From: Mel Gorman <mgorman@suse.de>
To: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Johannes Weiner <hannes@cmpxchg.org>,
Rik van Riel <riel@redhat.com>,
Andrew Morton <akpm@linux-foundation.org>,
George Spelvin <linux@horizon.com>,
Johannes Hirte <johannes.hirte@fem.tu-ilmenau.de>,
Tomas Racek <tracek@redhat.com>, Jan Kara <jack@suse.cz>,
Dave Hansen <dave@linux.vnet.ibm.com>,
Josh Boyer <jwboyer@gmail.com>,
Valdis Kletnieks <Valdis.Kletnieks@vt.edu>,
Jiri Slaby <jslaby@suse.cz>,
Thorsten Leemhuis <fedora@leemhuis.info>,
Zdenek Kabelac <zkabelac@redhat.com>,
Bruno Wolff III <bruno@wolff.to>, linux-mm <linux-mm@kvack.org>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
Subject: Re: kswapd craziness in 3.7
Date: Wed, 28 Nov 2012 16:42:18 +0000 [thread overview]
Message-ID: <20121128164217.GV8218@suse.de> (raw)
In-Reply-To: <20121128101359.GT8218@suse.de>
On Wed, Nov 28, 2012 at 10:13:59AM +0000, Mel Gorman wrote:
> On Tue, Nov 27, 2012 at 03:19:38PM -0800, Linus Torvalds wrote:
> > On Tue, Nov 27, 2012 at 2:26 PM, Johannes Weiner <hannes@cmpxchg.org> wrote:
> > > On Tue, Nov 27, 2012 at 05:02:36PM -0500, Rik van Riel wrote:
> > >>
> > >> Kswapd going crazy is certainly a large part of the problem.
> > >>
> > >> However, that leaves the issue of page_alloc.c waking up
> > >> kswapd when the system is not actually low on memory.
> > >>
> > >> Instead, kswapd is woken up because memory compaction failed,
> > >> potentially even due to lock contention during compaction!
> > >>
> > >> Ideally the allocation code would only wake up kswapd if
> > >> memory needs to be freed, or in order for kswapd to do
> > >> memory compaction (so the allocator does not have to).
> > >
> > > Maybe I missed something, but shouldn't this be solved with my patch?
> >
> > Ok, guys. Cage fight!
> >
> > The rules are simple: two men enter, one man leaves.
> >
>
> I'm fairly scorch damaged from this whole cycle already. I won't need a
> prop master to look the part for a thunderdome match.
>
> > And the one who comes out gets to explain to me which patch(es) I
> > should apply, and which I should revert, if any.
> >
>
> Based on the reports I've seen I expect the following to work for 3.7
>
> Keep
> 96710098 mm: revert "mm: vmscan: scale number of pages reclaimed by reclaim/compaction based on failures"
> ef6c5be6 fix incorrect NR_FREE_PAGES accounting (appears like memory leak)
>
> Revert
> 82b212f4 Revert "mm: remove __GFP_NO_KSWAPD"
>
> Merge
> mm: vmscan: fix kswapd endless loop on higher order allocation
> mm: Avoid waking kswapd for THP allocations when compaction is deferred or contended
>
and
mm: compaction: Fix return value of capture_free_page
but this one may already be in flight from Andrew's tree as he picked it
up already.
--
Mel Gorman
SUSE Labs
--
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/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
WARNING: multiple messages have this Message-ID (diff)
From: Mel Gorman <mgorman@suse.de>
To: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Johannes Weiner <hannes@cmpxchg.org>,
Rik van Riel <riel@redhat.com>,
Andrew Morton <akpm@linux-foundation.org>,
George Spelvin <linux@horizon.com>,
Johannes Hirte <johannes.hirte@fem.tu-ilmenau.de>,
Tomas Racek <tracek@redhat.com>, Jan Kara <jack@suse.cz>,
Dave Hansen <dave@linux.vnet.ibm.com>,
Josh Boyer <jwboyer@gmail.com>,
Valdis Kletnieks <Valdis.Kletnieks@vt.edu>,
Jiri Slaby <jslaby@suse.cz>,
Thorsten Leemhuis <fedora@leemhuis.info>,
Zdenek Kabelac <zkabelac@redhat.com>,
Bruno Wolff III <bruno@wolff.to>, linux-mm <linux-mm@kvack.org>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
Subject: Re: kswapd craziness in 3.7
Date: Wed, 28 Nov 2012 16:42:18 +0000 [thread overview]
Message-ID: <20121128164217.GV8218@suse.de> (raw)
In-Reply-To: <20121128101359.GT8218@suse.de>
On Wed, Nov 28, 2012 at 10:13:59AM +0000, Mel Gorman wrote:
> On Tue, Nov 27, 2012 at 03:19:38PM -0800, Linus Torvalds wrote:
> > On Tue, Nov 27, 2012 at 2:26 PM, Johannes Weiner <hannes@cmpxchg.org> wrote:
> > > On Tue, Nov 27, 2012 at 05:02:36PM -0500, Rik van Riel wrote:
> > >>
> > >> Kswapd going crazy is certainly a large part of the problem.
