When both CONFIG_IRQ_STACKS and SCS are enabled, also use a separate
per-CPU shadow call stack.
Signed-off-by: Sami Tolvanen <samitolvanen@google.com>
Tested-by: Nathan Chancellor <nathan@kernel.org>
Link: https://lore.kernel.org/r/20230927224757.1154247-13-samitolvanen@google.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
Implement CONFIG_SHADOW_CALL_STACK for RISC-V. When enabled, the
compiler injects instructions to all non-leaf C functions to
store the return address to the shadow stack and unconditionally
load it again before returning, which makes it harder to corrupt
the return address through a stack overflow, for example.
The active shadow call stack pointer is stored in the gp
register, which makes SCS incompatible with gp relaxation. Use
--no-relax-gp to ensure gp relaxation is disabled and disable
global pointer loading. Add SCS pointers to struct thread_info,
implement SCS initialization, and task switching
Signed-off-by: Sami Tolvanen <samitolvanen@google.com>
Tested-by: Nathan Chancellor <nathan@kernel.org>
Link: https://lore.kernel.org/r/20230927224757.1154247-12-samitolvanen@google.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
In Clang 17, -fsanitize=shadow-call-stack uses the newly declared
platform register gp for storing shadow call stack pointers. As
this is obviously incompatible with gp relaxation, in preparation
for CONFIG_SHADOW_CALL_STACK support, move global pointer loading
to a single macro, which we can cleanly disable when SCS is used
instead.
Link: https://reviews.llvm.org/rGaa1d2693c256
Link: a484e843e6
Signed-off-by: Sami Tolvanen <samitolvanen@google.com>
Tested-by: Nathan Chancellor <nathan@kernel.org>
Link: https://lore.kernel.org/r/20230927224757.1154247-11-samitolvanen@google.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
With CONFIG_IRQ_STACKS, we switch to a separate per-CPU IRQ stack
before calling handle_riscv_irq or __do_softirq. We currently
have duplicate inline assembly snippets for stack switching in
both code paths. Now that we can access per-CPU variables in
assembly, implement call_on_irq_stack in assembly, and use that
instead of redundant inline assembly.
Signed-off-by: Sami Tolvanen <samitolvanen@google.com>
Tested-by: Nathan Chancellor <nathan@kernel.org>
Reviewed-by: Guo Ren <guoren@kernel.org>
Link: https://lore.kernel.org/r/20230927224757.1154247-10-samitolvanen@google.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
Enhance the acpi_os_ioremap() to support opregions in MMIO space. Also,
have strict checks using EFI memory map to allow remapping the RAM similar
to arm64.
Signed-off-by: Sunil V L <sunilvl@ventanamicro.com>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Reviewed-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Acked-by: Conor Dooley <conor.dooley@microchip.com>
Link: https://lore.kernel.org/r/20231018124007.1306159-2-sunilvl@ventanamicro.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
After the vga console no longer relies on global screen_info, there are
only two remaining use cases:
- on the x86 architecture, it is used for multiple boot methods
(bzImage, EFI, Xen, kexec) to commucate the initial VGA or framebuffer
settings to a number of device drivers.
- on other architectures, it is only used as part of the EFI stub,
and only for the three sysfb framebuffers (simpledrm, simplefb, efifb).
Remove the duplicate data structure definitions by moving it into the
efi-init.c file that sets it up initially for the EFI case, leaving x86
as an exception that retains its own definition for non-EFI boots.
The added #ifdefs here are optional, I added them to further limit the
reach of screen_info to configurations that have at least one of the
users enabled.
Reviewed-by: Ard Biesheuvel <ardb@kernel.org>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Acked-by: Helge Deller <deller@gmx.de>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Link: https://lore.kernel.org/r/20231017093947.3627976-1-arnd@kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
On non-x86 architectures, the screen_info variable is generally only
used for the VGA console where supported, and in some cases the EFI
framebuffer or vga16fb.
Now that we have a definite list of which architectures actually use it
for what, use consistent #ifdef checks so the global variable is only
defined when it is actually used on those architectures.
Loongarch and riscv have no support for vgacon or vga16fb, but
they support EFI firmware, so only that needs to be checked, and the
initialization can be removed because that is handled by EFI.
IA64 has both vgacon and EFI, though EFI apparently never uses
a framebuffer here.
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Khalid Aziz <khalid@gonehiking.org>
Acked-by: Helge Deller <deller@gmx.de>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Link: https://lore.kernel.org/r/20231009211845.3136536-3-arnd@kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
If configuration options SOFTIRQ_ON_OWN_STACK and PREEMPT_RT
are enabled simultaneously under RISC-V architecture,
it will result in a compilation failure:
arch/riscv/kernel/irq.c:64:6: error: redefinition of 'do_softirq_own_stack'
64 | void do_softirq_own_stack(void)
| ^~~~~~~~~~~~~~~~~~~~
In file included from ./arch/riscv/include/generated/asm/softirq_stack.h:1,
from arch/riscv/kernel/irq.c:15:
./include/asm-generic/softirq_stack.h:8:20: note: previous definition of 'do_softirq_own_stack' was here
8 | static inline void do_softirq_own_stack(void)
| ^~~~~~~~~~~~~~~~~~~~
After changing CONFIG_HAVE_SOFTIRQ_ON_OWN_STACK to CONFIG_SOFTIRQ_ON_OWN_STACK,
compilation can be successful.
Fixes: dd69d07a5a ("riscv: stack: Support HAVE_SOFTIRQ_ON_OWN_STACK")
Reviewed-by: Guo Ren <guoren@kernel.org>
Signed-off-by: Jiexun Wang <wangjiexun@tinylab.org>
Reviewed-by: Samuel Holland <samuel@sholland.org>
Link: https://lore.kernel.org/r/20230913052940.374686-1-wangjiexun@tinylab.org
Cc: stable@vger.kernel.org
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
The RISC-V integer conditional (Zicond) operation extension defines
standard conditional arithmetic and conditional-select/move operations
which are inspired from the XVentanaCondOps extension. In fact, QEMU
RISC-V also has support for emulating Zicond extension.
Let us detect Zicond extension from ISA string available through
DT or ACPI.
Signed-off-by: Anup Patel <apatel@ventanamicro.com>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Reviewed-by: Conor Dooley <conor.dooley@microchip.com>
Acked-by: Palmer Dabbelt <palmer@rivosinc.com>
Signed-off-by: Anup Patel <anup@brainfault.org>
Extend the ISA string parsing to detect the Smstateen extension. If the
extension is enabled then access to certain 'state' such as AIA CSRs in
VS mode is controlled by *stateen0 registers.
Signed-off-by: Mayuresh Chitale <mchitale@ventanamicro.com>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Signed-off-by: Anup Patel <anup@brainfault.org>
The alternative stack checking in get_sigframe introduced by the Vector
support is not needed and has a problem. It is not needed as we have
already validate it at the beginning of the function if we are already
on an altstack. If not, the size of an altstack is always validated at
its allocation stage with sigaltstack_size_valid().
Besides, we must only regard the size of an altstack if the handler of a
signal is registered with SA_ONSTACK. So, blindly checking overflow of
an altstack if sas_ss_size not equals to zero will check against wrong
signal handlers if only a subset of signals are registered with
SA_ONSTACK.
Fixes: 8ee0b41898 ("riscv: signal: Add sigcontext save/restore for vector")
Reported-by: Prashanth Swaminathan <prashanthsw@google.com>
Signed-off-by: Andy Chiu <andy.chiu@sifive.com>
Link: https://lore.kernel.org/r/20230822164904.21660-1-andy.chiu@sifive.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
This commit comes at the tail end of a greater effort to remove the
empty elements at the end of the ctl_table arrays (sentinels) which
will reduce the overall build time size of the kernel and run time
memory bloat by ~64 bytes per sentinel (further information Link :
https://lore.kernel.org/all/ZO5Yx5JFogGi%2FcBo@bombadil.infradead.org/)
Remove sentinel element from riscv_v_default_vstate_table. This removal
is safe because register_sysctl implicitly uses ARRAY_SIZE() in addition
to checking for the sentinel.
Signed-off-by: Joel Granados <j.granados@samsung.com>
Signed-off-by: Luis Chamberlain <mcgrof@kernel.org>
and fix all in-tree references.
Architecture-specific documentation is being moved into Documentation/arch/
as a way of cleaning up the top-level documentation directory and making
the docs hierarchy more closely match the source hierarchy.
Signed-off-by: Costa Shulyupin <costa.shul@redhat.com>
Reviewed-by: Conor Dooley <conor.dooley@microchip.com>
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
Link: https://lore.kernel.org/r/20230930185354.3034118-1-costa.shul@redhat.com
Expose Zicboz through hwprobe and also provide a key to extract its
respective block size. Opportunistically add a macro and apply it to
current extensions in order to avoid duplicating code.
Signed-off-by: Andrew Jones <ajones@ventanamicro.com>
Reviewed-by: Evan Green <evan@rivosinc.com>
Reviewed-by: Conor Dooley <conor.dooley@microchip.com>
Link: https://lore.kernel.org/r/20230918131518.56803-11-ajones@ventanamicro.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
When Zicboz is present, enable its instruction (cbo.zero) in
usermode by setting its respective senvcfg bit. We don't bother
trying to set this bit per-task, which would also require an
interface for tasks to request enabling and/or disabling. Instead,
permanently set the bit for each hart which has the extension when
bringing it online.
This patch also introduces riscv_cpu_has_extension_[un]likely()
functions to check a specific hart's ISA bitmap for extensions.
Prior to checking the specific hart's bitmap in these functions
we try the bitmap which represents the LCD of extensions, but only
when we know it will use its optimized, alternatives path by gating
its call on CONFIG_RISCV_ALTERNATIVE. When alternatives are used, the
compiler ensures that the invocation of the LCD search becomes a
constant true or false. When it's true, even the new functions will
completely vanish from their callsites. OTOH, when the LCD check is
false, we need to do a search of the hart's ISA bitmap. Had we also
checked the LCD bitmap without the use of alternatives, then we would
have ended up with two bitmap searches instead of one.
Signed-off-by: Andrew Jones <ajones@ventanamicro.com>
Reviewed-by: Conor Dooley <conor.dooley@microchip.com>
Link: https://lore.kernel.org/r/20230918131518.56803-10-ajones@ventanamicro.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
commit c818fea83d ("riscv: say disabling zicbom if no or bad
riscv,cbom-block-size found") improved the error messages for
zicbom but zicboz was missed since its patches were in flight
at the same time. Get 'em now.
Signed-off-by: Andrew Jones <ajones@ventanamicro.com>
Reviewed-by: Conor Dooley <conor.dooley@microchip.com>
Link: https://lore.kernel.org/r/20230918131518.56803-9-ajones@ventanamicro.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
RISC-V software breakpoint trap handlers are used for {k,u}probes.
When trapping from kernelmode, only the kernelmode handlers should be
considered. Vice versa, only usermode handlers for usermode
traps. This is not the case on RISC-V, which can trigger a bug if a
userspace process uses uprobes, and a WARN() is triggered from
kernelmode (which is implemented via {c.,}ebreak).
The kernel will trap on the kernelmode {c.,}ebreak, look for uprobes
handlers, realize incorrectly that uprobes need to be handled, and
exit the trap handler early. The trap returns to re-executing the
{c.,}ebreak, and enter an infinite trap-loop.
The issue was found running the BPF selftest [1].
Fix this issue by only considering the swbp/ss handlers for
kernel/usermode respectively. Also, move CONFIG ifdeffery from traps.c
to the asm/{k,u}probes.h headers.
Note that linux/uprobes.h only include asm/uprobes.h if CONFIG_UPROBES
is defined, which is why asm/uprobes.h needs to be unconditionally
included in traps.c
Link: https://lore.kernel.org/linux-riscv/87v8d19aun.fsf@all.your.base.are.belong.to.us/ # [1]
Fixes: 74784081aa ("riscv: Add uprobes supported")
Reviewed-by: Guo Ren <guoren@kernel.org>
Reviewed-by: Nam Cao <namcaov@gmail.com>
Tested-by: Puranjay Mohan <puranjay12@gmail.com>
Signed-off-by: Björn Töpel <bjorn@rivosinc.com>
Link: https://lore.kernel.org/r/20230912065619.62020-1-bjorn@kernel.org
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
The current riscv boot protocol requires 2MB alignment for RV64
and 4MB alignment for RV32.
