public inbox for [email protected]
 help / color / mirror / Atom feed
From: Pavel Begunkov <[email protected]>
To: Palash Oswal <[email protected]>
Cc: Jens Axboe <[email protected]>,
	[email protected], LKML <[email protected]>,
	[email protected]
Subject: Re: INFO: task hung in io_uring_cancel_sqpoll
Date: Mon, 3 May 2021 11:45:26 +0100	[thread overview]
Message-ID: <[email protected]> (raw)
In-Reply-To: <CAGyP=7e-3QtS-Z3KoAyFAbvm4y+9=725WR_+PyADYDi8HYxMXA@mail.gmail.com>

On 5/3/21 5:41 AM, Palash Oswal wrote:
> On Mon, May 3, 2021 at 3:42 AM Pavel Begunkov <[email protected]> wrote:
>> The test might be very useful. Would you send a patch to
>> liburing? Or mind the repro being taken as a test?
> 
> Pavel,
> 
> I'm working on a PR for liburing with this test. Do you know how I can

Perfect, thanks

> address this behavior?

As the hang is sqpoll cancellations, it's most probably
fixed in 5.13 + again awaits to be taken for stable.

Don't know about segfaults, but it was so for long, and
syz reproducers are ofthen faults for me, and exit with 0
in the end. So, I wouldn't worry about it.

> root@syzkaller:~/liburing/test# ./runtests.sh sqpoll-cancel-hang
> Running test sqp[   15.310997] Running test sqpoll-cancel-hang:
> oll-cancel-hang:
> [   15.333348] sqpoll-cancel-h[305]: segfault at 0 ip 000055ad00e265e3 sp]
> [   15.334940] Code: 89 d8 8d 34 90 8b 45 04 ba 03 00 00 00 c1 e0 04 03 46
> All tests passed
> 
> root@syzkaller:~/liburing/test# ./sqpoll-cancel-hang
> [   13.572639] sqpoll-cancel-h[298]: segfault at 0 ip 00005634c4a455e3 sp]
> [   13.576506] Code: 89 d8 8d 34 90 8b 45 04 ba 03 00 00 00 c1 e0 04 03 46
> [   23.350459] random: crng init done
> [   23.352837] random: 7 urandom warning(s) missed due to ratelimiting
> [  243.090865] INFO: task iou-sqp-298:299 blocked for more than 120 secon.
> [  243.095187]       Not tainted 5.12.0 #142
> [  243.099800] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disable.
> [  243.105928] task:iou-sqp-298     state:D stack:    0 pid:  299 ppid:  4
> [  243.111044] Call Trace:
> [  243.112855]  __schedule+0xb1d/0x1130
> [  243.115549]  ? __sched_text_start+0x8/0x8
> [  243.118328]  ? io_wq_worker_sleeping+0x145/0x500
> [  243.121790]  schedule+0x131/0x1c0
> [  243.123698]  io_uring_cancel_sqpoll+0x288/0x350
> [  243.125977]  ? io_sq_thread_unpark+0xd0/0xd0
> [  243.128966]  ? mutex_lock+0xbb/0x130
> [  243.132572]  ? init_wait_entry+0xe0/0xe0
> [  243.135429]  ? wait_for_completion_killable_timeout+0x20/0x20
> [  243.138303]  io_sq_thread+0x174c/0x18c0
> [  243.140162]  ? io_rsrc_put_work+0x380/0x380
> [  243.141613]  ? init_wait_entry+0xe0/0xe0
> [  243.143686]  ? _raw_spin_lock_irq+0xa5/0x180
> [  243.145619]  ? _raw_spin_lock_irqsave+0x190/0x190
> [  243.147671]  ? calculate_sigpending+0x6b/0xa0
> [  243.149036]  ? io_rsrc_put_work+0x380/0x380
> [  243.150694]  ret_from_fork+0x22/0x30
> 
> Palash
> 

-- 
Pavel Begunkov

  reply	other threads:[~2021-05-03 10:45 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <[email protected]>
     [not found] ` <[email protected]>
     [not found]   ` <[email protected]>
2021-04-30 14:21     ` INFO: task hung in io_uring_cancel_sqpoll Palash Oswal
2021-04-30 14:33       ` Pavel Begunkov
2021-04-30 15:02         ` Palash Oswal
2021-04-30 15:07           ` Pavel Begunkov
2021-04-30 15:11             ` Palash Oswal
2021-04-30 18:34           ` Pavel Begunkov
2021-04-30 18:34             ` syzbot
2021-04-30 21:04             ` Pavel Begunkov
2021-05-01  4:51               ` Palash Oswal
2021-05-01  7:39                 ` Pavel Begunkov
2021-05-02 10:33           ` Pavel Begunkov
2021-05-02 15:01             ` Palash Oswal
2021-05-02 15:56               ` Pavel Begunkov
2021-05-02 22:12       ` Pavel Begunkov
2021-05-03  4:41         ` Palash Oswal
2021-05-03 10:45           ` Pavel Begunkov [this message]
     [not found]             ` <CAGyP=7eoSfh7z638PnP5UF4xVKcrG1jB_qmFo6uPZ7iWfu_2sQ@mail.gmail.com>
2021-05-13 21:28               ` Pavel Begunkov
2021-05-13 21:31                 ` Pavel Begunkov
2021-05-15  7:44                   ` Palash Oswal
2021-06-12 15:48 ` syzbot

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 \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    /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