> > >>
> > >> However, that leaves the issue of page_alloc.c waking up
> > >> kswapd when the system is not actually low on memory.
> > >>
> > >> Instead, kswapd is woken up because memory compaction failed,
> > >> potentially even due to lock contention during compaction!
> > >>
> > >> Ideally the allocation code would only wake up kswapd if
> > >> memory needs to be freed, or in order for kswapd to do
> > >> memory compaction (so the allocator does not have to).
> > >
> > > Maybe I missed something, but shouldn't this be solved with my patch?
> >
> > Ok, guys. Cage fight!
> >
> > The rules are simple: two men enter, one man leaves.
> >
>
> I'm fairly scorch damaged from this whole cycle already. I won't need a
> prop master to look the part for a thunderdome match.
>
> > And the one who comes out gets to explain to me which patch(es) I
> > should apply, and which I should revert, if any.
> >
>
> Based on the reports I've seen I expect the following to work for 3.7
>
> Keep
> 96710098 mm: revert "mm: vmscan: scale number of pages reclaimed by reclaim/compaction based on failures"
> ef6c5be6 fix incorrect NR_FREE_PAGES accounting (appears like memory leak)
>
> Revert
> 82b212f4 Revert "mm: remove __GFP_NO_KSWAPD"
>
> Merge
> mm: vmscan: fix kswapd endless loop on higher order allocation
> mm: Avoid waking kswapd for THP allocations when compaction is deferred or contended
>
and
mm: compaction: Fix return value of capture_free_page
but this one may already be in flight from Andrew's tree as he picked it
up already.
--
Mel Gorman
SUSE Labs
next prev parent reply other threads:[~2012-11-28 16:42 UTC|newest]
Thread overview: 129+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-11-27 20:48 kswapd craziness in 3.7 Johannes Weiner
2012-11-27 20:48 ` Johannes Weiner
2012-11-27 20:48 ` [patch] mm: vmscan: fix kswapd endless loop on higher order allocation Johannes Weiner
2012-11-27 20:48 ` Johannes Weiner
2012-11-27 20:58 ` kswapd craziness in 3.7 Linus Torvalds
2012-11-27 20:58 ` Linus Torvalds
2012-11-27 21:16 ` Rik van Riel
2012-11-27 21:16 ` Rik van Riel
2012-11-27 21:49 ` Johannes Weiner
2012-11-27 21:49 ` Johannes Weiner
2012-11-27 22:02 ` Rik van Riel
2012-11-27 22:02 ` Rik van Riel
2012-11-27 22:26 ` Johannes Weiner
2012-11-27 22:26 ` Johannes Weiner
2012-11-27 23:19 ` Linus Torvalds
2012-11-27 23:19 ` Linus Torvalds
2012-11-28 10:13 ` Mel Gorman
2012-11-28 10:13 ` Mel Gorman
2012-11-28 10:51 ` Thorsten Leemhuis
2012-11-28 10:51 ` Thorsten Leemhuis
2012-11-28 16:42 ` Mel Gorman [this message]
2012-11-28 16:42 ` Mel Gorman
2012-11-28 22:52 ` Andrew Morton
2012-11-28 22:52 ` Andrew Morton
2012-11-28 23:54 ` Mel Gorman
2012-11-28 23:54 ` Mel Gorman
2012-11-29 0:14 ` Andrew Morton
2012-11-29 0:14 ` Andrew Morton
2012-11-29 15:30 ` Thorsten Leemhuis
2012-11-29 15:30 ` Thorsten Leemhuis
2012-11-29 17:05 ` Johannes Weiner
2012-11-29 17:05 ` Johannes Weiner
2012-11-30 12:39 ` Thorsten Leemhuis
2012-11-30 12:39 ` Thorsten Leemhuis
2012-12-01 0:45 ` Johannes Weiner
2012-12-01 0:45 ` Johannes Weiner
2012-12-03 8:30 ` Thorsten Leemhuis
2012-12-03 8:30 ` Thorsten Leemhuis
2012-12-03 13:08 ` Fedora repo (was: Re: kswapd craziness in 3.7) Borislav Petkov
2012-12-03 13:08 ` Borislav Petkov
2012-12-03 19:42 ` kswapd craziness in 3.7 Johannes Weiner
2012-12-03 19:42 ` Johannes Weiner
2012-12-04 21:42 ` Johannes Weiner
2012-12-04 21:42 ` Johannes Weiner
2012-12-05 3:01 ` Bruno Wolff III
