linux/drivers/infiniband/hw/cxgb3
Steve Wise ec6995ddaa RDMA/cxgb3: Don't complete flushed send work requests twice
When the SQ is flushed, mark the flushed entries as not signaled so
the poll logic doesn't re-insert the CQ entry thinking its an out of
order completion.

The bug can cause the NFS/RDMA server to crash due to processing the
same completed work request twice.

Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
2009-04-29 15:15:59 -07:00
..
Kconfig
Makefile
cxio_dbg.c
cxio_hal.c RDMA/cxgb3: Don't complete flushed send work requests twice 2009-04-29 15:15:59 -07:00
cxio_hal.h RDMA/cxgb3: Handle EEH events 2009-03-30 08:37:56 -07:00
cxio_resource.c
cxio_resource.h
cxio_wr.h
iwch.c RDMA/cxgb3: Handle EEH events 2009-03-30 08:37:56 -07:00
iwch.h RDMA/cxgb3: Handle EEH events 2009-03-30 08:37:56 -07:00
iwch_cm.c RDMA/cxgb3: Adjust ORD/IRD (if needed) for peer2peer connections 2009-04-20 13:53:15 -07:00
iwch_cm.h RDMA/cxgb3: Release dependent resources only when endpoint memory is freed. 2009-03-30 08:37:59 -07:00
iwch_cq.c
iwch_ev.c
iwch_mem.c
iwch_provider.c
iwch_provider.h
iwch_qp.c RDMA/cxgb3: Don't zero QP attrs when moving to IDLE 2009-04-20 17:00:53 -07:00
iwch_user.h
tcb.h