In KEXEC_FILE path, the elf_find_pbase() function should align
the kexeced kernel entry according to the requirement, otherwise
the kexeced kernel would silently BUG at the setup_vm().
Fixes: 8acea455fa ("RISC-V: Support for kexec_file on panic")
Signed-off-by: Song Shuai <songshuaishuai@tinylab.org>
Link: https://lore.kernel.org/r/20230907103304.590739-3-songshuaishuai@tinylab.org
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
For readability and simplicity, cleanup the riscv_kexec_relocate code:
- Re-sort the first 4 `mv` instructions against `riscv_kexec_method()`
- Eliminate registers for debugging (s9,s10,s11) and storing const-value (s5,s6)
- Replace `jalr` with `jr` for no-link jump
I tested this on Qemu virt machine and works as it was.
Signed-off-by: Song Shuai <songshuaishuai@tinylab.org>
Link: https://lore.kernel.org/r/20230907103304.590739-2-songshuaishuai@tinylab.org
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
The current riscv boot protocol requires 2MB alignment for RV64
and 4MB alignment for RV32.
In KEXEC_FILE path, the elf_find_pbase() function should align
the kexeced kernel entry according to the requirement, otherwise
the kexeced kernel would silently BUG at the setup_vm().
Fixes: 8acea455fa ("RISC-V: Support for kexec_file on panic")
Signed-off-by: Song Shuai <songshuaishuai@tinylab.org>
Link: https://lore.kernel.org/r/20230906095817.364390-1-songshuaishuai@tinylab.org
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
Puranjay Mohan <puranjay12@gmail.com> says:
Here is some data to prove the V2 fixes the problem:
Without this series:
root@rv-selftester:~/src/kselftest/bpf# time ./test_tag
test_tag: OK (40945 tests)
real 7m47.562s
user 0m24.145s
sys 6m37.064s
With this series applied:
root@rv-selftester:~/src/selftest/bpf# time ./test_tag
test_tag: OK (40945 tests)
real 7m29.472s
user 0m25.865s
sys 6m18.401s
BPF programs currently consume a page each on RISCV. For systems with many BPF
programs, this adds significant pressure to instruction TLB. High iTLB pressure
usually causes slow down for the whole system.
Song Liu introduced the BPF prog pack allocator[1] to mitigate the above issue.
It packs multiple BPF programs into a single huge page. It is currently only
enabled for the x86_64 BPF JIT.
I enabled this allocator on the ARM64 BPF JIT[2]. It is being reviewed now.
This patch series enables the BPF prog pack allocator for the RISCV BPF JIT.
======================================================
Performance Analysis of prog pack allocator on RISCV64
======================================================
Test setup:
===========
Host machine: Debian GNU/Linux 11 (bullseye)
Qemu Version: QEMU emulator version 8.0.3 (Debian 1:8.0.3+dfsg-1)
u-boot-qemu Version: 2023.07+dfsg-1
opensbi Version: 1.3-1
To test the performance of the BPF prog pack allocator on RV, a stresser
tool[4] linked below was built. This tool loads 8 BPF programs on the system and
triggers 5 of them in an infinite loop by doing system calls.
The runner script starts 20 instances of the above which loads 8*20=160 BPF
programs on the system, 5*20=100 of which are being constantly triggered.
The script is passed a command which would be run in the above environment.
The script was run with following perf command:
./run.sh "perf stat -a \
-e iTLB-load-misses \
-e dTLB-load-misses \
-e dTLB-store-misses \
-e instructions \
--timeout 60000"
The output of the above command is discussed below before and after enabling the
BPF prog pack allocator.
The tests were run on qemu-system-riscv64 with 8 cpus, 16G memory. The rootfs
was created using Bjorn's riscv-cross-builder[5] docker container linked below.
Results
=======
Before enabling prog pack allocator:
------------------------------------
Performance counter stats for 'system wide':
4939048 iTLB-load-misses
5468689 dTLB-load-misses
465234 dTLB-store-misses
1441082097998 instructions
60.045791200 seconds time elapsed
After enabling prog pack allocator:
-----------------------------------
Performance counter stats for 'system wide':
3430035 iTLB-load-misses
5008745 dTLB-load-misses
409944 dTLB-store-misses
1441535637988 instructions
60.046296600 seconds time elapsed
Improvements in metrics
=======================
It was expected that the iTLB-load-misses would decrease as now a single huge
page is used to keep all the BPF programs compared to a single page for each
program earlier.
--------------------------------------------
The improvement in iTLB-load-misses: -30.5 %
--------------------------------------------
I repeated this expriment more than 100 times in different setups and the
improvement was always greater than 30%.
This patch series is boot tested on the Starfive VisionFive 2 board[6].
The performance analysis was not done on the board because it doesn't
expose iTLB-load-misses, etc. The stresser program was run on the board to test
the loading and unloading of BPF programs
[1] https://lore.kernel.org/bpf/20220204185742.271030-1-song@kernel.org/
[2] https://lore.kernel.org/all/20230626085811.3192402-1-puranjay12@gmail.com/
[3] https://lore.kernel.org/all/20230626085811.3192402-2-puranjay12@gmail.com/
[4] https://github.com/puranjaymohan/BPF-Allocator-Bench
[5] https://github.com/bjoto/riscv-cross-builder
[6] https://www.starfivetech.com/en/site/boards
* b4-shazam-merge:
bpf, riscv: use prog pack allocator in the BPF JIT
riscv: implement a memset like function for text
riscv: extend patch_text_nosync() for multiple pages
bpf: make bpf_prog_pack allocator portable
Link: https://lore.kernel.org/r/20230831131229.497941-1-puranjay12@gmail.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
Alexandre Ghiti <alexghiti@rivosinc.com> says:
The following KASLR implementation allows to randomize the kernel mapping:
- virtually: we expect the bootloader to provide a seed in the device-tree
- physically: only implemented in the EFI stub, it relies on the firmware to
provide a seed using EFI_RNG_PROTOCOL. arm64 has a similar implementation
hence the patch 3 factorizes KASLR related functions for riscv to take
advantage.
The new virtual kernel location is limited by the early page table that only
has one PUD and with the PMD alignment constraint, the kernel can only take
< 512 positions.
* b4-shazam-merge:
riscv: libstub: Implement KASLR by using generic functions
libstub: Fix compilation warning for rv32
arm64: libstub: Move KASLR handling functions to kaslr.c
riscv: Dump out kernel offset information on panic
riscv: Introduce virtual kernel mapping KASLR
Link: https://lore.kernel.org/r/20230722123850.634544-1-alexghiti@rivosinc.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
This resurrects the vector ptrace() support that was removed for 6.5 due
to some bugs cropping up as part of the GDB review process.
* b4-shazam-merge:
RISC-V: Add ptrace support for vectors
Link: https://lore.kernel.org/r/20230825050248.32681-1-andy.chiu@sifive.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
Prabhakar <prabhakar.csengg@gmail.com> says:
From: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
non-coherent DMA support for AX45MP
====================================
On the Andes AX45MP core, cache coherency is a specification option so it
may not be supported. In this case DMA will fail. To get around with this
issue this patch series does the below:
1] Andes alternative ports is implemented as errata which checks if the
IOCP is missing and only then applies to CMO errata. One vendor specific
SBI EXT (ANDES_SBI_EXT_IOCP_SW_WORKAROUND) is implemented as part of
errata.
Below are the configs which Andes port provides (and are selected by
RZ/Five):
- ERRATA_ANDES
- ERRATA_ANDES_CMO
OpenSBI patch supporting ANDES_SBI_EXT_IOCP_SW_WORKAROUND SBI is now
part v1.3 release.
2] Andes AX45MP core has a Programmable Physical Memory Attributes (PMA)
block that allows dynamic adjustment of memory attributes in the runtime.
It contains a configurable amount of PMA entries implemented as CSR
registers to control the attributes of memory locations in interest.
OpenSBI configures the PMA regions as required and creates a reserve memory
node and propagates it to the higher boot stack.
Currently OpenSBI (upstream) configures the required PMA region and passes
this a shared DMA pool to Linux.
reserved-memory {
#address-cells = <2>;
#size-cells = <2>;
ranges;
pma_resv0@58000000 {
compatible = "shared-dma-pool";
reg = <0x0 0x58000000 0x0 0x08000000>;
no-map;
linux,dma-default;
};
};
The above shared DMA pool gets appended to Linux DTB so the DMA memory
requests go through this region.
3] We provide callbacks to synchronize specific content between memory and
cache.
4] RZ/Five SoC selects the below configs
- AX45MP_L2_CACHE
- DMA_GLOBAL_POOL
- ERRATA_ANDES
- ERRATA_ANDES_CMO
----------x---------------------x--------------------x---------------x----
* b4-shazam-merge:
soc: renesas: Kconfig: Select the required configs for RZ/Five SoC
cache: Add L2 cache management for Andes AX45MP RISC-V core
dt-bindings: cache: andestech,ax45mp-cache: Add DT binding documentation for L2 cache controller
riscv: mm: dma-noncoherent: nonstandard cache operations support
riscv: errata: Add Andes alternative ports
riscv: asm: vendorid_list: Add Andes Technology to the vendors list
Link: https://lore.kernel.org/r/20230818135723.80612-1-prabhakar.mahadev-lad.rj@bp.renesas.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
Evan Green <evan@rivosinc.com> says:
The current setting for the hwprobe bit indicating misaligned access
speed is controlled by a vendor-specific feature probe function. This is
essentially a per-SoC table we have to maintain on behalf of each vendor
going forward. Let's convert that instead to something we detect at
runtime.
We have two assembly routines at the heart of our probe: one that
does a bunch of word-sized accesses (without aligning its input buffer),
and the other that does byte accesses. If we can move a larger number of
bytes using misaligned word accesses than we can with the same amount of
time doing byte accesses, then we can declare misaligned accesses as
"fast".
The tradeoff of reducing this maintenance burden is boot time. We spend
4-6 jiffies per core doing this measurement (0-2 on jiffie edge
alignment, and 4 on measurement). The timing loop was based on
raid6_choose_gen(), which uses (16+1)*N jiffies (where N is the number
of algorithms). By taking only the fastest iteration out of all
attempts for use in the comparison, variance between runs is very low.
On my THead C906, it looks like this:
[ 0.047563] cpu0: Ratio of byte access time to unaligned word access is 4.34, unaligned accesses are fast
Several others have chimed in with results on slow machines with the
older algorithm, which took all runs into account, including noise like
interrupts. Even with this variation, results indicate that in all cases
(fast, slow, and emulated) the measured numbers are nowhere near each
other (always multiple factors away).
* b4-shazam-merge:
RISC-V: alternative: Remove feature_probe_func
RISC-V: Probe for unaligned access speed
Link: https://lore.kernel.org/r/20230818194136.4084400-1-evan@rivosinc.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
The BPF JIT needs to write invalid instructions to RX regions of memory to
invalidate removed BPF programs. This needs a function like memset() that
can work with RX memory.
Implement patch_text_set_nosync() which is similar to text_poke_set() of
x86.
Signed-off-by: Puranjay Mohan <puranjay12@gmail.com>
Reviewed-by: Pu Lehui <pulehui@huawei.com>
Acked-by: Björn Töpel <bjorn@kernel.org>
Tested-by: Björn Töpel <bjorn@rivosinc.com>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/r/20230831131229.497941-4-puranjay12@gmail.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
The patch_insn_write() function currently doesn't work for multiple pages
of instructions, therefore patch_text_nosync() will fail with a page fault
if called with lengths spanning multiple pages.
This commit extends the patch_insn_write() function to support multiple
pages by copying at max 2 pages at a time in a loop. This implementation
is similar to text_poke_copy() function of x86.
Signed-off-by: Puranjay Mohan <puranjay12@gmail.com>
Reviewed-by: Pu Lehui <pulehui@huawei.com>
Reviewed-by: Björn Töpel <bjorn@rivosinc.com>
Tested-by: Björn Töpel <bjorn@rivosinc.com>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/r/20230831131229.497941-3-puranjay12@gmail.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
We can now use arm64 functions to handle the move of the kernel physical
mapping: if KASLR is enabled, we will try to get a random seed from the
firmware, if not possible, the kernel will be moved to a location that
suits its alignment constraints.
Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Tested-by: Conor Dooley <conor.dooley@microchip.com>
Tested-by: Song Shuai <songshuaishuai@tinylab.org>
Reviewed-by: Sami Tolvanen <samitolvanen@google.com>
Tested-by: Sami Tolvanen <samitolvanen@google.com>
Link: https://lore.kernel.org/r/20230722123850.634544-6-alexghiti@rivosinc.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
Dump out the KASLR virtual kernel offset when panic to help debug kernel.
