Commit 57dc541c authored Mar 15, 2022 by Chunsong Feng Committed by Tomasz Zawadzki Mar 16, 2022

rdma: set rnr_retry_count when rq is used



When rq is used, rnr_retry_count is not set. As a result, the initiator
cannot connect to the target after some error. Set the rnr_retry_count
to avoid infinite retries.

Fix issue #2429

Signed-off-by: Chunsong Feng <fengchunsong@huawei.com>
Change-Id: Iab03ea5ff10f5a9759c41f26589731aa92ef2f56
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/11946


Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Community-CI: Mellanox Build Bot

parent ca8e6d23

lib/nvmf/rdma.c

+2 −5

Original line number	Diff line number	Diff line
		@@ -1205,13 +1205,10 @@ nvmf_rdma_event_accept(struct rdma_cm_id id, struct spdk_nvmf_rdma_qpair rqpai
		}

		/* Configure infinite retries for the initiator side qpair.
		* When using a shared receive queue on the target side,
		* we need to pass this value to the initiator to prevent the
		* We need to pass this value to the initiator to prevent the
		* initiator side NIC from completing SEND requests back to the
		* initiator with status rnr_retry_count_exceeded. */
		if (rqpair->srq != NULL) {
		ctrlr_event_data.rnr_retry_count = 0x7;
		}

		/* When qpair is created without use of rdma cm API, an additional
		* information must be provided to initiator in the connection response:

Admin message