From: Wu Fengguang <fengguang.wu@intel.com>
To: Mel Gorman <mel@csn.ul.ie>
Cc: Andrew Morton <akpm@linux-foundation.org>,
linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org,
linux-mm@kvack.org, Dave Chinner <david@fromorbit.com>,
Chris Mason <chris.mason@oracle.com>,
Nick Piggin <npiggin@suse.de>, Rik van Riel <riel@redhat.com>,
Johannes Weiner <hannes@cmpxchg.org>,
Christoph Hellwig <hch@infradead.org>,
KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>,
KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>,
Andrea Arcangeli <aarcange@redhat.com>, Jan Kara <jack@suse.cz>
Subject: Re: [PATCH 12/14] vmscan: Do not writeback pages in direct reclaim
Date: Wed, 7 Jul 2010 13:03:38 +0800 [thread overview]
Message-ID: <20100707050338.GA5039@localhost> (raw)
In-Reply-To: <20100705134949.GC13780@csn.ul.ie>
Hi Mel,
> Second, using systemtap, I was able to see that file-backed dirty
> pages have a tendency to be near the end of the LRU even though they
> are a small percentage of the overall pages in the LRU. I'm hoping
> to figure out why this is as it would make avoiding writeback a lot
> less controversial.
Your intuitions are correct -- the current background writeback logic
fails to write elder inodes first. Under heavy loads the background
writeback job may run for ever, totally ignoring the time order of
inode->dirtied_when. This is probably why you see lots of dirty pages
near the end of LRU.
Here is an old patch for fixing this. Sorry for being late. I'll
pick up and refresh the patch series ASAP. (I made a mistake last
year to post too many patches at one time. I'll break them up into
more manageable pieces.)
[PATCH 31/45] writeback: sync old inodes first in background writeback
<https://kerneltrap.org/mailarchive/linux-fsdevel/2009/10/7/6476313>
Thanks,
Fengguang
next prev parent reply other threads:[~2010-07-07 5:03 UTC|newest]
Thread overview: 51+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-06-29 11:34 [PATCH 0/14] Avoid overflowing of stack during page reclaim V3 Mel Gorman
2010-06-29 11:34 ` [PATCH 01/14] vmscan: Fix mapping use after free Mel Gorman
2010-06-29 14:27 ` Minchan Kim
2010-07-01 9:53 ` Mel Gorman
2010-06-29 14:44 ` Johannes Weiner
2010-06-29 11:34 ` [PATCH 02/14] tracing, vmscan: Add trace events for kswapd wakeup, sleeping and direct reclaim Mel Gorman
2010-06-29 11:34 ` [PATCH 03/14] tracing, vmscan: Add trace events for LRU page isolation Mel Gorman
2010-06-29 11:34 ` [PATCH 04/14] tracing, vmscan: Add trace event when a page is written Mel Gorman
2010-06-29 11:34 ` [PATCH 05/14] tracing, vmscan: Add a postprocessing script for reclaim-related ftrace events Mel Gorman
2010-06-29 11:34 ` [PATCH 06/14] vmscan: kill prev_priority completely Mel Gorman
2010-06-29 11:34 ` [PATCH 07/14] vmscan: simplify shrink_inactive_list() Mel Gorman
2010-06-29 11:34 ` [PATCH 08/14] vmscan: Remove unnecessary temporary vars in do_try_to_free_pages Mel Gorman
2010-06-29 11:34 ` [PATCH 09/14] vmscan: Setup pagevec as late as possible in shrink_inactive_list() Mel Gorman
2010-06-29 11:34 ` [PATCH 10/14] vmscan: Setup pagevec as late as possible in shrink_page_list() Mel Gorman
2010-06-29 11:34 ` [PATCH 11/14] vmscan: Update isolated page counters outside of main path in shrink_inactive_list() Mel Gorman
2010-06-29 11:34 ` [PATCH 12/14] vmscan: Do not writeback pages in direct reclaim Mel Gorman
2010-07-02 19:51 ` Andrew Morton
2010-07-05 13:49 ` Mel Gorman
2010-07-06 0:36 ` KOSAKI Motohiro
2010-07-06 5:46 ` Minchan Kim
2010-07-06 6:02 ` KOSAKI Motohiro
2010-07-06 6:38 ` Minchan Kim
2010-07-06 10:12 ` Mel Gorman
2010-07-06 11:13 ` KOSAKI Motohiro
2010-07-06 11:24 ` Minchan Kim
2010-07-06 15:25 ` Mel Gorman
2010-07-06 20:27 ` Johannes Weiner
2010-07-06 22:28 ` Minchan Kim
2010-07-07 0:24 ` Mel Gorman
2010-07-07 1:15 ` Christoph Hellwig
2010-07-07 9:43 ` Mel Gorman
2010-07-07 12:51 ` Rik van Riel
2010-07-07 1:14 ` Christoph Hellwig
2010-07-08 6:39 ` KOSAKI Motohiro
2010-07-07 5:03 ` Wu Fengguang [this message]
2010-07-07 9:50 ` Mel Gorman
2010-07-07 18:09 ` Christoph Hellwig
2010-06-29 11:34 ` [PATCH 13/14] fs,btrfs: Allow kswapd to writeback pages Mel Gorman
2010-06-30 13:05 ` Chris Mason
2010-07-01 9:55 ` Mel Gorman
2010-06-29 11:34 ` [PATCH 14/14] fs,xfs: " Mel Gorman
2010-06-29 12:37 ` Christoph Hellwig
2010-06-29 12:51 ` Mel Gorman
2010-06-30 0:14 ` KAMEZAWA Hiroyuki
2010-07-01 10:30 ` Mel Gorman
2010-07-02 6:26 ` KAMEZAWA Hiroyuki
2010-07-02 6:31 ` KAMEZAWA Hiroyuki
2010-07-05 14:16 ` Mel Gorman
2010-07-06 0:45 ` KAMEZAWA Hiroyuki
2010-07-02 19:33 ` [PATCH 0/14] Avoid overflowing of stack during page reclaim V3 Andrew Morton
2010-07-05 1:35 ` KAMEZAWA Hiroyuki
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=20100707050338.GA5039@localhost \
--to=fengguang.wu@intel.com \
--cc=aarcange@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=chris.mason@oracle.com \
--cc=david@fromorbit.com \
--cc=hannes@cmpxchg.org \
--cc=hch@infradead.org \
--cc=jack@suse.cz \
--cc=kamezawa.hiroyu@jp.fujitsu.com \
--cc=kosaki.motohiro@jp.fujitsu.com \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mel@csn.ul.ie \
--cc=npiggin@suse.de \
--cc=riel@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 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).