From: Peter Zijlstra <peterz@infradead.org>
To: Jens Axboe <jens.axboe@oracle.com>
Cc: linux-kernel@vger.kernel.org, fengguang.wu@intel.com,
shaohua.li@intel.com, chris.mason@oracle.com, jack@suse.cz
Subject: Re: [PATCH 02/10] writeback: balance_dirty_pages() shall write more than dirtied pages
Date: Thu, 24 Sep 2009 17:03:34 +0200 [thread overview]
Message-ID: <1253804614.18939.35.camel@laptop> (raw)
In-Reply-To: <1253803236-20760-3-git-send-email-jens.axboe@oracle.com>
On Thu, 2009-09-24 at 16:40 +0200, Jens Axboe wrote:
> From: Wu Fengguang <fengguang.wu@intel.com>
>
> Some filesystem may choose to write much more than ratelimit_pages
> before calling balance_dirty_pages_ratelimited_nr(). So it is safer to
> determine number to write based on real number of dirtied pages.
>
> Otherwise it is possible that
> loop {
> btrfs_file_write(): dirty 1024 pages
> balance_dirty_pages(): write up to 48 pages (= ratelimit_pages * 1.5)
> }
> in which the writeback rate cannot keep up with dirty rate, and the
> dirty pages go all the way beyond dirty_thresh.
>
> The increased write_chunk may make the dirtier more bumpy.
> So filesystems shall be take care not to dirty too much at
> a time (eg. > 4MB) without checking the ratelimit.
>
> Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
next prev parent reply other threads:[~2009-09-24 15:03 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-09-24 14:40 [PATCH 0/10] Current writeback patch queue Jens Axboe
2009-09-24 14:40 ` [PATCH 01/10] fs: Fix busyloop in wb_writeback() Jens Axboe
2009-09-25 0:48 ` Wu Fengguang
2009-09-24 14:40 ` [PATCH 02/10] writeback: balance_dirty_pages() shall write more than dirtied pages Jens Axboe
2009-09-24 15:03 ` Peter Zijlstra [this message]
2009-09-24 14:40 ` [PATCH 03/10] writeback: stop background writeback when below background threshold Jens Axboe
2009-09-24 15:03 ` Peter Zijlstra
2009-09-24 15:18 ` Peter Zijlstra
2009-09-24 16:13 ` Jens Axboe
2009-09-24 16:26 ` Peter Zijlstra
2009-09-25 0:47 ` Wu Fengguang
2009-09-24 14:40 ` [PATCH 04/10] writeback: kupdate writeback shall not stop when more io is possible Jens Axboe
2009-09-24 14:40 ` [PATCH 05/10] writeback: cleanup writeback_single_inode() Jens Axboe
2009-09-24 14:40 ` [PATCH 06/10] writeback: improve readability of the wb_writeback() continue/break logic Jens Axboe
2009-09-25 0:52 ` Wu Fengguang
2009-09-24 14:40 ` [PATCH 07/10] writeback: get rid to incorrect references to pdflush in comments Jens Axboe
2009-09-24 14:40 ` [PATCH 08/10] writeback: move inodes from one super_block together Jens Axboe
2009-09-24 14:40 ` [PATCH 09/10] writeback: don't resort for a single super_block in move_expired_inodes() Jens Axboe
2009-09-24 14:40 ` [PATCH 10/10] writeback: make the super_block pinning more efficient Jens Axboe
2009-09-25 2:55 ` [PATCH 0/10] Current writeback patch queue Wu Fengguang
2009-09-25 4:06 ` Jens Axboe
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=1253804614.18939.35.camel@laptop \
--to=peterz@infradead.org \
--cc=chris.mason@oracle.com \
--cc=fengguang.wu@intel.com \
--cc=jack@suse.cz \
--cc=jens.axboe@oracle.com \
--cc=linux-kernel@vger.kernel.org \
--cc=shaohua.li@intel.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