[PATCH 2/2] nvme-rdma: Fix race at error recovery

[Israel Rukshin]

We need to stop rdma queues before canceling the requests.
With this approach we avoid the race between error recovery work
and rdma completions.
The commit fix a NULL deref of a request MR at nvme_rdma_process_nvme_rsp().

Signed-off-by: Israel Rukshin <israelr at mellanox.com>
Reviewed-by: Max Gurtovoy <maxg at mellanox.com>
---
 drivers/nvme/host/rdma.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c
index c1abfc8..56fccac 100644
--- a/drivers/nvme/host/rdma.c
+++ b/drivers/nvme/host/rdma.c
@@ -734,7 +734,6 @@ static struct blk_mq_tag_set *nvme_rdma_alloc_tagset(struct nvme_ctrl *nctrl,
 static void nvme_rdma_destroy_admin_queue(struct nvme_rdma_ctrl *ctrl,
 		bool remove)
 {
-	nvme_rdma_stop_queue(&ctrl->queues[0]);
 	if (remove) {
 		blk_cleanup_queue(ctrl->ctrl.admin_q);
 		nvme_rdma_free_tagset(&ctrl->ctrl, ctrl->ctrl.admin_tagset);
@@ -817,7 +816,6 @@ static int nvme_rdma_configure_admin_queue(struct nvme_rdma_ctrl *ctrl,
 static void nvme_rdma_destroy_io_queues(struct nvme_rdma_ctrl *ctrl,
 		bool remove)
 {
-	nvme_rdma_stop_io_queues(ctrl);
 	if (remove) {
 		blk_cleanup_queue(ctrl->ctrl.connect_q);
 		nvme_rdma_free_tagset(&ctrl->ctrl, ctrl->ctrl.tagset);
@@ -947,6 +945,7 @@ static void nvme_rdma_reconnect_ctrl_work(struct work_struct *work)
 	return;
 
 destroy_admin:
+	nvme_rdma_stop_queue(&ctrl->queues[0]);
 	nvme_rdma_destroy_admin_queue(ctrl, false);
 requeue:
 	dev_info(ctrl->ctrl.device, "Failed reconnect attempt %d\n",
@@ -963,12 +962,14 @@ static void nvme_rdma_error_recovery_work(struct work_struct *work)
 
 	if (ctrl->ctrl.queue_count > 1) {
 		nvme_stop_queues(&ctrl->ctrl);
+		nvme_rdma_stop_io_queues(ctrl);
 		blk_mq_tagset_busy_iter(&ctrl->tag_set,
 					nvme_cancel_request, &ctrl->ctrl);
 		nvme_rdma_destroy_io_queues(ctrl, false);
 	}
 
 	blk_mq_quiesce_queue(ctrl->ctrl.admin_q);
+	nvme_rdma_stop_queue(&ctrl->queues[0]);
 	blk_mq_tagset_busy_iter(&ctrl->admin_tag_set,
 				nvme_cancel_request, &ctrl->ctrl);
 	nvme_rdma_destroy_admin_queue(ctrl, false);
@@ -1725,6 +1726,7 @@ static void nvme_rdma_shutdown_ctrl(struct nvme_rdma_ctrl *ctrl, bool shutdown)
 {
 	if (ctrl->ctrl.queue_count > 1) {
 		nvme_stop_queues(&ctrl->ctrl);
+		nvme_rdma_stop_io_queues(ctrl);
 		blk_mq_tagset_busy_iter(&ctrl->tag_set,
 					nvme_cancel_request, &ctrl->ctrl);
 		nvme_rdma_destroy_io_queues(ctrl, shutdown);
@@ -1736,6 +1738,7 @@ static void nvme_rdma_shutdown_ctrl(struct nvme_rdma_ctrl *ctrl, bool shutdown)
 		nvme_disable_ctrl(&ctrl->ctrl, ctrl->ctrl.cap);
 
 	blk_mq_quiesce_queue(ctrl->ctrl.admin_q);
+	nvme_rdma_stop_queue(&ctrl->queues[0]);
 	blk_mq_tagset_busy_iter(&ctrl->admin_tag_set,
 				nvme_cancel_request, &ctrl->ctrl);
 	blk_mq_unquiesce_queue(ctrl->ctrl.admin_q);
@@ -2001,6 +2004,7 @@ static struct nvme_ctrl *nvme_rdma_create_ctrl(struct device *dev,
 	return &ctrl->ctrl;
 
 out_remove_admin_queue:
+	nvme_rdma_stop_queue(&ctrl->queues[0]);
 	nvme_rdma_destroy_admin_queue(ctrl, true);
 out_kfree_queues:
 	kfree(ctrl->queues);
-- 
1.8.3.1