From: Chad Talbott <ctalbott@google.com>
To: linux-kernel@vger.kernel.org
Cc: linux-mm@kvack.org, wfg@mail.ustc.edu.cn,
Martin Bligh <mbligh@google.com>,
Michael Rubin <mrubin@google.com>,
Andrew Morton <akpm@google.com>,
sandeen@redhat.com
Subject: Bug in kernel 2.6.31, Slow wb_kupdate writeout
Date: Tue, 28 Jul 2009 12:11:31 -0700 [thread overview]
Message-ID: <1786ab030907281211x6e432ba6ha6afe9de73f24e0c@mail.gmail.com> (raw)
I run a simple workload on a 4GB machine which dirties a few largish
inodes like so:
# seq 10 | xargs -P0 -n1 -i\{} dd if=/dev/zero of=/tmp/dump\{}
bs=1024k count=100
While the dds are running data is written out at disk speed. However,
once the dds have run to completion and exited there is ~500MB of
dirty memory left. Background writeout then takes about 3 more
minutes to clean memory at only ~3.3MB/s. When I explicitly sync, I
can see that the disk is capable of 40MB/s, which finishes off the
files in ~10s. [1]
An interesting recent-ish change is "writeback: speed up writeback of
big dirty files." When I revert the change to __sync_single_inode the
problem appears to go away and background writeout proceeds at disk
speed. Interestingly, that code is in the git commit [2], but not in
the post to LKML. [3] This is may not be the fix, but it makes this
test behave better.
Thanks,
Chad
[1] I've plotted the dirty memory from /proc/meminfo and disk write
speed from iostat at
http://sites.google.com/site/cwtlinux/2-6-31-writeback-bug
[2] git commit:
http://mirror.celinuxforum.org/gitstat/commit-detail.php?commit=8bc3be2751b4f74ab90a446da1912fd8204d53f7
[3] LKML post: http://marc.info/?l=linux-kernel&m=119131601130372&w=2
--
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>
next reply other threads:[~2009-07-28 19:11 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-07-28 19:11 Chad Talbott [this message]
2009-07-28 21:49 ` Bug in kernel 2.6.31, Slow wb_kupdate writeout Martin Bligh
2009-07-29 7:15 ` Martin Bligh
2009-07-29 11:43 ` Wu Fengguang
2009-07-29 14:11 ` Martin Bligh
2009-07-30 1:06 ` Wu Fengguang
2009-07-30 1:12 ` Martin Bligh
2009-07-30 1:57 ` Wu Fengguang
2009-07-30 2:59 ` Martin Bligh
2009-07-30 4:08 ` Wu Fengguang
2009-07-30 19:55 ` Martin Bligh
2009-08-01 2:02 ` Wu Fengguang
2009-07-30 0:19 ` Martin Bligh
2009-07-30 1:28 ` Martin Bligh
2009-07-30 2:09 ` Wu Fengguang
2009-07-30 2:57 ` Martin Bligh
2009-07-30 3:19 ` Wu Fengguang
2009-07-30 20:33 ` Martin Bligh
2009-08-01 2:58 ` Wu Fengguang
2009-08-01 4:10 ` Wu Fengguang
2009-07-30 1:49 ` Wu Fengguang
2009-07-30 21:39 ` Jens Axboe
2009-07-30 22:01 ` Martin Bligh
2009-07-30 22:17 ` Jens Axboe
2009-07-30 22:34 ` Martin Bligh
2009-07-30 22:43 ` Jens Axboe
2009-07-30 22:48 ` Martin Bligh
2009-07-31 7:50 ` Peter Zijlstra
2009-08-01 4:03 ` Wu Fengguang
2009-08-01 4:53 ` Wu Fengguang
2009-08-01 5:03 ` Wu Fengguang
2009-08-01 4:02 ` Wu Fengguang
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=1786ab030907281211x6e432ba6ha6afe9de73f24e0c@mail.gmail.com \
--to=ctalbott@google.com \
--cc=akpm@google.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mbligh@google.com \
--cc=mrubin@google.com \
--cc=sandeen@redhat.com \
--cc=wfg@mail.ustc.edu.cn \
/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).