linux/drivers/infiniband/hw/cxgb4
Steve Wise d9594d990a RDMA/cxgb4: Reset wait condition atomically
The driver was never really waiting for RDMA_WR/FINI completions
because the condition variable used to determine if the completion
happened was never reset, and this condition variable is reused for
both connection setup and teardown.  This causes various driver
crashes under heavy loads due to releasing resources too early.

The fix is to use atomic bits to correctly reset the condition
immediately after the completion is detected.

Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
2011-05-09 22:06:22 -07:00
..
cm.c RDMA/cxgb4: Reset wait condition atomically 2011-05-09 22:06:22 -07:00
cq.c RDMA/cxgb4: Centralize the wait logic 2010-09-28 10:46:34 -07:00
device.c RDMA/cxgb4: Initialization errors can cause crash 2011-05-09 22:06:22 -07:00
ev.c RDMA/cxgb4: Handle CPL_RDMA_TERMINATE messages 2010-09-28 10:46:32 -07:00
iw_cxgb4.h RDMA/cxgb4: Reset wait condition atomically 2011-05-09 22:06:22 -07:00
Kconfig Update broken web addresses in the kernel. 2010-10-18 11:03:14 +02:00
Makefile IB: Replace EXTRA_CFLAGS with ccflags-y 2010-10-23 13:45:03 -07:00
mem.c RDMA/cxgb4: Centralize the wait logic 2010-09-28 10:46:34 -07:00
provider.c RDMA/cxgb4: Export T4 TCP MIB 2010-10-22 21:57:26 -07:00
qp.c RDMA/cxgb4: Fix missing parentheses 2011-05-09 22:06:22 -07:00
resource.c RDMA/cxgb4: Support on-chip SQs 2010-09-28 10:46:35 -07:00
t4.h RDMA/cxgb4: Do CIDX_INC updates every 1/16 CQ depth CQE reaps 2011-03-14 12:09:11 -07:00
t4fw_ri_api.h RDMA/cxgb4: Add module option to tweak delayed ack 2010-07-21 10:53:52 -07:00
user.h RDMA/cxgb4: Support on-chip SQs 2010-09-28 10:46:35 -07:00