -30,8 +30,7 @@ * */ -#define KMSG_COMPONENT "IPVS" -#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt +#define pr_fmt(fmt) "IPVS: " fmt #include #include diff --git a/net/netfilter/ipvs/ip_vs_sh.c b/net/netfilter/ipvs/ip_vs_sh.c index 92e77d7a6b50..0e85e07e23b9 100644 --- a/net/netfilter/ipvs/ip_vs_sh.c +++ b/net/netfilter/ipvs/ip_vs_sh.c @@ -32,8 +32,7 @@ * */ -#define KMSG_COMPONENT "IPVS" -#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt +#define pr_fmt(fmt) "IPVS: " fmt #include #include diff --git a/net/netfilter/ipvs/ip_vs_sync.c b/net/netfilter/ipvs/ip_vs_sync.c index 5a0c6f42bd8f..54dd1514ac45 100644 --- a/net/netfilter/ipvs/ip_vs_sync.c +++ b/net/netfilter/ipvs/ip_vs_sync.c @@ -32,8 +32,7 @@ * Persistence support, fwmark and time-out. */ -#define KMSG_COMPONENT "IPVS" -#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt +#define pr_fmt(fmt) "IPVS: " fmt #include #include diff --git a/net/netfilter/ipvs/ip_vs_twos.c b/net/netfilter/ipvs/ip_vs_twos.c index 8d5419edde50..dbb7f5fd4688 100644 --- a/net/netfilter/ipvs/ip_vs_twos.c +++ b/net/netfilter/ipvs/ip_vs_twos.c @@ -4,8 +4,7 @@ * Authors: Darby Payne */ -#define KMSG_COMPONENT "IPVS" -#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt +#define pr_fmt(fmt) "IPVS: " fmt #include #include diff --git a/net/netfilter/ipvs/ip_vs_wlc.c b/net/netfilter/ipvs/ip_vs_wlc.c index 9fa500927c0a..9da445ca09a1 100644 --- a/net/netfilter/ipvs/ip_vs_wlc.c +++ b/net/netfilter/ipvs/ip_vs_wlc.c @@ -14,8 +14,7 @@ * Wensong Zhang : added any dest with weight=0 is quiesced */ -#define KMSG_COMPONENT "IPVS" -#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt +#define pr_fmt(fmt) "IPVS: " fmt #include #include diff --git a/net/netfilter/ipvs/ip_vs_wrr.c b/net/netfilter/ipvs/ip_vs_wrr.c index 85ce0d04afac..99f09cbf2d9b 100644 --- a/net/netfilter/ipvs/ip_vs_wrr.c +++ b/net/netfilter/ipvs/ip_vs_wrr.c @@ -13,8 +13,7 @@ * with weight 0 when all weights are zero */ -#define KMSG_COMPONENT "IPVS" -#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt +#define pr_fmt(fmt) "IPVS: " fmt #include #include diff --git a/net/netfilter/ipvs/ip_vs_xmit.c b/net/netfilter/ipvs/ip_vs_xmit.c index 95af252b2939..3162ce3c2640 100644 --- a/net/netfilter/ipvs/ip_vs_xmit.c +++ b/net/netfilter/ipvs/ip_vs_xmit.c @@ -21,8 +21,7 @@ * - the only place where we can see skb->sk != NULL */ -#define KMSG_COMPONENT "IPVS" -#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt +#define pr_fmt(fmt) "IPVS: " fmt #include #include diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c index e388de8dca09..f97f77b041d9 100644 --- a/net/smc/af_smc.c +++ b/net/smc/af_smc.c @@ -16,8 +16,7 @@ * based on prototype from Frank Blaschka */ -#define KMSG_COMPONENT "smc" -#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt +#define pr_fmt(fmt) "smc: " fmt #include #include From a6c121a2432eee2c4ebceb1483ccd4a50a52983d Mon Sep 17 00:00:00 2001 From: Robert Marko Date: Thu, 27 Nov 2025 12:44:35 +0100 Subject: [PATCH 1257/1381] net: phy: aquantia: check for NVMEM deferral Currently, if NVMEM provider is probed later than Aquantia, loading the firmware will fail with -EINVAL. To fix this, simply check for -EPROBE_DEFER when NVMEM is attempted and return it. Fixes: e93984ebc1c8 ("net: phy: aquantia: add firmware load support") Signed-off-by: Robert Marko Reviewed-by: Russell King (Oracle) Link: https://patch.msgid.link/20251127114514.460924-1-robimarko@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/phy/aquantia/aquantia_firmware.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/phy/aquantia/aquantia_firmware.c b/drivers/net/phy/aquantia/aquantia_firmware.c index bbbcc9736b00..569256152689 100644 --- a/drivers/net/phy/aquantia/aquantia_firmware.c +++ b/drivers/net/phy/aquantia/aquantia_firmware.c @@ -369,7 +369,7 @@ int aqr_firmware_load(struct phy_device *phydev) * assume that, and load a new image. */ ret = aqr_firmware_load_nvmem(phydev); - if (!ret) + if (ret == -EPROBE_DEFER || !ret) return ret; ret = aqr_firmware_load_fs(phydev); From 362d051c90b6e27b01dceed8c47ddfc86e60db2c Mon Sep 17 00:00:00 2001 From: Donald Hunter Date: Thu, 27 Nov 2025 12:34:59 +0000 Subject: [PATCH 1258/1381] tools: ynl: add schema checking Add a --validate flag to pyynl for explicit schema check with error reporting and add a schema_check make target to check all YNL specs. make -C tools/net/ynl schema_check make: Entering directory '/home/donaldh/net-next/tools/net/ynl' ok 1 binder.yaml schema validation not ok 2 conntrack.yaml schema validation 'labels mask' does not match '^[0-9a-z-]+$' Failed validating 'pattern' in schema['properties']['attribute-sets']['items']['properties']['attributes']['items']['properties']['name']: {'type': 'string', 'pattern': '^[0-9a-z-]+$'} On instance['attribute-sets'][14]['attributes'][22]['name']: 'labels mask' ok 3 devlink.yaml schema validation [...] Signed-off-by: Donald Hunter Link: https://patch.msgid.link/20251127123502.89142-2-donald.hunter@gmail.com Signed-off-by: Jakub Kicinski --- tools/net/ynl/Makefile | 20 +++++++++++++++++++- tools/net/ynl/pyynl/cli.py | 21 ++++++++++++++++----- 2 files changed, 35 insertions(+), 6 deletions(-) diff --git a/tools/net/ynl/Makefile b/tools/net/ynl/Makefile index a40591e513b7..b23083b2dfb2 100644 --- a/tools/net/ynl/Makefile +++ b/tools/net/ynl/Makefile @@ -12,6 +12,8 @@ endif libdir ?= $(prefix)/$(libdir_relative) includedir ?= $(prefix)/include +SPECDIR=../../../Documentation/netlink/specs + SUBDIRS = lib generated samples ynltool tests all: $(SUBDIRS) libynl.a @@ -54,4 +56,20 @@ install: libynl.a lib/*.h run_tests: @$(MAKE) -C tests run_tests -.PHONY: all clean distclean install run_tests $(SUBDIRS) + +schema_check: + @N=1; \ + for spec in $(SPECDIR)/*.yaml ; do \ + NAME=$$(basename $$spec) ; \ + OUTPUT=$$(./pyynl/cli.py --spec $$spec --validate) ; \ + if [ $$? -eq 0 ] ; then \ + echo "ok $$N $$NAME schema validation" ; \ + else \ + echo "not ok $$N $$NAME schema validation" ; \ + echo "$$OUTPUT" ; \ + echo ; \ + fi ; \ + N=$$((N+1)) ; \ + done + +.PHONY: all clean distclean install run_tests schema_check $(SUBDIRS) diff --git a/tools/net/ynl/pyynl/cli.py b/tools/net/ynl/pyynl/cli.py index ff81ff083764..af02a5b7e5a2 100755 --- a/tools/net/ynl/pyynl/cli.py +++ b/tools/net/ynl/pyynl/cli.py @@ -10,7 +10,7 @@ import sys import textwrap sys.path.append(pathlib.Path(__file__).resolve().parent.as_posix()) -from lib import YnlFamily, Netlink, NlError +from lib import YnlFamily, Netlink, NlError, SpecFamily sys_schema_dir='/usr/share/ynl' relative_schema_dir='../../../../Documentation/netlink' @@ -127,6 +127,7 @@ def main(): group.add_argument('--list-msgs', action='store_true') group.add_argument('--list-attrs', dest='list_attrs', metavar='OPERATION', type=str, help='List attributes for an operation') + group.add_argument('--validate', action='store_true') parser.add_argument('--duration', dest='duration', type=int, help='when subscribed, watch for DURATION seconds') @@ -168,15 +169,25 @@ def main(): if args.family: spec = f"{spec_dir()}/{args.family}.yaml" - if args.schema is None and spec.startswith(sys_schema_dir): - args.schema = '' # disable schema validation when installed - if args.process_unknown is None: - args.process_unknown = True else: spec = args.spec if not os.path.isfile(spec): raise Exception(f"Spec file {spec} does not exist") + if args.validate: + try: + SpecFamily(spec, args.schema) + except Exception as error: + print(error) + exit(1) + return + + if args.family: # set behaviour when using installed specs + if args.schema is None and spec.startswith(sys_schema_dir): + args.schema = '' # disable schema validation when installed + if args.process_unknown is None: + args.process_unknown = True + ynl = YnlFamily(spec, args.schema, args.process_unknown, recv_size=args.dbg_small_recv) if args.dbg_small_recv: From 129dc6075a150f82926ed10942f02f941985d51f Mon Sep 17 00:00:00 2001 From: Donald Hunter Date: Thu, 27 Nov 2025 12:35:00 +0000 Subject: [PATCH 1259/1381] tools: ynl: add a lint makefile target Add a lint target to run yamllint on the YNL specs. make -C tools/net/ynl lint make: Entering directory '/home/donaldh/net-next/tools/net/ynl' yamllint ../../../Documentation/netlink/specs/*.yaml ../../../Documentation/netlink/specs/ethtool.yaml 1272:21 warning truthy value should be one of [false, true] (truthy) make: Leaving directory '/home/donaldh/net-next/tools/net/ynl' Signed-off-by: Donald Hunter Link: https://patch.msgid.link/20251127123502.89142-3-donald.hunter@gmail.com Signed-off-by: Jakub Kicinski --- tools/net/ynl/Makefile | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/tools/net/ynl/Makefile b/tools/net/ynl/Makefile index b23083b2dfb2..7736b492f559 100644 --- a/tools/net/ynl/Makefile +++ b/tools/net/ynl/Makefile @@ -56,6 +56,8 @@ install: libynl.a lib/*.h run_tests: @$(MAKE) -C tests run_tests +lint: + yamllint $(SPECDIR) schema_check: @N=1; \ @@ -72,4 +74,4 @@ schema_check: N=$$((N+1)) ; \ done -.PHONY: all clean distclean install run_tests schema_check $(SUBDIRS) +.PHONY: all clean distclean install run_tests lint schema_check $(SUBDIRS) From acce9d7200e23de310c2cf8cd5439e86cd9ddc41 Mon Sep 17 00:00:00 2001 From: Donald Hunter Date: Thu, 27 Nov 2025 12:35:01 +0000 Subject: [PATCH 1260/1381] ynl: fix a yamllint warning in ethtool spec Fix warning reported by yamllint: ../../../Documentation/netlink/specs/ethtool.yaml 1272:21 warning truthy value should be one of [false, true] (truthy) Signed-off-by: Donald Hunter Link: https://patch.msgid.link/20251127123502.89142-4-donald.hunter@gmail.com Signed-off-by: Jakub Kicinski --- Documentation/netlink/specs/ethtool.yaml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Documentation/netlink/specs/ethtool.yaml b/Documentation/netlink/specs/ethtool.yaml index 05d2b6508b59..0a2d2343f79a 100644 --- a/Documentation/netlink/specs/ethtool.yaml +++ b/Documentation/netlink/specs/ethtool.yaml @@ -1269,7 +1269,7 @@ attribute-sets: - name: hist type: nest - multi-attr: True + multi-attr: true nested-attributes: fec-hist - name: fec From 1adc241f3940c172c8c674a62004d3c34b360ab3 Mon Sep 17 00:00:00 2001 From: Donald Hunter Date: Thu, 27 Nov 2025 12:35:02 +0000 Subject: [PATCH 1261/1381] ynl: fix schema check errors Fix two schema check errors that have lurked since the attribute name validation was made more strict: not ok 2 conntrack.yaml schema validation 'labels mask' does not match '^[0-9a-z-]+$' not ok 13 nftables.yaml schema validation 'set id' does not match '^[0-9a-z-]+$' Signed-off-by: Donald Hunter Link: https://patch.msgid.link/20251127123502.89142-5-donald.hunter@gmail.com Signed-off-by: Jakub Kicinski --- Documentation/netlink/specs/conntrack.yaml | 2 +- Documentation/netlink/specs/nftables.yaml | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/Documentation/netlink/specs/conntrack.yaml b/Documentation/netlink/specs/conntrack.yaml index bef528633b17..db7cddcda50a 100644 --- a/Documentation/netlink/specs/conntrack.yaml +++ b/Documentation/netlink/specs/conntrack.yaml @@ -457,7 +457,7 @@ attribute-sets: name: labels type: binary - - name: labels mask + name: labels-mask type: binary - name: synproxy diff --git a/Documentation/netlink/specs/nftables.yaml b/Documentation/netlink/specs/nftables.yaml index cce88819ba71..17ad707fa0d5 100644 --- a/Documentation/netlink/specs/nftables.yaml +++ b/Documentation/netlink/specs/nftables.yaml @@ -915,7 +915,7 @@ attribute-sets: type: string doc: Name of set to use - - name: set id + name: set-id type: u32 byte-order: big-endian doc: ID of set to use From bba18f3ba7cc1ca47428bc563fcc34e9e2ce0bf8 Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Thu, 27 Nov 2025 02:17:15 -0800 Subject: [PATCH 1262/1381] net: bnxt: extract GRXRINGS from .get_rxnfc Commit 84eaf4359c36 ("net: ethtool: add get_rx_ring_count callback to optimize RX ring queries") added specific support for GRXRINGS callback, simplifying .get_rxnfc. Remove the handling of GRXRINGS in .get_rxnfc() by moving it to the new .get_rx_ring_count(). This simplifies the RX ring count retrieval and aligns bnxt with the new ethtool API for querying RX ring parameters. Signed-off-by: Breno Leitao Link: https://patch.msgid.link/20251127-grxrings_broadcom-v1-1-b0b182864950@debian.org Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c index efb9bf20e66b..068e191ede19 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c @@ -1770,6 +1770,13 @@ static int bnxt_set_rxfh_fields(struct net_device *dev, return rc; } +static u32 bnxt_get_rx_ring_count(struct net_device *dev) +{ + struct bnxt *bp = netdev_priv(dev); + + return bp->rx_nr_rings; +} + static int bnxt_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd, u32 *rule_locs) { @@ -1777,10 +1784,6 @@ static int bnxt_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd, int rc = 0; switch (cmd->cmd) { - case ETHTOOL_GRXRINGS: - cmd->data = bp->rx_nr_rings; - break; - case ETHTOOL_GRXCLSRLCNT: cmd->rule_cnt = bp->ntp_fltr_count; cmd->data = bp->max_fltr | RX_CLS_LOC_SPECIAL; @@ -5616,6 +5619,7 @@ const struct ethtool_ops bnxt_ethtool_ops = { .set_channels = bnxt_set_channels, .get_rxnfc = bnxt_get_rxnfc, .set_rxnfc = bnxt_set_rxnfc, + .get_rx_ring_count = bnxt_get_rx_ring_count, .get_rxfh_indir_size = bnxt_get_rxfh_indir_size, .get_rxfh_key_size = bnxt_get_rxfh_key_size, .get_rxfh = bnxt_get_rxfh, From 335d78c6161b18e95f72404ee3e210e64b5f5adc Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Thu, 27 Nov 2025 02:17:16 -0800 Subject: [PATCH 1263/1381] net: bcmgenet: extract GRXRINGS from .get_rxnfc Commit 84eaf4359c36 ("net: ethtool: add get_rx_ring_count callback to optimize RX ring queries") added specific support for GRXRINGS callback, simplifying .get_rxnfc. Remove the handling of GRXRINGS in .get_rxnfc() by moving it to the new .get_rx_ring_count(). This simplifies the RX ring count retrieval and aligns bcmgenet with the new ethtool API for querying RX ring parameters. Signed-off-by: Breno Leitao Link: https://patch.msgid.link/20251127-grxrings_broadcom-v1-2-b0b182864950@debian.org Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/broadcom/genet/bcmgenet.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.c b/drivers/net/ethernet/broadcom/genet/bcmgenet.c index d99ef92feb82..05512aa10c20 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c +++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c @@ -1640,6 +1640,13 @@ static int bcmgenet_get_num_flows(struct bcmgenet_priv *priv) return res; } +static u32 bcmgenet_get_rx_ring_count(struct net_device *dev) +{ + struct bcmgenet_priv *priv = netdev_priv(dev); + + return priv->hw_params->rx_queues ?: 1; +} + static int bcmgenet_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd, u32 *rule_locs) { @@ -1649,9 +1656,6 @@ static int bcmgenet_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd, int i = 0; switch (cmd->cmd) { - case ETHTOOL_GRXRINGS: - cmd->data = priv->hw_params->rx_queues ?: 1; - break; case ETHTOOL_GRXCLSRLCNT: cmd->rule_cnt = bcmgenet_get_num_flows(priv); cmd->data = MAX_NUM_OF_FS_RULES | RX_CLS_LOC_SPECIAL; @@ -1700,6 +1704,7 @@ static const struct ethtool_ops bcmgenet_ethtool_ops = { .get_ts_info = ethtool_op_get_ts_info, .get_rxnfc = bcmgenet_get_rxnfc, .set_rxnfc = bcmgenet_set_rxnfc, + .get_rx_ring_count = bcmgenet_get_rx_ring_count, .get_pauseparam = bcmgenet_get_pauseparam, .set_pauseparam = bcmgenet_set_pauseparam, }; From 6f2e1c75bc5eb2237e03c134c32a2a352484881e Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Thu, 27 Nov 2025 14:08:48 +0200 Subject: [PATCH 1264/1381] net: dsa: introduce the dsa_xmit_port_mask() tagging protocol helper MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Many tagging protocols deal with the transmit port mask being a bit mask, and set it to BIT(dp->index). Not a big deal. Also, some tagging protocols are written for switches which support HSR offload (including packet duplication offload), there we see a walk using dsa_hsr_foreach_port() to find the other port in the same switch that's member of the HSR, and set that bit in the port mask too. That isn't sufficiently interesting either, until you come to realize that there isn't anything special in the second case that switches just in the first one can't do too. It just becomes a matter of "is it wise to do it? are sufficient people using HSR/PRP with generic off-the-shelf switches to justify add an extra test in the data path?" - the answer to which is probably "it depends". It isn't _much_ worse to not have HSR offload at all, so as to make it impractical, esp. with a rich OS like Linux. But the HSR users are rather specialized in industrial networking. Anyway, the change acts on the premise that we're going to have support for this, it should be uniformly implemented for everyone, and that if we find some sort of balance, we can keep everyone relatively happy. So I've disabled that logic if CONFIG_HSR isn't enabled, and I've tilted the branch predictor to say it's unlikely we're transmitting through a port with this capability currently active. On branch miss, we're still going to save the transmission of one packet, so there's some remaining benefit there too. I don't _think_ we need to jump to static keys yet. The helper returns a 32-bit zero-based unsigned number, that callers have to transpose using FIELD_PREP(). It is not the first time we assume DSA switches won't be larger than 32 ports - dsa_user_ports() has that assumption baked into it too. One last development note about why pass the "skb" argument when this isn't used. Looking at the compiled code on arm64, which is identical both with and without it, the answer is "why not?" - who knows what other features dependent on the skb may be handled in the future. Link: https://lore.kernel.org/netdev/20251126093240.2853294-4-mmyangfl@gmail.com/ Cc: "Alvin Šipraga" Cc: Chester A. Unal" Cc: "Clément Léger" Cc: Daniel Golle Cc: David Yang Cc: DENG Qingfang Cc: Florian Fainelli Cc: George McCollister Cc: Hauke Mehrtens Cc: Jonas Gorski Cc: Kurt Kanzenbach Cc: Linus Walleij Cc: Sean Wang Cc: UNGLinuxDriver@microchip.com Cc: Woojung Huh Signed-off-by: Vladimir Oltean Link: https://patch.msgid.link/20251127120902.292555-2-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- net/dsa/tag.h | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+) diff --git a/net/dsa/tag.h b/net/dsa/tag.h index 5d80ddad4ff6..cf52283fe9df 100644 --- a/net/dsa/tag.h +++ b/net/dsa/tag.h @@ -319,6 +319,24 @@ static inline void *dsa_etype_header_pos_tx(struct sk_buff *skb) return skb->data + 2 * ETH_ALEN; } +static inline unsigned long dsa_xmit_port_mask(const struct sk_buff *skb, + const struct net_device *dev) +{ + struct dsa_port *dp = dsa_user_to_port(dev); + unsigned long mask = BIT(dp->index); + + if (IS_ENABLED(CONFIG_HSR) && + unlikely(dev->features & NETIF_F_HW_HSR_DUP)) { + struct net_device *hsr_dev = dp->hsr_dev; + struct dsa_port *other_dp; + + dsa_hsr_foreach_port(other_dp, dp->ds, hsr_dev) + mask |= BIT(other_dp->index); + } + + return mask; +} + /* Create 2 modaliases per tagging protocol, one to auto-load the module * given the ID reported by get_tag_protocol(), and the other by name. */ From 621d06a40e4719fa76b8572046663d54e16425a5 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Thu, 27 Nov 2025 14:08:49 +0200 Subject: [PATCH 1265/1381] net: dsa: tag_brcm: use the dsa_xmit_port_mask() helper The "brcm" and "brcm-prepend" tagging protocols populate a bit mask for the TX ports, so we can use dsa_xmit_port_mask() to centralize the decision of how to set that field. The port mask is written u8 by u8, first the high octet and then the low octet. Cc: Florian Fainelli Cc: Jonas Gorski Signed-off-by: Vladimir Oltean Link: https://patch.msgid.link/20251127120902.292555-3-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- net/dsa/tag_brcm.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/net/dsa/tag_brcm.c b/net/dsa/tag_brcm.c index eadb358179ce..cf9420439054 100644 --- a/net/dsa/tag_brcm.c +++ b/net/dsa/tag_brcm.c @@ -92,6 +92,7 @@ static struct sk_buff *brcm_tag_xmit_ll(struct sk_buff *skb, { struct dsa_port *dp = dsa_user_to_port(dev); u16 queue = skb_get_queue_mapping(skb); + u16 port_mask; u8 *brcm_tag; /* The Ethernet switch we are interfaced with needs packets to be at @@ -119,10 +120,9 @@ static struct sk_buff *brcm_tag_xmit_ll(struct sk_buff *skb, brcm_tag[0] = (1 << BRCM_OPCODE_SHIFT) | ((queue & BRCM_IG_TC_MASK) << BRCM_IG_TC_SHIFT); brcm_tag[1] = 0; - brcm_tag[2] = 0; - if (dp->index == 8) - brcm_tag[2] = BRCM_IG_DSTMAP2_MASK; - brcm_tag[3] = (1 << dp->index) & BRCM_IG_DSTMAP1_MASK; + port_mask = dsa_xmit_port_mask(skb, dev); + brcm_tag[2] = (port_mask >> 8) & BRCM_IG_DSTMAP2_MASK; + brcm_tag[3] = port_mask & BRCM_IG_DSTMAP1_MASK; /* Now tell the conduit network device about the desired output queue * as well From e094428fb40c1b2cc98d97bc93db935aa65daa63 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Thu, 27 Nov 2025 14:08:50 +0200 Subject: [PATCH 1266/1381] net: dsa: tag_gswip: use the dsa_xmit_port_mask() helper The "gswip" tagging protocol populates a bit mask for the TX ports, so we can use dsa_xmit_port_mask() to centralize the decision of how to set that field. Cc: Hauke Mehrtens Cc: Daniel Golle Signed-off-by: Vladimir Oltean Link: https://patch.msgid.link/20251127120902.292555-4-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- net/dsa/tag_gswip.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/net/dsa/tag_gswip.c b/net/dsa/tag_gswip.c index 51a1f46a567f..5fa436121087 100644 --- a/net/dsa/tag_gswip.c +++ b/net/dsa/tag_gswip.c @@ -48,8 +48,7 @@ /* Byte 3 */ #define GSWIP_TX_DPID_EN BIT(0) -#define GSWIP_TX_PORT_MAP_SHIFT 1 -#define GSWIP_TX_PORT_MAP_MASK GENMASK(6, 1) +#define GSWIP_TX_PORT_MAP GENMASK(6, 1) #define GSWIP_RX_HEADER_LEN 8 @@ -61,7 +60,6 @@ static struct sk_buff *gswip_tag_xmit(struct sk_buff *skb, struct net_device *dev) { - struct dsa_port *dp = dsa_user_to_port(dev); u8 *gswip_tag; skb_push(skb, GSWIP_TX_HEADER_LEN); @@ -70,7 +68,7 @@ static struct sk_buff *gswip_tag_xmit(struct sk_buff *skb, gswip_tag[0] = GSWIP_TX_SLPID_CPU; gswip_tag[1] = GSWIP_TX_DPID_ELAN; gswip_tag[2] = GSWIP_TX_PORT_MAP_EN | GSWIP_TX_PORT_MAP_SEL; - gswip_tag[3] = BIT(dp->index + GSWIP_TX_PORT_MAP_SHIFT) & GSWIP_TX_PORT_MAP_MASK; + gswip_tag[3] = FIELD_PREP(GSWIP_TX_PORT_MAP, dsa_xmit_port_mask(skb, dev)); gswip_tag[3] |= GSWIP_TX_DPID_EN; return skb; From f59e44cc0d6c586d32732377d748045eeacc8cee Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Thu, 27 Nov 2025 14:08:51 +0200 Subject: [PATCH 1267/1381] net: dsa: tag_hellcreek: use the dsa_xmit_port_mask() helper The "hellcreek" tagging protocol populates a bit mask for the TX ports, so we can use dsa_xmit_port_mask() to centralize the decision of how to set that field. Cc: Kurt Kanzenbach Signed-off-by: Vladimir Oltean Link: https://patch.msgid.link/20251127120902.292555-5-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- net/dsa/tag_hellcreek.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/net/dsa/tag_hellcreek.c b/net/dsa/tag_hellcreek.c index 663b25785d95..544ab15685a2 100644 --- a/net/dsa/tag_hellcreek.c +++ b/net/dsa/tag_hellcreek.c @@ -20,7 +20,6 @@ static struct sk_buff *hellcreek_xmit(struct sk_buff *skb, struct net_device *dev) { - struct dsa_port *dp = dsa_user_to_port(dev); u8 *tag; /* Calculate checksums (if required) before adding the trailer tag to @@ -33,7 +32,7 @@ static struct sk_buff *hellcreek_xmit(struct sk_buff *skb, /* Tag encoding */ tag = skb_put(skb, HELLCREEK_TAG_LEN); - *tag = BIT(dp->index); + *tag = dsa_xmit_port_mask(skb, dev); return skb; } From ea659a9292b1df91abb80e56198d34f8c6697e84 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Thu, 27 Nov 2025 14:08:52 +0200 Subject: [PATCH 1268/1381] net: dsa: tag_ksz: use the dsa_xmit_port_mask() helper The "ksz8795", "ksz9893", "ksz9477" and "lan937x" tagging protocols populate a bit mask for the TX ports. Unlike the others, "ksz9477" also accelerates HSR packet duplication. Make the HSR duplication logic available generically to all 4 taggers by using the dsa_xmit_port_mask() function to set the TX port mask. Cc: Woojung Huh Cc: UNGLinuxDriver@microchip.com Signed-off-by: Vladimir Oltean Link: https://patch.msgid.link/20251127120902.292555-6-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- net/dsa/tag_ksz.c | 20 ++++---------------- 1 file changed, 4 insertions(+), 16 deletions(-) diff --git a/net/dsa/tag_ksz.c b/net/dsa/tag_ksz.c index 0b7564b53790..9170a0148cc4 100644 --- a/net/dsa/tag_ksz.c +++ b/net/dsa/tag_ksz.c @@ -120,7 +120,6 @@ static struct sk_buff *ksz_common_rcv(struct sk_buff *skb, static struct sk_buff *ksz8795_xmit(struct sk_buff *skb, struct net_device *dev) { - struct dsa_port *dp = dsa_user_to_port(dev); struct ethhdr *hdr; u8 *tag; @@ -131,7 +130,7 @@ static struct sk_buff *ksz8795_xmit(struct sk_buff *skb, struct net_device *dev) tag = skb_put(skb, KSZ_INGRESS_TAG_LEN); hdr = skb_eth_hdr(skb); - *tag = 1 << dp->index; + *tag = dsa_xmit_port_mask(skb, dev); if (is_link_local_ether_addr(hdr->h_dest)) *tag |= KSZ8795_TAIL_TAG_OVERRIDE; @@ -294,21 +293,12 @@ static struct sk_buff *ksz9477_xmit(struct sk_buff *skb, tag = skb_put(skb, KSZ9477_INGRESS_TAG_LEN); hdr = skb_eth_hdr(skb); - val = BIT(dp->index); - + val = dsa_xmit_port_mask(skb, dev); val |= FIELD_PREP(KSZ9477_TAIL_TAG_PRIO, prio); if (is_link_local_ether_addr(hdr->h_dest)) val |= KSZ9477_TAIL_TAG_OVERRIDE; - if (dev->features & NETIF_F_HW_HSR_DUP) { - struct net_device *hsr_dev = dp->hsr_dev; - struct dsa_port *other_dp; - - dsa_hsr_foreach_port(other_dp, dp->ds, hsr_dev) - val |= BIT(other_dp->index); - } - *tag = cpu_to_be16(val); return ksz_defer_xmit(dp, skb); @@ -371,8 +361,7 @@ static struct sk_buff *ksz9893_xmit(struct sk_buff *skb, tag = skb_put(skb, KSZ_INGRESS_TAG_LEN); hdr = skb_eth_hdr(skb); - *tag = BIT(dp->index); - + *tag = dsa_xmit_port_mask(skb, dev); *tag |= FIELD_PREP(KSZ9893_TAIL_TAG_PRIO, prio); if (is_link_local_ether_addr(hdr->h_dest)) @@ -436,8 +425,7 @@ static struct sk_buff *lan937x_xmit(struct sk_buff *skb, tag = skb_put(skb, LAN937X_EGRESS_TAG_LEN); - val = BIT(dp->index); - + val = dsa_xmit_port_mask(skb, dev); val |= FIELD_PREP(LAN937X_TAIL_TAG_PRIO, prio); if (is_link_local_ether_addr(hdr->h_dest)) From 84a60bbec50341ec49f9b4d56ca8dab9d99807f6 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Thu, 27 Nov 2025 14:08:53 +0200 Subject: [PATCH 1269/1381] net: dsa: tag_mtk: use the dsa_xmit_port_mask() helper The "mtk" tagging protocol populates a bit mask for the TX ports, so we can use dsa_xmit_port_mask() to centralize the decision of how to set that field. Cc: Chester A. Unal" Cc: Daniel Golle Cc: DENG Qingfang Cc: Sean Wang Signed-off-by: Vladimir Oltean Link: https://patch.msgid.link/20251127120902.292555-7-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- net/dsa/tag_mtk.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/dsa/tag_mtk.c b/net/dsa/tag_mtk.c index b670e3c53e91..dea3eecaf093 100644 --- a/net/dsa/tag_mtk.c +++ b/net/dsa/tag_mtk.c @@ -54,7 +54,8 @@ static struct sk_buff *mtk_tag_xmit(struct sk_buff *skb, * whether that's a combined special tag with 802.1Q header. */ mtk_tag[0] = xmit_tpid; - mtk_tag[1] = (1 << dp->index) & MTK_HDR_XMIT_DP_BIT_MASK; + mtk_tag[1] = FIELD_PREP(MTK_HDR_XMIT_DP_BIT_MASK, + dsa_xmit_port_mask(skb, dev)); /* Tag control information is kept for 802.1Q */ if (xmit_tpid == MTK_HDR_XMIT_UNTAGGED) { From a4a00d9e365a99188e40153b5672d184649eb14e Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Thu, 27 Nov 2025 14:08:54 +0200 Subject: [PATCH 1270/1381] net: dsa: tag_mxl_gsw1xx: use the dsa_xmit_port_mask() helper The "gsw1xx" tagging protocol populates a bit mask for the TX ports, so we can use dsa_xmit_port_mask() to centralize the decision of how to set that field. Cc: Hauke Mehrtens Cc: Daniel Golle Signed-off-by: Vladimir Oltean Link: https://patch.msgid.link/20251127120902.292555-8-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- net/dsa/tag_mxl-gsw1xx.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/net/dsa/tag_mxl-gsw1xx.c b/net/dsa/tag_mxl-gsw1xx.c index 701a079955f2..60f7c445e656 100644 --- a/net/dsa/tag_mxl-gsw1xx.c +++ b/net/dsa/tag_mxl-gsw1xx.c @@ -43,8 +43,8 @@ static struct sk_buff *gsw1xx_tag_xmit(struct sk_buff *skb, struct net_device *dev) { - struct dsa_port *dp = dsa_user_to_port(dev); __be16 *gsw1xx_tag; + u16 tag; /* provide additional space 'GSW1XX_HEADER_LEN' bytes */ skb_push(skb, GSW1XX_HEADER_LEN); @@ -55,9 +55,10 @@ static struct sk_buff *gsw1xx_tag_xmit(struct sk_buff *skb, /* special tag ingress */ gsw1xx_tag = dsa_etype_header_pos_tx(skb); gsw1xx_tag[0] = htons(ETH_P_MXLGSW); - gsw1xx_tag[1] = htons(GSW1XX_TX_PORT_MAP_EN | GSW1XX_TX_LRN_DIS | - FIELD_PREP(GSW1XX_TX_PORT_MAP, BIT(dp->index))); + tag = FIELD_PREP(GSW1XX_TX_PORT_MAP, dsa_xmit_port_mask(skb, dev)) | + GSW1XX_TX_PORT_MAP_EN | GSW1XX_TX_LRN_DIS; + gsw1xx_tag[1] = htons(tag); gsw1xx_tag[2] = 0; gsw1xx_tag[3] = 0; From 5733fe2a7ad11a2d1ef8aaa0e6693923b2cbbdc2 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Thu, 27 Nov 2025 14:08:55 +0200 Subject: [PATCH 1271/1381] net: dsa: tag_ocelot: use the dsa_xmit_port_mask() helper The "ocelot" and "seville" tagging protocols populate a bit mask for the TX ports, so we can use dsa_xmit_port_mask() to centralize the decision of how to set that field. This protocol used BIT_ULL() rather than simple BIT() to silence Smatch, as explained in commit 1f778d500df3 ("net: mscc: ocelot: avoid type promotion when calling ocelot_ifh_set_dest"). I would expect that this tool no longer complains now, when the BIT(dp->index) is hidden inside the dsa_xmit_port_mask() function, the return value of which is promoted to u64. Signed-off-by: Vladimir Oltean Link: https://patch.msgid.link/20251127120902.292555-9-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- net/dsa/tag_ocelot.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/net/dsa/tag_ocelot.c b/net/dsa/tag_ocelot.c index bf6608fc6be7..3405def79c2d 100644 --- a/net/dsa/tag_ocelot.c +++ b/net/dsa/tag_ocelot.c @@ -46,11 +46,10 @@ static void ocelot_xmit_common(struct sk_buff *skb, struct net_device *netdev, static struct sk_buff *ocelot_xmit(struct sk_buff *skb, struct net_device *netdev) { - struct dsa_port *dp = dsa_user_to_port(netdev); void *injection; ocelot_xmit_common(skb, netdev, cpu_to_be32(0x8880000a), &injection); - ocelot_ifh_set_dest(injection, BIT_ULL(dp->index)); + ocelot_ifh_set_dest(injection, dsa_xmit_port_mask(skb, netdev)); return skb; } @@ -58,11 +57,10 @@ static struct sk_buff *ocelot_xmit(struct sk_buff *skb, static struct sk_buff *seville_xmit(struct sk_buff *skb, struct net_device *netdev) { - struct dsa_port *dp = dsa_user_to_port(netdev); void *injection; ocelot_xmit_common(skb, netdev, cpu_to_be32(0x88800005), &injection); - seville_ifh_set_dest(injection, BIT_ULL(dp->index)); + seville_ifh_set_dest(injection, dsa_xmit_port_mask(skb, netdev)); return skb; } From 48afabaf4aaa2991df94cdfd554687b8c74b58be Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Thu, 27 Nov 2025 14:08:56 +0200 Subject: [PATCH 1272/1381] net: dsa: tag_qca: use the dsa_xmit_port_mask() helper The "qca" tagging protocol populates a bit mask for the TX ports, so we can use dsa_xmit_port_mask() to centralize the decision of how to set that field. Signed-off-by: Vladimir Oltean Link: https://patch.msgid.link/20251127120902.292555-10-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- net/dsa/tag_qca.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/net/dsa/tag_qca.c b/net/dsa/tag_qca.c index 0cf61286b426..6d56a28c914c 100644 --- a/net/dsa/tag_qca.c +++ b/net/dsa/tag_qca.c @@ -14,7 +14,6 @@ static struct sk_buff *qca_tag_xmit(struct sk_buff *skb, struct net_device *dev) { - struct dsa_port *dp = dsa_user_to_port(dev); __be16 *phdr; u16 hdr; @@ -26,7 +25,7 @@ static struct sk_buff *qca_tag_xmit(struct sk_buff *skb, struct net_device *dev) /* Set the version field, and set destination port information */ hdr = FIELD_PREP(QCA_HDR_XMIT_VERSION, QCA_HDR_VERSION); hdr |= QCA_HDR_XMIT_FROM_CPU; - hdr |= FIELD_PREP(QCA_HDR_XMIT_DP_BIT, BIT(dp->index)); + hdr |= FIELD_PREP(QCA_HDR_XMIT_DP_BIT, dsa_xmit_port_mask(skb, dev)); *phdr = htons(hdr); From 4abf39c8aef5a0d4ec83f82164f69a3fcf8c8491 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Thu, 27 Nov 2025 14:08:57 +0200 Subject: [PATCH 1273/1381] net: dsa: tag_rtl4_a: use the dsa_xmit_port_mask() helper MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The "rtl4a" tagging protocol populates a bit mask for the TX ports, so we can use dsa_xmit_port_mask() to centralize the decision of how to set that field. Cc: Linus Walleij Cc: "Alvin Šipraga" Signed-off-by: Vladimir Oltean Link: https://patch.msgid.link/20251127120902.292555-11-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- net/dsa/tag_rtl4_a.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/dsa/tag_rtl4_a.c b/net/dsa/tag_rtl4_a.c index feaefa0e179b..3cc63eacfa03 100644 --- a/net/dsa/tag_rtl4_a.c +++ b/net/dsa/tag_rtl4_a.c @@ -57,7 +57,7 @@ static struct sk_buff *rtl4a_tag_xmit(struct sk_buff *skb, out = (RTL4_A_PROTOCOL_RTL8366RB << RTL4_A_PROTOCOL_SHIFT); /* The lower bits indicate the port number */ - out |= BIT(dp->index); + out |= dsa_xmit_port_mask(skb, dev); p = (__be16 *)(tag + 2); *p = htons(out); From 5afe4ccc33f47ab84efab01d8f159542fd945b95 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Thu, 27 Nov 2025 14:08:58 +0200 Subject: [PATCH 1274/1381] net: dsa: tag_rtl8_4: use the dsa_xmit_port_mask() helper MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The "rtl8_4" and "rtl8_4t" tagging protocols populate a bit mask for the TX ports, so we can use dsa_xmit_port_mask() to centralize the decision of how to set that field. Cc: Linus Walleij Cc: "Alvin Šipraga" Signed-off-by: Vladimir Oltean Link: https://patch.msgid.link/20251127120902.292555-12-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- net/dsa/tag_rtl8_4.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/net/dsa/tag_rtl8_4.c b/net/dsa/tag_rtl8_4.c index 15c2bae2b429..2464545da4d2 100644 --- a/net/dsa/tag_rtl8_4.c +++ b/net/dsa/tag_rtl8_4.c @@ -103,7 +103,6 @@ static void rtl8_4_write_tag(struct sk_buff *skb, struct net_device *dev, void *tag) { - struct dsa_port *dp = dsa_user_to_port(dev); __be16 tag16[RTL8_4_TAG_LEN / 2]; /* Set Realtek EtherType */ @@ -116,7 +115,7 @@ static void rtl8_4_write_tag(struct sk_buff *skb, struct net_device *dev, tag16[2] = htons(FIELD_PREP(RTL8_4_LEARN_DIS, 1)); /* Zero ALLOW; set RX (CPU->switch) forwarding port mask */ - tag16[3] = htons(FIELD_PREP(RTL8_4_RX, BIT(dp->index))); + tag16[3] = htons(FIELD_PREP(RTL8_4_RX, dsa_xmit_port_mask(skb, dev))); memcpy(tag, tag16, RTL8_4_TAG_LEN); } From b33aa90e68b42fd2c6125c1173449eb8c1c97d2c Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Thu, 27 Nov 2025 14:08:59 +0200 Subject: [PATCH 1275/1381] net: dsa: tag_rzn1_a5psw: use the dsa_xmit_port_mask() helper MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The "a5psw" tagging protocol populates a bit mask for the TX ports, so we can use dsa_xmit_port_mask() to centralize the decision of how to set that field. Cc: "Clément Léger" Signed-off-by: Vladimir Oltean Link: https://patch.msgid.link/20251127120902.292555-13-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- net/dsa/tag_rzn1_a5psw.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/net/dsa/tag_rzn1_a5psw.c b/net/dsa/tag_rzn1_a5psw.c index 69d51221b1e5..10994b3470f6 100644 --- a/net/dsa/tag_rzn1_a5psw.c +++ b/net/dsa/tag_rzn1_a5psw.c @@ -39,7 +39,6 @@ struct a5psw_tag { static struct sk_buff *a5psw_tag_xmit(struct sk_buff *skb, struct net_device *dev) { - struct dsa_port *dp = dsa_user_to_port(dev); struct a5psw_tag *ptag; u32 data2_val; @@ -60,7 +59,7 @@ static struct sk_buff *a5psw_tag_xmit(struct sk_buff *skb, struct net_device *de ptag = dsa_etype_header_pos_tx(skb); - data2_val = FIELD_PREP(A5PSW_CTRL_DATA_PORT, BIT(dp->index)); + data2_val = FIELD_PREP(A5PSW_CTRL_DATA_PORT, dsa_xmit_port_mask(skb, dev)); ptag->ctrl_tag = htons(ETH_P_DSA_A5PSW); ptag->ctrl_data = htons(A5PSW_CTRL_DATA_FORCE_FORWARD); ptag->ctrl_data2_lo = htons(data2_val); From 3c1975bbdf92c952c8c17ed67aa652adc6b64962 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Thu, 27 Nov 2025 14:09:00 +0200 Subject: [PATCH 1276/1381] net: dsa: tag_trailer: use the dsa_xmit_port_mask() helper The "trailer" tagging protocol populates a bit mask for the TX ports, so we can use dsa_xmit_port_mask() to centralize the decision of how to set that field. Signed-off-by: Vladimir Oltean Link: https://patch.msgid.link/20251127120902.292555-14-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- net/dsa/tag_trailer.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/net/dsa/tag_trailer.c b/net/dsa/tag_trailer.c index 22742a53d6f4..4dce24cfe6a7 100644 --- a/net/dsa/tag_trailer.c +++ b/net/dsa/tag_trailer.c @@ -14,12 +14,11 @@ static struct sk_buff *trailer_xmit(struct sk_buff *skb, struct net_device *dev) { - struct dsa_port *dp = dsa_user_to_port(dev); u8 *trailer; trailer = skb_put(skb, 4); trailer[0] = 0x80; - trailer[1] = 1 << dp->index; + trailer[1] = dsa_xmit_port_mask(skb, dev); trailer[2] = 0x10; trailer[3] = 0x00; From 24099389a63fb9a3cf79d04f38c7892a9957858b Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Thu, 27 Nov 2025 14:09:01 +0200 Subject: [PATCH 1277/1381] net: dsa: tag_xrs700x: use the dsa_xmit_port_mask() helper The "xrs700x" is the original DSA tagging protocol with HSR TX replication support, we now essentially move that logic to the dsa_xmit_port_mask() helper. The end result is something akin to hellcreek_xmit() (but reminds me I should also take care of skb_checksum_help() for tail taggers in the core). The implementation differences to dsa_xmit_port_mask() are immaterial. Cc: George McCollister Signed-off-by: Vladimir Oltean Link: https://patch.msgid.link/20251127120902.292555-15-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- net/dsa/tag_xrs700x.c | 8 +------- 1 file changed, 1 insertion(+), 7 deletions(-) diff --git a/net/dsa/tag_xrs700x.c b/net/dsa/tag_xrs700x.c index 68d4633ddd5e..a05219f702c6 100644 --- a/net/dsa/tag_xrs700x.c +++ b/net/dsa/tag_xrs700x.c @@ -13,16 +13,10 @@ static struct sk_buff *xrs700x_xmit(struct sk_buff *skb, struct net_device *dev) { - struct dsa_port *partner, *dp = dsa_user_to_port(dev); u8 *trailer; trailer = skb_put(skb, 1); - trailer[0] = BIT(dp->index); - - if (dp->hsr_dev) - dsa_hsr_foreach_port(partner, dp->ds, dp->hsr_dev) - if (partner != dp) - trailer[0] |= BIT(partner->index); + trailer[0] = dsa_xmit_port_mask(skb, dev); return skb; } From 64b0d2edb61aba9009a1c957568ad674059841a0 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Thu, 27 Nov 2025 14:09:02 +0200 Subject: [PATCH 1278/1381] net: dsa: tag_yt921x: use the dsa_xmit_port_mask() helper The "yt921x" tagging protocol populates a bit mask for the TX ports, so we can use dsa_xmit_port_mask() to centralize the decision of how to set that field. Cc: David Yang Signed-off-by: Vladimir Oltean Link: https://patch.msgid.link/20251127120902.292555-16-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- net/dsa/tag_yt921x.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) diff --git a/net/dsa/tag_yt921x.c b/net/dsa/tag_yt921x.c index 995da44f0a2a..6bbfd42dc5df 100644 --- a/net/dsa/tag_yt921x.c +++ b/net/dsa/tag_yt921x.c @@ -38,14 +38,11 @@ #define YT921X_TAG_RX_CMD_FORWARDED 0x80 #define YT921X_TAG_RX_CMD_UNK_UCAST 0xb2 #define YT921X_TAG_RX_CMD_UNK_MCAST 0xb4 -#define YT921X_TAG_TX_PORTS_M GENMASK(10, 0) -#define YT921X_TAG_TX_PORTn(port) BIT(port) +#define YT921X_TAG_TX_PORTS GENMASK(10, 0) static struct sk_buff * yt921x_tag_xmit(struct sk_buff *skb, struct net_device *netdev) { - struct dsa_port *dp = dsa_user_to_port(netdev); - unsigned int port = dp->index; __be16 *tag; u16 tx; @@ -58,7 +55,8 @@ yt921x_tag_xmit(struct sk_buff *skb, struct net_device *netdev) /* VLAN tag unrelated when TX */ tag[1] = 0; tag[2] = 0; - tx = YT921X_TAG_PORT_EN | YT921X_TAG_TX_PORTn(port); + tx = FIELD_PREP(YT921X_TAG_TX_PORTS, dsa_xmit_port_mask(skb, netdev)) | + YT921X_TAG_PORT_EN; tag[3] = htons(tx); return skb; From 2c28ee720ad14f58eb88a97ec3efe7c5c315ea5d Mon Sep 17 00:00:00 2001 From: Hangbin Liu Date: Thu, 27 Nov 2025 14:33:10 +0000 Subject: [PATCH 1279/1381] selftests: bonding: add delay before each xvlan_over_bond connectivity check Jakub reported increased flakiness in bond_macvlan_ipvlan.sh on regular kernel, while the tests consistently pass on a debug kernel. This suggests a timing-sensitive issue. To mitigate this, introduce a short sleep before each xvlan_over_bond connectivity check. The delay helps ensure neighbor and route cache have fully converged before verifying connectivity. The sleep interval is kept minimal since check_connection() is invoked nearly 100 times during the test. Fixes: 246af950b940 ("selftests: bonding: add macvlan over bond testing") Reported-by: Jakub Kicinski Closes: https://lore.kernel.org/netdev/20251114082014.750edfad@kernel.org Signed-off-by: Hangbin Liu Link: https://patch.msgid.link/20251127143310.47740-1-liuhangbin@gmail.com Signed-off-by: Jakub Kicinski --- .../testing/selftests/drivers/net/bonding/bond_macvlan_ipvlan.sh | 1 + 1 file changed, 1 insertion(+) diff --git a/tools/testing/selftests/drivers/net/bonding/bond_macvlan_ipvlan.sh b/tools/testing/selftests/drivers/net/bonding/bond_macvlan_ipvlan.sh index c4711272fe45..559f300f965a 100755 --- a/tools/testing/selftests/drivers/net/bonding/bond_macvlan_ipvlan.sh +++ b/tools/testing/selftests/drivers/net/bonding/bond_macvlan_ipvlan.sh @@ -30,6 +30,7 @@ check_connection() local message=${3} RET=0 + sleep 0.25 ip netns exec ${ns} ping ${target} -c 4 -i 0.1 &>/dev/null check_err $? "ping failed" log_test "${bond_mode}/${xvlan_type}_${xvlan_mode}: ${message}" From e5235eb6cfe02a51256013a78f7b28779a7740d5 Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Thu, 27 Nov 2025 07:30:15 -0800 Subject: [PATCH 1280/1381] net: netpoll: initialize work queue before error checks Prevent a kernel warning when netconsole setup fails on devices with IFF_DISABLE_NETPOLL flag. The warning (at kernel/workqueue.c:4242 in __flush_work) occurs because the cleanup path tries to cancel an uninitialized work queue. When __netpoll_setup() encounters a device with IFF_DISABLE_NETPOLL, it fails early and calls skb_pool_flush() for cleanup. This function calls cancel_work_sync(&np->refill_wq), but refill_wq hasn't been initialized yet, triggering the warning. Move INIT_WORK() to the beginning of __netpoll_setup(), ensuring the work queue is properly initialized before any potential failure points. This allows the cleanup path to safely cancel the work queue regardless of where the setup fails. Fixes: 248f6571fd4c5 ("netpoll: Optimize skb refilling on critical path") Signed-off-by: Breno Leitao Link: https://patch.msgid.link/20251127-netpoll_fix_init_work-v1-1-65c07806d736@debian.org Signed-off-by: Jakub Kicinski --- net/core/netpoll.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/core/netpoll.c b/net/core/netpoll.c index 331764845e8f..09f72f10813c 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -554,6 +554,7 @@ int __netpoll_setup(struct netpoll *np, struct net_device *ndev) int err; skb_queue_head_init(&np->skb_pool); + INIT_WORK(&np->refill_wq, refill_skbs_work_handler); if (ndev->priv_flags & IFF_DISABLE_NETPOLL) { np_err(np, "%s doesn't support polling, aborting\n", @@ -591,7 +592,6 @@ int __netpoll_setup(struct netpoll *np, struct net_device *ndev) /* fill up the skb queue */ refill_skbs(np); - INIT_WORK(&np->refill_wq, refill_skbs_work_handler); /* last thing to do is link it to the net device structure */ rcu_assign_pointer(ndev->npinfo, npinfo); From ff736a286116d462a4067ba258fa351bc0b4ed80 Mon Sep 17 00:00:00 2001 From: Thorsten Blum Date: Wed, 26 Nov 2025 23:08:05 +0100 Subject: [PATCH 1281/1381] net: ipconfig: Replace strncpy with strscpy in ic_proto_name strncpy() is deprecated [1] for NUL-terminated destination buffers because it does not guarantee NUL termination. Replace it with strscpy() to ensure the destination buffer is always NUL-terminated and to avoid any additional NUL padding. Although the identifier buffer has 252 usable bytes, strncpy() copied only up to 251 bytes to the zero-initialized buffer, relying on the last byte to act as an implicit NUL terminator. Switching to strscpy() avoids this implicit behavior and does not use magic numbers. The source string is also NUL-terminated and satisfies the __must_be_cstr() requirement of strscpy(). Link: https://www.kernel.org/doc/html/latest/process/deprecated.html#strncpy-on-nul-terminated-strings [1] Signed-off-by: Thorsten Blum Link: https://patch.msgid.link/20251126220804.102160-2-thorsten.blum@linux.dev Signed-off-by: Jakub Kicinski --- net/ipv4/ipconfig.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/ipv4/ipconfig.c b/net/ipv4/ipconfig.c index 22a7889876c1..019408d3ca2c 100644 --- a/net/ipv4/ipconfig.c +++ b/net/ipv4/ipconfig.c @@ -1690,7 +1690,8 @@ static int __init ic_proto_name(char *name) *v = 0; if (kstrtou8(client_id, 0, dhcp_client_identifier)) pr_debug("DHCP: Invalid client identifier type\n"); - strncpy(dhcp_client_identifier + 1, v + 1, 251); + strscpy(dhcp_client_identifier + 1, v + 1, + sizeof(dhcp_client_identifier) - 1); *v = ','; } return 1; From cb2dc6d2869a4fb7ef8d792a81a74bc6f0958a72 Mon Sep 17 00:00:00 2001 From: Oliver Hartkopp Date: Sat, 29 Nov 2025 10:05:00 +0100 Subject: [PATCH 1282/1381] can: Kconfig: select CAN driver infrastructure by default The CAN bus support enabled with CONFIG_CAN provides a socket-based access to CAN interfaces. With the introduction of the latest CAN protocol CAN XL additional configuration status information needs to be exposed to the network layer than formerly provided by standard Linux network drivers. This requires the CAN driver infrastructure to be selected by default. As the CAN network layer can only operate on CAN interfaces anyway all distributions and common default configs enable at least one CAN driver. So selecting CONFIG_CAN_DEV when CONFIG_CAN is selected by the user has no effect on established configurations but solves potential build issues when CONFIG_CAN[_XXX]=y is set together with CANFIG_CAN_DEV=m Fixes: 1a620a723853 ("can: raw: instantly reject unsupported CAN frames") Reported-by: Vincent Mailhol Closes: https://lore.kernel.org/all/CAMZ6RqL_nGszwoLPXn1Li8op-ox4k3Hs6p=Hw6+w0W=DTtobPw@mail.gmail.com/ Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-kbuild-all/202511280531.YnWW2Rxc-lkp@intel.com/ Closes: https://lore.kernel.org/oe-kbuild-all/202511280842.djCQ0N0O-lkp@intel.com/ Closes: https://lore.kernel.org/oe-kbuild-all/202511282325.uVQFRTkA-lkp@intel.com/ Closes: https://lore.kernel.org/oe-kbuild-all/202511291520.guIE1QHj-lkp@intel.com/ Suggested-by: Marc Kleine-Budde Signed-off-by: Oliver Hartkopp Link: https://patch.msgid.link/20251129090500.17484-1-socketcan@hartkopp.net Signed-off-by: Marc Kleine-Budde --- include/linux/can/dev.h | 7 +++++++ net/can/Kconfig | 1 + 2 files changed, 8 insertions(+) diff --git a/include/linux/can/dev.h b/include/linux/can/dev.h index 52c8be5c160e..f6416a56e95d 100644 --- a/include/linux/can/dev.h +++ b/include/linux/can/dev.h @@ -111,7 +111,14 @@ struct net_device *alloc_candev_mqs(int sizeof_priv, unsigned int echo_skb_max, void free_candev(struct net_device *dev); /* a candev safe wrapper around netdev_priv */ +#if IS_ENABLED(CONFIG_CAN_NETLINK) struct can_priv *safe_candev_priv(struct net_device *dev); +#else +static inline struct can_priv *safe_candev_priv(struct net_device *dev) +{ + return NULL; +} +#endif int open_candev(struct net_device *dev); void close_candev(struct net_device *dev); diff --git a/net/can/Kconfig b/net/can/Kconfig index af64a6f76458..e4ccf731a24c 100644 --- a/net/can/Kconfig +++ b/net/can/Kconfig @@ -5,6 +5,7 @@ menuconfig CAN tristate "CAN bus subsystem support" + select CAN_DEV help Controller Area Network (CAN) is a slow (up to 1Mbit/s) serial communications protocol. Development of the CAN bus started in From e0e1b6db2e4b8fae44e222c188d3e96259d00c8e Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Asbj=C3=B8rn=20Sloth=20T=C3=B8nnesen?= Date: Wed, 5 Nov 2025 18:32:12 +0000 Subject: [PATCH 1283/1381] wireguard: netlink: enable strict genetlink validation MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit WireGuard is a modern enough genetlink family, that it doesn't need resv_start_op. It already had policies in place when it was first merged, it has also never used the reserved field, or other things toggled by resv_start_op. wireguard-tools have always used zero initialized memory, and have never touched the reserved field, neither have any other clients I have checked. Closed-source clients are much more likely to use the embeddedable library from wireguard-tools, than a DIY implementation using uninitialized memory. Signed-off-by: Asbjørn Sloth Tønnesen Signed-off-by: Jason A. Donenfeld --- drivers/net/wireguard/netlink.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/net/wireguard/netlink.c b/drivers/net/wireguard/netlink.c index 67f962eb8b46..8adeec6f9440 100644 --- a/drivers/net/wireguard/netlink.c +++ b/drivers/net/wireguard/netlink.c @@ -631,7 +631,6 @@ static const struct genl_ops genl_ops[] = { static struct genl_family genl_family __ro_after_init = { .ops = genl_ops, .n_ops = ARRAY_SIZE(genl_ops), - .resv_start_op = WG_CMD_SET_DEVICE + 1, .name = WG_GENL_NAME, .version = WG_GENL_VERSION, .maxattr = WGDEVICE_A_MAX, From aea199fa157164ed0824d14dddd6c6fa28e130e4 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Asbj=C3=B8rn=20Sloth=20T=C3=B8nnesen?= Date: Wed, 26 Nov 2025 17:35:33 +0000 Subject: [PATCH 1284/1381] wireguard: netlink: validate nested arrays in policy MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Use NLA_POLICY_NESTED_ARRAY() to perform nested array validation in the policy validation step. The nested policy was already enforced through nla_parse_nested(), however extack wasn't passed previously, so no fancy error messages. With the nested attributes being validated directly in the policy, the policy argument can be set to NULL in the calls to nla_parse_nested(). Signed-off-by: Asbjørn Sloth Tønnesen Signed-off-by: Jason A. Donenfeld --- drivers/net/wireguard/netlink.