Jason Gunthorpe 071ba4cc55 RDMA: Add rdma_connect_locked() ... 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: 2a7cec5381 ("RDMA/cma: Fix locking for the RDMA_CM_CONNECT state") Acked-by: Santosh Shilimkar <santosh.shilimkar@oracle.com> Acked-by: Jack Wang <jinpu.wang@cloud.ionos.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Max Gurtovoy <mgurtovoy@nvidia.com> Reviewed-by: Sagi Grimberg <sagi@grimberg.me> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>		2020-10-28 09:14:49 -03:00
..
af_rds.c
bind.c
cong.c	net: rds: delete duplicated words	2020-09-18 14:12:43 -07:00
connection.c
ib_cm.c	RDMA: Add rdma_connect_locked()	2020-10-28 09:14:49 -03:00
ib_frmr.c
ib_mr.h
ib_rdma.c
ib_recv.c	net/rds: suppress page allocation failure error in recv buffer refill	2020-10-09 12:32:03 -07:00
ib_ring.c
ib_send.c
ib_stats.c
ib_sysctl.c
ib.c
ib.h
info.c
info.h
Kconfig
loop.c
loop.h
Makefile
message.c
page.c
rdma_transport.c
rdma_transport.h
rdma.c	RDS: drop double zeroing	2020-09-20 19:09:11 -07:00
rds_single_path.h
rds.h
recv.c
send.c
stats.c
sysctl.c
tcp_connect.c
tcp_listen.c
tcp_recv.c
tcp_send.c
tcp_stats.c
tcp.c
tcp.h
threads.c
transport.c