From mboxrd@z Thu Jan  1 00:00:00 1970
From: Ondrej Zary <linux@rainbow-software.org>
Subject: Re: [git patches] IDE update
Date: Tue, 05 Jul 2005 12:19:25 +0200
Message-ID: <42CA5EAD.7070005@rainbow-software.org>
References: <200507042033.XAA19724@raad.intranet> <42C9C56D.7040701@tomt.net> <42CA5A84.1060005@rainbow-software.org> <20050705101414.GB18504@suse.de>
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1;
	format=flowed
Content-Transfer-Encoding: QUOTED-PRINTABLE
Return-path: <linux-ide-owner@vger.kernel.org>
Received: from mailhub3.nextra.sk ([195.168.1.146]:40461 "EHLO
	mailhub3.nextra.sk") by vger.kernel.org with ESMTP id S261800AbVGEKTd
	(ORCPT <rfc822;linux-ide@vger.kernel.org>);
	Tue, 5 Jul 2005 06:19:33 -0400
In-Reply-To: <20050705101414.GB18504@suse.de>
Sender: linux-ide-owner@vger.kernel.org
List-Id: linux-ide@vger.kernel.org
To: Jens Axboe <axboe@suse.de>
Cc: =?ISO-8859-1?Q?Andr=E9_Tomt?= <andre@tomt.net>, Al Boldi <a1426z@gawab.com>, 'Bartlomiej Zolnierkiewicz' <bzolnier@gmail.com>, 'Linus Torvalds' <torvalds@osdl.org>, linux-ide@vger.kernel.org, linux-kernel@vger.kernel.org

Jens Axboe wrote:
> On Tue, Jul 05 2005, Ondrej Zary wrote:
>=20
>>Andr=E9 Tomt wrote:
>>
>>>Al Boldi wrote:
>>>
>>>
>>>>Bartlomiej Zolnierkiewicz wrote: {
>>>>
>>>>
>>>>>>>On 7/4/05, Al Boldi <a1426z@gawab.com> wrote:
>>>>>>>Hdparm -tT gives 38mb/s in 2.4.31
>>>>>>>Cat /dev/hda > /dev/null gives 2% user 33% sys 65% idle
>>>>>>>
>>>>>>>Hdparm -tT gives 28mb/s in 2.6.12
>>>>>>>Cat /dev/hda > /dev/null gives 2% user 25% sys 0% idle 73% IOWAI=
T
>>>
>>>
>>>The "hdparm doesn't get as high scores as in 2.4" is a old discussed=
 to=20
>>>death "problem" on LKML. So far nobody has been able to show it affe=
cts=20
>>>anything  but that pretty useless quasi-benchmark.
>>>
>>
>>No, it's not a problem with hdparm. hdparm only shows that there is=20
>>_really_ a problem:
>>
>>2.6.12
>>root@pentium:/home/rainbow# time dd if=3D/dev/hda of=3D/dev/null bs=3D=
512
>>count=3D1048576
>>1048576+0 records in
>>1048576+0 records out
>>
>>real    0m32.339s
>>user    0m1.500s
>>sys     0m14.560s
>>
>>2.4.26
>>root@pentium:/home/rainbow# time dd if=3D/dev/hda of=3D/dev/null bs=3D=
512
>>count=3D1048576
>>1048576+0 records in
>>1048576+0 records out
>>
>>real    0m23.858s
>>user    0m1.750s
>>sys     0m15.180s
>=20
>=20
> Perhaps some read-ahead bug. What happens if you use bs=3D128k for
> instance?
>=20
Nothing - it's still the same.

root@pentium:/home/rainbow# time dd if=3D/dev/hda of=3D/dev/null bs=3D1=
28k=20
count=3D4096
4096+0 records in
4096+0 records out

real    0m32.832s
user    0m0.040s
sys     0m15.670s

--=20
Ondrej Zary