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/drivers/net/wireguard/netlink.c b/drivers/net/wireguard/netlink.c index 8adeec6f9440..97723f9c7998 100644 --- a/drivers/net/wireguard/netlink.c +++ b/drivers/net/wireguard/netlink.c @@ -18,6 +18,8 @@ #include static struct genl_family genl_family; +static const struct nla_policy peer_policy[WGPEER_A_MAX + 1]; +static const struct nla_policy allowedip_policy[WGALLOWEDIP_A_MAX + 1]; static const struct nla_policy device_policy[WGDEVICE_A_MAX + 1] = { [WGDEVICE_A_IFINDEX] = { .type = NLA_U32 }, @@ -27,7 +29,7 @@ static const struct nla_policy device_policy[WGDEVICE_A_MAX + 1] = { [WGDEVICE_A_FLAGS] = NLA_POLICY_MASK(NLA_U32, __WGDEVICE_F_ALL), [WGDEVICE_A_LISTEN_PORT] = { .type = NLA_U16 }, [WGDEVICE_A_FWMARK] = { .type = NLA_U32 }, - [WGDEVICE_A_PEERS] = { .type = NLA_NESTED } + [WGDEVICE_A_PEERS] = NLA_POLICY_NESTED_ARRAY(peer_policy), }; static const struct nla_policy peer_policy[WGPEER_A_MAX + 1] = { @@ -39,7 +41,7 @@ static const struct nla_policy peer_policy[WGPEER_A_MAX + 1] = { [WGPEER_A_LAST_HANDSHAKE_TIME] = NLA_POLICY_EXACT_LEN(sizeof(struct __kernel_timespec)), [WGPEER_A_RX_BYTES] = { .type = NLA_U64 }, [WGPEER_A_TX_BYTES] = { .type = NLA_U64 }, - [WGPEER_A_ALLOWEDIPS] = { .type = NLA_NESTED }, + [WGPEER_A_ALLOWEDIPS] = NLA_POLICY_NESTED_ARRAY(allowedip_policy), [WGPEER_A_PROTOCOL_VERSION] = { .type = NLA_U32 } }; @@ -467,7 +469,7 @@ static int set_peer(struct wg_device *wg, struct nlattr **attrs) nla_for_each_nested(attr, attrs[WGPEER_A_ALLOWEDIPS], rem) { ret = nla_parse_nested(allowedip, WGALLOWEDIP_A_MAX, - attr, allowedip_policy, NULL); + attr, NULL, NULL); if (ret < 0) goto out; ret = set_allowedip(peer, allowedip); @@ -593,7 +595,7 @@ static int wg_set_device(struct sk_buff *skb, struct genl_info *info) nla_for_each_nested(attr, info->attrs[WGDEVICE_A_PEERS], rem) { ret = nla_parse_nested(peer, WGPEER_A_MAX, attr, - peer_policy, NULL); + NULL, NULL); if (ret < 0) goto out; ret = set_peer(wg, peer); From 9755f9de8fac62ed36f1925d6996daf4060d8839 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Asbj=C3=B8rn=20Sloth=20T=C3=B8nnesen?= Date: Wed, 26 Nov 2025 17:35:34 +0000 Subject: [PATCH 1285/1381] wireguard: netlink: use WG_KEY_LEN in policies MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When converting the netlink policies to YNL, the constants used in the policy have to be visible to userspace. As NOISE_*_KEY_LEN isn't visible to userspace, change the policy to use WG_KEY_LEN, as also documented in the UAPI header: $ grep WG_KEY_LEN include/uapi/linux/wireguard.h * WGDEVICE_A_PRIVATE_KEY: NLA_EXACT_LEN, len WG_KEY_LEN * WGDEVICE_A_PUBLIC_KEY: NLA_EXACT_LEN, len WG_KEY_LEN * WGPEER_A_PUBLIC_KEY: NLA_EXACT_LEN, len WG_KEY_LEN * WGPEER_A_PRESHARED_KEY: NLA_EXACT_LEN, len WG_KEY_LEN [...] Add a couple of BUILD_BUG_ON() to ensure that they stay in sync. No behavioural changes intended. Signed-off-by: Asbjørn Sloth Tønnesen Signed-off-by: Jason A. Donenfeld --- drivers/net/wireguard/netlink.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/drivers/net/wireguard/netlink.c b/drivers/net/wireguard/netlink.c index 97723f9c7998..682678d24a9f 100644 --- a/drivers/net/wireguard/netlink.c +++ b/drivers/net/wireguard/netlink.c @@ -24,8 +24,8 @@ static const struct nla_policy allowedip_policy[WGALLOWEDIP_A_MAX + 1]; static const struct nla_policy device_policy[WGDEVICE_A_MAX + 1] = { [WGDEVICE_A_IFINDEX] = { .type = NLA_U32 }, [WGDEVICE_A_IFNAME] = { .type = NLA_NUL_STRING, .len = IFNAMSIZ - 1 }, - [WGDEVICE_A_PRIVATE_KEY] = NLA_POLICY_EXACT_LEN(NOISE_PUBLIC_KEY_LEN), - [WGDEVICE_A_PUBLIC_KEY] = NLA_POLICY_EXACT_LEN(NOISE_PUBLIC_KEY_LEN), + [WGDEVICE_A_PRIVATE_KEY] = NLA_POLICY_EXACT_LEN(WG_KEY_LEN), + [WGDEVICE_A_PUBLIC_KEY] = NLA_POLICY_EXACT_LEN(WG_KEY_LEN), [WGDEVICE_A_FLAGS] = NLA_POLICY_MASK(NLA_U32, __WGDEVICE_F_ALL), [WGDEVICE_A_LISTEN_PORT] = { .type = NLA_U16 }, [WGDEVICE_A_FWMARK] = { .type = NLA_U32 }, @@ -33,8 +33,8 @@ static const struct nla_policy device_policy[WGDEVICE_A_MAX + 1] = { }; static const struct nla_policy peer_policy[WGPEER_A_MAX + 1] = { - [WGPEER_A_PUBLIC_KEY] = NLA_POLICY_EXACT_LEN(NOISE_PUBLIC_KEY_LEN), - [WGPEER_A_PRESHARED_KEY] = NLA_POLICY_EXACT_LEN(NOISE_SYMMETRIC_KEY_LEN), + [WGPEER_A_PUBLIC_KEY] = NLA_POLICY_EXACT_LEN(WG_KEY_LEN), + [WGPEER_A_PRESHARED_KEY] = NLA_POLICY_EXACT_LEN(WG_KEY_LEN), [WGPEER_A_FLAGS] = NLA_POLICY_MASK(NLA_U32, __WGPEER_F_ALL), [WGPEER_A_ENDPOINT] = NLA_POLICY_MIN_LEN(sizeof(struct sockaddr)), [WGPEER_A_PERSISTENT_KEEPALIVE_INTERVAL] = { .type = NLA_U16 }, @@ -643,6 +643,9 @@ static struct genl_family genl_family __ro_after_init = { int __init wg_genetlink_init(void) { + BUILD_BUG_ON(WG_KEY_LEN != NOISE_PUBLIC_KEY_LEN); + BUILD_BUG_ON(WG_KEY_LEN != NOISE_SYMMETRIC_KEY_LEN); + return genl_register_family(&genl_family); } From 73af07d7f2f60914d9ad6b4ec102248196f8ceb9 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Asbj=C3=B8rn=20Sloth=20T=C3=B8nnesen?= Date: Wed, 26 Nov 2025 17:35:35 +0000 Subject: [PATCH 1286/1381] wireguard: netlink: convert to split ops MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This patch converts WireGuard from using the legacy struct genl_ops to struct genl_split_ops, by applying the same transformation as genl_cmd_full_to_split() would otherwise do at runtime. WGDEVICE_A_MAX is swapped for WGDEVICE_A_PEERS, while they are currently equivalent, then .maxattr should be the maximum attribute that a given command supports, and not change along with WGDEVICE_A_MAX. This is an incremental step towards adopting netlink policy code generated by ynl-gen, ensuring that the code and spec is aligned. This is a trivial patch with no behavioural changes intended. Signed-off-by: Asbjørn Sloth Tønnesen Signed-off-by: Jason A. Donenfeld --- drivers/net/wireguard/netlink.c | 16 +++++++++------- 1 file changed, 9 insertions(+), 7 deletions(-) diff --git a/drivers/net/wireguard/netlink.c b/drivers/net/wireguard/netlink.c index 682678d24a9f..e7efe5f8465d 100644 --- a/drivers/net/wireguard/netlink.c +++ b/drivers/net/wireguard/netlink.c @@ -616,28 +616,30 @@ static int wg_set_device(struct sk_buff *skb, struct genl_info *info) return ret; } -static const struct genl_ops genl_ops[] = { +static const struct genl_split_ops wireguard_nl_ops[] = { { .cmd = WG_CMD_GET_DEVICE, .start = wg_get_device_start, .dumpit = wg_get_device_dump, .done = wg_get_device_done, - .flags = GENL_UNS_ADMIN_PERM + .policy = device_policy, + .maxattr = WGDEVICE_A_PEERS, + .flags = GENL_UNS_ADMIN_PERM | GENL_CMD_CAP_DUMP, }, { .cmd = WG_CMD_SET_DEVICE, .doit = wg_set_device, - .flags = GENL_UNS_ADMIN_PERM + .policy = device_policy, + .maxattr = WGDEVICE_A_PEERS, + .flags = GENL_UNS_ADMIN_PERM | GENL_CMD_CAP_DO, } }; static struct genl_family genl_family __ro_after_init = { - .ops = genl_ops, - .n_ops = ARRAY_SIZE(genl_ops), + .split_ops = wireguard_nl_ops, + .n_split_ops = ARRAY_SIZE(wireguard_nl_ops), .name = WG_GENL_NAME, .version = WG_GENL_VERSION, - .maxattr = WGDEVICE_A_MAX, .module = THIS_MODULE, - .policy = device_policy, .netnsok = true }; From b8bcc17f583b87f9913659d69cd796fbc203e510 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Asbj=C3=B8rn=20Sloth=20T=C3=B8nnesen?= Date: Wed, 26 Nov 2025 17:35:36 +0000 Subject: [PATCH 1287/1381] wireguard: netlink: lower .maxattr for WG_CMD_GET_DEVICE MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Previously .maxattr was shared for both WG_CMD_GET_DEVICE and WG_CMD_SET_DEVICE. Now that it is split, then we can lower it for WG_CMD_GET_DEVICE to follow the documentation which defines .maxattr as WGDEVICE_A_IFNAME for WG_CMD_GET_DEVICE. $ grep -hC5 'one but not both of:' include/uapi/linux/wireguard.h * WG_CMD_GET_DEVICE * ----------------- * * May only be called via NLM_F_REQUEST | NLM_F_DUMP. The command * should contain one but not both of: * * WGDEVICE_A_IFINDEX: NLA_U32 * WGDEVICE_A_IFNAME: NLA_NUL_STRING, maxlen IFNAMSIZ - 1 * * The kernel will then return several messages [...] While other attributes weren't rejected previously, the consensus is that nobody sends those attributes, so nothing should break. Link: https://lore.kernel.org/r/aRyLoy2iqbkUipZW@zx2c4.com/ Suggested-by: Jason A. Donenfeld Signed-off-by: Asbjørn Sloth Tønnesen Signed-off-by: Jason A. Donenfeld --- drivers/net/wireguard/netlink.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/wireguard/netlink.c b/drivers/net/wireguard/netlink.c index e7efe5f8465d..c2d0576e96f5 100644 --- a/drivers/net/wireguard/netlink.c +++ b/drivers/net/wireguard/netlink.c @@ -623,7 +623,7 @@ static const struct genl_split_ops wireguard_nl_ops[] = { .dumpit = wg_get_device_dump, .done = wg_get_device_done, .policy = device_policy, - .maxattr = WGDEVICE_A_PEERS, + .maxattr = WGDEVICE_A_IFNAME, .flags = GENL_UNS_ADMIN_PERM | GENL_CMD_CAP_DUMP, }, { .cmd = WG_CMD_SET_DEVICE, From 7adf0efb41fc386daed9f74b9fe8d7e28f0f6f3d Mon Sep 17 00:00:00 2001 From: Colin Ian King Date: Fri, 28 Nov 2025 17:38:02 +0000 Subject: [PATCH 1288/1381] ynl: samples: Fix spelling mistake "failedq" -> "failed" There is a spelling mistake in an error message. Fix it. Signed-off-by: Colin Ian King Link: https://patch.msgid.link/20251128173802.318520-1-colin.i.king@gmail.com Signed-off-by: Jakub Kicinski --- tools/net/ynl/samples/tc-filter-add.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/net/ynl/samples/tc-filter-add.c b/tools/net/ynl/samples/tc-filter-add.c index 1f9cd3f62df6..97871e9e9edc 100644 --- a/tools/net/ynl/samples/tc-filter-add.c +++ b/tools/net/ynl/samples/tc-filter-add.c @@ -207,7 +207,7 @@ static int tc_filter_del(struct ynl_sock *ys, int ifi) req = tc_deltfilter_req_alloc(); if (!req) { - fprintf(stderr, "tc_deltfilter_req_alloc failedq\n"); + fprintf(stderr, "tc_deltfilter_req_alloc failed\n"); return -1; } memset(req, 0, sizeof(*req)); From d3fbfb8b2c4a352bf9167809d2e2e8f8e685d348 Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Fri, 28 Nov 2025 05:11:45 -0800 Subject: [PATCH 1289/1381] net: gianfar: convert to use .get_rx_ring_count Convert the gianfar driver to use the new .get_rx_ring_count ethtool operation instead of implementing .get_rxnfc for handling ETHTOOL_GRXRINGS command. This simplifies the code by removing the ETHTOOL_GRXRINGS case from the switch statement and replacing it with a direct return of the queue count. The driver still maintains .get_rxnfc for other commands including ETHTOOL_GRXCLSRLCNT, ETHTOOL_GRXCLSRULE, and ETHTOOL_GRXCLSRLALL. Signed-off-by: Breno Leitao Reviewed-by: Simon Horman Link: https://patch.msgid.link/20251128-gxring_freescale-v1-1-22a978abf29e@debian.org Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/freescale/gianfar_ethtool.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/freescale/gianfar_ethtool.c b/drivers/net/ethernet/freescale/gianfar_ethtool.c index 5fd1f7327680..6fa752d3b60d 100644 --- a/drivers/net/ethernet/freescale/gianfar_ethtool.c +++ b/drivers/net/ethernet/freescale/gianfar_ethtool.c @@ -1431,6 +1431,13 @@ static int gfar_set_nfc(struct net_device *dev, struct ethtool_rxnfc *cmd) return ret; } +static u32 gfar_get_rx_ring_count(struct net_device *dev) +{ + struct gfar_private *priv = netdev_priv(dev); + + return priv->num_rx_queues; +} + static int gfar_get_nfc(struct net_device *dev, struct ethtool_rxnfc *cmd, u32 *rule_locs) { @@ -1438,9 +1445,6 @@ static int gfar_get_nfc(struct net_device *dev, struct ethtool_rxnfc *cmd, int ret = 0; switch (cmd->cmd) { - case ETHTOOL_GRXRINGS: - cmd->data = priv->num_rx_queues; - break; case ETHTOOL_GRXCLSRLCNT: cmd->rule_cnt = priv->rx_list.count; break; @@ -1519,6 +1523,7 @@ const struct ethtool_ops gfar_ethtool_ops = { #endif .set_rxnfc = gfar_set_nfc, .get_rxnfc = gfar_get_nfc, + .get_rx_ring_count = gfar_get_rx_ring_count, .set_rxfh_fields = gfar_set_rxfh_fields, .get_ts_info = gfar_get_ts_info, .get_link_ksettings = phy_ethtool_get_link_ksettings, From b2d63392690156285e53562ed91b87131c9a97ac Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Fri, 28 Nov 2025 05:11:46 -0800 Subject: [PATCH 1290/1381] net: dpaa2: convert to use .get_rx_ring_count Convert the dpaa2 driver to use the new .get_rx_ring_count ethtool operation instead of implementing .get_rxnfc for handling ETHTOOL_GRXRINGS command. This simplifies the code by removing the ETHTOOL_GRXRINGS case from the switch statement and replacing it with a direct return of the queue count. The driver still maintains .get_rxnfc for other commands including ETHTOOL_GRXCLSRLCNT, ETHTOOL_GRXCLSRULE, and ETHTOOL_GRXCLSRLALL. Signed-off-by: Breno Leitao Reviewed-by: Simon Horman Link: https://patch.msgid.link/20251128-gxring_freescale-v1-2-22a978abf29e@debian.org Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/freescale/dpaa2/dpaa2-ethtool.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-ethtool.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-ethtool.c index 00474ed11d53..baab4f1c908d 100644 --- a/drivers/net/ethernet/freescale/dpaa2/dpaa2-ethtool.c +++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-ethtool.c @@ -711,6 +711,13 @@ static int dpaa2_eth_update_cls_rule(struct net_device *net_dev, return 0; } +static u32 dpaa2_eth_get_rx_ring_count(struct net_device *net_dev) +{ + struct dpaa2_eth_priv *priv = netdev_priv(net_dev); + + return dpaa2_eth_queue_count(priv); +} + static int dpaa2_eth_get_rxnfc(struct net_device *net_dev, struct ethtool_rxnfc *rxnfc, u32 *rule_locs) { @@ -719,9 +726,6 @@ static int dpaa2_eth_get_rxnfc(struct net_device *net_dev, int i, j = 0; switch (rxnfc->cmd) { - case ETHTOOL_GRXRINGS: - rxnfc->data = dpaa2_eth_queue_count(priv); - break; case ETHTOOL_GRXCLSRLCNT: rxnfc->rule_cnt = 0; rxnfc->rule_cnt = dpaa2_eth_num_cls_rules(priv); @@ -949,6 +953,7 @@ const struct ethtool_ops dpaa2_ethtool_ops = { .get_strings = dpaa2_eth_get_strings, .get_rxnfc = dpaa2_eth_get_rxnfc, .set_rxnfc = dpaa2_eth_set_rxnfc, + .get_rx_ring_count = dpaa2_eth_get_rx_ring_count, .get_rxfh_fields = dpaa2_eth_get_rxfh_fields, .set_rxfh_fields = dpaa2_eth_set_rxfh_fields, .get_ts_info = dpaa2_eth_get_ts_info, From ca8df5b877d4b027fcd3968c0519af2b33de4e8a Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Fri, 28 Nov 2025 05:11:47 -0800 Subject: [PATCH 1291/1381] net: enetc: convert to use .get_rx_ring_count Convert the enetc driver to use the new .get_rx_ring_count ethtool operation instead of implementing .get_rxnfc for handling ETHTOOL_GRXRINGS command. This simplifies the code in two ways: 1. For enetc_get_rxnfc(): Remove the ETHTOOL_GRXRINGS case from the switch statement while keeping other cases for classifier rules. 2. For enetc4_get_rxnfc(): Remove it completely and use enetc_get_rxnfc() instead. Now on, enetc_get_rx_ring_count() is the callback that returns the number of RX rings for enetc driver. Also, remove the documentation around enetc4_get_rxnfc(), which was not matching what the function did(?!). Signed-off-by: Breno Leitao Reviewed-by: Simon Horman Link: https://patch.msgid.link/20251128-gxring_freescale-v1-3-22a978abf29e@debian.org Signed-off-by: Jakub Kicinski --- .../ethernet/freescale/enetc/enetc_ethtool.c | 37 ++++++------------- 1 file changed, 11 insertions(+), 26 deletions(-) diff --git a/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c b/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c index 3e222321b937..fed89d4f1e1d 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c +++ b/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c @@ -633,6 +633,13 @@ static int enetc_set_cls_entry(struct enetc_si *si, return enetc_set_fs_entry(si, &rfse, fs->location); } +static u32 enetc_get_rx_ring_count(struct net_device *ndev) +{ + struct enetc_ndev_priv *priv = netdev_priv(ndev); + + return priv->num_rx_rings; +} + static int enetc_get_rxnfc(struct net_device *ndev, struct ethtool_rxnfc *rxnfc, u32 *rule_locs) { @@ -640,9 +647,6 @@ static int enetc_get_rxnfc(struct net_device *ndev, struct ethtool_rxnfc *rxnfc, int i, j; switch (rxnfc->cmd) { - case ETHTOOL_GRXRINGS: - rxnfc->data = priv->num_rx_rings; - break; case ETHTOOL_GRXCLSRLCNT: /* total number of entries */ rxnfc->data = priv->si->num_fs_entries; @@ -681,27 +685,6 @@ static int enetc_get_rxnfc(struct net_device *ndev, struct ethtool_rxnfc *rxnfc, return 0; } -/* i.MX95 ENETC does not support RFS table, but we can use ingress port - * filter table to implement Wake-on-LAN filter or drop the matched flow, - * so the implementation will be different from enetc_get_rxnfc() and - * enetc_set_rxnfc(). Therefore, add enetc4_get_rxnfc() for ENETC v4 PF. - */ -static int enetc4_get_rxnfc(struct net_device *ndev, struct ethtool_rxnfc *rxnfc, - u32 *rule_locs) -{ - struct enetc_ndev_priv *priv = netdev_priv(ndev); - - switch (rxnfc->cmd) { - case ETHTOOL_GRXRINGS: - rxnfc->data = priv->num_rx_rings; - break; - default: - return -EOPNOTSUPP; - } - - return 0; -} - static int enetc_set_rxnfc(struct net_device *ndev, struct ethtool_rxnfc *rxnfc) { struct enetc_ndev_priv *priv = netdev_priv(ndev); @@ -1335,6 +1318,7 @@ const struct ethtool_ops enetc_pf_ethtool_ops = { .get_rmon_stats = enetc_get_rmon_stats, .get_eth_ctrl_stats = enetc_get_eth_ctrl_stats, .get_eth_mac_stats = enetc_get_eth_mac_stats, + .get_rx_ring_count = enetc_get_rx_ring_count, .get_rxnfc = enetc_get_rxnfc, .set_rxnfc = enetc_set_rxnfc, .get_rxfh_key_size = enetc_get_rxfh_key_size, @@ -1363,7 +1347,7 @@ const struct ethtool_ops enetc4_ppm_ethtool_ops = { ETHTOOL_COALESCE_MAX_FRAMES | ETHTOOL_COALESCE_USE_ADAPTIVE_RX, .get_eth_mac_stats = enetc_ppm_get_eth_mac_stats, - .get_rxnfc = enetc4_get_rxnfc, + .get_rx_ring_count = enetc_get_rx_ring_count, .get_rxfh_key_size = enetc_get_rxfh_key_size, .get_rxfh_indir_size = enetc_get_rxfh_indir_size, .get_rxfh = enetc_get_rxfh, @@ -1386,6 +1370,7 @@ const struct ethtool_ops enetc_vf_ethtool_ops = { .get_sset_count = enetc_get_sset_count, .get_strings = enetc_get_strings, .get_ethtool_stats = enetc_get_ethtool_stats, + .get_rx_ring_count = enetc_get_rx_ring_count, .get_rxnfc = enetc_get_rxnfc, .set_rxnfc = enetc_set_rxnfc, .get_rxfh_indir_size = enetc_get_rxfh_indir_size, @@ -1413,7 +1398,7 @@ const struct ethtool_ops enetc4_pf_ethtool_ops = { .set_wol = enetc_set_wol, .get_pauseparam = enetc_get_pauseparam, .set_pauseparam = enetc_set_pauseparam, - .get_rxnfc = enetc4_get_rxnfc, + .get_rx_ring_count = enetc_get_rx_ring_count, .get_rxfh_key_size = enetc_get_rxfh_key_size, .get_rxfh_indir_size = enetc_get_rxfh_indir_size, .get_rxfh = enetc_get_rxfh, From 40d5ce4af2067aab275b8aef690e77455da5314a Mon Sep 17 00:00:00 2001 From: Yeounsu Moon Date: Mon, 1 Dec 2025 07:06:53 +0900 Subject: [PATCH 1292/1381] net: dlink: fix several spelling mistakes in comments This patch fixes multiple spelling mistakes in dl2k driver comments: - "deivices" -> "devices" - "Ttransmit" -> "Transmit" - "catastronphic" -> "catastrophic" - "Extened" -> "Extended" Also fix incorrect unit description: `rx_timeout` uses 640ns increments, not 64ns. - "64ns" -> "640ns" These are comment-only changes and do not affect runtime behavior. Signed-off-by: Yeounsu Moon Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20251130220652.5425-2-yyyynoom@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/dlink/dl2k.c | 8 ++++---- drivers/net/ethernet/dlink/dl2k.h | 2 +- 2 files changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/net/ethernet/dlink/dl2k.c b/drivers/net/ethernet/dlink/dl2k.c index 6e4f17142519..846d58c769ea 100644 --- a/drivers/net/ethernet/dlink/dl2k.c +++ b/drivers/net/ethernet/dlink/dl2k.c @@ -41,7 +41,7 @@ module_param(tx_flow, int, 0); module_param(rx_flow, int, 0); module_param(copy_thresh, int, 0); module_param(rx_coalesce, int, 0); /* Rx frame count each interrupt */ -module_param(rx_timeout, int, 0); /* Rx DMA wait time in 64ns increments */ +module_param(rx_timeout, int, 0); /* Rx DMA wait time in 640ns increments */ module_param(tx_coalesce, int, 0); /* HW xmit count each TxDMAComplete */ @@ -262,7 +262,7 @@ rio_probe1 (struct pci_dev *pdev, const struct pci_device_id *ent) np->link_status = 0; /* Set media and reset PHY */ if (np->phy_media) { - /* default Auto-Negotiation for fiber deivices */ + /* default Auto-Negotiation for fiber devices */ if (np->an_enable == 2) { np->an_enable = 1; } @@ -887,7 +887,7 @@ tx_error (struct net_device *dev, int tx_status) frame_id = (tx_status & 0xffff0000); printk (KERN_ERR "%s: Transmit error, TxStatus %4.4x, FrameId %d.\n", dev->name, tx_status, frame_id); - /* Ttransmit Underrun */ + /* Transmit Underrun */ if (tx_status & 0x10) { dev->stats.tx_fifo_errors++; dw16(TxStartThresh, dr16(TxStartThresh) + 0x10); @@ -1083,7 +1083,7 @@ rio_error (struct net_device *dev, int int_status) get_stats (dev); } - /* PCI Error, a catastronphic error related to the bus interface + /* PCI Error, a catastrophic error related to the bus interface occurs, set GlobalReset and HostReset to reset. */ if (int_status & HostError) { printk (KERN_ERR "%s: HostError! IntStatus %4.4x.\n", diff --git a/drivers/net/ethernet/dlink/dl2k.h b/drivers/net/ethernet/dlink/dl2k.h index 4788cc94639d..9ebf7a6db93e 100644 --- a/drivers/net/ethernet/dlink/dl2k.h +++ b/drivers/net/ethernet/dlink/dl2k.h @@ -270,7 +270,7 @@ enum _pcs_reg { PCS_ESR = 15, }; -/* IEEE Extened Status Register */ +/* IEEE Extended Status Register */ enum _mii_esr { MII_ESR_1000BX_FD = 0x8000, MII_ESR_1000BX_HD = 0x4000, From 4b1639cac04cbc0d19321fc26476ea4f89ff4c8a Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Thu, 27 Nov 2025 16:48:46 -0800 Subject: [PATCH 1293/1381] selftests: net: py: handle interrupt during cleanup Following up on the old discussion [1]. Let the BaseExceptions out of defer()'ed cleanup. And handle it in the main loop. This allows us to exit the tests if user hit Ctrl-C during defer(). Link: https://lore.kernel.org/20251119063228.3adfd743@kernel.org # [1] Reviewed-by: Simon Horman Reviewed-by: Petr Machata Link: https://patch.msgid.link/20251128004846.2602687-1-kuba@kernel.org Signed-off-by: Jakub Kicinski --- tools/testing/selftests/net/lib/py/ksft.py | 18 ++++++++++++++++-- 1 file changed, 16 insertions(+), 2 deletions(-) diff --git a/tools/testing/selftests/net/lib/py/ksft.py b/tools/testing/selftests/net/lib/py/ksft.py index ebd82940ee50..531e7fa1b3ea 100644 --- a/tools/testing/selftests/net/lib/py/ksft.py +++ b/tools/testing/selftests/net/lib/py/ksft.py @@ -163,7 +163,7 @@ def ksft_flush_defer(): entry = global_defer_queue.pop() try: entry.exec_only() - except BaseException: + except Exception: ksft_pr(f"Exception while handling defer / cleanup (callback {i} of {qlen_start})!") tb = traceback.format_exc() for line in tb.strip().split('\n'): @@ -333,7 +333,21 @@ def ksft_run(cases=None, globs=None, case_pfx=None, args=()): KSFT_RESULT = False cnt_key = 'fail' - ksft_flush_defer() + try: + ksft_flush_defer() + except BaseException as e: + tb = traceback.format_exc() + for line in tb.strip().split('\n'): + ksft_pr("Exception|", line) + if isinstance(e, KeyboardInterrupt): + ksft_pr() + ksft_pr("WARN: defer() interrupted, cleanup may be incomplete.") + ksft_pr(" Attempting to finish cleanup before exiting.") + ksft_pr(" Interrupt again to exit immediately.") + ksft_pr() + stop = True + # Flush was interrupted, try to finish the job best we can + ksft_flush_defer() if not cnt_key: cnt_key = 'pass' if KSFT_RESULT else 'fail' From aadff9f766399a350aa3e09fc671617bdc0b365f Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Thu, 27 Nov 2025 11:45:56 -0800 Subject: [PATCH 1294/1381] selftests: net: add a hint about MACAddressPolicy=persistent New NIPA installation had been reporting a few flaky tests. arp_ndisc_evict_nocarrier is most flaky of them all. I suspect that the flakiness is due to udev swapping the MAC addresses on the interfaces. Extend the message in arp_ndisc_evict_nocarrier to hint at this potential issue. Having the neigh get fail right after ping is rather unusual, unless udev changes the MAC addr causing a flush in the meantime. Link: https://patch.msgid.link/20251127194556.2409574-1-kuba@kernel.org Signed-off-by: Jakub Kicinski --- tools/testing/selftests/net/arp_ndisc_evict_nocarrier.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/testing/selftests/net/arp_ndisc_evict_nocarrier.sh b/tools/testing/selftests/net/arp_ndisc_evict_nocarrier.sh index 92eb880c52f2..00758f00efbf 100755 --- a/tools/testing/selftests/net/arp_ndisc_evict_nocarrier.sh +++ b/tools/testing/selftests/net/arp_ndisc_evict_nocarrier.sh @@ -75,7 +75,7 @@ setup_v4() { ip neigh get $V4_ADDR1 dev veth0 >/dev/null 2>&1 if [ $? -ne 0 ]; then cleanup_v4 - echo "failed" + echo "failed; is the system using MACAddressPolicy=persistent ?" exit 1 fi From 09339d0d83108fd4f59b94677ac99bde870aa0ae Mon Sep 17 00:00:00 2001 From: Alok Tiwari Date: Fri, 28 Nov 2025 00:52:57 -0800 Subject: [PATCH 1295/1381] l2tp: correct debugfs label for tunnel tx stats l2tp_dfs_seq_tunnel_show prints two groups of tunnel statistics. The first group reports transmit counters, but the code labels it as rx. Set the label to "tx" so the debugfs output reflects the actual meaning. Signed-off-by: Alok Tiwari Reviewed-by: Simon Horman Link: https://patch.msgid.link/20251128085300.3377210-1-alok.a.tiwari@oracle.com Signed-off-by: Jakub Kicinski --- net/l2tp/l2tp_debugfs.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/l2tp/l2tp_debugfs.c b/net/l2tp/l2tp_debugfs.c index 2d0c8275a3a8..5cfaab7d0890 100644 --- a/net/l2tp/l2tp_debugfs.c +++ b/net/l2tp/l2tp_debugfs.c @@ -163,7 +163,7 @@ static void l2tp_dfs_seq_tunnel_show(struct seq_file *m, void *v) seq_printf(m, " %d sessions, refcnt %d/%d\n", session_count, tunnel->sock ? refcount_read(&tunnel->sock->sk_refcnt) : 0, refcount_read(&tunnel->ref_count)); - seq_printf(m, " %08x rx %ld/%ld/%ld rx %ld/%ld/%ld\n", + seq_printf(m, " %08x tx %ld/%ld/%ld rx %ld/%ld/%ld\n", 0, atomic_long_read(&tunnel->stats.tx_packets), atomic_long_read(&tunnel->stats.tx_bytes), From ea2d3befcf29cca5221d0572d54e4d71b683175d Mon Sep 17 00:00:00 2001 From: David Yang Date: Sat, 29 Nov 2025 12:21:34 +0800 Subject: [PATCH 1296/1381] net: dsa: yt921x: Set ageing_time_min/ageing_time_max The ageing time is in 5s step, ranging from 1 step to 0xffff steps, so add appropriate attributes. Signed-off-by: David Yang Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20251129042137.3034032-1-mmyangfl@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/yt921x.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/dsa/yt921x.c b/drivers/net/dsa/yt921x.c index ebfd34f72314..d5fb17d2b6e0 100644 --- a/drivers/net/dsa/yt921x.c +++ b/drivers/net/dsa/yt921x.c @@ -2855,6 +2855,8 @@ static int yt921x_mdio_probe(struct mdio_device *mdiodev) ds->assisted_learning_on_cpu_port = true; ds->priv = priv; ds->ops = &yt921x_dsa_switch_ops; + ds->ageing_time_min = 1 * 5000; + ds->ageing_time_max = U16_MAX * 5000; ds->phylink_mac_ops = &yt921x_phylink_mac_ops; ds->num_ports = YT921X_PORT_NUM; From e3b8cbf40c6e60a7a935bd8980884d5741a7a77b Mon Sep 17 00:00:00 2001 From: Andre Carvalho Date: Sat, 29 Nov 2025 12:24:19 +0000 Subject: [PATCH 1297/1381] selftests: netconsole: remove log noise due to socat exit This removes some noise that can be distracting while looking at selftests by redirecting socat stderr to /dev/null. Before this commit, netcons_basic would output: Running with target mode: basic (ipv6) 2025/11/29 12:08:03 socat[259] W exiting on signal 15 2025/11/29 12:08:03 socat[271] W exiting on signal 15 basic : ipv6 : Test passed Running with target mode: basic (ipv4) 2025/11/29 12:08:05 socat[329] W exiting on signal 15 2025/11/29 12:08:05 socat[322] W exiting on signal 15 basic : ipv4 : Test passed Running with target mode: extended (ipv6) 2025/11/29 12:08:08 socat[386] W exiting on signal 15 2025/11/29 12:08:08 socat[386] W exiting on signal 15 2025/11/29 12:08:08 socat[380] W exiting on signal 15 extended : ipv6 : Test passed Running with target mode: extended (ipv4) 2025/11/29 12:08:10 socat[440] W exiting on signal 15 2025/11/29 12:08:10 socat[435] W exiting on signal 15 2025/11/29 12:08:10 socat[435] W exiting on signal 15 extended : ipv4 : Test passed After these changes, output looks like: Running with target mode: basic (ipv6) basic : ipv6 : Test passed Running with target mode: basic (ipv4) basic : ipv4 : Test passed Running with target mode: extended (ipv6) extended : ipv6 : Test passed Running with target mode: extended (ipv4) extended : ipv4 : Test passed Signed-off-by: Andre Carvalho Reviewed-by: Simon Horman Link: https://patch.msgid.link/20251129-netcons-socat-noise-v1-1-605a0cea8fca@gmail.com Signed-off-by: Jakub Kicinski --- tools/testing/selftests/drivers/net/lib/sh/lib_netcons.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/testing/selftests/drivers/net/lib/sh/lib_netcons.sh b/tools/testing/selftests/drivers/net/lib/sh/lib_netcons.sh index 87f89fd92f8c..ae8abff4be40 100644 --- a/tools/testing/selftests/drivers/net/lib/sh/lib_netcons.sh +++ b/tools/testing/selftests/drivers/net/lib/sh/lib_netcons.sh @@ -249,7 +249,7 @@ function listen_port_and_save_to() { # Just wait for 2 seconds timeout 2 ip netns exec "${NAMESPACE}" \ - socat "${SOCAT_MODE}":"${PORT}",fork "${OUTPUT}" + socat "${SOCAT_MODE}":"${PORT}",fork "${OUTPUT}" 2> /dev/null } # Only validate that the message arrived properly From 484f1176896e85b5ae4f04403c8cd5b79aced95e Mon Sep 17 00:00:00 2001 From: Zhangchao Zhang Date: Fri, 26 Sep 2025 18:10:46 +0800 Subject: [PATCH 1298/1381] Bluetooth: mediatek: add gpio pin to reset bt Support the platform Bluetooth to be reset by hardware pin, when a Bluetooth exception occurs, attempt to reset the Bluetooth module using the hardware reset pin, as this method is generally more stable and reliable than a software reset. If the hardware reset pin is not specified in the device tree, fall back to the existing software reset mechanism to ensure backward compatibility. Co-developed: Sean Wang Co-developed: Hao Qin Co-developed: Chris Lu Signed-off-by: Zhangchao Zhang Reviewed-by: AngeloGioacchino Del Regno Signed-off-by: Luiz Augusto von Dentz --- drivers/bluetooth/btusb.c | 13 +++++++++++++ 1 file changed, 13 insertions(+) diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c index fa683bb7f0b4..9bfa55cb7acf 100644 --- a/drivers/bluetooth/btusb.c +++ b/drivers/bluetooth/btusb.c @@ -2808,6 +2808,19 @@ static int btusb_mtk_reset(struct hci_dev *hdev, void *rst_data) btusb_stop_traffic(data); usb_kill_anchored_urbs(&data->tx_anchor); + /* Toggle the hard reset line. The MediaTek device is going to + * yank itself off the USB and then replug. The cleanup is handled + * correctly on the way out (standard USB disconnect), and the new + * device is detected cleanly and bound to the driver again like + * it should be. + */ + if (data->reset_gpio) { + gpiod_set_value_cansleep(data->reset_gpio, 1); + msleep(200); + gpiod_set_value_cansleep(data->reset_gpio, 0); + return 0; + } + err = btmtk_usb_subsys_reset(hdev, btmtk_data->dev_id); usb_queue_reset_device(data->intf); From 01622e9a5336015f657df423b700e488fbcd2f39 Mon Sep 17 00:00:00 2001 From: Javier Nieto Date: Mon, 29 Sep 2025 15:14:41 -0700 Subject: [PATCH 1299/1381] Bluetooth: hci_h5: avoid sending two SYNC messages Previously, h5_open() called h5_link_control() to send a SYNC message. But h5_link_control() only enqueues the packet and requires the caller to call hci_uart_tx_wakeup(). Thus, after H5_SYNC_TIMEOUT ran out (100ms), h5_timed_event() would be called and, realizing that the state was still H5_UNINITIALIZED, it would re-enqueue the SYNC and call hci_uart_tx_wakeup(). Consequently, two SYNC packets would be sent and initialization would unnecessarily wait for 100ms. The naive solution of calling hci_uart_tx_wakeup() in h5_open() does not work because it will only schedule tx work if the HCI_PROTO_READY bit is set and hci_serdev only sets it after h5_open() returns. This patch removes the extraneous SYNC being enqueued and makes h5_timed_event() wake up on the next jiffy. Signed-off-by: Javier Nieto Signed-off-by: Luiz Augusto von Dentz --- drivers/bluetooth/hci_h5.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/drivers/bluetooth/hci_h5.c b/drivers/bluetooth/hci_h5.c index d0d4420c1a0f..863ee93dd8a8 100644 --- a/drivers/bluetooth/hci_h5.c +++ b/drivers/bluetooth/hci_h5.c @@ -213,7 +213,6 @@ static void h5_peer_reset(struct hci_uart *hu) static int h5_open(struct hci_uart *hu) { struct h5 *h5; - const unsigned char sync[] = { 0x01, 0x7e }; BT_DBG("hu %p", hu); @@ -243,9 +242,11 @@ static int h5_open(struct hci_uart *hu) set_bit(HCI_UART_INIT_PENDING, &hu->hdev_flags); - /* Send initial sync request */ - h5_link_control(hu, sync, sizeof(sync)); - mod_timer(&h5->timer, jiffies + H5_SYNC_TIMEOUT); + /* + * Wait one jiffy because the UART layer won't set HCI_UART_PROTO_READY, + * which allows us to send link packets, until this function returns. + */ + mod_timer(&h5->timer, jiffies + 1); return 0; } From 97fdb2ea06efdde784480934d1ff60eb21e88b58 Mon Sep 17 00:00:00 2001 From: Javier Nieto Date: Mon, 29 Sep 2025 15:59:21 -0700 Subject: [PATCH 1300/1381] Bluetooth: hci_h5: implement CRC data integrity The UART-based H5 protocol supports CRC data integrity checks for reliable packets. The host sets bit 5 in the configuration field of the CONFIG link control message to indicate that CRC is supported. The controller sets the same bit in the CONFIG RESPONSE message to indicate that CRC may be used from then on. Tested on a MangoPi MQ-Pro with a Realtek RTL8723DS Bluetooth controller using the tip of the bluetooth-next tree. Signed-off-by: Javier Nieto Signed-off-by: Luiz Augusto von Dentz --- drivers/bluetooth/Kconfig | 1 + drivers/bluetooth/hci_h5.c | 42 ++++++++++++++++++++++++++++++++++---- 2 files changed, 39 insertions(+), 4 deletions(-) diff --git a/drivers/bluetooth/Kconfig b/drivers/bluetooth/Kconfig index 7df69ccb6600..c5d45cf91f88 100644 --- a/drivers/bluetooth/Kconfig +++ b/drivers/bluetooth/Kconfig @@ -188,6 +188,7 @@ config BT_HCIUART_3WIRE bool "Three-wire UART (H5) protocol support" depends on BT_HCIUART depends on BT_HCIUART_SERDEV + select CRC_CCITT help The HCI Three-wire UART Transport Layer makes it possible to user the Bluetooth HCI over a serial port interface. The HCI diff --git a/drivers/bluetooth/hci_h5.c b/drivers/bluetooth/hci_h5.c index 863ee93dd8a8..43371f389fd2 100644 --- a/drivers/bluetooth/hci_h5.c +++ b/drivers/bluetooth/hci_h5.c @@ -7,6 +7,8 @@ */ #include +#include +#include #include #include #include @@ -58,6 +60,7 @@ enum { H5_TX_ACK_REQ, /* Pending ack to send */ H5_WAKEUP_DISABLE, /* Device cannot wake host */ H5_HW_FLOW_CONTROL, /* Use HW flow control */ + H5_CRC, /* Use CRC */ }; struct h5 { @@ -141,8 +144,8 @@ static void h5_link_control(struct hci_uart *hu, const void *data, size_t len) static u8 h5_cfg_field(struct h5 *h5) { - /* Sliding window size (first 3 bits) */ - return h5->tx_win & 0x07; + /* Sliding window size (first 3 bits) and CRC request (fifth bit). */ + return (h5->tx_win & 0x07) | 0x10; } static void h5_timed_event(struct timer_list *t) @@ -361,8 +364,10 @@ static void h5_handle_internal_rx(struct hci_uart *hu) h5_link_control(hu, conf_rsp, 2); h5_link_control(hu, conf_req, 3); } else if (memcmp(data, conf_rsp, 2) == 0) { - if (H5_HDR_LEN(hdr) > 2) + if (H5_HDR_LEN(hdr) > 2) { h5->tx_win = (data[2] & 0x07); + assign_bit(H5_CRC, &h5->flags, data[2] & 0x10); + } BT_DBG("Three-wire init complete. tx_win %u", h5->tx_win); h5->state = H5_ACTIVE; hci_uart_init_ready(hu); @@ -426,7 +431,24 @@ static void h5_complete_rx_pkt(struct hci_uart *hu) static int h5_rx_crc(struct hci_uart *hu, unsigned char c) { - h5_complete_rx_pkt(hu); + struct h5 *h5 = hu->priv; + const unsigned char *hdr = h5->rx_skb->data; + u16 crc; + __be16 crc_be; + + crc = crc_ccitt(0xffff, hdr, 4 + H5_HDR_LEN(hdr)); + crc = bitrev16(crc); + + crc_be = cpu_to_be16(crc); + + if (memcmp(&crc_be, hdr + 4 + H5_HDR_LEN(hdr), 2) != 0) { + bt_dev_err(hu->hdev, "Received packet with invalid CRC"); + h5_reset_rx(h5); + } else { + /* Remove CRC bytes */ + skb_trim(h5->rx_skb, 4 + H5_HDR_LEN(hdr)); + h5_complete_rx_pkt(hu); + } return 0; } @@ -557,6 +579,7 @@ static void h5_reset_rx(struct h5 *h5) h5->rx_func = h5_rx_delimiter; h5->rx_pending = 0; clear_bit(H5_RX_ESC, &h5->flags); + clear_bit(H5_CRC, &h5->flags); } static int h5_recv(struct hci_uart *hu, const void *data, int count) @@ -687,6 +710,7 @@ static struct sk_buff *h5_prepare_pkt(struct hci_uart *hu, u8 pkt_type, struct h5 *h5 = hu->priv; struct sk_buff *nskb; u8 hdr[4]; + u16 crc; int i; if (!valid_packet_type(pkt_type)) { @@ -714,6 +738,7 @@ static struct sk_buff *h5_prepare_pkt(struct hci_uart *hu, u8 pkt_type, /* Reliable packet? */ if (pkt_type == HCI_ACLDATA_PKT || pkt_type == HCI_COMMAND_PKT) { hdr[0] |= 1 << 7; + hdr[0] |= (test_bit(H5_CRC, &h5->flags) && 1) << 6; hdr[0] |= h5->tx_seq; h5->tx_seq = (h5->tx_seq + 1) % 8; } @@ -733,6 +758,15 @@ static struct sk_buff *h5_prepare_pkt(struct hci_uart *hu, u8 pkt_type, for (i = 0; i < len; i++) h5_slip_one_byte(nskb, data[i]); + if (H5_HDR_CRC(hdr)) { + crc = crc_ccitt(0xffff, hdr, 4); + crc = crc_ccitt(crc, data, len); + crc = bitrev16(crc); + + h5_slip_one_byte(nskb, (crc >> 8) & 0xff); + h5_slip_one_byte(nskb, crc & 0xff); + } + h5_slip_delim(nskb); return nskb; From 33b2835f0b7e2a458473b0e3a23b54b92108b6b0 Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Tue, 2 Sep 2025 11:11:40 -0400 Subject: [PATCH 1301/1381] Bluetooth: HCI: Add initial support for PAST This adds PAST related commands (HCI_OP_LE_PAST, HCI_OP_LE_PAST_SET_INFO and HCI_OP_LE_PAST_PARAMS) and events (HCI_EV_LE_PAST_RECEIVED) along with handling of PAST sender and receiver features bits including new MGMG settings ( HCI_EV_LE_PAST_RECEIVED and MGMT_SETTING_PAST_RECEIVER) which userspace can use to determine if PAST is supported by the controller. Signed-off-by: Luiz Augusto von Dentz --- include/net/bluetooth/hci.h | 54 ++++++++++++++++++++++ include/net/bluetooth/hci_core.h | 12 +++++ include/net/bluetooth/mgmt.h | 2 + net/bluetooth/hci_event.c | 79 ++++++++++++++++++++++++++++---- net/bluetooth/hci_sync.c | 3 ++ net/bluetooth/iso.c | 25 ++++++++++ net/bluetooth/mgmt.c | 12 +++++ 7 files changed, 177 insertions(+), 10 deletions(-) diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h index cb4c02d00759..d883ad233ebc 100644 --- a/include/net/bluetooth/hci.h +++ b/include/net/bluetooth/hci.h @@ -647,6 +647,8 @@ enum { #define HCI_LE_EXT_ADV 0x10 #define HCI_LE_PERIODIC_ADV 0x20 #define HCI_LE_CHAN_SEL_ALG2 0x40 +#define HCI_LE_PAST_SENDER 0x01 +#define HCI_LE_PAST_RECEIVER 0x02 #define HCI_LE_CIS_CENTRAL 0x10 #define HCI_LE_CIS_PERIPHERAL 0x20 #define HCI_LE_ISO_BROADCASTER 0x40 @@ -2068,6 +2070,44 @@ struct hci_cp_le_set_privacy_mode { __u8 mode; } __packed; +#define HCI_OP_LE_PAST 0x205a +struct hci_cp_le_past { + __le16 handle; + __le16 service_data; + __le16 sync_handle; +} __packed; + +struct hci_rp_le_past { + __u8 status; + __le16 handle; +} __packed; + +#define HCI_OP_LE_PAST_SET_INFO 0x205b +struct hci_cp_le_past_set_info { + __le16 handle; + __le16 service_data; + __u8 adv_handle; +} __packed; + +struct hci_rp_le_past_set_info { + __u8 status; + __le16 handle; +} __packed; + +#define HCI_OP_LE_PAST_PARAMS 0x205c +struct hci_cp_le_past_params { + __le16 handle; + __u8 mode; + __le16 skip; + __le16 sync_timeout; + __u8 cte_type; +} __packed; + +struct hci_rp_le_past_params { + __u8 status; + __le16 handle; +} __packed; + #define HCI_OP_LE_READ_BUFFER_SIZE_V2 0x2060 struct hci_rp_le_read_buffer_size_v2 { __u8 status; @@ -2800,6 +2840,20 @@ struct hci_evt_le_ext_adv_set_term { __u8 num_evts; } __packed; +#define HCI_EV_LE_PAST_RECEIVED 0x18 +struct hci_ev_le_past_received { + __u8 status; + __le16 handle; + __le16 service_data; + __le16 sync_handle; + __u8 sid; + __u8 bdaddr_type; + bdaddr_t bdaddr; + __u8 phy; + __le16 interval; + __u8 clock_accuracy; +} __packed; + #define HCI_EVT_LE_CIS_ESTABLISHED 0x19 struct hci_evt_le_cis_established { __u8 status; diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index 0cb87687837f..1bd12c303e25 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -2053,6 +2053,18 @@ void hci_conn_del_sysfs(struct hci_conn *conn); #define sync_recv_capable(dev) \ ((dev)->le_features[3] & HCI_LE_ISO_SYNC_RECEIVER) #define sync_recv_enabled(dev) (le_enabled(dev) && sync_recv_capable(dev)) +#define past_sender_capable(dev) \ + ((dev)->le_features[3] & HCI_LE_PAST_SENDER) +#define past_receiver_capable(dev) \ + ((dev)->le_features[3] & HCI_LE_PAST_RECEIVER) +#define past_capable(dev) \ + (past_sender_capable(dev) || past_receiver_capable(dev)) +#define past_sender_enabled(dev) \ + (le_enabled(dev) && past_sender_capable(dev)) +#define past_receiver_enabled(dev) \ + (le_enabled(dev) && past_receiver_capable(dev)) +#define past_enabled(dev) \ + (past_sender_enabled(dev) || past_receiver_enabled(dev)) #define mws_transport_config_capable(dev) (((dev)->commands[30] & 0x08) && \ (!hci_test_quirk((dev), HCI_QUIRK_BROKEN_MWS_TRANSPORT_CONFIG))) diff --git a/include/net/bluetooth/mgmt.h b/include/net/bluetooth/mgmt.h index f5be96f08b9d..8234915854b6 100644 --- a/include/net/bluetooth/mgmt.h +++ b/include/net/bluetooth/mgmt.h @@ -119,6 +119,8 @@ struct mgmt_rp_read_index_list { #define MGMT_SETTING_ISO_BROADCASTER BIT(20) #define MGMT_SETTING_ISO_SYNC_RECEIVER BIT(21) #define MGMT_SETTING_LL_PRIVACY BIT(22) +#define MGMT_SETTING_PAST_SENDER BIT(23) +#define MGMT_SETTING_PAST_RECEIVER BIT(24) #define MGMT_OP_READ_INFO 0x0004 #define MGMT_READ_INFO_SIZE 0 diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c index 3838b90343d9..af34c9938509 100644 --- a/net/bluetooth/hci_event.c +++ b/net/bluetooth/hci_event.c @@ -5936,6 +5936,71 @@ static void hci_le_ext_adv_term_evt(struct hci_dev *hdev, void *data, hci_dev_unlock(hdev); } +static int hci_le_pa_term_sync(struct hci_dev *hdev, __le16 handle) +{ + struct hci_cp_le_pa_term_sync cp; + + memset(&cp, 0, sizeof(cp)); + cp.handle = handle; + + return hci_send_cmd(hdev, HCI_OP_LE_PA_TERM_SYNC, sizeof(cp), &cp); +} + +static void hci_le_past_received_evt(struct hci_dev *hdev, void *data, + struct sk_buff *skb) +{ + struct hci_ev_le_past_received *ev = data; + int mask = hdev->link_mode; + __u8 flags = 0; + struct hci_conn *pa_sync, *conn; + + bt_dev_dbg(hdev, "status 0x%2.2x", ev->status); + + hci_dev_lock(hdev); + + hci_dev_clear_flag(hdev, HCI_PA_SYNC); + + conn = hci_conn_hash_lookup_create_pa_sync(hdev); + if (!conn) { + bt_dev_err(hdev, + "Unable to find connection for dst %pMR sid 0x%2.2x", + &ev->bdaddr, ev->sid); + goto unlock; + } + + conn->sync_handle = le16_to_cpu(ev->sync_handle); + conn->sid = HCI_SID_INVALID; + + mask |= hci_proto_connect_ind(hdev, &ev->bdaddr, PA_LINK, + &flags); + if (!(mask & HCI_LM_ACCEPT)) { + hci_le_pa_term_sync(hdev, ev->sync_handle); + goto unlock; + } + + if (!(flags & HCI_PROTO_DEFER)) + goto unlock; + + /* Add connection to indicate PA sync event */ + pa_sync = hci_conn_add_unset(hdev, PA_LINK, BDADDR_ANY, + HCI_ROLE_SLAVE); + + if (IS_ERR(pa_sync)) + goto unlock; + + pa_sync->sync_handle = le16_to_cpu(ev->sync_handle); + + if (ev->status) { + set_bit(HCI_CONN_PA_SYNC_FAILED, &pa_sync->flags); + + /* Notify iso layer */ + hci_connect_cfm(pa_sync, ev->status); + } + +unlock: + hci_dev_unlock(hdev); +} + static void hci_le_conn_update_complete_evt(struct hci_dev *hdev, void *data, struct sk_buff *skb) { @@ -6412,16 +6477,6 @@ static void hci_le_ext_adv_report_evt(struct hci_dev *hdev, void *data, hci_dev_unlock(hdev); } -static int hci_le_pa_term_sync(struct hci_dev *hdev, __le16 handle) -{ - struct hci_cp_le_pa_term_sync cp; - - memset(&cp, 0, sizeof(cp)); - cp.handle = handle; - - return hci_send_cmd(hdev, HCI_OP_LE_PA_TERM_SYNC, sizeof(cp), &cp); -} - static void hci_le_pa_sync_established_evt(struct hci_dev *hdev, void *data, struct sk_buff *skb) { @@ -7206,6 +7261,10 @@ static const struct hci_le_ev { /* [0x12 = HCI_EV_LE_EXT_ADV_SET_TERM] */ HCI_LE_EV(HCI_EV_LE_EXT_ADV_SET_TERM, hci_le_ext_adv_term_evt, sizeof(struct hci_evt_le_ext_adv_set_term)), + /* [0x18 = HCI_EVT_LE_PAST_RECEIVED] */ + HCI_LE_EV(HCI_EV_LE_PAST_RECEIVED, + hci_le_past_received_evt, + sizeof(struct hci_ev_le_past_received)), /* [0x19 = HCI_EVT_LE_CIS_ESTABLISHED] */ HCI_LE_EV(HCI_EVT_LE_CIS_ESTABLISHED, hci_le_cis_established_evt, sizeof(struct hci_evt_le_cis_established)), diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c index 6e76798ec786..54ce93236551 100644 --- a/net/bluetooth/hci_sync.c +++ b/net/bluetooth/hci_sync.c @@ -4393,6 +4393,9 @@ static int hci_le_set_event_mask_sync(struct hci_dev *hdev) if (ext_adv_capable(hdev)) events[2] |= 0x02; /* LE Advertising Set Terminated */ + if (past_receiver_capable(hdev)) + events[2] |= 0x80; /* LE PAST Received */ + if (cis_capable(hdev)) { events[3] |= 0x01; /* LE CIS Established */ if (cis_peripheral_capable(hdev)) diff --git a/net/bluetooth/iso.c b/net/bluetooth/iso.c index 5859ec1c04dd..d0a79f601e8d 100644 --- a/net/bluetooth/iso.c +++ b/net/bluetooth/iso.c @@ -80,6 +80,7 @@ static struct bt_iso_qos default_qos; static bool check_ucast_qos(struct bt_iso_qos *qos); static bool check_bcast_qos(struct bt_iso_qos *qos); static bool iso_match_sid(struct sock *sk, void *data); +static bool iso_match_sid_past(struct sock *sk, void *data); static bool iso_match_sync_handle(struct sock *sk, void *data); static bool iso_match_sync_handle_pa_report(struct sock *sk, void *data); static void iso_sock_disconn(struct sock *sk); @@ -2090,6 +2091,16 @@ static bool iso_match_sid(struct sock *sk, void *data) return ev->sid == iso_pi(sk)->bc_sid; } +static bool iso_match_sid_past(struct sock *sk, void *data) +{ + struct hci_ev_le_past_received *ev = data; + + if (iso_pi(sk)->bc_sid == HCI_SID_INVALID) + return true; + + return ev->sid == iso_pi(sk)->bc_sid; +} + static bool iso_match_sync_handle(struct sock *sk, void *data) { struct hci_evt_le_big_info_adv_report *ev = data; @@ -2109,6 +2120,7 @@ static bool iso_match_sync_handle_pa_report(struct sock *sk, void *data) int iso_connect_ind(struct hci_dev *hdev, bdaddr_t *bdaddr, __u8 *flags) { struct hci_ev_le_pa_sync_established *ev1; + struct hci_ev_le_past_received *ev1a; struct hci_evt_le_big_info_adv_report *ev2; struct hci_ev_le_per_adv_report *ev3; struct sock *sk; @@ -2122,6 +2134,7 @@ int iso_connect_ind(struct hci_dev *hdev, bdaddr_t *bdaddr, __u8 *flags) * SID to listen to and once sync is established its handle needs to * be stored in iso_pi(sk)->sync_handle so it can be matched once * receiving the BIG Info. + * 1a. HCI_EV_LE_PAST_RECEIVED: alternative to 1. * 2. HCI_EVT_LE_BIG_INFO_ADV_REPORT: When connect_ind is triggered by a * a BIG Info it attempts to check if there any listening socket with * the same sync_handle and if it does then attempt to create a sync. @@ -2141,6 +2154,18 @@ int iso_connect_ind(struct hci_dev *hdev, bdaddr_t *bdaddr, __u8 *flags) goto done; } + ev1a = hci_recv_event_data(hdev, HCI_EV_LE_PAST_RECEIVED); + if (ev1a) { + sk = iso_get_sock(&hdev->bdaddr, bdaddr, BT_LISTEN, + iso_match_sid_past, ev1a); + if (sk && !ev1a->status) { + iso_pi(sk)->sync_handle = le16_to_cpu(ev1a->sync_handle); + iso_pi(sk)->bc_sid = ev1a->sid; + } + + goto done; + } + ev2 = hci_recv_event_data(hdev, HCI_EVT_LE_BIG_INFO_ADV_REPORT); if (ev2) { /* Check if BIGInfo report has already been handled */ diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c index 262bf984d2aa..7dbd3d57e66c 100644 --- a/net/bluetooth/mgmt.c +++ b/net/bluetooth/mgmt.c @@ -852,6 +852,12 @@ static u32 get_supported_settings(struct hci_dev *hdev) if (ll_privacy_capable(hdev)) settings |= MGMT_SETTING_LL_PRIVACY; + if (past_sender_capable(hdev)) + settings |= MGMT_SETTING_PAST_SENDER; + + if (past_receiver_capable(hdev)) + settings |= MGMT_SETTING_PAST_RECEIVER; + settings |= MGMT_SETTING_PHY_CONFIGURATION; return settings; @@ -937,6 +943,12 @@ static u32 get_current_settings(struct hci_dev *hdev) if (ll_privacy_enabled(hdev)) settings |= MGMT_SETTING_LL_PRIVACY; + if (past_sender_enabled(hdev)) + settings |= MGMT_SETTING_PAST_SENDER; + + if (past_receiver_enabled(hdev)) + settings |= MGMT_SETTING_PAST_RECEIVER; + return settings; } From c530569adc19b5f0c62955de41f067bad34e3fe0 Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Tue, 2 Sep 2025 11:14:28 -0400 Subject: [PATCH 1302/1381] Bluetooth: hci_core: Introduce HCI_CONN_FLAG_PAST This introduces a new device flag so userspace can indicate if it wants to enable PAST Receiver for a specific device. Signed-off-by: Luiz Augusto von Dentz --- include/net/bluetooth/hci_core.h | 1 + net/bluetooth/hci_sync.c | 55 +++++++++++++++++++++++++++++++- 2 files changed, 55 insertions(+), 1 deletion(-) diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index 1bd12c303e25..8c2235444808 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -166,6 +166,7 @@ enum hci_conn_flags { HCI_CONN_FLAG_REMOTE_WAKEUP = BIT(0), HCI_CONN_FLAG_DEVICE_PRIVACY = BIT(1), HCI_CONN_FLAG_ADDRESS_RESOLUTION = BIT(2), + HCI_CONN_FLAG_PAST = BIT(3), }; typedef u8 hci_conn_flags_t; diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c index 54ce93236551..ba6f13e9235c 100644 --- a/net/bluetooth/hci_sync.c +++ b/net/bluetooth/hci_sync.c @@ -4324,6 +4324,10 @@ static int hci_le_set_event_mask_sync(struct hci_dev *hdev) if (ll_privacy_capable(hdev)) hdev->conn_flags |= HCI_CONN_FLAG_ADDRESS_RESOLUTION; + /* Mark PAST if supported */ + if (past_capable(hdev)) + hdev->conn_flags |= HCI_CONN_FLAG_PAST; + /* If the controller supports Extended Scanner Filter * Policies, enable the corresponding event. */ @@ -7024,10 +7028,41 @@ static void create_pa_complete(struct hci_dev *hdev, void *data, int err) hci_dev_unlock(hdev); } +static int hci_le_past_params_sync(struct hci_dev *hdev, struct hci_conn *conn, + struct hci_conn *acl, struct bt_iso_qos *qos) +{ + struct hci_cp_le_past_params cp; + int err; + + memset(&cp, 0, sizeof(cp)); + cp.handle = cpu_to_le16(acl->handle); + /* An HCI_LE_Periodic_Advertising_Sync_Transfer_Received event is sent + * to the Host. HCI_LE_Periodic_Advertising_Report events will be + * enabled with duplicate filtering enabled. + */ + cp.mode = 0x03; + cp.skip = cpu_to_le16(qos->bcast.skip); + cp.sync_timeout = cpu_to_le16(qos->bcast.sync_timeout); + cp.cte_type = qos->bcast.sync_cte_type; + + /* HCI_LE_PAST_PARAMS command returns a command complete event so it + * cannot wait for HCI_EV_LE_PAST_RECEIVED. + */ + err = __hci_cmd_sync_status(hdev, HCI_OP_LE_PAST_PARAMS, + sizeof(cp), &cp, HCI_CMD_TIMEOUT); + if (err) + return err; + + /* Wait for HCI_EV_LE_PAST_RECEIVED event */ + return __hci_cmd_sync_status_sk(hdev, HCI_OP_NOP, 