Commit b4d30668 authored Jul 10, 2019 by lorneli Committed by Ben Walker Jul 23, 2019

nvmf/rdma: defer qp destruction until nvmf layer closes qp



Currently rqpair will be destroyed directly in ibv_poll_cq path
if it has been drained, regardless of whether there are outstanding
I/Os issued to bdev layer. So after outstanding I/Os completing,
spdk_nvmf_rdma_close_qpair will be called from nvmf layer, accessing
a destroyed qp.

This path defers qp destruction in nvmf_rdma_destroy_drained_qpair
func until nvmf layer closes qp.

Fixes 851

Change-Id: I8bcce66f8053ddb105702ac603d5d73af54bdcfc
Signed-off-by: lorneli <lorneli@163.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/461237


Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Seth Howell <seth.howell@intel.com>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>

parent 7dbdf87f

lib/nvmf/rdma.c

+6 −0

Original line number	Diff line number	Diff line
		@@ -2672,6 +2672,12 @@ static void nvmf_rdma_destroy_drained_qpair(void *ctx)
		}

		spdk_nvmf_rdma_qpair_process_pending(rtransport, rqpair, true);

		/* Qpair will be destroyed after nvmf layer closes this qpair */
		if (rqpair->qpair.state != SPDK_NVMF_QPAIR_ERROR) {
		return;
		}

		spdk_nvmf_rdma_qpair_destroy(rqpair);
		}