Signed-off-by: Zong Li <zong.li@sifive.com>
Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Tested-by: Conor Dooley <conor.dooley@microchip.com>
Tested-by: Song Shuai <songshuaishuai@tinylab.org>
Reviewed-by: Sami Tolvanen <samitolvanen@google.com>
Tested-by: Sami Tolvanen <samitolvanen@google.com>
Link: https://lore.kernel.org/r/20230722123850.634544-3-alexghiti@rivosinc.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
KASLR implementation relies on a relocatable kernel so that we can move
the kernel mapping.
The seed needed to virtually move the kernel is taken from the device tree,
so we rely on the bootloader to provide a correct seed. Zkr could be used
unconditionnally instead if implemented, but that's for another patch.
Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Tested-by: Conor Dooley <conor.dooley@microchip.com>
Tested-by: Song Shuai <songshuaishuai@tinylab.org>
Reviewed-by: Sami Tolvanen <samitolvanen@google.com>
Tested-by: Sami Tolvanen <samitolvanen@google.com>
Link: https://lore.kernel.org/r/20230722123850.634544-2-alexghiti@rivosinc.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
This patch add back the ptrace support with the following fix:
- Define NT_RISCV_CSR and re-number NT_RISCV_VECTOR to prevent
conflicting with gdb's NT_RISCV_CSR.
- Use struct __riscv_v_regset_state to handle ptrace requests
Since gdb does not directly include the note description header in
Linux and has already defined NT_RISCV_CSR as 0x900, we decide to
sync with gdb and renumber NT_RISCV_VECTOR to solve and prevent future
conflicts.
Fixes: 0c59922c76 ("riscv: Add ptrace vector support")
Signed-off-by: Andy Chiu <andy.chiu@sifive.com>
Link: https://lore.kernel.org/r/20230825050248.32681-1-andy.chiu@sifive.com
[Palmer: Drop the unused "size" variable in riscv_vr_set().]
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
Add required ports of the Alternative scheme for Andes CPU cores.
I/O Coherence Port (IOCP) provides an AXI interface for connecting external
non-caching masters, such as DMA controllers. IOCP is a specification
option and is disabled on the Renesas RZ/Five SoC due to this reason cache
management needs a software workaround.
Signed-off-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
Reviewed-by: Conor Dooley <conor.dooley@microchip.com>
Tested-by: Conor Dooley <conor.dooley@microchip.com> # tyre-kicking on a d1
Link: https://lore.kernel.org/r/20230818135723.80612-3-prabhakar.mahadev-lad.rj@bp.renesas.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
Now that we're testing unaligned memory copy and making that
determination generically, there are no more users of the vendor
feature_probe_func(). While I think it's probably going to need to come
back, there are no users right now, so let's remove it until it's
needed.
Signed-off-by: Evan Green <evan@rivosinc.com>
Reviewed-by: Conor Dooley <conor.dooley@microchip.com>
Link: https://lore.kernel.org/r/20230818194136.4084400-3-evan@rivosinc.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
Rather than deferring unaligned access speed determinations to a vendor
function, let's probe them and find out how fast they are. If we
determine that an unaligned word access is faster than N byte accesses,
mark the hardware's unaligned access as "fast". Otherwise, we mark
accesses as slow.
The algorithm itself runs for a fixed amount of jiffies. Within each
iteration it attempts to time a single loop, and then keeps only the best
(fastest) loop it saw. This algorithm was found to have lower variance from
run to run than my first attempt, which counted the total number of
iterations that could be done in that fixed amount of jiffies. By taking
only the best iteration in the loop, assuming at least one loop wasn't
perturbed by an interrupt, we eliminate the effects of interrupts and
other "warm up" factors like branch prediction. The only downside is it
depends on having an rdtime granular and accurate enough to measure a
single copy. If we ever manage to complete a loop in 0 rdtime ticks, we
leave the unaligned setting at UNKNOWN.
There is a slight change in user-visible behavior here. Previously, all
boards except the THead C906 reported misaligned access speed of
UNKNOWN. C906 reported FAST. With this change, since we're now measuring
misaligned access speed on each hart, all RISC-V systems will have this
key set as either FAST or SLOW.
Currently, we don't have a way to confidently measure the difference between
SLOW and EMULATED, so we label anything not fast as SLOW. This will
mislabel some systems that are actually EMULATED as SLOW. When we get
support for delegating misaligned access traps to the kernel (as opposed
to the firmware quietly handling it), we can explicitly test in Linux to
see if unaligned accesses trap. Those systems will start to report
EMULATED, though older (today's) systems without that new SBI mechanism
will continue to report SLOW.
I've updated the documentation for those hwprobe values to reflect
this, specifically: SLOW may or may not be emulated by software, and FAST
represents means being faster than equivalent byte accesses. The change
in documentation is accurate with respect to both the former and current
behavior.
Signed-off-by: Evan Green <evan@rivosinc.com>
Acked-by: Conor Dooley <conor.dooley@microchip.com>
Link: https://lore.kernel.org/r/20230818194136.4084400-2-evan@rivosinc.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
* Support for the new "riscv,isa-extensions" and "riscv,isa-base" device
tree interfaces for probing extensions.
* Support for userspace access to the performance counters.
* Support for more instructions in kprobes.
* Crash kernels can be allocated above 4GiB.
* Support for KCFI.
* Support for ELFs in !MMU configurations.
* ARCH_KMALLOC_MINALIGN has been reduced to 8.
* mmap() defaults to sv48-sized addresses, with longer addresses hidden
behind a hint (similar to Arm and Intel).
* Also various fixes and cleanups.
-----BEGIN PGP SIGNATURE-----
iQJHBAABCAAxFiEEKzw3R0RoQ7JKlDp6LhMZ81+7GIkFAmTx96kTHHBhbG1lckBk
YWJiZWx0LmNvbQAKCRAuExnzX7sYiVjRD/9DYVLlkQ/OEDJjPaEcYCP49xgIVUUU
lhs3XbSs2VNHBaiG114f6Q0AaT/uNi+uqSej3CeTmEot2kZkBk/f2yu+UNIriPZ9
GQiZsdyXhu921C+5VFtiI47KDWOVZ+Jpy3M1ll61IWt3yPSQHr1xOP0AOiyHHqe3
cmqpNnzjajlfVDoXPc2mGGzUJt/7ar4thcwnMNi98raXR5Qh7SP6rrHjoQhE1oFk
LMP3CHqEAcHE2tE4CxZVpc6HOQ5m0LpQIOK7ypufGMyoIYESm5dt/JOT4MlhTtDw
6JzyVKtiM7lartUnUaW3ZoX4trQYT5gbXxWrJ2gCnUGy3VulikoXr1Rpz0qfdeOR
XN8OLkVAqHfTGFI7oKk24f9Adw96R5NPZcdCay90h4J/kMfCiC7ZyUUI1XIa5iy1
np5pZCkf8HNcdywML7qcFd5n2O0wchyFnRLFZo6kJP9Ls5cEi6kBx/1jSdTcNgx/
fUKXyoEcriGoQiiwn29+4RZnU69gJV3zqQNLPpuwDQ5F/Q1zHTlrr+dqzezKkzcO
dRTV2d2Q4A5vIDXPptzNNLlRQdrc8qxPJ1lxQVkPIU4/mtqczmZBwlyY2u9zwPyS
sehJgJZnoAf+jm71NgQAKLck4MUBsMnMogOWunhXkVRCoZlbbkUWX4ECZYwPKsVk
W7zVPmLvSM0l5g==
=/tXb
-----END PGP SIGNATURE-----
Merge tag 'riscv-for-linus-6.6-mw1' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux
Pull RISC-V updates from Palmer Dabbelt:
- Support for the new "riscv,isa-extensions" and "riscv,isa-base"
device tree interfaces for probing extensions
- Support for userspace access to the performance counters
- Support for more instructions in kprobes
- Crash kernels can be allocated above 4GiB
- Support for KCFI
- Support for ELFs in !MMU configurations
- ARCH_KMALLOC_MINALIGN has been reduced to 8
- mmap() defaults to sv48-sized addresses, with longer addresses hidden
behind a hint (similar to Arm and Intel)
- Also various fixes and cleanups
* tag 'riscv-for-linus-6.6-mw1' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux: (51 commits)
lib/Kconfig.debug: Restrict DEBUG_INFO_SPLIT for RISC-V
riscv: support PREEMPT_DYNAMIC with static keys
riscv: Move create_tmp_mapping() to init sections
riscv: Mark KASAN tmp* page tables variables as static
riscv: mm: use bitmap_zero() API
riscv: enable DEBUG_FORCE_FUNCTION_ALIGN_64B
riscv: remove redundant mv instructions
RISC-V: mm: Document mmap changes
RISC-V: mm: Update pgtable comment documentation
RISC-V: mm: Add tests for RISC-V mm
RISC-V: mm: Restrict address space for sv39,sv48,sv57
riscv: enable DMA_BOUNCE_UNALIGNED_KMALLOC for !dma_coherent
riscv: allow kmalloc() caches aligned to the smallest value
riscv: support the elf-fdpic binfmt loader
binfmt_elf_fdpic: support 64-bit systems
riscv: Allow CONFIG_CFI_CLANG to be selected
riscv/purgatory: Disable CFI
riscv: Add CFI error handling
riscv: Add ftrace_stub_graph
riscv: Add types to indirectly called assembly functions
...
Jisheng Zhang <jszhang@kernel.org> says:
Currently, riscv defines ARCH_DMA_MINALIGN as L1_CACHE_BYTES, I.E
64Bytes, if CONFIG_RISCV_DMA_NONCOHERENT=y. To support unified kernel
Image, usually we have to enable CONFIG_RISCV_DMA_NONCOHERENT, thus
it brings some bad effects to coherent platforms:
Firstly, it wastes memory, kmalloc-96, kmalloc-32, kmalloc-16 and
kmalloc-8 slab caches don't exist any more, they are replaced with
either kmalloc-128 or kmalloc-64.
Secondly, larger than necessary kmalloc aligned allocations results
in unnecessary cache/TLB pressure.
This issue also exists on arm64 platforms. From last year, Catalin
tried to solve this issue by decoupling ARCH_KMALLOC_MINALIGN from
ARCH_DMA_MINALIGN, limiting kmalloc() minimum alignment to
dma_get_cache_alignment() and replacing ARCH_KMALLOC_MINALIGN usage
in various drivers with ARCH_DMA_MINALIGN etc.[1]
One fact we can make use of for riscv: if the CPU doesn't support
ZICBOM or T-HEAD CMO, we know the platform is coherent. Based on
Catalin's work and above fact, we can easily solve the kmalloc align
issue for riscv: we can override dma_get_cache_alignment(), then let
it return ARCH_DMA_MINALIGN at the beginning and return 1 once we know
the underlying HW neither supports ZICBOM nor supports T-HEAD CMO.
So what about if the CPU supports ZICBOM or T-HEAD CMO, but all the
devices are dma coherent? Well, we use ARCH_DMA_MINALIGN as the
kmalloc minimum alignment, nothing changed in this case. This case
can be improved in the future once we see such platforms in mainline.
After this patch, a simple test of booting to a small buildroot rootfs
on qemu shows:
kmalloc-96 5041 5041 96 ...
kmalloc-64 9606 9606 64 ...
kmalloc-32 5128 5128 32 ...
kmalloc-16 7682 7682 16 ...
kmalloc-8 10246 10246 8 ...
So we save about 1268KB memory. The saving will be much larger in normal
OS env on real HW platforms.
patch1 allows kmalloc() caches aligned to the smallest value.
patch2 enables DMA_BOUNCE_UNALIGNED_KMALLOC.
After this series:
As for coherent platforms, kmalloc-{8,16,32,96} caches come back on
coherent both RV32 and RV64 platforms, I.E !ZICBOM and !THEAD_CMO.
As for noncoherent RV32 platforms, nothing changed.
As for noncoherent RV64 platforms, I.E either ZICBOM or THEAD_CMO, the
above kmalloc caches also come back if > 4GB memory or users pass
"swiotlb=mmnn,force" to force swiotlb creation if <= 4GB memory. How
much mmnn should be depends on the specific platform, it needs to be
tried and tested all possible usage case on the specific hardware. For
example, I can use the minimal I/O TLB slabs on Sipeed M1S Dock.