0, NULL, + HCI_EV_LE_PAST_RECEIVED, + conn->conn_timeout, NULL); +} + static int hci_le_pa_create_sync(struct hci_dev *hdev, void *data) { struct hci_cp_le_pa_create_sync cp; - struct hci_conn *conn = data; + struct hci_conn *conn = data, *le; struct bt_iso_qos *qos = &conn->iso_qos; int err; @@ -7059,6 +7094,24 @@ static int hci_le_pa_create_sync(struct hci_dev *hdev, void *data) hci_update_passive_scan_sync(hdev); + /* Check if PAST is possible: + * + * 1. Check if an ACL connection with the destination address exists + * 2. Check if that HCI_CONN_FLAG_PAST has been set which indicates that + * user really intended to use PAST. + */ + le = hci_conn_hash_lookup_le(hdev, &conn->dst, conn->dst_type); + if (le) { + struct hci_conn_params *params; + + params = hci_conn_params_lookup(hdev, &le->dst, le->dst_type); + if (params && params->flags & HCI_CONN_FLAG_PAST) { + err = hci_le_past_params_sync(hdev, conn, le, qos); + if (!err) + goto done; + } + } + /* SID has not been set listen for HCI_EV_LE_EXT_ADV_REPORT to update * it. */ From d3413703d5f8b7d1e6f514f9440ed5da1bc30796 Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Fri, 5 Sep 2025 11:34:44 -0400 Subject: [PATCH 1303/1381] Bluetooth: ISO: Add support to bind to trigger PAST This makes it possible to bind to a different destination address after being connected (BT_CONNECTED, BT_CONNECT2) which then triggers PAST Sender proceedure to transfer the PA Sync to the destination address. Signed-off-by: Luiz Augusto von Dentz --- include/net/bluetooth/hci_core.h | 1 + include/net/bluetooth/hci_sync.h | 1 + net/bluetooth/hci_conn.c | 12 ++++ net/bluetooth/hci_sync.c | 92 ++++++++++++++++++++++++++++++ net/bluetooth/iso.c | 98 ++++++++++++++++++++++++++------ 5 files changed, 187 insertions(+), 17 deletions(-) diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index 8c2235444808..1f74722f3f4d 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -1602,6 +1602,7 @@ struct hci_conn *hci_bind_cis(struct hci_dev *hdev, bdaddr_t *dst, struct hci_conn *hci_bind_bis(struct hci_dev *hdev, bdaddr_t *dst, __u8 sid, struct bt_iso_qos *qos, __u8 base_len, __u8 *base, u16 timeout); +int hci_past_bis(struct hci_conn *conn, bdaddr_t *dst, __u8 dst_type); struct hci_conn *hci_connect_cis(struct hci_dev *hdev, bdaddr_t *dst, __u8 dst_type, struct bt_iso_qos *qos, u16 timeout); diff --git a/include/net/bluetooth/hci_sync.h b/include/net/bluetooth/hci_sync.h index e352a4e0ef8d..3133f40fa9f9 100644 --- a/include/net/bluetooth/hci_sync.h +++ b/include/net/bluetooth/hci_sync.h @@ -188,3 +188,4 @@ int hci_le_conn_update_sync(struct hci_dev *hdev, struct hci_conn *conn, int hci_connect_pa_sync(struct hci_dev *hdev, struct hci_conn *conn); int hci_connect_big_sync(struct hci_dev *hdev, struct hci_conn *conn); +int hci_past_sync(struct hci_conn *conn, struct hci_conn *le); diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c index 6fc0692abf05..4f9dc1435ccc 100644 --- a/net/bluetooth/hci_conn.c +++ b/net/bluetooth/hci_conn.c @@ -2245,6 +2245,18 @@ struct hci_conn *hci_bind_bis(struct hci_dev *hdev, bdaddr_t *dst, __u8 sid, return conn; } +int hci_past_bis(struct hci_conn *conn, bdaddr_t *dst, __u8 dst_type) +{ + struct hci_conn *le; + + /* Lookup existing LE connection to rebind to */ + le = hci_conn_hash_lookup_le(conn->hdev, dst, dst_type); + if (!le) + return -EINVAL; + + return hci_past_sync(conn, le); +} + static void bis_mark_per_adv(struct hci_conn *conn, void *data) { struct iso_list_data *d = data; diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c index ba6f13e9235c..65f2701beb49 100644 --- a/net/bluetooth/hci_sync.c +++ b/net/bluetooth/hci_sync.c @@ -7228,3 +7228,95 @@ int hci_connect_big_sync(struct hci_dev *hdev, struct hci_conn *conn) return hci_cmd_sync_queue_once(hdev, hci_le_big_create_sync, conn, create_big_complete); } + +struct past_data { + struct hci_conn *conn; + struct hci_conn *le; +}; + +static void past_complete(struct hci_dev *hdev, void *data, int err) +{ + struct past_data *past = data; + + bt_dev_dbg(hdev, "err %d", err); + + kfree(past); +} + +static int hci_le_past_set_info_sync(struct hci_dev *hdev, void *data) +{ + struct past_data *past = data; + struct hci_cp_le_past_set_info cp; + + hci_dev_lock(hdev); + + if (!hci_conn_valid(hdev, past->conn) || + !hci_conn_valid(hdev, past->le)) { + hci_dev_unlock(hdev); + return -ECANCELED; + } + + memset(&cp, 0, sizeof(cp)); + cp.handle = cpu_to_le16(past->le->handle); + cp.adv_handle = past->conn->iso_qos.bcast.bis; + + hci_dev_unlock(hdev); + + return __hci_cmd_sync_status(hdev, HCI_OP_LE_PAST_SET_INFO, + sizeof(cp), &cp, HCI_CMD_TIMEOUT); +} + +static int hci_le_past_sync(struct hci_dev *hdev, void *data) +{ + struct past_data *past = data; + struct hci_cp_le_past cp; + + hci_dev_lock(hdev); + + if (!hci_conn_valid(hdev, past->conn) || + !hci_conn_valid(hdev, past->le)) { + hci_dev_unlock(hdev); + return -ECANCELED; + } + + memset(&cp, 0, sizeof(cp)); + cp.handle = cpu_to_le16(past->le->handle); + cp.sync_handle = cpu_to_le16(past->conn->sync_handle); + + hci_dev_unlock(hdev); + + return __hci_cmd_sync_status(hdev, HCI_OP_LE_PAST, + sizeof(cp), &cp, HCI_CMD_TIMEOUT); +} + +int hci_past_sync(struct hci_conn *conn, struct hci_conn *le) +{ + struct past_data *data; + int err; + + if (conn->type != BIS_LINK && conn->type != PA_LINK) + return -EINVAL; + + if (!past_sender_capable(conn->hdev)) + return -EOPNOTSUPP; + + data = kmalloc(sizeof(*data), GFP_KERNEL); + if (!data) + return -ENOMEM; + + data->conn = conn; + data->le = le; + + if (conn->role == HCI_ROLE_MASTER) + err = hci_cmd_sync_queue_once(conn->hdev, + hci_le_past_set_info_sync, data, + past_complete); + else + err = hci_cmd_sync_queue_once(conn->hdev, hci_le_past_sync, + data, past_complete); + + if (err) + kfree(data); + + return err; +} diff --git a/net/bluetooth/iso.c b/net/bluetooth/iso.c index d0a79f601e8d..85fa9363b897 100644 --- a/net/bluetooth/iso.c +++ b/net/bluetooth/iso.c @@ -987,20 +987,14 @@ static int iso_sock_bind_bc(struct socket *sock, struct sockaddr_unsized *addr, return 0; } -static int iso_sock_bind_pa_sk(struct sock *sk, struct sockaddr_iso *sa, +/* Must be called on the locked socket. */ +static int iso_sock_rebind_bis(struct sock *sk, struct sockaddr_iso *sa, int addr_len) { int err = 0; - if (sk->sk_type != SOCK_SEQPACKET) { - err = -EINVAL; - goto done; - } - - if (addr_len != sizeof(*sa) + sizeof(*sa->iso_bc)) { - err = -EINVAL; - goto done; - } + if (!test_bit(BT_SK_PA_SYNC, &iso_pi(sk)->flags)) + return -EBADFD; if (sa->iso_bc->bc_num_bis > ISO_MAX_NUM_BIS) { err = -EINVAL; @@ -1023,6 +1017,77 @@ static int iso_sock_bind_pa_sk(struct sock *sk, struct sockaddr_iso *sa, return err; } +static struct hci_dev *iso_conn_get_hdev(struct iso_conn *conn) +{ + struct hci_dev *hdev = NULL; + + iso_conn_lock(conn); + if (conn->hcon) + hdev = hci_dev_hold(conn->hcon->hdev); + iso_conn_unlock(conn); + + return hdev; +} + +/* Must be called on the locked socket. */ +static int iso_sock_rebind_bc(struct sock *sk, struct sockaddr_iso *sa, + int addr_len) +{ + struct hci_dev *hdev; + struct hci_conn *bis; + int err; + + if (sk->sk_type != SOCK_SEQPACKET || !iso_pi(sk)->conn) + return -EINVAL; + + /* Check if it is really a Broadcast address being requested */ + if (addr_len != sizeof(*sa) + sizeof(*sa->iso_bc)) + return -EINVAL; + + /* Check if the address hasn't changed then perhaps only the number of + * bis has changed. + */ + if (!bacmp(&iso_pi(sk)->dst, &sa->iso_bc->bc_bdaddr) || + !bacmp(&sa->iso_bc->bc_bdaddr, BDADDR_ANY)) + return iso_sock_rebind_bis(sk, sa, addr_len); + + /* Check if the address type is of LE type */ + if (!bdaddr_type_is_le(sa->iso_bc->bc_bdaddr_type)) + return -EINVAL; + + hdev = iso_conn_get_hdev(iso_pi(sk)->conn); + if (!hdev) + return -EINVAL; + + bis = iso_pi(sk)->conn->hcon; + + /* Release the socket before lookups since that requires hci_dev_lock + * which shall not be acquired while holding sock_lock for proper + * ordering. + */ + release_sock(sk); + hci_dev_lock(bis->hdev); + lock_sock(sk); + + if (!iso_pi(sk)->conn || iso_pi(sk)->conn->hcon != bis) { + /* raced with iso_conn_del() or iso_disconn_sock() */ + err = -ENOTCONN; + goto unlock; + } + + BT_DBG("sk %p %pMR type %u", sk, &sa->iso_bc->bc_bdaddr, + sa->iso_bc->bc_bdaddr_type); + + err = hci_past_bis(bis, &sa->iso_bc->bc_bdaddr, + le_addr_type(sa->iso_bc->bc_bdaddr_type)); + +unlock: + hci_dev_unlock(hdev); + hci_dev_put(hdev); + + return err; +} + static int iso_sock_bind(struct socket *sock, struct sockaddr_unsized *addr, int addr_len) { @@ -1038,13 +1103,12 @@ static int iso_sock_bind(struct socket *sock, struct sockaddr_unsized *addr, lock_sock(sk); - /* Allow the user to bind a PA sync socket to a number - * of BISes to sync to. - */ - if ((sk->sk_state == BT_CONNECT2 || - sk->sk_state == BT_CONNECTED) && - test_bit(BT_SK_PA_SYNC, &iso_pi(sk)->flags)) { - err = iso_sock_bind_pa_sk(sk, sa, addr_len); + if ((sk->sk_state == BT_CONNECT2 || sk->sk_state == BT_CONNECTED) && + addr_len > sizeof(*sa)) { + /* Allow the user to rebind to a different address using + * PAST procedures. + */ + err = iso_sock_rebind_bc(sk, sa, addr_len); goto done; } From 14b06c3a88f7031d64fbce197fad1d400e507663 Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Fri, 26 Sep 2025 15:56:43 -0400 Subject: [PATCH 1304/1381] Bluetooth: HCI: Always use the identity address when initializing a connection This makes sure hci_conn is initialized with the identity address if a matching IRK exists which avoids the trouble of having to do it at multiple places which seems to be missing (e.g. CIS, BIS and PA). Signed-off-by: Luiz Augusto von Dentz --- include/net/bluetooth/hci_core.h | 4 +-- net/bluetooth/hci_conn.c | 43 ++++++++++++++++++++------------ net/bluetooth/hci_event.c | 20 +++++++-------- net/bluetooth/hci_sync.c | 2 +- 4 files changed, 40 insertions(+), 29 deletions(-) diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index 1f74722f3f4d..858b58206e80 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -1571,9 +1571,9 @@ int hci_le_create_cis_pending(struct hci_dev *hdev); int hci_conn_check_create_cis(struct hci_conn *conn); struct hci_conn *hci_conn_add(struct hci_dev *hdev, int type, bdaddr_t *dst, - u8 role, u16 handle); + u8 dst_type, u8 role, u16 handle); struct hci_conn *hci_conn_add_unset(struct hci_dev *hdev, int type, - bdaddr_t *dst, u8 role); + bdaddr_t *dst, u8 dst_type, u8 role); void hci_conn_del(struct hci_conn *conn); void hci_conn_hash_flush(struct hci_dev *hdev); diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c index 4f9dc1435ccc..c3f7828bf9d5 100644 --- a/net/bluetooth/hci_conn.c +++ b/net/bluetooth/hci_conn.c @@ -922,10 +922,12 @@ static int hci_conn_hash_alloc_unset(struct hci_dev *hdev) U16_MAX, GFP_ATOMIC); } -static struct hci_conn *__hci_conn_add(struct hci_dev *hdev, int type, bdaddr_t *dst, +static struct hci_conn *__hci_conn_add(struct hci_dev *hdev, int type, + bdaddr_t *dst, u8 dst_type, u8 role, u16 handle) { struct hci_conn *conn; + struct smp_irk *irk = NULL; switch (type) { case ACL_LINK: @@ -937,12 +939,14 @@ static struct hci_conn *__hci_conn_add(struct hci_dev *hdev, int type, bdaddr_t case PA_LINK: if (!hdev->iso_mtu) return ERR_PTR(-ECONNREFUSED); + irk = hci_get_irk(hdev, dst, dst_type); break; case LE_LINK: if (hdev->le_mtu && hdev->le_mtu < HCI_MIN_LE_MTU) return ERR_PTR(-ECONNREFUSED); if (!hdev->le_mtu && hdev->acl_mtu < HCI_MIN_LE_MTU) return ERR_PTR(-ECONNREFUSED); + irk = hci_get_irk(hdev, dst, dst_type); break; case SCO_LINK: case ESCO_LINK: @@ -960,7 +964,15 @@ static struct hci_conn *__hci_conn_add(struct hci_dev *hdev, int type, bdaddr_t if (!conn) return ERR_PTR(-ENOMEM); - bacpy(&conn->dst, dst); + /* If and IRK exists use its identity address */ + if (!irk) { + bacpy(&conn->dst, dst); + conn->dst_type = dst_type; + } else { + bacpy(&conn->dst, &irk->bdaddr); + conn->dst_type = irk->addr_type; + } + bacpy(&conn->src, &hdev->bdaddr); conn->handle = handle; conn->hdev = hdev; @@ -1059,7 +1071,7 @@ static struct hci_conn *__hci_conn_add(struct hci_dev *hdev, int type, bdaddr_t } struct hci_conn *hci_conn_add_unset(struct hci_dev *hdev, int type, - bdaddr_t *dst, u8 role) + bdaddr_t *dst, u8 dst_type, u8 role) { int handle; @@ -1069,16 +1081,16 @@ struct hci_conn *hci_conn_add_unset(struct hci_dev *hdev, int type, if (unlikely(handle < 0)) return ERR_PTR(-ECONNREFUSED); - return __hci_conn_add(hdev, type, dst, role, handle); + return __hci_conn_add(hdev, type, dst, dst_type, role, handle); } struct hci_conn *hci_conn_add(struct hci_dev *hdev, int type, bdaddr_t *dst, - u8 role, u16 handle) + u8 dst_type, u8 role, u16 handle) { if (handle > HCI_CONN_HANDLE_MAX) return ERR_PTR(-EINVAL); - return __hci_conn_add(hdev, type, dst, role, handle); + return __hci_conn_add(hdev, type, dst, dst_type, role, handle); } static void hci_conn_cleanup_child(struct hci_conn *conn, u8 reason) @@ -1410,14 +1422,13 @@ struct hci_conn *hci_connect_le(struct hci_dev *hdev, bdaddr_t *dst, if (conn) { bacpy(&conn->dst, dst); } else { - conn = hci_conn_add_unset(hdev, LE_LINK, dst, role); + conn = hci_conn_add_unset(hdev, LE_LINK, dst, dst_type, role); if (IS_ERR(conn)) return conn; hci_conn_hold(conn); conn->pending_sec_level = sec_level; } - conn->dst_type = dst_type; conn->sec_level = BT_SECURITY_LOW; conn->conn_timeout = conn_timeout; conn->le_adv_phy = phy; @@ -1587,7 +1598,7 @@ static struct hci_conn *hci_add_bis(struct hci_dev *hdev, bdaddr_t *dst, memcmp(conn->le_per_adv_data, base, base_len))) return ERR_PTR(-EADDRINUSE); - conn = hci_conn_add_unset(hdev, BIS_LINK, dst, HCI_ROLE_MASTER); + conn = hci_conn_add_unset(hdev, BIS_LINK, dst, 0, HCI_ROLE_MASTER); if (IS_ERR(conn)) return conn; @@ -1633,7 +1644,8 @@ struct hci_conn *hci_connect_le_scan(struct hci_dev *hdev, bdaddr_t *dst, BT_DBG("requesting refresh of dst_addr"); - conn = hci_conn_add_unset(hdev, LE_LINK, dst, HCI_ROLE_MASTER); + conn = hci_conn_add_unset(hdev, LE_LINK, dst, dst_type, + HCI_ROLE_MASTER); if (IS_ERR(conn)) return conn; @@ -1644,7 +1656,6 @@ struct hci_conn *hci_connect_le_scan(struct hci_dev *hdev, bdaddr_t *dst, conn->state = BT_CONNECT; set_bit(HCI_CONN_SCANNING, &conn->flags); - conn->dst_type = dst_type; conn->sec_level = BT_SECURITY_LOW; conn->pending_sec_level = sec_level; conn->conn_timeout = conn_timeout; @@ -1681,7 +1692,8 @@ struct hci_conn *hci_connect_acl(struct hci_dev *hdev, bdaddr_t *dst, acl = hci_conn_hash_lookup_ba(hdev, ACL_LINK, dst); if (!acl) { - acl = hci_conn_add_unset(hdev, ACL_LINK, dst, HCI_ROLE_MASTER); + acl = hci_conn_add_unset(hdev, ACL_LINK, dst, 0, + HCI_ROLE_MASTER); if (IS_ERR(acl)) return acl; } @@ -1750,7 +1762,7 @@ struct hci_conn *hci_connect_sco(struct hci_dev *hdev, int type, bdaddr_t *dst, sco = hci_conn_hash_lookup_ba(hdev, type, dst); if (!sco) { - sco = hci_conn_add_unset(hdev, type, dst, HCI_ROLE_MASTER); + sco = hci_conn_add_unset(hdev, type, dst, 0, HCI_ROLE_MASTER); if (IS_ERR(sco)) { hci_conn_drop(acl); return sco; @@ -1942,7 +1954,7 @@ struct hci_conn *hci_bind_cis(struct hci_dev *hdev, bdaddr_t *dst, cis = hci_conn_hash_lookup_cis(hdev, dst, dst_type, qos->ucast.cig, qos->ucast.cis); if (!cis) { - cis = hci_conn_add_unset(hdev, CIS_LINK, dst, + cis = hci_conn_add_unset(hdev, CIS_LINK, dst, dst_type, HCI_ROLE_MASTER); if (IS_ERR(cis)) return cis; @@ -2133,12 +2145,11 @@ struct hci_conn *hci_pa_create_sync(struct hci_dev *hdev, bdaddr_t *dst, bt_dev_dbg(hdev, "dst %pMR type %d sid %d", dst, dst_type, sid); - conn = hci_conn_add_unset(hdev, PA_LINK, dst, HCI_ROLE_SLAVE); + conn = hci_conn_add_unset(hdev, PA_LINK, dst, dst_type, HCI_ROLE_SLAVE); if (IS_ERR(conn)) return conn; conn->iso_qos = *qos; - conn->dst_type = dst_type; conn->sid = sid; conn->state = BT_LISTEN; conn->conn_timeout = msecs_to_jiffies(qos->bcast.sync_timeout * 10); diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c index af34c9938509..7c4ca14f13e5 100644 --- a/net/bluetooth/hci_event.c +++ b/net/bluetooth/hci_event.c @@ -2267,7 +2267,7 @@ static void hci_cs_create_conn(struct hci_dev *hdev, __u8 status) } else { if (!conn) { conn = hci_conn_add_unset(hdev, ACL_LINK, &cp->bdaddr, - HCI_ROLE_MASTER); + 0, HCI_ROLE_MASTER); if (IS_ERR(conn)) bt_dev_err(hdev, "connection err: %ld", PTR_ERR(conn)); } @@ -3123,7 +3123,8 @@ static void hci_conn_complete_evt(struct hci_dev *hdev, void *data, &ev->bdaddr, BDADDR_BREDR)) { conn = hci_conn_add_unset(hdev, ev->link_type, - &ev->bdaddr, HCI_ROLE_SLAVE); + &ev->bdaddr, 0, + HCI_ROLE_SLAVE); if (IS_ERR(conn)) { bt_dev_err(hdev, "connection err: %ld", PTR_ERR(conn)); goto unlock; @@ -3299,7 +3300,7 @@ static void hci_conn_request_evt(struct hci_dev *hdev, void *data, conn = hci_conn_hash_lookup_ba(hdev, ev->link_type, &ev->bdaddr); if (!conn) { - conn = hci_conn_add_unset(hdev, ev->link_type, &ev->bdaddr, + conn = hci_conn_add_unset(hdev, ev->link_type, &ev->bdaddr, 0, HCI_ROLE_SLAVE); if (IS_ERR(conn)) { bt_dev_err(hdev, "connection err: %ld", PTR_ERR(conn)); @@ -5670,14 +5671,13 @@ static void le_conn_complete_evt(struct hci_dev *hdev, u8 status, if (status) goto unlock; - conn = hci_conn_add_unset(hdev, LE_LINK, bdaddr, role); + conn = hci_conn_add_unset(hdev, LE_LINK, bdaddr, bdaddr_type, + role); if (IS_ERR(conn)) { bt_dev_err(hdev, "connection err: %ld", PTR_ERR(conn)); goto unlock; } - conn->dst_type = bdaddr_type; - /* If we didn't have a hci_conn object previously * but we're in central role this must be something * initiated using an accept list. Since accept list based @@ -5982,7 +5982,7 @@ static void hci_le_past_received_evt(struct hci_dev *hdev, void *data, goto unlock; /* Add connection to indicate PA sync event */ - pa_sync = hci_conn_add_unset(hdev, PA_LINK, BDADDR_ANY, + pa_sync = hci_conn_add_unset(hdev, PA_LINK, BDADDR_ANY, 0, HCI_ROLE_SLAVE); if (IS_ERR(pa_sync)) @@ -6515,7 +6515,7 @@ static void hci_le_pa_sync_established_evt(struct hci_dev *hdev, void *data, goto unlock; /* Add connection to indicate PA sync event */ - pa_sync = hci_conn_add_unset(hdev, PA_LINK, BDADDR_ANY, + pa_sync = hci_conn_add_unset(hdev, PA_LINK, BDADDR_ANY, 0, HCI_ROLE_SLAVE); if (IS_ERR(pa_sync)) @@ -6956,7 +6956,7 @@ static void hci_le_cis_req_evt(struct hci_dev *hdev, void *data, cis = hci_conn_hash_lookup_handle(hdev, cis_handle); if (!cis) { - cis = hci_conn_add(hdev, CIS_LINK, &acl->dst, + cis = hci_conn_add(hdev, CIS_LINK, &acl->dst, acl->dst_type, HCI_ROLE_SLAVE, cis_handle); if (IS_ERR(cis)) { hci_le_reject_cis(hdev, ev->cis_handle); @@ -7073,7 +7073,7 @@ static void hci_le_big_sync_established_evt(struct hci_dev *hdev, void *data, bt_dev_dbg(hdev, "ignore too large handle %u", handle); continue; } - bis = hci_conn_add(hdev, BIS_LINK, BDADDR_ANY, + bis = hci_conn_add(hdev, BIS_LINK, BDADDR_ANY, 0, HCI_ROLE_SLAVE, handle); if (IS_ERR(bis)) continue; diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c index 65f2701beb49..a36d2414a3ca 100644 --- a/net/bluetooth/hci_sync.c +++ b/net/bluetooth/hci_sync.c @@ -7013,7 +7013,7 @@ static void create_pa_complete(struct hci_dev *hdev, void *data, int err) goto unlock; /* Add connection to indicate PA sync error */ - pa_sync = hci_conn_add_unset(hdev, PA_LINK, BDADDR_ANY, + pa_sync = hci_conn_add_unset(hdev, PA_LINK, BDADDR_ANY, 0, HCI_ROLE_SLAVE); if (IS_ERR(pa_sync)) From f817db10dc80d5d1eece60d93eba11062afe5154 Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Mon, 29 Sep 2025 17:43:53 -0400 Subject: [PATCH 1305/1381] Bluetooth: ISO: Attempt to resolve broadcast address Broadcasters maybe using RPAs which can change over time and not matching the address used as destination in the socket, so this attempts to resolve the addresses then match with the socket address, in case that uses an indentity address, or then match the IRKs if both broadcaster and socket are using RPAs. Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/iso.c | 58 ++++++++++++++++++++++++++++++--------------- 1 file changed, 39 insertions(+), 19 deletions(-) diff --git a/net/bluetooth/iso.c b/net/bluetooth/iso.c index 85fa9363b897..fd56eebe9245 100644 --- a/net/bluetooth/iso.c +++ b/net/bluetooth/iso.c @@ -87,8 +87,8 @@ static void iso_sock_disconn(struct sock *sk); typedef bool (*iso_sock_match_t)(struct sock *sk, void *data); -static struct sock *iso_get_sock(bdaddr_t *src, bdaddr_t *dst, - enum bt_sock_state state, +static struct sock *iso_get_sock(struct hci_dev *hdev, bdaddr_t *src, + bdaddr_t *dst, enum bt_sock_state state, iso_sock_match_t match, void *data); /* ---- ISO timers ---- */ @@ -638,8 +638,8 @@ static struct sock *__iso_get_sock_listen_by_sid(bdaddr_t *ba, bdaddr_t *bc, * match func data - pass -1 to ignore * Returns closest match. */ -static struct sock *iso_get_sock(bdaddr_t *src, bdaddr_t *dst, - enum bt_sock_state state, +static struct sock *iso_get_sock(struct hci_dev *hdev, bdaddr_t *src, + bdaddr_t *dst, enum bt_sock_state state, iso_sock_match_t match, void *data) { struct sock *sk = NULL, *sk1 = NULL; @@ -651,8 +651,25 @@ static struct sock *iso_get_sock(bdaddr_t *src, bdaddr_t *dst, continue; /* Match Broadcast destination */ - if (bacmp(dst, BDADDR_ANY) && bacmp(&iso_pi(sk)->dst, dst)) - continue; + if (bacmp(dst, BDADDR_ANY) && bacmp(&iso_pi(sk)->dst, dst)) { + struct smp_irk *irk1, *irk2; + + /* Check if destination is an RPA that we can resolve */ + irk1 = hci_find_irk_by_rpa(hdev, dst); + if (!irk1) + continue; + + /* Match with identity address */ + if (bacmp(&iso_pi(sk)->dst, &irk1->bdaddr)) { + /* Check if socket destination address is also + * an RPA and if the IRK matches. + */ + irk2 = hci_find_irk_by_rpa(hdev, + &iso_pi(sk)->dst); + if (!irk2 || irk1 != irk2) + continue; + } + } /* Use Match function if provided */ if (match && !match(sk, data)) @@ -2012,6 +2029,7 @@ static void iso_conn_ready(struct iso_conn *conn) struct hci_ev_le_pa_sync_established *ev2 = NULL; struct hci_ev_le_per_adv_report *ev3 = NULL; struct hci_conn *hcon; + struct hci_dev *hdev; BT_DBG("conn %p", conn); @@ -2022,13 +2040,15 @@ static void iso_conn_ready(struct iso_conn *conn) if (!hcon) return; + hdev = hcon->hdev; + if (test_bit(HCI_CONN_BIG_SYNC, &hcon->flags)) { /* A BIS slave hcon is notified to the ISO layer * after the Command Complete for the LE Setup * ISO Data Path command is received. Get the * parent socket that matches the hcon BIG handle. */ - parent = iso_get_sock(&hcon->src, &hcon->dst, + parent = iso_get_sock(hdev, &hcon->src, &hcon->dst, BT_LISTEN, iso_match_big_hcon, hcon); } else if (test_bit(HCI_CONN_BIG_SYNC_FAILED, &hcon->flags)) { @@ -2036,12 +2056,12 @@ static void iso_conn_ready(struct iso_conn *conn) HCI_EVT_LE_BIG_SYNC_ESTABLISHED); /* Get reference to PA sync parent socket, if it exists */ - parent = iso_get_sock(&hcon->src, &hcon->dst, + parent = iso_get_sock(hdev, &hcon->src, &hcon->dst, BT_LISTEN, iso_match_pa_sync_flag, NULL); if (!parent && ev) - parent = iso_get_sock(&hcon->src, + parent = iso_get_sock(hdev, &hcon->src, &hcon->dst, BT_LISTEN, iso_match_big, ev); @@ -2049,7 +2069,7 @@ static void iso_conn_ready(struct iso_conn *conn) ev2 = hci_recv_event_data(hcon->hdev, HCI_EV_LE_PA_SYNC_ESTABLISHED); if (ev2) - parent = iso_get_sock(&hcon->src, + parent = iso_get_sock(hdev, &hcon->src, &hcon->dst, BT_LISTEN, iso_match_sid, ev2); @@ -2057,7 +2077,7 @@ static void iso_conn_ready(struct iso_conn *conn) ev3 = hci_recv_event_data(hcon->hdev, HCI_EV_LE_PER_ADV_REPORT); if (ev3) - parent = iso_get_sock(&hcon->src, + parent = iso_get_sock(hdev, &hcon->src, &hcon->dst, BT_LISTEN, iso_match_sync_handle_pa_report, @@ -2065,7 +2085,7 @@ static void iso_conn_ready(struct iso_conn *conn) } if (!parent) - parent = iso_get_sock(&hcon->src, BDADDR_ANY, + parent = iso_get_sock(hdev, &hcon->src, BDADDR_ANY, BT_LISTEN, NULL, NULL); if (!parent) @@ -2208,7 +2228,7 @@ int iso_connect_ind(struct hci_dev *hdev, bdaddr_t *bdaddr, __u8 *flags) */ ev1 = hci_recv_event_data(hdev, HCI_EV_LE_PA_SYNC_ESTABLISHED); if (ev1) { - sk = iso_get_sock(&hdev->bdaddr, bdaddr, BT_LISTEN, + sk = iso_get_sock(hdev, &hdev->bdaddr, bdaddr, BT_LISTEN, iso_match_sid, ev1); if (sk && !ev1->status) { iso_pi(sk)->sync_handle = le16_to_cpu(ev1->handle); @@ -2220,7 +2240,7 @@ int iso_connect_ind(struct hci_dev *hdev, bdaddr_t *bdaddr, __u8 *flags) ev1a = hci_recv_event_data(hdev, HCI_EV_LE_PAST_RECEIVED); if (ev1a) { - sk = iso_get_sock(&hdev->bdaddr, bdaddr, BT_LISTEN, + sk = iso_get_sock(hdev, &hdev->bdaddr, bdaddr, BT_LISTEN, iso_match_sid_past, ev1a); if (sk && !ev1a->status) { iso_pi(sk)->sync_handle = le16_to_cpu(ev1a->sync_handle); @@ -2233,7 +2253,7 @@ int iso_connect_ind(struct hci_dev *hdev, bdaddr_t *bdaddr, __u8 *flags) ev2 = hci_recv_event_data(hdev, HCI_EVT_LE_BIG_INFO_ADV_REPORT); if (ev2) { /* Check if BIGInfo report has already been handled */ - sk = iso_get_sock(&hdev->bdaddr, bdaddr, BT_CONNECTED, + sk = iso_get_sock(hdev, &hdev->bdaddr, bdaddr, BT_CONNECTED, iso_match_sync_handle, ev2); if (sk) { sock_put(sk); @@ -2242,10 +2262,10 @@ int iso_connect_ind(struct hci_dev *hdev, bdaddr_t *bdaddr, __u8 *flags) } /* Try to get PA sync socket, if it exists */ - sk = iso_get_sock(&hdev->bdaddr, bdaddr, BT_CONNECT2, + sk = iso_get_sock(hdev, &hdev->bdaddr, bdaddr, BT_CONNECT2, iso_match_sync_handle, ev2); if (!sk) - sk = iso_get_sock(&hdev->bdaddr, bdaddr, + sk = iso_get_sock(hdev, &hdev->bdaddr, bdaddr, BT_LISTEN, iso_match_sync_handle, ev2); @@ -2284,7 +2304,7 @@ int iso_connect_ind(struct hci_dev *hdev, bdaddr_t *bdaddr, __u8 *flags) u8 *base; struct hci_conn *hcon; - sk = iso_get_sock(&hdev->bdaddr, bdaddr, BT_LISTEN, + sk = iso_get_sock(hdev, &hdev->bdaddr, bdaddr, BT_LISTEN, iso_match_sync_handle_pa_report, ev3); if (!sk) goto done; @@ -2334,7 +2354,7 @@ int iso_connect_ind(struct hci_dev *hdev, bdaddr_t *bdaddr, __u8 *flags) hcon->le_per_adv_data_len = 0; } } else { - sk = iso_get_sock(&hdev->bdaddr, BDADDR_ANY, + sk = iso_get_sock(hdev, &hdev->bdaddr, BDADDR_ANY, BT_LISTEN, NULL, NULL); } From a3b76bf4c4c90994248e3211b300d2d9834874f9 Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Mon, 6 Oct 2025 16:32:24 -0400 Subject: [PATCH 1306/1381] Bluetooth: MGMT: Allow use of Set Device Flags without Add Device In certain cases setting devices flags like HCI_CONN_FLAG_PAST it shouldn't require to do Add Device first since it may not need to add an auto-connect policy, so this instead just automatically creates a hci_conn_params if one cannot be found using HCI_AUTO_CONN_DISABLED. Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/mgmt.c | 148 ++++++++++++++++++++++--------------------- 1 file changed, 76 insertions(+), 72 deletions(-) diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c index 7dbd3d57e66c..c11cdef42b6f 100644 --- a/net/bluetooth/mgmt.c +++ b/net/bluetooth/mgmt.c @@ -5122,6 +5122,69 @@ static void device_flags_changed(struct sock *sk, struct hci_dev *hdev, mgmt_event(MGMT_EV_DEVICE_FLAGS_CHANGED, hdev, &ev, sizeof(ev), sk); } +static bool is_connected(struct hci_dev *hdev, bdaddr_t *addr, u8 type) +{ + struct hci_conn *conn; + + conn = hci_conn_hash_lookup_ba(hdev, LE_LINK, addr); + if (!conn) + return false; + + if (conn->dst_type != type) + return false; + + if (conn->state != BT_CONNECTED) + return false; + + return true; +} + +/* This function requires the caller holds hdev->lock */ +static struct hci_conn_params *hci_conn_params_set(struct hci_dev *hdev, + bdaddr_t *addr, u8 addr_type, + u8 auto_connect) +{ + struct hci_conn_params *params; + + params = hci_conn_params_add(hdev, addr, addr_type); + if (!params) + return NULL; + + if (params->auto_connect == auto_connect) + return params; + + hci_pend_le_list_del_init(params); + + switch (auto_connect) { + case HCI_AUTO_CONN_DISABLED: + case HCI_AUTO_CONN_LINK_LOSS: + /* If auto connect is being disabled when we're trying to + * connect to device, keep connecting. + */ + if (params->explicit_connect) + hci_pend_le_list_add(params, &hdev->pend_le_conns); + break; + case HCI_AUTO_CONN_REPORT: + if (params->explicit_connect) + hci_pend_le_list_add(params, &hdev->pend_le_conns); + else + hci_pend_le_list_add(params, &hdev->pend_le_reports); + break; + case HCI_AUTO_CONN_DIRECT: + case HCI_AUTO_CONN_ALWAYS: + if (!is_connected(hdev, addr, addr_type)) + hci_pend_le_list_add(params, &hdev->pend_le_conns); + break; + } + + params->auto_connect = auto_connect; + + bt_dev_dbg(hdev, "addr %pMR (type %u) auto_connect %u", + addr, addr_type, auto_connect); + + return params; +} + static int set_device_flags(struct sock *sk, struct hci_dev *hdev, void *data, u16 len) { @@ -5165,9 +5228,16 @@ static int set_device_flags(struct sock *sk, struct hci_dev *hdev, void *data, params = hci_conn_params_lookup(hdev, &cp->addr.bdaddr, le_addr_type(cp->addr.type)); if (!params) { - bt_dev_warn(hdev, "No such LE device %pMR (0x%x)", - &cp->addr.bdaddr, le_addr_type(cp->addr.type)); - goto unlock; + /* Create a new hci_conn_params if it doesn't exist */ + params = hci_conn_params_set(hdev, &cp->addr.bdaddr, + le_addr_type(cp->addr.type), + HCI_AUTO_CONN_DISABLED); + if (!params) { + bt_dev_warn(hdev, "No such LE device %pMR (0x%x)", + &cp->addr.bdaddr, + le_addr_type(cp->addr.type)); + goto unlock; + } } supported_flags = hdev->conn_flags; @@ -7554,68 +7624,6 @@ static int get_clock_info(struct sock *sk, struct hci_dev *hdev, void *data, return err; } -static bool is_connected(struct hci_dev *hdev, bdaddr_t *addr, u8 type) -{ - struct hci_conn *conn; - - conn = hci_conn_hash_lookup_ba(hdev, LE_LINK, addr); - if (!conn) - return false; - - if (conn->dst_type != type) - return false; - - if (conn->state != BT_CONNECTED) - return false; - - return true; -} - -/* This function requires the caller holds hdev->lock */ -static int hci_conn_params_set(struct hci_dev *hdev, bdaddr_t *addr, - u8 addr_type, u8 auto_connect) -{ - struct hci_conn_params *params; - - params = hci_conn_params_add(hdev, addr, addr_type); - if (!params) - return -EIO; - - if (params->auto_connect == auto_connect) - return 0; - - hci_pend_le_list_del_init(params); - - switch (auto_connect) { - case HCI_AUTO_CONN_DISABLED: - case HCI_AUTO_CONN_LINK_LOSS: - /* If auto connect is being disabled when we're trying to - * connect to device, keep connecting. - */ - if (params->explicit_connect) - hci_pend_le_list_add(params, &hdev->pend_le_conns); - break; - case HCI_AUTO_CONN_REPORT: - if (params->explicit_connect) - hci_pend_le_list_add(params, &hdev->pend_le_conns); - else - hci_pend_le_list_add(params, &hdev->pend_le_reports); - break; - case HCI_AUTO_CONN_DIRECT: - case HCI_AUTO_CONN_ALWAYS: - if (!is_connected(hdev, addr, addr_type)) - hci_pend_le_list_add(params, &hdev->pend_le_conns); - break; - } - - params->auto_connect = auto_connect; - - bt_dev_dbg(hdev, "addr %pMR (type %u) auto_connect %u", - addr, addr_type, auto_connect); - - return 0; -} - static void device_added(struct sock *sk, struct hci_dev *hdev, bdaddr_t *bdaddr, u8 type, u8 action) { @@ -7727,17 +7735,13 @@ static int add_device(struct sock *sk, struct hci_dev *hdev, /* If the connection parameters don't exist for this device, * they will be created and configured with defaults. */ - if (hci_conn_params_set(hdev, &cp->addr.bdaddr, addr_type, - auto_conn) < 0) { + params = hci_conn_params_set(hdev, &cp->addr.bdaddr, addr_type, + auto_conn); + if (!params) { err = mgmt_cmd_complete(sk, hdev->id, MGMT_OP_ADD_DEVICE, MGMT_STATUS_FAILED, &cp->addr, sizeof(cp->addr)); goto unlock; - } else { - params = hci_conn_params_lookup(hdev, &cp->addr.bdaddr, - addr_type); - if (params) - current_flags = params->flags; } cmd = mgmt_pending_new(sk, MGMT_OP_ADD_DEVICE, hdev, data, len); From 577cf4c0a1e8471a0d6c0f36bb3716285e27ad5e Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Mon, 6 Oct 2025 16:53:35 -0400 Subject: [PATCH 1307/1381] Bluetooth: ISO: Fix not updating BIS sender source address The source address for a BIS sender/Broadcast Source shall be updated with the advertisement address since in case privacy is enabled it may use an RPA rather than an identity address. Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/iso.c | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+) diff --git a/net/bluetooth/iso.c b/net/bluetooth/iso.c index fd56eebe9245..4bce03c8a8d3 100644 --- a/net/bluetooth/iso.c +++ b/net/bluetooth/iso.c @@ -2034,6 +2034,25 @@ static void iso_conn_ready(struct iso_conn *conn) BT_DBG("conn %p", conn); if (sk) { + /* Attempt to update source address in case of BIS Sender if + * the advertisement is using a random address. + */ + if (conn->hcon->type == BIS_LINK && + conn->hcon->role == HCI_ROLE_MASTER && + !bacmp(&conn->hcon->dst, BDADDR_ANY)) { + struct hci_conn *bis = conn->hcon; + struct adv_info *adv; + + adv = hci_find_adv_instance(bis->hdev, + bis->iso_qos.bcast.bis); + if (adv && bacmp(&adv->random_addr, BDADDR_ANY)) { + lock_sock(sk); + iso_pi(sk)->src_type = BDADDR_LE_RANDOM; + bacpy(&iso_pi(sk)->src, &adv->random_addr); + release_sock(sk); + } + } + iso_sock_ready(conn->sk); } else { hcon = conn->hcon; From c126f98c011f5796ba118ef2093122d02809d30d Mon Sep 17 00:00:00 2001 From: Chris Lu Date: Wed, 15 Oct 2025 11:31:49 +0800 Subject: [PATCH 1308/1381] Bluetooth: btusb: MT7920: Add VID/PID 0489/e135 Add VID 0489 & PID e135 for MediaTek MT7920 USB Bluetooth chip. The information in /sys/kernel/debug/usb/devices about the Bluetooth device is listed as the below. T: Bus=06 Lev=01 Prnt=01 Port=00 Cnt=01 Dev#= 2 Spd=480 MxCh= 0 D: Ver= 2.10 Cls=ef(misc ) Sub=02 Prot=01 MxPS=64 #Cfgs= 1 P: Vendor=0489 ProdID=e135 Rev= 1.00 S: Manufacturer=MediaTek Inc. S: Product=Wireless_Device S: SerialNumber=000000000 C:* #Ifs= 3 Cfg#= 1 Atr=e0 MxPwr=100mA A: FirstIf#= 0 IfCount= 3 Cls=e0(wlcon) Sub=01 Prot=01 I:* If#= 0 Alt= 0 #EPs= 3 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=81(I) Atr=03(Int.) MxPS= 16 Ivl=125us E: Ad=82(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms E: Ad=02(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms I:* If#= 1 Alt= 0 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=83(I) Atr=01(Isoc) MxPS= 0 Ivl=1ms E: Ad=03(O) Atr=01(Isoc) MxPS= 0 Ivl=1ms I: If#= 1 Alt= 1 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=83(I) Atr=01(Isoc) MxPS= 9 Ivl=1ms E: Ad=03(O) Atr=01(Isoc) MxPS= 9 Ivl=1ms I: If#= 1 Alt= 2 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=83(I) Atr=01(Isoc) MxPS= 17 Ivl=1ms E: Ad=03(O) Atr=01(Isoc) MxPS= 17 Ivl=1ms I: If#= 1 Alt= 3 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=83(I) Atr=01(Isoc) MxPS= 25 Ivl=1ms E: Ad=03(O) Atr=01(Isoc) MxPS= 25 Ivl=1ms I: If#= 1 Alt= 4 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=83(I) Atr=01(Isoc) MxPS= 33 Ivl=1ms E: Ad=03(O) Atr=01(Isoc) MxPS= 33 Ivl=1ms I: If#= 1 Alt= 5 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=83(I) Atr=01(Isoc) MxPS= 49 Ivl=1ms E: Ad=03(O) Atr=01(Isoc) MxPS= 49 Ivl=1ms I: If#= 1 Alt= 6 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=83(I) Atr=01(Isoc) MxPS= 63 Ivl=1ms E: Ad=03(O) Atr=01(Isoc) MxPS= 63 Ivl=1ms I:* If#= 2 Alt= 0 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=(none) E: Ad=8a(I) Atr=03(Int.) MxPS= 64 Ivl=125us E: Ad=0a(O) Atr=03(Int.) MxPS= 64 Ivl=125us I: If#= 2 Alt= 1 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=(none) E: Ad=8a(I) Atr=03(Int.) MxPS= 64 Ivl=125us E: Ad=0a(O) Atr=03(Int.) MxPS= 64 Ivl=125us Signed-off-by: Chris Lu Reviewed-by: Paul Menzel Signed-off-by: Luiz Augusto von Dentz --- drivers/bluetooth/btusb.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c index 9bfa55cb7acf..8529da7fe296 100644 --- a/drivers/bluetooth/btusb.c +++ b/drivers/bluetooth/btusb.c @@ -621,6 +621,8 @@ static const struct usb_device_id quirks_table[] = { /* Additional MediaTek MT7920 Bluetooth devices */ { USB_DEVICE(0x0489, 0xe134), .driver_info = BTUSB_MEDIATEK | BTUSB_WIDEBAND_SPEECH }, + { USB_DEVICE(0x0489, 0xe135), .driver_info = BTUSB_MEDIATEK | + BTUSB_WIDEBAND_SPEECH }, { USB_DEVICE(0x13d3, 0x3620), .driver_info = BTUSB_MEDIATEK | BTUSB_WIDEBAND_SPEECH }, { USB_DEVICE(0x13d3, 0x3621), .driver_info = BTUSB_MEDIATEK | From 5a6700a31c953af9a17a7e2681335f31d922614d Mon Sep 17 00:00:00 2001 From: Chris Lu Date: Wed, 15 Oct 2025 11:31:50 +0800 Subject: [PATCH 1309/1381] Bluetooth: btusb: MT7922: Add VID/PID 0489/e170 Add VID 0489 & PID e170 for MediaTek MT7922 USB Bluetooth chip. The information in /sys/kernel/debug/usb/devices about the Bluetooth device is listed as the below. T: Bus=06 Lev=01 Prnt=01 Port=00 Cnt=01 Dev#= 2 Spd=480 MxCh= 0 D: Ver= 2.10 Cls=ef(misc ) Sub=02 Prot=01 MxPS=64 #Cfgs= 1 P: Vendor=0489 ProdID=e170 Rev= 1.00 S: Manufacturer=MediaTek Inc. S: Product=Wireless_Device S: SerialNumber=000000000 C:* #Ifs= 3 Cfg#= 1 Atr=e0 MxPwr=100mA A: FirstIf#= 0 IfCount= 3 Cls=e0(wlcon) Sub=01 Prot=01 I:* If#= 0 Alt= 0 #EPs= 3 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=81(I) Atr=03(Int.) MxPS= 16 Ivl=125us E: Ad=82(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms E: Ad=02(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms I:* If#= 1 Alt= 0 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=83(I) Atr=01(Isoc) MxPS= 0 Ivl=1ms E: Ad=03(O) Atr=01(Isoc) MxPS= 0 Ivl=1ms I: If#= 1 Alt= 1 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=83(I) Atr=01(Isoc) MxPS= 9 Ivl=1ms E: Ad=03(O) Atr=01(Isoc) MxPS= 9 Ivl=1ms I: If#= 1 Alt= 2 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=83(I) Atr=01(Isoc) MxPS= 17 Ivl=1ms E: Ad=03(O) Atr=01(Isoc) MxPS= 17 Ivl=1ms I: If#= 1 Alt= 3 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=83(I) Atr=01(Isoc) MxPS= 25 Ivl=1ms E: Ad=03(O) Atr=01(Isoc) MxPS= 25 Ivl=1ms I: If#= 1 Alt= 4 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=83(I) Atr=01(Isoc) MxPS= 33 Ivl=1ms E: Ad=03(O) Atr=01(Isoc) MxPS= 33 Ivl=1ms I: If#= 1 Alt= 5 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=83(I) Atr=01(Isoc) MxPS= 49 Ivl=1ms E: Ad=03(O) Atr=01(Isoc) MxPS= 49 Ivl=1ms I: If#= 1 Alt= 6 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=83(I) Atr=01(Isoc) MxPS= 63 Ivl=1ms E: Ad=03(O) Atr=01(Isoc) MxPS= 63 Ivl=1ms I:* If#= 2 Alt= 0 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=(none) E: Ad=8a(I) Atr=03(Int.) MxPS= 64 Ivl=125us E: Ad=0a(O) Atr=03(Int.) MxPS= 64 Ivl=125us I: If#= 2 Alt= 1 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=(none) E: Ad=8a(I) Atr=03(Int.) MxPS= 512 Ivl=125us E: Ad=0a(O) Atr=03(Int.) MxPS= 512 Ivl=125us Signed-off-by: Chris Lu Reviewed-by: Paul Menzel Signed-off-by: Luiz Augusto von Dentz --- drivers/bluetooth/btusb.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c index 8529da7fe296..229fac4c1d33 100644 --- a/drivers/bluetooth/btusb.c +++ b/drivers/bluetooth/btusb.c @@ -687,6 +687,8 @@ static const struct usb_device_id quirks_table[] = { BTUSB_WIDEBAND_SPEECH }, { USB_DEVICE(0x0489, 0xe153), .driver_info = BTUSB_MEDIATEK | BTUSB_WIDEBAND_SPEECH }, + { USB_DEVICE(0x0489, 0xe170), .driver_info = BTUSB_MEDIATEK | + BTUSB_WIDEBAND_SPEECH }, { USB_DEVICE(0x04ca, 0x3804), .driver_info = BTUSB_MEDIATEK | BTUSB_WIDEBAND_SPEECH }, { USB_DEVICE(0x04ca, 0x38e4), .driver_info = BTUSB_MEDIATEK | From 1fb0d830dab89d0dc99bb84a7087b0ceca63d2d8 Mon Sep 17 00:00:00 2001 From: Ravindra Date: Wed, 15 Oct 2025 15:09:02 +0530 Subject: [PATCH 1310/1381] Bluetooth: btintel_pcie: Support for S4 (Hibernate) During S4 (hibernate), the Bluetooth device loses power. Upon resume, the driver performs the following actions: 1. Unregisters hdev 2. Calls function level reset 3. Registers hdev Test case: - run command sudo rtcwake -m disk -s 60 Signed-off-by: Ravindra Signed-off-by: Kiran K Signed-off-by: Luiz Augusto von Dentz --- drivers/bluetooth/btintel_pcie.c | 41 ++++++++++++++++++++++++++++++++ drivers/bluetooth/btintel_pcie.h | 2 ++ 2 files changed, 43 insertions(+) diff --git a/drivers/bluetooth/btintel_pcie.c b/drivers/bluetooth/btintel_pcie.c index a075d8ec4677..f280bcc61bbf 100644 --- a/drivers/bluetooth/btintel_pcie.c +++ b/drivers/bluetooth/btintel_pcie.c @@ -825,6 +825,11 @@ static inline bool btintel_pcie_in_d0(struct btintel_pcie_data *data) return !(data->boot_stage_cache & BTINTEL_PCIE_CSR_BOOT_STAGE_D3_STATE_READY); } +static inline bool btintel_pcie_in_device_halt(struct btintel_pcie_data *data) +{ + return data->boot_stage_cache & BTINTEL_PCIE_CSR_BOOT_STAGE_DEVICE_HALTED; +} + static void btintel_pcie_wr_sleep_cntrl(struct btintel_pcie_data *data, u32 dxstate) { @@ -2532,6 +2537,8 @@ static int btintel_pcie_suspend_late(struct device *dev, pm_message_t mesg) dxstate = (mesg.event == PM_EVENT_SUSPEND ? BTINTEL_PCIE_STATE_D3_HOT : BTINTEL_PCIE_STATE_D3_COLD); + data->pm_sx_event = mesg.event; + data->gp0_received = false; start = ktime_get(); @@ -2581,6 +2588,20 @@ static int btintel_pcie_resume(struct device *dev) start = ktime_get(); + /* When the system enters S4 (hibernate) mode, bluetooth device loses + * power, which results in the erasure of its loaded firmware. + * Consequently, function level reset (flr) is required on system + * resume to bring the controller back into an operational state by + * initiating a new firmware download. + */ + + if (data->pm_sx_event == PM_EVENT_FREEZE || + data->pm_sx_event == PM_EVENT_HIBERNATE) { + set_bit(BTINTEL_PCIE_CORE_HALTED, &data->flags); + btintel_pcie_reset(data->hdev); + return 0; + } + /* Refer: 6.4.11.7 -> Platform power management */ btintel_pcie_wr_sleep_cntrl(data, BTINTEL_PCIE_STATE_D0); err = wait_event_timeout(data->gp0_wait_q, data->gp0_received, @@ -2589,6 +2610,26 @@ static int btintel_pcie_resume(struct device *dev) bt_dev_err(data->hdev, "Timeout (%u ms) on alive interrupt for D0 entry", BTINTEL_DEFAULT_INTR_TIMEOUT_MS); + + /* Trigger function level reset if the controller is in error + * state during resume() to bring back the controller to + * operational mode + */ + + data->boot_stage_cache = btintel_pcie_rd_reg32(data, + BTINTEL_PCIE_CSR_BOOT_STAGE_REG); + if (btintel_pcie_in_error(data) || + btintel_pcie_in_device_halt(data)) { + bt_dev_err(data->hdev, "Controller in error state for D0 entry"); + if (!test_and_set_bit(BTINTEL_PCIE_COREDUMP_INPROGRESS, + &data->flags)) { + data->dmp_hdr.trigger_reason = + BTINTEL_PCIE_TRIGGER_REASON_FW_ASSERT; + queue_work(data->workqueue, &data->rx_work); + } + set_bit(BTINTEL_PCIE_CORE_HALTED, &data->flags); + btintel_pcie_reset(data->hdev); + } return -EBUSY; } diff --git a/drivers/bluetooth/btintel_pcie.h b/drivers/bluetooth/btintel_pcie.h index 04b21f968ad3..48e1ae1793e5 100644 --- a/drivers/bluetooth/btintel_pcie.h +++ b/drivers/bluetooth/btintel_pcie.h @@ -464,6 +464,7 @@ struct btintel_pcie_dump_header { * @txq: TX Queue struct * @rxq: RX Queue struct * @alive_intr_ctxt: Alive interrupt context + * @pm_sx_event: PM event on which system got suspended */ struct btintel_pcie_data { struct pci_dev *pdev; @@ -513,6 +514,7 @@ struct btintel_pcie_data { u32 alive_intr_ctxt; struct btintel_pcie_dbgc dbgc; struct btintel_pcie_dump_header dmp_hdr; + u8 pm_sx_event; }; static inline u32 btintel_pcie_rd_reg32(struct btintel_pcie_data *data, From 88c6216a52ea592ec351dddd042ecb0247325be5 Mon Sep 17 00:00:00 2001 From: Ravindra Date: Wed, 15 Oct 2025 15:09:03 +0530 Subject: [PATCH 1311/1381] Bluetooth: btintel_pcie: Suspend/Resume: Controller doorbell interrupt handling Due to a hardware bug during suspend/resume, the controller may miss a doorbell interrupt. To address this, a retry mechanism has been added to inform the controller before reporting a failure. Test case: - run suspend and resume cycles. Signed-off-by: Ravindra Signed-off-by: Kiran K Signed-off-by: Luiz Augusto von Dentz --- drivers/bluetooth/btintel_pcie.c | 117 +++++++++++++++++++------------ drivers/bluetooth/btintel_pcie.h | 2 + 2 files changed, 75 insertions(+), 44 deletions(-) diff --git a/drivers/bluetooth/btintel_pcie.c b/drivers/bluetooth/btintel_pcie.c index f280bcc61bbf..b0ad3c759ef5 100644 --- a/drivers/bluetooth/btintel_pcie.c +++ b/drivers/bluetooth/btintel_pcie.c @@ -2524,6 +2524,48 @@ static void btintel_pcie_coredump(struct device *dev) } #endif +static int btintel_pcie_set_dxstate(struct btintel_pcie_data *data, u32 dxstate) +{ + int retry = 0, status; + u32 dx_intr_timeout_ms = 200; + + do { + data->gp0_received = false; + + btintel_pcie_wr_sleep_cntrl(data, dxstate); + + status = wait_event_timeout(data->gp0_wait_q, data->gp0_received, + msecs_to_jiffies(dx_intr_timeout_ms)); + + if (status) + return 0; + + bt_dev_warn(data->hdev, + "Timeout (%u ms) on alive interrupt for D%d entry, retry count %d", + dx_intr_timeout_ms, dxstate, retry); + + /* clear gp0 cause */ + btintel_pcie_clr_reg_bits(data, + BTINTEL_PCIE_CSR_MSIX_HW_INT_CAUSES, + BTINTEL_PCIE_MSIX_HW_INT_CAUSES_GP0); + + /* A hardware bug may cause the alive interrupt to be missed. + * Check if the controller reached the expected state and retry + * the operation only if it hasn't. + */ + if (dxstate == BTINTEL_PCIE_STATE_D0) { + if (btintel_pcie_in_d0(data)) + return 0; + } else { + if (btintel_pcie_in_d3(data)) + return 0; + } + + } while (++retry < BTINTEL_PCIE_DX_TRANSITION_MAX_RETRIES); + + return -EBUSY; +} + static int btintel_pcie_suspend_late(struct device *dev, pm_message_t mesg) { struct pci_dev *pdev = to_pci_dev(dev); @@ -2539,26 +2581,18 @@ static int btintel_pcie_suspend_late(struct device *dev, pm_message_t mesg) data->pm_sx_event = mesg.event; - data->gp0_received = false; - start = ktime_get(); /* Refer: 6.4.11.7 -> Platform power management */ - btintel_pcie_wr_sleep_cntrl(data, dxstate); - err = wait_event_timeout(data->gp0_wait_q, data->gp0_received, - msecs_to_jiffies(BTINTEL_DEFAULT_INTR_TIMEOUT_MS)); - if (err == 0) { - bt_dev_err(data->hdev, - "Timeout (%u ms) on alive interrupt for D3 entry", - BTINTEL_DEFAULT_INTR_TIMEOUT_MS); - return -EBUSY; - } + err = btintel_pcie_set_dxstate(data, dxstate); + + if (err) + return err; bt_dev_dbg(data->hdev, "device entered into d3 state from d0 in %lld us", ktime_to_us(ktime_get() - start)); - - return 0; + return err; } static int btintel_pcie_suspend(struct device *dev) @@ -2603,40 +2637,35 @@ static int btintel_pcie_resume(struct device *dev) } /* Refer: 6.4.11.7 -> Platform power management */ - btintel_pcie_wr_sleep_cntrl(data, BTINTEL_PCIE_STATE_D0); - err = wait_event_timeout(data->gp0_wait_q, data->gp0_received, - msecs_to_jiffies(BTINTEL_DEFAULT_INTR_TIMEOUT_MS)); + err = btintel_pcie_set_dxstate(data, BTINTEL_PCIE_STATE_D0); + if (err == 0) { - bt_dev_err(data->hdev, - "Timeout (%u ms) on alive interrupt for D0 entry", - BTINTEL_DEFAULT_INTR_TIMEOUT_MS); - - /* Trigger function level reset if the controller is in error - * state during resume() to bring back the controller to - * operational mode - */ - - data->boot_stage_cache = btintel_pcie_rd_reg32(data, - BTINTEL_PCIE_CSR_BOOT_STAGE_REG); - if (btintel_pcie_in_error(data) || - btintel_pcie_in_device_halt(data)) { - bt_dev_err(data->hdev, "Controller in error state for D0 entry"); - if (!test_and_set_bit(BTINTEL_PCIE_COREDUMP_INPROGRESS, - &data->flags)) { - data->dmp_hdr.trigger_reason = - BTINTEL_PCIE_TRIGGER_REASON_FW_ASSERT; - queue_work(data->workqueue, &data->rx_work); - } - set_bit(BTINTEL_PCIE_CORE_HALTED, &data->flags); - btintel_pcie_reset(data->hdev); - } - return -EBUSY; + bt_dev_dbg(data->hdev, + "device entered into d0 state from d3 in %lld us", + ktime_to_us(ktime_get() - start)); + return err; } - bt_dev_dbg(data->hdev, - "device entered into d0 state from d3 in %lld us", - ktime_to_us(ktime_get() - start)); - return 0; + /* Trigger function level reset if the controller is in error + * state during resume() to bring back the controller to + * operational mode + */ + + data->boot_stage_cache = btintel_pcie_rd_reg32(data, + BTINTEL_PCIE_CSR_BOOT_STAGE_REG); + if (btintel_pcie_in_error(data) || + btintel_pcie_in_device_halt(data)) { + bt_dev_err(data->hdev, "Controller in error state for D0 entry"); + if (!test_and_set_bit(BTINTEL_PCIE_COREDUMP_INPROGRESS, + &data->flags)) { + data->dmp_hdr.trigger_reason = + BTINTEL_PCIE_TRIGGER_REASON_FW_ASSERT; + queue_work(data->workqueue, &data->rx_work); + } + set_bit(BTINTEL_PCIE_CORE_HALTED, &data->flags); + btintel_pcie_reset(data->hdev); + } + return err; } static const struct dev_pm_ops btintel_pcie_pm_ops = { diff --git a/drivers/bluetooth/btintel_pcie.h b/drivers/bluetooth/btintel_pcie.h index 48e1ae1793e5..e3d941ffef4a 100644 --- a/drivers/bluetooth/btintel_pcie.h +++ b/drivers/bluetooth/btintel_pcie.h @@ -158,6 +158,8 @@ enum msix_mbox_int_causes { /* Default interrupt timeout in msec */ #define BTINTEL_DEFAULT_INTR_TIMEOUT_MS 3000 +#define BTINTEL_PCIE_DX_TRANSITION_MAX_RETRIES 3 + /* The number of descriptors in TX queues */ #define BTINTEL_PCIE_TX_DESCS_COUNT 32 From 8dbbb5423c0802ec21266765de80fd491868fab1 Mon Sep 17 00:00:00 2001 From: Chingbin Li Date: Mon, 6 Oct 2025 16:46:47 +0800 Subject: [PATCH 1312/1381] Bluetooth: btusb: Add new VID/PID 2b89/6275 for RTL8761BUV Add VID 2b89 & PID 6275 for Realtek RTL8761BUV USB Bluetooth chip. The information in /sys/kernel/debug/usb/devices about the Bluetooth device is listed as the below. T: Bus=01 Lev=01 Prnt=01 Port=02 Cnt=01 Dev#= 6 Spd=12 MxCh= 0 D: Ver= 1.10 Cls=e0(wlcon) Sub=01 Prot=01 MxPS=64 #Cfgs= 1 P: Vendor=2b89 ProdID=6275 Rev= 2.00 S: Manufacturer=Realtek S: Product=Bluetooth Radio S: SerialNumber=00E04C239987 C:* #Ifs= 2 Cfg#= 1 Atr=e0 MxPwr=500mA I:* If#= 0 Alt= 0 #EPs= 3 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=81(I) Atr=03(Int.) MxPS= 16 Ivl=1ms E: Ad=02(O) Atr=02(Bulk) MxPS= 64 Ivl=0ms E: Ad=82(I) Atr=02(Bulk) MxPS= 64 Ivl=0ms I:* If#= 1 Alt= 0 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=03(O) Atr=01(Isoc) MxPS= 0 Ivl=1ms E: Ad=83(I) Atr=01(Isoc) MxPS= 0 Ivl=1ms I: If#= 1 Alt= 1 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=03(O) Atr=01(Isoc) MxPS= 9 Ivl=1ms E: Ad=83(I) Atr=01(Isoc) MxPS= 9 Ivl=1ms I: If#= 1 Alt= 2 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=03(O) Atr=01(Isoc) MxPS= 17 Ivl=1ms E: Ad=83(I) Atr=01(Isoc) MxPS= 17 Ivl=1ms I: If#= 1 Alt= 3 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=03(O) Atr=01(Isoc) MxPS= 25 Ivl=1ms E: Ad=83(I) Atr=01(Isoc) MxPS= 25 Ivl=1ms I: If#= 1 Alt= 4 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=03(O) Atr=01(Isoc) MxPS= 33 Ivl=1ms E: Ad=83(I) Atr=01(Isoc) MxPS= 33 Ivl=1ms I: If#= 1 Alt= 5 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=03(O) Atr=01(Isoc) MxPS= 49 Ivl=1ms E: Ad=83(I) Atr=01(Isoc) MxPS= 49 Ivl=1ms Signed-off-by: Chingbin Li Signed-off-by: Luiz Augusto von Dentz --- drivers/bluetooth/btusb.