From: Dave Chinner <david@fromorbit.com>
To: Peter Zijlstra <peterz@infradead.org>
Cc: Chris Mason <chris.mason@oracle.com>,
Artem Bityutskiy <dedekind1@gmail.com>,
Jens Axboe <jens.axboe@oracle.com>,
linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org,
hch@infradead.org, akpm@linux-foundation.org, jack@suse.cz,
Theodore Ts'o <tytso@mit.edu>,
Wu Fengguang <fengguang.wu@intel.com>
Subject: Re: [PATCH 8/8] vm: Add an tuning knob for vm.max_writeback_mb
Date: Wed, 9 Sep 2009 11:53:59 +1000 [thread overview]
Message-ID: <20090909015359.GC7146@discord.disaster> (raw)
In-Reply-To: <1252428983.7746.140.camel@twins>
On Tue, Sep 08, 2009 at 06:56:23PM +0200, Peter Zijlstra wrote:
> On Tue, 2009-09-08 at 12:29 -0400, Chris Mason wrote:
> > Either way, if pdflush or the bdi thread or whoever ends up switching to
> > another file during a big streaming write, the end result is that we
> > fragment. We may fragment the file (ext4) or we may fragment the
> > writeback (xfs), but the end result isn't good.
>
> OK, so what we want is for a way to re-enter the whole
> writeback_inodes() path onto the same file, right?
No, that would take use back to the Bad Old Days where one large
file write can starve out the other 10,000 small files that need to
be written. The old writeback code used to end up in this way
because it didn't rotate large files to the back of the dirty inode
queue once wbc->nr_to_write was exhausted. This could cause files
not to be written back for tens of minutes....
Cheers,
Dave.
--
Dave Chinner
david@fromorbit.com
next prev parent reply other threads:[~2009-09-09 1:55 UTC|newest]
Thread overview: 67+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-09-08 9:23 [PATCH 0/8] Per-bdi writeback flusher threads v19 Jens Axboe
2009-09-08 9:23 ` [PATCH 1/8] writeback: get rid of generic_sync_sb_inodes() export Jens Axboe
2009-09-08 10:27 ` Artem Bityutskiy
2009-09-08 10:41 ` Jens Axboe
2009-09-08 10:52 ` Artem Bityutskiy
2009-09-08 10:57 ` Jens Axboe
2009-09-08 11:01 ` Artem Bityutskiy
2009-09-08 11:05 ` Jens Axboe
2009-09-08 11:31 ` Artem Bityutskiy
2009-09-08 9:23 ` [PATCH 2/8] writeback: move dirty inodes from super_block to backing_dev_info Jens Axboe
2009-09-08 9:23 ` [PATCH 3/8] writeback: switch to per-bdi threads for flushing data Jens Axboe
2009-09-08 13:46 ` Daniel Walker
2009-09-08 14:21 ` Jens Axboe
2009-09-08 9:23 ` [PATCH 4/8] writeback: get rid of pdflush completely Jens Axboe
2009-09-08 9:23 ` [PATCH 5/8] writeback: add some debug inode list counters to bdi stats Jens Axboe
2009-09-08 9:23 ` [PATCH 6/8] writeback: add name to backing_dev_info Jens Axboe
2009-09-08 9:23 ` [PATCH 7/8] writeback: check for registered bdi in flusher add and inode dirty Jens Axboe
2009-09-08 9:23 ` [PATCH 8/8] vm: Add an tuning knob for vm.max_writeback_mb Jens Axboe
2009-09-08 10:37 ` Artem Bityutskiy
2009-09-08 16:06 ` Peter Zijlstra
2009-09-08 16:29 ` Chris Mason
2009-09-08 16:56 ` Peter Zijlstra
2009-09-08 17:28 ` Chris Mason
2009-09-08 17:46 ` Peter Zijlstra
2009-09-08 17:55 ` Peter Zijlstra
2009-09-08 18:32 ` Peter Zijlstra
2009-09-09 14:23 ` Jan Kara
2009-09-09 14:37 ` Wu Fengguang
2009-09-10 15:49 ` Peter Zijlstra
2009-09-14 11:17 ` Jan Kara
2009-09-24 8:33 ` Wu Fengguang
2009-09-24 15:38 ` Peter Zijlstra
2009-09-25 1:33 ` Wu Fengguang
2009-09-29 17:35 ` Jan Kara
2009-09-30 1:24 ` Wu Fengguang
2009-09-30 11:55 ` Jan Kara
2009-09-30 12:10 ` Jens Axboe
2009-10-01 15:17 ` Wu Fengguang
2009-10-01 13:36 ` Wu Fengguang
2009-10-01 14:22 ` Jan Kara
2009-10-01 14:54 ` Wu Fengguang
2009-10-01 21:35 ` Jan Kara
2009-10-02 2:25 ` Wu Fengguang
2009-10-02 9:54 ` Jan Kara
2009-10-02 10:34 ` Wu Fengguang
2009-09-08 18:35 ` Chris Mason
2009-09-08 17:57 ` Chris Mason
2009-09-08 18:28 ` Peter Zijlstra
2009-09-09 1:53 ` Dave Chinner [this message]
2009-09-09 3:52 ` Wu Fengguang
2009-09-08 18:06 ` Theodore Tso
[not found] ` <20090908181937.GA11545@infradead.org>
2009-09-08 19:34 ` Theodore Tso
2009-09-09 9:29 ` Wu Fengguang
2009-09-09 12:28 ` Christoph Hellwig
2009-09-09 12:32 ` Wu Fengguang
2009-09-09 12:36 ` Artem Bityutskiy
2009-09-09 12:37 ` Jens Axboe
2009-09-09 12:43 ` Christoph Hellwig
2009-09-09 12:44 ` Jens Axboe
2009-09-09 12:51 ` Christoph Hellwig
2009-09-09 12:57 ` Wu Fengguang
-- strict thread matches above, loose matches on Subject: below --
2009-09-04 7:46 [PATCH 0/8] Per-bdi writeback flusher threads v18 Jens Axboe
2009-09-04 7:46 ` [PATCH 8/8] vm: Add an tuning knob for vm.max_writeback_mb Jens Axboe
2009-09-04 15:28 ` Richard Kennedy
2009-09-05 13:26 ` Jamie Lokier
2009-09-05 16:18 ` Richard Kennedy
2009-09-05 16:46 ` Theodore Tso
2009-09-07 19:09 ` Jan Kara
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=20090909015359.GC7146@discord.disaster \
--to=david@fromorbit.com \
--cc=akpm@linux-foundation.org \
--cc=chris.mason@oracle.com \
--cc=dedekind1@gmail.com \
--cc=fengguang.wu@intel.com \
--cc=hch@infradead.org \
--cc=jack@suse.cz \
--cc=jens.axboe@oracle.com \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=peterz@infradead.org \
--cc=tytso@mit.edu \
/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).