From: Mark Lord <lkml@rtr.ca>
To: jens.axboe@oracle.com
Cc: Rene Herman <rene.herman@gmail.com>, Tejun Heo <htejun@gmail.com>,
Alan Cox <alan@lxorguk.ukuu.org.uk>,
linux-kernel@vger.kernel.org, Linus Torvalds <torvalds@osdl.org>
Subject: Re: 2.6.20-rc2+: CFQ halving disk throughput.
Date: Mon, 01 Jan 2007 18:28:45 -0500 [thread overview]
Message-ID: <4599992D.8000607@rtr.ca> (raw)
In-Reply-To: <45998F62.6010904@gmail.com>
Rene Herman wrote:
> Tejun Heo wrote:
>
>> Everything seems fine in the dmesg. Performance degradation is
>> probably some other issue in -rc kernel. I'm suspecting recently
>> fixed block layer bug. If it's still the same in the next -rc,
>> please report.
>
> In fact, it's CFQ. The PATA thing was a red herring. 2.6.20-rc2 and 3
> give me ~ 24 MB/s from "hdparm t /dev/hda" while 2.6.20-rc1 and below
> give me ~ 50 MB/s.
>
> Jens: this is due to "[PATCH] cfq-iosched: tighten allow merge
> criteria", 719d34027e1a186e46a3952e8a24bf91ecc33837:
>
> http://www2.kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=719d34027e1a186e46a3952e8a24bf91ecc33837
>
>
> If I revert that one, I have my 50 M/s back. config and dmesg attached
> in case they're useful.
Wow.. same deal here -- sequential throughput drops from 40MB/sec to 28MB/sec
with CFQ -- whereas the anticipatory scheduler maintains the 40MB/sec.
Jens.. I wonder if the new merging test is a bit too strict?
There are four possible combinations, and the new code
allows merging for two of them: sync+sync and async+async.
But surely one of (not sure which) sync+async or async+sync may also be okay?
Or would it?
This is a huge performance hit.
Cheers
next prev parent reply other threads:[~2007-01-01 23:53 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-12-20 13:37 PATA -- pata_amd on 2.6.19 fails to IDENTIFY my DVD-ROM Rene Herman
2006-12-27 7:19 ` Tejun Heo
2006-12-27 17:50 ` Rene Herman
2006-12-28 2:24 ` Tejun Heo
2007-01-01 22:46 ` 2.6.20-rc2+: CFQ halving disk throughput Rene Herman
2007-01-01 22:56 ` Rene Herman
2007-01-01 23:30 ` Linus Torvalds
2007-01-01 23:28 ` Mark Lord [this message]
2007-01-02 8:34 ` Jens Axboe
2007-01-02 15:01 ` Mark Lord
2007-01-02 16:53 ` Tejun Heo
2007-01-02 17:35 ` Mark Lord
2007-01-02 17:53 ` Tejun Heo
2007-01-02 18:10 ` Mark Lord
2007-01-02 21:50 ` ATA_12 support for libata/ATAPI devices Mark Lord
2007-01-02 17:09 ` 2.6.20-rc2+: CFQ halving disk throughput Jan Engelhardt
2007-01-02 5:36 ` Andrew Morton
2007-01-02 8:44 ` Jens Axboe
2007-01-02 10:24 ` Rene Herman
2007-01-02 11:57 ` Jens Axboe
2007-01-02 12:10 ` Jens Axboe
2007-01-02 15:12 ` Mark Lord
2007-01-02 15:14 ` Jens Axboe
2007-01-02 15:40 ` Mark Lord
2007-01-02 15:45 ` 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=4599992D.8000607@rtr.ca \
--to=lkml@rtr.ca \
--cc=alan@lxorguk.ukuu.org.uk \
--cc=htejun@gmail.com \
--cc=jens.axboe@oracle.com \
--cc=linux-kernel@vger.kernel.org \
--cc=rene.herman@gmail.com \
--cc=torvalds@osdl.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.