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c index 229fac4c1d33..fe2692f5a8ab 100644 --- a/drivers/bluetooth/btusb.c +++ b/drivers/bluetooth/btusb.c @@ -785,6 +785,8 @@ static const struct usb_device_id quirks_table[] = { BTUSB_WIDEBAND_SPEECH }, { USB_DEVICE(0x2b89, 0x8761), .driver_info = BTUSB_REALTEK | BTUSB_WIDEBAND_SPEECH }, + { USB_DEVICE(0x2b89, 0x6275), .driver_info = BTUSB_REALTEK | + BTUSB_WIDEBAND_SPEECH }, /* Additional Realtek 8821AE Bluetooth devices */ { USB_DEVICE(0x0b05, 0x17dc), .driver_info = BTUSB_REALTEK }, From 0b00bee940cb16bfd42e764465705dbf2a95b55c Mon Sep 17 00:00:00 2001 From: Pascal Giard Date: Tue, 7 Oct 2025 11:33:15 -0400 Subject: [PATCH 1313/1381] Bluetooth: btusb: Reclassify Qualcomm WCN6855 debug packets Some Qualcomm Bluetooth controllers, e.g., QCNFA765 with WCN6855 chip, send debug packets as ACL frames with header 0x2EDC. The kernel misinterprets these as malformed ACL packets, causing repeated errors: Bluetooth: hci0: ACL packet for unknown connection handle 3804 This can occur hundreds of times per minute, greatly cluttering logs. On my computer, I am observing approximately 7 messages per second when streaming audio to a speaker. For Qualcomm controllers exchanging over UART, hci_qca.c already filters out these debug packets. This patch is for controllers not going through UART, but USB. This patch uses the classify_pkt_type callback to reclassify the packets with handle 0x2EDC as HCI_DIAG_PKT before they reach the HCI layer. This change is only applied to Qualcomm devices marked as BTUSB_QCA_WCN6855. Tested on: Thinkpad T14 gen2 (AMD) with QCNFA765 (0489:E0D0) Signed-off-by: Pascal Giard Reviewed-by: Paul Menzel Signed-off-by: Luiz Augusto von Dentz --- drivers/bluetooth/btusb.c | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+) diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c index fe2692f5a8ab..7d2a8103dab8 100644 --- a/drivers/bluetooth/btusb.c +++ b/drivers/bluetooth/btusb.c @@ -1137,6 +1137,24 @@ static void btusb_qca_reset(struct hci_dev *hdev) btusb_reset(hdev); } +static u8 btusb_classify_qca_pkt_type(struct hci_dev *hdev, struct sk_buff *skb) +{ + /* Some Qualcomm controllers, e.g., QCNFA765 with WCN6855 chip, send debug + * packets as ACL frames with connection handle 0x2EDC. These are not real + * ACL packets and should be reclassified as HCI_DIAG_PKT to prevent + * "ACL packet for unknown connection handle 3804" errors. + */ + if (skb->len >= 2) { + u16 handle = get_unaligned_le16(skb->data); + + if (handle == 0x2EDC) + return HCI_DIAG_PKT; + } + + /* Use default packet type for other packets */ + return hci_skb_pkt_type(skb); +} + static inline void btusb_free_frags(struct btusb_data *data) { unsigned long flags; @@ -4245,6 +4263,7 @@ static int btusb_probe(struct usb_interface *intf, data->recv_acl = btusb_recv_acl_qca; hci_devcd_register(hdev, btusb_coredump_qca, btusb_dump_hdr_qca, NULL); data->setup_on_usb = btusb_setup_qca; + hdev->classify_pkt_type = btusb_classify_qca_pkt_type; hdev->shutdown = btusb_shutdown_qca; hdev->set_bdaddr = btusb_set_bdaddr_wcn6855; hdev->reset = btusb_qca_reset; From 73d2d709cc172da53402c9b2ead39bebc774ea16 Mon Sep 17 00:00:00 2001 From: Ariel D'Alessandro Date: Wed, 1 Oct 2025 15:33:20 -0300 Subject: [PATCH 1314/1381] dt-bindings: net: Convert Marvell 8897/8997 bindings to DT schema Convert the existing text-based DT bindings for Marvell 8897/8997 (sd8897/sd8997) bluetooth devices controller to a DT schema. While here, bindings for "usb1286,204e" (USB interface) are dropped from the DT schema definition as these are currently documented in file [0]. [0] Documentation/devicetree/bindings/net/btusb.txt Signed-off-by: Ariel D'Alessandro Reviewed-by: Rob Herring (Arm) Signed-off-by: Luiz Augusto von Dentz --- .../net/bluetooth/marvell,sd8897-bt.yaml | 79 ++++++++++++++++++ .../devicetree/bindings/net/btusb.txt | 2 +- .../bindings/net/marvell-bt-8xxx.txt | 83 ------------------- 3 files changed, 80 insertions(+), 84 deletions(-) create mode 100644 Documentation/devicetree/bindings/net/bluetooth/marvell,sd8897-bt.yaml delete mode 100644 Documentation/devicetree/bindings/net/marvell-bt-8xxx.txt diff --git a/Documentation/devicetree/bindings/net/bluetooth/marvell,sd8897-bt.yaml b/Documentation/devicetree/bindings/net/bluetooth/marvell,sd8897-bt.yaml new file mode 100644 index 000000000000..a307c64cfa4d --- /dev/null +++ b/Documentation/devicetree/bindings/net/bluetooth/marvell,sd8897-bt.yaml @@ -0,0 +1,79 @@ +# SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause) +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/net/bluetooth/marvell,sd8897-bt.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Marvell 8897/8997 (sd8897/sd8997) bluetooth devices (SDIO) + +maintainers: + - Ariel D'Alessandro + +allOf: + - $ref: /schemas/net/bluetooth/bluetooth-controller.yaml# + +properties: + compatible: + enum: + - marvell,sd8897-bt + - marvell,sd8997-bt + + reg: + maxItems: 1 + + interrupts: + maxItems: 1 + + marvell,cal-data: + $ref: /schemas/types.yaml#/definitions/uint8-array + description: + Calibration data downloaded to the device during initialization. + maxItems: 28 + + marvell,wakeup-pin: + $ref: /schemas/types.yaml#/definitions/uint16 + description: + Wakeup pin number of the bluetooth chip. Used by firmware to wakeup host + system. + + marvell,wakeup-gap-ms: + $ref: /schemas/types.yaml#/definitions/uint16 + description: + Wakeup latency of the host platform. Required by the chip sleep feature. + +required: + - compatible + - reg + - interrupts + +additionalProperties: false + +examples: + - | + #include + + mmc { + vmmc-supply = <&wlan_en_reg>; + bus-width = <4>; + cap-power-off-card; + keep-power-in-suspend; + + #address-cells = <1>; + #size-cells = <0>; + + bluetooth@2 { + compatible = "marvell,sd8897-bt"; + reg = <2>; + interrupt-parent = <&pio>; + interrupts = <119 IRQ_TYPE_LEVEL_LOW>; + + marvell,cal-data = /bits/ 8 < + 0x37 0x01 0x1c 0x00 0xff 0xff 0xff 0xff 0x01 0x7f 0x04 0x02 + 0x00 0x00 0xba 0xce 0xc0 0xc6 0x2d 0x00 0x00 0x00 0x00 0x00 + 0x00 0x00 0xf0 0x00>; + marvell,wakeup-pin = /bits/ 16 <0x0d>; + marvell,wakeup-gap-ms = /bits/ 16 <0x64>; + }; + }; + +... diff --git a/Documentation/devicetree/bindings/net/btusb.txt b/Documentation/devicetree/bindings/net/btusb.txt index f546b1f7dd6d..a68022a57c51 100644 --- a/Documentation/devicetree/bindings/net/btusb.txt +++ b/Documentation/devicetree/bindings/net/btusb.txt @@ -14,7 +14,7 @@ Required properties: Also, vendors that use btusb may have device additional properties, e.g: -Documentation/devicetree/bindings/net/marvell-bt-8xxx.txt +Documentation/devicetree/bindings/net/bluetooth/marvell,sd8897-bt.yaml Optional properties: diff --git a/Documentation/devicetree/bindings/net/marvell-bt-8xxx.txt b/Documentation/devicetree/bindings/net/marvell-bt-8xxx.txt deleted file mode 100644 index 957e5e5c2927..000000000000 --- a/Documentation/devicetree/bindings/net/marvell-bt-8xxx.txt +++ /dev/null @@ -1,83 +0,0 @@ -Marvell 8897/8997 (sd8897/sd8997) bluetooth devices (SDIO or USB based) ------- -The 8997 devices supports multiple interfaces. When used on SDIO interfaces, -the btmrvl driver is used and when used on USB interface, the btusb driver is -used. - -Required properties: - - - compatible : should be one of the following: - * "marvell,sd8897-bt" (for SDIO) - * "marvell,sd8997-bt" (for SDIO) - * "usb1286,204e" (for USB) - -Optional properties: - - - marvell,cal-data: Calibration data downloaded to the device during - initialization. This is an array of 28 values(u8). - This is only applicable to SDIO devices. - - - marvell,wakeup-pin: It represents wakeup pin number of the bluetooth chip. - firmware will use the pin to wakeup host system (u16). - - marvell,wakeup-gap-ms: wakeup gap represents wakeup latency of the host - platform. The value will be configured to firmware. This - is needed to work chip's sleep feature as expected (u16). - - interrupt-names: Used only for USB based devices (See below) - - interrupts : specifies the interrupt pin number to the cpu. For SDIO, the - driver will use the first interrupt specified in the interrupt - array. For USB based devices, the driver will use the interrupt - named "wakeup" from the interrupt-names and interrupt arrays. - The driver will request an irq based on this interrupt number. - During system suspend, the irq will be enabled so that the - bluetooth chip can wakeup host platform under certain - conditions. During system resume, the irq will be disabled - to make sure unnecessary interrupt is not received. - -Example: - -IRQ pin 119 is used as system wakeup source interrupt. -wakeup pin 13 and gap 100ms are configured so that firmware can wakeup host -using this device side pin and wakeup latency. - -Example for SDIO device follows (calibration data is also available in -below example). - -&mmc3 { - vmmc-supply = <&wlan_en_reg>; - bus-width = <4>; - cap-power-off-card; - keep-power-in-suspend; - - #address-cells = <1>; - #size-cells = <0>; - btmrvl: bluetooth@2 { - compatible = "marvell,sd8897-bt"; - reg = <2>; - interrupt-parent = <&pio>; - interrupts = <119 IRQ_TYPE_LEVEL_LOW>; - - marvell,cal-data = /bits/ 8 < - 0x37 0x01 0x1c 0x00 0xff 0xff 0xff 0xff 0x01 0x7f 0x04 0x02 - 0x00 0x00 0xba 0xce 0xc0 0xc6 0x2d 0x00 0x00 0x00 0x00 0x00 - 0x00 0x00 0xf0 0x00>; - marvell,wakeup-pin = /bits/ 16 <0x0d>; - marvell,wakeup-gap-ms = /bits/ 16 <0x64>; - }; -}; - -Example for USB device: - -&usb_host1_ohci { - #address-cells = <1>; - #size-cells = <0>; - - mvl_bt1: bt@1 { - compatible = "usb1286,204e"; - reg = <1>; - interrupt-parent = <&gpio0>; - interrupt-names = "wakeup"; - interrupts = <119 IRQ_TYPE_LEVEL_LOW>; - marvell,wakeup-pin = /bits/ 16 <0x0d>; - marvell,wakeup-gap-ms = /bits/ 16 <0x64>; - }; -}; From b8414ba5a0e66ab0c5203cd015fc2d2948860af3 Mon Sep 17 00:00:00 2001 From: Sakari Ailus Date: Mon, 27 Oct 2025 15:35:38 +0200 Subject: [PATCH 1315/1381] Bluetooth: Remove redundant pm_runtime_mark_last_busy() calls pm_runtime_put_autosuspend(), pm_runtime_put_sync_autosuspend(), pm_runtime_autosuspend() and pm_request_autosuspend() now include a call to pm_runtime_mark_last_busy(). Remove the now-reduntant explicit call to pm_runtime_mark_last_busy(). Signed-off-by: Sakari Ailus Signed-off-by: Luiz Augusto von Dentz --- drivers/bluetooth/btmtksdio.c | 1 - drivers/bluetooth/hci_bcm.c | 6 +----- drivers/bluetooth/hci_h5.c | 2 -- drivers/bluetooth/hci_intel.c | 3 --- 4 files changed, 1 insertion(+), 11 deletions(-) diff --git a/drivers/bluetooth/btmtksdio.c b/drivers/bluetooth/btmtksdio.c index 62db31bd6592..fba3ab6d30a5 100644 --- a/drivers/bluetooth/btmtksdio.c +++ b/drivers/bluetooth/btmtksdio.c @@ -615,7 +615,6 @@ static void btmtksdio_txrx_work(struct work_struct *work) sdio_release_host(bdev->func); - pm_runtime_mark_last_busy(bdev->dev); pm_runtime_put_autosuspend(bdev->dev); } diff --git a/drivers/bluetooth/hci_bcm.c b/drivers/bluetooth/hci_bcm.c index fff845ed44e3..9286a5f40f55 100644 --- a/drivers/bluetooth/hci_bcm.c +++ b/drivers/bluetooth/hci_bcm.c @@ -326,7 +326,6 @@ static irqreturn_t bcm_host_wake(int irq, void *data) bt_dev_dbg(bdev, "Host wake IRQ"); pm_runtime_get(bdev->dev); - pm_runtime_mark_last_busy(bdev->dev); pm_runtime_put_autosuspend(bdev->dev); return IRQ_HANDLED; @@ -710,7 +709,6 @@ static int bcm_recv(struct hci_uart *hu, const void *data, int count) mutex_lock(&bcm_device_lock); if (bcm->dev && bcm_device_exists(bcm->dev)) { pm_runtime_get(bcm->dev->dev); - pm_runtime_mark_last_busy(bcm->dev->dev); pm_runtime_put_autosuspend(bcm->dev->dev); } mutex_unlock(&bcm_device_lock); @@ -748,10 +746,8 @@ static struct sk_buff *bcm_dequeue(struct hci_uart *hu) skb = skb_dequeue(&bcm->txq); - if (bdev) { - pm_runtime_mark_last_busy(bdev->dev); + if (bdev) pm_runtime_put_autosuspend(bdev->dev); - } mutex_unlock(&bcm_device_lock); diff --git a/drivers/bluetooth/hci_h5.c b/drivers/bluetooth/hci_h5.c index 43371f389fd2..96e20a66ecd1 100644 --- a/drivers/bluetooth/hci_h5.c +++ b/drivers/bluetooth/hci_h5.c @@ -616,7 +616,6 @@ static int h5_recv(struct hci_uart *hu, const void *data, int count) if (hu->serdev) { pm_runtime_get(&hu->serdev->dev); - pm_runtime_mark_last_busy(&hu->serdev->dev); pm_runtime_put_autosuspend(&hu->serdev->dev); } @@ -658,7 +657,6 @@ static int h5_enqueue(struct hci_uart *hu, struct sk_buff *skb) if (hu->serdev) { pm_runtime_get_sync(&hu->serdev->dev); - pm_runtime_mark_last_busy(&hu->serdev->dev); pm_runtime_put_autosuspend(&hu->serdev->dev); } diff --git a/drivers/bluetooth/hci_intel.c b/drivers/bluetooth/hci_intel.c index 1d6e09508f1f..20baf2895dec 100644 --- a/drivers/bluetooth/hci_intel.c +++ b/drivers/bluetooth/hci_intel.c @@ -280,7 +280,6 @@ static irqreturn_t intel_irq(int irq, void *dev_id) /* Host/Controller are now LPM resumed, trigger a new delayed suspend */ pm_runtime_get(&idev->pdev->dev); - pm_runtime_mark_last_busy(&idev->pdev->dev); pm_runtime_put_autosuspend(&idev->pdev->dev); return IRQ_HANDLED; @@ -371,7 +370,6 @@ static void intel_busy_work(struct work_struct *work) list_for_each_entry(idev, &intel_device_list, list) { if (intel->hu->tty->dev->parent == idev->pdev->dev.parent) { pm_runtime_get(&idev->pdev->dev); - pm_runtime_mark_last_busy(&idev->pdev->dev); pm_runtime_put_autosuspend(&idev->pdev->dev); break; } @@ -1003,7 +1001,6 @@ static int intel_enqueue(struct hci_uart *hu, struct sk_buff *skb) list_for_each_entry(idev, &intel_device_list, list) { if (hu->tty->dev->parent == idev->pdev->dev.parent) { pm_runtime_get_sync(&idev->pdev->dev); - pm_runtime_mark_last_busy(&idev->pdev->dev); pm_runtime_put_autosuspend(&idev->pdev->dev); break; } From 4a23ce935f74d19df48d1906497fad1cef391392 Mon Sep 17 00:00:00 2001 From: Max Chou Date: Tue, 28 Oct 2025 14:24:13 +0800 Subject: [PATCH 1316/1381] Bluetooth: btrtl: Add the support for RTL8761CUV Add support for RTL8761CUV BT controller on the USB interface. Do not apply IC_MATCH_FL_HCIVER when hci_ver is 0 in the ic_id_table. The device info from /sys/kernel/debug/usb/devices as below. T: Bus=02 Lev=01 Prnt=01 Port=00 Cnt=01 Dev#= 13 Spd=12 MxCh= 0 D: Ver= 1.10 Cls=e0(wlcon) Sub=01 Prot=01 MxPS=64 #Cfgs= 1 P: Vendor=0bda ProdID=c761 Rev= 2.00 S: Manufacturer=Realtek S: Product=Bluetooth Controller S: SerialNumber=10000 C:* #Ifs= 2 Cfg#= 1 Atr=a0 MxPwr=100mA I:* If#= 0 Alt= 0 #EPs= 3 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=81(I) Atr=03(Int.) MxPS= 64 Ivl=1ms E: Ad=02(O) Atr=02(Bulk) MxPS= 64 Ivl=0ms E: Ad=82(I) Atr=02(Bulk) MxPS= 64 Ivl=0ms I:* If#= 1 Alt= 0 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=83(I) Atr=01(Isoc) MxPS= 0 Ivl=1ms E: Ad=03(O) Atr=01(Isoc) MxPS= 0 Ivl=1ms I: If#= 1 Alt= 1 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=83(I) Atr=01(Isoc) MxPS= 9 Ivl=1ms E: Ad=03(O) Atr=01(Isoc) MxPS= 9 Ivl=1ms I: If#= 1 Alt= 2 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=83(I) Atr=01(Isoc) MxPS= 17 Ivl=1ms E: Ad=03(O) Atr=01(Isoc) MxPS= 17 Ivl=1ms I: If#= 1 Alt= 3 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=83(I) Atr=01(Isoc) MxPS= 25 Ivl=1ms E: Ad=03(O) Atr=01(Isoc) MxPS= 25 Ivl=1ms I: If#= 1 Alt= 4 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=83(I) Atr=01(Isoc) MxPS= 33 Ivl=1ms E: Ad=03(O) Atr=01(Isoc) MxPS= 33 Ivl=1ms I: If#= 1 Alt= 5 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=83(I) Atr=01(Isoc) MxPS= 49 Ivl=1ms E: Ad=03(O) Atr=01(Isoc) MxPS= 49 Ivl=1ms I: If#= 1 Alt= 6 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=83(I) Atr=01(Isoc) MxPS= 63 Ivl=1ms E: Ad=03(O) Atr=01(Isoc) MxPS= 63 Ivl=1ms Signed-off-by: Max Chou Signed-off-by: Luiz Augusto von Dentz --- drivers/bluetooth/btrtl.c | 16 +++++++++++++++- 1 file changed, 15 insertions(+), 1 deletion(-) diff --git a/drivers/bluetooth/btrtl.c b/drivers/bluetooth/btrtl.c index 52794db2739b..5603b282f9bc 100644 --- a/drivers/bluetooth/btrtl.c +++ b/drivers/bluetooth/btrtl.c @@ -72,6 +72,7 @@ enum btrtl_chip_id { CHIP_ID_8851B = 36, CHIP_ID_8922A = 44, CHIP_ID_8852BT = 47, + CHIP_ID_8761C = 51, }; struct id_table { @@ -230,6 +231,14 @@ static const struct id_table ic_id_table[] = { .cfg_name = "rtl_bt/rtl8761bu_config", .hw_info = "rtl8761bu" }, + /* 8761CU */ + { IC_INFO(RTL_ROM_LMP_8761A, 0x0e, 0, HCI_USB), + .config_needed = false, + .has_rom_version = true, + .fw_name = "rtl_bt/rtl8761cu_fw", + .cfg_name = "rtl_bt/rtl8761cu_config", + .hw_info = "rtl8761cu" }, + /* 8822C with UART interface */ { IC_INFO(RTL_ROM_LMP_8822B, 0xc, 0x8, HCI_UART), .config_needed = true, @@ -344,7 +353,8 @@ static const struct id_table *btrtl_match_ic(u16 lmp_subver, u16 hci_rev, (ic_id_table[i].hci_rev != hci_rev)) continue; if ((ic_id_table[i].match_flags & IC_MATCH_FL_HCIVER) && - (ic_id_table[i].hci_ver != hci_ver)) + (ic_id_table[i].hci_ver != hci_ver) && + (ic_id_table[i].hci_ver != 0)) continue; if ((ic_id_table[i].match_flags & IC_MATCH_FL_HCIBUS) && (ic_id_table[i].hci_bus != hci_bus)) @@ -662,6 +672,7 @@ static int rtlbt_parse_firmware(struct hci_dev *hdev, { RTL_ROM_LMP_8851B, 36 }, /* 8851B */ { RTL_ROM_LMP_8922A, 44 }, /* 8922A */ { RTL_ROM_LMP_8852A, 47 }, /* 8852BT */ + { RTL_ROM_LMP_8761A, 51 }, /* 8761C */ }; if (btrtl_dev->fw_len <= 8) @@ -1305,6 +1316,7 @@ void btrtl_set_quirks(struct hci_dev *hdev, struct btrtl_device_info *btrtl_dev) case CHIP_ID_8851B: case CHIP_ID_8922A: case CHIP_ID_8852BT: + case CHIP_ID_8761C: hci_set_quirk(hdev, HCI_QUIRK_WIDEBAND_SPEECH_SUPPORTED); /* RTL8852C needs to transmit mSBC data continuously without @@ -1524,6 +1536,8 @@ MODULE_FIRMWARE("rtl_bt/rtl8761b_fw.bin"); MODULE_FIRMWARE("rtl_bt/rtl8761b_config.bin"); MODULE_FIRMWARE("rtl_bt/rtl8761bu_fw.bin"); MODULE_FIRMWARE("rtl_bt/rtl8761bu_config.bin"); +MODULE_FIRMWARE("rtl_bt/rtl8761cu_fw.bin"); +MODULE_FIRMWARE("rtl_bt/rtl8761cu_config.bin"); MODULE_FIRMWARE("rtl_bt/rtl8821a_fw.bin"); MODULE_FIRMWARE("rtl_bt/rtl8821a_config.bin"); MODULE_FIRMWARE("rtl_bt/rtl8821c_fw.bin"); From 027473ef6a83f28cd6b12551d93dfa24662bdcce Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Tue, 28 Oct 2025 16:53:21 +0100 Subject: [PATCH 1317/1381] Bluetooth: MAINTAINERS: Add Bartosz Golaszewski as Qualcomm hci_qca maintainer There are no dedicated maintainers of Qualcomm hci_qca Bluetooth drivers, but there should be, because these are actively used on many old and new platforms. Bartosz Golaszewski agreed to take care of this code. Link: https://lore.kernel.org/r/CAMRc=MdqAATOcDPhd=u0vOb8nLxSRd7N8rLGLO8F5Ywq3+=JCw@mail.gmail.com/ Signed-off-by: Krzysztof Kozlowski Acked-by: Bartosz Golaszewski Signed-off-by: Luiz Augusto von Dentz --- MAINTAINERS | 1 + 1 file changed, 1 insertion(+) diff --git a/MAINTAINERS b/MAINTAINERS index 09932ab7e0e8..4b2603ed45f4 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -21071,6 +21071,7 @@ F: Documentation/devicetree/bindings/net/qcom,bam-dmux.yaml F: drivers/net/wwan/qcom_bam_dmux.c QUALCOMM BLUETOOTH DRIVER +M: Bartosz Golaszewski L: linux-arm-msm@vger.kernel.org S: Maintained F: drivers/bluetooth/btqca.[ch] From 56f765ce73b4303ec5d85439ca7ce1bc0736d18b Mon Sep 17 00:00:00 2001 From: Yang Li Date: Mon, 27 Oct 2025 14:10:02 +0800 Subject: [PATCH 1318/1381] Bluetooth: iso: fix socket matching ambiguity between BIS and CIS When both BIS and CIS links exist, their sockets are in the BT_LISTEN state. dump sock: sk 000000001977ef51 state 6 src 10:a5:62:31:05:cf dst 00:00:00:00:00:00 sk 0000000031d28700 state 7 src 10:a5:62:31:05:cf dst00:00:00:00:00:00 sk 00000000613af00e state 4 # listen sock of bis src 10:a5:62:31:05:cf dst 54:00:00:d4:99:30 sk 000000001710468c state 9 src 10:a5:62:31:05:cf dst 54:00:00:d4:99:30 sk 000000005d97dfde state 4 #listen sock of cis src 10:a5:62:31:05:cf dst 00:00:00:00:00:00 To locate the CIS socket correctly, check both the BT_LISTEN state and whether dst addr is BDADDR_ANY. Link: https://github.com/bluez/bluez/issues/1224 Signed-off-by: Yang Li Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/iso.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/net/bluetooth/iso.c b/net/bluetooth/iso.c index 4bce03c8a8d3..e36d24a9098b 100644 --- a/net/bluetooth/iso.c +++ b/net/bluetooth/iso.c @@ -2021,6 +2021,11 @@ static bool iso_match_pa_sync_flag(struct sock *sk, void *data) return test_bit(BT_SK_PA_SYNC, &iso_pi(sk)->flags); } +static bool iso_match_dst(struct sock *sk, void *data) +{ + return !bacmp(&iso_pi(sk)->dst, (bdaddr_t *)data); +} + static void iso_conn_ready(struct iso_conn *conn) { struct sock *parent = NULL; @@ -2105,7 +2110,7 @@ static void iso_conn_ready(struct iso_conn *conn) if (!parent) parent = iso_get_sock(hdev, &hcon->src, BDADDR_ANY, - BT_LISTEN, NULL, NULL); + BT_LISTEN, iso_match_dst, BDADDR_ANY); if (!parent) return; @@ -2374,7 +2379,7 @@ int iso_connect_ind(struct hci_dev *hdev, bdaddr_t *bdaddr, __u8 *flags) } } else { sk = iso_get_sock(hdev, &hdev->bdaddr, BDADDR_ANY, - BT_LISTEN, NULL, NULL); + BT_LISTEN, iso_match_dst, BDADDR_ANY); } done: From 32caa197b9b603e20f49fd3a0dffecd0cd620499 Mon Sep 17 00:00:00 2001 From: Max Chou Date: Wed, 5 Nov 2025 13:50:39 +0800 Subject: [PATCH 1319/1381] Bluetooth: btusb: Add new VID/PID 0x0489/0xE12F for RTL8852BE-VT Add the support ID(0x0489, 0xE12F) to usb_device_id table for Realtek RTL8852BE-VT. The device info from /sys/kernel/debug/usb/devices as below. T: Bus=04 Lev=02 Prnt=02 Port=05 Cnt=01 Dev#= 86 Spd=12 MxCh= 0 D: Ver= 1.00 Cls=e0(wlcon) Sub=01 Prot=01 MxPS=64 #Cfgs= 1 P: Vendor=0489 ProdID=e12f Rev= 0.00 S: Manufacturer=Realtek S: Product=Bluetooth Radio S: SerialNumber=00e04c000001 C:* #Ifs= 2 Cfg#= 1 Atr=e0 MxPwr=500mA I:* If#= 0 Alt= 0 #EPs= 3 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=81(I) Atr=03(Int.) MxPS= 16 Ivl=1ms E: Ad=02(O) Atr=02(Bulk) MxPS= 64 Ivl=0ms E: Ad=82(I) Atr=02(Bulk) MxPS= 64 Ivl=0ms I:* If#= 1 Alt= 0 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=03(O) Atr=01(Isoc) MxPS= 0 Ivl=1ms E: Ad=83(I) Atr=01(Isoc) MxPS= 0 Ivl=1ms I: If#= 1 Alt= 1 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=03(O) Atr=01(Isoc) MxPS= 9 Ivl=1ms E: Ad=83(I) Atr=01(Isoc) MxPS= 9 Ivl=1ms I: If#= 1 Alt= 2 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=03(O) Atr=01(Isoc) MxPS= 17 Ivl=1ms E: Ad=83(I) Atr=01(Isoc) MxPS= 17 Ivl=1ms I: If#= 1 Alt= 3 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=03(O) Atr=01(Isoc) MxPS= 25 Ivl=1ms E: Ad=83(I) Atr=01(Isoc) MxPS= 25 Ivl=1ms I: If#= 1 Alt= 4 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=03(O) Atr=01(Isoc) MxPS= 33 Ivl=1ms E: Ad=83(I) Atr=01(Isoc) MxPS= 33 Ivl=1ms I: If#= 1 Alt= 5 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=03(O) Atr=01(Isoc) MxPS= 49 Ivl=1ms E: Ad=83(I) Atr=01(Isoc) MxPS= 49 Ivl=1ms Signed-off-by: Max Chou Signed-off-by: Luiz Augusto von Dentz --- drivers/bluetooth/btusb.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c index 7d2a8103dab8..975e2817804f 100644 --- a/drivers/bluetooth/btusb.c +++ b/drivers/bluetooth/btusb.c @@ -585,6 +585,8 @@ static const struct usb_device_id quirks_table[] = { /* Realtek 8852BT/8852BE-VT Bluetooth devices */ { USB_DEVICE(0x0bda, 0x8520), .driver_info = BTUSB_REALTEK | BTUSB_WIDEBAND_SPEECH }, + { USB_DEVICE(0x0489, 0xe12f), .driver_info = BTUSB_REALTEK | + BTUSB_WIDEBAND_SPEECH }, /* Realtek 8922AE Bluetooth devices */ { USB_DEVICE(0x0bda, 0x8922), .driver_info = BTUSB_REALTEK | From f460768bba2a548abba1df0fd1df5ec77341a539 Mon Sep 17 00:00:00 2001 From: Max Chou Date: Wed, 5 Nov 2025 13:50:40 +0800 Subject: [PATCH 1320/1381] Bluetooth: btusb: Add new VID/PID 0x13d3/0x3618 for RTL8852BE-VT Add the support ID(0x13d3, 0x3618) to usb_device_id table for Realtek RTL8852BE-VT. The device info from /sys/kernel/debug/usb/devices as below. T: Bus=04 Lev=02 Prnt=02 Port=05 Cnt=01 Dev#= 86 Spd=12 MxCh= 0 D: Ver= 1.00 Cls=e0(wlcon) Sub=01 Prot=01 MxPS=64 #Cfgs= 1 P: Vendor=13d3 ProdID=3618 Rev= 0.00 S: Manufacturer=Realtek S: Product=Bluetooth Radio S: SerialNumber=00e04c000001 C:* #Ifs= 2 Cfg#= 1 Atr=e0 MxPwr=500mA I:* If#= 0 Alt= 0 #EPs= 3 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=81(I) Atr=03(Int.) MxPS= 16 Ivl=1ms E: Ad=02(O) Atr=02(Bulk) MxPS= 64 Ivl=0ms E: Ad=82(I) Atr=02(Bulk) MxPS= 64 Ivl=0ms I:* If#= 1 Alt= 0 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=03(O) Atr=01(Isoc) MxPS= 0 Ivl=1ms E: Ad=83(I) Atr=01(Isoc) MxPS= 0 Ivl=1ms I: If#= 1 Alt= 1 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=03(O) Atr=01(Isoc) MxPS= 9 Ivl=1ms E: Ad=83(I) Atr=01(Isoc) MxPS= 9 Ivl=1ms I: If#= 1 Alt= 2 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=03(O) Atr=01(Isoc) MxPS= 17 Ivl=1ms E: Ad=83(I) Atr=01(Isoc) MxPS= 17 Ivl=1ms I: If#= 1 Alt= 3 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=03(O) Atr=01(Isoc) MxPS= 25 Ivl=1ms E: Ad=83(I) Atr=01(Isoc) MxPS= 25 Ivl=1ms I: If#= 1 Alt= 4 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=03(O) Atr=01(Isoc) MxPS= 33 Ivl=1ms E: Ad=83(I) Atr=01(Isoc) MxPS= 33 Ivl=1ms I: If#= 1 Alt= 5 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=03(O) Atr=01(Isoc) MxPS= 49 Ivl=1ms E: Ad=83(I) Atr=01(Isoc) MxPS= 49 Ivl=1ms Signed-off-by: Max Chou Signed-off-by: Luiz Augusto von Dentz --- drivers/bluetooth/btusb.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c index 975e2817804f..05c4643686a3 100644 --- a/drivers/bluetooth/btusb.c +++ b/drivers/bluetooth/btusb.c @@ -587,6 +587,8 @@ static const struct usb_device_id quirks_table[] = { BTUSB_WIDEBAND_SPEECH }, { USB_DEVICE(0x0489, 0xe12f), .driver_info = BTUSB_REALTEK | BTUSB_WIDEBAND_SPEECH }, + { USB_DEVICE(0x13d3, 0x3618), .driver_info = BTUSB_REALTEK | + BTUSB_WIDEBAND_SPEECH }, /* Realtek 8922AE Bluetooth devices */ { USB_DEVICE(0x0bda, 0x8922), .driver_info = BTUSB_REALTEK | From 1216462f4e7c4b7e5cf31545fa5fa4d1c53214cb Mon Sep 17 00:00:00 2001 From: Max Chou Date: Wed, 5 Nov 2025 13:50:41 +0800 Subject: [PATCH 1321/1381] Bluetooth: btusb: Add new VID/PID 0x13d3/0x3619 for RTL8852BE-VT Add the support ID(0x13d3, 0x3619) to usb_device_id table for Realtek RTL8852BE-VT. The device info from /sys/kernel/debug/usb/devices as below. T: Bus=04 Lev=02 Prnt=02 Port=05 Cnt=01 Dev#= 86 Spd=12 MxCh= 0 D: Ver= 1.00 Cls=e0(wlcon) Sub=01 Prot=01 MxPS=64 #Cfgs= 1 P: Vendor=13d3 ProdID=3619 Rev= 0.00 S: Manufacturer=Realtek S: Product=Bluetooth Radio S: SerialNumber=00e04c000001 C:* #Ifs= 2 Cfg#= 1 Atr=e0 MxPwr=500mA I:* If#= 0 Alt= 0 #EPs= 3 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=81(I) Atr=03(Int.) MxPS= 16 Ivl=1ms E: Ad=02(O) Atr=02(Bulk) MxPS= 64 Ivl=0ms E: Ad=82(I) Atr=02(Bulk) MxPS= 64 Ivl=0ms I:* If#= 1 Alt= 0 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=03(O) Atr=01(Isoc) MxPS= 0 Ivl=1ms E: Ad=83(I) Atr=01(Isoc) MxPS= 0 Ivl=1ms I: If#= 1 Alt= 1 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=03(O) Atr=01(Isoc) MxPS= 9 Ivl=1ms E: Ad=83(I) Atr=01(Isoc) MxPS= 9 Ivl=1ms I: If#= 1 Alt= 2 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=03(O) Atr=01(Isoc) MxPS= 17 Ivl=1ms E: Ad=83(I) Atr=01(Isoc) MxPS= 17 Ivl=1ms I: If#= 1 Alt= 3 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=03(O) Atr=01(Isoc) MxPS= 25 Ivl=1ms E: Ad=83(I) Atr=01(Isoc) MxPS= 25 Ivl=1ms I: If#= 1 Alt= 4 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=03(O) Atr=01(Isoc) MxPS= 33 Ivl=1ms E: Ad=83(I) Atr=01(Isoc) MxPS= 33 Ivl=1ms I: If#= 1 Alt= 5 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=03(O) Atr=01(Isoc) MxPS= 49 Ivl=1ms E: Ad=83(I) Atr=01(Isoc) MxPS= 49 Ivl=1ms Signed-off-by: Max Chou Signed-off-by: Luiz Augusto von Dentz --- drivers/bluetooth/btusb.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c index 05c4643686a3..ef2b4d3996e6 100644 --- a/drivers/bluetooth/btusb.c +++ b/drivers/bluetooth/btusb.c @@ -589,6 +589,8 @@ static const struct usb_device_id quirks_table[] = { BTUSB_WIDEBAND_SPEECH }, { USB_DEVICE(0x13d3, 0x3618), .driver_info = BTUSB_REALTEK | BTUSB_WIDEBAND_SPEECH }, + { USB_DEVICE(0x13d3, 0x3619), .driver_info = BTUSB_REALTEK | + BTUSB_WIDEBAND_SPEECH }, /* Realtek 8922AE Bluetooth devices */ { USB_DEVICE(0x0bda, 0x8922), .driver_info = BTUSB_REALTEK | From a8b38d19857d42a1f2e90c9d9b0f74de2500acd7 Mon Sep 17 00:00:00 2001 From: Shuai Zhang Date: Sun, 9 Nov 2025 17:24:37 +0800 Subject: [PATCH 1322/1381] Bluetooth: btusb: add new custom firmwares The new platform uses the QCA2066 chip along with a new board ID, which requires a dedicated firmware file to ensure proper initialization. Without this entry, the driver cannot locate and load the correct firmware, resulting in Bluetooth bring-up failure. This patch adds a new entry to the firmware table for QCA2066 so that the driver can correctly identify the board ID and load the appropriate firmware from 'qca/QCA2066/' in the linux-firmware repository. Signed-off-by: Shuai Zhang Acked-by: Dmitry Baryshkov Signed-off-by: Luiz Augusto von Dentz --- drivers/bluetooth/btusb.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c index ef2b4d3996e6..683ac02e964b 100644 --- a/drivers/bluetooth/btusb.c +++ b/drivers/bluetooth/btusb.c @@ -3304,6 +3304,7 @@ static const struct qca_device_info qca_devices_table[] = { static const struct qca_custom_firmware qca_custom_btfws[] = { { 0x00130201, 0x030A, "QCA2066" }, + { 0x00130201, 0x030B, "QCA2066" }, { }, }; From bc6f557b3387d807e08ffb8d638f149fedb75bf6 Mon Sep 17 00:00:00 2001 From: Chethan T N Date: Fri, 7 Nov 2025 11:10:11 +0530 Subject: [PATCH 1323/1381] Bluetooth: btintel_pcie: Introduce HCI Driver protocol This patch adds the infrastructure that allow the user space program to talk to intel pcie driver directly for fetching basic driver details. The changes introduced are referred form commit 04425292a62c15 ("Bluetooth: Introduce HCI Driver protocol") Signed-off-by: Chethan T N Signed-off-by: Luiz Augusto von Dentz --- drivers/bluetooth/btintel_pcie.c | 59 ++++++++++++++++++++++++++++++++ 1 file changed, 59 insertions(+) diff --git a/drivers/bluetooth/btintel_pcie.c b/drivers/bluetooth/btintel_pcie.c index b0ad3c759ef5..2936b535479f 100644 --- a/drivers/bluetooth/btintel_pcie.c +++ b/drivers/bluetooth/btintel_pcie.c @@ -19,6 +19,7 @@ #include #include +#include #include "btintel.h" #include "btintel_pcie.h" @@ -2360,6 +2361,63 @@ static bool btintel_pcie_wakeup(struct hci_dev *hdev) return device_may_wakeup(&data->pdev->dev); } +static const struct { + u16 opcode; + const char *desc; +} btintel_pcie_hci_drv_supported_commands[] = { + /* Common commands */ + { HCI_DRV_OP_READ_INFO, "Read Info" }, +}; + +static int btintel_pcie_hci_drv_read_info(struct hci_dev *hdev, void *data, + u16 data_len) +{ + struct hci_drv_rp_read_info *rp; + size_t rp_size; + int err, i; + u16 opcode, num_supported_commands = + ARRAY_SIZE(btintel_pcie_hci_drv_supported_commands); + + rp_size = sizeof(*rp) + num_supported_commands * 2; + + rp = kmalloc(rp_size, GFP_KERNEL); + if (!rp) + return -ENOMEM; + + strscpy_pad(rp->driver_name, KBUILD_MODNAME); + + rp->num_supported_commands = cpu_to_le16(num_supported_commands); + for (i = 0; i < num_supported_commands; i++) { + opcode = btintel_pcie_hci_drv_supported_commands[i].opcode; + bt_dev_dbg(hdev, + "Supported HCI Drv command (0x%02x|0x%04x): %s", + hci_opcode_ogf(opcode), + hci_opcode_ocf(opcode), + btintel_pcie_hci_drv_supported_commands[i].desc); + rp->supported_commands[i] = cpu_to_le16(opcode); + } + + err = hci_drv_cmd_complete(hdev, HCI_DRV_OP_READ_INFO, + HCI_DRV_STATUS_SUCCESS, + rp, rp_size); + + kfree(rp); + return err; +} + +static const struct hci_drv_handler btintel_pcie_hci_drv_common_handlers[] = { + { btintel_pcie_hci_drv_read_info, HCI_DRV_READ_INFO_SIZE }, +}; + +static const struct hci_drv_handler btintel_pcie_hci_drv_specific_handlers[] = {}; + +static struct hci_drv btintel_pcie_hci_drv = { + .common_handler_count = ARRAY_SIZE(btintel_pcie_hci_drv_common_handlers), + .common_handlers = btintel_pcie_hci_drv_common_handlers, + .specific_handler_count = ARRAY_SIZE(btintel_pcie_hci_drv_specific_handlers), + .specific_handlers = btintel_pcie_hci_drv_specific_handlers, +}; + static int btintel_pcie_setup_hdev(struct btintel_pcie_data *data) { int err; @@ -2386,6 +2444,7 @@ static int btintel_pcie_setup_hdev(struct btintel_pcie_data *data) hdev->set_bdaddr = btintel_set_bdaddr; hdev->reset = btintel_pcie_reset; hdev->wakeup = btintel_pcie_wakeup; + hdev->hci_drv = &btintel_pcie_hci_drv; err = hci_register_dev(hdev); if (err < 0) { From 6f7cf13ef6b0fe2bdd539e5aa1b1fc8a1213cfc3 Mon Sep 17 00:00:00 2001 From: Ayaan Mirza Baig Date: Tue, 11 Nov 2025 19:50:41 +0530 Subject: [PATCH 1324/1381] drivers/bluetooth: btbcm: Use kmalloc_array() to prevent overflow Replace the open-coded multiplication in kmalloc() with a call to kmalloc_array() to prevent potential integer overflows. This is a mechanical change, replacing BCM_FW_NAME_LEN with the type-safe sizeof(*fw_name) as the element size Signed-off-by: Ayaan Mirza Baig Signed-off-by: Luiz Augusto von Dentz --- drivers/bluetooth/btbcm.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/bluetooth/btbcm.c b/drivers/bluetooth/btbcm.c index 3a3a56ddbb06..d33cc70eec66 100644 --- a/drivers/bluetooth/btbcm.c +++ b/drivers/bluetooth/btbcm.c @@ -642,7 +642,9 @@ int btbcm_initialize(struct hci_dev *hdev, bool *fw_load_done, bool use_autobaud snprintf(postfix, sizeof(postfix), "-%4.4x-%4.4x", vid, pid); } - fw_name = kmalloc(BCM_FW_NAME_COUNT_MAX * BCM_FW_NAME_LEN, GFP_KERNEL); + fw_name = kmalloc_array(BCM_FW_NAME_COUNT_MAX, + sizeof(*fw_name), + GFP_KERNEL); if (!fw_name) return -ENOMEM; From a106e50be74b0896583f4d010a69f9806e4194f4 Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Fri, 14 Nov 2025 09:29:28 -0500 Subject: [PATCH 1325/1381] Bluetooth: HCI: Add support for LL Extended Feature Set MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This adds support for emulating LL Extended Feature Set introduced in 6.0 that adds the following: Commands: - HCI_LE_Read_All_Local_Supported_­Features(0x2087)(Feature:47,1) - HCI_LE_Read_All_Remote_Features(0x2088)(Feature:47,2) Events: - HCI_LE_Read_All_Remote_Features_Complete(0x2b)(Mask bit:42) Signed-off-by: Luiz Augusto von Dentz --- include/net/bluetooth/hci.h | 23 ++++++ include/net/bluetooth/hci_core.h | 5 +- include/net/bluetooth/hci_sync.h | 2 + net/bluetooth/hci_event.c | 125 ++++++++++++++++++++++++------- net/bluetooth/hci_sync.c | 102 ++++++++++++++++++++++++- 5 files changed, 227 insertions(+), 30 deletions(-) diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h index d883ad233ebc..a27cd3626b87 100644 --- a/include/net/bluetooth/hci.h +++ b/include/net/bluetooth/hci.h @@ -653,6 +653,7 @@ enum { #define HCI_LE_CIS_PERIPHERAL 0x20 #define HCI_LE_ISO_BROADCASTER 0x40 #define HCI_LE_ISO_SYNC_RECEIVER 0x80 +#define HCI_LE_LL_EXT_FEATURE 0x80 /* Connection modes */ #define HCI_CM_ACTIVE 0x0000 @@ -2255,6 +2256,19 @@ struct hci_cp_le_set_host_feature { __u8 bit_value; } __packed; +#define HCI_OP_LE_READ_ALL_LOCAL_FEATURES 0x2087 +struct hci_rp_le_read_all_local_features { + __u8 status; + __u8 page; + __u8 features[248]; +} __packed; + +#define HCI_OP_LE_READ_ALL_REMOTE_FEATURES 0x2088 +struct hci_cp_le_read_all_remote_features { + __le16 handle; + __u8 pages; +} __packed; + /* ---- HCI Events ---- */ struct hci_ev_status { __u8 status; @@ -2937,6 +2951,15 @@ struct hci_evt_le_big_info_adv_report { __u8 encryption; } __packed; +#define HCI_EVT_LE_ALL_REMOTE_FEATURES_COMPLETE 0x2b +struct hci_evt_le_read_all_remote_features_complete { + __u8 status; + __le16 handle; + __u8 max_pages; + __u8 valid_pages; + __u8 features[248]; +} __packed; + #define HCI_EV_VENDOR 0xff /* Internal events generated by Bluetooth stack */ diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index 858b58206e80..4263e71a23ef 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -378,7 +378,7 @@ struct hci_dev { __u8 minor_class; __u8 max_page; __u8 features[HCI_MAX_PAGES][8]; - __u8 le_features[8]; + __u8 le_features[248]; __u8 le_accept_list_size; __u8 le_resolv_list_size; __u8 le_num_of_adv_sets; @@ -702,6 +702,7 @@ struct hci_conn { __u8 attempt; __u8 dev_class[3]; __u8 features[HCI_MAX_PAGES][8]; + __u8 le_features[248]; __u16 pkt_type; __u16 link_policy; __u8 key_type; @@ -2067,6 +2068,8 @@ void hci_conn_del_sysfs(struct hci_conn *conn); (le_enabled(dev) && past_receiver_capable(dev)) #define past_enabled(dev) \ (past_sender_enabled(dev) || past_receiver_enabled(dev)) +#define ll_ext_feature_capable(dev) \ + ((dev)->le_features[7] & HCI_LE_LL_EXT_FEATURE) #define mws_transport_config_capable(dev) (((dev)->commands[30] & 0x08) && \ (!hci_test_quirk((dev), HCI_QUIRK_BROKEN_MWS_TRANSPORT_CONFIG))) diff --git a/include/net/bluetooth/hci_sync.h b/include/net/bluetooth/hci_sync.h index 3133f40fa9f9..56076bbc981d 100644 --- a/include/net/bluetooth/hci_sync.h +++ b/include/net/bluetooth/hci_sync.h @@ -189,3 +189,5 @@ int hci_le_conn_update_sync(struct hci_dev *hdev, struct hci_conn *conn, int hci_connect_pa_sync(struct hci_dev *hdev, struct hci_conn *conn); int hci_connect_big_sync(struct hci_dev *hdev, struct hci_conn *conn); int hci_past_sync(struct hci_conn *conn, struct hci_conn *le); + +int hci_le_read_remote_features(struct hci_conn *conn); diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c index 7c4ca14f13e5..a9868f17ef40 100644 --- a/net/bluetooth/hci_event.c +++ b/net/bluetooth/hci_event.c @@ -2886,12 +2886,8 @@ static void hci_cs_le_read_remote_features(struct hci_dev *hdev, u8 status) hci_dev_lock(hdev); conn = hci_conn_hash_lookup_handle(hdev, __le16_to_cpu(cp->handle)); - if (conn) { - if (conn->state == BT_CONFIG) { - hci_connect_cfm(conn, status); - hci_conn_drop(conn); - } - } + if (conn && conn->state == BT_CONFIG) + hci_connect_cfm(conn, status); hci_dev_unlock(hdev); } @@ -3915,11 +3911,49 @@ static u8 hci_cc_le_setup_iso_path(struct hci_dev *hdev, void *data, return rp->status; } +static u8 hci_cc_le_read_all_local_features(struct hci_dev *hdev, void *data, + struct sk_buff *skb) +{ + struct hci_rp_le_read_all_local_features *rp = data; + + bt_dev_dbg(hdev, "status 0x%2.2x", rp->status); + + if (rp->status) + return rp->status; + + memcpy(hdev->le_features, rp->features, 248); + + return rp->status; +} + static void hci_cs_le_create_big(struct hci_dev *hdev, u8 status) { bt_dev_dbg(hdev, "status 0x%2.2x", status); } +static void hci_cs_le_read_all_remote_features(struct hci_dev *hdev, u8 status) +{ + struct hci_cp_le_read_remote_features *cp; + struct hci_conn *conn; + + bt_dev_dbg(hdev, "status 0x%2.2x", status); + + if (!status) + return; + + cp = hci_sent_cmd_data(hdev, HCI_OP_LE_READ_ALL_REMOTE_FEATURES); + if (!cp) + return; + + hci_dev_lock(hdev); + + conn = hci_conn_hash_lookup_handle(hdev, __le16_to_cpu(cp->handle)); + if (conn && conn->state == BT_CONFIG) + hci_connect_cfm(conn, status); + + hci_dev_unlock(hdev); +} + static u8 hci_cc_set_per_adv_param(struct hci_dev *hdev, void *data, struct sk_buff *skb) { @@ -4171,6 +4205,9 @@ static const struct hci_cc { sizeof(struct hci_rp_le_set_cig_params), HCI_MAX_EVENT_SIZE), HCI_CC(HCI_OP_LE_SETUP_ISO_PATH, hci_cc_le_setup_iso_path, sizeof(struct hci_rp_le_setup_iso_path)), + HCI_CC(HCI_OP_LE_READ_ALL_LOCAL_FEATURES, + hci_cc_le_read_all_local_features, + sizeof(struct hci_rp_le_read_all_local_features)), }; static u8 hci_cc_func(struct hci_dev *hdev, const struct hci_cc *cc, @@ -4325,6 +4362,8 @@ static const struct hci_cs { HCI_CS(HCI_OP_LE_EXT_CREATE_CONN, hci_cs_le_ext_create_conn), HCI_CS(HCI_OP_LE_CREATE_CIS, hci_cs_le_create_cis), HCI_CS(HCI_OP_LE_CREATE_BIG, hci_cs_le_create_big), + HCI_CS(HCI_OP_LE_READ_ALL_REMOTE_FEATURES, + hci_cs_le_read_all_remote_features), }; static void hci_cmd_status_evt(struct hci_dev *hdev, void *data, @@ -5645,6 +5684,7 @@ static void le_conn_complete_evt(struct hci_dev *hdev, u8 status, struct hci_conn *conn; struct smp_irk *irk; u8 addr_type; + int err; hci_dev_lock(hdev); @@ -5775,26 +5815,8 @@ static void le_conn_complete_evt(struct hci_dev *hdev, u8 status, hci_debugfs_create_conn(conn); hci_conn_add_sysfs(conn); - /* The remote features procedure is defined for central - * role only. So only in case of an initiated connection - * request the remote features. - * - * If the local controller supports peripheral-initiated features - * exchange, then requesting the remote features in peripheral - * role is possible. Otherwise just transition into the - * connected state without requesting the remote features. - */ - if (conn->out || - (hdev->le_features[0] & HCI_LE_PERIPHERAL_FEATURES)) { - struct hci_cp_le_read_remote_features cp; - - cp.handle = __cpu_to_le16(conn->handle); - - hci_send_cmd(hdev, HCI_OP_LE_READ_REMOTE_FEATURES, - sizeof(cp), &cp); - - hci_conn_hold(conn); - } else { + err = hci_le_read_remote_features(conn); + if (err) { conn->state = BT_CONNECTED; hci_connect_cfm(conn, status); } @@ -6608,7 +6630,6 @@ static void hci_le_remote_feat_complete_evt(struct hci_dev *hdev, void *data, conn->state = BT_CONNECTED; hci_connect_cfm(conn, status); - hci_conn_drop(conn); } } @@ -7186,6 +7207,50 @@ static void hci_le_big_info_adv_report_evt(struct hci_dev *hdev, void *data, hci_dev_unlock(hdev); } +static void hci_le_read_all_remote_features_evt(struct hci_dev *hdev, + void *data, struct sk_buff *skb) +{ + struct hci_evt_le_read_all_remote_features_complete *ev = data; + struct hci_conn *conn; + + bt_dev_dbg(hdev, "status 0x%2.2x", ev->status); + + hci_dev_lock(hdev); + + conn = hci_conn_hash_lookup_handle(hdev, __le16_to_cpu(ev->handle)); + if (!conn) + goto unlock; + + if (!ev->status) + memcpy(conn->le_features, ev->features, 248); + + if (conn->state == BT_CONFIG) { + __u8 status; + + /* If the local controller supports peripheral-initiated + * features exchange, but the remote controller does + * not, then it is possible that the error code 0x1a + * for unsupported remote feature gets returned. + * + * In this specific case, allow the connection to + * transition into connected state and mark it as + * successful. + */ + if (!conn->out && + ev->status == HCI_ERROR_UNSUPPORTED_REMOTE_FEATURE && + (hdev->le_features[0] & HCI_LE_PERIPHERAL_FEATURES)) + status = 0x00; + else + status = ev->status; + + conn->state = BT_CONNECTED; + hci_connect_cfm(conn, status); + } + +unlock: + hci_dev_unlock(hdev); +} + #define HCI_LE_EV_VL(_op, _func, _min_len, _max_len) \ [_op] = { \ .func = _func, \ @@ -7291,6 +7356,12 @@ static const struct hci_le_ev { hci_le_big_info_adv_report_evt, sizeof(struct hci_evt_le_big_info_adv_report), HCI_MAX_EVENT_SIZE), + /* [0x2b = HCI_EVT_LE_ALL_REMOTE_FEATURES_COMPLETE] */ + HCI_LE_EV_VL(HCI_EVT_LE_ALL_REMOTE_FEATURES_COMPLETE, + hci_le_read_all_remote_features_evt, + sizeof(struct + hci_evt_le_read_all_remote_features_complete), + HCI_MAX_EVENT_SIZE), }; static void hci_le_meta_evt(struct hci_dev *hdev, void *data, diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c index a36d2414a3ca..a9f5b1a68356 100644 --- a/net/bluetooth/hci_sync.c +++ b/net/bluetooth/hci_sync.c @@ -4011,8 +4011,19 @@ static int hci_le_read_buffer_size_sync(struct hci_dev *hdev) /* Read LE Local Supported Features */ static int hci_le_read_local_features_sync(struct hci_dev *hdev) { - return __hci_cmd_sync_status(hdev, HCI_OP_LE_READ_LOCAL_FEATURES, - 0, NULL, HCI_CMD_TIMEOUT); + int err; + + err = __hci_cmd_sync_status(hdev, HCI_OP_LE_READ_LOCAL_FEATURES, + 0, NULL, HCI_CMD_TIMEOUT); + if (err) + return err; + + if (ll_ext_feature_capable(hdev) && hdev->commands[47] & BIT(2)) + return __hci_cmd_sync_status(hdev, + HCI_OP_LE_READ_ALL_LOCAL_FEATURES, + 0, NULL, HCI_CMD_TIMEOUT); + + return err; } /* Read LE Supported States */ @@ -7320,3 +7331,90 @@ int hci_past_sync(struct hci_conn *conn, struct hci_conn *le) return err; } + +static void le_read_features_complete(struct hci_dev *hdev, void *data, int err) +{ + struct hci_conn *conn = data; + + bt_dev_dbg(hdev, "err %d", err); + + if (err == -ECANCELED) + return; + + hci_conn_drop(conn); +} + +static int hci_le_read_all_remote_features_sync(struct hci_dev *hdev, + void *data) +{ + struct hci_conn *conn = data; + struct hci_cp_le_read_all_remote_features cp; + + memset(&cp, 0, sizeof(cp)); + cp.handle = cpu_to_le16(conn->handle); + cp.pages = 10; /* Attempt to read all pages */ + + /* Wait for HCI_EVT_LE_ALL_REMOTE_FEATURES_COMPLETE event otherwise + * hci_conn_drop may run prematurely causing a disconnection. + */ + return __hci_cmd_sync_status_sk(hdev, + HCI_OP_LE_READ_ALL_REMOTE_FEATURES, + sizeof(cp), &cp, + HCI_EVT_LE_ALL_REMOTE_FEATURES_COMPLETE, + HCI_CMD_TIMEOUT, NULL); + + return __hci_cmd_sync_status(hdev, HCI_OP_LE_READ_ALL_REMOTE_FEATURES, + sizeof(cp), &cp, HCI_CMD_TIMEOUT); +} + +static int hci_le_read_remote_features_sync(struct hci_dev *hdev, void *data) +{ + struct hci_conn *conn = data; + struct hci_cp_le_read_remote_features cp; + + if (!hci_conn_valid(hdev, conn)) + return -ECANCELED; + + /* Check if LL Extended Feature Set is supported and + * HCI_OP_LE_READ_ALL_REMOTE_FEATURES is supported then use that to read + * all features. + */ + if (ll_ext_feature_capable(hdev) && hdev->commands[47] & BIT(3)) + return hci_le_read_all_remote_features_sync(hdev, data); + + memset(&cp, 0, sizeof(cp)); + cp.handle = cpu_to_le16(conn->handle); + + /* Wait for HCI_EV_LE_REMOTE_FEAT_COMPLETE event otherwise + * hci_conn_drop may run prematurely causing a disconnection. + */ + return __hci_cmd_sync_status_sk(hdev, HCI_OP_LE_READ_REMOTE_FEATURES, + sizeof(cp), &cp, + HCI_EV_LE_REMOTE_FEAT_COMPLETE, + HCI_CMD_TIMEOUT, NULL); +} + +int hci_le_read_remote_features(struct hci_conn *conn) +{ + struct hci_dev *hdev = conn->hdev; + int err; + + /* The remote features procedure is defined for central + * role only. So only in case of an initiated connection + * request the remote features. + * + * If the local controller supports peripheral-initiated features + * exchange, then requesting the remote features in peripheral + * role is possible. Otherwise just transition into the + * connected state without requesting the remote features. + */ + if (conn->out || (hdev->le_features[0] & HCI_LE_PERIPHERAL_FEATURES)) + err = hci_cmd_sync_queue_once(hdev, + hci_le_read_remote_features_sync, + hci_conn_hold(conn), + le_read_features_complete); + else + err = -EOPNOTSUPP; + + return err; +} From 525459da4bd62a81142fea3f3d52188ceb4d8907 Mon Sep 17 00:00:00 2001 From: Gongwei Li Date: Wed, 19 Nov 2025 15:33:38 +0800 Subject: [PATCH 1326/1381] Bluetooth: btusb: Add new VID/PID 13d3/3533 for RTL8821CE Add VID 13d3 & PID 3533 for Realtek RTL8821CE USB Bluetooth chip. The information in /sys/kernel/debug/usb/devices about the Bluetooth device is listed as the below. T: Bus=01 Lev=01 Prnt=01 Port=00 Cnt=01 Dev#= 2 Spd=12 MxCh= 0 D: Ver= 1.10 Cls=e0(wlcon) Sub=01 Prot=01 MxPS=64 #Cfgs= 1 P: Vendor=13d3 ProdID=3533 Rev= 1.10 S: Manufacturer=Realtek S: Product=Bluetooth Radio S: SerialNumber=00e04c000001 C:* #Ifs= 2 Cfg#= 1 Atr=e0 MxPwr=500mA I:* If#= 0 Alt= 0 #EPs= 3 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=81(I) Atr=03(Int.) MxPS= 16 Ivl=1ms E: Ad=02(O) Atr=02(Bulk) MxPS= 64 Ivl=0ms E: Ad=82(I) Atr=02(Bulk) MxPS= 64 Ivl=0ms I:* If#= 1 Alt= 0 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=03(O) Atr=01(Isoc) MxPS= 0 Ivl=1ms E: Ad=83(I) Atr=01(Isoc) MxPS= 0 Ivl=1ms I: If#= 1 Alt= 1 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=03(O) Atr=01(Isoc) MxPS= 9 Ivl=1ms E: Ad=83(I) Atr=01(Isoc) MxPS= 9 Ivl=1ms I: If#= 1 Alt= 2 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=03(O) Atr=01(Isoc) MxPS= 17 Ivl=1ms E: Ad=83(I) Atr=01(Isoc) MxPS= 17 Ivl=1ms I: If#= 1 Alt= 3 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=03(O) Atr=01(Isoc) MxPS= 25 Ivl=1ms E: Ad=83(I) Atr=01(Isoc) MxPS= 25 Ivl=1ms I: If#= 1 Alt= 4 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=03(O) Atr=01(Isoc) MxPS= 33 Ivl=1ms E: Ad=83(I) Atr=01(Isoc) MxPS= 33 Ivl=1ms I: If#= 1 Alt= 5 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=03(O) Atr=01(Isoc) MxPS= 49 Ivl=1ms E: Ad=83(I) Atr=01(Isoc) MxPS= 49 Ivl=1ms Signed-off-by: Gongwei Li Signed-off-by: Luiz Augusto von Dentz --- drivers/bluetooth/btusb.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c index 683ac02e964b..8ed3883ab8ee 100644 --- a/drivers/bluetooth/btusb.c +++ b/drivers/bluetooth/btusb.c @@ -504,6 +504,8 @@ static const struct usb_device_id quirks_table[] = { /* Realtek 8821CE Bluetooth devices */ { USB_DEVICE(0x13d3, 0x3529), .driver_info = BTUSB_REALTEK | BTUSB_WIDEBAND_SPEECH }, + { USB_DEVICE(0x13d3, 0x3533), .driver_info = BTUSB_REALTEK | + BTUSB_WIDEBAND_SPEECH }, /* Realtek 8822CE Bluetooth devices */ { USB_DEVICE(0x0bda, 0xb00c), .driver_info = BTUSB_REALTEK | From 56435627d90f252017237fa66a9fa081b11113f9 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Tue, 25 Nov 2025 12:35:07 +0200 Subject: [PATCH 1327/1381] net: pcs: lynx: accept in-band autoneg for 2500base-x Testing in two circumstances: 1. back to back optical SFP+ connection between two LS1028A-QDS ports with the SCH-26908 riser card 2. T1042 with on-board AQR115 PHY using "OCSGMII", as per https://lore.kernel.org/lkml/aIuEvaSCIQdJWcZx@FUE-ALEWI-WINX/ strongly suggests that enabling in-band auto-negotiation is actually possible when the lane baud rate is 3.125 Gbps. It was previously thought that this would not be the case, because it was only tested on 2500base-x links with on-board Aquantia PHYs, where it was noticed that MII_LPA is always reported as zero, and it was thought that this is because of the PCS. Test case #1 above shows it is not, and the configured MII_ADVERTISE on system A ends up in the MII_LPA on system B, when in 2500base-x mode (IF_MODE=0). Test case #2, which uses "SGMII" auto-negotiation (IF_MODE=3) for the 3.125 Gbps lane, is actually a misconfiguration, but it is what led to the discovery. There is actually an old bug in the Lynx PCS driver - it expects all register values to contain their default out-of-reset values, as if the PCS were initialized by the Reset Configuration Word (RCW) settings. There are 2 cases in which this is problematic: - if the bootloader (or previous kexec-enabled Linux) wrote a different IF_MODE value - if dynamically changing the SerDes protocol from 1000base-x to 2500base-x, e.g. by replacing the optical SFP module. Specifically in test case #2, an accidental alignment between the bootloader configuring the PCS to expect SGMII in-band code words, and the AQR115 PHY actually transmitting SGMII in-band code words when operating in the "OCSGMII" system interface protocol, led to the PCS transmitting replicated symbols at 3.125 Gbps baud rate. This could only have happened if the PCS saw and reacted to the SGMII code words in the first place. Since test #2 is invalid from a protocol perspective (there seems to be no standard way of negotiating the data rate of 2500 Mbps with SGMII, and the lower data rates should remain 10/100/1000), in-band auto-negotiation for 2500base-x effectively means Clause 37 (i.e. IF_MODE=0). Make 2500base-x be treated like 1000base-x in this regard, by removing all prior limitations and calling lynx_pcs_config_giga(). This adds a new feature: LINK_INBAND_ENABLE and at the same time fixes the Lynx PCS's long standing problem that the registers (specifically IF_MODE, but others could be misconfigured as well) are not written by the driver to the known valid values for 2500base-x. Co-developed-by: Alexander Wilhelm Signed-off-by: Alexander Wilhelm Signed-off-by: Vladimir Oltean Link: https://patch.msgid.link/20251125103507.749654-1-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- drivers/net/pcs/pcs-lynx.c | 77 +++----------------------------------- 1 file changed, 5 insertions(+), 72 deletions(-) diff --git a/drivers/net/pcs/pcs-lynx.c b/drivers/net/pcs/pcs-lynx.c index 677f92883976..73e1364ad1ed 100644 --- a/drivers/net/pcs/pcs-lynx.c +++ b/drivers/net/pcs/pcs-lynx.c @@ -40,12 +40,12 @@ static unsigned int lynx_pcs_inband_caps(struct phylink_pcs *pcs, { switch (interface) { case PHY_INTERFACE_MODE_1000BASEX: + case PHY_INTERFACE_MODE_2500BASEX: case PHY_INTERFACE_MODE_SGMII: case PHY_INTERFACE_MODE_QSGMII: return LINK_INBAND_DISABLE | LINK_INBAND_ENABLE; case PHY_INTERFACE_MODE_10GBASER: - case PHY_INTERFACE_MODE_2500BASEX: return LINK_INBAND_DISABLE; case PHY_INTERFACE_MODE_USXGMII: @@ -80,27 +80,6 @@ static void lynx_pcs_get_state_usxgmii(struct mdio_device *pcs, phylink_decode_usxgmii_word(state, lpa); } -static void lynx_pcs_get_state_2500basex(struct mdio_device *pcs, - struct phylink_link_state *state) -{ - int bmsr; - - bmsr = mdiodev_read(pcs, MII_BMSR); - if (bmsr < 0) { - state->link = false; - return; - } - - state->link = !!