* b4-shazam-merge:
riscv: enable DMA_BOUNCE_UNALIGNED_KMALLOC for !dma_coherent
riscv: allow kmalloc() caches aligned to the smallest value
Link: https://lore.kernel.org/linux-arm-kernel/20230524171904.3967031-1-catalin.marinas@arm.com/ [1]
Link: https://lore.kernel.org/r/20230718152214.2907-1-jszhang@kernel.org
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
Sami Tolvanen <samitolvanen@google.com> says:
This series adds KCFI support for RISC-V. KCFI is a fine-grained
forward-edge control-flow integrity scheme supported in Clang >=16,
which ensures indirect calls in instrumented code can only branch to
functions whose type matches the function pointer type, thus making
code reuse attacks more difficult.
Patch 1 implements a pt_regs based syscall wrapper to address
function pointer type mismatches in syscall handling. Patches 2 and 3
annotate indirectly called assembly functions with CFI types. Patch 4
implements error handling for indirect call checks. Patch 5 disables
CFI for arch/riscv/purgatory. Patch 6 finally allows CONFIG_CFI_CLANG
to be enabled for RISC-V.
Note that Clang 16 has a generic architecture-agnostic KCFI
implementation, which does work with the kernel, but doesn't produce
a stable code sequence for indirect call checks, which means
potential failures just trap and won't result in informative error
messages. Clang 17 includes a RISC-V specific back-end implementation
for KCFI, which emits a predictable code sequence for the checks and a
.kcfi_traps section with locations of the traps, which patch 5 uses to
produce more useful errors.
The type mismatch fixes and annotations in the first three patches
also become necessary in future if the kernel decides to support
fine-grained CFI implemented using the hardware landing pad
feature proposed in the in-progress Zicfisslp extension. Once the
specification is ratified and hardware support emerges, implementing
runtime patching support that replaces KCFI instrumentation with
Zicfisslp landing pads might also be feasible (similarly to KCFI to
FineIBT patching on x86_64), allowing distributions to ship a unified
kernel binary for all devices.
* b4-shazam-merge:
riscv: Allow CONFIG_CFI_CLANG to be selected
riscv/purgatory: Disable CFI
riscv: Add CFI error handling
riscv: Add ftrace_stub_graph
riscv: Add types to indirectly called assembly functions
riscv: Implement syscall wrappers
Link: https://lore.kernel.org/r/20230710183544.999540-8-samitolvanen@google.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
Chen Jiahao <chenjiahao16@huawei.com> says:
On riscv, the current crash kernel allocation logic is trying to
allocate within 32bit addressible memory region by default, if
failed, try to allocate without 4G restriction.
In need of saving DMA zone memory while allocating a relatively large
crash kernel region, allocating the reserved memory top down in
high memory, without overlapping the DMA zone, is a mature solution.
Hence this patchset introduces the parameter option crashkernel=X,[high,low].
One can reserve the crash kernel from high memory above DMA zone range
by explicitly passing "crashkernel=X,high"; or reserve a memory range
below 4G with "crashkernel=X,low". Besides, there are few rules need
to take notice:
1. "crashkernel=X,[high,low]" will be ignored if "crashkernel=size"
is specified.
2. "crashkernel=X,low" is valid only when "crashkernel=X,high" is passed
and there is enough memory to be allocated under 4G.
3. When allocating crashkernel above 4G and no "crashkernel=X,low" is
specified, a 128M low memory will be allocated automatically for
swiotlb bounce buffer.
See Documentation/admin-guide/kernel-parameters.txt for more information.
To verify loading the crashkernel, adapted kexec-tools is attached below:
https://github.com/chenjh005/kexec-tools/tree/build-test-riscv-v2
Following test cases have been performed as expected:
1) crashkernel=256M //low=256M
2) crashkernel=1G //low=1G
3) crashkernel=4G //high=4G, low=128M(default)
4) crashkernel=4G crashkernel=256M,high //high=4G, low=128M(default), high is ignored
5) crashkernel=4G crashkernel=256M,low //high=4G, low=128M(default), low is ignored
6) crashkernel=4G,high //high=4G, low=128M(default)
7) crashkernel=256M,low //low=0M, invalid
8) crashkernel=4G,high crashkernel=256M,low //high=4G, low=256M
9) crashkernel=4G,high crashkernel=4G,low //high=0M, low=0M, invalid
10) crashkernel=512M@0xd0000000 //low=512M
11) crashkernel=1G,high crashkernel=0M,low //high=1G, low=0M
* b4-shazam-merge:
docs: kdump: Update the crashkernel description for riscv
riscv: kdump: Implement crashkernel=X,[high,low]
Link: https://lore.kernel.org/r/20230726175000.2536220-1-chenjiahao16@huawei.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
Nam Cao <namcaov@gmail.com> says:
Simulate some currently rejected instructions. Still to be simulated are:
- c.jal
- c.ebreak
* b4-shazam-merge:
riscv: kprobes: simulate c.beqz and c.bnez
riscv: kprobes: simulate c.jr and c.jalr instructions
riscv: kprobes: simulate c.j instruction
Link: https://lore.kernel.org/r/cover.1690704360.git.namcaov@gmail.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
Some mv instructions were useful when first introduced to preserve a0 and
a1 before function calls. However the code has changed and they are now
redundant. Remove them.
Signed-off-by: Nam Cao <namcaov@gmail.com>
Reviewed-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Link: https://lore.kernel.org/r/20230725053835.138910-1-namcaov@gmail.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
These are the remaining few clean-ups of DT related includes which
didn't get applied to subsystem trees.
-----BEGIN PGP SIGNATURE-----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=eLMf
-----END PGP SIGNATURE-----
Merge tag 'devicetree-header-cleanups-for-6.6' of git://git.kernel.org/pub/scm/linux/kernel/git/robh/linux
Pull devicetree include cleanups from Rob Herring:
"These are the remaining few clean-ups of DT related includes which
didn't get applied to subsystem trees"
* tag 'devicetree-header-cleanups-for-6.6' of git://git.kernel.org/pub/scm/linux/kernel/git/robh/linux:
ipmi: Explicitly include correct DT includes
tpm: Explicitly include correct DT includes
lib/genalloc: Explicitly include correct DT includes
parport: Explicitly include correct DT includes
sbus: Explicitly include correct DT includes
mux: Explicitly include correct DT includes
macintosh: Explicitly include correct DT includes
hte: Explicitly include correct DT includes
EDAC: Explicitly include correct DT includes
clocksource: Explicitly include correct DT includes
sparc: Explicitly include correct DT includes
riscv: Explicitly include correct DT includes
("refactor Kconfig to consolidate KEXEC and CRASH options").
- kernel.h slimming work from Andy Shevchenko ("kernel.h: Split out a
couple of macros to args.h").
- gdb feature work from Kuan-Ying Lee ("Add GDB memory helper
commands").
- vsprintf inclusion rationalization from Andy Shevchenko
("lib/vsprintf: Rework header inclusions").
- Switch the handling of kdump from a udev scheme to in-kernel handling,
by Eric DeVolder ("crash: Kernel handling of CPU and memory hot
un/plug").
- Many singleton patches to various parts of the tree
-----BEGIN PGP SIGNATURE-----
iHUEABYIAB0WIQTTMBEPP41GrTpTJgfdBJ7gKXxAjgUCZO2GpAAKCRDdBJ7gKXxA
juW3AQD1moHzlSN6x9I3tjm5TWWNYFoFL8af7wXDJspp/DWH/AD/TO0XlWWhhbYy
QHy7lL0Syha38kKLMXTM+bN6YQHi9AU=
=WJQa
-----END PGP SIGNATURE-----
Merge tag 'mm-nonmm-stable-2023-08-28-22-48' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
Pull non-MM updates from Andrew Morton:
- An extensive rework of kexec and crash Kconfig from Eric DeVolder
("refactor Kconfig to consolidate KEXEC and CRASH options")
- kernel.h slimming work from Andy Shevchenko ("kernel.h: Split out a
couple of macros to args.h")
- gdb feature work from Kuan-Ying Lee ("Add GDB memory helper
commands")
- vsprintf inclusion rationalization from Andy Shevchenko
("lib/vsprintf: Rework header inclusions")
- Switch the handling of kdump from a udev scheme to in-kernel
handling, by Eric DeVolder ("crash: Kernel handling of CPU and memory
hot un/plug")
- Many singleton patches to various parts of the tree
* tag 'mm-nonmm-stable-2023-08-28-22-48' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm: (81 commits)
document while_each_thread(), change first_tid() to use for_each_thread()
drivers/char/mem.c: shrink character device's devlist[] array
x86/crash: optimize CPU changes
crash: change crash_prepare_elf64_headers() to for_each_possible_cpu()
crash: hotplug support for kexec_load()
x86/crash: add x86 crash hotplug support
crash: memory and CPU hotplug sysfs attributes
kexec: exclude elfcorehdr from the segment digest
crash: add generic infrastructure for crash hotplug support
crash: move a few code bits to setup support of crash hotplug
kstrtox: consistently use _tolower()
kill do_each_thread()
nilfs2: fix WARNING in mark_buffer_dirty due to discarded buffer reuse
scripts/bloat-o-meter: count weak symbol sizes
treewide: drop CONFIG_EMBEDDED
lockdep: fix static memory detection even more
lib/vsprintf: declare no_hash_pointers in sprintf.h
lib/vsprintf: split out sprintf() and friends
kernel/fork: stop playing lockless games for exe_file replacement
adfs: delete unused "union adfs_dirtail" definition
...
The DT of_device.h and of_platform.h date back to the separate
of_platform_bus_type before it was merged into the regular platform bus.
As part of that merge prepping Arm DT support 13 years ago, they
"temporarily" include each other. They also include platform_device.h
and of.h. As a result, there's a pretty much random mix of those include
files used throughout the tree. In order to detangle these headers and
replace the implicit includes with struct declarations, users need to
explicitly include the correct includes.
Link: https://lore.kernel.org/r/20230714174043.4040561-1-robh@kernel.org
Signed-off-by: Rob Herring <robh@kernel.org>
Currently, riscv defines ARCH_DMA_MINALIGN as L1_CACHE_BYTES, I.E
64Bytes, if CONFIG_RISCV_DMA_NONCOHERENT=y. To support unified kernel
Image, usually we have to enable CONFIG_RISCV_DMA_NONCOHERENT, thus
it brings some bad effects to coherent platforms:
Firstly, it wastes memory, kmalloc-96, kmalloc-32, kmalloc-16 and
kmalloc-8 slab caches don't exist any more, they are replaced with
either kmalloc-128 or kmalloc-64.
Secondly, larger than necessary kmalloc aligned allocations results
in unnecessary cache/TLB pressure.
This issue also exists on arm64 platforms. From last year, Catalin
tried to solve this issue by decoupling ARCH_KMALLOC_MINALIGN from
ARCH_DMA_MINALIGN, limiting kmalloc() minimum alignment to
dma_get_cache_alignment() and replacing ARCH_KMALLOC_MINALIGN usage
in various drivers with ARCH_DMA_MINALIGN etc.[1]
One fact we can make use of for riscv: if the CPU doesn't support
ZICBOM or T-HEAD CMO, we know the platform is coherent. Based on
Catalin's work and above fact, we can easily solve the kmalloc align
issue for riscv: we can override dma_get_cache_alignment(), then let
it return ARCH_DMA_MINALIGN at the beginning and return 1 once we know
the underlying HW neither supports ZICBOM nor supports T-HEAD CMO.
So what about if the CPU supports ZICBOM or T-HEAD CMO, but all the
devices are dma coherent? Well, we use ARCH_DMA_MINALIGN as the
kmalloc minimum alignment, nothing changed in this case. This case
can be improved in the future.
After this patch, a simple test of booting to a small buildroot rootfs
on qemu shows:
kmalloc-96 5041 5041 96 ...
kmalloc-64 9606 9606 64 ...
kmalloc-32 5128 5128 32 ...
kmalloc-16 7682 7682 16 ...
kmalloc-8 10246 10246 8 ...
So we save about 1268KB memory. The saving will be much larger in normal
OS env on real HW platforms.
