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.bu...@linux.intel.com> Signed-off-by: Jianchao Wang <jianchao.w.w...@oracle.com> --- V2: - Just not invoke nvme_remove_dead_ctrl cannot fix the hole completely. Move the nvme_kill_queues to nvme_remove_dead_ctrl based on Keith's suggestion - Patch comment changes drivers/nvme/host/pci.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c index fc33804..73a97fc 100644 --- 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); -- 2.7.4