(bmsr & BMSR_LSTATUS); - state->an_complete = !!(bmsr & BMSR_ANEGCOMPLETE); - if (!state->link) - return; - - state->speed = SPEED_2500; - state->pause |= MLO_PAUSE_TX | MLO_PAUSE_RX; - state->duplex = DUPLEX_FULL; -} - static void lynx_pcs_get_state(struct phylink_pcs *pcs, unsigned int neg_mode, struct phylink_link_state *state) { @@ -108,13 +87,11 @@ static void lynx_pcs_get_state(struct phylink_pcs *pcs, unsigned int neg_mode, switch (state->interface) { case PHY_INTERFACE_MODE_1000BASEX: + case PHY_INTERFACE_MODE_2500BASEX: case PHY_INTERFACE_MODE_SGMII: case PHY_INTERFACE_MODE_QSGMII: phylink_mii_c22_pcs_get_state(lynx->mdio, neg_mode, state); break; - case PHY_INTERFACE_MODE_2500BASEX: - lynx_pcs_get_state_2500basex(lynx->mdio, state); - break; case PHY_INTERFACE_MODE_USXGMII: case PHY_INTERFACE_MODE_10G_QXGMII: lynx_pcs_get_state_usxgmii(lynx->mdio, state); @@ -152,7 +129,8 @@ static int lynx_pcs_config_giga(struct mdio_device *pcs, mdiodev_write(pcs, LINK_TIMER_HI, link_timer >> 16); } - if (interface == PHY_INTERFACE_MODE_1000BASEX) { + if (interface == PHY_INTERFACE_MODE_1000BASEX || + interface == PHY_INTERFACE_MODE_2500BASEX) { if_mode = 0; } else { /* SGMII and QSGMII */ @@ -202,15 +180,9 @@ static int lynx_pcs_config(struct phylink_pcs *pcs, unsigned int neg_mode, case PHY_INTERFACE_MODE_1000BASEX: case PHY_INTERFACE_MODE_SGMII: case PHY_INTERFACE_MODE_QSGMII: + case PHY_INTERFACE_MODE_2500BASEX: return lynx_pcs_config_giga(lynx->mdio, ifmode, advertising, neg_mode); - case PHY_INTERFACE_MODE_2500BASEX: - if (neg_mode == PHYLINK_PCS_NEG_INBAND_ENABLED) { - dev_err(&lynx->mdio->dev, - "AN not supported on 3.125GHz SerDes lane\n"); - return -EOPNOTSUPP; - } - break; case PHY_INTERFACE_MODE_USXGMII: case PHY_INTERFACE_MODE_10G_QXGMII: return lynx_pcs_config_usxgmii(lynx->mdio, ifmode, advertising, @@ -271,42 +243,6 @@ static void lynx_pcs_link_up_sgmii(struct mdio_device *pcs, if_mode); } -/* 2500Base-X is SerDes protocol 7 on Felix and 6 on ENETC. It is a SerDes lane - * clocked at 3.125 GHz which encodes symbols with 8b/10b and does not have - * auto-negotiation of any link parameters. Electrically it is compatible with - * a single lane of XAUI. - * The hardware reference manual wants to call this mode SGMII, but it isn't - * really, since the fundamental features of SGMII: - * - Downgrading the link speed by duplicating symbols - * - Auto-negotiation - * are not there. - * The speed is configured at 1000 in the IF_MODE because the clock frequency - * is actually given by a PLL configured in the Reset Configuration Word (RCW). - * Since there is no difference between fixed speed SGMII w/o AN and 802.3z w/o - * AN, we call this PHY interface type 2500Base-X. In case a PHY negotiates a - * lower link speed on line side, the system-side interface remains fixed at - * 2500 Mbps and we do rate adaptation through pause frames. - */ -static void lynx_pcs_link_up_2500basex(struct mdio_device *pcs, - unsigned int neg_mode, - int speed, int duplex) -{ - u16 if_mode = 0; - - if (neg_mode == PHYLINK_PCS_NEG_INBAND_ENABLED) { - dev_err(&pcs->dev, "AN not supported for 2500BaseX\n"); - return; - } - - if (duplex == DUPLEX_HALF) - if_mode |= IF_MODE_HALF_DUPLEX; - if_mode |= IF_MODE_SPEED(SGMII_SPEED_2500); - - mdiodev_modify(pcs, IF_MODE, - IF_MODE_HALF_DUPLEX | IF_MODE_SPEED_MSK, - if_mode); -} - static void lynx_pcs_link_up(struct phylink_pcs *pcs, unsigned int neg_mode, phy_interface_t interface, int speed, int duplex) @@ -318,9 +254,6 @@ static void lynx_pcs_link_up(struct phylink_pcs *pcs, unsigned int neg_mode, case PHY_INTERFACE_MODE_QSGMII: lynx_pcs_link_up_sgmii(lynx->mdio, neg_mode, speed, duplex); break; - case PHY_INTERFACE_MODE_2500BASEX: - lynx_pcs_link_up_2500basex(lynx->mdio, neg_mode, speed, duplex); - break; case PHY_INTERFACE_MODE_USXGMII: case PHY_INTERFACE_MODE_10G_QXGMII: /* At the moment, only in-band AN is supported for USXGMII From 6ab578739a4c1f5ae3c5416d9c7339a2f50d83f1 Mon Sep 17 00:00:00 2001 From: Jeremy Kerr Date: Wed, 26 Nov 2025 16:48:54 +0800 Subject: [PATCH 1328/1381] net: mctp: test: move TX packetqueue from dst to dev To capture TX packets during a test, we are currently intercepting the dst->output with an implementation that adds the transmitted packet to a skb queue attached to the test-specific mock dst. The netdev itself is not involved in the test TX path. Instead, we can just use our test device to stash TXed packets for later inspection by the test. This means we can include the actual mctp_dst_output() implementation in the test (by setting dst.output in the test case), and don't need to be creating fake dst objects, or their corresponding skb queues. We need to ensure that the netdev is up to allow delivery to ndo_start_xmit, but the tests assume active devices at present anyway. Signed-off-by: Jeremy Kerr Link: https://patch.msgid.link/20251126-dev-mctp-test-tx-queue-v2-1-4e5bbd1d6c57@codeconstruct.com.au Signed-off-by: Jakub Kicinski --- net/mctp/test/route-test.c | 109 ++++++++++++++----------------------- net/mctp/test/utils.c | 45 ++++++--------- net/mctp/test/utils.h | 13 +---- 3 files changed, 59 insertions(+), 108 deletions(-) diff --git a/net/mctp/test/route-test.c b/net/mctp/test/route-test.c index be9149ac79dd..75ea96c10e49 100644 --- a/net/mctp/test/route-test.c +++ b/net/mctp/test/route-test.c @@ -20,7 +20,6 @@ struct mctp_frag_test { static void mctp_test_fragment(struct kunit *test) { const struct mctp_frag_test *params; - struct mctp_test_pktqueue tpq; int rc, i, n, mtu, msgsize; struct mctp_test_dev *dev; struct mctp_dst dst; @@ -43,13 +42,12 @@ static void mctp_test_fragment(struct kunit *test) dev = mctp_test_create_dev(); KUNIT_ASSERT_NOT_ERR_OR_NULL(test, dev); - mctp_test_dst_setup(test, &dst, dev, &tpq, mtu); + mctp_test_dst_setup(test, &dst, dev, mtu); rc = mctp_do_fragment_route(&dst, skb, mtu, MCTP_TAG_OWNER); KUNIT_EXPECT_FALSE(test, rc); - n = tpq.pkts.qlen; - + n = dev->pkts.qlen; KUNIT_EXPECT_EQ(test, n, params->n_frags); for (i = 0;; i++) { @@ -61,8 +59,7 @@ static void mctp_test_fragment(struct kunit *test) first = i == 0; last = i == (n - 1); - skb2 = skb_dequeue(&tpq.pkts); - + skb2 = skb_dequeue(&dev->pkts); if (!skb2) break; @@ -99,7 +96,7 @@ static void mctp_test_fragment(struct kunit *test) kfree_skb(skb2); } - mctp_test_dst_release(&dst, &tpq); + mctp_dst_release(&dst); mctp_test_destroy_dev(dev); } @@ -130,13 +127,11 @@ struct mctp_rx_input_test { static void mctp_test_rx_input(struct kunit *test) { const struct mctp_rx_input_test *params; - struct mctp_test_pktqueue tpq; struct mctp_test_route *rt; struct mctp_test_dev *dev; struct sk_buff *skb; params = test->param_value; - test->priv = &tpq; dev = mctp_test_create_dev(); KUNIT_ASSERT_NOT_ERR_OR_NULL(test, dev); @@ -147,13 +142,10 @@ static void mctp_test_rx_input(struct kunit *test) skb = mctp_test_create_skb(¶ms->hdr, 1); KUNIT_ASSERT_NOT_ERR_OR_NULL(test, skb); - mctp_test_pktqueue_init(&tpq); - mctp_pkttype_receive(skb, dev->ndev, &mctp_packet_type, NULL); - KUNIT_EXPECT_EQ(test, !!tpq.pkts.qlen, params->input); + KUNIT_EXPECT_EQ(test, !!dev->pkts.qlen, params->input); - skb_queue_purge(&tpq.pkts); mctp_test_route_destroy(test, rt); mctp_test_destroy_dev(dev); } @@ -182,7 +174,6 @@ KUNIT_ARRAY_PARAM(mctp_rx_input, mctp_rx_input_tests, static void __mctp_route_test_init(struct kunit *test, struct mctp_test_dev **devp, struct mctp_dst *dst, - struct mctp_test_pktqueue *tpq, struct socket **sockp, unsigned int netid) { @@ -196,7 +187,7 @@ static void __mctp_route_test_init(struct kunit *test, if (netid != MCTP_NET_ANY) WRITE_ONCE(dev->mdev->net, netid); - mctp_test_dst_setup(test, dst, dev, tpq, 68); + mctp_test_dst_setup(test, dst, dev, 68); rc = sock_create_kern(&init_net, AF_MCTP, SOCK_DGRAM, 0, &sock); KUNIT_ASSERT_EQ(test, rc, 0); @@ -215,11 +206,10 @@ static void __mctp_route_test_init(struct kunit *test, static void __mctp_route_test_fini(struct kunit *test, struct mctp_test_dev *dev, struct mctp_dst *dst, - struct mctp_test_pktqueue *tpq, struct socket *sock) { sock_release(sock); - mctp_test_dst_release(dst, tpq); + mctp_dst_release(dst); mctp_test_destroy_dev(dev); } @@ -232,7 +222,6 @@ struct mctp_route_input_sk_test { static void mctp_test_route_input_sk(struct kunit *test) { const struct mctp_route_input_sk_test *params; - struct mctp_test_pktqueue tpq; struct sk_buff *skb, *skb2; struct mctp_test_dev *dev; struct mctp_dst dst; @@ -241,13 +230,12 @@ static void mctp_test_route_input_sk(struct kunit *test) params = test->param_value; - __mctp_route_test_init(test, &dev, &dst, &tpq, &sock, MCTP_NET_ANY); + __mctp_route_test_init(test, &dev, &dst, &sock, MCTP_NET_ANY); skb = mctp_test_create_skb_data(¶ms->hdr, ¶ms->type); KUNIT_ASSERT_NOT_ERR_OR_NULL(test, skb); mctp_test_skb_set_dev(skb, dev); - mctp_test_pktqueue_init(&tpq); rc = mctp_dst_input(&dst, skb); @@ -266,7 +254,7 @@ static void mctp_test_route_input_sk(struct kunit *test) KUNIT_EXPECT_NULL(test, skb2); } - __mctp_route_test_fini(test, dev, &dst, &tpq, sock); + __mctp_route_test_fini(test, dev, &dst, sock); } #define FL_S (MCTP_HDR_FLAG_SOM) @@ -303,7 +291,6 @@ struct mctp_route_input_sk_reasm_test { static void mctp_test_route_input_sk_reasm(struct kunit *test) { const struct mctp_route_input_sk_reasm_test *params; - struct mctp_test_pktqueue tpq; struct sk_buff *skb, *skb2; struct mctp_test_dev *dev; struct mctp_dst dst; @@ -313,7 +300,7 @@ static void mctp_test_route_input_sk_reasm(struct kunit *test) params = test->param_value; - __mctp_route_test_init(test, &dev, &dst, &tpq, &sock, MCTP_NET_ANY); + __mctp_route_test_init(test, &dev, &dst, &sock, MCTP_NET_ANY); for (i = 0; i < params->n_hdrs; i++) { c = i; @@ -336,7 +323,7 @@ static void mctp_test_route_input_sk_reasm(struct kunit *test) KUNIT_EXPECT_NULL(test, skb2); } - __mctp_route_test_fini(test, dev, &dst, &tpq, sock); + __mctp_route_test_fini(test, dev, &dst, sock); } #define RX_FRAG(f, s) RX_HDR(1, 10, 8, FL_TO | (f) | ((s) << MCTP_HDR_SEQ_SHIFT)) @@ -438,7 +425,6 @@ struct mctp_route_input_sk_keys_test { static void mctp_test_route_input_sk_keys(struct kunit *test) { const struct mctp_route_input_sk_keys_test *params; - struct mctp_test_pktqueue tpq; struct sk_buff *skb, *skb2; struct mctp_test_dev *dev; struct mctp_sk_key *key; @@ -457,7 +443,7 @@ static void mctp_test_route_input_sk_keys(struct kunit *test) KUNIT_ASSERT_NOT_ERR_OR_NULL(test, dev); net = READ_ONCE(dev->mdev->net); - mctp_test_dst_setup(test, &dst, dev, &tpq, 68); + mctp_test_dst_setup(test, &dst, dev, 68); rc = sock_create_kern(&init_net, AF_MCTP, SOCK_DGRAM, 0, &sock); KUNIT_ASSERT_EQ(test, rc, 0); @@ -497,7 +483,7 @@ static void mctp_test_route_input_sk_keys(struct kunit *test) skb_free_datagram(sock->sk, skb2); mctp_key_unref(key); - __mctp_route_test_fini(test, dev, &dst, &tpq, sock); + __mctp_route_test_fini(test, dev, &dst, sock); } static const struct mctp_route_input_sk_keys_test mctp_route_input_sk_keys_tests[] = { @@ -572,7 +558,6 @@ KUNIT_ARRAY_PARAM(mctp_route_input_sk_keys, mctp_route_input_sk_keys_tests, struct test_net { unsigned int netid; struct mctp_test_dev *dev; - struct mctp_test_pktqueue tpq; struct mctp_dst dst; struct socket *sock; struct sk_buff *skb; @@ -591,20 +576,18 @@ mctp_test_route_input_multiple_nets_bind_init(struct kunit *test, t->msg.data = t->netid; - __mctp_route_test_init(test, &t->dev, &t->dst, &t->tpq, &t->sock, - t->netid); + __mctp_route_test_init(test, &t->dev, &t->dst, &t->sock, t->netid); t->skb = mctp_test_create_skb_data(&hdr, &t->msg); KUNIT_ASSERT_NOT_ERR_OR_NULL(test, t->skb); mctp_test_skb_set_dev(t->skb, t->dev); - mctp_test_pktqueue_init(&t->tpq); } static void mctp_test_route_input_multiple_nets_bind_fini(struct kunit *test, struct test_net *t) { - __mctp_route_test_fini(test, t->dev, &t->dst, &t->tpq, t->sock); + __mctp_route_test_fini(test, t->dev, &t->dst, t->sock); } /* Test that skbs from different nets (otherwise identical) get routed to their @@ -661,8 +644,7 @@ mctp_test_route_input_multiple_nets_key_init(struct kunit *test, t->msg.data = t->netid; - __mctp_route_test_init(test, &t->dev, &t->dst, &t->tpq, &t->sock, - t->netid); + __mctp_route_test_init(test, &t->dev, &t->dst, &t->sock, t->netid); msk = container_of(t->sock->sk, struct mctp_sock, sk); @@ -685,7 +667,7 @@ mctp_test_route_input_multiple_nets_key_fini(struct kunit *test, struct test_net *t) { mctp_key_unref(t->key); - __mctp_route_test_fini(test, t->dev, &t->dst, &t->tpq, t->sock); + __mctp_route_test_fini(test, t->dev, &t->dst, t->sock); } /* test that skbs from different nets (otherwise identical) get routed to their @@ -738,14 +720,13 @@ static void mctp_test_route_input_multiple_nets_key(struct kunit *test) static void mctp_test_route_input_sk_fail_single(struct kunit *test) { const struct mctp_hdr hdr = RX_HDR(1, 10, 8, FL_S | FL_E | FL_TO); - struct mctp_test_pktqueue tpq; struct mctp_test_dev *dev; struct mctp_dst dst; struct socket *sock; struct sk_buff *skb; int rc; - __mctp_route_test_init(test, &dev, &dst, &tpq, &sock, MCTP_NET_ANY); + __mctp_route_test_init(test, &dev, &dst, &sock, MCTP_NET_ANY); /* No rcvbuf space, so delivery should fail. __sock_set_rcvbuf will * clamp the minimum to SOCK_MIN_RCVBUF, so we open-code this. @@ -768,7 +749,7 @@ static void mctp_test_route_input_sk_fail_single(struct kunit *test) KUNIT_EXPECT_EQ(test, refcount_read(&skb->users), 1); kfree_skb(skb); - __mctp_route_test_fini(test, dev, &dst, &tpq, sock); + __mctp_route_test_fini(test, dev, &dst, sock); } /* Input route to socket, using a fragmented message, where sock delivery fails. @@ -776,7 +757,6 @@ static void mctp_test_route_input_sk_fail_single(struct kunit *test) static void mctp_test_route_input_sk_fail_frag(struct kunit *test) { const struct mctp_hdr hdrs[2] = { RX_FRAG(FL_S, 0), RX_FRAG(FL_E, 1) }; - struct mctp_test_pktqueue tpq; struct mctp_test_dev *dev; struct sk_buff *skbs[2]; struct mctp_dst dst; @@ -784,7 +764,7 @@ static void mctp_test_route_input_sk_fail_frag(struct kunit *test) unsigned int i; int rc; - __mctp_route_test_init(test, &dev, &dst, &tpq, &sock, MCTP_NET_ANY); + __mctp_route_test_init(test, &dev, &dst, &sock, MCTP_NET_ANY); lock_sock(sock->sk); WRITE_ONCE(sock->sk->sk_rcvbuf, 0); @@ -815,7 +795,7 @@ static void mctp_test_route_input_sk_fail_frag(struct kunit *test) KUNIT_EXPECT_EQ(test, refcount_read(&skbs[1]->users), 1); kfree_skb(skbs[1]); - __mctp_route_test_fini(test, dev, &dst, &tpq, sock); + __mctp_route_test_fini(test, dev, &dst, sock); } /* Input route to socket, using a fragmented message created from clones. @@ -833,7 +813,6 @@ static void mctp_test_route_input_cloned_frag(struct kunit *test) const size_t data_len = 3; /* arbitrary */ u8 compare[3 * ARRAY_SIZE(hdrs)]; u8 flat[3 * ARRAY_SIZE(hdrs)]; - struct mctp_test_pktqueue tpq; struct mctp_test_dev *dev; struct sk_buff *skb[5]; struct sk_buff *rx_skb; @@ -845,7 +824,7 @@ static void mctp_test_route_input_cloned_frag(struct kunit *test) total = data_len + sizeof(struct mctp_hdr); - __mctp_route_test_init(test, &dev, &dst, &tpq, &sock, MCTP_NET_ANY); + __mctp_route_test_init(test, &dev, &dst, &sock, MCTP_NET_ANY); /* Create a single skb initially with concatenated packets */ skb[0] = mctp_test_create_skb(&hdrs[0], 5 * total); @@ -922,7 +901,7 @@ static void mctp_test_route_input_cloned_frag(struct kunit *test) kfree_skb(skb[i]); } - __mctp_route_test_fini(test, dev, &dst, &tpq, sock); + __mctp_route_test_fini(test, dev, &dst, sock); } #if IS_ENABLED(CONFIG_MCTP_FLOWS) @@ -930,7 +909,6 @@ static void mctp_test_route_input_cloned_frag(struct kunit *test) static void mctp_test_flow_init(struct kunit *test, struct mctp_test_dev **devp, struct mctp_dst *dst, - struct mctp_test_pktqueue *tpq, struct socket **sock, struct sk_buff **skbp, unsigned int len) @@ -944,7 +922,7 @@ static void mctp_test_flow_init(struct kunit *test, * mctp_local_output, which will call dst->output on whatever * route we provide */ - __mctp_route_test_init(test, &dev, dst, tpq, sock, MCTP_NET_ANY); + __mctp_route_test_init(test, &dev, dst, sock, MCTP_NET_ANY); /* Assign a single EID. ->addrs is freed on mctp netdev release */ dev->mdev->addrs = kmalloc(sizeof(u8), GFP_KERNEL); @@ -965,16 +943,14 @@ static void mctp_test_flow_init(struct kunit *test, static void mctp_test_flow_fini(struct kunit *test, struct mctp_test_dev *dev, struct mctp_dst *dst, - struct mctp_test_pktqueue *tpq, struct socket *sock) { - __mctp_route_test_fini(test, dev, dst, tpq, sock); + __mctp_route_test_fini(test, dev, dst, sock); } /* test that an outgoing skb has the correct MCTP extension data set */ static void mctp_test_packet_flow(struct kunit *test) { - struct mctp_test_pktqueue tpq; struct sk_buff *skb, *skb2; struct mctp_test_dev *dev; struct mctp_dst dst; @@ -983,15 +959,15 @@ static void mctp_test_packet_flow(struct kunit *test) u8 dst_eid = 8; int n, rc; - mctp_test_flow_init(test, &dev, &dst, &tpq, &sock, &skb, 30); + mctp_test_flow_init(test, &dev, &dst, &sock, &skb, 30); rc = mctp_local_output(sock->sk, &dst, skb, dst_eid, MCTP_TAG_OWNER); KUNIT_ASSERT_EQ(test, rc, 0); - n = tpq.pkts.qlen; + n = dev->pkts.qlen; KUNIT_ASSERT_EQ(test, n, 1); - skb2 = skb_dequeue(&tpq.pkts); + skb2 = skb_dequeue(&dev->pkts); KUNIT_ASSERT_NOT_ERR_OR_NULL(test, skb2); flow = skb_ext_find(skb2, SKB_EXT_MCTP); @@ -1000,7 +976,7 @@ static void mctp_test_packet_flow(struct kunit *test) KUNIT_ASSERT_PTR_EQ(test, flow->key->sk, sock->sk); kfree_skb(skb2); - mctp_test_flow_fini(test, dev, &dst, &tpq, sock); + mctp_test_flow_fini(test, dev, &dst, sock); } /* test that outgoing skbs, after fragmentation, all have the correct MCTP @@ -1008,7 +984,6 @@ static void mctp_test_packet_flow(struct kunit *test) */ static void mctp_test_fragment_flow(struct kunit *test) { - struct mctp_test_pktqueue tpq; struct mctp_flow *flows[2]; struct sk_buff *tx_skbs[2]; struct mctp_test_dev *dev; @@ -1018,17 +993,17 @@ static void mctp_test_fragment_flow(struct kunit *test) u8 dst_eid = 8; int n, rc; - mctp_test_flow_init(test, &dev, &dst, &tpq, &sock, &skb, 100); + mctp_test_flow_init(test, &dev, &dst, &sock, &skb, 100); rc = mctp_local_output(sock->sk, &dst, skb, dst_eid, MCTP_TAG_OWNER); KUNIT_ASSERT_EQ(test, rc, 0); - n = tpq.pkts.qlen; + n = dev->pkts.qlen; KUNIT_ASSERT_EQ(test, n, 2); /* both resulting packets should have the same flow data */ - tx_skbs[0] = skb_dequeue(&tpq.pkts); - tx_skbs[1] = skb_dequeue(&tpq.pkts); + tx_skbs[0] = skb_dequeue(&dev->pkts); + tx_skbs[1] = skb_dequeue(&dev->pkts); KUNIT_ASSERT_NOT_ERR_OR_NULL(test, tx_skbs[0]); KUNIT_ASSERT_NOT_ERR_OR_NULL(test, tx_skbs[1]); @@ -1044,7 +1019,7 @@ static void mctp_test_fragment_flow(struct kunit *test) kfree_skb(tx_skbs[0]); kfree_skb(tx_skbs[1]); - mctp_test_flow_fini(test, dev, &dst, &tpq, sock); + mctp_test_flow_fini(test, dev, &dst, sock); } #else @@ -1063,7 +1038,6 @@ static void mctp_test_fragment_flow(struct kunit *test) static void mctp_test_route_output_key_create(struct kunit *test) { const u8 dst_eid = 26, src_eid = 15; - struct mctp_test_pktqueue tpq; const unsigned int netid = 50; struct mctp_test_dev *dev; struct mctp_sk_key *key; @@ -1080,7 +1054,7 @@ static void mctp_test_route_output_key_create(struct kunit *test) KUNIT_ASSERT_NOT_ERR_OR_NULL(test, dev); WRITE_ONCE(dev->mdev->net, netid); - mctp_test_dst_setup(test, &dst, dev, &tpq, 68); + mctp_test_dst_setup(test, &dst, dev, 68); rc = sock_create_kern(&init_net, AF_MCTP, SOCK_DGRAM, 0, &sock); KUNIT_ASSERT_EQ(test, rc, 0); @@ -1127,14 +1101,13 @@ static void mctp_test_route_output_key_create(struct kunit *test) KUNIT_EXPECT_FALSE(test, key->tag & MCTP_TAG_OWNER); sock_release(sock); - mctp_test_dst_release(&dst, &tpq); + mctp_dst_release(&dst); mctp_test_destroy_dev(dev); } static void mctp_test_route_extaddr_input(struct kunit *test) { static const unsigned char haddr[] = { 0xaa, 0x55 }; - struct mctp_test_pktqueue tpq; struct mctp_skb_cb *cb, *cb2; const unsigned int len = 40; struct mctp_test_dev *dev; @@ -1149,7 +1122,7 @@ static void mctp_test_route_extaddr_input(struct kunit *test) hdr.dest = 8; hdr.flags_seq_tag = FL_S | FL_E | FL_TO; - __mctp_route_test_init(test, &dev, &dst, &tpq, &sock, MCTP_NET_ANY); + __mctp_route_test_init(test, &dev, &dst, &sock, MCTP_NET_ANY); skb = mctp_test_create_skb(&hdr, len); KUNIT_ASSERT_NOT_ERR_OR_NULL(test, skb); @@ -1178,7 +1151,7 @@ static void mctp_test_route_extaddr_input(struct kunit *test) KUNIT_EXPECT_MEMEQ(test, cb2->haddr, haddr, sizeof(haddr)); kfree_skb(skb2); - __mctp_route_test_fini(test, dev, &dst, &tpq, sock); + __mctp_route_test_fini(test, dev, &dst, sock); } static void mctp_test_route_gw_lookup(struct kunit *test) @@ -1530,14 +1503,13 @@ static void mctp_test_bind_lookup(struct kunit *test) struct socket *socks[ARRAY_SIZE(lookup_binds)]; struct sk_buff *skb_pkt = NULL, *skb_sock = NULL; struct socket *sock_ty0, *sock_expect = NULL; - struct mctp_test_pktqueue tpq; struct mctp_test_dev *dev; struct mctp_dst dst; int rc; rx = test->param_value; - __mctp_route_test_init(test, &dev, &dst, &tpq, &sock_ty0, rx->net); + __mctp_route_test_init(test, &dev, &dst, &sock_ty0, rx->net); /* Create all binds */ for (size_t i = 0; i < ARRAY_SIZE(lookup_binds); i++) { mctp_test_bind_run(test, &lookup_binds[i], @@ -1557,7 +1529,6 @@ static void mctp_test_bind_lookup(struct kunit *test) skb_pkt = mctp_test_create_skb_data(&rx->hdr, &rx->ty); KUNIT_ASSERT_NOT_ERR_OR_NULL(test, skb_pkt); mctp_test_skb_set_dev(skb_pkt, dev); - mctp_test_pktqueue_init(&tpq); rc = mctp_dst_input(&dst, skb_pkt); if (rx->expect) { @@ -1591,7 +1562,7 @@ static void mctp_test_bind_lookup(struct kunit *test) for (size_t i = 0; i < ARRAY_SIZE(lookup_binds); i++) sock_release(socks[i]); - __mctp_route_test_fini(test, dev, &dst, &tpq, sock_ty0); + __mctp_route_test_fini(test, dev, &dst, sock_ty0); } static struct kunit_case mctp_test_cases[] = { diff --git a/net/mctp/test/utils.c b/net/mctp/test/utils.c index 35f6be814567..37f1ba62a2ab 100644 --- a/net/mctp/test/utils.c +++ b/net/mctp/test/utils.c @@ -13,7 +13,10 @@ static netdev_tx_t mctp_test_dev_tx(struct sk_buff *skb, struct net_device *ndev) { - kfree_skb(skb); + struct mctp_test_dev *dev = netdev_priv(ndev); + + skb_queue_tail(&dev->pkts, skb); + return NETDEV_TX_OK; } @@ -26,7 +29,7 @@ static void mctp_test_dev_setup(struct net_device *ndev) ndev->type = ARPHRD_MCTP; ndev->mtu = MCTP_DEV_TEST_MTU; ndev->hard_header_len = 0; - ndev->tx_queue_len = DEFAULT_TX_QUEUE_LEN; + ndev->tx_queue_len = 0; ndev->flags = IFF_NOARP; ndev->netdev_ops = &mctp_test_netdev_ops; ndev->needs_free_netdev = true; @@ -51,6 +54,7 @@ static struct mctp_test_dev *__mctp_test_create_dev(unsigned short lladdr_len, dev->ndev = ndev; ndev->addr_len = lladdr_len; dev_addr_set(ndev, lladdr); + skb_queue_head_init(&dev->pkts); rc = register_netdev(ndev); if (rc) { @@ -63,6 +67,11 @@ static struct mctp_test_dev *__mctp_test_create_dev(unsigned short lladdr_len, dev->mdev->net = mctp_default_net(dev_net(ndev)); rcu_read_unlock(); + /* bring the device up; we want to be able to TX immediately */ + rtnl_lock(); + dev_open(ndev, NULL); + rtnl_unlock(); + return dev; } @@ -79,26 +88,15 @@ struct mctp_test_dev *mctp_test_create_dev_lladdr(unsigned short lladdr_len, void mctp_test_destroy_dev(struct mctp_test_dev *dev) { + skb_queue_purge(&dev->pkts); mctp_dev_put(dev->mdev); unregister_netdev(dev->ndev); } -static const unsigned int test_pktqueue_magic = 0x5f713aef; - -void mctp_test_pktqueue_init(struct mctp_test_pktqueue *tpq) -{ - tpq->magic = test_pktqueue_magic; - skb_queue_head_init(&tpq->pkts); -} - static int mctp_test_dst_output(struct mctp_dst *dst, struct sk_buff *skb) { - struct kunit *test = current->kunit_test; - struct mctp_test_pktqueue *tpq = test->priv; - - KUNIT_ASSERT_EQ(test, tpq->magic, test_pktqueue_magic); - - skb_queue_tail(&tpq->pkts, skb); + skb->dev = dst->dev->dev; + dev_queue_xmit(skb); return 0; } @@ -169,11 +167,9 @@ struct mctp_test_route *mctp_test_create_route_gw(struct net *net, return rt; } -/* Convenience function for our test dst; release with mctp_test_dst_release() - */ +/* Convenience function for our test dst; release with mctp_dst_release() */ void mctp_test_dst_setup(struct kunit *test, struct mctp_dst *dst, - struct mctp_test_dev *dev, - struct mctp_test_pktqueue *tpq, unsigned int mtu) + struct mctp_test_dev *dev, unsigned int mtu) { KUNIT_EXPECT_NOT_ERR_OR_NULL(test, dev); @@ -183,15 +179,6 @@ void mctp_test_dst_setup(struct kunit *test, struct mctp_dst *dst, __mctp_dev_get(dst->dev->dev); dst->mtu = mtu; dst->output = mctp_test_dst_output; - mctp_test_pktqueue_init(tpq); - test->priv = tpq; -} - -void mctp_test_dst_release(struct mctp_dst *dst, - struct mctp_test_pktqueue *tpq) -{ - mctp_dst_release(dst); - skb_queue_purge(&tpq->pkts); } void mctp_test_route_destroy(struct kunit *test, struct mctp_test_route *rt) diff --git a/net/mctp/test/utils.h b/net/mctp/test/utils.h index 06bdb6cb5eff..4cc90c9da4d1 100644 --- a/net/mctp/test/utils.h +++ b/net/mctp/test/utils.h @@ -18,6 +18,8 @@ struct mctp_test_dev { unsigned short lladdr_len; unsigned char lladdr[MAX_ADDR_LEN]; + + struct sk_buff_head pkts; }; struct mctp_test_dev; @@ -26,11 +28,6 @@ struct mctp_test_route { struct mctp_route rt; }; -struct mctp_test_pktqueue { - unsigned int magic; - struct sk_buff_head pkts; -}; - struct mctp_test_bind_setup { mctp_eid_t bind_addr; int bind_net; @@ -59,11 +56,7 @@ struct mctp_test_route *mctp_test_create_route_gw(struct net *net, mctp_eid_t gw, unsigned int mtu); void mctp_test_dst_setup(struct kunit *test, struct mctp_dst *dst, - struct mctp_test_dev *dev, - struct mctp_test_pktqueue *tpq, unsigned int mtu); -void mctp_test_dst_release(struct mctp_dst *dst, - struct mctp_test_pktqueue *tpq); -void mctp_test_pktqueue_init(struct mctp_test_pktqueue *tpq); + struct mctp_test_dev *dev, unsigned int mtu); void mctp_test_route_destroy(struct kunit *test, struct mctp_test_route *rt); void mctp_test_skb_set_dev(struct sk_buff *skb, struct mctp_test_dev *dev); struct sk_buff *mctp_test_create_skb(const struct mctp_hdr *hdr, From 9bf66036d686b9a67000ba22bd94be13a4ea79ac Mon Sep 17 00:00:00 2001 From: Long Li Date: Wed, 26 Nov 2025 13:45:52 -0800 Subject: [PATCH 1329/1381] net: mana: Handle hardware recovery events when probing the device When MANA is being probed, it's possible that hardware is in recovery mode and the device may get GDMA_EQE_HWC_RESET_REQUEST over HWC in the middle of the probe. Detect such condition and go through the recovery service procedure. Signed-off-by: Long Li Reviewed-by: Haiyang Zhang Link: https://patch.msgid.link/1764193552-9712-1-git-send-email-longli@linux.microsoft.com Signed-off-by: Jakub Kicinski --- .../net/ethernet/microsoft/mana/gdma_main.c | 176 ++++++++++++++++-- include/net/mana/gdma.h | 12 +- 2 files changed, 170 insertions(+), 18 deletions(-) diff --git a/drivers/net/ethernet/microsoft/mana/gdma_main.c b/drivers/net/ethernet/microsoft/mana/gdma_main.c index 8fd70b34807a..efb4e412ec7e 100644 --- a/drivers/net/ethernet/microsoft/mana/gdma_main.c +++ b/drivers/net/ethernet/microsoft/mana/gdma_main.c @@ -15,6 +15,20 @@ struct dentry *mana_debugfs_root; +struct mana_dev_recovery { + struct list_head list; + struct pci_dev *pdev; + enum gdma_eqe_type type; +}; + +static struct mana_dev_recovery_work { + struct list_head dev_list; + struct delayed_work work; + + /* Lock for dev_list above */ + spinlock_t lock; +} mana_dev_recovery_work; + static u32 mana_gd_r32(struct gdma_context *g, u64 offset) { return readl(g->bar0_va + offset); @@ -387,6 +401,25 @@ EXPORT_SYMBOL_NS(mana_gd_ring_cq, "NET_MANA"); #define MANA_SERVICE_PERIOD 10 +static void mana_serv_rescan(struct pci_dev *pdev) +{ + struct pci_bus *parent; + + pci_lock_rescan_remove(); + + parent = pdev->bus; + if (!parent) { + dev_err(&pdev->dev, "MANA service: no parent bus\n"); + goto out; + } + + pci_stop_and_remove_bus_device(pdev); + pci_rescan_bus(parent); + +out: + pci_unlock_rescan_remove(); +} + static void mana_serv_fpga(struct pci_dev *pdev) { struct pci_bus *bus, *parent; @@ -419,9 +452,12 @@ static void mana_serv_reset(struct pci_dev *pdev) { struct gdma_context *gc = pci_get_drvdata(pdev); struct hw_channel_context *hwc; + int ret; if (!gc) { - dev_err(&pdev->dev, "MANA service: no GC\n"); + /* Perform PCI rescan on device if GC is not set up */ + dev_err(&pdev->dev, "MANA service: GC not setup, rescanning\n"); + mana_serv_rescan(pdev); return; } @@ -440,9 +476,18 @@ static void mana_serv_reset(struct pci_dev *pdev) msleep(MANA_SERVICE_PERIOD * 1000); - mana_gd_resume(pdev); + ret = mana_gd_resume(pdev); + if (ret == -ETIMEDOUT || ret == -EPROTO) { + /* Perform PCI rescan on device if we failed on HWC */ + dev_err(&pdev->dev, "MANA service: resume failed, rescanning\n"); + mana_serv_rescan(pdev); + goto out; + } - dev_info(&pdev->dev, "MANA reset cycle completed\n"); + if (ret) + dev_info(&pdev->dev, "MANA reset cycle failed err %d\n", ret); + else + dev_info(&pdev->dev, "MANA reset cycle completed\n"); out: gc->in_service = false; @@ -454,18 +499,9 @@ struct mana_serv_work { enum gdma_eqe_type type; }; -static void mana_serv_func(struct work_struct *w) +static void mana_do_service(enum gdma_eqe_type type, struct pci_dev *pdev) { - struct mana_serv_work *mns_wk; - struct pci_dev *pdev; - - mns_wk = container_of(w, struct mana_serv_work, serv_work); - pdev = mns_wk->pdev; - - if (!pdev) - goto out; - - switch (mns_wk->type) { + switch (type) { case GDMA_EQE_HWC_FPGA_RECONFIG: mana_serv_fpga(pdev); break; @@ -475,12 +511,48 @@ static void mana_serv_func(struct work_struct *w) break; default: - dev_err(&pdev->dev, "MANA service: unknown type %d\n", - mns_wk->type); + dev_err(&pdev->dev, "MANA service: unknown type %d\n", type); break; } +} + +static void mana_recovery_delayed_func(struct work_struct *w) +{ + struct mana_dev_recovery_work *work; + struct mana_dev_recovery *dev; + unsigned long flags; + + work = container_of(w, struct mana_dev_recovery_work, work.work); + + spin_lock_irqsave(&work->lock, flags); + + while (!list_empty(&work->dev_list)) { + dev = list_first_entry(&work->dev_list, + struct mana_dev_recovery, list); + list_del(&dev->list); + spin_unlock_irqrestore(&work->lock, flags); + + mana_do_service(dev->type, dev->pdev); + pci_dev_put(dev->pdev); + kfree(dev); + + spin_lock_irqsave(&work->lock, flags); + } + + spin_unlock_irqrestore(&work->lock, flags); +} + +static void mana_serv_func(struct work_struct *w) +{ + struct mana_serv_work *mns_wk; + struct pci_dev *pdev; + + mns_wk = container_of(w, struct mana_serv_work, serv_work); + pdev = mns_wk->pdev; + + if (pdev) + mana_do_service(mns_wk->type, pdev); -out: pci_dev_put(pdev); kfree(mns_wk); module_put(THIS_MODULE); @@ -541,6 +613,17 @@ static void mana_gd_process_eqe(struct gdma_queue *eq) case GDMA_EQE_HWC_RESET_REQUEST: dev_info(gc->dev, "Recv MANA service type:%d\n", type); + if (!test_and_set_bit(GC_PROBE_SUCCEEDED, &gc->flags)) { + /* + * Device is in probe and we received a hardware reset + * event, the probe function will detect that the flag + * has changed and perform service procedure. + */ + dev_info(gc->dev, + "Service is to be processed in probe\n"); + break; + } + if (gc->in_service) { dev_info(gc->dev, "Already in service\n"); break; @@ -1938,8 +2021,19 @@ static int mana_gd_probe(struct pci_dev *pdev, const struct pci_device_id *ent) if (err) goto cleanup_mana; + /* + * If a hardware reset event has occurred over HWC during probe, + * rollback and perform hardware reset procedure. + */ + if (test_and_set_bit(GC_PROBE_SUCCEEDED, &gc->flags)) { + err = -EPROTO; + goto cleanup_mana_rdma; + } + return 0; +cleanup_mana_rdma: + mana_rdma_remove(&gc->mana_ib); cleanup_mana: mana_remove(&gc->mana, false); cleanup_gd: @@ -1963,6 +2057,35 @@ static int mana_gd_probe(struct pci_dev *pdev, const struct pci_device_id *ent) disable_dev: pci_disable_device(pdev); dev_err(&pdev->dev, "gdma probe failed: err = %d\n", err); + + /* + * Hardware could be in recovery mode and the HWC returns TIMEDOUT or + * EPROTO from mana_gd_setup(), mana_probe() or mana_rdma_probe(), or + * we received a hardware reset event over HWC interrupt. In this case, + * perform the device recovery procedure after MANA_SERVICE_PERIOD + * seconds. + */ + if (err == -ETIMEDOUT || err == -EPROTO) { + struct mana_dev_recovery *dev; + unsigned long flags; + + dev_info(&pdev->dev, "Start MANA recovery mode\n"); + + dev = kzalloc(sizeof(*dev), GFP_KERNEL); + if (!dev) + return err; + + dev->pdev = pci_dev_get(pdev); + dev->type = GDMA_EQE_HWC_RESET_REQUEST; + + spin_lock_irqsave(&mana_dev_recovery_work.lock, flags); + list_add_tail(&dev->list, &mana_dev_recovery_work.dev_list); + spin_unlock_irqrestore(&mana_dev_recovery_work.lock, flags); + + schedule_delayed_work(&mana_dev_recovery_work.work, + secs_to_jiffies(MANA_SERVICE_PERIOD)); + } + return err; } @@ -2067,6 +2190,10 @@ static int __init mana_driver_init(void) { int err; + INIT_LIST_HEAD(&mana_dev_recovery_work.dev_list); + spin_lock_init(&mana_dev_recovery_work.lock); + INIT_DELAYED_WORK(&mana_dev_recovery_work.work, mana_recovery_delayed_func); + mana_debugfs_root = debugfs_create_dir("mana", NULL); err = pci_register_driver(&mana_driver); @@ -2080,6 +2207,21 @@ static int __init mana_driver_init(void) static void __exit mana_driver_exit(void) { + struct mana_dev_recovery *dev; + unsigned long flags; + + disable_delayed_work_sync(&mana_dev_recovery_work.work); + + spin_lock_irqsave(&mana_dev_recovery_work.lock, flags); + while (!list_empty(&mana_dev_recovery_work.dev_list)) { + dev = list_first_entry(&mana_dev_recovery_work.dev_list, + struct mana_dev_recovery, list); + list_del(&dev->list); + pci_dev_put(dev->pdev); + kfree(dev); + } + spin_unlock_irqrestore(&mana_dev_recovery_work.lock, flags); + pci_unregister_driver(&mana_driver); debugfs_remove(mana_debugfs_root); diff --git a/include/net/mana/gdma.h b/include/net/mana/gdma.h index a4cf307859f8..eaa27483f99b 100644 --- a/include/net/mana/gdma.h +++ b/include/net/mana/gdma.h @@ -382,6 +382,10 @@ struct gdma_irq_context { char name[MANA_IRQ_NAME_SZ]; }; +enum gdma_context_flags { + GC_PROBE_SUCCEEDED = 0, +}; + struct gdma_context { struct device *dev; struct dentry *mana_pci_debugfs; @@ -430,6 +434,8 @@ struct gdma_context { u64 pf_cap_flags1; struct workqueue_struct *service_wq; + + unsigned long flags; }; static inline bool mana_gd_is_mana(struct gdma_dev *gd) @@ -600,6 +606,9 @@ enum { /* Driver can send HWC periodically to query stats */ #define GDMA_DRV_CAP_FLAG_1_PERIODIC_STATS_QUERY BIT(21) +/* Driver can handle hardware recovery events during probe */ +#define GDMA_DRV_CAP_FLAG_1_PROBE_RECOVERY BIT(22) + #define GDMA_DRV_CAP_FLAGS1 \ (GDMA_DRV_CAP_FLAG_1_EQ_SHARING_MULTI_VPORT | \ GDMA_DRV_CAP_FLAG_1_NAPI_WKDONE_FIX | \ @@ -611,7 +620,8 @@ enum { GDMA_DRV_CAP_FLAG_1_HANDLE_RECONFIG_EQE | \ GDMA_DRV_CAP_FLAG_1_HW_VPORT_LINK_AWARE | \ GDMA_DRV_CAP_FLAG_1_PERIODIC_STATS_QUERY | \ - GDMA_DRV_CAP_FLAG_1_SKB_LINEARIZE) + GDMA_DRV_CAP_FLAG_1_SKB_LINEARIZE | \ + GDMA_DRV_CAP_FLAG_1_PROBE_RECOVERY) #define GDMA_DRV_CAP_FLAGS2 0 From aee0f01b4f118844cb126a5f27e91c2adf087c59 Mon Sep 17 00:00:00 2001 From: "Nikola Z. Ivanov" Date: Fri, 28 Nov 2025 09:25:44 +0200 Subject: [PATCH 1330/1381] team: Add matching error label for failed action Follow the "action" - "err_action" pairing of labels found across the source code of team net device. Currently in team_port_add the err_set_slave_promisc label is reused for exiting on error when setting allmulti level of the new slave. Signed-off-by: Nikola Z. Ivanov Reviewed-by: Simon Horman Reviewed-by: Jiri Pirko Link: https://patch.msgid.link/20251128072544.223645-1-zlatistiv@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/team/team_core.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/team/team_core.c b/drivers/net/team/team_core.c index 0a41d2b45d8c..4d5c9ae8f221 100644 --- a/drivers/net/team/team_core.c +++ b/drivers/net/team/team_core.c @@ -1231,7 +1231,7 @@ static int team_port_add(struct team *team, struct net_device *port_dev, if (err) { if (dev->flags & IFF_PROMISC) dev_set_promiscuity(port_dev, -1); - goto err_set_slave_promisc; + goto err_set_slave_allmulti; } } @@ -1258,6 +1258,7 @@ static int team_port_add(struct team *team, struct net_device *port_dev, return 0; err_set_dev_type: +err_set_slave_allmulti: err_set_slave_promisc: __team_option_inst_del_port(team, port); From c3b744fd20191963f1739498446b5efb812c99d8 Mon Sep 17 00:00:00 2001 From: Raju Rangoju Date: Sat, 29 Nov 2025 23:20:15 +0530 Subject: [PATCH 1331/1381] amd-xgbe: refactor the dma IRQ handling code path Refactor the DMA interrupt bottom-half handling to improve the readability, maintainability, without changing the intended behavior. Signed-off-by: Raju Rangoju Link: https://patch.msgid.link/20251129175016.3034185-2-Raju.Rangoju@amd.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/amd/xgbe/xgbe-drv.c | 62 ++++++++++++++++-------- 1 file changed, 41 insertions(+), 21 deletions(-) diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c index 0653e69f0ef7..aca1c57554d7 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c @@ -367,10 +367,11 @@ static irqreturn_t xgbe_ecc_isr(int irq, void *data) static void xgbe_isr_bh_work(struct work_struct *work) { struct xgbe_prv_data *pdata = from_work(pdata, work, dev_bh_work); - struct xgbe_hw_if *hw_if = &pdata->hw_if; - struct xgbe_channel *channel; - unsigned int dma_isr, dma_ch_isr; unsigned int mac_isr, mac_tssr, mac_mdioisr; + struct xgbe_hw_if *hw_if = &pdata->hw_if; + bool per_ch_irq, ti, ri, rbu, fbe; + unsigned int dma_isr, dma_ch_isr; + struct xgbe_channel *channel; unsigned int i; /* The DMA interrupt status register also reports MAC and MTL @@ -384,43 +385,62 @@ static void xgbe_isr_bh_work(struct work_struct *work) netif_dbg(pdata, intr, pdata->netdev, "DMA_ISR=%#010x\n", dma_isr); for (i = 0; i < pdata->channel_count; i++) { + bool schedule_napi = false; + struct napi_struct *napi; + if (!(dma_isr & (1 << i))) continue; channel = pdata->channel[i]; dma_ch_isr = XGMAC_DMA_IOREAD(channel, DMA_CH_SR); + + /* Precompute flags once */ + ti = !!XGMAC_GET_BITS(dma_ch_isr, DMA_CH_SR, TI); + ri = !!XGMAC_GET_BITS(dma_ch_isr, DMA_CH_SR, RI); + rbu = !!XGMAC_GET_BITS(dma_ch_isr, DMA_CH_SR, RBU); + fbe = !!XGMAC_GET_BITS(dma_ch_isr, DMA_CH_SR, FBE); + netif_dbg(pdata, intr, pdata->netdev, "DMA_CH%u_ISR=%#010x\n", i, dma_ch_isr); - /* The TI or RI interrupt bits may still be set even if using - * per channel DMA interrupts. Check to be sure those are not - * enabled before using the private data napi structure. - */ - if (!pdata->per_channel_irq && - (XGMAC_GET_BITS(dma_ch_isr, DMA_CH_SR, TI) || - XGMAC_GET_BITS(dma_ch_isr, DMA_CH_SR, RI))) { - if (napi_schedule_prep(&pdata->napi)) { - /* Disable Tx and Rx interrupts */ - xgbe_disable_rx_tx_ints(pdata); + per_ch_irq = pdata->per_channel_irq; - /* Turn on polling */ - __napi_schedule(&pdata->napi); - } + /* + * Decide which NAPI to use and whether to schedule: + * - When not using per-channel IRQs: schedule on global NAPI + * if TI or RI are set. + */ + if (!per_ch_irq && (ti || ri)) + schedule_napi = true; + + napi = per_ch_irq ? &channel->napi : &pdata->napi; + + if (schedule_napi && napi_schedule_prep(napi)) { + /* Disable interrupts appropriately before polling */ + xgbe_disable_rx_tx_ints(pdata); + + /* Turn on polling */ + __napi_schedule(napi); } else { - /* Don't clear Rx/Tx status if doing per channel DMA - * interrupts, these will be cleared by the ISR for - * per channel DMA interrupts. + /* + * Don't clear Rx/Tx status if doing per-channel DMA + * interrupts; those bits will be serviced/cleared by + * the per-channel ISR/NAPI. In non-per-channel mode + * when we're not scheduling NAPI here, ensure we don't + * accidentally clear TI/RI in HW: zero them in the + * local copy so that the eventual write-back does not + * clear TI/RI. */ XGMAC_SET_BITS(dma_ch_isr, DMA_CH_SR, TI, 0); XGMAC_SET_BITS(dma_ch_isr, DMA_CH_SR, RI, 0); } - if (XGMAC_GET_BITS(dma_ch_isr, DMA_CH_SR, RBU)) + if (rbu) pdata->ext_stats.rx_buffer_unavailable++; /* Restart the device on a Fatal Bus Error */ - if (XGMAC_GET_BITS(dma_ch_isr, DMA_CH_SR, FBE)) + if (fbe) schedule_work(&pdata->restart_work); /* Clear interrupt signals */ From ab96af7004c76de5fdca077b1e3732a7ec96f69b Mon Sep 17 00:00:00 2001 From: Raju Rangoju Date: Sat, 29 Nov 2025 23:20:16 +0530 Subject: [PATCH 1332/1381] amd-xgbe: schedule NAPI on Rx Buffer Unavailable (RBU) Under heavy load, Rx Buffer Unavailable (RBU) can occur if Rx processing is slower than network. When an RBU is signaled, try to schedule NAPI to help recover from such situation (including cases where an IRQ may be missed or such) Signed-off-by: Raju Rangoju Link: https://patch.msgid.link/20251129175016.3034185-3-Raju.Rangoju@amd.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/amd/xgbe/xgbe-drv.c | 19 +++++++++++++++---- 1 file changed, 15 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c index aca1c57554d7..3ddd896d6987 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c @@ -410,15 +410,29 @@ static void xgbe_isr_bh_work(struct work_struct *work) * Decide which NAPI to use and whether to schedule: * - When not using per-channel IRQs: schedule on global NAPI * if TI or RI are set. + * - RBU should also trigger NAPI (either per-channel or global) + * to allow refill. */ if (!per_ch_irq && (ti || ri)) schedule_napi = true; + if (rbu) { + schedule_napi = true; + pdata->ext_stats.rx_buffer_unavailable++; + } + napi = per_ch_irq ? &channel->napi : &pdata->napi; if (schedule_napi && napi_schedule_prep(napi)) { /* Disable interrupts appropriately before polling */ - xgbe_disable_rx_tx_ints(pdata); + if (per_ch_irq) { + if (pdata->channel_irq_mode) + xgbe_disable_rx_tx_int(pdata, channel); + else + disable_irq_nosync(channel->dma_irq); + } else { + xgbe_disable_rx_tx_ints(pdata); + } /* Turn on polling */ __napi_schedule(napi); @@ -436,9 +450,6 @@ static void xgbe_isr_bh_work(struct work_struct *work) XGMAC_SET_BITS(dma_ch_isr, DMA_CH_SR, RI, 0); } - if (rbu) - pdata->ext_stats.rx_buffer_unavailable++; - /* Restart the device on a Fatal Bus Error */ if (fbe) schedule_work(&pdata->restart_work); From 6f268e275c74dae0536e0b61982a8db25bcf4f16 Mon Sep 17 00:00:00 2001 From: Jonas Gorski Date: Fri, 28 Nov 2025 09:06:19 +0100 Subject: [PATCH 1333/1381] net: dsa: b53: fix VLAN_ID_IDX write size for BCM5325/65 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Since BCM5325 and BCM5365 only support up to 256 VLANs, the VLAN_ID_IDX register is only 8 bit wide, not 16 bit, so use an appropriate accessor. Fixes: c45655386e53 ("net: dsa: b53: add support for FDB operations on 5325/5365") Reviewed-by: Florian Fainelli Tested-by: Álvaro Fernández Rojas Signed-off-by: Jonas Gorski Link: https://patch.msgid.link/20251128080625.27181-2-jonas.gorski@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/b53/b53_common.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c index 72c85cd34a4e..7f24d2d8f938 100644 --- a/drivers/net/dsa/b53/b53_common.c +++ b/drivers/net/dsa/b53/b53_common.c @@ -1966,8 +1966,12 @@ static int b53_arl_op(struct b53_device *dev, int op, int port, /* Perform a read for the given MAC and VID */ b53_write48(dev, B53_ARLIO_PAGE, B53_MAC_ADDR_IDX, mac); - if (!is5325m(dev)) - b53_write16(dev, B53_ARLIO_PAGE, B53_VLAN_ID_IDX, vid); + if (!is5325m(dev)) { + if (is5325(dev) || is5365(dev)) + b53_write8(dev, B53_ARLIO_PAGE, B53_VLAN_ID_IDX, vid); + else + b53_write16(dev, B53_ARLIO_PAGE, B53_VLAN_ID_IDX, vid); + } /* Issue a read operation for this MAC */ ret = b53_arl_rw_op(dev, 1); From 9316012dd01952f75e37035360138ccc786ef727 Mon Sep 17 00:00:00 2001 From: Jonas Gorski Date: Fri, 28 Nov 2025 09:06:20 +0100 Subject: [PATCH 1334/1381] net: dsa: b53: fix extracting VID from entry for BCM5325/65 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit BCM5325/65's Entry register uses the highest three bits for VALID/STATIC/AGE, so shifting by 53 only will add these to b53_arl_entry::vid. So make sure to mask the vid value as well, to not get invalid VIDs. Fixes: c45655386e53 ("net: dsa: b53: add support for FDB operations on 5325/5365") Reviewed-by: Florian Fainelli Tested-by: Álvaro Fernández Rojas Signed-off-by: Jonas Gorski Link: https://patch.msgid.link/20251128080625.27181-3-jonas.gorski@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/b53/b53_priv.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/dsa/b53/b53_priv.h b/drivers/net/dsa/b53/b53_priv.h index 2bfd0e7c95c9..3b22e817fb28 100644 --- a/drivers/net/dsa/b53/b53_priv.h +++ b/drivers/net/dsa/b53/b53_priv.h @@ -350,7 +350,7 @@ static inline void b53_arl_to_entry_25(struct b53_arl_entry *ent, ent->is_age = !!(mac_vid & ARLTBL_AGE_25); ent->is_static = !!(mac_vid & ARLTBL_STATIC_25); u64_to_ether_addr(mac_vid, ent->mac); - ent->vid = mac_vid >> ARLTBL_VID_S_65; + ent->vid = (mac_vid >> ARLTBL_VID_S_65) & ARLTBL_VID_MASK_25; } static inline void b53_arl_to_entry_89(struct b53_arl_entry *ent, From 8e46aacea4264bcb8d4265fb07577afff58ae78d Mon Sep 17 00:00:00 2001 From: Jonas Gorski Date: Fri, 28 Nov 2025 09:06:21 +0100 Subject: [PATCH 1335/1381] net: dsa: b53: use same ARL search result offset for BCM5325/65 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit BCM5365's search result is at the same offset as BCM5325's search result, and they (mostly) share the same format, so switch BCM5365 to BCM5325's arl ops. Fixes: c45655386e53 ("net: dsa: b53: add support for FDB operations on 5325/5365") Reviewed-by: Florian Fainelli Tested-by: Álvaro Fernández Rojas Signed-off-by: Jonas Gorski Link: https://patch.msgid.link/20251128080625.27181-4-jonas.gorski@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/b53/b53_common.c | 18 +----------------- drivers/net/dsa/b53/b53_regs.h | 4 +--- 2 files changed, 2 insertions(+), 20 deletions(-) diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c index 7f24d2d8f938..91b0b4de475f 100644 --- a/drivers/net/dsa/b53/b53_common.c +++ b/drivers/net/dsa/b53/b53_common.c @@ -2125,16 +2125,6 @@ static void b53_arl_search_read_25(struct b53_device *dev, u8 idx, b53_arl_to_entry_25(ent, mac_vid); } -static void b53_arl_search_read_65(struct b53_device *dev, u8 idx, - struct b53_arl_entry *ent) -{ - u64 mac_vid; - - b53_read64(dev, B53_ARLIO_PAGE, B53_ARL_SRCH_RSTL_0_MACVID_65, - &mac_vid); - b53_arl_to_entry_25(ent, mac_vid); -} - static void b53_arl_search_read_89(struct b53_device *dev, u8 idx, struct b53_arl_entry *ent) { @@ -2746,12 +2736,6 @@ static const struct b53_arl_ops b53_arl_ops_25 = { .arl_search_read = b53_arl_search_read_25, }; -static const struct b53_arl_ops b53_arl_ops_65 = { - .arl_read_entry = b53_arl_read_entry_25, - .arl_write_entry = b53_arl_write_entry_25, - .arl_search_read = b53_arl_search_read_65, -}; - static const struct b53_arl_ops b53_arl_ops_89 = { .arl_read_entry = b53_arl_read_entry_89, .arl_write_entry = b53_arl_write_entry_89, @@ -2814,7 +2798,7 @@ static const struct b53_chip_data b53_switch_chips[] = { .arl_buckets = 1024, .imp_port = 5, .duplex_reg = B53_DUPLEX_STAT_FE, - .arl_ops = &b53_arl_ops_65, + .arl_ops = &b53_arl_ops_25, }, { .chip_id = BCM5389_DEVICE_ID, diff --git a/drivers/net/dsa/b53/b53_regs.h b/drivers/net/dsa/b53/b53_regs.h index 69ebbec932f6..505979102ed5 100644 --- a/drivers/net/dsa/b53/b53_regs.h +++ b/drivers/net/dsa/b53/b53_regs.h @@ -376,10 +376,8 @@ #define B53_ARL_SRCH_RSLT_MACVID_89 0x33 #define B53_ARL_SRCH_RSLT_MACVID_63XX 0x34 -/* Single register search result on 5325 */ +/* Single register search result on 5325/5365 */ #define B53_ARL_SRCH_RSTL_0_MACVID_25 0x24 -/* Single register search result on 5365 */ -#define B53_ARL_SRCH_RSTL_0_MACVID_65 0x30 /* ARL Search Data Result (32 bit) */ #define B53_ARL_SRCH_RSTL_0 0x68 From 85132103f700b1340fc17df8a981509d17bf4872 Mon Sep 17 00:00:00 2001 From: Jonas Gorski Date: Fri, 28 Nov 2025 09:06:22 +0100 Subject: [PATCH 1336/1381] net: dsa: b53: fix CPU port unicast ARL entries for BCM5325/65 On BCM5325 and BCM5365, unicast ARL entries use 8 as the value for the CPU port, so we need to translate it to/from 5 as used for the CPU port at most other places. Fixes: c45655386e53 ("net: dsa: b53: add support for FDB operations on 5325/5365") Signed-off-by: Jonas Gorski Reviewed-by: Florian Fainelli Link: https://patch.msgid.link/20251128080625.27181-5-jonas.gorski@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/b53/b53_priv.h | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-) diff --git a/drivers/net/dsa/b53/b53_priv.h b/drivers/net/dsa/b53/b53_priv.h index 3b22e817fb28..ae2c615c088e 100644 --- a/drivers/net/dsa/b53/b53_priv.h +++ b/drivers/net/dsa/b53/b53_priv.h @@ -344,12 +344,14 @@ static inline void b53_arl_to_entry_25(struct b53_arl_entry *ent, u64 mac_vid) { memset(ent, 0, sizeof(*ent)); - ent->port = (mac_vid >> ARLTBL_DATA_PORT_ID_S_25) & - ARLTBL_DATA_PORT_ID_MASK_25; ent->is_valid = !!