Commit e59ac513 authored Mar 12, 2019 by Seth Howell Committed by Jim Harris Mar 19, 2019

rdma: remove reqs from read/write queues in error



Not doing so can cause us to hit asserts during the shutdown path. This
should fix an intermittent failure we are seeing on the test pool where
we hit the assert rdma_req->state != RDMA_REQUEST_STATE_FREE in
spdk_nvmf_rdma_request_process.

Note that this problem doesn't cause any data corruption when debug is
not enabled, it just causes us to probcess a subset of commands through
the state machine one extra time suring qpair shutdown.

Change-Id: Ibc36bfea87ec4089b8e2c7a915f48714fddb0b09
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/447843


Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>

parent 9e5075a7

lib/nvmf/rdma.c

+4 −0

Original line number	Diff line number	Diff line
		@@ -1550,6 +1550,10 @@ spdk_nvmf_rdma_request_process(struct spdk_nvmf_rdma_transport *rtransport,
		if (rqpair->ibv_attr.qp_state == IBV_QPS_ERR \|\| rqpair->qpair.state != SPDK_NVMF_QPAIR_ACTIVE) {
		if (rdma_req->state == RDMA_REQUEST_STATE_NEED_BUFFER) {
		TAILQ_REMOVE(&rgroup->pending_data_buf_queue, rdma_req, link);
		} else if (rdma_req->state == RDMA_REQUEST_STATE_DATA_TRANSFER_TO_CONTROLLER_PENDING) {
		STAILQ_REMOVE(&rqpair->pending_rdma_read_queue, rdma_req, spdk_nvmf_rdma_request, state_link);
		} else if (rdma_req->state == RDMA_REQUEST_STATE_DATA_TRANSFER_TO_HOST_PENDING) {
		STAILQ_REMOVE(&rqpair->pending_rdma_write_queue, rdma_req, spdk_nvmf_rdma_request, state_link);
		}
		rdma_req->state = RDMA_REQUEST_STATE_COMPLETED;
		}