| Message ID | 20180227205633.8292-3-tatyana.e.nikolova@intel.com (mailing list archive) |
|---|---|
| State | Accepted |
| Delegated to: | Leon Romanovsky |
| Headers | show |
diff --git a/providers/i40iw/i40iw_uverbs.c b/providers/i40iw/i40iw_uverbs.c index bb3cfd43..d77a4691 100644 --- a/providers/i40iw/i40iw_uverbs.c +++ b/providers/i40iw/i40iw_uverbs.c @@ -342,6 +342,8 @@ int i40iw_upoll_cq(struct ibv_cq *cq, int num_entries, struct ibv_wc *entry) ret = iwucq->cq.ops.iw_cq_poll_completion(&iwucq->cq, &cq_poll_info); if (ret == I40IW_ERR_QUEUE_EMPTY) { break; + } else if (ret == I40IW_ERR_QUEUE_DESTROYED) { + continue; } else if (ret) { fprintf(stderr, PFX "%s: Error polling CQ, status %d\n", __func__, ret); if (!cqe_count)
Currently CQ poll returns an error if the QP associated with a CQE is destroyed although there can be more CQEs for other QPs which are still active. Fix this by skipping the CQE for the destroyed QP and continue with CQE processing. The patch is similar to driver commit f8a4e76c75e5 ("RDMA/i40iw: Fix for checking if the QP is destroyed") Fixes: faf6782ac8bc ("libi40iw: Initial public release of libi40iw-0.5.223") Reported-by: Andrew Boyer <Andrew.Boyer@dell.com> Signed-off-by: Tatyana Nikolova <tatyana.e.nikolova@intel.com> --- providers/i40iw/i40iw_uverbs.c | 2 ++ 1 file changed, 2 insertions(+)