(mac_vid & ARLTBL_VALID_25); ent->is_age = !!(mac_vid & ARLTBL_AGE_25); ent->is_static = !!(mac_vid & ARLTBL_STATIC_25); u64_to_ether_addr(mac_vid, ent->mac); + ent->port = (mac_vid >> ARLTBL_DATA_PORT_ID_S_25) & + ARLTBL_DATA_PORT_ID_MASK_25; + if (is_unicast_ether_addr(ent->mac) && ent->port == B53_CPU_PORT) + ent->port = B53_CPU_PORT_25; ent->vid = (mac_vid >> ARLTBL_VID_S_65) & ARLTBL_VID_MASK_25; } @@ -383,8 +385,11 @@ static inline void b53_arl_from_entry_25(u64 *mac_vid, const struct b53_arl_entry *ent) { *mac_vid = ether_addr_to_u64(ent->mac); - *mac_vid |= (u64)(ent->port & ARLTBL_DATA_PORT_ID_MASK_25) << - ARLTBL_DATA_PORT_ID_S_25; + if (is_unicast_ether_addr(ent->mac) && ent->port == B53_CPU_PORT_25) + *mac_vid |= (u64)B53_CPU_PORT << ARLTBL_DATA_PORT_ID_S_25; + else + *mac_vid |= (u64)(ent->port & ARLTBL_DATA_PORT_ID_MASK_25) << + ARLTBL_DATA_PORT_ID_S_25; *mac_vid |= (u64)(ent->vid & ARLTBL_VID_MASK_25) << ARLTBL_VID_S_65; if (ent->is_valid) From 3b08863469aa6028ac7c3120966f4e2f6051cf6b Mon Sep 17 00:00:00 2001 From: Jonas Gorski Date: Fri, 28 Nov 2025 09:06:23 +0100 Subject: [PATCH 1337/1381] net: dsa: b53: fix BCM5325/65 ARL entry multicast port masks We currently use the mask 0xf for writing and reading b53_entry::port, but this is only correct for unicast ARL entries. Multicast ARL entries use a bitmask, and 0xf is not enough space for ports > 3, which includes the CPU port. So extend the mask accordingly to also fit port 4 (bit 4) and MII (bit 5). According to the datasheet the multicast port mask is [60:48], making it 12 bit wide, but bits 60-55 are reserved anyway, and collide with the priority field at [60:59], so I am not sure if this is valid. Therefore leave it at the actual used range, [53:48]. The ARL search result register differs a bit, and there the mask is only [52:48], so only spanning the user ports. The MII port bit is contained in the Search Result Extension register. So create a separate search result parse function that properly handles this. Fixes: c45655386e53 ("net: dsa: b53: add support for FDB operations on 5325/5365") Reviewed-by: Florian Fainelli Signed-off-by: Jonas Gorski Link: https://patch.msgid.link/20251128080625.27181-6-jonas.gorski@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/b53/b53_common.c | 4 +++- drivers/net/dsa/b53/b53_priv.h | 25 +++++++++++++++++++++---- drivers/net/dsa/b53/b53_regs.h | 8 +++++++- 3 files changed, 31 insertions(+), 6 deletions(-) diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c index 91b0b4de475f..09a64812cd84 100644 --- a/drivers/net/dsa/b53/b53_common.c +++ b/drivers/net/dsa/b53/b53_common.c @@ -2119,10 +2119,12 @@ static void b53_arl_search_read_25(struct b53_device *dev, u8 idx, struct b53_arl_entry *ent) { u64 mac_vid; + u8 ext; + b53_read8(dev, B53_ARLIO_PAGE, B53_ARL_SRCH_RSLT_EXT_25, &ext); b53_read64(dev, B53_ARLIO_PAGE, B53_ARL_SRCH_RSTL_0_MACVID_25, &mac_vid); - b53_arl_to_entry_25(ent, mac_vid); + b53_arl_search_to_entry_25(ent, mac_vid, ext); } static void b53_arl_search_read_89(struct b53_device *dev, u8 idx, diff --git a/drivers/net/dsa/b53/b53_priv.h b/drivers/net/dsa/b53/b53_priv.h index ae2c615c088e..f4afbfcc345e 100644 --- a/drivers/net/dsa/b53/b53_priv.h +++ b/drivers/net/dsa/b53/b53_priv.h @@ -348,8 +348,8 @@ static inline void b53_arl_to_entry_25(struct b53_arl_entry *ent, ent->is_age = !!(mac_vid & ARLTBL_AGE_25); ent->is_static = !!(mac_vid & ARLTBL_STATIC_25); u64_to_ether_addr(mac_vid, ent->mac); - ent->port = (mac_vid >> ARLTBL_DATA_PORT_ID_S_25) & - ARLTBL_DATA_PORT_ID_MASK_25; + ent->port = (mac_vid & ARLTBL_DATA_PORT_ID_MASK_25) >> + ARLTBL_DATA_PORT_ID_S_25; if (is_unicast_ether_addr(ent->mac) && ent->port == B53_CPU_PORT) ent->port = B53_CPU_PORT_25; ent->vid = (mac_vid >> ARLTBL_VID_S_65) & ARLTBL_VID_MASK_25; @@ -388,8 +388,8 @@ static inline void b53_arl_from_entry_25(u64 *mac_vid, if (is_unicast_ether_addr(ent->mac) && ent->port == B53_CPU_PORT_25) *mac_vid |= (u64)B53_CPU_PORT << ARLTBL_DATA_PORT_ID_S_25; else - *mac_vid |= (u64)(ent->port & ARLTBL_DATA_PORT_ID_MASK_25) << - ARLTBL_DATA_PORT_ID_S_25; + *mac_vid |= ((u64)ent->port << ARLTBL_DATA_PORT_ID_S_25) & + ARLTBL_DATA_PORT_ID_MASK_25; *mac_vid |= (u64)(ent->vid & ARLTBL_VID_MASK_25) << ARLTBL_VID_S_65; if (ent->is_valid) @@ -414,6 +414,23 @@ static inline void b53_arl_from_entry_89(u64 *mac_vid, u32 *fwd_entry, *fwd_entry |= ARLTBL_AGE_89; } +static inline void b53_arl_search_to_entry_25(struct b53_arl_entry *ent, + u64 mac_vid, u8 ext) +{ + memset(ent, 0, sizeof(*ent)); + ent->is_valid = !!(mac_vid & ARLTBL_VALID_25); + ent->is_age = !!(mac_vid & ARLTBL_AGE_25); + ent->is_static = !!(mac_vid & ARLTBL_STATIC_25); + u64_to_ether_addr(mac_vid, ent->mac); + ent->vid = (mac_vid >> ARLTBL_VID_S_65) & ARLTBL_VID_MASK_25; + ent->port = (mac_vid & ARL_SRCH_RSLT_PORT_ID_MASK_25) >> + ARL_SRCH_RSLT_PORT_ID_S_25; + if (is_multicast_ether_addr(ent->mac) && (ext & ARL_SRCH_RSLT_EXT_MC_MII)) + ent->port |= BIT(B53_CPU_PORT_25); + else if (!is_multicast_ether_addr(ent->mac) && ent->port == B53_CPU_PORT) + ent->port = B53_CPU_PORT_25; +} + static inline void b53_arl_search_to_entry_63xx(struct b53_arl_entry *ent, u64 mac_vid, u16 fwd_entry) { diff --git a/drivers/net/dsa/b53/b53_regs.h b/drivers/net/dsa/b53/b53_regs.h index 505979102ed5..54b1016eb7eb 100644 --- a/drivers/net/dsa/b53/b53_regs.h +++ b/drivers/net/dsa/b53/b53_regs.h @@ -332,7 +332,7 @@ #define ARLTBL_VID_MASK_25 0xff #define ARLTBL_VID_MASK 0xfff #define ARLTBL_DATA_PORT_ID_S_25 48 -#define ARLTBL_DATA_PORT_ID_MASK_25 0xf +#define ARLTBL_DATA_PORT_ID_MASK_25 GENMASK_ULL(53, 48) #define ARLTBL_VID_S_65 53 #define ARLTBL_AGE_25 BIT_ULL(61) #define ARLTBL_STATIC_25 BIT_ULL(62) @@ -378,6 +378,12 @@ /* Single register search result on 5325/5365 */ #define B53_ARL_SRCH_RSTL_0_MACVID_25 0x24 +#define ARL_SRCH_RSLT_PORT_ID_S_25 48 +#define ARL_SRCH_RSLT_PORT_ID_MASK_25 GENMASK_ULL(52, 48) + +/* BCM5325/5365 Search result extend register (8 bit) */ +#define B53_ARL_SRCH_RSLT_EXT_25 0x2c +#define ARL_SRCH_RSLT_EXT_MC_MII BIT(2) /* ARL Search Data Result (32 bit) */ #define B53_ARL_SRCH_RSTL_0 0x68 From d39514e6a2d14f57830d649e2bf03b49612c2f73 Mon Sep 17 00:00:00 2001 From: Jonas Gorski Date: Fri, 28 Nov 2025 09:06:24 +0100 Subject: [PATCH 1338/1381] net: dsa: b53: fix BCM5325/65 ARL entry VIDs BCM5325/65's ARL entry registers do not contain the VID, only the search result register does. ARL entries have a separate VID entry register for the index into the VLAN table. So make ARL entry accessors use the VID entry registers instead, and move the VLAN ID field definition to the search register definition. Fixes: c45655386e53 ("net: dsa: b53: add support for FDB operations on 5325/5365") Signed-off-by: Jonas Gorski Reviewed-by: Florian Fainelli Link: https://patch.msgid.link/20251128080625.27181-7-jonas.gorski@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/b53/b53_common.c | 9 +++++++-- drivers/net/dsa/b53/b53_priv.h | 12 ++++++------ drivers/net/dsa/b53/b53_regs.h | 7 +++++-- 3 files changed, 18 insertions(+), 10 deletions(-) diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c index 09a64812cd84..ac995f36ed95 100644 --- a/drivers/net/dsa/b53/b53_common.c +++ b/drivers/net/dsa/b53/b53_common.c @@ -1853,19 +1853,24 @@ static int b53_arl_rw_op(struct b53_device *dev, unsigned int op) static void b53_arl_read_entry_25(struct b53_device *dev, struct b53_arl_entry *ent, u8 idx) { + u8 vid_entry; u64 mac_vid; + b53_read8(dev, B53_ARLIO_PAGE, B53_ARLTBL_VID_ENTRY_25(idx), + &vid_entry); b53_read64(dev, B53_ARLIO_PAGE, B53_ARLTBL_MAC_VID_ENTRY(idx), &mac_vid); - b53_arl_to_entry_25(ent, mac_vid); + b53_arl_to_entry_25(ent, mac_vid, vid_entry); } static void b53_arl_write_entry_25(struct b53_device *dev, const struct b53_arl_entry *ent, u8 idx) { + u8 vid_entry; u64 mac_vid; - b53_arl_from_entry_25(&mac_vid, ent); + b53_arl_from_entry_25(&mac_vid, &vid_entry, ent); + b53_write8(dev, B53_ARLIO_PAGE, B53_ARLTBL_VID_ENTRY_25(idx), vid_entry); b53_write64(dev, B53_ARLIO_PAGE, B53_ARLTBL_MAC_VID_ENTRY(idx), mac_vid); } diff --git a/drivers/net/dsa/b53/b53_priv.h b/drivers/net/dsa/b53/b53_priv.h index f4afbfcc345e..bd6849e5bb93 100644 --- a/drivers/net/dsa/b53/b53_priv.h +++ b/drivers/net/dsa/b53/b53_priv.h @@ -341,7 +341,7 @@ static inline void b53_arl_to_entry(struct b53_arl_entry *ent, } static inline void b53_arl_to_entry_25(struct b53_arl_entry *ent, - u64 mac_vid) + u64 mac_vid, u8 vid_entry) { memset(ent, 0, sizeof(*ent)); ent->is_valid = !!(mac_vid & ARLTBL_VALID_25); @@ -352,7 +352,7 @@ static inline void b53_arl_to_entry_25(struct b53_arl_entry *ent, ARLTBL_DATA_PORT_ID_S_25; if (is_unicast_ether_addr(ent->mac) && ent->port == B53_CPU_PORT) ent->port = B53_CPU_PORT_25; - ent->vid = (mac_vid >> ARLTBL_VID_S_65) & ARLTBL_VID_MASK_25; + ent->vid = vid_entry; } static inline void b53_arl_to_entry_89(struct b53_arl_entry *ent, @@ -381,7 +381,7 @@ static inline void b53_arl_from_entry(u64 *mac_vid, u32 *fwd_entry, *fwd_entry |= ARLTBL_AGE; } -static inline void b53_arl_from_entry_25(u64 *mac_vid, +static inline void b53_arl_from_entry_25(u64 *mac_vid, u8 *vid_entry, const struct b53_arl_entry *ent) { *mac_vid = ether_addr_to_u64(ent->mac); @@ -390,14 +390,13 @@ static inline void b53_arl_from_entry_25(u64 *mac_vid, else *mac_vid |= ((u64)ent->port << ARLTBL_DATA_PORT_ID_S_25) & ARLTBL_DATA_PORT_ID_MASK_25; - *mac_vid |= (u64)(ent->vid & ARLTBL_VID_MASK_25) << - ARLTBL_VID_S_65; if (ent->is_valid) *mac_vid |= ARLTBL_VALID_25; if (ent->is_static) *mac_vid |= ARLTBL_STATIC_25; if (ent->is_age) *mac_vid |= ARLTBL_AGE_25; + *vid_entry = ent->vid; } static inline void b53_arl_from_entry_89(u64 *mac_vid, u32 *fwd_entry, @@ -422,7 +421,8 @@ static inline void b53_arl_search_to_entry_25(struct b53_arl_entry *ent, ent->is_age = !!(mac_vid & ARLTBL_AGE_25); ent->is_static = !!(mac_vid & ARLTBL_STATIC_25); u64_to_ether_addr(mac_vid, ent->mac); - ent->vid = (mac_vid >> ARLTBL_VID_S_65) & ARLTBL_VID_MASK_25; + ent->vid = (mac_vid & ARL_SRCH_RSLT_VID_MASK_25) >> + ARL_SRCH_RSLT_VID_S_25; ent->port = (mac_vid & ARL_SRCH_RSLT_PORT_ID_MASK_25) >> ARL_SRCH_RSLT_PORT_ID_S_25; if (is_multicast_ether_addr(ent->mac) && (ext & ARL_SRCH_RSLT_EXT_MC_MII)) diff --git a/drivers/net/dsa/b53/b53_regs.h b/drivers/net/dsa/b53/b53_regs.h index 54b1016eb7eb..54a278db67c9 100644 --- a/drivers/net/dsa/b53/b53_regs.h +++ b/drivers/net/dsa/b53/b53_regs.h @@ -329,11 +329,9 @@ #define B53_ARLTBL_MAC_VID_ENTRY(n) ((0x10 * (n)) + 0x10) #define ARLTBL_MAC_MASK 0xffffffffffffULL #define ARLTBL_VID_S 48 -#define ARLTBL_VID_MASK_25 0xff #define ARLTBL_VID_MASK 0xfff #define ARLTBL_DATA_PORT_ID_S_25 48 #define ARLTBL_DATA_PORT_ID_MASK_25 GENMASK_ULL(53, 48) -#define ARLTBL_VID_S_65 53 #define ARLTBL_AGE_25 BIT_ULL(61) #define ARLTBL_STATIC_25 BIT_ULL(62) #define ARLTBL_VALID_25 BIT_ULL(63) @@ -353,6 +351,9 @@ #define ARLTBL_STATIC_89 BIT(14) #define ARLTBL_VALID_89 BIT(15) +/* BCM5325/BCM565 ARL Table VID Entry N Registers (8 bit) */ +#define B53_ARLTBL_VID_ENTRY_25(n) ((0x2 * (n)) + 0x30) + /* Maximum number of bin entries in the ARL for all switches */ #define B53_ARLTBL_MAX_BIN_ENTRIES 4 @@ -380,6 +381,8 @@ #define B53_ARL_SRCH_RSTL_0_MACVID_25 0x24 #define ARL_SRCH_RSLT_PORT_ID_S_25 48 #define ARL_SRCH_RSLT_PORT_ID_MASK_25 GENMASK_ULL(52, 48) +#define ARL_SRCH_RSLT_VID_S_25 53 +#define ARL_SRCH_RSLT_VID_MASK_25 GENMASK_ULL(60, 53) /* BCM5325/5365 Search result extend register (8 bit) */ #define B53_ARL_SRCH_RSLT_EXT_25 0x2c From 0b2b27058692d437b12d3f2a3bf0fa699af7376e Mon Sep 17 00:00:00 2001 From: Jonas Gorski Date: Fri, 28 Nov 2025 09:06:25 +0100 Subject: [PATCH 1339/1381] net: dsa: b53: allow VID 0 for BCM5325/65 Now that writing ARL entries works properly, we can actually use VID 0 as the default untagged VLAN for BCM5325 and BCM5365 as well. So use 0 as default PVID for all chips and do not reject VLAN 0 anymore, which we ignored since commit 45e9d59d3950 ("net: dsa: b53: do not allow to configure VLAN 0") anyway. Signed-off-by: Jonas Gorski Reviewed-by: Florian Fainelli Link: https://patch.msgid.link/20251128080625.27181-8-jonas.gorski@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/b53/b53_common.c | 8 +------- 1 file changed, 1 insertion(+), 7 deletions(-) diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c index ac995f36ed95..a1a177713d99 100644 --- a/drivers/net/dsa/b53/b53_common.c +++ b/drivers/net/dsa/b53/b53_common.c @@ -872,10 +872,7 @@ static void b53_enable_stp(struct b53_device *dev) static u16 b53_default_pvid(struct b53_device *dev) { - if (is5325(dev) || is5365(dev)) - return 1; - else - return 0; + return 0; } static bool b53_vlan_port_needs_forced_tagged(struct dsa_switch *ds, int port) @@ -1699,9 +1696,6 @@ static int b53_vlan_prepare(struct dsa_switch *ds, int port, { struct b53_device *dev = ds->priv; - if ((is5325(dev) || is5365(dev)) && vlan->vid == 0) - return -EOPNOTSUPP; - /* Port 7 on 7278 connects to the ASP's UniMAC which is not capable of * receiving VLAN tagged frames at all, we can still allow the port to * be configured for egress untagged. From 596c696301b1bcb9e482e22a4f7f582858eefbcc Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 1 Dec 2025 14:51:02 -0800 Subject: [PATCH 1340/1381] Revert "r8169: add DASH support for RTL8127AP" This reverts commit 17e9f841dd227a4dc976b22d000d5f669bc14493. Nathan reports error messages appearing in dmesg since commit under Fixes: [ 3.844125] r8169 0000:01:00.0 (unnamed net_device) (uninitialized): rtl_eriar_cond == 0 (loop: 100, delay: 100). [ 3.864844] r8169 0000:01:00.0 eth0: rtl_eriar_cond == 1 (loop: 100, delay: 100). [ 3.878825] r8169 0000:01:00.0 eth0: rtl_eriar_cond == 1 (loop: 100, delay: 100). [ 3.892632] r8169 0000:01:00.0 eth0: rtl_eriar_cond == 1 (loop: 100, delay: 100). [ 5.002551] r8169 0000:01:00.0 eth0: rtl_eriar_cond == 1 (loop: 100, delay: 100). [ 5.016286] r8169 0000:01:00.0 eth0: rtl_eriar_cond == 1 (loop: 100, delay: 100). [ 5.030027] r8169 0000:01:00.0 eth0: rtl_eriar_cond == 1 (loop: 100, delay: 100). Let's drop the bad change and revisit in the next release cycle. Repoted-by: Nathan Chancellor Link: https://lore.kernel.org/20251201224238.GA604467@ax162 Fixes: 17e9f841dd22 ("r8169: add DASH support for RTL8127AP") Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/realtek/r8169_main.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c index 630319604211..405e91eb3141 100644 --- a/drivers/net/ethernet/realtek/r8169_main.c +++ b/drivers/net/ethernet/realtek/r8169_main.c @@ -1512,7 +1512,6 @@ static enum rtl_dash_type rtl_get_dash_type(struct rtl8169_private *tp) case RTL_GIGA_MAC_VER_51 ... RTL_GIGA_MAC_VER_52: return RTL_DASH_EP; case RTL_GIGA_MAC_VER_66: - case RTL_GIGA_MAC_VER_80: return RTL_DASH_25_BP; default: return RTL_DASH_NONE; From e1de33c377b685298da406bb4838bd9814194f96 Mon Sep 17 00:00:00 2001 From: Gal Pressman Date: Sun, 30 Nov 2025 12:25:31 +0200 Subject: [PATCH 1341/1381] net/mlx5e: Use u64 instead of __u64 in ieee_setmaxrate Change upper_limit_mbps/gbps from __u64 to u64 to follow kernel coding conventions. Signed-off-by: Gal Pressman Reviewed-by: Nimrod Oren Signed-off-by: Tariq Toukan Link: https://patch.msgid.link/1764498334-1327918-2-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c b/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c index cf8f14ce4cd5..dd491fd8162b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c @@ -595,8 +595,8 @@ static int mlx5e_dcbnl_ieee_setmaxrate(struct net_device *netdev, struct mlx5_core_dev *mdev = priv->mdev; u8 max_bw_value[IEEE_8021QAZ_MAX_TCS]; u8 max_bw_unit[IEEE_8021QAZ_MAX_TCS]; - __u64 upper_limit_mbps; - __u64 upper_limit_gbps; + u64 upper_limit_mbps; + u64 upper_limit_gbps; int i; struct { int scale; From e1098bb02f2d9a85a127aecad6378e4f159acce5 Mon Sep 17 00:00:00 2001 From: Gal Pressman Date: Sun, 30 Nov 2025 12:25:32 +0200 Subject: [PATCH 1342/1381] net/mlx5e: Rename upper_limit_mbps to upper_limit_100mbps Clarify that the limit represents the threshold for using 100 Mbps units rather than a general Mbps limit. Signed-off-by: Gal Pressman Reviewed-by: Nimrod Oren Signed-off-by: Tariq Toukan Link: https://patch.msgid.link/1764498334-1327918-3-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c b/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c index dd491fd8162b..7127442f8003 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c @@ -595,7 +595,7 @@ static int mlx5e_dcbnl_ieee_setmaxrate(struct net_device *netdev, struct mlx5_core_dev *mdev = priv->mdev; u8 max_bw_value[IEEE_8021QAZ_MAX_TCS]; u8 max_bw_unit[IEEE_8021QAZ_MAX_TCS]; - u64 upper_limit_mbps; + u64 upper_limit_100mbps; u64 upper_limit_gbps; int i; struct { @@ -614,7 +614,7 @@ static int mlx5e_dcbnl_ieee_setmaxrate(struct net_device *netdev, memset(max_bw_value, 0, sizeof(max_bw_value)); memset(max_bw_unit, 0, sizeof(max_bw_unit)); - upper_limit_mbps = 255 * MLX5E_100MB; + upper_limit_100mbps = 255 * MLX5E_100MB; upper_limit_gbps = 255 * MLX5E_1GB; for (i = 0; i <= mlx5_max_tc(mdev); i++) { @@ -622,7 +622,7 @@ static int mlx5e_dcbnl_ieee_setmaxrate(struct net_device *netdev, max_bw_unit[i] = MLX5_BW_NO_LIMIT; continue; } - if (maxrate->tc_maxrate[i] <= upper_limit_mbps) { + if (maxrate->tc_maxrate[i] <= upper_limit_100mbps) { max_bw_value[i] = div_u64(maxrate->tc_maxrate[i], MLX5E_100MB); max_bw_value[i] = max_bw_value[i] ? max_bw_value[i] : 1; From 53f7a771285182be7bfba6d59ccdd0d47fc1a097 Mon Sep 17 00:00:00 2001 From: Gal Pressman Date: Sun, 30 Nov 2025 12:25:33 +0200 Subject: [PATCH 1343/1381] net/mlx5e: Use U8_MAX instead of hard coded magic number Replace hard coded 255 magic number with U8_MAX (the register field is 8 bits). Signed-off-by: Gal Pressman Reviewed-by: Nimrod Oren Signed-off-by: Tariq Toukan Link: https://patch.msgid.link/1764498334-1327918-4-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c b/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c index 7127442f8003..79f9d43b09b3 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c @@ -614,8 +614,8 @@ static int mlx5e_dcbnl_ieee_setmaxrate(struct net_device *netdev, memset(max_bw_value, 0, sizeof(max_bw_value)); memset(max_bw_unit, 0, sizeof(max_bw_unit)); - upper_limit_100mbps = 255 * MLX5E_100MB; - upper_limit_gbps = 255 * MLX5E_1GB; + upper_limit_100mbps = U8_MAX * MLX5E_100MB; + upper_limit_gbps = U8_MAX * MLX5E_1GB; for (i = 0; i <= mlx5_max_tc(mdev); i++) { if (!maxrate->tc_maxrate[i]) { From 87a5112bfc406a72c6bf1e8cdef9b0169dc1df6a Mon Sep 17 00:00:00 2001 From: Gal Pressman Date: Sun, 30 Nov 2025 12:25:34 +0200 Subject: [PATCH 1344/1381] net/mlx5e: Use standard unit definitions for bandwidth conversion MLX5E_100MB and MLX5E_1GB defines are confusing, MLX5E_100MB is not equal to 100 * MEGA, and MLX5E_1GB is not equal to one GIGA, as they hide the Kbps rate conversion required for ieee_maxrate. Replace hardcoded bandwidth conversion values with standard unit definitions from linux/units.h. Rename MLX5E_100MB/MLX5E_1GB to MLX5E_100MB_TO_KB/MLX5E_1GB_TO_KB to clarify these are conversion factors to Kbps, not absolute bandwidth values. Signed-off-by: Gal Pressman Reviewed-by: Nimrod Oren Signed-off-by: Tariq Toukan Link: https://patch.msgid.link/1764498334-1327918-5-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- .../net/ethernet/mellanox/mlx5/core/en_dcbnl.c | 17 +++++++++-------- 1 file changed, 9 insertions(+), 8 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c b/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c index 79f9d43b09b3..fddf7c207f8e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c @@ -31,14 +31,15 @@ */ #include #include +#include #include "en.h" #include "en/port.h" #include "en/port_buffer.h" #define MLX5E_MAX_BW_ALLOC 100 /* Max percentage of BW allocation */ -#define MLX5E_100MB (100000) -#define MLX5E_1GB (1000000) +#define MLX5E_100MB_TO_KB (100 * MEGA / KILO) +#define MLX5E_1GB_TO_KB (GIGA / KILO) #define MLX5E_CEE_STATE_UP 1 #define MLX5E_CEE_STATE_DOWN 0 @@ -572,10 +573,10 @@ static int mlx5e_dcbnl_ieee_getmaxrate(struct net_device *netdev, for (i = 0; i <= mlx5_max_tc(mdev); i++) { switch (max_bw_unit[i]) { case MLX5_100_MBPS_UNIT: - maxrate->tc_maxrate[i] = max_bw_value[i] * MLX5E_100MB; + maxrate->tc_maxrate[i] = max_bw_value[i] * MLX5E_100MB_TO_KB; break; case MLX5_GBPS_UNIT: - maxrate->tc_maxrate[i] = max_bw_value[i] * MLX5E_1GB; + maxrate->tc_maxrate[i] = max_bw_value[i] * MLX5E_1GB_TO_KB; break; case MLX5_BW_NO_LIMIT: break; @@ -614,8 +615,8 @@ static int mlx5e_dcbnl_ieee_setmaxrate(struct net_device *netdev, memset(max_bw_value, 0, sizeof(max_bw_value)); memset(max_bw_unit, 0, sizeof(max_bw_unit)); - upper_limit_100mbps = U8_MAX * MLX5E_100MB; - upper_limit_gbps = U8_MAX * MLX5E_1GB; + upper_limit_100mbps = U8_MAX * MLX5E_100MB_TO_KB; + upper_limit_gbps = U8_MAX * MLX5E_1GB_TO_KB; for (i = 0; i <= mlx5_max_tc(mdev); i++) { if (!maxrate->tc_maxrate[i]) { @@ -624,12 +625,12 @@ static int mlx5e_dcbnl_ieee_setmaxrate(struct net_device *netdev, } if (maxrate->tc_maxrate[i] <= upper_limit_100mbps) { max_bw_value[i] = div_u64(maxrate->tc_maxrate[i], - MLX5E_100MB); + MLX5E_100MB_TO_KB); max_bw_value[i] = max_bw_value[i] ? max_bw_value[i] : 1; max_bw_unit[i] = MLX5_100_MBPS_UNIT; } else if (maxrate->tc_maxrate[i] <= upper_limit_gbps) { max_bw_value[i] = div_u64(maxrate->tc_maxrate[i], - MLX5E_1GB); + MLX5E_1GB_TO_KB); max_bw_unit[i] = MLX5_GBPS_UNIT; } else { netdev_err(netdev, From 5e1bf5ae5e3ba3588b474669ba05f5d202003d84 Mon Sep 17 00:00:00 2001 From: Parthiban Veerasooran Date: Mon, 1 Dec 2025 08:53:45 +0530 Subject: [PATCH 1345/1381] net: phy: phy-c45: add SQI and SQI+ support for OATC14 10Base-T1S PHYs Add support for reading Signal Quality Indicator (SQI) and enhanced SQI+ from OATC14 10Base-T1S PHYs. - Introduce MDIO register definitions for DCQ_SQI and DCQ_SQIPLUS. - Add `genphy_c45_oatc14_get_sqi_max()` to return the maximum supported SQI/SQI+ level. - Add `genphy_c45_oatc14_get_sqi()` to return the current SQI or SQI+ value. - Update `include/linux/phy.h` to expose the new APIs. SQI+ capability is read from the Advanced Diagnostic Features Capability register (ADFCAP). If SQI+ is supported, the driver calculates the value from the MSBs of the DCQ_SQIPLUS register; otherwise, it falls back to basic SQI (0-7 levels). This enables ethtool to report the SQI value for OATC14 10Base-T1S PHYs. Open Alliance TC14 10BASE-T1S Advanced Diagnostic PHY Features Specification ref: https://opensig.org/wp-content/uploads/2025/06/OPEN_Alliance_10BASE-T1S_Advanced_PHY_features_for-automotive_Ethernet_V2.1b.pdf Signed-off-by: Parthiban Veerasooran Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20251201032346.6699-2-parthiban.veerasooran@microchip.com Signed-off-by: Jakub Kicinski --- drivers/net/phy/mdio-open-alliance.h | 13 +++ drivers/net/phy/phy-c45.c | 137 +++++++++++++++++++++++++++ include/linux/phy.h | 29 ++++++ 3 files changed, 179 insertions(+) diff --git a/drivers/net/phy/mdio-open-alliance.h b/drivers/net/phy/mdio-open-alliance.h index 6850a3f0b31e..449d0fb67093 100644 --- a/drivers/net/phy/mdio-open-alliance.h +++ b/drivers/net/phy/mdio-open-alliance.h @@ -56,6 +56,8 @@ /* Advanced Diagnostic Features Capability Register*/ #define MDIO_OATC14_ADFCAP 0xcc00 #define OATC14_ADFCAP_HDD_CAPABILITY GENMASK(10, 8) +#define OATC14_ADFCAP_SQIPLUS_CAPABILITY GENMASK(4, 1) +#define OATC14_ADFCAP_SQI_CAPABILITY BIT(0) /* Harness Defect Detection Register */ #define MDIO_OATC14_HDD 0xcc01 @@ -65,6 +67,17 @@ #define OATC14_HDD_VALID BIT(2) #define OATC14_HDD_SHORT_OPEN_STATUS GENMASK(1, 0) +/* Dynamic Channel Quality SQI Register */ +#define MDIO_OATC14_DCQ_SQI 0xcc03 +#define OATC14_DCQ_SQI_VALUE GENMASK(2, 0) + +/* Dynamic Channel Quality SQI Plus Register */ +#define MDIO_OATC14_DCQ_SQIPLUS 0xcc04 +#define OATC14_DCQ_SQIPLUS_VALUE GENMASK(7, 0) + +/* SQI is supported using 3 bits means 8 levels (0-7) */ +#define OATC14_SQI_MAX_LEVEL 7 + /* Bus Short/Open Status: * 0 0 - no fault; everything is ok. (Default) * 0 1 - detected as an open or missing termination(s) diff --git a/drivers/net/phy/phy-c45.c b/drivers/net/phy/phy-c45.c index f5e23b53994f..d48aa7231b37 100644 --- a/drivers/net/phy/phy-c45.c +++ b/drivers/net/phy/phy-c45.c @@ -1695,3 +1695,140 @@ int genphy_c45_oatc14_cable_test_start(struct phy_device *phydev) OATC14_HDD_START_CONTROL); } EXPORT_SYMBOL(genphy_c45_oatc14_cable_test_start); + +/** + * oatc14_update_sqi_capability - Read and update OATC14 10Base-T1S PHY SQI/SQI+ + * capability + * @phydev: Pointer to the PHY device structure + * + * This helper reads the OATC14 ADFCAP capability register to determine whether + * the PHY supports SQI or SQI+ reporting. + * + * SQI+ capability is detected first. The SQI+ field indicates the number of + * valid MSBs (3–8), corresponding to 8–256 SQI+ levels. When present, the + * function stores the number of SQI+ bits and computes the maximum SQI+ value + * as (2^bits - 1). + * + * If SQI+ is not supported, the function checks for basic SQI capability, + * which provides 0–7 SQI levels. + * + * On success, the capability information is stored in + * @phydev->oatc14_sqi_capability and marked as updated. + * + * Return: + * * 0 - capability successfully read and stored + * * -EOPNOTSUPP - SQI/SQI+ not supported by this PHY + * * Negative errno on read failure + */ +static int oatc14_update_sqi_capability(struct phy_device *phydev) +{ + u8 bits; + int ret; + + ret = phy_read_mmd(phydev, MDIO_MMD_VEND2, MDIO_OATC14_ADFCAP); + if (ret < 0) + return ret; + + /* Check for SQI+ capability + * 0 - SQI+ is not supported + * (3-8) bits for (8-256) SQI+ levels supported + */ + bits = FIELD_GET(OATC14_ADFCAP_SQIPLUS_CAPABILITY, ret); + if (bits) { + phydev->oatc14_sqi_capability.sqiplus_bits = bits; + /* Max sqi+ level supported: (2 ^ bits) - 1 */ + phydev->oatc14_sqi_capability.sqi_max = BIT(bits) - 1; + goto update_done; + } + + /* Check for SQI capability + * 0 - SQI is not supported + * 1 - SQI is supported (0-7 levels) + */ + if (ret & OATC14_ADFCAP_SQI_CAPABILITY) { + phydev->oatc14_sqi_capability.sqi_max = OATC14_SQI_MAX_LEVEL; + goto update_done; + } + + return -EOPNOTSUPP; + +update_done: + phydev->oatc14_sqi_capability.updated = true; + return 0; +} + +/** + * genphy_c45_oatc14_get_sqi_max - Get maximum supported SQI or SQI+ level of + * OATC14 10Base-T1S PHY + * @phydev: pointer to the PHY device structure + * + * This function returns the maximum supported Signal Quality Indicator (SQI) or + * SQI+ level. The SQI capability is updated on first invocation if it has not + * already been updated. + * + * Return: + * * Maximum SQI/SQI+ level supported + * * Negative errno on capability read failure + */ +int genphy_c45_oatc14_get_sqi_max(struct phy_device *phydev) +{ + int ret; + + if (!phydev->oatc14_sqi_capability.updated) { + ret = oatc14_update_sqi_capability(phydev); + if (ret) + return ret; + } + + return phydev->oatc14_sqi_capability.sqi_max; +} +EXPORT_SYMBOL(genphy_c45_oatc14_get_sqi_max); + +/** + * genphy_c45_oatc14_get_sqi - Get Signal Quality Indicator (SQI) from an OATC14 + * 10Base-T1S PHY + * @phydev: pointer to the PHY device structure + * + * This function reads the SQI+ or SQI value from an OATC14-compatible + * 10Base-T1S PHY. If SQI+ capability is supported, the function returns the + * extended SQI+ value; otherwise, it returns the basic SQI value. The SQI + * capability is updated on first invocation if it has not already been updated. + * + * Return: + * * SQI/SQI+ value on success + * * Negative errno on read failure + */ +int genphy_c45_oatc14_get_sqi(struct phy_device *phydev) +{ + u8 shift; + int ret; + + if (!phydev->oatc14_sqi_capability.updated) { + ret = oatc14_update_sqi_capability(phydev); + if (ret) + return ret; + } + + /* Calculate and return SQI+ value if supported */ + if (phydev->oatc14_sqi_capability.sqiplus_bits) { + ret = phy_read_mmd(phydev, MDIO_MMD_VEND2, + MDIO_OATC14_DCQ_SQIPLUS); + if (ret < 0) + return ret; + + /* SQI+ uses N MSBs out of 8 bits, left-aligned with padding 1's + * Calculate the right-shift needed to isolate the N bits. + */ + shift = 8 - phydev->oatc14_sqi_capability.sqiplus_bits; + + return (ret & OATC14_DCQ_SQIPLUS_VALUE) >> shift; + } + + /* Read and return SQI value if SQI+ capability is not supported */ + ret = phy_read_mmd(phydev, MDIO_MMD_VEND2, MDIO_OATC14_DCQ_SQI); + if (ret < 0) + return ret; + + return ret & OATC14_DCQ_SQI_VALUE; +} +EXPORT_SYMBOL(genphy_c45_oatc14_get_sqi); diff --git a/include/linux/phy.h b/include/linux/phy.h index 059a104223c4..fbbe028cc4b7 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -530,6 +530,30 @@ struct phy_c45_device_ids { struct macsec_context; struct macsec_ops; +/** + * struct phy_oatc14_sqi_capability - SQI capability information for OATC14 + * 10Base-T1S PHY + * @updated: Indicates whether the SQI capability fields have been updated. + * @sqi_max: Maximum supported Signal Quality Indicator (SQI) level reported by + * the PHY. + * @sqiplus_bits: Bits for SQI+ levels supported by the PHY. + * 0 - SQI+ is not supported + * 3 - SQI+ is supported, using 3 bits (8 levels) + * 4 - SQI+ is supported, using 4 bits (16 levels) + * 5 - SQI+ is supported, using 5 bits (32 levels) + * 6 - SQI+ is supported, using 6 bits (64 levels) + * 7 - SQI+ is supported, using 7 bits (128 levels) + * 8 - SQI+ is supported, using 8 bits (256 levels) + * + * This structure is used by the OATC14 10Base-T1S PHY driver to store the SQI + * and SQI+ capability information retrieved from the PHY. + */ +struct phy_oatc14_sqi_capability { + bool updated; + int sqi_max; + u8 sqiplus_bits; +}; + /** * struct phy_device - An instance of a PHY * @@ -626,6 +650,7 @@ struct macsec_ops; * @link_down_events: Number of times link was lost * @shared: Pointer to private data shared by phys in one package * @priv: Pointer to driver private data + * @oatc14_sqi_capability: SQI capability information for OATC14 10Base-T1S PHY * * interrupts currently only supports enabled or disabled, * but could be changed in the future to support enabling @@ -772,6 +797,8 @@ struct phy_device { /* MACsec management functions */ const struct macsec_ops *macsec_ops; #endif + + struct phy_oatc14_sqi_capability oatc14_sqi_capability; }; /* Generic phy_device::dev_flags */ @@ -2257,6 +2284,8 @@ int genphy_c45_an_config_eee_aneg(struct phy_device *phydev); int genphy_c45_oatc14_cable_test_start(struct phy_device *phydev); int genphy_c45_oatc14_cable_test_get_status(struct phy_device *phydev, bool *finished); +int genphy_c45_oatc14_get_sqi_max(struct phy_device *phydev); +int genphy_c45_oatc14_get_sqi(struct phy_device *phydev); /* The gen10g_* functions are the old Clause 45 stub */ int gen10g_config_aneg(struct phy_device *phydev); From 16416c8352875deca61d359e44a419fae0d8993c Mon Sep 17 00:00:00 2001 From: Parthiban Veerasooran Date: Mon, 1 Dec 2025 08:53:46 +0530 Subject: [PATCH 1346/1381] net: phy: microchip_t1s: add SQI support for LAN867x Rev.D0 PHYs Add support for Signal Quality Index (SQI) reporting in the Microchip T1S PHY driver for LAN867x Rev.D0 (OATC14-compliant) PHYs. This patch registers the following callbacks in the microchip_t1s driver structure: - .get_sqi - returns the current SQI value - .get_sqi_max - returns the maximum SQI value This enables ethtool to report the SQI value for LAN867x Rev.D0 PHYs. Signed-off-by: Parthiban Veerasooran Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20251201032346.6699-3-parthiban.veerasooran@microchip.com Signed-off-by: Jakub Kicinski --- drivers/net/phy/microchip_t1s.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/phy/microchip_t1s.c b/drivers/net/phy/microchip_t1s.c index 5a0a66778977..e601d56b2507 100644 --- a/drivers/net/phy/microchip_t1s.c +++ b/drivers/net/phy/microchip_t1s.c @@ -575,6 +575,8 @@ static struct phy_driver microchip_t1s_driver[] = { .get_plca_status = genphy_c45_plca_get_status, .cable_test_start = genphy_c45_oatc14_cable_test_start, .cable_test_get_status = genphy_c45_oatc14_cable_test_get_status, + .get_sqi = genphy_c45_oatc14_get_sqi, + .get_sqi_max = genphy_c45_oatc14_get_sqi_max, }, { PHY_ID_MATCH_EXACT(PHY_ID_LAN865X_REVB), From d973ac83ad0d0292c81d4bcbb4a335834f90aa5d Mon Sep 17 00:00:00 2001 From: David Yang Date: Mon, 1 Dec 2025 17:42:28 +0800 Subject: [PATCH 1347/1381] net: dsa: yt921x: Use *_ULL bitfield macros for VLAN_CTRL VLAN_CTRL should be treated as a 64-bit register. GENMASK and BIT macros use unsigned long as the underlying type, which will result in a build error on architectures where sizeof(long) == 4. Replace them with unsigned long long variants. Signed-off-by: David Yang Reviewed-by: Andrew Lunn Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20251201094232.3155105-2-mmyangfl@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/yt921x.h | 26 +++++++++++++------------- 1 file changed, 13 insertions(+), 13 deletions(-) diff --git a/drivers/net/dsa/yt921x.h b/drivers/net/dsa/yt921x.h index 44719d841d40..01ef623946fd 100644 --- a/drivers/net/dsa/yt921x.h +++ b/drivers/net/dsa/yt921x.h @@ -382,23 +382,23 @@ #define YT921X_FDB_HW_FLUSH_ON_LINKDOWN BIT(0) #define YT921X_VLANn_CTRL(vlan) (0x188000 + 8 * (vlan)) -#define YT921X_VLAN_CTRL_UNTAG_PORTS_M GENMASK(50, 40) +#define YT921X_VLAN_CTRL_UNTAG_PORTS_M GENMASK_ULL(50, 40) #define YT921X_VLAN_CTRL_UNTAG_PORTS(x) FIELD_PREP(YT921X_VLAN_CTRL_UNTAG_PORTS_M, (x)) -#define YT921X_VLAN_CTRL_UNTAG_PORTn(port) BIT((port) + 40) -#define YT921X_VLAN_CTRL_STP_ID_M GENMASK(39, 36) +#define YT921X_VLAN_CTRL_UNTAG_PORTn(port) BIT_ULL((port) + 40) +#define YT921X_VLAN_CTRL_STP_ID_M GENMASK_ULL(39, 36) #define YT921X_VLAN_CTRL_STP_ID(x) FIELD_PREP(YT921X_VLAN_CTRL_STP_ID_M, (x)) -#define YT921X_VLAN_CTRL_SVLAN_EN BIT(35) -#define YT921X_VLAN_CTRL_FID_M GENMASK(34, 23) +#define YT921X_VLAN_CTRL_SVLAN_EN BIT_ULL(35) +#define YT921X_VLAN_CTRL_FID_M GENMASK_ULL(34, 23) #define YT921X_VLAN_CTRL_FID(x) FIELD_PREP(YT921X_VLAN_CTRL_FID_M, (x)) -#define YT921X_VLAN_CTRL_LEARN_DIS BIT(22) -#define YT921X_VLAN_CTRL_INT_PRI_EN BIT(21) -#define YT921X_VLAN_CTRL_INT_PRI_M GENMASK(20, 18) -#define YT921X_VLAN_CTRL_PORTS_M GENMASK(17, 7) +#define YT921X_VLAN_CTRL_LEARN_DIS BIT_ULL(22) +#define YT921X_VLAN_CTRL_INT_PRI_EN BIT_ULL(21) +#define YT921X_VLAN_CTRL_INT_PRI_M GENMASK_ULL(20, 18) +#define YT921X_VLAN_CTRL_PORTS_M GENMASK_ULL(17, 7) #define YT921X_VLAN_CTRL_PORTS(x) FIELD_PREP(YT921X_VLAN_CTRL_PORTS_M, (x)) -#define YT921X_VLAN_CTRL_PORTn(port) BIT((port) + 7) -#define YT921X_VLAN_CTRL_BYPASS_1X_AC BIT(6) -#define YT921X_VLAN_CTRL_METER_EN BIT(5) -#define YT921X_VLAN_CTRL_METER_ID_M GENMASK(4, 0) +#define YT921X_VLAN_CTRL_PORTn(port) BIT_ULL((port) + 7) +#define YT921X_VLAN_CTRL_BYPASS_1X_AC BIT_ULL(6) +#define YT921X_VLAN_CTRL_METER_EN BIT_ULL(5) +#define YT921X_VLAN_CTRL_METER_ID_M GENMASK_ULL(4, 0) #define YT921X_TPID_IGRn(x) (0x210000 + 4 * (x)) /* [0, 3] */ #define YT921X_TPID_IGR_TPID_M GENMASK(15, 0) From 633b1d010ce838bd5e9dd71e738328e16e3df81d Mon Sep 17 00:00:00 2001 From: David Yang Date: Mon, 1 Dec 2025 17:42:29 +0800 Subject: [PATCH 1348/1381] net: dsa: yt921x: Add STP/MST support Support for STP/MST was deferred from the initial submission of the driver. Signed-off-by: David Yang Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20251201094232.3155105-3-mmyangfl@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/yt921x.c | 115 +++++++++++++++++++++++++++++++++++++++ drivers/net/dsa/yt921x.h | 9 +++ 2 files changed, 124 insertions(+) diff --git a/drivers/net/dsa/yt921x.c b/drivers/net/dsa/yt921x.c index d5fb17d2b6e0..12e1bd5a1061 100644 --- a/drivers/net/dsa/yt921x.c +++ b/drivers/net/dsa/yt921x.c @@ -2098,6 +2098,117 @@ yt921x_dsa_port_bridge_join(struct dsa_switch *ds, int port, return res; } +static int +yt921x_dsa_port_mst_state_set(struct dsa_switch *ds, int port, + const struct switchdev_mst_state *st) +{ + struct yt921x_priv *priv = to_yt921x_priv(ds); + u32 mask; + u32 ctrl; + int res; + + mask = YT921X_STP_PORTn_M(port); + switch (st->state) { + case BR_STATE_DISABLED: + ctrl = YT921X_STP_PORTn_DISABLED(port); + break; + case BR_STATE_LISTENING: + case BR_STATE_LEARNING: + ctrl = YT921X_STP_PORTn_LEARNING(port); + break; + case BR_STATE_FORWARDING: + default: + ctrl = YT921X_STP_PORTn_FORWARD(port); + break; + case BR_STATE_BLOCKING: + ctrl = YT921X_STP_PORTn_BLOCKING(port); + break; + } + + mutex_lock(&priv->reg_lock); + res = yt921x_reg_update_bits(priv, YT921X_STPn(st->msti), mask, ctrl); + mutex_unlock(&priv->reg_lock); + + return res; +} + +static int +yt921x_dsa_vlan_msti_set(struct dsa_switch *ds, struct dsa_bridge bridge, + const struct switchdev_vlan_msti *msti) +{ + struct yt921x_priv *priv = to_yt921x_priv(ds); + u64 mask64; + u64 ctrl64; + int res; + + if (!msti->vid) + return -EINVAL; + if (!msti->msti || msti->msti >= YT921X_MSTI_NUM) + return -EINVAL; + + mask64 = YT921X_VLAN_CTRL_STP_ID_M; + ctrl64 = YT921X_VLAN_CTRL_STP_ID(msti->msti); + + mutex_lock(&priv->reg_lock); + res = yt921x_reg64_update_bits(priv, YT921X_VLANn_CTRL(msti->vid), + mask64, ctrl64); + mutex_unlock(&priv->reg_lock); + + return res; +} + +static void +yt921x_dsa_port_stp_state_set(struct dsa_switch *ds, int port, u8 state) +{ + struct yt921x_priv *priv = to_yt921x_priv(ds); + struct dsa_port *dp = dsa_to_port(ds, port); + struct device *dev = to_device(priv); + bool learning; + u32 mask; + u32 ctrl; + int res; + + mask = YT921X_STP_PORTn_M(port); + learning = false; + switch (state) { + case BR_STATE_DISABLED: + ctrl = YT921X_STP_PORTn_DISABLED(port); + break; + case BR_STATE_LISTENING: + ctrl = YT921X_STP_PORTn_LEARNING(port); + break; + case BR_STATE_LEARNING: + ctrl = YT921X_STP_PORTn_LEARNING(port); + learning = dp->learning; + break; + case BR_STATE_FORWARDING: + default: + ctrl = YT921X_STP_PORTn_FORWARD(port); + learning = dp->learning; + break; + case BR_STATE_BLOCKING: + ctrl = YT921X_STP_PORTn_BLOCKING(port); + break; + } + + mutex_lock(&priv->reg_lock); + do { + res = yt921x_reg_update_bits(priv, YT921X_STPn(0), mask, ctrl); + if (res) + break; + + mask = YT921X_PORT_LEARN_DIS; + ctrl = !learning ? YT921X_PORT_LEARN_DIS : 0; + res = yt921x_reg_update_bits(priv, YT921X_PORTn_LEARN(port), + mask, ctrl); + } while (0); + mutex_unlock(&priv->reg_lock); + + if (res) + dev_err(dev, "Failed to %s port %d: %i\n", "set STP state for", + port, res); +} + static int yt921x_port_down(struct yt921x_priv *priv, int port) { u32 mask; @@ -2783,6 +2894,10 @@ static const struct dsa_switch_ops yt921x_dsa_switch_ops = { .port_bridge_flags = yt921x_dsa_port_bridge_flags, .port_bridge_leave = yt921x_dsa_port_bridge_leave, .port_bridge_join = yt921x_dsa_port_bridge_join, + /* mst */ + .port_mst_state_set = yt921x_dsa_port_mst_state_set, + .vlan_msti_set = yt921x_dsa_vlan_msti_set, + .port_stp_state_set = yt921x_dsa_port_stp_state_set, /* port */ .get_tag_protocol = yt921x_dsa_get_tag_protocol, .phylink_get_caps = yt921x_dsa_phylink_get_caps, diff --git a/drivers/net/dsa/yt921x.h b/drivers/net/dsa/yt921x.h index 01ef623946fd..61bb0ab3b09a 100644 --- a/drivers/net/dsa/yt921x.h +++ b/drivers/net/dsa/yt921x.h @@ -274,6 +274,13 @@ #define YT921X_VLAN_IGR_FILTER_PORTn(port) BIT(port) #define YT921X_PORTn_ISOLATION(port) (0x180294 + 4 * (port)) #define YT921X_PORT_ISOLATION_BLOCKn(port) BIT(port) +#define YT921X_STPn(n) (0x18038c + 4 * (n)) +#define YT921X_STP_PORTn_M(port) GENMASK(2 * (port) + 1, 2 * (port)) +#define YT921X_STP_PORTn(port, x) ((x) << (2 * (port))) +#define YT921X_STP_PORTn_DISABLED(port) YT921X_STP_PORTn(port, 0) +#define YT921X_STP_PORTn_LEARNING(port) YT921X_STP_PORTn(port, 1) +#define YT921X_STP_PORTn_BLOCKING(port) YT921X_STP_PORTn(port, 2) +#define YT921X_STP_PORTn_FORWARD(port) YT921X_STP_PORTn(port, 3) #define YT921X_PORTn_LEARN(port) (0x1803d0 + 4 * (port)) #define YT921X_PORT_LEARN_VID_LEARN_MULTI_EN BIT(22) #define YT921X_PORT_LEARN_VID_LEARN_MODE BIT(21) @@ -449,6 +456,8 @@ enum yt921x_fdb_entry_status { YT921X_FDB_ENTRY_STATUS_STATIC = 7, }; +#define YT921X_MSTI_NUM 16 + #define YT9215_MAJOR 0x9002 #define YT9218_MAJOR 0x9001 From 3b87e60d213150bc1f8475ec3c542f4c8cdd2e2c Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Sun, 30 Nov 2025 15:16:43 +0200 Subject: [PATCH 1349/1381] net: dsa: mt7530: unexport mt7530_switch_ops Commit cb675afcddbb ("net: dsa: mt7530: introduce separate MDIO driver") exported mt7530_switch_ops for use from mt7530-mmio.c. Later in the patch set, mt7530-mmio.c used mt7530_probe_common() to access the mt7530_switch_ops still from mt7530.c - see commit 110c18bfed41 ("net: dsa: mt7530: introduce driver for MT7988 built-in switch"). This proves that exporting mt7530_switch_ops was unnecessary, so unexport it back. Cc: DENG Qingfang Cc: Sean Wang Signed-off-by: Vladimir Oltean Reviewed-by: Daniel Golle Acked-by: Daniel Golle Acked-by: Chester A. Unal Link: https://patch.msgid.link/20251130131657.65080-2-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/mt7530.c | 3 +-- drivers/net/dsa/mt7530.h | 1 - 2 files changed, 1 insertion(+), 3 deletions(-) diff --git a/drivers/net/dsa/mt7530.c b/drivers/net/dsa/mt7530.c index 548b85befbf4..1acb57002014 100644 --- a/drivers/net/dsa/mt7530.c +++ b/drivers/net/dsa/mt7530.c @@ -3254,7 +3254,7 @@ static int mt7988_setup(struct dsa_switch *ds) return mt7531_setup_common(ds); } -const struct dsa_switch_ops mt7530_switch_ops = { +static const struct dsa_switch_ops mt7530_switch_ops = { .get_tag_protocol = mtk_get_tag_protocol, .setup = mt753x_setup, .preferred_default_local_cpu_port = mt753x_preferred_default_local_cpu_port, @@ -3291,7 +3291,6 @@ const struct dsa_switch_ops mt7530_switch_ops = { .conduit_state_change = mt753x_conduit_state_change, .port_setup_tc = mt753x_setup_tc, }; -EXPORT_SYMBOL_GPL(mt7530_switch_ops); static const struct phylink_mac_ops mt753x_phylink_mac_ops = { .mac_select_pcs = mt753x_phylink_mac_select_pcs, diff --git a/drivers/net/dsa/mt7530.h b/drivers/net/dsa/mt7530.h index 7e47cd9af256..3e0090bed298 100644 --- a/drivers/net/dsa/mt7530.h +++ b/drivers/net/dsa/mt7530.h @@ -939,7 +939,6 @@ static inline void INIT_MT7530_DUMMY_POLL(struct mt7530_dummy_poll *p, int mt7530_probe_common(struct mt7530_priv *priv); void mt7530_remove_common(struct mt7530_priv *priv); -extern const struct dsa_switch_ops mt7530_switch_ops; extern const struct mt753x_info mt753x_table[]; #endif /* __MT7530_H */ From a0244e76213980f3b9bb5d40b0b6705fcf24230d Mon Sep 17 00:00:00 2001 From: Xiaoliang Yang Date: Sun, 30 Nov 2025 15:16:44 +0200 Subject: [PATCH 1350/1381] net: hsr: create an API to get hsr port type MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Since the introduction of HSR_PT_INTERLINK in commit 5055cccfc2d1 ("net: hsr: Provide RedBox support (HSR-SAN)"), we see that different port types require different settings for hardware offload, which was not the case before when we only had HSR_PT_SLAVE_A and HSR_PT_SLAVE_B. But there is currently no way to know which port is which type, so create the hsr_get_port_type() API function and export it. When hsr_get_port_type() is called from the device driver, the port can must be found in the HSR port list. An important use case is for this function to work from offloading drivers' NETDEV_CHANGEUPPER handler, which is triggered by hsr_portdev_setup() -> netdev_master_upper_dev_link(). Therefore, we need to move the addition of the hsr_port to the HSR port list prior to calling hsr_portdev_setup(). This makes the error restoration path also more similar to hsr_del_port(), where kfree_rcu(port) is already used. Cc: Sebastian Andrzej Siewior Cc: Lukasz Majewski Signed-off-by: Xiaoliang Yang Signed-off-by: Vladimir Oltean Reviewed-by: Łukasz Majewski Link: https://patch.msgid.link/20251130131657.65080-3-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- include/linux/if_hsr.h | 9 +++++++++ net/hsr/hsr_device.c | 20 ++++++++++++++++++++ net/hsr/hsr_slave.c | 7 ++++--- 3 files changed, 33 insertions(+), 3 deletions(-) diff --git a/include/linux/if_hsr.h b/include/linux/if_hsr.h index d7941fd88032..f4cf2dd36d19 100644 --- a/include/linux/if_hsr.h +++ b/include/linux/if_hsr.h @@ -43,6 +43,8 @@ extern bool is_hsr_master(struct net_device *dev); extern int hsr_get_version(struct net_device *dev, enum hsr_version *ver); struct net_device *hsr_get_port_ndev(struct net_device *ndev, enum hsr_port_type pt); +int hsr_get_port_type(struct net_device *hsr_dev, struct net_device *dev, + enum hsr_port_type *type); #else static inline bool is_hsr_master(struct net_device *dev) { @@ -59,6 +61,13 @@ static inline struct net_device *hsr_get_port_ndev(struct net_device *ndev, { return ERR_PTR(-EINVAL); } + +static inline int hsr_get_port_type(struct net_device *hsr_dev, + struct net_device *dev, + enum hsr_port_type *type) +{ + return -EINVAL; +} #endif /* CONFIG_HSR */ #endif /*_LINUX_IF_HSR_H_*/ diff --git a/net/hsr/hsr_device.c b/net/hsr/hsr_device.c index 492cbc78ab75..d1bfc49b5f01 100644 --- a/net/hsr/hsr_device.c +++ b/net/hsr/hsr_device.c @@ -690,6 +690,26 @@ struct net_device *hsr_get_port_ndev(struct net_device *ndev, } EXPORT_SYMBOL(hsr_get_port_ndev); +int hsr_get_port_type(struct net_device *hsr_dev, struct net_device *dev, + enum hsr_port_type *type) +{ + struct hsr_priv *hsr = netdev_priv(hsr_dev); + struct hsr_port *port; + + rcu_read_lock(); + hsr_for_each_port(hsr, port) { + if (port->dev == dev) { + *type = port->type; + rcu_read_unlock(); + return 0; + } + } + rcu_read_unlock(); + + return -EINVAL; +} +EXPORT_SYMBOL(hsr_get_port_type); + /* Default multicast address for HSR Supervision frames */ static const unsigned char def_multicast_addr[ETH_ALEN] __aligned(2) = { 0x01, 0x15, 0x4e, 0x00, 0x01, 0x00 diff --git a/net/hsr/hsr_slave.c b/net/hsr/hsr_slave.c index 8177ac6c2d26..afe06ba00ea4 100644 --- a/net/hsr/hsr_slave.c +++ b/net/hsr/hsr_slave.c @@ -207,14 +207,14 @@ int hsr_add_port(struct hsr_priv *hsr, struct net_device *dev, port->type = type; ether_addr_copy(port->original_macaddress, dev->dev_addr); + list_add_tail_rcu(&port->port_list, &hsr->ports); + if (type != HSR_PT_MASTER) { res = hsr_portdev_setup(hsr, dev, port, extack); if (res) goto fail_dev_setup; } - list_add_tail_rcu(&port->port_list, &hsr->ports); - master = hsr_port_get_hsr(hsr, HSR_PT_MASTER); netdev_update_features(master->dev); dev_set_mtu(master->dev, hsr_get_max_mtu(hsr)); @@ -222,7 +222,8 @@ int hsr_add_port(struct hsr_priv *hsr, struct net_device *dev, return 0; fail_dev_setup: - kfree(port); + list_del_rcu(&port->port_list); + kfree_rcu(port, rcu); return res; } From bed59a86e91ad680331ecad4bf8717cbd4f5fc7f Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Sun, 30 Nov 2025 15:16:45 +0200 Subject: [PATCH 1351/1381] net: dsa: avoid calling ds->ops->port_hsr_leave() when unoffloaded This mirrors what we do in dsa_port_lag_leave() and dsa_port_bridge_leave(): when ds->ops->port_hsr_join() returns -EOPNOTSUPP, we fall back to a software implementation where dp->hsr_dev is NULL, and the unoffloaded port is no longer bothered with calls from the HSR layer. This helps, for example, with interlink ports which current DSA drivers don't know how to offload. We have to check only in port_hsr_join() for the port type, then in port_hsr_leave() we are sure we're dealing only with known port types. Signed-off-by: Vladimir Oltean Link: https://patch.msgid.link/20251130131657.65080-4-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- net/dsa/port.