From: Jeff Garzik <jeff@garzik.org>
To: Andreas Schnaiter <schnaiter@gmx.net>
Cc: linux-kernel@vger.kernel.org
Subject: Re: Linux 2.6.16 - SATA read performance drop ~50% on Intel 82801GB/GR/GH
Date: Tue, 11 Apr 2006 23:57:06 -0400 [thread overview]
Message-ID: <443C7A92.6010604@garzik.org> (raw)
In-Reply-To: <200604120136.28681.schnaiter@gmx.net>
Andreas Schnaiter wrote:
> Hi,
>
> after upgrading from linux 2.6.15.7 to 2.6.16.2 I noticed an extreme slowdown
> of the SATA disks on my system. Writing/reading a 8GB file showed that
> reading performes with less than half the speed on 2.6.16 (strangely hdparm
> shows almost no difference).
> The two affected disks are connected to the Intel 82801GB/GR/GH (ICH7 Family)
> Serial ATA Controller.
> Disks on the Silicon Image/Intel IDE Controllers are not affected.
> I didn't have the chance yet to test if this problem also exists on the
> Silicon Image SATA Controller.
hdparm usually skips the usual layers, and benchmarks the ATA command
submission itself. So if hdparm didn't change, that may indicate the
problem is either in the block (scheduler?) or filesystem layers.
Tests:
1) Try with Silicon Image controller, as you mentioned. Try in same
machine with same device, if possible.
2) Eliminate the filesystem layer by doing dd directly to the block
device (dd ... of=/dev/sda1) rather than dd'ing to a file on a filesystem.
3) Try decreasing the block size to 4k or 8k.
4) Finally, the thing that will help us kernel hackers the most, use
'git bisect' to definitively discover the changeset that causes your
problems.
Jeff
next prev parent reply other threads:[~2006-04-12 3:57 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-04-11 23:36 Linux 2.6.16 - SATA read performance drop ~50% on Intel 82801GB/GR/GH Andreas Schnaiter
2006-04-12 3:57 ` Jeff Garzik [this message]
2006-04-12 9:49 ` Jan Engelhardt
2006-04-12 21:22 ` Bill Davidsen
2006-04-14 23:00 ` Linda Walsh
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=443C7A92.6010604@garzik.org \
--to=jeff@garzik.org \
--cc=linux-kernel@vger.kernel.org \
--cc=schnaiter@gmx.net \
/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