From: Oleksandr Natalenko <oleksandr@natalenko.name>
To: linux-kernel@vger.kernel.org, Jens Axboe <axboe@fb.com>,
Christoph Hellwig <hch@lst.de>, Sagi Grimberg <sagi@grimberg.me>,
linux-nvme@lists.infradead.org,
David Jeffery <djeffery@redhat.com>,
Ming Lei <ming.lei@redhat.com>,
Laurence Oberman <loberman@redhat.com>,
Paolo Valente <paolo.valente@linaro.org>, Jan Kara <jack@suse.cz>,
Sasha Levin <sashal@kernel.org>,
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Keith Busch <kbusch@kernel.org>
Subject: New warning in nvme_setup_discard
Date: Thu, 15 Jul 2021 15:56:38 +0200 [thread overview]
Message-ID: <4729812.CpyZKHjjVO@natalenko.name> (raw)
Hello.
After a v5.13.2 massive update I encountered this:
```
[19231.556665] ------------[ cut here ]------------
[19231.556674] WARNING: CPU: 20 PID: 502 at drivers/nvme/host/core.c:850
nvme_setup_discard+0x188/0x1f0
...
[19231.556776] CPU: 20 PID: 502 Comm: kworker/20:1H Not tainted 5.13.2 #1
[19231.556780] Hardware name: ASUS System Product Name/Pro WS X570-ACE, BIOS
3601 05/26/2021
[19231.556784] Workqueue: kblockd blk_mq_run_work_fn
[19231.556789] RIP: 0010:nvme_setup_discard+0x188/0x1f0
[19231.556794] Code: 49 8b 44 24 10 4c 8b 90 40 0b 00 00 4c 2b 15 8f 09 d8 00
49 c1 fa 06 49 c1 e2 0c 4c 03 15 90 09 d8 00 4d 89 d0 e9 b9 fe ff ff <0f> 0b b8
00 00 00 80 49 01 c2 72 52 48 c7 c0 00 00 00 80 48 2b 05
[19231.556798] RSP: 0018:ffffaed2416efc00 EFLAGS: 00010287
[19231.556802] RAX: ffff8e67fb580000 RBX: ffff8e640bbe5240 RCX: 0000000000000020
[19231.556805] RDX: ffff8e67fb580000 RSI: 000000000000001f RDI: 0000000000000000
[19231.556808] RBP: ffff8e640bbe5388 R08: ffff8e677b580000 R09: 0000000008abb800
[19231.556811] R10: ffff8e677b580000 R11: 0000000000000400 R12: ffff8e6405999c00
[19231.556814] R13: 000000000000001f R14: ffff8e6405693000 R15: ffff8e640bbaf800
[19231.556816] FS: 0000000000000000(0000) GS:ffff8e6b0ef00000(0000) knlGS:
0000000000000000
[19231.556819] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[19231.556822] CR2: ffff888c76ece000 CR3: 000000047a184000 CR4: 0000000000350ee0
[19231.556825] Call Trace:
[19231.556830] nvme_setup_cmd+0x2d0/0x670
[19231.556834] nvme_queue_rq+0x79/0xc90
[19231.556837] ? __sbitmap_get_word+0x30/0x80
[19231.556842] ? sbitmap_get+0x85/0x180
[19231.556846] blk_mq_dispatch_rq_list+0x15c/0x810
[19231.556851] ? list_sort+0x21d/0x2f0
[19231.556856] __blk_mq_do_dispatch_sched+0x196/0x320
[19231.556860] __blk_mq_sched_dispatch_requests+0x14d/0x190
[19231.556864] blk_mq_sched_dispatch_requests+0x2f/0x60
[19231.556867] blk_mq_run_work_fn+0x43/0xc0
[19231.556871] process_one_work+0x24e/0x430
[19231.556876] worker_thread+0x54/0x4d0
[19231.556880] ? process_one_work+0x430/0x430
[19231.556883] kthread+0x1b3/0x1e0
[19231.556886] ? __kthread_init_worker+0x50/0x50
[19231.556889] ret_from_fork+0x22/0x30
[19231.556895] ---[ end trace d9abdf019a56b4c7 ]---
[19231.556906] blk_update_request: I/O error, dev nvme1n1, sector 632935424 op
0x3:(DISCARD) flags 0x0 phys_seg 31 prio class 0
```
or, in code:
```
850 if (WARN_ON_ONCE(n != segments)) {
851 if (virt_to_page(range) == ns->ctrl->discard_page)
852 clear_bit_unlock(0, &ns->ctrl->discard_page_busy);
853 else
854 kfree(range);
855 return BLK_STS_IOERR;
856 }
```
BFQ scheduler is in use.
Something similar was already fixed by a958937ff166fc60d1c3a721036f6ff41bfa2821,
but I do not have a multipath device here, it's just 2 NVMe SSDs in a soft
RAID10 with LUKS and LVM.
Any idea what this might mean? v5.13.2 brought some commit into a stable tree
that are, as I still suspect, causing unreproducible panics [1] [2].
Previously, I dropped that extra stuff from my kernel build and had no issues.
This time I also do not have any extra commits in the block layer, only those
that are in v5.13.2.
Thanks.
[1] https://lore.kernel.org/linux-block/3533087.dJKXTdksHR@spock/
[2] https://lore.kernel.org/linux-block/2957867.CS06ZTPI5V@spock/
--
Oleksandr Natalenko (post-factum)
next reply other threads:[~2021-07-15 13:56 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-07-15 13:56 Oleksandr Natalenko [this message]
2021-07-15 14:19 ` New warning in nvme_setup_discard Greg Kroah-Hartman
2021-07-15 14:21 ` Oleksandr Natalenko
2021-07-15 21:37 ` Laurence Oberman
2021-07-16 5:50 ` Oleksandr Natalenko
2021-07-16 2:16 ` Ming Lei
2021-07-16 5:53 ` Oleksandr Natalenko
2021-07-16 9:33 ` Ming Lei
2021-07-16 10:03 ` Oleksandr Natalenko
2021-07-16 10:41 ` Ming Lei
2021-07-16 12:56 ` Oleksandr Natalenko
2021-07-17 9:35 ` Ming Lei
2021-07-17 12:11 ` Oleksandr Natalenko
2021-07-17 12:19 ` Oleksandr Natalenko
2021-07-17 12:35 ` Oleksandr Natalenko
2021-07-19 1:40 ` Ming Lei
2021-07-19 6:27 ` Oleksandr Natalenko
2021-07-20 9:05 ` Oleksandr Natalenko
2021-07-21 8:00 ` Ming Lei
2021-07-27 15:12 ` Oleksandr Natalenko
2021-07-27 15:58 ` Ming Lei
2021-07-28 13:44 ` Oleksandr Natalenko
2021-07-28 15:53 ` Ming Lei
2021-07-28 16:38 ` Oleksandr Natalenko
2021-07-29 3:33 ` Ming Lei
2021-07-29 9:29 ` Ming Lei
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=4729812.CpyZKHjjVO@natalenko.name \
--to=oleksandr@natalenko.name \
--cc=axboe@fb.com \
--cc=djeffery@redhat.com \
--cc=gregkh@linuxfoundation.org \
--cc=hch@lst.de \
--cc=jack@suse.cz \
--cc=kbusch@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-nvme@lists.infradead.org \
--cc=loberman@redhat.com \
--cc=ming.lei@redhat.com \
--cc=paolo.valente@linaro.org \
--cc=sagi@grimberg.me \
--cc=sashal@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).