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/dsa/port.c b/net/dsa/port.c index 082573ae6864..ca3a7f52229b 100644 --- a/net/dsa/port.c +++ b/net/dsa/port.c @@ -1909,6 +1909,9 @@ void dsa_port_hsr_leave(struct dsa_port *dp, struct net_device *hsr) struct dsa_switch *ds = dp->ds; int err; + if (!dp->hsr_dev) + return; + dp->hsr_dev = NULL; if (ds->ops->port_hsr_leave) { From 30296ac7642652428396222e720718f2661e9425 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Sun, 30 Nov 2025 15:16:46 +0200 Subject: [PATCH 1352/1381] net: dsa: xrs700x: reject unsupported HSR configurations As discussed here: https://lore.kernel.org/netdev/20240620090210.drop6jwh7e5qw556@skbuf/ the fact is that the xrs700x.c driver only supports offloading HSR_PT_SLAVE_A and HSR_PT_SLAVE_B (which were the only port types at the time the offload was written, _for this driver_). Up until now, the API did not explicitly tell offloading drivers what port has what role. So xrs700x can get confused and think that it can support a configuration which it actually can't. There was a table in the attached link which gave an example: $ ip link add name hsr0 type hsr slave1 swp0 slave2 swp1 \ interlink swp2 supervision 45 version 1 HSR_PT_SLAVE_A HSR_PT_SLAVE_B HSR_PT_INTERLINK ---------------------------------------------------------------- user space 0 1 2 requests ---------------------------------------------------------------- XRS700X driver 1 2 - understands The switch would act as if the ring ports were swp1 and swp2. Now that we have explicit hsr_get_port_type() API, let's use that to work around the unintended semantical changes of the offloading API brought by the introduction of interlink ports in HSR. Fixes: 5055cccfc2d1 ("net: hsr: Provide RedBox support (HSR-SAN)") Cc: Lukasz Majewski Signed-off-by: Vladimir Oltean Reviewed-by: George McCollister Link: https://patch.msgid.link/20251130131657.65080-5-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/xrs700x/xrs700x.c | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/drivers/net/dsa/xrs700x/xrs700x.c b/drivers/net/dsa/xrs700x/xrs700x.c index 4dbcc49a9e52..0a05f4156ef4 100644 --- a/drivers/net/dsa/xrs700x/xrs700x.c +++ b/drivers/net/dsa/xrs700x/xrs700x.c @@ -566,6 +566,7 @@ static int xrs700x_hsr_join(struct dsa_switch *ds, int port, struct xrs700x *priv = ds->priv; struct net_device *user; int ret, i, hsr_pair[2]; + enum hsr_port_type type; enum hsr_version ver; bool fwd = false; @@ -589,6 +590,16 @@ static int xrs700x_hsr_join(struct dsa_switch *ds, int port, return -EOPNOTSUPP; } + ret = hsr_get_port_type(hsr, dsa_to_port(ds, port)->user, &type); + if (ret) + return ret; + + if (type != HSR_PT_SLAVE_A && type != HSR_PT_SLAVE_B) { + NL_SET_ERR_MSG_MOD(extack, + "Only HSR slave ports can be offloaded"); + return -EOPNOTSUPP; + } + dsa_hsr_foreach_port(dp, ds, hsr) { if (dp->index != port) { partner = dp; From 0e75bfe340bf05d1586eaf02942438573bda69e3 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Sun, 30 Nov 2025 15:16:47 +0200 Subject: [PATCH 1353/1381] net: dsa: add simple HSR offload helpers MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit It turns out that HSR offloads are so fine-grained that many DSA switches can do a small part even though they weren't specifically designed for the protocols supported by that driver (HSR and PRP). Specifically NETIF_F_HW_HSR_DUP - it is simple packet duplication on transmit, towards all (aka 2) ports members of the HSR device. For many DSA switches, we know how to duplicate a packet, even though we never typically use that feature. The transmit port mask from the tagging protocol can have multiple bits set, and the switch should send the packet once to every port with a bit set from that mask. Nonetheless, not all tagging protocols are like this, and sometimes the port is a single numeric value rather than a bit mask. For that reason, and also because switches can sometimes change tagging protocols for different ones, we need to make HSR offload helpers opt-in. For devices that can do nothing else HSR-specific, we introduce dsa_port_simple_hsr_join() and dsa_port_simple_hsr_leave(). These functions monitor when two user ports of the same switch are part of the same HSR device, and when that condition is true, they toggle the NETIF_F_HW_HSR_DUP feature flag of both net devices. Normally only dsa_port_simple_hsr_join() and dsa_port_simple_hsr_leave() are needed. The dsa_port_simple_hsr_validate() helper is just to see what kind of configuration could be offloadable using the generic helpers. This is used by switch drivers which are not currently using the right tagging protocol to offload this HSR ring, but could in principle offload it after changing the tagger. Suggested-by: David Yang Cc: "Alvin Šipraga" Cc: Chester A. Unal" Cc: "Clément Léger" Cc: Daniel Golle Cc: DENG Qingfang Cc: Florian Fainelli Cc: George McCollister Cc: Hauke Mehrtens Cc: Jonas Gorski Cc: Kurt Kanzenbach Cc: Linus Walleij Cc: Sean Wang Cc: UNGLinuxDriver@microchip.com Cc: Woojung Huh Signed-off-by: Vladimir Oltean Link: https://patch.msgid.link/20251130131657.65080-6-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- include/net/dsa.h | 9 +++++++ net/dsa/dsa.c | 65 +++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 74 insertions(+) diff --git a/include/net/dsa.h b/include/net/dsa.h index e40cdc12f7f3..cced1a866757 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -1322,6 +1322,15 @@ bool dsa_mdb_present_in_other_db(struct dsa_switch *ds, int port, const struct switchdev_obj_port_mdb *mdb, struct dsa_db db); +int dsa_port_simple_hsr_validate(struct dsa_switch *ds, int port, + struct net_device *hsr, + struct netlink_ext_ack *extack); +int dsa_port_simple_hsr_join(struct dsa_switch *ds, int port, + struct net_device *hsr, + struct netlink_ext_ack *extack); +int dsa_port_simple_hsr_leave(struct dsa_switch *ds, int port, + struct net_device *hsr); + /* Keep inline for faster access in hot path */ static inline bool netdev_uses_dsa(const struct net_device *dev) { diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c index 5b01a0e43ebe..a20efabe778f 100644 --- a/net/dsa/dsa.c +++ b/net/dsa/dsa.c @@ -9,6 +9,7 @@ #include #include +#include #include #include #include @@ -1766,6 +1767,70 @@ bool dsa_mdb_present_in_other_db(struct dsa_switch *ds, int port, } EXPORT_SYMBOL_GPL(dsa_mdb_present_in_other_db); +/* Helpers for switches without specific HSR offloads, but which can implement + * NETIF_F_HW_HSR_DUP because their tagger uses dsa_xmit_port_mask() + */ +int dsa_port_simple_hsr_validate(struct dsa_switch *ds, int port, + struct net_device *hsr, + struct netlink_ext_ack *extack) +{ + enum hsr_port_type type; + int err; + + err = hsr_get_port_type(hsr, dsa_to_port(ds, port)->user, &type); + if (err) + return err; + + if (type != HSR_PT_SLAVE_A && type != HSR_PT_SLAVE_B) { + NL_SET_ERR_MSG_MOD(extack, + "Only HSR slave ports can be offloaded"); + return -EOPNOTSUPP; + } + + return 0; +} +EXPORT_SYMBOL_GPL(dsa_port_simple_hsr_validate); + +int dsa_port_simple_hsr_join(struct dsa_switch *ds, int port, + struct net_device *hsr, + struct netlink_ext_ack *extack) +{ + struct dsa_port *dp = dsa_to_port(ds, port), *other_dp; + int err; + + err = dsa_port_simple_hsr_validate(ds, port, hsr, extack); + if (err) + return err; + + dsa_hsr_foreach_port(other_dp, ds, hsr) { + if (other_dp != dp) { + dp->user->features |= NETIF_F_HW_HSR_DUP; + other_dp->user->features |= NETIF_F_HW_HSR_DUP; + break; + } + } + + return 0; +} +EXPORT_SYMBOL_GPL(dsa_port_simple_hsr_join); + +int dsa_port_simple_hsr_leave(struct dsa_switch *ds, int port, + struct net_device *hsr) +{ + struct dsa_port *dp = dsa_to_port(ds, port), *other_dp; + + dsa_hsr_foreach_port(other_dp, ds, hsr) { + if (other_dp != dp) { + dp->user->features &= ~NETIF_F_HW_HSR_DUP; + other_dp->user->features &= ~NETIF_F_HW_HSR_DUP; + break; + } + } + + return 0; +} +EXPORT_SYMBOL_GPL(dsa_port_simple_hsr_leave); + static const struct dsa_stubs __dsa_stubs = { .conduit_hwtstamp_validate = __dsa_conduit_hwtstamp_validate, }; From 42e63b1373a32bd136dc51325684727f3b9a4986 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Sun, 30 Nov 2025 15:16:48 +0200 Subject: [PATCH 1354/1381] net: dsa: yt921x: use simple HSR offloading helpers Accelerate TX packet duplication with HSR rings. Cc: David Yang Signed-off-by: Vladimir Oltean Link: https://patch.msgid.link/20251130131657.65080-7-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/yt921x.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/net/dsa/yt921x.c b/drivers/net/dsa/yt921x.c index 12e1bd5a1061..1c511f5dc6ab 100644 --- a/drivers/net/dsa/yt921x.c +++ b/drivers/net/dsa/yt921x.c @@ -2874,6 +2874,9 @@ static const struct dsa_switch_ops yt921x_dsa_switch_ops = { /* mtu */ .port_change_mtu = yt921x_dsa_port_change_mtu, .port_max_mtu = yt921x_dsa_port_max_mtu, + /* hsr */ + .port_hsr_leave = dsa_port_simple_hsr_leave, + .port_hsr_join = dsa_port_simple_hsr_join, /* mirror */ .port_mirror_del = yt921x_dsa_port_mirror_del, .port_mirror_add = yt921x_dsa_port_mirror_add, From 4b65d445556d398b54fbeed8780ea9f07813f757 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Sun, 30 Nov 2025 15:16:49 +0200 Subject: [PATCH 1355/1381] net: dsa: ocelot: use simple HSR offload helpers Accelerate TX packet duplication with HSR rings. This is only possible with the NPI-based "ocelot" tagging protocol, not with "ocelot-8021q", because the latter does not use dsa_xmit_port_mask(). This has 2 implications: - Depending on tagging protocol, we should set (or not set) the offload feature flags. Switching tagging protocols is done with ports down, by design. Additional calls to dsa_port_simple_hsr_join() can be put in the ds->ops->change_tag_protocol() path, as I had originally tried, but this would not work: dsa_user_setup_tagger() would later clear the feature flag that we just set. So the additional call to dsa_port_simple_hsr_join() should sit in the ds->ops->port_enable() call. - When joining a HSR ring and we are currently using "ocelot-8021q", there are cases when we should return -EOPNOTSUPP (pessimistic) and cases when we shouldn't (optimistic). In the pessimistic case, it is a configuration that the port won't support even with the right tagging protocol. Distinguishing between these 2 cases matters because if we just return -EOPNOTSUPP regardless, we lose the dp->hsr_dev pointer and can no longer replay the offload later for the optimistic case, from felix_port_enable(). Signed-off-by: Vladimir Oltean Link: https://patch.msgid.link/20251130131657.65080-8-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/ocelot/felix.c | 70 +++++++++++++++++++++++++++++++++- 1 file changed, 69 insertions(+), 1 deletion(-) diff --git a/drivers/net/dsa/ocelot/felix.c b/drivers/net/dsa/ocelot/felix.c index 20ab558fde24..9e5ede932b42 100644 --- a/drivers/net/dsa/ocelot/felix.c +++ b/drivers/net/dsa/ocelot/felix.c @@ -1233,6 +1233,7 @@ static int felix_port_enable(struct dsa_switch *ds, int port, { struct dsa_port *dp = dsa_to_port(ds, port); struct ocelot *ocelot = ds->priv; + struct felix *felix = ocelot_to_felix(ocelot); if (!dsa_port_is_user(dp)) return 0; @@ -1246,7 +1247,25 @@ static int felix_port_enable(struct dsa_switch *ds, int port, } } - return 0; + if (!dp->hsr_dev || felix->tag_proto == DSA_TAG_PROTO_OCELOT_8021Q) + return 0; + + return dsa_port_simple_hsr_join(ds, port, dp->hsr_dev, NULL); +} + +static void felix_port_disable(struct dsa_switch *ds, int port) +{ + struct dsa_port *dp = dsa_to_port(ds, port); + struct ocelot *ocelot = ds->priv; + struct felix *felix = ocelot_to_felix(ocelot); + + if (!dsa_port_is_user(dp)) + return; + + if (!dp->hsr_dev || felix->tag_proto == DSA_TAG_PROTO_OCELOT_8021Q) + return; + + dsa_port_simple_hsr_leave(ds, port, dp->hsr_dev); } static void felix_port_qos_map_init(struct ocelot *ocelot, int port) @@ -2232,6 +2251,52 @@ static void felix_get_mm_stats(struct dsa_switch *ds, int port, ocelot_port_get_mm_stats(ocelot, port, stats); } +/* Depending on port type, we may be able to support the offload later (with + * the "ocelot"/"seville" tagging protocols), or never. + * If we return 0, the dp->hsr_dev reference is kept for later; if we return + * -EOPNOTSUPP, it is cleared (which helps to not bother + * dsa_port_simple_hsr_leave() with an offload that didn't pass validation). + */ +static int felix_port_hsr_join(struct dsa_switch *ds, int port, + struct net_device *hsr, + struct netlink_ext_ack *extack) +{ + struct ocelot *ocelot = ds->priv; + struct felix *felix = ocelot_to_felix(ocelot); + + if (felix->tag_proto == DSA_TAG_PROTO_OCELOT_8021Q) { + int err; + + err = dsa_port_simple_hsr_validate(ds, port, hsr, extack); + if (err) + return err; + + NL_SET_ERR_MSG_MOD(extack, + "Offloading not supported with \"ocelot-8021q\""); + return 0; + } + + if (!(dsa_to_port(ds, port)->user->flags & IFF_UP)) + return 0; + + return dsa_port_simple_hsr_join(ds, port, hsr, extack); +} + +static int felix_port_hsr_leave(struct dsa_switch *ds, int port, + struct net_device *hsr) +{ + struct ocelot *ocelot = ds->priv; + struct felix *felix = ocelot_to_felix(ocelot); + + if (felix->tag_proto == DSA_TAG_PROTO_OCELOT_8021Q) + return 0; + + if (!(dsa_to_port(ds, port)->user->flags & IFF_UP)) + return 0; + + return dsa_port_simple_hsr_leave(ds, port, hsr); +} + static const struct phylink_mac_ops felix_phylink_mac_ops = { .mac_select_pcs = felix_phylink_mac_select_pcs, .mac_config = felix_phylink_mac_config, @@ -2262,6 +2327,7 @@ static const struct dsa_switch_ops felix_switch_ops = { .get_ts_info = felix_get_ts_info, .phylink_get_caps = felix_phylink_get_caps, .port_enable = felix_port_enable, + .port_disable = felix_port_disable, .port_fast_age = felix_port_fast_age, .port_fdb_dump = felix_fdb_dump, .port_fdb_add = felix_fdb_add, @@ -2318,6 +2384,8 @@ static const struct dsa_switch_ops felix_switch_ops = { .port_del_dscp_prio = felix_port_del_dscp_prio, .port_set_host_flood = felix_port_set_host_flood, .port_change_conduit = felix_port_change_conduit, + .port_hsr_join = felix_port_hsr_join, + .port_hsr_leave = felix_port_hsr_leave, }; int felix_register_switch(struct device *dev, resource_size_t switch_base, From 6db31942e3470607be155d2c641c7a15da3c826c Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Sun, 30 Nov 2025 15:16:50 +0200 Subject: [PATCH 1356/1381] net: dsa: realtek: use simple HSR offload helpers MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit All known Realtek protocols: "rtl4a", "rtl8_4" and "rtl8_4t" use dsa_xmit_port_mask(), so they are compatible with accelerating TX packet duplication for HSR rings. Enable that feature by setting the port_hsr_join() and port_hsr_leave() operations to the simple helpers provided by DSA. Cc: "Alvin Šipraga" Signed-off-by: Vladimir Oltean Reviewed-by: Linus Walleij Link: https://patch.msgid.link/20251130131657.65080-9-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/realtek/rtl8365mb.c | 2 ++ drivers/net/dsa/realtek/rtl8366rb.c | 2 ++ 2 files changed, 4 insertions(+) diff --git a/drivers/net/dsa/realtek/rtl8365mb.c b/drivers/net/dsa/realtek/rtl8365mb.c index 964a56ee16cc..c575e164368c 100644 --- a/drivers/net/dsa/realtek/rtl8365mb.c +++ b/drivers/net/dsa/realtek/rtl8365mb.c @@ -2134,6 +2134,8 @@ static const struct dsa_switch_ops rtl8365mb_switch_ops = { .get_stats64 = rtl8365mb_get_stats64, .port_change_mtu = rtl8365mb_port_change_mtu, .port_max_mtu = rtl8365mb_port_max_mtu, + .port_hsr_join = dsa_port_simple_hsr_join, + .port_hsr_leave = dsa_port_simple_hsr_leave, }; static const struct realtek_ops rtl8365mb_ops = { diff --git a/drivers/net/dsa/realtek/rtl8366rb.c b/drivers/net/dsa/realtek/rtl8366rb.c index 8bdb52b5fdcb..d96ae72b0a5c 100644 --- a/drivers/net/dsa/realtek/rtl8366rb.c +++ b/drivers/net/dsa/realtek/rtl8366rb.c @@ -1815,6 +1815,8 @@ static const struct dsa_switch_ops rtl8366rb_switch_ops = { .port_fast_age = rtl8366rb_port_fast_age, .port_change_mtu = rtl8366rb_change_mtu, .port_max_mtu = rtl8366rb_max_mtu, + .port_hsr_join = dsa_port_simple_hsr_join, + .port_hsr_leave = dsa_port_simple_hsr_leave, }; static const struct realtek_ops rtl8366rb_ops = { From b6ad21ef286a00de103ef6d94292e4da649d6453 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Sun, 30 Nov 2025 15:16:51 +0200 Subject: [PATCH 1357/1381] net: dsa: lantiq_gswip: use simple HSR offload helpers Both the "gswip" and "gsw1xx" protocols use dsa_xmit_port_mask(), so they are compatible with accelerating TX packet duplication for HSR rings. Enable that feature by setting the port_hsr_join() and port_hsr_leave() operations to the simple helpers provided by DSA. Cc: Hauke Mehrtens Cc: Daniel Golle Signed-off-by: Vladimir Oltean Link: https://patch.msgid.link/20251130131657.65080-10-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/lantiq/lantiq_gswip_common.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/dsa/lantiq/lantiq_gswip_common.c b/drivers/net/dsa/lantiq/lantiq_gswip_common.c index 122ccea4057b..9da39edf8f57 100644 --- a/drivers/net/dsa/lantiq/lantiq_gswip_common.c +++ b/drivers/net/dsa/lantiq/lantiq_gswip_common.c @@ -1652,6 +1652,8 @@ static const struct dsa_switch_ops gswip_switch_ops = { .get_sset_count = gswip_get_sset_count, .set_mac_eee = gswip_set_mac_eee, .support_eee = gswip_support_eee, + .port_hsr_join = dsa_port_simple_hsr_join, + .port_hsr_leave = dsa_port_simple_hsr_leave, }; void gswip_disable_switch(struct gswip_priv *priv) From 4af9fa2ba65aa6bdad6f343b9e44704dce1ddf79 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Sun, 30 Nov 2025 15:16:52 +0200 Subject: [PATCH 1358/1381] net: dsa: mv88e6060: use simple HSR offload helpers The "trailer" tagging protocol uses dsa_xmit_port_mask(), which means we can offload HSR packet duplication on transmit. Enable that feature. Signed-off-by: Vladimir Oltean Link: https://patch.msgid.link/20251130131657.65080-11-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/mv88e6060.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/dsa/mv88e6060.c b/drivers/net/dsa/mv88e6060.c index 294312b58e4f..9c8ac14cd4f5 100644 --- a/drivers/net/dsa/mv88e6060.c +++ b/drivers/net/dsa/mv88e6060.c @@ -297,6 +297,8 @@ static const struct dsa_switch_ops mv88e6060_switch_ops = { .phy_read = mv88e6060_phy_read, .phy_write = mv88e6060_phy_write, .phylink_get_caps = mv88e6060_phylink_get_caps, + .port_hsr_join = dsa_port_simple_hsr_join, + .port_hsr_leave = dsa_port_simple_hsr_leave, }; static int mv88e6060_probe(struct mdio_device *mdiodev) From 017bcff7321af438f1190dbdf417815ea8ef60c0 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Sun, 30 Nov 2025 15:16:53 +0200 Subject: [PATCH 1359/1381] net: dsa: hellcreek: use simple HSR offload helpers The "hellcreek" tagging protocol uses dsa_xmit_port_mask(), which means we can offload HSR packet duplication on transmit. Enable that feature. Cc: Kurt Kanzenbach Signed-off-by: Vladimir Oltean Link: https://patch.msgid.link/20251130131657.65080-12-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/hirschmann/hellcreek.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/dsa/hirschmann/hellcreek.c b/drivers/net/dsa/hirschmann/hellcreek.c index e0b4758ca583..dd5f263ab984 100644 --- a/drivers/net/dsa/hirschmann/hellcreek.c +++ b/drivers/net/dsa/hirschmann/hellcreek.c @@ -1926,6 +1926,8 @@ static const struct dsa_switch_ops hellcreek_ds_ops = { .port_vlan_filtering = hellcreek_vlan_filtering, .setup = hellcreek_setup, .teardown = hellcreek_teardown, + .port_hsr_join = dsa_port_simple_hsr_join, + .port_hsr_leave = dsa_port_simple_hsr_leave, }; static int hellcreek_probe(struct platform_device *pdev) From 585943b7ad3092257ce2d696af32e2c783a566b8 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Sun, 30 Nov 2025 15:16:54 +0200 Subject: [PATCH 1360/1381] net: dsa: mt7530: use simple HSR offload helpers The "mtk" tagging protocol uses dsa_xmit_port_mask(), which means we can offload HSR packet duplication on transmit. Enable that feature. Cc: Daniel Golle Cc: DENG Qingfang Cc: Sean Wang Signed-off-by: Vladimir Oltean Acked-by: Chester A. Unal Link: https://patch.msgid.link/20251130131657.65080-13-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/mt7530.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/dsa/mt7530.c b/drivers/net/dsa/mt7530.c index 1acb57002014..b9423389c2ef 100644 --- a/drivers/net/dsa/mt7530.c +++ b/drivers/net/dsa/mt7530.c @@ -3290,6 +3290,8 @@ static const struct dsa_switch_ops mt7530_switch_ops = { .set_mac_eee = mt753x_set_mac_eee, .conduit_state_change = mt753x_conduit_state_change, .port_setup_tc = mt753x_setup_tc, + .port_hsr_join = dsa_port_simple_hsr_join, + .port_hsr_leave = dsa_port_simple_hsr_leave, }; static const struct phylink_mac_ops mt753x_phylink_mac_ops = { From 7271d4a08c399104de663762806a9a521352a0d1 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Sun, 30 Nov 2025 15:16:55 +0200 Subject: [PATCH 1361/1381] net: dsa: a5psw: use simple HSR offload helpers MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The "a5psw" tagging protocol uses dsa_xmit_port_mask(), which means we can offload HSR packet duplication on transmit. Enable that feature. Cc: "Clément Léger" Signed-off-by: Vladimir Oltean Link: https://patch.msgid.link/20251130131657.65080-14-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/rzn1_a5psw.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/dsa/rzn1_a5psw.c b/drivers/net/dsa/rzn1_a5psw.c index 1635255f58e4..4d857e3be10b 100644 --- a/drivers/net/dsa/rzn1_a5psw.c +++ b/drivers/net/dsa/rzn1_a5psw.c @@ -1035,6 +1035,8 @@ static const struct dsa_switch_ops a5psw_switch_ops = { .port_fdb_add = a5psw_port_fdb_add, .port_fdb_del = a5psw_port_fdb_del, .port_fdb_dump = a5psw_port_fdb_dump, + .port_hsr_join = dsa_port_simple_hsr_join, + .port_hsr_leave = dsa_port_simple_hsr_leave, }; static int a5psw_mdio_wait_busy(struct a5psw *a5psw) From 977839161f26862e77a2c9a46b4d7b0b00b20dad Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Sun, 30 Nov 2025 15:16:56 +0200 Subject: [PATCH 1362/1381] Documentation: net: dsa: mention availability of RedBox Since commit 5055cccfc2d1 ("net: hsr: Provide RedBox support (HSR-SAN)"), RedBox is available (including for offload in DSA). Update the DSA documentation that states it isn't. Signed-off-by: Vladimir Oltean Link: https://patch.msgid.link/20251130131657.65080-15-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- Documentation/networking/dsa/dsa.rst | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) diff --git a/Documentation/networking/dsa/dsa.rst b/Documentation/networking/dsa/dsa.rst index 7b2e69cd7ef0..92df551434f4 100644 --- a/Documentation/networking/dsa/dsa.rst +++ b/Documentation/networking/dsa/dsa.rst @@ -1104,12 +1104,11 @@ health of the network and for discovery of other nodes. In Linux, both HSR and PRP are implemented in the hsr driver, which instantiates a virtual, stackable network interface with two member ports. The driver only implements the basic roles of DANH (Doubly Attached Node -implementing HSR) and DANP (Doubly Attached Node implementing PRP); the roles -of RedBox and QuadBox are not implemented (therefore, bridging a hsr network -interface with a physical switch port does not produce the expected result). +implementing HSR), DANP (Doubly Attached Node implementing PRP) and RedBox +(allows non-HSR devices to connect to the ring via Interlink ports). -A driver which is able of offloading certain functions of a DANP or DANH should -declare the corresponding netdev features as indicated by the documentation at +A driver which is able of offloading certain functions should declare the +corresponding netdev features as indicated by the documentation at ``Documentation/networking/netdev-features.rst``. Additionally, the following methods must be implemented: From 4e4c00f34d5dbbb2b7174fc419747f0b3fe7a4d5 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Sun, 30 Nov 2025 15:16:57 +0200 Subject: [PATCH 1363/1381] Documentation: net: dsa: mention simple HSR offload helpers Keep the documentation up to date. Signed-off-by: Vladimir Oltean Link: https://patch.msgid.link/20251130131657.65080-16-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- Documentation/networking/dsa/dsa.rst | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/Documentation/networking/dsa/dsa.rst b/Documentation/networking/dsa/dsa.rst index 92df551434f4..5c79740a533b 100644 --- a/Documentation/networking/dsa/dsa.rst +++ b/Documentation/networking/dsa/dsa.rst @@ -1119,6 +1119,14 @@ methods must be implemented: - ``port_hsr_leave``: function invoked when a given switch port leaves a DANP/DANH and returns to normal operation as a standalone port. +Note that the ``NETIF_F_HW_HSR_DUP`` feature relies on transmission towards +multiple ports, which is generally available whenever the tagging protocol uses +the ``dsa_xmit_port_mask()`` helper function. If the helper is used, the HSR +offload feature should also be set. The ``dsa_port_simple_hsr_join()`` and +``dsa_port_simple_hsr_leave()`` methods can be used as generic implementations +of ``port_hsr_join`` and ``port_hsr_leave``, if this is the only supported +offload feature. + TODO ==== From d8e08149a5ed501d515dbe9f53cc46d25acb4a2a Mon Sep 17 00:00:00 2001 From: Florian Fuchs Date: Sun, 30 Nov 2025 20:41:55 +0100 Subject: [PATCH 1364/1381] net: ps3_gelic_net: Use napi_alloc_skb() and napi_gro_receive() Use the napi functions napi_alloc_skb() and napi_gro_receive() instead of netdev_alloc_skb() and netif_receive_skb() for more efficient packet receiving. The switch to napi aware functions increases the RX throughput, reduces the occurrence of retransmissions and improves the resilience against SKB allocation failures. Signed-off-by: Florian Fuchs Reviewed-by: Simon Horman Reviewed-by: Eric Dumazet Link: https://patch.msgid.link/20251130194155.1950980-1-fuchsfl@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/toshiba/ps3_gelic_net.c | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-) diff --git a/drivers/net/ethernet/toshiba/ps3_gelic_net.c b/drivers/net/ethernet/toshiba/ps3_gelic_net.c index 591866fc9055..d35d1f3c10a1 100644 --- a/drivers/net/ethernet/toshiba/ps3_gelic_net.c +++ b/drivers/net/ethernet/toshiba/ps3_gelic_net.c @@ -364,6 +364,7 @@ static int gelic_card_init_chain(struct gelic_card *card, * gelic_descr_prepare_rx - reinitializes a rx descriptor * @card: card structure * @descr: descriptor to re-init + * @napi_mode: is it running in napi poll * * return 0 on success, <0 on failure * @@ -374,7 +375,8 @@ static int gelic_card_init_chain(struct gelic_card *card, * must be a multiple of GELIC_NET_RXBUF_ALIGN. */ static int gelic_descr_prepare_rx(struct gelic_card *card, - struct gelic_descr *descr) + struct gelic_descr *descr, + bool napi_mode) { static const unsigned int rx_skb_size = ALIGN(GELIC_NET_MAX_FRAME, GELIC_NET_RXBUF_ALIGN) + @@ -392,7 +394,10 @@ static int gelic_descr_prepare_rx(struct gelic_card *card, descr->hw_regs.payload.dev_addr = 0; descr->hw_regs.payload.size = 0; - descr->skb = netdev_alloc_skb(*card->netdev, rx_skb_size); + if (napi_mode) + descr->skb = napi_alloc_skb(&card->napi, rx_skb_size); + else + descr->skb = netdev_alloc_skb(*card->netdev, rx_skb_size); if (!descr->skb) { descr->hw_regs.payload.dev_addr = 0; /* tell DMAC don't touch memory */ return -ENOMEM; @@ -464,7 +469,7 @@ static int gelic_card_fill_rx_chain(struct gelic_card *card) do { if (!descr->skb) { - ret = gelic_descr_prepare_rx(card, descr); + ret = gelic_descr_prepare_rx(card, descr, false); if (ret) goto rewind; } @@ -964,7 +969,7 @@ static void gelic_net_pass_skb_up(struct gelic_descr *descr, netdev->stats.rx_bytes += skb->len; /* pass skb up to stack */ - netif_receive_skb(skb); + napi_gro_receive(&card->napi, skb); } /** @@ -1069,7 +1074,7 @@ static int gelic_card_decode_one_descr(struct gelic_card *card) /* * this call can fail, propagate the error */ - prepare_rx_ret = gelic_descr_prepare_rx(card, descr); + prepare_rx_ret = gelic_descr_prepare_rx(card, descr, true); if (prepare_rx_ret) return prepare_rx_ret; From a8658f7bb6504af125242bb4d4e49ca032eccf91 Mon Sep 17 00:00:00 2001 From: Carolina Jubran Date: Sun, 30 Nov 2025 11:19:33 +0200 Subject: [PATCH 1365/1381] selftests: drv-net: Add devlink_rate_tc_bw.py to TEST_PROGS This makes devlink_rate_tc_bw.py present in the Makefile under the same directory. Signed-off-by: Carolina Jubran Reviewed-by: Cosmin Ratiu Reviewed-by: Nimrod Oren Link: https://patch.msgid.link/20251130091938.4109055-2-cjubran@nvidia.com Signed-off-by: Jakub Kicinski --- tools/testing/selftests/drivers/net/hw/Makefile | 1 + 1 file changed, 1 insertion(+) diff --git a/tools/testing/selftests/drivers/net/hw/Makefile b/tools/testing/selftests/drivers/net/hw/Makefile index 7c819fdfa107..9c163ba6feee 100644 --- a/tools/testing/selftests/drivers/net/hw/Makefile +++ b/tools/testing/selftests/drivers/net/hw/Makefile @@ -20,6 +20,7 @@ TEST_GEN_FILES := \ TEST_PROGS = \ csum.py \ devlink_port_split.py \ + devlink_rate_tc_bw.py \ devmem.py \ ethtool.sh \ ethtool_extended_state.sh \ From 2a60ce94c6e871cbe761e6019d40bb8cabb8d15d Mon Sep 17 00:00:00 2001 From: Carolina Jubran Date: Sun, 30 Nov 2025 11:19:34 +0200 Subject: [PATCH 1366/1381] selftests: drv-net: introduce Iperf3Runner for measurement use cases GenerateTraffic was added to spin up long-running iperf3 load, mainly to drive high PPS background traffic. It was never meant to provide stable throughput numbers, and trying to repurpose it for measurement does not make sense. Introduce Iperf3Runner to allow tests to split out server/client configuration, control start/stop, and collect JSON output for analysis. This makes it possible to measure bandwidth directly when validating egress shaping. GenerateTraffic stays as the background load generator, reusing the common iperf3 helpers under the hood. Signed-off-by: Carolina Jubran Reviewed-by: Cosmin Ratiu Reviewed-by: Nimrod Oren Link: https://patch.msgid.link/20251130091938.4109055-3-cjubran@nvidia.com Signed-off-by: Jakub Kicinski --- .../drivers/net/hw/lib/py/__init__.py | 5 +- .../selftests/drivers/net/lib/py/__init__.py | 5 +- .../selftests/drivers/net/lib/py/load.py | 84 +++++++++++++++++-- 3 files changed, 82 insertions(+), 12 deletions(-) diff --git a/tools/testing/selftests/drivers/net/hw/lib/py/__init__.py b/tools/testing/selftests/drivers/net/hw/lib/py/__init__.py index 0c61debf86fb..766bfc4ad842 100644 --- a/tools/testing/selftests/drivers/net/hw/lib/py/__init__.py +++ b/tools/testing/selftests/drivers/net/hw/lib/py/__init__.py @@ -28,7 +28,7 @@ try: ksft_setup, ksft_variants, KsftNamedVariant from net.lib.py import ksft_eq, ksft_ge, ksft_in, ksft_is, ksft_lt, \ ksft_ne, ksft_not_in, ksft_raises, ksft_true, ksft_gt, ksft_not_none - from drivers.net.lib.py import GenerateTraffic, Remote + from drivers.net.lib.py import GenerateTraffic, Remote, Iperf3Runner from drivers.net.lib.py import NetDrvEnv, NetDrvEpEnv __all__ = ["NetNS", "NetNSEnter", "NetdevSimDev", @@ -44,7 +44,8 @@ try: "ksft_eq", "ksft_ge", "ksft_in", "ksft_is", "ksft_lt", "ksft_ne", "ksft_not_in", "ksft_raises", "ksft_true", "ksft_gt", "ksft_not_none", "ksft_not_none", - "NetDrvEnv", "NetDrvEpEnv", "GenerateTraffic", "Remote"] + "NetDrvEnv", "NetDrvEpEnv", "GenerateTraffic", "Remote", + "Iperf3Runner"] except ModuleNotFoundError as e: print("Failed importing `net` library from kernel sources") print(str(e)) diff --git a/tools/testing/selftests/drivers/net/lib/py/__init__.py b/tools/testing/selftests/drivers/net/lib/py/__init__.py index d9d035634a31..8b75faa9af6d 100644 --- a/tools/testing/selftests/drivers/net/lib/py/__init__.py +++ b/tools/testing/selftests/drivers/net/lib/py/__init__.py @@ -44,10 +44,11 @@ try: "ksft_not_none", "ksft_not_none"] from .env import NetDrvEnv, NetDrvEpEnv - from .load import GenerateTraffic + from .load import GenerateTraffic, Iperf3Runner from .remote import Remote - __all__ += ["NetDrvEnv", "NetDrvEpEnv", "GenerateTraffic", "Remote"] + __all__ += ["NetDrvEnv", "NetDrvEpEnv", "GenerateTraffic", "Remote", + "Iperf3Runner"] except ModuleNotFoundError as e: print("Failed importing `net` library from kernel sources") print(str(e)) diff --git a/tools/testing/selftests/drivers/net/lib/py/load.py b/tools/testing/selftests/drivers/net/lib/py/load.py index c4e808407cc4..f181fa2d38fc 100644 --- a/tools/testing/selftests/drivers/net/lib/py/load.py +++ b/tools/testing/selftests/drivers/net/lib/py/load.py @@ -2,21 +2,89 @@ import re import time +import json from lib.py import ksft_pr, cmd, ip, rand_port, wait_port_listen -class GenerateTraffic: - def __init__(self, env, port=None): + +class Iperf3Runner: + """ + Sets up and runs iperf3 traffic. + """ + def __init__(self, env, port=None, server_ip=None, client_ip=None): env.require_cmd("iperf3", local=True, remote=True) - self.env = env - self.port = rand_port() if port is None else port - self._iperf_server = cmd(f"iperf3 -s -1 -p {self.port}", background=True) + self.server_ip = server_ip + self.client_ip = client_ip + + def _build_server(self): + cmdline = f"iperf3 -s -1 -p {self.port}" + if self.server_ip: + cmdline += f" -B {self.server_ip}" + return cmdline + + def _build_client(self, streams, duration, reverse): + host = self.env.addr if self.server_ip is None else self.server_ip + cmdline = f"iperf3 -c {host} -p {self.port} -P {streams} -t {duration} -J" + if self.client_ip: + cmdline += f" -B {self.client_ip}" + if reverse: + cmdline += " --reverse" + return cmdline + + def start_server(self): + """ + Starts an iperf3 server with optional bind IP. + """ + cmdline = self._build_server() + proc = cmd(cmdline, background=True) wait_port_listen(self.port) time.sleep(0.1) - self._iperf_client = cmd(f"iperf3 -c {env.addr} -P 16 -p {self.port} -t 86400", - background=True, host=env.remote) + return proc + + def start_client(self, background=False, streams=1, duration=10, reverse=False): + """ + Starts the iperf3 client with the configured options. + """ + cmdline = self._build_client(streams, duration, reverse) + return cmd(cmdline, background=background, host=self.env.remote) + + def measure_bandwidth(self, reverse=False): + """ + Runs an iperf3 measurement and returns the average bandwidth (Gbps). + Discards the first and last few reporting intervals and uses only the + middle part of the run where throughput is typically stable. + """ + self.start_server() + result = self.start_client(duration=10, reverse=reverse) + + if result.ret != 0: + raise RuntimeError("iperf3 failed to run successfully") + try: + out = json.loads(result.stdout) + except json.JSONDecodeError as exc: + raise ValueError("Failed to parse iperf3 JSON output") from exc + + intervals = out.get("intervals", []) + samples = [i["sum"]["bits_per_second"] / 1e9 for i in intervals] + if len(samples) < 10: + raise ValueError(f"iperf3 returned too few intervals: {len(samples)}") + # Discard potentially unstable first and last 3 seconds. + stable = samples[3:-3] + + avg = sum(stable) / len(stable) + + return avg + + +class GenerateTraffic: + def __init__(self, env, port=None): + self.env = env + self.runner = Iperf3Runner(env, port) + + self._iperf_server = self.runner.start_server() + self._iperf_client = self.runner.start_client(background=True, streams=16, duration=86400) # Wait for traffic to ramp up if not self._wait_pkts(pps=1000): @@ -61,7 +129,7 @@ class GenerateTraffic: def _wait_client_stopped(self, sleep=0.005, timeout=5): end = time.monotonic() + timeout - live_port_pattern = re.compile(fr":{self.port:04X} 0[^6] ") + live_port_pattern = re.compile(fr":{self.runner.port:04X} 0[^6] ") while time.monotonic() < end: data = cmd("cat /proc/net/tcp*", host=self.env.remote).stdout From cb1acbd30a42204b39e4846e1139b6fa9a40fafe Mon Sep 17 00:00:00 2001 From: Carolina Jubran Date: Sun, 30 Nov 2025 11:19:35 +0200 Subject: [PATCH 1367/1381] selftests: drv-net: Use Iperf3Runner in devlink_rate_tc_bw.py Replace the inline iperf3 subprocess and JSON parsing with Iperf3Runner. Signed-off-by: Carolina Jubran Reviewed-by: Cosmin Ratiu Reviewed-by: Nimrod Oren Link: https://patch.msgid.link/20251130091938.4109055-4-cjubran@nvidia.com Signed-off-by: Jakub Kicinski --- .../drivers/net/hw/devlink_rate_tc_bw.py | 70 ++++++++----------- 1 file changed, 29 insertions(+), 41 deletions(-) diff --git a/tools/testing/selftests/drivers/net/hw/devlink_rate_tc_bw.py b/tools/testing/selftests/drivers/net/hw/devlink_rate_tc_bw.py index ead6784d1910..16e5dda5bee1 100755 --- a/tools/testing/selftests/drivers/net/hw/devlink_rate_tc_bw.py +++ b/tools/testing/selftests/drivers/net/hw/devlink_rate_tc_bw.py @@ -64,6 +64,7 @@ from lib.py import KsftSkipEx, KsftFailEx, KsftXfailEx from lib.py import NetDrvEpEnv, DevlinkFamily from lib.py import NlError from lib.py import cmd, defer, ethtool, ip +from lib.py import Iperf3Runner class BandwidthValidator: @@ -139,8 +140,8 @@ def setup_vlans_on_vf(vf_ifc): Sets up two VLAN interfaces on the given VF, each mapped to a different TC. """ vlan_configs = [ - {"vlan_id": 101, "tc": 3, "ip": "198.51.100.2"}, - {"vlan_id": 102, "tc": 4, "ip": "198.51.100.10"}, + {"vlan_id": 101, "tc": 3, "ip": "198.51.100.1"}, + {"vlan_id": 102, "tc": 4, "ip": "198.51.100.9"}, ] for config in vlan_configs: @@ -224,13 +225,13 @@ def setup_devlink_rate(cfg): raise KsftFailEx(f"rate_set failed on VF port {port_index}") from exc -def setup_remote_server(cfg): +def setup_remote_vlans(cfg): """ - Sets up VLAN interfaces and starts iperf3 servers on the remote side. + Sets up VLAN interfaces on the remote side. """ remote_dev = cfg.remote_ifname vlan_ids = [101, 102] - remote_ips = ["198.51.100.1", "198.51.100.9"] + remote_ips = ["198.51.100.2", "198.51.100.10"] for vlan_id, ip_addr in zip(vlan_ids, remote_ips): vlan_dev = f"{remote_dev}.{vlan_id}" @@ -238,14 +239,13 @@ def setup_remote_server(cfg): f"type vlan id {vlan_id}", host=cfg.remote) cmd(f"ip addr add {ip_addr}/29 dev {vlan_dev}", host=cfg.remote) cmd(f"ip link set dev {vlan_dev} up", host=cfg.remote) - cmd(f"iperf3 -s -1 -B {ip_addr}",background=True, host=cfg.remote) defer(cmd, f"ip link del {vlan_dev}", host=cfg.remote) def setup_test_environment(cfg, set_tc_mapping=True): """ Sets up the complete test environment including VF creation, VLANs, - bridge configuration, devlink rate setup, and the remote server. + bridge configuration and devlink rate setup. """ vf_ifc = setup_vf(cfg, set_tc_mapping) ksft_pr(f"Created VF interface: {vf_ifc}") @@ -256,51 +256,39 @@ def setup_test_environment(cfg, set_tc_mapping=True): setup_bridge(cfg) setup_devlink_rate(cfg) - setup_remote_server(cfg) - time.sleep(2) + setup_remote_vlans(cfg) -def run_iperf_client(server_ip, local_ip, barrier, min_expected_gbps=0.1): +def measure_bandwidth(cfg, server_ip, client_ip, barrier): """ - Runs a single iperf3 client instance, binding to the given local IP. - Waits on a barrier to synchronize with other threads. + Synchronizes with peers and runs an iperf3-based bandwidth measurement + between the given endpoints. Returns average Gbps. """ + runner = Iperf3Runner(cfg, server_ip=server_ip, client_ip=client_ip) try: barrier.wait(timeout=10) except Exception as exc: raise KsftFailEx("iperf3 barrier wait timed") from exc - iperf_cmd = ["iperf3", "-c", server_ip, "-B", local_ip, "-J"] - result = subprocess.run(iperf_cmd, capture_output=True, text=True, - check=True) - try: - output = json.loads(result.stdout) - bits_per_second = output["end"]["sum_received"]["bits_per_second"] - gbps = bits_per_second / 1e9 - if gbps < min_expected_gbps: - ksft_pr( - f"iperf3 bandwidth too low: {gbps:.2f} Gbps " - f"(expected ≥ {min_expected_gbps} Gbps)" - ) - return None - return gbps - except json.JSONDecodeError as exc: - ksft_pr(f"Failed to parse iperf3 JSON output: {exc}") - return None + bw_gbps = runner.measure_bandwidth(reverse=True) + except Exception as exc: + raise KsftFailEx("iperf3 bandwidth measurement failed") from exc + + return bw_gbps -def run_bandwidth_test(): +def run_bandwidth_test(cfg): """ - Launches iperf3 client threads for each VLAN/TC pair and collects results. + Runs parallel bandwidth measurements for each VLAN/TC pair and collects results. """ - def _run_iperf_client_thread(server_ip, local_ip, results, barrier, tc_ix): - results[tc_ix] = run_iperf_client(server_ip, local_ip, barrier) + def _run_measure_bandwidth_thread(local_ip, remote_ip, results, barrier, tc_ix): + results[tc_ix] = measure_bandwidth(cfg, local_ip, remote_ip, barrier) vf_vlan_data = [ # (local_ip, remote_ip, TC) - ("198.51.100.2", "198.51.100.1", 3), - ("198.51.100.10", "198.51.100.9", 4), + ("198.51.100.1", "198.51.100.2", 3), + ("198.51.100.9", "198.51.100.10", 4), ] results = {} @@ -309,8 +297,8 @@ def run_bandwidth_test(): for local_ip, remote_ip, tc_ix in vf_vlan_data: thread = threading.Thread( - target=_run_iperf_client_thread, - args=(remote_ip, local_ip, results, start_barrier, tc_ix) + target=_run_measure_bandwidth_thread, + args=(local_ip, remote_ip, results, start_barrier, tc_ix) ) thread.start() threads.append(thread) @@ -320,10 +308,11 @@ def run_bandwidth_test(): for tc_ix, tc_bw in results.items(): if tc_bw is None: - raise KsftFailEx("iperf3 client failed; cannot evaluate bandwidth") + raise KsftFailEx("iperf3 failed; cannot evaluate bandwidth") return results + def calculate_bandwidth_percentages(results): """ Calculates the percentage of total bandwidth received by TC3 and TC4. @@ -398,10 +387,10 @@ def check_bandwidth_distribution(bw_data, validator): def run_bandwidth_distribution_test(cfg, set_tc_mapping): """ - Runs parallel iperf3 tests for both TCs and collects results. + Runs parallel bandwidth measurements for both TCs and collects results. """ setup_test_environment(cfg, set_tc_mapping) - bandwidths = run_bandwidth_test() + bandwidths = run_bandwidth_test(cfg) bw_data = calculate_bandwidth_percentages(bandwidths) test_name = "with TC mapping" if set_tc_mapping else "without TC mapping" print_bandwidth_results(bw_data, test_name) @@ -451,7 +440,6 @@ def main() -> None: ) if not cfg.pci: raise KsftSkipEx("Could not get PCI address of the interface") - cfg.require_cmd("iperf3", local=True, remote=True) cfg.bw_validator = BandwidthValidator() From 3796e549e305194ac379c843f0eec5e013276fc3 Mon Sep 17 00:00:00 2001 From: Carolina Jubran Date: Sun, 30 Nov 2025 11:19:36 +0200 Subject: [PATCH 1368/1381] selftests: drv-net: Set shell=True for sysfs writes in devlink_rate_tc_bw.py Commit 7c32f7a2d3db ("selftests: net: py: don't default to shell=True") changed the cmd() helper to avoid spawning a shell unless explicitly requested. The devlink_rate_tc_bw test enables SR-IOV by writing to the sriov_numvfs sysfs attribute using redirection. Without shell=True the redirection is not interpreted and the VF device never appears, causing the test to fail. Fix by explicitly passing shell=True in the two places that update sriov_numvfs. Signed-off-by: Carolina Jubran Reviewed-by: Cosmin Ratiu Reviewed-by: Nimrod Oren Link: https://patch.msgid.link/20251130091938.4109055-5-cjubran@nvidia.com Signed-off-by: Jakub Kicinski --- tools/testing/selftests/drivers/net/hw/devlink_rate_tc_bw.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tools/testing/selftests/drivers/net/hw/devlink_rate_tc_bw.py b/tools/testing/selftests/drivers/net/hw/devlink_rate_tc_bw.py index 16e5dda5bee1..df71936531cf 100755 --- a/tools/testing/selftests/drivers/net/hw/devlink_rate_tc_bw.py +++ b/tools/testing/selftests/drivers/net/hw/devlink_rate_tc_bw.py @@ -117,8 +117,8 @@ def setup_vf(cfg, set_tc_mapping=True): except Exception as exc: raise KsftSkipEx(f"Failed to enable switchdev mode on {cfg.pci}") from exc try: - cmd(f"echo 1 > /sys/class/net/{cfg.ifname}/device/sriov_numvfs") - defer(cmd, f"echo 0 > /sys/class/net/{cfg.ifname}/device/sriov_numvfs") + cmd(f"echo 1 > /sys/class/net/{cfg.ifname}/device/sriov_numvfs", shell=True) + defer(cmd, f"echo 0 > /sys/class/net/{cfg.ifname}/device/sriov_numvfs", shell=True) except Exception as exc: raise KsftSkipEx(f"Failed to enable SR-IOV on {cfg.ifname}") from exc From 9ecd05a2c872d5eb2ae360d2907cf9375cf2ef5a Mon Sep 17 00:00:00 2001 From: Carolina Jubran Date: Sun, 30 Nov 2025 11:19:37 +0200 Subject: [PATCH 1369/1381] selftests: drv-net: Fix and clarify TC bandwidth split in devlink_rate_tc_bw.py Correct the documented bandwidth distribution between TC3 and TC4 from 80/20 to 20/80. Update test descriptions and printed messages to consistently reflect the intended split. Signed-off-by: Carolina Jubran Reviewed-by: Cosmin Ratiu Reviewed-by: Nimrod Oren Link: https://patch.msgid.link/20251130091938.4109055-6-cjubran@nvidia.com Signed-off-by: Jakub Kicinski --- .../drivers/net/hw/devlink_rate_tc_bw.py | 26 +++++++++---------- 1 file changed, 13 insertions(+), 13 deletions(-) diff --git a/tools/testing/selftests/drivers/net/hw/devlink_rate_tc_bw.py b/tools/testing/selftests/drivers/net/hw/devlink_rate_tc_bw.py index df71936531cf..db24bfc51da4 100755 --- a/tools/testing/selftests/drivers/net/hw/devlink_rate_tc_bw.py +++ b/tools/testing/selftests/drivers/net/hw/devlink_rate_tc_bw.py @@ -21,21 +21,21 @@ Test Cases: ---------- 1. test_no_tc_mapping_bandwidth: - Verifies that without TC mapping, bandwidth is NOT distributed according to - the configured 80/20 split between TC4 and TC3 - - This test should fail if bandwidth matches the 80/20 split without TC + the configured 20/80 split between TC3 and TC4 + - This test should fail if bandwidth matches the 20/80 split without TC mapping - - Expected: Bandwidth should NOT be distributed as 80/20 + - Expected: Bandwidth should NOT be distributed as 20/80 2. test_tc_mapping_bandwidth: - Configures TC mapping using mqprio qdisc - Verifies that with TC mapping, bandwidth IS distributed according to the - configured 80/20 split between TC3 and TC4 - - Expected: Bandwidth should be distributed as 80/20 + configured 20/80 split between TC3 and TC4 + - Expected: Bandwidth should be distributed as 20/80 Bandwidth Distribution: ---------------------- -- TC3 (VLAN 101): Configured for 80% of total bandwidth -- TC4 (VLAN 102): Configured for 20% of total bandwidth +- TC3 (VLAN 101): Configured for 20% of total bandwidth +- TC4 (VLAN 102): Configured for 80% of total bandwidth - Total bandwidth: 1Gbps - Tolerance: +-12% @@ -402,10 +402,10 @@ def run_bandwidth_distribution_test(cfg, set_tc_mapping): def test_no_tc_mapping_bandwidth(cfg): """ - Verifies that bandwidth is not split 80/20 without traffic class mapping. + Verifies that bandwidth is not split 20/80 without traffic class mapping. """ - pass_bw_msg = "Bandwidth is NOT distributed as 80/20 without TC mapping" - fail_bw_msg = "Bandwidth matched 80/20 split without TC mapping" + pass_bw_msg = "Bandwidth is NOT distributed as 20/80 without TC mapping" + fail_bw_msg = "Bandwidth matched 20/80 split without TC mapping" is_mlx5 = "driver: mlx5" in ethtool(f"-i {cfg.ifname}").stdout if run_bandwidth_distribution_test(cfg, set_tc_mapping=False): @@ -419,13 +419,13 @@ def test_no_tc_mapping_bandwidth(cfg): def test_tc_mapping_bandwidth(cfg): """ - Verifies that bandwidth is correctly split 80/20 between TC3 and TC4 + Verifies that bandwidth is correctly split 20/80 between TC3 and TC4 when traffic class mapping is set. """ if run_bandwidth_distribution_test(cfg, set_tc_mapping=True): - ksft_pr("Bandwidth is distributed as 80/20 with TC mapping") + ksft_pr("Bandwidth is distributed as 20/80 with TC mapping") else: - raise KsftFailEx("Bandwidth did not match 80/20 split with TC mapping") + raise KsftFailEx("Bandwidth did not match 20/80 split with TC mapping") def main() -> None: From 5cc1bddcfeb298fe7e67952a50894de8afa46db5 Mon Sep 17 00:00:00 2001 From: Carolina Jubran Date: Sun, 30 Nov 2025 11:19:38 +0200 Subject: [PATCH 1370/1381] selftests: drv-net: Fix tolerance calculation in devlink_rate_tc_bw.py MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Currently, tolerance is computed against the TC’s expected percentage, making TC3 (20%) validation overly strict and TC4 (80%) overly loose. Update BandwidthValidator to take a dict of shares and compute bounds relative to the overall total, so that all shares are validated consistently. Signed-off-by: Carolina Jubran Reviewed-by: Cosmin Ratiu Reviewed-by: Nimrod Oren Link: https://patch.msgid.link/20251130091938.4109055-7-cjubran@nvidia.com Signed-off-by: Jakub Kicinski --- .../drivers/net/hw/devlink_rate_tc_bw.py | 74 ++++++++----------- 1 file changed, 30 insertions(+), 44 deletions(-) diff --git a/tools/testing/selftests/drivers/net/hw/devlink_rate_tc_bw.py b/tools/testing/selftests/drivers/net/hw/devlink_rate_tc_bw.py index db24bfc51da4..4e4faa9275bb 100755 --- a/tools/testing/selftests/drivers/net/hw/devlink_rate_tc_bw.py +++ b/tools/testing/selftests/drivers/net/hw/devlink_rate_tc_bw.py @@ -69,39 +69,35 @@ from lib.py import Iperf3Runner class BandwidthValidator: """ - Validates bandwidth totals and per-TC shares against expected values - with a tolerance. + Validates total bandwidth and individual shares with tolerance + relative to the overall total. """ - def __init__(self): + def __init__(self, shares): self.tolerance_percent = 12 - self.expected_total_gbps = 1.0 - self.total_min_expected = self.min_expected(self.expected_total_gbps) - self.total_max_expected = self.max_expected(self.expected_total_gbps) - self.tc_expected_percent = { - 3: 20.0, - 4: 80.0, - } + self.expected_total = sum(shares.values()) + self.bounds = {} + + for name, exp in shares.items(): + self.bounds[name] = (self.min_expected(exp), self.max_expected(exp)) def min_expected(self, value): """Calculates the minimum acceptable value based on tolerance.""" - return value - (value * self.tolerance_percent / 100) + return value - (self.expected_total * self.tolerance_percent / 100) def max_expected(self, value): """Calculates the maximum acceptable value based on tolerance.""" - return value + (value * self.tolerance_percent / 100) + return value + (self.expected_total * self.tolerance_percent / 100) - def bound(self, expected, value): - """Returns True if value is within expected tolerance.""" - return self.min_expected(expected) <= value <= self.max_expected(expected) - - def tc_bandwidth_bound(self, value, tc_ix): + def bound(self, values): """ - Returns True if the given bandwidth value is within tolerance - for the TC's expected bandwidth. + Return True if all given values fall within tolerance. """ - expected = self.tc_expected_percent[tc_ix] - return self.bound(expected, value) + for name, value in values.items(): + low, high = self.bounds[name] + if not low <= value <= high: + return False + return True def setup_vf(cfg, set_tc_mapping=True): @@ -353,38 +349,26 @@ def verify_total_bandwidth(bw_data, validator): """ total = bw_data['total_bw'] - if validator.bound(validator.expected_total_gbps, total): + if validator.bound({"total": total}): return - if total < validator.total_min_expected: + low, high = validator.bounds["total"] + + if total < low: raise KsftSkipEx( f"Total bandwidth {total:.2f} Gbps < minimum " - f"{validator.total_min_expected:.2f} Gbps; " - f"parent tx_max ({validator.expected_total_gbps:.1f} G) " + f"{low:.2f} Gbps; " + f"parent tx_max ({validator.expected_total:.1f} G) " f"not reached, cannot validate share" ) raise KsftFailEx( f"Total bandwidth {total:.2f} Gbps exceeds allowed ceiling " - f"{validator.total_max_expected:.2f} Gbps " - f"(VF tx_max set to {validator.expected_total_gbps:.1f} G)" + f"{high:.2f} Gbps " + f"(VF tx_max set to {validator.expected_total:.1f} G)" ) -def check_bandwidth_distribution(bw_data, validator): - """ - Checks whether the measured TC3 and TC4 bandwidth percentages - fall within their expected tolerance ranges. - - Returns: - bool: True if both TC3 and TC4 percentages are within bounds. - """ - tc3_valid = validator.tc_bandwidth_bound(bw_data['tc3_percentage'], 3) - tc4_valid = validator.tc_bandwidth_bound(bw_data['tc4_percentage'], 4) - - return tc3_valid and tc4_valid - - def run_bandwidth_distribution_test(cfg, set_tc_mapping): """ Runs parallel bandwidth measurements for both TCs and collects results. @@ -395,9 +379,10 @@ def run_bandwidth_distribution_test(cfg, set_tc_mapping): test_name = "with TC mapping" if set_tc_mapping else "without TC mapping" print_bandwidth_results(bw_data, test_name) - verify_total_bandwidth(bw_data, cfg.bw_validator) + verify_total_bandwidth(bw_data, cfg.traffic_bw_validator) - return check_bandwidth_distribution(bw_data, cfg.bw_validator) + return cfg.tc_bw_validator.bound({"tc3": bw_data['tc3_percentage'], + "tc4": bw_data['tc4_percentage']}) def test_no_tc_mapping_bandwidth(cfg): @@ -441,7 +426,8 @@ def main() -> None: if not cfg.pci: raise KsftSkipEx("Could not get PCI address of the interface") - cfg.bw_validator = BandwidthValidator() + cfg.traffic_bw_validator = BandwidthValidator({"total": 1}) + cfg.tc_bw_validator = BandwidthValidator({"tc3": 20, "tc4": 80}) cases = [test_no_tc_mapping_bandwidth, test_tc_mapping_bandwidth] From 6b0f4ca079dbe6ae4aa57e529d67c7dc00d63577 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Asbj=C3=B8rn=20Sloth=20T=C3=B8nnesen?