All of lore.kernel.org
 help / color / mirror / Atom feed
From: Dmitry Vyukov <dvyukov@google.com>
To: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp>
Cc: Michal Hocko <mhocko@kernel.org>,
	Sergey Senozhatsky <sergey.senozhatsky@gmail.com>,
	syzbot <syzbot+77e6b28a7a7106ad0def@syzkaller.appspotmail.com>,
	Johannes Weiner <hannes@cmpxchg.org>,
	Andrew Morton <akpm@linux-foundation.org>,
	guro@fb.com,
	"Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>,
	LKML <linux-kernel@vger.kernel.org>,
	Linux-MM <linux-mm@kvack.org>,
	David Rientjes <rientjes@google.com>,
	syzkaller-bugs <syzkaller-bugs@googlegroups.com>,
	Yang Shi <yang.s@alibaba-inc.com>,
	Sergey Senozhatsky <sergey.senozhatsky.work@gmail.com>,
	Petr Mladek <pmladek@suse.com>
Subject: Re: INFO: rcu detected stall in shmem_fault
Date: Wed, 10 Oct 2018 15:17:57 +0200	[thread overview]
Message-ID: <CACT4Y+bQOr=+e8y5so3V9K390uJnMy4+FR7_t9G7Fjn12S4hoA@mail.gmail.com> (raw)
In-Reply-To: <b7727ff0-b34f-25a2-b9e7-56e70d9349c4@i-love.sakura.ne.jp>

On Wed, Oct 10, 2018 at 3:10 PM, Tetsuo Handa
<penguin-kernel@i-love.sakura.ne.jp> wrote:
>>>>>>> Just flooding out of memory messages can trigger RCU stall problems.
>>>>>>> For example, a severe skbuff_head_cache or kmalloc-512 leak bug is causing
>>>>>>
>>>>>> [...]
>>>>>>
>>>>>> Quite some of them, indeed! I guess we want to rate limit the output.
>>>>>> What about the following?
>>>>>
>>>>> A bit unrelated, but while we are at it:
>>>>>
>>>>>   I like it when we rate-limit printk-s that lookup the system.
>>>>> But it seems that default rate-limit values are not always good enough,
>>>>> DEFAULT_RATELIMIT_INTERVAL / DEFAULT_RATELIMIT_BURST can still be too
>>>>> verbose. For instance, when we have a very slow IPMI emulated serial
>>>>> console -- e.g. baud rate at 57600. DEFAULT_RATELIMIT_INTERVAL and
>>>>> DEFAULT_RATELIMIT_BURST can add new OOM headers and backtraces faster
>>>>> than we evict them.
>>>>>
>>>>> Does it sound reasonable enough to use larger than default rate-limits
>>>>> for printk-s in OOM print-outs? OOM reports tend to be somewhat large
>>>>> and the reported numbers are not always *very* unique.
>>>>>
>>>>> What do you think?
>>>>
>>>> I do not really care about the current inerval/burst values. This change
>>>> should be done seprately and ideally with some numbers.
>>>
>>> I think Sergey meant that this place may need to use
>>> larger-than-default values because it prints lots of output per
>>> instance (whereas the default limit is more tuned for cases that print
>>> just 1 line).
>
> Yes. The OOM killer tends to print a lot of messages (and I estimate that
> mutex_trylock(&oom_lock) accelerates wasting more CPU consumption by
> preemption).
>
>>>
>>> I've found at least 1 place that uses DEFAULT_RATELIMIT_INTERVAL*10:
>>> https://elixir.bootlin.com/linux/latest/source/fs/btrfs/extent-tree.c#L8365
>>> Probably we need something similar here.
>
> Since printk() is a significantly CPU consuming operation, I think that what
> we need to guarantee is interval between the end of an OOM killer messages
> and the beginning of next OOM killer messages is large enough. For example,
> setup a timer with 5 seconds timeout upon the end of an OOM killer messages
> and check whether the timer already fired upon the beginning of next OOM killer
> messages.
>
>>
>>
>> In parallel with the kernel changes I've also made a change to
>> syzkaller that (1) makes it not use oom_score_adj=-1000, this hard
>> killing limit looks like quite risky thing, (2) increase memcg size
>> beyond expected KASAN quarantine size:
>> https://github.com/google/syzkaller/commit/adedaf77a18f3d03d695723c86fc083c3551ff5b
>> If this will stop the flow of hang/stall reports, then we can just
>> close all old reports as invalid.
>
> I don't think so. Only this report was different from others because printk()
> in this report was from memcg OOM events without eligible tasks whereas printk()
> in others are from global OOM events triggered by severe slab memory leak.

