diff options
| author | Israel Rukshin <israelr@mellanox.com> | 2018-06-19 15:34:11 +0300 |
|---|---|---|
| committer | Christoph Hellwig <hch@lst.de> | 2018-06-20 14:20:51 +0200 |
| commit | c947657b15379505a9bba36a02005882b66abe57 (patch) | |
| tree | 252bc9998774c6eb77d2e0dd31c850ec0d2db67d | |
| parent | 94e42213cc1ae41c57819539c0130f8dfc69d718 (diff) | |
nvme-rdma: Fix command completion race at error recovery
The race is between completing the request at error recovery work and
rdma completions. If we cancel the request before getting the good
rdma completion we get a NULL deref of the request MR at
nvme_rdma_process_nvme_rsp().
When Canceling the request we return its mr to the mr pool (set mr to
NULL) and also unmap its data. Canceling the requests while the rdma
queues are active is not safe. Because rdma queues are active and we
get good rdma completions that can use the mr pointer which may be NULL.
Completing the request too soon may lead also to performing DMA to/from
user buffers which might have been already unmapped.
The commit fixes the race by draining the QP before starting the abort
commands mechanism.
Signed-off-by: Israel Rukshin <israelr@mellanox.com>
Reviewed-by: Max Gurtovoy <maxg@mellanox.com>
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Christoph Hellwig <hch@lst.de>
| -rw-r--r-- | drivers/nvme/host/rdma.c | 8 |
1 files changed, 6 insertions, 2 deletions
diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c index f9affb71ac85..2815749f4dfb 100644 --- a/drivers/nvme/host/rdma.c +++ b/drivers/nvme/host/rdma.c @@ -728,7 +728,6 @@ out: 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 @@ out_free_queue: 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); @@ -1734,6 +1735,7 @@ 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); @@ -1745,6 +1747,7 @@ 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); @@ -2011,6 +2014,7 @@ 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_uninit_ctrl: nvme_uninit_ctrl(&ctrl->ctrl); |