Link: https://lore.kernel.org/linux-arm-kernel/20230524171904.3967031-1-catalin.marinas@arm.com/ [1]
Signed-off-by: Jisheng Zhang <jszhang@kernel.org>
Reviewed-by: Conor Dooley <conor.dooley@microchip.com>
Link: https://lore.kernel.org/r/20230718152214.2907-2-jszhang@kernel.org
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
Commit 883bbbffa5 ("ftrace,kcfi: Separate ftrace_stub() and
ftrace_stub_graph()") added a separate ftrace_stub_graph function for
CFI_CLANG. Add the stub to fix FUNCTION_GRAPH_TRACER compatibility
with CFI.
Reviewed-by: Kees Cook <keescook@chromium.org>
Tested-by: Nathan Chancellor <nathan@kernel.org>
Signed-off-by: Sami Tolvanen <samitolvanen@google.com>
Link: https://lore.kernel.org/r/20230710183544.999540-11-samitolvanen@google.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
With CONFIG_CFI_CLANG, assembly functions indirectly called
from C code must be annotated with type identifiers to pass CFI
checking. Use the SYM_TYPED_START macro to add types to the
relevant functions.
Reviewed-by: Kees Cook <keescook@chromium.org>
Tested-by: Nathan Chancellor <nathan@kernel.org>
Signed-off-by: Sami Tolvanen <samitolvanen@google.com>
Link: https://lore.kernel.org/r/20230710183544.999540-10-samitolvanen@google.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
Commit f0bddf5058 ("riscv: entry: Convert to generic entry") moved
syscall handling to C code, which exposed function pointer type
mismatches that trip fine-grained forward-edge Control-Flow Integrity
(CFI) checks as syscall handlers are all called through the same
syscall_t pointer type. To fix the type mismatches, implement pt_regs
based syscall wrappers similarly to x86 and arm64.
This patch is based on arm64 syscall wrappers added in commit
4378a7d4be ("arm64: implement syscall wrappers"), where the main goal
was to minimize the risk of userspace-controlled values being used
under speculation. This may be a concern for riscv in future as well.
Following other architectures, the syscall wrappers generate three
functions for each syscall; __riscv_<compat_>sys_<name> takes a pt_regs
pointer and extracts arguments from registers, __se_<compat_>sys_<name>
is a sign-extension wrapper that casts the long arguments to the
correct types for the real syscall implementation, which is named
__do_<compat_>sys_<name>.
Reviewed-by: Kees Cook <keescook@chromium.org>
Tested-by: Nathan Chancellor <nathan@kernel.org>
Signed-off-by: Sami Tolvanen <samitolvanen@google.com>
Link: https://lore.kernel.org/r/20230710183544.999540-9-samitolvanen@google.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
Andy Chiu <andy.chiu@sifive.com> says:
We add a vlenb field in Vector context and save it with the
riscv_vstate_save() macro. It should not cause performance regression as
VLENB is a design-time constant and is frequently used by hardware.
Also, adding this field into the __sc_riscv_v_state may benifit us on a
future compatibility issue becuse a hardware may have writable VLENB.
Adding and saving VLENB have an immediate benifit as it gives ptrace a
better view of the Vector extension and makes it possible to reconstruct
Vector register files from the dump without doing an additional csr read.
This patchset also sync the number of note types between us and gdb for
riscv to solve a conflicting note.
This is not an ABI break given that 6.5 has not been released yet.
* b4-shazam-merge:
RISC-V: vector: export VLENB csr in __sc_riscv_v_state
RISC-V: Remove ptrace support for vectors
Link: https://lore.kernel.org/r/20230816155450.26200-1-andy.chiu@sifive.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
We've found two bugs here: NT_RISCV_VECTOR steps on NT_RISCV_CSR (which
is only for embedded), and we don't have vlenb in the core dumps. Given
that we've have a pair of bugs croup up as part of the GDB review we've
probably got other issues, so let's just cut this for 6.5 and get it
right.
Fixes: 0c59922c76 ("riscv: Add ptrace vector support")
Reviewed-by: Maciej W. Rozycki <macro@orcam.me.uk>
Signed-off-by: Andy Chiu <andy.chiu@sifive.com>
Link: https://lore.kernel.org/r/20230816155450.26200-2-andy.chiu@sifive.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
The Kconfig refactor to consolidate KEXEC and CRASH options utilized
option names of the form ARCH_SUPPORTS_<option>. Thus rename the
ARCH_HAS_KEXEC_PURGATORY to ARCH_SUPPORTS_KEXEC_PURGATORY to follow
the same.
Link: https://lkml.kernel.org/r/20230712161545.87870-15-eric.devolder@oracle.com
Signed-off-by: Eric DeVolder <eric.devolder@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
On riscv, the current crash kernel allocation logic is trying to
allocate within 32bit addressible memory region by default, if
failed, try to allocate without 4G restriction.
In need of saving DMA zone memory while allocating a relatively large
crash kernel region, allocating the reserved memory top down in
high memory, without overlapping the DMA zone, is a mature solution.
Here introduce the parameter option crashkernel=X,[high,low].
One can reserve the crash kernel from high memory above DMA zone range
by explicitly passing "crashkernel=X,high"; or reserve a memory range
below 4G with "crashkernel=X,low".
Signed-off-by: Chen Jiahao <chenjiahao16@huawei.com>
Acked-by: Guo Ren <guoren@kernel.org>
Acked-by: Baoquan He <bhe@redhat.com>
Link: https://lore.kernel.org/r/20230726175000.2536220-2-chenjiahao16@huawei.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
The independent softirq stack uses s0 to save & restore sp, but s0 would
be corrupted when CONFIG_FRAME_POINTER=n. So add s0 in the clobber list
to fix the problem.
Fixes: dd69d07a5a ("riscv: stack: Support HAVE_SOFTIRQ_ON_OWN_STACK")
Cc: stable@vger.kernel.org
Reported-by: Zhangjin Wu <falcon@tinylab.org>
Signed-off-by: Guo Ren <guoren@linux.alibaba.com>
Signed-off-by: Guo Ren <guoren@kernel.org>
Tested-by: Drew Fustini <dfustini@baylibre.com>
Link: https://lore.kernel.org/r/20230716001506.3506041-3-guoren@kernel.org
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
The independent irq stack uses s0 to save & restore sp, but s0 would be
corrupted when CONFIG_FRAME_POINTER=n. So add s0 in the clobber list to
fix the problem.
Fixes: 163e76cc6e ("riscv: stack: Support HAVE_IRQ_EXIT_ON_IRQ_STACK")
Cc: stable@vger.kernel.org
Reported-by: Zhangjin Wu <falcon@tinylab.org>
Signed-off-by: Guo Ren <guoren@linux.alibaba.com>
Signed-off-by: Guo Ren <guoren@kernel.org>
Tested-by: Drew Fustini <dfustini@baylibre.com>
Link: https://lore.kernel.org/r/20230716001506.3506041-2-guoren@kernel.org
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
When we test seccomp with 6.4 kernel, we found errno has wrong value.
If we deny NETLINK_AUDIT with EAFNOSUPPORT, after f0bddf5058, we will
get ENOSYS instead. We got same result with commit 9c2598d435 ("riscv:
entry: Save a0 prior syscall_enter_from_user_mode()").
After analysing code, we think that regs->a0 = -ENOSYS should only be
executed when syscall != -1. In __seccomp_filter, when seccomp rejected
this syscall with specified errno, they will set a0 to return number as
syscall ABI, and then return -1. This return number is finally pass as
return number of syscall_enter_from_user_mode, and then is compared with
NR_syscalls after converted to ulong (so it will be ULONG_MAX). The
condition syscall < NR_syscalls will always be false, so regs->a0 = -ENOSYS
is always executed. It covered a0 set by seccomp, so we always get
ENOSYS when match seccomp RET_ERRNO rule.
Fixes: f0bddf5058 ("riscv: entry: Convert to generic entry")
Reported-by: Felix Yan <felixonmars@archlinux.org>
Co-developed-by: Ruizhe Pan <c141028@gmail.com>
Signed-off-by: Ruizhe Pan <c141028@gmail.com>
Co-developed-by: Shiqi Zhang <shiqi@isrc.iscas.ac.cn>
Signed-off-by: Shiqi Zhang <shiqi@isrc.iscas.ac.cn>
Signed-off-by: Celeste Liu <CoelacanthusHex@gmail.com>
Tested-by: Felix Yan <felixonmars@archlinux.org>
Tested-by: Emil Renner Berthing <emil.renner.berthing@canonical.com>
Reviewed-by: Björn Töpel <bjorn@rivosinc.com>
Reviewed-by: Guo Ren <guoren@kernel.org>
Link: https://lore.kernel.org/r/20230801141607.435192-1-CoelacanthusHex@gmail.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
commit 914d6f44fc ("RISC-V: only iterate over possible CPUs in ISA
string parser") changed riscv_fill_hwcap() from iterating over CPU DT
nodes to iterating over logical CPU IDs. Since this function runs long
before cpu_dev_init() creates CPU devices, it hits the fallback path in
of_cpu_device_node_get(), which itself iterates over the DT nodes,
searching for a node with the requested CPU ID. (Incidentally, this
makes riscv_fill_hwcap() now take quadratic time.)
riscv_fill_hwcap() passes a logical CPU ID to of_cpu_device_node_get(),
which uses the arch_match_cpu_phys_id() hook to translate the logical ID
to a physical ID as found in the DT.
arch_match_cpu_phys_id() has a generic weak definition, and RISC-V
provides a strong definition using cpuid_to_hartid_map(). However, the
RISC-V specific implementation is located in arch/riscv/kernel/smp.c,
and that file is only compiled when SMP is enabled.
As a result, when SMP is disabled, the generic definition is used, and
riscv_isa gets initialized based on the ISA string of hart 0, not the
boot hart. On FU740, this means has_fpu() returns false, and userspace
crashes when trying to use floating-point instructions.
Fix this by moving arch_match_cpu_phys_id() to a file which is always
compiled.
Fixes: 70114560b2 ("RISC-V: Add RISC-V specific arch_match_cpu_phys_id")
Fixes: 914d6f44fc ("RISC-V: only iterate over possible CPUs in ISA string parser")
Reported-by: Palmer Dabbelt <palmer@rivosinc.com>
Signed-off-by: Samuel Holland <samuel.holland@sifive.com>
Reviewed-by: Conor Dooley <conor.dooley@microchip.com>
Link: https://lore.kernel.org/r/20230803012608.3540081-1-samuel.holland@sifive.com
Cc: stable@vger.kernel.org
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
Petr Tesarik <petrtesarik@huaweicloud.com> says:
From: Petr Tesarik <petr.tesarik.ext@huawei.com>
The kexec_file_load(2) syscall does not work at least in some kernel
builds. For details see the relevant section in this blog post:
https://sigillatum.tesarici.cz/2023-07-21-state-of-riscv64-kdump.html
This patch series handles an additional relocation types, removes the need
to implement a Global Offset Table (GOT) for the purgatory and fixes the
placement of initrd.
* b4-shazam-merge:
riscv/kexec: load initrd high in available memory
riscv/kexec: handle R_RISCV_CALL_PLT relocation type
Link: https://lore.kernel.org/r/cover.1690365011.git.petr.tesarik.ext@huawei.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
When initrd is loaded low, the secondary kernel fails like this:
INITRD: 0xdc581000+0x00eef000 overlaps in-use memory region
This initrd load address corresponds to the _end symbol, but the
reservation is aligned on PMD_SIZE, as explained by a comment in
setup_bootmem().
It is technically possible to align the initrd load address accordingly,
leaving a hole between the end of kernel and the initrd, but it is much
simpler to allocate the initrd top-down.
Fixes: 838b3e2848 ("RISC-V: Load purgatory in kexec_file")
Signed-off-by: Torsten Duwe <duwe@suse.de>
Signed-off-by: Petr Tesarik <petr.tesarik.ext@huawei.com>
Cc: stable@vger.kernel.org
Reviewed-by: Conor Dooley <conor.dooley@microchip.com>
Link: https://lore.kernel.org/all/67c8eb9eea25717c2c8208d9bfbfaa39e6e2a1c6.1690365011.git.petr.tesarik.ext@huawei.com/
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
R_RISCV_CALL has been deprecated and replaced by R_RISCV_CALL_PLT. See Enum
18-19 in Table 3. Relocation types here:
https://github.com/riscv-non-isa/riscv-elf-psabi-doc/blob/master/riscv-elf.adoc
It was deprecated in ("Deprecated R_RISCV_CALL, prefer R_RISCV_CALL_PLT"):
a0dced8501
Recent tools (at least GNU binutils-2.40) already use R_RISCV_CALL_PLT.
