linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Rik van Riel <riel@redhat.com>
To: Mel Gorman <mgorman@suse.de>
Cc: Andrea Arcangeli <aarcange@redhat.com>,
	linux-mm@kvack.org, Hugh Dickins <hughd@google.com>,
	Richard Davies <richard@arachsys.com>,
	Shaohua Li <shli@kernel.org>, Rafael Aquini <aquini@redhat.com>
Subject: Re: [PATCH 1/7] mm: remove ZONE_RECLAIM_LOCKED
Date: Thu, 06 Jun 2013 13:37:28 -0400	[thread overview]
Message-ID: <51B0C8D8.7070708@redhat.com> (raw)
In-Reply-To: <20130606090430.GC1936@suse.de>

On 06/06/2013 05:04 AM, Mel Gorman wrote:
> On Wed, Jun 05, 2013 at 05:10:31PM +0200, Andrea Arcangeli wrote:
>> Zone reclaim locked breaks zone_reclaim_mode=1. If more than one
>> thread allocates memory at the same time, it forces a premature
>> allocation into remote NUMA nodes even when there's plenty of clean
>> cache to reclaim in the local nodes.
>>
>> Signed-off-by: Andrea Arcangeli <aarcange@redhat.com>
>
> Be aware that after this patch is applied that it is possible to have a
> situation like this
>
> 1. 4 processes running on node 1
> 2. Each process tries to allocate 30% of memory
> 3. Each process reads the full buffer in a loop (stupid, just an example)
>
> In this situation the processes will continually interfere with each
> other until one of them gets migrated to another zone by the scheduler.

This is a very good point.

Andrea, I suspect we will need some kind of safeguard against
this problem.

I can see how ZONE_RECLAIM_LOCKED may not be that safeguard,
but I do not have any good ideas on what it would be.

Should we limit zone reclaim to priority == DEF_PRIORITY, and
fall back if we fail to free enough memory at DEF_PRIORITY?

Does anybody have better ideas?

--
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>

  reply	other threads:[~2013-06-06 17:40 UTC|newest]

Thread overview: 48+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-06-05 15:10 [PATCH 0/7] RFC: adding compaction to zone_reclaim_mode > 0 Andrea Arcangeli
2013-06-05 15:10 ` [PATCH 1/7] mm: remove ZONE_RECLAIM_LOCKED Andrea Arcangeli
2013-06-05 19:23   ` Rik van Riel
2013-06-05 20:37   ` KOSAKI Motohiro
2013-06-05 20:51     ` Christoph Lameter
2013-06-05 21:03       ` KOSAKI Motohiro
2013-06-06 14:15         ` Christoph Lameter
2013-06-06 17:17           ` KOSAKI Motohiro
2013-06-06 18:16             ` Christoph Lameter
2013-06-05 21:33   ` Rafael Aquini
2013-06-06  9:04   ` Mel Gorman
2013-06-06 17:37     ` Rik van Riel [this message]
2013-06-14 16:16       ` Rik van Riel
2013-06-17  9:30         ` Mel Gorman
2013-06-17 18:12           ` Rik van Riel
2013-06-26 20:10             ` Andrea Arcangeli
2013-06-05 15:10 ` [PATCH 2/7] mm: compaction: scan all memory with /proc/sys/vm/compact_memory Andrea Arcangeli
2013-06-05 19:34   ` Rik van Riel
2013-06-05 21:39   ` Rafael Aquini
2013-06-06  9:05   ` Mel Gorman
2013-06-05 15:10 ` [PATCH 3/7] mm: compaction: don't depend on kswapd to invoke reset_isolation_suitable Andrea Arcangeli
2013-06-05 19:49   ` Rik van Riel
2013-06-26 20:38     ` Andrea Arcangeli
2013-06-06  9:11   ` Mel Gorman
2013-06-26 20:48     ` Andrea Arcangeli
2013-06-06 12:47   ` Rafael Aquini
2013-06-05 15:10 ` [PATCH 4/7] mm: compaction: reset before initializing the scan cursors Andrea Arcangeli
2013-06-05 20:04   ` Rik van Riel
2013-06-06  9:14   ` Mel Gorman
2013-06-06 12:49   ` Rafael Aquini
2013-06-05 15:10 ` [PATCH 5/7] mm: compaction: increase the high order pages in the watermarks Andrea Arcangeli
2013-06-05 20:18   ` Rik van Riel
2013-06-28 22:14     ` Andrea Arcangeli
2013-06-06  9:19   ` Mel Gorman
2013-06-05 15:10 ` [PATCH 6/7] mm: compaction: export compact_zone_order() Andrea Arcangeli
2013-06-05 20:24   ` Rik van Riel
2013-06-05 15:10 ` [PATCH 7/7] mm: compaction: add compaction to zone_reclaim_mode Andrea Arcangeli
2013-06-05 22:21   ` Rik van Riel
2013-06-06 10:05   ` Mel Gorman
2013-07-11 16:02     ` Andrea Arcangeli
2013-07-12 12:26       ` Hush Bensen
2013-07-12 16:01         ` Andrea Arcangeli
2013-07-12 23:23           ` Hush Bensen
2013-07-15  9:16             ` Andrea Arcangeli
2013-07-12 23:57   ` Hush Bensen
2013-07-15  9:25     ` Andrea Arcangeli
2013-06-06  8:53 ` [PATCH 0/7] RFC: adding compaction to zone_reclaim_mode > 0 Mel Gorman
2013-06-06 10:09 ` Mel Gorman

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=51B0C8D8.7070708@redhat.com \
    --to=riel@redhat.com \
    --cc=aarcange@redhat.com \
    --cc=aquini@redhat.com \
    --cc=hughd@google.com \
    --cc=linux-mm@kvack.org \
    --cc=mgorman@suse.de \
    --cc=richard@arachsys.com \
    --cc=shli@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).