= Date: Wed, 26 Nov 2025 17:35:37 +0000 Subject: [PATCH 1371/1381] wireguard: netlink: add YNL specification MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This patch adds a near[1] complete YNL specification for WireGuard, documenting the protocol in a machine-readable format, rather than comments in wireguard.h, and eases usage from C and non-C programming languages alike. The generated C library will be featured in a later patch, so in this patch I will use the in-kernel python client for examples. This makes the documentation in the UAPI header redundant, it is therefore removed. The in-line documentation in the spec is based on the existing comment in wireguard.h, and once released it will be available in the kernel documentation at: https://docs.kernel.org/netlink/specs/wireguard.html (until then run: make htmldocs) Generate wireguard.rst from this spec: $ make -C tools/net/ynl/generated/ wireguard.rst Query wireguard interface through pyynl: $ sudo ./tools/net/ynl/pyynl/cli.py --family wireguard \ --dump get-device \ --json '{"ifindex":3}' [{'fwmark': 0, 'ifindex': 3, 'ifname': 'wg-test', 'listen-port': 54318, 'peers': [{0: {'allowedips': [{0: {'cidr-mask': 0, 'family': 2, 'ipaddr': '0.0.0.0'}}, {0: {'cidr-mask': 0, 'family': 10, 'ipaddr': '::'}}], 'endpoint': b'[...]', 'last-handshake-time': {'nsec': 42, 'sec': 42}, 'persistent-keepalive-interval': 42, 'preshared-key': '[...]', 'protocol-version': 1, 'public-key': '[...]', 'rx-bytes': 42, 'tx-bytes': 42}}], 'private-key': '[...]', 'public-key': '[...]'}] Add another allowed IP prefix: $ sudo ./tools/net/ynl/pyynl/cli.py --family wireguard \ --do set-device --json '{"ifindex":3,"peers":[ {"public-key":"6a df b1 83 a4 ..","allowedips":[ {"cidr-mask":0,"family":10,"ipaddr":"::"}]}]}' [1] As can be seen above, the "endpoint" is only dumped as binary data, as it can't be fully described in YNL. It's either a struct sockaddr_in or struct sockaddr_in6 depending on the attribute length. Signed-off-by: Asbjørn Sloth Tønnesen Signed-off-by: Jason A. Donenfeld --- Documentation/netlink/specs/wireguard.yaml | 298 +++++++++++++++++++++ MAINTAINERS | 1 + include/uapi/linux/wireguard.h | 129 --------- 3 files changed, 299 insertions(+), 129 deletions(-) create mode 100644 Documentation/netlink/specs/wireguard.yaml diff --git a/Documentation/netlink/specs/wireguard.yaml b/Documentation/netlink/specs/wireguard.yaml new file mode 100644 index 000000000000..30479fc6bb69 --- /dev/null +++ b/Documentation/netlink/specs/wireguard.yaml @@ -0,0 +1,298 @@ +# SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause) +--- +name: wireguard +protocol: genetlink-legacy + +doc: | + **Netlink protocol to control WireGuard network devices.** + + The below enums and macros are for interfacing with WireGuard, using generic + netlink, with family ``WG_GENL_NAME`` and version ``WG_GENL_VERSION``. It + defines two commands: get and set. Note that while they share many common + attributes, these two commands actually accept a slightly different set of + inputs and outputs. These differences are noted under the individual + attributes. +c-family-name: wg-genl-name +c-version-name: wg-genl-version +max-by-define: true + +definitions: + - + name-prefix: wg- + name: key-len + type: const + value: 32 + - + name: --kernel-timespec + type: struct + header: linux/time_types.h + members: + - + name: sec + type: u64 + doc: Number of seconds, since UNIX epoch. + - + name: nsec + type: u64 + doc: Number of nanoseconds, after the second began. + - + name: wgdevice-flags + name-prefix: wgdevice-f- + enum-name: wgdevice-flag + type: flags + entries: + - replace-peers + - + name: wgpeer-flags + name-prefix: wgpeer-f- + enum-name: wgpeer-flag + type: flags + entries: + - remove-me + - replace-allowedips + - update-only + - + name: wgallowedip-flags + name-prefix: wgallowedip-f- + enum-name: wgallowedip-flag + type: flags + entries: + - remove-me + +attribute-sets: + - + name: wgdevice + enum-name: wgdevice-attribute + name-prefix: wgdevice-a- + attr-cnt-name: --wgdevice-a-last + attributes: + - + name: unspec + type: unused + value: 0 + - + name: ifindex + type: u32 + - + name: ifname + type: string + checks: + max-len: 15 + - + name: private-key + type: binary + doc: Set to all zeros to remove. + display-hint: hex + checks: + exact-len: wg-key-len + - + name: public-key + type: binary + display-hint: hex + checks: + exact-len: wg-key-len + - + name: flags + type: u32 + doc: | + ``0`` or ``WGDEVICE_F_REPLACE_PEERS`` if all current peers should be + removed prior to adding the list below. + enum: wgdevice-flags + - + name: listen-port + type: u16 + doc: Set as ``0`` to choose randomly. + - + name: fwmark + type: u32 + doc: Set as ``0`` to disable. + - + name: peers + type: indexed-array + sub-type: nest + nested-attributes: wgpeer + doc: | + The index/type parameter is unused on ``SET_DEVICE`` operations and is + zero on ``GET_DEVICE`` operations. + - + name: wgpeer + enum-name: wgpeer-attribute + name-prefix: wgpeer-a- + attr-cnt-name: --wgpeer-a-last + attributes: + - + name: unspec + type: unused + value: 0 + - + name: public-key + type: binary + display-hint: hex + checks: + exact-len: wg-key-len + - + name: preshared-key + type: binary + doc: Set as all zeros to remove. + display-hint: hex + checks: + exact-len: wg-key-len + - + name: flags + type: u32 + doc: | + ``0`` and/or ``WGPEER_F_REMOVE_ME`` if the specified peer should not + exist at the end of the operation, rather than added/updated and/or + ``WGPEER_F_REPLACE_ALLOWEDIPS`` if all current allowed IPs of this + peer should be removed prior to adding the list below and/or + ``WGPEER_F_UPDATE_ONLY`` if the peer should only be set if it already + exists. + enum: wgpeer-flags + - + name: endpoint + type: binary + doc: struct sockaddr_in or struct sockaddr_in6 + checks: + min-len: 16 + - + name: persistent-keepalive-interval + type: u16 + doc: Set as ``0`` to disable. + - + name: last-handshake-time + type: binary + struct: --kernel-timespec + checks: + exact-len: 16 + - + name: rx-bytes + type: u64 + - + name: tx-bytes + type: u64 + - + name: allowedips + type: indexed-array + sub-type: nest + nested-attributes: wgallowedip + doc: | + The index/type parameter is unused on ``SET_DEVICE`` operations and is + zero on ``GET_DEVICE`` operations. + - + name: protocol-version + type: u32 + doc: | + Should not be set or used at all by most users of this API, as the + most recent protocol will be used when this is unset. Otherwise, + must be set to ``1``. + - + name: wgallowedip + enum-name: wgallowedip-attribute + name-prefix: wgallowedip-a- + attr-cnt-name: --wgallowedip-a-last + attributes: + - + name: unspec + type: unused + value: 0 + - + name: family + type: u16 + doc: IP family, either ``AF_INET`` or ``AF_INET6``. + - + name: ipaddr + type: binary + doc: Either ``struct in_addr`` or ``struct in6_addr``. + display-hint: ipv4-or-v6 + checks: + min-len: 4 + - + name: cidr-mask + type: u8 + - + name: flags + type: u32 + doc: | + ``WGALLOWEDIP_F_REMOVE_ME`` if the specified IP should be removed; + otherwise, this IP will be added if it is not already present. + enum: wgallowedip-flags + +operations: + enum-name: wg-cmd + name-prefix: wg-cmd- + list: + - + name: get-device + value: 0 + doc: | + Retrieve WireGuard device + ~~~~~~~~~~~~~~~~~~~~~~~~~ + + The command should be called with one but not both of: + + - ``WGDEVICE_A_IFINDEX`` + - ``WGDEVICE_A_IFNAME`` + + The kernel will then return several messages (``NLM_F_MULTI``). It is + possible that all of the allowed IPs of a single peer will not fit + within a single netlink message. In that case, the same peer will be + written in the following message, except it will only contain + ``WGPEER_A_PUBLIC_KEY`` and ``WGPEER_A_ALLOWEDIPS``. This may occur + several times in a row for the same peer. It is then up to the receiver + to coalesce adjacent peers. Likewise, it is possible that all peers will + not fit within a single message. So, subsequent peers will be sent in + following messages, except those will only contain ``WGDEVICE_A_IFNAME`` + and ``WGDEVICE_A_PEERS``. It is then up to the receiver to coalesce + these messages to form the complete list of peers. + + Since this is an ``NLA_F_DUMP`` command, the final message will always + be ``NLMSG_DONE``, even if an error occurs. However, this ``NLMSG_DONE`` + message contains an integer error code. It is either zero or a negative + error code corresponding to the errno. + attribute-set: wgdevice + flags: [uns-admin-perm] + + dump: + pre: wg-get-device-start + post: wg-get-device-done + request: + attributes: + - ifindex + - ifname + reply: &all-attrs + attributes: + - ifindex + - ifname + - private-key + - public-key + - flags + - listen-port + - fwmark + - peers + - + name: set-device + value: 1 + doc: | + Set WireGuard device + ~~~~~~~~~~~~~~~~~~~~ + + This command should be called with a wgdevice set, containing one but + not both of ``WGDEVICE_A_IFINDEX`` and ``WGDEVICE_A_IFNAME``. + + It is possible that the amount of configuration data exceeds that of the + maximum message length accepted by the kernel. In that case, several + messages should be sent one after another, with each successive one + filling in information not contained in the prior. Note that if + ``WGDEVICE_F_REPLACE_PEERS`` is specified in the first message, it + probably should not be specified in fragments that come after, so that + the list of peers is only cleared the first time but appended after. + Likewise for peers, if ``WGPEER_F_REPLACE_ALLOWEDIPS`` is specified in + the first message of a peer, it likely should not be specified in + subsequent fragments. + + If an error occurs, ``NLMSG_ERROR`` will reply containing an errno. + attribute-set: wgdevice + flags: [uns-admin-perm] + + do: + request: *all-attrs diff --git a/MAINTAINERS b/MAINTAINERS index 09932ab7e0e8..8b44a380642c 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -27673,6 +27673,7 @@ M: Jason A. Donenfeld L: wireguard@lists.zx2c4.com L: netdev@vger.kernel.org S: Maintained +F: Documentation/netlink/specs/wireguard.yaml F: drivers/net/wireguard/ F: tools/testing/selftests/wireguard/ diff --git a/include/uapi/linux/wireguard.h b/include/uapi/linux/wireguard.h index 8c26391196d5..dee4401e0b5d 100644 --- a/include/uapi/linux/wireguard.h +++ b/include/uapi/linux/wireguard.h @@ -1,135 +1,6 @@ /* SPDX-License-Identifier: (GPL-2.0 WITH Linux-syscall-note) OR MIT */ /* * Copyright (C) 2015-2019 Jason A. Donenfeld . All Rights Reserved. - * - * Documentation - * ============= - * - * The below enums and macros are for interfacing with WireGuard, using generic - * netlink, with family WG_GENL_NAME and version WG_GENL_VERSION. It defines two - * methods: get and set. Note that while they share many common attributes, - * these two functions actually accept a slightly different set of inputs and - * outputs. - * - * WG_CMD_GET_DEVICE - * ----------------- - * - * May only be called via NLM_F_REQUEST | NLM_F_DUMP. The command should contain - * one but not both of: - * - * WGDEVICE_A_IFINDEX: NLA_U32 - * WGDEVICE_A_IFNAME: NLA_NUL_STRING, maxlen IFNAMSIZ - 1 - * - * The kernel will then return several messages (NLM_F_MULTI) containing the - * following tree of nested items: - * - * WGDEVICE_A_IFINDEX: NLA_U32 - * WGDEVICE_A_IFNAME: NLA_NUL_STRING, maxlen IFNAMSIZ - 1 - * WGDEVICE_A_PRIVATE_KEY: NLA_EXACT_LEN, len WG_KEY_LEN - * WGDEVICE_A_PUBLIC_KEY: NLA_EXACT_LEN, len WG_KEY_LEN - * WGDEVICE_A_LISTEN_PORT: NLA_U16 - * WGDEVICE_A_FWMARK: NLA_U32 - * WGDEVICE_A_PEERS: NLA_NESTED - * 0: NLA_NESTED - * WGPEER_A_PUBLIC_KEY: NLA_EXACT_LEN, len WG_KEY_LEN - * WGPEER_A_PRESHARED_KEY: NLA_EXACT_LEN, len WG_KEY_LEN - * WGPEER_A_ENDPOINT: NLA_MIN_LEN(struct sockaddr), struct sockaddr_in or struct sockaddr_in6 - * WGPEER_A_PERSISTENT_KEEPALIVE_INTERVAL: NLA_U16 - * WGPEER_A_LAST_HANDSHAKE_TIME: NLA_EXACT_LEN, struct __kernel_timespec - * WGPEER_A_RX_BYTES: NLA_U64 - * WGPEER_A_TX_BYTES: NLA_U64 - * WGPEER_A_ALLOWEDIPS: NLA_NESTED - * 0: NLA_NESTED - * WGALLOWEDIP_A_FAMILY: NLA_U16 - * WGALLOWEDIP_A_IPADDR: NLA_MIN_LEN(struct in_addr), struct in_addr or struct in6_addr - * WGALLOWEDIP_A_CIDR_MASK: NLA_U8 - * 0: NLA_NESTED - * ... - * 0: NLA_NESTED - * ... - * ... - * WGPEER_A_PROTOCOL_VERSION: NLA_U32 - * 0: NLA_NESTED - * ... - * ... - * - * It is possible that all of the allowed IPs of a single peer will not - * fit within a single netlink message. In that case, the same peer will - * be written in the following message, except it will only contain - * WGPEER_A_PUBLIC_KEY and WGPEER_A_ALLOWEDIPS. This may occur several - * times in a row for the same peer. It is then up to the receiver to - * coalesce adjacent peers. Likewise, it is possible that all peers will - * not fit within a single message. So, subsequent peers will be sent - * in following messages, except those will only contain WGDEVICE_A_IFNAME - * and WGDEVICE_A_PEERS. It is then up to the receiver to coalesce these - * messages to form the complete list of peers. - * - * Since this is an NLA_F_DUMP command, the final message will always be - * NLMSG_DONE, even if an error occurs. However, this NLMSG_DONE message - * contains an integer error code. It is either zero or a negative error - * code corresponding to the errno. - * - * WG_CMD_SET_DEVICE - * ----------------- - * - * May only be called via NLM_F_REQUEST. The command should contain the - * following tree of nested items, containing one but not both of - * WGDEVICE_A_IFINDEX and WGDEVICE_A_IFNAME: - * - * WGDEVICE_A_IFINDEX: NLA_U32 - * WGDEVICE_A_IFNAME: NLA_NUL_STRING, maxlen IFNAMSIZ - 1 - * WGDEVICE_A_FLAGS: NLA_U32, 0 or WGDEVICE_F_REPLACE_PEERS if all current - * peers should be removed prior to adding the list below. - * WGDEVICE_A_PRIVATE_KEY: len WG_KEY_LEN, all zeros to remove - * WGDEVICE_A_LISTEN_PORT: NLA_U16, 0 to choose randomly - * WGDEVICE_A_FWMARK: NLA_U32, 0 to disable - * WGDEVICE_A_PEERS: NLA_NESTED - * 0: NLA_NESTED - * WGPEER_A_PUBLIC_KEY: len WG_KEY_LEN - * WGPEER_A_FLAGS: NLA_U32, 0 and/or WGPEER_F_REMOVE_ME if the - * specified peer should not exist at the end of the - * operation, rather than added/updated and/or - * WGPEER_F_REPLACE_ALLOWEDIPS if all current allowed - * IPs of this peer should be removed prior to adding - * the list below and/or WGPEER_F_UPDATE_ONLY if the - * peer should only be set if it already exists. - * WGPEER_A_PRESHARED_KEY: len WG_KEY_LEN, all zeros to remove - * WGPEER_A_ENDPOINT: struct sockaddr_in or struct sockaddr_in6 - * WGPEER_A_PERSISTENT_KEEPALIVE_INTERVAL: NLA_U16, 0 to disable - * WGPEER_A_ALLOWEDIPS: NLA_NESTED - * 0: NLA_NESTED - * WGALLOWEDIP_A_FAMILY: NLA_U16 - * WGALLOWEDIP_A_IPADDR: struct in_addr or struct in6_addr - * WGALLOWEDIP_A_CIDR_MASK: NLA_U8 - * WGALLOWEDIP_A_FLAGS: NLA_U32, WGALLOWEDIP_F_REMOVE_ME if - * the specified IP should be removed; - * otherwise, this IP will be added if - * it is not already present. - * 0: NLA_NESTED - * ... - * 0: NLA_NESTED - * ... - * ... - * WGPEER_A_PROTOCOL_VERSION: NLA_U32, should not be set or used at - * all by most users of this API, as the - * most recent protocol will be used when - * this is unset. Otherwise, must be set - * to 1. - * 0: NLA_NESTED - * ... - * ... - * - * It is possible that the amount of configuration data exceeds that of - * the maximum message length accepted by the kernel. In that case, several - * messages should be sent one after another, with each successive one - * filling in information not contained in the prior. Note that if - * WGDEVICE_F_REPLACE_PEERS is specified in the first message, it probably - * should not be specified in fragments that come after, so that the list - * of peers is only cleared the first time but appended after. Likewise for - * peers, if WGPEER_F_REPLACE_ALLOWEDIPS is specified in the first message - * of a peer, it likely should not be specified in subsequent fragments. - * - * If an error occurs, NLMSG_ERROR will reply containing an errno. */ #ifndef _WG_UAPI_WIREGUARD_H From b5c5a82bf5cb96e14a6627ef21be962052a0c6d8 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Asbj=C3=B8rn=20Sloth=20T=C3=B8nnesen?= Date: Wed, 26 Nov 2025 17:35:38 +0000 Subject: [PATCH 1372/1381] wireguard: uapi: move enum wg_cmd MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This patch moves enum wg_cmd to the end of the file, where ynl-gen would generate it. This is an incremental step towards adopting an UAPI header generated by ynl-gen. This is split out to keep the patches readable. This is a trivial patch with no behavioural changes intended. Signed-off-by: Asbjørn Sloth Tønnesen Signed-off-by: Jason A. Donenfeld --- include/uapi/linux/wireguard.h | 15 ++++++++------- 1 file changed, 8 insertions(+), 7 deletions(-) diff --git a/include/uapi/linux/wireguard.h b/include/uapi/linux/wireguard.h index dee4401e0b5d..3ebfffd61269 100644 --- a/include/uapi/linux/wireguard.h +++ b/include/uapi/linux/wireguard.h @@ -11,13 +11,6 @@ #define WG_KEY_LEN 32 -enum wg_cmd { - WG_CMD_GET_DEVICE, - WG_CMD_SET_DEVICE, - __WG_CMD_MAX -}; -#define WG_CMD_MAX (__WG_CMD_MAX - 1) - enum wgdevice_flag { WGDEVICE_F_REPLACE_PEERS = 1U << 0, __WGDEVICE_F_ALL = WGDEVICE_F_REPLACE_PEERS @@ -73,4 +66,12 @@ enum wgallowedip_attribute { }; #define WGALLOWEDIP_A_MAX (__WGALLOWEDIP_A_LAST - 1) +enum wg_cmd { + WG_CMD_GET_DEVICE, + WG_CMD_SET_DEVICE, + + __WG_CMD_MAX +}; +#define WG_CMD_MAX (__WG_CMD_MAX - 1) + #endif /* _WG_UAPI_WIREGUARD_H */ From 8d974872ab29eeb93a5b0b698007257d8be07968 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Asbj=C3=B8rn=20Sloth=20T=C3=B8nnesen?= Date: Wed, 26 Nov 2025 17:35:39 +0000 Subject: [PATCH 1373/1381] wireguard: uapi: move flag enums MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Move the wg*_flag enums, so they are defined above the attribute set enums, where ynl-gen would place them. This is an incremental step towards adopting an UAPI header generated by ynl-gen. This is split out to keep the patches readable. This is a trivial patch with no behavioural changes intended. Signed-off-by: Asbjørn Sloth Tønnesen Signed-off-by: Jason A. Donenfeld --- include/uapi/linux/wireguard.h | 25 ++++++++++++++----------- 1 file changed, 14 insertions(+), 11 deletions(-) diff --git a/include/uapi/linux/wireguard.h b/include/uapi/linux/wireguard.h index 3ebfffd61269..a2815f4f2910 100644 --- a/include/uapi/linux/wireguard.h +++ b/include/uapi/linux/wireguard.h @@ -15,6 +15,20 @@ enum wgdevice_flag { WGDEVICE_F_REPLACE_PEERS = 1U << 0, __WGDEVICE_F_ALL = WGDEVICE_F_REPLACE_PEERS }; + +enum wgpeer_flag { + WGPEER_F_REMOVE_ME = 1U << 0, + WGPEER_F_REPLACE_ALLOWEDIPS = 1U << 1, + WGPEER_F_UPDATE_ONLY = 1U << 2, + __WGPEER_F_ALL = WGPEER_F_REMOVE_ME | WGPEER_F_REPLACE_ALLOWEDIPS | + WGPEER_F_UPDATE_ONLY +}; + +enum wgallowedip_flag { + WGALLOWEDIP_F_REMOVE_ME = 1U << 0, + __WGALLOWEDIP_F_ALL = WGALLOWEDIP_F_REMOVE_ME +}; + enum wgdevice_attribute { WGDEVICE_A_UNSPEC, WGDEVICE_A_IFINDEX, @@ -29,13 +43,6 @@ enum wgdevice_attribute { }; #define WGDEVICE_A_MAX (__WGDEVICE_A_LAST - 1) -enum wgpeer_flag { - WGPEER_F_REMOVE_ME = 1U << 0, - WGPEER_F_REPLACE_ALLOWEDIPS = 1U << 1, - WGPEER_F_UPDATE_ONLY = 1U << 2, - __WGPEER_F_ALL = WGPEER_F_REMOVE_ME | WGPEER_F_REPLACE_ALLOWEDIPS | - WGPEER_F_UPDATE_ONLY -}; enum wgpeer_attribute { WGPEER_A_UNSPEC, WGPEER_A_PUBLIC_KEY, @@ -52,10 +59,6 @@ enum wgpeer_attribute { }; #define WGPEER_A_MAX (__WGPEER_A_LAST - 1) -enum wgallowedip_flag { - WGALLOWEDIP_F_REMOVE_ME = 1U << 0, - __WGALLOWEDIP_F_ALL = WGALLOWEDIP_F_REMOVE_ME -}; enum wgallowedip_attribute { WGALLOWEDIP_A_UNSPEC, WGALLOWEDIP_A_FAMILY, From 88cedad45ba14097e06d2c9f6578688097a94691 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Asbj=C3=B8rn=20Sloth=20T=C3=B8nnesen?= Date: Wed, 26 Nov 2025 17:35:40 +0000 Subject: [PATCH 1374/1381] wireguard: uapi: generate header with ynl-gen MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Use ynl-gen to generate the UAPI header for WireGuard. The cosmetic changes in this patch confirms that the spec is aligned with the implementation. By using the generated version, it ensures that they stay in sync. Changes in the generated header: * Trivial header guard rename. * Trivial white space changes. * Trivial comment changes. * Precompute bitflags in ynl-gen (see [1]). * Drop __*_F_ALL constants (see [1]). [1] https://lore.kernel.org/r/20251014123201.6ecfd146@kernel.org/ No behavioural changes intended. Signed-off-by: Asbjørn Sloth Tønnesen Signed-off-by: Jason A. Donenfeld --- drivers/net/wireguard/netlink.c | 6 +++--- include/uapi/linux/wireguard.h | 38 ++++++++++++++++----------------- 2 files changed, 22 insertions(+), 22 deletions(-) diff --git a/drivers/net/wireguard/netlink.c b/drivers/net/wireguard/netlink.c index c2d0576e96f5..0ce0bda8c1ce 100644 --- a/drivers/net/wireguard/netlink.c +++ b/drivers/net/wireguard/netlink.c @@ -26,7 +26,7 @@ static const struct nla_policy device_policy[WGDEVICE_A_MAX + 1] = { [WGDEVICE_A_IFNAME] = { .type = NLA_NUL_STRING, .len = IFNAMSIZ - 1 }, [WGDEVICE_A_PRIVATE_KEY] = NLA_POLICY_EXACT_LEN(WG_KEY_LEN), [WGDEVICE_A_PUBLIC_KEY] = NLA_POLICY_EXACT_LEN(WG_KEY_LEN), - [WGDEVICE_A_FLAGS] = NLA_POLICY_MASK(NLA_U32, __WGDEVICE_F_ALL), + [WGDEVICE_A_FLAGS] = NLA_POLICY_MASK(NLA_U32, 0x1), [WGDEVICE_A_LISTEN_PORT] = { .type = NLA_U16 }, [WGDEVICE_A_FWMARK] = { .type = NLA_U32 }, [WGDEVICE_A_PEERS] = NLA_POLICY_NESTED_ARRAY(peer_policy), @@ -35,7 +35,7 @@ static const struct nla_policy device_policy[WGDEVICE_A_MAX + 1] = { static const struct nla_policy peer_policy[WGPEER_A_MAX + 1] = { [WGPEER_A_PUBLIC_KEY] = NLA_POLICY_EXACT_LEN(WG_KEY_LEN), [WGPEER_A_PRESHARED_KEY] = NLA_POLICY_EXACT_LEN(WG_KEY_LEN), - [WGPEER_A_FLAGS] = NLA_POLICY_MASK(NLA_U32, __WGPEER_F_ALL), + [WGPEER_A_FLAGS] = NLA_POLICY_MASK(NLA_U32, 0x7), [WGPEER_A_ENDPOINT] = NLA_POLICY_MIN_LEN(sizeof(struct sockaddr)), [WGPEER_A_PERSISTENT_KEEPALIVE_INTERVAL] = { .type = NLA_U16 }, [WGPEER_A_LAST_HANDSHAKE_TIME] = NLA_POLICY_EXACT_LEN(sizeof(struct __kernel_timespec)), @@ -49,7 +49,7 @@ static const struct nla_policy allowedip_policy[WGALLOWEDIP_A_MAX + 1] = { [WGALLOWEDIP_A_FAMILY] = { .type = NLA_U16 }, [WGALLOWEDIP_A_IPADDR] = NLA_POLICY_MIN_LEN(sizeof(struct in_addr)), [WGALLOWEDIP_A_CIDR_MASK] = { .type = NLA_U8 }, - [WGALLOWEDIP_A_FLAGS] = NLA_POLICY_MASK(NLA_U32, __WGALLOWEDIP_F_ALL), + [WGALLOWEDIP_A_FLAGS] = NLA_POLICY_MASK(NLA_U32, 0x1), }; static struct wg_device *lookup_interface(struct nlattr **attrs, diff --git a/include/uapi/linux/wireguard.h b/include/uapi/linux/wireguard.h index a2815f4f2910..a100b9715b08 100644 --- a/include/uapi/linux/wireguard.h +++ b/include/uapi/linux/wireguard.h @@ -1,32 +1,29 @@ -/* SPDX-License-Identifier: (GPL-2.0 WITH Linux-syscall-note) OR MIT */ -/* - * Copyright (C) 2015-2019 Jason A. Donenfeld . All Rights Reserved. - */ +/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause) */ +/* Do not edit directly, auto-generated from: */ +/* Documentation/netlink/specs/wireguard.yaml */ +/* YNL-GEN uapi header */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ -#ifndef _WG_UAPI_WIREGUARD_H -#define _WG_UAPI_WIREGUARD_H +#ifndef _UAPI_LINUX_WIREGUARD_H +#define _UAPI_LINUX_WIREGUARD_H -#define WG_GENL_NAME "wireguard" -#define WG_GENL_VERSION 1 +#define WG_GENL_NAME "wireguard" +#define WG_GENL_VERSION 1 -#define WG_KEY_LEN 32 +#define WG_KEY_LEN 32 enum wgdevice_flag { - WGDEVICE_F_REPLACE_PEERS = 1U << 0, - __WGDEVICE_F_ALL = WGDEVICE_F_REPLACE_PEERS + WGDEVICE_F_REPLACE_PEERS = 1, }; enum wgpeer_flag { - WGPEER_F_REMOVE_ME = 1U << 0, - WGPEER_F_REPLACE_ALLOWEDIPS = 1U << 1, - WGPEER_F_UPDATE_ONLY = 1U << 2, - __WGPEER_F_ALL = WGPEER_F_REMOVE_ME | WGPEER_F_REPLACE_ALLOWEDIPS | - WGPEER_F_UPDATE_ONLY + WGPEER_F_REMOVE_ME = 1, + WGPEER_F_REPLACE_ALLOWEDIPS = 2, + WGPEER_F_UPDATE_ONLY = 4, }; enum wgallowedip_flag { - WGALLOWEDIP_F_REMOVE_ME = 1U << 0, - __WGALLOWEDIP_F_ALL = WGALLOWEDIP_F_REMOVE_ME + WGALLOWEDIP_F_REMOVE_ME = 1, }; enum wgdevice_attribute { @@ -39,6 +36,7 @@ enum wgdevice_attribute { WGDEVICE_A_LISTEN_PORT, WGDEVICE_A_FWMARK, WGDEVICE_A_PEERS, + __WGDEVICE_A_LAST }; #define WGDEVICE_A_MAX (__WGDEVICE_A_LAST - 1) @@ -55,6 +53,7 @@ enum wgpeer_attribute { WGPEER_A_TX_BYTES, WGPEER_A_ALLOWEDIPS, WGPEER_A_PROTOCOL_VERSION, + __WGPEER_A_LAST }; #define WGPEER_A_MAX (__WGPEER_A_LAST - 1) @@ -65,6 +64,7 @@ enum wgallowedip_attribute { WGALLOWEDIP_A_IPADDR, WGALLOWEDIP_A_CIDR_MASK, WGALLOWEDIP_A_FLAGS, + __WGALLOWEDIP_A_LAST }; #define WGALLOWEDIP_A_MAX (__WGALLOWEDIP_A_LAST - 1) @@ -77,4 +77,4 @@ enum wg_cmd { }; #define WG_CMD_MAX (__WG_CMD_MAX - 1) -#endif /* _WG_UAPI_WIREGUARD_H */ +#endif /* _UAPI_LINUX_WIREGUARD_H */ From 3fd2f3d2f4259df19eec3ea5a188d7c50a37e216 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Asbj=C3=B8rn=20Sloth=20T=C3=B8nnesen?= Date: Wed, 26 Nov 2025 17:35:42 +0000 Subject: [PATCH 1375/1381] wireguard: netlink: generate netlink code MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This patch adopts netlink policies and command definitions generated by ynl-gen, thus completing the conversion to YNL. Given that the old and new policies are functionally identical and have just been moved to a new file, it serves to verify that the policies generated from the spec are identical to the previous policy code. The following functions are renamed: wg_get_device_dump() -> wg_get_device_dumpit() wg_set_device() -> wg_set_device_doit() The new files are covered by the existing drivers/net/wireguard/ pattern in MAINTAINERS. No behavioural changes intended. Signed-off-by: Asbjørn Sloth Tønnesen Signed-off-by: Jason A. Donenfeld --- drivers/net/wireguard/Makefile | 2 +- drivers/net/wireguard/generated/netlink.c | 73 +++++++++++++++++++++++ drivers/net/wireguard/generated/netlink.h | 30 ++++++++++ drivers/net/wireguard/netlink.c | 60 ++----------------- 4 files changed, 109 insertions(+), 56 deletions(-) create mode 100644 drivers/net/wireguard/generated/netlink.c create mode 100644 drivers/net/wireguard/generated/netlink.h diff --git a/drivers/net/wireguard/Makefile b/drivers/net/wireguard/Makefile index dbe1f8514efc..00cbcc9ab69d 100644 --- a/drivers/net/wireguard/Makefile +++ b/drivers/net/wireguard/Makefile @@ -13,5 +13,5 @@ wireguard-y += peerlookup.o wireguard-y += allowedips.o wireguard-y += ratelimiter.o wireguard-y += cookie.o -wireguard-y += netlink.o +wireguard-y += netlink.o generated/netlink.o obj-$(CONFIG_WIREGUARD) := wireguard.o diff --git a/drivers/net/wireguard/generated/netlink.c b/drivers/net/wireguard/generated/netlink.c new file mode 100644 index 000000000000..3ef8c29908c2 --- /dev/null +++ b/drivers/net/wireguard/generated/netlink.c @@ -0,0 +1,73 @@ +// SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause) +/* Do not edit directly, auto-generated from: */ +/* Documentation/netlink/specs/wireguard.yaml */ +/* YNL-GEN kernel source */ +/* YNL-ARG --function-prefix wg */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ + +#include +#include + +#include "netlink.h" + +#include +#include + +/* Common nested types */ +const struct nla_policy wireguard_wgallowedip_nl_policy[WGALLOWEDIP_A_FLAGS + 1] = { + [WGALLOWEDIP_A_FAMILY] = { .type = NLA_U16, }, + [WGALLOWEDIP_A_IPADDR] = NLA_POLICY_MIN_LEN(4), + [WGALLOWEDIP_A_CIDR_MASK] = { .type = NLA_U8, }, + [WGALLOWEDIP_A_FLAGS] = NLA_POLICY_MASK(NLA_U32, 0x1), +}; + +const struct nla_policy wireguard_wgpeer_nl_policy[WGPEER_A_PROTOCOL_VERSION + 1] = { + [WGPEER_A_PUBLIC_KEY] = NLA_POLICY_EXACT_LEN(WG_KEY_LEN), + [WGPEER_A_PRESHARED_KEY] = NLA_POLICY_EXACT_LEN(WG_KEY_LEN), + [WGPEER_A_FLAGS] = NLA_POLICY_MASK(NLA_U32, 0x7), + [WGPEER_A_ENDPOINT] = NLA_POLICY_MIN_LEN(16), + [WGPEER_A_PERSISTENT_KEEPALIVE_INTERVAL] = { .type = NLA_U16, }, + [WGPEER_A_LAST_HANDSHAKE_TIME] = NLA_POLICY_EXACT_LEN(16), + [WGPEER_A_RX_BYTES] = { .type = NLA_U64, }, + [WGPEER_A_TX_BYTES] = { .type = NLA_U64, }, + [WGPEER_A_ALLOWEDIPS] = NLA_POLICY_NESTED_ARRAY(wireguard_wgallowedip_nl_policy), + [WGPEER_A_PROTOCOL_VERSION] = { .type = NLA_U32, }, +}; + +/* WG_CMD_GET_DEVICE - dump */ +static const struct nla_policy wireguard_get_device_nl_policy[WGDEVICE_A_IFNAME + 1] = { + [WGDEVICE_A_IFINDEX] = { .type = NLA_U32, }, + [WGDEVICE_A_IFNAME] = { .type = NLA_NUL_STRING, .len = 15, }, +}; + +/* WG_CMD_SET_DEVICE - do */ +static const struct nla_policy wireguard_set_device_nl_policy[WGDEVICE_A_PEERS + 1] = { + [WGDEVICE_A_IFINDEX] = { .type = NLA_U32, }, + [WGDEVICE_A_IFNAME] = { .type = NLA_NUL_STRING, .len = 15, }, + [WGDEVICE_A_PRIVATE_KEY] = NLA_POLICY_EXACT_LEN(WG_KEY_LEN), + [WGDEVICE_A_PUBLIC_KEY] = NLA_POLICY_EXACT_LEN(WG_KEY_LEN), + [WGDEVICE_A_FLAGS] = NLA_POLICY_MASK(NLA_U32, 0x1), + [WGDEVICE_A_LISTEN_PORT] = { .type = NLA_U16, }, + [WGDEVICE_A_FWMARK] = { .type = NLA_U32, }, + [WGDEVICE_A_PEERS] = NLA_POLICY_NESTED_ARRAY(wireguard_wgpeer_nl_policy), +}; + +/* Ops table for wireguard */ +const struct genl_split_ops wireguard_nl_ops[2] = { + { + .cmd = WG_CMD_GET_DEVICE, + .start = wg_get_device_start, + .dumpit = wg_get_device_dumpit, + .done = wg_get_device_done, + .policy = wireguard_get_device_nl_policy, + .maxattr = WGDEVICE_A_IFNAME, + .flags = GENL_UNS_ADMIN_PERM | GENL_CMD_CAP_DUMP, + }, + { + .cmd = WG_CMD_SET_DEVICE, + .doit = wg_set_device_doit, + .policy = wireguard_set_device_nl_policy, + .maxattr = WGDEVICE_A_PEERS, + .flags = GENL_UNS_ADMIN_PERM | GENL_CMD_CAP_DO, + }, +}; diff --git a/drivers/net/wireguard/generated/netlink.h b/drivers/net/wireguard/generated/netlink.h new file mode 100644 index 000000000000..5dc977ee9e7c --- /dev/null +++ b/drivers/net/wireguard/generated/netlink.h @@ -0,0 +1,30 @@ +/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause) */ +/* Do not edit directly, auto-generated from: */ +/* Documentation/netlink/specs/wireguard.yaml */ +/* YNL-GEN kernel header */ +/* YNL-ARG --function-prefix wg */ +/* To regenerate run: tools/net/ynl/ynl-regen.sh */ + +#ifndef _LINUX_WIREGUARD_GEN_H +#define _LINUX_WIREGUARD_GEN_H + +#include +#include + +#include +#include + +/* Common nested types */ +extern const struct nla_policy wireguard_wgallowedip_nl_policy[WGALLOWEDIP_A_FLAGS + 1]; +extern const struct nla_policy wireguard_wgpeer_nl_policy[WGPEER_A_PROTOCOL_VERSION + 1]; + +/* Ops table for wireguard */ +extern const struct genl_split_ops wireguard_nl_ops[2]; + +int wg_get_device_start(struct netlink_callback *cb); +int wg_get_device_done(struct netlink_callback *cb); + +int wg_get_device_dumpit(struct sk_buff *skb, struct netlink_callback *cb); +int wg_set_device_doit(struct sk_buff *skb, struct genl_info *info); + +#endif /* _LINUX_WIREGUARD_GEN_H */ diff --git a/drivers/net/wireguard/netlink.c b/drivers/net/wireguard/netlink.c index 0ce0bda8c1ce..1da7e98d0d50 100644 --- a/drivers/net/wireguard/netlink.c +++ b/drivers/net/wireguard/netlink.c @@ -9,6 +9,7 @@ #include "socket.h" #include "queueing.h" #include "messages.h" +#include "generated/netlink.h" #include @@ -18,39 +19,6 @@ #include static struct genl_family genl_family; -static const struct nla_policy peer_policy[WGPEER_A_MAX + 1]; -static const struct nla_policy allowedip_policy[WGALLOWEDIP_A_MAX + 1]; - -static const struct nla_policy device_policy[WGDEVICE_A_MAX + 1] = { - [WGDEVICE_A_IFINDEX] = { .type = NLA_U32 }, - [WGDEVICE_A_IFNAME] = { .type = NLA_NUL_STRING, .len = IFNAMSIZ - 1 }, - [WGDEVICE_A_PRIVATE_KEY] = NLA_POLICY_EXACT_LEN(WG_KEY_LEN), - [WGDEVICE_A_PUBLIC_KEY] = NLA_POLICY_EXACT_LEN(WG_KEY_LEN), - [WGDEVICE_A_FLAGS] = NLA_POLICY_MASK(NLA_U32, 0x1), - [WGDEVICE_A_LISTEN_PORT] = { .type = NLA_U16 }, - [WGDEVICE_A_FWMARK] = { .type = NLA_U32 }, - [WGDEVICE_A_PEERS] = NLA_POLICY_NESTED_ARRAY(peer_policy), -}; - -static const struct nla_policy peer_policy[WGPEER_A_MAX + 1] = { - [WGPEER_A_PUBLIC_KEY] = NLA_POLICY_EXACT_LEN(WG_KEY_LEN), - [WGPEER_A_PRESHARED_KEY] = NLA_POLICY_EXACT_LEN(WG_KEY_LEN), - [WGPEER_A_FLAGS] = NLA_POLICY_MASK(NLA_U32, 0x7), - [WGPEER_A_ENDPOINT] = NLA_POLICY_MIN_LEN(sizeof(struct sockaddr)), - [WGPEER_A_PERSISTENT_KEEPALIVE_INTERVAL] = { .type = NLA_U16 }, - [WGPEER_A_LAST_HANDSHAKE_TIME] = NLA_POLICY_EXACT_LEN(sizeof(struct __kernel_timespec)), - [WGPEER_A_RX_BYTES] = { .type = NLA_U64 }, - [WGPEER_A_TX_BYTES] = { .type = NLA_U64 }, - [WGPEER_A_ALLOWEDIPS] = NLA_POLICY_NESTED_ARRAY(allowedip_policy), - [WGPEER_A_PROTOCOL_VERSION] = { .type = NLA_U32 } -}; - -static const struct nla_policy allowedip_policy[WGALLOWEDIP_A_MAX + 1] = { - [WGALLOWEDIP_A_FAMILY] = { .type = NLA_U16 }, - [WGALLOWEDIP_A_IPADDR] = NLA_POLICY_MIN_LEN(sizeof(struct in_addr)), - [WGALLOWEDIP_A_CIDR_MASK] = { .type = NLA_U8 }, - [WGALLOWEDIP_A_FLAGS] = NLA_POLICY_MASK(NLA_U32, 0x1), -}; static struct wg_device *lookup_interface(struct nlattr **attrs, struct sk_buff *skb) @@ -199,7 +167,7 @@ get_peer(struct wg_peer *peer, struct sk_buff *skb, struct dump_ctx *ctx) return -EMSGSIZE; } -static int wg_get_device_start(struct netlink_callback *cb) +int wg_get_device_start(struct netlink_callback *cb) { struct wg_device *wg; @@ -210,7 +178,7 @@ static int wg_get_device_start(struct netlink_callback *cb) return 0; } -static int wg_get_device_dump(struct sk_buff *skb, struct netlink_callback *cb) +int wg_get_device_dumpit(struct sk_buff *skb, struct netlink_callback *cb) { struct wg_peer *peer, *next_peer_cursor; struct dump_ctx *ctx = DUMP_CTX(cb); @@ -304,7 +272,7 @@ static int wg_get_device_dump(struct sk_buff *skb, struct netlink_callback *cb) */ } -static int wg_get_device_done(struct netlink_callback *cb) +int wg_get_device_done(struct netlink_callback *cb) { struct dump_ctx *ctx = DUMP_CTX(cb); @@ -502,7 +470,7 @@ static int set_peer(struct wg_device *wg, struct nlattr **attrs) return ret; } -static int wg_set_device(struct sk_buff *skb, struct genl_info *info) +int wg_set_device_doit(struct sk_buff *skb, struct genl_info *info) { struct wg_device *wg = lookup_interface(info->attrs, skb); u32 flags = 0; @@ -616,24 +584,6 @@ static int wg_set_device(struct sk_buff *skb, struct genl_info *info) return ret; } -static const struct genl_split_ops wireguard_nl_ops[] = { - { - .cmd = WG_CMD_GET_DEVICE, - .start = wg_get_device_start, - .dumpit = wg_get_device_dump, - .done = wg_get_device_done, - .policy = device_policy, - .maxattr = WGDEVICE_A_IFNAME, - .flags = GENL_UNS_ADMIN_PERM | GENL_CMD_CAP_DUMP, - }, { - .cmd = WG_CMD_SET_DEVICE, - .doit = wg_set_device, - .policy = device_policy, - .maxattr = WGDEVICE_A_PEERS, - .flags = GENL_UNS_ADMIN_PERM | GENL_CMD_CAP_DO, - } -}; - static struct genl_family genl_family __ro_after_init = { .split_ops = wireguard_nl_ops, .n_split_ops = ARRAY_SIZE(wireguard_nl_ops), From 9fefc78f7f02d71810776fdeb119a05a946a27cc Mon Sep 17 00:00:00 2001 From: Xiang Mei Date: Thu, 27 Nov 2025 17:14:14 -0700 Subject: [PATCH 1376/1381] net/sched: sch_cake: Fix incorrect qlen reduction in cake_drop MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit In cake_drop(), qdisc_tree_reduce_backlog() is used to update the qlen and backlog of the qdisc hierarchy. Its caller, cake_enqueue(), assumes that the parent qdisc will enqueue the current packet. However, this assumption breaks when cake_enqueue() returns NET_XMIT_CN: the parent qdisc stops enqueuing current packet, leaving the tree qlen/backlog accounting inconsistent. This mismatch can lead to a NULL dereference (e.g., when the parent Qdisc is qfq_qdisc). This patch computes the qlen/backlog delta in a more robust way by observing the difference before and after the series of cake_drop() calls, and then compensates the qdisc tree accounting if cake_enqueue() returns NET_XMIT_CN. To ensure correct compensation when ACK thinning is enabled, a new variable is introduced to keep qlen unchanged. Fixes: 15de71d06a40 ("net/sched: Make cake_enqueue return NET_XMIT_CN when past buffer_limit") Signed-off-by: Xiang Mei Reviewed-by: Toke Høiland-Jørgensen Link: https://patch.msgid.link/20251128001415.377823-1-xmei5@asu.edu Signed-off-by: Paolo Abeni --- net/sched/sch_cake.c | 60 ++++++++++++++++++++++++-------------------- 1 file changed, 33 insertions(+), 27 deletions(-) diff --git a/net/sched/sch_cake.c b/net/sched/sch_cake.c index 32bacfc314c2..d325a90cde9e 100644 --- a/net/sched/sch_cake.c +++ b/net/sched/sch_cake.c @@ -1597,7 +1597,6 @@ static unsigned int cake_drop(struct Qdisc *sch, struct sk_buff **to_free) qdisc_drop_reason(skb, sch, to_free, SKB_DROP_REASON_QDISC_OVERLIMIT); sch->q.qlen--; - qdisc_tree_reduce_backlog(sch, 1, len); cake_heapify(q, 0); @@ -1743,14 +1742,14 @@ static void cake_reconfigure(struct Qdisc *sch); static s32 cake_enqueue(struct sk_buff *skb, struct Qdisc *sch, struct sk_buff **to_free) { + u32 idx, tin, prev_qlen, prev_backlog, drop_id; struct cake_sched_data *q = qdisc_priv(sch); - int len = qdisc_pkt_len(skb); - int ret; + int len = qdisc_pkt_len(skb), ret; struct sk_buff *ack = NULL; ktime_t now = ktime_get(); struct cake_tin_data *b; struct cake_flow *flow; - u32 idx, tin; + bool same_flow = false; /* choose flow to insert into */ idx = cake_classify(sch, &b, skb, q->flow_mode, &ret); @@ -1823,6 +1822,8 @@ static s32 cake_enqueue(struct sk_buff *skb, struct Qdisc *sch, consume_skb(skb); } else { /* not splitting */ + int ack_pkt_len = 0; + cobalt_set_enqueue_time(skb, now); get_cobalt_cb(skb)->adjusted_len = cake_overhead(q, skb); flow_queue_add(flow, skb); @@ -1833,13 +1834,13 @@ static s32 cake_enqueue(struct sk_buff *skb, struct Qdisc *sch, if (ack) { b->ack_drops++; sch->qstats.drops++; - b->bytes += qdisc_pkt_len(ack); - len -= qdisc_pkt_len(ack); + ack_pkt_len = qdisc_pkt_len(ack); + b->bytes += ack_pkt_len; q->buffer_used += skb->truesize - ack->truesize; if (q->rate_flags & CAKE_FLAG_INGRESS) cake_advance_shaper(q, b, ack, now, true); - qdisc_tree_reduce_backlog(sch, 1, qdisc_pkt_len(ack)); + qdisc_tree_reduce_backlog(sch, 1, ack_pkt_len); consume_skb(ack); } else { sch->q.qlen++; @@ -1848,11 +1849,11 @@ static s32 cake_enqueue(struct sk_buff *skb, struct Qdisc *sch, /* stats */ b->packets++; - b->bytes += len; - b->backlogs[idx] += len; - b->tin_backlog += len; - sch->qstats.backlog += len; - q->avg_window_bytes += len; + b->bytes += len - ack_pkt_len; + b->backlogs[idx] += len - ack_pkt_len; + b->tin_backlog += len - ack_pkt_len; + sch->qstats.backlog += len - ack_pkt_len; + q->avg_window_bytes += len - ack_pkt_len; } if (q->overflow_timeout) @@ -1927,24 +1928,29 @@ static s32 cake_enqueue(struct sk_buff *skb, struct Qdisc *sch, if (q->buffer_used > q->buffer_max_used) q->buffer_max_used = q->buffer_used; - if (q->buffer_used > q->buffer_limit) { - bool same_flow = false; - u32 dropped = 0; - u32 drop_id; + if (q->buffer_used <= q->buffer_limit) + return NET_XMIT_SUCCESS; - while (q->buffer_used > q->buffer_limit) { - dropped++; - drop_id = cake_drop(sch, to_free); + prev_qlen = sch->q.qlen; + prev_backlog = sch->qstats.backlog; - if ((drop_id >> 16) == tin && - (drop_id & 0xFFFF) == idx) - same_flow = true; - } - b->drop_overlimit += dropped; - - if (same_flow) - return NET_XMIT_CN; + while (q->buffer_used > q->buffer_limit) { + drop_id = cake_drop(sch, to_free); + if ((drop_id >> 16) == tin && + (drop_id & 0xFFFF) == idx) + same_flow = true; } + + prev_qlen -= sch->q.qlen; + prev_backlog -= sch->qstats.backlog; + b->drop_overlimit += prev_qlen; + + if (same_flow) { + qdisc_tree_reduce_backlog(sch, prev_qlen - 1, + prev_backlog - len); + return NET_XMIT_CN; + } + qdisc_tree_reduce_backlog(sch, prev_qlen, prev_backlog); return NET_XMIT_SUCCESS; } From 108f9405ce81085284c7ab09b84784b94b611435 Mon Sep 17 00:00:00 2001 From: Xiang Mei Date: Thu, 27 Nov 2025 17:14:16 -0700 Subject: [PATCH 1377/1381] selftests/tc-testing: Test CAKE scheduler when enqueue drops packets MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Add tests that trigger packet drops in cake_enqueue(): "CAKE with QFQ Parent - CAKE enqueue with packets dropping". It forces CAKE_enqueue to return NET_XMIT_CN after dropping the packets when it has a QFQ parent. Signed-off-by: Xiang Mei Reviewed-by: Toke Høiland-Jørgensen Link: https://patch.msgid.link/20251128001415.377823-3-xmei5@asu.edu Signed-off-by: Paolo Abeni --- .../tc-testing/tc-tests/infra/qdiscs.json | 28 +++++++++++++++++++ 1 file changed, 28 insertions(+) diff --git a/tools/testing/selftests/tc-testing/tc-tests/infra/qdiscs.json b/tools/testing/selftests/tc-testing/tc-tests/infra/qdiscs.json index 0091bcd91c2c..47de27fd4f90 100644 --- a/tools/testing/selftests/tc-testing/tc-tests/infra/qdiscs.json +++ b/tools/testing/selftests/tc-testing/tc-tests/infra/qdiscs.json @@ -1005,5 +1005,33 @@ "teardown": [ "$TC qdisc del dev $DUMMY clsact" ] + }, + { + "id": "4366", + "name": "CAKE with QFQ Parent - CAKE enqueue with packets dropping", + "category": [ + "qdisc", + "cake", + "netem" + ], + "plugins": { + "requires": "nsPlugin" + }, + "setup":[ + "$TC qdisc add dev $DUMMY handle 1: root qfq", + "$TC class add dev $DUMMY parent 1: classid 1:1 qfq maxpkt 1024", + "$TC qdisc add dev $DUMMY parent 1:1 handle 2: cake memlimit 9", + "$TC filter add dev $DUMMY protocol ip parent 1: prio 1 u32 match ip protocol 1 0xff flowid 1:1", + "ping -I$DUMMY -f -c1 -s64 -W1 10.10.10.1 || true", + "$TC qdisc replace dev $DUMMY parent 1:1 handle 3: netem delay 0ms" + ], + "cmdUnderTest": "ping -I$DUMMY -f -c1 -s64 -W1 10.10.10.1 || true", + "expExitCode": "0", + "verifyCmd": "$TC -s qdisc show dev $DUMMY", + "matchPattern": "qdisc qfq 1:", + "matchCount": "1", + "teardown": [ + "$TC qdisc del dev $DUMMY handle 1: root" + ] } ] From 96a8395061358adcd4b6a4f0f4c8989ec69e8659 Mon Sep 17 00:00:00 2001 From: Tariq Toukan Date: Sun, 30 Nov 2025 12:13:36 +0200 Subject: [PATCH 1378/1381] net/mlx5e: Update XDP features in switch channels The XDP features state might depend of the state of other features, like HW-LRO / HW-GRO. In general, move the re-evaluation announcement of the XDP features (xdp_set_features_flag_locked) into the flow where configuration gets changed. There's no point in updating them elsewhere. This is a more appropriate place, as this modifies the announced features while channels are inactive, which avoids the small interval between channel activation and the proper setting of the XDP features. Signed-off-by: Tariq Toukan Reviewed-by: Dragos Tatulea Reviewed-by: William Tu Link: https://patch.msgid.link/1764497617-1326331-2-git-send-email-tariqt@nvidia.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/mellanox/mlx5/core/en.h | 2 +- .../net/ethernet/mellanox/mlx5/core/en_ethtool.c | 10 +--------- drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 13 ++++++------- drivers/net/ethernet/mellanox/mlx5/core/en_rep.c | 2 +- 4 files changed, 9 insertions(+), 18 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index 3ada7c16adfb..811178d8976c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -1249,7 +1249,7 @@ void mlx5e_netdev_attach_nic_profile(struct mlx5e_priv *priv); void mlx5e_set_netdev_mtu_boundaries(struct mlx5e_priv *priv); void mlx5e_build_nic_params(struct mlx5e_priv *priv, struct mlx5e_xsk *xsk, u16 mtu); -void mlx5e_set_xdp_feature(struct net_device *netdev); +void mlx5e_set_xdp_feature(struct mlx5e_priv *priv); netdev_features_t mlx5e_features_check(struct sk_buff *skb, struct net_device *netdev, netdev_features_t features); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c index fe67c73849f9..d3fef1e7e2f7 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -2294,7 +2294,6 @@ static int set_pflag_rx_striding_rq(struct net_device *netdev, bool enable) struct mlx5e_priv *priv = netdev_priv(netdev); struct mlx5_core_dev *mdev = priv->mdev; struct mlx5e_params new_params; - int err; if (enable) { /* Checking the regular RQ here; mlx5e_validate_xsk_param called @@ -2315,14 +2314,7 @@ static int set_pflag_rx_striding_rq(struct net_device *netdev, bool enable) MLX5E_SET_PFLAG(&new_params, MLX5E_PFLAG_RX_STRIDING_RQ, enable); mlx5e_set_rq_type(mdev, &new_params); - err = mlx5e_safe_switch_params(priv, &new_params, NULL, NULL, true); - if (err) - return err; - - /* update XDP supported features */ - mlx5e_set_xdp_feature(netdev); - - return 0; + return mlx5e_safe_switch_params(priv, &new_params, NULL, NULL, true); } static int set_pflag_rx_no_csum_complete(struct net_device *netdev, bool enable) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index e537df670758..f0f2bc7f317d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -3365,6 +3365,7 @@ static int mlx5e_switch_priv_params(struct mlx5e_priv *priv, } } + mlx5e_set_xdp_feature(priv); return 0; } @@ -3396,6 +3397,7 @@ static int mlx5e_switch_priv_channels(struct mlx5e_priv *priv, } } + mlx5e_set_xdp_feature(priv); if (!MLX5_CAP_GEN(priv->mdev, tis_tir_td_order)) mlx5e_close_channels(old_chs); priv->profile->update_rx(priv); @@ -4409,10 +4411,10 @@ static int mlx5e_handle_feature(struct net_device *netdev, return 0; } -void mlx5e_set_xdp_feature(struct net_device *netdev) +void mlx5e_set_xdp_feature(struct mlx5e_priv *priv) { - struct mlx5e_priv *priv = netdev_priv(netdev); struct mlx5e_params *params = &priv->channels.params; + struct net_device *netdev = priv->netdev; xdp_features_t val; if (!netdev->netdev_ops->ndo_bpf || @@ -4461,9 +4463,6 @@ int mlx5e_set_features(struct net_device *netdev, netdev_features_t features) return -EINVAL; } - /* update XDP supported features */ - mlx5e_set_xdp_feature(netdev); - return 0; } @@ -5859,7 +5858,7 @@ static void mlx5e_build_nic_netdev(struct net_device *netdev) netdev->netmem_tx = true; netif_set_tso_max_size(netdev, GSO_MAX_SIZE); - mlx5e_set_xdp_feature(netdev); + mlx5e_set_xdp_feature(priv); mlx5e_set_netdev_dev_addr(netdev); mlx5e_macsec_build_netdev(priv); mlx5e_ipsec_build_netdev(priv); @@ -5957,7 +5956,7 @@ static int mlx5e_nic_init(struct mlx5_core_dev *mdev, mlx5e_psp_register(priv); /* update XDP supported features */ - mlx5e_set_xdp_feature(netdev); + mlx5e_set_xdp_feature(priv); if (take_rtnl) rtnl_unlock(); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c index 0335ca8277ef..ee9595109649 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c @@ -867,7 +867,7 @@ static void mlx5e_build_rep_params(struct net_device *netdev) if (take_rtnl) rtnl_lock(); /* update XDP supported features */ - mlx5e_set_xdp_feature(netdev); + mlx5e_set_xdp_feature(priv); if (take_rtnl) rtnl_unlock(); From d4aa0cc9bd31f3e0cd5f067d649bf39135e4b46b Mon Sep 17 00:00:00 2001 From: Tariq Toukan Date: Sun, 30 Nov 2025 12:13:37 +0200 Subject: [PATCH 1379/1381] net/mlx5e: Support XDP target xmit with dummy program Save per-channel resources in default, in device and host memory. As no better API exist, make the XDP-redirect-target SQ available by loading a dummy XDP program. This improves the latency of interface up/down operations when feature is disabled. Perf numbers: NIC: Connect-X7. Setup: 248 channels, default mtu and rx/tx ring sizes. Interface up + down: Before: 2.246 secs After: 1.798 secs (-0.448 sec) Saves ~1.8 msec per channel. Signed-off-by: Tariq Toukan Reviewed-by: Dragos Tatulea Reviewed-by: William Tu Link: https://patch.msgid.link/1764497617-1326331-3-git-send-email-tariqt@nvidia.com Signed-off-by: Paolo Abeni --- .../net/ethernet/mellanox/mlx5/core/en_main.c | 23 +++++++++---------- 1 file changed, 11 insertions(+), 12 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index f0f2bc7f317d..6168f0814414 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -2612,7 +2612,7 @@ static int mlx5e_open_queues(struct mlx5e_channel *c, if (err) goto err_close_icosq_cq; - if (netdev_ops->ndo_xdp_xmit) { + if (netdev_ops->ndo_xdp_xmit && c->xdp) { c->xdpsq = mlx5e_open_xdpredirect_sq(c, params, cparam, &ccp); if (IS_ERR(c->xdpsq)) { err = PTR_ERR(c->xdpsq); @@ -4415,19 +4415,18 @@ void mlx5e_set_xdp_feature(struct mlx5e_priv *priv) { struct mlx5e_params *params = &priv->channels.params; struct net_device *netdev = priv->netdev; - xdp_features_t val; + xdp_features_t val = 0; - if (!netdev->netdev_ops->ndo_bpf || - params->packet_merge.type != MLX5E_PACKET_MERGE_NONE) { - xdp_set_features_flag_locked(netdev, 0); - return; - } + if (netdev->netdev_ops->ndo_bpf && + params->packet_merge.type == MLX5E_PACKET_MERGE_NONE) + val = NETDEV_XDP_ACT_BASIC | NETDEV_XDP_ACT_REDIRECT | + NETDEV_XDP_ACT_XSK_ZEROCOPY | + NETDEV_XDP_ACT_RX_SG; + + if (netdev->netdev_ops->ndo_xdp_xmit && params->xdp_prog) + val |= NETDEV_XDP_ACT_NDO_XMIT | + NETDEV_XDP_ACT_NDO_XMIT_SG; - val = NETDEV_XDP_ACT_BASIC | NETDEV_XDP_ACT_REDIRECT | - NETDEV_XDP_ACT_XSK_ZEROCOPY | - NETDEV_XDP_ACT_RX_SG | - NETDEV_XDP_ACT_NDO_XMIT | - NETDEV_XDP_ACT_NDO_XMIT_SG; xdp_set_features_flag_locked(netdev, val); } From 854858848bc7ac6ea3bf896ef1b541cb9d4bca20 Mon Sep 17 00:00:00 2001 From: Pavel Begunkov Date: Sun, 30 Nov 2025 23:35:16 +0000 Subject: [PATCH 1380/1381] net: page pool: xa init with destroy on pp init The free_ptr_ring label path initialises ->dma_mapped xarray but doesn't destroy it in case of an error. That's not a real problem since init itself doesn't do anything requiring destruction, but still match it with xa_destroy() to silence warnings. Signed-off-by: Pavel Begunkov Link: https://patch.msgid.link/02904c6d83dbe5cc1c671106a5c97bd93ab31006.1764542851.git.asml.silence@gmail.com Signed-off-by: Jakub Kicinski --- net/core/page_pool.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/core/page_pool.c b/net/core/page_pool.c index 1a5edec485f1..a085fd199ff0 100644 --- a/net/core/page_pool.c +++ b/net/core/page_pool.c @@ -307,6 +307,7 @@ static int page_pool_init(struct page_pool *pool, free_ptr_ring: ptr_ring_cleanup(&pool->ring, NULL); + xa_destroy(&pool->dma_mapped); #ifdef CONFIG_PAGE_POOL_STATS if (!pool->system) free_percpu(pool->recycle_stats); From 9954464d737dd12f12b274d3da46397e3656f079 Mon Sep 17 00:00:00 2001 From: Pavel Begunkov Date: Sun, 30 Nov 2025 23:35:17 +0000 Subject: [PATCH 1381/1381] net: page_pool: sanitise allocation order We're going to give more control over rx buffer sizes to user space, and since we can't always rely on driver validation, let's sanitise it in page_pool_init() as well. Note that we only need to reject over MAX_PAGE_ORDER allocations for normal page pools, as current memory providers don't need to use the buddy allocator and must check the order on init.i Suggested-by: Stanislav Fomichev Reviewed-by: Mina Almasry Signed-off-by: Pavel Begunkov Link: https://patch.msgid.link/77ad83c1aec66cbd00e7b3952f74bc3b7a988150.1764542851.git.asml.silence@gmail.com Signed-off-by: Jakub Kicinski --- net/core/page_pool.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/core/page_pool.c b/net/core/page_pool.c index a085fd199ff0..265a729431bb 100644 --- a/net/core/page_pool.c +++ b/net/core/page_pool.c @@ -301,6 +301,9 @@ static int page_pool_init(struct page_pool *pool, } static_branch_inc(&page_pool_mem_providers); + } else if (pool->p.order > MAX_PAGE_ORDER) { + err = -EINVAL; + goto free_ptr_ring; } return 0;