nvme: fix hang in remove path
authorMing Lei <ming.lei@redhat.com>
Fri, 2 Jun 2017 08:32:08 +0000 (16:32 +0800)
committerChristoph Hellwig <hch@lst.de>
Wed, 7 Jun 2017 09:08:49 +0000 (11:08 +0200)
We need to start admin queues too in nvme_kill_queues()
for avoiding hang in remove path[1].

This patch is very similar with 806f026f9b901eaf(nvme: use
blk_mq_start_hw_queues() in nvme_kill_queues()).

[1] hang stack trace
[<ffffffff813c9716>] blk_execute_rq+0x56/0x80
[<ffffffff815cb6e9>] __nvme_submit_sync_cmd+0x89/0xf0
[<ffffffff815ce7be>] nvme_set_features+0x5e/0x90
[<ffffffff815ce9f6>] nvme_configure_apst+0x166/0x200
[<ffffffff815cef45>] nvme_set_latency_tolerance+0x35/0x50
[<ffffffff8157bd11>] apply_constraint+0xb1/0xc0
[<ffffffff8157cbb4>] dev_pm_qos_constraints_destroy+0xf4/0x1f0
[<ffffffff8157b44a>] dpm_sysfs_remove+0x2a/0x60
[<ffffffff8156d951>] device_del+0x101/0x320
[<ffffffff8156db8a>] device_unregister+0x1a/0x60
[<ffffffff8156dc4c>] device_destroy+0x3c/0x50
[<ffffffff815cd295>] nvme_uninit_ctrl+0x45/0xa0
[<ffffffff815d4858>] nvme_remove+0x78/0x110
[<ffffffff81452b69>] pci_device_remove+0x39/0xb0
[<ffffffff81572935>] device_release_driver_internal+0x155/0x210
[<ffffffff81572a02>] device_release_driver+0x12/0x20
[<ffffffff815d36fb>] nvme_remove_dead_ctrl_work+0x6b/0x70
[<ffffffff810bf3bc>] process_one_work+0x18c/0x3a0
[<ffffffff810bf61e>] worker_thread+0x4e/0x3b0
[<ffffffff810c5ac9>] kthread+0x109/0x140
[<ffffffff8185800c>] ret_from_fork+0x2c/0x40
[<ffffffffffffffff>] 0xffffffffffffffff

Fixes: c5552fde102fc("nvme: Enable autonomous power state transitions")
Reported-by: Rakesh Pandit <rakesh@tuxera.com>
Tested-by: Rakesh Pandit <rakesh@tuxera.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
drivers/nvme/host/core.c

index a60926410438b98c2e414de081f7c8093bac5862..0f9cc0c55e153758b0f16c583870f466f40be808 100644 (file)
@@ -2438,6 +2438,10 @@ void nvme_kill_queues(struct nvme_ctrl *ctrl)
        struct nvme_ns *ns;
 
        mutex_lock(&ctrl->namespaces_mutex);
+
+       /* Forcibly start all queues to avoid having stuck requests */
+       blk_mq_start_hw_queues(ctrl->admin_q);
+
        list_for_each_entry(ns, &ctrl->namespaces, list) {
                /*
                 * Revalidating a dead namespace sets capacity to 0. This will