[PATCH] Revert "nvme-multipath: fix hang when disk goes live over reconnect"

Chaitanya Kulkarni chaitanyak at nvidia.com
Wed Mar 23 17:07:32 PDT 2022


On 3/23/22 17:06, Chaitanya Kulkarni wrote:
> This reverts commit d50c992edf10b95d2034097405c94fecfbe1ef7f which is
> causing following OOPs:
> <1>[    1.943642] BUG: kernel NULL pointer dereference, address: 0000000000000008
> <1>[    1.943645] #PF: supervisor read access in kernel mode
> <1>[    1.943646] #PF: error_code(0x0000) - not-present page
> <6>[    1.943648] PGD 0 P4D 0
> <4>[    1.943649] Oops: 0000 [#1] PREEMPT SMP NOPTI
> <4>[    1.943651] CPU: 0 PID: 7 Comm: kworker/u96:0 Not tainted 5.17.0-rc2nvme+ #58
> <4>[    1.943653] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.14.0-0-g155821a1990b-prebuilt.qemu.org 04/01/2014
> <4>[    1.943654] Workqueue: nvme-reset-wq nvme_reset_work [nvme]
> <4>[    1.943662] RIP: 0010:nvme_parse_ana_log+0x1e/0x160 [nvme_core]
> 
> <4>[    1.943670] RSP: 0018:ffffc90000043e00 EFLAGS: 00010286
> <4>[    1.943672] RAX: 0000000000000000 RBX: ffff888103fd4210 RCX: 0000000000000000
> <4>[    1.943673] RDX: ffffffffc00b62c0 RSI: ffffc90000043e44 RDI: ffff888103fd4210
> <4>[    1.943673] RBP: ffff888103fd4210 R08: 0000000000000001 R09: ffff888100051828
> <4>[    1.943674] R10: 0000000000000000 R11: fffffffffff4a904 R12: 0000000000000000
> <4>[    1.943675] R13: ffff88817daa6500 R14: 0000000000000000 R15: ffff88817daa6505
> <4>[    1.943677] FS:  0000000000000000(0000) GS:ffff888fff200000(0000) knlGS:0000000000000000
> <4>[    1.943678] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> <4>[    1.943678] CR2: 0000000000000008 CR3: 00000001788a4000 CR4: 0000000000350ef0
> <4>[    1.943680] Call Trace:
> <4>[    1.943683]  <TASK>
> <4>[    1.943684]  ? nvme_update_ns_ana_state+0x40/0x40 [nvme_core]
> <4>[    1.943690]  nvme_mpath_update+0x4a/0x70 [nvme_core]
> <4>[    1.943695]  nvme_start_ctrl+0x110/0x140 [nvme_core]
> <4>[    1.943700]  process_one_work+0x1af/0x380
> <4>[    1.943709]  worker_thread+0x50/0x3a0
> <4>[    1.943711]  ? rescuer_thread+0x370/0x370
> <4>[    1.943712]  kthread+0xe7/0x110
> <4>[    1.943714]  ? kthread_complete_and_exit+0x20/0x20
> <4>[    1.943716]  ret_from_fork+0x22/0x30
> <4>[    1.943719]  </TASK>
> 
> [0]kdb>
> 
> With this revert not testing can proceed forward.
> 

s/not/now/, can be do at the time of applying patch.

-ck




More information about the Linux-nvme mailing list