Kernels built with such binutils fail kexec_load_file(2) with:
kexec_image: Unknown rela relocation: 19
kexec_image: Error loading purgatory ret=-8
The binary code at the call site remains the same, so tell
arch_kexec_apply_relocations_add() to handle _PLT alike.
Fixes: 838b3e2848 ("RISC-V: Load purgatory in kexec_file")
Signed-off-by: Torsten Duwe <duwe@suse.de>
Signed-off-by: Petr Tesarik <petr.tesarik.ext@huawei.com>
Cc: Li Zhengyu <lizhengyu3@huawei.com>
Cc: stable@vger.kernel.org
Reviewed-by: Conor Dooley <conor.dooley@microchip.com>
Link: https://lore.kernel.org/all/b046b164af8efd33bbdb7d4003273bdf9196a5b0.1690365011.git.petr.tesarik.ext@huawei.com/
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
Since RISC-V Linux v6.4, the commit 3335068f87 ("riscv: Use
PUD/P4D/PGD pages for the linear mapping") changes phys_ram_base
from the physical start of the kernel to the actual start of the DRAM.
The Crash-utility's VTOP() still uses phys_ram_base and kernel_map.virt_addr
to translate kernel virtual address, that failed the Crash with Linux v6.4 [1].
Export kernel_map.va_kernel_pa_offset in vmcoreinfo to help Crash translate
the kernel virtual address correctly.
Fixes: 3335068f87 ("riscv: Use PUD/P4D/PGD pages for the linear mapping")
Link: https://lore.kernel.org/linux-riscv/20230724040649.220279-1-suagrfillet@gmail.com/ [1]
Signed-off-by: Song Shuai <suagrfillet@gmail.com>
Reviewed-by: Xianting Tian <xianting.tian@linux.alibaba.com>
Reviewed-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Link: https://lore.kernel.org/r/20230724100917.309061-1-suagrfillet@gmail.com
Cc: stable@vger.kernel.org
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
acpi_os_ioremap() currently is a wrapper to memremap() on
RISC-V. But the callers of acpi_os_ioremap() expect it to
return __iomem address and hence sparse tool reports a new
warning. Fix this issue by type casting to __iomem type.
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202307230357.egcTAefj-lkp@intel.com/
Fixes: a91a9ffbd3 ("RISC-V: Add support to build the ACPI core")
Signed-off-by: Sunil V L <sunilvl@ventanamicro.com>
Reviewed-by: Conor Dooley <conor.dooley@microchip.com>
Link: https://lore.kernel.org/r/20230724100346.1302937-1-sunilvl@ventanamicro.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
As it says on the tin, provide Kconfig option to control parsing the
"riscv,isa" devicetree property. If either option is used, the kernel
will fall back to parsing "riscv,isa", where "riscv,isa-base" and
"riscv,isa-extensions" are not present.
The Kconfig options are set up so that the default kernel configuration
will enable the fallback path, without needing the commandline option.
Suggested-by: Andrew Jones <ajones@ventanamicro.com>
Suggested-by: Palmer Dabbelt <palmer@rivosinc.com>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Signed-off-by: Conor Dooley <conor.dooley@microchip.com>
Link: https://lore.kernel.org/r/20230713-aviator-plausibly-a35662485c2c@wendy
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
To fully deprecate the kernel's use of "riscv,isa",
of_early_processor_hartid() needs to first try using the new properties,
before falling back to "riscv,isa".
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Signed-off-by: Conor Dooley <conor.dooley@microchip.com>
Link: https://lore.kernel.org/r/20230713-tablet-jimmy-987fea0eb2e1@wendy
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
Add support for parsing the new riscv,isa-extensions property in
riscv_fill_hwcap(), by means of a new "property" member of the
riscv_isa_ext_data struct. For now, this shadows the name of the
extension for all users, however this may not be the case for all
extensions, based on how the dt-binding is written.
For the sake of backwards compatibility, fall back to the old scheme
if the new properties are not detected. For now, just inform, rather
than warn, when that happens.
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Signed-off-by: Conor Dooley <conor.dooley@microchip.com>
Link: https://lore.kernel.org/r/20230713-vocation-profane-39a74b3c2649@wendy
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
Before adding more complexity to it, split riscv_fill_hwcap() into 3
distinct sections:
- riscv_fill_hwcap() still is the top level function, into which the
additional complexity will be added.
- riscv_fill_hwcap_from_isa_string() handles getting the information
from the riscv,isa/ACPI equivalent across harts & the various quirks
there
- riscv_parse_isa_string() does what it says on the tin.
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Signed-off-by: Conor Dooley <conor.dooley@microchip.com>
Link: https://lore.kernel.org/r/20230713-daylight-puritan-37aeb41a4d9b@wendy
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
So that riscv_fill_hwcap() can use riscv_isa_ext to probe for single
letter extensions, add them to it.
As a result, what gets spat out in /proc/cpuinfo will become borked, as
single letter extensions will be printed as part of the base extensions
and while printing from riscv_isa_arr. Take the opportunity to unify the
printing of the isa string, using the new member of riscv_isa_ext_data
in the process.
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Signed-off-by: Conor Dooley <conor.dooley@microchip.com>
Link: https://lore.kernel.org/r/20230713-despite-bright-de00ac888cc7@wendy
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
In riscv_fill_hwcap() riscv_isa_ext array can be looped over, rather
than duplicating the list of extensions with individual
SET_ISA_EXT_MAP() usage. While at it, drop the statement-of-the-obvious
comments from the struct, rename uprop to something more suitable for
its new use & constify the members.
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Signed-off-by: Conor Dooley <conor.dooley@microchip.com>
Link: https://lore.kernel.org/r/20230713-dastardly-affiliate-4cf819dccde2@wendy
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
To facilitate using one struct to define extensions, rather than having
several, shunt isa_ext_arr to cpufeature.c, where it will be used for
probing extension presence also.
As that scope of the array as widened, prefix it with riscv & drop the
type from the variable name.
Since the new array is const, print_isa() needs a wee bit of cleanup to
avoid complaints about losing the const qualifier.
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Reviewed-by: Evan Green <evan@rivosinc.com>
Signed-off-by: Conor Dooley <conor.dooley@microchip.com>
Link: https://lore.kernel.org/r/20230713-spirits-upside-a2c61c65fd5a@wendy
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
isa_ext_arr cannot be empty, as some of the extensions within it are
always built into the kernel. When this code was first added, back in
commit a9b202606c ("RISC-V: Improve /proc/cpuinfo output for ISA
extensions"), the array was empty and needed a dummy item & thus there
could be no extensions present. When the first multi-letter ones did
get added, it was Sscofpmf - which didn't have a Kconfig symbol to
disable it.
Remove this check, as it has been redundant since Sscofpmf was added.
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Signed-off-by: Conor Dooley <conor.dooley@microchip.com>
Link: https://lore.kernel.org/r/20230713-veggie-mug-3d3bf6787ae2@wendy
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
When filling hwcap the kernel already expects the isa string to start with
rv32 if CONFIG_32BIT and rv64 if CONFIG_64BIT.
So when recreating the runtime isa-string we can also just go the other way
to get the correct starting point for it.
Signed-off-by: Heiko Stuebner <heiko.stuebner@vrull.eu>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Reviewed-by: Evan Green <evan@rivosinc.com>
Co-developed-by: Conor Dooley <conor.dooley@microchip.com>
Signed-off-by: Conor Dooley <conor.dooley@microchip.com>
Link: https://lore.kernel.org/r/20230713-masculine-saddlebag-67a94966b091@wendy
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
ACPI ISA strings are based on a specification after Zicsr and Zifencei
were split out of I, so we shouldn't be treating them as part of I. We
haven't release an ACPI-based kernel yet, so we don't need to worry
about compatibility with the old ISA strings.
Fixes: 07edc32779 ("RISC-V: always report presence of extensions formerly part of the base ISA")
Reviewed-by: Conor Dooley <conor.dooley@microchip.com>
Reviewed-by: Sunil V L <sunilvl@ventanamicro.com>
Link: https://lore.kernel.org/r/20230711224600.10879-1-palmer@rivosinc.com
Cc: stable@vger.kernel.org
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
* A bunch of fixes/cleanups from the first part of the merge window,
mostly related to ACPI and vector as those were large.
* Some documentation improvements, mostly related to the new code.
* The "riscv,isa" DT key is deprecated.
* Support for link-time dead code elimination.
* Support for minor fault registration in userfaultd.
* A handful of cleanups around CMO alternatives.
-----BEGIN PGP SIGNATURE-----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=jBYf
-----END PGP SIGNATURE-----
Merge tag 'riscv-for-linus-6.5-mw2' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux
Pull more RISC-V updates from Palmer Dabbelt:
- A bunch of fixes/cleanups from the first part of the merge window,
mostly related to ACPI and vector as those were large
- Some documentation improvements, mostly related to the new code
- The "riscv,isa" DT key is deprecated
- Support for link-time dead code elimination
- Support for minor fault registration in userfaultd
- A handful of cleanups around CMO alternatives
* tag 'riscv-for-linus-6.5-mw2' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux: (23 commits)
riscv: mm: mark noncoherent_supported as __ro_after_init
riscv: mm: mark CBO relate initialization funcs as __init
riscv: errata: thead: only set cbom size & noncoherent during boot
riscv: Select HAVE_ARCH_USERFAULTFD_MINOR
RISC-V: Document the ISA string parsing rules for ACPI
risc-v: Fix order of IPI enablement vs RCU startup
mm: riscv: fix an unsafe pte read in huge_pte_alloc()
dt-bindings: riscv: deprecate riscv,isa
RISC-V: drop error print from riscv_hartid_to_cpuid()
riscv: Discard vector state on syscalls
riscv: move memblock_allow_resize() after linear mapping is ready
riscv: Enable ARCH_SUSPEND_POSSIBLE for s2idle
riscv: vdso: include vdso/vsyscall.h for vdso_data
selftests: Test RISC-V Vector's first-use handler
riscv: vector: clear V-reg in the first-use trap
riscv: vector: only enable interrupts in the first-use trap
RISC-V: Fix up some vector state related build failures
RISC-V: Document that V registers are clobbered on syscalls
riscv: disable HAVE_LD_DEAD_CODE_DATA_ELIMINATION for LLD
riscv: enable HAVE_LD_DEAD_CODE_DATA_ELIMINATION
...
Conor reports that risc-v tries to enable IPIs before telling the
core code to enable RCU. With the introduction of the mapple tree
as a backing store for the irq descriptors, this results in
a very shouty boot sequence, as RCU is legitimately upset.
Restore some sanity by moving the risc_ipi_enable() call after
notify_cpu_starting(), which explicitly enables RCU on the calling
CPU.
Fixes: 832f15f426 ("RISC-V: Treat IPIs as normal Linux IRQs")
Reported-by: Conor Dooley <conor@kernel.org>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/20230703-dupe-frying-79ae2ccf94eb@spud
Cc: Anup Patel <apatel@ventanamicro.com>
Cc: Palmer Dabbelt <palmer@rivosinc.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Tested-by: Conor Dooley <conor.dooley@microchip.com>
Link: https://lore.kernel.org/r/20230703183126.1567625-1-maz@kernel.org
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
As of commit 2ac8743437 ("RISC-V: split early & late of_node to
hartid mapping") my CI complains about newly added pr_err() messages
during boot, for example:
[ 0.000000] Couldn't find cpu id for hartid [0]
[ 0.000000] riscv-intc: unable to find hart id for /cpus/cpu@0/interrupt-controller
Before the split, riscv_of_processor_hartid() contained a check for
whether the cpu was "available", before calling riscv_hartid_to_cpuid(),
but after the split riscv_of_processor_hartid() can be called for cpus
that are disabled.
Most callers of riscv_hartid_to_cpuid() already report custom errors
where it falls, making this print superfluous in those case. In other
places, the print adds nothing - see riscv_intc_init() for example.
Fixes: 2ac8743437 ("RISC-V: split early & late of_node to hartid mapping")
Signed-off-by: Conor Dooley <conor.dooley@microchip.com>
Link: https://lore.kernel.org/r/20230629-paternity-grafted-b901b76d04a0@wendy
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
The RISC-V vector specification states:
Executing a system call causes all caller-saved vector registers
(v0-v31, vl, vtype) and vstart to become unspecified.
The vector registers are set to all 1s, vill is set (invalid), and the
vector status is set to Dirty.
That way we can prevent userspace from accidentally relying on the
stated save.