2012-12-05 3:01 ` Bruno Wolff III
2012-12-06 17:37 ` Bruno Wolff III
2012-12-06 17:37 ` Bruno Wolff III
2012-12-06 19:31 ` Linus Torvalds
2012-12-06 19:31 ` Linus Torvalds
2012-12-06 19:43 ` Rik van Riel
2012-12-06 19:43 ` Rik van Riel
2012-12-06 20:23 ` Johannes Weiner
2012-12-06 20:23 ` Johannes Weiner
2012-12-06 20:32 ` Rik van Riel
2012-12-06 20:32 ` Rik van Riel
2012-12-08 12:06 ` Zlatko Calusic
2012-12-08 12:06 ` Zlatko Calusic
2012-12-08 21:22 ` Zlatko Calusic
2012-12-08 21:22 ` Zlatko Calusic
2012-12-09 1:01 ` Linus Torvalds
2012-12-09 1:01 ` Linus Torvalds
2012-12-09 21:59 ` Zdenek Kabelac
2012-12-09 21:59 ` Zdenek Kabelac
2012-12-10 11:03 ` Mel Gorman
2012-12-10 11:03 ` Mel Gorman
2012-12-10 16:39 ` Johannes Weiner
2012-12-10 16:39 ` Johannes Weiner
2012-12-10 18:01 ` Mel Gorman
2012-12-10 18:01 ` Mel Gorman
2012-12-10 18:33 ` Zlatko Calusic
2012-12-10 18:33 ` Zlatko Calusic
2012-12-10 19:13 ` Linus Torvalds
2012-12-10 19:13 ` Linus Torvalds
2012-12-10 20:35 ` Zlatko Calusic
2012-12-10 20:35 ` Zlatko Calusic
2012-12-10 21:28 ` Linus Torvalds
2012-12-10 21:28 ` Linus Torvalds
2012-12-10 21:42 ` Borislav Petkov
2012-12-10 21:42 ` Borislav Petkov
2012-12-10 21:47 ` Linus Torvalds
2012-12-10 21:47 ` Linus Torvalds
2012-12-10 21:54 ` Borislav Petkov
2012-12-10 21:54 ` Borislav Petkov
2012-12-10 22:15 ` Zlatko Calusic
2012-12-10 22:15 ` Zlatko Calusic
2012-12-10 23:27 ` Hugh Dickins
2012-12-10 23:27 ` Hugh Dickins
2012-12-11 0:19 ` Zlatko Calusic
2012-12-11 21:56 ` Zlatko Calusic
2012-12-11 21:56 ` Zlatko Calusic
2012-12-19 22:24 ` Zlatko Calusic
2012-12-19 22:24 ` Zlatko Calusic
2012-12-10 18:29 ` Zlatko Calusic
2012-12-10 18:29 ` Zlatko Calusic
2012-12-06 8:09 ` Thorsten Leemhuis
2012-12-06 8:09 ` Thorsten Leemhuis
2012-11-27 21:29 ` Johannes Weiner
2012-11-27 21:29 ` Johannes Weiner
2012-11-28 13:35 ` Zdenek Kabelac
2012-11-28 13:35 ` Zdenek Kabelac
2012-11-28 14:04 ` Jiri Slaby
2012-11-28 14:04 ` Jiri Slaby
2012-11-28 9:45 ` Mel Gorman
2012-11-28 9:45 ` Mel Gorman
2012-12-03 15:23 ` Zdenek Kabelac
2012-12-03 15:23 ` Zdenek Kabelac
2012-12-03 19:18 ` Johannes Weiner
2012-12-03 19:18 ` Johannes Weiner
2012-12-04 9:05 ` Zdenek Kabelac
2012-12-04 9:05 ` Zdenek Kabelac
2012-12-04 9:15 ` Jiri Slaby
2012-12-04 9:15 ` Jiri Slaby
2012-12-04 16:11 ` Johannes Weiner
2012-12-04 16:11 ` Johannes Weiner
2012-12-04 16:22 ` Jiri Slaby
2012-12-04 16:22 ` Jiri Slaby
2012-12-04 19:50 ` Johannes Weiner
2012-12-04 19:50 ` Johannes Weiner
2012-12-08 10:35 ` Jiri Slaby
2012-12-08 10:35 ` Jiri Slaby
2012-12-04 16:15 ` Johannes Weiner
2012-12-04 16:15 ` Johannes Weiner
2012-12-06 13:51 ` Zdenek Kabelac
2012-12-06 13:51 ` Zdenek Kabelac
2012-12-03 13:14 ` Jiri Slaby
2012-12-03 13:14 ` Jiri Slaby
2012-12-04 8:55 ` Jiri Slaby
2012-12-04 8:55 ` Jiri Slaby
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=20121128164217.GV8218@suse.de \
--to=mgorman@suse.de \
--cc=Valdis.Kletnieks@vt.edu \
--cc=akpm@linux-foundation.org \
--cc=bruno@wolff.to \
--cc=dave@linux.vnet.ibm.com \
--cc=fedora@leemhuis.info \
--cc=hannes@cmpxchg.org \
--cc=jack@suse.cz \
--cc=johannes.hirte@fem.tu-ilmenau.de \
--cc=jslaby@suse.cz \
--cc=jwboyer@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=linux@horizon.com \
--cc=riel@redhat.com \
--cc=torvalds@linux-foundation.org \
--cc=tracek@redhat.com \
--cc=zkabelac@redhat.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 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.