nvme-pci: move nvme_kill_queues to nvme_remove_dead_ctrl

There is race between nvme_remove and nvme_reset_work that can
lead to io hang.

nvme_remove                    nvme_reset_work
                               -> nvme_remove_dead_ctrl
                                 -> nvme_dev_disable
                                   -> quiesce request_queue
                                 -> queue remove_work
-> cancel_work_sync reset_work
-> nvme_remove_namespaces
  -> splice ctrl->namespaces
                               nvme_remove_dead_ctrl_work
                               -> nvme_kill_queues
  -> nvme_ns_remove               do nothing
    -> blk_cleanup_queue
      -> blk_freeze_queue

Finally, the request_queue is quiesced state when wait freeze,
we will get io hang here. To fix it, move the nvme_kill_queues
from nvme_remove_dead_ctrl_work to nvme_remove_dead_ctrl.

Suggested-by: Keith Busch <keith.busch@linux.intel.com>
Signed-off-by: Jianchao Wang <jianchao.w.wang@oracle.com>
Reviewed-by: Keith Busch <keith.busch@intel.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>

diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c
index fc33804662e7bd35cfbacd93a26101bf23b3f43d..73a97fcea364af5782db501416ff114e8e8fde1b 100644 (file)
--- a/drivers/nvme/host/pci.c
+++ b/drivers/nvme/host/pci.c
@@ -2289,6 +2289,7 @@ static void nvme_remove_dead_ctrl(struct nvme_dev *dev, int status)
 
        nvme_get_ctrl(&dev->ctrl);
        nvme_dev_disable(dev, false);
+       nvme_kill_queues(&dev->ctrl);
        if (!queue_work(nvme_wq, &dev->remove_work))
                nvme_put_ctrl(&dev->ctrl);
 }
@@ -2405,7 +2406,6 @@ static void nvme_remove_dead_ctrl_work(struct work_struct *work)
        struct nvme_dev *dev = container_of(work, struct nvme_dev, remove_work);
        struct pci_dev *pdev = to_pci_dev(dev->dev);
 
-       nvme_kill_queues(&dev->ctrl);
        if (pci_get_drvdata(pdev))
                device_release_driver(&pdev->dev);
        nvme_put_ctrl(&dev->ctrl);