[regression] nvme pci sysfs remove operation hang during fio test observed on 5.16.0-rc1

Yi Zhang yi.zhang at redhat.com
Wed Nov 17 07:57:41 PST 2021


Hello

I found this regression issue on 5.16.0-rc1, pls check it and let me
know if you need any test/debug info for it, thanks.

[  369.657564] INFO: task main.sh:1695 blocked for more than 122 seconds.
[  369.664871]       Tainted: G S                5.16.0-rc1 #1
[  369.671106] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs"
disables this message.
[  369.679854] task:main.sh         state:D stack:    0 pid: 1695
ppid:  1692 flags:0x00004004
[  369.679863] Call Trace:
[  369.679866]  <TASK>
[  369.679872]  __schedule+0x3e3/0x9a0
[  369.679887]  ? _raw_spin_lock_irqsave+0x17/0x40
[  369.679893]  schedule+0x44/0xc0
[  369.679899]  blk_mq_freeze_queue_wait+0x62/0x90
[  369.679908]  ? finish_wait+0x80/0x80
[  369.679920]  del_gendisk+0x18d/0x210
[  369.679931]  nvme_ns_remove+0xc7/0x1d0 [nvme_core]
[  369.679954]  nvme_remove_namespaces+0xac/0xf0 [nvme_core]
[  369.679974]  nvme_remove+0x60/0x140 [nvme]
[  369.679984]  pci_device_remove+0x36/0xb0
[  369.679994]  device_release_driver_internal+0xf7/0x1e0
[  369.680005]  pci_stop_bus_device+0x69/0x90
[  369.680012]  pci_stop_and_remove_bus_device_locked+0x16/0x30
[  369.680018]  remove_store+0x75/0x90
[  369.680025]  kernfs_fop_write_iter+0x130/0x1c0
[  369.680035]  new_sync_write+0x122/0x1b0
[  369.680045]  vfs_write+0x1ba/0x2a0
[  369.680053]  ksys_write+0x59/0xd0
[  369.680058]  do_syscall_64+0x3a/0x80
[  369.680067]  entry_SYSCALL_64_after_hwframe+0x44/0xae
[  369.680073] RIP: 0033:0x7f99b1159648
[  369.680078] RSP: 002b:00007ffec2a09748 EFLAGS: 00000246 ORIG_RAX:
0000000000000001
[  369.680083] RAX: ffffffffffffffda RBX: 0000000000000002 RCX: 00007f99b1159648
[  369.680086] RDX: 0000000000000002 RSI: 0000564e26ba1b20 RDI: 0000000000000001
[  369.680089] RBP: 0000564e26ba1b20 R08: 000000000000000a R09: 00007f99b11ec620
[  369.680091] R10: 000000000000000a R11: 0000000000000246 R12: 00007f99b142c6e0
[  369.680094] R13: 0000000000000002 R14: 00007f99b1427880 R15: 0000000000000002
[  369.680099]  </TASK>
[  492.536344] INFO: task main.sh:1695 blocked for more than 245 seconds.
[  492.543638]       Tainted: G S                5.16.0-rc1 #1
[  492.549870] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs"
disables this message.
[  492.558617] task:main.sh         state:D stack:    0 pid: 1695
ppid:  1692 flags:0x00004004
[  492.558625] Call Trace:
[  492.558627]  <TASK>
[  492.558631]  __schedule+0x3e3/0x9a0
[  492.558639]  ? _raw_spin_lock_irqsave+0x17/0x40
[  492.558644]  schedule+0x44/0xc0
[  492.558650]  blk_mq_freeze_queue_wait+0x62/0x90
[  492.558655]  ? finish_wait+0x80/0x80
[  492.558663]  del_gendisk+0x18d/0x210
[  492.558671]  nvme_ns_remove+0xc7/0x1d0 [nvme_core]
[  492.558692]  nvme_remove_namespaces+0xac/0xf0 [nvme_core]
[  492.558710]  nvme_remove+0x60/0x140 [nvme]
[  492.558718]  pci_device_remove+0x36/0xb0
[  492.558724]  device_release_driver_internal+0xf7/0x1e0
[  492.558732]  pci_stop_bus_device+0x69/0x90
[  492.558737]  pci_stop_and_remove_bus_device_locked+0x16/0x30
[  492.558742]  remove_store+0x75/0x90
[  492.558748]  kernfs_fop_write_iter+0x130/0x1c0
[  492.558756]  new_sync_write+0x122/0x1b0
[  492.558763]  vfs_write+0x1ba/0x2a0
[  492.558769]  ksys_write+0x59/0xd0
[  492.558775]  do_syscall_64+0x3a/0x80
[  492.558780]  entry_SYSCALL_64_after_hwframe+0x44/0xae
[  492.558785] RIP: 0033:0x7f99b1159648
[  492.558789] RSP: 002b:00007ffec2a09748 EFLAGS: 00000246 ORIG_RAX:
0000000000000001
[  492.558794] RAX: ffffffffffffffda RBX: 0000000000000002 RCX: 00007f99b1159648
[  492.558797] RDX: 0000000000000002 RSI: 0000564e26ba1b20 RDI: 0000000000000001
[  492.558800] RBP: 0000564e26ba1b20 R08: 000000000000000a R09: 00007f99b11ec620
[  492.558803] R10: 000000000000000a R11: 0000000000000246 R12: 00007f99b142c6e0
[  492.558805] R13: 0000000000000002 R14: 00007f99b1427880 R15: 0000000000000002
[  492.558811]  </TASK>

-- 
Best Regards,
  Yi Zhang




More information about the Linux-nvme mailing list