linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Colin Cross <ccross@android.com>
To: Minchan Kim <minchan.kim@gmail.com>
Cc: Mel Gorman <mgorman@suse.de>,
	Andrew Morton <akpm@linux-foundation.org>,
	Pekka Enberg <penberg@cs.helsinki.fi>,
	KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>,
	Andrea Arcangeli <aarcange@redhat.com>,
	David Rientjes <rientjes@google.com>,
	LKML <linux-kernel@vger.kernel.org>,
	Linux-MM <linux-mm@kvack.org>
Subject: Re: [PATCH] mm: avoid livelock on !__GFP_FS allocations
Date: Tue, 15 Nov 2011 16:28:36 -0800	[thread overview]
Message-ID: <CAMbhsRSePzsN-4JXEEwFoaa9EhBfHQ11gsjqJCDzV2nonJ0DqQ@mail.gmail.com> (raw)
In-Reply-To: <20111116002235.GA10958@barrios-laptop.redhat.com>

On Tue, Nov 15, 2011 at 4:22 PM, Minchan Kim <minchan.kim@gmail.com> wrote:
> On Tue, Nov 15, 2011 at 05:36:56PM +0000, Mel Gorman wrote:
>> On Wed, Nov 16, 2011 at 01:13:30AM +0900, Minchan Kim wrote:
>> The impact would be that during the time between processes been frozen
>> and storage being suspended, GFP_NOIO allocations that used to call
>> wait_iff_congested and retry while kswapd does its thing will return
>> failure instead. These GFP_NOIO allocations that used to succeed will
>> now fail in rare cases during suspend and I don't think we want that.
>>
>> Is this what you meant or had you something else in mind?
>>
>
> You read my mind exactly!
>
> I thought hibernation process is as follows,
>
> freeze user processes
> oom_disable
> hibernate_preallocate_memory
> freeze kernel processes(include kswapd)
> pm_restrict_gfp_mask
> swsusp_save
>
> My guessing is hibernate_prealocate_memory should reserve all memory needed
> for hibernation for reclaimaing pages of kswapd because kswapd just would be
> stopped so during swsusp_save, page reclaim should not be occured.
>
> But being see description of patch, my guess seems wrong.
> Now the problem happens and it means page reclaim happens during swsusp_save.
> Colin or someone could confirm this?

The problem I see is during suspend, not hibernation.  The particular
allocation that usually causes the problem is the pgd_alloc for page
tables when re-enabling the 2nd cpu during resume, which is odd as
those same page tables were freed during suspend.  I guess an
unfreezable kernel thread allocated that memory between the free and
re-allocation.

> If so, could we reserve more memory when we preallocate hibernation memory
> for avoiding page reclaim without kswapd?
>
> --
> Kind regards,
> Minchan Kim
>

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

  reply	other threads:[~2011-11-16  0:28 UTC|newest]

Thread overview: 47+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-11-14 14:04 [PATCH] mm: avoid livelock on !__GFP_FS allocations Mel Gorman
2011-11-14 18:38 ` Andrea Arcangeli
2011-11-15 10:30   ` Mel Gorman
2011-11-14 23:03 ` Andrew Morton
2011-11-15 10:42   ` Mel Gorman
2011-11-15 15:43     ` Mel Gorman
2011-11-15 16:13 ` Minchan Kim
2011-11-15 17:36   ` Mel Gorman
2011-11-16  0:22     ` Minchan Kim
2011-11-16  0:28       ` Colin Cross [this message]
2011-11-16  0:45         ` Minchan Kim
2011-11-16  7:10           ` Pekka Enberg
2011-11-16 21:44             ` David Rientjes
2011-11-16 21:58               ` Rafael J. Wysocki
2011-11-16 22:07               ` Minchan Kim
2011-11-16 22:48                 ` David Rientjes
2011-11-15 21:40 ` David Rientjes
2011-11-16  9:52   ` Mel Gorman
2011-11-16 21:39     ` David Rientjes
  -- strict thread matches above, loose matches on Subject: below --
2011-10-25  6:39 Colin Cross
2011-10-25  7:40 ` Pekka Enberg
2011-10-25  7:51   ` Colin Cross
2011-10-25  8:08     ` Pekka Enberg
2011-10-25 22:12     ` David Rientjes
2011-10-25  9:09 ` Mel Gorman
2011-10-25  9:26   ` Colin Cross
2011-10-25 11:23     ` Mel Gorman
2011-10-25 17:08       ` Colin Cross
2011-11-01 12:28         ` Mel Gorman
2011-10-25 19:39       ` Pekka Enberg
2011-11-01 12:29         ` Mel Gorman
2011-10-25 19:29   ` Colin Cross
2011-10-25 22:18   ` David Rientjes
2011-10-26  1:46     ` Colin Cross
2011-10-26  5:47       ` David Rientjes
2011-10-26  6:12         ` David Rientjes
2011-10-26  6:16           ` Colin Cross
2011-10-26  6:24             ` David Rientjes
2011-10-26  6:26               ` Colin Cross
2011-10-26  6:33                 ` David Rientjes
2011-10-26  6:36                   ` Colin Cross
2011-10-26  6:51                     ` David Rientjes
2011-10-26  6:57                       ` Colin Cross
2011-10-26  7:10                         ` David Rientjes
2011-10-26  7:22                           ` Colin Cross
2011-11-01 12:36                             ` Mel Gorman
2011-10-25 22:10 ` David Rientjes

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=CAMbhsRSePzsN-4JXEEwFoaa9EhBfHQ11gsjqJCDzV2nonJ0DqQ@mail.gmail.com \
    --to=ccross@android.com \
    --cc=aarcange@redhat.com \
    --cc=akpm@linux-foundation.org \
    --cc=kamezawa.hiroyu@jp.fujitsu.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mgorman@suse.de \
    --cc=minchan.kim@gmail.com \
    --cc=penberg@cs.helsinki.fi \
    --cc=rientjes@google.com \
    /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).