Ack.
I guess I just hoped deep down that we somehow magically get rid of
all these reports with some simple change like this :)

  reply	other threads:[~2018-10-10 13:18 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-10-10  0:08 syzbot
2018-10-10  0:12 ` Tetsuo Handa
2018-10-10  4:11   ` David Rientjes
2018-10-10  7:55     ` Dmitry Vyukov
2018-10-10  9:13       ` Michal Hocko
2018-10-10  9:33         ` Dmitry Vyukov
2018-10-10  9:02     ` Michal Hocko
2018-10-10  8:59   ` Michal Hocko
2018-10-10 10:43     ` Tetsuo Handa
2018-10-10 11:35       ` Michal Hocko
2018-10-10 11:48         ` Sergey Senozhatsky
2018-10-10 12:25           ` Michal Hocko
2018-10-10 12:29             ` Dmitry Vyukov
2018-10-10 12:36               ` Dmitry Vyukov
2018-10-10 13:10                 ` Tetsuo Handa
2018-10-10 13:17                   ` Dmitry Vyukov [this message]
2018-10-11  1:17                   ` Sergey Senozhatsky
2018-10-10 15:17               ` Sergey Senozhatsky
2018-10-10 14:19         ` Tetsuo Handa
2018-10-10 15:11 ` [RFC PATCH] memcg, oom: throttle dump_header for memcg ooms without eligible tasks Michal Hocko
2018-10-10 15:11   ` Michal Hocko
2018-10-11  6:37   ` Tetsuo Handa
2018-10-11  6:37     ` Tetsuo Handa
2018-10-12 10:47     ` Tetsuo Handa
2018-10-12 10:47       ` Tetsuo Handa
2018-10-12 11:20   ` Johannes Weiner
2018-10-12 12:08     ` Michal Hocko
2018-10-12 12:10       ` Tetsuo Handa
2018-10-12 12:41         ` Johannes Weiner
2018-10-12 12:58           ` Tetsuo Handa
2018-10-13 11:09             ` Tetsuo Handa
2018-10-13 11:22               ` Johannes Weiner
2018-10-13 11:28                 ` Tetsuo Handa
2018-10-15  8:19                   ` Michal Hocko
2018-10-15 10:57                     ` Tetsuo Handa
2018-10-15 11:24                       ` Michal Hocko
2018-10-15 12:47                         ` Tetsuo Handa
2018-10-15 13:35                           ` Michal Hocko
2018-10-16  0:55                             ` Tetsuo Handa
2018-10-16  9:20                               ` Michal Hocko
2018-10-16 11:05                                 ` Tetsuo Handa
2018-10-16 11:17                                   ` Michal Hocko

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='CACT4Y+bQOr=+e8y5so3V9K390uJnMy4+FR7_t9G7Fjn12S4hoA@mail.gmail.com' \
    --to=dvyukov@google.com \
    --cc=akpm@linux-foundation.org \
    --cc=guro@fb.com \
    --cc=hannes@cmpxchg.org \
    --cc=kirill.shutemov@linux.intel.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mhocko@kernel.org \
    --cc=penguin-kernel@i-love.sakura.ne.jp \
    --cc=pmladek@suse.com \
    --cc=rientjes@google.com \
    --cc=sergey.senozhatsky.work@gmail.com \
    --cc=sergey.senozhatsky@gmail.com \
    --cc=syzbot+77e6b28a7a7106ad0def@syzkaller.appspotmail.com \
    --cc=syzkaller-bugs@googlegroups.com \
    --cc=yang.s@alibaba-inc.com \
    --subject='Re: INFO: rcu detected stall in shmem_fault' \
    /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

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.