IB/rdmavt: Avoid reseting wqe send_flags in unreserve

The wqe should be read only and in fact the superfluous reset of the
RVT_SEND_RESERVE_USED flag causes an issue where reserved operations
elicit a bad completion to the ULP.

The maintenance of the flag is now entirely within rvt_post_one_wr()
where a reserved operation will set the flag and a non-reserved operation
will insure the operation that is about to be posted has the flag reset.

Fixes: Commit 856cc4c237 ("IB/hfi1: Add the capability for reserved operations")
Reviewed-by: Don Hiatt <don.hiatt@intel.com>
Signed-off-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
This commit is contained in:
Mike Marciniszyn 2017-03-20 17:26:01 -07:00 committed by Doug Ledford
parent 5f14e4e667
commit 44dcfa4b18
2 changed files with 6 additions and 5 deletions

View File

@ -1772,10 +1772,13 @@ static int rvt_post_one_wr(struct rvt_qp *qp,
0);
qp->s_next_psn = wqe->lpsn + 1;
}
if (unlikely(reserved_op))
if (unlikely(reserved_op)) {
wqe->wr.send_flags |= RVT_SEND_RESERVE_USED;
rvt_qp_wqe_reserve(qp, wqe);
else
} else {
wqe->wr.send_flags &= ~RVT_SEND_RESERVE_USED;
qp->s_avail--;
}
trace_rvt_post_one_wr(qp, wqe);
smp_wmb(); /* see request builders */
qp->s_head = next;

View File

@ -2,7 +2,7 @@
#define DEF_RDMAVT_INCQP_H
/*
* Copyright(c) 2016 Intel Corporation.
* Copyright(c) 2016, 2017 Intel Corporation.
*
* This file is provided under a dual BSD/GPLv2 license. When using or
* redistributing this file, you may do so under either license.
@ -526,7 +526,6 @@ static inline void rvt_qp_wqe_reserve(
struct rvt_qp *qp,
struct rvt_swqe *wqe)
{
wqe->wr.send_flags |= RVT_SEND_RESERVE_USED;
atomic_inc(&qp->s_reserved_used);
}
@ -550,7 +549,6 @@ static inline void rvt_qp_wqe_unreserve(
struct rvt_swqe *wqe)
{
if (unlikely(wqe->wr.send_flags & RVT_SEND_RESERVE_USED)) {
wqe->wr.send_flags &= ~RVT_SEND_RESERVE_USED;
atomic_dec(&qp->s_reserved_used);
/* insure no compiler re-order up to s_last change */
smp_mb__after_atomic();