Rémi pointed out [1] that writing to the registers might be
superfluous, and setting vill is sufficient.
Link: https://lore.kernel.org/linux-riscv/12784326.9UPPK3MAeB@basile.remlab.net/ # [1]
Suggested-by: Darius Rad <darius@bluespec.com>
Suggested-by: Palmer Dabbelt <palmer@rivosinc.com>
Suggested-by: Rémi Denis-Courmont <remi@remlab.net>
Signed-off-by: Björn Töpel <bjorn@rivosinc.com>
Link: https://lore.kernel.org/r/20230629142228.1125715-1-bjorn@kernel.org
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
Add include of <vdso/vsyscall.h> to pull in the defition of vdso_data
to remove the following sparse warning:
arch/riscv/kernel/vdso.c:39:18: warning: symbol 'vdso_data' was not declared. Should it be static?
Signed-off-by: Ben Dooks <ben.dooks@codethink.co.uk>
Link: https://lore.kernel.org/r/20230616114357.159601-1-ben.dooks@codethink.co.uk
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
If there is no context switch happens after we enable V for a process,
then we return to user space with whatever left on the CPU's V registers
accessible to the process. The leaked data could belong to another
process's V-context saved from last context switch, impacting process's
confidentiality on the system.
To prevent this from happening, we clear V registers by restoring
zero'd V context after turining on V.
Fixes: cd05483724 ("riscv: Allocate user's vector context in the first-use trap")
Signed-off-by: Andy Chiu <andy.chiu@sifive.com>
Reviewed-by: Björn Töpel <bjorn@rivosinc.com>
Link: https://lore.kernel.org/r/20230627015556.12329-2-andy.chiu@sifive.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
The function irqentry_exit_to_user_mode() must be called with interrupt
disabled. The caller of do_trap_insn_illegal() also assumes running
without interrupts. So, we should turn off interrupts after
riscv_v_first_use_handler() returns.
Fixes: cd05483724 ("riscv: Allocate user's vector context in the first-use trap")
Signed-off-by: Andy Chiu <andy.chiu@sifive.com>
Reviewed-by: Björn Töpel <bjorn@rivosinc.com>
Link: https://lore.kernel.org/r/20230625155416.18629-1-andy.chiu@sifive.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
Jisheng Zhang <jszhang@kernel.org> says:
When trying to run linux with various opensource riscv core on
resource limited FPGA platforms, for example, those FPGAs with less
than 16MB SDRAM, I want to save mem as much as possible. One of the
major technologies is kernel size optimizations, I found that riscv
does not currently support HAVE_LD_DEAD_CODE_DATA_ELIMINATION, which
passes -fdata-sections, -ffunction-sections to CFLAGS and passes the
--gc-sections flag to the linker.
This not only benefits my case on FPGA but also benefits defconfigs.
Here are some notable improvements from enabling this with defconfigs:
nommu_k210_defconfig:
text data bss dec hex
1112009 410288 59837 1582134 182436 before
962838 376656 51285 1390779 1538bb after
rv32_defconfig:
text data bss dec hex
8804455 2816544 290577 11911576 b5c198 before
8692295 2779872 288977 11761144 b375f8 after
defconfig:
text data bss dec hex
9438267 3391332 485333 13314932 cb2b74 before
9285914 3350052 483349 13119315 c82f53 after
patch1 and patch2 are clean ups.
patch3 fixes a typo.
patch4 finally enable HAVE_LD_DEAD_CODE_DATA_ELIMINATION for riscv.
* b4-shazam-merge:
riscv: disable HAVE_LD_DEAD_CODE_DATA_ELIMINATION for LLD
riscv: enable HAVE_LD_DEAD_CODE_DATA_ELIMINATION
vmlinux.lds.h: use correct .init.data.* section name
riscv: vmlinux-xip.lds.S: remove .alternative section
riscv: move options to keep entries sorted
riscv: Fix orphan section warnings caused by kernel/pi
Link: https://lore.kernel.org/r/20230523165502.2592-1-jszhang@kernel.org
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
- Add new feature to have function graph tracer record the return value.
Adds a new option: funcgraph-retval ; when set, will show the return
value of a function in the function graph tracer.
- Also add the option: funcgraph-retval-hex where if it is not set, and
the return value is an error code, then it will return the decimal of
the error code, otherwise it still reports the hex value.
- Add the file /sys/kernel/tracing/osnoise/per_cpu/cpu<cpu>/timerlat_fd
That when a application opens it, it becomes the task that the timer lat
tracer traces. The application can also read this file to find out how
it's being interrupted.
- Add the file /sys/kernel/tracing/available_filter_functions_addrs
that works just the same as available_filter_functions but also shows
the addresses of the functions like kallsyms, except that it gives the
address of where the fentry/mcount jump/nop is. This is used by BPF to
make it easier to attach BPF programs to ftrace hooks.
- Replace strlcpy with strscpy in the tracing boot code.
-----BEGIN PGP SIGNATURE-----
iIoEABYIADIWIQRRSw7ePDh/lE+zeZMp5XQQmuv6qgUCZJy6ixQccm9zdGVkdEBn
b29kbWlzLm9yZwAKCRAp5XQQmuv6qnzRAPsEI2YgjaJSHnuPoGRHbrNil6pq66wY
LYaLizGI4Jv9BwEAqdSdcYcMiWo1SFBAO8QxEDM++BX3zrRyVgW8ahaTNgs=
=TF0C
-----END PGP SIGNATURE-----
Merge tag 'trace-v6.5' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace
Pull tracing updates from Steven Rostedt:
- Add new feature to have function graph tracer record the return
value. Adds a new option: funcgraph-retval ; when set, will show the
return value of a function in the function graph tracer.
- Also add the option: funcgraph-retval-hex where if it is not set, and
the return value is an error code, then it will return the decimal of
the error code, otherwise it still reports the hex value.
- Add the file /sys/kernel/tracing/osnoise/per_cpu/cpu<cpu>/timerlat_fd
That when a application opens it, it becomes the task that the timer
lat tracer traces. The application can also read this file to find
out how it's being interrupted.
- Add the file /sys/kernel/tracing/available_filter_functions_addrs
that works just the same as available_filter_functions but also shows
the addresses of the functions like kallsyms, except that it gives
the address of where the fentry/mcount jump/nop is. This is used by
BPF to make it easier to attach BPF programs to ftrace hooks.
- Replace strlcpy with strscpy in the tracing boot code.
* tag 'trace-v6.5' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:
tracing: Fix warnings when building htmldocs for function graph retval
riscv: ftrace: Enable HAVE_FUNCTION_GRAPH_RETVAL
tracing/boot: Replace strlcpy with strscpy
tracing/timerlat: Add user-space interface
tracing/osnoise: Skip running osnoise if all instances are off
tracing/osnoise: Switch from PF_NO_SETAFFINITY to migrate_disable
ftrace: Show all functions with addresses in available_filter_functions_addrs
selftests/ftrace: Add funcgraph-retval test case
LoongArch: ftrace: Enable HAVE_FUNCTION_GRAPH_RETVAL
x86/ftrace: Enable HAVE_FUNCTION_GRAPH_RETVAL
arm64: ftrace: Enable HAVE_FUNCTION_GRAPH_RETVAL
tracing: Add documentation for funcgraph-retval and funcgraph-retval-hex
function_graph: Support recording and printing the return value of function
fgraph: Add declaration of "struct fgraph_ret_regs"
* Support for ACPI.
* Various cleanups to the ISA string parsing, including making them
case-insensitive
* Support for the vector extension.
* Support for independent irq/softirq stacks.
* Our CPU DT binding now has "unevaluatedProperties: false"
-----BEGIN PGP SIGNATURE-----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=Fpq8
-----END PGP SIGNATURE-----
Merge tag 'riscv-for-linus-6.5-mw1' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux
Pull RISC-V updates from Palmer Dabbelt:
- Support for ACPI
- Various cleanups to the ISA string parsing, including making them
case-insensitive
- Support for the vector extension
- Support for independent irq/softirq stacks
- Our CPU DT binding now has "unevaluatedProperties: false"
* tag 'riscv-for-linus-6.5-mw1' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux: (78 commits)
riscv: hibernate: remove WARN_ON in save_processor_state
dt-bindings: riscv: cpus: switch to unevaluatedProperties: false
dt-bindings: riscv: cpus: add a ref the common cpu schema
riscv: stack: Add config of thread stack size
riscv: stack: Support HAVE_SOFTIRQ_ON_OWN_STACK
riscv: stack: Support HAVE_IRQ_EXIT_ON_IRQ_STACK
RISC-V: always report presence of extensions formerly part of the base ISA
dt-bindings: riscv: explicitly mention assumption of Zicntr & Zihpm support
RISC-V: remove decrement/increment dance in ISA string parser
RISC-V: rework comments in ISA string parser
RISC-V: validate riscv,isa at boot, not during ISA string parsing
RISC-V: split early & late of_node to hartid mapping
RISC-V: simplify register width check in ISA string parsing
perf: RISC-V: Limit the number of counters returned from SBI
riscv: replace deprecated scall with ecall
riscv: uprobes: Restore thread.bad_cause
riscv: mm: try VMA lock-based page fault handling first
riscv: mm: Pre-allocate PGD entries for vmalloc/modules area
RISC-V: hwprobe: Expose Zba, Zbb, and Zbs
RISC-V: Track ISA extensions per hart
...
- Parallel CPU bringup
The reason why people are interested in parallel bringup is to shorten
the (kexec) reboot time of cloud servers to reduce the downtime of the
VM tenants.
The current fully serialized bringup does the following per AP:
1) Prepare callbacks (allocate, intialize, create threads)
2) Kick the AP alive (e.g. INIT/SIPI on x86)
3) Wait for the AP to report alive state
4) Let the AP continue through the atomic bringup
5) Let the AP run the threaded bringup to full online state
There are two significant delays:
#3 The time for an AP to report alive state in start_secondary() on
x86 has been measured in the range between 350us and 3.5ms
depending on vendor and CPU type, BIOS microcode size etc.
#4 The atomic bringup does the microcode update. This has been
measured to take up to ~8ms on the primary threads depending on
the microcode patch size to apply.
On a two socket SKL server with 56 cores (112 threads) the boot CPU
spends on current mainline about 800ms busy waiting for the APs to come
up and apply microcode. That's more than 80% of the actual onlining
procedure.
This can be reduced significantly by splitting the bringup mechanism
into two parts:
1) Run the prepare callbacks and kick the AP alive for each AP which
needs to be brought up.
The APs wake up, do their firmware initialization and run the low
level kernel startup code including microcode loading in parallel
up to the first synchronization point. (#1 and #2 above)
2) Run the rest of the bringup code strictly serialized per CPU
(#3 - #5 above) as it's done today.
Parallelizing that stage of the CPU bringup might be possible in
theory, but it's questionable whether required surgery would be
justified for a pretty small gain.
If the system is large enough the first AP is already waiting at the
first synchronization point when the boot CPU finished the wake-up of
the last AP. That reduces the AP bringup time on that SKL from ~800ms
to ~80ms, i.e. by a factor ~10x.
The actual gain varies wildly depending on the system, CPU, microcode
patch size and other factors. There are some opportunities to reduce
the overhead further, but that needs some deep surgery in the x86 CPU
bringup code.
For now this is only enabled on x86, but the core functionality
obviously works for all SMP capable architectures.
- Enhancements for SMP function call tracing so it is possible to locate
the scheduling and the actual execution points. That allows to measure
IPI delivery time precisely.
-----BEGIN PGP SIGNATURE-----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=Y8if
-----END PGP SIGNATURE-----
Merge tag 'smp-core-2023-06-26' of ssh://gitolite.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull SMP updates from Thomas Gleixner:
"A large update for SMP management:
- Parallel CPU bringup
The reason why people are interested in parallel bringup is to
shorten the (kexec) reboot time of cloud servers to reduce the
downtime of the VM tenants.
The current fully serialized bringup does the following per AP:
1) Prepare callbacks (allocate, intialize, create threads)
2) Kick the AP alive (e.g. INIT/SIPI on x86)
3) Wait for the AP to report alive state
4) Let the AP continue through the atomic bringup
5) Let the AP run the threaded bringup to full online state
There are two significant delays:
#3 The time for an AP to report alive state in start_secondary()
on x86 has been measured in the range between 350us and 3.5ms
depending on vendor and CPU type, BIOS microcode size etc.
