mirror of https://github.com/torvalds/linux.git
071ba4cc55
There are two flows for handling RDMA_CM_EVENT_ROUTE_RESOLVED, either the
handler triggers a completion and another thread does rdma_connect() or
the handler directly calls rdma_connect().
In all cases rdma_connect() needs to hold the handler_mutex, but when
handler's are invoked this is already held by the core code. This causes
ULPs using the 2nd method to deadlock.
Provide a rdma_connect_locked() and have all ULPs call it from their
handlers.
Link: https://lore.kernel.org/r/0-v2-53c22d5c1405+33-rdma_connect_locking_jgg@nvidia.com
Reported-and-tested-by: Guoqing Jiang <guoqing.jiang@cloud.ionos.com>
Fixes:
|
||
|---|---|---|
| .. | ||
| Kconfig | ||
| Makefile | ||
| af_rds.c | ||
| bind.c | ||
| cong.c | ||
| connection.c | ||
| ib.c | ||
| ib.h | ||
| ib_cm.c | ||
| ib_frmr.c | ||
| ib_mr.h | ||
| ib_rdma.c | ||
| ib_recv.c | ||
| ib_ring.c | ||
| ib_send.c | ||
| ib_stats.c | ||
| ib_sysctl.c | ||
| info.c | ||
| info.h | ||
| loop.c | ||
| loop.h | ||
| message.c | ||
| page.c | ||
| rdma.c | ||
| rdma_transport.c | ||
| rdma_transport.h | ||
| rds.h | ||
| rds_single_path.h | ||
| recv.c | ||
| send.c | ||
| stats.c | ||
| sysctl.c | ||
| tcp.c | ||
| tcp.h | ||
| tcp_connect.c | ||
| tcp_listen.c | ||
| tcp_recv.c | ||
| tcp_send.c | ||
| tcp_stats.c | ||
| threads.c | ||
| transport.c | ||