#4 The atomic bringup does the microcode update. This has been
measured to take up to ~8ms on the primary threads depending
on the microcode patch size to apply.
On a two socket SKL server with 56 cores (112 threads) the boot CPU
spends on current mainline about 800ms busy waiting for the APs to
come up and apply microcode. That's more than 80% of the actual
onlining procedure.
This can be reduced significantly by splitting the bringup
mechanism into two parts:
1) Run the prepare callbacks and kick the AP alive for each AP
which needs to be brought up.
The APs wake up, do their firmware initialization and run the
low level kernel startup code including microcode loading in
parallel up to the first synchronization point. (#1 and #2
above)
2) Run the rest of the bringup code strictly serialized per CPU
(#3 - #5 above) as it's done today.
Parallelizing that stage of the CPU bringup might be possible
in theory, but it's questionable whether required surgery
would be justified for a pretty small gain.
If the system is large enough the first AP is already waiting at
the first synchronization point when the boot CPU finished the
wake-up of the last AP. That reduces the AP bringup time on that
SKL from ~800ms to ~80ms, i.e. by a factor ~10x.
The actual gain varies wildly depending on the system, CPU,
microcode patch size and other factors. There are some
opportunities to reduce the overhead further, but that needs some
deep surgery in the x86 CPU bringup code.
For now this is only enabled on x86, but the core functionality
obviously works for all SMP capable architectures.
- Enhancements for SMP function call tracing so it is possible to
locate the scheduling and the actual execution points. That allows
to measure IPI delivery time precisely"
* tag 'smp-core-2023-06-26' of ssh://gitolite.kernel.org/pub/scm/linux/kernel/git/tip/tip: (45 commits)
trace,smp: Add tracepoints for scheduling remotelly called functions
trace,smp: Add tracepoints around remotelly called functions
MAINTAINERS: Add CPU HOTPLUG entry
x86/smpboot: Fix the parallel bringup decision
x86/realmode: Make stack lock work in trampoline_compat()
x86/smp: Initialize cpu_primary_thread_mask late
cpu/hotplug: Fix off by one in cpuhp_bringup_mask()
x86/apic: Fix use of X{,2}APIC_ENABLE in asm with older binutils
x86/smpboot/64: Implement arch_cpuhp_init_parallel_bringup() and enable it
x86/smpboot: Support parallel startup of secondary CPUs
x86/smpboot: Implement a bit spinlock to protect the realmode stack
x86/apic: Save the APIC virtual base address
cpu/hotplug: Allow "parallel" bringup up to CPUHP_BP_KICK_AP_STATE
x86/apic: Provide cpu_primary_thread mask
x86/smpboot: Enable split CPU startup
cpu/hotplug: Provide a split up CPUHP_BRINGUP mechanism
cpu/hotplug: Reset task stack state in _cpu_up()
cpu/hotplug: Remove unused state functions
riscv: Switch to hotplug core state synchronization
parisc: Switch to hotplug core state synchronization
...
Select CONFIG_HAVE_LD_DEAD_CODE_DATA_ELIMINATION for RISC-V, allowing
the user to enable dead code elimination. In order for this to work,
ensure that we keep the alternative table by annotating them with KEEP.
This boots well on qemu with both rv32_defconfig & rv64 defconfig, but
it only shrinks their builds by ~1%, a smaller config is thereforce
customized to test this feature:
| rv32 | rv64
--------|------------------------|---------------------
No DCE | 4460684 | 4893488
DCE | 3986716 | 4376400
Shrink | 473968 (~10.6%) | 517088 (~10.5%)
The config used above only reserves necessary options to boot on qemu
with serial console, more like the size-critical embedded scenes:
- rv64 config: https://pastebin.com/crz82T0s
- rv32 config: rv64 config + 32-bit.config
Here is Jisheng's original commit-msg:
When trying to run linux with various opensource riscv core on
resource limited FPGA platforms, for example, those FPGAs with less
than 16MB SDRAM, I want to save mem as much as possible. One of the
major technologies is kernel size optimizations, I found that riscv
does not currently support HAVE_LD_DEAD_CODE_DATA_ELIMINATION, which
passes -fdata-sections, -ffunction-sections to CFLAGS and passes the
--gc-sections flag to the linker.
This not only benefits my case on FPGA but also benefits defconfigs.
Here are some notable improvements from enabling this with defconfigs:
nommu_k210_defconfig:
text data bss dec hex
1112009 410288 59837 1582134 182436 before
962838 376656 51285 1390779 1538bb after
rv32_defconfig:
text data bss dec hex
8804455 2816544 290577 11911576 b5c198 before
8692295 2779872 288977 11761144 b375f8 after
defconfig:
text data bss dec hex
9438267 3391332 485333 13314932 cb2b74 before
9285914 3350052 483349 13119315 c82f53 after
Signed-off-by: Zhangjin Wu <falcon@tinylab.org>
Co-developed-by: Jisheng Zhang <jszhang@kernel.org>
Signed-off-by: Jisheng Zhang <jszhang@kernel.org>
Reviewed-by: Guo Ren <guoren@kernel.org>
Tested-by: Bin Meng <bmeng@tinylab.org>
Reviewed-by: Kefeng Wang <wangkefeng.wang@huawei.com>
Tested-by: Nick Desaulniers <ndesaulniers@google.com> # build
Link: https://lore.kernel.org/r/20230523165502.2592-5-jszhang@kernel.org
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
ALTERNATIVE mechanism can't work on XIP, and this is also reflected by
below Kconfig dependency:
RISCV_ALTERNATIVE
...
depends on !XIP_KERNEL
...
So there's no .alternative section at all for XIP case, remove it.
Signed-off-by: Jisheng Zhang <jszhang@kernel.org>
Reviewed-by: Conor Dooley <conor.dooley@microchip.com>
Reviewed-by: Guo Ren <guoren@kernel.org>
Tested-by: Nick Desaulniers <ndesaulniers@google.com> # build
Link: https://lore.kernel.org/r/20230523165502.2592-3-jszhang@kernel.org
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
During hibernation or restoration, freeze_secondary_cpus
checks num_online_cpus via BUG_ON, and the subsequent
save_processor_state also does the checking with WARN_ON.
In the case of CONFIG_PM_SLEEP_SMP=n, freeze_secondary_cpus
is not defined, but the sole possible condition to disable
CONFIG_PM_SLEEP_SMP is !SMP where num_online_cpus is always 1.
We also don't have to check it in save_processor_state.
So remove the unnecessary checking in save_processor_state.
Fixes: c031721001 ("RISC-V: Add arch functions to support hibernation/suspend-to-disk")
Signed-off-by: Song Shuai <songshuaishuai@tinylab.org>
Reviewed-by: Conor Dooley <conor.dooley@microchip.com>
Link: https://lore.kernel.org/r/20230609075049.2651723-4-songshuaishuai@tinylab.org
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
guoren@kernel.org <guoren@kernel.org> says:
From: Guo Ren <guoren@linux.alibaba.com>
This patch series adds independent irq/softirq stacks to decrease the
press of the thread stack. Also, add a thread STACK_SIZE config for
users to adjust the proper size during compile time.
* b4-shazam-merge:
riscv: stack: Add config of thread stack size
riscv: stack: Support HAVE_SOFTIRQ_ON_OWN_STACK
riscv: stack: Support HAVE_IRQ_EXIT_ON_IRQ_STACK
Link: https://lore.kernel.org/r/20230614013018.2168426-1-guoren@kernel.org
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
Conor Dooley <conor@kernel.org> says:
From: Conor Dooley <conor.dooley@microchip.com>
Here are some bits that were discussed with Drew on the "should we
allow caps" threads that I have now created patches for:
- splitting of riscv_of_processor_hartid() into two distinct functions,
one for use purely during early boot, prior to the establishment of
the possible-cpus mask & another to fit the other current use-cases
- that then allows us to then completely skip some validation of the
hartid in the parser
- the biggest diff in the series is a rework of the comments in the
parser, as I have mostly found the existing (sparse) ones to not be
all that helpful whenever I have to go back and look at it
- from writing the comments, I found a conditional doing a bit of a
dance that I found counter-intuitive, so I've had a go at making that
match what I would expect a little better
- `i` implies 4 other extensions, so add them as extensions and set
them for the craic. Sure why not like...
* b4-shazam-merge:
RISC-V: always report presence of extensions formerly part of the base ISA
dt-bindings: riscv: explicitly mention assumption of Zicntr & Zihpm support
RISC-V: remove decrement/increment dance in ISA string parser
RISC-V: rework comments in ISA string parser
RISC-V: validate riscv,isa at boot, not during ISA string parsing
RISC-V: split early & late of_node to hartid mapping
RISC-V: simplify register width check in ISA string parsing
Link: https://lore.kernel.org/r/20230607-audacity-overhaul-82bb867a825f@spud
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
Add the HAVE_SOFTIRQ_ON_OWN_STACK feature for the IRQ_STACKS config, and
the irq and softirq use the same irq_stack of percpu.
Tested-by: Jisheng Zhang <jszhang@kernel.org>
Signed-off-by: Guo Ren <guoren@linux.alibaba.com>
Signed-off-by: Guo Ren <guoren@kernel.org>
Link: https://lore.kernel.org/r/20230614013018.2168426-3-guoren@kernel.org
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
The previous patch ("function_graph: Support recording and printing
the return value of function") has laid the groundwork for the for
the funcgraph-retval, and this modification makes it available on
the RISC-V platform.
We introduce a new structure called fgraph_ret_regs for the RISC-V
platform to hold return registers and the frame pointer. We then
fill its content in the return_to_handler and pass its address to
the function ftrace_return_to_handler to record the return value.
Link: https://lore.kernel.org/linux-trace-kernel/a8d71b12259f90e7e63d0ea654fcac95b0232bbc.1680954589.git.pengdonglin@sangfor.com.cn
Signed-off-by: Donglin Peng <pengdonglin@sangfor.com.cn>
Acked-by: Palmer Dabbelt <palmer@rivosinc.com>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Of these four extensions, two were part of the base ISA when the port was
written and are required by the kernel. The other two are implied when
`i` is in riscv,isa on DT systems.
There's not much that userspace can do with this extra information, but
there is no harm in reporting an ISA string that closer resembles the
current versions of the specifications either.
Signed-off-by: Conor Dooley <conor.dooley@microchip.com>
Link: https://lore.kernel.org/r/20230607-nest-collision-5796b6be8be6@spud
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
While expanding on the comments in the ISA string parsing code, I
noticed that the conditional decrement of `isa` at the end of the loop
was a bit odd.
The parsing code expects that at the start of the for loop, `isa` will
point to the first character of the next unparsed extension.
However, depending on what the next extension is, this may not be true.
Unless the next extension is a multi-letter extension preceded by an
underscore, `isa` will either point to the string's null-terminator or
to the first character of the next extension, once the switch statement
has been evaluated.
Obviously incrementing `isa` at the end of the loop could cause it to
increment past the null terminator or miss a single letter extension, so
`isa` is conditionally decremented, just so that the loop can increment
it again.
It's easier to understand the code if, instead of this decrement +
increment dance, we instead use a while loop & rely on the handling of
individual extension types to leave `isa` pointing to the first
character of the next extension.
As already mentioned, this won't be the case where the following
extension is multi-letter & preceded by an underscore. To handle that,
invert the check and increment rather than decrement.
Hopefully this eliminates a "huh?!?" moment the next time somebody tries
to understand this code.
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Signed-off-by: Conor Dooley <conor.dooley@microchip.com>
Reviewed-by: Sunil V L <sunilvl@ventanamicro.com>
Link: https://lore.kernel.org/r/20230607-estate-left-f20faabefb89@spud
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
Sami Tolvanen
Deepak Gupta
Sunil V L
Arnd Bergmann
Jiexun Wang
Chen Jiahao
Anup Patel
Mayuresh Chitale
Andy Chiu
Joel Granados
Costa Shulyupin
Andrew Jones
Björn Töpel
Song Shuai
Palmer Dabbelt
Puranjay Mohan
Alexandre Ghiti
Lad Prabhakar
Evan Green
Linus Torvalds
Nam Cao
Rob Herring
Jisheng Zhang
Eric DeVolder
Mingzheng Xing
Guo Ren
Celeste Liu
Samuel Holland
Torsten Duwe
Song Shuai
Justin Stitt
Conor Dooley
Heiko Stuebner
Marc Zyngier
Ben Dooks
Zhangjin Wu
Donglin Peng